Fish Physiology and Biochemistry vol. 13 no. 4 pp 309-316 (1994) Kugler Publications, Amsterdam/New York
Effects of unilateral ovariectomy on recruitment and growth of follicles in the rainbow trout, Oncorhynchus mykiss C.R. Tylerl, J.J. Naglerl, T.G. Pottinger 2 and M.A. Turner 1 Department of Biology and Biochemistry, Brunel University, Uxbridge, Middlesex. UB8 3PH. United Kingdom; 2 Institute of FreshwaterEcology, Windermere laboratory, Far Sawrey, Ambleside, Cumbria. LA22 OLP. United Kingdom Accepted: May 13, 1994 Keywords: unilateral ovariectomy, hypertrophy, recruitment, fecundity, follicle, rainbow trout
Abstract Virgin female rainbow trout, Oncorhynchus mykiss, were unilaterally ovariectomised at various stages of ovarian development to investigate the effect of the removal of one ovary on subsequent recruitment and growth of follicles in the remaining ovary. The right ovary was removed from groups of 12-15 fish, 12, 7 and 4 months before they were due to ovulate, and the gonadosomatic index and follicle number and size determined just prior to ovulation. There were no differences in fecundity or follicle size in fish unilaterally ovariectomised at 12 and 7 months prior to ovulation compared to the controls. However, in the females unilaterally ovariectomised 4 months prior to ovulation, the remaining ovary either had the normal number of follicles for a single ovary but of a significantly larger size than follicles in the controls, or alternatively had almost 70% more than the normal number of vitellogenic follicles but comprising two distinctly different size populations. Differences in plasma oestradiol-17[5 concentrations at the final sample were seen only in females unilaterally ovariectomised 4 months prior to ovulation, where the levels were significantly lower than both the sham operated and control fish (p < 0.05). These data show that in the rainbow trout, complete compensatory ovarian hypertrophy following unilateral ovariectomy can occur throughout a major part of ovarian development, but that follicle recruitment is limited to stages up to (and therefore fecundity is determined by) mid-vitellogenesis.
Introduction In mammals, removal of one ovary, resulting in the loss of a large pool of ovarian follicles, leads to compensatory hypertrophy in the remaining ovary, which often doubles its normal ovulation rate (Greenwald 1961; Peppler and Greenwald 1970; Hirshfield 1982; Meijs-Roelofs et al. 1982). Similarly, studies in a few species of lizards (Jones et al. 1977, 1978; Sinervo and Licht 1991) and amphi-
bians (Kjaer and Jorgensen 1971; Jorgensen 1973; Billeter and Jorgensen 1976) have shown that unilateral ovariectomy (ULO) results in a hypertrophy of the remaining ovary. Unilateral removal of one testis in fish was performed in the stickleback (Gasterosteus aculeatus) as early as 1931, and resulted in compensatory hypertrophy of the remaining testis (Craig-Bennet 1931). Information was not given, however, on the changes associated with the hypertrophy. Subse-
Correspondenceto: Dr. Charles Tyler, Department of Biology and Biochemistry, Brunel University, Uxbridge, Middlesex. UB8 3PH, United Kingdom, Tel. 0895 274000 ext. 2085, Fax 0895 274348.
310 quent studies in male rainbow trout demonstrated that removal of one testis, produced as much as a two-fold increase in the mass of the remaining testis (Robertson 1958) and that this hypertrophy resulted from actively dividing germ cells (Billard et al. 1981). There are very few studies on ULO in fish, but the available evidence indicates that a compensatory hypertrophy can occur (catfish, Heteropnuestesfossilis, Goswami and Sundararaj 1968a, b; tilapia, Tilapaiaurea, Dadzie and Hyder 1976). In the rainbow trout, during each maturational cycle the ovaries grow from less than 0.2% of the body weight up to around 20% just prior to ovulation, at which time a single batch of 2000-3000 eggs per kilogram of body weight, each measuring about 5 mm in diameter, is ovulated into the body cavity. Oocyte development is divided into primary and secondary growth and each of these periods are subdivided into a number of stages (Nagahama 1983). Growth occurs principally in the secondary growth phase during vitellogenesis, when follicles increase in size from around 0.5 mm to 5 mm in diameter. This increase is largely due to uptake of extra-ovarian protein from the blood (Tyler et al. 1988, 1990b; Tyler 1991). A previous study on the dynamics of follicle growth and recruitment in the rainbow trout (Tyler et al. 1990b) has shown that there is little, if any, change in the number of follicles recruited into the secondary growth phase from the time vitellogenesis is initiated through to when the eggs are finally ovulated 9 months later, suggesting that fecundity is normally determined at or before the onset of vitellogenesis. In fish, little is known about the factors that affect oogonial proliferation, the pool size and growth pattern of previtellogenic oocytes or follicular development. Furthermore, the determinants of fecundity have yet to be firmly established in any animal. The effects of ULO on follicle growth and recruitment make it a useful tool for examining potential factors affecting the growth and recruitment of secondary growth follicles. In this study, rainbow trout were unilaterally ovariectomised at various stages during ovarian development to assess if, and when, compensatory hypertrophy could occur during the ovarian growth cycle and, if compensation did take place, to investigate the regulatory
responses in the remaining ovary in terms of the number (fecundity) and size of vitellogenic follicles produced.
Materials and methods Fish maintenance One hundred and twenty virgin 2-year-old rainbow trout, of South African origin, were divided equally between four outdoor 15001 glass fibre tanks. Each tank was supplied with a constant flow of lake water (25 1min-'). Fish were marked with an identity number by careful scarification using a mounted needle. The exposed dermis was flooded with a concentrated solution of alcian blue. During the study, which lasted more than a year, fish required re-marking. Fish were fed five times weekly with commercial food at the manufacturers recommended rate.
Unilateralovariectomy operations The study was initiated one year prior to the expected ovulation date of the fish. One tank of fish were maintained as controls (group 1), while females in the other 3 tanks were operated on in December 1992 (group 2), May 1993 (group 3) and August 1993 (group 4), 12, 7 and 4 months before they were due to ovulate, respectively. At each of these time points, 12-15 females were unilaterally ovariectomised and a similar number sham operated. Operations were conducted under 2-phenoxyethanol anaesthesia (1.2000; Sigma Chemical Company, Poole, Dorset, U.K.). An incision (approximately 2-3 cm in length) was made behind the operculum in a dorso-ventral direction to a depth such that the body cavity was just penetrated, the right ovary was located and carefully pulled through the incision using a specially designed blunt glass hook and detached from the viscera using fine scissors. After removal of the ovary the incision was closed with 4-5 sutures and the wound was covered liberally with antiseptic cream. The ovary removed from each ULO fish was weighed,
311 fixed in 70% alcohol and fifty of the largest follicles were measured to the nearest 0.05 mm using a stereoscopic binocular microscope fitted with an eyepiece graticule. Sham operated fish were treated similarly, but the right ovary was left in situ. At each sampling time, all the females in every tank were blood sampled from the Cuverian sinus using heparinised syringes and weight (g) and fork length (cm) were recorded. Blood samples were spun at 2500 x g for 15 min at 4°C and the plasma withdrawn and frozen at -70 0 C until required.
Measurements of plasma levels of oestradiol-17/3 Plasma samples were radioimmunoassayed for oestradiol-173 according to Carragher et al. (1989).
Assessing the effects of unilateralovariectomy on fecundity At the termination of the experiment (mid-December 1993), when the fish were close to ovulation, all of the experimental females were killed, blood samples were removed, and fish weighed and measured. Some of the females within the study population had already ovulated, but these were excluded from the analyses. The coefficient of condition (condition factor; K) of fish was determined using the equation K = [fish weight (g)/fish length3 (cm)] x 100, according to Sumpter et al. (1991). In the control and sham operated females, both ovaries were removed and weighed separately. In the ULO females, the remaining left ovary was removed and weighed. In a few ULO females the right ovary had not been removed completely at the time of the operation and any fish in which the remaining portion of the right ovary exceeded 5% by weight of the remaining left ovary was excluded from the analyses. Gonadosomatic indices [GSI; (total ovary weight/total body weight) x 100] were calculated for each fish. The size of follicles in each ovary was determined by measuring the diameters of 20 follicles to the nearest 0.1 mm, as described above. The number of follicles in each ovary was determined by weighing
a portion of the left ovary (around 10% of the ovary weight), counting the number of follicles present and extrapolating the number for the weight of the whole ovary (see Tyler et al. 1990a). An exception to the analyses detailed above took place in females unilaterally ovariectomised 4 months prior to ovulation. In some of these fish two different sized populations of vitellogenic follicles were present, rather than one as in all of the other groups (see results), and therefore the ovaries from all females in group 4 were fixed in 70% alcohol and approximately 20% of the follicles counted and their diameters measured to the nearest 0.1 mm. From the measurements the number of follicles present in different size classes for the whole ovary was calculated and size - frequency histograms were produced (see Tyler et al. 1990b).
Statistical analyses In this study, data are presented as mean + SEM. Significant differences between the ULO, sham and control groups were determined by ANOVA (p < 0.05) followed by a multiple comparisons test (Fisher PLSD, p < 0.05). Significant differences between the weights of the right and left ovaries in control fish and differences in the levels of plasma oestradiol-17 between females with one and two populations of vitellogenic follicles in those unilaterally ovariectomised in August were determined using Students 't' tests. All comparisons were performed using a Statview 512+ programme on an Apple Macintosh personal computer.
Results The status of ovarian development, in terms of the GSI and the size of the largest follicles present, at the times of the operations are given in Fig. 1. In December 1992 the mean GSI was 0.19 ± 0.01 and most, if not all, of the follicles (which measured less than 0.65 mm in diameter), were previtellogenic (Tyler 1991). In May 1993, the GSI was 0.36 + 0.03 and most of the maturing follicles were vitellogenic (measuring up to 1.1 mm in diameter). In August
312 Gonadosomatic Index 0.19
+
0.01
0.36 + 0.03
1.24 + 0.06
30 go
Dates of operations
25
*
'i
_
December'92
X
2
o
[3 May 3
20
C,
Augt l3
[
00 D-
06 a '
.
15' 10'
0 5. nt .
IIIll l A_
.2
.-
.3
.4
.
.5
..
.6
.
in IIi . .
.7
,_
.6
.
..
.9
.
··
1
1.1
1.2
.
II .
1.3
.
.
1.4
.
1.
l
.
I11l I q
1. 6 1.7
F.. 1.8
.
1.9
2.0
Mean diameter of largest ovarian follicles (mm). Fig. 1. Status of ovarian development, in terms of the gonadosomatic index and the size of the largest ovarian follicles, in female rainbow trout at the times of conducting unilateral ovariectomy operations in December 1992, May 1993 and August 1993.
1993, the GSI was 1.24 ± 0.06 and the maturing
group of ULO females, the remaining ovary
follicles were in mid-vitellogenesis (measuring up to 2.1 mm in diameter). At the final sampling point there were no significant differences in body weight between the ULO, sham and control fish. Similarly, there were no differences in condition factor between the ULO, sham and control fish (1.51 + 0.02, 1.52 ± 0.02 and 1.49 ± 0.03, respectively). There were no significant differences between the weights of the left and right ovary in the control fish, indicating that removal of the right ovary in the ULO fish removed half the ovarian tissue. In females unilaterally ovariectomised at 12 and 7 months prior to ovulation, at the final sampling there were no significant differences in GSI, number of follicles and follicle size compared to the sham operated and control fish (Fig. 2). In females unilaterally ovariectomised 4 months prior to ovulation, there was a statistically significant reduction in both the GSI (8.2 0.7 [ULO] vs. 11.7 ± 1.14 [sham] and 13.8 + 2.1 [controls]) and the number of follicles (1021 + 102 [ULO] vs. 1650 ± 142 [sham] and 1832 + 134 [controls]; Fig. 2). In this
showed one of two growth patterns; either the normal number of follicles for a single ovary were present (778 ± 60) that were of a significantly larger size (mean follicle diameter in these fish, 4.83 ± 0.15 mm vs. 3.99 ± 0.17 mm in the shams and 4.19 ± 0.11 mm in the controls; p < 0.05) or there were approximately 70% more than the normal number of vitellogenic follicles in the remaining ovary (1264 + 133), but these comprised two distinctly different size populations. The mean follicle diameters of these two populations were 4.1 ± 0.21, which was not significantly different from the sham-operated and controls, and 2.73 ± 0.21 which was significantly smaller (p < 0.05; Fig. 3). During the study, plasma oestradiol-173 levels increased from 0.1-0.8 ng/ml in December 1992, to 0.5-5 ng/ml in May 1993, to 2-17 ng/ml in August 1993, and up to a maximum of 70 ng/ml at the final sampling in December 1993. There were no differences in the levels of plasma oestradiol-1713 between control, sham and ULO groups, with the exception of females in group 4 at the final sampling time; here plasma levels of oestradiol-17[
313
a
C
.2 U E o0 o o o
a o
--C 1 0
2
a
01 IE Z a i
a .2 o'3 I 2
V S
a a
z
Control
ULO
S ULO
S ULO
Dec. '92
s
May '93
Aug. '93
Dates of unilateral ovarlectomy Fig. 2. Gonadosomatic index and the number and mean diameter of follicles in females just prior to ovulation. S = sham operated, ULO = unilaterally ovariectomised. * significantly differed from the sham operated group in August and the controls, p < 0.05. n: controls - 8; Dec. S - 8, ULO - 9; May S - 13, ULO - 12; Aug. S - 10, ULO - 12.
were significantly lower in the ULO fish (12.6 + 2.84 ng/ml) compared to both the sham operated (33.7 4.8 ng/ml) and control fish (32.4 5.6). In this group of ULO females, there was also a significant difference in the level of plasma oestradiol-173 between the females containing two populations of differently sized vitellogenic follicles and those containing only one (21.3 + 1.3 ng/ml vs. 6.92 3.48 ng/ml, respectively p < 0.05). Discussion In animals, the number of ovulations and eggs oviposited has evolved to ensure optimum survival
of offspring in relation to environmental pressures, and the mean number of eggs ovulated per cycle is typical of each species (Jones 1978). The results of this study on the rainbow trout demonstrate that, even during vitellogenesis, removal of half the ovarian tissue is compensated for by hypertrophy of the remaining ovary, which doubles its production of vitellogenic follicles and thus maintains a constancy in the number of eggs produced. In addition, during the study, two females were found that naturally had only one ovary and their fecundity was similarly not affected. Compensatory ovarian hypertrophy is a phenomenon seen in a wide variety of animal species, but the mechanisms by which one ovary compensates for the absence or perhaps malfunction of the other probably differ. Compensatory hypertrophy in mammals may be achieved by two mechanisms, either by reduction in the rate of atresia of antral follicles present (Greenwald 1961; Jones 1978; Hirshfield 1982; Gosden et al. 1989), or by recruitment of an additional number of small antral follicles (Peppler and Greenwald 1970; Osman et al. 1982; Meijs-Roelofs et al. 1984). Unlike the situation in mammals, however, where the number of oogonia is set and indeed decreasing even before birth, in fish oogonial proliferation occurs throughout life (Billard et al. 1982; Billard 1987), although in the trout it is usually limited to short periods during or just after ovulation each year (Franchi et al. 1962). In the tilapia and Indian catfish, partial ovariectomy has been shown to increase the number of dividing oogonia (Goswami and Sundararaj 1968a; Dadzie and Hyder 1976). Similarly in this study on the rainbow trout, ULO may have activated germ cell division to reinitiate a 'normal' gametogenesis. In other batch spawning fish, however, it has been shown that in a vitellogenic ovary there is also an asynchronous population of pre-vitellogenic follicles and these may serve as a pool for further follicle recruitment into the maturing pool (see Wallace and Selman 1981). The ability of the trout ovary to recruit another population of follicles into the secondary growth phase, by whatever mechanism, even as late in development as mid-vitellogenesis, probably functions normally to balance any loss from the pool of developing follicles.
314
a
2 ar
0
a
0
a
IO
oI
Follicle dameter (mm)
Fig. 3. Variation in the sizes of follicles in females unilaterally ovarectomised in August. Fish 1-6 inclusive did not recruit a second population of follicles into vitellogenesis, fish 7-12 did. The number of follicles in an ovary in each 100 gm-diameter size class is expressed as a percentage of the total number of maturing follicles in that ovary.
Lacker et a. (1987), suggested that in mammals all follicles inherit the same developmental programme and the maturation rate of each follicle at any time depends on it's maturity and the concentration of circulating oestrogens and gonadotrophin. In the rainbow trout, in contrast, follicles of a very similar size appear to be able to grow at different rates (Tyler et al. 1990a). In this study, follicles recruited from oogonia/follicles in the previtellogenic pool into the secondary growth phase following unilateral ovariectomy in May could have differed in size by up to 200 thousand fold (by volume) compared to the larger vitellogenic population, yet they were the same size as their earlierdeveloping counterparts at ovulation, 7 months later. Enhanced growth rates of this 'new' population of developing follicles is likely to be a function, at least in part, of the more advanced reproductive status of the female (increased plasma levels of both
oestradiol-175, which in May were up to 5 ng/ml, and vitellogenin, etc.) compared to that which they would normally encounter at the beginning of the secondary growth phase. However, it is likely that there are also intraovarian factors that participate in the control of follicle growth (Tyler et al. 1990b). These data illustrate that growth of follicles can proceed at rates considerably faster than those normally observed. Mechanisms controlling follicle recruitment per se and recruitment after unilateral ovariectomy are yet to be determined in any animal. In a number of mammals, however, ULO induces an increase in levels of serum follicle stimulating hormone (FSH) and this in turn causes compensatory growth of follicles in the remaining ovary (Jones 1978; for review, Osman et al. 1982). Recent evidence suggests that sensory innervation may also play a role in ovarian growth and development (Deshmukh et al.
315 1992). In non-mammalian vertebrates there is evidence that growth hormone restores follicle recruitment in ULO toads on a restricted diet (a restricted diet inhibits ovarian hypertrophy; Billeter and Jorgensen 1976), but the absence of homologous FSH (gonadotrophin-I) assays, until very recently, have precluded assessing the role of gonadotrophin in follicle growth and recruitment. The compensatory systems controlling follicle numbers in the rainbow trout, however, appear to be limited to stages of ovary development up to around mid-vitellogenesis; half of the females unilaterally ovarectomised in August did not recruit a second population of follicles into the maturing pool and the other half of the females that did only partially compensated in that they recruited only 70%o more follicles into the maturing population rather than doubling the number. Females that did not recruit a second population of follicles into vitellogenesis produced follicles that were of a significantly larger size; 76% larger by volume compared to the controls. These data indicate that the size of the eggs produced in rainbow trout is determined, within the limits of variance set by genetic constraints, by the number of follicles undergoing vitellogenesis. The two populations of vitellogenic follicles found in half of the females unilaterally ovariectomised in group 4 were of widely divergent sizes at the final sampling (a mean difference in volume of over 3.5-fold). It has not been established whether a trout egg has to reach a critical size before it is ovulated and since none of the females in group 4 had ovulated, we were not able to determine the fate of the population of smaller follicles; they may ovulate as small eggs at the normal ovulation time, they may undergo atresia, or alternatively they may delay the timing of ovulation such that they are able to reach a size approaching that of their larger counterparts. The lower levels of plasma oestradiol-173 seen in the ULO females, compared to both the controls and sham operated fish in group 4 at the final sampling may simply be a reflection of the reduced amount of ovarian tissue (and hence a reduced amount of steroid-secreting granulosa and thecal cells). Within this group of ULO females there was also a clear difference in the levels of plasma
oestradiol-1713 between the females containing two populations of maturing follicles and those containing only one. The higher titre of plasma oestradiol-173 in the females containing two populations of maturing follicles probably results from both the greater number of vitellogenic follicles and that the population of smaller vitellogenic follicles would be more actively secreting oestradiol-171B compared to the larger follicles that were close to full maturity (see Kanamori et al. 1988 for seasonal pattern of oestradiol-173 secretion). The level of plasma oestradiol-17 in the ULO females in group 4 that contained only one population of maturing follicles is somewhat lower than that anticipated compared to the controls, even when the reduced amount of ovarian mass is accounted for, and this may indicate that these fish were closer to ovulation than their counterparts in the control groups (see Kanamori et al. 1988). Jansen (1994) suggested that in mammals, two ovaries exist for anatomical symmetry and a 'dab of insurance' should one fail, because for virtually all physiological purposes they act as one. This hypothesis seems also to apply to the rainbow trout, where females with only one ovary can carry out gametogenesis normally and are equally as fecund as females with two ovaries. Follicle recruitment into the secondary growth phase in the rainbow trout, and therefore a final decision, however, is limited to the stages of development up to mid vitellogenesis.
References cited Billard R. 1987. The reproductive cycle of the male and female brown trout (Salmo trutta fario); a quantitative study. Reprod. Nutr. Develop. 27: 29-44. Billard, R., Crim, L.W., Peter, R.E. and Breton, B. 1982. Longterm changes in plasma and pituitary GtH after castration of rainbow trout at an immature stage. In Reproductive Physiology of Fish. pp. 50. Edited by C.J.J. Ritcher and H.J.Th. Goos. Pudoc Press, Wageningen. Billeter, E. and Jorgensen, C.B. 1976. Ovarian development in young toads, Bufo bufo bufo (L.): Effect of unilateral ovariectomy, hypophysectomy, treatment with gonadotropin (hCG), growth hormone and prolactin, and importance of body growth. Gen. Comp. Endocrinol. 29: 531-544. Carragher, J.C., Sumpter, J.P., Pottinger, T.G. and Pickering,
316 A.D. 1989. The deleterious effects of cortisol implantation on reproductive function in two species of trout, Salmo truttaL. and Salmo gairdneri Richardson. Gen. Comp. Endocrinol. 76: 310-321. Craig-Bennett, A. 1931. The reproductive cycle of the three spined stickleback, Gasterosteusaculeatus Linn. Phil. Trans. Roy. Soc. London Ser. 219: 197-279. Dadzie, S. and Hyder, M. 1976. Compensatory hypertrophy of the remaining ovary and the effects of methallibure in the unilaterally ovariectomized Tilapia aurea. Gen. Comp. Endocrinol. 29: 433-440. Deshmukh, M.K., Krishna, R.N.S. and Subhedar, N. 1992. Ovarian stretch signals trigger volume increase in the magnocellular preoptic nuclear region in the brain of the catfish, Clariasbatrachus (Linn.). J. Exp. Zool. 263: 231-234. Franchi, L.L., Mandle, A.M. and Zuckermann, S. 1962. The development of the ovary and the process of oogenesis. In The Ovary. pp. 1-88. Edited by R.E. Jones. Academic Press, New York. Gosden, R.G., Telfer, E., Faddy, M.J. and Brook, D.J. 1989. Ovarian cyclicity and follicular recruitment in unilaterally ovariectomized mice. J. Reprod. Fertil. 87: 257-264. Goswami, S.V. and Sundararaj, B.I. 1968a. Compensatory hypertrophy of the remaining ovary after unilateral ovariectomy at various phases of the reproductive cycle of catfish, Heteropnueustes fossilis (Bloch). Gen. Comp. Endocrinol. 11: 401-413. Goswami, S.V. and Sundararaj, B.I. 1968b. Effect of estradiol benzoate, human chorionic gonadotrophin, and folliclestimulating hormone on unilateral ovariectomy-induced compensatory hypertrophy in catfish, Heteropnueustes fossilis (Bloch). Gen. Comp. Endocrinol. 11: 393-400. Greenwald, G.S. 1961. Quantitative study of follicular development in the ovary of the intact or unilaterally ovariectomised hamster. J. Reprod. Fertil. 2: 351-361. Hirshfield, A.N. 1982. Follicular recruitment in long term hemicastrate rats. Biol. Reprod. 27: 48-53. Jansen, R. 1994. Follicles in competition. Nature, Lond. 367: 10. Jones, R.E. 1978. Control of follicular selection. In The Vertebrate Ovary: Comparative Biology and Evolution. pp. 763788. Edited by R.E. Jones. Plenum Press, New York. Jones, R.E., Fitzgerald, K.T. and Duvall, D. 1978. Quantitative analysis of the ovarian cycle of the lizard Lepidodactylus lugubris. Gen. Comp. Endocrinol. 35: 70-76. Jones, R.E., Fitzgerald, K.T. and Tokarz, R.R. 1977. Endocrine control of clutch size in reptiles. VII. Compensatory ovarian hypertrophy following unilateral ovarectomy in Sceloporus occidentalis. Gen. Comp. Endocrinol. 31: 157160. Jorgensen, C.B. 1973. Pattern of recruitment of oocytes to second growth phase in normal toads, and in hypophysectomised toads Bufo bufo bufo (L), treated with gonadotrophin (hCG). Gen. Comp. Endocrinol. 21: 152-159. Kanamori, A.K., Adachi, S. and Nagahama, Y. 1988. Developmental changes in steroidogenic responses of ovarian follicles of amago salmon (Oncorhynchus rhodurus)to chum salmon
gonadotropin during oogenesis. Gen. Comp. Endocrinol. 72: 13-24. Kjaer, K. and Jorgensen, C.B. 1971. Effects of mammalian gonadotropins on ovary in the toad Bufo bufo bufo (L). Gen. Comp. Endocrinol. 17: 424-431. Lacker, H.M., Beers, W.H., Meuli, L.E. and Akin, E. 1987. A theory of follicle selection: I Hypotheses and examples. Biol. Reprod. 37: 570-580. Meijs-Roelofs, H.M.A., Kramer, P., Osman, P. and Sander, H.J. 1984. Compensatory ovulatory mechanisms operative after first ovulation in rats unilaterally ovariectomised prepuberally. Biol. Reprod. 31: 44-51. Meijs-Roelofs, H.M.A., Osman, P. and Kramer, P. 1982. Ovarian follicular development leading to first ovulation and accompanying gonadotrophin levels as studied in the unilaterally ovariectomized rat. J. Endocrinol. 92: 341-349. Nagahama, Y. 1983. The functional morphology of teleost gonads. In Reproduction-Endocrine Tissues and Hormones. pp. 223-276. Edited by W.S. Hoar, D.J. Randall and E.M. Donaldson. Academic Press, New York. Osman, P., Meijs-Roelofs, H.M.A. and Kramer, P. 1982. Compensatory follicle growth and ovulation after unilateral ovariectomy during the late prepubertal period in the rat. In Development and Function of Reproductive Organs. pp. 299-306. Edited by A.G. Byskov and H. Peters. Exerpta Medica, Amsterdam. Pepper, R.D. and Greenwald, G.S. 1970. Influence of unilateral ovariectomy on follicular development in cycling rats. Am. J. Anatomy 127: 9-14. Robertson, M.D. 1958. Accelerated development of testis after unilateral gonadectomy, with observations of the normal testes of rainbow trout. Bull. Fish Wildl. Serv. 58: 9-30. Sinervo, B. and Licht, P. 1991. Hormonal and physiological control of clutch size, egg size, and egg shape in side-blotched lizards (Uta stansburiana):Constraints on the evolution of lizard life histories. J. Exp. Zool. 257: 252-264. Sumpter, J.P., Lincoln, R.F., Bye, V.J., Carragher, J.F. and LeBail, P.Y. 1991. Plasma growth hormone levels during sexual maturation in diploid and triploid rainbow trout (Oncorhynchus mykiss). Gen. Comp. Endocrinol. 83: 103-110. Tyler, C.R. 1991. Vitellogenesis in Salmonids. In Reproductive Physiology of Fish. pp. 295-299. Edited by A.P. Scott, J.P. Sumpter, D. Kime and M.S. Rolfe. University of East Anglia Press, Norwich. Tyler, C.R., Sumpter, J.P. and Bromage, N.R. 1988. In vivo ovarian uptake and processing of vitellogenin in the rainbow trout, Salmo gairdneri. J. Exp. Zool. 246: 171-179. Tyler, C.R., Sumpter, J.P. and Handford, R. 1990a. The dynamics of vitellogenin sequestration into vitellogenic ovarian follicles of the rainbow trout, Salmo gairdneri.Fish Physiol. Biochem. 8: 211-219. Tyler, C.R., Sumpter, J.P. and Witthames, P.R. 1990b. The dynamics of oocyte growth during vitellogenesis in the rainbow trout, Salmo gairdneri. Biol. Reprod. 43: 202-209. Wallace, R.A. and Selman, K. 1981. Cellular and dynamic aspects of oocyte growth in teleosts. Am. Zool. 21: 325-343.
Effects of unilateral ovariectomy on recruitment and growth of follicles in the rainbow trout, Oncorhynchus mykiss C.R. Tylerl, J.J. Naglerl, T.G. Pottinger 2 and M.A. Turner 1 Department of Biology and Biochemistry, Brunel University, Uxbridge, Middlesex. UB8 3PH. United Kingdom; 2 Institute of FreshwaterEcology, Windermere laboratory, Far Sawrey, Ambleside, Cumbria. LA22 OLP. United Kingdom Accepted: May 13, 1994 Keywords: unilateral ovariectomy, hypertrophy, recruitment, fecundity, follicle, rainbow trout
Abstract Virgin female rainbow trout, Oncorhynchus mykiss, were unilaterally ovariectomised at various stages of ovarian development to investigate the effect of the removal of one ovary on subsequent recruitment and growth of follicles in the remaining ovary. The right ovary was removed from groups of 12-15 fish, 12, 7 and 4 months before they were due to ovulate, and the gonadosomatic index and follicle number and size determined just prior to ovulation. There were no differences in fecundity or follicle size in fish unilaterally ovariectomised at 12 and 7 months prior to ovulation compared to the controls. However, in the females unilaterally ovariectomised 4 months prior to ovulation, the remaining ovary either had the normal number of follicles for a single ovary but of a significantly larger size than follicles in the controls, or alternatively had almost 70% more than the normal number of vitellogenic follicles but comprising two distinctly different size populations. Differences in plasma oestradiol-17[5 concentrations at the final sample were seen only in females unilaterally ovariectomised 4 months prior to ovulation, where the levels were significantly lower than both the sham operated and control fish (p < 0.05). These data show that in the rainbow trout, complete compensatory ovarian hypertrophy following unilateral ovariectomy can occur throughout a major part of ovarian development, but that follicle recruitment is limited to stages up to (and therefore fecundity is determined by) mid-vitellogenesis.
Introduction In mammals, removal of one ovary, resulting in the loss of a large pool of ovarian follicles, leads to compensatory hypertrophy in the remaining ovary, which often doubles its normal ovulation rate (Greenwald 1961; Peppler and Greenwald 1970; Hirshfield 1982; Meijs-Roelofs et al. 1982). Similarly, studies in a few species of lizards (Jones et al. 1977, 1978; Sinervo and Licht 1991) and amphi-
bians (Kjaer and Jorgensen 1971; Jorgensen 1973; Billeter and Jorgensen 1976) have shown that unilateral ovariectomy (ULO) results in a hypertrophy of the remaining ovary. Unilateral removal of one testis in fish was performed in the stickleback (Gasterosteus aculeatus) as early as 1931, and resulted in compensatory hypertrophy of the remaining testis (Craig-Bennet 1931). Information was not given, however, on the changes associated with the hypertrophy. Subse-
Correspondenceto: Dr. Charles Tyler, Department of Biology and Biochemistry, Brunel University, Uxbridge, Middlesex. UB8 3PH, United Kingdom, Tel. 0895 274000 ext. 2085, Fax 0895 274348.
310 quent studies in male rainbow trout demonstrated that removal of one testis, produced as much as a two-fold increase in the mass of the remaining testis (Robertson 1958) and that this hypertrophy resulted from actively dividing germ cells (Billard et al. 1981). There are very few studies on ULO in fish, but the available evidence indicates that a compensatory hypertrophy can occur (catfish, Heteropnuestesfossilis, Goswami and Sundararaj 1968a, b; tilapia, Tilapaiaurea, Dadzie and Hyder 1976). In the rainbow trout, during each maturational cycle the ovaries grow from less than 0.2% of the body weight up to around 20% just prior to ovulation, at which time a single batch of 2000-3000 eggs per kilogram of body weight, each measuring about 5 mm in diameter, is ovulated into the body cavity. Oocyte development is divided into primary and secondary growth and each of these periods are subdivided into a number of stages (Nagahama 1983). Growth occurs principally in the secondary growth phase during vitellogenesis, when follicles increase in size from around 0.5 mm to 5 mm in diameter. This increase is largely due to uptake of extra-ovarian protein from the blood (Tyler et al. 1988, 1990b; Tyler 1991). A previous study on the dynamics of follicle growth and recruitment in the rainbow trout (Tyler et al. 1990b) has shown that there is little, if any, change in the number of follicles recruited into the secondary growth phase from the time vitellogenesis is initiated through to when the eggs are finally ovulated 9 months later, suggesting that fecundity is normally determined at or before the onset of vitellogenesis. In fish, little is known about the factors that affect oogonial proliferation, the pool size and growth pattern of previtellogenic oocytes or follicular development. Furthermore, the determinants of fecundity have yet to be firmly established in any animal. The effects of ULO on follicle growth and recruitment make it a useful tool for examining potential factors affecting the growth and recruitment of secondary growth follicles. In this study, rainbow trout were unilaterally ovariectomised at various stages during ovarian development to assess if, and when, compensatory hypertrophy could occur during the ovarian growth cycle and, if compensation did take place, to investigate the regulatory
responses in the remaining ovary in terms of the number (fecundity) and size of vitellogenic follicles produced.
Materials and methods Fish maintenance One hundred and twenty virgin 2-year-old rainbow trout, of South African origin, were divided equally between four outdoor 15001 glass fibre tanks. Each tank was supplied with a constant flow of lake water (25 1min-'). Fish were marked with an identity number by careful scarification using a mounted needle. The exposed dermis was flooded with a concentrated solution of alcian blue. During the study, which lasted more than a year, fish required re-marking. Fish were fed five times weekly with commercial food at the manufacturers recommended rate.
Unilateralovariectomy operations The study was initiated one year prior to the expected ovulation date of the fish. One tank of fish were maintained as controls (group 1), while females in the other 3 tanks were operated on in December 1992 (group 2), May 1993 (group 3) and August 1993 (group 4), 12, 7 and 4 months before they were due to ovulate, respectively. At each of these time points, 12-15 females were unilaterally ovariectomised and a similar number sham operated. Operations were conducted under 2-phenoxyethanol anaesthesia (1.2000; Sigma Chemical Company, Poole, Dorset, U.K.). An incision (approximately 2-3 cm in length) was made behind the operculum in a dorso-ventral direction to a depth such that the body cavity was just penetrated, the right ovary was located and carefully pulled through the incision using a specially designed blunt glass hook and detached from the viscera using fine scissors. After removal of the ovary the incision was closed with 4-5 sutures and the wound was covered liberally with antiseptic cream. The ovary removed from each ULO fish was weighed,
311 fixed in 70% alcohol and fifty of the largest follicles were measured to the nearest 0.05 mm using a stereoscopic binocular microscope fitted with an eyepiece graticule. Sham operated fish were treated similarly, but the right ovary was left in situ. At each sampling time, all the females in every tank were blood sampled from the Cuverian sinus using heparinised syringes and weight (g) and fork length (cm) were recorded. Blood samples were spun at 2500 x g for 15 min at 4°C and the plasma withdrawn and frozen at -70 0 C until required.
Measurements of plasma levels of oestradiol-17/3 Plasma samples were radioimmunoassayed for oestradiol-173 according to Carragher et al. (1989).
Assessing the effects of unilateralovariectomy on fecundity At the termination of the experiment (mid-December 1993), when the fish were close to ovulation, all of the experimental females were killed, blood samples were removed, and fish weighed and measured. Some of the females within the study population had already ovulated, but these were excluded from the analyses. The coefficient of condition (condition factor; K) of fish was determined using the equation K = [fish weight (g)/fish length3 (cm)] x 100, according to Sumpter et al. (1991). In the control and sham operated females, both ovaries were removed and weighed separately. In the ULO females, the remaining left ovary was removed and weighed. In a few ULO females the right ovary had not been removed completely at the time of the operation and any fish in which the remaining portion of the right ovary exceeded 5% by weight of the remaining left ovary was excluded from the analyses. Gonadosomatic indices [GSI; (total ovary weight/total body weight) x 100] were calculated for each fish. The size of follicles in each ovary was determined by measuring the diameters of 20 follicles to the nearest 0.1 mm, as described above. The number of follicles in each ovary was determined by weighing
a portion of the left ovary (around 10% of the ovary weight), counting the number of follicles present and extrapolating the number for the weight of the whole ovary (see Tyler et al. 1990a). An exception to the analyses detailed above took place in females unilaterally ovariectomised 4 months prior to ovulation. In some of these fish two different sized populations of vitellogenic follicles were present, rather than one as in all of the other groups (see results), and therefore the ovaries from all females in group 4 were fixed in 70% alcohol and approximately 20% of the follicles counted and their diameters measured to the nearest 0.1 mm. From the measurements the number of follicles present in different size classes for the whole ovary was calculated and size - frequency histograms were produced (see Tyler et al. 1990b).
Statistical analyses In this study, data are presented as mean + SEM. Significant differences between the ULO, sham and control groups were determined by ANOVA (p < 0.05) followed by a multiple comparisons test (Fisher PLSD, p < 0.05). Significant differences between the weights of the right and left ovaries in control fish and differences in the levels of plasma oestradiol-17 between females with one and two populations of vitellogenic follicles in those unilaterally ovariectomised in August were determined using Students 't' tests. All comparisons were performed using a Statview 512+ programme on an Apple Macintosh personal computer.
Results The status of ovarian development, in terms of the GSI and the size of the largest follicles present, at the times of the operations are given in Fig. 1. In December 1992 the mean GSI was 0.19 ± 0.01 and most, if not all, of the follicles (which measured less than 0.65 mm in diameter), were previtellogenic (Tyler 1991). In May 1993, the GSI was 0.36 + 0.03 and most of the maturing follicles were vitellogenic (measuring up to 1.1 mm in diameter). In August
312 Gonadosomatic Index 0.19
+
0.01
0.36 + 0.03
1.24 + 0.06
30 go
Dates of operations
25
*
'i
_
December'92
X
2
o
[3 May 3
20
C,
Augt l3
[
00 D-
06 a '
.
15' 10'
0 5. nt .
IIIll l A_
.2
.-
.3
.4
.
.5
..
.6
.
in IIi . .
.7
,_
.6
.
..
.9
.
··
1
1.1
1.2
.
II .
1.3
.
.
1.4
.
1.
l
.
I11l I q
1. 6 1.7
F.. 1.8
.
1.9
2.0
Mean diameter of largest ovarian follicles (mm). Fig. 1. Status of ovarian development, in terms of the gonadosomatic index and the size of the largest ovarian follicles, in female rainbow trout at the times of conducting unilateral ovariectomy operations in December 1992, May 1993 and August 1993.
1993, the GSI was 1.24 ± 0.06 and the maturing
group of ULO females, the remaining ovary
follicles were in mid-vitellogenesis (measuring up to 2.1 mm in diameter). At the final sampling point there were no significant differences in body weight between the ULO, sham and control fish. Similarly, there were no differences in condition factor between the ULO, sham and control fish (1.51 + 0.02, 1.52 ± 0.02 and 1.49 ± 0.03, respectively). There were no significant differences between the weights of the left and right ovary in the control fish, indicating that removal of the right ovary in the ULO fish removed half the ovarian tissue. In females unilaterally ovariectomised at 12 and 7 months prior to ovulation, at the final sampling there were no significant differences in GSI, number of follicles and follicle size compared to the sham operated and control fish (Fig. 2). In females unilaterally ovariectomised 4 months prior to ovulation, there was a statistically significant reduction in both the GSI (8.2 0.7 [ULO] vs. 11.7 ± 1.14 [sham] and 13.8 + 2.1 [controls]) and the number of follicles (1021 + 102 [ULO] vs. 1650 ± 142 [sham] and 1832 + 134 [controls]; Fig. 2). In this
showed one of two growth patterns; either the normal number of follicles for a single ovary were present (778 ± 60) that were of a significantly larger size (mean follicle diameter in these fish, 4.83 ± 0.15 mm vs. 3.99 ± 0.17 mm in the shams and 4.19 ± 0.11 mm in the controls; p < 0.05) or there were approximately 70% more than the normal number of vitellogenic follicles in the remaining ovary (1264 + 133), but these comprised two distinctly different size populations. The mean follicle diameters of these two populations were 4.1 ± 0.21, which was not significantly different from the sham-operated and controls, and 2.73 ± 0.21 which was significantly smaller (p < 0.05; Fig. 3). During the study, plasma oestradiol-173 levels increased from 0.1-0.8 ng/ml in December 1992, to 0.5-5 ng/ml in May 1993, to 2-17 ng/ml in August 1993, and up to a maximum of 70 ng/ml at the final sampling in December 1993. There were no differences in the levels of plasma oestradiol-1713 between control, sham and ULO groups, with the exception of females in group 4 at the final sampling time; here plasma levels of oestradiol-17[
313
a
C
.2 U E o0 o o o
a o
--C 1 0
2
a
01 IE Z a i
a .2 o'3 I 2
V S
a a
z
Control
ULO
S ULO
S ULO
Dec. '92
s
May '93
Aug. '93
Dates of unilateral ovarlectomy Fig. 2. Gonadosomatic index and the number and mean diameter of follicles in females just prior to ovulation. S = sham operated, ULO = unilaterally ovariectomised. * significantly differed from the sham operated group in August and the controls, p < 0.05. n: controls - 8; Dec. S - 8, ULO - 9; May S - 13, ULO - 12; Aug. S - 10, ULO - 12.
were significantly lower in the ULO fish (12.6 + 2.84 ng/ml) compared to both the sham operated (33.7 4.8 ng/ml) and control fish (32.4 5.6). In this group of ULO females, there was also a significant difference in the level of plasma oestradiol-173 between the females containing two populations of differently sized vitellogenic follicles and those containing only one (21.3 + 1.3 ng/ml vs. 6.92 3.48 ng/ml, respectively p < 0.05). Discussion In animals, the number of ovulations and eggs oviposited has evolved to ensure optimum survival
of offspring in relation to environmental pressures, and the mean number of eggs ovulated per cycle is typical of each species (Jones 1978). The results of this study on the rainbow trout demonstrate that, even during vitellogenesis, removal of half the ovarian tissue is compensated for by hypertrophy of the remaining ovary, which doubles its production of vitellogenic follicles and thus maintains a constancy in the number of eggs produced. In addition, during the study, two females were found that naturally had only one ovary and their fecundity was similarly not affected. Compensatory ovarian hypertrophy is a phenomenon seen in a wide variety of animal species, but the mechanisms by which one ovary compensates for the absence or perhaps malfunction of the other probably differ. Compensatory hypertrophy in mammals may be achieved by two mechanisms, either by reduction in the rate of atresia of antral follicles present (Greenwald 1961; Jones 1978; Hirshfield 1982; Gosden et al. 1989), or by recruitment of an additional number of small antral follicles (Peppler and Greenwald 1970; Osman et al. 1982; Meijs-Roelofs et al. 1984). Unlike the situation in mammals, however, where the number of oogonia is set and indeed decreasing even before birth, in fish oogonial proliferation occurs throughout life (Billard et al. 1982; Billard 1987), although in the trout it is usually limited to short periods during or just after ovulation each year (Franchi et al. 1962). In the tilapia and Indian catfish, partial ovariectomy has been shown to increase the number of dividing oogonia (Goswami and Sundararaj 1968a; Dadzie and Hyder 1976). Similarly in this study on the rainbow trout, ULO may have activated germ cell division to reinitiate a 'normal' gametogenesis. In other batch spawning fish, however, it has been shown that in a vitellogenic ovary there is also an asynchronous population of pre-vitellogenic follicles and these may serve as a pool for further follicle recruitment into the maturing pool (see Wallace and Selman 1981). The ability of the trout ovary to recruit another population of follicles into the secondary growth phase, by whatever mechanism, even as late in development as mid-vitellogenesis, probably functions normally to balance any loss from the pool of developing follicles.
314
a
2 ar
0
a
0
a
IO
oI
Follicle dameter (mm)
Fig. 3. Variation in the sizes of follicles in females unilaterally ovarectomised in August. Fish 1-6 inclusive did not recruit a second population of follicles into vitellogenesis, fish 7-12 did. The number of follicles in an ovary in each 100 gm-diameter size class is expressed as a percentage of the total number of maturing follicles in that ovary.
Lacker et a. (1987), suggested that in mammals all follicles inherit the same developmental programme and the maturation rate of each follicle at any time depends on it's maturity and the concentration of circulating oestrogens and gonadotrophin. In the rainbow trout, in contrast, follicles of a very similar size appear to be able to grow at different rates (Tyler et al. 1990a). In this study, follicles recruited from oogonia/follicles in the previtellogenic pool into the secondary growth phase following unilateral ovariectomy in May could have differed in size by up to 200 thousand fold (by volume) compared to the larger vitellogenic population, yet they were the same size as their earlierdeveloping counterparts at ovulation, 7 months later. Enhanced growth rates of this 'new' population of developing follicles is likely to be a function, at least in part, of the more advanced reproductive status of the female (increased plasma levels of both
oestradiol-175, which in May were up to 5 ng/ml, and vitellogenin, etc.) compared to that which they would normally encounter at the beginning of the secondary growth phase. However, it is likely that there are also intraovarian factors that participate in the control of follicle growth (Tyler et al. 1990b). These data illustrate that growth of follicles can proceed at rates considerably faster than those normally observed. Mechanisms controlling follicle recruitment per se and recruitment after unilateral ovariectomy are yet to be determined in any animal. In a number of mammals, however, ULO induces an increase in levels of serum follicle stimulating hormone (FSH) and this in turn causes compensatory growth of follicles in the remaining ovary (Jones 1978; for review, Osman et al. 1982). Recent evidence suggests that sensory innervation may also play a role in ovarian growth and development (Deshmukh et al.
315 1992). In non-mammalian vertebrates there is evidence that growth hormone restores follicle recruitment in ULO toads on a restricted diet (a restricted diet inhibits ovarian hypertrophy; Billeter and Jorgensen 1976), but the absence of homologous FSH (gonadotrophin-I) assays, until very recently, have precluded assessing the role of gonadotrophin in follicle growth and recruitment. The compensatory systems controlling follicle numbers in the rainbow trout, however, appear to be limited to stages of ovary development up to around mid-vitellogenesis; half of the females unilaterally ovarectomised in August did not recruit a second population of follicles into the maturing pool and the other half of the females that did only partially compensated in that they recruited only 70%o more follicles into the maturing population rather than doubling the number. Females that did not recruit a second population of follicles into vitellogenesis produced follicles that were of a significantly larger size; 76% larger by volume compared to the controls. These data indicate that the size of the eggs produced in rainbow trout is determined, within the limits of variance set by genetic constraints, by the number of follicles undergoing vitellogenesis. The two populations of vitellogenic follicles found in half of the females unilaterally ovariectomised in group 4 were of widely divergent sizes at the final sampling (a mean difference in volume of over 3.5-fold). It has not been established whether a trout egg has to reach a critical size before it is ovulated and since none of the females in group 4 had ovulated, we were not able to determine the fate of the population of smaller follicles; they may ovulate as small eggs at the normal ovulation time, they may undergo atresia, or alternatively they may delay the timing of ovulation such that they are able to reach a size approaching that of their larger counterparts. The lower levels of plasma oestradiol-173 seen in the ULO females, compared to both the controls and sham operated fish in group 4 at the final sampling may simply be a reflection of the reduced amount of ovarian tissue (and hence a reduced amount of steroid-secreting granulosa and thecal cells). Within this group of ULO females there was also a clear difference in the levels of plasma
oestradiol-1713 between the females containing two populations of maturing follicles and those containing only one. The higher titre of plasma oestradiol-173 in the females containing two populations of maturing follicles probably results from both the greater number of vitellogenic follicles and that the population of smaller vitellogenic follicles would be more actively secreting oestradiol-171B compared to the larger follicles that were close to full maturity (see Kanamori et al. 1988 for seasonal pattern of oestradiol-173 secretion). The level of plasma oestradiol-17 in the ULO females in group 4 that contained only one population of maturing follicles is somewhat lower than that anticipated compared to the controls, even when the reduced amount of ovarian mass is accounted for, and this may indicate that these fish were closer to ovulation than their counterparts in the control groups (see Kanamori et al. 1988). Jansen (1994) suggested that in mammals, two ovaries exist for anatomical symmetry and a 'dab of insurance' should one fail, because for virtually all physiological purposes they act as one. This hypothesis seems also to apply to the rainbow trout, where females with only one ovary can carry out gametogenesis normally and are equally as fecund as females with two ovaries. Follicle recruitment into the secondary growth phase in the rainbow trout, and therefore a final decision, however, is limited to the stages of development up to mid vitellogenesis.
References cited Billard R. 1987. The reproductive cycle of the male and female brown trout (Salmo trutta fario); a quantitative study. Reprod. Nutr. Develop. 27: 29-44. Billard, R., Crim, L.W., Peter, R.E. and Breton, B. 1982. Longterm changes in plasma and pituitary GtH after castration of rainbow trout at an immature stage. In Reproductive Physiology of Fish. pp. 50. Edited by C.J.J. Ritcher and H.J.Th. Goos. Pudoc Press, Wageningen. Billeter, E. and Jorgensen, C.B. 1976. Ovarian development in young toads, Bufo bufo bufo (L.): Effect of unilateral ovariectomy, hypophysectomy, treatment with gonadotropin (hCG), growth hormone and prolactin, and importance of body growth. Gen. Comp. Endocrinol. 29: 531-544. Carragher, J.C., Sumpter, J.P., Pottinger, T.G. and Pickering,
316 A.D. 1989. The deleterious effects of cortisol implantation on reproductive function in two species of trout, Salmo truttaL. and Salmo gairdneri Richardson. Gen. Comp. Endocrinol. 76: 310-321. Craig-Bennett, A. 1931. The reproductive cycle of the three spined stickleback, Gasterosteusaculeatus Linn. Phil. Trans. Roy. Soc. London Ser. 219: 197-279. Dadzie, S. and Hyder, M. 1976. Compensatory hypertrophy of the remaining ovary and the effects of methallibure in the unilaterally ovariectomized Tilapia aurea. Gen. Comp. Endocrinol. 29: 433-440. Deshmukh, M.K., Krishna, R.N.S. and Subhedar, N. 1992. Ovarian stretch signals trigger volume increase in the magnocellular preoptic nuclear region in the brain of the catfish, Clariasbatrachus (Linn.). J. Exp. Zool. 263: 231-234. Franchi, L.L., Mandle, A.M. and Zuckermann, S. 1962. The development of the ovary and the process of oogenesis. In The Ovary. pp. 1-88. Edited by R.E. Jones. Academic Press, New York. Gosden, R.G., Telfer, E., Faddy, M.J. and Brook, D.J. 1989. Ovarian cyclicity and follicular recruitment in unilaterally ovariectomized mice. J. Reprod. Fertil. 87: 257-264. Goswami, S.V. and Sundararaj, B.I. 1968a. Compensatory hypertrophy of the remaining ovary after unilateral ovariectomy at various phases of the reproductive cycle of catfish, Heteropnueustes fossilis (Bloch). Gen. Comp. Endocrinol. 11: 401-413. Goswami, S.V. and Sundararaj, B.I. 1968b. Effect of estradiol benzoate, human chorionic gonadotrophin, and folliclestimulating hormone on unilateral ovariectomy-induced compensatory hypertrophy in catfish, Heteropnueustes fossilis (Bloch). Gen. Comp. Endocrinol. 11: 393-400. Greenwald, G.S. 1961. Quantitative study of follicular development in the ovary of the intact or unilaterally ovariectomised hamster. J. Reprod. Fertil. 2: 351-361. Hirshfield, A.N. 1982. Follicular recruitment in long term hemicastrate rats. Biol. Reprod. 27: 48-53. Jansen, R. 1994. Follicles in competition. Nature, Lond. 367: 10. Jones, R.E. 1978. Control of follicular selection. In The Vertebrate Ovary: Comparative Biology and Evolution. pp. 763788. Edited by R.E. Jones. Plenum Press, New York. Jones, R.E., Fitzgerald, K.T. and Duvall, D. 1978. Quantitative analysis of the ovarian cycle of the lizard Lepidodactylus lugubris. Gen. Comp. Endocrinol. 35: 70-76. Jones, R.E., Fitzgerald, K.T. and Tokarz, R.R. 1977. Endocrine control of clutch size in reptiles. VII. Compensatory ovarian hypertrophy following unilateral ovarectomy in Sceloporus occidentalis. Gen. Comp. Endocrinol. 31: 157160. Jorgensen, C.B. 1973. Pattern of recruitment of oocytes to second growth phase in normal toads, and in hypophysectomised toads Bufo bufo bufo (L), treated with gonadotrophin (hCG). Gen. Comp. Endocrinol. 21: 152-159. Kanamori, A.K., Adachi, S. and Nagahama, Y. 1988. Developmental changes in steroidogenic responses of ovarian follicles of amago salmon (Oncorhynchus rhodurus)to chum salmon
gonadotropin during oogenesis. Gen. Comp. Endocrinol. 72: 13-24. Kjaer, K. and Jorgensen, C.B. 1971. Effects of mammalian gonadotropins on ovary in the toad Bufo bufo bufo (L). Gen. Comp. Endocrinol. 17: 424-431. Lacker, H.M., Beers, W.H., Meuli, L.E. and Akin, E. 1987. A theory of follicle selection: I Hypotheses and examples. Biol. Reprod. 37: 570-580. Meijs-Roelofs, H.M.A., Kramer, P., Osman, P. and Sander, H.J. 1984. Compensatory ovulatory mechanisms operative after first ovulation in rats unilaterally ovariectomised prepuberally. Biol. Reprod. 31: 44-51. Meijs-Roelofs, H.M.A., Osman, P. and Kramer, P. 1982. Ovarian follicular development leading to first ovulation and accompanying gonadotrophin levels as studied in the unilaterally ovariectomized rat. J. Endocrinol. 92: 341-349. Nagahama, Y. 1983. The functional morphology of teleost gonads. In Reproduction-Endocrine Tissues and Hormones. pp. 223-276. Edited by W.S. Hoar, D.J. Randall and E.M. Donaldson. Academic Press, New York. Osman, P., Meijs-Roelofs, H.M.A. and Kramer, P. 1982. Compensatory follicle growth and ovulation after unilateral ovariectomy during the late prepubertal period in the rat. In Development and Function of Reproductive Organs. pp. 299-306. Edited by A.G. Byskov and H. Peters. Exerpta Medica, Amsterdam. Pepper, R.D. and Greenwald, G.S. 1970. Influence of unilateral ovariectomy on follicular development in cycling rats. Am. J. Anatomy 127: 9-14. Robertson, M.D. 1958. Accelerated development of testis after unilateral gonadectomy, with observations of the normal testes of rainbow trout. Bull. Fish Wildl. Serv. 58: 9-30. Sinervo, B. and Licht, P. 1991. Hormonal and physiological control of clutch size, egg size, and egg shape in side-blotched lizards (Uta stansburiana):Constraints on the evolution of lizard life histories. J. Exp. Zool. 257: 252-264. Sumpter, J.P., Lincoln, R.F., Bye, V.J., Carragher, J.F. and LeBail, P.Y. 1991. Plasma growth hormone levels during sexual maturation in diploid and triploid rainbow trout (Oncorhynchus mykiss). Gen. Comp. Endocrinol. 83: 103-110. Tyler, C.R. 1991. Vitellogenesis in Salmonids. In Reproductive Physiology of Fish. pp. 295-299. Edited by A.P. Scott, J.P. Sumpter, D. Kime and M.S. Rolfe. University of East Anglia Press, Norwich. Tyler, C.R., Sumpter, J.P. and Bromage, N.R. 1988. In vivo ovarian uptake and processing of vitellogenin in the rainbow trout, Salmo gairdneri. J. Exp. Zool. 246: 171-179. Tyler, C.R., Sumpter, J.P. and Handford, R. 1990a. The dynamics of vitellogenin sequestration into vitellogenic ovarian follicles of the rainbow trout, Salmo gairdneri.Fish Physiol. Biochem. 8: 211-219. Tyler, C.R., Sumpter, J.P. and Witthames, P.R. 1990b. The dynamics of oocyte growth during vitellogenesis in the rainbow trout, Salmo gairdneri. Biol. Reprod. 43: 202-209. Wallace, R.A. and Selman, K. 1981. Cellular and dynamic aspects of oocyte growth in teleosts. Am. Zool. 21: 325-343.
