| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Nicholls State University,3 Thibodaux, Louisiana 70310
Lab IRTA-ICM,4 CMIMA, 08003-Barcelona, Spain
Reference Center in Aquaculture,5 Generalitat de Catalunya, Barcelona, Spain
Dipartimento di Scienze del Mare,6 Universita Politecnica delle Marche, 60131 Ancona, Italy
Department of Physiological Sciences,7 University of Florida, Gainesville, Florida 32611
Whitney Laboratory,8 University of Florida, St. Augustine, Florida 32086
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
cathepsin, developmental biology, free amino acids, gamete biology, gametogenesis, hydration, killifish, maturation-inducing steroid, oocyte development, proteolysis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A clear synopsis of precursor-product relationships in many teleosts when compared to those of tetrapods is further complicated by an additional YP processing that occurs in many teleost oocytes during meiosis reinitiation (oocyte maturation). The most striking difference in yolk profiles documented for Fundulus heteroclitus, a spawner of benthic eggs (benthophil species), concerns the disappearance of the oocyte largest 122-kDa YP (YP 122) and the concomitant appearance of smaller YP bands immediately before ovulation [27–29]. In other marine teleosts that spawn pelagic (buoyant) eggs (pelagophil species), a more pronounced proteolysis of YPs has been reported [8, 9, 17, 19, 29–36], which appears to be mediated by the lysosomal cysteine proteinase cathepsin L [37]. This enhanced proteolytic processing may be related to a unique preovulatory process that occurs in many marine teleost oocytes, termed hydration, which determines egg and early embryo survival. Near the time of germinal vesicle breakdown (GVBD), a rapid increase of oocyte volume occurs due to the uptake of water, which results in an increase of oocyte volume of approximately 2-fold in postmaturational oocytes from benthophil species, and up to 8-fold in pelagophil fish [8, 28, 32, 34, 38–43]. The major osmotic effectors for oocyte hydration are the accumulation of inorganic ions and the generation of abundant free amino acids (FAAs) derived from the proteolysis of YPs during maturation [8, 9, 19, 29, 32, 34–36, 39–42, 44–46]. The recent discovery of a novel water channel (aquaporin) in pelagophil oocytes and its role mediating water uptake into the oocyte during maturation, indicates that hydration of fish oocytes is a highly controlled mechanism mainly based on the interplay between YP hydrolysis and aquaporins [43].
To date, Vtg-YP precursor-product relationships and further processing of YPs during oocyte maturation and hydration have been reported for only two species of pelagophil teleosts [9, 19]. Because almost no information is available for marine benthophil teleosts, in which oocyte hydration is less pronounced and more directly associated with the accumulation of K+ [44, 45], we sought to characterize the Vg-YP relationships in F. heteroclitus, with an emphasis on the processing of YP 122 during oocyte maturation and the potential proteases involved. After the completion of the cDNA and putative protein sequences of two F. heteroclitus Vgs [47, 48], the necessary blueprint for a comparison with microsequencing data for oocyte and egg YPs was available. In this paper, we document internal and N-terminal amino acid sequences from seven isolated YPs, all of which can be positioned within the Vg1 and Vg2 predicted protein sequences. Our data suggest that the majority of YPs are derived from Vg1, and that a small amount are derived from Vg2, in contrast to what has been reported for pelagophil teleosts. Additionally, we show that the rapid processing of YP 122 during preovulatory oocyte hydration is associated with increased cathepsin B, rather than cathepsin L, enzyme activity.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Adult F. heteroclitus were collected from the salt marshes of northeast Florida and the Bay of Cádiz (Spain), and they were maintained under reproductively active conditions in the laboratory as described [49, 50]. Unless indicated otherwise, chemical reagents, culture medium, and hormones were purchased from Sigma (Madrid, Spain). All procedures for the sampling of fish and their killing were approved by the Ethical Committee from IRTA (Spain) and followed institutional animal care and use committee guidelines.
SDS-PAGE of Yolk Proteins and Microsequencing
Ovarian follicles and eggs were dissected from the ovary and placed in 1.5-ml Eppendorf tubes free of culture medium containing 100–750 µl of 1x Laemmli sample buffer [51]. The follicles were immediately homogenized with a sterile Kontes pestle (Iberlabo, Madrid, Spain) and heated for 5–10 min at 97°C. The tubes were cooled to room temperature and 10 U of Benzonase was added to the homogenates to digest DNA for approximately 20 min. The homogenates were then briefly centrifuged at 12 000 x g for 3 min, separating the dissolved yolk from the insoluble cellular debris. The supernatant was stored at –20°C or immediately run on SDS-PAGE according to basic procedures described by Laemmli [51] using 10% or 15% acrylamide minigels (7 x 10 cm). Molecular weight standards and follicle and egg homogenates (0.05–0.1 follicles/lane) were placed in wells and electrophoresed at constant voltage (180 V). Protein bands were visualized by fixing gels in 12% trichloroacetic for 1 h, overnight staining in 0.2% Coomassie blue R-350 (Amersham-Pharmacia Biotech) in 30% methanol plus 10% acetic acid, and final destaining in 25% methanol, 7% acetic acid.
Gradient gels (7%–20% acrylamide) for overall YP profiles were also run and calibrated as previously described [28]. For microsequencing, YPs were initially separated on 125 x 140 x 1.5 mm slab gels containing a 3.5% stacking gel overlaying separating gels from 7% for larger YPs to 12% for smaller YPs, and Tris-tricine running buffers [52]. After electrophoresis, proteins were electroblotted onto polyvinylidene difluoride (PVDF) membranes in buffer containing 10 mM MES (2-morpholinoethanesulfonic) pH 6, and 20% methanol at 20 V overnight. Protein bands were visualized by brief staining in 0.02% Coomassie blue in 40% methanol plus 5% acetic acid and rinsed in distilled water. Membranes were dried and stored at –20°C until individual bands were cut out and submitted for sequencing. N-terminal amino acid analysis were performed on PVDF-bound proteins using the Applied Biosystems Model 473a Sequencer [53] by the Protein Chemistry Core Facility of the University of Florida.
The two largest N-terminally blocked YPs (YP 122 and YP 103) were again electrophoresed, blotted onto PVDF, and subjected to cleavage [54] by 0.003 U/µg protein of endoproteinase LysC (Endo LysC; Promega, Madison, WI) in 50 mM Tris pH 8.8, 0.2 M ammonium bicarbonate, 0.1% SDS, and 0.1 mM EDTA. Protein fragments were then separated on Tris-tricine gels, blotted to PVDF, and visualized by silver staining [55]. Similar bands of 13 kDa (presumed to be identical) were isolated from the YP 122 and YP 103 digestion and once again submitted to the Protein Chemistry Core for N-terminal amino acid sequencing.
Sequencing data for Vg cDNAs along with YP microsequencing results were organized using the PC/GENE software package (Intelligenetics, Mountain View, CA). Prediction of signal peptides was carried out using the method described by von Heijne [56] (www.cbs.dtu.dk/services/SignalP). Consensus sites for proteolysis (PEST) were designated according to the algorithm described by Rogers et al. [57] (www.at.embnet.org/embnet/tools/bio/PESTfind).
Induction of Oocyte Maturation In Vitro
Females were collected from the tanks, killed by stunning and decapitation, and the ovaries immediately removed and placed in 60-mm diameter Petri dishes containing 10 ml of 75% Leibovitz L-15 medium with L-glutamine and 100 µg gentamicin/ml pH 7.5. Fully grown follicle-enclosed oocytes were manually isolated from the rest of the ovary using watchmaker forceps. Groups of 20–30 follicles were used for induction of oocyte maturation in vitro using 0.1 µg/ml of the naturally occurring maturation-inducing steroid (MIS) 17
,20ß-dihydroxy-4-pregnen-3-one (17,20ßP) following the procedures described by Greeley et al. [58]. Control (1% ethanol) and steroid-treated follicles were incubated at 25°C up to 48 h in a temperature-controlled incubator. The occurrence of oocyte maturation in vitro was scored by the incidence of GVBD. Samples of individual or groups of follicles undergoing oocyte maturation were frozen in liquid nitrogen and kept at –80°C for further SDS-PAGE analysis and determination of cathepsin enzyme activity. The time-course degradation of the YP 122 and YP 45 in vitro was monitored by measuring the intensity of the corresponding bands in SDS-PAGE using Quantity-One software (Bio-Rad Laboratories, Hercules, CA). The intensity was normalized with respect to that of the YP 77 band, which does not change during oocyte maturation.
Cathepsin L and B Enzyme Assays
Cathepsin L (CatL) and cathepsin B (CatB) enzymatic activities were determined in ovarian follicles undergoing oocyte maturation in vivo and in vitro. Follicles (50–100 mg) were placed in 2 volumes of distilled water, and the mixture was thoroughly homogenized and centrifuged at 14 000 x g for 10 min at 4°C. The supernatant was recovered and used for enzymatic assays. The protein content of the supernatants was measured using the Bio-Rad (Barcelona, Spain) Protein Assay adapted to microtiter plates using BSA as standard. The fluorescent assay for both CatL and CatB activity was based on that described by Barret and Kirschke [59], adapted for use in 96-well plates using Z-Arg-Arg-7-amino-4-methylcoumarin (AMC) and Z-Phe-Arg-AMC as substrates for CatB and CatL, respectively. Although Z-Arg-Arg-AMC appears to be a specific substrate for CatB, both CatB and CatL can catalyze the hydrolysis of Z-Phe-Arg-AMC [60]. Thus, the individual contribution of CatL to the hydrolysis of Z-Phe-Arg-AMC was estimated performing the CatL assay in the presence and absence of 2 µM CA-074, a highly specific CatB inhibitor [61]. The supernatant of the samples (2 and 4 µl) were diluted to 100 µl with 0.1% Brij 35, and 50 µl of activator buffer (340 mM sodium acetate, 60 mM acetic acid, 4 mM disodium-EDTA, 8 mM dithiothreitol, pH 5.5) was added. After 1 min to allow enzyme activation at 30°C, 50 µl of substrate solution (400 µM Z-Arg-Arg-AMC or 20 µM Z-Phe-Arg-AMC for CatB and CatL, respectively) was added and incubated during 30 min at 30°C. The AMC products were measured at 40-sec intervals using a Synergy HT temperature-controlled spectrofluorometer (BIO-TEK), with excitation set at 360/40 nm and emission at 460/40 nm. The specific activity was expressed as pmol AMC per minute per milligram of protein, using the slope of the kinetic curve between 5 and 20 min for the calculations of the AMC released.
Data Analysis
Data on YP 122 degradation and cathepsin enzyme activity are presented as the mean ± SEM. The data were statistically analyzed by one-or two-way analysis of variance after arcsine transformation when needed, and followed by the Duncan multiple-range test. All experiments were performed at least three times on different batches of ovarian follicles collected from reproductively active females.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The YPs typically found in F. heteroclitus oocytes and eggs are demonstrated in Figure 1, along with our designations of certain bands according to their apparent molecular mass as determined on gradient gels (see also [28, 29, 62]). At least nine YPs were resolved by SDS-PAGE, successfully blotted onto PVDF membranes, and submitted for protein sequencing by Edman degradation. Four YPs (YP 122, YP 103, YP 77, and YP 75) appeared to be N-terminally blocked, whereas five yielded N-terminal sequences (Table 1). Four of the five N-terminally sequenced YPs (YP 45, YP 42, YP 26, and YP 19) were successfully mapped to internal positions within Vg1 (Fig. 2), whereas the fifth (YP 69) mapped to the N-terminal position of Vg2 (VgII) [47, 48]. YP 42, obtained from eggs, overlaps YP 45, which is present only in oocytes, so it is apparently generated from YP 45 by cleavage of 26 amino acids (2.8 kDa) from the N-terminus.
|
|
|
To identify the origins of YP 122 and YP 103, each protein band was proteolytically cleaved with Endo Lys C. The digestion products were again separated on Tris-tricine gels and visualized by silver staining. The reaction with Endo LysC was confirmed as only a partial digestion by the isolation of peptide products larger than those predicted if cleavage had occurred at every lysine residue. The pattern of electrophoresed digestion products from YP 122 and YP 103 initially appeared to be identical, indicating that the two YPs originated from the same precursor molecule (not shown). However, a difference between the digestion products was discovered when the 13-kDa peptides derived from YP 122 and YP 103 Endo LysC digestion were sequenced. The 13-kDa band isolated from the YP 103 contained two peptides (YEFSDELLQTPLQLIKISD. . and LFESLVDSDKVVENPLLREVVFLGY. .) mapping near the N-terminal region of Vg1 (Fig. 2). The more heavily stained 13-kDa band isolated from YP 122 contained the same two peptides found in the YP 103 digestion, but also a third peptide (KY.AKHIGVGLKA.FKFASQ. .) that mapped much farther along the Vg1 sequence, beginning at residue 982 (Fig. 2). We interpret these data as evidence that YP 122 and YP 103 are identical Vg1-derived YPs except for a short 19-kDa extension at the C-terminus of YP 122, which contains the third Endo LysC digestion product. Therefore, we concluded that both YP 122 and YP 103 correspond to lipovitellin heavy chain of Vg1 (LvH1) [63] (Fig. 3).
|
The C-terminus of YP 103 was predicted to lie near the N-terminal residue obtained from YP 19 (Glu 965), suggesting that YP 103 and YP 19 result from the cleavage of YP 122. The estimated juncture between YP 103 and YP 19 lies at the midpoint of a predicted proteolytic site (PEST site; residues 952–974), receiving a score of +6.31 (Table 2). A score greater than zero denotes a possible PEST region, and a value greater than +5.0 is considered a site [57]. The purported cleavage site bisects the predicted PEST site, leaving the two resulting protein sequences with termini that do not surpass the cutoff value for valid PEST sites (Fig. 2). Thus, although YP 122 contains a PEST site, neither of its cleavage products, YP 103 and YP 19, do. Potential PEST sites were further investigated in fish and other vertebrate Vg full-length cDNA sequences available in GenBank (Table 2). In chicken [62], both Vg1 and Vg2 show two PEST sites each with a high score (
+9.41), while Xenopus laevis VgA2 [63], but not VgB1 [64], contains a sequence with a score (+4.81) very close to the cutoff value of +5.0. In fish, in addition to F. heteroclitus Vg1, PEST sites within Vg sequences with a score close or greater than +5.0 were found in Cypriniformes (Cyprinus carpio Vg and Danio rerio Vg [65]), Beloniformes (Oryzias latipes Vg1), and Petromyzontiformes (Ichthyomyzon unicuspis Vg [66]), which are freshwater or marine benthophil species. It is interesting that PEST sites occurred roughly in the same location along the Vg amino acid sequences of amphibians, aves, and fish. The Vgs of more advanced pelagophil teleosts, however, such as Perciformes (Oreochromis aureus Vg, Sillago japonica Vg, Acanthogobious flavimanus Vg [26]), Gadiformes (Melanogrammus aeglefinus Vg1 [VgA] and Vg2 [VgB] [19]), and Anguiliformes (Anguilla japonica Vg1 and Vg2), do not contain PEST sites. However, Vg from some benthophil species belonging to these orders, such as Japanese common goby (A. flavimanus), as well as from more primitive benthophil fish, such as trout (Oncorhynchus mykiss Vg [67]) and sturgeon (Acipenser transmontanus Vg [68]), show sequences with a score above zero, indicating potential PEST regions.
|
No attempt was made to obtain internal sequences for F. heteroclitus YP 77 and YP 75 due to limited material. Although we expect that the polyserine repeats present in both Vg1 and Vg2 are processed into true phosvitins (in which most of the serines are phosphorylated), we were unable to verify this by N-terminal sequencing. The highly negative charge of phosvitins prevents staining with Coomassie blue [69, 70]; we have found that it also apparently prevents their adherence to PVDF membranes used for microsequencing. Schematic models summarizing our positive data are provided in Figure 3.
Time Course of YP 122 Degradation and Cathepsin Enzyme Activity During Oocyte Maturation
To investigate the potential proteases involved in YP 122 degradation, we first determined the enzyme activities of the lysosomal cysteine proteinases CatB and CatL in follicle-enclosed oocytes undergoing meiotic maturation in vivo (Fig. 4). During maturation, F. heteroclitus oocytes increase in volume as a result of hydration (Fig. 4A, compare stage 1 and 4), and a typical coalescence of a high number of oil droplets from the periphery toward one pole of the oocyte is observed ([71]; Fig. 4A, stage 4). At the same time, GVBD occurs (Fig. 4A, stage 3,) and oocytes become more translucent with respect to prematurational oocytes (Fig. 4A, stage 4). The activity of cathepsin enzymes in maturing follicle-enclosed oocytes in vivo was determined after manual dissection from the ovary of spawning females. These experiments indicate that CatL activity progressively decreased (P < 0.01) as maturation proceeded (Fig. 4B), whereas CatB activity transiently increased (P < 0.01) at the time of GVBD and beginning of oil droplet coalescence, abruptly decreasing its activity thereafter when oocytes reached the maturation and ovulation stages (Fig. 4B).
|
To confirm the possible correlation between cathepsin enzyme activity and the degradation of YP 122, a series of experiments were carried out in which both SDS-PAGE analysis of YP 122 and CatL and CatB activities were determined in ovarian follicles undergoing meiosis reinitiation in vitro. Groups of fully grown (postvitellogenic) follicles were dissected from the ovary and incubated in the presence or absence of 0.1 µg/ml of 17,20ßP in ethanol up to 48 h. As oocytes underwent GVBD, individual follicles were separated and frozen, and the degradation of the YP 122 was determined in each follicle by SDS-PAGE followed by Coomassie blue staining. In both control and MIS-stimulated follicles, the YP 122 appeared to be degraded progressively with culture time, in agreement with that previously described [61] (Fig. 5, A and B). However, the time course of YP proteolysis revealed that in steroid-treated follicles the hydrolysis of YP 122 occurred faster and was more pronounced (P < 0.001), resulting in its complete disappearance in SDS-PAGE at 36 h of culture time (Fig. 5, A and B). In ethanol-treated follicles, however, the YP 122 was still clearly visible at 48 h (Fig. 5, A and B). A similar pattern of degradation of the YP 45 was found between ethanol- and 17,20ßP-treated follicles (data not shown). Therefore, these data indicated that MIS indeed stimulated the proteolysis of YPs in vitro along with the reinitiation of meiosis, although some degradation of YPs also occurred in the absence of steroid.
|
During the in vitro experiments described above, groups of follicles were also sampled for CatL and CatB activity at the same time points that follicles were chosen for SDS-PAGE analysis. In agreement with the in vivo data, CatL enzyme activity in MIS-treated oocytes decreased markedly (P < 0.001) during oocyte maturation, whereas in control follicles its activity remained approximately constant during the same culture period (Fig. 5C). However, the activity of CatB in maturing oocytes progressively increased up to the time of GVBD (Fig. 5C), in a close positive correlation with the degradation of YP 122 detected in these follicles (Fig. 5B). Control follicles also showed an increased activity of CatB throughout culture time, although, interestingly, they reached lower levels than those found in MIS-treated follicles (P < 0.001). Thus, the CatB activity determined in control follicles also appeared to be associated with a limited degradation of YP 122 (Fig. 5C).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the schematic representation that summarizes the data for F. heteroclitus (Fig. 3), the N-terminal positions of the various YPs were drawn to scale based on their absolute positions along the primary Vg sequences. The lengths for each YP, however, were based on size estimates derived from SDS-PAGE gradient gels. Although this procedure has been found to provide reasonable estimates for the sizes of highly modified YPs [72], the estimates are subject to the normal errors inherent in such methods, and the corresponding lengths drawn along the Vg sequences are enhanced by any posttranslational modifications present in the YPs. Therefore, the positions of the various C-termini should be considered tentative. Most likely, C-termini for YP 122 and YP 19 lie near the N-terminus of phosvitin (polyserine domain), while that for YP 26 lies near the C-terminus of Vg. A major factor that prevents the construction of a definitive map accounting for all YPs derived from the Vgs is the difficulty in microsequencing the phosphopeptides (phosvitins, phosvettes) derived from the polyserine domain. Although we expect that the polyserine repeats represented in both Vg1 and Vg2 cDNAs are processed into true phosvitins and phosvettes, we have been unable to verify this by N-terminal sequencing. Because phophopeptides can be visualized with a cationic dye (Stains-all), they have been documented in F. heteroclitus as 31-kDa (phosvitin) and 14-kDa (phosvette) bands in prematurational oocytes, with at least four smaller bands of approximately 24, 17, 10, and 8 kDa appearing in preparations from ovulated eggs [27].
We had initially assumed that YP 122 and YP 103, the most prominent YPs in extracts from postvitellogenic oocytes, were derived separately from two different Vgs, but internal sequences indicated that both YPs originate from LvH I (Fig. 3). We can thus conclude that Vg1 is truly the major YP precursor in F. heteroclitus. Interestingly, two subunits corresponding to LvH apparently derived from Vg1 have been found in vitellogenic ovaries of another benthophil teleost, the white perch (Morone americana) [10]. Our evidence also indicates that during the course of Vg1 processing into YPs, a proteolytic event that might be mediated by the lysosomal aspartic protease cathepsin D (CatD) as reported for other lower vertebrates [37, 73–75], generates approximately equimolar amounts of YP 122 and YP 109+YP 19. Thus, in addition to cleavage near residue 1080, approximately half of the incoming Vg1 sustains an additional scission between residues 954–955 by an alternative proteolytic process that remains to be defined. Similarly, YP 45 and YP 26 also overlap each other and both are found in the oocyte after the initial proteolytic processing of Vg1, so each apparently represents a proteolytic variant derived from the same precursor. Whether YP 122 is proteolytically generated together with YP 45, YP 26, or both remains to be defined. Another interesting aspect of the data are that the C-terminus of YP 45 is approximately 9 kDa shy of the Vg C-terminus. Thus, a 9-kDa piece of Vg appears to be cleaved from the C-terminus when YP 45 is generated and is unaccounted for. A similar situation has been found for the processing of X. laevis Vg [15].
One of the most pronounced changes that occurs in F. heteroclitus YPs is the complete disappearance of the LvH1 122 (YP 122) during the transformation of oocytes into mature, ovulated eggs [27–29, 62]. The proteolysis of LvH1 122 appears to be accelerated during MIS-induced oocyte maturation and hydration in vitro, although limited processing of this yolk product is observed in control follicles, which do not undergo maturation and marked growth due to hydration ([62]; Fig. 5A). Lysosomal cysteine proteinases, particularly CatL and CatB, as well as CatD, seem to be the major degradative enzymes in the yolk of invertebrate and vertebrate embryos [76–80], and available evidence might indicate a similar scenario for teleosts [74, 81– 84]. During teleost oocyte maturation, however, much less is known about the proteases involved in the proteolysis of YPs. Recent studies by Carnevali et al. [37] in the gilthead seabream (Sparus aurata), a pelagophil teleost, have found high levels of CatD and CatB activities in isolated early vitellogenic ovarian follicles, while the activity of CatL is higher than that of CatB and CatD in follicles at midvitellogenesis stage. Isolated ovarian CatL was able to cleave purified LvH in vitro under appropriate conditions, and thus the authors suggested that this protease is most likely the enzyme involved in the oocyte maturation-associated proteolysis of LvH [37]. These findings might agree with the high enzyme activity of CatL observed in seabream buoyant and fertilized eggs [85], and with other studies in trout showing high CatL mRNA levels in late vitellogenic oocytes [86]. However, in the pelagophil barfin flounder (Verasper moseri) CatB-like proteases have been suggested to be involved in the maturation-associated proteolysis of YPs [87]. In F. heteroclitus, the enzymatic activity of CatL dramatically decreases during oocyte maturation both in vivo and in vitro (Figs. 4B and 5C), even though its mRNA increases approximately 3-fold during MIS-induced meiotic maturation [50]. By contrast, the activity of CatB transiently increased during maturation, coincident with the time of maximum degradation of LvH1 122. Our observations thus suggest that CatB, rather than CatL, is the major enzyme involved in the proteolytic processing of LvH1 122 in F. heteroclitus.
The results reported here on YP processing in F. heteroclitus are in contrast to those described for pelagophil marine teleosts, such as barfin flounder [9] and haddock (Melanogrammus aeglefinus [19]). In these species, unlike in F. heteroclitus, LvH derived from Vg1 (VgA) and Vg2 (VgB) are equally represented in the pool of yolk products in postvitellogenic oocytes. During oocyte maturation, LvH1, together with phosvitins and ß'-components (ß'-Cs) derived from both Vgs, are extensively cleaved into FAAs, which are believed to generate the osmotic pressure for oocyte hydration, while dimeric LvH2 is dissociated into a monomer with a slightly reduced size that is stored for further embryonic development [9, 19, 88]. In F. heteroclitus, LvH2 (referred here as YP 69) is also mostly unaltered during oocyte maturation, and together with LvH1 103, is presumably stored for developing embryos. These different models for YP processing between pelagophil and benthophil teleosts are summarized in Figure 6, in which we constructed a flow chart describing the molecular alterations of Vg1 and Vg2 and derived YPs during vitellogenesis and oocyte maturation based on the present and previously published data [8, 9]. These models, however, may be more complicated, because in some teleost species the existence of a third phosvitinless Vg, which is incorporated by the oocyte and not cleaved into YPs, has been demonstrated [87], although there is presently no evidence for such Vg in barfin flounder, haddock, or F. heteroclitus. In any case, the causes for the apparent divergent mechanisms of YP processing during oocyte maturation between pelagophil and benthophil teleosts, which might involve the activation of different enzymatic machineries [37], remain intriguing. These may be related with the specific role or roles that the proteolysis of YPs is believed to play during oocyte hydration in these species. In pelagophil teleosts, FAAs released from the proteolysis of LvH1 provide most of the osmotic potential that drives water uptake into the oocyte [9, 19, 32, 42], whereas in F. heteroclitus, the hydration of the oocyte is primarily due to the influx of K+ ions, possibly via heterologous gap junctions [44, 45, 89]. Thus, a limited proteolysis of YPs when compared with pelagophil oocytes might occur in benthophil oocytes, which may otherwise be used to fulfill the demands of FAAs and small peptides for energy production and protein synthesis during embryo development.
|
The mechanisms for the differential processing of LvH1 122 and 103 in F. heteroclitus, as well as of LvH1 and LvH2 in pelagophil hydrating oocytes, are completely unknown. Differences in the amino acid sequence, secondary structure, or intracellular compartmentation of the LvHs within the same or separated yolk globules, might allow the protease or proteases apparently activated by the MIS to distinguish between LvHs [19]. In F. heteroclitus, a possible cause of the rapid and rather selective proteolysis of the LvH1 122 is the occurrence of a PEST site within its C-terminal tail. The proteolytically resistant LvH1 103 is identical to LvH1 122, except that it lacks the terminal domain where the PEST site occurs. PEST sites were initially defined as a conserved clustering of amino acids that occurs in cytoplasmic proteins known to be rapidly degraded [57, 90]. Common to all PEST sites, which are flanked by positively charged residues, are high local concentrations of Pro, Glu, Ser, Thr, or a combination of these. The LvH1 122, once generated from Vg1, is stable until oocytes undergo final maturation, so presumably the proteolysis-triggering PEST site becomes exposed to the appropriate protease only at this time due to a change in intracellular compartmentation. Of interest, PEST sites are mostly found in Vg sequences from Cypriniformes, Beloniformes, and Petromyzontiformes, which include benthophil fish species as F. heteroclitus (Table 2). By contrast, PEST sites are missing in Vg sequences from pelagophil teleosts, which might indicate the existence of a different proteolytic mechanism in these species. resulting in the complete degradation, unlike in benthophil fish, of the LvH1 yolk products. It thus appears that the MIS-activated specific intracellular mechanisms regulating the degradation of specific YPs during oocyte maturation need to be further investigated if we are to understand the underlying events leading to the hydration of fish oocytes and the maternal nutrient contribution to developing embryos.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Joan Cerdà, IRTA Lab, Room B46, CIMIMA-CSIC, Passeig Marítim, 37-49, 08003-Barcelona, Spain. FAX: 34 93 2309555; jcerda{at}icm.csic.es
Received: 28 February 2005.
First decision: 23 March 2005.
Accepted: 24 May 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
