Abstract

Nuclear receptors (NRs) comprise one of the most abundant classes of transcriptional regulators of metabolic diseases and have emerged as promising pharmaceutical targets. Small heterodimer partner (SHP; NR0B2) is a unique orphan NR lacking a DNA-binding domain but contains a putative ligand-binding domain. SHP is a transcriptional regulator affecting multiple key biological functions and metabolic processes including cholesterol, bile acid, and fatty acid metabolism, as well as reproductive biology and glucose-energy homeostasis. About half of all mammalian NRs and several transcriptional coregulators can interact with SHP. The SHP-mediated repression of target transcription factors includes at least three mechanisms including direct interference with the C-terminal activation function 2 (AF2) coactivator domains of NRs, recruitment of corepressors, or direct interaction with the surface of NR/transcription factors. Future research must focus on synthetic ligands acting on SHP as a potential therapeutic target in a series of metabolic abnormalities. Current understanding about the pleiotropic role of SHP is examined in this paper, and principal metabolic aspects connected with SHP function will be also discussed.

1. Introduction

Nuclear receptors (NRs) constitute a unique family of ligand-modulated transcription factors. NRs mediate cellular response to small lipophilic endogenous and exogenous ligands [1, 2] and are responsible for sensing a number of hormones, including steroid and thyroid hormones, and act as positive and negative regulators of the expression of specific genes [35]. Therefore, NRs play a central role in many aspects of mammalian development, as well as lipid homeostasis, physiology, and metabolism. NRs make up one of the most abundant classes of transcriptional regulators in the body and have emerged as promising pharmaceutical targets.

Classically, NRs consist of several functional domains, that is, a variable N-terminal ligand-independent transactivation domain (which often exhibits a constitutive transcription activation function (AF-1)), a highly conserved DNA-binding domain (DBD) that contains two zinc fingers, a hinge domain (a variable linker region), and a multifunctional C-terminal domain. Furthermore, the C-terminal domain includes the ligand binding (LBD), the dimerization interface, and the ligand-dependent transactivation domain AF-2 [1, 6].

Small heterodimer partner (SHP; NR0B2 for nuclear receptor subfamily 0, group B, member 2; MIM number 604630, 601665) is a member of the mammalian NR superfamily, due to the presence of a putative ligand-binding domain (LBD) [7]. SHP functions as a corepressor through heterodimeric interaction with a wide array of nuclear receptors and repressing their transcriptional activity. SHP achieves its goal via several members of the NR superfamily that are able to regulate SHP expression. However, SHP is also a unique and atypical NR because it lacks the classical DNA-binding domain (DBD), generally present in other NRs [8]. The NR0B family of NRs consists of 2 orphan receptors: SHP and DAX-1 (dosage-sensitive sex reversal adrenal hypoplasia congenita (AHC) critical region on the X chromosome, gene 1). DAX1 is a gene whose mutation causes the X-linked adrenal hypoplasia congenita [9] and is the only family member that lacks a conventional DBD. DAX-1 (NR0B1) is therefore seen as the closest relative of SHP in the NR superfamily [1012]. Both SHP and DAX-1 appear to be specific to vertebrates. In this respect, no homologous genes have been found in Drosophila melanogaster or Caenorhabditis elegans [12]. Whereas SHP is different from other conventional NRs both structurally and functionally, it acts as a ligand-regulated receptor in metabolic pathways [13]. SHP belongs to the orphan subfamily since there is no known ligand for this receptor, except for some retinoid-related molecules [14]. SHP inhibits transcriptional activation by working on several other nuclear receptors, that is, directly modulating the activities of conventional nuclear receptors by acting as an inducible and tissue-specific corepressor [12, 15]. The discovery of SHP dates back to 1996 [10]; since then, this orphan NR has been identified as a key transcriptional regulator of signaling pathways [8, 16] involving fundamental biological functions and metabolic processes. Such processes include cholesterol, bile acid and fatty acid metabolism, glucose and energy homeostasis, and reproductive biology [17]. Experiments performed by fluorescence in situ hybridization (FISH) analysis of the human metaphase chromosome have shown that SHP is found at a single locus on chromosome 1 at position 1p36.1 and consists of two exons and a single intron spanning approximately 1.8 kb with 257 amino acids in humans [18]. In mice and rats, SHP resides on chromosomes 4 and 5, respectively, both consisting of 260 amino acids. SHP expression is predominantly observed in the liver [10, 18], but it is also detected at lower levels in other tissues, including the pancreas, spleen, small intestine, colon, gallbladder, kidney, adrenal gland, ovary, lung, brainstem, cerebellum, heart, and thymus (Table 1) [1921].

The genomic structure and human SHP domain structure are depicted in Figure 1 [15]. SHP is indeed able to repress the transcriptional activities of its target NRs and transcriptional regulators through two functional Leu-Xaa-Xaa-Leu-Leu- (LXXLL-) like motifs [2224]. Such motifs appear to be essential for the interaction with the (activation function 2) AF-2 domains of several sets of NRs [22, 23]. The human SHP is enriched by another 12 amino acids [2536], and this region between helix 6 and 7 is also involved in the repression of the transactivation of NRs [37].

About half of all mammalian NRs and several transcriptional coregulators can interact with SHP [12]. Since SHP lacks DNA-binding domain, it exerts the inhibitory effects through protein-protein interaction [10]. SHP expression seems to follow a circadian rhythm in the liver, involving the CLOCK-BMAL1 pathway and suggesting that some of the regulatory functions of SHP and deriving functions must be temporal [19, 20, 38].

Gene expression of SHP is regulated by several factors including NRs, transcription factors, and a number of additional conditions and substances, as extensively reported in Table 2. Also, the central role of SHP is clear since this NR is able to act as a coregulator for wide range of targets, namely, NRs/transcription factors/transcriptional coregulators and few different molecules, as depicted in Table 3. In general, SHP acts as a repressor of the transcriptional activity of the specific interacting partner (via LBD of the partner and NR boxes of SHP) [12, 3943]. However, it is also demonstrated that SHP is able to upregulate gene transcription, as in the case of PPARα and PPARγ [4446] and NF-κB [44].

Both N-terminal NR interaction domain and C-terminal domain of SHP are important for repression [47, 48]. Overall, the SHP-mediated repression of target transcription factors occurs by at least three distinct transcriptional repression mechanisms (Figure 2).

A first mechanism involves direct interference with the AF-2 coactivator domain of NRs (competition for coactivator binding, leading to the repression of NR-mediated transcriptional activity). This is the case for the inhibition of estrogen receptor α (ERα) and estrogen receptor β (ERβ) [49].

A second mechanism for the SHP-mediated repression involves the recruitment of corepressors including direct interactions among mammalian homolog of the Saccharomyces cerevisiae transcriptional corepressor Sin3p (mSin3A), human Brahma (Brm), SWItch/Sucrose NonFermentable (SWI/SNF) complexes leading to the repression of cholesterol 7α-hydroxylase (CYP7A1) [50]).

A third mechanism of inhibition of SHP involves the direct interaction with the surface of NR or transcription factor, resulting in the blockade of DNA binding and the consequent inhibition of its transcriptional activity. This is the case for RAR-RXR heterodimers [10], PXR-RXR binding to DNA by SHP [1], interaction with hepatocyte nuclear factor (HNF4), or Jun family of the activator protein 1 (AP-1) transcription factor complex (JunD) [51, 52].

All three mechanisms might occur sequentially or alternatively according to type of cells and promoters [12].

Clearly, information on factors that increase or decrease SHP expression and that are regulated by SHP is essential for understanding the regulatory effects of this orphan NR. Few years of research have not been enough to identify a true ligand. Interestingly, it is suggested that targeting posttranslational modifications of SHP may be an effective therapeutic strategy. Selected groups of genes could be controlled to cure a vast range of metabolic and SHP-related diseases [53]. Overall, the huge amount of information on SHP function is currently available, making this NR essential in a number of functions involving cholesterol and bile acid metabolism, lipogenesis, glucose metabolism, steroid hormone biosynthesis, xenobiotic homeostasis/metabolism, and cell cycle.

In particular, the ability of SHP in interacting with different metabolic signaling pathways including bile acids and lipid homeostasis, fat mass, adipocytes, and obesity will be reviewed here.

2. Bile Acids and Lipid Homeostasis

The wide ability of SHP to target multiple genes in diverse signaling pathways points to the key role of SHP in various biological processes, including the metabolism of bile salts, glucose, and fatty acids. Both unique structure and functional properties account for the complexity of SHP signaling. Studies suggest that loss of SHP might positively affect cholesterol and bile acid homeostasis in pathophysiologically relevant conditions [54]. Bile acids (BAs) are amphipatic cholesterol metabolites which are synthesized in the liver, secreted into bile, stored in the gallbladder, and secreted postprandially into the duodenum. BAs are synthesized from cholesterol, and this pathway provides the elimination of excess cholesterol in the body [55]. Moreover, BAs should be seen as physiological detergents which, in the small intestine, are essential for the absorption, transport, and distribution of lipophilic molecules, including dietary lipids, steroids, and lipid-soluble vitamins. In the intestine, BAs undergo extensive metabolism by the intestinal microflora. A high efficient system is the enterohepatic circulation of BAs [55, 56], where more than 90–95% of BAs are returned to the liver from the terminal ileum via the portal vein. Thus, the concentration of BAs in serum, liver, and intestine is tightly regulated to prevent damage to enterohepatic tissues due to their strong detergent moiety [5759]. The major rate-limiting step in biosynthetic pathway of BAs in humans is initiated by cholesterol 7α-hydroxylase (CYP7A1), the microsomal P450 liver enzyme, to produce two primary BAs, cholic acid, and chenodeoxycholic acid, essential in the overall balance of cholesterol homeostasis. Sterol 12α hydroxylase (CYP8B1) catalyzes the synthesis of cholic acid, a step which determines the cholic acid to CDCA ratio in the bile [60]. Secondary bile acids (deoxycholic acid and lithocholic acid) and tertiary bile acids (ursodeoxycholic acid) in humans are produced following intestinal dehydroxylation of primary bile acids by intestinal bacteria [58, 61].

Regulation of BA biosynthesis is highly coordinated and is mediated by key NRs including the orphan receptor, liver receptor homologue-1 (LRH1; NR5A2), the hepatocyte nuclear factor 4α (HNF4α), SHP, and the bile acid receptor farnesoid X receptor (FXR; NR1H4). Thus, the activation of FXR initiates a feedback regulatory loop via induction of SHP, which suppresses LRH-1- and HNF4α-dependent expression of the two major pathway enzymes cholesterol 7hydroxylase (CYP7A1) and sterol 12 hydroxylase (CYP8B1).

The BA feedback regulation primarily occurs since BAs act as transcriptional regulators for the expression of the gene encoding CYP7A1. Both cholic acid and chenodeoxycholic acid function as endogenous ligands for the nuclear bile acid receptor FXR [62]. FXR expression is high in the intestine and liver, the two sites where BAs reach high concentrations to activate FXR. The transcription by FXR includes heterodimerization with retinoid X receptors (RXRs) in the cytoplasm, translocation into the nucleus, and binding to DNA response elements in the regulatory regions of target genes [63]. When the bind of BAs to FXR, SHP transcription is increased [60, 64, 65], this alteration leads to the inhibition of LRH-1 activity or HNF4α on the BA response elements (BAREs) of CYP7A1 and CYP8B1 promoters [64, 65]. In this scenario, BA synthesis is downregulated by a precise feedback regulatory mechanism, which represents the major pathway under normal physiological conditions [6466] (Figure 3). LRH1 is also a well-known activator of Shp gene transcription [64, 65], and this step leads to an autoregulatory loop of gene expression by SHP [42]. This step also includes the G protein pathway suppressor 2 (GPS2) interacting with FXR, LRH-1, and HNF4α to regulate CYP7A1 and CYP8B1 expression in human hepatocytes [67] (Table 3). A critical role in maintaining cholesterol homeostasis for CYP7A1 has been recently advocated in a model of in Cyp7a1-tg mice [68].

The hepatocyte nuclear factor-1α (HNF1α), which haploinsufficiency causes the Maturity-onset diabetes of the young type 3 (MODY3), also appears to modulate SHP expression via the FXR pathway. In this respect, HNF1α (−/−) mice displayed a defect in bile acid transport, increased bile acid and liver cholesterol synthesis, and impaired HDL metabolism [69].

A role for SHP in mediating the recruitment of mSin3A-Swi/Snf to the CYP7A1 promoter, with chromatin remodeling and gene repression, has been described. In HepG2 cells, Kemper et al. [50] have shown that bile acid treatment resulted in SHP-mediated recruitment of transcriptional coregulators mSin3A and Swi/Snf complex to the promoter, chromatin remodeling, and gene repression (Table 3). This is an additional mechanism involving transformation of nucleosome conformation for the repression by SHP of genes activated by various NRs. In line with such results, increased synthesis and accumulation of BAs occurs in SHP (−/−) mice, due to the loss of SHP repression and consequent derepression of the rate-limiting CYP7A1 and cholesterol 12α-hydroxylase (CYP8B1) (the rate-determining enzyme of the alternative but minor BA synthesis pathway) in the biosynthetic pathway [7072].

Mechanisms independent of the FXR/SHP/LRH pathway might also exist, since BAs feeding to SHP (−/−) mice reduced the levels of CYP7A1 mRNA to similar levels of control mice [70, 71]. Such SHP-independent and alternative pathways include the protein kinase C/Jun N-terminal kinase (PKC/JNK) pathway [73], the FXR/FGFR4 (FGF receptor 4) pathway [57, 74], the cytokine/JNK pathway [75], the pregnane X receptor (PXR) mediated pathway [76], and the JNK/c-Jun signaling pathway [77].

Another study demonstrated, in SHP (−/−) mice on a background of 129 strain, the protection against hypercholesterolemia in three different models: an atherogenic diet, hypothyroidism, and SHP (−/−) mice intercrossed with LDLR (−/−) mice (to generate SHP/LDLR double (−/−) mice in a mixed 129-C57BL/6 background). When fed an atherogenic diet, the latter strain was almost completely resistant to diet-mediated increases in triglyceride, very low-density lipoprotein (VLDL) cholesterol, and low-density lipoprotein (LDL) cholesterol but had an increase in high-density lipoprotein (HDL) cholesterol as compared with LDLR (−/−) mice. Such results point to the protection against dyslipidemia following the inhibition of hepatic SHP expression, although no antagonist ligands have yet been identified for SHP [78]. We have recently examined biliary lipid secretion and cholesterol gallstone formation in male SHP (−/−) and (+/+) mice before and during the feeding of a lithogenic diet for 56 days [79]. Deletion of the Shp gene significantly increased hepatic bile salt synthesis, and doubled the increase of biliary bile salt outputs in SHP (−/−) mice than in (+/+) mice. The intestinal bile acid pool size was significantly greater in SHP (−/−) mice than in (+/+) mice. These increased BAs are efficacious ligands of FXR and can stimulate the expression of intestinal fibroblast growth factor 15 (FGF15) in mice through the FXR signaling pathway, which is consistent with the expanded bile acid pool size in SHP (−/−) mice. At 14 days on the lithogenic diet, fasting gallbladder volume was significantly larger in SHP (+/+) mice than in (−/−) mice [80].

Indeed, FGF15/19 (mouse and human orthologs, resp.) is another FXR gene target in the intestine and appears to contribute to the fine tuning of bile acid synthesis in the liver. Thus, a model for FXR-mediated repression of bile acid synthesis should also take into account the bile acid-mediated activation of intestinal FXR and FGF15 in the small intestine (while the FXR-SHP pathway is activated in the liver). According to the most plausible view, FGF15 acts as a hormone to signal between intestine and liver. The secreted FGF15 by the intestine circulates to the liver, likely through the portal circulation or lymph flow [81], and induces the activation of FGFR4 in the liver. As shown in Figure 3, the FGF15/FGFR4 pathway synergizes with SHP in vivo to repress CYP7A1 expression [57]. In humans, a similar mechanism should involve the FGF19. Of note, activation of FXR transcription in the intestine protected the liver from cholestasis in mice by inducing FGF15 expression and reducing the hepatic pool of BA. This suggests a potential approach to reverse cholestasis in patients [82]. Hepatic fatty acid homeostasis is also regulated by SHP since regulating these genes involves in fatty acid uptake, synthesis, and export [8387]. In a study exploring global gene expression profiling combined with chromatin immunoprecipitation assays in transgenic mice constitutively expressing SHP in the liver, overexpression of SHP in the liver was associated with the depletion of the hepatic bile acid pool and a concomitant accumulation of triglycerides in the liver [84]. By contrast, fat accumulation induced by a high-cholesterol or high-fat diet is prevented by the deletion of SHP [88, 89]. The pleiotropic role of SHP can also be found in the case of nonalcoholic liver steatosis since OB/SHP double (−/−) mice (a model of severe obesity and insulin resistance) became resistant to liver steatosis and showed improved insulin sensitivity [86].

Another interesting role for SHP emerged after it was found that BAs negatively regulate the human angiotensinogen (ANG) gene. ANG is the precursor of vasoactive octapeptide angiotensin II, and BAs act through the SHP pathway by preventing hepatocyte nuclear factor-4 (HNF4) from binding to the human ANG promoter [90].

3. Fat Mass, Adipocytes, and Obesity

SHP appears to play a central role in obesity. Human obesity is considered a polygenic disorder characterized by partly known abnormal molecular mechanisms resulting in increased fat mass, with an imbalance between the energy acquired from nutrients that dissipated as heat (i.e., thermogenesis). In this respect, weight stability requires a balance between calories consumed and calories expended [91]. In adipose tissue depots, two main types of adipocytes exist, that is, brown adipocytes and white adipocytes. In several animal species, some adipose tissue sites mainly include brown adipocytes (BATs) and the other contains mainly white adipocytes (WATs). BAT dissipates chemical energy to produce heat either as a defense against cold [92] or as energy expenditure to compensate food intake [93, 94]. The unusual function of BAT might be better understood by considering that they share a common origin with myocytes [95, 96], and BAT was indeed considered something in between muscle and adipose tissue [95]. BAT is deemed as the major site for sympathetic (adrenergic) mediated adaptive thermogenesis; this pathway involves the uncoupling protein-1 (UCP1). WAT is mainly implicated in the regulation of lipid storage and catabolism but also in the synthesis and secretion of adipokines [97100]. While the percentage of young men with BAT is high, the activity of BAT is reduced in men who are overweight or obese [101]. Thermogenesis unequivocally exists in both humans and animals, and BAT is the major site of thermogenesis which can be increased by environmental factors (i.e., adaptive thermogenesis). In both human and animal species, dietary composition, chronic cold exposure, and exercise may increase thermogenesis [102]. As far as adipose tissue biology is concerned, SHP seems to play a distinct regulatory function in WAT, as compared with BAT. A number of experiments have focused on animal models of obesity and subtle molecular changes. SHP-deficient mice are protected against high-fat-diet-induced obesity [89].

Peroxisome proliferator-activated receptor (PPAR) γ coactivator-1 (PGC-1) family members are multifunctional transcriptional coregulators. PGC-1 acts as a molecular switch in several metabolic pathways. In particular, PGC-1α and PGC-1β regulate mitochondrial biogenesis, adaptive thermogenesis, fatty acid and glucose metabolism, fiber-type switching in skeletal muscle, peripheral circadian clock, and development of the heart [103]. In particular, SHP functions as a negative regulator of energy production in BAT [89] because SHP is a negative regulator of PGC-1α expression in BAT. In turn, PCG-1α is a coactivator of uncoupling protein 1 (UCP1) which plays a major role in energy dissipation as heat in multilocular BAT of different animal species and humans [104106]. Fat-specific (BAT) SHP-overexpressed transgenic mice had increased body weight and adiposity. Energy metabolism, however, was increased, and BAT cold exposure function was enhanced with activation of thermogenic genes and mitochondrial biogenesis (enhanced β1-AR gene expression and PGC1α). Compared with wild-type mice on a high-fat diet, SHP overexpression was associated with enhanced diet-induced obesity phenotype with weight gain, increased adiposity, and severe glucose intolerance. An additional feature of SHP transgenic mice was a decreased diet-induced adaptive thermogenesis, increased intake of food, and decreased physical activity [107]. This leads to the conclusion that, although expressed at low levels in fat, activation of SHP in adipocytes has a strong effect on weight gain and diet-induced obesity [107]. Moreover, if mechanisms linked to energy metabolism and the development of obesity are considered, SHP has distinct roles in WAT and BAT. As previously mentioned, while SHP deletion in obese leptin-deficient mice (ob/ob) prevented the development of nonalcoholic fatty liver and improved peripheral insulin sensitivity [86], SHP deletion did not overcome the severe obesity caused by leptin deficiency. A significant protective effect from obesity by SHP deficiency was likely associated with the low basal level of SHP expressed in fat. Adipogenesis appears to be influenced by SHP: when SHP was overexpressed in 3T3-L1 preadipocytes, cell differentiation was inhibited, as well as the accumulation of neutral lipids within the cells. Thus, SHP may act as a molecular switch governing adipogenesis. In particular, SHP appears to be a potent adipogenic suppressor, and preadipocytes are kept in an undifferentiated state through the inhibition of the adipogenic transcription factors and stimulators [108]. Further studies will address whether the loss of SHP function results in inhibition of lipid accumulation in adipocytes, similar to what is observed in hepatocytes. In a future clinical setting, treatment of obesity might also include drugs able to mimic or stimulate the effects of SHP. Mutations in the Shp gene have also been reported in patients with lipodystrophy carrying four different polymorphisms [109].

SHP mutations may not be considered a common cause of severe obesity. A number of important clinical studies have examined this issue (Table 4); however, Hung et al. [110] in UK examined the relationships between genetic variation in SHP and weight at birth, adiposity, and insulin levels in three different populations (the Genetics of Obesity Study) GOOS, the Avon Longitudinal Study of Parents and Children (ALSPAC), and the Ely studies). In the 329 cases of severe early-onset obesity (GOOS study), two novel and rare missense mutations (R34G and R36G) were identified which might in part contribute to obesity in the probands. Furthermore, two common polymorphisms, namely, G171A (12% of subjects with higher birth weight) and −195CTGAdel (16% of subjects with lower birth weight) were found. In the ALSPAC cohort of 1,079 children, the G171A variant was associated with increased body mass index and waist circumference together with higher insulin secretion 30 minutes after glucose load. Thus, whereas mutations in the Shp gene cannot be seen as a common cause of severe human obesity, genetic variation in the Shp gene locus may influence birth weight and have effects on body size. The effect might ultimately involve insulin secretion by the negative regulation between SHP and the hepatocyte nuclear factor-4α (HFN-4α), a transcription factor involved in differentiation and function of pancreatic β-cells [110].

A possibility is that decreased SHP expression or function results in increased HFN-4α activity with a cascade of events, including fetal hyperinsulinemia, and increased birth weight. At a later stage, sustained hyperinsulinemia might be responsible of insulin resistance and obesity of the adult [110].

Mutations in the Shp gene were also associated with influence on birth weight, mild obesity, and insulin levels in the study by Nishigori et al. on 274 Japanese subjects [111]. Mutations in several genes encoding transcription factors of the hepatocyte nuclear factor (HNF) cascade are associated with maturity-onset diabetes of the young (MODY). MODY is a monogenic form of early-onset diabetes mellitus (defective insulin secretion with normal body weight), and SHP is deemed as a plausible candidate MODY gene; this is because SHP is able to inhibit the transcriptional activity of the hepatocyte nuclear factor-4α (HFN-4α), a key member of the MODY regulatory network. Thus, further studies have looked for segregation of SHP mutations with MODY in a cohort of Japanese patients with early-onset diabetes. In this context, variants in SHP appeared to cosegregate with increased body mass index in families, thus contributing to obesity among Japanese subjects. Also, increased risk of morbidity was observed in another study from Japan, examining patients with type 2 diabetes and SHP mutations [112].

Major differences, however, might exist in the prevalence and function of SHP variants in different populations. Of note, the results from other Caucasian cohorts did not confirm the association between SHP mutation and obesity [113, 114]. Echwald et al. conducted an elegant study on the prevalence of SHP variants by single-strand conformational polymorphism and heteroduplex analysis among 750 Danish obese men with early-onset obesity [114]. As control, a cohort of 795 nonobese control subjects was genotyped using PCR-RFLP. Functional analyses of the identified coding region variants were performed in both MIN6-m9 and HepG2 cell lines. Five novel variants were identified (including 3 missense variants (c.100C>G [p.R34G], c.278G>A [p.G93D], and c.415C>A [p.P139H]) and 2 silent variants (c.65C>T [p.Y22Y] and c.339G>A [p.P113P])). The previously reported [111] c.512G>C [p.G171A] common polymorphism was identified; however, the prevalence of functional SHP variants associated with obesity was considerably lower among Danish subjects (1 out of 750 obese, none of control subjects), compared to the prevalence observed in Japan by Nishigori et al. [111]. Mitchell at al. [113] investigated SHP variants in 1927 UK subjects according to type 2 diabetes, obesity, and birth weight. Although reporting a raised body mass index among homozygous carriers of the 171A variant (<1%), this polymorphism was unlikely to be associated with all three conditions in Caucasians. Taken together, the above-mentioned studies suggest that the 171A variant might contribute only to subsets of polygenic obesity.

4. Other Functions of SHP

The existence of multiple interactions of SHP with NRs, transcription factors and transcriptional cofactors (Tables 2 and 3) points to the pleiotropic and central role of SHP in the body.

SHP has been hypothesized to act in glucose homeostasis via complex pathways involving the inhibition of glucocorticoid receptors (GR) in mammalian cells and the inhibition of PGC-1 gene, a coactivator of NRs important for gluconeogenic gene expression and the PGC-1-regulated phospho(enol)pyruvate carboxykinase (PEPCK) promoter. Such steps underscore a physiologically relevant role for SHP in modulating hepatic glucocorticoid action [22]. Following the bile acid-induced induction, SHP inhibited a number of other pathways, including the HNF4α-mediated transactivation of the PEPCK and fructose biphosphate (FBP) promoters, as well as the transactivation of the glucose-6-phosphatase (G6Pase) promoter mediated by Foxo1 [115]. The interaction between SHP inhibitory function and the 3 isoforms (α, β, and γ) of the hepatocyte nuclear factor-3 (HNF4) points to the regulatory role of SHP on gluconeogenesis [51]. A role for SHP in insulin secretion pathway has also been reported. Mutations in hepatocyte nuclear factor 1α (HNF-1α) is associated with maturity-onset diabetes of the young type 3. This condition depends on impaired insulin secretory response in pancreatic beta cells.

Indeed, loss of HNF-1α function in HNF-1α (−/−) mice resulted in altered expression of genes involved in glucose-stimulated insulin secretion, but also insulin synthesis, and beta-cell differentiation. Pancreatic islets of HNF-1α (−/−) mice showed a distinctive reduction of SHP expression and a downregulation of the HNF4α gene expression. Since SHP appears to repress its own transcriptional activation following heterodimerization with HNF4α, a feedback autoregulatory loop between SHP and HNF4α has been hypothesized [116]. Also, SHP likely functions as a negative regulator of pancreatic islet insulin secretion. SHP (−/−) mice were characterized by hypoinsulinemia, increased glucose-dependent response of islets, increased peripheral insulin sensitivity, and increased glycogen stores [117]. The role played by SHP in the regulation of hepatic gluconeogenesis has also emerged in a number of additional experiments. For example, the liver of SHP (−/−) mice showed increased glycogen stores [117], while hepatic Shp gene expression (induced by the antidiabetic biguanide drug metformin) was associated with inhibition of hepatic gluconeogenesis. Induction of SHP was achieved via AMP-activated protein kinase (AMPK) and associated with downregulation of essential gluconeogenic enzyme genes, that is, phosphoenolpyruvate carboxykinase (PEPCK), glucose-6-phosphatase (G6Pase) [118], and fructose-1,6-bisphosphatase (FBP1) [119].

PGC-1 gene is a coactivator of NRs, and this step is relevant for gluconeogenic gene expression. Yamagata et al. [119] showed that bile acid (chenodeoxycholic acid) was able to induce the downregulation of PGC-1 gene, and this mechanism involved forkhead transcription factors (Foxo1, Foxo3a, Foxo4) via a SHP-dependent manner.

Drug metabolism and detoxification might be regulated by SHP. This is also the case for excess BAs: the pregnane X receptor (PXR) induces CYP3A and inhibits CYP7α, both involved in biochemical pathways leading to the conversion of cholesterol into primary BAs, whereas CYP3A is also involved in the detoxification of toxic secondary bile acid derivatives. SHP acts as a potent repressor of PXR transactivation, and this finding suggests that PXR can act on both bile acid synthesis and elimination detoxification [1]. Additional mechanisms involved in the SHP-dependent control of pathways of drug metabolism have been identified. The expression of genes involved with the metabolism of xenobiotics might be regulated by SHP in the spleen acting on (aryl hydrocarbon receptor (AHR)/AHR nuclear translocator (ARNT)) AHR/ARNT heterodimers which, in turn, bind to xenobiotic response elements (XREs) at the level of specific DNA sequences [41]. A number of genes involved in hormone and drug metabolism would be expressed (i.e., UGT16, ALDH3, CYP1A1, CYP1A2, CYP1B1, etc.). SHP also appears to downregulate the constitutive-androstane-receptor- (CAR-) mediated CYP2B1 gene expression, induced by phenobarbital to form the CAR/RXR heterodimer which, in turn, binds to 2 DR-4 sites to form the phenobarbital responsive unit in the CYP2B gene [120] (Table 3). One role of SHP in steroidogenesis has been identified in the testes with influence on testosterone synthesis and germ cell differentiation [121] and in the intestine for glucocorticoid synthesis [122].

A role for SHP in cell proliferation and apoptosis signaling is emerging. Depending on the cell type, SHP seems to have both inhibitory and stimulatory effects on apoptosis. However, the manipulation of SHP through the synthetic ligands adamantyl-substituted retinoid-related (ARR) compounds 6-[3-(1-adamantyl)-4-hydroxyphenyl]-2-naphthalenecarboxylic acid (CD437/AHPN) and 4-[3-(1-adamantyl)-4-hydroxyphenyl]-3-chlorocinnamic acid (3-Cl-AHPC) induces apoptosis of a number of malignant cells (i.e., leukemia and breast carcinoma) both in vitro and in vivo [123, 124]. The complex mechanism implies binding of ARR and 3-Cl-AHPC to SHP with formation of a corepressor complex containing Sin3A and nuclear receptor corepressor (N-CoR) which activate local control of mitochondrial function and apoptosis, with a limiting function on tumorigenesis [17, 123] (Table 3). SHP appears to be also involved in DNA methylation and acting as a tumor suppressor, at least in the human and mouse livers [125127]. Whether manipulation of SHP will be helpful in the treatment of hepatic and other gastrointestinal cancers is still a matter of research. The recent finding that SHP negatively regulates TLR signaling to NF-κB has raised the interest for the role of SHP in mechanisms governing innate immunity. SHP appears to negatively regulate the expression of genes encoding inflammatory molecules. Of note, direct binding of NF-κB seems to occur in resting cells, while binding of SHP to TRAF6 occurs in LPS-stimulated cells [128, 129].

5. Conclusions and Perspectives

A remarkable number of metabolic functions in the body appear to be regulated by the orphan unique NR, small heterodimer partner SHP, which targets a complex set of genes in multiple pathways as a transcriptional corepressor (Figure 4). Pathways include fatty acid metabolism, glucose homeostasis, and drug-hormone detoxification. When looking at complex mechanisms leading to some important lipidopathies, that is, obesity and liver steatosis, enlightening data about the regulatory function of SHP are provided by studies using Shp-deleted and Shp-overexpressed animal models. Most likely, a condition of Shp deficiency might counteract lipid accumulation and improve plasma lipoprotein profiles. Further studies are urgently needed to confirm that such an important metabolic regulatory mechanism of SHP is true and has high translational value. To date, however, no synthetic antagonists or agonists for SHP are available, and one should keep in mind that rather divergent and somewhat elusive data have been observed regarding the loss of SHP function in humans and rodents. Thus, careful examination of subtle SHP intrinsic functions is essential to dissect potential modulatory pathways of SHP for a variety of metabolic abnormalities but also in tumorigenesis. Moreover, identifying specific endogenous ligands and synthetic agonists of SHP will pave the way to for therapeutic intervention. The effect of synthetic ligands on SHP modulation in hepatocytes and adipocytes, for example, might represent therapeutic tools for the treatment of constituents of the metabolic syndrome, namely, hypercholesterolemia, overweight obesity, and liver steatosis.

Abbreviations

AF1:Activation function 1
AF2:Activation function 2
AHR/ARNT:Aryl hydrocarbon receptor (AHR)/AHR nuclear translocator (ARNT)
ALDH3:Human aldehyde dehydrogenase 3 gene
AMPK:AMP-activated protein kinase
ARR:Adamantyl-substituted retinoid-related molecules
β1-AR:β1-adrenergic receptor
BAREs:Bile acids response elements
BAT:Brown adipose tissue
Brm:Human Brahma (a catalytic component of the SWI/SNF-related chromatin-remodeling complex)
CAR:Constitutive androstane receptor
CD437/AHPN:6-[3-(1-Adamantyl)-4-hydroxyphenyl]-2-naphthalenecarboxylic acid
3-Cl-AHPC:4-[3-(1-Adamantyl)-4-hydroxyphenyl]-3-chlorocinnamic acid
CYP3A:Hepatic cytochrome P-4503A
CYP1A1:Cytochrome P450, family 1, subfamily A, polypeptide 1
CYP1A2:Cytochrome P450, family 1, subfamily A, polypeptide 2
CYP1B1:Cytochrome P450, family 1, subfamily B, polypeptide 1
CYP2B1:Cytochrome P450, subfamily IIB
CYP7A1:Cholesterol 7 alpha-hydroxylase
CYP8B1:Cholesterol 12alpha-hydroxylase
DBD:DNA-binding domain
DAX-1:Dosage-sensitive sex reversal adrenal hypoplasia congenita (AHC) critical region on the X chromosome, gene 1
DR-4:Nuclear receptor half-site repeat
ER:Estrogen receptor
FBP1:Fructose-1,6-bisphosphatase
FGF:Fibroblast growth factor
FGFR:Fibroblast growth factor receptor
FISH:Fluorescence in situ hybridization
Foxo:Forkhead transcription factors
FXR:Farnesoid X receptor
G6Pase:Glucose-6-phosphatase
GPCR:G-protein-coupled Receptor
GPS2:G protein pathway suppressor 2
GR:Glucocorticoid receptors
HDL:High-density lipoprotein
HNF:Hepatocyte nuclear factor
JNK:Jun N-terminal kinase
JUN-D:Jun family of the activator protein 1 (AP-1) transcription factor complex
LBD:Ligand-binding domain
LDL:Low-density lipoprotein
LDLR:Low-density lipoprotein receptor
LRH1:Liver receptor homologue 1
3T3-L1:Mouse embryonic fibroblast adipose-like cell line.
LXXLL:Leu-Xaa-Xaa-Leu-Leu
mSin3A:Mammalian homolog of the Saccharomyces cerevisiae transcriptional corepressor Sin3p
N-CoR:Nuclear receptor corepressor
NRs:Nuclear receptors
PEPCK:Phosphoenolpyruvate carboxykinase
PGC-1:Peroxisome proliferator-activated receptor-γ (PPAR-γ) coactivator-1
PKC:Protein kinase C
PXR:Pregnane X receptor
RAR:Retinoid acid receptor
RNA Pol II:RNA polymerase II
RNF31Member of the ring-between-ring (RBR) family of E3 ubiquitin ligases
RXR:Retinoid X receptor
SHP:Small heterodimer partner
SMILE:SHP-interacting leucine zipper protein
SMRT:Silencing mediator of retinoid and thyroid hormone receptor
SRC-1:Steroid receptor coactivator-1
SWI/SNF:SWItch/Sucrose NonFermentable (yeast nucleosome remodeling complex composed of several proteins which are products of the SWI and SNF genes)
UCP1:Uncoupling protein-1
USF-1:Upstream stimulatory factor-1
TFIID:Transcription initiation factor II D
TFIIE:Transcription factor II E
UGT-16:Uridine 5′-diphospho (UDP) glucuronosyl transferase family member
VLDL:Very-low-density lipoprotein
XREs:Xenobiotic response elements
WAT:White adipose tissue.

Conflict of Interests

The authors declare that there is no conflict of interests.

Acknowledgments

This work was supported in part by research grants DK54012 and DK73917 (D. Q.-H. Wang) from the National Institutes of Health (US Public Health Service), FIRB 2003 RBAU01RANB002 (P. Portincasa) from the Italian Ministry of University and Research, and ORBA10ROPA from the University of Bari. P. Portincasa was a recipient of the short-term mobility grant 2005 from the Italian National Research Council (CNR). L. Bonfrate was a recipient of the travel grant for young investigators from the European Society for Clinical Investigation, 2011. Due to space limitations, the authors apologize to those whose publications related to the discussed issues could not be cited.