I Hf. 313

‘3
Em
C;

(a

 

LIBRARY
Michigan State
University

1 ' A

 

 

This is to certify that the

dissertation entitled

CHARACTERIZATION OF NUCLEAR RECEPTORS
OF THE ECDYSONE GENE REGULATORY HIERARCHY
DURING MOSQUITO VITELLOGENESIS

presented by

Chao Li

has been accepted towards fulfillment
of the requirements for

phD 638mm Genetics

fl! /42/%é/

Major professor

 

6-24-2001
Date

 

MSU it an Affirmative Action/Equal Opportunity lamination 0 - 12771

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8/01 cJClFiC/DateDuepss-sz

 

 

CHARACTERIZATION OF NUCLEAR RECEPTORS
OF THE ECDYSONE GENE REGULATORY HIERARCHY
DURING MOSQUITO VITELLOGENESIS

By

Chao Li

A DISSERTATION

Submitted to
Michigan State University
In partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Genetics Program

2001

Tilt

:I. Q .
ICELIJIIM I
& V

 

tepmdudm
that this 5
tanst‘ripnu
knouicdgc
meal is red
hamn and

precursors (

ABSTRACT
CHARACTERIZATION OF NUCLEAR RECEPTORS
OF THE ECDYSONE GENE REGULATORY HIERARCHY
DURING MOSQUITO VITELLOGENESIS
By

Chao Li

The insect steroid hormone, 20-hydroxyecdysone (20E), plays a critical role in
regulating widely diverse physiological functions of development, metamorphosis and
reproduction. The mode of 20E action is well understood in Drosophila development,
where this steroid hormone regulates genes via a complex hierarchy containing several
transcription factors belonging to the nuclear receptor superfamily. In contrast, our
knowledge of 20E action in insect reproduction has been limited. In mosquitoes, a blood
meal is required for egg development and as a consequence, they are vectors of numerous
human and domestic animal diseases. The process of production of yolk protein
precursors or vitellogenesis is central to egg development and is regulated by 20E. The
goal of my thesis has been characterization of nuclear receptors of the 20E gene
regulatory hierarchy involved in the mosquito vitellogenesis. The functional receptor for
2013, which is at the top of the hierarchy, is a heterodimer of two nuclear receptors, the
ecdysone receptor (EcR) and Ultraspiracle (USP). There are two isoforms for both EcR
and USP in mosquitoes, which are differentially expressed in a stage- and tissue-specific
manner during vitellogenesis. In vitro fat body culture experiments suggest that the
ligand, 20E itself, differentially regulates expression of AaEcR and AaUSP isoforms. In
Drosophila, the action of the EcR/USP heterodimer is mediated by the product of an

early gene E75, which is an orphan nuclear receptor. Analysis of the E75 gene has

  
   

Show Ii
xitc‘llogct‘
53:35. \I i'
guitars. .i
ctgiohm‘.
SCESIIIW It

10 directl} .

 

pm me! {tugs
competence
expressed h
are Iou. but
bod} cultur
Nth-31rd“ 3.1

ii‘Suils in St

demonstrat:

Ric-rem cr, 1

List

'fl
v

.Oprnc:
AHR‘sihfi

AHR‘S mi

It]

ECd'v 50178 F“
' g

Chao Li
Shown that the ZOE-regulated hierarchy is used reiteratively during mosquito
vitellogenesis. In the fat body, AaE 75 mRNA is not present during the previtellogenic
stage, while AaE 75 is highly expressed during the vitellogenic stage. In in vitro fat body
culture, AaE 75 is rapidly induced by 205 and this induction is not blocked by
cycloheximide. Dose response experiments have revealed that AaE 75 is ten fold more
sensitive to induction by 20E than the yolk protein gene, Vg. Furthermore, AaE75 is able
to directly activate the Vg promoter in cell transfection assays, suggesting its role as an
early gene. The mosquito becomes competent for a vitellogenic response to 20E after
previtellogenic development. The nuclear orphan receptor AaFTZ-F 1 serves as a
competence factor for the stage-specific response to 20E. AaFTZ-FI transcript is
expressed highly in the fat body during pre- and post-vitellogenic periods when 20E titers
are low, but it disappears in mid-vitellogenesis when 20E titers are high. In in vitro fat
body culture, AaFTZ-FI expression is inhibited by 20B and super-activated by its
withdrawal. Following in vitro AaFTZ-FI super-activation, a secondary 20E challenge
results in super-induction of the early AaE 75 gene and the late target VCP gene,
demonstrating involvement of AaF 72-F1 in developing additional 20E responsiveness.
Moreover, the onset of ZOE-response competence in the mosquito fat body is correlated
with the appearance of the functional AaFTZ-Fl protein at the end of previtellogenic
development. Finally, I have identified another mosquito nuclear orphan receptor
AHR78 that is expressed at a high level only in the ovary during the vitellogenic period.
AHR78 inhibits the DNA binding and transactivation activities of AaEcR/AaUSP on the

ecdysone response element of the Vg promoter.

lslnt‘crc
continua
disciplira.
acadcmit
enhanced
Dar id Ar:

Third I \k;

 

ACKNOWLEDGMENTS

I sincerely appreciate my major professor, Dr. Alexander Raikhel, for his guidance and
continued support throughout my graduate studies. His thoughtful, patient and
disciplined mentoring provided me invaluable guidance, while allowing sufficient
academic freedom and thus my interest and understanding of biomedical research were
enhanced. Second, I would like to thank my guidance committee, Drs. Donald Jump,
David Arnosti, and Ke Dong, for their help and suggestions during my five-year studies.
Third, I wish to thank the friends I made in the Raikhel lab and the Genetics Program for
their friendship and advice. Especially, I greatly appreciated Dr. Ken Miura for his
detailed teaching of molecular biology techniques at the beginning of my studies, Dr.
Sheng-fu Wang for his helpful discussions during my experiment difficulties, and Mr.
Alan Hays for culturing mosquitoes and other technical helps. I also thank people who
have been doing excellent collaborative work with me including Drs. Marianna
Kapitskaya, S-F. Wang, Jinsong Zhu, K. Miura, Avraham Biran, and Kook-Ho Cho. In
addition, I would like to thank Drs. William Pierceall and William Segraves at Yale
University for cloning the AaE 75 gene and editing my manuscripts. I am also very
grateful to Michel Trail for editing a manuscript, and S-F. Wang and Geoffrey Attardo
for editing my dissertation. Finally, I would like to thank my parents and brothers for
their love, support and encouragement from the beginning to the end of my graduate

studies.

list oil _
list of I.
List of A

Chapter 1

 

 

GAP-[CI 2.

7"."
Hi »_
‘riz‘r .1

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

List of Tables vii
List of Figures viii
List of Abbreviations xii
Chapter 1. Literature Review 1
Molecular mechanism of nuclear receptor action 2
Nuclear receptors in the insect ecdysone gene regulatory hierarchy --------- 1 1
Nuclear receptors in mosquito vitellogenesis 19
Rationale for current studies 23
Chapter 2. Characterization of two Isoforms of the Mosquito Ecdysone
Receptor (EcR) and Ultraspiracle (USP) 27
Abstract 28
Introduction 29
Materials and Methods 32
Results 38
Discussion 58
Chapter 3. Characterization of the Mosquito Early Gene E75 68
Abstract i 69
Introduction 70
Materials and Methods 73
Results 79
Discussion 99
106

Chapter 4. Cloning and Characterization of the Mosquito Early-late Gene HR3 -------

Chapter 5

Chaiter 6.

 

Re"

 

Abstract

 

Introduction

 

Materials and Methods

 

Results

 

Discussion

Chapter 5. Cloning and Characterization of the Mosquito F 72-F1 Gene

 

 

Abstract

 

Introduction

 

Materials and Methods

 

Results

 

Discussion

 

Chapter 6. Cloning and Characterization of the Mosquito HR 78 Gene

 

Abstract

 

Introduction

 

Materials and Methods

 

Results

 

Discussion

 

Chapter 7. Summary and Future Research Perspectives

 

References

vi

107

108

112

116

128

133

134

135

139

145

163

168

169

170

174

178

191

195

204

Chapter
Table 1.(

Chapter

I] ~-
“ 1
1.10:; l. .

 

Chapter:
Table l. .\

LIST OF TABLES

Chapter 1
Table 1. Conservation of nuclear receptors and cofactors between insects and
vertebrates l8

 

Chapter 2
Table 1. The equilibrium dissociation constants (Kd) of different DNA sequences
binding to AaEcR/USP-A and AaEcR/USP-B 57

 

Chapter 5
Table 1. Sequences of the PCR primers used in experiments 142

 

vii

 

C haptcr 2

Fit; 1. Ev

Fig. 2. D...

 

~ 0

(I .4
S7.
-1
r;

LIST OF FIGURES

Chapter 1

 

Fig. 1. Functional domains of nuclear receptors

Fig. 2. Activation and repression of a gene by nuclear receptors through

 

coactivators or corepressors

10

 

Fig. 3. Structures of major insect hormones and their vertebrate analogs

Fig. 4. Summary of events during the first cycle of vitellogenesis in the

 

anautogenous mosquito, Aedes aegypti

Chapter 2

 

Fig. 1. Expression profiles of USP and EcR mRNA during vitellogenesis

Fig. 2. Dose response of USP-B, EcR and VCP to 20E in in vitro fat body

12

20

40

43

 

organ culture

Fig. 3. Effect of 20-hydroxyecdysone and cycloheximide on the transcription of

 

USP-A, USP-B, EcR and VCP in the previtellogenic fat bodies

Fig. 4. Developmental profiles of USP proteins revealed by Western blot

 

Fig. 5. EcR/USP binding to a direct repeat DR-4

 

 

Fig. 6. EcR/USP binding to DR-12

 

Fig. 7. Direct measurement of equilibrium dissociation constants (Kd)

Fig. 8. EcR/USP-A transactivated a reporter gene with either IR-l or

 

DR-4 EcRES in CV-l cells

47

50

50

50

53

55

Fig. 9. AaEcR/USP-B mediated more efficient transactivation than EcR/USP-A ------- 57

Fig. 10. Characteristics of the AaEcR-A isoform

 

viii

66

Chapter 3

 

 

 

 

 

 

 

 

 

 

Fig. 1. Nucleotide and amino acid sequences of AaE75 81-83
Fig. 2. Genomic organization of AaE75 86
Fig. 3. Comparison of A edes aegypti and D. melanogaster E75

amino acid sequences 86
Fig. 4. AaE 75 transcripts in the mosquito ovary 88
Fig. 5. AaE75 transcripts in the mosquito fat body 88
Fig. 6. Accumulation of AaE75 mRNA in the mosquito fat body 90
Fig. 7. Expression of AaE75 RNA isoforms in the mosquito ovary during

the first vitellogenic cycle 92
Fig. 8. Accumulation of AaE75 mRNA in the mosquito ovaries 92
Fig. 9. AaE75 response to 20-hydroxyecdysone in in vitro fat body organ culture ------ 94
F ig.10. Effect of cycloheximide on the activation of the AaE 75 gene 94
F ig.1 1. Time course of AaE 75 induction in in vitro fat body organ culture --------------- 95
F ig.12. Action of E75 and EcR/USP on the Vg Promoter 98
Chapter 4

Fig. 1. Alignment of the deduced amino acid sequence of the mosquito

AHR3 cDNA with its insect homologues 118

 

Fig. 2. Domain comparison of the mosquito HR3 with selected members of the

 

ROR/HR3 subfamily of nuclear receptors 120

Fig. 3. AHR3 transcripts in the mosquito A edes aegypti 122

 

 

ChapterS
is l. 8141

Lilli!

 

\itc

.lJ

(It

(is

-1ni

a.

.' .7
(":4
0‘

E.‘.~
‘ . s:

1W
in n

Fifi”

(’7
m
(1.
r

in ll:-

Fig. 4. Expression of AHR3 in the mosquito ovary during the first

 

vitellogenic cycle 123
Fig. 5. Expression of AHR3 in the mosquito fat body during the first

vitellogenic cycle 123

 

Fig. 6. AHR3 response to 20-hydoxyecdysone in in vitro fat body organ culture---- 125
Fig. 7. Time course of AHR3 induction by 20E in in vitro fat body organ culture---- 125

Fig. 8. Effect of cycloheximide on the AHR3 induction by 20E in in vitro

 

fat body organ culture — 127

Chapter 5
Fig. 1. Summary of events during the first cycle of vitellogenesis in the

anautogenous mosquito, Aedes aegypti 137

 

Fig. 2. Northern blot analysis of stage-specific induction of the vitellogenin and

vitellogenic carboxypeptidase transcripts by the hormone 20E 146

 

Fig. 3. Alignment of the deduced amino acid sequence of the mosquito

 

AaFTZ-FI cDNA with its insect homologs 149
Fig. 4. AaFTZ-FI expression profiles in the female mosquito fat body and pupae-- 152
Fig. 5. In vitro repression of AaF TZ-F I expression by the hormone 20E, and super-

induction of AaF YZ-F 1 by withdrawal of 20E in fat body organ culture ---- 155
Fig. 6. Effects of cycloheximide on the 20E regulation of AaF TZ—F I

in in vitro fat body organ culture 157

 

Fig. 7. Secondary responses to 20E of the ecdysone-inducible genes

in in vitro fat body culture 159

 

Fig. 8. Identification of the functional AaFTZ-F 1 protein in fat body

 

nuclear extracts of previtellogenic female mosquitoes 161

Chapter 6
Fig. 1. Alignment of the deduced amino acid sequence of the mosquito

AHR78 cDNA with its insect homologs 179-180

 

Fig. 2. Domain comparison of the mosquito AHR78 with selected members

of the HR78/TR2 subfamily nuclear receptors 18]

 

Fig. 3. AHR 78 expression profiles in the female mosquito fat body and ovary ------ 183
Fig. 4. Induction of AHR78 mRNA by 20E in the ovary but not the fat body-------- 185
Fig. 5. Inhibition of EcR/USP binding to the natural EcREs by AHR78 ------------- 187
Fig. 6. AHR78 inhibition of ZOE-stimulated transactivation by AaEcR/AaUSP

in 82 cells 189-190

 

xi

   
 

ME. 21'-
A31 .lt..‘
AR: ant:
B-Cul. [
Bmzsfih.
Bp;bu\c
CAltd1
cDNAio
lhnzlhvn

 

DRidnect
ECRICCd};
EMS-‘11 Cl;
E‘REZCCQ
Emiethg
ER3 851

10 3

LIST OF ABBREVIATIONS
20E: 20-hydroxyecdysone
Aa: Aedes aegypti
AR: androgen receptor
B-Gal: B—galactosidase
Bm: Silkworm Bombyx mori
Bp: base pair
CAT: chloramphenicol aceyltransferase
cDNA: complementary DNA
Dm: Drosophila melanogaster
DNA: Deoxyrbonucleic acid
DR: direct repeat
EcR: ecdysteroid receptor
EMSA: electrophoretic mobility shift assay
EcRE: ecdysteroid response element
Eip: ecdysone inducible protein
ER: estrogen receptor
GR: glucocorticoid receptor
GST: glutathione S—transferase
HRE: hormone response element
Hsp: heat Shock protein 27
IR: inverted repeat

III: juvenile hormone

xii

Kb: kilobase pair

Kd: equilibrium dissociation constant
LBD: ligand binding domain

Luc: luciferase

MR: mineralocorticoid receptor
mRNA: messenger RNA

MurA: muristerone A

NR: nuclear hormone receptor
PAGE: polyacrylamide gel electrophoresis
PR: progesterone receptor

PBM: post blood meal

PCR: polymerase chain reaction
RAR: retinoid acid receptor

RNA: ribonucleic acid

RT: reverse transcription

RXR: retinoid X receptor

82: Drosoplzila Schneider cell line 2
SDS: sodium dodecyl sulfate

Tm: yellow mealworm Tenebrio moliter
TR: thyroid hormone receptor

USP: Ultraspiracle

VCP: vitellogenic carboxypeptidase

VCB: vitellogenic cathepsin-B

xiii

l'g: tit;
VDR: \'
\gR: it:

IPP: )t

 

 

Vg: vitellogenin
VDR: vitamin D receptor
VgR: vitellogenin receptor

YPP: yolk protein precursor

xiv

Chapter 1

Literature Review

  

piece»;
receptor
hindirtu:

J . t._
mt'iCkUi’t

these liga

 

and mind

titamm D

MOLECULAR MECHANISM OF NUCLEAR RECEPTOR ACTION

General structures and functions of nuclear receptors

Steroid/Thyroid hormones play critical roles in a wide array of biological
processes such as development, reproduction and metabolism. Classical nuclear hormone
receptors are ligand-activated transcription factors that activate gene expression upon
binding to their respective ligands. The ligands for nuclear receptors are small lipophilic
molecules such as steroids, retinoids, vitamin D, and thyroid hormone. The receptors for
these ligands are: the sex steroid receptors including estrogen (ER), progesterone (PR),
and androgen (AR) receptors; the adrenal steroid receptors including glucocorticoid (GR)
and mineralocorticoid (MR) receptors; and others including thyroid hormone (TR),
vitamin D (VDR), retinoic acid (RAR) and 9-cis retinoic acid (RXR), and ecdysone
(EcR) receptors. In addition, a large number of so-called nuclear orphan receptors have
been identified. While these receptors have similar structures to nuclear hormone
receptors, their ligands are unknown. The nuclear receptor superfamily is comprised of
both nuclear hormone receptors and orphan receptors (Tsai and O’Malley, 1994;
Mangelsdorf et al., 1995).

One striking structural characteristic of the nuclear receptor superfamily is that
most members contain five to six evolutionarily conserved functional domains, which can
be divided into five regions starting from the N-terminal: A/B, C, D, E and F (Fig. 1).
The A/B domain iS most variable among family members and contains the ligand-
independent transactivation function (AF 1). The C domain is also named the DNA
binding domain (DBD). The D domain acts as a hinge region between the C and E

domains. It contains a nuclear localization Signal, as well as corepressor binding sites on

.fl x20: 00 :BoE om? E 0:00 NA? 05. .qu 0% EEE» 5:05:
5:038: “5:53??me— 05 8 80:0: Nd? .:o:0:£ 5050383505 603302720? 0% 2 £80: E0820 m2 05 E TUE
dougcommgb Em :ESEBELoEESmS acumNtBE: 5083080: 305 E88: 0—005 80: mo 226:3 3:256: mm: Amy E050:
wEuEn Bum: 051 $8300: 320:: 088 E wEvEn 838380 :50 5:03:02 80—05: :8 60:an 80:03:00 068 0:5:00
ADV :Ew0: 0wE: 0f. .>__Eat0qsm 05 E 5:023:00 00:03:00 52%:— 05 mm: ADV E060: wEwEn <75 05. .EEEESQS 05
w:oEm 00:0:c0m DEER “Em 0E0 E ES, mEmEoc m “0:0 m2 d2 :0 REES“- Z 05 E3 0:080:00 05 E :32? .mEmEov 3:288:

_80>0m («0 “0008800 0:0 EEEESQE 003000: 50—03: 0:: .«0 30980:: 05. 6.003000.— ..a0_0=: he 055:3. _a:e:0:=m .—.w_.m

 

 

m L 1 _ :0=a>:0am:u..._.

 

 

 

 

_ L W L 5:09.85 0803.00-00

 

 

a _ £50335 300050.700

 

_ _ w=_0=£-<z:

 

_l _ H 52.305:—

 

T L :EEESQ am:

 

 

_ L “50:50:08.8:

j: :0:

0.80 Name. Tm<

 

 

 

 

 

 

‘1' 11 1,-1Iv

 

h m— : 0 m2

nuclear
(1811!.
is at the

11st: ar-

 

rerton l
fi

sequent-c

 

superlaml
Ti'picall}
Mitch an
080 cor
geemctn
10110“ in‘

rat

’)I\\r .
It‘gltfiin'x

mfirLE

” ‘ N
1119 ,
H‘ ALLC

nuclear hormone receptors. The E domain is characterized as the ligand-binding domain
(LBD), which binds to ligands and possesses activation functions (AF 2). The F domain

is at the very C-terminal and varies in size and primary sequence among the superfamily
(Tsai and O’Malley, 1994).

A central and common feature of nuclear receptors is their DBDS. The DBD
region has dual functions: hormone response element binding and dimerization. The
sequences in the DBD exhibit remarkable conservation among the nuclear receptor
superfamily, and this has proven useful for the cloning of many other nuclear receptors.
Typically, the DNA-binding region consists of 66 highly conserved amino acid residues,
which are referred to as the “core” region of the DBD in the nuclear receptors. The core
DBD contains two zinc-nucleated subdomains. Each zinc is coordinated in a tetrahedral
geometry by the sulfurs from conserved cysteines. The hinge region immediately
following the C-terminal of the DBD can also contribute significantly to the DNA
binding and target specificity of the receptor (Rastinejad, F., 1998; Luisi et al., 1991).

Most receptors function as homo- or heterodimers, binding selectively to gene-
regulatory sequences called hormone response elements (HRES). There are two distinct
regions that maintain the dimerization function. First, certain receptors can dimerize via
their LBDS in the absence of DNA. Second, more functionally relevant dimerization
interfaces are formed through a separate interface between the receptors’ DBDS only in
the presence of an appropriate HRE (Perlmann et al., 1993; Zechel et al., 1994; Hard et
aL,1990)

Nuclear receptors often bind to bipartite DNA targets as a dimer. Despite the large

number of nuclear receptors, only two types of consensus hexameric half-sites appear to

repeatr‘

AGAAt

 

COCPC
thhclt
113C P‘bk

Receptor

 

moscthrt
CUVC>PUI
$-193g
I
amour}
meadda
l .
mtduni

r \
tfixcplvr.

be used as targets. The two most common half-Site arrangements are the symmetric
repeat and the direct repeat with different inter-half-Site spacers. The steroid receptors
for glucocorticoids, mineralocorticoids, androgens, and progesterone, bind to 5’-
AGAACA-3’ half Sites; the estrogen receptor (ER), RXR, TR, VDR, RAR, PPAR,
COUP-TF, and NGFI-B preferentially bind to consensus 5’-AGGTCA-3’ half sites
(Klock et al., 1987; Umenoso et al., 1991). Three critical residues in the DBDS, called
the P-box, predispose receptors to preferentially bind the GR versus the ER elements.
Receptors recognizing the AGAACA half sites contain a GSXXV in their P-box, while
those that recognize AGGTCA half-sites typically contain EGXXG residues at the
corresponding positions. Thus, the P-box is critical for half-Site discrimination (Mader et
al., 1989).

The LBD of nuclear receptors is a large domain that sometimes encompasses
almost the entire C terminus of these proteins. The LBD has been credited with having
the additional functions of hormone-dependent dimerization and transactivation. Nuclear
localization and heat Shock protein association functions also require the LED of some
receptors. The binding of hormone results in receptor transformation and the initiation of
a series of events by which receptors activate the expression of target genes (Kumar et
al., 1987; Dobson et al., 1989; Allan et al., 1992).

Interaction with a molecular chaperone-containing heterocomplex (MCH) is a
pre-requisite for activation of steroid receptors. This interaction includes the molecular
chaperones (Hsp90 and Hsc70), molecular chaperone interacting proteins (Hop, Hip, and
p23), and peptidyl- prolyl isomerases (F KBP51, F KBP52, and Cyp40). In the case of PR

and GR, the MCH is required for maturation to the aporeceptor state (DeFranco et al.,

I‘ll'O t, B'

contort“

 

ehpICHZL

.111 Egttngq

 

basal trail.

Ohl'n (“Ck

1998). Association of hsp90 is required for steroid binding to GR both in vitro and in
vivo (Bresnick et al., 1989; Picard et al., 1990), and the binding of hsp90 to GR causes a

conformational change in the GR LBD (Stancato et al., 1996).

Roles of cofactors in gene activation and silencing by nuclear receptors

Upon hormone binding, receptors bind to DNA and activate target gene
expression. Early studies have demonstrated that PR, ER and GR activate transcription
by enhancing the formation of stable preinitiation complexes at their target promoters by
either direct interaction with components of the transcription machinery, or through an
intermediate factor such as a coactivator (Klein-Hitpass et al., 1990; Tsai et al., 1990;
Elliston et al., 1990). In the past few years, numerous nuclear receptor coactivators have
been identified. They interact directly with the activation domain of a nuclear receptor in
an agonist-dependent manner, and also interact with other factors and components of the
basal transcription machinery but do not enhance the basal transcription activity on their
own (Mckenna et al., 1999; Robyr et al., 2000; and Glass and Rosenfeld 2000).

The first cloned nuclear receptor coactivator, steroid receptor coactivator-1 (SRC-
l), represents one of the well-documented coactivators for nuclear receptors (Onate etal.,
1995). Many other subsequently identified coactivators with Structures and functions
Similar to SRC-1 comprise the SRC coactivator family. The SRC family can be divided
into three subfamilies: SRC-l/NCoA-l, SRC-2/GRIPl/TIF2/NcoA-2, and SRC-
3/p/CIP/RAC3/ACTR/AIB-l/TRAM-1. SRC coactivators Share an ability to stimulate
ligand-dependent transactivation by a rather large number of nuclear receptors in

transient overexpression experiments. These coactivators also exhibit structure

 

shnflar

boxcs.n

1.81% t'

 

ehccs"
mcconu
dantp"t
this in.“I
limit-err;
till (0ng
receptors.
hamsteras
addition, I

 

1X11 e! d!”

MI,
3132‘)”.

“é‘l—i‘il‘ix't

Similarities. A distinct feature is the presence of multiple LXXLL Signature motifs or NR
boxes, which comprise of determinants for direct interactions with the nuclear receptor
LBDS (Mckenna et al., 1999).

Several LBD crystal structures have established that upon ligand binding, the a
helices in the LBD, including the AF2 core (helix 12), undergo a major reorientation in
the context of the overall LBD structure. This reorientation forms part of a “charged
clamp” that accommodates SRC coactivators within a hydrophobic cleft of the LBD.
This interaction appears to occur through direct contacts with the LXXLL motif.
Furthermore, SRC coactivators interact with cointegrators CBP/p300 and together they
can coactivate synergistically, while C BP/p300 itself interacts directly with nuclear
receptors. SRC and CBP/p300 coactivators both possess intrinsic histone acetyl
transferase (HAT) activity and therefore can act in concert to remodel chromatin. In
addition, p/CIP and P/CAF coactivators are recruited to nuclear receptors in response to
hormone binding as part of a growing, HAT-containing, chromatin-remodeling complex
(Xu et al., 1999; Freedman 1999).

Moreover, nuclear receptors can repress transcription under various
circumstances. Repression occurs mostly in the absence of a ligand or when an
antagonist is bound to the receptor. It has been established that the unliganded TR is able
to bind DNA acting as a repressor. In the past several years, three active repressors have
been identified: SMRT (silencing mediator for retinoid and thyroid hormone receptor)
(Chen and Evans, 1995), N-COR (nuclear receptor corepressor) (Kurokawa et al., 1995;
Horlein et al., 1995) and SUN-COR (small ubiquitous nuclear corepressor) (Zamir et al

1997). These corepressors are able to interact with the unliganded TR and RAR

 

 

 

Fig. 2. Activation and repression of a gene by nuclear receptors through
coactivators or corepressors. The nuclear hormone receptor (NR) is associated with a
corepressor (N-COR, SMRT), which in turn recruits a histone deacctylase (HDAC)
through its interaction with Sin3. Deacctylation of histonc tails leads to transcriptional
repression. Addition of the ligand disrupts this repression complex in favor of the
association of a coactivator complex (SRC-l, P/CAF, p300/CBP, pC IP, and others).
These proteins possess a histone acetyltransferase activity that allows chromatin
decompaction through histone modifications. The interaction between the nuclear
receptor AF -2 domain and the coactivation complex occurs through the LXXLL motif in
many coactivators. The coactivator and corepressor complexes are represented with
dashed lines since their exact composition in vivo is not determined. Modified from

Robyr et al. (2000).

 

 

  

Histone deacetylatlon

\.

 

 

 
  

LiganN

Co-actlvators
recruitment

 

\) Co-re ressors
re ease

Hlstone acetylation

 

 

 

 

domah
uponlz,
dentun~
rerunr
ofSNIR

HDAC-

factor Til

Kills 1.

 

nuclear hl
Remittr

iRklb‘fl' t):

associated with RXR on DNA. The C terminus of N-CoR interacts with the hinge
domain and a portion of the LBD in TR and RAR. The silencing activity is abolished
upon ligand binding by the release of corepressors from the receptor. Further studies have
demonstrated that SMRT and N-CoR mediate transcription repression through
recruitment of a histone deacetylase complex. The N terminus repression domain (SD-1)
of SMRT interacts with Sin3A, which in turn associates with a histone deacetylase
HDAC-l through one of its two Silencing domains. Thus, chromatin modification
through deacetylation plays an active role in nuclear receptor mediated gene Silencing. In
addition, TR silencing is mediated by its direct interaction with the general transcription
factor TFIIB. This interaction is interrupted by thyroid hormone binding (Robyr et al.,
2000).

As a general model (Fig. 2), association of coactivators or corepressors with a
nuclear hormone receptor is a ligand-dependent process. Coactivators and corepressors
recruit HATS and HDACS, respectively, that finally result in gene activation or repression

(Robyr et al., 2000).

NUCLEAR RECEPTORS IN THE INSECT ECDYSONE GENE
REGULATORY HIERARCHY

In insects, two classes of lipophilic hormones control many events in
metamorphosis, development and reproduction. One class is the steroid hormones known
as ecdysteroids. The active form of ecdysteroid appears to be 20-hydroxyecdysone
(20E). The other class is the scsquiterpenoids also known as juvenile hormones (Fig. 3)

(Riddiford, 1985; Bownes, 1986; Dhadialla and Raikhel, 1994). During the last decade,

. I DJ."

:3:
a"

J ‘i"."..- i ll- .
' ”Pk U‘lgllnhi’l’l'i ' l

7‘ " 'I‘ki'i in I 'w ’I’nctiq '
if” :1 0 l1» 1 1“: h"... «an ""
.1 ‘

I-
-.
,.
:.-.
,;.
‘1
..
-‘..
..o
1".3‘
be.
u.“-
1“.
‘3‘
.".
.,E
0‘:
7‘
n.-
to~
58‘,»
16

 

 

' :...l . t
. l w-
ir‘hdu‘" \‘1 t: ‘01 "‘

”v ' . r . .L; . ‘i
.3. Lil .rr ' "u r. -,.H I-f I “hug-trio
‘fi'lr..m""'“ 1:.1i..t..m.r.\-" "F...” nu ‘
I . '.

_ u owl fl r“"._‘

it‘“

I I

 

 

20E Estradiol

0 COOH

Retinoic acid

Fig. 3. Structures of major insect hormones and their vertebrate analogs.
20-hydroxy-ecdysone (20B) and juvenile hormone 111 (JH III) are insect
hormones; Estradiol and 9-cis retinoic acid (retinoic acid) are two vertebrate

hormones. The arrow on 20E indicates the difference between ecdysone and
its active form (ZOE).

 

l

.1 ,,
moxie.

iNRlll
BUIII Ex I
domains

ECR 15011

 

.i. , .
31:931le

UHdcrth;

111611111100:

the molecular mechanism of 20E action has been studied in detail (Thummel, 1995;
Raikhel et al., 1999; Kozlova and Thummel, 2000; Riddiford et al., 2001), while the
molecular basis underlying juvenile hormone action remains obscure (Riddiford, 1996).

The functional receptor for 20-hydroxyecdysone is the heterodimer of EcR
(NRlHl) and Ultraspiracle (USP; NR 284) (Koelle et al., 1991; Yao et al., 1992, 1993).
Both EcR and USP have multiple isoforms that differ only in their N-terminal A/B
domains in many insects. In Drosophila, there exists one EcR gene, which encodes three
EcR isoforms: EcR-A, ~Bl and-82. EcR-A and EcR-B result from utilization of
alternative promoters while EcR-Bl and EcR-BZ are products of alternative Splicing
under the same promoter (Koelle etal., 1991; Talbot et al., 1993). The EcR isoforms
identified in other insects are named similarly to the Drosophila EcR isoforms.

In Drosophila, only one isoform of USP encoded by a Single gene is known (Oro
et al., 1990; Henrich er al., 1990; Shea et al., 1990). Yet two different molecular weight
proteins are found, possible due to the use of an alternative translational start site
(Henrich et al., 1994). By contrast, two USP isoforms have been found in three other
insects including, A edes aegypti (Kapitskaya et al., 1996), Manduca (Jindra et al., 1997),
and Chironomus tentans (Vogtli et al., 1999). It is unclear thus far whether these
isoforms result from alternative splicing of a single gene, or from different genes. The
A/B domains of Aedes USP-A, Chironomus USP-1, and Manduca USP-1 exhibit the
highest Similarity, particularly at the N-terminus, to that of the single USPS of
Drosophila, the Silkworm Bombyx mori (Tzertzinis et al., 1994), and Spruce budworm

Choristoneurafumiferana (Perera et al., 1999).

 

 

respon ,~ ~

x x l l . .. . .
. MW. NH 1. I. ll . u a W a e 1... n1 run iu
fit. .1 . r n n .4 Pl. r d H1 \ L \0 P a .\. i .L
10 flu l. J W .U 1 . . r . . C a: . 1h. a C
. h Oil Ill hit 351 tV c n I 9‘ 45 1 it It. \\ iL
«N “u Fl «\_ U n n “I ‘1 0 that H;
N. MN on U; D. . a fit it... - r I. - .C t .\ .0 F
A. «Au ydrU Q Alfis r I “L; iu AL ..l._ A\ . I .4 i“ . . ..
1W1. Di , 1 It. Up n... 3 R S 0: run 0.: ml. .\ h. «A. .Q

Both EcR and USP are required for 20E binding and binding to ecdysone
responsive elements (EcRE). The first EcRE was identified in the promoter of the heat
shock protein-27 gene in Drosophila (Riddihough and Pelman; 1987). It is an imperfect
palindrome with only a 1-bp spacer (hsp27). Several other EcREs have also been
identified in the regulatory regions of four Drosoplu'la genes: Eip28/29 (Cherbas et al.
1991), Fbp-l (Antoniewski er al., 1994), Sgs-4 (Lehmann and Korge, 1995), and Lsp-Z
(Antoniewski et al., 1995), each of which contains an imperfect inverted repeat with a 1-
bp spacer (IR-1). These findings suggest that natural Drosophila EcREs are
predominantly IR-ls. Nonetheless, artificial direct repeats (DR) with different spacers
appear to be functional as well (Homer et al., 1995). In addition, a direct repeat EcRE
with a 12-nucleotide spacer exists in the Drosophila 3C/ng (D’Avino etal., 1995).

In invertebrates, perhaps the most surprising discovery in the nuclear receptor
superfamily is the large number of genes that are regulated by ecdysone. Most of these
genes are nuclear orphan receptors, which are either induced or repressed by this
hormone at the level of transcription. In Drosophila, the discovery of most of these
genes arose from efforts to understand how ecdysone directs early stages of
metamorphosis. Early studies from polytene chromosome puffing patterns in larval
salivary glands have revealed that ecdysone signaling is transduced and amplified via a
two-step regulatory hierarchy (Ashbumer et al., 1974), which is widely cited as the
Ashbumer model. At the ends of larval and prepupal development, pulses of ecdysone
directly induce a small number of “early genes”, at least three of which encode
transcription factors: E74 (Burtis et al., 1990; Thummel et al., 1990), and two nuclear

receptors: E75 (Segraves and Hogness, 1990) and BR-C (DiBello etal., 1991). These

 

mdyg

S€CO ‘.

 

HOE'IIC‘»
d‘élll‘iCl .
first Zin,
“Nhthc

pl Unity“

1:35
a
‘i93e\

early genes both repress their own expression and induce more than 100 “late genes” as
secondarily responsive genes. Interestingly, the early genes are expressed widely, while
many late genes are expressed in a tissue-specific manner such as the Sgs-4 glue and L7]
genes (Thummel, 1996).

E75 is a complex ecdysone—inducible early gene that encodes three isoforms,
E75A, E758 and E75C as a result of alternative splicing (F eigl et al., 1989; Segraves and
Hogness, 1990). All three E75 isoforms are orphan receptors. E75A and E75C have
distinct A/B domains joined to the identical DBD and LBD. In contrast, E75C lacks the
first zinc finger of the DBD, suggesting that it is incapable of binding DNA. Consistent
with the Ashbumer model, several EcR/USP binding sites have been found in the E75
promoter (Talbot, 1993), and E75A protein is found to be bound to both early and late
gene loci (Hill et al., 1993). Moreover, recent studies have revealed that E75 is also
required for development of egg chambers. This function was determined in E75
germline mutants in which nurse cells degenerate (Buszczak et al., 1999).

There are some nuclear orphan receptors showing delayed early responses to
ecdysone. HR3 is one of those, and therefore is called an “early-late” gene in the
ecdysone gene regulatory hierarchy (Koelle et al., 1992). Drosophila HR3 (DHR3)
comprises a complex locus, encoding at least three different-sized classes of mRNAs.
The Drosophila “early-late” genes are induced by ecdysone, but differ from the early
genes in that they require ecdysone-induced protein for maximal levels of gene
expression (Stone and Thummel, 1993; Homer et al., 1995). Functional studies have

revealed that DHR3 represses the ecdysone induction of early genes and induces BFTZ-

15

 

subseu.

tempur-
l994l. .
in mid,”
in in 11;.

cud} 50ft;

hierarcha

 

 

Specific I‘
319903;") lL

ecdysone

ln

appear lo

'-
!

‘CCSPIOB
(Zcrth {If

The {Wm

:15th DH
ECR ESP
widens I

1995a: H

FU.

l

J
U? .

Fl, an orphan nuclear receptor that is essential for the appropriate response to the
subsequent prepupal pulse of ecdysone (White et al., 1997).

The orphan receptor Drosop/zila BFTZ-F 1 has been identified as providing a
temporal link between late larval and prepupal ecdysone responses (Woodard et al.,
1994). Drosophila BFTZ-Fl is expressed during the brief interval of low ecdysone titer
in midprepupae. Consistent with its expression profile, 20E represses BFTZ-F 1 transcript
in in vitro cultured salivary glands. Ectopic expression of BFTZ-Fl in flies enhances
ecdysone induction of the early genes; in BFTZ mutants the ecdysone-triggered genetic
hierarchy is severely attenuated. Thus, BFTZ-F 1 serves as a competence factor for stage-
specific transcription responses to ecdysone, and its expression in the midprepupae
appears to reset the system to allow reinduction of the early and late genetic responses to
ecdysone (Woodard et al., 1994; Broadus et al., I999).

In addition, it has been established that some insect nuclear orphan receptors
appear to be involved in negative regulation of ecdysone responses. Three orphan
receptors with this function have been identified, DHR38 (Sutherland et al., 1995), Svp
(Zelhof et al., 1995a), DHR78 (Fisk and Thummel, 1995, 1998; Zelhof et al., 1995b).
The formation of inactive heterodimers between DHR38 and USP, SVP and EcR, as well
as the DHR78 homodimer as a competitor, inhibit DNA binding and activation of
EcR/USP on the ecdysone responsive elements, providing different levels at which
ecdysone responses can be negatively controlled (Sutherland et al., 1995; Zelhof et al.,
1995a; Fisk and Thummel, 1995, 1998; Zelhofet al., 1995b).

Furthermore, recent studies on the molecular mechanism of EcR/USP action have

demonstrated that cofactors are involved in activation or repression of ecdysone gene

regulation. Two Drosophila co-repressors have been characterized to interact with
unliganded EcR/U SP: Alien (Dressel et'al., 1999) and SMRTER (silencing mediator for
RXR and TR-related ecdysteroid receptor interacting factor) (Tsai et al., 1999). These
proteins can interact directly with EcR, but not USP, in the absence of ligand. SMRTER
has a similar function but different structure relative to the vertebrate corepressors SMRT
and N-CoR. Antibody staining of salivary gland polytene chromosomes has displayed
that SMRTER co-localizes with EcR/USP, suggesting these proteins function together to
repress target gene expression. Like vertebrate SMRT and N-CoR, Drosophila SMRTER
mediates repression by interaction with Sin3A, a repressor to form a complex with
HDAC (Tsai et al., 1999). Drosophila Alien is another co-repressor for EcR/USP,
belonging to a novel class of co-repressors that are unrelated to SMRT and N-CoR.

Alien proteins are highly conserved in evolution, showing 90% identity between human
and fly. Vertebrate Alien interacts with TR only in the absence of hormone; Drosophila
Alien shows similar activity, mediating repression through the Sin3A pathway (Dressel et
aL,1999)

It was believed that there was no SRC family coactivator in Drosophila and yeast
(Freedman, 1999); however, a recent discovery has demonstrated that Drosophila TAI is
an EcR/USP coactivator, the homolog of A181 (SRC3) (Bai et al., 2000). Tai protein has
the conserved sequence and overall domain structure of a SRC family member,
containing bHLH, PAS and LXXLL motifs. TAI also shows a functional similarity with
SRC proteins. TAI directly interacts with EcR in a ligand dependent manner, and co-
localizes with EcR/USP in egg chambers and salivary gland polytene chromosomes.

Furthermore, TAI augments transcriptional activation by EcR/USP in cultured cells (Bai

tumult

int enci‘

  
   

Simildfi.

and th‘l'.

tertehr;

\
lnsect _\

et al., 2000). These findings provide the first evidence that nuclear receptor coactivators
exist in both vertebrates and invertebrate flies.

Based on findings to date, most insect nuclear receptors have vertebrate
homologs, exhibiting high levels of evolutionary conservation between vertebrates and
invertebrates in the nuclear receptor superfamily. These conservations include
similarities of protein structures and general molecular mechanisms of nuclear receptors

and their cofactors (Table 1).

Table 1. Conservation of nuclear receptors and cofactors between insects and

 

 

 

 

 

 

 

 

 

 

vertebrates

Insect Nuclear Receptor Vertebrate Homolog % DBD/LBD Identity
EcR BAR (FXR) 81/34
USP RXR 86/49
E75 Rev-Erb 79/35
HR3 RORa 75/20
FTZ-Fl SFl 88/33
HR78 TR2 74/26
HR38 NGFI-B 88/54
SVP COUP-TF 94/93
HNF4 HNF 4 90/67

 

Insect Co-activator Vertebrate Homolog Domain Similarity

 

 

 

TAI

AIBl

bHLH, PAS, LXXLL

 

Insect Co—repressor

Vertebrate Homolog

% Sequence identity

 

 

 

 

 

SMRTER SMRT/N-CoR High in SNOR,
SANT, ITS, LSD, and GSI
motifs

Alien Alien 90

 

18

 

 

Faster»

   
  
  

anauttg
female :
resuitzn:
llPP ,1 ;;

Raikhel.

WCIOL \
km?- PR
Control
H“;~"C‘L1Li
Immed‘.

“10$qu

NUCLEAR RECEPTORS IN MOSQUITO VITELLOGENESIS

Mosquitoes serve as vectors for many devastating human diseases (Collins and
Paskewitz, 1995; Butler et al., 1997; Beier, 1998) due to their ability to transmit
pathogens during blood feeding, which is required for their egg development. In the
anautogenous mosquitoes, vitellogenesis and egg maturation are initiated only after a
female mosquito ingests vertebrate blood. A blood meal triggers a hormonal cascade
resulting in production of 20E, the terminal signal, which activates yolk protein precursor
(YPP) genes in the fat body (Hagedom, 1983; 1985; Raikhel, 1992; Dhadialla and
Raikhel, 1994; Deitsch etal., 1995a; 1995b).

In the anautogenous mosquito Aedes aegypti, which is being studied as a model
vector, vitellogenesis proceeds through two periods (Fig. 4). The approximately 3-day-
long previtellogenic development of both the ovary and fat body proceeds under the
control of juvenile hormone (111) Ill (Gwartz and Spielman, 1973; F lannagan and
Hagedom, 1977). JH titers are high during the previtellogenic period and drop
immediately following the onset of vitellogenesis, which is initiated when the female
mosquito takes a blood meal (Shapiro et al., 1986). The fat body then produces YPP,
which are internalized by the developing oocytes (Raikhel, 1992). In addition to
vitellogenin (Vg), the major YPP, the mosquito fat body synthesizes three other YPPs,
vitellogenic carboxypeptidase (VCP), vitellogenic cathepsin B (VCB) (Cho et al., 1991;
1999), and lipophorin (Sun et al., 2000). YPP synthesis reaches its maximal levels at 24
hr post-blood meal (PBM) and then proceeds until 30-32 hr PBM, when it is rapidly

terminated (Raikhel, 1992).

 

 

20!

"I

ate 0e
‘:

\l'l‘

 

Fig. 4. S
mosquit
female.

 

519358un
ICCUT.\'(\1|
Sl'nthesi/
lB\1l, 15
351d ccd}
Hagt‘dun
1110,qu
‘s‘itcllogc;

1991). .\

E 1 A 88M :1 s 1,2 1L6 17,0214 2832 31643 433514 6109673

 

Hormones

10*: A W
m mnflfllflMMmL W

 

 

 

 

Late Genes
Vs

 

 

VC P

 

 

 

E 1 «3 33M ‘1 {5 12 1.6202512832364248 5JL6LQ'6 72

Days Hours

Fig. 4. Summary of events during the first cycle of vitellogenesis in the anautogenous
mosquito, Aedes aegypti. The previtellogenic period begins at eclosion (E) of the adult
female. During first ~2-3 days of post-eclosion life, the fat body becomes competent for
subsequent vitellogenesis. The female then enters a state-of-arrest; yolk protein
precursors, vitellogenin (Vg) and vitellogenic carboxypeptidase (VCP) are not
synthesized during this previtellogenic period. Only when the female takes a blood meal
(BM), is vitellogenesis initiated. Hormones: hormonal titers of juvenile hormone (JH)
and ecdysteroids (20E) in A. aegypti females (Hagedom et al., 1975; Shapiro and
Hagedom, 1982). Transcript profiles of yolk protein precursors (Late Genes) in the
mosquito fat body were determined by Northern blot and RT-PCR/Southem blot. Vg,
vitellogenin (Cho and Raikhel, 1992); VCP, vitellogenic carboxypeptidase (Cho et al.,
1991). Modified from Li et al., 2000.

20

 

rate of

C1331;

 

16\ 615 .;

role of.

shown 1

 

 

 

WP 11)

C}'C10ilc

Ecdysteroid titers in the female mosquito hemolymph positively correlate with the
rate of YPP synthesis in the fat body (Fig. 4). While the 20E titers are only slightly
elevated at 4 hr PBM, they begin to rise sharply at 6-8 hr PBM, and reach their maximum
levels at 18-24 hr PBM (Hagedom et al., 1975). Consistent with the initially proposed
role of 20E in activating vitellogenesis in anautogenous mosquitoes (Hagedom etal.,
1973; Hagedom and Fallon, 1973), experiments using an in vitro fat body culture have
shown that physiological doses of 20E (10'7 - 10'6 M) activate two YPP genes, Vg and
VCP (Deitsch el al., 1995a; 1995b). Utilization of the protein synthesis inhibitor,
cycloheximide (Chx), has demonstrated that the activation of YPP genes in the mosquito
fat body requires protein synthesis (Deitsch et al., 1995a). Thus, it is likely that a
regulatory cascade similar to that seen in Drosophila development mediates the action of
20E in the mosquito fat body.

The functional ecdysteroid receptor is a heterodimer of the ecdysone receptor
(EcR) and a retinoid X receptor (RXR) homolog, Ultraspiracle (USP) (Yao et al., 1992;
1993). In the mosquito A edes aegypti, one isoform of ecdysone receptor (AaEcR) and
two USP isoforms, AaUSP-A and AaUSP-B have been cloned (Cho et al., 1995;
Kapitskaya et al., 1996). The mosquito EcR/USP heterodimer has been shown to bind to
various EcREs to modulate ecdysone regulation of target genes (Wang et al., 1998). Two
classes of EcREs, inverted repeats (IR) and directed repeats (DR), with different spacers
are able to be bound by AaEcR/AaUSP. The IR-l (IR with a one nucleotide spacer) and
DR-4 (DR with a four nucleotide spacer) display optimal binding affinities in each class
of EcRE. Changing the half sites of the Drosop/zi/a natural hsp27 EcRE from imperfect

repeats to perfect repeats AGGTCA increases the DNA binding affinity of

to IRS.
I

pmmot.
may 6.
mm; :1; :

located .

the \‘g 1i

.

l

111 1058 0'

 

AaEcR/AaUSP 10 fold. The spacer-length requirement for DRs is less stringent relative
to IRs. These findings suggest that although EcREs identified in some Drosophilrz gene
promoters are predominantly le, some other native EcREs with different configurations
may exist in other insects (Wang et al., 1998). Indeed, most recent studies on the
mosquito Vg gene promoter have revealed that an imperfect DR-l EcRE (Vg EcRE) is
located at the 348 bp upstream of the Vg promoter, and AaEcR/AaUSP is able to bind to
the Vg EcRE. Moreover, deletion of the Vg EcRE from the Vg promoter-reporter results
in loss of activation by AaEcR/AaUSP in cultured cells (Martin et al., 2001).

AaUSP-A and AaUSP-B are the first two USP isoforms identified in insects
(Kapitskaya et al., 1996). AaUSP-A and AaUSP-B are identical proteins except at the N-
terminal regions of the A/B domains: the first 31 aa in AaUSP—A different from the first 6
aa in AaUSP-B. Both mosquito USP isoforms can form heterodimers with AaEcR and
bind to EcREs. In the fat body, AaUSP-A is expressed highly at the previtellogenic stage
and post-vitellogenic stage while AaUSP-B is expressed at relative high levels in the
vitellogenic stage (Kapitskaya et al., 1996). These data suggest that the two isoforms
may play different roles during mosquito vitellogenesis.

Previtellogenic, preparatory development is required for mosquito fat bodies to
attain both the responsiveness to 20E and competence for massive yolk protein synthesis
and secretion. Following this stage, mosquitoes enter a previtellogenic developmental
arrest, namely the state-of—arrest, which prevents the activation of YPP genes in the
previtellogenic female fat body prior to blood feeding (Raikhel and Lea, 1990). Recent
studies have demonstrated that AHR3 8, a mosquito homolog of DHR38 and vertebrate

NGFI-B, plays a crucial role in the state of arrest (Zhu et al., 2000). AHR38 is able to

22

“10]

I
disruptE
Ech_I

|
SW It"
heterot:
EcR L r
Ech5>

EXISA~

P081V11c

10 ecd} 5( I

Ii

("D

a
t

._ 1..
Lutdl‘ f

 

disrupt binding of EcR/USP to EcREs and inhibit transactivation of a reporter gene by
EcR/U SP in cultured cells. GST pulldown and co-immunoprecipitation experiments
have revealed that AaUSP, the obligatory heterodimer partner for AaEcR, forms a
heterodimer with AHR38 at the state-of-arrest, preventing the formation of a functional
EcR/U SP complex (Zhu et al., 2000). Consistent with these findings, binding of the
EcR/USP complex to EcREs in the fat body nuclear extracts is barely detectable by
EMSAs at the state-of-arrest; in contrast, a large amount of binding is visible in the 1hr
PBM nuclear extracts (Miura et al., 1999).

Another nuclear orphan receptor AaSVP, a homolog of Drosophila SVP and
vertebrate COUP-TF, has been cloned recently in our lab. AaSVP is highly conserved
with its homologs in term of amino acid sequence. AaSVP interacts with AaUSP in the
post vitellogenic period, suggesting its involvement in terminating vitellogenic responses
to ecdysone in mosquito fat bodies (Miura, Zhu, and Raikhel, unpublished data).

Finally, it is known that the mosquito Vg gene is activated in a tissue specific
manner. The nuclear orphan receptor HNF-4, a member of the Hepatocyte Nuclear
Factor 4 (HNF-4) family, is possibly implicated in controlling tissue specificity. Three
HN F -4 isoforms are expressed in mosquito fat bodies with different developmental
profiles during vitellogenesis, suggesting distinct functions for each isoform at different

stages of vitellogenesis (Kapitskaya et al., 1998).

RATIONALE FOR CURRENT STUDIES
One of the main research interests in Dr. Raikhel’s lab is to understand the

molecular mechanism underlying mosquito vitellogenesis. In his lab, several yolk

23

 

 

 

sex-5.:
which
prox id .
C001T3>

dCI'EiOp

 

 

 

EESCCIIK

protein precursor genes have been cloned, which are expressed in a stage-, tissue- and
sex-specific manner. These genes are activated in female fat bodies by the hormone 20E,
which is triggered by a blood meal. This system of stringently controlled gene regulation
provides an ideal model to study ecdysone gene regulation during insect reproduction. In
contrast to detailed studies on the ecdysone regulatory cascade in Drosophila
development and metamorphosis, however, the ecdysone gene regulatory hierarchy in
insect reproduction was poorly understood.

In mosquitoes, one isoform of EcR and two isoforms of USP have been cloned in
our lab, but some primary questions still need to be addressed. Three EcR isoforms have
been identified in Drosophila; similarly, two or three isoforms have been found in many
other insects. It has been demonstrated that the different EcR isoforms are expressed in
different tissues and mediate functionally different ecdysone responses. Therefore, in
mosquitoes, one of the key questions was whether any other EcR isoforms exist, and if
they do, whether they are involved in vitellogenesis. Secondly, two USP isoforms have
been identified in mosquitoes and they are expressed differentially during vitellogenesis.
AaUSP-A mRNAs in fat bodies are at high levels during previtellogenic and
postvitellogenic periods, but at low levels during the vitellogenic period. In contrast,
AaUSP-B is expressed at high levels during the vitellogenic stage. Correlated with
AaUSP mRN A levels, the two hormones, JHIII and 20E, display opposite titer dynamics
during pre- and vitellogenic stages. JHIII titers are high during the previtellogenic stage
and decline immediately following the onset of vitellogenesis. 20E remains at low levels
during the previtellogenic stage and rises sharply after a blood meal. A recent study

suggests that USP may be a candidate receptor for JH (Jones and Sharp, 1997).

24

Therefore, other more critical questions addressed in the first part of my research were
how the mosquito USP gene(s) are regulated during vitellogenesis and whether AaUSP-A
and AaUSP-B isoform expressions are under the control of JHIII or 20E.

The ecdysone gene regulatory hierarchy has been well documented in Drosophila
metamorphosis. This regulatory cascade contains many transcription factors, most of
which are nuclear orphan receptors. In mosquitoes, utilization of the protein synthesis
inhibitor, cycloheximide (Chx), has demonstrated that the activation of YPP genes in the
mosquito fat body requires protein synthesis (Deitsch et al., 19953). Thus, YPP genes
serve as the ecdysone regulated late target genes. It is likely that a regulatory cascade
similar to that seen in Drosophila development mediates the action of 20E in the
mosquito fat body. Among the components of the ecdysone gene regulatory hierarchy in
Drosophila, the early gene E75 and the early-late gene HR3 are two key ones. Therefore,
in the second and third parts of my studies, mosquito homologs of E75 and HR3 were
cloned and characterized as a beginning to elucidating the ecdysone regulation cascade in
mosquito vitellogenesis.

A preparatory developmental previtellogenic period is required for the mosquito
fat body to attain competence for ecdysone responsiveness (Sappington and Raikhel,
1999). F lannagan and Hagedom (1977) have demonstrated that the acquisition of
competence for the ecdysone response in the mosquito fat body is under the control of
juvenile hormone 111. However, the molecular mechanism underlying the change in
sensitivity to ecdysone in the mosquito fat body remains obscure. In the fourth part of
my research, I addressed the question of whether competence for the ecdysone response

is controlled at the transcriptional level or translational level. Furthermore, the mosquito

25

 

‘3 1“ 1" '.

mosq-

COHIIL'

 

111013.111.

 

DIE/Sop

 

homolog of BFTZ-Fl, a competence factor in the Drosophila ecdysone regulatory
network, was cloned. I then designed experiments to address the question of what role
mosquito FTZ-Fl plays during vitellogenesis, and whether the molecular mechanisms
controlling the stage-specific ecdysone responses are conserved during Drosophila
metamorphosis and mosquito vitellogenesis.
Finally, in the fifth part of my research, I cloned the mosquito homolog of

Drosophila HR 78, which is involved in negative regulation of ecdysone responses. I
designed experiments and answered the questions of whether mosquito HR78 represses

the DNA binding and transactivation functions of AaEcR/AaUSP, and what role it plays

during vitellogenesis.

26

I
.
.

g

‘4‘

6
o
a.

'_3”
’T
-vl
‘.
.3
.u
. .1
1
-..,
7.!
'.‘C
1..
_':_t
.,
,‘4.
."
-1.
-
u‘r
. v
t
..
.1
O
«-

 

oft

 

t.
11031 r951

 

 

Chapter 2*
Characterization of Two Isoforms

of the Mosquito Ecdysone Receptor (EcR) and Ultraspiracle (USP)

*Most results have originally been published in Dev.Biol.
Ref: Wang, s.-E.,Lj,_g,”,2hu,1.,Miura, K., Miksicek, R. 1., Raikhel, A. s. (2000).
Dev. Biol. 218, 99-113.

# Underlined two authors contributed equally to this work.

27

 

 

 

 

flprcss'
EcR L's
them 11
1131154“
”1938 r,

ABSTRACT

Ultraspiracle (USP), the insect homologue of the vertebrate retinoid X receptor
(RXR) is an obligatory dimerization partner for the ecdysteroid receptor (EcR). Two
USP isoforms, USP-A and USP—B, with distinct N-terrnini occur in the mosquito, Aedes
aegypti. In the fat body and ovary, USP-A mRNA is highly expressed during the pre—
and late-vitellogenic stages, corresponding to a period of low ecdysteroid titer, while
USP-B mRNA exhibits its highest levels during the vitellogenic period, correlating with a
high ecdysteroid titer. Remarkably, 20-hydroxyecdysone (20E) has opposite effects on
USP isoform transcripts in in vitro fat body culture. This steroid hormone up-regulates
USP-B transcription and its presence is required to sustain a high level of USP-B
expression. In contrast, 20E inhibits activation of USP-A transcription. Although
EcR/USP-A recognizes the same ecdysteroid responsive elements, EcR/USP-B binds
them with an affinity two—fold higher than that of EcR/USP-A. Likewise, EcR/USP-B
transactivates a reporter gene in CV-l cells two-fold more strongly than EcR/USP-A.
These results suggest that USP-B functions as a major heterodimerization partner for EcR

during the vitellogenic response to 20E in the mosquito.

28

was funk...

domai"

«r r)
tOlilLr"

 

WHEEL.

members 1

 

INTRODUCTION

In insects, as in higher metazoans, steroid-regulated gene expression is mediated
by a family of intracellular receptors, which belong to the superfamily of nuclear
receptors. These receptors have a highly conserved structure, including a DNA—binding
domain (DBD) and a multifunctional ligand-binding domain (LBD). In addition to
conferring ligand binding specificity, the LBD is responsible for dimerization,
transactivation, co-activator/co—repressor interactions, and in the case of vertebrate
steroid receptors, heat shock protein binding (O’Malley and Conneely, 1992;
Mangelsdorf et al., 1995; Onate et al., 1995). Nuclear receptors are often encoded by
several genes and each gene may produce several subtypes called isoforms due to
utilization of alternative promoters, splicing or polyadenylation signals (Keightley, 1998).
The major differences between isoforms appear to be centered in the N-terminal A/B
domain. This least conserved region of nuclear receptors contains the activation function
AF-l, which has been demonstrated to provide cell- and promoter-specific transactivation
properties (Kastner et al., 1990; Kuiper et al., 1996; Paech et al., 1997).

The major insect steroid hormones, ecdysteroids, are involved in regulating
development, molting, metamorphosis and reproduction (Riddiford, 1993b; Raikhel,
1992). DNA binding, ligand binding and transfection assays indicate that the functional
ecdysteroid receptor complex consists of two subunits, the ecdysteroid receptor itself
(EcR) and the retinoid X receptor homologue Ultraspiracle (USP), both of which are
members of the nuclear receptor superfamily (Koelle et al., 1991; Yao et al., 1992; 1993;
Thomas et al., 1993). The role of USP as a partner in the functional ecdysteroid receptor

was further substantiated by mutational analysis. Fly usp mutants show multiple defects

29

 

 

EUR 15
meni:
199'L;
Um I);

dmnmr

 

spficzng

{Esme-a

”c301“

31.199.

dfgcncr

DWI/1'),
1‘

in ecdysteroid-regulated gene expression at the larval-prepupal transition, indicating that
USP is required for ecdysteroid receptor activity in vivo (Hall and Thummel, 1998).

The pleiotropic effects of ecdysteroids are reflected by the existence of multiple
EcR isoforms. EcR isoforms were first identified in Drosophila (Talbot et al., 1993) and
then in other insects including Bombyx mori (Swever et al., 1995; Kamimura et al., 1996;
1997) and Manduca sexta (Fujiwara et al., 1995; Jindra et al., 1996a). The three
Drosophila EcR isoforms (EcR-A, EcR-Bl and EcR-B2) differ in the N-terminal A/B
domain, as a result of utilization of different transcription start sites and alternative
splicing. Distinct functions of DmEcR isoforms were first evidenced by their different
tissue- and stage-specific expression profiles (Talbot et al., 1993). In the Drosophila
nervous system, EcR-A and EcR-Bl isoforms have clearly distinct functions (Truman et
al., 1994). A high level of EcR-A expression in the ventral CNS correlates with rapid
degeneration of neurons after adult emergence (Robinow et al., 1993). Moreover,
Drosophila EcR mutations confirm that functional differences exist among receptor
isoforms (Bender et al., 1997). In particular, the EcR-Bl mutants can not survive
through metamorphosis, and EcR-Bl and EcR-B2 mutants fail to prune back larval-
specific dendrites to initiate larval neuron remodeling (Schubiger et al., 1998).

Unlike Drosophila USP for which a single form of mRNA has been identified
(Shea et al., 1990; Henrich et al., 1990, Oro et al., 1990), two USP isoforms are reported
in the mosquito Aedes aegypti (USP-A and USP-B, Kapitskaya et al., 1996) and in the
moth, M. sexta (USP-l and USP-2, Jindra et al., 1997). In both insects, USP isoforms
exhibit distinct N-terminal A/B domains similar to isoform variation in the EcR isoforms.

In A. aegypti, the N-terminal 31 aa in USP—A are different from the first 6 aa in USP-B.

30

 

 

 

 

50kt?

11101111 '

Bonn?
111 Dit-
Houcx

[SPtsr

1116! \1};

 

RRLS
P1017101:
31.1991
distinCt
\il'c‘llgg‘
€\pressi

Comm“:

Moreover, their 5’ untranslated regions (UTRs) are different, suggesting that these
isoforms are most likely derived from the same gene via utilization of alternative
promoters (Kapitskaya, et al., 1996).

In sharp contrast to the large body of information defining the roles of EcR
isoforms, little is known about the properties of USP isoforms. The lack of USP isoforms
in Drosophila makes it difficult to perform USP isoform-specific mutant analysis.
However, identification of two forms of USP from M. sexta has permitted the study of
USP isoform profiles during development (Jindra, 1997; Asahina et al., 1997; Hiruma et
al., 1999; Lan et al., 1999). The level of USP-2 mRNA correlates with the ecdysteroid
titer whereas USP-1 transcription is complementary to that of USP-2. However, it is the
EcR/USP-l complex, not the complex containing EcR/USP-2, that activates the MHR-3
promoter although MHR-3 mRNA appears following a rise of ecdysteroid levels (Lan et
al., 1999). Likewise, the two Aedes USP isoforms enabled us to investigate the roles of
distinct USP isoforms in the adult insect during reproduction. Indeed, mosquito
vitellogenesis has provided a perfect model to study ecdysteroid-regulated gene
expression during reproduction, as the endocrine release of ecdysteroids is tightly
controlled by a blood meal (Fig. l-A). The DNA binding properties of the Aedes
EcR/USP-B complex have been studied in detail: it binds to various ecdysteroid
responsive elements (EcRE) with the AGGTCA consensus half site arranged either as
direct repeats or inverted repeats. A l-bp spacer is optimal for inverted repeats, while a 4-
bp spacer is optimal for direct repeats. DNA binding activity is correlated with
transactivation as demonstrated by transfection assays in CV-l cells (Wang et al., 1998).

Only the heterodimeric complex containing the mosquito EcR and one of the mosquito

31

 

 

 

 

Cal

USP isoforms (either USP-A or USP-B) represents a functional receptor capable of
binding the EcRE or the ecdysteroid ligand (Kapitskaya et al., 1996).

Here, we report the developmental profiles of the mosquito USP isoforms, their
differential responses to 20—hydroxyecdysone (20E) as well as functional differences with
respect to their DNA binding and transactivation abilities. Taken together, these findings
suggest that USP isoforms clearly perform distinct functions during mosquito
vitellogenesis. Furthermore, these data suggest that it is USP-B that serves as a major

partner for EcR during the mosquito vitellogenic response to ecdysteroid.

MATERIALS AND METHODS

Animals

Mosquitoes, A. aegypti, were reared as described by Hays and Raikhel (1990).
Vitellogenesis was initiated in 3-5 day old adult female mosquitoes by blood feeding

them on rats.

RNA Isolation and RT -PCR/Southern Blot Analysis

Fat bodies and ovaries were dissected from female mosquitoes at different time
points ranging from 0- to 5-day after eclosion at the previtellogenic stage, or from
vitellogenic females ranging from 1 to 60 hours post-blood-meal (PBM). The temporal
profiles of transcript abundance for the mosquito EcR, USP-A and USP-B in the
mosquito fat body and ovary were examined using RT-PCR followed by Southern
blotting with specific radioactive probes. Total RNA was prepared from fat body

throughout the first vitellogenic cycle, using the guanidine isothiocyanate method as

32

described previously (Bose and Raikhel, 1988) with the modification that all isopropanol
precipitation steps were done without low-temperature incubation to avoid co-

precipitation of glycogen and salts. Alternatively, total RNA was extracted with TRIZOL
Reagent (Gibco-BRL). RNA yields were determined spectrophotometrically, absorbency

ratios A260/A280 and A260/A230 of RNA preparations were consistently above 1.7 and
2.0, respectively. Five microgram aliquots of each RNA preparation were reverse-
transcribed by the SuperscriptII reverse transcriptase (Gibco-BRL) using random
hexamer primers (Promega) in a reaction volume of 20 ul. The reverse transcription

product was diluted to 40 p] with TE buffer (10 mM Tris-Cl, 1 mM EDTA, pH 8.0) and

stored at -200C as a cDNA pool until use. For developmental profile analyses, 0.025 fat
body-equivalents of cDNA pools were used as PCR templates in the case of mosquito fat
body tissues whereas an equal amount of total RNA was used for reverse transcription in
the case of ovarian tissues. A 522bp USP—A specific fragment was amplified with the
primer pair: forward, 5’-TCATATCGTTCCGGAGATGTGG-3’ and reverse, 5’-
CCAATCCTGCCAGAGGTAGTG-3’. A 400bp USP-B specific fragment was amplified
with the primer pair, forward, 5’-CTTCTCACAAGAGGTGCTGAGO-3’and reverse, 5’-
TGGTATCCAACTGGAACTTGCG-3’. A 314bp AaEcR specific fragment was
amplified with the primer pair, forward, 5’-GAGGAAGATCAACATGACGTGC-3’ and
reverse, 5’-ACCGTGAGGGAGAACATCTGC-3’. A 463-bp VCP (vitellogenic
carboxypeptidase, Cho et al, 1991) specific fragment was amplified with the primer pair,
forward, 5’-AGCGCCCATTCTTGTTTGG-3’ and reverse,

S’-CAGCTCATACAGGTATTCTCC-3 ’.

33

tor 3

preps:
[SP-.1.

t‘ragm:

 

lllCorp.

550C lf‘

 

 

Thermal cycling conditions were as follows: the reaction was incubated at 94°C

for 3 min followed by 17 cycles of 94°C for 30 sec, 600C for 30 sec and 720C for 40 sec.
As a reference to the developmental changes of the fat body, the same cDNA was
subjected to 10 or 15 cycles of RT-PCR followed by Southern blotting with a primer pair
specific to the mosquito VCP. After PCR amplification, 10 pl each out of the total 25p]
reaction was fractionated on a 2% agarose gel, and transferred onto Hybond N+
membrane (Amersham) under alkaline conditions. Radioactive cDNA probes were
prepared from 15 ng each of the PCR fragments generated by the primer pairs of EcR,
USP-A, USP-B, or VCP with the corresponding plasmids as templates. These amplified

fragments were labeled by a random-primers DNA labeling system (Gibco BRL) to
incorporate [a-32P]-dATP (NEN). The membrane was hybridized with the probe at

650C in a solution of 5 x SSC, 5 x Denhardt' solution, 0.5%SDS, 25mM NaHzPO4,

25mM NazHPO4, 200 ug/ml of denatured salmon sperm DNA (Sigma) following

prehybridization. The membrane was washed twice at 650C in 2 x SSC, 0.1%SDS for 15
min, twice in 0.1 x SSC, 0.1%SDS for 15 min, then autoradiographed. RT-
PCR/Southem blotting of the total RNA preparations without the reverse transcriptase
did not result in any appreciable signals, indicating that contamination of RNA
preparations with genomic DNA fragments was virtually negligible (not shown). The
radioactive blots were quantified with a phosphorimager (Molecular Dynamics).

Examination of the EcR isoforms was performed in the same method as that in the
USP isoforms. When conducting this experiment, AaEcR is named as AaEcR-B, and the
new isoform is named AaEcR-A. The primers for AaEcR-A are: Forward, 5’-

TGACGGCCATTCCGGCTTC; Reverse, GAGACGAGTAGCCATTGGAC.

34

In vitro Fat Body Culture

To examine the transcriptional inducibility of mosquito EcR, USP-A and USP-B
by ecdysteroid, the abdominal walls with adhering fat body (hereafter referred to as the
fat body) were dissected from 3~5-day-old previtellogenic females and incubated in an
organ culture system as previously described (Deitsch et al., 1995a; Raikhel et al., 1997).
The fat bodies were incubated either with culture media in the absence of ecdysteroid or
in the presence of 10’6 M 20E (unless otherwise indicated). At every 4-hour interval after
incubation, 9 fat bodies of each time point were collected and immediately stored in
liquid nitrogen until RNA isolation. For the hormone withdrawal experiments, fat bodies
were washed 3 times with culture media after incubating with 10'6 M 20B for 4 hours and
then further incubated in hormone-free media. Total RNA extraction and RT-
PCR/Southem blot analyses were performed as described above.

To confirm that the transcriptional response was due to the fat body and not the
epidermis of the abdominal wall, the epidermis-free fat bodies were obtained by gently
vortexing the dissected abdominal walls. The epidermis-free fat body was incubated in
culture medium supplemented with 10'6 M 20E for 3-hr. The fat body adhering to the
abdominal wall was incubated under the same conditions. Similar amounts of fat body
tissue from both preparations were examined for the presence of two ZOE-induced genes,
an early gene E 75 and the late gene VCP (Deitsch et al., 1995a; Pierceall et al., 1999).
The levels of induction of both genes were similar in fat bodies of both preparations.
However, the epidermis-free fat body could be kept alive for only a short incubation time

(data not shown).

35

1T3115.

with .

C0110;

1111th

desert

 

 

 

gencrm
acre ex
6131, 1
7.53, 5

mcmbf‘;

monocj

To test the effect of the protein synthesis inhibitor cycloheximide (Chx) on
transcription, dissected 3-5 day old female mosquito fat bodies were pretreated for l-hr
with culture media containing lmM Chx, then further incubated with the same
concentration of Chx either without hormone or with 10'6 M 20E. Chx/hormone
withdrawal experiments were conducted as for the hormone withdrawal experiment

described above except that lmM Chx was added to all culture media.

Antibodies and Western Blot Analysis

The monoclonal antibody ABll against DmUSP (Christianson et al., 1992) was a
generous gift from Dr. F .C. Kafatos (EMBL, Heidelberg, Germany). Nuclear proteins
were extracted from the fat bodies of female mosquitoes as previously described (Miura
et al., 1999). Protein extracts corresponding to 50 mosquito equivalents were resolved on
7.5% SDS-polyacrylamide gels, followed by electro-blotting to polyvinylidene difluoride
membranes (Amersham). These membranes were probed with the anti-DmUSP
monoclonal antibody (1:200 dilution) and then incubated with a secondary anti-mouse
lgG antibodies (1:5,000 dilution) conjugated to horseradish peroxidase (Cappel).
Immune complexes were visualized by addition of SuperSignal Ultra Chemiluminescent

Substrate (Pierce) and autoradiography.

In vitro Protein Synthesis and Electrophoretie Mobility Shift Assay (EMSA)
Nuclear receptors were synthesized by the coupled in vitro transcription-
translation (TNT) system (Promega). The EcoRI fragment of mosquito USP-A cDNA in

pBluescript was subcloned into the EcoRI site of pGEM7Z (Promega). The same vector

36

W35 1.

 

M197;

1996.

which

 

11131111?

£00511;

ecdysr

 

 

Sf'nthcs
Pill} ltCr
measm

133.1113“;

Smith:

was used to subclone DmEcR and DmUSP cDNA fragments (Wang, et al., 1998)
whereas mosquito EcR and USP-B cDNAs were cloned into pGEM3Z (Kapitskaya et al.,
1996). All these expression vectors utilized the SP6 promoter to synthesize proteins,
which were first confirmed by 35S—Methionine labeling and SDS-PAGE following the
manufacturer’s instruction. EMSA and measurement of equilibrium dissociation
constants were conducted as described before (Wang et al., 1998). In brief, four
ecdysteroid response elements (EcREs):

IR,” -1 (agagacaagGGTTCAaTGCACTtgtccaat),

IRper-l (agcttcaaAGGTCAaTGACCTtgtccatcg),

DR-4 (aagcgaaAGGTCAaggaAGGTCAggaaaat)

and DR12 (aagcgaaAGGTCAagaggccaaagaAGGTCAggaaaat) were end-labeled with y’
3'ZP-ATP (DuPont NEN) with T4 polynucleotide kinase (Gibco BRL). Unless indicated,
50 fmol of labeled EcRE was incubated with 1 pl each of the indicated in vitro
synthesized nuclear receptor proteins. Binding reactions were resolved using a 6% native
polyacrylamide gel, which was then vacuum dried at 800C and exposed to X-ray film. To
measure the equilibrium dissociation constants, bound and free EcRE probes were
quantified with a phosphorimager, permitting the construction of saturation curves and

Scatchard analysis (Wang et al., 1998).

Construction of Reporter and Expression Plasmids and Cell T ransaetivation
Assay
Transactivation assays were conducted with the green African monkey kidney cell

line CV-l (ATCC). The EcoRI fragment of USP-A was subcloned into the mammalian

37

 

expr:
l'SP-i
the re:
)ll\'-
81 al..
11611 p'
follow

1 1 pt) 1‘."

 

ether-t
cells 11.
galdt‘lt ~
am It}.
C11 a,

teen 1

:1}
M,”

and L‘SP-

pnmtrs f

I
l I

 

expression vector pCDNA3. l/Zeo (+), which was also used to express mosquito EcR and
USP-B proteins (Wang et al., 1998). CMV-LacZ was used as a co-reporter to normalize
the reporter gene activities of the two reporter plasmids AMTV-SxIRhSP-l—CAT and A
MTV-3xDR4-CAT. Transfection assays were performed as described previously (Wang
et al., 1998). In brief, CV-l cells were maintained in DMEM with 10% calf serum. Six-
well plates were seeded at a density of 2x105 cells/well the day before transfection.
Following the manufacturer’s instruction, cells were transfected for 5 hours with
LipofectAMlNETM (Gibco BRL) and then added fresh medium together with either
ethanol vehicle or 1 pM Muristerone A. After a hormone treatment of 36 or 48 hours,
cells were harvested for CAT (chloramphenicol acetyl transferase) activity and [3-
galactosidase activity. CAT activity was normalized with B-galactosidase activity. CAT
activity was expressed as the percent of substrate converted to a product, and lunit of
CAT activity was defined as 1% of the chloramphenicol substrate converted to its

acetylated forms.

RESULTS

Tissue- and Stage-Specific mRNA Expression of the Mosquito USP Isoforms

It has been previously reported that in vitellogenic tissues, the fat body and ovary,
there are several transcripts corresponding to two Aedes aegypti USP isoforms (USP-A
and USP-B) (Kapitskaya et al., 1996). Here, more detailed analysis was undertaken for
the developmental changes in transcripts of the two USP isoforms in the fat body and
ovary during vitellogenesis. RT-PCR was performed utilizing USP isoform-specific

primers followed by Southem-Blot hybridization with isoform-specific probes (Fig. 1).

38

 

 

Fig.1. Expression profiles of USP and EcR mRN A during vitellogenesis. (A)
Hormone and Yolk protein mRNA profiles. The whole body ecdysteroid titer at various
time points (hours) in post vitellogenic female mosquito is converted to 20E equivalent
(Hagedom et al., 1975). JH (juvenile hormone) titer is from Shapiro et al., 1986. The
mRNA expression profiles of yolk protein Vg (Vitellogenin) and VCP (Vitellogenic
carboxypeptidase) are from Deitsch et al., 1995b. The titers are measured at various
days’ post-eclosion in previtellogenic mosquito and various hours’ post—blood meal in
vitellogenic mosquitoes. Other abbreviations: E, eclosion; BM: blood meal. (B) USP and
EcR mRNA profiles in the fat body. Total RNA was isolated from fat body dissected
from previtellogenic female mosquitoes at O-ld, 1-2d, or 3-5d after eclosion (columns 1-
3), or from vitellogenic mosquitoes 1h, 3h, 4h, 6h, 15h, 24h, 36h or 48h PBM (columns
4-1 1). (C). USP and EcR profiles in the ovary. Total RNA was isolated from ovaries
dissected from previtellogenic female mosquitoes, O-ld, 2—3d or 4-5d (columns 1-3) or
vitellogenic mosquitoes 1h, 6h, 12h, 24h, 36h, or 48h PBM (column 4-9). RNA samples
from each time point was subjected to reverse transcription and PCR amplification using
either one pair of primers specific to AaEcR or two pairs of primers specific to USP-A
and USP-B, respectively. Reverse transcribed products from equal number of fat bodies
(fat body profile) or equal amount 5 pg total RNA (ovary profile) were used for PCR
amplification. The PCR products were resolved by agarose gel electrophoresis, followed
by Southern blotting and autoradiography.

39

 

 

 

A Days llours
l’ost-liclosion Post-Blood-Meal
1% l ‘ 3 11M 4 X 12 16 20 24 28 32 36 42 4X 54 (IU OI) 72

b

L

Hormones
2015
III

Yolk Protein mRNA
Vs
VC P

 

 

 

 
 
 

 

 

 

 

 

B Day: Post-Ecloslon Hours Post-blood Meal
0-11-2 3-511 3 4 6 15 2436 48

3::3 ”23-3.1-3”

 

 

ECR p ---—-----m

 

 

Lane1234567891011

C Days Post-Ecloslon Hours Post-blood Meal
0.1 2-3 4-5 11 6 12 24 36 4a

USP-Ab ‘. C p I-

 

 

-u.

USP-B> v~m~-~..., g.

 

 

 

 

EcR> w..-“

Lane12 3 4 5 6 7 8 9

 

40

L A A l A I. I L 1 J 4 #1 .L.-_,-»__,

 

Expr’
expel
mat;
l
for t“!

SSIIlC

blot (1'

(its. 1
5

PI? 'lIL’

 

 

 

 

mmed

iFIS- 11

Expression of mosquito EcR was monitored in the same tissues as well. To minimize the
experimental variability between PCR reactions, pooled RNA samples from fat body or
ovary tissues were used in addition to low cycle numbers (17 cycles) for amplification.
For control purposes, Northern blot analysis performed in parallel on samples from the
same fat body or ovary tissues revealed identical mRNA profiles as RT-PCR/Southem
blot analysis (data not shown).

In the fat body, USP-A mRNA was highly expressed at 0-1 day after emergence
(Fig. 18, lane 1), however its levels gradually decreased over the next 5 days of the
previtellogenic period (Fig 18, lanes 2 and 3). The level of USP-A mRNA declined even
more dramatically immediately after a blood meal; it dropped at 3-hr PBM (Fig. 1B, lane
5) and remained low during the entire active vitellogenic period until 24-hr PBM (Fig 18,
lanes 6-9). The level of USP-A mRNA increased again in the fat body at the termination
of the vitellogenic period, 36-hr PBM (Fig. 18, lane 10). The expression pattern of USP-
B in the fat body was significantly different from that of USP-A. The level of the USP-B
transcript was relatively constant throughout the pre- and early vitellogenic periods, but
exhibited an obvious enhancement by 24- hr PBM (Fig 1B, lane 9). However, this
transcript declined at termination of vitellogenesis at 36-hr PBM (Fig. 13, lane 10).

The USP-A mRNA was abundant in ovaries of newly emerged female mosquitoes
(Fig 1C, lane 1). Its level increased during the second day of previtellogenic development
(Fig. 1C, lane 2), but decreased slightly in 4-5 day old previtellogenic females (Fig. 1C,
lane 3). As in the fat body, a blood meal resulted in a dramatic drop in the USP-A
mRNA level (Fig. 1C, lane 4) and the decline persisted throughout vitellogenesis (Fig.

1C, lanes 5-7). At 36-hr PBM, a small increase of USP-A mRNA was observed (Fig. 1C,

41

lane 8). Ovaries of newly emerged female mosquitoes also contained a detectable level
of USP-B mRNA (Fig. 1C, lane 1) which was enhanced in 2- to 3-day old ovaries (Fig
1C, lanes 2-3). A blood meal triggered a considerable increase in the USP-B mRNA
within one hour, and the level of USP-B transcript remained high during the first half of
the vitellogenic period (Fig.1C, lanes 4-5). Thus, our results clearly demonstrated that in
the adult mosquito female, the expression patterns of the two USP isoforms during the
first vitellogenic cycle were dramatically different in both the fat body and ovaries.
Levels of the EcR transcript were monitored in the same tissues and times as
those of the USP isoforms. The levels of EcR transcript were relatively constant in the
female fat body, showing elevations at the beginning of the pre-vitellogenic phase (Fig.
1B, lane 1), early vitellogenesis (Fig. 1B, lane 5) and post-vitellogenesis (Fig. 1B, lane
10). In the ovary, the EcR transcript level also exhibited three increases at approximately

similar times as in the fat body (Fig. 1C).

In Vitro E fleet of 2041 ydroxyecdysone on USP-A and USP-B Transcripts

The dramatic differences in the expression patterns of mosquito USP isoform
transcripts observed in vitellogenic tissues suggested that mosquito USP isoforms were
differentially regulated during vitellogenesis. I began studying the control of their
expression by examining the effect of 20E on these USP transcripts. These experiments
were conducted using an in vitro fat body culture system, which has been well-
established and is more suitable for these studies than small mosquito ovaries (Raikhel et
al., 1997). In addition, unlike ovaries containing several cell types, the fat body is a

relatively homogeneous tissue consisting of predominantly one cell type, trophocytes

42

 

 

mRNA (Fold-Induction. EcR 8- USP-B)

 

 

 

 

5 120
DEcR
4.5 T
2 4 'IVCP a?
8 1 0
.5 3.5 >_
g A... 80 g
u: 3 .§
.5 E
3 25 60 b
S 3
C 2 C
5 l t
L9. 1.5 4° 9.
§ g
a: 1 I!
E , 20 E
0.5
0 - L- 0

 

 

 

 

 

 

 

 

 

 

 

 

10-8 10-7 10-6 10-5
205 (M)

Fig. 2. Dose response of USP-B, EcR and. VCP to 20E in in vitro fat body organ
culture. Fat bodies dissected from nine previtellogenic female mosquitoes were cultured
for four hours in vitro either in the absence of hormone or in the presence of increasing
concentrations of 20E ranging from 10’9M-10'S M. Total RNA was isolated from these fat
bodies and subjected to RT-PCR with specific primers to EcR, USP-B and VCP,
respectively. The PCR products were resolved by the agarose gel electrophoresis,
followed by Southern blotting and autoradiography, and were quantified by the
phosphorimaging analysis. The levels of EcR and USP-B mRNA from fat bodies
cultured without hormone were defined as 1 unit. The maximum level of VCP mRNA
from fat bodies incubated with 10’6M 20E was defined as 100, because VCP mRNA level
was below detection in the previtellogenic fat bodies before the hormone treatment.

43

inctix

incub

sou:

lhctf

contr.

  
   
    
  

isoforr‘
tuhurc
afierin
express:
COEIhlLt
1311:11

Squares)
1

 

(Raikhel, 1992), making interpretation of results easier. Abdominal walls with adhering
fat bodies, isolated from previtellogenic female mosquitoes (3-5 days after eclosion) were
incubated in culture medium either in the absence of hormone or in the presence of
increasing amounts of 20E ranging from 10'9 M-lO'5 M (Fig. 2). After a 4-hour
incubation, these fat bodies were subjected to RNA isolation, RT-PCR and Southern
blotting analyses. USP-B and EcR were activated by 20E in a dose-dependent manner.
The mRNA levels of both EcR and USP-B were maximal at 10'6M 20E (Fig. 2). In
contrast, USP-A mRNA was not induced by any concentration of 20E (data not shown).
As a positive control, I monitored the 20E dose response of vitellogenic carboxypeptidase
(VCP) mRNA, one of the late target genes in the mosquito fat body (Cho et al., 1991;
Deitsch et al., 1995a,b). This ecdysteroid-responsive gene exhibited a peak induction
level at 10'6 M 20E (Fig. 2).

Next, I investigated the time course of 20E effects on the expression of USP
isoforms. In this experiment, dissected previtellogenic fat bodies were incubated in
culture media in the presence or absence of 10’6M 20E (Fig. 3, A-D). At 4-hour intervals
after incubation, USP transcripts were analyzed as described above. The effect of 20E on
expression of the two USP isoforms in cultured previtellogenic fat bodies was opposite:
continuous presence of 20E in the culture medium failed to induce the expression of
USP-A (Fig. 3A, open squires), while it activated USP-B expression (Fig. 3B, open
squares). USP-B transcript levels steadily increased with 5-, 7- and 10-fold enhancement
at 8-, 12- and 16-hr of incubation, respectively (Fig. 3B, open squares). When fat bodies
were incubated for four hours in the presence of 20B and then transferred into a hormone-

free medium, the level of the USP-A transcript rose dramatically (Fig. 3A, open

44

 

bUd.

triangles). In contrast, the USP-B transcript declined to its background level after an
initial rise at 4-hr (Fig. 38, open triangles).

The effect of 20E on the expression of AaEcR in in vitro cultured fat bodies was
different from that of both USP isoforms. Incubation in hormone-free medium for l6-hr
resulted in a steady increase from the high basal level of the EcR transcript present in the
previtellogenic fat body (Fig. 3C, open circles). Continuous incubation of the fat body in
20E-containing culture medium first led to a rapid increase of the AaEcR transcript level
by 4-hr incubation and then a considerably slower rise until 12-hr after incubation when
the EcR level reached a plateau (Fig. 3C , open squares). Interestingly, withdrawal of the
hormone after 4-hr of incubation caused the EcR transcript level to rise to a two-fold
higher level (Fig. 3C, open triangles).

As a positive control, I monitored the effect of 20E on VCP. Previtellogenic fat
bodies incubated in hormone-free media did not show any discernible level of VCP
expression (Fig. 3D). This is consistent with results reported before (Deitsch et al.,
1995a). Continuous incubation in the presence of 20E resulted in the robust activation of
VCP gene expression: the level of VCP mRNA doubled every 4 hours during the 12-hour
incubation period and then continued to increase to the end of incubation (Fig. 3D, open
squares). Hormone withdrawal after a 4-hr induction period reversed VCP gene
activation, and the VCP mRNA levels retumed to an undetectable level by l6-hr of

incubation (Fig. 3D, open triangles).

45

    

a... 1.2.1.; . . .uw. '1. 1. . l .

._ . .......... ......:....4.1. 4..
.1. .fiu.....$ if .99.... .. .4 .. .
.. .? ......... _ .
,.._ a”, .. . :2... f... ...:.u.u...1.....

. .11

 

 

 

Fig. 3. Effect of 20-hydroxyecdysone and cycloheximide on the transcription of
USP-A (A and E, respectively), USP-B (B and F), EcR (C and G) and VCP (D and
H) in the previtellogenic fat bodies. Previtellogenic fat bodies dissected from female
mosquitoes (3-5 day after eclosion) were incubated in culture media only (open circles),
with a 4-hour pulse treatment of 10'6M 20E (open triangles), with continuous 20E
treatment (open squares), with lmM Chx (closed circles), with lmM Chx and a 4-hour
pulse 20E treatment (closed triangles) or with lmM Chx and continuous 20E treatment
(closed squares) for 16 hours. At every 4-hour interval, a group of 9 fat bodies were
collected and subjected to RNA isolation, RT-PCR amplification and Southern blot
analysis. The intensity of bands from Southern blots was quantified by phosphorimaging.
For EcR, USP-A and USP-B, the level of amplified transcript detected from
previtellogenic fat bodies prior to any treatment was defined as 1 unit. For the VCP
mRNA, the level of amplified transcript detected from previtellogenic fat body treated
only with 20E for 16 hours was defined as 100 units. For comparison, the mRNA profiles
of fat bodies treated with 20E continuously were plotted in both, left and right, panels
(dotted curve with open squares).

46

 

20E/Chx

20E

 

--_ .. __.__q

       

 

 

a

.P

_S

"U

__

m

h

h

4

em.
mp.
mmm
CCC.
8 6 4 2 O
E

 

USP-A

 

——-a—— 20E-4h
- o - -2OE

——-o-—CM

00000000
4208642
4|

22635.28. <sz
A

16

12

16

12

 

llll
.

USP-B

 

 

 

 

 

USP-B

 

 

 

 

42086420
4|

€332.66“: <sz
B

16

12

 

EcR

 

 

 

 

EcR

 

8642086420

eczuaucwsot <sz
C

16

2

1

16

12

 

 

 

 

 

 

0.
.o

P .

C
V

.o.
0 0 0 0 0 0 0
2 0 8 6 4 2
H
o .
. . .

P

C .

V .6 .
0 0 0 w 0 0 0
2 0 8 4 2

.8256!
.6 283.: <sz
D

16

2

1

16

12

Culture Tlme (hours)

Culture Time (hours)

 

47

Fig. 3. Effect of 20-hydroxyecdysone and cycloheximide on the transcription of
USP-A (A and E, respectively), USP-B (B and F), EcR (C and G) and VCP (D and
H) in the previtellogenic fat bodies. Previtellogenic fat bodies dissected from female
mosquitoes (3-5 day after eclosion) were incubated in culture media only (open circles),
with a 4-hour pulse treatment of 10'6M 20E (open triangles), with continuous 20E
treatment (open squares), with lmM Chx (closed circles), with lmM Chx and a 4-hour
pulse 20E treatment (closed triangles) or with lmM Chx and continuous 20E treatment
(closed squares) for 16 hours. At every 4-hour interval, a group of 9 fat bodies were
collected and subjected to RNA isolation, RT-PCR amplification and Southern blot
analysis. The intensity of bands from Southern blots was quantified by phosphorimaging.
For EcR, USP-A and USP-B, the level of amplified transcript detected from
previtellogenic fat bodies prior to any treatment was defined as 1 unit. For the VCP
mRNA, the level of amplified transcript detected from previtellogenic fat body treated
only with 20E for 16 hours was defined as 100 units. For comparison, the mRN A profiles
of fat bodies treated with 20E continuously were plotted in both, left and right, panels
(dotted curve with open squares).

46

 

 

 

ipfionol

'C P (D and
iom female
tpen circles).
isZOE

nd a thou
E treatment
es were

:m blot
phorimaging.

re \"C P

d)‘ treated
INA profiles
11. panels

U

mRNA (Fold-Induction) mRNA (Fold4nduction) mRNA (Fold-Induction)

mRNA (Percent of

.a
b
O

d—A

ON
GO

Mulmum)

—o-— 20E-4h

~o-v20E

—o—CM

ZOE

 

USP-A

 

 

 

 

 

 

 

Culture Time (hours)

0 4 8 12 16

120
VCP
100 ID
60 l.
60
40 -'
05'

2o .

0 A

 

47

8 _.__._ _ ___-_
E -——i--— Chx/20E-4h

 

20E/Chx

+ Chx/20E

—-—O— Chx

 

 

 

 

H 120
100
so
60
4o
20

 

_.——.__a

 

 

 

 

 

 

 

 

0 4 8 12 16
VCP
,D
ID.
.0'
I '0'.

 

0 4 8 12 16

Culture Time (hours)

 

 

30E.

cult.

repr;
also t
tFto,

pellet:

 

E [fect of C ycloheximide on USP Transcription

To assess whether mosquito USP isoforms are under direct or indirect control of
ZOE, I utilized a protein synthesis inhibitor cycloheximide (Chx) in the in vitro fat body
culture experiments (Fig. 3, E-H). Previtellogenic Fat bodies were pretreated with culture
media containing lmM Chx, followed by further incubation with or without 10'6M 20E.
Fat bodies cultured continuously in the presence of lmM Chx exhibited elevated USP-A
transcription (near 4-fold induction, Fig 3E, closed circles), suggesting that Chx may
repress the expression of inhibitors blocking USP-A expression. A similar profile was
also observed when the fat bodies were incubated in media containing both 20B and Chx
(Fig. 3E, closed squares). Importantly, when the 20E-withdrawal experiment was
performed in the presence of Chx, no super-induction was observed in contrast to
samples treated for only 4-hr with 2013 alone (Fig. 3E, closed triangles).

The effect of Chx on the 20E induction of USP-B was different from that of USP-
A. Overall, the presence of Chx in the medium led to slightly higher levels of the USP-B
transcript than in similar treatments without Chx (Fig. BB and 3F). In the Chx-containing
medium, activation by 20E led to super-induction of USP-B mRNA (Fig. 3F), while its
withdrawal resulted in considerably reduced, but still elevated levels of USP-B transcript
(Fig. 3F, closed triangles).

Fat bodies treated with Chx displayed enhanced levels of EcR transcripts, higher
than in those incubated with a culture medium only (Fig.3C and 36). In the presence of
Chx, the effect of 20E on the AaEcR transcript in cultured fat bodies was very similar to

that observed for USP-B (Fig. 3F ). The level of VCP transcription was also monitored in

48

the

 

ml...

 

these fat bodies as a control. Addition of Chx to the culture medium resulted in complete

inhibition of VCP mRNA irrespective of the presence or absence of 20E (Fig. 3H).

Protein Profiles of USP Isoforms in the Fat Body

To determined USP protein profiles in the fat body during the vitellogenic cycle,
nuclear extracts were isolated from fat bodies of female mosquitoes at various
previtellogenic and vitellogenic stages. Nuclear proteins were then resolved by SDS-
PAGE and the USP proteins were detected using anti—DmUSP monoclonal antibody in a
Western blot assay. The 55-kDa USP-A and 52-kDa USP-B proteins produced from their
respective cDNAs using an in vitro TNT transcription-translation system were used as
positive controls (Fig. 4, lanes 1 and 2). Typically, the USP antibody recognized two
major bands in the fat body nuclear extracts: the upper band which displayed a mobility
similar to that of the TNT-produced USP-A, and the lower band which co-migrated with
the TNT-produced USP-B (Fig. 4, lanes 3-10). These upper and lower bands were
considered to be the in vivo counterparts of USP-A and USP-B, respectively.

There were no detectable USP-A and USP-B proteins in newly-eclosed, O-O.5 day
old female mosquitoes (Fig. 4, lane 3). Only a polypeptide of a considerably higher
mobility (43 KDa), as well as a minor band, was recognized by the USP antibody. The
identity of these lower molecular weight species and their relationship to mature USP-A
and USP-B are currently unknown. The USP-A protein was not detected in the fat body
of previtellogenic female mosquitoes until they were 3-5 days old, when a trace of USP-
A was visible. However, this protein dramatically increased during the early stage of

vitellogenesis, 3-6 hr PBM. Yet, by the peak of vitellogenic activity, 18-24 hr PBM, the

49

rm ,_
Dmes

 

TNT Days Hours
Lysate Post-Ecloslon Post-Blood Meal

 

 

‘F C? 13 is:

a. a. v3 '0 "c .x: “,1 _= _= J:

V3 W o, '2' V? ‘9 no so 00 o

D o -— m m -— m v o
L

 

55koav -
52koav - -
43kDa> - - .3 ' ‘4

Lane12345678910

Fig. 4. Developmental profiles of USP proteins revealed by Western blot. Fat body nuclear extracts
were isolated from female mosquitoes at previtellogenic stages 0—0.5 day, l-2 day, 3-5 day (lanes 3-5), or
post blood meal stages PBM 3-6h, l8-24h, 36h, 48h, 60h (lanes 6-10). Protein extracts, equivalent to 50 fat
bodies per lane, were loaded in each lane for SDS-PAGE and Western blot analysis with the anti-DmUSP
monoclonal antibody. In parallel, in vitro TNT expressed USP-A and USP-B proteins (lanes 1 and 2) were
analyzed as positive controls. Arrowheads indicate the sizes of major bands.

 

 

 

 

32p DR-4

2 1!: «fl

; 9; :z T— L‘

% A % if. j {g If

TNT Lys-te 2 S S S 5‘ _§ 3
DmUSP Ab - s ‘+—'. + - 4» -

-
U

 

 

 

       

 

                    

Lane 1 2 3 4 5 6 7 8 9 10

1112 13 14

Fig. 5. EcR/USP binding to a direct repeat DR-4. 32P labeled DR-4 EcRE was incubated with in vitro
synthesized AaEcR (lane 1), USP-B (lanes 2 and 3), AaEcR and USP-B (lanes 4 and 5), DmEcR (lane 6),
DmUSP (lanes 7 and 8), DmEcR and DmUSP (lanes 9 and 10), USP-A (lanes 13 and 14) or AaEcR and
USP-A (lanes 1 l and 12) and subjected to EMSA. Anti-DmUSP antibody was used for supershift assays to
verify binding specificity (lanes 4, 9 and l 1).

TNT Lycato AaEcR A-EcR DmEcR
AaUSP-A AaUSP—B DmUSP
DmUSP Ab '-—¥_ fi‘ 'ifi'v’ ‘—

 

" “‘ ' PO. -‘~
. w l

4.1.1, -
1 2 3 4 5 6

 

 

Lana
Fig. 6. EcR/USP binding to DR-12. ”P-labeled DR12 was incubated with in vitro synthesized AaEcR

and USP-A (Lanes l-2), AaEcR and USP-B (lanes 3-4) or DmEcR and DmUSP (Lane 5-6) and subjected
to EMSA. Supershift assays were conducted with anti-DmUSP antibody (Lanes 2, 4 and 6).

50

lei'1
lexc

deel

 

CUM]

“35 1

tr“) M ,

r
“1‘54

level of nuclear USP-A protein in the fat body drastically decreased (Fig. 4, lane 7). The
level of USP-A increased again at the late vitellogenic stage, 36-48 hr PBM, before
declining again at 60-hr PBM.

In contrast to USP-A, the USP-B protein appeared in the fat body of
previtellogenic, 1-2 day old mosquitoes and reached a high level in 3-5 day old females.
Moreover, the level of USP-B protein remained without any significant changes

throughout the entire vitellogenic and postvitellogenic periods (Fig. 4).

DNA Binding Properties of the E cR/USP-A Heterodimer: the E cR/USP-A
complex is Capable of Binding to Various EcREs

It has been shown before that both heterodimers, EcR/USP-A and EcR/USP-B,
bind to the hsp27 response element IR”“”-l (Kapitskaya et al., 1996). More recently, it
has been demonstrated that the EcR/USP—B heterodimer displays promiscuous binding
and transactivation properties towards DNA elements derived from the AGGTCA half-
site, arranged either as inverted repeats or as direct repeats (Wang et al., 1998). Here, it
was investigated whether the EcR/USP-A heterodimer exhibited DNA binding and
transactivation properties similar to those of the EcR/USP-B heterodimer.

First, EMSA was conducted with the inverted repeats containing the perfect
consensus sequence AGGTCA separated by a single nucleotide spacer (IRpcr-l). These
experiments showed that the EcR/USP-A heterodimer bound efficiently to IRp°'-1 (data

not shown).

Next, a direct repeat of the AGGTCA half-site with a four-nucleotide spacer (DR-

4 element) was tested. This sequence has been shown to have an optimal spacer among

51

 

 

 

 

lm

0’1

direct repeats for mosquito EcR/USP-B and Drosophila EcR/USP (Wang et al., 1998;
Antoniewski et al., 1996; Vogtli et al., 1998). TNT—expressed AaEcR, USP-A, USP-B,
DmEcR or DmUSP proteins alone did not exhibit any binding to DR-4 (Fig. 5, lanes 1-3,
6-8, 13 and 14). A combination of mosquito AaEcR and USP-B lysates yielded efficient
binding, whose specificity was confirmed by supershifts using the anti-DmUSP
monoclonal antibody (Fig. 5, lanes 4 and 5). Similar binding activity was detected for the
DmEcR/DmUSP heterodimer (Fig. 5, lanes 9 and 10). Likewise, the AaEcR/USP-A
heterodimer displayed similar specific binding to the DR-4 element as confirmed by a
supershift assay using anti-DmUSP monoclonal antibody (Fig 5, lanes 11 and 12).

Then it was investigated whether AaEcR/USP-A could bind to a direct repeat of
the AGGTCA half-site with a 12-nucleotide spacer (DR-12), which displays robust
binding activity with the AaEcR/USP-B and DmEcR/DmUSP heterodimers (D’Avino et
al., 1995; Wang et al., 1998). EMSA revealed that the AaEcR/USP-A bound the DR-12
element efficiently, and the specificity of binding was demonstrated by a supershift with
anti-DmUSP antibody (Fig 6).

Taken together, these experiments indicate that the AaEcR/USP-A complex, like
its AaEcR/USP-B complex counterpart, is capable of binding to various EcREs including

inverted and direct repeats as well as widely spaced repeats such as DR-12.

Estimation of Equilibrium Dissociation Constant (K4): AaEcR/USP-B Bound

on EcRE Twice as Strongly as AaEcR/USP-A
The similar binding properties of the AaEcR/USP-A and AaEcR/USP-B

heterodimers prompted us to investigate whether they might possess differential binding

52

AProbe Concen.(nM) 20 8.0 4.0 2.0 1.0 0.4

 

”a.

 

Q's" :1

 

 

 

 

La ne

B Bound (nl'l)
0.2

 

      

Boumvru-r
0 I) \ V

 

 

 

o ' 0 ()_N’
002‘»
0020 I
”HIS -
0.05
omn
””mn nm or on o:
lhmmi (n3!)
0 E A
0 5 10 15 20 25

EeRE(thsp-1) (an)

Fig.7, Direct measurement of equilibrium dissociation constants (Kd). Increasing
amounts of 32F labeled IRhSp-l EcRE probe ranging from 0.4nM-20nM was incubated
with in vitro synthesized AaEcR and USP-A proteins and subjected to EMSA. Bound and
free probes were quantified by phosphorimager analysis. Saturation curve (B) and
Scatchard plot (inset in panel B) were prepared to calculate the Kd. This experiment was
repeated three times and the mean value taken as the Kd.

53

all:

billt.

DR-

\\ he

a {‘11 7

 

affinities for EcREs. The AaEcR/USP-B heterodimer binds various EcREs with different
binding affinities. The equilibrium dissociation constants (de) for IRhSp—l, IRper-l and
DR-4 are 3.73 nM, 0.326 nM and 2.21nM, respectively (Wang et al., 1998). To evaluate

whether combinations of EcR with different USP isoforms exhibit different binding
affinities for these DNA elements, the Kd of AaEcR/USP-A to each [Rpm-1, IRhSp-land

DR-4 was estimated. The AaEcR/U SP heterodimer was incubated in the presence of

5x10'7 M 20E, with increasing concentrations of radioactive probes, IRhSp-l or DR—4.
Saturation binding analyses and Scatchard plots were used to estimate Kd values for the

heterodimeric pairs bound to thSp-l (Fig. 7) or DR-4 (not shown). These analyses
indicated that like the AaEcR/USP-B heterodimer, the AaEcR/USP-A bound to various
EcREs with different affinities. The differences in Kd values for these EcREs showed
that the binding affinity of EcR/USP-A to a DR-4 was about 2-fold higher than to an
IRhSp-l (Table 1), yet ten times weaker than to the IRper-l (not shown). More importantly,
these results demonstrated the AaEcR/USP-A complex bound to each EcRE with a 2-fold
lower affinity than the corresponding AaEcR/USP-B complex, suggesting that USP-B is

a more potent heterodimerization partner for EcR (Table 1).

USP-B is a More Potent EcR Partner than USP-A for Transactivation via
EcREs in C V-I Cells

Transactivation properties of two mosquito heterodimers were next compared
using the CV-l cell line. This mammalian cell line has no endogenous EcR and contains
very low endogenous level of RXR. It has been used to study transactivation of

EcR/USP from several arthropod species including Drosophila and mosquito (Thomas

54

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A a .m
DEtOH
3 Themes
g. 61
8
< 5
'8
E ‘*
é 3
’5 2“
:
5 o F‘- rtl :-
’5xtR”"’-1 ._.. ........................ .+.. ...................... ;. ........................ ; ................ 4. ........
EcR - - + - +
39§ET§ ‘ ............... T ...................... T. ......................... f ...................... f ......
B 09
,_ 0.3
,5
g 0.71
0.6.
3
:5, 0.5
3
g o...
E’ 03
E .
3 0.2-
'— 01
o Y . ..
EtOH MurA

Fig.8. EcR/USP-A transactivated a reporter gene with either IR-l or DR-4 EcREs
in CV-l cells. A. The AMTV-leRhsP-l-CAT reporter gene (1.2ug) and 0.4ug CMV-
LacZ were transfected into CV-l cells with 0.4ug of either AaEcR or USP-A plasmid or
0.4ug each of the two expression vectors. The expression vector pCDNA3.1/zeo (+) was
used as carrier DNA to equalize the total amount of plasmid so that each well received
2.4ug DNA. Cells receiving no exogenous DNA were used as a mock control. B. The
AMTV—3xDR-4-CAT reporter gene (1.2ug) and 0.4ug CMV-LacZ were transfected into
CV-l cells with 0.4ug of AaEcR and USP-A expression vectors. After transfection cells
were incubated with either vehicle (ethanol) or luM MurA at 37°C for 36 hours and
harvested for CAT activity and B-galactosidase activity, and CAT activity was
normalized with B-galactosidase activity. The experiments were repeated more than three
times.

55

et al., 1993; Yao et al., 1992; 1993; Wang et al., 1998). It has been shown previously
that transfection of AaEcR/USP—B into CV-l cells rendered them ecdysteroid responsive.
Furthermore, it has been shown that the levels of reporter transactivation by AaEcR/USP-
B correlated with the DNA binding affinities of this heterodimer when three distinct three
functional EcREs were compared (Wang et al., 1998).

The transactivation potential of the AaEcR/USP-A heterodimer was assessed with
the AMTV-leRhSp-l-C AT reporter plasmid, which contains five tandem repeats of IR”—
1. Transfection of CV-l cells with the reporter alone resulted in a very low basal level of
CAT activity (Fig. 8A). Co-transfection of the reporter with USP-A alone did not confer
MurA responsiveness. Weak response to MurA was detected in cells transfected with the
reporter gene and AaEcR expression vector, suggesting that AaEcR can heterodimerize
with endogenous RXR to activate this reporter gene. However, strong activation of the
reporter gene by 1 uM MurA (50-fold) was noted only in cells co-transfected with AaEcR
and USP-A expression vectors (Fig. 8A). This indicated that the AaEcR/USP-A
heterodimer formed a functional hormone-dependent transactivator in CV—l cells.

Next, to test whether the AaEcR/USP-A heterodimer could also activate a reporter
gene containing the DR-4 as an EcRE, the AMTV-3xDR-4-CAT reporter plasmid
containing three tandem repeats of DR-4 was transfected into CV—l cells. Co-
transfection of this reporter construct with either the AaEcR or the USP-A alone did not
render CV-l cells ecdysteroid-responsive (data not shown). However, co-transfecting the
reporter construct with both AaEcR and USP-A expression vectors rendered CV-l cells

highly responsive to the ecdysteroid MurA giving 6-fold induction (Fig. 8B).

56

Table 1. The equilibrium dissociation constants (Kd) of different DNA sequences
binding to AaEcR/USP-A and AaEcR/USP-B.

 

 

 

EcRE IR'””-l(nM)* DR-4 (nM)*
AaEcR/USP-A 7.36i0.81 4.06:0.05
AaEcR/USP-B 3.73:0.85 2.21:0.36

 

 

 

 

 

To measure the Kd values, in vitro synthesized AaEcR and USP-A were incubated with ,
increasing concentrations of 32F labeled IR,” -1 or DR-4 as described in Fig. 6. EMSA
and quantification for each element were performed a minimum of three times and the
mean values were taken as the Kd. For comparison, Kd values for EcR/USP-B to the
same elements are also listed (Wang et al., 1998. with permission).

* Mean of three determinations : SE

 

 

 

 

 

 

CAT Activity (Percent of Acetylatlon)

 

 

0 I 1 IT I l T I l l

USP (no) 0 (55 bis (1‘? .133 (go Kgo (19° b9o 19°
Fig. 9. AaEcR/USP-B mediated more efficient transactivation than EcR/USP-A.
Reporter plasmid (0.6ug), CMV-LacZ (0.2ug) and AaEcR plasmid (0.2ug) were
transfected into CV-l cells with increasing amounts of either USP-A or USP-B
expression vector ranging from 3.1ng-800ng. The expression vector pCDNA3.1/zeo (+)
was used as carrier DNA to equalize the total amount of plasmid so that each well
received 2.0ug DNA. Cells receiving no exogenous DNA were used as a mock control.
After transfection, cells were incubated with luM MurA at 370C for 48 hours and
harvested for CAT activity and B-galactosidase activity. CAT activity was normalized to
B-galactosidase activity. The experiments were repeated three times.

57

The transactivation potentials of the two heterodimers were then compared. In
this experiment (Fig. 9), a fixed amount of reporter plasmid (AMTV-SxIRhSPl-CAT,
600ng) and an AaEcR expression vector (200mg) were co-transfected into CV-l cells
along with increasing amounts of either USP-A or USP-B expression vectors, ranging
from 3 ng to 800 ng. Cells transfected with a reporter plasmid and the AaEcR expression
vector alone displayed less than 0.5 units of CAT activity (not shown). Co-transfecting
these cells with the USP-A plasmid increased the Mur A-dependant CAT activity in
proportion to the amount of USP plasmid added. The activation reached saturation with
200 ng of USP-A plasmid (Fig. 9). Co-transfecting the cells with USP—B instead of USP-
A resulted in higher Mur A-dependant activation of the reporter at lower levels of input
plasmid. Notably, the USP-B mediated transactivation reached its peak with about half
the amount of input plasmid compared to that of USP-A (lOOng versus 200 ng). In a
separate experiment, transactivation potentials of the two heterodimers were compared
using the AMTV-3xDR-4—CAT reporter plasmid. Co-transfection of this reporter with
the AaEcR and AaUSP-A expression plasmids consistently led to a lower level of
reporter induction than with the AaEcR and AaUSP-B plasmids (data not shown). Taken
together, these results further suggested that USP-B is a more potent partner than USP-A

for EcR mediated transactivation in CV-l cells.

DISCUSSION
In this work, the expression profiles of USP isoforms were characterized during
the first vitellogenic cycle in the female mosquito A. aegypti. The transcriptional

responses of USP isoforms to 20E were investigated, and the DNA-binding and

58

transactivation properties of the EcR/USP-A and EcR/USP-B heterodimers were
compared by employing EMSA and transfection assays. Importantly, here two mosquito
USP isoforms, USP-A and USP-B, displayed distinct stage-specific expression in two
vitellogenic tissues, the fat body and ovary. USP-A mRNA was highly expressed in
previtellogenic and late vitellogenic fat bodies and ovaries, coinciding with a period of
low ecdysteroid titer, whereas USP-B mRNA was enriched at the vitellogenic stage when
the ecdysteroid titer in the female mosquito is high. Significantly, USP-B, but not USP-
A mRNA levels, correlated with the level of EcR in both tissues.

The mRNA profiles of the USP isoforms strongly suggest that ecdysteroids be
involved in differential regulation of their respective promoters. Indeed, when I utilized
the fat body in vitro organ culture to investigate the effect of 20E on the transcription of
USP isoforms, these experiments unequivocally established ecdysteroid engagement in
controlling the transcription of both USP-A and USP-B. Remarkably, the effect of 20E
on USP-A expression was opposite to that of USP-B expression. Continuous exposure to
20E abrogated the transcription of USP-A. However, a 4-hour pulse treatment of fat
bodies with 20E followed by continued incubation in hormone-free medium resulted in a
drastic augmentation of USP-A transcription. Interestingly, these USP-A mRNA
responses to 20E in vitro are similar to those of AaFTZ-F 1. According to our
observations, AaFTZ-Fl is also suppressed by 20E, but is super-activated under
conditions of 20E withdrawal (Li C., Kapitskaya, M.Z. and Raikhel, A.S., unpublished
observation). FTZ-F 1 , an insect homologue of the steroidogenic factor 1 (SF-1), has
been implicated in regulating a stage-specific response to ecdysteroid (Woodard et al.,

1994). This similarity in the effect of 2013 suggests that, as in the case of FTZ-Fl, the

59

nuclear regulatory receptor, HR3 (Hormone Receptor 3) may be involved in mediating
the ZOE effect on USP-A. HR3 is involved in mediating 20E responsiveness in
DmSOphila (Lam et al., 1997; White et al., 1997). Moreover, the HR3 homologue has
been identified in the mosquito and has been found to respond positively to 20E in an in
vitro fat body culture system (Kapitskaya, M.Z., Li, C. and Raikhel, A.S., unpublished
results). The observation that Chx completely inhibited USP-A super-induction
following 20E withdrawal further substantiated the involvement of HR3 or other labile
factors in activation of USP-A expression.

In contrast to USP-A, the USP-B transcript was highly elevated by the continuous
presence of 2013 in an in vitro fat body assay. Furthermore, continuous exposure of the
fat body to 2013 was required for maximal activation of USP-B transcription. In the
presence of Chx, 20E super-induced USP-B transcription, suggesting that it represents an
early gene of the ecdysteroid gene regulatory hierarchy. One critical criterion to
differentiate early and late genes is the effect of a protein synthesis inhibitor such as Chx
on their response to hormone (Ashbumer et al., 1974; Huet et al., 1995). Vg and VCP
genes, which are transcriptionally induced by ecdysteroid but inhibited by Chx, are
typical late genes in the mosquito (Deitsch et al., 19953). In contrast, addition of
cycloheximide into the culture media super-induces AaE75, a mosquito early gene
(Pierceall et al., 1999). Of note, the effects of 20E on EcR in the mosquito fat body, as
well as the EcR mRNA response to Chx were generally similar to those of USP-B.
However, as in the case of EcR expression, a more complex pattern of response to 20E
compared to that of USP-B was observed (Fig. 3). This may reflect the presence of more

than one isoform of EcR as has been shown for M. sexta (Jindra et al., 1996a). In

60

r}.

lSU

llo

contrast to USP, only a single EcR isoform corresponding to Drosophila EcR—B1 has
been identified so far in the mosquito (Cho et al., 1995). However, our recent studies
suggest that there is at least one additional EcR isoform present in the mosquito during
vitellogenesis (Wang, S.F., Li, C., Zhu, 1.8. and Raikhel, A.S. unpublished data).

Interestingly, USP isoforms have been found to be differentially expressed during
the Manduca epidermis during larval molts (J indra et al., 1997). When the ecdysteroid
titer is high, the Manduca USPl transcript (the counterpart of USP-A) disappears and the
levels of the USP2 transcript (the counterpart of USP-B) are elevated, resembling the
USP isoform expression profiles in the adult female mosquito during vitellogenesis.
Moreover, in vitro exposure of the day-2 fourth instar larval epidermis to 20E led to the
decrease of Manduca USP-1 mRNA, whereas the effect of 20E on USP-2 mRNA was
opposite causing its levels to rise (Hiruma et al, 1999). A differential effect of 2013 has
also been reported for Manduca EcR isoforms (Jindra et al., 1996b). In vertebrates,
ligands can cause different or even opposite responses of various receptor isoforms
(Syvala et al., 1996; Paech et al., 1997; Keightley, 1998).

In the fat body of newly eclosed female mosquitoes, transcripts of both USP
isoforms and EcR are present with the USP-A transcript being particularly abundant.
However, our immunoblot analysis showed that the corresponding proteins were not
detectable at this stage in the fat body. A high mobility 43-kDa protein, recognized by the
USP monoclonal antibody, was observed in the newly eclosed female fat body. During
the extraction of the fat body nuclear proteins, a cocktail of proteinase inhibitors was
added and all samples were processed under identical conditions. Therefore, it is unlikely

that the 43-kDa protein is a degradation product of USP-A or USP-B. Our previous

61

 

Northern blot analyses have demonstrated the existence of four USP transcripts of
different sizes in both the fat body and ovary (Kapitskaya et al., 1996). Utilization of
isoform—specific probes derived from the 5’- and 3’- untranslated regions of USP-A and
USP-B revealed that the largest 3-kb transcript corresponds to USP-A, while two smaller
transcripts of 2.5 and 2.2 kb correspond to USP-B (Kapitskaya et al., 1996). The identity
of the smallest 1.8-kb transcript remains unknown. It is possible that a newly identified
43—kDa USP-related polypeptide is a product of the 1.8-kb transcript. This suggestion,
however, requires further study.

Traces of the USP-B protein appeared in 1-2 day old mosquitoes and reached high
levels in 3-5 day-old mosquitoes. By contrast, only a low level of the USP-A protein was
detectable in 3-5 day-old mosquitoes. EMSA utilizing nuclear extracts from different
stages of previtellogenic fat bodies also showed the appearance of functional AaEcR/U SP
heterodimers only in fat bodies of previtellogenic females 3-5 days after eclosion (Zhu,

J .S. and Raikhel, A.S., unpublished data). Thus, since the USP-B protein is the most
abundant USP isoform in the fat body of 3-5 day old previtellogenic females, an increase
in the USP-B protein along with EcR protein, which is also appeared at this stage, may
promote competence for the vitellogenic response to a blood meal (Zhu, J .S., Wang, S.F.,
Li, C. and Raikhel, A.S., unpublished data).

During the previtellogenic phase which covers the period from adult emergence to
3-5 days after eclosion, the female mosquito becomes competent for vitellogenic events
(reviewed by Hagedom, 1989; Raikhel, 1992). The previtellogenic phase is accompanied
by a rise in the juvenile hormone (JH) titer, which reaches its peak two days after

eclosion and then declines (Fig. l-A; Shapiro et al., 1986). Acquisition of competency to

62

 

 

 

Elia

vitellogenic events and responsiveness to ecdysteroid have been shown to be controlled
by JH (reviewed in Raikhel, 1992; Dhadialla and Raikhel, 1994). In the ovary, the
development of the receptor-mediated complex responsible for yolk protein uptake is
controlled by JH (Raikhel and Lea, 1985). In the fat body, JH-dependent events include
an increase in ploidy and proliferation of ribosomes (Dittmann et al., 1989; Raikhel and
Lea, 1990). However, the molecular mechanism of JH action in the mosquito is
unknown. In view of our observations on the delayed appearance of USP protein during
previtellogenic development, it is possible that .lH acts at the translational level regulating
coordinated production of a set of transcription factors required for rapid activation of
target genes in response to blood-meal induced hormones. Verification of this hypothesis
requires further study.

The level of the USP-A protein increased dramatically in the fat body in response
to a blood meal. However, at the active phase of vitellogenesis, 18-24 hr PBM, the USP-
A protein dropped to quite a low level, consistent with the decrease in USP-A mRNA
level, which was also absent at this time. In contrast to the fluctuation of USP-A mRN A
and protein levels during the vitellogenic stage, USP-B mRNA and protein levels are
expressed evenly high throughout this stage, suggesting once again that USP-B may be
the major partner for EcR function during vitellogenesis.

In order to provide additional insights into the possible roles of USP isoforms in
the mosquito, we have compared the DNA-binding and transactivation properties of the
EcR/USP-A heterodimer with those of the EcR/USP-B heterodimer which we have
characterized previously (Wang et al., 1998). Although the extent of our present analysis

has not been as detailed as for USP-B (Wang et al., 1998), it appears that USP-A can

63

effectively dimerize with EcR, and the resulting heterodimer exhibits a wide specificity
of EcRE recognition similar to that of the AaEcR/USP-B heterodimer. The EcR/USP-A
heterodimer bound inverted repeats (thSp-l and IRper-l), direct repeats (DR-4) and
widely spaced EcREs such as DR-12. However, by measuring binding affinities of both
heterodimers we have clearly shown that the EcR/USP-B consistently displayed higher
binding affinity than EcR/USP-A to either direct or inverted repeats. Moreover, although
EcR/USP-A mediates ligand-induced transactivation in CV-l cells from either IRhsP-l or
DR-4 EcREs, once again the EcR/USP-B heterodimer appears to be the more effective
transactivator. The functional difference between USP-A and USP-B in mosquito cells
could be more profound than observed in CV-l cells. The Manduca EcR heterodimerized
with either USP-1 or USP-2 bind IR,” -1 and another imperfect palindrome (EcREl)
from the MHR3 promoter. Yet only EcR-USP], but not EcR/USP—Z, efficiently
transactivated the MHR3 promoter in Manduca GV-l cells (Lan et al., 1999).

Thus, these findings have demonstrated that throughout the vitellogenic cycle, the
mosquito USP isoforms are differentially regulated at both the transcriptional and
translational levels. They exhibit strikingly different responses to 20E in fat body
cultured in vitro. Our functional studies of the DNA binding and transactivation
properties indicate that, although both isoforms form functional heterodimers with EcR
capable of recognizing a wide range of EcREs, USP-B forms a more efficient functional
ecdysteroid receptor than USP-A. Taken together, our findings strongly suggest that
USP-B is the major heterodimeric partner of EcR responsible for mediating the
ecdysteroid response during vitellogenesis in the mosquito. Elucidation of the possible

role of USP-A in the adult female mosquito requires further investigation.

64

Sequence comparison of the AaEcR with Drosophila EcR isoforms revealed that
the AaEcR shows high similarity with Drosophila EcR-B. Since many insect EcR
isoforms have been identified, it was likely to exist more than one EcR isoform in
mosquitoes. Indeed, most recently, the second AaEcR isoform has been identified in
mosquitoes by means of PCR amplification using degenerated primers based on highly
conserved regions in the EcR-A A/B domain between Drosophila and other insects (Fig.
10A; Wang, S-F, Li, C. and Raikhel, A., unpublished). Subsequently, the PCR generated
DNA fragment was used to screen the mosquito vitellogenic cDNA library. However,
extensive library screening failed to identify a cDNA clone in this library (Li, C. and
Raikhel, A., unpublished). When I performed PCR using a primer located on the PCR-
generated AaEcR-A fragment and a second primer on the 3’UTR of the AaEcR common
region, I amplified the entire nucleotide sequence of AaEcR-A from the vitellogenic-
staged fat body cDNA. This result further has confirmed existence of AaEcR-A in
mosquito fat bodies (now shown).

Next, I used AaEcR isoform-specific primers to examine the expression levels of
AaEcR-A in both fat bodies and ovaries (Fig. 10B). In fat bodies, AaEcR-A is barely
detectable during the previtellogenic stage; whereas, its levels increase after a blood meal
with a peak level at 24hr PBM followed by a rapid decline to the previtellogenic level at
48hr PBM. In ovaries, AaEcR-A is expressed at relatively high levels throughout the
previtellogenic and vitellogenic stages with elevated levels after a blood meal (Fig. 108).
Thus, the AaEcR-A expression levels appear to correlate with 20E levels during

vitellogenesis.

65

 

A.

E F

A/B C D
EcR-A m1 f _J:JH°mone

 

 

EcR-B l f H___.l.__lormone

 

 

Fat body Hours After a Blood Meal
PV 3 4 6 15 24 36 48
EcR-A W G - .. ....
EcR-B HQQQ~~-M
Ovary Hours After a Blood Meal
PV 6 12 18 24 30 36 42 48
EcR-A use- cm.-- 9
EcR-B ,, . ,- ,-
.. an-..‘ ”O
C, o 4 s 12 16 (hr) Culture
EcR-A - - m. 1“ fl

EcR-B . -‘ “1.“

Fig.10. Characteristics of the AaEcR-A isoform. (A). Schematic domain
structures of the mosquito EcR isoforms. (B). Expression profiles of
AaEcR-A in fat bodies and ovaries by RT-PCR/Southern analysis. (C).
Induction of AaEcR-A mRNA in the fat body culture.

66

The correlation of the expression pattern of AaEcR-A in the fat body with 20E
levels suggests that AaEcR-A expression may be under the control of 20E. To test this
hypothesis, previtellogenic fat bodies were incubated with 20E for 16hr, and the mRNA
levels of AaEcR-A were monitored during this period of time. Indeed, I detected that the
AaEcR-A mRNA levels rose gradually during the 16hr culture with a maximal level at
16hr, indicating that AaEcR-A is more inducible by 20E than the AaEcR-B (Fig. 10C).

In the fat body, the substantial low level of AaEcR-A expression during the
previtellogenic stage, and relatively high level expression only within a limited time
frame during the vitellogenic stage, suggest that AaEcR-A may not be a major receptor
for the ecdysone response during vitellogenesis. Elucidation of the possible role of

AaEcR-A in mosquito vitellogenesis requires further investigation.

ACKNOWLEDGMENTS

I thank Sheng-fu Wang for DNA binding and cell transfection assays and
initiation of the AaEcR isoform project, J insong Zhu for Western blot analysis and Ken
Miura for some RT-PCR experiments. I also thank Alan R. Hays’ assistance in mosquito
rearing and Dr. F. C. Kafatos for providing us with the Drosophila USP monoclonal

antibody. This work was supported by NIH grant Al-36959.

67

Chapter 3*

Characterization of the Mosquito Early Gene E 75

*Most results have originally been published in Mol. Cell. Endocrinol.
Ref: Pierceall, W: E., Li, Cf, Biran, A., Miura, K., Raikhel, A. S., and Segraves, W. A.
(1999). Mol. Cell Endocrinol. 150, 73-89.

" Underlined two authors contributed equally to this work.

68

 

 

d:

111,.)

L.
Inc

ABSTRACT

The steroid hormone ecdysone# controls genetic regulatory hierarchies underlying
insect molting, metamorphosis and, in some insects, reproduction. Cytogenetic and
molecular analysis of ecdysone response in Drosophila larval salivary glands has
revealed regulatory hierarchies including early genes which encode transcription factors
controlling late ecdysone response. In order to determine whether similar hierarchies
control reproductive ecdysone response, ecdysone-regulated gene expression in
vitellogenic mosquito ovaries and fat bodies has been investigated. Here, the homologue
of the Drosophila E 75 early ecdysone inducible gene has been identified in the yellow
fever mosquito Aedes aegypti, and show that, as in Drosophila, the mosquito homologue,
AaE 75, consists of three overlapping transcription units with three mRNA isoforms,
AaE75A, AaE7SB, and AaE75C, originating as a result of alternative splicing. All three
A dB 75 isoforms are induced at the onset of vitellogenesis by a blood meal-activated
hormonal cascade, and highly expressed in the mosquito ovary and fat body, suggesting
their involvement in the regulation of oogenesis and vitellogenesis, respectively. In vitro
fat body culture experiments demonstrate that AaE 75 isoforms are induced by 20-
hydroxyecdysone, an active ecdysteroid in the mosquito, independently of protein
synthesis. Furthermore, AaE75 is able to activate the Vg promoter in cell transfection
assays, suggesting its role as an early vitellogenic gene. These findings suggest that
related ecdysone-triggered regulatory hierarchies may be used reiteratively during

developmental and reproductive ecdysone response.

# Following the recommendation of Cherbas et al. (1984), ecdysone is used as the generic term for
hormones with the appropriate biological activities. The active hormone in the Drosophila salivary gland
hierarchies and mosquito vitellogenesis and oogenesis has been shown to be 20-hydroxyecdysone.

69

INTRODUCTION

The insect steroid hormone ecdysone controls developmental processes including
larval molting, metamorphosis and, in some insects, reproduction. In Drosophila, gene
regulation by ecdysone is best understood in the late larval and prepupal salivary gland,
where hierarchies of gene expression are reflected in the activities of polytene
chromosome puffs. Extensive characterization of puffing pattems has revealed that
ecdysone acts first to activate a small number of early ecdysone inducible genes, and that
their products control subsequent progression of the genetic response to ecdysone
(Ashbumer et al., 1974). Molecular characterization of this hierarchy has led to the
identification of the ecdysone receptor as a heterodimer of the EcR (Ecdysone Receptor)
and USP (Ultraspiracle) proteins, and has shown that the early genes are direct targets of
the ecdysone-receptor complex (Cherbas et al., 1991; Koelle et al., 1991; Yao et al.,
1992; Yao et al., 1993). Consistent with their proposed regulatory role, the early genes,
BR-C, E74 and E 75, each encode transcription factors which have been implicated in the
regulation of early and late gene expression (Burtis et al., 1990; Segraves and Hogness,
1990; Cherbas et al., 1991; Guay and Guild, 1991; Koelle et al., 1991; DiBello et al.,
1992; Hill et al., 1992; Yao et al., 1992, 1993; Fletcher and Thummel, 1995). The
expression of these genes in other tissues and at other stages of development is consistent
with the idea that these genes are also involved in controlling spatially and temporally
parallel regulatory hierarchies mediating ecdysone response in other tissues and stages
(Segraves, 1988; Thummel et al., 1990; Huet et al., 1993). Among likely ecdysone-
regulated processes in Drosophila, two of the least understood are the progression of

oogenesis and vitellogenesis. The ecdysone-inducibility of yolk proteins (Bownes, 1986)

70

and the female sterility of mutations affecting ecdysone synthesis and ecdysone response
are suggestive of a role for ecdysone in female reproduction, but little is known of the
specific role of ecdysone in these processes (Garen et al., 1977; Oro et al., 1992).

In contrast, the role of ecdysone in female reproduction is well established in
anautogenous mosquitoes, especially the yellow fever mosquito Aedes aegypti. In this
mosquito, initiation of egg maturation and vitellogenesis requires a blood meal, which
stimulates neuroendocrine production of the ecdysiotropic neuropeptide EDNH (egg
development neurosecretory hormone) or OEH (ovarian ecdysteroidogenic hormone)
(Lea, 1967, 1972; Matsumoto et al., 1989). This hormone stimulates ovarian ecdysone
production, which in turn controls the induction and progression of vitellogenesis and egg
development. In A. aegypti, ecdysone levels are low during the first 8-10 hr post blood
meal (PBM), with only a small peak at 4 hr. Thereafter, the ecdysone level increases
dramatically to a maximum at 16-20 hr PBM, then rapidly declines to‘previtellogenic
levels in preparation for the termination of vitellogenesis at 30-32h PBM (Hagedom,
1983, 1985; Hays and Raikhel, 1990; Raikhel, 1992). Aspects of the oogenic and
vitellogenic responses can be induced by exogenous 20-hydroxyecdysone administration
(Hagedom, 1983, 1985; Raikhel, 1992; Dhadialla and Raikhel, 1994), but the action of
20-hydroxyecdysone (20E) on the mosquito yolk protein genes, vitellogenin (Vg) and
vitellogenic carboxypeptidase (V CP) requires protein synthesis (Deitsch et al., 1995a).
This indicates that, as in the Drosophila salivary gland, ecdysone response occurs via a
cascade of regulatory events. Other functions associated with ecdysone during mosquito
egg maturation include the formation of vitelline envelope (Raikhel and Lea, 1991; Lin et

al., 1993) and the separation of the secondary follicle in the ovary (Beckemeyer and Lea,

71

1980). Thus, the mosquito provides an ideal opportunity to investigate hierarchical
control of ecdysone-induced gene expression during reproduction. Moreover, the critical
role of anautogenous mosquitoes as disease vectors amplifies the importance of
understanding this process, a potential target in the development of pest control
strategies.

Consistent with the proposed role of ecdysone in controlling oogenesis and
vitellogenesis, EcR and USP transcripts can be detected in pre- and post-vitellogenic
ovaries and fat bodies. Three size classes of transcripts for AaEcR, the A. aegypti
homologue of Drosophila EcR, are expressed in pre- and post-vitellogenic ovaries and fat
bodies (Cho et al., 1995). The other critical component, USP, is also expressed in these
tissues with one predominating in the previtellogenic stage and the other in the
vitellogenic stage (Kapitskaya et al., 1996).

Here, the A. aegypti homologue of the Drosophila E 75 gene has been identified
and characterized, a putative representative of the next level in the ecdysone response
hierarchy. As one of the classical early genes, E 75 has been proposed to play a role in
controlling ecdysone response in the late third instar salivary gland. E 75 transcripts are
induced during all developmental ecdysone pulses, and genetic analysis suggests that E 75
is required for all critical developmental ecdysone responses, during embryogenesis,
larval development and metamorphosis (W.A. Segraves, P. Jenik, C. F ichtenberg and C.
Hughes, unpublished). It is shown here that A. aegypti E 75 is expressed in ovary and fat
body following a blood meal, and that AaE 75 transcripts are ecdysone-inducible in
isolated fat body in vitro. These findings suggest that AaE 75 plays an important role in

mediating ovarian and fat body ecdysone response, and that ecdysone-triggered

72

regulatory hierarchies such as those implicated in the initiation of metamorphosis may be

reiteratively utilized in the control of reproductive ecdysone response.

MATERIALS AND METHODS

Degenerate PCR/In verse PCR amplifications

Several conserved regions with relatively low codon degeneracy, within the DNA
and hormone binding domains of Manduca sexta (Segraves and Woldin, 1993) and D.
melanogaster E 75 were used to design degenerate oligonucleotides for PCR
amplification of mosquito E 75 sequences (Figs. 1 and 2). Oligonucleotides used for
amplification of the 421 bp hormone binding domain fragment were E2F,
GNCTNATHTGYATGTTYGAY, annealing to common region nucleotides 560-579 and
E4R, CCACATYTGYTGNCKNAG, annealing to common region nucleotides 981-964
(Fig. 1). Each PCR amplification reaction contained ~lOO ng of genomic DNA, 100 ng of
each primer, 250 mM each dNTP, 2 mM MgC12, 33.5 mM Tris pH 8.8, 16.6 mM
ammonium sulfate, 10 mM b-mercaptoethanol, and 1.0 unit of Taq polymerase
(Amersham) in 32 ul final reaction volume, and was conducted for 35 cycles of 94°C for
1 min, 46°C for l min, 72°C for 1.5 min. If no product was observed after the initial
amplification, 1 ul of the initial reaction was diluted into fresh reaction mix and re-
amplified for 20 to 35 cycles. Where possible, re-amplifications were performed using
nested primer sets. For PCR amplifications using non-degenerate primers, annealing
temperatures were increased to 50-52°C and the extension cycle was lengthened to 3 min.
For mapping of genomic clones using PCR amplification, 10 ng of template DNA was

used in each amplification reaction.

73

 

Co
Ra
lle

ill“:

Plt

Inverse PCR templates were prepared by digestion with the appropriate restriction
enzyme, followed by ethanol precipitation in the presence of ammonium acetate, dilution
to a volume of 1 ml and self-ligation overnight at 18°C. The template DNA was then re-
precipitated and 250 ng was used in each PCR reaction. Oligonucleotides used for the
specific amplification of exon A0 sequences were GTCGAATTCDATRKNAGYTC,
annealing to exons A0 and Al from nucleotide 8 of exon A1 to nucleotide 5 88 of exon
A0, and 1N3, GTGTGTGCGGTGATAAAGC, annealing to nucleotides 28-46 of exon

A1.

cDNA and genomic libraries

An A. aegypti cDNA library prepared from poly (A+) selected mRNA from fat
body 0-48 hr post blood meal (PBM) (Cho et al., 1991) and an A. aegypti genomic library
constructed from Sau3A-digested DNA cloned into the AF ix II vector (Deitsch and
Raikhel, 1993) have been described previously. For size-selected genomic DNA
libraries, fragments of the desired size were excised from low melting point agarose gels,
digested with b-agarase (NEB), phenol/chloroform extracted, ethanol precipitated and
ligated into EcoRl-digested phage arms of the replacement vector 1607 (Davis et al.,
1980). Phage were packaged with Gigapak 11 XL packaging extract (Stratagene) and

propagated on E. coli C600 hflA to select for recombinant phage, which are cI-.
DNA sequencing

Vitellogenic fat body cDNA clones in XZAPII were propagated in Bluescript KS-

following in vivo excision of the recombinant Bluescript phagemid in the presence of a

74

helper filamentous phage (Stratagene). Sequence analysis of cDNA clones, genomic
clones and PCR products was performed by thermal cycle sequencing using a modified
Taq polymerase and chain terminating dideoxynucleotides (Perkin-Elmer). All sequence
data obtained from PCR-derived templates was confirmed by comparison to unamplified
genomic or cDNA sequence. Some preliminary and confirmatory sequence analysis was
generated on the IE] automated sequencing system at the Keck Biotechnology Center
(Yale University School of Medicine). Sequence data was assembled and analyzed using

Geneworks sequence analysis software.

Animals
Mosquitoes (Aedes aegypti) were reared as described (Hays and Raikhel, 1990).

Vitellogenesis was initiated in adult females 3-5 days after eclosion by allowing them to

feed on rats.

Northern blot analyses

Total RNA was prepared from ovaries or fat bodies dissected from female
mosquitoes 3-5 days old (previtellogenic), or at 6 hr intervals post blood meal (PBM),
using the guanidine isothiocyanate method (Cho et al., 1991). For analysis of
transcription events early after the onset of vitellogenesis, ovaries and fat bodies were
also dissected l, 3, and 4 hr PBM. Northern blot analyses were performed with the same
amounts of mosquito equivalents of total RNA (6-48 hr PBM ovaries) loaded on 1%
formaldehyde-agarose gels. Isoforrn-specific probes (designated A625 for AaE75A,

B660 for AaE75B and C350 or C796 for AaE75C) were derived from isoform-specific

75

exons. A625 contains all of the indicated exon A0 sequences. B660 extends from the
EcoRI site (nt 337) to the PstI site (nt 1007) of exon B 1. C796 was obtained by
amplification of exon CO sequences using oligonucleotides CF 1 and CR1 (Fig. 1). C350
is derived from an AaE75C cDNA clone, and extends from nucleotide 504 of exon C0 to
the EcoRI site formed at the junction of exons CO and Al (Fig. 1). The probe common to
all isoforms extends from an EcoRI site (nt 862) to the end of the indicated exon 4/5
sequences in the AaE7SB cDNA 2-3A6 (Fig. 1). All probes were used for hybridization
under high stringency conditions. The following cDNA fragments were used as probes to
control for stage- and tissue-specificity: mosquito vitellogenic carboxypeptidase (VCP),
the fat body-specific vitellogenic gene expressed exclusively during vitellogenesis (Cho
et al., 1991); the vitellogenin receptor (VgR), an ovary-specific gene expressed in the
germ-line cells, i.e., the oocyte and nurse cells (Sappington et al., 1996); the vitelline
envelope gene 15a-1, an ovary-specific gene expressed in the follicular epithelium (Lin et
al., 1993). In addition, a cDNA probe for A edes aegpti cytoplasmic actin (Deitsch et al.,

1995a,b) served as a control for a constitutively expressed gene.

Developmental RT-PCR analyses

The temporal changing profiles of the transcript abundance of the three E 75
isoforms were examined by using RT-PCR followed by Southern blotting with specific
radioactive probes. Total RNA was prepared from fat body or ovaries throughout the
first vitellogenic cycle as described above (Cho et al., 1991) with some modification in
which all isopropanol precipitation steps were done without low-temperature incubation

to avoid the co-precipitation of glycogen and salts. RNA preparations were determined

76

spectrophotometrically. A260/A280 and A260/A230 of RNA preparations were always
above 1.7 and 2.0, respectively and quite consistent irrespective of developmental stages
of the mosquitoes. Five microgram each of the RNA preparation was reverse-transcribed
by SuperscriptII reverse transcriptase (Gibco-BRL) and random hexamer (Promega) in a

reaction volume of 20 ul, diluted to 40 pl with TE buffer (10 mM Tris-Cl, pH 8.0,

containing 1 mM EDTA) and stored at -20°C as a cDNA pool until use. In the case of
mosquito tissues, 0.025 fat body- or ovary-equivalent cDNA pools ranging from 0.3 ul to
1.4 ul were used as PCR templates because of drastic changes in size of the tissue during
vitellogenesis. PCR-amplification of three isoform-specific DNA fragments was done
either simultaneously in a single tube with 25 ul of reaction mixture containing three
isoform-specific sense primers (75A-F, 7SB-F, and 75C-F) and one common antisense
primer (75com-R), or separately using only one set of primers. For the details of the
primers, see below.

Thermal cycling conditions were as follows: the reaction was incubated at 940C
for 3 min; this was followed by 18 (for fat body samples) or 16 (for ovarian samples)
cycles of 940C for 30 sec, 600C for 30 sec and 720C for 30 sec. As a reference of the
development in the fat body and ovary, the same cDNA was subjected to the 10 cycles of
RT-PCR followed by Southern blotting with a primer pair specific to the mosquito VCP
(Cho et al., 1991), and vitellogenin receptor (Sappington et al., 1996), respectively. After
PCR, 10 ul each of the reaction was fractionated 2.5 % agarose gel, transferred onto
Hybond N+ membrane (Amersham) under alkaline conditions. Radioactive cDNA probe
was prepared from 15 mg each of three PCR fragments generated by the primer pairs of

75A-F/75com-R (generating 517 bp fragment), 7SB-F/75com-R (394 bp) and 75C-

77

F/7SCom-R (280 bp), respectively, by random-priming labeling with [a32P]-dATP

(NEN), and the membrane hybridized with the probe at 65°C in a solution of 5 x SSPE, 5

x Denhardt' solution, 200 ug/ml of denatured salmon sperm DNA (Sigma) following

prehybridization. The membrane was washed twice at 65°C in 2 x SSPE for 15 min,
once in 1 x SSPE for 15 min and finally twice in 0.1 x SSPE for 15 min, all containing
0.1% SDS, then autoradiographed. PCR/Southem blotting of the total RNA preparations
without reverse transcription did not result in any appreciable signal on autoradiogram,
indicating that the contamination of genomic DNA fragments into the total RNA fraction
was virtually negligible (not shown). The radioactive blots were quantified with a
phosphorimager (Molecular Dynamics). Primers used for the developmental RT-PCR:

75A-F, TAGTGCAATCAACGTATACCAATC;

7SB-F, CGTGGAAGAAGACCACGATCG;

75C-F, AACATAACCACGTGACCTCAATG;

75com-R, CAAGGGCGATACTGAATTTTCTG.

In vitro fat body culture

To examine the ecdysone inducibility of the mosquito AaE 75 homologue, the
abdominal walls with adhering fat body (hereafter referred to as the fat body) were
dissected from 3 to 5-day-old previtellogenic females and incubated in an organ culture
system for 2 to 4 hr with increasing doses of 20E as previously described (Deitsch et al.,
l995a,b; Raikhel et al., 1997). Cycloheximide (Chx) (Calbiochem) was dissolved in
water immediately before use (Deitsch et al., 1995). Total RNA was extracted as

indicated above (Cho et al., 1991). As a control, RNA was extracted from fat bodies

78

incubated without the hormone or Chx. Northern blot or RT-PCR analyses were utilized
as described above. Hybridization using a probe for the ecdysone-inducible VCP gene
provided a positive control, while a cDNA probe for mosquito actin served as a control

for a constitutively expressed gene (Deitsch et al., 1995a,b).

Cell culture and transient transfection assay

Cell culture and transient transfection assay were performed basically as
previously described (Wang et al., 2000b). The pAc5-AaEcR and pAc5-AaUSPb
constructs contained the full length AaEcR and AaUSPb (Wang et al., 2000b). The Vg-
promoter-luc is the 2.1kb Vg promoter driving a luciferase reporter gene in the pGL3-
basic vector. The full-length of AaE75A was subcloned into the pAc5 NotI and ApaI

sites for transfection.

RESULTS

Cloning of an E 75 homologue from Aedes aegypti cDNA and genomic libraries

To identify the A. aegypti homologue of the Drosophila E 75 early ecdysone
inducible gene, a PCR-based strategy was utilized to take advantage of scattered highly
conserved regions in the DNA-binding domain and the E domain of the D. melanogaster
and Manduca sexta E75 proteins; the latter domain has strong similarity to the hormone
binding domain of receptors in the nuclear hormone receptor superfamily (Segraves and
Hogness, 1990; Segraves and Woldin, 1993). Several combinations of degenerate
primers within the hormone-binding domain amplified genomic DNA fragments of the
predicted size (see Materials and Methods). The largest of these, a 421 bp fragment, was

cloned and sequenced. It encodes a pollypeptide with ~90% amino acid identity to the

79

corresponding region of Drosophila E75. Thus the gene from which these sequences had
been amplified was tentatively named A. aegypti E 75 or AaE 75.

Using the internal sequence of the 421-bp fragment, a second set of
oligonucleotide primers was designed, directed away from one another, for use in inverse
PCR reactions. DNA prepared from a fat body cDNA library was cleaved with EcoRI,
recircularized and subjected to inverse PCR, yielding a 2.2 kb fragment. Direct sequence
analysis indicated that this fragment contained sequences extending 5' into the D region
which links the DNA and hormone binding domains and 3' through the carboxyl terminal
domain to the 3' untranslated region. Further PCR analysis of the cDNA library, using
degenerate primers for the DNA binding domain and primers derived from the sequence
of the 2.2 kb fragment, indicated that the fat body cDNA library contained cDNAs
extending at least into the DNA binding domain.

To obtain vitellogenic fat body cDNA clones, the 2.2-kb inverse PCR fragment
was used as a probe to screen approximately 2 x 10° recombinants. The cDNA clone
extending farthest in the 5' direction, 2-3A6, was sequenced in its entirety. Analysis of
this sequence indicated that this was a full length Aedes aegypti E 753 cDNA (Fig. 1,
exon BI and common region).

A second type of cDNA clone contained a region closely resembling the E 75 A1
exon, which encodes the first finger of the DNA binding domain, preceded by sequences
closely resembling the E75C-specific sequences of Drosophila. However, this clone
contained only 350 bp of E75C-specific sequence and appeared to be truncated at its 5'
end. Three sequential rounds of inverse PCR "walking" gave rise to an additional 2 kb of

AaE75C-specific genomic sequence. PCR of vitellogenic fat body cDNA using

80

mt.

 

(’1 D
TAWTACTCICICGCTCGGICMAGCTAGTGICGGTGIGCIMTIMUTGWTM(CWTWGUIWUWGWCWUCGCTMH 120
IQCLNKLTQDElDSlEI

1mmGTMTCGGTGATMCCMCWTMTUAWGCWGTMCCTWUGIWGTMUAWUGCCAGIGGKICMICAGGCGCCWGIWUTCC 240
RTFVIGDIQRIIYSPIIVOCKKIVKLEIASGLIQAPISSS

ATGGTMTUMTGGCCMC1106WWI(WWWMTMWCUCWCGIGCWGIGGCGCUHCWW 360
IVTSIGHHQLGQDRARIQIIAKPSIKPCENVAPVQAQEVII

CICATICGGUTWCWUKQIWWUIKMUKMQTWWCWCTTCCCATCGTCCGGT‘CIGICCICCTCICCMACICTMCGGT 4”
VIIDGRFYDLQIISIIASPSHHIHILQEPSIIIPVI.SSSPISIC

CI MTCATMCCMTGCCGGCACT AGI GMTWACCGGT MICCTCCMGCCWGCCCCCACT GCACGICCI CT CT CGCCMCACAATCATAGT 601m (CT ((601 EU
LIITNAGSSIIDTGIPPSIQPPPINHPVSTITIIVOAPPII

7501‘ —r
TCGWHCWWWKCCCGCUCCWWGCTMIUICWWIWTWIWGGGWIMEUCUWUWG 720
SAVSSSK011LPPPPATIIIVQQQHQHQHQQQQQHINVTSI

r——nr
06CUCGCGCCKCCCUCWMKCWAUATWGGTMTWGCC IGTGGCWMWWIWGCGTCCMTICWGCCCWCIAG M1
QPPPPPPPVCAGSIGGIGGAGIIAKIKSAHEEPSSSIIPDI.

 

runnnwmttummrmnnwmcnummurmmmmmmmmnmnuncm 120
HtSDISYQQSVIKEITIIVVQVQKCSQSQI

 

ISA-f
CWTMMCWWWWGGTCCCWKGCGWWCWICWCGGCCGCWWAGIWTCMCGTAMCA 240
PVISYIAETGEPCIPISTTV‘RIEEVIPPKEFIVVQSIYT

§
ATGICCTMMTTATWATWIATTGTWWCICCACCWCTAIGCTGGCTGG‘GITIAICIACCMCTUTUTCTCGCWGMCGC 360
llllPKI'i‘lSRKNIVLESSPPHKPILAGDYIPAHHlHQQQQI

mmutmmummmumammtmmumm ((GGIATIMTCCMCGTCTTCTCC“(16601010111010 480
QlAlllll0FSKGPVLLVPKTEDCVIYSHPTPSSPTGHHNIF

(“T“CI'ICICCMCTCWTGCKGCUTUGGCICCTCCTGTGGATAIMCCCCGGIGTIAIWTMCAGCGIIGCQWUMWMCWCICMWAG 6“
SVASPAPIHHHQAAPVDIKPPVIIDISVASSSKGEPDLNI

MAI

MITCGACGGMCCAU.‘GIACIWCGTCICIGC‘CIWMMC‘ITCCGGI'ITCGCIKGGTCTGCKTCCIGCWGGCTGCMG 89
FDGT'TVL(RVCCDKASGFNYGVHSCEGCK

Fig. 1. Nucleotide and amino acid sequences of AaE75. Sequences for the
transcription unit-specific exons, C0, A0, A1 and Bl, and for the common region, shared
by all three AaE75 transcription units, are shown. Oligonucleotides used for RT-PCR
analysis and other probe generation are indicated above the nucleotide sequence. Introns
separating common region exons 2, 3 and 4/5 are indicated by vertical arrows.

81

0101131

CMTTCGGCACCACATCCT CCCCCCT C61 C61 MACCTT CT (CT CGCCCT CMCCMACCCCCT (GT CGCCCITT MGCT CTAGCT ACT GCGT CGT CAGT GCCT CT CAGCCACATATACC
TCMCAAMMATMCCMCATCCMTCCMGGATCTCTGCCAMGTGTCCAMCTCCGTATCMTCMGCGGCMGTCATCCCCCAGCMTACACACAGCTGTAUACTTATCACCMG

TGCMTACATCGGGACT GMGMCGMGGCCGGCACTI CACAGCT ACMMMGACTT CAAGT CCATTCATCCMGT ACT CCAACAGT CCMCI CT CMTTCCCCT CAAMTCCCATGT C
11 G C

CAGTGCACCACGAGCCCGCCACIAACCAGMATCCICGGCCGCGGCGCICAACACTCACMMCTCIGMAG'ITCCMGCAGTCACTCTAGTITMACMTACTMTCATACCACCMCA
AVQQEPASIEKSSAAALKTEKSLKVPSSQCSLK111110115!

GTACCCCIAGCACCAGCACCMCACCCTTMCMGGCCCACAGIGCACIAGITAAMTTCTTCMTCCGCTCCCCTGMTACTTCMATTTCATCACACTCTACMCGCMCTTCCGCTG
SSCSSSSASLKKAHSAIVKIlESAPlNSSItlTLYNGSSC

GCACT CGTTCIACT ACCACT ACT MTAGT ACT ACT ACT CMAGT MICMGT ACT ACCT CCT MGGAGGAMCCMMTCACI GT GIT AGCAAGT AGIT CCAGT CAMCCCT CCMCATC

CSGSSSSSNSSSSQSNQVVAPKEETKISVLASSSSQTVED

m-r ‘0
CTCACGCCCGCMATICAMACMCCCCCMCATCGATTTCIGCCCGTGCMCMCACCACCATCCCCAMCAGTCCATTAGTG CATGCCAACCGGGCTCGTCTCCCTCAGTAGATGATA

PNACKLKTTPKIOFCPIKKITIAKEIISACERGSSASVDD

 

ACMTCMCCT CACCAGCACCMCMCMCCCGT CACCCAGT 6T GCCAIACT ACACCMCACT CCCACAMCACCCAACCGCACCATCCCGGAGI CGCCACT CCT C GCACACTT CCCATC
KIQPQQQQEECVSQCAIlOEQSHKQPTCCCRSRESSQSSD

ACT CGT CCCT GT CCGCATCT CCCCAGCCCACCACCCATCACACCT GCT GCAGCMTACTT GCAGTT CCAAIGGMCCAGT AGCCACI CCICTTCC CACTATCI CATCAAIGATCT CT GCA
115$LSASPCPTSODSCCSNSCSSNCSSSDCCSDYLINDLC

mcmwcmrrcwmm 1112
KQFEEILCEOH

cm region

 

‘
CGTTT CTT CACGCGCT CCATCCAGCACAMATICACT ATCGCCCTT CT ACT MAMCCAGCAGT CT ACT ATACT CCGMTAMCCCTMTCGCT GCCAGT ATTCT GAMMA
CFFRR SIIQQK QYRPCTKAQQC lRIIRNR CQYC°IGIGCTKKTCc

I
ATCCCCGTTGGAATGACI’CGCGACGCGGICACATTCGGTCGTGTGCCGMCCGCCMAAGGCCCGCATCCTCGCGCCCATGCAGCAAACCACCCAGM
IA VGISADAVRFGRVPKAEKARI lAAIQQSTquccmgwccamccum

GCCACCCAGCTCCACCATCAGCCCCCTCTCCTCGCGCCCCTCCTCCCACCCCACATGCACACGTGTCAGTTCACGCGTCAGMGGTCACTTCCATGCGGCMCCT
ATE DQPIII.LAAVLllAllltlllCEliTllEKVT$11thR°ATCACGCCTCCCIST

I
ACCIACTCCATCCCTACTTTCCCATCTCCACTCMCCCCCCCCCACACCIACAATEACACCAGCAGTTCAGCCAACGCTTCCCCCAIGTMTTCGTGGCGICATTCACTTT
YSSIP'I'I.ACP1.IIPAPELQEQEl’lellFAlilllIlGVIDliGAmgnAG

ATCCCCGCCTTCCACATGCTCACCCAEGACGATMGITCACCCICCTCMCCCCGGCCTGTTCCATGCGCTCTTCCIGCGTCTGATTTGCATCTTCGACTCGCACATAMCICCATCATC
FlQI TDQD HLLKAGLFDALFVRLICIFDSQIN II

TGCCTGAATCCCCAGGTCATCCCCCGCCATACCATTCACMCCCTGCCMCGCACGTTTCCTGCTCGATICGACGTTCAACTTTGCCCMCCMTCMCTCGATCMTCTGACCCACCCC
C ICQVIIRRDTIQACAIARFLVDSTFIFAER1111 SAN LT DA

240

1080

240

720

WATCGCTCT CIT CT CT CCCATTCT CTT GATCACGCCCGACCCACCGCCACT GCGCMTCT GCAGIT CATCCMCCCATCT ACT CCMCCT CMCGCTT GCCT CCACT CGATCCT ATCC m

EIC1FCAIVlITPORPCLRNVELIERIYSKLKAClQSIVS

Fig. 1 (cont’d).

82

CMMCCCGCCCHTMCCCACMTTCATCCACCMCTGCTCCGMCACTACCGGATCTGCGTACGCTCAGCACCTTGCACACGGAGMGCTCGTAGTGTTCCCMCGCAGCATAMGAA
QIAPDKPEFIQELLRTLPOLRTLSTLHIEKLVVFRTEHKE

CT CCT ACCCCACCACATCT CCMCCCT CMCMCATCT CGCAAACACT CCCACCT CCMCACCT GCACCT CCCACGGCMTMTGT CCAGCATGT AGCCMCAGI CCCATGAGCT C661 A
I.RQIIAIIAEEDLAKSPSSITISCDGINVEDVAKSPASSV

TCCAGT ACT CAST CCGCCGMACMCCT CCCACI ACT CGCATATTCCTT CCT CCTT GACCGCTT CGCCGCCT CI GCT GGCCCCT ACT CT GI CCGGT CMTGCCCT ATCCGC CATCGGCCA
SSIESAETTSDYSNIPSSLSASAPLLAATLSGQCPIRHRA

AGCACCGGTTCGTCCGCACAACATCACATCATTCGTGGIACSGCCCATCTCGCACAGAACCCATTGACCATCACACCCCITATCCCATCGGTTGGMCACATCACGTCCGGTACCGGAM
SSCSSAEDDIICGTAHLAQNCLTITPVIRSVGTHHVRYRK

CTACACTCACCCACCGATTCC CGTATTGMTCTGGMACCMAMCACGAC CACAMCCAGTCAGCAGTGGATCGTCIICTTGCTCMGTCCACCATCTTCCCTACACCAICAATCCGAC
lOSPTDSCIESGIEKHDHKPVSSCSSSCSSPRSSLEDQSD

CACMCCCTCACATCGTGCMMCAICCCCCICCTCAACCCGGTCTTGCAGGCACCTCCCCTGTACCATACCAACACCCTGATGCACGMCCCTACMACCCCACAACAMTTCCGTGCA
DKRNIVEAHPVLKRVLQAPPLYDTNSLIDEAYKPHKKFRA

ATGCCT CACACCCAMCTCMGCCCAGCCT CCT CCMGTT CCT (CT (CT CT ACAATCACT CT CT CATCACATTCGC CCMACCATCCT CCAGCT CTT CCCCATCTT CCCTGGT ATCGT CA
11 SEA EPAPS SSSSTITCSSASPKP SSSSSPSSVVSS

CACACCACCACTATGCCTCMTCACCTCCCCACCMCCCCTTTCCCACCITCATATCCATTTMCCCCACCGTI’CMTCACTCACCACACCMACCCTTCACCACCACCATCATCAICAC
11555116 CQSPPQQPLSQCHIHLTRPLAQSPHQSLHllllllllllll

(ACT CGGGMCT CCT CCCGCT AGCCATCATCAGI CCT CTT CCCT ATCCAGCACCCATTCCGT GIT CGCCMATCCCT CATGCAGCAACCCACGATCACT CCGCAACAGATCMACCTT CC
HSGTPPAS llSllQS $lSSTll SVLAKS EQPTATPEQIKRS

CACATCATCCACAACTACATCATCCCACAGTCCCACCMTCTCACCACCATCCCCMGCGCACAGTTCGTACCGCAGTGGTCTCCTACTTTCCAGCCCACCATCCTCATCGGTCTCGCCA
DIINNYIIRESQQSQHNPCAQSSYRSCLLVCSPRSSSVSP

CT GACCT CCACCAGT AGCACT ACCAGT MCT CATCCT CCT CCGGCAGTGCTI GT CCATTCT CACCCMTAGCAGCAGCAGCAGCI CCCACACI GCACCCACMCCGI'T ATCACCACCACC
lTSSSSSSSISSSSCSCCPFSANSSSSRIQSGPTSVITTT

CCCCACCAMCT CCTT CT CCATCCCT CMTTCCACCCGMGT GCT CATACTT CCT CCI CTT CCAGT ACT GCCAGCACCAGT ACT ACCATCCCCT ACTTCCAGT CACCACATTCCACATCG
RQQTPSPSLISTCSADTSSSSSTASSSTSIIYFQSPIISTS

ATCT CCCCACCGCT GGCCCT CATCCCTCMCMGCCACCAGCACMCCACCGCATCCATTGCCT CCCCI (ACT (CST 6111’ CACGGCACCCT CTCCCT CC CC'I CCCCT CATCCACCTT CM
ISPPVAVIAQQASSTTTCSICSPQSVVTAPSPSPRLIElQ

CIGCACATCGCTCCCGCTTCCACTTCCTCCCGCACCACCCATCCCCCCCTCMCCTCTCCAMMAICCCCCTCCCCTTCCCCACCICCCCACCMTCMCCTCCTCCTCCTCCACAGCC
VDIACGSTSSGTTDAPLNLSKKSPSPSPARHQSISSSSIA

ACCATCCATAAGGT CCT ACT CfigAfiCGT MGCCACCCCCACCACCATTACMCMCMCMCMCT ACT ACT ACT ACT MGT CATCGACAAAMTCMCI ACT ACT AMCAMACAAGCC
T I 11 K V I. 1. A

CITTMACACGCAGATCAGTTTACACMCTGTACAMCTITMCATAGTMCACTAGTCACTATTGCTACGCGCTACCCATTTACMTTTTATAMTCACAMCCCCACCCCCCCCCCAC

GCICCCCICACGCCCACTCCTCCCGAATTC 2790

Fig. 1 (cont’d).

83

ton

1.200

1320

1440

1920

2160

2640

2760

oligonucleotides based on this sequence indicated that the cDNA library contained
sequences extending nearly to the start of the AaE75C open reading frame, but screening
of an additional 2 X 10° cDNAs did not reveal any full length E75C cDNAs. Therefore
inverse PCR on EcoRI-digested recircularized templates was utilized to isolate the C-
specific region from the rare or truncated cDNA or cDNAs responsible for this signal. At
the 3' end of the isolated sequences, the presence of an EcoRI site derived from the exon
CO-exon Al splice (see Fig. l) confirms that these sequences derive from spliced RNA
and not from genomic DNA. At the 5' end, sequence analysis indicates that these
sequences extend to within 30 bp of a putative initiator Met closely preceded by an in-
frame stop codon (Fig. 1). RT-PCR analysis indicates, however, that the 5' end of
AaE75C likely lies significantly upstream of this region (data not shown).

No AaE75A-specific cDNAs were identified in the screens described above. In
order to isolate the AaE75A-specific coding region from the fat body library, primer sets
were designed to amplify specifically AaE75A sequences. First, primers annealing
within the A1 exon were used for an initial round of inverse PCR amplification, then a
predicted exon AO-specific degenerate primer was used to re-amplify these products (see
Materials and Methods). This yielded 183 bp of sequence closely matching predicted
AaE75A-specific sequences. Primers derived from this sequence were used in a second
inverse PCR reaction on recircularized EcoRI-digested fat body cDNA library material,

giving rise to a fragment extending beyond the 5' end of the AaE75A-specific ORF (Fig.

l).

84

Defining the Aedes aegypti E 75 genomic structure

To elucidate the genomic structure of the AaE 75 gene, the cDNA fragments
described above were used as probes to identify genomic clones from size-selected and
complete A. aegvpti genomic libraries, yielding a collection of overlapping genomic
clones within each region. Exons were localized by restriction mapping, hybridization
and PCR using exon-specific and vector primers (Fig. 2). The intron between exons A0
and A1 is approximately 15.6 kb in length and the intron between the AaE7SB-specific
region and exon 2 is 150-200 bp in length.

The failure of clones A8A and 810 to cross-hybridize indicated that there are
additional intron sequences separating these two clones. Genomic clones A8A and 810
were attempted to use to define the length of this intron through genomic walking. Three
rounds of chromosomal walking from the A8A side yielded an additional 20+ kb of
genomic DNA before the presence of repetitive sequences prevented further walking.
Screening with a 5' terminal fragment of clone B 10 yielded no additional clones,
indicating a potential gap in the genomic library. While the complete length of the intron
between the A1 and B1 exons thus cannot be determined, these findings suggest that it is
in excess of 30 kb.

Within the common region, restriction mapping, hybridization, PCR and direct
sequence analysis showed that there is no intron separating the hormone binding domain
and the carboxyl terminal F domain, but that there is an intron of approximately 3.8 kb
separating exon 3 from exon 4/5. To determine the length of the intron separating exon 2
from exon 3, a 5' terminal fragment of clone E7 was used to isolate a clone, L13, which

also hybridized with the 810 clone. Further restriction mapping and Southern blot

85

co A0 A1 312 345

0111575 0 0111575 A DmElS B

 

 

 

co ‘° “ 312 3 us
me “7% “a“
l—‘—“-‘-* A8 F- J. 810 : . . :57; in U,
C13 L13 Hull—1
5: l I I l l J 1 . g I 13.
10k!)

 

Fig. 2. Genomic organization of AaE75. The intron-exon structure and partial restriction map
of genomic clones are shown. The DNA binding domain is encoded within exon Al and exon 2.
The region of the steroid receptor hormone binding domain is located within the first ~200 amino
acids of exon 4/5. The intron-exon map of D. melanogaster E 75 (DmE 75) is shown for
comparison. EcoRI sites are indicated on the genomic clones.

lsoform-specific domains

 

     

AaE75A shared common region domains

 

AaE75C

AaE7SB

Fig. 3. Comparison of Aedes aegypti and D. melanogaster E75 amino acid sequences.
Schematic representation of sequence identity within structural domains of the A. aegypti E75 A,
E75B, and E75C-specific regions as well as the DNA binding domain (C), linking region (D),
hormone binding domain (E), and repressor region (F). Overall percentage of amino acid identity
between A. aegypti and D. melanogaster E75 domains are presented below each region. Sub-
regions showing high levels of identity within less conserved domains are indicated as dark

stippled bars.

86

analysis allowed alignment of these clones and showed that the intron separating exons 2

and 3 is approximately 17 kb in length (Fig. 2).

Evolutionary conservation of the dipteran E 75 proteins

The predicted amino acid sequences of the AaE75A, AaE7SB and AaE75C
proteins are shown above the corresponding nucleic acid sequences (Fig. 1). Fig. 3
schematically depicts the comparison between these sequences and those of E75 A, B and
C from D. melanogaster. The DNA binding, linking, and hormone binding regions
display high levels of identity with Drosophila E75: 100%, 91% and 93%, respectively,
at the amino acid level. The terminal domains are substantially more substantially
divergent, but show a similar organization and contain several highly conserved
subregions. More significantly, the amino terminal domains specific to the AaE75A,
AaE7SB and AaE75C isoforms are clearly conserved with each showing ~30% identity

to the amino terminal domain of the corresponding Drosophila protein.

AaE 75 transcript analysis

The transcript size of the AaE 75 gene was examined by the Northern blot analysis
of vitellogenic tissues of female mosquitoes during the first cycle of vitellogenesis.
Mosquito ovarian RNA 24-30 hours PBM was hybridized with isoform-specific probes
(A625 for AaE75A, B660 for AaE75B and C350 or C796 for AaE75C; see Materials and
Methods), revealing two transcripts per isoform (Fig. 4): 6.2 and 5, 6.9 and 5.6, and 6.8
and 5.4 kb, for isoforms A, B, and C, respectively. When a probe to the region common

to all three AaE 75 isoforms was utilized, 5 bands were observed (Fig. 4). These findings

87

Probe A B C Com

 

   

‘L‘A
“on .3.
9.5——
7'5 _6_2 -6.9 —6.8 ‘3'
4.4— l --5.0 —5-5 —5.4 -
2.4—
1.4—

Fig. 4. AaE 75 transcripts in the mosquito ovary. Total RNA was extracted from
ovaries dissected from vitellogenic mosquitoes 24-30 hr after a blood meal. Northern
blots were hybridized with AaE75 isoform-specific probes. (Lane A, AaE75A-specific
probe; Lane B, AaE7SB-specific probe; Lane C, AaE75C-specific probe; Com, common
region AaE75 probe).

A
1" ( u
_ 'l— '1;
9 5 " ‘
_—
- t

7.5— ; i _7.2
. -. _-, ’ —6.1

4.4—— y; .. ,3

2.4—— "

1.4— ;.

 

 

 

Fig. 5. AaE75 transcripts in the mosquito fat body. Total RNA was extracted from fat
bodies dissected from vitellogenic mosquitoes 24-30 hr after a blood meal. Northern
blots were hybridized with AaE75 isoform-specific probes. (Lane A, AaE75A-specifrc
probe; Lane B, AaE75B-specific probe; Lane C, AaE75C-specific probe).

88

are consistent with the presence of several different 3' ends in the cDNA clones analyzed
(data not shown), suggesting that these RNA size-classes arise from utilization of
alternative polyadenylation signals, as in the case of Drosophila E 75 (Segraves and
Hogness, 1990).

In the fat body, transcript size was also established using Northern blot
hybridization of RNA from the vitellogenic fat bodies 24-30 hr PBM with isoform-
specific AaE 75 probes. As in the ovary, two transcripts are clearly observed for the
AaE75A and AaE75C isoforms. Similarly, it is likely that two transcripts existed for
AaE75B (Fig. 5). Overall, the level of E 75 RNA in the fat body was considerably lower
than in the ovary, making the Northern blot analysis a relatively undesirable alternative

for further analysis in this tissue.

Time course of AaE 75 expression in the fat body and ovary during
vitellogen esis

To determine the temporal expression pattern of AaE 75 mRN A levels in the
vitellogenic tissues, the fat body and ovary, RT-PCR analysis was primarily used.
Complementary DNA was reverse transcribed from fat body total RNA and then
amplified in PCR reactions using isoform-specific primers (see Fig. 1 and Materials and
Methods). In both vitellogenic tissues, AaE 75 expression was induced at the onset of
vitellogenesis by a blood meal-activated hormonal cascade, and AaE 75 isoforms were
highly expressed in the mosquito ovary and fat body, suggesting an involvement in the
regulation of oogenesis and vitellogenesis, respectively. In the fat body (Figs. 6a and 6b),

E 75 transcripts exhibit a sharp rise immediately after the onset of vitellogenesis reaching

89

(a).
Days after Eclosion Hours after Blood Meal

o-11-23-sl1 3 4 6 15 24 3648

 

 

E75A—\ ..... a.--
E7SB— "' --.8

~a-“e- .

517 bp
394 bp

 

 

 

 

 

137st F280 bp
VCP ]—456 bp
(b).
Em 3
E“ E

 

 

 

 

PV in 30 411 611 15h 24h 36h 4811

Time After a Blood Meal

Fig. 6. Accumulation of AaE75 mRNA in the mosquito fat body. This experiment
was performed by Southern blotted RT-PCR (see Materials and Methods). The results
are expressed normalized units of the highest readings per fat body and represent means
of three determinations :t SE. Shown also are the results of a RT-PCR with VCP, which
serves to illustrate the course of vitellogenesis in the fat body. A. a representative
Southern blot of the RT-PCR; B. quantitative analysis of the data.

90

a peak at 3-4 hr PBM when ecdysone titer shows a first small peak (Hagedom et al.,
1975). After the initial drop, levels of AaE75A and AaE7SB transcripts gradually
increased again reaching a second peak at the 24 hr PBM when ecdysone levels just have
passed their major peak, at 18-20 hr PBM (Hagedom et al., 1975). The second gradual
increase in AaE 75 transcripts closely correlated with the accumulation of RNA for YPPs
in the fat body as demonstrated by the VCP transcript kinetics. The second peak of
isoforms A and C transcripts was not as pronounced as that of isoform B. After the
second peak, the amounts of A01? 75 transcripts decreased to very low levels at the end of
vitellogenesis, 42 and 48 h PBM (Figs. 6a and 6b).

The temporal expression of the AaE 75 gene in the ovary was first examined by
Northern blot analyses utilizing total ovarian RNA for time points spanning the first
vitellogenic cycle from 0 to 48 hr PBM. For each time point, the same number of ovary-
equivalents of RNA was loaded, and blots were probed with the three isoform-specific
A aE 75 probes. To provide a reference with follicle development, two ovary-specific
probes were utilized: vitellogenin receptor VgR cDNA, (Sappington et al., 1996), which
is specific to the germ-line cells (i.e., the oocyte and nurse cells); and the VE gene 15a-1
(Lin et al., 1993), which is specifically expressed in the follicular epithelium surrounding
the germ cells in the mosquito follicle. These Northern blots showed that transcripts for
all three A015 75 isoforms exhibited similar patterns of accumulation and disappearance in
the mosquito ovary (Fig. 7). The AaE 75 transcripts accumulated slowly during the first
10-12 hr PBM; however AaE75A and Aa7SB transcripts exhibited dramatic
accumulation during the next 12 hr, reaching a peak by 24 hr PBM. The amount of these

AaE 75 transcripts declined between 24 and 36 hr PBM, the time coinciding with the

91

hours PBM

E758

E75C

V9
Receptor

. 6 , Envelope

Fig. 7. Expression of AaE75 RNA isoforms in the mosquito ovary during the first
vitellogenic cycle. Total RNA was extracted from the same amount of ovary equivalents
dissected from previtellogenic mosquitoes 3-5 days after eclosion and from vitellogenic
mosquitoes at the indicated hours after a blood meal. Northern blots were hybridized
with AaE75 isoform-specific probes (see Materials and Methods), and with Vitellogenin
Receptor (VgR) and Vitelline Envelope (VE) cDNA clones.

 

Developmental Profile of E75 in Mosquito Ovary

140 I ““1 12
l +E7SA

120]l /§\I l-l-E753 10 a:
76100 4 ' \ ‘\ l— ‘5750 9 -
380i / / J_\\\L__Y_9_E_6 EE’
“'3 sol / / ~- % <
LU l 4 g 2

401; l o CE

2

20 4 “L 2

o l- 0

 

 

 

PV 6h 12h 18h 24h 30h 36h 42h 48h
Tlme after a Blood Meal

Fig. 8. Accumulation of AaE75 mRNA in the mosquito ovaries, quantified by
phosphorimager analysis of Southern blotted RT-PCR products (see Materials and
Methods). The results represent means of three determinations i S.E. Shown also are
the results of a RT-PCR with VgR, which serves to illustrate the course of ovarian
development.

92

termination of vitellogenesis and cessation of yolk accumulation by the ovary, and was
followed by a more gradual decline over the next 12 hr, reaching low levels by 48 hr
PBM. The AaE75C showed only a modest increase with a small peak at 24 hr PBM (Fig.
7). The onset of high AaE 75 expression in the vitellogenic ovary closely paralleled that
of the ecdysone-inducible VE genes. However, the VE gene transcripts remain in the
ovary at peak levels well beyond the peak of AaE 75 expression (Fig.7). The overall
expression profiles of AaE 75 transcripts resembled even more closely the vitellogenic
profile of VgR expression (Sappington et al., 1996), which increased to a peak at 24 hr
PBM, and then declined during the latter portion of vitellogenesis (Fig. 7). In contrast to
AaE 75, however, VgR was also expressed at a low level in previtellogenic ovaries (not
shown). The quantitative analysis of the temporal expression of AaE 75 mRNA levels in
the vitellogenic ovary, performed using the RT-PCR analysis as described for the fat

body, is in agreement with the Northern blot analysis (Fig. 8).

Response of AaE 75 to ecdysone in vitro

The pattern of AaE 75 transcript kinetics in the mosquito ovary and fat body
following a blood meal suggests that ecdysone produced in response to a blood meal is
likely responsible for the induction of the AaE 75 gene. To determine whether AaE 75
induction following a blood meal is ecdysone-mediated, we have tested the in vitro effect
of 20-hydroxyecdysone on the appearance of AaE 75 mRNA in isolated previtellogenic
female fat body, under conditions previously demonstrated in vitro induction of the VCP
gene (Deitsch et al., l995a) (Fig. 9). As it has been previously established (Deitsch et al.,

1995a), the levels of the V CP mRNA increased in response to increasing concentration of

93

C.)
01

800

700 CI E75A

 

 

 

 

 

 

- so 7;

a BE758 3
E 600 BE75C . 25%
g 500 IActin %
m .ch 20 >
."2 on
1.1.1 400 c
'g 15 '8
.5 300 <
‘6 1:
E 10 .315
g 200 E
100 . 5 z

0 _-_.=n J—Eh: 49m , 0
PV 0 10‘ 10‘

208 (M)

Fig. 9. AaE75 response to 20-hydroxyecdysone in in vitro fat body organ culture.
The transcriptional response of the mosquito E75 homologue to 20-hydroxyecdysone was
examined in an in vitro culture system of female previtellogenic fat body. Total RNA
also was extracted from fat bodies dissected from previtellogenic (PV) mosquitoes to
provide a control for the in vitro system. RT-PCR and Southern hybridization were
performed with isoform-specific probes for AaE75A, AaE7SB, or AaE75C (see Materials
and Methods, and Fig. 6). RT-PCR for VCP provided a positive control, while Actin was
used as control for loading.

 

 

 

 

 

 

 

 

 

 

 

 

 

700 35
__ 600 DE75A 30
2 135758 .5
e: 0
_l 500 . SE75C . 25 <
g IActin 06 '6
3‘, 400 Ich 20 8 e):
in > -‘
'0 300 15 g ‘2‘
£3 .5 a:
'5 3
g 200 10 E
o O
_ . z
- - + + 20 E (10‘ M)
- + — + CHX(10'°M)

Fig. 10. Effect of cycloheximide (Chx) on the activation of the AaE 75 gene. The
previtellogenic fat bodies were first pre-incubated in the culture medium for 1 hr in the
absence or presence of 10'5 M Chx, and then continued incubating with the same dose of
Chx but with the addition of 10'6 M 20E for 2 hr. RT-PCR and Southern hybridization
were performed with isoform-specific probes for AaE75A, AaE75B, or AaE75C (see
Materials and Methods, and Fig. 6).

94

 

 

3w

 

 

 

 

 

 

 

 

a 600 .
3 500 .
.. 400 ‘
E 300 <
O
z 200 1

100

o . -
PV 1 2 4 a

(a) Culture Time (h)

1200

+E7sA
'°°° ‘ +5758

 

+ E75C
-_______l-"‘W

 

 

 

o 4 -—-—'"—""—' ‘ ' . e: . 4L . :1
PV 1 2 4 6
(b) Culture Tlme (h)

Fig. 11. Time course of AME 75 induction in in vitro fat body organ culture. Total
RNA was extracted from previtellogenic (PV) fat bodies or fat bodies incubated in the
presence of ZO-hydroxyecdysone for the indicated times. RT-PCR and southern blotting
were performed with isoform-specific probes for AaE75A, AaE7SB or AaE75C (see
Section 2, Fig. 6). VCP was included as a positive control. The results represent the
means of three independent experiments +SE: (a). RNA accumulation of E75A, E7SB,
E75C and VCP during exposure to 10 '7 M ZO-hydroxyecdysone; (b) RNA accumulation
of E75A, E753, E75C and VCP during exposure to 10 '7 M ZO-hydroxyecdysone and

10 '5 M cycloheximide

95

20-hydroxyecdysone, reaching maximum at 10’6 to 10'5 M of the hormone in the medium
(not shown). The levels of actin mRNA remained relatively constant. The mRNA levels
of all three AaE 75 isoforms increased in a dose-dependent manner in response to 20-
hydoxyecdysone. The levels of AaE75A and AaE7SB were elevated significantly, while
only small increase was observed for AaE75C mRNA. Importantly, the AaE 75
transcripts reached their maximal levels at the concentration of 20-hydroxyecdysone in
the culture medium at least 10-fold lower (10'7 M) than that for the VCP and Vg genes
(Fig. 9).

To determine the requirement of protein synthesis for activation of the E 75 early
gene in the mosquito fat body by 20E, the previtellogenic fat bodies were incubated with
cycloheximide (Chx), a protein synthesis inhibitor which reversibly represses protein
synthesis in this tissue in a dose-dependent manner (Deitsch et al., l995a). The fat
bodies were first pre-incubated in the culture medium for 1 hr in the absence or presence
of 10'5 M Chx, and then continued incubating with the same dose of Chx but with the
addition of 10'6 M 20B for 2 hr. Transcripts for all three AaE 75 isoforms appeared in
response to 2013 in the presence of Chx, in contrast to the V CP gene, which was inhibited
by Chx (Fig 10). Incubation of the non-stimulated, previtellogenic fat bodies in the
presence of Chx alone activated neither the AaE 75 gene nor VCP (not shown). The time
course of response also differentiated between AaE75 and VCP. AaE75 transcripts are
induced rapidly and reach a peak by 2—4 h before beginning to regress. In contrast, VCP
transcripts continue to accumulate to 6 h and beyond in response to hormonal stimulation.

When the accumulation of early gene protein products is blocked by cycloheximide, both

96

the markcq‘

blocked (F

A (‘I_
On
been idem:
unpublishe
promoter, (

AaETSA w

 

containing
activated 11
increased i
(Fig-l2). 1
30E USlnt:y

Ne
the largct
EXPTCSSjO
With the ‘
Signll‘ica,
EcR,-US}
amouM (

present \

the marked regression of early gene expression and the increase in VCP expression are

blocked (Fig. 1 1).

Activation of the Vg promoter by E 75 in 52 cells

On the Vg promoter, a number of potential transcription factor binding sites have
been identified including E75 and EcR/U SP (Martin et al., 2001; Martin and Raikhel,
unpublished). To examine whether E75 is capable of inducing activation of the Vg
promoter, cell transfection assays were performed. The expression plasmid carrying
AaE75A was transfected into Drosophila 82 cells along with the reporter plasmid
containing the Vg-promoter-luciferase. Either in the presence or absence of 20E, E75A
activated the reporter gene to more than two fold (Fig. 12, lOOng), and the activation
increased in a dose-dependent manner in response to increasing amounts of E75A
(F ig.12). Interestingly, higher levels of activation were detected without 20E than with
20E using this 2.1kb Vg promoter construct (Fig. 12, -ZOE and +20E).

Next, to investigate whether E75 affects the EcR/USP-induced transactivation of
the target Vg gene, which has been reported previously (Martin et al., 2001), the
expression plasmids carrying EcR/U SP and E75A were co-transfected into SZ cells along
with the Vg-luciferase plasmid. In the presence of 20E, the reporter gene activities were
significantly enhanced when both EcR/USP and E75A were present, compared with
EcR/U SP or E75A alone (Fig. 12). This synergistic activation was dependant on the
amount of E75A, exhibiting stronger activation when higher amounts of E75A were

present with constant amount of EcR/USP (l lng) (Fig. 12).

97

 

200

O
5
4|

0
0

1

33:3. .03..

0
5

 

 

 

 

 

 

 

O 0.4 1.2 3.7 1 1 33 100
E75A amount (ng)

20E

F’

Fig.12. Action of E75 and EcR/U SP on the Vg Promoter.

Increasing amounts of the expression plasmid for AaE75A were
cotransfected with the 2.1kb-Vg promoter-luciferase construct and a constant
amount of AaEcR/AaUSP plasmids (+EcRUSP) into Drosophila 82 cells in
the presence (+20E) or absence of 20E (-20E). Reporter gene activities were
measured after 1 day culture (see Materials and Methods).

 

 

 

 

Vg Promoter

 

98

 

vitellocm

Tl

 

molecular‘
control of
1992 t. L':
09mm hi

identificc

transcri p

These results indicate that the early gene E 75 may be involved in the activation of
the Vg gene and be part of the ecdysone gene regulatory hierarchy during mosquito

vitellogenesis.

DISCUSSION

The yellow fever mosquito Aedes aegypti is an ideal system for investigating the
molecular basis of ecdysone action in insect reproduction, because of the stringent
control of egg maturation and vitellogenic events triggered by a blood meal (Raikhel,
1992). Using PCR-based strategies to isolate genomic and cDNA clones, the Aedes
aegypti homologue of the Drosophila early ecdysone-inducible gene E75 has been
identified. Its genomic organization, including the structure of the three overlapping
transcription units AaE75A, AaE7SB, and AaE75C, is highly conserved, with the exon-
intron structure of all three transcription units closely resembling that seen in Drosophila
(F ig.1). By contrast, in lepidoptera there is an altered structure of the E758 transcription
unit, in which the E7SB-specific region is fused to exon 2, and E75C transcripts have
never been observed (Zhou et al., 1998; Segraves and Woldin, 1993; Jindra et al., 1994a).
These differences are paralleled by a striking difference in the length of both the E75A
and E758 transcript units. In Drosophila and mosquito, the E75A transcript units, are,
e.g. 50 and >6Skb, respectively, comparing to ~16 kb in Manduca (Zhou et al., 1998)

A similar relationship is seen for the E75 proteins, in which the conservation of
E75 sequences among species is consistent with their phylogenetic relatedness. For
example, within the largest block of conserved sequence, the E region, a comparison of

AaE75 to E75 from Drosophila, another dipteran, shows approximately half as much

99

dit'ergen
proteins t
fitmlferw
each shox
family of

Ti
conscrx'at
resulting i

consemm

 

 

Conserved
Smaller an
1993; Jind
and AaE7:
Proteins b

Clean), C0:

These Obs
lSofOrmS Z
fUnQnOnaj

Ea
OfuleuoE
eCdFSOnQ

Critical to

divergence (93% identity) as the comparison of either diperan E75 protein to E75
proteins of lepidopteran Manduca Sexta, Galleria mellonella and Chortstoneura
fumiferana (87% identity). The mosquito, Drosophila and lepidopteran E75 E regions
each show approximately 36—37% identity within the E region to the Rev—erb/RVR
family of mammalian E75 homologues (Miyajima et al., 1989; Lazar et al., 1990).

The comparison between AaE75 and Drosophila E75 also shows clear
conservation of each of the three independent isoforms; the AaE75A, AaE7SB and
AaE75C-specific amino termini each show several clusters of conserved amino acids,
resulting in overall identity of 31, 30 and 33% respectively. While evolutionary
conservation of E75A- and E7SB-specific sequences has been seen in Manduca, the
conserved regions consisted of only a small number of amino acids within the much
smaller amino terminal domains of the lepidopteran E75 proteins (Segraves and Woldin,
1993; Jindra et al., 1994a; Palli et al., 1997; Zhou et al., 1998). In contrast, the AaE75A-
and AaE7SB-specific regions much more closely resemble those of the Drosophila
proteins both in size and amino acid sequence. The AaE75C-specific region is also
clearly conserved, and represents the first E75C isoform seen outside of Drosophila.
These observations extend our knowledge of the evolutionary divergence of E75
isoforms and support the hypothesis that these represent evolutionarily conserved
functionally independent forms of the E75 protein.

Each of the three AaE 75 transcription units is expressed in' the fat body and ovary
of vitellogenic female mosquitoes, and in vitro culture experiments indicate that AaE 75 is
ecdysone-inducible in isolated fat body. These findings suggest that E 75, which plays a

critical role in the control of ecdysone response during Drosophila development, is also

100

involved
gene ma}
signaling
be expres-
ecdysone
govemed

In

onset of \'

 

 

Peak (Ha:
increase (1
gradually
20E lCVel;
Hou‘eycr‘
3CCUmula
in the ind
b69n fOUI

involved in the control of reproductive ecdysone response in the mosquito. The AaE 75
gene may be one of the key genes responsible for controlling downstream ecdysone
signaling in the regulation of vitellogenic genes in the mosquito. Insofar, E 75 appears to
be expressed and required for all ecdysone responses; it may thus be an integral part of an
ecdysone response cascade used reiteratively in developmental and reproductive events
governed by this hormone.

In the fat body (Fig. 6), E 75 transcripts exhibit a sharp rise immediately after the
onset of vitellogenesis reaching a peak at 3 hr PBM when ZOE titer shows a first small
peak (Hagedom etal., 1975). In contrast, the YPP transcripts show only a small gradual
increase during first 4 hr PBM. After the initial drop, levels of AaE 75 transcripts
gradually increase again reaching second peak of their expression at the 24 hr PBM when
20E levels have passed their major peak, at 18 hr PBM (Hagedom et al., 1975).
However, the second gradual increase in AaE 75 transcripts closely correlates with the
accumulation of RNA for YPPs in the fat body, suggesting that AaE 75 may be involved
in the indirect regulation of these genes by 20E (Fig. 6). Indeed, E75 binding sites have
been found in the regulatory regions of mosquito YPP genes, Vg and VCP (D. Martin and
A. Raikhel, unpublished observation).

While the transient early induction of AaE 75 may readily be explained in the
context of E 75's role in Dr050phila developmental ecdysone response hierarchies, the
sustained later response during the period of peak vitellogenesis is less easily understood.
Elucidating the conditions, which allow this sustained response of the AaE 75 gene will
no doubt be critical to our understanding of the hormonal regulation of developmental

events which require prolong expression of genes at a high level, such as vitellogenic

IO]

genes in 1
high levc
ecdysone
vitellogei

I I;
correspon
induction

that AaE 7

 

tissues in
al,1997;!
1996‘. Zhu
mOSquito
ImPOrtant
VitellOng
‘M \‘8.10"
participm
le\'€l of C
fiI‘SI AUE
are harui.
Mean}? 2
Die/0,20%
requiri‘m

Oflhe 1a]

genes in the mosquito fat body. Interestingly, AaE7SB reinduction reaches a particularly
high level. Recent studies of White et al. (1997), showing the role of 13758 in attenuating
ecdysone response, suggest that AaE7SB may be acting here to help terminate some
vitellogenic events in the mosquito fat body.

In vitro organ culture experiments have demonstrated that 20E induces transcripts
corresponding to all three AaE 75 isoforms in the previtellogenic fat body. The 20E
induction of the AaE 75 gene was independent of protein synthesis, thus demonstrating
that AaE 75 acts as an early gene in ecdysone hierarchy, as previously seen for larval
tissues in other insects (Segraves and Hogness, 1990; Jindra and Riddiford, 1996; Palli et
al., 1997; Zhou et al., 1998). Unlike Galleria and Manduca E 75 (Jindra and Riddiford,
1996; Zhou et al., 1998), in which E75A exhibited higher sensitivity to 20E than E7SB,
mosquito E 75 isoform transcripts show similar sensitivity to this hormone in vitro.
Importantly, however, AaE 75 is ten-fold more sensitive to 20E induction than
vitellogenic genes, VCP and Vg (Fig. 10). AaE 75 transcripts were induced by 20E at 10'
7M vs. 10'6M required for in vitro induction of VCP and Vg, suggesting that AaE 75 may
participate in activation of genes important for vitellogenic events preceding the high
level of expression of YPP genes. This suggestion is agreement with the early time of the
first AaE 75 expression peak in the fat body in vivo, when the transcripts for YPP genes
are hardly detectable. Previously, a similar concentration-dependent pattern of activation
of early and late chromosome puffs and genes by 2013 has been observed in Drosophila
melanogaster (Karim and Thummel, 1992; Russell and Ashbumer, 1996), reflecting the
requirement for a threshold level of early gene activation required for maximal induction

of the late genes.

[02

by 24 hr
of .405 75
response
transcript
and AaE—
considera
body, wh‘.
the ovary

rthiireme

 

“19 semi-i
exprCSSiQr

Th
abOut its l"-

SallVar}.

‘1

(-

(I

decline as
unUSualh,

plaUSibic .

As in the fat body, the highest level of AaE 75 expression in the ovary was reached
by 24 hr PBM. More sensitive RT-PCR analysis indicates a much earlier detectable level
of AaE 75, as early as 1 hr PBM, in conjunction with the initial onset of ecdysone
response. However, in contrast to the fat body, there was no evident early AaE 75
transcript peak in the ovary soon after the onset of vitellogenesis. In addition, AaE75A
and AaE7SB transcripts were present in the ovary at the high levels until 42 hr PBM,
considerably longer than in the fat body. It is important to remember that unlike the fat
body, which is relatively homogeneous tissue consisting predominantly of one cell type,
the ovary contains both somatic and germ-line tissues which may have distinct AaE 75
requirements. Indeed, the kinetics of expression of the ecdysone-inducible genes, VgR
the germ-line cells and VE in the follicular epithelium, are different. Thus, the
expression patterns of AaE 75 transcripts in these ovarian tissues may differ as well.

The timing of AaE 75 expression within the ovary raises interesting questions
about its position within the reproductive ecdysone response hierarchy. In Drosophila
salivary glands, E 75 transcripts peak within approximately 2-4 hr, and then begin to
decline as a result of negative autoregulation. One factor proposed to influence the
timing of ecdysone response in Drosophila is the time required to transcribe the
unusually long early genes, E74, BR-C and E 75 (Thummel, 1992). It also seems
plausible that the full progression of vitellogenic ecdysone response may be most closely
analogous to the series of ecdysone responses in the Drosophila third instar larva and
prepupa, in which each ecdysone pulse results in the synthesis of factors required to set
the stage for next (Hagedom et al., 1975), rather than to a single discrete ecdysone

response.

l03

levels oi'I
acting as
during ti:
equivale:

broader t

 

 

 

 

autoregul
Suggests

gene isoft
difference
\vithin cm
active dur
Ofecdygo
during 00
lWham [Q
relam'e u.

have dl lTC'

While these factors may contribute to the relatively slow appearance of high
levels of AaE 75 transcripts, the broad, sustained response suggests that AaE 75 may be
acting as an early-late in this system, such that it is further induced rather than repressed
during the late ecdysone response. Hierarchically, the early and early-late genes are
equivalent (Woodard et al., 1994); however, the early-late genes are expressed over a
broader time frame than the transiently expressed early genes which are subject to
autoregulatory repression. Comparison between different tissues in the Drosophila larva
suggests that there can be substantial differences in the kinetics with which various early
gene isoform levels rise and fall (Huet et al., 1993; I995); presumably, these reflect
differences in the temporal requirements of these genes for the regulation of target genes
within each tissue. Comparison of AaE 75 expression to the expression of other genes
active during the peak of vitellogenesis suggests that it may be required for the regulation
of ecdysone induced genes, such as vitelline envelope protein, for a sustained period
during oogenesis. The specific AaE 75 isoforms which might be involved in this process
remain to be determined; although AaE75C expression is low in the ovary as a whole
relative to fat body, the ovary consists of both somatic and germ line tissues which may
have different AaE 75 requirements.

Despite these intriguing differences, the present findings suggest that elements of
20E-triggered regulatory hierarchies are used reiteratively during development and
reproduction. Indeed, E 75 is thus far expressed and required for all characterized 20E
responses; it may be an integral part of all or nearly all 20E-responsive regulatory
hierarchies. Parallel genetic studies in Drosophila support the suggestion that

reproductive and developmental ZOE responses are controlled by closely related primary

l04

response

BR-C C;

et al.. i9

 

mid-00g
in mosqt
reproduc
strengths
possible i
ofinsect

CVolved.

’7-

genOmic
Brad D,
Slipping
ArizOn:
Support

WAS

responses. Of the three transcription factor-encoding classical early genes, E 75, E74 and
BR-C, each is involved in the regulation of oogenesis as well as development (Buszczak
etal., 1999). Remarkably, E 75 and E74 appear to be involved in progression through a
mid-oogenesis checkpoint, which may be analogous to the previtellogenic ovarian arrest
in mosquitoes, suggesting that the fundamental mechanisms underlying the control of
reproductive 20E response may be highly conserved. By combining the complementary
strengths of A. aegipti physiological studies and Drosophila genetics, it should be
possible to dissect the mechanisms of through which 20E and E 75 control critical aspects
of insect reproduction, and to understand the ways in which these mechanisms have

evolved.

ACKNOWLEDGMENTS

I thank William Pierceall and William Segraves at Yale University for cDNA and
genomic cloning of AaE 75, Avraham Biran for Northern blot. I also thank Judy Cole and
Brad Dunlap for assistance in the isolation and analysis of genomic clones, Dr. Tom
Sappington for critical reading of the manuscript, and Dr. H. Hagedom (University of
Arizona) for the gift of a mosquito vitelline envelope gene probe. This work was
supported by a grant from the National Institutes of Health (A136959) to A.S.R. and

W.A.S. W.E.P was supported in part by the Seessel Anonymous Fund.

105

Chapter 4*
Cloning and Characterization

of the Mosquito Early-late Gene HR3

*Originally published in Mol. Cell. Endocrinol.
Ref: Kapitskaya, m, 2., Li, g“, Miura, 1c, Segraves, w., and Raikhel, A. s. (2000).
Mol. Cell. Endocrinol. 160, 25-37.

# Underlined two authors contributed equally to this work.

106

 

controllii
metamon‘_
understai
deveIOpn‘
important
molecular
meal-actii
hierarchy

eXpressioi

 

been prev
regUlaIOQ
representc
”‘6 reprod
COnsenan
regulating
AHR3, a h
inducible 3
female mC
ec9).-”Stem,

culture ex;

acme ecdv

ABSTRACT

The insect steroid hormone, 20-hydroxyecdysone (20E), is a key factor
controlling critical developmental events of embryogenesis, larval molting,
metamorphosis, and, in some insects, reproduction. Our lab is interested in
understanding the molecular basis of the steroid hormone ecdysone action in insect egg
development. The yellow fever mosquito, Aedes aegypti, in addition to being an
important vector of human diseases, represents an outstanding model for studying
molecular mechanisms underlying egg maturation due to stringently controlled, blood
meal-activated reproductive events in this insect. To elucidate the genetic regulatory
hierarchy controlling the reproductive ecdysone response, ecdysone-regulated gene
expression in vitellogenic mosquito ovaries and fat bodies has been investigated. It has
been previously demonstrated the conservation of a primary ecdysone-triggered
regulatory hierarchy, implicated in development of immature stages of Drosophila,
represented by the ecdysone receptor/Ultraspiracle complex and an early gene E 75 during
the reproductive ecdysone response (Chapter 2, 3). The present study demonstrates that
conservation of the factors involved in the ecdysone-responsive genetic hierarchy
regulating female reproduction extends beyond the early genes. Here, it is a report on
AHR3, a highly conserved homologue of the Drosophila HR3 early-late ecdysone-
inducible gene in the mosquito. AHR3 is expressed in both vitellogenic tissues of the
female mosquito, the fat body and the ovary. The expression of AHR3 correlates with the
ecdysteroid titer, reaching a peak at 24 hr after a blood meal. Moreover, in vitro fat body
culture experiments demonstrate that the kinetics and dose response of AHR3 to 20E, an

active ecdysteroid in the mosquito, is similar to those of the late vitellogenic genes rather

l07

than the
activatit
inhibitor

regulatir‘

T
dCVClOpn
some insl

Analysis

 

 

 

larval ant
Small nUl‘
large Hun
hierarcm
hcterodir.
the reunc
FunhL’rm
Sindeed
IranSCn-p
Segraves
01., 1991
UE[\,\-Ork

”D the m

than the early E 75 gene. However, as shown for other early and early-late genes, the 20E
activation of AHR3 is not inhibited by the presence of cycloheximide, a protein synthesis
inhibitor. Taken together, these findings strongly suggest AHR3 involvement in

regulating the vitellogenic response to ecdysone in the adult mosquito.

INTRODUCTION

The insect steroid hormone 20—hydroxyeedysone (20E) regulates multiple
developmental processes including embryogenesis, larval molting, metamorphosis and, in
some insects, reproduction (Bownes, 1986; Riddiford, 1993a,b; Thummel, 1995).
Analysis of the ecdysone effect on polytene chromosome puffing patterns in the late
larval and prepupal salivary gland of Drosophila suggested that the initial activation of a
small number of early ecdysone-inducible genes leads to subsequent inducement of a
large number of late target genes (Ashbumer et al., 1974). Elucidation of this genetic
hierarchy at the molecular level has led to the identification of the ecdysone receptor as a
heterodimer of two nuclear receptors, the ecdysone receptor (EcR) and the homologue of
the retinoid X receptor, Ultraspiracle (USP) (Koelle et al., 1991; Yao et al., 1992; 1993).
Furthermore, these studies have shown that the action of the ecdysone-receptor complex
is indeed mediated by early genes such as the BR-C, E74 and E 75 genes encoding
transcription factors involved in regulation of late gene expression (Burtis et al., 1990;
Segraves and Hogness, 1990; Thummel et al., 1990; Guay and Guild, 1991; DiBello et
al., 1991; Fletcher and Thummel, 1995). The ecdysone-mediated genetic regulatory
network is further refined by the presence of a set of genes which are involved in setting

up the timing and stage—specificity of gene activation by this genetic hierarchy

108

(Thummel, 1997). The orphan nuclear receptor DHR3, an early-late gene with induction
characteristics including properties of both early and late genes, is one of the key genes in '
the ecdysone-mediated genetic regulatory network. Recent studies have revealed that
during insect metamorphosis DHR3 has a dual role in repressing the early genes while
activating another orphan nuclear receptor, flF TZ—F 1, which in turn provides competence
for the stage-specific late prepupal response to 20E (Woodard et al, 1994; Lam et al,
1997; White et al, 1997; Broadus et al., 1999). DHR3 homologues cloned from several
Lepidopteran insects are highly conserved and their expression suggests HR3
involvement in mediating ecdysone response during Lepidopteran development and
metamorphosis (Palli et al., 1992; I996; Jindra et al., 1994b). Recent analysis of
Drosophila DHR3 mutants has revealed that DHR3 function is, in addition, required to
complete embryogenesis (Carney et al., 1997), prepupal—pupal transition and
differentiation of adult structures during Drosophila metamorphosis (Lam et al., 1999).
In contrast to the wealth of information concerning the molecular basis of the
ecdysone-mediated genetic hierarchy during development and metamorphosis,
comparatively little is known about the ecdysone-regulated aspects of insect female
reproduction, oogenesis and vitellogenesis. In Drosophila, ecdysone involvement in
female reproduction has been suggested by the female sterility of mutations affecting
ecdysone synthesis and ecdysone response (Garen et al., 1977; Oro et al., 1992), by the
ecdysone-inducibility of yolk protein genes (Bownes, 1986) and most recently by studies
implicating the ecdysone response hierarchy in egg chamber maturation (Buszczak et al.,
1999). The role of ecdysone in female reproduction is best documented in anautogenous

mosquitoes, especially the yellow fever mosquito, Aedes aegypti. In anautogenous

109

mosquitoes, a blood meal is required for egg maturation. Ingestion of blood stimulates
the brain to secrete the ecdysiotropic neuropeptide, egg development neurosecretory
hormone or ovarian ecdysteroidogenic hormone (Lea, 1967; 1972; Matsumoto et al.,
1989; Brown et al., 1998). These hormones activate the ovary to produce ecdysone,
which controls the synchronous progression of egg development in the ovary and the
synthesis of yolk protein precursors by the fat body. In A. aegypti, ecdysteroid levels are
relatively low during the first 8 hr post blood meal (PBM), with only a small peak at 4 hr.
Later, the ecdysone level rapidly increases to the maximum at 18-24 hr PBM, then
abruptly declines to the previtellogenic level in preparation for the termination of
vitellogenesis at 30-32 hr PBM (Hagedom, 1983; 1985; Raikhel, 1992; Dhadialla and
Raikhel, 1994). Action of 20-hydroxyecdysone on the mosquito yolk protein genes,
vitellogenin (Vg) and vitellogenic carboxypeptidase (VCP), requires protein synthesis,
suggesting indirect control by the genetic regulatory cascade involving early genes
(Deitsch et al., 1995a,b). In the mosquito ovary, established functions associated with
ecdysone include the formation of the vitelline envelope (Raikhel and Lea, 1991; Lin et
al., 1993) and the separation of the secondary follicle in the ovary (Beckemeyer and Lea,
1980). Thus, the mosquito provides an ideal system to investigate control of ecdysone-
induced gene expression during female reproduction. Furthermore, the critical role of
anautogenous mosquitoes as disease vectors increases the significance of elucidating the
ecdysone-mediated gene regulatory hierarchy, a potential target in the development of
novel strategies for vector control.

In full agreement with the suggested role of ecdysone in regulating female

reproduction, ecdysone receptor transcripts can be detected in pre- and vitellogenic

llO

ovaries and fat bodies (Cho et al., 1995). Transcripts encoding two different USP
isoforms are differentially expressed in these tissues (Kapitskaya et al., 1996; Wang et
al., 2000a). The mosquito EcR/U SP heterodimer has been shown to bind to various
ecdysone response elements (EcREs) to modulate ecdysone regulation of target genes
(Wang et al., 1998). Recently, the A. aegypti homologue of the Drosophila E 75 gene, a
putative representative of the next level in the ecdysone response hierarchy, has been
identified and characterized. A. aegypti E 75, AaE 75, is expressed in ovary and fat body
following a blood meal. AaE 75 transcripts, corresponding to three E75 isoforms, are
ecdysone-inducible in isolated fat body cultured in vitro. These findings suggest that
AaE 75 mediates ovary and fat body ecdysone responses, and that ecdysone-triggered
regulatory hierarchies such as those implicated in the initiation of metamorphosis may be
reiteratively utilized in the control of the reproductive ecdysone response (Pierceall et al.,
1999).

The present study demonstrates that conservation of the factors involved in the
ecdysone genetic hierarchy regulating female reproduction extends beyond the early
genes. It is shown here that a highly conserved homologue of DHR3 is expressed in the
vitellogenic tissues of the female mosquito, the fat body and the ovary. The expression
of the mosquito homologue, which is named AHR3, correlates with the titer of 20E,
peaking at 24 hr PBM. Furthermore, 20E activates the AHR3 transcript in the
previtellogenic fat body in vitro; this activation is not inhibited by cycloheximide, a
protein inhibitor, suggesting that AHR3 is acting as an early-late gene in this system.
Taken together, these findings strongly suggest AHR3 involvement in regulating the

vitellogenic response to ecdysone in the adult mosquito.

Ill

MATERIALS AND METHODS

Animals

Mosquitoes, A.aegypti, were reared according to Hays and Raikhel (1990).
Larvae were fed on a standard diet as described before (Lea, 1964). Vitellogenesis was

initiated by allowing females 3-5 days after eclosion to feed on an anesthetized white rat.

Materials

The RNA ladder was purchased form Life Technologies, Inc.; Sequenase was
from U. S. Biochemical Corp.; restriction enzymes were from Boehringer Mannheim. All
primers used in PCR were ordered from Gibco-BRL. Superscript II reverse transcriptase
and Taq DNA polymerase were from Gibco-BRL; random hexamer primers from
Promega. Perkin-Elmer was the source of reagents for the polymerase chain reaction
(PCR). MSI CO supplied nitrocellulose-blotting membranes. Radionucleotides for
labeling of probes and DNA sequencing were from NEN life Science Products. All other

reagents were of analytical grade from Sigma or Baker.

Cloning, sequencing, and 5 ’RA CE of cDNA

The cDNA fragment of AHR3 was first obtained by PCR, for which degenerate
primers were designed based on the sequences of its homologues in other insect species.
Amplification was achieved in a Perkin Elmer thermal cycler using the template cDNA
reverse-transcribed from 20 pg of total RNA prepared from the fat bodies of vitellogenic

female mosquitoes. The PCR-generated fragment was used as a probe to screen a

112

XZAPII cDNA library, which was prepared from the fat bodies of vitellogenic female
mosquitoes 6-48 hr post-blood meal (PBM) as previously reported (Cho and Raikhel,
1992). Several positive cDNA clones were subsequently isolated and sequenced in W.M.
Keck Foundation Biotechnology Resource Laboratory at Yale University. Analyses of
nucleotide and deduced amino acid sequences were performed using the software of
GCG (Genetics Computer Group, University of Wisconsin, Madison).

5’ RACE (Rapid Amplification of cDNA Ends) was performed with a 5’ RACE
kit from GIBCO/BRL using the manufacturer’s protocol. The first strand of cDNA for
the mosquito HR3 gene was prepared with the gene-specific primer HR3-SP1:
ATTTGCGCTAACATGCTATCG and total RNA from post blood meal 24 hr fat bodies.
After the first round PCR, the gene-specific primer HR3-SP2:
CAGCCATTTCAAGTTCACTACG for HR3 was used for the second round nested PCR.
The PCR fragments positively confirmed by Southern Blot hybridization were then

subcloned into pGEM-T-easy vectors (Promega) and sequenced.

Northern blot hybridization

Total RNA was isolated from mosquitoes of different stages and tissues using the
guanidine isothiocyanate method as described previously (Bose and Raikhel, 1988).
Polyadenylated mRNA was isolated using Niomag oligo (dT)20 magnetic beads and the
manufacturer’s protocols (PerSeptive Diagnostics, Inc.). For Northern blot analysis, total
or poly(A)+ RNA was separated by electrophoresis in 1.2% agarose/formaldehyde gels in

MOPS buffer, blotted to a nitrocellulose membrane, and hybridized with 32P-labed DNA

113

probes under high stringency conditions. Autoradiography was conducted at -80°C using

intensifying screens.

R T-PCR/Southern blot analyses

The temporal profile of transcript abundance of HR3 in the mosquito fat body was
examined by RT-PCR followed by Southern blotting with the gene-specific radioactive
probe. Total RNA was prepared from fat bodies throughout the first vitellogenic cycle,
as described in Northern Blot above, with the modification that all isopropanol
precipitation steps were done without low-temperature incubation to avoid co-
precipitation of glycogen and salt. RNA concentration was determined
spectrophotometrically; Absorbance ratios, A260/A280 and A260/A230, in RNA
preparations were always above 1.7 and 2.0, respectively, quite consistent irrespective of
the developmental stages of mosquitoes. Five microgram aliquots of each RNA
preparation were reverse-transcribed by SuperscriptII reverse transcriptase (Gibco-BRL)
and random hexamer (Promega) in a reaction volume of 20 pl. After the reaction, the

volume of each sample was brought to 40 pl with TE buffer (10 mM Tris-Cl, lmM

EDTA, pH 8.0). cDNA samples were stored at -200C until use. In the case of mosquito
tissues, 0.025 fat body- or pupae-equivalent cDNA pools were used as PCR templates.

See Fig. l for the details ofthe primer design.
Thermal cycling conditions were as follows: the samples were incubated at 940C

for 3 min, followed by 18 cycles of 940C for 30 sec, 60°C for 30 sec and 720C for 30
see. As a reference to the developmental change of the fat body, the samples of the same

cDNAs were subjected to the 10 cycles of PCR with a primer pair specific to the

114

mosquito vitellogenic carboxypeptidase; the reaction was followed by Southern blotting
hybridization using a V CP specific probe generated by PCR using DNA from a VCP
cDNA clone as a template (Cho et al., 1991). The primers for V CP were: forward, 5’-
AGCGCCCATTCTTGTTTGG-3 ’ and reverse, 5 ’-CAGCTCATACAGGTATTCTCC-3 ’.
RT-PCR/Southem blotting of the total RNA preparations without reverse transcriptase
did not result in any appreciable signal, indicating that contamination of RNA
preparations with genomic DNA fragments was virtually negligible (not shown). The

radioactive blots were quantified with a phosphorimager (Molecular Dynamics).

In vitro fat body culture

To examine the effect of ecdysone on expression of the mosquito HR3 gene, the
abdominal walls with adhering fat bodies (hereafter referred to as the fat body) were
dissected from 3-5-day-old previtellogenic females and incubated in an organ culture
system as previously described (Raikhel et al., 1997). The fat bodies were incubated
with 20E either for 16 hr, or after 4hr with 20E followed by washing with the hormone-
free medium 3 times and 12hr continuing incubation in the hormone-free medium. At
every 4 hr interval, a sample of 9 fat bodies was collected and stored in liquid nitrogen
until use. Total RNA was extracted as described in RT-PCR above. As a control, RNA
was extracted from fat bodies incubated in the hormone-free medium. RT-PCR/Southem
blot analyses were utilized as described above. Hybridization using probes for the
ecdysone-inducible VCP gene was provided as a positive control.

To test the effect of the protein synthesis inhibitor cycloheximide (Chx) on

transcription, dissected 3-5 day old female mosquito fat bodies were pretreated with

115

culture media containing lmM Chx for 1 hr then further incubated with the same
concentration of Chx either without or with 106 M 20E. The Chx/hormone withdrawal
experiment was conducted as described for the hormone withdrawal experiment above

except that lmM Chx was added to all culture media.

RESULTS

Isolation and characterization of a cDNA clone encoding the mosquito AHR3
nuclear receptor

To obtain a mosquito HR3 clone, degenerate primers were designed, based on the
amino acid comparison of its insect homologues. The sense primer Pr-l conformed to
the conserved sequence QIEIIPCK adjacent to the C (DNA-binding) domain; the
antisense primer Pr—2 corresponded to the sequence EKVEDEVR from the conserved
part of the D domain of these nuclear receptors (Fig. 1). After a mosquito PCR-fragment
of the expected size, 285 bp, was sequenced and confirmed to encode a DHR3-related
polypeptide, it was used as a probe to screen the mosquito fat body cDNA library (see
Materials and Methods). The cDNA clone with the largest insert of 2.56 kb (designated
as AHR3 clone) was fully sequenced. This cDNA clone contained an open reading frame
(ORF) of 1.42-kb and 3’-untranslated region (3’-UTR) of 1.14 kb. It did not have in-
frame stop-codons at the 5’-end, indicating the ORF could potentially be truncated at the
5’-terminus; it also lacked a canonical site of polyadenylation and poly(A)-tail at the 3’-
end.

In the isolated AHR3 cDNA clone, there were 20 base pairs without any in-frame

termination codon located upstream of potential translation initiation site ATG. In order

116

Fig.1. Alignment of the deduced amino acid sequence of the mosquito AHR3 cDNA
with its insect homologues: DHR3 from Drosophila (Koelle et al, 1992), MHR3 from
Manduca (Palli et a1, 1992), and GHR3 from Galleria (Jindra et al, 1994b). Asterisks or
hats indicate amino acids, which are identical or similar, respectively. The borders of
DNA-binding domain (DBD) and putative ligand-binding domain (LBD) are marked
with square braces. The P-box in DBD is in bold. The carboxy-terminal extension
(CTE) of DBD, defined according to Giguere et. al. (1994), is double-underlined. The
conserved 8-amino acid stretch at the C-termini of LBD is in bold. Only part of A/B-
domains, adjacent to DBDs, are shown for DHR3, MHR3, and GHR3 (notice a high
percentage of identical amino acids between lepidopteran MHR3 and GHR3). The
vertical arrowhead marks a position from which a high similarity between different HR3
receptors starts. Horizontal double arrows indicate sequences corresponding to the
degenerate sense primer (Pr-1) and antisense primer (Pr-2) used for AHR3 cloning by
PCR. An arrow shows the sequence-specific antisense PCR-primer Pr-3, and two black
dots stand for two Xhol restriction sites of the AHR3 cDNA, which were utilized to
prepare a single-stranded DNA hybridization probe (see Fig. 3, 4). Two arrows indicate
the sequence-specific sense (Pr-4) and antisense (Pr-5) primers used for RT-PCR analysis
(see Fig. 5, 6, 7, 8). The sequence of the mosquito AHR3 has been submitted to the
GenBank (http;//www.ncbi.nlm.nih.gov1; its accession number is AF230281.

117

 

[——— DBD
Pr-l P154 P'box

1 MLRDAPNRSELEMAVSSTVFDSMLAQIEIIPCKVCGDKSSGVHYGVITCIGCKGFFRR 58

19 HANNLGQSNVQSPAGQNNSSGSIKAQIEIIPCKVCGDKSSGVHYGVITCIGCKGFFRR 77
73 SSEGMFGP-ISGMFMDKKAANSIRAQIEIIPCKVCGDKSSGVHYGVITCIGCKGFFRR 130
76 SSEGMFGSSISGMFMDKKAANSIRAQIEIIPCKVCGDKSSGVHYGVITCEGCKGFFRR 134

* **********************************

Pr-S Pr-2

DEBS

iiii

iiii

iiii iiii

iiii

iiii

SQSSVVNYQCPRNKQCVVDRVNRNRCQYCRLQKCLKLGMSRDAVKFGRMSKKQREKVEDEVRFH
SQSSVVNYQCPRNKQCVVDRVNRNRCQYCRLQKCLKLGMSRDAVKFGRMSKKQREKVEDEVRFH
SQSTVVNYQCPRNKACVVDRVNRNRCQYCRLQKCLKLGMSRDAVKFGRMSKKQREKVEDEVRYH
SQSTVVNYQCPRNKACVVDRVNRNRCQYCRLQKCLKLGMSRDAVKFGRMSKKQREKVEDEVRYH

***A********** ***********************************************A*

 

 

o
RAQMRAQSDAAPDSSVFDTQ--TPSSSDQLHHGGYNGY-A--Y-NNEV---GYGSP---YGYST
RAQMRAQSDAAPDSSVYDTQ--TPSSSDQLHHNNYNSY-SGGYSNNEV---GYGSP---YGYSA
KAQMRAQNDAAPDS-VYDAQQQTPSSSDQFH-GHYNGY-P ----------- GYASPLSSYGYNN

RAQMRAQTDAAPDS-VYDAQQQTPSSSDQFH—GHYNGYPG--Y-RSPLSSYGYGGN---AGPAL
A******A****** *A*A* *******A* A **A* **AA *
S----VTPQ -------- QTMGYDISADYVDSTTTYEPRST--IIDSDFISGHTEGDINDVLIKT
s—--—VTPQ -------- QTMQYDISADYVDSTT-YEPRST--IIDPEFIS-HADGDINDVLIKT
AGPPLTSNMSSIQAQPQAQQPY---—DYADSTTTYEPKQPG-YLDADFI-GQVEGDISKVLVKS
TSNMNIQPQAP ------ QQQPYDVSADYVDSTTAYEPKQTGGFLDADFI-GHAEGDISKVLVKS
A **A**** *it““ A* A** ***A **A*A

[— LBD

LAEAHANTNHKLEIVHDMFRKSQDVTRIMYYKNMSQEELWLDCAEKLTAMIQQIIEFAKLIPGF
LAEAHANTNTKLEAVHDMFRKQPDVSRILYYKNLGQEELWLDCAEKLTQMIQNIIEFAKLIPGF
LAERHANTNPKLEYINEMFSKPQDVSKLLFYNSLTYEEMWLDCADKLTQMIQNIIEFAKLIPGF
LAEAHANTNPKLEYVHEMFRKPPDVSKLLFYNSMTYEEMWLDCADKLTGMIQNIIEFAKLIPGF

*** ***** *** AAA** * *iAAAAA*AAAA **A*****A*** ***A***********

MRLSQDDQILLLKTGSFELAIVRMSRLMDLSTNSVLYGDIMLP-QEVFYTSDSFEMKLVACIFE
MRLSQDDQILLLKTGSFELAIVRMSRLLDLSQNAVLYGDVMLP-QEAFYTSDSEEMRLVSRIFQ
MKLTQDDQILLLKSGSFELAIVRLSRLIDVNREQVLYGDVVLPIRECVHARDPRDMALVSGIFE
MKLSQDDQILLLKSGSFELAIVRLSRLIDVNREQVLYGDVVLPIRECVHARDPRDVTLVVGIFD

*A*A*********A*********A***A*AA AA*****AA** A* AA * AA **A **A

Pr-3

TAKSIAELKLTETELALYQ LVLLWPERNGVRGNTEIQRLFEMSMSAIRQEIEANHAPLKGDVT
TAKSIAELKLTETELALYQSLVLLWPERNGVRGNTEIQRLFNLSMNAIRQELETNHAPLKGDVT
AAKSIARLKLTESELALYQSLVLLWPERHGVMGNTEIRCLFNMSMSAMRHEIEANHAPLKGDVT
AAKTIERLKLTETELALYQSLVLLWPERHGVRGNPEIQCLFNMSMAAMRHEIESNHAPLKGDVT

A**A* A*****A***************A** ** *** **AA**A*A*A*A*A**********

o
VLEILLNKIPTFRELSIMHMEALQKFKQDHPQYVFPALYKILFSIDSQQDLMT
VLDTLLNNIPNFRDISILHMESLSKFKLQHPNVVFPALYKILFSIDSQQDLT
VLDTLLAKIPTFRELSLMHLGALSRFKMTHPHHVFPALYKELFSLDSVLDYTHG
VLDTLLAKIPTFRELSLMHLEALCRFKAAHPHHVFPALYKILFSLDSVLDYTHG

*iA ** A**A**AA*AA*A A* A** **A ***********A** i

468
487
548
557

118

122
141
194
198

174
199
245
257

224
244
302
311

288
308
366
375

351
371
430
439

415
435
494
503

to identify the missing 5’-end of AHR3 cDNA, S’RACE PCR was carried out using total
RNA from the fat body 24 hr post blood meal (PBM) as a template (see Materials and
Methods). A 200-bp fragment was the major amplification product (data not shown). To
verify the absence of genomic contamination, 5’RACE was performed under the same
conditions except without superscript II reverse transcriptase. This control did not result
in any amplification product, indicating that the 200-bp PCR fragment originated from
mRNA. This fragment was subcloned into an pGEM-T-easy vector (Promega) and
sequenced. It shared an expected 36-bp sequence overlapping with the S’end of the
AHR3 cDNA clone. This observation confirmed that the 200-bp fragment was indeed a
portion of the mosquito HR3 cDNA. There were two in-frame termination codons, TAG
and TGA, located at the 45 bp and 42 bp upstream of the initiation site ATG, suggesting
that the 5’ end of the AHR3 ORF was complete.

Conceptual translation of the mosquito HR3 clone showed that it exhibited a
structural organization similar to that of other members in the HR3-subfamily of nuclear
receptors (F i gs.l and 2). The most conserved regions were the DNA-binding domain
(DBD) and so—called carboxy-terminal extension (CTE) of the DNA-binding domain
(Figsl and 2). As illustrated in Fig.1 and Fig.2, the mosquito AHR3 had the P-box
which shared 100%, 97% and 76% identical amino acids in the DBD, and 100%, 91%
and 74% identical amino acids in CTE with Drosophila HR3, Chorz'stoneura HR3 and
human homologue RORa, respectively. This indicated that the AHR3 and other HR3
receptors could bind to similar target DNA-sites.

A putative LBD of AHR3 did not show high similarity with those of the nematode

CeHR3 and human RORa counterparts, though it was still significantly conserved within

119

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

HI<——>I<——>I< >
AHR3 DBD! E H
31 66 23 114 234
DHR3 300%] g 33%
SO 66 23 115 233
MHR3 {2% g 61%
103 66 23 120 236
CHR3 -7%‘ g 48%
101 66 23 121 235
CeHR3 591% g
127 66 23 115 222
RON“

 

 

 

   

Fig. 2. Domain comparison of the mosquito HR3 with selected members of the
ROR/HR3 subfamily of nuclear receptors. Shown are the HR3 nuclear receptors:
AHR3 from Aedes aegypti (this study), DHR3 from Drosophila (Koelle et al, 1992),
MHR3 from Manduca (Palli et al, 1992), CHR3 from Chortstoneura (Palli et al, 1996),
CeHR3 from nematode Caenorhabditis elegans (Kostrouch et al, 1995), and human
RORal (Giguere et al, 1994). The letters above the mosquito HR3 protein denote
functional domains. The D-domain is divided into the CTE (carboxy-terminal extension)
and H (hinge) subdomains. Numbers below proteins indicate the number of amino acids
in each subdomain. The percentages of identities between corresponding domains of
AHR3 and other homologues are indicated within each domain; the ratios were calculated
as a number of amino acids identical between comparative sequences divided by the total
amino acid number of the shorter sequence.

120

insect LBDs (Fig. 2). Notably, there were 8 conserved amino acids in all members of the
HR3 subfamily at the C-terminus of LBD (shown in bold in Fig. 1). They contained a
stretch LYKELF similar to the AF-2 core motif (Beato et al, 1995), which strongly
suggested the functional significance of this sequence. The high similarity between
HR38 started in A/B-domains from the 7th amino acid preceding DBDS; however, the

rest of the HR3 A/B domain was poorly conserved in all members of the insect HR3.

The AHR3 transcripts and the time course of AHR3 expression in the mosquito

ovary and fat body during the first vitellogenic cycle

To determine the size of AHR3 transcripts, Northern analysis was conducted for
poly (A) RNA of the ovary and fat body isolated from females in the middle of the
vitellogenic stage (24 hr PBM), when female mosquitoes were exposed to a high
concentration of the steroid hormone 20E. Analysis of the ovarian RNA revealed the
presence of a major 7.0 kb transcript, and two minor transcripts of 9.5 kb and 5.9 kb in
size. In the vitellogenic fat body, the same three AHR3 transcripts were observed. In
contrast to the ovary, however, all three transcripts were expressed evenly in the fat body.
Interestingly, in pupae only the major 7.0 kb and minor 9.5 kb transcripts were present
(Fig. 3).

Northern blot analysis of AHR3 RNA in the ovary from female mosquitoes taken
at different stages of vitellogenesis showed a transcriptional pattern of AHR3 with a
prominent peak at 24-30 hr PBM, soon after a peak of 20E titer in the mosquito
hemolymph (Fig. 4).

Due to the relatively low abundance of AHR3 mRNA, the RT-PCR analysis was

used for elucidation of the AHR3 mRNA kinetics in the female mosquito fat body during

121

0v FB P

 

 

_
8.5 kb
9.5 kb — 7.0 kb — a
_ " N
7.0 kb — 5.9 kb it
5.9 kb '—

 

 

 

 

 

 

 

 

Fig. 3. AHR3 transcripts in the mosquito Aedes aegypti. Northern blot analysis of poly(A)
RNA extracted from ovaries (CV) and fat bodies (F B) of female mosquitoes, and whole body
pupae (P). A 0.82 kb XhoI-fragment of the AHR3 cDNA (marked by two black dots in Fig. 1)
was a template for a sequence-specific anti-sense primer Pr-3 (shown in Fig. l) to prepare a
hybridization probe by PCR. The resulting single-stranded DNA-probe of 0.68 kb
corresponded to the C-terminal part of the hinge-subdomain and the N-terminal part of the
LBD of the mosquito HR3.

122

PBM (h)
PV 6 1 2 1 8 24 30 36 42

 

 

9.” _

- - “Ir
7.0%.! -i n
5..“ ‘1

 

 

 

tubuacautidfivo.

 

Fig. 4. Expression of AHR3 in the mosquito ovary during the first vitellogenic cycle.
Northern blot analysis of total RNA extracted from ovaries of female mosquitoes
dissected before (PV) or after a blood meal (PBM) at the indicated times. Hybridization
was performed as described in Fig. 3. The bottom lanes present ribosomal RNA in the
same gel stained by ethidium bromide prior to blotting.

PV ((1) PBM Q1)
0-1 1-2 3-5 1 3 4 6 15 24 36 48

(A) .. C.-

 

 

 

 

 

 

....

Fig. 5. Expression of AHR3 in the mosquito fat body during the first vitellogenic
cycle. Total RNA, extracted from mosquito fat bodies, was reverse transcribed and
cDNA equivalents to 0.025 mosquito fat body were subjected to PCR, using primers
specific to either AHR3 or VCP. The resulting PCR-fragments were hybridized with 32P-
labeled gene-specific DNA-probes. (A) RT-PCR/Southem analysis of the AHR3
expression. The 0.22 kb PCR-fragment was generated with the AHR3-specific primers
Pr-4 and Pr—S (shown in Fig. 1). (B) RT-PCR/Southem analysis of the VCP expression.
VCP is a late gene regulated by 20E in the mosquito fat body (Deitsch et al., 1995a).

 

 

(B)

 

123

the first vitellogenic cycle. AHR3 mRNA was absent during the previtellogenic period,
the transcript could be detected only after the onset of vitellogenesis triggered by a blood
meal. A peak of AHR3 expression was clearly seen at 24 hr PBM. Determination of the
VCP mRNA levels in the same samples clearly showed that this expression of AHR3

mRNA coincided with the peak expression of yolk protein genes (Fig. 5).

In vitro 20E induction of AHR3 mRNA in the female mosquito fat body

To determine whether the positive correlation between the AHR3 expression and
the peak of ecdysteroids observed in vivo reflected the involvement of 20E in regulation
of expression of the AHR3 gene, I investigated the patterns of AHR3 transcript
accumulation in in vitro cultured fat bodies in response to 20E. For these experiments,
previtellogenic fat bodies of 3-5 days after eclosion were incubated in the culture medium
in the presence or absence of the hormone. The RT-PCR analysis was used to monitor the
transcript levels in the cultured tissues (see Materials and Methods). In the preliminary
experiments it was found that the AHR3 transcript, which was absent in the
previtellogenic fat body, appeared after exposure of the tissue to 20E (not shown). In the
next series of experiments, I determined the 20E concentration required for maximal
activation of AHR3 in the fat body by culturing this tissue with increasing concentrations
of the hormone. Fat bodies from previtellogenic 3-5 day-old females were used to
determine the background levels of transcripts for an early gene E 75, a late gene V CP, a
housekeeping gene actin, and the early-late gene under investigation, AHR3. Actin was
expressed in detectable amounts in previtellogenic fat bodies. Culturing in the medium
without the hormone did not result in any measurable increase in mRNA of all these

genes (Fig. 6). The levels of actin mRNA remained relatively constant in the presence of

124

140

 

  
 

 

  

 

 

     

IE75A,
120 1
135753;
53400 IHRS ; “
V IVCP . 33
s ' s
2 § . \‘s
- °°* \ Se ss
E ‘° R \\ R
20 \ s \
o, . It s s s
PV 0 10* 10'7 10‘ 10“

ZOE (M)

Fig. 6. AHR3 response to 20-hydoxyecdysone (20E) in in vitro fat body organ
culture. The transcriptional response of AHR3 to 20E was examined in an in vitro
culture system of female previtellogenic fat bodies (Deitsch et al., 1995a; Raikhel etal.,
1997). Total RNA was extracted from untreated previtellogenic (PV) fat bodies, or from
fat bodies incubated for 2 hr at the indicated 20E concentrations. RT-PCR/Southem blot
hybridization was performed using probes specific for AHR3 or each other gene (see
Materials and Methods, and Fig. 1). AaE 75A and AaE 75B provided controls for an early
gene, VCP for a late 20E-responsive gene. The results represent the average of three
independent experiments : S.D.

 

 

 

 

 

 

 

 

 

 

 

 

 

   

(a) Culture Time (hr) (b) Culture Tlme (hr)

Fig.7. Time course of AHR3 induction by 20E in in vitro fat body organ culture.
Accumulation of AHR3 mRNA (A) and VCP mRN A (B) durin culturing in the
hormone-free medium (CM), in the continuous presence of 10' M 20E (20E), or 4 hr
exposure to 20E followed by culturing in the hormone-free medium (20E-4hr). RT-
PCR/Southem blot hybridization was performed using probes specific for either AHR3,
or VCP as a late 20E-responsive gene. The results represent the average of three
independent experiments i SD.

125

every tested concentration of 20E (not shown). In contrast, E 75, HR3 and VCP genes
responded to increasing concentrations of 20E in the culture medium by dramatic rise in
their mRNA levels. As it has been previously established, the VCP mRNA reached its
maximal levels at 10'6 M of 20E (Deitsch et al., l995a), while E 75A and E 758 mRNAs at
10'7 M of 20E (Pierceall et al., 1999). The mRNA levels of AHR3 increased in a dose-
dependent manner in response to 20E. The AHR3 transcript reached its maximal level at
the 20E concentration of 10'6 M (Fig. 6).

Culturing the previtellogenic fat bodies in the hormone-free medium up to 16 hr
did not induce AHR3 and VCP mRNAs (Fig. 7). In contrast, AHR3 mRNA appeared
rapidly in the fat bodies cultured in the medium containing 10'6 M 20E. The levels of
AHR3 mRN A then gradually increased over the l6-hr culturing period and did not show
any regression (Fig. 7A). The VCP mRNA induction profile was similar to that of AHR3,
the levels of this transcript increased continuously throughout the entire 16 hr incubation
with 10'6 M 20E (Fig. 7B). In the next experiment, the fat bodies were cultured in this
hormone-containing medium for only 4 hrs, washed three times with the hormone—free
medium and continued incubation in the hormone-free medium for 12 more hrs. Both
AHR3 and V CP mRNA levels increased after first the 4 hr of incubation, however, their
levels decreased following withdrawal of the hormone from the medium. Interestingly,
the AHR3 mRNA level dropped more rapidly that that of VCP (Fig.7A and 7B).

To determine the requirement of protein synthesis for activation and maintenance
by 20E of the HR3 early-late gene in the mosquito fat body, the previtellogenic fat bodies
were incubated with cycloheximide (Chx), a protein synthesis inhibitor which reversibly

represses protein synthesis in this tissue in a dose-dependent manner (Deitsch eta1.,

126

 

"-l—o— Chx/205.411
+ Chx/20E
—a— Chx

 

--.--20E

 

 

Relative mRNA Level

 

 

 

 

 

 

 

 

 

 

 

 

(a) Culture Time (hr)
-—O——Chx/20E-4h
+ChXIZOE
—o—Chx
- - .- - -20E

120
'15 100 . I
g vcp
< 3° ‘ . ' '
2 p'
e .. ..
o '.
. 4o « .'
e ,-
g 20 " ' ' I '
_ _ . I
o .W

0 4 8 12 16

(b) Culture Time (hr)

Fig. 8. Effect of cycloheximide on the AHR3 induction by 20E in in vitro fat body
organ culture. Accumulation of AHR3 mRNA (A) and VCP mRNA (B) during culturing
with 10'3 M Chx in the hormone-free medium (Chx), in the continuous presence of 10’6
M 20E (Chx/20E), or in the 4 hr exposure to 20E followed by culturing in the hormone-
free medium (Chx/20E-4hr). Dashed lines in (A) and (B), marked (20E), are controls
from the previous experiment (Fig. 7). RT-PCR/Southem blot hybridization was
performed using probes specific for either AHR3, or VCP as a late 20E-responsive gene.
The results represent the average of three independent experiments j; SD.

127

l995a,b). The fat bodies were first pre-incubated in the culture medium for 1 hr in the
presence of 10'3 M Chx, and then continued incubation with the same dose of Chx but
with the addition of 10'6 M 20B for various times (Fig. 8). Incubation of the
previtellogenic fat bodies in the presence of Chx in the hormone-free medium induced
neither the AHR3 gene nor VCP. Addition of 10’6 M 20B to the Chx-containing medium
activated the AHR3 gene in contrast to the VCP gene, which was inhibited by Chx.
Furthermore, the AHR3 gene was induced to a higher level by Chx/20E than by 20E
alone. Likewise, withdrawal of 20E from the Chx—containing medium after the first 4 hrs
of incubation, caused a slower decline in the AHR3 mRNA level in fat bodies (Fig. 8A)

than withdrawal of 20E from the Chx-free medium (Fig. 7A).

DISCUSSION

In this work, cloning and characterization of the mosquito nuclear receptor HR3
(AHR3) is reported. Analysis of the AHR3 deduced amino acid sequence showed that it
was highly similar to its insect counterparts. The most conserved were the DNA-binding
domain (DBD) and so-called carboxy-terminal extension (CTE) of the DNA-binding
domain (F igs.l and 2). Giguere et al. (1994, 1995) have performed a detailed analysis of
RORa nuclear receptor and have shown that this human HR3 courterpart binds DNA as a
monomer, recognizing an asymmetric response element (RORE) composed of two
distinct half-sites, a 6 bp AT-rich sequence at the 5’-end and a PuGGTCA core half-site
motif at the 3’-end (Giguere et al, 1994). The specificity of RORa binding to the RORE
is provided by amino acids from the P-box and the CTE: the P-box recognizes the core
half-site motif, and a group of amino acids from the CTE recognizes the AT-rich half—site

(Giguere et al, 1995). Subsequently, the presence of similar DNA-sites has been shown

128

for insect HR3s from Drosophila (Lam et al, 1997; Homer et al, 1995) and
Choristoneura (Palli et al, 1996). The mosquito AHR3 had a highly conserved P-box
and CTE, suggesting that it likely recognized a similar target DNA-site as a monomer.

The ROR/HR3 transcription factors are classified as orphan nuclear receptors due
to the lack of any known ligand (Mangelsdorf et al, 1995). The lack of ligand-dependent
transactivation could lead to the relaxation in conservation of the orphan receptors’
LBDS, which in nuclear receptors are generally responsible for dimerization and ligand-
binding functions. This may be one reason why putative LBD of AHR3 did not show a
high similarity with those of nematode CeHR3 and human RORa counterparts (Giguere
et al, 1994; Kostrouch et al, 1995), though it was still significantly conserved among
HR3 receptors (Fig. 2).

The high similarity between HR3s started in A/B-domains from the 7th amino
acid preceding DBDs. Interestingly, an intron-exon junction (marked by arrowhead in
Fig. 1) has been determined at this position after the structure of DHR3 and human RORa
genes were characterized (Koelle et al, 1992; Giguere et al, 1994). The rest of the
mosquito A/B domain was poorly conserved, as is consistent with the generally low
conservation of highly variable nuclear receptor A/B domains. However, in the case of
the ROR/HR3 subfamily, analysis of the A/B-domains merits special attention. In fact,
cloning of three human RORa isoforms, which differ only in their A/B-domains, has
revealed that the distinct A/B-domains modulate the DNA-binding specificity of each
RORor isoform (Giguere et al, 1994; McBroom et al 1995). So far no similar HR3
isoform variation has been demonstrated in insects. In this context, it is interesting to

note that in lepidopteran HR3 receptors from Manduca (Palli et al, 1992), Galleria

129

(Jindra et al, 1994b), and Choristoneura (Palli et al, 1996), isolated from larval stages of
these insects, the A/B-domains are not only similar in size but also possess a high
percentage of identical amino acids (Palli et al, 1996). The mosquito HR3 is the first
HR3 isolated from an adult insect. It had a distinct A/B-domain and might represent a
distinct adult-specific HR3 isoform.

Although, HR3 was one of the first insect nuclear receptors to be cloned (Palli et
al., 1992; Koelle et al., 1992), understanding of its role in regulating the ecdysone-
mediated genetic network has been achieved only recently (Lam et al, 1997; White et al,
1997). These studies have demonstrated that the orphan nuclear receptor DHR3 is one of
the key genes in the ecdysone-mediated genetic regulatory network. During Drosophila
metamorphosis, DHR3 has a dual role in repressing the early genes while activating
another orphan nuclear receptor, flFIZ—Fl , which has been implicated in providing
competence for stage-specific responses to ecdysone (Lam et al, 1997; White et al,
1997). DHR3 homologues cloned from several Lepidopteran insects are highly
conserved, and their expressions suggest HR3 involvement in mediating ecdysone
response during Lepidopteran development and metamorphosis (Palli et al., 1992; 1996;
Jindra et al., 1994b). Moreover, recent analysis of Drosophila DHR3 mutants revealed
that DHR3 function is required to complete embryogenesis (Carney et al., 1997),
prepupal-pupal transition and differentiation of adult structures during Drosophila
metamorphosis (Lam et al., 1999). In this study, the novel evidence provided strongly
suggests that the involvement of HR3 in the ecdysone gene regulatory hierarchy during

female reproduction.

130

In insects, transcription of the HR3 gene is under the regulation of 20E (Homer et
al, 1995; Palli et al, 1996;1indra et al, 1994b; Palli et al, 1992). Expression analysis of
the mosquito HR3 concurs with the studies in other insects, which have previously shown
the hallmarks of early-late gene expression, direct positive regulation of the HR3
expression by 20E followed by a sustained expression during the late ecdysteroid
response. The in vivo developmental profiles of AHR3 in both fat body and ovary
showed that the peak of the AHR3 transcript coincided with a. high titer of 20E in
hemolymph. In in vitro cultured fat bodies, in contrast to the early gene AaE 75, which
showed rapid peak activation by 20E at 2~4 hr of culturing (Pierceall et al., 1999), the
AHR3 induction slowly and continuously increase in its mRNA level up to 16 hr of
culturing. This pattern of AHR3 activation by 20E was consistent with observations on
the successive puffing expression of the corresponding loci 7SB (encoding E 75) and 46F
(encoding DHR3) in the polytene chromosomes (Ashbumer, 1972; Walker and
Ashbumer, 1981), and with findings in cultured Drosophila organs (Homer et al., 1995)
and salivary glands (Huet et al., 1995). Likewise, in the cultured silk glands of Galleria
mellonella, 20E induction of GHR3 mRNA has been observed 2 hr after the peak of
E 75A mRNA (Jindra and Riddiford, 1996). Our in vitro study suggested that delayed
induction of the early-late HR3 gene was due to the relatively low sensitivity of its
maximal response to 20E, which was more similar to late genes rather than to early
genes. It seems likely that the high level of 20E needed for maximal response is required
to trigger the expression of factors that contribute to the sustained induction, presumably

through elements in addition to the EcREs responsible for early expression. However,

131

like other early-late genes, AHR3 is also directly regulated by 20E; in contrast to the late
VCP gene, AHR can be activated by 20E in the absence of new protein synthesis.
Previously, I have described on the expression and function of the ecdysone
receptor genes AaEcR and AaUSP and the induction of the early ecdysone-inducible gene
AaE 75 in conjunction with the initiation of vitellogenic ecdysteroid response. Here, this
study demonstrates that conservation of ecdysone responsive genes, originally elucidated
in development of immature stages of Drosophila, extends to the next level in this
hierarchy, to the early-late genes involved in the transition from the early to the late
ecdysone response. The expression pattern of the early-late gene AHR3 suggests that it
may play a central role in mediating this transition during the vitellogenic response to

ecdysone in the adult mosquito.

ACKNOWLEDGMENTS
I thank Marianna Kapitskaya for partial cloning of AHR3 and Northern analysis,
Ken Miura for RT-PCR analysis, and William Segraves for critical reading. This work

was supported by a grant from the National Institutes of Health (A136959).

132

Chapter 5*

Cloning and Characterization of the Mosquito FTZ-FI Gene

*Modified from the original publication in Dev. Biol.
Ref: Li, C., Kapitskaya, M. Z., Zhu, 1., Miura, K., Segraves, W., Raikhel, A. S. (2000).

Dev. Biol. 224, 96-110

133

ABSTRACT

In the mosquito Aedes aegypti, the adult female becomes competent for a
vitellogenic response to ecdysone after previtellogenic development. Here, BFTZ-F 1, the
nuclear receptor implicated as a competence factor for stage-specific responses to
ecdysone during Drosophila metamorphosis, serves a similar function during mosquito
vitellogenesis. AaFTZ-Fl is expressed highly in the mosquito fat body during pre- and
post-vitellogenic periods when ecdysteroid titers are low. The mosquito AaFIZ-FI
transcript nearly disappears in mid-vitellogenesis when ecdysteroid titers are high. An
expression peak of HR3, a nuclear receptor implicated in the activation of flF IZ-F I in
Drosophila, precedes each rise in mosquito F IZ-F 1 expression. In in vitro fat body
culture, AaFTZ-FI expression is inhibited by 20-hydroxyecdysone (20E) and super-
activated by its withdrawal. Following in vitro AaFTZ—FI super-activation, a secondary
20E challenge results in super-induction of the early AaE 75 gene and the late target VCP
gene. Electrophoretic mobility shift assays show that the onset of ecdysone—response
competence in the mosquito fat body is correlated with the appearance of the functional
AaFTZ-Fl protein at the end of the previtellogenic development. These findings suggest
that a conserved molecular mechanism for controlling stage specificity is reiteratively

used during metamorphic and reproductive responses to ecdysone.

134

INTRODUCTION

Competence to respond to specific developmental signals allows the spatially and
temporally restricted utilization of global signals within a multicellular organism. A
signal molecule, which is spread throughout an organism via the circulatory system,
elicits a specific response only from competent cells and tissues, leaving other parts of an
organism unaffected. Recent studies have given rise to considerable insight into the
molecular mechanisms underlying competence. One of the general means for
accomplishing competence is the time- and cell-specific expression of a receptor or
receptor cofactor needed to render a cell responsive to a corresponding hormonal signal
(Shi et al., 1996; Puigserver et al., 1998). Regulation of competence can also be
achieved through chromatin remodeling via involvement of linker histories, which repress
transcription (Steinbach et al., 1997; Lee and Archer, 1998). Finally, the stage-
specificity of a response to a hormonal signal can be determined by a specialized
competence factor. Drosophila metamorphosis provides one of the best-studied
examples of how competence affects the specificity of response to a hormonal signal. In
this organism, the BFTZ-Fl orphan nuclear receptor functions as a competence factor
which directs the developmental transition from larval to pupal-specific gene expression
in response to the insect steroid hormone ecdysone (Woodard et al., 1994; Broadus et al.,
1999)

Mosquitoes serve as vectors for many harmful human diseases (Collins and
Paskewitz, 1995; Butler et al., 1997; Beier, 1998) because they require blood feeding for
their egg development. In the so-called anautogenous mosquitoes, vitellogenesis and egg

maturation are initiated only after a female mosquito ingests vertebrate blood. A blood

135

meal triggers a hormonal cascade resulting in production of 20-hydroxyecdysone (20E),
the terminal signal, which activates yolk protein precursor (YPP) genes in the fat body
(Hagedom, 1983; 1985; Raikhel, 1992; Dhadialla and Raikhel, 1994; Deitsch et al.,
l995a;1995b)

In the anautogenous mosquito Aedes aegipti, which is being used as a model
vector, vitellogenesis proceeds through two periods (Fig. 1). The approximately 3-day-
long previtellogenic development of both the ovary and fat body proceeds under the
control of juvenile hormone (.1 H) 111 (Gwartz and Spielman, 1973; Flannagan and
Hagedom, 1977). The JH titers are high during the previtellogenic period and drop
immediately following the onset of vitellogenesis, which is initiated when the female
mosquito takes a blood meal (Shapiro et al., 1986). The fat body then produces yolk
protein precursors (YPP), which are internalized by the developing oocytes (Raikhel,
1992). In addition to vitellogenin (Vg), the major YPP, the mosquito fat body
synthesizes two other YPPs, vitellogenic carboxypeptidase (VCP) and vitellogenic
cathepsin B (VCB) (Cho et al., 1991; 1999). YPP synthesis reaches its maximal levels at
24 hr post-blood meal (PBM) and then proceeds until 30-32 hr PBM, when it is rapidly
terminated (Raikhel, 1992).

The titers of ecdysteroid in the mosquito female hemolymph are correlated with
the rate of YPP synthesis in the fat body (Fig. 1). While the 20E titers are only slightly
elevated at 4 hr PBM, they begin to rise sharply at 6—8 hr PBM, and reach their maximum
levels at 18-24 hr PBM (Hagedom et al., 1975). Consistent with the initially proposed
role of 20E in activating vitellogenesis in anautogenous mosquitoes (Hagedom et al.,

1973; Hagedom and Fallon, 1973), experiments using an in vitro fat body culture have

136

rm
N
l—bJ
03
Z
J;
l-oo

12 6 g) 4 218312 3642 4854 go 6672

L J J

Homiones

20F. 4— W
1" WIDE W

Nuclear Receptors

 

 

 

 

AaEcR M—
A-USP- _— ,— A—

AalTSPb

AHR3 _ A L

Late Genes
Vs

 

 

 

 

 

 

 

 

VCP

 

 

er .2 3.81M {31.21.6302518323942485.46.06.672

 

Days Hours

Fig. 1. Summary of events during the first cycle of vitellogenesis in the
anautogenous mosquito, Aedes aegvpti. The previtellogenic period begins at eclosion
(E) of the adult female. During first ~2-3 days of post-eclosion life, the fat body becomes
competent for subsequent vitellogenesis. The female then enters a state-of-arrest; yolk
protein precursors, vitellogenin (Vg) and vitellogenic carboxypeptidase (VCP) are not
synthesized during this previtellogenic period. Only when the female takes a blood meal
(BM), vitellogenesis is initiated. Hormones: hormonal titers of juvenile hormone (JH)
and ecdysteroids (20E) in A. aeg’pti females (Hagedom et al., 1975; Shapiro and
Hagedom, 1982). Transcript profiles of nuclear receptors (Nuclear Receptors) in the
mosquito fat body were determined by RT-PCR/Southem blot; Transcript profiles of yolk
protein precursors (Late Genes) in the mosquito fat body were determined by Northern
blot and RT-PCR/Southem blot. AaEcR, ecdysone receptor; AaUSPa, Ultraspiracle
isoform A; AaUSPb, Ultraspiracle isoform B (Wang et al., 2000); AaE75A, early gene
AaE 75 isoform A (Pierceall et al., 1999); AHR3, early-late gene AHR3 (Kapitskaya et
al., 2000); Vg, vitellogenin (Cho and Raikhel, 1992); VCP, vitellogenic carboxypeptidase
(Cho et al., 1991).

137

shown that physiological doses of 20B (10’7 - 10'6 M) activate two YPP genes, Vg and
VCP (Deitsch et al., 19953; 1995b). Utilization of the protein synthesis inhibitor,
cycloheximide (Chx), has demonstrated that the activation of YPP genes in the mosquito
fat body requires protein synthesis (Deitsch et al., l995a). Thus, it is likely that a
regulatory cascade similar to that seen in Drosophila development mediates the action of
20E in the mosquito fat body.

The functional ecdysteroid receptor is a heterodimer of the ecdysone receptor
(EcR) and a retinoid X receptor (RXR) homolog, Ultraspiracle (USP) (Yao et al., 1992;
1993). In the mosquito Aedes aegypti, one isoform of ecdysone receptor (AaEcR) and
two USP isoforms, AaUSP-A and AaUSP—B have been cloned (Cho et al., 1995;
Kapitskaya et al., 1996). The mosquito EcR/U SP heterodimer has been shown to bind to
various ecdysone response elements (EcREs) to modulate ecdysone regulation of target
genes (Wang et al., 1998). A mosquito homolog of the Drosophila early gene E 75 is
involved in mediating the ecdysteroid response during vitellogenesis (Pierceall et al.,
1999). Furthermore, a highly conserved homolog of the DHR3 orphan nuclear receptor
previously implicated in activation of Drosophila flFTZ-Fl is expressed in vitellogenic
tissues, the fat body and ovary (Kapitskaya et al., 2000).

A preparatory, developmental, previtellogenic period is required for the mosquito
fat body to attain competence for ecdysone responsiveness (Sappington and Raikhel,
1999). F lannagan and Hagedom (1977) have demonstrated that the acquisition of
competence for ecdysone response in the mosquito fat body is under the control of
juvenile hormone 111. However, the molecular mechanisms underlying the acquisition of

sensitivity to ecdysone in the mosquito fat body remain obscure.

138

Here this chapter reports that BFTZ-F l , the orphan nuclear factor implicated as a
competence factor for stage-specific responses to ecdysone during Dr050phila
metamorphosis, likely serves a similar function during mosquito vitellogenesis. A
homolog of Drosophila MTZ—FI is expressed relatively highly in the mosquito fat body
during pre- and post-vitellogenic periods when ecdysteroid titers are low. In in vitro fat
body culture, AaFTZ—FI expression is inhibited by 20B and super-activated by its
withdrawal. Moreover, afier the in vitro super-activation of AaF 72-F1 expression in the
previtellogenic mosquito fat body, a secondary 20E challenge results in super-induction
of the early AaE 75 gene and the late target VCP gene. Although AaFIZ—FI mRNA is
abundant even in newly emerged mosquitoes, electrophoretic mobility shift assays show
that it is the appearance of functional AaFTZ-F 1 protein that correlates with the onset of
ecdysone-response competence in the mosquito fat body. Our findings suggest that the
regulation and function of F TZ-F 1 during mosquito reproduction closely resemble those
shown at the onset of Drosophila metamorphosis, and that FTZ-Fl may therefore be part
of a conserved and broadly utilized molecular mechanism controlling the stage specificity

of ecdysone responses.

MATERIALS AND METHODS

Animals

Mosquitoes, Aedes aegypti, were reared according to the method of Hays and
Raikhel (1990). Larvae were fed on a standard diet as previously described (Lea, 1964).
Vitellogenesis was initiated by allowing females, 3-5 days after eclosion, to feed on an

anesthetized white rat.

139

Materials

The RNA ladder was purchased from Gibco—BRL; Sequenase, from U. S.
Biochemical Corp.; restriction enzymes, from Boehringer Mannheim. All primers used
in PCR were ordered from Gibco-BRL. Superscript II reverse transcriptase and Taq
DNA polymerase were from Gibco-BRL; random hexamer primers, from Promega.
Perkin-Elmer was the source of reagents for the polymerase chain reaction (PCR). MSI
CO and Amersham Pharrnacia Biotech supplied nitrocellulose and Hybond-N+ blotting
membranes, respectively. Radionucleotides for labeling probes and DNA sequencing
were from NEN Life Science Products. All other reagents were of an analytical grade

from Sigma or Baker.

cDNA cloning

Degenerate primers based on the sequences of F TZ—F 1 insect homologs were used
for PCR to obtain a mosquito homolog cDNA fragment. Amplification was carried out
in a Perkin Elmer thermal cycler using template cDNA produced by reverse transcription
from total RNA prepared from the fat bodies of vitellogenic female mosquitoes. The
PCR-generated fragment was used as a probe to screen a XZAPII cDNA library prepared
from the fat bodies of vitellogenic female mosquitoes during 6-48 hr post-blood meal
(PBM) as previously reported (Cho and Raikhel, 1992). Several positive cDNA clones
were subsequently isolated and sequenced at W.M. Keck Foundation Biotechnology

Resource Laboratory, Yale University. Analyses of nucleotide and deduced amino acid

140

sequences were performed using the software of GCG (Genetics Computer Group,

University of Wisconsin, Madison).

Northern blot analysis

Total RNA was isolated by using the guanidine isothiocyanate method as
described previously (Bose and Raikhel, 1988), or using TRIZOL Reagent (Gibcol-
BRL). Polyadenylated mRNA was isolated using Niomag oligo (dT)20 magnetic beads
and the manufacturer’s protocols (PerSeptive Diagnostics, Inc.). For Northern blot
analysis, total or poly(A) RNA was separated by electrophoresis in 1.2%
agarose/formaldehyde gels in MOPS buffer, blotted to nitrocellulose membranes, and
hybridized with 32P-labeled DNA probes (Random Primers DNA Labeling Systems,
Gibco-BRL) under high stringency conditions. Autoradiography was conducted at -80°C

with intensifying screens.

RT -PCR/Southern blot

RT-PCR/Southem analyses were performed basically as previously described
(Pierceall et al., 1999). Total RNA was prepared from fat bodies throughout the first
vitellogenic cycle, as described in the Northern blot procedure above, with the
modification that all isopropanol precipitation steps were done without low-temperature
incubation, to avoid co-precipitation of glycogen and salt. RNA concentration was

determined spectrophotometrically; absorbance ratios, A260/A280 and A260/A230, in

RNA preparations were always above 1.7 and 2.0, respectively. Five microgram aliquots

of each RNA preparation were reverse-transcribed by Superscript II reverse transcriptase

141

(Gibco-BRL) and random hexamer (Promega) in a reaction volume of 20 it]. After this
reaction, the volume of each sample was brought to 40 ul with TE buffer (10 mM Tris-
Cl, lmM EDTA, pH 8.0). cDNA samples were stored at -200C until use. In the analyses
of developmental profiles, 0.025 fat body- or pupa-equivalent cDNA pool was used as a
PC R template for each point. Thermal cycling conditions were as follows: 940C for 3
min, followed by 18 cycles of 940C for 30 sec, 600C for 30 sec and 720C for 30 see. As
a reference to the developmental change of the fat body, the samples of the same cDNAs
were subjected to the 13 cycles of PCR with a primer pair specific to the mosquito
vitellogenic carboxypeptidase (V CP). The reaction was followed by Southern blot
hybridization using a VCP specific probe, which was generated by PCR using cloned

VCP cDNA as a template (Cho et al., 1991). The PCR primers for each gene examined

in this paper were listed in Table 1.

Table 1

 

Primers Forward Reverse

 

AaE75A TAGTGCAATCAACGTATACCAATC GGCACACGACCGAATCTGAC

AaE7SB CGTGGAAGAAGACCACGATCG GGCACACGACCGAATCTGAC
AHR3 GTGTGCGGCGACAAGTCGTC TCGACCGAATTTGACCGCGTC
AaFTZ-Fl TGAAGGTGGACGACCAGATGAA TGTTCTGCAGCTCGTTGAAGTG
VCP AGCGCCCATTCTTGTTTGG C AGCTCATACAGGTATTCTCC
V g TAC TTGAAGAC AAGATGC TAGC G CGTTCTTGTAACTGTAGCCTTG

 

142

RT-PCR/ Southern blotting of total RNA preparations without reverse
transcriptase did not result in any appreciable signal, indicating that contamination of
RNA preparations with genomic DNA fragments was virtually negligible (not shown).
As a quantitative control for the RT-PCR, each cDNA was diluted to 1/2, 1/4, 1/8, and
1/16, and PCR amplifications of the cDNA dilutions were monitored by different PCR
cycles ranging from 13 to 22 cycles followed by Southern blot quantification. The
analysis of normalized results showed that RT-PCR amplifications were linear (not
shown). To further test whether results of the RT-PCR/Southem analyses correlate with
those from the Northern blot analysis, both of these techniques were used in parallel to
measure the VCP mRNA levels in fat bodies during the vitellogenic cycle as well as over
the course of in vitro treatments with 20-hydroxyecdysone. VCP mRNA levels were
analyzed by either Northern blot or RT-PCR/Southern blot analyses. In both cases, the
results were quantified using a phosphorimager (Molecular Dynamics). The Northern

blot and RT-PCR/Southem blot analyses yielded very similar results (data not shown).

In vitro organ culture

The abdominal walls with adhering fat bodies (hereafter referred to as the fat
body) were dissected from 0~0.S-day or 3~5-day-old previtellogenic females and
incubated in an organ culture system as previously described (Raikhel et al., 1997). For
the stage-specific responses to 20E, the dissected fat bodies were exposed to 10'6 M 20B
for 6 hr; for all other cultures, the fat bodies were incubated with 20E either for 16 hr, or
for 4 hr with 20E followed by washing three times with hormone-free medium and an

additional 12 hr incubation in hormone-free medium. At every 4 hr interval, a sample of

143

nine fat bodies was collected and stored in liquid nitrogen until use. As a control, the fat
bodies were incubated in hormone-free medium. RT-PCR/Southern blot analyses were
utilized as described above. Hybridization using probes for the ecdysone-inducible VCP
gene was provided as a positive control.

To test the effect of the protein synthesis inhibitor cycloheximide (Chx) on the
expression of each gene, dissected fat bodies from 3-5 day old female mosquitoes were
pre-treated with culture medium containing 10'4 M Chx for 1 hr, then further incubated
with the same concentration of Chx either with or without 10'6 M 20E. The
C hx/hormone withdrawal experiment was conducted as described previously in the
hormone withdrawal experiment above except that 104 M Chx was added to all culture
media.

The experiments of secondary responses to 20E or Chx/20E in cultured fat bodies
were an extension of the 20E-withdrawal experiments described above. After the initial 4
hr exposure to 20E followed by 12 hr culturing without 20E, the fat bodies were exposed
again to 20E, or pre-treated with Chx for 1 hr and then exposed to 20E/Chx for an
additional 4 or 16 hr. In addition to these two experiments, as the third experiment, after
pre-treatment with Chx for 1 hr and the initial 4 hr exposure to 20E/Chx followed by 12
hr culturing with Chx only, the fat bodies were exposed again to 20E/Chx for an
additional 4 or 16 hr. RT-PCR/Southem blot were utilized to analyze the responses of

each gene under investigation.

144

Electrophoretic mobility shift assays

Electrophoretic mobility shift assays (EMSA) were carried out as described
previously (Wang et al., 1998). The nucleotide sequence of the FIRE used in EMSA is
5’GCAGCACCGTCTCAAGGTCGCCGAGTAGGAGAA3’ (Ohno et al., 1994). The
FIRE oligo was purified before EMSA. The in vitro synthesized AaFTZ-FI protein was
made by the TNT system (Promega). Fat body nuclear extracts (FBNE) were prepared as
previously described (Miura et al., 1999). The DmFTZ-Fl antibody was a generous gift

from Dr. Carl Thummel (University of Utah).

RESULTS

E cdysone-responsive vitellogenic competence in the mosquito fat body

Previously, Flanagan and Hagedom (1977) have shown that the fat body of newly
emerged female mosquitoes is not responsive to 20E in vitro with respect to vitellogenin
protein synthesis. However, responsiveness to this hormone can be detected in the fat
body as early as 12 hr post-eclosion. Furthermore, the levels of the 20E-activated Vg
synthesis in the fat body in vitro reach its maximum in 36-48-hr-old mosquitoes. Here, I
utilized the fat body from mosquitoes of different post-eclosion ages and monitored the
appearance of mRNAs for two YPP genes, Vg and VCP, in response to 20E in vitro.
These experiments confirmed that the fat body of a newly emerged mosquito was not
competent to respond to hormonal stimulation and that the YPP genes were not activated
(Fig. 2). In contrast, both Vg and VCP genes were expressed in the fat body of 3-5 day
old females treated with 10‘6 M 20B in vitro (Fig. 2). Moreover, the RT-PCR analysis of

both Vg and VCP mRNAs showed that the ability of the YPP genes to respond to

145

0-0.5d 3-5d
+ - + - 20E

 

6.5kb « urn V9

 

 

 

 

1.5kb - ' , VCP

_.....

Fig. 2. Northern blot analysis of stage-specific induction of the vitellogenin (Vg) and
vitellogenic carboxypeptidase (VCP) transcripts by the hormone 20E. Nine
previtellogenic fat bodies of different stages were cultured in the presence (+) or absence
(-) of 10‘6 M 20B for 6 hr. Total RNA from 3 fat bodies (~10 ug) was loaded in each
lane. Hybridization was performed with the 32P- labeled Vg or VCP probe (see Materials
and Methods). Ribosomal RNA (rRNA) stained with ethidium bromide is shown as a
loading control. 0-0.5d: fat bodies from newly eclosed previtellogenic females, 0-12 hr
post-eclosion; 3-5d: fat bodies from previtellogenic females, 3—5 day post-eclosion.

 

 

 

146

hormonal activation increased gradually over the previtellogenic period from 1 to 3 day
post-eclosion (not shown). These data indicated that the ecdysone-responsive

vitellogenic competence in the mosquito fat body was under transcriptional control.

Isolation and characterization of a cDNA clone encoding the mosquito AaF T Z-
F1 nuclear receptor

To investigate whether the mosquito homolog of the competence factor BFTZ-Fl
could be involved in the control of competence in the mosquito fat body, a mosquito
homolog of Drosophila BFTZ-F l was cloned. The initial isolation of a cDNA fragment
encoding the mosquito F IZ-F I homolog was achieved through a PCR-based approach.
A pair of degenerate PCR-primers used to amplify cDNA of the mosquito F IZ—F 1
homolog is shown in Fig. 3. After a mosquito PCR-fragment of the appropriate size, 266
bp, was sequenced and confirmed to encode a polypeptide of the expected sequence, it
was used as a probe to screen a vitellogenic mosquito fat body cDNA library. Several
positive clones were identified, the largest of which contained an insert of 4.6 kb. The
first methionine residue at the N-terminal region of the ORF (811 aa) was located at 1.4
kb downstream of this cDNA clone. Before the first Met, there were several in-frame
stop codons in the 5’UTR, suggesting that the first Met was the translation start site.
There was a poly (A) tail at the 3’UTR end of this clone.

The conceptual translation of AaFTZ-Fl amino acid sequence revealed distinctly
high similarity with its insect counterparts of Drosophila and Bombyx (Lavorgna et al.,
1991; 1993; Sun et al., 1994), particularly in the DBD and the adjacent stretch of 29

amino acids known as the F TZ-Fl box (Fig. 3). According to recent data, members of

147

Fig. 3. Alignment of the deduced amino acid sequence of the mosquito AaFTZ-FI
cDNA with its insect homologs: DmBFTZ-Fl from Drosophila (Lavorgna et al., 1993)
and BmFTZ-F I from Bombyx (Sun et al., 1994). Asterisks or carets indicate amino acids
which are identical or similar, respectively. The borders of DNA-binding domain (DBD)
and putative ligand-binding domain (LBD) are marked with square braces. The P-box in
the DBD is in bold. The F TZ-F 1 -box is double-underlined. Two vertical arrowheads
flank the non-conserved insertions in the hinge-regions of insect FTZ-F ls. Horizontal
arrows indicate sequences corresponding to the degenerate sense primer (Pr-l) and
antisense primer (Pr-2) used for AaFTZ—Fl cDNA cloning. A black dot marks an EcoRV
restriction site. Two other arrows represent the regions of the sequence-specific sense
(Pr-3) and antisense (Pr-4) primers utilized in RT-PCR analysis of AaF TZ-F I . The Pr-4
was also used to generate a single-stranded DNA-probe for Northern blot (see Fig. 4A).
Most part of A/B domains in FTZ-Fls are not shown due to very low similarity. The
complete sequence of mosquito AaFTZ—Fl has been submitted to the GenBank

(http://wwwngbi.nlm.nih.govl; its accession number is AF274870.

148

 

m2 -r1
mane 4'1
m2 4'1

AaFTZ-FI

mflnz -F1

BmFTZ-FI

MHZ-F1
mflnz -F1

BmFTZ-Ffi

m2 -1"1
magma-F1

BmFTZ-Fi

amz 4'1
MHz-r1
amnz -r'1

m2 -r'1
mflnz -F1
mz-n

MHZ-F1
mgr-1'2 -1r1

BmFTZ-Ffl

mz-n
mBHz-m

BmFTZ-Ffl

m2 -r1
mama-F1
m2 -r'1

m2 -r1
magma-r1
mam-r1

l—* DBD
Pr-l I P-box

335 ELCPVCGDKVSGYHYGLLTCESCKGFFKRTVQ

281 ELCPVCGDKVSGYHYGLLTCESCKGFFKRTVQ
87 ELCPVCGDKVSGYHYGLLTCESCKGFFKRTVQ

********************************

4—1

NKKVYTCVAERQCHIDKTQRKRCPYCRFQKCLEVGMKLEAVRADRMRGGRNKFGPMYKRDR

NKKVYTCVAERSCHIDKTQRKRCPYCRFQKCLEVGMKLEAVRADRMRGGRNKFGPMYKRDR
NKKVYTCVAERACHIDKTQRKRCPFCRFQKCLDVGMKLEAVRADRMRGGRNKFGPMYKRDR

***********A************A******ir"****************************

ARKLQIMRQRQLAIQALRGSIG-GGVGIGQLGSDGASAQLQGIDYHQPYSNMHIKQEIQIP

ARKLQVMRQRQLALQALRNSMGPDIKPTPI--SPG --------- YQQAYPNMNIKQEIQIP
ARKLQMMRQRQIAVQTLRGSLGDGGLVLG-FGSP -------------- YTAVSVKQEIQIP

*****A*****A*A*A**A*A* * * *******
—

V' o
QVSSLTSSPDSSPSPIAIALGQVNPQTTISLSSGVGGNGSSGGGSGGGGSNNSSGAGTGGS
QVSSLTQSPDSSPSPIAIALGQVNASTGGVIATPMNAGTGGSGGGGLNGPSSVGNGNSSNG
QVSSLTSSPESSPGP---ALLRAQPQPPQPPPPPTHDKWEAH -------------------
******A**A***A* **

V
NSTSTAIQNQLSESKLWMSANSTTA ---------------------------- SPHSLSPK

SSNGNNNSSTGNGTSGGGGGNNAGGGGGGTNSNDGLHRNGGNDSSSCHEAGIGSLQNTADS
————————————————————————————————————————————————————— SPHSASPD

*AAA

AFSFDSGTN—---PTSTADSGNPADTLRVSPMIRDFVQSIDDREWQTSLFSLLQSQSYNQC

KLCFDSGTH----PSSTADA--LIEPLRVSPMIREFVQSIDDREWQTQLFALLQKQTYNQV
AFTFDTQSNTAATPSSTAEA-TSTETLRVSPMIREFVQTVDDREWQNALFGLLQSQTYNQC

*A***A’\ A *‘k‘k‘k****A***AA****‘k'kAA‘k'kA'k‘kir/‘i/‘ir'k‘k

r——> LBD “'3
EVDLFEL-MCKVLDQNLFSQVDWARNTIFFKDLKVDDQMKLLQNSWSDMLVLDHIHQRLH

EVDLFELLMCKVLDQNLFSQVDWARNTVFFKDLKVDDQMKLLQHSWSDMLVLDHLHHRIH
EVDLFEL-MCKVLDQNLFSQVDWARNTVFFKYLKVDDQMKLLQDSWSVMLVLDHLHQRMH

******* *******************A*** *‘k*********"*** ******A*A*A*

Pr-4
NGLPDETTLHNGQKFDLLGLGLLGVPSLABHFNELQNKLQELKFDVGDYICMKFLLLLNP

NGLPDETQLNNGQVFNLMSLGLLGVPQPGDYFNELQNKLQDLKFDMGDYVCMKFLILLNP
NGLPDETTLHNGQKFDLLCLGLLGVPSLADHFNELQNKLAELKFDVPDYICVKFMLLLNP

******* *A'k‘kir *A*’\ *******A AAA‘k‘ki'k'k'k‘k'ki'A‘k‘k'k‘kA **A*A**AA****

A **A AA

VEVRGITNRKTVVEGYENVQAALLDYTLTCYPSVPDKFSKLLSIIPEIHAMATRGEEHLY

—SVRGIVNRKTVSEGHDNVQAALLDYTLTCYPSVNDKFRGLVNILPEIHAMAVRGEDHLI
-EVRGIVNVKCVREGYQTVQAALLDYPYL-LSTIQDKFGKLVMVVPEIHALRL-GEKSTC

****A* * * **AAA******** AA *** *A AA*****A **
IKHCAGSAPSQTLLMEMLHAKRK 840

TCTPSTVPAVRPPKRCSWRCCTPSARDRGRENVTRNT 816
TSGIVQARHLPRLFSWKCCTQNANLEVPVTNKVEELRSAKPRRHHNK 555

149

366

312
118

Pr—2

427

373
179

487

423
225

548

484
264

581

545
272

638

600
332

697

660
391

757

720
451

817

779
508

the FTZ-F l subfamily bind as a monomer to a 5’-PyCAAGGPyCPu-3’ DNA response
element. The P-box from the DBD recognizes the core motif 5’-AGGPyCPu-3’, and the
F TZ-F 1 -box recognizes the first three base pairs 5’—PyCA-3’ of the target DNA-element
(Ueda and Hirose, 1991; Wilson et al., 1993; Ueda et al., 1992). The mosquito AaFTZ-
F 1 had a P-box with conserved CfiCKfi amino acids, which were characteristic for
members of the FTZ-F 1 subfamily. The DBD and FTZ-F 1 box of AaFTZ-Fl were
highly conserved relative to the corresponding sequences in insect homologs, and also
shared a high similarity with those in vertebrate FTZ-F l homologs, exhibiting 83%
identity with mouse LRH-l (Tugwood et al., 1991) and 85% identity with bovine Ad4BP
(Honda et al., 1993).

In the hinge subdomain, which connects the F TZ-Fl box and the putative LBD,
significant conservation was retained only among insect FTZ-Fls. The conserved
sequences were interrupted by a non-conserved spacer sequence (marked with vertical
arrowheads in Fig. 3) in each insect receptor. The A/B-domains of the insect FTZ-Fl
receptors were quite divergent (not shown).

Comparison of the mosquito LBD with the LBDs of insect and non-insect FTZ-Fl
homologs showed that there was limited identity in LBDs between mosquito FTZ-F 1 and
its vertebrate counterparts, exhibiting only 46% identity with mouse LRH-l and 42%
with bovine Ad4BP. However, a high level of identity was retained in the LBDs of insect

FTZ-F ls, showing 70% with DmFTZ-F l and 66% with BmFTZ-Fl.

150

Temporal expression of A aF T Z-F I in the mosquito fat body during the first
vitellogenic cycle

To determine the pattern of AaF 72-F1 transcripts in the female fat body,
Northern blot was performed for analysis of poly(A) RNA isolated from 100 fat bodies at
each time point. Three critical time points during mosquito vitellogenesis were chosen:
the early previtellogenic stage at 1 day post emergence (Fig. 4A, PV), the late
vitellogenic stage at 36 hr PBM (Fig. 4A, PBM 36h) when the 20E titers are low, and the
middle vitellogenic stage at 18 hr PBM (Fig.4A, PBM 18h) when 20E is at the peak
level. The 10 kb and 5.2 kb AaFTZ-FI mRNAs were detected in the fat body of 1d PV
and 36h PBM mosquitoes; in contrast, the fat bodies at 18 hr PBM contained no
detectable AaFTZ-Fl message (Fig. 4A, PBM 18h).

In order to examine the AaFIZ-FI developmental profile in this vitellogenic
tissue in more details, a more sensitive RT-PCR analysis was utilized. First, RT-PCR
was conducted by two pairs of primers located in the 5’UTR or between the hinge region
and LBD, and it appeared that the expression patterns of AaFTZ-F 1 during vitellogenesis
were the same (not shown). In the later experiments, the primer pair in the hinge region
and LBD was chosen. As demonstrated in Fig. 4B, the level of the AaF 72-F1 transcript
was high in the fat bodies of newly eclosed females, and its level declined slightly over
the course of previtellogenic development. After the onset of vitellogenesis, triggered by
a blood meal, the AaF 72-F1 transcript level dropped dramatically during the first 4 hr
PBM and then stayed at background levels until 24 hr PBM. Then, it rose again,
reaching the previtellogenic levels at 36 hr PBM, and stayed at a high level for the next

12 hr of post-vitellogenesis. Previously, it has been demonstrated that the AHR3 gene is

151

 

 

 

 

 

 

 

 

 

 

 

A __EEM_.
PV 18h 35h
1on1: d R
5.2kb —‘ “e.
8 PV (d) PBM (h)
0-1 1-2 3-5 1 3 4 6 15 24 33 48
b n - “ — ~— “-flj FTZ-F1

 

 

 

i
c l . «an. - VCP

C 0 6 12 1 8 24 30 36 (h)

"-1

* m...

Fig. 4. AaFTZ-FI expression profiles in the female mosquito fat body and pupae. (A)
Northern blot analysis of AaF TZ—F 1 in fat bodies. Poly(A) RNA was extracted from fat bodies
of 100 female mosquitoes at each point, PV: previtellogenic stage 0-1 day post-eclosion; PBM
18h or 36h: post blood meal 18 or 36 hours. The region between EcoRV site and Pr-4 (Fig.3) of
AaFTZ-Fl was used as a probe for hybridization. (B) Temporal expression of AaF IZ-F I in the
fat body during the first vitellogenic cycle by RT-PCR/Southem analyses. A. The early-late gene
AHR3 expression profile. The 0.22 kb PCR product was generated with the AHR3-specific
primers and hybridized with a 32P-labeled AHR3 specific DNA probe (see Methods and
Materials). B. AaFTZ-Fl expression profile. The 0.2 kb PCR-fragment was generated with the
AaFTZ-FI-specific primers Pr-3 and Pr-4 and hybridized with a 32P AaFIZ-FI specific probe
(shown in Fig. 3). C. Expression profile of the mosquito vitellogenic carboxypeptidase (VCP)
(Cho et al., 1991). VCP is shown as an independent control for RT-PCR/Southem analyses of an
ecdysone-inducible late gene in the mosquito fat body. (C) Temporal expression of AaF TZ—F 1
during pupa development analyzed by RT-PCR/Southem analyses. Expression profiles of AHR3
(HR3) and AaFTZ—FI (F TZ-F l) are shown. Each lane contained the same mosquito pupa
equivalents of total RNA subjected to RT-PCR/Southem analyses.

 

 

 

 

 

 

 

 

 

 

152

expressed in a narrow window correlated with the high levels of ecdysteroid at 18-24 hr
PBM in the mosquito fat body (Kapitskaya et al., 2000). The post-vitellogenic rise in
AaFTZ-FI mRNA levels occurred immediately after the AHR3 mRNA peak and in
correlation with a drop in the ecdysteroid titer (Fig. 4B).

The high level of the AaFIZ-FI transcript in the fat body of the early
previtellogenic females suggested that it might be activated prior to emergence by a
falling ecdysteroid titer in late pupae. To test this hypothesis, RT-PCR was performed
for analysis of the levels of A aF TZ-F 1 and AHR3 transcripts during development of
female mosquito pupae. Developing female pupae were collected at 6 hr intervals, from
0 hr (newly-pupated) to the final collection of late pupae at 36 hr post-pupation (adult
mosquitoes normally emerge within the next 6 hr). RT-PCR analysis was then performed
using RNA collected at different intervals of pupal development. AHR3 transcript levels
increased during early and mid-pupal stages, reaching its peak in 24 hr old pupae. By 30
hr, the AHR3 transcript level decreased dramatically and remained low in late pupae. The
A aF TZ-F 1 transcript level exhibited a sharp increase in 18-24 hr old pupae and remained

at the elevated levels until the end of pupal development (Fig. 4C).

Dual effects of 20E on AaF T Z—F 1 mRNA in cultured mosquito fat bodies

To determine whether the negative correlation between AaFTZ-FI expression and
the levels of ecdysteroids observed in vivo reflected the involvement of 20E in regulation
of the A aF IZ-F 1 gene, I investigated the patterns of AaF TZ—F 1 transcript accumulation
in in vitro cultured fat bodies. Previtellogenic fat bodies, 3-5 days after eclosion, were

incubated for up to 16 hr in culture medium either in the presence or absence of 10'6 M

153

20E. RT-PCR analysis was used to monitor the transcript levels of AaF IZ-F 1 in the
cultured organs (see Material and Methods). Fat bodies from previtellogenic 3-5 day-old
females were used to determine the background levels of transcripts for each ecdysone-
regulated gene including two isoforms of an early gene E 75 (AaE 75A and AaE75B), an
early—late gene AHR3, and a late gene VCP. Culturing the previtellogenic fat bodies in
hormone-free medium up to 16 hr did not induce AaE 75, AHR3, or VCP mRNAs (Fig. 5).
The AaF 72-F1 mRNA level, which was relatively high in the previtellogenic fat body,
was not effected by the 16 hr incubation in hormone-free medium (Fig. 5, FTZ-F 1). In
contrast to A aE 75, AHR3, and VCP genes, which were induced by the presence of 20E in
culture medium, the AaFTZ-FI mRNA dropped to an undetectable level after a 12 hr
incubation with 10'6M 20E (Fig. 5, 2013).

In the next experiment, fat bodies were cultured in hormone-containing medium
for only 4 hr, then washed three times in hormone-free medium and incubated in
hormone-free medium for an additional 12 hr. Both AaE 75A and AaE75B levels
dramatically increased after a 4 hr incubation in 20E-containing medium, and then
dropped sharply following the withdrawal of the hormone from medium (Fig. 5, E75A,
E75B). AHR3 and VCP mRNA levels, which rose moderately during 4 hr incubation
with 20E, also declined (Fig. 5, HR3, VCP). In contrast, the AaFIZ-FI mRNA exhibited
a considerable elevation of its level following the withdrawal of 20E from the incubating
medium (Fig. 5, FTZ-Fl).

To determine the requirement of protein synthesis for 20E-dependent activation or
repression of the AaF 72-F1 gene in the mosquito fat body, the previtellogenic fat bodies

were incubated with cycloheximide (Chx), a protein synthesis inhibitor which reversibly

154

120

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100« E75A
E 80
% +205.“
g 60 +205
5 +0117!
2 40
82
20
° - =- — L‘543—J
o 4
120 o 12 16
<
z
o:
— —o—205.4n
g +205
:5 p—o—cm
33
100.
g so
5 41—20541
g 60 +205
é —a—CM
40
a?
20«
0
120
g 100 FTZ-F1
a: 801
E 60
i 40
39
20.
0
0 4 a 12 1a
120
100
g VCP
{E 5° +205.“
2 60 +205
g —o—CM
5 40.
BE
20
o .. er - a
o 4 a 12 16

Culture Tlme (hr)

Fig. 5. In vitro repression of AaF TZ—F 1 expression by the hormone 20E, and super-
induction of AaF T Z-F I by withdrawal of 20E in fat body organ culture. RT-
PCR/Southem analyses were performed for each ecdysone-regulated gene in mosquitoes
as follow: early gene AaE 75A (E75A) and AaE75B (E75B) isoforms (Pierceall et al.,
1999); early-late gene AHR3 (HR3) (Kapitskaya et al., 2000); AaFIZ-FI (this paper);
late gene VCP (VCP) (Deitsch et al., l995a; l995b). For the in vitro fat body culture and
RT-PCR/Southem analyses, see Methods and Materials. Closed squares represent a
continuous treatment with 20E (20E); open squares, a 4hr treatment with 20E followed
by an additional culture in 20E-free medium (20E-4h); open triangles, a control medium
without 20E (CM). Each point represents an average of three independent experiments :
SD.

155

represses YPP protein synthesis in this tissue in a dose-dependent manner (Deitsch et al.,
l995a; 1995b). The fat bodies were first pre-incubated in hormone-free culture medium
for 1 hr in the presence of 104M Chx, and then incubated with the same dose of Chx but
with the addition of 10'6M 20E for various times (Fig. 6). While incubation of the
previtellogenic fat bodies in the presence of Chx in the hormone-free medium did not
induce AaE 75, AHR3, and VCP mRNAs, the AaFIZ-FI transcript levels were
considerably elevated (Fig. 6, Chx). Addition of 10‘6 M 20E to Chx-containing medium
activated A 0E 75 and AHR3 genes to the levels much higher than those attained with 20E
alone; the levels of AaF TZ-F l mRN A were reduced compared with those in hormone-
free, Chx-containing medium (Fig. 6, Chx/20E). Withdrawal of 20E from Chx-
containing medium after 4-hr incubation caused a slower decline in the AaE 75A, AaE75B
and AHR3 mRNA levels than 20E withdrawal from Chx-free medium (Fig. 5, 20E-4h;
Fig. 6, Chx/2OE-4h). Interestingly, the AaFTZ—F! gene was super-induced under the
same conditions (Fig. 6, FTZ-Fl). The V CP gene was not responsive to any stimulation

in the presence of Chx (Fig. 6, VCP).

In vitro enhancement of ecdysone-responsive vitellogenic competence following
over-expression of AaF T Z-F I

Withdrawal of 20E from the fat body culture after an initial 4 hr incubation in the
presence of 20E resulted in a considerable elevation of AaF TZ—F 1 mRNA (Fig. 5, FTZ-
Fl, 20E-4h). If AaFTZ-F l is a competence factor for 20E sensitivity, one might expect
that these conditions should lead to enhanced responsiveness of the fat body to 20E. To

test this hypothesis, I re-introduced the fat body, after treatment for 16 hr as in Fig. 5

156

 

Relative RNA Level

 

 

——o-— Chx/20E-4h
—.—— Chx/206

 

 

 

 

 

 

Relative RNA Level

 

 

 

-—-O— Chx/2054b
+ Chx/ZOE
—o— Chx

- - O - -ZOE

 

 

 

 

Relative RNA Level

 

 

 

 

Relative RNA Level

 

 

 

 

—-o-—Chx/205-4h
+0hx/206
—o—-Chx

- - D - -20E

 

 

 

 

--.'°2OE

 

—o— Chx/206411
+ Chx/205
—o— Chx

 

 

 

 

 

 

E VCP
—’ 801
.5. 1"
CE 50 .‘
O .'
E. 40«
8 -'
Ct 20 .'
0‘...
O -" - ‘ m
0 4 8 12

Culture Tlme (hr)

Fig. 6. Effects of cycloheximide (Chx) on the 20E regulation of AaF TZ—F 1 in in vitro
fat body organ culture. The mRNA levels of each AaE 75A (E75A), AaE7SB (E75B),
AHR3 (HR3), AaFIZ-FI (FTZ-Fl) or VCP (VCP) gene were examined during culturing
with 104 M Chx in hormone-free medium (Chx), in the continuous presence of 104 M
Chx and 10'6 M 20B (Chx/20E), or in the 4 hr exposure to 10'4 M Chx and 10'6 M 2013
followed by culturing witth“1 M Chx in hormone-free medium (Chx/20E-4h). Dashed
lines (20B) are from the previous experiment (Fig. 5, 20E), and all other data are relative
to these values. RT-PCR/Southem blot hybridization were performed using a probe
specific for each 20E-responsive gene. Each point represents an average of three

independent experiments i SD.

157

16

 

—o—- Chx/2064b
—-.-— Chx/20E
-——o—— Chx

 

 

--..-.205

 

 

(20E-4h) with the elevated level of AaF IZ-F 1 mRNA, to a 20E-containing medium and
examined its capacity for a secondary 20E-activation of the early E 75, early-late HR3,
and late VCP genes (Fig. 7). In the first experiment (Fig. 7A), the tissue was incubated
in the presence of 10'6 M 20E for 4 hr or for 16 hr. The 4-hr secondary incubation
resulted in super-induction of the early AaE 75 gene, with the level of AaE 75A mRNA
more than three-fold higher than that seen in the 4-hr primary 20E-incubation (Fig. 7A).
The HR3 and VCP transcript levels were also increased compared to the primary
incubation. However, their levels increased even further after a 16-hr secondary
incubation. In particular, this secondary incubation resulted in a 2-fold induction of the
late VCP gene compared to that in the primary 20E incubation (Fig. 7A). Thus, this test
suggests that the fat body with the experimentally increased level of AaF 72-F1 mRN A
was competent to activate the early, early-late, and late 20E-responsive genes to a much
higher level.

Next, I modified the second part of the experiment. As in the first experiment, the
previtellogenic female fat bodies were first incubated in 20E-containing medium for 4 hr,
followed by 12 hr incubation in the hormone-free medium. However, the secondary 20E
incubations were performed in the presence of 10‘4 M Chx (Fig. 78). Under these
conditions, A aE 75A showed even higher induction after 4-hr incubation (8-fold).
Furthermore, AaE 75A transcript continued accumulation reaching the l6-fold level of
increase after 16-hr incubation. Likewise, the early-late HR3 gene transcript showed
continuing accumulation. In contrast, the secondary activation of the late VCP gene was

inhibited (Fig.7B).

158

 

  

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

 

A 400
} i-576A'
g a°° ' l-HRS
lDVCP
E 100 l m
0 , L :h -
Oh 4 h 16 h + Oh + 4n +16h
1st Responoe (20E) 2nd Response (ZOE)
a 2400
2000 - -
3 1.575%
g 1600 l-HR3
1200 lDYCP .
‘ 800
400
O e M
Oh 4 h 16 h + on + 411 +16h
1st Response (ZOE) 2nd Response (20E+Chx)
1000
“-5752; l
i-HRS ,
19.1.1931

 

 

NormalzedRNAlevelo
e§§§§

 

 

on 4 h 16 h + on + 4n + 16h
1st Response (20E+Chx) 2nd Response (205+Chx)

 

 

Fig. 7. Secondary responses to 20E of the ecdysone-inducible genes in in vitro fat body culture. The
primary response experiments were performed as previously described (Fig. 5, 20B, or Fig. 6, Chx/20E)
and presented for comparison with the secondary responses. The secondary response experiments were
performed following the 20E withdrawal experiment: the previtellogenic competent fat body was first
incubated with 10'6 M 20B for 4 hr and then in hormone-free medium for 12 hr in the absence or presence
of 10" M Chx (Fig. 5, 20E-4h, or Fig. 6, Chx/20E-4h, respectively). The secondary responses were
evaluated in such a fat body after additional incubation with 20E for 4 hr or 16 hr in the absence (20E) or
presence of Chx (20E + Chx). RT-PCR/Southern analyses were conducted for the cultured fat bodies. (A)
Super-induction of ecdysone-inducible genes (E75A, HR3 and VCP) in the secondary responses to 20E.
Both the primary and secondary response experiments were performed in the presence of 10" 20B. (B)
Effects of Chx on the secondary responses of 20E-responsive genes. The primary response experiment was
performed as in Panel A. The 20E withdrawal prior to the secondary incubation was performed in the
presence of 10‘ M 20E only (as in Fig. 5, 20E-4h), while the secondary response experiment was in the
presence of 10" M Chx and 10'6 M 20B. (C) Abolishment of the secondary responses of 20E-inducible
genes by Chx present in the initial 20E withdrawal incubation. In the primary responses, E 75 and HR3
genes were super-induced by 20E in the presence of 10" M Chx. However, after the 20E-withdrawal
experiment performed in the presence of the same concentration of Chx, as in Fig. 6, Chx/20E-4h, the
secondary induction of E 75 and HR3 genes was inhibited. For each 20E-inducible gene, data were
normalized based on the I6-hr 20E induction in the primary response, which was referred to as 100. Each
point represents an average of three independent experiments : SD.

159

In the final experiment, I conducted both the initial (Fig. 6, F TZ-Fl, Chx/20E-4h)
and secondary incubations in the presence of 10“t M Chx. Under these conditions,
activation of all 20E-responsive genes during the secondary response was abolished (Fig.
7C). This suggests that the AaFTZ-F 1 protein synthesis resulted from the initial
incubation without Chx was required for the secondary activation of the 20E-responsive

genes.

The onset of ecdysone-response competence was correlated with the appearance
of functional AaF T Z-F 1 protein

The fat body of a newly emerged female mosquito was not competent to respond
to 20E, and the competence was acquired by this tissue as a result of the previtellogenic
development (Fig. 2). However, the AaFTZ-FI gene was highly expressed in this tissue
throughout the previtellogenic development (Fig. 4B, FTZ-Fl). To resolve this apparent
contradiction, investigation was conducted on whether the AaFTZ-Fl protein was present
at the first day after eclosion, or the appearance of the functional AaFTZ-Fl protein could
correlate with the acquisition of competence for 20E-vitellogenic response by this tissue.
To test this hypothesis, EMSA was performed for nuclear extracts from previtellogenic
fat bodies and the FTZ-Fl recognition element (FIRE) from in the upstream regulatory
region of Drosophilafushi tarazu (ftz) gene. This 9-bp binding sequence for FTZ-F 1 (5’-
YCAAGGYCR-B’) was identified by an in vitro DNase I footprinting assay (Ohno et al.,
1994)

First, it was evaluated whether the in vitro translated AaFTZ-F 1 protein was able

to bind to the FIRE (Fig. 8A, lane 2). The specificity of binding was confirmed by

160

A B

TNT-beta FTZ-Fl - + + + + + -----
3—5d PV FBNE ------ + + + + + TNT-beta FTZ-Fl - + - -
unlabeledFlRE --+----+--- 0-0.5dPVFBNE -'+-
nonspecificDNA ---+----+-- 3-5dPV FBNE ---+
anti-DmFTZ—Fl ----+----+-
prc-immunc scrum - +

    

 

 

 

123456789101

 

1234

Fig. 8. Identification of the functional AaFTZ-Fl protein in fat body nuclear
extracts (F BNE) of previtellogenic female mosquitoes. (A) EMSA analysis of the
AaFTZ-Fl protein translated in vitro (TNT-beta FTZ-Fl) or in the FBNE from
mosquitoes, 3-5 days after eclosion. The protein samples were incubated with the 32P-
labeled FTZ-F 1 response element, F 1RE. The specificity of the interaction between the
FIRE and AaFTZ-F l was examined by the addition of SO-fold molar excess of the
unlabeled FIRE (lanes 3 and 8) or a double-stranded nonspecific DNA (lanes 4 and 9).
In addition, anti-DmFTZ-F 1 antibodies (lanes 5 and 10) and pre-immune serum (lanes 6
and 1 l) were added to the reactions. (B) EMSA analysis of the AaFTZ-F 1 protein in

F ENE at different stages of the previtellogenic mosquitoes: lane 1, negative control; lane
2, TNT—AaFTZ-F 1; lane 3, FBNE from newly eclosed previtellogenic females, 0-12 hr
post-eclosion; lane 4, FBNE from previtellogenic females, 3-5 days post-eclosion.

l6]

competition with an excess amount of the specific unlabeled F 1RE (Fig. 8A, lane 3) and
by co-incubation with anti-DmFTZ-Fl antibodies (Fig. 8A, lane 5). Both of these
treatments eliminated the binding complex. The EMSA analyses were further carried out
by incubating the fat body nuclear extracts (F BNE) of the 3-5-day old female mosquitoes
with 32P-labeled F 1 RE. Two retarded DNA-protein complexes were observed (Fig. 8A,
lane 7), one of which was similar in mobility to the complex formed by F 1RE and in vitro
translated AaFTZ-Fl protein. Both bands formed by the nuclear extract were removed
by competition with an excess amount of the specific unlabeled F 1RE (Fig. 8A, lane 8) or
by co-incubation with anti-DmFTZ-Fl antibodies (Fig. 8A, lane 10), indicating that these
complexes were the specific binding of AaFTZ-F 1 to F 1RE.

Next, the DNA binding activity of AaFTZ-F l in the FBNE of the female
mosquitoes at O-O.5 day after eclosion was tested. No AaFTZ-Fl-containing complex
was detected in the F BNE of the newly-eclosed females (Fig. 8B, lane 3). In contrast, the
AaFTZ-Fl binding was clearly seen in the 3-5 day-old FBNE used as a positive control
in the same experiment (Fig. SB, lane 4). Testing of the FBNE from mosquitoes of
intermediate previtellogenic ages, 1 and 2 day-old, was inconclusive due to variability in
development and limitation in the sensitivity of the assays. Interestingly, when the FBNE
from the vitellogenic mosquitoes were used in EMSA, the results showed that the
intensity of the AaFTZ-Fl-containing complexes decreased sharply after a blood meal
and disappeared completely by the peak of vitellogenesis. The AaFTZ-Fl-containing
complexes appeared again in the FBNE from the post-vitellogenic mosquitoes (data not

shown). Thus, these analyses confirmed that the acquisition of competence for 20E-

162

vitellogenic response by the mosquito fat body correlated with the appearance of the

functional AaFTZ-Fl protein.

DISCUSSION

The Drosophila F TZ-F 1 gene, located within the 75CD mid-prepupal salivary
gland polytene chromosome puff, includes two overlapping transcription units encoding
(1F TZ-Fl and BFTZ-F 1, two orphan nuclear receptor isoforms differing only in the N-
terminal part of their A/B domains (Ueda et al., 1990; Lavorgna et al., 1991; 1993).
BFTZ-Fl has been shown to play a central role in the acquisition of competence for the
late prepupal ecdysone response at the onset of Drosophila metamorphosis (Woodard et
al., 1994; Broadus et al., 1999). This nuclear receptor is repressed by ecdysone and
expressed during the low hormone titer interval in mid-prepupae (Lavorgna et al., 1993;
Woodard et al., 1994). lmmunocytochemical staining of polytene chromosomes with
anti-BFTZ-F 1 antibodies shows that BFTZ-F 1 protein is bound to several sites
corresponding to ecdysone-regulated genes, including the early genes, E 75 and E 74,
suggesting its direct involvement in regulation of these genes (Lavorgna et al., 1993).
Indeed, ectopic over-expression of BFTZ-Fl significantly increases the ecdysone
induction of the early genes BR-C, E 74A and E 75A (Woodard et al., 1994). Finally, a
recent study has shown that the ecdysone-triggered hierarchy, which directs the stage-
specific metamorphic transition, is attenuated in BFTZ-Fl mutants (Broadus et al., 1999).
These observations support the initial suggestion made from studying ecdysone responses
of puffs of Drosophila salivary gland polytene chromosomes, which has showed that

competence is acquired during the mid-prepupal period for re-induction of the BR-C,

163

E 74A and E 75A early puffs (Richards, 1976a; 1976b; 1976c), and clearly define BFTZ-Fl
as a key competence factor for stage—specific ecdysone responses in Drosophila
metamorphosis.

Ecdysone plays a critical role in controlling oogenesis and vitellogenesis in flies
and mosquitoes (Bownes, 1986; Hagedom, 1985; Raikhel, 1992; Buszczak et al., 1999;
Raikhel et al., 1999); however, limited information has been available concerning the
molecular basis of these ecdysone-mediated events. In particular, an important question
to be answered is how the same hormone, which regulates larval molts and pupal
metamorphosis, activates entirely different sets of target genes in the reproductive tissues
of an adult insect. In the case of mosquito, it is shown here that AaFTZ-F I likely serves
as a competence factor for the stage-specific responses to ecdysone in the fat body during
vitellogenesis. A homolog of Drosophila flFTZ-FI is expressed at relatively high levels
in the mosquito fat body during pre- and post-vitellogenic periods when ecdysteroid titers
are low. However, the AaFTZ-FI transcript nearly disappears during the vitellogenic
period with high ecdysteroid levels.

In Drosophila, another orphan nuclear receptor (DHR3) is implicated in

regulation of flF TZ—F 1 gene transcription. During Drosophila metamorphosis, DHR3 has

dual roles in both repressing the early genes and activating flFYZ—FI transcription (Lam
et al., 1997; White et al., 1997). The mosquito HR3 homolog, AHR3, has been cloned
(Kapitskaya et al., 2000), and its expression analysis concurs with the studies on
Drosophila metamorphosis, suggesting its role in AaFTZ-FI gene activation. The in vivo
developmental profiles of AHR3 in both the fat body and ovary have shown that the peak

of the AHR3 transcript coincides with a high ecdysteroid titer in the hemolymph

164

(Kapitskaya et al., 2000). In this study, the developmental profile of AHR3 in pupae was
also examined. Remarkably, the peak expression of AHR3 preceded expression of the
AaFTZ-FI gene in pupae and in the adult vitellogenic tissues, the fat body (this study)
and ovary (Kapitskaya, M., Li, C., and Raikhel, A., unpublished observation). The
sequential expression of AHR3 and AaFTZ-FI in the late pupa-early adult suggests that
these factors are involved in setting up stage-specific responses to ecdysone in the
previtellogenic fat body for the first vitellogenic cycle. Likewise, expression of AHR3
and AaFYZ-FI during the late vitellogenic period is likely preparing the fat body for a
second vitellogenic cycle.

Experiments with in vitro organ culture strongly supported the hypothesis that
AaFTZ-F 1 expression is involved in the acquisition of competence for ecdysone response
in the adult mosquito fat body. In the previtellogenic fat body, AaFTZ—FI expression was
inhibited by 20E and strongly activated by its withdrawal. Similarly, in the silkworm
Bombyx mori, BmFTZ-Fl is inducible by a decline in the ecdysteroid titer (Sun et al.,
1994). Addition of Chx into the medium in withdrawal experiments also led to super-
induction of the mosquito F TZ-F 1 , suggesting that AaFTZ-FI expression is repressed by
one or more unstable repressors. Importantly, after in vitro over-expression of AaF TZ-F 1
in the previtellogenic mosquito fat body, a secondary 20E challenge resulted in super-
induction of the early AaE 75 gene and the late V CP gene. These experiments are
consistent with in vitro organ culture experiments in Drosophila which have shown that
,6FTZ—Fl is repressed by ecdysone, and that in vivo ectopic expression of ,677 TZ-F 1 in
late-third instar larvae enhances early gene induction (Woodard et al., 1994). The

acquisition of competence for ecdysone responsiveness by the mosquito fat body occurs

165

over the first two days of the previtellogenic development (Flanagan and Hagedom,
1977; this study). Although the newly emerged mosquitoes were not competent for
ecdysone response, AaFTZ-F] mRNA was abundant in their fat bodies. The explanation
for this apparent contradiction was revealed by EMSA analysis of AaFTZ-F 1 DNA
binding activity, which showed that the appearance of functional AaFTZ-F 1 protein,
rather than its mRNA, was correlated with the ecdysone-response competence in the
mosquito fat body. The previtellogenic development of the mosquito fat body, which
involves an increase in ploidy (Dittmann et al., 1989) and ribosome proliferation (Raikhel
and Lea, 1991), is controlled by J H III. Flanagan and Hagedom (1977) have shown that
the fat body responsiveness to ecdysone is also under JH regulation. Further studies are
required to elucidate whether JH is involved in the post-transcriptional control of AaFTZ-
F1 activity.

In conclusion, our findings indicate that a conserved molecular mechanism for the
stage specificity may be used during insect metamorphic and reproductive responses to
ecdysone. However, the unique feature of mosquito vitellogenesis is the apparent
involvement of post-transcriptional control of AaFTZ-F 1 in the acquisition of the
ecdysone-response competence. Elucidation of the molecular mechanism underlying this
unique characteristic of mosquito vitellogenesis represents an exciting and challenging

task for future research.

ACKNOWLEDGMENTS
I thank Marianna Kapitskaya for partial cloning of AaFTZ-F 1, J insong Zhu for

EMSA analysis and Ken Miura for RT/PCR analysis. I also thank William Segraves for

166

critical reading, Alan Hays for excellent technical assistance, Michel Trail for editing the
manuscript, and Carl Thummel for his generous gift of Drosophila FTZ-F 1 antibodies.

This grant was supported by NIH grant AI-36959.

167

Chapter 6

Cloning and Characterization of the Mosquito HR 78 Gene

168

ABSTRACT

The ecdysone gene regulatory hierarchy is mediated by a number of transcription
factors belonging to the nuclear receptor superfamily. Here, AHR 78 has been identified,
which is the mosquito homolog of Drosophila hormone receptor 78 (DHR78) and its
vertebrate counterpart, testicular receptor 2 (TR2). A full-length cDNA clone was
obtained from a mosquito vitellogenic fat body cDNA library. The complete ORF of
AHR78, consisting of 630 amino acids, exhibits the highest homology to DHR78 and
TR2. The AHR78 DBD displays 95% identity (98% similarity) with the DHR78 DBD
and 79% identity (89% similarity) with the TR2 DBD. During the previtellogenic period,
AHR 78 is expressed at extremely low levels in the fat body and previtellogenic ovary,
however, the AHR78 transcript is considerably elevated in the ovary during the
vitellogenic period, reaching its peak level at 24 hr post blood meal, which correlates
with the expression peak of the vitellogenin gene (Vg), a target of 20E in the fat body. In
in vitro culture, AHR 78 mRNA is induced by 20E in the ovary but not in the fat body.
Furthermore, AHR78 protein is capable of inhibiting the DNA binding and
transactivation activities of AaEcR/AaUSP on various ecdysone response elements
(EcRE), including Drosophila hsp27- and Eip28/29—EcREs, and the Aedes Vg-EcRE.
These results suggest that AHR78 may be involved in inhibition of the Vg gene

expression in the ovary during vitellogenesis.

169

INTRODUCTION

The major insect steroid hormone, 20E, is involved in regulating multiple
developmental processes including embryogenesis, larval molting, metamorphosis and
reproduction (Bownes, 1986; Raikhel, 1992; Riddiford, 1993a,b; Thummel, 1995). The
functional ecdysteroid receptor is a heterodimer of the ecdysone receptor (EcR) and a
retinoid X receptor (RXR) homolog, Ultraspiracle (USP) (Koelle et al., 1991; Yao et al.,
1992; 1993). EcR and USP play the key roles in the ecdysone gene regulatory
hierarchies during Drosoplzila metamorphosis (Bender et al., 1997; Hall and Thummel
1998). In addition, the formation of inactive heterodimers between DHR38 and USP,
SVP and EcR, as well as the DHR78 homodimer as a competitor, inhibit DNA binding
and activation by EcR/USP on the ecdysone responsive elements, providing different
levels at which ecdysone responses can be negatively controlled (Sutherland et al., 1995;
Zelhofet al., 1995a; Fisk and Thummel, 1995, 1998; Zelhof et al., 1995b).

In contrast to the large body of information concerning the molecular basis of the
ecdysone-mediated gene regulation during development and metamorphosis,
comparatively little is known about the ecdysone-regulated aspects of insect female
reproduction, especially the negative regulation of ecdysone responses. The mosquito is
an ideal model animal to study ecdysone-mediated gene regulation during reproduction.
This is due to the fact that mosquito reproduction is triggered by blood feeding, which is
a requirement for egg development. In addition, mosquitoes act as vectors for many
devastating human diseases, which increases the significance of current research (Collins
and Paskewitz, 1995; Butler etal., 1997; Beier, 1998). In the anautogenous mosquito

Aedes aegypti, the process of vitellogenesis is the cornerstone of egg maturation.

170

Vitellogenin is initiated only after a female mosquito ingests vertebrate blood. A blood
meal triggers a hormonal cascade which results in production of 20E, the terminal signal
that activates yolk protein precursor (YPP) genes in the fat body (Hagedom, 1983; 1985;
Raikhel, 1992; Dhadialla and Raikhel, 1994; Deitsch et al., 1995a; 1995b). Vitellogenin
(Vg) is one of the major YPPs. In addition, the mosquito fat body synthesizes several
other YPPs, vitellogenic carboxypeptidase (VCP), vitellogenic cathepsin B (VCB) and
lipophorin (Cho et al., 1991, 1999; Sun et al., 2000). YPP synthesis reaches maximal
levels at 24 hr post-blood meal (PBM) and then proceeds until 30-32 hr PBM, when it is
rapidly terminated (Raikhel, 1992). YPP genes are activated in the fat body and their
protein products are translocated through the hemolymph and finally internalized into the
developing oocytes (Raikhel, 1992).

The titers of ecdysteroid in the female mosquito hemolymph positively correlate
with the rate of YPP synthesis in the fat body. While 20E titers are only slightly elevated
at 4 hr PBM, they begin to rise sharply at 6-8 hr PBM, and reach their maximum levels at
18-24 hr PBM (Hagedom et al., 1975). Consistent with the initially proposed role of 20E
in activating vitellogenesis in anautogenous mosquitoes (Hagedom et al., 1973;
Hagedom and Fallon, 1973), experiments using an in vitro fat body culture have shown
that physiological doses of 20E (10'7 - 10’6 M) activate two YPP genes, Vg and VCP
(Deitsch et al., l995a; 1995b). Utilization of the protein synthesis inhibitor,
cycloheximide (Chx), has demonstrated that the activation of YPP genes in the mosquito
fat body requires protein synthesis (Deitsch et al., 1995a). Thus, it is likely that a
regulatory cascade similar to that seen in Drosophila development mediates the action of

20E in the mosquito fat body.

171

In A. aegypti, two isoforms of both the ecdysone receptor (AaEcR-A and AaEcR-
B) and Ultraspiracle (AaUSP-A and AaUSP-B). have been identified. Also AaEcR-B and
AaUSP-B function as major heterodimerization partners during vitellogenesis (Cho et al.,
1995; Kapitskaya et al., 1996; Wang et al., 2000a; Wang, Li, and Raikhel, unpublished
data). Mosquito EcR and USP are expressed in pre- and vitellogenic fat bodies and
ovaries (Cho et al., 1995; Wang et al., 2000a). The AaEcR/AaUSP heterodimer has been
shown to bind to various ecdysone response elements (EcREs) to modulate ecdysone
regulation of target genes (Wang et al., 1998). Most recently, we have demonstrated that
the Vg S’- regulatory region contains a functional EcRE that is necessary to confer its
responsiveness to 20E. AaEcR/AaUSP binds directly to the Vg EcRE, and activates the
Vg promoter in the presence of 20E in Drosophila 82 cells (Martin et al., 2001).

Furthermore, studies in the previous chapters have demonstrated that

conservation of the ecdysone-responsive gene regulatory hierarchy extends beyond EcR
and USP. A mosquito homolog of the Drosophila early gene E 75 is involved in
mediating the ecdysteroid responses during vitellogenesis (Pierceall et al., 1999).
Mosquito AaFTZ-F 1 serves as a competence factor for the vitellogenic response to
ecdysone, which is implicated in activation of AaE75 and YPP genes (Li et al., 2000).
The orphan nuclear receptor AHR3, a mosquito homolog of DHR3, has also been cloned,
and is thought to be involved in the activation of AaF TZ-F 1 during vitellogenesis
(Kapitskaya et al., 2000; Li et al., 2000). In addition, AHR3 8, the mosquito homolog of
DHR38 and vertebrate NGFI-B, inhibits the ecdysone response in the mosquito fat body
during the previtellogenic stage by strongly interacting with the AaUSP protein and

preventing the formation of a functional ecdysteroid receptor (Zhu et al., 2000). Thus,

172

despite the functional diversity of genes targeted by ecdysone-mediated gene regulation,
some molecular mechanisms of the ecdysone gene regulatory hierarchy appear to be
conserved between mosquito vitellogenesis and Drosophila metamorphosis.

In this study, I report the cloninglof mosquito HR78, a homolog of Drosophila
DHR78 (Zelhof et al., I995a; Fisk and Thummel, 1995) and vertebrate TR2 (Chang and
Kokontis, 1988). DHR78 is an orphan receptor, which can inhibit ecdysone signaling
and is required for ecdysone-regulated gene expression in mid-third instar larvae in
Drosophila (Fisk and Thummel, 1995, 1998; Zelhof et al., 1995b). While the mosquito
AHR78 DBD is highly conserved relative to the DHR78 DBD, the LBDs in this
subfamily show less similarity relative to the other nuclear receptor subfamilies,
suggesting that the AHR78 subfamily has evolved rapidly, particular in the E domain.
AHR78 is expressed at extremely low levels in the fat body. In the ovary, AHR78
mRNA levels are relatively low during the previtellogenic stage; however, its levels
increase dramatically during the vitellogenic stage with a peak level at 24 hr post blood
meal. Moreover, AHR78 inhibits binding by EcR/USP to natural EcREs and inhibits the
transactivation of EchU SP in cell transfection assays. Most significantly, AHR78
inhibits DNA binding'and transactivation of EcR/USP on the Vg promoter, suggesting
that AHR78 may be involved in inhibition of Vg gene activation in the ovary during

vitellogenesis.

173

MATERIALS AND METHODS

Animals

Mosquitoes, A edes aegypti, were reared according to the method of Hays and
Raikhel (1990). Larvae were fed on a standard diet as previously described (Lea, 1964).

Vitellogenesis was initiated by allowing females, 3-5 days afier eclosion, to feed on an

anesthetized white rat.

cDNA cloning

Degenerate primers based on the specific-conserved sequences in DBDS between
DHR 78 and mTRZ were designed for PCR to obtain a mosquito homolog HR78 cDNA
fragment. Two forward degenerate primers, AHR78-F1 (GCN WSI GGI MGI CAY
TAY GII G) and AHR78-F 2 (TGY GAR GGI TGY AAR GGN TTY TT), and one
reversed primer, AHR78-R (TGR CAI CKR TTI CKR TGR TGY TT) were utilized for
nested PCR. Amplification was carried out in a Perkin Elmer thermal cycler using a
template of cDNA produced by reverse transcription from total RNA prepared from the
fat bodies of the vitellogenic female mosquitoes. The PCR—generated fragment was used
as a probe to screen a kZAPII cDNA library prepared from the fat bodies of vitellogenic
female mosquitoes during 6-48 hr post-blood meal (PBM) as previously reported (Cho
and Raikhel, 1992). Several positive cDNA clones were subsequently isolated and the
one with longest cDNA insert was sequenced at W.M. Keck Foundation Biotechnology
Resource Laboratory, Yale University. Analyses of nucleotide and deduced amino acid
sequences were performed using the software of GCG (Genetics Computer Group,

University of Wisconsin, Madison).

174

 

 

RT-PCR/Southern blot

RT-PCR/Southem analyses were performed as previously described (Li et al.,
2000). Total RNA was prepared from fat bodies throughout the first vitellogenic cycle
by TRIZOL Reagent (Gibco BRL). After reverse-transcription by Superscript II reverse
transcriptase (Gibco BRL), 0.025 fat body equivalent of the cDNA pool was used as a

PCR template for each point for the analyses of developmental profiles. Thermal cycling
conditions were as follows: 940C for 3 min, followed by 18 cycles of 94°C for 30 sec,

600C for 30 sec and 720C for 30 sec. The PCR primers for AHR78 are AHR78-F 33
(GCT TC C CAA CTA CCT GAG TGC G) and AHR78-R73 (GCC AGC TGG CCG
CTA CTC T), both of which are located in the ligand binding domain of AHR78. As a
reference to the developmental change of the fat body and the ovary, the samples of the
same cDNAs were subjected to the 13 cycles of PCR with each primer pair specific to the
mosquito vitellogenic carboxypeptidase (VCP) or vitellogenin receptor (VgR). Southern
blot hybridization was performed under high stringency conditions using each gene-

specific probe, which was generated by PCR using each cloned cDNA as a template.

In vitro fat body and ovary culture

Mosquito fat body culture experiments were performed as previously described
(Li et al., 2000). The ovary culture experiments were performed according to the method
by Sappington et a1. (1997). After culturing, RT-PCR/Southem blot were utilized to

analyze the mRNA levels.

175

Electrophoretic mobility shift assays

 

Electrophoretic mobility shift assays (EMSA) were carried out as described
previously (Wang et al., 2000a). Nuclear receptor proteins were synthesized in vitro
using a coupled transcription-translation (TNT) kit from Promega. AaEcR and AaUSPb
proteins were expressed with the SP6 promoter from the vectors pGEM3Z-AaEcR and

pGEM7Z-AaUSPb, respectively. AHR78 protein was expressed from pBluescriptSK-

 

AHR78 with the T3 promoter. The double strands of EcREs are: Vg-EcRE,
agctgthGACCTagcgggAGGCCAaTGGTCGagt; Hsp27,

agagacaagGGTTCAaTGCACTtgtccaat; Eip28/29, ttaaAGGATCtTGACCCcaaT
GAACthat. Five pmols of each EcRE was end-labeled with 7-321) ATP (Dupont NEN)
by T4 polynucleotide kinase (Gibco BRL). In each binding reaction, typically 1 pl TNT
protein and 100 fmol labeled EcRE were incubated in total 20ul of binding buffer
containing 20mM HEPES (pH7.5), 2mM DTI‘, lOOmM KCl, 7.5% Glycerol, 0.1% NP-
40, 2 pg of poly (dI-dC)o poly (dI-dC) (Pharrnacia Biotech). In some experiments, the
binding buffer also contained 2 pg of non-specific single stranded DNA to eliminate non-
specific binding (Wang et al., 1998). The binding reaction was conducted for 30 min at

room temperature.

Cell culture and transient transfection assay
Cell culture and transient transfection assay were performed as previously
described (Wang et al., 2000b). The Drosophila S2 cell line (Invitrogen) was

maintained at 22-24 °C in Schneider Drosophila medium supplemented with 10% heat-

176

inactivated fetal bovine serum, penicillin (100U/ml), and streptomycin (IOOug/ml)
(Gibco BRL). Transfection was conducted with Cellfectin (Gibco BRL). Typically,
lOOng of the luciferase reporter plasmid and 12.5 ng of the reporter pAc5-LacZ plasmid
as well as a total 180ng of all receptors and pAc5 plasmids, were mixed in a 48-well plate
with 2 ug of Cellfectin in a total volume of ZOul, and then incubated at room temperature
for 30min. The transfection cocktail was overlaid with 200 pl of 82 cells (2x106 cells/ml)
in Drosophila serum-free medium. Transfection was terminated 12 hr later by adding
200ul 20% fetal bovine serum media containing either 2><10'6 M 20E or an equal amount
of vehicle ethanol. After treatment with the hormone 20E for 27hr, cells were collected
and lysed by 100 pl lxreporter lysis buffer (Promega) with one cycle of freezing and
thawing. Reporter gene assays were conducted as described by the Promega firefly
luciferase reporter and B-galactosidase systems. A luminometer (Turner Designs model
TD20e) was used to detect luciferase activity with 10 sec pre-delay and 30 sec integration
time. Luciferase activities were normalized by B-galactosidase activities in each
experiment. The reporter plasmids AMTV-Eip28/29-luc and AMTV-hsp27-luc were kind
gifts from M. Mckeown (Salk Institute, San Diego, CA). The AMTV-Eip28/29-luc
construct contains four copies of Eip28/29 EcRE and the AMTV-hsp27—luc has five
copies of hsp27 EcRE. The pAc5-AaEcR and pAc5-AaUSPb constructs contained the
full length AaEcR and AaUSPb (Wang et al., 2000b). The pAc5-AHR78 plasmid was
constructed by digesting pBluesript SK-AHR78, which was directly from the mosquito
lZaplI cDNA library, with XbaI and ApaI and subcloning the ORF into the pAcS
XbaI/Apal sites. The Vg-luc plasmid was the Vg promoter (-348 to +115bp) subcloned

in the pGL3—basic vector (Promega).

177

 

RESULTS

Cloning and characterization of AHR 78 cDNA

Three degenerate primers were designed (see Materials and Methods), based on
specific-conserved regions of the DNA binding domain (DBD) in the DHR78/TR2
family. After the first round of PCR amplification (using the fat body cDNA as a
template) by the primer pair AHR78-Fl and AHR78-R, and purification of the amplified
DNA fragment, the second round of nested PCR amplification was conducted using the
primers AHR78-F2 and AHR78-R. The final PCR amplified DNA fragment (~110 bp)
was then subcloned into a pGEM-T vector and sequenced. The sequencing data indicated
that this DNA fragment highly resembled the DHR78/TR2 family, and the deduced
amino acid sequence was more similar to that of DHR78 than that of any other nuclear
receptors in the GeneBank. Subsequently, this DNA fragment was used as a probe to
screen the mosquito fat body cDNA library that was generated from fat bodies of 6-48hr
PBM mosquitoes by the poly (T) method (Cho and Raikhel, 1992).

Several positive clones were identified by screening of the cDNA library. The
clone with the largest insert was completely sequenced on both strands. This clone of a
2.8 kb insert contained an open reading frame (ORF) of 630 amino acids. There were 3
stop codons before the first putative start codon (ATG), and polyadenylation signaling
(AATAAA) after the ORF, suggesting the ORF is complete (the GeneBank accession
number is ......... ).

The conceptual translation of AHR78 amino acid sequence revealed distinctly

high similarity with its insect counterparts of Drosophila and Tenebrio molitor (Zelhof

178

 

AHR78
TmHR78
DHR78

AHR78
TmHR78
DHR78

AHR78
TmHR78
DHR78

AHR78
TmHR78
DHR78

AHR78
TmHR78
DHR78

AHR78
TmHR78
DHR78

AHR78
TmHR78
DHR78

AHR78
TmHR78
DHR78

AHR78
TmHR78
DHR78

AHR78
TmHR78
DHR78

AHR78
TmHR78
DHR78

l
~~~MEPHNFD

~~~~~~~~~~

MDGVKVETFI

5 DBD
ELCLVCGDRA
ELCVVCEDRA
ELCLVCGDRA

101

HRNRCQYCRL
HRNRCQYCRL
HRNRCQFCRL

151

NPNSKYNPHR
NIGNNYNSNS
NGSSTYLSGK

201

ELAVANLSKR
DLGLQFLNKR
GLNFAELTQT

251

PAD.SIAALM
SMD.S .....
SSTLCLQLLA

301

H.8LPMETTD
D.ALIM....
HKGLQILQPI

351
....NGRDDD
....AA....
FECGGNRSGG
401
ETGSRLLFAT

ESGSRLLFLS
ETGSRIIFLT

451

LAFNTVMLAL
LSLSTILSAL
LSVPTIIGQF

501

TDQEFAYMRL
DEHEYAILKA
TDMEFGLLRL

IKPNLQQMKL
~~~~~~~mI
KSEENRAMPL

SGRHYGAISC
SGRHYGAISC
SGRHYGAISC

4—1
QKCLACGMRS
QKCLACGMRS
QKCLBSGMRS

SGYQQGRGKG

A..SKSPPAP
IGNSGSSDLP

MATQQQQQ

QLGPTAAAAV

SNSMEKNLIC
..ARDKQLIS
QNQLERN.GN

RDDDCLNYDY
.DEKC..YEY
SDEAINEAVF

VHWMKKNHLF
IHWTRNIPAF
IHTLRKVPVF

IQYMKNLILN
ISHLHTLIAQ
IQSTRQLADI

LCIFNPDNIL
ITLFSADQ..
ILLFNP....

EPGTQQFGSS
z...asrxsz
recesasccr

EGCKGFFKRS
EGCKGFFKRS
EGCKGFFKRS

DSVQHERKPI
DSVQHERKPI
DSVQHERKPI

...KEYHNEQ
...KIF..IR
HSVKAESAPR

sanwsssrwr
ranspsgasr
QQQQHQQSGS

NMSGLGPPPQ
TNTGAG....
LNFNAGEVPT

NSLEFIQNLE
KALDTMARVQ
stxpzcnsz

GSLELSDNCI
EGPILQEQHI
EQDLLTDVQC

sunsnsrosz
QYLTTETQIT
EQLEAHTQVK

KKXGSDVINY
DKMSATKVKQ
DKIEPLKISK

QDNVKNQHLA
PDILLRKHVE
...TLFQHRK

179

SLDDKGPSLQ
ILSDX...IR
PLPGGGVGMG

IRKQLGYQCR

IRKQLGYQCR
IRKQLGYQCR

IDKKDG....
IDKKDY....
VDRKEGIIAA

KSNTTVTPSY
KDLSTDSPGL
LQCTARQQRA

SSNPFSSLHS

YSPDIPKADP

QPPPSLEPPE
......... E
ALPTTSTMGL

HELNNSVNNN
C.LNGT....
AEDSGTEDAV

SFDIQVPNVL
SFNLQIPGPV
AFHVQPPTLV

LMRQTWPELF
LLRGCWBELF
LLRGVWPALM

LTKYILLIQE
VSDHIVKLQD
MANLTRTLHD

KIQDMVLSSF
KLQEKSFQAL
E ...... RSL

50
HLSSLNSMSI
NNSNNLCLTV
AGA8.AILSV

100
GSMNCEVTKH
GSKNCEVTKH
GAMNCEVTKH

150
...... SGGP
...... 8N..
AGSSSTSGGG

200
LNLFQGFNLA
LP..APFNPC
FNLNAEYIPM

250
FKSQDDRQPT
...EDD...V
EDDEDDSMDN

300
LPSKPTPPSS
L ........ 8
IQSSLDMRVI

350
YGIKSELN..
..DL88LT..
DAELEHMELD

400
PNYLSAHYVC
ppvnnrurrc
asrnnrarvc

‘-’*Lae

450
MIGLAQSSGQ
TLGLAQCSQT
AIALAQCQGQ

500
FVSDVQKLNL
YANTMNRLNV
FVQELQSLDV

550
RDYYKQKHSR
KT! .......
RGYVRRVQLY

551 600
AHR78 LMBS.RSEEN IRGGQDDDDD YYESSHHQQQ RQQRHNLEQR LVTILMKLPT

TmHR78 .......... VHNSFPDDTD ................... R FPRLLLRLPP
DHR78 ALSSLRRQGG IGGG ....................... ERR FNVLVARLLP
601 650

AHR78 LRALNSKKDL EDLFFSTLIG QVQIESVEM! ILQTNDG.AT SFSNL.VRNY
TmHR78 LRGLEPLV.L EELFEAGLIG QVQIDSVIPY ILRMGNGMPT PTSNRHVKSE
DHR78 LSSLDA.EAMIEELFFANLVG QMQMDALIPF IEMTSNTSGL ~~~~~~~~~~

651 668
AHR78 arsarpvesc cameo!)
TmHR78 gunmen“ ~~~~~~~~

 

Fig. 1. Alignment of the deduced amino acid sequence of the mosquito AHR78
cDNA with its insect homologs: DHR78 from Drosophila (Fisk and Thummel, 1995;
Zelhof, et al., 1995b) and TmHR78 from T enebrio molitor (Mouillet et al., 1999). The
borders of DNA-binding domain (DBD) and putative ligand-binding domain (LBD) are
marked with square braces. The complete nucleotide sequence of AHR 78 has been
submitted to the GenBank (http;//www,ngbi,nlm.nih.gov1; its accession number is ..........

 

180

 

 

.8538 8:05 2: mo
.382: Boa 05:8 .38 05 .3 e336 32.0368 033.8988 Boston 32:52 mesa 05:8 .8 confine a ma coon—=28 803 move.— ofi EEEop some 555

@8865 2a mezwoBEo: 550 one why—*2 mo mEmEoo w:_e=oqmetoo 5053 Gotta—EV. .3 85:82 .8 mowfieooeom of. .3382 :08 E memos oEEm
Mo 0382. 05 8865 2:205 323 £3832 .mEmEoe _mcouoee 828p 5805 mum: 8:588 05 goes Betc— ef. Ammo— .mzeoxov. can mess“:
NM... 8308 one .332 do 3 uoEoN ”32 ._oEE=nH EB 02“: SEQSEQ Eob mummo .2635 £23 agave 83V 80¢ mummies ”£9308 320::
mum: 05 2a :Bozm .3358.— ..335: gags—.4. «MP—.35:— 2: we 22.52: 138.3. :33 why—E 33.08:. 2: no eemtanfieo 5355 .N .wE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

am _
3...: $3 2...:
Se _ _
”DE:
:3 _
_ ”Ema.

 

181

et al., 1995a; Fisk and Thummel, 1995; Mouillet et al., 1999) (Fig. 1). AHR78 contains
the classic nuclear receptor functional domains: A/B domain, DNA binding domain
(DBD) or C domain, hinge region or D domain, ligand binding domain (LBD) or E
domain, and F domain. Comparison of AHR78 with Drosophila DHR78 and its
vertebrate counterpart, human testicular receptor 2 (TR2) showed that these proteins were
of similar length in regards to their amino acid sequences: AHR78, 630aa; DHR78 601aa;
mTR2, 590aa. Moreover, the AHR78 DBD is highly conserved in relation to the DBDs
in DHR 78 and TR2, exhibiting 95% identity (98% similarity) and 79% identity (89%
similarity), respectively. The LBD of AHR78 exhibits 33% identity (47% similarity)
with DHR78 and 30% identity (41% similarity) with mTRZ. The other domains of this

family are relatively less conserved (Fig. 2).

AHR 78 is highly expressed in the vitellogenic ovary, but not the fat body

To examine the AHR78 expression profiles, RT-PCR was utilized to analyze the
mRNA levels during mosquito vitellogenesis. The two critical vitellogenic tissues, the
fat body and ovary, were chosen for examination of AHR78 expression levels during the
first vitellogenic cycle. In the fat body, traces of AHR78 were expressed at the relatively
constant levels throughout the previtellogenic and vitellogenic stages. Vitellogenic
carboxypeptidase (VCP), one of the major YPPs (Cho et al, 1991), was used as a positive
expression control in the fat body. In the ovary, AHR78 mRNA levels were considerably
higher than those in the fat body, with dramatic increases after a blood meal peaking at
18—24h PBM, followed by a gradual decline after 30hr PBM (Fig. 3). Interestingly,

expression dynamics of AHR78 in the ovary correlates with 20E level changes during

182

 

(A) W!
PV ((1) Post Blood Meal (hr)

0-1 1-2 3-5 1 3 6 15 24 36 48

 

 

VCP

 

(B) 91313! Post Blood Meal (hr)

 

PV 612 18 24 30 36 42 48

“M--- HR73

Fig. 3. AHR 78 expression profiles in the female mosquito fat body and ovary. (A)
Temporal expression of AHR 78 in the fat body during the first vitellogenic cycle
examined by RT-PCR/Southern analyses. The expression profiles of AHR 78 (upper
panel) and vitellogenic carboxypeptidase (VCP) as a control (lower panel) were
examined (see Materials and Methods). (B) Temporal expression of AaHR 78 in the
ovary during the first vitellogenic cycle. The expression profiles of AHR 78 (upper panel)
and vitellogenin receptor (VgR) as a control (lower panel) were examined (see Materials
and Methods). Each lane contained the same mosquito equivalents of total RNA
subjected to RT-PCR/Southem analyses. PV: previtellogenic stage.

183

 

this period as well as the accumulation of yolk protein precursors by the oocytes. As a
positive control, the expression profile for vitellogenic receptor (VgR) (Sappington et al
1996) was provided for the ovary. In addition, AHR78 expression patterns and levels in
both the fat body and ovary were monitored by using different pairs of PCR primers,
which were located in either the 5’UTR or LBD. These results were the same, suggesting
that the expression profiles of AHR78 detected by the RT-PCR method were specific to

the AHR78 gene (not shown).

AHR 78 is inducible by 20E in the ovary but not in the fat body

To investigate whether the high levels of AHR78 mRNA in the ovary during the
vitellogenic stage are under the control of 20E, ovaries of previtellogenic 3-5d-old or 2h
PBM were incubated with 20E for 6h followed by RT-PCR/Southem analysis. The
AHR78 mRNA levels increased significantly after the 20E treatment in both
previtellogenic and 2h PBM ovaries (Fig. 48). For control, treatment of ovaries with
ethanol, which is solvent for 20E, did not show any effect on AHR78 mRNA levels (Fig.
48).

Next, fat bodies were tested in a parallel experiment. After previtellogenic 3-5d-
old fat bodies were cultured with 10'6 M 20E for 4b and 8h, AHR78 did not exhibit
detectable changes in mRNA levels (Fig. 4A). In the control, VCP mRNA was highly

induced by 20E in the same experiment.

184

 

 

A Fat Body Culture

 

 

 

205 - 4h 8h
AHR78
B PV Ovary 2hPBM Ovary
EtOH - + + - + +
205 - - + - - +
n- “mat?!“ we. . I: AHR78
3; _ __,____ _____3

Fig. 4. Induction of AHR78 mRNA by 20E in the ovary but not the fat body.

A). The AHR78 gene was not responsive to 20E in the fat body. Previtellogenic 3-5d-old
female fat bodies were incubated with 10'6 M 20E for 4 or 8 hr followed by RT-
PCR/Southern analysis. The VCP gene was used as a positive control. B). Induction of
AHR78 mRNA in the ovary. Previtellogenic 3-5d-old female ovaries (PV Ovary) or 2h-
PBM ovaries were treated with 20E or 20E solvent ethanol (EtOH) for 6hr, and then
analyzed by RT—PCR/Southem blot.

185

 

 

AHR 78 inhibits the DNA binding of EcR/USP to EcREs

To test the effect of AHR78 on EcR/USP DNA binding, EMSA was performed
using two well characterized EcREs, hsp27 and Eip28/29. In the absence of AHR78,
AaEcR/AaUSPb bound to hsp27 (F ig.5A, lane 4, 7) as previously reported (Wang et al.,
1998). Addition of increasing amounts of AHR78 to the binding reactions dramatically
decreased EcR/USP binding to EcRE (lane 8, 9). In the control, addition of another
nuclear orphan receptor AaE75A did not have any effect on DNA binding of mosquito
EcR/U SP (Fig. 5A, lane 10, 1 l). Unexpectedly, AHR78 alone was not able to bind to
hsp27 (Fig. 5A, lane 12).

When testing the effect of AHR78 on EcR/U SP DNA binding by using the
Eip28/29 EcRE, results similar to those in the hsp27 experiment were obtained: addition
of AHR78 dramatically decreased the DNA binding of EcR/U SP (Fig. SB, lane 8, 9).

Furthermore, I examined whether AHR78 had a similar effect on EcR/U SP
binding to Vg-EcRE, an EcRE identified in the Vg promoter (Martin, et al., 2001), and
found that AHR78 inhibited the binding of EcR/USP to the Vg-EcRE (Fig. 5C, lane 2, 3).
Moreover, I tested whether 20E had any effect on the AHR78 inhibiting action. As
shown in Fig. 5C, Lane 7, in the presence of 20E, inhibition of AHR78 on the EcR/USP
DNA-binding was similar to that seen in the absence of 20E (Fig. 5C, lane 3). As a
control, addition of increasing amounts of another nuclear orphan receptor AaFTZ-Fl
protein into the binding reactions did not show any detectable effect in the absence of
20E (Fig. 5C, lane 4, 5). Likewise, in the presence of 20E (Fig. 5C, lane 8), AaFTZ-F 1

protein did not affect the DNA binding of EcR/USP.

186

 

(A) EcR - + - + + + + + + + + -
USP - - + + + + + + + + + -
HR78 - - - - - - 0 l 2 - — +
E75 - - - — - - - .. .. 1 2 _
Ab-DmUSP - - - - + - - - .. - ' - _
Competitor - - - — - + — _ - - - ..
Hsp27-EcRE + + + + + + + + + + + +

—'9 u—e e—o H H

Lane 1 2 3 4 5 6 7 8 9 10 ll 12
Hsp27 EcRE: agagacaagma I GCACTtgtccaat

(C)

(B)
EcR + + + + + + + +
SCSI; + J i i i ' : : : USP + + + + + + + +
was .' _ L 0 I 2 HR78 . 1 2 . - - 2 +
Ab-DmUSP . . . + - - - . , FrZ-Fl - - - 1 2 - - 2
Competitor - - - - + . . - . 20E - - - - - + + +
Eip28/29 EcRE + + + + + + + + + VgEcRE + + + + + + + +
H ‘ :
Lane 1 2 3 4 5 6 7 8 9 Lane 1 2 3 4 5 6 7 8

Eip28/29 EcRE: ttaaAGGATS ITGAC CC caaTGAAC that Vg EcRE: agctgthGACCTagcgggAQQCCAaTgiQngagt

Fig. 5. Inhibition of EcR/USP binding to the natural EcREs by AHR78. In vitro-translated AaEcR and
AaUSP proteins alone or along with in vitro translated AHR78 protein were incubated with 32P-labeled
different natural EcREs. The reaction mixtures were subjected to EMSA and autoradiography. (A)
Inhibition of EcR/USP binding to the hsp27-EcRE by AHR78. In addition to 1 ul of each TNT-AaEcR and
AaUSP (lane 4 and 7), 1 ul or 2 ul of TNT-AHR78 (lane 8 and 9) or AaE75A (lane 10 and 11) were added
to the reactions. For controls, antibody against Drosophila USP supershified the EcR/USP binding (lane
5); 50 times molar excess of cold hsp27 EcRE was added as competition (lane 6). All reactions were in the
absence of 20E. The arrow indicates the specific binding. (B) Inhibition of EcR/USP binding to the
Eip28/29-ECRE by AHR78. This was a similar experiment to that in the panel A except that Eip28/29-
EcRE was used as 32P-labeled or cold oligonucleotides. (C) Inhibition of EcR/USP binding to the mosquito
Vg-ECRE by AHR78. This was a similar experiment to that in the panel A except that Vg-EcRE was used

as 32P-labeled oligonucleotides, and some experiments in the presence of 5x 1045M 20E (lane 6, 7 and 8) as
well as AaFTZ-F 1 protein as controls (lane 4 and S).

187

AHR 78 inhibits 20E-stimulated transactivation by E cR/USP

To investigate whether AHR78 affects the EcR/USP-induced transactivation of
target genes, a cell transfection assay was utilized. The reporter plasmid Eip-luc
containing the Drosophila Eip28/29-EcRE, which has been previously shown to be
activated by the mosquito EcR/U SP (Wang et al., 2000b), was transfected into
Drosophila 32 cells alone or along with the expression plasmids carrying AaEcR, AaUSP
or AHR 78 cDNAs. After transfection, cells were incubated either in the absence of
hormone or in the presence of 20E at 10'6M for 27 hr. When no EcR/ USP plasmid was
transfected, cells receiving Eip-luc alone exhibited a low level of 20E-stimulated reporter
gene activities, which was presumably due to the presence of endogenous Drosophila
EcR/U SP in $2 cells (Fig. 6A, column I). Transfection of both mosquito EcR and USP
plasmids did not affect this baseline activity in the absence of 20E (Fig. 6A, column 2);
however, addition of 20E resulted in 2.5 fold induction in reporter gene activities
(F i g.6A, column 2). When increasing amounts of AHR78 plasmid was co-transfected
with EcR/ USP plasmids, the induction was reduced continuously with virtually no
induction at 8-fold molar excess of the AHR78 plasmid (Fig. 6A, column 5).
Interestingly, in the absence of 20E, AHR78 also reduced the basal reporter activities by
up to 50% (Fig. 6A, column 5).

Similarly, using another hsp27-EcRE construct in the AMTV promoter driving a
luciferase reporter, co-transfection of AHR 78 plasmids along with EcR/USP plasmids
also reduced 20E-stimulated activation. The effect of AHR78 was less dramatic in the

hsp27-luc construct than that in the Eip28/29-luc construct, presumably due to the high

188

 

(A)

300

 

 

200 ~

 

D(-)ZOE
.(+)20E

100 4

Normalized Luc. Unit

 

 

 

r—F

 

 

A MTV-Eip28/29 Luc.

Fig 6. AHR78 inhibition of 20E-stimulated transactivation by AaEcR/AaUSP in 82
cells. (A) AHR78 inhibition of 20E-stimulated transactivation of the Eip-luc by
AaEcR/AaUSP in 82 cells. The reporter plasmids (Eip-luc,) were transfected into
Drosophila 82 cells with no expression plasmid (column 1), or with AaEcR/AaUSP
expression plasmids (lane 2) and increasing molar excesses of AHR78 expression
plasmids (lane 3, 4 and 5) (see Materials and Methods). Afier transfection, cells were
incubated either in the absence or in the presence of hormone 20E at 10'6M for 27 hr and
harvested for B-galactosidase and luciferase activity assays. Luciferase activity was
normalized with B-galactosidase activity.

189

 

 

(B)

 

500

400 -

 

300 - El (-)20E

I (+)ZOE

 

200 ~

 

 

 

100 -

 

 

 

Normalized Luc. Unit

 

 

 

 

 

Vg ECRE
350 Luc.

Fig. 6. (Cont’d). B) AHR78 inhibition of 20E-stimulated transactivation of the Vg-luc by
AaEcR/AaUSP in 82 cells. This experiment was performed similarly to the panel A
except that the Vg-luc reporter plasmid was used in this experiment. Each value
represents the mean :SD (n=3).

190

 

affinity of hsp27-EcRE with EcR/U SP and, as a result of it, high background levels of
reporter gene activities in this experiment (not shown).

In addition to examining the effect of AHR78 on the two Drosophila EcREs, I
investigated the effect of AHR78 on the Vg-EcRE, which has been demonstrated to be
responsive to EcR/USP (Martin et al., 2001). The reporter activities, driven by a 348bp
upstream region of the Vg promoter, increased by 3-fold in responsive to EcR/U SP in the
presence of 20E (Fig. 68, columns 1 and 2.). The activation by 20E was not affected by
a small amount of AHR78 (Fig. 68, column 3); whereas it was decreased significantly
upon co-transfection of increasing amounts of AHR78 (Fig. 63, columns 4 and 5). Other
promoter-reporter constructs consisting of either 2kb, lkb or 600bp of the Vg promoter
yielded less conclusive results (not shown). This is likely due to the complexity of the
Vg promoter and interfering factors in the 82 cells, making the effect of AHR78
invisible. Taken together, these results suggest that AHR78 inhibits the EcR/U SP

transactivation on various EcREs, most importantly on the native Vg EcRE.

DISCUSSION

In this study, I report the cloning and characterization of AHR78, the mosquito
homolog of HR78 in other insects and vertebrate TR2. A search of the gene data bank
using the full AHR78 amino acid sequence revealed that AHR78 exhibits the highest
similarity with the members of this HR78/TR2 subfamily, rather than other subfamilies
of the nuclear receptor superfamily (not shown). AHR78 consists of 5 classical
functional domains of the nuclear receptors (Fig. 2). While DBDS of this subfamily are

highly conserved, LBDS are less similar relative to other members of the nuclear receptor

191

 

superfamily. Although the AHR78 LBD shows 33% identity with DHR78, this is the
closest homology it shares among nuclear receptors. Similarly, T enebrio molitor
TmHR78 shows only 31% identity with DHR78 LBD (Mouillet et al., 1999). These
results suggest that members of the HR78/TR2 subfamily of nuclear receptors have
evolved rapidly, particularly in this domain.

Analysis of expression profiles of HR 785 during insect development reveals the
similar temporal correlations with ecdysteroid levels. In the mosquito fat body, AHR78
exhibited only negligible expression levels throughout the previtellogenic and
vitellogenic stages. In contrast, in the ovary, AHR78 levels rose dramatically after a
blood meal with peak expression occurring at 18-24 hr PBM correlating positively with
the ecdysteroid peak in the mosquito (Fig. 3). During Drosophila metamorphosis,
DHR78 is expressed throughout development with peaks of expression in third instar
larvae and prepupae that is correlated with known ecdysteroid pulses (Fisk and Thummel,
1995; Zelhof et al., 1995b). Similar correlation between expression levels and
ecdysteroid variations has been reported for TmHR78 (Mouillet et al., 1999). The
correlations between expression levels of HR78 and ecdysteroid levels further support
that AHR78 is a member of the HR78 family, despite low conservation levels of the LBD
amino acid sequences.

In Drosophila, it has been demonstrated by DHR78 mutants that DHR78
functions at the top of the ecdysteroid regulatory hierarchies. DHR78 mutations block
ecdysteroid-regulated gene expression in mid-third instar larvae (Fisk and Thummel,
1998). In addition, DHR78 protein binds as a homodimer to a subset of EcR/U SP

binding sites in vitro; therefore, it has been proposed that DHR78 can function to regulate

192

 

ecdysone responses negatively (Thummel, 1995). However, in mosquitoes direct binding
of AHR78 to EcREs was not observed. This was probably due to sensitivity of the
experiment, or the evolutionary divergence among insect HR78, particularly in LBDS.
One of the LBD functions in a classical nuclear receptor is to form a receptor homodimer.
The less conserved LBDS in the HR78 subfamily may explain why AHR78 did not bind
to EcREs directly. In mosquitoes, the direct interaction of AHR78 and EcR/USP was
observed in vitro in the GST-pulldown assays (Li, C. and Raikhel, A, unpublished). This
result suggested that AHR78 negatively regulated the ecdysone responses in a different
way. The direct interaction of AHR78 and EcR/USP may prevent the AaEcR/AaUSP
binding to EcREs. Although it has not been reported in other insects that HR78 proteins
can interact directly with EcR or USP, DHR78 protein is bound to many ecdysteroid-
regulated puff loci, suggesting that DHR78 may interact directly with EcR or USP (Fisk
and Thummel, 1998). In addition, consistent with findings in Dros0phila DHR78,
AHR78 inhibits the 20E-induced activation of target genes in cell transfection assays
(Fisk and Thummel, 1995; Zelhof et al., 1995b; this paper).

One unique characteristic of AHR78 is that it is highly expressed in the ovary
during the vitellogenic stage (Fig. 3B). YPP genes are silent in the ovary despite the
presence of both EcR and USP (Wang et al., 2000a). Taken together with the role of
AHR78 in inhibition of the Vg promoter, it suggests that AHR78 may be responsible for
the tissue-specific expression of the YPP genes, such as the inactivity of the Vg gene in
the ovary. Further elucidation of the functions of AHR78 may provide new insights in
our understanding the nature of ecdysone gene regulatory hierarchies in general as well

as ecdysteroid regulation of mosquito vitellogenesis in particular.

193

 

ACKNOWLEDGMENTS

This work was supported by a grant (A136959) from the National Institutes of
Health. I thank Dr. Kook-Ho Cho for performing the ovary culture experiment, Mr. Alan
Hays for his excellent assistance in maintaining the mosquito culture, Dr. M. Ken for
contributive help at the initiation of this project, Dr. F. Kafatos for the generous gift of
anti-DmUSP antibodies, Dr. M. McKeown for providing reporter constructs, and Dr. D.

Martin for providing Vg-luc constructs.

194

 

Chapter 7

Summary and Future Research Perspectives

195

Summary and Future Research Perspectives

Ecdysone gene regulation is one of best systems for studying eukaryotic
transcription regulation. In Drosophila, this system has been studied for several decades,
and recent discoveries have made significant advances in the molecular mechanism of the
ecdysone gene regulatory hierarchies during metamorphosis. In contrast, in other insects,
though ecdysone is one of the key hormones and regulates many events including
development, metamorphosis and reproduction, the molecular basis of the ecdysone gene
regulation hierarchy was poorly understood. This is particularly so in ecdysone gene
regulation of insect reproduction. This insufficiency has made the general fundamental
molecular mechanism of ecdysone gene regulation in insects unclear.

In this dissertation, I have reported the extensive investigation of the molecular
basis of ecdysone gene regulation in female mosquito reproduction. Characterizations of
many elements of the ecdysone gene regulation cascade have demonstrated that the
general molecular mechanism underlying ecdysone gene regulation appears to be
conserved between mosquito vitellogenesis and fly metamorphosis. These suggest that
despite the diversities of physiological functions of ecdysone in insects, the fundamental
molecular features are considerably similar. In the future research, I propose to study the

following aspects:

Regulatory regions of the Aa USP gene(s).
As described in Chapter 2, mosquito USP displays interesting characteristics: 1).
The presence of two isoforms with distinct N-termini, 2). Differential expression and

regulation by 20E of the two mosquito USP isoforms during vitellogenesis. 3). Direct

196

 

 

regulation of AaUSP-B by 20E and indirect regulation of AaUSP-A by 20E. The
AaUSP-A expression profile negatively correlates with ecdysteroid titers while the
AaUSP-B isoform expression shows positive correlation with ecdysteroid titers during
mosquito vitellogenesis. In vitro organ culture experiments have suggested that the
differential expression of mosquito USP isoforms is controlled by 20E. AaUSP-A
expression is upregulated by the falling 20E titer while AaUSP-B is upregulated by the
rising 20E titer. Moreover, the upregulation of AaUSP-B is enhanced in the presence of
cycloheximide (Chx), a protein synthesis inhibitor; whereas the upregulation of AaUSP-
A is abolished in the presence of Chx. These interesting differential regulation features
lead us to ask the questions: what is the molecular basis of remarkable differential
regulation of the two USP isoforms? How does the gene structure of USP facilitate this
regulation? Does it result from alternative splicing of a single gene or from two
separated genes? In insects, most known nuclear receptor isoforms are generated by
alternative splicing and/or differential promoter utilization, though in vertebrates many
nuclear receptor isoforms result from different genes. Examination of the mosquito USP
gene will definitely provide more information about gene structure and regulation in
general as well as those specifically involved in vitellogenesis.

The clear answers to the questions above will first rely on genomic cloning of the
mosquito USP gene. Using the two Aa USP cDNA clones to screen the genomic library
can be the method to obtain the genomic structure of AaUSP. Genomic Southern blot
can also be utilized to perform hybridization with probes generated by different S’UTR or
N-termini of AaUSP cDNAs. In this way, it will be known whether the two AaUSP

isoforms are derived from different genes or from different promoters of a single gene.

197

Moreover, further identification of the transcriptional units of Aa USP is necessary, which
can be done by primer extension or RNase protection assays. The next step will be to
clone the promoter regions of the two AaUSP isoforms and to functionally study the
differential regulation of AaUSP by 20E.

Once the two promoters are identified, I propose to construct a promoter-reporter
plasmid and co-transfect them with EcR/USP into the Drosophila 82 cell line. We will
then be able to examine how the two USP promoters respond to the presence or absence
of 20E. Furthermore, a series of deletions of the AaUSP promoters will be constructed to
map the differential responsive regions on the different promoters. EMSAs can be
performed to test whether there is a direct binding site of EcR/U SP on the AaUSP-B
promoter. Through this procedure, we will be able to obtain the answer concerning the

ecdysone responsive regions of two AaUSP isoforms.

The early gene (E 75) regulation in the ecdysone genetic regulatory hierarchy.

As described in Chapter 3, the mosquito homolog of Drosophila E 75 has been
cloned, which is a crucial early gene involved in Drosophila metamorphosis. E75
belongs to the nuclear receptor superfamily and has a counterpart in vertebrates, Rev-erb.
Previous data has demonstrated that E 75 is directly induced and responsive earlier than
Vg in the mosquito fat body by the steroid hormone 20E, suggesting that E 75 is involved
in the activation and maintenance of the mosquito vitellogenic response to ecdysone.
Next, I propose to use the biochemical and genetic approaches to investigate how E 75 is

regulated and how it controls its target genes during mosquito vitellogenesis.

198

 

 

Three cDNA isoforms and the genomic DNA of mosquito E 75 have been cloned.
I propose to extend these studies by identifying the transcription start site of each
transcriptional unit, and further to identify the promoter of mosquito E 75.

As shown in Chapter 3, E 75 gene expression is induced by 20E. After identifying
the E 75 promoter, we can scan the promoter sequence for putative EcR/U SP binding sites
using the consensus binding site of EcR/U SP as a reference. These putative binding sites
will be tested using in vitro DNA binding assays to identify the EcR/U SP binding sites on
the E 75 promoter. Furthermore, I propose to test whether these binding sites are
functional in vivo by cell transfection assays, using reporter genes driven by the different
E 75 promoter deletions, co-transfected with EcR/USP expression plasmids. Moreover,
we can study the molecular basis of the auto-inhibition of the E75 gene. It has been
demonstrated that the induction by 20E of E 75 is significantly increased in the presence
of cycloheximide, a protein synthesis inhibitor, suggesting that the E75 protein likely
represses its own gene promoter. I propose to utilize the same strategy used above to
identify potential binding sites of E75 on its own gene promoter, and test the action of the
E75 protein on its own promoter by cell transfection assays. Finally, we can characterize
the interaction between E75 and EcR/U SP on the E 75 promoter, and investigate how
activation by EcR/U SP and repression by E75 on the E 75 gene promoter lead to sustained
expression of E 75.

The induction of the E 75 gene positively correlates with that of vitellogenin gene
both in vitro and in vivo. The putative E75 binding sites have been found on the Vg
promoter (Martin and Raikhel, unpublished). Furthermore, it has been shown in cell

transfection assays using the Vg promoter carrying a reporter gene that E75 up-regulates

199

 

the reporter expression (Chapter 3). Next, I propose to identify the functional binding
sites of E75 on the Vg promoter, and then compare the differences between each E75
responsive element on the E 75 and Vg promoters. Additionally, we can test whether the
different responsive elements result in different E75 activities. As shown in Chapter 3,
EcR/U SP and E75 synergistically activated the Vg promoter in Drosophila S2 cells.
Next, I propose to characterize the potential different molecular mechanisms of the action
of both EcR/U SP and E75 on different promoters (Vg and E 75). Furthermore, we can

use Drosophila to further characterize mosquito E 75, by crossing the Drosophila E 75

 

mutant or the ectopically inducible E 75 Drosophila with the mosquito-Vg-promoter-
transgenic Drosophila, both of which are available. In the future, if possible, we can
make a transgenic mosquito to characterize the action of E75 and EcR/U SP on the Vg
promoter, so that we can obtain a much better understanding of the molecular mechanism

of indirect ecdysone control of vitellogenic genes.

Molecular basis of competence and its regulation for vitellogenic responses to
ecdysone.

Results from AaFTZ-Fl studies (Chapter 5) indicate that AaFTZ-Fl serves as a
competence factor for controlling vitellogenic responses to ecdysone during the
previtellogenic period. This is based on two lines of evidence: one is that in in vitro fat
body culture, following the high level of AaFTZ-F lexpression, induction of E75A and
VCP by 20E increases significantly; the second lines of evidence is that binding of
nuclear extracts from newly enclosed, incompetent female fat bodies to the F TZ-Fl

responsive element is undetectable, while binding is clearly detected in nuclear extracts

200

 

from 3~5-day-old competent fat bodies. These observations lead us to believe that
competence of mosquito vitellogenic responses to ecdysone correlates with the
appearance of functional AaFTZ-F 1 protein. Since AaFTZ-Fl mRNA is abundant
throughout the entire previtellogenic stage, regulation of AaFTZ-Fl binding activities
could be by either translational and/or post-translational control. Preliminary Western
blot analysis with Drosophila F TZ-Fl antibodies indicated that AaFTZ-F 1 protein was
present during the entire previtellogenic stage. This result needs to be confirmed by
Western blot using antibodies specific to AaFTZ-Fl. If the possibility of translational
control is ruled out, we can focus on examining post-translational control. A common
post-translation modification is phosphorylation. We will first determine if this
regulating system is being mediated by phosphorylation. Interestingly, steroidogenic
factor 1 (SF -1), a vertebrate homolog of the insect FTZ-F l , has been demonstrated to be
post-translationally regulated. Activation of SF -1 mediated transcription via interaction
with nuclear receptor coactivators depends on the phosphorylation of a single serine

residue in the AF -1 domain of SF -1 (Hammer et al., 1999). Moreover, the silkworm

BmFTZ-F 1 activation function is mediated by the coactivator MBFI, which stabilizes the

BmFTZ-F 1 -DNA complex (Li et al., 1994). To test whether phosphorylation of AaFTZ-

F1 contributes to vitellogenic competence, we can introduce protein kinases or their

inhibitors into EMSA experiments; or alternatively, in organ culture experiments, we can

enhance/ inhibit the phosphorylation in the fat body by kinases or their inhibitors.

Furthermore, another interesting question is whether the acquisition of

competence via AaFTZ-Fl is under hormonal control. Early studies have established that

the acquisition of competence by the mosquito fat body for the vitellogenic response to

201

 

Pm

 

ecdysone is controlled by juvenile hormone (JH) (F lannagan and Hagedom, 1977).
Treatment of previtellogenic incompetent fat bodies with JH renders this tissue capable of
synthesizing vitellogenin protein in response to 20E (F lannagan and Hagedom, 1977; Ma
etal., 1988). Our preliminary experiments have revealed that after incubation of newly
eclosed incompetent fat bodies with 10'6 M JHIII for 24 hr followed by addition of 10'6 M
20E, the induction of VCP mRNA was significantly increased, as shown by RT-
PCR/Southem blot (Li C, and Raikhel, A., unpublished). Therefore, one interesting
question is whether JH-controlled competence and AaFTZ-F 1 -involved competence are
in the same pathway. To answer this question, we can perform EMSAs to examine the
DNA binding activity of nuclear extracts from newly eclosed, 24hr JH III-treated female
fat bodies. This experiment will provide direct evidence of whether the active DNA-
binding AaFTZ-Fl protein results from JH action in the fat body.

In vertebrates, recent discoveries have established that two ligands, oxysterols and
25-Hydroxycholesterol, are capable of stimulating SF -1 dependant transcription. This
suggests that SF -1 may be a nuclear hormone receptor rather than an orphan receptor
(Bertherat, 1998). Therefore, a more interesting question for AaFTZ-F 1 will be whether
it is a J H receptor. A simple experiment can be performed by cell transfection assay,
using a promoter-reporter construct containing FTZ-F l responsive elements. Co-
transfection of both this reporter construct and AaFTZ-F 1 expression plasmids into
Drosophila 82 cells (and/or CV-l cells) in the presence or absence of JHIII or its analog

methoprene should show if FTZ-Fl is responsive to JH or its analogs.

202

 

AHR 78 function in vitellogenic ovaries

In Chapter 6, AHR78 studies have demonstrated that AHR78 is expressed at very
low levels in the fat body and the previtellogenic ovary, but expressed at considerably
high levels in the vitellogenic ovary. Functional analysis has established that AHR78
acts as a repressor, inhibiting DNA binding and transactivation of EcR/U SP. These
results suggest that AHR78 may be involved in inhibiting activation of YPP genes in the
ovary. To further support this hypothesis, first, we should specify the localization of
AHR78 expression within the ovary due to the fact that the mosquito ovary contains
several different tissues including oocytes, nurse cells and follicle cells. The main
challenge of this experiment lies in the relatively low levels of expression of AHR78
mRNA. An AHR78 anti-sense mRN A probe can be used in in situ hybridization
experiments. Antibodies against the AHR78 can also be raised to examine the AHR78
protein levels in the ovary. Second, further EMSA analysis of AHR78 inhibition of
EcR/USP binding to EcREs will be conducted by nuclear extracts from both fat bodies
and ovaries. Previous studies in our lab have successfully detected specific binding to
EcREs of EcR/U SP from nuclear extracts in both fat bodies and ovaries (Miura et al.,
1999). In this proposed study, in vitro produced AHR78 protein will be added into the
binding reactions to test whether it will abolish the native EcR/USP proteins binding to
EcREs. Furthermore, preliminary GST pull-down experiments indicate that AHR78
interacts with EcR and USP. Next, co-immunoprecipitation experiments will be designed
to examine whether the interactions of AHR78 with EcR/USP occur in vivo. These
studies will provide more valuable information regarding the potential inhibitory role of

AHR78 on EcR/USP action. '

203

 

 

 

REFERENCES

204

REFERENCES

Allan, G.F., Leng, X., Tsai, S.Y., Weigel, N.L., Edwards, D.P., Tsai, M.J., and O'Malley,
B.W. (1992). Hormone and antihorrnone induce distinct conformational changes
which are central to steroid receptor activation. J. Biol. Chem. 267, 19513-19520.

Antoniewski, C., Laval, M., Dahan, A., and Lepesant, J .A. (1994). The ecdysone
response enhancer of the Fbpl gene of Drosophila melanogaster is a direct target
for the EcR/USP nuclear receptor. Mol. Cell. Biol. 14, 4465-4474.

Antoniewski, C., O'Grady, M.S., Edmondson, R.G., Lassieur, S.M., and Benes, H.
(1995). Characterization of an EcR/U SP heterodimer target site that mediates

ecdysone responsiveness of the Drosophila Lsp-2 gene. Mol. Gen. Genet. 249,
545-556.

Antoniewski, C., Mugat, B., Delbac, F. and Lepesant, J.-A. (1996). Direct repeats bind
the EcR/USP receptor and mediate ecdysteroid responses in Drosophila
melanogaster. Mol. Cell. Biol. 16, 2977-2986.

Asahina, M., Jindra, M., and Riddiford, L. M. (1997). Developmental expression of
Ultraspiracle proteins in the epidermis of the tobacco homworm, Manduca sexta,

during larval life and the onset of metamorphosis. Dev. Genes Evol. 207, 381-
388.

Ashbumer, M. (1972). Puffing patterns in melanogaster and related species. Results
Probl. Cell. Difler. 4, 101-151. Review.

Ashbumer, M., Chihara, C., Meltzer, P., and Richards, G. (1974). Temporal control of
puffing activity in polytene chromosomes. Cold Spring Harb. Symp. Quant. Biol.
38, 655-662.

Bai, 1., Uehara, Y., and Montell, DJ. (2000). Regulation of invasive cell behavior by
taiman, a Drosophila protein related to AIB l , a steroid receptor coactivator
amplified in breast cancer. Cell 103, 1047-1058.

Beato, M., Herrlich, P., and Schutz, G. (1995). Steroid hormone receptors: many actors in
search of a plot. Cell. 83,851-7. Review.

Beckemeyer, E. and Lea, A. (1980). Induction of follicle separation in the mosquito by
physiological amounts of ecdysterone. Science 209, 819-821.

Beier, J. C. (1998). Malaria parasite development in mosquitoes. Annu. Rev. Entomol. 43,
519-543.

Bender, M., Imam, F. B., Talbot, W.S., Ganetzky, B., and Hogness, D. S. (1997).

Drosophila ecdysone receptor mutations reveal functional differences among
receptor isoforms. Cell 91, 777-788.

205

 

Bertherat, J. (1998). The nuclear receptor SF-l (steroidogenic factor-1) is no longer an
orphan. Eur. J. Endocrinol. 138, 32-33. Review.

Bose, S. G., and Raikhel, A. S. (1988). Mosquito vitellogenin subunits originate from a
common precursor. Biochem. Biophys. Res. Commun. 155, 436-42.

Bownes, M. (1986). Expression of genes coding for vitellogenin (yolk protein). Ann. Rev.
Entomol. 31, 507-531.

Bresnick, E.H., Dalman, F.C., Sanchez, E.R., and Pratt, W.B. (1989). Evidence that the
90-kDa heat shock protein is necessary for the steroid binding conformation of the
L cell glucocorticoid receptor. J. Biol. Chem. 264, 4992-4997.

Broadus, J ., McCabe, J. R., Endrizzi, B., Thummel, C. S, and Woodard, C. T. (1999). The
Drosophila beta F TZ-Fl orphan nuclear receptor provides competence for stage-
specific responses to the steroid hormone ecdysone. Mol. Cell 3, 143-149.

Brown, M.R., Graf, R., Swiderek, K.M., Fendley, D., Stracker, T.H., Champagne, DE,
and Lea, AD. (1998). Identification of a steroidogenic neurohormone in female
mosquitoes. J. Biol. Chem. 273, 3967-3971.

Burtis, K. C., Thummel, C. S., Jones, C. W., Karim, F. D., and Hogness, D. S. (1990).
The Drosophila 74EF puff contains E74, a complex ecdysone-inducible gene that
encodes two ets-related proteins. Cell 61, 85-99.

Buszczak, M., Freeman, M., Carlson, J ., Cooley, L., and Segraves, W. A. (1999).
Ecdysone response genes govern egg chamber development during mid-oogenesis
in Drosophila. Development 26, 4581-4589.

Butler, D., Maurice, J ., and O'Brien, C. (1997). Vaccines: a roller-coaster of hopes.
Nature 386, 537-538.

Carney, G.E., Wade, A.A., Sapra, R., Goldstein, ES, and Bender, M. (1997). DHR3, an
ecdysone-inducible early-late gene encoding a Drosophila nuclear receptor, is
required for embryogenesis. Proc. Natl. Acad. Sci. USA 94, 12024-12029.

Chang C. and Kokontis J. (1988). Identification of a new member of the steroid receptor
super-family by cloning and sequence analysis. Biochem. Biophys. Res. Commun.
155, 971-977.

Chen, J .D. and Evans, RM. (1995). A transcriptional co-repressor that interacts with
nuclear hormone receptors. Nature 377, 454-457.

Cherbas, L., Fristrom, J .W., and O'Connor, J .D. (1984). The action of ecdysone in
imaginal discs and Kc cells of Drosophila melanogaster. In: Biosynthesis,

206

 

 

metabolism and mode of action of invertebrate hormones (Hoffman, J and
Porchet, M., eds.) pp. 305-322. Springer-Verlag, Berlin.

Cherbas, L., Lee, K., and Cherbas, P. (1991). Identification of ecdysone response
elements by analysis of the Eip28/29 gene. Genes Dev. 5, 120-131.

Cho, W. L, Tsao, S. M., Hays, A. R., Walter, R., Chen, J. S., Snigirevskaya, E. S., and
Raikhel, A. S. (1999). Mosquito cathepsin B-like protease involved in embryonic
degradation of vitellin is produced as a latent extraovarian precursor. J. Biol.

Chem. 274, 13311-13321.

Cho, W. -L., Kapitskaya, M. Z., and Raikhel, A. S. (1995). Mosquito ecdysteroid
receptor: analysis of the cDNA and expression during vitellogenesis. Insect
Biochem. Mol. Biol. 25, 19-27.

Cho, W.-L., and Raikhel, A. S. (1992). Cloning of cDNA for mosquito lysosomal
aspartic protease. J. Biol. Chem. 267, 21823-21829.

Cho, W.-L., Deitsch, K. W., and Raikhel, A. S. (1991). Extraovarian protein accumulated
in mosquito oocytes is a serine carboxypeptidase activated in embryos. Proc.
Natl. Acad. Sci. USA 88, 10821-10824.

Christianson, A.M. K., King, D.L., Hatzivassiliou, E., Casas, J .E., Hallenbeck, P.L.,
Nikodem, V.M., Mitsialis, S.A., and Kafatos, F .C. (1992). DNA binding and

heteromerization of the Drosophila transcription factor chorion factor
l/ultraspiracle. Proc. Natl. Acad. Sci. USA. 89, 11503-11507.

Collins, F. H., and Paskewitz, S. M. (1995). Malaria: current and future prospects for
control. Annu. Rev. Entomol. 40, 195-219.

D'Avino, P. P., Crispi, S., Cherbas, L., Cherbas, P., and Furia, M. (1995). The moulting
hormone ecdysone is able to recognize target elements composed of direct
repeats. Mol. Cell. Endocrinol. 113, 1-9.

Davis, R.W., Botstein, D., and Roth, J .R. (1980). Advanced Bacterial Genetics. Cold
Spring Harbor Laboratory, Cold Spring Harbor.

DeFranco, D.B., Ramakrishnan, C. and Tang, Y. (1998). Molecular chaperones and
subcellular trafficking of steroid receptors. J. Steroid Biochem. Mol. Biol. 65, 51-
58. Review.

Deitsch, K. W., Chen, J .-S., and Raikhel, A. S. (l995a). Indirect control of yolk protein

genes by 20-hydrocyecdysone in the fat body of the mosquito, Aedes aegypti.
Insect Biochem. Mol. Biol. 25, 449-454.

207

Deitsch, K. W., Dittmer, N., Kapitskaya, M. Z., Chen, J .-S., Cho, W.-L., and Raikhel, A.
S. (1995b). Regulation of gene expression by 20-hydroxyecdysone in the fat body
of Aedes aegxpti (Diptera: Culicidae). Eur. J. Entomol. 92, 237-244.

Deitsch, K.W. and Raikhel, AS. (1993). Cloning and analysis of the locus for mosquito
vitellogenic carboxypeptidase. Insect Bioch. Molec. Biol. 2, 205-213.

Dhadialla, T. S., and Raikhel, A. S. (1994). Endocrinology of mosquito vitellogenesis. In
“Perspectives in Comparative Endocrinology” (K. G. Davey, R. E. Peter, and S.
S. Tobe, Eds.), pp. 275-281. National Research Council, Canada.

DiBello, P. R., Withers, D.A., Bayer, C. A., Fristrom, J. W., Guild, G. M. (1991). The
Drosophila Broad-Complex encodes a family of related, zinc finger-containing
proteins. Genetics 129, 385-397.

Dittmann, F., Kogan, P. H., and Hagedom, H. H., (1989). Ploidy levels and DNA
synthesis in fat-body cells of the adult mosquito, A edes Aegypti — the role of
Juvenile-Hormone. Arch. Insect Biochem. 12, 133-143.

Dobson, A.D., Conneely, O.M., Beattie, W., Maxwell, B.L., Mak, P., Tsai, M.J.,
Schrader, W.T. and O'Malley, B.W. (1989). Mutational analysis of the chicken
progesterone receptor. J. Biol. Chem. 264, 4207-4211.

Dressel, U., Thormeyer, D., Altincicek, B., Paululat, A., Eggert, M., Schneider, 8.,
Tenbaum, S.P., Renkawitz, R. and Baniahmad, A. (1999). Alien, a highly
conserved protein with characteristics of a corepressor for members of the nuclear
hormone receptor superfamily. Mol. Cell. Biol. 19, 3383-3394.

Elliston, J .F ., Fawell, S.E., Klein-Hitpass, L., Tsai, S.Y., Tsai, M.J., Parker, MG. and
O'Malley, B.W. (1990). Mechanism of estrogen receptor-dependent transcription
in a cell-free system. Mol. Cell. Biol. 10, 6607-6612.

F eigl, G., Gram, M., and Pongs, O. (1989). A member of the steroid hormone receptor
gene family is expressed in the 20-OH-ecdysone inducible puff 75B in
Drosophila melanogaster. Nucleic Acids Res. 17, 7167-7178.

Fisk, G. J. and Thummel, C. S. (1995). Isolation, regulation, and DNA-binding properties
of three Drosophila nuclear hormone receptor superfamily members. Proc. Natl.
Acad. Sci. USA 92, 10604-8.

Fisk, G. J. and Thummel, C. S. (1998). The DHR78 nuclear receptor is required for
ecdysteroid signaling during the onset of Drosophila metamorphosis. Cell 93,
543-55.

Flanagan, T. R., and Hagedom, H. H. (1977). Vitellogenin synthesis in the mosquito: the
role of juvenile hormone in the development of responsiveness to ecdysone.

208

 

 

 

Physiol. Entomol. 2, 173-178.

Fletcher, J .C. and Thummel, CS. (1995). The Drosophila E 74 gene is required for proper
stage- and tissue-specific transcription of ecdysone-regulated genes at the onset of
metamorphosis. Development 121, 1411-1421.

Freedman LP. (1999). Increasing the complexity of coactivation in nuclear receptor
signaling. Cell 97, 5-8. Review.

Fujiwara, H., Jindra, M., Newitt, R., Palli, S. R., Hiruma, K., and Riddiford, L. M.
(1995). Cloning of an ecdysone receptor homolog from Manduca sexta and the

developmental profile of its mRNA in wings. Insect Biochem. Mol. Biol. 25, 845- 4
856.

 

Garen, A.L., Kauvar, L., and Lepesant, J .-A. (1977). Roles of ecdysone in Drosophila
development. Proc. Natl. Acad. Sci. USA 74, 5099-5103.

Giguere, V., McBroom, L.D., and Flock, G. (1995). Determinants of target gene

specificity for RORal: monomeric DNA binding by an orphan nuclear receptor.
Mol. Cell. Biol. 15, 2517-2526.

Giguere, V., Tini, M., Flock, G., Ong, E., Evans, R.M., Otulakowski, G. (1994). Isoforrn-
specific amino-terminal domains dictate DNA-binding properties of ROR alpha, a
novel family of orphan hormone nuclear receptors. Genes Dev. 8, 538-553.

Glass, GK. and Rosenfeld, MG. (2000). The coregulator exchange in transcriptional
functions of nuclear receptors. Genes Dev. 14, 121-141. Review. ~

Guay, P. and Guild, GM. (1991). The ecdysone-induced puffing cascade in Drosophila
salivary glands: a Broad-Complex early gene regulates interrnolt and late gene
transcription. Genetics 129, 169-175.

 

Gwardz, R.W., and Spielman, A. (1973). Corpus allatum control of ovarian development
in Aedes aegypti. J. Insect Physiol. 19, 1441-1448.

Hagedom, H. H. (1989). Physiological roles of hemolymph ecdysteroids in adult insect.
In “ Ecdysone, From Chemistry to Mode of Action”. (J. Koolman, Eds.), pp.279-
289. Thieme Medical Publishers, Inc., New York.

Hagedom, H. H., and Fallon, A. M. (1973). Ovarian control of vitellogenin synthesis by
the fat body in Aedes aegypti. Nature 244, 103-105.

Hagedom, H. H., Fallon, A. M., and Laufer, H. (1973). Vitellogenin synthesis by the fat

body of the mosquito Aedes aegypti: evidence of transcriptional control. Dev.
Biol. 31, 285-294.

209

Hagedom, H., O'Connor, J ., Fuchs, M., Sage, B., Schlaeger, D., and Bohm, M. (1975).
The ovary as a source of a-ecdysone in an adult insect. Proc. Natl. Acad. Sci.
USA 72, 3255-3259.

Hagedom, H.H. (1983). The role of ecdysteroids in the adult insect. In: Endocrinology of
Insects (Downer, R. and Laufer, H., eds.) pp. 271-304. Liss, New York.

Hagedom, H.H. (1985). The role of ecdysteroids in reproduction. In: Comprehensive
Insect Physiology, Biochemistry and Pharmacology, Vol. 8 (Kerkut, G. and
Gilbert, L., eds.) pp. 205-261. Pergamon, Oxford.

Hall, B. L., and Thummel, C. S. (1998). The RXR homolog ultraspiracle is an essential
component of the Drosophila ecdysone receptor. Development 125, 4709-4717.

Hammer, G.D., Krylova, 1., Zhang, Y., Darimont, B.D., Simpson, K., Weigel, N.L., and
Ingraham, HA. (1999). Phosphorylation of the nuclear receptor SF-l modulates

cofactor recruitment: integration of hormone signaling in reproduction and stress.
Mol. Cell. 3, 521-526.

Hard, T., Kellenbach, E., Boelens, R., Maler, B.A., Dahlman, K., Freedman, L.P.,
Carlstedt-Duke, J ., Yamamoto, K.R., Gustafsson, J .A. and Kaptein, R. (1990).

Solution structure of the glucocorticoid receptor DNA-binding domain.
Science 249, 157-160.

Hays, A. R. and Raikhel, A. S. (1990). A novel protein produced by the vitellogenic fat
body and accumulated in mosquito oocytes. Roux ’s Arch. Dev. Biol. 199, 114-
121.

Henrich, V. C., Sliter, T. J., Lubahn, D. B., Maclntyre, A., and Gilbert, L. I. (1990). A
steroid/thyroid hormone receptor superfamily member in Drosophila

melanogaster that shares extensive sequence similarity with a mammalian
homologue. Nucleic Acids Res. 18, 4143-8.

Henrich, V.C., Szekely, A.A., Kim, S.J., Brown, N.E., Antoniewski, C., Hayden, M.A.,
Lepesant, J .A., and Gilbert, L1. (1994). Expression and function of the

ultraspiracle (usp) gene during development of Drosophila melanogaster. Dev.
Biol. 165, 38-52.

Hill, R.J., Segraves, W.A., Choi, D., Underwood, P.A., and Macavoy, E. (1993). The
reaction with polytene chromosomes of antibodies raised against Drosophila
E75A protein. Insect Biochem. Molec. Biol. 23, 99-104.

Hiruma, K., Shinoda, T., Malone, F., and Riddiford, L. M. (1999). Juvenile hormone
modulates 20-hydroxyecdysone—inducible ecdysone receptor and ultraspiracle
gene expression in the tobacco hornworrn, Manduca sexta. Dev. Genes Evol. 209,
18-30.

210

 

‘

 

Honda, 8., Morohashi, K., Nomura, M., Takeya, H., Kitajima, M., and Omura, T. (1993).
Ad4BP regulating steroidogenic P-450 gene is a member of steroid hormone
receptor superfamily. J. Biol. Chem. 268, 7494-502.

Horlein, A.J., Naar, A.M., Heinzel, T., Torchia, J., Gloss, B., Kurokawa, R., Ryan, A.,
Kamei, Y., Soderstrom, M., Glass, GK. and Rosenfeld, MG. (1995). Ligand-
independent repression by the thyroid hormone receptor mediated by a nuclear
receptor co-repressor. Nature 377, 397-404.

Homer, M.A., Chen, T., Thummel, CS. (1995). Ecdysteroid regulation and DNA binding
properties of Drosophila nuclear receptor superfamily members. Dev. Biol. 168,
490-502.

Huet, F ., Ruiz, C., and Richards, G. (1993). Puffs and PCR: the in vivo dynamics of early
gene expression during ecdysone responses in Drosophila. Development 118,
613-627.

Huet, F ., Ruiz, C., and Richards, G. (1995). Sequential gene activation by ecdysone in
Drosophila melanogaster: the hierarchical equivalence of early and early late
genes. Development 121, 1 195-1204.

Jindra, M., Huang, J. Y., Malone, F ., Asahina, M., and Riddiford, L. M. (1997).
Identification and mRNA developmental profiles of two ultraspiracle isoforms in
the epidermis and wings of Manduca sexta. Insect Mol. Biol. 6, 41-53.

J indra, M., Malone, F ., Hiruma, K., and Riddiford, L. M. (1996). Developmental profiles
and ecdysteroid regulation of the mRNAs for two ecdysone receptor isoforms in
the epidermis and wings of the tobacco homworrn, Manduca sexta. Dev. Biol.

180, 258-272.

Jindra, M. and Riddiford, L.M. (1996). Expression of ecdysteroid regulated transcripts in
the silk gland of the wax moth, Galleria mellonella. Dev. Gen. Evol. 206, 305-
314.

J indra, M., Sehnal, F ., and Riddiford, L.M. (1994a). Isolation, characterization and
developmental expression of the ecdysone-induced E75 gene of the wax moth
Galleria mellonella. Eur. J. Biochem. 221, 665-675.

Jindra, M., Sehnal, F ., and Riddifird, L.M. (1994b). Isolation and developmental
expression of the ecdysteroid-induced GHR3 gene of the wax moth Galleria
mellonella. Insect Biochem. Mol. Biol. 24, 763-773.

Jones, G. and Sharp, RA. (1997). Ultraspiracle: an invertebrate nuclear receptor for
juvenile hormones. Proc. Natl. Acad. Sci. USA. 94, 13499-13503.

211

 

 

Kamimura, M., Tomita, S., and Fujiwara, H. (1996). Molecular cloning of an ecdysone
receptor (BI isoform) homologue from the silkworm, Bombyx mori, and its
mRNA expression during wing disc development. Comp. Biochem. Physiol. B
Biochem. Mol. Biol., 113, 341-347.

Kamimura, M., Tomita, S., Kiuchi, M., and Fujiwara, H. (1997). Tissue-specific and
stage-specific expression of two silkworm ecdysone receptor isoforms --
ecdysteroid-dependent transcription in cultured anterior silk glands. Eur. J.
Biochem. 248, 786-793.

Kapitskaya, M. Z., Li, C., Miura, K., Segraves, W., and Raikhel, A. S. (2000). Expression
of the early-late gene encoding the nuclear receptor HR3 suggests its involvement

in regulating the vitellogenic response to ecdysone in the adult mosquito. Mol.
Cell. Endocrinol. 160, 25-37.

Kapitskaya, M., Wang, S. -F., Cress, D. E., Dhadialla, T. S., and Raikhel, A. S. (1996).
The mosquito ultraspiracle homologue, a partner of ecdysteroid receptor
heterodimer: cloning and characterization of isoforms expressed during
vitellogenesis. Mol. Cell. Endocrinol. 121, 119-132.

Kapitskaya, M.Z., Dittmer, N.T., Deitsch, K.W., Cho, W.L., Taylor, D.G., Leff, T., and
Raikhel, AS. (1998). Three isoforms of a hepatocyte nuclear factor-4

transcription factor with tissue- and stage-specific expression in the adult
mosquito. J. Biol. Chem. 273, 29801-29810.

Karim, F. D. and Thummel, CS. (1992). Temporal coordination of regulatory gene
expression by the steroid hormone ecdysone. EMBO J. 11, 4083-4093.

Kastner, P., Krust, A., Turcotte, B., Stropp, U., Tora, L., Gronemeyer. H., and Chambon,
P. (1990). Two distinct estrogen-regulated promoters generate transcripts
encoding the two functionally different human progesterone receptor forms A and
B. EMBO J. 9, 1603-14.

Keightley, M. C. (1998) Steroid receptor isoforms: exception or rule? Mol. Cell
Endocrinol. 137, 1-5

Klein-Hitpass, L., Tsai, S.Y., Weigel, N.L., Allan, G.F., Riley, D., Rodriguez, R.,
Schrader, W.T., Tsai, M.J. and O'Malley, B.W. (1990). The progesterone receptor
stimulates cell-free transcription by enhancing the formation of a stable
preinitiation complex. Cell 60, 247-257.

Klock, G., Strahle, U., and Schutz, G. (1987). Oestrogen and glucocorticoid responsive
elements are closely related but distinct. Nature. 329, 734-736.

Koelle, M. R., Talbot, W. S., Segraves, W..A., Bender, M. T., Cherbas, P., and Hogness,
D.S. (1991). The Drosophila EcR gene encodes an ecdysone receptor, a new
member of the steroid receptor superfamily. Cell 67, 59-77

212

Koelle, M.R., Segraves, W.A., and Hogness, D.S. (1992). DHR3: a Drosophila steroid
receptor homolog. Proc. Natl. Acad. Sci. USA 89, 6167-6171.

Kostrouch, Z., Kostrouchova, M., and Rall, J .E. (1995). Steroid/thyroid hormone receptor
genes in Caenorhabditis elegans. Proc. Natl. Acad. Sci. USA 92, 156-159.

Kozlova, T. and Thummel, C. S. (2000). Steroid regulation of postembryonic
development and reproduction in Drosophila.Trends Endocrinol. Metab. 11, 276-
280. Review.

Kuiper, G. G., Enmark, E., Pelto-Huikko, M., Nilsson, S., and Gustafsson, J. A. (1996).
Cloning of a novel receptor expressed in rat prostate and ovary. Proc. Natl. Acad.
Sci. USA 93, 5925-30.

Kumar, V., Green, S., Stack, G., Berry, M., Jin, J.R., and Chambon, P. (1987). Functional
domains of the human estrogen receptor.
Cell 51, 941-951.

Kurokawa, R., Soderstrom, M., Horlein, A., Halachmi, S., Brown, M., Rosenfeld, M.G.,
and Glass, CK. (1995). Polarity-specific activities of retinoic acid receptors
determined by a co-repressor. Nature 377, 451-454.

Lam, G., Hall, B.L., Bender, M., and Thummel, CS. (1999). DHR3 is required for the
prepupal-pupal transition and differentiation of adult structures during Drosophila
metamorphosis. Dev. Biol. 212, 204-16.

Lam, G.T., J iang, C., and Thummel, CS. (1997). Coordination of larval and prepupal
gene expression by the DHR3 orphan receptor during Drosophila metamorphosis.
Development 124, 1757-1769.

Lan, Q., Hiruma, K., Hu, X., Jindra, M., and Riddiford, L. M. (1999). Activation of a
delayed-early gene encoding MHR3 by the ecdysone receptor heterodimer EcR-
Bl-USP-l but not by EcR-Bl-USP-Z. Mol. Cell Biol. 19, 4897-4906.

Lavorgna, G., Karim, F. D., Thummel, C. S., and Wu, C. (1993). Potential role for a
FTZ-Fl steroid receptor superfamily member in the control of Drosophila
metamorphosis. Proc. Natl. Acad. Sci. USA 90, 3004-3008.

Lavorgna, G., Ueda, H., Clos, J ., and Wu, C. (1991). FTZ-Fl, a steroid hormone
receptor-like protein implicated in the activation of fushi tarazu. Science 252,

848-851.

Lazar, M.A., Hodin, R.A., Cardona, G., and Chin, W.W. (1990). Gene expression from
the c-erbAa/Rev-erbAa genomic locus. J. Biol. Chem. 265, 12859-12863.

213

Lea, A. (1967). The medial neurosecretory cells and egg maturation in mosquitoes. J.
Insect Phyio]. 13, 419-429.

Lea, A. O. (1964). Studies on the dietary and endocrine regulation of autogenous Aedes
taeniorhynchus. J. Med. Entomol. 1, 40-44.

Lea, AD. (1967). The medial neurosecretory cells and egg maturation in mosquitoes. J.
Insect. Phyiol. 13, 419-429.

Lea, AD. (1972). Regulation of egg maturation in the mosquito by the neurosecretory
system: the role of the corpus cardiacum. Gen. Comp. Endocrinol. Suppl. 3, 602-
608.

Lee, H. L, and Archer, T. K. (1998). Prolonged glucocorticoid exposure
dephosphorylates histone H1 and inactivates the MMTV promoter. EMBO J. 17,
1454-1466.

Lehmann, M. and Korge, G. (1995). Ecdysone regulation of the Drosophila Sgs-4 gene is
mediated by the synergistic action of ecdysone receptor and SEBP 3. EMBO J.
14, 716-726.

Li, C., Kapitskaya, M. Z., Zhu, J ., Miura, K., Segraves, W., Raikhel, A. S. (2000).
Conserved molecular mechanism for the stage specificity of the mosquito
vitellogenic response to ecdysone. Dev. Biol. 224, 96-110.

Li, F. Q., Ueda, H., and Hirose, S. (1994). Mediators of activation of fushi tarazu gene
transcription by BmFTZ-F1.Mol. Cell. Biol. 14, 3013-3021.

Lin, Y., Hamblin, M.T., Edwards, M.J., Barillas-Mury, C., Kanost, M.R., Knipple, D.C.,
Wolfner, M.F., and Hagedom, H.H. (1993). Structure, expression, and hormonal
control of genes from the mosquito, Aedes aegypti, which encode proteins similar
to the vitellin membrane proteins of Drosophila melanogaster. Dev. Biol. 155,

558-568.

Luisi, B.F., Xu, W.X., Otwinowski, Z., Freedman, L.P., Yamamoto, K.R., and Sigler,
RB. (1991). Crystallographic analysis of the interaction of the glucocorticoid
receptor with DNA. Nature 352, 497-505.

Ma, M., Zhang, J.-Z., Gong, H., and Gwardz, R. (1988). Permissive action of juvenile
hormone on vitellogenin production by the mosquito, A edes aegypti. J. Insect
Physiol. 34, 593-596.

Mader, S., Kumar, V., dc Vemeuil, H., and Chambon, P. (1989). Three amino acids of

the oestrogen receptor are essential to its ability to distinguish an oestrogen from a
glucocorticoid-responsive element. Nature 338, 271-274.

214

Mangelsdorf, D. J., Thummel, C., Beato, M., Herrlich, P., Schutz, G., Umesono, K.,
Blumberg, B., Kastner, R, Mark, M., Chambon, P., and Evans, R. M. (1995). The
nuclear receptor superfamily: the second decade. Cell 83, 835-839.

Martin, D., Wang, S., and Raikhel, A. S. (2001). The vitellogenin gene of the mosquito
Aedes aegypti is a direct target of ecdysteroid receptor. Mol. Cell. Endocrinol.
173, 75-86.

Matsumoto, S., Brown, M., Suzuki, A., and Lea, A. (1989). Isolation and characterization
of ovarian ecdysteroidogenic hormones from the mosquito, Aedes aegypti. Insect
Biochem. 19, 651-656.

McBroom, L.D., Flock, G., and Giguere, V. (1995). The non-conserved hinge region and
distinct amino-terminal domains of the RORa orphan nuclear receptor isoforms
are required for proper DNA bending and RORa-DNA interactions. Mol. Cell.
Biol. 15, 796-808.

McKenna, N.J., Lanz, RB, and O'Malley, B.W. (1999). Nuclear receptor coregulators:
cellular and molecular biology. Endocr. Rev. 20, 321—44. Review.

Miura, K., Wang, S.-F., and Raikhel, A. R. (1999). Two distinct subpopulations of
ecdysone receptor complex in the female mosquito during vetellogenesis. Mol.
Cell. Endocrinol. 156, 111-120.

Miyajima, N., Horiuchi, R., Shibuya, Y., F ukushige, S., Matsubara, K., Toyoshima, K.,
and Yamamoto, T. (1989). Two erbA homologs encoding proteins with different
T3 binding capacities are transcribed from opposite DNA strands of the same
genetic locus. Cell 57, 31-39.

Mouillet, J. F., Bousquet, F ., Sedano, N., Alabouvette, J ., Nicolai, M., Zelus, D., Laudet,
V. and Delachambre, J. (1999). Cloning and characterization of new orphan
nuclear receptors and their developmental profiles during Tenebrio

metamorphosis. Eur. J. Biochem. 265, 972-981.

Ohno, C. K., Ueda, H., and Petkovich, M. (1994). The Drosophila nuclear receptors
FTZ-Fl alpha and FTZ-Fl beta compete as monomers for binding to a site in the
fushi tarazu gene. Mol. Cell. Biol. 14, 3166-3175.

O'Malley, B. W. and Conneely, O. M. (1992). Orphan receptors: in search of a unifying
hypothesis for activation. Mol. Endocrinol. 6, 1359-1361.

Onate, S. A., Tsai, S. Y., Tsai, M. J., and O'Malley, B. W. (1995). Sequence and

characterization of a coactivator for the steroid hormone receptor superfamily.
Science 270, 1354-1357.

215

Oro, A. E., McKeown, M., and Evans, R. M. (1990). Relationship between the product of
the Drosophila ultraspiracle locus and the vertebrate retinoid X receptor. Nature
347, 298-301.

Oro, A.E., McKeown, M., Evans, RM. (1992). The Drosophila retinoid X receptor
homolog ultraspiracle functions in both female reproduction and eye
morphogenesis. Development 115, 449-462.

Paech, K., Webb, P., Kuiper, G. G., Nilsson, S., Gustafsson, J ., Kushner, P.J., and
Scanlan, TS. (1997). Differential ligand activation of estrogen receptors ER-or
and ER-B at AP] sites. Science 277, 1508-1510.

Palli, S.R., Hiruma, K., Riddiford, L.M. (1992). An ecdysteroid-inducible Manduca gene
similar to the Drosophila DHR3 gene, a member of the steroid hormone receptor
superfamily. Dev. Biol. 150, 306-318.

Palli, S.R., Ladd, T.R., Ricci, A.R., Sohi, SS, and Retnakaran, A. (1997). Cloning and
developmental expression of Choristonerura hormone receptor 75: A homolog of
the Drosophila E75A gene. Dev. Genet. 20, 36-46.

Palli, S.R., Ladd, T.R., Sohi, S.S., Cook, B.J., and Retnakaran, A. (1996). Cloning and
developmental expression of Chortstoneura hormone receptor 3, an ecdysone-
inducible gene and a member of the steroid hormone receptor superfamily. Insect
Biochem. Mol. Biol. 26, 485-499.

Perera, S.C., Ladd, T.R., Dhadialla, T.S., Krell, P.J., Sohi, S.S., Retnakaran, A., and
Palli, SR. (1999). Studies on two ecdysone receptor isoforms of the spruce
budworm, Choristoneurafumiferana. Mol. Cell. Endocrinol. 152, 73-84.

Perlmann, T., Rangarajan, P.N., Umesono, K., and Evans, RM. (1993). Determinants for
selective RAR and TR recognition of direct repeat HREs. Genes Dev. 7, 1411-
1422.

Picard, D., Khursheed, B., Garabedian, M.J., F ortin, M.G., Lindquist, S., and Yamamoto,
KR. (1990). Reduced levels of hsp90 compromise steroid receptor action in vivo.
Nature 348, 166-168.

Pierceall, W. E., Li, C., Biran, A., Miura, K., Raikhel, A. S., and Segraves, W. A. (1999).
E75 expression in mosquito ovary and fat body suggests reiterative use of
ecdysone-regulated hierarchies in development and reproduction. Mol. Cell
Endocrinol. 150, 73-89.

Puigserver, P., Wu, Z., Park, C. W., Graves, R., Wright, M., and Spiegelman, B. M.

(1998). A cold-inducible coactivator of nuclear receptors linked to adaptive
thermogenesis. Cell 92, 829-839.

216

Raikhel, A. S. (1992). Vitellogenesis in mosquitoes. Advance Disease Vector Res. 9, l-
39.

Raikhel, A. S., and Lea, A. O. (1985). Hormone-mediated formation of the endocytic
complex in mosquito oocytes. Gen. Comp. Endocrinol. 57, 422-433.

Raikhel, A. S., and Lea, A. O. (1990). Juvenile hormone controls previtellogenic
proliferation of ribosomal RNA in the mosquito fat body. Gen. Comp.
Endocrinol. 77, 423-434.

Raikhel, A. S., Deitsch, K. W., and Sappington, T. W. (1997). Culture and analysis of
the insect fat body. In “The Molecular Biology of Insect Disease Vectors, A
Methods Manual”. (J. M. Crampton, C. B. Beard, and C. Louis, Eds.), pp 507-
522. Chapman & Hall, London, UK.

Raikhel, A.S., Deitsch, K.W., and Sappington, T.W. (1997). Culture and analysis of the
insect fat body. In: The molecular biology of insect disease vectors. A methods
manual (Crampton, J .M., Beard, CB. and Louis, C., eds.) pp. 507-522. Chapman
and Hall, London.

Raikhel, AS. and Lea, AD. (1991). Control of follicular epithelium development and
vitelline envelope formation in the mosquito; role of juvenile hormone and 20-
hydroxyecdysone. Tissue Cell 23, 577-591.

Raikhel, A.S., Miura, K., and Segraves, W.A.(l999). Nuclear receptors in mosquito
vitellogenesis. Amer. Zoo]. 39, 722-735.

Rastinejad, F. (1998). Structure and function of the steroid and nuclear receptor DNA
binding domain. In “Molecular Biology of Steroid and Nuclear Hormone
receptors” (Leonard P. Freedman, Eds), pp. 105-132. Birkhauser Boston,
Cambridge.

Richards, G. (19763). The control of prepupal puffing patterns in vitro; implications for
prepupal ecdysone titers in Drosophilia melanogaster. Dev. Biol. 48, 191-195.

Richards, G. (1976b). Sequential gene activation by ecdysone in polytene chromosomes
of Drosophila melanogaster. IV. The mid prepupal period. Dev. Biol. 54, 256-
263.

Richards, G. (1976c). Sequential gene activation by ecdysone in polytene chromosomes
of Drosophila melanogaster. V. The late prepupal puffs. Dev. Biol. 54, 264-275.

Riddiford, L. M. (1985). Hormone action at the cellular level. In: “Comprehensive insect

physiology, biochemistry, and pharmacology” (G.A. Kerkut and LI. Gibert,
Eds), Vol. 8, pp. 37-84. Plenum, New York.

217

Riddiford, L. M. (1993a). Hormone and Drosophila development in “The Development of
Drosophila melanogaster”, (M. Bate, and A. Martinez Eds.), Vol. 2, pp.899-939.
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.

Riddiford, L.M. (1993b). Hormone receptors and the regulation of insect metamorphosis.
Receptor 3, 203-209. Review.

Riddiford, L.M. (1996). Juvenile hormone: The status of its “Status Quo” action. Arch.
Insect Biochem. Physiol. 32, 271-286.

Riddiford, L.M., Cherbas, P., and Truman, J.W. (2001). Ecdysone receptors and their
biological actions. Vitam. Harm. 60,1-73. Review.

Riddihough, G. and Pelham, H.R.B. (1987). An ecdysone responsive element in the
Drosophila hsp27 promoter. EMBO J. 6, 3729-3 734.

Robinow, S., Talbot, W. S., Hogness, D. S., and Truman, J. W. (1993). Programmed cell
death in the Drosophila CNS is ecdysone-regulated and coupled with a specific
ecdysone receptor isoform. Development 119, 1251-1259.

Robyr, D., Wolffe, AP. and Wahli W. (2000). Nuclear hormone receptor coregulators in
action: diversity for shared tasks. Mol. Endocrinol. 14, 329-347. Review.

Russell, S. and Ashbumer, M. (1996). Ecdysone-regulated chromosome puffing in
Drosophila melanogaster. IN: Metamorphosis (eds. Gilbert, L.I., Tata, J .R., and
Atkinson, B.G.), pp. 109-144, Academic Press.

Sappington, T. W. and Raikhel, A. S. (1999). Aedes aegypti. In “Encyclopedia of
Reproduction” (E. Knobil and J. D. Neill, Eds), Vol. 1, pp.61-77. Academic
Press, San Diego.

Sappington, T. W., Brown M. R. and Raikhel, A. S. (1997). Culture and analysis of insect
ovaries. In “Molecular Biology of Insect Disease Vectors: A methods manual”
(J .M. Crampton, C.B. Beard and C.Louis, Eds), Chapter 42, pp.523-535.
Chapman & Hall.

Sappington, T.W., Kokoza, V.A., Cho, W.-L., and Raikhel, AS. (1996). Molecular
characterization of the mosquito vitellogenin receptor reveals unexpected high
homology to the Drosophila yolk protein receptor. Proc. Natl. Acad. Sci. USA 93,
8934-8939.

Schubiger, M., Wade, A. A., Carney, G. E., Truman, J. W., and Bender, M. (1998).

Drosophila EcR-B ecdysone receptor isoforms are required for larval molting and
for neuron remodeling during metamorphosis. Development 125, 2053-2062.

218

Segraves, W.A. (1988). Molecular and genetic analysis of the E75 ecdysone-responsive
gene of Drosophila melanogaster. Ph.D. thesis, Stanford University, Stanford,
California.

Segraves, W.A. and Hogness, D.S. (1990). The E75 ecdysone-inducible gene responsible
for the 75B early puff in Drosophila encodes two new members of the steroid
receptor superfamily. Genes Dev. 4, 204-219.

Segraves, W.A. and Woldin, CW. (1993). The E75 gene of Manduca sexta and
comparison with its Drosophila homolog. Insect Biochem. Molec. Biol. 23, 91-97.

Shapiro, A.B., Wheelock, G. D., Hagedom, H. H., Baker, F. C. Tsai, L. W., and
Schooley, D. H. (1986). Juvenile hormone and the juvenile hormone esterase in
adult female of the mosquito Aedes aegypti. J. Insect Physiol. 32, 867-877.

Shapiro, J. P. and Hagedom, H. H. ( 1982). Juvenile hormone and the development of
ovarian responsiveness to a brain hormone in the mosquito, Aedes aegipti. Gen.
Comp. Endocrinol. 46, 176-183.

Shea, M. J., King, D. L., Conboy, M. J., Mariani, B. D., and Kafatos, F. C. (1990).
Proteins that bind to Drosophila chorion cis-regulatory elements: a new C2H2

zinc finger protein and a C2C2 steroid receptor-like component. Genes Dev. 4,
1 128-1 140.

Shi, Y. B., Wong, J ., Puzianowska-Kuznicka, M., and Stolow, M. A. (1996). Tadpole
competence and tissue-specific temporal regulation of amphibian metamorphosis:
roles of thyroid hormone and its receptors. Bioessays 18, 391-399.

Stancato, L.F., Silverstein, A.M., Gitler, C., Groner, B., and Pratt, W.B. (1996). Use of
the thiol-specific derivatizing agent N-iodoacetyl-3-[1251]iodotyrosine to
demonstrate conformational differences between the unbound and hsp90—bound
glucocorticoid receptor hormone binding domain. J. Biol. Chem. 271, 8831-8836.

Steinbach, O. C., Wolffe, A. P., and Rupp, R. A. (1997). Somatic linker histones cause
loss of mesodermal competence in Xenopus. Nature 389, 395-399.

Stone, BL. and Thummel, CS. (1993). The Drosophila 78C early late puff contains E78,
an ecdysone-inducible gene that encodes a novel member of the nuclear hormone
receptor superfamily. Cell 75, 307—320.

Sun, J ., Hiraoka, T., Dittmer, N.T., Cho, K.H., and Raikhel, AS. (2000). Lipophorin as a
yolk protein precursor in the mosquito, Aedes aegypti. Insect Biochem. Mol. Biol.
30,1161-1171.

219

Sun, G. C., Hirose, S., and Ueda, H. (1994). Intermittent expression of BmFTZ-F l, a
member of the nuclear hormone receptor superfamily during development of the
silkworm Bombyx mori. Dev. Biol. 162, 426-437.

Sutherland, J .D., Kozlova, T., Tzertzinis, G., Kafatos, F .C. (1995). Drosophila hormone
receptor 38: a second partner for Drosophila USP suggests an unexpected role for

nuclear receptors of the nerve growth factor-induced protein B type. Proc. Natl.
Acad. Sci. USA 92, 7966-70.

Swevers, L., Drevet, J. R., Junke, M. D., and latrou, K. (1995). The silkmoth homolog of
the Drosophila ecdysone receptor (B1 isoform): cloning and analysis of

expression during follicular cell differentiation. Insect Biochem. Mo]. Bio]. 25,
857-866.

Syvala, H., Pekki, A., Blauer, M., Pasanen, S., Makinen, E., Ylikomi, T., and Tuohimaa,
P. (1996). Hormone-dependent changes in A and B forms of progesterone
receptor. J. Steroid. Biochem. Mol. Biol. 58, 517-524.

Talbot, W. S., Swyryd EA, and Hogness D.S. (1993). Drosophila tissues with different
metamorphic responses to ecdysone express different ecdysone receptor isoforms.
Cell 73, 1323-1337.

Thomas, H. E., Stunnenberg, H. G., and Stewart, A. F. (1993). Heterodimerization of the
Drosophila ecdysone receptor with retinoid X receptor and ultraspiracle. Nature
362, 471-475.

Thummel, CS. (1992). Mechanisms of transcriptional timing in Drosophila. Science
255, 39-40.

Thummel, CS. (1995). From embryogenesis to metamorphosis: the regulation and
function of Drosophila nuclear receptor superfamily members. Cell 83, 871-877.
Review.

Thummel, CS. (1996). Files on steroids--Drosophila metamorphosis and the
mechanisms of steroid hormone action. Trends Genet. 12, 306-310. Review.

Thummel, CS. (1997). Dueling orphans--interacting nuclear receptors coordinate
Drosophila metamorphosis. Bioessays 19, 669-672. Review.

Thummel, C.S., Burtis, K. C., and Hogness, D.S. (1990). Spatial and temporal patterns of
E 74 transcription during Drosophila development. Cell 61, 101-111.

Truman, J. W., Talbot, W. S., Fahrbach, S. E., and Hogness, D. S. (1994). Ecdysone
receptor expression in the CNS correlates with stage-specific responses to
ecdysteroids during Drosophila and Manduca development. Development 120,
219-234.

220

Tsai, C.C., Kao, H.Y., Yao, T.P., McKeown, M., and Evans, R.M. (I999). SMRTER, a
Drosophila nuclear receptor coregulator, reveals that EcR-mediated repression is
critical for development. Mol. Cell. 4, 175-186.

Tsai, M.J. and O'Malley, B.W. (1994). Molecular mechanisms of action of
steroid/thyroid receptor superfamily members. Annu. Rev. Biochem. 63, 451-486.
Review.

Tsai, S.Y., Srinivasan, G., Allan, G.F., Thompson, E.B., O'Malley, B.W., and Tsai, M.J.
(1990). Recombinant human glucocorticoid receptor induces transcription of
hormone response genes in vitro. J. Biol. Chem. 265, 17055-17061.

Tugwood, J. D., Issemann, I., and Green, S. (1991). Liver receptor homologue protein.
GenBank DNA Database Accession No. M81385.

Tzertzinis, G., Malecki, A., and Kafatos, F.C. (1994). BmCFl, a Bombyx mori RXR-type
receptor related to the Drosophila ultraspiracle. J. Mol. Bio]. 238, 479-486.

Ueda, H. and Hirose, S. (1991). Defining the sequence recognized with BmFTZ-Fl, a
sequence specific DNA binding factor in the silkworm, Bombyx mori, as revealed
by direct sequencing of bound oligonucleotides and gel mobility shifi competition

analysis. Nucleic Acids Res. 19, 3689-3693.

Ueda, H., Sonoda, S., Brown, J. L., Scott, M. P., and Wu, C. (1990). A sequence-specific
DNA-binding protein that activates fushi tarazu segmentation gene expression.
Genes Dev. 4, 624-635.

Ueda, H., Sun, G. C., Murata, T., and Hirose, S. (1992). A novel DNA-binding motif
abuts the zinc finger domain of insect nuclear hormone receptor FTZ-Fl and
mouse embryonal long terminal repeat-binding protein. Mol. Cell. Biol. 12, 5667-
5672.

Umesono, K., Murakami, K.K., Thompson, CC, and Evans, R.M. (1991). Direct repeats
as selective response elements for the thyroid hormone, retinoic acid, and vitamin
D3 receptors. Cell 65, 1255-1266.

Vogtli, M., Elke, C., Imhof,.M.O., and Lezzi, M. (1998). High level transactivation by
the ecdysone receptor complex at the core recognition motif. Nucleic Acids Res.
26, 2407-2414.

Vogtli, M., Imhof, M.G., Brown, N.E., Rauch, P., Spindler-Barth, M., Lezzi, M., and
Henrich, V.C. (1999). Functional characterization of two Ultraspiracle forms
(CtUSP-l and CtUSP-Z) from Chironomus tentans. Insect Biochem. Mol. Bio]. 29,
931-942.

221

Walker, V.K. and Ashbumer, M. (1981). The control of ecdysterone-regulated puffs in
Drosophila salivary glands. Cell 61, 101-111.

Wang, S.-F., Li, C., Zhu, J., Miura, K., Miksicek, R. J ., and Raikhel, A. S. (20003).
Differential expression and regulation by 20-hydroxyecdysone of mosquito
ultraspiracle isoforms. Dev. Biol. 218, 99-113.

Wang, S. F ., Ayer, S., Segraves, W. A., Williams, D. R., and Raikhel, A. S. (2000b).
Molecular determinants of differential ligand sensitivities of insect ecdysteroid
receptors. Mol. Cell. Biol. 20, 3870-3879.

Wang, S.-F., Miura, K., Miksicek, R. J ., Segraves, W. A., and Raikhel, A. S (1998). DNA
binding and transactivation characteristics of the mosquito ecdysone receptor-
ultraspiracle complex. J. Biol. Chem. 273, 27531-27540.

White, K.P., Hurban, P., Watanabe, T., and Hogness, D.S. (1997). Coordination of
Drosophila metamorphosis by two ecdysone-induced nuclear receptors. Science
276,114-117.

Wilson, T. E., F ahmer, T. J ., and Milbrandt, J. (1993). The orphan receptors NGFI-B and
steroidogenic factor I establish monomer binding as a third paradigm of nuclear
receptor-DNA interaction. Mo]. Cell. Biol. 13, 5794-5804.

Woodard, C. T., Baehrecke, E. H., and Thummel, C. S. (1994). A molecular mechanism
for the stage specificity of the Drosophila prepupal genetic response to ecdysone.
Cell 79, 607-615.

Xu, L., Glass, C.K., and Rosenfeld, MG. (1999). Coactivator and corepressor complexes
in nuclear receptor function. Curr. Opin. Genet Dev. 9, 140-147.

Yao, T.-P., Forman, B.M., Jiang, Z., Cherbas, L, Chen, J.-D., McKewon, M., Cherbas, P.,
and Evans, R.M. (1993). Functional ecdysone receptor is the product of EcR and
Ultraspiracle genes. Nature 336, 476-479.

Yao, T.-P., Segraves, W.A., Oro, A.E., McKeown, M., and Evans, R.M. (1992).
Drosophila ultraspiracle modulates ecdysone receptor function via heterodimer
formation. Cell 71, 63-72.

Zamir, 1., Dawson, J., Lavinsky, R.M., Glass, C.K., Rosenfeld, M.G., and Lazar, MA.
(1997). Cloning and characterization of a corepressor and potential component of

the nuclear hormone receptor repression complex. Proc. Natl. Acad. Sci. USA. 94,
14400- 14405.

Zechel, C., Shen, X.Q., Chen, J .Y., Chen, Z.P., Chambon, P., and Gronemeyer, H.
(1994). The dimerization interfaces formed between the DNA binding domains of
RXR, RAR and TR determine the binding specificity and polarity of the full-
length receptors to direct repeats. EMBO J. 13, 1425-1433.

222

Zelhof, A. C., Yao, T. P., Chen, J. D., Evans, R. M., and McKeown, M. (1995a). Seven-
up inhibits ultraspiracle-based signaling pathways in vitro and in vivo. Mol. Cell.
Biol. 15, 6736-45.

Zelhof, A. C., Yao, T. P., Evans, R. M., and McKeown, M. (l995b). Identification and
characterization of a Drosophila nuclear receptor with the ability to inhibit the
ecdysone response. Proc. Natl. Acad. Sci. USA 92, 10477-81.

Zhou, B., Hiruma, K., Jindra, M., Shinoda, T., Segraves, W.A., Malone, F ., and
Riddiford, L.M. (1998). Regulation of the transcription factor E75 by 20-
hydroxyecdysone and juvenile hormone in the epidermis of the tobacco
homworm, Manduca sexta, during larval molting and metamorphosis. Dev. Biol.

193, 127-138.
Zhu, J ., Miura, K., Chen, L., and Raikhel, A. S. (2000). AHR38, a homolog of NGFI-B,

inhibits formation of the functional ecdysteroid receptor in the mosquito Aedes
aegypti. EMBO J. 19, 253-62.

223