For antiretroviral therapy (ART) to be successful, it is essential that drugs reach their pharmacological site of action. Since the target of protease inhibitors (PIs) is within HIV infected cells, the intracellular (IC) concentrations of PIs are crucial in the understanding of therapeutic failure. Previous investigations have demonstrated a differential PI intracellular accumulation in peripheral blood mononuclear cells (PBMCs) (Khoo et al., 2002) and in lymphoblastoid cell lines (Jones et al., 2001). As HIV primarily targets CD4+ lymphocytes, it is important to investigate the IC concentration of PIs in cellular subsets. In addition, lymphocyte subsets differentially express transporters such as P-glycoprotein (P-gp), and consequently may influence IC PI accumulation in different cell subsets.

Human PBMCs were isolated from buffy coats obtained from the local blood bank (n=12) by density gradient centrifugation and separated by magnetic MACS® microbeads into total PBMCs, CD4+, CD4-, CD8+ and CD8- subsets. Cells, (10x10⁶) were incubated (2h, 37°C, 5% CO₂) with PIs [10µM, ³H-ritonavir (RTV), ³H-indinavir (IDV), ³H-nelfinavir (NFV), ¹⁴C-lopinavir (LPV) and ¹⁴C-saqiunavir (SQV); specific activity 0.1µCi] in a 10ml final incubation volume. Cells were centrifuged and the IC (following methanol extraction at 4°C) and extracellular (EC) concentration of drug determined by scintillation counting. P-gp expression on lymphocyte subsets was determined by dual colour flow cytometry. The partition coefficient (Log₁₀ P value) of the PIs was assessed by n-octanol/water shake flask method and HPLC-MS/MS. Data expressed as ratio of IC/EC.

Differential accumulation of the PIs was observed in each cell subset with the hierarchy; NFV>SQV>RTV=LPV>IDV, and values of 15.8 ± 7.5, 7.7 ± 4.6, 4.6 ± 2.1, 3.6 ± 1.9 and 0.7 ± 0.1 in total lymphocytes, 22.1 ± 9.0, 11.6 ± 4.9, 7.2 ± 2.4, 7.9 ± 2.9 and 0.56 ± 0.2 in CD4+ cells and 13.8 ± 7.7, 7.8 ± 3.4, 4.9 ± 2.4, 5.1 ± 0.9 and 0.6 ± 0.2 in CD8+ cells. In addition a differential subset accumulation of each PI was observed, generally with PI accumulation greater in CD4+ cells than CD8+ cells. No relationship was observed between PI accumulation and lymphocyte expression of P-gp. The Log10 partition coefficient of the PIs demonstrated the following rank order NFV>SQV>APV= LPV> RTV>IDV with values of 2.9 ± 0.03, 1.9 ± 0.06, 1.7 ± 0.05, 1.7 ± 0.05, 1.2 ± 0.03 and 0.9 ± 0.01 respectively.

This study demonstrates a differential PI accumulation in lymphocyte subsets with the hierarchy reflecting the rank order of the lipophilicity and previously reported PI accumulation data. Interestingly, a differential subset accumulation of drug was observed with CD4+ cells containing greater PI accumulation than CD8+ cells. This suggests that other factors such as active transport are important for accumulation in addition to passive diffusion. However, no relationship between PI accumulation and P-gp expression was evident, which is in contrast to cell line studies that overexpress efflux transporters. It may be that combinations of influx and efflux transporters are important for PI accumulation.

Khoo, S. H. et al (2002) Antimicrob. Agents Chemother. 46, 3228-3235.
Jones, K. et al (2001) AIDS. 15, 675-681.