Correction: Hepatic alveolar echinococcosis infection induces a decrease in NK cell function through high expression of NKG2A in patients

Abuduaini AbuliziTalaiti TuerganPaizula ShalayiadangChuanshan ZhangRuiqing ZhangTiemin JiangQiang GuoHui WangLiang LiRenyong LinYingmei ShaoTuerganaili Aji^*

• The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China

Human alveolar echinococcosis (AE) is a potentially lethal zoonosis caused by the cestode Echinococcus multilocularis (E. multilocularis) [1]; it is prevalent mainly in Western China, the Middle East and Central Europe [2, 3]. China accounts for 91% of the global AE burden every year; thus, sustained efforts have been made to prevent, control, and manage this disease[2]. Humans can contract AE through contaminated food or water, and the liver is the primary infected organ. If not treated in time, the disease infiltrates and consequently leads to critical involvement of the vasculature and to jaundice, cirrhosis, and other clinical symptoms, which can result in liver failure or even death[4]. The fatality rate of untreated or inadequately treated patients with human alveolar echinococcosis is 90% at 10–15 years after diagnosis[4, 5]. Hence, it is regarded as a “parasitic cancer”[6].The immune mechanism of E. multilocularis larvae is important for self-healing or persistent chronic infection after they enter the liver. In hepatic AE patients and animal models, long-term parasitism by larvae is mediated mainly by regulatory T cells and related cytokines, such as IL-10 and TGF-β[7]. Our recent studies demonstrated the potential importance of the remaining Th subsets, such as Th17, Treg and Th9, in E. multilocularis infection[8]. Our data indicated that E. multilocularis can induce T-cell exhaustion through the inhibitory receptor TIGIT and that blocking this checkpoint may reverse the functional impairment of T cells and represent a possible approach to immunotherapy against AE[9]. In addition, E. multilocularis vesicular fluid increases PD-L1 and CTLA-4 expression on T cells[10].As important innate immune cells, NK cells constitute the first line of defence against infection and tumours. Many related studies have shown its special role in the tumour progression of liver cancer patients[11]. The ability of intrahepatic NK cells to secrete IFN-γ and induce cytotoxicity is decreased significantly in hepatocellular carcinoma patients[12]. The intracellular parasite Toxoplasma gondii impairs the ability of NK cells to recognize target cells and reduces the secretion of IFN-γ in the host[13]. Malaria might increase the ability of NK cells to secrete IFN-γ via KIR/HLA molecules in early-stage infection[14], and Leishmania can inhibit the proliferation of NK cells[15]. Extracellular parasites: When the percentage of NK cells decreases, the expression of NKG2D, CD69 and Ly49A is upregulated in individuals infected with Angiostrongylus cantonensis[16], and Schistosoma japonicum inhibits liver fibrosis by activating liver NK cells and increasing IFN-γ secretion[17]. Several related studies have shown that in patients with hepatic alveolar echinococcosis, the activity of NK cells in the peripheral blood is significantly downregulated, and E. multilocularis vesicular fluid might inhibit the activation and proliferation of NK cells in mice[18].NKG2A is specifically expressed on the surface of some lymphocytes, such as NK cells, T cells and NKT cells, and transduces inhibitory signals[19]. High expression of NKG2A induces functional downregulation of NK cells and is related to poor prognosis in patients with hepatic cellular carcinoma[20]. Currently, the newest research reports that NKG2A is a new checkpoint, and blocking NKG2A with monolizumab could promote the antitumour immune activity of CD8+ T cells and NK cells, which could be used as a supplement to the first generation of cancer immunotherapy[11]. Our recent study demonstrated that a reduction in NK cell frequency and increased NKG2A might result in low cytotoxic activity through decreased IFN-γ secretion during E. multilocularis infection[18]. Therefore, studies on the correlation between the expression of NKG2A on NK cells and the pathological and clinical parameters of hepatic AE are very important.However, some research has focused on immune interactions in E. multilocularis infection. NK cells are involved in E. multilocularis infection both in vitro and in vivo. However, very little is known regarding the possibility of immune dysfunction in hepatic AE patients. In this study, we demonstrated that NKG2A expression in hepatic NK cells and its functional exhaustion, such as decreased IFN-γ production and decreased cellular cytotoxicity. These findings may indicate the existence of a negative regulatory mechanism in exhausted NK cells as a result of the increased expression of the inhibitory receptor NKG2A in hepatic AE patients.Patients and methods Liver tissues from 36 hepatic AE patients and fresh hepatic AE liver tissue samples were obtained from 10 hepatic AE patients who underwent radical hepatectomy between 2012 and 2015 in the Department of Hepatobiliary and Echinococcosis Surgery Department of the First Affiliated Hospital of Xinjiang Medical University[21-23]. In accordance with the location of surgical samples from patients with hepatic AE, the liver tissue adjacent to the lesion (0.5 cm) and its volume of approximately 2 cm × 2 cm × 2 cm were selected as the close liver lesion tissue (CLT), and the corresponding normal liver tissue approximately 3–5 cm away from the lesion was taken as the distal liver tissue (DLT). All fresh tissues were used for phenotypic analysis, and most of them were also used for intracellular cytokine analysis if they had a sufficient number of cells. All patients provided written informed consent in accordance with the Helsinki Declaration. The protocols for all study cohorts were approved by the Ethics Board of Xinjiang Medical University.Flow cytometryLiver tissue-infiltrating leukocytes were obtained as previously described[9]. To digest the samples, they were cut into small pieces and digested in RPMI 1640 (HyClone Laboratories, USA) supplemented with collagenase IV (10 mg/mL, Sigma‒Aldrich) and DNase I (33.3 mg/mL, Sigma‒Aldrich, United States) at 37 °C for 1–2 h. The peripheral lymphocytes, liver tissue-infiltrating leukocytes and NK cells from the in vitro cultures were stained with fluorochrome-conjugated Abs and then analysed through flow cytometry[24]. Abs against the following proteins were used for staining: CD3 (UCHT1), CD56 (HCD56), CD16 (3G8), CD69 (FN50), NKG2D (1D11), IFN-γ (4S. B3), TNF-α (MAb11), perforin (dG9), granzyme B (GB11) (BioLegend, United States), and NKG2A (131411) (R&D Systems, United States). The stained cells were analysed using an LSRFortessa flow cytometer (Becton Dickinson, United States), and the data were analysed using FlowJo analysis software V10 (Treestar, United States).ImmunohistochemistryThe paraffin-embedded sections were dewaxed in xylene and rehydrated with distilled water. Following incubation with antibodies against human NKG2A (PA5-21949, Thermo Fisher, United States), adjacent sections were stained with the DAB Peroxidase Substrate Kit (SK-4100) (Vector Laboratories, United States). Positive and negative controls were tested before formal staining. The Olympus Optical Microscope Cellsens scanning software (Olympus, Japan) was used to perform scanning analysis, and the positive staining surface around the lesion was calculated[25].NK cell purificationPeripheral blood was obtained from healthy controls, and purified NK cells were enriched from whole blood via RosetteSep™ Human NK Cell Cocktail (50 μl/ml) (Stem Cell Technologies, Canada). The cells were incubated in medium alone or with recombinant human IL-12 p40 (200-12p40) (10 ng/mL; Peprotech, United States) plus recombinant human IL-15 (200-15) (10 ng/mL; Peprotech, United States), with IL-12+IL-15 plus recombinant human TGF-β1 (1 ng/mL; PeproTech, United States), or with IL-12+IL-15 plus EMP (Echinococcus multilocularis protein). The protein was extracted and quantified after grinding Echinococcus multilocularis. NK cells were stimulated for 24 h at 37 °C in an atmosphere of 5% CO2. NK cell viability was evaluated via the trypan blue exclusion method and analysed via flow cytometry after incubation.Statistical analysisStatistical analysis was performed using GraphPad Prism 7.0 (GraphPad Software, San Diego, CA). The Wilcoxon nonparametric statistical test or Mann–Whitney nonparametric statistical test was used when there were more than two groups. P < 0.05 was considered statistically significant. (P values are presented as * P < 0.05; ** P < 0.01; *** P < 0.001; **** P < 0.0001).Increased NKG2A expression in NK cells from the close lesion tissue of hepatic AE patientsTo investigate the expression of NKG2A in intrahepatic NK cells, we analysed the close lesion tissue (CLT) and distal lesion tissue (DLT) of liver AE lesions. As shown in Fig. 1A, 1B and 1C, the percentage of NKG2A+ NK cells in the CLT was significantly greater than that in the DLT. As previously demonstrated[26], we analysed the expression of NKG2A on CD56dim and CD56bright NK cells in DLT and CLT, and reported that the percentage of NKG2A+CD56dim NK cells in CLTs was significantly greater than that in DLTs. There was a negative correlation between the expression of NKG2A on NK cells in the CLT and alkaline phosphatase (ALP) (P = 0.0234, r = -0.7212)(Fig 1D), but there was no correlation between alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels(Fig 1E-G). These results suggest that the percentage of NKG2A+ NK cells in the CLT of hepatic AE patients was significantly increased, which was mainly manifested by the upregulation of NKG2A expression on CD56dim NK cells.Increased expression of NKG2A in NK cells induced a significant reduction in IFN-γ production in close lesion tissue compared with distal lesion tissue from hepatic AE patientsThe function of NK cells is affected by combining both activating receptors and inhibitory receptors. According to previous studies[27], NKG2A-induced phenotypic changes might be accompanied by functional alterations in NK cells. Therefore, we analysed IFN-γ, TNF-α, Granzyme B and perforin production in NK cells (Fig. 2). There was a significant reduction in IFN-γ production in the CLT compared with DLT (Fig. 2A). However, there was no significant difference in Granzyme B, perforin or TNF-α levels (Fig. 2B, 2C, 2D). A negative correlation was detected between IFN-γ production and NKG2A expression in NK cells from the CLT (Fig. 2E). In addition, the capacity of NKG2A+ NK cells from CLT regions to produce IFN-γ and granzyme B was also significantly decreased (Fig. 2F, 2G). However, the ability of NKG2A+ NK cells to produce TNF-α and perforin was not significantly different between the CLT and the DLT (Fig. 2H, 2I). In addition, the ability of NKG2A+ NK cells to secrete IFN-γ and Granzyme B was, surprisingly, significantly lower than that of NKG2A- NK cells (Fig. 3A, 3B). In contrast, NKG2A+ NK cells produced much higher perforin levels than NKG2A+ NK cells did (Fig. 3C), but there is there was no significant difference produced TNF-α levels(Fig. 3D) . These findings suggest that the increased expression of NKG2A on NK cells in the CLT decreased the ability of NK cells to secrete IFN-γ.Higher NKG2A expression is correlated with lesion activity and fibrosis in hepatic AE patientsAfter the expression of NKG2A in CLT and DLT tissues from 36 patients with hepatic AE was examined, our results suggested that the expression of NKG2A was significantly increased in the CLT (Fig. 4A, Fig. 4B). We also analysed the correlation between the ratio of the NKG2A expression area in the CLT and DLT tissues (CLT/DLT) and the PET-CT value (SUVmax) and found a positive correlation between NKG2A expression and the PET-CT value (SUVmax) (P=0.0065, R=0.735) (Fig. 4C). In this study, according to liver tissue samples from patients with hepatic AE who underwent PET‒CT examination, Masson staining was used to detect the collagen area (fibrotic area) around the lesion, and α-SMA immunohistochemical staining was used to detect the positive expression of hepatic stellate cells. A PET‒CT (SUVmax) value greater than or equal to 3.0 was included in the high-activity lesion group, and a value less than 3.0 was included in the low-activity lesion group. The results revealed that the area of fibrosis around lesions was significantly lower in high-activity lesions than in low-activity lesions (Fig. 4D, 4E), and a negative correlation existed between the NKG2A+ expression area and the fibrosis prelesion area (Fig. 4F, 4G) (these results need to be further verified by expanding the sample size). These results suggest that the expression of NKG2A in the CLT was upregulated and was positively correlated with the activity of the lesion. Moreover, the greater activity of the lesion and lower degree of fibrosis around the lesion result in the lesion not being fully limited by the fibrous layer, which induces faster growth of the hepatic AE lesion.Compared with those of CD56bright NK cells, the percentages of NK cells are predominantly decreased on CD56dim NK cellsHuman NK cells are divided into two subsets, and their main functions are not exactly the same: CD56bright NK cells are responsible for cytokine secretion, whereas CD56dim NK cells are responsible for cytotoxicity[28]. We analysed the percentage of NKG2A expression in different subsets of NK cells in hepatic AE patients. The gating strategy used to separate the CD56bright and CD56dim NK cells is shown in Fig. 5A. We observed that the percentages of total NK cells (Fig. 5B) and CD56bright NK cells (Fig. 5C) in the CLT and DLT tissues were not significantly different. However, the percentage of CD56dim NK cells among total NK cells decreased significantly in the CLT (Fig. 5D). These results further confirmed that the decreased percentage of the NK cell subset was mainly CD56dim NK cells in the CLT tissues of hepatic AE patients.DiscussionOur study demonstrated the obviously increased expression of NKG2A in NK cells in the CLT, which might induce a significant reduction in IFN-γ production from closely related lesion tissue; moreover, increased NKG2A expression is correlated with lesion activity and fibrosis in hepatic AE patients. According to present research reports, this is the first report on hepatic NK cells and their related functions in hepatic AE patients.The percentage of human liver NK cells among total lymphocytes is approximately 25%~40%[29], and approximately 90% of NK cells in the peripheral blood and spleen belong to the CD56dim subset, whereas only 50% of NK cells in the liver belong to the CD56dim subset. NK cell subsets have different effects on the proliferation response, cytotoxicity, cytokine production, and expression of NK cell receptors and adhesion molecules[29]. NK cells, as critical components of the innate immune system, are important effector lymphocyte populations involved in antitumour and anti-infection immunity[30]. However, in the context of tumours and chronic infections, NK cells exhibit an exhausted status similar to that of exhausted T cells, resulting in poor effector function and an altered phenotype. In ovarian cancer, the expression of NKG2A is upregulated, which induces NK cell failure[28]. The upregulation of NKG2A on NK cells might indicate antitumour immune tolerance and promote tumour metastasis in lung cancer patients[31]. Previous studies have shown that the NK toxicity of peripheral blood mononuclear cells (PBMCs) in AE patients is lower than that in nonparasitic biliary disease patients[32]. Another recent study revealed that the vesicular fluid of E. multilocularis has an inhibitory effect on the activation and proliferation of NK cells in human PBMCs[10]. According to previous relevant studies and our research, the increased expression of NKG2A on NK cells in the CLT of hepatic AE patients might be the main mechanism mediating NK cell dysfunction. Monalizumab treatment can restore CD107 and IFN-γ production in NK cells against various tumor cells[33]. When class I MHC ligands of NK cell inhibitory receptors are downregulated, which commonly occurs in tumour cells, the loss of inhibitory signals and the resulting unabated positive signalling also leads to NK cell activation (IFN-γ and TNF-α). This phenomenon is referred to as the ‘missing-self’ response[34]. So we also demonstrated that the ability of NK cells to secrete IFN-γ and granzyme B in the CLT of a lesion is significantly lower than that in the DLT. Moreover, the percentage of NKG2A+NK cells in the CLT may be negatively correlated with the percentage of IFN-γ secretion. In other words, the inhibitory molecule NKG2A is upregulated on NK cells in the CLT of the lesion, which is the area with the strongest inflammatory immune response, possibly leading to the downregulation of NK cell function. Humans are occasional intermediate hosts, and the severity of liver AE in humans is caused by the continuous asexual reproduction of E. multilocularis and strong inflammatory granuloma infiltration around it, which leads to pathological damage to the liver after parasitic infection. In this study, we analysed the positive expression of NKG2A in the CLT and DLT surrounding liver AE lesions using immunohistochemical staining techniques. The results suggest that the positive expression of NKG2A in the CLT was significantly greater than that in the DLT and that there might be a positive correlation between NKG2A (CLT/DLT) and PET-CT (SUVmax) values. Reuter[35] used positron emission tomography (PET) to track the metabolic changes in 18F-FDG in liver AE lesions, and the active area around the liver AE lesion exhibited differences in energy metabolism. Therefore, at present, PET‒CT (SUVmax) values can be used to effectively determine the activity of AE lesions. On the basis of our research results, the greater the expression of NKG2A in the surrounding tissues of the lesion is, the greater the activity of the lesion, the significantly lower the degree of fibrosis around the lesion, the weaker the fibrous layer to restrict the lesion, and the greater the speed at which the lesion grows and invades the surrounding normal liver tissue.In summary, infection of the host with E. multilocularis leads to infiltration of inflammatory cells, proliferation of fibroblasts and promotion of organ fibrosis. Moreover, the fibrous layer is involved in the inflammatory response around the larvae. On the one hand, this can effectively prevent the parasite from receiving a sufficient immune response from the host, which is beneficial for parasite growth and continued parasitism. On the other hand, it limits the parasite's continued growth, invasion, and growth of new metastatic lesions. Our research revealed that after infection with E. multilocularis, the increased expression of NKG2A on NK cells leads to the downregulation of IFN-γ secretion. The increase in NKG2A expression in the tissue near the lesion in patients with liver AE is similar to that observed in animal experiments, leading to a decrease in the function of NK cells, which secrete IFN-γ and granzyme B, and a reduction in fibrosis around the lesion may increase lesion activity and growth. However, further research is needed to explore the detailed interaction mechanism involved and we need to expand the sample size to further validate the conclusion.