UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
Limitations of current antiretroviral therapy in HIV-1 infection: the search for new
strategies
Sankatsing, S.U.C.
Publication date
2004
Link to publication
Citation for published version (APA):
Sankatsing, S. U. C. (2004). Limitations of current antiretroviral therapy in HIV-1 infection: the
search for new strategies.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.
CHAPTERR 10
Antivirall activity of HIV-1 protease inhibitors
nelfinavirr and indinavir in vivo is not influenced by
P-glycoproteinn activity on CD4
+T-cells
S.U.C.. Sankatsing, M. Cornelissen, N. Kloosterboer, K.M.LL Crommentuyn, T.M. Bosch, F.P. Mul, S. Jurriaans,
A.D.R.. Huitema, J.H. Beijnen, J.M.A. Lange, J.M. Prins, H.. Schuitemaker
Abstract t
Background:Background: The drug transporter P-glycoprotein (P-gp) can limit the uptake
off protease inhibitors (Pis) from the gut and reduces intracellular accumulation off Pis, which could limit the antiretroviral effect of a PI containing regimen for HIV-1.. However, P-gp expression on T cells can also reduce HIV-1 replication.. Here we studied the net effect of P-gp on the intracellular HIV-1 RNAA and DNA load in vivo.
Methods:Methods: From 27 HIV-1 patients (13 without therapy and 14 on a PI
containingg regimen) a blood sample was taken. CD4+ T cells were isolated andd subsequently sorted in naive (CD45RO) and memory (CD45RO+) subsetss with either high P-gp activity (P-gphigt1) or low Pgp activity (P-gplow). In thesee subsets unspliced HIV-1 RNA as a measure for HIV-1 replication, and thee HIV-1 DNA load were determined using a quantitative real-time (RT)-PCR.
Results:Results: In patients on a PI regimen intracellular unspliced HIV-1 RNA was
significantlyy lower in P-gph,gh naive CD4+ cells compared with P-gplow naive CD4++ cells (p=0.04). The same trend was seen in naive CD4+ cells of treatmentt naive patients. In both treated and untreated patients HIV-1 DNA levelss were significantly lower in P-gphigh than in P-gplow memory CD4+ cells (p=0.022 and p=0.04 respectively).
Conclusion:Conclusion: High cellular P-gp activity coincides with a lower intracellular
HIV-11 load in vivo. As this observation was made in therapy naive and PI treated patients,, the potential efflux function of P-gp on Pis seems clinically less relevantt than the effect of P-gp on intracellular HIV-1 replication.
EffectEffect ofP-gp activity in vivo
Introduction n
P-glycoproteinn (P-gp) is a cellular plasma membrane protein encoded by the multii drug resistance gene 1 (MDR-1), which was first discovered in 1976 1. It functionss as an ATP-dependent efflux pump for a wide range of compounds 2. Thee function of P-gp is being thoroughly investigated in the oncology field, becausee it may confer resistance to anti-cancer therapy by pumping cytostatic drugss out of tumour cells 3'4. Inhibition of P-gp activity has been studied in cancerr patients who received chemotherapy and this resulted in an improved relapse-freee period and overall survival, which was contributed to the inhibition off P-gp induced drug efflux from tumour cells 5"7.
Thee protease inhibitors (Pis) used for the treatment of HIV-1 infection 8 are substratess for P-gp 9"15. Given the distribution of P-gp in the human body 16, P-gpp may limit the oral uptake of Pis from the gut and the penetration of Pis into soo called sanctuary sites like the central nervous system 14,17. It has also been demonstratedd in vitro that P-gp can counteract the accumulation of Pis in CD4++ T-cells 18"20, which are one of the primary target cells for HIV-1 replication,, thereby limiting the antiretroviral effect of Pis 15. Taken together, P-gpp activity might create sanctuary sites for HIV-1 and facilitate residual low-levell HIV-1 replication during highly active antiviral therapy 21. Inhibition of P-gpp activity to improve the antiviral activity of Pis has been considered 16. However,, P-gp has also been demonstrated to decrease HIV-1 replication in vitro,, by interfering with HIV-1 fusion with the plasma membrane and also at a laterr step in the viral life cycle 22,23. It is therefore of great importance to elucidatee the exact effects of P-gp activity on HIV-1 replication and PI therapy inn vivo 16. Here we compared the levels of HIV-1 RNA and DNA in purified CD4++ T cells with high or low P-gp activity that were obtained from therapy naïvee HIV-1 infected patients and from patients receiving a therapy regimen thatt included the Pis nelfinavir (NFV) or indinavir (IDV).
Methods s
Patients Patients
Peripherall blood was collected from 27 HIV-1 infected patients, of whom 13 weree therapy naive and 14 were on a Pl-containing regimen (consisting of
eitherr NFV or IDV boosted with ritonavir) for at least 8 weeks. From every patientt 200 ml_ EDTA-anticoagulated blood was taken once. The Medical Ethicss Committee of our hospital approved the study and written informed consentt was obtained from all patients.
CellCell isolation
Peripherall blood mononulear cells (PBMC) were isolated from fresh EDTA bloodd by Ficoll density gradient centrifugation. CD4+ cells were isolated by positivee selection using anti-CD4 coated magnetic beads (Macs CD4 microBeads,, Miltenyi biotec, Bergisch Gladbach, Germany). All cell isolation stepss were performed on ice with chilled buffers supplemented with 10u.M -verapamill hydrochloride (Sigma, St. Louis, MO), to minimise P-gp activity. Previouss studies have used gp membrane expression as a measure for P-gpp activity, but P-gp expression not necessarily predicts P-gp activity 16. Therefore,, for purification of CD4+ cells with high or low P-gp activity, we incubatedd CD4+ T lymphocytes in medium (iscove's modified dulbecco's mediaa supplemented with 10% fetal calf serum) with 0.5 ug/ml rhodamine 123 (rho123)) (Sigma, St. Louis, MO) for 1 hour at 37"C. Subsequently, cells were washedd and resuspended in medium after which rho123 influx was immediatelyy established by FACS analysis. Next, CD4+ cells loaded with rho1233 were incubated overnight in medium at . Cells were then stained withh monoclonal antibodies directed against CD4 (PercP conjugated, Becton Dickinson,, San Jose, Ca) and CD45RO (APC conjugated, Becton Dickinson) andd sorted on a MoFlo High Speed Cell Sorter using a 4 way sort system (DakoCytomation,, Fort Collins, Colorado) in a CD45RO+rho123low, CD45RO+rho123high,, CD45ROrho123low and a CD45ROrho123high CD4+ T celll population. High rho123 Mean Fluorescence Intensity (MFI) reflected low P-gpp activity and low rho123 MFI reflected high P-gp activity. Therefore we referr to these cell populations as CD45RO+P-gphigh, CD45RO+P-gplow, CD45ROP-gphighh and CD45RO" P-gplow
EffectEffect ofP-gp activity in vivo
PlasmaPlasma HIV-1 RNA
HIV-11 RNA levels in EDTA plasma were measured using the Versant HIV-1 RNAA 3.0 assay (bioMérieux, Boxtel, the Netherlands) according to the manufacturer'ss instructions, with a detection limit of 50 copies/ml.
QuantitativeQuantitative Real-time PCR for measurements of cell associated HIV-1 DNA andand unspliced HIV-1 RNA
HIV-11 proviral DNA and unspliced intracellular HIV-1 RNA were measured usingg real-time quantitative PCR assays.
Primerss were designed with Primer Express software (ABI) and amplify a 79-bpp amplicon of the p24 part of the GAG gene. 5' primer, forward:
5'AAAGAGACCATCAATGAGGAAGC3'' and 3'primer: 5TCTGCCTGGTGCAATAGG3'' fluorescent probe
5'-6FAM-TGCACTGGATGCACTCTATCCCATT-TAMRA-3'.. Primers were empirically testedd and detect both HIV-1 RNA and HIV-1 proviral DNA.
Afterr sorting, 106 P-gph,gh or P-gplow cells were lysed in guanidine thiocyanate buffer.. For the quantitative HIV-1 RNA/DNA assay 2.5x105 cells were used andd nucleic acids were isolated by a silica-based method 24. RT-reaction was carriedd out in duplex, according to the ABI Taqman Reverse Transcription reagentss kit (ABI). Each RT reaction contained 1 pi 10xRT buffer, 2.2 pi MgCb (25mM),, 2 pi dNTPs (10mM) 1.5 pi random hexamers (50uM), 0.2 Rnase inhibitorr (20U/pl), 0,35 pi multiscribe RT (50U/pl), 0,35 pi H20 and 2.5 pi
samplee eluate (= 10,000 cell equivalents). The RT-reaction started with 10 minutess 25 , 30 minutes 48 C and 5 minutes . The PCR reaction was donee according to Platinum® Quantitative PCR Supermix UDG (Invritogen life technologies,, Carlsbad, USA). Each reaction contained 10 pi RT-reaction or 100 pi eluate (= 40,000 cell equivalents) and 40 pi of PCR mixture consisting of 255 pi supermix, 3.6 mM MgCb, 0.9 uM forward primer and reverse primer, 0.2 pMM Taqman probe and 1pl ROX reference dye <50X concentration). PCR conditionss consisted of an activation step of UDG (2 min, 50 , activation of Platinum®® Taq DNA polymerase (10 min, ) followed by 45 cycles of 15 secc at C and 1 min at .
Seriall diluted Scott Layne particles , in vitro transcibed HIV-1 RNA or plasmidd HIV-1 DNA were also subjected to the above RT-PCR or PCR to obtainn standard curves. No significant differences were observed in the standardss for RNA versus DNA (RNA standard: y = -1.422Ln(x) + 40,387 versuss DNA standard: y = -1.404Ln(x) +40.202).
Ass a control a real time PCR human p actin was used in which 5 ul (20,000 celll equivalents) sample eluate was tested according to the manufacturer's instructionss (Applied Biosystems). The p actin reaction conditions were the samee apart from the primer and probe concentrations (0.3 uM forward and reversee primers and 0.2 uM Taqman probe). A standard curve from 1.67 x 102 -- 1.67 x 104 copies of p actin included in the kit was used as input (r^O.99). Thee average Ct value for the P actin was 22.5 0.52. Ct value. The lower limit off detection of the RT-PCR is 25 copies of RNA and 5 copies DNA. Samples weree considered negative if the Ct value exceeded 45 cycles.
ConcentrationsConcentrations ofnelfinavir and indinavir
Forr measuring PI concentrations in plasma and PBMCs, plasma was separatedd from EDTA blood by centrifugation and stored at -20 C until use andd PBMCs were isolated as described above. Drug concentrations of NFV andd IDV in plasma and in a cell pellet with a median of 6.6 x106 cells were simultaneouslyy measured with a validated method using high-performance liquidd chromatography coupled with electrospray tandem mass spectrometry (LC-MS/MS)266 and expressed in mg/L.
IdentificationIdentification ofP-gp polymorphisms
Genotypingg for single nucleotide polymorphisms (SNP) in exons 12, 21 and 266 was performed. Genomic DNA was isolated from EDTA blood using the Boomm method 24. SNPs were determined with a PCR assay followed by DNA sequencingg according to a modified assay that has been described in detail in previouss studies 27,28.
StatisticalStatistical analyses
EffectEffect ofP-gp activity in vivo
Thee intracellular unspliced HIV-1 RNA and HIV-1 DNA load in P-gphigh and P-gploww cells were compared with the Wilcoxon test. Pearson correlation was usedd to study the correlation between plasma HIV-1 RNA and intracellular unsplicedd HIV-1 RNA and DNA, and the correlation between intracellular unsplicedd HIV-1 RNA and DNA and the plasma concentrations of NFV and IDV.. The Kruskal-Wallis test was used to study the relation between allelic variantss of P-gp polymorphisms and plasma HIV-1 RNA in therapy naïve patientss and in patients using NFV and the relation between allelic variants andd plasma NFV concentrations.
Results s
Patients Patients
Alll patients were male with a median age of 43 (interquartile range 34-53) years.. The therapy regimen of the patients and their plasma HIV-1 RNA concentrationss at the time of blood sampling are given in table 1. In 4 of these 144 patients who received a P! containing regimen, plasma HIV-1 RNA was detectablee (table 1). In the untreated patients the plasma HIV-1 RNA concentrationn ranged between 3.62 and 5.00 log™ copies/mL.
ExEx vivo comparison of intracellular ievels of unspliced HIV-1 RNA and proviral
DNADNA in CD4+ T cells with high or low P-gp activity
Wee first set up a method to sort cells on the basis of P-gp activity. Purified CD44 cells from our study subjects were incubated with rho123 for 1 hour after whichh intracellular rho123 uptake was established by FACS analysis (Fig 1a). Afterr further overnight incubation at 37 , cells were stained with antibodies directedd against CD45RO, to allow a rough discrimination between memory (CD45RO+)) and naive (CD45RO) T cells. Using a FACsorter, naive (CD45RO)) and memory (CD45RO+) CD4+ T cells with either high (P-gphiQh) or loww (P-gplow) P-gp activity were purified (Fig 1b). When cells from P-gp knockoutt mice 29 were used in this assay, no efflux of rho123 could be measured,, indicating that our assay is indeed specific for monitoring P-gp activityy (data not shown). One could argue that rho123 and PI may compete forr P-gp and consequently that PI levels in the patient directly influence the
effluxx of rho123 in our assay. However, the presence of PI added in vitro to thee cultures of purified cells had no effect on the efflux of rho123 during the overnightt incubation period (data not shown).
Tablee 1 Therapy used by the patients and their plasma HIV-1 RNA concentration at thee time of blood sampling
Patient t 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 10 0 11 1 12 2 13 3 14 4 Therapy y NFV,, ddl, d4T NFV,, RTV, ddl, d4T NFV,, CBV NFV,, TZV NFV,, CBV NFV,, CBV IDV,, RTV, CBV IDV,, RTV, CBV IDV,, RTV, CBV IDV,, RTV, CBV IDV,, RTV, 3TC, d4T NFV,, 3TC, d4T NFV,, 3TC, d4T, NVP NFV,, 3TC, d4T
Plasmaa HIV-1 RNA (log10 copies/ml)
<1.70 0 <1.70 0 <1.70 0 <1.70 0 <1.70 0 <1.70 0 <1.70 0 <1.70 0 <1.70 0 <1.70 0 4.15 5 3.97 7 4.60 0 3.39 9
NFV=nelfinavir,, IDV= indinavir, RTV=ritonavir, ddl=didanosine, d4T=stavudine, CBV=combivirr (zidovudine and lamuvidine), TZV=trizivir (zidovudine, lamivudine and abacavir),, 3TC=lamivudine, NVP=nevirapine.
Wee observed a statistically significant difference in intracellular unspliced RNA betweenn P-gph,gh and P-gplow cells in naïve CD4+ T cells of patients treated withh a PI containing regimen. In this group the median unspliced intracellular HIV-11 RNA was 0.70 (interquartile range, IQR, 0.70-0.94) logio copies/104 P-gphighh cells compared with a median of 1.13 (IQR 0.70-1.94) log™ copies/104 P-gploww cells (p=0.04; table 2, figure 2). The same trend, albeit not statistically significant,, was observed in intracellular HIV-1 RNA levels in P-gphigh and P-gploww naïve CD4+ T cells of untreated patients (table 2; figure 2). Intracellular HIV-11 RNA levels in P-gphigh and P-gplow cells were indifferent in memory CD4++ T cells of both therapy naïve patients and patients receiving a PI containingg therapy regimen (table 2; figure 2).
EffectEffect of P-gp activity in vivo
Figuree 1 Representative rhodamine 123 staining showing P-gp activity in CD4+ T cells
-*—» » c c O O O O 10 0 f * H w * # f f W < H , , i o '' 102 10S r h o 1 2 3 ( M F I ) ) io o rho1233 (MFI)
A.. Intracellular rho123 uptake by CD4* T cells after 1 hour of incubation (M1=98.6%) B. Rho1233 efflux after overnight incubation at 37 C of CD45RO stained CD4+ T cells. UL: CD45RO+ ww represents the P-gphigh memory CD4+ T cells; LL: CD45RO"rho123low representss the P-gphigh naive CD4+ T cells; UR: CD45RO+rho123high represents the P-gp'0™ memoryy CD4+ T cells; LR: CD45RO"rho123high represents the P-gplow naïve CD4+ T cells.
Percentagee of gated cells are depicted in the upper right corner of graph. Using MoFlo Sorter gatedd cells were isolated in the four different fractions for further analysis.
AA significant difference in HIV-1 DNA levels was observed between P-gphigh andd P-gplow cells with a memory phenotype in therapy naïve patients and in patientss on a PI containing regimen (figure 2). In the P-gphigh memory cells of naivee patients, the HIV-1 DNA load was 0.70 (IQR 0.70-2.03) log™ copies/104 cellss as compared to 1.38 (IQR 0.74-2.16) log™ copies/104 P-gplow memory cellss (p=0.04; table 2, figure 2). In patients on a PI regimen, the median provirall DNA load in P-gphigh memory cells was 0.74 (IQR 0.70-0.99) log™ copies/1044 cells as compared to a median of 0.95 (IQR 0.70-1.36) log™ provirall DNA copies/104 P-gplow memory cells (p=0.02; table 2, figure 2).
Separatee analyses which excluded those patients who were unsuccessfully treatedd with a PI containing regimen, as reflected by a detectable plasma HIV-11 RNA load (n= 4), provided identical results (data not shown).
10 o o c c C D D
I I
C D D C D D1 1
x : : Q . . C O O O O o o c c C O O < < Z Z or r 0 0 O O "a . . C O O c c 3 3 > > C M M a> > o o SS " b ££ ' — Q .. "5 5 O )) .3 ? ** ! O O z z a> > O O11
'
Q .. ^ B O )) — 11 a . u o 6 ) ) 'Q . . "a i i o o 11 S Q __ " R << o ZZ e fcfc 8 » "OO — "a .. 2 COO 4 > CC ü = >> | o Q .. ^ ?55 £ uu ~~ o *2 2 C D D "5 5 5 a . . 0.55 5 0.04 4 0.70(0.70-1.77) ) 1.38(0.74-2.16) ) 0.70(0.70-1.60) ) 0.700 (0.70-2.03) 0.09 9 0.31 1 1.411 (0.80-3.11) 2.111 (1.16-3.09) 1.00(0.70-2.83) ) 2.09(0.70-3.16) )Therapyy naive patients (n=13) Naïvee CD4 cells + Memoryy CD4+ cells 0.50 0 0.02 2 0.700 (0.40-0.70) 0.95(0.70-1.36) ) 0.700 (0.30-0.70) 0.744 (0.70-0.99) 0.04 4 0.14 4 1.13(0.70-1.94) ) 1.53(1.00-2.25) ) 0.700 (0.70-0.94) 1.07(0.70-1.58) ) PII treated patients (n=14)
Naïvee CD4 cells + Memoryy CD4 cells +
EffectEffect of P-gp activity in vivo
Figuree 2 Intracellular unspliced HIV-1 RNA and HIV-1 DNA load in naive and
memoryy CD4+ cells with high P-gp activity (P-gph,gh) or low P-gp activity (P-gplow) of patientss that are without therapy (naive), or on a PI containing regimen
So So %% 'S 3 88 s aa g üü .9! SS &2
Naivee cells naïve patients
Naivee cells PI patients
Naivee cells naive patients
ÓÓ 2 mm Q. 2 cc « "SS "5 3 uu & «« o. 2
Memoryy cells naïve patients
Memoryy cells PI patients
P-gph»ghh P-gplow
Memoryy cells PI patients
P-gphighh P-gplow ZZ * 5 oo S
33 I
3 SS .a rara o . 2 —— o cc u —— 1Naivee cells PI patients Memoryy cells PI patients
CorrelationCorrelation between plasma HIV-1 RNA concentration and intracellular unsplicedunspliced RNA and DNA load in therapy naive patients
Theree was a correlation between plasma HIV-1 RNA and unspliced RNA in memoryy cells with low P-gp activity (r=0.61;p=0.04). There was however no correlationn between plasma HIV-1 RNA levels and the intracellular unspliced RNAA load in naive cells with either high or low P-gp activity and memory cells withh high P-gp activity (table 3).
Theree was also a correlation between the plasma HIV-1 RNA concentration andd the proviral DNA load in all CD4+ T cell subsets tested, with the exception off P-gphi9h naive CD4+ cells (table 3).
Tablee 3 Correlation between plasma HIV-1 RNA concentration and intracellular unsplicedd HIV-1 RNA and DNA load in the different T-cell subsets, in therapy naive patients s
Celll subset
Naivee CD4+ cells P-gpnign Naivee CD4+ cells P-gplow Memoryy CD4+ cells P-gph'9h Memoryy CD4+ cells P-gplow
Unsplicedd RNA r r 0.39 9 0.52 2 0.57 7 0.61 1 P P 0.24 4 0.09 9 0.07 7 0.04 4 DNA A r r 0.39 9 0.64 4 0.76 6 0.61 1 P P 0.24 4 0.03 3 0.007 7 0.04 4 rr = correlation coefficient
ConcentrationsConcentrations ofnelfinavir and indinavir
Alll but one patient who were receiving a PI containing therapy regimen had NFVV or IDV plasma concentrations above the minimum desired concentration. Onee patient had a plasma IDV concentration of 0.23 mg/L at 24 hours after finall intake, which could be attributed to a missed dosage. This patient howeverr still had an undetectable plasma HIV-1 RNA load.
Forr none of the cell subsets we found a correlation between plasma concentrationss of NFV and IDV and unspliced HIV-1 RNA load. The same wass true for the plasma concentrations of NFV and IDV and HIV-1 proviral DNAA load (data not shown). Intracellular PI concentrations unfortunately could nott be quantified in our patients, because the amount of available cells for thesee analyses was too low.
EffectEffect ofP-gp activity in vivo
PolymorphismPolymorphism of P-gp
Polymorphismss In exon 12, 21, and 26 of MDR-1, the gene encoding P-gp, havee been associated with P-gp function 27. In a total of 16 patients (10 therapyy naive patients and 6 patients who were on a PI containing regimen) wee determined the previously described polymorphisms in exons 12, 21 and 26.. Allele frequencies of C1236T in exon 12 were 28.6% CC, 57.1% CT and 14.3%% TT. At position 2677 in exon 21 the allele frequencies of G2677T were 26.7%% GG, 53.3% GT and 20% TT and for C3435T in exon 26 it was 25% CC,, 50% CT and 25% TT. There were no differences in plasma viral load betweenn groups of individuals with different allelic variants of exons 12, 21 andd 26 in either therapy naive patients (p=0.91, p=0.77 and p=0.81, respectively)(figuree 3) or patients on a NFV containing regimen (p=0.14, p=0.144 and p=0.14, respectively). Moreover, there was no difference in plasmaa NFV concentrations between individuals with different allelic variants off exon 12, 21 and 26 (p=0.22, p=0.22 and p=0.20). No data on the allelic frequencyy of exon 12, 21, and 26 were available from patients on an IDV containingg regimen. Finally, there was no correlation between P-gp genotype andd percentage of P-gphigh cells (data not shown).
Discussion n
Basedd on experimental in vitro data and animal studies it has been suggested thatt P-gp may reduce intracellular concentrations of Pis in vivo, thereby compromisingg the effectiveness of a PI containing antiretroviral therapy regimenn 14-16-20. other in vitro studies have implicated that P-gp might interfere withh HIV-1 replication 22,23, which would result in a lower HIV load. Here we madee an attempt to elucidate the net effect of P-gp activity on the efficacy of antiretrovirall therapy and on the viral load in therapy naive individuals. We comparedd intracellular viral RNA and proviral DNA levels in cells with either highh (P-gph'9h) or low (P-gplow) activity obtained from the same patient. This approachh circumvented the potential problem of large interpatient variability in P-gpp activity.
Exonn 12C1236T 55 0 E E "33 5 5 o o Q_ _ O O O O g>> 5.0 < < Z Z CC CC T -- 4 5 > > I I (0 (0 E E (0(0 4.0 D. . 3.5 5 :: . - . Exonn 21 G2677T Exonn 26 C3435T
Figuree 3 Plasma HIV-1 RNA (log10
copies/ml)) in therapy naive patientss with the respective P-gp polymorphismss of exon 12 (upper figure),, exon 21 (middle figure) andd exon 26 (lower figure). Horizontall lines represent the median,, blocks represent the interquartilee range.
EffectEffect ofP-gp activity in vivo
Itt has been reported that P-gp activity is higher in memory T cells than in naïvee T cells 30. As patients may differ in their naïve/memory T cell ratio, analysiss of the CD4+ T cell subset in total could have introduced a bias in our results.. Therefore we separately analysed the naïve and memory CD4+ T cell subsets. .
Levelss of intracellular unspliced RNA were lower in patients who were successfullyy treated with a PI containing regimen as compared to patients whoo were not successfully treated or patients not receiving any therapy. However,, even in these successfully treated patients with a plasma HIV-1 RNAA load below the limit of detection intracellular unspliced RNA could still be measured,, which is in accordance with previous studies 31~33. Our assay does nott allow the discrimination between genomic viral RNA and unspliced messengerr RNA. As both RNA species are relatively unstable, the presence off either form of RNA is a reflection of ongoing viral replication.
Wee observed in patients receiving PI therapy a significantly lower level of intracellularr RNA in naïve T cells with high P-gp activity. The same trend was observedd in therapy naïve individuals. Moreover, in both therapy naïve patientss and patients receiving antiretroviral therapy, the HIV-1 proviral DNA loadd was significantly lower in P-gphigh memory cells than in P-gplow memory cells.. This observation is in line with the in vitro observation that HIV-1 replicationn is lower in cells with high P-gp activity, which may be caused by thee interference of P-gp with the fusion of HIV-1 22,23, although there are indicationss that P-gp can also inhibit HIV-1 replication in later steps of the replicationn cycle 22. These findings are also in accordance with a previous studyy demonstrating an inverse correlation between P-gp activity and plasma HIV-11 RNA 34. However, we cannot conclude a causal relationship, as P-gp activityy may for instance coincide with reduced coreceptor expression levels andd consequently a reduced HIV-1 susceptibility.
Wee did not observe an association between P-gp activity and intracellular HIV-11 RNA load in the memory CD4+ T cell subset. As HIV-1 replication in
memoryy T cells is most optimally supported by cellular factors, a potential inhibitoryy effect of P-gp may be too small to achieve any measurable effect. Ourr data do implicate that the potential effect of P-gp on lowering the intracellularr PI concentrations is limited in vivo, as this would have resulted in aa higher intracellular HIV load.
Previously,, a correlation has been demonstrated between a P-gp polymorphismm at position 3435 in exon 26 and P-gp activity. A homozygous T genotypee was associated with a lower plasma nelfinavir concentration in vivo
35
.. In 16 individuals of our study population, the allele frequencies of polymorphismss in exons 12, 21 and 26 of P-gp were comparable with those publishedd before 36. We did not find a relation between P-gp polymorphisms andd P-gp activity, plasma nelfinavir concentrations or plasma HIV-1 RNA load, whichh is in contrast with previous studies. As these other studies included moree patients, the absent correlation in our present analysis may be due to loww patient numbers.
Inn conclusion, we have shown here that high cellular P-gp activity coincides withh a lower intracellular HIV-1 load in vivo. As this observation was made in therapyy naive and PI treated patients, the potential efflux function of P-gp on Piss seems clinically less relevant. Although our study does not exclude the potentiall beneficial effect of P-gp inhibitors on potentiating PI concentrations inn sanctuary body sites, the clinical application of such inhibitors to potentiate intracellularr PI concentrations should be reconsidered.
Acknowledgements s
Thiss study was financially supported by a private, non-commercial foundation thatt wishes not to be named and a grant from the Landsteiner Foundation for
Bloodd Transfusion Research (LSBR) project number 0024. We thank Alfred H. Schinkel,, Division of Experimental Therapy, The Netherlands Cancer Institute forr helpful discussions.
EffectEffect ofP-gp activity in vivo
References s
1.. Juliano RL, Ling V. A surface glycoprotein modulating drug permeability in Chinesee hamster ovary cell mutants. Biochim Biophys Acta. 1976;455:152-162. .
2.. Sharom FJ, Yu X, Doige CA. Functional reconstitution of drug transport and ATPasee activity in proteoliposomes containing partially purified P-glycoprotein.. J Biol Chem. 1993;268:24197-24202.
3.. Gottesman MM, Pastan I. Biochemistry of multidrug resistance mediated by thee multidrug transporter. Annu Rev Biochem. 1993;62:385-427.
4.. Krishna R, Mayer LD. Multidrug resistance (MDR) in cancer. Mechanisms, reversall using modulators of MDR and the role of MDR modulators in influencingg the pharmacokinetics of anticancer drugs. Eur J Pharm Sci. 2000;11:265-283. .
5.. Belpomme D, Gauthier S, Pujade-Lauraine E et al. Verapamil increases the survivall of patients with anthracycline-resistant metastatic breast carcinoma.
AnnAnn Oncol. 2000;11:1471-1476.
6.. List AF, Kopecky KJ, Willman CL et al. Benefit of cyclosporine modulation of drugg resistance in patients with poor-risk acute myeloid leukemia: a Southwestt Oncology Group study. Blood. 2001;98:3212-3220.
7.. Millward MJ, Cantwell BM, Munro NC et al. Oral verapamil with chemotherapy forr advanced non-small cell lung cancer: a randomised study. Br J Cancer. 1993;67:1031-1035. .
8.. Flexner C. HIV-protease inhibitors. N Engl J Med. 1998;338:1281-1292. 9.. Clayette P, Jorajuria S, Dormont D. Significance of P-glycoprotein for the
pharmacologyy and clinical use of HIV protease inhibitors. AIDS. 2000;14:235-236. .
10.. Gutmann H, Fricker G, Drewe J et al. Interactions of HIV protease inhibitors withh ATP-dependent drug export proteins. Mol Pharmacol. 1999;56:383-389. 11.. Huisman MT, Smit JW, Schinkel AH. Significance of P-glycoprotein for the
pharmacologyy and clinical use of HIV protease inhibitors. AIDS. 2000; 14:237-242. .
12.. Huisman MT, Smit JW, Wiltshire HR et al. P-glycoprotein limits oral availability,, brain, and fetal penetration of saquinavir even with high doses of ritonavir.. Mol Pharmacol. 2001;59:806-813.
13.. Kim AE, Dintaman JM, Waddell DS et al. Saquinavir, an HIV protease inhibitor,, is transported by P-glycoprotein. J Pharmacol Exp Ther. 1998;286:1439-1445. .
14.. Kim RB, Fromm MF, Wandel C et al. The drug transporter P-glycoprotein limitss oral absorption and brain entry of HIV-1 protease inhibitors. J Clin
Invest.Invest. 1998;101:289-294.
15.. Lee CG, Gottesman MM, Cardarelli CO et al. HIV-1 protease inhibitors are substratess for the MDR1 multidrug transporter. Biochemistry. 1998;37:3594-3601. .
16.. Sankatsing SUC, Beijnen JH, Schinkel AH et al. P-glycoprotein in Human Immunodefiencyy Virus 1 Infection and Therapy. Antimicrob Agents
Chemother.Chemother. 2004;48:1073-1081.
17.. Washington CB, Wiltshire HR, Man M et al. The disposition of saquinavir in normall and P-glycoprotein deficient mice, rats, and in cultured cells. Drug
MetabMetab Dispos. 2000;28:1058-1062.
18.. Jones K, Hoggard PG, Sales SD et al. Differences in the intracellular accumulationn of HIV protease inhibitors in vitro and the effect of active transport.. AIDS. 2001;15:675-681.
19.. Jones K, Bray PG, Khoo SH et al. P-glycoprotein and transporter MRP1 reducee HIV protease inhibitor uptake in CD4 cells: potential for accelerated virall drug resistance? AIDS. 2001;15:1353-1358.
20.. Meaden ER, Hoggard PG, Newton P et al. P-glycoprotein and MRP1 expressionn and reduced ritonavir and saquinavir accumulation in HIV-infected individuals.. J Antimicrob Chemother. 2002;50:583-588.
21.. Saksena NK, Potter SJ. Reservoirs of HIV-1 in vivo: implications for antiretrovirall therapy. AIDS Rev. 2003;5:3-18.
22.. Lee CG, Ramachandra M, Jeang KT et al. Effect of ABC transporters on HIV-11 infection: inhibition of virus production by the MDR1 transporter. FASEB J. 2000;14:516-522. .
23.. Speck RR, Yu XF, Hildreth J et al. Differential effects of p-glycoprotein and multidrugg resistance protein-1 on productive human immunodeficiency virus infection.. J Infect Dis. 2002;186:332-340.
24.. Boom R, Sol CJ, Salimans MM et al. Rapid and simple method for purification off nucleic acids. J Clin Microbiol. Microbiol. 1990;28:495-503.
EffectEffect ofP-gp activity in vivo
25.. Layne SP, Merges MJ, Dembo M et al. Factors underlying spontaneous inactivationn and susceptibility to neutralization of human immunodeficiency virus.. Virology. 1992;189:695-714.
26.. Crommentuyn KM, Rosing H, Nan-Offeringa LG et al. Rapid quantification of HIVV protease inhibitors in human plasma by high-performance liquid chromatographyy coupled with electrospray ionization tandem mass spectrometry.. J Mass Spectrom. 2003;38:157-166.
27.. Hoffmeyer S, Burk O, von Richter O et al. Functional polymorphisms of the humann multidrug-resistance gene: multiple sequence variations and correlationn of one allele with P-glycoprotein expression and activity in vivo.
ProcProc Natl Acad Sci USA. 2000;97:3473-3478.
28.. Kim RB, Leake BF, Choo EF et al. Identification of functionally variant MDR1 alleless among European Americans and African Americans. Clin Phannacol
Ther.Ther. 2001;70:189-199.
29.. Schinkel AH, Smit JJ, van Tellingen O et al. Disruption of the mouse mdMa P-glycoproteinn gene leads to a deficiency in the blood-brain barrier and to increasedd sensitivity to drugs. Cell. 1994;77:491-502.
30.. Gupta S, Gollapudi S. P-glycoprotein (MDR 1 gene product) in cells of the immunee system: its possible physiologic role and alteration in aging and humann immunodeficiency virus-1 (HIV-1) infection. J Clin Immunol. 1993;13:289-301. .
31.. Furtado MR, Callaway DS, Phair JP et al. Persistence of HIV-1 transcription inn peripheral-blood mononuclear cells in patients receiving potent antiretroviraltherapy.. N EnglJ Med. 1999;340:1614-1622.
32.. Sharkey ME, Teo I, Greenough T et al. Persistence of episomal HIV-1 infectionn intermediates in patients on highly active anti-retroviral therapy. Nat
Med.Med. 2000;6:76-81.
33.. Fraser C, Ferguson NM, Anderson RM. Quantification of intrinsic residual virall replication in treated HIV-infected patients. Proc Natl Acad Sci USA. 2001;98:15167-15172. .
34.. Hulgan T, Donahue JP, Hawkins C et al. Implications of T-cell P-glycoprotein activityy during HIV-1 infection and its therapy. J Acquir Immune Defic Syndr. 2003;34:119-126. .
35.. Fellay J, Marzolini C, Meaden ER et al. Response to antiretroviral treatment inn HIV-1-infected individuals with allelic variants of the multidrug resistance transporterr 1: a pharmacogenetics study. Lancet. 2002;359:30-36.
36.. Haas DW, Wu H, Li H et al. MDR1 gene polymorphisms and phase 1 viral decayy during HIV-1 infection: an adult AIDS Clinical Trials Group study. J
