Abstract
The aim of this study was to evaluate the presence of mutations and polymorphisms associated with drug resistance among HIV-1-infected patients in proviral DNA and viral RNA extracted from PBMCs and plasma, respectively, in 34 HIV-1-infected patients (11 naive and 23 receiving HAART). Additional drug resistance mutations were found in only one compartment in 14 of 23 treated patients. Mutations conferring resistance to an additional drug were found in plasma in only 7 of 23 patients. A greater number of differences was found in strains in patients infected for at least more than 9 years, compared to naive patients and patients for whom the time since the first diagnosis was lower (p < 0.02). This study confirms the usefulness of simultaneous testing of different compartments for assessing drug resistance in the pol region and suggests that the heterogeneity observed in different compartments might be increased with time of infection and treatment experience.
T
The selection of the best viral compartment for assessing resistance to both reverse transcriptase (RT) and protease HIV inhibitor genotyping is still a matter of debate. Testing for drug resistance mutations in proviral DNA extracted from peripheral blood mononuclear cells (PBMCs) or viral RNA extracted from plasma seems to provide generally concordant results. 2 Some authors recommend simultaneous genotypic determination in both compartments for a more complete determination of drug resistance mutations in the quasispecies viral repertoire. 3 However, contradictory results have been reported on the usefulness of each compartment, depending, for example, on whether naive or treated patients are studied. We hypothesized that the results may also depend on the time on infection and the treatment of infected patients. The aim of this study was to evaluate genotypic resistance and the presence of mutations and polymorphisms associated with drug resistance among HIV-1-infected patients in proviral DNA and viral RNA extracted from PBMCs and plasma, and to correlate the differences observed with the time of infection and treatment experience.
A total of 34 HIV-1-infected patients were studied, after signed consent. Of them, 11 patients were naive, infected with strains not harboring drug resistance mutations, and 23 were receiving HAART. Since the time of infection was unknown for most of the patients, the time of first diagnosis of HIV-1 was used as an estimation of the time of infection. Around 15 ml of blood with EDTA anticoagulant was extracted for plasma collection. PBMCs were also collected and stored at −70°C until use. Proviral DNA and/or RNA were extracted from PBMCs or plasma with commercial kits (QIAamp DNA Blood mini Kit and QIAamp Viral RNA Kit, QIAGEN, Germany). Around 1500 nt of the pol gene were amplified by nested polymerase chain reaction (PCR) or RT and nested PCR, as described previously.
4
PCR-purified fragments were sent to the Macrogen Sequencing Service (Macrogen, Korea) for sequencing. The electropherotypes of the sequences were visually inspected for the presence of ambiguous nucleotides (quasispecies). Nucleotide alignments and homology trees were performed using DNAMAN Version 5.2.2. (Lynnon Bio Soft, Canada). Sequences were submitted to a genotype sequence algorithm (Stanford University HIV Drug Re-sistance Database (
Table 1 shows a comparative analysis of polymorphisms and quasispecies found in proviral DNA and viral RNA obtained from PBMCs and plasma, respectively. All viral strains belong to HIV-1 subtype B. A high percentage of identity was found between the sequences of the two compartments, irrespective of the time of infection. Significant differences were found between the two compartments in the identification of mutations conferring drug resistance. Additional mutations were found, frequently as quasispecies, in both compartments, for 14 out of the 23 treated patients. Mutations conferring resistance to a drug in addition to those found in the other compartment were found in plasma in only 7 of 23 patients. In three of these seven patients, the mutation conferring resistance to an additional drug was M184V (Table 2). These three patients were receiving 3TC at the time of genotypic testing. HIV192 was infected with a virus that exhibited the highest number of differences in major mutations (seven) and circulated in the patient with the longest presumed time of infection (Table 2). Strain HIV158 showed mutations only in plasma associated with RT resistance, although the HAART regimen at the moment of testing did not include such drug inhibitors. The patient infected with this strain previously experienced multiple virological failures and several drug regimens before this analysis.
Nucleotide identity between proviral DNA and viral RNA sequence.
Drug resistance differences found only in proviral DNA from PBMCs (D) or viral RNA from plasma (R). Major mutations conferring resistance to an additional drug than the one identified in the other compartment are shown in bold.
Number of total mutations or polymorphisms observed in only one compartment.
Number of ambiguous sites observed in each compartment.
na, not available at the time of genotypic testing.
The mean of differences in polymorphism was significantly higher than the mean of strains found in naive treated patients with less than 5 years of diagnosis (Student's t test, p < 0.02).
Additional drug resistance mutations found in only one compartment are shown in bold.
Additional drug resistance conferred by mutations found exclusively in the plasma compartment.
HAART treatment at the moment of genotypic testing.
Only the strains infecting naive patients without drug resistance were included, to reduce the probability of analyzing viruses previously subjected to drug selective pressure. When the drug resistance mutations and other polymorphisms identified by the Alamos database were compared in the two compartments, a greater number of differences was found in strains in patients infected for at least more than 9 years, compared to naive patients (mean differences: 9.4 vs. 4.7, p < 0.02) (Table 1). In addition, more differences were found in the strains infecting highly experienced patients (infected for at least more than 9 years) compared to the strains infecting patients for whom the time since the first diagnosis was lower (mean differences: 9.4 vs. 4.0, p < 0.02). At the same time, no particular association was found in the number of quasispecies that was detected in the strains in each compartment. Finally, no relationship was found between the differences observed between the two compartments and the viral load (Table 1).
Plasma remains the compartment of choice for genotypic testing, although PBMCs harbor archival proviral DNA, and their utility in drug resistance testing has not been completely explored. Several factors may influence the detection of drug resistance mutations in each compartment, such as viral load and the time of infection, since the exposure to long periods of drugs and to diverse HAART regimens may favor the presence of an archival reservoir of drug resistance mutations, particularly in the PBMC compartment. This study is in agreement with previous ones showing that ideally both compartments should be analyzed for a more complete identification of HIV-1 drug resistance mutations in patients on HAART. 3,5,6 Although the number of patients studied is small in our study, it is in agreement with the large study of Turriziani et al. 6 and it is more common to identify additional mutations conferring drug resistance in the plasma compartment, compared to the PBMC compartment, when analyzing treated patients. Quan et al. 7 found greater diversity of the quasispecies population in the PBMC compartment compared to plasma in one patient who previously received several HAART regimens. However, several studies of treated patients show the identification of mutations in the plasma compartment, not necessarily found in PBMC proviral DNA. A possible explanation for this finding is that the resistant variants observed in the plasma compartment were derived from other additional in vivo sources to circulating leukocytes. 2,5
In addition, the probability of detection of a specific drug resistance mutation might be greater in plasma under the selective pressure of the drug, whereas the detection of the same mutation seems to be more independent on the HAART regimen at the moment of testing, for the PBMC compartment. 6 An example of this is the finding that three of seven patients harbored viruses with the M184V mutation conferring resistance to lamivudine only in the plasma compartment, coincident with the use of this drug in HAART at the moment of testing. On the other hand, the PBMC compartment might bring additional information more frequently in naive patients infected with drug-resistant strains. 3,7 –9 We could not analyze this particular aspect in this study, as only one naive patient studied harbored drug-resistant virus and this patient was not included in this study. For this single patient, the buffy compartment was a better predictor of drug resistance (data not shown).
A greater degree of polymorphic divergence was observed between the two compartments if the strains analyzed were infecting patients with longer times of diagnosis of infection, compared to either naive patients or patients with a presumed shorter time of infection. In Venezuela, patients present for diagnosis of HIV-1 frequently at a late stage of disease 10 ; thus the time of diagnosis is often very near the time of the beginning of HAART treatment. As we hypothesized, the number of differences in drug resistance mutations or polymorphisms in the polymerase increased with the time of infection and drug treatment experience. Thus, the contradictory results reported in the literature might be in part due to the fact that this parameter was not considered in the comparisons. Studies on patients with a history of pre-HAART drug resistance indicate that even after years of successful HAART, the wild-type viral strains circulating before the initiation of the therapy as well as all the different drug-resistant viral strains that appear over time remain detectable in the proviral reservoir for a long period of time. 11 The longer the time of infection and the longer the time under treatment, the more diverse this repertoire would be.
In conclusion, a high degree of heterogeneity was found between the proviral DNA found in PBMCs and viral RNA from plasma in the pol region of HIV-1 circulating in long-term infected patients, which might be associated with a longer time of divergent evolution of the viral strains and the selective pressure exerted by HAART. This study supports previous ones suggesting the utility of genotypic testing in both compartments for a more accurate determination of HAART-associated drug resistance mutations.
Footnotes
Acknowledgments
This work was supported by Grant G-2005000394 and Misión Ciencia from MPPCyT, Venezuela.
Disclosure Statement
No competing financial interests exist.