HIV-1 protease inhibitors (PI) block the processing of the Pr55Gag and Pr160Gag/Pol polyprotein precursors by protease. Highly active antiretroviral therapies that contain PI have led to dramatic decreases in plasma viremia and corresponding reductions in morbidity and mortality rates . However, the relatively rapid emergence of viruses that are resistant to these compounds substantially limits the long-term treatment efficacy of currently available regimens [2,3].
In the present study, analysis was performed on HIV-1 variants derived previously by selection in vitro for resistance to amprenavir (formerly VX-478 or 141 W94) [4,5]. Viruses selected in vitro for decreased susceptibility to amprenavir  carried protease mutations similar to those observed in patients treated with amprenavir . Isolates from amprenavir-treated patients exhibit decreased amprenavir susceptibility and have mutations in the HIV-1 protease gene, resulting in amino acid substitutions, primarily at positions M46I/L, I47V, I50V, I54L/V, and I84V, as well as mutations in the Gag p1/p6 cleavage site . Varying degrees of HIV-1 cross-resistance among PI have been reported previously [6,8]. However, PI cross-resistance in HIV-1 isolates from amprenavir-treated patients has not been fully evaluated.
The accumulation of mutations in virus grown in increasing concentrations of amprenavir in vitro  provided an opportunity to dissect the interrelated effects of mutations on amprenavir resistance and replication capacity. Cross-resistance to other PI, including the newest compound available for widespread use (formulated in a fixed combination with ritonavir as KaletraTM), lopinavir, was also studied.
Materials and methods
Viruses emerged during serial passages of HIV-1IIIB in the presence of increasing concentrations of amprenavir, as previously reported . The sequence of emergence of mutant virus was L10F (M1c+), L10F/I84V (M2c+), L10F/M46I/I50V (M4c+) and L10F/M46I/I47V/I50V (M5c+). Each of these viruses had a L449F mutation in the Gag p1/p6 site that is cleaved by protease (designated by ‘+’ in the names). Gag-Pol polymerase chain reaction (PCR) products amplified from these amprenavir-selected HIVIIIB isolates were cloned into a genomic proviral vector (pJM11ΔGPR)  derived from HIVNL4−3. This produced the following recombinant viruses: L10F/I84V (M2r−), L10F/I84V/Gag L449F (M2r+), L10F/M46I/I50V/Gag L449F (M4r+) and L10F/M46I/I47V/I50V/Gag L449F (M5r+). The ‘−’ indicates a lack of the mutation in Gag p1/p6 cleavage site (e.g. the wild-type 449L). To serve as a control, a recombinant virus containing the corresponding Gag-Pol fragment of HIV-1IIIB was cloned into the same vector (IIIB−). A mutant virus containing protease I50V was constructed by site-directed mutagenesis in HIVNL4−3 isogenic background . Pol and Gag (p7, p1 and p6) sequences in all viral vectors were confirmed before virus production. Stocks of virus were generated in MT-2 cells by electroporation. The 50% tissue culture-infectious dose (TCID50) of each virus stock was determined by endpoint dilution in MT-2 cells .
A rapid recombinant assay was used to measure the drug susceptibility of the HIVIIIB variants selected in vitro, the recombinant HIVNL4−3 virus clones derived from each HIVIIIB isolate, and isogenic viruses derived by site-directed mutagenesis (PhenoSense HIV; ViroLogic, South San Francisco, CA, USA) . This assay involves the construction of resistance test vectors, which consist of a pool of recombinant HIV-1 containing Gag (3′-end from p7), protease and reverse transcriptase sequences derived from the virus sample that is being evaluated. Resistance test vectors also contain a luciferase reporter gene replacing env to monitor a single round of virus replication. The susceptibility of resistance test vectors to a panel of HIV-1 PI was compared with a reference vector containing the protease and reverse transcriptase sequences derived from HIV-1NL4−3. Two independent measurements of each viral isolate were obtained. The correlation in the susceptibility (IC50) between different drugs (measuring the degree of cross-resistance) was estimated using Spearman's rank correlation test.
Replication capacity was measured using a modified version of the PhenoSense drug-susceptibility assay [13,14]. The relative replication capacity of the virus was determined by measuring the amount of luciferase activity produced 72 h after infection in the absence of drug. Replication capacity is expressed as the percentage of the luciferase activity produced by the vectors containing mutant Gag-Pol sequences compared with the luciferase activity from vectors containing the HIV-1NL4−3 Gag-Pol reference sequences (100%). Two replicates were performed for each mutant. The replicative capacity of each mutant, measured as a percentage of that of wild type, was normalized using a square-root transformation. Differences between viral strains in their replication capacity were tested by fitting a mixed effects model using maximum likelihood, which assumed that the replication capacity of each strain was a fixed effect, and between-replicate error was a random effect.
Inocula of 3000 TCID50 of each virus were used to infect 3 × 106 MT-2 cells (multiplicity of infection of 0.001) as previously described . Cultures were set up in 24-well plates, with five replicates per mutant. p24 growth kinetics was analysed by fitting a linear model to the log-transformed p24 data from the five replicates for each mutant by maximum likelihood. For a given mutant, the growth rate of p24 was assumed to be the same between replicates, but the initial level of p24 was allowed to vary between replicates according to a normal distribution. Missing values were assumed to be missing at random. This model was found to give a good fit to the data.
Growth competition assays were performed in MT-2 cells in the absence of drug as described elsewhere . Infections were initiated with unequal amounts of two competing virus variants; typically 20 and 80%, based on virus infectivity titrations. MT-2 cells were also separately infected with mutants in the absence of wild-type virus to evaluate the potential for true genetic reversion over the course of the experiment. Aliquots containing cells were taken at days 1, 4, 7, 14 and 21, and total infected cell DNA was extracted (Qiamp DNA blood Mini kit; Qiagen, Hilden, Germany). DNA was PCR-amplified in triplicate including the 3′-end of Gag and the whole protease coding region. PCR products from the same sample were pooled, cloned into pGEM T-Easy vector System II (Promega, Madison, WI, USA) and transformed into Escherichia coli competent cells. Twenty recombinant plasmids were used per timepoint to determine the proportion of mutant versus wild-type virus in each competitive culture timepoint. Genotyping was performed using the d-Rhodamine Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems, Warrington, UK) and subsequent electrophoresis in an automated sequencer (either an ABI PRISM 310 or 377; Applied Biosystems).
Protein maturation analysis
Virion-associated protease activity was determined by Western blot analysis using Gag-specific antibodies. After budding from the cell, virion maturation is completed through the cleavage of the Gag polyprotein by the viral protease into the structural proteins, matrix, capsid and nucleocapsid. For sample preparation, the viruses were collected by centrifugation and the pellet lysed directly in loading buffer and analysed by sodium-dodecylsulphate–polyacrylamide gel electrophoresis. The Gag cleavage products were visualized on the Western blots using antibodies against capsid protein, p24. Defects in protease activity result in the accumulation of p41, a partly cleaved precursor, which was quantified and expressed as a percentage of Gag intermediates and compared with the reference virus (HIVNL4−3).
All statistical analyses were performed in the R language (http://www.r-project.org).
Production of recombinant viruses
Each recombinant virus vector produced viruses, except for the protease single mutant I50V, which had been constructed by site-directed mutagenesis and confirmed by DNA sequencing (Fig. 1a). The recombinant virus M5r+ produced virus after transfection, but did not replicate to a titer sufficient to set up competitive cultures. We also detected a genotypic change in M2r− with respect to its parental virus M1c+: M1c+ had only the protease substitution L10F in the presence of the (p1/p6) Gag cleavage site mutation L449F, whereas M2r− had the mutations L10F/I84V in the absence of any Gag cleavage site mutation (Table 1 and Fig. 2). This difference seems likely to be caused by the selection of a different virus in the process of molecular cloning from a culture that contained a mixed population of different mutants as well as the wild type.
Phenotypic susceptibility of protease mutants
The drug susceptibility of the virus isolates selected in vitro during passage in increasing concentrations of amprenavir (`c’ series), and the recombinant viruses derived from them (`r’ series), were assayed against a panel of PI (Table 1). Resistance to amprenavir increased as mutations accumulated. The acquisition of mutations M46I, I50V and subsequently I47V by an L10F/I84V virus increased the IC50 from 8.4-fold (M2c+) to 48-fold (M5c+). Notably, amprenavir-resistant viruses also had a parallel loss of susceptibility to lopinavir. The IC50 of this drug increased from 7.7-fold in M2c+ to 31-fold in M5c+. We also observed moderate cross-resistance to ritonavir (five- to sevenfold-increase in IC50 for M4c+, M4r+ and M5c+) and weak cross-resistance to nelfinavir, predominantly as a result of the low level of nelfinavir resistance (IC50 4.1) in the M5c+ strain. All drug-resistant mutants remained susceptible to saquinavir and indinavir. It should be noted that the only difference between M2r−and M2r+, the cleavage site mutation L449F, increases PI resistance.
Viral replication capacity
The viral replication capacity of amprenavir-selected viruses in the absence of drugs was measured using several approaches, including a single-cycle replication capacity assay, replication kinetics as measured by the production of Gag p24 antigen in parallel cultures, as well as competitive virus replication. Based on the single-cycle replication capacity assay amprenavir- selected virus isolates and their cloned recombinant viruses exhibited less than 10% of the replication capacity of the wild-type reference strain (HIV-1NL4−3) in the absence of drug, with the exception of M1c+ (Table 1). There were significant differences between the replication capacities of the different amprenavir-selected strains. The replication capacity of M1c+ was significantly greater (35.9%; linear mixed-effects model, P < 0.0001) and the replication capacity of M4r+ was significantly lower (2%;P = 0.03) than the rest of the strains (M2c+, M4c+, M5c+, M2r−, M2r+: range 4–6.3%) in the absence of drug (Table 1). Replication capacity data was also obtained in the presence of different amprenavir concentrations by a different analysis of the drug susceptibility assays. Although the viruses with higher amprenavir IC50 had a relative replication advantage over wild type in higher amprenavir concentrations, the same rank order of relative replication capacity of these mutants was seen in lower amprenavir concentrations as in the absence of drug (Fig. 3).
Gag polypeptide processing of viruses produced in the single-cycle assay was evaluated by determining the accumulation of a p41 (MA/CA) intermediate that precedes the production of the fully processed p24 CA (Table 1). The p41 in virions increased from 7% in M1c+ to 21% in M5c+, relative to the reference virus. The accumulation of the incompletely processed Gag p41 peptide correlated with the increase in IC50 to amprenavir and lopinavir (Spearman's rank correlation coefficient 0.87, P = 0.02), although there was no significant difference in replication capacity, as measured using the single-cycle assay.
The replication capacity of these mutants in the absence of drug was also determined by the kinetics of Gag p24 antigen production (Fig. 1b). The rate of increase of p24 did not differ significantly between HIV-1NL4−3 and IIIB− (data not shown) and between the mutants M2r− and M2r+. Overall, the rate of p24 increase for M2r− and M2r+ was approximately 75% that of wild type, and the rate of p24 increase for M4r+ was approximately 50% that of wild type. This assay suggested the order (in decreasing replication capacity): IIIB− > (M2r−, M2r+) > M4r+, consistent with the results of the single-cycle assay.
A rank order of replication capacity among the recombinant viruses in the absence of drug was also determined using competition cultures. Each mutant virus was competed against a wild-type reference virus (IIIB−) and each of the other mutant viruses. At least 20 molecular clones were sequenced to measure the relative frequency of wild type versus mutant virus at days 1, 7, 14 and 21 of each competitive culture. We found a progressive decrease in the relative replication capacity of recombinant virus clones, in the absence of drug, corresponding to viral isolates selected in vitro at increasing concentrations of amprenavir (Fig. 4). The replication of M4r+ was the most impaired based on competition with wild-type and M2r+ viruses. M2r+ replicated less well than the wild-type or M2r− viruses. These competition results in the absence of drug were consistent with the single cycle assay data and the p24 growth kinetics; however, in the competitive cultures the M2r− had a higher replication capacity than M2r+.
Site-directed mutagenesis was performed to introduce the mutation I50V in pNL4-3. Recovered clones were sequenced and the presence of the I50V as a unique change in Gag and protease was confirmed. Three independent clones were transfected three different times in MT-2 cells in the presence of adequate positive controls. The cultures never produced detectable amounts of p24 antigen. In an independent laboratory (ViroLogic) similar results were obtained with a totally independent procedure using different site-directed protease 50 mutant clones (N. Parkin and R. Ziermann, unpublished observation). The data obtained from all of the experiments allowed us to define the following replication capacity rank: wild type > M1c+ > M2r− > M2r+,c+ > M4r+,c+ > M5c+ > I50V.
We evaluated the PI susceptibility and replication capacity of a series of HIV-1 variants that were progressively selected in vitro by the serial passage of HIV-1IIIB in CEM-SS cells in the presence of increasing concentrations of amprenavir . After seven passages, a variant with the PRL10F/I84V mutation dominated the population. Two passages later the dominant variant contained L10F/M46I/I47V/I50V mutations . Viruses with a similar spectrum of mutations have been reported by others after selection in vitro by amprenavir , and have been observed in viruses from patients failing amprenavir treatment [7,17].
In the present study, we used this series of amprenavir-selected viruses, as well as recombinant virus clones derived from them, to characterize the PI drug resistance/cross-resistance patterns and the replication capacity of amprenavir-selected mutants.
The accumulation of two mutations in the protease (L10F/I84V), along with the gag L449F cleavage site mutation, resulted in moderate levels of resistance to amprenavir and lopinavir. Transitioning from the double protease mutant (L10F/I84V) to the triple (L10F/M46I/I50V) or the quadruple (L10F/M46I/I47V/I50V) protease mutants increased lopinavir cross-resistance from 7.7- to a 19- and 31-fold change in IC50, respectively. This observation is discordant with the current genotypic rule, which specifies that multiple protease mutations (six or more) are needed to confer at least 10-fold resistance to lopinavir . Similar drug-susceptibility results were obtained for drug-selected virus populations (M1c+, M2c+, M4c+ and M5c+) and molecular clones derived from each population (M2r−, M2r+ and M4r+). Intermediate levels (< 10-fold) of cross-resistance to ritonavir and nelfinavir were also observed after the accumulation of three mutations in the protease. Less cross-resistance was observed to saquinavir and indinavir, in agreement with previous studies . These data warrant consideration, on the basis of resistance testing, of saquinavir/ritonavir and indinavir/ritonavir, as well as lopinavir/ritonavir, as components of salvage regimens for patients who have failed a previous regimen containing amprenavir with virus genotypes such as those studied here. However, the current data identifying new HIV mutation patterns conferring cross-resistance to lopinavir point out the need for further studies correlating genotype and phenotype before relying only on genotyping for prediction of cross-resistance to lopinavir in clinical specimens.
The data presented here are consistent with a recent report that 59% of the individuals who had taken amprenavir-based regimens (ACTG 347 ) and switched to a four-drug regimen (indinavir, nevirapine, stavudine and lamivudine) achieved subsequent durable virological suppression after one year of follow-up . Whether or not the mutation patterns associated with low replication capacity in this study were frequent in protocol ACTG347 and resulted in a better response to a salvage treatment including indinavir requires further investigation.
The relative replicative capacity of the amprenavir-selected mutants in the absence of drug was examined using several different methods: a single-cycle replication assay, which measures luciferase production or the accumulation of partly processed Gag (p41), replication kinetics, and replication competition. The single-cycle replication capacity assay revealed large replication impairment (> 90%) for all mutants tested except for M1c+ (L10F), which had a replication capacity of 36% with respect to the wild-type virus. This assay proved to be a rapid and reproducible method for measuring replication capacity, which can also be performed in the presence of drug. Although the viruses grew better in the presence of high concentrations of amprenavir as more mutations accumulated, the same relative ranking of replication capacity (diminishing replication capacity as mutations accumulated) was seen in either low concentrations of amprenavir or in the absence of drug. In this assay the accumulation of partly processed Gag p41 (i.e. impaired protease activity), increased with the number of protease mutations (M5c+ > M4c+ > M2c+ > M1c+ > wild type). The very low replication capacity values in the single-cycle assay contrasted with the fairly good replication in other culture conditions (such as in the p24 production assay in Fig. 1). The clinical relevance of the single-cycle assay is not yet established and it should not be assumed that viruses with very low replication capacity in that assay will not replicate in vivo.
Subtle differences in replication capacity were more precisely defined by repeated testing in culture-based assays. A clear ranking of decreasing replication capacity was obtained by measuring p24 antigen production after transfection with normalized amounts of DNA. p24 antigen production kinetics after infection with normalized TCID50 confirmed the spectrum of replication capacity from wild-type to M4r+, and was in agreement with previous observations . Competitive cultures, despite being more laborious and time consuming, established an accurate ranking of replication capacity in the absence of drug (wild type > M2r− > M2r+ > M4r+) by allowing direct comparison of each mutant against the wild type or other mutants. Protease mutations were neither gained nor lost during the course of the assay. Recombinant virus clones (`r’ series) were used when measuring replication capacity in culture-based assays, to eliminate the outgrowth of minor variants present in the pool of drug-selected isolates (`c’ series) that might complicate the competition in the absence of drug (data not shown). Such outgrowth of minor wild-type virus has been observed in vivo during treatment interruptions in HIV-1-infected patients with treatment failure [13,22]. The use of virus infectivity titrations, rather than virion particle counts, to measure the inocula of the two variants in the competition cultures probably underestimated replication capacity differences because a greater number of the less replicative virions would have been used. Nevertheless, differences in replication capacity were observed despite this more stringent methodology.
The PI susceptibility of the I50V mutant in the HIVNL4−3 background could not be tested because this virus did not replicate (Fig. 1a). Other authors have shown that HIV-1IIIB harbouring the single substitution I50V replicates and remains susceptible to amprenavir in culture . The inability of the I50V virus to replicate in this system may relate to the different HIV-1 genetic background or the cells used . We speculate that a protease I50V single mutant in the HIVNL4−3 background is unable to replicate.
In this study, the progressive accumulation of protease mutations did not re-establish viral replication capacity. This is in contrast with earlier studies of PI-selected mutants, in which later mutations were seen to compensate for the replication capacity impairment associated with the first mutation [10,24]. However, the shift from a L10F/I84V viral population to viral variants containing several mutations, including I50V, suggests that protease L10F, M46I and the Gag L449F may work as compensatory mutations for protease I50V. The accumulation of protease I47V further increased amprenavir and lopinavir resistance but reduced viral replication capacity. Biochemical studies have also shown that the addition of M46I and I47V improves the replication of I50V mutant viruses . Protease M46I has been suggested to play a compensatory role in resistance development .
Notably, the only difference between M2r− and M2r+ was the accumulation of Gag L449 in the P1’ residue of the p1/p6 cleavage site. This cleavage site mutation decreased amprenavir and lopinavir susceptibility of the L10F/I84V virus from 3.7- and 2.8-fold to 4.6- and 6.2-fold, respectively. This Gag substitution had previously been reported to reduce susceptibility to the PI lopinavir or BILA 1906 BS, but not to saquinavir or ritonavir [27–29]. This cleavage site mutation incompletely restored virus replication capacity of an amprenavir-selected isolate .
Amprenavir-selected viruses with fewer than five mutations can exhibit a greater than 10-fold increase in IC50 to amprenavir and also to lopinavir. This suggests that a given number of resistance mutations in protease may not be an adequate criterion for lopinavir resistance, and suggests that the use of lopinavir/ritonavir, after an amprenavir-containing regimen fails, may be better guided by phenotypic resistance testing. This approach, however, will be limited by the possibility that some clinical specimens may not be able to be phenotyped. The single-cycle growth assay for replication capacity yielded similar results to competitive cultures in the absence of drug, although small relative differences were more apparent with competitive cultures and absolute levels of replication were higher. These data suggest further study of cross-resistance to lopinavir in specimens from patients failing amprenavir, and provide initial data on concordance between a rapid single-round growth assay and a slower, more laborious measure of replication capacity.
Amprenavir-selected viruses were kindly provided by Dr R. Byrn (Vertex Pharmaceuticals, Inc.).
1. Palella FJ, Delaney KM, Moorman AC. et al
. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators.
N Engl J Med 1998, 338: 853–860.
2. Erickson JW, Gulnik SV, Markowitz M. Protease inhibitors. resistance, cross-resistance, fitness and the choice of initial and salvage therapies.
AIDS 1999, 13 (Suppl. A) : S189–S204.
3. Boden D, Markowitz M. Resistance to human immunodeficiency virus type 1 protease inhibitors. Antimicrob Agents Chemother 1998, 42: 2775–2783.
4. St Clair MH, Millard J, Rooney J. et al
. In vitro antiviral activity of 141W94 (VX-478) in combination with other antiretroviral agents. Antiviral Res 1996, 29: 53–56.
5. Kim EE, Baker CT, Dwyer MD. et al
. Crystal structure of HIV-1 protease in complex with VX-478, a potent and orally bioavailable inhibitor of the enzyme. J Am Chem Soc 1995, 117: 1181–1182.
6. Partaledis JA, Yamaguchi K, Tisdale M. et al
. In vitro selection and characterization of human immunodeficiency virus type 1 (HIV-1) isolates with reduced sensitivity to hydroxyethylamino sulfonamide inhibitors of HIV-1 aspartyl protease. J Virol 1995, 69: 5228–5235.
7. De Pasquale MP, Murphy R, Kuritzkes D, et al
. Resistance during early virological rebound on amprenavir plus zidovudine plus lamivudine triple therapy or amprenavir monotherapy in ACTG protocol 347.
In:2nd International Workshop on HIV Drug Resistance and Treatment Strategies
. Lake Maggiore, Italy, 1998 [Abstract 71].
8. Tisdale M, Myers RE, Maschera B, Parry NR, Oliver NM, Blair ED. Cross-resistance analysis of human immunodeficiency virus type 1 variants individually selected for resistance to five different protease inhibitors. Antimicrob Agents Chemother 1995, 39: 1704–1710.
9. Martinez-Picado J, Sutton L, De Pasquale MP, Savara AV, D'Aquila RT. Human immunodeficiency virus type 1 cloning vectors for antiretroviral resistance testing. J Clin Microbiol 1999, 37: 2943–2951.
10. Martinez-Picado J, Savara A, Sutton L, D'Aquila RT. Replicative fitness of protease inhibitor-resistant mutants of human immunodeficiency virus type 1. J Virol 1999, 73: 3744–3752.
11. Japour AJ, Mayers DL, Johnson VA. et al
. Standardized peripheral blood mononuclear cell culture assay for determination of drug susceptibilities of clinical human immunodeficiency virus type 1 isolates. The RV-43 Study Group, the AIDS Clinical Trials Group Virology Committee Resistance Working Group.
Antimicrob Agents Chemother 1993, 37: 1095–1101.
12. Petropoulos CJ, Parkin NT, Limoli KL. et al
. A novel phenotypic drug susceptibility assay for human immunodeficiency virus type 1. Antimicrob Agents Chemother 2000, 44: 920–928.
13. Deeks SG, Wrin T, Liegler T. et al
. Virologic and immunologic consequences of discontinuing combination antiretroviral-drug therapy in HIV-infected patients with detectable viremia. N Engl J Med 2001, 344: 472–480.
14. Gamarnik A, Wrin T, Ziermann R, et al
. Drug resistance is associated with impaired protease and reverse transcriptase function and reduced replication capacity: characterization of recombinant viruses derived from 200 HIV-1 infected patients.
In:4th International Workshop on HIV Drug Resistance and Treatment Strategies
. Sitges, Spain, 2000 [Abstract 116].
15. Martinez-Picado J, Savara AV, Shi L, Sutton L, D'Aquila RT. Fitness of human immunodeficiency virus type 1 protease inhibitor-selected single mutants. Virology 2000, 275: 318–322.
16. Hellmann N, Johnson P, Petropoulos C. Validation of the performance characteristics of a novel, rapid phenotypic drug susceptibility assay, PhenoSense HIV.
In:39th Interscience Conference on Antimicrobial Agents and Chemotherapy
. San Francisco, 1999 [Abstract 418].
17. Schmidt B, Korn K, Moschik B, Paatz C, Uberla K, Walter H. Low level of cross-resistance to amprenavir (141W94) in samples from patients pretreated with other protease inhibitors. Antimicrob Agents Chemother 2000, 44: 3213–3216.
18. Kempf D, Brun S, Rode R, et al
. Identification of clinically relevant phenotypic and genotypic breakpoints for ABT-378/r in multiple PI-experienced, NNRTI-naive patients.
In:4th International Workshop on HIV Drug Resistance and Treatment Strategies
. Sitges, Spain, 2000 [Abstract 89].
19. Murphy RL, Gulick RM, DeGruttola V. et al
. Treatment with amprenavir alone or amprenavir with zidovudine and lamivudine in adults with human immunodeficiency virus infection. AIDS Clinical Trials Group 347 Study Team.
J Infect Dis 1999, 179: 808–816.
20. Gulick RM, Smeaton LM, D'Aquila RT. et al
. Indinavir, nevirapine, stavudine, and lamivudine for human immunodeficiency virus-infected, amprenavir-experienced subjects. AIDS Clinical Trials Group protocol 373.
J Infect Dis 2001, 183: 715–721.
21. Alford JL, McQuaid TJ, Partadelis JA, Markland W, Byrn RA. Amprenavir resistant mutants show reduced viral fitness: correlation of protease mutations, cleavage site mutations, enzyme kinetics, viral polyprotein processing and viral growth kinetics.
In:3rd International Workshop on HIV Drug Resistance and Treatment Strategies
. San Diego, USA, 1999 [Abstract 44].
22. Miller V, Sabim C, Hertogs K. et al
. Treatment interruptions in HIV-1 infected patients with treatment failure: virological and immunological effects. AIDS 2000, 14: 2857–2867.
23. Rose RE, Gong Y-F, Greytok JA. et al
. HIV-1 viral background plays a major role in development of resistance to protease inhibitors. Proc Natl Acad Sci U S A 1996, 93: 1648–1653.
24. Nijhuis M, Schuurman R, de Jong D. et al
. Increased fitness of drug resistant HIV-1 protease as a result of acquisition of compensatory mutations during suboptimal therapy. AIDS 1999, 13: 2349–2359.
25. Pazhanisamy S, Stuver CM, Cullinan AB, Margolin N, Rao BG, Livingston DJ. Kinetic characterization of human immunodeficiency virus type-1 protease-resistant variants. J Biol Chem 1996, 271: 17979–17985.
26. Ho DD, Toyoshima T, Mo H. et al
. Characterization of HIV-1 variants with increased resistance to a C2-symmetric protease inhibitor. J Virol 1994, 68: 2016–2020.
27. Carrillo A, Stewart KD, Sham HL. et al
. In vitro selection and characterization of human immunodeficiency virus type 1 variants with increased resistance to ABT-378, a novel protease inhibitor. J Virol 1998, 72: 7532–7541.
28. Doyon L, Croteau G, Thibeault D, Poulin F, Pilote L, Lamarre D. Second locus involved in HIV-1 resistance to protease inhibitors. J Virol 1996, 70: 3763–3769.
29. Mammano F, Petit C, Clavel F. Resistance-associated loss of viral fitness in human immunodeficiency virus type 1: phenotypic analysis of protease and gag coevolution in protease inhibitor-treated patients. J Virol 1998, 72: 7632–7637.
30. Robinson LH, Myers RE, Snowden BW, Tisdale M, Blair ED. HIV type 1 protease cleavage site mutations and viral fitness: implications for drug susceptibility phenotyping assays. AIDS Res Hum Retroviruses 2000, 16: 1149–1156.