Physiology of the female reproductive cycle

As in other mammals, the reproductive tract in female elephants is composed of a pair of ovaries (located near the kidneys), the Fallopian tubes, a uterus with two cornua or horns, and a chamber that leads into the urogenital passage that opens externally at the vulva. A striking feature of the reproductive system in elephants is the long (nearly 1 m) urogenital canal, a common passage for both the genital and urinary tracts that opens at a position anterior to the hind legs. The clitoris is also unusually pronounced in cows; this has often been a source of incorrect sexing of animals in the field, of young captives, or even of museum specimens.

Typically, the estrous cycle in a mammal involves a follicular phase with growth of the vesicular ovarian (or so-called Graafian) follicles containing the ovum (egg); ovulation, which is release of the egg into the fallopian tubes; the luteal phase, in which the corpus luteum is formed inside the ruptured follicle; and finally, if the egg is not fertilized, the degeneration of the corpus luteum to start a new follicular phase. Several hormones mediate this process; the principal hormones are the two gonadotropins, follicle-stimulating hormone (FSH) and luteinizing hormone (LH), secreted by the pituitary gland in the brain, and the two female sex hormones, estrogen and progesterone, produced by the ovaries. The levels of FSH and LH are relatively low during most of the follicular phase, but rise abruptly when the rate of estrogen secretion by the growing follicle increases. The rise in LH is quantitatively higher than that of FSH and induces maturation of the follicle, with ovulation occurring soon after the LH surge. Following ovulation, the corpus luteum, a glandular tissue that forms inside the ruptured follicle, now begins to produce estrogen and another steroid hormone, progesterone. Monitoring the levels of these hormones in the bloodstream or their metabolites in the urine of an animal helps to reconstruct the estrous cycle of the species.

The estrous cycle in elephants is still rather poorly understood. Overt signs of estrus in female elephants are lacking. Some early behavioral observations during 1969-1970 of captive cow elephants and attendant bulls in Sri Lanka (Ceylon) indicated an estrous cycle of 18-27 days, with an average of 22 days. Bulls showed distinct sexual interest in the cows at these intervals, and the animals mated more often during this time. Data on vaginal cytology and urinary estrogens from elephants in western zoos also pointed to an approximately 3-week estrous cycle.

Until this time, it was difficult to detect progesterone in elephants because the levels of this hormone were probably very low. The development of sensitive radioimmunoassay techniques, however, changed this situation. In 1975, Edward Plotka and associates reported the first reliable detection of progesterone and estrogen in the blood of captive Asian and African elephants in various reproductive states. No clear relationship between these hormone levels and the reproductive state of the animal could be discerned in these early experiments. A major shift in our knowledge of the estrous cycle in elephants occurred when Michael Schmidt and colleagues, working at the Washington Park Zoo (Portland, Oregon) with Asian elephants, announced at a meeting in 1981 that the duration of the cycle was about 16 weeks; this was based on clearer progesterone profiles in several cows (fig. 3.1).

Puberty can be considered as the onset of the first estrous cycle, as evidenced by the development of a large follicle, while sexual maturity is the age at first ovulation seen from the presence of at least one corpus luteum. By the above criteria, there may be a difference of 2-4 years between puberty and sexual maturity in female elephants. The age at sexual maturity itself varies quite widely in populations of African and Asian elephants. This seems related to local climate, forage resources, nutritional plane, and other ecological factors (chapter 7). In the wild, female elephants typically attain sexual maturity between 11 and 14 years of age, although the mean age of maturity in populations ranges from 9 years to as high as 22 years. Reports of 6-year-old elephants attaining sexual maturity under captive conditions are exceptional. Some define sexual maturity as the age at first conception, but this may not occur during the first ovulation.

The follicular phase in elephants is also referred to as the nonluteal or interluteal phase. The early histological studies of ovaries obtained from elephants shot during control operations in Africa are still the main sources of information of follicular and luteal structures. These descriptions have been provided mainly by J. S. Perry, Roger Short, John Hanks, Irvin Buss, Norman Smith, and Richard Laws. The ovaries of a mature elephant, pregnant or otherwise, contain multiple follicles in different stages of development. The sizes of follicles reach 30 mm in diameter, with pubertal animals having at least one follicle greater than 6 mm. The dominant follicles in mature animals generally vary from 10 to 20 mm.

Changes in steroid and peptide hormone profiles and their relationship to follicular development are still rather unclear. Some of the early measurements of FSH in the blood of African elephants showed no clear patterns or apparent relationship to age. The more detailed work of Janine Brown and associates in captive Asian elephants reported in 1991 provides a much clearer picture of FSH profiles. A cyclic pattern of FSH lasting 12-14 weeks was seen, with low levels (below 10 ng/ml) during the late follicular and early luteal phases, followed by elevated levels (typically 25-40 ng/ml) over a long period (7-8 weeks) during the late luteal and early follicular phases. FSH levels were also inversely related to those of another peptide hormone, inhibin (as expected,

Figure 3.1

Weekly mean (±standard error) progesterone and estradiol concentrations in blood of six Asian elephant cows for 15 estrous cycles. The average frequency of positive urine tests by bulls during the same cycle is also shown. Week 0 is designated as the week preceding one in which progesterone exceeded 100 pg/ml. (From Hess et al. 1983. Reproduced with the permission of the Society for the Study of Reproduction, Inc., U.S.A.)


Figure 3.1

Weekly mean (±standard error) progesterone and estradiol concentrations in blood of six Asian elephant cows for 15 estrous cycles. The average frequency of positive urine tests by bulls during the same cycle is also shown. Week 0 is designated as the week preceding one in which progesterone exceeded 100 pg/ml. (From Hess et al. 1983. Reproduced with the permission of the Society for the Study of Reproduction, Inc., U.S.A.)

because inhibin suppresses FSH), with a similar 12-14 week cycle. The inter-luteal or follicular period itself is of a short 4-6-week duration.

After it became apparent from the work of David Hess and the Schmidts that the estrous cycle in Asian elephants is likely to be much longer than earlier suspected, further work on hormone profiles of both species by Edward Plotka and associates confirmed that the cycle was indeed 14-16 weeks. They proposed that elevated estrogen levels, although often obscure, reflect possibly five successive waves of follicular growth with a periodicity of about 3 weeks throughout the cycle, and that the clear progestin cycle of 14-16 weeks represents the culmination of one of these waves in ovulation and formation of a functional corpus luteum. They could only occasionally detect multiple peaks of LH within a progestin cycle. Later work, however, clearly indicated more than one LH surge during a cycle, but this did not occur throughout the cycle. LH profiles of African elephants obtained by N. Kapustin and coworkers showed that each ovarian cycle was characterized by two LH surges, separated by 3 weeks, within the follicular phase, and that the second peak was associated with an increase in progestin.

The dynamics of the primary class of female sex hormones, the estrogens, during the estrous cycle are confusing. The concentrations of circulating estra-diol (the predominant estrogen in most mammals) are very low (usually under 15 pg/ml) and seemingly fluctuate without any pattern. Only the study by N. Kapustin's team indicates small but significant estradiol peaks prior to each of the two LH peaks. Estrogen profiles presented by K. Taya and associates for Asian elephants in the Tokyo Zoo have also been interpreted by Keith Hodges to contain multiple peaks within the interluteal phase. It is possible that measurements of conjugated estradiol (e.g., to protein) may yield more meaningful results than profiles of the free hormone.

The present evidence, therefore, strongly suggests that the timing of ovulation is closely linked to the increase of progestin and to the second of the twin LH surges toward the end of the interluteal phase. Other possibilities cannot be ruled out in view of the multiple LH peaks and possible follicular waves.

There are also conflicting views as to whether the elephant releases one egg (monovular) or several eggs (polyovular) during a cycle. Roger Short followed an African cow in estrus; the elephant was then shot, and the ovaries were collected. One of the ovaries contained "one fresh haemorrhagic ovulation point, 9 mm diameter," which was taken to be evidence for a single follicular rupture. The low frequency (about 1%) of twinning recorded in both African and Asian elephant populations also seems to argue for monovulation. On the other hand, the large numbers of corpora lutea (CL) present in the ovaries of both pregnant and nonpregnant elephants suggest that the animal is polyovula-tory. After examining the corpora lutea in ovaries of several hundred animals culled in Zambia, John Hanks and Roger Short concluded that the elephant could be either monovular or polyovular. Of course, it is entirely possible that an animal producing only one offspring at a time may be releasing several eggs during a cycle, but only one is fertilized.

If an elephant does not conceive, the luteal phase lasts for about 8-11 weeks, although a shortened phase of only 2-3 weeks has also been noticed. The multiple, large CL in the ovary are a feature the elephant shares with another large mammal, the whale. The CL range in size from 2 to 35 mm in diameter. Both small and large CL bear stigmata of ovulation; thus, size cannot be used to distinguish between primary and accessory lutea. The early biochemical studies that accompanied the histological examinations of fresh ovaries either failed to detect progesterone or recorded extremely low levels of this hormone in the luteal tissue. Roger Short and John Hanks thus advanced the hypothesis that the accumulation of CL from one cycle to another was necessary for production of sufficient quantities of progesterone to sustain pregnancy. So far, there is no evidence for any progressive increase in progesterone concentrations with successive nonfertile cycles. It is not clear why multiple CL should persist structurally, but not functionally, from one cycle to another.

Recent work by Keith Hodges and associates has shown that it is not progesterone, but two other forms of progestin, 5a-pregnane-3,20-dione (5a-DHP) and 3a-hydroxy-5a-pregnan-20-one (5a-P-3a-OH), which are the principal progestins synthesized by CL of African elephants (culled at Kruger, South Africa). In both the elephant species, 5a-DHP is also the predominant progestin in blood, with its concentration 10-20-fold that of progesterone in the African species. The affinity of this 5a-reduced steroid to an endometrial receptor also confirms its biological activity.

Attention therefore now is shifting from progesterone to obtaining profiles of these 5a-reduced progestins during the ovarian cycle of elephants. Already, one investigation by Hodges's team led by Michael Heistermann has shown that 5a-DHP profiles in the blood of captive African elephants are more pronounced than those of progesterone during the luteal phase (table 3.1). This could provide a much clearer indication of the time of ovulation. Similarly, 5a-P-3-OH showed a clear cyclic pattern in urine, thereby providing a means to monitor luteal function noninvasively. In Asian elephants, the profiles of the related 17a-hydroxyprogesterone (17a-OHP) in blood and of pregnanetriol in

Table 3.1

Concentration of progestins in the blood of captive African elephants during the estrous cycle.

Table 3.1

Concentration of progestins in the blood of captive African elephants during the estrous cycle.

5a-DHP (ng/

Ovarian Phase

Progesterone (ng/ml)


Follicular phase

0.14 ± 0.01

1.43 ± 0.11

Midluteal phase

0.71 ± 0.06

13.6 ± 1.09

Luteal/follicular levels



Source: Based on Heistermann et al. (1997). 5a-DHP, 5a-pregnane-3, 20-dione.

Source: Based on Heistermann et al. (1997). 5a-DHP, 5a-pregnane-3, 20-dione.

urine were found by Cheryl Niemuller, also working with Hodges, to be useful in monitoring the ovarian cycle.

Prolactin, a peptide hormone normally elevated during pregnancy, is also known to play a regulatory role in the ovarian cycle of mammals, especially in relation to environmental factors such as photoperiod. One investigation by Ursula Bechert and colleagues on nonpregnant African elephants showed elevated levels of prolactin during the follicular phase and reduced levels during the luteal phase, the inverse of the progestin pattern. This suggested that pro-lactin could play a role in regulating ovarian function in the elephant.

Based on the information available until about 1998, a model of the ovarian cycle in elephants was proposed by Keith Hodges. In 2000, Janine Brown updated this model to arrive at the following picture of the estrous cycle (fig. 3.2). During the luteal phase, the elevated levels of progestins inhibit follicular development and the release of LH. The subsequent decrease in progestin restores follicular activity. The elevated levels of FSH at the beginning of the non-luteal phase initiate two successive waves of follicular development, each of which culminates in a distinct LH peak. The follicles that develop during the first wave do not result in ovulation, but regress and form CL, which secrete hormones later in the cycle. Over the next 3 weeks, another wave of follicular development results in the formation of one large follicle ("Graafian" follicle) that releases an egg about 24 hours after the second LH surge. The elevated estrogens, which have low concentrations that are difficult to measure, may trigger the LH surges during each follicular wave. The Graafian follicle may also secrete small amounts of progestins in the period leading up to ovulation. After ovulation, the progestin levels rise along with the maturation of CL, followed by a gradual rise in FSH that peaks at the end of the luteal phase.

A key question regarding the elephant's estrous cycle is the significance of the two or more follicular waves during the interluteal phase. A functional explanation may be that the first, nonovulatory, follicular wave may be needed to form accessory CL for producing the small preovulatory progestin rise required for ovulation. An evolutionary explanation could be that this serves as an early signal to a bull of impending ovulation in a particular cow. In regions of low elephant density, it may be advantageous for a bull to be associated with one family group for an extended period and for a cow to maintain a bull's interest over this period to derive reproductive benefits. With a short duration of 2-4 days during an estrous cycle in which a cow elephant can conceive, a missed mating opportunity would mean a delay of 15 weeks before another possibility of conception arises. Perhaps Richard Barnes's model of bull association with an elephant herd (chapter 4) should be revisited.

Should an elephant conceive, the resulting pregnancy proceeds through a gestation of 20-22 months. The persistence of multiple CL through pregnancy has been emphasized by practically all the histological investigations of culled elephants. Some studies did not detect any corpora albicantia in the ovaries of elephants, even during the second half of pregnancy, implying that no regression of CL had occurred. More recently, D. J. de Villiers and coworkers at

Figure 3.2

Model for the elephant estrous cycle. Smoothed profiles of luteinizing hormone (LH), progestins, follicular-stimulating hormone (FSH), and inhibin in blood serum are shown based on various studies of Asian and African elephants. Time 0 corresponds to the presumed time of ovulation. (Based on Brown 2000. Reproduced from Zoo Biology with the permission of John Wiley and Sons.)

Figure 3.2

Model for the elephant estrous cycle. Smoothed profiles of luteinizing hormone (LH), progestins, follicular-stimulating hormone (FSH), and inhibin in blood serum are shown based on various studies of Asian and African elephants. Time 0 corresponds to the presumed time of ovulation. (Based on Brown 2000. Reproduced from Zoo Biology with the permission of John Wiley and Sons.)

Kruger found that the volume of CL, but not the number, decreases, indicating a slow regression through gestation.

Hormonal regulation of pregnancy in elephants is even less understood than hormonal regulation of the ovarian cycle. The CL appear to be most active hormonally between 3 and 15 months of gestation. No consistent changes in the levels of blood progesterone have been recorded, although one study on the Asian elephant by Cheryl Niemuller recorded up to a twofold rise during pregnancy. She also demonstrated that changes in the ratio of blood progesterone to 17a-OHP from as early as the second or third week of conception could be a valuable tool in early diagnosis of pregnancy.

What supports CL function through pregnancy is not clear. No consistent patterns are seen in gonadotropins, a possible contender for this role. A clear surge is seen, however, of prolactin, which is known to support luteal function in other mammals. First investigated by A. S. McNeilly and team, the levels of prolactin rise substantially (20- to 100-fold) in the blood of both elephant species, although the timing of its initial increase at 4-6 months into pregnancy is considered too late to support CL activity beyond the normal luteal phase duration (which should occur at 2-3 months). The origin of this prolac-tin is not known; it is unlikely to be entirely from the pituitary gland. The placenta itself may play a role in progestin production, although the evidence is lacking.

As during the ovarian cycle, the levels of free estrogens remain low with no apparent pattern. Conjugated estrogens, however, increase in both blood and urine during the second half of gestation. There is an interesting suggestion that the ovaries of the developing fetus may be hormonally active. Limited data show an enlargement of the ovaries in the fetus to adult sizes during late gestation, followed by regression before birth. During the week preceding parturition, there is the anticipated decrease in progesterone levels to baseline values.

Was this article helpful?

0 0

Post a comment