.fifififiuflmfigfimfi

lpofi oil 0"
\v v

Pa

8.
. 1 . #Mflrm.:fi5u
.v . if“!
.nyIMJfi .

«Ul‘mm. Jn.
“fly.“hhfivir.
I ~03“: .n. l l,”
RE; 3%.... $6.. fl.

5. .. r!

it: A
25...“; : ..
;

51..
oralrc).
J31}?
. 4': v o A

. 0D
‘1 I.
vii]
y‘tl'
. .. .l’lll’l
.... A:
.. 2.. a 2.
~ :.

 

 

0.!7’.‘.ly. .. vll‘ifiuf 1“.
.Q'vtaillll: O’ltvl‘: t 01)..”-
.lo 1. . 31...: AI! '0 Ointlrl CID... "L
:11: III}! trialloltr ‘.90.::l.ufnz .1 . 1-.52 7 v ..
. .... o .r‘.... 5...! . .
u . . . ,
HQ
2 v V.: ,t
vnti I:

. S. .I «I:
- , , ‘ is... .I 31!. gt
. . it. I. u v 1:? .4 .. . .. 43 I Ibrb‘ ‘ i
u in. a it :1.v.v )rifl‘fi.l.n.l . . a g
a I. .u V 3 . .11..“‘53' .315... ‘51.. n..\ V .
v..- , J.» ti. l,‘..ln....-5n.€li'. I o Fungi. » .gfig
flaring .. é. ass-u. ESE. gaining .553...

 

 

THESIS

LIBRARIES

W "

MICMGAN STATE uNNERS"Y \

l\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

3 1293 00793 7

This is to certify that the

dissertation entitled
CHHQFICTEQIZGT/ON OP Acyt. «24411157;
PQOTEIN GENE! Filo/'1 3.4.”.Eia9PiL—i Mflflfi/fi
Q.) Hé‘yA/H. mes COLD/78!")
presentedby

flLe'N/(F) HLouse‘r-QAAq/ac

has been accepted towards fulfillment
of the requirements for

PA' ‘0 degreein 09”. 0F ROTH/UV 19M)
flm/vr PATHOLOGV

 

 

\(‘w

Major professor Q‘ 7
Date 7/? {A V
I r

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

 

 

 

 

V—W

 

k

LIBRARY

Michigan State

University

_.___.

 

_/

PLACE IN RETURN BOX to remove this checkout frorn your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE ll

 

'étL—Ls—qggb

 

 

 

 

 

 

LA
:

 

 

 

 

 

 

 

fi—j

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

——l

MSU I. An Affirmative ActiorVEquel Opportunity Inditution

1

CWMHJ

 

CHARACTERIZATION OF ACYL CARRIER PROTEIN GENES FROM
ARABIDOPSIS THALIANA (L.) HEYNH. var. COLUMBIA

by

Alenka Hloueek-Radojcic

DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Botany and Plant Pathology

1992

ABSTRACT

CHARACTERIZATION OF ACYL CARRIER PROTEIN GENES FROM
ARABIDOPSIS THALIANA (L.) HEYNH. VAR. COLUMBIA

BY

Alenka Hlousek-Radojcic

Multiple ACP isoforms have been characterized from Arabidopsis thaliana
(L.) Heynh. var. Columbia. Immunoblot analysis and radiolabeling of ACPs
via acylation of ACPs in crude protein extracts revealed the presence of
a minimum of one tissue-specific and at least three constitutively
expressed ACP isoforms in Arabidopsis leaves, roots, and developing and

dry seeds.

Two ACP clones, ACPI and ACPZ have been isolated and characterized from
Arabidopsis genomic and cDNA libraries, respectively. The genomic clone
ACPI encodes a 137 amino acids long preACP-l, having a 54 amino acids long
transit peptide. Based on comparisons with the Brassica campestris cDNA
ACP clone, ArabidOpsis ACPI has three introns. Intron I is located in the
transit peptide coding region, whereas introns II and III are located in
the coding region for mature ACP-l between amino acids 1 and 2 and 42 and
43, respectively. The Arabidopsis ACPZ cDNA clone encodes a protein with
a 52 amino acid transit peptide and an 83 amino acid mature protein. ACPI
and ACPZ have 70% identity at both nucleic acid and amino acid levels.
DNA blot analysis indicated that Arabidopsis has a minimum of three ACP

genes. Based on RNA blot analysis, the ACPI and ACPZ genes are expressed

in leaves, roots, and developing seeds.

Expressing coding regions for mature ACPs in Escherichia coli, three
Arabidopsis ACP isofonms were partially purified on 03-52 columns and
identified by immunoblotting with antiserum to spinach ACP-I. Even the
Arabidopsis ACP-2 and ACP-3 isoforms that differ in a single amino acid,
were resolved by native and denaturing 1M urea PAGE. ACP-l, ACP-Z and
ACP-B were found to comigrate with different Arabidopsis ACP isoforms
found in leaves, roots, and seeds, suggesting that all three of them are

constitutively expressed ACP isoforms.

To peace.

ACKNOWLEDGMENTS

I would like to thank to Dr. J.B. Ohlrogge for having made my work and
stay exciting. His support, advice and incredible amount of patience and
understanding were invaluable. Thanks to all the members of the
Ohlrogge's lab from the past and present; their support made my life so
much easier. very special thanks to Dusty, Jan, Katherine and Paul, who
were with me even when the exciting results were replaced by big

disappointment.

My thanks also go to the members of my guidance committee, Drs. Barbara
Sears, William Smith and Chris Somerville for their continuous support
during my stay as a graduate student. Thanks to Chris Somerville for
giving me the opportunity to feel the excitement of being briefly a part

of his group.

I would like to thank my parents, Emilija and Zdenko, and to my brother,
2vonimir, for their love and long-distance support. I would like to give
a special thanks to Mirna for the wittiness and joy, and to Alan for his
smiles and relatively peaceful nights. Finally, my greatest appreciation

goes to Zlatko for his constant love and encouragement.

vii

TABLE OF CONTENTS

Page
Li.t Of T‘hl.'. O O O O O O O O O O O 0 O O O O O O O O O O O O 0 O O 0 O O O O O O O O O O O O O O O O O O O O O O O O O 0 v
List of Figures. . . . . . ............................................... . iv

Li.t Ofnt.v1at10n'000000......OOOOOOOOOOOOO00.0.00... ...... 00...... ii

...:

Chapter one: Introduction................... ....... .. ..... .. .........
Fatty acid synthesis...............................................
Functions of ACP in lipid biosynthesis.............................
Functions of ACP other than fatty acid biosynthesis... ........ .....
Biochemical characteristics and structure of ACPs..................
Subcellular localization of acyl carrier proteins........ ...... .... 12
Isoforms of ACPs and their intercellular localization.... ....... ... 14
Molecular biology of acyl carrier proteins......................... 16
Bibliography....................................................... 20

\iU'IH

\0

Chapter two: Isolation and characterisation of an ACP cDMA and a
genomic clone from Arabidqpsis thaliana............................ 25
Abstract........................................................... 25
Introduction.................................... .......... ......... 25
Experimental procedures.... ............. ... ......... . ......... ..... 28
Results............................................................ 37
Discussion......................................................... 57
Bibliography....................................................... 64

Chapter three: Characterization of Arabidopsis acyl carrier proteins. 68
Abstract........................................................... 68
Introduction....................................................... 69
Experimental procedures..................... .................. ..... 71
Results......... ................................................. .. 82
Discussion................. .......... ........... ............ ....... 95

Chapter four: Summary and Perspectives...............................103
Summary.............................................. ..... .........103
Perspectives.............. ....................................... ..104
Bibliography.......................................................108

AmndixOOOOOO.....OOOOOOOOOOOO0.00.00.00.00.0.0.0.0000...00.0.0.0...110

vi

LIST OF TABLES

Page

Table 1: DNA percentage identity of the transit peptide and
mature ACP coding regions................................ 49

Figure

Figure

Figure
Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

10

11

12

13

14

15

LIST OF FIGURES

Page
Schematic representation of fatty acid biosynthesis in

planta.’0....00......0.00..................OOOOOOOOOOOOOOOOO 2

Schematic representation of the phosphopantetheine binding
region of acyl carrier protein.............................. 4

Comparison of the amino acid sequences of mature ACPs....... 10
Restriction map of pAD4 and sequencing strategy............. 39

DNA and deduced amino acid sequence of Arabidopsis ACPI

genQOOOOOOOOOOOOO......OOOOOOOOOOOO......OOOOOOOOOOOO0.0....41
Restriction map of pALlZO-E and sequencing strategy......... 44

DNA and deduced amino acid sequence of Arabidopsis ACPZ cDNA
Clone, PAleo-E.0......OO......OOOOOOOOOOOOOOOOOO0.0.0.0.... 46

DNA blot hybridization analysis of ACP genes in Arabidopsis. 52

RNA blot hybridizations of the total RNA purified from
Arabidopsis leaves, roots and developing seeds.............. 55

Schematic representation of the subcloning strategy for
expression of mature coding regions of Arabidopsis genomic
ACP-loloneinEO caliOOOOOOOOOOOOIOO ...... ......OOOOOOOOOOO 74

Schematic representation of the subcloning strategy for
expression of mature coding regions of Arabidopsis ACP-1 and
ACP-3 inEe caliOOOOOO0.........I.........OOOOOOOOOOOOOOO... 78

Immunoblot analysis of ACP isoforms in Arabidopsis root, seed
and leaf protein extracts resolved by SDS/PAGE.............. 83

Immunoblot analysis of ACP isoforms in Arabidopsis root, seed
and leaf protein extracts resolved by native PAGE........... 85

Immunoblot analysis of ACP isoforms in Arabidopsis root, leaf
and seed protein extracts resolved by denaturing 1M urea

PAGEOOOO0.0......00.0.00...00.00.000.000...0.00.00.00.00000086

Immunoblot analysis of DE-52 column fractions from the
purification of mature ACP-l expressed in E.coli and
reBOIVBdeSDS/PAGEOO.....OOOOOOOOOOOOOO....OOOOOO0.0...... 88

iv

Figure 16 Purification of Arabidopsis ACP-1 expressed in E.coli
transformedwithpAD4cmOOO............OOOOOOOOOOOOOO0.....0. 89

Figure 17 Purification of Arabidopsis ACP-2 expressed in E. coli
transformedWithNDlzoOOOOOOOOOOOOOOO00.0.00...0.0.0.000... 91

Figure 18 Palmitoylation of Arabidopsis ACP isoforms in root, dry seed
and leaf protein extracts................................... 94

iii

amp 0.000.000.00.00.....
BCIP 0......0....0..00.0

Blotto 000.00.000.00....

5x Denhardt's .. ........

nspc ...................
ddI-Izo
on

EDTA0000.00.0..00000000

GUS 0.0.0.0..00000000000

IPTG 0......O0...0.00..0

L30... ....... 00.00.0000.

LB plates ..... .........
MES ....................
MOPS ...................
NBT ............ ..... ...
NEM ....................
PAGE ...................

PCR 000.00.000.0000...0.

LIST OF ABBREVIATIONS

ampicillin
bromochloroindolyl phosphate

5% (w/v) nonfat dry milk (Carnation Co., Los
Angeles, CA) in TBST

0.1% (w/v) bovine serum albumin, 0.1% (w/v) PVP,
0.1 % (w/v) Ficoll 6000

diethyl pyrocarbonate

distilled water

double distilled water

dithiothreitol

(ethylenedinitrilo)- N,N,N',N'-tetraacetic acid

ethylene glycol-bis(b-aminoethyl ether)-
N,N,N',N'—tetraacetic acid

b-glucuronidase
isopropyl-b-D-thiogalactopyranoside

10 g Bacto-tryptone (Difco), 5 g yeast extract
(Difco), 5 9 NaCl in 1 l ngD

LB broth with 10 g Bacto-agar (Difco)
2-(N-Morpholino)ethanesulfonic acid
(3-[N-Morpholino]propanesulfonic acid
nitroblue tetrazolium
N-ethylmaleimide

polyacrylamide gel electrophoresis

polymerase chain reaction

ii

PEG 0 0 O O O . . 0 0 . 0 0 O . O 0 O 0 O O PelyethYIene glyco1
PVP 0 0 O O O O 0 O O O 0 0 O O O 0 0 O 0 0 DOIYViny1pyrr01 idone
SDS ........ ............ sodium dodecyl sulphate

SM buffer . .0 O... O O... 0. 100 w NQCI' lo M M9804, 50 mM Tris H01, pH 7.5,
0.01 % gelatin

1x SSPE ......OOOOOOO... 0.15 M NaCI’ 0.01 M NaH2P04' 0.001 M EDTA’
disodium salt, pH 7.4.

TBST OOOOOOOOOOOOOOOOOOO lomMTria HCl' pH 8.0, lsomMNaCl' 0.05% (V/V)
Tween 20

TCA .................... trichloroacetic acid
TE 0.00....00...00...... lomMTria 8C1, PH 704’ lmMEDTA

TEA-Ac ................. triethylammonium acetate

CBAPTERONE

INTRODUCTION

Fatty acid synthesis

Fatty acid biosynthesis occurs through the sequential elongation of acyl
chains with two carbon units (52). The initial reactions in fatty acid
biosynthesis start with a transfer of a malonyl group to the
phosphopantetheine arm of acyl carrier protein (ACP), followed by a
condensation reaction with acetyl-CoA (22). The four carbon acetoacetyl
moiety, while still bound to ACP, goes through ketoreduction, dehydration
and enoyl reduction steps (52). For the synthesis of 16 or 18 carbon
fatty acids this four step cycle is repeated seven or eight times,
respectively (Fig.1). Acyl carrier protein, via the thioester bond that
is formed between the carboxyl group of the acyl chain and the sulfhydryl
group of the phosphopantetheine (Fig. 2), ”activates" the acyl group and
carries it from one catalytic domain to the other, thereby functioning as

a cofactor in fatty acid biosynthesis (30).

Although the set of reactions in fatty acid biosynthesis is similar in all
living organisms, the molecular organization of fatty acid syntheses
differs (56). In animals, fatty acid synthase is a multifunctional

protein (referred to as type I) which is composed of seven catalytic

Figure 1. Schematic representation of fatty acid and some steps of lipid
biosynthesis in higher plants. AAT - acetyl-CoAzACP transacylase; ACC -
acetyl-CoA carboxylase; ACP - acyl carrier protein; AT - acyltransferase;
DH - B-hydroxyacyl-ACP dehydrase; DGD - digalactosyl diacylglycerol; DS -
stearoyl-ACP desaturase; ER - enoyl-ACP reductase; GBP - glycerol-3-
phosphate; us I - B-ketoacyl-ACP synthase I; KAS II — fl-ketoacyl-ACP
synthase II; was III - B-ketoacyl-ACP synthase III; KR - B-ketoacyl-ACP
reductase; LPA - lysophosphatidic acid; MAT - malonyl-CoAzACP
transacylase; MGD - monogalactosyl diacylglycerol; PA - phosphatidic acid;
PG - phosphatidylglycerol; SI. - sulphoquinovosyl diacylglycerol; TE -
thioesterase; 16:0-CoA - palmitoyl-CoA; 18:0-CoA - stearoyl-CoA;

 

PLASTID

CYTOPLASM

acteyl-CoA —ACC—> malonyl-00A

16:0-CoA
18:1 -CoA

    

Prosthetic group structure of acyl carrier protein

4'-Wm

 

 

043
l
O—CH2— c- CHOHCO M-iCH.Ci-l.OON-iCi-12<3Hz SH
| l .
ova—9:0 043 3 ‘ ‘W‘
I
o
I
m2
I
/Ala— Aep— N-i - CH -CO— Leu —Asp\)~
9‘“ Ser »\
/ L
o 9/
0 \
" e
/ I.

\

Figure 2. Prosthetic group structure of acyl carrier protein. 4'-
phosphopantetheine group is bound to a serine residue in the middle of the
mature ACP.

5

domains and a domain for ACP. The functional enzyme is present as a

homodimer. Yeast also has a type I organization, but the FAS is composed
of two subunits, a and B. The a subunit contains ACP, B-ketoacyl
reductase and B—ketoacyl synthase domains, whereas the B subunit contains
domains for the remaining five enzymes: the acetyl transacylase, the enoyl
reductase, the dehydratase and the malonyl/palmitoyl transacylase. In
plants and most prokaryotes fatty acid synthase is of the nonassociated
type (type II). Reactions are catalyzed by six separate enzymes and the
protein cofactor, ACP (30). The functional organization of the enzymes in

vivo is unknown.

Functions of ACP in lipid biosynthesis

The principal products of de novo fatty acid biosynthesis in plants are
palmitic and stearic acids (52). However, plant membranes are very rich
in mono- and polyunsaturated fatty acids (58). Therefore, the majority of
palmitic and stearic acids undergo further modifications, some of which
are ACP mediated. McKeon & Stumpf (28) reported that stearoyl-ACP
desaturase, which is involved in the biosynthesis of oleic acid, prefers
stearoyl-ACP over stearoyl-CoA as a substrate. The enzyme is reportedly
found in a soluble fraction (47). Partially purified stearoyl-ACP
desaturase requires soluble ferredoxin, ferredoxin NADPH reductase and
molecular oxygen (28). Based on all these characteristics, the stearoyl-

ACP desaturase is considered to be localized in the chloroplast stroma.

6
In addition to the biosynthesis of fatty acids, plastids are also the site
of biosynthesis of some plastidial membrane glycerolipids (41). However,
only a fraction of acyl groups synthesized in plastids are used as
substrates in the plastidial reactions of glycerolipid biosynthesis. The
remaining portion is transported into the extraplastidial compartment to
enter the cytoplasmic pathway of lipid biosynthesis. Acyl-ACPs entering
the plastidial pathway of glycerolipid biosynthesis are substrates of two
acyltransferases. The glycerol-B-phosphate acyltransferase is soluble
(5) , whereas monoacyl glycerol-B-phosphate acyltransferase is localized in
the inner chloroplast membrane (1). In contrast to cytoplasmic
acyltransferases, both plastidial acyltransferases prefer acyl-ACPs over
the acyl-CoA substrates. In addition, the plastidial acyltransferases
also have different affinities for acyl groups bound to ACP. The
acyltransferase that transfers the acyl chain to the first position of the
glycerol backbone prefers oleoyl-ACP, whereas the second transferase,
which is responsible for the acylation of the second carbon on the
glycerol backbone, favors palmitoyl-ACP (9). As a result of those
specificities, plastidial glycerolipids have a distinct positional
distribution of fatty acids when compared with cytoplasmatically

synthesized glycerolipids (41).

Acyl groups entering the cytoplasmic pathway of lipid biosynthesis must be
transported through the plastid envelope. This process starts with the
hydrolysis of the acyl group from ACP, presumably followed by the
transport of free fatty acid into the cytoplasm. On the outer plastid

envelope membrane, acyl-CoA synthase esterifies free fatty acids to CoA,

7
providing substrates for extraplastidial lipid biosynthesis. Hydrolysis
of acyl-ACPs occurs in plastids by the action of an acyl-ACP hydrolase
(thioesterase) (52). Partial purification of the enzyme from avocado
mesocarp revealed the presence of a hydrolase that in addition to a
preferential use of acyl—ACPs, showed great selectivity toward the chain
length and the level of desaturation. Kinetic studies indicated that the

preferred substrate was oleoyl-ACP (32).

Functions of ACP other than in fatty acid biosynthesis

In addition to the central position of ACP in fatty acid and lipid
metabolism, ACP-like proteins have been detected as important factors in
other reactions in prokaryotes and eukaryotes. Recently, acyl-ACP was
found to be an activating factor in haemolysin A synthesis (20). Based on
the analysis of amino acid sequences, the first 20 amino acids of the so
called, "cytoplasmic activating factor", were identical to E.coli ACP.
The activation of prohaemolysin.A into haemolysin.A, required the transfer
of an acyl group from acyl-ACP to one of the internal amino acids of the
prohaemolysin A. ACP was also found to be an important factor in the
synthesis of membrane-derived oligosaccharides in E.coli (53), although,
both apo and holo-ACPs appeared 'to be <equally functional in these

transglucosylation reactions (54).

ACP-like domains have been detected in polyketide synthases (PKS) (18).

In some organisms polyketide synthases are multifunctional proteins (7),

8
whereas in others they are of a dissociable type (18). ACP domains of PKS
have similar functions to ACPs in FAS, carrying acyl groups from one
catalytic unit to the other. Recently, gene sequence information for the
polyketide synthase from the Saccharopolyspora srythrea has revealed six
repeated motifs whose sequences are very similar to the enzymatic domains
of the PKS and FAS components (39). Each motif was found to contain

coding regions for B-ketoacyl synthase, acyltransferase and ACP, whereas

ketoreductase and dehydrase sequences were found only in some motifs.

Characterization of the nodulation operon region, which is confined to the
symbiotic plasmid in Rhizobium leguminosarum, revealed that an ACP-like
protein is the product of nod? gene (48). The gene is induced by plant
root exudate and nodD protein. Therefore, has been postulated that the
nodF encoded ACP could serve as a cofactor in the synthesis of cyclic

glucans required for nodulation (48).

Acyl ACPs have been found as a part of the peripheral arm of the
NAnflzubiquinone reductase in Neurospora crassa (43) and in bovine heart
mitochondria (42). The ACP isolated from N. crassa was partially acylated
with myristic and hydroxymyristic acid, which has led to speculation that
mitochondrial fatty acid synthase provides some of the mitochondrial
lipids (29). Independent analysis of the samples from bovine heart
mitochondria under reducing and non-reducing conditions revealed that
complex I contained only acylated-ACPs (42). Given that some other
subunits of the mitochondrial respiratory complex have been involved in

reactions other than respiration e.g. complex III in proteolytic

9
processing of mitochondrially imported proteins, it is possible that the
peripheral arm of complex I is involved in mitochondrial fatty acid

biosynthesis.

Biochemical characteristics and structure of ACPs

The first evidence for the existence of ACP in plants was obtained in 1964
by Overath and Stumpf (35). They observed that a soluble, heat stable
factor similar to F.coli ACP was required for high rates of fatty acid
synthesis in extracts of avocado mesocarp. A few years later, Simoni et
al. (49) purified acyl carrier protein from spinach and avocado,
determined the amino acid composition and confirmed its activity in fatty

acid synthesis.

Acyl carrier proteins from plants and prokaryotes are highly similar
proteins (Fig.3). They are 83 - 90 amino acid long, with molecular
weights of around 10,000. All mature ACPs have a high content of acidic
residues and a low content of hydrophobic and aromatic residues. As a
result of their amino acid composition, ACPs are acidic proteins and
therefore are highly soluble. They also bind low amounts of SDS and
migrate»on SDS-PAGE gels differently from their expected.molecular'weight.
ACPs have a low isoelectric point (3.9-4.2) and are stable to heat

treatment (30).

10

Figure 3. Comparison of the amino acid sequences of plant, animal and
prokaryotic ACPs. Conserved amino acids surrounding phosphopantetheine
binding site are marked with capital letters. The 4'-phosphopantetheine
group is attached to a serine residue (*) near the middle of the
polypeptide chain. A.th.- Arabidopsis thaliana (L.) Heynh. var. Columbia;
B.n.- Brassica napus; B.c.- Brassica campestris; S.o.- Spinacia oleracea;
H.v.- Hordeum vulgare; Z.m.- Zea mays; Cy. sp.- Cylindrotheca sp.; Cr. -
Cryptomonas; A.v.- Anabaena variabilis ; Rh.m.- Rhizobium meliloti; R.sp.-
R. leguminosarum; S.e. - Saccharopolyspora erythraea; E.coli- Escherichia
coli; N.c.- Neurospora crassa ; B.t.- Bos taurus; c.l.- chicken liver. cp
- chloroplastidial; mt - mitochondrial; monocot - monocotyledonous plants;
dicot -dicotyledonous plants; prokaryot - prokaryotes.

11

s.>oenx_eee—ouxee.0..xc......ebt—E>.n>o.e.no.>U>OEJeO.OJOe.eeeec_eesn.>o..cesee).0.ceouee
e>>exxee.>o>.eoeflt.xeen.ua.eeo.eo!tli..U>OOJeO.OJo¢E.eec>e.xeeo.rom.x_s>.).nx.ee.-_eeeue

cevnoe..mesexneee.ouenxun.e.e.eee.IE>>U>hOJe0.0Joce.ze—e—.cxnoc>xnaoe.>ue..e.eenxx.‘ooee

eatoc.>u_eea>—~_leeeeun.e~u.eee.li..U>hOJeOIOJve>.eICC—eeeox>o_oeo..18).....e
eeeeexoeeeeu..e>.o_~.!t>ro>_e.oe_oeoe>ntoseoe_ee>.u>.04eOEUJpe..eeee—.oeee.o.oee.>oe.se..ex.Pt
anon..o_...-oua.o.o.oee.aE,.w>»04¢0¢OJee.....eusxeoo>e.coo>.x:>.eo.e.
eoexe.sleeno>._.eoeevoa.e>o.eee.’t>.u>h04e0e04pu..eeoeeexee0>o.ee.>.xxesee.eeat.
u..o.>ee>>oo.cc_ce3eue~:6e.x.o>.¢e_93.>9e.¢4e0.04e..epo—u.eo.e>ea.eoceex>.xc.ee..e.u.colt
s.«peee0)ou.ereeDn_e.t.eee_IE).U>»OJeOeOJDce.eeOIu>ounceee.ue.s.18):..aeeoe

ee>xoe.evaevc.ce.oeeeooe.e.¢.eee.IE>.U>hOJeOQOJoce.cevxuscexv>o.oee._.U>80e_e¢ec
xen.t¢e.>c.ene>a—.xeemzue.e.c.e>e.2E..U..OJeOeOJUxO.teoe.sxeeoee.fleee.xv.s1...le

exoes.xe._eeeue>-e..aeeeee>e.e.cee.eE>.u>h04eOIOJeeoxeeouaearea-e.Uxxs.eo>xo~.'xxee
eexee>.zn._«eaves-e.aoeepeese.e.eee.DE>.U>»OJeOeOJDe.x.noose.oenee.cxxs.eo>xe>—tuxe
10>.ee..¢eeoe>oe..oeeeeese.o.eee.06>.U>h04e0e04oe..eeoe>e.noe.e.oex>.uo)xe>.ezxee
e.xe—s.xe._:eeev>—e.ecee-Onsu.cases.9E>.m>hOJeOIOJeeoxeee-)aunease.Uxx>.Eo>sv>uexereO.Iee

xceo.ee.Eeeees.uu..oeceee>.>e.eee.IE>.U>»OJeOIO.ee.x.eevexeeee.e.eexe.ne>x¢2eeasee
are..oe.>eeeoo....eoexeensc.e.eee.e£>_u>p04o0e041e.e-eeussseee.e.xex>.eo>xe..exxe

xs.e>.ee..eeeeesue.xeexeeefie.n.eee.e£>.u>»04eoeoJoe.xueessxsoex.e_oxxs.se)xe>ueexee

xxe!l.ee..eeeoe>ue.xeexe0!€o.0.eee.96>.u>h04e0104v>.x.ee>>connx.e_oxx>.ee>xe>ueexee

let....Oe..lleflesue.eoezOOlEe.o.eee.9E».U>h04e0e0400slueOu.00eex.e.01~>>00>tv>-em800

sexea..ee..leeoe>~e.eoeseeefle.o.eee.ofie.w>bOJeOIOJee.xuelu.eueea_e.cx.>>e0>xv>uenmee

xxec..ee..eeeoesue.roezeeeia.seeee.oE>_u>hOJeOeOJoeexueessxxneu_e.oxx>.ee>ae.«ease.

C

«005.63 Eon oEEc d0< 239: 05 .0

Q0(
ltiQU(

QIIQU(

A04
AU(
AU(

0AU(

.QU(
AU¢

aoiaus
condos

A04
...IAU(
..IQU(
.II0(

..IIU(
.IAU(
10¢
AU(
nInO(
~lflU(
'IAU(

i 6:3

sҤ
a «fit

E

0,)

IIE£ ..
eve-ev‘ooo
(‘14

208:2?

>
0

113193
ICXNDUCKU

adds
zomp

 

lOKJQHOJd E1
2
<

FZ<Hu

12

From structure/function analysis, it appears that the amino terminus is
functionally more important than the carboxyl terminus (12). Deletion of
more than 3 residues from the amino terminus of E.coli ACP resulted in a
reduction of biological activity» However, deletion of more than 10 amino
acid residues from the carboxyl terminus had no effect on ACP function
(12). Halo ACPs have 4'-phosphopantetheine covalently bound to a serine
residue through a phosphodiester bond (Fig. 2). While actively

participating in fatty acid biosynthesis, acyl residues are bound to the

terminal sulfhydryl of the prosthetic group via a thioester bond (30).

Predictions of protein secondary structure from the amino acid sequence
data, combined with NMR studies and nuclear Overhauser data originally
suggested that B.coli ACP has four a helices (13). However, more recent
proton NMR analyses showed that the second helix is missing (23). A
strong structural homology between B.coli and spinach ACP was recently
shown by NMR studies (23) indicating that plant ACPs are more similar to
prokaryotic ACPs than to the ACP regions of the animal fatty acid

synthases.

Subcellular localisation of acyl carrier proteins

ACP is one of the most abundant proteins in E.coli (30), but its
subcellular localization is still not defined. In 1970, Van der Bosch et
a1. (55) concluded from the electron microscopic analysis of an B.coli

auxotroph grown on [B - 3H] alanine, that ACP is associated with cell

13
membranes. Contrary to these data, immunoelectron microscopical analysis,
performed by Jackowski et al. (21), showed that ACP could be found only in
the cytoplasm and did not bind to membranes. Reevaluation of the same
issue combined the findings of the two groups in one. According to Bayan
8: Therisod (3), the majority of ACP in E.coli is localized in the
cytoplasm. However, a small fraction (approximately 1.5% - 3% of the ACP
pool) remains associated‘with protein.binding sites in the inner membrane.
The authors postulated that ACP binding sites in inner membranes are
enzymes involved in phospholipid and membrane - bound oligosaccharide
biosynthesis. In support of this theory, they reported that the number of
ACP binding sites increases in microsomal vesicles isolated from
overproducers of the glycerol-3-phosphate acyltransferase (GPAT) (4).
GPAT is localized in the inner membrane and catalyzes the transfer of
fatty acid chains from acyl-ACP to the sn-1 position of glycerol-3-

phosphate in the synthesis of membrane phospholipids.

In animals and fungi, the multifunctional FAS is localized in the
cytoplasm (56) . Recently, ACPs have been detected as separate proteins in
N. crassa mitochondria (43, 29) and in mitochondria of bovine heart muscle

(42).

Initial reports on the localization of FAS in plants associated fatty acid
synthesis with different plant organelles, due to inappropriate methods of
organelle separation. In 1979, Ohlrogge et al. (34) reported that the
majority of ACP is localized in plastids. However, they cautioned that

because of the large proportion of the cell volume that was occupied by

14
plastids minor amounts of ACP present in some other organelles might not
be detected. Recently, upon the localization of mitochondrial ACP in N.
crassa, Brody and coworkers (29) were able to identify ACP-like proteins
in mitochondrial fractions of pea leaves and potato tubers (6). These
proteins were characterized as putative ACPs based on their cross-
imunoreactivity and on the changed mobilities on PAGE upon deacylation of

the protein fraction.

The highly soluble nature of ACP implied that plant ACPs are probably
localized in plastid stroma. However, immunogold labeling studies using
spinach ACP-I antibodies in Brassica leaves detected most of the ACP
associated with the thylakoid membranes (50). Lack of more extensive
studies on the localization not only of ACP but also of other fatty acid
synthase components prevents us from making firm conclusions regarding the

intracellular and intraorganellar localization of ACPs in plants.

Isoforms of acyl carrier proteins and their intercellular localisation

Soon after the report that ACP from spinach leaves might be represented by
more than one isoform (49), several laboratories reported that multiple
ACP isoforms are present, not only in spinach (26), but also in barley
(27), castor bean and soybean (24). Recent data which are based on the
labeling of ACPs with.[3H] palmitate, detected multiple ACP isoforms even

in simple multicellular plants (2).

15
ACP isoforms from different plants are immunologically similar (31).
Antibodies raised against spinach ACP-I cross-react with spinach ACP-II
and with other plant and prokaryotic ACPs. However, according to the
results of a competitive binding assay, spinach ACP-II could compete with
only 40% of the ACP-I binding sites (31). The same antibodies were even
less efficient in recognizing ACP isoforms from prokaryotes like E. coli,

Anabaena variabilis and Synechocystis 6803 (10).

Tissue-specific expression of ACP isoforms was initially observed in
spinach, soybean and castor bean (24). Based on SDS-PAGE analysis, these
plants have a single ACP isoform expressed in a constitutive manner, in
leaves, roots and seeds, and a single leaf-specific ACP isoform. Barley,
however, has three ACP isoforms expressed in leaves (27). Their relative
abundances vary greatly with the age of leaf cells (2). The predominant
ACP isoform found in young regions of the leaves, becomes even more
abundant as the leaf matures. The two minor ACP isoforms from the young
leaf areas are less abundant as the leaf tissue gets older. These results
suggest that the expression of the three ACP isoforms in barley leaves is
developmentally regulated. Multiple ACP isoforms also have been detected
in rapeseed (44). Cloning of ACP from a rapeseed seed cDNA library
revealed the presence of a large ACP family that was mainly expressed in
seeds. Low levels of the expression of the "seed-specific ACPs” were also

detected in leaves (40).

In contrast to higher plants, prokaryotes and unicellular eukaryotes were,

up to now, considered to have a single form of ACP. These conclusions

16
were based on the extensive analysis of the E; coli ACP (30), on the
electrophoretic analysis of protein extracts from unicellular algae,
Chlamydomonas reinhardtii and Dunaliella tertiolecta (2), and from
photosynthetic cyanobacteria, Synechocystis strain 6803 (2, 22) and
Agmnellum quadruplicatum (2), and on ACP sequence information from
Streptomyces erythraeus (11). However, recently, two ACP isoforms were
reported in Rhizobiumu One ACP isoform, is the product of the nod! gene,
which is part of the nodulation. operon (48), and the other is a
constitutively expressed ACP isoform (36). It appears that the two
isoforms have different functions. The inducible ACP isoform seems to be
involved in the synthesis of membrane-derived oligosaccharides, whereas

the constitutive ACP isoform is part of the fatty acid synthase (48).

Molecular biology of acyl carrier proteins

Early studies characterizing ACPs suggested that plant ACPs are nuclear
encoded proteins. These findings were based on differences observed in
the sizes of the in vitro mRNA translation products compared to purified
ACPs, after immunoprecipitation with antibodies to ACP (31). Isolation
and characterization of other ACP clones confirmed that eukaryotic ACPs
are nuclear encoded proteins. Amino acid sequence information of the
purified ACP isoforms when compared to deduced amino acid sequences from
cDNA and genomic clones indicated that spinach ACP-I (26, 45) and ACP-II
(31, 46), barley major leaf ACP isoform (27, 16, 17) and rapeseed ACP

isoform (44), are translated as precursor proteins with approximately 51-

17
59 amino acid long transit peptides. Furthermore, studies confirmed that
the transit peptide of the precursor ACP is processed upon import into the
chloroplasts. Very recently, an ACP-like sequence has been detected in
chloroplast genome of the marine diatoms Cylindrotheca sp. Strain N1 (19)
and of the Cryptomonas sp. (57). However, a DNA sequence search for ACP-
like sequences of the tobacco chloroplast genome did not reveal any

sequences similar to ACPs.

Sequence information for cDNA ACP clones from dicots [spinach ACP-I (45)
and ACP-II (46), seed ACP isoforms from rapeseed (44), B. campestris seed
(8)], and monocots [barley ACP-I (16), ACP-II (15), ACP-III (16), and
maize (51)] is available. In addition to the sequence of the B.coli ACP
gene (38), several genomic clones have been sequenced from barley (17), B.
napus (40) and Arabidopsis (37, 25). The known plant ACP genes have very
similar structures. 'The coding regions are interrupted‘with three introns
whose positions are almost completely conserved. Intron I is located in
the transit peptide coding region. Its position varies in the different
ACPs by a few amino acid residues. In the barley Ach and Ac13 genes, the
second intron is positioned between the last amino acid of the transit
peptide and the first amino acid of the mature ACP isoform, whereas in
barley AclI, intron II is in between the first and second amino acid of
the mature ACP-I. The ACP 09 gene from.Brassica has a similar position of
the intron as barley Acll. Intron III in all known ACP genes is located
only 9 base pairs downstream of the serine codon which is the binding site
of the phosphopantetheine prosthetic group. The entire region surrounding

the prosthetic group attachment site is highly conserved (30). The

18

introns from the dicot ACPs are shorter than the barley introns.

The expression patterns of spinach ACP isoforms were confirmed by Schmid
and Ohlrogge (46) at the mRNA level. Sensitive RNase protection assays
were used to detect ACP-II mRNAs in leaves, roots and developing seeds and
ACP-I mRNA in leaves. With the isoform specific probes they were able to
quantify steady-state levels of both ACP-I and ACP-II mRNA. In leaves,
each of the two isoforms represented approximately 0.1 mol% of the
polyadenylated RNA pool and the molar ratio between ACP-I and ACP-II
messages was 0.9 (46). The measurements of steady-state levels of ACP
mRNA in both spinach and soybean leaves revealed that the abundance of ACP
message is higher in younger leaves than in older leaves (14). In the
same study the highest relative abundance of ACP'mRNA in soybean seeds was
shown to be 0.027% at 20 days after flowering. In both leaves and seeds,
changes in the level of ACP mRNA were accompanied by changes in the
protein levels, suggesting that increases in ACP concentration in

developing seeds are a result of de novo protein synthesis.

The transcription start sites have been determined for B. napus ACP 05
gene and for barley Ac11 and Ac13 genes. In B. napus, the 5' leader
sequence is 69 nucleotides long. The transcription start site sequence,
GGCATCA, shares five out of seven nucleotides with a proposed consensus
transcriptional start sequence (40). In both barley ACP genes, Acll and
Ac13, the three transcriptional start sites have been located (17). The
major transcriptional start sites in Ac11 and Ac13 genes result in 97 and

84 nucleotide long RNA leader sequences, respectively. In addition, minor

19
transcripts with 90 nucleotide and 144 nucleotide long leader sequences
were detected in AclI, whereas in Ac13, the two minor leader sequences
were 77 and 140 nucleotides long. Preliminary results indicated that
Arabidopsis ACPZ message has a 123 nucleotides long leader sequence (25).
No multiple transcripts were reported in the study. Putative multiple
polyadenylation sites were found for barley Ac12 and Ac13 genes (16,17)

and for spinach ACP-II (46).

The three barley ACP genes have been localized on chromosomes using the
Chinese Spring wheat - Betzes barley addition lines (17, 15). According
tar DNA. blot hybridizations, Ac11 and .Ach are on a chromosome 7,
approximately 10 kb apart, and Ac13 is on the chromosome 1. All three
genes were reported to be present as single copy genes. Arabidopsis ACPZ
and ACP3 genes are linked on a 18 kb DNA fragment (25). However, no
analysis has been performed to assess the number of copies for either one
of them. Quantitative DNA blot hybridizations revealed a large ACP family
in B. napus genome indicating that some genes exist in multiple copies
(44). Even in spinach, which has only two to three ACP isoforms
distinguishable by PAGE, DNA blot hybridizations suggest the presence of
a larger gene family (46). DNA sequence comparisons indicate that
Arabidopsis ACPZ and ACP3 genes are products of a recent gene duplication
event. The degree of identity between B. campestris, B. napus and
Arabidopsis ACPI genes suggests that they all share a common ancestral

gene which diverged from the two other known Arabidopsis ACP genes.

10.

11.

20

Bibliography

Andrews, J., Ohlrogge, J.B. and Keegstra, K. (1985) Final step of
phosphatidic acid synthesis in pea chloroplasts occurs in the inner
envelope membrane. Plant Physiol. 78; 459-465.

Battey, J.F. and Ohlrogge, J.B. (1990) Evolutionary and tissue-
specific control of expression of multiple acyl-carrier protein
isoforms in plants and bacteria. Planta 180; 352-360.

Bayan, N. and Therisod, H. (1989) Membrane-binding sites for acyl
carrier protein in Escherichia coli. FEBS Lett. 253; 221-225.

Bayan, N. and Therisod, H. (1989) Evidence for interactions of acyl
carrier protein with glycerol-B-phosphate acyltransferase, an inner
membrane protein of Escherichia coli. FEBS Lett. 255; 330-334.

Bertrams, M. and Heinz, E. (1981) Positional specificity and fatty
acid selectivity of purified sn-glycerol-3-phosphate acyltransferase
from chloroplasts. Plant Physiol. 68; 653-657.

Chuman, L. and Brody, S. (1989) Acyl carrier protein (ACP) is present
in the mitochondria of plants and eucaryotic micro-organisms. Eur. J.
Biochem. 184; 643-649.

Cortes, J., Haydock, S.F., Roberts, G.A., Bevitt, D.J. and Leadlay,
P.F. (1990) An unusually large multifunctional polypeptide in the
erythromycin-producing polyketide synthase of SaccharOpolyspora
erythraea. Nature 348; 176-178.

de Silva, J., Loader, N.M., Jarman, C., Windust, J.H.C., Hughes, 8.
and Safford, R. (1990) The isolation and sequence analysis of two
seed-expressed acyl carrier protein genes from Brassica napus. Plant
Mol. Biol. 14; 537-548.

Frentzen, M., Heinz, E., McKeon, T.A. and Stumpf, P.K. (1983)
Specificities and selectivities of glycerol-3-phosphate
acyltransferase and monoacylglycerol-3-phosphate acyltransferase from
pea and spinach chloroplasts. Eur. J. Biochem. 129; 629-636.

Froehlich, J.E., Poorman, R., Reardon, E., Barnum, S.R. and Jaworski,
J.G. (1990) Purification and characterization of acyl carrier protein
from two cyanobacteria species. Eur. J. Biochem. 193; 817-825.

Hale, R.S., Jordan, K.P. and Leadlay, F. (1987) A small discrete acyl

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

21

carrier protein is involved in de novo fatty acid biosynthesis in
Streptomyces erythraeus. FEBS Lett. 224; 133-136.

Hancock, W.S., Marshall, G.R. and Vagelos, P.R. (1973) Acyl Carrier
Protein. xx Chemical synthesis and characterization of analogues of
acyl carrier protein. J. Biol. Chem. 248; 2424-2434.

Holak, T.A., Nilges, M., Prestegard, J.H., Gronenborn, A.M. and.Clore,
G.M. (1988) Three-dimensional structure of acyl carrier protein in
solution.determined by nuclear magnetic resonance and the combined use
of dynamical stimulated annealing and distance geometry. Eur. J.
Biochem. 175;9-15.

Hannapel, D.J. and Ohlrogge, J.B. (1988) Regulation of acyl carrier
protein messenger RNA levels during seed and leaf development. Plant
Physiol. 86; 1174-1178.

Hansen, L. (1987) Three cDNA clones for barley leaf acyl carrier
proteins I and III. Carlsberg Res. Commun. 52; 381-392.

Hansen, L. and Kauppinen, s. (1991) Barley acyl carrier protein II:
nucleotide sequence of cDNA clones and chromosomal location of the
Ac12 gene. Plant Physiol. 97; 472-474.

Hansen, L. and von Wettstein-Knowles, P. (1991) The barley genes Acll
and Ac13 encoding acyl carrier protein I and III are located on
different chromosomes. Mol. Gen. Genet. 229; 467-478.

Hopwood, D.A. and Sherman, D.H. (1990) Molecular genetics of
polyketides and its comparison to fatty acid biosynthesis. Annu. Rev.

Hwang, S.-R. and Tabitas, P.R. (1991) Acyl carrier protein-derived
sequence encoded by the chloroplast genome in the marine diatom
Cylindrotheca sp. Strain Nl. J. Biol. Chem. 266; 13492-13494.

Issartel, J.-P., Koronakis, V. and Hughes, C. (1991) Activation of
Escherichia coli prohaemolysin to the mature toxin by acyl carrier
protein dependent fatty acylation. Nature 351; 759-761.

Jackowski, S., Edwards, H.H., Davis, D. and Rock, C.O. (1985)
Localization of acyl carrier protein in Escherichia coli. J.
Bacteriology 162; 5-8.

Jaworski, J.G., Post-Beittenmiller, D. and Ohlrogge, J.B. (1992)
Acetyl-acyl carrier protein is not a major intermediate in fatty acid
biosynthesis in plants. submitted

Kim, Y., Ohlrogge, J.B. and Prestegard, J.H. (1990) Motional effects
on NMR structural data. Comparison of spinach and Escherichia coli
acyl carrier protein. Biochem. Pharmacol. 40; 7-13.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

22

Kuo, T.M. and Ohlrogge, J.B. (1984) The primary structure of spinach
acyl carrier protein. Arch. Biochem. Biophys. 234; 290-296.

Lamppa, G. and Jacks, C. (1991) Analysis of two linked genes coding
for the acyl carrier protein (ACP) from Arabidopsis thaliana
(columbia). Plant Mol. Biol. 16; 469-474.

Matsumura, S. and Stumpf, P.R. (1968) Fat metabolism in higher plants.
xxxv. Partial primary structureeof spinach acyl carrier protein. Arch.
Biochem. Biophys. 125; 932-941.

Hoj, P.B. and Svendsen, I. (1984) Barley chloroplasts contain two acyl
carrier proteins coded for by two different genes. Carlsberg
Res.Commun. 49; 483-492.

McKeon, T.A. and Stumpf, P.R. (1982) Purification and characterization
of the stearoyl-acyl carrier protein desaturase and the acyl-acyl
carrier protein thioesterase from maturing seeds of safflower. J.
Biol. Chem. 257; 12141-12147.

Mikolajczyk, S. and Brody, s. (1990) De novo fatty acid synthesis
mediated by acyl-carrier protein in Neurospora crassa mitochondria.
Eur. J. Biochem. 187; 431-437.

Ohlrogge, J.B. (1987) Biochemistry of plant acyl carrier proteins. In
The Biochemistry of Plants. A Comprehensive Treatise. Vol.9. Stumpf,
P.R. ed., Academic Press Inc., Harcourt BraceiJovanovich, Orlando, pp.
137-158.

Ohlrogge, J.B. and Ruo, T.M. (1985) Plants have isoforms for acyl
carrier protein that are expressed differently in different tissues.
J. Biol. Chem. 260; 8032-8037.

Ohlrogge, J.B., Shine, W.E. and Stumpf, P.R. (1978) Fat metabolism in
higher plants. Characterization of plant acyl-ACP and acyl-CoA
hydrolases. Arch. Biochem. Biophys. 189; 382-391.

Ohlrogge, J.B. and Kuo, T.M. (1985) Plants have isoforms for acyl
carrier protein that are expressed differently in different tissues.

Ohlrogge, J.B., Kuhn, D.N. and Stumpf, P.R. (1979) Subcellular
localization of acyl carrier protein in leaf protoplasts of Spinacia
oleracea. Proc. Natl. Acad. Sci. USA 76; 1194-1198.

Overath, P. and Stumpf, P.R. (1964) Fat metabolism in higher plants.
XXIII. Properties of a soluble fatty acid synthetase from avocado

Platt, M.W., Miller, K.J., Lane, W.S. and Kennedy, E.P. (1990)
Isolation and characterization of the constitutive acyl carrier
protein from Rhizobium meliloti. J. Bacteriol. 172; 5440-5444.

37.

38.

39.

40.

41.

42.

43.

440

45.

46.

47.

48.

23

Post-Beittenmiller, M.A., Hlousek-Radojcic, A. and Ohlrogge, J.B.
(1989) DNA sequence of a genomic clone encoding an Arabidopsis acyl
carrier protein (ACP). Nucleic Acids Res. 17; 1777.

Rawlings, M. and.Cronan,Jr., J.E. (1992) The gene encoding Escherichia
coli acyl carrier protein lies within a cluster of fatty acid
biosynthetic genes. J. Biol. Chem. 267; 5751-5754.

Revill, W.P. and Leadlay, P.F. (1991) Cloning, characterization, and
high-level expression in Escherichia coli of the Saccharopolyspora
erythrea gene encoding an acyl carrier protein potentially involved in
fatty acid biosynthesis. J. Bacteriology 173; 4379-4385.

Rose, R.E., DeJesus, C.E., Moylan, S.L., Ridge, N.P., Scherer, D.E.
and Knauf, V.C. (1987) The nucleotide sequence of a cDNA clone
encoding acyl carrier protein (ACP) from Brassica campestris seeds.
Nucleic Acids Res. 15; 7197.

Roughan, P.G. and Slack, C.R. (1982) Cellular organization of
glycerolipid metabolism. Ann. Rev. Plant Physiol. 33; 97-132.

Runswick, M.J., Fearnley, I.M., Skehel, J.M. and Walker,.J.E. (1991)
Presence of an acyl carrier protein in NADqubiquinone oxidoreductase
from bovine heart mitochondria. FEBS Lett. 286; 121-124.

Sackmann, 0., Zensen, R., Rohlen, D., Jahnke, U. and Weiss, H. (1991)
The acyl-carrier protein in Neurospora crassa mitochondria is a
subunit of NADH:ubiquinone reductase (complex 1). Eur. J. Biochem.
136;

Safford, R., Windust, J.H.C., Lucas, C., De Silva, J., James, G.M.,
Hellyer, A., Smith, C.G., Slabas, A.R. and Hughes, S.G. (1987)
Plastid-localized seed acyl-carrier protein of Brassica napus is
encoded by'a distinct, nuclear multigene family. Eur. J. Biochem. 174;
287-295.

Scherer, D.E. and Rnauf, V.C. (1987) Isolation of a cDNA clone for the
acyl carrier protein-I of spinach. Plant Mol. Biol. 9;127-134.

Schmid, K.M. and Ohlrogge, J.B. (1990) A root acyl carrier protein-II
from spinach is also expressed in leaves and seeds. Plant Mol. Biol.
15; 765-778.

Shanklin, J. and Somerville, C. (1991) Stearoyl-acyl-carrier-protein
desaturase from higher plants is structurally unrelated to the animal
and fungal homologs. Proc. Natl. Acad. Sci. USA 88; 2510-2514.

Shearman, C.A., Rossen,, Johnston, A.W. and Downie, J.A. (1986) The
Rhizobium legumonisarum nodulation gene nodF encodes a polypeptide
similar to acyl carrier protein and is regulated by nodD plus a factor
in pea root exudate. EMBO J. 5; 647-652.

49.

50.

51.

52.

530

54.

55.

56.

57.

580

24

Simoni, R.D., Criddle, R.S. and Stumpf, P.K. (1967) Fat metabolism in
higher plants. XXXI. Purification and properties of plant and
bacterial acyl carrier proteins. J. Biol. Chem. 242; 573-581.

Slabas, A.R. and Smith, C.G. (1988) Immunogold localization of acyl
carrier protein in plants and Escherichia coli: Evidence for membrane
association in plants. Planta 175; 145-152.

Souciet, G. and Weil, J.-H. (1992) Molecular cloning and sequencing of
a cDNA encoding the maize (Zea mays L.) acyl carrier protein (ACP);
comparison to other ACP genes and ACPs. Plant Sci. 81; 215-225.

Stumpf, P.R. (1987) Biosynthesis of saturated fatty acids. In The
Biochemistry of Plants. A Comprehensive Treatise. Vol.9 Lipids:
Structure and Function. Stumpf, P.K. ed., Academic Press, Inc.
Harcourt Brace Jovanovich, Orlando, pp. 121 - 134.

Therisod, H., Weissborn, A.C. and Kennedy, E.P. (1986) An essential
function for acyl carrier protein in the biosynthesis of membrane-
derived oligosaccharides of Escherichia coli. Proc. Natl. Acad. Sci.
USA 83; 7236-7240.

Therisod, H. and Kennedy, E.P. (1987) The function of acyl carrier
protein in the synthesis of membrane derived oligosaccharides does not
require its phosphopantetheine prosthetic group. Proc. Natl. Acad.
Sci. USA 84; 8235-8238.

Van den Bosch, H., Williamson, J.R. and Vagelos, P.R. (1970)
Localization of acyl carrier protein in Escherichia coli. Nature 228;
338-341.

Wakil, S.J. (1989) Fatty acid synthase, a proficient multifunctional
enzyme. Biochemistry 28; 4523-4530.

Wang, S. and Liu, x.-Q. (1991) The plastid genome of Cryptomonas O
encodes an hsp70-like protein, a histone-like protein, and an acyl
carrier protein. Proc. Natl. Acad. Sci. USA 88; 10783-10787.

Whitaker, B.D. (1986) Fatty acid composition of polar lipids in fruit
and leaf chloroplasts of "16:3"- and "18:3"- plant species. Planta
169; 313-319.

CHAPTER TWO
ISOLATION AND CHARACTERIZATION OF AN ACP cDNA AND A GENONIC CLONE FROM

ARABIDOPSIS I'HALIANA

Abstract

I have isolated genomic and cDNA ACP clones from A sep6-lac5 and 1 - ZAP
(Stratagene) libraries, respectively. The two Arabidopsis ACPs, ACPI and
ACP2, shared 35% identical nucleotides in the transit peptide coding
region and 69% identical nucleotides in the coding region for mature ACP.
By using gene specific probes derived from these two clones for the
hybridization of genomic DNA blots, a minimum of three ACP genes in
Arabidopsis has been detected. According to RNA blot analysis, ACPI and
ACPZ genes are expressed in leaves, roots and developing seeds. However,
the possibility of the probes hybridizing to very similar, but not

identical ACP RNA messages, cannot be ruled out.

Introduction

Mouse ear cress, better known as Arabidopsis thaliana (L.) Heynh. var.
Columbia, is a small cruciferous weed, closely related to economically

important crops from the genus Brassica (38). The small size, short

25

26
generation time and.only five chromosomeslof Arabidopsis (29,11) attracted
the attention of plant geneticists who developed its extensive genetic
map (21) . The haploid genome of Arabidopsis has approximately 1x108 base-
pairs (1), with almost no repetitive DNA (37). Because it is the smallest
known genome in higher plants, it has been postulated that Arabidopsis may
have the minimum number of genes necessary for the growth of a plant. The
initial characterizations of several Arabidopsis genes have supported that
theory. For example, only one aldehyde dehydrogenase gene was identified
in the Arabidopsis genome, whereas two or three have been found in other
higher plants (7). Chlorophyll a/b light-harvesting protein, that has a
large gene family with eight and a minimum of sixteen genes in Pisum and
Petunia, respectively, has only three genes in Arabidopsis (26).
Similarly, the gene family for the 12S-globulin seed-storage protein genes
has a lower number of genes in Arabidopsis than in other plants (29).
However, recently several large gene families in Arabid0psis have been
characterized that were of comparable size to the gene families fromiother
plants. Both a and B-tubulin genes in Arabidopsis are organized in large
gene families that have a nunimum of six and nine genes, respectively
(22,46). Even though it appears that more exceptions to the general rule
of smaller gene families in Arabidopsis will be found, it is still likely

that most Arabidopsis gene families have a reduced number of genes.

For many years Arabidopsis has been a model plant for the analysis of
lipid metabolism. A large number of mutants in various steps of lipid
biosynthesis provided valuable information on the biosynthesis of lipids

(32). In Arabidopsis, as in other higher plants, the membrane lipids are

27
synthesized by two complementary pathways that are present in plastidial
and cytoplasmic compartments (40,6) . The origin of the synthesis of
particular glycerolipids such as phosphatidylglycerol,
monogalactosyldiacylglycerol, digalactosyldiacylglycerol and sulfolipids,
could be revealed from the positional distribution of 16 carbon and 18
carbon fatty acids (48) . The glycerolipids that are made predominantly in
plastids contain substantial amounts of hexadecatrienoic fatty acid at the
sn-2 position. On the contrary, the glycerolipids made in the
endoplasmatic reticulum contain predominantly a-linolenate at the same
position. As a result of those differences the group of plants whose
plastid membrane lipids are assembled in the extraplastidial compartment
are named "18:3 plants", as opposed to the "16:3" plants, where plastids
contribute to the synthesis of glycerolipids. In ArabidOpsis, a "16:3"
plant according to Browse et al. (6), 38% of de novo synthesized fatty
acids goes into the plastidial pathway and more than half of the fatty
acid pool (56%) that is exported to the cytoplasm re-enters the
chloroplasts where it contributes to the production of thylakoid
glycerolipids. Arabidopsis leaf lipids contain predominantly
polyunsaturated 18 carbon fatty acids, whereas its seeds have in addition

to the 18 carbon polyunsaturates, 20 and 22 carbon fatty acids (5).

In spite of the extensive work on lipid metabolism in Arabidopsis, very
little information is available on the regulation of fatty acid
biosynthesis. Having in mind the advantages of Arabidopsis for molecular
biological and biochemical studies, I wanted to characterize Arabidopsis

ACP genes as a first step that will bring us closer to the understanding

28
of the regulation of the expression of the components of fatty acid
synthase. Therefore, I began to characterize acyl carrier proteins in
Arabidopsis, so they can be used as markers in the analysis of fatty acid
biosynthesis in plants (34). It has been known that Arabidopsis, like
other multicellular plants, has multiple ACP isoforms that appear to be
expressed in a tissue-specific manner (2). However, the specific
functions of ACP isoforms have not yet been revealed even in spinach where
the characterization of ACPs led to the amino acid sequencing of the ACP-I
(24) and ACP-II (33) isoforms and to the cloning of their cDNA clones

(43,44).

Experimental procedures

Materials. An Arabidopsis genomic library in 1 sep6-lac5 was provided by

Chris Somerville. An Arabidopsis expression library in A-ZAP (Stratagene)
prepared from mRNA isolated from cold-acclimated plants was a gift from
Mike Tomashow. A cDNA ACP clone from Brassica napus seeds, pBN45, was
donated by Richard Safford. A 66-nucleotide long oligomer, complementary
to the region encoding the 22 amino acids surrounding the
phosphopantetheine binding site (15), was aigift from Penny von Wettstein-

Knowles and Lars Hansen.

Plant material. Arabidopsis thaliana (L.) Heynh. ecotype Columbia plants
were grown under a 16 h light/dark cycle either in a soil or in hydroponic

culture. Plants grown in the soil were watered with a Hoagland solution

29

(5 M 10103, 2.5 mM m4, pH 5.5, 2 mM M9804, 2 mM Ca(NO3)2, 2.5 mM serum,
70 pm H3803, 14 um MnClz, 0.5 mt CuSO;,, 1 mt 2nso,,, 0.2 mt NaMoO4, 10 mM
NaCl, 0.01 [IN CoClZ) once a week. Hydroponic culture was set in a
Hoagland solution with a continuous aeration and weekly changing of the

solution. Seeds of the hydroponically grown plants were germinated on a

sheet of a cheese cloth that was 0.5 cm above the water layer.

Preparation of hybridization probes. DNA fragments that were used for
random primer labeling were resolved by agarose gel electrophoresis using
Sea Plaque low-gelling agarose (FMC Bioproducts) in Tris-Acetate buffer.
The band of interest was cut from the gel, melted at 70°C, and used
directly for labeling. Random primer labeling reactions were carried out
with a mix of random hexamers, Klenow fragment and (32P)dCTP, according to
the manufacturers' protocols (US Biochemical Corp.). Oligonucleotides
were either random primer labeled with [32P1CTP or end-labeled with
[32P1ATP using Klenow fragment according to the methods described in

Maniatis (42).

Sense and antisense RNA probes were made according to the protocols
included with a Promega Riboprobe kit, utilizing either T3, T7 or Sp6 RNA
polymerase. For templates, recombinant plasmids, were CsCl purified and
linearized with one of the restriction enzymes from the polylinker region
(42). The enzymes were inactivated by proteinase K treatment and removed
by phenol] chloroform extraction. The DNA fragments were recovered by
ethanol precipitation. Incorporation of the’(38P)UTP into RNA probes was

determined.by Cerenkov counting of aliquots spotted on glass fiber filters

30
and precipitated with 5% TCA containing 20 mM sodium pyrophosphate (4).
The length of the probes was confirmed by resolving labeled RNA probes in

the 6 M urea PAGE followed by autoradiography.

Screening of the Arabidopsis genomic library. 1x106 plaque forming units
of the A library were adsorbed to 5x106 E.coli LE392 in SM buffer for 20
minutes at 37°C. 3ml of 0.7% top agarose, that was cooled to 55°C, was
mixed with cells and the mixture was spread over the LB plates. Plates
were then incubated at 37°C for 14-18 hours. When the clear plaques became
visible, plates were either stored at 4°C or processed immediately. The
nitrocellulose circles were marked and placed face-down onto a plate for
1 minute. Replica filters were left on the top agar layer for 90 seconds.
Each filter was carefully removed and placed face-up for 1 minute on a
sheet of filter paper saturated with 0.5 M NaOH, 1.5 M NaCl for DNA
denaturation. Nitrocellulose filters were then neutralized with an
incubation on 1.5 M NaCl, 0.5 M Tris HCl, pH 8.0 for 5 minutes. Following
the neutralization step, filters were rinsed in 2x SSPE and baked for 2

hours at 80°C in a vacuum oven.

Nitrocellulose filters were prehybridized at 42°C in 7% SDS, 2x SSPE, 30%
formamide for 30 minutes to 5 hours. The prehybridization mixture was
removed and a the same buffer and [32PJUTP labeled RNA probe derived from
pBN45 plasmid (2x106 CPM) were added to the container. Hybridization was
performed at 42°C for 12-24 hours. Filters were washed twice at room
temperature with 2x SSPE, 0.2% SDS and once at 50°C for 30 minutes in 1x

SSPE, 0.2% SDS. Filters were exposed to film for 2 days with intensifying

31

screens at -70°C .

Antibody screening of the Arabidopsis cDNA expression library. The cDNA
library made in l-ZAP was grown in E. coli strain BB4 (Stratagene) . Cells
that were prepared for the adsorption of viruses were grown in TBMg +
maltose media and harvested by centrifugation at 3000 rpm for 10 minutes.
The pellet was resuspended in 5 ml 10 mM MgSO,,. Viral dilutions that were
made in SM buffer were mixed with E.coli cells and incubated with shaking
at 37°C for 20 minutes. 0.7% molten agarose, cooled to 55°C, was mixed
with cells and the mixture was poured on the agar plate. Plaques were
grown at 42°C for 6-8 hours. Dry nitrocellulose filters that were
saturated with 10 mM IPTG solution, were placed onto the plates. IPTG
induced the expression of the proteins during the incubation of the plates
at 37°C for 2-4 hours. At the end of incubation, filters were transferred
to petri dishes. Proteins were fixed on the filters with 3% TCA for 5
minutes. Residual amounts of TCA were removed by rinsing filters with dHZO

three times.

Before imunoblotting with spinach ACP-I or recombinant spinach ACP-I
antibodies, filters were blocked for 2 hours in Blotto. 1:250 dilution of
the primary antibody was prepared in Blotto, and filters were
imunoblotted for one hour. At the end of incubation with primary
antibodies, filters were washed 3 times in TBST for 10 minutes each.
Secondary goat anti-rabbit IgG antibodies were diluted to 1:1000 and added
to the filters. After one hour of incubation, the antibody solution was

removed and filters were washed three times for 10 minutes with TBST and

32
once with alkaline phosphatase buffer (100 mM Tris Hcl, pH9.5, 100 mM
NaCl, 5 mM MgClz). Color development was done in alkaline phosphatase
buffer with NET (50 mg/ml) and BCIP (50 mg/ml) that were dissolved in 70%
and 100% N,N-dimethylformamide, respectively. The color development was
stopped by thoroughly rinsing filters with any). Positive plaques were
picked from the plates, and phage was extracted from the cells in SM

buffer supplemented with a few drops of chloroform.

Putative positive plaques were rescreened with anti-spinach ACP-I
antibodies, and replica filters were probed with.[3aP]ATP end-labeled 66-
nucleotide long oligomer. Phagemids from the plaques that gave positive
signal with antibodies and with the 66-mer were isolated and pBluescript
plasmids were rescued using R408 helper phage according to the Stratagene
protocol. Positive plasmids were confirmed by probing with random primed

(321: ) dCTP-labeled 66-mer .

Preparation of 1 DNA. The putative positive plaques were grown in 2 ml
LB liquid cultures. 1/20 volume of the phage stock, that was prepared in
1 ml of su, was mixed with 500 #1 SM and with 1x108 bacteria, and
incubated at 37° for 20 minutes. 2 ml of LB media, that was supplemented
with MgSOl, to 5mM, was added to the mixture of bacteria and phage, and left
to grow overnight at 37°C at constant shaking. After 10 hours of growth,
approximately four drops of chloroform were added to the cells, which.were
left in the shaker for an additional 15 minutes. Following centrifugation
at 3,000 rpm for 5 minutes, the supernatant was transferred to another

tube where it was treated with DNase and RNase for 20 minutes at 37°C.

33

RNase was inhibited with the addition of 40 pl DEPC. Following the
addition of 75 ul of 10% SDS and 375 #1 of 2 M Tris HCl, 0.2 M EDTA, the

solution was mixed and placed on ice for 30 seconds. 250 pl of 8 M
potassium acetate was added to the solution which was then mixed by
inversion, and the solution was put on ice for an additional 30 seconds.
Proteins were pelleted by centrifugation at 10,000 rpm for 15 minutes.
DNA from the supernatant was precipitated with 2 volumes of ethanol and
spun down at 10,000 rpm for 10 minutes. The DNA pellet was then washed

with 70% ethanol and resuspended in TE buffer.

DNA subcloning and sequencing. DNA of the putative 1 ACP clone was
digested with EcoRI and subcloned into pBS(+) (42). A restriction map of
the 4.8 kb EcoRI fragment, which hybridized to the [3SPJRNA probe derived
from pBN45, was constructed using single, double and triple digests with
several restriction enzymes. DNA fragments were resolved by 1% agarose
gel electrophoresis and DNA blots were hybridized with.[3aP]-labeled 66-

mar .

A putative ACP clone that was isolated from the 1 ZAP expression library
was digested with EcoRI. The EcoRI fragment that hybridized to the [32F]-
labeled 66-mery was subcloned into jpBS(+)KS for further restriction

mapping and sequencing.

DNAs of plasmid clones that were used for sequencing were purified either

by CsCl gradient purification or by a "mini" preparation method that was

34
followed by PEG precipitation of DNA (42) . DNA was alkaline denatured and
sequenced following the dideoxy chain termination method using either G
mix or I mix and Sequenase 2.0 according to the U.S. Biochemical

protocols.

Preparation of Arebddopsis genomic DNA. Either fresh or frozen aerial
parts of 5 week old Arabidopsis plants were used for the isolation of
genomic DNA. 10 g of tissue were ground with mortar and pestle in liquid
nitrogen. The frozen powder was added to 10 ml of homogenization buffer
(0.15 M Tris HCl, pH 8.5, 0.1 M EDTA, 2% N-lauroylsarcosine, sodium salt,
0.1 mg/ml proteinase K). The'mixture was incubated at 37°C for 30 minutes
with gentle stirring» Cell debris‘was removed by centrifugation at 10,000
rpm for 10 ndnutes. The supernatant was transferred to another tube,
while the pellet was again resuspended in homogenization buffer and
incubated at 37°C for 30 minutes. After a second centrifugation the
supernatants were combined, and the DNA.was precipitated with 2 volumes of
100% ethanol at 40C for 2-3 hours. DNA was pelleted by centrifugation at
5,000 rpm for 60 minutes. The pellet was resuspended in 10 mM Tris HCl,
pH 8.0, 1 mM EDTA and DNA was purified on a CsCl gradient containing
ethidium bromide, for 18 hours at 45,000 rpm. Upon completion of the
centrifugation, the ethidium bromide was extracted from the gradient with
water-saturated 1-butanol. DNA was precipitated with 70% ethanol at 49c
for 3-4 hours and after collection by centrifugation the pellet was
resuspended in TE buffer. Quantification of DNA was performed by

measuring absorbence at 260 nm and 280 nm.

35
Preparation of Arabidopsis RNA. Five to six week old plants were used for
the extraction of total RNA from leaves. Developing seeds were collected
from the seed pods with pale green to green seeds from the 5- to 8- week-
old Arabidopsis plants. Two-week-old hydroponic cultures of Arabidopsis
were used as a source of roots for RNA isolation. All tissues were flash

frozen in liquid nitrogen and stored at -70°C until used.

Total RNA was extracted from 400 mg of root material and 200 mg of leaf or
developing seed as described in Hall et al. (16) with some modifications.
Tissue was ground in a microfuge tube with a minipestle in a
homogenization buffer (0.2 M Tris HCl, pH 9.0, 0.4 mM NaCl, 25 mM EGTA, 1%
(w/v) SDS) which was preheated in boiling water. Sequentially, 5 mg/ml of
PVP (mol wt 360,000) and 0.5 mg/ml proteinase K were added to the
homogenate. After incubation at 37°C for 1 hour, proteinase K was
inhibited by the addition of 12 mg/ml BaClz and incubation on ice for 15
minutes. At the end of incubation, KCl was added to a final concentration
of 143 mM, and the homogenate was centrifuged for 10 minutes at 15,000
rpm. RNA from the supernatant was precipitated overnight in 2 M LiCl.
After centrifugation for 20 minutes at 15,000 rpm, the pellet was rinsed
with 2 M LiCl and resuspended in DEPC-treated water. Total RNA was stored
in ethanol at -70°C. Before use, the RNA was dissolved in DEPC-treated

water.

DNA and RNA gel electrophoresis and blotting techniques. Purified
plasmid, phage or genomic DNAs were digested with one or more restriction

endonucleases, according to the conditions recommended by the

36
manufacturer, and were resolved in a 0.8 % or 1% agarose gel in TE buffer
under constant current (60 - 120 mA). The upper region of the gel
containing the high molecular weight DNA fragments was exposed to UV for
5 minutes to nick the DNA and enhance transfer of the large size DNA
fragments (42) . Prior to DNA transfer to a nitrocellulose filter, DNA was
denatured by soaking the gel in 1.5 M NaCl, 0.5 M NaOH for 30 minutes.
The gel was neutralized by a 30 minute incubation in 0.5 M Tris HCl, pH
7.5, 1.5 M NaCl. The transfer of DNA was done in 10x SSPE. The filter
was baked for an hour at 80°C in a vacuum. In some cases DNA was
transferred from the gels to the nitrocellulose by alkaline transfer.
Gels were rinsed for 15 minutes in deo, and the transfer and simultaneous
denaturation of DNA was done in 0.4 M NaOH. Upon completion of the

transfer, the membrane was briefly washed in 2x SSPE and air dried.

Total RNA was subjected to denaturing formaldehyde/formamide 1% agarose
gel electrophoresis (42). Following electrophoresis, the gel was washed
in DEPC-treated water for 15 minutes. The transfer of RNA to the Zeta
probe (Bio Rad) nylon membrane was done in 5 mM NaOH. Membranes were

first air dried and then vacuum dried at 80°C for 1 hour.

Prehybridization and hybridization of the DNA blots were done at 50°C for
16 and 24 hours, respectively, in 2xSSPE, 7% (w/v) SDS, 30% (v/v)
formamide. ' [32P1UTP-labeled RNA probes (2x106 CPM) were added to the
hybridization mix. Following hybridization, the DNA blot was washed twice
at room temperature for 5 minutes in 2xSSPE, 0.2 % (w/v) SDS, and once for

30 minutes at 50°C in 1x SSPE, 0.2% (w/v) SDS. Blots were exposed to films

37

for variable time periods at -70°C with intensifying screens.

RNA blots were prehybridized in SxSSPE, 1% (w/v) SDS, 15x Denhardt's, 50%
(v/v) formamide and 0.5 mg/ml salmon sperm DNA at 65°C for 6 to 16 hours
and then hybridized with a [32P]UTP-labeled antisense RNA probe in the
fresh prehybridization buffer for 24 hours at 65°C. The filter was washed
with two five minute washes at room temperature in 2x SSPE, 0.1 % (w/v)

SDS, and with one, 30 minute wash at 65°C in 0.2x SSPE, 0.1% (w/v) 803.

RESULTS

Structure of the m1 gene. Approximately 1x10S plaques of the Arabidopsis
genomic library were screened with a [32P1RNA probe prepared from the B.
napus ACP cDNA clone, pBN45, as a heterologous probe (41) . DNA was
isolated from three positive clones and was digested with EcoRI.
Fragments were resolved on an agarose gel, transblotted to nitrocellulose
and hybridized against the same probe used for library screening. EcoRI
fragments that hybridized to the probe were subcloned into the pBS(+)
vector (Fig. 4). The subclones, pAD4 and pl-S, were further analyzed.
A partial restriction map of the 2 kb EcoRI fragment of pl-S, followed by
DNA blot hybridizations, identified a region of the clone that hybridized

to the probe. However, after sequencing 95% of this region there was no

significant identity with any known ACP sequences. Therefore, it was

concluded that the clone p1-5 was a false positive.

38
Single, double and triple digests with 6 restriction endonucleases of the
4.8 kb EcoRI fragment from the pAD4 were resolved on agarose gels. Gels
were transblotted to nitrocellulose and the DNA was hybridized to [3EP1-
labeled pBN45 derived RNA probe. Information from the agarose gels and
DNA blot hybridizations was used for the construction of a restriction.map
and for the generation of subcloning strategy (Fig. 4). Double stranded
DNA sequencing of approximately 30% of the region of the 4.8 kb EeoRI
fragment revealed a full length ACP gene (Fig. 5), termed ACPI (36, see

Appendix).

Based on DNA sequence comparisons with previously published ACP cDNA
sequences from B. napus (41) and from B. campestris (39), it was deduced
that the ACPI gene has three AT-rich introns. One intron is positioned in
the transit peptide coding region, and the two other introns are in the
coding region for mature.ACP (Fig. 5). Intron-I is located between codons
for amino acids 16 and 17. Intron-II separates the first and the second
amino acid of the'mature polypeptide, whereas intron-III is located within
the conserved prosthetic group attachment region. The longest intron is
intron-I: 445 bp. Introns II and III are similar in size, 80 and 76 bp,
respectively. Splice site sequences are similar to the consensus
sequences for the intron splice sites from dicots (13). The sequence
surrounding thermost likely translational start codon correlates well with
both Kozak's (23) and Lutcke's (27) consensus sequences for the
translational start site from animal and plant genes, respectively. At
106 bp upstream from the translational start site is a putative TATA box

(23). A sequence that resembles a polyadenylation site consensus sequence

39

Figure 4. Restriction map of pAD4 and sequencing strategy. RI - EcoRI,
N - NcoI; H - HindIII; B - 8911; S - SacI; Sl - SalI. Exons are marked
with closed boxes and introns are marked with open boxes. ATG - putative
translational start codon; TAA - putative translational stop codon; Ser -
prosthetic group binding site.

4O

 

ii

 

 

 

 

 

 

 

lllv
T All
lv Iv lllv
E I _m w z m m m I z E
_ _ o _ I ll. _ F _ _ _
nx mo tom. .1 4,
<5. mm o:
as
Al.
m1
.2
aw

VD<Q to >ooembm mcficosoom 0cm omE cocoEmmE

41

Figure 5. DNA and deduced amino acid sequence of Arabidopsis ACP] gene.
The coding region of the pre ACP-1 is marked with capital letters. The
coding region for mature ACP-1 is underlined with a solid line. Putative
TATA box is underlined with a dotted line. Prosthetic group attachement

site is marked with a bold letter (S). Putative polyadenylation consensus
sequence is underlined with a dashed line.

42

DNA and demoed amino acid «meme of Arab/dope]: AGP- I gene

CO!COOIOOI.‘......O.CI°I.IO..IOO..I..IDCQIII...‘IOOOOCOIOOOOIOCOIIOI90°C!OOOICOIICIII.IOO...O
CCOIIOIOIIIOOIIOOOIICOICDICOOOIDIIDOOIOOOOIIOIOCLLQKQIIIOIIIOCIOOIOOOIOOOCIOIICOCOCIC‘IIOIIC.

M A T O F S A
ctccgccctctctccccatctctttcoacaoatctcttctctctctcototttcacgaaeca ATO OCO ACT CAA TTC AOC OCT

8 V S L O T S C L
TCT OTC TCA TTO CAA ACT TCT TOT CTO gtaatcaeatcttcttctttcattttaacaatteccttaoaoatcoctttgetctta

coattcatoatcatttcccgttgtttccatocctoctcttottetcotatataaactcctgacocatotttattatotgtaacetagaagatc
ttcoctttaotaetgatcagttttetgtttttottccaatoattttatttctagttteaaattctgtgatcogcaecacatgoeatteaaatc
cgtggttttotottctogttotaatctegatctacagtceattoaatttatgettgtgoaagtoaatgcttacttttggctotoaateattgt
ocatggacattcattgoctaegaatttoocttgaottatoeaaettcagaacaaagttcttatoaetocaatoctttttttttttgtgtgtgt

A T T R I S F O K P A L I S N H G K T
ttgatgaaecaaetag OCA ACA ACA AGO ATT AOT TTC CAA AAO CCA OCT TTO ATT TCC AAC CAT GOA AAO ACT

N L S F N L R R S l D s H R L S v s C A
AAT CTA TCC TTC AAC CTC COC COT TCA ATC CCA TCT GOO GOO CTC TCT OTT TCT TOC OCO atetgagcatttt

.A, .K a E T l
ttttcaaotttottaatctetatatatesssctcaaaottttaeaeoacattatatttootttetao OCA AAA CAA GAO ACG ATA

.E K’ V' 5 A I K K' K' £1 L $41.L .f P D K' .K V' V .A .E T
GAG AAA GTG TCT GCT ATA GTT AAG AAG CAA CTA TCA CTT ACA CCG GAT AAA AAA GTC GTT GCA GAA ACC

23;, F A a 1 fit A a I 1 a 7
AAA TTT OCT OAC CTT GOA OCA OAT TCT CTC OAC ACO tatccattcactcaagtogcaattaacecaeattcttaettttct

9' E L V Al 6 1_ E' E: E' F M l a in
toatectgetcaectoegtgttttrcttcag OTT GAG ATA OTA ATG OOT TTA GAO OAA GAG TTT AAC ATC CAA ATG

A. E .E K A, O .K L .A T Y’ .E 0 .A A .E L l E E L 41 .N
GCC GAA GAG AAA GCA CAG AAG ATT GCC ACA GTT GAG CAA GCT GCT GAA CTC ATT GAA GAG CTC ATC AAC

4£__JL__£L. 0
GAO AAO AAO tee ttttegctttateeeetgcccttegg!e‘pctaeeeeeoeeeecoeceeeeeeeeecegggcgttgagtttgttttc

IIiSIOiI'OIII'CICIOICIIIIICIIIIOIOIO'CIIOCOIOIC‘OCCIIIOICCCII‘OO'15.9!IICIOCIIOISICCICOOIICIC
IO!!!

43

(AATAAA) (18) is only 25 bp downstream from the stop codon.

Structure of the m2 cDNA clone. 61:10" plaques from the Arabidopsis cDNA
expression library made in A - ZAP were screened using two rabbit antisera

prepared against spinach ACP-I (2S) and recombinant spinach ACP-I (3).

Twelve putative positive clones were further screened with anti-spinach
ACP-I antibodies and with [32P1ATP end-labeled 66-mer. 'rwo recombinant
phagemids that were picked as the putative ACP clones were excised
according to the Stratagene protocol. DNA was digested with EcoRI and
analyzed on an agarose gel. Only one of the two clones hybridized to the
probe. Partial restriction mapping revealed that the clone, pAL120-7, has
two EcoRI fragments. Based on the DNA blot hybridization data the 0.7 kb
.EcoRI fragment, which hybridized to the end-labeled 66-mer, was subcloned.
A partial restriction map of the newly formed pALlZO-E (Fig. 6) was used
to develop a strategy for sequencing. DNA sequencing of the EcoRI
fragment of the pALlZO-E clone revealed a 716 bp fragment with an open

reading frame encoding 136 amino acids (Fig. 7).

Both the nucleic acid and deduced amino acid sequences have 70% identity
with Arabidopsis ACP-1, indicating that these two clones encode
substantially different ACPs (Table 1). 16 out of 19 amino acids from the
highly conserved region surrounding the prosthetic group binding site of
ACP-2 were identical to the amino acid composition of the same region from
other plant ACPs (Fig. 3). ACP-2 has a 52 amino acid transit peptide
based on the sequence comparisons between the putative translational start

site and N-terminal amino acid sequences of other mature ACPs. Recently,

44

Figure 6. Restriction map and sequencing strategy for pALlZO-E. RI -
EcoRI; H - HindIII; s - SacI. ATG - the putative translational start

codon; TAG - the putative translational stop codon; Ser - prosthetic group
binding site.

 

/\
A
/\
W
A

V
A

I E m I E

5
4
ox F .O _ ..0w 3?. 009020 _

@(P 3.83 gong 0P<

 

46

Figure 7. DNA and deduced amino acid sequence of an Arabidopsis ACPZ cDNA
clone, pALlZO-E. The mature ACP-2 coding region is underlined with a
solid line. The putative prosthetic group binding site is marked with a
bold S (8). The putative termination codon is marked with an asterix.
The putative polyadenylation consensus sequence is marked with a dashed

line.

M
ata

sct

ctt

9!!

GO.

A 5
act tcc

09‘ 3..

Get ICC

V H
91¢ 00.

L e
ctt 09‘

E A!

47

mummmmuwAa-zmm

I
0!!

9t!

COC

K
CID

A
act

A

A
act

tta

0
caa

0
act

E

A
act

aac

act

etc

I
tcc

E

S
tct

p

‘1!

alt

TC.

6!!

K

A 8
act tct

s N
aac aat

S C
tcc tac

ctc II.

D r
aat aca

A O

l
0!!

G

OCT

9‘0

5

s L
tcc ctt

A K
aca aaa

A 0
act aac

E I

r

0.9 .1.

I A

CCCCTCOCctTCTQCOCCTCC‘TCICTCTCOCtCCOI'CTC'G‘ICO.‘TCITTCOTTCT

O A R P R O L I A
caa act cat cct cac caa cta aca aca

s S L 8 F N L R O
aac aat ctt aac tta aat ctc cac caa

P E r V' D
cct aaa aca ate 68¢

.K V C A
aaa ata tat 06!

E I T A A T
aaa att acc act acc

K F A
06C .0. {'1 CC!

V AV’ 6 L E
at!

E E F 6
ata aaa tta aaa aaa aaa ttt aaa

 

T V’ E O A A A L I

 

If!

E
0.9

9;! 3‘9

E L

OCO

L_ F _E
aaa ctc tta tta aaa aaa

_K

CC. 0.9

A, .K
066 0.0

tct

fl

atc OCG

aca "T 9.9 6;. 96. act aca CTC It!

1.9 ..‘O'C!T‘0!‘OCIttCOCOOOOOIOOGIIGOIGTCIOOOCCC

aaaccactatcttattatttcattaactaaaaaacaattatatctattaaaaaattttatattattttaaaaaaattattacaa
aacttatatctaacttttatctttcttctaaaacattttactttcaaccttcatcaaactctcttacaaaacttatataatctc
atttcctagltqg’aatatttttccctttoaaa

48
the sequence of an Arabidopsis genomic ACP clone, A1, analogous to ACPZ
has been reported (25). However, the genomic ACP clone did not have any
apparent polyadenylation consensus sequences. The ACP-2 cDNA clone,
pALl20-E, has 78 base pairs of additional sequence information at the 3'
nontranslated end, and a putative poly(A+) consensus site (18) is located

220 nucleotides away from the stop codon.

All known Brassicaceous ACPs have an identical sequence of seven amino
acids (met-ala-glu-glu-lys-ala-gln) located seventeen amino acids
downstream from the amino acid serine to which the prosthetic group binds
(Fig. 3). In addition, both Arabidopsis ACPs have higher sequence
similarity at the carboxy terminal regions (79% over the last 28 amino
acids) than do the amino terminal regions (54% over the first 28 amino
acids of the mature protein). Similar levels of sequence conservation

have been found in other plant ACPs.

DNA. blot hybridizations. Arabidopsis genomic DNA. was digested to
completion with EcoRI and HindIII. DNA fragments were resolved by 1%
agarose gel electrophoresis and blotted to nitrocellulose membranes.
Identical DNA blots were probed with. [33P]UTP labeled RNA probes
synthesized in vitro fromieither a full length Arabidopsis cDNA.ACP clone,
pALl20-E, or from a SalI deletion, pAD430-A, of the ACP] genomic
.Arabidopsis clone, pAD4. The pAD430-A.plasmid contained the complete ACPI
coding region‘with introns. ‘Under the hybridization conditions used these

two clones do not cross-hybridize.

49

Table 1. DNA percentage identity of the transit peptide and mature ACP
coding regions. Arabidopsis AD4 ACP — genomic Arabidopsis ACPI clone
(36); Arabidopsis AL 120-E ACP — cDNA Arabidopsis ACPZ (25); Arabidopsis
A2 ACP - genomic Arabidopsis ACP3 clone (25); B. napus ACPOS - genomic ACP
clone (9); B. campestris cDNA ACP - seed cDNA ACP clone (39); S. oleracea
ACP-I - leaf cDNA ACP clone (43); S. oleracea ACP-II - root cDNA ACP clone
(44); H. vulgare ACP-I - leaf cDNA ACP clone (15).

SO

 

 

 

 

 

 

 

 

 

 

 

057%)»; £7 E“ g" g g? 8‘ §= §‘ §5
“600.0 33 3% :3 .3 $3 $1 8 1&5 $2
@065 < < at on as o:

A $3?" 35 35 57 42 35 33 37 40
4.351?” 59 90 40 40 35 37 37 37
£31?" 6 9 9 6 35 37 35 37 35 33
W 83 71 72 73 37 37 35 39
366°37'23”“ 82 69 7o 87 37 39 37 37
.66.. 57 60 60 63 62 35 39 35
5.6116“ 58 62 61 63 62 64 35 40
”.39.” 60 61 60 65 65 65 61 40
“Am"fi‘ 59 67 64 63 62 61 63 67

 

 

 

 

 

 

 

 

 

 

 

51
The ACP] probe hybridized to a single fragment in both EcoRI and HindIII
digested genomic DNA (Fig. 8b). The 4.8 kb EcoRI and the 3.2 kb HindIII
fragments hybridizing to the probe were identical in size to the EcoRI and
HindIII fragments identified by restriction mapping, respectively. These
data indicated that the .Arabidopsis genome has a single-copy gene

homologous to the pAD4 clone.

When Arabidopsis genomic DNA was digested with HindIII or EcoRI,
respectively, there were three or two fragments detected on DNA blots
probed with ACPZ clone (Fig. 8a). The sequence information of the two
Arabidopsis ACP genomic clones that were recently reported (25) helped to
unravel the information provided by the DNA blot hybridizations. One of
the two Arabidopsis ACP genomic clones isolated by Lamppa and Jacks (25),
A1, is the genomic equivalent of the ACP2 cDNA clone, pALlZO-E, described
here. In addition, sequence information confirms that Al is linked to an
ACPJ gene (named A2 by Lamppa et al.). The linked genes, which have more
than 90% sequence identity, both have a.HindIII enzyme recognition site 60
nucleotides downstream of the translational start site in the transit
peptide coding region. In contrast, only the ACP3 gene has an EcoRI site
in its 5'untranslated region. Therefore, the three fragments detected in
the HindIII digest and two fragments from the EcoRI digest represent ACPZ

and ACP3 genes.

Genomic DNA blots were also probed with random primer labeled 66-mer. At
lower stringency conditions, the probe hybridized to a large number of

bands. Most of the hybrids melted at increasing temperatures. Even

52

Figure 8. DNA blot hybridization analysis of ACP genes in Arabidopsis.

Arabidopsis genomic DNA was digested either with.aindIII (lane 1) or EcoRI
(lane 2). EcoRI digests of pAL120-E and AD430A were used as controls.
Identical blots were probed either with [ aP1-labeled RNA probe derived
from ACPZ cDNA clone, pALlZO-E (blot A) or ACPI genomic clone, pAD43OA

(blot 8). Positions of the A HindIII/EcoRI size markers are labeled on
the right hand side.

53

u - --—5.0
-3.4

-1.9
—1.6

—1.3

‘n'fi

probed with: ACP-2 ACP—1

54
though the 66-mer was found to be a relatively universal ACP probe for the
cDNA clones, in hybridizations against genomic sequences it is a weak
probe. ‘The oligonucleotide isidesigned.to complement the highly conserved
region that surrounds the phosphopantetheine binding site of the ACP.
However, in seven sequenced plant genomic ACP clones (two from barley
(15), three from Arabidopsis (25,36), and two from B. napus (9)), this
region is interrupted by an intron. Even though it is not known whether
this occurs in every ACP gene, the poor hybridization to DNA blots with

labeled 66-mer strongly suggests that it may be.

RNA blot hybridizations. Both immunoblot and RNA blot analysis of ACP
expression in spinach revealed. the presence of tissue-specific and
constitutive ACP isoforms (44,43). Similarly, immunoblot results of
proteins separated on SDS PAGE suggested that Arabidopsis might also have
tissue-specific and constitutive ACP isoforms (2). In order to address
this question, total RNAs isolated from leaf, root and developing seed
material were size-fractionated on denaturing formaldehyde/formamide 1%
agarose gels. The RNAs were blotted to nitrocellulose and the filters
were hybridized with [32PJUTP-labeled antisense RNA probes made either from
pALlZO-E or from pAD430-A. In both cases, RNA probes hybridized to the
same length.mRNA in RNA extracts from all three tissues (Fig.9). However,
because of the high degree of identity between ACP genes, detected mRNA
may result either from a single gene expressed constitutively in all three
tissues or from very similar ACP genes (e.g. ACPZ and ACP3) expressed in

a tissue-specific manner.

55

Figure 9. RNA blot hybridizations of the total RNA purified from
Arabidopsis leaves, roots and developing seeds. Total RNA was size-
fractionated on formaldehyde/formamide agarose gels. RNA blots were
probed with either [RH-labeled antisense RNA probes derived from ACPZ
cDNA clone, pALlZO-E (blot A) or ACPI genomic clone, pAD43OA (blot B).
Lanes 1 to 3 contain total RNA from leaves (L), roots (R) and developing

seeds (S), respectively.

56

_ 1.9kb
- 1.4kb

- 0.9kb

 

probed with:

57

DISCUSSION

Two.ACP clones were isolated by screening the Arabidopsis genomic and cDNA
libraries, respectively. The 70% nucleic acid and deduced amino acid
sequence identities found between these two clones indicated that they

encode different ACP isoforms. Recently, Lamppa and Jacks (25) have
characterized two Arabidopsis ACP genomic clones, A1 and A2. The A1 clone
is a genomic analog of the ACPZ cDNA clone, pALl20-E, and thus in further
text, A1 and A2 are referred to as ACPZ and ACPJ (see Appendix). ACPZ and
ACP3 genes are linked and have only 10% difference in their nucleic acid
sequences. In addition to the three characterized Arabidopsis ACP genes,
two more ACP isoforms have been identified recently. The sequence of 25
amino acids of the amino terminus for the major leaf-specific Arabidopsis
ACP isoform has been obtained, and the nucleotide sequence for what might
be an Arabidopsis mitochondrial ACP isoform, has been characterized (Dave
Shintani and John Ohlrogge, personal comm.). Therefore, ACP isoforms in
Arabidopsis are encoded by a gene family containing at least five ACP

genes.

To date, seven ACP genomic clones [Acll and Ac13 from barley (15), ACPI
(36), ACRZ and ACPJ (25) from Arabidopsis and ACPOS and ACPO9 genomic
clones from B. napus (9)] have been sequenced from higher plants. As
mentioned earlier Arabidopsis ACPZ and.ACP3 are linked. Similarly, barley
Ac11 and Ach have been located on the same chromosome, not more than 10
kb apart (15). DNA blot hybridizations indicated that genes encoding

spinach ACP-I and ACP-II isoforms may also be linked (44).

58
All characterized ACP genes have three introns whose positions are almost
completely conserved” ‘Two introns (II and III) are in identical positions
in the coding region for mature ACPs. The position of intron I in the
transit peptide differs by a few amino acids, which is probably a result
of the variable length of the ACP transit peptides. Assuming that
Arabidopsis ACPs have alanine in the first two positions of the mature
ACP, intron II is located between the first and the second amino acid.
From the known amino acid sequences for either amino terminus or complete
mature ACP of spinach ACP-I (24) and ACP-II (33), barley ACP-I and ACP-II
(17), and B.napus seed ACP (45), ACP isoforms can have either two (spinach
ACP-II, barley ACP-I and B. napus ACP) or one (spinach ACP-I, barley ACP-
II) alanine at the amino terminus. Spinach ACP-I is the only known ACP
isoform that has alanine both as the last amino acid in the transit
peptide and, as noted before, as the first amino acid of the mature ACP.
In addition, since the last three amino acids in the ACP transit peptide
seem to be conserved (44), it is likely that Arabidopsis ACPI and ACPZ
have alanines at the first two positions of the mature ACP. A similar
organization with an intron.positioned after the second amino acid of the
mature protein has been found in the rch multigene family (10). Intron
III has been found in the middle of the highly conserved region
surrounding the phosphopantetheine binding site. Even though in most
genes, such highly conserved regions are not interrupted by introns, some
members of the chalcone synthase multigene family (31) and the
triosephosphate isomerase gene from Aspergillus nidulans (28) have an

intron in the conserved active site region.

59
Transit peptides of the Arabidopsis ACPs, like other ACP transit peptides,
have a high content of hydroxyamino acids. At the carboxy terminus of the
transit peptides an amino acid motif (I/V S C) appears to be conserved in
the majority of known ACPs. The same amino acid motif was found in some
other proteins that are imported into plastids and was proposed to be a
consensus sequence for the cleavage site (12) . However, upon more
detailed analysis of the amino acid sequences of transit peptides of

nonhomologous plastid imported proteins, that motif was not found (47).

Even though Arabidopsis and Brassica ACP transit peptides share a few
identical amino acid motifs, the overall low amino acid sequence identity,
35-40%, is similar to the low sequence identities of transit peptides of
other plastid imported proteins (20). Heijne et a1. (47) have analyzed
amino acid sequences of over twenty nonhomologous transit peptides and.did
not find any of the "homology blocks” reported by Tobin (19). Heijne and
coworkers (47) pointed out that the structures of transit peptides differ
between stroma-targeted proteins and thylakoid-targeted proteins. Transit
peptides of stroma-targeted proteins have three distinct regions. The
amino terminal region contains many Pro and.Gly residues and the first ten
amino acids are mostly uncharged. The central region is rich in Ser and
is of variable length. The carboxy-terminal region contains an increased

proportion of Arg residues and can form an amphiphilic B-strand. In
contrast, transit peptides of proteins targeted to the thylakoids have
mosaic-structures. As stroma targeted ACPs, Arabidopsis ACP-1, B. napus
ACP embryo cDNA and spinach ACP-I transit peptides are consistent with

these guidelines. However, the transit peptides of other plant ACPs lack

60

an arginine-rich carboxy terminal region.

The Arabidopsis ACPZ and ACP3 genes are very similar (90% at the nucleic
acid level) (Table 1). Even their transit peptides are highly identical
at both amino acid and nucleic acid levels. The high degree of identity
between these two clones indicates that ACP2 and ACP3 are result of a very
recent gene duplication event. Similarly, coding regions of the B. napus
and B. campestris ACPs have high degrees of identity. Since B. napus is
an allotetraploid originating from a cross between B. campestris and B.
oleracea, the characterized ACP gene from B. campestris is probably the
ancestor of the ACPOS gene from B. napus. Based on the high sequence
identities of Arabidopsis ACPI with B. napus embryo cDNA ACP (41) and B.

campestris cDNA ACP (39), the three ACP genes are also closely related.

The 5' untranslated region of Arabidopsis ACPI is G+C rich. In contrast,
the 5' untranslated regions of Arabidopsis ACPZ and ACP3 are rich in
thymidines (25). Even though the primary structures of the promoter
regions of ACPI and ACPZ are different, RNA blot hybridizations appeared
to have similar expression. patterns for both .ACPI and .ACPZ genes.
However, we should keep in mind that ACP-1 and ACP-2 probes may hybridize
to similar ACP messages on RNA blots. Therefore, the differences in the
promoter sequences might result in different expression patterns of ACPI,
ACPZ and ACP3, that could not be detected by standard RNA blot analysis.
Similarly to Arabidopsis, the two ACP genes from barley, Acll and Ac13,
have different promoter sequences (47). The proximal region of the barley

Ac13 promoter has a higher G+C content than does that of AclI. In

61
addition, three GC elements that are very similar to the recognition
sequences of the Spl factor of RNA polymerase II (30) were found only in
Ac13. These features lead to the conclusion that the minor leaf ACP
isoform, that is encoded by Ac13 gene, may be a constitutively expressed

ACP isoform, similar to the ACP-II isoform from spinach (44).

The positions of the putative RNA.polyadenylation consensus sequences (18)
in the three Arabidopsis ACP genes differ. In Arabidopsis ACPI, the
putative polyadenylation site is only 24 nucleotides away from the
translational stop site. The ACPZ gene has an identical sequence 220
nucleotides downstream from the stop codon. Two putative sequences,
similar to the polyadenylation consensus sequences were found in the ACP3
gene, 232 and 259 nucleotides 3' 10f its translational stop codon.
Possibleemultiple polyadenylation sites were also found in barley Ach and

Ac13 genes (15), as well as in spinach ACP-II (44).

In spite of the small genome size of Arabidopsis, ACP isoforms in that
crucifer are apparently encoded by a large gene family. In B. napus, the
gene family of seed-expressed ACP isoforms is estimated to have
approximately 35 genes (41). Even.when we take into account the fact that
B. napus is an allotetraploid, the number of ACP genes is still high. In
spinach, too, the low number of ACP isoforms identified by inumnoblot
analysis (24) differs from the information.obtained from DNA hybridization
studies. Evidence from DNA hybridizations indicates that more than two
ACP genes are present in the spinach genome (44), the authors have

suggested that some of them may be pseudogenes and/or some may encode

62

identical proteins.

Arabidopsis ACP-1 and ACP-2 derived nucleic acid probes turned out to be
very specific, detecting only ACP genes that are identical (ACPI and ACPZ)
or very similar to the probes (ACPJ). The use of a less specific probe
such as the 66-mer (15) that is complementary to the highly conserved
region that surrounds the prosthetic group binding site was ineffective
with genomic DNA, because of the position of the third intron. In all ACP
genes that have been characterized so far, intron III is located in the
middle of that highly conserved region. Therefore, under low stringency
conditions, the 66-mer hybridized to a large number of fragments, whereas
under increased stringency washing conditions, most of the hybrids were

melted.

In conclusion, I have characterized two ACP clones, ACPI and ACPZ which
have 70% identity at the nucleotide level. DNA blot hybridizations with
gene specific probes could detect three ACP genes. However, the use of a
less specific probe and continued work by Dave Shintani indicated that
Arabidopsis ACP isoforms are encoded by a gene family that has more than
three genes. The two ACP genes, ACPI and ACPZ, seem to be expressed in
leaves, roots and developing seeds. Even though their expression pattern
is similar to the constitutive expression pattern of the spinach ACP-II
isoform, the possibility that similar but not identical ACP messages were
detected could not be ruled out. Additional characterization of the ACP
clones combined with the RNase protection analysis of their expression

patterns will be necessary for the complete characterization of the ACP

gene family in Arabidopsis.

63

10.

64

Bibliography

Arumuganathan, R. and Earle, B.D. (1991) Estimation of nuclear DNA
content of plants by flow cytometry. Plant Molecular Biology Reporter
9; 229-241.

Battey, J.F. and Ohlrogge, J.B. (1990) Evolutionary and tissue-
specific control of expression of multiple acyl-carrier protein
isoforms in plants and bacteria. Planta 180; 352-360.

Beremand, P.D., Elmore, B.D., Dziewanowska, K. and.Guerra, D.J. (1989)
Expression of an active spinach acyl carrier protein-I/protein A gene
fussion. Plant Mol. Biol. 12; 95-104.

Berger, S.L. (1987) Quantifying 32P-labeled and unlabeled nucleic
acids. Heth. Enzymology, Vol.152. Berger, S. L. and Ximmel, A.R.,
Edge, ppe 49-54e

Browse, J., McCourt, P. and Somerville, C. (1986) A mutant of
Arabidopsis deficient in C13,3 and C16,; leaf lipids. Plant Physiol. 81;
859-864.

Browse, J., Warwick, N., Somerville, C.R. and Slack, C.R. (1986)
Fluxes through the prokaryotic and eukaryotic pathways of lipid
synthesis in the "16:3” plant Arabidopsis thaliana. Biochem. J. 235;
25-31e

Chang, C. and Meyerowitz, E.M. (1986) Molecular cloning and DNA
sequence of the Arabidopsis thaliana alcohol dehydrogenase gene. Proc.
Nth. ACQd. 3C1. USA 83; 1408-14120

Chang, C., Bowman, J.L., DeJohn, A.W., Lander, E.S. and Meyerowitz
E.H. (1988) Restriction fragment length polymorphism linkage map for
Arabidopsis thaliana. Proc. Natl. Acad. Sci. USA 85; 6856-6860.

de Silva, J., Loader, N.M., Jarman, C., Windust, J.H.C., Hughes, 8.
and Safford, R. (1990) The isolation and sequence analysis of two
seed-expressed acyl carrier protein genes from Brassica napus. Plant
Mol. Biol. 14; 537-548.

Dean, C., Pichersky, E. and Dunsmuir, P. (1989) Structure, evolution,
and regulation of Rch genes in higher plants. Annu. Rev. Plant
Physiol. Plant Mol. Biol. 40; 415-439.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

65

Estelle, M.A. and Somerville, C.R. (1986) The mutants of Arabidopsis.
Trends in Genet. 4; 89-93.

Gavel, Y. and von Heijne, G. (1990) A conserved cleavage—site motif in
chloroplast transit peptides. FEBS Lett. 261; 455-458.

Hanley, B.A. and Schuler, M.A. (1988) Plant intron sequences: evidence
for distinct groups of introns. Nucleic Acids Res. 16; 7159-7176.

Hansen, L. and von Wettstein-Knowles, P. (1991) The barley genes Acll
and Ac13 encoding acyl carrier proteins I and III are located on
different chromosomes. Mol. Gen. Genet. 229; 476-478.

Hansen, L. (1987) Three cDNA clones for barley leaf acyl carrier
proteins I and III. Carlsberg. Res. Commun. 52; 381-392.

Hall, T.C., Ma, Y., Buchbunder, B.U., Pyne, J.W., Sun, S.M. and Bliss,
F.A. (1978) Messenger RNA for G1 protein of French bean seeds: cell-
free translation and product characterization. Proc. Natl. Acad. Sci.
USA 75; 3196-3200.

Hoj, P.B. and Svendsen, I. (1983) Barley acyl carrier protein: its
amino acid sequence and assay using purified malonyl-CoA:ACP
transacylase. Carlsberg Res. Commun. 48; 285-305.

Joshi, C.P. (1987) Putative polyadenylation signals in nuclear genes
of higher plants: a compilation and analysis. Nucleic Acids Res. 15,
9627-9640.

Karlin—Neumann, G.A. and Tobin, E.M. (1986) Transit peptides of
nuclear-encoded chloroplast proteins share a common amino acid
framework. EMBO J. 5; 9—13.

Keegstra, R., Olsen, L.J. and Theg, S.M. (1989) Chloroplastic
precursors and their transport across the envelope membranes. Annu.
Rev. Plant Physiol. Plant Mol. Biol. 40; 471-501.

Koorneef, M., van Eden, J., Hanhart, C.J., Stam, P., Braaksma, F.J.
and Feenstra, W.J. (1983) Linkage map of Arabidopsis thaliana. J.
Heredity 74; 265-272.

Kopczak, S.D., Haas, N.A., Hussey, F.J., Silflow, C.D. and Snustad,
D.P. (1992) The small genome of Arabidopsis contains at least six
expressed a-tubulin genes. Plant Cell 4; 539-547.

Kozak, M. (1989) The scanning model for translation: an update. J.
Biol. Chem. 108; 229-241.

Kuo, T.M. and Ohlrogge, J.B. (1984) The primary structure of spinach
acyl carrier protein. Arch. Biochem. Biophys. 234; 290-296.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

66

Lamppa, G. and Jacks, C. (1991) Analysis of two linked genes coding
for the acyl carrier protein (ACP) from Arabidopsis thaliana
(Columbia). Plant Mol.Biol. 16; 469-474.

Leutwiler, L.s., Meyerowitz, E.M. and.Tobin, E.M. (1986) Structure and
expression of three light-harvesting chlorophyl a/b binding proteins
in ArabidOpsis thaliana. Nucleic Acids Res. 14; 4051-4064.

Lutcke, H.A., Chow, K.C., Micke, E.S., Moss, K.A., Kern, H.F. and
Scheele, G.A. (1987) Detection of AUG initiation codons differs in
plants and animals. EMBO J. 6; 43-48.

McKnight, G.L., O'Hara, P.J. and Parker, M.L. (1986) Nucleotide
sequence of the triosephosphate isomerase gene from Aspergillus
nidulans: Implication for a differential loss of introns. Cell 46;
143-147.

Meyerowitz, E.M. (1987) Arabidopsis thaliana. Ann. Rev. Genet. 21; 93-
111.

Mitchell, P.J. and Tjian, R. (1989) Transcriptional regulation in
mammalian cells by sequence-specific DNA binding proteins. Science
245; 371-378.

Niesbach-Klogsen, U., Barzen, E., Bernhardt, J., Rohde, N., Schwarz-
Sommer, 2., Reif, H.J., Wienand, U. and Saedler, H. (1987) Chalcone
synthase genes in plants: a tool to study evolutionary relationships.
J. Mol. Evol. 26; 213-225.

Ohlrogge, J.B., Browse, J. and Somerville, C.R. (1991) The genetics of
plant lipids. Biochem. Biophys. Acta 1082; 1-26.

Ohlrogge, J.B. and Kuo, T.M. (1985) Plants have isoforms for acyl
carrier protein that are expressed differently in different tissues.
J. Biol. Chem. 260; 8032-8037.

Ohlrogge, J.B. (1987) Biochemistry of plant acyl carrier proteins. In
The Biochemistry of plants. A comprehensive treatise. Vol.9. P.R.
Stumpf, ed., Academic Press, Hartcourt Brace Jovanovich, Orlando,
pp.l37- 158.

Ohlrogge, J.B., Kuhn, D.N. and Stumpf, P.R. (1979) Subcellular
localization of acyl carrier protein in leaf protoplasts of Spinacia
oleracea. Proc. Natl. Acad. Sci. USA 76; 1194-1198.

Post-Beittenmiller, M.A., Hlousek-Radojcic, A. and Ohlrogge, J.B.
(1989) DNA sequence of a genomic clone encoding an Arabidopsis acyl
carrier protein (ACP). Nucleic Acids Res. 17; 1777.

Pruitt, R.E. and Meyerowitz, E.M. (1986) Characterization of the
genome of Arabidopsis thaliana. J. Mol. Biol. 187; 169-183.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

67

Redei, G.P. (1974) Arabidopsis thaliana. In Handbook of Genetics
Vol.2. King, R.C. Ed., Plenum Press, New York, pp. 151-180.

Rose, R.E., DeJesus, C.E., Moylan, S.L., Ridge, N.P., Scherer, D.E.
and Xnauf, V.C. (1987) The nucleotide sequence of a cDNA clone
encoding acyl carrier protein (ACP) from Brassica campestris seeds.
Nucleic Acids Res. 15; 7197.

Roughan, P.G. and Slack, C.R. (1982) Cellular organization of
glycerolipid metabolism. Ann. Rev. Plant Physiol. 33; 97-132.

Safford, R., Windust, J.H.C., Lucas, C., DeSilva, J., James, C.M.,
Hellyer, A., Smith, C.G., Slabas, A.R. and Hughes, S.G. (1988)
Plastid-localised seed acyl carrier protein of Brassica napus is
encoded by a distinct, nuclear multigene family. Eur. J. Biochem.
174; 287-295.

Sambrook, J., Fritsch, E.F. and.Maniatis, T. (1989) Molecular Cloning.
A Laboratory Manual, 2“ ed., Cold Spring Harbor; Lab. Press.

Scherer, D.E. and Knauf, V.C. (1987) Isolation of a cDNA clone for the
acyl carrier protein-I of spinach. Plant Mol. Biol. 9; 127-134.

Schmid, K.M. and Ohlrogge, J.B. (1990) A root acyl carrier protein-II
from spinach is also expressed in leaves and seeds. Plant Mol. Biol.
15; 765-778.

Slabas, A.R., Harding, J., Hellyer, A., Roberts, P. and Bambridge,
H.E. (1987) Induction, purification and characterization of acyl
carrier protein from developing seeds of oil seed rape (Brassica
napus). Biochem. Biophys. Acta 921; 50-

Snustad, D.P., Haas, N.A., Kopczak, S.D. and Silflow, C.D. (1992) The

small genome of Arabidopsis contains at least nine expressed B-tubulin
genes. Plant Cell 4; 549-556.

von Heijne, G., Steppuhn, J. and Herrmann, G. (1989) Domain structure
of mitochondrial and chloroplast targeting peptides. Eur. J. Biochem.
180; 535-545.

Whitaker, B.D. (1986) Fatty-acid composition of polar lipids in fruit
and leaf chloroplasts of "16:3"- and "18:3"-plant species. Planta 169;
313-319.

CRARACTBRIIRIION OF ARABIDOPSIS RCYL CARRIER PROTEINS

ABSTRACT

Arabidopsis tissues reveal a complex pattern of expression of ACP
isoforms. Immunoblot analysis of the leaf, root and seed protein extracts
separated by native PAGE and denaturing 1M urea PAGE indicated that each
tissue, in addition to the constitutive ACP isoforms, expresses one or
more tissue-specific isoforms. The inmunoblot identifications of ACP
bands were confirmed by acylation of ACP extracts with E. coli acyl-ACP

synthetase.

Coding regions for the mature protein of the three characterized
Arabidopsis.ACP clones (ACPI, ACPZ and.ACP3) were expressed in.Escherichia
coli. Recombinant Arabidopsis ACPs were partially purified on an ion-
exchange DE-52 column. It was shown that each of the three ACPs have
different mobilities on polyacrylamide gel electrophoresis. In addition,
each of the identified ACP isoforms commigrated with a different

Arabidopsis ACP isoform expressed in leaves, roots and developing seeds.

68

69

Introduction

Acyl carrier proteins are’one of the best analyzed components of the fatty
acid synthase in plants (20). However, although a large number of cDNA
and genomic ACP clones have been characterized to date, only a few ACP
isoforms have been identified and correlated with the corresponding DNA
sequences. In spinach, ACP-I and ACP-II cDNA clones (30,31) and the
mature isoforms (17,22) have been characterized. The ACP-I isoform is a
spinach major leaf isoform, whereas ACP-II is a minor isoform. The
separation of barley ACP isoforms on DEAE columns (14) and the
palmitoylation of ACPs with (“Clpalmitate, revealed the presence of the
three ACP isoforms (3). For the major leaf ACP-I isoform, and one of the
minor leaf isoforms, ACP-II, amino acid sequences have been obtained (14).
Recently, three cDNA.and two genomic ACP clones were isolated from barley.
The cDNA clones encode ACP-I (12), ACP-II (11), and a third isoform,
designated ACP-III (12), whose expression has not yet been defined. The
two barley ACP genomic clones, Acl1 and Ac13, encode ACP-I and ACP-III
isoforms (13). The analysis of 5'untranslated regions of barley Ac13
revealed several DNA motifs similar to the ones found in mammalian
housekeeping genes. Therefore, the authors proposed that the Ac13 gene

might be constitutively expressed as is the case with spinach ACP-II.

Initial work on the characterization of the Arabidopsis ACP isoforms (3)
revealed a similar pattern of expression that of spinach and barley. In
Arabidopsis leaves, one major and two minor isoforms were detected by

polyclonal antibodies raised against spinach ACP-I. Identical or similar

70
RNA messages to the Arabidopsis ACPI and ACPZ mRNAs were shown to be
expressed in leaves, roots and developing seeds. Because ACP-l and ACP-2
RNA probes did not cross-hybridize with each other, under the conditions
used, described in the Methods section of the Chapter two, it could be
concluded that at least two different ACP isoforms were expressed in
leaves, roots and seeds. However, only a single ACP isoform was detected
by spinach ACP-I antibodies in Arabidopsis root protein extracts that were
resolved by denaturing SDS PAGE (3). Because of the difference in the
number of ACP isoforms present in roots at the RNA level with the number
of detected isoforms by spinach ACP-I antibodies, I wanted to use a

different system for a detailed analysis of Arabidopsis ACP isoforms.

According to Ohlrogge et al. (21), the ACP concentration in spinach leaf
chloroplasts was calculated to be approximately 8 MM, which is 10 to 25

times lower than the estimated ACP concentration in E. coli (20). Thus,
the purification of ACP isoforms from Arabidopsis tissues for the amino
acid sequencing and for the biochemical characterization of the proteins
would require large amounts of the tissue to start with. For that reason,
coding regions for mature ACP-1, ACP-2 and ACP-3 were expressed in E. coli
as a source' of the .Arabidopsis isoforms for the protein analysis.
According to Guerra et al. (4), the expression of synthetic spinach ACP-I
clone in E. coli at levels above that of constitutive E. coli ACP did not
significantly affect bacterial growth. Similarly, it was expected that
Arabidopsis ACP isoforms would not affect the growth of bacterial cells
either. Therefore, using Arabidopsis ACPs expressed in E. coli as

standards, I wanted to define the isoforms encoded by the three

71
characterized Arabidopsis ACPJ, ACP2 and ACP3 genes and to determ ine the
number of ACP isoforms expressed in protein extracts in different

Arabidopsis tissues.

Experimental procedures

Material. Arabidopsis thaliana (L.) Heynh. var. Columbia plants were
grown either in soil or in hydroponic cultures as described in Chapter 1.
E. coli DH5a strain was used for the subcloning procedures and for the

expression of the mature Arabidopsis ACPs.

Subcloning and sequencing. Restriction digestion analysis of the plasmids
was done according to the enzyme manufacturers' protocols. Plasmid DNA
purification, DNA agarose gel analysis and subcloning of the fragments
were performed as described in the Methods section of Chapter 1.
Sequencing of the generated plasmids was done following the U.S.
Biochemical protocol for double stranded DNA sequencing using dideoxy

chain termination sequencing method and Sequenase 2.0.

Oligonucleotide purification. Oligonucleotides that were used for
polymerase chain reactions (PCR) were desalted on C-18 Sep-Pak Cartridges
according to the manufactureres' protocols with some modifications.
Cartridges were first rinsed with 10 ml of methanol, then with 10 ml of
ddHyD and finally with 10 ml of 50 mM TEA-Ac. 3 ml of oligonucleotide

solution that was prepared by mixing 500 pl of the original

72
oligonucleotide solution made in ddeO, with 2500 #1 of 50 mM TEA-Ac, was
loaded on the column. Salts were washed off the column with 10 ml ddHfiD,
and Oligonucleotides were eluted from the column with 3 ml of 50 mM TEA-
Ac:methanol (1:1). 1 ml fractions were collected, and the concentration
of the oligonucleotide solution was calculated from measurements of the UV

absorbance at 260 nm.

Polymerase chain reaction (PCR). PCR was done with plasmid DNA templates
(pA1120-E, pJD120, pAD4) using Taq DNA polymerase (Perkin Elmer Cetus)
according to the instructions provided by the manufacturer. Initial
melting of the DNA template was done at 94°C for 3 minutes. Thirty cycles
of 1 minute of melting at 94°C, two minute annealing at 37°C and two minute
extension at 72°C, were followed by seven minute complete extension at
72°C. PCR amplified fragments were analyzed in Tris-borate buffer either

by l - 1.2 % agarose gel electrophoresis or by PAGE.

Expressison of Arabidopsis ACP-l, ACP-2 and ACP-3 isoforms in B.coli.
(1) Construction of the expression plasmid with ACP-1 mature coding
region. The coding region for mature ACP-1 was produced by PCR-mediated
amplification of the third and fourth exons from the genomic clone, pAD4.
Four 21-nucleotide oligomers complementary to the 5' and 3' regions of the
two exons were designed with each carrying a different restriction site
(NcoI, XhoI, Sell or HindIII). The XhoI and SalI sites that were
introduced at the 3' and 5' ends of the coding regions of the third and
fourth exons, respectively, were destroyed following ligation. Codon

degeneracy was used to preserve the amino acid sequence of the generated

73
mature ACP-1, while the two restriction sites were introduced into the
coding regions to enable the ligation of the third and fourth exons (Fig.
10). The recombinant fragment was subcloned into the pTrc 99B expression
vector (1), and the newly generated plasmid, pAD4cm, was introduced into

B.coli DHsa strain.

(ii) Construction of the expression plasmid with mature ACP-2 coding
region. A 24 nucleotide oligonucleotide and the universal sequencing
primer were used as primers to amplify a 0.5 kb fragment encoding mature
ACP-2 protein. The 24-mer introduced an NcoI site at the codon for the
first amino acid of the mature protein. The amplified fragment was
subcloned by blunt end ligation into pBS(+) vector. Following digestion
with Noel and EcoRI, the fragment was excised and subcloned into pTrc 99B

expression vector (Fig.11a). The resulting plasmid, pJD120, was

introduced into B.coli DH5a strain.

(iii) Construction of the expression plasmid with ACP-3 mature coding
region. A third ACP gene, A2, described by Lamppa and Jacks (19), differs
from ACP-2 in the mature coding sequence by only a single amino acid. At
position 80, ACP-2 has Phe, whereas ACP-3 has Leu. Therefore, site-
directed mutagenesis was used to change the amino acid Phe to generate a
recombinant plasmid with a coding region for the mature ACP-3. A 22
nucleotide long primer that contained the codon for Leu instead of the
codon for Phe at amino acid 80, and universal sequencing primer were used
to PCR amplify a SacI/EcoRI fragment of the pAllZO-E clone. The fragment

was then digested with both enzymes and subcloned into the truncated

74

Figure 10. Schematic representation of the subcloning strategy for
expression of mature coding regions of Arabidopsis genomic ACP-1 clone in
E. coli. A) Schematic representation of the structure of Arabidopsis.ACP1
gene. ATG - putative translational start site; Ser - prosthetic group
binding site; TAA - translational stop codon. B) Subcloning of the PCR
fragments into the pTrc99B expression vector. JO24 2
5'ttgccatggcggcaaaacaagagacg3'; J025 = 5'atggtcgaccgtgtcgagagaatcth';
J026 = 5'cagctcgagatagtaatgggtttaga3'; J027 = 5'ttaaagcttaaattacttcttct
cgt3'. RI = EcoRI; H = HindIII; S = SacI; N = NcoI; B = BclII; Sl - SalI;
x = XhoI. Solid black regions in the map denote coding sequences, white
boxes denote intervening sequences, and lines represent 5' and 3'
untranslated sequences.

75

 

. .

X
Z
“I

Z

 

a
I
o
-----
eeeee
e s
o
a
'

as
s
sssss
sees
sees
as
as
as

sea

sass
ale
0
sssssssssss
ssssssssssssssssssss

 

 

76
pJD120 plasmid to replace its original fragment (Fig.11b). The newly

generated plasmid, pMZlZOL, was introduced into B.coli strain DH5a.

The fidelity of the amino acid sequence of the three mature ACP isoforms
was confirmed by sequencing portions of the generated clones using the

dideoxy chain termination sequencing method.

Expression of nature Arabidopsis ACP isoforms in E. coli and their
purification. E. coli transformed with either pAD4cm (ACP-1) or pJD120
(ACP-2) was grown at 37°C in 1 l of LB supplemented with 100 [Lg/ml
ampicillin and 0.5 mM IPTG to ODam of 4.7. ACPs were isolated according
to the procedure of Guerra et al. (4) with a few modifications. Cells
were collected by centrifugation at 600 rpm for 30 minutes. The cell
pellet was resuspended in 10 volumes of homogenization buffer (0.1M Tris
HCl, pH 7.0, 20 mM glycine, 1 mM EDTA, 1 mM DTT). After the addition of
0.3 mg of lysozyme, the homogenate was stirred for 2 hours. At the end of
the second hour, 1/2 volume of 0.5% (v/v) Triton x—1oo was added and the
stirring continued. Bacterial DNA was degraded by incubation of the
homogenate with 100 pg of DNase for 10 minutes with continued stirring.
Following a few bursts with the polytron, a cell free supernatant was
obtained by centrifugation at 10,000 rpm for 15 minutes. The supernatant
was adjusted to 60% saturation with (NH4)ZSO4 and left overnight at 4°C
with constant stirring. Precipitated proteins were collected by
centrifugation at 10,000 rpm for 15 minutes. Soluble proteins from the
supernatant were precipitated by adding TCA to 5% followed by incubation

on ice for one hour. After centrifugation at 10,000 rpm for 15 minutes,

77
the pellet was redissolved in 20 mM MES, pH 6.1, containing 1 mM DTT. The
pH was brought to 7.2, and the solution was applied to a DE-52 column
which was equilibrated with 20 mM MES, pH 6.1, containing 1 mM DTT. Once
the protein extract was applied to the column, the column was washed with
the same buffer that was used for its equilibration. Proteins were eluted
with 100 ml of a NaCl linear gradient (0 M to 0.5 M) made in 20 mM MES, pH
6.1, containing'l mM DTT; 1-2 ml fractions were collected. Fractions were

assayed for ACP by an acyl-ACP synthetase assay (27).

A 300 m1 LB culture of E. coli cells transformed with pMZl20L (ACP-3) was
induced by 0.5 mM IPTGw The soluble fraction containing ACP-3 isoform'was
obtained by grinding the cell pellet in 2.5 % TCA (1:5, w/v). Tissue
debris and precipitated proteins were removed by centrifugation at 15,000
rpm for 5 minutes. By resuspending the pellet in 50 mM MOPS, pH 6.8, ACP
was brought back to solution. The solution was again centrifuged for 5
minutes at 15,000 rpm to remove insoluble material. For deacylation of
ACP, the pH of the solution was brought to between 8.5 and 9.0 and DTT was
added to 100 mM. The extract was then incubated for 10 minutes at 37°C.
Following precipitation of proteins in 10% TCA, ACPs were dissolved in 50

mM mops, pH 6.4-6.8.

Polyacrylanide gel electrophoresis and innunoblot analysis. SDS/PAGE was
performed as described by Laemmli (18). Proteins were resolved on a 3%
polyacrylamide/0.4% bisacrylamide stacking and 15% polyacrylamide/0.4%
bisacrylamide 1.5 mm thick resolving gel in a 0.025 M Tris HCl, pH 9.0,

0.19 M glycine, 10 % (w/v) SDS buffer system. Gels were run at 30 mA/gel

78

Figure 11. Schematic representation of the strategy for expression of the
mature coding regions of Arabidopsis ACP-2 (pALl20-E) and ACP-3 in E3
coli. A) Structure of the pAL120-E cDNA clone. ATG - putative
translational start site; Ser - prosthetic group binding site; TAG -
translational stop codon. B) Strategy used for the PCR amplification and
cloning of the mature ACP-2 coding region and 3' nontranslated region.
C) Mature ACP-3 was constructed by replacing the SacI/EcoRI fragment from
pJDl20 with a PCR-amplified SacI/EcoRI fragment that contained the codon
for Leu instead of the codon for Phe (marked with a bold letters in the
sequence of the primer and with a star in the figures). J021 -
5'ccatgggctgcaaaacctgagaca3; JO43 = 5'tgaggagctcttgttggaaaagB'; UP 8
5'actggccgtcgttttac3'; RI = EcoRI; H = HindIII; S = SacI; N - NcoI.

79

 

 

a)
pAL 120-E
ATG mgfl": TAG 0.1 kb
Jr 1 J, ‘—
I . T I -
n1 3 RI 3
g:
12.5

 

 

0.0120

c)

RI

 

80

for 5 hours. Prior to loading, ACP samples were reduced with 50 mM DTT.

Native PAGE and denaturing urea PAGE were prepared according to Rock &
Cronan (26). A 13% polyacrylamide/0.4% bis(acrylamide) resolving gel was
overlayered with 3% polyacrylamide/bis(acrylamide) stacking gel. Urea
gels were a modification of native PAGE. 1M urea was included both in the
stacking and in the resolving gel as well as in an sample buffer. Samples
were deacylated in 100 mM DTT at pH 8.5 - 9.0, and ACPs were blocked with
20 mM NEM. Gels were run in a 0.025 M Tris HCl, pH 9.0, 0.19 M glycine

buffer system at 30 mA/gel for 3 hours.

The gels were electrophoretically transblotted to 0.2 um nitrocellulose
filter in a 39 mM glycine, 48 mM Tris HCl, pH 9.0, 20% (v/v) methanol
system at 0.8 mA/mm2 of the gel surface. Proteins were fixed onto the
filter with a 3 minute incubation in 5% (w/v) TCA, followed by thorough

washing in ngD.

Nitrocellulose was blocked with Blotto for 2 hours. Spinach ACP-I
antibodies were added to the fresh TBST buffer in a dilution of 1:500.
Filters were immunoblotted for one hour. The primary antibody solution
was poured off and filters were washed three times with TBST for 10
minutes each. Secondary; goat. anti(rabbit IgG)-alkaline-phosphatase
conjugated antibodies (1:2,000), were added to fresh blocking solution and
incubated for one hour. At the end of the incubation, the antibody
solution was poured off and filters were washed three times in TBST for 10

minutes each. Immunostaining was carried out in alkaline phosphatase

81
buffer with NBT and BCIP dissolved in 70% and 100% N,N-dimethylformamide,

respectively. Color development was stopped by thorough rinsing in dngD.

Preadsorbing of the antibodies with E. coli extract. A 25 ml liquid
culture of E. coli cells was inoculated from a single cell colony and left
to grow overnight at 37°C with constant shaking. Cells were then pelleted
at 6,000 rpm for 10 minutes. The pellet was resuspended in 1 ml PBS, pH
7.0 and incubated in a boiling water bath for 10 minutes. Following a 5
minute sonication, the insoluble material was pelleted by centrifugation
at 13,000 rpm for 5 minutes. 1 ml of the supernatant was mixed with 20 ml

of 1:500 primary anti-spinach ACP-I antibodies for 40 minutes before use.

Radiolabeling od ACPs. Arabidopsis ACP isoforms that were expressed and
purified from E. coli, as well as fractions that were assayed for ACPs,
were acylated with [“C]palmitate (56 mCi/mmol) in a reaction catalyzed by
an acyl-ACP synthetase (27). ACPs from the Arabidopsis leaf, root and
seed material were labeled with uniformly labeled [U-14C]palmitate (>500
mCi/mmol). The enzyme was partially purified from E. coli cells by John
Shanklin according to Rock and Cronan (27). The reaction mix contained
0.1 M Tris HCl, pH 8.0, 10 mM M9012, 0.4 M LiCl, 3x10S DPM [1‘C]-16:0, 5 mM
ATP, 5 mM DTT, sample and acyl-ACP synthetase. After one hour incubation
at 37°C, half of the reaction mix was spotted onto DE 81 filter discs that
were left to dry in a hood. Unincorporated label was washed off the
filters with three changes of 80% isopropanol, 20% phosphate saline buffer
[10 mM KPO4, pH 6.0, 0.1 M NaCl]. Following the addition of the

scintillation fluid, radioactivity was counted in a Beckman scintillation

82

counter and the acylated ACPs were resolved by native PAGE.

RESULTS

Previous studies have shown that polyclonal spinach ACP-I antibodies could
detect one and three ACP isoforms in Arabidopsis root and leaf protein
extracts, respectively, that were resolved in an SDS\PAGE system (3). The
analysis of protein extracts from Arabidopsis developing seeds with anti-
spinach ACP-I antibodies detected a single ACP isoform commigrating with
a root and with a minor leaf ACP isoform (Fig. 12). Therefore, based on
the SDS\PAGE immunoblot analysis, Arabidopsis tissues appeared to have
three ACP isoforms. However, it was noticed that mobilities of the ACPs
in SDS\PAGE were affected by the content of charged and hydrophobic amino
acids in the mature ACP (20), suggesting that different isoforms with a
similar' percentage of hydrophobic, acidic and basic residues would
comigrate in that gel system. In addition, the ACP isoforms that differ
in a single amino acid, like Arabidopsis mature ACP-2 and ACP-3 isoforms

(19), would probably not be separated by SDS/PAGE.

Native polyacrylamide gels developed by Rock and Cronan (26) and
denaturing urea PAGE were shown to have high resolving power for the
intermediates of fatty acid synthesis (24). The immunoblot analysis of
protein extracts from Arabidopsis leaf, root and developing seed tissues
that were resolved in either native/PAGE or urea PAGE indicated that each

tissue has a minimum of four ACP isoforms (Fig. 13,14).

83

E.co|i ACP

-..
,. s

SDS PAGE

Figure 12. Immunoblot analysis of ACP isoforms in Arabidopsis root, seed
and leaf protein extracts. Crude protein extracts were resolved in SDS-
15%PAGE. Blots were probed with antiserum to spinach ACP—I. L =
Arabidopsis leaf extract; R = Arabidopsis root extract; S = Arabidopsis
seed extract.

84
Surprisingly, leaves, roots and seeds had at least three comigrating ACP
isoforms, indicating that the genes encoding these isoforms might be
constitutively expressed. In addition, tissue-specific ACP isoforms were
detected in leaves, roots and seeds. The relative abundance of different
ACP isoforms varied in different tissues. The most abundant isoform in
leaves appeared to be the tissue-specific isoform. However, in roots, all
four detected putative ACP isoforms were relatively equally abundant,
whereas in seeds the three ”constitutive" ACP isoforms were more abundant

than the ones that were expressed in a tissue-specific manner.

Even though both ACPI and ACPZ mRNAs appeared to be expressed in leaves,
roots and developing seeds, their deduced amino acid sequences could not
reveal the identity of the isoforms they were encoding. Therefore, as an
initial step in defining the protein products that will enable analysis of
different functions of ACP isoforms in Arabidopsis, I have expressed
mature ACP-l, ACP-2 and ACP-3 isoforms in E. coli. For the expression of
mature ACP-1, PCR-mediated amplification of the third and the fourth exons
was followed by their sequential subcloning into the E. coli expression
vector, pTrc 99B, yielding pAD4cm (Fig. 10). The coding region for the
mature .ACP-2 was PCR-amplified and subcloned into the same vector,
yielding pJD120 (Fig. 11a ). Site-directed mutagenesis was used for the
change of the amino acid Phe at position 80 of the ACP-2, into the amino
acid Leu of the ACP-3. A DNA fragment from pJD120, that contained a codon
for Phe, was replaced by a PCR amplified fragment that contained a codon
for Leu, yielding pM2120L (Fig. 11b). Fidelity of the final subcloning

products was confirmed by sequencing portions of the fragments.

'—

ACP-2
ACP-3

I
Q.
L R DvS DrS 5:)

er

, .
*4 . .
... - - — ‘ACP—1
.M‘ ‘II.) "AfijD-2
-—

native PAGE

Figure 13. Immunoblot analysis of ACP isoforms in Arabidopsis root, seed
and leaf protein extracts. Crude protein extracts were separated using
native PAGE. Blots were probed with antiserum to spinach ACP-I. L =
Arabidopsis leaf extract; R = Arabidopsis root extract; DvS = Arabidopsis
developing seed extract; Drs = Arabidopsis dry seed extract; ACP-l =
purified Arabidopsis ACP-1 standard; ACP-2 = purified Arabidopsis ACP-2
standard; ACP-3 = E. coli extract containing expressed Arabidopsis ACP-3
isoform.

86

ACP—2
ACP-3
ACP- 1

LRS

- . L '
ACP—2 — - -
ACP-3 —
ACP-1 — -

 

Figure 14. Immunoblot analysis of ACP isoforms in Arabidopsis root, seed
and leaf protein extracts. Crude protein extracts were separated using
denaturing 1M urea PAGE. Blots were probed with anti-spinach ACP-I
antibodies. L = Arabidopsis leaf extract; R = Arabidopsis root extract; 5
= Arabidopsis dry seed extract; ACP-1 = purified Arabidopsis ACP-1
standard; ACP-2 = purified Arabidopsis ACP-2 standard; ACP-3 = E. coli
extract containing expressed Arabidopsis ACP-3 isoform.

87
E. coli DHsarcells were transformed.with one of the generated plasmids and

after induction with 5 mM IPTG the recombinant Arabidopsis ACPs were
expressed 1.5 to 3.3 fold over the basal level of expression of the E.
coli ACP. The Arabidopsis ACP-1 and ACP-2 isoforms were purified from the
E. coli ACP on a DE-52 column in a 0 to 0.5 M NaCl gradient. The ACP
activity in different fractions was measured using an acyl-ACP synthetase
(EC 6.2.1.) assay (27). Because the elution profile did not reveal any
information on the nature and origin of particular isoforms, samples from
the fractions with ACP activity were separated in SDS/PAGE and identified
by immunoblotting. Identical filters were immunobloted with either
spinach ACP-I antibodies or with the same antibodies that were blocked
with E. coli extract. In both cases, immunoblots revealed that
Arabidopsis ACP-1 (Fig. 15) and ACP-2 isoforms (data not shown) eluted
from the DE-52 column with a lower salt concentrations than E. coli ACPs.
E. coli expressed Arabidopsis mature ACP-1 was eluted with a higher salt
concentration than the Arabidopsis ACP-2 (0.3 vs 0.22 M NaCl), whereas E.
coli ACPs were eluted with a similar salt concentrations in both

purifications (0.35 and 0.4 M NaCl) (Fig. 16 and 17).

Mature.Arabidopsis.ACP-1, ACP-2 and ACP-3 isoforms that were purified from
the E. coli cells were used as standards in the protein analysis of the
Arabidopsis ACP isoforms. The three ACP isoforms were easily resolved on
both native PAGE and 1M urea PAGE. Even the ACP-2 and ACP-3 isoforms that
differ by only a single amino acid at position 80, were separated from
each other (Fig. 13 and 14). The mature ACP-1, ACP-2 and ACP-3 proteins

were found to commigrate with three putative ACP isoforms detected in

88

E.coli tfn. w pAD4cm E.coli tfn. w pAD4cm

.coli extract
E.coli ACP
before
column
fract. #33
fract. #57
.coli extract

ACP

fract. #33
fract #57

E.coli
before
column

(I
”I

anti-spinach ACP—l antibodies anti-spinach ACP—l antibodies
blocked w E.co|i prot. extr.

Figure 15. Immunoblot analysis of the column fractions from the
purification of Arabidopsis ACP—1 expressed in E. coli. Proteins were
separated by SDS-PAGE and ACPs were detected with anti-spinach ACP-I
antibodies that were either preblocked with E. coli protein extracts or
were not preblocked. Due to the high salt concentration migration of the
Arabidopsis ACP-1 in fraction #33 was slightly altered.

89

Figure 16. Purification of Arabidopsis ACP-l expressed in E. coli.
Protein extracts of IPTG induced E. coli cells that were transformed with
pAD4cm, were applied to a DE-52 ion exchange column. 1 ml fractions were
collected and assayed for ACP using [“C]palmitate and E. coli acyl-ACP

synthetase. The elution peaks of the Arabidopsis ACP-1 and E. coli ACP
are indicated by the arrows.

[“CIpalmitoyl-ACP (cbmx 1000)

6

0

90

Arabidopsis ACP-1 purification on 05-52 column
Acyl ACE? synthetase assay

ACP- 1
‘1!

Eco" ACP

 

 

 

l l l 1 l l 1 I
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 59 73 77
Fraction number

91

Figure 17. Purification of Arabidopsis ACP-2 expressed in E. coli.
Protein extracts of IPTG induced E. coli cells that were transformed with
pJD120, were applied to a DE-52 ion exchange column. 4 ml fractions were
collected and assayed for ACP using [“C]palmitate and E. coli acyl-ACP

synthetase. The elution times of the Arabidopsis .ACP-2 and E. coli ACP
are indicated by the arrows.

["C]palmitoyl-ACP (dpmx 1000)

12

1O

 

92

Arabidopsis ACP-2 purification on DE-52 column
Acyl =-ACP synthetase assay
ACP-2

Ecoli ACP

i

11"11111

4 8121620242832364044485256
Fraction No.

93
immunoblots of protein extracts of Arabidopsis leaves, roots, developing
and dry seeds, indicating that ACPI, ACPZ and ACP3 genes are expressed in
a constitutive manner. Based on relative abundances, ACP-l, ACP-2 and
ACP-3 isoforms appear to be three of the four more abundant isoforms in

seeds, whereas in leaves they are less abundant isoforms.

Some of the proteins that were detected by polyclonal spinach ACP-I
antibodies in immunoblots might possibly be cross-reacting non-ACP
proteins. In order to further evaluate the number of ACP isoforms in
Arabidopsis tissues, ACPs were specifically labeled using E. coli acyl ACP
synthetase (27) and ["‘CJpalmitate. ACP-enriched leaf, root and seed
protein fractions were acylated and the proteins were separated on native
PAGE. Proteins were transferred to nitrocellulose membranes and following
autoradiography, putative ACPs on the blot were detected by spinach ACP-I
antibodies. The autoradiogram showed.that.Arabidopsis leaf, root and seed
tissues have expressed multiple ACP isoforms, some of which are tissue-
specific (Fig. 18). A smaller number of ACP isoforms detected on an
autoradiogram compared to the number of putative deacylated ACP isoforms
on the immunoblots, could be a result of the changes in the mobilities
that occur upon acylation of ACPs (27). The acyl group is presumably
positioned in the hydrophobic cleft made by three a helical regions of the
ACP (15). Upon acylation the conformation of the ACP changes, resulting

in the changed mobility on native PAGE.

The [“C]palmitoylated ACP-1 was separated on native PAGE from the

[1‘C1palmitoylated ACP-2 isoform (Fig. 18). The ACP-2 isoform was found

94

— O. — Q
30 30
< <
9 9
R S L ‘-' 7"
[“Cbalmitoyl D .. ..... -
ACP—1 N
["Cbalmltoyl D .. «. - ..

ACP-2 -

Figure 18. Palmitoylation of Arabidopsis ACPs in root, dry seed and leaf
protein extracts. ACP-enriched crude protein extracts were acylated using
[“C]palmitate and E. coli acyl-ACP synthetase. Acyl-ACPs were separated
by native PAGE. Proteins were blotted to nitrocellulose membrane and
exposed to a film for 1 week. L = Arabidopsis leaf extract; R =
Arabidopsis root extract; 5 = Arabidopsis dry seed extract; [14C]palmitoyl
ACP-1 = Arabidopsis [14C]-palmitoyl-ACP-l standard; [14C]palmitoyl ACP-2 =
Arabidopsis [ 14C ] -palmitoyl-ACP-2 standard .

95

in all three tissues, whereas the ACP-1 isoform was found in leaves and

seeds. We cannot rule out the possibility that, due to poorer isolation
or acylation of ACP-1, that ACPl was not recovered as efficiently from the
root extracts as from the other tissues. An alternative explanation is
that an isoform other than ACP-1 which is expressed in roots comigrates
with deacylated ACP-1 on native PAGE. Such an isofonm upon
palmitoylation, might undergo slightly different conformational change

from the ACP-1 and be resolved from it upon native PAGE.

As a result of the analysis of the expression of ACP isoforms in
Arabidopsis tissues, several constitutive and tissue-specific ACP isoforms
were detected by anti-spinach ACP-I antibodies and by autoradiography in
leaves, roots and seeds. Based on the labeling studies with acyl-ACP
synthetase and on the immunoblot analysis, ACP-2 is clearly expressed in
a constitutive manner. Similarly, ACP-1 and ACP-3 seem to be
constitutively expressed isoforms. However, due to the complex pattern of
ACP isoforms expressed in Arabidopsis tissues, for the complete
characterization of their expression patterns, labeling studies at the
protein level and the analysis of the expression of each isoform at the

mRNA level using specific probes will need to be done.

DISCUSSION

In this study Arabidopsis mature ACP isoforms, ACP-1 (23), ACP-2 (this

study, 19) and .ACP-3 (19) and their' expression patterns have been

96
identified. Because of the low abundance of ACPs in plant tissues (21),
instead of purifying.ACP isoforms from.Arabidopsis tissues, coding regions
for the mature ACPs have been expressed in E. coli cells. Protein
extracts from Arabidopsis leaf, root and seed material and purified E.
coli expressed Arabidopsis ACPs were analyzed by the combination of PAGE,

immunoblotting and radiolabeling studies.

Rock and Cronan (26) reported that E. coli ACP comigrates in SDS/PAGE with
proteins of MW 20,000, even though its calculated MW is 8,847. They
pointed out that the highly acidic nature and the low number of
hydrophobic residues of E. coli ACP probably decrease the efficiency of
binding of SDS. As a result, the protein migrated more slowly than one
would expect from its molecular weight. A similar effect was reported for
plant ACP isoforms. For example, barley ACP-I of MW 9,500, comigrates
with E. coli ACP on SDS/PAGE (14), whereas spinach ACP-I (MW 8708) and
ACP-II (MW 8,899), comigrate with proteins of MW 14,500 and 16,000
(17,22), respectively. However, mature Arabidopsis ACP-1 (MW 9,159) and
ACP-2 (MW 8,962) isoforms, comigrate with proteins of MW similar to their
estimated molecular weight. The differences in the behaviour of ACPs on
SDS/PAGE might be in the ratio of acidic and basic amino acids. E. coli
ACP has more acidic and fewer basic amino acid residues than either

Arabidopsis ACP-l or ACP-2 (20 vs 18 acidic and 7 vs. 11 or 13 basic).

Initially, three ACP isoforms that were detected on immunoblots of leaf,
root and seed protein extracts resolved by SDS/PAGE (3, Fig. 12)

suggesting that Arabidopsis might have a small ACP gene family. However,

97
a minimum of seven putative ACP isoforms were separated in the protein
extracts of these tissues in native and denaturing urea/PAGE (Fig. 13 and
14). Even though the resolving power of native and urea PAGE systems was
such that ACP isoforms that differ in a single amino acid were resolved,
we still cannot exclude the possibility that some ACP isoforms with
conservative amino acid changes comigrated. Because of the stability of
ACPs and of the preparation of the protein samples multiple bands detected
by spinach ACP-I antibodies were probably not a result of the proteolysis
of ACP isoforms. As in Arabidopsis, SDS/PAGE analysis of the B. napus
seed protein extract revealed only a single ACP band (33) . However,
Safford et al. characterized eight B. napus ACP cDNA clones, encoding six
different ACPs (29). Even though no information on the abundance of any
of these isoforms is available, the results suggest that six B. napus seed
ACP isoforms commigrate in SDS/PAGE system. In addition to Arabidopsis,
B. napus, Avena sativa and H. vulgare'were found to have more than two ACP
isoforms expressed in leaf tissue (3). Even some non-vascular plants such
as Polytrichum and Marchantia, were reported to haveia.minimum of four and
six ACP isoforms in leaf tissues, respectively (3). Therefore, the
occurence of more than two ACP isoforms does not appear to be restricted
only to the Brassicaceae family. To date, spinach is the only higher
plant that has two ACP isoforms that are resolved by native/PAGE, by
SDS/PAGE and by urea/PAGE (17,22,24). There, ACP-I and ACP-II isoforms
were found in leaves, whereas only ACP-II was found in roots and seeds.
However, work of Schmid and Ohlrogge (31) indicated that more than two
genes might be encoding ACP-I and ACP-II isoforms. As they pointed out,

the apparent simplicity at the protein level, might not be represented at

98
the gene level, where different members of the spinach ACP gene family

could be differently regulated by spatio-temporal factors.

Based on the immunoblot analysis of the protein extracts resolved either
by native or by denaturing urea PAGE, Arabidopsis ACP isoforms encoded by
ACPI , ACPZ and ACP3 appear to be expressed in a constitutive manner.
Specific labeling of ACPs with E. coli acyl-ACP synthetase and
["‘Clpalmitate in leaf, root and seed protein extracts confirmed the
presence of the ACP-2 isoform in all these tissues. Similarly, a recent
report of the expression pattern of the b-glucuronidase reporter gene,
which was under the control of the ACP-2 promoter in transformed tobacco

detected the ACPZ gene promoter driven expression of GUS in every tissue

(2)-

Polyclonal anti-spinach ACP-I antibodies, that were used in this study,
have poorer cross-reactivity with prokaryotic and monocotyledonous ACPs
than with dicotyledonous ACPs (9). Therefore, some of the epitopes on the
Arabidopsis ACP isoforms are not recognized by the antibodies. However,
by using saturating levels of the antibodies in the immunoblot analysis,
the intensities of the immunoblot signals represent approximate relative
abundances of the ACP isoforms. Hence, in ArabidOpsis leaves, the three
constitutively expressed ACP isoforms are less abundant than the leaf
specific isoforms, similarly to the constitutive spinach ACP-II isoform.
In contrast to spinach, Arabidopsis appears to have isoforms that are

specifically expressed even in roots and seeds.

99
Gene families are common in the plant kingdom (25). Isoforms and isozymes
can be localized either in the same (6, 32) or in different cellular
compartments (7). While the functional importance of the isoforms in
different cellular compartments generally lies in performing similar
functions that have been compartmentalized, the functions of different
isoforms localized within the same organelle are still unknown. ACP
isoforms appear to have a complex distribution within plant cells.
Spinach ACP-I and ACP-II (21) and barley ACP-I and ACP-II (14) are found
in chloroplasts. ACP-like proteins have been detected in mitochondria
(5). Even though in vitroidata, indicated that oleoyl-ACP acyltransferase
and glycerol-3-P acyltransferase are differentially efficient with two
spinach ACP isoforms, ACP-I and ACP-II (10), no in vivo evidence to date
supports that hypothesis. The predictions of the secondary structures of
known ACP isoforms reveal very similar arrangements of the a helical and
b turn regions. Moreover, between 54% and 64% of both ACP-I and ACP-II
pools occur in a nonesterified form (24). The small variations in the
secondary structures of ACP isoforms might still be reflected in a
different effectiveness with components of the fatty acid synthase. At
the same ‘time, .ACP isoforms ‘that are located in ‘the same cellular
compartment might still be capable of substituting for each other in the
reactions of fatty acid and lipid biosynthesis. In this case the presence
of a large ACP gene family, as well of the other components of fatty acid
synthase, might permit plasticity in the expressional patterns in a

spatial and/or temporal manner.

10.

100

Bibliography

Amann, E., Ochs, B. and Abel, K.-J. (1988) Tightly regulated tac
promoter vectors useful for the expression of unfused and fused
proteins in Escherichia coli. Gene 69; 301-315.

Baerson, S.R. and Lamppa, G.K. (1992) A complex expression pattern is
directed by the Arabidopsis acyl carrier protein Al gene promoter in
transgenic tobacco. 2r" Steenbock Symposium, Cellular Communication
in Plants, May 31-June 2, 1992, University of Wisconsin.- Madison, 14.

Battey, J.F. and Ohlrogge, J.B. (1990) Evolutionary and tissue-
specific control of expression of multiple acyl-carrier protein
isoforms in plants and bacteria. Planta 180; 352-360.

Beremand, P.D., Hannapel, D.J., Guerra, D.J., Kuhn, D.N. and.Ohlrogge,
J.B. (1987) Synthesis, cloning and expression in Escherichia coli of
a spinach acyl carrier protein-I gene. Arch. Biochem. Biophys. 256;

90-100.

Chuman, L. and Brody, S. (1989) Acyl carrier protein is present in the
mitochondria of plants and eucaryotic micro-organisms. Eur. J.
Biochem. 184; 643-649.

Dean, C., Pichersky, E. and Dunsmuir, P. (1989) Structure, evolution,
and regulation of Rch genes in higher plants. Annu. Rev. Plant
Physiol. Plant Mol. Biol. 40; 415-439.

Edwards, J., Walker, E.L. and Coruzzi, G.M. Cell-specific expression
in transgenic plants reveals nonoverlapping roles for chloroplast and
cytosolic glutamine synthetase. Proc. Natl. Acad. Sci. USA 87; 3459-
3463.

Fosket, D.E. and Morejohn, L.C. (1992) Structural and functional
organization of tubulin. Annu. Rev. Plant Physiol. Plant Mol. Biol.
43; 201-240.

Froehlich, J.E., Poorman, R, Reardon, E., Barnum, S. and Jaworski,
J.G. (1990) Purification and characterization of acyl carrier protein
from two cyanobacteria species. Eur. J. Biochem. 193; 817-825.

Guerra, D.J., Ohlrogge, J.B. and Frentzen, M. (1986) Activity of acyl
carrier protein isoforms in reactions of plant fatty acid metabolism.
Plant Physiol. 82; 448-453.

11.

12.

l3.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

101

Hansen, L. and Kauppinen, S. (1991) Barley acyl carrier protein II:
Nucleotide sequence of cDNA clones and chromosomal location of the
Ach gene. Plant Physiol. 97; 472-474

Hansen, L. (1987) Three cDNA clones for barley leaf acyl carrier
proteins I and III. Carlsberg. Res. Commun. 52; 381-392.

Hansen, L. and von Wettstein-Knowles, P. (1991) The barley genes A011
and Ac13 encoding acyl carrier proteins I and III are located on
different chromosomes. Mol. Gen. Genet. 229; 476-478.

Hoj, P.B. and Svendsen, I. (1983) Barley acyl carrier protein: its
amino acid sequence and assay using purified malonyl-CoA:ACP
transacylase. Carlsberg Res. Commun. 48; 285-305.

Holak, T.A., Nilges, M., Prestegard, J.H., Gronenborn, A.M. and Clore,
G.M. (1988) Three-dimensional structure of acyl carrier protein in
solution determined by nuclear'magnetic resonance and the combined.use
of dynamical simulated annealing and distance geometry. Eur. J.
Biochem. 175; 9-15.

Kopczak, S.D., Haas, N.A., Hussey, P.J.m Silflow, C.D. and Snustad,
D.P. (1992) The small genome of Arabidopsis contains at least six

expressed a-tubulin genes. The Plant Cell 4; 539-547.

Kuo, T.M. and Ohlrogge, J.B. (1984) The primary structure of spinach
acyl carrier protein. Arch. Biochem. Biophys. 234; 290-296.

Laemmli, U.K. (1970) Cleavage of structural proteins during the
assembly of the head of bacteriphage T4. Nature 227; 680-685.

Lamppa, G. and Jacks, C. (1991) Analysis of two linked genes coding
for the acyl carrier protein (ACP) from Arabidopsis thaliana
(Columbia). Plant Mol.Biol. 16; 469-474.

Ohlrogge, J.B. (1987) Biochemistry of plant acyl carrier proteins. In
The Biochemistry of Plants. A Comprehensive Treatise. Vol.9.
P.K.Stumpf, ed., Academic Press, Hartcourt Brace Jovanovich, Orlando,
pp.137- 158.

Ohlrogge, J.B., Kuhn, D.N. and Stumpf, P.K. (1979) Subcellular
localization of acyl carrier protein in leaf protoplasts of Spinacia
oleracea. Proc. Natl. Acad. Sci. USA 76; 1194-1198.

Ohlrogge, J.B. and Kuo, T.M. (1985) Plants have isoforms for acyl
carrier protein that are expressed differently in different tissues.

Post-Beittenmiller, M.A., Hlousek-Radojcic, A. and Ohlrogge, J.B.
(1989) DNA sequence of a genomic clone encoding an Arabidopsis acyl
carrier protein (ACP). Nucleic Acids Res. 17; 1777.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

102

Post-Beittenmiller, D., Jaworski, J.G. and Ohlrogge, J.B. (1991) In
vivo pools of free and acylated acyl carrier proteins in spinach. J.

Raff, E.C., Diaz, H.B., Hoyle, H.D., Hutchens, J.A., Kimble, M., Raff,
R.A., Rudolph, J.E. and Subler, M.A. (1987) Origin of multiple gene
families: Are there both functional and regulatory constraints? In
Development as an evolutionary process. Proceedings of the meeting
held at the Marine Biological Laboratory, Woods Hole, Massachusetts,
Aug. 23 & 24, 1985, R.A. Raff and E.C. Raff, eds., Alan R. Liss, Inc.,
pp. 203-238.

Rock, C.O. and Cronan, J.E. (1981) Acyl carrier protein from
Escherichia coli. M. Enzymology 71; 341-351.

Rock, 0.0. and Cronan, J.E. (1979) Re-evaluation of the solution
structure of acyl carrier protein. J. Biol. Chem. 254; 9778-9785.

Rock, C.O. and Cronan, J.E. (1979) Solubilization, purification and
salt activation of acyl-acyl carrier protein synthetase from
Escherichia coli. J. Biol. Chem. 254; 7116-7122.

Safford,R., Windust, J.H.C., Lucas, C., DeSilva, J., James, C.M.,
Hellyer, A., Smith, C.G., Slabas, A.R. and Hughes, S.G. (1988)
Plastid-localised seed acyl carrier protein of Brassica napus is
encoded by a distinct, nuclear multigene family. Eur. J. Biochem.
174; 287-295.

Scherer, D.E. and Knauf, V.C. (1987) Isolation of a cDNA clone for the
acyl carrier protein-I of spinach. Plant Mol. Biol. 9; 127-134.

Schmid, K.M. and Ohlrogge, J.B. (1990) A root acyl carrier protein-II
from spinach is also expressed in leaves and seeds. Plant Mol. Biol.
15; 765-778.

Simpson, J., Van Montagu, M. and Herrera-Estrella, M. (1986)
Photosynthesis-associated gene families: Differences in response to
tissue-specific and environmental factors. Science 233; 34-233.

Slabas, A.R., Harding, J., Hellyer, A., Roberts, P. and Bambridge,
H.E. (1987) Induction, purification and characterization of acyl
carrier protein from developing seeds of oil seed rape (Brassica
napus). Biochem. Biophys. Acta 921; 50-

Snustad, D.P., Haas, N.A., Kopczak, S.D. and Silflow, C.D. (1992) The
small genome of Arabidopsis contains at least nine expressed B-tubulin
genes. The Plant Cell 4; 549-556.

103

CHAPTER POUR

SUMMARY AND PERSPECTIVES

Summary

A characterization of several Arabidopsis ACP isoforms at both the DNA and
protein level presented in this dissertation can be summarized as follows:

1).A genomic ACP clone encoding the ACP-1 isofonm was isolated
using B. napus cDNA ACP clone as a source of a heterologous probe.

2) The structure of the Arabidopsis ACP gene was determined by
comparing the cDNA ACP sequences from B. napus and B. campestris with the
Arabidopsis ACPI gene. The Arabidopsis ACPI gene encodes a protein that
has a 54 amino acid long transit peptide and an 83 amino acid long mature
protein. The gene contains one intron in the transit peptide coding
region and two introns in the coding region for the mature ACP.

3) A cDNA ACP clone that encodes the Arabidopsis ACP-2 isoform was
isolated from an expression library using anti-spinach ACP-I antibodies.

4) The ACPZ cDNA clone encodes a 134 amino acid long ACP that has
a 52 amino acid long transit peptide.

5) The ACP gene family in Arabidopsis has a minimum of three ACP
genes.

6) RNA blot analysis indicated that Arabidopsis ACP-l and ACP-2

might be constitutively expressed. In addition, a minimum of two ACP

104
isoforms are expressed in leaves, roots and seeds.

7) It was shown that native PAGE and urea PAGE should be preferred
systems for the analysis of Arabidopsis ACP isoforms because of their
better resolving power than SDS PAGE. Even ACP isoforms that differ by a
single amino acid, like ACP-2 and.ACP-3, were resolved in either native or
urea PAGE systems.

8) Arabidopsis mature ACP isoforms encoded by ACP], ACPZ and ACP3
genes are expressed in a constitutive manner.

9) Based on the immunoblot and radiolabeling studies Arabidopsis
leaf, root and seed tissues each have expressed tissue-specific ACP

isoforms.

Perspectives

Plant fatty acid synthase provides fatty acids for the cell membrane
synthesis, for the synthesis of cuticular waxes in epidermis and for the
synthesis of storage lipids in developing seeds (6,10) . While maintenance
of cell membranes requires constitutive activity of the components of
fatty acid synthase, biosynthesis of waxes and storage lipids are tissue-

specific and/or developmentally regulated events.

To date, multiple isoforms or isozymes have been identified for some of
the components of fatty acid synthase. One and two isozymes of the
malonyl-CoAzACP transacylase, that were expressed in a tissue-specific

manner, have been found in soybean seed and leaf tissues, respectively

 

105
(3). Very recently, three dimeric forms of the B-ketoacyl ACP synthase
I, composed of two different polypeptides, have been detected in barley
leaves (9). Multiple isoforms of a protein cofactor, ACP, that are
expressed in a tissue-specific'manner in higher plants, have been reported
even in simple multicellular algae (1). It seems likely that
isoforms/isozymes of some or of all components of plant fatty acid
synthase have different activities in reactions of fatty acid and some
steps of lipid biosynthesis, although direct in vivo evidence for this

possibility is lacking.

The work presented in this dissertation initiated characterization of an
ACP gene family in Arabidopsis. ACP nucleotide sequences for ACPI and
ACPZ genes will allow‘thorough characterization of theiexpression.patterns
of ACP isoforms in Arabidopsis at the cellular level, using either in situ
hybridizations or the analysis of the expression of a reporter gene driven
by either one of the promoters. The sequences also provide information on
the suitable approach that needs to be undertaken for the identification

of other members of ACP gene family.

Very little is known about the regulation of the expression of fatty acid
synthase in plants. Sequence information for the promoter regions of the
two characterized.ACP’genes in.Arabidopsis‘will permit initial analysis of
the regulatory elements, that can be extended as additional ACP genes are
identified. In order to perform these experiments one should look for the
expression patterns of chimeric genes comprised of different deletions of

the promoter regions fused to the coding region of reporter genes (such as

106

B - glucuronidase).

Recently, several groups have reported the involvement of ACP in reactions
other than fatty acid and lipid biosynthesis in prokaryotes, such as
activation of prohaemolysin (4) and synthesis of membrane derived
oligosacharides (11). ACP has been characterized as a subunit of an
NADH:ubiquinone oxidoreductase in both N. crassa (8) and in bovine heart
mitochondria (7). However, its function in the respiratory complex is not
known, Different ACP isoforms have been differentially efficient with the
enzymes at the branching point in the initial steps of glycerolipid
biosynthesis in vitro (2). Therefore, in order to address the question of
the functional importance of ACP isofomms in plant metabolism, with a
special emphasis on fatty acid and lipid biosynthesis, the expression of
particular ACP isoforms should be inhibited by expressing antisense RNA
ACP constructs. Identification of the ACP isoforms as provided in this
thesis, encoded by the ACPI, ACP2 and ACP3 genes will enable easy
detection of the inhibited isoforms by antibodies. Arabidopsis plants
with an almost complete inhibition of one or more of the ACP isoforms can
be analysed in detail. In addition, the plants carrying two or three
different ACP antisense clones could be obtained by crossing the
transformants that have a single ACP antisense clones. This way, it will
be possible to analyse the effects of the inhibition of either all

constitutive and/or tissue-specific ACP isoforms.

Availability of the expression clones with coding regions for the three

mature Arabidopsis ACP isoforms provides an opportunity to analyze the

107
secondary structure of an ACP. ‘The site-directed.mutagenesis of R-spinach
ACP-I isoform confirmed that Ser 38 is a phosphopantetheine binding site
(5). Similarly, site-directed mutagenesis of other regions of ACP
isoforms in combination.with the deletional analysis of central, amino and
carboxyl terminal regions might provide an information on the functional
meortance of different areas of ACP in the interaction with components of

fatty acid synthase.

Arabidopsis, like Neurospora, seems to have a mitochondrial ACP isoform
(Dave Shintani person. commun.). Availability of the clones for the
choloroplast-targeted and mitochondrial-targeted ACP isoforms will provide
tools for the analysis of the transport mechanism of the isoforms into
different organelles. In addition, it will be possible to tackle the
question of the cellular location of the attachment of the prosthetic

group to ACP.

 

10.

108

Bibliography

Battey, J.F. and Ohlrogge, J.B. (1990) Evolutionary and tissue-
specific control of expression of multiple acyl-carrier protein
isoforms in plants and bacteria. Planta 180; 352-360.

Guerra, D.J., Ohlrogge, J.B. and Frentzen, M. (1986) Activity of acyl
carrier protein isoforms in reactions of plant fatty acid metabolism.
Plant Physiol. 82; 448-453.

Guerra, D.J. and Ohlrogge, J.B. (1986) Partial purification and
characterization of two forms of malonyl-coenzyme A:acyl carrier
protein transacylase fromisoybean leaf tissue. Arch. Biochem. Biophys.
246; 274-285.

Issartel, J.-P., Koronakis, V. and Hughes, C. (1991) Activation of
Escherichia coli prohaemolysin to the mature toxin by acyl carrier
protein-dependent fatty acylation. Nature 351; 759-761.

Jaworski, J.G., Post-Beittenmiller, M.A. and Ohlrogge, J.B. (1989)
Site-directed mutagenesis of the spinach acyl carrier protein-I
prosthetic group attachment site. Eur. J. Biochem. 184; 603-609.

Ohlrogge, J.B., Browse, J. and Somerville, C.R. (1991) The genetics of
plant lipids. Biochim. Biophys. Acta 1082; 1-26.

Runswick, M.J., Fearnley, I.M., Skehel, J.M. and Walker, J.E. (1991)
Presence of an acyl carrier protein in NADH:ubiquinone oxidoreductase
from bovine heart mitochondria. FEBS Lett. 286; 121-124.

Sackmann, U., Zensen, R., Rohlen, D., Jahnke, U. and Weiss, H. (1991)
The acyl-carrier protein in Neurospora crassa mitochondria is a
subunit of NADH:ubiquinone reductase (complex I). Eur. J. Biochem.
136; 1-7.

Siggaard-Andersen, M., Kauppinen, S. and Wettstein-Knowles, P. von
(1991) Primary structure of a cerulenin binding B-ketoacyl-[acyl
carrier protein] synthase from barley chloroplasts. Proc. Natl. Acad.
Sci. USA 88; 4114-4118.

Slabas, A.R. and Fawcett, T. (1992) The biochemistry and molecular
biology of plant lipid biosynthesis. Plant Mol. Biol. 19; 169-191.

 

109

11. Therisod, H., Weissborn, A.C. and Kennedy, E.P. (1986) An essential
function for acyl carrier protein in the biosynthesis of membrane-
derived oligosaccharides of Escherichia coli. Proc. Natl. Acad. Sci.
USA 83; 7236-7240.

APPENDIX

APPENDIX

ACP nomenclature

In spinach, the leaf-specific ACP isoform is designated as ACP-I and the
constitutive ACP isoform as ACP-II (1). Barley ACP-I and ACP-II isoforms
appear to be expressed in leaves (2) . However, no information is
available on the expression of these two isoforms in other tissues.
Because of the apparent similarities in the nomenclature with acid
(phosphatases, characterized.barley ACP genes were named AclI and Acl3 (3).
The analysis of ACP isoforms in Arabidopsis revealed the probable presence

of more than 6 ACP isoforms.

Therefore, we suggest that the following nomenclature be applied to
specify genes and isoforms for acyl carrier proteins in Arabidopsis in
order to: (i) avoid confusion because the complete information on the
expression patterns is not available yet for Arabidopsis ACP genes and
(ii) to establish an easy system for their naming since the number of ACP
isoforoms in Arabidopsis is high. Following these guidelines, each ACP
isoform should be labeled with "ACPn", where n is an Arabic number. The
number should define the ACP isoform according to time of publication of
either nucleic acid or amino acid sequence. (By numbering different
isoforms using Arabic numerals instead of Roman numerals we will avoid

correlation with spinach and barley isoforms in the terms of their

110

111
expressiom.patterns and‘will provide a simpler system for the nwmerization
of large number of isoforms). Wild type genes should be labeled in upper
case italics, mutant genes in lower case italics and protein isoforms in
capital letters. According to these guidelines, the first published ACP
gene from Arabidopsis (4) would be ACPI, its mutant would be acpI and the

isoform encoded by that gene ACP-1.

112

Bibliography

Hansen, L. and von Wettstein-Knowles, P. (1991) The barley genes Ac11
and Ac13 encoding acyl carrier proteins I and III are located on
different chromosomes. Mol. Gen. Genet. 229; 476-478.

Hoj, P.B. and Svendsen, I. (1983) Barley acyl carrier protein: its
amino acid sequence and assay using purified malonyl-CoA:ACP
transacylase. Carlsberg Res. Commun. 48; 285-305.

Ohlrogge, J.B. (1987) Biochemistry of plant acyl carrier proteins. In
The Biochemistry of Plants. A Comprehensive Treatise. Vol.9.
P.K.Stumpf, ed., Academic Press, Hartcourt Brace Jovanovich, Orlando,
pp.137- 158.

Post-Beittenmiller, M.A., Hlousek-Radojcic, A. and Ohlrogge, J.B.
(1989) DNA sequence of a genomic clone encoding an Arabidopsis acyl
carrier protein (ACP). Nucleic Acids Res. 17; 1777.

 

iiiiiiiiiiiiiiii