Secondary Logo

Journal Logo


CD4 T-cell hyperactivation and susceptibility to cell death determine poor CD4 T-cell recovery during suppressive HAART

Massanella, Martaa,*; Negredo, Eugeniab,*; Pérez-Álvarez, Nuriab,c; Puig, Jordib; Ruiz-Hernández, Raula; Bofill, Margaritaa,d; Clotet, Bonaventuraa,b; Blanco, Juliàa

Author Information
doi: 10.1097/QAD.0b013e328337b957
  • Free



Shortly after the introduction of HAART, it was reported that some treated HIV-infected patients showing satisfactory virological responses failed to increase their absolute counts of CD4 T cells [1]. The unexpected behavior of these patients, known as immunological nonresponders or discordant patients, has been widely studied from different viewpoints. On one hand, several groups have identified clinical predictors, such as the nadir CD4 T-cell count, for the default in CD4 T-cell recovery [2,3] and have reported how discordant responses negatively impact morbidity and mortality [4–6]. On the other hand, abundant work has aimed to understand the virological and immunological basis for this default. In general, reduced thymic function, resulting in low production of new T cells [7,8] and an increase in CD4 T-cell destruction revealed by a high sensitivity to ex-vivo cell death [7,8], has been shown to contribute to this failure (excellently reviewed in [5]). However, the ultimate cause of the discordant response and the relative contribution of poor T-cell production and increased destruction remain unknown. In contrast, it seems clear that the increased destruction of CD4 T cells is intimately related to immune hyperactivation, as activated cells are prone to undergo apoptosis ex vivo[9]. Thus, the simplest scenario suggests that homeostatic CD4 T-cell proliferation is activated in response to low naive T-cell production. However, other factors modulate CD4 T-cell turnover. In particular, bacterial translocation has been shown to increase in animal models after the ablation of thymic function [10] and in discordant patients [11]. Finally, the existence of low level [12,13] or latent undetectable ongoing viral replication [14] has also been described to influence the activation and death of CD4 T cells and, hence, may be a determinant in the discordant response.

A better knowledge of the factors influencing CD4 T-cell recovery could help to define optimal therapeutic strategies for poor immunological responders. Therefore, we sought to determine the individual contribution of bacterial translocation, CD4 T-cell production, activation and destruction to the discordant phenotype. In this study, we compared the perturbations in the CD4 and CD8 T-cell compartments in a large cohort of HIV-infected patients on stable suppressive HAART with satisfactory or unsatisfactory CD4 T-cell recovery. Different statistical approaches support the notion that immune activation and sensitivity to cell death, specifically in the CD4 T-cell compartment, are the main determinants of CD4 T-cell recovery and should be preferentially targeted in treatment strategies for discordant patients.

Patients and methods

Patients and samples

A cross-sectional, descriptive, comparative study was performed to analyze the immune status of patients with a discordant immune response. The Institutional Review Board of our center approved the study (EO code: EO-07-024), and all patients provided their written informed consent.

A total of 247 patients consecutively seen in our HIV Outpatient Unit during a period of 12 months (November 2007–November 2008) were screened for the study. Of the 247 patients, 230 met the inclusion criteria: confirmed diagnosis of HIV infection, continuous HAART with sustained undetectable HIV-1 RNA (plasma viral load <50 copies/ml) for at least the past 2 years (minimum of four determinations during this time period) and good antiretroviral treatment (ART) adherence. ART regimens included two nucleoside reverse transcriptase inhibitors (NRTIs)/nucleotide reverse transcriptase inhibitors and one boosted-ritonavir protease inhibitor or one non-NRTI (nevirapine or efavirenz). Chemotherapy or interferon/ribavirin treatments and a history of opportunistic infections or hepatic cirrhosis during the past 2 years were the exclusion criteria.

We considered patients concordant, with favorable virologic and immunologic responses, if they showed CD4 T-cell counts of more than 400 cells/μl. Conversely, discordant patients with a favorable virologic response, but an unfavorable immunologic response, persistently showed CD4 T cells less than 350 cells/μl. Clinical and demographic data were collected from medical records.

A 40 ml single blood extraction was performed on all of the participants. Blood was collected in Vacutainer tubes (Becton Dickinson, Madrid, Spain) and immediately processed either for immunophenotype or plasma and peripheral blood mononuclear cell (PBMC) isolation. Plasma was obtained by centrifugation of whole blood at 1200g for 10 min and immediately stored at −80°C until use. PBMCs were obtained from cell concentrates layered on Ficoll–Hypaque (Atom Reactiva, Barcelona, Spain) density gradients and either used immediately for ex-vivo cell death assays or stored in liquid nitrogen.

CD4 and CD8 T-cell immunophenotype

We stained fresh blood samples with the following antibody combinations: CD95–fluorescein isothiocyanate (FITC), PD-1–phycoerythrin (PE), human leukocyte antigen (HLA)-DR–peridinin chlorophyll protein complex (PerCP), CD3–allophycocyanin (APC)–Cy7, CD4–APC and CD8–PE–Cy7, which was designed to evaluate activation, and CD45RA–FITC, CD31–PE, CD38–PerCP, CD3–APC–Cy7, CD4–APC and CD8–PE–Cy7, which was designed to evaluate CD4 T-cell production (thymic output) and CD38 expression in CD45RA cells. An unstained control and a control antibody combination containing CD3–APC–Cy7, CD4–APC and CD8–PE–Cy7 were performed for all samples. All antibodies were from Becton Dickinson. Briefly, 20 μl of blood was incubated [15 min at room temperature] with the different antibody combinations in V-bottomed 96-well plates. After incubation, blood was lysed with 200 μl of FACS Lysing solution (Becton Dickinson) for 30 min at room temperature, washed in PBS and resuspended in PBS containing 1% formaldehyde. Cells were acquired in a LSRII flow cytometer (Becton Dickinson) coupled with a high-throughput screening (HTS) loader. At least 10 000 lymphocytes were collected for each sample and analyzed with FLowJo software (Tree Star, Inc., Ashland, Oregon, USA) using the following strategy: lymphocytes were gated according to morphological parameters, and CD3+ cells were identified and gated as CD4+CD8 or CD8+CD4. Double-positive cells were excluded from the analysis.

Evaluation of cell death

We cultured 200 000 PBMCs in Roswell Park Memorial Institute medium containing 10% fetal calf serum in 96-well plates for 24 h. At that time, cells were incubated with 40 nmol/l of the potentiometric mitochondrial probe DIOC6 (Molecular Probes; Invitrogen, Carlsbad, California, USA) and CD3–APC–Cy7, CD4–APC and CD8–PE–Cy7 antibodies for 1 h at 37°C. Samples were then acquired in a LSRII flow cytometer (Becton Dickinson) coupled with a HTS loader. At least 10 000 lymphocytes were collected for each sample and analyzed with FLowJo software (Tree Star, Inc.). After gating the CD3+CD4+ or the CD3+CD8+ subsets, dead cells were identified by their low DIOC6 fluorescence [15].

Determination of plasma lipopolysaccharide and soluble CD14

Lipopolysaccharide (LPS) levels were quantified in plasma samples diluted with endotoxin-free water (dilutions ranging from 1/10 to 1/100) and treated at 75°C for 10 min to inactivate plasma proteins. Samples were assayed in duplicate using a limulus amebocyte lysate assay (Charles River Laboratories International, Inc., Wilmington, Massachusetts, USA) according to the manufacturer's protocol.

Soluble CD14 (sCD14) levels, a surrogate marker of bacterial translocation [16], were quantified in plasma samples using an ELISA assay (Diaclone, Stamford, Connecticut, USA). Plasma samples were diluted (1/50) and tested in duplicate. Plasma from healthy individuals were used as a control and yielded lower levels of sCD14 (5.7 ± 1.6 μg/ml) than samples from HIV-infected patients (8.4 ± 2.1 μg/ml).

Statistical analysis

Continuous variables were expressed as the median (interquartile range) and compared using nonparametric tests (Mann–Whitney and Wilcoxon tests for nonpaired and paired data, respectively) because the parameters were not normally distributed. The Kruskal–Wallis test was used for multiple comparisons. Discrete variables were described as percentages (number of patients) and compared using the chi-squared or Fisher exact test. Pearson's correlation coefficient was calculated to assess the association between variables. The percentage of correct classification of the predicted response by the observed variable was used to assess the performance of the models. Univariate and multivariate logistic regressions were fitted to predict the probability of a discordant response. Statistical analyses were performed using SPSS software version 15.0 (SPSS, Inc., Chicago, Illinois, USA) with univariate two-tailed significance levels of 5%. Graphs were plotted with GraphPad Prism version 5 (GraphPad Software, Inc., San Diego, California, USA). P values are presented in scientific notation when lower than 0.0001 to precisely describe significant differences.


Patient characteristics

We included 230 patients in the study: 95 were defined as discordant and 135 were defined as concordant. Most of the demographic and clinical parameters were well balanced among both groups. However, minimal but significant differences were observed in CD8 T-cell absolute counts and in time on therapy. As expected, significantly lower levels of CD4 T-cell counts (absolute and percentage) and lower nadir CD4 T-cell counts were observed in discordant patients compared with concordant patients. These data are summarized in Table 1.

Table 1
Table 1:
Patient description.

Analysis of T-cell activation

Hyperactivation of CD8 T cells is a feature of HIV infection that determines both progression to AIDS in untreated patients and response to HAART in treated patients [17,18]. Therefore, we first analyzed different parameters of CD8 T-cell activation in our patients. The expression of the CD38 marker was significantly higher in discordant patients as compared with concordant patients (P = 1.55 × 10−6, Fig. 1a). However, as the percentage of CD38+CD45RA+ cells was similar in both groups (P = 0.188), differences were restricted to the CD45RA CD8 T-cell subset (P = 1.25 × 10−5, Fig. 1a). The expression of the death receptor Fas (CD95) and the activation marker HLA-DR as well as the frequency of activated, apoptosis-prone double-positive HLA-DR+CD95+ cells were also higher in CD8 T cells from discordant patients (P = 0.0027, P = 2.45 × 10−5 and P = 6.34 × 10−5, respectively, Fig. 1a).

Fig. 1
Fig. 1:
Analysis of activation markers in CD8 and CD4 T cells. (a) Frequency of CD38+ (upper left), CD38+CD45RA+ (upper middle) and CD38+CD45RA (upper right) in CD8 T cells. The percentage of CD95+ (lower left), HLA-DR+ (lower middle) and double HLA-DR+CD95+ (lower right) cells in the CD8 T-cell population is also shown. (b) Frequency of CD38+ (upper left), CD38+CD45RA+ (upper middle) and CD38+CD45RA (upper right) in CD4 T cells. The percentage of CD95+ (lower left), HLA-DR+ (lower middle) and double HLA-DR+CD95+ (lower right) cells in the CD4 T-cell population is also shown. Individual measurements of concordant (green symbols) and discordant patients (red symbols) with median values (black lines) and interquartile ranges (bars) are shown. P values were calculated according to the Mann–Whitney test. HLA-DR, human leukocyte antigen DR.

Apart from CD8 T-cell activation, the level of activation of CD4 T cells has also been correlated to disease progression and treatment success [19]. Importantly, CD4 T-cell activation is one of the major driving forces behind HIV replication. When we measured CD4 T-cell activation, the expression of CD38 was comparable in both groups. However, discordant patients showed a lower percentage of CD38+CD45RA+ CD4 T cells (P = 2.12 × 10−6, Fig. 1b) and a higher percentage of CD38+CD45RA CD4 T cells (P = 3.51 × 10−8, Fig. 1b), probably reflecting a reduced naive compartment and a higher level of activation in CD45RA cells. The hyperactivation in discordant patients was further confirmed by the increased frequency of CD95 or HLA-DR expressing cells and by the percentage of activated, apoptosis-prone HLA-DR+CD95+ cells (P = 7.19 × 10−10, P = 2.71 × 10−15 and P = 8.24 × 10−18, respectively, Fig. 1b). Similar results were observed when PD-1, a death receptor used as a marker of cell exhaustion [20], was analyzed. The percentage of PD-1+ cells and HLA-DR+PD-1+ cells was also higher in discordant patients (P = 0.0033 and P = 1.75 × 10−11, respectively, data not shown). In summary, these data confirm the existence of perturbations in T-cell homeostasis that strongly affect the CD4 T-cell compartment.

Analysis of CD4 T-cell production and destruction: relationship with activation

New CD4 T cells are produced by the thymus and expanded either by cytokine-driven proliferation or postthymic selection. The CD31 antigen can be used to identify these mechanisms of naive T-cell homeostasis. The CD45RA+CD31+ cells contain 95% of total T-cell receptor excision circles (TRECs) in CD4 T cells [21] and have been defined as thymic-naive CD4 T cells [22], although it may also contain cells that proliferate in response to cytokines [23]. Conversely, the CD4+CD45RA+CD31 population has been used as surrogate marker of cells expanded by postthymic selection, as their TREC content is lower as compared with the CD4+CD45RA+CD31+ population [21]. When analyzed in our cohort, the frequency of CD4+CD45RA+ cells was lower in discordant patients (P = 6.68 × 10−6, Fig. 2a). Lower levels of CD4+CD45RA+CD31+ cells (P = 0.0002, Fig. 2a) and CD4+CD45RA+CD31 cells (P = 1.19 × 10−5 data not shown) were also observed in discordant patients. However, the frequency of CD31 cells in the CD45RA+CD4+ T cells showed no significant difference between groups (Fig. 2a). Consistent with previous analyses [7,8], these data may indicate that despite similar levels of postthymic selection, limited CD4 T-cell production or cytokine-driven proliferation impedes the proper maintenance of the naive CD4 T-cell pool in discordant patients.

Fig. 2
Fig. 2:
Analysis of CD4 T-cell production and destruction. (a) Percentages of CD45RA+ (upper left) and CD45RA+CD31+ (recent thymic emigrants, upper middle) in the CD4 T-cell population. Percentage of CD31 in the CD45RA+ CD4 T-cell population was used as a surrogate marker of peripheral naive cell expansion (upper right). Cell death that occurred in the 24-h ex-vivo PBMC cultures was analyzed in the CD4 (lower left) and CD8 T-cell subsets (lower right). Individual measurements of concordant (green symbols) and discordant patients (red symbols) with median values (black lines) and interquartile ranges (bars) are shown. P values were calculated according to the Mann–Whitney test. (b) Dot plots of thymic production (CD4+CD45RA+CD31+) and activation (CD4+HLA-DR+CD95+, left) and ex-vivo CD4 T-cell death (right). Red and green symbols correspond to discordant and concordant patients, respectively. Pie charts show the percentage of discordant (red) and concordant patients (green) in each quadrant. HLA-DR, human leukocyte antigen DR; PBMC, peripheral blood mononuclear cell

We next analyzed the level of spontaneous cell death occurring in CD4 and CD8 T cells in ex-vivo PBMC cultures. As described by others for reduced cohorts [7], discordant patients showed higher levels of CD4 T-cell death, which reached a high degree of significance (P = 2.5 × 10−12, Fig. 2a). In contrast, CD8 T cells from concordant and discordant patients showed no significant differences in sensitivity to ex-vivo cell death (Fig. 2a) despite differences observed in activation status.

To confirm the role of CD4 T-cell production and destruction in the recovery of CD4 T cells, we stratified all patients (n = 230) according to their increase in CD4 T cells, which was measured as the difference between their current and nadir CD4 T-cell counts (see Figure, Supplemental Digital Content 1, Higher increases in CD4 T cells were associated with higher thymic output and lower cell activation and death (P = 7.9 × 10−5, P = 1.5 × 10−10 and P = 8.9 × 10−12, respectively). However, none of these parameters was different in patients with the lowest CD4 T-cell recovery (increase of <100 CD4 T cells/μl) compared with patients with a CD4 T-cell gain between 100 and 200 cells/μl (see Figure, Supplemental Digital Content 1, Thus, to further explore the contribution of each factor to CD4 T-cell recovery, we performed a double analysis of correct classification rates. In this analysis, data on CD4 T-cell production from concordant and discordant patients were plotted against CD4 T-cell activation or death (Fig. 2b). In each case, four quadrants were defined according to the median values of each parameter. As expected, patients showing low CD4 T-cell production and high activation/death were mostly discordant, whereas patients showing high CD4 T-cell production and low activation/death were mostly concordant. Patients with high activation or death values were mostly discordant patients, irrespective of their level of thymic output (upper quadrants). Conversely, patients with low activation or death values were mostly Concordant in both the low and high thymic production groups (lower quadrants, Fig. 2b). This analysis indicates that the discordant phenotype is better identified by activation and death values than by CD4 T-cell production.

Bacterial translocation

Having identified CD4 T-cell activation as a major determinant of discordant response, we next evaluated its relationship with bacterial translocation. Because LPS measurements in plasma yielded hardly detectable and inconsistent values, we evaluated the level of sCD14 as a surrogate marker of bacterial translocation. This marker of monocyte activation showed higher levels in discordant patients (P = 0.002, Fig. 3a) and an inverse correlation with the percentage of CD4 T cells (Fig. 3b), but appeared to be unrelated to the percentage of thymic CD45RA+CD31+ CD4 T cells (data not shown). In contrast, sCD14 positively correlated with CD4 T-cell activation (CD38+CD45RA and HLA-DR+CD95+ CD4 T cells) and cell death (Fig. 3b). However, the correlation coefficients were low (r < 0.3) in all cases. Furthermore, similar levels of sCD14 were found in the different groups of patients defined by the increase in CD4 T cells (P = 0.652, see Figure, Supplemental Digital Content 1,, suggesting a secondary role for bacterial translocation in CD4 T-cell recovery.

Fig. 3
Fig. 3:
Plasma levels of soluble CD14. (a) Individual measurements of sCD14 in plasma from concordant (green symbols) and discordant patients (red symbols) with median values (black lines) and interquartile ranges (bars) are shown. P values were calculated according to the Mann–Whitney test. (b) Correlation of sCD14 data from all patients with the percentage of CD4 T cells (upper left), the frequency of activated CD4 T cells measured as CD38+CD45RA cells (upper right) or HLA-DR+CD95+ cells (lower left) and the extent of ex-vivo CD4 T-cell death (lower right). Pearson's correlation coefficients and P values are indicated in each plot. HLA-DR, human leukocyte antigen DR; sCD14, soluble CD14.

Analysis of the predictors of discordant response

In a final analysis, we assessed the odds ratio of being discordant for each measured parameter (Fig. 4a). First, a univariate analysis ruled out any effect of demographic parameters (age and sex) and confirmed the predominant role of the nadir CD4 T-cell count as predictor of discordance. CD8 T-cell death was unrelated to immune recovery, whereas activation markers in the CD8 T-cell subset were significant but poorly predictive of the discordant phenotype. However, the same parameters measured in CD4 T cells appeared to be more relevant. Although CD4 T-cell production (CD45RA+CD31+ cells) and CD38 expression in CD45RA cells also showed significant but poor predictive values; the percentage of CD4+HLA-DR+CD95+ cells, CD4 T-cell death and sCD14 were the best immunological predictors of the discordant condition. The impact of the nadir CD4 T-cell count was confirmed by categorizing all patients into high (n = 84) and low (n = 146) nadir groups (cutoff 200 cells/μl). Patients with low nadir showed higher rate of discordance (58 vs. 12% in the high nadir group, P = 6.5 × 10−12), lower CD4 T-cell counts (absolute and percentage, P = 2.8 × 10−14 and P = 3.8 × 10−12, respectively), higher levels of sCD14 (P = 0.007) and activation in both CD4 (HLA-DR+CD95+ cells, P = 4.9 × 10−5) and CD8 T cells (CD38+CD45RA, P = 0.001). Interestingly, no differences were found in the frequency of CD45RA+CD31+ CD4 T cells (P = 0.376) between both groups.

Fig. 4
Fig. 4:
Predictive factors for unsatisfactory immune recovery. The ability of the indicated parameters to predict the probability of discordance was assessed in univariate (a) and multivariate analyses, which included most of parameters from the univariate analysis (b). Symbols denote OR values, and lines indicate 95% CIs. OR values of Nadir CD4 T cells were calculated for decreases of 50 cells/μl. Asterisks denote statistical significance, * P < 0.05, ** P < 0.005. CI, confidence interval; HLA-DR, human leukocyte antigen DR; OR, odds ratio.

To definitively ascertain the role of activation and death in CD4 T-cell recovery, we performed a multivariate analysis in which the discordant condition was analyzed with respect to CD4 T-cell production (CD45RA+CD31+ cells), expansion (CD45RA+CD31 cells), activation (HLA-DR+CD95+ cells) and death. This analysis identified activation and death as the best predictors of a discordant response (data not shown). The analysis was further refined by adding nadir CD4 T-cell count, age, plasma sCD14 and CD8 T-cell activation to the regression model. Nadir CD4 T-cell values and, again, ex-vivo CD4 T-cell death and activation were the only predictors for a discordant response to HAART (Fig. 4b). Taken together, our data suggest that CD4 T-cell recovery during HAART seems to be a matter of increased destruction rather than reduced production of CD4 T cells.


Extensive characterization of virologically suppressed, poor immunological responders, HIV-infected individuals has provided compelling evidence for decreased thymic production and increased activation and apoptosis [5]. However, the contribution of each one of these factors to poor CD4 T-cell recovery remains unsolved. Our experimental approach was designed to answer this question; the number of patients analyzed gave the study the necessary power to address the relative weight of CD4 T-cell production and destruction. First, we validated our patient cohort by identifying previously described criteria for the discordant response. Indeed, nadir CD4 T-cell count, CD4 T-cell production (CD45RA+CD31+CD4 T cells), CD8 and CD4 T-cell activation (HLA-DR and/or CD95 expression and CD38 expression in CD45RA cells) and CD4 T-cell sensitivity to ex-vivo cell death showed significant differences between concordant and discordant patients.

This first analysis revealed that perturbations in T-cell homeostasis that were observed in discordant patients mainly affected CD4 T cells. Consistently, univariate analyses indicated that the differences in the CD8 T-cell compartment were much less predictive of discordance than those observed in CD4 T cells. The analysis of this latter subset confirmed that the production of new cells (CD4+CD45RA+CD31+) was not a determinant factor for a discordant response. This observation is consistent with the fact that age, which inversely correlated with CD4+CD45RA+CD31+ cells, showed no predictive value in the univariate analysis. Conversely, activation markers (CD4+ HLA-DR+CD95+ cells) and the extent of ex-vivo cell death, along with nadir CD4 T-cell counts, were the best predictors for an unsatisfactory immune reconstitution.

Bacterial translocation, coinfections, HAART composition and cryptic viral replication are some of the factors that may explain the increased activation and death of CD4 T cells in discordant patients. Bacterial translocation was analyzed as a first candidate because a low nadir CD4 T-cell count could be associated with deteriorated gut-associated lymphoid tissue, high circulating bacterial bioproducts [10] and, consequently, increased CD4 T-cell activation [24] and apoptosis. However, the level of plasma sCD14 in our patients, which was a good predictor for a discordant response in the univariate analysis, showed no relevance in the multivariate approach. Correlation analyses showed no relationship between the level of sCD14 and the recovery of CD4 T cells or the frequency of naive cells, although some degree of correlation was observed with the percentage of CD4 T cells, the expression of activation markers and ex-vivo CD4 T-cell death. Taken together, these data do not support a specific role of naive T cells in maintaining immune function in the gut [10] and reveal a weak impact of bacterial bioproducts on immune activation in chronically infected, virologically suppressed patients. Thus, bacterial translocation failed to fully explain the discordant immune recovery.

Increased activation and death of CD4 T cells in discordant patients may rely on coinfections [25] or the effects of HAART on cell viability [26,27]. However, we ruled out these possibilities in a parallel analysis in which we showed no effect of hepatitis C virus infection or drug regimens on CD4 T-cell recovery [28]. Thus, the ultimate cause of immune discordance remains elusive. In this regard, the role of cryptic viral replication in discordant responses may be worth exploring and might explain some markers associated with immune discordance (frequency of CD4+CCR5+ T cells, cellular proviral load, monocyte turnover and function). Indeed, CD8 T-cell activation in virologically suppressed patients appears to be independent of residual viremia [29], whereas particularly sensitive to cryptic viral replication (Buzon et al., in preparation). Therefore, it is possible that in discordant patients, higher CD4 T-cell activation and proviral DNA [8] favor cryptic HIV replication in lymphoid tissue, limiting CD4 T-cell recovery and normalization of CD8 T-cell activation. The analysis of this hypothesis is strongly limited by the availability of appropriate technical tools. In the absence of data on viral replication in lymphoid tissue in our patients, ultrasensitive plasma viral load could be used as an alternate surrogate marker [14]. However, it is still unclear whether this parameter, which also quantifies continuous reservoir draining, actually reflects ongoing viral replication [14,29–32]. Currently, clinical intensification strategies and an active search of new markers of viral replication (episomal HIV DNA, activation markers) may be required to definitively ascertain the role of viral replication in discordant responses.

Alternative approaches to identify additional immunological factors not detected by our cross-sectional analysis include retrospective analysis of patients reaching satisfactory or unsatisfactory immune recovery from a low nadir CD4 T-cell count.


We show that CD4 T-cell destruction that is driven by cellular activation is the major cause of poor immune recovery in discordant patients. The ultimate causes of CD4 T-cell activation appear to be intimately associated with the nadir CD4 T-cell count, but do not rely exclusively on homeostasis and bacterial bioproducts.


This work was supported by unrestricted grants from GILEAD and MERCK and by funds from the FIPSE project 06/3600, the Spanish AIDS network ‘Red Temática Cooperativa de Investigación en SIDA (RD06/0006)’ and the European network NEAT. We thank all the patients who participated in the study.

J.B. is a researcher from Fundació Institut de Recerca en Ciències de la Salut Germans Trias i Pujol and is supported by the ISCIII and the Health Department of the Catalan Government (Generalitat de Catalunya). M.M. is supported by a predoctoral grant from Generalitat de Catalunya and the European Social Fund. This work is part of the PhD thesis that M.M. conducts at Pompeu Fabra University.

M.M. designed and performed immunophenotype, cell death and sCD14 assays. N.P.-A. performed statistical analyses. J.P. recruited patients and collected clinical data. M.B., R.R.-H. and B.C. designed immunophenotype panels and interpreted data. E.N. and J.B. designed, coordinated the work and wrote the manuscript. All authors read and approved the final version of the manuscript.

There are no conflicts of interest.


1. Piketty C, Castiel P, Belec L, Batisse D, Si Mohamed A, Gilquin J, et al. Discrepant responses to triple combination antiretroviral therapy in advanced HIV disease. AIDS 1998; 12:745–750.
2. Florence E, Lundgren J, Dreezen C, Fisher M, Kirk O, Blaxhult A, et al. Factors associated with a reduced CD4 lymphocyte count response to HAART despite full viral suppression in the EuroSIDA study. HIV Med 2003; 4:255–262.
3. Kaufmann GR, Bloch M, Finlayson R, Zaunders J, Smith D, Cooper DA. The extent of HIV-1-related immunodeficiency and age predict the long-term CD4 T lymphocyte response to potent antiretroviral therapy. AIDS 2002; 16:359–367.
4. Moore RD, Gebo KA, Lucas GM, Keruly JC. Rate of comorbidities not related to HIV infection or AIDS among HIV-infected patients, by CD4 cell count and HAART use status. Clin Infect Dis 2008; 47:1102–1104.
5. Gazzola L, Tincati C, Bellistri GM, Monforte A, Marchetti G. The absence of CD4+ T cell count recovery despite receipt of virologically suppressive highly active antiretroviral therapy: clinical risk, immunological gaps, and therapeutic options. Clin Infect Dis 2009; 48:328–337.
6. Piketty C, Weiss L, Thomas F, Mohamed AS, Belec L, Kazatchkine MD. Long-term clinical outcome of human immunodeficiency virus-infected patients with discordant immunologic and virologic responses to a protease inhibitor-containing regimen. J Infect Dis 2001; 183:1328–1335.
7. Benveniste O, Flahault A, Rollot F, Elbim C, Estaquier J, Pedron B, et al. Mechanisms involved in the low-level regeneration of CD4+ cells in HIV-1-infected patients receiving highly active antiretroviral therapy who have prolonged undetectable plasma viral loads. J Infect Dis 2005; 191:1670–1679.
8. Marchetti G, Gori A, Casabianca A, Magnani M, Franzetti F, Clerici M, et al. Comparative analysis of T-cell turnover and homeostatic parameters in HIV-infected patients with discordant immune-virological responses to HAART. AIDS 2006; 20:1727–1736.
9. Gougeon ML, Lecoeur H, Dulioust A, Enouf MG, Crouvoiser M, Goujard C, et al. Programmed cell death in peripheral lymphocytes from HIV-infected persons: increased susceptibility to apoptosis of CD4 and CD8 T cells correlates with lymphocyte activation and with disease progression. J Immunol 1996; 156:3509–3520.
10. Bourgeois C, Hao Z, Rajewsky K, Potocnik AJ, Stockinger B. Ablation of thymic export causes accelerated decay of naive CD4 T cells in the periphery because of activation by environmental antigen. Proc Natl Acad Sci U S A 2008; 105:8691–8696.
11. Marchetti G, Bellistri GM, Borghi E, Tincati C, Ferramosca S, La Francesca M, et al. Microbial translocation is associated with sustained failure in CD4+ T-cell reconstitution in HIV-infected patients on long-term highly active antiretroviral therapy. AIDS 2008; 22:2035–2038.
12. Kitchen CM, Philpott S, Burger H, Weiser B, Anastos K, Suchard MA. Evolution of human immunodeficiency virus type 1 coreceptor usage during antiretroviral Therapy: a Bayesian approach. J Virol 2004; 78:11296–11302.
13. Vaamonde CM, Hoover DR, Anastos K, Tan T, Shi Q, Gao W, et al. Factors associated with poor immunologic response to virologic suppression by highly active antiretroviral therapy in HIV-infected women. AIDS Res Hum Retroviruses 2006; 22:222–231.
14. Mavigner M, Delobel P, Cazabat M, Dubois M, L'Faqihi-Olive FE, Raymond S, et al. HIV-1 residual viremia correlates with persistent T-cell activation in poor immunological responders to combination antiretroviral therapy. PLoS One 2009; 4:e7658.
15. Blanco J, Barretina J, Clotet B, Este JA. R5 HIV gp120-mediated cellular contacts induce the death of single CCR5-expressing CD4 T cells by a gp41-dependent mechanism. J Leukoc Biol 2004; 76:804–811.
16. Brenchley JM, Price DA, Schacker TW, Asher TE, Silvestri G, Rao S, et al. Microbial translocation is a cause of systemic immune activation in chronic HIV infection. Nat Med 2006; 12:1365–1371.
17. Bofill M, Mocroft A, Lipman M, Medina E, Borthwick NJ, Sabin CA, et al. Increased numbers of primed activated CD8+CD38+CD45RO+ T cells predict the decline of CD4+ T cells in HIV-1-infected patients. AIDS 1996; 10:827–834.
18. Giorgi JV, Hultin LE, McKeating JA, Johnson TD, Owens B, Jacobson LP, et al. Shorter survival in advanced human immunodeficiency virus type 1 infection is more closely associated with T lymphocyte activation than with plasma virus burden or virus chemokine coreceptor usage. J Infect Dis 1999; 179:859–870.
19. Barretina J, Blanco J, Bonjoch A, Llano A, Clotet B, Este JA. Immunological and virological study of enfuvirtide-treated HIV-positive patients. AIDS 2004; 18:1673–1682.
20. Day CL, Kaufmann DE, Kiepiela P, Brown JA, Moodley ES, Reddy S, et al. PD-1 expression on HIV-specific T cells is associated with T-cell exhaustion and disease progression. Nature 2006; 443:350–354.
21. Kimmig S, Przybylski GK, Schmidt CA, Laurisch K, Mowes B, Radbruch A, Thiel A. Two subsets of naive T helper cells with distinct T cell receptor excision circle content in human adult peripheral blood. J Exp Med 2002; 195:789–794.
22. Kohler S, Thiel A. Life after the thymus: CD31+ and CD31 human naive CD4+ T-cell subsets. Blood 2009; 113:769–774.
23. Azevedo RI, Soares MV, Barata JT, Tendeiro R, Serra-Caetano A, Victorino RM, Sousa AE. IL-7 sustains CD31 expression in human naive CD4+ T cells and preferentially expands the CD31+ subset in a PI3K-dependent manner. Blood 2009; 113:2999–3007.
24. Catalfamo M, Di Mascio M, Hu Z, Srinivasula S, Thaker V, Adelsberger J, et al. HIV infection-associated immune activation occurs by two distinct pathways that differentially affect CD4 and CD8 T cells. Proc Natl Acad Sci U S A 2008; 105:19851–19856.
25. Nunez M, Soriano V, Lopez M, Ballesteros C, Cascajero A, Gonzalez-Lahoz J, Benito JM. Coinfection with hepatitis C virus increases lymphocyte apoptosis in HIV-infected patients. Clin Infect Dis 2006; 43:1209–1212.
26. Negredo E, Molto J, Burger D, Viciana P, Ribera E, Paredes R, et al. Unexpected CD4 cell count decline in patients receiving didanosine and tenofovir-based regimens despite undetectable viral load. AIDS 2004; 18:459–463.
27. Badley AD. In vitro and in vivo effects of HIV protease inhibitors on apoptosis. Cell Death Differ 2005; 12(Suppl 1):924–931.
28. Negredo E, Massanella M, Puig J, Pérez-Álvarez N, Gallego-Escudero JM, Villarroya J, et al.Poor CD4 T-cell recovery in virologically suppressed HIV-infected patients is predicted by nadir CD4 T-cell count and determined by high CD4 T-cell intrinsic apoptosis. Clinical implications.Clin Infec Dis 2010; in press.
29. Steel A, Cox AE, Shamji MH, John L, Nelson M, Henderson DC, et al. HIV-1 viral replication below 50 copies/ml in patients on antiretroviral therapy is not associated with CD8+ T-cell activation. Antivir Ther 2007; 12:971–975.
30. Ruiz L, van Lunzen J, Arno A, Stellbrink HJ, Schneider C, Rull M, et al. Protease inhibitor-containing regimens compared with nucleoside analogues alone in the suppression of persistent HIV-1 replication in lymphoid tissue. AIDS 1999; 13:F1–F8.
31. Dinoso JB, Kim SY, Wiegand AM, Palmer SE, Gange SJ, Cranmer L, et al. Treatment intensification does not reduce residual HIV-1 viremia in patients on highly active antiretroviral therapy. Proc Natl Acad Sci U S A 2009; 106:9403–9408.
32. Shiu C, Cunningham CK, Greenough T, Muresan P, Sanchez-Merino V, Carey V, et al. Identification of ongoing human immunodeficiency virus type 1 (HIV-1) replication in residual viremia during recombinant HIV-1 poxvirus immunizations in patients with clinically undetectable viral loads on durable suppressive highly active antiretroviral therapy. J Virol 2009; 83:9731–9742.

bacterial translocation; cell death; discordant patients; immune activation; thymus

Supplemental Digital Content

© 2010 Lippincott Williams & Wilkins, Inc.