Comparative miRNA Expression Profile Analysis of Squamous Cell Carcinoma and Peritumoral Mucosa from the Meso- and Hypopharynx

Abstract

Background/Aim: The role of microRNAs (miRNA) in carcinogenesis is related to their genome-regulatory function. The aim of the present study was to identify and compare miRNA expression signatures of meso- and hypopharynx squamous cell cancers in consideration of the cancer field hypothesis. Patients and Methods: Thirteen snap-frozen biopsy series of tumors and peritumoral tissues from the meso- and hypopharynx were analyzed regarding their miRNA expression with quantitative real-time polymerase chain reaction (qRT-PCR). Results: Significant miRNA expression changes of miR-21, -27a, -34a, -143 and -146a were found in peritumoral tissues that were dependent from the tumor location and the distance from the primary tumor site. Conclusion: miRNA expression analysis was found to be appropriate for molecular segregation of tumor location and peritumoral tissue segments, and appears to be a promising marker for cancer field characterization.

In Europe, according to the age-standardized rates per 100,000 population, Hungary had the highest estimated incidence, mortality and prevalence in lip, oral cavity and pharyngeal cancers for both sexes. Based on the cancer registerty held by the National Institute of Oncology, in 2010 and 2011 meso- and hypopharyngeal carcinomas counted more than 1,200 cases. The number of diagnosed mesopharyngeal carcinomas in our clinical site has doubled, while hypopharyngeal carcinoma cases have tripled since 1983 (Figure 1 and Figure 2). Incidence of meso- and hypopharyngeal carcinomas is constantly growing, while the rate of mortality is above 50% due to late-stage recognition. In our clinical practice, malignant cancers of the meso- and hypopharyngeal regions are most specific in the 55-60 age range population, although the appearance of the different pharyngeal carcinomas has been observed in a growing number at younger ages in the past few years. It is important to note that the cancers affecting only one region, having good operability options and a better prognosis are disappearing, while the ones involving more than one region, with aggressive spreading and progressive prognosis are becoming more frequent. Our goal was to find and study a molecular marker that could be used as a predictive and a prognostic tool to complete clinical diagnostics.

Figure 1. Number of malignant mesopharyngeal tumors in the Department of Otorhinolaryngology and Head and Neck Surgery, Pécs, Hungary between 1983-2012.

Figure 2. Number of malignant hypopharyngeal tumors in the Department of Otorhinolaryngology and Head and Neck Surgery, Pécs, Hungary between 1983-2012.

In recent years micro-RNAs (miRNA) have been excessively studied. These small, 19-25 base-pair-long RNAs are transcribed from an intra- or intergenic segment of DNA and are not translated into proteins. Matured miRNA binds to a ribonucleotid complex, called RISC (RNA induced silencing complex), and drives the regulation of mRNA translation, either by binding to the target mRNA’s 3’ UTR (untranslated region) segment, or to 5’ UTR or just directly to the ORF (open reading frame). This way translation is blocked or endonucleases are activated causing the degradation of mRNA target (1,2). An estimated 50-60% of the human genome is regulated by miRNAs. A specific mRNA can be regulated by multiple miRNAs, and a miRNA can modulate translation of more then hundred mRNAs (1). This RNA interference mechanism establishes precise modulation of gene expression, harmonic regulation of cellular functions as cell division, differentiation, apoptosis and intermediary metabolism (3). miRNAs can behave either as oncogenes (oncomirs) or tumor suppressors according to which target mRNA is regulated by them (1). miRNAs can be detected in blood serum, saliva and other body fluids (4-9). Cell-free, matured miRNAs – because of their small size – can easily escape from endonuclease activity, this characteristic make them well reproducible and detectable in laboratory experiments (10).

Clinical management of HNSCC is complicated by the high rate of local tumor recurrence or development of a second primary tumor. The development of a recurrent tumor is strongly supported by both the Somatic Mutation Theory and the Tissue Organization Field Theory. Peritumoral tissues which might appear normal on macroscopical and histopathological evaluation can show several genetic and epigenetic alterations: DNA mutations, chromosomal aberrations, genomic instability, changes in DNA methylation and miRNA dysregulation. Our aim was to analyze miRNA expression pattern changes of meso- and hypopharyngeal mucosal tissues surrounding the primary tumors within and beyond the margin of surgical removal, to be able to model the extension field of modified molecular regulatory around tumorous tissue organization (11,12).

Patients and Methods

Sample collection. Samples of meso- and hypopharynx carcinomas were collected from patients diagnosed and operated on at the Department of Otorhinolaryngology, Head and Neck Surgery of the Clinical Center at University of Pécs. All removed tissue blocks were verified by histopatolocigal examination. During tissue sample collection we followed a standard mapping biopsy strategy and removed small fractions (approx. 0,5-1 mg) of mucosal tissues from the tumor edge (0), 1cm (1) and 2 cm (2) distance from the macroscopic tumor margin and from intact mucosal tissue being at least 3 cm (3) distance from the tumor margin. During the study 52 tissue samples, originating from 13 patients, were collected. Five patients suffered from hypo- and eight from mesopharnyx carcinoma. On clinicopathological classification tumors were listed into TNM II, III and IV stage squamous cell carcinomas.

RNA isolation. Tissue samples from mapping biopsy were immediately snap-frozen to –80˚C following collection until molecular laboratory processing. Sixty μg of tissue sample were homogenized in 150 μl lysis puffer (HighPure miRNA Isolation Kit, catalogue number: 05080576001, Roche, Mannheim, Germany) by using MagNa Lyzer Green Beads (Roche) cheramic beads tubes in MagNA Lyzer (Roche) shaking homogenizer. Total RNA was isolated with downstream application of the High Pure miRNA isolation kit (Roche), following the manufacturer’s instructions. RNA quality was checked by nano-drop absorption photometry and only the RNA fractions with (260/280 nm A>1,9) were used for reverse transcription.

cDNA synthesis and RT PCR. Five μg/μl of the RNA templates was used for cDNA synthesis with the Universal cDNA synthesis kit (Quiagen, Woburn, MA, USA), applying the random hexamer priming, included in the kit. cDNA samples were evaluated for miRNA expression in Roche LC480 system (Roche). For miRNA expression analysis specific primers were chosen from the universal miCURY LNA primer set (Exiqon, Vedbaek, Denmark) according to: hsa-miR-21, -27a, -34a, -143, -146a, -148a, -155, -221. Universal miCURY LNA 5S rRNA and U6 snRNA were used as controls for relative quantification. PCR mastermix contained 2 μl specific primer mix, 8 μl cDNS-template and 10 μl LC480 SYBR Green I Master mix in a total volume of 20 μl. Amplification was carried out on a 8×12 plate according to the following design: 6 tumor specific miRNA and two internal controls, that were examined on ten nucleic acid samples with unknown concentrations, on one set of positive samples with known concentrations and against one set of negative controls.

Statistical analysis. Relative quantification results were calculated by Exor4.0 software of LC480 (Roche), using the ΔΔ-CP method. Calculated relative quantification rates grouped according to tumor and peritumoral tissues were used for further statistical analysis by two tailed two sample T-test, as well as binary logistic regression- using the software SPSS 21.0.

Results

During miRNA expression analysis of meso- and hypopharyngeal mapping biopsies from tumor and peritumoral tissues, we found perceptible differences in characterizing the miRNA expression of mucosa surrounding the tumor tissue. Mucosal tissues located 1 cm away from surgical excision line show only a little difference in expression rate pattern from the tumor itself, while those 2 and 3 cm away from the tumor differ prominently. The samples taken from 2 cm and 3 cm distance from the tumor center shared many similarities in miRNA expression patterns. In the different line segments of peritumoral mucosa a specific miRNA expression pattern has been found. miR-21, -27a and -146a showed a decreasing relative expression rate scaling from the tumor tissues, where their expression is high, towards the peritumoral tissues where their expression is gradually lower according to distance. miR-34a and -143 was found to have similar but inverse expression pattern being the lowest in the tumor and higher in the tumor surrounding tissues. We could not observe such expression changes according to miR-148a, -155 and 221 (Figure 3).

Figure 3. The standard mapping biopsy - 1: tumor biopsy, 2: 1 cm 3: 2 cm, 4: 3 cm distance from the macroscopic tumor margin in the normal mucosa.

Analysing the miRNA expression reflected to the global expression measured, in cancer cells miR-21, miR-27a, miR-221 give nearly 50% of the cells’ total miRNA expression. This number is barely 15% in cells localized in tissues, which can be found outermost from the tumor tissue. While miR-34a and miR-143 show exactly the opposite; their expression rates are below 10% in tumor tissue, and growing up step by step to 40% as we are going further from the tumor. Mucosa being 1 cm far from surgical excision line shows little difference from the tumor itself in the expression rate referred on each other, while in the distances of 2 and 3 cm these rates differ prominently. Sample from 2 cm presents miRNA expression pattern similarities with intact mucosa in the distance of 3 cm. Expression differences are the greatest between the intact mucosa from 3 cm and the tumor (Figure 4).

Figure 4. miRNA expression related to their global expression in the tumor and 1 cm, 2 cm, 3 cm distance from the tumor.

Tumor and peritumoral samples of meso- and hypopharynx location showed significantly different miRNA expression rates on binary logistic regression analysis according to miR-21, miR-143, miR-155 and miR-221. In samples from hypopharynx miR-21, 143 and 155 have shown significantly higher expression rates, while expression of miR-221 was detectable solely in samples collected from the mesopharynx (p<0.05).

Study of tumor and peritumoral tissue samples – as in hypo- so in mesopharynx –, tumor and microscopically-intact tissue from 1 cm showed that they are both epigenetically highly similar (Figure 5 and Figure 6).

Figure 5. Mesopharyngeal carcinoma and peritumoral tissue (Photograph is published with the written permission of Orvosi Hetilap).

Figure 6. Hypopharyngeal carcinoma and peritumoral tissue (Photograph is published with the written permission of Orvosi Hetilap).

Discussion

Carcinogenesis of oral, laryngeal, meso- and hypopharyngeal squamous cell carcinomas are often described by the cancer field or as we nowadays call it, tissue organization field theory (13-14). This theory states that; the enviromental factors that can initiate tumor development, and the factors, defining individual sensibility can make characteristic changes genomically in the phenotypically-intact mucosa cells. Precursor lesions, primary tumors, recurrent tumors or even a second primary tumor can develop on this genetically or epigenetically changed tissue organization field (Figure 7). This theory can explain the clinical features of these tumors such as rapid progression, high recurrence rates, occurrence of second primary tumors at a notably high rate.

Figure 7. Tissue organization field theory.

miRNAs are defined by individual genetics, but they also play an important role as morphostatic factors on the epigenetic level and they have been proven to be promising in the molecular characterisation of numerous type of cancers (14-16). We found interesting and characteristic differences between miRNA expression rate changes in meso- and hypopharyngeal tumors and peritumoral, macroscopically -intact tissues. We took the advantage of these stable and well reproducible molecules to identify and describe the cancer field, which clearly showed correlation to the changes in the miRNA expression patterns.

miR-21 was found to be statistically significantly elevated in all peritumoral tissues compared to 3-cm distant mucosa. Previous studies have already proven the role of miR-21 in the modified regulation of p53-dependent apoptosis and cell-cyle arrest in oral squamo-cellular carcinomas (17-22). In our study, miR-21 showed a significantly elevated expression rate in meso- and hypopharyngeal carcinomas. This expression rate was lowered in intact tissues around the mesopharyngeal carcinomas, while in hypopharyngeal carcinomas this rate was found to be significantly lower (3 cm) from the tumor itself. Logistic regeression analysis revealed that in hypopharyngeal carcinomas miR-21 expression rates were significantly higher than in mesopharyngeal carcinomas.

miR-34a and miR-143 have been shown to have tumor-suppressor activities. They can block those proteins and their translations, which play important roles in tumor cell proliferation and neoangiogenesis. Their lowered expression rates were identified in HPV associated head and neck cancers (23-28). In our study these miRNAs also showed lowered expression rates. In mesopharyngeal carcinomas these miRNAs were found 2 and 3 cm from the tumor, while in hypopharyngeal ones miR-34a could be detected in tissues just 1 cm from the tumor. The loss of miR-34a’s expression rate in these tumors is strongly associated with the tumor phenotype.

miR-146a and miR-148a are effective blockers of promoter-specific repressors, that play a key role in NF-kB (nuclear factor-kappa B) anti-apoptotic signaling pathway (29,30). These miRNAs play an important role at those signaling points which are shared by both the inflammatory and carcinogenetic pathways. In our study we found both miRNAs’ expression characteristic of hypophyrangeal carcinomas. These rates were also observed in the peritumoral tissues and even areas further way from the tumors.

miR-155 has been identified in laryngeal squamous cell carcinomas and can be seen to promote tumor invasion (31-33). In our study miR-155 has shown significant expression rates in tumors, and could be also seen in tissues (2 cm) away from the tumor.

miR-221 was seen in correlation to cancers with bad prognosis and the estimated stage of invasion (22,34,35). In our research we found that miR-221 was the characteristic miRNA in mesopharyngeal carcinomas and in their peritumoral tissues, this miRNA could not be detected in hypopharyngeal carcinomas. it is possible, that in the hypopharynx a variant miRNA with a different nucleotid sequence matures, which we were not able to efficiently amplify by our specific primer sequences. Our further studies will focus on this miRNA to throw more light on this difference.

miR-27a is one of the least studied miRNAs in the literature (36,37). During our research we found that this miRNA showed very similar behavior both in meso- and hypopharyngeal carcinomas as miR-21.

The site-specific comparison of meso- and hypopharyngeal samples mostly resulted in finding the presence of miR-221, and the differences between miR-21, -143, -155 expression rates. miR-21, -143 and -155 markers showed higher expression rates in hypopharyngeal squamous cell carcinomas, while the rate of miR-221 was significantly higher in mesopharyngeal ones. The difference in life style could be a possible explanation to this.

We observed that in miRNA expression rates, some characteristic patterns have been found to be common in tumor tissues, and in the intact mucosa at varying distances from the tumor. We have also found identical profiles and some features which distinguish meso- and hypopharyngeal carcinomas on miRNA expression level. Analysis of miRNA expression signatures was an efficient method to segregate tumor locations and peritumoral tissues. miRNA profile analysis is a promising marker for cancer field characterization. Future application of this method could enable early detection of the disturbed epigenetic regulation in peritumoral tissue fields showing unchanged macroscopic appearance that can be helpful in molecular diagnostics, prognostics and cancer risk assessment (38).

Acknowledgements

This work was supported by: Research Found of Pécs University Faculty of Medicine KA-2016-19 and the European Union and the State of Hungary, co-financed by the European Social Fund in the framework of TÁMOP 4.2.4.A/2-11-1-2012-0001 ‘National Excellence Program’. Proposal number: A2-ACSJD-13-0349.

We would like to thank Tamas Toth for his assistance provided by the editing of the figures.