Methylation level of ASC/TMS1 and MyD88 genes in healthy larynx and laryngeal carcinoma tissue

There are no biomarkers for diagnosis, prognosis and treatment of patients with laryngeal carcinoma. Methylation changes of ASC/TMS1 and MyD88 genes, in healthy and cancer tissue, might be related with development and progression of cancer. The study explored is there a difference in gene’s methylation in healthy and tumor tissue and does it correlate with protein expression. The total of 36 patients were enrolled in the study. Methylation of bisulphite converted DNA was quantied by pyrosequencing in fresh frozen cancer and adjacent non-malignant tissues. The overall methylation of MyD88 gene is signicantly higher in healthy tissue and this nding correlates with protein expression and the overall methylation of ASC/TMS1 gene is unchanged but the protein expression of ASC/TMS1 is signicantly higher in cancer. The methylation status of the ASC/TMS1 and MyD88 genes are promising prognostic biomarker candidates and may lead to earlier detection of laryngeal cancer.


Introduction
Laryngeal cancer (LC) is a second most common cancer of Head and Neck Cancer (HNC) group which develops from laryngeal squamous cell epithelium. It comprises 40% of all HNC and 1-2% of all human malignancies [1]. The most important risk factors in LC development are tobacco smoking and excessive alcohol consumtion comprising very important synergistic effect [2,3]. In past 20 years there is a slight decrease in the overall incidence of LC due to global anti-smoking compaign especially in USA and European countries. Unfortunatelly it has no signi cant in uence on overall patient survival which is not markedly improved as there is still a great number of local and regional recurrences, advanced metastases and secondary primary tumors [4]. Patient prognosis depends mostly on staging status at presentation [5][6][7]. Last few decades it became obvious that chronic in ammation predisposes to different forms of cancer. Pathogens as bacteria and viruses activate innate immune mechanisms which enables their elimination from the organism [8]. Some pathogens can change innate immune mechanisms and lead to the chronic in ammation. Chronic in ammation has protumorigenic effect and enhances the risk of tumorigenesis [9]. Important examples are H. pylori infection and development of gastric carcinoma [10], Hepatitis B and C virus infection and hepatocellular carcinoma [11], Human papilloma virus in development of oropharyngeal carcinoma [12,13]. The pro le of expression of different molecules engaged in activation and modulation of immune response and in ammation, their enhanced accumulation and decreased activation determine the balance between protumorigenic and antitumorigenic microenvironment [8]. Cytokines control the in ammation through activation of different signal pathways by inducing protumorigenic immunity (IL -12, TRAIL, IFNγ), tumor development (IL-6, IL-17, IL-23), tumor growth and survival (TRAIL, FasL, TNF-α, EGFR, TGF-β). Generally, we can say that carcinoma is result of the change in signal pathways which participate in the cell proliferation regulation. Key factors in relationship between chronic in ammation and carcinogensis are transcription factors (NF-κB i STAT3) and proin ammation cytokines (IL-1β, IL-6, IL-23 i TNF-α) [14].
Innate immunity mechanism is the rst line of organism defence and it is regulated by germline-encoded pattern recognition receptors (PRRs) expressed by cells, including macrophages, monocytes, dendritic cells, neutrophils and epithelial cells. The most studied PRRs are membrane-bound Toll-like receptors (TLRs) which recognize pathogen-associated molecular patterns (PAMPs) and intracellular NOD-like receptors (NLRs) which recognize PAMPs as well as danger-associated molecular patterns (DAMPs).
Activation of those receptors lead to complex signal pathway activation with resulting transcriptional factors NF-κB i AP-1 activation and synthesis and producing of proin ammatory cytokines and chemokines. Activation of NLRs results in in ammasomes formation which enables inactive to active caspase transformation and pro-IL-1β maturation. NLRs and TLRs pathways are connected so activation of TLRs pathway enhance producing of pro-IL-1β which mature by NLRs signaling pathway activation [15].
There are two genes, ASC/TMS1 (Apoptosis-associated speck like protein containing a caspase recruiting domain/Target of methylation induced silencing-1) gene and MyD88 (Myeloid differentiation primary response 88) gene, which both encode adaptor molecules with important roles in signal pathways activated by organism response to in ammation or injury.
Adaptor molecule ASC is involved in signal pathway characterized by formation of molecular platforms in ammasomes which activate NLRs and control maturation and secretion of proin ammatory cytokines interleukin-1β (IL-1β) and interleukin-18 (IL-18).
From all NLRs NLRP3 is currently the most fully characterized in creation of NLRP3 in ammasome. NLRP3 in ammasome consists of NLRP3 scaffold, the ASC adaptor molecule and caspase-1. It is activated upon exposure to speci c ligands like pathogens, toxins, crystals, PAMPs and DAMPs. Upon NLRP3 activation, NLRP3 oligomerization leads to the PYD domain clustering homotypic interaction with the PYD domain of adaptor molecule ASC, and interaction of the CARD domain of adaptor molecule ASC with CARD domain of the procaspase-1. Procaspase-1 permits autocleavage and formation of the active caspase-1 which leads to the maturation and secretion of interleukin-1β (IL-1β) [8, 15,16].
From the literature it is known that the mechanisms involved in NLRP3 activation are not well established and still intensely debated but mostly acceptable are three main models. First model explain that extracellular ATP stimulates opening of ATP-dependant ion channel of the cell membrane and in that way pore formation in the cell membrane allows direct NLRP3 in ammasome activation by NLRP3 cytosol agonists. Second model propose as activators particulate structures (urates, silica, asbestos) which activate phagocytes and by lysosomal damage release lysosomal contents consequently activating NLRP3 in ammasome formation. According to third model all NLRP3 agonists trigger the generation of reactive oxygen species (ROS) and this pathway activate NLRP3 in ammasome formation [17].
It is well known that some epigenetic mechanisms, like DNA methylation, can change gene expression by methylation of CpG islands in the regulatory regions of the genome. As those islands, rich with CG dinucleotides (> 55%) are situated in regulatory gene regions they affect the transcription of genes. If CpG islands are unmethylated genes are expressed and if they are hypermethylated the expression of the genes is repressed (gene silencing). The in ammation induce DNA methyltransferase (DNMT) dependent DNA methylation which subsequently leads to silencing of the genes associated with carcinogenesis [19].
In this study, we hypothesized that the changes in the methylation status of promoter regions of ASC/TMS1 and MyD88 genes, responsible for activation and regulation of in ammation in healthy and laryngeal cancer tissue, might be related with development and progression of the laryngeal cancer. The aim of our study is to investigate is there a difference in the methylation status of this genes in healthy and tumor tissue and does it correlate with protein expression.
In recent literature the methylation status of this genes has not yet been analyzed in laryngeal cancer and the pyrosequencing method used to determinate methylation status of this genes has not been used in other cancers studies.

Material And Methods
In this study, during routine surgical procedure in the treatment of laryngeal cancer, we have collected 36 samples of cancer tissue with related histologically healthy/not tumor involved tissue. All samples of cancer and healthy/not tumor involved tissue were analyzed histopathologically. Cancer tissue was analyzed to determine the value of cancer density and only cancer tissue with > 70% of tumor cells was used for further analysis. Adjacent healthy/not tumor involved tissue was analyzed to con rm lack of malignant cells. We started research after applying and receiving approval from Ethics Committee of University Hospital Center Zagreb and all procedures followed were in the accordance with the ethical standards of the Committee. The study was performed in the Department of Otorhinolaryngology and Head and Neck Surgery University Hospital Center Zagreb and all molecular researches were done in the Department for Molecular Medicine, Laboratory for Advanced Genomics Ruđer Bošković Institute in Zagreb. All patients signed informed consent prior to a surgical intervention.

Analysis of DNA Methylation by Pyrosequencing
The methylation status of both genes promoters was analyzed by pyrosequencing method which as quantitative analysis of DNA CpG methylation enables precise expression of percentage of methylated DNA in the analyzed sample tissue. This method offers a robust, versatile platform with rapid quantitative results without the onerous time commitment, high costs and technical di culty of alternative methods.
First step was isolation of genomic DNA of interest by using commercially available DNA isolation kit (Qiagen, Hilden, Germany; DNeasy Blood and Tissue Kit). Genomic DNA of interest was submitted to bisul te conversion in which treating genomic DNA with sodium bisul te selectively converts cytosine to uracil; however 5-methylcytosine was protected from deamination and the CG sequence was preserved in downstream reactions. Unmethylated cytosine through conversion by deamination into uracil was nally converted into thymine with epigenetic change in DNA sequence. Bisul te conversion was performed by using commercially available kit (Qiagen, Hilden, Germany; Epitect® 96 bisul te kit). DNA sequence for polymerase chain reaction (PCR) primers were acquired from free genetic sequence database GenBank (National Center for Biotechnology Information, NCBI). Identi cation of MyD88 gene in the database was NG_016964 and of ASC/TMS1 gene was AF184072.1. Primer designing started by determination of promoter DNA sequence of interest of ASC/TMS1 gene and MyD88 gene. After detecting region of interest (ROI) of both genes primers were designed. Primers for PCR of bisul te converted DNA and pyrosequencing were designed using PyroMark Assay Design Software 2 − 0 (Qiagen, Hilden; Germany).
Following pyrosequencing, pyrograms were analyzed by Pyro Q-CpG™ program (Qiagen, Hilden; Germany). This program can automatically determine individual methylation frequencies for all CpG islands as well as to evaluate the assays quality. Pyrograms which did not satisfy by quality were excluded from the study.
The results of the pyrosequencing of DNA sequence showed percentage of DNA methylation rate in analyzed sequence by using paired t-test (p-values < 0.05 were statistically signi cant) in GraphPad prism 5.0 program. Extraction of total proteins from laryngeal cancer tissue and healthy/not tumor involved tissue Proteins from cancer tissue and healthy/not tumor involved tissue were extracted by Passive Lysis 5x Buffer (Promega, Fitchburg, Wisconsin, USA). By Western Blotting Procedure particular proteins were identi ed from the mixture of protein extracted from the samples. By SDS-PAGE (Sodium dodecyl sulfate polyacrylamide gel electrophoresis) proteins were separated according their electrophoretic mobility.
Following separation, proteins were transferred from gel to nitrocellulose membrane using Trans Blot Turbo™ Transfer System (BIO-RAD, Hercules, California, USA). Finally, protein of interest were labeled by primary and secondary antibodies and the signal was detected by the method of chemiluminescence using reagent Clarity™ Western ECL Blotting Substrates (BIO-RAD, Hercules, California, USA). In order to quantify detected signal we used free ImageJ 1.49v software. The values of tested proteins (MyD88 and ASC/TMS1) were normalized with the values of control protein (vinkulin) from the adjacent gel. Normalized values were statistically analyzed in GraphPad Prism 5.0 program (GraphPad Software, San Diego, USA) by using 2-way analysis of variance (2way ANOVA test) and paired t -test. P-values less than 0,05 were regarded as statistically signi cant.

Results
The total of 36 patients participated in this study; 33 male patients and 3 female patients. All of them had squamous cell carcinoma (SCC) of the larynx with different grades of differentiation from grade I to grade III. As the most important variable for overall survival of those patients beside differentiation grade is tumor localization, tumor size, and existence or absence of tumor metastasis and in all patients the disease staging was de ned using UICC TNM (Union for International Cancer Control Tumor Node Metastasis) classi cation.
Results of analysis of MyD88 gene methylation status in healthy/not tumor involved and tumor tissue To accomplish differential methylation analysis of MyD88 gene in healthy/not tumor involved and tumor tissue, the methylation status of CpG islands of gene promoter region was performed. (Table 1) According to nucleotide sequence of DNA in MyD88 gene 3 regions of interest (ROI) were de ned. In region of interest 1 (ROI_1) we detected 4 CpG islands, in region of interest 2 (ROI_2) 3 CpG islands, and in region of interest 3 (ROI_3) 3 CpG islands. In promoter region of MyD88 gene in ROI_1 in CpG1 island we found that the average methylation status is slightly higher in healthy/not tumor involved tissue than in cancer tissue without statistically signi cant difference (p > 0,05; p = 0,276). In CpG2 island we found that the average methylation status is statistically signi cantly lower in cancer tissue than in healthy/not tumor involved tissue with statistical signi cance of p = 0,0267 (p < 0,05). In CpG3 and CpG4 islands of promoter region of MyD88 gene there was no statistically signi cant diffrence in methylation status between cancer and healthy/not tumor involved tissue of the larynx with p-values p = 0,9803 (p > 0,05) and p = 0.2048 (p > 0,05), respectively. In ROI_2 the methylation status of CpG5 island and CpG6 island showed there was no statistically signi cant difference between cancer and healthy/not tumor involved tissue of the larynx with p-values p = 0,2302 (p > 0,05) and p = 0,1870 (p > 0,05), respectively. In the same ROI_2 in the promoter region of MyD88 gene in CpG7 island we found statistically signi cant difference in the methylation status between cancer and healthy/not tumor involved tissue. We found that the methylation status of CpG7 in promoter region of MyD88 gene was statistically signi cantly lower in cancer tissue in relation to healthy/not tumor involved tissue according to p-value, p = 0,0174 (p < 0,05). In the ROI_3 in CpG8 and CpG9 islands of promoter region of MyD88 gene there was no statistically signi cant difference in the methylation status between cancer and healthy/not tumor involved tissue with p-values, p = 0,1740 (p > 0,05) and p = 0,1255 (p > 0,05), respectively. In the CpG10 island of promoter region of MyD88 gene we found that cancer tissue is statistically signi cantly hypomethylated in relation to healthy/not tumor involved tissue of the larynx with p-value p = 0,0407 (p < 0,05) The purpose of this analysis was to de ne overall methylation status of the promoter region of the MyD88 gene. The results of this analysis showed there was statistically signi cant difference in overall methylation status between cancer and healthy/not tumor involved tissue in the promoter region of MyD88 gene. The cancer tissue was statistically signi cantly hypomethylated in relation to healthy/ not tumor involved tissue of the larynx with p-value p = 0,0061. In overall difference in the methylation status signi cant contribution was obtained in 3 CpG islands: CpG2(p = 0,0267), CpG7(p = 0,0174), and CpG10(p = 0,0407). (Fig. 1) Results of analysis of ASC/TMS1 gene methylation status in healthy/not tumor involved and tumor tissue To accomplish differential methylation analysis of ASC/TMS1 gene in healthy/not tumor involved and tumor tissue, the methylation status of CpG islands of gene promoter region was performed. (Table 2) According to nucleotide sequence of DNA in gene only 1 region of interest (ROI) was de ned. In this ROI_1 3 CpG islands were detected. The methylation status of CpG1 island in the promoter region of ASC/TMS1 gene showed there was no statistically signi cant difference in the methylation status between healthy/not tumor involved and cancer tissue of the larynx according to p-value p = 0,0797 (p > 0,05). In the CpG2 and CpG3 islands of the promoter region of the ASC/TMS1 gene there was no statistically signi cant difference between methylation status of the cancer and healthy/not tumor involved tissue of the larynx with p-values p = 0,2665 (p > 0,05), and p = 0,3526 (p > 0,05), respectively.
The results of the analysis of the overall methylation status of CpG islands in the promoter region of ASC/TMS1 gene showed no statistically signi cant difference between cancer and healthy/not tumor involved tissue of the larynx p = 0,1562 (p > 0,05). (Fig. 2) As promoter region methylation is one of the mechanism of gene expression control the purpose of the analysis on the protein level was to investigate if there is a correlation of promotor gene methylation and protein expression.

Results of analysis of MyD88 protein expression in healthy/not tumor involved and tumor tissue
To accomplish MyD88 protein expression we used Western blot method applied on 21 samples of healthy/not tumor involved and tumor tissue. The results were statistically analysed and we found statistically signi cant difference in MyD88 protein expression between healthy/not tumor involved and tumor tissue of the larynx with p < 0,0001. (Fig. 3) Results of analysis of ASC/TMS1 protein expression in healthy/not tumor involved and tumor tissue ASC/TMS1 protein expression was analysed by Western blot method applied on 23 samples of healthy/not tumor involved and tumor tissue. The results were statistically analysed and we found statistically signi cant difference in MyD88 protein expression between healthy/not tumor involved and tumor tissue of the larynx with p = 0,0157. (Fig. 4)

Discussion
Head and Neck Squamous Cell Carcinoma (HNSCC) which comprise laryngeal carcinoma is frequent cancer with unfortunatelly increasing rate of incidence [20]. According to our knowledge based on diagnosis and treatment of this cancer we still need development of molecular targets, signaling molecular pathways and reliable diagnostic modality in discovering early disease state. In most European countries according to the data of World Health Organisation (WHO) the incidence of laryngeal cancer is decreasing in last two decades [21]. A slight decrease in the overall incidence of HNSCC comprising laryngeal cancer has been detected mostly due to global anti-smoking and anti-alcohol consumption compaign [22]. The relationship between chronic in ammation, innate immunity and carcinogensis is not new and it is widely recognized but most of the molecular mechanisms involved in this relationship are still not revealed. Oncogenes associated with different molecular patterns and functions leed to proin ammatory response and in ammation following infection is associated with carcinogenesis [21]. Chronic in ammation triggers increase the risk of cancer formation and development like infection (Helicobacter pylori for gastric cancer and gastric mucosal lymphoma, Human papilloma virus for cervical and oropharyngeal cancer, Hepatitis B/C virus for hepatocellular cancer), immune deregulation and autoimmune diseases and unknown origin infections [23]. The key mechanisms include transcription factors activation by the activation of the receptors of innate immune system resulting in proin ammatory cytokines secretion. Accordingly, we can say that molecular pathways related to cancer and in ammation are crucial factors in carcinogenesis and genetic instability of important molecules in this process are related to cancer development. It is the reason why molecular pathways related to cancer and in ammation present the aim of research of new strategies for cancer treatment and prevention. The innate immunity system is very important part of immune system and comprises mechanisms to defend organism from dangerous signals like pathogenic microbes or cellular stress signals. The innate immunity response is based on activation of an array of germline-encoded pattern-recognition receptors (PRRs) and by ligand activation of those receptors the cells which expressed receptors are activated. It is well known that innate immunity receptors play a role in cancerogenesis and metastatic cancer as chronic in ammation is induced by innate immunity system and stimulates occurence and development of cancer. 24 The important role in activation and regulation of activation of innate immunity receptors have adaptor molecules of signal pathways. In our study we investigated the correlation of methylation status and expression of MyD88 gene and ASC/TMS1 gene, adaptor molecules, in TLR and NLR signal pathways and we wanted to search if the methylation status of this genes can be used as potential prognostic biomarker. We hypothesized that changed methylation status of promoter regions of genes involved in activation and regulation of the in ammation, in healthy/not tumor involved and cancer tissue of the larynx, can be related with occurence and progression of laryngeal cancer. We think that results of our research represent important contribution in investigation of adaptor molecules methylation status as yet in the literature there is no data considering MyD88 gene methylation status in cancer. Also, methodology used in this research is innovative as we used in our analysis of the methylation status the pyrosequencing method which is the "golden standard" method in methylation status determination. All results of ASC/TMS1 gene methylation status in recent literature were de ned only by methylationspeci c PCR.
The results of this study con rmed that there is signi cant difference in the methylation status of the promoter region of CpG islands of MyD88 gene between healthy/not tumor involved and cancer tissue of the larynx. Precisely, the results showed that the cancer tissue of the larynx is statistically signi cantly hypomethylated according to the healthy/not tumor involved tissue. Moreover, results of this research indicated that the overall rate of methylation status of the promoter region of MyD88 gene is statistically signi cantly reduced in cancer tissue according to healthy/not tumor involved tissue. Furthermore, MyD88 protein expression level, explored to de ne the correlation between methylation status and protein expression, strongly con rmed the correlation between methylation and expression. According to the results, reduced methylation of the promoter region of MyD88 gene in DNA isolated from cancer tissue corresponds to higher expression of this protein in cancer tissue as opposite to healthy/not tumor involved tissue higher methylation corresponds to reduced expression of MyD88 protein. The author showed that diethylnitrosamine (DEN) induced Hepatocellular carcinoma (HCC) is associated with increased serum interleukin-6 (IL-6) concentration whose production is largerly dependent on MyD88 signaling pathway [27]. Chefetz et al. demonstrated that during surgical debulking or chemotherapy in ovarian cancer, cellular debris released during the process of injury and cell death may activate the TLR2-MyD88-NFκB pathway in a speci c cancer cell population, the EOC stem cells, leading to tumor repair and hence disease recurrence [28]. Higgins et al. analzyed TLR-driven in ammation in a mouse model for stage IV breast cancer. The author found that high mobility group box 1 protein (HMGB1) was produced by 4T1 in a Myd88-dependent manner and that neutralizing extracellular HMGB1 resulted in 50% decrease in growth of the tumor cells [29]. Unlike other authors, Kinowaki et al. reported that tumor progression is associated with reduced expression of MyD88 in hepatocellular carcinoma (HCC) [30]. Pyrosequencing results with the aim of determining the methylation status of promoter region of MyD88 gene showed the difference in overall level of methylation status between healthy/not tumor involved tissue and cancer tissue. The overall diffrence level is statistically signi cant and three CpG islands highly contribute to this difference; CpG2, CpG7, and CpG10. The other tested CpG islands did not expressed the difference in the overall methylation level between healthy/not tumor involved and cancer tissue. Our results are signi cant as for the rst time according to the literature the methylation status of promoter region of MyD88 gene between healthy/not tumor involved and cancer tissue was analyzed. The result revealed statistically signi cant difference in the methylation status, MyD88 gene is hypomethylated in cancer tissue, and the results correlates with protein expression. Our results also showed there is a pattern of methylation of promoter region of MyD88 gene as only CpG2, CpG7, and CpG10 islands contributed to the result. If there is tissue speci c methylation status of CpG islands for the same gene, than, in the future, according to this model, we can be able to differentiate second primary lung cancer from solitary pulmonary metastasis in laryngeal cancer. It can be a diagnostic problem as patients suffering from laryngeal cancer have a great risk to develop primary lung cancer due to extensive tobacco smoking, or to develop solitary pulmonary metastasis which can sometimes comprise di culties in obtaining satisfactory biopsy. The results of the methylation status of promotor region of the ASC/TMS1 gene showed no statistically signi cant difference in the methylation status of different CpG islands between healthy/not tumor involved and cancer tissue. In addition, there is no signi cant difference in overall methylation status between healthy/not tumor involved and cancer tissue. However, according to our results there is signi cant difference in expression of ASC/TMS1 gene between healthy/not tumor involved and cancer tissue, expression of protein ASC is signi cantly higher in cancer tissue. Adaptor molecule ASC/TMS1 plays important role in proin ammatory signaling pathway and cancer related in ammation. Many authors analyzed ASC/TMS1 gene expression in divers cancers but the methylation status of this gene with correlation of the expression between healthy/not tumor involved and laryngeal cancer tissue was not yet published. Salaminen et al. examined the mechanisms involved in the epigenetic regulation of ASC/TMS1 as well as their signi cance in the coordination of apoptosis and in ammasome functions [31]. Hong et al. demonstrated that restoration of ASC expression in human colorectal cancer DLD-1 cells, in which ASC/TMS1 gene is silenced by aberrant methylation, potentiated cell death mediated by DNA damaging agent [32]. According to Virmani et al. methylation of ASC/TMS1 gene correlated inversely with expression in the pathogenesis of small cell and non small lung and breast cancers. Xiaofang et al. analyzed methylation status of tumor supressor genes, including ASC/TMS1, in cholangiocarcinoma. The authors revealed that the methylation of DAPK, p14 ARF , and ASC/TMS1 in cholangiocarcinoma is a common event. Furthermore, p53 mutation combined with DAPK, p14 ARF , and/or ASC methylation correlates with malignancy and poor prognosis [33]. Unlike former authors Kordi Tamandani et al. described the methylation status of TMS1/ASC and CASP8 genes in cervical cancer. In this study the authors concluded that the frequency of TMS1/ASC and CASP8 genes methylation in cervical cancer is rare (< 6%), and plays no critical role in development of cervical cancer [34]. Mirza et al. observed signi cant correlation between promoter methylation of ASC/TMS1 gene and loss of protein expression in breast cancer con rming that promoter methylation is an important mechanism for transcriptional silencing of these genes in breast cancer [35]. Deswaerte et al. revealed a protumorigenic role in gastric cancer for the key in ammasome adaptor ASC/TMS1 and its effector cytokine IL18. Genetic ablation of ASC/TMS1 in the spontaneous mouse model of intestinal-type gastric cancer suppressed tumorigenesis by augmenting caspase-8-like apoptosis in the gastric epithelium, independently from effects on myeloid cells and mucosal in ammation. The authors concluded that in ammasome activation that elevates IL18 helps drive gastric cancer by protecting cancer cells against apoptosis, with potential implications for new therapeutic strategies in this setting [36]. Machida et al. studied proapoptotic ASC/TMS1 gene in lung cancer and used the ndings to develop a sputum marker. The authors concluded that hypermethylation of ASC/TMS1 gene is a potential marker for late-stage lung cancer and, in sputum, could predict prognosis in patients resected for early-stage disease [37].
Results of the analysis of the methylation status of ASC/TMS1 gene in our study are different related to results from published articles. According to our results there is no statistically signi cant difference in the methylation status between healthy/not tumor involved and cancer tissue. This can be explained by the fact that in this study the methylation status was determined using pyrosequencing method which is considerably more precise than commonly used methods, like methylation-speci c PCR.
In addition, in this study there is no correlation between methylation status of ASC/TMS1 gene and protein expression which can indicate that protein expression undergoes alternative post-transcriptional mechanisms of regulation.

Conclusion
In conclusion, the main contribution of the present study is in determination of the methylation status of MyD88 gene and ASC/TMS1 gene by pyrosequencing method which is still considered to be the "golden standard" in methylation status analysis. Furthermore, we found statistically signi cant difference in the methylation status of MyD88 gene between healthy/not tumor involved and laryngeal cancer tissue. We also showed that regarding gene/protein MyD88 there is a positive correlation between methylation status and gene expression -hypermethylation correlates with reduced protein expression. It is worth mentioning that regarding gene/protein ASC/TMS1 there is no positive correlation of methylation status and protein expression. In fact, our results showed there is no statistically signi cant difference in gene methylation status but there is signi cantly higher protein expression in laryngeal cancer tissue. The pattern of methylation which comprises the tissue speci c methylation status indicates potential biomarkers in differential diagnosis of Head and Neck and lung squamous cell carcinoma.

Declarations
Acknowledgements: none Funding: This research did not receive any speci c grant from funding agencies in the public, commercial, or not-for-pro t sectors.
Availability of data and materials: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
There is statistically signi cant difference in ASC/TMS1 protein expression between healthy/not tumor involved and tumor tissue. The values are shown as mean value ±SEM

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.