Expression of MCM2 as a Proliferative Marker in Actinic Keratosis and Cutaneous Squamous Cell Carcinoma

Abstract

Background/Aim: Minichromosome maintenance protein 2 (MCM2) can be considered an indicator of cancer clinical outcome. In this study, we tried to estimate the usefulness of assessing MCM2 protein expression in actinic keratosis (AK) and cutaneous squamous cell carcinoma (cSCC).

Materials and Methods: The study included 22 lesions of AK, 57 of cSCC and 17 tissue samples of the healthy skin.

Results: Higher average expression of MCM2 protein in cSCC and AK was demonstrated in comparison to healthy skin (p=0.01). Likewise, the level of MCM2 expression differed statistically significantly (p=0.02) between SCC, AK, and healthy skin. Significant correlations between MCM2 expression and Ki-67 and p53 antigen were found (r=0.51, p=0.01; r=0.45, p=0.04 respectively) in AK lesions, however these relationships were not noted in cSCC.

Conclusion: MCM2 is overexpressed in both AK and cSCC lesions, however this protein cannot be considered an important indicator of proliferation in cSCC.

Actinic keratosis (AK) is one of the most common skin conditions with continuously increasing incidence. Histologically, AK is characterized by a proliferation of atypical keratinocytes limited to the epidermis and is mostly considered as a pre-neoplastic lesion which can transform into invasive squamous cell carcinoma (1,2). AK presents as erythematous, rough, scaly plaques which develop on the skin chronically exposed to ultraviolet radiation (UV). The correlation of increased AK prevalence in areas of high UV-exposure has been shown in a previous study (2). Age over 80 years, male sex, fair skin type and immunosuppression due to organ transplant belong also to the other known risk factors of this condition (3,4). Typical clinical variant of this condition may progress into thickened or hypertrophic (keratinized) lesions. Other subtypes of AK include lichen planus-like keratosis, pigmented actinic keratosis, actinic cheilitis or cutaneous horn (5). Without early detection and consequent treatment, the possibility of transformation into cutaneous squamous cell carcinoma (cSCC) is about 10% (6). It is believed that mutations in p53 and following inactivation of the p16 tumor suppressor are responsible for this progression (7).

Cutaneous squamous cell carcinoma (cSCC) is the second most common skin cancer and represents 20% of all non-melanoma skin cancers (8). The main cause of cSCC is cumulative UV radiation. Similarly, ionizing radiation, genodermatoses, human papillomavirus (HPV), arsenic, polycyclic aromatic hydrocarbons, immunosuppression, and chronic ulcers have proven as important factors in the etiopathogenesis of this tumour (9). The cancer mostly infiltrates local tissue and regional lymph nodes, though distant metastasis is possible (8,10). Clinical and histological type of tumor, histological differentiation, tumor size, localization, depth of infiltration and TNM stage are counted as factors with a great prognostic value of cSCC (11,12).

Minichromosome maintenance protein 2 (MCM2) is involved in DNA replication and belongs to a group of proteins which were first identified in the yeast cells of Saccharomyces cervisiae (13). The best known among these proteins are the MCM2-7 which are responsible for initiation and inhibition of DNA replication in all eucaryotic cells (14,15). MCM2-7 proteins were distinguished on the basis of the structure of their protein chain and the presence of specific MCM domain. This domain includes motifs which are characteristic to the ATPase and enables binding and hydrolysis of the ATP molecules (16,17). These proteins are crucial for the synthesis and elongation of DNA. First of all, MCM proteins form the prereplication complex with the regulatory protein Cdc6 and the replication factor Cdt1which both create the origin recognition complex. In this way they initiate DNA replication only once in every cell cycle (18,19). MCM proteins are very stable and active during all phases of the cell cycle, but their expression is reduced when the cell achieves the state of quiescence, differentiation, or senescence (20,21).

Due to the presence of MCM proteins in dividing cells and their absence in the dormant cells, it is possible to use that group of proteins as cell proliferative markers. Therefore, the assessment of MCM protein expression in a range of malignancies, has recently been the subject of research. The data obtained over the last few years showed a dysregulated pattern of MCM protein expression. Moreover, the level of expression of MCM2 protein in some cancers can be considered as an indicator of clinical outcome (22). Breast cancer, brain tumors, non-small cell lung cancer, prostate cancer and also bladder, esophageal, renal cell and oral squamous cell carcinoma belong to this group (23-29). Numerous studies point out that MCM proteins are also more specific and more sensitive than other typical proliferative markers (30,31). In most studies, multifactorial analysis reveals that the expression of MCM proteins is an adverse prognostic factor in reference to patients’ survival time.

In the field of dermatology MCM expression was analyzed in numerous skin conditions, i.e., in melanocytic nevi, cutaneous melanoma (32,33), Merkel cell carcinoma (34), T-cell lymphoproliferative skin disorders (35) or basal cell carcinoma (BCC) (36). In spite of well documented MCM expression in various human cancers, further investigations in a field of skin lesions including cSCC and AK are needed.

The aim of the study was to assess MCM2 protein expression in cSCC, AK and in the healthy skin. The research also included the evaluation of MCM2 expression depending on clinical and histological parameters. Moreover, the correlations between the expression of MCM2 and other proliferative markers, namely Ki-67 antigen and p53 protein were analyzed. All these steps are crucial to estimate the usefulness of the examination of MCM2 protein expression in assessing the stage of cSCC and AK in comparison to the other well-known proliferation markers.

Materials and Methods

Materials. The study involved 96 patients. On the basis of histological examination 22 lesions were diagnosed as AK and 57 as cSCC. In the group of patients with AK there were 17 women and 5 men aged 40-85 years (mean age 71.3±11.4 years). The average duration of the disease was 27.3±18.2 months. Erythematous type of AK was recognized in 8 patients (64%) and keratinized type in 14 patients (36%). In the group with cSCC diagnosis there were 23 women and 34 men aged 27-89 years (mean age 71.7±14.2 years) and the average duration of the disease was 255±47.1 months. Among all cSCC lesions keratinized (cSCC keratodes) (n=39; 68.5%), non-keratinized (cSCC akeratodes) (n=10; 17.5%) and mixed neoplasms, consisting of both SCC and BCC (n=8; 14%) were distinguished. In the terms of clinical picture, the study included 39 cases of hypertrophic cSCC and 17 cases of ulcerative cSCC. In one of the cases information about morphological type was not found. The control group consisted of 7 women and 10 men aged 12-82 years (mean age: 59.1±17.9 years). The tissue samples of the healthy skin were taken from patients undergoing plastic surgeries.

The research obtained ethical approval by the Wroclaw Medical University Ethics Committee (Consent no. KB-371/2010). Skin biopsies were taken from the patients after obtaining the informed consent. The material was fixed in 10% neutral buffered formalin and impregnated in paraffin. From each biopsy 4 μm thick sections were cut and stained by hematoxylin and eosin for histological examination.

Basic evaluations. AK lesions were divided in terms of clinical type into erythematous or keratinized. All cases of SCC were characterized by the assessment of: clinical and histological type, Broder Histological Classification, as well as intensity and depth of the inflammation. The inflammation was estimated as mild when only single inflammatory cells where visible, as moderate when numerous inflammatory cells were observed and as severe in the case of a massive inflammatory infiltrate. The depth of inflammation was evaluated on the basis of affecting the dermis or the dermis and subcutaneous tissue. Other data included in the study was: exposure to UV radiation, localization and size of the skin lesion, duration of the skin condition, age, and sex of the patients.

Immunohistochemistry. The sections of the tissue material samples were subjected to three separate immunohistochemical reactions using specific monoclonal antibodies in order to determine the expression of MCM2 protein, Ki -67 antigen and p53 protein. Paraffin-embedded tissue sections were deparaffinized in xylene and rehydrated in ethanol. Then they were incubated in hydrogen peroxide. The epitopes were revealed using heated citrate buffer solution for 20 min. After cooling, the slides were rinsed with running and distilled water. The activity of endogenous peroxidase was blocked with hydrogen peroxide. Subsequently slides were rinsed in phosphate-buffered saline. Consecutive sections were incubated using different, specific mouse monoclonal antibodies. Incubation was performed separately using mouse monoclonal anti-MCM2 antibodies (Clone CRCT21; code MCL, Dako, Glostrup, Denmark) at dilution 1:30 for 1 h, mouse monoclonal anti-Ki-67 antibodies (Clone MIB-1; code M7240, Dako) at dilution 1:50 for 30 min and mouse monoclonal anti-p53 antibodies (Clone DO-7, Dako) at dilution 1:50 for 30 min.

In the next steps the incubation with biotinylated antibodies (LSAB+, yellow colour, Dako), then with streptavidin-horseradish peroxidase complex (LSAB+, red colour, Dako) and finally with DAB chromogen was performed. The sections were counterstained with hematoxylin.

The evaluation of MCM2 protein, p53 antigen and Ki-67 expression was carried out under the light microscope. The percentage of positive cells in relation to the total number of cancer cells was assessed. Additionally, the average amount of positive cells for the MCM2 protein was computed in three different fields of each sample. The level of the expression of all three markers was evaluated using a scale from 0 to 3 points. The reaction was evaluated as negative – 0 (no immunostained cells), weak – 1 (≤25% immunostained cells), moderate – 2 (26-50% immunostained cells) or strong – 3 (>50% immunostained cells).

Statistical analysis. The obtained data were analyzed using Statistica 6.0 Windows Software. When comparing two groups of the patients, if the variables had a distribution close to normal, possible differences between analyzed groups were verified with Student’s t-test. In case of variables with a nonparametric distribution Mann-Whitney U-test was used. In case of comparing 3 groups, Kruskal-Wallis test was employed. The χ² test was used in comparison between qualitative variables. The quantitative dependencies between the analyzed parameters were proved with the Spearman’s rank correlation coefficient. p-Value of ≤0.05 was considered to be statistically significant.

Results

The average expression of MCM2 protein in cSCC and AK was significantly higher than in healthy skin. In cSCC it was 50.9%±19.6%, in AK 43.2%±17.3% and in healthy skin 35.0%±14.9% (p=0.01 for both comparisons) (Table I). Likewise, the level of MCM2 expression differed statistically significantly (p=0.02) between SCC, AK, and healthy skin (Figure 1). The strong expression was found in 47.4% (n=27) cases of cSCC, in 22.7% (n=5) cases of AK and only in 17.7% (n=3) samples of healthy skin. The moderate expression appeared in 40.3% (n=23), 63.6% (n=14) and 52.9% (n=9) while weak expression was revealed in 5.3% (n=3), 9.1% (n=2) and 29.4% (n=5) respectively. The expression was not found in 7% (n=4) cases of cSCC and in 4.6% (n=1) cases of AK (Figure 2).

Table I. Average expression of minichromosome maintenance protein 2 (MCM2) in cutaneous squamous cell carcinoma (cSCC), actinic keratosis (AK) and healthy skin.

Figure 1. Level of minichromosome maintenance protein 2 expression in cutaneous squamous cell carcinoma (cSCC), actinic keratosis (AK) and healthy skin.

Figure 2. Minichromosome maintenance protein 2 expression in cutaneous squamous cell carcinoma (cSCC), actinic keratosis (AK) and healthy skin. (a) Weak MCM2 expression in the healthy skin (≤25% immunostained cells), 200× magnification. (b) Strong MCM2 expression in the AK (>50% immunostained cells), 200× magnification. (c) Strong MCM2 expression in cSCC (>50% immunostained cells), 200× magnification.

MCM2 expression in healthy skin. All 17 healthy skin biopsies showed positive MCM2 expression. No statistically significant correlations were found between the level of MCM2 expression and the area where the biopsy was taken, sex or age of the patients were found.

MCM2 expression in AK. Statistically significant differences in expression of MCM2 protein between clinical subtypes of AK were not observed. Not only average expression, but also the level of MCM2 expression did not significantly differ in the tissue material from erythematous and keratinized type of AK. Similarly, there were no statistically significant differences in MCM2 expression in actinic keratosis lesions from skin exposed and unexposed to the ultraviolet radiation. Moreover, the study did not show any dependencies between MCM2 expression and diameter of the lesion, duration of the disease, localization of the skin lesion and patient’s age (details not shown). The average MCM2 expression in women with AK was 38.2%±15.7% while in men 59.0%±13.1% (p=0.02).

MCM2 expression in cSCC. The average expression of MCM2 protein in keratinized (keratodes) type of cSCC was 52.7%±19.2% and in non-keratinized (akeratodes) type was 41.3%±19.2%. In cases of mixed type, it amounted to 55.2%±21.3%. Relevant differences in the level of MCM2 protein expression depending on the histological type of cancer were not observed.

In cSCC keratodes type the lack of expression as well as a weak expression was found in 5.1% (n=2) of the patients. The moderate expression was noted in 41.1% (n=16) of the cases and the strong in 48.7%. In akeratodes cSCC the lack of expression and the weak expression was demonstrated in 10% (n=1), the moderate in 50% (n=5) and the strong in 30% (n=3) of the cases. The mixed type showed in 12.5% (n=1) the lack of expression, in 25% (n=2) moderate expression and in 62.5% (n=5) strong expression of MCM2 protein. The results also did not show a significant dependence between the average expression of MCM2 protein as well as level of the expression and the clinical type of cSCC.

The average expression in hyperkeratotic lesions was 52.7%±19.2% and in ulcerative lesions it was 41.3±19.2%. There were no significant differences in the expression of MCM2 protein on the basis of histological grading assessed by Broder’s Classification both in average and in the level of MCM2 protein expression. However, the study revealed a trend (p=0.06) for higher MCM2 expression in tumors with less intensive inflammation. Lower expression of MCM2 was observed in the tumors with severe intensity of the inflammation (mean 41.2%±19.2%), while in tumor with moderate intensity of inflammation the expression was higher (mean 49.3%±19.0%). The strongest expression (mean 60.0%±18.7%) was observed in tumors with mild intensity of the inflammation. Similar tendency was noticed in the level of the expression (details not shown).

The depth of inflammation process and the exposure of lesions to ultraviolet radiation did not affect the expression of MCM2 protein. The study did not show any correlation between MCM2 expression and the following parameters: size of tumor, duration of the disease, localization of the tumor and patient’s age (details not shown). However, we showed statistically significant dependence between patients’ sex and MCM2 protein expression (p=0.03). The average expression in men was 55.6%±18.1% whereas in women it was 44.0%±20.2%.

Correlation between the expression of MCM2 protein and Ki-67 and p53 antigen. In healthy skin there was no positive correlation between MCM2 protein expression and p53 or Ki-67 antigen. In the AK lesions the study revealed significant positive correlations between the expression of MCM2 protein and both antigens – Ki-67 and p53 (r=0.51, p=0.01; r=0.45, p=0.04 respectively). Analyzing different types of AK correlations between MCM2 protein expression and Ki-67 antigen in keratinized AK subtype was proved (r=0.6, p=0.02). No other relationships were found (detailed data not shown).

In cSCC correlation between MCM2 protein expression and Ki-67 antigen was positive, however it did not reach statistical significance (r=0.24, p=0.07). Nevertheless, in hyperkeratotic cSCC lesions MCM2 expression correlated significantly with Ki-67 expression (r=0.32, p=0.04). We were unable to show the correlation between the expression of MCM2 protein in cSCC biopsies and p53 antigen (details not shown).

Discussion

The early diagnosis and treatment of cancer is essential to improve patient prognosis. Some replication proteins are over-expressed in tumor cell lines what makes them a potential biomarker for diagnosing malignancies and predicting patient prognosis. Over the last years, researchers studied the mechanisms of DNA synthesis very precisely. MCM proteins are an essential component of this process, and their usefulness as new markers of cell proliferation was lately a subject of detailed studies (37). These proteins are very active in cycling cells, but they are not detected when the cell achieves the state of quiescence, differentiation, or senescence (38,39). This makes them more specific markers of cell proliferation then conventional biomarkers. Data on several types of carcinomas suggest that antibodies against MCM proteins allow to identify a greater number of cells per cycle then antibodies against Ki-67 (15). The expression of MCM proteins were thoroughly evaluated in range of malignancies in terms of patient management and survival outcomes according to various clinical and pathogenetic features (22-29). Likewise, in the dermatological field MCM proteins expression was investigated. Gambichler et al. (33) considered the MCM protein as a promising additive tool for distinguishing benign from malignant melanocytic skin lesions. The pattern of MCM expression was also estimated in nonmelanoma epithelial skin cancers and precancerous lesions, such as AK, Bowen disease or Bowenoid papulosis (36,40-41). The studies revealed the presence of this proteins not only in proliferative and malignant skin lesions but also in normal epidermis (39). Due to involvement of MCM proteins in DNA replication and cell cycle regulation they may be a good indicators of cancer cells and carcinogenesis. So far performed studies demonstrated that MCM protein expression can indicate dysplasia and preinvasive and invasive malignant tissue. Therefore, it can potentially be useful in investigating cutaneous SCC progression, which originates from an uncontrolled proliferation of atypical epidermal keratinocytes.

This study confirmed the expression of MCM2 protein in all specimens of healthy skin, which is in agreement with previous studies, which revealed the expression of MCM2 proteins in basal cell layer of the epidermis (36). Moreover, the expression of MCM2 protein was significantly higher in AK and cSCC compared to the control group. AK is considered as a pre-neoplastic lesion which can transform into invasive squamous cell carcinoma. So far carried out studies established increased level of MCM2 expression in this skin conditions. Shin et al. (40) proved usefulness of MCM2 protein as a reliable marker for atypical proliferating cells in AK. In their study MCM2 expression showed significant differences among adjacent grades of AK with positive correlation between the AK grade and the expression of this marker. The mean number of immunoreactive cells for MCM2 was higher in KIN II than in KIN I and in KIN III than in KIN II. Moreover, this study showed stronger correlation of MCM2 proteins with the grade of histological differentiation then in other proliferative markers: Ki-67 and proliferating cell nuclear antigen (PCNA). Additionally, Stojkovic-Filipovic et al. (42) confirmed this finding showing increased expression of MCM2, MCM5 and MCM7 protein from KIN1 to KIN3. Our results did not show correlation between MCM2 protein expression and clinical type, UV exposure, diameter of the lesion, duration of the disease, localization of the skin lesion or patient’s age. These clinical parameters were not included into analysis in previous studies.

We demonstrated MCM2 expression in 93% cases of cutaneous cSCC, what makes this result higher compared to 66.7% according to Abdou et al. (36). Freeman et al. (39) showed that in all types of squamous carcinoma included into their study (skin, cervix, lung, and esophagus) the expression of MCM2 protein varied with the grade of the tumor. Cases of moderately and poorly differentiated cSCC showed higher expression of MCM2 then well-differentiated cSCC. This tendency was not confirmed in subsequent research. Torres-Rendon et al. (43) demonstrated the lack of correlation between the grade of oral squamous cell carcinoma differentiation and the expression level of the biomarkers. In this study the expression of MCM2 was not only examined but also of Ki-67 and geminin protein. On the other hand, Stojkovic-Filipovic et al. (42) investigated the relationship between the level of MCM expression and cSCC thickness. They found significantly stronger correlation of MCM expression in cSCC cases thicker than 6 mm, that can imply association of MCM proteins with the high-risk factors in cSCC. The study also proved higher percentage of MCM expression in poorly differentiated cSCC then in well-differentiated cSCC. In the present study analysis of clinical and histological subtypes of SCC did not demonstrate a significant influence of these parameters on MCM2 expression. The expression of this marker in keratinized, non- keratinized and in mixed-type cancer (BCC/cSCC) did not differ significantly. There also was no difference in hyperkeratotic and ulcerative cSCC and despite one exception, no correlations between the protein expression and other examined features were found. There was a marked tendency for higher protein expression in tumors with less severe inflammation. The more severe the inflammation of the tumor, the lower the expression of the protein. Our study did not establish dependency between expression of the marker and Broder’s classification, depth of the infiltration, UV exposure, size, or localization of cSCC lesion. The lack of differences in expression of MCM2 protein according to mentioned above histological and clinical parameters does not allow to consider this protein as a significant prognostic marker of cSCC. A positive correlation of MCM2 expression with Ki-67 and p53 was stronger in AK. A weaker correlation was observed with regard to cSCC, what can be associated with the large diversity of the tumor in terms of histology.

In conclusion, we present a result of multifactorial analysis of MCM2 protein expression in AK and cSCC lesions. This is one of the few studies that takes into account such a large number of clinical and histological factors. In so far published research MCM2 protein has been recognized as a valuable prognostic marker in many malignancies. This aspect in the field of skin non-melanoma skin cancers is still not fully confirmed. Some results suggest MCM2 protein as a reliable marker for diagnosing and grading AK, others emphasize its association with high-risk prediction in SCC (40,42). Due to lack of dependency of MCM2 expression according to histological and clinical factors our findings do not allow to confirm this protein as a significant prognostic marker of cutaneous squamous cell carcinoma.

Conflicts of Interest

The Authors declare no conflicts of interest.

Authors’ Contributions

Conceptualization, A.R., K.Ś., P.D., A.P. and J.S.; methodology, A.R., K.Ś., P.D., A.P. and J.S.; software, A.R. and A.P; validation, A.R., K.Ś., P.D., A.P. and J.S.; formal analysis, A.R., K.Ś., P.D., A.P. and J.S; investigation, A.R., K.Ś., P.D., A.P. and J.S; resources, A.R.; data curation, A.R. and A.P. ; writing—original draft preparation, A.R. and K.Ś.; writing—review and editing, A.R., K.Ś., P.D., A.P. and J.S.; visualization, A.R., K.Ś., P.D., A.P. and J.S.; supervision, P.D. and J.S.; project administration, P.D. and J.S.; funding acquisition, P.D. and J.S.; All Αuthors have read and agreed to the published version of the manuscript.