Abstract 356

In healthy individuals human cytomegalovirus (hCMV) infection is asymptomatic or a mild illness. In immunosuppressed patients after hematopoietic cell transplantation, however, CMV is potentially life-threatening complication. We have previously shown that upon mouse CMV infection, a population of Ly49H+ NK cells expand and are responsible for disease clearance through the induction of a “memory NK cell response”. Whether similar events occur in hCMV infection is unknown. Some studies suggest that the C-type lectin-like receptor NKG2C and the killer cell immunoglobulin-like receptor (KIR) family may be involved. We studied NK cell function after hCMV reactivation. Cohort 1 consisted of patients who received umbilical cord blood grafts (n=23). Recipients were CMV-seronegative (n=10) and -seropositive (n=13), with 8 patients developing detectable hCMV in the blood 22 to 82 days after transplant. For this cohort, peripheral blood mononuclear cells (PBMC) were collected at day 28, day 60 and day 100 after transplant. Cryopreserved PBMC were thawed and rested overnight in cytokine-free media. After incubation with K562 cells a functional analysis was performed using multi-color flow cytometry for CD107a (a surrogate for cytotoxicity), IFNγ, CD56, CD3, CD158a, CD158b, CD158e, CD57, NKG2A, and NKG2C. In cohort 1 the frequency of NKG2C+ NK cells significantly increased between day 28 and day 60 (10±1.3% v 26±5%, p=0.01), and this increase persisted at day 100 in seropositive patients who reactivated CMV (32±4.8%, p=0.001). No change in NKG2C expression was detected in seropositive patients who did not reactivate CMV or in seronegative recipients, showing that this response was specific. There was no difference in the expression of CD107a by NKG2C+ NK cells in the three groups. In marked contrast, IFNγ production by NKG2C+ NK cells was low prior to CMV reactivation (0.8±1.4%) and increased at day 60 and day 100 (7.9±2.1%, p=0.012). The IFNγ response was specific to NKG2C+ NK cells as the NKG2C population produced minimal IFNγ at day 60 and day 100 (NKG2C+: 7.9±2.1% vs. NKG2C: 0.4±0.25%, p=0.003). The frequency of NKG2C+ NK cells with increased cytokine production correlated with CMV reactivation in this cohort. The second cohort included PBMCs collected at the time of viral reactivation (monitored weekly) and 2, 4, and 8 weeks later (n=10). The frequency of NKG2C+ NK cells increased following the detection hCMV viremia. The peak response (i.e., rise in NKG2C+ NK cells compared to baseline) was at 4 weeks post infection (23±3.7% vs. 13±1.8%, p=0.046), followed by contraction at 8 weeks. NKG2C+NK cells produced significantly more IFNγ than did NKG2C cells at 2, 4 (NKG2C+: 9.4±1.2% vs. NKG2C: 2.4±0.3%, p=0.0002) and 8 weeks post infection. The maximum IFNγ production occurred at 4 weeks, with the peak NKG2C+ frequency. We have previously demonstrated that KIR expression is essential to educate NK cells to produce IFNγ. Therefore, we measured coexpression of NKG2C and KIR in these cohorts. Four weeks after infection, NK cells that coexpressed NKG2C and KIR produced significantly more IFNγ than did the KIR+ NKG2CNK cells (11.4±1.66% vs. 5.6±0.83%, p=0.0078). Importantly, only the NKG2C+NK cells expressing KIR that recognized recipient MHC ligands made IFNγ. In contrast, NKG2C+ NK cells coexpressing KIR that did not recognize recipient ligands were hyporesponsive. Lastly, we tested whether coexpression of CD57, a phenotypic marker of NK cells maturity, might be present on NKG2C+ cells after CMV reactivation. Over the 8 weeks following CMV reactivation the NKG2C+ NK cells but not the NKG2CNK cells progressively acquired CD57, suggesting that CD57 may mark the human memory NK cell in response to CMV reactivation. In summary, our data demonstrate that CMV reactivation induces an NK cell IFNγ response specific to NKG2C+cells educated by recipient KIR ligands that then acquire CD57+. These findings support the emerging concept that hCMV specific, innate memory cell populations could be exploited for therapeutic purposes.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution