Uric acid transporters hiding in the intestine

Abstract

Context: Hyperuricaemia is known as an abnormally increased uric acid level in the blood. Although it was observed many years ago, since uric acid excretion via the intestine pathway accounted for approximately one-third of total elimination of uric acid, the molecular mechanism of ‘extra-renal excretion’ was poorly understood until the finding of uric acid transporters.

Objective: The objective of this study was to gather all information related to uric acid transporters in the intestine and present this information as a comprehensive and systematic review article.

Methods: A literature search was performed from various databases (e.g., Medline, Science Direct, Springer Link, etc.). The key terms included uric acid, transporter and intestine. The period for the search is from the 1950s to the present. The bibliographies of papers relating to the review subject were also searched for further relevant references.

Results: The uric acid transporters identified in the intestine are discussed in this review. The solute carrier (SLC) transporters include GLUT9, MCT9, NPT4, NPT homolog (NPT5) and OAT10. The ATP binding cassette (ABC) transporters include ABCG2 (BCRP), MRP2 and MRP4. Bacterial transporter YgfU is a low-affinity and high-capacity transporter for uric acid.

Conclusion: The present review may be helpful for further our understanding of hyperuricaemia and be of value in designing future studies on novel therapeutic pathways.

Keywords:

• Extra-renal excretion
• gout
• hyperuricaemia

Introduction

Gout, a painful form of inflammatory arthritis, emerged as a ‘wealthy’ epidemic in the eighteenth century England. The epidemiological link between hyperuricaemia and gout was established nearly two centuries ago (Richette & Bardin 2010). An abnormally high level of blood uric acid is known as hyperuricaemia. With the formation of urate (the ionized form of uric acid) crystals, hyperuricaemia causes gout in the joints and predisposes to nephrolithiasis in the kidney. Recent evidence (Lima et al. 2015; Perez-Ruiz et al. 2015) is accumulating indicating a role for hyperuricaemia in the genesis and progression of hypertension, diabetes mellitus, metabolic syndrome, renal diseases, cardiovascular diseases and even the aging process (Campion et al. 1987; Iwama et al. 2012; Krishnan et al. 2014). In both academic and clinical practice, there has been increased interest in hyperuricaemia worldwide. Uric acid has now been identified as a predictor of incident events of these diseases (So & Thorens 2010).

The molecular physiology of uric acid homeostasis

Uric acid is produced by humans as the final metabolic product of purines, both endogenous and exogenous. Uric acid is a weak acid (pK_a 5.8) and has low solubility in water as well as in plasma. At physiological pH, uric acid circulates mainly in the form of urate. Hyperuricaemia pathogenicity is associated with low solubility of urate in the extracellular environment leading to crystal formation. The levels of urate in the blood are maintained by the balance between uric acid generation and excretion (de Oliveira & Burini 2012). An alteration in this balance may account for hyperuricaemia, so either decreased excretion, increased synthesis or both will lead to elevated plasma uric acid concentrations (hyperuricaemia). Although dietary, genetic, or disease-related excesses in uric acid production underlie hyperuricaemia in some cases, practically the main cause for hyperuricaemia is lower excretion of uric acid. Among 65 hyperuricaemic patients analyzed for uric acid metabolism, 9.2% had over-production type, 80.0% had under-excretion type and 10.8% had a mixed type (Oka et al. 2014).

In the blood, urate would theoretically reach saturation in the concentration of 6.4 mg/dL, which may not occur because protein (mainly albumin) bound urate increases its solubility (Susic & Frohlich 2015). Similarly, the binding to protein is necessary for uric acid to be cleared from the circulation. Excretion of uric acid requires specialized transporters located in renal tubule cells and intestinal epithelial cells. The importance of the transporters in excretion of uric acid has become increasingly evident.

Uric acid transport by the kidney has been investigated for many years. The kidney plays a crucial role in maintaining plasma urate levels through complex transepithelial transport systems that promote both re-absorption and secretion of uric acid (Hediger et al. 2005). The primary transporter involved in uric acid re-absorption is urate transporter 1 (URAT1), which is only expressed in the kidney (Enomoto et al. 2002). Molecular identification of URAT1 promoted the advances of new transporters involved in renal uric acid transport. Many of transporters have been identified and well reviewed (Hediger et al. 2005; Richette & Bardin 2010; So & Thorens 2010). Recently, genome-wide association studies indicated a substantial association between uric acid concentration and single nucleotide polymorphisms (SNPs) of transporter-coding genes such as SLC2A9 (GLUT9), ABCG2 (BCRP), SLC17A1 (NPT1), SLC17A3 (NPT4), SLC17A4 (NPT5), SLC22A11 (OAT4), SLC22A12 (URAT1) and SLC16A9 (MCT9). GLUT9 (glucose transporter 9) has unexpectedly been shown to play a critical role in uric acid re-absorption at the proximal tubule, probably more prominent than its partner URAT1 (Sakurai 2013).

Uric acid transporters in the intestine

The intestine can also secrete uric acid. As far back as half a century ago, it was estimated that approximately one-third of uric acid was excreted into the gut and was further metabolized by resident bacteria (Buzard et al. 1952; Sorensen 1965). In oxonate-treated rats, significant amounts of externally administered and endogenous uric acid were recovered in the intestinal lumen, while biliary excretion was minimal. Accordingly, direct intestinal secretion was thought to be a substantial contributor to extra-renal elimination of uric acid (Hosomi et al. 2012). It is not difficult to imagine that the intestinal contribution of uric acid excretion is an important alternative pathway during renal insufficiency. Enhanced enteric excretion of uric acid may account for the lack of significant hyperuricaemia in chronic renal failure (Vaziri et al. 1995). In most gouty subjects, extra-renal disposal of uric acid is even greater. In some patients with renal insufficiency, enteral uricolysis constitutes the major process of uric acid disposal (Sorensen 1975). However, there had been few mechanistic studies on the intestinal ‘extra-renal excretion’ pathway of uric acid before the finding of UA transporters.

In the intestine, purine compounds are metabolized to uric acid, which is transported. The small intestine is a major site of uric acid production in the body (Oh et al. 1967). Furthermore, the small intestine represents the primary site of absorption for all ingested compounds. Intestinal transporters play significant roles in the absorbing process. P-Glycoprotein (P-gp), the first clinically important drug transporter, was identified in 1986 (Roninson et al. 1986).

An interest for other intestinal transporters is now emerging. The largest human intestinal transporter database (>5000 interaction entries for >3700 molecules) have been assembled and curated (Sedykh et al. 2013). Among those transporters mediating the cellular uptake and efflux of a broad variety of endogenous compounds, drugs and their metabolites, solute carriers (SLCs) and ATP binding cassette (ABC) transporters represent two superfamilies in humans. ABC transporters (e.g., multidrug resistance proteins, MRPs) utilize the energy from ATP hydrolysis and function as efflux transporters. SLC transporters (e.g., organic anion transporters, OATs) do not rely directly on ATP hydrolysis and are largely (but not exclusively) uptake importers moving small molecules from the extracellular milieu into cells (Estudante et al. 2013). While much research has been conducted focusing on the role of uric acid transporters in the kidney, less is known about the importance of uptake and efflux transporters identified in the intestine. The molecular nature of uric acid transporters in the intestine has so far not been investigated in detail.

ABCG2/BCRP

The ATP-binding cassette, subfamily G, 2 (ABCG2) was originally identified in placental tissue as a xenobiotic transporter in relation to the development of multidrug resistance of breast cancer cells. It was, therefore, also termed ‘breast cancer resistance protein’ (BCRP) (Doyle et al. 1998). ABCG2 is expressed on the apical membrane in tissues including kidney and intestine. The story of ABCG2 and uric acid began with a genome-wide association study, which indicated that ABCG2 played a crucial role of uric acid homeostasis through both renal and extra-renal uric acid excretion (Dehghan et al. 2008; Nakayama et al. 2011). At least 10% of all gout cases in whites are attributable to rs2231142, a common functional polymorphism in ABCG2 (Woodward et al. 2009). ABCG2 dysfunction caused renal uric acid under-excretion and induced hyperuricaemia even if the renal uric acid overload was not remarkable (Matsuo et al. 2014).

Recent studies suggest that ABCG2 has an important role in intestinal excretion. Given higher expression of ABCG2 in the intestine, it is likely that the hypoactive variant of ABCG2 leads to decreased uric acid excretion into the intestine rather than decreased uric acid excretion from the kidney (Sarkadi et al. 2004). In Caco-2 cells, uric acid showed a polarized flux from the basolateral to apical side, and this flux was almost abolished in the presence of an ABCG2 inhibitor (Hosomi et al. 2012). ABCG2 knockout mice significantly exhibited a reduction of intestinal uric acid excretion and an increased plasma concentration of uric acid (Ichida et al. 2012). In the 5/6 nephrectomy rat model of chronic kidney disease, serum UA did not increase despite the decrease in UA excretion. Meanwhile, ABCG2 was significantly increased in the intestine (Yano et al. 2014). Increased serum uric acid in patients with ABCG2 dysfunction could be explained by the decreased excretion of uric acid from the intestine (Takada et al. 2014). Patients with mild to severe ABCG2 dysfunction accounted for 88.2% of early-onset gout patients. Severe ABCG2 dysfunction particularly increased the risk of early-onset gout (Matsuo et al. 2013). These results suggest that decreased intestinal uric acid excretion caused by ABCG2 dysfunction might be a common mechanism of hyperuricemia.

Multidrug resistance proteins (MRPs, ABCCs)

MRP2 and MRP4 both have been shown to transport p-aminohippurate (PAH), the classical substrate used in the characterization of organic anion transport in renal proximal tubular cells. Various inhibitors of MRP2-mediated PAH transport also inhibited MRP4. However, MRP4 has higher affinity for PAH and is expressed more than MRP2 in the kidney (Smeets et al. 2004). With respect to renal physiology and pharmacology, particular attention should be paid to MRP2 and MRP4. In the apical membrane of kidney proximal tubules, MRP4 mediates ATP-dependent uric acid transport through multiple allosteric binding sites. MRP4 appears to control ATP dependent uric acid extrusion from the cells into the tubular lumen and thus contributes to uric acid excretion. Human MRP4, but not MRP2, expressed in HEK293 cells, mediates ATP-dependent efflux of uric acid (Van Aubel et al. 2004). Cultured avian proximal tubules lack re-absorptive uric acid transport, knockdown of MRP4 expression reduces uric acid secretion by 65% (Bataille et al. 2008).

MRPs forms 1 to 6 are present in intestinal epithelial cells (Grandvuinet et al. 2012). In jejunal tissue, MRP2 is highly expressed while MRP4 is expressed to a low extent (Hilgendorf et al. 2007). In cultured chicken proximal tubules, RT-PCR revealed mRNA for MRP2- and MRP4-like organic anion transporters in avian proximal epithelium. Luminal application of MK-571 (a known substrate for MRP2 and MRP4) dramatically reduced both uric acid secretory and re-absorptive flux (Dudas et al. 2005).

Glucose transporter 9 (GLUT9)

GLUT9 was initially identified by sequence similarity with the glucose transporter. Differing from other GLUTs, GLUT9 is most probably not involved in glucose and/or fructose transport (Mueckler & Thorens 2013). The relationship between GLUT9 and uric acid has recently been highlighted. Common variation in the SLC2A9 gene that encodes GLUT9 might explain 1.7–5.3% of the variance in serum uric acid concentrations, following a genome-wide association scan (Vitart et al. 2008). Despite the ever-increasing list of hyperuricaemic genes, variation in SLC2A9 remains the major single genetic determinant of serum uric acid, followed closely by the effects of variation in the ABCG2 gene (Mandal & Mount 2015). Mice with systemic knockout of Glut9 displayed moderate hyperuricaemia, massive hyperuricosuria and mild renal insufficiency (Preitner et al. 2009).

Human GLUT9 has two splice variants with different expression patterns: GLUT9a and GLUT9b. The two splice variants are expressed differentially within polarized cells, with GLUT9a localized predominantly on the basolateral surfaces and GLUT9b expressed on apical surfaces. The two isoforms could not be distinguished by the use of antibody, so the polarized distribution of the two splice variants is not confirmed in vivo (Augustin et al. 2004). GLUT9 is widely expressed, with substantial transcript levels primarily in kidney and liver, but is present at low levels in several other tissues, including small intestine and colon. GLUT9 is also expressed in articular cartilage, cytokines could induce expression of GLUT9 in chondrocytes, and thus it is possible that GLUT9 plays a role in gout (Phay et al. 2000). Because GLUT9 is present in the intestine, its genetic inactivation may lower intestinal excretion of uric acid, although this has not been formally tested. Enterocyte uric acid metabolism could potentially be targeted to modulate or prevent the development of the metabolic syndrome. Glut9-deficient mice developed impaired enterocyte uric acid transport kinetics, hyperuricaemia, hyperuricosuria, spontaneous hypertension, dyslipidaemia and elevated body fat (DeBosch et al. 2014).

Monocarboxylate transporter 9 (MCT9)

MCT9, a member of the monocarboxylate co-transporter family, was ubiquitously expressed and was especially expressed at a high level in the kidney (Halestrap & Price 2004). SLC16A9, encoding MCT9, was identified from analysis of the human genomic expressed sequence tag (EST) databases (Halestrap & Price 1999). Meta-analysis of genome-wide association studies revealed a relationship between SLC16A9 and serum uric acid concentrations (Phipps-Green et al. 2016). Genotyping of rs12356193 within SLC16A9 was associated with dl-carnitine and propionyl-l-carnitine concentrations, which in turn were associated with serum UA levels. It implies that MCT9 could indirectly affect extra-renal urate excretion (Kolz et al. 2009). Another missense variant of MCT9, rs2242206, significantly increased the risk of renal overload gout but not of renal underexcretion gout. Although the function of MCT9 remains not fully unclear, MCT9 might have a possible physiological role in urate excretion from human intestinal epithelial cells where MCT9 expression is observed (Nakayama et al. 2013).

Na⁺/phosphate co-transporters

The Na⁺/phosphate co-transporters (NPTs), an SLC17 family of proteins initially characterized as phosphate carriers, mediate the transport of organic anions. Three SLC17 family members associated with serum uric acid concentrations have been identified through genomic analysis, and designated NPT1 (SLC17A1), NPT4 (SLC17A3) and NPT homolog (SLC17A4). Human sodium phosphate transporter 4 (hNPT4) is localized at the apical side of renal tubules and functions as a voltage-driven uric acid transporter. It is likely to act as a common secretion route for both drugs and may play an important role in diuretics-induced hyperuricaemia (Jutabha et al. 2010). Actually, NPT4 is present in the small intestine too (Reimer & Edwards 2004).

SLC17A4 protein was isolated from a human intestine mucosal. The amino acids sequence shows 48% overall similarity with the renal NPT1 protein (Shibui et al. 1999). Since NPT1 is responsible for renal uric acid extrusion, heterologous expression reveals functional characteristics similar to those of NPT1 and NPT4, indicating the possible involvement of the NPT homolog (NPT5) in uric acid extrusion from the intestinal tract (Togawa et al. 2012).

Organic anion transporter 10 (OAT10)

The OAT subfamily (SLC22A) constitutes roughly half of the SLC22 transporter family. Most of OATs are highly expressed in the kidneys. Nine OATs were functionally identified (OAT1-7, Oatv1 and URAT1). Some OATs have a strong selectivity for uric acid (Nigam et al. 2015). Human organic anion transporter (hOAT10) is highly expressed in the kidney and to a weaker extent in the intestine. From a cyclosporine A-induced hyperuricaemia, hOAT10 was identified as a new uric acid transporter (Bahn et al. 2008).

Bacterial uric acid transporters

The intestine is the main site of uricolysis (the degradation of uric acid). Some disappearance of uric acid was observed in cultures of Escherichia coli, Aerobacter aerogenes and Paracolobactrum species, isolated from human feces and from intestinal contents of rats (Lau & Wiseman 1964). In the lower intestinal tract, uric acid is exposed to a large population of bacteria that can use the uric acid as a metabolic substrate (Braun & Campbell 1989). When uric acid was administered to human subjects by intravenous injections, about 80% of the uric acid was excreted in the urine. If administered by mouth, uric acid was extensively degraded in the gastrointestinal tract, presumably by bacterial action (Wyngaarden & Stetten 1953). The ubiquitous nucleobase–ascorbate transporter (NAT/NCS2) family includes more than 2000 members, but only 15 have been characterized experimentally. Among Escherichia coli members, the xanthine permeases XanQ and XanP are functionally known. Of the remaining members, YgfU is closely related in sequence and genomic locus with XanQ. YgfU is a proton-gradient dependent, low-affinity and high-capacity transporter for uric acid (Papakostas & Frillingos 2012).

Perspectives

The increasing knowledge of uric acid transporters sheds light on the causes of hyperuricemia. Identification of the renal transporter URAT1 in 2002 paved the way for successive clarification of the uric acid transport system. Novel technologies and various useful public databases have improved our understanding of the physiological roles of transporters. Although the intestinal contribution in uric acid excretion has been recognized over the past 50 years, molecular information of intestinal uric acid transport is obtained quite recently, with ABCG2 being the best characterized to date. The battle to ferret out uric acid transporters hiding in the intestine has just begun.

Actually, the history of transporter discovery in the intestine (from P-gp) is longer than that in the kidney. As outlined in this review, some transporters participating in uric acid transport in the kidney are also expressed in the intestine (Table 1). However, are the similar mechanisms of these transporters for uric acid transport in the intestine? There is much more we need to address. Future research will continue to investigate the precise role of the known transporters and to search for candidates related to intestinal uric acid handling. The pathophysiological roles of intestinal uric acid transporters need to be unveiled, especially during renal insufficiency processes. In view of a wider space and a wide variety of bacteria in the intestine, it seems likely that transporters in the intestine would be much more complex than in the kidney. Studying transporter interaction profiles can be critical to fully understand the precise mechanism governing the bi-directional transport of uric acid. The interaction of transporter with other associated proteins has not been functionally confirmed yet, as uric acid transporters represents an important challenge. The entire picture of uric acid metabolism will advance our knowledge and give a wider selection of effective treatments for hyperuricemia.

Table 1. Uric acid transporters in the intestine.

Name	Encoding gene	Studying stage	References
ABCG2/BCRP	ABCG2	In vitro, in vivo, in clinic	Sarkadi et al. (2004), Nakayama et al. (2011), Hosomi et al. (2012), Ichida et al. (2012), Matsuo et al. (2014), Takada et al. (2014) and Yano et al. (2014)
MRP2	ABCC2	In vivo	Dudas et al. (2005), Hilgendorf et al. (2007) and Grandvuinet et al. (2012)
MRP4	ABCC4	In vivo	Dudas et al. (2005), Hilgendorf et al. (2007) and Grandvuinet et al. (2012)
GLUT9	SLC2A9	In vivo	Phay et al. (2000) and, DeBosch et al. (2014)
MCT9	SLC16A9	In clinic	Nakayama et al. (2013)
NPT4	SLC17A3	No study	Reimer and Edwards (2004)
NPT homolog (NPT5)	SLC17A4	In vivo	Shibui et al. (1999) and Togawa et al. (2012)
OAT10	SLC22A13	In vitro, in vivo	Bahn et al. (2008)
YgfU	ygfU	In vitro	Papakostas and Frillingos (2012)

Disclosure statement

The authors report that they have no conflict of interest.

Funding

This project was granted financial support from China Postdoctoral Science Foundation, No. 2015M571253.