5 Far from the Madding Crowd
The thin grasses, more or less coating the hill, were touched by the wind in breezes of differing powers, and almost of differing natures—one rubbing the blades heavily, another raking them piercingly, another brushing them like a soft broom. The instinctive act of humankind was to stand and listen, and learn how the trees on the right and the trees on the left wailed or chaunted to each other in the regular antiphonies of a cathedral choir; how hedges and other shapes to leeward then caught the note, lowering it to the tenderest sob; and how the hurrying gust then plunged into the south, to be heard no more. The sky was clear—remarkably clear—and the twinkling of all the stars seemed to be but throbs of one body, timed by a common pulse.
—Thomas Hardy (1840–1928), Far from the Madding Crowd
Neuroendocrinology is about things that matter for our health and well-being: stress and appetite, metabolism, body rhythms, growth, and all aspects of reproduction: the reproductive cycle, sexual behavior, pregnancy and parturition, lactation and maternal behavior. In the hypothalamus, specialized neurons control the secretion of pituitary hormones that act on the gonads, the thyroid and adrenal glands, the kidneys, liver, heart and gut. These neuroendocrine systems offer a window on the brain. Their products can be measured relatively easily and, with wit and ingenuity, their effects are determinable. The activity of these neurons is interpretable to a degree equaled in few other areas of neuroscience.
If we ask of any neuron in the brain, what does it really do, the answer is often frustratingly incomplete. Even if we know how it responds to stimuli, what it makes and to which other neurons it talks, we might yet not know what it does that matters. But we can know much of what, for example, vasopressin and oxytocin neurons do even before we know how they do it. They send axons to the posterior pituitary gland, and what they secrete from there can be measured in the blood and has demonstrable consequences for important physiological functions. Vasopressin regulates the kidneys, and oxytocin regulates milk letdown from the mammary gland and controls uterine contractions during parturition—we now know that they both do much, much more than this, but when I started out as a neuroendocrinologist we knew just these things for sure.
These neurons once seemed unusual in that they use peptides for signaling. Peptides are much larger than conventional neurotransmitters, and hard to make. A peptide is a chain of amino acids, and which particular amino acids and in which order they are assembled is determined by the gene for that peptide. Oxytocin has nine amino acids and a molecular weight of about 1,000 daltons, and the oxytocin gene must produce it as a fragment of a much larger precursor peptide, which has a molecular weight of about 23,000 daltons. The precursor contains the oxytocin sequence and other sequences that determine what the cell will do with the oxytocin. Part of the precursor determines that it will be packaged into vesicles to be transported to sites where the vesicles can be released. Another part is important for folding the precursor in a way that enables it to aggregate with other precursor molecules so they can be packaged compactly in a vesicle. The vesicles also contain enzymes that cleave oxytocin from the rest of the precursor. It’s complicated and expensive, but the final product is a powerful molecule: oxytocin will survive in the extracellular fluid for much longer than ordinary neurotransmitters, it can act on cells at much lower concentrations, and it acts at sophisticated receptors that have a range of properties through which they control complex signaling pathways within those target cells.
When I began as a neuroscientist, oxytocin was one of only a handful of peptides known to be released by neurons. Now, peptides are no longer rare and curious: more than a hundred are so far known to be released by neurons, to act not only on other neurons but also on the even vaster numbers of nonneuronal cells in the brain. Some peptides control the blood flow that carries the oxygen that neurons consume. Others control the growth of synapses between neurons and the removal of disused synapses. Many regulate the expression of genes in the cells that they target. Some control the shape as well as the function of the glial cells that are so abundant in the brain, which cleanse the external environment and provide growth factors and metabolites to support neuronal function. Many influence neuronal excitability directly, or modulate the effects of neurotransmitters.
One striking characteristic of peptides is their ability to orchestrate behavior—to coordinate different systems to evoke a coherent, adaptive, organismal response, be it maternal behavior, aggression, sexual arousal, or behaviors associated with hunger and thirst: foraging, feeding, and drinking. An injection of NPY (neuropeptide Y) into the brain will provoke feeding, and an injection of α-MSH (α-melanocyte-stimulating hormone) will stop it. Orexin can wake a sleeping beauty. Angiotensin can make a horse drink. Vasopressin can make her turn to violence, and oxytocin make him maternal. The ability of peptides to evoke such dramatic “global” consequences invites us to look for things that are special about peptide signaling, and for this, the neuroendocrine systems have been a rich source of inspiration.
The hypothalamus is at the base of the brain. If you curl your tongue back as far as you can and press it on the roof of your mouth, the hypothalamus will be almost on the tip of your tongue. Your pituitary gland sits just a little farther back, below the hypothalamus and connected to it by the neural stalk, a thin strip of tissue that contains blood vessels and a bundle of axons. The gland in humans is described in Wikipedia as being the size of a pea. So common is this description that it seemed likely to be wrong, as I confirmed by examining a selection of peas. Wikipedia also gives the weight of the human pituitary as about half a gram, and in this it is more correct. The pituitary in a human is at least five times the average size of my peas; it is more like the size of a chickpea.
The pituitary has three distinctive lobes that produce nine main hormones, and the anterior lobe makes six of these. Luteinizing hormone (LH) and follicle-stimulating hormone (FSH) are gonadotropins, and regulate the gonads—the ovaries and testes. Prolactin regulates the production of milk by the mammary gland. Growth hormone acts on the liver to stimulate the production of insulin-like growth factor 1, which promotes bone growth and muscle development. Thyroid-stimulating hormone regulates the thyroid gland, which controls our metabolic rate by secreting other hormones that act on almost all of the cells in our body, and which are also critically important for brain development. Adrenocorticotropic hormone (ACTH), secreted in response to stressors, regulates the production of glucocorticoid hormones by the adrenal gland, and these affect blood glucose levels, fat metabolism, and blood flow to muscles. These six pituitary hormones are made in five types of endocrine cells: LH and FSH are both made by one of these cell types, the gonadotrophs. The gonadotrophs, lactotrophs, somatotrophs, thyrotrophs, and corticotrophs are all regulated by factors released from hypothalamic neurons into blood vessels at the base of the brain that communicate between the hypothalamus and the pituitary. This was first shown by Geoffrey Harris, who, by transplanting the pituitary gland to various sites in the body, and by severing the blood supply to it from the hypothalamus, showed that the pituitary could survive without its blood supply from the hypothalamus, but could not function.1
The intermediate lobe of the pituitary contains cells that contain α-MSH, which promotes the production of the pigment melanin in the skin. These cells (melanotrophs) are directly innervated by axons from the hypothalamus—as are endocrine cells in the anterior pituitary of fish.
The posterior lobe of the pituitary (also called the neural lobe or neurohypophysis) is also innervated by neurons of the hypothalamus, but it has no endocrine cells: the axonal endings of these neurons secrete oxytocin and vasopressin directly into the systemic circulation.2 This system has been the focus of intense interest for many years, because of its unique tractability. The neurons are large, and they are aggregated in compact nuclei in the hypothalamus, making them relatively easy to recognize and study. They make huge amounts of their product—enough to be active throughout the body, making them amenable to studying the regulation of peptide synthesis, metabolism, and transport. Their axons are compactly and conveniently aggregated in the neural lobe away from any other neuronal elements, making it possible to study their properties and how electrical activity in them regulates secretion. And what they secrete is readily measurable in the blood and in the brain and can be related directly to physiological functions (figure 5.1).
The neuroendocrine systems of the hypothalamus. The anterior pituitary contains endocrine cells that produce six major hormones. The secretion of these hormones is controlled by “hypothalamic hormones” released by small populations of neurons in different regions of the hypothalamus. LH and FSH secretion is controlled by GnRH from neurons in the rostral hypothalamus; growth hormone secretion by GHRH from the arcuate nucleus and by somatostatin from the periventricular nucleus; prolactin secretion by dopamine from the arcuate nucleus; TSH secretion by TRH from the paraventricular nucleus; and ACTH secretion by CRF, also from the paraventricular nucleus. The posterior pituitary contains no endocrine cells: here, the two hormones oxytocin and vasopressin are secreted from the axonal endings of magnocellular neurons of the supraoptic and paraventricular nuclei. Between the posterior and anterior lobes is the intermediate lobe (not shown here). It contains cells that produce MSH, and these are directly innervated by another population of dopamine cells in the arcuate nucleus.
Neuroendocrinology has its icons, and the milk-ejection reflex is one. The Origin of the Milky Way, painted by Tintoretto in about 1575, hangs in the National Gallery in London; it shows Jupiter holding the infant Hercules to Juno’s left breast. Some milk from that breast spurts up to form the Milky Way; but milk also spurts from the unsuckled breast. From this we might correctly conclude that, while the letdown of milk is a response to the suckling of an infant, this response is not local to the breast that is suckled—it involves a systemic mediator. That mediator is oxytocin, and it causes cells of the mammary gland to secrete milk into a collecting duct. Oxytocin is secreted into the blood from the posterior pituitary but is made in the hypothalamus, in the magnocellular neurons of the supraoptic and paraventricular nuclei. When young suckle, oxytocin is secreted as a pulse into the blood; that pulse is the result of nearly synchronous activation of every magnocellular oxytocin neuron in an intense burst of spike activity that lasts for little more than a second.
Oxytocin has many roles. It is secreted during sexual activity: it modulates the lordosis reflex, the posture that in many female mammals is the signature of their willingness to mate; it promotes penile erection in males; and it is secreted during parturition, with potent effects on the uterus. In rats, it is also secreted in response to food intake: it promotes sodium excretion, affects gut motility, and induces satiety—the feeling of fullness after a meal that blunts our appetite. It also has a fascinating range of behavioral effects: in the spinal cord it modulates pain, in the amygdala it suppresses fear, in the paraventricular nucleus it dampens responses to stressors, and best known of all, it influences social attachment. Mice that lack oxytocin can still mate, get pregnant, and give birth; oxytocin is important for these things but not essential for them. However, their young will not survive unless they are cross-fostered; their mother will nurse them but cannot feed them: the milk-ejection reflex absolutely requires oxytocin.3
As well as showing that the anterior pituitary is controlled by hypothalamic “releasing factors,” Geoffrey Harris helped build our understanding of the posterior pituitary. He knew that oxytocin could stimulate milk letdown in lactating animals, he saw that this could be measured as an increase in pressure inside the mammary gland, and, with Barry Cross, he showed that electrical stimulation of the neural stalk increased intramammary pressure. Harris and Cross proposed that, in response to suckling, oxytocin is secreted as a result of an increase in the spiking activity of hypothalamic neurons, propagated along the axons that pass through the neural stalk.4,5 When Barry Cross moved to the Chair of Anatomy at Bristol, he set out to test this by recording the spiking activity of oxytocin neurons using microelectrodes.
The supraoptic nucleus of the hypothalamus contains only oxytocin and vasopressin neurons, all of which project to the posterior pituitary. However, this small site is surrounded by other neurons, and a microelectrode aimed there, however carefully, will not always hit its target. The breakthrough came from finding a way to be certain that the microelectrode has reached the intended target—from finding a way to identify the neuroendocrine neurons. In 1966 Kinji Yagi showed that electrical stimulation of the neural stalk would trigger spikes that were conducted not only “down” the axon, toward the nerve endings—but also “up” the axon (antidromically), toward the cell bodies. This made it possible to identify the neurons that project to the pituitary,6 and thus to study them in living animals.
In Barry Cross’s department, Dennis Lincoln and his PhD student Jon Wakerley set out to study the milk-ejection reflex.7,8 In anesthetized lactating rats, they cannulated a mammary gland to record intramammary pressure and placed a stimulating electrode on the neural stalk. They then guided a microelectrode into the supraoptic nucleus, using landmarks on the skull to determine the appropriate point of entry of the microelectrode into the brain. When they encountered a neuron, they could tell whether it was a supraoptic neuron by whether it displayed antidromic spikes when the neural stalk was stimulated.
After finding a supraoptic neuron, they allowed a litter of pups to suckle. The first surprise was not how strongly the suckling affected supraoptic neurons but how little effect it had. Some speeded up slightly, others slowed down, but generally nothing much happened. Some “phasic” neurons fired with long bursts separated by long silent periods; many others fired slowly and apparently randomly. After about 15 minutes of suckling, some of the slow-firing cells showed an explosive burst of spikes followed about ten seconds later by a sharp increase in intramammary pressure. These milk-ejection bursts recurred every few minutes while the pups were suckling, and each was accompanied by a stretch reflex in which the pups extended their forelegs stiffly to press on the mammary gland while they sucked.
Milk-ejection bursts last on average for just two seconds and contain about 100 spikes; within the first few spikes the bursts reach a peak of intensity that is sustained for about half a second, then wanes, and is followed by several seconds of silence. This pattern is stereotyped; the first bursts are smaller, but once the reflex is established, successive bursts in the same cell are very consistent in size and shape. The bursts occur at almost exactly the same time in all of the magnocellular oxytocin neurons. Other neurons in the supraoptic nucleus that did not participate in these bursts could be assumed to be vasopressin cells, including most of the phasic neurons.
Thus came the realization that, in response to suckling, oxytocin is secreted in pulses that result from synchronized bursts of spike activity. While the reflex is a response to the sucking of pups at the nipples, the activity of oxytocin cells does not passively follow this input. Rather, the input “permits” the oxytocin neurons to display bursts that are generated by some deterministic process. There is, however, something special about the suckling input: many other stimuli activate oxytocin cells strongly but never produce bursts.
This discovery—and it was a discovery, for although it may now seem that the nature of the reflex could have been predicted—came as a complete surprise. After these experiments, no study of any hormone secretion was complete without studying not just how much hormone was secreted but also the pattern in which it was secreted. Other workers began taking frequent samples to study hormone secretion and found that LH, FSH, prolactin, ACTH, thyroid-stimulating hormone, and growth hormone were also secreted in pulses. Two questions came to dominate neuroendocrinology: why are hormones secreted in pulses, and how are the pulses generated.
Geoffrey Harris had noted that electrical stimulation of the neural stalk would evoke an increase in intramammary pressure only if high frequencies of stimulation were used. This is partly because the response of the mammary gland has a narrow dynamic range. Oxytocin causes milk letdown by causing myoepithelial cells of the mammary gland to contract: but for any particular myoepithelial cell this is pretty much an all-or-nothing response. There must be enough oxytocin before it has any effect, and a maximum effect is achieved at a dose that is not much more than just enough.9 In a lactating rat, the maximum effect is achieved when about 2 ng of oxytocin is injected intravenously as a bolus—the same amount as is secreted after each milk-ejection burst. If the same dose is given slowly it has little effect, and if much higher doses are infused slowly, then, after an initial peak the intramammary pressure diminishes even while oxytocin is still being infused, and the gland becomes insensitive to oxytocin. Thus the mammary gland requires pulsatile secretion of oxytocin.
Many organs and tissues desensitize when exposed to continuously high levels of a hormone. However, there are exceptions: in late pregnancy, oxytocin precipitates uterine contractions that continue in the presence of high levels of oxytocin. Moreover, in rats, oxytocin has another function; at low concentrations it stimulates sodium excretion (natriuresis), partly by an effect on the kidneys and partly by stimulating the secretion of a natriuretic hormone from the heart. This natriuretic effect requires continuous exposure to oxytocin, and the effect is graded according to concentration. Thus, whether pulses are necessary or not depends upon the properties of the target tissues, which implies that there has been co-evolution of the properties of the target organ and the patterning of hormone secretion.
There is another phenomenon behind Harris’s observation. How much oxytocin is secreted by electrical stimulation depends on the frequency of stimulation: each spike within a burst releases much more oxytocin than the isolated spikes that occur between bursts.10 The background activity of oxytocin neurons is about 2 spikes per second, and during a burst it rises to 50–100 spikes per second. But the bursts occur only every few minutes, last for only about two seconds, and are followed by up to ten seconds of silence. So, overall, suckling hardly increases the spiking activity at all; the reflex involves more a reorganization of activity than an amplification, and it is effective because bursts are so efficient at releasing oxytocin. Again, this is true but not inevitable. Stimulus-secretion coupling—the manner in which electrical activity affects secretion—is often nonlinear but the nonlinearities are different in different systems. The properties of the terminals and the mechanisms of burst generation have coevolved.
Oxytocin’s role in milk ejection is indispensable: mothers that lack oxytocin cannot feed their young. By contrast, although oxytocin is named for its effects on uterine contractility (from the Greek for “quick birth”), mice that lack oxytocin deliver their young relatively normally. In 1941 James Ferguson in Toronto reported that distension of the uterus and cervix could induce oxytocin secretion in the pregnant rabbit: this Ferguson reflex is now known to be a feature of all mammals, and since oxytocin can stimulate uterine contractions, which in turn stimulate oxytocin secretion, it is a rare example of a biological positive feedback system.11 Nevertheless, in the absence of oxytocin, other mechanisms can ensure a successful delivery. In the year that Ferguson described his reflex, Dey and colleagues reported on the effects of lesions to the neural stalk in pregnant guinea pigs that prevent any secretion of oxytocin.12 Of 16 labors that they studied, 10 were prolonged and difficult but 6 were normal.
Geoffrey Harris had shown that electrical stimulation of the neural stalk could evoke uterine contractions in pregnant rabbits, but being a cautious and skeptical scientist, he remained unsure whether this meant that oxytocin was really secreted during parturition, or whether it was merely a “pharmacological” effect, of no physiological significance. His experiments prompted a trainee obstetrician, Mavis Gunther, to write to the British Medical Journal.13 She had attended labor in a woman who was still breast-feeding a previous child, and had noticed that, during each uterine contraction, beads of milk appeared at the nipples. Many factors were known that could elicit uterine contractions but only oxytocin was known to induce milk letdown, so Gunther proposed that uterine contractions must have provoked the secretion of oxytocin, which therefore must act in a positive-feedback manner to provoke further contractions and thereby support parturition.
However, soon it came to be recognized that the placenta of pregnant women synthesized an enzyme, oxytocinase, that could degrade oxytocin, and that plasma concentrations of oxytocinase increased dramatically toward term.14 This was strange: if oxytocin was important for parturition, why did the placenta produce an enzyme that destroys it?
In the 1980s, in Barry Cross’s former department in Bristol, Alastair Summerlee and his colleagues recorded the electrical activity of oxytocin cells in conscious rats and rabbits throughout parturition and subsequent lactation.15–18 The milk-ejection reflex in conscious rats was essentially identical to that described in the anesthetized rat, and similar bursts occurred during parturition; these were linked to the delivery of the young. In rabbits, things were only slightly different. Whereas rats allow their young to suckle continuously for long periods, rabbits nurse theirs for just a few minutes each day. Their oxytocin cells showed bursts that looked like those in rats, but each period of suckling was associated with several bursts in quick succession, and the bursts continued for a while after suckling, apparently in response to grooming of the nipples by the doe.
Parturition in both rats and rabbits was also accompanied by bursts that were associated with pulses of oxytocin secretion, and with delivery of the young. The recognition that oxytocin secretion was pulsatile during parturition cast a new light on the role of oxytocinase, for while high concentrations diminish the basal levels of oxytocin, they also “sharpen” pulses of oxytocin by reducing their half-life. By frequent blood sampling combined with methods to inactivate oxytocinase in the samples, Anna-Rita Fuchs and coworkers showed that spontaneous delivery in women is also accompanied by pulsatile secretion of oxytocin.19
But are the pulses necessary? This was less clear, as the uterus continues to contract in the continued presence of oxytocin. Nevertheless, in the rat, pulses are indeed a more effective way for oxytocin to drive parturition. In my lab at Babraham, Simon Luckman showed this by first interrupting parturition with morphine (a potent inhibitor of oxytocin neurons).20 He showed that normal parturition could be reinstated with pulses of oxytocin given every ten minutes, but not by the same amount given by continuous infusion.
The trigger for parturition varies between species, but in all mammals oxytocin (or in marsupials, its ortholog, mesotocin) has a role.21 Oxytocin is not essential—other mechanisms can compensate for its absence—but it is secreted in large amounts during labor, it acts on a uterus that expresses greatly increased numbers of oxytocin receptors, and blocking either oxytocin secretion or its actions slows parturition.
The explosive nature of milk-ejection bursts suggested that some positive feedback might be involved in generating them, and Barry Cross and his colleagues set about to test whether oxytocin itself could excite the oxytocin cells.22 They recorded from magnocellular neurons in rats and rabbits and gave oxytocin either intravenously or directly to the neurons (by iontophoresis, which involves expelling drugs from a micropipette). The results were disconcerting: direct exposure to oxytocin excited many of the magnocellular neurons, but when given intravenously, even very large doses of oxytocin had no effect.
When these experiments were conducted there was no evidence that oxytocin was released in the brain, and Barry Cross recognized that the ineffectiveness of intravenous oxytocin meant that oxytocin secreted from the pituitary probably did not find its way back into the brain. He thus thought that the direct effects of oxytocin on oxytocin cells were of no physiological significance. He was right on the first issue: there is a blood-brain barrier to oxytocin, later demonstrated compellingly by Wim Mens and his colleagues in Utrecht, who injected massive amounts of oxytocin intravenously into rats and showed that, although plasma concentrations increased a thousandfold, levels in the cerebrospinal fluid scarcely changed.23 But Cross was wrong on the second issue.
This became apparent when Philippe Richard and Marie-José Freund-Mercier in Strasbourg showed that small amounts of oxytocin injected into the brain dramatically facilitated the milk-ejection reflex.24 While it might have been expected that oxytocin would excite oxytocin neurons, this is not what they saw. Oxytocin had no effect in rats that were not being suckled and only slight effects on the background activity of oxytocin cells in rats that were. But, for 15 to 30 minutes after injecting oxytocin, they saw an increase in the size and frequency of milk-ejection bursts. These effects could be evoked by injecting as little as 12 picograms into the brain—a tiny amount (figure 5.2).
The milk-ejection reflex, uncovered by Wakerley and Lincoln using electrophysiological studies in anesthetized rats. In response to suckling, oxytocin cells discharge intermittently in brief synchronized bursts that evoke secretion of pulses of oxytocin, and these pulses induce abrupt episodes of milk letdown, reflected by increases in intramammary pressure. This reflex is dramatically potentiated by injections of very small amounts of oxytocin into the brain. This was the first clear demonstration that peptides, released centrally, can have potent physiological actions.
The milk-ejection reflex had seemed to be a product of a sophisticated neuronal network that transformed a fluctuating sensory input (from the suckling of hungry young) into stereotyped bursts that were synchronized among the entire population of oxytocin cells. Yet injecting oxytocin into the brain with no spatial or temporal sophistication resulted in an apparently specific modulation of this reflex, and did so at impressively low doses.
Did these observations reflect an action of oxytocin on oxytocin cells? Oxytocin produced simultaneous bursts, but the neurons do not communicate with each other either by chemical synapses or by electrical synapses. There seems to be no physical connection between the neurons in one supraoptic nucleus and those in the other, which are separated by the third cerebral ventricle. If there is positive feedback, why weren’t bursts triggered by any stimulus that excited oxytocin cells? Bursts occurred during suckling and also during parturition, but never in response to any of a growing catalog of other kinds of stimuli that could excite oxytocin cells.
And yet, further experiments of Philippe Richard, Françoise Moos, and their colleagues showed that the effects of oxytocin on the milk-ejection reflex must reflect an action on oxytocin cells.25 First, they used the technique of push-pull perfusion to sample fluid from the supraoptic nucleus during suckling. This involves placing a probe into the brain that consists of two concentric cannulae, one to slowly infuse artificial cerebrospinal fluid, and one to collect fluid at the same rate. Substances released in that region of the brain would mix with the infused fluid and be recovered by the collecting cannula. The researchers saw that oxytocin release was increased even before any increase in the blood was detected, and hence before the oxytocin cells showed any increase in spiking activity. Second, they found that small amounts of an oxytocin antagonist microinjected into just one supraoptic nucleus would block the reflex. To a physiologist, experiments with antagonists have a special significance: they reveal the actions of an agonist produced by the brain itself—an endogenous agonist. Thus oxytocin release in the hypothalamus was essential, and there had to be some communication between the nuclei if an intervention in one could block bursts in the other. Third, they studied isolated oxytocin cells in vitro and used a probe to measure the intracellular calcium concentration—a dye that produces a fluorescent signal when the calcium concentration increases. With this, they showed that oxytocin indeed acted directly on oxytocin cells: the oxytocin cells responded by releasing calcium from their intracellular stores.
These experiments were the first convincing demonstration of a physiological role for any peptide in the brain. They did not explain the milk-ejection reflex but defined the questions that had to be answered before it could be explained. Building that explanation took another twenty years. The questions posed had no precedent in our understanding. Where did the oxytocin that was released in the supraoptic nucleus come from, if not from synapses? What triggered its release, if that release was not governed by spiking activity? What synchronized the oxytocin cells, if they were not linked by either synapses or electrical junctions?
We now know that more than a hundred neuropeptides are expressed in different neuronal populations, and that most if not all neurons in the entire brain release one or more peptide messengers as well as a conventional neurotransmitter. Because peptides have a long half-life and act at receptors at very low concentrations, their actions are not confined to targets adjacent to the site of release. Importantly, peptides in the brain often have organizational and activational roles that seem more like the roles of hormones in the periphery.
Whereas conventional neurotransmitters are packaged in small vesicles that are released only at synapses, peptides are packaged in large vesicles in all parts of a neuron, including in the cell body and dendrites, and can be released from any of these. Peptides can be released by spiking activity, but, whereas at a synapse one spike might release one synaptic vesicle, it may take many hundreds of spikes to release one peptide-containing vesicle. Some stimuli can alter the availability of vesicles for release, and this changes the functional connectivity of neurons. Some stimuli, by mobilizing intracellular signaling cascades, can cause peptide release from dendrites without any release of conventional transmitters from axon terminals—which lack local calcium stores. Peptide signals can be long-lasting and can act at considerable distances from their site of release. And some neurons can change their phenotype in different physiological states, expressing different peptides.
This view of the brain is different from the conventional portrait. It shows the hypothalamus as a “Europe” of the brain, a confusion of small nations, each noisy, heterogeneous, and sometimes strident. Each nation contains multiple clans that use a variety of languages and other signals that act at diverse spatial and temporal scales to communicate with other clans and neuronal nations.
Notes