| |
TMC-125: new NNRTI- resistance and patient study results
|
| |
| |
TMC125 Displays a High Genetic Barrier to the Development of Resistance: Evidence from In Vitro Selection Experiments
Journal of Virology, October 2005, p. 12773-12782, Vol. 79, No. 20
Johan Vingerhoets,1 Hilde Azijn,1 Els Fransen,1 Inky De Baere,1 Liesbet Smeulders,1 Dirk Jochmans,1 Koen Andries,2 Rudi Pauwels,1, and Marie-Pierre de Bethune1*
Tibotec, Mechelen, Belgium,1 and Johnson & Johnson Pharmaceutical Research & Development, Beerse, Belgium2
TMC-125, new NNRTI for NRTI Resistance
Study data in patients reported in Nov 2005 at HIV Conference EACS in Dublin
TMC-125: efficacy & safety in highly resistant patients (11/18/05)
TMC-125, new NNRTI for Resistance (11/18/05)
ABSTRACT
TMC125 is a potent new investigational nonnucleoside reverse transcriptase inhibitor (NNRTI) that is active against human immunodeficiency virus type 1 (HIV-1) with resistance to currently licensed NNRTIs. Sequential passage experiments with both wild-type virus and NNRTI-resistant virus were performed to identify mutations selected by TMC125 in vitro. In addition to "classic" selection experiments at a low multiplicity of infection (MOI) with increasing concentrations of inhibitors, experiments at a high MOI with fixed concentrations of inhibitors were performed to ensure a standardized comparison between TMC125 and current NNRTIs.
Both low- and high-MOI experiments demonstrated that the development of resistance to TMC125 required multiple mutations which frequently conferred cross-resistance to efavirenz and nevirapine. In high-MOI experiments, 1 μM TMC125 completely inhibited the breakthrough of resistant virus from wild-type and NNRTI-resistant HIV-1, in contrast to efavirenz and nevirapine. Furthermore, breakthrough of virus from site-directed mutant (SDM) SDM-K103N/Y181C occurred at the same time or later with TMC125 as breakthrough from wild-type HIV-1 with efavirenz or nevirapine.
The selection experiments identified mutations selected by TMC125 that included known NNRTI-associated mutations L100I, Y181C, G190E, M230L, and Y318F and the novel mutations V179I and V179F. Testing the antiviral activity of TMC125 against a panel of SDMs indicated that the impact of these individual mutations on resistance was highly dependent upon the presence and identity of coexisting mutations. These results demonstrate that TMC125 has a unique profile of activity against NNRTI-resistant virus and possesses a high genetic barrier to the development of resistance in vitro.
INTRODUCTION
Nonnucleoside reverse transcriptase inhibitors (NNRTIs) are common components of therapy for antiretroviral-naive human immunodeficiency virus type 1 (HIV-1)-infected patients. They have synergistic activity with nucleoside reverse transcriptase inhibitors (NRTIs) (7, 9, 35), high potency that is comparable to that of protease inhibitors (PIs) (27, 29, 30), proven durability (K. Tashima, S. Staszewski, M. Nelson, A. Rachlis, D. Skiest, R. Stryker, L. Bessen, V. Wirtz, S. Overfield, and D. Sahner, Abstr. XV Int. AIDS Conf., abstr. TuPeB4547, 2004), and are associated with fewer long-term metabolic complications than most PIs (32). However, the utility of the currently licensed NNRTIs is limited by the relatively easy selection of single mutations that alone can confer large reductions in susceptibility to the drug: i.e., they are characterized by a low genetic barrier to the development of resistance. This disadvantage is compounded by cross-resistance across the class (4, 11), which makes sequential therapy with the currently licensed NNRTIs clinically inappropriate (3). Resistance to NNRTIs among HIV-1-positive patients receiving antiretroviral therapy is widespread, and the prevalence of transmitted NNRTI resistance is also increasing in some populations (14, 17, 26, 33). Therefore, there is a growing clinical need for new NNRTIs with more robust resistance profiles and with activity against HIV strains resistant to the currently licensed NNRTIs.
TMC125 is a new investigational diarylpyrimidine NNRTI that was selected in the course of lead compound optimization of agents specifically chosen for their activity against NNRTI-resistant HIV-1 (2). It is highly active against wild-type HIV-1, with a 50% effective concentration (EC50) of 1.4 nM, and it retains activity (EC50 < 100 nM) against 97% of a large number of NNRTI-resistant clinical isolates (2). Structure determination by crystallization and modeling of the interactions between TMC125 and reverse transcriptase (RT) indicate multiple binding conformations which may account for this unique antiviral activity profile (10). In a 7-day clinical trial, TMC125 was well tolerated and demonstrated significant and rapid antiviral activity in patients with high levels of phenotypic NNRTI resistance (13).
An understanding of the mechanisms of development of resistance is essential in order to maximize the clinical benefit of new antiretroviral agents. In vitro selection experiments can be a useful indicator of the genetic barrier to the development of resistance and the resistance profile of a drug. They may therefore guide the selection of drug combinations and the effective sequencing of drugs in patients.
The objectives of this study were to characterize the resistance profile of TMC125 with respect to the specific mutations and patterns of mutations selected in vitro by the drug and to determine their contribution to resistance to TMC125 and currently licensed NNRTIs. "Classic" selection experiments were performed with a low multiplicity of infection (MOI; expressed as 50% cell culture infectious dose [CCID50]/cell) and increasing NNRTI concentrations to optimize the escape of resistant virus populations. In addition, we developed an experimental system that used a high MOI and fixed concentrations of the selecting NNRTI (H. Azijn, M.-P. de Bethune, K. Hertogs, K. Andries, P. Janssen, and R. Pauwels, Abstr. 4th Int. Workshop HIV Drug Resist., abstr. 86, 1995). This system permits standardization and thus allows comparison between NNRTIs to provide meaningful insight into the genetic barrier to the development of resistance for the different inhibitors. It also provides resistance data that complement those from low-MOI experiments, and combined these data may be more predictive of in vivo resistance profiles. Selection was performed both from wild-type HIV-1 strain LAI and from strains with common NNRTI resistance-associated mutations. Single genomes were analyzed using limiting dilution and clonal sequencing. To investigate the significance of selected mutations, a panel of site-directed mutants (SDMs) was tested for sensitivity to TMC125 and efavirenz.
RESULTS
In vitro selection starting from wild-type virus and using a high MOI.
Selection experiments using a high MOI and fixed concentrations of inhibitors can be standardized to allow a more accurate comparison of the different NNRTIs with TMC125 in vitro. Figure 1 shows the comparison of the time taken for virus breakthrough to occur in cultures infected with wild-type HIV-1 LAI and exposed to different concentrations of TMC125, efavirenz, or nevirapine (one of two experiments is shown). Complete virus breakthrough occurred rapidly with 1.0 μM nevirapine (13x EC50) or with 1.0 μM efavirenz (1,000x EC50), and similar results were obtained with higher concentrations (10.0 μM for nevirapine and 5.0 μM for efavirenz) (data not shown). In contrast, no virus breakthrough was observed in cultures treated with 1.0 μM TMC125 (710x EC50). Virus breakthrough that occurred with lower concentrations of TMC125 (Table 1 and Fig. 1) was delayed compared with that seen with 1.0 μM nevirapine or efavirenz. A comparable delay in virus breakthrough was observed when in vitro selection experiments were conducted with two wild-type recombinant clinical isolates (data not shown).
The virus strains selected by TMC125 harbored Y181C alone or in combination with V179V/I or L100L/I and T386A (Table 1), only the latter of which had high-level resistance to TMC125 (FC = 78) and efavirenz (FC = 210). All viruses selected by TMC125 were resistant to nevirapine (FC 80).
In vitro selection starting from wild-type virus and using a low MOI.
Cells were infected with wild-type HIV-1 LAI at low MOI and cultured in the presence of progressively increasing concentrations of TMC125 in two separate experiments. Table 2 details the genotype and phenotype of the virus populations that escaped at various time points. In the first experiment, virus escaped after 7 days at 10 nM TMC125 and by day 35 was able to replicate in 200 nM TMC125. The virus strains selected initially acquired E40K and Y181C mutations, followed by M230I (by day 35) and V179F (by day 39). Virus with all four mutations was less susceptible to TMC125 (FC = 150) and nevirapine (FC > 130) but remained susceptible to efavirenz (FC = 3.0). With increasing concentrations, FC values for both TMC125 and efavirenz increased.
A different pattern of mutations was selected by TMC125 in the second experiment (Table 2). Y181C and T386A were initially selected by day 15 at 40 nM TMC125 and persisted in all subsequently selected strains. The additional acquisition of G190E by day 35 at 1.0 μM TMC125 was associated with a decrease in susceptibility to TMC125 (FC = 160). FC values for efavirenz remained lower than those for TMC125 until the acquisition of G190E. All selected viruses were resistant to nevirapine (FC > 130).
As a control, a low-MOI selection experiment was performed from wild-type HIV-1 LAI with efavirenz (data not shown). The selection of L100I alone conferred resistance to efavirenz (FC = 32), but not to TMC125 (FC = 2), and very-high-level resistance to efavirenz (FC > 10,000) developed with the additional accumulation of G190E, V276I, and T386A. The presence of G190E correlated with the development of high-level resistance to efavirenz. This is similar to when TMC125 was used as the selecting inhibitor, where G190E in the presence of other mutations was associated with resistance.
Highly resistant virus strains selected in the low-MOI experiments were subjected to clonal analysis. Table 3 details the large number of different clones selected by TMC125 from wild-type virus by day 59 in the second experiment. All clones contained Y181C (except one with Y181S), G190E, and T386A.
In vitro selection starting from mutant virus and using a high MOI.
When cells infected with efavirenz- and nevirapine-resistant SDM-K103N were cultured in the presence of 1.0 μM TMC125, there was again no virus breakthrough in two separate experiments (Fig. 2A). At 40 and 200 nM TMC125, 100% virus breakthrough occurred by 11 to 13 days and by 28 days, respectively, or not at all. Virus selected from SDM-K103N by TMC125 acquired the additional mutations Y181C/Y with L100L/I and Y181C with M230M/I (Table 1). The FC for TMC125 remained <10 in all selected virus populations, in comparison with FCs of 31 for efavirenz and nevirapine.
There was no virus breakthrough of SDM-Y181C when infected cells were cultured with 1.0 μM TMC125 in two separate experiments (Fig. 2B). At 40 and 200 nM TMC125, 100% virus breakthrough occurred by 7 to 9 days and by 11 to 13 days, respectively. Virus acquiring V179I or V179F had an FC of 16 or 32, respectively, for TMC125 but remained sensitive to efavirenz.
TMC125 was active against SDM-K103N/Y181C (FC = 4), but selection of replicating populations occurred at all concentrations of TMC125 (Fig. 2C). Additional mutations conferring resistance to TMC125 were L100I/E194G, with or without V189I (Table 1).
In vitro selection starting from mutant virus and using a low MOI.
Low-MOI selection experiments using TMC125 were also performed starting with SDM-K103N (Table 4) and with HIV-1 strain r9599 (V179I/YI81C) (Table 5). The time scale of selection was similar to that observed when starting with wild-type virus. Virus populations escaping from SDM-K103N with high-level resistance to TMC125 had a minimum of four additional mutations, of which L100I, E138G, and Y181C occurred most frequently. With r9599, the selection of L234I and Y318F in combination with the starting V179I/YI81C was sufficient to confer high-level resistance to TMC125.
Clonal analysis of highly resistant virus populations selected from SDM-K103N and r9599 by TMC125 is detailed in Tables 4 and 5, respectively. Starting from SDM-K103N, the mutations L100I, E138G, V179I, and Y181C were commonly observed together. Starting from r9599 (V179I/Y181C), all clones except one contained L234I and Y318F, and L100I or V108I were also observed in multiple clones.
Antiviral activity of TMC125 and efavirenz against SDMs.
A panel of SDMs was constructed to study the role of specific RT mutations selected by TMC125 and was tested for sensitivity to TMC125 and efavirenz (J. Vingerhoets, I. De Baere, H. Azijn, T. Van den Bulcke, P. McKenna, T. Pattery, R. Pauwels and M.-P. de Bethune, Abstr. 11th Conf. Retrovir. Opportun. Infect., abstr. 621, 2004) (Table 6). With the exception of M230L, and in contrast to efavirenz, an FC of >10 for TMC125 only occurred in the presence of multiple mutations, such as L100I/K103N/Y181C (FC = 43).
As seen in virus populations selected in high-MOI experiments, mutations at position 179 in combination with Y181C reduced the sensitivity of virus to TMC125, especially V179F (FC = 130), which was selected less frequently than V179I. Y181C and V179F appeared to have a synergistic effect on the decreased susceptibility of TMC125.
Y318F, which was selected from r9599 (V179I/Y181C) in the low-MOI experiments, was associated with modest FCs for TMC125 when in combination with either Y181C or K103N as an SDM, but the combination of all three mutations K103N/Y181C/Y318F increased the FC to 11.
TMC125 and efavirenz had good antiviral activity against SDM-L74V and SDM-L74V/G190E. This suggests that resistance to TMC125 conferred by G190E on many of the strains selected from wild-type HIV-1 depended upon the presence of at least one other mutation, especially Y181C, although many other mutations not known to be associated with NNRTI resistance were also frequently present with G190E (Tables 1 and 2). In a similar way, SDM-Y181C/L234I was sensitive to TMC125 (FC = 1.4), suggesting that if L234I contributed to the resistance of clones selected from r9599-the context in which it was identified-this depended upon the presence of V179I and/or Y318F.
SDM-M230L confirmed that mutations at this position, as observed in resistant strains selected both from wild-type HIV-1 (low-MOI experiment) and from SDM-K103N (high-MOI experiment), can confer resistance to both TMC125 (FC = 13) and efavirenz (FC = 15).
DISCUSSION
The experimental conditions in vitro may affect the selection of mutations in HIV-1 resistant to NNRTIs; accordingly the predictive value of in vitro data for the mutations and mutation patterns that will be selected in vivo by NNRTIs may be limited. The current study sought to maximize the clinical relevance of in vitro data by developing an additional and complementary experimental system that used a high MOI and a fixed concentration of RT inhibitor.
Several different factors may contribute to discrepancies between in vitro experiments and clinical observations and between different in vitro experiments. The limited predictive value of in vitro experiments is illustrated by the example of the investigational NNRTI capravirine: in vitro selection of V106A/F227L and K103T/V106A/L234I did not predict the substitutions which were selected in vivo at positions 101, 108, 190, and/or 188 (34; K. E. Potts, T. Fujiwara, A. Sato, J. Cao, R. L. Jackson, J. Isaacson, O. Maldonado, B. Atkinson, B. Wang, T. Nash-Alexander, and A. K. Patick, Abstr. 3rd Int. Workshop HIV Drug Resist. Treatment Strategies, abstr. 15, 1999; J. Hammond, R. Pesano, P. Hawley, and A. Patick, XIII Int. HIV Drug Resist. Workshop, abstr. 15, 2004). The specific selection conditions may influence the genotype of emerging resistant virus. Two pathways of resistance to efavirenz were described in one study by using different experimental conditions: L100I/V179D/Y181C and L100I/V108I (5, 34). Of these mutations, only L100I arose in our selection experiments with efavirenz and none of these in vitro pathways was predictive of the K103N mutation most frequently observed in patients failing efavirenz therapy (5).
Classical low-MOI in vitro selection experiments identify mutations that can confer resistance when viral escape through the steady accumulation of mutations is facilitated by suboptimal increasing concentrations of inhibitor. The present study sought to broaden this scope by developing a complementary experimental system that used a high MOI and a fixed concentration of RT inhibitor. A high-MOI experiment can be more readily standardized than a low-MOI experiment and therefore enables the in vitro performance of different NNRTIs to be compared. At the same time, it maximizes genetic diversity, thus mimicking the dynamics of quasispecies evolution in infected individuals, where 109 to 1010 new viruses are produced daily with a mutation rate of 10-4 to 10-5 (20, 22, 25). The fixed drug concentrations in the high-MOI experiments (as opposed to gradually increasing concentrations in the low-MOI experiments) also approximate more closely to the in vivo context of regular, fixed dosages. As mentioned above, the predictive value of in vitro selection experiments with NNRTIs is limited, and while a combination of experimental approaches represents a more comprehensive investigation of a drug's in vitro resistance profile, this may not correspond to the selection of resistance in vivo. Data from the ongoing TMC125 phase II clinical trials program are required for the analysis of clinical resistance.
The additional clinical consideration of preexisting mutations in both treatment-naive and treatment-experienced patients was addressed by selecting virus from NNRTI-resistant strains. Preexisting mutations can influence the development of resistance to a subsequent drug, although they are not necessarily associated with in vitro resistance to that drug. For example, virus with the Y181C mutation that is typically selected by nevirapine is fully susceptible to efavirenz in vitro, but attempts to sequence efavirenz in patients failing nevirapine-based antiretroviral therapy and showing this mutation have been largely unsuccessful (3, 8, 28). The existence of baseline Y181C may also influence the identity of mutations selected by efavirenz, its presence being associated with mutations at position 190 rather than L100I or K103N (1).
The high-MOI experiments confirmed that TMC125 has a resistance profile that is distinct from that of efavirenz and nevirapine and showed that it can greatly delay the selection of resistant mutants from both wild-type strains and SDMs carrying K103N and Y181C in comparison with these NNRTIs. Taken together with the appearance of multiple mutations in the resistant strains, as opposed to a single mutation for the current NNRTIs, this points to TMC125 having an increased genetic barrier to the development of resistance in vitro.
In the low-MOI experiments, several distinct pathways to the development of resistance to TMC125 were identified. The two key patterns selected from wild-type HIV-1 incorporated the NNRTI resistance-associated combinations Y181C/M230L and Y181C/G190E, together with various changes not previously associated with NNRTI resistance, such as L31I, E40K, A62V, L74V, V90I, V179F, and T386A. Experiments with SDMs suggested a contributory role for T386A in TMC125 resistance and demonstrated the importance of V179F, in combination with Y181C, in conferring resistance to TMC125 but not to efavirenz. The role of V179I is not clear, and further studies, including resistance monitoring in the current clinical trials with TMC125, will generate more clinically relevant information.
The existence of multiple pathways to high-level resistance to TMC125 is supported by a recent report of similar TMC125 selection experiments in which a broad range of clonal diversity was detailed (J. E. Brillant, K. Klumpp, S. Swallow, T. Mirzadegan, N. Cammack, and G. Heilek-Snyder, Abstr. XII Int. HIV Drug Resist. Workshop, abstr. 16, 2004). This study confirmed our findings with the selection of V179F, Y181C, and M230L by TMC125 and also noted substitutions at position 138, also seen in many of the clones selected from SDM-K103N.
Several mutations, selected from SDM-K103N and the recombinant virus r9599, are not well established NNRTI mutations. For example, residue E138 of the p51 subunit of RT forms part of the binding pocket of NNRTIs, and mutations at this position have been associated with resistance to experimental TSAO [2',5'-bis-O-(tert-butyldimethylsilyl)-3'-spiro-5'-(4"-amino-1",2"-oxathiole-2",2"-dioxide)-pyrimidine] RT inhibitors (6) but not to currently licensed NNRTIs (24). L234I is known to confer resistance to the investigational NNRTI capravirine (12), but its effect on other NNRTIs requires further study. Our observation of Y318F only in highly resistant virus in the presence of V179I/Y181C is consistent with the fact that Y318F is rarely reported in the absence of other major NNRTI resistance-associated mutations. This is a mutation in the NNRTI-binding pocket which on its own causes high-level resistance to delavirdine and in combination with other NNRTI resistance mutations contributes to resistance to efavirenz and nevirapine (16).
Notably, all pathways to the development of resistance to TMC125 in the low-MOI experiments comprised at least three mutations, including Y181C. Consequently TMC125-selected virus populations were all highly resistant to nevirapine. In contrast, the pathways to efavirenz and TMC125 resistance, respectively, were distinct. Virus that was highly resistant to TMC125 continued to accumulate a variety of further mutations not known to confer NNRTI resistance. An example is L74V, which is known to compensate for the severe impairment in the replicative capacity of virus with G190E (19). Whether any more of these mutations could have a role in modifying the replication rate of the virus is under investigation.
Taken together these data demonstrate that in vitro TMC125 can suppress the selection of resistant virus from an NNRTI-resistant virus population and that it has an increased genetic barrier to the development of resistance as compared with efavirenz or nevirapine. The degree of concordance between the mutations selected in high-MOI experiments and those selected in low-MOI experiments suggests that TMC125 has a consistent in vitro resistance profile relatively unaffected by the MOI. The mutations selected by TMC125 in vitro that have been identified as contributing to resistance to TMC125 (FC > 10) were L100I, V179F, V179I, Y181C, G190E, M230L, and Y318F. Other mutations including V189I, E194G, L234I, and T386A may have a role in resistance or may compensate for changes in the replication rate of virus strains and require further investigation. The prevalence of these mutations in the clinical sample database of NNRTI-resistant viruses (described in Materials and Methods) was variable (Table 6) but was less than 2% for all single mutations and combinations of mutations with an FC of >10 for TMC125.
The ability of TMC125 to escape the effects of drug resistance mutations has been explored in crystal structure studies (10, 31). TMC125 and other diarylpyrimidine analogues bind RT in a horseshoe-shaped configuration with two "wings" connected by a pyrimidine ring "body." A wing can rotate to accommodate certain substitutions of the surrounding RT residues without significantly affecting the efficacy of TMC125 because the other wing is able to make compensatory rotations to reoptimize its own binding without incurring a large energy penalty. The compact structure of these inhibitors also permits reorientation of the pyrimidine ring body to compensate for the repositioning of a wing. A range of single mutations in RT can confer resistance to the currently licensed NNRTIs. In contrast, the crystallography and structural data suggest that single mutations do not prohibit effective binding of TMC125 to RT but merely reduce the number of alternative binding configurations available, thereby giving it less scope to accommodate additional RT mutations. This supports our finding that resistance to TMC125 requires multiple mutations, and it may provide a rationale for particular resistance-associated combinations of mutations when these affect both wings of RT-bound TMC125, such as Y318F and L234I.
In conclusion, TMC125 is a potent and unique new investigational NNRTI characterized by a high genetic barrier to the development of resistance in vitro. It is active against viruses with either of the signature mutations for efavirenz and nevirapine and can prevent the emergence of resistant HIV-1 from them. We have demonstrated that TMC125 selects for a variety of NNRTI-associated mutations in vitro but that resistance generally requires the presence of at least two of these mutations. Several novel substitutions, most notably V179F, were identified in resistant strains and are the subject of further studies.
In vivo data from short-term trials with TMC125 in treatment-naive and treatment-experienced HIV-1-infected subjects with NNRTI resistance support the clinical relevance of these results (13, 15).
MATERIALS AND METHODS
SDMs and recombinant clinical isolates. Mutant RT coding sequences were constructed in a pGEM vector containing HIV-1 HXB2 protease (PR) and RT coding sequences, using the QuikChange site-directed mutagenesis kit (Stratagene) with high-performance liquid chromatography-purified primers. Plasmids were sequenced to verify mutations, and SDM virus stocks were generated by the recombination of mutant PR-RT sequences with a PR-RT-deleted HIV-1 HXB2 proviral clone (18). The NNRTI-resistant recombinant clinical isolate r9599, which harbors the mutations V179I and Y181C, was constructed by the Antivirogram method (18). MT4 cells were cotransfected with sample-derived viral PR and RT coding sequences and an HIV-1 HXB2-derived proviral clone with deletions in the PR and RT coding region (18).
In vitro selection experiments. MT4 or MT4-long terminal repeat (LTR)-enhanced green fluorescent protein (EGFP) cells were infected with HIV-1 wild-type strain LAI, r9599, or SDMs carrying the K103N mutation (SDM-K103N), the Y181C mutation (SDM-Y181C), or both (SDM-K103N/Y181C).
For the high-MOI experiments, infection was performed at an MOI of >1 CCID50/cell, corresponding to a total viral inoculum of 109 RNA copies, using 1.5 x 106 cells, in the presence of 40 nM, 200 nM, or 1.0 μM TMC125 or 1.0 μM efavirenz or nevirapine. Cultures were subcultivated in the presence of the same concentration of NNRTI every 3 to 4 days up to a maximum of 30 days and examined for signs of virus replication. At 100% virus breakthrough, the supernatant was collected and stored as a new virus strain.
For the low-MOI experiments, infection was performed at an MOI of 0.01 to 0.001 CCID50/cell in the presence of the NNRTI at 2x to 3x the EC50. This corresponded to an initial infection rate of 1%. Cultures were subcultivated in the presence of the same concentration of NNRTI every 3 to 4 days and microscopically examined for signs of virus replication. At 100% virus breakthrough, the supernatant was collected and stored as a new virus strain. New virus strains were used to infect fresh cultures in the presence of a higher concentration of NNRTI. The procedure was repeated up to a concentration of 15 μM, which represented >10,000 times the starting concentration for the wild-type strain and SDM-K103N or >900 times the starting concentration for r9599. In this way, there was a gradual selection of variants able to grow in the presence of high inhibitor concentrations.
Viral breakthrough was assessed by scoring microscopically the cytopathogenicity (cytopathic effect; MT4 and MT4-LTR-EGFP cells) and virus-induced fluorescence (MT4-LTR-EGFP cells only) in all experiments. A viral breakthrough of 100% was defined as microscopical evidence of extensive viral replication in all cell clusters (full cytopathic effect).
Virus stocks for further phenotypic characterization were produced from the new virus strains by infecting MT4 cells in the absence of inhibitor. Stocks were titrated and subsequently phenotyped. Virus for genotyping new strains was obtained either from the harvested supernatant or from the subsequently produced virus stock.
Genotyping and phenotyping. Viral RNA was extracted from culture supernatant or virus stock using the QIAamp viral RNA extraction kit (QIAGEN). cDNA encompassing part of polymerase was generated with Expand RT (Boehringer Mannheim), followed by amplification of the PR and RT regions by nested PCR as described elsewhere (18). PCR products were genotyped using the Big Dye terminator kit (Applied Biosystems), and sequences were resolved on a high-throughput automated ABI PRISM 3700 or 3730-XL DNA analyzer. Sequencing results are reported as amino acid changes compared with the wild-type HIV-1 (clone HXB2) reference sequence (21).
Some of the virus strains obtained after in vitro selection were subjected to limiting dilution and clonal analysis as described elsewhere (23; L. Michiels, E. Fordel, A. Scholliers, E. Gubbels, J. Vingerhoets, S. Bloor, A. Brophy, A. Takagi, K. Hertogs, F. Maldarelli, J. Coffin, B. Larder, and M. Van Houtte, Abstr. 4th Int. Workshop HIV Drug Resist. Treatment Strategies, abstr. 77, 2000). The principle of limiting dilution sequencing is based on a Poisson's distribution: if no more than one-third (30%) of replicate amplification reactions (PCR) are positive, there is an 80% likelihood that PCR products are the result of the amplification of a single sequence. To obtain the 30% positive PCRs, two consecutive dilutions of the cDNA were made. First, a threefold dilution series was made (10 replicates per cDNA and per dilution), and the dilution corresponding to 30% positive PCRs was determined. In the second step, 96 replicates of the PCR were performed at this determined dilution. Typically, 25 to 30 positive PCRs of the cDNA region encompassing PR and the first 400 codons of RT were obtained and sequenced.
Phenotyping was performed by the cell-based Antivirogram assay method, using either viability measurements from a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) colorimetric assay (28) or a fluorescent reporter-gene assay, as described by Hertogs et al. (18) with the modifications of Pauwels et al. (R. Pauwels, K. Hertogs, S. Kemp, S. Bloor, K. Van Acker, J. Hansen, W. De Beukeleer, C. Roelant, B. Larder, and P. Stoffels, Abstr. 2nd Int. Workshop HIV Drug Resist. Treatment Strategies, abstr. 51, 1998).
Fold changes in EC50 values (FCs) of virus to the test drug were calculated by dividing the mean or median EC50 for the tested virus by the mean or median EC50 for the wild-type virus tested in parallel, respectively.
Prevalence of mutations.
The prevalence of mutations was determined in a database containing 7,144 clinical samples submitted for resistance testing and found to be resistant to at least one of the current NNRTIs, defined as having an FC above the biological cutoff values of the Antivirogram (FC = 6, 8, and 10 for efavirenz, nevirapine, and delavirdine, respectively).
|
|
| |
| |
|
|
|
