Introduction
Despite the existence of several effective preventive vaccines [
1], cervical cancer remains the most common malignancies of the female genital tract. According to recent GLOBOCAN statistics, there are approximately 604,000 new cases of cervical cancer and 342,000 deaths annually [
2]. Persistent infection with high-risk subtype Human Papillomaviruses (HPV) [
3,
4], especially HPV16, is a precursor of precancerous lesions of the cervix, also known as cervical intraepithelial neoplasia (CIN), which eventually leads to cancer in many cases [
5,
6]. Natural killer (NK) cells are an important component of innate immunity and represent the first line of defense against viral infections. Receptors, activators or inhibitors, expressed on the surface of NK cells can contribute to the NK cell activation [
7]. Research has shown that NK cells appear in early HPV-infected lesions, can be essential to the pathogenesis of the lesions [
8]. Most HPV infections are transient and self-limiting, nevertheless, in approximately 1–2% of the affected it might persist, suggesting that HPV may have evolved an escape strategy in response to the pursuit of NK cells by regulating its activation [
9,
10]. Currently, the drivers for HPV to escape NK cell immune response in the occurrence and progression of CIN and cervical cancer are still unclear, especially the action mechanism of different NK cell subsets in the progression of HPV-related CIN.
In recent years, with the development of immune checkpoint blockade treatment, NK cells have shown promise for targeted therapy [
11]. Previous studies have demonstrated that NK cells play a pivotal role in the immune response against early-stage HPV-associated CIN, especially in low-grade CIN tissues with a relatively low viral load. Besides, NK cells, especially the CD56
brightNK cells with immunomodulatory properties, can infiltrate and attack cancerous tissues more easily. Unlike B cells and T cells, the activation of NK cells relies heavily on the recognition of major histocompatibility complex (MHC)-I-like molecules by inhibitory receptors to distinguish between “self” and “non-self” [
7]. In fact, many virus-infected cells and cancer cells down-regulate the expression of MHC-I molecules to escape the detection of cytotoxic CD8
+T cells. Remarkably, NK cells can recognize and respond to those target cells with “missing-self” phenotype and attack them [
7,
12‐
14]. However, despite the loss expression of MHC-I molecules, NK cells still show reduced function in patients with CIN and cervical cancer [
15]. Based on this, we believe that this may be associated with the non-MHC-1 molecular pathway described by previous researchers [
16]. TIGIT, KLRG1, Siglec-7, LAIR-1, and CD300a are important non-MHC-I inhibitory receptors on NK cells, involved in the NK cell-mediated regulation of various diseases [
17‐
21]. However, it remains unclear whether non-MHC-I inhibitory receptors are expressed in HPV16(+) CIN, or whether the activation of related pathways suppresses NK cell function and is associated with impaired viral clearance.
To assess the roles of NK cells and these inhibitory receptors in the immunopathogenesis of HPV16(+) CIN, we examined the subset distribution, Interferon-γ (IFN-γ) and Tumor necrosis factor-α (TNF-α) expression levels and inhibitory immunophenotype expression levels of peripheral blood NK cells from HPV16(+) women with different CIN grades or HPV(−) women. We also measured levels of ligands corresponding to these inhibitory receptors in tissues of cervical. We found that the upregulation expression of TIGIT, KLRG1 on CD56brightNK cells and its ligands Poliovirus receptor (PVR), E-Cadherin, N-Cadherin is directly associated with HPV16(+) CIN, suggesting that NK cell surface inhibitory receptors may be the key components responsible for the compromised viral clearance due to impaired immunity in HPV16(+) subjects with CIN.
Materials and methods
Study participants and procedures
All the participants were recruited from the Fourth Medical Center of the Chinese PLA (People’s Liberation Army) General Hospital, Beijing, China. HPV samples and biopsy specimens were obtained during the second colposcopy for female subjects whose liquid-based cytology results during the first visit returned positive for atypical squamous cells, low-grade squamous intraepithelial lesion, or high-grade squamous intraepithelial lesion. These specimens were subjected to independent histological analysis by three pathologists and categorized as no intraepithelial neoplasia (CIN 0), low-grade intraepithelial neoplasia (CIN I), and high-grade intraepithelial neoplasia (CIN II–III). The exclusion criteria were as follows: pregnancy, other infectious diseases, acute reproductive tract infections, receiving anti-inflammatory therapy, tumors, immunodeficiency, and receiving immunosuppressive agents. The final cohort of 82 participants was divided into four groups: (1) 26 HPV16 subjects with CIN 0; (2) 17 HPV16(+) subjects with CIN 0; (3) 16 HPV16(+) subjects with CIN I; (4) 23 HPV16(+) subjects with CIN II–III. In addition, archival paraffin-embedded tissue samples were also collected and categorized into the above-mentioned four groups, each of which comprised 10 cases. All participants recruited in this study were aged between 20 and 60 (37.68 ± 10.36 years) and had signed an informed consent form. This study was reviewed and approved by the hospital ethics committee: Ethics Committee of the Fourth Medical Center of PLA General Hospital (2021KY006-HS001).
Polymerase chain reaction (PCR) for HPV DNA typing
Cervical cytobrushes used to collect cervical exfoliated cells were placed into 10 ml gigenespecimen transport medium (Qiagen) and DNA was extracted from the cervical samples using the DNeasy tissue kit (Qiagen), according to the manufacturer’s instructions. The DNA was then subjected to qualitative PCR using specific primers. The β-Globin was amplified to verify that the samples contained DNA of sufficient quality and quantity for the HPV test. Water was used as a negative control in each PCR experiment.
Flow cytometry
Blood samples (100 µl) collected from the participants were incubated for 25 min at 25 °C in a flow cytometry tube with the following specific antibodies (5 µl each): CD3 (563219, BD), CD16 (302018, Biolegend), CD56 (555518, BD), TIGIT (12-9500-42, eBioscience), DNAM-1 (559788, BD), NKG2A (130-113-565, Miltenyi Biotec), LAIR-1 (ab27744, Abcam), Siglec-7 (ab200558, Abcam), and KLRG1 (25–9488-42, eBioscience). In addition, an isotype control was prepared for each sample. After incubation, the reaction mixture in each tube was mixed thoroughly with 2 ml of hemolysin. The samples were then washed for flow cytometry analysis (BD FACSCanto™ Flow Cytometer) within 4 h.
Cytokine flow cytometry assay
3 μl of blood samples were collected from the participants and subjected to density gradient centrifugation to isolate peripheral blood mononuclear cells (PBMC). After being washed and resuspended, freshly isolated PBMCs were seeded at a density of 106 cells/well and incubated with 100 ng/ml of phorbol myristate acetate (524400, sigma), 1 μg/ml of ionomycin (407951, sigma), and 10 μg/ml of brefeldin A (555029, BD) at 37 °C and 5% CO2 for 6 h. Subsequently, the stimulated cells were incubated with 5 μl of CD56, CD3, CD16, and isotype controls (as mentioned above) in the dark at 25 °C for 25 min for cell surface labeling. The cells were then fixed and permeabilized using Fixation/Permeabilization Solution kit (555028, BD) and incubated in the dark with 10 μl of anti-IFN-γ (552887, BD) and 8 μl of anti-TNF-α (562083, BD) at 4 °C for 30 min. After washing, the samples were loaded for flow cytometry assay.
Immunohistochemical (IHC) assay
Cervical biopsy samples were routinely embedded in paraffin and sectioned at a thickness of 3 μm, followed by hematoxylin–eosin (HE) staining. The HE-stained tissue sections were then examined by three experienced pathologists. Prior to IHC staining, the tissue sections were dewaxed in xylene and rehydrated in graded alcohol. The tissue sections were incubated with the following antibodies: PVR (ab267389, diluted at a ratio of 1:300, Abcam), E-Cadherin and N-Cadherin (MAB0738 and MAB0571, respectively, diluted at a ratio of 1:200, MXB Biotechnologies), followed by IHC staining using the streptavidin-peroxidase conjugate method. Negative controls were prepared with non-immune isotype-specific sera used in place of primary antibodies, while previously positively-stained tissue sections for the labeled proteins were taken as positive controls.
Statistical analysis
Statistical analysis was performed using GraphPad 8.0.1. Flow cytometry data were expressed as percentages and median fluorescence intensity (MFI). A nonparametric test (Kruskal–Wallis test) was used for median comparisons and completed with Dunn’s multiple comparisons tests. In experiments involving histology or IHC, the figures shown are representative of the tissue sections. The Allred scoring system was used for IHC staining quantification [
22]. Proportion scores were: 0, none; 1, less than one percent; 2, one percent to one tenth; 3, one tenth to one third; 4. one third to two thirds; 5, Two thirds or more. The intensity scores were: 1 for weak, 2 for medium, and 3 for strong. We evaluated 10 high-power fields (× 400 magnification) in each sample. Added the average of the intensity score and the proportional score as the Allred total score (range = 0–8). The image was captured with an Olympus BX40F microscope (Olympus). All statistical assessments were two-sided, and statistical significance was defined as
P < 0.05.
Discussion
Persistent high-risk HPV infection is the most important risk factor of cervical cancer, and the virus clearance barrier caused by immune escape is a key link in the persistent infection of HPV [
23]. Prompt detection and clearance of HPV by the acquired immune system are rarely attained, as HPV is an epitheliotropic virus that often establishes local infections. NK cells are incapable of preventing persistent HR-HPV infections among specific populations despite exhibiting effective antiviral activity during the early stage of viral infection. Thus, an in-depth investigation of the evasion mechanism from NK cell-mediated immune responses by HPV is of great clinical significance for the development of therapeutic vaccines and drugs, the elimination of persistent HPV infection, and the effective prevention of cervical cancer [
24].
In this study, we investigated the expression of inhibitory receptors and secretion levels of IFN-γ and TNF-α in CD56bright and CD56dim NK cells from HPV(−) women or HPV16(+) women with different CIN grades. The results revealed that the HPV16(+) CIN groups had a significantly increased number of CD56bright NK cells and significantly reduced ability to secrete IFN-γ. Further, the expression of the inhibitory receptors TIGIT and KLRG1, as well as their ligands PVR, N-Cadherin and E-Cadherin were significantly increased. It would be speculated that TIGIT, KLRG1 and its ligands PVR, E-cadherin and N-cadherin are involved in the immune response of CD56brightNK cells in HPV16(+) CIN, affecting the immune status of patients in different clinical stages to a certain extent, and promoting the progression of CIN.
NK cells play a key role in the early control of viral infections and tumor immunosurveillance [
25,
26]. CD56
dim NK cells, which account for approximately 90% of circulating NK cells, carry high levels of perforin and granzyme in the cytoplasm and primarily exert cytotoxic effects [
27]. CD56
bright NK cells, which account for about 10% of circulating NK cells and are widely present in the human liver and uterus, secrete high levels of cytokines and are key components of the first line of host defense against viral infections. Previous studies have shown that CD56
bright NK cells perform important functions in the clearance of HPV-, hepatitis B virus-, and respiratory syncytial virus-infected cells in the early stages, as well as in the prevention of persistent viral infections [
28‐
31]. In some diseases, the CD56
bright NK cell subset may be selectively expanded due to its unique cytokine profile, which is consistent with our findings [
32]. Our study showed that subjects in the HPV16(+) CIN II–III group had a significantly increased number of circulating CD56
bright NK cells. Similarly, the number of circulating NK cells increased, with significantly elevated secretion levels of IFN-γ and TNF-α in the early stage of acute HBV infection. However, different ligands and cytokines render NK cells tolerant to HBV and reduce their antiviral capacity following the progression to chronic infection [
33,
34].
In this regard, we speculated that the infection of HPV may locally inhibit the immunomodulatory activity of CD56
bright NK cells via certain mechanisms, leading to the compensatory proliferation of this particular NK cell subpopulation. Our subsequent analysis of the ability of the circulating NK cells to secrete effector cytokines showed that CD56
bright and CD56
dim NK cells in healthy subjects exhibited a remarkable ability to secrete IFN-γ and TNF-α, respectively. However, NK cells, particularly CD56
bright NK cells, in subjects with CIN exhibited a significantly reduced ability to secrete IFN-γ, which is consistent with previous findings [
35,
36]. Taken together, these findings indicate that the effector function of CD56
bright NK cells is reduced during the progression of HPV16(+) CIN, and that the increased number of CD56
bright NK cells did not contribute to the timely clearance of virus-infected cells.
NK cell activation is balanced by signals transduced by activating or inhibitory receptors [
37]. Under normal conditions, activation relies heavily on the inhibitory receptor to recognize MHC-I molecules to distinguish “self” from “non-self”. Decreased expression or structural abnormalities in MHC-I molecules on the surface of virus-infected cells and tumor cells affect the recognition of ligands by corresponding inhibitory receptors on NK cells, thereby activating NK cells. HPV-mediated immune evasion of CIN and cervical cancer is often accompanied by downregulation of the MHC-I molecules and NK cells would be activated to mediate HPV clearance in this case [
14]. However, findings from previous studies have revealed varying degrees of reduction in NK cells function from subjects with HPV-mediated CIN which suggests the involvement of alternative mechanisms to prevent NK cell activation. Some studies have reported that during infections with down-regulated expression of MHC-I molecules, non MHC-I specific inhibitory receptors participate in virus immune escape [
38].
Studies have shown that NK cells express some important non-MHC-I inhibitory receptors, such as TIGIT, KLRG1 et al. [
19,
39‐
44]. It is worth pointing out that CD56
bright NK cells also highly express a non-canonical MHC-I receptor NKG2A. In order to identify the inhibitory receptors that affect the function of CD56
brightNK cells in the progression of HPV16(+) CIN, we screened some of candidates recrptor: TIGIT, KLRG1, Siglec-7, LAIR-1, CD300a and NKG2A, and all data were shown in Table
1. Siglec-7 is a member of the immunoglobulin-type lectin (Siglec) family and is highly expressed on NK cells. The ligand of Siglec-7 is sialylated glycans. Studies have shown that NK cell-mediated cytotoxicity was enhanced by desialylation and blockade of Siglec-7 in renal cell carcinoma, melanoma, colon adenocarcinoma, cervical cancer, and chronic myeloid leukemia [
19,
39]. NKG2A recognizes the non-classical MHC-I molecule HLA-E, and the expression of NKG2A is found to be up-regulated in peripheral blood NK cells of patients with chronic hepatitis. Antiviral therapy against NKG2A in patients with chronic hepatitis causes NK cells to produce more IFN-γ [
40]. At present, humanized anti-NKG2A monoclonal antibody (IPH2201, monalizumab) has been used in several clinical trials [
41]. Leukocyte-associated immunoglobulin-like receptor-1 (LAIR-1) widely exists in immune cells, and the ligand of LAIR-1 is collagen, indicating that extracellular matrix may have the function of immune regulation. Relevant studies have shown that LAIR-1 can mediate immune escape in oral squamous cell carcinoma and hepatocellular carcinoma [
42,
43]. Phosphatidylserine (PS)/phosphatidylethanolamine (PE) interacts with CD300a, revealing its role in immune function regulation and involvement in disease host responses. CD300a exhibits therapeutic potential in HIV infection, CMV and Pichinde virus infection [
44]. However, our screening results showed that Siglec-7, NKG2A, LAIR-1 and CD300a were not significantly different in our group. These receptors may be less responsive in the early stages of chronic HPV16 infection. Alternatively, the progression from CIN to cervical cancer takes a certain amount of time, and the pathological changes of the two disease states are quite different. Our future studies will further explore the expression of these receptors in cervical cancer.
Our results confirmed an increased proportion of TIGIT expressed on CD56
bright NK cells in HPV16(+) CIN. Although there was no statistically significant, the MFI of TIGIT showed a consistent trend with percentage across groups. We consider that persistent HPV infection may preferentially stimulate the overall increase in the proportion of TIGIT expressed on CD56
bright NK cells, and individual differences between samples may also be the reason for this phenomenon. Previous studies have shown that TIGIT is involved in the immune response to tumor and viral infections. Johnston et al. [
45] had previously demonstrated that TIGIT was significantly upregulated in tumor-infiltrating lymphocytes in endometriosis, breast cancer, clear cell renal cell carcinoma, non-small cell lung cancer, and colorectal cancer. In addition, patients infected with human immunodeficiency virus (HIV) had a higher TIGIT expression level in NK cells than HIV. In this study, researchers found that high levels of TIGIT reversibly inhibited IFN-γ production in the NK cells [
46]. Previous studies have indicated that the inhibitory interaction of the TIGIT-PVR axis is critical for the functional heterogeneity of NK cells, by which it negatively regulates cell viability and induces immune tolerance [
33,
34]. We observed that PVR, the ligand of TIGIT, had a relatively high expression level in HPV16(+) subjects with CIN. Therefore, we speculated that PVR might respond to TIGIT, although we do not have sufficient evidence to prove that TIGIT-PVR is involved in HPV16 immune evasion. Our results could hypothesis that the TIGIT-PVR pathway may be an important negative regulator for inducing HPV16 immune evasion, which will be our important research work in the future. In fact, clinical trials of anti-TIGIT antibodies are ongoing. For instance, the Phase II CITYSCAPE trial evaluated the combination of tiragolumab (anti-TIGIT antibody) and atezolizumab (anti-PD-L1 antibody) in the treatment of non-small cell lung cancer with encouraging results [
47].
KLRG1 is the other important non-MHC-I specific receptor on NK cells. Cadherin type I, such as -E, N cadherin, has been identified as the ligand of KLRG1 [
48]. KLRG1 is a transmembrane receptor of the lectin-like superfamily predominantly mediating inhibitory effect through the cytoplasmic ITIM motif. Previous studies showed that KLRG1 was rapidly upregulated on NK cells, CD8
+ and CD4
+ T cells during chronic infections with the human cytomegalovirus, Epstein-Barr virus, and HIV [
48,
49]. Additionally, it was found that the expression of KLRG1 increased in chronic hepatitis C [
50]. Ito et al. [
49] showed that the interaction of cadherin with KLRG1 activated the inhibitory pathway of NK cells. Our results showed the expression of KLRG1 on circulating CD56
bright NK cells was increased primarily in HPV16(+) subjects with CIN 0 and CIN I. Although its characteristic expression in peripheral blood circulation was not found in high-grade CIN. It has been reported that KLRG1 is widely used as a differentiation marker, and the expression of KLRG1 increases with differentiation. The highest percentage of KLRG1 expression is observed on highly differentiated end-stage cells [
51], which may support our experimental results. We observed that HPV16(+) CD56
brightNK cells in CIN II–III displayed abnormal increase in number and decreased function, implying more poorly differentiation of CD56
brightNK cells compared to HPV16(+) low-grade group. Cervical histochemical results showed that the expression of its ligands E, N-cadherin in CIN II–III was significantly increased, and E-Cadherin was also increased in CIN I. It may be noted that even though the expression of E-cadherin is reduced in most tumor studies to increase cell invasiveness, correlation studies showed that some tumors may paradoxically use KLRG1-cadherin to evade immune surveillance. For example, in invasive breast cancer, E-cadherin was down-regulated initially by some tumor cells to acquire metastatic potential, and then it was re-expressed to promote adhesion and evade immune attack [
52]. It is still noteworthy that the KLRG1 expression on circulating CD56
dim NK cells was increased significantly, which needs to be investigated in future experiments to clarify the effect of CD56
dim NK cells on HPV16(+) CIN.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.