HBD3 Induces PD-L1 Expression on Head and Neck Squamous Cell Carcinoma Cell Lines

Human β-defensin 3 (HBD3) is an antimicrobial peptide up-regulated in the oral tissues of individuals with head and neck squamous cell carcinomas (HNSCC) and oral squamous cell carcinomas (SCC) and present in high concentrations in their saliva. In this study, we determined if HBD3 contributes to HNSCC pathogenesis by inducing programmed death-ligand 1 (PD-L1) expression on HNSCC cell lines. For this, SCC cell lines SCC4, SCC15, SCC19, SCC25, and SCC99 (5.0 × 104 viable cells) were used. Cells were incubated with IFNγ (0.6 µM) and HBD3 (0.2, 2.0, or 20.0 µM) for 24 h. Cells alone served as controls. Cells were then treated with anti-human APC-CD274 (PD-L1) and Live/Dead Fixable Green Dead Cell Stain. Cells treated with an isotype antibody and cells alone served as controls. All cell suspensions were analyzed in a LSR II Violet Flow Cytometer. Cytometric data was analyzed using FlowJo software. Treatment with IFNγ (0.6 µM) increased the number of cells expressing PD-L1 (p < 0.05) with respect to controls. Treatment with HBD3 (20.0 µM) also increased the number of cells expressing PD-L1 (p < 0.05) with respect to controls. However, treatment with IFNγ (0.6 µM) was not significantly different from treatment with HBD3 (20.0 µM) and the numbers of cells expressing PD-L1 were similar (p = 1). Thus, HBD3 increases the number of cells expressing PD-L1. This is a novel concept, but the role HBD3 contributes to HNSCC pathogenesis by inducing PD-L1 expression in tumors will have to be determined.


Introduction
Head and neck squamous cell carcinomas (HNSCC) and oral squamous cell carcinomas (SCC) are neoplasms of oral tissues. Their onset and associated mutational profiles are often associated with prior alcohol consumption, tobacco use, and human papillomavirus infections [1][2][3]. In the most recent worldwide study in 2012, there were~300,000 cases of HNSCC and SCC (2.1% of total cancer cases in the world, the sixth most common cancer by incidence worldwide) and~145,000 associated deaths (1.8% of the world total) [4]. In the USA, there were 53,000 estimated new cases of HNSCC and SCC in 2019 (3.0% of new cancer cases) and 10,860 estimated deaths in 2019 (1.8% of all cancer deaths) [5]. The 5 years survival rate is 65.3% [5]. Only modest improvements in the survival rate are seen with chemoradiation, surgical resection, reconstructive methods, and biological treatments [6]. HNSCCs arise through the accumulation of genetic and epigenetic changes in genes acting in cancer-associated signaling pathways [1,7,8]. HNSCC cells produce a variety of immunosuppressive cytokines, chemokines, and biomarkers [9,10]. Among these is programmed death-ligand 1 (PD-L1) [10].
PD-L1 is a 33.28 kDa type I transmembrane protein expressed on the surface of immune and non-immune cells [11][12][13]. It is a co-inhibitory immune checkpoint protein that binds to the programmed death-1 (PD-1) receptor on T-cells [14]. The interaction of PD-L1 with PD-1 regulates the balance between co-stimulatory and co-inhibitory immune signals, maintains the breadth and magnitude of the immune response, maintains self-tolerance, prevents adverse autoimmune inflammatory events, protects the host from uncontrolled immune responses to pathogens, and prevents inflammatory tissue damage. Increases in PD-L1 expression can occur on SCC cells [8] as a result of mutations in tumor cell signaling pathways or exposure of tumor cells to inflammatory cytokines IL-1, IL-6, GM-CSF, IFNγ, TNFα, and VEGF [15][16][17][18][19] and the gamma-chain cytokines IL-2, IL-7, IL-10, IL-15, and IL-21 [20]. The latter group plays a role in peripheral T-cell expansion and survival. The presence of PD-L1 affects T-cell responsiveness in a quantitative manner [21]. A high level of PD-L1 expression increases impairment of T-cell survival and activity. Thus, a high level of PD-L1 expression on the surface of tumor cells inhibits the activation, expansion, and effector functions of T-cells [22][23][24] and helps SCC cells evade normal anti-tumor immune mechanisms.
HBD3 is a potent host defense peptide [25]. It has 45 amino acid residues and a monoisotopic mass of 5157.7 Da [26]. Its lysine and arginine residues gives it a strong positive charge [27]. HBD3 is expressed in mucosal epithelial cells and keratinocytes, including cells and tissues in the oral cavity [28,29]. It is present in gingival crevicular fluid and saliva.
HBD3 enhances adaptive immunity [40]. It interacts with G protein-coupled receptors on immature dendritic cells, particularly CCR6 and induces NK cell activation, IFNγ secretion, and mature dendritic cell dependent cytolytic function [41].
In this study, we used SCC cell lines (Table 1). We report that HBD3 induces the expression of PD-L1 and this is a novel concept. However, to what extent HBD3 contributes to HNSCC and SCC pathogenesis by inducing PD-L1 expression in tumors will have to be determined. Cell lines were derived from HNSCC and SCC malignant tumors [19,42]. (TNM, where T describes the size of the primary tumor, N describes the nearby lymph nodes, and M describes the metastasis).

Results
Bates et al. found that PD-L1 was present in differing concentrations on the surface of HNSCC cell lines in Table 1 by ELISA, and we summarized this information in Table 2 [10]. PD-L1 concentrations in cell lysates ranged from 79.67 to 539.79 pg/mL. In this study, we also found that PD-L1 is present on the surface of HNSCC cell lines by flow cytometry (Figure 1). The percent of SCC4, SCC15, SCC19, SCC25, and SCC99 cells with PD-L1 expression (e.g., no-treatment control group) ranged from 1.79 to 89.40% (Table 3). Values (pg/mL) were from our recent work used as the basis for the article by Bates et al. [10].

Discussion.
PD-L1 is an important immune checkpoint molecule in cancer pathogenesis regulating both tumor-intrinsic signaling and adaptive immunosuppression. It is induced by inflammatory and gamma-chain cytokines [15][16][17]20]. However, little is known about other factors in the oral cavity that also influence PD-L1 expression in SCC. Here we demonstrate that HBD3, a host defense peptide with diverse innate immune activities [29], is one of these factors. HBD3 is present in nasal mucus,  The percent of cells with PD-L1 expression increased when treated with 0.6 µM IFNγ, which was used as a positive control, compared to the untreated controls. The percent increase in the staining of cells (with respect to the no-treatment controls) varied from 5.8% for SCC19 to 657.5% for SCC15 (Table 3).
Similarly, the percent of cells with PD-L1 expression increased when treated with 0.2, 2.0, or 20.0 µM HBD3 compared to the untreated controls. At 20.0 µM HBD3, the percent increase in the staining of cells (with respect to the no-treatment controls) varied from 7.9% for SCC19 to 109.0% for SCC15 (Table 3). HBD3 had a dose dependent increase in PD-L1 expression. The percent of SCC99 cells with PD-L1 expression increased with increasing concentrations of HBD3 (Figure 1g).
We performed Wilcoxon signed rank tests to compare differences between each of the treatment groups and the control group, as well as differences between the two treatment groups (Table 3). Treatment A (20 µM HBD3) was significantly elevated compared to the control group: p = 0.03 (one-sided test). Treatment B (0.6 µM IFNγ) was also significantly elevated compared to the control group: p = 0.03 (one-sided test). However, no significant difference was shown between Treatment A (20 µM HBD3) and Treatment B (0.6 µM IFNγ) (two-sided p = 1). Multiple comparisons were not adjusted.

Discussion
PD-L1 is an important immune checkpoint molecule in cancer pathogenesis regulating both tumor-intrinsic signaling and adaptive immunosuppression. It is induced by inflammatory and gamma-chain cytokines [15][16][17]20]. However, little is known about other factors in the oral cavity that also influence PD-L1 expression in SCC. Here we demonstrate that HBD3, a host defense peptide with diverse innate immune activities [29], is one of these factors. HBD3 is present in nasal mucus, saliva, and gingival crevicular fluid at concentrations as high as 6.2 mg/mL [43]. However, it is also present in abnormally high concentrations in the saliva of patients with oral cancers [44,45] likely associated with the intense inflammation associated with oral cancer in these individuals [44]. In our work, we have shown that HBD3 can bidirectionally regulate chemokine and cytokine responses [36,37]. Therefore, it is conceivable that elevated concentrations of HBD3, like that in oral cancer [44,45] can induce PD-L1 expression in SCC cells. To what extent it is involved in enhancing the suppression of tumor-specific T-cell activity and contributes to an immunosuppressive environment in the tissues of the oral cavity will have to be determined.
Recently, Ghosh and colleagues assessed the dysregulation of HBD3 in oral squamous cell carcinoma (OSCC) [46]. They noted that HBD3 is up regulated in OSCC tissues with respect to healthy oral mucosa [44] and is expressed in the cytoplasm of OSCC cells [47]. In HPV infected oral epithelial cells, oncogene E6 increases HBD3 mRNA and peptide expression and tumor suppressor p53 is inhibited by E6 and blocks HBD3 expression [45,46,48].
HBD3 regulates many pathways including those that likely influence PD-L1 expression. Also in our work, we found that HBD3 at low concentrations attenuates pro-inflammatory agonist-induced chemokine and pro-inflammatory cytokine responses and at high concentrations, enhances agonist-induced chemokine and pro-inflammatory cytokine responses [36,37]. HBD3 binds to G-protein coupled receptor CCR6, which stimulates Gi and Gq respectively and activates the LYN/SYK/PLC/PKC-DAG pathway converging at the activation of AKT, NFKB, and NFAT. Future studies will determine whether HBD3-induced PD-L1 expression and IFNγ-induced PD-L1 expression share similar pathways.

PD-L1 Inducers
Human β defensin-3 (HBD3, catalog no. 300-52) was purchased from Peprotech, Rocky Hill, NJ, USA. The concentrations of HBD3 (0.2, 2.0, or 20.0 µM) used in this study were similar to those concentrations used in our previous studies [36] and those concentrations used in similar studies by others [53].
Recombinant human IFNγ (catalog no. 300-02) was purchased from Peprotech, Rocky Hill, NJ, USA. IFNγ is a well-known PD-L1 inducer [15] and was used as a positive control.

Immunohistochemistry
A double-sandwich ELISA (MyBioSource, Inc., San Diego, CA, USA) was used to detect human papillomavirus (HPV) antigen tissue culture media and cell lysates as previously described [10].
Immunohistochemistry (IHC) was used to detect p16 Ink4a in cell lines as previously described [10]. Cell pellets were fixed in 10% neutral buffered formalin, immobilized in agar, embedded in paraffin, and sectioned. Sections were deparaffinized and stained with an antibody to p16 Ink4a . Squamous cell carcinoma of the uterus was used as a positive tissue control.

Induction of PD-L1 Expression
SCC cells (5.0 × 10 4 viable cells) were incubated without and with 0.6 µM IFNγ or 0.2, 2.0, or 20.0 µM HBD3 for 24 h. The number of cells expressing PD-L1 was determined by flow cytometry as described below in our laboratory [19].

Detection of Cells Expressing PD-L1 by Flow Cytometry
Cells were first stained with Live/Dead Fixable Green Dead Cell Stain (BD Biosciences, San Jose, CA, USA), then stained with anti-human APC-CD274 (563741 PD-L1, BD Pharmingen, San Jose, CA, USA), and then examined using an LSR II Violet Flow Cytometer (BD Biosciences). Cells stained with an isotype control (APC mouse IgG1 κ, BD Pharmingen, San Jose, CA, USA) served as a control to account for any non-specific binding. Cells not stained were also included as controls. Flow cytometric data was analyzed using FlowJo software (Tree Star, Inc., Ashland, OR) [19].

Statistical Analysis
Wilcoxon signed rank tests were performed to compare between each of the treatment groups and the control group, as well as between the two treatment groups. Treatment A was 20.0 µM HBD3 was and Treatment B was 0.6 µM IFNγ. Multiple comparisons were not adjusted.

Conclusions
In conclusion, HBD3 is thought to have a role in the pathogenesis of oral cancers [44][45][46]. Here we show that HBD3 increases the number of HNSCC cells expressing PD-L1. However, to what extent HBD3 contributes to HNSCC pathogenesis by inducing PD-L1 expression in tumors will have to be determined. Funding: This research was funded by the National Institutes of Health grants NIH, NIDCR R01 DE014390 and NIH, NIDCR T90 DE023520.