Abstract
Background/Aim: The High-Mobility Group A1 (HMGA1) protein has been implicated in human malignancies, playing an important role in cancer proliferation, angiogenesis and metastasis. Increased HMGA1 expression has been found in skin mouse tumors, whereas Hmga1-null mice were protected against skin carcinogenesis. Here, we examined the expression of HMGA1 in human skin tumors, squamous cell carcinoma and basal cell carcinoma. Materials and Methods: Tumor and normal skin tissues from 15 affected patients were surgically excised, and mRNA and protein extraction was performed. mRNA and protein content for both HMGA1 and MMP-11, a proteinase enzyme that plays a role in tumor development and progression, was measured by real-time PCR and western blotting, respectively. Data were analyzed by the SPSS software. Results: HMGA1 mRNA and protein expression patterns were higher in neoplastic skin lesions, compared to normal skin (p<0.001). Similar results were observed for MMP-11. Conclusion: Our data confirm previous observations in mice studies, and suggest that HMGA1 and MMP-11 may play a key role in the proliferation and progression of skin tumors in humans.
The High Mobility Group A1 (HMGA1) protein is a non-histone nuclear factor that interacts with the narrow minor groove of AT-rich regions of DNA, modulating cell cycle-related chromosomal changes, DNA replication and repair, molecular chaperoning function, and transcription of many genes (1, 2). In recent years, HMGA1 has also emerged as a novel transcriptional regulator of glucose homeostasis and metabolism (3-7), acquiring a growing interest in this context, being involved in many aspects of human health and disease (8), including insulin-resistant type-2 diabetes mellitus (3, 9), and the metabolic syndrome (10, 11). In addition, HMGA1 is causally linked to neoplastic transformation and tumor development, its expression strongly correlates with metastatic potential of certain tumors, and it is thought to be a hallmark of a wide variety of cancers (12). For example, an increased expression of HMGA1 has been reported to correlate with the grade of malignancy of murine skin carcinoma (13). Consistently with this finding, protection against chemical-induced skin carcinogenesis was reported in Hmga1-null mice (14). However, although skin cancers are the most common cancers in humans (15, 16), to date, no data exist in current literature addressing the relationship between HMGA1 expression and human skin cancer.
Skin can be affected by both melanoma cancer (MC) and non-melanoma cancer (NMC). NMCs include basal cell carcinoma (BCC), and squamous cell carcinoma (SCC). The incidence of NMC is 18-20 times higher than that of MC (15, 16); as a result, health-related costs for NMC in U.S.A. are 6-7 times greater than those for treating MC (17), thus becoming a serious challenge for the public health system. BCC originates from stem cells in the hair follicle infundibulum or interfollicular epidermis and is the most common skin cancer worldwide, accounting for almost 90% of skin tumors (18). It does not metastasize, thus is not lethal, but it is locally invasive and can cause facial disfigurement for its frequent facial localization (19). SCC accounts for about 20% of skin cancers (20), and is responsible for the majority of deaths caused by NMC (21). Although most SCCs can be removed surgically, over 10% of them tend to be more aggressive with a greater tendency to metastasize (21).
Although our understanding on the pathogenesis of NMC has increased in the last years, further studies about the molecular mechanisms underlying the pathogenesis of these tumors are still necessary in order to improve prevention and treatment. In line with this need, several lines of research have been recently directed at investigating the role of matrix metalloproteinases (MMPs) in NMC. MMPs are zinc-containing endopeptidases able to degrade various components of extracellular matrix proteins. They are up-regulated in the skin in response to multiple stimuli, including UV radiation, oxidative stress and inflammation (22). By affecting various processes that are related to tumor initiation/progression, growth and angiogenesis, MMPs have been involved in photocarcinogenesis of NMC (BCC and SCC) (23), whose treatment options may include MMPs specific inhibitors (23, 24).
In view of the above considerations, and on the basis of evidence that a link between HMGA1 and members of the MMP family exists during the neoplastic transformation process (25), here we examined the expression of HMGA1 and MMP-11 in tumoral and normal tissues of patients with NMC.
Patients and Methods
Patients and samples. A total of 15 consecutive, unrelated subjects with NMC (10 BCC and 5 SCC) were enrolled between September 2016 and January 2017 at the “Operative Unit of Plastic Surgery”, University of Catanzaro, Italy. All subjects came from Calabria, Southern Italy, a region with temperate climate conditions, mostly sunny during the year. Clinical measurements and biochemical analyses were carried out before surgical intervention. Fasting glycemia, total and HDL-cholesterol, triglycerides, creatinine, lactate dehydrogenase and erythrocyte sedimentation rate were obtained in all patients with no caloric intake for at least 8 h. Both tumor and surrounding normal tissues were obtained from all subjects, and each specimen was divided into two equal parts, one of which was snap frozen in liquid nitrogen for protein and RNA extraction, while the other was analyzed by histopathological examination and immunohistochemical analysis to confirm the diagnosis. The study was approved by the local ethics committee, Regione Calabria Comitato Etico Sezione Area Centro (protocol registry n. 116 of May 14, 2015), and informed consent was obtained from patients before surgery.
mRNA and protein extraction. Both total RNA and protein were extracted from healthy and tumor tissues, using the AllPrep DNA/RNA/Protein Mini kit according to the manufacturer's instructions (Qiagen, Valencia, CA, USA). Briefly, 25 mg of tissue sample was homogenized in RLT buffer (supplied with the kit), the lysate was transferred into an RNeasy spin column for RNA binding, and the flow-through waste was collected in a tube for recovering of total proteins, followed by elution of RNA from the column.
Real-time PCR and immunoblot analysis. RNA abundance was measured by a NanoDrop spectrophotometer (Thermo Fisher Scientific, Inc., Waltham, MA, USA), and its quality confirmed on agarose gel. cDNA was synthesized from 1 μg of total RNA using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA), following the manufacturer's instructions. A real-time thermocycler (Eppendorf Mastercycler ep realplex ES) was used to perform quantitative PCR. SYBR Green fluorescence was measured, and relative quantification was made against the RPS9 cDNA, used as an internal standard (5). PCR primers for human HMGA1, RPS9, and MMP-11 were previously reported (26, 27). All PCR reactions were done in triplicate.
Final protein concentration was determined using the Bradford's method. As the tissue biopsy specimens were very small in size, a sufficient amount of protein extract was obtained only from four tumors and relative control tissues. For HMGA1 protein expression, western blot (WB) analysis was performed as previously described (28), using a polyclonal-specific antibody against HMGA1 (29). WB for MMP-11 was performed as previously reported, by employing an anti-MMP-11 specific antibody (27).
Statistical analysis. Initially, each quantitative trait was tested for normality of distribution, using the Shapiro-Wilk normality test. Continuous values between two groups were compared by using the non-parametric Mann-Whitney test. The Spearman's rank correlation analysis was used to explore the correlations between HMGA1 and MMP-11 levels and clinical, biochemical and cancer features. Instead, linear regression analysis was employed to measure the association of HMGA1 mRNA levels and MMP-11 mRNA amounts. A p-value of <0.05 (two-tailed) was considered statistically significant. Data were analyzed with SPSS 20.0 software (SPSS Inc., Chicago, IL, USA).
Results
Characteristics of the study group. As reported in Table I, the enrolled population included six females and nine males, with ages ranging from 56 to 92 years, and a body mass index between 21.7 and 29.3 Kg/m2. Four of them were affected by type 2 diabetes, twelve had hypertension and eight underwent hypolipidemic treatment (Table I). Skin tumors were mostly localized in the head; ten at the face region and four in the scalp. The maximum diameter of lesion ranged from 5 to 80 mm, and none of them was treated with chemotherapy. When clinical and biochemical data from patients with BCC were compared with those obtained from patients with SCC, no significant differences were observed, nor among continuous variables, neither among categorical values (Table II).
HMGA1 mRNA expression. HMGA1-specific transcripts were detectable in both cancer and normal adjacent tissues, as revealed by RT-PCR. As illustrated in Figure 1A and B, HMGA1 mRNA expression was higher in tumor tissue from both BBC and SCC when compared with that of paired normal skin tissues (p<0.001). Of note, when HMGA1 mRNA abundance in BCC was compared with that in SCC, mRNA levels in the latter were higher (p<0.001, Figure 1B), thus suggesting the hypothesis that HMGA1 expression may directly correlate with tumor aggressiveness.
Clinical and histopathological features of enrolled patients.
MMP-11 mRNA expression. MMP-11 mRNA levels were simultaneously measured. As shown in Figure 2A and B, MMP-11 mRNA expression was considerably higher in tumor tissues from both BBC and SCC compared to respective normal skin samples (p<0.001). As for the expression of HMGA1, MMP-11 mRNA abundance in SCC was significantly higher than that in BCC (p<0.001, Figure 2B), thus supporting the notion that up-regulation of both HMGA1 and MMP-11 in NMC may play a role in tumor invasion and migration.
HMGA1 and MMP-11 protein levels. HMGA1 and MMP-11 protein expression was therefore assessed in extracts from BCC and SCC and compared to that from the surrounding normal tissue. As Shown in Figure 3, HMGA1 and MMP-11 protein expression levels were higher in BCC and SCC than in adjacent non-cancerous control tissue. Also in this case, the levels of these two proteins were higher in SCC as compared to BCC (p<0.05), thus paralleling the results obtained with mRNA expression analysis.
Determinants of the HMGA1 and MMP-11 expression. The Spearman univariate correlation analysis was employed to correlate HMGA1 and MMP-11 mRNA expression levels with either clinical, biochemical or tumor features. All continuous variables indicated in Tables I and II were tested. The only correlation observed was that between HMGA1 and MMP-11 mRNA and protein expression values (ρ=0.788, p<0.0001). This association was confirmed in a linear regression analysis (t=9.554, p<0.0001), after log-transformation of the values, including HMGA1 as independent variable and MMP-11 as dependent variable. Also, the value of adjusted R-squared (R2=0.866) indicated that the variability of MMP-11 was well related to the variability of HMGA1. The inclusion of possible confounders (e.g. sex, age, BMI) in the regression model did not modify the results.
Comparison of clinical and biochemical features among BCC and SCC groups.
HMGA1 gene expression in BCC and SCC samples and adjacent normal tissue. A) HMGA1 mRNA levels in tumor tissue of each patient (gray bars), as measured by qRT-PCR. Levels are expressed as fold of increment with respect to each relative control (black bars), which is assigned an arbitrary value of 1. B) Comparison of HMGA1 mRNA levels in tumor tissue from skin BCC (n=10) and SCC (n=5). (CTRL, control). Results are the mean±s.d. of three independent measurements from each patient. *p<0.05 vs. control; **p<0.05 vs. BCC tumor tissue.
Discussion
The role of HMGA1 in tumorigenesis is well supported (12). In fact, HMGA1 positively regulates a variety of genes involved in tumor growth, invasion, migration, neoangiogenesis, epithelial-mesenchymal transition and cancer metastasis (12). Also, overexpression of HMGA1 occurs in a wide range of human cancers, including breast, lung, thyroid, bladder, prostate, pancreas, stomach, colon, kidney, uterus, and hepatocellular carcinomas, as well as tumors of the hematopoietic system (12). For the first time in the present study, we show that overexpression of HMGA1 also occurs in skin tumors, specifically in BCC and SCC non-melanoma skin cancers.
The incidence of NMCs is increasing, perhaps because of the increased UV exposure, which can provide an explanation for the frequent localization of these tumors in the face and neck, two areas of the body that receive heavy UV exposure (19, 30, 31). The worldwide trend toward a significant increase in the older populations may represent another critical factor influencing growth and development of NMCs (32). It is well known that age-related skin changes, due to both genetic and environmental factors, can profoundly influence cell cycle and apoptosis, DNA stability and repair, and cellular metabolism (33). In this regard, it must be noted that an involvement of HMGA1 in DNA repair has been reported (34). In particular, it has been demonstrated that overexpression of HMGA1 predisposes to tumorigenesis by repressing the xeroderma pigmentosum complementation group A gene (XP-A gene), which encodes for a nuclear protein involved in DNA excision repair (34). Therefore, it is tempting to hypothesize that overexpression of HMGA1, as documented in our study, may compromise the repair of UV-induced damaged DNA, thereby promoting neoplastic transformation of the skin.
MMP-11 gene expression in BCC and SCC samples and adjacent normal tissue. A) MMP-11 mRNA levels in tumor tissue of each patient (white bars), as measured by qRT-PCR. Levels are expressed as fold of increment with respect to each relative control (black bars), which is assigned an arbitrary value of 1. B) Comparison of MMP-11 mRNA levels in tumor tissue from skin BCC (n=10) and SCC (n=5). (CTRL, control). Results are the mean±s.d. of three independent measurements from each patient. *p<0.05 vs. control; **p<0.05 vs. BCC tumor tissue.
HMGA1 and MMP-11 protein expression in BCC and SCC samples and adjacent normal tissue. HMGA1 and MMP-11 protein expression was performed in extracts of surrounding normal tissue (CTRL, n=4) and tumor tissues from BCC (n=2) and SCC (n=2). Representative WBs of HMGA1 and MMP-11 out of three performed for each condition are shown in the autoradiograms. Bar graphs above the autoradiograms are derived from densitometric scanning of WBs, using the ImageJ software program. Data are mean±s.d. of three independent experiments. β-Actin, control of protein loading. *p<0.05 vs. control.
In addition, here we demonstrate that MMP-11 is also up-regulated in NMCs compared to the surrounding normal tissue. Interestingly, our findings also reveal that both HMGA1 and MMP-11 are higher in SCC than in BCC. Previous studies in this context have demonstrated an increased expression of some MMPs in NMC (22). In particular, overexpression of MMP-1 has been reported at the invasive front of BCC (35), and it has been associated with the initial steps of tumor growth in SCC (36). Excessive expression of MMP-2 has been detected in the stroma of SCC compared with BCC, suggesting a role for this MMP in the different pattern of invasion found in these two tumors (37). MMP-9 was found in the stromal fibroblasts surrounding the tumor invasion sites in both infiltrating BCC and SCC (38), while another study reported an elevation of both MMP-9 and MMP-2 in SCC versus BCC (39). Other studies have proved that MMP-13 can be involved in BCC and SCC angiogenesis with different mechanisms (40, 41). Therefore, all these evidence clearly indicates that MMPs might have a crucial role in the regulation of growth and development of NMC, letting hypothesize that factors produced by UV-damaged skin stromal cells are critical for tumorigenesis (42). Several studies, including one of our own, indicate that HMGA1 seems to be involved in the up-regulation of some MMPs, in particular MMP-2, MMP-9 and MMP-11 (27, 43-46). By positively affecting the expression of MMPs, our data in the present study may contribute to provide a mechanistic hypothesis for the role of HMGA1 in BCC and SCC tumorigenesis. Interestingly, our findings indicate that MMP-11 is much more expressed in SCC than in BCC, and this could have a role in the major aggressiveness of SCC and its metastasizing ability. If this observation will be further confirmed, MMP-11 could be used as marker to distinguish self-limiting skin tumors from more aggressive non-melanoma skin cancers.
As a limitation of this work, the small sample size of the study must be pointed out, which does not allow to exclude definitely a β error in correlation analysis. Also, due to the small quantity of specimens collected, expression analysis of other genes, as well as proteins, was not performed in all tissues. In conclusion, our data indicate that HMGA1 is overexpressed in NMCs, and this increase parallels the increase in MMP-11 expression. Although further studies are necessary to confirm and extend these data, our findings provide two new hallmarks of NMC and suggest possible adjunctive pathogenetic mechanisms for these tumors.
Footnotes
↵* These Authors contributed equally to this study.
Conflicts of Interest
The Authors do not have any conflicts of interest to disclose.
- Received November 14, 2017.
- Revision received December 8, 2017.
- Accepted December 11, 2017.
- Copyright© 2018, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
