THE ANATOMICAL RECORD 219:193-203 (1987) Vasoactive Intestinal Peptide-Immunoreactive Nerves in the Rat Kidney DAVID S. KNIGHT, JOHN A. BEAL, ZHI PING YUAN, AND TIMOTHY S. FOURNET Department of Anatomy, School of Medicine, Louisiana State University, Shreveport, LA 71130 ABSTRACT An indirect immunohistochemical method in which an avidin-biotinylated horseradish peroxidase complex is bound to the secondary antibody was used to visualize vasoactive intestinal peptide-immunoreactive (VIPD nerves in the rat kidney. Rats were perfused with 4% paraformaldehyde or 2% paraformaldehyde + 0.15% picric acid in 0.1 M phosphate buffer, then transferred to the buffer. After 24-48 hours, the kidneys were sectioned with a Vibratome at 200 or 300 pm and incubated in the primary antiserum for 18 hours at room temperature. A sparse plexus of VIPI nerves innervates the rat renal calyx. Some VIPI nerves innervate interlobar arteries and each succeeding segment of the arterial tree including afferent arterioles, but most innervate arcuate and interlobular arteries. VIPI axons do not innervate each arcuate artery or each interlobular branch of an arcuate artery with equal density. Although some axons follow each interlobular branch, most form a dense plexus on only one or two branches. Therefore, most VIPI nerves in the rat kidney innervate a restricted segment of the renal arterial tree. These nerves may be efferent and may selectively dilate arcuate and smaller arteries, or they may be afferent and may sense local changes in mechanical or chemical parameters. The rat kidney is innervated by both sensory and postganglionic autonomic nerves in which several neuroactive substances have been identified. Catecholamine-containing nerves form a uniformly dense plexus on the arteries and also innervate the veins of the rat kidney Gjungqvist and Wagermark, 1970).These nerves are acetylcholinesterase-positive, but there probably are no intrarenal cholinergic nerves (Barajas et al., 1976). Intrarenal nerves immunoreactive for substance P, vasoactive intestinal peptide and calcitonin gene-related peptides have been reported (Barajas et al., 1983; Ferguson and Bell, 1985; Su et al., 1986), but the relative densities of peptidergic nerves on different parts of the arterial tree have not been thoroughly studied. Whether each peptide is synthesized by a separate nerve population or coexists with other peptides or with catecholamine, and whether each is contained in sensory or autonomic nerves are points of interest that require further study. Immunohistochemical methods have been used to visualize vasoactive intestinal peptide (VIP)in both central and peripheral neurons, including those that innervate the heart and blood vessels where it is released by nerve stimulation and probably causes vasodilation (Della et al., 1983; Eklund et al., 1979; Heistad et al., 1980; Jarhult et al., 1980; Shimizu and Taira, 1979). Although several workers failed to detect VIP-immunoreactive (VIPI) nerves in pig, cat, rat, mouse, and guinea pig kidneys (Alm et al., 1980; Larsson et al., 1977; Uddman et al., 1981), Hokfelt et al. (1978) showed that VIPI nerves innervate arteries in the guinea pig renal cortex, and Forssman et al. (1982)found such nerves in kidneys of the guinea pig, rat, cat, dog, pig, and Tupaia and 0 1987 ALAN R. LISS, INC. noted that intrarenal distribution is subject to species variation. Barajas et al. (1983)visualized VIPI terminals in dog and rat kidneys. Most of these terminals innervate large arteries in the inner cortex and juxtamedullary region, but there are, in the dog kidney, some terminals near glomeruli or the origins of arterioles. Della et al. (1983) in a survey of VIPI nerves associated with the guinea pig cardiovascular system found a sparse-to-moderate number of terminals on the renal artery, but few fibers on the renal vein and concluded VIPI terminals are likely to be efferent and to effect vasodilation when stimulated. Renal arterial terminals are of moderate density compared with the most densely innervated arteries supplying the gut, brain, and reproductive organs. In the present study, rat kidneys were serially sectioned and immunolabeled to allow visualization and mapping of intrarenal VIPI nerves. The purpose of this study is to visualize VIPI nerves selectively in thick sections to reveal nerve terminal distribution and to illustrate relative densities on specific parts of the arterial tree. MATERIALS AND METHODS Twelve Wistar rats of either sex (244-410 g) were used in this study. Two fixation protocols were tried. Kidneys were fixed by transcardial perfusion with 200 ml of 0.1 M Sorensen’s buffer + 0.5% procaine followed by 500 ml of either 4% paraformaldehyde or 2% paraformaldehyde 0.15% picric acid in cold Sorensen’s buffer. After removal, kidneys fixed in 4% paraformaldehyde were fixed + Received October 20, 1986; accepted May 8, 1987. Fig. 1. Sparse plexus of VIPI nerves (arrows) on the renal calyx. Renal tubules X256. (RT),renal calyx (RC). Fig. 2. Large perivascular nerve bundle (NB) (Zontaining several VIPI axons (arrows). Arteriole (A). x256. in cold fixative for 2 hours, and those fixed in 2% paraformaldehyde 0.15% picric acid in cold fixative overnight were then transferred to cold Sorensen’s buffer. After 24-48 hours in cold buffer, the kidneys were sectioned with a Vibratome a t 200 or 300 pm and collected in the buffer. Sections of some kidneys were treated with 0.01% sodium borohydride in water (Farr and Nakane, + 1981). All sections were presoaked for 1-2 hours in 0.1 M Sorensen’s buffer containing 0.3%Triton X-100, 5% bovine serum albumin, and 2.5% normal goat serum (NGS) prior to incubation in primary antiserum. The antiserum to vasoactive intestinal peptide (VIP, INC 39H2T) was diluted 1:2,000 in 0.1 M Sorensen’s buffer containing 0.3% Triton X-100 and 1%NGS. Ten milli- VIP-IMMUNOREACTIVE NERVES 4 Fig. 3. VIPI nerves innervating an arteriolar branch of an interlobar artery. Arteriole (A), interlobar renal artery @A). ~ 2 5 6 . Fig. 4. Camera lucida drawing of VIPI nerves innervating intrarenal arteries. Arcuate artery (ARC), interlobular artery (INT), VIPI nerve terminals (T), glomerulus (GI. X42. Inset X3.2. 195 196 D.S. KNIGHT, J.A. BEAL, Z.P. YUAN, AND T.S. FOURNET liters of the diluted antiserum was absorbed with 0.2 mg of rat kidney powder for 30 minutes at room temperature and filtered (0.2 pm Gelman) prior to use. Kidney sections were incubated in the primary antiserum for 18 hours at room temperature, rinsed in buffer, and treated with biotinylated goat anti-rabbit IgG, then the avidinbiotinylated horseradish peroxidase (HRP) complex (Hsu et al., 1981) as prescribed in the Vectastain procedure (Vector Laboratories, Burlingame, CA). Control sections were incubated for the same time period in normal (nonimmune) rabbit serum or in specific primary antiserum preabsorbed with 1 x lOP4M synthetic VIP then processed as above. HRP was visualized by reaction for 10 minutes in 0.05% diaminobenzidine in Sorensen’sbuffer (pH 7.3) containing 0.01% H202. The free-floating sections were then rinsed in buffer, dehydrated in graded ethanols, placed in xylene, and mounted on slides in Permount. A Leitz OrthopladOrthomat microscope and 80A filter were used to photograph specifically stained nerves. A Leitz Dialux microscope equipped with a drawing tube was used to trace the nerves innervating selected arteries. Arterial segments whose nerve terminals were drawn lay wholly within the 300 pm tissue section so that drawings illustrate terminals around the circumference of the vessel. Kidney sections from which these articles were selected were photographed so that the location, orientation, and approximate size of the vessel could be illustrated. RESULTS Incubation in either specific primary antiserum or nonimmune control antiserum results in nonspecific binding of serum components to thick ascending and descending limbs of the loops of Henle. This causes moderate background staining in the outer medullary zone, whereas background staining in the inner medulla and cortex is light. Nevertheless, only neural elements stain intensely after the treatment described above. A sparse plexus of vasoactive intestinal peptide-immunoreactive (VIPI) nerves innervates the rat renal calyx (Fig. 1). Collagenous cords extend from the renal pelvis into the parenchyma, and some VIPI axons extend along and lie parallel with these cords; however, few if any innervate the surrounding tubules. Some VIPI nerves innervate interlobar arteries and each succeeding segment of the arterial tree including afferent arterioles. These nerves do not innervate efferent arterioles, although some juxtamedullary arterioles that divide to form medullary vascular bundles are innervated. There are VIPI nerves associated with most intrarenal arteries. Near the renal hilus, large perivascular nerve bundles contain VIPI axons most of which remain within the nerve bundles throughout this region without ending or giving rise to axon collaterals (Fig. 2). These axons do not form compact VIPI nerve bundles but are dispersed throughout the more numerous axons not reactive for VIP. Most intrarenal VIPI axons are periarterial, but few such axons innervate interlobar arteries; these do not form a regularly spaced pattern. Some axons leave the main nerve bundles to innervate arteriolar branches of interlobar arteries (Fig. 3 ) or to form a sparse, localized plexus on the outer surface of the media of an interlobar artery. Most VIPI axons innervate arcuate and interlobular arteries (Figs. 4-8).Axons on some arcuate arteries are evenly distributed (Figs. 4, 61,whereas those on other such vessels are clustered near the points where smaller arteries branch (Figs. 5, 7). Such terminal clusters extend onto some branches so that middle to distal segments of arcuate arteries and proximal segments of interlobular arteries branching from them have the densest and most complex VIPI plexuses of any intrarenal vessels (Figs. 4-7, 9). These arteries are from 0.1 to 0.3 mm in diameter in the perfusion-fixed rat kidneys examined in this study. VIPI axons do not innervate each arcuate artery or each interlobular branch of an arcuate artery with equal density. Although some axons follow each interlobular branch, most form a dense plexus on only one or two branches (Fig. 7). Two arcuate arteries are illustrated in Figures 5 and 7. Innervation of the two vessels differs in that VIPI axons in Figure 5 are concentrated on the parent arcuate artery, whereas most of those in Figure 7 innervate two interlobular branches whose parent vessel has few axons. The more regular innervation pattern of the arcuate and interlobular arteries illustrated in Figure 4 was less frequently seen. Arterioles also branch directly from arcuate arteries, and many such arterioles have more axons than larger branches nearby (Figs. 6,10, arrows). Axons on sparselyinnervated arteries appear to ramify at branch points (Figs. 4,8, arrows). Distal parts of the arterial tree (afferent arterioles and distal segments and smaller divisions of interlobular arteries) are sparsely innervated by VIPI nerves. Most VIPI axons end along afferent arterioles proximal to the juxtaglomerular apparatus, but some give rise to terminals near the glomerular vascular poles (Figs. 4,7, 11). Few VIPI terminals invade the tubules surrounding small cortical arteries. Although no VIPI axons innervate efferent arterioles, some axons that appear to arise from the nerve plexuses of nearby arcuate arteries enter the outer stripe of the outer zone to innervate arteriolae and venae rectae of some medullary vascular bundles (Fig. 12). DISCUSSION Vasoactive intestinal peptide-immunoreactive nerves innervate the rat renal calyx and intrarenal arterial tree. Dolezel(1975)described the renal connective tissue skeleton, a system of collagenous cords extending from the renal pelvis into the parenchyma, and noted that it is innervated by monoaminergic and acetylcholinesterase-containing nerves. He suggested that some of these terminals are sensory and detect changes in pressure in the renal pelvis or parenchyma. VIPI nerves associated with this system are sparse and are oriented mainly parallel with the parenchymal extensions. VIPI terminals innervate interlobar arteries and succeeding segments of the arterial tree including afferent arterioles. Nerve plexus density on different parts of the arterial tree varies such that most terminals innervate arcuate and interlobular arteries, while arteries proximal and distal to these vessels have fewer terminals. This contrasts with noradrenergic nerves, which, in the adult rat, form a uniformly dense plexus on all intrarenal arteries as far distally as the afferent arterioles with some terminals innervating efferent arterioles and med- VIP-IMMUNOREACTIVE NERVES 197 5 Fig. 5. Camera lucida drawing of VIPI nerves innervating intrarenal arteries. Arrow indicates nerve terminals near points where smaller arteries branch. Arcuate artery (ARC). x42. Inset x3.2. Fig. 6. Camera lucida drawing of VIPI nerves innervating intrarenal arteries. Arcuate artery (ARC), arteriole (A). ~ 4 2Inset . ~3.2. ullary vascular bundles as well (Ljungqvist and Wagermark, 1970; McKenna and Angelakos, 1968; Cowen et al., 1982; Dhall et al., 1986). Although there are some VIPI terminals on most intrarenal arteries, they do not uniformly innervate all vessels of the same size or type. For example, this study shows that VIPI terminals form a moderately dense plexus on the initial segments of some interlobular arteries, but other such vessels are only sparsely innervated, and some are devoid of any such terminals. Some arteriolar branches of interlobar arteries are innervated in preference to the parent artery. Many arteriolar branches of arcuate arteries also 198 D.S. KNIGHT, J.A. BEAL, Z.P. YUAN, AND T.S. FOURNET ARC 8 Fig. 7. Camera lucida drawing of VIPI nerves innervating intrarenal arteries. Arrow indicates nerve terminals near points where smaller arteries branch. Arcuate artery (ARC), interlobular artery (INT), VIPI nerve terminals (T), glomerulus (GI. X42. Inset ~ 3 . 2 . Fig. 8. Camera lucida drawing of VIPI nerves innervating intrarenal arteries. x432. Inset x3.2. VIP-IMMUNOREACTIVE NERVES Fig. 9. Plexus of VIPI nerves innervating an artery in the juxtamedullary renal cortex. Renal artery (RA). x256. Fig. 10. VIPI nerves innervating an arteriole (A) in the juxtamedullary renal cortex. Arcuate artery (ARC), convoluted tubules (CT). ~ 3 0 7 . 199 200 D.S. KNIGHT, J.A. BEAL, Z.P. YUAN, AND T.S. FOURNET Fig. 11. VIPI nerve terminals (arrows) near the vascular pole of a glomerulus (GI. ~ 4 0 0 . Fig. 12. VIPI nerve terminals (arrows) innervating a medullary vascular bundle in the outer stripe of the outer zone. Straight tubules (ST).X256. VIP-IMMUNOREACTIVE NERVES appear to receive more axons than larger branches nearby, but this is less evident because of the generally greater numbers of VIPI axons on the arcuate arteries. VIPI neurons whose terminals innervate the rat kidney may be a population of cells separate from those containing norepinephrine or other neurotransmitters. Neither chemical nor surgical destruction of noradrenergic axon terminals reduces vascular VIPI nerve plexus density in several organs (Costa and Furness, 1982; Della et al., 19831, and renal arterial noradrenergic and VIPI nerves have different growth patterns during development (Della et al., 1983). While there is evidence for coexistence of VIP and acetylcholine in some neurons (Lundberg et al., 1979a,b),many cholinergic nerves do not contain VIP (Lundberg et al., 19821, and it is doubtful whether there are any cholinergic nerves in the rat kidney (Barajas et al., 1976; DiBona, 1982). There are multiple peptides in various combinations in some sensory and autonomic neurons (Dalsgaard et al., 1983; Leah et al., 1985; Tuchscherer and Seybold, 1985). Substance P-immunoreactive nerves innervate the rat kidney, but it seems unlikely that this peptide and VIP coexist in intrarenal nerves, as VIP1 nerve distribution differs from that of substance P-immunoreactive nerves, most of which innervate interlobar vessels in the rat (Ferguson and Bell, 1985). The distribution pattern of calcitonin gene-related peptideimmunoreactive nerves in the urinary system is similar to that of substance P (Su et al., 1986). Most evidence suggests that VIPI terminals originate in autonomic ganglia and that VIP is released by nerve stimulation in several peripheral organs where it relaxes vascular and nonvascular smooth muscle and excites secretory epithelial cells (Dalsgaard et al., 1983; Elfvin, 1980;Fahrenkrug, 1982;Schultzberg et al., 1980). There are VIPI cell bodies in spinal and nodose ganglia and terminals in the superficial dorsal horn so that VIP may be a primary afferent neurotransmitter concentrated in the sacral spinal cord (Loren et al., 1979;Lundberg et al., 1978; Honda et al., 1983; Kawatani et al., 1983; Salt and Hill, 1983). Dun (1983), on the basis of the origin and termination of VIPI nerves in the guinea pig, suggested that VIPI terminals in the inferior mesenteric ganglion are sensory and mediate reflex modulation of ganglionic transmission in response to stimulation of gut mechanoreceptors (Dalsgaard et al., 1983).There are VIPI terminals in the cat carotid body, but whether they are sensory or autonomic is unknown (Lundberg et al., 1979a). Several workers have shown that VIPI nerves are vasodilatory, as VIP is released by nerve stimulation and the exogenous peptide dilates many blood vessels (Fahrenkrug, 1979; Lundberg et al., 1982; Said, 1982). VIP infused into the renal artery increases renin secretion, but the mechanism of action is unknown (Porter and Ganong, 1982; Rosa et al., 1985). It appears to be independent of renal baroreflex or macula densa mechanisms, renal arterial dilation or changes in plasma potassium concentration (Porter and Ganong, 1982;Rosa et al., 1985). VIP appears to act directly on juxtaglomerular cells to increase renin secretion, and the effect may be, in part, prostaglandin-mediated (Calam et al., 1982). VIP influences the hypothalamus and modulates the secretory activity of several endocrine glands, suggesting multiple roles for VIP in hormonal control sys- 201 tems (Hakanson et al., 1982; Polak and Bloom, 1982). Increased solute excretion after VIP infusion into the renal artery in the absence of significant changes in total renal vascular resistance, perfusion flow rate, or glomerular filtration rate suggests a direct effect on renal tubular reabsorption (Rosa et al., 1985).Failure to increase phosphate excretion suggests that the site of action is distal to the proximal tubule, but the small number of vascular bundles in which VIPI terminals were identified in this study would not seem to support a significant role for medullary parts of the nephron in tubular responses to neurally released VIP. On the other hand, there are few VIPI terminals near other segments of the nephron, so that any tubular effects of VIP released by renal nerve stimulation may involve some indirect mechanism. The distribution of VIPI terminals observed in the present work raises two questions. Why do most intrarenal VIPI terminals innervate a restricted region of the arterial tree, and why are some arcuate and interlobular branches relatively well innervated whereas others have few terminals? If these nerves are autonomic, evidence suggests that they release VIP, which relaxes smooth muscle cells or stimulates cell secretion. Intrarenal VIPI terminal distribution shown in this paper would apparently limit the principal effects to some segments of arcuate and interlobular arteries. However, peptides are effective in low concentration, are not recaptured by nerve terminals, and have a long time course of action postsynaptically so that sustained release may increase VIP concentration in renal arterial blood by diffusion through the vascular wall as with other neurally released substances (Bennett, 1972; Bradford, 1986) and may cause effects downstream directly or by modulation of catecholamine release or other mechanisms (Bradford, 1986; Krieger, 1983). Noradrenergic nerves innervate renal arterioles, juxtaglomerular apparati, and some tubules but probably are not involved in the renin response to VIP (Calam et al., 1982). It is evident from the present study and that of Barajas et al. (1983) that VIPI axons are dispersed throughout some large and small intrarenal nerve bundles in which interaction with noradrenergic axons is possible. These or other mechanisms may, in part, regulate arteriolar smooth muscle tone, renin release, and renal tubular transport in response to neurally released VIP, as VIPI terminals do not innervate most arterioles and tubules in the rat kidney. Intrarenal sensory nerves are both mechanoreceptive (Astrom and Craaford, 1967)and chemoreceptive (Recordati et al., 19821, but data on sensory receptor distribution and ultrastructure and neurotransmitter synthesis are minimal (Barajas and Wang, 1978; DiBona, 1982; Kuo et al., 1984). Unilateral stimulation of these receptors may cause contralateral renal vascular constriction, pressor or depressor responses systemically, or a change in nerve activity or catecholamine content in hypothalamic nuclei involved in body sodium and water balance (Aars and Akre, 1970; Calaresu and Ciriello, 1981a,b; Calaresu et al., 1976; Colindres et al., 1980; Katholi et al., 1983;Recordati et al., 1982).Ifintrarenal VIPI nerves are sensory, the partial innervation of arcuate and interlobular arteries in the inner cortex and juxtamedullary zone may suggest that sensory input originating at these strategic points in the arterial tree is adequate to moni- 202 D.S. KNIGHT, J.A. BEAL, Z.P. YUAN, AND T.S. FOURNET tor blood flow, blood pressure, or other parameters involved in effecting these responses. In the rabbit renal artery, nerve growth is continuous from birth to age 6 months after which time a gradual nerve loss occurs even though the proportion of varicosities to nerve fibers increases (Cowen et al., 1982).Innervation density of guinea pig renal arteries gradually decreases postnatally (Gallen et al., 1982). There is no evidence as to whether intrarenal VIPI axon terminals in rats form static plexuses or are involved in the kind of nerve growth or regression that occurs in these species, but either process could explain the nonuniform distribution of axon terminals (Dhall et al., 1986). Most VIPI nerves in the rat kidney innervate a restricted segment of the renal arterial tree. These nerves may be efferent and may selectively dilate arcuate and smaller arteries, or they may be afferent and may sense local changes in mechanical or chemical parameters. ACKNOWLEDGMENT This work was supported by NIH grant HLB34431. LITERATURE CITED Aars, H., and S. Akre (1970) Reflex changes in sympathetic activity and arterial blood pressure evoked by afferent stimulation of the renal nerve. Acta Physiol. Scand., 78:184-188. Alm, P., J. Alumets, R. Hakanson, Ch. Owman, N.-0. Sjoberg, F. Sundler, and B. Walles (1980) Origin and distribution of VIP (vasoactive intestinal polypeptide) nerves in the genito-urinary tract. Cell Tissue Res., 205:337-347. Astrom, A,, and J. Craaford (1967) Afferent activity recorded in the kidney nerves of rats. Acta Physiol. Scand., 7O:lO-15. Barajas, L., and P. Wang (1978) Myelinated nerves of the rat kidney. J. Ultrastruct. Res,. 65:148-162. Barajas, L., P. Wang, and S. DeSantis (1976) Light and electron microscopic localization of acetylcholinesterase activity in the rat renal nerves. Am. J. Anat., 147r219-234. Barajas, L., K.N. Sokolski, and J. Lechago (1983)Vasoactive intestinal polypeptide-immunoreactivenerves in the kidney. Neurosci. Lett., 43r263-260. Bennett, M.R. (1972) Autonomic Neuromuscular Transmission. Cambridge University Press, London. Bradford, H.F. (1986) Chemical Neurobiology. W.H. Freeman, New York. Calam, J., R. Dimaline, W.S. Peart, and R.J. Unwin (1982)Vasoactive intestinal polypeptide (VIP) and renin release in the conscious rabbit. J. Physiol. (Lond.), 329r66p. Calaresu, F.R., and J. Ciriello (1981a) Renal afferent nerves affect discharge rate of medullary and hypothalamic single units in the cat. J.A.N.S., 3:311-320. Calaresu, F.R., and J. Ciriello (1981b)Altered concentration of catecholamines in the hypothalamus of the rat after renal denervation. Can. J. Physiol. Pharmacol., 59:1274-1277. Calaresu, F.R, A. Stella, and A. Zanchetti (1976) Haemodynamic responses and renin release during stimulation of afferent renal nerves in the cat. J. Physiol. (Lond.),255:187-700. Colindres, R.E., W.S. Spielman, N.G. Moss, W.W. Harrington, andC.W. Gottschalk (1980)Functional evidence for renorenal reflexes in the cat. Am. J. Physiol., 239:F265-270. Costa, M., and J.B. Furness (1982) The origins, pathways and termination of nerves with VIP-like immunoreactivity in the guinea-pig small intestine. Neuroscience 8365-676. Cowen, T., A.J. Haven, C. Wen-&in, D.D. Gallen, F. Franc, and G. Burnstock (1982) Development and aging of perivascular adrenergic nerves in the rabbit. A quantitative fluorescence histo-chemical study using image analysis. J.A.N.S., 5317-336. Dalsgaard, C.-J., T. Hokfelt, M. Schultzberg, J.M. Lundberg, L. Terenius, G.J. Dockeray, and M. Goldstein (1983) Origin of peptidecontaining fibers in the inferior mesenteric ganglion of the guineapig: Immunohistochemical studies with antisera to substance P, enkephalin, vasoactive intestinal polypeptide, cholecystokinin and bombesin. Neuroscience, 9r191-211. Della, N.G., R.E. Papka, J.B. Furness, and M. Costa (1983)Vasoactive intestinal peptide-like immunoreactivity in nerves associated with the cardiovascular system of guinea-pigs. Neuroscience, 9t605-619. Dhall, U., T.Cowen, A.J. Haven, and G. Burnstock (1986)Perivascular noradrenergic and peptide-containing nerves show different patterns of changes during development and aging in the guinea pig. J.A.N.S., 16:109-126. DiBona, G.F. (1982) The functions of the renal nerves. Rev. Physiol. Biochem. Pharmacol., 94r75-181. Dolezel, S. (1975) The connective tissue skeleton in the mammalian kidney and its innervation. Acta Anat. (Basel),93:194-209. Dun, N.J. (1983) Peptide hormones and transmission in sympathetic ganglia. In: Autonomic Ganglia. L.-G. Elfvin, ed. John Wiley and Sons, Chichester, pp. 345-366. Eklund, S., M. Jodal, 0. Lundgren, and A. Sjogvist (1979) Effects of vasoactive intestinal polypteptide on blood flow, motility and fluid transport in the gastrointestinal tract of the cat. Acta Physiol. Scand., 105:461-468. Elfvin, L.-G. (1980) Morphological studies on central and peripheral connections of sympathetic ganglia. In: Histochemistry and Cell Biology of Autonomic Neurons, SIF Cells and Paraneurons. 0. Eranko, S. Soninila, and H. Paivarinata, eds. Raven Press, New York, pp. 335-340. Fahrenkrug, J. (1979) Vasoactive intestinal polypeptide: measurement, distribution and putative neurotransmitter function. Digestion, 19:149-169. Fahrenkrug, J. (1982) VIP as a neurotransmitter in the peripheral nervous svstem. In: Vasoactive Intestinal PeDtide: Advances in Peptide HLrmone Research Series. Sami I. Saib, ed. Raven Press, New York, pp 361-372. Farr, A.G., and P.K. Nakane (1981) Immunohistochemistrg with enzyme labeled antibodies: A brief review. J. Immunol: Methods, 47r129-144. Ferguson, M., and C . Bell (1985) Substance P-immunoreactive nerves in the rat kidney. Neurosci. Lett., 60:183-188. Forssman, W.G., D. Hock, and J. Meta (1982) Peptidergic innervation of the kidney. Neurosci. Lett., IO:S183. Gallen, D.D., T. Cowen, S.G. Griffith, A.J. Haven, and G. Burnstock (1982) Functional and nonfunctional nerve-smooth muscle transmission in the renal arteries of the newborn and adult rabbit and guinea pig. Blood Vessels, 19:237-246. Hakanson, R., F. Sundler, and R. Uddman (1982) Distribution and topography of peripheral VIP nerve fibers: Functional implications. In Vasoactive Intestinal Peptide: Advances in Peptide Hormone Research Series. S.I. Said, ed. Raven Press, New York, pp. 121-144. Heistad, D.D., M.L. Marcus, S.I. Said, and P.M. Gross (1980) Effect of acetylcholine and vasoactive intestinal peptide on cerebral blood flow. Am. J. Physiol., 239:H73-H80. Hokfelt, T.,M. Schultzberg, R. Elde, G. Nilsson, L. Terenius, S. Said, and M. Goldstein (1978) Peptide neurons in peripheral tissues including the urinary tract: Immunohistochemical studies. Acta Pharmacol. Toxicol. (Copenh.), 43r79-89. Honda, C.N., M. Rethelyi, and P. Petrusz (1983) Preferential immunohistochemical localization of vasoactive intestinal polypeptide (VIP) in the sacral spinal cord of the cat: Light and electron microscopic observations. J. Neurosci., 3r2183-2196. Hsu, S.-M.,L. Raine, and H. Fanger (1981)Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: A comparison between ABC and unlabeled antibody (PAP) procedures. J. Histochem. Cytochem., 29:577-580. Jarhult, J., P. Helstrang, and F. Sundler (1980) Immunohistochemical localization and vascular effects of vasoactive intestinal polypeptide in skeletal muscle of the cat. Cell Tissue Res., 207:55-64. Katholi, R.E., G.R. Hageman, P.L. Whitlow, and W.T. Woods (1983) Hemodynamic and afferent renal nerve responses to intrarenal adenosine in the dog. Hypertension 5 (Suppl. It149-154. Kawatani, M., I.P. Lowe, I. Nadelhaft, C. Morgan, and W.C. DeGroat (1983) Vasoactive intestinal polypeptide in visceral afferent pathways to the sacral spinal cord of the cat. Neurosci. Lett., 42r311316. Krieger, D.T. (1983) Brain peptides: What, where and why. Science 222975-985. .. ---. Kuo, D.C., J.J. Oravitz, R. Eskay, and W.C. DeGroat (1984) Substance P in renal afferent perikarya identified by retrograde transport of fluorescent dye. Brain Res., 323:168-171. Larsson, L.I., J. Fahrenkrug, and O.B. Schaffalitzky de Muckadell (1977) Vasoactive intestinal polypeptide occurs in nerves of the female genitourinary tract. Science, 197r1374-1375. Leah, J.D., A.A. Cameron, and P.J. Snow (1985) Neuropeptides in physiologically identified mammalian sensory neurones. Neurosci. Lett., 56:257-263. Ljungqvist, A., and J. Wagermark (1970) The adrenergic innervation of intrarenal glomerular and extra glomerular circulatory routes. Nephron, 7:218-229. VIP-IMMUNOREACTIVE NERVES Loren, I., P.C. Emson, J. Fahrenkrug, A. Bjorklund, J. Alumets, R. Hakanson, and F. Sundler (1979) Distribution of vasoactive intestinal polypeptide in the rat and mouse brain. Neuroscience, 4:19531976. Lundberg, J.M., T. Hokfelt, M. Schultzberg, G. Norell, K. Uvnas-Wallensten, L. Terrenius, A. Dahlstrom, J.F. Rehfeld, and S. Said (1978) Pathways of peripheral peptide neurons with reference to the vagus nerve. Neurosci. Lett. [Suppl.], 1:224. Lundberg, J.M., T. Hokfelt, J. Fahrenkrug, G. Nilsson, and L. Terenius (1979a) Peptides in the cat carotid body (glomus caroticum): VIP-, enkephalin-, and substance P-like immunoreactivity. Acta Physiol. Scand., 107:279-281. Lundberc J.M.. T. Hokfelt, M. Schultzberg, K. Uvnas-Wallensten, C. Kohl&, and S.I. Said (1979133 Occurrence of vasoactive intestinal polypeptide (VIP)-like immunoreactivity in certain cholinergic neurons of the cat: Evidence from combined immunohistochemistry and acetylcholinesterase staining. Neuroscience, 4:1539-1559. Lundberg, J.M., A. Anggard, J. Fahrenkrug, 0. Johansson, and T. Hokfelt (1982)Vasoactive intestinal polypeptide in cholinergic neurons of exocrine glands. In Vasoactive Intestinal Peptide: Advances in Peptide Hormone Research Series. S.I. Said, ed. Raven Press, New York, pp. 373-389. McKenna, O.C., and E.T. Angelakos (1968) Adrenergic innervation of the canine kidney. Circ. Res., 22:345-354. Polak, J.M., and S.R. Bloom (1982)Distribution and tissue localization of VIP in the central nervous system and in seven peripheral organs. In: Vasoactive Intestinal Peptide: Advances in Peptide Hormone Research Series. S.I. Said, ed. Raven Press, New York, pp. 107-120. Porter, J.P., and W.F. Ganong (1982) Relation of vasoactive intestinal polypeptide to renin secretion. In: Vasoactive Intestinal Peptide: Advances in Peptide Hormone Research Series. S.I. Said, ed. Raven Press, New York, pp. 285-297. 203 Recordati, G., S.Genovesi, and D. Cerati (1982)Renorenal reflexes in the cat elicited stimulation of renal chemoreceptors. J.A.N.S., 6:127-142. Rosa, R.M., P. Silva, J.S. Stoff, and F.H. Epstein (1985) Effect of vasoactive intestinal peptide on isolated perfused rat kidney. Am. J. Physiol., 249:E494-E497. Said, S.I. (1982) Vasodilator action of V I P introduction and general consideration. In: Vasoactive Intestinal Peptide: Advances in Peptide Hormone Research Series. S.I. Said, ed. Raven Press, New York, pp 145-148. Salt, T.E., and R.G. Hill (1983) Neurotransmitter candidates of somatosensory primary afferent fibres. Neuroscience, 10:1083-1103. Schultzberg, M., T. Hokfelt, and J.M. Lundberg (1980)Peptide neurons in the autonomic nervous system. In: Histochemistry and Cell Biology of Autonomic Neurons, SIF Cells and Paraneurons. 0. Eranko, S. Soninila, and H. Paivarinata, eds. Raven Press, New York, pp. 341-348. Shimizu, T., and N. Taira (1979)Assessment of the effects of vasoactive intestinal peptide (VIP)on blood flow through and salivation of the dog salivary gland in comparison with those of secretin, glucagon and acetylcholine. Br. J. Pharmacol., 65683-687. Sn, H.C., J. Wharton, J.M. Polak, P.K. Mulderry, M.A. Ghatei, S.J. Gibson, G. Terenghi, J.F. B. Morrison, J. Ballesta, and S.R. Bloom (1986) Calcitonin gene-related peptide immunoreactivity in afferent neurons supplying the urinary tract combined retrograde tracing and immunohistochemistry. Neuroscience, 18:727-747. Tuchscherer, M.M., and V.S. Seybold (1985) Immunohistochemical studies of substance P, cholecystokinin-octapeptide and somatostatin in dorsal root ganglia of the rat. Neuroscience, 14:593-605. Uddman. R.. J. Alumets. L. Edvinsson. R. Hakanson. and F. Sundler (1981)VIP nerve fibres around peripheral blood vessels. Acta Physid.Scand., 112:65-70.