Medical Physiology, 3rd Edition

Intermediate- and Long-Term Control of the Circulation

In addition to the rapidly acting neural mechanisms that control the total peripheral resistance and cardiac output, humoral controls contribute to the homeostasis of the circulation. In most instances, these control systems operate on a time scale of hours or days, far more slowly than the neurotransmitter-mediated reflex control by the CNS.

Two classes of humoral controls influence the circulation:

1. Vasoactive substances released in the blood, or in the proximity of vascular smooth muscle, modulate the vasomotor tone of arteries and veins, affecting blood pressure and the distribution of blood flow.

2. Nonvasoactive substances, which act on targets other than the cardiovascular system, control the effective circulating volume by modulating ECF volume. By determining the filling of the blood vessels, these nonvasoactive agents also modulate the mean arterial pressure and cardiac output.

Endocrine and paracrine vasoactive compounds control the circulatory system on an intermediate- to long-term basis

Vasoactive substances, both endocrine and paracrine, cause blood vessels to contract or to relax (Table 23-3). In many instances, paracrine control dominates over endocrine control. The chemical messengers controlling the blood vessels can be amines, peptides, or proteins; derivatives of arachidonic acid; or gases such as NO.

TABLE 23-3

Vasoactive Compounds



Epinephrine (via α1 receptors)

Epinephrine (via β2 receptors)











Biogenic Amines

Monoamines may be either vasoconstrictors (epinephrine and serotonin) or vasodilators (histamine).

1. Epinephrine. The source of this hormone is the adrenal medulla (see pp. 1030–1033). Epinephrine binds to α1 receptors on VSMCs with less affinity than norepinephrine, causing vasoconstriction (see Table 20-7), and with high affinity to β2 receptors on VSMCs, causing vasodilation (see Table 20-7). Because β2 receptors are largely confined to the blood vessels of skeletal muscle, the heart, the liver, and the adrenal medulla itself, epinephrine is not a systemic vasodilator. Epinephrine also binds to β1 receptors in the heart, thereby increasing the heart rate and contractility (see p. 542). For the cardiovascular system, the effects of catecholamines originating from the adrenal medulla are usually minor compared with those of the norepinephrine released from the sympathetic nerve endings.

2. Serotonin. Also known as 5-hydroxytryptamine (5-HT; see Fig. 13-8B), this monoamine is synthesized by serotonergic nerves, enterochromaffin cells, and adrenal chromaffin cells. 5-HT is also present in platelets and mast cells. Serotonin binds to 5-HT2A and 5-HT2B receptors on VSMCs, causing vasoconstriction (see Table 20-8). Circulating serotonin is generally not involved in normal systemic control of the circulation but rather in local control. Serotonin is particularly important in vessel damage, where it contributes to hemostasis (see p. 439).

3. Histamine. Like serotonin, histamine (see Fig. 13-8B) may also be present in nerve terminals. In addition, mast cells release histamine in response to tissue injury and inflammation. Histamine binds to H2receptors on VSMCs, causing vasodilation (see Table 20-8). Although histamine causes vascular smooth muscle to relax, it causes visceral smooth muscle (e.g., bronchial smooth muscle in asthma) to contract.


Vasoactive peptides may be either vasoconstrictors or vasodilators (see Table 23-3).

1. Angiotensin II (ANG II). Part of the renin-angiotensin-aldosterone cascade (see pp. 841–842), ANG II, as its name implies, is a powerful vasoconstrictor. The liver secretes angiotensinogen into the blood. The enzyme renin, released into the blood by the kidney, then converts angiotensinogen to the decapeptide ANG I. Finally, angiotensin-converting enzyme (ACE), which is present primarily on endothelial cells, particularly those of the lung, cleaves ANG I to the octapeptide ANG II. Aminopeptidases further cleave it to the heptapeptide ANG III, which is somewhat less vasoactive than ANG II. imageN23-12


Metabolism of the Angiotensins

Contributed by Emile Boulpaep, Walter Boron

The liver synthesizes and releases into the blood the α2-globulin angiotensinogen (Agt), which is a plasma glycoprotein that consists of 452 amino acids. Its molecular weight ranges from 52 to 60 kDa, depending on the degree of glycosylation. Angiotensinogen belongs to the serpin (serine protease inhibitor) superfamily of proteins, which also includes antithrombin III (see p. 446 as well as Tables 18-4 and 18-5). The liver contains only small stores of angiotensinogen, which it constitutively secretes. Production by the liver is greatly increased during the acute-phase response (see Box 18-1). Angiotensinogen is synthesized in several tissues other than liver. In addition to the 52- to 60-kDa form of angiotensinogen, a high-molecular-weight (HMW) angiotensinogen complex of 450 to 500 kDa is also present in plasma. Polymorphisms within the angiotensinogen gene may contribute to normal variations in arterial blood pressure and a tendency to develop hypertension.

The juxtaglomerular cells of the kidney—also called granular cells (see p. 727)—are specialized smooth-muscle cells of the afferent arteriole that synthesize and release both the glycoprotein renin (molecular weight 37 to 40 kDa)—pronounced ree-nin—and its inactive precursor prorenin, which is the major circulating form. Prorenin-activating enzymes on endothelial cells convert this prorenin to renin. The kidney is the major source of circulating prorenin/renin, and the liver is responsible for removing renin from the circulation. The half-life of renin in plasma is 10 to 20 minutes. Renin is an aspartyl proteinase that cleaves a leucine-valine bond near the amino (N) terminus of angiotensinogen to release a decapeptide called angiotensin I (ANG I), which is not biologically active. By an alternative pathway, nonrenin proteases can also produce ANG I. The full sequence of ANG I is Asp-Arg-Val-Tyr-Ile-His-Pro-Phe-His-Leu:


(NE 23-5)

Angiotensin-converting enzyme (ACE; molecular weight ~200 kDa) is produced by and attached to endothelial cells. ACE is a dipeptidyl carboxypeptidase (a zinc peptidase) that cleaves ANG I by removing the carboxy-terminal (C-terminal) dipeptide histidine-leucine and producing the octapeptide angiotensin II (ANG II). The ACE cleavage site is a phenylalanine-histidine bond. The sequence of ANG II is Asp-Arg-Val-Tyr-Ile-His-Pro-Phe:


(NE 23-6)

ANG II has a half-life in blood of 1 to 3 minutes, indicating that a large fraction is removed in a single pass through the circulation. ANG II acts on G protein–coupled receptors known as AT1 and AT2. In addition, two other receptors are less well characterized: AT3 and AT4. ANG IV (see below) can bind to AT4 receptors.

By an alternative pathway, non-ACE proteases can also convert ANG I to ANG II. Conversely, note that ANG I is not a specific substrate for ACE, which can cleave other peptides, including bradykinin (see pp. 553–554), enkephalins, and substance P.

Aminopeptidase A (also called angiotensinase A, or glutamyl aminopeptidase) further cleaves the aspartate-arginine bond on ANG II to produce the heptapeptide angiotensin III (ANG III, also called ANG-(2-8)), which has the sequence Arg-Val-Tyr-Ile-His-Pro-Phe. Ang III, like ANG II, can also bind to AT receptors.

Aminopeptidase B (also called angiotensinase B, or arginyl aminopeptidase) finally cleaves an arginine-valine bond on ANG III to produce the hexapeptide angiotensin IV (ANG IV, also called ANG-(3-8)). The sequence of ANG IV is Val-Tyr-Ile-His-Pro-Phe. This metabolite is inactive.

Finally, another ANG metabolite has received attention in recent years. ANG-(1-7) consists only of the first seven amino acids of ANG I*Asp-Arg-Val-Tyr-Ile-His-Pro. This heptapeptide can arise from ANG I by at least three routes:


(NE 23-7)

Note that, by the above nomenclature, ANG I is ANG-(1-10), and ANG II is ANG-(1-8). The first and third pathways involve a new enzyme called ACE2. The second is catalyzed by any in a family of enzymes called neutral endopeptidases (NEPs). ANG-(1-7) can bind to a G protein–coupled receptor called the Mas receptor and—when acting on the cardiovascular system—can elicit effects opposite those of ANG II.


Donoghue M, Hsieh F, Baronas E, et al. A novel angiotensin-converting enzyme-related carboxypeptidase (ACE2) converts angiotensin I to angiotensin 1-9. Circ Res. 2000;87:E1–E9.

Ferrario CM, Chappell MC. Novel angiotensin peptides. Cell Mol Life Sci. 2004;61:2720–2727.

Gurley SB, Allred A, Le TH, et al. Altered blood pressure responses and normal cardiac phenotype in ACE2-null mice. J Clin Invest. 2006;116:2218–2225.

Yagil Y, Yagil C. Hypothesis: ACE2 modulates blood pressure in the mammalian organism. Hypertension. 2003;41:871–873.

*ANG-(1-7) is not to be confused with another heptapeptide metabolite of ANG II, namely, ANG III—also known as ANG-(2-8). ANG III has actions similar to those of ANG II, but is weaker (see p. 553).

In VSMCs, ANG II binds to G protein–coupled AT1A receptors, activating phospholipase C, raising [Ca2+]i, and leading to vasoconstriction (see Table 20-8). However, ANG II is normally not present in plasma concentrations high enough to produce systemic vasoconstriction. In contrast, ANG II plays a major role in cardiovascular control during blood loss (see p. 585), exercise, and similar circumstances that reduce renal blood flow. Reduced perfusion pressure in the kidney causes the release of renin (see p. 841). Plasma ANG II levels rise, leading to an intense vasoconstriction in the splanchnic and renal circulations. The resulting reduced renal blood flow leads to even more renin release and higher ANG II levels, a dangerous positive-feedback system that can lead to acute renal failure. A widely studied model of hypertension that demonstrates the importance of this mechanism is the Goldblatt model for hypertension (Box 23-1).

Box 23-1


Hypertension is found in ~20% of the adult population. It can damage endothelial cells, producing a number of proliferative responses, including arteriosclerosis. In the long term, hypertension can lead to coronary artery disease, myocardial infarction, heart failure, stroke, and renal failure. In the great majority of cases, hypertension is the result of dysfunction of the mechanisms used by the circulation for the long-term rather than short-term control of arterial pressure. In fact, chronically hypertensive patients may have diminished sensitivity of their arterial baroreceptors.

Most people with an elevated blood pressure have “primary hypertension,” in which it is not possible to identify a single, specific cause. Renal artery stenosis, which compromises renal blood flow, is the most common cause of secondary hypertension. An experimental equivalent of renal artery stenosis is the “one-clip two-kidney” model of hypertension first described by Goldblatt (see Box 40-2). This model does not explain most cases of hypertension, but it does give us our best description of the pathophysiological mechanism involved in at least some patients with elevated blood pressure. The most common cause of renal artery stenosis is the narrowing of the renal artery by atherosclerotic plaque. Fibromuscular disease of the renal arterial wall can also be responsible, usually in young women, as can any space-occupying lesion (e.g., metastatic cancer or benign cysts). If the stenosis is removed by angioplasty or surgery, and if preliminary test results show that the stenosis is the likely cause of the elevation in blood pressure, then a significant percentage of such patients will experience resolution of their hypertension.

The cumulative obstruction of smaller arteries and arterioles may also produce hypertension, as is often seen in diseases of the renal parenchyma or any end-stage renal disease. Conversely, constriction of larger vessels proximal to the kidneys can also cause hypertension, as is the case with coarctation of the aorta, a congenital malformation that constricts flow through the aorta to the lower parts of the body. Another cause of secondary hypertension is chronic volume overload. Volume overload can be acquired, such as in primary aldosteronism (caused by either a benign adenoma or bilateral hyperplasia) and pheochromocytoma (a tumor of the adrenal medulla that releases excessive amounts of catecholamines into the circulation). Volume overload can be genetic, as in rare mendelian forms of hypertension such as Liddle disease imageN23-14 and pseudohypoaldosteronism type 2.


Liddle Disease

Contributed by Emile Boulpaep, Walter Boron

Liddle disease is caused by a gain-of-function mutation in the β or γ subunit of the epithelial Na+ channel ENaC. For example, the critical regions in the β subunit are critical for endocytosis or proteasomal degradation of the channel. Thus, mutations result in an overabundance of ENaC at the apical membrane, resulting in excessive Na+ reabsorption, an increase in effective circulating volume, and hypertension. The disease is readily diagnosed by determining whether the hypertension is reversed by the drug amiloride, which antagonizes ENaC. Indeed, amiloride is an effective therapy for Liddle disease.

For a description of ENaC, see Table 6-2, family No. 14.


Hansson JH, Nelson-Williams C, Suzuki H, et al. Hypertension caused by a truncated epithelial sodium channel gamma subunit: Genetic heterogeneity of Liddle syndrome. Nat Genet. 1995;11:76–82.

Liddle GW, Bledsoe T, Coppage WS. A familial renal disorder simulating primary aldosteronism but with negligible aldosterone secretion. Trans Assoc Am Physicians. 1963;76:199–213.

Schild L, Canessa CM, Shimkets RA, et al. A mutation in the epithelial sodium channel causing Liddle disease increases channel activity in the Xenopus laevis oocyte expression system. Proc Natl Acad Sci U S A. 1995;92:5699–5703.

Shimkets RA, Warnock DG, Bositis CM, et al. Liddle's syndrome: Heritable human hypertension caused by mutations in the beta subunit of the epithelial sodium channel. Cell. 1994;79:407–414.

ANG II has a range of other effects—besides direct vasoactive effects—that indirectly increase mean arterial pressure: (1) ANG II increases cardiac contractility. (2) It reduces renal plasma flow, thereby enhancing Na+ reabsorption in the kidney. (3) As discussed in the next section, ANG II and ANG III also stimulate the adrenal cortex to release aldosterone. (4) In the CNS, ANG II stimulates thirst and leads to the release of another vasoconstrictor, AVP. (5) ANG II facilitates the release of norepinephrine by postganglionic sympathetic nerve terminals. (6) Finally, ANG II also acts as a cardiac growth factor (see Box 22-3).

2. Arginine vasopressin (AVP). The posterior pituitary releases AVP, also known as antidiuretic hormone. AVP binds to V1A receptors on VSMCs, causing vasoconstriction (see Table 20-8), but only at concentrations higher than those that are strongly antidiuretic (see pp. 817–819). Hemorrhagic shock causes enhanced AVP release and a vasoconstriction that contributes to a transient restoration of arterial pressure (see pp. 584–585).

3. Endothelins (ETs). Endothelial cells produce ETs (see pp. 461–463) that bind to ETA receptors on VSMCs, causing vasoconstriction (see Table 20-8). Although, on a molar basis, ETs are the most powerful vasoconstrictors, it is not clear whether these paracrine agents play a dominant role in overall blood pressure homeostasis.

4. Atrial natriuretic peptide (ANP or ANF). Released from atrial myocytes in response to stretch (see p. 547), this 28–amino-acid peptide binds to ANP receptor A (NPR1) on VSMCs, which is membrane-bound guanylyl cyclase, causing vasodilation (see Table 20-8). Although ANP lowers blood pressure, its role in the overall regulation of mean arterial pressure is doubtful. Because ANP also has powerful diuretic and natriuretic actions, it ultimately reduces plasma volume and therefore blood pressure. imageN23-13


Crosstalk Between Atrial Natriuretic Peptide and Endothelin

Contributed by Emile Boulpaep

Atrial Natriuretic Peptide (ANP) is also involved in an intriguing feedback loop involving endothelin. ANP stimulates endothelin formation by endothelial cells, but endothelin is itself a secretagogue for the atrial myocytes, causing them to release ANP. Thus, a vasodilator (ANP) promotes the release of a vasoconstrictor (ET), which in turn promotes the release of the original vasodilator.

5. Kinins. At least three different kinins exist: (1) the nonapeptide bradykinin, which is formed in plasma; (2) the decapeptide lysyl-bradykinin, which is liberated from tissues; and (3) methionyl-lysyl-bradykinin, which is present in the urine. These kinins are produced by the breakdown of kininogens, catalyzed by kallikreins—enzymes that are present in plasma and in tissues such as the salivary glands, pancreas, sweat glands, intestine, and kidney. Kallikreins form in the blood from the following cascade: Plasmin acts on clotting factor XII, releasing fragments with proteolytic activity. These factor XII fragments convert an inactive precursor, prekallikrein, to kallikreins. The kinins formed by the action of these kallikreins are eliminated by the kininases (kininase I and II). Kininase II is the same as ACE. Thus, the same enzyme (ACE) that generates a vasoconstrictor (ANG II) also disposes of vasodilators (bradykinin). Bradykinin binds to B2 receptors on endothelial cells, causing release of NO and prostaglandins and thereby vasodilation (see Table 20-8). Like histamine, the kinins relax vascular smooth muscle but contract visceral smooth muscle.


Many tissues synthesize prostaglandins, derivatives of arachidonic acid (see pp. 61–62). Prostacyclin (prostaglandin I2 [PGI2]) binds to prostanoid IP receptors on VSMCs, causing strong vasodilation (see Table 20-8). PGE2 binds to prostanoid EP2 and EP4 receptors on VSMCs, also causing vasodilation (see Table 20-8). It is doubtful that prostaglandins play a role in systemic vascular control. In veins and also in some arteries, arachidonic acid or Ca2+ ionophores cause endothelium-dependent contractions. Because cyclooxygenase inhibitors prevent this vasoconstrictor response, venous endothelial cells probably metabolize arachidonic acid into a vasoconstrictive cyclooxygenase product, presumably thromboxane A2.

Nitric Oxide

Nitric oxide synthase (NOS) produces NO from arginine in endothelial cells (see p. 480). NO activates the soluble guanylyl cyclase in VSMCs, causing vasodilation (see Table 20-8). Although NO is a powerful paracrine vasodilator, it is not clear that it plays an important role in overall blood pressure homeostasis.

Pathways for the renal control of ECF volume are the primary long-term regulators of mean arterial pressure

The volume of the ECF includes both the blood plasma and the interstitial fluid. The small solutes in the plasma and interstitial fluid exchange freely across the capillary wall, so that the entire ECF constitutes a single osmotic compartment. Because the plasma volume is a more or less constant fraction (~20%) of the ECF volume, changes in the ECF volume produce proportional changes in plasma volume. Thus, assuming that the compliance of the vasculature is constant (see Equation 19-5), such an increase in plasma volume will lead to an increase in transmural blood pressure.

We saw in Chapter 20 that the Starling forces across a capillary (i.e., the hydrostatic and colloid osmotic pressure differences) determine the traffic of fluid between the plasma and the interstitial fluid (see p. 471). Thus, alterations in the Starling forces acting across the capillary wall can affect the plasma volume and therefore blood pressure.

Because of the importance of the ECF volume and Starling forces in determining the plasma volume, one might expect that the body would have specific sensors for ECF volume, interstitial fluid volume, and blood volume. However, the parameter that the body controls in the intermediate- and long-term regulation of the mean arterial pressure is none of these but rather a more vague parameter termed the effective circulating volume. The effective circulating volume is not an anatomical volume but the functional blood volume that reflects the extent of tissue perfusion, as sensed by the fullness or pressure in the vessels. The control mechanisms that defend effective circulating volume include the two classes of stretch receptors described in this chapter. First, the high-pressure receptors in the carotid sinus and aorta (see pp. 534–536) do double duty. In the short term, these baroreceptors regulate blood pressure by their direct cardiovascular effects, as was already discussed. In the longer term, they regulate effective circulating volume. Second, the low-pressure receptors—located in the pulmonary artery, the junction of the atria with their corresponding veins, the atria themselves, and the ventricles (see pp. 546–547)—regulate effective circulating volume by direct and indirect effects on the cardiovascular system. In addition to these already familiar receptors involved in neural control of the circulation, other sensors monitor effective circulating volume (see Table 40-2): the baroreceptors in the renal artery, the stretch receptors in the liver, and the atrial myocytes themselves as well as—to some extent—osmoreceptors in the CNS.

As we will see in Chapter 40, these sensors of effective circulating volume send signals to the dominant effector organ—the kidney—to change the rate of Na+ excretion in the urine. These signals to the kidney follow four parallel effector pathways (see p. 834): (1) the renin–ANG II–aldosterone axis, (2) the autonomic nervous system, (3) the posterior pituitary that releases AVP, and (4) the atrial myocytes that release ANP. Of these four parallel pathways, the most important is the renin-angiotensin-aldosterone system. By regulating total body Na+ content, the kidney determines ECF volume. Therefore, the kidney ultimately governs the blood volume and is thus the principal agent in the long-term control of mean arterial pressure.