Share this article on:

Effects of HIV-1 Infection on Lymphocyte Phenotypes in Blood Versus Lymph Nodes

Yang, Otto O MD*†; Ferbas, John J PhD*; Hausner, Mary Ann MS*; Hultin, Lance E BS*; Hultin, Patricia M*; McFadden, David MD‡; Sawicki, Mark MD‡; Detels, Roger MD§; Majchrowicz, Martin MPH*; Matud, Jose L MS*; Giorgi, Janis V PhD*†∥; Jamieson, Beth D PhD*

JAIDS Journal of Acquired Immune Deficiency Syndromes: 15 August 2005 - Volume 39 - Issue 5 - pp 507-518
Basic Science

Most immunopathogenesis studies of HIV-1 use peripheral blood. Most lymphocytes reside in lymphoid tissues, however, and the extent to which blood mirrors tissues is unclear. Here, we analyze lymphocytes in blood and lymph nodes of HIV-1-uninfected and -infected persons. Baseline comparison of node and blood lymphocytes in seronegative persons demonstrates a lower ratio of CD8+ versus CD4+ T lymphocytes, a lower number of effector cells (CD28) within the CD8+ compartment, and greater activation (D-receptor [DR+]) within the CD4+ compartment. In infected versus uninfected persons, nodes exhibit elevated CD8+ T lymphocytes with an increased memory-effector phenotype (CD62L/CD45RA) and activation (CD38+ and DR+) but minimal differences in the CD4+ compartment. Changes attributable to HIV-1 infection are markedly greater in node lymphocytes than in blood. Comparisons of CD8+ T-lymphocyte parameters and viremia in infected persons reveal positive correlations of CD38+ expression on cells in blood and nodes and a negative correlation of terminal effector cells (CD62L/CD45RA+) in the nodes to viremia. Multiple linear regression analysis indicates that CD38 expression on node (not blood) CD8+ T lymphocytes is the sole independent predictor for viremia. Thus, blood indirectly reflects processes in lymphoid tissues, and caution should be applied when interpreting immunopathogenesis studies of blood.

From the *Department of Medicine, Geffen School of Medicine, UCLA Medical Center, University of California, Los Angeles, CA; †Department of Microbiology, Immunology, and Molecular Genetics, University of California, Los Angeles, CA; ‡Department of Surgery, Geffen School of Medicine, UCLA Medical Center, University of California, Los Angeles, CA; and §UCLA School of Public Health, University of California, Los Angeles, CA.


Received for publication February 11, 2005; accepted April 27, 2005.

Supported by National Institute of Allergy and Infectious Diseases grants AI037613 (J. V. Giorgi) and AI043203 (O. O. Yang).

Reprints: Otto O. Yang, Division of Infectious Diseases, 37-121 CHS, UCLA Medical Center, 10833 LeConte Avenue, Los Angeles, CA 90095 (e-mail:

The accessibility of peripheral blood has limited most HIV-1 pathogenesis studies in humans to assays of blood, under the assumption that this compartment adequately reflects the overall interactions of virus, target cells, and immune responses. Indeed, valuable correlations have been determined, such as the usefulness of the blood CD4+ T-lymphocyte concentration in predicting the risk of infection and viremia level in predicting the rate of disease progression. Such findings have suggested that peripheral blood can serve as a window for viewing systemic pathologic processes in HIV-1 infection.

The CD4+ T lymphocytes that are the predominant source of HIV-1 replication in vivo mostly reside in lymphoid tissues, however. It has been estimated that only 2% of the total lymphocyte population is found in peripheral blood1; therefore, the blood is probably a minor site of viral replication within the host. Data suggest that more vigorous viral replication occurs in lymphoid tissues than is reflected by the degree of viremia,2,3 indicating that tissues are the major reservoir of viral replication. Furthermore, infected CD4+ T lymphocytes are rare in the peripheral circulation even when viremia is high,4 suggesting that virion production occurs outside this compartment.

Although CD8+ T lymphocytes are believed to have an important role in reducing HIV-1 replication in vivo, quantitative correlates of immunity have remained elusive. Qualitative experiments have shown that CD8+ T-lymphocyte depletion of simian immunodeficiency virus (SIV)-infected macaques results in uncontrolled viral replication in vivo,5-7 but comprehensive attempts to correlate the magnitude or breadth of the HIV-1-specific CD8+ T-lymphocyte response in blood to viremia have been disappointing.8,9 A more global measure of CD8+ T lymphocytes, the activation status of these cells (in blood) as reflected by CD38 expression, has been noted to correlate with viremia and to predict disease progression.10,11 Viremia is a better predictor of disease than CD38 expression on CD8+ T lymphocytes,12 however, and the relative relationships of CD38 expression, CD8+ T-lymphocyte activity, and viral replication are unknown.

An important factor that is poorly understood is the extent to which measurements of CD8+ lymphocytes in blood reflect antiviral activity at the major sites of viral replication. Although some data indicate that HIV-1-specific CD8+ T-lymphocyte targeting is similar between blood and lymph node compartments,13 the potential differences remain poorly understood. Here, we perform detailed assessments of T-lymphocyte subpopulations and compare blood and lymph nodes in HIV-1-infected and -uninfected persons.

Back to Top | Article Outline


Study Subjects

The HIV-1-seropositive subjects were men in the Los Angeles arm of the Multicenter AIDS Cohort Study (MACS), with the exception of 2 women (Table 1). The seronegative control subjects were men in the MACS cohort with recent and past high-risk exposure to HIV-1 selected on the basis of high infection-risk exposure, defined as exceeding the number of anal insertive partners corresponding to the 90th percentile of all seronegative men in the MACS during the 2.5-year interval from 1982 through 1985, and continued high-risk exposure, as indicated by reported sexual behavior at MACS visits from 1993 through 1994, where exposures are defined as unprotected anal receptive intercourse with an HIV-1-infected sexual partner. High-risk seronegative subjects were chosen as the appropriate control group, because the sexual exposure status of individuals (independent of HIV-1 infection) has been correlated with changes in T-lymphocyte subsets compared with low-risk persons.14 All subjects provided informed consent under University of California Institutional Review Board (IRB)-approved protocols.

Back to Top | Article Outline

Collection of Peripheral Blood Mononuclear Cells

Peripheral blood mononuclear cells (PBMCs) were obtained from heparinized blood specimens within 1 hour of phlebotomy by density gradient centrifugation on Lymphoprep (Nycomed Pharma, Oslo, Norway), and all processing for immunologic assays was completed within 6 hours.

Back to Top | Article Outline

Isolation of Lymph Node Mononuclear Cells

Superficial inguinal lymph nodes were removed surgically through sterile operative procedures for subsequent isolation of mononuclear cells. Each node was bisected, and the cut surface of each half was mechanically disrupted through 50-gauge stainless steel mesh. The resulting lymph node mononuclear cell (LNMC) suspension was passed through a 70-μm cell strainer (Becton Dickinson, Franklin Lakes, NJ).

Back to Top | Article Outline

HIV-1 Quantitation in Blood

HIV-1 RNA in plasma was quantitated by using the Amplicor HIV-1 Quantitation Kit (Roche, Branchburg, NJ) per the manufacturer's recommended protocol.

Back to Top | Article Outline

Phenotypic Analysis of T Lymphocytes by Flow Cytometry

T-cell subset phenotyping by cell surface staining of freshly isolated cells was performed as previously described.15 All monoclonal antibodies were purchased as conjugates of fluorescein isothiocyanate (FITC), phycoerythrin (PE), or peridinin chlorophyll protein (PerCP) from Becton Dickinson Immunocytometry Systems (San Jose, CA). Two-color (FITC/PE) or 3-color (FITC/PE/PerCP) antibody combinations were used to stain for expression of CD3/CD4, CD3/CD8, CD4/CD28/CD3, CD8/CD28/CD3, CD45RA/CD62L/CD4, CD45RA/CD62L/CD8, human leukocyte antigen (HLA)-D-related (DR)/CD38/CD4, and HLA-DR/CD38/CD8 and were immediately analyzed on a FACScan flow cytometer (BDIS, San Jose, CA). Cell surface CD38 molecule numbers were estimated as previously reported,16 using a bead standard. To ensure accurate quantitation of staining, the cytometer was standardized for fluorescence intensity each day using glutaraldehyde-fixed chicken red blood cells (Biosure, Grass Valley, CA).

For intracellular detection of perforin, 150 μL of whole blood or 5 × 105 node cells were first surfaced stained with anti-CD3-PE and anti-CD8-PerCP. After surface staining, OrthoPermaFix solution (Ortho Diagnostics, Raritan, NJ) was added for 40 minutes at room temperature, followed by 2 washes. The cells were then stained for 40 minutes at room temperature with perforin-FITC or IgG2-FITC isotype control (Ancell, Bayport, MN) in the presence of human AB serum. The cells were washed twice and run immediately on a FACSCalibur flow cytometer (BDIS). Because of high variability in background staining, only CD8+ T cells that expressed high levels of perforin were graded as positive for this study. The cutoff was determined in previous control studies by gating on CD28/CD8+ natural killer (NK) lymphocytes, which express high levels of perforin.

Back to Top | Article Outline

Statistical Analysis

Most comparisons between HIV-1-seropositive and -seronegative groups were performed using bootstrap resampling,17 adjusting traditional t test P values (designated Padj) for multiple comparisons (14 variables for each CD4+ and CD8+ T-lymphocyte subset measured in peripheral blood and lymph node sample types), interdependence of certain variables, and nonnormal distribution of several variables (Table 2). All compared arms included at least 7 independent measurements for each group (see Table 2). Comparisons were performed using the Multitest procedure in SAS (SAS Institute Cary, NC). This program also was used to evaluate for correlations among immune markers in the peripheral blood, lymph nodes, and viremia (Table 3). For these comparisons, viremia was quantified as the logarithm10 of HIV-1 genomes per milliliter of plasma; measurements in infected persons below the limit of detection were assigned a value of 2 log10 units. Identified correlations were confirmed by additional testing using a Markov Chain/Monte Carlo simulation (WinBUGS; Imperial College of Science, Technology, and Medicine, London, England), in which values for viremia below the limit of detection (400 copies/mL of plasma) were considered censored data in the range of 0 to 400.18

Back to Top | Article Outline


HIV-1 Infection Preferentially Affects CD8+ but Not CD4+ T-Lymphocyte Phenotypes in Peripheral Blood

Several markers on the peripheral blood CD8+ and CD4+ T lymphocytes of HIV-1-seronegative and -seropositive persons were assessed and compared. The seronegative persons were highly exposed yet uninfected men from the MACS serving as controls for the infected persons, most of whom were also men from the MACS (see Table 1). The peripheral CD8+ T-lymphocyte pool was larger in the seropositive versus the seronegative persons (median ± SD absolute CD8+ T-lymphocyte counts of 1098 ± 473 cells/mm3 vs. 719 ± 281 cells/mm3; P = 0.024, Padj = 0.173), and the CD4+ T-lymphocyte pool was significantly smaller (524 ± 213 cells/mm3 vs. 812 ± 200 cells/mm3; P = 0.001, Padj = 0.016). This was also reflected in the percentages of CD8+ (P < 0.001, Padj < 0.001) and CD4+ (P = 0.001, Padj = 0.010) cells within the total lymphocyte pool (Fig. 1). These populations then were analyzed individually for subsets of CD62L/CD45RA, CD28, and CD38/DR expression.

Within the blood CD8+ T-lymphocyte compartment (see Fig. 1A), the major difference was the activation status. The memory and naive populations defined by CD62L/CD45RA were similar between seropositive and seronegative persons. Differentiation of effector cells, as reflected by CD28 status, was also similar. The activation status, as defined by CD38/DR markers of activation, showed that the CD38+ subsets were higher in seropositive persons, however, as well as the total percentage of cells that were CD38+ (P = 0.003, Padj = 0.029). In contrast to CD38, total DR+ cell percentages showed a weak trend toward higher DR+ cells in seropositive persons, possibly suggesting differential activation attributable to infection.

Interestingly, the blood CD4+ T-lymphocyte compartment (see Fig. 1B) was more similar between seropositive and seronegative subjects. Again, memory and naive subsets and CD28 status were similar. There was a trend toward higher CD38+ but not DR+ subsets in the seropositive subjects (P = 0.039, Padj = 0.275 for total CD38+ percentage). These data indicated that HIV-1 infection results in more dramatic changes within the CD8+ T-lymphocyte compartment than within the CD4+ T-lymphocyte compartment, suggesting preferential responsiveness of CD8+ T lymphocytes to infection.

Back to Top | Article Outline

CD8+ and CD4+ T-Lymphocyte Concentrations and Phenotypes Vary Between Peripheral Blood and Lymph Nodes in Uninfected Subjects

To define the baseline status of T lymphocytes in lymph nodes in the absence of perturbation by HIV-1 infection, the subpopulations of CD8+ and CD4+ T lymphocytes were compared in the blood and lymph nodes of the HIV-1-seronegative control subjects (Fig. 2). Overall, the percentages of CD8+ T lymphocytes were lower and those of CD4+ T lymphocytes were higher in lymph nodes compared with blood (Padj < 0.001 for each). In the CD8+ compartment (see Fig. 2A), the CD62L/CD45RA and CD38/DR subsets were similar, but a markedly lower percentage of CD28 effector cells were found in lymph nodes versus blood (Padj < 0.001). By contrast, the CD4+ compartment showed more differences (see Fig. 2B), revealing higher CD38+/DR+ (P = 0.001, Padj = 0.012), lower CD38/DR (P = 0.006, Padj = 0.042), higher CD38/DR+ (P = 0.002, Padj = 0.018), and higher total DR+ (P = 0.001, Padj = 0.012) cell percentages in lymph nodes, suggesting more activation marked by DR but not CD38 expression. Although there was a suggestion for a trend of more CD28/CD4+ effector cells in blood (P = 0.0269, Padj = 0.132), this difference was less marked than that noted for CD8+ T lymphocytes. Overall, these observations suggested that lymph nodes contain a lower percentage of CD8+ T lymphocytes than blood and these tend to be CD28+ and that lymph nodes contain a higher percentage of CD4+ T lymphocytes than blood and these tend to be activated as demonstrated by DR but not CD38 expression.

Back to Top | Article Outline

HIV-1 Infection Preferentially Increases Activated Effector Subsets of CD8+ but Not CD4+ T Lymphocytes in Lymph Nodes

These cell subsets were also evaluated in lymph nodes from HIV-1-seropositive persons to examine the effects of infection. The percentage of CD8+ T lymphocytes overall was increased (Padj < 0.001), and there were marked differences among differentiation and activation parameters of lymph node CD8+ T lymphocytes as a result of infection (Fig. 3A). Percentages of CD8+ memory-effector lymphocytes as defined by CD62L/CD45RA (P = 0.003, Padj = 0.023) and CD28 (P = 0.009, Padj = 0.060)19-21 phenotypes were increased. Activation, as represented by percentages of CD38+ (Padj < 0.001) and DR+ (Padj = 0.023) cells, was increased, in large part because of a larger CD38+/DR+ subset (Padj < 0.001). By contrast, there was a lower percentage of total CD4+ T lymphocytes in the lymph nodes of infected versus uninfected persons (Padj < 0.001), but the phenotypes were remarkably similar, with the exception of an elevated percentage of the CD38+/DR+ (P = 0.001, Padj = 0.016) subset (Figure 3B). These findings further confirmed greater phenotypic changes in the CD8+ than CD4+ T lymphocyte compartment attributable to HIV-1 infection.

Back to Top | Article Outline

HIV-1-Induced Changes in CD8+ T-Lymphocyte Phenotypes are Much Greater in Lymph Nodes Than in Blood

To compare quantitatively the degree of change in various T-lymphocyte subsets in infected versus uninfected persons, the ratio of each subset in infected versus uninfected persons was calculated for CD8+ (Fig. 4A) and CD4+ (Fig. 4B) T lymphocytes in the blood and lymph nodes. This analysis revealed that the changes in the CD8+ T-lymphocyte compartment attributable to infection occurred in the same subsets in the blood and lymph nodes but that they were of greater magnitude in the lymph nodes. In particular, the percentages of CD28, CD38+/DR+, and total CD38+ subsets were increased many fold in the lymph nodes versus the blood. There were fewer changes in the CD4+ T lymphocytes, which, overall, were similar between the blood and lymph nodes.

Back to Top | Article Outline

Concentration of Perforin-Expressing CD8+ T Lymphocytes in Blood and Lymph Nodes of Infected Subjects Is Not Less Than That of Uninfected Subjects

Because reduced perforin expression in lymph node CD8+ T lymphocytes has been proposed as a mechanism of immune failure in HIV-1 infection,22-26 this parameter was assessed in the blood and lymph nodes of the subjects (Fig. 5). Intracellular staining of CD8+ T lymphocytes in the blood and lymphs node of uninfected persons revealed markedly lower percentages of perforin-expressing cells in the lymph nodes compared with the blood (mean of 0.2% vs. 27.2% in the 4 tested subjects; P = 0.007). In HIV-1-infected persons, there was a similar relation between the lymph nodes and blood (mean of 1.5% vs. 35.9% in the 11 tested subjects; P < 0.001). Comparison of the uninfected and infected persons indicated a trend for increased perforin-expressing CD8+ T lymphocytes in both compartments, particularly the lymph node (blood, P = 0.177; lymph node, P = 0.032). Further evaluation of the infected subjects revealed that freshly isolated unstimulated bulk CD8+ T lymphocytes in both compartments possessed HIV-1-specific lytic activity directed against Gag, Pol, Env, and Nef (by chromium release assays using recombinant vaccinia virus-infected autologous transformed B cells, data not shown). Neither perforin levels nor bulk HIV-1-specific cytolytic activity related to viremia or CD8+ T-lymphocyte phenotypes in these subjects (not shown). Overall, these data did not demonstrate any quantitative reduction of perforin expression in the blood or lymph node CD8+ T lymphocytes of HIV-1-infected subjects, although they did not exclude the possibility that this unaltered perforin expression was inappropriately low in the setting of infection.

Back to Top | Article Outline

Blood CD8+ T-Lymphocyte Compartment Imprecisely Mirrors That in Lymph Nodes of Infected Subjects

Because most prior phenotypic studies of CD8+ T lymphocytes in HIV-1 infection have been performed using blood, the relationships of the CD8+ T-lymphocyte subsets in the lymph node to those in the peripheral blood were therefore assessed for the HIV-1-infected persons (Fig. 6). The overall percentages of total CD8+ T lymphocytes (see Fig. 6) were roughly correlated between the 2 compartments (r2 = 0.323, P = 0.018). The subsets defined by CD62L/CD45RA, CD28, and CD38/DR were also roughly correlated (see Fig. 6; r2 range: 0.271-0.684). Despite these correlations, however, subset interrelations within each compartment showed variation. Comparing CD38 expression with the percentage of CD62L/CD45RA+ (terminal effector) cells (Fig. 7), a significant correlation was seen in the lymph nodes but not in blood (r2 < 0.001, P = 0.970 for blood; r2 = 0.644, P = 0.002 for lymph node). These results indicated that although the percentages of T-lymphocyte subsets in the blood roughly reflect the populations in the lymph nodes, interrelations in the lymph node may be not be evident in the blood.

Back to Top | Article Outline

CD8+ T-Lymphocyte Phenotypes in Lymph Nodes Correlate More Closely to Viremia Than Those in Blood

Because most viral replication occurs in lymphoid tissues, the relationship of viremia to CD8+ T-lymphocyte subsets in the blood and lymph nodes of seropositive subjects was evaluated (Fig. 8). Examination of CD38 expression confirmed a weakly positive correlation of expression on blood CD8+ cells to viremia (see Fig. 8A; r2 = 0.220, P = 0.058) but a much stronger association of lymph node CD8+ cells (see Fig. 8B; r2 = 0.505, P = 0.001). Further examining the percentage of terminal effector CD8+ T lymphocytes (CD62L/CD45RA+) in blood, there was no clear correlation to viremia (see Fig. 8C; r2 = 0.080, P = 0.374), whereas the percentages in the lymph node showed a robust negative correlation (see Fig. 8D; r2 = 0.520, P = 0.008), indicating that the lymph node but not blood concentration of these effector cells is an important determinant of antiviral activity. The memory-effector subsets (CD62L/CD45RA) in the blood and lymph nodes were somewhat positively associated with viremia (r2 = 0.250 and 0.288, P = 0.098 and 0.072 for blood and lymph node, respectively; not shown), suggesting antigenic dependence of the frequency of these cells but not antiviral activity.

A multiple linear regression model was applied to CD38 expression on lymphocytes in the blood and lymph nodes, the percentage of CD62/CD45RA+ lymphocytes in the lymph node, and viremia (see Table 3). This model suggested that CD38 expression on CD8+ T lymphocytes in the lymph node (but not blood) was the sole significant predictor of viremia (P = 0.046). This agreed with a subsequent Markov Chain/Monte Carlo simulation. This model therefore suggested that the activation status of CD8+ T lymphocytes as reflected by CD38 expression is an independent predictor of immune containment of HIV-1 replication in lymph nodes.

Back to Top | Article Outline


The interaction of CD8+ T lymphocytes with HIV-1 in vivo, although believed to be crucial in pathogenesis, remains obscure. One necessary assumption for most studies of this interaction is that examinations of cells and virus in the blood provide an accurate representation of processes in the whole in vivo environment. Because lymphocytes, which are major targets of infection and effectors of cellular immunity against HIV-1, traffic between blood and tissues, it is reasonable to assume that T lymphocytes in tissue and blood compartments somewhat mirror each other. Supporting this view, a detailed study of CD8+ T lymphocytes in HIV-1-infected individuals demonstrates that the targeting of HIV-1-specific CD8+ T lymphocytes is highly similar in blood and lymph nodes.13 It is less clear, however, whether the functional status of the T lymphocytes in the blood and lymph nodes is also similar. Multiple studies have sought to investigate CD8+ T-lymphocyte function or phenotype as a determinant for the efficacy of the immune response against HIV-1, but most have been performed using peripheral blood, which is not the major site of viral replication.

Our data indicate that there are baseline differences between T lymphocytes in the blood and lymph nodes of HIV-1-uninfected persons. There is a much lower ratio of CD8+ to CD4+ T lymphocytes. There are fewer differentiated CD8+ effector T lymphocytes, as reflected by the higher percentage of CD28 expression, but the activation status is similar. Among CD4+ T lymphocytes, there is, again, a lower percentage of CD28 effector cells but also more marked differences in the activation status, notable for a greater percentage of HLA-DR-expressing cells. Overall, there are more differences in the CD4+ T-lymphocyte compartment than in the CD8+ T-lymphocyte compartment when comparing lymph node and blood in the absence of HIV-1 infection.

Comparing the lymph node T-lymphocyte populations between HIV-1-seronegative and -seropositive persons, the CD8+ T lymphocytes are predominantly affected, similar to blood but to a greater degree. The CD28 and CD38+/HLA-DR+ population expansions are proportionally greater in the lymph node versus the blood, indicating that analysis of these subsets in the blood of HIV-1-infected persons underestimates the changes in the total lymphocyte pool. Changes in the CD4+ lymphocyte phenotypes, by contrast, are far more similar between the blood and lymph node. HIV-1 infection therefore provokes greater alterations in the CD8+ T lymphocytes of lymph nodes than in those of peripheral blood.

It has been suggested that reduced perforin expression in blood22-25 and lymph node26 CD8+ T lymphocytes may be a mechanism of immune failure in HIV-1 infection. We find that the mean perforin expression level is much lower in the lymph node than in the blood of HIV-1-infected persons but that this difference is similar to that in HIV-1-seronegative subjects. It seems that the percentages of perforin-expressing CD8+ T lymphocytes might, in fact, be somewhat increased in HIV-1 infection; thus, depletion compared with the uninfected state does not seem to be a mechanism of immune failure in HIV-1 infection. These data do not exclude inadequate elevation of perforin-expressing cells or reduction of perforin expression in the HIV-1-specific subset of CD8+ T lymphocytes, however.

Interestingly, global assessments of CD8+ T-lymphocyte phenotypes in the lymph node do correlate with plasma viremia. There is a positive correlation of CD38 expression and viremia in agreement with the prior observation of CD38 expression on peripheral blood CD8+ T lymphocytes correlating to viremia.12 Expression levels on cells in the blood and lymph nodes are correlated, but statistical analysis suggests that the predictive value of CD38 expression on cells in the lymph nodes for viremia is independent of that in blood. Furthermore, the percentage of CD8+ terminal effector cells of the CD62L/CD45RA+ phenotype in the lymph node is inversely correlated to viremia, but this correlation is not observed in blood. The percentage of these cells is also closely inversely correlated with CD38 expression levels within the lymph node but, again, not in blood. Thus, despite general correlations of cell populations between blood and the lymph node, the interrelations of populations within the lymph node are obscured in blood.

The mechanism behind the relationships of CD38 expression on CD8+ T lymphocytes within the lymph node to viremia is unclear. Prior work has demonstrated a clear phenomenon of a positive correlation between blood CD8+ T-lymphocyte CD38 expression and viremia as well as disease progression.10-12 Our data suggest that this correlation is more direct in lymph nodes, supporting a relationship based on the interaction of these cells with HIV-1. It has been hypothesized that the expression of CD38 could be a marker of ineffective CD8+ T lymphocytes and therefore directly associated with defective responding cells or expressed secondary to excessive immune activation because of general immune failure.27 Prior correlative work using blood supports the latter hypothesis, finding that CD38 expression is a predictor of disease progression independent of viremia or CD4+ T-lymphocyte concentration.12 The observation that CD38 expression on CD4+ T lymphocytes also predicts disease progression28,29 lends additional support to this concept of CD38 being a general marker for immune failure through immune activation, independent of viral replication.

Our data on T lymphocytes in lymph nodes provide evidence for the possibility that CD38 expression may be more directly associated with CD8+ T-lymphocyte antiviral function. CD38+/CD8+ T lymphocytes are believed to be more prone to apoptosis.30-32 Also, multiple functions in T lymphocytes have been ascribed to the CD38 molecule itself, including enzymatic activity of protein ribosylation,32 serving as an adhesive by binding CD31 on endothelial cells,33 acting as a costimulatory receptor,34,35 and perhaps assisting in nucleotide scavenging in exhausted cells.35 Whether expression of the CD38 molecule has direct functional consequences or serves as an indirect marker for the status of CD8+ T lymphocytes remains to be determined.

The inverse correlation of lymph node CD8+ T lymphocytes of the CD62L/CD45RA+ phenotype to viremia is a novel finding. These cells are believed to be terminally differentiated effector cells that are high-efficiency killers,19 and skewed maturation of HIV-1-specific CD8+ T lymphocytes causing an arrest at the CD45RA preterminally differeniated stage has been proposed as a specific mechanism of immune failure in HIV-1 infection.20 Although this potential perturbation of CD8+ T-lymphocyte development has been demonstrated by comparing the phenotypes of HIV-1-specific versus cytomegalovirus (CMV)-specific CD8+ T lymphocytes in HIV-1-infected persons, direct evidence of functional consequences has been lacking. Our data demonstrate a significant negative correlation of CD62L/CD45RA+ CD8+ T lymphocytes in the lymph node to viremia, suggesting that these cells may be involved directly in viral suppression in vivo. This correlation, although clear with cells from the lymph node, is not observed with cells from the blood, supporting the hypothesis that the key HIV-1 immune interactions occur in lymphoid tissues and not in blood.

It is unclear to what degree CD38 expression and CD62L/CD45RA+ differentiation of CD8+ T lymphocytes are independent or interrelated. It has been observed by chromium release assays on sorted cells that the CD38+/HLA-DR+ compartment contains more HIV-1-specific CD8+ T-lymphocyte activity than the CD38+/HLA-DR or CD38/HLA-DR+ compartment.36 These data do not exclude the possibility that CD38 is a marker for ineffective CD8+ T lymphocytes; however, because we find no association of the CD38/DR+ population with reduced viremia, this seems unlikely. Nevertheless, these results must be interpreted with caution, because the distribution of CD38 and DR expression on HIV-1-specific CD8+ T lymphocytes may not be reflected by that of the CD8+ T-lymphocyte population as a whole. More detailed studies are required to define whether these factors are mechanistically distinct in their associations with viremia.

Finally, a factor that could affect our results is antiretroviral drug therapy. The individuals in this study had highly suppressed or varying levels of viremia attributable to treatment. Such treatment can reduce the levels of activation37 and HIV-1-specific cellular immunity.38-40 Thus, it is possible that treatment of the research subjects blunted cellular activation and reduced the differences we detected. Another consideration is whether drug therapy contributed to the differences noted. Further study of untreated individuals is necessary to discriminate these possibilities.

In summary, we find that there are potentially important differences between T lymphocytes in the blood and lymph nodes. Baseline differences are notable in the ratio and phenotypes of CD4+ and CD8+ T lymphocytes in HIV-1-uninfected persons. HIV-1 infection particularly affects the phenotype of the CD8+ lymphocytes, and the changes seen in the peripheral blood underestimate those seen in lymph nodes. Although there are general correlations between T-lymphocyte populations in the blood and lymph nodes, the interaction of CD8+ T lymphocytes with HIV-1 is likely best seen with cells from lymph nodes, which are the predominant source of viral replication. These data introduce caveats in the interpretation of pathogenesis and vaccine studies of blood.

Back to Top | Article Outline


1. Westermann J, Pabst R. Distribution of lymphocyte subsets and natural killer cells in the human body. Clin Investig. 1992;70:539-544.
2. Pantaleo G, Cohen OJ, Schacker T, et al. Evolutionary pattern of human immunodeficiency virus (HIV) replication and distribution in lymph nodes following primary infection: implications for antiviral therapy. Nat Med. 1998;4:341-345.
3. Pantaleo G, Menzo S, Vaccarezza M, et al. Studies in subjects with long-term nonprogressive human immunodeficiency virus infection. N Engl J Med. 1995;332:209-216.
4. Simmonds P, Balfe P, Peutherer JF, et al. Human immunodeficiency virus-infected individuals contain provirus in small numbers of peripheral mononuclear cells and at low copy numbers. J Virol. 1990;64:864-872.
5. Jin X, Bauer DE, Tuttleton SE, et al. Dramatic rise in plasma viremia after CD8(+) T cell depletion in simian immunodeficiency virus-infected macaques. J Exp Med. 1999;189:991-998.
6. Matano T, Shibata R, Siemon C, et al. Administration of an anti-CD8 monoclonal antibody interferes with the clearance of chimeric simian/human immunodeficiency virus during primary infections of rhesus macaques. J Virol. 1998;72:164-169.
7. Schmitz JE, Kuroda MJ, Santra S, et al. Control of viremia in simian immunodeficiency virus infection by CD8+ lymphocytes. Science. 1999;283:857-860.
8. Addo MM, Yu XG, Rathod A, et al. Comprehensive epitope analysis of human immunodeficiency virus type 1 (HIV-1)-specific T-cell responses directed against the entire expressed HIV-1 genome demonstrate broadly directed responses, but no correlation to viral load. J Virol. 2003;77:2081-2092.
9. Betts MR, Ambrozak DR, Douek DC, et al. Analysis of total human immunodeficiency virus (HIV)-specific CD4+ and CD8+ T-cell responses: relationship to viral load in untreated HIV infection. J Virol. 2001;75:11983-11991.
10. Liu Z, Hultin LE, Cumberland WG, et al. Elevated relative fluorescence intensity of CD38 antigen expression on CD8+ T cells is a marker of poor prognosis in HIV infection: results of 6 years of follow-up. Cytometry. 1996;26:1-7.
11. Liu Z, Cumberland WG, Hultin LE, et al. Elevated CD38 antigen expression on CD8+ T cells is a stronger marker for the risk of chronic HIV disease progression to AIDS and death in the Multicenter AIDS Cohort Study than CD4+ cell count, soluble immune activation markers, or combinations of HLA-DR and CD38 expression. J Acquir Immune Defic Syndr Hum Retrovirol. 1997;16:83-92.
12. Giorgi JV, Lyles RH, Matud JL, et al. Predictive value of immunologic and virologic markers after long or short duration of HIV-1 infection. J Acquir Immune Defic Syndr. 2002;29:346-355.
13. Altfeld M, van Lunzen J, Frahm N, et al. Expansion of pre-existing, lymph node-localized CD8+ T cells during supervised treatment interruptions in chronic HIV-1 infection. J Clin Invest. 2002;109:837-843.
14. Killian MS, Monteiro J, Matud J, et al. Persistent alterations in the T-cell repertoires of HIV-1-infected and at-risk uninfected men. AIDS. 2004;18:161-170.
15. Yang OO, Boscardin WJ, Matud J, et al. Immunologic profile of highly exposed yet HIV type 1-seronegative men. AIDS Res Hum Retroviruses. 2002;18:1051-1065.
16. Iyer SB, Hultin LE, Zawadzki JA, et al. Quantitation of CD38 expression using QuantiBRITE beads. Cytometry. 1998;33:206-212.
17. Westfall PH, Young SS. Resampling-Based Multiple Testing. New York: John Wiley & Sons; 1993.
18. Putter H, Heisterkamp SH, Lange JMA, et al. A Bayesian approach to parameter estimation in HIV dynamical models. Stat Med. 2002;21:2199-2214.
19. Hamann D, Baars PA, Rep MH, et al. Phenotypic and functional separation of memory and effector human CD8+ T cells. J Exp Med. 1997;186:1407-1418.
20. Champagne P, Ogg GS, King AS, et al. Skewed maturation of memory HIV-specific CD8 T lymphocytes. Nature. 2001;410:106-111.
21. Appay V, Dunbar PR, Callan M, et al. Memory CD8+ T cells vary in differentiation phenotype in different persistent virus infections. Nat Med. 2002;8:379-385.
22. Migueles SA, Laborico AC, Shupert WL, et al. HIV-specific CD8(+) T cell proliferation is coupled to perforin expression and is maintained in nonprogressors. Nat Immunol. 2002;3:1061-1068.
23. Haridas V, McCloskey TW, Pahwa R, et al. Discordant expression of perforin and granzyme A in total and HIV-specific CD8 T lymphocytes of HIV infected children and adolescents. AIDS. 2003;17:2313-2322.
24. Zhang D, Shankar P, Xu Z, et al. Most antiviral CD8 T cells during chronic viral infection do not express high levels of perforin and are not directly cytotoxic. Blood. 2003;101:226-235.
25. Heintel T, Sester M, Rodriguez MM, et al. The fraction of perforin-expressing HIV-specific CD8 T cells is a marker for disease progression in HIV infection. AIDS. 2002;16:1497-1501.
26. Andersson J, Kinloch S, Sonnerborg A, et al. Low levels of perforin expression in CD8+ T lymphocyte granules in lymphoid tissue during acute human immunodeficiency virus type 1 infection. J Infect Dis. 2002;185:1355-1358.
27. Liu Z, Cumberland WG, Hultin LE, et al. CD8+ T-lymphocyte activation in HIV-1 disease reflects an aspect of pathogenesis distinct from viral burden and immunodeficiency. J Acquir Immune Defic Syndr Hum Retrovirol. 1998;18:332-340.
28. Benito JM, Zabay JM, Gil J, et al. Quantitative alterations of the functionally distinct subsets of CD4 and CD8 T lymphocytes in asymptomatic HIV infection: changes in the expression of CD45RO, CD45RA, CD11b, CD38, HLA-DR, and CD25 antigens. J Acquir Immune Defic Syndr Hum Retrovirol. 1997;14:128-135.
29. Giorgi JV, Hultin LE, McKeating JA, 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.
30. Prince HE, Jensen ER. HIV-related alterations in CD8 cell subsets defined by in vitro survival characteristics. Cell Immunol. 1991;134:276-286.
31. Gougeon ML, Lecoeur H, Dulioust A, 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.
32. Grimaldi JC, Balasubramanian S, Kabra NH, et al. CD38-mediated ribosylation of proteins. J Immunol. 1995;155:811-817.
33. Prabhakar P, Laboy JI, Wang J, et al. Effect of NADH-X on cytosolic glycerol-3-phosphate dehydrogenase. Arch Biochem Biophys. 1998;360:195-205.
34. Ausiello CM, la Sala A, Ramoni C, et al. Secretion of IFN-gamma, IL-6, granulocyte-macrophage colony-stimulating factor and IL-10 cytokines after activation of human purified T lymphocytes upon CD38 ligation. Cell Immunol. 1996;173:192-197.
35. Bofill M, Parkhouse RM. The increased CD38 expressed by lymphocytes infected with HIV-1 is a fully active NADase. Eur J Immunol. 1999;29:3583-3587.
36. Ho HN, Hultin LE, Mitsuyasu RT, et al. Circulating HIV-specific CD8+ cytotoxic T cells express CD38 and HLA-DR antigens. J Immunol. 1993;150:3070-3079.
37. Lange CG, Lederman MM, Madero JS, et al. Impact of suppression of viral replication by highly active antiretroviral therapy on immune function and phenotype in chronic HIV-1 infection. J Acquir Immune Defic Syndr. 2002;30:33-40.
38. Rinaldo CR Jr, Huang XL, Fan Z, et al. Anti-human immunodeficiency virus type 1 (HIV-1) CD8(+) T-lymphocyte reactivity during combination antiretroviral therapy in HIV-1-infected patients with advanced immunodeficiency. J Virol. 2000;74:4127-4138.
39. Kalams SA, Goulder PJ, Shea AK, et al. Levels of human immunodeficiency virus type 1-specific cytotoxic T-lymphocyte effector and memory responses decline after suppression of viremia with highly active antiretroviral therapy. J Virol. 1999;73:6721-6728.
40. Ogg GS, Jin X, Bonhoeffer S, et al. Decay kinetics of human immunodeficiency virus-specific effector cytotoxic T lymphocytes after combination antiretroviral therapy. J Virol. 1999;73:797-800.
Back to Top | Article Outline


The MACS (available at: includes the following individuals:

Baltimore, The Johns Hopkins University Bloomberg School of Public Health: Joseph B. Margolick (Principal Investigator), Haroutune Armenian, Barbara Crain, Adrian Dobs, Homayoon Farzadegan, Nancy Kass, Shenghan Lai, Justin McArthur, and Steffanie Strathdee

Chicago, Howard Brown Health Center, The Feinberg School of Medicine, Northwestern University, and Cook County Bureau of Health Services: John P. Phair (Principal Investigator), Joan S. Chmiel (Co-Principal Investigator), Sheila Badri, Bruce Cohen, Craig Conover, Maurice O'Gorman, Frank Pallela, Daina Variakojis, and Steven M. Wolinsky

Los Angeles, University of California, UCLA Schools of Public Health and Medicine: Roger Detels and Beth Jamieson (Principal Investigators), Barbara R. Visscher (Co-Principal Investigator), Anthony Butch, John Fahey, Otoniel Martínez-Maza, Eric N. Miller, John Oishi, Paul Satz, Elyse Singer, Harry Vinters, Otto Yang, and Stephen Young

Pittsburgh, University of Pittsburgh, Graduate School of Public Health: Charles R. Rinaldo (Principal Investigator), Lawrence Kingsley (Co-Principal Investigator), James T. Becker, Phalguni Gupta, John Mellors, Sharon Riddler, and Anthony Silvestre

Data Coordinating Center (Baltimore), The Johns Hopkins University Bloomberg School of Public Health: Alvaro Muñoz (Principal Investigator), Lisa P. Jacobson (Co-Principal Investigator), Haitao Chu, Stephen R. Cole, Janet Schollenberger, Eric Seaberg, Michael Silverberg, and Sol Su

National Institutes of Health (Bethesda), National Institute of Allergy and Infectious Diseases: Carolyn Williams

National Cancer Institute: Jodi Black

Cited By:

This article has been cited 1 time(s).

JAIDS Journal of Acquired Immune Deficiency Syndromes
Characterization of Baseline Intestinal Mucosal Indices of Injury and Inflammation in Men for Use in Rectal Microbicide Trials (HIV Prevention Trials Network-056)
McGowan, I; Elliott, J; Cortina, G; Tanner, K; Siboliban, C; Adler, A; Cho, D; Boscardin, WJ; Soto-Torres, L; Anton, PA
JAIDS Journal of Acquired Immune Deficiency Syndromes, 46(4): 417-425.
PDF (122) | CrossRef
Back to Top | Article Outline

CD8+ T lymphocytes; peripheral blood; lymph node; HIV-1

© 2005 Lippincott Williams & Wilkins, Inc.