Novel Insights into the Organic Solute Transporter Alpha/Beta, OSTα/β: From the Bench to the Bedside

Abstract

Organic solute transporter alpha/beta (OSTα/β) is a heteromeric solute carrier protein that transports bile acids, steroid metabolites and drugs into and out of cells. OSTα/β protein is expressed in various tissues, but its expression is highest in the gastrointestinal tract where it facilitates the recirculation of bile acids from the gut to the liver. Previous studies established that OSTα/β is upregulated in liver tissue of patients with extrahepatic cholestasis, obstructive cholestasis, and primary biliary cholangitis (PBC), conditions that are characterized by elevated bile acid concentrations in the liver and/or systemic circulation. The discovery that OSTα/β is highly upregulated in the liver of patients with nonalcoholic steatohepatitis (NASH) further highlights the clinical relevance of this transporter because the incidence of NASH is increasing at an alarming rate with the obesity epidemic. Since OSTα/β is closely linked to the homeostasis of bile acids, and tightly regulated by the nuclear receptor farnesoid X receptor, OSTα/β is a potential drug target for treatment of cholestatic liver disease, and other bile acid-related metabolic disorders such as obesity and diabetes. Obeticholic acid, a semi-synthetic bile acid used to treat PBC, under review for the treatment of NASH, and in development for the treatment of other metabolic disorders, induces OSTα/β. Some drugs associated with hepatotoxicity inhibit OSTα/β, suggesting a possible role for OSTα/β in drug-induced liver injury (DILI). Furthermore, clinical cases of homozygous genetic defects in both OSTα/β subunits resulting in diarrhea and features of cholestasis have been reported. This review article has been compiled to comprehensively summarize the recent data emerging on OSTα/β, recapitulating the available literature on the structure-function and expression-function relationships of OSTα/β, the regulation of this important transporter, the interaction of drugs and other compounds with OSTα/β, and the comparison of OSTα/β with other solute carrier transporters as well as adenosine triphosphate-binding cassette transporters. Findings from basic to more clinically focused research efforts are described and discussed.

1. Introduction

Nearly two decades ago, organic solute transporter alpha/beta (OSTα/β/SLC51A/B) was identified in a screen of a hepatic cDNA library from the little skate (Leucoraja erinacea) (Wang, Seward, Li, Boyer, & Ballatori, 2001), but even today this heteromeric transport protein is relatively poorly understood and understudied. OSTα/β has been detected on the basolateral membrane of epithelial cells in tissues ranging from the zona reticularis of the adrenal cortex to the renal tubules and the rectum (Ballatori, et al., 2005; Fang, et al., 2010; Uhlen, et al., 2015), with the highest mRNA and protein levels found in the small intestine (ileum and duodenum). In conjunction with the apical sodium-dependent bile acid transporter (ASBT/SLC10A2) on the luminal membrane of the intestinal epithelial cells, intestinal OSTα/β localized on the basolateral membrane plays a key role in the reabsorption and enterohepatic circulation of bile acids (Ballatori, Fang, Christian, Li, & Hammond, 2008; Frankenberg, et al., 2006; Rao, et al., 2008; Sultan, et al., 2018). Furthermore, OSTα/β-mediated efflux of bile acids protects the ileal epithelium against intracellular bile acid accumulation and intestinal injury in mice (Ferrebee, et al., 2018). While OSTα/β is known primarily for its important role in the transport and homeostasis of bile acids, other steroids and some drugs also have been identified as OSTα/β substrates (Ballatori, et al., 2005; Wang, et al., 2001). In addition, some drugs/xenobiotics inhibit the transport function of OSTα/β (Malinen, Ali, Bezencon, Beaudoin, & Brouwer, 2018; Malinen, Kauttonen, et al., 2019). Although assessing interactions with OSTα/β is not yet a requirement in the drug development pipeline, the International Transporter Consortium has acknowledged the potential of OSTα/β-mediated drug interactions (Kenna, et al., 2018; Zamek-Gliszczynski, et al., 2018). OSTα/β is also expressed in other organs central to drug absorption, distribution, metabolism, excretion and toxicity (ADME-Tox) including the kidneys and liver. Interestingly, hepatic OSTα/β is markedly upregulated in certain liver diseases (Malinen, et al., 2018; Soroka, Ballatori, & Boyer, 2010), and when bile flow is interrupted in humans or rodents (Boyer, et al., 2006; Chai, et al., 2015; Schaap, van der Gaag, Gouma, & Jansen, 2009).

To date, review articles on OSTα/β have focused on the role of OSTα/β as a bile acid transporter (Ballatori, et al., 2009; Dawson, Hubbert, & Rao, 2010; Soroka, Ballatori, et al., 2010), summarizing the expression (Ballatori, 2005, 2011; Ballatori, Christian, Wheeler, & Hammond, 2013), structure (Dawson, et al., 2010) and regulation of OSTα/β (Ballatori, et al., 2013; Ballatori, et al., 2009). Recently, OSTα/β has received renewed attention in relation to its potential roles in nonalcoholic steatohepatitis (NASH), bile acid-related metabolic disorders such as cholestatic liver disease, obesity and diabetes, as well as in drug-induced liver injury (DILI). The present review provides an update on OSTα/β from a pharmaceutical perspective. Recent data emerging on this transporter are highlighted, and available information on the structure-function and expression-function relationships of OSTα/β, the regulation of this important transporter, the interaction of drugs and other compounds with OSTα/β, and the comparison of OSTα/β with other solute carrier (SLC) and adenosine triphosphate (ATP)-binding cassette (ABC) transporters is summarized.

2. Expression of OSTα/β

2.1. Co-expression of SLC51A/OSTα and SLC51B/OSTβ mRNAs and Proteins

Transporters properly expressed on the basolateral or apical plasma membranes are important determinants of the disposition of endogenous and exogenous compounds in the body. Human SLC51A and SLC51B genes are transcribed on chromosomes 3q29 and 15q22.31, respectively. Both OSTα and OSTβ subunits are translated in the endoplasmic reticulum and translocated onto the basolateral plasma membrane of epithelial cells to form a functional transporter (Ballatori, et al., 2005; Boyer, et al., 2006; Dawson, et al., 2005). It seems that OSTα and OSTβ stabilize each other in mammalian cells to form the heteromeric OSTα/β protein complex (Li, Cui, Fang, Lee, & Ballatori, 2007). Protein expression of OSTα and OSTβ in co-expressing human embryonic kidney (HEK) 293 cells decreased only modestly when cells were treated for 24 hr with the protein synthesis inhibitor cycloheximide, suggesting a half-life of OSTα/β beyond 24 hr when both subunits were co-expressed (Li, et al., 2007). Pulse-chase experiments supported a half-life of OSTα beyond 24 hr when co-expressed with OSTβ, and revealed an OSTα half-life of ~2 hr in the absence of OSTβ (Dawson, et al., 2010). Furthermore, both subunits are required to enable the transport function of OSTα/β at the plasma membrane (Wang, et al., 2001). The necessary co-expression for transport function was confirmed with transport studies in transfected African green monkey kidney fibroblast-like COS-7 cells: increased uptake of taurocholate (TCA) and estrone sulfate (ES) was evident in cells expressing both OSTα and OSTβ as compared to cells transfected with only one of the subunits (Sun, et al., 2007). Similarly, HEK 293 and Madin-Darby canine kidney (MDCK) cells only express OSTα and OSTβ protein at the plasma membrane when both genes are transfected simultaneously (Dawson, et al., 2005). In renal and intestinal tissue of OSTα-knockout mice, SLC51B mRNA is present, while OSTβ protein was not detected (Li, et al., 2007). Interestingly, in clawed frog (Xenopus laevis) oocytes, both OSTα and OSTβ subunits were able to separately reach the plasma membrane when singly expressed, although each subunit alone lacked transporter activity (Seward, Koh, Boyer, & Ballatori, 2003). Although both subunits are needed for functional OSTα/β on the plasma membrane, OSTα and OSTβ can be expressed at different protein levels, as evidenced in many tissues (Ballatori, 2005) [Table 1; (Uhlen, et al., 2015)]. These differences may be explained by unknown OSTα/β stoichiometry on the plasma membrane, and/or variable intracellular expression of the subunits.

Table 1.

Protein Levels of Human OSTα and OSTβ in Various Tissues and Cell Types from Healthy Individuals.

Tissue	Adrenal Gland	Appendix	Cerebellum	Colon
Cell Type	Glandular Cells	Glandular Cells	Purk inje Cells	Glandular Cells	Peripheral Nerve/Ganglion
OSTα Level	Not detected*	Medium	Not detected*	Low	Low
OSTβ Level	Not detected*	Medium	Not detected*	Medium	Not detected
Tissue	Epididymis	Hippocampus	Kidney	Liver
Cell Type	Glandular Cells	Neuronal Cells	Cells in Tubules	Bile Duct Cells	Hepatocytes
OSTα Level	Low	Not detected*	Low	Not detected*	Low
OSTβ Level	Not detected	Not detected*	Low	Not detected	Not detected
Tissue	Prostate	Rectum	Seminal Vesicle	Skin
Cell Type	Glandular Cells	Glandular Cells	Glandular Cells	Keratinocytes	Epidermal Cells
OSTα Level	Low	Low	Low	Low	Low
OSTβ Level	Not detected	Medium	Not detected	Not detected	Not detected
Tissue	Small Intestine	Stomach	Testis
Cell Type	Glandular Cells	Glandular Cells	Cells in Seminiferous Ducts	Leydig Cells
OSTα Level	Medium	Low	Not detected	Medium
OSTβ Level	High	Low/Medium	Low	Not detected

Data on protein levels were obtained from The Human Protein Atlas version 19 and Ensembl version 92.38 and were determined using microarray-based immunohistochemistry. A color scale was applied to each column separately, with dark green and red denoting the highest and lowest levels of expression, respectively. When comparing this table to the main text, discrepancies can be noted with other publications, which may be due to differences in sample preparation, antibodies utilized, data analysis, and/or other procedures (e.g., colorimetric versus fluorescence-based detection, with varying sensitivities).

^*Although The Human Protein Atlas did not report positive staining for OSTα and/or OSTβ in adrenal glandular cells, cerebellar cells, hippocampal cells or bile duct cells, previous studies revealed that OSTα and OSTβ levels were detected in the zona reticularis of the adrenal gland, Purkinje cells of the cerebellum, and cornu ammonis cells of the hippocampus (Fang, et al., 2010), while OSTα levels have been reported in cholangiocytes (Ballatori, et al., 2005). Various other cell types not included in the table also were evaluated by The Human Protein Atlas. OSTα and OSTβ protein were not detectable in adipose tissue (adipocytes), appendix (lymphoid tissue), bone marrow (hematopoietic cells), breast (adipocytes, glandular and myoepithelial cells), bronchus (respiratory epithelial cells), caudate (glial and neuronal cells), cerebellum (granular and molecular layer cells), cerebral cortex (endothelial, glial, neuronal cells and neuropil), colon (endothelial cells), endometrium (endometrial stroma and glandular cells), esophagus (squamous epithelial cells), Fallopian tube (glandular cells), gallbladder (glandular cells), heart muscle (myocytes), hippocampus (glial cells), kidney (glomerular cells), lung (macrophages and pneumocytes), lymph node (non-germinal center cells), nasopharynx (respiratory epithelial cells), oral mucosa (squamous epithelial cells), ovary (ovarian stroma cells), pancreas (exocrine glandular cells and islets of Langerhans), parathyroid gland (glandular cells), placenta (trophoblastic cells), salivary gland (glandular cells), skeletal muscle (myocytes), skin (fibroblasts, Langerhans and melanocytes), smooth muscle (smooth muscle cells), soft tissue (chondrocytes, fibroblasts and peripheral nerve), spleen (red pulp cells), thyroid gland (glandular cells), tonsil [(non)-germinal center and squamous epithelial cells], urinary bladder (urothelial cells), uterine cervix (glandular and squamous epithelial cells), and vagina (squamous epithelial cells). OSTα protein was not detectable in lymph node (germinal center cells) and spleen (white pulp cells). OSTβ protein was not detectable in placenta (decidual cells).

2.2. Tissue expression of SLC51A/OSTα and SLC51B/OSTβ

The expression of OSTα/β mRNA and protein has been reviewed in several species (Ballatori, 2005, 2011; Ballatori, et al., 2013). More recently, the Human Protein Atlas project [www.proteinatlas.org; (Uhlen, et al., 2015)] has shown that OSTα and OSTβ protein (Table 1), as well as SLC51A (14 splice variants) and SLC51B (one transcript) mRNA (Table 2) are expressed in various human tissues and cell types. In tissues, the expression levels of SLC51A/OSTα and SLC51B/OSTβ are usually not equal. Furthermore, when comparing various publicly available resources on OSTα and OSTβ protein levels, some discrepancies are evident, as described below – particularly for those tissues where expression is relatively low, which could be the result of differences in procedures (e.g., tissue handling and storage, manual evaluation of immunohistochemical data) or antibody quality (e.g., selectivity and specificity).

Table 2.

mRNA Isoform Expression of Human SLC51A and SLC51B in Various Tissues and Cell Types from Healthy Individuals.

	SLC51A Level																SLC51B Level
Tissue / Cell Type	ENST 00000 296327	ENST 00000 415111	ENST 00000 416660	ENST 00000 428985	ENST 00000 442203	ENST 00000 471430	ENST 00000 472653	ENST 00000 475271	ENST 00000 475672	ENST 00000 476129	ENST 00000 479732	ENST 00000 484407	ENST 00000 492794	ENST 00000 496737	Average	Sum	ENST 00000 334287
Adipose Tissue	0.07	0.42	0.04	0.06	0.14	0.04	0.00	0.06	0.25	0.06	1.45	0.05	0.74	1.51	0.35	4.89	0.86
Adrenal Gland	3.40	0.66	0.68	0.43	0.34	1.02	0.41	0.00	1.03	0.28	2.13	5.66	1.38	1.86	1.38	19.29	1.32
Appendix	1.02	0.59	0.16	0.12	0.80	0.49	0.58	0.00	0.54	0.08	1.45	0.70	0.63	2.79	0.71	9.96	4.48
Basophil	0.00	0.00	0.00	0.00	0.59	0.00	0.00	0.00	0.00	0.00	0.21	0.00	0.00	0.93	0.12	1.72	0.01
Bone Marrow	0.52	1.16	0.91	1.37	3.02	1.00	1.05	0.14	0.58	3.70	1.17	5.29	1.38	4.54	1.84	25.82	0.53
Breast	0.00	0.22	0.00	0.06	0.26	0.00	0.03	0.04	0.00	0.00	0.00	0.26	0.39	0.11	0.10	1.38	0.06
Cerebral Cortex	0.00	0.33	0.10	0.03	0.20	0.45	0.07	0.00	0.41	0.17	1.24	0.13	0.62	1.68	0.39	5.45	5.95
Cervix, Uterine	0.60	0.17	0.00	0.13	0.29	0.11	0.00	0.00	0.00	0.13	0.07	0.10	1.07	0.00	0.19	2.67	59.69
Classical Monocyte	0.00	0.04	0.00	0.00	0.24	0.00	0.00	0.01	0.06	0.01	0.31	0.00	0.08	0.33	0.08	1.06	0.00
Colon	9.30	0.37	0.02	0.24	0.31	0.66	0.87	0.09	0.35	0.27	0.67	3.73	5.63	0.61	1.65	23.14	40.73
Duodenum	28.07	0.60	0.11	0.58	1.65	3.51	3.11	0.00	1.22	1.82	1.78	7.61	7.44	3.38	4.35	60.87	85.37
Endometrium	0.04	0.53	0.00	0.04	0.29	0.00	0.02	0.02	0.22	0.00	1.22	0.22	0.62	1.75	0.36	4.97	5.52
Eosinophil	0.22	0.01	0.00	0.00	0.33	0.00	0.00	0.00	0.00	0.03	0.34	0.24	0.00	0.71	0.13	1.87	0.00
Epididymis	0.97	0.00	0.00	0.00	0.41	0.66	0.30	0.00	0.00	0.00	0.30	1.58	0.00	0.14	0.31	4.37	0.39
Esophagus	0.00	0.61	0.00	0.00	0.13	0.27	0.03	0.10	0.49	0.00	2.44	0.00	1.01	2.16	0.52	7.23	2.71
Fallopian Tube	2.08	0.30	0.00	0.16	0.32	0.25	0.25	0.00	0.25	0.06	0.59	0.95	0.73	0.37	0.45	6.30	6.01
Gallbladder	0.13	0.84	0.08	0.00	0.25	0.18	0.00	0.02	0.44	0.08	1.95	0.10	0.34	3.47	0.56	7.88	4.33
gdTCR	0.00	0.01	0.00	0.00	0.34	0.00	0.00	0.00	0.12	0.00	0.62	0.00	0.22	0.56	0.13	1.87	0.03
Heart Muscle	0.00	0.09	0.11	0.00	0.10	0.07	0.02	0.00	0.23	0.11	1.78	0.04	0.57	0.85	0.28	3.98	1.66
Intermediate Monocyte	0.00	0.02	0.06	0.00	0.16	0.00	0.00	0.00	0.04	0.00	0.27	0.00	0.07	0.28	0.06	0.91	0.00
Kidney	4.42	0.71	0.04	0.21	0.17	0.37	0.08	0.40	0.51	0.11	0.85	3.28	3.65	0.59	1.10	15.37	19.35
Liver (Fetal)	Detected*																Detected*
Liver	31.95	1.48	0.00	1.42	0.75	2.96	1.73	0.88	0.98	0.45	0.56	18.97	24.38	0.40	6.21	86.90	0.93
Lung	0.20	0.36	0.16	0.03	0.24	0.26	0.13	0.03	0.22	0.10	1.22	0.32	0.30	1.38	0.35	4.95	9.37
Lymph Node	0.00	0.70	0.04	0.01	0.35	0.08	0.00	0.03	0.16	0.11	1.18	0.46	0.65	3.37	0.51	7.14	0.75
MAIT T-cell	0.00	0.01	0.00	0.00	0.26	0.00	0.00	0.00	0.01	0.00	0.43	0.00	0.02	0.20	0.07	0.93	0.03
Memory B-cell	0.00	0.01	0.00	0.00	0.22	0.00	0.00	0.00	0.04	0.00	0.29	0.00	0.12	0.30	0.07	0.98	0.00
Memory CD4 T-cell	0.00	0.01	0.02	0.00	0.20	0.00	0.00	0.00	0.13	0.00	0.48	0.00	0.16	0.39	0.10	1.39	0.02
Memory CD8 T-cell	0.00	0.00	0.00	0.00	0.22	0.00	0.00	0.00	0.00	0.00	0.25	0.00	0.00	0.20	0.05	0.68	0.02
Myeloid DC	0.00	0.00	0.00	0.00	0.09	0.00	0.00	0.00	0.05	0.01	0.29	0.00	0.12	0.34	0.06	0.90	0.01
Naive B-cell	0.00	0.03	0.00	0.00	0.46	0.00	0.00	0.00	0.07	0.00	0.34	0.00	0.14	0.36	0.10	1.40	0.00
Naive CD4 T-cell	0.00	0.02	0.00	0.00	0.30	0.05	0.00	0.00	0.00	0.00	0.40	0.03	0.02	0.29	0.08	1.10	0.01
Naive CD8 T-cell	0.00	0.08	0.01	0.00	0.31	0.00	0.00	0.00	0.24	0.01	0.53	0.00	0.28	0.45	0.14	1.91	0.02
Neutrophil	0.00	0.03	0.14	0.00	1.24	0.00	0.01	0.00	0.07	0.00	0.73	0.00	0.07	0.64	0.21	2.93	0.03
NK-cell	0.00	0.09	0.00	0.00	0.54	0.00	0.00	0.00	0.03	0.00	0.43	0.00	0.00	0.51	0.11	1.59	0.00
Non-Classical Monocyte	0.00	0.05	0.02	0.00	0.19	0.02	0.00	0.00	0.08	0.01	0.57	0.00	0.11	0.57	0.12	1.63	0.01
Ovary	0.07	0.89	0.10	0.00	0.11	0.39	0.00	0.00	0.38	0.11	1.00	6.21	0.35	0.76	0.74	10.36	1.52
Pancreas	0.00	0.00	0.00	0.01	0.11	0.00	0.00	0.00	0.05	0.00	0.08	0.06	0.23	0.32	0.06	0.86	0.05
Parathyroid Gland	3.18	1.94	0.00	0.00	0.00	3.39	0.56	0.00	0.53	0.00	0.11	3.51	1.94	0.21	1.10	15.37	0.00
Pituitary Gland	Detected*																Detected*
Placenta	0.01	0.34	0.00	0.00	0.25	0.04	0.02	0.06	0.20	0.02	0.88	0.09	0.29	0.73	0.21	2.93	1.00
Plasmacytoid DC	0.00	0.00	0.00	0.00	0.40	0.00	0.00	0.00	0.00	0.01	0.28	0.00	0.08	0.36	0.08	1.14	0.00
Prostate	0.16	0.41	0.00	0.05	0.37	0.13	0.05	0.05	0.35	0.33	0.87	0.16	0.75	0.94	0.33	4.62	2.94
Rectum	0.53	0.07	0.00	0.20	0.11	0.05	0.09	0.00	0.00	0.12	0.09	0.60	0.66	0.10	0.19	2.61	31.41
Salivary Gland	0.21	0.30	0.00	0.04	0.08	0.14	0.01	0.04	0.06	0.19	0.44	0.06	0.18	0.57	0.17	2.33	0.47
Seminal Vesicle	0.06	0.07	0.00	0.00	0.23	0.00	0.04	0.00	0.00	0.00	0.00	0.03	0.06	0.03	0.04	0.53	0.94
Skeletal Muscle	0.03	0.21	0.38	0.00	0.40	0.15	0.06	0.00	0.36	0.29	1.20	0.24	0.33	0.83	0.32	4.50	0.38
Skin	0.03	1.35	0.00	0.00	0.59	0.26	0.03	0.16	1.00	0.08	4.27	0.14	1.86	6.55	1.17	16.32	1.03
Small Intestine	65.43	1.49	0.15	4.22	4.28	6.97	8.62	0.60	1.53	3.79	2.11	20.32	16.80	5.20	10.11	141.52	118.33
Smooth Muscle	0.31	0.04	0.00	0.12	0.21	0.10	0.21	0.00	0.00	0.02	0.00	0.58	0.35	0.00	0.14	1.94	2.56
Spleen	0.00	0.93	0.04	0.00	0.42	0.00	0.04	0.00	0.30	0.29	1.28	0.03	0.44	3.13	0.49	6.89	1.20
Stomach	0.15	0.32	0.03	0.00	0.18	0.00	0.00	0.00	0.16	0.05	0.91	0.03	0.18	1.33	0.24	3.35	0.80
T-reg	0.00	0.02	0.00	0.00	0.34	0.00	0.00	0.00	0.04	0.01	0.14	0.00	0.07	0.24	0.06	0.85	0.00
Testis	1.23	1.02	2.67	0.75	1.18	1.00	0.44	0.48	2.66	0.24	1.63	12.32	6.06	1.75	2.39	33.44	13.12
Thyroid Gland	0.51	1.03	0.00	0.00	0.44	0.13	0.20	0.12	0.17	0.32	1.48	0.57	0.70	1.42	0.51	7.10	1.16
Tonsil	0.22	0.22	0.00	0.04	0.07	0.13	0.02	0.12	0.06	0.00	0.06	1.01	0.26	0.12	0.17	2.35	0.06
Total PBMC	0.00	0.01	0.00	0.00	0.07	0.01	0.00	0.00	0.02	0.00	0.13	0.00	0.06	0.08	0.03	0.38	0.01
Urinary Bladder	0.04	0.37	0.00	0.00	0.21	0.00	0.00	0.00	0.26	0.00	0.78	0.14	0.00	1.43	0.23	3.23	3.19
Average	2.77	0.39	0.11	0.18	0.46	0.45	0.34	0.06	0.30	0.24	0.83	1.71	1.51	1.14	0.75	10.50	7.69

Data on transcript levels were obtained from The Human Protein Atlas version 19 and Ensembl version 92.38, and were determined using next-generation sequencing-based RNA-sequencing. Values are presented as transcripts per million. A color scale was applied to each column separately, with green and red denoting the highest and lowest levels of expression, respectively.

^*Some tissues were not analyzed in The Human Protein Atlas for SLC51A and SLC51B mRNA levels, but positive levels were detected in another study (Seward, et al., 2003). According to ensembl.org, only ENST00000296327 (340 amino acids), ENST00000415111 (23 amino acids), ENST00000416660 (66 amino acids) and ENST00000428985 (212 amino acids) are protein-coding SLC51A transcripts. gdTCR, gamma delta T cell; DC, dendritic cell; MAIT T-cell, mucosal associated invariant T-cell; NK-cell, natural killer cell; T-reg, regulatory T cell; PBMC, peripheral blood mononuclear cell.

2.2.1. SLC51A and SLC51B mRNA expression in tissues

In one of the first OSTα/β studies, human SLC51A and SLC51B mRNA was found in a variety of tissues, primarily the testis, colon, liver, fetal liver, small intestine, kidney, adrenal gland and ovary, and at lower levels in the heart, lung, brain, pituitary gland, uterus, prostate and adipose tissue (Seward, et al., 2003). Some SLC51A mRNA was observed in the human thyroid and mammary glands, but SLC51B mRNA was below the detection limit (Seward, et al., 2003). Differences in SLC51A mRNA expression among the duodenum, terminal ileum and colon were negligible in a study with eight healthy human subjects, and a similar trend among these tissues was observed for SLC51B mRNA expression (Schwarz, 2012). This study also showed higher mRNA expression of SLC51A compared to SLC51B in the human liver; the opposite was observed in the human kidney (Schwarz, 2012), in agreement with a previous study (Ballatori, et al., 2005) and the Human Protein Atlas project (Table 2). The Human Protein Atlas project detected some level (e.g., ≥0.01 transcripts per million RNA molecules per sample) of SLC51A and SLC51B expression in nearly all analyzed cell types and tissues (Table 2). Compared to the average expression of SLC51A mRNA in all analyzed human tissue/cellular samples, SLC51A mRNA expression was higher in the adrenal gland, bone marrow, colon, duodenum (small intestine), kidney, liver, parathyroid gland, skin and testis. SLC51B mRNA expression was higher in the uterine cervix, colon, duodenum (small intestine), kidney, lung, rectum and testis, compared to the average expression of SLC51B mRNA in all analyzed human tissue/cellular samples. The abundance of human SLC51A and SLC51B mRNA compared to other transporters in histologically normal tissues (small intestine, liver and kidney) is depicted in Figure 1.

Figure 1.

Abundance of Human SLC51A and SLC51B mRNAs Compared to Other Transporters in Histologically Normal Small Intestine, Liver and Kidney. The data were obtained from The Human Protein Atlas version 19 and Ensembl version 92.38, and represent consensus normalized expression levels from three transcriptomics datasets (HPA, GTEx and FANTOM5). Data are expressed as a percentage of total tissue transporter abundance.

In the mouse, the highest expression of SLC51A and SLC51B mRNA was found in the ileum, followed by the jejunum (Dawson, et al., 2005). In pigs, both SLC51A and SLC51B mRNA had the highest intestinal expression in the ileum, followed by the jejunum, duodenum, cecum and colon (Fang, et al., 2018). Interestingly, while SLC51A and SLC51B mRNA was detected in the small intestines, colons and kidneys from mice, rats and humans, the transcripts were nearly undetectable in mouse and rat liver, while these transcripts were detectable in human liver (Ballatori, et al., 2005; Schwarz, 2012; Seward, et al., 2003). Laser capture microdissection-isolated mouse Purkinje and hippocampal cells showed mRNA expression of SLC51A and SLC51B (Fang, et al., 2010).

2.2.2. OSTα and OSTβ protein expression in tissues

Studies on protein expression and localization have been limited due to the unavailability of commercial OSTα/β antibodies until relatively recently. The first OSTα/β protein expression studies were performed with antibodies produced in-house (Ballatori, et al., 2005; Dawson, et al., 2005; Li, et al., 2007). In those early studies, OSTα and/or OSTβ protein expression was reported in ADME-Tox organs, including the human ileum, kidney and liver using indirect immunofluorescence (Ballatori, et al., 2005). In the human liver, OSTα protein was expressed in hepatocytes and cholangiocytes (Ballatori, et al., 2005). Using these same in-house-produced OSTα and OSTβ antibodies, immunolocalization experiments revealed that human OSTα and OSTβ protein levels were detectable in the steroidogenic Purkinje cells and cornu ammonis cells of the cerebellum and hippocampus, respectively, as well as in the zona reticularis of the human adrenal gland (Fang, et al., 2010). In another study with healthy human subjects and using custom-made antibodies, western blot analysis demonstrated that OSTα protein levels were higher in the colon than the terminal ileum and duodenum (Schwarz, 2012). More recently, a quantitative proteomic analysis identified similarly high levels of OSTα and OSTβ protein in four jejunum and twelve ileum samples from macroscopically normal tissue; these tissues were resected from healthy subjects and individuals with underlying inflammatory bowel disease, colon cancer or ischemia (Couto, et al., 2020). Protein levels of OSTα and OSTβ were higher than P-glycoprotein (P-gp/MDR1/ABCB1). New antibodies have been produced as part of the Human Protein Atlas project (Uhlen, et al., 2015), which confirmed, using microarray-based immunohistochemistry, the protein expression of OSTα and OSTβ in several tissues and cell types in which expression was reported previously (Table 1). For example, expression of OSTα and/or OSTβ in hepatocytes, renal tubular cells and intestinal cells was reported before (Ballatori, et al., 2005), and confirmed in the Human Protein Atlas project. The project revealed the highest expression of OSTα protein (medium level) in the duodenum (small intestine), appendix and testis, while the highest expression of OSTβ protein (high level) was found in the duodenum (small intestine), followed by the appendix, colon, rectum and stomach (medium level). Expression of one or both subunits was detected in other tissues (low level), while undetectable levels were reported in the majority of tested cell types. Interestingly, while OSTα protein was detected previously in human cholangiocytes (Ballatori, et al., 2005), the Human Protein Atlas project reported undetectable levels in bile duct cells (Table 1). Furthermore, protein levels of OSTα and OSTβ were reported in cerebellar, hippocampal and adrenal glandular cells (Fang, et al., 2010), but neither of the subunits were detected in these cell types in the Human Protein Atlas project. It is unclear whether these discrepancies are due to antibody quality, or procedural and/or sample differences.

In addition to protein analyses in humans, early studies found OSTα and/or OSTβ protein expression in the rat and/or mouse small intestine, colon, kidney and cholangiocytes by immunoblot analysis and/or tissue immunolocalization (Ballatori, et al., 2005; Dawson, et al., 2005). Furthermore, murine OSTα and OSTβ protein localization was shown in Purkinje cells and cornu ammonis cells (Fang, et al., 2010).

2.3. Expression in ADME-Tox models

Cellular and in silico ADME-Tox models play an important role in drug development and pharmacology, allowing for the evaluation of ADME-Tox properties of drug candidates prior to undertaking time- and resource-intensive animal studies and clinical trials. These systems can be used to assess the impact of transporters (and metabolizing enzymes, among other factors) on the disposition of compounds, and data generated using in vitro and/or in silico models can be extrapolated to humans (Alqahtani, 2017). SLC51A and SLC51B mRNA has been detected in several cell lines widely used in drug development, including Caco-2, HEK 293, HeLa and HepG2, and in cell lines used as models for in vitro pharmacology studies, including MCF7, PC-3 and SH-SY5Y [Table 3; (Uhlen, et al., 2015)]. While SLC51A mRNA was found to be highest in REH cells (a childhood B acute lymphoblastic leukemia cell line), SLC51B transcripts were not detectable in these cells. Conversely, while SLC51B mRNA was highest in Caco-2, HepG2 and EFO-21 cells (epithelial colorectal adenocarcinoma, well-differentiated hepatocellular carcinoma, and metastatic ovarian serous cystadenocarcinoma cell lines, respectively), SLC51A mRNA was relatively low in these cells, albeit detectable. A similar trend between SLC51A and SLC51B gene expression levels in Caco-2 cells was observed in another study, in which SLC51B mRNA levels were ~11-fold higher than SLC51A levels (Li, et al., 2012). Similarly, in the human and rodent colon (Ballatori, et al., 2005), and particularly the human rectum (Table 2), SLC51B mRNA levels were found to be higher than SLC51A mRNA levels.

Table 3.

mRNA Expression of Human SLC51A and SLC51B in Various Cell Lines.

	SLC51A Level			SLC51B Level
Cell Line	TPM	pTPM	NX	TPM	pTPM	NX
A-431	0.4	0.5	1.0	0.0	0.0	0.0
A549	0.7	0.8	1.6	1.3	1.6	1.4
AF22	0.6	0.8	1.4	0.2	0.3	0.2
AN3-CA	0.7	0.9	1.7	0.7	0.9	0.7
ASC diff	0.0	0.0	0.0	0.0	0.0	0.0
ASC TERT1	0.3	0.3	1.0	0.1	0.1	0.1
BEWO	0.0	0.0	0.1	0.4	0.5	0.4
BJ	0.3	0.4	0.8	0.0	0.0	0.0
BJ hTERT+	0.0	0.0	0.0	0.0	0.0	0.0
BJ hTERT+ SV40 Large T+	0.0	0.0	0.1	0.2	0.2	0.2
BJ hTERT+ SV40 Large T+ RasG12V	0.0	0.0	0.0	0.0	0.0	0.0
Caco-2	0.2	0.3	0.7	14.9	18.3	16.5
CAPAN-2	0.2	0.2	0.4	0.0	0.1	0.0
Daudi	0.0	0.0	0.1	0.0	0.1	0.1
EFO-21	0.8	1.0	1.9	11.0	13.5	11.1
fHDF/TERT166	0.4	0.4	1.1	0.0	0.0	0.0
HaCaT	0.3	0.4	0.9	0.0	0.0	0.0
HAP1	0.0	0.1	0.2	0.0	0.0	0.0
HBEC3-KT	0.0	0.0	0.0	0.0	0.0	0.0
HBF TERT88	0.0	0.0	0.0	0.0	0.0	0.0
HDLM-2	0.3	0.4	0.7	0.1	0.1	0.1
HEK 293	1.2	1.5	2.8	0.2	0.3	0.2
HEL	0.4	0.5	1.0	0.0	0.0	0.0
HeLa	0.2	0.3	0.8	0.1	0.1	0.1
HepaRG	Detected*			Detected*
HepG2	0.5	0.7	1.7	10.4	12.7	13.8
HHSteC	0.1	0.1	0.2	0.1	0.1	0.1
HL-60	0.5	0.6	1.4	0.1	0.2	0.2
HMC-1	0.5	0.7	1.4	0.0	0.0	0.0
HSkMC	0.1	0.1	0.3	0.0	0.0	0.0
hTCEpi	0.0	0.0	0.0	0.0	0.0	0.0
hTEC/SVTERT24-B	0.0	0.1	0.2	0.1	0.1	0.1
hTERT-HME1	0.4	0.5	1.3	0.0	0.0	0.0
HuH-7	Detected*			Detected*
HUVEC TERT2	0.0	0.1	0.2	0.0	0.0	0.0
K-562	0.1	0.1	0.2	0.2	0.2	0.2
Karpas-707	1.0	1.7	3.3	0.1	0.2	0.2
LHCN-M2	0.0	0.0	0.0	0.0	0.0	0.0
MCF7	0.2	0.3	0.6	0.3	0.4	0.4
MOLT-4	0.0	0.0	0.1	0.0	0.1	0.0
NB-4	0.3	0.4	1.2	0.2	0.3	0.3
NTERA-2	0.0	0.0	0.1	0.4	0.5	0.4
PC-3	0.8	1.0	2.0	0.4	0.5	0.4
REH	3.4	4.4	7.9	0.0	0.0	0.0
RH-30	0.0	0.0	0.1	0.0	0.0	0.0
RPMI-8226	0.1	0.2	0.5	0.6	0.8	0.8
RPTEC TERT1	0.0	0.0	0.0	0.3	0.3	0.4
RT4	0.3	0.4	0.9	0.0	0.0	0.0
SCLC-21H	0.2	0.2	0.3	0.8	1.0	0.6
SH-SY5Y	0.2	0.3	0.5	0.5	0.7	0.5
SiHa	0.2	0.3	0.5	0.0	0.0	0.0
SK-BR-3	1.1	1.3	3.2	0.0	0.0	0.0
SK-MEL-30	0.1	0.1	0.3	0.0	0.0	0.0
T-47d	1.0	1.2	2.3	0.4	0.5	0.4
THP-1	0.1	0.2	0.5	0.0	0.0	0.0
TIME	0.2	0.2	0.5	0.0	0.0	0.0
U-138 MG	0.7	0.9	1.6	0.3	0.3	0.3
U-2 OS	0.3	0.4	0.7	0.0	0.0	0.0
U-2197	0.5	0.7	1.3	0.1	0.1	0.1
U-251 MG	0.3	0.4	0.9	0.1	0.1	0.1
U-266/70	1.1	1.8	3.3	0.1	0.2	0.2
U-266/84	1.3	2.0	4.0	0.2	0.4	0.3
U-698	0.2	0.2	0.6	0.1	0.2	0.2
U-87 MG	0.5	0.6	1.1	0.0	0.1	0.0
U-937	0.2	0.3	0.6	0.1	0.1	0.1
WM-115	0.2	0.3	0.6	0.1	0.2	0.1

Data on transcript expression levels per gene in 64 cell lines were obtained from The Human Protein Atlas version 19 and Ensembl version 92.38, and were determined using next-generation sequencing-based RNA-sequencing. Values are presented as transcripts per million (TPM) RNA molecules per sample, protein-coding transcripts per million (pTPM) and normalized expression (NX). NX data for every gene in each sample were obtained by normalizing TPM data using trimmed mean of M values (Robinson & Oshlack, 2010), followed by Pareto scaling (van den Berg, Hoefsloot, Westerhuis, Smilde, & van der Werf, 2006). A color scale was applied to each column separately, with green and red denoting the highest and lowest levels of expression, respectively.

^*HepaRG and HuH-7 cell lines both express SLC51A and SLC51B mRNA (Landrier, et al., 2006; Sissung, et al., 2019), but were not included in the current dataset from the Human Protein Atlas.

Caco-2 cells form tight junctions and resemble enterocytes when appropriately differentiated and polarized. OSTα/β has been suggested to facilitate basolateral uptake of TCA and ES transport in Caco-2 cells (Grandvuinet & Steffansen, 2011), while inhibitors of OSTα/β reduced basolateral uptake of ES in these cells (Grandvuinet, Gustavsson, & Steffansen, 2013). A role for OSTα/β as an uptake transporter in the basolateral-to-apical transport of rosuvastatin was speculated in Caco-2 cells (Li, et al., 2012). Furthermore, a novel physiologically-based pharmacokinetic model of rosuvastatin disposition incorporating contributions of OSTα/β among a few other transporters improved predictions of interactions with rifampin, compared to previous models in which OSTα/β was not included (Wang, Zheng, & Leil, 2017). Studies using P-gp-expressing cells, including Caco-2 cells and human primary proximal tubule cells, have shown that the kinetic models for the P-gp substrate digoxin needed to incorporate one or two basolateral uptake transporters in addition to a passive permeability component in order to adequately describe the observed data (Chaudhry, et al., 2018; Ellens, Meng, Le Marchand, & Bentz, 2018). A role for OSTα/β as an uptake transporter has been proposed for the interpretation of kinetic digoxin data in human primary proximal tubule cells in the presence of verapamil (Ellens, et al., 2018), and OSTα/β may play a role in digoxin disposition in Caco-2 cells; single concentration studies in X. laevis oocytes suggested that digoxin was an OSTα/β substrate (Seward, et al., 2003; Wang, et al., 2001).

The hepatoma cell lines HepG2 as well as HuH-7 and HepaRG cells (two additional well-differentiated hepatocellular carcinoma cell lines) express SLC51A and SLC51B mRNA in standard cell culture conditions. Furthermore, HepG2 and HuH-7 cells express OSTα and OSTβ at the protein level (Boyer, et al., 2006; Landrier, Eloranta, Vavricka, & Kullak-Ublick, 2006; Malinen, Ito, Kang, Honkakoski, & Brouwer, 2019; Schaffner, et al., 2015; Sissung, et al., 2019; Xu, Sun, & Suchy, 2014), and have been used to study this transporter. Particularly the effect of farnesoid X receptor (FXR) ligands on SLC51A/B mRNA and OSTα/β protein expression has been evaluated in HepG2 and HuH-7 cells (Boyer, et al., 2006; Landrier, et al., 2006; Schaffner, et al., 2015; Soroka, Xu, Mennone, Lam, & Boyer, 2008; Xu, et al., 2014). Furthermore, protein expression of OSTα, OSTβ, and the organic anion transporting polypeptide (OATP/SLCO) 1B3, as well as transport of the SLC substrate dehydroepiandrosterone sulfate (DHEAS) increased over time in HuH-7 cell cultures (Malinen, Ito, et al., 2019).

3. Endogenous and exogenous OSTα/β substrates and inhibitors

OSTα/β is an important transporter for bile acids and other steroid substrates. The first OSTα/β-mediated transport studies were performed in X. laevis oocytes injected with synthetic transcripts coding for skate OSTα and OSTβ protein (Wang, et al., 2001); some endogenous and exogenous OSTα/β substrates and inhibitors were identified in these studies. Subsequent studies have discovered several novel OSTα/β substrates and inhibitors.

3.1. OSTα/β substrates

In X. leavis oocytes, the heteromeric OSTα/β transported the endogenous compounds TCA, ES and prostaglandin E₂, but p-aminohippurate and S-dinitrophenyl glutathione were not substrates (Seward, et al., 2003; Wang, et al., 2001). Subsequent studies in mice and oocytes confirmed TCA and ES as OSTα/β substrates, and identified DHEAS as a substrate (Ballatori, et al., 2008; Fang, et al., 2010). Pregnenolone sulfate was shown to be an OSTα/β substrate, although pregnenolone and dehydroepiandrosterone were not (Fang, et al., 2010). Furthermore, individual bile acid species have been evaluated as OSTα/β substrates, consistent with the important role this transporter plays in bile acid homeostasis. OSTα/β transported unconjugated, taurine- and glycine-conjugated forms of cholate, chenodeoxycholate (CDCA), deoxycholate (DCA) and lithocholate, as well as the taurine- and glycine-conjugated forms of ursodeoxycholate (Table 4) (Ballatori, et al., 2005; Suga, Yamaguchi, Ogura, & Mano, 2019).

Table 4.

In Vitro Substrates of OSTα/β and Other Transporters.

OSTα/β Substrate	OSTα/β Km (μM)	OSTα/β-expressing System	Also a Substrate for These Transporters	References
1. Bile Acids
CA	>400*	MDCKII-transfected	NTCP, ASBT, OATP1B1, OATP1B3, MRP4, BCRP	(Balakrishnan, Wring, & Polli, 2006; Blazquez, et al., 2012; Craddock, et al., 1998; Cui, Konig, Leier, Buchholz, & Keppler, 2001; Dong, Ekins, & Polli, 2015; Ho, Leake, Roberts, Lee, & Kim, 2004; Rius, Hummel-Eisenbeiss, Hofmann, & Keppler, 2006; Suga, et al., 2019; Suga, et al., 2017)
CDCA	23.0	MDCKII-transfected	OATP1B1, OATP1B3	(Suga, et al., 2019; Suga, et al., 2017)
DCA	14.9	MDCKII-transfected	OATP1B1, OATP1B3	(Suga, et al., 2019; Suga, et al., 2017)
GCA	>1,000*	MDCK-transfected; MDCKII-transfected	NTCP, ASBT, OATP1B1, OATP1B3, BSEP, MRP3, MRP4, BCRP	(Balakrishnan, et al., 2006; Ballatori, et al., 2005; Blazquez, et al., 2012; Hayashi, et al., 2005; Kramer, et al., 1999; Meier, Eckhardt, Schroeder, Hagenbuch, & Stieger, 1997; Rius, et al., 2006; Suga, et al., 2019; Suga, et al., 2017; Zeng, Liu, Rea, & Kruh, 2000)
GCDCA	864.2	MDCK-transfected; MDCKII-transfected	NTCP, ASBT, OATP1B1, OAT 1B3, MRP4, BSEP	(Balakrishnan, et al., 2006; Ballatori, et al., 2005; Craddock, et al., 1998; Dong, et al., 2015; Hayashi, et al., 2005; Rius, et al., 2006; Suga, et al., 2019; Suga, et al., 2017)
GDCA	586.4	MDCK-transfected, MDCKII-transfected	ASBT, OATP1B1, OATP1B3, MRP4	(Balakrishnan, et al., 2006; Ballatori, et al., 2005; Craddock, et al., 1998; Rius, et al., 2006; Suga, et al., 2019; Suga, et al., 2017)
GLCA	12.8	MDCKII-transfected	NTCP, ASBT, OATP1B1, OATP1B3	(Balakrishnan, et al., 2006; Dong, et al., 2015; Suga, et al., 2019; Suga, et al., 2017)
GUDCA	>1,000*	MDCK-transfected, MDCKII-transfected	NTCP, ASBT, OATP1B1, OATP1B3, BSEP, MRP4	(Balakrishnan, et al., 2006; Ballatori, et al., 2005; Craddock, et al., 1998; Dong, et al., 2015; Hayashi, et al., 2005; Maeda, Kambara, Tian, Hofmann, & Sugiyama, 2006; Rius, et al., 2006; Suga, et al., 2019; Suga, et al., 2017)
LCA	NA	MDCKII-transfected	-	(Suga, et al., 2019)
TCA	351–698; >10,000*	X. laevis oocytes-injected, MDCK-transfected, HeLa-transfected; COS-transfected; Flp-In 293-transfected; MDCK-transfected; MDCKII-transfected	NTCP, ASBT, OAT3, OATP1A2, OATP1B1, OATP1B3, OATP2B1, BSEP, MRP3, MRP4, BCRP	(Abe, et al., 2001; Balakrishnan, et al., 2006; Ballatori, et al., 2005; Blazquez, et al., 2012; Cha, et al., 2001; Choi, et al., 2011; Craddock, et al., 1998; Hallen, Bjorquist, Ostlund-Lindqvist, & Sachs, 2002; Hayashi, et al., 2005; Kim, et al., 1999; Kramer, et al., 1999; Leslie, Watkins, Kim, & Brouwer, 2007; Love, et al., 2001; Malinen, et al., 2018; Meier, et al., 1997; Nozawa, Imai, Nezu, Tsuji, & Tamai, 2004; Rius, et al., 2006; Seward, et al., 2003; Suga, et al., 2019; Suga, et al., 2017; Sultan, et al., 2018; van de Wiel, et al., 2018; Visser, et al., 2010; Wang, et al., 2001; Zhang, et al., 2003)
TCDCA	723.7	MDCK-transfected; MDCKII-transfected	NTCP, ASBT, OATP1B1, OATP1B3, MRP4, BSEP	(Balakrishnan, et al., 2006; Ballatori, et al., 2005; Dong, et al., 2015; Kis, et al., 2009; Kramer, et al., 1999; Meier, et al., 1997; Rius, et al., 2006; Suga, et al., 2019; Suga, et al., 2017)
TDCA	>2,000*	MDCK-transfected, MDCKII-transfected	ASBT, OATP1B1, OATP1B3, BSEP	(Balakrishnan, et al., 2006; Ballatori, et al., 2005; Hayashi, et al., 2005; Kramer, et al., 1999; Suga, et al., 2019; Suga, et al., 2017)
TLCA	23.9	MDCKII-transfected	NTCP, ASBT, OATP1B1, OATP1B3	(Balakrishnan, et al., 2006; Dong, et al., 2015; Suga, et al., 2019; Suga, et al., 2017)
TUDCA	>2,000*	MDCK-transfected; MDCKII-transfected	NTCP, ASBT, OATP1A2, OATP1B1, OATP1B3, BSEP, MRP4	(Balakrishnan, et al., 2006; Ballatori, et al., 2005; Dong, et al., 2015; Hayashi, et al., 2005; Kramer, et al., 1999; Maeda, et al., 2006; Meier, et al., 1997; Rius, et al., 2006; Suga, et al., 2019; Suga, et al., 2017)
2. Other Endogenous Compounds
DHEAS	1.5	X. laevis oocytes-injected; Flp-In 293-transfected	NTCP, OAT2, OAT3, OAT4, OAT7, OATP1A2, OATP1B1, OATP2B1, MRP1, MRP2, MRP4, BCRP	(Ballatori, et al., 2005; Burckhardt, 2012; Cha, et al., 2001; Cha, et al., 2000; Cui, et al., 2001; Fang, et al., 2010; Grube, et al., 2007; Hagenbuch & Stieger, 2013; Kobayashi, et al., 2005; Kullak-Ublick, et al., 1998; Malinen, Kauttonen, et al., 2019; Miyajima, Kusuhara, Fujishima, Adachi, & Sugiyama, 2011; Pfeifer, Bridges, Ferslew, Hardwick, & Brouwer, 2013; Pizzagalli, et al., 2003; Yamada, et al., 2007; Zelcer, et al., 2003)
ES	320	X. laevis oocytes-injected; Flp-In 293-transfected	NTCP, OAT2, OAT3, OAT4, OAT7, MATE1, MATE2-K, MRP1, BCRP	(Ballatori, et al., 2005; Burckhardt, 2012; Burckhardt & Burckhardt, 2011; Cha, et al., 2000; Cui, et al., 2001; Grube, et al., 2007; Hagenbuch & Stieger, 2013; Hirano, Maeda, Shitara, & Sugiyama, 2006; Ho, et al., 2004; Imai, et al., 2003; Kobayashi, et al., 2005; Kullak-Ublick, et al., 2001; Lu, Gonzalez, & Klaassen, 2010; Malinen, et al., 2018; Miyazaki, et al., 2005; Noe, Portmann, Brun, & Funk, 2007; Nozawa, et al., 2004; Pizzagalli, et al., 2003; Qian, Song, Cui, Cole, & Deeley, 2001; Seward, et al., 2003; Terada & Inui, 2008; Tsuda, et al., 2007; Wang, et al., 2001; Yamaguchi, et al., 2010; Yamashita, et al., 2006; Zhou, Tanaka, Pan, Ma, & You, 2004)
PGE2	NA	X. laevis oocytes-injected	OCT1, OCT2, OAT1, OAT2, OAT3, OAT4, OATP2B1, MRP4	(Kimura, et al., 2002; Nishio, et al., 2000; Reid, et al., 2003; Seward, et al., 2003; Wang, et al., 2001)
PREGS	6.9	X. laevis oocytes-injected	NTCP, OAT4, OATP2B1	(Fang, et al., 2010; Grube, Köck, Karner, et al., 2006; Kimura, et al., 2002)
3. Drugs
Atorvastatin	NA	HeLa-transfected	NTCP, OATP1B1, OATP1B3, OATP2B1, P-gp, MRP2, BCRP	(Choi, et al., 2011; Ellis, Hawksworth, & Weaver, 2013; Grube, Köck, Oswald, et al., 2006; Hochman, et al., 2004; Kameyama, Yamashita, Kobayashi, Hosokawa, & Chiba, 2005; Karlgren, et al., 2012; Keskitalo, et al., 2009; Lau, Huang, Frassetto, & Benet, 2007; Schwarz, 2012)
Digoxin	NA	X. laevis oocytes-injected	OATP1B3, OATP4C1, P-gp, MDR3	(Kullak-Ublick, et al., 2001; Seward, et al., 2003; Smith, et al., 2000; Troutman & Thakker, 2003; Wang, et al., 2001)
Docetaxel	NA	HeLa-transfected	OATP1A2, OATP1B1, OATP1B3, P-gp	(de Graan, et al., 2012; Durmus, et al., 2014; Iusuflfet al., 2015; Lee, Leake, Kim, & Ho, 2017; Schwarz, 2012; Shirakawa, et al., 1999; Yamaguchi, et al., 2008)
Rosuvastatin	NA	HeLa-transfected	NTCP, OAT3, OATP1A2, OATP1B1, OATP1B3, OATP2B1, P-gp, MRP2, MRP4, BCRP	(Choi, et al., 2011; Dong, et al., 2015; Ellis, et al., 2013; Ho, et al., 2006; Keskitalo, et al., 2009; Kikuchi, et al., 2006; Kitamura, Maeda, Wang, & Sugiyama, 2008; Lu, et al., 2015; Pfeifer, et al., 2013; Schwarz, 2012; Windass, Lowes, Wang, & Brown, 2007)
Sulfasalazine	NA	HeLa-transfected	OATP2B1, MRP2, BCRP	(Schwarz, 2012; Urquhart, et al., 2008; Kusuhara, et al., 2012; Dahan & Amidon, 2009)

^*Reported K_m values were higher than the maximum substrate concentration tested. CA, cholate; CDCA, chenodeoxycholate; DHEAS, dehydroepiandrosterone sulfate; ES, estrone sulfate; GCA, glycocholate; GCDCA, glycochenodeoxycholate; GDCA, glycodeoxycholate; GLCA, glycolithocholate; GUDCA, glycoursodeoxycholate; K_m, substrate concentration at one-half of the maximum velocity; LCA; lithocholate; NA, data not available in the literature; PGE₂, prostaglandin E₂; PREGS, pregnenolone sulfate; TCA, taurocholate; TCDCA, taurochenodeoxycholate; TDCA, taurodeoxycholate; TLCA, taurolithocholate; TUDCA, tauroursodeoxycholate

Single concentration studies in X. laevis oocytes suggested that digoxin is an OSTα/β substrate (Seward, et al., 2003; Wang, et al., 2001). In a study with Caco-2 cells and OSTα/β inhibitors causing cis-inhibition (i.e., inhibition from the extracellular side of the cell), a plausible role for OSTα/β in mediating the basolateral-to-apical transport of rosuvastatin was inferred (Li, et al., 2012). This speculation is supported by experiments showing that rosuvastatin uptake was increased in OSTα/β-overexpressing HeLa cells compared to control cells (Schwarz, 2012). However, the role of OSTα/β in rosuvastatin absorption was less clear based on studies in OSTα knockout mice (Schwarz, 2012). Atorvastatin, sulfasalazine and docetaxel, but not the structurally related paclitaxel, were transported by OSTα/β in HeLa cells (Schwarz, 2012).

Compounds that have been identified as OSTα/β substrates are also substrates for other SLC and some ABC transporters (Table 4). In terms of transport of exogenous compounds, OSTα/β is most similar to OATP1B3 and P-gp, which also transport atorvastatin, rosuvastatin, docetaxel and digoxin. With regards to endogenous compounds, OATP1B1 and OATP1B3 typically transport OSTα/β substrates, but Na⁺-taurocholate co-transporting polypeptide (NTCP/SLC10A1), ASBT, the multidrug resistance-associated protein (MRP/ABCC) 4, breast cancer resistance protein (BCRP/ABCG2) and the bile salt export pump (BSEP/ABCB11) also have affinity for multiple OSTα/β substrates.

OSTα/β is reported to be a low affinity/high capacity transporter for multiple substrates, particularly bile acid species (Malinen, et al., 2018; Suga, et al., 2019). In systems in which both high affinity/low capacity transporters and low affinity/high capacity transporters are present, high affinity transporters typically play a more dominant role in transport at low substrate concentrations (Lin & Smith, 1999). However, when substrate concentrations are substantially elevated, high capacity transporters become more dominant. In situations when high affinity transporters are not able to handle increased substrate concentrations, a high capacity transporter may be induced to become the dominant transporter for substrates in that system. For example, the induction of OSTα/β in hepatocytes under cholestatic conditions (see Section 8) suggests that when hepatocellular bile acids are elevated, OSTα/β is upregulated and becomes the dominant transporter to efflux bile acids across the basolateral plasma membrane.

3.2. OSTα/β inhibitors

More compounds have been evaluated as inhibitors of OSTα/β than as substrates. Thus far, relatively few compounds have been shown to inhibit OSTα/β compared to other SLC transporters. Initial studies in X. laevis oocytes using TCA and ES as the substrates suggested that spironolactone, digoxin, probenecid and indomethacin, in addition to various endogenous bile acids (e.g., the skate bile acid scymnol sulfate) and other steroid molecules, were OSTα/β inhibitors at concentrations ≥200 μM; bromosulfophthalein inhibited OSTα/β at 100 μM (Wang, et al., 2001). However, these compounds were tested only at a single concentration. These findings were largely reproduced in a follow-up study with X. laevis oocytes (Seward, et al., 2003). The inhibitory effect of digoxin, bromosulfophthalein and probenecid on OSTα/β-mediated TCA uptake was not reproduced in a more recent study using OSTα/β-overexpressing Flp-In 293 cells, whereas spironolactone and indomethacin inhibited OSTα/β function (Malinen, et al., 2018). In addition to these compounds, the BSEP and MRP4 inhibitor troglitazone sulfate (Funk, Ponelle, Scheuermann, & Pantze, 2001; Yang, Pfeifer, Köck, & Brouwer, 2015), the MRP3 inhibitor fidaxomicin (Ali, Welch, Lu, Swaan, & Brouwer, 2017) and the BSEP inhibitor ethinyl estradiol (Morgan, et al., 2013) inhibited OSTα/β-mediated DHEAS and/or TCA uptake (Malinen, et al., 2018; Malinen, Kauttonen, et al., 2019). In a fluorescent resonance energy transfer-based, high-throughput screen, 1,280 compounds were tested as OSTα/β inhibitors using taurochenodeoxycholate as the substrate, but only a single compound, clofazimine, was found to be an OSTα/β-specific inhibitor (van de Wiel, de Waart, Oude Elferink, & van de Graaf, 2018). Additional studies revealed that 25 μM sulfasalazine, and both unconjugated and glucuronidated ezetimibe inhibited OSTα/β-mediated transport of TCA (Schwarz, 2012).

Several purported OSTα/β substrates (e.g., digoxin, sulfasalazine, multiple bile acid species) also inhibit transport of other OSTα/β substrates, most likely via competitive inhibition, although most of these studies evaluated inhibitors only at a single concentration. Theoretically, all substrates become (competitive) inhibitors at sufficiently high concentrations. To the knowledge of the authors, the majority of the inhibitors listed in this section have not been studied as substrates. One explanation for this is that substrate quantification requires compound-specific analytical methods or generation of a stable, labeled substrate.

3.3. Current limitations in studying OSTα/β substrates and inhibitors

The published in vitro OSTα/β studies attempting to identify OSTα/β substrates or inhibitors may have been limited by a variety of factors. For instance, the functional evaluation of one transporter may be confounded when the in vitro system is influenced by a second transporter such as co-expression systems involving OSTα/β and ABST in MDCK-II epithelial cells (Ballatori, et al., 2005; van de Wiel, et al., 2018), or OSTα/β and NTCP in U-2 OS (van de Wiel, et al., 2018) or HeLa (Schwarz, 2012) cells, even when cells expressing only ASBT or NTCP are used as controls, respectively. This is especially important to consider because OSTα/β can function as both an uptake and efflux transporter. A novel fluorescent resonance energy transfer-based bile acid sensor (van de Wiel, et al., 2018) elegantly measured bile acid-mediated intracellular activation of FXR, but evaluated the inhibition of OSTα/β-mediated bile acid transport only indirectly using taurochenodeoxycholate as the substrate. Additionally, since there is a relationship between transporter expression and kinetic parameters (Balakrishnan, et al., 2007; Kalvass & Pollack, 2007; Tachibana, et al., 2010), a variable extent of transporter expression at the plasma membrane such as in the OSTα/β-overexpressing Flp-In 293 cells (Malinen, et al., 2018; Malinen, Kauttonen, et al., 2019) can lead to under- or overestimation of substrate parameters such as the maximum transport rate achieved (V_max), or inhibitor parameters such as the half-maximal inhibitory concentration (IC₅₀). Additionally, poor solubility of a substrate may limit the range of concentrations that can be studied, and hamper accurate determination of kinetic parameters of the transporter in the particular cell system.

4. Transport mechanism of OSTα/β

4.1. Transport direction in vivo and in vitro

OSTα/β belongs to the SLC transporter family, and it is thought to function primarily as an efflux transporter in vivo on the basolateral membrane of enterocytes involved in the enterohepatic recycling of bile acids (Dawson, et al., 2010), and in the cholestatic liver to protect hepatocytes from toxic bile acid accumulation (Boyer, et al., 2006; Chai, et al., 2015; Malinen, et al., 2018). OSTα/β expressed on the basolateral membrane of kidney cells may play a role in salvaging bile acids that escaped hepatic extraction (Dawson, et al., 2010). Interestingly, however, in OSTα^−/− mice that underwent bile duct ligation, adaptive responses in the kidney, including reduced apical ASBT and increased apical MRP2 and MRP4 protein levels, resulted in increased urinary excretion of bile acids; no compensatory increase in basolateral MRP3 was observed in the kidneys of these mice (Soroka, Mennone, Hagey, Ballatori, & Boyer, 2010).

Despite the hypothesized, primary role of OSTα/β as an efflux transporter in vivo, the majority of OSTα/β-based transport studies in vitro have evaluated the uptake function of OSTα/β. Some studies utilizing cells expressing both OSTα/β and a different transporter capable of uptake [e.g., ASBT or NTCP (Ballatori, et al., 2005; Schwarz, 2012; van de Wiel, et al., 2018)] have attempted to analyze the efflux direction of OSTα/β by comparing transport in OSTα/β/ASBT- or OSTα/β/NTCP-expressing cells with cells only expressing ASBT or NTCP, respectively. In addition, a study evaluating the hepatobiliary uptake and efflux kinetics of TCA disposition showed that human hepatocytes in which OSTα/β expression was induced exhibited significantly increased basolateral efflux clearance of TCA (Guo, LaCerte, Edwards, Brouwer, & Brouwer, 2018). Although other factors could have contributed to the observed increase in basolateral efflux of TCA, these data suggest a plausible role for OSTα/β as a bile acid efflux transporter in hepatocytes under the conditions used in this study.

Some transporters demonstrate symmetric transport (e.g., the same K_m for both uptake and efflux directions); however, this is not necessarily the case for all bidirectional transporters (Baird, et al., 2004; Bosdriesz, et al., 2018; Elbing, et al., 2004; Maier, Volker, Boles, & Fuhrmann, 2002). It is unclear how the uptake kinetics of OSTα/β compares to the efflux kinetics. Examination of differences in OSTα/β-mediated substrate transport in both directions warrants further investigation, since kinetic parameters determined for the uptake direction may not accurately reflect the transporter’s function in the physiologically more relevant efflux direction. Unfortunately, to date it has been inherently more complex to study a bidirectional (ATP-independent) transporter in isolation in the efflux direction, because it involves loading cells with a probe substrate prior to initiation of the efflux phase, which may introduce additional experimental variability. The existing in vitro methodologies for evaluation of efflux kinetics result in high system-to-system and lab-to-lab variability (Heikkinen, Korjamo, Lepikko, & Monkkonen, 2010; Korjamo, Kemilainen, Heikkinen, & Monkkonen, 2007).

4.2. Driving force for transport

Experiments in X. laevis oocytes indicated that OSTα/β transport operated by facilitated diffusion (Ballatori, et al., 2005; Seward, et al., 2003). However, the question remains whether OSTα/β functions as a uniporter, symporter, or antiporter. Early studies suggested that ion (Na⁺, K⁺, H⁺, Cl⁻) gradients, ATP and pH (Ballatori, et al., 2005; Seward, et al., 2003; Wang, et al., 2001) did not alter OSTα/β-mediated uptake of model substrates, implying that OSTα/β most likely functions as a uniporter that is regulated by the substrate concentration on either side of the plasma membrane. However, recent functional OSTα/β in vitro studies have been performed in K⁺-rich buffers (Malinen, Kauttonen, et al., 2019; Sultan, et al., 2018), sometimes in the complete absence of Na⁺. Extracellular replacement of sodium chloride (NaCl) with choline chloride (C₅H₁₄NOCl) stimulated OSTα/β-mediated uptake of probe substrates in OSTα/β-overexpressing Flp-In 293 cells (Malinen, et al., 2018), suggesting that the regulation of OSTα/β transport may be ion-dependent. Furthermore, low extracellular pH conditions also stimulated OSTα/β-mediated TCA uptake in OSTα/β-overexpressing Flp-In 293 cells in addition to low Na⁺ conditions (Malinen, et al., 2018).

Transport mediated by some other members of the SLC transporter family is influenced by extracellular pH, including OATP (SLCO) (Kobayashi, et al., 2003; Leuthold, et al., 2009; Stieger & Hagenbuch, 2014) and monocarboxylate transporters (MCT/SLC16A) (Halestrap & Price, 1999). Mechanisms of transport are fundamental to our understanding of individual transporters and transport protein families, and could be linked to health and disease, but there are still many knowledge gaps that need to be addressed in future investigations, particularly with respect to OSTα/β.

5. Structure of OSTα/β

In contrast to many well-studied transporters, OSTα/β consists of two different protein subunits, OSTα and OSTβ, encoded by the SLC51A and SLC51B genes, respectively. When OSTα/β was discovered in the little skate, OSTα protein consisting of 352 amino acids with seven putative transmembrane domains (TMDs) and OSTβ protein consisting of 182 amino acids and one or two putative TMDs were identified (Wang, et al., 2001). Human OSTα is a protein of 340 amino acids with seven potential TMDs, whereas OSTβ consists of 128 amino acids with a single predicted TMD (Christian, Li, Hinkle, & Ballatori, 2012). X-ray crystallography and high-resolution cryo-electron microscopy studies of OSTα/β are lacking, but valuable information on OSTα/β structure has been generated using other approaches, including topology-prediction algorithms, protein subunit expression and mutagenesis. Differences in the prediction algorithms of available topology-prediction models estimate five-to-seven TMDs in OSTα. However, the transmembrane hidden Markov model (Krogh, Larsson, von Heijne, & Sonnhammer, 2001), one of the most commonly applied topology-prediction tools used by the Universal Protein Resource (UniProt), among several other models predict that OSTα likely contains seven TMDs, and OSTβ only one (Figure 2). Although OSTα and OSTβ depend on each other for stable expression and function at the plasma membrane, the interaction of the two subunits is unresolved, and the exact stoichiometry of the OSTα/β heteromer is unknown. Protein expression studies suggest that OSTα/β may exist in the form of a heterodimer (i.e., OSTα-OSTβ), or as a heteromultimer consisting of several units of OSTα and/or OSTβ, such as OSTα-OSTα-OSTβ (Li, et al., 2007; Schwarz, 2012). Also, there is some evidence suggesting that OSTα is able to form homodimers (Li, et al., 2007). Whether disproportionate regulation in the expression of the OSTα and OSTβ subunits impacts the type of heteromeric interaction and function of OSTα/β is unknown.

Figure 2.

Membrane Topology of OSTα and OSTβ According to the Transmembrane Hidden Markov Model (TMHMM). The figures of the two p otein subunits in (A) were generated using Protter v1.0, an open-source tool for visualization of the extracellular, transmembrane and intracellular domains of membrane proteins. In (B), red text represents extracellular residues; bold black text represents transmembrane residues; blue text represents intracellular residues. *, amino acids with common missense mutations in the general population (see Tables 7 and 8); †, homozygous mutations p.Q186* (premature st p codon) in SLC51A and p.F27fs (frame shift leading to premature stop codon at position 50) in SLC51B were found in the first cases of OSTα (Gao, et al., 2019) and OSTβ (Sultan, et al., 2018) deficiency, respectively.

Several studies have evaluated the role of structural domains essential for the interaction between OSTα and OSTβ. The first of these studies examined the impact of truncating the amino- (N-) and carboxy- (C-)terminal fragments of OSTα on the interaction with OSTβ (Sun, et al., 2007). While the C-terminal tail (23 amino acids) and five amino acids of OSTα’s predicted last TMD were not essential for the interaction with OSTβ, truncation of the N-terminal tail (48 amino acids) and two amino acids of OSTα’s predicted first TMD eliminated the protein interaction, and resulted in intracellular accumulation of both subunits (Sun, et al., 2007).

When mouse OSTα and OSTβ are co-expressed, the OSTα subunit undergoes N-glycosylation-dependent maturation (Dawson, et al., 2005). N-glycosylation is typically associated with the Asn-X-Ser/Thr motif, where X can be any amino acid (Soroka, et al., 2008), except proline. While the mouse, rat and skate have this consensus motif in the N-terminus of the OSTα subunit, the human protein lacks this motif (Soroka, et al., 2008). N-glycosylation also appears to take place in human OSTα (Soroka, et al., 2008), as reported for some other proteins lacking the consensus motif (Valliere-Douglass, et al., 2009). Western blot experiments indicated that mouse and human OSTα have an unglycosylated form (~31 kDa), a core precursor, immature form (~35 kDa) found in the endoplasmic reticulum that is likely glycosylated, and a mature, complex glycoprotein (~40 kDa) (Dawson, et al., 2005; Soroka, et al., 2008). While the mature, glycosylated form of OSTα is generated upon co-expression of OSTβ, immunoprecipitation experiments and studies with the glycosylation inhibitor tunicamycin have shown that OSTα glycosylation is not required for heteromerization with OSTβ, nor for expression at the plasma membrane (Dawson, et al., 2005; Soroka, et al., 2008).

Multiple research groups have examined the structural domains of the relatively small OSTβ subunit. Two studies have indicated that the N-terminal tail of OSTβ is critical for transport activity and membrane localization (Sun, et al., 2012; Xu, et al., 2016). The deletion of 35 amino acids of OSTβ’s N-terminus prevented proper interaction with OSTα and expression at the membrane (Sun, et al., 2012). In a follow-up study, substituting the two leucines in the highly conserved acidic di-leucine motif (-EL₂₀L₂₁EE) at the extracellular N-terminus of OSTβ with two alanines abolished OSTβ’s association with OSTα, and resulted in a lack of expression of both subunits at the plasma membrane (Xu, et al., 2016). However, a more recent study with chimeras of OSTβ and the single TMD receptor activity-modifying protein 1 suggests that OSTβ’s TMD is important in the interaction with OSTα and for maintaining OSTα/β transport function (Christian & Hinkle, 2017). Replacement of OSTβ’s extracellular N-terminal domain by a segment of receptor activity-modifying protein 1 resulted in no loss of function, while the cytoplasmic C-terminal domain demonstrated an involvement with OSTα association (Christian & Hinkle, 2017). Finally, there is clinical evidence from two brothers with OSTβ deficiency (leading to congenital diarrhea and mild cholestasis) that OSTβ’s amino acid sequence from codon position 27 onward, which includes a portion of OSTβ’s extracellular N-terminus as well as OSTβ’s entire predicted TMD and C-terminus, is important for OSTα/β protein expression and transport activity (Sultan, et al., 2018).

More work is needed to elucidate the exact stoichiometry of OSTα/β and the roles and structures of the individual subunits, including the identification of potential sites for, and types of, post-translational modifications besides glycosylation. Molecular modeling approaches may lead to a better understanding of OSTα/β function in health and disease. For example, crystal structures of acetyl CoA synthetase and leucyl-tRNA synthetase have been used to construct a homology model of human OSTβ to clarify the impact of a synthetic mutation (Xu, et al., 2016). Establishing the overall structure of OSTα/β will likely facilitate ongoing and stimulate novel investigations of OSTα/β, and aid in predicting the potential for drug- and/or disease-mediated alterations in function.

6. Genetic variation in SLC51A and SLC51B

Transcript ENST00000296327 is the protein-coding SLC51A transcript with the highest overall expression in the human body (Table 2); this transcript codes for the 340 amino acid form of OSTα. Three other SLC51A transcripts (ENST00000415111, ENST00000416660 and ENST00000428985) also contain an open reading frame, but it is unclear whether these shorter protein-coding sequences (23, 66 and 212 amino acids, respectively) result in functional protein. OSTβ only has one reported transcript (ENST00000334287). Numerous common and rare variants have been detected in SLC51A and SLC51B by the 1000 Genomes Project, the Exome Aggregation Consortium, and others. These variants are located in many types of transcript structure locations, with intron variants being the most common, followed by 239 SLC51A and 92 SLC51B missense (i.e., resulting in an amino acid substitution) variants (Table 5). Some of these are located within the splice site region, within 1–3 bases of the exon boundary. A variety of tools have been employed to predict the functional consequences of these missense variants, including Sorting Intolerant From Tolerant [SIFT; (Kumar, Henikoff, & Ng, 2009)], Polymorphism Phenotyping v2 [PolyPhen-2; (Adzhubei, et al., 2010)], Rare Exome Variant Ensemble Learner [REVEL; (Ioannidis, et al., 2016)] and MutationAssessor (Reva, Antipin, & Sander, 2011) that use tool-specific algorithms to arrive at predictions (Table 6). Only 1 out of 239 SLC51A missense variants (rs939885, V202I) and 1 out of 92 SLC51B missense variants (rs537053592, D10A) are common with a minor allele frequency (MAF) ≥ 0.01 (Tables 7–8). Both of these are predicted to be tolerated by SIFT, benign by PolyPhen-2, likely benign by REVEL, and neutral by MutationAssessor. Only some rare variants found in both genes are predicted to have potentially harmful consequences to the function of the protein. Although these rare variants have been detected in a few individuals (e.g., by the Exome Aggregation Consortium), their health status related to bile acid homeostasis or liver disease is unknown. Further work is needed to obtain more information on the physiological/pathophysiological consequences of the different variants found in both genes.

Table 5.

Distinct Genetic Variants Reported in the General Population in SLC51A and SLC51B Categorized by Variant Type.

Genetic Variant Type	SLC51A	SLC51B
3’ UTR Variant	44	73
5’ UTR Variant	83	68
Coding Sequence Variant	68	26
Frameshift Variant	10	9
Inframe Deletion	5	2
Inframe Insertion	2	1
Intron Variant	4,263	1,798
Missense Variant	226	89
Missense Variant~Splice Region Variant	13	3
Protein Altering Variant	1	0
Splice Acceptor Variant	4	3
Splice Acceptor Variant~Intron Variant	2	0
Splice Donor Variant	7	2
Splice Region Variant~5’ UTR Variant	0	2
Splice Region Variant~Coding Sequence Variant	7	0
Splice Region Variant~Intron Variant	46	15
Splice Region Variant~Synonymous Variant	8	1
Start Lost	0	2
Start Retained Variant	1	0
Stop Gained	6	1
Stop Gained~Frameshift Variant	1	0
Stop Lost	1	0
Synonymous Variant	103	28
Total	4,901	2,123

The data were obtained from Ensembl version 92.38. UTR, untranslated region.

Table 6.

Tools Predicting the Functional Consequence of Distinct Missense Variants Reported in the General Population in SLC51A and SLC51B.

		SLC51A		SLC51B
Prediction Tool	Classification of Variant Tolerability	Missense Variant	Missense Variant~Splice Region Variant	Missense Variant	Missense Variant~Splice Region Variant
SIFT	Tolerated	104	6	52	0
	Tolerated - Low Confidence	1	0	0	0
	Deleterious - Low Confidence	1	0	0	0
	Deleterious	120	7	37	3
PolyPhen-2	Benign	113	7	55	1
	Possibly Damaging	38	2	19	1
	Probably Damaging	75	4	15	1
REVEL	NA	0	2	0	1
	Likely Disease Causing	14	2	0	0
	Neutral	9	0	17	1
MutAs	NA	0	2	0	1
	Low	41	0	39	0
	Medium	176	11	33	1

Data were obtained from Ensembl version 92.38. Predictions of whether the observed missense variants in SLC51A and SLC51B are harmful to the function of the protein were made based on four tools: SIFT, Sorting Intolerant From Tolerant; PolyPhen-2, Polymorphism Phenotyping v2; REVEL, Rare Exome Variant Ensemble Learner; MutAs, MutationAssessor. NA, not available.

Table 7.

Common Variants in SLC51A.

Variant ID	Alleles	MAF Allele	MAF	Class	Type	Variant ID	Alleles	MAF Allele	MAF	Class	Type
rs111754451	A/-	-	0.01	Deletion	Intron Variant	rs111996482	C/T	T	0.04	SNP	Intron Variant
rs59792018	C/G/T	G	0.01	SNP	Intron Variant	rs78843541	T/C	C	0.05	SNP	Intron Variant
rs559074935	A/G	G	0.01	SNP	Intron Variant	rs182482018	A/G	G	0.06	SNP	Intron Variant
rs144777514	T/C	C	0.01	SNP	Intron Variant	rs78622644	T/G	G	0.06	SNP	Intron Variant
rs12633332	G/C/T	T	0.01	SNP	Intron Variant	rs138047603	A/C	C	0.07	SNP	Intron Variant
rs74350809	G/A	A	0.01	SNP	Intron Variant	rs78736929	T/C/G	C	0.08	SNP	Intron Variant
rs73083261	C/T	T	0.011	SNP	Intron Variant	rs75096911	C/T	T	0.09	SNP	Intron Variant
rs115944258	G/A	A	0.011	SNP	Intron Variant	rs12638322	C/T	T	0.1	SNP	Intron Variant
rs78135195	G/A	A	0.011	SNP	Intron Variant	rs35551560	G/A/T	A	0.12	SNP	Intron Variant
rs115746134	A/G	G	0.011	SNP	Intron Variant	rs11185518	C/T	T	0.17	SNP	Intron Variant
rs143984572	A/G	G	0.012	SNP	Intron Variant	rs78771288	A/G	G	0.18	SNP	Intron Variant
rs9835817	T/C	C	0.012	SNP	Intron Variant	rs56030157	C/A/G/T	T	0.21	SNP	5’ UTR Variant
rs142534043	C/T	T	0.012	SNP	Intron Variant	rs78530091	C/T	T	0.24	SNP	Intron Variant
rs144713505	C/T	T	0.013	SNP	Intron Variant	rs4916521	G/A	A	0.25	SNP	Intron Variant
rs116735912	C/T	T	0.013	SNP	Intron Variant	rs35516030	G/-	-	0.26	Deletion	Intron Variant
rs146021046	G/T	T	0.013	SNP	Intron Variant	rs73212130	G/A/C	C	0.26	SNP	Intron Variant
rs147722295	C/T	T	0.013	SNP	Intron Variant	rs73083248	A/G	G	0.26	SNP	Intron Variant
rs367976475	C/T	T	0.013	SNP	Intron Variant	rs71323710	A/G/T	G	0.27	SNP	Intron Variant
rs115878588	G/A	A	0.014	SNP	Intron Variant	rs7429803	C/G	G	0.3	SNP	Intron Variant
rs111732200	C/T	T	0.014	SNP	Intron Variant	rs72611184	C/T	T	0.31	SNP	Intron Variant
rs116446766	C/T	T	0.015	SNP	Intron Variant	rs13064065	G/A	A	0.31	SNP	Intron Variant
rs191597067	A/G	G	0.015	SNP	Intron Variant	rs7625886	T/C	T	0.31	SNP	Intron Variant
rs138984296	G/A	A	0.015	SNP	Intron Variant	rs73083250	G/A	A	0.32	SNP	Intron Variant
rs73083235	G/T	T	0.016	SNP	Intron Variant	rs1543975	T/C	C	0.34	SNP	Intron Variant
rs138755866	G/A	A	0.016	SNP	Intron Variant	rs7641135	T/C	C	0.37	SNP	Intron Variant
rs76992230	C/T	T	0.017	SNP	Intron Variant	rs67044678	G/A	A	0.37	SNP	Intron Variant
rs111386747	C/T	T	0.017	SNP	Intron Variant	rs56867360	C/T	T	0.39	SNP	Intron Variant
rs114844186	G/A	A	0.018	SNP	Intron Variant	rs9840089	G/A	G	0.39	SNP	Intron Variant
rs73083243	G/A	A	0.019	SNP	Intron Variant	rs1875088	G/A	A	0.39	SNP	Intron Variant
rs9844310	C/T	T	0.019	SNP	Intron Variant	rs7637009	C/G	G	0.39	SNP	Intron Variant
rs112960794	A/G	G	0.02	SNP	Intron Variant	rs72611185	T/A/C	C	0.39	SNP	Intron Variant
rs79398212	T/C	C	0.021	SNP	Intron Variant	rs67261052	C/T	T	0.39	SNP	Intron Variant
rs142969419	G/A/C	C	0.021	SNP	Intron Variant	rs60276076	C/G/T	G	0.39	SNP	Intron Variant
rs112174518	A/G	G	0.021	SNP	Intron Variant	rs56653551	A/G	A	0.4	SNP	Intron Variant
rs114618254	G/T	T	0.022	SNP	Intron Variant	rs61608982	G/A	A	0.4	SNP	Intron Variant
rs113316551	T/C	C	0.023	SNP	Intron Variant	rs11185519	T/A/C	A	0.41	SNP	Intron Variant
rs34352044	C/T	T	0.023	SNP	Synonymous Variant (S30)	rs56875789	C/T	T	0.41	SNP	Intron Variant
rs76496703	T/C	C	0.023	SNP	Intron Variant	rs7638797	A/C	C	0.41	SNP	Intron Variant
rs199580327	T/-	-	0.024	Deletion	Intron Variant	rs9343	A/G	A	0.41	SNP	3’ UTR Variant
rs117874525	C/T	T	0.027	SNP	Intron Variant	rs1476331	T/C	T	0.43	SNP	Intron Variant
rs118117782	T/G	G	0.027	SNP	Intron Variant	rs939885	G/A	A	0.46	SNP	Missense Variant (V202I)
rs73083237	G/A/C	A	0.028	SNP	Intron Variant	rs68119320	C/G/T	T	0.46	SNP	Intron Variant
rs73083239	G/A	A	0.028	SNP	Intron Variant	rs4916441	A/C/T	T	0.47	SNP	Intron Variant
rs181512712	C/T	T	0.029	SNP	5’ UTR Variant	rs58021463	A/G	G	0.47	SNP	Intron Variant
rs116216921	G/A/C/T	C	0.031	SNP	Intron Variant	rs4916531	A/G	A	0.48	SNP	Intron Variant
rs149309867	G/A	A	0.035	SNP	Intron Variant	rs1476332	C/G/T	C	0.48	SNP	Intron Variant
rs113705710	C/G	G	0.036	SNP	Intron Variant	rs4916440	T/A/C/G	T	0.48	SNP	Intron Variant
rs73083244	A/G	G	0.037	SNP	Intron Variant	rs6807596	G/A	G	0.48	SNP	Intron Variant
rs112216889	G/A/C	C	0.039	SNP	Intron Variant	rs4916519	C/G/T	C	0.49	SNP	Intron Variant
rs116434121	C/T	T	0.041	SNP	Intron Variant	rs12491227	G/A	G	0.49	SNP	Intron Variant
rs73212134	G/A	A	0.041	SNP	Intron Variant	rs17852687	T/C	C	0.5	SNP	Synonymous Variant (L225)

Data were obtained from Ensembl version 92.38. MAF, minor allele frequency; SNP, single-nucleotide polymorphism.

Table 8.

Common Variants in SLC51B.

Variant ID	Alleles	MAF Allele	MAF	Class	Type	Variant ID	Alleles	MAF Allele	MAF	Class	Type
rs151094607	G/A	A	0.012	SNP	Intron Variant	rs2946676	G/A/C	G	0.086	SNP	Intron Variant
rs542776151	T/G	G	0.014	SNP	Intron Variant	rs2414870	C/T	C	0.086	SNP	Intron Variant
rs2414871	G/A	A	0.023	SNP	Intron Variant	rs2919349	G/A	G	0.086	SNP	Intron Variant
rs2919347	T/C	C	0.023	SNP	Intron Variant	rs56390972	C/T	C	0.103	SNP	Intron Variant
rs143801232	AGG/-	-	0.024	Deletion	Intron Variant	rs10851744	C/G/T	C	0.114	SNP	Intron Variant
rs2919352	A/G	G	0.024	SNP	3’ UTR Variant	rs537053592	A/C/T	C	0.122	SNP	Missense Variant (D10A)
rs2946674	C/G/T	T	0.025	SNP	Intron Variant	rs8026292	A/G	G	0.35	SNP	Intron Variant
rs142907492	C/T	T	0.025	SNP	Intron Variant	rs6494510	T/C	C	0.352	SNP	Intron Variant
rs74981303	C/A/T	T	0.025	SNP	Intron Variant	rs4238399	C/T	T	0.412	SNP	Intron Variant
rs2946675	T/C	C	0.028	SNP	Intron Variant	rs12906276	G/T	T	0.423	SNP	Intron Variant
rs2919351	T/C/G	C	0.028	SNP	Intron Variant	rs28688080	T/A/C	C	0.457	SNP	Intron Variant
rs2919348	A/G	G	0.031	SNP	Intron Variant	rs11071825	T/A/C	T	0.471	SNP	Intron Variant
rs1670	T/C	C	0.068	SNP	3’ UTR Variant

Data were obtained from Ensembl version 92.38. MAF, minor allele frequency; SNP, single-nucleotide polymorphism.

A recent description of two brothers with a homozygous mutation in SLC51B leading to a frameshift at codon 27 (in the extracellular N-terminus of OSTβ) and a premature stop at codon 50 (in OSTβ’s predicted TMD) constitutes the first two clinical cases of a genetic defect in one of the OSTα/β subunits (Sultan, et al., 2018). These brothers were diagnosed with chronic diarrhea, features of cholestatic liver disease (e.g., elevated serum gamma-glutamyltransferase activity) and severe fat-soluble vitamin deficiency. Since previous studies have shown the importance of OSTβ’s TMD for proper interaction with OSTα, and the importance of several amino acids in OSTβ’s intracellular C-terminus for proper orientation in the plasma membrane (Christian & Hinkle, 2017; Christian, et al., 2012), it is not surprising that this frameshift mutation (p.F27fs) would lead to defective OSTα/β transport. In vitro studies with OSTα/β-overexpressing COS cells containing this mutation showed a lack of protein expression of both OSTα and OSTβ subunits, and a strong reduction in TCA uptake compared to OSTα/β-overexpressing cells without this mutation. The observation that these first two clinical cases of OSTα/β dysfunction have a homozygous deletion in a SLC51B codon coupled with the fact that both parents reported no gastrointestinal symptoms, and had normal serum liver chemistries, suggests that only one functional copy of SLC51B may be needed for normal health. Another case report of a male Pakistani toddler (2.5 years old) with cholestasis, elevated liver enzymes and congenital diarrhea, revealed that this patient had a homozygous mutation in SLC51A leading to a premature termination at codon 186 (Gao, et al., 2019). Based on the finding that his unaffected parents and at least one of his siblings were heterozygous for this mutation, it appears that only one functional SLC51A allele is required for normal OSTα/β function.

7. Mechanisms of OSTα/β induction

Transporter expression is tightly regulated at the transcriptional level by nuclear receptors (NRs), including the aryl hydrocarbon receptor, constitutive androstane receptor (CAR), nuclear factor erythroid-2-related factor 2, pregnane X receptor (PXR), peroxisome proliferator-activated receptors, and hepatocyte nuclear factors among others (Ferslew, Köck, & Brouwer, 2014) (Table 9). Transcriptional regulation of OSTα/β has been described previously, including the involvement of various NRs and information on promoter binding sites (Ballatori, et al., 2013; Ballatori, et al., 2009). In the present review, the most recent reports on this subtopic are highlighted. The bile acid-activated FXR is a primary NR that is responsible for the induction of OSTα/β (Soroka, Ballatori, et al., 2010). FXR is involved in the regulation of several other transporters playing important roles in bile acid disposition such as NTCP, OATP1B3, BSEP, and MRP2 (Ferslew, et al., 2014); thus, in this manner bile acids are involved in the regulation of their own levels. Activation of hepatic FXR leads to decreased protein levels of bile acid uptake transporters while efflux transporters, including OSTα/β, are induced. In addition to the regulation of bile acid transporters, hepatic FXR regulates bile acid biosynthesis by inducing expression of small heterodimer partner (SHP), a repressor of gene transcription, and indirectly downregulating the key enzyme cytochrome P450 (CYP) 7A1 that metabolizes cholesterol into the bile acid precursor 7α-hydroxycholesterol (Goodwin, et al., 2000). Furthermore, bile acids in the gastrointestinal tract can activate intestinal FXR leading to induction of the hormone fibroblast growth factor (FGF) 19 in humans (or FGF15 in mice). Secreted FGF15/19 activates the receptor tyrosine kinase FGF receptor 4 present on hepatocytes, resulting in subsequent hepatic CYP7A1 downregulation that appears to be SHP-independent; in addition, cholestatic bile acids in the human liver may activate hepatic FXR and lead to FGF19 and FGF receptor 4 signaling in an autocrine or paracrine fashion (Chiang, 2009). In vitro and mouse studies suggest that OSTα/β is strictly dependent on FXR (Boyer, et al., 2006; Liu, et al., 2014; Soroka, et al., 2008; Zollner, et al., 2006).

Table 9.

Comparison of OSTα/β to Other Transporters Sharing the Same Substrate(s).

Protein	Gene(s)	Cellular Localizationa	Transport Direction	Driving Force	Transcriptional Regulation	References
OSTα/β	SLC51A, SLC51B	Basolateral (intestine, liver, kidney)	Uptake and Efflux	Substrate gradient	FXR, CAR, PXR, HNF1α, HNF4α, LXRα, RARα, RXRs	(Ballatori, et al., 2005; Okuwaki, et al., 2007; Petrov, et al., 2020; Schaffner, et al., 2015; Seward, et al., 2003; Teng & Piquette-Miller, 2007; Wang, et al., 2001; Xu, et al., 2014)
NTCP	SLC10A1	Basolateral (liver)	Uptake	Na+-dependent (symporter)	FXR, PXR, HNF4α, RXR	(Denson, et al., 2001; Ho, et al., 2004)
ASBT	SLC10A2	Apical (intestine, kidney)	Uptake	Na+-dependent (symporter)	FXR	(Craddock, et al., 1998; Hallen, et al., 2002; Kramer, et al., 1999; Love, et al., 2001)
OCT1	SLC22A1	Basolateral (intestine, kidney, liver)	Uptake and Efflux	Substrate gradient, membrane potential-influenced	PXR, PPAR, HNF1α, HNF4α	(O’Brien, et al., 2013; Saborowski, Kullak-Ublick, & Eloranta, 2006)
OCT2	SLC22A2	Basolateral (kidney)	Uptake and Efflux	Substrate gradient	HNF1, HNF4α	(Popowski, et al., 2005)
OAT1	SLC22A6	Basolateral (liver)	Uptake	Anion/Dicarboxylate exchanger	LXR, HNF1α/β	(Jin, Kikuchi, Saji, Kusuhara, & Sugiyama, 2012; Kittayaruksakul, Soodvilai, Asavapanumas, Muanprasat, & Chatsudthipong, 2012)
OAT2	SLC22A7	Basolateral (kidney, liver), apical (kidney)	Uptake and Efflux	Anion/Dicarboxylate exchanger	FXR, RARα, RXRα, HNF4α	(Fork, et al., 2011; Kobayashi, et al., 2005; Le Vee, Jouan, Stieger, & Fardel, 2013; Shen, et al., 2015)
OAT3	SLC22A8	Basolateral (kidney)	Uptake	Anion/Dicarboxylate exchanger	HNF1α/β	(Burckhardt & Burckhardt, 2011; Jin, et al., 2012; Kikuchi, et al., 2006)
OAT4	SLC22A11	Apical (kidney)	Uptake and Efflux	Anion/Dicarboxylate exchanger	HNF1α/β	(Burckhardt, 2012; Jin, et al., 2012; Zhou, et al., 2004)
OAT7	SLC22A9	Basolateral (liver)	Uptake	Fatty acid exchanger	HNF1α, HNF4α	(Klein, et al., 2010; Emami Riedmaier, et al., 2016)
MATE1	SLC47A1	Apical (kidney, liver)	Efflux (Uptake at ≥pH 7.4)	pH-dependent, H+ antiporter	HNF4α	(Lu, et al., 2010; Tanihara, et al., 2007; Tsuda, et al., 2007)
MATE2-K	SLC47A2	Apical (kidney)	Efflux	pH-dependent, H+ antiporter	NA	(Tanihara, et al., 2007; Terada & Inui, 2008)
OATP1A2	SLCO1A2	Apical (intestine, liver, kidney), Basolateral (liver)	Uptake (and Efflux)	Unclear	SXR, PXR, VDR	(Eloranta, Hiller, Juttner, & Kullak-Ublick, 2012; Lee, et al., 2005; Miki, et al., 2006)
OATP1B1	SLCO1B1	Basolateral (liver)	Uptake	Substrate gradient, pH-influenced	FXR, PXR, LXRα, HNF1α, HNF3β, HNF4α	(Gui, et al., 2008; Kamiyama, et al., 2007; Ohtsuka, et al., 2006; Rodrigues, et al., 2019)
OATP1B3	SLCO1B3	Basolateral (liver)	Uptake (and Efflux)	Substrate gradient, pH-influenced	FXR, HNF1α, HNF3β, HNF4α	(Gui, et al., 2008; Kamiyama, et al., 2007; Ohtsuka, et al., 2006; Rodrigues, et al., 2019)
OATP2B1	SLCO2B1	Apical (intestine), Basolateral (liver, intestine)	Uptake	Substrate gradient, pH-influenced	FXR, PXR, HNF1α, HNF3β, HNF4α	(Hagenbuch & Stieger, 2013; Ohtsuka, et al., 2006; Rodrigues, et al., 2019; Yamada, et al., 2007)
OATP4C1	SLCO4C1	Basolateral (kidney)	Uptake	Substrate gradient	AhR	(Mikkaichi, et al., 2004; Suzuki, et al., 2011)
P-gp	ABCB1	Apical (intestine, kidney, liver)	Efflux	ATP-dependent	CAR, PXR, HNF1α, VDR	(Callaghan, Crowley, Potter, & Kerr, 2008; Chan, Hoque, Cummins, & Bendayan, 2011; Chow, Durk, Cummins, & Pang, 2011; Kimura, Kioka, Kato, Matsuo, & Ueda, 2007; Rodrigues, et al., 2019)
MDR3	ABCB4	Apical (liver)	Efflux	ATP-dependent	FXR, LXR, PPAR	(Smith, et al., 2000)
BSEP	ABCB11	Apical (liver)	Efflux	ATP-dependent	FXR, RXR	(Hayashi, et al., 2005; Kis, et al., 2009)
MRP1	ABCC1	Basolateral (intestine, kidney)	Efflux	ATP-dependent, GSH-influenced	PXR, CAR	(Bakos & Homolya, 2007)
MRP2	ABCC2	Apical (intestine, kidney, liver)	Efflux	ATP-dependent	FXR, CAR, PXR, HNF4α, VDR, NRF2	(Arana, Tocchetti, Rigalli, Mottino, & Villanueva, 2016; Rodrigues, et al., 2019)
MRP3	ABCC3	Basolateral (liver, intestine, kidney)	Efflux	ATP-dependent	PXR, CAR, VDR	(Rodrigues, et al., 2019; Zeng, et al., 2000)
MRP4	ABCC4	Apical (intestine, kidney), Basolateral (liver)	Efflux	ATP-dependent, GSH-influenced	FXR, CAR, PPAR	(Lai & Tan, 2002)
BCRP	ABCG2	Apical (intestine, liver)	Efflux	ATP-dependent	CAR, PXR, PPAR	(Rodrigues, et al., 2019; Lin, et al., 2017)

^aTissue localization: intestine, kidney and/or liver

ABC, ATP-binding cassette transporter; AhR, aryl hydrocarbon receptor; ATP, adenosine triphosphate; CAR, constitutive androstane receptor; FXR, farnesoid X receptor; GSH, reduced glutathione; HNF, hepatocyte nuclear factor; K_m, substrate concentration at one-half of the maximum velocity; LXR, liver X receptor; NA, data not available in the literature; NRF2, nuclear factor erythroid-2-related factor 2; PPAR, peroxisome proliferator-activated receptor; PXR; pregnane X receptor; RAR, retinoid acid receptor; RXR, retinoid X receptor; SLC, solute carrier transporter; SXR, steroid and xenobiotic receptor; VDR, vitamin D receptor

Treatment of primary human hepatocytes with the potent FXR agonist CDCA, a prevalent unconjugated primary bile acid produced in the liver, and obeticholic acid (OCA), a semi-synthetic bile acid used to treat primary biliary cholangitis (PBC), under review for the treatment of NASH, and in development for the treatment of other metabolic disorders, induced SLC51A/OSTα and SLC51B/OSTβ at the mRNA and protein level (Guo, et al., 2018; Jackson, Freeman, & Brouwer, 2016; Liu, et al., 2014). Overall, the secondary bile acid DCA, produced by gut bacteria via dehydroxylation of primary bile acids, and the primary bile acid cholic acid, showed an intermediate ability to regulate FXR target genes (Liu, et al., 2014). While CDCA-mediated induction of SLC51A and SLC51B mRNA was observed in human ileal biopsies (Landrier, et al., 2006), OCA-mediated induction of the SLC51A and SLC51B genes was reported in human liver slices (Ijssennagger, et al., 2016). CDCA also induced SLC51A and SLC51B mRNA in HepG2 and HuH-7 cell lines (Landrier, et al., 2006; Schaffner, et al., 2015; Xu, et al., 2014).

Apart from bile acids, additional compounds, NRs or conditions have been tested as potential mediators of OSTα/β induction. For example, in pigs fed with corn-soybean meal diets containing 5% apple pectin compared to pigs fed cornstarch as the control, ileal, cecal and colonic SLC51A and SLC51B mRNA expression trended higher (Fang, et al., 2018). In primary human hepatocytes cultured under hypoxic conditions, a characteristic of cholestasis, OSTα/β expression was increased. Under low oxygen conditions, hypoxia-inducible factor-1α binds to hypoxia responsive elements in the gene promoters of both OSTα/β subunits (Schaffner, et al., 2015). In a humanized liver mouse model, bile acids were elevated compared to the non-humanized control model; furthermore, hepatic human SLC51A and SLC51B mRNA was increased in the humanized liver mouse model compared to human liver (Chow, et al., 2017). Intestinal murine SLC51A mRNA was elevated in the humanized liver mouse model compared to the non-humanized control model. Some of these alterations in expression may be due to miscommunication between human hepatocytes and the murine intestinal cells. PXR may be involved in negatively regulating SLC51A and SLC51B, among other FXR-regulated genes (Ballatori, et al., 2009). For instance, in cholic acid-fed wild-type mice, the PXR activator 5-pregnen-3β-ol-20-one-16α-carbonitrile downregulated Slc51a and Slc51b gene expression (Teng & Piquette-Miller, 2007). Furthermore, gene expression of Slc51a and Slc51b was elevated in PXR^−/− mice compared to wild-type control mice. Antibiotics can result in bile acid elevations by enhancing bile acid synthesis; however, in mice treated with ampicillin or bacitracin/neomycin/streptomycin, ileal OSTα protein was not altered even though ASBT protein levels were elevated compared to control mice (Miyata, Hayashi, Yamakawa, Yamazoe, & Yoshinari, 2015). Experiments with Caco-2 cells demonstrated that liver X receptor-α plays a role in the transcriptional regulation of mouse OSTα/β (Okuwaki, et al., 2007). In the LEE-1 cell line derived from a little skate embryo, the eicosanoid precursor arachidonic acid induced OSTβ, while a lipid mixture containing arachidonic acid induced OSTα (Hwang, Parton, Czechanski, Ballatori, & Barnes, 2008). Studies in HuH-7 and HepG2 cells have suggested that OSTβ is regulated by retinoic acid receptor-α and CAR heteromerized with retinoid X receptor-α (Xu, et al., 2014). Furthermore, OSTα/β can be regulated by ligands of the vitamin D receptor and the glucocorticoid receptor in a species-dependent manner (Khan, Chow, Porte, Pang, & Groothuis, 2009).

Aside from extensive research on transcriptional regulation, the post-translational regulation of OSTα/β has been evaluated. For instance, it is evident that OSTα undergoes N-glycosylation-dependent maturation (see Section 5), although identification of other post-translational modifications warrants further investigation. Finally, whether microRNA-mediated post-transcriptional control of gene expression plays a role in regulating OSTα and OSTβ levels is unknown, but could be involved in organ-specific differences in OSTα/β protein expression.

8. Altered hepatic OSTα/β expression in cholestatic liver conditions, NASH and other disorders

Studies in humans and non-clinical models have demonstrated that cholestatic conditions lead to adaptive responses in bile acid synthesis and transporters (Boyer, et al., 2006; Schaap, et al., 2009; Trauner, Wagner, Fickert, & Zollner, 2005). For instance, treatment of the HepG2 human hepatoma cell line with CDCA (50 μM, 24 hr) increased OSTα, OSTβ, SHP and BSEP mRNA, and OSTα and OSTβ proteins (Boyer, et al., 2006). Furthermore, hepatic transcripts of FGF19, the bile acid synthesis enzyme CYP7A1, and several transporters including OSTα, OSTβ, MRP3 and OATP1B1 were significantly altered in patients with extrahepatic cholestasis (i.e., obstruction of bile flow in the ducts from the liver to the duodenum) compared to control subjects (Schaap, et al., 2009). In several other human cholestatic disorders, OSTα and/or OSTβ are substantially induced, suggesting an important detoxifying role for hepatic OSTα/β-mediated efflux when bile constituents (e.g., bile acids) accumulate in the liver. While OSTα protein usually has low hepatic expression and OSTβ protein tends to be nearly undetectable in healthy livers, hepatic OSTα and particularly OSTβ protein were highly expressed in patients with PBC (Boyer, et al., 2006; Malinen, et al., 2018); elevations of SLC51A mRNA and especially SLC51B mRNA have been reported in PBC. In obstructive cholestatic human livers, both SLC51B mRNA and OSTβ protein were found to be significantly elevated compared to control livers, while SLC51A mRNA was unchanged, and OSTα protein was significantly decreased (Chai, et al., 2015). Most recently, SLC51B mRNA and OSTβ protein were found to be significantly increased in livers from patients with NASH, with OSTα protein also trending higher in NASH livers (Malinen, et al., 2018). NASH, the most severe form of non-alcoholic fatty liver disease, is increasing at an alarming rate with the obesity epidemic. Unlike obstructive/extrahepatic cholestasis and PBC, NASH is not typically characterized by clinical symptoms of cholestasis; nevertheless, some patients with NASH exhibit cholestasis (Sorrentino, et al., 2005). The reduction of bile flow in cholestasis can lead to elevated serum bile acid concentrations, potentially attributable, in part, to overexpression of hepatic OSTα/β and other basolateral efflux transporters serving to protect hepatocytes from toxicity. In addition to overexpression of hepatic OSTα/β in NASH, serum bile acid concentrations also are elevated in NASH (Ferslew, et al., 2015). The increase in hepatic OSTα/β expression in individuals with cholestatic disorders could mean that drug interactions with OSTα/β could affect hepatic OSTα/β-mediated transport, which may increase susceptibility to toxicity.

While SLC51A mRNA was largely unchanged in the livers from common bile duct-ligated mice or rats, SLC51B mRNA increased compared to sham-treated rodents (Boyer, et al., 2006). However, despite the lack of alterations in SLC51A mRNA, hepatic OSTα protein was induced in the common bile duct-ligated rodents. These studies suggest that biliary obstruction induces hepatic SLC51B/OSTβ in both humans and rodents, but that obstructive cholestatic human livers and livers from common bile duct-ligated rodents may regulate SLC51A/OSTα differently, likely due to important species differences. Apart from hepatic conditions, SLC51A and SLC51B mRNA was elevated in the livers from rats with chronic renal failure, a condition associated with increased serum bile acid concentrations and changes in bile acid homeostasis (Gai, et al., 2014).

9. Potential role of OSTα/β in drug development

Although OSTα/β has not been a direct target for new therapeutics, the FXR agonist OCA, which is used for the treatment of PBC, significantly induced OSTα/β and increased the intrinsic basolateral efflux clearance of TCA in sandwich-cultured human hepatocytes (Guo, et al., 2018). However, it is unclear to what extent the enhanced efflux mediated by OSTα/β plays a role in OCA-based PBC treatment. It has been hypothesized that therapeutically inducing OSTα/β expression in cholestasis can protect the liver from hepatotoxicity (Zollner, et al., 2006), and this still appears to be a plausible hepatoprotective mechanism against bile acid-mediated liver injury. Apart from inducing OSTα/β expression, in vitro uptake studies have shown that human OSTα/β transport activity can be stimulated by various compounds (Malinen, et al., 2018); whether stimulation of OSTα/β transport also occurs in vivo remains to be elucidated.

Studies in mice have indicated that inhibition of OSTα/β may have therapeutic purposes [e.g., by activating intestinal FXR and enhancing bile acid signaling (van de Wiel, et al., 2018)], but it is important to emphasize that rodents have a distinct bile acid profile and different bile acid regulation compared to humans (Chiang, 2009; Garcia-Canaveras, Donato, Castell, & Lahoz, 2012; Hofmann, 2004, 2009). Therefore, rodent models may not accurately predict human cholestatic conditions.

Aside from potentially targeting OSTα/β for pharmacotherapeutic purposes, OSTα/β may play an important role in predicting xenobiotic-induced hepatotoxicity in drug development. DILI is a major public health concern for patients and healthcare providers, a primary safety issue in drug development, and a frequent reason for drug withdrawal from the market (Mosedale & Watkins, 2017; Onakpoya, Heneghan, & Aronson, 2016; Perez & Briz, 2009; Temple & Himmel, 2002). Among the various mechanisms underlying DILI, inhibition of bile acid transport and alterations in bile acid homeostasis account for almost half of the DILI cases (Weaver, et al., 2019). Since OSTα/β is important in bile acid homeostasis, it is plausible that proper functioning of this transporter protects patients from DILI. As discussed, the transport function of OSTα/β is inhibited by some drugs associated with DILI (Malinen, Kauttonen, et al., 2019), but interactions with OSTα/β are not evaluated routinely due to lack of clinical evidence. The potential role of OSTα/β in bile acid-mediated DILI remains to be elucidated.

10. Conclusion/outlook

The observed expression of OSTα/β in the intestine, liver and kidneys, the major organs determining the absorption, metabolism and elimination of drugs, the upregulation of OSTα/β in hepatic diseases, the interaction of OSTα/β with drugs, and clinical case reports of diarrhea and cholestasis in patients with homozygous genetic defects in OSTα/β have increased awareness among clinicians and scientists about this understudied transporter. Prior reviews on OSTα/β focused on the role of this protein in bile acid and steroid metabolite transport. The present review provides current knowledge on the pharmacological role of OSTα/β, as well as the expression, structure and function of OSTα/β.

OSTα/β is expressed on the basolateral membrane of various types of epithelial cells and is composed of two subunits, OSTα and OSTβ. The need for the expression of two separate proteins for transporter function on the basolateral membrane separates OSTα/β from the well-known drug transporters. Interestingly, in intestinal cells where the physiological function of OSTα/β is evident under basal conditions, or in hepatocytes under cholestatic conditions, the expression of OSTβ protein is higher than OSTα protein, while the opposite tends to be true in the tissues where the role of OSTα/β is still unclear. Although OSTα/β-mediated transport appears to be largely dependent on substrate concentration, a role for pH and some ions has been noted.

Studies during the last two decades have shown that OSTα/β is an important transporter for the disposition of endogenous compounds, particularly relating to the enterohepatic circulation of bile acids, but the involvement of OSTα/β in drug-drug/drug-bile acid interactions has not been thoroughly characterized. It is now evident that OSTα/β can transport probe substrates that are shared by other clinically relevant drug transporters (Table 4). OSTα/β appears to be a low affinity/high capacity transporter for several substrates (see Section 3.1). This is a complicating factor when trying to determine the extent of involvement of OSTα/β in the transport of, or interaction with, a specific compound of interest. To compare substrate affinities of OSTα/β to other transporters of interest, it is ideal to evaluate these transporters under identical experimental conditions, which is not always possible.

The inhibition of OSTα/β by drugs associated with hepatotoxicity suggests a possible role of OSTα/β in DILI; however, the number of compounds interrupting OSTα/β transport is relatively low even though more than a thousand compounds have been evaluated as inhibitors. This low identification rate may correspond to actual physiological phenomena or it may reflect the lack of sensitivity of in vitro systems used thus far to study OSTα/β. The presence of OSTα/β is often ignored in models predicting disposition or toxicity of test compounds, potentially generating a misleading interpretation of the pharmacokinetics, pharmacodynamics and safety of compounds. Future investigations designed to determine the three-dimensional structure of OSTα/β and to identify novel OSTα/β substrates and inhibitors will aid in the development and use of structure-activity relationship models, and advance knowledge regarding the role of OSTα/β in drug disposition and drug interactions.

List of Abbreviations*

ASBT/SLC10A2

apical sodium-dependent bile acid transporter

ATP

adenosine triphosphate

ADME-Tox

absorption, distribution, metabolism, excretion and toxicity

BCRP/ABCG2

breast cancer resistance protein

BSEP/ABCB11

bile salt export pump

C-

carboxy-

CAR

constitutive androstane receptor

CDCA

chenodeoxycholate

CYP

cytochrome P450

DCA

deoxycholate

DHEAS

dehydroepiandrosterone sulfate

DILI

drug-induced liver injury

ES

estrone sulfate

FGF

fibroblast growth factor

FXR

farnesoid X receptor

HEK

human embryonic kidney

MAF

minor allele frequency

MCT/SLC16A

monocarboxylate transporters

MDCK

Madin-Darby canine kidney

MRP/ABCC

multidrug resistance-associated protein

N-

amino-

NASH

nonalcoholic steatohepatitis

NTCP/SLC10A1

Na⁺-taurocholate cotransporting polypeptide

NR

nuclear receptor

OATP/SLCO

organic anion transporting polypeptide

OCA

obeticholic acid

OSTα/SLC51A

organic solute transporter alpha

OSTβ/SLC51B

organic solute transporter beta

P-gp/MDR1/ABCB1

P-glycoprotein

PBC

primary biliary cholangitis

PolyPhen-2

Polymorphism Phenotyping v2

PXR

pregnane X receptor

REVEL

Rare Exome Variant Ensemble Learner

SIFT

Sorting Intolerant From Tolerant

SLC

solute carrier

TCA

taurocholate

TMD

transmembrane domain

*In this review, all letters of mRNA and protein names for all species are in uppercase. Gene names are italicized, with only the first letter in uppercase for mouse gene names.