Heat Shock Protein 27 (HSP27) as a Potential Prognostic Marker: Immunohistochemical Analysis of Oral Epithelial Dysplasia and Oral Squamous Cell Carcinoma

Aim: To evaluate and correlate the expression of heat shock protein 27 (HSP27) in oral epithelial dysplasia (ED) and oral squamous cell carcinoma (OSCC). Materials and methods: This immunohistochemical study of HSP27 expression was performed on 45 samples retrieved from the departmental archives. It included 15 cases of oral ED, 15 cases of OSCC and 15 cases of epithelial hyperplasia (EH). The staining intensity and distribution were scored. The expression was compared between the study groups. Kruskal-Wallis Test, Mann-Whitney U Test and Chi-square tests were performed for statistical analysis using SPSS v21.0 (IBM Corp., Armonk, NY, USA). Results: There was a statistically significant difference in HSP27 staining parameters between EH, oral ED and OSCC. There was no significant difference between oral ED and OSCC. Conclusion: HSP27 expression shows enhanced expression in oral ED and OSCC. Its expression should be investigated using larger sample sizes with clinico-pathological correlation to prove its efficiency as a prognostic marker. It will help us in defining treatment modalities so that mortality and morbidity associated with OSCC could be reduced.


Introduction
In the global scenario, cancer of the head and neck is ranked third among common malignancies noted in developing countries. Squamous cell carcinoma is the most frequent with a 46% survival rate [1]. Eighty-six to 95% of such head and neck malignancies arise from the surface epithelium. Oral squamous cell carcinoma (OSCC) can involve any part of the oral cavity and oropharynx but predominantly involves the tongue. A multitude of risk factors and genetic determinants play a critical role in its pathogenesis [2,3].
Most patients with OSCC report at advanced stages that require radical surgery, higher doses of radiotherapy and chemotherapy that severely impacts speech and swallowing and compromises quality of life. Though multiple tumor markers have potential, an OSCC screening tool with an accurate diagnostic and prognostic marker is still a mirage [4].
Oral potentially malignant disorders have varied clinical presentations and it is difficult to predict the evolution of malignant change. Prognosis could be determined by histopathology and expression of molecular markers. The heat shock protein (HSP) family promotes growth and survival by enhancing stress resistance in tumor cells. HSPs also increase tumor immunity by stimulation of innate immune responses [1].
These proteins, primarily the HSP70 family, HSP90 family, and HSP27, are essential for normal homeostasis of the body. HSPs prevent protein coagulation and maintain multi-protein complexes in stress-induced states. HSPs regulate refolding of damaged proteins and destruction of misfolded/short-lived proteins. HSPs exert their anti-apoptotic effect by interacting with Bax, Cytochrome C and DAXX [5]. The overabundance of HSPs in stressed and transformed cells could favor anti-apoptotic activity [6].
Elevated expression of HSP27 either individually or in combination has been widely reported in cancers of breast, uterus and kidney, osteosarcoma and leukemias [6]. HSPs are produced transiently after cellular stressor events like heat shock or chemical and oxidative stress. HSPs are often overexpressed in tumor cells and reduce the response to chemotherapy with their anti-apoptotic capabilities [7,8]. HSP expression is labeled as a marker of the malignant potential of oral epithelial lesions [9].
The interesting possibility is that elevated HSP levels in tumor cells could impart a selective pro-survival edge and aid in carcinogenesis [10][11][12].

Materials And Methods
This retrospective study with a total of 45 specimens of formalin-fixed paraffin-embedded blocks of previously diagnosed patients were taken from the department archives. Of these 45 specimens, three groups were formed, Group 1: Oral Squamous Cell Carcinoma (OSCC), Group 2: Oral Epithelial Dysplasia (ED) and Group 3: Epithelial Hyperplasia (EH) with 15 samples for each group. Sections of each tissue (3-4 micrometers thick) were prepared with a Leica RM2245 microtome (Leica, Wetzlar, Germany).
One section was taken on a normal slide for routine hematoxylin and eosin stain as described in Bancroft [13]. Another section taken on a Poly-L-Lysine coated slide was stained for immunohistochemistry of HSP27, as per manufacturer's instructions (Biogenex Inc, Houston, TX, USA). Harris Hematoxylin was used as the counterstain. Appropriate positive and negative controls were utilized for every batch.
Photomicrographs were captured with an LX400 Labomed Inc. (Los Angeles, CA, USA) microscope and assessed for staining characteristics. Histopathological diagnosis of stained slides of OSCC, ED and EH were evaluated by three independent Oral Pathologists for representative histopathology, according to the World Health Organization classification.
The staining intensity was graded on a scale of 0-4 and the mean percentage of positive cells in 5 high power fields showing maximum staining of HSP27 was used to grade 1-4. Staining Percentage was given as Score 1 for 0-25% stained cells, Score 2 for 25-50% stained cells, Score 3 for 50-75% stained cells and Score 4 for 75-100% stained cells [1]. Staining Intensity was given as Score 0 for No detectable immunoreactivity, Score 1 for Very low staining, Score 2 for Low staining, Score 3 for Moderate staining and Score 4 for High staining [1].
Staining index (SI) was calculated by multiplying staining percentage and staining intensity. Mean values of the staining index along with staining intensity (SIN) and percentage of stained cells (PSC) by three observers were tabulated and graphically plotted [1].
Difference in the staining index of epithelial cells between groups was analyzed with Kruskal-Wallis Test, Mann-Whitney U Test and Chi-square tests performed using SPSS v21.0 (IBM Corp., Armonk, NY, USA). p ≤ 0.05 was considered statistically significant.

Results
In normal epithelium the staining was present only in the basal cells. Among 45 cases of in the present study, 15 cases were oral ED (with five cases each of mild dysplasia, moderate dysplasia and severe dysplasia), 15 cases were OSCC (with five cases each of well differentiated SCC, moderately differentiated SCC and poorly differentiated SCC) and 15 had EH ( Table 1).  Mean staining index was highest in OSCC (6.07) followed by ED (2.40) and least in EH (1) (  Mann-Whitney Test was performed between three groups. HSP27 expression was higher in ED when compared with EH with p value of 0.027. Comparison with EH and OSCC was also significant with p value of 0.003. But, the comparison between ED and OSCC was insignificant with p value of 0.062 ( Table 3).

Discussion
Family of heat shock proteins (HSPs) can expedite growth and survival of tumor cells due to their cytoprotective properties. HSP27 has an anti-apoptotic role with its mitochondrial interplay and can promote cancer progression. Increase in free radicals production leads to enhanced HSP27 expression to protect the cells from oxidative damage. Tumor cells of OSCC show higher levels of free radicals and could be responsible for amplified HSP27 response [1,2]. Increase in anti-apoptotic nuclear factor kappa-light chain enhancer of activated B cells (NF-κB) activity through enhanced degradation of its main inhibitor, inhibitor of kappa B (I-κB) may contribute to the anti-apoptotic effect of HSP27 [4]. Research has revealed a robust correlation between OSCC and HSP27. Variations in HSP27 have been implied in carcinogenesis [11,12].
According to the studies conducted by Seoane et al. [14], Ito et al. [15], Leonardi et al. [16], Tekkesin et al. [3], and Lo et al. [12], the expression of HSP27 increases in dysplasia and OSCC when compared to its expression in normal and hyperplastic epithelium. Another role for HSP27, that of facilitating the activation of the ubiquitin-proteasome pathway [17]. Increased HSP27 expression in oral lichen planus when compared to normal epithelium was shown by Bramanti et al. [18] and Tyagi et al. [19].
In our study, epithelial cells in OSCC showed higher intensity than the ED, whereas HSP27 staining of epithelial cells was minimal in case of EH. Kruskall-Wallis test showed that there was statistically significant difference among the three study groups in the mean staining index of epithelial cells for HSP27. Furthermore, Mann Whitney test for mean staining index of epithelial cells for HSP27 showed significant differences in EH and ED (p=0.027), EH and OSCC (p=0.002) and insignificant differences between ED and OSCC (p=0.062). Consistent with the above studies, our study also had progressive increase in expression of HSP27 among hyperplastic, dysplastic and carcinomatous lesions.
When the expression level of HSP27 in our study was compared within grades of dysplasia, the results show insignificant decrease in expression from mild to moderate dysplasia and insignificant increase from moderate to severe dysplasia. It could be attributed to the smaller sample size in our study. Expression level of HSP27 decreases with the increase in grades of dysplasia as stated by Leonardi et al. [16]. The downregulation in dysplasia could indicate the defective cytoprotection against mutagenic environmental factors and thus augment the transformation of oral epithelial dysplasia into OSCCs.
Our results were consistent with these studies but the results were insignificant when all grades of OSCC were compared for the expression level of HSP27. Our study results show higher expression of HSP27 in MDSCC as compared to PDSCC. Nakajima et al. [20], Muzio et al. [21], Wang et al. [2] and Leonardi et al. [16] reported inverse relation of expression of HSP27 with increase in grades of OSCC and stated that expression of HSP27 decrease as the grades of OSCC increase from WDSCC to PDSCC.
Prognostic role of HSP27 in OSCC is still ambiguous. Some studies have reported enhanced survival with HSP27 expression but inverse correlation has been reported too. Inconsistent values could be due to multifactorial oncogenesis [2,22]. Several proteomic studies identified HSP27 as a highly overexpressed protein in OSCC. Proteomic study done by Wang et al. [2] also demonstrated that higher HSP27 expression is associated with tongue SCC invasion and metastasis.
Reduced expression of HSP27 was reported as a negative prognostic factor in oesophageal carcinoma but overexpression was related poor diagnosis in astrocytoma [4]. Ito et al. [15] reported that HSP27 did not have any correlation with staging, histological grade, survival and metastatic potential. Mese et al. [23] reported reduced HSP27 expression in PDSCC than WDSCC but did not have any interrelation with clinical staging. Zhu et al. [24] observed HSP27 overexpression in a head and neck squamous cell carcinoma from a metastatic lymph node, but it was rarely expressed in the primary cancer cells from the same patient. HSP27 exists as large oligomers that reorganize to dimers and tetramers after phosphorylation. Increased HSP levels in tumor cells could be attributed to increased expression of HER & c-Myc and loss of p53 function. HSP27 enhance tumor cell growth and promote tumor cell survival by inhibition of caspase-dependent apoptosis. HSP27 has a dual role that can impart resistance of chemotherapy and can also induce CD8+ T cell-mediated anti-tumor immune responses if released into circulation [25].
Wei et al. performed a meta-analysis of HSP expression in oral cancer [22]. The pooled hazards ratio revealed that HSP27 protein overexpression significantly increased the survival rate of oral cancer patients. HSP27 might have different roles in different subtypes of oral cancers, and tumor stage might affect the HSP27 expression.
Little progress has been made in raising survival rates over the previous 30 years, despite numerous advancements in treatment. Therefore, preventing it and developing technology that makes early identification of OSCC easier that could increase survival and quality of life [1]. There is a need for extensive studies with large sample size and detailed clinical information, especially clinico-pathological data to observe the effects of HSP family proteins in various parameters of oral cancer including risk, progression, and prognosis [22][23][24][25][26][27][28][29].
Tekkesin et al. [3] reported higher HSP27 expression in leukoplakia and OSCC than in normal mucosa but no difference in HSP27 expression was apparent between OSCC and leukoplakic lesions. Similarly, we have observed significant increase in HSP27 expression from ED and OSCC but did not observe any significant difference in HSP27 levels between these two groups.

Limitations
Our study had a smaller sample size among the study groups. Other HSP family proteins like HSP70 and HSP90 could be studied. A multi-centric study with detailed clinico-pathological correlation is needed to ascertain the actual role of HSP27 expression among various grades of dysplasia and subtypes of OSCC.
Once the HSP27 expression is validated, it could be used to devise a prognostic model in the future. Variations in HSP27 levels at the time of diagnosis and post-radiotherapy or post-chemotherapy will add valuable information in establishing a treatment protocol for patients with OSCC.

Conclusions
In cases of OSCC, expression of HSP27 should be investigated more thoroughly using larger sample sizes of OSCC with a complete record of etiological factors and clinical findings. Detailed immunohistochemical analysis of HSP27 expression could highlight its role in ED and OSCC. HSP27 would definitely prove to be a good prognostic marker and help us in defining treatment modalities for OSCC, so that mortality and morbidity associated with OSCC could be reduced.

Additional Information Disclosures
Human subjects: Consent was obtained or waived by all participants in this study. Saveetha Dental College Institutional Human Ethics Committee issued approval IHEC/SDC/PhD/OPATH-2212/22/001. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.