1.4.“; r‘
:5”. .333; an"

k
..:s. .. W
mm: “Jim ”Elm:

    
 
  
  

 

   

 
 

     
 

  

  
 
  

. «4. u. m ~.. "I:
1~n§:‘:€q 52.1. ‘9‘""""v
b‘t $.32“ .3” $3.?“ 5-.“
. ”W W,
. , . I ‘ ‘ til“ K E
. , . ‘7 , 'A; . . ., . . , -A~.-»...L. :
. , > ' . > 5. v -.t(‘u_yl.uc.m
> ‘ _ . . -
. , , s ‘ a. : ‘y'z‘éx 1:58"
.5 . .. .
. k
v
.
. ,
l‘ '
.
4
q , AM“
I: Z. \ ,nw' 5...»:

 
    

 

(«A-luv- -V\-§w:h~h:wm 5.
~ - “ii-L: ‘1... gig; ‘tigt
.‘ ”"~-‘

“1.1. .

  
  

n

J.
tflflfl‘f'l‘w'f

 

1‘

. -
.u

{viiin «~35
.um. {FISH 1:“

     

 

:rtu

“4,
‘I

    
      
   
    
     

n . t.. W -‘-
.. . ‘ "‘ A - «m. 5.. : " “N - '1..MR"’.2"”"
Y'-~I~:Aw.lul.u:,. , _ ' ~ v u .. \ W ,_
.. . a. v: , w y W - s

 

 

 
 
 
 

 

\

1 3

'E.

"‘ 3.3.4.:
7’" \41 Ci”;-
_.1 ‘5 he}? ‘11::
. ‘.

. .

(

. .

.‘ ‘

H‘- h

.I P

M

F‘

  

 

.
v
{ 4"“2‘fié‘.‘ wan“
., -. . w” .5» .
‘Ir a 3‘ a ..-. fl
'3', ‘ I - ‘v.
t.»

  

 

:-

  

‘Mu .n‘.

 

 

 

 

 

 

 

 

nu ~ '4: n‘ ‘ ‘

s.‘.'l‘|ll'u(“u~v r

 

 

 

           

   

     

 

 

 

   

 

   

   

     
  
 

 

 

 

     

 

 

 

 
     

   
 

 

 

4 a ~ . 'Q
. ' It . ~....:..(.‘.
‘ r .: ...
. . .. _ , - . , ‘
| ' ~ . ' . .
.,' I n ‘
.. ,_ ‘ _
,
L" ' ‘ .. . _ y .I ‘ . ,
n '. "‘ . ,, ‘
V .u > ,' ,' 4
. . . ¢ . " . "
u ., l - . - q .. .’ .
-.‘ W. I I .1. ~ 11" . V ‘
~ :1 ‘ ‘ '
~' ' . '. .' a
v , ' , "
‘ ‘. , ‘~. .‘.
. ‘ v .‘
‘ ‘ _ . , ' . :3»... .1.‘
. ' ' ' -.. . 7v .. "If.” I... M
. . ,. . _ n .. .’ v I “ v -- ‘ nv‘fl Or‘qwrmvtnl c a“-
. . .’ . . ~ on 4—.“
II‘ c n . - . v
. ' I , .
. .- ..“ -. .

          

MICHI IGAN STAT

mu Ill/IllIll/I/I/lI/l" :llll‘lll‘

 

 

Isl/Ill

 

 

This is to certify that the

dissertation entitled

Characterization and CDNA Cloning of

Proteases Involved in Mosquito Vitello-

- sis
presented by gene

CHO , WEN- LONG

has been accepted towards fulfillment
of the requirements for

Ph. D. degree in Genetics

WWW fife/WK

Dr. Alexander S. Raikhel

Major professor

 

Date May 20: 1992

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

 

 

 

 

 

 

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE ill RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

. , hav- }' 4:
'. A II. 4:.
we - aw
, , l '5' ab

 

     

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1‘1: 1i

MSU is An Affirmative Action/Equal Opportunity institution
cMMMt

 

i——.—. -

CHARACTERIZATION AND cDNA CLONING OF PROTEASES
iNVOLVED IN THE MOSQUITO VITELLOGENESIS
BY
Wen-Long Cho

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Genetics Graduate Program

1 992

ABSTRACT

CHARACTERIZATION AND cDNA CLONING OF PROTEASES
INVOLVED iN THE MOSQUITO VITELLOGENESIS
By
Wen-Long Cho

In the mosquito. Aedes aegypti, a lysosomal aspartic protease (mLAP),
which is involved in the termination of vitellogenesis in the fat body, was purified
and characterized. The native molecular weight of the purified mLAP
determined by PAGE under non denaturing conditions was 80,000. The enzyme
was resolved into a single 40 kDa peptide with SDS-PAGE under either reducing
or non-reducing conditions. This mosquito LAP has a pl of 5.4 and its optimal
condition for enzymatic reaction is pH 3.0 at 45°C using hemoglobin as a
substrate. The cDNA corresponding to mLAP was isolated and sequenced. The
deduced amino acid sequence exhibits 92% and 81% similarity to human
cathepsin D (H00) and E (HCE) respectively. Kinetic analysi of mLAP in the
mosquito fat body at the mRNA and protein levels revealed a 12 hr lag in its
translation. Translational regulation of mLAP mRNA may occur. The 5'-
untranslated region of mLAP mRNA is similar to elements conferring negative

translational control by steroids.

A cDNA encoding a mosquito vitellogenic 53 kDa protein was cloned and
sequenced. This cDNA hybridizes to a 1.5 kb mRNA present only in the fat body
of vitellogenic females. The analysis of the deduced amino acid sequence from

this cDNA indicates that it is a serine carboxypeptidase.

The enzymatic property of this mosquito vitellogenic carboxypeptidase (VCP)
was identified using an assay with a serine protease inhibitor. [3H] diisopropyl
fluorophosphate. The finding that an enzyme required for embryonic
development is synthesized outside of oocytes represents a biological

phenomenon previously unknown.

ACKNOWLEDGMENTS

I sincerely appreciate Dr. Alex Raikhel's guidance. support. and
encouragements . I would like to thank the members of my committee, Dr.
Tom Friedman. Dr. Will Kopachik. Dr. Natasha Raikhei and Dr. Charles
Sweeley. for their comments. suggemions and careful reading of my
manuscripts. I would like to extend my special thanks to Dr. Tochi Dhadialla
and Mr. Alan Hays for their expert technical instructions in protein
biochemistry. I appreciate the help in reviewing manuscripts from my fellow
students, Kirk Deitsch and Neal Dittumer. i would also like to thank all the
people from Alex's laboratory for their friendship.

Finally, I would like to thank my mother for her understanding and my

wife. Chuen-Hui Su whom I owe my greatest appreciation for her love and
making my life more comfortable during this period.

IV

CONTENTS

Page
LIST OF TABLES ............................................................................................ Vll
LIST OF FIGURES .......................................................................................... VIII
ABBREVIATIONS ............................................................................................ Xi
CHAPTER 1: introduction ........................................................................... 1
References ........................................................................... 13
CHAPTER 2: Purification and Characterization of a Lysosomal
Aspartic Protease with Cathepsin D Activity from the
Mosquito ............................................................................... 15
Abstract ................................................................................. 16
Introduction ........................................................................... 17
Experimental Procedures ..................................................... 19
Results .................................................................................. 27
Discussion ............................................................................ 58
References ........................................................................... 61
CHAPTER 3: Cloning of cDNA for Mosquito Lysosomal Aspartic
Protease ............................................................................... 63
Abstract ................................................................................ 64
Introduction .......................................................................... 65
Experimental Procedures ..................................................... 67
Results and Discussion ........................................................ 82
References ........................................................................... 105

CHAPTER 4:

CHAPTER 5:

An Extraovarian Protein Accumulated in Mosquito
Oocytes is a Carboxypeptidase Activated in

Embryos .............................................................................. 109
Abstract ............................................................................... 110
introduction ......................................................................... 1 1 1
Experimental Procedures .................................................... 1 12
Results and Discussion ....................................................... 115
References .......................................................................... 131
Summary and Future Research Prospects ......................... 134
References .......................................................................... 141

LIST OF TABLES

Page
Table 1-1 Mosquito-borne diseases mediated by arboviruses .................. 4
Table 2-1 Purification of mosquito lysosomal aspartic protease (mLAP).... 28
Table 3-1 Homology of mosquito lysosomal aspartic protease (mLAP)
to aspartic proteases ................................................................. 75
Table 4—1 Homology of mosquito vitellogenic carboxypeptidase (VCP)
to serine carboxypeptidases ........................................................ 123
Table 5-1 Comparison among mosquito lysosomal aspartic protease
(mLAP), human cathepsin D (HCD) and human cathepsin E
(HCE) .............................................................................................. 136

VII

LIST OF FIGURES

Page

CHAPTER 1
Figure 1 Jungle (A) and urban (B) cycles of yellow fever and dengue

fever ............................................................................................... 3
Figure 2 Summary of eventns during the first cycle of egg maturation

in anautogeneous mosquitoe. Aedes aegypti ............................. 6
Figure 3 A schematic representation of biosythetic pathway for

vitellogenin in mosquito fat bodies.
CHAPTER 2
Figure 1 Elution profile of mosquito proteins on S-sepharose fast flow

column (1.0 cm x 10 cm) ............................................................... 29
Figure 2 SDS-PAGE (10-15%) of mosquito lysosomal aspartic

protease with cathepsin D activity after various purification

steps as revealed with silver staining ........................................... 31
Figure 3 Non-denaturing polyacrylamide gel electrophoresis of

purified enzyme stained with silver ............................................... 34
Figure 4 Two dimensional gel electrophoresis of proteins in

fractions pooled after S-sepharose chromatography

from the lane 4 in Fig. 2 ............................................................ 36
Figure 5 Determination of isoelectric point (pl) of mosquito aspartic

protease ......................................................................................... 38
Figure 6 Effect of pH on the activity of mosquito aspartic protease ............. 41
Figure 7 Effect of temperature on the activity of mosquito aspartic

protease ......................................................................................... 43
Figure 8 Michaelis-Menten and Lineweaver-Burk plots for mosquito

aspartic protease with increasing concentrations of

hemoglobin .................................................................................... 45
Figure 9 lmmunoblot analysis of polyclonal antibodies against the

mosquito aspartic protease ........................................................... 48

VIII

Figure 10 Protein and enzyme distributions after isopycnic
centrifugation on 30% Percoll gradients of the mosquito

organelles ...................................................................................... 50
Figure 11 lmmunoblot analysis of fractions from Percoll gradients
of the mosquito organelles ........................................................... 54

Figure 12 Comparison of N-terminai amino acid sequence of
mosquito aspartic protease with sequences of
vertebrate cathepsin D and cathepsin E ............................ 56

CHAPTER 3
Figure 1 Design of primers for amplification of mosquito lysosomal
aspartic protease (mLAP) cDNA

fragment by the polymerase chain reaction (PCR) ..................... 69
Figure 2 Results of the PCR amplification and isolation of the

mLAP cDNA fragment ................................................................. 70
Figure 3 Sequencing strategy of mLAP cDNA clone ................................. 73

Figure 4 Northern blot analysis of V9 mRNA in the mosquito fat body
during vitellogenesis .................................................................... 77

Figure 5 Western blot analysis of V9 in the mosquito fatbody
during vitellogesis ........................................................................ 80

Figure 6 Nucleotide and deduced amino acid sequence of
mosquito lysosomal aspartic protease (mLAP) ............................ 83

Figure 7 Hydropathy plot of the deduced amino acid sequence
of mLAP ......................................................................................... 86

Figure 8 Alignment of amino acid sequences of mLAP. human
cathepsin D (HCD) and human cathepsin E (HCE) ..................... 89

Figure 9 Sequence alignment of regions of aspartic proteases
corresponding to the cleavage sites of cathepsins D ................. 92

Figure 10 Relationship between members of a family of aspartic
proteases ...................................................................................... 93

Figure 11 Expression of mLAP in the mosquito fat body during
vitellogenesis ................................................................................ 96

Figure 12 Kinetics of mLAP and V9 at the mRNA and protein levels
in the mosquito fat bodies during vitellogenesis .........................
Figure 13 Putative steroid regulatory elements in the 5'-untranslated
region of mLAP mRNA ................................................................
CHAPTER 4
Figure 1 Northern blot analysis of sex- and stage-specific
expression of the vitellogenic carboxypeptidase (VCP)
mRNA transcript ........................................................................
Figure 2 Northern blot analysis of tissue-specific expression of the
vitellogenic carboxypeptidase (VCP) mRNA transcript ........
Figure 3 Nucleotide and deduced amino acid sequences of
the mosquito vitellogenic carboxypeptidase (VCP) ...............
Figure 4 Alignment of amino acid sequences of the mosquito
carboxypeptidase (VCP) and the wheat carboxypeptidase Y
homolog (WCP) ........................................................................
Figure 5 immunoblot (A) and [3H]-diisopropyl fluorophosphate

binding (B) analyses of VCP during mosquito embryonic

development ...........................................................................

98

. 102

116

124

ABBREVIATIONS

CAT ........................................................... Chloram phenicol acetyltransferase
CNP .............................................. 2', 3'-cyclic nucleotide 3'-phosphodiesterase
DFP ..................................................................... Diisopropyi fluorophosphate
D‘IT ...................................................................................... 1 .4-Dithiothreitol
ECL ................................................................ Enhanced chemiluminescence
EDTA .............................................................. Ethylenedinitrilo tetraacetic acid
606 ............................................... Com putsoftware from University of Wisconsin

Genetics Computer Group
HCD ............................................................................... Human cathepsin D
HCE ................................................................................. Human cathepsin E
20-HE ........................................................................... 20-hydroxyecdysone
IEF .................................................................................. isoelectric focusing
kDa .............................................................................................. kilo-Dalton
MBP ................................................................................ Myelin basic protein
mLAP ..................................................... Mosquito lysosomal aspartic protease
PBM ..................................................................................... Post blood meal
PCR. ...................................................................... Polymerase chain reaction
pl .......................................................................................... isoelectric point
808 ................................................................................... Sodium dodecyl sulfate
SOS-PAGE .......................................... SDS polyacrylamide gel electrophoresis
VCP .................................................................. Vitellogenic carboxypeptidase
Vg .............................................................................................. Vitellogenin
WCP ........................................................................ Wheat carboxypeptidase

CHAPTER 1

INTRODUCTION

1. Mosquito-borne diseases:

Mosquito-borne human diseases remain a major international health
problem. It is estimated that there are 200 million new cases of malaria'annually
(Kemp et al., 1987). Another disease, filariasis. carried by mosquitoes, affects
more than 100 million people annually. in addition, many arboviral diseases,
such as yellow fever, dengue fever and encephalitis, mediated by mosquito
transmission are still jeopardizing human life (Tabachnick, 1991).

Malaria is a mosquito-borne infection caused by protozoa of the genus
Plasmodium. Four species of malaria infect human: P. falciparum, P. vivax, P.
ovale and P. malariae. Control measures directed at the parasite itself (drug
therapy) or the mosquito (insecticide) have eradicated or controlled malaria in
some countries, but have failed in much of the tropical world. Most forms of
parasites are intracellular and therefore not directly accessible to the human
immune system. Because of antigenic variation and [or strain difference, an
effective vaccine is still being developed (Kemp et al., 1987).

Many arboviruses. such as dengue and yellow fever viruses, appear to be
capable of indefinite survival in insects by transovarial transmission from one
generation to the next or to be maintained in infection cycles (Fig. 1) without the
intervention of the natural vertebrate or invertebrate hosts. (White and Fenner,
1986).

In addition to dengue fever and yellow fever, many other arbovirus
infections causing human diseases are mediated by mosquitoes as shown in
table 1 (White and Fenner, 1986). Control of arboviral disease rests upon (1)
vector control and (2) vaccination. Temporally, these two strategies effectively
control most of lethal encephalitis and hemorrhagic fevers. Nevertheless, the
insecticide resistance of mosquitoes and side effects of vaccination are still

potential problems for control of these diseases.

..o>o_ Sacco EB .93 32% Co 3.96 g :35 was g 295.. P .9“.

E

‘55.: >35:

$2.... d V v

9933. 83$ S 983 83$ 95322 2:63 .2

9882' ‘gagoz ’ ~

5E3: >252

  

  

Tile 1. Mosquito-borne diseases mediated by arboviruses.

Human Alphaviruses

VERTEBRATE
RESERVIOR

   
   
 

     
  

VIRUS DISTRIBUTION INVERTEBRATE
I Vector

 

 

 

 

 

 

} Chikungunya Asia,Africa Mosquito Man Fever,arthritis, ,
' hemorrhagic fever 1
O'nyong-nyong Africa Mosquito Man Fever,arthritis
Ross River Australia, Mosquito Marsupials Arthritis
Pacific
Sindbis Africa,Asia, Mosquito Birds Fever
Australia
Easter equine Americas Mosquito Birds Encephalitis
encephalitis
Western Americas Mosquito Birds Encephalitis
equine
encephalitis

 

 

 

 

. Human Flaviviruses ‘

 

 

 

 

 

 

 

 

l
l VIRUS DISTRIBUTION lNVERTEBRATE VERTEBRATE DISEASE
1 VECTOR RESERVOIR IN MAN
Yellow fever Africa, Mosquito Monkeys, Hepatities,
South America man hemorrhagic fever
Dengue 1-4 Asia, Pacific, Mosquito Monkeys, Fever,
Caribbean, man hemorrhagic fever
Africa
Japanese East and SE Mosquito Birds, pigs Encephalitis
encephalitis Asia
Murray Australia, Mosquito Birds Encephalitis
valley New Guinea
encephalitis
West Nile Africa, Europe, Mosquito Birds Fever,
fever Middle East encephalitis
St. Louis Americas Mosquito Birds Encephalitis
encephalitis
Rocio Brazil Mosquito Birds Encephalitis
Venezuelan Americas Mosquito Rodents, Encephalitis
equine horses
encephalitis

 

 

 

 

 

  

An understanding of the molecular basis of major physiological processes in

mosquitoes could lead to develop novel strategies in vector management.

2. Vitellogenic cycle of mosquito, Aedes aegypti:

The maintenance and dispersal of mosquito-borne diseases depend upon
the successful reproduction of mosquitoes. The foundation of the reproductive
cycle is vitellogenesis involving massive production of yolk protein precursors
and their accumulation by developing oocytes. The vitellogenesis of mosquitoes
is triggered by a blood meal and, as a consequence, is linked to transmission of
pathogens. Therefore, elucidation of aspects related to vitellogenesis is critical
for the successful development of vector control.

The mosquito fat body is a major organ involved in the synthesis of
macromolecules and the metabolism of nutrients and other compounds. These
functions are hormonally controlled and change successively according to the
demands of the insect at different stages. in adult mosquitoes, the fat body is
located in the head, thorax and predominantly in the abdomen. The primary
important function of the fat body of an oogenic female is the synthesis of the
yolk protein precursor, vitellogenin (V9) and other proteins crucial for oocyte
maturation.

in the female of the yellow fever mosquito, Aedes aegypti, used for this
study, egg maturation proceeds through two developmental phases (Fig 2). The
previtellogenic period begins at the emergence of the adult female and it is
completed within 60-72 hr. During this phase, both fat bodies and ovaries
become competent for subsequent vitellogenesis. The female then enters a

state-of-arrest. The ingestion of vertebrate blood initiates the vitellogenesis.

Fig. 2. Summary of events during the first cycle of egg maturation in the
anautogeneous mosquito, Aedes aegypti.
The first cycle proceeds through two developmental periods. The previtellogenic
period begins at the eclosion of the adult female and it includes preparation
stage and arrest stage. The vitellogenic period is initiated by a blood meal. It is
divided into a synthetic stage (synthesis) and a termination stage (termination).
The relative concentrations of Juvenile hormone lll (JHlll) and 20-
hydroxyecdysone (20-HE) measured in mosquito whole bodies are indicated in
both panels. In fat bodies, the cathepsin D-like activity (CD), the rate of
vitellogenic carboxypeptidase (VCP) synthesis and the rate of vitellogenin (Vg)
synthesis, are relatively presented in the right panel. The big arrow indicates
that the developmental cycle continues from previtellogenic stage to vitellogenic
stage when the mosquito receives a blood meal. (With modifications from

Hagedorn et at, Raikhel, 1992 and Shapiro et at, 1986)

 

 

 

ao> no a> mzeu 21.. =. 1..
.35. coca < .2? Que: 58.3 55‘ 9.5:
8 we 8 8 a 8 «a o o 8 R 3 R o
. ...... r ...... .I ..... k .I .. _ . 1 . _ . u . HILVIN . _ . _ . _ 4

s
It
II
cccccc
I:
was.
IIIII
.....
llllll

 

n .
.9 .-
.r .
a .
a .
I .
r .
r .
r .
a .
e

 

 

 

 

 

'83:..th 2% ii 322 ll _ Sage

 

 

 

 

 

 

 

 

 

 

The vitellogenic period proceeds for 48 hr and it is divided into a synthesis stage
and a termination stage based on the V9 synthesis in the fat body (Raikhel,
1992)

3. Honnonai regulation of fat body activities during vitellogenesis:

In the adult mosquito of A. aegypti, the levels of juvenile hormone lll
(JHlll) produced by the corpora allata (CA) were measured in whole bodies
(Shapiro et al., 1986). The amount of JHlll rises from 0.7 nglg body weight
(gbw) to 7.5 nglgbw in 2 days after eclosion, then declines slowly over 5 days in
non-blood fed mosquitoes. After a blood meal, the levels of JHlll decrease
rapidly to 2.3 nglgbw during the first 3 hr, followed by a slow decrease to the
lowest point (0.4 nglgbw) at 24 hr PBM. From 48 hr PBM on, JHlll starts to rise
again until 96 hr. (Fig 2). Moreover, JHlll is shown to stimulate the increase of
ploidy and the proliferation of ribosomes of fat bodies (trophocytes) during
previtellogenic development (Dittmann et al., 1989; Raikhel and Lea, 1990).

The ovaries of adult vitellogenic mosquitoes synthesize and secrete
ecdysone which is converted to an active hormone, 20-hydroxyecdysone (20-
HE), by fat bodies (Hagedorn, 1989). The levels of ecdysteroid in A. aegypti
were measured in whole body preparations by Hagedorn et al. (1975). in
mosquitoes, the levels of ecdysteroid are low during the first 8-10 hr after a
blood meal with only a small peak at 4 hr PBM. Thereafter, it rises dramatically
to the highest level at 16-20 hr PBM, then declines rapidly to previtellogenic
levels (Fig. 2).

In mosquitoes, A. aegypti, 20—HE was proposed as the primary factor that
initiates the V9 synthesis in fat bodies (Fallon, 1986; Hagedorn and Fallon, 1973;
Hagedorn et al., 1973). Dhadialla and Raikhel (1990) used an in vitro system to
verify that Vg production can be initiated by physiological doses of 20-HE. This

hormone affects the transcription of V9 mRNA and resulted in greater
accumulation of pro-Vg than its mature subunits (Dhadialla and Raikhel, 1990).
The recent finding of elements similar to Drosophila ecdysteroid response
elements in regulatory sequences of V9 genes of Anopheles and Aedes
(Romans and Miller, unpublished; Romans and Hagedorn, unpublished)

strengthens the proposed role of 20-HE in V9 gene expression.

4. Biosynthesis of mosquito vitellogenin in the fat body:

The major protein synthesized by fat bodies for egg development is Vg.
Hagedorn et al.(1973, 1985) demonstrated that Vg synthesis in the mosquito fat
bodies starts at 3-4 hr post blood meal (PBM), reaches the highest level at 24-28
hr then declines to background levels by 36-40 hr PBM (Fig 2).

The stages in the biosynthesis of V9, starting from mRNA to the formation
of the mature secreted protein, have been elucidated (Bose and Raikhel, 1988;
Dhadialla and Raikhel, 1990; Raikhel el al., 1990). The proposed pathway for
V9 biosynthesis in the mosquito fat body is shown in Fig. 3. The precursor of
primary translated Vg has a M = 224,000. It is processed through a series of
complicated modifications, such as glycosylation, phosphorylation, proteolytic
cleavage and sulfation then it is transformed into mature secreted Vg subunits of
200- and 66- kDa.

10

V9 mRNA 6.5 kb

V

 

Pre pro Vg

(ER)

Pro V9

 

 

 

 

224 kDa
1. Translation
2. co-tranelational
glycosylation
v
3. Poet-translational 250 kDa
phoephorylation
4 Cleavage
5. Pboepborylation
l
190 kDa
6. Sulfation |
v
200 kDa
7. Secretion
of Large subunit
nature vg
Fig. 3.

vitellogenin in noequito tat bodies

Raikhel,

1990).

v (Golgi)
66 kDa

 

Small subunit

 

A aohenatio repreantation of bioeyntbetio pathway for

(From Dhadial la and

ll

5. Identification of an additional novel vitellogenic protein:

Analysis of proteins secreted by mosquito fat bodies has demonstrated
that several proteins, in addition to V9, are synthesized and secreted into the
hemolymph during vitellogenesis. One of these proteins is accumulated by
ovaries similar to V9 (Hays and Raikhel, 1990).

This protein has a molecular weight of 53,000 as resolved by SDS-PAGE
under reducing conditions. It is glycosylated, but not detectably phosphorylated
or sulfated. Radioimmunoassay, using antibodies against this 53-kDa protein,
has demonstrated that it is synthesized only by the fat body of vitellogenic
female mosquitoes. Both the synthesis and secretion of this protein could be
stimulated by a physiological dose of 20-HE in the previtellogenic female fat
body cultured in vitro. The properties of this sex-, stage-, and tissue- specific
protein, including its size, kinetics, and regulation of its synthesis drew our

attention to investigate its expressional control during vitellogenesis.

6. Lysosomal activity in the fat body:
The lysosomal system plays an important role in the turnover and

catabolism of various subcellular components (Locke and Collins, 1980;
Brainton, 1981; Farquhar and Palade, 1981; Glaumann et al., 1981). During
metamorphosis and egg maturation of insects, when the fat body switches from
one function to another, lysosomal activity of fat bodies rises sharply and cellular
organelles undergo massive autophagical degradation (Locke and Collins, 1980;
Raikhel and Lea, 1983; Dean etal., 1985; Raikhel, 1986a, 1986b).

Activities of several lysosomal enzymes were found to increase in the
mosquito fat body during the cessation of vitellogenesis (Fig. 2: Raikhel, 1986a;
1986b). Moreover, analysis using video-enhanced fluorescent microscopy and

electron microscopy revealed that lysosomes execute two important functions

12

during the termination of V9 synthesis: (1) interruption of V9 secretion by
degrading the Vg-containing secretory granules, (2) destruction of the
biosynthetic machinery, rough endoplasmic reticulum (RER) and Golgi complex,
for trophocyte remodelling (Raikhel, 1986a; 1986b). Regulation of lysosomal
activity is stimulated by the high titer of V9 concentration in hemolymph at the
termination stage of vitellogenesis (Raikhel, 1986c). in contrast, 20-HE was
shown to initiate lysosomal activity in the fat body of immature insects (Sass and
Kovacs 1980; Locke and Collins, 1980; Tojo et al., 1981). Therefore, 20-HE may
also be involved in the stimulation of lysosomal activity in fat bodies of adult
mosquitoes.

To understand the regulation of lysosomal activity in the female fat body
at molecular level, we chose a lysosomal enzyme, cathepsin D, which is well
characterized in the vertebrate system, for this project. Therefore, we decided to
purify a cathepsin D-like enzyme from mosquitoes then characterized it and
analyzed its expression in the fat body. Moreover, we planed to investigate the
expression and regulation of this enzyme at the molecular level through the

cDNA cloning, RNA analysis and studies of control mechanisms.

13

REFERENCES

Bose, S. G. and Raikhel, A. S. (1988) Biochem. Biophys. Res. Comm. 155, 436-
42.

Brainton, D. F. (1981) J. Cell Biol, 91, 66s-76s.

Dean, R. L., Locke, M. and Collins, D. V. (1985) In: Comprehensive Insect
Physiology Biochemistry and Pharmacology (Eds.) Kerkut, G. A. and
Gilbert, L. l., Pergamon Press, Oxford, New York, Toronto, Sydney, Paris,
Frankfurt. pp. 155-210.

Dhadialla, T. S. and Raikhel, A. S. (1990) J. Biol. Chem. 265, 9924-9933.

Dittmann, F., Kogan, P. H., and Hagedorn, H. H. (1989) Arch. Insect Biochem.
Physiol, 12, 133-143.

Fallon, A. M. (1986) Exp. Cell Res. 166, 535-542.
Farquhar, M. G. and Palade, G. E. (1981) J. Cell Biol, 91, 773-1035.

Glaumann, H., Ericsson, J. L. T., and Marzeila, L. (1981) Int. Rev. Cytol. 73,
149-182.

Hagedorn, H. H., O'Conner, J. D., Fuchs, M. 8., Sage, B., Schlaeger, D. A., and
Bohm, M. K. (1975) Proc. Natl. Acad. Sci. USA 72, 3255-3259.

Hagedorn, H. H. (1985) In: Comprehensive lnseect physiology, Biochemistry,
and Pharmacology, (eds) Kerkut, G. A., Gilbert, L. I, Vol. 8. pp.205-261,
Pergamon Press, Oxford.

Hagedorn, H. H., Fallon, A. M., and Laufer, H. (1973) Develop. Biol, 31, 285-
294.

Hagedorn, H. H. (1989) in: Ecdysone from Chemistry to Mode of Action (ed.)
Koolman, J. Thieme Medical Publishers, Inc., New York, pp. 279-289.

Hagedorn, H. H. and Fallon, A. M. (1973) Nature, 244, 103-105.
Hays, A. R. and Raikhel, A. S. (1990) Roux's Arch. Dev. Biol, 199, 1 14-121.

Kemp, D. J., Coppel, R. L. and Ander, R. F. (1987) Ann. Rev. Microbiol. 41, 181-
208.

l4

Locke, M. and Collins, J. V. (1980) In: Pathobiology of Cell Membrane,
Academic Press, New York, Vol. 2, pp. 223-248.

Raikhel, A. S. (1986a) J. Insect Physiol. 32, 297-604.

Raikhel, A. S. (1986b) Tissue & Cell, 18, 125-142.

Raikhel, A. S. (1986c) in: Host Regulated Developmental Mechanisms in Vector
Arthropods. (Eds) Borovsky and A. Spielman, University of Florida, pp25-
31.

Raikhel, A. S. and Lea, A. O. (1983) Tissue & Cell, 15, 281-300.

Raikhel, A. S. and Lea, A. O. (1990) Gen. Comp. Endocrinol. 77, 423-434.

Raikhel, A. S., Dhadialla, T. 8., Che, W. -L., Hays, A. R. and Koller, C. N. (1990)
In: Molecular Insect Science (eds) Hagedorn, H. H. Hildebrand, J. G.,
Kidwell, M. G., and Law, J. H. Plenum Press, New York, pp. 147-154.

Raikhel, A. S. (1992) Adv. Disease Vector Res., 9, (in press).

Sass, M. and Kovacs, J. (1980) J. Insect Physiol 26, 569-577.

Shapiro, A. B., Wheelock, G. D., Hagedorn, H. H., Baker, F. C., Tsai, L. W., and
Schooley, D. A. (1986) J. Insect Physiol, 32, 867-877.

Tabachnick, W. J. (1991) American Entomologist, 37, 14-24.
Tojo, S., Kiguchi, K and Kimura, S. (1981 ) J. Insect Physiol. 27, 491-

Wite, D. O. and Fenner, F. J. (1986) In: Medical Virology, Academic Press,
Edition 3. Pp. 479-506.

CHAPTER 2*

PURIFICATION AND CHARACTERIZATION OF A LYSOSOMAL
ASPARTIC PROTEASE WITH CATHEPSIN D ACTIVITY FROM
THE MOSQUITO

*Originally published in Insect Biochem.
Reference: Cho, W. -L., Dhadialla, T. S. and Raikhel, A. S. (1991)
insect Biochem. 21, 165-176.

15

16

ABSTRACT

A lysosomal aspartic protease with cathepsin D activity, from the
mosquito, Aedes aegypti, was purified and characterized. its isolation involved
ammonium sulfate (30%-50%) and acid (pH 2.5) precipitations of protein extracts
from whole previtellogenic mosquitoes followed by cation exchange
chromatography. Purity of the enzyme was monitored by SDS-PAGE and silver
staining of the gels. The native molecular weight of the purified enzyme as
determined by polyacrylamide gel electrophoresis under non-denaturing
conditions was 80,000. SDS-PAGE resolved the enzyme into a single
polypeptide with Mr=40,000 suggesting that it exists as a homodimer in its non-
denatured state. The pl of the purified enzyme was 5.4 as determined by
isoelectric focusing gel electrophoresis. The purified enzyme exhibits properties
characteristic of cathepsin D. it utilizes hemoglobin as a substrate and its
activity is completely inhibited by pepstatin-A and 6M urea but not by 10 mM
KCN. Optimal activity of the purified mosquito aspartic protease was obtained at
pH 3.0 and 45°C. With hemoglobin as a substrate the enzyme had an apparent
Km of 4.2 uM. Polyclonal antibodies to the purified enzyme were raised in
rabbits. The specificity of the antibodies to the enzyme was verified by
immunoblot analysis of crude mosquito extracts and the enzyme separated by
both non-denaturing and SDS-PAGE. Density gradient centrifugation of
organelles followed by enzymatic and immunoblot analyses demonstrated the
lysosomal nature of the purified enzyme. The N-terrninal amino acid sequence
of the purified mosquito lysosomal protease (19 amino acids) has 74% identity

with N-terminal amino acid sequence of porcine and human cathepsins D.

17

INTRODUCTION

The lysosomal system plays a significant role in the regulation of
important physiological processes in both immature and adult insects. The fat
body, which is functionally analogous to the vertebrate liver, undergoes dramatic
remodelling associated with metamorphosis and the termination of egg
maturation cycles. It is during these periods that the lysosomal activity rises
sharply and cellular organelles undergo massive autophagical degradation
(Locke, 1980; Raikhel and Lea, 1983; Dean etal, 1985; Raikhel, 1986a; 1986b).
in adult mosquitoes, there is a dramatic increase in lysosomal activity in the fat
body during cessation of synthesis of the yolk protein precursor, vitellogenin
(Raikhel, 1986a). This lysosomal activity is directed towards specific
degradation of organelles involved in biosynthesis and secretion of vitellogenin
(Raikhel, 1986b).

It is important to elucidate the molecular mechanisms regulating this
specific lysosomal activity. Cathepsin D, which is one of the most abundant
lysosomal proteases (Barrett, 1970), is a good candidate for developing a
molecular probe for these studies. Furthermore, the kinetics of the specific
activity of this enzyme in the mosquito fat body were demonstrated and the
factor triggering the rise in cathepsin D activity identified (Raikhel, 1986a;
1986c). Although cathepsins D from various tissues of many vertebrates have
been extensively characterized (Kirschke and Barrett, 1987), little is known about
characteristics of similar enzymes from insect tissues. The only report on the
purification and partial characterization of a cathepsin D-like enzyme was from
pupae of the blowfly, Aldrichina grahami (Kawamura et al., 1987). We report
here the isolation and characterization of a lysosomal aspartic protease from the

mosquito, Aedes aegypti. This lysosomal protease exhibits unique features by

18

being similar in its molecular composition to a vertebrate cathepsin E, but having
enzymatic properties of cathepsin D. Furthermore, its N-terminal amino acid
sequence has 74% identity with N-terminal amino acid sequence of porcine and

human cathepsins D but only limited similarity with cathepsin E.

l9

EXPERIMENTAL PROCEDURES

Materials

Bovine cathepsin D, bovine hemoglobin, pepstatin A, pepstatin A-
agarose, S-sepharose, concanavalin A-sepharose, Percoll and chicken egg
albumin were purchased from Sigma Chemical 00.. SDS-PAGE molecular
weight markers, ampholytes and protein assay dye reagent were from Bio-Rad
Laboratories. Non-denaturing PAGE high molecular weight markers, pl
calibration kit (pl 3-10), PhastGel IEF 3-9 were obtained from Pharmacia.
lmmobilon transfer membrane (PVDF), 0.45-um pore size, were from Millipore.

All other reagents employed were of analytical grade from Sigma and Baker 00..

Insects

Headless, previtellogenic mosquitoes (Aedes aegyptr), frozen in liquid
nitrogen and shipped on dry ice, were kindly provided by Dr A.O. Lea
(Department of Entomology, University of Georgia). The frozen mosquitoes
were stored at 40°C until needed. Vitellogenic mosquitoes were obtained from
our laboratory reared colony. To initiate vitellogenesis, mosquitoes were fed on

rats.

Purification of the Mosquito Aspartic Protease

All the purification steps were carried out at 40C. About 20 g of frozen
mosquitoes were homogenized in 100 ml homogenization buffer ( 0.2 M sodium
acetate, pH 3.5 ,0.15 M NaCI and 0.02% sodium azide). The homogenate was
centrifuged at 27,000 xg for 30 min. The supernatant was collected and
recentrifuged at 60,000 xg for 1 hr. Proteins in the supernatant thus obtained

were precipitated with 30 to 50 percent ammonium sulfate. The precipitated

20

proteins were dissolved in 10 ml of the above homogenization buffer. The
protein solution was dialyzed against 3 x 1 liter of sodium phosphate buffer (0.1
M Na phosphate, pH 2.5, 0.15 M NaCI). Peptides precipitated during dialysis
were removed by centrifugation at 20,000xg. The supernatant was concentrated
by ultrafiitration using Amicon YM-10 membrane and the buffer in the protein
solution exchanged with 0.05 M Na-citrate buffer, pH 4.0, containing 0.02%
NaN3 in the same ultrafiltration cell. About 5-10 ml of the concentrated solution
was centrifuged at 20,000 xg to remove any aggregated proteins before applying
to a S-Sepharose fast flow column (1.0 x 10 cm) equilibrated with 0.05 M Na-
citrate buffer, pH 4.0, 0.02% NaN3. Unbound proteins, which did not contain
detectable cathepsin D activity, were washed from the column with the
equilibration buffer. About 50% of bound cathepsin D activity was eluted with the
equilibration buffer containing 0.1 M NaCI. Fractions with cathepsin D activity
were pooled and concentrated by ultrafiltration using Amicon YM-10 membranes
and Centricon-30 concentrators (Amicon). SDS-PAGE followed by silver
staining showed that at this stage of purification, the concentrated fraction
predominantly consisted of a single polypeptide with only a few minor
contaminating peptides. Further purification of the enzyme was achieved by its
electroelution from polyacrylamide gels after electrophoresis.

Although, finally the above scheme was adopted for the purification of a
mosquito aspartic protease with cathepsin D activity, gel filtration and affinity
chromatographic techniques (below) were also attempted to purify the enzyme.
These steps were extremely useful in establishing the identity of the aspartic

protease on $08 gels as well as revealing some of its characteristics.

Gel permeation chromatography

The supernatant obtained after acid precipitation of the initial mosquito

21

homogenate was dialysed against 0.15 M sodium phosphate buffer, pH 7.0, 0.15
M NaCl. 0.02% NaN3 and concentrated by ultrafiltration. Four milligram protein
was applied onto a Sephadex G-100-120 column (2.5 cm x 50 cm) equilibrated

and eluted with the dialysis buffer.

Pepstatin-afi'inity chromatography

Fractions containing cathepsin D activity, collected after gel permeation,
were pooled and concentrated by ultrafiltration. During concentration, the buffer
was exchanged by washing three times with 50 ml pepstatin A-agarose binding
buffer (0.05 M Na-acetate, pH 3.5, 0.2 M NaCl, 0.02% NaN3) and then applied
as such onto a pepstatin A-agarose column (1 cm x 10 cm) equilibrated with
binding buffer. Unbound proteins were washed out completely with the same
buffer. The bound protein, an aspartic protease with cathepsin D activity, was
eluted with a high pH and high salt buffer (0.05 M Tris-HCI, pH 8.5, 0.6 M NaCI).
A complete elution of bound proteins from the column was achieved by using a
chaotropic buffer (0.12 M Tris-HCI, pH 6.8, 4% 808, 0.15 M 011').

Concanavalin A-Sepharose chromatography

Fractions containing cathepsin D activity, collected after cation-exchange
chromatography (above) were loaded on a Con A-Sepharose column (1.0 cm x
10 cm) equilibrated with Con A buffer (0.2 M Tris-HCI, pH 7.4, 0.5 M NaCl, 10
mM CaClz, 10 mM MgCl2 and 0.02% NaN3). The mosquito aspartic protease
was eluted with 0.1 M methyl or-D-mannopyranoside in Con A buffer.

Assay of cathepsin D activity
Whole mosquitoes or their dissected body parts were homogenized in the

homogenization buffer. The homogenate was centrifuged at 13,600 xg for 20

22

min and the supernatant was assayed. The proteolytic activity of cathepsin D-
like enzymes was measured in an incubation mixture (200 pl) containing 10 pl of
extract or fraction obtained after column chromatography, 140 pl of 0.2 M citrate-
phosphate buffer, pH 2.9 or 0.6 M Na-acetate buffer, pH 2.4, and 50 ul of 2%
(w/v) aqueous hemoglobin solution as substrate resulting in a final pH 3.0. After
1 hour of incubation at 45°C, the reaction was stopped by addition of 40 pl of
18% (w/v) ice-cold trichloroacetic acid (TCA) to precipitate peptides. After 5 min
on ice the precipitated peptides were centrifuged at 12,000 xg for 5 min. The
digested small peptides released in 100 pl of supernatant were mixed with 700 pl
of 3% TCA and the protein content determined by the Bio-Rad protein assay. To
determine the activity specific to cathepsin D, 1 pl of 1.5 mM pepstatin A, an
inhibitor of cathepsin D activity, was added to the control samples prior to the
addition of hemoglobin. One unit of cathepsin D-Iike enzyme is defined as the

amount required to release 1 pg of small peptides from the substrate in 1 h.

Assay for acid phosphatase activiity

Acid phosphatase was assayed as described by Raikhel (1986a). The
reaction mixture, containing 50 ul sodium citrate buffer (pH 4.8), 50 ul 0.45 p-
nitrophenol phosphate, and 20 [.11 sample, were incubated at 37°C for 30 min.
The reaction was terminated by addition of 60 pl of 0.5 N NaOH and then the

absorbence was measured at 420 mn.

Protein determination

The concentration of soluble peptides was measured by a protein-dye
assay (Bradford, 1976) using Bio-Rad reagents and bovine serum albumin
(BSA) as a standard.

23

Poiyacryiamide gel electrophoresis

The purity and the apparent molecular weight of the enzyme during its
fractionation was monitored by SDS-PAGE on 10-15% gradient gels unless
specified otherwise under reducing conditions (Laemmli, 1970) and the peptides
visualized by silver staining according to the supplier's instructions (Sigma). The
native molecular weight of the enzyme was estimated by electrophoresis under

non-denaturing conditions on 5-15% gradient polyacrylamide gels.

Isoeiectric focusing (IEF)
The isoelectric point (pl) of native form of the aspartic protease was
determined by electrophoresis on IEF PhastGel (pH 3-9) using Phast-System

(Pharmacia). Standard pl marker proteins were also separated on each gel.

Two-dirnensional gel electrophoresis

Tube gels for the first dimension (IEF) were prepared by modification of
O'Farrell's (1975) protocol. The gel mix was made by dissolving 5.5 g urea in
1.0 ml 40% acrylamidel5% bisacrylamide, 2.0 ml 10% NP-40, 0.4 ml Ampholyte
5-7, 0.1 ml Ampholyte 3.5-10 and 2.5 ml double distilled water, The anode
solution was 85% phosphoric acid and the cathode solution was 20 mM NaOH.
The gels were pre-run at 200 V for 15 min, at 300 V for 30 min and then at 400 V
for 30 min. Protein samples were then loaded and focusing was carried out at
400 V for 16 hours followed by 1 hour at 800 V. Polypeptides separated in the
first dimension were then separated by SDS-PAGE in the second dimension

using 10-15% gradient slab gels.

Electroelution of the anzyrne from polyacrylamide gels
The 40 kDa enzyme band was cut from $08 gels and electroeluted with

24

lSCO Model 1750 Electrophoretic Concentrator. According to the
manufacturer's instructions, 40 mM Tris-Acetate buffer containing 2 mM EDTA
and 4 mM Tris-Acetate buffer containing 0.2 mM EDTA (pH 8.6) were used for
outer and inner compartments of the concentrator, respectively. The addition of
0.1% lithium dodecyl sulfate to both sample cup and inner cathode compartment

buffers improved the efficiency of elution.

Production of polyclonal antibodies

The enzyme purified by electroelution from polyacrylamide gels was used
to produce polyclonal antibodies in rabbits. Female New Zealand white rabbits
were injected with 40 pg enzyme mixed with Freund's Complete Adjuvant (1:1
vlv). The rabbits were boosted at two weekly intervals with the enzyme
preparation mixed with Freund's incomplete Adjuvant (1 :1 vlv). The rabbits were
bled two weeks after the second booster injection and the sera were collected.
The immunoglobulin (lgG) fraction was precipitated from the sera with 35%
saturated ammonium sulfate solution. The precipitated lgG was solubilized in
and dialysed against 0.2 M Na-phosphate, pH 7.8, 1 mM EDTA, 0.02% NaN3.
and stored at -20 DC in 0.5 mg/ml aliquots. The specificity of the antibodies was

determined by immunoblot analysis.

lmmunoblot

The proteins resolved by SDS-PAGE or non-denaturing PAGE were
transferred to nylon-reinforced nitrocellulose (NitroScreen West, Dupont)
according to Burnette (1981). After the transfer, proteins were visualized by
Ponceau S stain and the molecular weight standards were marked. The blots
were blocked with 5% non-fat dry milk in sodium phosphate buffer (10 mM

25

sodium phosphate, 0.9 % NaCl, pH 7.2) for 1.5 h. Primary antibodies were
applied at a 1:500 dilution in blocking solution containing 0.3% Tween-20. The
primary antibodies were detected using goat anti-rabbit antibodies conjugated
with alkaline phosphatase (1:1,000 dilution). The antigen-antibody complex was
then visualized by a substrate developing cocktail containing 0.005% 5-bromo-4-
chloro-3-indolye phosphate, P-toluidine salt, and 0.01% Nitroblue tetrazolium-

sodium.

Subceliuiar fractionation

The separation of lysosomes was performed according to Alquier et
al(1985). The mosquitoes were homogenized in 10 mM Tris buffer, pH 7.4
containing 0.25 M sucrose (TS buffer) using a glass/Teflon Potter-Elvehjem
homogenizer. The resulting tissue homogenate was centrifuged at 800 xg for 20
min. The supernatant was collected and centrifuged at 26,000 xg for 20 min.
The pellet was resuspended in 0.5 ml TS buffer and then mixed with 10 ml 30%
Percoll in TS buffer. The suspension was centrifuged in a fixed-angled rotor
(Beckman Type 65) at 60,000 xg for 45 min. A test tube prepared as for the
samples but containing density marker beads (Pharmacia) was also centrifuged
simultaneously to obtain a density profile. The gradients were collected into 21
fractions. The fractions corresponding to cathepsin and acid phosphatase
activity were pooled. Although, Percoll in the gradient fractions did not effect the
enzyme activity assays, it was removed by filteration through glass microfibre
filters (GF/D, Whatman) before samples were used for SDS-PAGE.

N-terrninai amino acid sequencing
After S-sepharose chromatography, the partially purified enzyme (50 pM)
was resolved by 10-15% gradient SDS-PAGE and then electroblotted onto

26

PVDF membrane (Matsudaira, 1987). The membrane was stained with 0.1%
Coomassie Blue R-250 in 50% methanol for 5 min, and then destained in 50%
methanol/10% acetic acid for 10 min at room temperature. After rinsing with
deionized H20 for 5-10 min, the membrane was air dried and the area with the
enzyme band was cut out with a clean razor. The band on the cut membrane
was sequenced on an Applied Biosystem model 477A protein sequencer directly
and resulting PTH-amino acids were analyzed on an on-line 120A analyzer
(Macromolecular Facilities, Department of Biochemistry, Michigan State
University). The computer-aided comparison of the mosquito aspartic protease
sequence with known cathepsin D and cathepsin E sequences from other
species was done using a GCG program from the University of Wisconsin
Genetics Computer Group and FASTP program (PIR, NBRF, Georgetown
University, Washington 0.0.). Two libraries were used for this comparison:
GenEMBL protein library containing 28,968 sequences and Amino Acid Bank

Protein library containing 5,415 sequences.

27

RESULTS

Purification of the mosquito aspartic protease with cathepsin D activity

In our initial attempts to purify mosquito cathepsin D, we utilized
ammonium sulfate precipitation of proteins from crude extracts of headless
previtellogenic adult mosquitoes followed by gel permeation chromatography.
Fractions from gel permeation column containing cathepsin activity were pooled
together and applied on pepstatin A-agarose affinity column. The protein eluted
from the affinity column by the high-salt elution buffer, when analyzed by SDS-
PAGE, revealed a single band with an apparent molecular weight of 40,000 (Fig.
2, lane 6). The identity of the 40 kDa peptide as cathepsin was additionally
verified by inhibition of its activity with pepstatin A in the enzymatic assay.
However, the yield by the affinity chromatographic step was extremely low.
Even after pH of the elution buffer was elevated to 9.0, less that 10% of
cathepsin activity eluted from the column and the enzyme rapidly lost its activity.
When the bound protein was denatured and completely eluted from the column
by the chaotropic buffer, the 40 kDa peptide was revealed by SDS-PAGE.
Therefore, due to low yields and increased instability of the enzyme, this
procedure was not used. Similarly, peptstatin A-affinity chromatography could
not be used for purification of a cathepsin D-like enzyme from another insect,
blowfly, Aidricina grahami (Kawamura et al., 1987). in contrast, this method was
effectively used to purify mammalian cathepsins (Afting and Becker, 1981;
Takahashi and Tang, 1981).

Finally, cation-exchange chromatography was used for routine purification

of the mosquito enzyme after two precipitation steps of the crude extracts.

Table 1. Purification of mosquito lysosomal aspartic protease.

 

Procedures

    
 

 

1 . Crude extract

__T___

(me)

1 Protein

Activity
(Units)

28

 

Activity

 

(Units/poi

     

Yield

(96)

     

 

2. (NH412804 ppt

   

69.680

0.333

80

 

3. Acid precipitation

65.325

2.56

75

 

 

4. Ultrafiltration

 

34,545

8.7

40

 

  

5. S-sepharose

 

6,000

 

 

461.0

m

 

6.9

Purification

 

    
     
 

      
 

 

     

29

Fig. 1: Elution profile of mosquito proteins on S-sepharose fast fiow column
(1.0 cm x 10 cm).

Sample, proteins obtained after acid precipitation step (Fig. 2, lane 3). Arrows
indicate elution with, (1) start buffer, 0.05 M Na-citrate, pH 4.0, 0.02% NaN3 and
(2) elution buffer, start buffer containing 0.1 M NaCI. Flow rate, 25 ml/h.

Fraction size, 1 ml. A bar denotes the fractions pooled for further analyses.

0.5

30

(mu 969) eoueqlosqv

-0.1

‘3!
o
l

 

Hr" CD-Activity

 

 

 

3'0 40
‘ Fraction Number

2'0

I
10

 

 

——— (mu 093) eoueqaosqv aanelag

31

Fig. 2: SDS-PAGE (10-15%) of mosquito lysosomal aspartic protease with
cathepsin D activity after various purification steps as revealed with silver
staining.

Lanes: (1) soluble proteins of headless previtellogenic mosquitoes; (2) 30-50%
ammonium sulfate precipitated proteins from step 1; (3) supernatant after acid
precipitation at pH 2.5 of proteins from step 2; (4) pooled fractions numbers 10-
45 containing cathepsin D activity as shown by dotted profile in Fig. 1; (5) pooled
fractions indicated by a bar in Fig. 1 after S-sepharose chromatography; and (6)
purified enzyme after affinity chromatography on pepstatin A-sepharose. The
molecular weights on the right, in order of decreasing Mr, are of phosphorylase

b, BSA, ovalbumin, carbonic anhydrase, soybean trypsin inhibitor and lysozyme.

32

123456

‘1
Ii

(kDa)

n97-4
'66.2

1.45

'31
~21.5

 

h'14|.4

 

33

Results obtained after such a purification are summarized in Table 1 and Fig. 2.
Precipitation of the crude extract with 30-50% ammonium sulfate followed by
lowering the pH of the resuspended proteins to 2.5 resulted in a 23-fold increase
in the purity of the enzyme. A further 4000-fold purification of the enzyme with a
7% yield of the original extract was achieved by cation-exchange
chromatography of the acid precipitated protein extract on a S-sepharose
column. A typical elution profile of proteins from the S-sepharose column is
shown in Fig. 1. The fractions containing peak cathepsin activity (indicated by a
bar in Fig. 1) were highly enriched for the enzyme as judged by electrophoresis
(Fig. 2).

Electrophoretic analysis of the purified mosquito enzyme

The purified enzyme was analyzed by polyacrylamide gel electrophoresis
under both denaturing and non-denaturing conditions. In either case, the
enzyme resolved into a single protein band. By SDS-PAGE the mosquito
enzyme had an apparent molecular weight of 40,000 (Fig. 2). Its native
molecular weight was estimated to be 80,000 (Fig. 3). To confirm that the 80
kDa protein band had cathepsin D activity, it was cut from the gel and the gel
piece homogenized in buffer was used for cathepsin D enzymatic assay. The
supernatant separated from the homogenized gel showed cathepsin D activity
(results not shown).

Pooled fractions obtained after cation-exchange chromatography,
containing partially purified enzyme, were also analyzed by 2-dimensionai gel
electrophoresis. After electrophoresis in the second dimension under reducing

conditions a predominant spot with Mr=40,000 was revealed (Fig. 4).

34

Fig. 3: Non-denaturing polyacrylamide gel electrophoresis of purified
enzyme stained with silver.
Lanes: (1) mosquito aspartic protease; (MW) molecular weight markers (from

top to bottom), thyroglobulin, ferritin, catalase, aldolase, bovine serum albumin

and chicken egg albumin.

 

35

1 MW (kDa)
.- «A _- 669
e-e - 440
i I - 232

H P140

 

 

 

36

Fig. 4: Two dimensional gel electrophoresis of proteins in fractions pooled
after S-sepharose chromatography lane 4 in Fig. 2).

First dimension (1-D): IEF separation with ampholyte, pH 5-8; Second dimension
(2-D): 10-15% gradient SDS-PAGE. The arrow shows the mosquito enzyme.

The molecular weights on the right are the same as in Fig. 2.

37

vé—I

m.—NI

nvl

N601
vKol

Goo:

 

4m)

8 s

32

 

min

L).

It' I '1')

38

Fig. 5: Determination of lsoelectric point (pl) of mosquito aspartic protease.

Lanes: (A) purified enzyme; (B) pl markers on the right and on the graph: (1)
trypsinogen, 9.3; (2) lentil lectin-basic band, 8.65; (3) lentil lectin-middle band,
8.45; (4) lentil lectin-acidic band, 8.15; (5) myoglobin-basic band, 7.35; (6)
myoglobin-acidic band, 6.85; (7) human carbonic anhydrase B, 6.55; (8) bovine
carbonic anhydrase B, 5.85; (9) beta-Iactoglobulin A, 5.2; (10) soybean trypsin
inhibitor, 4.55; (11) amyloglucosidase, 3.5. The pl of the mosquito enzyme, 5.4,

was determined from the plot on the left.

39

A83 ovofieo Seam conga:—
ov on

 

 

 

 

s._...._....._..
Pwl . o.nv..o+xoonno_.olu> a
91.. I.
all -

I“ In
or. III
10
w" -
, Th
.- v
to
«I C
mhu ~ -o
Fl
m <

[(1

40

isoelectric point of the purified enzyme

Analytical IEF of the enzyme on pH 3-9 Phast IEF gels showed that the
native enzyme has a pl of 5.4 (Figure 5). A similar pl value for the partially
purified enzyme was obtained by two dimensional gel electrophoresis when it

was separated in the first dimension (IEF) in the presence of 6 M urea (Fig. 4).

Evidence for glycosylation of the purified enzyme

Fractions obtained after cation-exchange chromatography (lane 4 in Fig.
2) and enriched for cathepsin D activity were applied on Con A-Sepharose
column. The absence of cathepsin activity in the unbound fraction and its
presence in fractions eluted with Con-A buffer containing 0.1 M methyl or-D-
mannopyranoside indicated that the enzyme is a glycoprotein with mannose on

the glycosyl part.

Enzymatic properties of the purified enzyme

The purified enzyme displayed a pH optimum close to 3.0 when assayed
with hemoglobin as a substrate (Fig. 6). At 4°C it maintained its activity between
pH 2.5 - 4.0, but lost its activity and degraded rapidly at pH higher than 7.

The optimum temperature for the hydrolysis of hemoglobin by the purified
enzyme was 45°C as shown in Fig. 7. The enzyme was inactivated at
temperatures above 60°C.

The purified mosquito enzyme was further characterized by its activity
towards hemoglobin as a substrate. The activity of the purified enzyme (0.5 pg)
was assayed with concentrations of hemoglobin increasing from 0 pM to 77.5 pM
(based on the molecular weight of hemoglobin to be 64,500). The Michaelis
Constant (Km), determined by the Michaelis-Menten plot as well as Lineweaver-

Burk plot, was calculated to be 4.2 pM (Fig. 8).

" IL

,a .3”! :11.)

I;

f. a.

Q\ . \
1':

1.

I r

e.

‘Q
30

41

Fig. 6: Effect of pH on the activity of mosquito aspartic protease.

The pH dependence of cathepsin D activity was determined by using stock
solutions of 0.6 M phosphoric acid, 0.6 M acetic acid, 0.6 M KH2PO4 and 0.6M
Tris-base to prepare buffers with different pH (1.5-9.0). The assay mixture was
140 III of desired pH buffer, 10 III (0.5 ug) of purified enzyme and 50 ul of 2%

hemoglobin. Each point is a mean 1 S.E.M. of 3 assays.

42

 

 

 

 

0.5

r n I r
x: m.

o
(mu 969) some

I 1
N

O
QJOSQV

0.1 -

 

0.0

6.0 7.0 8.0 9.0

pH value

3.0 4.0 5.0

2.0

43

Fig. 7. Effect of temperature on the activity of mosquito aspartic protease.

The reaction mixtures were the same as described in figure 3, except that only
0.6 M Na-acetate buffer, pH 2.4, was used in all reactions. Incubation of the
reaction mixtures was for 1 hr at the indicated temperatures. Each point is a

mean 1 S.E.M of 3 determinations.

44

o:

 

oofi _

om

_

om

_

—

oo .ehfieuomaoe

on

_

n

on

_

4‘

ohv
_

om.

_

cm

I'IT

 

 

ed

«to

45

Fig. 8. Michaelis-Menten and Linaweaver-Burk plots for mosquito aspartic
protease with increasing concentrations of hemoglobin.

The reaction mixture contained 10 ul (0.5 ug) of the purified mosquito enzyme,
140 III of 0.6 M acetate buffer (pH 2.4) and 50 ul of hemoglobin (0-77.5 uM).
The incubation time was 10 min. Inset: Double-reciprocal (Lineweaver-Burk) plot

of the kinetics of mosquito aspartic protease using the Michaelis-Menten data.

M

A23 Encamoaom

 

 

 

won

 

was

(20!!) MI

  

H

    

on cm on 3. on cm ca . o
_ ___..__.L___...L_r._L_._L._.._.__.__C
rose a)
can an; on col coal onml
_ _ _ _ p p _ _ _ _ _ [0
nmmm€¢ + xmm—mhwd I >1
I2 -
tom

Io.“

In:

low

 

 

mm

47

The activity of the mosquito enzyme to hydrolyze hemoglobin was also
tested in the presence of various inhibitors. In contrast to pepstatin A (7.5 pM)
and urea (6 M) which completely inhibited the enzyme activity, KCN (10 mM) did

not have any inhibitory effect.

Production and characterization of polyclonal antibodies against the mosquito
aspartic protease

Polyclonal antibodies against the gel-purified mosquito aspartic protease
were produced in rabbits. The specificity of the produced antibodies was
established by immunoblot analysis. Both crude mosquito extracts and purified
aspartic protease were resolved by SDS-PAGE and non-denaturing PAGE and
transferred to nitrocellulose membranes. These protein blots were probed with
antibodies to the enzyme. The antibodies reacted with 80 kDa and 40 kda
bands in both samples separated on non-denaturing and SOS-polyacrylamide
gels, respectively (Fig. 9). The molecular weights of these protein bands
correspond to the native and the reduced state of the aspartic protease (Figs. 3
and 4).

Tissue and subcellular distribution of the aspartic protease

Female mosquitoes 40 h post blood feeding were used to analyze the
distribution of the enzyme in different tissues. Although cathepsin D activity was
present in all assayed tissues (whole body, head and thorax, isolated abdomens
free of mid-gut and ovaries, gut and ovaries) it was the highest in isolated
abdomens, consisting largely of adhering fat body. When the same tissue
extracts were analyzed on immunoblots using antibodies against the purified
aspartic protease, the enzyme was detected in all of them. However, it was

clearly the most abundant in the isolated abdomen extracts (not shown).

aw

48

Fig. 9. Immunoblot analysis of polyclonal antibodies against the mosquito
aspartic protease.

A. Non-denaturing PAGE; B. SDS-PAGE under reducing conditions. Lane 1,
crude tissue extract after acid precipitation; 2, purified enzyme. Molecular weight
standards on the right are as in Figs. 1 and 2. Eighty and 40 refer to the native

and sub-unit molecular weight (kDa) of the enzyme, respectively.

80-

1

r-45

 

49

SDS

40— ‘-

(kDa)

'-914

-66.2

-21.5

—14.4

 

 

50

Fig. 10. Protein and enzyme distributions after isopycnic centrifugation on
30% Percoll gradients of the mosquito organelles.

A, Density and protein profiles; B and C, acid phosphatase (-A-) and cathepsin D
(---) activity distribution before and after sonication of organelle preparation,

respectively.

51

1.16

ello QE\uv hfimdoa

'—1.14
—112
-110
.08

06

04
”-102
100

all... Ana 03; occasions:
w m.

0.

-o.25
—o.2o
—o.1$
—0.10

»
Av

«lie AS: ONE oodenuomn:

w. m w.
o o 0
_ _ _ _ . _ u

 

 

lI'II'I'I

 

 

 

 

1.4

____________

2 0. a 6 4 2

I. I e o. o. 0.
blue A~1\u3 530.5

 

 

 

 

0.00

.|.. Anna .33 0033.32:

._ _ _ _ . _ _ _

. m m. m. m
0 0 O 0

I. As: many ocean—.32;

_
o
1.
0

 

0.00

Fraction Number

52

in order to locate the subcellular source of cathepsin D activity in our
purified preparations, we centrifuged TS buffer extracted mosquito homogenates
(Materials and Methods) on isopycnic Percoll gradients. The distribution of
protein, acid phosphatase (a lysosomal marker) and cathepsin D activities are
reported in Fig. 10. The protein profile exhibited a large peak at low density.
Both acid phosphatase and cathepsin D activities banded at a density higher
than where the proteins floated on the gradients. The density (1.058 glml) at
which the two enzymes peaked was reproducible in several experiments.
However, in some experiments, in addition to this stable peak of enzyme activity
another peak with acid phosphatase and cathepsin D activities was obtained
which banded at the same position where peak protein concentration was (not
shown). We argued that this second peak of enzymatic activity at lower density
resulted from enzymes released from damaged lysosomes. In order to verify
this, the organelle preparation was sonicated prior to Percoll gradient
centrifugation. Analysis of such Percoll gradients showed a considerable
reduction of the high density acid phosphatase/cathepsin D activity peak and an
increase in the lower density enzymatic peak (Fig. 10). Based on these
experiments, we conclude that the higher density enzyme activity peak identified
by its content of acid hydrolases and its sensitivity to sonication represents
lysosomes.

Next, the fractions from Percoll gradient were subjected to immunoblot
analysis using antibodies against the purified aspartic protease. The enzyme

was only detected in fractions which exhibited cathepsin D activity (Fig. 11).

Analysis of N-tenninai amino acid sequence
The N-terminal amino acid sequence of the electroblotted mosquito

enzyme was determined. The 19 N-terminal amino acid residues of the mosquito

53

enzyme were compared with sequences from two computer protein libraries with
GCG and FASTP programs. The results show that the mosquito sequence has
the highest functional similarity with porcine and human cathepsin D.
Comparison of the aligned amino acid sequences revealed a 74% identity
between N-terminal amino acid sequence of the mosquito aspartic protease and
porcine and human cathepsins D (Fig. 12A), and a 53% identity between N-
terminal amino acid sequence of the mosquito enzyme and an internal sequence

of human cathepsin E (Fig. 128).

Fig. 11. Immunoblot analysis of fractions from Percoll gradients of the
mosquito organelles.

Samples of fractions from Percoll gradient (Fig. 10) were separated on 12%
polyacrylamide-SDS gels, transferred onto nitrocellulose membrane and probed
with antibodies to the purified mosquito aspartic protease. Lanes 1, 2 and 3
correspond to fractions 5, 15 and 21, respectively, in Fig. 10 A, B. Lane 4,
sample used for isopycnic centrifugation. Molecular weight standards are as in

Fig. 2.

 

 

55

" (kDa)

-97.4
-66.2

-45

. “ ~31

 

l-21.5

 

 

 

55

(kDa)

-97.4
-66.2

-45

-21.5

 

 

 

56

Fig. 12. Comparison of N-tenninai amino acid sequence of mosquito aspartic
protease with sequences of vertebrate cathepsin DUI) and cathepsin E (B).

The marks, (:) and (.), between aligned sequences indicate identical and
functionally related residues, respectively. Amino acids are represented by

standard one-letter abbreviations.

57

IIZOQAMOHmHBONHNIOAMZHAmflKdmOQSmUm

IIIIIIIIIIH40>HO¢DA>2mAmmm>m0

IIUOQmBUHUHHOHMOGGXHZKflbflmHQOMB><Q><Om>III

lIUOQmBUHUHflOHHOQQIHZKflbflmHWO

IIIIIIIIII H<0>NO<OA>2mAANm>m0

um awuaununo afiuam

“eaeeuoum mad onwaerz
2:

no cwumununu sees:
”a cannonuno ucwuuom
.eaeeuouum and ou«:erz
3:

58

DISCUSSION

In this report, we present data on the purification and characterization of
an aspartic lysosomal protease from the mosquito, Aedes aegypti. In general, its
enzymatic characteristics are similar to those reported for cathepsins D from
mammalian tissues (Takahashi and Tang, 1981; Kirschke and Barrett, 1987) and
a cathepsin D-like enzyme from the blowfly, A. grahami (Kawamura et al., 1987).
The mosquito enzyme hydrolyzed hemoglobin as a substrate with pH and
temperature optima of 3.0 and 450C, respectively. Yamamoto et al. (1978)
established differences between the activities of rat spleen cathepsin D and E
based on inhibition produced by pepstatin A, urea or KCN. We found that in the
case of the mosquito enzyme, complete inhibition with pepstatin-A and urea and
not with KCN indicated that our purified enzyme was cathepsin D. Furthermore,
and more importantly, a comparison of 19 residues of the N-terminal sequence
of mosquito aspartic protease revealed a high degree of sequence and
functional similarity with porcine and human cathepsins D. Similar comparison
with the recently published amino acid sequences of cathepsins E (Azuma et al.,
1989; Yonezawa et al., 1990) showed lesser degree of similarity. To our
knowledge this is the first time that the N-tenninal sequence of an insect aspartic
protease has been obtained.

The mosquito aspartic protease had an apparent molecular weight of
about 80,000 as determined by PAGE under non-reducing conditions and
consisted of two subunits identical in molecular weight (Mr=40,000). For the
blowfly cathepsin D-like enzyme, an apparent molecular weight of 41,000 was
estimated by SDS-PAGE, but the native molecular weight was not determined
(Kawamura et al., 1987). Of the two mammalian aspartic proteases termed

cathepsin D and E, the former has a molecular weight of about 40,000-50,000

I"

59

and exists as a monomer in its native state (Barrett, 1977). Whereas, cathepsin
E has a molecular weight of about 86,000-100,000 in its native form and consists
of two identical subunits (Yonezawa et al., 1987; Muto et al., 1987).

The homogeneity of the purified mosquito aspartic protease after cation-
chromatographic step was confirmed by Isoelectric-focusing and electrophoresis
under both denaturing and non-denaturing conditions. The existence of
isoenzymes for the blowfly (Kawamura et al., 1987) and mammalian (Barrett,
1970; Smith and Turk, 1974; Whitaker and Sayer, 1979; Yamamoto et al., 1979)
cathepsins D was demonstrated based on pi differences between isoenzymes.
However, the mosquito aspartic protease in our purified preparations displayed a
single pl of 5.4, indicating the absence of isoenzymes. This conclusion was also
supported by kinetic data analysis (Bell and Bell, 1988).

The purified mosquito aspartic protease from the mosquito was shown to
be a lysosomal enzyme. it has ubiquitous distribution in mosquito tissues. The
enzyme was the most abundant in the fat body which is characterized by a high
lysosomal activity (Raikhel, 1986a). Subcellular fractionation using Percoll
gradients demonstrated the lysosomal nature of this enzyme. It was co-locaiized
with the activity peak of lysosomal acid hydrolysis, acid phosphatase and
cathepsin D.

The aspartic protease that we have purified from the mosquito is,
therefore, characterized by unique features. Being a dimer, consisting of two
identical subunits, this enzyme is similar to a mammalian cathepsin E which is an
intracellular but not a lysosomal enzyme associated with lymphoid tissues
(Yonezawa et al., 1988). The mosquito enzyme is, however, a lysosomal
enzyme and has enzyme characteristics similar to those of a lysosomal
cathepsin D. Furthermore, the N-terminal of the mosquito protease has a much

higher degree of similarity with cathepsin D than cathepsin E.

60

ACKNOWLEDGEMENTS

We thank Dr. Arden O. Lea for providing us with large quantities of
mosquitoes for enzyme purification, Mr. Alan R. Hays for the assistance and
advice on the production of antibodies, and Dr. Charles Sweeley for his critical
reading of the manuscript. This research was supported by grants Al-24716

from the National Institutes of Health and MSU Biotechnology Center to ASR

61

REFERENCES

Afting E. G. and Becker M. L. (1981) Biochem. J. 197, 519-522.

Alquier C., Guenin P., Munari-Silem, Y., Audebert C. and Rousset B. (1985).
Biochem. J. 232, 529-537.

Azuma, T., Pals, G., Mohandas, T. K, Couvreur J. M., and Taggart, R. T. J.
Biol. Chem. 264, 16748-16753.

Barrett A. J. (1970) Biochem. J. 117, 601-607.

Barrett, A. J. (1977) In: Proteinases in Mammalian Cells and Tissue (ed. A. J.
Barrett), pp. 209-248, North-Holland Publishing 00., Amsterdam.

Bell J. E. and Bell E. T. (1988) In: Proteins and Enzymes. pp. 370-376.
Englewood Cliffs, New Jersey.

Bradford M. M. (1976) Analyt. Biochem. 72, 248-254.

Burnette, W. N. (1981). Anai.Biochem.112,195-203.

Dean R. L., Locke M. and Collins J. V. (1985) In: Comprehensive insect
Physiology Biochemistry and Phannacoiogy. (Eds. Kerdut G. A. and
Gilbert L. l.), Vol. 3, pp.155-210. Pergamon Press, Oxford, New York.

Kawamura M., Wadano A. and Miura K (1987) insect Biochem. 17:77-83.

Kirschke H. and Barrett A. J. (1987) in: Their Role in Protein Breakdown. (Eds.
Glaumann H. and Ballard F. J.), pp. 222-238, Academic Press, New York.

Laemmli U. K. (1970) Nature 227, 680-685.

Locke M. (1980) The cell biology of fat body development. In: insect Biology in
the Future (Eds. Locke M. and Smith D. 8.), pp. 227-252 Academic Press.
New York.

Matsudaira P .(1987) J. Biol. Chem. 262, 10035-10038.

Muto M, Yamamoto M., and Tani S. (1987) J. Biochem. 101, 1069-1075.

O'Farrel P. H. (1975) J. Biol. Chem. 250, 4007-4021.

Raikhel A. S. (1986a) J. Insect Physiol. 32, 597-604.

62

Raikhel A. S. (1986b) Tiss. Ceil18, 125-142.

Raikhel A. S. (1986c) in: Host regulated Developmental Mechanism in Vector
Arthropods. Proceedings of the Vero Beach Symposium (Eds. Spielman,
A. and Borovsky, 0.), pp.25—31, University of Florida.

Raikhel A. S. and Lea A. O. (1983) Tiss. Celi15, 281-300.

Smith R. and Turk V. (1974) Eur. J. Biochem. 48, 245-254.

Takahashi T. and Tang J. (1981) Methods Enzymoi. Vol 80, 565-581.

Whitaker J. N. and Seyer J. M. (1979) J. Neurochem. 32, 325-333.

Yamamoto K., Katsuda N. Himeno M. and Kato K (1979) Eur. J. Biochem. 95,
459-467.

Yamamoto K., Katsuda N., and Kato K (1978) Eur. J. Biochem. 92, 499-508.

Yonezawa S., Tanaka T., Muto N. and Tani S. (1987) Biochem. Biophys. Res.
Commun. 144, 1251-1256.

Yonezawa, S., Fujii K, Maejima Y., Tamoto K, Mori Y., and Muto N. (1988).
Arch. Biochem. Biophy. 267, 176-183.

Yonezawa, S., Takahashi T., lchinose M., Miki K, Tanaka J., and Gasa S.
(1990). Biochem. Biophys. Res. Commun. 166, 1 032-1 038.

CHAPTER 3‘

CLONING OF CDNA FOR MOSQUITO LYSOSOMAL ASPARTIC
PROTEASE

SEQUENCE ANALYSIS OF INSECT LYSOSOMAL ENZYME SIMILAR TO

CATHEPSINS D AND E

* This manuscript is submitted to The Journal of Biological Chemistry.

63

ABSTRACT

A CDNA coding for the lysosomal aspartic protease, from the mosquito
(mLAP). was cloned and sequenced. The mLAP cDNA is 1,420 base pairs long
with an open reading frame of 387 amino acids. The deduced amino acid
sequence contains a signal pre-propeptide sequence of 18 amino acids followed
by 369 amino acids with a 35 amino acid. putative pro-enzyme domain in the N-
terminal. The amino acid sequence of mLAP is 92% and 81% similar to human
cathepsin D and cathepsin E, respectively. Typical cleavage sites for cathepsin
D processing into light and heavy chains are lacking in mLAP. A single
glycosylation site occurs in the mLAP sequence, at a position corresponding to
the first glycosylation site of cathepsins D. The mLAP sequence shares putative
phosphorylation determinants, which in cathepsins D are linked to the formation
of mannose-6-phosphate.

In the mosquito fat body, lysosomal enzymes specifically degrade
organelles involved in the biosynthesis and secretion of vitellogenin (V9). The
mLAP mRNA accumulates to its highest level 24 hr after initiation of V9
synthesis and 12 hr before the peak of mLAP protein accumulation and its
enzymatic activity. Translational regulation of mLAP mRNA may occur. The 5'-
untranslated region of mLAP mRNA is similar to elements conferring negative

translational control by steroids.

65

INTRODUCTION

Although lysosomes are ubiquitous cellular organelles, in several tissues
they participate in various developmental and physiological processes (Smith
and Farquhar, 1966; Glaumann etal, 1981; Orci et al., 1984). In the fat body of
insects, the lysosomal system is involved in cellular remodelling, which is
associated with metamorphosis and termination of egg maturation cycles (Locke,
1980; Dean et al., 1985; Raikhel, 1986a; 1986b; 1992). The fat body of insects
is a functional analogue to the vertebrate liver. It is responsible for metabolism
and storage of carbohydrates, lipids and proteins, and synthesis and regulation
of hemolymph proteins and sugars (Wyatt, 1980). In oogenic females an
important function of the fat body is synthesis of yolk protein precursors, mainly
vitellogenin (Vg) (Kunkel and Nordin, 1985; Raikhel and Dhadialla, 1992). In the
mosquito fat body, the rise in specific activities of several lysosomal enzymes
coincides with a dramatic decline in V9 synthesis (Raikhel 1986a). Fluorescent
and electron microscopic analysis revealed two important roles of lysosomes
during the lamination of V9 production: (1) interruption of V9 secretion by
degrading the Vg-containing secretory granules; (2) destruction of the
biosynthetic machinery, RER and Golgi complexes, and subsequent remodelling
of fat body trophocytes (Raikhel, 1986b).

Due to the medical importance of mosquitoes as vectors of numerous
devastating human diseases, the elucidation of the biochemical and molecular
basis of vitellogenesis and egg maturation is critical to the successful
development of novel strategies in vector management. Mechanisms for altering
the lysosomal activity in the mosquito fat body to cause premature interruption of
vitellogenesis and egg maturation in mosquitoes may emerge.

In mammalian cells, targeting of newly synthesized enzymes to the

lysosome depends on phosphorylated mannose oligosaccharides (Kornfeld and
Mellman, 1989). In yeast and plants, the signal targeting enzymes to the
vacuole, the equivalent of lysosomes, is contained in polypeptide domains
(Chrispeels and Raikhel, 1992; Bednarek and Raikhel, 1992). Nothing is known
about the mechanism of lysosomal targeting in insects. The mechanisms may
be different in insects because of inability to modify high-mannose sugar
residues to complex sugars (Nordin et al., 1984; Osir et al., 1986; Nagao et al.,
1987)

A lysosomal aspartic protease from the mosquito Aedes aegypti was
purified and characterized (Cho et al., 1991a). The structure of this enzyme is
similar to mammalian cathepsin E, in native molecular weight of 80,000 with two
identical 40-kDa subunits. The purified enzyme, however, exhibits properties
characteristic of cathepsin D. This mosquito aspartic protease does not have
isozymes, and its pl is 5.4. Density gradient centrifugation of the organelles,
followed by enzymatic and immunoblot analyses, localized the enzyme in
lysosomes.

Here, we report the cloning and analysis of the cDNA coding for mLAP.
This is the first report of the sequence for an insect lysosomal protease which
provides insight into the evolution of aspartic proteases. The. temporal
expression of mLAP mRNA in mosquito fat bodies during the V9 synthetic cycle
shows the peak of mLAP mRNA occurs 12 hr before the peak of mLAP protein
and enzymatic activity. This suggests possible translational regulation of mLAP
in the fat body. The 5'-untranslated region of mLAP mRNA has sequences
which are similar to those implicated into negative translational control by

steroids.

67

EXPERIMENTAL PROCEDURES

Animals

Mosquitoes, Aedes aegypti, were reared as described by Hays and
Raikhel (1990). Larvae were fed on a standard diet (Lea, 1964). Vitellogenesis
in adult females 3-5 days after eclosion was initiated by feeding them on rats.
in the anautogenous mosquito, A. aegypti, vitellogenesis proceeds through two
distinct developmental stages. As a result of the previtellogenic stage, which
begins at adult eclosion and is completed within 72 hr, the tissues participating in
vitellogenesis, the fat body and the ovaries, become competent: the fat body for
the synthesis of yolk protein precursors and the oocytes for their internalization.
The ingestion of vertebrate blood triggers a cascade of hormonal signals that
culminate in the massive synthesis of V9 and other yolk protein precursors in the
fat body and their accumulation by the oocytes. The peak of these activities
occurs at 24 hr after the initiation of vitellogenesis by a blood meal. Later, these
activities dramatically decline. In the fat body this decline in the production of V9
and other yolk protein precursors coincides with the rising activity of lysosomal

enzymes (Raikhel, 1992).

Materials

Molecular weight markers for SDS-PAGE were purchased from Bio-Rad
Laboratories. RNA ladder ( 0.24 - 9.5 kb) was from Bethesda Research
Laboratories (BRL). Modified T7 DNA polymerase, Sequenase, was supplied by
United State Biochemical Co. Horseradish peroxidase-conjugated affinity-
purified rabbit anti-mouse lgG and goat anti-rabbit lgG were from Organon
Teknika Corp. The enhanced chemiluminescence (ECL) Western blotting

detection system was purchased from Amersham Corp. Perkin Elmer Cetus

68

was the source of the reagents for the PCR work. [32deATP ( 3,000 Cilmmol)
for labeling and [35S]dATP (1,000-1,500 Cilmmol) for sequencing were from
New England Nuclear (DuPont). All other reagents used were of analytical

grade from Sigma Chemical Co. and Baker Co.

Poiyrnerase Chain Reaction

The N-terminal amino acid sequence of mLAP (Cho et al., 1991a) was
used to design the 25 bp sense primer (primer-1). The 28 bp anti-sense primer
(primer-2) was synthesized based on the amino acid sequence of the first
catalytic center conserved among aspartic proteases (Azuma ei al., 1989). Both
primers include a Xbal restriction site at their 5' ends (Fig. 1). The primers were
synthesized at the Macromolecular Structure, Sequencing, and Synthesis
Facility of Michigan State University.

DNA amplification was carried out on a Perkin Elmer thermal cycler with
mosquito cDNA templates. The latter was obtained from 20 ug of total RNA
isolated from the fat bodies of female mosquitoes 36 hr post-blood meal. All
steps were performed according to the manufacturer's protocol (Cetus Co). The
first five cycles of the primary amplification were carried out at 94°C for 40 sec,
at 48°C for 2 min, and at 72°C for 3 min; then the reaction was shifted to the
second set of conditions for 20 additional cycles at 94°C for 40 sec, at 53°C for
2 min, and at 72°C for 3 min. The final polymerase extension step was carried
out at 72°C for 10 min. A 96 base pair fragment was predominantly amplified
during the first PCR cycle (Fig. 2A). This fragment was gel-purified from 10%
acrylamide gel and was used for the secondary amplification. For the secondary
am plification, twenty-five cycles were carried out at 94°C for 40 sec, at 60°C for

2 min, and at 72°C for 3 min. The conditions for the final extension reaction

69

 

 

 

 

 

(A)
1 10
GPVPEPLSNYLDAQYYGAI EQLGSSNLWVP ---------
Primer-1 Primer-2
(3)
Primers Primer Sequence Degeneracy
Primer-1 5' CGTQTAGAGATGCNCAATATTATGGG 3' 64
Xbal C G C C
Primer-2 5' AT'LCAIQATTNCTNCTNCCNGTATCAAA 3' 16,384
Xbal GA GA G G

 

 

 

Fig. 1. Design of primers for amplification of mLAP cDNA fragment by the
polymerase chain reaction. (A) The primer-1 is based on the underlined
stretch of amino acids from the N-terminal sequence of mLAP (Cho et al.,
1991a). The primer-2 is based on the underlined sequence from the first
catalytic center of human cathepsin D which is conserved among aspartic
proteases (Faust et al., 1985). (B) Nucleotide sequences of primer-1 and
primer-2. Sequences which are underlined and marked by Xbal indicate the
sites for this restriction enzyme. Numbers indicate the level of degeneracy for

each primer.

 

70

Fig. 2. Results of the PCR amplification and isolation of the mLAP cDNA
fragment. (A) The first PCR amplification using: lane 1, cDNA from vitellogenic
fat bodies and both primers; lane 2, the cDNA and the primer-1; lane 3, the
cDNA and the primer-2; lane 4, both primers without the cDNA; lane 5, the cDNA
without primers. A predominantly amplified band of 96 base pairs is marked by
this number on the left. (B) The secondary PCR amplification utilizing the 96 bp
cDNA fragment and both primers is shown in lane 1. Lane 2 shows a 86 bp
cDNA fragment which was obtained after subcloning in pUC 119 and releasing
by the restriction enzyme. The molecular sizes of both fragments are shown on
the left. Lane Ma in A is the 1 kb DNA ladder (BRL). lane Mb in A and lane M in
B are DNA markers made from pBR322 DNA digested with Alu l. The sizes of

the markers are indicated by the numbers.

71

Ice

I3

'8

l3

col lom
3|. loo.

 

an. N F .2 an. as £2 92 m ¢ n N w is

72

were the same as in the primary amplification (Fig. 28).

Construction of the cDNA library from fat bodies of vitellogenic female
mosquitoes

Total RNA from mosquito fat bodies was isolated by the guanidine
thiocyanate method as described by Bose and Raikhel (1988). Mosquito fat
bodies for cDNA library construction were collected from females at 0 to 48 hr
after a blood meal with 6 hr intervals. Polyadenylated mRNA [poly(A)+RNA] was
obtained from total RNA by chromatography on an oligo(dT)-callulose column
(Ausubei ei al., 1990). The integrity of poly(A)+ RNA was confirmed by Northern
blot hybridization, using both V9 and mLAP probes. A cDNA library was
constructed from 20 ug of mosquito poly (A)"' RNA in a AZAP ll vector
(Stratagene Co). The oligo-dT- primed cDNA was fractionated to collect
fragments bigger than 400 base pairs; these were then inserted into the EcoRI
site of the lZAP ll. The titer of the unamplifiad and amplified libraries are 1.7 X
106 pfu/ml and 6 X 109 pfulmi, respectively.

Cloning and sequencing of thM

The 96-bp amplified cDNA fragment which was obtained by the PCR, was
digested with Xbal. The released 86-bp fragment (Fig. 10) was subcloned into
pUC 119 and sequenced. After positive confirmation of its sequence, this 86 bp
mLAP cDNA fragment was used for screening of the AZAP ll cDNA library. With
this probe, 40 putative mLAP clones were isolated from the cDNA library. The
cDNA clone with the longest insert (1420 base pairs) was sequenced, using a
single-strand dideoxy chain lamination method (Sanger at al., 1977; Sambrook
et al., 1989). The results were confirmed by sequencing in both directions (Fig.

3).

73

 

 

 

 

 

 

 

 

 

 

 

 

 

200bp

Fig. 3. Sequencing strategy of mLAP cDNA clone.

Numbers at the top correspond to the nucleotides in the cDNA. The coding
region is indicated by a box. Positions of EcoR1 restriction sites are denoted.
Horizontal arrows indicate the direction and extent of each sequencing

determination.

74

Analysis of nucleotide and amino acid sequences of mLAP

The analysis of the deduced amino acid sequence of mLAP was
performed by using the FASTA program (University of Wisconsin Genetics
Computer Group Software), according to the algorithm of Lipman and Pearson
(1985). The deduced amino acid sequence of mLAP was compared to deduced
amino acid sequences translated from GenBank (release 68.0) and Swissprot
(release 18.0) databases (Table 1). The initial score was calculated using the
best sequence alignment of two sequences. The optimized score, calculated by
considering insertions or deletions, gives a batter indication of functional
relations between proteins. The statistical significance of the scores was
evaluated by Z value [(similarity score - mean of random scores)l(standard
deviation of random scores); Z > 10 indicates statistical significance]. The
similarities between mLAP and other sequences were calculated as the
percentage of amino acid number having identical residues or functional
substitutions, relative to the total number of amino acids in mLAP (387).

The dandrogram showing the relationship between members of a family
of aspartic proteases was generated by the pairwise alignment of several
aspartic proteases, with a computer GCG program (University of Wisconsin
Genetics Computer Group Software). The similarity scores were used to create
a clustering order that was represented as a dandrogram. Aspartic proteases
revealing higher similarity scores to mLAP as indicated in Table 1, were chosen
for this analysis. The following sequences were used in this analysis: mLAP,
human cathepsin D (Faust et al., 1985), pig cathepsin D (Shawale and Tang,
1984), mouse cathepsin D (Grusby at al., 1990), Barley aspartic protease
(Runaberg-Roos at al., 1991), human cathepsin E (Azuma at al., 1989), bovine
chymosin A (Harris at al., 1982), human papsinogen A (Sogawa at al., 1983),

human renin (lmai at al., 1983), yeast protainase A (Ammerar at al., 1986), and

cure. «a

75

Table 1. Homology of mLAP to aspartic proteases.

 

SOURCE PROTEIN SCORE‘ Z” IDENTITY

 

 

 

 

 

 

 

 

 

  
  
  

 

 

 

VALUE %

Mouse *Cathapsin D 1,177 143.74 55.8
Human “Cathepsin D 1,170 137.56 57.8
Porcine *Cathapsin D 1,143 160.49 52.7
I Human Cathepsin E 988 100.46 43.4
Yeast *Protainasa A 885 164.48 42.1
Human Ranin 894 116.29 40.5
Human Papsinogan A 833 78.90 42.1
Barley *Asp Protease 775 40.00 35.1
Bovine Chymosin A 763 39.00 47.8
Rhizopus Asp Protease 549 40.70 _ 31.0

 

 

 

 

 

 

 

'Only optimized scores are presented.

"2 value is a statistical significance of the optimized score
(2 > 10 indicates statistical significance).

*Indicatas the lysosomal or vacuolar enzyme.

76

Rhizopus aspartic protease (Horiuchi at al., 1988). The dandrogram
demonstrates the relationship among proteins based on their similarity of amino
acid sequences and it also implies the evolutionary relationship among these

proteins.

Preparation of RNA and Northern blot analysis of mLAP and V9 mRNAs

The isolation of total and polyadenylylatad RNA from mosquito fat bodies
was performed according to the method described by Bose and Raikhel (1988).
Fat bodies were dissected from mosquito females collected from 0 to 48 hr after
a blood meal with 3 hr or 6 hr intervals. Non-blood fed mosquitoes used for
experiments were 3 to 4 day old. The abdominal wall with the attached fat body
was removed and used for the isolation of the RNA and protein. For the
Northern analysis, the total RNA was separated by 1.2% fonnaldehyde/agarose
gel electrophoresis, transferred to nitrocellulose membrane, and hybridized with
either a [32P]-labeled 593 bp fragment of mLAP cDNA or a [32Pj-labaled 1.8 kb
fragment of V9 gene A1 (gift from HR. Hagedorn, University of Arizona) under
high-stringency conditions (Fig. 4).

For quantitation, radioactive Northern blots of mLAP and V9 were
exposed on ”phosphor" screens (Molecular Dynamics Co) for several hours and
than the screens were scanned with "Phosphor” Imager (Molecular Dynamics).

The data were analyzed using the lmagaQuantTM program.

Protein preparation and Western blot analysis of mLAP and Vg

Mosquito fat bodies were homogenized in the homogenization buffer: 100
mM sodium phosphate buffer, pH 7.0, 150 mM NaCl, 0.02% sodium azide. The
homogenization buffer was supplemented with a mixture of protease inhibitors: 1

mM phanylmethylsulfonyl fluoride, 5 mM a-amino-caprionic acid, 1 mM

77

Fig. 4. Northern blot analysis of Vg mRIM in mosquito fat bodies during
vitellogenesis.

Fat bodies, dissected from non-blood fed mosquitoes (N) and from vitellogenic
mosquitoes at indicated hours after a blood meal, were used for extractions of
total RNA Total RNA was resolved by 1.2% forrnaldehyda agarose gel
electrophoresis. The amount of RNA corresponding to the fat body of one
mosquito was loaded into each lane. The RNA was transferred to a
nitrocellulose membrane. A 1.8 kb fragment of the mosquito Vg gene A1,
labeled with [3213], was used for hybridization. The molecular weight of V9 RNA

is marked on the right.

78

ST

 

at”.
. . .e e.
. . r
as Q . U I. .3

3 3 on an R a Z 3 2 e m... z

79

benzamidine, 10 ug/ml aprotinin, and 2 ug/ml each antipain, leupaptin and
chymostatin. The homogenate was centrifuged at 27,216 x g for 30 min. The
supernatant was used for 9% SDS-PAGE under reducing conditions. The
amount of total protein corresponding to the fat body of one mosquito was
loaded into each lane. Resolved proteins were transferred to nitrocellulose
membrane using Bio-Rad Blotting Apparatus. Mosquito LAP was detected with
polyclonal antibodies (Cho et al., 1991a) and a ECL Western blotting detection
system. Vitellogenin was detected with a mixture of two monoclonal antibodies
(ratio 1:1), both directed against a small Vg subunit, 65 kDa (clone 261 and
B11D12; Raikhel at al., 1986) and the same detection system (Fig. 5). The ECL
Western blotting detection system was applied using the manufacturer's protocol
(Amersham).

Chemiluminescent immunoblots were exposed to X-ray films. The films
with bands below saturated exposure were subjected to Computing
Dansitometar Model 300A (Molecular Dynamics). The quantitation was
performed with the program that was used for analyzing the Northern blot data.

80

Fig. 5. Western blot analysis of Vg in the mosquito fat body during
vitellogenesis.

Fat bodies collected as described in Fig. 4 were subjected to protein extraction.
Total proteins were resolved by 9% SDS-PAGE under reduction conditions. The
amount of total proteins equal to the fat body of one mosquito was loaded into
each lane. Vitellogenin is detected with monoclonal antibodies against to a small
Vg subunit. The molecular weight markers are in order of decreasing Mr

phosphatase b, bovine serum albumin, ovalbumin, and carbonic anhydrase.

81

3.

N34
:7

3.:

 

 

‘ r

3 3 cm em 3 fl 2 m m... z

82

RESULTS AND DISCUSSION

1. ANALYSIS OF cDNA CODING FOR MOSQUITO LYSOSOMAL
ASPARTIC PROTEASE

Cloning and analysis of mLAP cDNA

The cDNA encoding mLAP was cloned by an approach using the
polymerase chain reaction (PCR) and screening of a AZAP ll cDNA library
specific to the fat body of vitellogenic female mosquitoes.

The PCR primers were designed and synthesized based on the amino
acid sequence of the N-terminal of mLAP (primer 1), and the first catalytic
domain of aspartic proteases (primer 2). A 96 base pair fragment was obtained
by PCR using these primers (for more details see "EXPERIMENTAL
PROCEDURES”). The sequence of the 96 bp fragment revealed a 70% identity
with human cathepsin D. Both ends of this fragment matched the PCR primer
sequences. The PCR-amplified fragment was subcloned in pUC119 and a 86 bp
fragment was released by enzyme digestion. On a Northern blot of total RNA
from vitellogenic fat bodies 36 hr post-blood meal (PBM), it hybridized with a 1.5
kb mRNA. This probe was used to screen a lZAP— ll cDNA library specific to the
vitellogenic fat bodies. 0f forty putative mLAP clones obtained the clone with the
longest insert (1,420 base pairs) was sequenced. The sequence of this cDNA,
obtained with the strategy shown in Fig. 3, is presented in Fig. 6. There is an
open reading frame of 1,164 nucleotides and a polyadenylation signal (AATAAA)
at 19 nucleotides upstream of the poly(A) tail.

Analysis of the deduced amino acid sequence of mLAP

The predicted amino acid sequence of mLAP is 387 residues (predicted

83

Fig. 6. Nucleotide and deduced amino acid sequence of mosquito lysosomal
aspartic protease (mLAP).

The putative signal peptide is indicated in bold letters. The putative pro-enzyme
sequence is underlined. The amino acid sequence matching the N-tarminus of
the purified mLAP, determined by microsequencing is boxed. A potential
glycosylation residue (N) is marked by a solid square. Amino acids (D) of
conserved catalytic sites are marked by diamond-shaped marks. The putative
polyadenylylation signal (AATAAA) is underlined by a thick broken line. Positive

numbers are counted from the first amino acid of the mature enzyme.

61

121
-S3

181
-37

241
-17

301
361
24

421
44

481
64

S41
84

601
104

661
124

721
144

781
164

841
184

901

204

961
224

1021
244

1081
264

1141
284

1201
304

1261

.324 .

1321
1381

84

GGCACGAGCGGCACGAGGGGTTTCACAGCAAGTAGAACGTCCTTTTTATTGTTCCTCTTT
CCAAAGGAAGAAICCTTCTAASAGCAGACGTTGAOGAIAAGCCTTTGAAATCATCGCTCT

CTAGCCGGCGAGAIGCTAATTAAATCAAIIATTGCCCTCGTTTGCTTGGCCGTTCTATCC
I. L I K .8 I I A L V C L A V"L 8

CAGGCGGACTTTGTTAGAGTTCAGCTGCATAAAACTGAAAGTGCGCGTCAGCATTTTCGA
Q A 2 z 2 B V 9 L a E I 3 § A B Q N P .8

umracacacoeacamamrmocmcraraarccmrarmc
IL V D '1' l J J_2_L_B_II_.IS_¥_I!_8_Y_§ G

, LAA‘A_ 1, .1 l; 1 A” hmmrmcnu
P £4:P .L § A! x L Q A Q, I X I T I G T

CCACCGCAGAGCTTCAAAGTTGTGTTCGAIACGGGATCATCTAACCTTTGGGTGCCCTOG
P P Q S P K V V P D T G S S N L N V P S
4

 

 

 
   

 

 

AAGGAGTGCTCATTCACCAACAECGCTTGCTTGATGCACAACAAATACAATGCCAAGAAG
K 8 C S P T N I A C L N H 'N K Y N A K K

TCATOGAOGTTOGAAAAGAACGGAACAGCTTTCCAIATTGAATAIGGAICTGGTAGCTTA
S S I P B N = G T A P N I Q Y G S G S L

TCTGGTTACTTGTCAACTGACACCGTTGGTTTGGGAGGGGTTTCOGTIACGAAACAAACC
S 'G Y L S T D T V G L G G V S V T R Q T

TTCGCTGAAGCCAIGAASGAACCAGGAITGGTATTOGTTGCGGCCAAGTTTGACGGAAIT
P A B A I N E P G L V P V A A K P D G I

CTCGGAITAGGCTACAGCTOGAITTCAGTAGATGGCGTCGTACCAGTATTCTACAATATG
L‘ G L G Y S S I S V D G V V P V P Y N H

TTCAACCAGGGTCTCAIGGAIGCTCOOGTTTTCTCTTTCTATTTGAAICGTGATCCAAGT
P N Q G L I D A P V P S P Y L N R D P S

GCTGCTGAGGGTGGCGAAATTAITTTCGGTGGAICAGACTCGAAIAAGTATACTGGGGAC
A A. I G G B I I P G G S D 8 N K Y T G D

TTTACTTAICTGTCGGTGGACCGIAAAGCCIACTGGCAAITCAAAAIGGACTCOGTTAAG
P T; Y L S V D R X A Y W Q P K N D S V X

GTTGGCGAIACTGAGTTCflGCAACAAIGGAIGCGAAGCAAITGCCGAIACCGGCAOCAGC
V G D T 8 P C N N G C E A I A D T G T S
4

TTGAITGCCGGCCCAGTG1CGGAGGTCAOOGCTATCAAGAAGGCTAIOGGTGGCACTCCT

L I A ,G P V 8 8 V T A I N K A I G G T P .

ATTA2GAAOGGAGAAIAGAflGGITGACTGCTOGTTGAITCCCAAACTGOCAAAGAPCTCA
I N N .G P. I N V’ D C 8 L I P X L P K I S

TTOGTTTTGGGAGGAAAAICAITGGAICTOGAAGGTGCTGATTACGTACTGOGTGTGGCT
P V L G G K 8 P D L B G A D Y V' L R V A

CAAAIGGGtAAAAGCAICTGCCTGPCTGGGTTCAIGGGAAIGGAIAITCCAOOGGCIAAI
Q N G K I I C L S G P N G I D I P P P N

GGACCGTTGTGGAITTTGGGAGAOGTTTTCATPGGTAAAIAITACACOGAATTCGAIATG
G P L N I L G D V P I G K Y Y T B P D N

GGCAAIGAICGCGTTGGAITTGCCACTGCTGTCTAAAGAAITGATAGAITTGTATTGGTA
G N D R V G P A T A V

AAAAIACCTGCACAITTCCAGTTCCAAAAATATTATTAGAAAGTGTGCATTACTGAAAA!

AAAATGTGAGTTAAACTOGTGCCAAAAAAAAAAAAAAAAA

60
120

180
-38

240
-18

300
360
23

420
43

480
63

S40
83

600
103

660
123

720
143

780
163

840
183

900
203

960
223

1020
243

1080
263

1140
283

1200
301

1260
323

1320
334

1380
1420

85

Mr=41,890). Hydropathy analysis of the deduced amino acid sequence (Kyte
and Doolittle 1982) indicated the N-terminal portion of mLAP (amino acid
residues -53 to -36) is highly hydrophobic. This feature is typical of a signal pre-
propeptide sequence (Fig. 7). A 18-residue N-terminal portion of mLAP is a
putative signal peptide, according to the (-3, -1) rule for signal sequence
cleavage sites (von Heijne, 1986).

Lysosomal aspartic proteases, such as cathepsins D, are synthesized as
pro-enzymes and activated in lysosomes by cleavage of pro-enzyme sequences
(Yonezawa et al., 1988). A comparison of the deduced amino acid sequence of
mLAP with the N-terminal sequence of a mature mLAP (Cho et al., 1991a),
suggests that 35 amino acids of mLAP is a putative pro.enzyme sequence (Fig.
6). The mature subunit form of the mLAP, therefore, is 334 amino acids with the
predicted M, of 35,812 Da. The mLAP subunit, estimated using reducing SDS-
PAGE, is 40 kDa (Cho et al., 1991a). There is a single potential N-linked
glycosylation site at the position Ash-7O (Fig. 6). We expected a glycosylation
site because mLAP is a glycoprotein (Cho et al., 1991a). The increase in mature

mLAP Mr is likely from glycosylation.

Comparison of the mLAP amino acid sequence to aspartic protease
sequences

Comparison of the mLAP amino acid sequence to deduced amino acid
sequences of lysosomal and non-lysosomal aspartic proteases revealed
significant similarities (Table 1). High similarities exist between mLAP and
cathepsins D from several mammalian species, human cathepsin E, and yeast
vacuolar proteinase A (Table 1). The sequence identity is 58% between mLAP
and human cathipsin D (HCD), and 44% between mLAP and human cathepsin E

(HCE). However, considering both identical and conserved replacements, the

86

Fig. 7. Hydropathy plot of the deduced amino acid sequence of mLAP.
The hydropathy plot was obtained by using the algorithm of Kyte and Doolittle
(1982). The ordinate gives the hydropathic index. The abscissa indicates the

amino acid position corresponding to the deduced sequence in Fig. 6.

87

can

NV" 52. «v—

 

 

 

nan

oudnzfiomnrz

 

o~noxmomn>z

no 5% n: mm.-

1 ‘

o..~..
9?
a...

0.."

 

.o.~

.ogw

 

 

h 3016‘:

$30 .35! 80.: 00:31.00 duos 0.38. woos—won

nouvvuuoon use ovaxv >zb¢homn>=

87

van

«6N

NV"

50—

NV— 56 A? m. an-

 

 

 

‘

 

o~a~=mo¢a>= . T .

.0...“

.04”
0~ao=ho¢n¥z

 

 

h 30cc“:
«:90 ace! long oocosvoo duos acute vooavon

nonvvuuooa v3. ovmxv >=F¢moma>z

88

similarity is 92% between mLAP and HCD and 81% between mLAP and HCE
(Fig. 8). Although the putative cleavage sites for the signal peptide and the pro-
enzyme in mLAP are at the positions similar to those in HCD (Faust et al., 1985),
these sites are not conserved in the HCE sequence (Athauda et al., 1990;
Azuma et al., 1989).

Catalytic activity of aspartic proteases depends on two aspartic acid
residues at active centers (Shewale and Tang, 1984). These two catalytic
centers with aspartic acid residues (Asp-33 and Asp-219) as well as surrounding
residues, are conserved in mLAP (Fig. 8).

Human CD has two N-linked glycosylation sites, but mLAP and HCE each
possesses only one glycosylation site. The position of the mLAP glycosylation
site is identical to the first HCD site but not to the HCE one. In mammalian
lysosomal enzymes, the mannose 6-phosphate residues serve as the
recognition marker for the targeting of these enzymes to lysosomes (Kornfeld,
1987). Lysine 203 and amino acids 265-292 of HCD are required for
phosphorylation of its high-mannose oligosaccharides (Baranski et al., 1990;
1991). In contrast, the first phosphorylation determinant is missing and the
second determinant is not well conserved in HCE. Cathepsin E is a non-
lysosomal enzyme which is associated with either the plasma membrane or
cytosol in gastric mucosa and neutrophils (Ueno et al., 1989; lchimaru et al.,
1990). The first determinant (Lys 203 of HCD) is found at Lys 192 in mLAP.
The residues surrounding Lys 192 are conserved in mLAP (Fig. 8). The second
phosphorylation determinant of mLAP (C-253 to L-283) is 43.3% identical to the
corresponding sequence of HCD (C-265 to L-293). Considering functional
substitutions, the similarity between these sequences of mLAP and HCD
becomes 86.7% (Fig. 8).

89

Fig. 8. Alignment of amino acid sequences of mLAP. human cathepsin D
(HCD) and human cathepsin E (HCE).

Amino acid sequences deduced from corresponding cDNAs were aligned by the
FASTA computer program. Vertical dash lines indicate identical residues. and
colons denote functional substitutions. Only a portion of HCE sequence is
presented. The small arrow indicates the putative cleavage site for the signal
peptide. The arrowhead refers to the cleavage sites for the pro-enzyme. The
potential glycosylation residues are marked with thick lines. The conserved
enzyme catalytic residues are denoted with diamond-shaped marks.
Phosphorylation determinants, lysine 203 and the amino acid stretch 265-292 of

HCD, are marked by asterisks. Dashed lines represent gaps.

90

H-CD MQPSSLLPLALCLLAAPASALVRIPLHKPTSIRRTMSEVGGSVEDLIAKGPVSKISQAVP

 

:':::' 3" : :::::"::"' :' I: :::I::::::' 'I:
I I' I II II III I I I
ILAP NLIKSIIAL-VCLAVLSQADPVRVQLHKIBSARQHPRNVDTBIKQLRL ----- KYN----
I. I
.z: :: :.: : z: 3:: : z: I: ::
H-CE-LLELGEAPGSLHRVPLRRHPSLKKKLRARSQLSEPWKSHNLDH‘IQPTE SC
' 4
H-CD AVTEGPIPEVLKNYMDAQYYGEIGIGTPPQCPTVVPDTGSSNLWVPSIHCKLLDIACWIH
== ' '-:"=' m: :lIIlI tztzlt Hum: I w: s!“ a
:I I II I ‘I II III I I I

II I I .I III I III
II 3I3I3I 33II III3 I3I 3I II IIJIIIIII I I3 II I
H-CE -SHDQSAKEPLINYLDMEYFGTISIGSPPQNFTVIPDTGSSNLWVPSVYC--TSPACKTH

H-CD HKYNSDKSSTYVKNGTSPDIHYGSGSLSGYLSQDTVSVPOQSASSASALGGVKVERQVPG

 

Iloo'III. IIIIoloI.IIIIIIII II IIIo: :” I3: 3
II" III‘ IIII‘ ‘I‘IIIIIIII II III' :I
ILAP NKYNAKKSSTPBKNGTAPHIQYGSGSLSGILSTDTVGL GGV
:::::::III ::: I :I IIIIzII II ::I I:: :IzzI: I I:
H-CE SRPQPSQSSH SQPGQSPSIQYGTGSLSGIIGADQVSV EGLTVVGQQPG

 

M-CD EATRQPGITPIAAKPDGILGMAYPRISVNNVLPVPDNLMQQKLVDQNIPSPYLSRDPDAQ

-II=8=Il=tIzIIIIIIIII::I:=III::I:III IxzsI Isl: :IIIII3III=I:
mp ntmmmmoxmmrssxsmmmoonmnvg'smpm

. III .II.I. III I. .. . . I II. .I I.
I3 III -II-I-III IIIIII°I°'I°II° III II-3I 3I3I I3II I-333I333

H-CE ESVTEPGQTPVDAEPDGILGLGYPSLAVGGVTPVPDNNHAQNLVDLPMPSVYMSSNPBGG

* O
H-CD PGGELMLGGTDSKI!KGSLSYLNVTRXAYWQVHLDQVEVASGLTLCKEGCEAIVDTGTSL
IIIzzzItzII: I:I=::II=I=IIIIII HI IzIzsz :zlzztlIIhII III
ILAP BGGBIIPGGSDSNKYTGDPTYLSVDRKAYWQPKIDSVKVGDT-BPCNNGCBAIADTGTBL
IIIIII I I I IIIIII.IIIII
3 3I3II I I 3333I3333 33I33 3IIII3 33I 33II3I II33II3 II3IIIIII

H-CB AGSELIPGGYDHSHPSGSLNWVPVTKQAYWQIALDNIQVGGTVHPCSEGOQAIVDTGTSL

****************************
H-CD MVGPVDEVRELQKAIGAVPLIQGEYHIPCEKVSTLPAITLKLGGKGYKLSPEDYTLKVSQ
2=III=II H: 'IIIHIHsIIII: I: us“ Is: IIIIH = =II=IslzI
ILA? IABPVSEVIAINKAIGGTPIINGBYNVDCSLIPKLPKISPVLGGKSPDLBGADYVLRVAQ
I:II :: :::::III::I: :III IzI: : :Izz :I:::I H: : :zIzI
H-CB ITGPSDKIKQLQNAIGAAPV-DGBYAVECANLNVHPDVTPTINGVPYTLSPTAYTLLDPV

H-CD AGKTLCLSGPMGHDIPPPSGPLWILGDVPIGRYYTVPDRDNNRVGPAEAARL
III3IIIIIII3IIIII3III lll:l:|l3:l: II33I 3:lll 3I
I I I IIII IIIII III III I II I II I III
ILAP IIGKIICLSGPIIGIDIPPPNGPLWILGDVPIGm‘l‘BPDIIGNDRVGPAIAV
I3 3I II :3II3II3II:II::I::: 33:3 :33I'3III333I:
I I II II II I II I III I
H-CE DGNQPCSSGPQGLDIHPPAGPLWILGDVPIRQPYSVPDRGNNRVGLAPAVP

56
56

64

116
105
113

176
165
173

236
224
233

296
284
292

348
334
343

91

This high conservation of determinants of oligosaccharide phosphorylation
suggests mLAP possesses a mannose-6-phosphate residue. The actual status
of the mLAP oligosaccharide moiety, with respect to its phosphorylation. requires
further investigation.

Multiple sequence alignment of both lysosomal (group I) and non-
lysosomal (group II) aspartic proteases were done (Fig. 9). Only vertebrate
cathepsins D have an additional stretch of amino acids the ”GI-hairpin” which is a
proteolytic processing region (Yonezawa et al., 1988). In vertebrate lysosomes,
the mature cathepsins D are generated after two steps: the first removes their
pro-enzyme sequences, and the second cleaves at the (II-hairpin to form the
heavy and light chain subunits (Faust et al., 1985; Conner, 1989). Proteolytic
conversion may be important for the stability of the cathepsin D tertiary structure
(Yonezawa et al., 1988). The maturation of the other aspartic proteases is
different. The yeast vacuolar proteinase A is cleaved once at the pro-enzyme
sequence (Faust and Kornfeld, 1989). The barley aspartic protease does not
have a cleavage site similar to that of cathepsin D. Rather it is cleaved at new
sites, resulting in a heterodimeric form of an active mature enzyme (Runaberg-
Ross et al., 1991). The mLAP forms a dimer from two identical 40-kDa subunits
(Cho et al., 1991a).

Tang and Wong (1987) suggested all aspartic proteases are derived from
the same ancestral enzyme. lt is clear that mLAP is most similar to cathepsin D.
Mosquito LAP, however, shares with cathepsin E the dimeric structure, low
optimal catalytic pH value, and lack of a III-hairpin processing sequence. The
pairwise computer comparison provided further evidence as for the evolutionary
relationships between these members of a family of aspartic proteases. A

dendrogram, generated by this analysis, is shown in Figure 10.

92

 

 

 

 

 

 

 

 

GROUP - I

Mosquito Aspartic Protease --dtvgl ggvs--
Human Cathepsin D --dtvsv P C Q's A S S A s AIL ggvk--
Mouse Cathepsin D --dtvsv P C K S D Q S K A ——- rgik--
Pig Cathepsin D --dtvsv P CIN 8 A L S G'V ——— ggik--
Bovine Cathepsin D --dtvsv P C N P 8'8 8'8 P ——— ggvt--
Barley Aspartic Protease --dsvtv gdlv--
Yeast Proteinase A --dt1si gdlt-—
GROUP-II

Human Cathepsin E --dquv eglt--
Human Renin --diitv ggit--
Human Pepsinogen A --dtqu ggis--
Bovine Chymosin A --dtvtv sniv--
Rhizopus Aspartic Protease --dnvn1 9911--

Fig. 9. Sequence alignment of regions of aspartic proteases corresponding
to the cleavage sites of cathepsins D.

Several aspartic proteases were compared by a multiple alignment computer
program. Corresponding partial sequences are presented. Lower case letters
indicate homologous residues, capital letters are additional residues found in
cathepsins D. Arrowheads denote the identified processing sites for the heavy
and light chains of cathepsins D. Solid lines indicate gaps. The sequences of
the following enzymes were used: mLAP. cathepsins D (Shewale and Tang,
1984; Faust et al., 1985; Grusby et al., 1990). yeast protainase A (Ammere et
al., 1986). cathepsin E (Azuma at al., 1989), renin (lmai et al., 1983),
papsinogen (Sogawa et al., 1983), barley-grain aspartic protease (Runeberg-
Roos et al., 1991), bovine chymosin (Harris 9! al., 1982) and rhizopus aspartic
protease (Horiuchi et al., 1988). Group I includes aspartic proteases of

lysosomal/vacuolar origin, group II non-lysosomal and secretory enzymes.

93

Fig. 10. Relationship between members of a family of aspartic proteases.

The dendrogram of aspartic proteases was generated by pairwise comparison of
deduced amino acid sequences of aspartic proteases. The distance along the
vertical axis is proportional to the difference between sequences. Asterisks

indicate lysosomal or vacuolar enzymes.

94

ouncuoum onuuoau< osmouwnm

4 suecaeuoum vane»

cacom seas:

oaoououm cauuomnt hoauam

sea-acne nausea-d assoeoeha ouusvsox

a cannonvso oesox

a cauaoeuao use

a cannonuoo cuss:

m sqeaonuou coesm

4 couosaumom ones:

4 canoehno esa>om

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

95

ll. EXPRESSION OF MOSQUITO LYSOSOMAL ASPARTIC PROTEASE IN
THE FAT BODY DURING VITELLOGENESIS

In fat bodies of vitellogenic female mosquitoes, the lysosomal activity
increases dramatically by the end of Vg synthesis. The specific activities of
several lysosomal enzymes, including cathepsin D-like activity, rise at 24-26 hr
after the initiation of V9 synthesis by a blood meal and reach a maximum at 36-
42 hr PBM (Raikhel, 1986a). In this study, we determined the expression
kinetics of the mLAP mRNA and compared it to the kinetics of mLAP protein in
mosquito fat bodies during the V9 synthetic cycle. As a control, we monitored
the kinetics of V9 at both mRNA and protein levels in fat bodies for the same
time periods.

Mosquito LAP mRNA levels increased between 6 and 12 hr PBM. The
levels were highest at 24 hr PBM, and then gradually declined to the background
level by 48 hr PBM (Figs. 11 and 12). In contrast, mLAP protein levels
monitored by immunoblot increased 12 hr later than mLAP mRNA (Figs. 11 and
12). Mosquito LAP protein was elevated at 24 hr PBM, but the highest level was
between 36 hr and 42 hr PBM. The increase in mLAP protein paralleled the
activity of a cathepsin D-like enzyme (Raikhel, 1986a).

We examined the accumulation of V9 mRNA and V9 protein during Vg
synthesis in the mosquito fat body. Unlike mLAP, both Vg mRNA and V9 protein
accumulated at similar rates reaching maximal levels at 24 hr PBM, and then
rapidly declining (Fig. 12). Racioppi et al. (1986) reported that the amount of V9
mRNA reached its peak levels at 36 hr PBM. Accoring to our data, however,

both Vg mRNA and V9 protein have similar kinetics and peak at 24 hr PBM.

96

Fig. 11. Expression of mLAP in the mosquito fat body during vitellogenesis.

Fat bodies, dissected from non-blood fed mosquitoes (N) and from vitellogenic
mosquitoes at indicated hours after a blood meal, were used for extractions of
total RNA and proteins. (A) Total RNA was resolved by 1.2% formaldehyde
agarose gel electrophoresis. The amount of RNA corresponding to the fat body
of one mosquito was loaded into each lane. The RNA was transferred to a
nitrocellulose membrane, and hybridized with a [32P1-Iabeled 593 bp mLAP
probe. A RNA ladder (BRL) is shown to the right. (8) Total proteins were
resolved by 9% SDS-PAGE under reducing conditions. The amount of total
protein equal to the fat body of one mosquito was loaded into each lane.
Polyclonal antibodies against mLAP were used for the immunoblot which was
processed as described in "Experimental Procedures”. The molecular weight
markers are in order of decreasing Mr phosphatase b, bovine serum albumin,

ovalbumin, and carbonic anhydrase.

97

N 0.5 6 12 1824 3036 42 48
(kb)

~9.5
*7.5

4.4

“-2.4

 

-0.24
Ill)
in?
-66.2

’0“- ‘—-’-

 

98

Fig. 12. Kinetics of mLAP and V9 at the mRNA and protein levels in the
mosquito fat bodies during vitellogenesis.

Quantitation of MRNA (-o-) and proteins (-V-) were performed using methods as
described in ”Experimental procedures". Quantitation is based on Northern and
Western blots of Figs. 4, 11 and 13. The upper panel is mLAP and the lower
panel is Vg. Numbers on the bottom indicate the time after a blood meal when

fat bodies were collected.

99

Ac_goi.n<l.mo_xv £28 35%
O O O O O

60

 

      

 

 

 

5 4 3 2 1| 0
a \V _ _ D
I \V 1
l V I
— E
O O O O O O
5 2 9 6 3

A<szroLno§v 36ch 232%

Aseoiqunog £28 magma

 

 

 

 

 

14—VG

_ _ _ _ _ _
2 O 8 6 4 2 O

16

A<zmerokno§v €28 252$

18 24 30 36 42 48

12
Hours After A Bldod Meal

6

O

100

Both Vg mRNA and mLAP mRNA reached their highest level 24 hr PBM.
The hemolymph titer of an insect steroid hormone, 20-hydroxyecdysone (20-HE)
is highest right before the peaks of mLAP and V9 mRNAs. at 18-20 hr PBM.
After this time, 20-HE levels dramatically drop (Hagedorn, 1985). Since Vg
synthesis is likely regulated at the transcriptional level by 20-HE (Hagedorn.
1985; 1989; Raikhel, 1992), 20-HE may also regulate mLAP gene expression.
The 12 hr delay in the increase of mLAP protein, however, suggests mLAP
synthesis is regulated at the translational level.

Steroid hormones have been implicated in regulation of gene expression
at the transcriptional and post-transcriptional levels. Steroid hormones affect the
stability of mRNAs (Brock and Shapiro, 1983; Gordon et al., 1988; Nielsen and
Shapiro, 1990) and also alter translation rates of certain mRNAs (Meyuhas et al.,
1987; Kumar et al., 1989; Verdi et al., 1989; Verdi and Campagnoni, 1990).
Steroids effects can be either positive or negative.

The expression kinetics of mLAP at mRNA and protein levels suggests
that 20-HE could be involved in the repressing translation of mLAP mRNA. It is
unlikely that the stability of mLAP mRNA is affected since its level exhibits a
steady decline after the peak, at 24 hr PBM (Fig. 12).

101

The translational inhibition of mLAP mRNA by 20-l-lE could be released when
the titer of 20-HE declined 24 hr post-blood meal.

Verdi and Campagnoni (1990) identified a steroid modulatory element in
the 5'-untranslated region of the myelin basic protein (MBP) mRNA. This
element is involved into a positive regulation of translation by hydrocortisone.
Putative modulatory sequences have been also identified in the 5'-untranslated
region of 2',3'-cyclic nucleotide 3'-phosphodiesterase (CNP) mRNA whose
translation is inhibited by hydrocortisone (Verdi and Campagnoni, 1990). We
compared the 5'-untranslated region of mLAP mRNA to those of GNP and MBP.
Three segments of the 5'-untranslated region of mLAP mRNA were found to be

similar to sequences in CNP mRNA (Fig. 13).

102

65 80 90 101
mLAP ---AGGAAGAAUCCUUCUAauagcagacGUUGACGAUAAG -----
CNP ---UCCUUCUA --------- GUUGAaGAUAcG -----
116 123 140 151
65 80 90 101
mLAP ---AGGAAGAAUCCUUCUAauagcagacGUUGACGAUAAG -----
CNP ---AGGAAGAA----
167 174
34 42
mLAP ---AGAACGUCC---
MBP AGAAgacCC---
1 9

Fig. 13. Putative steroid regulatory elements in the 5'-untranslated region of
rnLAPrnRNML

The 5'-untranslated region of mLAP mRNA was compared to those of the
mRNAs, which are translationally regulated by a steroid hormone: the rat 2', 3‘-
cyclic nucleotide 3'-phosphodiesterase (CNP) mRNA translation of which is
negatively regulated by hydrocortisone (CNP cDNA sequence is taken from
Bernier et al., 1987); mouse myelin basic protein (MBP) mRNA is positively
regulated by hydrocortisone (MBP cDNA sequence is taken from Newman et al.,
1987). Numbers indicate the position of the nucleotides from the beginning of

each cDNA sequence. Capital letters denote identical nucleotides.

103

Two 8-nucleotide segments of mLAP mRNA, located in tandem at positions 65
to 80, were identical to the segments of GNP mRNA. The third 12-nucleotide
segment of mLAP mRNA, located in positions 90-102, was only different in two
nucleotides from the CNP element. There is also a 9-nucleotide segment of the
5'-region of mLAP mRNA which has a similarity with the steroid modulatory
element of MBP (Fig. 13). Interestingly, the mLAP mRNA segment is different in
three nucleotides, two of which (GA) were shown to be essential for the MBP
element to function as a positive modulatory element (Verdi and Campagnon,
1990). Although final understanding of the functional significance of the putative
steroid regulatory elements in the 5'-untranslated region of mLAP mRNA
requires direct experimental proof, our finding provides strong evidence for
involvement of 20-HE in the negative translational regulation of this mosquito
lysosomal enzyme.

In conclusion, our analysis of mLAP cDNA has built a foundation for future
studies of the regulation of the expression of mLAP at both transcriptional and
translational levels. With the availability of mLAP cDNA, we can also address an
important question about the mechanism of targeting of lysosomal enzymes in

insects.

104

Acknowledgments

We thank Dr. Henry Hagedorn for his kind gift of the gene probe of
mosquito vitellogenin; and Drs. Will Kopachik and Suzanne Thiem for their

critical reading of the manuscript.

105

REFERENCES

Ammerer, 6., Hunter, C. P., Rothman, J. H., Saari, G. C., Valls, L. A., and
Steven, T. H. (1986) Mol. Cell. Biol, 6, 2490-2499.

Athauda, S. B. P., Matsuzaki, 0., Kageyama, T., and Takahashi, K (1990)
Biochem. Biophys. Res. Comm, 168, 878-885.

Ausubel, F. M., Brent, R., Kingstone, R. E., Moore, D. D., Seidman, J. G., Smith,
J. A., and Struhl, K. (1990) In: Current Protocols in Molecular Biology ,
Green Publishing Associates and Wiley-lnterscience, Vol. 1, pp. 4.51-
4.53, New Yolk

Azuma, T., Pals, G., Mohandas, T. K, Couvreur, J. M., and Taggar, R. T. (1989)
J. Biol. Chem, 264, 16748-16753.

Baranski, T. J., Koelsch, G., Hartsuck, J. A., and Kornfeld, S. (1991) J. Biol.
Chem, 266, 23365-23372.

Baranski, T. J., Faust, P. L., and Konfeld, S. (1990) Cell, 63, 281-291.

Bednarek, S. Y., and Raikhel, N. V. (1992) Plant Mol. Biol, (in press).

Bernier, L., Alvarez, F., Norgord, E. M., Raible, D. W., Mentaberry, A.,
Schembri,. J. G., Sabatini, D. D., and Colman, D. R. (1987) J. Neurosci.,
7, 2703-2710.

Bose, S. G., and Raikhel, S. A. (1988) Biochem. Biophys. Res. Commun., 155,
436-442.

Brock, M. L., and Shapiro, D. J. (1983) Cell, 34, 207-214.

Cardelli, J. A., and Dimond, R. L. (1988) In: Protein Transfer and Organelle
Biosynthesis (Das, R. C., and Robbins, P. W., eds) pp. 363-399,
Academic Press.

Cho. W. L., Dhadialla, T. S., and Raikhel, A. S. (1991a) Insect Biochem., 21,
165-176.

Cho. W. L., Deitsch, K W., and Raikhel, A. S. (1991b) Proc. Natl. Acad. Sci, 88,
10821-10824.

Chrispeels, M. J., and Raikhel, N. V. (1992) Cell, 68, 613-616.

106

Conner, G. E. (1989) Biochem. J,. 263, 601-604.

Dean, R. L., Collins, J. V., and Locke, M. (1985) In: Comprehensive Insect
Physiology, Biochemistry and Pharmacology (Kerkut, G. A., and Gilbert,
L. l., eds) Vol. 3, pp. 155-210. Pergamon Press, Oxford.

Faust, P. L., and Kornfeld, s. (1989) J. Biol. Chem, 264, 479-488.

Faust, P. L., Kornfeld, S., and Chirgwin, J. M. (1985) Pro. Natl. Acad. Sci., 82,
4910-4914.

Glaumann, H., Ericsson, J. L. E., and Marzeila, L. (1981) Intern. Rev. Cytol., 73,
149-182.

Gordon, D. A., Shelness, G. S., Nicosia, M., and Williams, D. L. (1988) J. Biol.
Chem, 283, 2625-2631.

Grusby, M. J., Mitchell, S. C., and Glimcher, L. H. (1990) Nucl. Acids Res., 13,
4008.

Hagedorn, H. H. (1985) In: Comprehensive Insect Physiology. Biochemistry,
and Pharmacology (Kerkut, G. A., and Gilbert, L. l., eds) Vol. 8, pp. 205-
261, Pergamon Press, Oxford .

Hagedorn, H.H. (1989) In: Ecdysone from Chemistry to Mode of Action
(Koolman J., ed.) Thieme Medical Publishers, Inc. New York, pp. 279-
289.

Harris, T. J. R., Lowe, P. A., Lyons, A., Thomas, P. G., Eaton, M. A. W., Millican,
‘ T. A., Patel, T. P., Bose, C. 0., Carey, N. H., and Doel, M. T. (1982) Nucl.
Acids Res., 10, 2177-2187.

Hays, A. R., and Raikhel, A. S. (1990) Roux's Arch. Dev. Biol, 199, 114-121.

Horiuchi, H., Yanai, K, Okazaki, T., Takagi, M., and Yano, K. (1988) J.
Bacteriol, 170, 272-278.

lchimaru, E., Sakai, H., Saku, T., Kunimatsu, K, Kato, Y., Kato, l., and
Yamamoto, K (1990) J. Biochem., 108, 1009-1015.

lmai, T., Miyazaki, H., Hirose, S., Hori, H., Hayashi, T., Kageyama, R., Ohkubo,
H., Nakanishi, S., and Murakami, K, (1983) Proc. Natl. Acad. Scl, 80,
7405-7409

Kornfeld, S. (1987) FASEB J., 1, 462-468.

107

Kornfeld, S., and Mellman, I. (1989) Annu. Rev. Cell Biol, 5, 483-525.

Kumar, S., Cole, R., Chiappelli, F., and De Vellis, J. (1989) Proc. Natl. Acad.
Sci, 86, 6807-6811..

Kunkel, J. G., and Nordin, J. H. (1985) In: Comprehensive Insect Physiology,
Biochemistry, and Pharmacology (Kerkut, G. A., and Gilbert, L. l., eds)
Vol. 1, pp. 83-111, Pergamon Press, Oxford.

Kyte, J., and Doolittle, R. F. (1982) J. Mol Biol, 157, 105-132.

Lea, A. O. (1964) J. Med. Ent., 1, 40-44.

Lipman, D. J... and Pearson, W. R. (1985) Science, 227, 1435-1441.

Locke, M. (1980) In: Insect Biology in Future (Locke, M., and Smith, D. S., eds)
pp. 227-252, Academic Press.

Meyuhas, 0., Thompson, JR, E. A., and Perry, R. (1987) Mol Cell. Biol, 7,
2691-2699.

Nagao, E., Takahashi, N., and Chino, H. (1987) Insect Biochem., 4, 531-538.

Newman, 8., Kitamura, K., and Campagnoni, A. T. (1987) Proc. Natl. Acad. Sci,
84. 886-890.

Nielsen, D. A., and Shapiro, D. J. (1990) Mol Cell. Biol, 10, 371-376.

Nordin, J. H., Gochoco, C. H., Wojchowski, D. M., and Kunkel, J. G. (1984)
Comp. Biochem. Physiol, 798, 379-390.

Orci, L., Ravazzola, M., Amherdt, M., Yanaihara, C., Yanaihara, N., Halban, P.,
Renold, A. E., and Perrelet, A. (1984) J. Cell. Biol, 98, 222-228.

Osir, E. 0., Anderson, D. R., Grimes, W. J., and Law, J. H. (1986) Insect
Biochem., 3, 471-478.

Racioppi, J. V., Gemmill, R. M., Kogan, P. H., Calvo, J. M., and Hagedorn, H. H.
(1986) Insect Biochem., 16, 255-262.

Raikhel, A. S., and Dhadialla, T. S. (1992) Annu. Rev. Entomol, 37, 217-251.

Raikhel, A. S. (1992) Adv. Disease Vector Res., 9, in press.

108

Raikhel, A. S. (1986a) J. Insect Physiol, 32, 597-604.
Raikhel, A. S. (1986b) Tissue & Cell, 18, 125-142.
Raikhel, A. S., Pratt, L. H., Lea, A. O. (1986) J. Insect Physiol, 32, 879-890.

Runeberg-Roos, P., Tormakangas, K, and Ostman, A. (1991) Eur. J. Biochem.,
202, 1021-1027.

Sambrook, J., Fritch, E. F., and Maniatis, T. (1989) In: Molecular Cloning: A
Laboratory Manual (Cold Spring Harbor La., Cold Spring Harbor, NY) 2nd
edition.

Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. Acad. Sci, 74,
5463-5467.

Shewale, J. G., and Tang, J. (1984) Proc. Natl. Acad. Sci, 81, 3703-3707.
Smith, R. E., and Farquhar, M. G. (1966) J. Cell Biol, 31, 319-345.

Sogawa, K, Fujii-Kuriyama, Y., MizuKami, Y., lchihara, Y., and Takahashi, K.
(1983) J. Biol. Chem, 258, 5306-5311.

Tang, J., and Wong, R. N. S. (1987) J. Cell. Biochem., 33, 53-63.

Ueno, E., Sakai, H., Kato, Y., and Yamamoto, K (1989) J. Biochem., 105, 878-
882.

Verdi, J. M., Kampf, K, and Campagnoni, A. T. (1989) J. Neurochem., 52, 321-
324.

Verdi, J. M., and Campagnoni, A. T. (1990) J. Biol. Chem, 285, 20314-20320.
von Heijne, G. (1986) Nucleic Acids Res., 14, 4683-4690.

Wyatt, G. R. (1980) In: Insect Biology in Future (Locke, M., and Smith, D. S.,
eds) pp. 201-225, Academic Press.

Yonezawa, S., Takahashi, T., Wang, X. -J., Wong, R. N. S., Hartsuck, J. A., and
Tang, J. (1988) J. Biol. Chem, 263, 16504-16511.

CHAPTER 4‘

AN EXTRAOVARIAN PROTEIN ACCUMULATED IN MOSQUITO OOCYTES
IS A CARBOXYPEPTIDASE ACTIVATED IN EMBRYOS

* Originally published in Proc. Natl. Acad. Sci.
Reference: Cho, W. -L., Deitsch, K. W., and Raikhel, A. S. (1991) Proc. Natl.
Acad. Sci.,88,10821-10824.

109

110

ABSTRACT

We report a novel phenomenon previously unknown for oviparous
animals; in mosquitoes, a serine carboxypeptidase is synthesized extraovarially
and then internalized by oocytes. The cDNA encoding mosquito vitellogenic
carboxypeptidase (VCP) was cloned and sequenced. The VCP cDNA hybridizes
to a 1.5 kb mRNA present only in the fat body of vitellogenic females. The
deduced amino acid sequence of VCP shares significant homology with
members of a serine carboxypeptidase family. Binding assays using a serine
protease inhibitor, [3H]-diisopropyl fluorophosphate (DFP), showed that VCP is
activated in eggs at the onset of embryonic development. Activation of VCP is
associated with the reduction in its size from 53-kDa (inactive pro-enzyme) to 48-
kDa (active enzyme). The active 48-kDa form of VCP is maximally present at

the middle and disappears by the end of embryonic development.

111

INTRODUCTION

In oviparous animals, development of an embryo depends upon utilization
of yolk proteins accumulated by oocytes during vitellogenesis. In most oviparous
animals, nematodes, arthropods and vertebrates, the task of producing massive
amounts of yolk protein precursors is performed by extraovarian tissues (1-5).
The developing oocytes of these animals are highly specialized for the specific
accumulation of extraovarian yolk protein precursors. The process of receptor-
mediated endocytosis by which yolk protein precursors are accumulated in
oocyte yolk bodies is well characterized for insects and vertebrates (6, 7). In
contrast, the mechanisms of degradation of yolk proteins have received
considerably less attention. Several enzymes that participate in hydrolysis of
yolk proteins in insect embryos have been recently described (8-11). However,
the origin of these enzymes and the time of their delivery into yolk bodies are
unknown.

In this paper, we report a novel biological phenomenon previously
unknown for insects or other oviparous animals. In vitellogenic female
mosquitoes, an extraovarian tissue, the fat body, synthesizes not only the yolk
protein precursor, vitellogenin, but also a pro-enzyme of a serine
carboxypeptidase. We have named it ”vitellogenic carboxypeptidase” (VCP).
This enzyme is internalized by developing oocytes, accumulated in yolk bodies

and activated during embryonic development.

112

EXPERIMENTAL PROCEDURES

Animals
Mosquitoes Aedes aegypti were reared as described by Hays and Raikhel
(12). The embryonic stages of A. aegypti were determined according to

Raminani and Cupp (13).

Cloning and sequencing of cDNA

A cDNA clone of VCP was isolated by immunoscreening of a lambda gt11
library generated from fat bodies of vitellogenic female mosquitoes. This cDNA
clone, pVCP-1, was used to isolate another cDNA clone, pVCP-2, encoding the
5'end 1020 bp of VCP from the same library. The former cDNA clone was
sequenced with the dideoxy chain termination method in both directions (14).

The missing part of the 3' and sequence was obtained by the polymerase
chain reaction using 20 nucleotides at the 3'end of pVCP-1 and poly-d(T)17 as
primers including the Xbal digestion site at the 5' ends of both primers (15).
Amplification was achieved by using cDNA reverse transcribed from 20 pg of
total RNA prepared from whole bodies of mosquitoes 24 hr post blood meal as
templates. The specific amplified band, identified by Northern blot hybridization
to 1.5 kb mRNA, was subcloned into pUC119 for sequencing. Similar to the
pVCP-2 cDNA clone, the 3'-end VCP cloned was sequenced in both directions.

Protein purification for partial sequencing
Mosquito ovaries, 48 hr post blood meal, were homogenized in 20 mM
Tris, 150 mM NaCl, pH 7.5, 0.02% sodium azide. The vitellogenic

carboxypeptidase was isolated from the protein extract by anion-exchange,

113

affinity and hydrophobic chromatographic steps using DEAE Sepharose CL-4B,
concanavalin A and Phenyl Sepharose columns, respectively. A fraction with
nearly pure protein was resolved by SDS-PAGE and subjected to peptide
sequencing using the method and facilities described by Cho et al. (16).

Purification and analysis of RNA

Isolation of total and polyadenylated RNA [poly(A)+RNA] from mosquito
tissues was performed as previously described (17).

In vitro translation of mosquito poly(A)*RNA was performed as described
by Bose and Raikhel (17) using rabbit reticulocyte lysate (Promega Biotech).
The VCP precursor was identified in in vitro translation products of poly(A)"’RNA
by immunoprecipitation with anti-VCP polyclonal antibodies (12).

The 840 bp insert from pVCP-1, a putative VCP cDNA clone, was used for
hybrid selection of VCP mRNA from the fat body poly(A)+RNA according to
Sambrook et al. (15). The hybrid selected mRNA was translated in vitro using
rabbit reticulocyte lysate and the protein product resolved by SDS-PAGE. The
pUC119 plasmid without an insert was used for hybrid selection of mRNA as
control.

For the Northern analysis, total RNA was separated by 1.2 %
formaldehyde agarose gel electrophoresis, transferred to nitrocellulose paper
and hybridized first to a [32P1-labeled 840 bp VCP cDNA insert from pVCP-1
under high stringency conditions. RNA hybridized to the VCP cDNA probe was
visualized by autoradiography of the blot. The probe was then stripped from the
blot (14) and the blot rehybridized with a [32Pl-labeled 1.8 kb fragment of
vitellogenin gene A1 (gift of Dr. H.H. Hagedorn) under high stringency

conditions.

114

Imm unoblot Analysis
Preparation of proteins, SDS-PAGE and immunoblotting were performed
as described by Hays and Raikhel (12).

Binding to Protein Inhibitor

Protein extracts, prepared as for immunoblotting but without any protease
inhibitors, were incubated with 0.1 p 0in l [3H]-DFP (Dupont) for 1 hour at room
temperature. These mixtures were then separated by SDS-PAGE under

reducing conditions and processed for fluorography.

Analysis of deduced amino acid sequence

The analysis of deduced amino acid sequence was performed using
FASTA program (GCG software) according to the algorithm by Lipman and
Pearson (18). The deduced amino acid sequence of VCP was compared with
amino acid sequences from the NBRF Protein Data Base (release 21.0) and with
deduced amino acid sequences translated from GeneBank (release 66.0) and
EMBL (release 19.0) Data Bases. The initial score was calculated from the best
sub-sequence alignment of two sequences. The optimized score, calculated by
considering insertions or deletions, gives a better indication about functional
relation between proteins. The statistical significance of the scores was
evaluated by 2 value [(similarity score - mean of random scores)l(standard
deviation of random scores); 2 > 10 indicates statistical significance]. The
similarities between VCP and other sequences were calculated as the
percentage of amino acid number having identical residues or functional

substitutions to the total number of amino acids in VCP (441).

115

RESULTS AND DISCUSSION

Recently, a female-specific protein (Mr=53,000), initially called 53KP, has
been found in the mosquito, Aedes aegypti. Like vitellogenin, this protein is
synthesized by the fat body of vitellogenic females under the control of 20-
hydroxyecdysone. The kinetics of the 53KP production by the vitellogenic fat
body is also similar to that of vitellogenin: it is produced as early as 4 hr and
reaches its peak near 24 hr after the initiation of vitellogenesis. Synthesis then
drops to low levels by 36 hr and declines to background levels by 48 hr. This
protein is secreted to the hemolymph and is selectively accumulated in yolk
bodies of developing oocytes (12). Data presented here show that this protein
(VCP) is a serine carboxypeptidase.

The cDNA encoding VCP was cloned by a combination of
immunoscreening of a lambda gt11 cDNA library and the polymerase chain
reaction. The identity of the cDNA was confirmed by direct sequencing of the N-
terminus of the purified VCP and by in vitro translation of hybrid selected mRNA
(data not shown).

Northern blot analysis demonstrated that transcription of the VCP mRNA
is limited to only female fat bodies and is initiated after a blood meal. Similar to
vitellogenin mRNA, the amount of VCP mRNA in the fat body is maximal at the
peak of the protein production, 24 hr after a blood meal, then it declines to
background levels by 48 hr after a blood meal (Figs. 2 and 3).

The sequence of full-length cDNA encoding VCP, confirmed by the
sequencing of cDNA clones in both directions, is presented in Fig. 1. The size of
VCP mRNA of 1.5 kb, estimated by Northern blot analysis, is in agreement with
the 1,511 bp mRNA estimated from the cDNA sequence. The VCP cDNA has a

single open reading frame that encodes a protein of 441 amino acids with a

116

Fig. 1. Northern blot analysis of sex- and stage-specific expression of the
VCP mRNA transcript.

Total RNA was extracted from whole bodies of male mosquitoes (lane 1) and fat
bodies of female mosquitoes before a blood meal (lane 2) or 24 and 48 hours
post blood meal (lanes 3 and 4, respectively). RNA ladder (BRL) was resolved in
lane M. A - agarose gel stained with ethidium bromide; B - hybridization with
0.84 kb fragment of VCP cDNA; C -hybridization with 1.8 kb fragment of

mosquito vitellogenin A1 gene.

117

vow

w

vamp

 

vamp:

Iva.

Iv.—
Ivan
lied
\mN
/ m6

9.

.. i "It"? 3?”

l‘Ir 3"..” r...'~

37 3111' I:

05

Q"-’ I
A .\
I.
. .VI

118

Fig. 2. Northern blot analysts of tissue-spectflc expression of the VCP mRNA
transcript.

Total RNA was extracted from ovaries (lane 1), fat bodies (lane 2), and midgut
(lane 3) of female mosquitoes 24 hours post blood meal. RNA ladder (BRL) was
resolved in lane M. A -agarose gel stained with ethidium bromide; B -
hybridization with 0.84 kb fragment of VCP cDNA

M123

 

119

 

1

2

120

Fig. 3. Nucleotide and deduced amino acid sequences of mosquito VCP.

The amino acid sequence matching the N-terminus determined from purified
VCP is underlined by a solid line. The signal peptide is boxed. The circled
amino acid (N) is a potential glycosylation site. A putative polyadenylation signal,

AATAAA, is shaded.

H

n H H

U

H H H
bu an an uH wO we we no no sq Mu um Hm Hm HA Hb HO
OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH

H H
OH

H
MIN HO‘ ‘00 db U'IO UN H01 \00 \Ith mm UN H0 \00 4‘ 01¢ IAIN H0 ‘00 duh mm UN HO

OH OH

9
U
G
H

1441
1501

121

moccacrccmrmcmrnmumW
Wmccrmmmrcwmmcnma
_ g 'r I. w N: P Y is is I.

ATGCGAGGATCGGCGTCTCCTCGTCGTCCAGGTGAAAGTGGTGAACCTTTGTTCCTGACT
M I! G S 11 § 2 I! R P (I E S (3 E P I; F I. T

 

 

 

ccmnnmflmnmnxnxxmmmunsmmnmncnuxmaamumcmamamaumnxmxw
P L I. Q D (3 K I I! E A II N X 11 R V’I' H P

1unTnawmmmmmmnmmmnmcmxncnmmxnmccancmnxmmmxncumnwc
M I. S S ‘I 8 53 Y S (3 P I! T V'I) A II H II S

hymnstunacmmmmmnwumcanmmummmxmmmmmmmxmcamuucrmnm
N L I? F W 1"] P A I! N N I! E Q I\ P I I; V

1mmcmxmmxmmnwammflammaucmnnrmnamusrnnmmmauunmumc
W I. Q G»(3 P G 11 S S I. P G I! F E I! N (3 P

‘rmmwmumcmumc laMmanawmuncannmmxmmcflxxmwmmmmxmm
F [I I H I! N S \I R Q I! B Y £3 W II Q ll H
cmuunxunwaunmumxnxmmnmemunemumcmnmmmumsmnmmcxmma
I! M I: Y I I) N P II G T'(3 F 53.? T I) S I) E

csrmmmammnmmnmmmmmwsnmnnnmmxnnnxmmmmnmmrunmamumc
G 1! S T 1' E E i! G E II L M II P I (IIQ P

rmannannncmxmxnnnrmmmmmmauumcnmwnnwomnmmmmnmxnmw
F ‘I L 1? P N I. L K.II P P 1! I S (3 B 53'! G

tmnmAmnnsmmnmxxrmnammmMmwuxmuummaammmxxmmxnmmmum
(3 K F If P A l'(3 Y A II N S (2 S Q I’ K I

Amunaamemummxmmnnmmmmmxnmummmucaxucmuxmmmnmmnmc
N L (2 G L 11 I G D (3 Y T I) P L 1' Q I.IN Y

GGAGAATACCTGTATGAGCTGGGCCTGATOGATTTGAAOGGAAGAAAGAAGTTOGAOGAG
E I! L Y I! L G I. I D I. N Gr]! K II F I) E

GATACGGCTGCTGCCATOGCCTGTGCCGAACGTAAGGACATGAAGTGCGCCAACCGCCTT
D 1?.A A II I A (2 A E I! K D I! K (2 A 011R L

ammumswmnfirnxmuxmcmxwnxmcmxmmunmmnncmummmnmmaxwa
I (2 G I. F D (3 L D (3 Q 3 £3 Y P II R \V‘T G

ITcmnmanmcmmmmnmaMmamxmmmmnmxmsmxmnmumnkmmnawxns
P 8 £3 Y Y 1' P I 81¢} D E 13 S K (2 D S I] L

mmxmsrnxucmammomxmwmnmomnmemxmmammnrummmuuncoanmc
N E I? L 8 II P E II R R.(3 I H \I G E I; P P

cmanmmcnwmsenummmuusmmmnmnmxnxusnummmnmmcnnncmuum
N I) S I) G I! N I( V II E I! L £3 E I) T I. D I?

mmmcuxmuxmmammmmmflxncmncmnmoaxnuxncrmnmcuunannm
‘V A I? W \I S K I. L S I! Y R.‘I L 3"! N (3 Q

rnmmcmmnaouxmmcnmoaHunaaxnommnnccnnunsmmummcrmummr
I. D I II C II Y P T V'I) P L I! K I! P I? D

ccanummmmsnummxnaxxmxmnunmxmnmaxmcnmmmaxmammxnas
(3 D E Y II R A. R E II I ‘V D I! K £3 P

Ggmmmfi?fl£OafmFNQPGHHBGMNWGMfiCflHflCMHflAdflXfiGMflENHKI

mumsaxmmaumoaumxnxmcmnmcmuacmumccunmnnmmaumammun

 

AAMMANUQA

at

H
10° 10° ‘00 10° ‘00 \DO

3...

M

ION 0m GO Ofi #0 MN 00‘ GO 01h non NM

00 0° 0° DO 100

U

38

10000

p
#1” b” UN UH NO U0 U10 ”0 NO Ms) Ms) Nat H01 HUI HO H0 H“

H H H H H
MN om no me #0 NM om no a.
100 I00 ’00 00 I00 100 ‘00

he
I.)
O
O

1440
1500

122

deduced molecular weight of 50,153. Cell-free translation of VCP mRNA in vitro
revealed that the VCP precursor is a 50-kDa polypeptide (data not shown).
Hydropathy analysis (19, 20) of the deduced amino acid sequence has shown
that VCP exhibits properties typical of a secretory protein. The amino acid
sequence of VCP has only one potential glycosylation site at position Asn135.
Glycosylation of VCP via high mannose, which accounts for about 2 kDa of the
protein molecular mass, was demonstrated previously (12).

In other insects, small extraovarian proteins are also involved in formation
of yolk protein reserves. Microvitellogenin (31 kDa) of Manduca sexta and
Hyalophora cecropia, and the 30-kDa protein of Bombyx mori are produced by
the fat body and deposited in oocytes (21-23). The nucleotide sequence of
Manduca microvitellogenin cDNA has 70% similarity to the cDNA sequence
coding for the Bombyx 30-kDa protein, indicating a close evolutionary
relationship between these proteins (24-26).

Comparison of amino acid sequences has not revealed any similarity
between mosquito VCP and either Manduca microvitellogenin or Bombyx 30-
kDa protein. Unexpectedly, the VCP sequence exhibited significant homology
with members of a family of serine carboxypeptidases (Table 1). The homology
between the amino acid sequences of mosquito VCP and these
carboxypeptidases is the highest at the N-terminal portion which includes two
conserved domains (27-32). The highest identity (28.1%) was found in a stretch
of 392 amino acids of VCP and wheat carboxypeptidase Y homolog (WCP) (Fig.
4). However, considering both identical and conservative replacements,
similarity between VCP and WCP is 62.3%. In mosquito VCP the conserved
domains are at positions Trp11o - Ser11g and Gly205 - Gly21o (Fig. 4).

123

Table 1 . Homology of mosquito VCP to serine carboxypeptidase

Protein Similarity
Value
62. 3 .

 

     

   

 

 

 

 

 

 

. Wheat Carboxypeptidase Y
Homolog
I Mouse Protective Protein 434 54.8 45.6
Human Protective Protein 422 59.2 48.2
Barley Carboxypeptidase l 394 42.9 56.7
Yeast Carboxypeptidase Y 383 43.2 31.4
Barley Serine 380 49.1 35 .4
Carboxypeptidase ll
Yeast KEX1 339 40.0 42.8
Carboxypeptidase

 

 

 

 

 

 

Only optimized score are presented in this table; 2 value is a statistical
significance of optimized scores (2 > 10 indicates statistical significance)

124

Fig. 4. Alignment of amino acid sequences of mosquito VCP and wheat
carboxypeptidase Y homolog (WCP).

Amino acid sequences deduced from corresponding cDNAs were aligned by
FASTA computer program. Vertical lines indicate identical residues and colons
denote functional substitutions. Only portions of both amino acid sequences,
flanking the overlapping regions, are presented. Two conserved domains found
in sequences of serine carboxypeptidases are shaded. The serine protease

catalytic center is marked by an asterisk.

VCP

WCP

VCP

WCP

VCP

WCP

VCP

WCP

VCP

WCP

VCP

WCP

VCP

wcp

VCP

WCP

125

LWNPYKKLMRGSASPRRPGESGEPLFLTPLLQDGKIEEARNKARVNHPNLSSVESYSGFM

I. I
33 .33 . 3 3 3 33333 313

FPGAQAERLIRALNLLPGRPRRGLGAGAEDVAPGQLLERRVTLPGLPEGVGDLGHHAGYY

 
        

EYSWHQNHHMIYIDNPVGTGPSFTDSDEGYSTNEEHVGENLMKPIQQFFVLFPNLLKHPF

I IIIII I I I I I I
333.33 33' 33. 33.....3333. 3 3 3.3 I333. 3.3. I. ”33 .
KFGWDKISNIIFVDPATGTGFSYSSDDRDTRHDEAGVSNDLYDFLQVFPKKHPEPVKNDF

AFASRVHQGNKKNEGTHINLKGFAIGNGLTDPAIQYKAYTDYALDNN

 

LI---DLNGRKKFDEDTAAAIA-CAERKDMKC--ANRLIQGLFDGLDGQESYFKKVTGFS
I I I I I
. 3 3. 3 .l I33 3 3. . 33 333' 333 33331 3
LIQXADYDRINKFIPPCEFAIKLOGTDGKASCMAAYMVCNSIFNSI ------- NKLVGTK

SYYNFIXGDEE--SKQDSVLMEFLSNPEVRKGIHVGELPFHDSDGHNKVAEMLSEDTLDT
3:}: I3 :3 3 3 I I 33333 31:33: I}: I 333 3: 3 3 3: 3 3

NYYDVRKECEGKLCYDFSNLEKFPGDKAVRQAIGVGDIEFVSCS--TSVYQAMLTDWMRN

VAPWVSKLLSH-YRVLFYNGQLDIICAYPMTVDFLMKMPFDGDSEYKRANREIYRVDRKS

33 33 (I33 3I'3I3I: I3I'33 3 33 3} 33:33: 333 3 3 I:

LEVGIPALLEDGINVLIYAGEYDLICNWLGNSRWVHSMEWSGQKDFAKTAESSPLVDDAQ

PGTRSGLVVCKRC

AGVLKSHGALSFLKVHNAGHMVPMDQPKAALEMLRRFTQGKLKESVPEEEPATTFYAA

81
91

141
150

201
210

257
270

311
323

369
381

428
441

441
499

126

The Ser207 of the VCP second domain corresponds to Ser257 of yeast
carboxypeptidase Y which was shown to be the catalytic center of serine
carboxypeptidases (33). In contrast to mosquito VCP and WCP (Fig. 4), all other
members of the serine carboxypeptidase family have three conserved domains
(27-32). Functional implications of these differences are not clear. Unlike serine
endopeptidases, the importance of three domains for carboxypeptidase activity
has not been established (34). Similar to all other serine carboxypeptidases in
the family (27-32), the amino acid sequence of mosquito VCP does not have any
significant homology with any known sequences of trypsin-like serine
endopeptidases or esterases.

lnternalization of VCP in mosquito oocytes occurs without any changes in
its size (12). The molecular mass of VCP, however, decreases by 0.5-1.0 kDa at
the onset of embryonic development (Fig. 5). As embryogenesis progresses,
VCP undergoes further reduction in its size and another, immunologically related
48-kDa band appears (Fig. 5). By the end of embryonic development, VCP
degrades into smaller peptides which disappear once the first instar larva
hatches (Fig. 5).

Radiolabeled serine protease inhibitor, [3H]DFP, which binds to the active
center of serine proteases (35), binds weakly to VCP in oocytes (Fig. 5; lanes 1).
The intensity of binding to VCP, however, increases at the onset and reaches
maximum in the middle of embryogenesis, when the inhibitor binds to a VCP
band of 48 kDa (Fig. 5). The binding of DFP to VCP can be inhibited by prior
treatment of protein extracts with other serine protease inhibitors,
phenylmethylsulfonyl fluoride or leupeptin (date not shown). Based on above
results, we conclude that VCP is synthesized by the fat body and internalized by
oocytes as a pro-enzyme which is then activated in eggs at the onset of

em bryogenesis.

127

Fig. 5. lmmunoblot (A) and [3Hl-DFP binding (B) analyses of VCP during
mosquito embryonic development

The lanes in both panels contain protein extracts from: 1, mosquito ovaries
removed 24 hr post blood meal during peak yolk accumulation; 2, 0-3 hr post
oviposition eggs at the onset of embryogenesis; 3, 44-47 hr post oviposition
eggs during mid embryogenesis; 4, 94-97 hr post oviposition eggs at the end of
embryogenesis; 5, first instar larva. In both figures, the mosquito VCP bands are
shown by arrows. The high molecular weight polypeptide bound to DFP (Fig.
4B) is not immunologically related to VCP. In both panels, proteins were
resolved by SDS-PAGE on 9% gels under reducing conditions. The molecular
weight standards in order of decreasing Mr are phosphorylase B, bovine serum

albumin, ovalbumin and carbonic anhydrase (Bio-Rad).

128

.m31

«can

¢sol

 

an...

 

 

.5

 

 

 

129

The activation of VCP is associated with the increase in its electrophoretic
mobility. It is not clear, however, whether this increase in VCP mobility is due to
its proteolytic cleavage or deglycosylation. The mouse carboxypeptidase
(protective protein) is activated as a result of cleavage and DFP binds only to the
activated carboxypeptidase subunit which contains the serine catalytic center
(31,32).

Some of the serine carboxypeptidases with which VCP shares significant
homology, such as human and mouse protective proteins and yeast KEX1
carboxypeptidase, are implicated in proteolytic activation of a number of
enzymes or other biologically active molecules (29-32). The mosquito VCP
could play a similar role by activating hydrolytic enzymes which are involved in
degradation of yolk proteins in developing embryos. Alternatively, it could
function as an exopeptidase in sequential degradation of vitellogenin.

The mosquito VCP is different from proteases known to hydrolyze yolk
proteins in insect embryos: an acidic cathepsin B-like protease of Drosophila, a
thiol protease or cathepsin L-like protease of Bombyx(8-11). Similar to mosquito
VCP, all these proteases are deposited as pro-enzymes in yolk bodies of eggs
and activated in embryos. Drosophila cathepsin B-like protease is activated as a
result of proteolysis by a serine protease (Mr=25,000) which is also present in
egg yolk bodies (9). It is not known, however, whether any of these enzymes
are of extraovarian origin or whether they are synthesized by developing oocytes
themselves. The mosquito VCP, therefore, is the first example among oviparous
animals of a proteolytic enzyme produced by an extraovarian tissue and

accumulated by oocytes for use in embryonic development.

130

ACKNOWLEDGEMENTS

We thank Dr. T. S. Dhadialla and Mr. A. R. Hays for their advice in protein
purification, Dr. H. H. Hagedorn for his generous gift of vitellogenin gene probe
and Drs. T. S. Dhadialla, T. B. Friedman, J. R. Miller and N. V. Raikhel for their
critical reading of the manuscript. This work was supported by grants from the
National Institutes of Health and from the Biotechnology Center of the Michigan

State University to ASR.

131

REFERENCES

1. Kunkel, J.G. 8 Nordin, J.H. (1985) In: Comprehensive Insect Physiology,
Biochemistry and Pharmacology, eds. Kerkut, GA. 8 Gilbert, L.I.
(Plenum, New York), Vol. 1, pp. 83-111.

2. Wyatt, GR. (1988) Can. J. Zool, 86, 2600-2610.

3. Dhadialla, T.S. 8 Raikhel, AS. (1990) J. Biol. Chem, 265, 9924-9933.

4. Kanost, M.R., Kawooya, J.K Law, J.H., Ryan, R.O., Van Heusden, MC, 8
Ziegler, R. (1990) Adv. Insect Physiol, 22, 299-396.

5. Wahli, W. (1988) TIC 4, 227-232.

6. Ferenz, H.-J. (1990) In: Molecular Insect Science, eds. Hagedorn, H.H.,
Hildebrand, J.G., Kidwell, MG. 8 Law, J.H. (Plenum Press, New York and
London), pp. 131-138.

7. Raikhel, AS. 8 Dhadialla, TS. (1992) Annu. Rev. Entomol, 37, 217-251.

8. Medina, M., Leon, P. 8 Vallejo, CG. (1988) Arch. Biochem. Biophys., 263,
355-363.

9. Medina, M. 8 Vallejo, CG. (1989) Insect Biochem., 19, 687-691.

10.Indrasith, L.S., Sasaki, T. 8 Yamashita, O. (1988) J. Biol. Chem, 263, 1045-
1051.

11.Kageyama, T. and Takahashi, S. (1990) Eur. J. Biochem., 193, 203-210.
12.Hays, AR. 8 Raikhel, AS. (1990) Roux's Arch. Dev. Biol, 199, 114-121.

13.Raminani, L.N. 8 Cupp. E.W. (1978) Int. J. Insect Morphol. Embryol, 7, 273-
296.

14.Sambrook, J., Fritsch, E.F. 8 Maniatis, T. (1989) In: Molecular Cloning: A
Laboratory Manual, Second Edition (Cold Spring Harbor Lab., Cold Spring
Harbor, NY).

15.Lee, C.C., Wu, X., Gibbs, R.A., Cook, R.G., Muzny, D.M. 8 Caskey, CT.
(1988) Science, 239, 1288-1291.

16.Cho, W.-L., Dhadialla, T.S. 8 Raikhel, AS. (1991) Insect Biochem., 21, 165-

132

176.

17.Bose, S.G. 8 Raikhel, AS. (1988) Biochem. Biophys. Res. Commun., 155,
436-442.

18.Lipman, DJ. 8 Pearson, W.R. (1985) Science, 227, 1435-1441.
19.Kyte, J. 8 Doolittle, RF. (1982) J. Mol Biol, 157, 105-132.
20.von Heijne, G. (1983) Eur. J. Biochem., 133, 17-21.

21.Kawooya, J.K, Osir, ED. 8 Law, J.H. (1986) J. Biol. Chem, 261, 10844-
10849.

22.Zhu, J., Indrasith, L.S. 8 Yamashita, O. (1986) Biochim. Biophys. Acta., 882,
427-436.

23.Kulakosky, RC. 8: Telfer, W.H. (1987) Insect Biochem., 17, 845-858.

24.Sakai, N., Mori, S., lzumi, S., Haino-Fukishima, K, Ogura, T., Maekawa, H. 8
Tomino, S. (1988) Biochim. Biophys. Acta, 949, 224-232.

25.Wang, X.-Y., Cole, KD. 8 Law, J.H. (1988) J. Biol. Chem, 263, 8851-8855.
26.Wang, X.-Y., Cole, KD. 8 Law, J.H. (1989) Gene, 80, 259-268.

27.Baulcombe, D.C., Backer, RF. 8 Jarvis, MG. (1987) J. Biol. Chem, 262,
13726-13735.

28.Sorensen, S.B., Svendsen, LB. 8 Breddam, K. (1987) Carlsberg Res.
Commun., 52, 285-295.

29.Dmochowska, A., Dignard, D., Henning, D., Thomas, BY 8 Bussey, H.
(1987) Cell, 50, 573-584.

30.Valls, L.A., Hunter, C.P., Rothman, J.H. 8 Stevens, TH (1987) Cell, 48, 887-
887.

31.Galijart, N.J., Gillemans, N., Harris, A., van der Horst, G.T.J., Verheijen, F.W.,
Galjaard, H. 8 d'Azzo, A. (1988) Cell, 54, 755-764.

32.Galijart, N.J., Gillemans, N., Meijer, D. 8 d'Azzo, A. (1990) J. Biol. Chem,
265, 4678-4684.

33.Hayashi, R., Moore, S. 8 Stein, W.H. (1973) J. Biol. Chem, 248, 8366-8369.

133

34.Wilson, KP., Liao, D.-l., Bullock, T., Remington, SJ. 8 Breddam, K (1990) J.
Mol. Biol, 211, 301-303.

35.Scheiner, OJ. 8 Ouigley, JP. (1982) Anal. Biochem., 122, 58-69.

CHAPTER 5

SUMMARY AND FUTURE RESEARCH PROSPECTS

I34

135

1. Mosquito lysosomal aspartic protease:

The termination stage of mosquito vitellogenesis involves the cessation of
V9 gene expression and the dramatic increase of lysosomal activity. This
lysosomal activity may interrupt Vg secretion by degrading the Vg-containing
granules and cause trophocyte remodeling by autophagocytosis of biosynthetic
machinery (Raikhel, 1986a, 1986b). In my study, one of the lysosomal enzymes,
aspartic protease, was purified. The protein characterization and cDNA
sequence analysis of mLAP reveal that mLAP shares similarity to both
cathepsins D and E as listed in Table-1. The result implies that mLAP may be
similar to the ancestral protein of vertebrate cathepsins D and E. During
evolution, this protein lost a phosphorylation determinant resulting in a different
subcellular distribution than lysosomal enzymes. In contrast, an insertion was
introduced causing an additional processing step in vertebrate cathepsin D.

During the termination stage of vitellogenesis, mLAP levels rise 10 fold
over the levels of the synthetic stage. Based on the profiles of mLAP mRNA
(Fig.3 of Chapter 3) and 20-HE concentration (Fig.2 of Chapter 1), we postulated
that 20-HE may regulate mLAP gene expression. This hypothesis will be
investigated at the transcriptional level by using in vivo and in vitro systems.
Microsurgical removal and the culture of mosquito fat body will be used to
analyze the effect of 20-HE on mLAP transcription. Moreover, when the
upstream region of mLAP gene is cloned, the prospective promoter and
enhancer regions will be cloned in pCAT-Enhancer vectors and pCAT-Promoter
vectors for functional analysis of regulatory elements in eukaryotic cell lines.
(Rosenthal, 1987). The position and the size of the cis- regulatory elements of
the mLAP gene will be localized and determined with combinative
oligonucleotide-directed deletions (Wallrath and Friedman, 1992). To identify

putative tissue specific regulatory elements, these sequence will be inserted into

136

 

 

 

 

 

 

 

 

 

 

 

 

Table 1. Comparison among mLAP, HCD and HCE.
l mLAP HCD HCE
- Native M, 80 kDa 47 kDa 76 kDa
. a M, of subunits 40 kDa 14 kDa 8t 31 kDa 38 kDa
: (Homodimer) (Dimer)
n Optimal reaction pH 3.0 3.5 3.0-3.2
I i lnsertional cleavage No Yes No
~ ‘ sites
i , N-linked 1 2 1
l l Glycosylation
j ‘ Isa-enzymes No Yes (3) Yes (2)

7 I pl 5.4 5.7; 6; 6.6 4.1-4.6

8 : Inhibition by 10 mM No No Yes

i KCN
9 Inhibition by 6 M Yes Yes No
? urea
10 Localization Lysosome Lysosome Cytosol or
membrane
bound

g 11 I Distribution Ubiquitous Ubiquitous Tissue specific
' 12 92% 81%

 

 

' Amino acid similarity to mLAP

 

 

137

P-elements containing a reporter gene, lac Z, for D. melanogaster gerrnline
transformation (Wallrath et al., 1990). The expression can be analyzed by the
color response caused by the 8-galactosidase activity in dissected tissue from
transformed flies. If the regulatory elements linked to the upstream region of the
lac Z gene are functional in the specific tissue, the tissue will turn blue when it is
stained with X-gal .

The increased translation of mLAP mRNA is postponed for 12 hr after
mRNA levels increase. We have found putative steroid hormone regulation
elements in the 5'-untranslated region of mLAP mRNA. This finding suggests
that 20-HE may inhibit the translation of mLAP mRNA between 12 hr and 24 hr
PBM. This regulation of mRNA translation by a steroid hormone is different from
the receptor-mediated regulation of gene expression. It demonstrates that a
steroid hormone can regulate the transcription in the nucleus but may also be
involved with the translation in the cytosol (Verdi and Campagnoni, 1990). In our
case, 20-HE may interact with the 5' untranslated region of mLAP mRNA to
inhibit or decrease the translation. To study this mechanism, the approach used
by Verdi and Campagnoni (1990) will be adopted. The mosquito in vitro
translational system will be prepared from a mosquito cell line (Gillies and
Stollar, 1981). The cDNA corresponding to the 5' untranslated region of mLAP
mRNA will be ligated to the upstream coding region of CAT, and inserted into a
pBluescript phagemid. With this construct, the fusion RNA can be generated in
an in vitro system with T7 RNA polymerase. To increase the stability of mRNA
and enhance the efficiency of translation, mCAPTM RNA Capping kit from
Stratagene will be used. The translational regulation of 20-HE on the 5'
untranslated region of mLAP mRNA will then be analyzed using a mosquito in
vitro translation system by measuring CAT activity without the endogenous

interference.

138

In vertebrates, the targeting of lysosomal enzymes from their site of
synthesis in RER to lysosomes is mainly directed by mannose-6-phosphate
receptors (cation dependent and cation independent receptors) which recognize
the mannose-6-phosphate marker exposed on lysosomal enzymes (Kornfeld,
1987). Lysosomal enzymes contain common protein determinants that are
recognized by lysosomal enzyme N-acetylglucosamine-1-phosphotransferase
and UDP-GlcNAc (UDP-linked N-acetylglucosamine) in the formation of
mannose-6-phosphate residues (Baranski et al., 1990). Two determinants for
mannose-6 phosphorylation similar to those of human cathepsin D are found in
the mLAP amino acid sequence. The cDNA for an insect lysosomal enzyme
which is well characterized at the protein level, can be used to investigate this
sorting mechanism in insect cells. The first step is to analyze whether mLAP is
phosphorylated at its glycan. Next we need to understand whether the N-linked
phosphorylated glycan is essential for transport of mLAP to lysosomes. To do
this, polyclonal antibodies which can recognize the native form of mLAP are
required. With the purification protocol that I developed for mLAP, it is difficult to I
isolate enough mLAP for antibody production. The coding sequence of mLAP
cDNA will be cloned into a baculovirus expression vector under the control of the
polyhedrin promoter. The overexpressed protein therefore can be purified easily
for polyclonal antibody production. Before a serious of analysis, a glycosylation
inhibitor, tunicamycin, will be used to examine whether it can interrupt the
transport of mLAP. Furthermore, to verify that mLAP is phosphorylated at the
oligosaccharide chain, mLAP will be labeled with [32P] orthophosphoric acid in
fat bodies cultured in vitro and then mLAP will be immunoprecipitated. If mLAP
is phosphorylated at the glycan only, the treatment of endoglycosidase H will
remove the radioactivity from mLAP. The result can be simply examined with
SDS-PAGE. If the peptide back bone of mLAP is phosphorylated too, than high

139

performance liquid chromatography (HPLC) has to be used for detecting the
[32F] labeled oligosaccharide. If mLAP is proved to be phosphorylated at the
oligosaccharide, site-directed mutagenesis will be used to change the potential
N-linked glycosylation residue (Asn-70). The mutated and wild type cDNAs will
be constructed into plasmids containing an inducible Drosophila hsp70 promoter
respectively. The DNA prepared from these constructs will be used to transfect
a mosquito cell line, As. albopictus, for transient expression (Durbin and Fallon,
1985). The difference of mLAP activity and mLAP protein detected in culture
medium between wild type and mutant DNA transfections will elucidate the
importance of phosphorylated glycan in insect lysosomal targeting. Alternatively,
the mutated cDNA and wild type mLAP cDNAs can also be subcloned into a
baculoviral vector under the control of early gene promoter, lE-1, or even the
polyhedrin promoter for expression in an insect cell line (Steiner at al, 1988;
Hammock et al., 1990 ). The amount of secreted mLAP proteins and the mLAP
activity measured in culture medium will allow us to understand the role of
phosphorylated mannose(s) in targeting of insect lysosomal enzymes. More
investigations of insect lysosomal enzyme trafficking can be achieved by
methods as described in Baranski et al (1990 and 1991).

2. Mosquito vitellogenic carboxypeptidase:

Cloning and sequencing of an additional vitellogenic protein, VCP, have
confirmed that it is synthesized extraovarially in a sex-, tissue- and stage-
specific manner. Even more importantly , analysis of its cDNA sequence has
lead to a discovery of a novel biological phenomenon previously unknown for
any oviparous animals. It is the first observation that an enzyme, participating in
embryonic development, is synthesized outside of the developing oocytes.

In the mosquito, VCP is synthesized as a proenzyme and processed to an

140

active enzyme during embryonic development. To analyze this activation
mechanism, it is important to purify the active form of the enzyme from embryos
and study its enzymatic properties. Furthermore, the subcellular localization of
VCP in embryos and the substrate determination for VCP are also essential for
the understanding of its physiological function in embryonic development.

The synthesis of VCP in fat bodies is controlled by 20-HE (Hays and
Raikhel, 1990). Further analysis of the regulatory mechanism depends upon
cloning of the corresponding gene(s). VCP is a tissue- sex- and stage- specific
expressed gene(s). Therefore, the regulatory elements of VCP gene(s) will be
constructed in plasmids containing CAT or luciferase genes. DNA prepared from
these constructs will be subjected to in vitro transcription with nuclear extract
prepared from mosquito fat bodies 24 hr PBM with a method modified from
Drosophila extract (Heiermann and Pongs, 1985; Kamakaka et al., 1991). The
transcripts will be used for in vitro translation with rabbit reticulocyte lysate. The
functional regulatory elements will be detected from enzymatic activities of CAT
or luciferase. In addition, those elements inserted into P-elements containing a
reporter gene, lac Z, for D. melanogaster embryo transformation as described in
mLAP project will be also an alternative tool for us to examine the sex-, tissue-
and stage- specific regulations of VCP gene(s) in Drosophila (Wallrath et al.,
1990)

The cornerstone of mosquito reproduction is vitellogenesis. Mosquito
LAP and VCP are essential proteases involved in vitellogenesis. Therefore, to
fully understand the regulation of these two proteins at the molecular level will be
a potential tool for the development of strategies for the interruption of

vitellogenesis or egg development in the mosquito population.

141

REFERENCES

Baranski, T. J., Faust, P. L., and Kornfeld, S. (1990) Cell, 63, 281-291.

Baranski, T. J., Koelsch, G., Hartsuck, J. A., and Kornfeld, S. (1991) J. Biol.
Chem. 266, 23365-23372.

Durbin, J. E., and Fallon, A. M. (1985) Gene 36, 173-178.
Gillies, S., and Stoller, V. (1981) J. Biol. Chem, 256, 13188-13192
Hays, A. R., and Raikhel, A. S. (1990) Roux's Arch. Dev. Biol, 199, 114-121.

Hammock, B. D., Bonning, B. C., Possee, R. D., Hanzlik, T. N., and Maeda, S.
(1990) Nature 344, 458-461

Heiermann, R., and Pongs, O. (1985) Nucl. Acids Res. 13, 2709-2730.

Kamakaka, R. T., Tyree, C. M., and Kadonaga, J. T.(1991) Proc. Natl. Acad. Sci.
88, 1024-1028.

Kornfeld, S. (1987) FASEB J. 1, 461 -468.

Raikhel, A. S. (1986a) J. Insect Physiol. 32, 597-604.
Raikhel, A. S. (1986b) Tissue 8 Cell, 18, 125—142.
Rosenthal, N. (1987) Math. Enzymol. 152, 714-720.

Steiner, H., Pohl, G., Gunne, H., Hellers, M., Elhammer, A., and Hansson, L.
(1988) Gene 73, 449-457.

Verdi, J. M. , and Campagnoni, A. T. (1990) J. Biol. Chem, 265, 20314-20320.

Wallrath, L.L., Burnett, J. B., and Friedman, T. B. (1990) Mol. Cell. Biol. 10,
5114-5127.

Wallrath, LL, and Friedman, T. B. (1992) Biotech. 12, 214-216.