NKp30~+NK Cells Are Associated With HBV Control During Peg-interferon-alpha-2b Therapy Of Chronic Hepatitis B

CHB (Chronic hepatitis B) has nearly affected 240 million people worldwide. The current treatments for CHB include IFN-α (interferon), Peg-IFN-α (pegylated interferon) and five oral NAs (nucleotides analogues). But the anti-HBe seroconversion rate is about 30% with Peg-IFN-α and 20% with NAs. Consequently, it is necessary to improve the clinical efficacy of Peg-IFN-α. It is key point to understand the immune mechanisms underlying the differences between responders and non-responders.NK (natural killer) cells generally account for about 30-50% of intrahepatic lymphocytes. Peg-IFN-a can activate NK cells in a direct or in a indirect manner. Furthermore many receptors regulating NK cells activity are benefited for control HBV. In humans NK cells differentiates two subsets by CD56 expression. CD56dim NK cells accounting for 85-90% of all NK cells mediate cytotoxicity. CD56bright NK cells accounting for 5-10% of all NK cells secrete large amounts of cytokines, such as TNF-α and IFN-γ. The activating and inhibitory receptors expression on NK cells can regulate the function of target cytolysis or cytokines secretion.The aim of our study is to evaluate the relationship between NK cells and the response to Peg-IFN-α-2b treatment. In this research we recruited 102 HBeAg-positive CHB patients. They received standard of care treatment with Peg-IFN-α-2b for 1 year and with half a year observation period. Before and after the start of Peg-IFN-α-2b treatment peripheral blood samples from the patients were obtained. Peripheral NK cells were phenotypically and functionally measured. Virological assessments were also determined at different time points. The major findings of our research are shown as follows:1. Alteration in the numbers of NK cells, especially CD56bright NK subsets associated with treatment responseWe longitudinally analyzed and compared the frequencies and absolute numbers of NK cells from SR patients and NR patients. The absolute numbers of NK cells in SR patients were higher than in NR patients from 3 to 12 months. Compared to the baseline, the absolute numbers of NK cells decreased during Peg-IFN-a-2b therapy in NR patients, but not in SR patients. During treatment with Peg-IFN-a-2b, the SR patients gained a higher proportion of NK cells than did NR patients, as observed from 3 to 9 months. Compared to the baseline, there was a statistically significant increase in proportions of NK cells of SR patients starting at 3,6, and 9 months for those patients undergoing therapy. A different trend was found in NR subjects, in which a statistically significant decline in proportions of NK cells occurred, starting at 3 months after the start of therapy and continuing through follow-ups. Additionally, we detected two subsets of NK cells during Peg-IFN-a-2b therapy. The absolute numbers and percentages of CD56bright NK cells significantly increased during Peg-IFN-a-2b therapy. Moreover, the absolute numbers of CD56bright NK cells in SR patients were higher than those in NR patients from 0 to 12 months.2. Persistently increased NKp30+ NK cells were found in patients throughout the course of Peg-IFN-a-2b therapy, especially in SR patientsWe investigated the dynamics of the NK cell phenotypes influenced by Peg-IFN-a-2b. We analyzed the proportion of NCR-expressing NK cells at different time points. The percentage of NK cells expressing the activating natural cytotoxicity receptor (NCR) NKp30 increased significantly. Notably, it was observed that the absolute number of NKp30+ NK cells was significantly higher in SR patients than in NR patients from 3 to 12 months. However, the expression of other NCRs (NKp44 and NKp46) and the activating C-type lectin receptor NKG2D were not significantly different between the SR and NR patients during therapy. To further confirm the relationship between NKp30+ NK cells and virological outcome, we observed the clinical data from the longitudinal cohort. We found a more substantial reduction in HBsAg in the SR patients, but not in NR patients. Spearman’s rank correlation coefficient between decrease in log10 HBV-DNA (or HBeAg) and increase in NKp30+ NK cells was determined at 12 months. We found that the decrease in HBV-DNA and HBeAg correlated with the therapy-induced increase of NKp30+ NK cells in SR patients, but not in NR patients.3. Peg-IFN-a-2b potently increased the proliferation of NKp30+ NK cells by promoting IL-15 productionTo investigate whether the increase of circulating NKp30+ NK cells was due to enhanced proliferation, Ki-67 expression was examined. Ki-67 expression was significantly higher in NKp30+ NK cells after the start of Peg-IFN-a-2b therapy than at the baseline. Moreover, the NKp30 expression on NK cells was related to Ki-67 expression. We next investigated whether this increase in NKp30 expression on NK cells was induced by Peg-IFN-a-2b or IL-2/IL-15 in vitro. In the current study, NKp30 expression on NK cells was predominantly driven by IL-15, which could not be induced by a Peg-IFN-a-2b stimulus for 24 or 48 h in vitro. Next, we investigated which cells were responsible for the production of IL-15. Biologically active IL-15 is presented on the cell surface as a complex with the IL-15Ra chain (CD215), usually on macrophage/monocytes or dendritic cells. We observed that the pro-inflammatory CD14+CD16+ monocytes and classical CD14+CD16- monocytes expressed high levels of CD215 (IL-15Rα). Furthermore, in SR patients, the CD215 expression on two subsets of monocytes cells were significantly higher than in NR patients. The study’s data suggest that monocyte-deriving IL-15 plays an important role in regulating the capacity of NK cells to proliferate more rapidly in SR patients than in NR patients.4. Peg-IFN-a-2b therapy maintained NKp30+ NK cells’ anti-viral function in SR patientsTo investigate the dynamics of alterations in the NK cell effector function, we analyzed the proportion of NK cells expressing IFN-y and CD107a using PMA/Ion as a stimulant in vitro. There is a significant difference in the frequency of polyfunctional IFN-y+CD107+ NK cells in responders and non-responders. Furthermore, the proportions of NK cells expressing CD107a were also higher in the SR patients. However, the IFN-y-expressing NK cells were not different between the SR and NR patients. Together, these results indicate that IFN-y and CD107a-co-expressing NK cells and CD107a-expressing NK cells were beneficial for CHB patients’responses to Peg-IFN-α-2b therapy. To determine the function of NKp30+ NK cells from the SR and NR patients, PBMCs were stimulated with IL-12 ex vivo. We found the percentage of NKp30+ IFN-γ+ NK cells was significantly higher in SR patients than in NR patients from 6 to 9 months. Furthermore, we examined the differences between IFN-y-producing NKp30+ NK cells in SR and NR patients at the end of therapy. As expected, the percentage of IFN-y-producing NKp30+ NK cells was higher in SR patients than in NR patients.5. SR patients gained CD62L+ NKp30+ NK cells during Peg-IFN-a-2b therapyWe analyzed the proportion and total numbers of NKp30+ NK cells expressing CD62L in CD56bright and CD56dim populations. At the end of therapy, the total numbers of CD56bright NKp30+CD62L+and CD56dim NKp30+CD62L+NK cells were higher in SR patients than in NR patients. Furthermore, the percentage of CD56dim NKp30+ CD62L+NK cells was significantly higher in SR patients at the end of treatment. These results indicate that more CD56bright NKp30+NK cells and CD56dimNKp30+ NK cells exhibited the CD62L phenotype after the start of Peg-IFN-a therapy, which was beneficial for producing IFN-y.6. Peg-IFN-a-2b increased NKG2A expression on NKp30+NK cells, which negatively affected the clinical outcomeRecently, we reported that HBV infection increased NKG2A expression on NK cells of mice and humans, reducing their ability to clear HBV. Thus, we investigated whether this impaired capacity to inhibit HBV was affected by altered expression of inhibitory NK cell receptors in NR patients after the start of Peg-IFN-a therapy. The dynamics of NKG2A, Tim-3, and PD-1 expression on NK cells were observed. Tim-3 and PD-1 are two important receptors on exhausted HBV-specific CD8+ T cells. In our study, the inhibitory receptor NKG2A was markedly down-regulated on NK cells after the start of Peg-IFN-a-2b therapy in SR patients, but not in NR patients. There was no decrease in Tim-3 or PD-1 expression on NK cells during the Peg-IFN-α therapy. We observed the NKG2A expression on the NKp30+NK cells between the SR and the NR patients during therapy (Figure 6C). We found that the percentage of NKG2A+ NKp30+NK cells was higher in NR patients than in SR patients during Peg-IFN-α-2b therapy, especially at later periods (6,9, and 12 months) and during follow-up (15 and 18 months). By contrast, the percentage of NKG2A-NKp30+ NK cells was higher in SR patients than in NR patients. Furthermore, this percentage of NKG2A+ NKp30+ NK cells increase was significant in the NR patients. The summation of inhibitory signals, when not opposed by activating signals, led to an inhibition of NK cell function. We hypothesized that a low expression of NKG2A in combination with a high level of NKp30 mediated a protective effect in the SR patients. To address this problem, an in vitro co-culture experiment was performed. The purified CD3- CD56+ NK cells from the SR or NR patients were co-cultured with K562 cells. We found a higher percentage of CD107a expressing CD56bright NKp30+NK cells and CD56dim NKp30+ NK cells compared with the NR patients. In conclusion, the high level of NKG2A expression on NKp30+ NK cells might have inhibited the degranulation capacity of NK cells in the NR patients, which negatively affected the clinical outcome.