Lactation

The mammary gland is a modified apocrine/exocrine sweat gland, unique to mammals, built from branching ducts that terminate in grape-like clusters of secretory sacs called alveoli. Each alveolus is lined by a single layer of milk-secreting alveolar (luminal) epithelial cells and wrapped in a basket of contractile myoepithelial cells, which sit between the epithelium and the basement membrane. Milk made by the epithelial cells accumulates in the alveolar lumen; it only leaves the gland when the myoepithelial cells contract, squeezing it into the duct system and out through the nipple. This anatomical division — a secretory cell type and a separate contractile cell type — is the physical basis for the two-hormone control system described below. Two processes must therefore be distinguished sharply, because students often blur them. Lactogenesis (milk synthesis) is the continuous biochemical manufacture of milk components — lactose, milk proteins, and lipid droplets — inside alveolar cells, driven principally by the hormone prolactin. Milk ejection (the let-down reflex) is the acute, mechanical expulsion of already-synthesized milk from the alveolar lumen into the ducts and out of the nipple, driven by oxytocin acting on myoepithelial cells. Prolactin makes milk over hours; oxytocin moves milk in seconds. A mother can have a breast full of milk (adequate lactogenesis) but be unable to deliver it to a nursing infant if let-down fails (inhibited oxytocin release, often from stress) — a distinction with direct clinical relevance to breastfeeding difficulties. Both hormones nonetheless originate from the same stimulus: mechanical stimulation of sensory nerve endings in the nipple and areola by an infant's suckling. This sensory signal ascends the spinal cord to the hypothalamus and there splits into two parallel pathways — one that lifts the tonic inhibition on prolactin-secreting cells of the anterior pituitary, and one that triggers phasic, pulsatile secretion of oxytocin from the posterior pituitary. The elegance of the system is that a single behavioral act — a hungry infant nursing — recruits exactly the two hormonal responses needed to build the next feed's supply and deliver the current one.

1. Mammary gland development — priming the factory. The gland develops in three broad phases, each governed by a different hormonal cocktail.

• Embryonic/prepubertal: A rudimentary ductal tree forms along the embryonic milk line under local mesenchymal-epithelial signaling, present in both sexes at birth but arrested until puberty.
• Puberty: Rising ovarian estrogen drives ductal elongation and branching (ductal morphogenesis) into the fat pad, aided by growth hormone and locally produced insulin-like growth factor 1 (IGF-1). Progesterone plays a smaller role here; no alveoli form yet.
• Pregnancy: This is the decisive phase. Rising estrogen and, especially, progesterone (from the corpus luteum and then the placenta) drive extensive side-branching and, critically, differentiation of alveolar buds into full secretory lobuloalveolar units. Prolactin, rising steadily across pregnancy, and human placental lactogen (hPL), secreted by the syncytiotrophoblast in large quantities, act together with cortisol and insulin to complete alveolar differentiation. This full structural maturation, called mammogenesis, produces a gland dense with alveoli ready to secrete — but during pregnancy itself, actual copious milk secretion is held back.

2. Why milk does not flow during pregnancy — the progesterone brake. Prolactin receptor density on alveolar cells rises across pregnancy, and prolactin levels themselves climb into the hundreds of ng/mL by term, yet full-volume milk secretion does not begin. The reason is that high circulating progesterone, sustained by the placenta, blocks prolactin's ability to induce the final step of secretory differentiation — specifically, progesterone prevents prolactin from up-regulating tight-junction closure between alveolar cells and from triggering full expression of milk-protein genes. This is the physiological basis of lactogenesis I: the mammary epithelium becomes capable of secretion (colostrum, rich in immunoglobulins, begins to accumulate) but is prevented from switching to full-volume, lactose-rich mature milk.

3. Delivery of the placenta — the trigger for lactogenesis II. At parturition, expulsion of the placenta causes a precipitous fall in circulating progesterone (and estrogen) within 24–72 hours, while prolactin remains high. With the progesterone brake removed, prolactin now drives full activation of the secretory machinery: tight junctions close completely, lactose synthase activity rises sharply (lactose is the principal osmotic driver of milk volume), and copious milk production begins — lactogenesis II, clinically recognized as "the milk coming in," typically 30–72 hours postpartum.

4. Prolactin-driven milk synthesis — the cellular pathway. Inside the alveolar epithelial cell, prolactin binds a transmembrane prolactin receptor and signals principally through the JAK2–STAT5 pathway: receptor dimerization activates the associated Janus kinase 2, which phosphorylates STAT5 transcription factors; phospho-STAT5 dimers translocate to the nucleus and drive transcription of milk-protein genes (casein, α-lactalbumin, whey acidic protein) and of enzymes for lactose and lipid synthesis. α-Lactalbumin is a key regulatory molecule: it is a modifier subunit of the enzyme lactose synthase, redirecting galactosyltransferase to make lactose from glucose and UDP-galactose instead of its default reaction. Milk fat is synthesized as triglyceride droplets that bud from the endoplasmic reticulum and are released by an unusual mechanism in which the droplet is enveloped by a piece of the apical plasma membrane as it exits the cell (apocrine-like secretion). Milk proteins and lactose, dissolved in an aqueous vesicle, are released by conventional exocytosis. The finished secretion — water, lactose, protein, fat, minerals, immunoglobulins — accumulates in the alveolar lumen between feeds.

5. Prolactin secretion control — the neuroendocrine loop. Prolactin-secreting lactotrophs of the anterior pituitary are normally under tonic inhibitory control by dopamine, released from hypothalamic tuberoinfundibular neurons into the hypophyseal portal circulation, acting on D2 receptors on lactotrophs to suppress prolactin release. This is unusual: for most anterior pituitary hormones the hypothalamus provides a releasing factor, but for prolactin the dominant hypothalamic signal is inhibitory. Suckling-generated afferent signals reaching the hypothalamus transiently suppress dopaminergic neuron firing, releasing the lactotrophs from inhibition and producing a pulse of prolactin secretion within minutes of nursing onset. Because each nursing bout regenerates a prolactin pulse, frequent, effective suckling is what sustains ongoing milk synthesis — this is the basis of the demand-driven nature of lactation ("supply follows demand").

6. A local autocrine brake: the whey protein FIL. Milk synthesis is also locally self-limiting. A whey protein called feedback inhibitor of lactation (FIL) accumulates in the alveolar lumen alongside milk; when milk is not removed (breast becomes full), FIL concentration rises and acts locally, within that alveolus, to suppress further milk secretion — independent of systemic hormones. Frequent, complete removal of milk (frequent, effective feeding or pumping) dilutes FIL and sustains high local synthesis rates; this explains why breasts that are drained more often, or more completely, produce more milk over time, and why milk supply is gland-specific (one breast can produce more than the other depending on local drainage).

7. Oxytocin-driven milk ejection — the neuroendocrine reflex. Suckling stimulates mechanoreceptors in the nipple-areolar complex; afferents travel via the spinal cord to the hypothalamus, where they activate magnocellular neurons of the paraventricular and supraoptic nuclei whose axons project directly to the posterior pituitary. Action potentials in these neurons trigger calcium-dependent exocytosis of oxytocin from axon terminals directly into the systemic circulation (oxytocin here behaves as a classical neurohormone, not a neurotransmitter at this site). Oxytocin travels in the blood to the breast, binds oxytocin receptors on myoepithelial cells, and via a Gq-PLC-IP3-calcium pathway triggers contraction of these smooth-muscle-like cells, which squeeze the alveoli and expel milk into the ducts, out through the nipple — the milk-ejection reflex or let-down. This reflex has three hallmark features:

• It is fast: milk ejection begins within about 30–60 seconds of suckling onset.
• It is pulsatile: oxytocin is released in discrete bursts, producing intermittent, not continuous, ejection.
• It is conditionable: the sight, sound, or even thought of the infant can trigger let-down in experienced nursing mothers, because higher cortical and limbic inputs (a learned, classically conditioned stimulus) converge on the same hypothalamic magnocellular neurons — demonstrating that this reflex, though hormonally executed, is neurally gated and can be facilitated or, importantly, inhibited by stress, catecholamines, and anxiety (via competing sympathetic vasoconstriction and central inhibition of oxytocin neurons).

8. Lactational amenorrhea. Suckling-driven neural input to the hypothalamus, especially via elevated prolactin, suppresses pulsatile release of gonadotropin-releasing hormone (GnRH), which in turn suppresses pulsatile LH secretion. Without adequate LH pulses, ovarian follicles fail to mature and ovulation is suppressed, delaying return of menses postpartum — the physiological basis of the Lactational Amenorrhea Method (LAM) of contraception, effective only under strict conditions (exclusive breastfeeding, high frequency, day and night, within 6 months postpartum, before menses returns).

The two-hormone architecture of lactation was not obvious; it emerged from decades of comparative bioassay, ablation, and electrophysiology experiments.

Ott and Scott (1910) and Schafer and Mackenzie (1911) — discovery of a milk-ejecting factor. Isidor Ott and John Scott, followed independently by Edward Schafer and Joseph Mackenzie, showed that injecting extracts of the posterior pituitary (not the anterior lobe) into lactating animals caused a rapid rise in intramammary pressure and expulsion of milk, even without direct neural stimulation of the gland. This localized the milk-ejection factor to the posterior pituitary, distinct from any anterior pituitary secretion, decades before the hormone was chemically identified as oxytocin by Vincent du Vigneaud in 1953 (who also synthesized it — the first peptide hormone ever synthesized, work that earned him the 1955 Nobel Prize in Chemistry).

Cross and Harris (1952) — electrical stimulation and the neurohormonal reflex. Brian Cross and Geoffrey Harris, working with lactating rabbits, showed that electrical stimulation of the hypothalamus (specifically the paraventricular/supraoptic region) caused a measurable rise in intramammary pressure with a short latency, mimicking natural milk ejection, and that this effect was abolished by prior removal of the posterior pituitary or by sectioning the pituitary stalk (which disconnects hypothalamic neurons from their axon terminals in the posterior lobe). This experiment was decisive: it proved the reflex is neurohumoral — a neural signal from suckling is converted into a hormonal signal (oxytocin) released from the posterior pituitary into the blood, rather than milk ejection being achieved by direct nerve-to-muscle innervation of the myoepithelial cells (which receive essentially no functional motor innervation).

Cowie and Folley, and the demonstration of a prolactin-inhibiting factor. Work through the 1950s–1970s (notably by Cowie, Folley, and later the identification of dopamine's role by Neill and colleagues) established that anterior pituitary tissue transplanted away from the hypothalamus (e.g., under the kidney capsule) continues to secrete prolactin at a high, unregulated rate — unlike other anterior pituitary hormones, which fall to near zero when disconnected from hypothalamic control. This "ectopic pituitary" experiment was the classical proof that the hypothalamus normally inhibits prolactin secretion (rather than stimulating it, as it does for ACTH, TSH, GH, and the gonadotropins). Subsequent pharmacological studies showed that dopamine, delivered via the hypophyseal portal blood, is the principal physiological prolactin-inhibiting factor: dopamine agonists (e.g., bromocriptine) suppress prolactin and can stop lactation, while dopamine antagonists (e.g., some antipsychotics) elevate prolactin and can cause inappropriate lactation (galactorrhea) — a clinically observed confirmation of the mechanism.

McNeilly and colleagues (1980s) — suckling frequency and prolactin pulses in humans. Direct measurement of plasma prolactin in nursing mothers showed a sharp pulse of prolactin secretion beginning within minutes of suckling onset and returning to baseline over roughly an hour, with pulse amplitude correlating with suckling duration and frequency, not with milk yield per se — supporting the idea that suckling stimulus intensity, not milk removal itself, drives the prolactin surge, while local milk removal (via FIL) separately controls the immediate rate of synthesis in each gland.

Wakerley and Lincoln (1973) — electrophysiological recording of oxytocin neurons during suckling. John Wakerley and David Lincoln recorded extracellularly from single supraoptic and paraventricular neurons in lactating rats during suckling and discovered that oxytocin-secreting neurons fire in a highly characteristic way: a slow, gradually accelerating basal discharge is periodically interrupted by a synchronized, high-frequency burst of firing across the whole population of oxytocin neurons, lasting only a few seconds, occurring roughly every 5–10 minutes and corresponding precisely, in time, to a measurable pulse of milk ejection. This was one of the clearest demonstrations in mammalian neuroendocrinology that a behaviorally relevant hormone pulse is generated by synchronized neural bursting, directly linking single-unit electrophysiology to a whole-organ physiological output (milk ejection), and it remains a textbook model system for studying neurosecretory pulse generation.

Example 1 — Reasoning through a stalled milk-ejection reflex (clinical/physiological case). A new mother reports that her breasts feel full and heavy, and pumping with a hospital-grade pump yields almost nothing, though the infant, when nursing directly, appears to swallow normally. Two nights later, in a calm, private room with her infant skin-to-skin, pumping yields 90 mL in ten minutes. Reason through the mechanism.

• Fullness with poor pump output but adequate output during direct, comfortable nursing indicates that milk synthesis (lactogenesis) is intact — the gland is full of milk — but milk ejection is failing under one set of conditions and succeeding under another.
• Milk ejection depends on oxytocin release from hypothalamic magnocellular neurons, a reflex that is facilitated by the conditioned cues of the infant (smell, skin contact) and inhibited by sympathetic activation and psychological stress (anxiety, unfamiliar pump noise, clinical environment, pain).
• Stress-induced catecholamine release can also cause local vasoconstriction in the breast, reducing oxytocin delivery to myoepithelial cells, and elevated cortisol/adrenergic tone can centrally suppress oxytocin neuron firing.
• The improvement with skin-to-skin contact and a calm environment is exactly what is predicted if the defect is in the neurally gated, conditionable oxytocin reflex rather than in prolactin-driven synthesis — consistent with the Cross–Harris model in which milk ejection, unlike synthesis, requires an intact, unstressed neurohypophyseal reflex arc.
• Clinical conclusion: this is inhibited let-down, not low milk supply; management targets the reflex (relaxation, skin-to-skin, familiar cues, sometimes exogenous oxytocin nasal spray in refractory cases) rather than measures to boost prolactin or synthesis.

Example 2 — Quantitative reasoning: energy cost of milk production. A lactating woman produces 780 mL of milk per day, energy content ≈ 0.67 kcal/mL (roughly 670 kcal/L, typical of mature human milk). Assume mammary conversion efficiency (dietary/stored energy → milk energy) is about 80%, reflecting real physiological estimates for milk synthesis efficiency. Step 1 — Energy content of milk produced per day:

Step 2 — Maternal energy cost, accounting for conversion efficiency:

Step 3 — Compare to typical recommended lactation increment. Standard dietary guidance recommends roughly +450 to +500 kcal/day above non-pregnant, non-lactating needs for the first six months of exclusive breastfeeding. Step 4 — Reconcile the gap: the calculated maternal cost (≈650 kcal/day) exceeds the typical recommended dietary increment (≈450–500 kcal/day) by roughly 150–200 kcal/day. This gap is physiologically real and is covered by mobilization of maternal fat stores accumulated during pregnancy — pregnancy weight gain includes a deliberate energy reserve (commonly several kilograms of fat) precisely to subsidize the energetic cost of lactation, so a nursing mother is not expected to eat the full incremental cost in food; a substantial fraction is drawn from her own adipose reserves, a mechanistic link between Pregnancy Physiology fat deposition and postpartum lactation energetics.

Sir Henry Hallett Dale and, more directly relevant here, Geoffrey Wingfield Harris are central figures, but the physiologist whose career most defines the modern understanding of lactation's neuroendocrine reflex is Geoffrey Harris, often called the father of neuroendocrinology, whose work with Brian Cross established that the hypothalamus governs pituitary hormone release through a combination of neural and vascular (portal blood) connections — the conceptual framework this entire article rests on. Equally central is Vincent du Vigneaud, the chemist who determined the structure of oxytocin and then achieved its total synthesis, proving definitively that a discrete nonapeptide, not some vaguer pituitary "principle," was the milk-ejecting hormone. Harris trained as a physiologist in Cambridge and Oxford in the 1930s–40s, at a time when the relationship between the brain and the pituitary gland was almost entirely unknown; the pituitary was widely viewed as an autonomous "master gland" controlling the body's other endocrine organs with little input from the nervous system. Harris's insight — built on careful anatomical and experimental work on the hypophyseal portal vessels and on hypothalamic stimulation and lesion studies — was that the hypothalamus is itself the master controller, communicating with the anterior pituitary via portal blood-borne releasing and inhibiting factors, and with the posterior pituitary via direct neurosecretory axons. Lactation, with its clean division between a hypothalamic-inhibitory anterior pathway (prolactin/dopamine) and a hypothalamic-neurosecretory posterior pathway (oxytocin), became one of the paradigm systems used to establish this general principle across all of neuroendocrinology.

Clinical lactation medicine. Understanding the prolactin/oxytocin split underlies modern management of breastfeeding difficulties: dopamine agonists (bromocriptine, cabergoline) are used to suppress lactation (e.g., after stillbirth, in non-breastfeeding mothers, or to treat pathological galactorrhea from prolactinoma or antipsychotic drugs), while synthetic oxytocin nasal sprays or careful attention to a calm feeding environment are used to facilitate let-down in mothers with inhibited ejection.

Prolactinoma and endocrinology. Pituitary prolactin-secreting adenomas cause pathological galactorrhea and, because prolactin excess suppresses GnRH, amenorrhea and infertility in non-pregnant women — a direct clinical extension of the same suckling-suppression-of-GnRH mechanism that underlies lactational amenorrhea.

Public health and child survival. The WHO recommendation of exclusive breastfeeding for the first six months of life rests on this physiology: immunoglobulin-rich colostrum in the first days, and continually adapting mature milk thereafter, deliver passive immunity and optimal nutrition that formula cannot fully replicate; understanding supply-and-demand dynamics (frequent removal → sustained prolactin pulses → sustained synthesis) underlies lactation-support counseling used globally to prevent early breastfeeding failure.

Agriculture and dairy science. Commercial dairying is applied lactation physiology at industrial scale: milking machines are timed and designed to trigger the oxytocin-mediated let-down reflex efficiently, and selective breeding, along with the historically controversial use of recombinant bovine somatotropin (rbST, which raises IGF-1 and indirectly milk yield in cattle), are direct downstream applications of the synthesis-and-ejection framework described here.

Contraception and reproductive health programs. The Lactational Amenorrhea Method is deployed as a low-cost, hormone-suppression-based contraceptive in resource-limited settings, directly leveraging the suckling-induced suppression of GnRH pulsatility described above, though its time-limited reliability requires careful patient counseling.

Conservation and comparative physiology. Comparative studies of lactation across species (e.g., extraordinarily fat-rich milk in marine mammals such as seals, versus dilute, frequent-feeding milk in species like humans and rabbits) inform conservation breeding programs and zoo animal husbandry, where mimicking natural suckling frequency and hormonal cues is essential for successful hand-rearing or maternal bonding of endangered species.

Ask the tutor: How does prolactin's suppression of GnRH during lactation compare mechanistically to the suppression seen in hyperprolactinemia from a pituitary prolactinoma?