American Journal of Primatology Supplement 1:15-25 (1982) Ovulation Detection and Artificial Insemination K.G. GOULD Yerkes Regional Primate Research Center, Emory University, Atlanta Methods of detecting ovulation and conducting artificial insemination in great apes are reviewed and described. Ovulation time can be detected to some extent by measuring timing relative to such cyclic correlates as menstruation, perineal swelling, body temperature, viscosity of cervical mucus, and behavior (copulation, masturbation, aggressivity). The most precise methods are observation of ovarian morphology using laparoscopy or ultrasonography, but the most practical method for great apes is the use of recently developed rapid hemagglutination inhibition test for luteinizing hormone. Using this test, three chimpanzees and one gorilla have been successfully artificially inseminated. Artificial insemination in great apes involves collection of semen by automasturbation or rectal probe electro-ejaculation. Females are anesthetized and fractionated ejaculate is placed on the cervical 0s. Complicating factors include potential effect of anesthesia, variation in semen quality, frequency of A1 required, and the apparent inherent infertility potential of great apes and humans. Key words: ovulation detection, artificial insemination, great apes, Gorilla, Pan, rapid LH test, infertility OVULATION DETECTION Whenever new, and more complicated procedures are advocated for the husbandry and breeding of any captive animal population it is legitimate to question the benefits to be derived from such procedures. Careful application of a program for ovulation detection in the apes will provide information on the entire menstrual cycle and the incidence of pregnancy (which is still difficult to ascertain in the gorilla), and it will also provide indirect evidence for the clinical normality of the males involved. Baseline information on the hormonal changes occurring during the menstrual cycle is available for all great ape species, although there is less information than is desirable [Collinset al, 1975; Graham, 1970; Nadler et al, 1979; Reyes et al, 19751. The changes during the menstrual cycle are basically similar to those in the human, even with regard to the metabolism of sex steriods. Figure 1 summarizes the basic hormone patterns during the menstrual cycle for the various species. Salient features of the cycle include elevation of estrogen during the follicular phase, with a maximum value just prior to elevation of luteinizing hormone levels in the peak associated with induction of ovulation. Progesterone, which is elevated in the normal Received January 2, 1981; accepted March 21. 1981. Address reprint requests to Dr. K.G. Gould. Yerkes Regional Primate Research Center. Emory University, Atlanta, GA 30322. 0275-2565/82/00Sl-0015$03.500 1982 Alan R. Liss, Inc. 16 Gould Phase OV Phase Fig. 1. Idealized alternation in circulating hormones and basal body temperature during the menstrual cycle of the great apes and humans. postovulatory or luted phase, begins to rise just prior to the dramatic increase in luteinizing hormone, although the significance of this rise in triggering or assisting the LH surge is not clear. Table I lists certain physical characteristics and physiological markers that can be of use in monitoring the menstrual cycle. Menstruation Menstruation is a useful external marker in the chimpanzee and, to a certain extent, in the gorilla and orang-utan, although in the latter two species overt menses may not be evident and a test for blood in the urine must be used. One such test is "Hemastix" (upjohn Company), which is very sensitive and has been demonstrated to be useful in this application. The normal menstrual periods in the chimpanzee, orang-utan, and gorilla are approximately 31, 26, and 37-45 days, respectively. Perineal Swelling Alteration in the size of the perined sex skin is very readily observed in the chimpanzee, and when scored on a scale from 0 to 4 can be used to identify phases of the menstrual cycle, as it provides an indirect measure of circulating estrogen levels [Fig. 21. The absolute size of the perineal swelling varies between animals, and it is the relative change in size that is important for cycle monitoring. The size of the perineal swelling has been shown to be increased by estrogen and reduced by progesterone [Graham, 1972; 19731. The reduction of swelling size around midcycle is therefore associated with increasing progesterone levels of the luteal phase. Accordingly, the last day of maximum swelling has been associated with the time of ovulation. The practical use of this measure is discussed further later. In the gorilla, labial swelling is detectable and can be used for monitoring of the cycle. However, as demonstrated by Nadler et a1 119791, the swelling is restricted to the labia and can be monitored by measurement of the length of the vulval cleft and eversion of the vulval lining. Monitoring of this change is possible Ovulation and Insemination 17 TABLE I. Physical and Behavioral Signs for Cycle Monitoring Menstruation Perineal swelling Basal body temperature Cervical mucus changes Masturbatiodcopulation 4c - PAN 201 c .- - I W 3 v, O W .-c 4- 2- t a 4 cm 2 4 cm 2 mi 1,111I I I I I I m I I I I I I I I I I L I 111111 I 1 1 I I I I 1 -22 -18 -14 -10 -6 -2 0 2 6 10 DAYS Fig. 2. Diayamatic representation of changes in perineal swelling or labial size in chimpanzee and gorilla, respectively. only with cooperative animals. In the organ-utan there is no discernible perineal change associated with the menstrual cycle. Body Temperature Basal body temperature is widely used in the human as a means for monitoring menstrual cyclicity and timing ovulation. As a result of the thermogenic action of progesterone, the onset of the luted phase is associated with an increase in body temperature (Fig. 1).A similar shift in body temperature can be demonstrated in the chimpanzee. This is not, however, a practical method for cycle monitoring, as it requires provision of an indwelling temperature sensitive telemetry device [Graham et al, 19771 (Fig. 3). Cervical Mucus The viscosity of cervical mucus is much reduced around the time of ovulation, and air dried mucus demonstrates a characteristic ferningpattern [Papanicolaou, 1946; Singh & Mancusco, 19761. Ovulation time in the human has also been predicted on the basis of alteration in physical properties of vaginal and cervical mucus. Once again, however, it is not practical to obtain suitable samples for analysis from the great apes (Figs. 4 and 5 ) . 18 Gould Fig.3. Indwelling telemetry device for transmission of basal body temperature before shielding and encapsulation [Warner and Graham, 1975). Maria Iyo! % Water 9o I U Sexual 4Swelling 2- -16 -12 -8 -4 0 +4 +I0 DAY OF CYCLE Fig. 4. Alteration in cervical mucus composition during the chimpanzee menstrual cycle. Ovulation and Insemination 19 Fig. 5. Ultrastructure of cervical mucus as seen by scanning electron microscope. The size of the pores between fibers changes during the menstrual cycle and is maximal around the time of owlation. This specimen is from early follicular phase cervical mucus. Behavior Behavioral changes, including masturbation and, in the gorilla, copulatory activity, tend to be associated with the preovulatory estrogen rise in the midcycle, periovulatory period. An increase in sexual activity and aggressive behavior has been observed in many, but not all, female chimpanzees. Copulatory behavior tends to be restricted to the periovulatory period in the gorilla and, interestingly, in the orangutan under conditions in which access to the male is controlled by the female [Nadler et al, 1981 this issue]. Masturbatory activity in the female gorilla may precede the LH surge by 2 to 3 days [O’Donohue,personal communication, 19801. It is important to identify parameters of menstrual cyclicity that have predictive value with regard to timing of ovulation and insemination. The simplest method for the chimpanzee is monitoring of sexual swelling and menstruation over a number of cycles. However, the time of ovulation does not bear as predictable a relationship to the detumescence of the sex skin swelling for an individual animal as can be measured using the LH peak, - ie., within 1 2 to 24 hours [Gould & Faulkner, in press]. Furthermore, if advanced planning is required for anesthesia of the female for insemination, the normal one to three days variation in duration of maximum swelling for a given cycle further confounds efforts at prediction. This simple approach is of less ready application in the gorilla, in which species monitoring of menses, together with careful observation of behavior, is of most value. In the orang-utan, one is frequently reduced to monitoring of menses alone. The most precise indications of impending ovulation include the analysis of the hormone changes that most consistently precede ovulation or the observation of ovarian morphology. Again, for the species under consideration, for logistic reasons, direct 20 Gould ovarian observation by laparoscopy [Graham, 19761 or indirect observation by ultrasonography [Smith et al, 19801are only rarely possible. Similarly, predictive use of serum hormone levels is impractical because of the chemical restraint involved in recovery of serum from most individuals. This makes it necessary to use analysis of urinary hormones for ovulation prediction. The most frequently used assays for this purpose are of estrone glucuronide, the major conjugated form of estrogen in ape urine, and of luteinizing hormone. The application of estrogen and progestin assays to assist in breeding programs is described elsewhere in these proceedings [Lasley, 19811. Rapid LH Test We have developed a rapid hemagglutination-inhibition test, which was originally designed to detect pregnancy in nonhuman primates, so that it will provide a measurement of LH in the apes and human. This application of the rapid LH test has been evaluated in 26 chimpanzees, six gorillas, and four orangutans, and positive results have been obtained. Details of the verification procedures have been published elsewhere [Gould & Faulkner, 1979; 19811. The chimpanzee usually shows a positive LH test for two to three days each cycle. The period of positive response in the gorilla may occur subsequent to masturbatory behavior, which may be associated with elevated estrogen. The correlation of the graded response (0-4) of the rapid test with urinary LH levels is good for steps 0-2 (r = 0.92; P 5 0.001; n = 60), but for a reason that has not yet been determined, not so good for the stronger readings. This problem, however, is of academic interest insofar as detection of the time of the peak is concerned. ARTIFICIAL INSEMINATION Timing The rapid test has been used in timing of laparoscopy for experimental procedures, and for timing of artificial insemination. It is the latter application that is of greater interest to participants in this workshop. Pregnancy has been achieved in three chimpanzees and one gorilla with single or double inseminations by this method [Gould & Faulkner, in press; Martin et al, 19781.The success rate of artificial insemination in the great apes has not been as high as we would like, and after a brief review of the procedures used for artificial insemination, the reasons for this situation are considered. Semen Collection The procedures used for semen collection include automasturbation, artificial vagina, and rectal probe electroejaculation (RPE).Semen collected from different sources has some different characteristics (Table 11),but semen from both RPE and automasturbation has been shown to be fertile. Deposition The site of deposition of semen in artificial insemination is worthy of some consideration. Natural deposition in the apes places the coagulum high in the vagina, on the surface of the cervix. This, then, is the most direct route for artificial placement. In order to enhance the number of sperm present at the cervix it has been our practice to place the semen in two fractions. After collection, the semen is allowed to stand at 37 "C for up to one hour to permit liquefaction to occur. The liquid fraction is placed into the external cervical os, and the coagulated portion is added. In the gorilla, semen collected by RPE is often fluid, and the incubation step is not required. I t is frequently noted that acertain portion of the chimpanzee coagulum will never liquefy. Direct insemination into the uterus can be used when it appears advantageous to bypass the cervix for reasons of unsuitable cervical mucus etc. If intrauterine insemination is to be used, then consider- 8.14+.58(27) 5.93 h.42 (15) Masturbated samples Artificial vagina 1 7.72 & 1.09 Electroejaculated samples 2.10 12.42 0.80 8.50 3.00 8.60 0.25 12.54 00.25 12.54 Range *Artificial vagina 1 = 48 cm in length. 'Artificial vagina 2 = 5 cm in length. "+ SE (mean + standard error of the mean). 3.58 1.07 (9) Artificial vagina !P * 6.57*.43 (64) Total samples * SEa Fructose (mgiml) TABLE 11. Semen Characteristics Range 17.20h.25 (9) 17.29 *.24 (5) 16.26 h.26 (2) 17.49+.35(8) 16.06 18.24 16.56 17.84 16.00 16.52 16.32 19.52 17.24+.16(24) 16.00 19.52 *SE Acid phosphatase (unitsiml) 0.2 5.6 1.0 8.0 1.02 +.34(9) 0.1 3.0 2.28 +.52 (10) 0.5 5.0 2.40 +.55 (12) 3.78+.44(25) 0.1 8.0 Range 2.77+.28(56) *SE Volume (cc) 84.47 *8.46 (2) 58.74 *9.81(8) 70.35 k4.05 (2) - 64.96*7.12(12) SE % Live 76.00 92.93 0.00 95.65 65.7 75.00 0.00 95.65 Range 2.50~t.27(8) 3.25*.25(8) 3.17&.31(6) 3.00+.38(11) 1 3 2 4 2 4 1 4 1 4 Range 2.88&.18(34) +SE Motility (0-4) c h3 U 9E $ 1 Be 22 Gould ation must be given to the possibility of immunization to inseminated spermatozoa. This possibility can be minimized by using spermatozoa, that have been washed thoroughly by passage through a Ficoll gradient, as recommended by Harrison . Prior to insemination urine samples are collected at 24 hour intervals for LH detection. Artificial insemination is performed when the test is positive. Procedure The routine procedure for artificial insemination at the Yerkes Primate Center involves immobilization of the recipient female with ketamine HCI and collection of a venous blood sample for analysis of steroid and protein hormones as an adjunct to verification of the rapid LH tests used to time the procedure. The animal is placed in the prone position and the external genitalia are cleaned with soap, betadine, and water. A lubricated cylindrical plastic speculum approximately 8 inches long by 1%inches in diameter is placed in the vagina and advanced to the fornix, thus effectively centering and exposing the cervix. A headlamp is an invaluable adjunct to this procedure (Fig. 6). We routinely use a sodium bicarbonate douche at the cervix prior to placement of semen during insemination. This involves delivery and aspiration of approximately 100 ml of a 7% solution of NaHCO, in water directed from a syringe against the cervical 0s. This procedure has been adopted because of the demonstrated efficacy of such a douche in improving sperm penetration through the cervix in human females whose reduced fertility is associated with reduced penetration of sperm through the cervix [Ansari et al, 19801. As stated above, the liquid portion of the semen sample is then placed via a 1 ml pipette into the external cervical 0s and the coagulum placed deep in the fornix with a plunger that fits through the center of the speculum. The animal remains recumbent during the initial recovery from anesthesia, which is a period of approximately 20-30 minutes. Potential modifications of this procedure for artificial insemination include clomid therapy to induce ovulation at a predictable time Fig. 6 . Artificial insemination of the chimpanzee. Note the use of a headlamp. Liquefied semen is being placed on the cervical 0s with a 1 ml pipette. Ovulation and Insemination 23 in the cycle. Such therapy for apes could consist of 50 mg of clomid on days 7 through 12, after menstruation, with administration of 3000 IU HCG intramuscularly 24 hours after the last administration of clomid, and artificial insemination 20 hours after the injection of hCG. The procedure can be modified by placement of washed sperm directly into the uterine cavity. Other routes for insemination, such as the intraperitoneal route, have been recorded in the literature but, on the basis of very limited studies, have not been shown to offer any advantage over vaginal deposit in the great apes. [Gould, unpublished data]. Complicating Factors Table I11 lists four areas of concern with regard to the less-than-optimum success rate of artificial insemination thus far obtained. The great apes uniformly require chemical restraint before artificial insemination can be performed. I t has been our experience that, although chimpanzee females can be trained to tolerate insertion of speculi and probes into the vagina around the time of ovulation, and to tolerate administration of innocuous douches while conscious, administration of semen or seminal plasma is accompanied by a violent aversion response on the part of the animal, necessitating restraint for adequate placement of the specimen in this species. Anesthesia. The effect of anesthesia per se upon ovulation and establishment of a competent luted phase has yet to be determined precisely. There is some evidence [Soules et al, 1980;Lasley, 1981 this issue] of interference with the normal hormone patterns of the periovulatory period by anesthetic agents. Against this, however, is experimental evidence documenting the occurrence of ovulation during cycles in which other primate species (macaques)were frequently anesthetized for laparoscopy to determine ovarian and follicular morphology throughout the menstrual cycle. Furthermore, luteinization occurs after laparoscopy and ovariectomy, as evidenced by occurrence of menses at a predicted time after ovulationlovectomy. The possible involvement of anesthetic agents with subsequent fertility must, however, be further investigated. Frequency. Arising from the necessity for anesthesia for artificial insemination is a constraint upon the frequency of insemination that is logistically practical within a single cycle. It is normal in the human for insemination to be performed on three or four occasions in each cycle, and even with this regimen pregnancy may not be achieved in 3 to 6 insemination cycles. Because it is necessary to restrict food and fluid intake prior to anesthesia, it is rarely possible to inseminate a female ape on more than two occasions in any one cycle, which emphasizes the necessity for precise timing of the insemination if pregnancy is to ensue. Semen quality. Semen quality is another factor that is difficult to control when artificial insemination is employed for propagation of endangered species. It has been well documented in nonprimates, particularly cattle and rabbits, that semen from different males is of different quality when it is tested by means of competitive fertilization with semen from other males or by fertility subsequent to frozen storage. Rigorous selection is practiced to ensure that only semen of the highest quality is used for donor insemina- TABLE 111. Artificial Insemination -The Problems Anesthesia Frequency Semen quality Inherent fertility 24 Gould tion in the human and for storage and dissemination in the bovine. When referring to atificial insemination, particularly in the gorilla, we do not enjoy the luxury of a large male population from which to select the most appropriate and most fertile donors. It is therefore possible that we are using less-than-optimally fertile semen samples for insemination in gorillas, and possibly in chimpanzees as well. Inherent infertility. The last complicating factor for successful artificial insemination that will be mentioned here is that of the inherent fertility of great apes and humans. Evidence has been presented that the rate of attrition among early developmental stage embryos in the human can approach 60% [Leridon, 19731. Such embryonic loss within the first week of development will remain undetected by routine observation of such parameters as incidence of menses subsequent to insemination. Early embryonic death, if it occurs to a significant extent in the great apes, would result in a reduced incidence of established pregnancy subsequent to artificial insemination. If early embryonic death is a reflection of a metastable microenvironment around the egg, which is especially susceptible to disruption by such insults as anesthesia, it could be that artificial insemination is adversely affected by anesthesia, as in the first problem mentioned above. CONCLUSION I t is fair to state that relatively simple and effective methods are available for monitoring the regularity and incidence of menstrual cyclicity in the great ape species. Use of these methods will provide potential improvement in conception rates from both natural and artificial breeding. While artificial insemination is not to be advocated as a panacea or as a replacement for natural breeding programs, it is evident that there will be significant advantages to the use of artificial insemination in situations where the male of an established pair is known to be infertile or where behavioral characteristics of the “breeding pair” preclude natural copulatory behavior, and such alternatives as breeding loans, with their attendent expense, loss of time, and quarantine period, are considered impractical. Consideration should also be given to use of artificial insemination with frozenhhawed semen as a method for introducing new genetic material from the wild population or across national boundaries. REFERENCES Ansari, A.H.; Gould, K.G.; Ansari, V.M. Sodium bicarbonate douching for improvement of the postcoital test. FERTILITY AND STERILITY 33: 608-612,1980. Collins, D.C.; Graham, C.E.; Preedy, J.R.K. Identification and measurement of urinary estrone, estradiol-178, estriol, pregnanediol and androsterone during the menstrual cycle of the orangutan. ENDOCRINOLOGY 96: 93-101, 1975. Gould, K.G.; Faulkner, J.R. Rapid detection of the midcycle LH peak in the chimpanzee. Paper presented at the SEVENTH CONGRESS OF THE INTERNATIONAL PRIMATOLOGICAL SOCIETY, Bangalore, India, 1979. Gould, K.G.; Faulkner, J.R. Development, validation and application of a rapid method for detection of ovulation in great apes and the human. FERTILITY AND STERILITY 35: 676-682.1981, Graham, C.E. Reproductive physiology of the chimpanzee. Pp. 183-220 in THE CHIMPANZEE, vol. 2. G.H. Bourne, ed. Basel, Karger. 1970. Graham, C.E. Menstrual cycle plasma proges- terone levels in the chimpanzee: Relation to endometrial histology and sexual swelling. BIOLOGY OF REPRODUCTION 7: 122, 1972. Graham, C.E. Chimpanzee endometrium and sexual swelling during menstrual cycle or hormone administration. FOLI A PRIMATOLOGIA 19: 458-468,1973. Graham, C.E. Technique of laparoscopy in the chimpanzee. JOURNAL OF MEDICAL PRIMATOLOGY 5: 111-123,1976. Graham, C.E.; Warner, H.; Misener, J.;Collins, D.C.; Preedy,J.R.K. The associationbetween basal body temperature, sexual swelling and urinary gonadal hormone levels in the menstrual cycle of the chimpanzee. JOURNAL OF REPRODUCTION AND FERTILITY 50: 23-28.1977. Harrison, R.A.P. A highly efficient method for washing mammalian spermatozoa. JOURNAL OF REPRODUCTION AND FERTILITY 48: 347-353,1976. Leridon, H. Pp. 13-27 in LES ACCIDENTS CHROMOSOMIQUES DE LA REPRODUCTION. Paris, Centre International de l’Enfance, 1973. Ovulation and Insemination Martin, D.E.; Graham, C.E.; Gould, K.G. Successful artificialinsemination in the chimpanzee. SYMPOSIUM OF THE ZOOLOGICAL SOCIETY OF LONDON no. 43: 249-260, 1978. Nadler, R.D.; Graham, C.E.; Collins, D.C.; Gould, K.G. Plasma gonadotropins, prolactin, gonadal steorids, and genital swelling during the menstrual cycle of lowland gorillas. ENDOCRINOLOGY 105: 290-296, 1979. Papanicolaou, B.N. A general survey of the vaginal smear and its use in research and diagnosis. AMERICAN JOURNAL OF OBSTETRICS AND GYNECOTAOGY 51: 316-328, 1946. Reyes, F.I.; Winter, J.S.D.; Faiman, C.; Hosbson, W.C. Serial serum levels of gonadotropins, prolactin and sex steroids in the nonpregnant and pregnant chimpanzee. ENDOCRINOLOGY 96: 1447,1975. 25 Singh, E.H.; Mancusco, S. Constituents of fern formation in human cervical mucus. MEDIKON 5: 19,1976. Smith, D.H.; Picker, R.H.; Sinosich, M.; Saunders, D.M. Assessment of ovulation by ultrasound and estradiol levels during spontaneous and induced cycles. FERTILITY AND STERILITY 33: 387-390, 1980. Soules, M.R.; Sutton, G.P.; Hammond, C.B.; Haney, A.F. Endocrine changes a t operation under general anesthesia: Reproductive hormone fluctuations in young women. FERTILITY AND STERILITY 33: 364-371, 1980. Warner, H.; Graham, C.E. Telemetry system for continuous recording of body temperature in large, unrestrained mammals. IRCS MEDICAL SCIENCE (Research on: Biomedical Techniques; Biophysics, Membrane and Cell Biology; Reproduction, ObstetricsiGynecology) 3: 214, 1975.