Natural killer (NK) cells are an important component of innate immunity and are functionally classified into three subsets: the cytotoxic subset (CD3negCD16posCD56pos), the cytokine producing subset (CD3negCD16posCD56bri)  and the dysfunctional subset (CD3negCD16posCD56neg) . The dysfunctional NK subset expands and the cytotoxic NK subset decreases in patients with chronic HIV-1 infection compared with those without HIV infection . Unlike T and B cells, NK cells do not require gene rearrangements for their receptor maturation [3,4]. The functions of NK cells are regulated by a number of MHC-I-specific inhibitory NK receptors (iNKRs) and natural cytotoxicity receptors (NCR) . The NKG2A/CD94 heterodimer, a C-type lectin receptor, is a key iNKR on NK cells for specifically recognizing HLA-E molecules on target cells and imparting inhibitory signals  to prevent the lyses of normal lymphoid cells [7,8]. Similar to NKG2A/CD94, the NKG2C/CD94 heterodimer also recognizes HLA-E, but it delivers activation signals to NK cells [5,6]. Normally, the affinity between NKG2A/CD94 complex and HLA-E interaction is 6-fold stronger than that between NKG2C/CD94 heterodimer and HLA-E. Therefore, NKG2C/CD94 complexes usually fail to bind to HLA-E and NK cells remain inhibited [9–11].
Studies have indicated that HLA-E may play an important role in the escape of CD4+ cells from NK-cell-mediated lyses . HIV-1 infection selectively down-regulates the expression of HLA-A and HLA-B on CD4+ T cells but does not significantly affect HLA-C and HLA-E [12,13]. Blocking the inhibitory receptors for HLA-C and HLA-E on NK cells induces NK-cell-mediated cytoxicity against CD4+ T cells . Another study showed that HIV-1 infection induces a switch from an inhibitory receptor (NKG2A) to an activating receptor (NKG2C) on NK cells , thereby resulting in an imbalance between NKG2A- and NKG2C-expressing cells. However, it remains unclear why an increase of NKG2C fails to result in enhanced attack on HIV-1 infected CD4+ T cells by NK cells when the HLA-E expression level is maintained. In this study, we have demonstrated that HIV-1 infection is associated with higher levels NKG2A in the cytotoxic NK subset in HIV-1-infected patients with more advanced disease.
One hundred and fifty-seven patients were enrolled in this study, including 93 patients with chronic HIV-1 infection (infection for more than three months), 20 patients with AIDS and 43 HIV-1 seronegative individuals. All HIV-1 seropositive patients tested positive for HIV-1 by ELISA (Genscreen; Bio-Rad, Hercules, California, USA) and positive test results were confirmed by western blot (GenelabsR Diagnostics HIV BLOT2.2, MP Biomedicals, Solon, OH, USA). Viral load was quantified by fluorescent real-time PCR (LightCycler; Roche, Basel, Switzerland; PCR Fluorogence Diagnostic, PG BioTech, Shenzhen, Guangdong, China) (Table 1). The patients with chronic HIV-1 infection were categorized into three groups: CD4 cell count > 500 cells/μl group (n = 44); CD4 cell count 200–500 cells/μl group (n = 49); and CD4 cell count < 200 cells/μl group (n = 20), which is classified as AIDS. To determine the influence of viremia on NK receptors, each of the first two groups were further classified into subgroups according to their viral loads: aviremic (n = 24) and viremic (n = 20) subgroups for the CD4 cell count > 500 cells/μl group and aviremic (n = 17) and viremic (n = 32) subgroups for CD4 cell count 200–500 cells/μl group. High viremia [viral load (VL) > 1 × 105 copies/ml] and at least one opportunistic infection were observed in all patients with CD4 cell count < 200 cells/μl. To determine the influence of HIV-1 infection time on NK receptors, patients with chronic infection were further classified into two subgroups according to their infection time determined by a sera negative cohort followed every six months: an early chronic infection subgroup (n = 39, infection time less than 2 years) and a chronic infection subgroup (n = 54, infection time more than 2 years). All patients in the study were naive to antiretroviral therapy.
The study protocol was approved by the Institutional Review Board of National Center for AIDS/Sexually Transmitted Disease Control and Prevention, Chinese Center for Disease Control and Prevention, and all patients provided written informed consent before sample collection.
Flow cytometric analysis
Blood samples were collected in EDTA and sent to the State Key Laboratory for Infectious Disease Prevention and Control in China CDC within 8 h. One milliliter of whole blood was used for CD4+ T cell counting and for NK cell-surface molecular analysis using flow cytometry (see below). The plasma was separated and stored at −70°C for viral load quantification. Peripheral blood mononuclear cells (PBMC) were isolated by density gradient centrifugation.
TruCount tubes and TriTEST reagents (CD3-FITC/CD4-PE/CD45-PerCP) were purchased from BD Biosciences, San Jose, CA, USA for counting CD4+ T cells. The NK cells were defined with CD3-APC (BD), CD16-FITC (BD), CD56-PC5 (BD), NKG2A-PE (R&D Systems, Minneapolis, MN, USA), and NKG2C-PE (R&D). Red blood cell lysis buffer was purchased from BD Biosciences. All samples were acquired with FACSCalibur (BD) and 104 lymphocytes were collected per sample.
Natural killer cell cytotoxic function assay
Freshly isolated PBMC were used for quantifying NK cell cytotoxicity against K562 cells pulsed with the fluorescent dye 5,6-carboxyfluorescein diacetate, succinimidyl ester (CFSE, Sigma-Aldrich Chemical, St. Louis, MO, USA) and propidium iodide (PI; Sigma-Aldrich Chemical, St. Louis, MO, USA), as described previously [16,17]. Briefly, K562 cells were washed twice with RPMI 1640 without fetal calf serum (FCS) and resuspended in RPMI 1640 at a density of 1 × 107 cells/ml, and then stained with CFSE at a final concentration of 7.5 μmol/l for 10 min at 37°C and 5% CO2 in the absence of FCS. After labeling, cells were washed twice and incubated for 1 h in complete medium at 37°C and 5% CO2. 2 × 104 K562 target cells were incubated with 4 × 105 fresh PBMC in a 96-well U-bottom plate for 4 h and stained with PI. The apoptotic K562 target cells were shown as PI+ CFSE+ on the histogram with PI and CFSE as axis. The percentage of NK-specific lysis of K562 cells was calculated by using the following formula: NK-specific lysis = [(percentage of PI+ cells in experimental setting – percentage of PI+ cells in negative control) / (100 - percentage of PI+ cells in negative control)] × 100 .
All data are displayed as mean ± SD. Differences between groups are compared using the Student's t-test, with a significance threshold of α < 0.05. Pearson's two parametric correlation analysis and Spearman nonparametric rank test was used in the regression analysis (SPSS Inc, Chicago, Illinois).
Lower levels of cytotoxic natural killer cells in patients with the deteriorated clinical status
After using anti-CD3 antibodies to distinguish CD3neg NK cells from CD3pos T cells, NK cells are functionally classified into three subsets according to their expression level of CD16 and CD56, including CD3negCD16posCD56pos cytotoxic NK cells (80–90% of total NK cells in HIV-1 seronegative participants), CD3negCD16posCD56bri cytokine producing NK cells (2–10% of total NK cells in HIV-1 seronegative participants) and CD3negCD16posCD56neg NK cells (described as the dysfunctional NK cells in HIV/AIDS patients, usually 2–10% of total NK cells in HIV-1 seronegative participants). In this study, the percentage of cytotoxic CD3negCD16posCD56pos NK cells was significantly lower even in the HIV-1 infected group with a CD4+ T-cell count > 500 cells/μl, compared to HIV-1 seronegative group (P < 0.001), and the percentage of these cells was even lower in patients with deteriorated clinical status. In contrast, the percentage of dysfunctional CD3negCD16posCD56neg NK cells was higher in patients with advanced disease (P < 0.001), whereas the proportion of CD3negCD16posCD56bri NK cells remained constant (Fig. 1a). In accordance with a previous report that the skewed composition of NK cell subsets has been observed since acute HIV-1 infection , our study indicated that the impairment of NK cells occurred during the early stages of HIV-1 infection and is continuously aggravated with decreasing CD4+ T cells.
NKG2A expression was higher in functional cytotoxic NK cells and lower in dysfunctional natural killer cells
As NKG2A and NKG2C play an important role in determining the functionality of NK cells , we compared the expression of NKG2A and NKG2C in cytotoxic NK cells (CD3negCD16posCD56pos) among HIV-infected populations with different clinical status. The percentage of NKG2A+ cytotoxic NK cells in HIV-1 seronegative individuals (n = 43, range: 6.02– 78.01%, mean: 26.64%) was similar to the HIV-1-infected group with CD4 cell count > 500 cells/μl (n = 44; range: 4.44–100%, mean: 25.17%). The percentage of NKG2A+ cytotoxic NK cells in patients with CD4 200–500 cells/μl (n = 49, range: 0.72–100%, mean: 40.85%, P < 0.001) and AIDS (n = 20, range: 23.63–100%, mean: 48.28%, P < 0.001,) were significantly higher, (Fig. 1b, left panel). Interestingly, the percentage of NKG2C+ NK cells in the cytotoxic NK subsets in all HIV-1 clinical stage was higher than that in the HIV-1 seronegative group (Fig. 1b, right panel). These data suggested that the proportion of NKG2A cytotoxic NK cells increases with the decrease of CD4+ T cells, as does the proportion of NKG2C+. To further confirm this observation, we examined NKG2A and NKG2C expression in groups with different HIV infection time, by comparing the early chronic HIV-infected group (n = 39, infected for 3–24 months), the chronic HIV-infected group (n = 55, infected > 24 months) and the AIDS group (n = 20, CD4 cell count < 200 cells/μl). A similar pattern was observed (Fig. 1c).
To explain the controversy between the higher percentage of NKG2A in cytotoxic NK subset and the decrease of NKG2A proportion in total NK cells as previously reported [15,19], we further examined the percentage of NKG2A+ cells in the dysfunctional NK cells (CD3negCD16posCD56neg NK cells), which are profoundly increased with disease progression (Fig. 1a). Our study showed that the percentage of NKG2A+ was significantly lower in the dysfunctional NK cells in patients at all stages of HIV-1 infection, but there were no differences in the percentage of NKG2C in dysfunctional NK cells in individuals with and without HIV-1 infection. (Fig. 1d). The low level of cytotoxic NK cells and the high proportion of dysfunctional NK cells was associated with the low level of NKG2A in total NK cells (data not shown).
The effect of viremia on the expression of NKG2A in the cytotoxic natural killer subset
Previous studies have reported that HIV viremia affects the expression of NKG2A [15,19]. Therefore, we investigated the expression of NKG2A in HIV infected patients with and without viremia. No significant difference in NKG2A expression was observed in aviremic and viremic subgroups of patients with CD4 cell count > 500 cells/μl. In patients with CD4+ T cell counts 200–500 cells/μl, the level of NKG2A cytotoxic NK cells is significantly higher in the viremic subgroups than that in the aviremic subgroup (P < 0.05) (Fig. 2a). These data suggest that viremia may not affect NKG2A expression in a short period of time but may increase the expression of NKG2A, after a prolonged time of interaction with NK cells. In contrast, no statistical differences in NKG2C expression were observed between aviremic and viremic subgroups of HIV positive patients with either CD4 cell count > 500 or 200–500 cells/μl l (Fig. 2a).
Natural killer cytotoxicity was lower in those with more advanced clinical status
We next examined the cytotoxicity of NK cells in patients with deteriorated clinical status. Since cytotoxicity against K562 cells is mediated by cytotoxic NK cells , we evaluated this subset's functionality by testing the cytolytic activity of PBMCs on K562 cells . As shown in Fig. 2b, NK cytotoxic activity against K562 cells was one-third lower in the HIV-1-infected group with CD4 cell count > 500 cells/μl l compared to HIV-1 seronegative group. NK cytotoxic activity was further decreased in those with advanced clinical status, reaching the lowest level in the AIDS group (E: T ratio = 20:1). These data indicate that the cytotoxic function of NK cells was impaired in all HIV-1 infection stages and was associated with disease progression.
The reverse association between the expression of NKG2A in cytotoxic natural killer subset and CD4 cell count.
Considering the NKG2A might play a role in the escape of HIV-1 infected CD4 cells from NK-mediated cytotoxicity in HIV-1 infected individuals, we performed a regression analysis of the relationship between the percentage of NKG2A in cytotoxic NK subset and total CD4 cell count. A reverse association between the percentage of NKG2A in cytotoxic NK subset and total CD4 cell count was observed in all HIV-1 positive groups (n = 113, r = –0.36) (Fig. 3a). Further analysis showed higher reverse association between NKG2A and CD4 cells count in the viremic group (n = 70, r = -0.40) (Fig. 3b) than in the aviremic group (n = 41, r = –0.20) (Fig. 3c). The results suggest that NKG2A might contribute to the escape of HIV-1 infected CD4 from NK-mediated cytotoxicity in viremic patients and consequently result in the persistent infection and loss of CD4 cells.
This study demonstrated that the proportion of cytotoxic NK cells was low in patients with increased dysfunctional NK cells during HIV-1 infection. NKG2A expression was high in the cytotoxic NK subset with advanced clinical status but significantly low in dysfunctional NK subset. The low level of cytotoxic NK cells and higher levels of NKG2A in cytotoxic NK subset could be associated with the observed reduction of NK-mediated cytotoxicity and thereby may contribute to protecting HIV-1 infected CD4 cells from attacking by NK cells.
It has been reported that HIV-1 infected CD4 cells may be able to escape the attack from cytotoxic T lymphocytes (CTL) and NK cells through the selective effect of HIV-1 on HLA expression in CD4 cells . HIV-1 down-regulates HLA-A and -B but does not significantly affect HLA-C and -E [12,20]. Bonaparte and coworkers have showed that NK cells lacking inhibitory receptors to HLA-C and HLA-E kill HIV-1 infected CD4 cells . Martini and coworkers' study showed that NKG2A-CD94/HLA-E interaction might contribute to NK cell dysfunction in destroying the HIV infected CD4 cells . It is known that NKG2A/CD94 is one of the key inhibitory receptors for HLA-E on NK cells and the interaction between them delivers inhibitory signals to and thereby disarms NK cells. Therefore, NKG2A-CD94/HLA-E interaction might play an important role in protecting HIV-1-infected CD4+ T cells from NK cells attack. However, previous studies have also shown that the proportion of NKG2A cells in total NK cells was decreased in HIV-1-infected patients, whereas NKG2C cells were increased [15,19], which might result in increased NK cell-cytotoxicity against host lymphocytes. Those previous studies are unable to explain the fact that HIV-1 infected CD4 cells consistently escape from the attack of NK cells . Since CD3negCD16posCD56pos NK cells are the NK cells responsible for cytotoxic attack , we hypothesized that HIV-1 infected CD4 cells are protected from NK-mediated cytotoxicity via the inhibition of cytotoxic function in CD3negCD16posCD56pos NK cells. Since NKG2A is a key receptor for the delivery of inhibitory signals to cytotoxic NK cells (CD3negCD16posCD56pos NK cells), whereas NKG2C transmits an activation signal, we studied the expression of NKG2A and NKG2C in the cytotoxic NK subset in HIV-1 infected patients at different stages of disease progression. Our data showed that NKG2A cytotoxic NK subset was higher in patients with advanced clinical status, and there was a reverse association between the percentage of NKG2A in cytotoxic NK subset and the total CD4 cell count. All the results together suggest that NKG2A may be involved in inhibiting the action of cytotoxic NK cells against HIV-1-infected CD4 cells by the interaction with HLA-E consequently resulting in the persistent infection and loss of CD4 cells. Moreover, we also observed that NKG2C-expressing cells were higher in chronic HIV-1 infection but not in AIDS patients. The binding affinity between NKG2A/CD94 and HLA-E is 6-fold stronger than that between NKG2C/CD94 and HLA-E, so it is unlikely that NKG2C/CD94 would compete with NKG2A/CD94 for HLA-E binding to deliver activating signals [10,11]. Although a high level of NKG2C expression in cytotoxic NK subset was observed during HIV-1 infection progression, the high level of NKG2A expression on cytotoxic NK cells might overwhelm the effect caused by NKG2C. This might explain the significant decrease of cytotoxic function of NK cells in HIV-1 infected individuals.
Our study also suggests that the expansion of dysfunctional NK cells during HIV-1 infection and the decrease of NKG2A in this population may account for the low level of NKG2A in total NK cells. Viremia may increase the expression of NKG2A over a prolonged period of interaction with NK cells but have no effect on NKG2C. However, due to the limited number of participants and the use of a cross-sectional methodology in this study, our conclusions need to be confirmed in a larger sample followed through time. In addition, a method to assess the activities of cytotoxic NK cells against HIV-1-infected CD4 cells should be established to gain direct evidence for how the increase of NKG2A expression influences the destruction of HIV-1-infected CD4 cells by NK cells.
Overall, NKG2A expression was high in the cytotoxic NK subset of HIV-1 infected patients with advanced clinical status. Since the interaction between NKG2A-CD94 and HLA-E delivers inhibitory signals to NK cells, elevated NKG2A expression may be associated with escape of HIV-1-infected CD4 cells from the attack of NK cells in patients with late stage HIV infection.
We thank Amy Wu for editorial help, and Bo Sheng for assistance for technical support.
Sponsorship: This project was supported by the China Comprehensive Integrated Programs for Research on AIDS (CIPRA U19AI51915), the National Institute of Allergy and Infectious Diseases, National Institute of Health, USA, China National Key Technologies R&D Program of the 10th Five-Year Plan (No.2004BA719A01) and by China National Program on Key Basic Research Projects (973 Program) (No. 2005CB522903), the China Ministry of Science and Technology, China.
1. Cooper MA, Fehniger TA, Caligiuri MA. The biology of human natural killer-cell subsets. Trends Immunol 2001; 22:633–640.
2. Mavilio D, Lombardo G, Benjamin J, Kim D, Fullman D, Marcenaro E, et al. Characterization of CD56-/CD16+ natural killer (NK) cells: a highly dysfunctional NK subset expanded in HIV-infected viremic individuals. Proc Natl Acad Sci U S A 2005; 102:2886–2891.
3. Reynolds CW, Bonyhadi M, Herberman RB, Young HA, Hedrick SM. Lack of gene rearrangement and mRNA expression of the beta chain of the T cell receptor in spontaneous rat large granular lymphocyte leukemia lines. J Exp Med 1985; 161:1249–1254.
4. Trinchieri G. Biology of natural killer cells. Adv Immunol 1989; 47:187–376.
5. Lanier LL. NK cell receptors. Annu Rev Immunol 1998; 16:359–393.
6. Lee N, Llano M, Carretero M, Ishitani A, Navarro F, Lopez-Botet M, et al. HLA-E is a major ligand for the natural killer inhibitory receptor CD94/NKG2A. Proc Natl Acad Sci USA 1998; 95:5199–5204.
7. Leibson PJ. Cytotoxic lymphocyte recognition of HLA-E: utilizing a nonclassical window to peer into classical MHC. Immunity 1998; 9:289–294.
8. Lanier LL. Follow the leader: NK cell receptors for classical and nonclassical MHC class I. Cell 1998; 92:705–707.
9. Vales-Gomez M, Reyburn HT, Erskine RA, Lopez-Botet M, Strominger JL. Kinetics and peptide dependency of the binding of the inhibitory NK receptor D94/NKG2-A and the activating receptor CD94/NKG2-C to HLA-E. EMBO J 1999; 18:4250–4260.
10. Lanier LL. Face off–the interplay between activating and inhibitory immune receptors. Curr Opin Immunol 2001; 13:326–331.
11. Kaiser BK, Barahmand-Pour F, Paulsene W, Medley S, Geraghty DE, Strong RK. Interactions between NKG2x immunoreceptors and HLA-E ligands display overlapping affinities and thermodynamics. J Immunol 2005; 174:2878–2884.
12. Cohen GB, Gandhi RT, Davis DM, Mandelboim O, Chen BK, Strominger JL, et al. The selective downregulation of class I major histocompatibility complex proteins by HIV-1 protects HIV-infected cells from NK cells. Immunity 1999; 10:661–671.
13. Nattermann J, Nischalke HD, Hofmeister V, Kupfer B, Ahlenstiel G, Feldmann G, et al. HIV-1 infection leads to increased HLA-E expression resulting in impaired function of natural killer cells. Antivir Ther 2005; 10:95–107.
14. Bonaparte MI, Barker E. Killing of human immunodeficiency virus-infected primary T-cell blasts by autologous natural killer cells is dependent on the ability of the virus to alter the expression of major histocompatibility complex class I molecules. Blood 2004; 104:2087–2094.
15. Mela CM, Burton CT, Imami N, Nelson M, Steel A, Gazzard BG, et al. Switch from inhibitory to activating NKG2 receptor expression in HIV-1 infection: lack of reversion with highly active antiretroviral therapy. AIDS 2005; 19:1761–1769.
16. Marcusson-Stahl M, Cederbrant K. A flow-cytometric NK-cytotoxicity assay adapted for use in rat repeated dose toxicity studies. Toxicology 2003; 193:269–279.
17. Kottilil S, Chun TW, Moir S, Liu S, McLaughlin M, Hallahan CW, et al. Innate immunity in human immunodeficiency virus infection: effect of viremia on natural killer cell function. J Infect Dis 2003; 187:1038–1045.
18. Alter G, Teigen N, Davis BT, Addo MM, Suscovich TJ, Waring MT, et al. Sequential deregulation of NK cell subset distribution and function starting in acute HIV-1 infection. Blood 2005; 106:3366–3369.
19. Mavilio D, Benjamin J, Daucher M, Lombardo G, Kottilil S, Planta MA, et al. Natural killer cells in HIV-1 infection: dichotomous effects of viremia on inhibitory and activating receptors and their functional correlates. Proc Natl Acad Sci U S A 2003; 100:15011–15016.
20. Ahmad R, Sindhu ST, Tran P, Toma E, Morisset R, Menezes J, et al. Modulation of expression of the MHC class I-binding natural killer cell receptors, and NK activity in relation to viral load in HIV-infected/AIDS patients. J Med Virol 2001; 65:431–440.
21. Martini F, Agrati C, D'Offizi G, Poccia F. HLA-E up-regulation induced by HIV infection may directly contribute to CD94-mediated impairment of NK cells. Int J Immunopathol Pharmacol 2005; 18:269–276.
© 2007 Lippincott Williams & Wilkins, Inc.