PD-1 and PD-L1 Expression Levels as a Potential Biomarker of Chronic Rhinosinusitis and Head and Neck Cancers

Inflammation is an etiological factor of various chronic diseases contributing to more than 50% of worldwide deaths. In this study, we focus on the immunosuppressive role of the programmed death-1 (PD-1) receptor and its ligand (PD-L1) in inflammatory-related diseases, including chronic rhinosinusitis and head and neck cancers. The study included 304 participants. Of this number, 162 patients had chronic rhinosinusitis with nasal polyps (CRSwNP), 40 patients had head and neck cancer (HNC) and there were 102 healthy subjects. The expression level of the PD-1 and PD-L1 genes in the tissues of the study groups was measured by qPCR and Western blot methods. The associations between the age of the patients and the extent of disease and genes’ expression were evaluated. The study showed a significantly higher mRNA expression of PD-1 and PD-L1 in the tissues of both the CRSwNP and HNC patient groups compared to the healthy group. The severity of CRSwNP significantly correlated with the mRNA expression of PD-1 and PD-L1. Similarly, the age of the NHC patients influenced PD-L1 expression. In addition, a significantly higher level of PD-L1 protein was noticed also for both the CRSwNP and HNC patient groups. The increased expression of PD-1 and PD-L1 may be a potential biomarker of inflammatory-related diseases, including chronic rhinosinusitis and head and neck cancers.


Introduction
According to statistics, more than 50% of worldwide deaths are related to inflammatory diseases. Inflammation is an important process involving the induction of various cells and mediators. That mechanism protects the human body against numerous harmful biological, chemical or physical stimuli. Normally, inflammatory response is characterized by the temporal upregulation of inflammatory activity that resolves after dangers have subsided and is followed by the restoration of cellular homeostasis. However, inflammatory response shifts from short-lived to chronic may lead to alterations in all tissues. Furthermore, impaired immune response increases susceptibility to infection and even tumor. There are numerous inflammation-related disorders, and they include chronic rhinosinusitis but also head and neck cancers [1][2][3][4].
Rhinosinusitis (RS) is defined as inflammation of the mucosal membranes in the paranasal sinuses and nasal cavity. According to the European Position Paper on Rhinosinusitis and Nasal Polyps (EPOS), published in 2020, RS is classified into acute rhinosinusitis (ARS) and chronic rhinosinusitis (CRS) [5]. CRS is characterized by the presence of at least two out of four cardinal symptoms (i.e., facial pain/pressure, hyposmia/anosmia, nasal drainage and nasal obstruction) for at least 12 consecutive weeks, in addition to

Materials
The following tissue samples were obtained from each study group: Group I-fragment of the mucosa from the lower nasal concha, collected during endoscopic septoplasty; Group II-polyp tissue of paranasal sinus collected during FESS; Group III-tumor tissue collected during surgical procedures. Immediately after collection, the tissue fragments were placed in Eppendorf tubes with 1 mL of RNAlater fluid (Qiagen, Hilden, Germany), preventing RNA degradation, and frozen at −20 • C. The expressions of the PD-1 and PD-L1 genes were evaluated in the materials.

Total RNA Extraction and cDNA Generation
Total RNA extractions were carried out using RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The RNA content and purity were measured using a PicoDrop spectrophotometer (Picodrop Limited, Hinxton, UK). The quality of the RNA samples was analyzed by measuring the ratio of absorptions at 260/280 nm. The purified total RNA was immediately used for cDNA synthesis or stored at −80 • C.
cDNA was generated with a Maxima First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's protocol. One milligram of total RNA was used as starting material; reverse transcription was performed in conditions optimized for use with this kit (25 • C for 10 min, 50 • C for 30 min and 85 • C for 5 min). The cDNA samples were kept frozen at −20 • C.

Real-Time PCR
The TaqMan gene expression experiments were performed in 10 µL reactions including 50 ng cDNA, 5 µL TaqMan Universal PCR Master Mix and 0.5 µL appropriate TaqMan Gene Expression Assay (Thermo Fisher Scientific, Waltham, MA, USA). The specific Taq-Man assays used in this study were programmed cell death 1 (PDCD1, Hs01550088_m1), programmed cell death 1 ligand 1 (PDCD1LG1, Hs01125301_m1) and β-actin (ACTB, Hs01060665_g1) as the endogenous control. The TaqMan PCR assays were performed on a 7900HT Fast Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA, USA) in FastGene Fast 96-well PCR plates (Nippon Genetics Europe GmbH, Düren, Germany). The following thermal cycling specifications were performed: 20 s at 95 • C and 40 cycles each for 3 s at 95 • C and 30 s at 60 • C. The ∆∆CT method was used [36].

Western Blot Analysis
Fifty milligrams of proteins were resolved on SDS electrophoresis gels with the use of standard procedures [37]. Immunoblotting was performed with anti-PD-1 rabbit antibody, PRS4065 (Merck, Darmstadt, Germany) or with anti-PD-L1 rabbit antibody, PRS4059 (Merck, Darmstadt, Germany). Final detection was achieved with the appropriate secondary antibody coupled to horseradish peroxidase (HRP), A0545 and anti-rabbit IgG-HRP (Merck, Darmstadt, Germany), followed by luminol-based chemiluminescent substrate (Cyanagen, Bologna, Italy) and visualized by a gel imaging system for chemiluminescence G: Box Chemi XR5 (Syngene, Cambridge, UK).
Equal loading of the proteins was additionally checked by the immunodetection of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mouse monoclonal antibody, MAB 374 (Merck Millipore, Burlington, MA, USA) and anti-mouse IgG-HRP, with A4416 as a secondary antibody (Sigma-Aldrich, Saint Louis, MO, USA), and visualized by G: Box Chemi XR5 (Syngene, Cambridge, UK). For the analysis of the protein band density, ImageJ 1.53k software (Wayne Rasband, USA) was used.

Statistics
The results are expressed as mean values ± standard deviation (SD). The differences between the samples were tested using one-way analysis of variance (ANOVA) with Tukey's post hoc test after confirming the normality of the distribution and the homogeneity of variances. The normality of distribution was appraised using the Shapiro-Wilk test. The homogeneity of the variances was tested using Levene's test.
The Spearman correlation coefficient (r) was estimated to determine the linear association between the following variables: patient's age and mRNA expression of the PD-1 and PD-L1 genes, extent of the disease and the mRNA expression of the PD-1 and PD-L1 genes. The outcome results were interpreted according to the degree of association as strong (r = 0.7-1), moderate (r = 0.5-0.7), or low (r = 0.3-0.5) after taking significant correlation (p < 0.05) values into consideration.
A multiple linear regression model was implemented to predict the values of the dependent variables (i.e., mRNA expression level of PD-1 and PD-L1 genes) starting from knowledge of the independent variables (i.e., age and extent of the diseases). The equation for a multiple linear regression is: where y is the mRNA expression level of the PD-1 and PD-L1 genes, β 0 is the intercept value, x i are two independent variables, β i are the estimated regression coefficients of the respective independent variables, and ε is the model error, i.e., the variation in our prediction of y with respect to the real value.
A level of p < 0.05 was considered statistically significant. All statistical computations were carried out in Statistica 13.0 software (StatSoft Polska, Cracow, Poland).

mRNA Expression Level of the PD-1 and PD-L1 Genes in Tissues from CRSwNP and NHC Patients
There were significant differences in the PD-1 and PD-L1 (p < 0.001) mRNA expression levels between polyp and normal tissues, with a 4.2-fold and 2.3-fold increase in the polyp tissues, respectively. Moreover, significant differences in the PD-1 and PD-L1 (p < 0.01) mRNA expression levels between tumor and normal tissues, with a 4.5-fold and 4.6-fold increase in the tumor tissues, were found, respectively. These data are presented in Figure 1.
The results are expressed as mean values ± standard deviation (SD). The differences between the samples were tested using one-way analysis of variance (ANOVA) with Tukey's post hoc test after confirming the normality of the distribution and the homogeneity of variances. The normality of distribution was appraised using the Shapiro-Wilk test. The homogeneity of the variances was tested using Levene's test.
The Spearman correlation coefficient (r) was estimated to determine the linear association between the following variables: patient's age and mRNA expression of the PD-1 and PD-L1 genes, extent of the disease and the mRNA expression of the PD-1 and PD-L1 genes. The outcome results were interpreted according to the degree of association as strong (r = 0.7-1), moderate (r = 0.5-0.7), or low (r = 0.3-0.5) after taking significant correlation (p < 0.05) values into consideration.
A multiple linear regression model was implemented to predict the values of the dependent variables (i.e., mRNA expression level of PD-1 and PD-L1 genes) starting from knowledge of the independent variables (i.e., age and extent of the diseases). The equation for a multiple linear regression is: where y is the mRNA expression level of the PD-1 and PD-L1 genes, β0 is the intercept value, xi are two independent variables, βi are the estimated regression coefficients of the respective independent variables, and ε is the model error, i.e., the variation in our prediction of y with respect to the real value.
A level of p < 0.05 was considered statistically significant. All statistical computations were carried out in Statistica 13.0 software (StatSoft Polska, Cracow, Poland).

mRNA Expression Level of the PD-1 and PD-L1 Genes in Tissues from CRSwNP and NHC Patients
There were significant differences in the PD-1 and PD-L1 (p < 0.001) mRNA expression levels between polyp and normal tissues, with a 4.2-fold and 2.3-fold increase in the polyp tissues, respectively. Moreover, significant differences in the PD-1 and PD-L1 (p < 00.1) mRNA expression levels between tumor and normal tissues, with a 4.5-fold and 4.6fold increase in the tumor tissues, were found, respectively. These data are presented in Figure 1.

mRNA Expression Level of the PD-1 and PD-L1 Genes in the Different Subgroups of CRSwNP and HNC Patients
The PD-1 and PD-L1 gene expression was evaluated in the different subgroups of the CRSwNP patients, including subjects with chronic rhinosinusitis with nasal polyps, chronic rhinosinusitis with nasal polyp and allergy and chronic rhinosinusitis with nonsteroidal anti-inflammatory drug (NSAID)-exacerbated respiratory disease (N-ERD). In addition, PD-1 and PD-L1 gene expression in nonsmoking and smoking groups of HNC patients was also assessed. The data is presented in Table 2. In patients with CRSwNP and coexisting allergy, the mRNA level of PD-1 in the polyp tissue was significantly elevated in comparison to that observed in the polyp tissue collected from patients with CRSwNP alone (p < 0.001).
In smoking patients with HNC, the mRNA level of PD-L1 in the tumor tissue was significantly elevated in comparison to that observed in tumor tissue collected from nonsmoking HNC patients (p = 0.037).

Correlation between the mRNA Expression Level of the PD-1 and PD-L1 Genes and the Ages of the CRSwNP and HNC Patients
In patients with CRSwNP, the mRNA levels of both the PD-1 and PD-L1 genes in the polyp tissue were not correlated with the disease severity calculated using the Kennedy scores. Conversely, in patients with HNC, the mRNA level of the PD-L1 gene in the tumor tissue were correlated with the age of patients (p = 0.048). These data are presented in Figure 2.

Correlation between the mRNA Expression Level of the PD-1 and PD-L1 Genes and the Severity of the CRSwNP and HNC Diseases
In patients with CRSwNP, the mRNA levels of both the PD-1 and PD-L1 genes (p < 0.001) in the polyp tissue were correlated with the disease severity calculated using the Kennedy scores. Conversely, a similar correlation was not observed for patients with HNC calculated using the TNM staging system. These data are presented in Figure 3.

Correlation between the mRNA Expression Level of the PD-1 and PD-L1 Genes and the Severity of the CRSwNP and HNC Diseases
In patients with CRSwNP, the mRNA levels of both the PD-1 and PD-L1 genes (p < 0.001) in the polyp tissue were correlated with the disease severity calculated using the Kennedy scores. Conversely, a similar correlation was not observed for patients with HNC calculated using the TNM staging system. These data are presented in Figure 3.

Correlation between the mRNA Expression Level of the PD-1 and PD-L1 Genes and the Severity of the CRSwNP and HNC Diseases
In patients with CRSwNP, the mRNA levels of both the PD-1 and PD-L1 genes (p < 0.001) in the polyp tissue were correlated with the disease severity calculated using the Kennedy scores. Conversely, a similar correlation was not observed for patients with HNC calculated using the TNM staging system. These data are presented in Figure 3.   Table 3 shows the multiple linear regression analysis. According to the model investigated, a variable Kennedy scale demonstrated a significant contribution to the mRNA expression level of the PD-L1 gene in the polyp tissue from CRSwNP patients. Moreover, the variable age demonstrated a significant contribution to the mRNA expression level of the PD-L1 gene in the tumor tissue from HNC patients. Table 3. Multiple linear regression analysis for age and extent of the diseases on the mRNA expression level of the PD-1 and PD-L1 genes.

PD-1 and PD-L1 Protein Expression Level Analysis
An increased level of PD-L1 protein in the tissues from CRSwNP and HNC patients in comparison to the control subjects was confirmed. These data are presented in Figure 4.  Table 3 shows the multiple linear regression analysis. According to the model investigated, a variable Kennedy scale demonstrated a significant contribution to the mRNA expression level of the PD-L1 gene in the polyp tissue from CRSwNP patients. Moreover, the variable age demonstrated a significant contribution to the mRNA expression level of the PD-L1 gene in the tumor tissue from HNC patients.

PD-1 and PD-L1 Protein Expression Level Analysis
An increased level of PD-L1 protein in the tissues from CRSwNP and HNC patients in comparison to the control subjects was confirmed. These data are presented in Figure 4.   Significant differences in the PD-L1 (p < 0.01) protein level between the polyp and normal tissues, with a 4.9-fold increase in the polyp tissues, were observed. Moreover, significant differences in the PD-L1 (p < 00.1) protein level between tumor and normal tissues, with a 5.7-fold increase in the tumor tissues, were found. The PD-1 protein level in both polyp and cancer tissues was also slightly increased in comparison to the control tissue, but the increase was not statistically significant.

Discussion
In our study, we investigated PD-1 and PD-L1 expression patterns in the tissues of CRSwNP patients. The data show increased mRNA and protein levels of PD-L1 in polyp tissues from patients with CRSwNP compared to the control group. The expression of PD-1 and its ligand PD-L1 in CRSwNP patients is poorly investigated. It is indicated that PD-1 expression is elevated in tissue infiltrating the T cells of CRSwNP patients [38]. Shi et al. have shown that the mRNA level of PD-L1 significantly was increased in eosinophilic CR-SwNP [39]. Furthermore, Krohn et al. have demonstrated an increased mRNA expression of PD-1 in tissue homogenates from patients with CRSwNP compared to controls. Moreover, the mRNA level was associated with disease severity calculated using Lund-Mackay scores on CT scans [40]. According to our data, the severity of the CRSwNP calculated using Kennedy scores of CT scans was significantly correlated with the expression of both the PD-1 and PD-L1 genes.
The PD-1/PD-L1 axis was mainly explored in cancer cells. Various signaling pathways upregulate PD-1/PD-L1 expression, including the phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway, mitogen-activated protein kinase (MAPK) pathway, Janus kinase/signal transducers and activators of transcription (JAK/STAT) pathway, WNT signaling pathway and nuclear factor kappa light-chain enhancer of activated B cells (NF-κB) pathway. The induction of the PI3K/AKT pathway can promote PD-L1 expression. Increased extrinsic signaling, as well as a decreased level of negative regulators, such as phosphatase and tensin homolog (PTEN), may occur [41,42]. In addition, the data indicate that the inhibition of the MAPK pathway may markedly prevent the expression of PD-L1 [43,44]. Moreover, the results confirm that the JAK/STAT pathway stimulates PD-L1 expression in cancers [45,46]. Other studies revealed the presence of WNT activity and PD-L1 expression crosstalk [47,48]. Additionally, a large body of evidence indicates that the induction of NF-κB signals promotes PD-L1 expression [49,50]. Actually, PD-L1 expression is upregulated by different molecular mechanisms, including inflammatory signaling, but also by genetic modifications, oncogenic pathways, microRNA or protein-level regulation [51].
PD-1/PD-L1 inhibit the activity of T cells and prevent effector immune cells from killing cancer cells. PD-1/PD-LI inhibitors suppress the PD-1/PD-L1 pathway to restore the normal immune function of T cells. Commonly administered PD-1/PD-L1 inhibitors in the treatment of cancers include, among others, nivolumab, pembrolizumab, atezolizumab, durvalumab and avelumab [52,53]. However, only a subset of PD-L1-positive patients benefits from PD-1/PD-L1 immune therapies. The immunotherapy resistance mechanisms are as follows: changes in T-cell function or proliferation, alternative immunological checkpoints, tumor cell adaptation and alterations of the tumor microenvironment. Therefore, understanding the mechanisms of PD-L1 regulation is helpful to enhance the treatment effect [51,[54][55][56][57].
On the one hand, studies indicate that the PD-L1 protein level in tumor cells is positively correlated with a higher tumor grade in breast cancer patients [58] and in squamous cell carcinoma of vulva patients [59]. In addition, upregulated PD-1 protein expression in T cells may be correlated with immune evasion in gastric cancer patients [60]. Chemotherapy combined with immunotherapy had better overall survival (OS) in nonsmall cell lung cancer patients regardless of the PD-L1 expression level [61]. However, a very high PD-L1 expression was associated with an OS benefit in patients with advanced non-small cell lung cancer receiving PD-L1 inhibitor monotherapy [62]. A study conducted by Taube et al. highlights a therapeutic opportunity in administering factors inhibiting the PD-1 pathway to patients with metastatic melanoma [63]. On the other hand, the PD-L1 expression of tumor cells was associated with improved OS in pulmonary squamous cell carcinomas [64]. A similar trend was also observed for patients with colorectal [65] and ovarian [66] cancers, as well as esophageal squamous cell carcinoma [67]. At present, it is not known why PD-L1 expression is associated with a favorable prognosis in certain cancers and a negative prognosis in others. The prognostic impact of this marker is based on incompletely elucidated mechanisms occurring in the tumor microenvironment.
In our study, we investigated PD-1 and PD-L1 expression patterns in the tissues of HNC patients. The research shows increased mRNA and protein levels of PD-L1 in tumor tissues obtained from patients with HNC compared to the control group. Moreover, a statistically significant higher mRNA expression of PD-1 was observed in patients with CRSwNP and allergy compared to CRSwNP alone, whereas a higher mRNA expression of PD-L1 was observed in smoking patients compared to nonsmoking. Melliante et al. point out that immunotherapy has a higher efficacy in smoking patients. This may be attributed to an increase in the number of genetic mutations, which leads to greater immunogenicity [51]. According to our data, the ages of the HNC patients correlated with the expression of the PD-L1 gene. There was no correlation between TNM staging and the expression of the PD-1 and PD-L1 genes. However, it has been suggested that PD-L1 mRNA expression may be associated with increased local lymphocytic infiltration and OS [68].
Our research supports the role of the PD-1/PD-L1 axis as an immune modulator in patients with HNC. Lyu et al. demonstrated that PD-L1 expression is upregulated in the tumor tissue of HNCs. PD-1/PD-L1 expression was significantly associated with radiosensitivity, and high PD-1 expression was strongly related to HPV/p16-positive HNCs [69]. On the one hand, Lin et al. found that a higher level of PD-L1 in HNC tissues was associated with distant metastases and worse outcomes [70]. Similarly, Moratin et al. concluded that elevated PD-L1 expression in human oral squamous cell carcinomas was correlated with tumor size, stage, regional metastases and worse OS [71]. On the other hand, Müller et al. found that PD-L1 expression in tumor tissue is correlated with better OS [72]. Other reports show that higher PD-1 and PD-L1 expression was associated with a better outcome with fewer local and distant recurrences, mainly in HPV-positive patients [73,74]. Hong et al. indicate that HPV-and PD-L1-positive patients with tonsillar cancer had a better OS, longer progression-free survival and decreased risk of death [75]. Moreover, it was indicated that the therapeutic response of HNC patients with an elevated expression of PD-L1 to immunotherapy was better [76]. Chen et al. reported that HNC patients with elevated PD-L1/PD-1 expression tend to have better OS outcomes and a lower probability of recurrence, providing more evidence for the PD-1-targeted immunotherapy of HNCs [77]. On the contrary, Kim et al. indicated that tumor cell expression of PD-L1 is not associated with clinical characteristics and the prognosis of HNC patients [78]. Therefore, the current results are still controversial.
According to the hypothesis, the studies showed that the level of expression of PD-1 and PD-L1 is significantly higher in both patients with NHC and CRSwNP compared to controls. However, our study's limitation is the small sample size; therefore, further investigation is needed. Elucidating the role of the PD-1/PD-L1 axis may, in the future, facilitate the implementation of effective treatments for immune-related diseases, such as CRS and HNC [79,80].

Conclusions
Immune-related diseases are becoming common both in male and female patients. Searching for new potential biomarkers for the early detection of these diseases is extremely important, as it will allow to immediately monitor their development. Our studies showed an increased expression of the PD-1 and PD-L1 genes, which may be a potential biomarker of chronic sinusitis with nasal polyps.