Determining the substrate specificity of SLCO1B3 for antiretroviral drugs using a X. laevismodel

Wai San Kwan1, Darren Moss1, Ruben Hartkoorn1, Enrique Salcedo-Sora2, Pat Bray2, Saye Khoo1, David Back1, Andrew Owen1

1University of Liverpool, Liverpool, UK, 2Liverpool School of Tropical Medicine, Liverpool, UK

Previously, we have shown that HIV protease inhibitors (PIs) are substrates for SLCO1A2 and SLCO1B1. SLCO1B3 is another major hepatic influx transporter and is known to transport similar xenobiotics (e.g. statins) to SLCO1B1 (1). However, there are also substrates (e.g. digoxin) which are specific to SLCO1B3 (2). Here we have investigated the influence of SLCO1B3 on the cellular accumulation of antiretroviral PIs saquinavir (SQV), lopinavir (LPV), darunavir (DRV) and non-nucleoside reverse transcriptase inhibitors nevirapine (NVP), efavirenz (EFV) and rilpivirine (TMC278) using the X laevis oocyte model.

SLCO1B3 was amplified from Huh7 cDNA (and a Kozak sequence added prior to the 5’ start codon of the gene) and cloned into the pCRII-TOPO vector. The SLCO1B3 insert was verified by sequencing and non-synonymous mutations were corrected using site-directed mutagenesis. The sequence was verified to NM_019844 prior to subcloning into the pBluescriptII-KSM vector (which contains 5’ and 3’ X. laevis β-globin UTR flanking the site of insertion). SLCO1B3-KSM cRNA was generated from the correct plasmids by in vitro transcription.

X. laevis oocytes lobes were extracted from an adult female X. laevis frog and individual oocytes were isolated using the platinum loop extraction method. Oocytes were maintained in modified Barth’s solution overnight. Healthy X. laevis oocytes were selected and injected with SLCO1B3 cRNA (± 50 ng) or water. The optimal uptake of a positive control substrate, [3H]-estrone-3-sulphate ([3H]-E3S) was determined to be 5 days post injection. The substrate specificity of SLCO1B3 for antiretrovirals was then studied in SLCO1B3-KSM injected and water injected oocytes. Following maintenance of the oocytes (18ºC, 5 days), SLCO1B3-KSM and water injected oocytes were incubated (1h, shaking, room temperature) with [3H]-E3S (1μM, 0.33μCi/ml), and the antiretrovirals, [3H]-SQV (1μM, 0.33μCi/ml), [3H]-LPV (1μM, 0.33μCi/ml), [14C]-DRV (3μM, 0.12μCi/ml), [3H]-NVP (1μM, 0.33μCi/ml), [14C]-EFV (1μM, 0.33μCi/ml) and [3H]-rilpivirine (3μM, 0.06μCi/ml). Radioactivity was measured by liquid scintillation spectrometry and statistical analysis was conducted using a paired t-test (StatsDirect 2.6).

SLCO1B3-KSM injected oocytes had a trend towards higher [3H]-E3S accumulation compared to water injected oocytes (0.50 ± 0.20 vs. 0.21 ± 0.01 pmol/oocyte, n=4, p=0.06). A significant increase in accumulation was observed for SQV (2.17 ± 0.61 vs. 1.62 ± 0.45 pmol/oocyte, n=4, p<0.05) with a trend towards higher LPV (2.30 ± 0.66 vs. 1.69 ± 0.78 pmol/oocyte, n=4, p=0.07) and DRV (0.63 vs. 0.39 pmol/oocyte, n=1). A significant increase of rilpivirine accumulation was observed between SLCO1B3-KSM injected oocytes when compared to water injected oocytes (37.0 ± 12.40 vs. 16.2 ± 3.50 pmol/oocyte, n=4, p<0.05). There were no differences in EFV or NVP accumulation between SLCO1B3-KSM and water injected oocytes.

In conclusion, these data show that PIs and rilpivirine are substrates for SLCO1B3. Further functional and pharmacogenetic studies are now required to determine the importance of SLCO1B3 for the clinical efficacy of these drugs in patients.