Abstract
Background/aim: SRY-related HMG box protein 18 (SOX18) is a transcription factor involved in a range of physiological processes, including differentiation of endothelial cells during new vessel formation. Numerous studies are being conducted to determine its role in carcinogenesis. Materials and Methods: Thirty-five cases of squamous cell carcinoma (SCC), 61 cases of basal cell carcinoma (BCC), 15 cases of actinic keratosis (AK) and 15 normal skin (NS) cases were examined in the study. Expression of SOX18 was investigated with immunohistochemistry and light optic microscopy. The obtained results were subjected to statistical analysis including available clinicopathological data. Results: Nuclear expression of SOX18 was shown in vascular endothelial cells, basal layer cells of NS epidermis, as well as in AK, BCC and SCC cancer cells. Expression of SOX18 in SCC, BCC and AK cells was significantly higher than in NS (p<0.01, p<0.001 and p<0.01, respectively). Additionally, higher expression of SOX18 in BCC than in SCC cells (p<0.001) was observed. Conclusion: SOX18 may play a role in the development of BCC and SCC. Further studies with the use of additional markers tested at the mRNA and protein level are necessary for better explanation of SOX18 function in cancer transformation.
Nowadays, due to rapidly increasing incidence, malignant skin neoplasms are an important medical, economical and social problem. Based on the World Health Organization Classification of Tumours, the most common malignant skin tumours are: basal cell carcinoma (BCC), squamous cell carcinoma (SCC) and melanoma malignum (1). BCC and SCC are jointly referred to as non-melanoma skin cancers (NMSCs) and originating from stratified squamous epithelium cells, as in contrast to melanoma, which is of neuroectodermal origin (1, 2). In 2012, there were 5.4 million NMSC cases, including 3.3 million of BCC and SCC (3).
Actinic keratosis (AK) is a common intraepithelial proliferative lesion classified as a preinvasive cancer (4). Up to 20% of all AK have a tendency towards transformation into SCC and for patients diagnosed with at least 10 AK lesions, the 5-year risk of development of SCC is approximately 14% (5-7). The development of AK is induced by frequent skin exposure to intense ultraviolet radiation (UV) (8). An increased number of NMSCs was also shown in patients with an excessive exposure to UV radiation, both in UV-B and UV-A range (9, 10). This was probably associated with an increased risk of DNA mutations in keratinocytes, particularly in TP53 gene, as well as with induction of local immunosuppressive state within the skin as a result of such radiation. In turn, UV reduces the body's ability for an effective immune response against cancer cells (11, 12). It was also shown that long-term exposure to UV radiation is associated primarily with an increase in the incidence of SCC (7).
BCC originates from epidermal basal layer cells. Uncovered skin exposed long-term to UV radiation, mainly in the head and neck area, is a typical localization for BCC (85% of cases), with about 30% of cases located within the skin in the nasal area (6). Since BCC rarely metastasises, it belongs to the so-called ‘locally malignant lesions’. Growing, it infiltrates local tissues therefore causing their destruction and leading to the deformation of deeper structures, including cartilage and bone tissue (13, 14). In turn, SCC (also called carcinoma spinocellulare) is derived from the stratum spinosum and is much more aggressive in comparison to BCC (7). Under microscopy, areas of hyperkeratosis and keratin pearls (keratodes type) are often present, especially in cancer classified as highly differentiated (1). SCC morbidity increases with age, mostly after the sixth decade of life, and predominantly occurs in the areas of the skin exposed long-term to intense sunlight (backs of hands, head and face skin, lips and auricles). Clinical presentation of SCC is highly diversified, ranging from asymptomatic to moderately severe lesions, and up to ulcerative and bleeding exophytic tumours (14).
Clinical characteristic of patients with squamous cell carcinoma (SCC), basal cell carcinoma (BCC), actinic keratosis (AK) and normal skin (NS).
In the all abovementioned dysplastic lesions, i.e. AK, BCC and SCC, treatment of choice is their complete surgical removal. Further therapeutic decisions are based on the results of histopathological examination, which, especially in the case of SCC, may be associated with the need for adjuvant therapy with the use of radio- and phototherapy (7).
SRY-related HMG box protein 18 (SOX18) encoded by SOX18 gene is responsible for the interaction with specific DNA sequences (5′-A/T A/T CAAA/TG-3′), thus functioning as a transcription factor (15). It is a part of a family of SOX factors consisting of eight groups (A–H), which are classified based on high homology of structure and function (16). SOX18 belongs to group F and is involved in maturation and differentiation of endothelial cells in the prenatal period and postnatally in angiogenesis. The analogy of SOX18 function with widely known inductors of angiogenesis i.e. proteins from the vascular endothelial growth factor family (VEGF), has been shown (17). Expression of SOX18 was observed in cardiomyocytes (particularly in the walls of ventricles and interventricular septum), cells of gastric and jejune mucosa, as well in pneumocytes. SOX18 is probably not only engaged in wingless-type MMTV integration site family (WNT) signalling pathways, but also in β-catenin/T-cell factor complex-dependent transcriptional regulation (16, 18).
Taking into account that silencing of SOX18 gene may cause inhibition of proliferation, migration and invasion of, e.g. hepatocellular carcinoma cells, and that expression of SOX18 is increased in numerous cancer types e.g. gastric stromal tumours, pancreatic, liver, ovary and breast cancer, it seems that this protein may play an important role in the processes of cancer promotion and progression (18-24). To date, no expression of SOX18 in BCC and SCC has been described and such research, as well as their results, may advance the knowledge on the role of SOX18 in NMSCs.
Materials and Methods
Patients. Material for the research was obtained between 2005 and 2007 from the Department of Dermatology, Venereology and Allergology of Wroclaw Medical University, Poland. A total of 126 archival paraffin blocks were used: 96 cases of invasive skin cancer (35 of SCC and 61 of BCC), 15 cases of preinvasive skin cancer (AK) and 15 cases of normal skin (NS) collected during dermatosurgical procedures. Available clinicopathological data included: patient's age and sex, time between diagnosis and treatment, histological type and clinical stage, localization of primary lesion and presence of inflammation (Tables I, II and III).
Immunohistochemistry (IHC). The reactions were conducted on 4 μm- thick paraffin sections mounted on SuperFrost Plus microscope slides (Menzel Gläser, Braunschweig, Germany). Deparaffinization, re-hydration and exposure of the epitopes were performed with use of PreTreatment Link Rinse Station and Target Retrieval Solution (pH 9, 97°C, 20 min). Activity of endogenous peroxidase was blocked by 5 min exposure to Peroxidase-Blocking Reagent. All sections were rinsed with wash buffer and incubated for 20 min at room temperature with the primary monoclonal antibodies directed against SOX18 (clone D-8, 1:25; Santa Cruz Biotechnology, Santa Cruz, CA, USA). Secondary goat anti-mouse antibodies coupled to a dextran core, linked to horseradish peroxidase were applied. Next, visualization was performed using the EnVision™ FLEX+ system according to the manufacturer's instructions. The IHC reactions were performed in an Autostainer Link48 (DakoCytomation) automated staining platform. The reactions were visualized using 3,3’-diaminobenzidine tetrachlorohydrate (DAB+ chromogen). All slides were counterstained with Mayer's hematoxylin and subsequently closed with SUB-X Mounting Medium in Cover Slipper (DakoCytomation). All reagents, except primary antibody against SOX18 were obtained from DakoCytomation, Glostrup, Denmark.
Analysis of IHC reactions. The intensity of IHC reactions was evaluated with the use of an Olympus BX41 microscope (Olympus, Tokyo, Japan). Nuclear expression of the protein was assessed in the cells that exhibited colour reaction in three regions with the highest intensity of expression (hot spots), by calculating the percentage of cells with positive IHC reaction.
Statistical analysis. Prism 5.0 programme (GraphPad, La Jolla, CA, USA) was used for statistical calculations. Analysis of normal distribution of SOX18 expression was performed with the use of Shapiro–Wilk test. Non-parametric Kruskal–Wallis test was used for the evaluation of the average differences in SOX18 expression in groups and then multiple comparisons of mean ranks for all samples were performed (Dunn's test). In all evaluated cases, statistically significant p-values were accepted at less than 0.05.
Results
Nuclear expression of SOX18 was shown in vascular endothelial cells, basal layer cells of NS epidermis, as well as in AK, BCC and SCC cancer cells (Figure 1). Intensity of observed SOX18 expression varied between cancer cells in AK (mean±standard deviation: 53.07%±30.27), SCC (47.76%±29.08) and BCC cases (73.50%±19.80), while in NS it was found rarely (4.86%±2.74). Statistically significant difference was shown between the levels of expression of SOX18 in each of the groups (AK, BCC, SCC) vs. NS (p<0.01, p<0.001, p<0.01, respectively), wherein the level of expression was higher in cancerous lesions. Additionally, higher expression of SOX18 in BCC than in SCC cells (p<0.05) was found. The comparisons of BCC vs. AK and SCC vs. AK were not statistically significant.
According to the available clinicopathological data shown in the tables, no statistically significant differences were found in SOX18 expression.
Discussion
There exist numerous studies regarding the role of SOX18 in carcinogenesis and its possible usage in diagnostic and therapeutic process. Expression of this protein was shown in many malignant neoplasms e.g. ovarian, breast, gastric, lung and liver cancer (18-24).
In ovarian cancer, SOX18 expression was found in the nuclei of vascular and lymphatic endothelial cells, as well as in the nuclei and cytoplasm of cancer cells. In tumor cells, the expression was significantly higher in more advanced clinical stages of the disease (stage III and IV according to the International Federation of Gynaecology and Obstetrics). Moreover, SOX18 expression was associated with shorter patient survival (22). Similarly, in invasive ductal breast carcinoma (IDC), SOX18 expression was observed in endothelial and cancer cells. Increased nuclear expression of SOX18 in IDC cells was positively correlated with more advanced clinical stage of the disease and positive status of human epidermal growth factor receptor 2 (23). In the case of gastric cancer, the expression of SOX18 was significantly higher in cancer cells in comparison to surrounding noncancerous mucosa. Gastric tumors with SOX18 overexpression were characterized by cancer cells invading lymphatic vessels and the presence of metastasis in lymph nodes. Furthermore, the recurrence-free survival time and the 5-year survival rate were significantly decreased in those cases (21). Similar correlation was discovered as a result of the research performed with the use of hepatocellular carcinoma cell line, where overexpression of SOX18 was positively correlated with shorter patient survival. Moreover, its lowered expression was associated with decreased tumor invasiveness and aggressiveness (18). The above observations were confirmed by research on cases of non-small cell lung cancer. The level of SOX18 in such tumours was significantly higher in cancer cells in comparison to non-cancerous surrounding tissue. Additionally, survival was significantly shorter in patients with higher expression of SOX18 (24).
Pathological characteristics of the patients with squamous cell carcinoma (1).
Pathological characteristics of the patients with basal cell carcinoma (1).
Immunohistochemical expression of SRY-related HMG box protein 18 (SOX18) in normal skin (A, B), actinic keratosis (C, D), squamous cell carcinoma (E, F) and basal cell carcinoma (G, H). Magnification ×200.
The above facts may indicate that SOX18 plays a role in the progression of these cancer types. Our NMSC observations also confirm those findings. Higher expression of SOX18 in the cases of preinvasive cancer in the form of AK, as well as invasive SCC and BCC in comparison to normal skin may suggest that SOX18 is involved in the progression of specific types of skin cancer.
Studies on new markers involved in skin carcinogenesis may provide valuable information regarding the mechanisms of cancer formation, thus contributing to a better understanding of this process, and in the future, to more precise diagnosis and more effective therapy. Our results may facilitate distinguishing of NMSCs (BCC and SCC) and AK from other noncancerous skin lesions, which may prove very useful at early stages of formation of dysplastic skin lesions.
Acknowledgements
This work was funded by scientific grant no. ST 810 of the Wroclaw Medical University.
- Received February 8, 2016.
- Revision received March 23, 2016.
- Accepted March 28, 2016.
- Copyright© 2016 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
