Guanylin and uroguanylin are homologous peptides that are structurally similar to the heat-stable enterotoxin (STa) produced by Escherichia coli bacteria (1,2). These peptides bind to an intestinal receptor, guanylate cyclase C (GC-C), and initiate a signal transduction cascade through a guanosine-3′, 5′-cyclic monophosphate (cGMP)-mediated mechanism. This, in turn, results in activation of the cystic fibrosis transmembrane conductance regulator and secretion of Cl− and HCO3− into the intestinal lumen. A number of observations suggest that guanylin and uroguanylin may have physiological functions in balancing the enteral salt intake and the renal salt excretion. For example, STa has been shown to regulate renal sodium tubular transport (3). Intravenous infusion of STa in rats and opossums results in diuresis, natriuresis and kaliuresis, which led some investigators to suggest that uroguanylin might act as part of an intestinal-renal axis of sodium homeostasis (3).
We have also recently shown that salt excretion is impaired in the uroguanylin-deficient mouse challenged with a short-term enteral (but not with an intravenous) salt loading (4). Because the uroguanylin knockout mouse also has diminished expression of guanylin, we could not exclude the possibility that guanylin is also an intestinal natriuretic peptide. Lastly, we and other investigators have suggested the presence of an alternate (non-GC-C) receptor for these ligands, and recent electrophysiological studies have demonstrated a cGMP-independent pathway for uroguanylin in immortalized human kidney epithelial (IHKE-1) cells and cortical-collecting ducts (5,6). It is unknown whether the action of uroguanylin in response to an enteral salt loading is mediated through GC-C.
Thus, to investigate the mechanisms by which these ligands might regulate renal fluid and electrolyte balance, we tested the hypothesis that uroguanylin-deficient (but not guanylin- or GC-C-deficient) mice would demonstrate impaired salt excretion on a high-salt diet.
Metabolic Balance Studies
Adult uroguanylin-, guanylin- and GC-C-deficient mice (4,7,8) and their wild-type littermates were housed individually in metabolic cages where animals had access to deionized water and food ad libitum, as previously described (9). Intake of food and water, along with animal weight, was measured daily. The mice were acclimated to the diet at least 2 days before the metabolic balance study. The mice were fed a control diet containing 1% NaCl (Harlan Teklad, Madison, WI) for the first 3 study days, and then by a 5%-NaCl diet (Harlan Teklad) for the next 3 days (days 4-6). Urine and feces were collected every 24 hours over the 6 consecutive days. Fecal samples were weighed and suspended in 0.75N nitric acid at 4°C. After centrifugation, an aliquot of the supernatant was used to determine Na+ and K+ content using a flame photometer (model 480; Corning Medical and Scientific, Medfield, MA) (9). The urine recovered during each 24-hour period was also aliquoted, and Na+ and K+ content was measured by using flame photometer. All mice in these studies were treated under the protocols approved by the Cincinnati Children's Hospital Medical Center and/or the University of Cincinnati Institutional Animal Care and Use Committee.
Urinary cGMP Studies
Urine was frozen at −70°C and centrifuged before being analyzed for cGMP level measurement using a commercial enzyme immunoassay kit (Assay Designs, Inc, Ann Arbor, MI).
Western Blot Analysis
Blood was obtained by tail bleedings and collected in hematocrit tubes to obtain serum. Mouse serum (volume, 1 μL) was run on a 4% to 12% polyacrylamide gel, transferred to a nylon membrane and probed with prouroguanylin antibody 6910 (gift of Michael Goy, University of North Carolina), as previously described (10). Signal was visualized on a Kodak X-OMAT AR film using a commercially available chemiluminescent kit (NEN Life Science Products, Boston, MA). As a control for loading, the blots were reevaluated with an actin probe (gift of J.L. Lessard, Children's Hospital Research Foundation, Cincinnati, OH).
Enteral Sodium Loading
Uroguanylin knockout and wild-type mice were prepared for short-term enteral salt loading as previously described (4). Briefly, the mice were fasted overnight, anesthetized with ketamine (concentration, 50 μg/g intraperitoneally) and thiobutabarbital (Inactin; RBI, Natick, MA; 100 μg/g intraperitoneally) and placed on a temperature-controlled table. The mice were further prepared for gastric gavage by inserting a polyethylene tubing (PE-50) orally into the esophagus, advancing into the stomach and securing with a 5-0 silk ligature. A 1.0-mL solution of 300 mmol/L NaCl was delivered through the gastric catheter for 2 to 3 minutes. The animals were killed and the kidneys were harvested for microscopy 20 minutes later.
Light and Electron Microscopy
The tissues were harvested and fixed in paraformaldehyde solution for light microscopy. The kidneys submitted for electron microscopy studies were sliced in 1-mm cubes, fixed for 1 hour in 1% osmium tetroxide in 0.1 mol/L cacodylate solution, dehydrated in graded ethanols and embedded in LX 112 plastic. Sections approximately 50-nm thick were cut using a Reichert Ultracut S ultramicrotome and stained with either 1% uranyl acetate for 60 minutes and lead citrate for 3 minutes or with silver stain. The grids were examined at 60 kV in a Zeiss EM 10 electron microscope and photographed using Kodak print film.
Mouse kidneys were fixed in situ by placing the isolated kidney in a small Plexiglas cup and bathing it in PLP fixative (2% paraformaldehyde, 75 mmol/L lysine and 10 mmol/L Na-periodate, pH 7.4) for 20 minutes. The kidneys were then removed and cut in half on a midsagittal plane and postfixed in PLP for another 3 to 4 hours. The fixed tissue was rinsed twice with phosphate-buffered saline solution (PBS), cryoprotected by incubation overnight in 30% sucrose in PBS, embedded in Tissue-Tek OCT compound (Sakura Finetek, Torrance, CA) and frozen in liquid nitrogen. Cryosections (thickness, 5 μm) were cut using a cryomicrotome (Mikron Instruments, San Marcos, CA) and transferred to Fisher Superfrost Plus-charged glass slides and air dried. For immunofluorescence labeling, the sections were rehydrated in PBS for 10 minutes, followed by 10-minute washing with 50 mmol/L NH4Cl in PBS, then with 1% sodium dodecyl sulfate in PBS for 4 minutes for antigen retrieval. Sodium dodecyl sulfate was removed by two 5-minute washes in PBS, and the sections were blocked with 1% bovine serum albumin (BSA) in PBS to reduce background. Dual labeling was performed by incubating with polyclonal antiserum NHE3-C00 at 1:100 dilution, or villin (Immunotech, Chicago, IL) at 1:1000 dilution or horseradish peroxidase (Sigma) at 1:100 dilution in 1% BSA/PBS for 1.5 hours at room temperature. After being washed 3 times for 5 minutes in PBS, the sections were incubated with a mixture of fluorescein isothiocyanate-conjugated goat antirabbit (Cappel Research Products, Durham, NC) and Alexa 568-conjugated goat antimouse (Molecular Probes, Eugene, OR) secondary antibodies at 1:100 dilution in 1% BSA in PBS for 1 hour, washed 3 times with PBS, mounted in Prolong Antifade (Molecular Probes) and dried overnight at room temperature. Slides were viewed with a Nikon PCM quantitative measuring high-performance confocal system equipped with filters for both fluorescein isothiocyanate and TRITC fluorescence attached to a Nikon TE300 Quantum upright microscope. Images were acquired with Simple PCI C-Imaging Hardware and Quantitative Measuring Software and processed with Adobe PhotoDeluxe (Adobe Systems, Mountain View, CA).
Student t test or analysis of variance was used, where appropriate, and repeated-measures analysis of variance with 2 main effects and an interaction term was used to determine if there were differences among experimental groups over time in the total amount of sodium and potassium excreted.
Over a 6-day study period, knockout and littermate control mice for each genotype were given a diet containing 1% NaCl for days 1 to 3, and then a diet containing 5% NaCl for days 4 to 6. To compare the total Na+ excretion in each group, it was necessary to determine their food and water intake during the study period (Fig. 1). As shown in Figure 1A, uroguanylin knockout mice consumed less food and drank less water when administered with the control 1%-NaCl diet. Either as a consequence of or a cause for decreased water intake, uroguanylin knockout mice also produced less urine on the diet containing 1% NaCl. On a 5%-NaCl diet, uroguanylin knockout and wild-type mice both drank significantly more water compared with mice administered with a 1%-NaCl diet. However, food and water consumption, urine volume and fecal weight were all similar between uroguanylin knockout and wild-type mice on this high-salt diet.
As shown in Figure 1B, guanylin knockout and wild-type mice consumed similar amounts of food and water on both the 1%- and 5%-NaCl diets. Guanylin knockout mice produced more urine on the 5%-NaCl diet and more stool on both diets than did the control mice.
As shown in Figure 1C, there were no significant differences in any of the metabolic parameters between knockout and wild-type GC-C mice on either diet.
Na+ and K+ Excretion
Na+ concentration was measured by flame photometry in urine and fecal samples that were collected every 24 hours during the study period. Total Na+ excretion was calculated by combining urinary excretion (approximately 90% of the total) with fecal excretion. As shown in Figure 2A, there was a slight tendency for lower Na+ excretion in the uroguanylin knockout mice compared with wild-type mice, while on the normal-salt diet, perhaps reflecting the slight decrease in food and water intake shown in Figure 1, the difference was significant on day 3 (P < 0.05). After the administration of the high-salt diet, total Na+ excretion increased in both genotypes when NaCl consumption was similar in uroguanylin knockout and wild-type mice. Notably, although the total Na+ excretion in the uroguanylin wild-type mice rose rapidly and achieved maximal levels by the first day, the total Na+ excretion in the uroguanylin knockout mice did not. Total Na+ excretion in uroguanylin knockout animals tended to be less on the first day of the high-salt diet (P = 0.06), was significantly less on the second day of the high-salt diet (P < 0.006) and was similar to that of wild-type animals on the third day of the high-salt diet (Fig. 2A). In contrast, guanylin and GC-C knockout mice did not show any impairment in Na+ excretion compared with their wild-type littermates on either diet (Figs. 2B, C). Unrelated to dietary changes in Na+ and consistent with our hypothesis that uroguanylin is also a kaliuretic agent, there was also a trend (data not shown) for uroguanylin knockout mice to excrete less K+ in both stool (P = 0.06) and urine (P = 0.05).
We wished to determine whether a high-salt diet resulted in increased levels of urinary cGMP, which might indicate activation by uroguanylin of GC-C or another cGMP-producing receptor. As expected, urinary cGMP levels in uroguanylin wild-type mice were higher than in uroguanylin knockout mice, although differences did not achieve statistical significance (Fig. 3). Neither uroguanylin knockout nor wild-type animals demonstrated an increased level of urinary cGMP in response to a high-salt diet (Fig. 3).
Western Blot Determination of Circulating Prouroguanylin
Uroguanylin was measured by Western blot analysis on days 2, 4 and 6 in wild-type animals. As a control, no staining was seen in serum from knockout animals at any time (data not shown). Circulating prouroguanylin levels decreased by 40% on the first day of the high-salt diet (day 4) and returned to baseline by day 6 (Fig. 4).
Light and Electron Microscopy Renal Sections
As part of our evaluation to understand the altered salt handling in uroguanylin knockout mice, we performed light and electron microscopy of renal sections harvested from knockout and wild-type kidneys. Kidney sections from uroguanylin, guanylin and GC-C knockout animals seemed normal by light microscopy. By electron microscopy, an increase in vacuolization in some proximal tubule sections was observed at baseline in the uroguanylin knockout animals, but not the other phenotypes. This vacuolization was more pronounced after 3 days on a high-salt diet (data not shown). To further explore this finding, we next subjected uroguanylin knockout and wild-type mice to a short-term enteral salt loading. After enteral salt loading, there was no change in the light or electron microscopy of guanylin or GC-C knockout or any wild-type mice. However, within 20 to 30 minutes, there was marked vacuolization in the proximal tubules of the uroguanylin knockout mice by light microscopy (Fig. 5) and electron microscopy (Figs. 6A-C). Other cell types in the kidney seemed normal.
Proximal Sodium Hydrogen Exchanger 3 Distribution
To explore potential mechanisms that may contribute to the observed vacuolization after a salt loading, we evaluated the distribution of NHE in proximal tubules by immunofluorescence before and after enteral NaCl administration. In both uroguanylin knockout and wild-type animals under control conditions, NHE3 seems localized largely in the apical microvillar membranes of proximal tubules, with only small amounts appearing at the base of the microvilli or retracted into a submicrovillar pool. After the administration of NaCl in wild-type mice (Fig. 7, left), NHE3 in most proximal tubules seemed to be substantially retracted from the microvilli into a subapical pool, in a response similar to that described previously in rats after short-term exposure to high-perfusion pressure (11). In contrast with uroguanylin knockout mice, the subapical redistribution of NHE3 was less apparent and is absent in some cases, as shown in Figure 7 (right). Whether uroguanylin deficiency leads to derangement of that promote retraction of NHE3 from the apical microvilli awaits a more thorough and quantitative analysis.
There is good evidence that guanylin and uroguanylin have both local intestinal (paracrine) and endocrine functions, forming a potential enteric-renal link to coordinate salt ingestion with natriuresis (for reviews, see Forte et al (3,12-14)). Both peptides, particularly uroguanylin, are postulated to function as intestinally derived natriuretic hormones because both circulate in the bloodstream (15,16) and because uroguanylin and guanylin mRNA are increased by high-salt diets (17,18). However, it is more likely that uroguanylin would serve a role as an intestinal natriuretic factor. In addition to high salt concentration, uroguanylin is upregulated in the circulation of patients with chronic renal failure and congestive heart failure (19-21). Furthermore, the localization of uroguanylin in enterocytes, goblet and enterochromaffin cells of the proximal small intestine is consistent with the luminal and systemic secretion of enteric uroguanylin under the influence of dietary salt (10,20,22,23). In contrast, guanylin is not expressed at equally high levels in the proximal small intestine (23). Moreover, in mammals, the kidney-specific activity of uroguanylin depends on the ninth amino acid residue. Because mammalian guanylin with an aromatic residue at this position is sensitive to a chymotrypsinlike endopeptidase located on the brush border of proximal tubules (12,13,24), filtered guanylin in the urine is quickly metabolized. In contrast, mammalian uroguanylin has asparagine at this position, which confers resistance to this enzyme. Thus, mature uroguanylin is abundantly excreted into the urine and could act on the renal tubules from the luminal side, although the understanding of the site and the mechanism of action of uroguanylin in the kidney is lacking (22,23,25,26).
If uroguanylin and/or related peptides are to serve as postulated intestinal natriuretic peptides, then it should be possible to demonstrate that exogenous administration can induce renal sodium excretion. Indeed, there are reports that guanylin and uroguanylin can initiate diuretic, natriuretic and kaliuretic responses both in vivo and ex vivo in rats and mice (21,27,28), with uroguanylin being substantially more potent than guanylin in eliciting these renal effects. It remains to be determined whether the changes in circulating uroguanylin, within the physiological range, can influence sodium excretion and the mechanism of action. Our use of a uroguanylin knockout mouse is a loss-of-function approach to determining the physiological function of uroguanylin. Finally, although increases in the urinary excretion and renal expression of uroguanylin have been demonstrated in response to altered dietary NaCl intake, changes in plasma concentrations have not been (17,20).
Therefore, we examined the impact of uroguanylin deficiency in mice given a salt challenge. Sodium excretion in wild-type mice achieved maximal levels by the first day of the challenge, whereas total sodium excretion in the uroguanylin knockout mice did not. In contrast, guanylin and GC-C knockout mice did not show any impairment in Na+ excretion when compared with their wild-type littermates. This extends our previous observation (4) in uroguanylin knockout mice by showing that the impaired salt handling is likely due to the deficiency of uroguanylin alone and not due to the loss of guanylin, and is mediated via a novel non-GC-C-mediated mechanism. This is also consistent with the observation that guanylin, uroguanylin and STa, when administered in pharmacological doses, act on the mouse kidney, in part, through a GC-C-independent mechanism (28). The lack of elevation of urinary cGMP after a change from a 1%- to a 5%-salt diet is also consistent with uroguanylin acting on a non-cGMP receptor (5,6), but is in contrast to the observations that urinary cGMP is elevated in rats after a switch from a nearly salt-free diet (salt concentration, 0.08%) to a high-salt diet (salt concentration, 4%). Our experimental approach started with standard rat chow, which contains a higher (1%) salt content and could, in part, explain this difference. However, because the alternate uroguanylin receptor has not yet been identified, the possibility remains that uroguanylin stimulates a non-cGMP-dependent signaling cascade.
Despite the impact of uroguanylin on renal salt handling, we were unable to demonstrate an increase in circulating prouroguanylin after a salt loading. In contrast, Kinoshita et al (20). demonstrated an increase in circulating uroguanylin level after a salt loading, but there was a strong correlation with impaired renal function. In human studies, biologically active uroguanylin accounts for approximately 60% of the circulating uroguanylin, the remainder being the 10-kDa prourouanylin precursor. Our inability to demonstrate an increase in circulating prouroguanylin level in the mouse after a salt loading suggests the possibilities that the levels of circulating uroguanylin are not mirrored by circulating prouroguanylin or that the natriuretic effect is mediated locally by renal uroguanylin, which has been shown to be increased after salt challenge in mouse kidney (17). Others have suggested that the natriuretic factor within prouroguanylin is distinct from uroguanylin itself (Michael Goy, personal communication). Further confirmation of these hypotheses will depend on the development of appropriate immunoassays for biologically active uroguanylin or other circulating natriuretic hormones related to this pathway (29).
Although a role for uroguanylin in an intestinal-renal axis is supported by these findings, interpretation of all of these observations is made difficult by the lack of convincing evidence that the only identified uroguanylin receptor, GC-C, plays a physiological role in the kidney in rodents or humans (30,31). Nevertheless, we sought to explore one potential mechanism by which the loss of uroguanylin might result in impaired salt handling. We have demonstrated a distinct light and electron microscopic phenotype in the proximal convoluted tubule of uroguanylin knockout mice: formation of distinct cytoplasmic inclusions or vacuoles in the knockout mice after enteral salt loading. The lack of a membrane limiting these vacuoles and the absence of lipid or other storage products suggest that they are compartments filled with excess water and solute. This intoxication of the proximal convoluted tubule is further evidence of the involvement of uroguanylin in a novel signal transduction pathway in the kidney but does not directly demonstrate the mechanism for impaired salt handling. Our observation that there may be impaired redistribution of NHE3 in the proximal convoluted tubule of uroguanylin knockout mice suggests one possible mechanism for impaired salt handling. Specifically, the normal redistribution of NHE3 in response to a salt loading might explain the reduced tubular reabsorption of Na+ or the diminished Na+/H+ exchange observed after salt challenge. The lack of redistribution of NHE3 in the uroguanylin knockout mouse could similarly explain the elevated tubular reabsorption of Na+ (greater Na+/H+ exchange) in this model after a salt loading (4).
In conclusion, there is growing evidence that uroguanylin plays a significant role in the maintenance of fluid and electrolyte homeostasis via actions in the kidney. This has broad implications for the role of uroguanylin, first, as a physiological modulator of natriuresis and, second, as a potential therapeutic modality for restoration of appropriate salt handling.
1. Currie MG, Fok KF, Kato J, et al. Guanylin: an endogenous activator of intestinal guanylate cyclase. Proc Natl Acad Sci U S A
2. Hamra FK, Forte LR, Eber SL, et al. Uroguanylin
: structure and activity of a second endogenous peptide that stimulates intestinal guanylate cyclase. Proc Natl Acad Sci U S A
3. Forte LR, Fan X, Hamra FK. Salt and water homeostasis: uroguanylin
is a circulating peptide hormone with natriuretic activity. Am J Kidney Dis
4. Lorenz JN, Nieman M, Sabo J, et al. Uroguanylin
knockout mice have increased blood pressure and impaired natriuretic response to enteral NaCl load. J Clin Invest
5. Sindic A, Basoglu C, Cerci A, et al. Guanylin, uroguanylin
, and heat-stable enterotoxin activate guanylate cyclase C and/or a pertussis toxin-sensitive G protein in human proximal tubule cells. J Biol Chem
6. Sindic A, Hirsch JR, Velic A, et al. Guanylin and uroguanylin
regulate electrolyte transport in isolated human cortical collecting ducts. Kidney Int
7. Steinbrecher KA, Wowk SA, Rudolph JA, et al. Targeted inactivation of the mouse guanylin gene results in altered dynamics of colonic epithelial proliferation. Am J Pathol
8. Mann EA, Jump ML, Wu J, et al. Mice lacking the guanylyl cyclase C receptor are resistant to STa-induced intestinal secretion. Biochem Biophys Res Commun
9. Meneton P, Schultheis PJ, Greeb J, et al. Increased sensitivity to K+
deprivation in colonic H,K-ATPase-deficient mice. J Clin Invest
10. Steinbrecher KA, Mann EA, Giannella RA, et al. Increases in guanylin and uroguanylin
in a mouse model of osmotic diarrhea are guanylate cyclase C-independent. Gastroenterology
11. Yang LE, Maunsbach AB, Leong PK, et al. Differential traffic of proximal tubule Na+
transporters during hypertension or PTH: NHE3
to base of microvilli vs. NaPi2 to endosomes. Am J Physiol Renal Physiol
12. Forte LR, London RM, Freeman RH, et al. Guanylin peptides: renal actions mediated by cyclic GMP. Am J Physiol Renal Physiol
13. Forte LR, London RM, Krause WJ, et al. Mechanisms of guanylin action via cyclic GMP in the kidney. Annu Rev Physiol
14. Forte LR. A novel role for uroguanylin
in the regulation of sodium balance. J Clin Invest
15. Hess R, Kuhn M, Schulz-Knappe P, et al. GCAP-II: isolation and characterization of the circulating form of human uroguanylin
. FEBS Lett
16. Kuhn M, Raida M, Adermann K, et al. The circulating bioactive form of human guanylin is a high molecular weight peptide (10.3 kDa). FEBS Lett
17. Potthast R, Ehler E, Scheving LA, et al. High salt intake increases uroguanylin
expression in mouse kidney. Endocrinology
18. Carrithers SL, Jackson BA, Cai WY, et al. Site-specific effects of dietary salt intake on guanylin and uroguanylin
mRNA expression in rat intestine. Regul Pept
19. Kinoshita H, Nakazato M, Yamaguchi H, et al. Increased plasma guanylin levels in patients with impaired renal function. Clin Nephrol
20. Kinoshita H, Fujimoto S, Nakazato M, et al. Urine and plasma levels of uroguanylin
and its molecular forms in renal diseases. Kidney Int
21. Greenberg RN, Hill M, Crytzer J, et al. Comparison of effects of uroguanylin
, guanylin, and Escherichia coli
heat-stable enterotoxin STa in mouse intestine and kidney: evidence that uroguanylin
is an intestinal natriuretic hormone. J Investig Med
22. Nakazato M, Yamaguchi H, Date Y, et al. Tissue distribution, cellular source, and structural analysis of rat immunoreactive uroguanylin
23. Whitaker TL, Witte DP, Scott MC, et al. Uroguanylin
and guanylin: distinct but overlapping patterns of messenger RNA expression in mouse intestine. Gastroenterology
24. Hamra FK, Fan X, Krause WJ, et al. Prouroguanylin and proguanylin: purification from colon, structure, and modulation of bioactivity by proteases. Endocrinology
25. Nakazato M, Yamaguchi H, Kinoshita H, et al. Identification of biologically active and inactive human uroguanylins in plasma and urine and their increases in renal insufficiency. Biochem Biophys Res Commun
26. Fujimoto S, Kinoshita H, Hara S, et al. Immunohistochemical localization of uroguanylin
in the human kidney. Nephron
27. Fonteles MC, Greenberg RN, Monteiro HS, et al. Natriuretic and kaliuretic activities of guanylin and uroguanylin
in the isolated perfused rat kidney. Am J Physiol
28. Carrithers SL, Ott CE, Hill MJ, et al. Guanylin and uroguanylin
in mice lacking guanylyl cyclase-C receptor. Kidney Int
29. Fukae H, Kinoshita H, Fujimoto S, et al. Changes in urinary levels and renal expression of uroguanylin
on low or high salt diets in rats. Nephron
30. Laney DW Jr, Mann EA, Dellon SC, et al. Novel sites for expression of an Escherichia coli
heat-stable enterotoxin receptor in the developing rat. Am J Physiol
31. Schulz S, Green CK, Yuen PS, et al. Guanylyl cyclase is a heat-stable enterotoxin receptor. Cell