CMV infection in kidney transplant recipients (KTRs) has been associated with an increased risk for graft loss and reduced host survival. CMV promotes persistent expansions of NK cells expressing the CD94/NKG2C receptor. The NKG2C (KLRC2) gene is frequently deleted, and copy number influences the adaptive response of NKG2C+ NK cells. The distribution of NKG2C+ NK cells and NKG2C genotypes (NKG2C+/+, NKG2C+/del, NKG2Cdel/del) were studied in cross-sectional (n = 253) and prospective (n = 122) KTR cohorts. Assessment of CMV viremia was restricted to symptomatic cases in the retrospective study, but was regularly monitored in the prospective cohort. Overall, the proportions of NKG2C+ NK cells were significantly higher in KTRs who had suffered posttransplant symptomatic CMV infection in the cross-sectional study. Yet, along the prospective follow-up (3, 6, 12, and 24 mo), posttransplant NKG2C+ NK cell expansions were not observed in every patient with detectable viremia who received preemptive antiviral therapy, suggesting that the adaptive NK cell response may be inversely related with the degree of CMV control. Remarkably, the incidence of posttransplant viremia was reduced among cases with high pretransplant levels of NKG2C+ NK cells. The NKG2C genotype distribution was comparable in KTR and healthy controls, and greater proportions of NKG2C+ cells were detected in NKG2C+/+ than in NKG2C+/del patients. Yet, a trend toward increased NKG2C+/del and reduced NKG2C+/+ frequencies associated with symptomatic infection was appreciated in both cohorts. Altogether, our results indirectly support that adaptive NKG2C+ NK cells are involved in the control of CMV in KTRs.

Human CMV infection in immunocompetent individuals is generally asymptomatic and the virus establishes a life-long latent infection, undergoing reactivations that allow its successful transmission, reaching a high prevalence in all human populations (1, 2). Immunosuppression of kidney transplant recipients (KTRs) favors CMV infection or reactivation, which increases the rate of graft loss and reduced host survival, requiring antiviral therapy (35). Prophylactic antiviral drugs are administered to CMV KTRs transplanted from a CMV+ donor or KTRs receiving intensive immunosuppression. In most CMV+ KTRs, preemptive therapy upon detection of CMV viremia, regardless of clinical expression, is currently used with positive results; yet, patients who might spontaneously control the infection are exposed to antiviral drug toxicity and the overall costs are increased (4). The identification of biomarkers predicting the risk for posttransplant CMV infection is warranted to improve its clinical management. Assessing the pretransplant frequency of T cells specific for viral Ags (IE1 and pp65) has been proposed to fulfill this goal (6).

CMV infection promotes the expansion of a mature NK cell subset that displays high surface expression levels of the CD94/NKG2C activating lectin-like receptor (NKG2Cbright). Expansions of NKG2C+ cells have been reported in CMV+ healthy adults and children (79), in CMV coinfected individuals suffering from other viral infections (1012), as well as in CMV-infected immunocompromised individuals, including KTRs (13) and hematopoietic stem cell transplant recipients (14, 15). Remarkably, the magnitude of the NK cell subset redistribution appears rather variable in healthy blood donors and persists under steady-state conditions (8). This adaptive NK cell response is reminiscent of that mediated in mice by “memory” NK cells expressing the Ly49H+ NK cell receptor (NKR) specific for the m157 murine CMV glycoprotein (16).

Additional characteristics of adaptive NK cells include: 1) expression of killer Ig-like receptor (KIR) and Ig-like transcript 2 (ILT2) (LILRB1 or LIR1) and lack of NKG2A inhibitory NKRs for HLA class I; 2) acquisition of CD57, a late differentiation marker shared with cytotoxic T cells; 3) downregulation of activating NKR (i.e., NKp46, NKp30), CD161, and Siglec7 molecules; and 4) stable surface levels of NKG2D and CD16 (8). Recently, epigenetic downregulation of some transcription factors and signaling molecules, for example, FcεRI γ-chain (FcRγ), has been reported (17, 18). The variability of some of these phenotypic features (e.g., CD57 expression or FcRγ loss) in expanded NKG2C+ cells from different individuals suggests that they are gradually acquired along late differentiation (19).

NKG2Cbright NK cells are functionally mature, capable of efficiently mediating cytotoxicity and cytokine production (e.g., TNF-α and IFN-γ) triggered via CD94/NKG2C (8). Moreover, Ab-dependent stimulation via CD16 (FcγRIIIa) activates adaptive NKG2C+ NK cells against CMV-infected cells (2022). Yet, the molecular and cellular mechanisms underlying the generation of adaptive NKG2Cbright NK cells in response to CMV and their role in the control of the infection remain unclear (8, 16, 23, 24). A deletion of the NKG2C gene (officially termed KLRC2) determines the existence of three genotypes (i.e., NKG2C+/+, NKG2C+/del, and NKG2Cdel/del) in populations of different ethnic origin (25, 26). NKG2C copy number has been positively correlated with expression levels and function of the CD94/NKG2C receptor, as well as with steady-state numbers of circulating adaptive NKG2C+ NK cells in healthy individuals (27, 28).

Expansions of NKG2Cbright NK cells were reported in KTRs with CMV viremia (13), but there is no information on their implication in antiviral defense. To address this issue, we analyzed the distribution of adaptive NK cells and the NKG2C genotype in relation with CMV infection in cross-sectional and prospective KTR cohorts. Our results support an involvement of adaptive NKG2C+ NK cells in the control of CMV infection in renal transplantation.

In a retrospective study, 253 patients transplanted between August 2006 and February 2013, under immunosuppression with tacrolimus, mycophenolate, and steroids, with a graft functioning >3 mo were enrolled. NKG2C genotyping was carried out in 238 cases and peripheral blood immunophenotyping in 191 patients at a median posttransplant time of 21 mo (interquartile range [IQR], 19–25).

In a second prospective study, 122 patients transplanted from March 2013 and receiving the same immunosuppressive treatment with tacrolimus, mycophenolate, and steroids were enrolled. All of them were genotyped for NKG2C, and NK cell phenotypic analysis was performed before and at different time points (3, 6, 12, and 24 mo) after transplantation. Both studies were approved by the Hospital Research Ethical Committee (CEIC-Parc de Salut Mar number 2010/3904/I and 137/C/2012), and patients signed an informed consent.

All patients were transplanted across a negative complement-dependent cytotoxicity cross-match with donor lymphocytes. HLA typing and complement-dependent cytotoxicity cross-match, as well as periodical detection of anti-HLA Abs, were carried out as described previously (29). A database including information on demographics, type of donor, number of transplant, initial and follow-up immunosuppression, delayed graft function and acute rejection episodes, donor and recipient CMV serology, and CMV viremia was generated.

Viremia was defined as the detection of CMV DNA in blood samples by a standardized commercial quantitative PCR (qPCR) assay (COBAS AmpliPrep, Cobas TaqMan; Roche Diagnostics) and expressed as copies per milliliter. In the retrospective study, for a subset of KTRs (n = 49) transplanted between 2006 and 2007, viremia was indirectly assessed by detection of pp65 antigenemia (Monofluo Kit CMV; Bio-Rad). Symptomatic CMV infection was defined as evidence of CMV replication/antigenemia together with clinical expression, including viral syndrome and invasive tissue disease (30).

In the retrospective study, posttransplant detection of CMV viremia was limited to patients who experienced development of symptoms and/or analytical changes compatible with infection. In the prospective study, prophylactic valganciclovir was administered for 90 d to CMV recipients of CMV+ donors, and to CMV+ recipients receiving thymoglobulin induction or any antirejection treatment. Posttransplant CMV viremia was regularly monitored every 2 wk, for 3 mo in recipients without prophylaxis and for 2 mo after prophylaxis ended in the remaining cases. All patients with compatible symptoms and any degree of replication received specific treatment. For asymptomatic cases, antiviral therapy was administered in seronegative KTRs with >500 copies per milliliter and in seropositive patients with >1000 copies per milliliter.

Phenotypic analysis was performed by flow cytometry in fresh peripheral blood samples, obtained by venous puncture in EDTA tubes as described previously (29). Samples were pretreated with saturating concentrations of human aggregated Igs to block FcγR and then labeled with different Ab combinations to define total T and NK cell subsets. For direct immunofluorescence staining, we used the following Abs: allophycocyanin-CD3 (clone SK7), PerCP-CD45 (clone 2D1), and FITC-CD56 (clone NCAM16.2) from BD Biosciences (San Diego, CA). NK cells were defined as CD3CD56+ lymphocytes, and NK cell subsets were identified by indirect immunofluorescence staining with Abs specific for NKG2C (clone MAB1381; R&D Systems, Minneapolis, MN), NKG2A (clone Z199; provided by Dr. A. Moretta), CD161 (clone HP-3G10; generated in our laboratory), and KIR; in that case, staining was performed using a mixture of Abs specific for KIR3DL1/L2 and 2DS4 (clone 5.133; provided by Dr. M. Colonna), KIR2DL2/S2/L3 (clone CH-L; provided by Dr. S. Ferrini), KIR3DL1 (clone DX9; provided by Dr. L.L. Lanier), and KIR2DL1/S1/S4 (clone HP-3E4; generated in our laboratory). PE-conjugated F(ab′)2 rabbit anti-mouse Ig (Dako, Glostrup, Denmark) was used as secondary Ab. After washing and erythrocyte lysis, samples were acquired with a FACSCanto cytometer, and data were analyzed with the DIVA software (BD Biosciences). Absolute numbers of cells were calculated from blood counts obtained in parallel. DNA was isolated from total blood using the Puregene, BloodCore kit B (Qiagen). NKG2C zygosity was assessed as previously described (26).

Categorical variables are expressed as percentages. Continuous variables are expressed as mean and SD or median (IQR), according to normal or nonnormal distributions. Normal distribution was assessed by normal Q-Q plots. Univariate analysis was performed by Student t, Mann–Whitney U, Friedman, χ2, or Fisher’s tests, as appropriate.

The Cox proportional hazards model was used to test the association between the %NKG2C+ NK cells and CMV infection, adjusted for confounders. Linearity assumption of the model was tested: none of the nonlinear parameters achieved statistical significance when continuous variables were included in the model using a smoothing spline. The proportional hazard assumption for all variables included in the model was also tested.

To find the best cutoff for the %NKG2C+ NK cells that discriminated KTRs at higher and lower risk for development of CMV infection, a bootstrapping procedure (31) was used performing 1000 replicates. In each replicate, a sample with replacement of size equal to the study sample was selected. In each replicate, several cutoffs, ranging from 1 to 80, were tested. The one that achieved maximum likelihood of the model was retained. The median of those 1000 cutoff values was considered the best, and the 2.5 and 97.5 percentiles as the limits of the 95% confidence interval (CI).

Analyses were performed using IBM SPSS Statistics for Windows, Version 21.0 (IBM, Armonk, NY) and R version 3.0.1 (R: a language and environment for statistical computing; R Foundation for Statistical Computing, Vienna, Austria).

As a first approach to assess the adaptive NK cell response to CMV in KTR, a retrospective study was carried out in a cohort of 253 cases immunosuppressed at transplantation who did not receive CMV prophylaxis (Supplemental Table I). Viremia was confirmed in 47 KTRs (18.6%) with clinically recorded symptomatic infection, who therefore received antiviral therapy. KTR groups stratified according to CMV infection were similar in donor and recipient CMV serology, delayed graft function, or acute rejection; unadjusted death-censored graft survival was lower in patients with CMV infection than in those without infection (p = 0.026).

Information on posttransplant distribution of the NKG2C+ NK cell subset was obtained in 191 cases, complemented by the analysis of additional NKRs (i.e., NKG2A, CD161, and KIR). Significantly higher proportions of NKG2C+ and concomitantly lower proportions of NKG2A+ NK cells were detected in KTRs who had developed posttransplant symptomatic CMV infection (n = 30), as compared with the remaining cases (n = 161) in whom the putative incidence of subclinical infection was undefined (Table I). No significant differences were observed in the distribution of CD161+ and KIR+ NK subsets (data not shown). In a Cox regression analysis, thymoglobulin induction (hazard ratio [HR], 2.65; 95% CI, 1.13–6.21; p = 0.025) and the percentage of NKG2C+ NK cells (HR, 1.021; 95% CI, 1.005–1.037; p = 0.010) were found to be independently and directly associated with symptomatic CMV infection; the association with age detected in the univariate analysis was not significant (data not shown). This cross-sectional study further supported the relationship between CMV and the adaptive expansion of NKG2C+ NK cells in immunosuppressed KTR, which was detectable several years after transplant (median follow-up at analysis: 54 mo; Supplemental Table I).

The retrospective study did not provide information on the incidence of subclinical CMV replication nor on pretransplant levels of NKG2C+ NK cells. To overcome these limitations, we conducted a prospective study assessing the course of the adaptive NK cell response in a second KTR cohort (N = 122) treated similarly with tacrolimus, mycophenolate, and steroids (Table II). A total of 109 cases (89%) were CMV+ at transplantation, and 21 received prophylactic valganciclovir. Posttransplant CMV viremia was regularly monitored by PCR, being detected at a median of 58 d (IQR, 41–112) in 39 cases (32%), who subsequently received antiviral therapy.

The NK cell immunophenotype was studied pretransplant and posttransplant at different time points. In agreement with previous reports in healthy donors, baseline NKR distribution revealed increased proportions of NKG2C+ cells and lower percentages of NKG2A+ cells in CMV+ compared with CMV KTRs, without significant differences in other NK cell subsets (Table III). Remarkably, pretransplant NKG2C+ NK cells appeared significantly higher in cases without posttransplant viremia (Table IV). This observation was confirmed in a reanalysis excluding D/R cases and KTRs with thymoglobulin induction and/or valganciclovir prophylaxis, comprising D+/R cases, thus ruling out the influence of these putative confounding variables (Table IV). Cox analysis (Table V) further supported the inverse association of pretransplant %NKG2C+ cells with CMV viremia, independently of confounding variables.

A cutoff to define pretransplant proportions of NKG2C+ NK cells that discriminated better KTRs at higher and lower risk for development of CMV viremia was calculated (24%), as described in 2Materials and Methods. The incidence of posttransplant CMV infection was significantly lower in the group with greater proportions of NKG2C+ cells (Fig. 1); actually, viremia was detected in 6 of 37 cases, including 1 receiving intensive immunosuppression, in contrast with 33 positive PCR events detected in 85 KTRs with <24% pretransplant NKG2C+ NK cells. Sensitivity analysis restricted to CMV-seropositive KTRs not receiving valganciclovir prophylaxis (n = 97) was carried out. Results were consistent with the primary analysis, and Kaplan–Meier estimates were fairly similar to those observed in the whole cohort (p = 0.011, data not shown). These observations pointed out a remarkable association of high baseline levels of NKG2C+ NK cells with a reduced risk for posttransplant CMV infection.

Overall, NKG2C+ NK cells remained rather stable along the follow-up in KTRs without detectable CMV viremia, including cases with low or high baseline proportions of NKG2C+ cells. By contrast, a progressive increase in the proportions of NKG2C+ cells was perceived in KTRs with CMV viremia, reaching statistical significance at 24 mo (Table VI). An analysis of individual evolution patterns revealed that only some KTRs with a positive PCR for CMV, who subsequently received preemptive antiviral therapy, developed posttransplant expansions of NKG2C+ NK cells during the follow-up; representative examples are displayed in Fig. 2.

NKG2C gene copy number variation influences expression levels and function of the receptor, as well as steady-state numbers of NKG2C+ NK cells, which tend to be higher in NKG2C+/+ than in NKG2C+/del healthy subjects. As shown in Table VII, the NKG2C genotype distributions appeared in both KTR cohorts comparable with those found in healthy donors. Moreover, the relation of NKG2C copy number with the magnitude of steady-state NKG2C+ NK cell expansions reported in CMV+ healthy donors (27, 28) was confirmed pretransplant and posttransplant in the prospective and retrospective KTR cohorts, respectively. In both groups, significantly higher proportions of NKG2C+ NK cells were detected in NKG2C+/+ than in NKG2C+/del cases (Fig. 3).

Considering the relation of the NKG2C genotype with the expansion of NKG2C+ adaptive NK cells, and their apparent negative association with the risk for posttransplant CMV infection, we addressed whether this event might be also under the influence of NKG2C copy number. In the retrospective cohort, genotype distributions in KTR with symptomatic CMV infection were compared with the remaining cases. In the prospective study, the genotype distribution was compared according to the incidence of viremia, as well as analyzing separately symptomatic cases categorized with the same criteria applied for the cross-sectional study. In both cohorts, higher and lower frequencies of NKG2C+/del and NKG2C+/+ genotypes were respectively detected in CMV+ KTR cases with symptomatic posttransplant infection (Table VII). Even though the differences did not reach statistical significance, the coincident trend indirectly suggested a relation of the NKG2C genotype with the clinical impact of posttransplant infection. Paradoxically, the homozygous NKG2C deletion appeared underrepresented among cases with symptomatic infection, particularly in the retrospective cohort; yet, the low frequency of this genotype rendered the observation difficult to interpret. Inclusion of the NKG2C genotype to the multivariate Cox model for CMV risk (Table V) did not alter the results (NKG2C+/+ genotype versus others: HR, 0.88; 95% CI, 0.44–1.75; p = 0.719), and the significance for basal NKG2C+ NK cells (HR, 0.96, 95% CI 0.94–0.98; p = 0.006) was maintained.

In this report we explored the relation of adaptive NKG2C+ NK cells with the response of KTRs to CMV, based on the complementary study of two cohorts. In a first cross-sectional analysis, patients who had experienced symptomatic posttransplant CMV infection displayed after several years higher proportions of NKG2C+ NK cells, in line with a previous report relating the expansion of adaptive NK cells with CMV infection in KTRs (8, 13). The interpretation of these results was ambiguous, because cases with pretransplant and posttransplant expansions of NKG2C+ cells were not discriminated and asymptomatic infections were not identified. In a second prospective study, KTRs were regularly monitored for CMV viremia, assessing their NK cell immunophenotype pretransplant and posttransplant. Along the follow-up, postinfection adaptive NK cell expansions were detected only in some cases with viremia and, exceptionally, in KTRs with negative CMV-specific PCR during the follow-up, supporting their development in response to untreated subclinical CMV replication. These observations indicated that detection of CMV viremia in KTRs followed by preemptive antiviral therapy was not indefectibly associated with an expansion of NKG2C+ cells, and suggested that a prompt CMV control may hamper or delay the adaptive NK cell response in KTRs. No significant correlation was found between the peak of viral load and posttransplant levels of NKG2C+ NK cells (data not shown). It is plausible that persistent/recurrent low levels of viral replication, after withdrawal of antiviral therapy, may account for the delayed progressive NKG2C+ NK cell expansions observed in some patients, revealing an inefficient control of the pathogen.

Remarkably, the incidence of posttransplant CMV viremia in the prospective study was lower in CMV+ KTRs displaying high pretransplant levels of circulating NKG2C+ NK cells, suggesting that they might play a protective role. This effect was confirmed to be independent of thymoglobulin induction and/or antiviral prophylaxis. Age is a known variable related with CMV infection in kidney transplantation, as observed in both cohorts (Table II, Supplemental Table I). Yet, Cox analysis indicated that the relation between the levels of % NKG2C+ cells and the risk for posttransplant CMV viremia was independent of age. Actually, there is no evidence supporting a relation between aging and the development of adaptive NKG2C+ cells in CMV+ individuals (32). Despite the limited in vitro direct response of adaptive NK cells against CMV-infected cells, presumably under the influence of viral immune evasion mechanisms, they have been shown to efficiently mediate virus-specific Ab-dependent cytotoxicity and cytokine production triggered through the FcγRIIIa (CD16) receptor (2022). It is plausible that pretransplant expansions of NKG2C+ cells, in combination with specific IgG, may contribute to posttransplant CMV control in seropositive KTRs, partially compensating for T cell immunosuppression. In this regard, CD16 downregulation and expression of activation markers in NK cells have been reported in KTRs (33). In some cases, this phenotypic profile might reflect an IgG-dependent NK cell activation triggered by CMV or donor-specific alloantibodies (29). On the other hand, CMV-induced adaptive NKG2C+ NK cells may be eventually boosted through CD16 along the Ab-dependent response to other pathogens (18). As indicated by sequential monitoring, regular immunosuppressive therapy (i.e., tacrolimus and mycophenolate) did not affect posttransplant levels of adaptive NK cells in PCR-negative KTRs. Yet, CMV viremia was observed in a case with high basal levels of NKG2C+ NK cells receiving intensive antirejection therapy, which included long-lasting treatment for lupus before transplantation and thymoglobulin induction. Further studies are warranted to assess the impact of different immunosuppressive regimens on the basal levels, development, and effector functions of adaptive NK cells.

Our study was focused on the analysis of NKG2C+ cell expansion as a conventional hallmark of the adaptive NK cell response to CMV. Detection of other phenotypic features in adaptive NKG2C+ NK cells (e.g., CD57 expression and FcRγ loss), reflecting their late differentiation, is variable (9, 18, 19, 27). Assessing the relation of these markers with the putative antiviral effect of adaptive NK cells deserves attention in larger KTR cohorts. In contrast, the association of high pretransplant levels of adaptive NKG2C+ NK cells with a lower risk for CMV infection might indirectly reflect the involvement of other immune effector cells (34). In particular, the levels of pretransplant CMV-specific TcRαβ+ T cells have been reported to predict the risk for posttransplant infection (6, 35), and increased numbers of TcRγδ+ T cells were associated with CMV control (36). Recent reports support that adaptive NK cells and Ag-specific T cells may develop independently in response to CMV (37, 38), and further studies are warranted to evaluate their relative contribution to the control of the viral infection in KTRs.

Previous studies revealed an influence of NKG2C gene copy number on surface expression levels and function of the activating CD94/NKG2C receptor, as well as on steady-state numbers of circulating NKG2C+ NK cells, which were reduced in NKG2C+/del as compared with NKG2C+/+ healthy individuals (27, 28); moreover, the lack of NKG2C was suggested to be associated with altered control of CMV in childhood (27, 28). The NKG2C genotype relation with the magnitude of the adaptive NK cell response was confirmed in both KTR cohorts, which displayed an overall distribution of the NKG2C deletion comparable with that of the control population. By contrast, an increased frequency of the NKG2C+/del genotype and a reciprocal reduction of NKG2C+/+ cases were detected among KTRs suffering symptomatic CMV infection. Although these differences did not reach statistical significance, the coincident trend in both cohorts suggested that NKG2C copy number might subtly influence the outcome of CMV infection in KTRs, modulating the magnitude and function of the adaptive NKG2C+ NK cell response (8).

Should NKG2C+ cells play a relevant role in the posttransplant response to CMV, an increased frequency of the NKG2Cdel/del genotype associated with infection would be predictable. Yet, an opposite reduction of the homozygous deletion was unexpectedly observed among this KTR group, particularly in the retrospective study. The relatively low frequency of the NKG2Cdel/del genotype in the general population (25, 26, 28) together with the limited number of cases with CMV symptomatic infection limits the conclusiveness of this analysis, which should be validated in a larger study with a longer follow-up. Yet, this intriguing observation might reflect a reduced graft survival in NKG2Cdel/del KTRs with symptomatic CMV replication, consistent with a potential role of adaptive NKG2C+ NK cells in antiviral defense. Alternatively, the homozygous deletion might be associated with a lower incidence of CMV infection. In this regard, expansions of NKG2C NK cells expressing activating KIRs were identified in NKG2Cdel/del hematopoietic stem cell transplant recipients suffering CMV infection (39), supporting the existence of alternative adaptive NK cell subsets in individuals with KIR-B haplotypes (i.e., which encode more activating KIR), a genetic feature previously associated with a lower incidence of CMV infection in KTR (40, 41). Nevertheless, no relation between expansions of NKG2C+ cells and KIR-A and KIR-B haplotypes was previously observed (7); moreover, NKG2C NK cell populations displaying adaptive phenotypic features independently of activating KIR expression have been recently reported (19, 42).

In summary, to our knowledge, this study provides the first evidence indirectly supporting a role of adaptive NKG2C+ NK cells in defense against CMV in KTRs. The association of high pretransplant levels of NKG2C+ NK cells with a reduced incidence of posttransplant CMV viremia suggests that they may confer some protection against viral reactivation or reinfection. From a practical standpoint, high baseline levels of NKG2C+ cells might predict a lower risk for posttransplant CMV replication and disease in KTRs receiving regular immunosuppression, particularly in NKG2C+/+ CMV+ patients. This information might contribute to stratifying KTRs for cost-effective use of CMV prophylaxis, which is not free of adverse events (43, 44). In contrast, posttransplant expansions of adaptive NKG2C+ NK cells are reactive to the incidence of CMV infection and, once established, may contribute to controlling viral replication. In that case, the individual evolution pattern of the adaptive NK cell response might become helpful to monitor CMV control. Yet, the observational design and relatively reduced number of cases are limitations of this report. Larger prospective studies comparatively analyzing pretransplant adaptive NK cells and CMV-specific T lymphocytes are warranted, because both might become complementary useful biomarkers in the clinical management of the viral infection in KTRs.

We thank Dulce Soto, Judit Rigol, Andrea Vera, Sara Alvarez, Anna Faura, and Rosa Causadias for technical help.

This study was supported by a coordinated research project from Fundació La Marató de TV3 (137/C/2012) and by grants from Plan Estatal I+D Retos (Grant SAF2013-49063-C2-1-R), the Spanish Ministry of Economy and Competitiveness (FEDER), the EC FP7–Spanish Ministry of Economy and Competitiveness Infect–ERA program (Grant PCIN-2015-191-C02-01), FIS–FEDER Grant PI13/00598, Intensification Programs (Spanish Ministry of Health ISCIII), and RedinRen Grant RD12/0021/0024. A.M. is supported by the Asociación Española Contra el Cáncer. D.R.-P. performed this study as part of her doctoral thesis at the University Autónoma Barcelona.

The online version of this article contains supplemental material.

Abbreviations used in this article:

CI

confidence interval

FcRγ

FcεRI γ-chain

HR

hazard ratio

IQR

interquartile range

KTR

kidney transplant recipient

NKR

NK cell receptor

qPCR

quantitative PCR.

1
Mocarski
E. S.
,
Shenk
T.
,
Griffiths
P. D.
,
Pass
R. F.
.
2013
.
Cytomegaloviruses
. In
Fields Virology
, 6th Ed.
Knipe
D. M.
,
Howley
P. M.
, eds.
Lippincott Williams & Wilkins
,
Philadelphia
, p.
1960
2015
.
2
Cannon
M. J.
,
Schmid
D. S.
,
Hyde
T. B.
.
2010
.
Review of cytomegalovirus seroprevalence and demographic characteristics associated with infection.
Rev. Med. Virol.
20
:
202
213
.
3
Sagedal
S.
,
Hartmann
A.
,
Nordal
K. P.
,
Osnes
K.
,
Leivestad
T.
,
Foss
A.
,
Degré
M.
,
Fauchald
P.
,
Rollag
H.
.
2004
.
Impact of early cytomegalovirus infection and disease on long-term recipient and kidney graft survival.
Kidney Int.
66
:
329
337
.
4
Razonable
R. R.
,
Hayden
R. T.
.
2013
.
Clinical utility of viral load in management of cytomegalovirus infection after solid organ transplantation.
Clin. Microbiol. Rev.
26
:
703
727
.
5
Tong
C. Y.
,
Bakran
A.
,
Peiris
J. S.
,
Muir
P.
,
Herrington
C. S.
.
2002
.
The association of viral infection and chronic allograft nephropathy with graft dysfunction after renal transplantation.
Transplantation
74
:
576
578
.
6
Bestard
O.
,
Lucia
M.
,
Crespo
E.
,
Van Liempt
B.
,
Palacio
D.
,
Melilli
E.
,
Torras
J.
,
Llaudó
I.
,
Cerezo
G.
,
Taco
O.
, et al
.
2013
.
Pretransplant immediately early-1-specific T cell responses provide protection for CMV infection after kidney transplantation.
Am. J. Transplant.
13
:
1793
1805
.
7
Gumá
M.
,
Angulo
A.
,
Vilches
C.
,
Gómez-Lozano
N.
,
Malats
N.
,
López-Botet
M.
.
2004
.
Imprint of human cytomegalovirus infection on the NK cell receptor repertoire.
Blood
104
:
3664
3671
.
8
López-Botet
M.
,
Muntasell
A.
,
Vilches
C.
.
2014
.
The CD94/NKG2C+ NK-cell subset on the edge of innate and adaptive immunity to human cytomegalovirus infection.
Semin. Immunol.
26
:
145
151
.
9
Béziat
V.
,
Liu
L. L.
,
Malmberg
J. A.
,
Ivarsson
M. A.
,
Sohlberg
E.
,
Björklund
A. T.
,
Retière
C.
,
Sverremark-Ekström
E.
,
Traherne
J.
,
Ljungman
P.
, et al
.
2013
.
NK cell responses to cytomegalovirus infection lead to stable imprints in the human KIR repertoire and involve activating KIRs.
Blood
121
:
2678
2688
.
10
Björkström
N. K.
,
Lindgren
T.
,
Stoltz
M.
,
Fauriat
C.
,
Braun
M.
,
Evander
M.
,
Michaëlsson
J.
,
Malmberg
K. J.
,
Klingström
J.
,
Ahlm
C.
,
Ljunggren
H. G.
.
2011
.
Rapid expansion and long-term persistence of elevated NK cell numbers in humans infected with hantavirus.
J. Exp. Med.
208
:
13
21
.
11
Béziat
V.
,
Dalgard
O.
,
Asselah
T.
,
Halfon
P.
,
Bedossa
P.
,
Boudifa
A.
,
Hervier
B.
,
Theodorou
I.
,
Martinot
M.
,
Debré
P.
, et al
.
2012
.
CMV drives clonal expansion of NKG2C+ NK cells expressing self-specific KIRs in chronic hepatitis patients.
Eur. J. Immunol.
42
:
447
457
.
12
Gumá
M.
,
Cabrera
C.
,
Erkizia
I.
,
Bofill
M.
,
Clotet
B.
,
Ruiz
L.
,
López-Botet
M.
.
2006
.
Human cytomegalovirus infection is associated with increased proportions of NK cells that express the CD94/NKG2C receptor in aviremic HIV-1-positive patients.
J. Infect. Dis.
194
:
38
41
.
13
Lopez-Vergès
S.
,
Milush
J. M.
,
Schwartz
B. S.
,
Pando
M. J.
,
Jarjoura
J.
,
York
V. A.
,
Houchins
J. P.
,
Miller
S.
,
Kang
S. M.
,
Norris
P. J.
, et al
.
2011
.
Expansion of a unique CD57⁺NKG2Chi natural killer cell subset during acute human cytomegalovirus infection.
Proc. Natl. Acad. Sci. USA
108
:
14725
14732
.
14
Della Chiesa
M.
,
Falco
M.
,
Podestà
M.
,
Locatelli
F.
,
Moretta
L.
,
Frassoni
F.
,
Moretta
A.
.
2012
.
Phenotypic and functional heterogeneity of human NK cells developing after umbilical cord blood transplantation: a role for human cytomegalovirus?
Blood
119
:
399
410
.
15
Foley
B.
,
Cooley
S.
,
Verneris
M. R.
,
Pitt
M.
,
Curtsinger
J.
,
Luo
X.
,
Lopez-Vergès
S.
,
Lanier
L. L.
,
Weisdorf
D.
,
Miller
J. S.
.
2012
.
Cytomegalovirus reactivation after allogeneic transplantation promotes a lasting increase in educated NKG2C+ natural killer cells with potent function.
Blood
119
:
2665
2674
.
16
O’Sullivan
T. E.
,
Sun
J. C.
,
Lanier
L. L.
.
2015
.
Natural killer cell memory.
Immunity
43
:
634
645
.
17
Schlums
H.
,
Cichocki
F.
,
Tesi
B.
,
Theorell
J.
,
Béziat
V.
,
Holmes
T. D.
,
Han
H.
,
Chiang
S. C.
,
Foley
B.
,
Mattsson
K.
, et al
.
2015
.
Cytomegalovirus infection drives adaptive epigenetic diversification of NK cells with altered signaling and effector function.
Immunity
42
:
443
456
.
18
Lee
J.
,
Zhang
T.
,
Hwang
I.
,
Kim
A.
,
Nitschke
L.
,
Kim
M.
,
Scott
J. M.
,
Kamimura
Y.
,
Lanier
L. L.
,
Kim
S.
.
2015
.
Epigenetic modification and antibody-dependent expansion of memory-like NK cells in human cytomegalovirus-infected individuals.
Immunity
42
:
431
442
.
19
Muntasell
A.
,
Pupuleku
A.
,
Cisneros
E.
,
Vera
A.
,
Moraru
M.
,
Vilches
C.
,
López-Botet
M.
.
2016
.
Relationship of NKG2C copy number with the distribution of distinct cytomegalovirus-induced adaptive NK cell subsets.
J. Immunol.
196
:
3818
3827
.
20
Costa-García
M.
,
Vera
A.
,
Moraru
M.
,
Vilches
C.
,
López-Botet
M.
,
Muntasell
A.
.
2015
.
Antibody-mediated response of NKG2Cbright NK cells against human cytomegalovirus.
J. Immunol.
194
:
2715
2724
.
21
Zhang
T.
,
Scott
J. M.
,
Hwang
I.
,
Kim
S.
.
2013
.
Cutting edge: antibody-dependent memory-like NK cells distinguished by FcRγ deficiency.
J. Immunol.
190
:
1402
1406
.
22
Wu
Z.
,
Sinzger
C.
,
Frascaroli
G.
,
Reichel
J.
,
Bayer
C.
,
Wang
L.
,
Schirmbeck
R.
,
Mertens
T.
.
2013
.
Human cytomegalovirus-induced NKG2C(hi) CD57(hi) natural killer cells are effectors dependent on humoral antiviral immunity.
J. Virol.
87
:
7717
7725
.
23
Gumá
M.
,
Budt
M.
,
Sáez
A.
,
Brckalo
T.
,
Hengel
H.
,
Angulo
A.
,
López-Botet
M.
.
2006
.
Expansion of CD94/NKG2C+ NK cells in response to human cytomegalovirus-infected fibroblasts.
Blood
107
:
3624
3631
.
24
Rölle
A.
,
Pollmann
J.
,
Ewen
E. M.
,
Le
V. T.
,
Halenius
A.
,
Hengel
H.
,
Cerwenka
A.
.
2014
.
IL-12-producing monocytes and HLA-E control HCMV-driven NKG2C+ NK cell expansion.
J. Clin. Invest.
124
:
5305
5316
.
25
Miyashita
R.
,
Tsuchiya
N.
,
Hikami
K.
,
Kuroki
K.
,
Fukazawa
T.
,
Bijl
M.
,
Kallenberg
C. G.
,
Hashimoto
H.
,
Yabe
T.
,
Tokunaga
K.
.
2004
.
Molecular genetic analyses of human NKG2C (KLRC2) gene deletion.
Int. Immunol.
16
:
163
168
.
26
Moraru
M.
,
Cañizares
M.
,
Muntasell
A.
,
de Pablo
R.
,
López-Botet
M.
,
Vilches
C.
.
2012
.
Assessment of copy-number variation in the NKG2C receptor gene in a single-tube and characterization of a reference cell panel, using standard polymerase chain reaction.
Tissue Antigens
80
:
184
187
.
27
Muntasell
A.
,
López-Montañés
M.
,
Vera
A.
,
Heredia
G.
,
Romo
N.
,
Peñafiel
J.
,
Moraru
M.
,
Vila
J.
,
Vilches
C.
,
López-Botet
M.
.
2013
.
NKG2C zygosity influences CD94/NKG2C receptor function and the NK-cell compartment redistribution in response to human cytomegalovirus.
Eur. J. Immunol.
43
:
3268
3278
.
28
Goodier
M. R.
,
White
M. J.
,
Darboe
A.
,
Nielsen
C. M.
,
Goncalves
A.
,
Bottomley
C.
,
Moore
S. E.
,
Riley
E. M.
.
2014
.
Rapid NK cell differentiation in a population with near-universal human cytomegalovirus infection is attenuated by NKG2C deletions.
Blood
124
:
2213
2222
.
29
Crespo
M.
,
Yélamos
J.
,
Redondo
D.
,
Muntasell
A.
,
Perez-Saéz
M. J.
,
López-Montañés
M.
,
García
C.
,
Torio
A.
,
Mir
M.
,
Hernández
J. J.
, et al
.
2015
.
Circulating NK-cell subsets in renal allograft recipients with anti-HLA donor-specific antibodies.
Am. J. Transplant.
15
:
806
814
.
30
Humar
A.
,
Michaels
M.
AST ID Working Group on Infectious Disease Monitoring
.
2006
.
American Society of Transplantation recommendations for screening, monitoring and reporting of infectious complications in immunosuppression trials in recipients of organ transplantation.
Am. J. Transplant.
6
:
262
274
.
31
Efron
B.
,
Gong
G.
.
1983
.
A leisurely look at the bootstrap, the jackknife, and cross-validation.
Am. Stat.
37
:
36
49
.
32
López-Botet
M.
,
Muntasell
A.
,
Martínez-Rodríguez
J. E.
,
López-Montañés
M.
,
Costa-García
M.
,
Pupuleku
A.
.
2016
.
Development of the adaptive NK cell response to human cytomegalovirus in the context of aging.
Mech. Ageing Dev.
158
:
23
26
.
33
Hoffmann
U.
,
Neudörfl
C.
,
Daemen
K.
,
Keil
J.
,
Stevanovic-Meyer
M.
,
Lehner
F.
,
Haller
H.
,
Blume
C.
,
Falk
C. S.
.
2015
.
NK cells of kidney transplant recipients display an activated phenotype that is influenced by immunosuppression and pathological staging.
PLoS One
10
:
e0132484
.
34
Vieira Braga
F. A.
,
Hertoghs
K. M.
,
van Lier
R. A.
,
van Gisbergen
K. P.
.
2015
.
Molecular characterization of HCMV-specific immune responses: parallels between CD8(+) T cells, CD4(+) T cells, and NK cells.
Eur. J. Immunol.
45
:
2433
2445
.
35
Lúcia
M.
,
Crespo
E.
,
Cruzado
J. M.
,
Grinyó
J. M.
,
Bestard
O.
.
2014
.
Human CMV-specific T-cell responses in kidney transplantation; toward changing current risk-stratification paradigm.
Transpl. Int.
27
:
643
656
.
36
Kaminski
H.
,
Garrigue
I.
,
Couzi
L.
,
Taton
B.
,
Bachelet
T.
,
Moreau
J. F.
,
Déchanet-Merville
J.
,
Thiébaut
R.
,
Merville
P.
.
2016
.
Surveillance of γδ T cells predicts cytomegalovirus infection resolution in kidney transplants.
J. Am. Soc. Nephrol.
27
:
637
645
.
37
Bigley
A. B.
,
Spielmann
G.
,
Agha
N.
,
O’Connor
D. P.
,
Simpson
R. J.
.
2016
.
Dichotomous effects of latent CMV infection on the phenotype and functional properties of CD8+ T-cells and NK-cells.
Cell. Immunol.
300
:
26
32
.
38
Bengnér
M.
,
Béziat
V.
,
Ernerudh
J.
,
Nilsson
B. O.
,
Löfgren
S.
,
Wikby
A.
,
Malmberg
K. J.
,
Strindhall
J.
.
2014
.
Independent skewing of the T cell and NK cell compartments associated with cytomegalovirus infection suggests division of labor between innate and adaptive immunity.
Age (Dordr.)
36
:
571
582
.
39
Della Chiesa
M.
,
Falco
M.
,
Bertaina
A.
,
Muccio
L.
,
Alicata
C.
,
Frassoni
F.
,
Locatelli
F.
,
Moretta
L.
,
Moretta
A.
.
2014
.
Human cytomegalovirus infection promotes rapid maturation of NK cells expressing activating killer Ig-like receptor in patients transplanted with NKG2C-/- umbilical cord blood.
J. Immunol.
192
:
1471
1479
.
40
Hadaya
K.
,
de Rham
C.
,
Bandelier
C.
,
Bandelier
C.
,
Ferrari-Lacraz
S.
,
Jendly
S.
,
Berney
T.
,
Buhler
L.
,
Kaiser
L.
,
Seebach
J. D.
, et al
.
2008
.
Natural killer cell receptor repertoire and their ligands, and the risk of CMV infection after kidney transplantation.
Am. J. Transplant.
8
:
2674
2683
.
41
Stern
M.
,
Hadaya
K.
,
Hönger
G.
,
Martin
P. Y.
,
Steiger
J.
,
Hess
C.
,
Villard
J.
.
2011
.
Telomeric rather than centromeric activating KIR genes protect from cytomegalovirus infection after kidney transplantation.
Am. J. Transplant.
11
:
1302
1307
.
42
Liu
L. L.
,
Landskron
J.
,
Ask
E. H.
,
Enqvist
M.
,
Sohlberg
E.
,
Traherne
J. A.
,
Hammer
Q.
,
Goodridge
J. P.
,
Larsson
S.
,
Jayaraman
J.
, et al
.
2016
.
Critical role of CD2 co-stimulation in adaptive Natural Killer cell responses revealed in NKG2C-deficient humans.
Cell Reports
15
:
1088
1099
.
43
Owers
D. S.
,
Webster
A. C.
,
Strippoli
G. F.
,
Kable
K.
,
Hodson
E. M.
.
2013
.
Pre-emptive treatment for cytomegalovirus viraemia to prevent cytomegalovirus disease in solid organ transplant recipients.
Cochrane Database Syst. Rev.
2
:
CD005133
10.1002/14651858
.
44
Blumberg
E. A.
,
Hauser
I. A.
,
Stanisic
S.
,
Mueller
E.
,
Berenson
K.
,
Gahlemann
C. G.
,
Humar
A.
,
Jardine
A. G.
.
2010
.
Prolonged prophylaxis with valganciclovir is cost effective in reducing posttransplant cytomegalovirus disease within the United States.
Transplantation
90
:
1420
1426
.

The authors have no financial conflicts of interest.

Supplementary data