Defensins play a role as peptide antibiotics and constitute an important element of anti-infectious protection (1). Apart from a direct antimicrobial action, defensins also have many other functions, such as antifungal and antiviral activities, and chemotactic activity towards dendritic cells, T cells, basal cells and neutrophils. They may induce the production of certain chemokines and proinflammatory mediators; regulate complement activity; inhibit fibrinolysis and the production of certain glucocorticosteroids; intensify the proliferation of lymphocytes, endothelial cells and fibroblasts; bind and neutralize endotoxins; stimulate healing of sores; induce degranulation of basal cells and show antitumor activity (2). There are three groups of defensins – defensin α and β, and ? defensins – generated during the cyclisation of the two α defensins (1). In humans, six α (HNP 1-6 – human neutrophil peptide) and four β defensins (HBD 1-4) have been identified (3). Changed expression level of defensins was reported in many pathologies, including dermatological lesions such as psoriasis, acne vulgaris and atopic dermatitis (3-5). Examinations of cancer samples such as oesophageal squamous cell carcinoma, primary cutaneous squamous cell carcinoma (SCC) and basal cell carcinoma (BCC) also showed changes in defensins profile, suggesting their potential participation in these pathologies (6-8).

Knowledge of molecular markers of tumour transformation is essential for malignant diseases because it enables a diagnosis to be made of a disease already at the level of molecular changes, and it aids cause-directed and efficient therapy. Non-melanoma skin cancers are still problematic in diagnosis because of clinical features resembling other pathologies i.e. keratoacanthoma. Keratoacanthoma (KA) is a relatively low-grade malignancy that histologically and clinically resembles well-differentiated SCC (9, 10). It develops on unchanged skin in the form of a dome shaped, well-limited, hard nodule, usually with the same colour as the surrounding skin. The central part of the tumour has a crater-like depression with embanked margins, filled up with a keratin mass (11, 12). KA is characterised by sometimes fast growth and a tendency to regression with scarring (11). BCC is the most common type of epithelial cancer; it is characterised by low clinical and histological malignancy and slow growth and rarely metastasises (13, 14). It is usually found on unchanged skin of the face and head and spreads in multiple directions with edgings. The multiple and multifocal presence of BCC is classified as nevoid BCC syndrome (Gorlin-Goltz syndrome) (13, 14). Clinically, the changes can be highly differentiated, superficial, corneous, nevus-like, scleroderma-like, nodular or sore-like (15). SCC is the second most frequent (after BCC) skin cancer. It originates from squamous epithelium and develops from premalignant states (leucoplakia). It is characterised by fast growth and metastases to regional lymph nodes. Clinical forms of SCC (sore and papular) morphologically resemble KA and BCC (9, 16). Therefore additional markers could facilitate diagnosis of non-melanoma skin pathologies.

Significant risk factors for BCC, SCC and KA are: UV radiation, ionized radiation, post-infectious and post-drug immunosuppression, chemical carcinogens (arsenic, wood tars), old age, male gender, light skin type and human papilloma virus infection (8, 9, 16-20). Most of these factors influence immunological response. Due to its multidirectional activity, defensins may be involved not only in the molecular pathogenesis of non-melanoma skin cancers, but also in the immunological reaction leading to self-regression of KA. Understanding of the molecular mechanisms underlying KA regression may contribute to the development of an efficient treatment for skin cancers. Better knowledge about defensins level in non-melanoma skin pathologies could improve their diagnosis and could affect better understanding of molecular mechanism underlying these pathologies. Till this time there is no research comparing defensins profile in these three non-melanoma pathologies (KA, SCC and BCC) in one study.

The aim of this work was to evaluate the mRNA profile of defensin family transcripts as an additional molecular marker of nonmelonoma skin pathologies.

The study included a group of 39 patient (16 females and 23 males; mean age 71.8 ± 9.5 years) diagnosed and treated in the Dermatology Clinics and Department of Medical University of Silesia in Katowice. Based on clinical and pathomorphological examination, 9 cases were kerathoacanthoma (KA), 11 were squamous cell carcinoma (SCC) and 19 were basal cell carcinoma (BCC). The changes were located on the skin of the face and head. The biopsies for the pathomorphological and molecular examination were obtained from the centre of the tumour and the margins of healthy tissues where no tumour cells were found. After surgical removal, the samples were immediately preserved in the RNA stabilisation reagent RNAlater (Qiagen GmbH, Hilden, Germany). For microarray analysis 19 samples (6 KA, 3 SCC, 7 BCC and 4 margins of healthy tissues) were selected. The study was approved by the Bioethical Commission of the Medical University of Silesia. All of the patients were informed about the research and signed an informed consent form.

Total RNA was extracted from the tissue samples using TRIZOL^® reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s instructions. Total RNA extracts were treated with DNase I (MBI Fermentas, Vilnius, Lithuania) and purified with a RNeasy Mini Kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer’s protocol. The quality of RNA was estimated by electrophoresis on 1% agarose gel stained with ethidium bromide. The RNA concentration was determined on the basis of absorbance values at 260 nm using a Gene Quant Pro spectrophotometer (LKB Biochrom Ltd., Cambridge, UK).

The analysis of the expression profile of defensin-related genes was performed using commercially available oligonucleotide microarrays HG-U133A (Affymetrix, Santa Clara, CA) according to the manufacturer’s protocol. For finding significant genes comparative analysis was performed with the use of GeneSpring 12.6.1 platform (Agilent Technologies, Inc., Santa Clara, CA, USA) and PL-Grid Infrastructure. The differences were analysed using the Oneway ANOVA test with Benjamini-Hochberg Multiple Testing Correction and TukeyHSD Post Hoc test. Genes were considered as potentially differentiating when FC ≥ 1.1 (fold change) and the significance level was set at p < 0.05.

The levels of the DEFB4A and β-actin transcripts were evaluated with the use of the real-time QRT-PCR TaqMan technique. The quantitative analysis was carried out using a Sequence Detector ABI PRISM™ 7000 (Applied Biosystems, Foster, CA, USA). Amplification was performed with the use of commercially available oligonucleotide primers specific for DEFB4A and β-actin genes (DEFB4A: TaqMan Gene Expression Assay defensin, beta 4; β-actin: TaqMan B-actin Detection Reagents Kit, Applied Biosystems, Inc., Foster, CA, USA) and QuantiTect Probe RT-PCR Kit (Qiagen GmbH, Hilden, Germany). For the assay, positive (β-actin mRNA) and negative (no template) controls were used. The thermal profile for one-step RT-PCR was as follows: 50°C for 30 min for reverse transcription, 95°C for 15 min, 45 cycles at 94°C for 15 s and at 60°C for 60 s. The standard curve was created for β-actin cDNA (TaqMan DNA Template Reagents Kit, Applied Biosystems, Inc., Foster, CA, USA). A standard curve was then generated by plotting the Ct values against the log of the known amount of the β-actin cDNA copy number. The mRNA copy numbers of the gene examined were recalculated per 1 μg of the total RNA. The RT-PCR products and the molecular weight marker pBR 322/Hae III (Fermentas International Inc., Ontario, Canada) were separated on 8% polyacrylamide gel and visualised with silver salts. The length of the amplified fragments was assessed by analysis with GelScan v.1.45 software (Kucharczyk TE, Warsaw, Poland). The statistical analysis of the real-time QRT-PCR results was performed with the use of Statistica version 9.0 software (StatSoft Inc., Oklahoma, USA). The one-way ANOVA followed by Tukey posthoc analysis of the logarithmic parameters were used to assess differences in the expression of the studied genes between KA, SCC and BCC. The differences between the tumour and the margin of the normal tissue were analysed using the t-test. All of results were expressed as means ± SD. The significance level was set at p < 0.05.