Analysis of Changes in the Expression of Selected Genes from the ABC Family in Patients with Triple-Negative Breast Cancer

Triple-negative breast cancer (TNBC) is characterized by a lack of expression of hormone receptors (estrogen and progesterone), as cancer cells also do not overexpress the HER2 receptor. Due to their molecular profile, treatments for this type of breast cancer are limited. In some cases, the pharmacotherapy of patients with TNBC is hindered by the occurrence of multidrug resistance, which is largely conditioned by proteins encoded by genes from the ABC family. The aim of our study was to determine the expression profile of 14 selected genes from the ABC family using real-time PCR in 68 patients with TNBC by comparing the obtained results with clinical data and additionally using bioinformatics tools (Ualcan and The Breast Cancer Gene Expression Miner v4.8 (bc -GenExMiner v4.8)), as well as by comparing experimental data with data in the Cancer Genome Atlas (TCGA) database. Based on the conducted studies, we found different levels of gene expression depending on the age of patients, tumor sizes, metastases to lymph nodes, cell infiltration into adipose tissue, tumor stages, or lymphovascularinvasion. The results of the presented studies demonstrate the effect of the expression level of the studied genes on the clinical course and prognosis of patients with TNBC, and suggest how profiling the expression level of genes from the ABC family may be a useful tool in determining personalized TNBC treatment.


Introduction
Triple-negative breast cancer (TNBC) is characterized by a negative expression of the progesterone receptor (PR), the estrogen receptor (ER), and human epidermal growth factor receptor 2 (HER2) [1]. TNBC, as one of the molecular subtypes of breast cancer (BC), accounts for about 10-20% of all diagnosed cases [2][3][4][5]. TNBC is most often diagnosed in premenopausal women under 40 years of age [6]. TNBC is an aggressive type of BC; has a worse prognosis than other molecular subtypes of breast cancer; and is characterized by earlier recurrences and metastases, shorter patient survival, and high mortality [7][8][9][10]. Currently, the basic method of treating TNBC is chemotherapy; due to the lack of expression of ER, PR, and HER2 receptors, hormonal therapy and targeted molecular therapy do not show satisfactory results [11]. Unfortunately, many patients who receive chemotherapy develop drug resistance, which is a major obstacle to successful cancer treatment. Resistance to the applied cytostatics may account for up to 90% of cases of ineffective chemotherapy [12,13].
Multidrug resistance (MDR) is defined as the insensitivity of cancer cells to the applied chemotherapeutic drugs, which differ in their mechanism of action, structural structure, and target point. MDR is one of the main causes of chemotherapy failure and is responsible In patients with triple-negative breast cancer, the highest mean expression level compared to the normal tissue surrounding the tumor was noted for the ABCC3 gene (M = 0.038216) (Figure 1a), while the lowest mean expression level was recorded for the ABCC6 gene (M = −0.949418) (Figure 1a). For all marked genes from the ABC family, a reduced average level of expression was observed compared to the control, except for the ABCC3 gene (M = 0.038216) (Figure 1a). Descriptive statistics for genes are presented in Table 1.
gene (M = −0.949418) (Figure 1a). For all marked genes from the ABC family, a red average level of expression was observed compared to the control, except for the AB gene (M = 0.038216) (Figure 1a). Descriptive statistics for genes are presented in Tab   Data obtained from the TCGA database also showed a statistically significant lowered average expression level of the ABCB1, ABCC4, ABCC6, and ABCG2 genes in patients with breast cancer compared to the control group (Figures 1 and 2). In contrast to the average expression level of the ABCC3 gene ( Figure 2) obtained during the experiments, the data obtained from the TCGA database showed a decrease in the average expression level of this gene, but these values were not statistically significant compared to the control group. The average expression levels of the ABCA3, ABCB4, ABCB9, ABCC1, ABCC5, and ABCC10 genes (Figures 1 and 2) obtained from the TCGA database were elevated compared to normal tissue, which was not observed in the results of the experimental part. Inconsistencies in the data obtained experimentally in comparison with the data deposited in the TCGA may result from differences in the expression levels of estrogen, progesterone, and HER2 receptors in cancers in patients from the study group.
2.2. The Relationship between the Selected Genes from the ABC Family and a Comparison of the Results Obtained from the Experimental Part with the Bioinformatics Analysis Data Results from the TCGA and SCAN-B Database A correlation analysis of the mean expression values of selected genes from the ABC family showed that there is a statistically significant positive correlation between all the tested genes. The strongest positive correlations occur between the genes ABCC1 and ABCA2 (r = 0.72), ABCC2 and ABCA2 (r = 0.75), ABCC5 and ABCA2 (r = 0.72), ABCG2 and ABCA2 (r = 0.70), ABCC5 and ABCA3 (r = 0.73), ABCC1 and ABCB9 (r = 0.72), ABCC1 and ABCC10 (r = 0.71), ABCC4 and ABCC1 (r = 0.80), ABCC5 and ABCC1 (r = 0.81), ABCG2 and ABCC1 (r = 0.74), ABCC5 and ABCC4 (r = 0.71), and ABCG2 and ABCC5 (r = 0.74). The lowest values of the correlation coefficient (r = 0.19) were found for the ABCC3 and ABCB9 genes and the ABCC4 and ABCC11 genes ( Figure 3a).
Data obtained from the TCGA and the SCAN-B database showed a statistically significantly higher average expression level of the ABCG2 (p = 0.0465), ABCC11 (p < 0.0001), ABCC6 (p < 0.001), ABCC5 (p = 0.0012), and ABCB9 (p = 0.0152) genes in patients over 51 years of age than in younger patients, while for the ABCC10 (p < 0.0001), ABCC4 (p < 0.0001), ABCC1 (p = 0.0087), and ABCB1 (p = 0.0014) genes, the mean expression level was higher in patients 51 years of age or younger ( Figures S1-S3). Data on the ABCB9 and ABCB1 genes confirmed our results, while for the ABCC4, ABCC6, and ABCC11 genes, the average expression levels depending on the age group generated from TCGA and the SCAN-B database differed from the experimental data. Differences in the obtained results may be due to the different sizes of the study groups or different age selection criteria when separating the groups for analysis.

Invasion of the Fat Tissue
Statistical analysis performed with the U Mann-Whitney test showed a statistically significantly higher expression level of the ABCB9 (p = 0.000), ABCC1 (p = 0.009), and ABCC6 (p = 0.000) genes in patients without tumor cell invasion into fat tissue. The average expression level of the ABCA2 (p = 0.000), ABCB1 (p = 0.000), and ABCG2 (p = 0.001) genes was statistically significantly higher in patients with diagnosed tumor cell infiltration into fat tissue compared to patients without tumor cells in the fat tissue ( Figure 5).   In the TCGA database, no information was found on the expression level of the examined genes depending on the presence or absence of tumor cell infiltration in fat tissue.

Lymphovascular Invasion
The analysis performed with the U Mann-Whitney test showed that the average expression level of the ABCA2 (p = 0.01) and ABCB1 (p = 0.001) genes was statistically significantly higher in patients associated with the lymphatic vessel infiltration of cancer cells. The average expression level of the ABCA3 (p = 0.007), ABCB9 (p = 0.000), ABCC1 (p = 0.000), ABCC4 (p = 0.000), ABCC5 (p = 0.000), and ABCC6 (p = 0.000) genes was statistically significantly higher in patients without tumor cell infiltration in the lymph vessels ( Figure 6).

Lymphovascular Invasion
The analysis performed with the U Mann-Whitney test showed that the average expression level of the ABCA2 (p = 0.01) and ABCB1 (p = 0.001) genes was statistically significantly higher in patients associated with the lymphatic vessel infiltration of cancer cells. The average expression level of the ABCA3 (p = 0.007), ABCB9 (p = 0.000), ABCC1 (p = 0.000), ABCC4 (p = 0.000), ABCC5 (p = 0.000), and ABCC6 (p = 0.000) genes was statistically significantly higher in patients without tumor cell infiltration in the lymph vessels ( Figure  6).  The statistical analysis of data from the TCGA and the SCAN-B database showed a statistically significant increase in the expression of the ABCB9 (p < 0.0001), ABCC2 (p = 0.0114), ABCC3 (p < 0.0001), ABCC5 (p < 0.0001), ABCC10 (p = 0.0269), and ABCC11 (p < 0.0001) genes in patients with lymphovascular invasion (nodal status). The data obtained experimentally for the ABCB9 and ABCC3 genes differ from the data generated from the TCGA and the SCAN-B database ( Figures S4-S6).

The Scarff-Bloom and Richardson (SBR) Grading System
The H Kruskal-Wallistest, together with the analysis of multiple comparisons, showed statistically significant differences in the expression of the ABCA2, ABCA3, ABCB1, and ABCC5 genes in patients from the SBR1 and SBR2 groups and the SBR1 and SBR3 groups, while there was no statistically significant difference in the patients classified into the SBR2 and SBR3 groups. Statistically significant differences in the expression of the ABCB4 gene were noted in patients from the SBR1 and SBR2 groups and the SBR2 and SBR3 groups. In the case of the average value of ABCC10 gene expression, statistically significant differences were observed among patients from the SBR2 and SBR3 groups. For the mean value of ABCC4 gene expression, statistically significant differences were observed in patients from the SBR1 and SBR2 groups, the SBR1 and SBR3 groups, and the SBR2 and SBR3 groups. Statistically significant differences in the mean expression of ABCC11 and ABCC2 genes were observed in patients from the SBR1 and SBR2 groups and the SBR2 and SBR3 groups (Table 2, Figure 7).

Metastases to the Lymph Nodes
Statistical analysis showed statistically significant differences in the expression level the ABCA3, ABCB9, and ABCC4 genes in patients classified into the pN0 and pN2 groups, the pN0 and pN3 groups, the pN1 and pN2 groups, and the pN1 and pN3 groups. Statistically significant differences were also observed in patients classified into the pN0 and pN2 groups, the pN0 and pN3 groups, the pN1 and pN2 groups, the pN1 and pN3 groups, and the pN2 and pN3 groups in terms of the expression level of the ABCC11, ABCC1, ABCC2, and ABCC6 genes. Statistically significant differences in the expression levels of the ABCB1, ABCC3, and ABCC5 genes were found in patients from the pN0 and pN3 groups, the pN1 and pN3 groups, and the pN2 and pN3 groups. The H Kruskal-Wallis test with multiple comparisons also showed statistically significant differences in the level of ABCB4 gene expression in patients from the pN0 and pN1 groups, the pN0 and pN2 groups, the pN0 and pN3 groups, the pN1 and pN3 groups, and the pN2 and pN3 groups. For the ABCC10 gene, a significant statistical difference in the level of expression occurred in patients from the pN0 and pN1 groups and the pN0 and pN3 groups. The ABCA2 gene expression values differed statistically significantly in patients in the pN0 and pN1 groups, the pN0 and pN2 groups, the pN1 and pN2 groups, the pN1 and pN3 groups, and the pN2 and pN3 groups. For the ABCG2 gene, statistical significance was demonstrated in the pN0 and pN2 groups, the pN0 and pN3 groups, the pN1 and pN3 groups, and the pN2 and pN3 groups (Figure 8, Table S1). groups depending on the presence of metastases to lymph nodes (pN0-no metastases to the regional lymph nodes, pN1-micrometastases or metastases in 1-3 axillary lymph nodes, pN2-metastases in 4-9 axillary lymph nodes, and pN3-metastases in 10 or more axillary lymph nodes).
Statistical analysis of data obtained from TCGA confirmed statistically significant differences in the expression level of the ABCB1 and ABCC3 genes in patients classified into pN0 and pN3 groups, the ABCA2 gene in patients from the pN0 and pN1 groups, the ABCB9 gene in patients from the pN0 and pN3 groups, the ABCC4 gene in patients from the pN0 and pN3 groups and the pN1 and pN3 groups, and the ABCC6 gene in patients from the pN0 and pN3 groups and the pN1 and pN3 groups. In addition, the analysis using the Ualcan online tool showed statistically significant differences in the expression level of the ABCB9 gene in patients from the pN0 and pN1 groups and the pN1 and pN2 groups; the ABCC3 gene in patients from the pN0 and pN1 groups and the pN1 and pN2 groups; and the ABCC5 gene in patients from the pN0 and pN1 groups, the pN0 and pN2 groups, and the pN1 and pN2 groups, which was not shown in the experimental part (Table S2).

Primary Tumor Size
The analysis showed statistically significant differences in the mean expression of the ABCA2, ABCA3, ABCB4, and ABCC3 genes in patients classified into the T1 and T3 groups and the T2 and T3 groups. Differences in the mean expression values of the ABCB1, ABCC2, ABCC5, ABCC6, and ABCG2 genes were statistically significant for patients from the T1 and T2 groups and the T1 and T3 groups. In the case of the ABCB9 and ABCC4 genes, there was a statistically significant difference in the level of expression in patients belonging to the T2 and T3 groups. The difference in the mean expression level of the ABCC10 gene was statistically significant in the case of patients from the T1 and T2 groups and the T2 and T3 groups, while a significant difference in expression for the ABCC11 Figure 8. Mean expression values of selected genes from the ABC family in patients classified into 4 groups depending on the presence of metastases to lymph nodes (pN0-no metastases to the regional lymph nodes, pN1-micrometastases or metastases in 1-3 axillary lymph nodes, pN2metastases in 4-9 axillary lymph nodes, and pN3-metastases in 10 or more axillary lymph nodes).
Statistical analysis of data obtained from TCGA confirmed statistically significant differences in the expression level of the ABCB1 and ABCC3 genes in patients classified into pN0 and pN3 groups, the ABCA2 gene in patients from the pN0 and pN1 groups, the ABCB9 gene in patients from the pN0 and pN3 groups, the ABCC4 gene in patients from the pN0 and pN3 groups and the pN1 and pN3 groups, and the ABCC6 gene in patients from the pN0 and pN3 groups and the pN1 and pN3 groups. In addition, the analysis using the Ualcan online tool showed statistically significant differences in the expression level of the ABCB9 gene in patients from the pN0 and pN1 groups and the pN1 and pN2 groups; the ABCC3 gene in patients from the pN0 and pN1 groups and the pN1 and pN2 groups; and the ABCC5 gene in patients from the pN0 and pN1 groups, the pN0 and pN2 groups, and the pN1 and pN2 groups, which was not shown in the experimental part (Table S2).

Primary Tumor Size
The analysis showed statistically significant differences in the mean expression of the ABCA2, ABCA3, ABCB4, and ABCC3 genes in patients classified into the T1 and T3 groups and the T2 and T3 groups. Differences in the mean expression values of the ABCB1, ABCC2, ABCC5, ABCC6, and ABCG2 genes were statistically significant for patients from the T1 and T2 groups and the T1 and T3 groups. In the case of the ABCB9 and ABCC4 genes, there was a statistically significant difference in the level of expression in patients belonging to the T2 and T3 groups. The difference in the mean expression level of the ABCC10 gene was statistically significant in the case of patients from the T1 and T2 groups and the T2 and T3 groups, while a significant difference in expression for the ABCC11 gene was demonstrated between patients from all groups, separated on the basis of tumor size (Figure 9, Table 3). gene was demonstrated between patients from all groups, separated on the basis of tumor size ( Figure 9, Table 3). Figure 9. The average expression level of the tested genes from the ABC family in patients classified into the T1, T2, or T3 groups, depending on the size of the primary tumor (T1-patients with a primary tumor size ≤ 20 mm, T2-patients with a primary tumor size > 20 mm but ≤50 mm, and T3-patients with a primary tumor size greater than 50 mm).  The Cancer Genome Atlas database does not contain information on the expression level of the tested genes from the ABC family in classified patients according to the size of the primary tumor.

Discussion
Breast cancer is characterized by high heterogeneity in terms of the clinical course, prognosis, method of treatment, and histopathological features, which is why it is so important to determine the subtype of the cancer before choosing a treatment [28]. The selection of the appropriate therapy in breast cancer is based on a combination of the cancer staging system, the patient's age at diagnosis, the tumor histotype and stage, the hormone receptor status, and the TNM classification. The implementation of effective treatment with a low level of toxicity in patients requires the development of selective and individualized therapies based on the molecular and clinical features of the tumor. The development of such therapies should be based on the knowledge of the benefits and potential toxic effects of each therapeutic regimen [29].
One of the greatest challenges of 21st century medicine is the selection of appropriate pharmacotherapy that would allow for the remission of the disease. One of the reasons Figure 11. Survival curves (Kaplan-Meier plotter) including an assessment of the prognostic significance of the ABCC4 (a), ABCC5 (b), ABCC6 (c), ABCC10 (d), ABCC11 (e), and ABCG2 (f) genes in patients with breast cancer (different molecular subtypes). The data were generated using the online tool Ualcan.

Discussion
Breast cancer is characterized by high heterogeneity in terms of the clinical course, prognosis, method of treatment, and histopathological features, which is why it is so important to determine the subtype of the cancer before choosing a treatment [28]. The selection of the appropriate therapy in breast cancer is based on a combination of the cancer staging system, the patient's age at diagnosis, the tumor histotype and stage, the hormone receptor status, and the TNM classification. The implementation of effective treatment with a low level of toxicity in patients requires the development of selective and individualized therapies based on the molecular and clinical features of the tumor. The development of such therapies should be based on the knowledge of the benefits and potential toxic effects of each therapeutic regimen [29].
One of the greatest challenges of 21st century medicine is the selection of appropriate pharmacotherapy that would allow for the remission of the disease. One of the reasons breast cancer chemotherapy fails is the occurrence of the phenomenon of multidrug resistance associated with proteins encoded by genes from the ABC family [30].
In our study, we showed that the expression level of the ABCA2, ABCA3, ABCB1, ABCB4, ABCB9, ABCC1, ABCC2, ABCC4, ABCC5, ABCC6, ABCC10, ABCC11, and ABCG2 genes was reduced in relation to the normal tissue surrounding the tumor (control) and only the average level of expression ABCC3 gene was elevated compared to controls (Figures 1 and 2). Analyzing clinical data, we showed that an expression of the ABCA3, ABCB9, ABCC2, and ABCC4 genes was statistically significantly higher in patients over 50 years of age (Figure 4). A higher expression level of the ABCA2, ABCB1, and ABCG2 genes was found in patients with a confirmed invasion of neoplastic cells into fat tissue ( Figure 5), and the expression of the ABCA2 and ABCB1 genes was higher in patients with lymphovascula invasion (Figure 6). Statistically significant differences in the expression of the examined genes were also observed in patients classified into groups depending on the size of the primary tumor, metastases to lymph nodes, or the SBR scale (Figures 7-9).
The expression level of the ABCB1 gene that we determined confirms the results obtained by João Marcos de Azevedo Delou and colleagues. Researchers characterized changes in the expression level of the ABCB1 gene in a breast tumor compared to benign breast tissue surrounding the tumor. In total, 712 women were qualified for the study, 62 of whom were diagnosed with triple-negative breast cancer. The analysis of the results showed that a decrease in the expression of the ABCB1 gene in patients diagnosed with triple-negative breast cancer is associated with the early stage of the disease and shorter survival of patients. The authors do not provide a possible causal mechanism explaining the effect of a decrease in the level of ABCB1 gene expression on the proliferation, migration, survival, or invasion of cancer cells. It is believed that a decrease in the ABCB1 gene expression in breast cancer cells may be a consequence of genomic instability in cancer cells, especially in triple-negative breast cancer [31].
In 2013, Hlaváč and colleagues published the results of their research on the search for new biomarkers of prognosis and/or predicting drug resistance among genes belonging to the ABC family in breast cancer patients (without division into molecular subtypes). For this purpose, the expression profile of 49 genes belonging to the ABC family was analyzed in the neoplastic tissue of breast cancer and in the normal tissue surrounding the tumor, treated at the Clinic Oncology Surgery in Prague. The obtained results indicated a significant increase in the expression level of the ABCA2, ABCA3, ABCB9, ABCC4, ABCC1, ABCC5, ABCC11, and ABCC10 genes in tumors compared to control tissues. For the ABCB1 and ABCC6 genes, the expression level was significantly reduced in the tumors. There were no significant changes in expression between the tumor and control tissues in the ABCB4, ABCA13, ABCC2, ABCC3, and ABCG2 genes. The expression levels of the ABCC10 and ABCC1 genes were higher in ER-negative patients than in ER-(+) patients. The expression level of the ABCC11 gene was higher in tumors with ER receptor expression. The expression levels of the ABCC11 and ABCA2 genes were significantly higher in patients with an expression of the progesterone receptor [32]. Differences in the gene expression levels in the work described above and in the assays performed by us may result from the fact that the level of expression was determined in a specific molecular subtype of BC-TNBC in our studies; however, in the described work, there is no separation of TNBC samples.
Balaji and colleagues examined the expression level of the ABCC1 and ABCC3 genes in breast cancer patients (without division into molecular subtypes) and assessed their role in inducing the phenomenon of resistance to anticancer drugs. They observed that the expression level of the ABCC1 and ABCC3 genes is elevated in breast cancer tissue, especially after treatment with anticancer drugs. The overexpression of the ABCC1 and ABCC3 genes led to a decrease in drug retention. The study showed that, like the ABCC1 gene, the ABCC3 gene was overexpressed in stage III primary breast cancer. Importantly, in in vivo and in vitro studies, drug treatment led to a further increase in ABCC1 and ABCC3 gene expression levels. In our studies, we also showed an increased expression level of the ABCC3 gene in the tumor tissue of patients with TNBC compared to controls, which may indicate the gene's role in inducing multidrug resistance in breast cancer cells. [33].
In conclusion, based on the experiments carried out, it can be concluded that the expression level of selected genes from the ABC family is associated with the molecular subtype of breast cancer as well as clinical features. Gene expression profiling is an important tool for assessing the genetic diversity of breast cancer and may influence the selection of appropriate effective therapy for women with triple-negative breast cancer. The high costs of the expression level profiling of all genes from the ABC family in patients with TNBC prevent the use of this method in routine hospital practice. However, based on literature data on the determination of the expression level of selected proteins encoded by genes from the ABC family in patients with TNBC and the results presented by us, it can be assumed that good prognostic markers of the course of TNBC and response to treatment in oncological patients can determine the expression level of the ABCC1, ABCC11, and ABCG2 genes [20][21][22][23][24][25][26]. More large-scale prospective studies as well as in vitro and in vivo functional studies are needed to assess the potential of ABC family genes as markers of the clinical course of TNBC and treatment responses in cancer patients.

Characteristics of Patients Qualified for the Study
Sixty-eight patients diagnosed with triple-negative breast cancer treated at the Oncology Center in Lublin were qualified for the study. The patients participating in the study gave their informed consent, and the study was approved by the Ethics Committee of the Medical University of Lublin (decision number: KE-0254/216/2014). The research was carried out in accordance with the Helsinki Declaration. The patients included in the study had no comorbidities, did not receive neoadjuvant chemotherapy, and did not indicate the presence of neoplasms in their family members in the history. Detailed clinical data of the patients are presented in Table 4. To determine the histological type of breast cancer, pathological lymph node metastases (pTNM), and tumor advancement, the 4th edition Classification of Tumors by the World Health Organization (WHO) for breast tumors [34], the 7th edition of the TNM classification of the American Joint Committee on Cancer (AJCC) [35], and the Scarff-Bloom and Richardson (SBR) scale were used [27].

Preparation of Material for Research
During the surgery, a fragment of the tumor (test sample) and tissue surrounding the tumor (control sample) were collected from the patients qualified for the study. The samples were subjected to histopathological evaluation to confirm the presence or absence of neoplastic cells. The collected tissue fragments were placed in sterile containers in an RNA-later solution (Invitrogen, Carlsbad, CA, USA) and stored at −20 • C.

Tissue Homogenization
The collected tissues were homogenized using the Precellys 24 homogenizer (Bertin-Instruments, Montigny-le-Bretonneux, France) with the Cryolys cooling option, enabling work with thermosensitive molecules. Tissue disintegration was achieved using a disintegrating material in the form of stainless steel beads (TK Biotech, Warszawa, Poland) placed in homogenized biological material.

RNA Isolation and cDNA Reverse Transcription
The collected tissue fragments were homogenized using a Precellys 24 homogenizer (Bertin-Instruments, Montigny-le-Bretonneux, France) with the Cryolys cooling option. The isolation of genetic material was performed using the modified Chomczyński and Sacchi method [36]. The quality and quantity of the isolated RNA was determined using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). The reverse transcription process was performed according to the manufacturer's instructions using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystem, Foster City, CA, USA).
The expression levels of the tested genes were determined using a 384-well TaqMan™ ® Human ABC Transporter Array (Applied Biosystems, Foster City, CA, USA) based on the manufacturer's protocol. The endogenous control in the presented studies was 18S-Hs99999901_s1. The level of gene expression using the real-time PCR method was determined according to the previously described procedure [37].

Statistical Analysis of Data
Statistica v.13.3, DisPlayr and GraphPad v.5.01 were used in the statistical analysis and graphic design (p < 0.05 was assumed statistically significant). The U Mann-Whitney test and the Kruskall-Wallis H test with multiple comparisons were used to calculate the differences in the expression level between genes and the r-Spearman coefficient was used with a heatmap correlation matrix for correlation analysis.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.