Purification of CD4+ T-cell subsets
Cryopreserved PBMCs were obtained from indvidual 1 between 1989 and 2009, and from individuals 2–5 between 1998 and 1999. CD4+ T cells were isolated from PBMCs by negative selection using CD4+ T-cell isolation kit II (Miltenyi Biotec Incorporated, Auburn, California, USA). CD4+ T-cell subsets (naive, central memory and effector memory) were obtained by FACS-sorting with the BD FACSAria II SORP (BD Biosciences, San Jose, California, USA) cell sorter on the basis of the expression of CD27 and CD45RO. The mAbs used were CD3 Pacific Blue (UCHT1); CD4 allophycocyanin (APC)-Cy7 (RPA-T4); CD45RO APC (UCHL1); and CD27 FITC (L128). All antibodies were obtained from BD Biosciences. Gating strategy used for the FACS-sorting of CD4+ T-cell subsets is depicted in Supplemental Figure 1, http://links.lww.com/QAD/A489.
Analysis of HIV-1 integration sites
Analysis of integration sites was performed by an inverse PCR method as previously described , except that BclI  was used for digestion of genomic DNA. Primer pairs used for the inverse PCR are listed in supplemental Table 1, http://links.lww.com/QAD/A489. The inverse PCR was performed with the Expand High Fidelity PCR System (Roche Applied Science, Indianapolis, Indiana, USA). Each round of PCR consisted of 25 cycles, with the initial denaturation at 94°C for 2 min, followed by 25 cycles of denaturation at 94°C for 15 s, annealing at 50°C for 30 s and extension at 72°C for 2 min, with the final extension at 72°C for 7 min. The amplicons captured the junction between the 5′ end of HIV-1 genome and host cell DNA. On average, 11 independent inverse PCR reactions (range: 3–34) were performed for each time point. The DNA was sequenced with the ABI BigDye Terminator v3.1 Ready Reaction Cycle Sequencing Kit (Applied Biosystems, Foster City, California, USA) and analysed with the ABI PRISM 3130xl Genetic Analyzer (Applied Biosystems). The human genomic sequence in each inverse PCR product was identified as a unique best-hit by BLAT search at the UCSC Bioinformatics Human Genome database (http://www.genome.ucsc.edu; Feb. 2009: GRCh37/hg19). The integration site in the SMC5 gene was independently confirmed by a custom-designed PCR that specifically amplifies the gene segment spanning the integration site in the host genome and the HIV-1 protease gene using the primer sets listed in supplemental Table 1, http://links.lww.com/QAD/A489.
Measurements of levels of HIV-DNA in CD4+ T-cell subsets
Limiting dilution analysis was performed on serial five-fold dilutions of CD4+ T-cell genomic DNA by a nested PCR method as previously described . Location of the PCR primers is shown in Supplemental Figure 3, http://links.lww.com/QAD/A489. The number of HIV-DNA copies was calculated according to the Poisson distribution formula. The HIV-1 copies were then normalized to total genomic DNA input (assuming a cellular genome weights 6 pg) and expressed as HIV-1 copies per 106 CD4+ T cells.
Hypermutation analysis in the HIV-1 protease region of the pol gene
Sequences were screened for evidence of human APOBEC3G-mediated editing/hypermutation, defined as a mutational process in which G-to-A transitions far exceed all other mutations , using the HYPERMUT software (http://www.hiv.lanl.gov) .
Site-directed mutagenesis and preparation of virus stocks
Construction of an HIV-1 variant containing a stop codon (TGG-to-TAG change) at codon 42 (W42Stop) in the protease gene was conducted by site-directed mutagenesis as previously described [24,25]. Additional experimental details are included in the supplementary information, http://links.lww.com/QAD/A489.
Detection of HIV-1 proteins by western blot
Viral lysates for the wild-type and the W42Stop mutant were adjusted to yield equivalent levels of protein (2 μg). Antibodies used in the western blot assay were a rabbit anti-PR antibody (kindly provided from Dr Robert Gorelick, Leidos Biomedical Research Incorporated); mouse monoclonal antip24 antibody (PerkinElmer, Santa Clara, California, USA, NEA9306); mouse monoclonal anti-RT antibody (PerkinElmer; NEA9304); rabbit anti-INT (kindly provided from Dr Robert Gorelick); and goat anti-ENV antibody (Abcam, Cambridge, Massachusetts, USA, ab21179).
Electron microscope images
Prior to transfection, HEK293T cells (0.1 × 106 cells/well) were seeded onto six-well plates in DM-10 media and incubated at 37°C over night. Equal amounts (2.5 μg of DNA) of pNL4.3WT or pNL4.3W42Stop were used for transfection of HEK293T cells with 7.5 μl of the Transit LT. Approximately 24 h after transfection, the cells were fixed with PBS containing 2% glutaraldehyde and 0.1 mol/l sodium cacodylate trihydrate for 1 h at room temperature and then processed for electron microscopy analysis as previously described .
Assessment of replication competency of the wild-type and mutant HIV-1 recombinant viruses
Virus stocks for the wild-type and the mutant HIV-1 containing the W42Stop mutation were treated with DNase I (Life Technologies, Grand Island, New York, USA) to remove contaminating plasmid DNA. An aliquot of virus that was heat-inactivated at 99°C for 15 min served as a negative control for infection. A total of 8 × 106 PHA-stimulated CD4+ T cells were exposed to equal amounts of the wild-type or the W42Stop viruses (adjusted by virion protein levels to 100 ng) at 37°C for 2 h. At days 1 and 7 postinfection, cells were harvested and washed with ice-cold PBS. The cells were then treated with a 25 mmol/l glycine-150 mmol/l NaCl solution for 10 min on ice in order to remove any extracellular virions. The presence of the HIV-DNA in target cells was determined by PCR amplification of the HIV-1 protease gene as described above. Levels of p24 in culture supernatants were measured by a PerkinElmer HIV-1 p24 antigen ELISA kit.
Long-term persistence of an integrated HIV-1 provirus containing a pretermination stop codon at position 42 of the HIV-1 protease
In 1998, while monitoring the presence of cell-associated HIV-RNA in PBMCs from HIV-infected individuals with plasma viral load less than 50 copies/ml to look for an evidence of residual viral replication, we coincidentally identified one individual whose cell-associated HIV-RNA contained an unusually high percentage (∼50% of amplicons examined) of a stop codon at position 42 of the HIV-1 protease. We also found the same stop codon, TGG (W, tryptophan) to TAG (stop) change, at the exactly same position in cell-associated HIV-DNA from the matching time point and named this provirus ‘W42Stop’. This provirus was found in PBMCs of this patient from 1991 to 2005. Given the fact that this naturally occurring replication-incompetent provirus would initially only have been able to infect a single cell, we sought to determine whether or not we could prove this assumption, and, if true, use this provirus to follow the fate of this cell and its progeny (Fig. 1a). Utilizing an inverse PCR technique , we were able to map the integration site to an intron between exons 11 and 12 of the SMC5 (Structural maintenance of chromosome 5) gene on chromosome 9 (9q21.12) (Fig. 1c). The same site was identified in 17 independent experiments utilizing samples that spanned 15 years (Fig. 1a and d). In contrast, analysis of integration sites for three other proviral species revealed three different chromosomal sites of integration: 6q15 (BACH2), 19q13 (RPS19) and 7q36 (intergenic region) (Fig. 1a). Having established that the W42Stop provirus had a single integration site and thus represented the marking of a single cell at a discrete point in time, we looked to see whether we could follow the fate of that cell and its progeny.
Naive, central memory and effector memory CD4+ T-cell subsets were examined for the presence of W42 and W42Stop HIV-DNA at various time points (Fig. 1b). HIV-DNA was detected in all three subsets examined. The frequency (copies/106 cells) of total HIV-DNA (integrated and unintegrated forms combined) was the highest in effector memory CD4+ T cells, followed by central memory, and the lowest in naive CD4+ T cells at most time points (Fig. 1b). The W42Stop provirus was only detected in the effector memory subset during the entire 17-year observational period (1993–2009). The frequency of effector memory cells harbouring the W42Stop was initially 100 copies/106 effector memory cells (0.01%) in 1993 and increased to 10 000 copies/106 effector memory cells (1% of the cells) in 1999, coincident with 3 years of effective cART (Fig. 2a and b). Following this time, the frequency of cells harbouring the W42Stop mutation underwent a logarithmic decay with a functional half-life of 11.1 months. Given that human effector memory CD4+ T cells have previously been reported to be short-lived (14–21 days)  with half-lives ranging from 3 to 6 days [11,12], our finding is quite surprising. Examination of the HIV-1 protease sequences (297 bp long) derived from the effector memory cells during 1993–2009 revealed that the nucleotide sequences for the W42Stop provirus (a total of 98 sequences) were identical (Supplemental Figure 1, http://links.lww.com/QAD/A489). This finding further confirms our hypothesis that the HIV-1 provirus with the W42Stop originated from a single event and was then replicated through somatic cell division.
The W42Stop mutation leads to a noninfectious form of HIV-1
To further prove that the W42Stop provirus did not encode for a replication-competent virus, we next conducted a series of experiments using site-directed mutagenesis to determine the impact of the W42Stop mutation on the replication competency of the encoded virus. A schematic diagram depicting the experiment design is shown in Fig. 3a. Unlike the wild-type virus with its dense core (Fig. 3b), the mutant viruses had a ring or doughnut-shaped structure, due to the failure of protease to cleave the gag-pol precursor (Fig. 3c). The protein compositions of the wild-type and the mutant viral particles were analysed by western blotting of sucrose-purified virions (Fig. 3d). The wild-type virions expressed the envelope, Gag-Pol precursor and cleaved Gag-Pol proteins. The mutant virions expressed envelope but only contained unprocessed and/or partially processed forms of Gag-Pol precursors. Thus, the lack of an intact HIV-1 protease in the W42Stop virions led to severe disruption of proteolytic processing of the Gag-Pol polyproteins. The proteolytic processing of the Env glycoprotein precursor, which is carried out by the host cell proteases, was unaffected in both the wild-type and the mutant virions. Of note was the observation that although RNA was contained in both wild-type and mutant viruses (Fig. 3e), only wild-type virus was able to replicate. The replication-competencies of the wild-type and the mutant viruses were assessed by culturing these viruses with CD4+ T cells and measuring levels of HIV-1 p24 antigen in the culture supernatants (Fig. 3f). No HIV-1 p24 antigen was detected in the culture supernatants of the mutant virus. The inability of the mutant virus to replicate was further confirmed by the absence of amplifiable HIV-DNA in the target CD4+ T cells (Fig. 3g). Taken together, these data demonstrated that the W42Stop virus is replication-incompetent.
A high prevalence of lethal mutations in the peripheral blood pool of HIV-DNAs and cell-associated HIV-RNAs
Although the primary focus of this study was to examine the fate of an integrated provirus with a particular stop codon mutation in the protease gene (the W42Stop) in effector memory CD4+ T cells, we also observed through additional detailed sequencing that such genetic defects could be detected in HIV-1 proviral DNA in high percentages of all sequenced proviruses from peripheral blood lymphocytes. Analysis of the 297-bp HIV-1 protease gene amplified from the cells of the study participant (Individual 1) collected in 1997 revealed that 14% (three of 22) of HIV-DNA from naive CD4+ T cells and 38% (eight of 21) of HIV-DNA from central memory CD4+ T cells contained premature stop codons within this 297-bp region (Fig. 4). A detailed examination of the dideoxynucleotide content of the stop-codons within the region protease revealed that all of the premature stop codons detected in naive and central memory CD4+ T cells were consistent with the action of APOBEC-mediated hypermutation [27–29]. Examination of date-matched plasma samples found no evidence of premature stop codons within the plasma viral RNA sequences (Fig. 4). In individual 1, premature stop codons were detected as early as 1993 and persistently detected for over a 17-year period in all three CD4+ T-cell subsets examined (Fig. 5a). Overall, 21% of HIV-DNA sequences derived from this 3% of the HIV-1 provirus genome (Fig. 5c) in naive and central memory CD4+ T cells contained premature stop codons. An additional four patients (individuals 2–5), all of whom were on suppressive antiretroviral therapy with plasma HIV-RNA levels less than 50 copies/ml, were chosen for analysis of the presence of cell-associated HIV-RNA with genetic defects. All four harboured cell-associated HIV-1 RNA containing premature stop codons in this same 297-bp region at frequencies ranging from 10 to 21% (Fig. 5b). A fairly high prevalence of HIV-1 proviruses containing lethal mutations has also been reported by others in patients with various virological profiles [30–36]. Therefore, the occurrence of high levels of lethal mutations in the peripheral blood pool of HIV-1 proviruses is likely to be a generalized phenomenon.
In this study, taking advantage of an experiment of nature, we have demonstrated the long-term persistence of effector memory CD4+ T cells in a patient with HIV-1 infection and a lack of reversion of these cells to either central memory or naive CD4+ T cells during their decay. We also report the occurrence of high levels of lethal mutations in the peripheral blood pool of HIV-1 proviral DNAs.
Determining the lifespan of human T cells is challenging and the reports in the literature in this area are inconsistent [1–3]. The majority of studies have utilized pulse-chase assays with markers such as BrdU or 2H-glucose [8–12] or relied upon following cells bearing lethal chromosomal lesions induced by therapeutic radiation . Although these methods are powerful in analysing kinetic behaviour of the labelled cells, they are unable to distinguish dividing cells from dying cells, as both lead to dilution of the marker. The present study took advantage of the chance observation of the presence of a unique replication-incompetent provirus in the effector memory cells of an HIV-infected individual who has been followed for over 20 years to address the question of the lifespan of these cells. Following mapping of the integration site of this provirus to confirm that it represented a single integration event, we were able to follow its path over a period of 17 years. As this virus had become part of the genome of the cell and faithfully replicated during each cell division, it allowed us to fully monitor the fate of this cell and its progeny from the time of proviral integration. Using this approach, we demonstrate that effector memory CD4+ T cells can persist over 17 years (longer than previous estimates of 2–3 weeks) . In the present study, we estimated the functional half-life for the effector memory CD4+ T-cell clone to be 11.1 months. However, the 11.1-month half-life could be an overestimate if the effector memory clone was continuing to divide after 1999 when we detected decay of the clone. Currently, we do not have the ability to assess whether continuing cell division took place for the effector memory clone after 1999. As demonstrated by this study, measuring the exact ‘intermitotic’ half-life of individual clones remains a challenge. We also demonstrate that human effector memory CD4+ T cells undergo substantial clonal expansion and can exist for years without detectably reverting to naive or central memory phenotypes. It is theoretically possible that central memory CD4+ T cells residing in lymphoid tissue harbour this same provirus and thus we cannot definitively state that there has been no reversion from effector memory to central memory.
Given the temporal relationship between the expansion of this clone and levels of the HIV-1 provirus, it is likely that the target of this cell is HIV-1 or an accompanying opportunistic pathogen such as cytomegalovirus (CMV). These results indicate that active immune responses may in some instances persist over decades and are consistent with studies of primary CMV infection in renal transplants. In those studies, the frequency of CMV-specific CD4+ T cells increased from less than 0.05 to 1.24% (range: 0.21–1.60%, n = 7) during primary infection, and the circulating CMV-specific CD4+ T cells were predominantly an effector memory phenotype and persisted for at least 120 weeks .
An additional finding of interest was the identification of lethal mutations in HIV-1 peripheral blood proviral DNA occurring at a high frequency. It is important to point out that our study only examined 297 bp of a 9800-bp genome. Given that we observed that approximately 20% of proviruses harboured a stop codon within these 279 base pairs, only equivalent to 3% of the entire HIV-1 genome, it is quite plausible that far more proviruses would be found to harbour a stop codon or other genetic defects somewhere in their full-length HIV-1 genome. Many of these defective proviruses cannot encode a replication-competent virus but retain the ability to encode the production of RNA (as shown in this study) and protein. It is tempting to speculate that the ongoing production of these ‘foreign materials’ within CD4+ T cells, even in patients with plasma virus loads less than 50 copies/ml, may play some role in the ongoing immune activation that is characteristic of HIV-1 infection. Using extensive sequencing, it has been recently suggested that the size of the latent reservoir of HIV-1 in peripheral blood may be 60 times greater than previously thought . Combined with our observation of long-term persistence of replication-incompetent integrated HIV-1 provirus and associated RNA transcripts, it is increasingly clear that estimation of the size of HIV-1 reservoir, the characterization of viral latency and the decay dynamics of HIV-1 may be more complex than currently appreciated. More detailed analyses utilizing genome-wide sequencing of HIV-1 may provide a clearer understanding of the genetic integrity of HIV-1 proviruses in the peripheral blood pool.
We thank Richard Davey, Colleen Hadigan, Joseph Kovacs, Frank Maldarelli and Mary Wright for providing patient specimens; Robin Dewar, Helene Highbarger, Akram Shah for running the Abbott HIV-1 assay; Xiaojun Hu and Da Wei Huang for helpful discussion; Robert Gorelick for providing antibodies for western blot; and Kunio Nagashima for electron microscopy analysis.
This work was funded through the intramural research programme of the National Institute of Allergy and Infectious Diseases of the National Institutes of Health (Bethesda, Maryland) and in part with federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. HHSN261200800001E. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products or organizations imply endorsement by the U.S. Government.
H.I. and H.C.L. designed the study, analysed the data and prepared this manuscript. H.I., T.I. and A.H. performed experiments. V.N., R.A.L. B.D. and T.I. contributed to the design of some experiments; J.W.A. performed T-cell sorting. C.A.R. led the identification of patient specimens.
Conflicts of interest
The authors declare no competing financial interests.
1. Sprent J. T and B memory cells
2. Crotty S, Ahmed R. Immunological memory in humans
. Semin Immunol
3. Ahmed R, Gray D. Immunological memory and protective immunity: understanding their relation
4. Sanders ME, Makgoba MW, Shaw S. Human naive and memory T cells: reinterpretation of helper-inducer and suppressor-inducer subsets
. Immunol Today
5. Mackay CR. Immunological memory
. Adv Immunol
6. Tough DF, Sprent J. Life span of naive and memory T cells
. Stem Cells
7. Michie CA, McLean A, Alcock C, Beverley PC. Lifespan of human lymphocyte subsets defined by CD45 isoforms
8. Neese RA, Misell LM, Turner S, Chu A, Kim J, Cesar D, et al. Measurement in vivo of proliferation rates of slow turnover cells by 2H2O labeling of the deoxyribose moiety of DNA
. Proc Natl Acad Sci U S A
9. McCune JM, Hanley MB, Cesar D, Halvorsen R, Hoh R, Schmidt D, et al. Factors influencing T-cell turnover in HIV-1-seropositive patients
. J Clin Invest
10. Macallan DC, Asquith B, Irvine AJ, Wallace DL, Worth A, Ghattas H, et al. Measurement and modeling of human T cell kinetics
. Eur J Immunol
11. Macallan DC, Wallace D, Zhang Y, De Lara C, Worth AT, Ghattas H, et al. Rapid turnover of effector-memory CD4(+) T cells in healthy humans
. J Exp Med
12. Srinivasula S, Lempicki RA, Adelsberger JW, Huang CY, Roark J, Lee PI, et al. Differential effects of HIV viral load and CD4 count on proliferation of naive and memory CD4 and CD8 T lymphocytes
13. Lanzavecchia A, Sallusto F. Dynamics of T lymphocyte responses: intermediates, effectors, and memory cells
14. Bell EB, Westermann J. CD4 memory T cells on trial: immunological memory without a memory T cell
. Trends Immunol
15. Sprent J. Lifespans of naive, memory and effector lymphocytes
. Curr Opin Immunol
16. Sallusto F, Geginat J, Lanzavecchia A. Central memory and effector memory T cell subsets: function, generation, and maintenance
. Annu Rev Immunol
17. Robertson JM, MacLeod M, Marsden VS, Kappler JW, Marrack P. Not all CD4+ memory T cells are long lived
. Immunol Rev
18. Hellerstein MK, Hoh RA, Hanley MB, Cesar D, Lee D, Neese RA, et al. Subpopulations of long-lived and short-lived T cells in advanced HIV-1 infection
. J Clin Invest
19. Han Y, Lassen K, Monie D, Sedaghat AR, Shimoji S, Liu X, et al. Resting CD4+ T cells from human immunodeficiency virus type 1 (HIV-1)-infected individuals carry integrated HIV-1 genomes within actively transcribed host genes
. J Virol
20. Bastie-Sigeac F, Lucotte G. Optimal use of restriction enzymes in the analysis of human DNA polymorphism
. Hum Genet
21. Imamichi H, Crandall KA, Natarajan V, Jiang MK, Dewar RL, Berg S, et al. Human immunodeficiency virus type 1 quasi species that rebound after discontinuation of highly active antiretroviral therapy are similar to the viral quasi species present before initiation of therapy
. J Infect Dis
22. Vartanian JP, Meyerhans A, Sala M, Wain-Hobson S. G-->A hypermutation of the human immunodeficiency virus type 1 genome: evidence for dCTP pool imbalance during reverse transcription
. Proc Natl Acad Sci U S A
23. Rose PP, Korber BT. Detecting hypermutations in viral sequences with an emphasis on G --> A hypermutation
24. Imamichi T, Berg SC, Imamichi H, Lopez JC, Metcalf JA, Falloon J, et al. Relative replication fitness of a high-level 3′-azido-3′-deoxythymidine-resistant variant of human immunodeficiency virus type 1 possessing an amino acid deletion at codon 67 and a novel substitution (Thr-->Gly) at codon 69
. J Virol
25. Brann TW, Dewar RL, Jiang MK, Shah A, Nagashima K, Metcalf JA, et al. Functional correlation between a novel amino acid insertion at codon 19 in the protease of human immunodeficiency virus type 1 and polymorphism in the p1/p6 Gag cleavage site in drug resistance and replication fitness
. J Virol
26. Gonda MA, Aaronson SA, Ellmore N, Zeve VH, Nagashima K. Ultrastructural studies of surface features of human normal and tumor cells in tissue culture by scanning and transmission electron microscopy
. J Natl Cancer Inst
27. Sheehy AM, Gaddis NC, Choi JD, Malim MH. Isolation of a human gene that inhibits HIV-1 infection and is suppressed by the viral Vif protein
28. Lecossier D, Bouchonnet F, Clavel F, Hance AJ. Hypermutation of HIV-1 DNA in the absence of the Vif protein
29. Harris RS, Bishop KN, Sheehy AM, Craig HM, Petersen-Mahrt SK, Watt IN, et al. DNA deamination mediates innate immunity to retroviral infection
30. Janini M, Rogers M, Birx DR, McCutchan FE. Human immunodeficiency virus type 1 DNA sequences genetically damaged by hypermutation are often abundant in patient peripheral blood mononuclear cells and may be generated during near-simultaneous infection and activation of CD4(+) T cells
. J Virol
31. Kieffer TL, Kwon P, Nettles RE, Han Y, Ray SC, Siliciano RF. G-->A hypermutation in protease and reverse transcriptase regions of human immunodeficiency virus type 1 residing in resting CD4+ T cells in vivo
. J Virol
32. Piantadosi A, Humes D, Chohan B, McClelland RS, Overbaugh J. Analysis of the percentage of human immunodeficiency virus type 1 sequences that are hypermutated and markers of disease progression in a longitudinal cohort, including one individual with a partially defective Vif
. J Virol
33. Fourati S, Lambert-Niclot S, Soulie C, Malet I, Valantin MA, Descours B, et al. HIV-1 genome is often defective in PBMCs and rectal tissues after long-term HAART as a result of APOBEC3 editing and correlates with the size of reservoirs
. J Antimicrob Chemother
34. Land AM, Ball TB, Luo M, Pilon R, Sandstrom P, Embree JE, et al. Human immunodeficiency virus (HIV) type 1 proviral hypermutation correlates with CD4 count in HIV-infected women from Kenya
. J Virol
35. Gandhi SK, Siliciano JD, Bailey JR, Siliciano RF, Blankson JN. Role of APOBEC3G/F-mediated hypermutation in the control of human immunodeficiency virus type 1 in elite suppressors
. J Virol
36. Eyzaguirre LM, Charurat M, Redfield RR, Blattner WA, Carr JK, Sajadi MM. Elevated hypermutation levels in HIV-1 natural viral suppressors
37. Sester M, Sester U, Gartner BC, Girndt M, Meyerhans A, Kohler H. Dominance of virus-specific CD8 T cells in human primary cytomegalovirus infection
. J Am Soc Nephrol
38. Ho YC, Shan L, Hosmane NN, Wang J, Laskey SB, Rosenbloom DI, et al. Replication-competent noninduced proviruses in the latent reservoir increase barrier to HIV-1 cure
genetic marker; HIV-1; in-vivo persistence of CD4-positive T lymphocytes; integration; provirus; replication-incompetent
Supplemental Digital Content
© 2014 Lippincott Williams & Wilkins, Inc.