The Role of Uroguanylin in Regulation of Ion Transport in Salivary Glands

Abstract

Objectives

Guanylin peptides are considered to be the only intrinsic regulators of salivary glands secretion. Therefore, the aim of this study was to determine the effects of systemic uroguanylin (UGN) of the salivary flow and ion composition. Besides, the objective was to investigate whether those effects include activation of guanylate cyclase C (GC-C).

Material and Methods

This study was conducted on 7 months old C57Bl6NCrl (wild type, WT) and GC-C knockout (KO) mice. Salivary flow rate and ion composition were determined after pilocarpine stimulation with UGN (30 µg/animal) or saline i.p. application. The expression of mRNA for AQPs, NHEs, NBCn1, Slc26a3/a6 and CFTR were determined by qPCR in submandibular salivary glands.

Results

When applied i.p., UGN decreased the pilocarpine stimulated saliva flow rate and increased the concentration of Na⁺, H⁺ and Cl^-. In GC-C KO mice, UGN showed no effect on saliva flow rate, while the concentrations of Na⁺, H⁺ and Cl^- are the same in GC-C KO littermates when compared to WT mice. UGN increased expression of Slc26a6 while in GC-C KO mice Slc26a6 had a higher expression when compared to WT mice, suggesting involvement of GC-C independent signalling pathway for UGN. The difference in Slc26a6 in GC-C KO mice is not unique for salivary glands because it was also found in duodenum and kidney cortex.

Conclusions

The effects of UGN via basolateral membrane of salivary glands cells have not been considered up to date. In our study, UGN, when applied i.p., decreased salivary flow rate, pH, and changed the composition of other ions. Therefore, plasma UGN, an hour after a meal, could have physiological and pathological importance (development of cavities, inflammations or demineralizations), and the inhibition of systemic UGN effects could be considered a new approach in treatment of those conditions.

Introduction

Uroguanylin (UGN) is a member of the guanylin peptides family which belongs to the family of naiuretic peptides. UGN is a small peptide composed of 16 amino acids, and it has two disulfide bonds (1). The gene for UGN (GUCA2B) is located at the 1^st chromosome in humans and 4^th chromosome in mice (2). UGN is expressed in the intestine and, after a meal, it is secreted in gut lumen and blood (3). In the intestine lumen, UGN binds to guanylate cyclase C (GC-C) receptor located at the luminal membrane of enterocytes, which leads to an increase in intracellular cGMP concentration (4, 5). Finally, UGN stimulates Cl^- and HCO₃^- secretion via the Cystic fibrosis transmembrane conductance regulator (CFTR) and Slc26a3/a6 and inhibition of Na⁺ reabsorption by Na⁺/H⁺ exchanger (4-10).

UGN and its GC-C dependent signalling pathway are also expressed in the brain, kidneys, heart, pancreas, lungs, reproductive system, spleen, lymph nodes, lungs and airways as well as in salivary glands (parotid and submandibular glands) (6-8, 11-21).

The existence of guanylin peptides and the GC-C signalling pathway in salivary glands has been known for more than two decades. Kulaksiz et al. showed the expression of guanylin peptides, GC-C, cGMP dependent protein kinase II, and CFTR in human parotid and submandibular glands. Due to the importance of CFTR in salivary glands physiology, detection of differences in saliva composition can serve as diagnostic tools for cystic fibrosis. Saliva as diagnostic tool can be also used in various systemic diseases (Sjogren's syndrome, cardiovascular diseases, diabetes or idiopathic infertility) (22, 23). GC-C is located at the apical membrane of the ducts of salivary glands. Its agonists, guanylin and UGN, are found in the cells of the intralobular and interlobular ducts in small vesicles at the apical part of the secretory epithelial cells (12, 24). The physiological function of guanylin peptides and regulation of their expression is unknown.

In addition to the GC-C dependent signalling pathway, guanylin peptides activate yet another signalling pathway which is located at the basolateral membrane of the target epithelial cells allowing hormones from the blood to exert their function (25). The existence of this, cGMP independent signalling pathway, has been found in the kidneys, intestine, and brain, Activation of this signalling pathway leads not only to an increase in intracellular Ca²⁺ concentration but also a decrease in intracellular cAMP concentration (26-31).

Since guanylin peptides are considered to be potential intrinsic regulators of salivary glands secretion (12), the aim of this study was to determine the effects of systemic UGN of the salivary flow and ion composition. Also, the aim of the present study was to investigate if the UGN functions via GC-C or other signalling pathway by performing experiments on GC-C KO mice.

Material and methods

Ethical approval: Experimental procedures used in this study were approved by the National Ethical Committee Ministry of Agriculture (UP/I-322-01/22-01/36) and the School of Medicine University of Zagreb (641-01/22-02/01). All experiments were performed in accordance with the ARRIVE guidelines. Our research adhered to the ethical guidelines of the Ethical Codex of Croatian Society for Laboratory Animal Science. Special efforts were made to minimize animal suffering and reduce the number of animals used.

Animals: Experiments were performed on the C57Bl6NCrl (wild type, WT) male mice and littermates derived from GC-C knock out mice (GC-C KO) 28.7 ± 1.3 weeks old. GC-C KO mice were a donation from Dr. K. A. Steinbrecher (Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, USA) and were crossbred with the C57Bl6NCrl strain. The obtained heterozygotes were crossbred with each other, after which their offspring were genotyped and knockout (GC-C KO) and wild type (GC-C WT) littermates were used in this study. The animals were housed in our animal facilities (HR-POK-006) under controlled environmental conditions and under the 12-hour light-dark cycle. All animals were fed ad libitum with standard rodent chow and had free access to water.

Sialometry: Before sialometry (measurement of stimulated saliva flow rate), mice were fasted for 2 hours, with free access to water to exclude saliva production due to feeding. Animals were first anesthetized with an i.p. injection of ketamine and xylazine (80-100 mg/kg, 5-10 mg/kg, respectively) (IACUC Guidelines: Anesthesia) and immobilized on a heated pad to prevent anesthesia-related hypothermia. After induction of anesthesia, to experimental group of WT mice (n = 8) human UGN (30 µg/animal in 250 μl of saline) (PeptaNova GmbH, Sandhausen, Germany) were administered i.p., while the control group of WT mice (n = 8) received i.p. 250 μl of saline (0.9% NaCl; Croatian Institute for Transfusion Medicine, Zagreb, Croatia). Both GC-C WT (n = 7) and GC-C KO (n = 6) mice were administered the same amount of UGN as the experimental group of WT mice. After fifteen minutes, to stimulate salivation, the cholinergic agonist, pilocarpine hydrochloride (0.001mg/g; Fagron Hrvatska d.o.o., Donja Zelina, Croatia), was administered i.p. (32).

After fifteen minutes, the animals were positioned so that the animal's head was tilted lower than the tail to reduce the possibility of deglutition and aspiration of saliva. Saliva was collected with filter paper (Medical Intertrade, Sveta Nedelja, Croatia) cut into strips positioned behind the upper and lower incisors. Saliva was collected for 15 min, after which the strips were inserted in test tubes (ThermoFisher Scientific, Waltham, Massachusetts, USA) and centrifuged (15 000 g, 24°C). Supernatant was collected and the amount of saliva was determined gravimetrically (Kern & Sohn GmbH, Balingen, Germany). The collected samples were stored at -20°C for further analysis. The saliva flow rate was calculated by g of animal weight in 15 minutes.

Still in anesthesia, both experimental groups of WT mice were sacrificed by cervical dislocation 1h after UGN or saline application while GC-C WT and GC-C KO animals 3h after UGN application to determine the effect of the GC-C independent signalling pathway. Submandibular salivary glands, duodenum and kidneys were isolated. The kidney cortex was carefully dissected and the tissue was flash frozen in liquid nitrogen and stored at -80°C until use.

pH and ion measurements in saliva: pH was measured potentiometrically using dedicated pH-sensors on the blood gas analyzed GEM Premier 4000 (Instrumentation Laboratory, Bedfor, USA). Concentrations of Na⁺, K⁺ and Cl^- were determined by indirect potentiometry using original manufacturer's reagents and protocols on the fully-automated analyzer Alinity c (Abbott Laboratories, Chicago, USA).

Expression of ion transporters and channels for water mRNA: Total RNA was isolated in the following manner. Tissue samples (100 mg) were homogenized in 1 mL Trizol solution (ThermoFisher Scientific) with an ultrasonicator (QSonica CL188, Newtown, Connecticut, USA). After addition of 0.2 mL of chloroform, (Kemika, Zagreb. Croatia), the tissue samples were incubated for 5 min at room temperature and centrifuged (15 min, +4°C, 12 000 g). Supernatant containing RNA was transferred to a new test tube and 0.5 mL of isopropyl alcohol (Carl Roth, Karlsruhe, Germany) was added. After centrifugation (15 min, +4°C, 12 000 g), the supernatant was removed and the precipitate was washed with 1 mL of ethanol (75%; Claro-Prom, Zagreb, Croatia). The samples were briefly centrifuged (5 min, +4°C, 8 000 g), ethanol was removed and the samples were left to air-dry. Isolated RNA was dissolved in sterile dH₂O and quantified with the NanoDrop ND-1000 spectrophotometer (Marshall Scientific LLC, Hampton, New Hampshire, USA).

RNA (1 µg) was transcribed into complementary DNA (cDNA) using the GoScriptTM Reverse Transcription System (Madison, Wisconsin, USA) in the following way: 5 µL RNA (1 µg), Oligo(dt)primers (1 µL) and H₂O were incubated at 70°C for 5 min and subsequently cooled to +4°C. Afterwards, GoScript reaction buffer (4 µL), MgCl₂ (2 µL), nucleotide mixture (1 µL), ribonuclease inhibitors (0.5 µL), GoScript reverse transcriptase (1 µL) and H₂O (6.5 µL) were added. The reaction mixture was then incubated at 25°C (5 min), 42°C (60 min) and 70°C (15 min).

TaqMan Real-Time PCR Assay (ThermoFisher Scientific) was used for quantitative expression of ion transporters and channels for water. Probes specific to aquaporins 1 (AQP1; Mm00431834_m1), 3 (AQP3; Mm01208559_m1) and 5 (AQP5; Mm00437578_m1), sodium-hydrogen exchanger 1 (NHE1; Mm00444270_m1) and 3 (NHE3; Mm01352473_m1), electroneutral sodium-bicarbonate cotransporter 1 (NBCn1; Mm01310972_m1), solute carrier family 26 members 3 (Slc26a3; Mm00445313_m1) solute carrier family 26 members 6 (Slc26a6; Mm00506742_m1), and cystic fibrosis transmembrane conductance regulator (CFTR; Mm00445197_m1) were used to quantify gene expression. As a housekeeping gene, we used beta-actin (ActB; Mm00607939_s1). A reaction mixture, containing TaqMan Master Mix (10 µL), labelled primers (1 µL), cDNA (1 µL) and H₂O (8 µL) was added to a 96-well plate, briefly centrifuged (5 min, 250 g) and placed in the 7500 Real-Time PCR System (ThermoFisher Scientific). The obtained results were normalized according to the tissue expression of ActB.

Statistical analysis: The Kolmogorov-Smirnov test was used to test the normal distribution. Student’s unpaired t tests were used with each effect compared with its own control. The data were presented as mean ± SEM. p < 0.05 was considered statistically significant. A correlation was calculated using the Pearson’s correlation test. The GraphPad Instat statistical software (GraphPad Software, Boston, MA, USA) was used for statistical analyses.

Results

Uroguanylin decreases the pilocarpine stimulated saliva flow rate and changes saliva ion composition

The ion content of the saliva of seven months old WT mice, upon pilocarpine stimulation is: pH = 7.94 ± 0.03; [Na⁺] = 48 ± 5 mmol/L; [K⁺] = 38 ± 4 mmol/L; [Cl^-] = 86 ± 3 mmol/L (n = 6 - 8).

UGN was applied i.p. and the pilocarpine stimulated saliva production and saliva composition was determined 1 h after application. UGN inhibits salivary production in WT mice (WT Control: 8.9 ± 0.4; WT UGN: 7.5 ± 0.3 mL/g body weight/15 min, n = 8, p = 0.022). In animals which do not have GC-C, this effect was abolished (GC-C WT UGN: 7.6 ± 0.5; GC-C KO UGN: 9.2 ± 0.3 mL/g body weight/15 min, n = 6, p = 0.023) (Figure 1A). UGN also decreased a pH (WT Control: 7.94 ± 0.03; WT UGN: 7.68 ± 0.06, n = 6 and 7 respectfully, p = 0.004, with an increase in H⁺ concentration from 11.5 to 20.1 nM/L), but this effect is still present in GC-C KO mice (GC-C WT: 7.67 ± 0.07; GC-C KO: 7.75 ± 0.06 n = 7 and 6 respectfully, p = 0.419), suggesting involvement of GC-C independent signalling pathway.

Figure 1.

Uroguanylin decreased pilocarpine stimulated saliva flow rate and pH via different signalling pathways. A: Effects of uroguanylin (UGN) on saliva production was not present in guanylate cyclase C (GC-C) knockout mice missing GC-C receptor (GC-C KO). B: UGN decreased pH, but this effect is still present in GC-C KO mice. The results are presented as mean ± SEM, n = 6-8. *p < 0.05 statistically significant when compared to WT control; **p < 0.05 statistically significant when compared to GC-C WT which are siblings (littermates) of GC-C KO mice but still have GC-C receptor. WT – wild type mice

UGN also changed saliva ion composition by increasing Na⁺ and Cl^- concentration (Na⁺: WT Control: 48 ± 5; WT UGN: 60 ± 3 mmol/L, n = 8, p = 0.046; Cl^-: WT Control: 86 ± 3 WT UGN: 95 ± 2 mmol/L, n = 7 and 8 respectively, p = 0.030). Those effects are still present in GC-C KO mice (Na⁺: GC-C WT: 64 ± 6; GC-C KO: 60 ± 3 n = 7 and 6 respectfully, p = 0.562; Cl^-: GC-C WT: 99 ± 5; GC-C KO: 94 ± 3 n = 7 and 6 respectfully, p = 0.483). UGN had no effect on the saliva concentration of K⁺ (Figure 2).

Figure 2.

Uroguanylin increased Na⁺ and Cl^- concentration via GC-C independent signalling pathway. Uroguanylin (UGN) increased concentration of Na⁺ and Cl^- and that effect is still present in GC-C KO mice. UGN did not change K⁺ concentration. The results are presented as mean ± SEM, n = 6-8. *p < 0.05 statistically significant when compared to WT control; WT – wild type mice

Positive correlation of the salivary flow vs pH and K⁺ concentrations in UGN stimulated mice

Passing through ducts of salivary glands, Na⁺ and Cl^– are absorbed and K⁺ and HCO₃^- are secreted. In the human parotid gland, the concentration of Na⁺ and Cl^– is in positive correlation, and K⁺ is in negative correlation to saliva flow rate (33). In this study, after pilocarpine stimulation of salivary production in 7 months old WT control mice, there was no statistically significant correlation of ion concentration with salivary flow (Table 1).

Table 1. Positive correlation of the salivary flow and pH and K⁺ concentrations in UGN stimulated mice.

Ion vs flow		pH	Na+	K+	Cl-
WT Control	r = p =	0.7921 0.0603	-0.0522 0.9021	0.2047 0.6268	0.1296 0.7817
WT UGN	r = p =	0.5320 0.0412	0.4845 0.0672	0.5582 0.0306	0.4931 0.0618
GC-C KO UGN	r = p =	0.4584 0.3605	-0.2265 0.6659	0.0824 0.8766	-0.4250 0.4008

GC-C – guanylate cyclase C, GC-C KO – mice missing GC-C, UGN - uroguanylin,

WT – wild type of mice

Upon UGN stimulation, pH and K⁺ concentration in pilocarpine stimulated saliva were in positive correlation with salivary flow (Table 1, p < 0.05 statistically significant correlation). The changes in pH could occur due to changes in the function of H⁺ or HCO₃^- transporters.

Uroguanylin increased expression of Slc26a6 in submandibular salivary glands via GC-C independent signaling pathway

In our study, the effects of the peptide hormone UGN on expression of mRNA for channels for water (AQP 1, 3 and 5), sodium-hydrogen exchanger isoform 1 (NHE1) and 3 (NHE3), sodium-bicarbonate cotransporter (NBCn1), members of solute carrier family 26 (a3 and a6) and cystic fibrosis transmembrane conductance regulator (CFTR) were determined. The expression of AQP3 and Slc26a3 was not found in mouse submandibular salivary glands.

Of all tested AQPs and ion transporters, one hour after i.p. application of UGN expression of Slc26a6 increased two times (p = 0.027). Similarly, the Slc26a6 expression was higher in GC-C KO mice compared to their GC-C WT littermates (p = 0.024) (Figure 2).

It is not surprising that, although the UGN application did not change NHE3 expression in submandibular salivary glands (p = 0.416), GC-C KO mice had two-time increase in expression of NHE3 compared with GC-C WT animals (p = 0.016).

GC-C regulates expression of Slc26a6 in salivary glands, kidney cortex and duodenum

To determine if effects of UGN are unique for salivary glands, we investigated differences in expression of Slc26a6 in duodenum and kidney cortex of GC-C KO mice, where, in physiological conditions, the GC-C dependent signaling pathway for UGN exists. Similar to submandibular salivary gland, the expression of Slc26a6 was higher in duodenum and kidney cortex of GC-C KO mice than in GC-C WT mice (p = 0.022 and p = 0.038, respectively, Figure 4). Another Cl-/HCO3- exchanger, Slc26a3, exists in duodenum, and its expression is also higher in animals missing GC-C (Figure 4).

Figure 4.

Expression of members of Slc26 family increased in submandibular salivary gland, duodenum and kidney cortex of animals missing GC-C. *p < 0.05 statistically significant when compared to GC-C WT mice (n = 6-10). The results are presented as mean ± SEM. GC-C KO - guanylate cyclase C (GC-C) knockout mice missing GC-C receptor

Discussion

The effect of the natriuretic peptides on salivary glands, including guanylin peptides and other naturistic peptides such as the atrial natriuretic peptide (ANP), as an agonist of guanylate cyclase A (GC-A), has been known for a while (34). After parasympathetic stimulation, in submandibular glands, ANP increases salivary secretion, decreases Na⁺ and increases K^+. However, in parotid glands, ANP increases salivary flow rate and Na⁺ and K⁺ concentration (35). Therefore, the aim of this study was to determine the effects of GC-C agonist, UGN, on saliva flow rate and ion composition.

In this study saliva flow rate upon pilocarpine stimulation and ion composition corresponded to previously published results (36, 37). It is a well-known fact that a decrease in saliva flow rate leads to a decrease in Na⁺ and Cl^- concentration, pH and an increase in K⁺ concentration. UGN decreased saliva flow rate and pH. It also increased the saliva concentration of Na⁺ and Cl^-, which could not be explained by changes in saliva flow rate.

The effects of the UGN on flow rate is GC-C dependent since the UGN effect is gone in GC-C KO mice compared to their GC-C WT littermates which is not the case for the changes of pH, Na⁺ and Cl^- concentration (Figure 1 and 2). Furthermore, upon UGN stimulation both pH and K⁺ concentrations were in positive correlation to saliva flow rate. Correlation was not found in GC-C KO mice suggesting the involvement of GC-C signaling pathway (Table 1).

To determine the possible mechanism of UGN action in salivary glands, the expressions of ion transporters and AQPs were determined in submandibular glands of WT and GC-C KO mice. In our study mRNA for AQP3 was not found although it has been previously shown (38). The expression of mRNA for all tested isoforms of AQPs did not differ between WT Control and WT UGN animals as well as between GC-C WT and GC-C KO mice (Figure 3). A possible mechanism of decrease in salivary flow rate due to UGN action remains unclear. This mechanism possibly involves translocation of AQPs in apical membrane of the cells. However, all changes in mRNA expression were not tested in this study.

Figure 3.

Uroguanylin increased mRNA expression of Slc26a6 in submandibular salivary glands via GC-C independent signalling pathway. A: Effects of uroguanylin (UGN) on expression of channels for water (AQPs), sodium-hydrogen exchanger isoform 1 (NHE1) and 3 (NHE3), sodium-bicarbonate cotransporter (NBCn1), solute carrier family 26 member 6 (Slc26a6), and cystic fibrosis transmembrane conductance regulator (CFTR) are shown. *p < 0.05 statistically significant when compared to WT Control animals, n = 7-9. B: Missing GC-C in submandibular glands of GC-C KO mice lead to an increase in NHE3 and Slc26a6 expression (*p < 0.05 statistically significant when compared to GC-C WT mice, n = 7-10). The results are presented as mean ± SEM. GC-C KO - guanylate cyclase C (GC-C) knockout mice missing GC-C receptor, 26A6 – Slc26a6, WT – wild type mice

Nonetheless, one member of Slc26 family, also Cl^-/HCO₃^- exchanger, pendrin (Slc26a4), is down regulated by UGN (39), previous research has not shown any involvement of UGN in regulation of Slc26a3 nor Slc26a6. In our study, one hour after i.p. application, UGN increased expression of mRNA for Slc26a6, however, GC-C KO mice also showed an increase in mRNA expression when compared to their WT littermates, thus suggesting possible effects of several signalling pathways for UGN on Slc26a6 expression. There are few possible explanations for this hypothesis. Protein kinase C (activated by Ca²⁺) inhibits Slc26a6 function in the intestine cell line (40) but the downstream part of Ca²⁺ signaling pathway increases expression of Slc26a6 on the apical membrane of pancreatic or salivary ducts cells (41), which is in line with effects of UGN presented in this study. Furthermore, Slc26a6 is located in acinar cells as well as in ductal cells of submandibular salivary glands (42), thus suggesting possible effects of i.p. UGN on acinar cells via GC-C independent signalling pathway.

GC-C is located at apical membrane of the ductal cell of salivary glands (24). Since the expression of Slc26a6 is increased in GC-C KO mice, its activation might lead to a decrease in its expression. The involvement of GC-C in regulation of Slc26a6 was confirmed in the duodenum and kidney cortex (Figure 4). In addition to effects on Slc26a6, in duodenum of GC-C KO, mice expression of Slc26a3 was higher than in their WT littermates.

It is known that UGN, when activating GC-C, inhibits NHE3 in apical membrane of kidney and intestine cells (10, 43). It seems that the regulation is similar in salivary glands ducts as well because NHE3 expression is higher in submandibular glands of GC-C KO mice (Figure 3). Furthermore, it is not surprising that when UGN was applied i.p., mRNA expression for NHE3 did not change since the GC-C is located in apical membranes of salivary duct cells (24) which i.p. UGN cannot reach.

Conclusion

The most important conclusions of this research are: UGN, when applied i.p., changed the salivary flow rate, pH and ion concentrations, thus suggesting its possible involvement in regulation of postprandial salivary glands function. The changes of ion composition of saliva by GC-A agonists did not match the changes by GC-C agonist (UGN). The effects of the UGN on saliva flow rate are GC-C dependent, which is not the case with the changes of H⁺, Na⁺ and Cl^- concentrations. In mice missing the GC-C, the expression of Slc26a3 and/or a6 was higher in salivary glands, duodenum and kidney cortex suggesting wildly spread regulation of this ion transporter by GC-C. I.p. applied UGN did not decrease the NHE3 expression which was higher in GC-C KO mice.