usu IcAnAfflflnutlvo man-l 0W '"'”"‘”°"
W‘

_,_—-—

 

 

A RAB-RELATED GTP-BINDING PROTEIN IN SCHISTOSOMA MANSON!

By

Ingrid Katharina Loeffler

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Pharmacology and Toxicology

1 996

ABSTRACT
A RAB-RELATED GTP-BINDING PROTEIN IN SCHISTOSOMA MANSON!
By

Ingrid Katharina Loeffier

Schistosomiasis is a parasitic disease initiated by the deposition of eggs
in host tissues by the blood fluke Schistosoma. A gene encoding a low-
molecular-weight GTP-binding protein (LMWGP) was cloned from Schistosoma
mansoni using a polymerase chain reaction (PCR)-based strategy. The gene
was termed smrab (Schistosoma mansoni Lab-related protein). Northern blot
analysis hybridizes smrab cDNA with a 1.5-kb band of mRNA; this mRNA is
abundantly expressed in male schistosomes, while only slightly in females. The
deduced amino acid sequence of smrab shares ca. 70% homology with that of
several rab-related LMWGPs. Smrab terminates in a CCXXX motif, which is one
of several signals for post-translational isoprenoid modification by geranylgeranyl
protein transferase (GGPT) type II. A GGPT assay with in-vitro translation
product confirms that smrab is geranylgeranylated. Recombinant expression of
smrab in the pET3a expression vector yields insoluble protein which migrates as
a 23-kDa band on SDS-PAGE. N-tenninal sequence information of the

recombinant protein matches the predicted amino acid sequence of smrab.

GTP-binding analysis indicates that the recombinant protein binds GTP.
Therefore, smrab meets the criteria recently established for acceptance into the
ras superfamily of GTP-binding proteins [61].

Immunohistochemical analysis with a polyclonal antibody to smrab (or-
pep) localizes the protein to subtegumental structures. The distribution of the
protein recognized by or-pep changes when schistosomes are pre-incubated with
mevinolin, a drug which inhibits prenylation. Implications of these
immunohistochemical findings towards the elucidation of the function of smrab

are discussed.

AKNOWLEDGEMENTS

I would like to aknowledge the support of my advisor, Dr. James Bennett.
His efforts toward finding people who could help me learn molecular biology and
his patience with my slow and frustrating progress through this project is greatly
appreciated. I am also grateful for the opportunity which he provided for me to
attend the Biology of Parasitism course at the Marine Biological Laboratory at
Woods Hole, MA.

I also thank the members of my guidance committee, Drs. Peggy
Contreras, Jim Galligan, and Donna Koslowsky, for their advice and
encouragement, and for their readiness to discuss my work when- and however
often I asked for their input.

Finally, I extend my futile and sincere thanks to the many, many mice who
had to suffer schistosomiasis and then died for our efforts to better understand

the parasite which causes this disease.

TABLE OF CONTENTS

List of Tables ............................................. viii
List of Figures ............................................. ix
List of Abbreviations ........................................ x
Introduction ............................................... 1
A. Schistosome biology .................................. 1

1. Classification ................................... 1

2. Life cycle ...................................... 1

3. Vitellogenesis ................................... 3

4. The tegument ................................... 4
B. Pathogenesis of schistosomiasis ........................ 10

C. Background experiments .............................. 11

1. The mevalonate pathway and mevinolin .............. 11

2. Prenylation of two groups of
schistosome proteins .............................. 15

3. GTP binding by a 25-kDa group of

schistosomal proteins ............................. 16
D. The mevalonate pathway and protein prenylation ........... 16
E. The ras superfamily of GTP-binding proteins ............... 20

Wired.)

1. Ras ........................................... 25

2. Rho ........................................... 27

3. Art ............................................ 29

4. Ran ........................................... 32

5. Rab ........................................... 34

F. Hypothesis ......................................... 36
Materials and Methods ...................................... 37
A. Schistosomes ....................................... 37
B. DNA and RNA isolation from S. mansoni .................. 37
C. Cloning a rab-related gene from S. mansoni ............... 39
D. Northern blot analysis ................................. 42
E. Expression of the recombinant smrab protein ............... 43

F. Decreasing inclusion body formation and
solubilization and refolding of recombinant

smrab .............................................. 45
G. N-terminal amino acid sequencing ....................... 46
H. Schistosome homogenate ............................. 47
l. Mevinolin treatment ................................... 47
J. GTP binding analysis .................................. 48
K. In-vitro transcription, translation and

geranylgeranylation ................................... 49
L. Polyclonal antibody preparation ......................... 50

vi

W91)

M. Western blot analysis ................................. 51
N. Immunohistochemistry ................................ 52
Results ................................................ 54
A. Sequence analysis of smrab ............................ 54
B. Northern blot analysis ................................. 61
C. Expression of recombinant smrab protein ................. 64
D. GTP binding analysis ................................. 73
E. Post-translational processing of smrab by GGPT ............ 74
F. Immunohistochemistry ................................ 75
Discussion ............................................... 86
Bibliography ..................... ' ......................... 105

vii

LIST OF TABLES

1. Consensus sequence motifs in the ras
superfamily of GTP-binding proteins ........................ 23

2. Nucleotide and deduced amino acid sequence
of smrab ............................................. 56

3. Amino acid sequence alignment of smrab
and homologous rab proteins ............................. 58

4. Amino acid sequence alignment of the L7
"specificity domain" ..................................... 6O

viii

LIST OF FIGURES

1. Life cycle of Schistosoma mansoni .......................... 2

2. The schistosome tegument ................................ 6

3. The mevalonate pathway ................................. 12
4. The GTPase cycle ....................................... 21
5. The structure of H-ras .................................... 24
6. Northern blot analysis of smrab ............................. 62
7. SDS-PAGE of recombinant smrab .......................... 66
8. GTP binding blot ........................................ 69
9. Geranylgeranylation of smrab .............................. 71
10. Western blot with or—pep antibodies ......................... 76
11-15. lmmunolocalization of smrab with or-pep antibodies .......... 80
11. Fresh male schistosomes ................................. 81
12. Mevinolin-treated male schistosomes ........................ 82
13. Fresh female schistosomes ............................... 83
14. Mevinolin-treated female schistosomes ...................... 84
15. Effect of in-vitro incubation on or-pep labeling ................. 85

LIST OF ABBREVIATIONS
(in alphabetical order)

a-pep: affinity-purified anti-peptide IgG

BME: beta-mercaptoethanol

5'RACE: 5' Rapid Amplification of cDNA Ends

bp: base pair

CAPS: 3-(cyclohexylamino)-1-propanesulfonic acid
DTT: dithiothreitol

EDTA: (ethylenedinitriIo)tetraacetic acid

EGTA: (ethylenebis[oxyethylenenitrilo])-tetraacetic acid
ER: endoplasmic reticulum

F PP: famesyl pyrophosphate

FPT: famesyl protein transferase

GAP: GTPase activating protein

GDP: guanine nucleotide diphosphate

GGPP: geranylgeranyl pyrophosphate

GGPT: geranylgeranyl protein transferase

GST: glutathione S-transferase

GTP: guanine nucleotide triphosphate

HEPES: N-2-Hydroxyethyl piperazine-N'—2-ethane sulfonic acid
HMG-CoA reductase: hydroxy-methyl-glutaryl coenzyme A reductase
HRP: horse radish peroxidase

IPTG: isopropyl-B-D-thiogalactopyranosid

kb: kilobase

LMWGP: low-molecular-weight GTP-binding protein
mlv: multilaminate vesicle

NADPH: nicotinamide adenine dinucleotide phosphate (reduced)
ORF: open reading frame

PBS: phosphate-buffered saline

PCR: Polymerase Chain Reaction

PMSF: phenyl methyl sulfonyl fluoride

SDS: sodium dodectyl sulfate

SDS-PAGE: SDS-polyacrylamide gel electrophoresis
smrab: Schistosoma mansoni [ab-related protein

UV: ultraviolet

INTRODUCTION

Schistosomiasis is a parasitic disease of humans and many domestic and wild
animals. The World Health Organization estimates that 500-600 million people
are at risk for the disease annually, with a morbidity of 200 million and a 10%
mortality [131]. The high morbidity of the disease in humans and their livestock,
its debilitating chronicity, and its especially marked prevalence in areas with
agricultural and water development projects makes schistosomiasis a disease of

great public health and economic importance.

9 S I . | I . |

E | El '[i I'

Schistosomes are blood flukes of the order Digenea which are obligate
parasites belonging to the class Trematoda. There are at least 19 species of
schistosome, of which five cause most of the human disease. Schistosoma
mansoni is responsible for intestinal schistosomiasis in Africa, the Middle East
and Latin America. (This is the species studied in our laboratory, and

"schistosome" will subsequently refer to S. mansoni in this manuscript.)

AJLLiIeJMzLe

adult worm pair
in definitive host

portal/nesenteric
/ venules \
via produce
lungs & 300-3000 eggs

liver / per day

50%
I stay in host 1

cercariae liver 50%
leave snail & penetrate b excreted via
host in water intestines

’I

asexual fresh water
reproduction where niracidiun
hatches

\ W... /0

‘infects snail

Eigum Life cycle of Schistosoma mansoni.

3

The basic life cycle of S. mansoni is diagrammed in Figure 1. Eggs hatch
miracidia in fresh water: miracidia infect snails, undergo asexual reproduction,
develop into cercariae. Cercariae leave the snails and penetrate the skin of the
person in the water, then migrate to the portal and mesenteric venules via lungs
and liver. Adult worms have an average life span of 3-5 years, although they
may live as long as 35 years. The male and female parasite are separate but
remain in copula throughout their life span. The female is held in the male's
gynecophoric canal and oviposits ca. 300 fertilized eggs per day into the
mesenteric venules. Approximately half of these eggs pass out of the host via
feces and continue the life cycle. The remaining half becomes trapped in the
liver and initiates the process of pathogenesis which characterizes
schistosomiasis (see 8., below).

Am

The physiological process of egg production by the female schistosome
has been described [127]. Briefly, an ovum is released from the ovary under the
control of a sphincter muscle. The ovum passes down the oviduct into an area
known as the ootype in which the eggs are formed. Several mature vitelline
cells, produced in the vitelline glands, are secreted through a vitelline canal into
the ootype where they join the ovum. Ovum and vitelline cells are subjected to
secretions from the Mehlis' glands which feed into the ootype. The nature or
purpose of these secretions has not been well-defined, though there is some

evidence that they have mucus and lipoprotein components [127]. Mucus may

4

be important in lubrication to assist the egg's passage through the uterus. The
lipoprotein is believed to form a membrane around the vitelline cells and to serve
as a template for eggshell deposition [127]. In the ootype the vitelline cells
secrete globules of eggshell precurser which enclose the ovum and vitelline cells
in the egg. The material remaining in the vitelline cells serves as yolk within the
egg. The completed egg then passes through the uterus and from it into the
environment.

Although very little is understood about the biochemistry of the
mechanisms involved in egg production, the processes obviously require highly-
efficient and productive protein synthetic mechanisms.

WHERE!

The schistosome tegument covers the external surface of the adult worm
(Figure 2). It is composed of an acellular syncytium bounded by lipid
membranes and embeds phospholipid- and protein-containing inclusion bodies.
As in most flatworms, the tegument of schistosomes is a very active system,
critical in nutrient uptake and defense against host immune surveillance. The
tegument ultrastructure has been well described with electron micrographic
studies performed in the 1960's and '70's [57,95,124,126,142]. Although there
are regional differences along the length of the worm, the tegument is similar in
male and female worms. The male curls around the body of the female, who is
situated in his gynecophoric canal, and exposes his dorsal surface to the host's

bloodstream and endothelium. The surface plasmalemma consists of two lipid

5

bilayers which project invaginations into the syncytium of the tegument beneath
it. Within the ground substance composing the syncytium are two types of
inclusion bodies, both of which are generated by the tegumental cytons beneath
the muscle layers. These bodies are the more common discoid granules and the

larger multilaminate vesicles (mlv). The thickness of the syncytial layer varies
from 0.9-3.0 pm. The bounding membrane of the inner surface of the tegument

is a single membrane (as opposed to the double apical layer), and invaginates
upward into the syncytium from the basilar lamina. Beneath the basilar lamina a
thin layer of circular muscle and a thicker layer of longitudinal muscle constitute
the musculature of the schistosome and vary in thickness over the organism's
body (thicker dorsally). Medial to the muscle layers, with cytoplasmic
connections extending laterally into the tegument, are the tegumental cell
bodies. These cells are nucleated and synthesize the materials necessary to
maintain and disguise the tegument (Figure 28 & C). The turnover of the
tegument surface of the schistosome is integral to the mechanism by which the
parasite evades the host's immune system. The schistosome tegument is
therefore a highly-dynamic system, constantly sloughing and replenishing. The
discoid granules maintain the integrity of the tegument ground substance
glycocalyx, while the largely phospholipid contents of the mlv's replace inner
surface membrane lamellae as the outer layers slough away [142]. Vlfilson &
Barnes [144] measured the half-life of the membranocalyx to be only 2-3 hours.

Later studies by others indicated that the rate at which membranocalyx turns

figure], The schistosome tegument.

A: Transmission electron micrograph (13,600x) of the surface of a male
Schistosoma mansoni, showing the outer surface (OS), surface channels (SC),
multilaminate vesicles (MLV), discoid granules (DG), spines (8), circular muscle
(CM), longitudinal muscle (LM), and dorso-ventral muscle (DVM).

B: Transmission electron micrograph of the tegument and subtegument of a
schistosome demonstrating a tegumental cyton and its cytoplasmic connection to
the surface of the worm (stained black). Horse radish peroxidase was injected
into the tegument and penetrated the cyton via the cytoplasmic connection

(courtesy of R. Fax and R. Fetterer). Bar = 2 pm.

C: Schematic diagram of the schistosome tegument and related structures. OM:
outer membrane, a double lipid bilayer; SYN: syncytium; lM: inner membrane,
single lipid bilayer; BL: basilar lamina; CM: circular muscle; LM: longitudinal
muscle; TC: tegumental cyton; CC: cytoplasmic connection.

 

Ekrgy_e_2. The schistosome tegument.

a
Figure 2. The schistosome tegument

 

9

over varies with the external environment of the parasite [109]. In these
experiments, schistosomes incubated in vitro remained viable and retained their
membrane for as long as two weeks, while in vivo 50% of the parasites had
completely replaced their outer membrane within three hours. Furthermore,
membrane turnover appeared to occur at different rates along the worrn's
surface, being lost first at the sucker end. Work by Kusel and Mackenzie [72]
evaluating the release of protein antigens into culture medium suggested that
two populations of proteins are released from the surface membrane. Using a
double isotope labeling technique, investigators discerned a group of proteins
which appear to aggregate in the surface membrane, which turn over at a
homogeneous rate, and which were thus presumed to function in a maintenance
capacity. These proteins are sloughed relatively slowly into culture medium at a
rate consistent with membrane turnover. A second group of proteins was found
to turn over much more quickly and was proposed to be actively secreted into
the parasite's environment.

Protein supplying the considerable requirement of the tegument's flux is
synthesized only in the tegumental cell bodies situated beneath the muscle
layers [142]. The cytoplasm of these cells contain large numbers of ribosomes,
abundant Golgi apparatus, and ribonucleic acid, none of which are found
elsewhere in the tegument structure [142], and all of which are important
components of synthetic and secretory processes. Wilson & Barnes [144]

describe two types of Golgi apparatus: one apparently responsible for the

1O

generation of discoid granules, the other for mlvs. While ribosomes synthesize
the protein components of the vesicles, the Golgi process these proteins,
elaborate the vesicular membranes, and generate the polysaccharide
components of the cells' secretions. Additionally, it has been suggested that
Golgi play a role in recycling tegument membrane [144]. The thickness of the
tegument membrane remains constant over time, indicating that some
process(es) must occur to remove the outer layers as new ones are replenished
on the inner surface. Membrane may simply be sloughed; alternatively it may
also be actively recycled by the Golgi apparatus or internalized for autolysis.

Cellular mechanisms driving and regulating protein synthesis and
secretion and the activities of the Golgi apparatus are gradually becoming better-
understood. The rab group of low-molecular-weight GTP-binding proteins is very
important in these processes (discussed below). Interference with the function of
rab and other members of the ras superfamily of GTP-binding proteins has been
shown to disrupt systems dependent on them [5,53,120]. Applied to the
regulation of schistosome fecundity and viability, this concept is the premise for
the research described here.
E E || . I l . | . .

As mentioned above, about half of the eggs produced by the female
schistosome make their way from the portal and mesenteric vasculature where
they are released to the liver. Here they are trapped and elicit an immune

response from the host which results in the formation of a granuluma around

11

each egg. Fibrous scar tissue eventually replaces the cellular granulomas.
Enough such scars obstruct portal flow and interfere with hepatic function,
resulting in portal hypertension, hepato- and splenomegaly, ascites, and hepatic
insufficiency, which characterize intestinal schistosomiasis.
WW

Since the hosts' immune response to the eggs of the schistosome is the
immediate cause of the disease, the aim in controlling schistosomiasis is to
minimize the number of eggs to which the host is exposed. This can be
achieved with three strategies: 1) preventing infection of the host by the parasite;
2) reducing the number of worms that reach egg-producing maturity; and 3)
reducing the number of eggs produced by each worm pair. In the context of the
latter strategy, work in our laboratory has focused on elucidating the
mechanisms which control egg production in the schistosome, and attention has
come to the role of the mevalonate pathway.

Q I II I I II I . I.

The mevalonate pathway is responsible for the synthesis or lipid
modification of cholesterol, ubiquinones, dolichols, isopentyl adenine and the
isoprenoids famesyl pyrophosphate and geranylgeranyl pyrophosphate (Figure
3). Hydroxymethylglutaryl-CoA (HMG-CoA) reductase catalyzes the conversion
of HMG-CoA to mevalonate, and is the rate-limiting enzyme in the the pathway.
Work in our laboratory with S. mansoni has shown that mevinolin, a competitive

inhibitor of HMG-CoA reductase, reduces egg production by the female parasite

12

figured. The mevalonate pathway. Three molecules of acetyI-CoA are
converted into 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA), then to mevalonate.
The latter reaction is the rate-limiting step in the mevalonate pathway. Mevinolin
and compactin competitively inhibit HMG-CoA reductase and can thereby block
synthesis of all end products of the pathway. Famesyl pyrophosphate (F PP) and
geranylgeranyl pyrophosphate (GGPP) are 15- and 20-carbon prenyl chains,
respectively. In addition to famesylating proteins, FPP is further processed for
the production of squalene, a precurser to cholesterol, the retinal isoprenoic
acids, and the prenyl chain of heme A. Two isomers of GGPP are formed (all-
trans and cis-trans); only the all-trans isomer is believed to attach to proteins
[125]. GGPP gives rise to dolichols, long-chain isoprenoic acids, and the side
chain of ubiquinones. (From reference 47).

13

 

 

Ketone Bodes
/ 3m 2W
“0' . Mo. H
‘ m m ..
"ozc OOSCoA -‘:- 'w' 02¢ (31120H
HMO-00A Mevalonate
. (‘aATP
AA - ‘ :W CADWP”
OPP T
Dlrnethylalyl-PP Isopemenyr-PP
/ 7 Tm
/ / PP Cholesterol
tRNA t
Geranyl-PP --.... ' i
Pinyin-dune .
”cowl-PP /Squalene
lsoprenolc
/ / / PP /

Famesyl-PP \ Home a
MTW \

/ / / / OPP

Geranylgeranyl-PP

//\\

mg]. W Ubiqulnones Proteins

14

in vitro and in vivo, without interfering with other aspects of the fluke's
physiology [135,136]. VandeWaa et al. [136] have shown that incubation of
schistosome worm pairs in mevalonate or in famesol stimulates egg production
and reverses the inhibitory effects of mevinolin on egg production. Schistosomes
do not synthesize cholesterol [87], and egg-laying assays with dolichols,
ubiquinones, isopentyl adenine and juvenile hormones demonstrated no effect
on parasite egg production nor on mevinolin‘s inhibition of egg production [136].
Furthermore, it was found that radiolabeled acetate or mevalonate can be
converted by schistosomes to non-sterol lipid; the process is blocked by
mevinolin [136]. On the the basis of these experimental results, VandeWaa et
al. [136] postulated that a mevalonate-derived non-sterol lipid plays a critical role
in schistosome egg production.

Vitelline cells of female schistosomes and tegumental and subtegumental
structures of both sexes were examined by electron microscopy following
treatment of worms with mevinolin [16; I.K. Loeffler and J.L. Bennett,
unpublished observations]. Compared to sections from control (untreated)
worms, the cells in these tissues are enlarged and swollen. They are also
crowded with large, homogeneously-stained vesicular structures. These
changes suggest that mevinolin treatment results in inhibition of vesicular
transport mechanisms in the schistosome. Mevalonate or a derivative thereof is
vital to the parasite, as worm burdens in infected mice can be reduced by as

much as 96% with oral administration of high doses of mevinolin [15].

15
CZE II. [I [I'! I.

The identity of the mevalonate-derived non-sterol lipid which appears to
mediate the effects of mevalonate and mevinolin on schistosome egg production
was further investigated. Following incubation of schistosomes in [3H]
mevalonate, soluble and membrane-bound fractions of the schistosome
homogenate were separated by denaturing polyacrylamide gel electrophoresis
(SDS-PAGE). Radiolabeled proteins migrated at ca. 25 and 46 kDa. These
proteins were electroluted from the gel, and the lipid group cleaved off and
identified as geranylgeranol and famesol (per protein group, respectively) by
high-perfonnance-liquid chromatography [16]. In subsequent experiments
schistosome homogenate was incubated with either [3H]famesyl pyrophosphate
(FPP) or [3H]geranylgeranyl pyrophosphate (GGPP), followed by electrophoresis
of homogenate proteins by SDS-PAGE. It was found that F PP labeled a 46-kDa
group of proteins, while GGPP labeled a 25-kDa group. These experiments
demonstrated that schistosomes have prenyltransferase enzymes, and that their
endogenous proteins undergo post-translational lipid modification.

BZA-58 (Genentech, San Francisco, CA) is a prenyltransferase inhibitor
with some apparent preference for F PT, but which also inhibits GGPT [61].
Schistosomes were incubated for 48 hours in buffer supplemented with [3Hjmev
alonate with or without 10‘5M BZA-5B. Worms incubated with BZA-SB showed a
67-90% decrease in egg production, and analysis of homogenate by SDS-PAGE

and HPLC revealed that prenylation of both the 25-kDa and 46-kDa group of

16

proteins was blocked (J. Bennett, unpublished observations). This experiment
confirmed that the observed link between the mevalonate pathway and egg
production is via a prenyltransferase, and suggests that the morphologic effects
of mevinolin observed on electron micrographs are also related to inhibition of a

prenyltransferase.

nun.- -_, nl-oroo '0 011-. 00‘1.

Schistosome homogenate which had been incubated with either [3H]FPP
or [3H]GGPP was separated by SDS-PAGE. Southwestern blot analysis of the
transfered proteins with [or-32P]GTP revealed strong GTP-binding by the 25-kDa,
geranylgeranylated group [16].

From the findings summarized here, it is reasonable to speculate that the
link between the mevalonate pathway and the regulation of egg production in the
schistosome may be via prenylated low-molecular-weight GTP-binding proteins

(LMWGPs).

D II I I II I I . I I.

Studies in the late 1970's and early 1980's indicated the importance of the
mevalonate pathway in the regulation of cell growth, proliferation and
morphology [5,46,49,53,59,78,98]. Inhibitors of HMG-CoA reductase were used
to establish that blocking the mevalonate pathway in mammalian cells arrests

cell cycling in vitro and suppresses tumor growth in vivo [78]. Cellular

17

morphologic changes such as rounding and neurite outgrowth were observed to
accompany the interference with cell cycling. Adding mevalonate back to the
culture media was found to abrogate these effects. However, just as VandeWaa
et al. demonstrated with schistosome egg production, a similar abrogation of the
effects of cell cycling and morphology could not be achieved with addition of
cholesterol, ubiquinones, dolichols or isopentyl adenine to cell media [8,79]. The
nature of the specific end-products of the mevalonate pathway mediating the
observed effects remained elusive until Schmidt et al. in 1984 traced
radiolabeled mevalonate to cellular proteins in the form of post-translational
isoprenoid modifications [116]. Since then, isoprenylated proteins have been
localized to various cellular compartments, and include nuclear lamins, mating
factors of yeast and fungi, insect juvenile hormones, the gamma subunit of
heterotrimeric GTP-binding proteins, and LMWGPs [47,48,49,78].

Isoprenoid modification of proteins occurs post-translationally in a three-
or four-step process. The enzymes famesyl protein transferase (FPT) and
geranylgeranyl protein transferase (GGPT) attach famesyl (15-carbon chain) and
geranylgeranyl (20-carbon chain), respectively, to the cysteine in a specific
carboxy-tenninal motif of the target protein via a thioether bond. F PT recognizes
proteins ending in the CAAX motif, where C Is a cysteine residue, A is an
aliphatic amino acid and X is the C-terrninal amino acid [53,80]. GGPT type I
also targets a CAAX motif, however with the important distinction that the X

residue is a Ieucine rather than a cysteine, alanine, serine or methionine as in

18
famesylated proteins [48]. GGPT type II modifies unique COOH-terminal motifs

which occur on the rab subfamily of LMWGPs [38,69,121]. These proteins
terminate in XXCC, XCXC, CCXX, or, more rarely, in CCXXX or CXXX motifs
[39]. Geranylgeranylation of two cysteines within the motif has been
demonstrated in at least one rab protein terminating in XCXC [38]. The COOH-
terminal tetrapeptide sequence is both necessary and sufficient for prenylation
by FPT and GGPT type I, with the terminal amino acid residue of the sequence
signalling prenyltransferase specificity. This has been demonstrated with
mutational analysis experiments in which the terminal X residue of the CAAX
motif was interchanged and which showed that the single mutation was sufficient
to confer enzyme specificity [68]. (T his requirement for specificity has recently
been brought into question by Desolms et al. [29] who describe a group of
peptide amines which lack the X residue of the CAAX motif but which still show a
100-fold selectivity for prenylation by F PT over GGPT type I.) Furthermore,
CAAX tetrapeptides serve as substrates and as competitive substrate inhibitors
for F PT and GGPT type I in in-vitro systems [50; Bennett, unpublished results].
Geranylgeranylation by GGPT type II, however, requires more than the COOH-
terminal motif. Studies with ras and rab chimeric proteins in which the terminal
CAAX and CC/CXC sequences were exchanged suggested that the enzyme
required additional sequence information and recognition of protein conformation
to signal the appropriate modification [65]. Further work demonstrated that a

conserved N-terrninal motif in the protein is necessary to signal prenylation by

19

GGPT type II, although it alone is also insufficient to signal the modification
[113]. Indicating even further complexity in translational modification by this
enzyme are the results of experiments with a specific LMWGP, rab5, indicating
that the protein must by in the GDP-bound conformation for interaction with the
prenyltransferase [1 12].

Following attachment of the lipid moiety to the protein, a membrane-
associated endoprotease removes the amino acids distal to the modified
cysteine [11]. The modified cysteine is then methylated by a specific protein
methyltransferase which methylates both famesylated and geranylgeranylated
proteins [11]. In some instances an upstream cysteine is also palmitoylated.
The completed lipid modification process renders the protein significantly more
hydrophobic at its COOH terminus than was the original translation product.

The function of protein prenylation remains less clear, although the
demonstration that the transforming ability of oncogenic ras proteins is blocked
by preventing ras prenylation first emphasized the importance of prenylation in
cellular function [141]. It has been well-established with several proteins that
preventing their prenylation changes their intracellular distribution and renders
them dysfunctional or altogether non-functional [66,70,78,125]. More
specifically, prenylation appears to be necessary for protein-lipid (membrane)
and protein-protein (effector) interaction [5,11,53,81,115]. Two models have
been proposed to describe the role of prenylation in protein function: the "dumb"

and "smart lipid" models [115]. In the former the prenyl group functions merely

20

to provide a lipophilic affinity between the protein and the membrane, without
specificity for the target membrane. Observations involving some ras-related
LMWGPs, for example, indicate that changing the nature of the lipid moiety
attached to the protein (FPP vs. GGPP) does not change the protein's
localization and function [115,125]. Alternatively, the "smart lipid" model
ascribes a targeting function to the prenyl group, such that the lipid modification
guides the protein to a specific location. Experiments with other prenylated
proteins indicate that interactions between the prenyl group and the membrane
lipid may be important in orienting the protein to its specific site on the

membrane, and hence for its function [22,115].

E II I 'I [GIE-l' I. I.
The ras superfamily of GTP-binding proteins is ubiquitous in eukaryotic
cells. Throughout the phylogenetic tree of organisms, these proteins appear to
be involved in the control of a diverse array of cellular functions including
differentiation, proliferation, intracellular vesicular trafficking, NADPH oxidation,
and cytoskeletal organization. The members of this large group of LMWGPs
share several definitive features: 1) they are 20-29 kDa in size; 2) they bind and
hydrolyze GTP; and 3) they maintain between and within their subfamilies

characteristic, conserved amino acid sequence motifs. They also share a similar

21

lnactlve

+ GAP

.0! NO

 

MThe GTPase cycle. The inactive LMWGP (shown square) releases
GDP and binds GTP with the help of guanine nucleotide release proteins
(GnRPs). Binding GTP changes the conformation of the protein and enables it
to interact with its effectors. Intrinsic GTPase, enhanced by GTPase activating
proteins (GAPs), then hydrolyzes GTP and returns the protein to its resting state.

22

three-dimensional structure and biochemical mechanism regulating the GTPase
switch which controls their activity.

Ras-related proteins function as binary switches by virtue of the GTPase
cycle (Figure 4). In the GDP-bound form the protein is inactive; upon exchange
of the nucleotide for GTP, the protein is activated for interaction with its effectors
until intrinsic GTPase activity hydrolyzes the GTP back to GDP, and the protein
returns to its resting state. Guanine nucleotide release proteins (GNRPs)
catalyze the release of GDP from the inactive protein, allowing its replacement
by GTP which, under ambient conditions, has a higher affinity for the protein
than does GDP. The low endogenous GTPase activity of these proteins is
catalyzed by GTPase-activating proteins (GAPs), which enhance the hydrolytic
function up to 105-fold. Careful regulation of this cycle is critical for normal
function of GTP-binding proteins. Failure to hydrolyze GTP is the driving
mechanism for oncogenic ras proteins, for instance, and allows their activity to
go unchecked, resulting in unregulated cell growth and proliferation [5,7].

Ras-related proteins range in overall amino acid sequence homology from
less than 30% within the superfamily to over 85% within subfamilies. However,
they all share highly-conserved amino acid sequence motifs in regions involved
in nucleotide binding, GTP-induced conformational change, and GTP hydrolysis
(see Table 1) [7]. Notably, these regions are all located in loops on one side of
the ras protein molecule [7 .30]. On the basis of these conserved domains,

overall sequence homologies, and similar biochemical mechanisms between

23

n>xzcoq> moomeoz>cm >m>m>amfl> >muexo boooo>q>qx can

ocxz<u>q oooo>nz>um >2mmHemHu HHueuo acoao>qum mum

axmzo>AH moocensqmq >2mm>em>> Havana ocooo>H>qx ozu

noxzo>qq moocenzch mo>mHemH> qummo >omcquaax can

coxzo>q> moocenHHoq xmcmHemcz qumuo >oomo>>>qx mmu

QxMZOOOO NOOdNnOOhO BIAJNVIQ ...mMO xwxxwoooox mdmcmmcoo
4-0 m-o «-0 H-o

.HDH .Hm um ocuzom EOHm voummpd .mvoo Uflom ocfiEm Hmuuoalmamcflm

one >9 Umumcmflmmv who mvflom OGHE¢ .mmdvflmmn Deaflnmonvxs "b “moSmeMH vanonmonv>£

no inflow ocflEm xcm ”x .HmmH mcofiuduflquSm Uflom ocflsm ©o>uomcoo "mnmuuoa wmmo

lemma: Hmnuo .mmsvflmMH vflom OCHEm HMUAUchH "mmhu naom .EoSU cmo3umn 0m mmoa oafl£3
.mcflmuoum UmumHmurmmu wo mmflHflEmMndm GHQDHB Um>ummcoo >H£mfl£ ma mocmsvmm mufl “mooH ma
mnu CH mofla mum .>ua>fluom 0flu>HOMU>£ mam How HMflucmmmm ma coH many amazomaoe nflmuoum
map QDH3 :m2 m0 :ofluomnoucfl How >Mmmmmomc ohm mm was NO mcoflmmm .Homtba mcflvcfln

mBO cfl Um>ao>cfl mum woodmmu HDOM Had .xaflemwuomdm own map mo mwflHflEmwQSm HSOm

can no some Scum mo>HDMucmmoHQMH m0 elm smdonnu H10 mconmH mo mucmecmflam mum :3onm
.mcflououm mnfivcfinumaw mo kHflEMMHmQSm mm“ on» Ca mmwuoe mocmsvmm mamcmmcou IAIuHQMH

24

 

 

 

 

 

 

Eigurei The structure of H-ras. Ribbons represent [3 sheets; coils represent a
helices. Loops and [3 sheets are numbered sequentially from the amino to the
carboxy termini. A: Schematic three-dimensional structure of human ras protein.
The guanine nucleotide binding pocket is indicated with the symbols
representing bound nucleotide: guanine base, ribose and phosphates are shown
as rectangular block, pentagonal block, and spheres, respectively.
B:Topological structure of the same molecule. (From reference 30.)

25

subfamilies, the three—dimensional structure of these proteins is proposed to be
very similar [5]. The crystal structures for both the GTP- and GDP-bound
conformations of human H-ras have been resolved (Figure 5), and sites for
molecular interactions identified [30,88]. Conformational differences between the
active and inactive states of the protein are localized to two regions: residues
30-38 of the L2 loop and residues 60-76 composing the L4 loop and its following
short alpha helix. The "on" and "off" states of the protein switch mechanism
appear to be induced by the gamma phosphate of GTP. Guanine nucleotide
binds in a pocket formed by loops L1, L2, L7 and L9. Loop L1 is located near
the phosphates of the bound nucleotide and is believed to be the location of the
GTPase catalytic site. Interestingly, most of the point mutations which activate
proto-oncogenic ras are located within loops L1, L4 and L7; i.e, in positions
where they may directly affect nucleotide binding and GTPase activity. De Vos
et al. [30] suggest that loop L4, although not in contact with the nucleotide, is in
direct contact with loop L1, and mutations in L4 may thereby affect GTP binding
or hydrolysis indirectly.

Based upon sequence homology and extrapolated structural similarity, the
ras-related LMWGPs are divided into five subfamilies: ras, rho, arf, ran and rab

[61].

E._1.._Ba§

Ras proteins are involved in an apparently universal and evolutionarily-

26

conserved cellular signalling pathway which integrates various signals received
from the cell surface and directs them to the cell nucleus for the regulation of
cellular differentiation and proliferation [5.36.53]. Recent work with a variety of
eukaryotic organisms has implicated ras as a critical molecular link between the
pathways activated by cell surface receptor tyrosine kinases and a serine/
threonine kinase cascade leading to the regulation of gene expression in the
nucleus [5,36,89]. The activating signals and the outcome of subsequent ras
activation are diverse [36]. Activators of the cell surface tyrosine kinase
receptors include a variety of cell modulators such as growth factors,
neurotrophic factors, and factors involved in the immune response [81]. In the
fission yeast Sacchanomyces cereviciae ras proteins control mitotic cell cycling in
response to changing nutritional status of the environment. In this same
organism, ras also regulates meiosis, sporulation and cell shape, as well as the
response to mating pheromone. Ras activation is required by Caenorhabditis
slogans for proper vulval development. Eye development of the fruitfly
Dmsophila melanogaster also utilizes this convergent pathway. Mammalian ras
genes are required for normal cell growth and differentiation in response to
mitogens [33] and have recently received considerable attention because of their
cell-transfomning abilities. These genes have been well-characterized as proto-
oncogenes, and, when over-activated, become oncogenic [2]. Notably, single
point mutations at one of several positions can activate proto-oncogenic ras

(codons 12, 13, 15, 16, 62, 64, 66, 125, and 128 of H-ras [30]). The protein's

27

hyperactivity is the result of a failure to hydrolyze GTP, which can be due to the
structure of ras itself (see above) or to a misfunction in the GAP responsible for
enhancing GTPase activity. Ras oncogenes play a critical role in human
carcinogenesis, most notably in over 50% of colon and 90% of pancreatic
cancers [45]. Ras proteins are famesylated and this post-translational
modification is necessary for both normal and oncogenic protein function
[54,141]. This requirement is currently under intense investigation as a
potentially exploitable target for controlling oncogenic ras and its tumorigenic

effects [45].

ELEM

Members of the rho group of proteins are approximately 30% homologous
to ras and, like ras, have a CAAX box at the carboxy-tenninus [53]. Also like ras,
mutation of the glycine corresponding to Gly12 in H-ras activates rho in a
manner similar to activation of proto-oncogenic ras [5,53]. Mutant recombinant
rho has been used to investigate the function of this group of proteins: they
appear to play an essential role in cytoskeletal organization. Early experiments
with Clostridium botulinum showed that rho is readily ADP-ribosylated and that
cells treated with C. botulinum toxin undergo morphologic changes consistent
with disassembly of actin microfilaments. Further investigation revealed that rho
proteins respond to extracellular signalling factors to regulate cellular responses

which ultimately involve the polymerization of actin microfilaments at the plasma

28

membrane [31]. These conclusions were supported by later work using mutant
rho proteins and cell microinjection and rescue techniques [5,53]. Elicited mast
cell secretion was recently shown to be regulated by rho and its close relative,
rac [104]. In keeping with rho's cytoskeletal association, the process appears to
depend on the centripetal redistribution of filamentous actin in response to cell
stimulation. Rho and rac have been demonstrated to work together to
coordinate this activity in mast cells [104].

Recently, investigators began to study the role of rho proteins (specifically
rhoB) in the signalling pathways activated by ultraviolet (UV) light, as increasing
evidence suggests overlap between these pathways and those activated by
growth factors [41]. It has been found that transcription and translation of rhoB
does increase rapidly on exposure of cells to UV light, suggesting a role for this
protein in genotoxic damage [41]. Rho proteins have also been implicated in the
regulation of calcium sensitivity in smooth muscle contraction [5].

In mammalian cells rho has been localized to both the cytosol and the
plasma membrane, as well as to compartments associated with the endocytic
recycling pathway. RHO1, the rho homolog in yeast, is found primarily in the
Golgi stacks [5].

Rac proteins constitute a subgroup of the rho subfamily of related
LMWGPs. These proteins are important not only in actin filament formation, and
hence in cytoskeletal maintenance, but also in the formation of reactive oxygen

species by phagocytic leukocytes [5,31]. Macrophages, neutrophils and

29

eosinophils kill bacterial invaders with toxic oxygen metabolites generated by
NADPH oxidase. This is a membrane-associated multimeric enzyme which
assembles on phagocytic activation. One of the subunits of the complex is rac
[5,31], which controls oxidase activation. Interestingly, the chemotactic activity of

stimulated leukocytes appears to be under control of rho [25].

E._3.._A[f

Arf was initially discovered as a protein cofactor necessary for ribosylation
of the heterotrimeric GTP-binding protein G, by cholera toxin (hence its acro
nymic designation: ADP-ribosylation factor) [62]. Several other arf proteins have
been identified since then, and they appear to be ubiquitous in eukaryotic cells
with a redundant pattern of gene expression [6,63]. The amino acid sequence of
individual arf proteins is highly conserved evolutionarily, suggesting that the
function of these proteins is basic and essential to cellular physiology [63,139].

At 20-21 kDa, arf proteins are the smaller of the ras superfamily of GTP-
binding proteins. Three biochemical properties distinguish arf from other
subfamilies of ras-related proteins [6]. First, binding of GTP, and hence activation
of the protein, depends on association with phospholipids. Second, arf binds
lipid membranes and associates with its effectors only in the GTP-bound form;
GTP hydrolysis to the inactive GDP-bound state releases the protein into the
soluble fraction. Third, investigators have been unable to measure any basal

GTPase activity in arf proteins. On the basis of these and other observations

30

involving the interaction of arf with specific phospholipid, investigators have
proposed a model in which arf both responds to and mediates the membrane
phospholipid environment in a GTP-dependent fashion [6,139].

Arf proteins are located on Golgi membranes and non-clathrin-coated
vesicles which bud from the surface of Golgi membranes. Nucleotide exchange
factors associated with the Golgi membrane enhance the replacement of GDP
with GTP on arf, which then enables the GTP-activated protein to bind the
membrane and to execute its function. As stated above, arf activation is
dependent not only on GTP binding, but also on its association with
phospholipids. The lipid modification enabling arf proteins to associate with
membranes is unique to LMWGPs and more akin to the modification found on
the alpha subunit of heterotrimeric GTP-binding proteins: instead of carboxy-
terrninal isoprenylation, arfs are myristoylated at the amino terminus. The
glycine in the second position of the amino acid sequence of arf proteins, directly
following the initiating methionine, is the necessary substrate for myristoyl CoA
protein N-myristoyltransferase (NMT). As with the carboxy-terminal motifs which
allow prenyltransferases to identify their targets for prenylation, the amino acids
surrounding the myristoylated glycine also influence recognition of arf proteins
by NMT [139].

The amino terminus of arf is integral to the protein's function. It is the site
for lipid modification, enabling the protein to bind its effector membrane.

Following a loop composed of the first ca. ten residues, it also twists the next six

31

residues into an alpha helix. This helix stabilizes the GTP-bound form of the
protein and appears to act as a GTP-dependent switch mechanism for effector
interaction [6,1 39].

Unlike other ras-related proteins, arf exhibits no measurable basal GTPase
activity. Arf proteins rely upon arf-specific GTPase activating proteins (arf-GAPs)
to terminate their activity and to allow their release from Golgi membranes to
complete the characteristic GTP-binding protein cycle [139]. In arf proteins an
aspartate residue occupies the position comparable to that of glycine 12 of ras,
which is essential for the GTPase activity of that protein. This difference is
unlikely to explain the lack of GTPase acitiviy in art, however, as amino acid
substitutions in this position are found in other LMWGPs as well (e.g., in rab),
which demonstrate GTPase activities comparable to those of ras [147]. Arf-GAP
activation, like activation of art itself, depends on phosphatidylinositol, and is
intimately associated with the activity of arf. Arf was recently discovered to
synergize with a cytosolic factor to stimulate phospholipase D [74], which
catalyzes the converstion of phosphatidylcholine to phosphatidic acid. The
product of phospholipase D activity (phosphatidic acid) and the enzyme's
substrate (phosphatidylcholine) respectively stimulate and inhibit arr-GAP
activity.

Arf proteins have been implicated in a number of seemingly disparate
roles, including the regulation of intracellular traffic in both endocytic and exocytic

pathways, vesicle formation and coat assembly, and activation of adenyl cyclase

32

(via 6,) [6.63.139]. All these activities, however, are associated with arf's
reversible and GTP-dependent association with intracellular membranes. Arf has
been identified as one of the several coat proteins associated with Golgi
vesicles, and is known to be required for the GTP-dependent assembly of the
coat protein complex (coatomer) [6,139]. Studies have also demonstrated the
essential role of arf in the transport of molecules from the endoplasmic reticulum
(ER) to the Golgi, as well as within the Golgi stack [119]. This role appears to be
associated with arfs involvement in vesicle fusion, as the importance of arf has
also been observed in the events of endosome fusion during endocytosis and

post-mitotic nuclear vesicle fusion [6,139].

35.33:!

Ran was originally identified in teratocarcinoma cells and named TC4 [34].
Since then its yeast, protozoan and mammalian counterparts have been
discovered [17,34,40,107,129] and have been assigned the acronym ran (Las-
related nuclear protein); hence the common presentation of this group 'of
proteins as ran/T C4.

The members of this subfamily are unique in the ras superfamily of binding
proteins in that 1) they are localized to the cell nucleus; and 2) they lack the
characteristic cysteine-dominated prenylation domain at the carboxy terminus
[3]. This sequence is replaced instead by a highly-acidic DEDDDL sequence

which is not prenylated. The C-terrninal DEDDDL sequence stabilizes the

33

binding of GDP to the ran protein and, like the post-translational isoprenoid
modification necessary for other ras-related proteins, is important in regulating
the interaction of GTP-bound ran with its many effector proteins [107].

Ran is a very abundant protein, two orders of magnitude more so than ras
(107 vs. 3x105 copies per HeLa cell) [92]. This high concentration appears to be
essential for normal cell function, as a genetically-engineered strain of S. pombe
which produced only half the amount of spi1 (its ran homologue) than the wild
type strain was markedly defective in cell cycle regulation and nuclear structure
[92]. Ran proteins are best-known for their essential role in nuclear import and
export [92,107]. Transportation of proteins and ribonucleoproteins across the
nuclear envelope is an active and highly-selective process. The nuclear pore
complex contains pieces of the required transport machinery, but soluble factors
are required as well. Moore and Blobel [91] described two cytosolic fractions
necessary for nuclear importation in Xenopus oocytes; fraction 8 was found to
contain two components, one of which was identified as the GTP-binding protein
ran. It was further found, in keeping with ran as a LMWGP, that the protein had
to be in the GTP-bound form and required GTP hydrolysis for successful nuclear
importation to occur [91,92]. Similar findings have supported these conclusions
in HeLa cells, as well [92]. More recently, it was found that export of proteins
from the nucleus also requires activated ran [94]. The role of ran proteins has
also been implicated in processing RNA and exporting it from the nucleus, and in

regulating DNA replication and cell cycling [92,107].

34

E21331:

Rab proteins constitute the largest subfamily of ras-related LMWGP's, with
over 30 identified to date. They share ca. 30% amino acid sequence homology
with ras and as much as 75% or better with each other. They have been
discovered in all eukaryotic organisms examined to date and their names vary
with the organism (e.g., YPT1 and SEC4 of Saccharomyces cereviciae; ora1 and
2 of the electric eel, Dyscopyge ommata; SAS1 and 2 of the slime mold,
Dictyostelium discoideum); those of mammalian origin are called rab, however,
and hence the general name for the subfamily.

Rab proteins have two structural characteristics which distinguish them
from other members of the ras superfamily [53]. First, they have a CCXXICXC
motif at the carboxy terminus for signalling prenylation instead of the
characteristic CAAX box (discussed above). Second, they do not have the
highly-conserved glycine at residue 12 of H-ras which, when mutated in ras,
activates the protein to an oncogenic capability. Instead, the comparable
position is usually held by a serine [110]. Rab proteins nonetheless have
GDP/GTP exchange rates and GTPase activities similar to ras [147].

Rab proteins function in the regulation of intracellular trafficking and
secretion. Early gene disruption experiments with YPT1 of S. cereviciae
identified the protein as essential for cell viability and showed that loss of YPT1
gene function resulted in abnormal cell budding and compromised cytoskeletal

integrity [118]. Further work demonstrated that mutation of the gene results in

35

accumulation of membranes and vesicles in the cell cytoplasm, abnormal ER
and Golgi structures and defects in secretion [122]. th1 and its mammalian
counterparts have been localized to the Golgi apparatus and to ER-Golgi carrier
vesicles [122]. On the basis of this localization, it was originally suggested, and
is now well-established, that the protein is necessary for early secretory events
[5,122]. Recent studies indicate that ypt1/rab proteins participate in regulating
the direction and fidelity of protein sorting in the ER-Golgi pathways [60].

SEC4 was the second rab-related protein identified and was isolated from
temperature-sensitive S. cereviciae secretion mutants which accumulated
secretory proteins and post-Golgi vesicles [43,110]. The gene was cloned [110],
and since then the role of the protein and its mammalian counterparts has been
described in Golgi secretory events [1 ,52,138].

With the exception of rab3, which is found only in neural tissue, rab
proteins are ubiquitous across eukaryotic organisms and cell types [24,53]. They
are localized to cellular compartments of the endo- and exocytic pathways, as
well as to those involved in the processing of proteins from the ER through the
Golgi stacks. The functions of these proteins have been studied using a variety
of methods, including cell-free transport assays and overexpressed wild
type/mutant recombinant proteins in intact cells, as well as antisense RNA and
inhibitory peptide techniques [5,26,76,102]. The large and rapidly-increasing
number of proteins in this subfamily of LMWGPs gives some suggestion of the

diversity of their function. Their roles can be categorized in three essential

36

cellular activities: 1) endocytic recycling of vesicles and proteins from the cell
surface (e.g., rabs 4, 5 [51,1341); 2) exocytosis and secretion (e.g., rabs 3, 8
[18,75]); and 3) protein transport through the ER-Golgi apparatus (e.g., rabs 1, 2,

6, 9 [77.84.101.1301).

Eetlxnszthesis

The observations in our laboratory suggest a link between mevalonate and
the regulation of egg production in Schistosoma mansoni via the isoprenoid end-
products of the mevalonate pathway. Characteristics of the schistosome
proteins identified by isoprenoid modification and GTP binding further suggest
that the mevalonate-mediated effect may occur at least in part in the context of
small GTP-binding proteins. As discussed, these proteins occur ubiquitously
throughout eukaryotic organisms, and their roles are essential to cell growth,
proliferation, and function. Given the female schisosome's fantastic rate of egg
production and the relentless turnover of the tegument's antigenic presentation in
(especially) the male, the schistosome would appear to exercise a tremendous
protein synthetic activity. It was considered very likely, then, that LMWGPs,
particularly the rab-related, play essential roles in the fecundity and viability of
this parasite. The work described here is an account of the cloning,
identification, and partial characterization of one such rab-related LMWGP in S.
mansoni; work undertaken as part of an effort to better understand the

physiology of this complex and elusive creature.

MATERIALS AND METHODS

Aishistosomes

The schistosomes used in this work were the Puerto Rican strain of
Schistosoma mansoni. White female laboratory mice (ICR; Harlan, Sprague-
Dawley, lnc., Indianapolis, IN) and Biomphalan‘a glabrata snails served as the
definitive and intermediate hosts, respectively, for the parasite. Snails were
infected with schistosome metacercariae; thirty days later, these snails were
induced to shed cercairae by exposure to complete darkness for 24 hours,
followed by re-exposure to light. Mice were infected at 15-20 grams body weight
by intraperitoneal injections of 250-300 cercariae collected from snails. Six
weeks after mouse infection, schistosomes reached sexual maturity and were
surgically harvested from the mesenteric veins and hepatic portal vein of the

mice.

E DIIE IBII!’ II. Ii S .

DNA was isolated according to modifications of the procedure described
by Strauss [128]. Briefly, 250 mg schistosomes (500 worm pairs) were frozen,
then powdered with mortar and pestel under liquid nitrogen. The powdered

worms were transfered to a Dounce homogenizer in homogenizing buffer (50mM

37

38
TrIS°HCl, pH 8.0; 50mM EDTA; 100mM NaCI) with 0.1% SDS and 3-4

mg Proteinase K. Following homogenization, the mixture was transfered to a
small flask and incubated 16-24 hrs. at 50°C. The schistosome digest was
extracted once with phenol/chlorofonnfisoamylalcohol and preciptated with 7.5M
ammonium acetate and ethanol for 1 hr. at -20°. Following centrifugation, the
DNA pellet was washed in 70% ethanol and resuspended in 5 mL TE buffer
(10mM Tris-HCI, pH 8.0; 1mM EDTA). The impure DNA was incubated at 37°C
for 1 hr with 2 mg RNAase A and 0.1% SDS; then for another 1-2 hrs. at 50°C
with the addition of 1 mg Proteinase K. Phenol/chloroform extraction and
ammonium acetate precipitation were repeated and the washed DNA pellet
dissolved in TE pH 8.0 at 4°C for several hours. Concentration and purity of the

isolated DNA were assessed spectrophotometrically at 260 and 280 nm (1
0.0.26‘) z 50 jug/mL DNA; Ame/A230 ratio approximately 1.8).

RNA isolation was performed using a guanidinium/chloroform extraction
method. Again, 500 schistosome worm pairs were snap-frozen in liquid nitrogen.
(For separate isolation of male and female RNA the worms were separated prior
to freezing.) The frozen pellets were ground under liquid nitrogen and transfered
to GT buffer (4.24M guanidinium thiocyanate; 0.1M Tris-HCI, pH 7.5; 0.5%
sarkosyl; 1M BME). The suspension was vortexed vigorously, then passed
through a 22-G needle. PhenoI/chIoroforrnlisoamylalcohol extraction and sodium
acetate precipitation were performed simultaneously, after which RNA was

precipitated from the aqueous supernatant overnight at -20°C with the addition

39

of an equal volume of isopropanol. The pellet was spun out and resuspended in
GT buffer. An equal volume isopropanol was added and precipitation was
repeated. Following a wash in 75% ethanol, the RNA pellet was dissovled in
RNAase-free water at 65°C. Aliquots of the isolated RNA are electrophoretically
separated on an agarose-forrnaledehyde gel, where a 2.0 kb band dominating a
smeared ladder pattern is characteristic of schistosome total RNA.

Concentration was assessed spectrophotometrically at 260 nm (1 O.D.260 z 40

pg/mL).

WWW

Degenerate oligonucleotide primers were designed to correspond to two
highly-conserved GTP-binding regions of LMWGPs: DTAGQE (pGA[CT]
AC[AGCT] GC[AGCT] GG[AGCT] CA[AG] GAO“) and NKC/SDM (pTCC AT[AG]
TC[AGC'I] [GC][AT][CT] TT[AG] TTOH) (positions 75-80 and 133-137 of yeast
SEC4, respectively [110). When combined with genomic DNA from S. mansoni,
a PCR yielded a product which migrated at approximately 200 bp on an agarose
gel. The band was excised from the gel and re-amplified with the same primers.
The resulting PCR product was cloned into the pCRTM1000 vector (Invitrogen,
San Diego, CA) and sequenced by the dideoxynucleotide method using the
Sequenase® kit (USB, Cleveland, OH). The clone was identified as clone 26.
The cloning and sequencing procedure was repeated using first-strand cDNA

synthesized from S. mansoni-derived RNA as template for the PCR.

40

An adult schistosome cDNA library packaged in lambda gt11 [27] was
probed with the cDNA insert of clone 26 using the plaque-lift technique [111].
Briefly, 10‘ plaque forming units of the lambda bacteriophage containing the

library were adsorbed onto strain Y1090 E. coli cells, plated on LB agar

supplemented with 0.2% maltose (w/v), 0.01M M9304 and 50 ug/mL ampicillin,

and incubated at 37°C until plaques in the bacterial lawn began to touch one
another. DNA represented by these plaques was lifted onto Hybond N+ nylon
membrane (Amersham, Arlington Hts., IL) in duplicate, denatured in 0.5M NaOH;
1.5M NaCl, neutralized in 1M TrisoHCl, pH 8.0; 1 .5M NaCl, and cross-linked onto
the membrane with W light (Stratalinker®, Stratagene, LaJolla, CA). Clone 26
was labeled with digoxigenin (GeniusTM System, Boehringer-Mannheim,
Indianapolis, IN) according to the manufacturer‘s instructions. Hybridization of
clone 26 with the plaque lifts was then carried out according to suggestions of
Boehringer-Mannheim for digoxigenin-labeled probes. Briefly, membranes were
prehybridized in 5x SSC (from 20x SSC: 3M NaCI; 0.3M sodium citrate; pH 7.5),
1% Blocking Reagent, 0.1% N-lauroylsarcosine and 0.02% sodium dodectyl
sulfate (SDS) for 2 hrs. at 65°C. Hybridization occured in the same buffer with
the addition of 200 ng labeled probe (0.5 - 2 pmol/mL buffer) at 65°C for 16 hrs.
Highest-stringency washes were in 0.5x SSC, 0.1% SDS at 65°C. Membranes
were then processed for chemiluminescent detection and exposed to auto
radiographic film. Bacteriophage plaques hybridizing with the labeled probe

were identified and purified according to standard techniques [111]. The cDNA

41

inserts from the purified lambda gt11 clones were amplified by PCR using
primers annealing to the flanking regions of the vector. The PCR products were
subcloned into pCRTMII vector (lnvitrogen), sequenced with the Sequenase ® kit,
and the sequences analyzed with the GCG software program [44]. One of the
identified clones, named clone 73, was further investigated.

5' Rapid Amplification of cDNA Ends (5'RACE) was performed according
to modifications of the technique described by Osamu et al. [100].
Oligonucleotide primers mentioned heretofore were designed using the OLIGO
software program [99] and synthesized by the Macromolecular Structure Facility
at Michigan State University. Using Superscriptmll Reverse Transcriptase
(GIBCO-BRL, Gaithersburg, MD), cDNA was generated from total schistosome
RNA with a specific antisense oligonucleotide primer
(,CGGTGAGGAAAGTAATCGCOH) annealing to the region immediately
downstream of the ORF of clone 73. A tailing reaction with terminal transferase
(Promega, Madison, WI) added approximately 15 deoxycytidine residues to the
5'end of the cDNA. This tailed cDNA was then used as template in a PCR with a
5' poly-G "anchor primer" and a 3' internal primer
(,GTTCCTGGCCTGCAGTATOH) specific to the DTAGQE region of clone 73.
The resulting PCR product was cloned into pCR-ScriptTMSK+ (Stratagene) and
sequenced with the Sequenase ® kit.

A final PCR was prepared to yield the complete cDNA representing the

sought schistosome gene. A specific primer was designed to anneal to the 5'

42
end of the ORF of the sequence identified with 5'RACE and contained a 5' tall

representing the Ndel restriction enzyme recognition site (bold) for subsequent
cloning purposes (DATCATATGATGGCAAAAAAGTCTTACGACOH). The 3'
primer was the same as that described above for the reverse transcription
reaction, with the exception that it had a BamHI linker at its 5' end, also for
unidirectional cloning purposes. The template first-strand cDNA for the PCR was
that used in the 5'RACE procedure, described above. The 645-bp PCR product
was cloned into pCR-ScripthK+ (Stratagene) and sequenced. It was identified

as smrab.

WEE

15 [19 male and 30 [19 female S. mansoni total RNA were separated
electrophoretically on a HEPES-buffered 1.5% agarose gel containing 16.2%
formaldehyde (vlv). RNA was transfered for 24 hrs. to Hybond N+ (Amersham)
nylon membrane in phosphate transfer buffer (7.9mM NazHPO,,, 17.1mM
NaHzPO4) and cross-linked to the membrane with UV light (Stratalinker ®,
Stratagene). The membrane was pre-hybridized for 2 hrs. in high-salt
hybridization buffer (1% BSA; 20mM sodium phosphate buffer pH 7.4; 15%
forrnamide (vlv); 1mM EDTA; 7% SDS; [Na']=750mM). Hybridization was carried
out for 16 hrs. at 42°C in 15 mL fresh hybridization buffer with 4x10° dpm labeled
probe. Washes were at 65°C in 0.5x SSC buffer. The membrane was wrapped

wet in SealWrapT'“I and exposed to autoradiographic film at -80°C.

43

Radiolabeled probes were prepared as follows. For the clone 73 probe,
plasmid preps of the pClel clone were prepared using the WizardTM Minipreps
columns (Promega) and the insert removed with EcoRI (Boehringer-Mannheim).
Following electrophoresis to separate insert from plasmid DNA, the insert was
eluted from a low-melt agarose gel using the QIAEX system (Qiagen, Chat
sworth, CA) and labeled with [32P]a-dCTP (3000 Cilmmole; DuPont-NEN,

Boston, MA) using a random primers labeling kit (GIBCO-BRL) to a specific
activity of 1.0-1 .5x10° dpm/piL. A 333-bp cDNA fragment of the schistosome

glutathione S-transferase (GST) gene was generated by PCR to use as a
positive and quality control for the Northern blots. The oligonucleotide primers
for this PCR were designed according to the nucleotide sequence of S. mansoni
GST [145] and correspond to positions 280-298 and 562-613 for the 5'
(,TGTTGGGTGGTTGTCCTAAOH) and 3' (pGTAG'ITCTTGA'ITGGTGGTAAAOH)

primers, respectively.

EE . [II I'I I I.

The pET3a expression plasmid was obtained from Novagen (Madison,
WI). The smrab insert was cut from plasmid preps of the pCR-ScriptTM-smrab
construct with Ndel and BamHI restriction enzymes (Boehringer-Mannheim) and
ligated into pET3a digested with the same enzymes. DH5or cells (GlBCO-BRL)
were transformed with the pET3a-smrab construct and sequence analysis

confirmed the presence of the full length of the smrab insert in frame and

44

downstream of the T7 promoter region in pET3a. pET3a-smrab was
subsequently transformed into BL(21)DE3, NovaBlue(DE3) and
BL21(DE3)pLysE cells (Novagen) for expression of the smrab protein. Similar
cloning procedures (using appropriate restriction enzymes) were followed with
pGEX4T-1 (Pharrnacia, Uppsala, Sweden) and pWSC (ATCC, Rockville, MD)
expression systems.

Optimal conditions for harvest of the recombinant protein from

BL21(DE3)pLysE cells were found to be as follows. One mL overnight culture of

transformed cells (grown in LB medium with 50 jug/mL carbenicillin and 34

uglmL chloramphenicol at 37°C, shaking at 300 rpm) was diluted 1:100 in fresh

medium and antibiotic. Growth was continued at 37°C, 300 rpm until the culture
reached an O.D.600 = 0.5, at which time translation of the recombinant protein
was induced with the addition of 0.4mM IPTG. Growth was continued under the
same conditions for 3.5 hrs. Cultures were then chilled on ice and cells pelleted
at 4°C and 2,500xg. Pellets were resuspended in 3 volumes (wlv) lysis buffer
(50mM Tris-HCI, pH 8.0; 100mM NaCI; 1mM EDTA) with the addition of 2.5mM
DTT and 0.5mM phenylmethylsulfonylfluoride (PMSF). The suspension was

rapidly frozen in a dry ice/methanol bath and thawed at room temperature to lyse
cells. DNAsel was added to 150 ugImL and incubation continued at room
temperature until the suspension lost viscosity. Inclusion bodies were pelleted

by centrifugation at 12,000xg at 4°C for 10 minutes and resuspended in 3

volumes (original pellet w/v) lysis buffer. The pellet was washed with the

45
addition of 0.5% Triton X-100, 9mM EDTA and 2.5mM DTT, incubated at room

temperature 5-10 minutes and recentrifuged at 12,000xg. To denature the
inclusion body pellet the washed pellet was suspended in 9 volumes (original
pellet wlv) 8M urea and 1mM DTT in lysis buffer and left at room temperature 16
hrs. Insoluble material was removed by another 12,000xg centrifugation at 4°C
and the supernatant used in subsequent manipulations.

Aliquots of total cell protein (following DNAse treatment) and of the soluble
and insoluble fractions (following the Triton X-100 wash step of the inclusion
body pellet) were analyzed by standard SDS-polyacrylamide gel electrophoresis
(PAGE) followed by staining in Coomassie Blue.

Cultures of untransformed BL(21)DE3pLysE cells and transformed,

uninduced cells were used as controls.

.fi‘ .H . .H ”A. ”I“ .u A. . .. .. .H I“ - . I.“ .
recombinantsmrab

Transformed cells were grown at 30°C pre- and/or post-induction rather
than at the usual 37°C. Experiments toward the same goal were performed
substituting minimal media (supplemented M9 medium or LB broth
supplemented with 0.3, 0.6, and 1.0M sorbitol) for LB broth at both 37°C and
30°C. Cells were harvested and processed as described above.

An optimal centrifugation rate was determined for collection of the

inclusion body pellet. Cell lysate was subjected to a range of centrifugation

46

rates, from 500xg to 12,000xg, followed by evaluation of pellet and supernatant
by SDS-PAGE. Wash conditions were evaluated for maximal removal of
bacterial protein and minimal loss of recombinant protein. Inclusion body pellets
were washed with 0.5% Triton X-100 or in 0.5, 1.0, 2.0 or 5.0M urea. Again,
pellets and supernatant were evaluated by SDS-PAGE.

Solubilization of inclusion bodies was attempted under the following
conditions: 8M urea or 6M guanidinium HCI 1 1mM DTT; 0.1, 1.0 or 2.0% SDS 1
5% BME; and these ten conditions for 1, 4, 8 or 16 hrs. Additionally, urea-
solubilized inclusion bodies were further denatured in KHZPO4, pH 10.7 buffer for
30 minutes, followed by readjustment of pH to 8.0 and dialysis against GTP-
binding protein refolding buffer (25mM HEPES, pH 7.5; 200mM NaCI; 5mM
MgClz; 0.1mM GDP; 1mM DTT).

To refold the denatured protein, the inclusion body pellet solubilized in BM
urea or SM guanidinium HCI were dialyzed against GTP-binding protein refolding
buffer or against the same buffer containing 1M NaCI. Inclusion bodies
solubilized in 2% SDS/5% BME were dialyzed against 50mM ammonium
bicarbonate either directly, or following elution from a 625-150 Sephadex column

(Sigma, St. Louis, MO).

Ell-l 'l . 'I .

A 15 “L aliquot of urea-solubilized inclusion bodies harboring smrab

protein was run onto a 12.5% SDS-PAG minigel (Mini-PROTEAN ® II, BioRad,

47

Richmond, CA). Protein was then transfered from the gel to lmmobilon-PSQ
(Millipore, Bedford, MA) PVDF membrane in 10mM CAPS, pH 11.0; 10%
methanol transfer buffer (Mini Trans-Blot ®, BioRad). The membrane was
stained with Ponceau-S and the 25-kDa band excised with a razor blade.
Sequencing of the protein on the membrane slice was carried out at the
Macromolecular Structure Facility (Michigan State University). Briefly, protein
was extracted from the membrane with a solution of 5% Tween20 in 20mM
ammonium bicarbonate for 60 minutes at 37°C. Tricarboxylic acid was then
added to a final concentration of 15% w/v, incubated 2 hours, then centrifuged to
recover precipitated protein. The protein pellet was washed in acetone and
dried, then digested with trypsin for 18 hours at 37°C (4% trypsin in 100mM
ammonium bicarbonate, pH 8.1). The digested protein fragments were

separated by high-perforrnance liquid chromatography.

WW

Schistosome homogenate was prepared by suspending freshly-harvested
worms in homogenizing buffer (50 mM Tris-HCI, pH 8.0; 1 mM MgCl2; 2 mM
DTT; 1 mM EGTA: 0.25% lubrol; 1 jug/mL leupeptin) and homogenized on ice.
The homogenate was centrifuged at 100,000xg for 20 minutes at 4°C. The

supernatant was aspirated and the pellet resuspended in Laemmli buffer [111].

48

Schistosomes were picked from mice into sterile RPMl-1640 medium
(GIBCO-BRL) warmed to room temperature. Worms were then transfered to
sterile incubation medium (RPMI 1640 with L-glutamine, without NaHCO3
(GIBCO-BRL) buffered to pH 7.4 with 20mM HEPES buffer; 50% horse serum;
50pM BME; 200U/mL penicillin; 0.2 mglmL streptomycin; 0.5 pig/mL
amphotericin B). Schistosomes were rinsed several times in this medium to
remove any mouse tissue. Worms were then transfered to autoclaved, capped
Erlenmeyer flasks containing fresh incubation medium (50 mL medium per 15
worm pairs). For treatment of worms with mevinolin, the incubation medium was
supplemented with 10'5M mevinolin (obtained from A.W. Alberts, Merck, Sharp
and Dohme Research Labs). Flasks were placed in a shallow 37°C water bath

and incubated with gentle agitation at 20 oscillations per minute for 48-72 hours.

IEIEI' I. I.

GTP-binding blots were prepared according to modification of the
technique described by Coulter et al. [21], in which protein is de- and re-natured
in situ prior to exposure to radiolabeled nucleotide. Proteins were separated by
SDS-PAGE and transfered to lmmobilon-P (Millipore) PVDF membrane as
described above. The membrane was soaked for 30 minutes at room
temperature in binding buffer (50mM NaHzPO4; 10mM MgCIz; 2mM DTT; 0.3%
Tween-20) with 8M urea. The protein was renatured on the membrane by

serially diluting the urea with equal volumes binding buffer until the concentration

49

of urea was less than 0.1 M. The membrane was incubated in fresh binding

buffer at 4°C overnight, then for 2 hrs. at room temperature in binding buffer
containing 3 pCilmL FZPja-GTP (DuPont-NEN; 3000 Cilmmole). The membrane

was washed extensively in binding buffer, dried, and exposed to

autoradiographic film.

Kill .I. I II' I I II.
The 645 bp smrab cDNA was amplified in a PCR using linearized pCR-

Scriptm-smrab construct as template, T7 promoter primer (5' primer), and the
smrab 3' primer described for the reverse transcriptase reaction, above. (The T7
primer was purchased from Stratagene; sequence: I,GTAATACGACTCACTAT
AGGGCOH.) RNA was transcribed from this PCR product with Ambion's Mega
scriptTM kit (Ambion, Austin, TX). Aliquots of this RNA were then added to a
rabbit reticulocyte lysate system (Retic Lysate IVTTM kit, Ambion) with or without
[35$]methionine (DuPont-NEN; 1175 Cilmmole) to yield [assj-labeled and
unlabeled smrab translation product. Concentration of in-vitm-translated smrab
was calculated from [35$]methionine incorporation into trichloroacetic acid-
insoluble product and the number of methionine residues in the smrab peptide.
The geranylgeranylation assay was performed according to the procedure
described by Sanford et al. [113]. [3SS]-labeled and unlabeled smrab from the in-

vitm translation procedure was diluted to 10nM in GGPT buffer (12mM Tris, pH

7.5; 0.6mM DTT; 3 mM MgCl2; 50 uglmL RNAase A) and 40% (v/v) reticulocyte

50
lysate in two Eppendorf tubes. [3H]GGPP (American Radiolabeled Chemicals,

Inc. (ARC), St. Louis, MO; 15 Cilmmol) was added to the tube with "cold"

translation product and "cold" GGPP (ARC) to that with [35$]-labeled translation
product at 10 uM to a final reaction volume of 20/.iL. Prenylation was initiated by
incubation at 37°C and continued for 2 hrs. The reaction was stopped on ice
and the products diluted with the addition of 20 14L Laemmli buffer. A 12.5%

SDS-PAG was prepared and loaded with aliquots of unprocessed [3SS]-Iabeled
translation product, [assj-labeled GGPT reaction product and [3H]-labeled GGPT
reaction product. Following electrophoresis, the gel was processed for

fluorography and dried prior to exposure to autoradiographic film.

IEII IEI'II |'[- II III)

Rabbit polyclonal antibodies were generated to the synthetic 14-mer
peptide HRLAEPPNSISGTC (amino acid residues 188-201 of smrab). The
peptide was synthesized by Research Genetics (Huntsville, AL) and conjugated
to Keyhole Limpet Hemocyanin (KLH; Pierce, Rockford, IL) with a sulfo-m-
Maleimidobenzoyle-N-hydroxysulfosuccinimide ester crosslinker (Pierce),
following the protocol provided in the crosslinker kit. Following conjugation, the
KLH-peptide was dialyzed against PBS (140mM NaCI; 27mM KCI; 10mM
NazHPO4; 1.76mM KH2P04; pH 7.2) to remove unconjugated peptide and EDTA
prior to injection into the rabbits.

Two female New Zealand White rabbits weighing approximately 2.5 kg

51

were used for generation of antibodies. KLH-conjugated peptide was mixed with
Freund's Complete Adjuvant (Gibco) BRL) (50% vlv) for delivery of 500 pg
antigen by subcutaneous injection. Pre-immune serum was drawn prior to
injection of antigen. Rabbits were boosted with 250 pg antigen in F reund's

Incomplete Adjuvant in two-week intervals; immune serum was drawn prior to
each injection. Immunity was assessed by dot-blot and Western blot analysis
(see below). Satisfactory immune response was detected in the fourth immune
serum.

The lgG fraction of the pre-immune serum was purified on a Protein A
column, following a standard protocol [55]. Antibody concentration in the pooled
fractions was measured by spectrophotometric absorption at 280 nm and
averaged 2.0 mglmL. These IgG fractions were used in subsequent
experiments.

The 196 fraction of anti-peptide serum specifically recognizing the 14-mer
peptide was affinity-purified with Pierce's lmmunoPure ® Ag/Ab Immobilization
Kit. Antibody concentration at 280 nm was 0.2 mglmL. This specific lgG fraction
(c-pep) was used in subsequent experiments (Western blotting and

immunohistochemistry).

Mzflestemflotjnalysis
Dot blots were used to screen for the presence of antibody recognizing the

14-mer peptide in or-pep immune serum. Briefly, 25 pg peptide and 2.5 pg KLH

52

were dotted onto nitrocellulose and air-dried. The blotting procedure was
followed according to specifications in the ECLTM Western Blotting kit (Amer
sham). Antiserum was diluted 1:100. Horse-radish peroxidase (HRP)-
conjugated goat-anti-rabbit antibody (Sigma) was used for chemiluminescent
detection of bound primary antibody.

Western blots were prepared by first separating proteins of antigenic
interest by SDS-PAGE, and then electrophoretically transfering these proteins
onto lmmobilon-P (Millipore) membrane, as described above. Again, membranes
were processed according to the protocol outlined in the ECLT'VI kit and exposed

to autoradiographic film for detection of labeled bands. Primary antibody was

used at the following dilutions: pre-immune lgG @ 125000 (0.4 pg/mL); oi-pep @

121000 (2 pg/mL). The HRP-conjugated secondary antibody was used at 1:1000

dilution.

II I I . I I . I

Whole Schistosoma mansoni were picked from mice, rinsed in PBS and
fixed for 3 hours at room temperature in 10% neutral-buffered formalin (Surgi
path, Richmond, IL). Fixed worms were blocked in paraffin and sliced into 2 pm
sections, which were transfered onto poly-L-Iysine-coated glass slides. Slides
were stored at 4°C prior to and following processing for immunohistochemical
detection. Processing was done at room temperature.

Sections were de-paraffinized in xylene for 20 minutes, then rinsed in

53

100% and 95% ethanol for 3 minutes each. Sections were rehydrated in PBST
(PBS with 0.3% Tween-20) for 20 minutes and then blocked in PBST with 5%

normal goat serum (Gibco-BRL) and 0.1% bovine serum albumin (BSA) for an

additional 20 minutes. Primary antibody was diluted to 3 pg/mL in a buffer

consisting of PBST with 2% normal goat serum and 0.1% BSA. 200 pL diluted

antibody was placed directly on the slide and incubated for 2 hours. Slides were

washed for 10 minutes with three changes of PBST, then incubated for 1 hour
with 200 pL Cy3-conjugated goat-rabbit antibody (Jackson Immunoresearch

Laboratories, Inc.) diluted 1:1000 in the antibody-dilution buffer described above.
Slides were washed again in PBST as described. A drop of buffered glycerol
(glycerol diluted 2:1 with 0.5M Na2003 buffered to pH 8.6 with 0.5M NaHCOs)
was placed on each slide near the specimens, and covered with a glass
coverslip. Labeled specimens were examined with a Nikon HFX-ll fluorescent
microscope, using a Nikon DM580 G-1A (green) filter. Specimens were
subjected to immunohistochemical analysis with, or-pep (affinity- purified) lgG,
pre-immune lgG, and affinity-purified a-pep IgG which had been pre-incubated
with 10-fold excess 14-mer peptide for 1 hour prior to application to specimens.
Fresh worms were examined, as were worms which had been treated in-vitro for
60 hours with 10‘5M mevinolin, and worms which had incubated under the same

conditions but without mevinolin.

RESULTS

AWL

A gene encoding a rab-related LMWGP was successfully cloned from
Schistosoma mansoni, using a PCR-based strategy. Degenerate oligonucleotide
primers designed to anneal with two highly-conserved regions in ras-related
LMWGPs were used to screen genomic schistosome DNA in an initial PCR. The
resulting DNA segments were inserted into a plasmid vector and sequenced. A
search of the GenBankm-EMBL sequence databank [44] confirmed the cloning
of a 195-bp sequence stretching between the two primers which shares a 62%
amino acid sequence homology with rab-related proteins. One of these PCR-
generated clones, clone 26, was subsequently used for cDNA library screening
(see below). The PCR was repeated with cDNA substituted for genomic DNA in
order to preclude the presence of an intron within clone 26, which could
complicate its use as a probe against cDNA libraries. Analysis of the resulting
cDNA clones proved identical to that of clone 26.

Clone 26 was used to search an adult S. mansoni cDNA library for larger
segments of the corresponding gene. This screen identified several distinct
lambda gt11 clones. On the basis of sequence comparison, one of these, clone

73, was found to represent all but approximately 60 N-tenninal amino acids of a

54

55

rab-related protein sequence.

A common deficiency of cDNA libraries is the representation of genes with
truncated 5' ends: this is also the case with clone 73. 5' RACE was performed in
order to identify the N-terrninus of the gene corresponding to this clone. An
antisense oligonucleotide primer was designed to generate specific first-strand
cDNA representing only the ORF of clone 73. This cDNA was tailed, and an
internal primer was used to generate the relatively short, double-stranded PCR
product. Sequence analysis of several identical clones of this DNA segment
suggested that their 204-bp ORF corresponds to the N-terrninus of the sought
schistosome LMWGP.

Finally, a specific sense primer recognizing the 5' terminus of the 5'RACE
product and the antisense primer used to generate the first-strand cDNA for the
5'RACE reaction yielded a 645-bp PCR product. Again, this cDNA was cloned
for sequence analysis. One of these clones, named smrab, was studied further
and appeared to represent the full-length cDNA corresponding to a schistosome
LMWGP.

The library-screening and 5'RACE steps yielded a number of other clones
which represented smaller sections of the ORFs of clones 73 and the 5'RACE
clone (respectively), but which also gave some sequence information on the 5'
and 3' untranslated regions of the smrab gene. The definitive 5' and 3' (poly-A)

ends of the gene were not identified; however the gene appears longer than

56

Iablej, Nucleotide and deduced amino acid sequence of smrab. Motifs
serving as templates for the oligonucleotide primers used in the original PCR
to generate clone 26 are marked with a solid line. Sequence constituting
clones 26 and 73 are marked with dotted and dashed lines, respectively.

1 at tgg
6 aag cgg tgt ttt caa cct tgc tga gat ctt act aat

42 atg gca aaa aag tct tac gac gca ctg ttt aaa at
1 M A K K S Y D A L F K I

78 ctg ctc att gga gat tct ggt gtt ggc aaa act tgc
13 L L I G D S G V G K T C

114 cta ctc gtt cgt tat gta gaa gag tct ttt gta cca
25 L

150 acc ttc att tca ata gga atc gat ttt aag att
37 T F

186 aaa aca att gag
49 K T l E

gaa gga aag aaa ata aag cta

 

222 cag ata tgg gat act gca ggc cag gaa cgc ttt cat
61 Q l W D T A G O E R F H

 

eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

258 aca att acc tcc tct tac tat cgc ggt gcc atg ggt
73 T l T Y

 

eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

294 ata atg ttg ata tac gat att act agt cgt caa acg
85 I M L I Y D l T S R Q T

 

330 ttt gat aac gtc cag acg tgg atg aat agg acc ttt
97 F N T

 

eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

109 F L A S N E V E K L L l

 

oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo

 

57

 

 

 

 

 

 

 

 

W

402 got at aag tgt gac atg gcg cat aac cga gtt gta
121 A N K C D M A H N R V V
438 tct tat gaa gag ggt cgg caa aaa gca gag gaa tat
133 S Y E E G R O K A E E Y
474 999 att ggt ttc ctg gaa aca agt gca aag tcg tca
145 G I G F L E T S A K S S
510 acg aat gta cac aag gcc ttc gag gaa ctg act cgc
157 T N V H K A F E E L T R
546 gcc atc ctc cat aac agt tct ata cgt aga cat gag
169 A I L H N S S l R R H E
582 tcg tct acg gtg act caa gtc cac cgg tta gcc gaa
181 S S T V T O V H R L A E
618 cct cca aat tcc att tct gga acg tgc tgt agt aca
193 P P N S l S G T C C S T
654 gga tag tcc ttc aag cga tta ctt tcc tca ccg ttt
205 G "

690 tcc ttt caa tat taa ctt atg gta atg ctt ata aat
726 aca atg ttg cga ttc atc ata aga tcc ata ttt tca
762 gag tcc cgc caa aag aca ttc aga gtt tta gtt tct
798 gac tcg taa ttt ggt tta atc ttt gtg atg agg tgc
834 ttc tgt cat act ctg gac cta cga ttc acg ttc ggt
870 tgt gat gaa aca ctt gcg cta gcg ttg aaa ttc ctt
906 gtt ata cca cac taa atg ttt ttc cac ttc ata taa
942 agc cta caa agc caa tta gtc acc att tac ttt tta
978 at gta aat ata ttt gaa ttc cgc cga tac tga cgg
1014 gct cca gga gtc gtc gcc acc

58

labial Amino acid sequence alignment of smrab and homologous rab
proteins. Percent sequence identity and similarity (Dayhoff grouping 28]) of
smrab with each protein are indicated at the end of each sequence. Conserved
GTP-binding domains are marked with a solid bar; regions representing loop L2,
oi-helix 3, and loop L7 are indicated (dashed lines). Origin of compared proteins:
ora1 & 2 - Discopyge ommata [96]; SAS1 & 1 - Dictyostelium discoideum [114];
SEC4 - Saccharomyces cereviciae [110]; yptl - Mus musculus [56]; rab10 -
Canis canis [13]; rab13 - Homo sapiens [146].

lam Amino acid sequence alignment of smrab and homologous rab proteins.

oral
rablO
oral
rabl3
sasl
sasl
yptl
sec4
smrab

oral
rablO
oral
rabl3
sasl
sasl
yptl
sec4
smrab

oral
rablO
ora1
rab13
sasl
sasl
yptl
sec4
smrab

oral
rablO
oral
rabl3
sasl
sasl
yptl
sec4
smrab

.MAKKTYDLL
.MAKKTYDLL

............ MAKTYDYL
............ MAKAYDHL
..... MTSPA TNKPAAYDFL
..... MTSPA TNKSAAYDYL
............ MNPEYDYL
MSGLRTVSAS SGNGKSYDSI

.MAKKSYDAL

59

FKLLLIGDSG VGKTCVLFRF SDDAFNTTFI
FKLLLIGDSG VGKTCVLFRF SDDAFNTTFI
FKLLLIGDSG VGKTCLLFRF SEDAFNTTFI

FKLLLIGDSG VGKTCLIIRF

AEDNFNNTYI

VKLLLIGDSG VGKSCLLLRF SDGSFTPSFI

IKLLLIGDSG VGKSCLLLRF
FKLLLIGDSG VGKSCLLLRF
MKILLIGDSG VGKSCLLVRF
FKILLIGDSG VGKTCLLVRY

STIGIDFKIK TVELHGKKIK
STIGIDFKIK TVELQGKKIK
STIGIDFKIR TVELDGKKIK
STIGIDFKIR TVDIEGKKIK
ATIGIDFKIR TIELEGKRIK
TTIGIDFKIR TIELEGKRIK
STIGVDFKIR TIELDGKTIK
TTIGIDFKIK TVDINGKKVK
STIGIDFKIK TIELEGKKIK

_ .... Ll----

DITNAKSFEN
DITNGKSFEN
DITNEKSFDN
DITDEKSFEN
DVTDEKSFGS
DVTDEKSFGN
DVTDOESFNN
DVTDERTFTN
DITSRQTFDN

ISKNLRNIDE
ISKHLRNIDE
IKNNIRNIEE
IQNWMKSIKE
IRNHIRNIEQ
IRNNIRNIEQ
VKOWLQEIDR
IKQNFKTVNE
VQTWMNRTFS

LOIWDTAGQE

LQIHDTAGQE
LQIHDTAGQE
LQVWDTAGQE
LQIWDTAGOE
LQINDTAGQE
LQIWDTAGQE
LQLHDTAGQE
LQINDTAGQE

HANEDVERML
HANEDVERML
HASSDVERMI
NASAGVERLL
HASDSVNKML
HATDSVNKML
YASENVNKLL
HANDEAQLLL
LASNEVEKLL

------- c3----~-.----L7----

IAREHAIRFF
IAREHGIRFF
LAIDYGIKFL
LAREHGIRFF
LADEYGIKFL
LADEYGIKFL
FADSLGIPFL
LAKELGIPFI
KAEEYGIGFL

ISTTGGGTGL
ISSGGGVTGH
AVDKLKSPPK
PSTDLKTCDK
CITP ..... N
QPGTNLGANN
SNVKIQSTPV

ETSAKANINI

ETSAKVNINI
ETSAKSSINV
ETSAKSSMNV
ETSAKNSVNV
ETSAKNSINV
ETSAKNATNV
ESSAKNDDNV
ETSAKSSTNV

KK ........
KSK .......
KPSQKKKQLS
KNTNK .....
NKKNT .....
NKKKA .....
KQSGGG....

EKAFLTLAED
EKAFLTLAED
EEAFITLARD
DEAFSSLARD
EEAFIGLAKD
EEAFISLAKD
EOSFMTMAAE
NEIFFTLAKL
HKAFEELTRA

..CCS.
..CC..
FRCSLL
..CSLG
..CC..
..CC..
..CC..

ISINSGSGNS SKSN ........ CC..
QVHRLAEPPN SISGT ....... CCSTG

RFHTITTSYY
RFHTITTSYY
RFRTITTAYY
RFKTITTAYY
RFRTITTAYY
RFRTITTAYY
RFRTITSSYY
RFRTITTAYY
RFHTITSSYY

LGNKCDMDDK
LGNKCDMNEK
LGNKCDMEAK
IGNKCDMTEK
IGNKCDMAEK
VGNKCDLTTK
VGNKSDM.ET
IANKCDMAHN

ILOKTPVKE.
ILRKTPVKE.
IMTKLNKKM.
ILLKSGGRR.
IKKRMIDTP.

IKKRMIDTP. ..

IKKRMGPGA.
IQEKIDSNKL
ILHNSSIRR.

SEDSFTPSFI
ADDTYTESYI
VEDKFNPSFI
VEESFVPTFI

---L1--

RGAMGIMLVY
RGAMGIMLVY
RGAMGIMKVY
RGAMGIILVY
RGAMGILLVY
RGAMGILLVY
RGAHGIIVVY
RGAMGIILVY
RGAMGIMLIY

RVVPKGKGEQ
RQVSKERGEK
RKVOKEQADK
KVVDSSRGKS
KVVDSSRGKS
KVVDYTTAKE
RVVTADQGEA
RVVSYEEGRQ

...PDRENVD
...PNSENVD
...NENSLOE
...SGNGNKP
...NDPDHTI
.NEQPQVV
...TAGGAEK
VGVGNGKEGN
...HESSTVT

50

100

LGNKCDMEDK RVVLKSKGEQ 150

100

60

mm Amino acid sequence alignment of the L7 "specificity domain".

9.: . o . 'J .I
ora1 E H A N E D V E R 33 89
rab10 E H A N E D V E R 33 89
ora1 E H A S S D V E R 56 89
rab13 E N A S A G V E R 56 89
sasl Q H A S D S V N K 44 78
sasl Q H A T D S V N K 33 78
yptl R Y A S E N V N K 44 89
sec4 E H A N D E A Q L 11 78
smrab S L A S N E V E K

61

1034 bp (Table 2). Deduced amino acid sequence analysis of smrab reveals an
ORF of 205 amino acids, which is predicted to fold into a protein of
approximately 22.6 kDa (Tables 2 and 3). Comparison of this sequence with the
GenBankTM-EMBL database indicates a 66-71% similarity and 49-59% identity
(Dayhoff grouping [28]) with rab-related LMWGPs of several species: ora1 & 2 of
the electric eel, Discopyge ommata [96]; SASl & 2 of the slime mold,
Dictyostelium discoideum [114]; SEC4 of the fission yeast, Saccharomyces
cereviciae [110]; and mammalian ypt1, rab10 and rab13 [13,56,146]. Sequence
homology in the C-terrninal third of the peptide is characteristically poor. Smrab
shares the four highly-conserved GTP-binding domains of proteins in the ras-
related superfamily. Within the analyzed group of rab-related proteins, smrab
maintains an especially high sequence homology in regions believed to
represent the effector domain (loop L1) and the specificity domains (loop L7 and
c-helix 3) of the proteins [35,102]. The effector domain is the region which
interacts with the molecules (usually proteins) relaying and receiving signals to
and from the rab protein. The specificity domain is discussed further in the
Discussion section; it is the region of the molecule believed to impart functional
specificity upon the particular rab protein. Consistent with its identity as a rab-
related LMWGP, smrab terminates in a geranylgeranylation signal motif. In this

case it is the relatively rare CCXXX sequence.

WWW

62

Eigufl Northern blot analysis of smrab. Schistosomal RNA was probed with
[32P]-Iabeled clone 73 (A) or with the [32P]-labeled glutathione S-transferase PCR

fragment (B). Lanes marked F and M respectively represent female and male
RNA. Markers indicate molecular size in kilobases.

63

Figure 6. Northern blot analysis of smrab.

.77kb-

1.4 - l'.

0.7s _

   

e4

Hybridization of schistosomal RNA with clone 73, which represents the C-
tenninal 71% of smrab, reveals a single band of approximately 1.5 kb (Figure 6).
The mRNA represented by this band appears to be far more abundant in male
than in female schistosomes. Control hybridizations with a PCR-generated
fragment of schistosomal GST resulted in the expected single 1.0-kb band [132];
GST message also appears to be generated to a greater degree in male than in

female worms, although the discrepancy is far less so than with smrab.

CE '[ l'l I I.

Expression of recombinant smrab was attempted in three different
expression vectors and 5 strains of E. coli cells. Only one combination of
these,the pET3a-smrab construct in BL(11)DE3pLysE cells, produced the protein
to any degree, and this quite abundantly, generating approximately 50% of total
cell protein as smrab. However, the recombinant protein was retrievable from
bacterial cells only in the insoluble form of precipitated inclusion bodies (Figure
7). These complexes form as newly-synthesized, denatured, recombinant
protein co-precipitates with ribosomes, nucleic acids and other bacterial
cytoplasmic proteins. Strong hydrophobic interactions between protein strands
and disulfide bond formation appear to be important in many instances [83].
Numerous theories attempt to explain the formation of these poorly-soluble
protein “clumps”, ranging from too much protein being synthesized too rapidly to

reaction by cells to a toxic product: in short, the cause of inclusion body

65

formation is not fully understood. It does appear to be involve more than simply
high localized concentration of protein, as very strong solvents are usually
required to solubilize the complexes (e.g., 6M guanidinium Hcl, 8M urea, 1%
SDS) [82]. The inclusion bodies formed in the course of generating smrab were
subjected extensively to strategies which others have employed in efforts to
denature and refold similarly-insoluble proteins [4,82,83,105]. Transformed
bacteria were grown at lower temperatures and in minimal or osmotically-
challenging media. The rationale for this is that such conditions are designed to
slow cell growth, and thereby to slow the rate of protein synthesis and the
liklihood of inclusion body formation. The recombinant protein synthesized under
these restrictive conditions appeared no less soluble.

Following cell harvest, optimal conditions for centrifugation rates, washing
inclusion body pellets, solubilizing pellets, and reducing environment (BME, DTT)
were determined. Inclusion body protein denatured only under extreme
conditions (eg. 8M urea). Following denaturation, protein refolding was
attempted under various dialysis conditions, all of which resulted in precipitation
of the protein within a few hours. Some work has indicated that bacterial
impurities in the protein preparation may promote improper folding and re-
precipitation of the recombinant protein [82]. To test the applicability of this
observation to smrab, the inclusion body pellet was solubilized in 1% SDS with
5% BME (protein suspensions in BM urea and 6M guanidinium HCI clogged the

column) and filtered through a Sephadex G15-150 column. This too, however,

66

Eigure 2, SDS-PAGE of recombinant smrab. Recombinant protein was
generated from the pET3a-smrab expression vector construct in
BL11(DE3)pLysE cells. Proteins were separated electrophoretically and stained
with Coomassie Blue.

Lane 1: Molecular weight marker (kDa).

Lanes 2- 4: Total, soluble, and insoluble protein fractions, respectively, of
untransformed, induced cells (negative control). The arrow indicates the 15.6
'kDa band representing chloramphenicol acetyl transferase (CAT). Note that
CAT is primarily in the soluble fraction.

Lanes 5-7: Total, soluble, and insoluble protein fractions, respectively, of the
pET3a-smrab expression product. The arrow indicates smrab: note this is
primarily in the insoluble fraction.

67

Figure 7. SDS-PAGE of recombinant smrab.

~lH
lid

‘. v

9

 

68

failed to prevent re-precipitation of the protein during subsequent dialysis.

Marston et al. [83] were successful in enhancing solubilization of a
recombinant protein in high-pH buffer prior to refolding by re-adjustment of pH
and dialysis. The rationale for this procedure is that the high-pH conditions
promote the disruption of non-covalent interactions in the inclusion bodies and
allow more efficient thio-disulfide interchange. This was attempted with smrab
and, again, proved unsuccessful in keeping the recombinant protein in a soluble
state.

All subsequent procedures involving recombinant smrab protein were thus
performed using the urea-solubilized inclusion body pellet.

Recombinant, urea-solubilized smrab migrates at the expected ca. 13-kDa
size on SDS-PAGE. The migration pattern does vary from gel to gel, however,
positioning the smrab band anywhere between 13 and 30 kDa (e.g., Figure 7
lane 7 vs. Figure 8 lane 1). This variation is most likely related to the insolubility
of the recombinant protein in any environment other than 8M urea. Much of the
urea in the sample loaded onto the gel precipitates immediately in the well,
thereby leaving inconsistent amounts of urea to migrate with the protein into the
gel, and resulting presumably in variations in the folding characteristics (and
hence mobility) of the protein as it moves through the gel matrix.
Chloramphenicol acetyl transferase (CAT) expressed by BL11(DE3)pLysE cells
is a 15.6 KDa protein which threatened to obscure visualization of the similarly-

sized smrab protein on SDS-PAGE. CAT is, however, a largely soluble protein

69

F'E

Elm GTP binding blot. Proteins were separated by SDS-PAGE, transfered
to PVDF membrane and de-and re-natured in situ as described in the text.
Membrane-bound proteins were labeled with [32Pjoi-GTP, then processed for
autoradiography.

Lane 1: Recombinant rat rab1b (positive control) [66].

Lane 2: Recombinant smrab.

Lanes 3,4: Pellet and supernatant, respectively, of 100,000xg fractionation of
schistosome homogenate.

Lane 5: Total cell protein from untransformed BL11(DE3)pLysE cells (negative
control). Band represents bacterial elongation and initiation factors which bind
GTP (molecular weights 45-55 kDa).

70

Figure 8. GTP binding blot.

 

i3 .
.13 inc
ice ... . is
m .9. cl in; N P .e.

71

EiguLei. Geranylgeranylation of smrab. The recombinant protein was
translated in an in-vitro rabbit reticulocyte lysate system with or without [35S]-
methionine. Labeled protein was then incubated with "cold" GGPP; unlabeled
protein was incubated with [3H]GGPP in the geranylgeranylation assay
described in Materials & Methods. Free [3SS]-methionine and small translation
products with nonspecificaIly-incorporated [35S] appear at the bottom of lanes 1
and 1; free and nonspecifically-bound [3H]GGPP is found at the bottom of lane
3.

Lane 1: [35$]-Iabeled translation product (unprocessed). Small arrow marks
band representing smrab protein.

Lane 2: [sssl-labeled translation product reacted with "cold" GGPP. Note greater
band intensity in region just above the large smrab band, co-migrating with
adjacent [3H] band in Lane 3.

Lane 3: Unlabeled translation product reacted with [3H]-GGPP. Large arrow
indicates the band representing geranylgernylated smrab.

72

Figure 9. Geranylgeranylation of smrab.

 

73

while smrab is not, and therefore the two are easily distinguishable (see Figure
7). Recombinant smrab does not transfer to nitrocellulose using conventional
Tris-glycine-methanol buffer, presumably because the protein precipitates in the
gel under these buffer conditions. High-pH buffer conditions (pH 11.0 CAPS
buffer; 10% methanol) and replacement of nitrocellulose with a specialized PVDF
membrane (lmmobilon P50, Millipore) results in transfer of approximately 50 -
75% of the protein (estimated from the amount of protein staining in the gel
following transfer vs. that staining on the membrane). This successful transfer
allowed N-tenninal sequence analysis of the recombinant protein, which confirms
the sequence of the first 19 amino acids to match that of the deduced sequence

of smrab (MAKKSYDALFKILLIGDSG).

D EIE-l . I. I .
A modification of conventional nucleotide blotting methods succeeded in
demonstrating the ability of the recombinant smrab protein to bind GTP (Figure
8). Urea-solubilized protein was separated by SDS-PAGE, transfered to PVDF
membrane, and then underwent de- and renaturation in situ. Processed in this
manner, recombinant smrab binds GTP avidly. The same blot demonstrates the
abundant presence of LMWGPs in schistosome homogenate. Southwestern
blotting was also performed with more conventional methods which did not
process the protein for denaturation and refolding once it had been transfered to

the PVDF membrane. These experiments were unsuccessful in demonstrating

GTF

blnc

bill

If"

as

74

GTP binding by smrab. Endogenous proteins in schistosome homogenate do
bind GTP without the modifying steps in the procedure [16], indicating that the
failure to label recombinant protein is not due to a problem with the nucleotide

binding technique.

EE H II. I . E II EEEI
The identity of smrab as a rab-related protein with a CCXXX carboxy-
terrninal sequence motif dictates that the protein is geranylgeranylated. To
ascertain this empirically, recombinant smrab was subjected to a GGPT assay
as described in Materials & Methods. For this assay the recombinant protein
was specially generated in a rabbit reticulocyte lysate in-vitro translation system
because efforts by others have demonstrated that rab proteins which are
normally geranylgeranylated are resistant to this modification if they have been
translated in bacterial expression systems [58]. Indeed, efforts to demonstrate
the isopreny-lation of bacterially-expressed recombinant smrab failed. (These
experiments were further complicated by the problem of insolubility encountered
with this protein.) The electrophoretic separation of the products of the GGPT
assay (Figure 9) reveal a single [3H]-labeled band migrating at a slightly heavier
molecular weight than that representing the [assl-labeled, unprocessed
translation product. A similarly-sized band is evident in the lane loaded with
[”Sl-labeled smrab that had been reacted with "cold" GGPP. These data

demonstrate the ability of smrab to serve as substrate for GGPT.

75
E I I . I I . I

Affinity-purified polyclonal antibodies were generated which recognize smrab
specifically on Western blot analysis. The peptide sequence which served as the
epitope for these antibodies was selected on the basis of the following
considerations. First, the tertiary structure of the protein molecule was
considered. Extrapolating from the structure described for H-ras [5,67], the
peptide segment is located on the surface of the protein molecule. Second, the
surface probability, antigenicity, hydrophobicity and hydrophilicity of the peptide
were predicted with several computer programs (GCG's PepPlot and
PeptideStructure [44] and Hydrophobic Cluster Analysis [42]). The surface
location of the peptide is important in enabling antibodies to recognize the native
protein molecule by this peptide epitope. The single, terminal cysteine in the
sequence was placed for the convenience of conjugating the peptide to a larger,
immunogenic molecule (KLH) by the sulfliydryl group of this cysteine.

Initial screening of antisera with dot blots for recognition of the 14-mer
peptide indicates a strong response with the fourth anti-peptide serum. Specific
anti-peptide IgG (oi-pep) from this antiserum was isolated on a column of
immobilized peptide and tested for smrab recognition by Western blotting.
Figure 10 demonstrates that affinity-purified or-pep antibodies selectively
recognize recombinant smrab and rab5, but not rab1b. In schistosome
homogenate, a-pep identifies a single band of ca. 35 kDa in the pellet fraction of

the 100,000xg centrifugate, and nothing in the supernatant.

76

EiguLe_L0_, Western blot with a-pep (A) and pre-immune (8) antibodies.
Molecular size is indicated in kDa.

Lane 1: Recombinant rat rab1b (control) [66].
Lane 2: Untransformed BL11(DE3)pLysE cells.
Lane 3: Recombinant smrab generated in BL11(DE3)pLysE cells.

Lanes 4,5: Pellet and supernatant, respectively, of 100,000xg centrifugation of
schistosome homogenate.

Lane 6: Recombinant human rab5 [112].

77

Figure 10. Western blot with c-pep antibodies.

[0. 1 2 3 4 5 6
' V
66..
4s.
A 31-. 5 -31
21.5,, b ! _21.s
14.5. i
h L 5 J l
'
B 31- - 31
-215
14.5.,

78

Lubrol was added to the schistosome homogenizing buffer to solubilize
membrane-bound proteins in an effort to make them more accessible to
antibodies in the soluble fraction, but this did not change the distribution of the
protein identified by oi-pep antibodies.

An effort was also made to generate antibodies to the inclusion body pellet,
with the rationalization that such polyclonal antibodies may recognize smrab with
greater specificity than antibodies raised to a peptide. Western blot analysis of
the lgG fraction of this anti-inclusion-body antiserum, however, showed no such
specificity, and work with this antiserum was not pursued further.

The affinity-purified or-pep antibodies clearly recognize specific structures in
paraffin-embedded sections of schistosomes (Figures 11-15). Negative control
incubations with pre-immune IgG or with a-pep IgG which had been pre-
incubated with excess peptide resulted in the homogeneous, nonspecific labeling
of visceral structures in the worms. Fresh worms labeled with immune (c-pep)
lgG, however, gave a distinct, punctate pattern in subtegumental structures (in
addition to the homogeneous visceral fluorescence seen in the negative
controls). In both male and female sections or-pep forms a distinct line of
fluorescence which follows the outline of the worm surface in a subtegumental
layer (Figures 11 & 13). Labeling in the males appears to be greater than in the
females, although the location of the target(s) recognized by the antibodies
appears to be similar between the two sexes. In mevinolin-treated male

schistosomes (Figure 11), distribution of the fluorescent label presents more

abu

WOI

lnr

lha

h0l

an:

Inc

79

abundantly and in a less even and confined distribution than in untreated male
worms. Likewise, sections from female schistosomes which had been incubated
in mevinolin (Figure 14) indicate a more diffuse distribution of bound antibody
than sections from untreated females. Labeling in these treated females,
however, does not appear to be increased over that in control sections; it
anything, it may even be more sparse. Sections from worms which had been
incubated like the mevinolin-treated worms, but with the omission of mevinolin in
the incubation medium, were also examined (Figure 15). The pattern of antibody
label in these sections is less intense and less delineated than in fresh-worm
sections, with occasional and irregular clusters of fairly concentrated
fluorescence. Label in these sections is also greater in those from male than

from female worms.

F.saab“w

80

Ejguresjjfii Immunolocalization of smrab with anti-pep antibodies. Paraffin-
embedded sections of male and female Schistosoma mansoni were incubated
with primary rabbit antibody, followed by secondary Cy3-conjugated goat-anti=
rabbit antibody. Primary antibody is a-pep lgG; negative controls are pre-
immune lgG or a-pep lgG which had been inactivated by pre-incubation with
excess peptide. Bar = 50 pm.

WW a)pre-immune lgG; b)pre-incubated ci-pep

lgG; c)a-pep lgG: note subtegumental immunofluorescence.

EignLerzMeximlinJreaMmalejchislosgmes. Male schistosomes were

incubated in 10‘5M mevinolin for 60 hours prior to fixation and sectioning. a)pre-
immune lgG; b)pre-incubated a-pep lgG; c)oi-pep lgG: note greater intesity of
fluorescent label and its relatively irregular distribution.

MW 80 Dre-immune lgG; Mere-incubated ei-
pep lgG; c,d)oi-pep lgG: note the distinct subtegumental line of immuno-
fluorescence.

W (as in Figure 12)- a)Pre-
immune IgG; b)pre-incubated a-pep lgG; c,d)or-pep lgG: note redistribution of
immunofluorescence from the distinct line seen with fresh worms to the diffuse
pattern seen here.

WWW Male and female

schistosomes which had been incubated for 60 hours under the same conditions
as the treated worms, with the omission of mevinolin in the incubation medium.
a)male (left) curled around a cross-sected female, labeled with pre-immune lgG;
b)male section labeled with oi-pep lgG; c)male (top) and female sections labeled
with a-pep lgG. lmmunofluorescence is less than in fresh and mevinolin-treated
worms, but its pattern is still relatively linear as in fresh worms.

81

. I

 

.. C

l

. . . a.
ll .. .9: .....i .. . ..i .
(Q Q; In 1‘. ,i.....JIdv.I‘ h s i.

mmaomoumHSQm meE mepb

—.—. 0.; (A;

82

 

 

onOmoumflnom mHmE
ccumopuicflaocfl>oz

N Peppflh

mQEOmOumflflum QHNEQM Smwhh "thaflh

 

lmmrtcmci‘aTltfi ‘FIIIIIIH a

 

VFopaflp

WOEOmOUWHd—Um OHNEQM UOUNOHUICHHOCflKVOZ

 

DISCUSSION

This work employed a PCR-based strategy to clone a novel low-molecular-
weight GTP-binding protein from S. mansoni. It is the first such gene to be
cloned from the schistosome; indeed, only one other rab-related protein has
been cloned from a parasite [9]. Criteria for classification of a protein as a
member of the ras-related superfamily of GTP-binding proteins are as follows: 1)
protein size of 10 - 19 kDa; 1) sequence homology in highly-conserved regions
of monomeric GTP-binding proteins; and 3) demonstration of conserved
sequence motifs characteristic of subfamilies of ras-related proteins [61].

The cloned schistosomal LMWGP introduced here as smrab has an open
reading frame of 105 amino acids, which calculates to a predicted protein size of
ca. 11.6 kDa. Recombinantly-expressed smrab supports this calculation by co-
migrating with molecular weight markers to the expected position on SDS-PAGE.
Northern blot analysis conflrrns the schistosomal origin of the smrab clone. The
1.5 kb size of the identified mRNA suggests the presence of untranslated
regions in the message, as the ORF is only 611 bp long. The nucleotide
sequence presented in Figure 1 indicates a particularly long 3' untranslated
region, which is consistent with schistosome genomic structure [64].

Sequence analysis of smrab shows a high overall and L7-specific sequence

86

87

identity with rab-related proteins. It also indicates the maintenance of
characteristically-conserved GTP-binding and effector regions of ras-related
proteins. The ability of LMWGPs to bind and to hydrolyze GTP is necessary for
regulated protein activity, and the regions of the molecule involved in these
functions are very similar within the ras superfamily of GTP-binding proteins. The
GTP-binding analysis described here demonstrates that recombinant smrab
does indeed bind GTP, lending further credence to its description as a GTP-
binding protein.

Further support for the identity of smrab as a rab-related protein is found in
its CCXXX carboxy terminus. As discussed in the Introduction, ras-related
LMWGPs require the post-translational addition of an isoprenoid moiety in order
to function properly. The COOH-terminal motif of the proteins is considered
important in their identification and modification by the correct prenyltransferase
enzyme. Rab proteins are usually modified by geranylgeranyl protein
transferase (GGPT) type II; five different C-tenninal cysteine motifs have been
identified in these proteins [39]. CCXXX is a relatively unusual terminus,
identified in only 8% of rab-related proteins [39]. Evidence for the post-
translational modification of smrab with a geranylgeranyl group is presented by
the succesle incorporation of [3H]GGPP into in-vitro-translated smrab.

The smrab cDNA was subcloned into bacterial expression systems for the
purpose of generating enough protein to subject it to biochemical evaluation.

These efforts yielded stubbome-insoluble recombinant protein. Insolubility

88

appears to be a common complication of recombinant proteins expressed in
bacterial cells, and workers have employed numerous strategies, with varied
success, to decrease inclusion body formation during protein synthesis and to
denature and refold proteins following harvest from bacterial cells [4,82,105]. In
spite of exhaustive experimentation with these methods, the smrab protein
precipitated repeatedly. Nonetheless, biochemical functions characterizing
smrab as a rab-related GTP-binding protein were successfully demonstrated by
using the alternative techniques described.

In summary, smrab meets the criteria for classification as a member of the
ras-related superfamily, and may be considered a rab-related LMWGP.

Rab proteins constitute the largest branch of the ras superfamily of proteins
(more than 30 identified to date). They are involved in the regulation of
intracellular transport and secretion. Specifically, their participation is implicated
in three pathways: 1) transport of proteins through the endocytic pathway; 1)
transport from the endoplasmic reticulum through the Golgi stacks and recycling
within this pathway; and 3) secretion of vesicles from the cell membrane
[5,53,86,119]. Researchers have speculated on the function of newly-identified
proteins on the basis of their sequence similarity with rab proteins of known
function [90,96,114,146]. More specifically, the L7 loop region has been
implicated in confering functional specificity to rab-related proteins [35] (This
observation may have to be regarded with some care in its generalization, as

other experiments with highly-homologous rab proteins using genetic chimaera

88

appears to be a common complication of recombinant proteins expressed in
bacterial cells, and workers have employed numerous strategies, with varied
success, to decrease inclusion body formation during protein synthesis and to
denature and refold proteins following harvest from bacterial cells [4,82,105]. In
spite of exhaustive experimentation with these methods, the smrab protein
precipitated repeatedly. Nonetheless, biochemical functions characterizing
smrab as a rab-related GTP-binding protein were successfully demonstrated by
using the alternative techniques described.

In summary, smrab meets the criteria for classification as a member of the
ras-related superfamily, and may be considered a rab-related LMWGP.

Rab proteins constitute the largest branch of the ras superfamily of proteins
(more than 30 identified to date). They are involved in the regulation of
intracellular transport and secretion. Specifically, their participation is implicated
in three pathways: 1) transport of proteins through the endocytic pathway; 1)
transport from the endoplasmic reticulum through the Golgi stacks and recycling
within this pathway; and 3) secretion of vesicles from the cell membrane
[5,53,86,119]. Researchers have speculated on the function of newly-identified
proteins on the basis of their sequence similarity with rab proteins of known
function [90,96,114,146]. More specifically, the L7 loop region has been
implicated in confering functional specificity to rab-related proteins [35] (This
observation may have to be regarded with some care in its generalization, as

other experiments with highly-homologous rab proteins using genetic chimera

ass
labl
p2‘

[35

Sll

89

and complementation assays have failed to confirm the basis for such an
assumption [18,1141.) The nine amino acids of L7 constitute what has been
labeled the "specificity domain" and, on the basis of structural similarity with
p21”, are believed to interact with membranes or membrane-localized proteins
[35]. Experiments with chimeric genes have suggested that this region
cooperates with the effector region (loop L1) and a carboxy-terminal membrane
targeting signal to regulate the function of the particular rab protein [14,35]. The
L1 effector domain interacts with guanine nucleotide exchange factors (GEFs)
and other effector proteins and is required for regulation of GTP hydrolysis
[7,102].

Moore et al. [90] have grouped sequences of ras-related proteins by
similarity in the a-helix 3 - loop L7 region and maintain that this region is
conserved within rab subclasses, while differing between functionally-distinct
subclasses. Sequence homology within the L7 region of the rab subclasses in
this analysis range from 33-100% identity and 78-100% similarity (Dayhoff
grouping [28]). The L7 region of smrab is compared with that of its rab-related
homologues in Table 4. Identities and similarities within this region keep well
within the range of those in the subclassification analysis of Moore et al. [90].
On the basis of sequence homology of this functional specificity region, then,
smrab may function most like ora1 of the electric eel (Discopyge ommata) and
mammalian rab13. The function of neither of these proteins has been

ascertained, although the role of rab13 has been proposed in a post-Golgi,

90

exocytic activity similar to that of yeast SEC4 on the basis of localization studies
[36]. Smrab also shares a high overall and L7-specific homology with ypt1,
which is understood to function in vesicle transport between the endoplasmic
reticulum and the cis-Golgi compartment [117,122]. These two activities describe
two of the known functional categories for rab proteins, the third involving
endocytic recycling from the plasma membrane.

Polyclonal antibodies were developed to a carboxy-terminal sequence 14
amino acids in length and unique to smrab. Their specific recognition of smrab
was tested by Western blot, and the results sufficiently confounding that
subsequent immunohistochemical analysis with or-pep antibodies must be
considered with some trepidation. Anti-pep antibodies recognize recombinant
smrab on Western blot analysis, but identify a protein in the membrane fraction
of schistosome homogenate which migrates as though it were 10-11 kDa larger
than the recombinant. One explanation for this may be that ci-pep antibodies
recognize a protein unrelated to smrab which exhibits an epitope similar to the
14-mer peptide synthesized for the generation of oi-pep. A search of the entire
GenBankm-EMBL database [44] indicates that the 14-mer peptide shares at
most ca. 50% similarity with segments of various unrelated proteins from a range
of species. Consecutive, identical amino acids in homologous sequences occur
in groups of no more than 3 or 4. Antibodies are generated to epitopes which
are minimally 6-7 amino acids long, so for this reason it is unlikely that the

polyclonal antibody population constituting oi-pep lgG contains antibodies which

91

recognize a protein other than smrab. Peptides may be very different in amino
acid sequence, however, yet share three-dimensional characteristics which are
similar enough to elicit cross-recognition by the same antibody. If this were the
case here, one would expect to see at least two bands on the Western blot, one
of which is the same size as recombinant smrab. One might argue that the
concentration of smrab is so low in schistosome homogenate as to be invisible
on Western blot, leaving only the 35-kDa band. Northern blot analysis indicates
much the contrary, however, in its demonstration of abundant smrab mRNA in
male schistosomes especially. A remote possibility with which to explain the 35-
kDa band in the membrane fraction of schistosome homogenate may be that the
protein is actually smrab, but that its apparently very firm membrane association
slows its migration through the gel and causes it to appear larger than it is.
However, one would expect this membrane attachment to break as the sample is
boiled in 1% SDS prior to being loaded onto the gel. Furthermore, treatment of
schistosomes with mevinolin prior to homogenization should prevent
geranylgeranylation of the protein, and, consequently, should change its
distribution from the membrane to the cytoplasmic fraction. Incubation of worms
in 10‘5 M mevinolin for 48 hours, however, does not change the outcome of the
Western blot from the results presented in Figure 10 (data not shown). It should
be noted that the geranylgeranylated protein (Figure 9, lane 3) migrates at ca. 33
kDa, which is somewhat closer to the 35-kDa band representing endogenous

schistosome protein than that representing recombinant smrab on Western blot.

92

It is worthy of suggestion that the recombinant geranylgeranylated protein, which
was generated as a soluble protein in a eukaryotic expression system, holds a
configuration closer to the native than the urea-solubilized, bacterialIy-expressed
protein. The migration pattern of the endogenous protein may more realistically
be compared, therefore, to that of the geranylgeranylated protein in Figure 9
than to that of the recombinant smrab of lane 3 in Figure 10 or of lane 7 in Figure
7. Yet, compounding the evidence against the identity of the 35-kDa band as
smrab is the fact that the GTP-binding blot (Figure 8) does not identify a protein
of that size. The balance of the evidence, then, speaks against the
identification of the 35-kDa protein labeled in Figure 10 as endogenous,
schistosomal smrab. Rather, it suggests that or-pep recognizes something other
than a prenylated GTP-binding protein in schistosome homogenate. This is
rather troubling, as it casts doubt upon the legitimacy of interpretations of the
immunohistochemical findings reported here. Identification of the enigmatic 35-
kDa protein may help to sort out the situation; this may be done by subjecting the
protein on the membrane to N-terrninal amino acid sequencing, although the
results of this procedure are likely to be confounded by impurities in the protein
sample obtained from blotted schistosome homogenate.

There remains, nonetheless, the feeble, though well-founded (albeit
unofficially, as negative results are generally not reported) argument that results
of Western blot experiments often do not match those of in-situ

immunohistochemical experiments with the same antibodies. This, together with

93

the fact that a-pep identifies only one band per lane on the Western blot of
Figure 10, allows some room for assertion that a-pep does indeed recognize an
endogenous rab-related protein on schistosome sections. Immunohistochemical
experiments were thus pursued using a-pep, keeping in mind that specificity of
this antibody preparation is not entirely certain. The authenticity of the labeling
pattern observed with a-pep may be verified with in-situ hybridization of labeled
smrab cDNA to RNA. Other procedures for ascertaining the specificity of the
antibody preparation may include immunoprecipitation of radiolabeled proteins
from schistosome homogenate, or elution of such proteins off of a column of
immobilized antibody. (The former procedure was attempted for this work but
failed to provide conclusive results.)

Anti-pep antibodies also exhibit some cross-reactivity with recombinant rab
proteins other than smrab, as evidenced by their recognition of H-rab5 (but not
rab1b) on Western blot. (Interestingly, H-rab5 does not share any sequence
homology with the 14-mer peptide used to generate oi-pep antibodies.)
Therefore, a-pep labeling on schistosome sections may represent more rab-
related proteins than just smrab. If there is such cross-reactivity, however, it
appears to be confined to the same subtegumental location, providing further
evidence to support the suggestion that the a-pep antibodies are labeling
proteins confined to the tegumental cytons. (It should be noted that vitellariae in
female sections label strongly with both negative control lgG and or-pep lgG, so

or-pep may be binding something there as well, but its recognition is obscured by

94

nonspecific binding.) Future work with confocal and transmission electron
microscopy will allow more specific identification of the structures labeled with a-
pep.

lmmunohistochemical characterization of sectioned, paraffin-embedded
worms with antibodies to a carboxy-terminal epitope of smrab (c-pep) localizes
the endogenous protein to subtegumental structures in the schistosome. In
these the exact identity of the structures labeled with the a-pep antibodies is not
discernible. Considering the anatomy of the schistosome, however, and the role
of rab-related proteins in protein synthesis and vesicle transport, the proteins
identified by the antibodies are most likely found in the tegumental cytons. As
described earlier, these are the only nucleated cells near the tegument in the
schistosome, and are highly active in supplying the surface of the worm with
proteins and vesicles necessary for its viability and antigenic variation. The rapid
turnover of the tegument membranocalyx and its associated vesicles [143,144]
implies the requirement for rapid and copious protein synthesis, which in turn
suggests the presence and activity of rab proteins.

By definition, smrab most probably works within one of the three categories
of intracellular activity described earlier. In the schistosome tegument this
translates into a variety of possibilities. In the tegumental cyton, the protein may
regulate the processing of newly-translated proteins, which are, in turn,
transported to the tegument for replenishment of the tegumental syncytium or for

variation of the antigenic face of the schistosome surface. In an endocytic

95

capacity, smrab may be important in restructuring the tegument and in taking up
molecules from the environment. This latter activity may serve purposes of
nourishment, environmental surveillance, and communication between worms
(discussed below). Exocytic roles for smrab may involve escorting vesicles to
the plasma membrane of the tegumental cyton or through the cytoplasmic
connections for release of their contents into the tegumental syncytium or onto
the surface of the worm.

The specific function of smrab may be characterized from the results of a
variety of experiments. To start with, a great deal can be learned from careful
localization of the protein. In order to do this, a specific and accurate method
must first be developed to identify endogenous smrab. A polyclonal antibody
may certainly serve this purpose, although its specificity may be difficult to prove,
as shown by the work presented here. In-situ hybridization of labeled smrab
cDNA to schistosome RNA may also prove informative. Ideally, to study the
subcellular localization and function of smrab, one would first over-express the
protein in yeast or mammalian (e.g., CHO or Swiss 3T3) cells. Although this
would remove the protein from the schistosome environment, these cells are, like
those of the schistosome, eukaryotic, and may be reasonably expected to
maintain smrab in its subcellular compartment. Over-expressing smrab in such a
system makes the protein amenable to a variety of experimental techniques.
The protein may be expressed with an epitope tag such as Influenza virus

Hemagglutinin (HA) or F luorescein isothiocyanate (FITC) [18, 133], and a

96

commercially-available antibody to the tag used to mark and/or to block the
activity of the protein. Localization should be done at light, confocal, and
electron microscopic levels.

Once smrab can be reliably identified, colocalization studies may be done for
more accurate subcellular localization of the protein. NBD-ceramide, for
instance, is a routinely-used lipid stain for the Golgi apparatus; antibodies to the
B-COP protein also mark Golgi stuctures [18]. Transferrin is an intrinsic marker
for the early endocytic pathway in yeast and CHO cells [134]. Confocal
microscopic techniques are especially useful for dissection and fusion of three-
dimensional images of cells subjected to such colocalization studies. As an
extension of this work at a molecular level, cross-linking experiments [71,85] and
the two-hybrid system [20] may be used to identify some of the structures with
which smrab associate. The two-hybrid system offers the advantage of obtaining
a cloned gene of the interacting protein (which, on the other hand, may present
an enigma of its own); its drawback is that it identifies only protein effectors and
would overlook, for instance, vesicular structures.

Such localization and colocalization studies would allow placement of smrab
into a subcellular category; i.e., in which of the three categories of rab activity it is
most likely involved.

Preliminary studies of smrab function are also best undertaken in eukaryotic
single-cell systems amenable to controlled molecular and biochemical

manipulation. In a eukaryotic expression system, functional mutants may be

97

generated expressing defective smrab. The protein may cany a mutation in its
DNA structure or may be functionally arrested in a GDP- or GTP-bound state.
The effects of these defects on the cell’s protein synthetic, endocytic, and
exocytic activities can then be determined with assays evaluating these
pathways [77,84, 97, 130,134]. These in-vitro experiments would identify the
cellular systems on which to focus when the work is taken to the schistosome in
vivo.

To characterize the activities of smrab in the schistosome, again, selective
markers and inhibitors for smrab must be created. The antibodies used in initial
localization experiments would certainly be useful in following changes in the
distribution of smrab as its activities are hindered. From the in-vitro studies
described above, the subcellular system in which smrab moves would be known;
with experiments approaching the question from the other end, this system can
be inhibited (e.g., by blocking export from the Golgi apparatus with Brefeldin A)
and the effect of this inhibition on the distribution of smrab observed. The
protein may, for instance, aggregate in the Golgi structures or around vesicles;
endo- or exocytic vesicles colocalized with smrab may accumulate; movement of
smrab with vesicles through cytoplasmic connections may be inhibited. Using
the in-situ hybridization technique, changes in smrab mRNA levels may also be
evaluated, providing clues toward the mechanisms regulating expression of the
protein. These experiments would legitimize the extrapolation of results from in-

vitro experiments in single-cell systems to the whole schistosome.

98

Subsequently, antisense RNA or peptide inhibitors targeting the effector
domains of smrab may selectively inhibit the expression and activity,
respectively, of the protein in the living worm. Preliminary work to develop these
inhibitors is again best done in eukaryotic, single-cell systems. These
techniques have been successfully employed to characterize functions of various
rab proteins [26.76.102.108], albeit in systems much simpler than that of an
intact schistosome. With a peptide inhibitor, again, changes in distribution of
endogenous smrab may provide important information regarding its activities in
the schistosome. Functional assays reflecting the results of blocking the
expression and/or activity of smrab must, however, also be performed. The
premise for the experiments discussed here is the assumption that smrab works
in the capacity of the schistosome tegument. Therefore, the effects of smrab
inhibition on tegumental turnover must be evaluated, and may be done using
methods already described by others [72.109.143.144]. Morphologic changes in
the tegument observed with the electron microscope may also prove informative
[10,142,143]. Significant effects on membrane structure and on turnover of
tegumental proteins might suggest the involvement of smrab in maintenance of
the tegument structure and/or in antigenic variation of its surface. Effects on
these two systems might be discerened from one another with methods
evaluating changes in the release of antigens from the schistosome surface [73].

The proteins identified by the ci-pep antibodies appear to be more abundant

in the subtegument of male schistosomes than in females. This observation is

99

consistent with the results of the Northern blot. which indicate that, although the
smrab gene is constituitively expressed in both sexes of schistosomes, it is done
so to afar greater extent in males than in females. While the general structure of
the tegument is similar in male and female schistosomes [148], the female is
significantly protected from the environment by her position in the gynecophoric
canal of the male. This arrangement suggests that a disproportionate burden of
antigenic defense is placed on the male. If smrab is involved in tegumental
maintenance as proposed, the difference in immunohistochemical findings
between male and female worms may be considered in the context of a
presumably greater tegumental activity in the male schistosome.

In some sections of male worms or-pep labeling appears more vivid on the
ventral side of the worm, e.g., that adjacent to the female (Figures 110 and 150).
Confirmation of this observation requres further careful investigation; if they are
accurate, however. they allow interesting speculation on a potiential role of
smrab in the communication between the male and female of a worm pair. Work
by numerous investigators [21.38.93.137] established that development and
sexual maturation of the female schistosome depends on her continued physical
contact with the male. Investigators have been unable. however, to identify the
nature of communication between worms of a pair. Most efforts towards this
goal have focused on the identification of a protein messenger passing from the
male to the female [103]. It is just as likely, though, that the signal is a non-

protein molecule or even an electron current. On the reaching speculation that

100

smrab is involved in this male-female communication, one might propose its role
in a number of capacities. In the male schistosome it may regulate the synthesis
of the molecular signal and its processing through the Golgi stack; it may aid in
its packaging into vesicles and their escort to the surface of the tegument; it may
regulate the release of the vesicular contents; it may work in a feedback network
regulating synthesis and release of the molecule(s) via an endocytic function.
Alternatively, on the female side. smrab may regulate the endocytic uptake of the
signalling molecule(s) through the tegument. Definitive evidence for the
involvement of smrab in the communication between male and female
schistosomes would involve detailed experiments of the sort discussed above, in
which the activity of smrab is selectively inhibited, and the effect of this inhibition
on female development, maturation, and fecundity with and without the presence
of the male characterized. This information. combined with results of in-vitro
experiments determining the function of smrab on a subcellular level, would then
provide important clues towards identification of the nature of the language
exchanged between male and female schistosomes.

The immunohistochemical findings in schistosomes which had been treated
with mevinolin are fascinating. Mevinolin inhibits the synthesis of the isoprenoids
FPP and GGPP by inhibition of the rate-limiting enzyme of the mevalonate
pathway, HMG-CoA reductase. Work in our laboratory has indicated that
prenyltransferase activity in schistosome homogenate increases with the addition

of mevinolin, suggesting that the limitation of lipid substrate increases the

101

synthesis of protein substrate for the enzymes (J.L. Bennett. unpublished
observations). Additionally, preventing the isoprenylation of smrab would be
expected to release the protein from its lipid-anchored site into the cytoplasmic
fraction, thereby changing the labeling distribution on immunohistochemical
analysis. Observations consistent with both of these points are observed in the
comparison of sections from worms which had undergone in-vitro incubation with
and without mevinolin. Incubation of worms without mevinolin appears to
decrease the amount of protein recognized by a-pep (especially in females), and
also results in a somewhat irregular distribution of the protein. This observation
associates 60-hour in-vitro incubation of schistosomes with changes in the
biochemistry of the protein(s) recognized by or-pep. Others have reported similar
findings in the description of morphologic changes which schistosomes undergo
in vitro, especially in the tegumental structures, and presumably in adaptation to
their new and antigentically less-demanding environment [10]. Addition of
mevinolin to the incubation medium markedly increases antibody label over that
in both fresh and incubated, untreated worms (especially in males). Fluorescent
label in sections from treated worms is denser and less linearly distributed than
in sections from control worms, suggesting both increased target protein and an
altered localization thereof. Therefore, mevinolin treatment appears to stimulate
synthesis of the smrab protein (albeit indirectly). As distribution of the protein is
altered in the incubated, untreated worms. it is difficult to say from these sections

whether mevinolin enhances this effect. Northern blot analysis with RNA

102

isolated from fresh worms. from incubated, untreated worms, or from mevinolin-
treated worms may help to determine more precisely how expression of smrab
changes under these conditions and how much of any change may be ascribed
to the effects of mevinolin. This may also help to discern whether the effects
observed with immunohistochemistry are due more to changes in protein
synthesis in response to mevinolin treatment or to altered distribution of the
protein. Furthermore. mevinolin-treated worms may be subjected to localization
and colocalization experiments as described above. Results of these
experiments would provide information on the effects of inhibition of prenylation
on smrab activity. They may also be evaluated in the context of smrab-inhibition
studies, such as those described earlier with antisense RNA and peptide
inhibitors.

The described changes in antibody labeling with incubation and mevinolin
treatment appear in both male and female schistosomes. Sections from
females, however, show consistently less label than those from males. This
difference is especially prominent in the response of worms to mevinolin
treatment, which may suggest a slower rate of smrab turnover (synthesis) in
females compared to that in males. Other investigators have noted differences
between male and female schistosomes in the effect of various drugs on the
tegument [12,140]. They suggested that this differential response may be
attributable to specific cellular differences in the tegument between male and

female worms, which in turn again supports the conclusions drawn from the

103

immunohistochemical observations described here. These observations have
yet to be verified with more reliable localization experiments, as well as with
Northern blot and in-situ hybridization analysis. If they are accurate. however,
they provide important clues towards our understanding of the roles of
isoprenylation and the smrab protein in S. mansoni and in the adaptive
strategies of this parasite, and of the differences in these roles between the two
sexes.

The context of the work in which the search for a LMWGP was initiated, and
which resulted in the cloning of smrab, was the investigation of the mechanisms
regulating egg production in the female schistosome. The identification of the
novel schistosome protein described here has guided our attention instead to the
mechanisms regulating the diverse activities of the highly-dynamic tegument in
this parasite. As mentioned previously, it should be considered that the
development and maintenance of sexual maturity in the female schistosome
depends on the physical proximity of the male [93]. Therefore, any contribution
by smrab towards the viability of the male schistosome presumably affects
female fecundity indirectly. Demonstration of such a link would nicely
complement our original interest in schistosome egg production.

Future studies such as those described in these last several pages should
provide considerable insight towards understanding the function of smrab in
Schistosoma mansoni. Characterization of smrab and other rab-related proteins

in the schistosome promise to further our understanding of the molecular

104

mechanisms regulating such processes as fecundity and antigenicity in this
parasite, and will prove invaluable toward the ultimate goal of controlling the

disease caused by the parasite.

BIBLIOGRAPHY

10.

BIBLIOGRAPHY

Balch, W.E. (1990) Small GTP-binding proteins in vesicular transport.
TIBS. 15, 473-477.

Barbacid, M. (1987) ras genes. Annu. Rev. Biochem. 56, 779-817.

Bischoff, PR. and Postingl, H. (1991) Mitotic regulator protein RCC1 is
complexed with a nuclear ras-related polypeptide. Proc. Natl. Acad. Sci.
USA. 88, 10830-10834.

Blackwell, JR. and Horgan, R. (1991) A novel strategy for production of a
highly expressed recombinant protein in an active form. F E88. 195, 10-
1 1.

Bokoch, GM. and Der, C.J. (1993) Emerging concepts in the R33
superfamily of GTP-binding proteins. FASEB J. 7, 750-759.

Boman. AL. and Kahn, RA. (1995) Art proteins: the membrane traffic
police? TIBS. 10, 147-150.

Bourne. H.R.. Sanders, DA. and McCormick. F. (1991) The GTPase
superfamily: conserved structure and molecular mechanism. Nature. 349.
1 17-1 16.

Brown, MS. and Goldstein, J.L. (1980) Multivalent feedback regulation of
HMG CoA reductase, a conotrol mechanism coordinating isoprenoid
synthesis and cell growth. J. Lipid Res. 11, 505-517.

Cappai, R., Osborn, A.H., Gleeson, PA, and Handman, E. (1993)
Cloning and characterization of a Golgi-associated GTP-binding protein
homologue from Leishmania major. Mol. Biochem. Parasitol. 61, 73-81.

Carlisle, 8.. Weisberg, LS. and Bently, AG. (1983) Schistosoma
mansoni: morphologic changes induced by maintenance in vitro. J.

105

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

1 06
Parasitol. 69, 319-334.

Casey, P.J., Moomaw. J.F., Zhang, F.L., Higgins, J.B. and Thissen, J.A.
(1994) Prenylation and G protein signaling. Recent Prog. Horm. Res. 49,
1 15-138. ’

Chappell, L.H. (1987) The interaction between drugs and parasites.
Parasitology. 96, $167-$193.

Chavrier, P., Vingron, M., Sander, C., Simons. K., and Zerial, M. (1990)
Molecular cloning of YPT1/SEC4-related cDNAs from an epithelial cell
line. Mol. Cell Biol. 10, 6578-6585.

Chavrier, P., Gorvel, J-P, Stelzer, E., Simons. K.. Gruenberg, J. and
Zerial, M. (1991) Hypervariable C-terrninal domain of rab proteins acts as
a targeting signal. Nature. 353, 769-771.

Chen, 6., Foster. LA. and Bennett, J.L. (1990) Antischistosomal action of
mevinolin: evidence that 3-hydroxy-glutaryl-coenzyme a reductase activity
in Schistosoma mansoni is vital for parasite survival. Naunyn-
Schmiedeberg's Arch. Pharrnacol. 341, 477-481.

Chen, G. and Bennett, J.L. (1993) Characterization of mevalonate-labeled
lipids isolated from parasite proteins in Schistosoma mansoni. Mol.
Biochem. Parasitol. 59, 187-191 .

Chen, L.M.. Chern, Y., Orig, SJ. and Tai, J.H. (1994) Molecular cloning
and characterization of a ras-related ran/tc4spi1 subfamily in Giardia
Iamblia. J. Biol. Chem. 169, 17197-17304.

Chen, Y-T., Holcomb, C and Moore, H-P.H. (1993) Expression and
localization of two low molecular weight GTP-binding proteins, rab8 and
rab10. by epitope tag. Proc. Natl. Acad. Sci. USA. 90, 6508-6511.

Chien, C.-T., Bartel, P.L., Stemglanz, R. and Fields. S. (1991) The two-
hybrid system: a method to identify and clone genes for proteins that
interact with a protein of interest. Proc. Natl. Acad. Sci. USA. 88, 9578-
9581.

Clough. ER. (1981) Morphology of reproductive organs and oogenesis in
bisexual and unisexual transplants of mature Schistosoma mansoni
females. J. Parasitol. 67, 535-539.

Coulter, L.J. and Hide, G. (1995) Trypanosoma brucei: characterization of

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

107

a life cycle stage-specific G-protein. Exp. Parasit. 80, 308-318.

Cox, A.D., Hisaka, M.M., Buss, J.E. and Der, C.J. (1991) Specific
isoprenoid modification is required for function of normal, but not
oncogenic, ras protein. Mol. Cell Biol. 11, 1606-1615.

Cox, AD. (1995) Mutation and analysis of prenylation signal sequences.
Methods in Enzymology. 150, 105-1 1 1.

Culine, S., Rousseau-Merck, M-F., Honoré, N., Nezelof, C. and Olofsson,
B. (1991) Specific expression of the ras-related rab3a gene in human
normal and malignant neuroendocrine cells. Cancer. 70, 1551-1556.

Dagher, M.C.. Fuchs, A., Boun'neyster, N., Jouan, A. and \fignais, P.V.
(1995) Small proteins and the neutrophil NADPH oxidase. Blochemie. 77,
7-8.

Davidson, J.S., Eales, A., Roeske, R.W. and Millar, RP. (1993) Inhibition
of pituitary hormone exocytosis by a synthetic peptide related to the rab
effector domain. F E88. 316. 119-111.

Davis, R.E., Davis. A.H., Carroll, S.M., Rajkovic, A. and Rottman, FM.
(1988) Tandemly repeated exons encode 81-base repeats in multiple,

developmentaly regulated Schistosoma mansoni transcripts. Mol. Cell
Biol. 8, 4745-4755.

Dayhoff, M.O., Schwartz, RM. and Orcutt, BC. (1978) A model of
evolutionary change in proteins. In: Atlas of Protein Sequence and
Structure (Dayhoff, M.O.. ed.). Vol 5, suppl. 3, pp. 345-351.

Desolms, S.J., Deanna, A.A.. Giuliani, E.A., Graham, S.L.. Kohl, N.E.,
Mosser, S.D., Oliff, A.l., Pompliano, D.L., Rands, E., Scholz, T.H.,
Wiggins, J.M., Gibbs, J.B. and Smith, R.L. (1995) Pseudopeptide
inhibitors of protein famesyltransferase. J. Med. Chem. 38, 3967-3971.

deVos, A.M.. Tong, L., Milbum, M.V., Matias, P.M., Jancarik, J., Noguchi,
S., Nishimura, S., Miura, K., Ohtsuka. E., and Kim, S-H. (1988) Three-
dimensional structure of an oncogene protein: catalytic domain of human
c-H-ras p11. Science. 139. 888-893.

Diekmann, D.. Abo. A., Johnston. 0., Segal. AW. and Hall, A. (1994)
Interaction of rac with p67phox and regulation of phagocytic NADPH
oxidase activity. Science. 165, 531-533.

32.

33.

35.

36.

37.

38.

39.

40

41.

42.

108

dos Reis, M.G., Davis, R.E., Hardeep, S.. Skelly, P.J. and Shoemaker,
CB. (1993) Characterization of the Schistosoma mansoni gene encoding

the glycolytic enzyme, triosephosphate isomerase. Mol. Biochem.
Parasitol. 59, 155-141.

Downward, J. (1991) Regulatory mechanisms for ras proteins. BioEssays.
14, 177-184.

Drivas, G.T.. Shih, A., Coutavas, E., Rush, MG. and D'Eustachio, P.
(1990) Characterization of four novel ras-like genes expressed in a human
teratocarcinoma cell line. Mol. Cell. Biol. 10, 1793-1798.

Dunn, B.. Steams. T. and Botstein, D. (1993) Specificity domains
distinguish the ras-related GTPases th1 and Sec4. Nature. 361, 663-
665.

Egan, SE. and Weinberg, RA. (1993) The pathway to signal
achievement. Nature. 365, 781-783.

Erasmus, DA. (1973) A comparative study of the reproductive system of
mature, immature, and unisexual females of Schistosoma mansoni.
Parasitology. 67. 165-183.

Famsworth, C.C., Kawata, M., Yoshida. Y., Takai, Y., Gelb, M.H. and
Glomset, J.A. (1991) C terminus of the small GTP-binding protein smg
p15A contains two geranylgeranylated cysteine residues and a methyl
ester. Proc. Natl. Acad. Sci. USA. 88, 6196-6100.

Famsworth, C.C., Seabra, M.C.. Ericsson, L.H., Gelb, M.H and Glomset,
J.A. (1994) Rab geranylgeranyl transferase catalyzes the
geranylgeranylation of adjacent cysteins in the small GTPases Rab1A,
Rab3A and Rab5A. Proc. Natl. Acad. Sci USA. 91 , 1 1963-1 1967.

Field. M.C., Field, H. and Boothroyd, J.C. (1995) A homologue of the
nuclear GTPase Ran/T C4 from Trypanosoma bmcei. Mol. Biochem.
Parasit. 69, 131-134.

Fritz, G., Kaina, B. and Aktories. K. (1995) The ras-related small GTP-
binding protein rhoB is immediate-early inductible by DNA damaging
treatments. J. Biol. Chem. 170, 15171-15177.

Gaboriaud, C., Bissery, V., Benchetrit, T., Momon, JP. (1987)
Hydrophobic cluster analysis: an efficient new way to compare and
analyze amino acid sequences. FEBS lett. 114, 149-155.

43.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

109

Gallwitz. D., Donath, C. and Sander, C. (1983) A yeast gene encoding a
protein homologous to the human c-haslbas proto-oncogene product.
Nature. 306, 704-707.

Genetics Computer Group Sequence Analysis Software Package Version
7.0. Devereux, Haeberli, Smithies. (1991) A comprehensive set of
sequence analysis programs for the VAX. Nucl. Acids Res. 11(1), 387-
395.

Gibbs, J.B., Oliff, A and Kohl, NE. (1994) Famesyltransferase inhibitors:
ras research yields a potential cancer therapeutic. Cell. 77, 175-178.

Gibbs, J.B., Pompliano, D.L., Mosser, S.D., Rands, E., Lingham, R.B.,
Singh, S.B., Scolnick, E.M., Kohl, NE. and Oliff, A. (1993) Selective
inhibition of famesyl-protein transferase blocks ras processing in vivo. J.
Biol. Chem. 168(11), 7617-7610.

Glomset, J.A., Gelb, M.H. and Famsworth, CC. (1990) Prenyl proteins in
eukaryotic cells: a new type of membrane anchor. TI BS. 15, 139-141.

Glomset, J.A., Gelb, M.H. and Famsworth, CC. (1991)
Geranylgeranylated proteins. Biochem. Soc. Trans. 10, 479-484.

Goldstein, J.L. and Brown. MS. (1990) Regulation of the mevalonate
pathway. Nature. 343,415-430.

Goldstein, J.L., Brown, M.S., Stradley, S.J., Reiss. Y. and Gierasch, L.M.
(1991) Nonfamesylated tetrapeptide inhibitors of protein
famesyltransferase. J. Biol. Chem. 166. 15575-15578.

Gorvel, J-P., Chavrier, P., Zerial, M. and Gruenberg, J. (1991) rab5
controls early endosome fusion in vitro. Cell. 64, 915-915.

Goud, B.. Salminen, A., Walworth, NC. and Novick, P.J. (1988) A GTP-
binding protein required for secretion rapidly associates with secretory
vesicles and the plasma membrane in yeast. Cell. 53, 753-768.

Hall, A. (1990) The cellular functions of small GTP-binding proteins.
Science. 149, 635-640.

Hancock. J.F., Magee, A.l.. Childs, J.E. and Marshall, C.J. (1989) All ras
proteins are polyisoprenylated but only some are palmitoylated. Cell. 57,
1 167-1 177.

55.

56.

57.

58.

59.

60.

61.

62.

63.

65.

110

Harlow, E. and Lane. 0., eds. (1988) Antibodies, a Laboratory Manual.
Cold Spring Harbor Laboratory, New York, NY.

Haubruck, H., Disela, C., Wagner, P. and Gallwitz, D. (1987) The ras-
related ypt1 protein is an ubiquitous eukaryotic protein: isolation and
sequence analysis of mouse cDNA clones highly homologous to the yeast
YPT1 gene. EMBO J. 6, 4049-4053.

Hockley, D.J. and McLaren, D.J. (1973) Schistosoma mansoni: changes
in the outer membrane of the tegument during development from cercaria
to adult worm. lnternatl. J. Parasit. 3, 13-15.

Hoffenberg, S., Sanford, J.C., Liu, 8.. Daniel, D.S., Tuvin, M., Knoll, B.J.,
Wessling-Resnick, M.. and Dickey, B.F. (1995) Biochemical and functional
characterization of a recombinant GTPase, rab5. and two of its mutants.
J. Biol. Chem. 170, 5048-5056.

James, G.L., Goldstein, J.L., Brown, M.S., Rawson. T.E., Somers, T.C.,
McDowell, R.S., Crowley, C.W., Lucas, B.K., Levinson, AD. and Marsters.
J.C. (1993) Benzodiazepine peptidomimetics: potent inhibitors of ras
famesylation in animal cells. Science. 160. 1937-1941.

Jedd, G., Richardson, 0., Litt, R. and Segev, N. (1995) The ypt1 GTPase
is essential for the first two steps of the yeast secretory pathway. J. Cell
Biol. 131, 583-590.

Kahn, R.A.. Der, OJ. and Bokoch, GM. (1991) The ras superfamily of
GTP-binding proteins: guidelines on nomenclature. FASEB J. 6, 1511-
1513.

Kahn, RA. and Gilman, AG. (1986) The protein cofactor necessary for
ADP-ribosylation of Gs by cholera toxin is itself a GTP-binding protein. J.
Biol. Chem. 161. 7906-7911.

Kahn, R.A.. Kern, F.G., Clark, J., Gelmann, ER and Rulka, C. (1991)
Human ADP-ribosylation factors, a functionally conserved family of GTP-
binding proteins. J. Biol. Chem. 166, 1606-1614.

Khalife, J., Trottein, F.. Schacht, A.M., Godin, 0., Pierce, R.J. and Capron,
A. (1993) Cloning of the gene encoding a Schistosoma mansoni antigen
homologous to human RolSS-A autoantigen. Mol. Biochem. Parasitol. 57,
193-101.

Khosravi-Far, R., Clark, G.J., Abe, K., Cox. A.D.. McLaine, T., Lutz, R.J..

66.

67.

68.

69.

70.

71.

72.

73.

74.

111

Sinensky. M. and Der. C.J. (1991) Ras (CXXX) and Rab (CC/CXC)
prenylation signal sequences are unique and functionally distinct. J. Biol.
Chem. 167, 14363-14368.

Khosravi-Far. R., Lutz, R.J., Cox. A.D., Conroy, L.. Boume, J.R.,
Sinensky. M., Balch, W.E., Buss, J.E. and Der, C.J. (1991) Isoprenoid
modification of rab proteins terminating in CC or CXC motifs. Proc. Natl.
Acad. Sci. USA. 88, 6164-6168.

Kim, S.-H., de Vos, A.M., Tong, L.. Milbum, M.V., Matias, P.M.. Jancarik,
J., Ohtsuka. E. and Nishimura, S. (1988) ras Oncogene proteins: three-
dimensional structures, functional implications, and a model for signal
transducer. Cold Spring Harbor Symposium on Quatitative Biology. 53
pt.1. 173-81.

Kinsella, B.T., Erdman, RA. and Maltese, W.A. (1991) Posttranslational
modification of Ha-ras p11 by famesyl versus geranylgeranyl isoprenoids
is determined by the COOH- terminal amino acid. Proc. Natl. Acad. Sci.
USA. 88, 8934-8938.

Kinsella, B.T. and Maltese. W.A. (1991) rab GTP-binding proteins
implicated in vesicular transport are insoprenylated in-vitro at cysteines
within a novel carboxy-terminal motif. J. Biol. Chem. 166, 8540-8544.

Kinsella, B.T and Maltese, W.A. (1991) Rab GTP-binding proteins with
three different carboxyl-terrninal cysteine motifs are modified in-vivo by
10-carbon isoprenoids. J. Biol. Chem. 167(6), 3940-3945.

Kurzchalia, T.V., Gorvel, J.-P., Dupree. P., Paron, R., Kellner, R.,
Houthaeve, T., Gruenberg, J. and Simons. K. (1991) Interactions of rab5
with cytosolic proteins. J. Biol. Chem. 167, 18419-18413.

Kusel, JR. and Mackenzie. PE. (1975) The measurement of the relative
turnover rates of proteins of the surface membranes and other fractions of
Schistosoma mansoni in culture. Parasitology. 71, 161-173.

Kusel, J.R., Mackenzie, PE. and McLaren, D.J. (1975) The release of
membrane antigens into culture by adult Schistosoma mansoni.
Parasitology. 71, 147-159.

Lambeth, J.D., Kwak, J-Y., Bowman, E.P., Perry, 0., Uhlinger, DJ and
Lopez. I. (1995) ADP-ribosylation factor functions synergistically with a
50-kDa cytosolic factor in cell-free activation of human neutrophil
phospholipase D. J. Biol. Chem. 170, 1431-1434.

75.

76.

77.

78.

79.

80.

81.

82.

83.

85.

86.

112

Lledo, P-M., Johannes, L., Vernier. P., Zorec, R., Darchen, F., Vincent, J-
D.. Henry, J-P. and Mason, W. (1994) Rab3 proteins: key players in the
control of exocytosis. TINS 17, 416-431.

Lledo, P-M.. Vernier. P., Vincent. H-D., Mason, WT and Zorec, R. (1993)
Inhibition of Rab3b expression attenuates Ca1+-dependent exocytosis in
rat anterior pituitary cells. Nature. 364, 540-544.

Lombardi, D., Soldati, T., Riederer. M.A., Goda, Y., Zerial, M. and Pfeffer,
SR. (1993) Rab9 functions in transport between late endosomes and the
trans Golgi network. EMBO J. 11, 677-681.

Maltese, W.A. (1990) Posttranslational modification of proteins by
isoprenoids in mammalian cells. FASEB J. 4, 3319-3318.

Maltese, W.A. and Sheridan, KM. (1987) lsoprenylated proteins in
cultured cells: subcellular distribution and changes related to altered
morphology and growth arrest induced by mevalonate deprivation. J. Cell
Physiol. 133, 471-481.

Marshall, C.J. (1993) Protein prenylation: a mediator of protein-protein
interactions. Science. 159, 1865-1866.

Marshall, MS. (1993) The effector interactions of p11ras. TIBS. 18, 150-
154.

Marston, F AG. and Hartley. D.L. (1990) Solubilization of protein
aggregates. In: Guide to Protein Purification (Deutscher, M.P.. ed.). pp.
164-176.

Marston, F.A.O., Lowe, P.A., Doel. M.T., Schoemaker, J.M., White, S. and
Angal. S. (1984) Purification of calf prochymosin (prorennin) synthesized
in Escherichia coli. Bio/Technology. 1. 800-804.

Martinez, 0., Schmidt, A., Salaméro, J., Hoflack, B., Roa, M. and Goud,
B. (1994) The small GTP-binding protein rab6 function in intra-Golgi
transport. J. Cell Biol. 1 17, 1575-1588.

McKieman, C.J., Brodyk. W.H. and Macara, LG. (1993) The rab3a
GTPase interacts with multiple factors thorugh the same effector domain.
J. Biol. Chem. 168, 14449-14451.

Melancon, P., Glick, B.S.. Malhotra, V., Weidman. P.J.. Serafini. T.,
Gleason, M.L., Orci, L. and Rothman, J.E. (1987) Involvement of GTP-

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

113

binding "G" proteins in transport through the Golgi stack. Cell. 51, 1053-
1061.

Meyer, F., Meyer, H., Beuding, E. (1970) Lipid metabolism in the parasitic
and free-living flatworms Schistosoma mansoni and Dugesia
domtocephala. Biochim. Biophys. Acta. 1 10, 157-166.

Milbum, M.V., Tong, L., deVos, A.M., Bri'Jnger, A., Yamaizumi, Z.,
Nishimura, S. and Kim, S-H. (1990) Molecular switch for signal
transduction: structural differences between active and inactive froms of
protooncogenic ras proteins. Science. 147, 939-945.

Moodie, S.A., Willumsen, B.M., Weber, M.J. and Wolfman, A. (1993)
Complexes of ras-GTP with raf-1 and mitogen-activated protein kinase
kinase. Science. 160, 1658-1661.

Moore, l., Schell, J. and Palme, K. (1995) Subclass-specific sequence
motifs identified in rab GTPases. TIBS. 10, 10-11.

Moore, MS. and Blobel. G. (1993) The GTP-binding protein Ran/T C4 is
required for protein import into the nucleus. Nature. 365. 661-663.

Moore, MS. and Blobel, G. (1994) A G protein involved in
nucleocytoplasmic transport: the role of ran. TIBS. 19, 111-116.

Moore, D.V., Yolles, T. K. and Meleney. HE. (1954) The relationship of
male worms to the sexual development of female Schistosoma mansoni.
J. Parasitol. 40. 166-185.

Moroianu. J. and Blobel, G. (1995) Protein export from the nucleus
requires the GTPase ran and GTP hydrolysis. Proc. Natl. Acad. Sci. USA.
91, 4318-4311.

Morris, GP. and Threadgold, LT. (1968) Ultrastructure of the tegument of
adult Schistosoma mansoni. J. Parasit. 54, 15-17.

Ngsee, J.K., Elferink, LA. and Scheller, RH. (1991) A family of ras-like
GTP-binding proteins expressed in electromotor neurons. J. Biol. Chem.
166, 1675-1680.

Oberhauser, A.F., Monck. J.R., Balch, W.E. and Femadez, J.M. (1991)
Exocytic fusion is activated by rab3a peptides. Nature. 360, 170-173.

O'Donnell, M.P., Kasiske, B.L., Kim. Y., Atluru, D. and Keane, W.F. (1993)

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

114

Lovastatin inhibits proliferation of rat mesangial cells. J. Clin. Invest. 91,
83-87.

OLIGO 4.1 Primer Analysis Software. Copyright 1989-1991 Wojciech
Rychlik. Published by National Bioscience, |nc.. Plymouth, MN.

Osamu, O., Dorit, R.L. and Gilbert, W. (1989) One-sided polymerase
chain reaction: the amplification of cDNA. Proc. Natl. Acad. Sci, USA. 86,
5673-5677.

Plutner, H., Cox, A.D.. Pind, S., Khosravi-Far. R.. Boume, J.R.,
Schwaninger. R., Der, OJ. and Balch, W.E. (1991) Rab1 b regulates
vesicular transport between the endoplasmic reticulum and successive
Golgi compartments. J. Cell Biol. 1 15, 31-43.

Plutner, H.. Schwaninger. R., Pind, S. and Balch, W.E. (1990) Synthetic
peptides of the rab effector domain inhibit vesicular transport through the
secretory pathway. EMBO J. 9, 1375-1383.

Poppiel, l. and Basch. PF. (1984) Putative polypeptide transfer from male
to female Schistosoma mansoni. Molec. Biochem. Parasit. 11, 179-188.

Price. L.S., Norman, J.C., Ridley, A.J. and Koffer, A. (1995) The small
GTPases Rae and Rho as regulators of secretion in mast cells. Current
Biology. 5, 68-73.

Pullen, J.K., Liang, S-M., Blake, M.S., Mates, S. and Tai, J.Y. (1995)
Production of Haemophilus influenzae type-b porin in Escherichia coli and
its folding into trimeric form. Gene. 151, 85-88.

Ren. M., Coutavas, E.. D'Eustachio, P. and Rush, MG. (1994) Effects of
mutant Ran/T C4 proteins on cell cycle progress. Mol. Cell. Biol. 14, 4116-
41 14.

Richards, S.A., Lounsbury, KM. and Macara, LG. (1995) The C terminus
of the nuclear RAN/T C4 GTPase stabilizes the GDP-bound state and
mediates interactions with RCC1, RAN-GAP, and HTF9A/RANBP1. J.
Biol. Chem. 170. 14405-14411.

Richmond, J. and Haydon, PG. (1993) Rab effector domain peptides
stimulate the release of neurotransmitter from cell cultured synapses.
FEBS. 316, 114-130.

Ruppel, A. and McLaren, D.J. (1986) Schistosoma mansoni: surface

110.

111.

112.

113.

114.‘

115.

116.

117.

118.

119.

120.

115

membrane stability in vitro and in vivo. Experim. Parasit. 61, 113-136.

Salminen, A. and Novick, P.J. (1987) A ras-like protein is required for a
post-Golgi event in yeast secretion. Cell. 49, 517-538.

Sambrook,J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning. a
Laboratory Manual, 1nd edition. Cold Spring Harbor Laboratory Press,
Cold Spring Harbor, New York.

Sanford, J.C., Pan. Y. and Wessling-Resnick, M. (1993) Prenylation of
Rab5 is dependent on guanine nucleotide binding. J. Biol. Chem. 168,
1 3773-1 3776.

Sanford, J.C., Pan, Y., Wessling-Resnick, M. (1995) Properties of rab5 N-
tenninal domain dictate prenyulation of C-terrninal cysteines. Molec. Biol.
Cell. 6, 71-85.

Saxe, SA. and Kimmel, AR. (1990) SAS1 and SAS1, GTP-binding
protein genes in Dictyostelium discoideum with sequence similarities to

essential genes in Saccharomyces cerevisiae. Mol. Cell Biol. 10, 1367-
1 378.

Schafer, W.R. and Rine, J. (1991) Protein prenylation: genes. enzymes,
targets and functions. Annual Rev. Genet. 30. 109-137.

Schmidt, R.A.. Schneider, OJ. and Glomset, J.A. (1984) Evidence for
post-translational incorporation of a product of mevalonic acid into Swiss
3T3 cell proteins. J. Biol. Chem. 159, 10175-10180.

Schmitt, H.D., Puzicha, M. and Gallwitz, D. (1988) Study of the ras-
related YPT1 gene product in yeast suggests a role in the regulation of
intracellular calcium. Cell. 53, 635-647.

Schmitt, H.D., Wagner, P., Pfaff, E. and Gallwitz, D. (1986) The ras-
related YPT1 gene product in yeast: a GTP-binding protein that might be
involved in microtubule organization. Cell. 47, 401-411.

Schwaninger. R., Plutner, H., Bokoch, GM. and Balch, W.E. (1991)
Multiple GTP-binding proteins regulate vesicular transport from the ER to
Golgi membranes. J. Cell Biol. 119, 1077-1096.

Seabra, M.C.. Brown, MS. and Goldstein, J.L. (1993) Retinal
degeneration in choroideremia: deficiency of rab geranylgeranyl
transferase. Science. 159:377-384.

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

131.

116

Seabra, M.C., Goldstein, J.L.. Slidhof, TC. and Brown, MS. (1991) Rab
geranylgeranyl transferase. J. Biol. Chem. 167. 14497-14503.

Segev, N.. Mulholland, J. and Botstein, D. (1988) The yeast GTP-binding
YPT1 protein and a mammalian counterpart are associated with the
secretion machinery. Cell. 51, 915-914.

Shoemaker, C.B., Ramachandran, H., Landa, A., dos Reis, MG. and
Stein, L.D. (1991) Alternative splicing of the Schistosoma mansoni gene

encoding a homologue of epidermal growth factor receptor. Mol. Biochem.
Parasitol. 53, 17-31.

Silk, M.H., Spence, I.M. and Gear, J.H.S. (1969) Ultrastructural studies of
the blood fluke - Schistosoma mansoni l. The integument. S. African J.
Med. Sci. 34, 1-10.

Sinensky. M. and Lutz, R.J. (1991) The prenylation of proteins.
BioEssays. 14(1), 15-31.

Smith. J.H.. Reynolds, ES. and von Lichtenberg, F. (1969) The
integument of Schistosoma mansoni. Am. J. Trop. Med. Hygiene. 18, 18-
49.

Smyth. J.D. The Physiology of Trematodes. W.H. Freeman & Co., San
Francisco. 1966. pp. 59-65.

Strauss, W. (1994) Preparation of genomic DNA from mammalian tissue.
In: Current Protocols in Molecular Biology, Vol. 1 (Ausbel, F. et al., eds.),
pp. 1.1.1-1.1.3. John Wiley & Sons. Inc.. New York.

Sultan. A.A., Richardson, W.A., Alano, P., Amot, DE. and Doerig, C.
( 1994) Cloning and characterisation of a Plasmodium falciparum
homologue of the ran/T C4 signal transducing GTPase involved in cell
cycle control. Mol. Biochem. Parasitol. 65, 331-338.

Tisdale, E.J., Boume, J.R., Khosravi-Far. R.. Der, OJ. and Balch, W.E.
(1991) GTP-binding mutants of rab1 and rab1 are potent inhibitors of
vesicular transport from the endoplasmic reticulum to the Golgi complex.
J. Cell Biol. 119. 749-761 .

Tropical Disease Research, Twelfth Programme Report. UNDPNVorld
BankNVHO Special Programme for Research and Training in Tropical
Diseases. 1995. World Health Organization. Geneva, Switzerland.

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

117

Trottein, F., Kieny, M.P., Venivaerde, C., Torpier, 6., Pierce, R.J., Balloul,
J-M., Schmitt, D., Lecocq, J-P. and Capron, A. (1990) Molecular cloning
and tissue distribution of a 16-kilodalton Schistosoma mansoni glutathione
S-transferase. Mol. Biochem. Parasitol. 41, 35-44.

Van Dam, G.J., Bogitsh, B.J., Fransen. J.A.M., Komelis, D.. van Zeyl,
R.J.M. and Deelder, AM. (1991) Application of the FlTC-Anti-FlTC-Gold
system to Ultrastructural localization of antigens. J. Histochem. and
Cytochem. 39. 1715-1718.

van der Sluijs. P., Hull, M., Webster, P., Male, P., Goud. B. and Mellman,
l. (1991) The small GTP-binding protein rab4 controls an early sorting
event on the endocytic pathway. Cell. 70, 719-740.

VandeWaa, EA. and Bennett, J.L. (1987) HMG-CoA reductase: role in
schistosome egg production. In: Molecular Strategies of Parasite Invasion
(Agabian, M. ed.), pp. 617-640. Liss, New York.

VandeWaa, E.A., Mills, 6., Chen, 6., Foster, LA. and Bennett, J.L. (1989)
Physiological role of HMG-CoA reductase in regulating egg production by
Schistosoma mansoni. Am. J. Physiol. 157, R618-615.

Vogel, H. (1941) Uber der Enfluss des Geschlechtspartners auf
Wachstum und Entwicklung bei Bilharzia mansoni und bei
Kreuzpaarungen zwischen verschiedenen Bilharzia-Aden. Z. Bakteriol. I.
Orig. 148, 78-96.

Walworth, N.C.. Brennwald, P., Kabcenell, A.K., Garrett, M. and Novick,
P. (1991) Hydrolysis of GTP by sec4 protein plays an important role in
vesicular transport and is stimulated by a GTPase—activating protein in
Saccharomyces cerevisiae. Molec. Cell. Biol. 1 1, 1017-1018.

Welsh. C.F., Moss, J and Vaughan, M. (1994) ADP-ribosylation factors: a
family of _10-kDa guanine nucleotide-binding proteins that activate
cholera toxin. Molec. Cell. Biochem. 138, 157-166.

West, P.M.. Kunz, SS. and Miller, KR. (1994) Activation of protein
kinase C by phorbol esters disrupts the tegument of Schistosoma
mansoni. Parasitology. 109, 461 -468.

Vlfillumsen. B.M., Norris, K., Papageorge, A.G.. Hubbert, ML. and Lowy,
DR. (1984) Harvey murine sarcoma virus p11 ras protein: biological and
biochemical significance of the cysteine nearest the carboxy terminus.
EMBOJ. 3, 1581-1585.

142.

143.

144.

145.

146.

147.

148.

118

Wilson, RA. and Barnes. P.E. (1974a) The tegument of Schistosoma
mansoni: observations on the formation. structure and composition of

cytoplasmic inclusions in relation to tegument function. Parasitology. 68,
1 39-1 58.

Wilson, RA. and Barnes, P.E. (1974b) An in-vitro investigation of dynamic
processes occurring in the schistosome tegument, using compounds
known to disrupt secretory processes. Parasitology. 68, 159-170.

Wilson, RA. and Barnes, PE. (1977) The formation and turnover of the
membranocalyx on the tegument of Schsistosoma mansoni. Parasitology.
74, 61-71.

Wright, M.D., Harrison, R.A.. Melder, A.M., Newport, GR. and Mitchell,
OF. (1991) Another 16-kilodalton glutathione S-transferase of
Schistosoma mansoni. Mol. Biochem. Parasitol. 49, 177-180.

Zahraoui, A., Joberty, G.. Arpin, M., Fontaine, J.J., Hellio. R., Tavitian. A.
and Louvard, D. (1994) A small rab GTPase is distributed in cytoplasmic
vesicles in non polarized cells but colocalizes with the tight junction
marker ZO-1 in polarized epithelial cells. J. Cell Biol. 114, 101-115.

Zaharaoui, A., Touchot, N., Chardin, P. and Tavitian. A. (1989) The
human rab genes encode a family of GTP-binding proteins related to
yeast YPT1 and SEC4 products involved in secretion. J. Biol. Chem. 164,
11394-11401.

Zhong. C., Skelly, P.J., Leaffer, D., Cohn, R.G., Caulfield, JP. and
Shoemaker, CB. (1995) lmmunolocalization of a Schistosoma mansoni
facilitated diffusion glucose transporter to the basal, but not the apical.
membranes of the surface syncytium. Parasitology. 110, 383-394.

MICHIGAN STATE UNIV. LIBRARIES
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
31293013904481