Natural killer (NK) cells play a crucial role in the first line of defense against pathogen-infected cells.1 Within the context of HIV infection, the ability of NK cells to mediate antibody-dependent cellular cytotoxicity (ADCC) is critical for protection from disease acquisition and progression.2 Indeed, the possible correlation between protection and ADCC-mediating antibodies in the RV144 trial3 has increased awareness toward the protective role of ADCC-mediating antibodies and the contribution of effector NK cells.
HIV infection results in chronic immune activation4 and persistent CD4+ and CD8+ T-cell activation correlate directly with NK cell activation.5 Indeed, HIV infection also leads to changes in NK cell subsets, phenotype, and cytolytic function.6–9 The changes in NK cell homeostasis is particularly pronounced in the chronic phase,6 although changes are found to have occurred during the acute stage of HIV infection.7 NK cells are often divided into CD56neg and CD56pos subsets. The dysfunctional CD56neg NK cells expand during chronic HIV infection and display a subset with a significantly lower cytolytic activity and ability to secrete cytokines compared with CD56pos NK cells.10 In turn, CD56pos NK cells are often divided into CD56dim and CD56bright NK cells, which are known for their cytolytic activity and regulatory role, respectively.7 CD56dim cells express high levels of CD16 (FcγRIIIa), a receptor responsible for triggering ADCC.4,11
HIV infection also leads to changes in the NK cell phenotype. In individuals with a chronic HIV infection, the frequency of NK cells expressing CD57, CD27, CD70, and CCR7 is upregulated.8,9,12 CD57 expression is linked to the maturation of NK cells12,13 and is acquired on CD56dim NK cells upon activation. Moreover, CD57pos NK cells are more responsive to signaling through the ADCC-activating CD16 receptor compared with CD57neg NK cells.13 CD57 expression also correlates with the expression of Granzyme B.14 As the CD57 receptor, the CD27 receptor is linked to NK cell maturation. In peripheral blood, the surface density of the CD27 receptor decreases with NK cell maturation, and the lack of CD27 expression on NK cells is associated with high cytolytic function.15 In addition, the majority of CD27neg NK cells belong to the CD56dim population, and the highest CD16pos expression is found to be on CD27neg NK cells. The CD27 receptor is mainly expressed on the CD56bright NK cell population, which is poorly cytotoxic.16 CD70 is a ligand of the CD27 receptor. It has been suggested that IL-7 may contribute to the upregulation of CD70, thereby acting in NK cell dysfunction in HIV infection.8 CCR7 is a known receptor for lymphoid tissue homing, and CD56bright NK cells express CCR7, in line with this subset being the main population present in lymph nodes.17
Moreover, impaired NK cell function during HIV infection has been linked to the reduced surface expression of the natural cytotoxicity receptor NKp46.18 The peripheral blood NK cells of HIV-infected individuals express a reduced surface density of NKp46 compared with healthy individuals,18 and AIDS patients have significantly lower NKp46 expression compared with non-AIDS patients.5
It has been shown that the suppression of HIV viral load by antiretroviral therapy in part restores lytic activity19 and NK cell-mediated killing.20 In this longitudinal study, we evaluated how ADCC function changes in HIV-infected individuals before and after highly active antiretroviral therapy (HAART). We found that the ability of effector cells to mediate ADCC was improved already after 6 months of therapy. In contrast, the ability of antibodies to mediate ADCC was unchanged or decreased. Although the improvement in the ability of NK cells to mediate ADCC did not correlate with general immune restoration, as measured by increases in CD4+ T-cell counts, it did correlate with a normalization in the frequency of NK cells expressing CCR7 and CD27. In turn, these changes correlated with increased cytotoxic capacity.
Study Individuals and Informed Consent
Peripheral blood mononuclear cells (PBMCs) were obtained from HIV-1–infected individuals followed at the University hospital of Copenhagen and Hvidovre Hospital. The study was approved by the National Committee for Health Research Ethics of the Danish Ministry of Health (H-3-2011-089), and individuals were included after obtaining written and verbal informed consent. HIV-negative plasma was obtained from HIV-negative individuals enrolled in another study approved by the National Committee for Health Research Ethics of the Danish Ministry of Health (H-3-2013-104). HIV immunoglobulin G (HIVIG) was obtained from the NIH AIDS Research and Reagent Program. A single buffy coat from a HIV-negative individual was obtained from the Danish blood bank, and PBMCs were obtained by density gradient centrifugation and cryopreserved until use. Table 1 outlines the clinical characteristics of the included male patients. Plasma viral load was quantified by the COBAS Ampliprep/COBAS TaqManHIV-1 Test (version 2.0 system; Roche Diagnostics, Copenhagen, Denmark). Absolute CD4+ T-cell counts were determined using the FACS Count system (BD Bioscience, San Jose, CA) according to the manufacturer's protocol. High-sensitivity C-reactive protein (CRP) was determined using the Brahms CRPus Kryptor (Thermo Scientific, Hennigsdorf, Germany) according to the manufacturer's protocol. If CRP was undetectable, the detection limit of the assay (<0.06 mg/L) is shown in Table 1 and used for calculations.
PBMCs were used as the source of NK cells. Cryopreserved PBMCs were thawed and rested overnight at 2 × 106 cells per milliliter in R10 (RPMI, Gibco; Life Technologies, Naerum, Denmark) supplied with 10% fetal bovine serum (Gibco) and 1% penicillin–streptomycin (Gibco) at 37°C and 5% CO2.
CEM.NKRCCR5 cells21 were coated with recombinant gp120 (0.01 mg/mL) and used as target cells. Recombinant gp120 HIV-1 representing the envelopes of subtype B (BaL; Immune Technology, New York, NY) and subtype AE (CM243; Science Protein, Meiden, CT) were used. The protein used for coating the target cells was matched for the patients' HIV-1 subtype (Table 1).
The ADCC-GranToxiLux assay (OncoImmunin, Gaithersburg, MD) was performed as described.22 In brief, CEM.NKRCCR5 cells were coated with gp120 and labeled with TFL4 and NFL1. The effector (E) and target (T) cells were tested in an E:T ratio of 30:1. The plasma/antibodies were tested in 5-fold dilutions starting at 1:300. Direct NK cell activity was tested and excluded for each PBMC donor. The cells were acquired using BD LSRII and analyzed using FlowJo (Treestar version 8.8.7; Treestar, Ashland, OR). ADCC for each patient was defined as the highest percentage of Granzyme B activity of the effector cells after background subtraction at a certain dilution before and after 6 months of HAART. If the dilutions were conflicting, we choose the dilution at which ADCC was the highest.
Staining of NK Cell Receptors
Thawed PBMCs (1 × 106) were incubated for 20 minutes at room temperature in the dark and in the presence of LIVE/DEAD Fixable Dead Cell Stain Kit marker (Life Technologies), anti-CD3 PerCP (BD Biosciences), anti-CD8 Qdot605 (Life Technologies), anti-CD56 AlexaFluor488 (eBioscience, San Diego, CA), anti-CD16 APC-H7 (BD Biosciences), anti-NKp46 PECy7 (BD Biosciences), anti-CD57 (BD Biosciences), anti-CD27 AlexaFluor700 (eBioscience), anti-CCR7 PerCP/Cy5.5 (BioLegend, London, United Kingdom), and anti-CD70 PE (BD Biosciences) antibodies. The antibodies were pretitrated to obtain the optimal concentration for staining. Subsequently, cells were washed and fixed using BD stabilizing fixative (BD Biosciences). The stained cells were acquired using BD LSRII and analyzed using FlowJo. For detailed gating strategies, see Supplemental Digital Content 1 (http://links.lww.com/QAI/A590). The NK cells were defined as CD3neg CD56pos/neg CD16pos. The total NK cell percentages and NK cell subsets are expressed as a percentage of the total lymphocytes.
Anti-HIV gp120 Enzyme-Linked Immunosorbent Assay
Anti-gp120–specific IgG subtype titers were determined by enzyme-linked immunosorbent assay, as previously described.23 The data were read and illustrated as the absorbance at the same plasma dilution before and at 6 months after HAART. The assay was modified by the use of 4 different conjugates: mouse anti-human IgG1, IgG2, IgG3, and IgG4 (Life Technologies).
The data analysis was performed using GraphPad Prism 6 software (version 6.0c; GraphPad Software Inc, La Jolla, CA), and the medians and interquartile ranges were calculated. The Wilcoxon matched-pairs signed-rank test was used to test the difference between before HAART and 6 months after HAART. A Spearman 2-tailed test was used to test for correlations. The Mann–Whitney 2-tailed test was used to test the differences between ADCC responders and ADCC nonresponders.
ADCC Activity of NK Cells Increases After 6 Months of HAART
ADCC activity in HIV-infected individuals was compared before and after 6 months of HAART in 3 different setups: (1) the ability of patients' peripheral blood mononuclear (PBM) effector cells (as a source of NK cells) to mediate ADCC was tested using the patients' PBMCs together with HIVIG, (2) the ability of the patients' antibodies to perform ADCC was tested using the patients' plasma together with healthy donor PBMCs, and (3) the ability of the patients' antibodies and patients' PBM effector cells to perform ADCC was tested in an autologous model.
The ADCC activity of PBMCs increased after 6 months of HAART (P = 0.0494; Figure 1A). The 12 individuals showing an increase in ADCC activity are here defined as ADCC responders (Fig. 1A), and the 7 individuals with a decrease in PBM effector cell ADCC activity are defined as ADCC nonresponders (Fig. 1A).
In contrast, the ability of patients' antibodies (ie, plasma) to mediate ADCC activity was not significantly different after 6 months of HAART (P = 0.9756; Fig. 1B). For 10 individuals, an increase was found in the ability of their antibodies to mediate ADCC after 6 months of HAART compared with a decrease for 9 individuals (Fig. 1B). Individuals defined as ADCC responders in Figure 1A are not necessary ADCC responders when considering antibody function. Interestingly, the IgG1, IgG2, and IgG3 anti-HIV antibody titers were significantly decreased after 6 months of HAART (P = 0.0186, P = 0.0149, and P = 0.0003, respectively; see Supplemental Digital Content 2A-C,http://links.lww.com/QAI/A590), whereas no change was found for the anti-HIV IgG4 antibody titers (P = 0.1531; see Supplemental Digital Content 2D,http://links.lww.com/QAI/A59).
The ability of the patients' PBM effector cells and antibodies in combination to mediate ADCC in an autologous model was not significantly different after 6 months of HAART (P = 0.5678; Fig. 1C). The responses in the autologous model seemed to be most influenced by the effector cells, with 8 of the 12 ADCC responders also displaying an increase in ADCC activity in this model. Regarding the 4 ADCC responders, the decreased activity observed in this autologous model may be explained by the decreased ability of their antibodies to mediate ADCC at 6 months after HAART (Fig. 1B).
Difference Between ADCC Responders and Nonresponders Is not Explained by Plasma Viral Load, CD4+ T-Cell Count, or NK Cell Count
The HIV-positive participants varied in regard to viral loads and CD4+ T-cell count at the time of treatment initiation (Table 1). All responded to treatment, with a decrease in viral replication and increasing CD4+ T-cell counts after 6 months (Table 1). The viral load and CD4+ T-cell count were not different at the time of treatment initiation between ADCC responders and nonresponders (P = 0.8930 and P = 0.8540, respectively, data not shown). The decrease in viral load (data not shown) or increase in CD4+ T-cell count after 6 months of HAART was not significantly different between the ADCC responders and nonresponders (P = 0.7892; Fig. 2A). The immune activation, as measured by CRP, was not different between responders and nonresponders (P > 0.9999, data not shown), and no correlation was found to ADCC (P = 0.5708, data not shown).
NK cells are often subdivided into CD56dim CD16pos and CD56neg CD16pos NK cells. No changes in the proportion of these different subsets before and after 6 months of HAART were detected (data not shown). As CD16 is the major NK cell receptor responsible for ADCC, we therefore focused our analysis of NK cells on CD3neg CD16pos cells (see Supplemental Digital Content 1,http://links.lww.com/QAI/A59). No difference was found in the median proportion of CD16pos NK cells before HAART (7.2%, interquartile range: 4.0–14.8) versus after 6 months of HAART (5.0%, 2.4–10.2) (P = 0.4835), and no difference was found between the ADCC responders and nonresponders (P = 0.6819; Fig. 2B). Similarly, the median number of CD16pos NK cells was unchanged before and at 6 months after therapy initiation at 0.12 × 109/L (0.05–0.32) and 0.012 × 109/L (0.04–0.27) (P = 0.7983), respectively, and no difference was found in the delta percentage numbers of CD16pos NK cells between the 2 groups (P = 0.0978; Fig. 2C). Moreover, the difference between the ADCC responders and nonresponders could not be explained by a higher cell surface expression of CD16 after 6 months of HAART, as the delta mean fluorescence intensity of CD16 was not significantly different between the 2 groups (P = 0.0670; Fig. 2D).
In ADCC Responders, the Expression of CD27 and CCR7 on CD16pos NK Cells Is Decreased After 6 Months of Treatment
The frequency of CD16pos NK cells expressing CCR7, CD27, NKp46, CD57, and CD70 was determined to investigate whether the frequency of NK cells expression changed after 6 months of HAART and whether this contributed to the difference between the ADCC responders and ADCC nonresponders. The responders exhibited a significant decrease in the frequency of NK cells expressing CCR7 (P = 0.0269; Fig. 3A) and CD27 (P = 0.0122; Fig. 3B), which was not observed in the nonresponders. In contrast, the CD16pos NK cell expression of NKp46 was unchanged in the ADCC responder group but was significantly decreased in the nonresponder group after 6 months of HAART (Fig. 3C). No change in expression was observed for the frequency of NK cells expressing CD57 or CD70 in the ADCC responders and nonresponders (Figs. 3D, E). Also, the number of CD57 and CD70 expressing NK cells were unchanged after 6 months of HAART. However, pooling all individuals, we observed a significant downregulation in the median frequency of CD16pos NK cells expressing CD70, from 13.9% (8.4–15.7) to 11.1% (4.3–16.4) (P = 0.0165, data not shown), after 6 months of HAART.
Interestingly, there was an inverse correlation between the frequency of CD16pos NK cells expressing CCR7 and Granzyme B activity at 6 months after treatment initiation (P = 0.0021; Fig. 4A). The same tendency was found for the frequency of NK cell expression of CD27 (P = 0.0563; Fig. 4B). For the CD56dim CD16pos NK cell subset, the inverse correlation between CD27 and Granzyme B was significant (P = 0.0481, data not shown). Such correlations were not observed before HAART or for the other markers (data not shown). This indicates that normalization of the NK cell receptor expression can explain the difference between ADCC responders and ADCC nonresponders.
In this longitudinal study, ADCC activity in HIV-1–positive individuals was evaluated before and after 6 months of HAART. Comprehensive analyses were performed using 3 different setups to study NK cells alone, antibodies alone, and the combination of effector cells and antibodies in an autologous model. After 6 months of therapy, evidence of improvement in NK cell function was shown, whereas no change was observed in ADCC-mediating antibodies. Individuals with an increase in ADCC activity were defined as ADCC responders, and individuals with no improvement were defined as ADCC nonresponders.
We were intrigued to further explore what distinguishes an ADCC responder. Most likely, the most obvious explanation would be that the improved NK cell function in ADCC responders was a reflection of general immune restoration, as measured by increases in CD4+ T-cell counts and reductions in plasma viral load during HAART. However, there was no correlation between ADCC and immune restoration, and neither of these parameters differed between the ADCC responders and ADCC nonresponders. Moreover, the individuals' nadir CD4+ T-cell count did not influence improvement in NK cell function, as measured by ADCC. The individuals included in this study did not necessarily have a very low CD4+ T-cell count when initiating therapy. Possibly the differences seen after 6 months of HAART would have been even larger if more severely immunocompromised individuals would have been included.
As the NK cell subset distribution is known to change during HIV infection, with an increase in dysfunctional CD56neg NK cells,10 we further evaluated whether any normalization in NK cell subsets (CD56dim CD16pos and CD56neg CD16pos NK cells) after HAART could explain the difference between improved ADCC activity. The ADCC-mediating ability of NK cells was significantly increased after 6 months of HAART, despite an unchanged NK cell subset distribution. In agreement, another study assessing NK cell cytotoxicity in HIV-infected individuals after antiretroviral therapy found that NK cell-mediated killing was increased,20 without changes in the NK cell subset distribution. The changes in NK cell homeostasis are reported to occur during acute infection,24 and several studies indicate that the normalization of NK cell subsets requires years of HAART.8,19,20,25 This has led to speculations on the dichotomy between functional and phenotypic recovery in NK cell subsets after antiretroviral therapy.20 Here, a relatively short period of HAART, that is, 6 months, was studied, which already improved ADCC NK cell function.
The CD16 receptor is the NK cell receptor known to trigger an ADCC response; accordingly, this study focused on NK cells expressing this receptor. However, the improved ability of HIV-infected individuals' effector PBMCs to mediate ADCC could not be explained by a change in the average proportion or the number of NK cells expressing CD16 or the surface density of CD16 after 6 months of HAART.
As the PBMC effector function increased during 6 months of HAART, despite no correlation with immune restoration or change in the subset distribution, a more in-depth analysis of the receptor expression of CD16pos NK cells was performed. The frequency of NK cells expressing CCR7, CD27, CD57, CD70, and NKp46 was determined before and after 6 months of HAART. CCR7, CD27, CD57, and CD70 are all known to be upregulated during HIV infection,8,12,13,15,26 whereas NKp46 is downregulated.18,27 In the ADCC responders, there were a significant decrease in the frequency of NK cells expressing CCR7 and CD27. In line with this, there was an inverse correlation between the frequency of CD56dim CD16pos NK cell expressing CCR7 and CD27 and the percent Granzyme B activity after 6 months of HAART. This finding is supported by other studies in which the lack of CD27 expression on NK cells was associated with high cytolytic function.15 In addition, a correlation between the frequency of fewer NK cells expressing CCR7 and Granzyme B expression has previously been indicated.26 However, previous studies have been unable to show the correlation between NK cell phenotype normalization and improved ADCC function during HAART that was observed in the present longitudinal study.
The frequency of NK cell expression of CD57 and CD70 did not differ between ADCC responders and ADCC nonresponders. Nevertheless, a significant increase in the frequency of CD57-expressing CD56neg CD16pos NK cells was found when combining all individuals (ADCC responders and nonresponders, data not shown) after 6 months of HAART. This increase could indeed influence the cytotoxicity of this population, which is generally known to be dysfunctional yet still able to mediate ADCC. Again, pooling all individuals, there was a significant downregulation in the frequency of CD70-expressing CD16pos NK cells. CD70 has been linked to NK cell dysfunction, being upregulated by IL-7 during HIV disease and downregulated during HAART.8 Measurement of IL-7 in this study would have been interesting considering the changes seen in CD70 expression; however, due to limited amount of plasma, this was not possible. A downregulation of CD70-expressing CD16pos NK cells may result in higher cytotoxic activity.8
The reduced surface expression of NKp46 is associated with impaired cytolytic function in viremic HIV-infected individuals.18 The results presented here indicate a relationship between decreases in the frequency of NKp46-expressing NK cells and reduced Granzyme B activity, as the NKp46-expressing CD16pos NK cells declined in the ADCC nonresponder group; this downregulation in the frequency of NKp46-expressing NK cells was not observed in the ADCC responder group.
Although NK cell function improved during HAART, no changes were observed in the ability of antibodies to mediate ADCC. Interestingly, a significant decrease in the anti-gp120 antibody binding titer of IgG1, IgG2, and IgG3 after 6 months of HAART was observed, which did not seem to affect ADCC activity. Despite ADCC assay differences used in previous studies, a correlation between IgG titers and ADCC was not found.28,29 In a recent study, ADCC antibody titers correlated with plasma HIV RNA, but at the same time, the ADCC antibody titers were significantly lower in viremic patients despite high viral loads compared with HIV controllers.30 This dichotomy suggests that the development of sufficient amounts of ADCC antibodies requires adequate levels of CD4+ T helper cells, simultaneous with the presence of antigen to maintain antibody-producing B cells.30 The very low viral replication in patients receiving HAART may be too low for the stimulation of plasma cells to produce high titers of ADCC-protective antibodies.
Finally, in our autologous model of antibodies in combination with effector PBMCs, no significant increase in ADCC was found after 6 months of HAART. Other studies have shown that despite the well-described defects in various NK cell functions during progressive HIV,10,31 the ability of NK cells to secrete Granzyme B was significantly higher in HIV-infected individuals compared with healthy donors.32 This suggests that NK cells could be preactivated and capable of responding to self-antibodies in vivo.32 Improvement in NK functionality after 6 months of HAART in combination with autologous antibodies may be the optimal and most physiological relevant combination to achieving higher ADCC activity. Nonetheless, this was not addressed in our study.
In conclusion, NK cell cytotoxic function, as measured by ADCC, was improved after 6 months of HAART, despite the lack of normalization in the NK subset distribution. Knowing that the normalization of NK cell subsets requires years of HAART,8,19,20,25 it is an important finding that the ability to mediate ADCC was improved already after 6 months not at least in the context of a therapeutic vaccine aiming at inducing ADCC-mediating antibodies. A correlation between cytotoxic activity and immune restoration was not found. Moreover, the nadir CD4+ T-cell count did not influence ADCC activity. However, we did find changes in the NK cell receptor expression indicating the positive influence of HAART because the frequency of NK cells expressing CCR7 and CD27 was altered toward normalization. These changes in NK cell phenotype correlated with increased cytotoxic activity.
The authors thank the studied individuals for their participation in this study. They also thank Birgit Knudsen for technical assistance, Kristina Thorsteinsson for help with the inclusion of a patient, and Michael Christiansen and Pia Lind for technical assistance with high-sensitivity CRP measurements. The NIH AIDS Research and Reagent Program donated standard antibodies.
1. Vivier E, Raulet DH, Moretta A, et al.. Innate or adaptive immunity? The example of natural killer cells. Science. 2011;331:44–49.
2. Fauci AS, Mavilio D, Kottilil S. NK cells in HIV infection: paradigm for protection or targets for ambush. Nat Rev Immunol. 2005;5:835–843.
3. Haynes BF, Gilbert PB, McElrath MJ, et al.. Immune-correlates analysis of an HIV-1 vaccine efficacy trial. N Engl J Med. 2012;366:1275–1286.
4. Paiardini M, Muller-Trutwin M. HIV-associated chronic immune activation. Immunol Rev. 2013;254:78–101.
5. Bisio F, Bozzano F, Marras F, et al.. Successfully treated HIV-infected patients have differential expression of NK cell receptors (NKp46 and NKp30) according to AIDS status at presentation. Immunol Lett. 2013;152:16–24.
6. Alter G, Teigen N, Davis BT, et al.. Sequential deregulation of NK cell subset distribution and function starting in acute HIV-1 infection. Blood. 2005;106:3366–3369.
7. Milush JM, Lopez-Verges S, York VA, et al.. CD56negCD16(+) NK cells are activated mature NK cells with impaired effector function during HIV-1 infection. Retrovirology. 2013;10:158.
8. Titanji K, Sammicheli S, De Milito A, et al.. Altered distribution of natural killer cell
subsets identified by CD56, CD27 and CD70 in primary and chronic human immunodeficiency virus-1 infection. Immunology. 2008;123:164–170.
9. Naranbhai V, Altfeld M, Karim SS, et al.. Changes in natural killer cell
activation and function during primary HIV-1 Infection. PLoS One. 2013;8:e53251.
10. Mavilio D, Lombardo G, Benjamin J, 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.
11. Hessell AJ, Hangartner L, Hunter M, et al.. Fc receptor but not complement binding is important in antibody protection against HIV. Nature. 2007;449:101–104.
12. Focosi D, Bestagno M, Burrone O, et al.. CD57+ T lymphocytes and functional immune deficiency. J Leukoc Biol. 2010;87:107–116.
13. Lopez-Verges S, Milush JM, Pandey S, et al.. CD57 defines a functionally distinct population of mature NK cells in the human CD56dimCD16+ NK-cell subset. Blood. 2010;116:3865–3874.
14. Chattopadhyay PK, Betts MR, Price DA, et al.. The cytolytic enzymes granyzme A, granzyme B, and perforin: expression patterns, cell distribution, and their relationship to cell maturity and bright CD57 expression. J Leukoc Biol. 2009;85:88–97.
15. Vossen MT, Matmati M, Hertoghs KM, et al.. CD27 defines phenotypically and functionally different human NK cell subsets
. J Immunol. 2008;180:3739–3745.
16. Silva A, Andrews DM, Brooks AG, et al.. Application of CD27 as a marker for distinguishing human NK cell subsets
. Int Immunol. 2008;20:625–630.
17. Fehniger TA, Cooper MA, Nuovo GJ, et al.. CD56bright natural killer cells are present in human lymph nodes and are activated by T cell-derived IL-2: a potential new link between adaptive and innate immunity. Blood. 2003;101:3052–3057.
18. De Maria A, Fogli M, Costa P, et al.. The impaired NK cell cytolytic function in viremic HIV-1 infection is associated with a reduced surface expression of natural cytotoxicity receptors (NKp46, NKp30 and NKp44). Eur J Immunol. 2003;33:2410–2418.
19. Weber K, Meyer D, Grosse V, et al.. Reconstitution of NK cell activity in HIV-1 infected individuals receiving antiretroviral therapy
. Immunobiology. 2000;202:172–178.
20. Chehimi J, Azzoni L, Farabaugh M, et al.. Baseline viral load and immune activation determine the extent of reconstitution of innate immune effectors in HIV-1-infected subjects undergoing antiretroviral treatment. J Immunol. 2007;179:2642–2650.
21. Howell DN, Andreotti PE, Dawson JR, et al.. Natural killing target antigens as inducers of interferon: studies with an immunoselected, natural killing-resistant human T lymphoblastoid cell line. J Immunol. 1985;134:971–976.
22. Pollara J, Hart L, Brewer F, et al.. High-throughput quantitative analysis of HIV-1 and SIV-specific ADCC-mediating antibody responses. Cytometry A. 2011;79:603–612.
23. Borggren MVL, Andresen BS, Grevstad B, et al., Fomsgaard a optimization of HIV-1 envelope DNA vaccine candidates within three different animal models, guinea pigs, rabbits and cynomolgus macaques. Vaccines. 2013;1:305–327.
24. Montaldo E, Del Zotto G, Della Chiesa M, et al.. Human NK cell receptors/markers: a tool to analyze NK cell development, subsets and function. Cytometry A. 2013;83:702–713.
25. Bayigga L, Nabatanzi R, Sekiziyivu PN, et al.. High CD56++CD16- natural killer (NK) cells among suboptimal immune responders after four years of suppressive antiretroviral therapy
in an African adult HIV treatment cohort. BMC Immunol. 2014;15:2.
26. Hong HS, Ahmad F, Eberhard JM, et al.. Loss of CCR7 expression on CD56(bright) NK cells is associated with a CD56(dim)CD16(+) NK cell-like phenotype and correlates with HIV viral load. PLoS One. 2012;7:e44820.
27. Parsons MS, Tang CC, Jegaskanda S, et al.. Anti-HIV antibody-dependent activation of NK cells impairs NKp46 expression. J Immunol. 2014;192:308–315.
28. Ljunggren K, Broliden PA, Morfeldt-Manson L, et al.. IgG subclass response to HIV in relation to antibody-dependent cellular cytotoxicity
at different clinical stages. Clin Exp Immunol. 1988;73:343–347.
29. Norley SG, Mikschy U, Werner A, et al.. Demonstration of cross-reactive antibodies able to elicit lysis of both HIV-1- and HIV-2-infected cells. J Immunol. 1990;145:1700–1705.
30. Lambotte O, Pollara J, Boufassa F, et al.. High antibody-dependent cellular cytotoxicity
responses are correlated with strong CD8 T cell viral suppressive activity but not with B57 status in HIV-1 elite controllers. PLoS One. 2013;8:e74855.
31. Kottilil S, Shin K, Jackson JO, et al.. Innate immune dysfunction in HIV infection: effect of HIV envelope-NK cell interactions. J Immunol. 2006;176:1107–1114.
32. Chung AW, Rollman E, Center RJ, et al.. Rapid degranulation of NK cells following activation by HIV-specific antibodies. J Immunol. 2009;182:1202–1210.