Objectives: To determine the effect of the protease inhibitors ritonavir, nelfinavir and indinavir on the P-glycoprotein (P-gp)-mediated transport of saquinavir in Caco-2 cell monolayers. To study the modulation of P-gp function in human lymphocytes by saquinavir, ritonavir, nelfinavir and indinavir.
Methods: We examined the effect of the protease inhibitors on P-gp function in human lymphocytes by using Rhodamine123 (Rh123; a fluorescent substrate of P-gp) by flow cytometry. Efflux of Rh123 correlates with P-gp function and inhibition of P-gp results in dye retention. Verapamil, a P-gp modulator and inhibitor of active transport at 4°C was used as a positive control. The transport of [14C]saquinavir (1μM) across Caco-2 cell monolayers was investigated, alone and in the presence of verapamil and ketoconazole (500μM) and the protease inhibitors at 100μM. Caco-2 cells are an in vitro model of the intestinal epithelium that is widely used for the study of P-gp function. The transport of saquinavir was determined in both the apical to basolateral (AP-BL) and basolateral to apical (BL-AP) directions.
Results: Saquinavir and ritonavir (10μM) markedly inhibited Rh123 efflux with an increase in fluorescence intensity similar to that obtained with verapamil. A small but statistically significant increase in fluorescence intensity was observed with nelfinavir; however indinavir did not modulate Rh123 efflux. In Caco-2 cells the apparent permeability coefficient for BL-AP efflux of saquinavir exceeded that for AP-BL efflux by a factor of 26: this is indicative of an active efflux pump. Known P-gp modulators caused a decrease in BL-AP efflux and an increase in AP-BL transport. The protease inhibitors displayed some P-gp modulation with ritonavir having the most potent effect.
Conclusions: We have demonstrated that saquinavir is a substrate for P-gp and that ritonavir, nelfinavir and indinavir modulate P-gp function in both human lymphocytes and Caco-2 cells.
From the Department of Pharmacology and Therapeutics, University of Liverpool, UK.
Sponsorship: Supported by the Medical Research Council, UK.
Correspondence to D. J. Back, Department of Pharmacology and Therapeutics, New Medical Building, University of Liverpool, Ashton Street, Liverpool, L69 3GE, UK.
Received: 27 January 1999; revised: 26 April 1999; accepted: 17 May 1999.
HIV-1 protease inhibitors have made a tremendous impact as part of triple combination therapy for the treatment of HIV infection. However there are a number of problems associated with use of the protease inhibitors.
Saquinavir has a low and variable oral bioavailability (hard gel 4%; soft gel 12% ), which is a consequence of multiple factors including extensive intestinal and hepatic first-pass metabolism by cytochrome P-450 (CYP) 3A4. Ritonavir, indinavir and nelfinavir are also predominately metabolized by CYP isoforms (including CYP3A4) with ritonavir being a potent inhibitor of both CYP3A4 and CYP2D6. It is important to maintain therapeutic levels of antiretroviral drugs as numerous reports have indicated the rapid development of HIV-1 strains that are resistant to them.
The overlapping substrate specificity between CYP3A and an active efflux pump, P-glycoprotein (P-gp) suggests that P-gp may contribute to the variation in bioavailability of drugs previously attributed to CYP3A.
P-gp is a 170kDa transmembrane glycoprotein which functions as an ATP-dependent drug efflux pump. It is expressed in several normal tissues including the apical membrane of intestinal epithelial cells. The human colon adenocarcinoma cell line (Caco-2) is a well-established in vitro model of the intestinal epithelium that expresses P-gp on its apical surface. This widely used model has been used to characterize a number of P-gp substrates and inhibitors including saquinavir and ritonavir.
Because many of the drugs used in the treatment of HIV infection and the diseases associated with it are CYP3A4 substrates or inhibitors, adverse drug-drug interactions may occur when these drugs are co-administered. This may also be true for drugs that are substrates for P-gp. It is therefore necessary to determine the interaction of the protease inhibitors with P-gp when given in combination. We studied the modulating effects of the protease inhibitors ritonavir, nelfinavir and indinavir on saquinavir, P-gp-mediated transport in Caco-2 cells.
An additional problem associated with the protease inhibitors is the existence of sanctuary sites such as resting latently infected CD4 cells. These sanctuary sites prevent complete eradication of the virus. It has been suggested that the affinity of antiretroviral drugs for P-gp may be responsible for the existence of these sanctuary sites.
Because P-gp is expressed in human lymphocytes and monocytes, which are major targets of HIV-1, efflux of the protease inhibitors from these cells may lead to the development of resistant strains of the virus. Flow cytometric analysis of Rhodamine123 (Rh123) efflux from blood cells can be used to investigate P-gp function. We examined the effect of the protease inhibitors on Rh123 efflux in human peripheral blood mononuclear cells (PBMC) obtained from healthy volunteers.
Dye efflux studies in PBMC
PBMC (3×105/ml) were isolated from blood samples obtained from healthy volunteers and loaded with Rh123 (1.5μg/ml; 25 min; 4°C) in RPMI 1640 media. Cells were washed twice with ice-cold media andincubated at 37°C or 4°C for 3 h in 10ml of dye-free media to allow dye efflux. Parallel experiments were performed at 37°C in the presence of verapamil (30μM), ritonavir, indinavir and nelfinavir (1 and 10μM).
Flow cytometric analysis of Rh123-labelled cells
Analysis was conducted on a FACScan flow cytometer (Becton Dickinson, Mount view, California, USA). For each sample 10000 events were collected. Rh123 fluorescence was followed in channel 1 (FL1). Lymphocytes were electronically gated and the amount of Rh123 fluorescence plotted as a histogram of FL1 staining. Data acquisition was performed using the computer program WINMDI version 2.6 to determine median FL1 fluorescence values.
Caco-2 cell culture
Caco-2 cells were cultured in Dulbecco‚s modified Eagle‚s medium and maintained at 37°C in a 10% CO2 incubator. Following confluence, the cells were split using 0.25% trypsin-EDTA, plated onto polycarbonate membrane transwells (0.4μm, 4.7cm2 growth area, 24mm insert diameter) at a density of 0.5×106 cells/cm2. The medium was changed every 2-3 days and the cells were used for transport experiments between 14 and 21 days post-seeding. The transepithelial electrical resistance (TER) across the cell monolayers was monitored using a Millicell-ERS (electrical resistance system) to assess cell monolayer integrity.
Transport experiments with Caco-2 cells
The cell monolayers were equilibrated in warm (37°C) transport medium (Hanks‚ balanced salt solution containing 10mM HEPES and buffered to pH7.4 with sodium hydroxide). The medium in the apical (for apical to basolateral transport) or basolateral (for basolateral to apical transport) chamber was removed and replaced with an equal volume of pre-warmed transport medium containing [14C]saquinavir (1μM; 20.38μCi/μmol). For inhibition studies the P-gp modulators (verapamil, 500μM; ketoconazole, 500μM) and protease inhibitors (ritonavir, nelfinavir and indinavir; 100μM) were added to the drug-containing transport medium.
At 30, 60, 90, 120 and 150 min, 500μl samples were taken from the receiver chamber and analysed using a liquid scintillation counter. Apparent permeability coefficients (Papp cm/s) were determined from the cumulative transport data.
Data were analysed by analysis of variance (ANOVA) followed by a Bonferroni t test.
Efflux of Rh123 from human lymphocytes
Lymphocytes that have incorporated the fluorescent dye, Rh123 after loading appear as a single broad peak (Fig. 1, inset a). After incubation at 37°C for 3 h in dye-free medium these cells have a bimodal distribution of Rh123 fluorescence indicating differential efflux of Rh123 in subsets of lymphocytes (Fig. 1, inset b). Those lymphocytes which retain Rh123 have high fluorescence intensity (FL1), and those which efflux Rh123 have low fluorescence intensity.
Temperature-dependence of Rh123 efflux and the effect of verapamil and protease inhibitors
Rh123 efflux was temperature sensitive with a significant increase (sevenfold; P<0.01) in the median fluorescence intensity at 4°C relative to the control lymphocytes (3 h incubation at 37°C, Fig. 1). This indicates that an energy-dependent active transport mechanism is involved in the efflux of Rh123.
To determine whether P-gp was responsible for the efflux of Rh123, stained lymphocytes were incubated in the presence of the P-gp modulator verapamil (30μM). This resulted in a 7.5 fold increase in median fluorescence intensity of the lymphocytes relative to the control cells (Fig. 1).
Saquinavir and ritonavir (10μM) markedly inhibited Rh123 efflux (Fig. 1) with a similar increase in fluorescence intensity to that obtained with verapamil. A small but significant increase in fluorescence intensity was observed with nelfinavir, however no significant effect was observed with indinavir.
Transport of saquinavir by Caco-2 cells
The cumulative BL-AP efflux of [14C]saquinavir (1μM) exceeded the AP-BL transport in Caco-2 cell monolayers (Fig. 2). This is indicative of an active efflux pump. Further evidence for the involvement of P-gp was the inhibition of BL-AP efflux of saquinavir by known P-gp modulators, verapamil and ketoconazole with a subsequent increase in AP-BL transport. Fig. 2 indicates Papp values for AP-BL and BL-AP transport for saquinavir alone and in the presence of P-gp modulators. The mean Papp value for BL-AP efflux of saquinavir exceeded that of AP-BL transport by a factor of 26 (206.7 versus 8.0×10-7cm/s respectively). In the presence of verapamil the ratio of Papp values was decreased to 3.6. A similar effect was observed with ketoconazole (ratio, 4).
Effect of the protease inhibitors on [14C]saquinavir transport
The effect of ritonavir, nelfinavir and indinavir at 100μM on the transport of saquinavir across Caco-2 cell monolayers is shown in Fig. 2. The protease inhibitors reduced the cumulative BL-AP efflux and increased the AP-BL transport of saquinavir. Ritonavir had the most marked inhibitory effect on the BL-AP efflux, reducing the Papp value for saquinavir by 59% with a sixfold increase in the AP-BL Papp. The overall Papp ratio was reduced from 26 (saquinavir alone) to 1.8. Corresponding ratios for indinavir and nelfinavir were 4.9 and 8.3 respectively. Monolayers used in these studies had a TER of 669±57Ω/cm2 (mean±SD, n=36) which remained essentially unchanged throughout the experiments.
We have shown that the protease inhibitors are modulators of P-gp function in human lymphocytes with the rank order of potency ritonavir>saquinavir>nelfinavir>indinavir. This further confirms studies performed in cultured cells indicating the protease inhibitors as P-gp modulators. Rh123 fluorescence was distributed bimodally in control lymphocytes indicating a differential efflux of Rh123. This is in agreement with previous reports indicating the heterogeneous expression of P-gp in subsets of human lymphocytes. Future studies are required to determine the modulation of Rh123 efflux in lymphocyte subsets and to correlate this with P-gp expression. Inhibition of P-gp was highly variable between cells obtained from different individuals confirming the inter-individual variability in P-gp expression amongst lymphocytic classes.
At present the concentrations of the protease inhibitors achieved inside their target cells in vivo remains unknown. Protease inhibitors inhibit HIV replication within lymphocytes and monocytes, hence the intracellular drug concentration will be an important determinant of in vivo efficacy. Lee etal.  showed that in KB-V1 cells which over-express P-gp, ritonavir, saquinavir and indinavir had a reduced inhibitory effect on HIV-1 replication. A limitation of the present study is the use of blood samples from healthy volunteers. As P-gp expression is reported to be altered in HIV infection, increasing with disease progression, further studies are required in lymphocytes obtained from HIV-positive patients.
This study also provides further evidence that saquinavir is a good substrate for P-gp in Caco-2 cells as indicated by a 26-fold difference in its directional transport. The affinity of saquinavir for P-gp may contribute to its low oral bioavailability as this pump acts to efflux xenobiotics from the enterocyte back into the intestinal lumen. As the substrate specificities of CYP3A4 and P-gp have been suggested to overlap, an inter-relationship between P-gp-mediated efflux and metabolic transformation of saquinavir by CYP3A4 may occur.
The effect of ketoconazole, a potent competitive CYP3A4 inhibitor, on saquinavir transport in our study was similar to that obtained with verapamil, the first P-gp modulator to be identified. Data from the present and other studies  point to ketoconazole being an inhibitor of P-gp. Therefore reduction of the oral clearance of CYP3A4 substrates previously attributed to CYP3A4 inhibition may also be mediated by its interaction with intestinal P-gp.
Previous clinical studies have shown that the increase in the area under the plasma saquinavir concentration time curve is greater in the presence of ritonavir than ketoconazole although they are both potent CYP3A4 inhibitors. This may be explained by their relative inhibition of P-gp. Our results indicate that ritonavir is a more potent P-gp inhibitor than ketoconazole. Thus inhibition of P-gp-mediated efflux by ritonavir may be a major contributor to the clinically observed 20-fold increase in the area under the plasma saquinavir concentration time curve. Alsenz etal.  previously suggested that ritonavir had a higher affinity for P-gp than saquinavir. In comparison, nelfinavir modulated P-gp function to a lesser extent than ritonavir, which may partly explain the smaller (fivefold) increase in the bioavailability of saquinavir when these drugs are co-administered.
In conclusion we have demonstrated that saquinavir is a substrate for P-gp and that ritonavir, nelfinavir and indinavir modulate P-gp function in both lymphocytes and Caco-2 cells.
Caco-2 cells, [14C]saquinavir, nelfinavir and saquinavir were generous gifts from Roche Products Ltd. (Welwyn Garden City, Hertfordshire, UK). Merck & Co, (West Point, Pasadena, USA) donated indinavir. Abbott (Chicago, Illinois, USA) donated ritonavir and ketoconazole was a gift from Janssen (Beerse, Belgium).
1. Perry C, Noble S. Saquinavir soft gel capsule formulation. Drugs
2. Fitzsimmons ME, Collins JM. Selective biotransformation of the human immunodeficiency virus protease inhibitor saquinavir by human small intestinal cytochrome P-450. Drug Metab Dispos
3. Eagling VE, Back DJ, Barry MG. Differential inhibition of cytochrome P450 isoforms by the protease inhibitors ritonavir, saquinavir and indinavir. Br J Clin Pharmacol
4. Condra JH, Schleif WA, Blahy OM, et al. In vivo emergence of HIV-1 variants resistant to multiple protease inhibitors. Nature
5. Wacher VJ, Wu CY, Benet LZ. Overlapping substrate specificities and tissue distribution of cytochrome P4503A and P-glycoprotein: implications for drug delivery and activity in cancer chemotherapy. Mol Carcinogen
6. Thiebaut F, Tsuruo T, Hamada H, et al. Cellular localisation of the multidrug-resistance gene product P-glycoprotein in normal human tissues. Proc Natl Acad Sci USA
7. Hunter J, Jepson MA, Tsuruo T, et al. Functional expression of P-glycoprotein in apical membranes of human intestinal Caco-2 cells. J Biol Chem
8. Alsenz J, Steffen H, Rainer A. Active apical secretory efflux of the HIV protease inhibitors saquinavir and ritonavir in Caco-2 cell monalayers. Pharm Res
9. Hoetelmans RMW. Sanctuary sites in HIV-1 infection. Antiviral Ther
10. Chaudhary PM, Mechetner EB, Roninson IB. Expression and activity of the multidrug resistance P-glycoprotein in human peripheral blood lymphocytes. Blood
11. Lee CGL, Gottesman MM, Cardarelli CO, et al. HIV-1 protease inhibitors are substrates for the MDR1 multidrug transporter. Biochemistry
12. Andreana A, Aggarwal S, Gollapudi S, et al. Abnormal expression of a 170-kilodalton P-glycoprotein encoded by MDR1 gene, a metabolically active efflux pump, in CD4+ and CD8+ T cells from patients with human immunodeficiency virus type 1 infection. AIDS Res Hum Retroviruses
13. Tsuruo T, Iida H, Tsukagoshi S, Sakurai Y. Overcoming of vincristine resistance in P388 leukemia in vivo and in vitro through enhanced cytotoxicity of vincristine and vinblastine by verapamil. Cancer Res
14. Siegsmund MJ, Cardarelli C, Aksentijevich I, et al. Ketoconazole effectively reverses multidrug resistance in highly resistant KB cells. J Urol
15. Merry C, Barry MG, Mulcahy F, et al. Saquinavir pharmacokinetics alone and in combination with ritonavir in HIV-infected patients. AIDS
16. Merry C, Barry MG, Mulcahy F, et al. Saquinavir pharmacokinetics alone and in combination with nelfinavir in HIV-infected patients. AIDS