NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Kopp UC. Neural Control of Renal Function. San Rafael (CA): Morgan & Claypool Life Sciences; 2011.

Cover of Neural Control of Renal Function

Neural Control of Renal Function.

Show details

Chapter 2Neuroanatomy


Nerves carrying fibers that run to or from the kidney are derived from the celiac plexus and its subdivisions, lumbar splanchnic nerves (thoracolumbar paravertebral sympathetic trunk), and the intermesenteric plexus (superior mesenteric ganglion). The celiac plexus consists of the aorticorenal ganglion, the celiac ganglion, and the major splanchnic nerves. There are species differences between the distribution between paravertebral (thoracic, lumbar) sympathetic chain ganglia and prevertebral celiac plexus ganglia. Whereas in the rat and hamster, the majority of the cell bodies of renal postganglionic sympathetic fibers are localized in ipsilateral paravertebral sympathetic chain ganglia, in the cat and monkey, only about half of the cell bodies of the postganglionic sympathetic fibers are localized on the ipsilateral side. The renal sympathetic nerves originate from the intermediolateral column of the spinal cord from T9 to T13. Several nuclei in the brainstem project to intermediolateral column, including the medullary rapae nuclei, rostral ventrolateral medulla, A5 cell group, and the paraventricular hypothalamic nucleus. Importantly, baroreceptor regulation of sympathetic outflow is regulated through rostral ventrolateral medulla, where separate subgroups of neurons preferentially control sympathetic outflow to various organs, including the kidney [58, 94].

Analysis of the extrinsic renal nerve fibers, i.e., those close to the renal hilus, showed that the majority of the renal nerves are unmyelinated, 96% unmyelinated versus 4% myelinated in the rat [58]. The mean diameters of the unmyelinated and myelinated renal nerve fibers are 1.3 and 3.1 μM, respectively. A similar distribution between unmyelinated and myelinated fibers was shown in the mouse [76].


The nerves innervating the kidneys are either efferent or afferent nerves. The efferent nerves derive from the neuraxis, i.e., para- and prevertebral ganglia, and enter the hilus of the kidney along the renal artery and vein. The afferent renal nerves travel from the kidney toward the dorsal root ganglia along the spinal cord. The first part of this book will focus on the efferent renal innervation. The afferent renal innervation of the kidney and their role in the water and salt homeostasis will be discussed toward the end of the book. The efferent renal nerves are postganglionic, and the majority of these are adrenergic, i.e., they contain norepinephrine varicosities at their nerve terminals. An important neurotransmitter role for norepinephrine was supported by the observations that decreasing renal sympathetic nerve activity to zero by chronic renal denervation reduced renal tissue norepinephrine concentration by >95% [16, 64]. Conversely, increasing renal sympathetic nerve activity by renal sympathetic nerve stimulation increased norepinephrine concentration in renal venous blood [27, 135, 193]. There is no evidence for parasympathetic innervation of the kidney.

The efferent sympathetic renal nerves are distributed to all segments of the intrarenal vasculature in the renal cortex and outer medulla, including the interlobar, arcuate, and interlobular arteries and the afferent and efferent glomerular arterioles. There are few sympathetic nerves in the inner medulla. Of the renal vasculature, the highest innervation has been observed along the afferent glomerular arterioles followed by the efferent glomerular arterioles [4, 11] (Figure 2.1).

FIGURE 2.1. Sympathetic nerve fibers (N), identified with an antibody against tyrosine hydroxylase, are located along the afferent and efferent arteriole close to the glomerulus (G).


Sympathetic nerve fibers (N), identified with an antibody against tyrosine hydroxylase, are located along the afferent and efferent arteriole close to the glomerulus (G).

As the afferent arterioles approach the glomerulus, the phenotype of the vascular smooth muscle cells changes to dense cored granules. These are the granular cells of the juxtaglomerular apparatus which contain renin. There is clear evidence for these cells to be innervated by sympathetic nerves [5].

Likewise, all parts of the nephron are innervated by sympathetic nerves [610] (Figure 2.2). Direct renal tubular innervation has been demonstrated in various species, including man. There is a synaptic contact between renal sympathetic nerve fiber varicosities and renal tubular epithelial cell basolateral membranes. Quantitative analysis of the distribution and density of neuroeffector junctions showed that the greatest numbers of neuroeffector junctions are in the proximal tubule followed by the thick ascending limb of Henle's loop, the distal convoluted tubule, and the collecting duct. As to density (i.e., number of innervated tubular profiles/number of tubular profiles examined), this was greatest in the thick ascending limb of Henle's loop, followed by the distal convoluted tubule, the proximal tubule, and the collecting duct [7].

FIGURE 2.2. Sympathetic nerves terminals are adjacent to the basolateral membrane of the tubules.


Sympathetic nerves terminals are adjacent to the basolateral membrane of the tubules. Shown are nerve varicosities (N) containing neurotransmitter vesicles (v) close to a proximal tubule (PT) (left panel) and a cortical collecting duct (CCD) (right panel) (more...)

There is a regional distribution of the renal innervation from outer cortex to inner medulla. The level of efferent renal sympathetic innervation is greatest at the corticomedullary border and decreases in the superficial cortical region and in the deeper regions of the medulla. More recent studies have also identified sympathetic nerves in the renal pelvic wall [149] (Figure 2.3).

FIGURE 2.3. Sympathetic nerve fibers (arrows), identified with an antibody against the norepinephrine transporter (NE-t) are located among the smooth muscle fibers in the renal pelvic wall.


Sympathetic nerve fibers (arrows), identified with an antibody against the norepinephrine transporter (NE-t) are located among the smooth muscle fibers in the renal pelvic wall.

Ultrastructural studies have identified two types of axons innervating vessels and renal tubules of the rabbit and rat kidney [175, 186]. Type I has continuous microtubules through the varicosities, whereas type II does not. The range of the diameter of type I and type II axons does not overlap, which suggests the important hypothesis that they subserve different renal functions. In support of this hypothesis, it has been observed within the rabbit kidney that type I and type II axons are differentially distributed among the various neuroeffectors. The distribution of type I axons is greater on afferent than efferent glomerular arterioles, juxtaglomerular cells, and proximal tubular cells. The density of type II axons on afferent arterioles, being much less than that of type I axons, is similar to that on efferent arterioles, juxtaglomerular cells, and proximal tubular cells. In this context, it is of interest that studies in rats showed that although most effectors, i.e., renal vasculature, juxtaglomerular cells, and nephrons, are innervated by separate nerve fibers, i.e., the renal nerves synapsing with only one of them, there is evidence for the same nerve fibers synapsing with more than one effector [58].


It is important to recognize that most recordings of efferent renal sympathetic nerve activity (ERSNA) are derived from multifiber preparations, which consist of numerous single nerve fibers. Although the firing frequency in each single nerve fiber is quite low, large numbers of fibers may fire at the same time resulting in discharges of summed spikes that can be more easily recorded. One of the most characteristic features of ERSNA is its pulse synchronous discharges mediated to a large extent by activation of the arterial baroreceptors. However, it is important to note that not each cardiac cycle has to be associated with a burst of ERSNA activity.

In anesthetized animals, measurements of multifiber ERSNA is accomplished by recording an amplified signal from an electrode placed around the central portion of a cut renal nerve bundle. In conscious animals, renal nerve activity is recorded from an electrode placed around intact nerve (i.e., not cut). Thus, these recordings represent both ERSNA and afferent renal nerve activity (ARNA). However, in most basal conditions, these recordings most likely represent ERSNA because of the much lower ARNA versus ERSNA in the nonstressed conscious state. For obvious reasons, similar techniques for assessing renal nerve activity cannot be used in humans. Instead, sympathetic nerve activity to various organs, including the kidney, can be estimated by measuring norepinephrine spillover from a specific organ into plasma. During infusion of a small amount of tritiated norepinephrine, norepinephrine concentration is measured in venous outflow from the organ, e.g., the kidney, together with organ plasma flow. Norepinephrine spillover from the kidney is measured by isotope dilution: [(CVCA) + CAE] × renal plasma flow, where CV and CA represent norepinephrine concentration in renal venous and arterial plasma, respectively, and E the fractional extraction of tritiated norepinephrine by the kidney [71]. It is important to note that urinary norepinephrine concentration reflects plasma norepinephrine concentration [135] and is thus not a good estimate of ERSNA.

Studies of the effects of increases in ERSNA on renal function have most commonly applied electrical square-wave signals at different intensities and frequencies via electrodes placed around the peripheral portion of one renal nerve bundle. Numerous studies have shown that this technique results in frequency-dependent changes in renin secretion, urinary sodium excretion, and renal blood (Figure 1.1), resembling those elicited by reflex renal nerve stimulation and thus suggesting that this technique mimics physiological changes in ERSNA. However, analysis of the ERSNA signals would suggest that a more physiological signal would be a diamond-shaped signal [62], with the amplitude of the first and last spikes in each burst being smaller than those in the middle of the burst. Also, these bursts are not of constant frequency and amplitude since ERSNA is under tonic control of the central nervous system. Using digital methods to construct stimulus patterns that reproduced multifiber ERSNA, i.e., the diamond-shaped signal, the changes in renal blood flow and urinary sodium excretion produced by electrical stimulation using the diamond-wave pattern were compared with those produced by the square-wave pattern. The two different signals were matched for total integrated voltage. Interestingly, renal nerve stimulation with the diamond-wave pattern characteristics resulted in greater decreases in renal blood flow and urinary sodium excretion than nerve stimulation with the square-wave characteristics (Figure 2.4). Among the possible contributors to the more enhanced functional responses to the diamond-wave stimulation may be the intermittent character of the diamond-shaped signal. At the same number of pulses/s, intermittent renal nerve stimulation produces greater renal vascular responses compared to continuous stimulation. The longer the rest period between bursts, the greater the renal vascular response. These data suggested that the increased rest time increased the time for possible repletion of the neurotransmitter within the renal sympathetic nerve terminals minimizing possible exhaustion of neurotransmitter release mechanisms. Taken together, these studies suggested that electrical renal nerve stimulation using a more physiological stimulation pattern produces greater effects on both the renal vasculature and tubules than renal nerve stimulation using the more commonly used square-wave pattern.

FIGURE 2.4. Comparing the effects of diamond- vs square-wave stimuli on (A) renal blood flow (RBF) and (B) urinary sodium excretion showed that the decreases in renal blood flow and urinary sodium excretion in response to renal nerve stimulation were greater with the diamond-square pattern than the square-wave pattern.


Comparing the effects of diamond- vs square-wave stimuli on (A) renal blood flow (RBF) and (B) urinary sodium excretion showed that the decreases in renal blood flow and urinary sodium excretion in response to renal nerve stimulation were greater with (more...)


As discussed above, there is strong evidence for the primary neurotransmitter released by the renal sympathetic nerves being norepinephrine. Surgical removal of the renal nerves results in marked reduction of renal tissue norepinephrine content (>90%), and stimulation of the renal nerves increases renal venous outflow of norepinephrine [16, 27, 64, 135]. Norepinephrine exerts various effects on renal function by activation of α- and β-adrenoceptors (vide infra). Norepinephrine is synthesized in neural tissue from the amino acid tyrosine. Tyrosine is converted into L-DOPA (3,4-dihydroxy-L-phenylalanine) by tyrosine hydroxylase. This is the rate-limiting step. L-DOPA is decarboxylated to dopamine by aromatic acid decarboxylase. Dopamine is converted to norepinephrine by dopamine β hydroxylase (Figure 2.5). Norepinephrine is transported into synaptic vesicles in the nerve terminals from which it is eventually released and acts on presynaptic and postsynaptic adrenoceptors.

FIGURE 2.5. Synthesis of norepinephrine in neural cells.


Synthesis of norepinephrine in neural cells.

There is anatomical evidence for the presence of neuropeptide Y (NPY) in renal sympathetic nerves that is co-released with norepinephrine during increases in ERSNA [58]. However, the physiological role for NPY in the control of renal function is still unclear. Although NPY is present in most sympathetic nerves innervating the renal vasculature and renal tubules, the high-intensity renal nerve stimulation required to release NPY raises the question of the physiological significance of NPY in these nerve terminals. Other cofactors released during activation of the renal sympathetic nerves are the purine and pyridine nucleotides, the most significant of which is ATP. Interestingly, in contrast to NPY, studies in isolated rat and mouse kidneys show that ATP contributes to the vasoconstrictor responses to renal sympathetic nerve stimulation at relatively low frequencies [222, 243]. In addition, the presence of purinergic P2 receptors on principal and intercalated cells of the collecting duct close to neural varicosities suggests that ATP may contribute to neurally induced sodium reabsorption [174, 243]. However, the functional importance of these receptors is currently unclear.

Although there is solid evidence for dopamine receptors in renal tissue, there is no evidence for neural release of dopamine, as none of the renal functional responses to renal sympathetic nerve stimulation is affected by dopamine receptor antagonists [58, 262]. The majority of intrarenal dopamine is synthesized from circulating L-DOPA by the proximal convoluted tubule and released locally to exert its actions in a paracrine and autocrine fashion. Basal circulating levels of dopamine are too low to stimulate dopamine receptors on renal vascular tissue. However, there is considerable evidence for circulating dopamine increasing urinary sodium excretion, its natriuretic actions being amplified in conditions of high sodium dietary intake.

Although acetylcholinesterase has been found in renal tissue, there is little evidence for cholinergic nerves in the kidney. The renal functional effects of renal sympathetic nerve stimulation are not affected by cholinergic antagonists. However, the presence of tubular acetylcholinesterase suggests that acetylcholine can be synthesized in the kidney from circulating choline.


As stated above, the primary neurotransmitter in sympathetic nerves is norepinephrine, which acts on the renal vasculature, renal tubules, and the juxtaglomerular cells to cause renal vasoconstriction, sodium reabsorption, and increased renin secretion rate, respectively, by activation of a range of adrenoceptors (Figure 2.6). The adrenoceptors originally identified as α- and β-adrenoceptors have subsequently been further classified in a range of subtypes following numerous pharmacological and molecular biological studies. There are α1- and α2-adrenoceptors and β1-, β2-, and β3-adrenoceptors. In addition, various α1 and α2 subtypes have been identified and are referred to as α1A-, α1B-, and α1D-adrenoceptors and α2A-, α2B-, and α2C-adrenoceptors, respectively. The various adrenoceptor subtypes are G-protein-coupled receptors. The binding of the ligand, norepinephrine, to its receptor elicits a signaling cascade involving various intracellular messengers dependent upon which adrenoceptor subtype is activated.

FIGURE 2.6. The major effects of increases in renal sympathetic nerve activity involve increases in renin secretion rate by activation of β1-adrenoceptors, decreases in urinary sodium excretion by activation of α1B-adrenoceptors, and decreases in renal blood flow by activation of α1A-adrenoceptors.


The major effects of increases in renal sympathetic nerve activity involve increases in renin secretion rate by activation of β1-adrenoceptors, decreases in urinary sodium excretion by activation of α1B-adrenoceptors, and decreases in (more...)

Although with different affinities, all three subtypes of the α1-adrenoceptors are coupled to the Gq/11 protein pathway leading to activation of phospholipase C (PLC), which in turn leads to increased intracellular calcium and activation of protein kinase C (PKC). Activation of PKC may interact with the phospholipase A2/arachidonic acid intracellular cascade leading to formation of eicosanoids [102, 247]. Furthermore, activation of α1-adrenoceptors may also activate the MAP kinase pathway, which over a longer time frame determines the rate of growth and hypertrophy of the vascular smooth muscle cells [248]. The distribution of the various α1 subtype adrenoceptors in the kidney varies. α1A- and α1B-adrenoceptors are equally distributed in the cortex and outer stripe of the medulla, whereas the α1B-adrenoceptor subtype predominates in the inner stripe of the medulla [78]. Most studies would suggest that the renal vasoconstrictor responses to norepinephrine are mainly mediated by activation of α1A-adrenoceptors with activation of α1B-adrenoceptors playing a lesser role. Activation of α1B-adrenoceptors is thought to play a greater role in the effects of norepinephrine on tubular sodium reabsorption. The norepinephrine-mediated activation of α1-adrenoceptors on the vascular smooth muscle cells causes a contraction that increases both afferent and efferent arteriolar resistance, which will reduce renal blood flow and glomerular filtration rate; the magnitude of contraction is dependent on the level of renal sympathetic nerve activation [58]. Concerning the role for α1D-adrenoceptors, studies in bladder tissue may indicate a role for α1D-adrenoceptors in mediating the activation of mechanosensitive nerves [116] in the renal pelvic area (vide infra).

All of the α2-subtype adrenoceptors are coupled to Gi/o proteins [102, 247] leading to inhibition of adenylyl cyclase, the N-type calcium (Ca++) voltage channel and/or activation of the G-protein-coupled inwardly rectifying K channels [22, 209, 242]. Stimulation of α2A- and α2C-adrenoceptors on the presynaptic nerve terminals leads to decreases in norepinephrine release, an import negative feedback mechanism involved in controlling norepinephrine release from sympathetic nerve terminals. Norepinephrine is reported to have higher affinity for α2C-adrenoceptors, which are activated by lower action potential frequencies [102, 209]. Furthermore, there are studies which suggest that activation of α2B-adrenoceptors, located on vascular smooth muscle cells, increases arterial pressure [209].

Regarding the tubular effects produced by increases in ERSNA, there is considerable evidence for α1-adrenoceptors, located on the basolateral side of the tubular cells, regulating sodium and water reabsorption. Most of the studies have focused on the proximal tubules where catecholamines are found to stimulate basolateral membrane Na+/K+-ATPase [2], the primary mechanism pumping sodium out of the cell [87]. The concomitant entry of sodium from the tubular lumen across the apical membrane into the epithelial cells is carried out by the sodium–hydrogen exchanger (NHE3), whose activity is increased by catecholamine stimulation of α1A- and α1B-adrenoceptors [168, 169]. NHE3 embedded in the apical membrane of proximal tubule epithelial cells is a key protein determining the rate of sodium entry into the cell and, hence, overall tubular fluid reabsorption. Acute elevations in renal perfusion pressure cause a relocation of the NHE3 protein to the subapical region, consistent with a reduction in proximal tubular fluid reabsorption. By contrast, challenges that cause an increased renal sympathetic nerve activity are associated with increased insertion of NHE3 in the main areas of the microvilli to enhance fluid reabsorption. As yet, evidence for a direct action of the renal sympathetic nerves on the distribution of NHE3 has not been forthcoming.

Further down the nephron, i.e., in the distal tubule and collecting duct, α2-adrenoceptors are the predominant subtype [206, 208]. As discussed above, activation of these adrenoceptors leads to reduced activation of adenylyl cyclase and thus cAMP generation. The main effects of activation of these receptors appear to be to reduce/buffer the effects of other agents, e.g., the antidiuretic hormone and PGE2 [208], which are known to activate the adenylyl cyclase/cAMP/protein kinase A pathway. Furthermore, there are reports that suggest that activation of α-adrenoceptors, presumably α2-adrenoceptors, decreases Na/K/2Cl protein expression in the thick ascending limb. These studies may explain that the α2-adrenoceptor mediated decreases in Cl flux in this tubular segment. What is currently not known is whether the norepinephrine-mediated decrease in Cl flux is a direct effect or requires the involvement of other factors, including nitric oxide (NO) [210].

β1-adrenoceptors are present in high density on the renin-containing juxtaglomerular cells in the afferent arterioles [58]. Stimulation of these receptors results in increased activity of adenylyl cyclase leading to activation of the cAMP/PKA transduction cascade eventually resulting in increased renin secretion rate, the magnitude of which is dependent on the intensity of ERSNA [120, 133]. β2-adrenoceptors have been localized to the collecting ducts, and they also use cAMP as a second messenger system. Their function at this site is unclear.


Angiotensin (ANG) II plays a significant role in facilitating neurotransmission. Stimulation of ANG II type 1 (AT1) receptors on presynaptic nerve terminals enhances norepinephrine release. This action of ANG II appears to be a generalized phenomenon throughout the central and peripheral nervous systems. ANG II also operates at sympathetic ganglia [1] and neuroeffector junctions [48, 51]. In anesthetized animals, ANG II has a tonic effect on the presynaptic release of norepinephrine. Blocking AT1 receptors with losartan decreases proximal sodium reabsorption. Also, losartan reduces the renal vasoconstrictor responses to increases in ERSNA. However, whether ANG II modulates renal function by stimulating presynaptic receptors in conscious animals or man is unclear.

There is evidence that the presynaptic effects of ANG II may be modified by an interaction with other autocrine and paracrine factors existing in the renal interstitium, including NO and reactive oxygen species, which can act at the neuroeffector junction.

NO is a paracrine factor that modulates neurotransmission at the neuroeffector junction. NO is generated via the action of nitric oxide synthase (NOS) enzymes, which exist in three isoforms, neuronal (nNOS, NOSI), inducible (iNOS, NOSII), and endothelial (eNOS, NOSIII), which are all expressed within the kidney. The eNOS isoform has been reported to be present in the endothelial cells of the renal vasculature and glomerular capillaries [3, 184]. The nNOS isoform has been identified on or close to renal sympathetic and sensory nerves [148, 170], in the renal tubules, and in the macula densa region [253]. The inducible isoform, which is primarily activated following an inflammatory challenge, has been found to be constitutively active in the medullary regions [131]. The site where NO is generated at the neuroeffector junction is unclear, but NO may exert its actions on both neurotransmitter release as well as the postsynaptic membrane to modify the response. In anesthetized dogs, inhibiting NOS enhanced the renal vasoconstrictor and antinatriuretic responses to renal nerve stimulation at low stimulation frequencies, which did not reduce renal blood flow in the absence of NOS blockade [66, 180]. These findings would be consistent with NO inhibiting presynaptic release of norepinephrine. However, this has not been a consistent finding. There are reports in rats suggesting that NO exerts a facilitatory modulation at the neuroeffector junction, i.e., facilitates the presynaptic release of norepinephrine [237]. The understanding of the possible significance of NO at the neuroeffector junction is made more complex by the fact that NO also may have a direct effect on tubular fluid reabsorption. Proximal tubular fluid reabsorption is increased by NO inhibition and depressed following administration of a NO donor [67, 200]. Taken together, it is currently unclear whether NO increases or decreases presynaptic norepinephrine release. It cannot be excluded that the various results reported in the literature are, at least in part, due to the experimental conditions under which the studies were performed.

There have been a number of investigations to determine whether oxidative stress and specifically reactive oxygen species, such as superoxide anions, may affect renal sympathetic nerve activity. It is important to recognize that reactive oxygen species can act at two levels, either directly on the sympathetic nerves at the ganglia or neuroeffector junction, or indirectly by scavenging NO and reducing its bioavailability. Systemic administration of the superoxide dismutase mimetic, tempol, to enhance scavenging of superoxide anions, reduces arterial pressure and renal sympathetic nerve activity, the effects being greater in hypertensive rats [224, 256]. This suggests that in states of oxidative stress, which characterizes hypertension, increased generation of superoxide anions could elevate the level of sympathetic neural outflow to various organs, including the kidney. Shokoji et al. [224] developed a technique whereby drugs could be directly applied to the renal sympathetic nerve fibers proximal to the recording electrodes. They showed that this local administration of tempol reduced renal sympathetic nerve activity, which was also evident if a NOS blocker was coadministered. Conversely, application of an inhibitor of superoxide dismutase, diethyl-thiocarbamate (DETC), increased renal sympathetic nerve activity. A voltage-gated potassium channel blocker completely prevented the DETC-induced increases in ERSNA. These results suggest that reactive oxygen species play a role in the regulation of peripheral sympathetic nerve activity and at least part of this mechanism is mediated through voltage-gated potassium channels. Together, these reports highlight the fact that the level of oxidative stress, and thus the generation of superoxide anions in the environment, may alter the magnitude of renal sympathetic nerve activity. What are not currently known are the functional consequences of these observations to the neural control of renin release, tubular reabsorption, and renal hemodynamics.

Copyright © 2011 by Morgan & Claypool Life Sciences.
Bookshelf ID: NBK57242


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...