Background
The prevalence of obesity and metabolic syndrome has increased at an alarming rate. By the year 2030, the number of adults with either type-1 or type-2 diabetes is estimated to be greater than 350 million [
1]. Adult onset type-2 diabetes (T2DM) constitutes over 90% of all diabetes cases and is characterized by insulin resistance, abnormal insulin secretion, or both. Of these cases, it is estimated that 16% of people have undiagnosed or poorly managed diabetes (NIDDK National Health Interview survey, 2007–2009).
It is well documented that Type-2 diabetes and hepatic steatosis are co-present [
2]. The incidence of non-alcoholic fatty liver disease (NAFLD) is prevalent in 40 to 70% of patients with T2DM [
3,
4]. This type of liver disease originates as hepatic steatosis, and can progress to non-alcoholic steatohepatitis (NASH), cirrhosis, and end stage liver failure [
5]. T2DM-related NAFLD is not fully understood, but it is known that leptin and insulin are important mediators in the progression of NAFLD [
6]. Leptin is a hormone secreted by adipocytes, which binds to the leptin receptor and increases partitioning of fatty acids towards oxidation instead of triacylglycerol formation [
7]. In mice and rats, leptin deficiency causes hyperphagia and obesity [
8]. Moreover, the lack of leptin action causes increased insulin secretion, which is hypothesized to cause insulin resistance in rodents and humans [
9]. Insulin resistance syndrome is hypothesized to cause NAFLD and augment progression to NASH [
10].
T2DM and hepatic steatosis are modeled by a variety of diet and genetically modified rodent models. Db/db mice (BKS.Cg-m +/+ Leprdb/J) mice possess a spontaneous diabetes (Db) mutation in the leptin receptor. Db/db mice are insulin resistant, hyperinsulinemic, hyperglycemic, glucose intolerant, and possess abnormal islet cell morphology [
11‐
13]. They become hyperinsulinemic from 10–14 days after birth; and exhibit significant weight gain with abnormally high triglycerides and low- and very low-density lipoproteins at 3 to 4 weeks of age. Hyperglycemia appears after 4–6 weeks of age.
Other mouse models of obesity, diabetes, and NAFLD exhibit altered transporter expression in liver and kidney [
14]. Transporters are membrane proteins, which facilitate chemical transport into and out of cells [
15]. Organic anion transporting polypeptides, organic anion transporters and organic cation transporters are often referred to as “uptake transporters”. They are predominantly localized to the basolateral membrane and extract chemicals from blood into hepatocytes (as reviewed by [
15,
16], but it should be noted that Oatps are localized to the luminal membrane in kidney [
17]. Transporters that are members of the ATP-binding cassette (Abc) superfamily facilitate efflux of chemicals out of cells; and include Multidrug resistance proteins (Abcbs), Multidrug resistance-associated proteins (Abcc), Bile salt-export pump (Abcb11), and Breast cancer resistance protein (Abcg2). In liver, Abcc2, Abcg2 and Abcbs are localized to the canalicular membrane and facilitate biliary excretion of chemicals. Abcc1, 3–6 are localized sinusoidally and/or basolaterally, and efflux chemicals from hepatocytes into blood. In kidney, organic anion and cation transporters contribute to renal clearance, along with organic anion transporting polypeptides and Abcc transporters for determining the urinary excretion of many endogenous chemicals and xenobiotics.
There is evidence in rodents and humans that obesity, NAFLD, and NASH may increase susceptibility to drug-induced liver disease (DILI) [
18] and exhibit altered excretion of acetaminophen [
19]. Early studies demonstrated that obese overfed rats, which display NAFLD, were more sensitive to acetaminophen (APAP)-induced liver toxicity [
18]. Other studies have demonstrated that obese rats exhibited increased furosemide-induced renal and hepatic toxicity [
20], as well as gentamicin-induced nephrotoxicity [
21]. More recently, studies documented higher serum and urinary levels of APAP glucuronide (APAP-G) in children with NAFLD, as compared to controls, after a single dose of APAP [
22].
Because obese and diabetic people comprise a significant portion of the population within the United States, there is a growing need to better predict drug clearance, DILI, adverse drug effects, and drug efficacy in this population. As transporters comprise a significant mechanism by which multiple drugs undergo hepatic and renal clearance, it is imperative to determine whether diabetes affects transporter expression. The purpose of this study was to compare drug transporter expression levels in normal and diabetic mice and illustrate that the disposition of a prototypical Abcc substrate is altered. The study herein thoroughly characterizes drug transporter expression in the db/db model, which can provide guidance for disposition/toxicology studies in diabetics. In the present study, transporter mRNA and protein expression was markedly changed in db/db mice, which exhibit a severe diabetes phenotype and NAFLD. Moreover increased excretion of APAP metabolites into urine was observed in db/db mice.
Discussion
The current study demonstrates that db/db mice, which are a widely used rodent model of diabetes with excessive weight gain and NAFLD, display profound alteration of transporter expression in both liver and kidney at the level of mRNA and protein expression. These observations are in agreement with [
14] and [
30]. Increased urine APAP-G and –S levels were also observed, which consistent with enhanced APAP-G disposition observed in other rodent steatosis models [
19]. Slco1a1 expression was markedly downregulated in livers and kidneys of db/db mice. As Slco1a1 mediates transport of wide variety of anionic, cationic, zwitterionic, as well as, neutral chemicals [
31], a significant decrease in Slco1a1 expression in liver and kidney could cause marked changes in pharmacokinetics and toxicity in the db/db mouse model. Along with Slco1a1, Slco1b2 protein expression was significantly decreased in livers of db/db female mice. In mice, Slco1a1, transports similar substrates as SLCO1A2, 1B1 and 1B3 in humans [
32]. As Ppar-α has a central role in the down regulation of Slco1a1 in mouse liver [
33,
34], and is upregulated in db/db liver, according to present study as well as previous findings [
35], it is possible that the observed downregulation is via a Ppar-α mediated mechanism. Also, as Fxr has been observed to be decreased in NALFD [
36], it is possible Fxr-dependent mechanisms regulate Slco expression. Fxr regulates mouse Slco1a1, 1a4 and 1a5 [
37]. Pxr also regulates Slco1a4 expression in mice [
38]. Similarly, human SLCO1B3 and 1A2 is regulated, in part, by FXR [
39]. However, db/db mice did not demonstrate any significant differences in mRNA expression of Fxr and Pxr in liver, suggesting that in the observed Slco decrease in Db/Db mice may be due to Ppar-α activation, and not Pxr and Fxr alterations. These observed changes in Slco expression in db/db mice could be predicative of SLCO expression changes in livers of diabetic humans. Further studies, which reveal nuclear receptor binding to specific response elements present in Slco promoters, will further elucidate how these transporters are regulated in leptin/leptin receptor deficient diabetes models.
The regulation of renal transporter expression in mouse models of diabetes and obesity remains limited. Data in this manuscript and Cheng et al. [
14] indicate that a severe diabetes phenotype alters renal transporter expression. It is intriguing that kidney transporter expression was substantially altered in this model, but minimal changes in renal pathology were observed. In humans SLC22A6 and SLC22A7 are predominant transporters localized to the basolateral membrane of renal proximal tubule cells [
40]. The SLCs transport certain antibiotics like benzylpenicillin, antivirals and NSAIDs (Non-steroidal anti-inflammatory drugs). Slc22a7 expression was virtually undetectable in db/db male and female mice as compared to respective C57BKS controls, indicating the possibility of different renal elimination of substrates such as antibiotics, antivirals and non-steroidal anti-inflammatory drugs in this model. Slc22a6 and Slc22a2 expression was also downregulated in db/db mice, especially males. The mechanism for the observed Slc downregulation was not determined, however HNF1 has been described to regulate human and mouse SLC22A7/Slc22a7 and HNF4 has been described to regulate SLC22A7 in kidney [
41,
42].
Efflux transporters, in general, were upregulated in livers of db/db mice. Abcc3 transports mono-ionic bile acids such as glycocholate and taurocholate [
43], as well as glucuronide or glutathione conjugates of certain drugs (e.g. APAP-G and morphine-3-glucuronide) [
44]. Abcc3 and 4 expressions were significantly upregulated in db/db mice livers, in both genders. Abcc4 also transports bile acids, antiviral drugs, and cyclic nucleotides [
15], but also contributes to the basolateral excretion of APAP-S [
45,
46]. Reisman et al. demonstrated increased plasma APAP-G and APAP-S concentrations correspond with increased Abcc3 and 4 protein expression, respectively [
47]. Additionally, in a rat model of NASH, it was observed that increased Abcc3 expression enhanced urinary excretion of APAP-G [
19]. Increased expression of Abcc3 and/or Abcc4 is associated with enhanced excretion of APAP metabolites [
19,
48]. In the present study, db/db mice had higher amounts of APAP-G and -S metabolites in urine, which was consistent with increased hepatic Abcc3 expression, and increased hepatic and renal Abcc4 expression. The reasons for higher excretion of APAP-G and APAP should be due to enhanced production of APAP-G and –S and/or enhanced basolateral excretion. Db/db mice also display increase in mRNA expression of the enzymes responsible for production of major conjugation metabolites like Ugt1a6 and Sult1a1 compared to C57BKS mice livers (Figure
8). Therefore, enhanced excretion of glucuronide and sulfate metabolites was expected. Overall, this data is consistent with published findings in children with NAFLD [
22]. Increased APAP-G levels were observed in plasma and urine samples from children presenting with NAFLD [
22]. Abcc1, 2, 4, and Abcg2 mRNA and/or protein expression was increased in liver, which is consistent with what was observed in livers of T2DM rats [
49]. Abcc1 and Abcg2, along with Abcb1, can transport the antidiabetic drug rosiglitazone [
50]. Severe liver injury has been reported in a person with T2DM [
51] and cholestatic injury has also been observed after rosiglitazone therapy [
52] – both suggesting hepatic clearance is necessary. Perhaps, differences in expression of these transporters in the diabetic liver could contribute to decreased hepatic clearance of rosiglitazone. An interesting observation is that rosiglitazone increases the incidence of cardiovascular disease in diabetic patients [
53]. As its use is still approved, determining whether diabetes could impede rosiglitazone clearance is important for predicting persons at risk.
The transporters analyzed in this study are known to be regulated by different mechanisms, involving various transcription factors such as Ppar-α, Pxr, constitutive androstane receptor (Car), nuclear factor E2-related factor 2 (Nrf2), Fxr, and Hepatocyte nuclear factor 1-alpha (Hnf-1α). Li and Klaassen (2004) showed that HNF1α levels are critical for constitutive expression of Slco1b2 in mouse liver [
54]. Also Slc22a6 and Slc22a7 expression in mouse kidneys is downregulated by targeted disruption HNF1α [
55]. Significantly reduced expression of Slco1a1 in liver, along with Slc22a7 in kidney in db/db mice suggests that HNF1α levels or binding is decreased in these mice. Similarly, Abcc3 and Abcc4 efflux transporter expression is regulated in part by Nrf2-keap1 pathway in liver [
24]. The present study clearly demonstrates that Abcc2-4 were upregulated in livers of db/db mice, which suggests activation of the Nrf2 and/or constitutive androstane pathways in these mice. Increased mRNA expression of Nrf2 and its target gene Gclc indicate that Nrf2-keap1 pathway is likely activated in db/db mice. The Nrf2-keap1 pathway is activated during periods of oxidative stress [
56]. Also as reviewed by Rolo and Palmeira, diabetes is typically accompanied by increased production of free radicals, present findings suggests that oxidative stress may be present in diabetic liver [
57]. Together, the data presented argue for additional future studies to better define nuclear receptor pathways that are upregulated in leptin/leptin receptor deficient models, which will aid in better understanding receptor-mediated mechanisms, which could regulate transporter expression in steatosis and T2DM. As reviewed by Klaassen and Slitt [
38], Car and Pxr are also known for regulating Abcc2, 3, 5, 6 and Abcc2, 3 respectively. The observed increase in Abcc2, 3, 5, and 6 expression could be attributed to the observed increased in Car expression and activity, as shown in Figure
7.
Similar to the liver, transporter expression is markedly altered in kidneys of db/db mice. Maher and colleagues showed that targeted disruption in Hnf1α significantly downregulated Slc22a6, 7 and 8 and Slco1a1 mRNA in mice kidneys [
55]. This indicates that db/db mice might have differential expression or binding of Hnf1α. Also, these mice have severe hyperglycemia. During normal course, almost all of the glucose is absorbed from the nephrons during urine formation. But due to overwhelming amounts of glucose in glomerular filtrate, kidneys are unable to absorb it and thus excrete glucose in urine. This hyperglycemic urine may cause some alterations in transporter expression in kidneys.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
VRM performed all experiments with mRNA and protein expression and immunohistochemistry, and drafted the manuscript. XW analyzed urine samples for APAP and metabolites. PET developed method for APAP analysis by HPLC. ALS, LMA and VRM designed the experiment, and contributed to writing of manuscript. All authors read and approved the final manuscript.