Purification, characterization and immunoassay of ... - Springer Link

Fish Physiology and Biochemistry vol. 12 no. 1 pp 31-46 (1993) Kugler Publications, Amsterdam/New York

Purification, characterization and immunoassay of striped bass (Morone saxatilis) vitellogenin Yunxia Taol, Akihiko Hara 2, Ronald G. Hodson l, L. Curry Woods III3 and Craig V. Sullivan IDepartment of Zoology, North CarolinaState University, Raleigh, North Carolina27695, U.S.A.; 2 Nanae Fish Culture Experiment Station, Faculty of Fisheries, Hokkaido University, Nanae, Kamedagun, Hokkaido 041-11, Japan; 3 Maryland AgriculturalExperiment Station, Crane Aquaculture Facility, P.O. Box 1475, Baltimore Maryland 21203, U.S.A. Accepted: January 9, 1993 Keywords: striped bass, reproduction, vitellogenin, vitellogenesis Abstract The egg yolk precursor, vitellogenin (VTG), was purified from blood plasma of striped bass by chromatography on hydroxylapatite or DEAE-agarose. The fish were first implanted with estradiol-173 (E2), which induced vitellogenesis. A rabbit antiserum (a-FSPP) raised against plasma from mature female striped bass, and then adsorbed with mature male plasma, was used to detect female-specific plasma protein (FSPP) in the chromatography fractions. Striped bass VTG (s-VTG) was collected from the peak fraction that was induced by E2, reacted with a-FSPP, and contained all detectable phosphoprotein. It appeared as a single band (Mr = 170,000) in SDS-PAGE or Western blots using a-FSPP, and as a pair of closely-spaced phospholipoprotein bands in native gradient-PAGE, suggesting that there is more than one circulating form of s-VTG. The relationship of s-VTG to the yolk proteins was verified using a-FSPP. The antiserum reacted with the main peak from gel filtration of saline ovary extracts, and it specifically immunostained the two main bands in Western blots of the extracts and the yolk granules of mature oocytes. The amino acid composition of s-VTG was similar to that of VTG from other fish and Xenopus. A radial immunodiffusion assay for sVTG was developed using a-FSPP and purified s-VTG as standard. The s-VTG was not detected in blood plasma of males, immature females, or regressed adult females, but plasma s-VTG levels were highly correlated with plasma E 2 and testosterone levels, and oocyte growth, in maturing females. The results indicate that the maturational status of female striped bass can be identified by s-VTG immunoassay.

Introduction Striped bass (Morone saxatilis) and their hybrids (genus Morone) support important commercial and sport fisheries in the United States. They are widely cultured to restock fisheries and for market (Hodson et al. 1987). Researchers have just begun to develop and reproduce captive striped bass broodstocks (Blythe 1992; Woods et al. 1989, 1992;

Hodson and Sullivan 1993; Woods and Sullivan 1993), and they need a reliable means to identify the maturational status of females. The presence of vitellogenin in the blood of female fish has promise as a marker of the onset and course of maturation. Vitellogenin (VTG) is synthesized by the liver and secreted into the blood from where it is taken up by the maturing oocytes to be processed into egg yolk, accounting for most

Correspondence to: Dr. Craig V. Sullivan, Campus Box 7617, Department of Zoology, North Carolina State University, Raleigh, NC 27695, U.S.A.: Telephone (919) 515-7186; Telefacs (919) 515-5327.

32 oocyte growth (Byrne et al. 1989; Selman and Wallace 1989). Hepatic vitellogenesis is triggered in fishes by estradiol-173 (E 2) (Mommsen and Walsh 1988). In female striped bass, oocyte growth (vitellogenesis) proceeds, and blood levels of E2 are elevated, for many months before spawning (Berlinsky and Specker 1991; Blythe 1992; Woods and Sullivan 1993). The objectives of this study were to: 1) purify striped bass vitellogenin (s-VTG), 2) develop an antiserum and immunoassay for measuring it in blood plasma, and 3) evaluate the potential of the assay for assessing maturation of female striped bass. Another objective was to obtain information on the biochemical characteristics of s-VTG. Biochemical knowledge of fish vitellogenins is developed mainly for salmonids (Tyler 1991) and very limited for more advanced teleosts like striped bass (order Perciformes). We purified s-VTG and developed an immunoassay for it. Our results show that plasma s-VTG concentrations increase continuously during ovarian maturation of striped bass, and it was possible to identify maturing (vitellogenic) females several months before the spawning season by s-VTG assay. The biochemical data indicate that s-VTG is based on a - 170 kDa polypeptide, may circulate as at least 2 native lipophosphoproteins, is immunologically related to the yolk proteins, and has an amino acid composition similar to other vertebrate vitellogenins. Some of these results were previously reported in abstract (Sullivan et al. 1991).

Materials and methods

and Sullivan 1993), immediately mixed with aprotinin (Sigma; I T.I.U.-ml- 1 blood), and centrifuged at 1000 x g for 5 min to separate the plasma which was either directly used for s-VTG purification, or frozen on dry ice and stored at - 70 0C for later use. Ovaries were excised from mature wild fish, immediately frozen in liquid nitrogen, and stored at - 70°C until used to prepare saline ovary extracts (Hara 1987).

Preparationof antiserum An antiserum against blood plasma proteins of mature female striped bass was prepared by immunizing a rabbit with pooled plasma samples obtained from several wild fish, and it was targeted at the female-specific plasma proteins (FSPP) by adsorption with an equal volume of plasma pooled from several mature males (Hara 1987). The adsorbed antiserum was designated anti-FSPP (a-FSPP).

Induction of vitellogenesis Vitellogenesis was induced in 3-year-old immature fish of both sexes (mean weight 1.6 0.2 kg) and in 5-year-old adult males (mean weight 3.3 0.5 kg). Lengths of silastic tubing were sealed at one end, filled with 15 mg of E 2 (Sigma), and implanted into the dorsal musculature of each fish. At 2-4 week intervals, the fish were anesthetized (11 ppm quinaldine sulfate) and injected intraperitoneally with E2 (5 mg-kg - 1 body weight) prepared as a 50 mg-ml-1 stock solution in propylene glycol. After 2-3 injections, their blood plasma was harvested for vitellogenin purification.

Sample collection The striped bass were domestic broodstock reared at the Pamlico Aquaculture Center (PAC) or the Crane Aquaculture Facility (CAF) as described previously (Woods et al. 1989, 1992; Hodson and Sullivan 1993), or wild fish captured in commercial pound nets on their spawning migration up the Roanoke River (North Carolina). Blood samples were obtained from their caudal vessels (Woods

Vitellogenin purification In preliminary studies, s-VTG was purified as described by Hara (1987), from 2 ml of plasma by chromatography on a 2.6 x 13 cm column of hydroxylapatite (Bio-Rad) equilibrated with 0.2 M potassium phosphate buffer (pH 6.8). Chromatography was performed at 4C with a flow rate of

33 42 ml-hr - l. Proteins were eluted stepwise with 75 ml each of 0.2 M, 0.4 M, 0.6 M, 1.2 M, and 2.0 M potassium phosphate buffer (pH 6.8) and 5.0 ml fractions were collected. The s-VTG was later routinely purified as described by Kishida et al. (1992), from 1 ml of blood plasma of E 2-treated fish by anion exchange chromatography performed on 2.6 x 24.5 cm column of DEAE-agarose (Bio-Rad). The column was first eluted with 25 mM Tris-HCI buffer containing 0.07 M NaCl (pH 7.5). It was then eluted with a gradient from 0.07 M to 0.5 M NaCl (150 ml:150 ml, v/v) in the same buffer and 5 ml fractions were collected. Chromatography was performed at 4°C with a flow rate of 40 ml-hr- 1 . Plasma samples were always dialyzed overnight at 0°C against the starting buffer before chromatography, and aprotinin was added to the chromatography buffers at a concentration of 0.04 T.I.U..ml-1.

Polyacrylamide gel electrophoresis

Vitellogenin detection

Proteins separated in SDS-PAGE gels were transferred to nitrocellulose membranes in transfer buffer (50 mM Tris, 380 mM glycine, 0.1% SDS, 20% methanol, pH 8.3) using a PhastTransferTM Semi-dry Transfer Kit (Pharmacia Inc.) fitted to the PAGE unit. Non-specific binding sites were blocked by incubating the membranes in a solution of 5% normal goat serum and 0.02% NaN 3 in PBS (0.14 M NaC1, 3 mM KCI, 10 mM Na 2 HPO4 , 2 mM KH 2PO 4, pH 7.2) overnight with agitation. Then membranes were washed twice for 5 min in PBS, incubated with agitation for 2 h in primary antibody solution (a-FSPP diluted 1:100 v/v in 5% normal goat serum in PBS) at 370 C, and washed 4 times for 5 min each in PBS. They were then incubated for 1 h at 37C in a secondary antibody solution of goat anti-rabbit immuno-y-globulin conjugated to horseradish peroxidase (Sigma) diluted 1:2000 v/v in PBS containing 5% normal goat serum. Antigen-antibody complexes were visualized by the peroxidase reaction using chloronaphthol reagent (Harlow and Lane 1988).

The a-FSSP antiserum was used in a single radial immunodiffusion assay (Mancini et al. 1965) to monitor FSPP in the chromatography fractions. The gel contained 1.0% a-FSPP and precipitation rings were measured to the nearest 0.01 mm using electronic digital calipers (Mitutoyo). Fractions were also assayed for trichloroacetic acid (TCA)precipitable phosphoprotein phosphorus using a procedure modified from Gamst and Try (1980). Three ml of each fraction were mixed with 1 ml of 48% TCA, incubated overnight at 4°C, and centrifuged for 30 min at 1000 x g. The precipitated sample was washed twice in ethanol:ethyl ether (3:1, v/v) solution and once in 2 ml of ethyl ether, and then dried overnight at 20 0C. After addition of 0.2 ml of 2 N NaOH, the samples were heated to 100°C for 15 min and cooled to 200 C for 15 min before 0.2 ml of 2 N HCI was added to neutralize each sample. Sample solution (0.3 ml) was then mixed with 0.3 ml of molybdate solution (2.4 mM ammonium molybdate in 0.5 M sulfuric acid and 1% Triton X-100) followed by 0.4 ml of citric acid solution (0.2 M citric acid in 0.5 M sulfuric acid and 1% Triton X-100), and the absorbance of the mixture at 366 nm was recorded.

Blood plasma, ovary extracts, and purified s-VTG were submitted to sodium dodecyl sulfate-4-15% polyacrylamide gradient gel electrophoresis (SDSPAGE), or to native gradient-PAGE, on a Phast SystemTM automated electrophoresis unit (Pharmacia) using the manufacturers recommended sample preparation and run conditions (Phast SystemTM Separation Technique File No. 130). Gels were routinely stained with Coomassie brilliant blue (Pharmacia; Phast SystemTM Development Technique File No. 200). Native gradient-PAGE gels were sometimes stained for phosphoprotein with methyl green as described by John and Thomas (1973) or for lipoprotein with Sudan black B as described by Prat et al. (1969). Western blots

Size-exclusion chromatography One ml of purified s-VTG solution (- 1 mg-ml- 1) was added to a 1.6 x 68 cm gel filtration column

34 of Sepharose 6B (Bio-Rad) equilibrated with 20 mM Tris-HCI buffer (pH 8.0) containing 2% NaCI, 0.1% NaN 3, and 0.04 T.I.U. aprotinin ml- l . The s-VTG samples were previously dialyzed overnight at 0°C against the starting buffer. Chromatography was performed at 4C with a flow rate of 15 ml-hr - 1 and 2.5 ml fractions were collected. Molecular mass markers (Pharmacia) were aldolase (158 kDa), catalase (232 kDa), ferritin (440 kDa), and thyroglobulin (669 kDa).

Amino acid composition The amino acid composition of purified s-VTG was evaluated on a Beckman Model 6300 amino acid analyzer. Ninety gl of purified s-VTG solution (1 mg-ml-1) was dried under vacuum in a hydrolysis tube. After adding 6 N HCl, the tube was sealed under vacuum and heated at 100°C for 20 h. The solution was lyophilized and redissolved in Beckman dilution buffer, the ninhydrin-derivatized amino acids were detected at 570 nm and 440 nm, and the peaks were integrated for quantification. Approximately 1.80-1 of the total s-VTG sample was analyzed in triplicate and the average percentage compositions were recorded.

Ovarian histology Ovarian tissue was obtained by biopsy (Rees and Harrell 1990) and fixed in Bouin's solution at 0° C for up to 24 h. Samples were washed several times in tap water, dehydrated in graded ethanol series, cleared in methyl salicylate, infiltrated with and embedded in paraffin (m.p. 55-57°C), sectioned at 10 itm, stained with hematoxylin and eosin, and counterstained with periodic acid Schiff's reagent (PAS). Oocyte diameters were measured using a microscope fitted with a calibrated ocular micrometer, and the oocytes were classified as primary growth, early secondary growth, or vitellogenic using the criteria of Berlinsky and Specker (1991). Another category of oocytes was classified as mature based upon their large size ( 750 jim) and the close proximity of the donor females to the

spawning season (April) when they were sampled (March). Immunocytochemistry Ovary sections were immunostained with a-FSPP using a Vectastain ABC kit and a Vectastain DAB substrate kit for horseradish peroxidase (Vector Laboratories) as described by Nozaki et al. (1988). Primary antibody solution was prepared by diluting a-FSPP 200 x in a 1:1 (v/v) mixture of 1 o BSA in PBS and immature male striped bass liver extract, followed by addition of 0.05 mg/ml of rat liver powder (Sigma) and vortex mixing, incubating the mixture for 2 h at 4°C, centrifuging it at 2,500 x g for 20 min, and collecting the supernatant as the primary antibody solution. Liver extract was prepared by homogenizing liver samples pooled from several immature male striped bass in 4 volumes of PBS at 0°C with a motor-driven Teflon pestle in a Potter-Elvehjem homogenizer (4 strokes, clearance 0.10-0.15 mm), centrifuging the homogenate at 3,000 x g at 0-4°C for 30 min, and collecting the supernatant as the final extract. Assay of blood plasma vitellogenin, E2, and testosterone s-VTG was assayed in 5 tl of sample or standard solution by single radial immunodiffusion using aFSPP and purified s-VTG standards calibrated by Bio-Rad Protein assay (Bio Rad Inc.) with bovine serum albumin (Sigma Chemical Co.) as the reference protein. Standard s-VTG concentrations ranged from 0 to 180 g.-ml- 1 in assay buffer (0.5% bovine y-globulin in 10 mM Tris-HCI buffer containing 0.15 M NaCl, pH 7.5). Plasma was diluted 1:4 in assay buffer before analysis. Plasma levels of E2 and testosterone (T) were measured in highly specific radioimmunoassays (Woods and Sullivan 1993). Statistical analyses Data on blood plasma s-VTG and hormone concentrations in maturing females was submitted to

35 o

E C

0.3

A 0.2 St.rt NaCI Gr4di.nt

o01-

o 00

I100

ZOO

300

400

500

0

E c

0.2

C

0.1

4

d e~

to

D

>'

0.0 0

100

, 200

, 300

. 400

500

Ec

E

2.0

E

C

1.5 1.0

ID

0.5

o 0

1 0

200

300

400

S0

Volume (ml)

Fig. 1. Purification of vitellogenin from 1 ml of blood plasma of striped bass by ion exhange chromatography on a DEAE agarose column. The gradient from 0.07 to 0.50 M NaCl (150 ml:150 ml, v/v) began at the position indicated by the arrow (panel A). The elution profile of blood plasma sampled from a male striped bass before (panel A) and two weeks after (panel B) it was implanted with estradiol-17/3 is shown (open circles). The immunoreactivity of the fractions to the antiserum (a-FSPP) raised against mature female-specific plasma proteins (panel B, closed circles) and the relative level of TCA-precipitable phosphoprotein phosphorus (panel C) are also indicated.

repeated measures analyses of variance (ANOVA) and differences in mean concentrations between sampling times were identified by least squares means comparison. The level of significance for all statistical tests was p _ 0.05.

Results s-VTG was purified from the blood plasma of E2 -treated striped bass by anion exchange on DEAE-agarose. Figure 1 shows the elution pattern on DEAE-agarose of plasma from an immature male striped bass before and two weeks after it was treated with E2 . A prominent peak eluting at

Fig. 2. Pattern in SDS 4-15% gradient PAGE of blood plasma from a male striped bass before implantation (lane 1; 8 g protein) and two weeks after it was implanted (lane 3; 16 g protein) with estradiol-17/3 (E2 ), and the corresponding Western blots for these samples (lanes 2 and 4, respectively). The electrophoretic pattern of 2 y/g of purified stiped bass vitellogenin (sVTG) is also shown (lane 5). It was collected as the central fraction of the E2-induced, female-specific, phosphoprotein peak seen when these same plasma samples were subjected to anion exchange chromatography on DEAE agarose (Fig. 1). Numbers on the right side of the figure indicate the position of the molecular mass markers (Pharmacia): myosin (212 kDa), a2-macroglobulin (170 kDa), /-galactosidase (116 kDa), transferrin (76 kDa) and glutamic dehydrogenase (53 kDa).

- 380 ml was initially absent (Fig. la), but was induced by E 2-treatment, reacted positively to the antiserum (a-FSPP) against female-specific plasma proteins (Fig. lb), and contained all of the detectable TCA-precipitable phosphoprotein phosphorus (Fig. c). This result was typical of pooled plasma samples used for routine s-VTG purification. SDSPAGE of the plasma from this same fish (Fig. 2) revealed a prominent band (Mr = 170,000) that was initially absent (lane 1) but induced by E2 (lane 3), was specifically recognized by a-FSPP in Western blots (lane 4 versus lane 2), and corresponded in position to the single band produced by the peak E 2-induced, female-specific, phosphoprotein fraction from anion exchange chromatography (lane 5), suggesting that this chromatography fraction contained purified s-VTG with a primary - 170 kDa polypeptide subunit. Another faint band (Mr = 110,000) appeared to be induced by E2 treatment (Fig. 2, lane 3). In

36

Fig. 3. Pattern in native 4-15% gradient PAGE of s-VTG. The electrophoretic pattern of s-VTG purified by anion exhange on DEAE agarose is shown before (lane 1; 4 ug protein) and after (lane 2; 4 Ig protein) it was frozen to - 700 C. Also shown is the pattern of vitellogenin purified by affinity chromatography on a column of hydroxylapatite before (lane 3; 8 g protein) and after (lane 4; 8 g protein) it was frozen. Numbers on the right side of the figure indicate the position of the molecular mass markers (Pharmacia): thyroglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), lactate dehydrogenase (140 kDa) and bovine serum albumin (67 kDa).

preliminary experiments, when purification procedures were done at room temperature without aprotinin, it was sometimes the major E 2-induced band and could be detected, albeit weakly, in the Western blots. We consider this band to represent a degradation product of s-VTG, formation of which can be avoided by use of protease inhibitors and low temperature. The prominent band (Mr 140,000) observed after this fish was treated with E 2 was present to a lesser degree before it was treated with E2 , was not routinely induced in other fish by E2 -treatment, and was not recognized by aFSPP in Western blots, indicating that it represents a protein unrelated to s-VTG. The s-VTG was also purified from plasma of E2 -treated fish by chromatography on hydroxylapatite. Plasma was fractionated into four peaks on a the hydroxylapatite column with stepwise elution. Most proteins (O.D. 280) were present in a major peak eluted by 0.2 M potassium phosphate, a smaller broad peak was eluted by 0.6 M potassium phosphate, and progressively larger, very sharp

peaks were eluted by 1.2 and 2.0 M potassium phosphate, respectively. In SDS-PAGE, this fourth peak appeared to be s-VTG, the single E 2-induced band (Mr = 170,000) specifically recognized by aFSPP in Western blots. However, s-VTG purified on hydroxylapatite showed a strong tendency to precipitate from solution and was very difficult to resolubilize, so we abandoned the method in favor of anion exchange on DEAE-agarose which eliminated the precipitation problems. The s-VTG, freshly purified on either hydroxylapatite or DEAE-agarose, appeared in native gradient-PAGE as a pair of closely spaced bands (Mr 340,000; Fig. 3, lane 1) judged to be phospholipoproteins because they stained positively for phosphorus using methyl-green and for lipid using Sudan black-B (data not shown). After being frozen at - 700 C for from 1 to several months, the s-VTG samples showed a very faint additional band (Mr = 600,000; lane 2), suggesting that freezing leads to formation or reformation of a higher molecular weight aggregate of the molecule. The two s-VTG bands were difficult to resolve when high concentrations of s-VTG ( 5 mg/ml) were subjected to native gradient-PAGE (lane 3), and concentrated s-VTG samples yielded several bands of even higher apparent molecular weight after freezing (lane 4), showing the strong tendency of the molecule to form aggregates. To avoid aggregation, we routinely used freshly purified s-VTG at concentrations 1 mg-ml - '. Size exclusion chromatography of purified sVTG on Sepharose 6B revealed a single broad symmetrical peak that reacted with the a-FSPP (Fig. 4). In repeated trials (N = 5), the average estimated molecular weight of the peak fraction was 600,000 (range 550,000 to 640,000). Two minor peaks, one eluting near the Vo and the other near the Vt, were also sometimes present but showed low or no reactivity with a-FSPP. The peak eluting near the Vt was generally not seen in freshly purified s-VTG samples, but appeared sometimes at low levels in frozen samples, suggesting that it results from slight degradation of the s-VTG molecule during freezing and thawing. Likewise, the peak eluting near the Vo was present in varying degrees. It was very prominent in highly concentrated samples of

37 YV

Vo ±

232

6691 440 1158 't+ 't '

0.4

Vt

0.30.2 C0

C4

d

0.1-

=,.

-4

O

4

-2

1 ff~`· . v . . 20

-12 -10 -8 -6

. 40

Volume

..

--- -

60

80

.

----

1 00

I I

°

--

--

12

0

E

(ml)

Fig. 4. Estimation of the molecular weight of native striped bass vitellogenin by gel filtration on a column of Sepharose 6B. The arrows indicate the position and mass of the molecular weight markers and vitellogenin (Vg), or the void volume (V0 ) and total volume (Vt) of the column. Also shown is the immunoreactivity of the fractions to the antiserum (a-FSPP) against female-specific plasma proteins of striped bass.

s-VTG (> 5 mg-ml- 1) after they were frozen, and reacted only slightly to a-FSPP under these conditions, but it was greatly diminished or absent in samples freshly purified on DEAE-agarose and maintained at concentrations < 1 mg-mg-l. This peak apparently results from aggregation of s-VTG with concomitant loss of relative immunoreactivity. Purified s-VTG was immunologically related to ovarian proteins using a-FSPP. Striped bass oocytes that we classified as mature are similar to the 'tertiary yolk granule stage' oocytes of the closely related European sea bass (Dicentrarchuslabrax) (Mayer et al. 1988). Egg yolk is present both in the membrane-bound yolk globules and as a non-membrane-bound amorphous mass in the outer cortex. In immunocytochemistry of such oocytes, a-FSPP strongly stained the yolk globules and stained the amorphous egg yolk with less intensity, but it did not react with other oocyte components or areas of the sections including the cortical alveoli, lipid droplets, chorion, follicle cells and extracellular space (Fig. 5). Sections of ovaries from immature females did not react at all to a-FSPP, nor was any immunostaining observed in sections of ovaries from mature females when normal rabbit serum was substituted for a-FSPP in the primary antibody solution, or when a-FSPP was preadsorbed with sVTG before immunostaining. The a-FSPP also reacted strongly and specifically with the main broad peak (Mr = 380,000) in the elution pattern

Fig. 5. Immunostaining pattern of sections of mature oocytes of striped bass using the antiserum (a-FSPP) against mature female-specific plasma proteins of striped bass. The yolk granules (G) and amorphous yolk (Y) show positive immunoreactivity, whereas the cortical alveoli (A), follicle cells (F), chorion (C), extracellular space (E) and lipid droplets (L) do not (see Results). Horizontal bar = 15 m.

of saline ovary extracts subjected to gel filtration on Sepharose 6B (data not shown). In Western blots of the extracts (Fig. 6), the s-VTG band was absent but a-FSPP reacted specifically with the two main bands (Mr = 94,000 and 86,000; lane 1) seen in the electrophoretic pattern of extracts of mature

38

Fig. 6. Western blot (lane 1) and corresponding electrophoretic pattern (lane 2; 15 Ag protein) of saline extracts of a mature ovary from a female striped bass, and the electrophoretic pattern of a saline extract of an ovary from an immature female (lane 3; 15 ug protein), after SDS 4-15°70 gradient PAGE. No immunostaining was observed in the Western blot of the electrophoregram of immature ovary extract (not shown). Numbers on the right side of the figure indicate the position of the molecular mass markers (Pharmacia): phosphorylase-B (94 kDa), bovine serum albumin (67 kDa), ovalbumin (43 kDa), carbonic anhydrase (30 kDa), soybean trypsin inhibitor (20 kDa), and calactalbumin (14 kDa).

ovaries (lane 2) but not immature ovaries (lane 3). The amino acid composition of purified s-VTG is shown in Table 1 compared with compositions reported for vitellogenins of some other fishes or for Xenopus. The relative abundance of specific amino acids in s-VTG is very similar to that reported for the other species. In particular, the vitellogenins are particularly rich in non-polar amino acids, especially alanine, valine, isoleucine, and leucine, which together comprise approximately 38% of the total in s-VTG, possibly relating to its function as a lipoprotein. In contrast, the common polar amino acid, glycine, is relatively low in abundance in the VTGs as compared with other proteins. Values for serine are lower in the fish VTGs as compared to Xenopus. As suggested by the Vlaming et al. (1980), the length of the phosphoserine-rich phosvitin domain of fish VTG may be approximately half of that in Xenopus. The abundance of non-polar amino acids in the fish VTGs was greater than that for Xenopus, which may relate to their higher lipid

content (Campbell and Idler 1980; de Vlaming et al. 1980; Hori et al. 1979; Redshaw and Follet 1971). Cysteine was detected in all of the fish VTGs, suggesting that disulfide bonds may contribute to their secondary structure. The radial immunodiffusion assay was validated for measuring s-VTG in blood plasma of striped bass. It was sensitive and had a wide working range from 4.5 to 180 Ag s-VTG.ml-l of undiluted plasma. The lower limit of detection of s-VTG, estimated as the lowest concentration in undiluted plasma that could be routinely distinguished from assay buffer blanks, was 4.5 g.-mg- 1 . The diameter of the precipitation rings produced by purified s-VTG were directly proportional to the concentration of 0.96) and serial diluthe s-VTG standard (R 2 tions of a plasma sample pooled from several vitellogenic females ran parallel to the standard curve over a 100 fold range of concentrations (0.05-5.0 1 of plasma). Recovery of various concentrations (4.5, 9, 18, 36, 72, 90, 120, 150 and 180 Ag-ml-') of purified s-VTG standard added to plasma sampled from mature (spermiating) males was 100% (P < 0.05), and s-VTG was never detected in plasma of males or immature females. The interassay coefficient of variation was 16.2% (N = 12) and the within assay coefficient of variation was 14.9% (N = 19). A sample of plasma pooled from several vitellogenic females was prepared, aliquoted, and stored at - 70C. Individual aliquots of this sample were thawed, run in each assay, and used to correct for any interassay variation. Plasma s-VTG, E 2 and T levels were measured in a group of 8-year-old fish (N=3) that were individually sampled at each of four stages of oocyte development: primary growth, early secondary growth, vitellogenic, and mature (Fig. 7). Plasma sVTG levels increased with the stage of ovarian maturation and covaried with plasma levels of E2 and T (Table 2), but s-VTG was not detected (< 4.5 y/g-ml-') in blood plasma of females sampled in summer (August), when their most mature oocytes were still in the primary growth stage (Fig. 7a). Plasma levels of E2 and T were also low ( 0.2 ng-ml- 1) in these fish. Plasma s-VTG levels were detectable (- 0.1 mg-ml- 1) in these same females when they were again sampled in autumn (October)

39 Table 1. The amino acid compositions of some vitellogenins Percentage of total amino acids

Amino acid

Aspartic acid Threonine Serine Proline Glutamic acid Glycine Alanine Cysteine/2 Valine Methionine Isoleucine Leucine Tyrosine Phenylalanine Lysine Histidine Arginine

Striped bass

Japanese eel

Rainbow trout

Carp

Xenopus

7.6 5.2 7.2 4.3 8.3 4.2 11.9 1.0 7.8 2.8 7.0 10.8 3.4 3.4 7.3 3.1 4.9

7.3 5.3 5.8 4.7 11.8 5.6 18.0 0.6 6.1 2.8 4.9 7.8 2.8 3.8 5.9 2.0 4.8

8.4 5.0 7.5 5.2 11.5 4.2 11.7 1.2 7.1 2.6 5.5 9.5 3.0 4.0 7.1 2.1 4.5

6.7 5.4 7.6 5.9 11.8 5.1 12.6 0.1 6.3 1.9 5.4 10.5 2.8 2.8 6.3 3.4 5.0

8.7 5.2 10.1 4.9 13.5 4.9 8.0 6.0 2.5 4.8 8.3 3.0 3.6 7.8 3.5 5.1

The results for striped bass are from this study. Those for Japanese eel and rainbow trout are from Hara et al. (1980) and Hara and Hirai (1978), respectively. Results for carp are from Tyler and Sumpter (1990), and those for Xenopus are from de Vlaming et al. (1980).

at a time when their most mature oocytes were in early secondary growth (Fig. 7a). Plasma hormone levels were still low in these fish, although slightly higher than when they were sampled before. Intermediate levels of s-VTG, E 2 and T were observed in these females when they were sampled in winter (January) with clearly vitellogenic oocytes (Fig. 7b, c). Plasma s-VTG levels (0.8 mg-ml- 1) and hormone levels (0.7-2.4 ng-ml- 1) were highest in females with mature oocytes (Fig. 7d), that were sampled in spring (March) just before the spawning season (Table 2). A similar pattern of change in circulating s-VTG and hormone levels was observed in a group of 7-year-old females (N = 12) repeatedly sampled for blood through two reproductive cycles (Fig. 8). After the 1989 spawning season, plasma s-VTG levels were non-detectable in females in summer, increased to low levels (- 0.1 mg-ml- 1) by late October and continued to increase after that until maximum levels (- 0.7 mg-ml - l ) were attained in early March. Plasma s-VTG levels decreased continuously as the spawning season progressed, to low levels (- 0.1 mg-ml - l) observed in mid-May in

spent females. In both 1989 and 1990 peak plasma s-VTG levels were observed early in or just prior to the start of the spawning season, on April 19 and February 27, respectively. E 2 and T levels were measured in these same plasma samples and have been reported elsewhere (Woods and Sullivan 1993). Plasma levels of s-VTG and E2 covaried in these females and they were highly correlated (R 2 =0.68; p _ 0.01).

Discussion In the present study, we purified a plasma protein from striped bass, identified it as striped bass vitellogenin (s-VTG), and developed an immunoassay for quantifying it in blood plasma of naturally maturing females as a measure of their maturational status. We also obtained some fundamental information on the biochemistry of s-VTG, a necessary step for more detailed investigation of oogenesis in striped bass. A number of criteria were used as evidence that the protein we purified is VTG. First, it was specific

40

Fig. 7. Histological appearance of striped bass oocytes at the specific stages of development identified in this study. The ovarian tissue samples were from the same fish for which blood plasma levels of estradiol 17-3, testosterone, and vitellogenin are given in Table 2. Panel A shows representative oocytes in the primary growth (pg) and early secondary growth (sg) stages. Horizontal bar = 30 am. Panel B shows vitellogenic (Vg) oocytes, and Panel C shows a mature oocyte with a migrating germinal vesicle (gv) in the process of final oocyte maturation. Horizontal bars = 100 m. Panel D is a higher magnification of part of a vitellogenic oocyte showing the yolk granules (y) and lipid droplets (1) in the peripheral ooplasm. Horizontal bar = 40 tam. Sections were stained with hematoxylin and eosin counterstained with PAS. Table 2. Levels of estradiol-173 (ng.ml-1), testosterone (ng-ml- ) and vitellogenin (g.ml - ') in blood plasma of female striped bass (N = 3) that were repetitively sampled on various dates and at different stages of oocyte development (PG, primary growth; ESG, early secondary growth; VTG, vitellogenic; MTR, mature). The mean SEM is shown Date

Estradiol-17,3

Testosterone

Vitellogenin

Oocyte stage

Aug 20 Oct 29 Jan 04 Mar 01

0.1 0.4 0.7 2.4

0.2 0.1 0.3 + 0.2 0.4 + 0.2 0.7 + 0.2

ND' 104 + 79 689 + 92 840 + 54

PG ESG VTG MTR

+ + + +

0.1 0.1 0.1 0.7

Data are shown as mean + SEM; IND indicates non detectable (< 4.5

g. ml-l).

41

cbmonpldkajqreihfg

1.0 E

0.8

E 0.6 ,C_ 01 0

0.4

.5:

0.2

k

I

q f

0.0

Jul. 1989

Sept.

Nov.

Sampling Month

Jan.

Mar.

May

1990

Fig. 8. Changes in mean blood plasma levels of vitellogenin in female striped bass broodstock sampled repetitively through two reproductive cycles. Symbols represent the mean value for 12 females. Brackets indicate SEM. Letter superscripts identify the mean value on each sampling date. Letters underlined by a common line indicate means that are not significantly different (p > 0.05). Heavy horizontal bars at the top of the figure indicate the spawning period, defined as the interval during which male fish of this stock were observed to be spermiating.

to maturing females. The a-FSPP antiserum, which was raised against blood plasma from mature female bass and targeted at the female-specific plasma proteins by preadsorption with male plasma, was used in a radial immunodiffusion assay calibrated with purified s-VTG as standard. Using the assay, s-VTG was found to be naturally present only in females showing signs of advanced maturation. These included the presence of yolk globules in their ooplasm, along with elevated levels of E2 and its precursor, T, in their blood plasma (Specker et al. 1987; Berlinsky and Specker 1991; Blythe 1992; Woods and Sullivan 1993). VTGs are generally considered to be strongly expressed only in females, although they have been reported to be present at low levels in males of some fish species (Sumpter 1991; Ding et al. 1989). The lack of immunoreactivity of plasma from males or immature fish indicates that the a-FSPP does not cross-react with any non-vitellogenic proteins. Within the limits of detection of our assay, s-VTG can be considered to be a female-specific plasma protein in striped bass. While s-VTG was normally absent in plasma of males or immature females, electrophoretic analyses showed that it was induced in these fish after

they were implanted with E 2. Inducibility by estrogens is a general feature of vitellogenesis in fishes (Mommsen and Walsh 1988). The s-VTG stained positively for phosphorus with methyl green in native gradient-PAGE, and it was the only TCAprecipitable phosphoprotein detected in E 2-treated animals. Fish VTGs are known to be highly phosphorylated (Campbell and Idler 1980; de Vlaming et al. 1980; Norberg and Haux 1985). They are the most abundant plasma protein and the major source of protein-bound phosphorus in blood plasma from strongly vitellogenic females (Ng and Idler 1983; Craik and Harvey 1984; Bjornsson and Haux 1985). It is also likely that the s-VTG is a lipoprotein, since it stained positively with Sudan black B in native gradient-PAGE and had a highly hydrophobic amino acid composition. Fish VTGs are known to contain a high percentage of lipids (- 20%), present mainly as phospholipids and triglycerides (Norberg and Haux 1985). The overall amino acid composition of s-VTG closely resembled those reported previously for VTGs of some other fish or Xenopus. The s-VTG was immunologically related to the egg yolk proteins. In Western blots of blood plasma, the antiserum used to measure s-VTG (a-

42 FSPP) immunostained only the E2 -induced band (Mr = 170,000) corresponding to the single band produced by purified s-VTG. This same antiserum immunocytochemically stained the yolk globules in mature striped bass oocytes, reacted with the main peak (Mr = 380,000) from gel filtration of ovary extracts, and specifically stained the two main bands (Mr 84,000 and 96,000) in Western blots of the extracts. The s-VTG band was itself absent in SDS-PAGE and corresponding Western blots of the extracts, suggesting that s-VTG is rapidly taken into the oocytes and efficiently broken down into smaller yolk components. Fish, amphibian, and avian vitellogenins are taken up by the developing oocytes and enzymatically modified to form mainly two yolk proteins, lipovitellin and phosvitin (Bergink and Wallace 1974; Deeley et al. 1975; Wallace 1985; Selman and Wallace 1989) along with some minor components. Evidence of a role for the s-VTG in ovarian growth was provided by the continuous rise in its concentration in blood plasma of maturing female striped bass. The increase in circulating s-VTG occurred during the main period of oocyte growth when histological signs of vitellogenesis become increasingly apparent, and it coincided with parallel increases in blood plasma levels of E 2 and its precursor, T. Vitellogenesis is induced by E2, and increases in circulating levels of E 2 are known to occur during vitellogenesis in many other teleosts (Fostier et al. 1983). The relationship between stages of ovarian development and plasma E 2 and T levels that we observed in the captive striped bass broodstock are similar to those reported by Berlinksy and Specker (1991) for wild striped bass sampled from various coastal waters and spawning areas. Our confidence that the maturational changes in ovarian histology and blood hormone and s-VTG levels that we observed in the domestic stock represent "normal" development, comparable to that in natural fish populations, is increased by our ability to successfully ovulate selected brookstock with egg fertilization and hatch rates and fry production rates equivalent to those obtained using wild females (Woods and Sullivan 1993). The s-VTG was initially purified from blood

plasma of E2-treated immature fish by chromatography on hydroxylapatite, but it showed a strong tendency to form strings of precipitate that were difficult to resolubilize. This problem persisted even when the amount of s-VTG applied to the column was reduced by using less plasma or by using plasma from naturally mature females, which have lower circulating s-VTG levels than generally seen E2-induced fish (up to 60 mg s-VTG-ml-l). We later routinely purified s-VTG by ion exchange on DEAE-agarose (Kishida et al. 1992), which yielded useful concentrations of s-VTG ( 1 mg-ml- 1) that did not precipitate from solution. To avoid degradation of s-VTG, the proteolytic enzyme inhibitor, aprotinin, was added to blood at the time of collection and also to the chromatography buffers, and purification procedures were carried out at 0-4 0 C. Fish VTGs can be highly susceptible to enzymatic degradation, and aprotinin has been used before to avoid proteolysis of mummichog (Fundulus heteroclitus), rainbow trout (Oncorhynchus mykiss), and carp (Cyprinus carpio) VTGs (Wallace and Selman 1982; Norberg and Haux 1985; Tyler and Sumpter 1990). The s-VTG was collected from the central portion of the single symmetrical chromatographic peak that was induced by E2, reacted with a-FSPP and contained all of the detectable TCA-precipitable phosphoprotein phosphorus. It appeared in SDS-PAGE as a single band with an estimated molecular weight of 170,000, the only E2 -induced band recognized by the a-FSPP antiserum in Western blots of the blood plasma. The homogeneous elution profile of the s-VTG and its appearance as single band in SDS-PAGE indicate an absence of proteolysis and suggest that the primary subunit of s-VTG may be a - 170 kDa polypeptide. This result corroborates the recent report of Kishida et al. (1992), who used DEAE-agarose to purify an sVTG with an apparent 170 kDa primary subunit from blood plasma of E2 -induced striped bass. This value is very similar to estimates of the molecular weight (170-200 kDa) of the rainbow trout VTG monomer (Chen 1983; Valotaire et al. 1984), and similar to VTG subunit molecular weights (85-200 kDa) reported for several other fishes (tabulated by Mommsen and Walsh 1988; Silver-

43 sand and Haux 1989; Bradley and Grizzle 1989; Ding et al. 1989). In native gradient-PAGE, s-VTG appeared as two closely spaced bands which stained positively for lipid and phosphorus, indicating that there may be at least two forms of s-VTG in the blood plasma of maturing female striped bass. The existence of two distinct VTGs was also recently reported for tilapia (Oreochromisaureus) (Ding et al. 1989). The basis for the multiplicity of native s-VTG, whether due to different degrees of post-translational processing (phosphorylation, lipidation or glycosylation) of a single s-VTG, or to expression of multiple VTG genes as seen in some other vertebrates (reviewed by Byrne et al. 1989; Wahli et al. 1981; Wang and Williams 1982), remains to be verified. If there are multiple s-VTG genes, then their products are polypeptides of very similar molecular weight. The apparent molecular weight of the s-VTG in native gradient-PAGE (340,000) falls within the range reported for the native dimeric VTGs of a variety of teleosts (Wiegand 1982; Hara 1987; Mommsen and Walsh 1988; Silversand and Haux 1989; Tyler and Sumpter 1990), including a recent 300,000; Kishida et al. report on s-VTG (Mr 1992). However, it should be taken as a minimum estimate since VTGs are more highly charged than the molecular mass markers used, and s-VTG likely migrated through the electrophoresis gel anomalously fast. Assuming that s-VTG, like other fish VTGs, contains about 20% lipid, and taking 170 kDa as the mass of the denaturated and delipidated s-VTG monomer, the native s-VTG dimer(s) should have a size of about 425 kDa. The molecular weight of intact s-VTG was approximately 600,000 when estimated by gel filtration on Sepharose 6B, but this is likely an overestimate. Gel filtration frequently overestimates the weight of nonspherical proteins such as VTG, and the pronounced tendency of sVTG to form aggregates (Fig. 3) complicates accurate determination of its native molecular weight by gel filtration. The propensity for aggregation is not unique to s-VTG. When purified rainbow trout or Xenopus VTGs are subjected to analytical ultracentrifugation, multiple peaks are resolved, indicating that polymerization is common feature of

VTG, at least under experimental manipulation (Campbell and Idler 1980; Redshaw and Follet 1971). We utilized the purified s-VTG to develop a single radial immunodiffusion assay for measuring its concentration in striped bass blood plasma. Although this method is not as sensitive as some radioimmunoassays (RIAs; Norberg and Haux 1985; Sumpter 1991; Tyler and Sumpter 1990) or enzyme linked-immunosorbent assays (ELISAs; Nunez-Rodriguez et al. 1989; Mannos et al. 1991; Kishida et al. 1992) that have been used to measure fish VTGs, it is simple, inexpensive, accurate and sensitive enough to measure a greater than 200-fold increase in circulating s-VTG during the reproductive cycle of female striped bass. Provided that the gels are prepared beforehand, it can easily be run in the field, which would be difficult with a RIA or ELISA. The level of sensitivity of the assay is more than adequate for the purpose of detecting vitellogenesis in female striped bass. It appears to be able to identify the sex of adults, i.e., discriminate between maturing (vitellogenic) females and immature fish or males, whenever vitellogenesis is strongly operative. Our results indicate that vitellogenesis occurs through most of the year (October-April) in this striped bass broodstock. Plasma levels of s-VTG varied from non-detectable (< 4.5 14g.ml- 1) in summer to nearly 1 mg-ml-1 in March, just prior to spawning. Peak s-VTG levels in naturally maturing striped bass are similar to those reported for some other teleosts (Copeland and Thomas 1988; Nunez-Rodriguez et al. 1989; Ouchi et al. 1989), and similar to recently reported plasma s-VTG levels in mature female striped bass quantified by specific ELISA (Kishida et al. 1992), although several-fold less than peak levels reported for salmonids (50 mg-ml-1; reviewed by Tyler 1991). Using the s-VTG assay it was possible to detect maturing female striped bass up to 7 months before the spawning season. This capability can be extremely useful for species, like striped bass, that are not sexually dimorphic. Plasma levels of s-VTG increased continuously during ovarian maturation, indicating that s-VTG assay has promise as a means for non-destructively evaluating the degree of

44 maturation in individual females. Gonadal biopsy is a less attractive means to assay reproductive development, especially in fish at early stages of maturation, since it can damage sensitive gonadal ducts or otherwise injure valuable broodstock.

Acknowledgements We thank Dr. J. Specker and M. Kishida for helpful discussions on vitellogenin purification, Dr. Y. Nagahama for providing the specific testosterone antiserum, J. Colby, L. Jackson and W. King for their help with fish sampling, histological technique and RIAs, respectively, and M. McCarthy, D. Theisen, M. Weber and T. Goff for assistance with care and sampling of the fish. We acknowledge Dr. M. Nozaki for providing the immunostaining protocol, and Dr. S. Vigna for the amino acid analysis. This work was supported by grants from the University of North Carolina Sea Grant College Program #NA86AA-D-SG062 and #NA90AA-DSG062 and the National Coastal Resources Research and Development Institute grant #NA87AAD-SG065 (Contract #2-5606-22-2). Parts of this work are included in the Ph.D. dissertation research of Y. Tao (Dept. Zoology, North Carolina State University).

References Bergink, E.K. and Wallace, R.A. 1974. Precursor-product relationship between amphibian vitellogenin and the yolk proteins, lipovitellin and phosvitin. J. Biol. Chem. 249: 28972903. Berlinsky, D.L. and Specker, J.L. 1991. Changes in gonadal hormones during oocyte development of striped bass (Morone saxatilis). Fish Physiol. Biochem. 9: 51-62. Bjornsson, B.T. and Haux, C. 1985. Distribution of calcium, magnesium and inorganic phosphate in plasma of estradiol17/3 treated rainbow trout. J. Comp. Physiol. B. 155: 347352. Blythe, W.G. 1992. Induced maturation and spawning of striped bass, Moronesaxatilis, exposed to 6-, 9-, and 12-month photothermal regimes. M.Sc. Thesis, Virginia Polytechnic Institute and State University. Bradley, J.T., and Grizzle, J.M. 1989. Vitellogenin induction by estradiol in channel catfish, Ictalurispunctatus.Gen. Comp. Endocrinol. 73: 28-39.

Byrne, B.M., Gruber, M. and Ab, G. 1989. The evolution of egg yolk proteins. Prog. Biophys. Mol. Biol. 53: 33-69. Campbell, C.M. and Idler, D.R. 1980. Characterization of an estradiol-induced protein from rainbow trout serum as vitellogenin by the composition and radioimmunological cross reactivity to ovarian yolk fractions. Biol. Reprod. 22: 605-617. Chen, T.T. 1983. Identification and characterization of estrogen-responsive gene products in the liver of rainbow trout. Can. J. Biochem. Cell Biol. 61: 802-810. Copeland, P.A. and Thomas, P. 1988. The measurement of plasma vitellogenin levels in a marine teleost, the spotted seatrout (Cynoscion nebulosus) by homologous radioimmunoassay. Comp. Biochem. Physiol. 91B: 17-23. Craik, J.C.A. and Harvey, S.M. 1984. The magnitudes of three phosphorus-containing fractions in the blood plasma and mature eggs of fishes. Comp. Biochem. Physiol. B. 78: 539-543. Deeley, R.G., Mullinix, K.P., Wetekam, W., Kronenberg, H.M., Meyers, M., Eldridge, J.D. and Goldberger, R.F. 1975. Vitellogenin synthesis in the avian liver. J. Biol. Chem. 25: 9060-9066. de Vlaming, V.L., Wiley, H.S., Delahuntry, G. and Wallace, R.A. 1980. Goldfish (Carassiusauratus) vitellogenin: Induction, isolation, properties and relationship to yolk proteins. Comp. Biochem. Physiol. 67B: 613-623. Ding, J.L., Hee, P.L. and Lam, T.J. 1989. Two forms of vitellogenin in the plasma and gonads of male tilapia (Oreochromis aureus). Comp. Biochem. Physiol. 93B: 363-370. Fostier, A., Jalabert, B., Billiard, R., Breton, B. and Zohar, Y. 1983. The gonadal steroids. In Fish Physiology. Vol. 9A, pp. 277-372. Edited by W.S. Hoar, D.J. Randall and E.M. Donaldson. Academic Press, New York. Gamst, 0. and Try, K. 1980. Determination of serum phosphate without deproteinization by ultraviolet spectrophotometry of the phosphomolybdic acid complex. J. Clin. Lab. Invest. 40: 483-486. Hara, A. 1987. Studies on female-specific serum proteins (vitellogenin) and egg yolk proteins in teleosts: immunochemical, physiochemical and structural studies. Mem. Fac. Fish, Hokkaido Univ. 34: 1-59. Hara, A. and Hirai, H. 1978. Comparative studies on immunochemical properties of female-specific serum protein and egg yolk proteins in rainbow trout (Salmo gairdneri). Comp. Biochem. Physiol. 59B: 339-343. Hara, A., Yamauchi, K. and Hirai, H. 1980. Studies on femalespecific serum protein (vitellogenin) and egg yolk protein in Japanese eel (Anguilla japonica). Comp. Biochem. Physiol. 65B: 315-320. Harlow, E. and Lane, D. 1988. Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory, New York. Hodson, R.G., Smith, T., McVey, J., Harrell, R. and Davis, N. 1987. Hybrid striped bass culture: status and perspective. University of North Carolina Sea Grant College Program North Carolina. Hodson, R.G. and Sullivan, C.V. 1993. Induced maturation and

45 spawning of domestic and wild striped bass (Morone saxatilis) broodstock with implanted GnRH analogue and injected hCG. J. Aquacult. Fish. Managem. 24: 271-280. Hori, S.H., Kodama, T. and Tanahashi, K. 1979. Induction of vitellogenin synthesis in goldfish by massive doses of androgens. Gen. Comp. Endocrinol. 37: 306-320. John, A.C. and Thomas, F.R. 1973. Staining of phospho-proteins on acrylamide gel electrophoregrams. Anal. Biochem. 54: 386-394. Kishida, M., Anderson, T.R. and Specker, J.L. 1992. Induction by -estradiol of vitellogenin in striped bass (Morone saxatilis): characterization and quantification in plasma and mucus. Gen. Comp. Endocrinol. 88: 29-39. Mancini, G., Carbonara, A.O. and Heremans, J.F. 1965. Immunological quantitation of antigens by single radial immunodiffusion. Immunochemistry 2: 235-254. Mannos, E., Zanuy, S., Carillo, M., Nunez Rodriguez, J. and Le Menn, F. 1991. Quantification by ELISA of vitellogenin levels in sea bass (Dicentrarchuslabrax L.) maintained under different photoperiods. In Proc. IV Int. Symp. Reproductive Physiology of Fish. p. 325. Edited by A.P. Scott, J.P. Sumpter, D.E. Kime and M.S. Rolfe. FishSymp 91, Sheffield. Mayer, I., Shackley, S.E. and Ryland, J.S. 1988. Aspects of the reproductive biology of the bass, Dicentrarchuslabrax L. I. An histological and histochemical study of oocyte development. J. Fish Biol. 33: 609-622. Mommsen, T.P. and Walsh, P.J. 1988. Vitellogenesis and oocyte assembly. In Fish Physiology. Vol. 9A, pp. 347-406. Edited by W.S. Hoar, D.J. Randall and E.M. Donaldson. Academic Press, New York. Ng, T.B. and Idler, D.R. 1983. Yolk formation and differentiation in teleost fishes. In Fish Physiology. Vol. 9A, pp. 373-404. Edited by W.S. Hoar, D.J. Randall and E.M. Donaldson. Academic Press. New York. Norberg, B. and Haux, C. 1985. Induction, isolation and characterization of the lipid content of plasma vitellogenin from two Salmo species: rainbow trout (Salmo gairdneri)and sea trout (Salmo trutta). Comp. Biochem. Physiol. 81B: 869-876. Nozaki, M., Miyata, K., Oota, Y., Gorbman, A. and Plisetskaya, E.P. 1988. Different cellular distributions of two somatostatins in brain and pancreas of salmonids, and their separate associations with insulin and glucagon secreting cells. Gen. Comp. Endocrinol. 69: 267-280. Nunez-Rodriguez, J.N., Kah, O., Geffard, M. and Le Menn, F. 1989. Enzyme-linked immunosorbent assay (ELISA) for sole (Solea vulgaris) vitellogenin. Comp. Biochem. Physiol. 92B: 741-746. Ouchi, K., Hara, A., Adachi, S., Arimoto, M., Mizuta, Y. and Nagahama, Y. 1989. Changes in serum concentrations of steroid hormones and vitellogenin during the period of vitellogenesis in the yellowtail, Seriolaquinqueradiata. Bull. Nansei Regional Fish. Res. Laboratory 22: 1-11. Prat, J.P., Lamy, J.N. and Weill, J.D. 1969. Coloration des lipoproteins apres electrophoreses en gel de ployacrylamid. Bull. Soc. Chim. Biol. 51: 1367.

Redshaw, M.R. and Follet 1971. The crystaline yolk-platelet proteins and their soluble plasma precursor in an amphibian, Xenopus laevis. Biochem. J. 124: 759-766. Rees, R.A. and Harrell, R.M. 1990. Artificial spawning and fry production of striped bass and hybrids. In Culture and Propagation of Striped Bass. pp. 43-72. Edited by R.M. Harrell, J.H. Kerby and R.V. Minton. American Fisheries Society, Bethesda. Selman, K. and Wallace, R.A. 1989. Cellular aspects of oocyte growth in teleosts. Zool. Sci. 6: 211-231. Silversand, C. and Haux, C. 1989. Isolation of turbot (Scophthalmus maximus) vitellogenin by high-performance anion-exchange chromatography. J. Chrom. 478: 387-397. Specker, J.L., Berlinsky, D.L., Bibb, H.D. and O'Brien, J.F. 1987. Oocyte development in striped bass: factors influencing estimates of age at maturity. Am. Fish. Soc. Symp. 1: 162-174. Sullivan, C.V., Tao, Y., Hodson, R.G., Hara, A., Bennett, R.O. and Woods, L.C. III. 1991. In Proc. IV Int. Symp. Reproductive Physiology of Fish. pp. 315-317. Edited by A.P. Scott, J.P. Sumpter, D.E. Kime and M.S. Rolfe. FishSymp 91, Sheffield. Sumpter, J.P. 1991. The purification, radioimmunoassay and plasma levels of vitellogenin from the rainbow trout, Salmo gairdneri.In Current Trends in Comparative Endocrinology. Edited by B. Lofts and W.N. Holmes. Vol. 1, pp. 355-357. Hong Kong University Press, Hong Kong. Tyler, C. 1991. Vitellogenesis in Salmonids. In Proc. IV Int. Symp. Reproductive Physiology of Fish. pp. 295-299. Edited by A.P. Scott, J.P. Sumpter, D.E. Kime and M.S. Rolfe. FishSymp 91, Sheffield. Tyler, C.R. and Sumpter, J.P. 1990. The development of a radioimmunoassay for carp, Cyprinus carpio, vitellogenin. Fish. Physiol. Biochem. 8: 129-140. Valotaire, Y., Tenniswood, M., Le Guellec, C. and Tata, J.R. 1984. The preparation and characterization of vitellogenin messenger RNA from rainbow trout (Salmo gairdneri). Biochem. J. 217: 73-77. Wang, S.-Y. and Williams, D.L. 1982. Biosynthesis of vitellogenin. Identification and characterization of nonphosphorylated precursors to avian vitellogenin I and vitellogenin II. J. Biol. Chem. 257: 3837-3846. Wahli, W., Dawid, I.B., Ryffel, G.U. and Weber, R. 1981. Vitellogenesis and the vitellogenin gene family. Science 212: 298-304. Wallace, R.A. 1985. Vitellogenesis and oocyte growth in nonmammalian vertebrates. In Developmental Biology. Vol. 1, pp. 127-177. Edited by L.W. Browder. Plenum Press, New York. Wallace, R.A. and Selman, K., 1982. A new procedure for the isolation of intact vitellogenin from teleosts. In Reproductive Physiology of Fish. pp. 161. Edited by C.J.J. Richter and H.J.Th. Goos. Pudoc, Wageningen. Wiegand, M.D. 1982. Vitellogenesis in fishes. In Reproductive Physiology of Fish. pp. 136-146. Edited by C.J.J. Richter and H.J.Th. Goos. Pudoc, Wageningen.

46 Woods, J.C. III, Bennett, R.O. and Sullivan C.V. 1992. Reproduction of a domestic striped bass broodstock. Prog. Fish Cult. 54: 184-188. Woods, J.C. III and Sullivan, C.V. 1993. Reproduction of striped bass (Moronesaxatilis) broodstock: monitoring matu-

ration and hormonal induction of spawning. J. Aquacult. Fish. Managem. 24: 213-224. Woods, L.C. III, Woiwode, J.G., McCarthy, M.A., Theisen, D.D. and Bennett, R.O. 1989. Noninduced spawning of captive striped bass in tanks. Prog. Fish Cult. 52: 201-202.