| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Neurobiochemistry, Tel Aviv University, Tel Aviv 66978, Israel
b Israel Oceanographic and Limnological Research, 81080 Haifa, Israel
c Génomique et Physiologie des Poissons, USC INRA, Université Bordeaux I, 33405 Talence cedex, France
d Department of Applied Biological Chemistry, The University of Tokyo, Bunkyo-ku, Tokyo 113, Japan
ABSTRACT
Penaeoid shrimp oocytes nearing the completion of oogenesis are enveloped in an acellular vitelline envelope and possess extracellular cortical rods (CRs) that extended into the cortical cytoplasm. These cortical specializations are precursors of the jelly layer (JL) of the egg. In searching for highly expressed mRNAs during oogenesis in the marine shrimp (Penaeus semisulcatus), two related cDNAs have been isolated that encode a mature protein of 250 amino acid residues. The deduced amino acid sequences revealed the presence of repeated cysteine-rich domains that are related to the chitin-binding domains of insect intestinal peritrophins. Similar cysteine-rich domains were reported in insect intestinal mucin, crustacean tachycitin, and invertebrate chitinases. The shrimp ovarian peritrophin (SOP) is glycosylated and can bind chitin when extracted from CRs. Its apparent molecular mass in SDS-PAGE is 29–35 kDa and 33–36 kDa, under nonreducing or reducing conditions, respectively. SOP is a major protein of CRs and the JL, and was immunodetected in ovaries; purified CRs; fertilized eggs that were surrounded by a JL matrix; and in the cloudy, whitish flocculent material appearing in sea water immediately after spawning. Immunolocalization in tissue sections determined that SOP was present in oocyte cytoplasm and in extraoocytic CRs. Shrimp expressed SOP mRNA in ovaries at all oocyte developmental stages, whereas expression in the hepatopancreas was restricted to vitellogenic stages. SOP mRNA was abundant in the shrimp ovary and was detected before the presence of the corresponding protein. This is the first demonstration that a protein with similar features to insect intestinal peritrophins is a component of CRs and is therefore a main precursor of the JL of spawned shrimp eggs.
oocyte development, ovum
INTRODUCTION
The final stages of oocyte maturation in penaeid shrimp are characterized by the appearance of rod-like bodies arranged radially around the periphery of the oocyte plasma membranes [1–13]. These cortical rods (CRs) or cortical specializations are located in extracellular pockets or crypts formed by the invagination of the oolemma into the egg cortex. The crypts elongate and extend toward the nucleus as maturation progresses and are apically overlain by the vitellin membrane that is formed prior to oocyte activation and spawning. The crypts contain the highly organized and tightly packed CRs, appearing as "bottle-brush" or "feathery" structures, and constituting around 10% of the oocyte volume [1]. Spawning and direct contact of the spawned eggs with sea water leads to the expulsion of CRs, lifting of the vitellin envelope [11], and formation of a corona composed of a flocculent matrix around the egg [3] consisting of a jelly layer (JL). Beneath the JL, the egg is covered by a surface coat until the formation of the hatching envelop a few minutes later and, therefore, the JL functions as the only external protective coat of eggs immediately after spawning. Around 45 min after spawning the JL matrix starts falling off the eggs and exocytosis of dense and ring vesicles leads to the formation of the hatching envelope. At the same time, a flocculent material appears in the sea water [3, 11, 14, 15].
The biochemical composition of the shrimp CR and the nature of the JL precursors are still poorly understood. Precursors isolated from whole mature ovaries are composed of approximately 70%–75% protein and 25%–30% carbohydrate [8]. A cDNA coding for a putative JL precursor protein was cloned from an ovarian cDNA library of Penaeus vannamei [1]. An antibody raised against the cDNA translation product expressed in bacteria showed that the 200-kDa ovarian polypeptide accumulated in CR during ovarian development [1, 10]. Unfortunately, the sequence of this clone has never been published or deposited in databases.
In searching for highly expressed mRNAs during oogenesis in the marine shrimp (P. semisulcatus), we have isolated two related cDNA clones coding for 29-kDa deduced proteins, which were termed shrimp ovarian peritrophins (SOPs). Analysis of the deduced amino acid sequence revealed similar structural features of SOP with insect intestinal peritrophins [16–18]. In addition, we demonstrate that SOP is a component of the CR precursor of the JL in shrimp eggs, it is glycosylated, and binds chitin, confirming a function deduced from its primary structure.
MATERIALS AND METHODS
Animals
Adult marine shrimps P. semisulcatus (de Haan) were caught in Haifa Bay, Israel, and kept in 3000-L maintenance tanks in running sea water as described previously [12, 19].
Collection of Eggs
Two or three ovulating females were placed in 25-L cylindrical tanks with conical bottom outlets. After spawning, eggs were collected from the bottom outlet of the tank.
Determination of Oocyte Diameter
Average oocyte diameter (AOD) was determined on ovarian tissue after resuspension in sea water [20]. Three different ovarian developmental stages were defined: previtellogenic (AOD <100 µm), vitellogenic (AOD, 100–300 µm), and late vitellogenic (AOD >300 µm). CRs in pockets were first observed microscopically in live samples from late vitellogenic ovaries with AOD exceeding 450 µm.
Molecular Cloning of SOP cDNAs
Total ovarian cDNAs were synthesized from 1 µg of poly(A)+ RNA extracted from vitellogenic shrimp ovaries using the Riboclone cDNA synthesis systems (Promega, Madison, WI). The cDNA was ligated to 
gt 11 as cloning vector using EcoRI linkers. A total of 50 000 plaques were screened using standard methods [21], with a major 1-kilobase (kb) cDNA band, identified on 1% agarose gel after ethidium bromide staining, used as a probe [22]. As will be shown in the present publication, this band, which was previously suggested to code for vitellin, is probably SOP. Two positive plaques were purified after the first round and one of them was cloned into pBluescript KS (+). This clone served as a probe for a second round of screening, which resulted in more than 10 positive plaques. Preliminary sequence analysis was performed on all 10 positive clones. Two clones that showed differences were fully sequenced and later named SOP1 and SOP2. Nine out of the 10 plaques were similar to SOP1.
Sequence Analysis
Sequence alignment and comparisons were performed with BESTFIT program of Genetics Computer Group software [23]. Database searches were performed using the BLASTP suite of programs and the BLOSUM62 matrix [24]. Multiple local alignments were constructed with the CLUSTALW program [25]. Signal sequence prediction was performed according to that described by Nielsen et al. ([26;] http://www.genome.cbs.dtu.dk/service/SignalP/). Scans for a pattern in SWISS-PROT and TREMBL were performed using the ScanProsite tool of the Expasy Molecular Biology Server (http://www.expasy.ch/).
RNA Isolation and Northern Blot Analysis
Total RNAs were extracted from ovaries and hepatopancreas (QuickPrep total RNA extraction kit; Pharmacia Biotech, Uppsala, Sweden) according to the manufacturer's instructions [27]. RNAs (20 µg) were separated on 1% agarose gels containing formaldehyde and transferred to Hybond-N membrane (Amersham, Little Chalfont, Buckinghamshire, England). The SOP1 clone from P. semisulctus and rDNA from P. aztecus (which was kindly provided by Dr. Trisha Spears, Department of Biological Science, Florida State University, Tampa, FL) were 32P-labeled by random primers using a NEBlot kit (New England Biolabs Inc., Beverly, MA) and served as probes.
Reverse Transcriptase-Polymerase Chain Reaction
Total RNAs isolated from ovaries and hepatopancreas at various molting cycle stages were subjected to reverse transcriptase-polymerase chain reaction (RT-PCR) amplifications after synthesis of a specific oligonucleotide primer set designed according to the SOP-1 cDNA sequence. The sense oligonucleotide primer was 5'-AGATGAGGTCGAATACT-3' (nucleotides 1 to 17) and the antisense primer was 5'-GAAGAGGAAGGGTAAGGG-3' (nucleotides 835 to 852), which gave an amplified product of 805 base pairs (bp). The first-strand cDNA synthesis was performed using the SuperScriptTM Preamplification System for First-Strand cDNA Synthesis (Gibco BRL, Paisley, UK). To each 10 µg of total RNA, 0.5 µg of oligo(dT)12–18 were added and brought to a final volume of 12 µl. The samples were incubated at 70°C for 10 min and kept on ice for at least 1 min. A 20-µl solution containing 20 mM Tris-HCl pH 8.4, 50 mM KCl, 2.5 mM MgCl2, 0.5 mM of each dNTP, 10 mM dithiothreitol, and 200 units of SuperScript II RT were added, incubated at 42°C for 50 min, and the reactions were terminated at 70°C for 15 min. The samples were chilled on ice and then 2 units of E. coli RNase H were added and the tubes and incubated at 37°C for 20 min. The PCR reactions were performed on 4 µl of each first-strand reaction, 100 µl of a solution containing 10 mM Tris-HCl pH 8.4, 50 mM KCl, 2.5 mM MgCl2, 0.5 nM of each dNTP, 100 pmol of each oligonucleotide primer, 0.5 unit of DynaZymeTM II DNA polymerase (Finnzymes Oy, Espoo, Finland). The PCR reactions were performed over 40 cycles, each cycle consisted of denaturation for 1 min at 94°C, annealing for 1.5 min at 55°C, and extension for 2 min at 72°C. Aliquots of 20 µl from each amplified PCR reaction were analyzed on 1% agarose gels.
Isolation of Cortical Rods
CRs were isolated as previously described [8] from late vitellogenic ovaries (AOD = 470 µm).
Isolation of Proteins from Sea Water Obtained after Spawning and Containing a Flocculent Material
A dense, cloudy, whitish flocculent material appeared immediately after spawning at the bottom of the tank and indicated the occurrence of spawning. A few minutes following spawning the eggs reached the bottom of the tank. One liter of sea water containing the flocculent material was collected as soon as it appeared, filtered to avoid contamination with eggs, aliquoted into 50-ml tubes, and stored at -70°C. Aliquots of sea water containing the flocculent material were lyophilized and the dried material was dissolved in 10 mM PBS saline buffer (150 mM NaCl, 10 mM phosphate buffer [mixture of 8:2 ratio of NaH2PO4:NaH2PO4] pH 7.4) up to 10 ml. Sonication was performed (4 x 30 sec), 100% trichloro-acetic acid was added to a final concentration of 15%, and the samples were kept on ice for 1 h. After a centrifugation of 30 min at 17 000 x g, the pellets were washed with cold ether and resuspended into PBS containing 2% SDS.
Protein Analyses
Ovarian and hepatopancreas tissue fragments were collected from females in previtellogenic, vitellogenic, and late vitellogenic (with CRs) ovarian developmental stages. They were homogenized in 10 mM PBS containing 1 mM of phenylmethylsulfonyl fluoride (PMSF) and 1 µg/ml of aprotinin (Sigma, St. Louis, MO). Samples were centrifuged for 10 min at 4°C at 10 000 x g and the supernatants were analyzed by polyacrylamide gel electrophoresis (PAGE). Purified CRs were homogenized in 10 mM PBS pH 7.4 containing 2% SDS, sonicated (4 x 30 sec), and applied on a 12% or 15% SDS-PAGE [28] before or after reduction by ß-mercaptoethanol. Protein concentration was determined using the Bradford protein assay (Bio-Rad, Richmond, CA).
N-Terminal Amino Acid Sequencing
CR extracts were subjected to SDS-PAGE under reducing conditions, transferred to PVDF membranes, and stained with Coomassie Blue G. The SOP band was eluted and N-terminal sequence was determined as described previously [29].
Preparation of Rabbit Specific Antisera Against SOP
The first specific polyclonal antiserum (a-SOP) was generated against the synthetic peptide, Ser-Val-Thr-Thr-Asp-Asn-His-Pro-Tyr-Ser-Lys-Leu-Cys linked to bovine serum albumin as an immunogen. The peptide sequence was the N-terminal end of the mature SOP determined after microsequencing of the protein purified from CRs and in accordance with the sequence deduced from its cDNA (see Results). The second specific polyclonal antiserum (a-SOP-35) was generated against the 
35-kDa major protein band observed in the flocculent material from sea water after subjecting it to SDS-PAGE. After verifying that the protein band cross-reacted with the a-SOP in Western immunoblot analysis, it was cut out from gels for injection into rabbits.
Immunoblotting Procedure
Following SDS-PAGE performed under reduced or nonreduced conditions, proteins were transferred to a nitrocellulose membrane, blocked with 10% fat-free milk, and incubated with rabbit a-SOP or a-SOP-35 antisera. Goat anti-rabbit immunoglobulin G (IgG) conjugated with horseradish peroxidase (HRP) or with alkaline phosphatase (Sigma) were used as secondary antibodies. Preliminary results showed that the a-SOP antiserum cross-reacted with SOP in immunoblots prepared from SDS-PAGE performed under reduced conditions, whereas a-SOP-35 antiserum cross-reacted with SOP in immunoblots prepared from SDS-PAGE performed either under reduced or nonreduced conditions.
Analysis of SOP Carbohydrates
CRs and JL were subjected to 15% SDS-PAGE under reducing conditions and transferred to nitrocellulose membranes. Binding to concanavalin A (Con A; Sigma) or wheat germ agglutinin (WGA; Sigma) was tested in protein blots [30], except that Con A-HRP or WGA-HRP were used in these experiments. Samples were also subjected to digestion with endoglycosidase H (Endo H) or N-glycosidase F (Endo F; according to the manufacturer's instructions, Boehringer-Mannheim, Mannheim, Germany), prior to SDS-PAGE and transfer to nitrocellulose membranes. Membranes were incubated with Con A-HRP, WGA-HRP, or subjected to Western blot analyses with a-SOP-35 antiserum.
Binding of SOP to Chitin
CR, JL, or ovarian homogenates (50–100 µg in each sample) were suspended in PBS containing 1 mM PMSF and each sample was incubated with 0.2 ml of affinity matrix of chitin beads (New England Biolabs), which was previously equilibrated with PBS. The mixture was incubated overnight at 4°C on a 360° rotary shaker. Samples were centrifuged (14 000 x g; Eppendorf microfuge), the supernatants containing the unbound proteins were collected, and the chitin beads were extensively washed with >x10 beads volume of 20 mM Hepes pH 8.0, 500 mM NaCl, and 0.1 mM EDTA. Proteins bound to the chitin beads were eluted with SDS sample buffer and boiled for 5 min before being subjected to 15% SDS-PAGE under reducing conditions. Total proteins were visualized with Coomassie Blue staining and SOP was detected by Western blot analyses using a-SOP antiserum.
Immunofluorescence Labeling Analysis
Shrimp ovarian fragments were fixed overnight with 4% paraformaldehyde in PBS, embedded in paraffin, and serial sections of 7 µm in thickness were prepared. The sections were pretreated for 10 min with 1% SDS in PBS at room temperature to enhance epitope exposure, and were then blocked for 1 h at 37°C in 10% normal calf serum. This blocking agent was also used to dilute the primary and secondary antibodies. Sections were incubated for 3 h at 37°C with rabbit preimmune serum or a-SOP-35 antiserum at 1:50 dilution. After washing, the sections were treated for 2 h at 37°C with Alexa 488-conjugated goat anti-rabbit IgG (Molecular Probes, Eugene, OR) or Cy2-conjugated goat anti-rabbit IgG (Jackson ImmunoResearch Laboratories, West Grove, PA) at 1:200 dilution. Immunostaining was analyzed on a conventional fluorescence microscope (Zeiss Axiophot, Oberkochen, Germany).
RESULTS
Molecular Cloning and Characterization of SOP cDNAs
Two different cDNA clones, SOP1 and SOP2, were isolated from a shrimp ovarian cDNA library (Fig. 1). For the SOP1 clone, information was obtained from nucleotides -2 to +928, excluding the poly(A) tail, and an open reading frame (ORF) of 831 nucleotides was shown coding for 277 amino acids (from nucleotides +1 to +831). A polyadenylation consensus signal (AATAAA) was found 13 nucleotides upstream from the poly(A) tail. By comparison with SOP1, SOP2 appeared as an incomplete cDNA clone with information obtained from nucleotide +4, thus excluding the initiator methionine. An ORF of 828 nucleotides was shown coding for 276 amino acids. The DNA sequence of the SOP1 and SOP2 clones was found to be heterologous at 29 positions in the translated part of the cDNA, resulting in 10 differences between the deduced amino acid sequences. These differences identified in the ORF together with the presence of gaps after alignment of the 3'-untranslated part of the cDNAs (Fig. 1) strongly suggest that the two cDNAs were products of two different genes. A signal peptide of 19 amino acids could be predicted by using a neural network method resulting in an SOP1 of 258 amino acids with an estimated molecular mass of 29 141 daltons. By microsequencing of the SOP protein purified from CR (see below), we determined an amino-terminal sequence of Ser-Val-Thr-Thr-Asp-Asn, leading to a deduced mature SOP1 protein of 250 amino acids with an estimated molecular mass of 28 156 daltons. These results suggest the cleavage of a presegment of prepro SOP generating a proprotein that differs from its mature form by the presence of an octopeptide amino terminal extension (Fig. 1) and a reduction of the calculated molecular mass of 1 kDa. There was one potential N-linked glycosylation site, at position 159 of SOP1, a site conserved in SOP2. Deduced amino acid composition of SOP1 revealed a high proportion of cysteine (9.3%), glycine (7.8%), and aspartic acid (7%); and a very low proportion of tryptophan (0.8%), histidine (1.9%), and methionine (2.3%).
|
Domain Structure of SOP cDNA and Similarities with Other Proteins
The deduced amino acid sequence of SOP1 was used in order to screen the GenBank nonredundant database. Similar results were obtained with SOP2-deduced amino acid sequence and were therefore not presented in the manuscript. The best match found using the BLASTP suite of programs with SOP1 was human fibrillin-2 precursor (score 54, expected value 1e-06). However, the significant sequence similarity with fibrillin was primarily due to the presence of multiple epidermal growth factor-like cysteine-rich domains in fibrillin rather than an overall sequence similarity. CG13805 gene product of Drosophila melanogaster (accession number AE003474) and peritrophin-44 (PM44, accession number L25106) from the peritrophic membrane of the larvae of the fly Lucilia cuprina were the second and third significant best scores obtained (score 54, expected value 2e-06 and score 50, expected value 3e-05). The characteristic feature of PM44 is the presence of five nonidentical but related cysteine-rich domains with a common register of six-cysteine residues. These repeats were also found in peritrophin-48 (accession numbers U79715 and AF030557) and peritrophin-95 (accession number U23828). The overall organization of PM44 was conserved in SOP1 (Fig. 2A) with a conservation of the six-cysteine consensus sequences in the cysteine-rich regions I to III (except cysteine 1 of domain III of SOP1; 28% identity for 156 amino acids compared of domains I to III; Fig. 2B). Repeats IV and V of SOP1 were imperfect repeats. The IV repeat of SOP1 contained cysteine numbers 1, 3, 4, and 5, whereas repeat V contained cysteine numbers 4 and 5. Comparison of SOP1 and PM44 internal repeats I to III indicated a strong conservation of an aromatic amino acid between cysteines 2 and 3, and 4 and 5 (Fig. 2B).
|
Searching SWISS-PROT and TREMBL with the following consensus sequence C-X(2)-[YFW]-X(2)-C-X(9,14)-C-X(5)-[FWY]-X(6)-C, which contains cysteine numbers 2, 3, 4, and 5 of the cysteine-rich repeat, was used in order to find related proteins to SOP or PM44 and containing at least one cysteine-rich repeat. Five matches were found in other proteins from the peritrophic membrane, the insect intestinal mucins, IIM14 and IIM22, of Trichoplusia ni larvae. In IIM, the cysteine-rich domains are interdispersed in the molecule (Fig. 2A). Other repeated regions are present in IIM, between domains I and II or IV and V, but they are not related with the cysteine-rich domain of SOP or PM44. In addition to the presence of the cysteine-rich domain in SOP and in peritrophic membrane proteins, the pattern was matched with invertebrate chitinases (e.g., 13 times for 35 sequences extracted with the consensus sequence from TREMBL). This is consistent with the significant scores found using BLASTP with some chitinases (e.g., chitinase PjCHI3 from the prawn, P. japonicus [score 41, expected value 0.016]). The six-cysteine consensus sequence was generally located in the C-terminal part of chitinases, so outside the enzyme catalytic domain (Fig. 2A). One cysteine-rich domain was also found in tachycitin, a small antimicrobial protein with chitin-binding activity found in horseshoe crab hemocytes.
Expression of the SOP mRNA in Shrimp Ovary and Hepatopancreas
The site of SOP mRNA expression was revealed by Northern blot analysis using SOP1 clone as a probe of RNA extracted from ovaries and hepatopancreas. Preliminary Northern blot analyses showed similar results with SOP1 and SOP2 probes. The differential expression, if any, of the two putative genes could not be discriminated at this stage due to the small number of differences found between SOP1 and SOP2 cDNA sequences. Ovaries show a single 1-kb SOP transcript that was expressed at a very high level during previtellogenesis (Fig. 3a, lane 1) and to a lower extent during vitellogenesis and CR ovarian stage (Fig. 3a, lanes 2 and 3, respectively). Control hybridization was performed with a 1023-bp shrimp rRNA and showed a similar expression pattern at the three ovarian developmental stages (Fig. 3b). RT-PCR analysis was performed on total RNAs extracted from ovaries and hepatopancreas collected from females 4, 6, 10, and 14 days after molting. A specific 805-bp DNA fragment was amplified in ovarian tissue at all developmental stages (Fig. 4, lanes 1 to 4). However, only a trace amount of amplified product was obtained 14 days after molting in comparison with Day 10 (Fig. 4, lanes 4 and 3). The same PCR product was obtained with RNA extracted from hepatopancreas at vitellogenic stage (Fig. 4, lanes 6 and 7) but not with previtellogenic and late vitellogenic stages (Fig. 4, lanes 5 and 8, respectively). The integrity of the RNA extracts used in these reactions is shown in the lower panel of Figure 4.
|
|
Immunodetection of SOP
In order to detect SOP in shrimp tissues, extracts prepared from ovarian homogenates were subjected to SDS-PAGE under reduced conditions (Fig. 5A) and immunodetection was performed with a-SOP (Fig. 5B) antiserum. The Western blot analysis showed that SOP was present in the ovary during vitellogenic and late vitellogenic stages. The a-SOP antiserum did not detect any proteins in previtellogenic stages in SDS-PAGE under reducing conditions. At the same reproductive stages, no immunoreactions were obtained with homogenates or samples of hepatopancreas, intestine, or hemolymph (data not shown). To investigate the occurrence of SOP in eggs after spawning, eggs were collected at several embryonic developmental stages and were subjected to a Western blot analysis using the a-SOP antiserum. The flocculent material appearing immediately after spawning in sea water was also collected for this analysis (see Materials and Methods). A strong immunoreaction was demonstrated in eggs when surrounded by the JL and in the proteins extracted from the lyophilized flocculent material appearing in sea water (Fig. 6, a and b; lanes 1 and 2, respectively). A faint immunoreaction was detected with homogenates of one- or two-cell embryos that lack the JL (Fig. 6b, lanes 3–4, respectively). No immunoreaction was found in later stages of eggs with fully developed embryos, shortly before hatching (data not shown). These results indicated that SOP is associated with ovaries and with eggs immediately after spawning when they are surrounded by a JL that dissociates shortly after initiation of embryonic development. Because JL originates from CRs [1–13], CRs were purified and the homogenates were subjected to PAGE under reduced conditions (Fig. 7A). Western blot analyses showed the presence of SOP using a-SOP and a-SOP-35 antisera (Fig. 7, B and C). The apparent molecular mass of SOP in extracts from CRs, JL, and ovaries in 15% SDS-PAGE ranged from 33 to 36 kDa under reducing conditions and from 29 to 35 kDa under nonreducing conditions, respectively, in five replicate preparations (data not shown). A 79-kDa protein cross-reacted with a-SOP (Fig. 7B) and a 122-kDa protein faintly cross-reacted with a-SOP-35 (Fig. 7C), possibly indicating the presence of SOP polymers in the CR.
|
|
|
Immunolocalization of SOP in Shrimp Oocytes
Figure 8 shows the immunolocalization of SOP on serial sections at different ovarian developmental stages. At the end of previtellogenesis, serial sectioning of the ovarian lamellae demonstrated a strong immunofluorescence labeling in the larger oocytes located at the periphery of the ovarian lamellae (Fig. 8A). Previtellogenic oocytes in the center of the lamellae were not immunoreactive and remained unlabeled during vitellogenesis (Fig. 8, A and B). At this stage, considerable enlargement of the peripheral oocytes were observed, and both vitellogenic and atretic oocytes were labeled with the a-SOP-35 antiserum (Fig. 8B). Higher magnification revealed that SOP was specifically present in the oocyte cytoplasm compartment with no label detected in the nucleus or in the surrounding follicular cells (Fig. 8C). In oocytes containing CRs, immunoreactivity was demonstrated in both the cytoplasm and in the CRs (Fig. 8D). SOP immunoreactivity was not found in sections incubated with the preimmuned serum, whereas a limited autofluorescence of the yolk globules was observed (Fig. 8E).
|
SOP Carbohydrates and Binding to Chitin
In order to determine whether SOP was glycosylated, binds chitin (or both), as suggested from its primary structure, the following experiments were performed with CR and JL preparations. Glycosylation was determined by binding to the lectin Con A. Con A was found to bind to CR and JL proteins including SOP, and these were removed after incubation of samples with either Endo H or Endo F (Fig. 9). Endo F releases common class N-glycans from the protein backbone, whereas Endo H cleaves GlcNac-GlnNac bonds from aspargine-linked glycan chains of glycoproteins. Immunoblots show a reduction (1 kDa) in the apparent molecular mass of SOP in CR and JL after enzymatic digestion with Endo H or Endo F (Fig. 10), indicating the presence of N-linked high mannose oligosaccharides in SOP. Only weak binding was revealed in CR with WGA before and after treatment with Endo H or Endo F (data not shown). In order to determine whether SOP binds chitin, CR, JL, and ovarian homogenates were incubated with chitin binding beads. SOP from CRs was found to bind chitin beads, whereas SOP from either ovarian homogenates or JL (Fig. 11) did not bind to chitin beads under the same experimental conditions (see Discussion for rationale).
|
|
|
DISCUSSION
Previous studies have shown the presence of a major 1-kb cDNA band after cDNA synthesis from poly(A)+ RNAs extracted from vitellogenic shrimp ovaries [22]. It was concluded that this abundant transcript could be linked to yolk protein deposition occurring during the rapid enlargement of the oocytes [22]. Here we describe the identification and molecular cloning of this hitherto unknown protein, termed shrimp ovarian peritrophin, or SOP, which is not related to vitellin or vitellogenin, as suggested previously [22].
The structural organization of the product specified by the isolated SOP cDNAs revealed the presence of repeated cysteine-rich domains that are related to the chitin-binding domains of insect intestinal peritrophins (Fig. 2). The green bottle fly Lucilia cuprina PM44 is an archetypal molecule of the peritrophins, it is glycosylated, and contains five related cysteine-rich domains each characterized by a similar distribution of cysteine residues [16]. Structurally related domains were reported in peritrophins from several insects and a number of proteins with similar structural characteristics are predicted from the genome of Caenorhabditis elegans and D. melanogaster [17, 18]. Because PM44 bind chitin, it is assumed that the chitin-binding capability of the molecule is due to the presence of at least one repeat or a certain number of repeats of the cysteine-rich domain [16]. The repeated domain structure may give enhanced strength to the interaction between the protein and the long oligomers of N-acetyl glucosamine that make chitin fibrils [17]. Similar cysteine-rich domains defined as invertebrate chitin-binding domain [31] or peritrophin-A domain [17] were also found in insect intestinal mucins (IIM14 and IIM22 of T. ni larvae), outside the catalytic domain of invertebrate chitinases and in tachycitin, a small antimicrobial protein of horseshoe crab hemocytes (Fig. 2). These proteins bind chitin [16, 32, 33]. It has been speculated [17] that the conserved aromatic amino acids in peritrophin-A domain bind GlcNac within chitin. Moreover, the three disulfide bonds in the domain may constrain the polypeptide to present the aromatic amino acids on the protein surface for interaction with the ring structures of the sugars. Chitin binding was demonstrated for SOP present in CR but not with the protein from JL or ovarian homogenates (Fig. 11), indicating a difference in the chitin-binding site accessibility between these sources. SOP in ovaries may have not been fully processed and therefore does not yet show abundant chitin binding sites, whereas chitin may have occupied all available sites on SOP of the JL. SOP from CRs and JL was found to bind the lectin Con A (Fig. 9), but no significant binding was observed for WGA, a lectin that is assumed also to bind chitin [34], suggesting the disruption of protein-chitin interactions by the detergent in SDS-PAGE. The presence of chitin-like glycoprotein linked to carbohydrate in the extracellular coat of crustacean eggs has been demonstrated [35, 36].
The calculated molecular mass of mature SOP protein deduced from its cDNA and after microsequencing of its N-terminal end is not in full agreement with the apparent molecular mass of SOP extracted from CR and JL, as determined by SDS-PAGE. This difference is probably due to post-translational modifications of SOP. SOP is synthesized with a signal sequence of 19 amino acid residues and secreted in a precursor form, which is further modified post-translationally by cleavage of the proform into the mature product found in the CR. Also, SOP was found as a N-glycan glycosylated protein rich in mannose (Figs. 9 and 10), leading to an increase in the apparent molecular mass in CR and JL. The apparent unusual higher molecular mass of SOP in SDS-PAGE under reducing conditions (33–36 kDa) compared with that observed with nonreducing conditions (29–35 kDa) may indicate an abnormal SDS binding as regularly observed for glycoproteins [37]. It should be noted that the mobility of a protein on SDS-PAGE can be affected by its chemical composition and structure [38], including single amino acid substitutions [39].
In shrimp ovary, SOP mRNA is expressed at all ovarian stages (Figs. 3 and 4), whereas in hepatopancreas, SOP transcripts were detected only during the vitellogenic stage (Fig. 4 and [22]). However, the expression of SOP in the hepatopancreas may be modulated by molting stage in addition to oocyte diameter (Fig. 4) and are not necessarily correlated with vitellogenesis. Western blotting analysis revealed that SOP is present in the ovary mainly during vitellogenic and late vitellogenic stages, including those showing CR (Fig. 3), but not in hepatopancreas, intestine, or hemolymph (data not shown). SOP transcripts were detected in large amounts in the ovary at stages prior to the detection of the proteins in ovarian homogenates (Figs. 3 and 5). Thus, there is no evidence at this stage that the SOP protein is found outside vitellogenic ovaries and its function in the hepatopancreas remains unknown at this stage. Localization of SOP by immunofluorescence indicates that SOP is very abundant in the cytoplasm of vitellogenic oocytes and could be detected at the same time in the extracellular CR during the late vitellogenenic stage (Fig. 8). These results strongly suggest that SOP is synthesized by the oocyte, and not the surrounding follicular cells, to be exported and deposited to the CR extracellular compartment.
It is well known that oocytes of several animal groups are surrounded by extracellular coats containing carbohydrates, but several biochemical differences between penaeid and other animal ova JL have been noted [8, 11, 13]. Following contact with sea water, Penaeus ova undergo a massive release of extracortical jelly precursor material found in the CR, which is transformed into a layer of jelly-like material surrounding the ova [11]. The amino acid composition of isolated jelly precursor from P. aztecus revealed, as for the product specified by SOP cDNA, high ratios of cysteine, glycine, and aspartic acid, and low ratios of histidine and methionine [8]. Massive release of SOP found in the CR during spawning of eggs from the ovary was demonstrated after analysis of the flocculent material appearing in sea water immediately after spawning. The flocculent material consisted mainly of a 30- to 33-kDa protein that immunoreacted with a-SOP antiserum (Fig. 6). The disintegration of the JL shortly after spawning [3] could explain the massive appearance of SOP in the flocculent material in sea water. SOP is present in eggs surrounded by the JL, but not in eggs with developing embryos that lack the JL (Fig. 6). JL has an important and critical role during early embryonic development. At this stage, embryos have no protection except the JL, because the assembly of hatching envelope is still incomplete [11].
In conclusion, we demonstrated that the product specified by an abundant shrimp ovarian transcript presents similar chitin-binding features to insect intestinal peritrophins. This protein is secreted by the oocyte as a major component of the extraoocytic CR and is therefore a precursor of the JL of shrimp eggs.
ACKNOWLEDGMENTS
We thank Dr. Trisha Spears, Department of Biological Science, Florida State University, Tampa, Florida, for the generous gift of rDNA from P. aztecus. Many thanks are due to Dr. Nechama Smorodinsky, Department of Cell Research and Immunology, Hybridoma Unit, Tel Aviv University, Tel Aviv, Israel, for preparing polyclonal a-SOP and a-SOP-35 antisera. The skillful technical assistance of Chantal Ballagny, Orna Gibson, and Isabelle Galais, is gratefully acknowledged.
FOOTNOTES
First decision: 11 August 2000.
1 This work was supported by grant 93-0083/1 from the Binational Science Foundation and grant 8938-1-97 from the Israel Ministry of Science. The nucleotide sequences reported in this paper have been submitted to the GenBank/EBI databank with accession numbers AF095580 for SOP1 and AF095581 for SOP2 cDNA sequences. 
2 Correspondence: Esther Lubzens, Israel Oceanographic and Limnological Research, Tel-Shikmona, P.O. Box 8030, Haifa 31080, Israel. FAX: 972 4 8511911; esther{at}ocean.org.il
Accepted: November 10, 2000.
Received: June 26, 2000.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  K. Yamano, G.-F. Qiu, and T. Unuma Molecular Cloning and Ovarian Expression Profiles of Thrombospondin,a Major Component of Cortical Rods in Mature Oocytes of Penaeid Shrimp,Marsupenaeus japonicus Biol Reprod, June 1, 2004; 70(6): 1670 - 1678. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
