Observations on variability in LH release and fertility during oestrus in the domestic cat (Felis catus)

Transcription

1 Observations on variability in LH release and fertility during oestrus in the domestic cat (Felis catus) T. E. Glover, P. F. Watson and R. C. Bonney Department ofphysiology, Royal Veterinary College, Royal College Street, London NW1 OTU, and * Wellcome Laboratories, Institute ofzoology, Zoological Society of London, Regent's Park, London NW14RY, U.K. Summary. Hormonal changes, behaviour, ovulation and fertility were examined in response to coitus at two different times during oestrus in the female domestic cat housed in conditions of natural light (N 13). On Day 2 = or Day 4/5 of oestrus females were allowed 1 copulation in 15 min (single matings) or 2\p=n-\3 copulations in 30 min (multiple matings). Plasma LH, oestradiol-17\g=b\and progesterone concentrations during the 24-h period after coitus were measured by radioimmunoassay; ovulation was assumed to have occurred if progesterone values were elevated 7\p=n-\30days after coitus. With the exception of 2 out of 3 animals receiving single matings on Day 2 of oestrus, all animals showed subsequent elevated progesterone values. Females receiving multiple matings had significantly greater releases of LH as measured by the area under the curve than those receiving single matings. There was significantly greater variability in the LH response of queens on Day 2 of oestrus compared to those on Day 4/5 for peak values and area under the curve; the only failure in release of LH was in queens on Day 2. Oestradiol levels did not differ significantly between Day 2 and Day 4/5 of oestrus. Progesterone values remained < 1 ng/ml for 24 h after coitus. Both LH peak values and area under the curve were significantly greater for animals that became pregnant. There were also significant differences in coital behaviour between queens on Day 2 and those on Day 4/5 of oestrus. It is suggested that LH release responsiveness rises during early oestrus so that the magnitude of an LH surge for a given stimulation depends on when, during oestrus, coitus occurs. Further major LH releases are then suppressed by a refractory state. Introduction The domestic cat is a reflex or induced ovulator (Longley, 1911), the magnitude and duration of the LH surge being influenced by the number of and intervals between copulations (Concannon, Hodgson & Lein, 1980; Wildt, Seager & Chakraborty, 1980; Johnson & Gay, 1981b) and by the day of oestrus when animals are mated on sequential days of oestrus (Banks & Stabenfeldt, 1982). Variations in ovulatory response (Banks & Stabenfeldt, 1982) and oestradiol concentrations (Paape, Shille, Seto & Stabenfeldt, 1975; Verhage, Beamer & Brennen, 1976; Shille, Lundstrom & Stabenfeldt, 1979; Banks & Stabenfeldt, 1982) during oestrus have also been reported. Elevation of peripheral plasma progesterone concentrations occur only after mating which results in ovulation and the formation of corpora lutea (Paape et al, 1975; Shille & Stabenfeldt, 1979). This present work was part of a study begun in 1979 to investigate the feasibility of utilizing the domestic cat as are reproductive model for non-domestic felids (Watson & Glover, 1980). Our aim Present address: Department of Chemical Pathology, St Mary's Hospital Medical School, London W2 1PG, U.K.

2 was to investigate possible variations in reproductive cycles that might affect controlled or captive breeding under conditions that would permit comparisons with non-domestic felids. Animals Materials and Methods Maintenance. The cats were part of a closed colony attended daily by a technician, and fed a commercial canned diet. They were maintained in a natural light cycle, with unrestricted access to water and litter trays. Unmated females were housed in a harem arrangement with free access to an outdoor run and separated from the males by a double wire partition. Mated females were housed individually in a separate nursery. Males were housed individually. Behaviour. Females were monitored daily in the presence of a male between 10:00 and 12:30 h by a trained animal technician. Females were judged to be approaching oestrus when lordosis, rolling and/or rubbing were detected. The coital behaviours observed were: the time of approach by the male (i.e. the time from release of the male to mounting), the duration of mounting, and the duration of the coital vocalization (yowl) which corresponds with the duration of intromission (Whalen, 1963). Mating regimen. After the first oestrus of the season, females were mated in the period of January-July between 10:30 and 12:30 h, within that section of the cattery utilized for the males. The animals were permitted either a single mating or multiple mating (2 or 3 intromissions within 30 min). Successful intromission was judged by the utterance of a characteristic loud vocalization by the female, subsequent disengagement of the male, and typical post-coital behaviour (Michael, 1961, 1973; Whalen, 1963). Fertility. The proportion of queens littering and the litter size were recorded as indicators of fertility. Blood sampling. Cephalic venepuncture was performed at 15 min and immediately before mating (-15 and 0 min samples), and at 15 (single mating only), 30, 45, 60, 90, 120, 180, 240min and 24 h post coitum, with two subsequent samples collected 1 week apart between 7 and 30 days post coitum. Samples obtained after mating ( min) were obtained from animals restrained with ketamine-hcl (10mg/kg; Vetalar: Park Davis & Co., Pontypool, Gwent NP4 0YH, U.K.). Blood samples (2-3 ml) were collected into heparin-primed tubes, centrifuged, and the plasma removed and stored frozen until assayed. Hormone assays and ovulation AU the samples from a given animal were assayed at one time. LH. Plasma LH was determined by a double-antibody radioimmunoassay for ovine LH using an antiserum (GDN15) provided by Dr G. Niswender. Radioimmunoassays using this antiserum have been described in detail by Niswender, Reichert, Midgley & Nalbandov (1969) and Millar & Aehnelt (1977). The antiserum does not cross-react with other pituitary hormones and has been used to measure LH in a variety of mammalian species (Miller & Aehnelt, 1977) including the domestic cat (Concannon et al., 1980). Ovine LH (LER-1056-C2), provided by Dr L. Reichert, was iodinated by the chloramine method of Greenwood, Hunter & Glover (1963). The antiserum was used at a final dilution of 1: at approximately 40% binding in the absence of unlabelled hormone. Ovine LH, NIH-LH-S21 (biopotency 2-5 NIH-LH-S1), was used as a reference prep aration. The details of the method were as described by Follett, Scanes & Cunningham (1972). The standard curve ranged from 5 to 0019 ng/ml and the detection limit of the assay in this study was 0-2 ng/ml. The intra- and inter-assay coefficients of variation were 5-9% (n = 9) and 11-7% (n = 13) respectively, measured near the midpoint (50% binding) of the standard curve. When a range of volumes ( µ ) of a single plasma sample were assayed then a linear relationship was

3 obtained between volume and amount of hormone measured (correlation coefficient, r 0-999); = the linear regression was described by 8-85* 217. = Oestradiol-17ß. Plasma oestradiol- 17ß concentration was measured as previously reported (Bonney, Moore & Jones, 1981) using antiserum raised in a rabbit to oestradiol-17ß-6-(0- carboxymethyl)-oxime-bovine serum albumin (Steranti Research Ltd, St Albans, Herts, U.K.) diluted according to the manufacturer's instructions. Cross-reactions with other oestrogens were: oestrone, 0-6%; oestradiol-17a, 0-5%, and oestriol, 0-4%. Procedural losses were determined and corrected for in each assay by the recovery of [3H]oestradiol added to a pool of cat plasma (mean recovery; %). Buffer blanks were extracted with each batch of samples and a correction was made when they exceeded the limit of sensitivity of the assay (10-3 ± 2-7 pg/tube; 13). The = inter- and intra-assay coefficients of variation calculated from estimations of a cat plasma pool in a single assay and in 10 separate assays were 9-3% (n 11) and 6-6% respectively, measured = near the midpoint of the standard curve. Accuracy was determined by the addition of known amounts of oestradiol (200, 100, 50 and 25 pg/tube) to male cat plasma. The linear regression for estimated oestradiol against added oestradiol was y 0-92* = Progesterone. Concentrations of progesterone were measured as reported elsewhere (Bonney et al, 1981). The final dilution of antiserum used was 1:8000. Inter- and intra-assay coefficients of variation, as established by repeated assay of a cat plasma pool in a single assay and in separate assays, were 14-3% (n 11) and 91% (n 5) respectively, measured = = near the midpoint of the standard curve. Buffer blanks were extracted with each assay but were always less than the sensitivity of the assay ( pg/tube, 10). Accuracy = was determined by the addition of known amounts of progesterone (1, 0-5, 0-25 and 0125 ng/tube) to male cat plasma. The linear regression for estimated progesterone against added progesterone was y 0-90x = Ovulation was assumed to have occurred when there was a post-copulatory rise in plasma pro gesterone concentration to > 5 ng/ml from two samples collected 1 week apart between 7 and 30 days/705? coitum. Experimental design Two groups of animals mated at different stages of oestrus were compared for LH release, oestradiol and progesterone levels, coital behaviour, ovulation, and fertility. Seven animals were mated on Day 2 of oestrus (Group I) and 6 on Day 4/5 (Group II). Each group was further divided into those animals mated singly ( 1) and those mated multiply ( 2-3 within 30 min). Day 1 of oestrus was considered to be the first day on which the female displayed characteristic hind-limb treading movements in the presence of the male. (Treading rather than the more commonly used criterion of mounting was used to avoid a premature LH release which might occur in response to mounting (Concannon et al, 1980).) Three aspects of the LH profiles were considered: the maximal or peak value recorded, the time to attain the peak value, and the area under the curve measured up to 4 h after mating. Changes in oestradiol and progesterone concentrations with mating were considered and pre-mating values of oestradiol were compared between Groups I and II. Statistical analysis The hormone data were analysed using a two-tailed variance ratio test and Student's unpaired t test. When variances differed significantly between groups a modified / test was used (Snedecor & Cochran, 1967). The possibility of correlation between LH release and the pre-mating oestradiol value was examined by means of a correlation coefficient. LH Results ng/ml before The LH release profiles are shown in Text-fig. 1. All animals had LH levels below 1 mating, with the exception of Females 5 and 7 for which the values immediately before mating were

4 ( ui/ßu) HT

5 and 6-5 ng/ml respectively. Two animals failed to release LH in response to mating (Nos 9 and 4, both in the single mating, Day 2 group); all other females had an increase in LH concentrations after mating. Excluding these 2 animals, the mean area under the curve was significantly less for singly mated animals than for multiply mated animals (P < 005), but there was no significant difference in mean peak values (10-80 ± 3-93 ng/ml and ± 5-33 ng/ml respectively, mean + s.e.m.) or in the time to reach peak values. In comparing Groups I and II no significant differences were detected in mean peak values, mean time to reach peak values, or area under the curve. How ever, within the sub-group receiving an adequate stimulus (i.e. multiple matings) there was signifi cantly greater variability (P < 005) in peak values of Group I females (range: ng/ml) compared with those of Group II females (range: ng/ml) and also for the area under the curve(p < 005). Oestradiol and progesterone The pre-mating oestradiol values were extremely variable, but the mean values for Groups I and II did not differ significantly. In ovulating queens, concentrations of oestradiol were lower 24 h post coitum than before mating ( compared with 620 ± 5-41 pg/ml, respectively, a differ ence which approached significance, 010 > > 005). Progesterone values for all females from before mating to 24 h after mating remained at 1 ng/ml or less. Correlation coefficients between pre-mating levels of oestradiol and the LH peak values or the area under the curve were not significant. Ovulation ng/ml in the The 2 animals that failed to give an LH response had progesterone values < 1 30-day period after mating, whereas the remaining animals had progesterone values between 11-3 and 37-8 ng/ml, and had therefore probably formed corpora lutea (Table 1). Fertility Pseudopregnancy (considered herein to be the occurrence of elevated plasma progesterone values without subsequent parturition) occurred in 5/11 queens (2/5 Group I, 3/6 Group II; Table 1). Multiply mated animals had greater fertility than singly mated animals (5/7 littering versus 1/6). Table 1. Ovulation and fertility of cats mated singly or multiply on Day 2 (Group I) or Day 4/5 (Group II) of oestrus (see text for details of determining day of oestrus) Mating regimen Group Animal Ovulation Pregnancy Litter size Single I II Multiple I II

6 Among multiply mated queens pseudopregnancy occurred only in Group I. Both the mean LH peak value and the mean area under the curve for the animals that became pregnant were significantly higher than those for the females that became pseudopregnant (P < 005). Behaviour There was a significant difference in the mean duration of coital mountings between Groups I and II (P < 001; Table 2), the animals in Group II remaining mounted longer. Significantly greater variances for the time of approach to mounting (P < 001) and the duration of the vocalization (P < 005) were found in Group I compared with Group II. Table 2. Coital behaviour of cats mated on Day 2 (Group I, = 6, = 12) or Day 4/5 (Group II, = 7, = 11) of oestrus Time of approach of male to mounting the female (sec) Duration of mounting (sec) Duration of the vocalization by the female (sec) Values are mean + s.e.m. Group I ± 6-3 Group II 5-2 ± ± ± 20 Discussion A significantly greater amount of LH was released (as measured by the area under the curve) in those females receiving multiple matings than in those mated singly, and the peak values of females receiving multiple matings were generally higher than in those females receiving single matings, although the difference did not attain statistical significance. A failure of LH response was only found in Group I (single matings) which corresponds with the data of Banks & Stabenfeldt (1982) who found a lack of LH response only in early oestrus in some females. As one of the purposes of this study was to gather data which could be compared with data obtained from non-domestic felids, it was felt desirable to gather the data under conditions similar to those used for non-domestic felids, i.e. anaesthesia. The anaesthetic used was ketamine-hcl which is one of the anaesthetic agents most widely used with non-domestic felids. The observations presented here on the LH surge response correlate well with studies in which anaesthesia was not used (Concannon et al, 1980; Johnson & Gay, 1981b; Banks & Stabenfeldt, 1982), and give no grounds for believing that the surge response was suppressed by the administration of ketamine, although suppression of tonic LH release in the cat has been demonstrated by Johnson & Gay (1981a). The dosage used in their work, however, was far greater than that used in the present study (30 mg ketamine/kg body weight compared with 10 mg/kg). The significantly greater variability in the LH responses of animals mated earlier in oestrus compared to the LH responses of those mated later we consider to be of biological importance. Our data suggest that it is an inherent feature of the oestrous cycle in the cat that there is considerable individual variation in responsiveness of the LH release mechanism early in oestrus. The respon siveness rises in early oestrus, but the rise need not coincide with the onset of behavioural signs (Johnson & Gay, 1981b; Banks & Stabenfeldt, 1982; see above, Group I, single matings). The responsiveness of the LH release mechanism could diminish within about 24 h giving a single day on which an ovulatory surge could be elicited. Alternatively, the LH release responsiveness might remain high for several days in the absence of a mating stimulus, but a refractory state could follow the ovulatory LH surge. Either of these alternatives would be consistent with the data of Banks &

7 Stabenfeldt (1982) who found, when mating cats daily, that the general pattern during oestrus was a single major LH surge, preceded and followed by lesser, non-ovulatory surges or by no surge at all. Our data, however, support the second alternative rather than the first. In the early stages of oestrus considerable variability in response to a comparable stimulus was seen between animals since they varied in the stage of responsiveness. Were the high LH release responsiveness to be con fined to a single 24 h period, a large between-animal variability in response would be expected throughout oestrus. However, in our study at mid-oestrus a comparable stimulus evoked a con sistently high LH release, suggesting that all the animals had reached peak responsiveness. Fur thermore, a refractory state as implied in the second alternative, was also suggested by Johnson & Gay (1981b) and by the work of Concannon, Lein & Hodgson (1982) who found that queens mated ad libitum for 35 h starting on Day 2 of oestrus showed no further increases of LH in response to copulation(s) after the 10th hour. This failure of LH response was not due to pituitary depletion of LH; when LHRH was administered at the 36th hour a rise in LH was obtained (Concannon et al, 1982). We have no evidence whether, in the absence of mating, the duration of responsiveness would extend until the cessation of behavioural signs of oestrus, and/or whether it may vary with the individual and cycle. Two females (Nos 5 and 7) had elevated LH concentrations before mating, raising the possibility of spontaneous LH release. While some induced ovulators have an incidence of spontaneous LH release and ovulation (Milligan, 1982), such has not been reported for the cat, apart from Bonnet (1897, quoted in Longley, 1911). In the harem situation in which the animals were housed the females had free access to one another, and some animals were seen either to mount or to submit to mounting during oestrus, including these two females. It is possible that the LH levels were due to such homosexual behaviour, since Concannon et al (1980) reported LH responses from mounting without intromission. Both of these animals were in Group II in which LH responsiveness is likely to have been high. The total lack of progesterone response over the 24 h from copulation agrees with other find ings (Shille et al, 1979; Shille, Munro, Walker-Farmer, Papkoff & Stabenfeldt, 1983; Concannon et al, 1980; Wildt et al, 1980; Wildt, Chan, Seager & Chakraborty, 1981) and indicates no preovulatory luteinization of the follicle, such as has been found in the bitch (Concannon, Cowan & Hansel, 1979). The behavioural observations and fertility results also support the concept that responsiveness at oestrus is not constant, with the greater variability in behavioural responses in early oestrus and the longer duration of mounting in Group II probably reflecting the increased sexual activity of females in mid-oestrus. However, more sophisticated behavioural studies are required to verify this suggestion. The only failures to show evidence of ovulation were in Group I (Nos 9 and 4 which showed no LH surge). Of the multiply mated females, all of which subsequently revealed elevated progesterone values, pseudopregnancies were found only in Group I. Indeed, all pseudopregnancies were associated with conditions of reduced LH response or responsiveness, i.e. Group I or singly mated animals. It is possible that some or all of these animals failed to ovulate but developed luteinized follicles. However, in an earlier laparoscopie study all cats that showed an LH surge ovulated, and in individual queens all observed antral follicles subsequently formed corpora lutea (Wildt et al, 1980). An alternative explanation, that a diminished LH surge may be adequate to induce ovulation but insufficient to promote maturation of the oocyte and/or fertilization and subsequent development, deserves further study. The cat most probably has several overlapping, but variable and not necessarily coinciding, fac tors affecting fertility during oestrus, including sexual behaviour, oestradiol concentrations and LH responsiveness. Oestradiol concentrations rise through pro-oestrus and are probably not constant during oestrus (Verhage et al, 1976; Shille et al, 1979; Banks & Stabenfeldt, 1982). Oestradiol might be expected to influence the uterine or tubai environment and sperm and egg transport. The most fertile matings would occur when all of these factors are favourable, a situation which is

8 probably less of a problem in the natural environment than in a controlled or captive environment, where environmental cues may be abnormal and interfere with the co-ordination of events. Mating (natural or artificial) towards the middle of oestrus probably yields the greatest chances of conception. We are indebted to Mrs Anne Anderson for expert management of the cat colony and for technical assistance with this work. References Banks, D.H. & Stabenfeldt, G. (1982) Luteinizing hor mone release in response to coitus on consecutive days of estrus. Biol. Reprod. 26, Bonney, R.C, Moore, H.D.M. & Jones, D.M. (1981) Plasma concentrations of oestradiol-17ß and pro gesterone, and laparoscopie observations of the ovary in the puma (Felis concolor) during oestrus, pseudopregnancy and pregnancy. J. Reprod. Fert. 63, Concannon, P., Cowan, R. & Hansel, W. (1979) LH release in ovariectomized dogs in response to estro gen withdrawal and its facilitation by progesterone. Biol. Reprod. 20, Concannon, P., Hodgson, B. & Lein, D. (1980) Reflex ovulation in estrous cats following single and multiple copulations. Biol. Reprod. 23, Concannon, P., Lein, D. & Hodgson, B. (1982) Sexual behavior and self-limiting reflex LH release during extended periods of ad libitum copulatory activity in domestic cats. J. Anim. Sci. 55, Suppl. 1, 344, Abstr. FoUett, B.K., Scanes, CG. & Cunningham, F.J. (1972) Radioimmunoassay for avian LH. J. Endocr. 51, Greenwood, F.C., Hunter, W.M. & Glover, J.S. (1963) The preparation of 131I-labelled human growth hor mone of high radioactivity. Biochem. J. 89, Johnson, L.M. & Gay, V.L. (1981a) Luteinizing hormone in the cat. I. Tonic secretion. Endocrinology 109, Johnson, L.M. & Gay, V.L. (1981b) Luteinizing hor mone in the cat. II. Mating-induced secretion. Endo crinology 109, Longley, W.H. (1911) The maturation of the egg and ovulation in the domestic cat. Am. J. Anat. 12, Michael, R.P. (1961) Observations upon the sexual behavior of the domestic cat (Felis catus L.) under laboratory conditions. Behaviour 18, Michael, R.P. (1973) The effects of hormones on sexual behavior in the female cat and rhesus monkey. In Handbook of Physiology, Section 7, Vol. II, Female Reproductive System: Part I, pp Ed. R. O. Greep. American Physiological Society, Washington, D.C. Millar, R.P. & Aehnelt, C. (1977) Application of ovine luteinizing hormone (LH) radioimmunoassay in the quantitation of LH in different mammalian species. Endocrinology 101, Milligan, S.R. (1982) Induced ovulation in mammals. Oxford Revs Reprod. Biol. 4, Niswender, G.D., Reichert, L.E., Midgley, A.R., Jr & Nalbandov, A.V. (1969) Radioimmunoassay for bov ine and ovine luteinizing hormone. Endocrinology 84, Paape, S.R., Shille, V.M., Seto, H. & Stabenfeldt, G.H. (1975) Luteal activity in the pseudopregnant cat. Biol. Reprod. 19, Shille, V.M. & Stabenfeldt, G.H. (1979) Luteal function in the domestic cat during pseudopregnancy and after treatment with prostaglandin F2a. Biol. Reprod. 21, Shille, V.M., Lundstrom, K.E. & Stabenfeldt, G.H. (1979) Follicular function in the domestic cat as determined by estradiol-17ß concentrations in plasma: relation to estrous behaviour and cornification of exfoliated vaginal epithelium. Biol. Reprod. 21, Shille, V.M., Munro, C, WaUter-Farmer, S., Papkoff, H. & Stabenfeldt, G.H. (1983) Ovarian and endocrine responses in the cat after coitus. /. Reprod. Fert. 69, Snedecor, G.W. & Cochran, W.G. (1967) Statistical Methods, 6th edn. Iowa State University Press, Ames. Verhage, H.G., Beamer, N.B. & Brennen, R.M. (1976) Plasma levels of estradiol and progesterone in the cat during polyestrus, pregnancy and pseudopregnancy. Biol. Reprod. 14, Watson, P.F. & Glover, T.E. (1980) The domestic animal model as a basis for artificial breeding in non-domestic animals. Proc. 9th Int. Congr. Anim. Reprod. & A.I., Madrid2, Whalen, R. (1963) Sexual behaviour of cats. Behaviour 20, Wildt, D.E., Seager, S.W.J. & Chakraborty, P.K. (1980) Effect of copulatory stimuli on incidence of ovulation and on serum luteinizing hormone in the cat. Endo crinology 107', Wildt, D.E., Chan, S.Y.W., Seager, S.W.J. & Chakraborty, P.K. (1981) Ovarian activity, circulat ing hormones and sexual behavior in the cat. I. Rela tionships during the coitus-induced luteal phase and the estrous period without mating. Biol. Reprod. 25, Received 17 December 1984