Animal Reproduction

A female rhesus monkey whose hypothalamus can no longer drive her pituitary is given the missing hormone, GnRH, through a drip. Given steadily, it fails. Given as a short burst once an hour, it works. Why would the timing of a signal matter more than its amount?

The answer follows a look at how animals make offspring.

Two ways to make offspring

Asexual reproduction makes offspring from one parent without fertilization, so the young carry that parent’s genes. In budding a new individual grows out of the parent, as in Hydra; in fission the parent splits in two, as in some sea anemones and flatworms; in fragmentation a piece breaks off and regrows a whole body, as in some sea stars. In parthenogenesis an egg develops without being fertilized.

Sexual reproduction joins two gametes, haploid sex cells (cells with one set of chromosomes), from two parents in fertilization, giving a diploid zygote with two sets. Meiosis, the cell division that halves the chromosome number and reshuffles the parents’ alleles, is taught in full in chromosomes and meiosis.

Asexual reproduction needs no mate. Sex carries what John Maynard Smith called the twofold cost of males. Suppose, in a made-up population, every female leaves two surviving young. An asexual female’s two are both daughters, and each will bear young; a sexual female’s are on average one daughter and one son. All else equal, the asexual line doubles each generation while the sexual line stays the same size. Sex also means finding a mate; what it offers is new combinations of alleles. Why sex persists despite its costs is still argued; see life histories and selection in ecological context.

Some animals use both: water fleas (Daphnia) and many aphids produce generations of females by parthenogenesis, then males and sexual eggs, often as the season turns. Hermaphrodites make both eggs and sperm. Earthworms make both at the same time, and two worms usually exchange sperm. Some fish change sex: clownfish start as males and the largest can become female, while many wrasses change from female to male. How sex is determined is in sex linkage and beyond.

Eggs in water, eggs in bodies

External fertilization happens outside the female’s body, usually in water. Sea urchins broadcast-spawn, releasing eggs and sperm into the sea, where many gametes never meet and eggs and larvae are exposed to predators. In most frogs, the male clasps the female (amplexus) and releases sperm over her eggs as she lays them.

Internal fertilization happens inside the female’s body. Sperm arrive close to the eggs, and on land the fertilized egg is protected from drying. But it says where sperm meet egg, not how the young are born. Birds fertilize internally and then lay shelled eggs. Most mammals are viviparous (the young develop inside the mother and are born live), yet monotremes such as the platypus lay eggs, and some sharks and lizards bear live young.

Making sperm and eggs

Spermatogenesis happens in the seminiferous tubules, coiled tubes inside each testis. At the tubule wall, spermatogonia (diploid stem cells, 46 chromosomes in humans) divide by mitosis, renewing themselves and supplying cells for meiosis. A primary spermatocyte (46 chromosomes) completes meiosis I to give two secondary spermatocytes (23 chromosomes each), and meiosis II gives four spermatids. Spermiogenesis remodels each spermatid into a sperm with a head and a tail. Sertoli cells lining the tubule support the developing cells; Leydig cells, between the tubules, make testosterone. Sperm mature in the epididymis, a coiled duct on the testis, and pass on through the vas deferens; the seminal vesicles and prostate add fluid to make semen. Heller and Clermont estimated the whole process at about 64 days in men; later estimates differ.

Oogenesis in human females starts before birth. In the fetal ovary, oogonia divide by mitosis and become primary oocytes, which enter meiosis I and arrest in prophase I. Each waits, for years or decades, in a follicle (the oocyte plus supporting cells). After puberty, usually one follicle per cycle matures, and its oocyte finishes meiosis I shortly before ovulation. The division is unequal: nearly all the cytoplasm goes to the secondary oocyte, and the other chromosome set leaves in a small polar body. The secondary oocyte is released arrested in metaphase II and finishes meiosis II only if fertilized, giving the egg and a second polar body. Adult human ovaries are not known to make new oocytes; a 2020 single-cell study found no support for germline stem cells there, though negative evidence cannot prove absence.

Worked problem. The inputs are invented for practice. (a) Suppose 120 human primary spermatocytes complete both meiotic divisions. Each gives 4 spermatids: 120 x 4 = 480 spermatids. (b) Suppose a hypothetical 75 percent survive maturation into sperm: 480 x 0.75 = 360 sperm, so 480 - 360 = 120 are lost. Check: 360 + 120 = 480. This counts cells in the testis, not a semen sample. (c) Now suppose 120 human primary oocytes completed meiosis. Each gives one egg, so 120 eggs, each with two or three polar bodies (the first polar body may or may not divide again). This is idealized: in humans meiosis II finishes only on fertilization. (d) Check the chromosome numbers: every primary spermatocyte and primary oocyte carries 46 chromosomes, and every spermatid and egg 23. The halving is the same in both sexes; the yield per meiosis is four against one.

Process

spermatogenesis oogenesis

First polar body divides again (oogenesis only)

yes no

Primary cells in the batch 120 Hypothetical survival, spermatid to sperm 75% (spermatogenesis only)

In this sketch: invented inputs; adult human; schematic cells, not to scale. The sperm count is of cells in the testis, not a semen sample. The oocyte count is idealized, because in humans meiosis II finishes only on fertilization.

Rule to keep: the halving from 46 to 23 is the same in both sexes; the yield per meiosis is four against one.

Gametes with a chromosome too many or too few are covered in chromosome changes.

The axis and its pulses: Knobil’s monkeys

The gonads are run by the hypothalamic-pituitary-gonadal axis. Neurons in the hypothalamus release gonadotropin-releasing hormone (GnRH) into portal vessels that carry it to the anterior pituitary (the route is in hypothalamus and autonomic control). GnRH triggers release of two gonadotropins, pituitary hormones that act on the gonads: luteinizing hormone (LH) and follicle-stimulating hormone (FSH). Their receptors on gonad cells belong to the family covered in G-protein-coupled receptors.

In human males, LH acts on Leydig cells, which make testosterone. FSH acts on Sertoli cells, which, with testosterone, support spermatogenesis. Two brakes close the loop. Testosterone acts back on the hypothalamus and pituitary and lowers gonadotropin release, and Sertoli cells release inhibin B, which lowers FSH. That is negative feedback, in which the product of a pathway turns the pathway down, as in the endocrine system.

GnRH is released in pulses. In 1978 Belchetz, Plant, Nakai, Keogh and Knobil asked whether the pulses matter, using female rhesus monkeys in which lesions of the hypothalamus had abolished LH and FSH secretion. The pituitary was intact, so they could replace GnRH by intravenous infusion, either continuously or as brief infusions once an hour, and measure LH and FSH. If the pituitary read only the total amount, the steady drip should have worked at least as well. It did not: hourly pulses restored LH and FSH secretion, and continuous infusion failed to sustain it. The authors reported that switching back to pulses restored secretion, and concluded that the pattern of GnRH, not its total amount, controls the pituitary.

The experiment shows pattern-reading by the pituitary in this primate, not where the pulses come from or what interval is normal in humans. In mice, Clarkson and colleagues (2017) recorded kisspeptin neurons in the arcuate nucleus of the hypothalamus; brief light-driven activation of those neurons evoked LH pulses, and silencing them suppressed the pulses. Whether they do so in primates is a separate question. The lesson: continuous GnRH is not “more GnRH”; held steady, it turns the pituitary down.

The cycle and pregnancy

In human females the axis runs two linked cycles: the ovarian cycle (the follicles) and the uterine cycle (the uterine lining, or endometrium).

In the follicular phase, FSH recruits a group of follicles and usually one becomes dominant. LH drives theca cells to make androgens, and FSH-supported granulosa cells convert them to estradiol. Rising estradiol rebuilds the endometrium after menstruation (the proliferative phase) and, for most of this phase, brakes the axis. But when estradiol stays high late in the phase, its effect switches to positive feedback, in which a signal boosts its own trigger: LH release rises steeply in the LH surge, which triggers ovulation, the release of the secondary oocyte.

After ovulation, in the luteal phase, the emptied follicle becomes the corpus luteum, secreting progesterone and estradiol. Progesterone converts the thickened endometrium into a lining that can receive an embryo, the secretory phase, and it brakes LH and FSH again. Estradiol and progesterone act mainly through receptors inside the cell, covered in nuclear receptors. If no embryo implants, the corpus luteum regresses, progesterone falls, and the lining breaks down in menstruation, which begins the next cycle.

The textbook 28-day cycle with day-14 ovulation is a drawing convention. Analysing 612,613 cycles from 124,648 users of a cycle-tracking app, Bull and colleagues found a mean cycle of 29.3 days, a mean follicular phase of 16.9 days and a mean luteal phase of 12.4 days, and only 13 percent of cycles were 28 days long. The sample is self-selected.

If an embryo implants, its outer cell layer (the trophoblast, which forms much of the placenta) secretes human chorionic gonadotropin (hCG). Acting through the LH receptor, hCG keeps the corpus luteum making progesterone, so the lining is not shed. From about weeks 7 to 9 of gestation, the placenta itself becomes the main source of progesterone. Placental exchange of oxygen and antibodies is in oxygen transport and adaptive immunity.

Most mammals have estrous cycles instead: the endometrium is generally reabsorbed rather than shed, and females are usually receptive mainly around ovulation. Menstruation is known in humans, apes and Old World monkeys, and a few other mammals, such as some bats and the spiny mouse. The embryo’s next steps are in animal development, and the plant counterpart is plant reproduction.

Design problem

Lutes and colleagues (Nature, 2010) measured chromosome volume in DAPI-stained oocyte nuclei: oocyte nuclei of the parthenogenetic Aspidoscelis tesselata had chromosomes about 2.2 times the volume of those of the sexual A. gularis, which the authors read as a doubling of DNA. In body cells (nucleated red blood cells), DNA content differed by less than 1 percent. In another parthenogen, A. neomexicana, chromosome-specific FISH probes showed that the partners paired in meiosis were sister chromosomes (copies of the same chromosome), not the two different homologues a sexual animal pairs. The oocytes also appeared to show the structures of meiotic pairing and crossing over. How can these all-female lizards keep meiosis and still pass on the mother’s genetic makeup, and what does each measurement rule out?

One way to do it

A doubled chromosomal DNA content suggests one extra round of DNA replication before meiosis, so the oocyte enters meiosis with a doubled chromosome set. The body-cell measurement rules out a simply bigger genome: body cells differ by under 1 percent, so the doubling belongs to the germ line.

In a sexual animal the maternal and paternal versions of each chromosome pair, and crossing over mixes them. If sister copies pair instead, crossing over swaps segments between identical copies and changes nothing, and the two divisions bring the doubled set back to the ordinary diploid number. The egg keeps both of the mother’s different versions of each chromosome (her heterozygosity) while meiosis and crossing over still happen.

Limits: the evidence is from these Aspidoscelis species, sister pairing was shown only for chromosomes the probes could tell apart, and other parthenogens may differ.