REVIEW URRENT C OPINION
Recent advances in understanding the STSL locus and ABCG5/ABCG8 biology Shailendra B. Patel a,b
Purpose of review To provide an update on recent advances made in our mechanistic and pathophysiological understanding of the rare human disease Sitosterolemia, the role of ABCG5/ABCG8 in sterol trafficking and how newer data implicate a more wider role in the body. Recent findings Sitosterolemia is caused by a genetic defect of sterolins (ABCG5/ABCG8) mapped to the STSL locus. Polymorphic variations in STSL have been linked to lipid levels and gallstone disease in whites. Newer studies now link this locus to a more diverse ethnic group for gallstone disease, susceptibility to biliary cancer, and show variants that alter sterolin function. Intriguingly, carriers of a mutant allele seem to show protection against carotid wall disease. Although the ‘promoter’ region of the STSL is minimal, regulatory regions responsive to liver X receptor have remained elusive, but no longer; two intronic regions in ABCG8 have now been identified. Xenosterol accumulation leads to loss of abdominal fat, infertility, and premature death. Xenosterol accumulation in mouse platelet membranes leads to platelet hyperactivation, increased microparticle formation, and reduced aIIbb3 surface expression. In humans, phytosterols may promote liver injury in parenteral nutrition-associated liver disease. Summary Progress in understanding sterolin function is beginning to show that xenosterols can be toxic and are involved on pathogenesis, and the role of ABCG5/ABCG8 may extend into other metabolic processes by altering intracellular sterol metabolism. Keywords ABCG5, phytosterols, pleiotropy, toxicity, xenosterols
INTRODUCTION The rare disease of ‘b’-Sitosterolemia was identified in 1974 in a classical article by Bhattacharyya and Connor [1], the disease locus, STSL, was mapped to human chromosome 2p21 in 1998 [2], and the genetic basis was fully elucidated by 2001 [3–6]. This advance led to uncovering a very fundamental physiological pathway that had not been previously conceptualized and led to a rewrite of pathways about sterol entry and excretion involving the intestine and liver. From studies carried out characterizing patients with this disease, as well as from in-vitro and model organism studies over the ensuing decade, we have learnt that two highly homologous genes, ABCG5 and ABCG8, comprise the STSL locus, arranged in a head-to-head organization and seem to be ‘promoterless’ [6], these genes encode for ATP-binding cassette half-transporters ABCG5 (sterolin-1) and ABCG8 (sterolin-2), which are obligate heterodimers and require each other for
maturation and expression at the apical cell surface of hepatocytes and enterocytes [7–11]; sterolins function to extrude all sterols, but seem to have a predilection for xenosterols [12]; mutations in either subunit lead to accumulation of xenosterols from the diet, of which sitosterol is the major species, hence the disease name Sitosterolemia [6]; in Sitosterolemia one can observe tendon and tuberous xanthomas, premature atherosclerosis and death, hemolysis, macrothrombocytopenia, a Clement J. Zablocki Veterans Affairs Medical Center, Milwaukee, Wisconsin, USA and bDivision of Endocrinology, Metabolism and Clinical Nutrition, Medical College of Wisconsin, Milwaukee, Wisconsin, USA
Correspondence to Shailendra B. Patel, Division of Endocrinology, Metabolism and Clinical Nutrition, Medical College of Wisconsin, HRC 4850, 8701 Watertown Plank Rd, Milwaukee, WI 53226, USA. Tel: +1 414 955 5645; e-mail: [email protected] Curr Opin Lipidol 2014, 25:169–175 DOI:10.1097/MOL.0000000000000071
0957-9672 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-lipidology.com
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Lipid metabolism
KEY POINTS The STSL locus plays a fundamental role in regulating dietary xenosterols and cholesterol. This locus manifests true pleiotropy affecting the blood, heart, liver, fat, and perhaps the central nervous system. The presence of biologically active common polymorphisms at this locus suggests that there may be both protective and deleterious effects that are contextual.
and rarely, endocrine and liver dysfunction [13]; in mice deficient in sterolins [14–17], all of the biochemical alterations are well recapitulated and biliary studies show severe impairment of cholesterol excretion (and overexpression of sterols lead to supersaturation of bile with cholesterol); the STSL is highly polymorphic in humans with polymorphisms in both ABCG5 and ABCG8 [6]; variations at STSL have been shown in Genome-wide Association Study (GWASs) to contribute to variations in lipid levels as well as gallstone disease [18–29]; gene expression of ABCG5 and ABCG8 is induced strongly by liver X receptor (LXR) activation [30], although liver receptor homologue (LRH) [31], hepatocyte nuclear factor 4 alpha (HNF-4a), and GATA4 [32] have also been implicated; and finally, blocking sterol entry at the intestinal level with the drug Ezetimibe can ameliorate the sitosterolemia in humans [33] and mice [34]. In normal humans, fortification of the diet with phytosterols reduces dietary cholesterol absorption, an effect known for several decades and preceding the discovery of Sitosterolemia [35–37]. This effect is mediated primarily by sitosterol competing with cholesterol for entry into micelles in the proximal small intestine [37]. Finally, plant sterols are present in large quantities in parenteral nutrition preparations, and in children both hemolysis and parenteral nutrition associated liver dysfunction/ disease (PNALD) has been reported [38], with the implication that plant sterols may be playing a role in these processes. Against this background, studies published over the last 2 years or so have now added increased insight into these physiological processes. These are grouped according to broad subject categories.
bile, and thus, it is not surprising that one would implicate sterolins as playing a role, as their major function seems to be extrusion of free cholesterol into bile at the hepatocyte canalicular membrane. The first genetic linkage came from QTL mapping of the Lith9 locus to the murine STSL site [39], followed by a GWAS that linked ABCG8 to gallbladder disease in humans [18]. Since then, several studies have replicated these genetic links, the latest of which show that the genetic predisposition is replicable in very diverse human populations. Worryingly, while gallstone disease is arguably not as dangerous with current modern medicine, biliary tree cancer risk seems to be elevated four-fold by variants at the STSL site [40 ]. Are these genetic variants altering sterolin function? Two articles have now examined not only the link between the ABCG8 variant 19H (D is the control amino acid) and propensity to form gallstones, and more importantly, both have attempted to establish alteration of sterolin function as a result of this amino acid change. Renner et al. [41 ] using individuals who were undergoing colonoscopies obtained ileal biopsies, as well as plasma to study the effect of D19 versus H19. Ileal mRNA expression (using qRT-PCR) was unaffected by genotype status for ABCG5, ABCG8, or NPC1L1. However, using plasma surrogate sterol markers, H19 carriers showed lower sitosterol and campesterol ratios to cholesterol (interpreted as lower cholesterol absorption) compared to the control D19 individuals [41 ]. How this is tied to increased gallstone risk is not clear, but the power of this study was the apparent lack of an effect on gene expression changes by the genetic variant. Much more exciting is the study by von Kampen et al. [42 ] who not only revalidated the genetic D19H linkage on over 4000 individuals to gallstone disease, but showed in an in-vitro assay that the H19 variant led to a 3.2-fold increase in cholesterol export (from HEK cells transfected with both ABCG5 and ABCG8). The R50C variant in ABCG5 was also genetically implicated, although this variant was not shown to increase cholesterol efflux. This is the first time a genetic variant at the STSL site is shown to have direct functional consequences [42 ]. Variants at the STSL locus have been linked to lipids levels previously, but now a large consortium has re-examined the relationship between serum phytosterol levels, coronary artery disease (CAD) prevalence using a genome-wide scan. Teupser et al. [43 ] used the KORA and CARLA cohorts, in which not only were DNA and blood available for analyses, some individuals had undergone partial liver resection and seemingly health liver areas were available for RNA extraction. Their data showed that common variants at the STSL site were linked to &
&
&
&&
&&
&
THE ROLE OF STSL IN GALLSTONE, BILIARY CANCER, AND CAROTID ARTERY DISEASE One of the factors that is important in formation of gallstones is the supersaturation of cholesterol on 170
www.co-lipidology.com
Volume 25 Number 3 June 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
STSL locus and ABCG5/ABCG8 biology Patel
variations in serum phytosterol levels (note that phytosterols can only come from the diet). Enticingly, single nucleotide polymorphisms (SNPs) that were associated with increased phytosterol level (by inference, reduced sterolin function) were also associated with increased prevalence of CAD, and one allele that was associated with reduced phytosterol level (increased sterolin function) was associated with reduced CAD. Furthermore, in hepatic samples, one allele (in intron 6 of ABCG8) that suggested reduced sterolin function (increased serum phytosterols) showed reduced mRNA expression (see also below on transcriptional regulation). Unfortunately, since these variants also affect serum cholesterol levels, no causative links should be made between CAD and plant sterols. These authors [43 ] also report a linkage of the ABO locus and phytosterols, but this is beyond this author’s expertise. More perplexing is a study that reports that a mutant change in ABCG8 is now associated with some protection against carotid artery disease [44 ]. The Old Order Amish harbor the mutation G574R [6,45], which, in a homozygous state, has been shown to not only result in Sitosterolemia but also premature atherosclerotic death. Horenstein et al. [44 ] studied all carriers of this mutation (n ¼ 110) and used noncarriers (n ¼ 181) from the same community and found increased plasma phytosterols but significantly decreased carotid intima-media wall thickness (a measure of atherosclerotic risk). This association withstood adjustments in standard lipid measure, suggesting at least that plasma cholesterol/lipoproteins were not involved in this observation, although markers of cholesterol synthesis were reduced in carriers. The changes in plant sterols in R574 carriers, although significantly increased by 33% compared to G574, was still relatively small and well within a clinically normal value (
Recent advances in understanding the STSL locus and ABCG5/ABCG8 biology Shailendra B. Patel a,b
Purpose of review To provide an update on recent advances made in our mechanistic and pathophysiological understanding of the rare human disease Sitosterolemia, the role of ABCG5/ABCG8 in sterol trafficking and how newer data implicate a more wider role in the body. Recent findings Sitosterolemia is caused by a genetic defect of sterolins (ABCG5/ABCG8) mapped to the STSL locus. Polymorphic variations in STSL have been linked to lipid levels and gallstone disease in whites. Newer studies now link this locus to a more diverse ethnic group for gallstone disease, susceptibility to biliary cancer, and show variants that alter sterolin function. Intriguingly, carriers of a mutant allele seem to show protection against carotid wall disease. Although the ‘promoter’ region of the STSL is minimal, regulatory regions responsive to liver X receptor have remained elusive, but no longer; two intronic regions in ABCG8 have now been identified. Xenosterol accumulation leads to loss of abdominal fat, infertility, and premature death. Xenosterol accumulation in mouse platelet membranes leads to platelet hyperactivation, increased microparticle formation, and reduced aIIbb3 surface expression. In humans, phytosterols may promote liver injury in parenteral nutrition-associated liver disease. Summary Progress in understanding sterolin function is beginning to show that xenosterols can be toxic and are involved on pathogenesis, and the role of ABCG5/ABCG8 may extend into other metabolic processes by altering intracellular sterol metabolism. Keywords ABCG5, phytosterols, pleiotropy, toxicity, xenosterols
INTRODUCTION The rare disease of ‘b’-Sitosterolemia was identified in 1974 in a classical article by Bhattacharyya and Connor [1], the disease locus, STSL, was mapped to human chromosome 2p21 in 1998 [2], and the genetic basis was fully elucidated by 2001 [3–6]. This advance led to uncovering a very fundamental physiological pathway that had not been previously conceptualized and led to a rewrite of pathways about sterol entry and excretion involving the intestine and liver. From studies carried out characterizing patients with this disease, as well as from in-vitro and model organism studies over the ensuing decade, we have learnt that two highly homologous genes, ABCG5 and ABCG8, comprise the STSL locus, arranged in a head-to-head organization and seem to be ‘promoterless’ [6], these genes encode for ATP-binding cassette half-transporters ABCG5 (sterolin-1) and ABCG8 (sterolin-2), which are obligate heterodimers and require each other for
maturation and expression at the apical cell surface of hepatocytes and enterocytes [7–11]; sterolins function to extrude all sterols, but seem to have a predilection for xenosterols [12]; mutations in either subunit lead to accumulation of xenosterols from the diet, of which sitosterol is the major species, hence the disease name Sitosterolemia [6]; in Sitosterolemia one can observe tendon and tuberous xanthomas, premature atherosclerosis and death, hemolysis, macrothrombocytopenia, a Clement J. Zablocki Veterans Affairs Medical Center, Milwaukee, Wisconsin, USA and bDivision of Endocrinology, Metabolism and Clinical Nutrition, Medical College of Wisconsin, Milwaukee, Wisconsin, USA
Correspondence to Shailendra B. Patel, Division of Endocrinology, Metabolism and Clinical Nutrition, Medical College of Wisconsin, HRC 4850, 8701 Watertown Plank Rd, Milwaukee, WI 53226, USA. Tel: +1 414 955 5645; e-mail: [email protected] Curr Opin Lipidol 2014, 25:169–175 DOI:10.1097/MOL.0000000000000071
0957-9672 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-lipidology.com
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Lipid metabolism
KEY POINTS The STSL locus plays a fundamental role in regulating dietary xenosterols and cholesterol. This locus manifests true pleiotropy affecting the blood, heart, liver, fat, and perhaps the central nervous system. The presence of biologically active common polymorphisms at this locus suggests that there may be both protective and deleterious effects that are contextual.
and rarely, endocrine and liver dysfunction [13]; in mice deficient in sterolins [14–17], all of the biochemical alterations are well recapitulated and biliary studies show severe impairment of cholesterol excretion (and overexpression of sterols lead to supersaturation of bile with cholesterol); the STSL is highly polymorphic in humans with polymorphisms in both ABCG5 and ABCG8 [6]; variations at STSL have been shown in Genome-wide Association Study (GWASs) to contribute to variations in lipid levels as well as gallstone disease [18–29]; gene expression of ABCG5 and ABCG8 is induced strongly by liver X receptor (LXR) activation [30], although liver receptor homologue (LRH) [31], hepatocyte nuclear factor 4 alpha (HNF-4a), and GATA4 [32] have also been implicated; and finally, blocking sterol entry at the intestinal level with the drug Ezetimibe can ameliorate the sitosterolemia in humans [33] and mice [34]. In normal humans, fortification of the diet with phytosterols reduces dietary cholesterol absorption, an effect known for several decades and preceding the discovery of Sitosterolemia [35–37]. This effect is mediated primarily by sitosterol competing with cholesterol for entry into micelles in the proximal small intestine [37]. Finally, plant sterols are present in large quantities in parenteral nutrition preparations, and in children both hemolysis and parenteral nutrition associated liver dysfunction/ disease (PNALD) has been reported [38], with the implication that plant sterols may be playing a role in these processes. Against this background, studies published over the last 2 years or so have now added increased insight into these physiological processes. These are grouped according to broad subject categories.
bile, and thus, it is not surprising that one would implicate sterolins as playing a role, as their major function seems to be extrusion of free cholesterol into bile at the hepatocyte canalicular membrane. The first genetic linkage came from QTL mapping of the Lith9 locus to the murine STSL site [39], followed by a GWAS that linked ABCG8 to gallbladder disease in humans [18]. Since then, several studies have replicated these genetic links, the latest of which show that the genetic predisposition is replicable in very diverse human populations. Worryingly, while gallstone disease is arguably not as dangerous with current modern medicine, biliary tree cancer risk seems to be elevated four-fold by variants at the STSL site [40 ]. Are these genetic variants altering sterolin function? Two articles have now examined not only the link between the ABCG8 variant 19H (D is the control amino acid) and propensity to form gallstones, and more importantly, both have attempted to establish alteration of sterolin function as a result of this amino acid change. Renner et al. [41 ] using individuals who were undergoing colonoscopies obtained ileal biopsies, as well as plasma to study the effect of D19 versus H19. Ileal mRNA expression (using qRT-PCR) was unaffected by genotype status for ABCG5, ABCG8, or NPC1L1. However, using plasma surrogate sterol markers, H19 carriers showed lower sitosterol and campesterol ratios to cholesterol (interpreted as lower cholesterol absorption) compared to the control D19 individuals [41 ]. How this is tied to increased gallstone risk is not clear, but the power of this study was the apparent lack of an effect on gene expression changes by the genetic variant. Much more exciting is the study by von Kampen et al. [42 ] who not only revalidated the genetic D19H linkage on over 4000 individuals to gallstone disease, but showed in an in-vitro assay that the H19 variant led to a 3.2-fold increase in cholesterol export (from HEK cells transfected with both ABCG5 and ABCG8). The R50C variant in ABCG5 was also genetically implicated, although this variant was not shown to increase cholesterol efflux. This is the first time a genetic variant at the STSL site is shown to have direct functional consequences [42 ]. Variants at the STSL locus have been linked to lipids levels previously, but now a large consortium has re-examined the relationship between serum phytosterol levels, coronary artery disease (CAD) prevalence using a genome-wide scan. Teupser et al. [43 ] used the KORA and CARLA cohorts, in which not only were DNA and blood available for analyses, some individuals had undergone partial liver resection and seemingly health liver areas were available for RNA extraction. Their data showed that common variants at the STSL site were linked to &
&
&
&&
&&
&
THE ROLE OF STSL IN GALLSTONE, BILIARY CANCER, AND CAROTID ARTERY DISEASE One of the factors that is important in formation of gallstones is the supersaturation of cholesterol on 170
www.co-lipidology.com
Volume 25 Number 3 June 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
STSL locus and ABCG5/ABCG8 biology Patel
variations in serum phytosterol levels (note that phytosterols can only come from the diet). Enticingly, single nucleotide polymorphisms (SNPs) that were associated with increased phytosterol level (by inference, reduced sterolin function) were also associated with increased prevalence of CAD, and one allele that was associated with reduced phytosterol level (increased sterolin function) was associated with reduced CAD. Furthermore, in hepatic samples, one allele (in intron 6 of ABCG8) that suggested reduced sterolin function (increased serum phytosterols) showed reduced mRNA expression (see also below on transcriptional regulation). Unfortunately, since these variants also affect serum cholesterol levels, no causative links should be made between CAD and plant sterols. These authors [43 ] also report a linkage of the ABO locus and phytosterols, but this is beyond this author’s expertise. More perplexing is a study that reports that a mutant change in ABCG8 is now associated with some protection against carotid artery disease [44 ]. The Old Order Amish harbor the mutation G574R [6,45], which, in a homozygous state, has been shown to not only result in Sitosterolemia but also premature atherosclerotic death. Horenstein et al. [44 ] studied all carriers of this mutation (n ¼ 110) and used noncarriers (n ¼ 181) from the same community and found increased plasma phytosterols but significantly decreased carotid intima-media wall thickness (a measure of atherosclerotic risk). This association withstood adjustments in standard lipid measure, suggesting at least that plasma cholesterol/lipoproteins were not involved in this observation, although markers of cholesterol synthesis were reduced in carriers. The changes in plant sterols in R574 carriers, although significantly increased by 33% compared to G574, was still relatively small and well within a clinically normal value (
