Loss of organic anion transporting polypeptide 1B3 function causes marked delay in indocyanine green clearance without any clinical symptoms

Abbreviations

ICG

indocyanine green

NTCP

sodium‐taurocholate cotransporting polypeptide

OATP

organic anion transporting polypeptide

P1‐P4

patients 1‐4

The indocyanine green (ICG) retention test is widely used for preoperative evaluation of liver function. Individuals who have markedly poor ICG clearance without severe liver diseases are defined as patients with a “constitutional ICG excretory defect.”1 However, the underlying molecular mechanisms remain unknown. We hypothesized that the ICG clearance defect involved impaired function of hepatic uptake transporters such as organic anion transporting polypeptide 1B1 (OATP1B1, gene symbol SLCO1B1), OATP1B3 (SLCO1B3), and sodium‐taurocholate cotransporting polypeptide (NTCP, SLC10A1). Homozygous inactivation of both SLCO1B1 and SLCO1B3 causes Rotor syndrome, for which three disease‐causing haplotypes linked to SLCO1B1 and SLCO1B3 are documented.2 We also reported a Japanese population‐specific haplotype: a homozygous c.1738C>T (p.R580X) nonsense mutation in SLCO1B1, and a homozygous insertion of a 6.5‐kbp long interspersed element (LINE‐1, L1) retrotransposon in intron 5 of SLOC1B3 that induced exon skipping and a premature stop codon that generated truncated proteins.3 A recent case report showed that homozygous c.755G>A (p.R252H) missense mutation in SLC10A1 led to the functional loss of NTCP resulting in hypercholanemia.4 Here, by genotyping these transporters, we report cases of an ICG clearance defect due to a deficiency of OATP1B3.

Cases

Four patients (P1‐P4) underwent an ICG retention test—popular in Japan—for assessing liver function. All four patients exhibited extremely high ICG retention at 15 minutes, ranging from 70.8% to 90.7% (Table 1A). ICG elimination rate constants for P3 and P4 were markedly low. Background liver diseases included primary biliary cholangitis (Ludwig stage I), fatty liver, and hepatocellular carcinoma. P4 was taking metformin, telmisartan, rosuvastatin, and alogliptin, while the other patients were not taking any drugs. Except for slight elevations in serum alanine aminotransferase and alkaline phosphatase levels, laboratory data were normal. Liver biopsy denied the presence of liver cirrhosis or cholestasis in all patients.

Table 1.

Patient Characteristics and Genotyping of SLCO1B1, SLCO1B3, and SLC10A1

Subject	Age (years)	Gender	Liver Disease	Platelets (×104/μL)	Albumin (g/dL)	ALT (IU/L)	ALP (IU/L)	T. Bil (mg/dL)	D. Bil (mg/dL)	Total bile acid (μmol/L)	PT‐INR	ICG‐ R15 (%)	ICG‐k
A. Patient Characteristics a
P1	44	F	PBC (stage I)	12.6	4.3	63	417	0.7	0.2	3.3	0.88	90.7	NA
P2	58	F	PBC (stage I)	23.6	4.1	32	443	0.8	0.3	2.5	0.95	79.3	NA
P3	62	M	Fatty liver	23.4	4.4	38	313	0.8	0.3	1.0	0.97	86	0.024
P4	53	F	HCC	27.1	4.6	27	201	1.4	0.1	5.2	1.26	70.8	0.023

Subject	SLCO1B1							SLCO1B3		SLC10A1
	c.388A>G	c.571T>C	c.757C>T	c.1738C>T	chr12: 21239315T>C	chr12: 21239628T>G	chr12: 21239652A>G	c.1747+1	L1	c.225G>A	c.263T>C	c.755G>A
	p.N130D	p.L191L	p.R253X	p.R580X	3′UTR	3′UTR	3′UTR	G>A	insertion	p.T75T	p.I88T	p.R252H
	rs2306283	rs4149057	rs183501729	rs71581941		rs4149087	rs4149088			rs4646285	rs148467625	rs147226818
B. Genotyping of SLCO1B1, SLCO1B3, and SLC10A1 b
P1	A/G	T/C	C/C	C/C	T/C	T/T	A/A	G/G	L1/L1	G/A	T/C	G/G
P2	G/G	T/T	C/C	C/C	T/T	T/G	A/G	G/G	L1/L1	G/G	T/T	G/G
P3	G/G	T/T	C/C	C/C	T/T	T/G	A/G	G/G	L1/L1	G/G	T/T	G/G
P4	G/G	T/T	C/C	C/C	T/T	T/G	A/G	G/G	L1/L1	G/G	T/T	G/G

^aNormal reference values of laboratory data: platelets, 13.0‐36.9 × 10⁴/μL; albumin, 3.8‐5.2 g/dL; ALT, 5‐40 IU/L; ALP, 115‐359 IU/L; total bilirubin, 0.3‐1.2 mg/dL; direct bilirubin, <0.4 mg/dL; total bile acid, <10 μmol/L; PT‐INR, 0.85‐1.15; ICG‐R15, <10%; ICG‐k, 0.179‐0.199.

^bc.757C>T and c.1738C>T in SLCO1B1 and c.1747+1G>A and L1 insertion in SLCO1B3 are variants associated with Rotor syndrome, and 755G>A in SLC10A1 is a variant associated with hypercholanemia (shown in bold). The following variants have not been reported to be pathogenic: c.388A>G, c.571T>C, chr12:21239315T>C, chr12:21239628T>G and chr12:21239652A>G in SLCO1B1, and c.225G>A and c.263T>C in SLC10A1.

Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; D. Bil, direct bilirubin; HCC, hepatocellular carcinoma; ICG‐R15, indocyanine green retention rate at 15 minutes; ICG‐k, indocyanine green elimination rate constant; NA, not available; PBC, primary biliary cholangitis; PT‐INR, international normalized ratio of prothrombin time; T. Bil, total bilirubin.

Sequencing of all SLCO1B1 and SLC10A1 exons after obtaining written informed consent did not reveal any pathogenic variants, whereas all patients were homozygous for the L1 insertion in intron 5 of SLOC1B3 (Table 1B). We confirmed the lack of hepatic OATP1B3 expression by immunohistochemistry in three patients (Fig. 1).

Figure 1.

Hepatic OATP1B3 protein expression in patients P1, P2, and P4 and a control. Positive staining for OATP1B3 is observed along the basolateral membrane of hepatocytes in centrilobular areas in control liver tissue (antibody HPA004943; Sigma‐Aldrich, St. Louis, MO). In contrast, OATP1B3 expression is not detected in any of the patients. Scale bar, 200 μm. Inset, high magnification.

Discussion

OATP1B3 and NTCP can transport ICG in vitro.5 An unrevealed deficiency of NTCP in addition to the absence of OATP1B3 might have contributed to the lack of ICG uptake in our patients. However, this seems unlikely because NTCP deficiency–associated hypercholanemia4 was not observed (Table 1A). ICG is reported to accumulate in some hepatocellular carcinoma nodules after intravenous injection. Hepatocellular carcinoma tissues with ICG accumulation had stronger OATP1B3 expression than those without ICG accumulation, suggesting the involvement of this transporter in ICG uptake.

We conclude that OATP1B3 is the major hepatic transporter for ICG uptake and that its deficiency profoundly impairs ICG clearance. A solitary defect in OATP1B3 does not appear to cause conjugated hyperbilirubinemia as reported.2 The allele frequency of the L1 insertion in SLCO1B3 was 0.054 in a Japanese population,3 suggesting that homozygosity occurs in approximately 1 in 400 individuals. Because OATP1B3 is involved in the hepatic uptake of various drugs such as statins, bosentan, olmesartan, erythromycin, and docetaxel, the risk of drug‐induced toxicity is likely increased in these individuals.

Patients with impaired function of a specific transporter, such as those reported here, can provide important insights into understanding the complicated hepatic transport systems.