Absolute CD4 T cell numbers recover more rapidly following suppression of HIV replication by highly active antiretroviral therapy (HAART) in children compared with adults. In contrast, CD8 T cell numbers do not change over time during treatment .
CD8 cytotoxic T lymphocytes are recognized as important in combating intracellular infections, such as with viruses. Using a combination of phenotypic markers, CD8 T cells can be divided into different subsets. In this study, the phenotypic marker CD27, which is downregulated upon interaction with its ligand CD70, is used with CD45RA to divide CD8 T cells into four different subsets: CD8+CD45RA+CD27+ (naive), CD8+CD45RA−CD27+ (memory), CD8+CD45RA−CD27− (memory/effector) and CD8+CD45RA+CD27− (effector) T cells . Although the sequential development of CD8 T cells during primary infection in mouse was shown to go from naive to effector and then to memory , during chronic viral infections in humans, a sequential maturation pattern has been suggested to follow the path naive to memory and via memory/effector to effector .
Herpesviruses, such s cytomegalovirus (CMV), Epstein–Barr virus (EBV) and varicella zoster virus, establish latency after initial infection. These latent infections have the potential to reactivate. CMV is a frequent infection in HIV-1-infected children [5,6]. In the pre-HAART era, children infected with both CMV and HIV-1 were more likely to have disease progression than children who were seropositive only for HIV-1 . Even in the era of HAART, CMV is associated with an increased risk of disease progression to AIDS and decreased survival [8,9].
In a healthy pediatric population, CMV seroprevalence at the age of 18 years was found to reach 75% . In this control population, it has been shown that CMV infection is associated with the outgrowth of CD8+CD45RA+CD27− effector T cells. Since we and others have found that CMV-specific CD8 T cells are preferentially of the CD45RA+CD27− phenotype [4,10–12], it seems likely that CMV relates to these CD8 effector T cell expansions.
The present study analyses the effect of CMV infection and replication on CD8 T cell differentiation in HIV-1-infected children taking HAART. To our knowledge, this is the first study to describe CMV-related immune reactivity in HIV-1-infected children associated with persistent CMV replication.
The Amsterdam pediatric HIV-1 cohort consists of children and young adolescents under the age of 18 years. The present study comprised all children who started HAART between 1997 and 2002. The Medical Ethical Committee approved this study and all caregivers gave written informed consent.
Blood and urine samples were obtained at each visit at intervals of 2–4 months. Serology for CMV was performed at the start of HAART. If positive, the test was repeated at least every 48 weeks thereafter. If negative, serological analysis was performed upon each visit until seroconversion.
CMV antibodies were defined by Axsym assays (Abbott Diagnostics, Amstelveen, the Netherlands) and expressed as arbitrary units. All tests were performed following the instructions of the manufacturer. Seropositivity was defined by the presence of a positive specific IgG after the age of 12–18 months or by CMV-specific IgG remaining positive during follow-up in order to exclude the confounding contribution of maternal antibodies in the very young.
Culture of cytomegalovirus
Patient urine was cocultivated with human diploid fibroblasts for culture of CMV according to standard procedures. Repeated cultures were done prospectively in 32 CMV-seropositive children. Prolonged CMV shedding was defined as the presence of at least two positive CMV cultures after more than 36 weeks on HAART. A patient was defined as definitively negative if two urine cultures were negative at an interval of 2 months after more than 36 weeks on HAART, while secreting CMV at the start of HAART.
Lymphocyte subsets and enumeration of cytotoxic effector cells
Numbers of B cells (CD19), T cells (CD3) and subsets (CD3+CD4+ and CD3+CD8+) were determined by standard procedures using FACScan (Becton Dickinson, San Jose, California, USA). Activation and maturation of the CD4 and CD8 T cells were determined using monoclonal antibodies against HLA-DR (Becton Dickinson), CD38, CD45RA (Coulter Immunology, Paris, France) and CD27 (Sanquin Reagents, Amsterdam, the Netherlands).
Flow cytometry and intracellular interferon-γ staining after antigen-specific stimulation
Peripheral blood mononuclear cells were stimulated either with 2 μg/ml 15-mer 11 amino acid residue overlap CMV-derived pp65 peptide pool or with CMV lysate 10 μl/ml (Microbix Biosystems, Toronto, Canada) in the presence of anti-CD28 (Sanquin Reagents) and CD49d (Becton Dickinson, San Jose, California, USA) for 6 h. After 1 h, brefeldin A was added. Subsequently, cells were stained with anti-CD27, anti-CD45RO and either anti-CD8 or anti-CD4; after fixation and permeabilization, the cells were stained intracellularly with anti-IFN-γ (Becton Dickinson). A FACSCalibur flowcytometer (Becton Dickinson) was used to acquire 200 000–400 000 events. For clarity of interpretation, CD45RO+ and CD45RO− was considered as CD45RA− and CD45RA+, respectively. For intracellular IFN-γ measurements, the number of responding T cells was calculated after subtracting negative control values.
Plasma HIV-1 RNA determination
Plasma HIV-1 RNA concentration was determined either using Nuclisens HIV-1 RNA QT (Biomérieux, Boxtel, the Netherlands) or Versant HIV-1 RNA 3.0 (Bayer, Tarrytown, New York, USA). All tests were performed according to the instructions of the manufacturers. HIV-1-RNA < 400 copies/ml was considered undetectable.
Statistical analyses were performed using SPSS for Windows version 11.5 (SPSS, Chicago, Illinois, USA). All P values were two-tailed and values < 0.05 were considered statistically significant. Continuous data were analysed using a Mann–Whitney U-test. Categorical data were compared with a Fisher's exact test. A paired sample t-test was used to analyse differences between baseline and 48 weeks of follow-up. Correlation was tested using the Spearman's correlation test. Univariable and multivariable logistic regression models were used to identify independent factors associated with T cell numbers above the median of the group, using a stepwise backward model.
Population characteristics and response to antiretroviral therapy
The study comprised 52 HIV-1-infected children who started HAART between 1997 and 2002. All children had completed a follow-up of 48 weeks on treatment at the time of analysis. Table 1 shows the baseline characteristics of the cohort. Plasma HIV-1 RNA was initially suppressed below detection levels in 49 of 52 children in a median of 9.1 weeks [interquartile range (IQR), 3.4–18.9]. After 48 weeks, 44 children still had undetectable plasma HIV RNA. Median CD4 T cell numbers increased from 480 to 1185 × 106 cells/l during 48 weeks of follow-up (P < 0.001).
Cytomegalovirus infection and CD8 T cell differentiation
Since CMV infection was associated with faster progression to AIDS and with the outgrowth of CD8 effector T cells, CMV-seropositive children in our cohort were compared with CMV-seronegative children.
Thirty-seven children had contracted CMV before initiation of HAART. Three patients seroconverted during treatment and all others remained seronegative during follow-up. Median age at baseline did not differ between CMV-seropositive and CMV-seronegative children (4.7 versus 4.8 years; P = 0.5), as was the case for sex (male patients 43% versus 47%; P = 1.0). Plasma HIV-1 RNA at baseline did not differ between CMV-seropositive and CMV-seronegative children (5.0 versus 4.6 log copies/ml; P = 0.12). Furthermore, children not able to suppress HIV replication during 48 weeks on HAART were equally present in the CMV-seropositive and CMV-seronegative group (5/37 versus 1/12; P = 0.54). Table 2 shows immunophenotypic comparisons of CMV-seropositive and CMV-seronegative children. The absolute CD8 effector T cell number was higher in the CMV-seropositive compared with the CMV-seronegative children at baseline (P = 0.009); this persisted after 48 weeks (P < 0.001) and was still present after 96 weeks on HAART (median 406 versus 53 × 106 cells/l; P = 0.001; data not shown). Comparing the fraction of each of the subsets showed that in CMV-seropositive children at baseline only CD8+CD45RA+CD27− effector T cells were higher (Table 2). After 48 weeks on HAART, the CD8+CD45RA−CD27− memory/effector T cells were also higher. In contrast, the naive CD8+CD45RA+CD27+ T cells were lower in CMV-seropositive children.
Cytomegalovirus-associated outgrowth of effector T cells and chronic immune activation
To test whether CMV infection was associated with chronic activation of the immune system, the activation markers HLA-DR and CD38 were analysed on CD4 and CD8 T cells. CD4+HLA-DR+CD38+ T cells at start of HAART were higher in CMV-seropositive than in CMV-seronegative children (P = 0.02). CD8+HLA-DR+CD38+ T cells were increased, almost significantly, in CMV-seropositive compared with CMV-seronegative individuals at baseline (P = 0.07) (Table 2). Differences in both CD4 and CD8 compartments disappeared during follow-up.
At baseline, absolute numbers of CD8 effector T cells correlated with CD4 T cells (r = 0.45; P = 0.005) and activated (HLA-DR+CD38+) CD4 and CD8 T cells (r = 0.71 and r = 0.82; P < 0.001), which persisted in the activated CD8 T cells after 48 weeks (r = 0.61; P < 0.001). In contrast, age and plasma HIV-1 RNA did not correlate with CD8 effector T cell numbers.
Primary cytomegalovirus infection during HAART
Three girls seroconverted for CMV during treatment with HAART. In patient A and B, the number of CD8+CD45RA+CD27− T cells increased and stabilized at a level above baseline (Fig. 1a,b). Patient C developed acute CMV infection when she had a CD4 cell count of only 6 × 106 cells/l. The number of CD8+CD45RA+CD27− T cells did not increase above the cut-off of 20 × 106 cells/l (Fig. 1c). She was treated with ganciclovir. Hepatic dysfunction and cachexia progressed, and she developed a fatal, multidrug-induced liver failure.
Cytomegalovirus serology and shedding in urine during latency
After infection, CMV remains dormant in the kidneys. Urinary shedding is a marker of CMV replication. After congenital infection, it can be found for up to 10 years . In healthy asymptomatic children and adolescents, urine samples are intermittently positive for CMV up to 30 weeks after primary infection .
Based on this report, an extended period of 36 weeks as arbitrary cut-off for persistent CMV shedding was chosen. Regular urine tests for CMV were prospectively performed in 32 of 37 CMV-seropositive children: 13 (41%), 11 boys and two girls, had persistent CMV shedding and 19 (59%), three boys and 16 girls, had negative urine cultures (P < 0.001). CMV secretors were younger than non-secretors (median 5.3 versus 1.0 years; P = 0.001) (Fig. 2a). This difference in age and gender was also present when only the 27 children who were vertically HIV-1 infected were analysed (P = 0.007 and P = 0.002, respectively). Furthermore, CMV-specific IgG increased in CMV secretors and stabilized in non-secretors (median 73.3 versus 0.00 arbitrary units/ml; P = 0.02) (Fig. 2b). Among patients with prolonged viral secretion, five were positive for CMV DNA, as measured by quantitative polymerase chain reaction on whole blood, compared with none in the non-secretors.
Apart from the single patient mentioned earlier and illustrated in Fig. 1c, none of the patients developed clinical CMV-related disease that needed treatment under HAART.
Cytomegalovirus shedding in the urine and T cells
At baseline, total numbers of CD4, CD8 and CD8 effector T cells were not different between secretors and non-secretors (Fig. 2c, left panel). In contrast, 36 weeks after the initiation of HAART, CMV-secreting patients had a higher number of total CD4 T cells (P = 0.01). This was also true for total CD8 T cells (P = 0.003), CD8+CD45RA+CD27− effector T cells (P = 0.01) (Fig. 2c, right panel) and CD8+CD45RA+CD27+ naive T cells (median 1067 versus 484 × 106 cells/l; P = 0.006) in CMV secretors versus non-secretors. These differences persisted until 96 weeks of follow-up (data not shown).
The relative fractions of each of the subsets within the CD8 T cells were not different between secretors and non-secretors at any time (data not shown). This finding underscores the idea that continuous replication is associated with absolute numbers of effector T cells instead of relative changes in subset distribution.
The median number of CD8+CD45RA+CD27− T cells (421 × 106 cells/l) and naive CD8+CD45RA+CD27+ T cells (710 × 106 cells/l) at week 36 for all CMV-seropositive children was used to define binary variables, above or under the median. In a univariable analysis, prolonged CMV shedding [odds ratio (OR), 7.9; 95% confidence interval (CI), 1.1–56.1; P = 0.04] and male gender (OR, 7.5; 95% CI, 1.3–43.0; P = 0.02) were associated with a higher chance of increased numbers of CD8+CD45RA+CD27− T cells. In contrast, age, HIV RNA at 36 weeks, Centers for Disease Control and Prevention classification at presentation, and prior varicella zoster virus or EBV infection gave no higher chance of CD8+CD45RA+CD27− numbers > 421 × 106 cells/l. A multivariable regression model showed that CMV secretion was the only predictor of having a high number of CD8 effector T cells at week 36 (OR, 7.9; P = 0.04). There was no two-way interaction found between gender and CMV secretion.
In a univariable analysis, male gender (OR, 5.1; 95% CI, 1.0–26.8; P = 0.06) and age (OR, 0.8; 95% CI, 0.6–0.9; P = 0.007) were associated with a higher number of naive CD8 T cells and not CMV shedding. In a multivariable analysis, age was the only independent predictor of the number of naive CD8 T cells (OR, 0.6; 95% CI, 0.3–0.9; P = 0.03). Therefore, in contrast to CD8 effector T cells, high naive CD8 T cells were independently associated with younger age. The same association was found in a multivariable analysis for the total (P = 0.03), naive (P = 0.04) and memory (P = 0.04) CD4 T cells at 36 weeks.
Cytomegalovirus-specific T cell responses
The number of IFN-γ-producing T cells can be used as a measure of the number of virus-specific T cells present in the blood. To study CMV-specific T cell immunity, IFN-γ production by CD4 and CD8 T cells was measured after stimulation with either CMV lysate or a peptide pool derived from the immunodominant pp65 antigen in 16 CMV-seropositive children: eight secretors and eight non-secretors. Neither absolute numbers nor percentages of CMV-specific IFN-γ-producing CD4 T cells were different in secretors versus non-secretors (Fig. 3a, and data not shown). In contrast to the increased in total CD8 T cell numbers and its CD45RA+CD27− effector subset in CMV secretors, the numbers of CMV-specific IFN-γ-producing CD8 T cells were lower in CMV secretors compared with non-secretors (median 6.1 versus 13.6 × 106 cells/l; P = 0.02; Fig. 3a). Also, lower numbers of IFN-γ-producing CD8+CD45RA+CD27− effector T cells were found in CMV secretors compared with non-secretors (median 1.0 versus 3.8 × 106 cells/l; P = 0.04; Fig. 3b), as was true for the CD8+CD27− T cells in children with prolonged CMV shedding (median 2.0 versus 6.0 × 106 cells/l; P = 0.01). Whereas IFN-γ-producing cells were equally detected in the CD27-positive (median 53.5%) and the CD27-negative subset (median 46.8%) in patients with complete CMV suppression, patients with persistent CMV shedding showed a difference in favour of the CD27-positive (median 62%) over the CD27-negative subset (median 36.7%) (P = 0.04) (Fig. 3c). Age was not associated with the number of IFN-γ-producing CD8+CD45RA+CD27− T cells. Together, these findings may suggest incomplete functional differentiation of CMV-specific CD8 T cells despite a higher frequency of total CD8 T cells with an effector phenotype.
As a control, the EBV-specific CD8 T cell responses in the same patients upon stimulation with an EBV-lytic antigen-derived BZLF-1 peptide pool were compared between the two groups; there were no differences in the numbers of IFN-γ-producing T cells (data not shown).
In the present study, the kinetics of CD8+CD45RA+CD27− effector T cells were analysed in HIV-1-infected children treated with HAART. In healthy children, a significant association between the number of circulating CD8+CD45RA+CD27− T cells and CMV seropositivity was found . In HIV-1-infected children, we now demonstrate that the outgrowth of these CD8 effector T cells is similarly related to CMV, as is further exemplified by the kinetics of these cells in patients with acute CMV infection under HAART.
In our cohort, there was no difference in CD4 or CD8 T cell numbers at baseline and during follow-up between the CMV-seropositive and CMV-seronegative group, although CD4 and CD8 T cells were more activated in children with prior CMV infection, as indicated by CD38 and HLA-DR expression. This difference in activation state between these two groups disappeared after initiation of HAART. A correlation between activated CD8 T cells and CD8 effector T cells was found both at baseline and after 48 weeks of follow-up. In contrast, such correlation was not found with plasma HIV RNA at any time. These data demonstrate that in HIV-1-infected children, apart from the effects of HIV itself, ongoing CMV replication may contribute to chronic alteration of the immune system.
CMV-seropositive but otherwise healthy children have been found to have a median of 67 × 106 cells/l CD8 effector T cells (mean 85 × 106 cells/l) , which is much lower than the median of 369 × 106 cells/l at baseline and 323 × 106 cells/l after 48 weeks HAART found in our cohort of HIV-1-infected children. In the same study, it was found that children who had primary CMV infection prior to organ transplant had a median of 74 × 106 cells/l CD8 effector T cells. In contrast, children who had primary CMV infection during treatment with immunosuppressive therapy had a median of 413 × 106 cells/l CD8 effector T cells. This suggests that the ability of the immune system to suppress CMV is inversely correlated with the number of CD8 effector T cells. This is in line with our finding that HIV-1-infected CMV-seropositive children had higher numbers of CD8 effector T cells and some of them showed continuing replication of CMV and mucosal shedding that was associated with the outgrowth of this subset.
Of the prospectively tested HIV-1-infected children, 41% showed persistent CMV secretion in the urine for more than 36 weeks after start of HAART, irrespective of plasma HIV RNA at baseline or after 48 weeks. CMV replication was reflected by persistent secretion of CMV in the urine, and periodic CMV DNA in the peripheral blood. Furthermore, CMV secretors showed increasing titres of CMV-specific IgG and increased numbers of CD8 effector T cells while CMV-specific IFN-γ-producing CD8 T cells were reduced, when compared with non-secreting patients in our cohort. These data suggest inadequate cellular immunity to CMV in children with prolonged secretion. Tu et al.  found that, after CMV infection in very young children, CD4 T cell responses were diminished in a selected group that secreted CMV after 1–2 years, while CD8 T cell responses were comparable to those in adults. In contrast, our results in HIV-1-infected children showed that CMV secretion was associated with a decreased number of functional CMV-specific IFN-γ-producing CD8 T cells in the presence of equal numbers of CMV-specific CD4 T cells.
There are several possible explanations for our findings. First, reduced numbers of CMV-specific CD8 T cells can be explained by differences in distribution of these cells over the various anatomical compartments. CMV-specific cells may have become trapped in the target organs and draining lymph nodes, whereas the increase in CD8 effector T cells could represent an epiphenomenon. However, CMV-specific T cells are preferentially found in the peripheral blood instead of extravascular tissues , and redistribution did not seem to affect (CMV-specific) CD4 effector T cells in our study. Moreover, CMV-specific IFN-γ-producing CD8 effector T cells were not able to suppress CMV replication completely.
An alternative explanation would relate to CMV specificity and responsiveness. The number of CMV-specific IFN-γ-producing T cells was significantly lower in children with prolonged CMV shedding. Despite a higher frequency of CD8 effector T cells, incomplete functional differentiation of CMV-specific CD8 T cells may be present . This is in line with the finding that, in HIV-1-infected male adults, progressors to AIDS with CMV end-organ disease showed increased CD8 T cells that were positive for the CMV-specific tetramer, but fewer CMV-specific IFN-γ-producing CD8 T cells .
Since HLA typing precluded the use of standard tetramer staining (i.e., HLA-A2, HLA-B7) to enumerate CMV-specific CD8 T cells in our cohort, a functional read-out for CMV-specific activity was used instead. Therefore, we cannot discriminate between the possibilities of an increase in CMV-related CD8 T cells with a virus specificity different from CMV pp65; an increased number of dysfunctional CMV pp65-directed T cells; or an indirect bystander phenomenon, being CMV related yet with little or no CMV specificity.
Other cellular functions remain to be studied. Virus-specific peptide-induced IFN-γ production correlated with cytotoxicity against target cells loaded with the same peptides . We describe, for the first time, that numbers of functional CMV-specific CD8 T cells are reduced in CMV-shedding children compared with children who can suppressed CMV replication.
Recent experimental studies [20–23] have revealed that the ability of ‘unhelped’ memory CD8 T cells to produce IFN-γ when restimulated was strongly reduced compared with ‘helped’ memory CD8 T cells. These experimental studies also demonstrated that restored CD4 help (as seen under HAART) of previously ‘unhelped’ memory CD8 T cells did not remedy the defective CD8 T cell response [20–25]. If CMV-specific CD4 T cells are defective before HAART, CD8 T cell responses start to fail and viral replication returns. In such a scenario, the increased antibody response in secretors may act to contain replication .
In addition, the naive CD8 T cells and the total, memory and memory–effector CD4 T cells seemed to expand more strongly in CMV secretors than in non-secretors, but multivariable analysis demonstrated that this association was, in contrast to CD8 effector T cells, confounded by age and not related to CMV. The expansion of naive T cells upon HAART is assumed to be largely antigen-independent. Cytokines, such as interleukin-7 and, to a lesser extent, interleukin-15, may play a role [26–29]. Whether CMV infection and prolonged shedding results in increased cytokine levels or a different cellular sensitivity affecting selective outgrowth of certain T cell subsets, remains to be determined.
In conclusion, our findings demonstrate that, similar to healthy age-matched controls , in HIV-1-infected children CMV infection is associated with the outgrowth of CD8+CD45RA+CD27− effector T cells, which is not seen with other herpesviruses tested nor with HIV-1 itself. Endogenous stimulation of the immune system by persistent CMV secretion results in progressively increasing CMV-specific IgG and higher numbers of CD8 effector T cells. Despite these increases, the CMV-specific IFN-γ-producing CD8 T cell response is diminished, which could explain the inability to suppress CMV completely in 41% of HIV-infected children, irrespective of HIV RNA and immune reconstitution under HAART.
We are indebted to Atie van der Plas and Eugenie le Poole for their care for the children at the outpatient clinic; to Marijke Roos and Margreet Westerlaken for T lymphocyte immunophenotyping and processing of patients’ samples; to René van Lier and Frank Miedema for critically reading and commenting on the manuscript; to Florian Kern for providing the pp65 peptide pool, and to Erwan Piriou for providing the BZLF-1 peptide pool.
Sponsorship: This research has been funded by grant number 2002 7006 from AIDS fonds, the Netherlands.
Note: Bekker and Bronke contributed equally to the work presented.
1. van Rossum AM, Scherpbier HJ, van Lochem EG, Pakker NG, Slieker WA, et al
. Therapeutic immune reconstitution in HIV-1-infected children is independent of their age and pretreatment immune status. AIDS 2001; 15:2267–2275.
2. Haman 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 1997; 186:1407–1418.
3. Wherry EJ, Teichgraber V, Becker TC, Masopust D, Kaech SM, Antia R, et al
. Lineage relationship and protective immunity of memory CD8 T cell subsets. Nat Immunol 2003; 4:225–234.
4. Appay V, Dunbar PR, Callan M, Klenerman P, Gillespie GM, Papagno L, et al
. Memory CD8+ T cells vary in differentiation phenotype in different persistent virus infections. Nat Med 2002; 8:379–385.
5. Chandwani S, Kaul A, Bebenroth D, Kim M, DiJohn D, Fidelia A, et al
. Cytomegalovirus infection in human immunodeficiency virus type 1-infected children. Pediatr Infect Dis J 1996; 15:310–314.
6. Kitchen BJ, Engler HD, Gill VJ, Marshall D, Steinberg SM, Pizzo PA, et al
. Cytomegalovirus infection in children with human immunodeficiency virus infection. Pediatr Infect Dis J 1997; 16:358–363.
7. Kovacs A, Schluchter M, Easley K, Demmler G, Shearer W, La Russa P, et al
. Cytomegalovirus infection and HIV-1 disease progression in infants born to HIV-1-infected women. Pediatric Pulmonary and Cardiovascular Complications of Vertically Transmitted HIV Infection Study Group. N Engl J Med 1999; 341:77–84.
8. Erice A, Tierney C, Hirsch M, Caliendo AM, Weinberg A, Kendall MA, et al
. Cytomegalovirus (CMV) and human immunodeficiency virus (HIV) burden, CMV end-organ disease, and survival in subjects with advanced HIV infection (AIDS Clinical Trials Group Protocol 360). Clin Infect Dis 2003; 37:567–578.
9. Deayton JR, Sabin CA, Johnson M, Emery V, Wilson P, Griffiths PD. Importance of cytomegalovirus viraemia in risk of disease progression and death in HIV-infected patients receiving highly active antiretroviral therapy. Lancet 2004; 363:2116–2121.
10. Kuijpers TW, Vossen MT, Gent MR, Davin JC, Roos MT, Wertheim-van Dillen PM, et al
. Frequencies of circulating cytolytic, CD45RA+CD27−CD8+ T lymphocytes depend on infection with CMV. J Immunol 2003; 170:4342–4348.
11. van Leeuwen EM, Gamadia LE, Baars PA, Remmerswaal EB, ten Berge IJ, van Lier RA. Proliferation requirements of cytomegalovirus-specific, effector-type human CD8+ T cells. J Immunol 2002; 169:5838–5843.
12. 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 2000; 192:63–75.
13. Noyola DE, Demmler GJ, Williamson WD, Griesser C, Sellers S, Llorente A, et al
. Cytomegalovirus urinary excretion and long term outcome in children with congenital cytomegalovirus infection. Congenital CMV Longitudinal Study Group. Pediatr Infect Dis J 2000; 19:505–510.
14. Zanghellini F, Boppana SB, Emery VC, Griffiths PD, Pass RF. Asymptomatic primary cytomegalovirus infection: virologic and immunologic features. J Infect Dis 1999; 180:702–707.
15. Tu W, Chen S, Sharp M, Dekker C, Manganello AM, Tongson EC, et al
. Persistent and selective deficiency of CD4+ T cell immunity to cytomegalovirus in immunocompetent young children. J Immunol 2004; 172:3260–3267.
16. Ellefsen K, Harari A, Champagne P, Bart PA, Sekaly RP, Pantaleo G. Distribution and functional analysis of memory antiviral CD8 T cell responses in HIV-1 and cytomegalovirus infections. Eur J Immunol 2002; 32:3756–3764.
17. van Baarle D, Kostense S, van Oers MH, Hamann D, Miedema F. Failing immune control as a result of impaired CD8+ T-cell maturation: CD27 might provide a clue. Trends Immunol 2002; 23:586–591.
18. Bronke C, Palmer NM, Jansen CA, Westerlaken GH, Polstra AM, Reiss P, et al
. Dynamics of cytomegalovirus (CMV)-specific T cells in HIV-1-infected individuals progressing to AIDS with CMV end-organ disease. J Infect Dis 2005; 191:873–880.
19. Goulder PJ, Tang Y, Brander C, Betts MR, Altfeld M, Annamalai K, et al
. Functionally inert HIV-specific cytotoxic T lymphocytes do not play a major role in chronically infected adults and children. J Exp Med 2000; 192:1819–1832.
20. Shedlock DJ, Shen H. Requirement for CD4 T cell help in generating functional CD8 T cell memory. Science 2003; 300:337–339.
21. Sun JC, Bevan MJ. Defective CD8 T cell memory following acute infection without CD4 T cell help. Science 2003; 300:339–342.
22. Bourgeois C, Rocha B, Tanchot C. A role for CD40 expression on CD8+ T cells in the generation of CD8+ T cell memory. Science 2002; 297:2060–2063.
23. Janssen EM, Lemmens EE, Wolfe T, Christen U, von Herrath MG, Schoenberger SP. CD4+ T cells are required for secondary expansion and memory in CD8+ T lymphocytes. Nature 2003; 421:852–856.
24. Bachmann MF, Hunziker L, Zinkernagel RM, Storni T, Kopf M. Maintenance of memory CTL responses by T helper cells and CD40–CD40 ligand: antibodies provide the key. Eur J Immunol 2004; 34:317–326.
25. Klenerman P. Commentry: T cells get by with a little help from their friends. Eur J Immunol 2004; 34:313–316.
26. Geiselhart LA, Humphries CA, Gregorio TA, Mou S, Subleski J, Komschlies KL. IL-7 administration alters the CD4:CD8 ratio, increases T cell numbers, and increases T cell function in the absence of activation. J Immunol 2001; 166:3019–3027.
27. Berard M, Brandt K, Bulfone-Paus S, Tough DF. IL-15 promotes the survival of naive and memory phenotype CD8+ T cells. J Immunol 2003; 170:5018–5026.
28. Alves NL, Hooibrink B, Arosa FA, van Lier RA. IL-15 induces antigen-independent expansion and differentiation of human naive CD8+ T cells in vitro
. Blood 2003; 102:2541–2546.
29. Napolitano LA, Grant RM, Deeks SG, Schmidt D, de Rosa SC, Herzenberg LA, et al
. Increased production of IL-7 accompanies HIV-1-mediated T-cell depletion: implications for T-cell homeostasis. Nat Med 2001; 7:73–79.
30. Centers for Disease Control and Prevention. 1994 revised classification system for human immunodeficiency virus infection in children less than 13 years of age. Official authorized addenda: human immunodeficiency virus infection codes and official guidelines for coding and reporting ICD-9-CM. MMWR 1994; 43:1–19.