CHAPTER 5 FEEDBACK CONTROL IN ENDOCRINE SYSTEMS
Principles and Practice of Endocrinology and Metabolism
CHAPTER 5 FEEDBACK CONTROL IN ENDOCRINE SYSTEMS
DANIEL N. DARLINGTON AND MARY F. DALLMAN
Localization of Components
Amplification of Output
Loads on Endocrine Systems
Examples of Feedback in Endocrine Systems
Hypothalamic–Pituitary–Target Endocrine Feedback
Feedback Effects on Amplitude and Pulse Frequency: The Gonadotropin System
Parturition and Suckling
Endocrine Cells as Receptors and Effectors
Receptors in Proximity to the Endocrine Gland
Receptors that can Initiate Multisystem Responses
The function of endocrine systems is to maintain homeostasis of the organism and to perpetuate the species. Homeostasis is maintained by the continual adjustment of nervous and endocrine systems in response to a changing environment. Because most endocrine systems are intimately related to the central nervous system and the autonomic nervous system (see Chap. 8,Chap. 9 and Chap. 85), a continual bidirectional flow of information—one relaying the conditions in the organism, and the other directing neural and endocrine responses—is present and functions to maintain appropriate operating conditions.
To produce stable systems, it has been useful to apply the knowledge gained from the engineering discipline to physiologic control systems. The quantitative relations among system components in engineering and the mathematical description of these have been useful in describing biologic control systems.1 The notion of a setpoint reference signal, which can be explicitly delineated in an engineered system, is, in most biologic systems, an apparent setpoint, deduced by the behavior of the system, and probably normally arises from the interaction of a hormone or metabolite with receptors and cellular function.
Feedback is a process within a system that occurs when the product or result of activity in the system modifies the factors that produce that product or result. The result can modify these factors in either of two ways: (a) by stimulating the factors to generate more product (positive feedback), or (b) by inhibiting the factors so that fewer products are generated (negative feedback).
Mammalian organisms are composed of many feedback systems. Each system has its own specific function and is self-regulating. However, all of the systems are interconnected, making the overall system extremely complex. Usually, negative feedback is used to maintain a variable within a range that is advantageous for optimal cellular function. When an internal or external stimulus perturbs a regulated variable, neural or endocrine responses (or both) occur that counteract the disturbance and return the variable to within its normal range. Such performance requires a means to determine the level of activity or the concentration of the variable that is regulated.
The components of a simple endocrine feedback system are shown in Figure 5-1. A variable is monitored by a receptor that is sensitive to changes in that variable. The receptor is connected to a structure or structures that process and integrate the signal generated by alterations in the variable. The signal processor includes a reference setpoint and a comparator element that compares the receptor input with the reference signal. The difference between input from the receptor and the setpoint is generated by the comparator, and the resulting error signal then modifies the existing activity of the effector endocrine gland. If the error signal is positive, the effector gland is stimulated to secrete, and the concentration of hormone in the circulation increases to levels that are adequate to effect the normalization of the perturbed regulated variable. Because many hormones are secreted into the circulation, the concentration of hormone achieved is affected by the rate of clearance and by the amount of hormone that is bound in the plasma compartment by a specific carrier protein (i.e., only the free, unbound hormone is available to cellular receptors).
FIGURE 5-1. Model for a simple endocrine feedback system consisting of a variable being monitored by a receptor, a signal processor that analyzes the signal, and an endocrine gland with a secretion rate that is regulated by the processor. The concentration of free hormone in the circulation depends on the clearance of the hormone and the presence and concentration of specific binding protein in the circulation. Any change in the variable is sensed, and a response (hormone) is elicited to bring the variable back within its “normal” range.
LOCALIZATION OF COMPONENTS
The receptor, processing unit, and endocrine gland may all reside within a single cell or in a group of similar cells functioning in concert, or they may involve entirely separate and widely dispersed units. For example, the B cell (b cell, insulin-producing) of the pancreas is directly sensitive to changes in extracellular glucose concentration (see Chap. 134). An increase in blood glucose is thus sensed by the B cell and increases the rate of insulin synthesis and secretion. The increased concentration of insulin causes increased glucose transport into those cells that contain insulin receptors and a reduction in blood glucose concentration. The normalization of glucose concentration eliminates the stimulus to the B cell.
Arterial blood pressure is monitored by mechanoreceptors in the carotid artery and aortic arch. The central nervous system processes and integrates afferent information from the mechano-receptors. The autonomic nervous and various endocrine systems represent the “gland” that secretes norepinephrine and various hormones; when there is a marked decrease in arterial pressure, the release of these hormones effectively restores arterial blood pressure to normal.
AMPLIFICATION OF OUTPUT
All feedback systems amplify their outputs (glandular secretions) to correct for the effects of perturbations. The effectiveness of this amplification in maintaining a regulated variable within the normal range is termed the gain of the system. The gain of a system is determined by comparing the magnitude of excursion of a regulated variable under conditions in which the feedback loop is abolished with the magnitude of excursion under conditions in which the feedback loop is intact.
Figure 5-2 shows a theoretical situation in which the control of glucose is examined. At the moment of onset of the glucose infusion, the pancreatic B cells are either removed from the system or left itact. In the absence of the B cells, insulin secretion would not occur, and the blood glucose concentration would increase because of the glucose infusion in proportion to the infusion rate. The steady-state glucose concentration would be determined by the volume into which glucose is distributed and the rate at which glucose is lost from that volume of distribution in the absence of insulin. With the B cells of the pancreas in the system (i.e., under closed-loop conditions), the infusion-induced increase in blood glucose concentration would increase the concentration of glucose bathing the pancreatic B cell and cause secretion of insulin. The elevated insulin concentration activates increased target cell uptake of glucose, lowering the circulating glucose levels. Glucose concentration in the open-loop condition increases well above the normal range; whereas under closed-loop control, glucose concentration is maintained close to the normal range.
FIGURE 5-2. Gain in an endocrine feedback system. The diagram shows the (hypothetical) blood glucose (top) and insulin (bottom) concentrations before and during infusion of glucose. The gain is defined as the ratio of the difference between the responses (perturbed glucose levels) of the open-loop and closed-loop conditions (arrow b) and the difference between the response of the closed-loop glucose and the preinfusion levels (arrow a). The gain is an index of feedback loop effectiveness. (Redrawn from Guyton AC. Textbook of medical physiology, 5th ed. Philadelphia: WB Saunders, 1976.).
The error in the regulation of blood glucose is the difference between the preinfusion value (see Fig. 5-2, dotted line) and the new level (see Fig. 5-2, arrow a). The degree by which the variable (glucose) is controlled is the difference between the open-loop glucose concentration and the closed-loop glucose concentration (see Fig. 5-2, arrow b). The gain of the system can then be calculated as the ratio between arrow b and arrow a.2 In this example, the difference between the open-loop and closed-loop response is 10 units, and the difference between the closed-loop response and preinfusion levels is 2 units. The gain of this system is, therefore, 10/2, or 5 (usually denoted as a negative number to show normalization). Thus, for each unit increase of the controlled variable under closed-loop conditions, there would be a five-fold change if the feedback loop were open.
The effectiveness of feedback can also be described in terms of the time elapsed between the onset of the stimulus (the change in the regulated variable) and the effect of the system response. This is a function of the time constant of each component of the feedback loop. In a multicomponent feedback system, the component with the slowest time constant determines the delay time between stimulus and response (time constant of the system). Figure 5-3 compares the imagined performance of two systems, one with a fast-time constant (left) and the other with a slow-time constant (right). Time constants may clarify the requirement for the hierarchical feedback loops that are so commonly encountered in biologic systems.
FIGURE 5-3. Comparison of the effectiveness of feedback systems (hypothetical) with fast- and slow-time constants. Notice that for the same stimulus, the feedback system with the fast-time constant regulates the variable near normal levels and does not overshoot or under-shoot as does the feedback system with the slow-time constant.
The time constants for transmission of neural information are in the milliseconds to seconds range. Peptide and protein hormones, acting on receptors in the plasma membrane to alter intracellular concentrations of a second messenger, exert many of their primary effects on cellular function over the range of seconds to minutes. Still slower, in hours, are the effects of steroid and thyroid hormones, which are mediated by alterations in the rate of gene transcription and translation, and of protein synthesis. Interestingly, those hormones with the longest time constants (adrenal and gonadal steroids, hepatic somatomedins) are secreted by glands controlled by other hormones with faster time constants, and these systems exhibit high-gain feedback control. Moreover, the feedback inhibition exerted by these hormones with characteristically slow target effects may be mediated by mechanisms with faster time constants than those of their target effects. Thus, the overall performance of a relatively slow system may appear tighter than anticipated and more like that of a system with an overall fast-time constant.
For instance, the secretion of glucocorticoids is regulated by a cascade system in which corticotropin-releasing hormone (CRH), secreted from the median eminence of the hypothalamus in response to appropriate stimulation, acts on corticotropes in the anterior pituitary to cause synthesis and secretion of adrenocorticotropic hormone (ACTH). ACTH, likewise, acts on the zona fasciculata cells of the adrenal cortex to stimulate the synthesis and secretion of cortisol and corticosterone (see Chap. 14). The initiation of a stimulus to CRH until a rise in circulating concentration of cortisol or corticosterone occurs requires <5 minutes. The minimum time required for cortisol to increase gluconeogenesis, enabling increased hepatic glucose secretion, may be ~2 hours. If insulin-induced hypoglycemia were the stimulus that provoked CRH secretion, return of the extracellular glucose concentration to normal would relieve the stimulus to CRH. However, in the absence of a more rapid cortisol feedback effect on the hypothalamic CRH-secreting cells and on the corticotrope cells that secrete ACTH, there would clearly be an overshoot in the amount of cortisol secreted in response to the stimulus of hypoglycemia because of the lag time between the elevated concentrations of cortisol in the circulation and the effect of these elevated levels on peripheral target cells. The direct feedback effects of cortisol, the other steroid hormones, and thyroid hormones on their trophic hormonal controllers allow tighter control of these systems.
LOADS ON ENDOCRINE SYSTEMS
Endocrine systems have the ability to restore homeostasis in response to varying degrees of perturbation. The load on the system is the degree or strength of the perturbation. In response to the perturbation, the endocrine system adjusts the amount of “free” circulating hormone in an attempt to regain homeostasis. The amount of free circulating hormone (or hormonal load) is dependent on the secretion and clearance rates of the hormone, and the production and clearance rates of the specific hormone-binding proteins. Newly secreted free hormone is immediately accessible to clearance from the circulation by metabolism in liver, kidney, and other tissues, and by filtration and excretion by the kidney. Moreover, many hormones are bound in the circulation by high-affinity specific binding globulins. The “bound” hormone is unavailable for diffusion into the interstitial fluid and cannot affect the target organ or tissue. However, the bound and free hormone exists in an equilibrium in plasma such that as free hormone is cleared, bound hormone becomes free. Bound hormone acts as a storage of free hormone. If, for any reason, the rate of metabolism or clearance changes, or if the production or clearance rates of its specific binding hormone changes, then the input signal to the endocrine gland that is producing the hormone must change to accommodate the change in load of free hormone.
EXAMPLES OF FEEDBACK IN ENDOCRINE SYSTEMS
Importantly, our current state of knowledge does not allow us to apply all these principles to most endocrine systems. Nonetheless, there is plentiful evidence for the existence of feedback regulation in all endocrine systems. Knowledge of this feed-back has been used in the design and interpretation of many clinical tests for the determination of endocrine disorders.
HYPOTHALAMIC–PITUITARY–TARGET ENDOCRINE FEEDBACK
One of the most clinically obvious and simplest forms of negative feedback control in endocrine systems involves suppression of the secretion of a trophic factor (or hormone) by the hormone it stimulates. For example, hormone A stimulates the secretion of hormone B, which in turn suppresses the secretion of hormone A. Hormone B may suppress the secretion of hormone A by acting directly on the cells that secrete A, or indirectly, by acting on the cells (or neurons) that stimulate the secretion of A. This type of control is exemplified in the relations between the hypothalamus, anterior pituitary gland, and peripheral endocrine glands controlled by pituitary hormones. The hypothalamus secretes neurohormones that stimulate (or inhibit) the secretion of specific anterior pituitary hormones, which in turn stimulate a peripheral target gland to secrete hormone and, with sufficient stimulation, to grow (see Chap. 8 and Chap. 9).
In Figure 5-4, CRH-containing neurons in the hypothalamus release CRH into the hypophysial portal system.3 ACTH released from corticotropes in response to CRH stimulates cortisol synthesis and secretion from the adrenal cortex. Cortisol acts to inhibit the secretion of ACTH from the corticotrope and to inhibit CRH secretion from the hypothalamic neuron, and it may also act on extrahypothalamic sites that regulate CRH synthesis and secretion.4,5 Clinically, the feedback effects of cortisol are important. Long-term therapy with pharmacologic amounts of glucocorticoids suppresses ACTH secretion to the extent that the adrenal cortices atrophy and become unresponsive to ACTH. The atrophic adrenal cortex does not secrete normal quantities of cortisol, and the abrupt discontinuation of exogenous glucocorticoids may lead to a patient who displays all the signs of cortisol deficiency (see Chap. 76 and Chap. 78).
FIGURE 5-4. Simplified feedback loop in the hypothalamic–pituitary–adrenal cortical system. (ACTH, adrenocorticotropin hormone; CRH, corticotropin-releasing hormone; pit., pituitary.) The dotted arrows show the sites of feedback by cortisol.
The feedback effects of cortisol on ACTH release is an example of long-loop feedback: secretion of the peripheral gland affecting the secretion of the pituitary trophic hormone. Long-loop feedback occurs in most of the anterior pituitary hormone systems and is most apparent when the capacity for hormone synthesis or secretion in the peripheral target gland is compromised or abolished. The magnitude of the effects of inhibiting or removing the long-loop feedback signal on circulating concentrations of the appropriate pituitary trophic hormone (i.e., the effects of opening the feedback loop) is large. The dramatic increases in the circulating concentrations of the pituitary trophic hormone that occur under conditions in which there is an abnormally low feedback signal from the target endocrine gland are useful clinically in distinguishing between a primary and a secondary disturbance in an endocrine system. For instance, hypothyroidism could arise from a primary disturbance in the synthesis of thyroxine in the thyroid gland, or it could result from lack of stimulation of the thyroid gland by thyrotropin (thyroid-stimulating hormone [TSH]). In both cases, circulating thyroxine concentrations would be low; if the defect were due to a primary thyroidal disturbance, TSH concentrations would be high, whereas if the defect were due to lack of TSH secretion, circulating concentrations would be low.
Evidence also exists for short-loop inhibition of secretion of hypothalamic-releasing hormones by the trophic hormones of the anterior pituitary. For example, growth hormone secreted by somatotropes in the anterior pituitary stimulates secretion of insulin-like growth factors from the liver and other peripheral tissues (see Chap. 12 and Chap. 173). The somatomedins exhibit long-loop feedback on growth hormone secretion. However, growth hormone has an effect on the hypothalamic-releasing and release-inhibiting hormones that regulate its secretion (Fig. 5-5). There is evidence suggesting that this feedback can occur either by inhibiting the secretion of growth hormone-releasing hormone or by stimulating the secretion of growth hormone-inhibiting hormone (somatostatin) into the hypophysial portal circulation.6,7 There is growing evidence that similar short-loop feedback circuits exist for the other hypophysial hormones. Quantitatively, these appear to be less important than the long-loop feedback, and they may serve as a fine-tuning system within the central nervous system.
FIGURE 5-5. Long- and short-loop feedback. Neurons in the hypothalamus release somatostatin and growth hormone-releasing hormone (GHRH) into the circulation that bathe the growth hormone-secreting cell (somatotrope). Growth hormone stimulates somatomedin C formation in peripheral tissues. Both growth hormone and somatomedins act at the level of the hypothalamus to inhibit growth hormone release.
Ultrashort-loop feedback is suspected to occur when a hormone acts on its own cell type to inhibit further secretion of itself. There is little evidence for this phenomenon; however, it could be of considerable importance in the regulation of secretion of posterior pituitary hormones. In the hypothalamus, collateral axons of oxytocin and vasopressin neurons make contact with other oxytocin and vasopressin neurons.8,9 These cell–cell interactions could help to sharpen and synchronize the secretion of hormone in a given cell type. In the anterior pituitary, vasoactive intestinal peptide (VIP) stimulates secretion of prolactin from lactotropes. However, this effect may be autocrine/paracrine in nature, because lacto-tropes express and secrete both VIP and prolactin into the extra-cellular fluid (Fig. 5-6). Thus, VIP can then stimulate secretion of VIP/prolactin from its own or neighboring lactotropes.10,11 and 12
FIGURE 5-6. Ultrashort-loop feedback. Vasoactive intestinal peptide (VIP) is secreted from lactotropes with prolactin in the anterior pituitary. VIP can stimulate further secretion of VIP/prolactin from the same cell or from neighboring lactotropes (autocrine and paracrine).
FEEDBACK EFFECTS ON AMPLITUDE AND PULSE FREQUENCY: THE GONADOTROPIN SYSTEM
Many, and perhaps most, anterior pituitary hormones are secreted in a pulsatile manner (i.e., surges in secretion occur at regular intervals). Evidence suggests that this arises from synchronous bursts of activity in hypothalamic neurons that secrete the appropriate releasing factors. The gonadal endocrine system has been best studied from this point of view. The gonadotropin-releasing hormone (GnRH) is released in bursts that appear to drive secretion of luteinizing hormone (LH) in episodes (see Chap. 16). In humans and in subhuman primates, GnRH and LH concentrations in hypophysial and systemic blood demonstrate cycles with periods of 2 to 4 hours. LH stimulates testosterone secretion in men and ovulation, luteal formation, and estrogen and progesterone secretion in women.
NEGATIVE FEEDBACK BY SEX STEROIDS
Testosterone, estrogen, and progesterone exert long-loop feed-back on the cycle amplitude and frequency of GnRH and LH. Testosterone and progesterone decrease the frequency and amplitude of the GnRH cycles and thus lower the mean concentration of LH in the circulation. The absence of both hormones leads to acceleration of the cycle frequency and amplitude. The effects of estradiol are complex. Low concentrations of estradiol or high concentrations of estradiol in the presence of progesterone inhibit both the frequency and amplitude of LH secretory pulses, thus decreasing both minimum and maximum concentrations of LH in the circulation.13,13a,14
POSITIVE FEEDBACK BY SEX STEROIDS
Conversely, during the late follicular phase of the menstrual cycle, high concentrations of unopposed estradiol increase pulse frequency and pulse amplitude of LH.13 Thus, a single hormone (estradiol) exerts either negative or positive effects on its trophic hormone, depending on the duration, concentration, and conditions of exposure. The unstable behavior of the gonadal system in this example of positive feedback is characteristic of positive feedback systems. The positive effect of estradiol leads to increasingly frequent and large pulses of LH that in turn lead to increasing secretion of estradiol until the system becomes too unstable to exist, and ovulation occurs. Because the ovum-containing follicle was the primary source of estradiol-secreting granulosa cells, ovulation disrupts the system, and the whole process can begin again (see Chap. 94 and Chap. 95).
PARTURITION AND SUCKLING
The female reproductive system provides two other good examples of positive feedback: parturition- and suckling- induced oxytocin secretion. In both of these examples, there is a neuroendocrine control loop. In the former, stretch receptors in the uterine cervix are stimulated by presentation of the infant’s head; the receptors, in turn, activate neural pathways in the spinal cord and the brainstem that cause secretion of oxytocin from the posterior pituitary. Oxytocin interacts with receptors on the uterine myometrium to cause uterine contraction and further stretch of the uterine cervix. The positive feedback cycle is ended with delivery of the baby and cessation of uterine stretch (see Chap. 25 and Chap. 109). With suckling, stimulation of the nipple activates stretch receptors in the nipples that stimulate spinal cord and brainstem pathways to oxytocin secretion. Oxytocin then causes myoepithelial cell contraction in milk ducts of the breast and milk letdown. The cycle is interrupted when suckling stops (see Chap. 25 and Chap. 106).
All controlled variables of an organism are monitored by receptors and processors that ensure that the variables are maintained within a relatively narrow normal range. When an internal or external stimulus changes or perturbs a regulated variable, the change is sensed and corrected by the nervous and endocrine systems. Some examples of variables that are regulated largely by activity in the autonomic nervous system and by hormones include blood pressure, blood oxygen and carbon dioxide tension, extracellular fluid volume, tissue substrates, metabolites and ions, the composition of filtrate in the kidney tubules, blood flow to different vascular beds, and enzyme production by exocrine glands. All of these variables are regulated by the feedback arrangement described earlier (see Fig. 5-1). Each variable is monitored by a sense organ that is connected to an information processor that controls neural or glandular secretion. The sense organ can be a modified structure that is (a) connected to the central nervous system, (b) connected to the gland, or (c) an integral part of the endocrine effector cell. The effector organ can be an endocrine gland (adrenal cortex, pancreas) or an autonomic postganglionic nerve terminal. When the concentration or level of function of a regulated variable is perturbed, a neural or hormonal response occurs that rectifies the perturbation.
ENDOCRINE CELLS AS RECEPTORS AND EFFECTORS
Some blood-borne agents (nonhormonal) have profound effects on the secretion of certain endocrine gland cells. The action of glucose on the pancreatic B cell was mentioned earlier. The overall effect of glucose-stimulated insulin secretion is to reduce elevated blood glucose levels to the normal range by removing glucose from the blood and facilitating storage in other tissues. Clearly, the action of insulin reduces the stimulus to its secretion. Another example of this kind is the direct effect of potassium on aldosterone secretion from cells of the adrenal zona glomerulosa (see Chap. 79). Small increases in the extra-cellular fluid concentration of potassium increase aldosterone synthesis and secretion. Aldosterone facilitates secretion of potassium by cells of the distal tubule of the kidney, thus reducing the extracellular fluid concentration of this cation.15 This type of feedback regulation also exists for gastrointestinal secretions. The gut hormones are secreted from polarized cells in the intestine that are stimulated by gut contents on the luminal surface and that secrete hormones from the basal surface.
Examples also can be given of direct stimulation of one endocrine gland that causes recruitment of another hormonal system. The chief cells of the parathyroid glands are directly sensitive to the concentration of ionized calcium in the extracellular fluid (Fig. 5-7; see Chap. 51). If the plasma concentration of ionized calcium decreases, the chief cells respond to the decrease in calcium with an increased rate of secretion of parathyroid hormone. Parathyroid hormone facilitates calcium reabsorption by the renal tubules and increases the release of calcium from bone. This hormone also facilitates the formation of 1,25-dihydroxycholecalciferol.16 This vitamin D metabolite further facilitates calcium release from bone and increases intestinal absorption of calcium (see Chap. 54). Although the regulation of the ionized calcium concentration in plasma is more complex than described here (see Chap. 49), this exemplifies how hormonal systems can be recruited for regulation of a variable.
FIGURE 5-7. Feedback regulation of plasma calcium levels. A decrease in extracellular fluid ionized calcium concentration (Ca2+) stimulates the release of parathyroid hormone phosphate acts to increase plasma calcium by facilitating calcium reabsorption from kidney, calcium release from bone, and formation of 1,25-dihydroxycholecalciferol (1,25[OH]2D) from 25-hydroxycholecalciferol (25[OH]D). (It also attenuates formation of the inactive form of the hormone, 24,25-dihydroxy-cholecalciferol [24,25(OH)2D].) 1,25(OH)2D increases calcium release from bone and the intestinal absorption of calcium. The dotted lines show the restitution of plasma calcium concentration.
RECEPTORS IN PROXIMITY TO THE ENDOCRINE GLAND
Endocrine tissue may modify its hormonal secretory rate as a regulated variable changes but may not actually sense the change itself. For example, the macula densa comprises a specialized cell type that senses sodium or chloride transport from the distal tubular filtrate of the kidney (Fig. 5-8). The macula densa is adjacent to the juxtaglomerular cells in the afferent arterioles entering the Bowman capsule. A fall in plasma sodium concentration is reflected in the sodium concentration in the tubular ultrafiltrate. A decrease in sodium delivery to the distal tubule is sensed by the macula densa cells. The macula densa stimulates the juxtaglomerular cells to secrete renin.17 Renin, released into the general circulation, cleaves the terminal 10 amino acids from angiotensinogen, producing angiotensin I. Angiotensin I is converted in the circulation (particularly as it passes through the pulmonary vascular bed) to angiotensin II. Angiotensin II stimulates the secretion of aldosterone from the cells of the adrenal zona glomerulosa, and the effect is the stimulation of sodium reabsorption and potassium and hydrogen excretion in the renal distal tubule.15 The net movement of sodium into the extracellular fluid compartment increases the plasma sodium concentration and relieves the stimulus to renin secretion.
FIGURE 5-8. Feedback regulation of plasma sodium levels and the existence of two feedback loops in the system (dotted lines). As plasma sodium [Na+] falls, the macula densa in the renal distal tubule senses the sodium fall in the filtrate and stimulates renin release from the juxtaglomerular (JG) cells. Renin enzymatically converts angiotensinogen (in the blood) into angiotensin I. Angiotensin I is enzymatically converted to angiotensin II, which stimulates aldosterone release. Aldosterone facilitates sodium reabsorption from the distal tubules, which leads to an increase in plasma sodium. (c.e., converting enzyme; stim., stimulates.)
The regulation of renin secretion, however, involves two feedback loops. Aldosterone facilitates sodium reabsorption and thus removes the initial stimulus to renin secretion; however, angiotensin II also directly inhibits renin secretion from the juxtaglomerular cells (see Fig. 5-8). Again, as with the hypothalamic– pituitary target gland cascades, the hormonal feed-back may serve to inhibit the system more rapidly than would occur with the full restitution of plasma sodium concentration, thus preventing excessive secretion of aldosterone and gaining greater efficiency of the system (see Chap. 79 and Chap. 183).
RECEPTORS THAT CAN INITIATE MULTISYSTEM RESPONSES
Multisystem control involves specialized receptor organs that are directly connected to or embedded in central components of the autonomic nervous system. Organizationally, the autonomic pathways in the medulla, pons, midbrain, and hypothalamus constitute this central system. Afferent input is provided from chemoreceptors, high- and low-pressure baroreceptors, and other peripheral receptors by way of the spinal cord and cranial nerves. Chemosensitive structures exist in the brain and include the circumventricular organs, glucose receptors and osmoreceptors in the hypothalamus, medullary cells that are sensitive to pH, and cells throughout this region of the brain that contain receptors for steroid hormones4,18,19 and insulin.20 The types of response generated by alterations in input from these receptors tend to involve multiple systems (see Chap. 8 and Chap. 9).
The efferent sympathetic and parasympathetic components of the autonomic nervous system innervate all endocrine tissue and can modify the secretion of glands. For example, the well-known cephalic phase of digestion includes vagal stimulation of insulin secretion, which occurs well in advance of substrate (glucose) stimulation of the pancreatic B cell. Insulin secretion, however, is just one of a set of autonomically mediated responses that occurs to the sight, smell, or taste of food. Also included in this phase of digestion is increased salivation, increased gastric motility and acid secretion, and increased gastrin secretion.
Stimulation of both central and peripheral receptors occurs with signals that regulate energy balance. In addition to the neural and hormonal feedback information from the alimentary system that occurs with ingestion of food, occupancy of central receptors for glucocorticoids and insulin appears to be critical for overall regulation of energy balance.20,21 Glucocorti-coids and insulin, apparently through their actions on the central nervous system,20,21 are, respectively, stimulatory and inhibitory to food intake; by contrast, their peripheral effects are, respectively, inhibitory and stimulatory to energy storage.
It appears likely that the reciprocal effects of glucocorticoids and insulin on food intake are mediated, in part, through their opposing actions on the orexigenic peptide that is synthesized in the arcuate nuclei of the hypothalamus, neuropeptide Y.20,21 Insulin inhibits20 and glucocorticoids stimulate21 neuropeptide Y synthesis and secretion. Neuropeptide Y stimulates food intake in satiated animals and, over the long term, causes obesity. Axons containing neuropeptide Y, in turn, innervate nearby hypothalamic cell groups that are known to constitute a neural network that determines food intake.
The metabolic effects of glucocorticoids and insulin on energy balance (see Chap. 72 and Chap. 135) are antagonistic to each other and are opposite in direction to their central, antagonistic effects on energy acquisition. Thus, this bihormonal effector and signaling system serves as a relatively simple overall regulator of energy stores.
Stimulation of peripheral receptors occurs with severe hemorrhage (Fig. 5-9). A sudden decrease in arterial blood pressure and venous return occasioned by a rapid hemorrhage of 20% of the blood volume is registered by the peripheral stretch receptors in the great veins and atria and in the aortic arch and carotid arteries. The information leads to increased autonomic sympathetic activity and the secretion of epinephrine from the adrenal medulla. Secreted catecholamines act to increase heart rate and stroke volume (by way of chronotropic and inotropic actions on cardiac muscle) and to increase peripheral vascular resistance by constriction of vascular smooth muscle22,23 (see Chap. 85).
FIGURE 5-9. Example of complex feedback regulation of blood pressure and blood volume after hemorrhage. (Ang, angiotensin; Symp. Act., sympathetic activity.)
Increased sympathetic outflow to the juxtaglomerular cells of the kidney and decreased renal perfusion pressure or flow cause renin secretion24,25 and, thus, an increase in circulating angiotensin II concentration. Angiotensin II acts directly on arterial smooth muscle to cause vasoconstriction and, therefore, increased arterial blood pressure. Angiotensin II, by way of an action on the subfornical organ, also stimulates the conscious desire to drink fluids and further augment sympathetic activity.19,26 Moreover, angiotensin II acts on the adrenal glomerulosal cells to stimulate the secretion of aldosterone,27 which causes increased reabsorption of sodium by the kidney and aids in the restoration of the extracellular fluid volume.
The hemorrhage-induced alteration in atrial baroreceptor input to the brain also causes stimulation of vasopressin secretion, which causes direct constriction of the vascular smooth muscle and, through its action on kidney, antidiuresis (see Chap. 25 and Chap. 206). All of these hormonal actions contribute to the maintenance of cardiac function and adequate perfusion pressure.28,29 In addition, the behavioral and renal actions of the hormones tend to restore extracellular fluid volume to normal. Clearly, in some situations, if one hormonal system is removed, the other systems “take over” to maintain arterial blood pressure within its normal range. However, all the systems are necessary to combat severe changes in the system. For example, resting arterial pressure is not affected by pharmacologic blockade of the receptors for vasopressin or angiotensin II. However, the administration of these same blocking agents or blockade of the sympathetic nervous system seriously impairs the restitution of arterial pressure after hemorrhage.30
The endocrine responses to hemorrhage enumerated earlier and in Figure 5-9 do not represent a complete list of the hormonal changes elicited by hemorrhage-induced changes in baroreceptor input to hypothalamic control systems. Increased secretion of ACTH, growth hormone, and pancreatic glucagon occurs, and pancreatic insulin secretion decreases. After the surgical removal of baroreceptor input to the central autonomic pathways, hemorrhage no longer elicits either vasopressin or ACTH secretion, strongly suggesting that it is this information that drives the responses.
Feedback information allows the control systems to adjust appropriately to the internal conditions and to diminish over-shoots in endocrine responses that are occasioned by long delays between hormonal secretion and target tissue response.
Houk JC. Control strategies in physiological systems. FASEB J 1988; 2:97.
Guyton AC. Textbook of medical physiology, 9th ed. Philadelphia: WB Saunders, 1996:7.
Engler D, Redei E, Kola I. The corticotropin-release inhibitory factor hypothesis: a review of the evidence for the existence of inhibitory as well as stimulatory hypophysiotropic regulation of adrenocorticotropin secretion and biosynthesis. Endocrine Rev 1999; 20:460.
Keller-Wood ME, Dallman MF. Corticosterone inhibition of ACTH secretion. Endocr Rev 1984; 5:1.
Dallman MF, Akana SF, Levin N, et al. Corticosteroids and the control of function in the hypothalamic-pituitary-adrenal (HPA) axis. Ann NY Acad Sci 1994; 746:22.
Frohman LA, Jansson J. Growth-hormone releasing hormone. Endocr Rev 1986; 7:223.
Muller EE. Neural control of somatotropic function. Physiol Rev 1987; 67:962.
Dreifuss JJ, Kelly JS. Recurrent inhibition of antidromically identified rat supraoptic neurons. J Physiol (Lond) 1972; 200:87.
Moos F, Freund-Merier MJ, Guerne Y, et al. Release of oxytocin and vasopressin by magnocellular nuclei in vitro: specific facilitatory effects of oxytocin on its own release. J Endocrinol 1984; 102:63.
Denef C. Paracrine interactions in the anterior pituitary. Clin Endocrinol Metab 1986; 15:1.
Schwartz J, Cherny R. Intercellular communication within the anterior pituitary influencing the secretion of hypophysial hormones. Endocr Rev 1992; 13:453.
Lambertz SW, Macleod RM. Regulation of prolactin secretion at the level of the lactotroph. Physiol Rev 1990; 70:279.
Plant TM. Gonadal regulation of hypothalamic gonadotropin-releasing hormone release in primates. Endocr Rev 1986; 7:75.
Evans JJ. Modulation of gonadotropin levels by peptides acting at the anterior pituitary gland. Endocr Rev 1999; 20:46.
Vitale ML, Chiocchio SR. Serotonin, a neurotransmitter involved in the regulation of luteinizing hormone release. Endocr Rev 1993; 14:480.
Carey RM, Sen S. Recent progress in the control of aldosterone secretion. Recent Prog Horm Res 1986; 42:251.
MacDonald BR. Parathyroid hormone, prostaglandins and bone resorption. World Rev Nutr Diet 1986; 47:163.
Schnermann J. Juxtaglomerular cell complex and the regulation of renal salt. Am J Physiol 1998; 274:2263.
Dallman MF. Viewing the ventromedial hypothalamus from the adrenal gland. Am J Physiol 1984; 246:R1.
Reid IA. Actions of angiotensin II on the brain: mechanisms and physiologic role. Am J Physiol 1984; 246:F533.
Schwartz MW, Figlewicz DP, Baskin DG, et al. Insulin in the brain: a hormonal regulator of energy balance. Endocr Rev 1992; 13:387.
Dallman MF, Strack AM, Akana SF, et al. Feast and famine: critical role of glucocorticoids with insulin in daily energy flow. Front Neuroendocrinol 1993; 14:303.
Darlington DN, Tehrani MJ. Blood flow, vascular resistance, and blood volume after hemorrhage in conscious adrenalectomized rat. J Appl Physiol 1997; 83:1648.
Darlington DN, Jones RO, Marzella L, Gann DS. Changes in regional vascular resistance and blood volume after hemorrhage in fed and fasted awake rats. J Appl Physiol 1995; 78:2025.
Darlington DN, Keil LC, Dallman MF. Potentiation of hormonal responses to hemorrhage and fasting, but not hypoglycemia in conscious adrenalectomized rats. Endocrinology 1989; 125:1398.
Darlington DN, Chew G, Ha T, et al. Corticosterone, but not glucose, treatment enables fasted adrenalectomized rats to survive moderate hemorrhage. Endocrinology 1990; 127:766.
Thrasher TN, Keenan CR, Ramsay DJ. Cardiovascular afferent signals and drinking in response to hypotension in dogs. Am J Physiol 1999; 277:R795.
Burnay MM, Python CP, Vallotton MB, et al. Role of the capacitative calcium influx in the activation of steroidogenesis by angiotensin-II in adrenal glomerulosa cells. Endocrinology 1994; 135:751.
Dampney RA. Functional organization of central pathways regulating the cardiovascular system. Physiol Rev 1994; 74:323.
Herd JA. Cardiovascular response to stress. Physiol Rev 1991; 71:305.
Rossi NF, Shrier RW. Role of arginine vasopressin in regulation of systemic arterial blood pressure. Annu Rev Med 1986; 37:13.