Medical Genetics Unit, Department of Gynecologic, Obstetric and Pediatric Sciences, University of Bologna, Policlinico S. Orsola-Malpighi, via Massarenti 9, Bologna 40138, Italykerry.rhoden@unibo.it
December 2011
The solute carrier family-5 member-8 (SLC5A8), identified simultaneously as a transporter and as a tumour suppressor (Rodriguez et al., 2002; Li et al., 2003), has drawn attention for its potential role in tumorigenesis at several sites, and as a potential prognostic marker and therapeutic target in neoplastic disease.
SLC5A8 expression and function in normal tissues
SLC5A8 belongs to the SLC5 family of sodium-coupled transporters which includes at least 12 structurally-related members with diverse tissue distribution and substrate specificity. SLC5A8, in particular, is a sodium-coupled monocarboxylate transporter (also known as SMCT1) predominantly found in the small intestine, colon, thyroid gland, kidney and salivary glands, and to a lesser extent in the retina and brain (Rodriguez et al., 2002; Gopal et al., 2004; Takebe et al., 2005; Iwanaga et al., 2006; Martin et al., 2006; Gopal et al., 2007; Martin et al., 2007; Frank et al., 2008). Electrophysiological and radiotracer studies in cells expressing recombinant SLC5A8 have demonstrated its ability to transport monocarboxylates such as butyrate, proprionate, acetate, lactate, pyruvate, and nicotinate (a B-complex vitamin) (Coady et al., 2004; Miyauchi et al., 2004; Gopal et al., 2004; Gopal et al., 2005), as well as ketone bodies and the amino acid derivative pyroglutamate (Martin et al., 2006; Miyauchi et al., 2010). Substrates are cotransported with sodium into cells, following the inward electrochemical gradient for sodium ions maintained by the sodium-potassium ATPase. Transport is electrogenic, due to a 2:1 Na+:monocarboxylate stoichiometry that results in the transfer of net positive charge into cells (Coady et al., 2007).
The major cotransported substrate for SLC5A8 likely varies from one tissue to another. In the colon, bacterial fermentation of unabsorbed carbohydrates and dietary fiber generates elevated levels of short chain fatty acids (SCFA), primarily acetate, proprionate and butyrate, all of which are SLC5A8 substrates in cellular models. SCFA are necessary for optimal colonic health and are thought to play a significant role in the prevention of gastrointestinal disorders, cancer and cardiovascular disease (Topping and Clifton, 2001; Wong et al., 2006). These effects result primarily from uptake and subsequent metabolism by colonocytes, although SCFA and their metabolites also target other tissues. Butyrate, in particular, is the major fuel for colonocyte metabolism, promotes colonocyte differentiation, and modulates colonic blood flow and electrolyte and water uptake. Acetate is the primary substrate for cholesterol synthesis in the liver, whereas proprionate is a substrate for hepatic gluconeogenesis and inhibits cholesterol synthesis.
SCFA are considered to be the primary substrates for colonic SLC5A8, however, recent data indicates that butyrate and proprionate transport in the colon is not altered in SLC5A8 knockout mice, probably reflecting dominant uptake by other transporters (e.g. SCFA/HCO3- exchange) or by non-ionic diffusion (Frank et al., 2008). Although this result may question the role of SLC5A8 in SCFA uptake in vivo, gene knockout may invoke compensatory mechanisms that mask the true physiological role of the gene product in question. In contrast, SLC5A8 knockout significantly attenuated lactate transport by colonic tissues, suggesting a role for SLC5A8 in intestinal lactate absorption, for example under pathological conditions of bacterial overgrowth leading to D-lactic acidosis (Frank et al., 2008).
Lactate is the preferred substrate for SLC5A8 in the kidney and salivary glands (Gopal et al., 2004; Frank et al., 2008); indeed, SLC5A8 knockout mice manifest higher urinary and salivary lactate concentrations compared to wild-type animals, suggesting that SLC5A8 contributes to lactate reabsorbtion by both organs (Frank et al., 2008). Renal SLC5A8 also mediates the reabsorption of nicotinate, the ionic form of nicotinic acid (vitamin B3), an essential vitamin for the normal function of all cells (Gopal et al., 2005), and of pyroglutamate, a byproduct of glutathione metabolism (Miyauchi et al., 2010).
Neuronal SLC5A8 contributes to the uptake of lactate and ketone bodies, used as an energy source in the brain under physiological and pathological conditions (Martin et al., 2006). Normally, lactate is the primary metabolic fuel for neurones and derives from the circulation or is generated from glucose by astrocytes. In contrast, ketone bodies are metabolic substrates for neurones during conditions of limited glucose availability such as pregnancy, starvation and uncontrolled diabetes. A similar role for SLC5A8 in the transport of lactate and ketone bodies has been proposed in the retina, in both neurons and retinal epithelial cells (Martin et al., 2007).
The preferred physiological substrate for SLC5A8 in the thyroid gland in unclear. SLC5A8 was first identified on the apical membrane of thyroid follicular cells, and was proposed to contribute to iodide flux into the thyroid lumen for incorporation into thyroglobulin, the precursor of thyroid hormones (Rodriguez et al., 2002). Subsequent studies, however, have shown that SLC5A8 does not transport iodide and SLC5A8 knockout mice have normal thyroid function, leaving the role of SLC5A8 in the thyroid gland an open question (Coady et al., 2004; Miyauchi et al., 2004; Paroder et al., 2006; Frank et al., 2008).
SLC5A8 methylation and silencing in cancer
Independent of its discovery as a solute carrier, SLC5A8 was also identified by Li et al. (2003) as a candidate tumour suppressor gene whose silencing by aberrant methylation is a common and early event in human colon neoplasia. Indeed, epigenetic modifications are a common feature of cancer and are thought to contribute to cancer initiation and progression (Sharma et al., 2010; Hatziapostolou and Iliopoulos, 2011). Epigenetic modifications (DNA methylation, histone acetylation and methylation, chromatin remodelling, and miRNA deregulation) interact with genetic alterations to disrupt gene function. In mammals, methylation of cytosine residues occurs at CpG dinucleotides concentrated in regions of large repetitive sequences, and in CpG islands located in gene promoters. CpG methylation of repetitive elements helps maintain genomic stability, whereas methylation of promoter CpG islands results in transcriptional silencing. In normal differentiated tissues, most CpG sites in the genome are methylated, whereas most gene promoter CpG islands are unmethylated. In contrast, the methylation landscape of the cancer genome is reversed, with global hypomethylation accompanied by hypermethylation of promoter CpG islands. Whereas global DNA hypomethylation increases genomic instability and activates proto-oncogenes, site-specific promoter hypermethylation contributes to tumorigenesis by silencing tumour suppressor genes.
SLC5A8 expression is suppressed in colon cancer, at both the transcriptional and protein level, and this effect is thought to be secondary to SLC5A8 promoter methylation (Li et al., 2003; Dong et al., 2005; Paroder et al., 2006; Thangaraju et al., 2008; Brim et al., 2011). Indeed, the SLC5A8 promoter region is unmethylated in the normal colon mucosa, and is frequently methylated in primary colon cancers, colon adenomas, and aberrant crypt foci (the earliest detectable morphologic abnormality of the colonic epithelium), suggesting that SLC5A8 promoter hypermethylation is an early event in colon tumorigenesis (Li et al., 2003).
Many colon cancer cells lines are also characterized by reduced SLC5A8 expression and promoter hypermethylation (Li et al., 2003). Expression is reactivated following treatment with the demethylating agent 5-azacytidine, confirming that he loss of gene expression is secondary to methylation. SLC5A8 expression is also restored by deletion of DNMT1, suggesting that methylation and therefore silencing is mediated by DNA methyltransferase-1 (Thangaraju et al., 2008).
SLC5A8 promoter methylation and gene silencing has also been demonstrated in various non-colonic neoplasms, including thyroid cancer (Lacroix et al., 2004; Porra et al., 2005; Hu et al., 2006; Schagdarsurengin et al., 2006), breast cancer (Thangaraju et al., 2006), gastric cancer (Ueno et al., 2004), brain cancer (Hong et al., 2005), prostate cancer (Park et al., 2007), pancreatic cancer (Park et al., 2008), head and neck squamous cell carcinoma (Bennett et al., 2008), and acute myeloid leukemia (Whitman et al., 2009). Furthermore, SLC5A8 expression is silenced in several non-colonic cancer cell lines and is restored by demethylating agents, suggesting that methylation-induced silencing of SLC5A8 is a common feature of many types of cancer.
SLC5A8 as a tumour suppressor
A role for SLC5A8 as a tumour suppressor was first suggested by the demonstration that ectopic expression of the gene in SLC5A8-deficient colon cancer cell lines reduces colony formation in vitro, but has no effect on the growth of SLC5A8-proficient cell lines; furthermore, cell lines with restored SLC5A8 expression have a reduced ability to form xenograft tumours in athymic mice (Li et al., 2003). These findings suggest that SLC5A8 methylation and silencing confers a specific growth advantage in the subset of colon cancers in which this locus is inactivated. SLC5A8 over-expression in a head and neck squamous carcinoma cell line also decreases colony growth, suggesting that SLC5A8 is a tumour suppressor at other cancer sites (Bennett et al., 2008).
Role of butyrate: The tumour suppressive function of SLC5A8 in the colon is thought to be secondary to the uptake of SCFAs, particularly butyrate, rather than a direct effect of SLC5A8 itself (Ganapathy et al., 2005; Gupta et al., 2006; Ganapathy et al., 2008). Butyrate is abundant in the colonic lumen (5-15 mM) as a result of bacterial fermentation of undigested organic matter, and is the major metabolic fuel for the colonic epithelium, vital for its normal growth and differentiation in vivo (Roediger, 1982). Butyrate has anticarcinogenic properties and has been shown to inhibit proliferation, and induce differentiation and apoptosis of cancer cells in vitro, including colorectal cancer cells (Kruh, 1982; Tsao et al., 1983; Augeron and Laboisse, 1984; Hague et al., 1993; Heerdt et al., 1994; Hague et al., 1995). Epidemiological studies have long demonstrated the protective effect of dietary fibre against colon cancer (Kim, 2000), and the generation of butyrate by bacterial fermentation is thought underlie this effect. Butyrate is a known inhibitor of histone deacetylase (HDAC), and as such regulates gene expression through epigenetic mechanisms involving the acetylation status of histones. HDAC inhibitors enhance the acetylation of lysine residues, weakening the interaction between histones and DNA, thereby facilitating transcription. HDAC inhibitors have been shown to cause growth arrest and apoptosis in a variety of tumours (Marks et al., 2001). Thus, the protective effect of dietary fibre against colon cancer is thought to be due, at least partially, to butyrate-mediated HDAC inhibition (Gupta et al., 2006). Other major SCFAs generated in the colon (proprionate and acetate) are less effective than butyrate in terms of HDAC inhibition and in terms of their anti-tumorigenic effects, consistent with the hypothesis that the protective effect of butyrate against colon cancer is related to its ability to inhibit HDAC (Hinnebusch et al., 2002). The role of SLC5A8 in tumour suppression in the colon by butyrate has been studied in vitro through ectopic expression of SLC5A8 in colon cancer cells in which SLC5A8 is completely silenced. Thus, re-expression of SLC5A8 in SLC5A8-silenced colon cancer cells induces apoptosis, but only when butyrate is present in the culture medium (Thangaraju et al., 2008). Furthermore, HDAC activity is high in SLC5A8-silenced colon cancer cells, and butyrate reduces HDAC activity only following SLC5A8 re-expression. These findings suggest that SLC5A8 per se without its transport function is not a tumour suppressor, but that SLC5A8/butyrate-induced apoptosis in tumour cells involves entry of butyrate into cells via SLC5A8 and subsequent inhibition of HDACs (Ganapathy et al., 2008). In contrast, animal studies using SLC5A8 knockout mice have failed to confirm the role of SLC5A8/butyrate in colon carcinogenesis. Treatment of SLC5A8 -/- knockout mice with carcinogens and breeding to the APCmin mouse line (which is highly susceptible to spontaneous intestinal adenoma formation) did not reveal a higher incidence of tumour formation, suggesting that SLC5A8 has no apparent role in the prevention of colon tumour formation and growth, at least in this model (Frank et al., 2008). However, butyrate transport by colonic tissues from SLC5A8 knockout mice was not impaired, suggesting that other pathways of butyrate uptake dominate, and may confer protection against tumorigenesis in these animals. Furthermore, butyrate has anti-proliferative and pro-apoptotic effects on a wide variety of cancer cells that do not express SLC5A8, suggesting that butyrate can enter cells and exert an anti-tumorigenic effect independently of SLC5A8.
Role of pyruvate: The expression of SLC5A8 in various normal tissues and its silencing in different cancers raises the possibility that other SLC5A8 substrates may be involved in tumour suppression outside the colon. In an attempt to identify alternative tumour suppressor SLC5A8 substrates, Thangaraju et al. (2006) focused their attention on pyruvate, a ubiquitous metabolite present in the circulation at concentrations of 100 uM, and a normal supplement of cell culture media. SLC5A8 is silenced in MCF-7 breast carcinoma cells through methylation. Ectopic expression of SLC5A8 in MCF-7 cells and exposure to pyruvate induces apoptosis and inhibits colony formation; in contrast, exposure of such cells to lactate, another SLC5A8 substrate, has no effect (Thangaraju et al., 2006). The apoptotic response of SLC5A8-expressing MCF-7 cells to pyruvate is accompanied by up-regulation of proapoptotic factors (p53, Bax, Bak, TRAIL, TRAILR1, and TRAILR2) and down-regulation of antiapoptotic factors (Bcl2, Bcl-W, and survivin), whereas, the expression of apoptosis-related genes is not affected by lactate. Pyruvate, but not lactate, inhibits HDAC activity with a similar potency as butyrate, supporting the hypothesis that tumour suppression by SLC5A8 is due to the uptake of substrates such as pyruvate that alter the expression of apoptosis-related genes by modifying the acetylation status of histones (Thangaraju et al., 2006).
Clinical and therapeutic implications
SLC5A8 expression correlates with survival in colon cancer suggesting a clinical utility as prognostic marker (Paroder et al., 2006). SLC5A8 protein expression is significantly reduced or absent in Duke C (locally advanced lymph-node-positive) colorectal cancer, irrespective of the differentiation status of tumours. Patients with low SLC5A8-expressing tumours show shorter disease-free and overall survival compared with patients with higher SLC5A8 expression, suggesting that SLC5A8 expression is a favourable indicator of colorectal cancer prognosis (Paroder et al., 2006).