您的位置： 百康网 > 期刊 > 内科学 > 《传染病学杂志》 > 2003年2月第3期 > 正文
Combination Antiretroviral Therapy Results in a Rapid Increase in T Cell Receptor Variable Region Repertoire Diversity within CD45RA CD8 T Cells in Human Immunodeficiency VirusInfected Children
http://www.100kang.com 2007-5-9 15:15:27 Combination


Departments of 1Pediatrics, Division of Immunology and Infectious Diseases, 2Pathology, Immunology, and Laboratory Medicine, and 3Statistics, University of Florida, College of Medicine, Gainesville

Received 25 January 2002; revised 3 October 2002; electronically published 24 January 2003.

Human immunodeficiency virus (HIV) type 1 disrupts the T cell receptor (TCR) variable region (V)  repertoire in CD8 T cells by impairing thymic capacity and skewing postthymic cellular maturation. The TCR repertoire was examined using spectratyping of CDR3 length diversity within CD45RA and CD45RO CD8 T cells in HIV-infected and healthy children. In healthy children, CDR3 lengths displayed Gaussian distribution in both CD45RA and CD45RO subsets. V families in HIV-infected children displayed a large proportion of perturbations in both subsets. High virus load and advanced immunosuppression correlated with increased perturbations within CD45RA but not CD45RO CD8 T cells. After therapy and virus suppression, there was rapid reestablishment of Gaussian distributions in CD45RA cells. HIV-1–induced disruption of TCR diversity within CD45RA CD8 T cells correlates with disease progression. Suppression of viral replication by treatment results in the rapid correction of TCR diversity in this CD8 subset because of emergence of new T cells from the thymus.

　

    Presented in part: 8th Conference on Retroviruses and Opportunistic Infections, Chicago, 4–8 February 2001 (abstract 686).
     Patients and/or their legal guardians provided informed consent. Guidelines for research involving human subjects of the US Department of Health and Human Services and the institutional review board of the University of Florida were followed.
     Financial support: Public Health Service Awards (grants RO1 HD32259 and RO1 AI 47723); National Institutes of Health (grant RR0082 sponsored the General Clinical Research Center); Elizabeth Glaser Pediatric AIDS Foundation (grant PG-50956); University of Florida Interdisciplinary Center for Biological Research.
     Reprints or correspondence: Dr. John W. Sleasman, Div. of Allergy and Immunology, Dept. of Pediatrics, University of South Florida, College of Medicine, 801 6th St. S, St. Petersberg, FL 33704 ().

     Human immunodeficiency virus (HIV) type 1 infection results in perturbations of T cell receptor (TCR) variable region (V)  repertoire within the CD8 T cells of infected children and adults [1–4]. Although CD8 T cells play an important role in the suppression of viral replication, virus-induced anergy leads to defective T cell responses and contributes to the failure of immunity to control infection [5, 6]. The prognostic significance of alterations within the TCR repertoire remains uncertain, particularly in HIV-infected children whose T cell dynamics are affected by their high thymic capacity and immaturity of the cellular immune response [7–9]. HIV-1 infection skews the maturation of antigen-specific CD8 T cells and alters their homing and effector cell functions [10]. This is one mechanism that enables HIV-1 to evade the T cell immune response [11]. CD8 T lymphocytes are composed of different subpopulations on the basis of their stage of differentiation, activation state, cellular phenotype, and functional capacity [12, 13]. There is a dynamic steady state among CD8 T cells as they emigrate from the thymus, encounter antigens, become activated to proliferate as memory T cells, or differentiate into cytotoxic effector T cells [5, 11]. CD8 T cells can be subdivided into different maturation lineages on the basis of the expression of the different isoforms of the leukocyte common antigen CD45 [14, 15]. The high-molecular-weight isoform, CD45RA, is expressed on both naive and antigen-primed effector CD8 T cells [12, 16]. Activated CD8 T cells predominantly express the low-molecular-weight isoform, CD45RO [17]. The isoforms of CD45 can be used to separate CD8 T cells into distinct populations, which allows for the accurate assessment TCR diversity within different CD8 T cell subsets [18, 19].

     Recombination of germ line–encoded V, diversity, and joining (J) region gene elements that make up the  chain of the TCR establishes antigen specificity and diversity of the cellular immune response [20–22]. Recombination forms the hypervariable fragment CDR3 region, which varies in both amino acid sequence and length [23]. Variation in CDR3 length can be exploited to monitor changes within the TCR repertoire as CD8 subpopulations differentiate during antigen-induced T cell activation [19, 24, 25]. Patterns of clonal dominance within CD8 T cells are associated with slower disease progression in HIV-infected children [4]. However, limited oligoclonal expansion of HIV-1–specific CD8 T cells during the acute phase of infection is associated with the emergence of viral immune escape, high levels of viral replication, and disease progression [26]. This apparent paradox forms the basis of the present study, which examines TCR diversity within different CD8 T cell subpopulations before and after antiretroviral therapy. Effective control of viral replication should lead to the reestablishment of a normal, Gaussian distribution of CDR3 lengths in HIV-infected children, similar to that in adults [2, 25]. However, TCR diversity within different CD8 T cell subsets has not been performed. Our study examines a relationship among perturbations in TCR V repertoire, virus load, and the extent of CD4 T cell attrition. In addition, changes in TCR diversity within CD45RA and CD45RO CD8 T cells after effective control of viral replication were evaluated.

SUBJECTS, MATERIALS, AND METHODS

     Subjects.     The study consisted of a cross-sectional analysis of blood samples obtained from 16 HIV-infected and 10 healthy age-matched children. Samples were obtained from HIV-infected children who were at different Centers for Disease Control and Prevention clinical and immune stages of HIV disease [27] and who were receiving different antiretroviral regimens. Samples were obtained from 6 subjects before and after the initiation of combination antiretroviral therapy that included 2 different nucleotide reverse-transcriptase inhibitors plus a protease inhibitor. The median age of the study cohort was 6 years (range, 2–16 years) for the healthy children and 7.5 years (range, 4–16 years) for the HIV-infected children (P = .38). Healthy children selected for study had no underlying medical condition, recent illnesses, or immunizations. Plasma HIV-1 RNA copy number was determined using the Roche Amplicor 1.0 assay, according to the manufacturer's protocol (Roche Molecular Systems). The sensitivity of the assay has 50 copies/mL as the lower limit of detection. T cell subsets were determined by use of flow cytometry [28].

     Isolation and purification of T lymphocyte subsets.     Peripheral blood mononuclear cells (PBMC) were isolated by ficoll-hypaque centrifugation. Separation of T cell subpopulations from PBMC was done as described elsewhere [18, 19, 29]. CD8 T cells were selected using magnetic multisort microbeads coated with an anti-CD8 antibody (Miltenyi Biotech) and a high-gradient magnetic separation column (Miltenyi Biotech). Microbeads coated with monoclonal antibody (MAb) and mononuclear cells were mixed and incubated at 4°C for 15 min. Labeled CD8 T cells, which adhered to the column, were washed 3 times with wash buffer and collected. Purified CD8 cells were released from the beads by using release reagent at 4°C for 15 min. The CD8 T cell subpopulations were initially separated into CD45RA, and, subsequently, the CD45RO subpopulation was isolated using microbeads coated with the appropriate MAbs (Miltenyi Biotech). The purity of each T cell subpopulation was >95%, as determined by flow cytometry and molecular analysis, with <1% contamination with the reciprocal subpopulation [18, 30].

     Extraction of mRNA and synthesis of cDNA.     Lymphocyte mRNA was isolated and purified using procedures described elsewhere [31, 32]. In brief, mRNA was extracted from 1–5 × 105 purified T lymphocytes by using guanidinium isothiocyanate and a spin column with oligo (dT)–cellulose (mRNA purification kit; Pharmacia). First-strand cDNA was synthesized from 0.2 to 1 g of mRNA by mixing with 1 g of random hexanucleotides (Promega) and 5 L of supertranscriptase II (Gibco BRL) in a total volume of 20 L at 37°C for 1 h. The reaction was terminated by heating to 70°C for 10 min.

     Analysis of CDR3 length using spectratyping.     CDR3 size analysis within TCR– chain was performed using 2-step polymerase chain reaction (PCR) [19, 33]. Separate amplification reactions for 22/24 human V gene families were carried out in 25-L reactions containing 0.5 L (20 M) of each forward V and reverse C primers, 0.5 L (10 mM) of dNTPs, 2.5 L of 10× PCR buffer, 20.5 L of sterile H2O, 0.25 L of cDNA, and 1.25 U of Taq polymerase. Amplifications were started with an initial denaturing step of 3 min at 95°C followed by 35 cycles of 95°C for 1 min, 55°C for 1 min, and 72°C for 1 min, with a final extension step of 7 min. Primer sequences have been published elsewhere [19].

     Second-round PCR was done with first-round products as the template and a pair of nested PCR primers. The forward primer (VN) was nested 3′ to the first round V-specific primer, and the reverse primer (CN) was located 88 bp away from the CDR3 region [19]. The reverse primer (CN) contained a nontemplate sequence GTTTCTT, which was placed on the 5′ end of the CNs primer and labeled at the 5′ end with a blue fluorescent dye (6-FAM; Applied Biosystems). To increase amplifying specificity, the primer concentrations were reduced to 5 M in a 25-L reaction. Second-round amplification was performed at 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s for 20 cycles. The fluorescent PCR product and size marker (red, ROX400; Applied Biosystems) were mixed with formamide and denatured at 94°C for 2 min. Samples were loaded on 6% acrylamide sequencing gel (National Diagnostic) on a 24-lane ABI model 373 DNA sequencer (Applied Biosystems) and run for 6 h.

     Quantitative analysis of CDR3 length profiles.     The area under the curve (AUC) for each V family was calculated by using ABI PRISM GeneScan analysis software (Applied Biosystems), which calculates the relative intensity of each product. To quantify CDR3 length distribution, we adapted a method of establishing a standard CDR3 length distribution profile described elsewhere [2]. In brief, an average distribution of CDR3 lengths within CD8 CD45RA or CD45RO T cells from 10 uninfected, age-matched children was used to establish a standard distribution for each V family. This was used for comparison to evaluate CDR3 length distribution for each V family in samples from HIV-1–infected individuals. AUC was translated into a probability distribution, P(i) = Ai/(Ai), using the fraction of the area (Ai) under the profile for each CDR3 length i, from minimal to maximal lengths in steps of 3 nt [2]. The extent of perturbation for each CDR3 length in a TCR profile was calculated for each Vk on the basis of the distance (D) between the probability distributions of sample i and control, D(i) = P(i) - P(i). The sum of the absolute value of the distances, D = 100i  D(i)  /2, in each family over all TCR lengths i yields the percentage perturbation of that TCR profile. Normal distribution with no perturbation corresponds to no difference (D = 0%) between control and sample probability distributions. For uninfected subjects, their TCR profile in CD8 CD45RA T cells had an average D of 14% ± 3% per V family. Therefore, the perturbation of CDR3 profile in CD8 CD45RA T cells was defined as D >23% (D + 3 SD) per family. The perturbation in CD8 CD45RO T cells was defined as D >25% (D + 3 SD), with an average D of 17.5% ± 2.5% per family. All analysis was performed using SAS software (SAS Institute).

     Sequencing CDR3 segments.     First-round templates for individual families were amplified using identical forward and reverse V PCR primers that lacked the fluorescent label used for spectratyping. The second-round products were used for TA cloning and sequencing. The PCR products were purified using a gel extraction kit (Qiagen), according to the manufacture's recommendation. Products were directly cloned into the TA cloning vector PCR 2.1 (Invitrogen), sequenced by fluorescent dideoxy terminators, and analyzed on an Applied Biosystems model 337A automated sequencer. Sequence of the CDR3 amino acid was analyzed using DNAMAN software (Lynnon).

     PCR amplification of T cell–receptor excisional circles (TRECs) in CD8 CD45RA T cells.     Purified CD8 CD45RA T cells were lysed in 100 L of K buffer (containing 100 g/mL proteinase K) at a concentration of 5000 cells/L for 1 h at 56°C and then for 10 min at 95°C to inactivate the protease. A pair of coding joint primers within the Rec-Jα region, 5′-CAT ATG GAA GAA TGA ATC AAT TAA TG-3′ and 5′-GAA TCT CTC ACT GAT AAT TTT CAG-3′, were used to amplify TRECs. Primers for -actin consisted of 5′-CTA GAA GCA TTT GCG GTG GAC-3′ and 5′-GAA ACT ACC TTC AAC TCC ATC ATG-3′. PCRs were prepared in final volume of 50 L, including 1 L of coding joint primers (10 pmol each) or 1 L of -actin primers (5 pmol each), 1 L of dNTPs (10 mM), 5 L of 10× PCR buffer, 37.5 L of sterile H2O, 4 L of the cell lysates, and 0.5 L of Taq DNA polymerase. The PCRs were performed in a cycle sequencer (Model 9600; Perkin-Elmer) with an initial 3 cycles: 97°C for 1 min, 55°C for 2 min, and 72°C for 1 min, followed by 37 cycles: 94°C for 1 min, 55°C for 2 min, and 72°C for 1 min. Approximately 20 L of the PCR products was separated on 2% agarose gel in Tris-acetate/EDTA electrophoresis buffer. Gels were stained with ethidium bromide for visualization under UV light.

     Immunofluorescence staining and flow cytometric analysis.     Lymphocyte subset enumeration was performed by standard 3-color flow cytometry analysis of PBMC using a FACScan (Becton Dickinson). Enumeration of CD3-, CD4-, and CD8-expressing T cells was carried out as described elsewhere [28]. Further analysis of CD8 T cell subpopulations to enumerate CD8 CD45RA CD11a T cells was assessed using MAbs directly conjugated with peridininchlorophyll protein, fluorescein isothiocyanate (FITC), or phycoerythrin. These MAbs were anti-CD8 (Leu2a; Becton Dickinson), anti-CD45RA FITC (Leu18; Becton Dickinson), and anti-CD11a (Becton Dickinson). T lymphocytes were gated, and a minimum of 15,000 cells/sample were acquired. The data were analyzed using CELLQUEST software (Becton Dickinson).

     Statistical analyses.     Statistical differences between study groups and changes in CDR3 length after therapy were determined using Sigma Stat (Jandel Scientific). A paired Student's t test was used to compare difference in CDR3 length distribution between healthy and HIV-infected children. Student's t test was also used to compare therapy-induced changes in CD4 T cell counts, virus load, and changes in CDR3 lengths. In addition, linear-regression analyses were performed to determine the relationship between virus load or CD4 T cell percentages with the degree of CDR3 length perturbations in CD45RA and CD45RO CD8 T cells from HIV-infected children.

RESULTS

     Differences in CDR3 length distribution in CD45RA and CD45RO CD8 T cells from healthy and HIV-infected children.     TCR CDR3 length distributions for each of 22 V families were determined among 16 HIV-infected children and 10 age-matched healthy children. An average D for all CDR3 profiles in CD8 CD45RA T cells from HIV-infected children was 26.5% ± 6.5%/V family, compared with 14.3% ± 3% from uninfected healthy control subjects (P < .0001). The average D in CD8 CD45RO T cells in HIV-infected children was 35% ± 9%/V family, compared with 17.5% ± 2.5% in the healthy control subjects (P < .0001).

     A summary of perturbations in all V families in healthy and HIV-1–infected children is shown in . A perturbation for any particular V family, in either CD45RA or CD45RO CD8 T cells, is defined as an average D > 3 SD, compared with the average of CDR3 length distributions of that individual V family from healthy children. If all the CDR3 lengths are within 3 SD of the average D, then the family is considered to have a Gaussian distribution of CDR3 lengths. The TCR Vfamilies within the CD8 T cell subsets from healthy children had fewer perturbations than did subsets from HIV-infected children. The median number of V families with perturbations in either CD45RA or CD45RO T cells was 5.5/healthy child, compared with 18 V families/HIV-infected child (P < .001). Among the healthy children, the frequency of V families with perturbations was similar in the CD454RA (median, 2.0; quartile range, 1.0–5.0) and CD45RO cells (median, 3.0; quartile range, 2.0–5.0), respectively (P = .938). In comparison to healthy children, HIV-infected children had significantly greater numbers of V families with perturbations in both CD8 subsets (CD45RA cells median, 10.0; quartile range, 5.5–14.0 and CD45RO cells median, 14.0; quartile range, 8.5–19.5) (P < .01 for each subset, HIV-infected vs. healthy children). Perturbations were more frequent in the CD45RO than the CD45RA CD8 T cells from HIV-infected children (P = .004).

fig.ommitted

Figure 1.        Frequency of CDR3 length perturbation among 22 variable region (V)  subfamilies within CD45RA and CD45RO subsets of CD8 T cells in human immunodeficiency virus (HIV)–infected children vs. control subjects. Subjects P1–P6 have not received combination antiretroviral therapy; subjects P7–P16 are receiving treatment. Gray boxes with "A" represent a perturbation within CD45RA T cells; hatched boxes with "O" represent perturbation within CD45RO T cells; "A/O" represents perturbations within both CD45RA and CD45RO T cells. Ctr, control; Pt, patient; TCR, T cell receptor.

     Perturbations within both CD45RA and CD45RO subsets of the same V family occurred in only 3 (1.4%) of 220 V families evaluated among healthy children. In contrast, 116 (33%) of 352 separate analyses of CDR3 length from the HIV-infected children showed simultaneous perturbations in both CD45RA and CD45RO CD8 T cells in the same V family (P < .001, Student's t test).  shows an example of CDR3 length distributions for 12 different V families from a healthy  and an HIV-infected  child. In the healthy child, CDR3 lengths within both CD45RA and CD45RO CD8 T cell subsets display predominantly Gaussian distributions, although perturbations within the CD45RO subsets can be observed, as is shown in V16 and V24 (, asterisks). In HIV-infected children, patterns of CDR3 perturbations occurred in CD45RA or CD45RO CD8 T cell subpopulations. The patterns of CDR3 lengths within the subsets were variable. In some families, such as V3, V13, V16, or V21, perturbations occurred within CD45RO cells, whereas a Gaussian pattern for the same families appeared in the CD45RA cells. The reciprocal pattern was also observed for V1, V2, V9, and V14. A pattern observed almost exclusively in HIV-infected children was perturbations within both CD45RA and CD45RO T cell subsets as in V7, V15, and V18.

fig.ommitted

Figure 2.        CDR3 length distributions of selected variable region (V)  subfamilies within CD45RA and CD45RO CD8 T cell subsets in a human immunodeficiency virus (HIV)–infected and a healthy child. A, CDR3 amino acid length distributions (X-axis) compared in relative fluorescent intensity (FI) (Y-axis) for 12 representative V families from CD45RA and CD45RO CD8 T cells from a healthy child. B, CDR3 length distribution in the same V families from an HIV-infected child of the same age. *, Predominant CDR3 lengths (>23% of average distance in CD8 CD45RA T cells or >35% of average distance in CD8 CD45RO T cells).

     Oligoclonal expansions of TCR V families within CD45RA and CD45RO CD8 T cell subsets.     Most V families that displayed perturbations  had single predominant CDR3 lengths that accounted for the deviation from Gaussian distribution. To evaluate the composition of perturbations in V families, CDR3 regions expressed by an individual V family were sequenced after amplification and cloning [34]. Representative families that were sequenced are shown in . CDR3 clones from the CD8+CD45RA V18 family had a Gaussian distribution of 8–14 aa. Each CDR3 region was composed of a different sequence, which indicates a polyclonal profile ( upper panel). In contrast, V18 TCR within the CD45RO cells from the same HIV-ifected subject expressed a 10-aa predominant CDR3 length that was identical among 15 of 22 clones, which indicates an oligoclonal profile ( lower panel). A similar result was found in the V22 TCR family . In this example, 18 (72%) of 25 V22 clones in the CD45RA subset expressed a identical 12-aa predominant CDR3 length ( upper panel), whereas all 11 V22 clones from the CD45RO subset, which displayed a Gaussian distribution of CDR3 lengths, had unique amino acid sequences ( lower panel). Among both HIV-infected and healthy children, sequencing revealed that the predominant CDR3 length was always oligoclonal (data not shown).

fig.ommitted

Figure 3.        Patterns of CDR3 lengths distribution and amino acid sequences of cloned variable region (V)  polymerase chain reaction products from CD45RA and CD45RO CD8 T cells from human immunodeficiency virus (HIV)–infected children. CDR3 amino acid lengths (X-axis), compared with relative fluorescence intensity (FI) (Y-axis). The V, CDR3, and joining region (J)  amino acid sequences and length with relative clonal frequency are shown on the right columns. A, V18 CDR3 length distribution and amino acid sequences from an HIV-infected child in which CD45RA cells are Gaussian and CD45RO T cells have a predominant CDR3 lengths of 10 aa. B, V22 for the same child in which CD45RA cells contains a predominant peak and CD45RO cells are Gaussian. C, V16 CDR3 lengths in CD45RA and CD45RO cells that have predominant peaks of the same length and amino acid sequence. D, V7 CDR3 lengths in which the predominant peaks in CD45RA and CD45RO CD8 T cells has different lengths. Bold type denotes predominant sequences.

     Predominant CDR3 peaks of 8 aa occurred within both CD45RA and CD45RO T cells of the V16 family from an HIV-infected child . Five of 10 clones from the CD45RA subset were identical in amino acid sequence to 9 of 10 clones from the CD45RO subset. In contrast, CD45RA and CD45RO T cells in the same V7 family contained discordant CDR3 lengths with distinct amino acid sequences . In CD45RA cells, the predominant 11-aa V7 sequence made up approximately half (11/20) of the CDR3 clones ( upper panel) but was undetectable among the V7 CDR3 clones from the CD45RO subset ( lower panel). Yet predominant 9-aa CDR3 length in the CD45RO T cells was detected as a minor sequence (2/20) in the CD45RA T cells. These findings indicate that, during the course of HIV-1 infection, the same expanded T cell clone can be detected simultaneously within different CD8 T cell subsets.

     Relationship between V perturbations and immune or viral status among HIV-infected children.     A cross-sectional analysis that evaluated a relationship between the percentage of CD4 T cells and the extent of V families with perturbations within CD8 subsets was performed in 16 HIV-infected children who had varying CD4 T cell counts and HIV loads. The extent of V families with perturbations was inversely correlated with CD4 percentage among the CD45RA subset (r = -0.68; P < .01)  but failed to show a significant relationship among the CD45RO subset (r = -0.08; P = .76) . There was a positive correlation between increased average D, with perturbations and plasma virus burden among the CD45RA subset (r = 0.53; P = .03)  but not in the CD45RO subset (r = -0.04; P = .88) .

fig.ommitted

Figure 4.        Relationships between the frequency of variable region (V)  CDR3 length perturbations in CD45RA and CD45RO CD8 T cell subsets to the CD4 T cell percentage and plasma virus load in a cohort of human immunodeficiency virus (HIV)–infected children. A and B, Relationship between the percentage of average distance (D) of CDR3 length profile per subject (X-axis) in CD8 CD45RA T cells (A) and CD8 CD45RO T cells (B) to the CD4 T cell percentage (Y-axis). C and D, Relationship between HIV-1 load, calculated in log10 RNA copies/mL, and perturbations in CD45RA and CD45RO T cells. Correlation coefficient and P values were determined using linear regression.

     Antiretroviral therapy restores CDR3 length diversity in CD8 CD45RA T cells.     Changes in CDR3 length diversity over time was evaluated among 5 therapy-naive children who were treated successfully with combination antiretroviral therapy . At the start of therapy, the mean virus burden was 3.9 ± 3.6 log10 copies/mL, and the mean CD4 T cell percentage was 31.4% ± 4.6%. Treatment produced a decline in virus burden to <50 copies/mL within 8–16 weeks in all 5 children. A sixth child with advanced disease and severe immunosuppression prior to treatment, who did not respond to therapy, was included as a comparison. In this child, virus burden declined ∼2 logs (from 5.8 log10 copies/mL before therapy to 3.9 log10 copies/mL) by 60 weeks of treatment, whereas the percentage of CD4 T cell increased from pretherapy levels of 7% to only 8% after 60 weeks.

fig.ommitted

Figure 5.        Changes of T cell receptor (TCR) CDR3 length profile within CD8 CD45RA and CD8 CD45RO T cells after highly active antiretroviral therapy with different time points in human immunodeficiency virus (HIV)–infected cells. HIV-infected children (as shown in ) prior to (week 0) and after treatment. Gray boxes with "A" represent a perturbation within CD45RA T cells; hatched boxes with "O" represent perturbation within CD45RO T cells; "A/O" represents perturbations within both CD45RA and CD45RO T cells. Pt, patient; V, variable region .

     In the 5 successfully treated children, the pretherapy V average D within CD45RO T cells did not change significantly after treatment (P > .15). In contrast, the average D improved significantly in the CD45RA T cells, from 36.9% ± 2.3% before therapy to 17.4% ± 1.7% after therapy (P < .001). As shown in , the average number of the perturbed V families among CD8 CD45RA T cells per subject prior to treatment was 11.0 ± 4.0, which declined to 2.0 ± 1.0 after 8–16 weeks of successful therapy (P = .003). Among the children who experienced suppression of viral replication, this improvement in CDR3 distribution was not observed in CD8 CD45RO T cells (P = .69).

     As shown in the representative subject in , effective control of viral replication resulted in the emergence of new CDR3 lengths among V1, V9, V11, and V22 families in CD45RA cells. In each V family, the predominant CDR3 length (10, 11, 9, or 12 aa, respectively) persisted over the course of therapy ( asterisks). Yet the average D of the predominant peaks decreased with the emergence of new CDR3 lengths . The predominant peaks represented 65% ± 0.16% of the average distances prior to therapy but only 19% ± 6% of the average D as virus load declined to <50 copies/mL (P < .006). Gaussian distributions of CDR3 lengths reappeared within 4–8 weeks after the initiation of therapy.

fig.ommitted

Figure 6.        Changes in CDR3 length distribution in CD45RA CD8 T cells after therapy. CDR3 lengths distributions prior to (0 weeks) and at 4 and 8 weeks after the initiation of combination antiretroviral therapy for a representative subject. Therapy resulted in a decline in virus load to undetectable levels (<50 copies/mL) by 4 weeks of treatment. A, Changes in CDR3 length distributions in 4 representative variable region (V)  families from CD45RA CD8 T cells. B, Therapy induced changes over time in the percentage of average distance (D) for the predominant peak (*) in each V family. FI, fluorescent intensity.

     To evaluate changes in TCR clonality after therapy, CDR3 products from amplification of V22 within CD8 CD45RA cells prior to and after 4 and 8 weeks of treatment were analyzed. In the representative subject shown in  prior to therapy, 80% of V22 CDR3 sequences were composed of an identical 12-aa length, which represented oligoclonality of a V22 to J2.5 recombination ( upper panel). The predominant 12-aa peak declined to 20% of the sequenced CDR3 lengths after only 4 weeks of therapy ( middle panel) and to 10% of the CDR3 sequences after 8 weeks of treatment ( lower panel). The new sequences identified within the emerging CDR3 lengths contained different amino acids from those detected at 4 weeks after therapy. The shift to a Gausssian distribution of CDR3 lengths resulted from emergence of new and more diverse CDR3 lengths.

fig.ommitted

Figure 7.        Analysis of the changes in clonotypes and CDR3 length-restricted patterns within CD45RA CD8 T cells after combination therapy. Polymerase chain reaction products for CDR3 lengths amplified, cloned, and sequenced from variable region (V) 22 CD45RA CD8 T cells from a human immunodeficiency virus–infected child receiving combination antiretroviral therapy. CDR3 lengths distribution are shown in the left side of the figure, and the V, CDR3, and joining region (J)  sequences with amino acid length and relative clonal frequency are shown on the right side. The changes CDR3 length and frequency of different clonotypes prior to therapy (0 weeks) and at 4 and 8 weeks after therapy are shown as the designated time points. FI, fluorescent intensity.

     Change in thymic output in HIV-infected children after antiretroviral therapy.     CD8 CD45RA T cells are a heterogeneous population consisting of both new thymic emigrants and terminally differentiated effector T cells [11]. To determine whether the changes in TCR diversity within this population were due to increased thymic output, the proportion of naive CD8 CD45RA CD11adim and relative number of CD45RA cells containing TREC were evaluated before and after treatment. Three subjects who had adequate sample availability were assessed. The relative proportions of CD8 CD45RA T cells expressing CD11adim (naive) or CD11abright (effector) was determined using flow cytometry analysis. Prior to therapy, the number and percentage of CD8 CD45RA expressing CD11adim in the 2 successfully treated subjects was 43% (220 cells/L) and 41% (210 cells/L), respectively . By 8–16 weeks after therapy, there was nearly a doubling in the number of CD45RA CD11adim CD8 T cells, as well as an increase in the percentage of cells to 68% (445 cells/L) and 62% (388 cells/L), respectively. The subject who failed to respond to antiretroviral treatment continued to display low numbers and percentages of CD11adim cells, with no change after treatment (31%–27%).

fig.ommitted

Figure 8.        Determination of T cell receptor excision circles (TRECs) within CD8 CD45RA T cells in human immunodeficiency virus (HIV)–infected children after combination therapy. Peripheral blood mononuclear cells (PBMC) were obtained before and after antiretroviral therapy from 3 HIV-infected children. Lanes 1 and 2, pretherapy and 8 weeks posttherapy samples from subject P1; lanes 3, 4, and 5, pretherapy and 4 and 8 weeks posttherapy samples from subject P3. Both children showed optimal control of viral replication after treatment. Lanes 6 and 7, pretherapy and 60 weeks posttherapy samples from a subject (P2) for whom virus replication was not suppressed with treatment. Lane 8, polymerase chain reaction–negative control. A, Changes in the absolute number (cells/L; top row) and percentages (bottom row) of CD8 CD45RA cells that express CD11adim, as determined using flow cytometry. B and C, Total cellular DNA from purified CD8 CD45RA T cells were amplified using primers for TREC coding joint and primers for -actin. B, Amplified products for TRECs; C, -actin. ND, not determined.

     To further assess changes in thymic output before and after therapy, an analysis of TRECs was performed using DNA from purified CD8 CD45RA T cells from the same 3 subjects. Therapy-induced virus suppression in subjects P1 and P3 resulted in increased detection of TRECs at 8–12 weeks after the initiation of treatment ( lanes 1–5). Increases in detectable TRECs were not seen in the subject (P2) for whom therapy failed ( lanes 6 and 7). Results of -actin amplification  indicated that similar amounts of input cellular DNA were amplified in all subjects. Together, these findings suggest that the reestablishment of TCR diversity in CD45RA cells is due to increases in thymic output of naive CD8 T cells.

DISCUSSION

     HIV-1 infection affects CD8 T cell development and function in several ways. Virus impairs thymic output, thus decreasing the number of thymic emigrants in both CD4 and CD8 T cell lineages [35]. In addition, ongoing viral replication induces a chronic state of T cell activation that leads to clonal expansion and, ultimately, clonal exhaustion due to persistent antigenemia [36]. In this way, HIV-1 skews CD8 T cell maturation and leads to an accumulation of activated CD8 T cells that cannot effectively control viremia [10]. Virus-induced aberrations within CD8 T cells are also evident when examining their TCR V repertoire [1, 2]. However, the impact of viral replication on the TCR V repertoire within CD8 T cells of different maturational stages and activation states has not been evaluated. The present study provides the first detailed examination of virus-induced alterations of the TCR diversity within different CD8 T cell subsets.

     We examined CDR3 length distribution within CD45RA and CD45RO CD8 T cells, because these CD45 isoforms provide the best initial dissection of the CD8 T cell subpopulations on the basis of activation state and stage of differentiation [12, 14, 37, 38]. Both CD8 and CD4 T cells undergo a postthymic differentiation that is driven by antigenic activation [39]. CD45RA CD8 T cells are composed of both naive T cells that have recently left the thymus and terminally differentiated effector T cells [12]. CD45RO CD8 T cells are functionally and phenotypically more homogenous and are composed primarily of activated and proliferating T cells [12]. Further subdivision of CD8 CD45RA CD8 T cells into naive and effector phenotypes can be made on the basis of the expression of CCR7, CD27, CD103, and CD11a [12, 14, 40]. Antigen activation results in the down-regulation of CCR7, CD107, and CD45RA and the up-regulation of CD11a and CD45RO [10, 12, 13, 38, 41, 42]. Differentiation into effector CD8 cells is associated with expression of CD11abright, reexpression of CD45RA, and loss of expression of CD27 and CD45RO [12, 37, 42, 43]. These cell-surface markers can be applied to further track CD8 T cell activation and differentiate naive T cells from activated effector cells.

     A comprehensive assessment of the differences in TCR CDR3 length diversity in different CD8 T cell subsets from both HIV-infected and healthy children has not been carried out; to our knowledge, our study is the first. Previous examinations of CDR3 length perturbations in CD8 T cells from healthy adults have found a limited degree of clonal expansions [2, 44, 45]. Furthermore, on the basis of comparisons between young and elderly adults, there appears to be an age-related increase in clonal T cell expansions in CD8 T cells [46]. Separation into CD45RA and CD45RO subsets enables a more sensitive evaluation of the cellular phenotypes undergoing clonal expansion [19]. In healthy children, CDR3 length distributions are predominantly Gaussian in both CD45RA and CD45RO CD8 T cells, although perturbations occasionally occur in either subset, an indication that clonal expansion can be detected in healthy children, perhaps in response to previous infection or antigenic stimulation.

     In comparison to the age-matched, healthy children, there is a 3–4-fold higher incidence of CDR3 length perturbations in both the CD45RA and CD45RO T cells in the HIV-infected group. The greater extent of perturbations within CD45RO cells may reflect the higher levels of activation and proliferation in this subset, compared with the CD45RA T cells, and a persistence of oligoclonal expansions within this subset [12]. Several different patterns of TCR perturbations were identified within the CD45 subpopulations from HIV-infected children. Perturbations could be seen in either CD45RA or CD45RO cells with the reciprocal population remaining Gaussian. In contrast to healthy children, HIV-infected children displayed a high proportion of V families that contained simultaneous perturbations in both CD45RA and CD45RO T cells, often due to the expansion of the same CDR3 length within an individual V family. In every instance, the predominant CDR3 length contained the same amino acid sequence in both subsets. It is unlikely that the oligoclonal expanison is due to a failure to adequately separate the subsets. There was no evidence of identical oligoclonal expansion in other perturbed V families from the same individual. This finding provides direct evidence that, during HIV infection, T cells expressing the same antigen receptor predominate in both CD45RA and CD45RO subsets. Concurrent expansion of the same T cell clone within both subpopulations supports the model of aberrant transition from CD8 CD45RO to CD8 CD45RA T cell phenotypes during the course of HIV-1 infection [10]. HIV-1–induced disruption of the normal distribution of CD8 TCR repertoire within CD45RA and CD45RO T cell subsets represents a persistent oligoclonal expansion during the course of disease that is likely to contribute to impaired cytotoxic CD8 T cell function [1, 10, 12, 13, 37].

     There is a striking correspondence among high virus load, low CD4 T cell count, and disruption of the V repertoire within CD45RA cells but not within CD45RO T cell subset. The results create a paradox if, as suggested by other investigators, the emergence of CD8 T cell clonal dominance is associated with greater control of viral replication [4]. We did not find that clonal dominance within either CD45RA or CD45RO CD8 T cells predicted better control of viral replication. The reasons for the discrepancy may reside in the phenotype of the expanded CD8 T cells. Further analysis of the phenotype and antigen specificity of the expanded CD8 T cells during the course of HIV infection may clarify this apparent discrepancy and provide important clues into the nature of the T cell response that best controls viral replication.

     The extent of CDR3 perturbations within CD45RA cells in the infected children was unexpected and may be due to distinct HIV-induced pathogenic processes in vivo. The first possibility is viral disruption of thymic ontogeny resulting in an aberrant TCR repertoire within naive CD8 T cells as they emigrate from the thymus [47]. The second possibility is that HIV-1 results in an abnormal expansion of antigen-primed effector CD45RA CD8 T cells [6, 10, 12, 38, 41]. The correlation between perturbations in CD45RA cells and high virus loads and advanced immune suppression supports the concept that HIV-induced T cell anergy within effector T cells contributes to the loss of T cell control of viral replication [6, 13, 48]. A more complete understanding of the virus-induced alterations in the immunoregulatory events that influence the changes in the TCR repertoire within these T cell subsets is an essential next step in unraveling the mechanism by which cytotoxic T lymphocytes control HIV-1 replication.

     After combination antiretroviral therapy with subsequent control of viral replication, there is a rapid reestablishment of TCR diversity within the CD45RA but not the CD45RO CD8 T cells. Early increases in CD8 T cells, B cells, and CD4 T cells after the initiation of antiretroviral therapy are thought to be due primarily to lymphocyte recirculation after treatment-induced declines in viral burden [8, 9, 49]. However, our previous studies showed that children display an early increase in naive CD4 T cells, as well as increases in total CD8 T cell numbers after therapy [28]. We were able to carry out a limited evaluation using a small number of subjects to determine the cellular origin of the CD8 CD45RA T cells that contributed to the reestablishment of TCR diversity. The assessment of the changes in TRECs and CD11adim expression within the CD45RA CD8 T cell subset demonstrated a dramatic increase in naive CD8 T cells. This finding supports a thymic origin of the new CDR3 lengths. The rapid restoration of the Gaussian distribution indicates that HIV-infected children, unlike adults, can rapidly restore their TCR repertoire after therapy [50]. However, sequence analysis of the CDR3 lengths within the CD45RA cells indicates that the oligoclonal expanded cells can persist for many weeks after therapy, even as TCR diversity within the CD45RA cells increases.

     Recent analysis of the HIV-specific cytotoxic T lymphocyte response has indicated that a large proportion of activated CD8 T cells are directed toward HIV-1 antigenic epitopes [5, 51, 52]. Our study shows that effective control of viral replication by antiretroviral therapy leads to a significant decrease in the levels of oligoclonal expansion within the V repertoire of the CD45RA but not CD45RO T cells. Early reestablishment of TCR diversity in CD45RA cells after optimal therapy may have prognostic significance in predicting whether an HIV-infected individual will be able sustain control viral replication after the initiation of treatment. The persistent perturbations in the CD45RO subset may reflect delayed reestablishment of the V repertoire in this subpopulation. Effective antiretroviral therapy that leads to a rapid decline and optimal control of viral replication results in increases in both CD4 and CD8 T cell populations [28, 49]. Therapy-induced increases in T cell numbers are also associated with an increase in TCR diversity and improved T cell function [53]. The present study expands these earlier observations by showing that HIV infection disrupts the TCR repertoire within both CD45RA and CD45RO CD8 T cell subpopulations. The degree of TCR perturbations within the CD45RA CD8 T cells correlates significantly with the degree of immune suppression and viral burden. In HIV-infected children, control of viral replication through combination antiretroviral therapy resulted in a rapid increase in TCR diversity within the CD45RA CD8 T cells, most likely because of early posttherapy increases in thymic output.

Acknowledgments

     We thank William G. Farmerie, Dan Brazeau, and Ginger Clark, for their help with the GeneScan analysis; Melissa Chen, for her assistance in the flow cytometry analysis; and Diana Nolte, for her help in preparing the manuscript.

References

　
1.  Pantaleo G, Demarest JF, Soudeyens H, et al. Major expansion of CD8+ T cells with a predominant V  usage during the primary immune response to HIV. Nature 1994; 370:463–7.
2.  Gorochov G, Neumann AU, Kereveur A, et al. Perturbation of CD4+ and CD8+ T-cell repertoires during progression to AIDS and regulation of the CD4+ repertoire during antiviral therapy. Nat Med 1998; 4:215–21.
3.  Wilson JD, Ogg GS, Allen RL, et al. Oligoclonal expansions of CD8+ T cells in chronic HIV infection are antigen specific. J Exp Med 1998; 188:785–90.
4.  Than S, Kharbanda M, Chitnis V, Bakshi S, Gregersen PK, Pahwa S. Clonal dominance patterns of CD8 T cells in relation to disease progression in HIV-infected children. J Immunol 1999; 162:3680–6.
5.  Walker PR, Ohteki T, Lopez JA, MacDonald HR, Maryanski JL. Distinct phenotypes of antigen-selected CD8 T cells emerge at different stages of an in vivo immune response. J Immunol 1995; 155:3443–52.
6.  Lewis DE, Tang DS, Adu-Oppong A, Schober W, Rodgers JR. Anergy and apoptosis in CD8+ T cells from HIV-infected persons. J Immunol 1994; 153:412–20.
7.  Luzuriaga K, Holmes D, Hereema A, Wong J, Panicali DL, Sullivan JL. HIV-1–specific cytotoxic T lymphocyte responses in the first year of life. J Immunol 1995; 154:433–43.
8.  Douek DC, McFarland RD, Keiser PH, et al. Changes in thymic function with age and during the treatment of HIV infection. Nature 1998; 396:690–5.
9.  Douek DC, Koup RA, McFarland RD, Sullivan JL, Luzuriaga K. Effect of HIV on thymic function before and after antiretroviral therapy in children. J Infect Dis 2000; 181:1479–82.
10.  Champagne P, Ogg GS, King AS, et al. Skewed maturation of memory HIV-specific CD8 T lymphocytes. Nature 2001; 410:106–11.
11.  McMichael AJ, Rowland-Jones SL. Cellular immune responses to HIV. Nature 2001; 410:980–7.
12.  Hamann D, Baars PA, Rep MH, et al. Phenotypic and functional separation of memory and effector human CD8+ T cells. J Exp Med 1997; 186:1407–18.
13.  Weekes MP, Carmichael AJ, Wills MR, Mynard K, Sissons JG. Human CD28-CD8+ T cells contain greatly expanded functional virus–specific memory CTL clones. J Immunol 1999; 162:7569–77.
14.  Sohen S, Rothstein DM, Tallman T, Gaudette D, Schlossman SF, Morimoto C. The functional heterogeneity of CD8+ cells defined by anti-CD45RA (2H4) and anti-CD29 (4B4) antibodies. Cell Immunol 1990; 128:314–28.
15.  Merkenschlager M, Terry L, Edwards R, Beverley PC. Limiting dilution analysis of proliferative responses in human lymphocyte populations defined by the monoclonal antibody UCHL1: implications for differential CD45 expression in T cell memory formation. Eur J Immunol 1988; 18:1653–61.
16.  Akbar AN, Salmon M, Janossy G. The synergy between naive and memory T cells during activation. Immunol Today 1991; 12:184–8.
17.  Slifka MK, Whitton JL. Activated and memory CD8+ T cells can be distinguished by their cytokine profiles and phenotypic markers. J Immunol 2000; 164:208–16.
18.  Sleasman JW, Leon BH, Aleixo LF, Rojas M, Goodenow MM. Immunomagnetic selection of purified monocyte and lymphocyte populations from peripheral blood mononuclear cells following cryopreservation. Clin Diagn Lab Immunol 1997; 4:653–8.
19.  Kou ZC, Puhr JS, Rojas M, McCormack WT, Goodenow MM, Sleasman JW. T-cell receptor V repertoire CDR3 length diversity differs within CD45RA and CD45RO T-cell subsets in healthy and human immunodeficiency virus–infected children. Clin Diagn Lab Immunol 2000; 7:953–9.
20.  Bogue M, Roth DB. V(D)J recombination. Curr Opin Immunol 1996; 8:175–80.
21.  Prochnicka-Chalufour A, Casanova JL, Avrameas S, Claverie JM, Kourilsky P. Biased amino acid distributions in regions of the T cell receptors and MHC molecules potentially involved in their association. Int Immunol 1991; 3:853–64.
22.  Toyonaga B, Yoshikai Y, Vadasz V, Chin B, Mak TW. Organization and sequences of the diversity, joining, and constant region genes of the human T-cell receptor  chain. Proc Natl Acad Sci USA 1985; 82:8624–8.
23.  Raaphorst FM, Kaijzel EL, van Tol MJ, Vossen JM, van den Elsen PJ. Non-random employment of V  6 and J  gene elements and conserved amino acid usage profiles in CDR3 regions of human fetal and adult TCR  chain rearrangements. Int Immunol 1994; 6:1–9.
24.  Pannetier C, Even J, Kourilsky P. T-cell repertoire diversity and clonal expansions in normal and clinical samples. Immunol Today 1995; 16:176–81.
25.  Kostense S, Raaphorst FM, Joling J, et al. T cell expansions in lymph nodes and peripheral blood in HIV-1–infected individuals: effect of antiretroviral therapy. AIDS 2001; 15:1097–107.
26.  Pantaleo G, Demarest JF, Schacker T, et al. The qualitative nature of the primary immune response to HIV infection is a prognosticator of disease progression independent of the initial level of plasma viremia. Proc Natl Acad Sci USA 1997; 94:254–8.
27.  Centers for Disease Control and Prevention. Revised classification system for human immunodeficiency virus infection in children less than 13 years of age. MMWR 1994; 43:1–12.
28.  Sleasman JW, Nelson RP, Goodenow MM, et al. Immunoreconstitution after ritonavir therapy in children with human immunodeficiency virus infection involves multiple lymphocyte lineages. J Pediatr 1999; 134:597–606.
29.  Aleixo LF, Goodenow MM, Sleasman JW. Molecular analysis of highly enriched populations of T-cell–depleted monocytes. Clin Diagn Lab Immunol 1995; 2:733–9.
30.  Sleasman JW, Aleixo LF, Morton A, Skoda SS, Goodenow MM. CD4+ memory T cells are the predominant population of HIV-1–infected lymphocytes in neonates and children. AIDS 1996; 10:1477–84.
31.  Chen ZW, Kou ZC, Shen L, et al. An acutely lethal simian immunodeficiency virus stimulates expansion of V7- and V14-expressing T lymphocytes. Proc Natl Acad Sci USA 1994; 91:7501–5.
32.  Chen ZW, Kou ZC, Lekutis C, et al. T cell receptor V  repertoire in an acute infection of rhesus monkeys with simian immunodeficiency viruses and a chimeric simian-human immunodeficiency virus. J Exp Med 1995; 182:21–31.
33.  Kou ZC, Halloran M, Lee-Parritz D, et al. In vivo effects of a bacterial superantigen on macaque TCR repertoires. J Immunol 1998; 160:5170–80.
34.  Chene G, Binquet C, Moreau JF, et al. Changes in CD4+ cell count and the risk of opportunistic infection or death after highly active antiretroviral treatment. Groupe d'Epidemiologie Clinique du SIDA en Aquitaine. AIDS 1998; 12:2313–20.
35.  McCune JM. The dynamics of CD4+ T-cell depletion in HIV disease. Nature 2001; 410:974–9.
36.  Lieberman J, Shankar P, Manjunath N, Andersson J. Dressed to kill? A review of why antiviral CD8 T lymphocytes fail to prevent progressive immunodeficiency in HIV-1 infection. Blood 2001; 98:1667–77.
37.  Faint JM, Annels NE, Curnow SJ, et al. Memory T cells constitute a subset of the human CD8+CD45RA+ pool with distinct phenotypic and migratory characteristics. J Immunol 2001; 167:212–20.
38.  Roos MT, van Lier RA, Hamann D, et al. Changes in the composition of circulating CD8+ T cell subsets during acute Epstein-Barr and human immunodeficiency virus infections in humans. J Infect Dis 2000; 182:451–8.
39.  Clement LT, Yamashita N, Martin AM. The functionally distinct subpopulations of human CD4+ helper/inducer T lymphocytes defined by anti-CD45R antibodies derive sequentially from a differentiation pathway that is regulated by activation-dependent post-thymic differentiation. J Immunol 1988; 141:1464–70.
40.  McFarland RD, Douek DC, Koup RA, Picker LJ. Identification of a human recent thymic emigrant phenotype. Proc Natl Acad Sci USA 2000; 97:4215–20.
41.  Altman JD, Moss PA, Goulder PJ, et al. Phenotypic analysis of antigen-specific T lymphocytes. Science 1996; 274:94–6.
42.  Fiorentini S, Licenziati S, Alessandri G, et al. CD11b expression identifies CD8+CD28+ T lymphocytes with phenotype and function of both naive/memory and effector cells. J Immunol 2001; 166:900–7.
43.  Hamann D, Kostense S, Wolthers KC, et al. Evidence that human CD8+CD45RA+CD27- cells are induced by antigen and evolve through extensive rounds of division. Int Immunol 1999; 11:1027–33.
44.  Fitzgerald JE, Ricalton NS, Meyer AC, et al. Analysis of clonal CD8+ T cell expansions in normal individuals and patients with rheumatoid arthritis. J Immunol 1995; 154:3538–47.
45.  Gea-Banacloche JC, Weiskopf EE, Hallahan C, et al. Progression of human immunodeficiency virus disease is associated with increasing disruptions within the CD4+ T cell receptor repertoire. J Infect Dis 1998; 177:579–85.
46.  Schwab R, Szabo P, Manavalan JS, et al. Expanded CD4+ and CD8+ T cell clones in elderly humans. J Immunol 1997; 158:4493–9.
47.  Kourtis AP, Ibegbu C, Nahmias AJ, et al. Early progression of disease in HIV-infected infants with thymus dysfunction N Engl J Med 1996; 335:1431–6 (erratum: N Engl J Med 1997; 336:595).
48.  Margolick JB, Munoz A, Donnenberg AD, et al. Failure of T-cell homeostasis preceding AIDS in HIV-1 infection. The Multicenter AIDS Cohort Study. Nat Med 1995; 1:674–80.
49.  Autran B, Carcelain G, Li TS, et al. Positive effects of combined antiretroviral therapy on CD4+ T cell homeostasis and function in advanced HIV disease. Science 1997; 277:112–6.
50.  Connors M, Kovacs JA, Krevat S, et al. HIV infection induces changes in CD4+ T-cell phenotype and depletions within the CD4+ T-cell repertoire that are not immediately restored by antiviral or immune-based therapies. Nat Med 1997; 3:533–40.
51.  Ogg GS, Jin X, Bonhoeffer S, et al. Quantitation of HIV-1–specific cytotoxic T lymphocytes and plasma load of viral RNA. Science 1998; 279:2103–6.
52.  Wilson JD, Ogg GS, Allen RL, et al. Direct visualization of HIV-1–specific cytotoxic T lymphocytes during primary infection. AIDS 2000; 14:225–33.
53.  Rosenberg ES, Billingsley JM, Caliendo AM, et al. Vigorous HIV-1–specific CD4+ T cell responses associated with control of viremia. Science 1997; 278:1447–50.


  
《传染病学杂志》2003年2月第187卷第3期