Short-Chain Fatty Acids Reduced Renal Calcium Oxalate Stones by Regulating the Expression of Intestinal Oxalate Transporter SLC26A6

ABSTRACT Renal calcium oxalate (CaOx) stone is a common urologic disease with a high prevalence and recurrence rate. However, short-chain fatty acids (SCFAs) are less often reported in the prevention of urolithiasis. This study aimed to explore the effect of SCFAs on the renal CaOx stone formation and the underlying mechanisms. Ethylene glycol was used to induce renal CaOx crystals in rats. SCFAs (acetate, propionate, or butyrate) were added as supplements to the drinking water with or without antibiotics. Because intestinal oxalate transporters SLC26A6 and SLC26A3 regulate the excretion and absorption of oxalate in the intestine, we injected adeno-associated virus 9 (AAV9)-SLC26A6-shRNA (short hairpin RNA) and AAV9-SLC26A3 into the tail vein of rats to suppress SLC26A6 and overexpress SLC26A3 expression in the intestine, respectively, to explore the role of SLC26A3 and SLC26A6 (SLC26A3/6) in the reduction of renal CaOx crystals induced by SCFAs. Results showed that SCFAs reduced renal CaOx crystals and urinary oxalate levels but, however, increased the abundance of SCFA-producing bacteria and cecum SCFA levels. SCFA supplements still reduced renal crystals and urinary oxalate after gut microbiota depletion. Propionate and butyrate downregulated intestinal oxalate transporter SLC26A3 expression, while acetate and propionate upregulated SLC26A6 expression, both in vivo and in vitro. AAV9-SLC26A3 exerted a protective effect against renal crystals, while AAV9-SLC26A6-shRNA contributed to the renal crystal formation even though the SCFAs were supplemented. In conclusion, SCFAs could reduce urinary oxalate and renal CaOx stones through the oxalate transporter SLC26A6 in the intestine. SCFAs may be new supplements for preventing the formation of renal CaOx stones. IMPORTANCE Some studies found that the relative abundances of short-chain-fatty-acid (SCFA)-producing bacteria were lower in the gut microbiota of renal stone patients than healthy controls. Our previous study demonstrated that SCFAs could reduce the formation of renal calcium oxalate (CaOx) stones, but the mechanism is still unknown. In this study, we found that SCFAs (acetate, propionate, and butyrate) reduced the formation of renal calcium oxalate (CaOx) crystals and the level of urinary oxalate. Depleting gut microbiota increased the amount of renal crystals in model rats, and SCFA supplements reduced renal crystals and urinary oxalate after gut microbiota depletion. Intestinal oxalate transporter SLC26A6 was a direct target of SCFAs. Our findings suggested that SCFAs could reduce urinary oxalate and renal CaOx stones through the oxalate transporter SLC26A6 in the intestine. SCFAs may be new supplements for preventing the formation of renal CaOx stones.

ABSTRACT Renal calcium oxalate (CaOx) stone is a common urologic disease with a high prevalence and recurrence rate. However, short-chain fatty acids (SCFAs) are less often reported in the prevention of urolithiasis. This study aimed to explore the effect of SCFAs on the renal CaOx stone formation and the underlying mechanisms. Ethylene glycol was used to induce renal CaOx crystals in rats. SCFAs (acetate, propionate, or butyrate) were added as supplements to the drinking water with or without antibiotics. Because intestinal oxalate transporters SLC26A6 and SLC26A3 regulate the excretion and absorption of oxalate in the intestine, we injected adeno-associated virus 9 (AAV9)-SLC26A6-shRNA (short hairpin RNA) and AAV9-SLC26A3 into the tail vein of rats to suppress SLC26A6 and overexpress SLC26A3 expression in the intestine, respectively, to explore the role of SLC26A3 and SLC26A6 (SLC26A3/6) in the reduction of renal CaOx crystals induced by SCFAs. Results showed that SCFAs reduced renal CaOx crystals and urinary oxalate levels but, however, increased the abundance of SCFA-producing bacteria and cecum SCFA levels. SCFA supplements still reduced renal crystals and urinary oxalate after gut microbiota depletion. Propionate and butyrate downregulated intestinal oxalate transporter SLC26A3 expression, while acetate and propionate upregulated SLC26A6 expression, both in vivo and in vitro. AAV9-SLC26A3 exerted a protective effect against renal crystals, while AAV9-SLC26A6-shRNA contributed to the renal crystal formation even though the SCFAs were supplemented. In conclusion, SCFAs could reduce urinary oxalate and renal CaOx stones through the oxalate transporter SLC26A6 in the intestine. SCFAs may be new supplements for preventing the formation of renal CaOx stones. IMPORTANCE Some studies found that the relative abundances of short-chain-fattyacid (SCFA)-producing bacteria were lower in the gut microbiota of renal stone patients than healthy controls. Our previous study demonstrated that SCFAs could reduce the formation of renal calcium oxalate (CaOx) stones, but the mechanism is still unknown. In this study, we found that SCFAs (acetate, propionate, and butyrate) reduced the formation of renal calcium oxalate (CaOx) crystals and the level of urinary oxalate. Depleting gut microbiota increased the amount of renal crystals in model rats, and SCFA supplements reduced renal crystals and urinary oxalate after gut microbiota depletion. Intestinal oxalate transporter SLC26A6 was a direct target of SCFAs. Our findings suggested that SCFAs could reduce urinary oxalate and renal CaOx stones through the oxalate transporter SLC26A6 in the intestine. SCFAs may be new supplements for preventing the formation of renal CaOx stones.

RESULTS
Oral administration of SCFAs reduced renal CaOx crystals and urinary oxalate level. We successfully developed renal CaOx stone model rats (ethylene glycol [EG] group) with 1% (vol/vol) EG drinking water. To explore the roles of three main SCFAs produced by gut microbiota (acetic acid, propionic acid, and butyric acid) in reducing renal stones, we provided three subgroups of model rats with sodium acetate (EG 1 Acetate group), sodium propionate (EG 1 Propionate group), or sodium butyrate (EG 1 Butyrate group) in drinking water. As we previously reported, hematoxylin and eosin (HE) staining showed that there were crystals in the lumen of the renal tubules with tubular dilation in model rats, whereas administration of one of the three SCFAs (acetate, propionate, or butyrate) reduced the renal crystals and the abnormality of renal histologic structure (27 (Fig. 1C). These data indicated that SCFA treatment significantly reduced urinary oxalate levels and renal CaOx crystals.
Oral administration of SCFAs regulated expression of the intestinal oxalate transporters SLC26A3/6. We next examined whether the decreased urinary oxalate level was associated with the level of intestinal oxalate transporters SLC26A3/6 at mRNA and protein levels using quantitative real-time PCR (qRT-PCR) assay and Western blot assay by Student's t test. The expression of SLC26A3 in the cecum at mRNA (2.7-fold, P , 0.001) and protein (2.6-fold, P = 0.047) levels was higher in the EG group than the control group (Fig. 2E, G, and H), whereas the level of SLC26A3 decreased in the ileum at mRNA (0.1-fold, P , 0.001) and protein (0.6-fold, P = 0.038) levels ( Fig. 2A, C, and D) and in the colon at mRNA (0.1-fold, P , 0.001) and protein (0.5-fold, P = 0.035) levels ( Fig. 2I, K, and L) after the administration of butyrate and in the cecum after the administration of acetate (mRNA level: 0.2-fold, P , 0.001; protein level: 0.3-fold, P = 0.045), propionate (mRNA level: 0.4-fold, P , 0.001; protein level: 0.3-fold, P = 0.033), or butyrate (mRNA level: 0.1-fold, P , 0.001; protein level: 0.2-fold, P = 0.022) (Fig. 2E, G, and H). The expression of SLC26A6 at mRNA and protein levels was not different between the control group and the EG group (P . 0.05). Acetate increased the expression of SLC26A6 in the ileum (mRNA level: 3.5-fold, P , 0.001; Short-Chain Fatty Acids Reduced Renal Stones protein level: 2.1-fold, P = 0.017) and cecum (mRNA level: 1.8-fold, P = 0.001; protein level: 1.3-fold, P = 0.030) (Fig. 2B to D and F to H). The expression of SLC26A6 in the ileum (mRNA level: 2.7-fold, P , 0.001; protein level: 2.1-fold, P = 0.033) and cecum (mRNA level: 1.5-fold, P , 0.001; protein level: 1.6-fold, P = 0.044) also increased after the administration of propionate (Fig. 2B to D and F to H). Immunofluorescence assay also showed that butyrate (0.6-fold, P = 0.032) decreased intestinal SLC26A3 protein expression, while acetate (2.0-fold, P , 0.001) and propionate (1.9-fold, P , 0.001) increased intestinal SLC26A6 protein expression in rats (see Fig. S1 in the supplemental material). These data showed that administration of SCFAs could upregulate SLC26A6 and downregulate SLC26A3 expression in the intestine.
Oral administration of SCFAs changed gut microbiota and SCFA levels in the content of cecum. After the administration of SCFAs, we also examined the gut microbiota using 16S rRNA gene sequencing. Beta diversity analysis evaluated by principalcoordinate analysis (PCoA) (Bray-Curtis) showed the apparent variation of the gut microbiota from the above five groups, of which the samples could be classified into five clusters (Fig. 3A, analysis of similarity [ANOSIM], R = 0.365, P = 0.001; ADONIS test, R 2 = 0.272, P = 0.001). Linear discriminant analysis effect size (LEfSe) analysis showed that the relative abundances of Lachnospiraceae (P = 0.010), Ruminococcus (P = 0.006), Eubacterium (P = 0.003), and Prevotellaceae (P = 0.028), which can produce SCFAs, were higher in rats receiving acetate, propionate, or butyrate (Fig. 3B). Moreover, the administration of propionate also increased the abundance of O. formigenes (P = 0.007) (Fig. S2). We obtained metabolic pathway information using Tax4Fun. Metabolic pathway analysis based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) database showed that propionate increased the pathway ko00680 (methane metabolism), which is involved in the production of acetic acid (P = 0.017) (Fig. 3C).
Student's t test showed that the level of acetic acid in the content of cecum increased in rats administered propionate compared with model rats (1.05 versus 0.71 mg/mg, P , 0.001) (Fig. 4A). Both cecal propionic acid (EG 1 Propionate versus EG: 0.51 versus 0.30 mg/mg, P = 0.001; EG 1 Butyrate versus EG: 0.53 versus 0.30 mg/mg, P , 0.001) and butyric acid (EG 1 Propionate versus EG: 0.65 versus 0.40 mg/mg, P = 0.020; EG 1 Butyrate versus EG: 0.71 versus 0.40 mg/mg, P = 0.013) increased in propionate-and butyrate-supplemented groups over the model group ( Fig. 4B and C). These results suggested that administration of SCFAs also changed the gut microbiota and SCFA levels in the content of cecum.
Depletion of gut microbiota increased CaOx crystal deposition in kidneys. To determine whether gut microbiota also regulated the formation of renal stones, we used antibiotics (Abx) to reduce the intestinal bacterial load and examined the change of crystal deposition in kidneys. Alpha diversity analysis using the sobs index showed that the administration of antibiotics significantly decreased the richness of gut microbiota in model rats (Fig. 5A). Student's t test showed that model rats with antibiotics had higher VK staining scores than those without antibiotics (34,851 versus 15,789, P = 0.006) ( Fig. 6A and C). The urinary oxalate level was also higher in model rats with antibiotics than those without (436 versus 364 mg/ml, P = 0.028) (Fig. 6D). These results suggested that the depletion of gut microbiota could promote the urinary oxalate level and the formation of renal crystals.
Supplementation of SCFAs with depletion of gut microbiota decreased renal crystals via regulating expression of intestinal oxalate transporters SLC26A3/6. To explore whether SCFAs decreased the renal stone formation via the gut microbiota, the gut microbiota of these five groups were depleted by antibiotics. The alpha diversity (sobs index) of gut microbiota of these five groups with antibiotics was much lower than that of those without antibiotics (P , 0.05) (Fig. 5A). The SCFAs in the content of cecum did not increase after administration of acetate, propionate, or butyrate ( Fig. 5B to D).
The protective effect of SCFAs against renal CaOx stones was attenuated by suppression of intestinal SLC26A6 but not by overexpression of intestinal SLC26A3. To verify the role of intestinal oxalate transporters SLC26A3 and SLC26A6 in the reduction of renal CaOx crystals induced by SCFAs, SLC26A3 was overexpressed by adeno- and SLC26A6 at mRNA (I and J) and protein (K and L) level in colon. At mRNA level, five rats of each group were examined, and the duplication was set as 3. At protein level, three rats of each group were examined. The relative protein expression of control group was set as 1. Student's t test was applied. P . 0.05 (n.s.), P , 0.05 (*), P , 0.01 (**), P , 0.001 (***).
HE and VK staining showed that the decrease of SLC26A6 expression or the increase of SLC26A3 expression in the intestine alone did not increase the amount of CaOx crystals in kidney nor the level of urinary oxalate (P . 0.05) with Student's t test. However, after the administration of EG, the decrease of intestinal SLC26A6 expression contributed to more renal crystals (HE scores: 6 versus 10.2, P = 0.010; area values in VK staining: 5,266.73 versus 15,804.3, P , 0.001) and higher urinary oxalate level (318 versus 476 mg/ml, P = 0.049). In contrast to the results mentioned above, renal crystals also existed after the decrease of intestinal SLC26A6 expression even though acetate, propionate, or butyrate was provided. In addition, the urinary oxalate levels between EG group (318 mg/ml) and A61EG1Acetate group (288 mg/ml), A61EG1Propionate group (245 mg/ml), or A61EG1Butyrate group (292 mg/ml) were not significantly different (P . 0.05). Interestingly, the increase of intestinal SLC26A3 expression decreased the renal crystals (HE scores: 6 versus 3, P = 0.021; area values in VK staining: 5,266.73 versus 3,041.73, P , 0.001). However, the urinary oxalate levels were not significantly different between EG group and EG1A3 group (318 versus 319 mg/ml, P . 0.05). There was no crystal in the kidney in the A31EG1Acetate, A31EG1Propionate, and A31EG1Butyrate groups. The results of HE and VK staining and urinary oxalate level were shown in Fig. 10. These results revealed that the intestinal SLC26A6 played a

DISCUSSION
Recently, several studies have demonstrated that renal stones may be associated with gut microbiota (9,10,39,40). Our previous study showed that the abundance of SCFA-producing bacteria was lower in the gut microbiota of patients with renal CaOx stones than in healthy people (41), which indicated a possible preventive role of SCFAs in renal stone formation. In this study, the results showed that SCFAs, including acetate, propionate, and butyrate, could reduce renal CaOx stones and urinary oxalate levels via regulating oxalate transporter (SLC26A3/A6) expression in rats, and the phenomenon also existed after the depletion of gut microbiota. The protective role of SCFAs against renal CaOx stones was attenuated after the decrease of intestinal SLC26A6 expression. Intestinal SLC26A3 also participated in the formation of renal crystals but not by regulating the oxalate metabolism.
SLC26 is a family of anion transporters with 10 different subtypes, of which SLC26A3 and SLC26A6 significantly affected the oxalate transportation in the intestine (16). Many studies have already verified the causal relationship between oxalate transporter SLC26, oxalate metabolism, and renal CaOx crystals. In the past 2 decades, five studies reported that SLC26A6 knockout mice developed renal CaOx stones, hyperkalemia, and hyperoxaluria due to the enhanced net oxalate absorption in the ileum (19,20,(42)(43)(44). Our study also verified that the decrease of intestinal SLC26A6 induced more renal crystals. We also found that acetate or propionate could increase the expression of SLC26A6 in the ileum, while this phenomenon weakened in cecum and colon. It may be due to the varied levels of SLC26A6 expression in different parts of the intestine. SLC26A6 has been demonstrated to have higher levels of expression in the small intestine than in the colon (18).
Another study showed that all segments of the intestine in SLC26A3 knockout mice exhibited net oxalate secretion, and 24-h urinary oxalate level decreased by 66% compared with control mice (21). However, our study found that the increase of intestinal SLC26A3 expression did not increase the urinary oxalate level. In contrast, it reduced renal crystals. It was reported previously that intestinal SLC26A3 was in charge of bicarbonate ion secretion and chloride ion and water absorption and that patients with SLC26A3 mutations would suffer from acidic diarrhea and systemic alkalosis (17). Thus, we hypothesized that the overexpression of SLC26A3 in the intestine may lead to less excretion of bicarbonate ions into urine and decrease renal CaOx crystal formation. On the other hand, SLC26A3 also played an important role in maintaining the intestinal epithelial barrier function (45). The mutation of SLC26A3 was associated with inflammatory bowel disease (IBD) (46). Additionally, IBD patients had a 2-fold-higher risk of urolithiasis than non-IBD individuals (47). Thus, mutation of intestinal SLC26A3 may be a risk factor for renal CaOx stones. These pieces of evidence indicated that the intestinal SLC26A3 also played an important role in preventing, rather than contributing to, the formation of renal CaOx stones.
Gut microbiota was also related to many diseases, including metabolic syndrome, autoimmune diseases, colorectal cancer, and inflammatory bowel disease (48)(49)(50)(51). In addition, it was reported that the relative abundance of O. formigenes was lower in the gut microbiota of renal stone patients than in healthy people (52). However, the application of probiotics containing O. formigenes for decreasing urinary oxalate levels was still controversial (25,26). Intriguingly, in our study, we found that the administration of propionate could elevate the abundance of O. formigenes in model rats with decreased urinary oxalate level. With the development of 16S rRNA gene sequencing, some studies found that apart from O. formigenes, there were also some other different gut bacteria between renal stone patients and healthy controls (10,39,40). For example, Bacteroides are higher in renal stone patients, whereas Faecalibacterium, Eubacterium hallii, Dorea, Ruminiclostridium, and Fusicatenibacter, which can produce SCFAs, are higher in healthy people. These studies indicated that the disturbance of the gut microbiota might be associated with the formation of renal stones. In our study, we found that SCFA-producing bacteria increased after the administration of SCFAs. The depletion of gut microbiota by antibiotics promoted the deposition of renal crystals. These pieces of evidence indicated that gut microbiota plays a critical role in preventing renal stones. Nevertheless, we also found that oral supplementation with SCFAs of rats with the gut microbiota depleted could also reduce the risk of renal stones. This study suggested that SCFAs could be used as prebiotics to prevent renal stones even though the gut microbiota disorder existed. Nevertheless, further drug toxicology evaluations and clinical trials are needed before SCFAs can be used as treatments for renal CaOx stones. To this end, further efforts to reveal the details of how SCFAs and the gut microbiome function in renal stone alleviation are required.
In conclusion, SCFAs could reduce urinary oxalate and renal CaOx stones through the oxalate transporter SLC26A6 in the intestine. SCFAs may be new supplements for preventing the formation of renal CaOx stones.

MATERIALS AND METHODS
Renal stone model and treatment regimes. Rats (Sprague-Dawley rats, 6 weeks old, male), obtained from Chengdu Dossy Experimental Animals Co., Ltd. (Chengdu, Sichuan, China), were acclimatized for 1 week before the experiment, followed by a 4-week cohousing period in the specific-pathogen-free animal facility at the Animal Experiment Center of West China Hospital, Sichuan University. All rats were singly housed and provided with sterilized food and water. The West China Hospital of Sichuan University Medical Research Ethics Committee approved the study (2017063A).
After 4 weeks, we collected 24-h urine and blood plasma. Renal and intestinal (ileum, cecum, and colon) tissues were fixed in 10% formaldehyde and embedded in paraffin. The remaining parts of these tissues, the corresponding intestinal content, and urine and blood plasma samples were all stored at 280°C for future analyses.
Cell culture experiments. The human intestinal epithelial line Caco-2 is a kind of human colon adenocarcinoma cell line, which has a similar structure and function as intestinal epithelial cells. Caco-2 cells were cultured in 1640 medium (HyClone, Logan, UT, USA) with 10% fetal bovine serum (Gibco, Carlsbad, CA) in a cell incubator, which was set at 37°C and 5% CO 2 . Different levels of oxalic acid (0, 0.01, 0.05, 0.1, 0.5, and 1 mM) and/or SCFAs (0, 0.1, 1, 5, 10, and 20 mM) were added into the Caco-2 cell culture system for 24 h, and then the SLC26A3/6 expression was measured.
Examination of renal crystals. Three-to 4-mm sections of the kidney were prepared and stained with HE and VK according to the protocols of staining kits (Solarbio, Beijing, China). The HE staining scoring system reported by Xiang et al. (54) and the Von Kossa staining scoring method using Image Pro Plus6 software were separately applied as semiquantitative and quantitative methods to evaluate the renal crystals. Higher scores represented more crystals in the kidney.