The advent of highly active antiretroviral therapy (HAART) resulted in dramatic changes in the disease course of HIV, but long-term therapy is limited by the requirement of strict daily adherence, the occurrence of drug toxicity, elevated treatment costs, and sustained risks of the emergence of drug resistance . In this context, structured treatment interruption may help to alleviate some of the problems associated with continuous HAART. However, the data to justify structured treatment interruption as a safe and effective approach remain controversial [2,3]. HIV-specific CD8 T lymphocytes are important in the antiviral response . In chronically HIV-infected patients, the expanded HIV-specific cytotoxic lymphocytes (CTL) persist at high frequencies and 1–2% of all circulating CD8 T cells selectively recognize a dominant HIV epitope . However, these CTL fail to control the infection completely. Interestingly, a skewed maturation of HIV-specific CTL has recently been demonstrated, indicating that the HIV-specific CTL pool is largely composed of pre-terminally differentiated CCR7− CD45RA− CD8+ T lymphocytes with poor cytotoxic activities . CCR7− effector T cells have lost the ability for autocrine proliferation, display specialized function such as IFN-γ production, and acquire mature cytolytic function after CD45RA expression [7,8]. Moreover, T cell activation induces a transient increase in CD27 expression that gradually switches off on effector cells . Altogether, monitoring both CCR7 and CD27 surface expression allows one to discriminate among different effector or long-term memory T cell subsets.
In this ‘pilot’ study, we analysed the influence of structured treatment interruption on effector/memory dynamics of CD8 T cell subsets in chronically HIV-infected patients showing a rapid or delayed viral rebound. Moreover, the CD8 T cell reactivities to mitogen and HIV-Gag peptide stimulation were compared.
Materials and methods
Asymptomatic chronically HIV-1-infected patients were recruited from the National Institute for Infectious Diseases ‘Lazzaro Spallanzani'. The criteria for inclusion in the study were: (a) HAART (at least 2 years of successful virus suppression); (b) stable CD4 T cell counts above 500 cells/μl for at least 12 months before entry; (c) undetectable HIV-RNA plasma levels (< 50 copies/ml by branched DNA) for at least 12 months before entry; (d) HAART discontinuation for side-effects or spontaneous request. Two patients treated with dual nucleoside reverse transcriptase inhibitors (NRTI) who fulfilled these requirements were included in the study. All selected patients underwent a single-cycle structured treatment interruption consisting of at least one month of discontinuation. HAART re-introduction was decided according to the current guidelines on antiretroviral therapy (CD4 T cells < 350 cells/μl or HIV RNA > 30 000 copies/ml). Twenty-six patients completed the study protocol. Two groups were selected on the basis of plasma viral rebound and were included in the present analysis. Group A (n = 14) included patients with a rapid viral rebound (within one month) and group B (n = 6) was composed of patients with a delayed or no viral rebound (after 4 months). Six patients presenting with intermediate behaviour were not considered. A rapid HIV-RNA rebound greater than 30 000 copies/ml was observed within 28.7 ± 1.8 days in group A patients. In group B, three patients showed no HIV-RNA rebound after 507.3 ± 199.5 days and three patients presented with an HIV-RNA rebound greater than 30 000 copies/ml at 159.3 ± 21.5 days. A clinical and immunological follow-up in group A was performed at the time of HAART suspension (t 0), after one month from suspension (t 1), at the resumption of HAART (t 2), and after 30 days from HAART resumption (t 3). Group B patients were monitored at HAART suspension (t 0) and after every month during suspension (t 1a, t 1b, t 1c, etc.). All patients gave written informed consent and the protocol was approved by the Institute's Ethical Committee.
Monocolonal antibodies and flow cytometry
Monoclonal antibodies (mAb) coupled with fluorescein, phycoerythrin, phycoerythrin-cyanin 5.1 (PE-Cy5) and allophycocyanin were combined for simultaneous staining. The anti-human antibodies used in this study were: anti-CD4 (IgG1, clone RPA-T4), anti-CD8 (IgG1, clone RPA-T8), anti-CD27 (IgG1, clone M-T271), anti-CD45RA (IgG2b, clone HI100), anti-IFN-γ mAb (IgG1, clone B27), anti-perforin mAb (IgG2b, clone δG9), and control mAb (IgG1 clone MOPC-21). The purified anti-CCR7 (IgM, clone 2H4) was detected using biotin-conjugated rat anti-mouse IgM (IgG2a, clone R6-60.2) and streptavidin phycoerythrin. All mAb were obtained from Becton Dickinson, Mountain View, CA, USA. Stainings for membrane or intracellular antigens were performed as described . Control for non-specific staining was monitored with isotype-matched mAb and non-specific staining was always subtracted from the specific results. Flow cytometric analysis was performed on a Facscalibur flow cytometer (Becton Dickinson). At least 100 000 live events were acquired, gated on small viable lymphocytes. Data files were analysed using CellQuest software (Becton Dickinson).
Cell isolation and stimulation
Peripheral blood mononuclear cells were obtained using standard Ficoll–Hypaque (Pharmacia, Uppsala, Sweden) density centrifugation and frozen in dimethyl sulphoxide 10% and fetal calf serum 90% at −80°C. Briefly, 1 × 106 thawed cells in 1 ml of complete RPMI 1640, 10% v/v heat-inactivated fetal calf serum, 2 mM l-glutamine, 50 U/ml penicillin and 50 μg/ml streptomycin, were incubated with 1 μg each of anti-CD28 and CD49d mAb (IgG1 clone CD28.2, and IgG1 clone 9F10, Becton Dickinson) and pooled Gag peptides (1 μg of each peptide) or with phorbol myristate acetate (50 ng/ml) and ionomycin (10 μg/ml). The pool of Gag peptides included 28 different 15-mers and was designed on the most conserved area of Gag gene product  and was purchased from Sigma-Genosys Cambridge, UK. Cells incubated with only anti-CD28 and CD49d were included in every experiment as control samples. The cultures were finally incubated at 37°C in a 5% carbon dioxide incubator for 1 h, followed by an additional 5 h incubation in the presence of 10 μg/ml brefeldin-A to inhibit cellular exocytosis (Sigma, St Louis, MO, USA).
Sequence analysis of Gag HIV-1 protein and peptide design
HIV-1 Gag protein sequences were downloaded from the GeneBank protein database. A total of 814 protein sequences were analysed and aligned using Antheprot software . The consensus sequences of the most conserved areas were submitted for the identification of HLA peptide-binding motifs on the HIV Immunology Database (http://hiv-web.lanl.gov/immunology/) . Subsequently, the areas of Gag protein identified as containing a large number of peptide-binding motifs were analysed using the quantitative implemented HLA peptide-binding motif databases syfpeithi (http://www.syfpeithi.de)  and bimas (http://bimas.dctr.nih.gov/molbio/hla_bind/)  for prediction and scoring of each putative epitope for all available HLA class I alleles in the two databases. A further analysis was performed by paproc (http://www.paproc.de)  in order to identify the most probable area of proteasome processing to design properly overlapping peptides containing more than one binding profile. At the end of the analysis each selected 15-mer contained one or more epitopes putatively able to bind, with a minimum of 30% of the maximum binding of any allele belong to an HLA-class I serological specificity, at least two different HLA serological specificities for each HLA class I locus (A, B or C) and recognized by HLA class I gene products from two different loci.
The peptides were purchased from Sigma-Genosys (Cambridge, UK) as free amino acids. All synthetic peptides were purified by reverse-phase chromatography to more than 90% purity. Sequence and purity were confirmed by mass spectrometry and analytical reverse-phase chromatography. Lyophilized peptides were resuspended in dimethyl sulphoxide at stock concentrations of 10 mg/ml for each peptide.
Differences among groups were evaluated using Wilcoxon's paired test.
Clinical, immunological and virological parameters of HIV structured treatment interruption study groups
HIV structured treatment interruption patients were divided into two groups as described in the Methods section (Table 1): one group of patients with a rapid viral rebound (group A) and a second group with a delayed or no viral rebound (group B). In group A (Fig. 1a), the viral rebound was paralleled by a significant decrease in CD4 T cells (t 0 versus t 1, 734 ± 51 cells/μl and 502 ± 38 cells/μl, respectively, P = 0.001, see Table 2). Interestingly, the CD8 T cell subset was significantly increased (t 0 versus t 1, 949 ± 80 cells/μl and 1338 ± 135 cells/μl, respectively, P = 0.01), showing that CD8 T cells are rapidly expanded during HIV replication. After 42.3 ± 23.7 days of structured treatment interruption (t 2), the viral and immunological parameters were similar to the situation described at t 1. The reintroduction of HAART was able effectively to suppress plasma viraemia, restoring the CD8 T cell numbers to the starting point. Moreover, CD4 T cells were reduced after structured treatment interruption and therapy reintroduction (t 3 versus t 0, P = 0.02). In contrast, group B showed no significant CD4 or CD8 T cell modulation after one month of structured treatment interruption (CD4 cell count t 0 versus t 1, 947 ± 196 cells/μl and 903 ± 176 cells/μl, respectively; CD8 cell count t 0 versus t 1, 926 ± 146 cells/μl and 985 ± 129 cells/μl, respectively; Fig. 1b and Table 2). Neither in group A nor group B have we observed any significant change in CD4 T cell effector/memory phenotypes, or in the CD4 T cell reactivities in terms of proliferation or IFN-γ production in response to HIV-1 p24 antigen or to cytomegalovirus protein (data not shown).
CD8 T cell dynamics during structured treatment interruption
As CD8 T cells were rapidly expanded in group A, the frequency of Gag-specific CD8 T lymphocytes was measured as IFN-γ production by flow cytometry. The number of Gag-specific CD8 T cells was significantly increased by HAART interruption (Fig. 2a, t 0 versus t 1, P = 0.01), indicating that the expanded pool of CD8 T cells contains HIV-specific CD8 T lymphocytes releasing IFN-γ. At t 2, the plasma viral load was still high and was similar to the situation at t 1 (Fig. 1a). In contrast, the frequency of Gag-specific CD8 T lymphocytes dropped back to the baseline levels observed in the absence of viral replication during HAART (t 1 versus t 2, P = 0.03, Fig. 2a), suggesting an exhaustion of the circulating pool of HIV-specific CD8 T cells in the absence of viral suppression. After HAART resumption, the frequency of Gag-specific CD8 T lymphocytes was significantly augmented (t 0 versus t 3, P = 0.01), confirming that structured treatment interruption may influence the number of HIV-specific CD8 T cells [17,18]. As shown in Fig. 2a, the CD8 T cell response to mitogen stimulation was also increased by HAART interruption (t 0 versus t 1, P = 0.01), suggesting that the expanded pool of CD8 T cells is functional and able to release IFN-γ upon stimulation. At t 2, the expanded mitogen-reactive CD8 T cell pool dropped back to baseline levels observed in the absence of viral replication during HAART (t 0), similar to Gag-specific CD8 T lymphocytes, confirming a general exhaustion of the circulating CD8 T cell pool. No significant changes in CD8 T cell reactivity were observed in group B patients independently of HAART interruption, with a stable baseline level of 0.5 ± 0.2 HIV Gag specific CD8 T cells/μl and 363.3 ± 181.4 phorbol myristate acetate–ionomycin-reactive CD8 T cells/μl.
Fig. 2b shows the frequency of HIV-specific CD8 T lymphocytes releasing IFN-γ from a representative donor (t 1, DA18). Moreover, Fig. 2c shows that the expression of CD45RA and CCR7 molecules on gated CD8 T cells (R2, Fig. 2b) allows for the discrimination of four specific CD8 T cell subsets: (i) CD45RA+ CCR7+ naive cells; (ii) CD45RA− CCR7+ long-term memory cells; (iii) CD45RA− CCR7− pre-terminally differentiated effector cells; and (iv) CD45RA+ CCR7− terminally differentiated effector cells. In contrast, gating on IFN-γ-producing HIV-specific CD8 T cells (R3, Fig. 2b) revealed that this subset mainly expressed the CD45RA− CCR7− phenotype of pre-terminally differentiated effector cells (94.2%, Fig. 2d). As shown in Fig. 2e, the general expansion of CD8 T cells after structured treatment interruption was largely composed of immature CD8 cells expressing the CD45RA− CCR7− phenotype (Fig. 2e, t 0 versus t 1, P = 0.03). Interestingly, almost half of the accumulated CD45RA− CCR7− CD8+ T cells also expressed the CD27 molecules (Fig. 2f), supporting a block at specific stages of CD8 T cell differentiation including CD27+ CD45RA− long-term memory cells. No significant changes in effector/memory CD8 T cell subsets were observed in group B patients independently of HAART interruption, with a stable baseline level of 667 ± 235 CCR7− CD45RA− CD8 T cells/μl, 345 ± 36 CCR7+ CD45RA− CD8 T cells/μl, 8 ± 3 CCR7− CD45RA+ CD8 T cells/μl, and 10 ± 7 CCR7+ CD45RA+ CD8 T cells/μl.
Both in HIV-infected individuals (groups A and B) and in healthy donors, the intracellular amount of perforin was ≅ 25% lower in CD45RA− CCR7− CD8 T cells than in terminally differentiated CD45RA+ CCR7− CD8 T lymphocytes (Fig. 3) as previously reported by others .
The majority of chronically infected patients undergoing structured treatment interruption have not been able to achieve durable virological control . However, most of the HIV-infected population residing in developing countries have no access to continuous antiretroviral treatment, and chronically HIV-infected patients treated with antiretroviral therapy may experience long-term side-effects, underlining the importance of reducing the patient's reliance on HAART.
HIV-specific CTL are involved in the control of HIV-replication . In HIV chronic infection, an inverse correlation between CTL and viral load was observed . Both virus and CTL decline to low levels after successful antiretroviral therapy . These observations suggest that the maintenance of circulating HIV-specific CTL effectors is dependent on persistent antigenic stimuli. Structured treatment interruption has thus been proposed as a boost for the immune system . Recent studies have shown that both CD4 and CD8 T cell responses increase after the viral rebound induced by structured treatment interruption. However, this immunity was only transient, and decreased after HAART resumption as a normal memory response to the withdrawal of antigen [17,18]. Accordingly, we observed that the frequencies of CD8 T cells releasing IFN-γ after mitogen-induced or Gag-specific stimulation were both significantly increased during HAART interruption. In particular, the frequency of HIV-specific IFN-γ−producing CD8 T cells were greater at t 3 after the reintroduction of HAART than at t 0. However, IFN-γ production by CD8 T cells was associated with the failure of viral control, suggesting that the quality but not the quantity of HIV-specific CD8 T lymphocytes may determine the outcome of the antiviral response.
A high viral burden in the presence of major HIV-specific CD8 T cell expansion is present in most HIV-infected patients, which suggests possible CTL effector function impairment . Similarly, a functional deterioration of SIV-specific CD8 T cells was found during the chronic phase of infection . In HIV-infected patients, the analysis of CD8 T cell effector phenotypes showed that the HIV-specific CD8 T cell pool is predominantly composed of pre-terminally differentiated CD45RA− CCR7− cells, demonstrating a skewed maturation of HIV-specific memory CD8 Tcells . Our data show that the intracellular amount of perforin was approximately 25% lower in CD45RA− CCR7− CD8 T cells than in terminally differentiated CD45RA+ CCR7− CD8 T lymphocytes. This observation was consistent with previously reported data . Moreover, a large fraction of these HIV-specific CD8 T cells express significantly lower levels of perforin that is linked with CD27 persistent expression, confirming an impaired maturation and defective cytolytic activity . In this context, our data indicate that an accumulation of pre-terminally differentiated CCR7− CD27+/− CD45RA− CD8 T cells was associated with the viral rebound during structured treatment interruption, demonstrating that failure to control viraemia is related to the accumulation of CD8 T cells at a specific stage of differentiation.
In chronically HIV-infected patients, the immune system is progressively damaged, and the quality of immune reconstitution appears to be different from that obtainable during acute infection . Most structured treatment interruption patients experiencing a rapid viral rebound have sustained HIV-specific CD8 T cells producing IFN-γ, but the majority of these cells present a phenotype that has been shown not to be functionally cytotoxic, leaving the patient with high numbers of non-functional virus-specific CD8 T cells. This impaired CTL function and accumulation of pre-terminally differentiated CD8 T cells may be a consequence of an increased turnover of terminally differentiated CD8 T cells or a lack of antigen-specific CD4 T cell helper activity. Adjuvant therapies, such as the use of immune modulators, may restore CTL effector functions and improve the quality of the antiviral immune response.
The authors would like to thank Dr E. Girardi, Professor M. Malkovsky and Dr M.Wallace for critical reading of the manuscript; M. Brescia for technical assistance and M. Lupi, G. Cianca, C. Vasile, L. Bolognesi and D. Menna for clinical support.
Sponsorship: This study has been supported by grants from the Ministero della Salute and Istituto Superiore della Sanità (grant nos. 30C23 and 45D/1.21).
1.Markowitz M. Resistance, fitness, adherence, and potency: mapping the paths to virologic failure.JAMA
2.Lori F, Lisziewicz J. Structured treatment interruptions for the management of HIV infection.JAMA
3.Montaner LJ. Structured treatment interruptions to control HIV-1 and limit drug exposure.Trends Immunol
4.McMichael AJ, Rowland-Jones SL. Cellular immune responses to HIV.Nature
5.Ogg GS, Jin X, Bonhoeffer S, Dunbar PR, Nowak MA, Monard S, et al
. Quantitation of HIV-1-specific cytotoxic T lymphocytes and plasma load of viral RNA.Science
6.Champagne P, Ogg GS, King AS, Knabenhans C, Ellefsen K, Nobile M, et al
. Skewed maturation of memory HIV-specific CD8 T lymphocytes.Nature
7.Campbell JJ, Butcher EC. Chemokines in tissue-specific and microenvironment-specific lymphocyte homing.Curr Opin Immunol
8.Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions.Nature
9.Hintzen RQ, de Jong R, Lens SM, Brouwer M, Baars P, van Lier RA. Regulation of CD27 expression on subsets of mature T-lymphocytes.J Immunol
10.Amicosante M, Gioia C, Montesano C, Casetti R, Topino S, D'offizi G, et al
. Computer-based design of an HLA-haplotype and HIV-clade independent CTL assay for monitoring HIV-specific immunity.Mol Med
11.Betts MR, Casazza JP, Koup RA. Monitoring HIV-specific CD8+ T cell responses by intracellular cytokine production.Immunol Lett
12.Deleage G, Combet C, Blanchet C, Geourjon C. antheprot: an integrated protein sequence analysis software with client/server capabilities.Comput Biol Med
13.Brander C, Walker BD. T lymphocyte responses in HIV-1 infection: implications for vaccine development.Curr Opin Immunol
14.Rammensee H, Bachmann J, Emmerich NP, Bachor OA, Stevanovic S. syfpeithi: database for MHC ligands and peptide motifs.Immunogenetics
15.Parker KC, Bednarek MA, Coligan JE. Scheme for ranking potential HLA-A2 binding peptides based on independent binding of individual peptide side-chains.J Immunol
16.Nussbaum AK, Kuttler C, Hadeler KP, Rammensee HG, Schild H. paproc: a prediction algorithm for proteasomal cleavages available on the World Wide Web.Immunogenetics
17.Ortiz GM, Wellons M, Brancato J, Vo HT, Zinn RL, Clarkson DE, et al
. Structured antiretroviral treatment interruptions in chronically HIV-1-infected subjects.Proc Natl Acad Sci U S A
18.Mollet L, Li TS, Samri A, Tournay C, Tubiana R, Calvez V, et al
. Dynamics of HIV-specific CD8+ T lymphocytes with changes in viral load. The RESTIM and COMET Study Groups.J Immunol
19.Hamann D, Baars PA, Rep MH, Hooibrink B, Kerkhof-Garde SR, Klein MR, et al.Phenotypic and functional separation of memory and effector human CD8+ T cells.J Exp Med
20.Miller V. Structured treatment interruptions in antiretroviral management of HIV-1.Curr Opin Infect Dis
21.Betts MR, Krowka JF, Kepler TB, Davidian M, Christopherson C, Kwok S, et al.Human immunodeficiency virus type 1-specific cytotoxic T lymphocyte activity is inversely correlated with HIV type 1 viral load in HIV type 1-infected long-term survivors.AIDS Res Hum Retroviruses
22.Kostense S, Ogg GS, Manting EH, Gillespie G, Joling J, Vandenberghe K, et al
. High viral burden in the presence of major HIV-specific CD8(+) T cell expansions: evidence for impaired CTL effector function.Eur J Immunol
23.Vogel TU, Allen TM, Altman JD, Watkins DI. Functional impairment of simian immunodeficiency virus-specific CD8+ T cells during the chronic phase of infection.J Virol
24.Appay V, Nixon DF, Donahoe SM, Gillespie GM, Dong T, King A, et al
. HIV-specific CD8(+) T cells produce antiviral cytokines but are impaired in cytolytic function.J Exp Med