*Article* **Identification by MicroRNA Analysis of Environmental Risk Factors Bearing Pathogenic Relevance in Non-Smoker Lung Cancer**

**Alberto Izzotti 1,2 , Gabriela Coronel Vargas 3 , Alessandra Pulliero 3 , Simona Coco 4 , Cristina Colarossi 5 , Giuseppina Blanco 5 , Antonella Agodi 6 , Martina Barchitta 6 , Andrea Maugeri 6 , CT-ME-EN Cancer Registry Workers 7,† , Gea Oliveri Conti 6, \* , Margherita Ferrante 6,7 and Salvatore Sciacca 5**


**Abstract:** MicroRNA and DNA adduct biomarkers may be used to identify the contribution of environmental pollution to some types of cancers. The aim of this study was to use integrated DNA adducts and microRNAs analyses to study retrospectively the contribution of exposures to environmental carcinogens to lung cancer in 64 non-smokers living in Sicily and Catania city near to the Etna volcano. MicroRNAs were extracted from cancer lung biopsies, and from the surrounding lung normal tissue. The expression of 2549 human microRNAs was analyzed by microarray. Benzo(a)Pyrene-DNA adducts levels were analyzed in the patients' blood by HPLC−fluorescence detection. Correlations between tetrols and environmental exposures were calculated using Pearson coefficients and regression variable plots. Compared with the healthy tissue, 273 microRNAs were downregulated in lung cancer. Tetrols levels were inversely related both with the distance from Etna and years since smoking cessation, but they were not significantly correlated to environmental exposures. The analysis of the microRNA environmental signatures indicates the contribution of environmental factors to the analyzed lung cancers in the following decreasing rank: (a) car traffic, (b) passive smoke, (c) radon, and (d) volcano ashes. These results provide evidence that microRNA analysis can be used to retrospectively investigate the contribution of environmental factors in human lung cancer occurring in non-smokers.

**Keywords:** no-smokers lung cancer; microRNA; DNA adducts; environmental risk factors

#### **1. Introduction**

The lung epithelium undergoes a series of morphological changes before becoming invasive, such as hyperplasia, metaplasia, and finally dysplasia and in situ carcinoma. The two main types of lung cancer are small and non-small cell lung cancer (NSCLC), accounting for 80% to 85% of all cases. The three most common histological forms of

**Citation:** Izzotti, A.; Coronel Vargas, G.; Pulliero, A.; Coco, S.; Colarossi, C.; Blanco, G.; Agodi, A.; Barchitta, M.; Maugeri, A.; CT-ME-EN Cancer Registry Workers; et al. Identification by MicroRNA Analysis of Environmental Risk Factors Bearing Pathogenic Relevance in Non-Smoker Lung Cancer. *J. Pers. Med.* **2021**, *11*, 666. https://doi.org/10.3390/

Academic Editor: Soterios A. Kyrtopoulos

jpm11070666

Received: 11 June 2021 Accepted: 12 July 2021 Published: 15 July 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

NSCLC are epidermoid or squamous cell carcinoma, large cell carcinoma, and adenoma; among them, adenocarcinoma accounts for 40% of all lung cancer cases. [1] Environmental factors, including smoking, diet, and environmental carcinogens, are important in the pathogenesis of cancers through epigenetic modifications.

Interestingly, some miRNAs are dysregulated in NSCLC, which may be indicative of disease status or therapeutic outcome. [2]. It is nowadays well established that environmental pollutants alter the microRNA machinery, a situation resulting in adaptive effects in the case of short-term exposures and adverse effects in the case of long-term exposure [3]. MicroRNAs (miRNAs) are little non-coding RNA molecules that have different regulatory roles in cell differentiation, proliferation, and survival. miRNAs can inhibit complementary mRNA targets, regulating translation through RNA degradation, and are found to be dysregulated in numerous diseases, including cancer, frequently altered owing to mutations or transcriptional changes of the enzymes that regulate miRNA biogenesis [4]. However, approximately 25% of lung cancer cases worldwide, mainly adenocarcinoma, cannot be attributed to tobacco smoking; lung cancer in never smokers is the seventh leading cause of cancer deaths worldwide [5]. According to clinical experience, a different epidemiology and natural history are observed between lung cancers in never smokers and those in smokers [6], suggesting that lung cancer in never smokers is a "different" disease, with a specific etiology and molecular differences. A still unsolved problem is the evaluation of the environmental contribution to lung cancer in non-smokers. Concerning the use of miRNAs as biomarkers for lung cancer, it has been noticed that miR-146a-5p, miR-324-5p, miR-223-3p, and miR-223-5p may regulate cancer-associated gene expression, as they are down expressed in the normal bronchial airways of smokers with lung cancer [7]. Our previous research addresses the identification of a reliable cluster of miRNAs to be used as early cancer predictors, considering the high heterogeneity of lung cancer patients. Differences between miRNA profiles based on gender have being suggested in animal models of adenoma-free and adenoma bearing mice exposed to mainstream cigarette smoking [8]. Moreover, the early diagnosis of lung cancers using miR-33a-5p and miR-128-3p signatures has being proposed, as they are linked to tumor suppression processes. [9] Different response to air pollutants as particulate matter, ultrafine particles, nitrogen oxides, black carbon, and carbon oxides (CO and CO2) may be related to a different expression of miR-92a-3p, miR-484, and miR-186-5p, linking traffic-related exposure to disease risk [10]. It is known that microRNA and other molecular alterations (oncogene mutations, DNA adducts, transcriptional silencing activation, and proteosome alteration) induced by environmental pollution are quite specific, as each pollutant preferentially alters the expression of a cluster of identifiable molecular fingerprints [11]. This issue has been explored in a peculiar environmental situation characterized by the presence of an active volcano (Etna) near to the analyzed population (Sicily, Italy). Indeed, volcanic dust from Etna has been related to a higher risk of pleural mesothelioma, thyroid cancer [12], and other non-malignant respiratory diseases [13], as well as having a possible pathogenic role in the epidemiology of amyotrophic lateral sclerosis [14] and neurodegenerative diseases [15]. Etna's volcanic dust is also a vector of atmospheric pollutants, such as polycyclic aromatic hydrocarbons and particulates rich on mercury [16]. Furthermore, in a recent study, the surface reactivity of ash from Etna's activity was characterized and, although most of the released elements are below the Italian legal limits, a few inorganic elements (B, Cd, Ni, and As) are released in a higher level than permitted, with possible negative consequence for human health [17]. miR-19a, miR-30e, miR-335, and miR451a in peripheral blood have been suggested as potential biomarkers of radon radiation damage [18]. To the best of our knowledge, the correlation between volcanic ash exposure and miRNA alterations has never been explored.

To ascertain whether or not there was an association between the cancer-related pattern of microRNA alteration induced by passive smoke exposure, airborne car traffic pollution, volcano ash, and radon exposure, we herein present a retrospective study to investigate the correlation between miRNAs expression and DNA adducts in lung cancer tissue

and healthy tissue in non- and former-smokers in order to shed light on the differential contribution of environmental factors to the lung carcinogenesis process. The aim of this paper is to use integrated DNA adducts and miRNA analyses in order to shed light on the differential contribution of environmental factors to lung carcinogenesis in non-smokers.

The presented approach integrates both molecular biomarkers (DNA adducts) and post-transcriptional regulation analysis (miRNAs expression in lung tissue) in order to shed light on the differential contribution of environmental factors to lung carcinogenesis in non- and former-smokers, ranking each environmental risk factor, mainly including passive smoke, car traffic pollution, volcano ashes, and radon. The presented approach can contribute to prioritize public health intervention for the primary prevention of lung cancer in non-smokers.

#### **2. Materials and Methods**

#### *2.1. Patient's Recruitment and Sampling*

Patient recruitment was carried out in four hospitals of Catania (University Hospital "G. Rodolico, San Marco"; "Garibaldi-Nesima" Hospital; "Cannizzaro" Hospital; and "Morgagni" Clinic) and "San Vincenzo" Hospital of Taormina (Messina province). The study protocol was performed according to the Declaration of Helsinki and was approved by the Ethic Committee Catania 1 (n. 11,778 released on 17 March 2015) and Ethic Committee Catania 2 (346/C.E. released on 28 May 2015), respectively.

The criteria used for the patient enrolment was as follows: over 18 years of age, have lung cancer for which surgery treatment has been indicated, have been non-smokers or former smokers for at least 5 years, and have signed the written informed consent. No restriction was made regarding the sex of patients or the morphology of the reported neoplastic lesions. Both the neoplastic and healthy tissue samples were taken from the same patient and the tissue samples were obtained directly from the pathological anatomies of the hospitals involved in the project. Instead, the blood samples were collected by the thoracic surgery units of the hospitals. The interviews were carried out directly in the thoracic surgery wards by the cancer registry doctors involved in the study. A total of 64 patients were finally enrolled. All of the patients lived near the Etna volcano (average 56 km away, min 13 km away, and max 152 km away), their average age was 69.02 years old (min 43 years old and max 84 years old), 34.4% were female, and 20.3% of patients died within three years after the biopsy. A total of 15 subjects had never smoked, while 20 were former smokers for more than 20 years, 13 for 11 to 19 years, and 9 for 10 to 5 years, and the smoking habits data was missed for 14 patients. Data were collected by trained personnel using a semi-structured questionnaire to obtain information on the sociodemographic and lifestyle data, including smoke history, nutrition, home characteristic, and home location (for Radon and urban traffic pollution exposure evaluation) (Figure 1).

#### *2.2. Lung Biopsies Collection*

Lung biopsy specimens (*n* = 64) were collected at the onset of disease from patients who were diagnosed with lung cancer between 2015 and 2018, and were referred to the Catania, Messina, Enna Cancer Registry, Italy. The study was approved by Ethics committee—informed consent was obtained by "G. Rodolico—San Marco" University Hospital. All patients were classified as cases according to the 2021 ICD-10-CM Diagnosis Code C34.90. microRNA were comparatively evaluated in the cancer and surrounding normal tissue, as identified by intra-surgery histopathological analysis.

#### *2.3. miRNA Extraction*

The total RNA was extracted from the lung biopsies using a standardized protocol that combined a phenol/guanidine-based lysis of samples and silica-membrane-based purification. In brief, 30 mg of the starting material was first disrupted and homogenized in 700 µL of the QIAzol Lysis Reagent, using the TissueRuptor II (Qiagen, Milan, Italy) for 20–40 s. Next, the total RNA was purified from the homogenate using the miRNeasy

Mini Kit (Qiagen, Milan, Italy), as described by the manufacturer's protocol. Purification of RNA was automated on the QIAcube instrument (Qiagen, Milan, Italy).

**Figure 1.** Patients enrolled-characteristics and sample sizes of analytical determinations carried out.

#### *2.4. miRNA-Microarray and Bioinformatic Analyses*

—informed consent was obtained by "G. Rodolico—San Marco" University Hospi- The miRNA expression profiling was carried out with the Agilent platform following the miRNA Microarray protocol v.3.1.1 (Agilent Technologies, Santa Clara, CA, USA). Briefly, 50 ng of total RNA, containing miRNAs and Spike-in controls underwent dephosphorylation and a labelling step with Cyanine 3-pCp. The Cy3-labeled RNA was then purified using a Micro Bio-Spin P-6 Gel Column (Bio-Rad Laboratories, Inc., Hercules, CA, USA) and hybridized on Human miRNA microarray slide 8 × 60 K (Agilent Technologies; including 2549 miRNAs, miRBase 21.0) at 55 ◦C for 20 h. After washing, the slides were scanned with a G2565CA scanner (Agilent Technologies) and the images were extracted by Feature Extraction software v.10 (Agilent Technologies). The microarray raw data were previously deposited in the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/; accessed on 8 March 2021, GEO accession number GSE169587, ID: 200169587) for a previous study from the same authors [4].

– ' The bioinformatic analyses of the microarray data were performed with GeneSpring software (GeneSpring Multi-Omic Analysis version 14.9 Build 11,939 by Agilent Technologies). For each specimen, the intensities of the replicated spots on each array were log transformed and averaged. All the lung-tissue-miRNA raw data files from the Agilent Technologies Microarray Scanner System G2565CA were imported to GeneSpring using miRNA analysis type, Technology 70156\_v21\_0, without baseline transformation. Data processing was performed by 3D principal component analysis (PCA) scores and Hierarchical Clustering.

Comparisons between sets of data were done by evaluating the fold variations. A volcano plot *t*-test analysis for all the miRNA entities between the averaged tumoral and healthy tissues was run, using fold change ≥2 and *p*-value ≤ 0.05, without multiple test corrections as the threshold values.

miRNAs related to five different environmental exposures (Environmental Exposure miRNA Signature) were determined analysing lung cancer related miRNAs comparatively in exposed vs. non-exposed subjects. Environmental exposures were determined for each patient using the questionnaires information according to (a) passive smoking at home, (b) passive smoking at work, (c) vehicle traffic at home, (d) distance (km) from the Etna volcano, and (e) radon risk according to home type.

To understand the relationship between environmental exposure signatures and their biological significance in lung cancer tissues, a target detection for each environmental exposure signature was done using the TargetScan prediction database. This database was chosen as it is the most updated database, and the number of target genes are reported for different cut-offs. The most interesting genes targeted by environmental exposure miRNA signatures that are potentially related with each environmental exposure were identified.

After the target detection, a prediction model was build using a Neural Network class prediction algorithm for each Environmental Exposure miRNA Signature to test the overall accuracy prediction for the chosen miRNAs.

#### *2.5. Validation of Microaarray Results by qPCR Analysis*

Microarray results for let-7a and miR-15 were further validated by qPCR on a subset of 20 patients for which enough RNA was still available. These miRNAs were selected because of their relevance lung carcinogenesis. The total RNA (10 ng) was reverse transcribed using miR-specific stem-loop RT primers (TaqMan MicroRNA Assays; Applied Biosystems, Thermo-Fisher) and components of the High Capacity cDNA Reverse Transcription kit (Life Technologies) according to the manufacturer's protocols.

The expression levels of individual miRNAs were detected by subsequent RQ-PCR using TaqMan MicroRNA assays (Life Technologies) and a Rotor Gene 3000 PCR System Corbett (Qiagen) using standard thermal cycling conditions in accordance with manufacturer recommendations. The PCR reactions were performed in triplicate in final volumes of 30 µL, including inter-assay controls (IAC) in order to account for variations between runs. RT-PCR (TaqMan MicroRNA Assays; Applied Biosystems, Thermo-Fisher) was used to quantify the expression of let-7a and miR-15 according to the manufacturer's instructions. To normalize the data for quantifying miRNAs, the universal small nuclear RNU38B (RNU38B Assay ID 001004; Applied Biosystems) as an endogenous control was used.

#### *2.6. Benzo[a]Pyrene-DNA Adduct Levels in Human White Blood Cells*

Hydrolyzed BPDE adducts or Tetrol I-1 and Tetrol II-2 were analyzed in lymphocyte DNA through the modified High-Performance Liquid Chromatography–Fluorescence (HPLC−FL) method described by Alexandrov et al. [19] and Oliveri Conti et al. [20].

Briefly, lymphocytes were separated from whole blood samples using HISTOPAQUE-1077 (Sigma-Aldrich Chemie Gmbh, Munich, Germany). The lymphocyte DNA was extracted using the DNeasy Blood and Tissue kit according to customer's procedure (Qiagen, Milan, Italy).

Hence, DNA was subjected to a procedure of hydrolysis and purification, and Tetrols were quantified according to the methodology of Oliveri Conti et al. [20]. HCl, also if hypergrade certified, can contain traces of fluorescent active contaminants that could interfere with the peak detection of the studied analytes and reduce analytical sensitivity of the method.

To avoid this important bias in the sample preparative phase, all of the HCl impurities visibly reactive to the FL detector were deleted by chemical purification [20]. To improve the sensibility of detection, Thermo Scientific™ PEEK Capillary Tubing (0.005 in) was used. The extracted and purified DNA was dissolved in 1 mL of water and analyzed in a Varian Pro Star System HPLC using a TOSOH (C18 RP 25 × 0.46 cm, 5 µm) column with the following elution program: 15 min with 20% water/acetonitrile of equilibrium phase, 5 min with 20% water/acetonitrile and 60 min to acetonitrile (100%) (slop of 1) and, finally 10 min to 100% acetonitrile.

An isocratic program (0.85 mL/min) was used, and the FL detector (FLD) was programmed to 344 nm (ext.) and 388 nm (em.) for the excitation and emission wavelengths, respectively. The sensitivity of the FLD was fixed to a high modality. The wavelength of

UV−VIS detector (UV) was set at 238 nm, permitting the dual detection of both Tetrols (I-1 and II-2).

The chromatographic system was calibrated using external certified pure standards of Tetrol I-1 and Tetrol II-2 (purity 99.0%) (Chemical Carcinogen Reference Standard Repository, Kansas City, MO, USA).

Recoveries were 94% and 82% for Tetrol I-1 and Tetrol II-2, respectively. The processing of reagent blank disclosed no trace of Tetrol I-1 and Tetrol II-2. The linearities (*R 2* ) obtained of FLD were 0.9980 and 0.9990 for Tetrol I-1 and Tetrol II-2, respectively. For UV, the *Rs<sup>2</sup>* were 0.9850 e 0.9803 for Tetrol I-1 and Tetrol II-2, respectively. MDL were 2.0 pg/mL and 3.1 pg/mL for Tetrol I-1 and Tetrol II-2, respectively. The validated method permitted detecting Tetrol I-1 and Tetrol II-2 in a minimum of 3µg of extracted DNA.

#### *2.7. Statistical Analysis*

The statistical significance of the differences between groups was evaluated by ANOVA, followed by Student's *t*-test for unpaired data. *p*-values lower than 0.05 were regarded as statistically significant. Correlations (i.e., Pearson coefficients and regression variable plots) between Benzo[a]Pyrene-DNA adducts and the different environmental exposures were calculated with IBM SPSS statistics (Version 22).

#### **3. Results**

#### *3.1. Comparison of miRNA Profile between Healthy and Cancer Tissue in Lung*

The scatter plot analysis of the miRNA-arrays comparing healthy and lung cancer tissues presents a general trend toward down regulation in cancer tissue, as indicated by the slope of the black regression line (Figure 2a). The volcano plot analysis highlighted a list of 273 miRNAs that were altered more than two folds and above the statistical significantly threshold (*p* < 0.05) in cancer vs. healthy tissue. Of these miRNAs, 222 were down-regulated (blue dots) and 51 were up-regulated (red dots) (Figure 2b). This group represents the Lung Cancer Related miRNAs. This list is reported in the supplementary material (Supplementary Table S1) and includes well established oncogenic miRNAs, such as an extensive downregulation of the whole let-7 miRNA family and of the miR-34 family, an established effector of p53.

The lung cancer related miRNAs downregulation trend was well distinguishable between the cancer and healthy tissue, as also indicated by the hierarchical cluster analysis (Figure 3a), where healthy tissue profiles (yellow bar) were clustered in the upper part of the hierarchical tree separately from cancer tissue profiles (blue bar). Colour range indicates lung cancer related miRNAs' intensity signal.

In the principal component analysis of variance (Figure 3b), healthy tissue samples (yellow dots) were mainly clustered in the lower left part of the 3D space. Instead, cancer tissue samples (blue dots) were mainly located in the upper right part of the 3D space.

The most significant predicted target genes for each of the lung cancer related miRNAs were identified using the TOP-Go Bioconductor R package and REVIGO online tool. The tree map of the most representative biological processes for lung cancer related miRNAs is reported in (Supplementary Figure S1). The most representative biological processes were as follows: regulation of RNA splicing, tissue migration, monosaccaride transmembrane transport, protein modification by small protein conjugation or removal, and cellular protein-containing complex assembly.

qPCR analyses performed for let-7a confirmed the downregulation of this miRNAs in cancer vs. healthy lung tissues of the same patient (Figure 4).

**Figure 2.** (**a**) Scatter plot analysis comparing miRNA expression (dots) according to their level of expression in healthy (vertical axis) vs. cancer (horizontal axis) tissues of the examined patients. The miRNA colour reflects the signal intensity (red is high, yellow is intermediate, and blue is low). The diagonal green lines indicate the two-fold variation interval. The best-fit regression line is reported in black. Its slope towards the horizontal axis reflects the overall downregulation of miRNA expression in cancer compared with healthy lung tissue. (**b**) Volcano plot analysis identifying miRNAs with an altered expression above two-fold (horizontal axis) and above the statistical significance threshold (*p* < 0.05) (vertical axis) in cancer vs. healthy lung tissue, either downregulated (blue) or upregulated (red).

**Figure 3.** (**a**) Clustering hierarchical analysis reporting the expression of the 273 lung cancer related miRNAs in healthy tissue (vertical axis, yellow bar) and lung cancer tissue (vertical axis, blue bar) in the 50 samples tested (horizontal lines). (**b**) Principal component analysis of variance identifying the samples from healthy tissues (yellow dots) and cancer tissues (blue dots) according to the variance of the expression of the 273 lung cancer related miRNAs. dots size = principal component analysis score.

**Figure 4.** Let-7a expression in healthy (blue) vs. cancer (red) lung tissue as evaluated by qPCR in 20 patients (*x*-axis). miRNA expression intensity is expressed in fluorescent units (*y*-axis).

On an average, Let-7a expression was down-regulated in cancer vs. healthy tissues by 6.2 ± 1.4 fold as evaluated by qPCR, and 7.4 ± 2.2 fold as evaluated by microarray.

#### *3.2. miRNA Profile Was Related with Cancer Histotype*

The miRNA expression was different between small cell lung cancer (SCLC) and NSCLC, as shown by the scatter plot analysis (Figure 5a). The volcano plot analysis indicated that 26 out of the 273 cancer related miRNAs were differentially expressed between SCLC and NSCL (Figure 5b). Of these miRNAs, 25 were up-regulated in NSCLC compared with SCLC and one was down-regulated. The identity of these 26 miRNAs permitted distinguishing between these two main cancer histotypes, and is reported in the supplementary material (Supplementary Table S2).

#### *3.3. B(a)P-DNA Adducts and Environmental Exposures*

The ANOVA analysis did not detect a statistically significant difference between the B(a)P-DNA adducts levels under different environmental exposures, including passive smoking at home, passive smoking at work, radon risk related to home type, and vehicle traffic at home (Figure 6). The linear regression analysis showed that the level of B(a)P-DNA adducts in the lymphocytes was inversely related with the distance from the Etna volcano (Figure 6e) and years since smoking cessation (Figure 6f).

‐ ‐ **Figure 5.** (**a**) Scatter plot analysis comparing miRNA expressions (dots) according to their level of expression in SCLC (vertical axis) vs. NSCLC (horizontal axis). miRNA color reflects the level of expression (red is high, yellow is intermediate, and blue is low). The diagonal lines indicate the two-fold variation interval. (**b**) Volcano plot analysis identifying miRNAs whose expression was altered more than two-fold (horizontal axis) and above the statistical significance threshold (*p* < 0.05) (vertical axis) in SCLC vs. NSCLC cancer downregulated (blue) or upregulated (red).

‐

≥ *≤*

**Figure 6.** Box plot analysis for total BaP DNA adducts in: (**a**) passive smoking at home, (**b**) passive smoking at work, (**c**) radon risk home type, (**d**) vehicle traffic at home, and linear regression for (**e**) distance from the Etna volcano, and (**f**) years since smoking cessation.

#### *3.4. Contribution of Environmental Exposures to Lung Carcinogenesis as Inferred from miRNA Profiling*

Five miRNA signatures were obtained comparing the miRNA expression in the tumoral lung tissue between patients undergoing a low or high exposure for each one of the environmental exposures. The cancer related miRNAs included in each miRNA environmental signature were used to run a volcano plot analysis (FC ≥ 2, *p* ≤ 0.05). These environmental signatures were further integrated with the established miRNA, as linked with the specific exposure from the available literature.

The number of miRNAs composing each environmental signature was as follows: (a) passive smoke at home, *n* = 8; (b) passive smoke at work, *n* = 1; (c) vehicle traffic, *n* = 53; (d), distance from the Etna volcano, *n* = 21; and (e) radon risk, *n* = 19. The volcanos plot analyses comparing miRNA expression in the lungs between unexposed vs. exposed subjects for each environmental signature are reported in Figure 7.

**Figure 7.** *t*-test volcano plot analyses identifying the miRNA environmental signatures (among the 273 lung cancer related miRNAs).

Accurate questionnaire data were available for 38 out of the 50 patients for whom the microRNA microarray data were collected. A volcano plot *t*-test using each EES was run to identify the altered miRNAs per patient.

Environmental exposure miRNA signatures were compared by Venn diagram analysis, with the miRNAs composing the individual cancer-related signature of each patient. This approach was used to identify the relative contribution of the environmental risk factors to cancer development in each patient (Figure 8).

The number of patients with a higher number of differentially regulated miRNAs than the median value for each environmental risk factor was as follows: (a) 38 for passive smoke, (b) 38 for vehicle traffic, (c) 21 for distance from the Etna volcano, and (d) 13 for radon risk

The targeted genes for each environmental exposure signature were analyzed. The number of target genes for each signature and different *p*-values cut-off are summarized in Table 1.

Most of the genes targeted by the environmental exposure miRNA signatures with statistical significance are also expressed in the lung tissue. The gene RIF (gene reference into function) and RPKM (reads per kilobase of transcript, per million mapped reads) values in the lung tissue for each targeted gene were found using the NCBI gene database

(https://www.ncbi.nlm.nih.gov/gene) accessed on 11 December 2021. These findings are reported in Table 2.

**Figure 8.** Number of altered environmental exposure miRNAs (*y*-axes, color codes) per patient (*x*-axes). Median of the summation of each environmental exposure miRNA signature are shown as horizontal colored lines. Patients having environmental risk factors above the median value underwent lung cancer contribution by this risk factor.

**Table 1.** Number of predicted target genes (Targetscan database) according to each environmental exposure with a *p*-value cut off.


The most significant targeted genes for each signature are summarized in Table 2.




#### *3.5. Evaluation of Environmental Exposure miRNA Signatures Efficacy by Neural Network Analysis*

GeneSpring 14.9 was used to build a prediction model to validate the accuracy of each of the environmental exposure signatures. The network was tested in tumoral tissue. The accuracy of the environmental signature increased compared with the number of miRNAs included, as follows: (a) remarkably, for passive smoking at home, distance from the Etna volcano, and home type radon risk; (b) slightly for vehicle traffic and distance from the Etna volcano; and (c) remained equal for passive smoking at work (Table 3).


**Table 3.** Results of the neural network class prediction for each environmental exposure miRNA signature.

#### *3.6. Environmental Exposure miRNA Signatures and B(a)P-DNA Adduct Levels*

The weight of different exposures to lung cancer were profiled by analyzing the B(a)P-DNA adduct levels in the lymphocytes and miRNAs profiles in lung cancer together. For this purpose, we ranked the four exposures on the basis of the number of altered miRNAs. Correlation tests (Pearson and Spearman's Rho) were run to evaluate whether or not B[a]P-DNA adducts were related with environmental exposure miRNA signatures. The B[a]P-DNA adduct levels were correlated only with the passive smoking miRNA signature (*p*-value = 0.049) (Table 4).

**Table 4.** Correlation between EESs miRNA and B(a)P-DNA adduct levels. (\*) Statistical significance *p* < 0.05.


\* Statistical significance *p* < 0.05.

#### **4. Discussion and Conclusions**

Our results provide evidence that miRNAs are massively deregulated in lung cancer compared with the surrounding normal tissue. This finding is in line with other studies [21]. A major problem in using miRNA analysis for lung cancer prediction and early diagnosis is the reproducibility of the results and the invasiveness of the biopsy approach. Our cancer related miRNAs signature at least in part overlaps with the most common lung cancer miRNA-related signatures found in the literature. Indeed, 27 miRNAs were included in our cancer related signature and also in other cancer miRNA-related signatures found in literature [22]. The miRNAs downregulated in lung cancer tissue included established antioncogenic miRNA such as let-7, miR-30, miR-34, and miR-140. The comparison of DNA adducts and miRNA expression provided evidence that post-transcriptional alteration is massive in lung cancer, while DNA adducts alteration of an environmental origin is detectable, but only at a very low level. DNA adducts are a hallmark of the environmental contribution to the analyzed cancers, with particular reference to environmental sources of polycyclic aromatic hydrocarbons derived from combustion, such as car traffic and passive smoke. However, BaP-DNA adducts are poor predictors of cancer because (a) they can be removed by DNA repair [23], (b) the resulting mutation can be silenced thus not having phenotypical or functional consequences, and (c) the bearing cell can be removed by apoptosis. Conversely, miRNA alterations are necessary to develop lung cancer. These alterations are initially adaptive, aimed at activating defensive processes counteracting the damage induced by environmental pollutants such as DNA repair,

mutation silencing, and apoptosis. However, whenever the environmental exposure lasts for decades, microRNA alterations become irreversible and commit cells to the occurrence of lung cancer environmental risk factors, inducing specific alterations in the microRNA machinery depending on the specificity of the environmental pollutant involved [3–11]. Accordingly, microRNA alteration is more predictive of lung cancer occurrence than genomic alterations. Indeed, microRNAs have been proposed as a tool for the early diagnosis of cancer or to identify subjects at a high risk for cancer development needing to undergo personalized cancer screening with a high frequency and sensitivity.

The negative correlation observed between the adduct level and some environmental exposures (i.e., traffic and volcano distance) is in line with the results previously published by other research groups reporting that populations undergoing long-term exposure to environmental pollution develop resistance mechanisms [24]. These events are referred to as adaptive events triggered by heterogeneous exposures [25].

The biological function of miRNAs identified as lung cancer contributors in environmental signatures reflect the pivotal role of the damage to the microRNA machinery during the carcinogenesis process. These events have been previously analyzed in detail during lung carcinogenesis in mice [8,26].

The results presented herein provide evidence that miRNA alteration in lung cancer results from exposure to environmental factors. This situation results in miRNA failure to control pivotal defensive mechanisms against cancer, mainly including oncogene suppression, cell adhesion and differentiation maintenance, cell cycle blockage, DNA and protein repair, intracellular signaling, and apoptosis.

The obtained results provide evidence that miRNA signatures may be used to identify the comparative contribution of environmental factors to lung carcinogenesis in humans. According to our environmental exposure miRNA signatures results, the contribution of the analysed environmental factors was, in decreasing order, car traffic, passive smoke, volcano, and radon. These results may be useful for stakeholders to prioritize public health intervention for the primary prevention of lung cancer in non-smokers. Indeed, it is commonly thought that the contribution of volcano ash is the main public health problem in Catania, one of the few cities in the world to be directly exposed to volcano emissions located in its near proximity. However, the obtained results indicate that the main public health problem in this town is the car traffic. Accordingly, preventive measures, such as the substitution of old with new cars characterized by low emission rates, are urgently required. In the second rank, there is passive smoke, which is a problem to be faced by stronger information campaigns and other measures (such as the increase of cigarettes price), which appears to be urgent in a country (i.e., Italy) still having the prevalence of 13 million of smokers out of a total population of 60 million.

In conclusion, the results presented herein provide more evidence that the analyses of epigenetic components may be used to face public health issues related with cancer prevention [27], with particular reference to the identification of the environmental risk factors to be prevented.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/jpm11070666/s1, Figure S1: Revigo TreeMap of GO-BP analysis. Most significative BPs are classified by colour with BPs subset, Table S1: *Cancer Related miRNAs*, Table S2: Significant altered miRNAs.

**Author Contributions:** Conceptualization, A.I. and S.S.; methodology, A.I., G.C.V., S.S. and A.P.; CT-ME-EN Cancer Registry Authors, G.O.C., M.F. and A.A.; validation, G.O.C., A.P., G.C.V., S.C. and M.B.; investigation, Cancer Registry Authors, G.O.C., M.B., A.M., C.C. and G.B.; resources, S.S. and M.F.; data curation, G.O.C., A.P., G.C.V. and A.I.; writing—original draft preparation, A.I., G.C.V., A.P. and G.O.C.; writing—review and editing, all authors; visualization, all authors; supervision, A.I. and M.F.; project administration, A.I. and S.S.; funding acquisition, S.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by Regional CT-ME-EN-Cancer Register fund-2015, and partially by the Italian Association for Cancer Research (AIRC, IG2017-20699 to A.I.) and the Italian Association for Cancer Research (AIRC, IG2017-20699 to A.I.).

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board Ethics Committees (n. 11778 released on 17 March 2015. and 346/C.E. released on 28 May 2015).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

**Acknowledgments:** We would like to thank all co-authors including the staff from CT-ME-EN Cancer Registry Group for their contribution: Andrea Benedetto (andrea.benedetto@registrotumoriintegrato.it) Marine Castaing (marinecastaing@hotmail.com), Alessia Di Prima (alessia26d@gmail.com), Paolo Fidelbo (paolo.fidelbo@gmail.com), Antonella Ippolito (antonellaippolito81@gmail.com), Eleonora Irato (eleonorairato@hotmail.it), Anna Leone (annaleone76@yahoo.it), Fiorella Paderni (fpaderni@gmail.com), Paola Pesce (paolapesce@tiscali.it), Alessandra Savasta (alessandrasavasta@virgilio.it), Carlo Sciacchitano (carlo@carlosciacchitano.it), Antonietta Torrisi (torrisidora@gmail.com), Antonina Torrisi (torrisinina@gmail.com), Massimo Varvarà (max.varvara@libero.it), Carmelo Viscosi (ettore.viscosi@gmail.com).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Margherita Ferrante 1,2, \* , Antonio Cristaldi <sup>1</sup> and Gea Oliveri Conti 1**


**Abstract:** The daily environmental exposure of humans to plasticizers may adversely affect human health, representing a global issue. The altered expression of microRNAs (miRNAs) plays an important pathogenic role in exposure to plasticizers. This systematic review summarizes recent findings showing the modified expression of miRNAs in cancer due to exposure to plasticizers. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology, we performed a systematic review of the literature published in the past 10 years, focusing on the relationship between plasticizer exposure and the expression of miRNAs related to cancer. Starting with 535 records, 17 articles were included. The results support the hypothesis that exposure to plasticizers causes changes in or the deregulation of a number of oncogenic miRNAs and show that the interaction of plasticizers with several redundant miRNAs, such as let-7f, let-7g, miR-125b, miR-134, miR-146a, miR-22, miR-192, miR-222, miR-26a, miR-26b, miR-27b, miR-296, miR-324, miR-335, miR-122, miR-23b, miR-200, miR-29a, and miR-21, might induce deep alterations. These genotoxic and oncogenic responses can eventually lead to abnormal cell signaling pathways and metabolic changes that participate in many overlapping cellular processes, and the evaluation of miRNA-level changes can be a useful target for the toxicological assessment of environmental pollutants, including plastic additives and plasticizers.

**Keywords:** plasticizers; cancer; microRNA; in vitro study; PRISMA

#### **1. Introduction**

The continuous daily environmental exposure of humans to different chemicals may adversely affect human health, thus representing a global issue [1–11]. In the last decade especially, ecological and epidemiological studies have focused on the presence of plastics and their additives in food and the environment [6,12–14].

Plasticizers are added to plastics to increase their flexibility, durability, and pliability. A large broad of molecules are used by the polymer industry, including phthalates, bisphenolates, flame retardants, metals, parabens, polychlorinated biphenyls, tributyltin, organophosphate esters, etc.

Among the plasticizers, phthalates, are the most widely used in polyvinyl chloride (PVC), polyethylene terephthalate (PET), polyvinyl acetate (PVA), and polyethylene (PE). Phthalates can be found in toys, personal care products, food packages, paints, pharmaceuticals and drugs, medical devices, catheters, blood transfusion devices, cosmetics, and PVC products for home furnishing such as PVC films for floors or household accessories [12,15,16].

Bisphenol A (BPA), another plasticizer, is the major component in the manufacture of epoxy and polycarbonate plastics and flame retardants. The uses of BPA include coatings for PVC water pipe walls, plastic bottles for water, baby bottles, food packaging, receipt inks, cosmetics, plastic toys, etc. BPA has drawn attention from public health and governmental agencies due to its widespread use. Also, BPA exerts genotoxic and carcinogenic

**Citation:** Ferrante, M.; Cristaldi, A.; Oliveri Conti, G. Oncogenic Role of miRNA in Environmental Exposure to Plasticizers: A Systematic Review. *J. Pers. Med.* **2021**, *11*, 500. https://doi.org/10.3390/ jpm11060500

Academic Editors: Alessandra Pulliero and Alberto Izzotti

Received: 2 April 2021 Accepted: 28 May 2021 Published: 2 June 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

activities due to the similarity of its chemical structure that resembles that of diethylstilbestrol, an accredited human carcinogen [17,18]. Plasticizer exposure has been reported as a reproductive and developmental toxicant and carcinogen [19,20].

Plasticizers have been associated with hormone-sensitive cancers such as breast, prostate, endometrial, ovarian, testicular and thyroid cancers, but also with non-hormonally sensitive cancers such as cervical and lung cancers, osteosarcoma, and meningioma [21].

Based on the evidence that is already available, several countries regulated some plasticizers, especially in food and human consume products. Some laws and limits are formulated under which certain plasticizers are permitted to be used in the plastics industry, for instance, the USA Consumer Product Safety Improvement Act (CPSIA) [22], the Proposition 65 list of California State (PROP65) [23], the European Commission Regulation (EC) no. 372/2007 of 2 April 2007 [24], etc.

However, the Chang and Flow study [25] showed that also subchronic exposures to DEHP and DiNP in adulthood lead both to immediate and long-term reproductive consequences in female mice.

The altered expression of microRNAs (miRNAs) represents an epigenetic mechanism that exerts an important pathogenic role linked to exposure to environmental pollutants with several pathological outcomes, including cancer promotion and development [26–28]. These miRNAs are very-short RNA, ranging from 19 to 25 nucleotides in size, that regulate the post-transcriptional silencing of target genes. A single miRNA can target hundreds of mRNAs and influence the expression of a large number of genes often involved in several important functional pathways [26]. The miRNAs are differentially regulated in various types of cancer, including ovarian, liver, gastric, pancreatic, esophageal, colorectal, breast, and lung cancers [29].

An emerging hypothesis explores the supposed coordination between miRNA-mediated gene control and splicing events in gene regulatory networks [27]. Several studies suggest that the maturation of miRNAs may depend on splicing factors [30]. However, microRNA modification results in carcinogenesis only when other molecular changes occur simultaneously, such as suppression of the inhibition of the expression of mutated oncogenes, the formation of microRNA adducts, p53-microRNA interconnections, and alterations of the *Dicer* function [26].

Due to the considerable stability of miRNAs, they are measurable both in blood and tissues and are therefore eligible as potential biomarkers for several non-communicable diseases, including cancer [29].

This systematic review summarizes recent findings showing the aberrant expression of miRNAs in cancer due to plasticizer exposure. We further discuss the challenges in environmental-miRNA research because this approach can be key for understanding the mechanism of cancer pathophysiology but also for early screening and/or personalized cancer therapy.

#### **2. Materials and Methods**

A brief critical review of scientific papers from the last ten years selected using the PubMed, Scopus, and Web of Science databases was carried out. The "Preferred Reporting Items for Systematic Reviews and Meta-Analysis" (PRISMA) methodology was applied in this study.

To assess the influence of plasticizer exposure on the expression of miRNAs in humans, all original articles published from 1 January 2010 to 29 December 2020 were selected based on the following criteria:


We searched the databases for controlled randomized studies, cohort studies, case– control studies, case reports, and in vitro studies. Only original articles written in English were collected for the PRISMA review.

We excluded papers that did not include original data such as informative reviews, commentaries, and editorials. Systematic reviews and meta-analyses were not eligible, but their references were screened for recovery of eligible studies missing in the databases.

Two investigators (A.C. and G.O.C.) screened all citations for potentially eligible studies and extracted data independently. Disagreements were adjudicated by a third investigator (M.F.).

The research was conducted using keywords including "Plasticizers and microRNA", "Plasticizers and oncogenic microRNA", "additives of plastic and miRNA cancer", "endocrine disrupting chemicals and miRNA cancer", "BPA and miRNA cancer", "di-n-butyl phthalate and miRNA cancer", "DBP and miRNA cancer", "monobutyl phthalate and miRNA cancer", "MBP and miRNA cancer", "Organophoshorus flame retardants and miRNA cancer", "flame retardants and miRNA cancer", "di(2-ethylhexyl) phthalate and miRNA cancer", "DEHP and miRNA cancer", "mono-(2-ethylhexyl) phthalate and miRNA cancer", "MEHP and miRNA cancer", "methylparaben and miRNA cancer", "parabens and miRNA cancer", "phthalates and miRNA cancer", "environmental phenols and miRNA cancer", "organophosphate esters and miRNA cancer", "Tributyltin and miRNA cancer", "PCBs and microRNAs cancer", "butylbenzyl phthalate and miRNA cancer", "PVC and miRNA cancer".

We also used combinations of the keywords, such as "Plasticizers and oncogenic effects" and "Plasticizers and microRNA changes".

All eligible studies were evaluated as eligible using a modified Newcastle Ottawa scale (rating system to score each study) [4].

#### **3. Results**

#### *3.1. Literature Inclusion Criteria*

Our initial search produced 477 potential references (Figure 1) from databases and 58 references from other sources.

Starting with 535 records, we identified in the first phase 322 records after the deletion of 213 duplicates. In addition, via evaluation of the title and abstract, we screened 57 records. Of these, 40 records were excluded due to the absence of some criteria of eligibility (reporting on miRNAs or indications, evidence about exposure to plasticizers, cancer tissue or cells analysis reporting, correct sample size, in vivo or in vitro study, statistical analysis of data, analysis of plasticizers in biological tissues) as listed in Figure 1. Finally, we included 17 studies in the systematic review. These studies used various approaches or study designs, but all focused on the effects of exposure to plasticizers on outcomes defined as "oncogenic miRNA identification and their down- or upregulation description".

#### *3.2. Summary of Literature Included*

Figure 1 describes the findings of the performed collection and screening procedures. We included the following studies, in chronological order: Tilghman et al., 2012 [31], Meng et al., 2013 [32], Li et al., 2014 [33], Kim et al., 2015 [34], Buñay et al., 2017 [35], Chang et al., 2017 [36], Chou et al., 2017 [37], Hui et al., 2018 [38], Wu et al., 2018 [39], Yin et al., 2018 [40], Scarano et al., 2019 [41]; Wang et al., 2019 [42], Zhu et al., 2019 [43], Cui et al., 2019 [44], Chorley et al., 2020 [45]; Duan et al., 2020 [46], and Zota et al., 2020 [47].

All the studies are in vitro cell studies on rats, mice, or human cancer cell lines. Several plasticizers were studied through controlled in vitro exposure as reported in Table 1.

**Table 1.** The included studies and their results ++ .





++ List of studies are organized according to chronological order in subgroup on the type of plasticizer/cocktail of plasticizers alphabetical basis basis. \* AS52-mutant cell (ASMC) clones; \*\* Human endometrial cancer cell line; <sup>+</sup> Human Hemangioma cells; <sup>β</sup> Acute myeloid leukemia; § Human ovarian cancer cell lines; <sup>ˆ</sup> Human breast cancer cells; <sup>ρ</sup> (ERα-negative and estrogen-resistant); NE: no effect; UP: upregulated, Down: downregulated, ◦ Human hepatocellular carcinoma; <sup>Σ</sup> Human oral squamous cell carcinoma; <sup>π</sup> Human prostate cancer cells; \$ cancer cells; a (10−<sup>4</sup> or 10−<sup>5</sup> M); <sup>b</sup> (10−<sup>6</sup> to 10−<sup>11</sup> M); Mono-ethylhexyl phthalate (MEHP); Bis (2-ethylhexyl) phthalate (DEHP); diethyl-phthalate (DEP); dibutyl phthalate (DBP); di-isobutyl-phthalate (DiBP), butylbenzyl-phthalate (BBzP); di-isononyl-phthalate (DiNP); benzyl butyl phthalate (BBP); 4-nonylphenol (NP); 4-tert-octylphenol (OP); dinoctyl phthalate (DNOP); Tris (1,3-dichloro-2-propyl) phosphate (TDCIPP) = organophosphate flame retardants. Observational Research on Genes and the Environment (FORGE) study; ΣDEHP = Sum of 21 phthalates and metabolites; ΣAA phthalates = Sum of 31 antiandrogenic phthalate metabolites.

Human endometrial, hemangioma, acute myeloid leukemia, ovarian, breast, hepatocellular, oral squamous, and prostate cancer cells lines were evaluated.

#### *3.3. Detailed Overview of the Literature Included*

#### 3.3.1. In Vitro Studies

Tilghman et al. (2012) [31] studied the effects of BPA (10 µM) and DDT (10 µM) on miRNA regulation and expression levels in hormone-responsive human breast cancer cells. The MCF-7 breast cancer cell line showed that both pollutants increased the expression of *ER* receptor target genes, including the progesterone receptor, bcl-2, and trefoil factor. The revealed miRNAs (27) were outlined in the exposed cells (miR-21, miR-638, miR-663, miR-1915, let-7g, let-7c, miR-923, miR-93, miR-320a, miR-1308, let-7f, miR-15b, miR-1275, miR-27b, miR-222, miR-193a-5p, miR-16, miR-26b, miR-149, miR-92a, miR-99b, miR-92b, miR-342-3p), of which several were upregulated and downregulated according to Table 1.

Several genes were differentially regulated by the compounds. For example, *Jun* and *Fas* genes were increased approximately 1.8- and 1.5-fold by BPA, but were relatively unchanged by DDT. The onco-miR-21 is an estrogen-regulated miRNA that plays an important role in breast cancer. In this study, miR-21 expression was downregulated by BPA, and several members of the let-7 family (let-7a, let-7b, let-7c, let-7d, let-7e and let-7f), were downregulated (*p* < 0.05) by all treatments. In contrast, miR-638 (*p* < 0.005), miR-663 (*p* < 0.005), and miR-1915 (*p* < 0.005) were upregulated by BPA and DDT.

Chou et al. (2017) [37] investigated the role of BPA exposure in the disruption of miRNA regulation and whether the related gene expression is decisive for carcinogenic progression. This study was carried out using human endometrial cancer RL95-2 cells and treatment with low to moderate BPA concentrations (10, 10<sup>3</sup> , and 10<sup>5</sup> nM).

Chou and colleagues reported that BPA exposure reduced miR-149 expression, downregulating the DNA repair gene *ARF6* (ADP ribosylation factor 6) and tumor protein p53 (*TP53*) and upregulating *CCNE2* (cyclin E2). The results of the study also showed that BPA was able to increase miR-107 to suppress hedgehog signaling factors, acting as a suppressor of fused homologs (*SUFU*) and GLI family zinc finger 3 (*GLI3*) and providing proof of the potential epigenetic mechanism of BPA exposure on endometrial carcinogenesis risk. In fact, miR107, miR149, miR200c, miR203, miR205, and miR765 changed the expression of some genes (*TP53*, *JUN*, *LAMB4*, *CCCDC6*, *PRKCA*, *STAT1*, *SUFU*, *CXCL8*, *DVL1*, *GLI3*, *CRK*, *LAMC1*, *MAPK1*, *MAPK9*) involved in the cancer pathway, recording a significant fold change of *N* > 2.0 compared to the control.

This study permitted the discovery and identification of five relevant pathways for potential BPA-induced endometrial cancer progression, including the cancer pathway, hedgehog pathway, cell cycle, adherens junction, and MAPK signaling pathway. In addition, *TP53*, *GLI3*, *CCNE2*, *CRK*, *KIF23*, *SAMD2*, *CCDC6*, *FZD3*, *ARF6*, *MAPK9*, *SUFU*, *PRC1*, *MDM2*, *SMAD4*, *DVL1*, *EGLN1*, *JUN*, *MYC*, *LAMC1*, *PRKACA*, and *STAT1* were genes that overlapped and were expressed significantly differently in these five pathways.

Meng et al. (2013) [32] developed an miRNA biosensor and applied this novel tool to detect miRNA-21 extracted from human hepatocarcinoma BEL-7402 cells and human mastocarcinoma MCF-7 cells and their expression under in vitro exposure to BPA. Normal human hepatic L-02 cells, BEL-7402 cells, and MCF-7 cells were incubated with 100 µM BPA at the same concentration for three and five days, respectively. The expression profiles of miRNA-21 in BEL-7402 and MCF-7 became 1.415-fold and 1.468-fold higher than that of normal L-02 cells, respectively, showing that the miRNA expression levels of cancer cells were upregulated compared to normal cells.

Li et al. (2014) [33] studied how microRNAs are involved in curcumin-mediated protection from BPA-associated induced effects on a breast cancer MCF-7 cell line. The MCF-7 cell line was exposed to BPA for 4 days. The results showed that BPA exhibited estrogenic activity by increasing the proliferation of estrogen-receptor-positive MCF-7 human breast cancer cells and promoting the transition of the cells from the G1 to S phase. Curcumin was able to inhibit the proliferative effects of BPA on MCF-7 cells. In addition, the BPA-induced upregulation of oncogenic miR-19a and miR-19b and the dysregulated expression of miR-19-related downstream proteins, including *PTEN*, *p-AKT*, *p-MDM2*, *p53*, and proliferating cell nuclear antigen, were sufficiently reversed by curcumin. Furthermore, Li and colleagues highlighted the important role of miR-19 in BPAmediated MCF-7 cell proliferation, suggesting for the first time that curcumin modulates the miR-19/*PTEN*/*AKT/p53* axis to exhibit its protective effects against BPA-associated breast cancer.

Kim et al. (2015) [34] used HepG2 cells that are widely used as a model system for studies of liver metabolism and genotoxicity. In particular, the authors determined the role of BPA exposure in the epigenetically affected expression of miR-22. The authors found methylated Chr17:1565786-1565940 regions (promoter sites for miR-22) in the normal samples, but unmethylated ones in samples exposed to BPA. Kim et al. identified seven differentially expressed miRNAs, including miR-22, in the BPA-exposed sample vs. the control. Notably, in samples exposed to BPA, miR-22 showed a 3.38-fold upregulation compared to normal samples. The study results highlight the regulation of miR-22 expression via hypomethylation of the promoter region due to BPA exposure.

Hui et al. (2018) [38] focused on BPA and ovarian cancer. This study was performed using in vitro exposure to BPA (10 or 100 nM) or 0.1% DMSO for 24 h using human ovarian adenocarcinoma SKOV3 cells, and then, the global gene expression profile was determined via high-throughput RNA sequencing. Transcriptomic analysis revealed 94 different expression genes related to tumorigenesis and metastasis.

The authors revealed the upregulation of miR-21-5p and miR-222-3p, also reporting that BPA (10 and 100 nM) increased migration and invasion as well as induced epithelial to mesenchymal transitions in SKOV3 and A2780 cells. Accordingly, doses of BPA found in the environment are capable of activating the regular Wnt signaling pathway. This study analyzed the possible mechanisms underlying the effects of BPA on ovarian cancer. Environmentally relevant doses of BPA modulated the gene expression profile and promoted the progress of epithelial to mesenchymal transition via the canonical Wnt signaling pathway of ovarian cancer.

Wu et al. (2018) [39] showed that BBP induced the proliferation of both ER(+) MCF-7 and ER(−) MDA-MB-231 breast cancer cells. This was proven by the increased cell viability, the transition of the cell cycle from the G1 to the S phase, the upregulation of *PCNA* and *Cyclin D1*, and the downregulation of *p21*. Moreover, BBP modulated the expression of the oncogenic miR-19a/b and *PTEN/AKT/p21* axis, revealing that miR-19 plays a crucial role in the promoting effect of BBP on breast cancer cells via the targeting of PTEN 3'UTR. These findings provide an important tool for targeted cancer intervention.

Zhu et al. (2019) [43] investigated the role of BBP in the cell proliferation of prostate cancer cells. Human prostate cancer LNCaP and PC-3 cell lines were exposed to low doses (0, 10−<sup>4</sup> , 10−<sup>5</sup> , 10−<sup>6</sup> , 10−<sup>7</sup> and 10−<sup>8</sup> mol/L) of BBP for 6 days. Zhu's results showed that 10−<sup>6</sup> and 10−<sup>7</sup> mol/L BBP increased the expression of *cyclinD1* and *PCNA*, decreased *p21* expression, and induced cell growth in both LNCaP and PC-3 cells vs. the control group. Furthermore, the authors found that BBP significantly downregulated the expression of miR-34a, along with upregulating miR-34a target gene *c-Myc*. Via cell transfection of an miR-34a mimic and inhibitor, the authors demonstrated that, in prostate cancer cells, the BBP trigger promoted cell proliferation mediated through the miR-34a/c-myc axis.

Duan et al. (2020) [45] also studied the effects of BBP on human acute monocytic leukemia AML U937 (isolated from the histiocytic lymph), Raji (lymphoblast Burkitt's lymph), and HL-60 (a promyelocytic cell line) cell lines, and normal blood cells.

BBP doses of 10−<sup>9</sup> and 10−<sup>4</sup> M were used for investigating the potential effect of BBP on the malignancy of AML cells. Instead for carrying out a mechanistic study, a dose of 10−<sup>8</sup> M was used. The authors examined the effects of BBP on the proliferation of AMLU937, Raji, and HL-60 cell lines. Moreover, the authors verified BBP's perturbation of treated U937 cells against the efficacy of chemotherapy using a double-exposure with increasing concentrations of daunorubicin or cytarabine with or without 10−<sup>8</sup> M BBP.

The results revealed that (10−<sup>8</sup> M) BBP can induce the proliferation and reduce the chemotherapy sensitivity of acute monocytic leukemia cells. *PDK1*, *PDK2*, *PDK3*, *PDK4*, *PDP2*, and *PDPR* genes can regulate the glucose metabolism and glycolysis of cancer

cells. In fact, cancer cells are characterized by high rates of glycolysis. The pyruvate dehydrogenase kinase (*PDK*) supports these energetic needs and also favors apoptosis resistance. Duan showed that BBP increased the expression of *PDK4* and *PDP2* in U937 cells, while in Raji cells, BBP only increased the expression of *PDK4*. This study confirmed that BBP can decrease the expression of miR-15b-5p, while it had no effect on miR-182 in both U937 and Raji cells. The overexpression of miR-15b-5p can abolish the BBP-induced mRNA and protein expression of *PDK4* in U937 cells. Furthermore, the inhibitor of miR-15b-5p can increase the mRNA and protein expression of *PDK4* in U937 cells.

Duan's results suggested that the downregulation of miR-15b-5p was involved in BBP-induced *PDK* and demonstrated that BBP can increase the mRNA stability of *PDK4* via the downregulation of miR-15b-5p.

Hence, BBP had no effect on the transcription and protein stability of *PDK4*; however, it significantly increased the mRNA stability of *PDK4.*

In Yin et al. (2018) [41], the global alterations of miRNA and mRNA expression in juvenile rat Sertoli cells (SCs) treated with 0.1 mM MBP were evaluated. Yin's results revealed that miR-3584-5p and miR-301b-3p were upregulated and their common target gene, Dexamethasone-induced Ras-related protein 1 (*Rasd1*), was downregulated. SC proliferation induced by low MBP concentration in vitro could be mediated by *Rasd1* regulation of the *ERK1/2* signaling pathway. These results represent a possible avenue to apply personalized medicine screening and therapy in testicular tumors induced by exogenous chemicals.

Cui et al. (2019) [44] studied the potential influence of MEHP, DEHP, DCHP, and BBP on the progression of hemangioma, one of the most common tumors of infancy. This in vitro study was carried out using hemangioma cells. The authors found that 100 nM of BBP can significantly trigger the migration and invasion of hemangioma cells, also inducing the overexpression of *Zeb1,* a powerful transcription factor for cell migration and invasion, via miR-655 suppression or downregulation. As 100 nM of BBP might also be found in human tissues, the potential health risks of BBP, particularly for oncologic HA patients, should be given more attention.

#### 3.3.2. In Vivo Studies

Buñay et al. (2017) [35] studied the consequences of chronic exposure to a mixture of phthalates and alkylphenols for the testes of male mice and in particular, reported changes in the expression patterns of miRNA/isomiRs, which act as regulators of gene expression in the testes. Additionally, damage to the testis and changes in the genes responsible for encoding proteins involved in the biogenesis, processing, editing, stability, or degradation of miRNAs were assessed. Buñay et al. carried out a case-control exposure study on a mix of phthalates and alkylphenols using adult male mice.

The exposed mice showed the degeneration of seminiferous tubules and hypertrophy/hyperplasia in Leydig cells and also an increase in exfoliation of germ cells of seminiferous tubules that close the lumen or showed fully closed tubules. Regarding mRNA levels, the authors report that the miRNAs of *Star* and *Cyp17a1* and *Sp1* and *Cyp11a1* were upregulated and downregulated, respectively. Instead, no significant differences in *Hsd3b1* mRNA expression were detected.

The authors quantified the mRNA expression levels of genes encoding proteins that are involved in pri-miRNA processing (*Drosha*), nuclear export (*Xpo5*), stability/degradation (*Lin28*, *Zcchc11*, *Zcchc6*, and *Snd1*), editing (*Adar*) and processing of pre-miRNAs (*Dicer*, *Ago2*).

A significant increase in the mRNA levels of *Drosha*, *Adar*, and *Zcchc11* in the testes of exposed mice was found compared to control mice, contrary to *Zcchc6*, *Dicer*, *Xpo5*, *Ago2*, *Lin28b,* and *Snd1*, which showed no differences.

miR20b-5p and miR-1291, which are implicated in cancer, and miR-3085-3p, implicated in inflammation, were all downregulated. miR-1291 targets DNA methyltransferases (*Dnmt3a*, *Dnmt3b*) that are involved in (de novo) histone methylation, genomic imprinting, X-chromosome inactivation, and testicular germ cell tumors due to exposure to alkylphenols. In addition, *Ccnd2*, *Ccnd1*, and *Raf1* are targets of the downregulated miR-15b-5p in exposed mice, and these targets are implicated in cancer and cell cycle regulation. Hence, this study suggests that the downregulation of sncRNAs through miR-1291 related to exposure to plasticizer mixtures might promote changes in the DNA methylation pattern, causing the epigenetic transmission of several diseases, including cancer.

In Chang et al. (2017) [36], the role of MEHP-induced reactive oxygen species (ROS) for genotoxicity was explained. Mono-ethylhexyl phthalate (MEHP) is a metabolite of DEPH. The toxicity of MEHP is more potent than that of DEPH. Chang's study provided evidence of the carcinogenicity of MEHP in Chinese hamster AA8, UV5, and EM9 ovary cells, as well as its ability to induce epigenetic modifications.

The cell lines were exposed to 0, 10, 25, and 50 mM MEHP. However, at 50 mM MEHP, all the cells died. The protection was not significant at 25 mM MEHP, and, even after exposure to a lower dose of MEHP (1 mM), the PARP-1-KD cells had a higher level of single-strand breaks. The subsequent *gpt* gene sequencing used to analyze the mutation points on the genes of AS52 mutant cells (ASMC) showed that 90% of all mutations were single-base pair substitutions, especially G:C to A:T mutations. Independent AS52-mutant cell clones were collected and used to perform sequential in vivo xenograft tumorigenic studies, and 4 of 20 clones had aggressive tumor growth. The study also showed that miR-let-7a and miR-125b has been downregulated in ASMC, which might raise oncogenic MYC and RAS levels and promote the activation of the *ErbB* pathway. The mutagenic pathway of MEHP can probably be triggered via the generation of ROS, causing base excision damage and resulting in carcinogenicity.

Wang et al. (2019) [48] sought to evaluate the capability of MEHP to promote the proliferation of oral cancer through an in vitro/in vivo study using human oral squamous carcinoma (OSCC) (human OSCC SCC-4, SCC-9, and SCC-25) cells and cell nuclear antigen (PCNA). SCC-4 cancer cells (2 × 106 per mouse) were diluted in 100 µL of normal medium and a researcher injected these subcutaneously into the left flank of each mouse to obtain OSCC cancer xenografts. When the tumor grew to 100 mm<sup>3</sup> , the mice of the MEHPs group were treated with MEHP (4 mg per kg, body weight) via intratumoral injection four times every three days. Tumor volume was measured every three days and, at the end of the experiment, mice were sacrificed and the xenograft tumors were removed to measure the expression of miRNAs and proteins.

The authors supported their hypothesis with results that showed the proliferation of oral cancer via MEHP through the downregulation of miR-27b-5p and miR-372-5p. In addition, MEHP induced the expression of *c-Myc*, which can suppress the transcription of miR-27b-5p in OSCC cells. Therefore, Wang's study showed that MEHP can promote the growth and progression of OSCC via the downregulation of miR-27b-5p and miR-372-5p.

Scarano et al. (2019) [41] studied the genome-wide levels of mRNAs to determine if perinatal exposure to a phthalate mixture in pregnant rats was capable of modifying gene expression during the prostate development of the filial generation. The study sought to determine the epigenetic role of these pollutants in prostate cancer.

Pregnant female Sprague Dawley rats were exposed daily (from gestational day 10 to postnatal day 21) to a mixture of phthalate by gavage and were suppressed after. Four groups were established—a control group exposed only to corn oil; (T1) 20 mg of the mixture (20 mg/kg/day); (T2) 200 mg of the mixture (200 mg/kg/day); and (T3) 200 mg of the mixture (200 mg/kg/day). The cocktail contained DEHP, DEP, DBP, DiBP, BBzP, and DiNP. The two lower doses mimicked daily human exposure levels based on the amount of DEHP, and the higher dose was selected to compare our results with those of similar phthalate studies. Rats from groups T1 to T3 received the respective doses of the phthalate cocktail prepared with 21% DEHP, 35% DEP, 15% DBP, 8% DiBP, 5% BBzP, and 15% DiNP. After birth, the number of F1 offspring per litter was reduced to 8 (at a 1:1 ratio between males and females whenever possible), and litters with fewer than six pups were suppressed.

miRNAs in the treated groups versus the control were upregulated in T1 vs. C and in T2 vs. C. miR-141-3p was exclusively upregulated in the T1 vs. C group, whereas other

miRNAs, such as miR-30d-5p, were deregulated in both groups with weak but significant alterations in gene expression. miRNA-184 was upregulated in all treatment groups vs. C. Among the possible targets for miR-141-3p (53 targets), 51 were downregulated. The MiR-NAs differentially expressed in the prostate tissue of these exposed animals were elicited in Table 1. Scarano's study, based on the evaluation of miRNAs and histopathological and immunostaining analyses, support the hypothesis of the epigenetic role of phthalate in prostate oncogenesis.

Chorley et al. (2020) [46] measured liver and blood miRNAs in male B6C3F1 mice exposed both to a known chemical activator of the peroxisome proliferator-activated receptor alpha (PPARα) and DEHP, respectively, for 7 and 28 days at concentrations of 0, 750, 1500, 3000, and 6000 ppm through oral exposure (feed). The PPARα pathway is a common target of several environmental chemicals. At the highest DEHP dose tested, 61 miRNAs were altered after 7 days, and 171 miRNAs after 28 days of exposure, with 48 overlapping miRNAs. Analysis of the 48 common miRNAs indicated the enrichment in PPARα–related targets and other pathways related to liver injury and cancer. The experiment was repeated using mmu-miRs-182-5p and -378a-3p analysis for DEHP, as well as di-n-octyl phthalate (DNOP) and n-butyl benzyl phthalate (BBP), two other related phthalates with weaker PPARα activity.

The results showed that the deregulatory potency of DEHP was superior to DNOP and BBP, and mmu-miRs-125a-5p, -182-5p, -20a-5p, and -378a-3p showed a clear dose relation linked to the PPARα pathway. These findings also highlight the putative miRNA biomarkers, as well as the stratified chemical potency of plasticizers and environmental pollutants in general.

Zota et al. (2020) [47] conducted the only human study included in this review. The Fibroids Observational Research on Genes and the Environment (FORGE) study involved 45 women living in Washington, DC, from 2014–2017. Eligible women were nonpregnant, pre-menopausal, English-speaking, and ≥18 years of age. The authors quantified the expression levels of 754 miRNAs in fibroid tumor samples and analyzed spot urine samples for phthalate metabolites collected from women undergoing surgery for fibroid treatment.

Associations between the miRNA levels in fibroids and phthalate biomarkers were also evaluated using a linear regression adjusted for age, race/ethnicity, and body mass index (BMI), and all the statistical tests were adjusted for multiple comparisons.

Fibroid tissues were collected during hysterectomy or myomectomy procedures. For patients with multiple fibroids, only the largest fibroid was sampled.

In addition, the evaluation of single metabolites was carried out, including diethyl phthalate (DEP), monoethyl phthalate (MEP), di-n-butyl phthalate (DnBP), mono-n-butyl phthalate (MnBP), mono-hydroxybutyl phthalate (MHBP), diisobutyl phthalate (DiBP), monoisobutyl phthalate (MiBP), mono-hydroxyisobutyl phthalate (MHiBP), butylbenzyl phthalate (BBzP), monobenzyl phthalate (MBzP), DnOP, mono(3-carboxypropyl) phthalate (MCPP), diisononyl phthalate (DiNP), monocarboxyoctyl phthalate (MCOP), diisodecyl phthalate (DiDP), monocarboxynonyl phthalate (MCNP), di(2-ethylhexyl) phthalate (DEHP), mono(2-ethylhexyl) phthalate (MEHP), mono(2-ethyl-5-hydroxyhexyl) phthalate (MEHHP), mono(2-ethyl-5-oxohexyl) phthalate (MEOHP), and mono(2-ethyl-5 carboxypentyl) phthalate (MECPP). The authors also calculated two summary measures, the molar sum of DEHP metabolites (ΣDEHP)21 and a potency-weighted sum of antiandrogenic phthalate metabolites (ΣAA phthalates).

The fibroid characteristics were similar across racial/ethnic groups. Phthalate exposure was ubiquitous in the enrolled woman, but nine phthalate metabolites were detected in >90% of participants. However, MEP levels were significantly higher in Black women. The enrolled women were Black (62%), overweight or obese (76%), privately insured (64%), and undergoing a myomectomy (58%). Compared with White/Latina women, Black women were more likely to be obese, publicly insured, and undergoing hysterectomy. The miRNA profiles detected were surprising with respect to social determinants.

A total of 35 miRNAs were underexpressed, and 39 miRNAs were overexpressed in fibroids rather than myometrium. Also, the expression of miR-10a-5p, miR-10a-3p, miR-140-3p, miR-144-5p, miR-150-5p, miR-205-5p, miR-27a-5p, miR-29b-2-5p, miR-29c-5p, miR-451a, and miR-95-3p was three-fold greater in myometrium; while expressions of miR-135a-5p, miR-135b-5p, miR-137-3p, miR-302b-3p, miR-335-3p, miR-34a-5p, miR-34a-3p, miR-34b-5p, miR-34c-5p, miR-483-5p, miR-488-3p, miR-488-5p, miR-508-3p, miR-577, miR-592, miR-651-5p, miR-885-5p, and miR-9-3pthese miRNAs were three-fold greater in fibroids.

The authors found 285 significant associations between phthalate biomarkers and miRNAs (*p* < 0.05), 34 of which were significant at *p* < 0.005.

After adjusting for multiple testing, we found two miRNAs associated with phthalate biomarkers—MHBP, associated with an increase in miR-10a-5p of 0.76 (95% CI = (0.40, 1.11)), and MEHHP, associated with miR-577 (β = 1.06, 95% CI = (0.53, 1.59)). Eight phthalate-miRNA associations varied significantly between White/Latina and Black women, and among these, there was an association between MBzP and miR-494-3p. Also, among white/Latina women, there were associations between MCPP and miR-337-5p; MBzP and miR-1227-3p; MEP and miR-645; MEP and miR-564; MEP and miR-374-5p; MEHP and miR-128-3p; and MEHP and miR-337-3p. Ten miRNAs were significantly associated with phthalate biomarkers either in the main analysis or in racial groups (miR-10a-5p, miR-577 miR-494-3p, miR-337-5p, miR-1227-3p, miR-645, miR-564, miR-374a-5p, miR-128-3p, miR-337-3p).

Zota et al. identified 923 mRNA targets that were experimentally observed or highly predicted targets of the 10 miRNAs, but 3 miRNAs (miR-10a-5p, miR-128-3p, miR-494-3p) were significantly associated with multiple fibroid-related processes, including angiogenesis, apoptosis, the proliferation of connective tissues, cell viability, tumorigenesis of the reproductive tract, and smooth muscle tumors.

miR-10a, miR-150, miR-29b, miR-29c, and miR-451 were underexpressed, and miR-34a was overexpressed in fibroids. The authors reported that miR-10a-5p expression in particular is associated with concentrations of MHBP, an oxidative metabolite of DnBP which is found in some personal care products, demonstrating that the epigenome is sensitive to interactions between chemical and non-chemical stressors, but also to social determinants that can influence a wide range of physical and social environmental exposures altering the biological response to environmental pollutants. On the basis of these results, the lack of human studies needs to be addressed urgently.

#### **4. Discussion and Conclusions**

The epigenetic effects of environmental chemicals such as plasticizers, including BPA and phthalates, on DNA methylation, as well as the expression of miRNAs, have substantiated our knowledge about the etiology of chronic diseases in humans, such as cancer. Evidence from in vitro and in vivo models has proved that epigenetic modifications due to exposure to common environmental pollutants can induce alterations in gene expression that may persist throughout life, increasing susceptibility to cancer. Epigenetics can affect the gene expression profiles of various organs and tissues. Among the phthalates, BPA, DEHP, MEHP, DBP, BBP, and MBP were found to cause 1232 and 265 interactions with the same genes and proteins, respectively.

This systematic review shows that miRNA-based diagnostic models can predict several targets of cancerous organs targets in humans with high accuracy. Also, the evidence regarding the carcinogenicity of several plasticizers was further supported by expression studies, permitting the future use of specific miRNA as valuable predictor or screening method for early diagnosis biomarkers as showed by Meng et al. study [32].

The use of profiling of miRNA as screening test through the high-throughput omic methods (microarrays and real-time quantitative PCR or qPCR, as well as real time PCR and nextgeneration sequencing) should be improved and applied in molecular early diagnosis to identify novel oncogenes, mechanisms, and/or pathways in which a stimuli, whether genetic or environmental, exerts a change on cell physiology to an oncological status.

Although the use of miRNAs is currently applied as a basic science tool, the overall miRNA's gene expression is moving from research laboratories to the large-scale clinical trials for the validation of a new diagnostic tool or for allowing clinical states to be determined in diseases such as cancer or other miRNA-diseases or altered gene expression related diseases. The use of miRNAs, as non-invasive tool of early diagnosis, need to be implemented in the clinical approach and miRNAs may be promising and effective candidates in the development of highly sensitive, noninvasive biomarkers for tumors screening prevention.

The miRNA-level changes can be useful for the toxicological assessment of several environmental pollutants, including plastic additives and plasticizers.

In this review, we showed that the interaction of plasticizers with several redundant miRNAs such as let-7f, let-7g, miR-125b, miR-134, miR-146a, miR-22, miR-192, miR-222, miR-26a, miR-26b, miR-27b, miR-296, miR-324, miR-335, miR-122, miR-23b, miR-200, miR-29a, and miR-21 might induce deep alterations in miRNA-mediated regulation and functions. These genotoxic and oncogenic responses can eventually lead to abnormal cell signaling pathways and metabolisms that participate in many intercrossed or overlapped cellular processes.

BPA induces the hypomethylation of histone promoter regions, indicating methylation changes as one of the possible mechanisms of BPA-induced adverse effects on carcinogenesis. BPA is also involved in the downregulation of gene repair ARF6 (involved in cell differentiation, apoptosis, and cell regulation), *TP53* (a tumor suppressor gene also referred to as the "Guardian of the Genome"), and over-regulates *CCNE2,* which is able to interact with *CDKN1A* and *CDKN1B* proteins, and with *CDK3*. The aberrant expression of *CCNE2* is a cause of cancer [48].

Phthalates downregulated the activity of some miRNAs (see Tab.1) implicated in cell cycle regulation and cancer. Additionally, the activation and overexpression of *ErbB*, *PPARα* pathways, the generation of ROS, and the overexpression of *Zeb1* (a transcription factor involved in cell migration and invasion) resulted from phthalate exposure.

It is important to note that the machinery by which plasticizers alter the epigenetic assets of cells require further study to elucidate the biology and biochemistry relatively to epigenetic alterations but also to disease-associated epigenetic alterations. A better understanding of these mechanisms will lead to better prediction of the health effects of plasticizers, allowing more targeted, easy, and appropriate disease-prevention and therapy strategies [49,50].

The lack of human studies needs to be addressed. Experimental evidence will permit the proposal of dedicated epidemiological studies to evaluate the real effects of plasticizers on human health, especially for cancer derived by microplastics and their plasticizers that are yet to be properly studied by oncologists yet.

**Author Contributions:** Conceptualization, M.F. and G.O.C.; Methodology, G.O.C.; Validation, M.F., A.C. and G.O.C.; Formal analysis, M.F., A.C. and G.O.C.; Investigation, M.F. and G.O.C.; Resources, M.F.; Data curation, A.C. and G.O.C.; Writing—original draft preparation, M.F. and G.O.C.; Writing review and editing, M.F. and G.O.C.; Visualization, M.F., A.C. and G.O.C.; Supervision, M.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

