https://orcid.org/0009-0007-3472-2916
Figures
Abstract
Breast cancer is defined as the unchecked growth of breast cells, with imbalances in prostaglandin and steroid hormone metabolism contributing to disease risk by altering prostaglandin types and forms (strong and weak) of steroid hormones. The AKR1C3 enzyme plays a key role in managing these metabolic processes. This study investigated the association between the AKR1C3 gene polymorphism (rs12529) and the risk of developing breast cancer in Bangladeshi individuals. A case-control investigation was conducted with a total of 620 samples, involving 310 individuals diagnosed with breast cancer and 310 healthy subjects. Herein, DNA extraction was performed via an organic process, whereas genotyping was employed via the PCR‒RFLP technique. Statistical assessments were conducted to analyze the association of polymorphisms, while molecular dynamics simulation and diverse computational techniques were employed to anticipate the functional and structural impacts of the SNP. Our study discovered that the rs12529 polymorphism of the AKR1C3 gene has an enhanced risk of susceptibility to breast malignancy (p = 0.016, OR = 1.97, 95% CI = 1.22 to 3.16 for the GG genotype in additive model 2). The recessive model (GG vs CC+CG) also showed an enhanced risk of susceptibility to breast malignancy (p = 0.0004, OR = 1.95, 95% CI = 1.40 to 2.73). In both premenopausal women and postmenopausal women, the GG genotype (for the recessive model) significantly increased breast cancer risk by 1.92-fold and 1.95-fold, respectively. However, no significant associations were observed regarding tumor grade or size in breast cancer development. In-silico analyses indicated that the H5Q (rs12529) mutation may decrease protein stability but is typically tolerated or functionally neutral. Molecular dynamics simulations revealed that H5Q leads to increased structural fluctuations and surface exposure, potentially causing the mutant AKR1C3 enzyme to operate differently from the wild type. In conclusion, rs12529 significantly increases the incidence of breast cancer in the population of Bangladesh. Computational analyses further revealed that the H5Q (rs12529) mutation in AKR1C3 leads to decreased stability and altered functional changes with notable conformational changes.
Citation: Shaon MA, Ansari F, Mahmud Z, Tamanna S, Ibrahim MNB, Alam NF, et al. (2025) Aldo-keto reductase family 1 member C3 (AKR1C3) gene polymorphism (rs12529) is associated with breast cancer in Bangladeshi population: A case-control study and computational investigation. PLoS One 20(6): e0318079. https://doi.org/10.1371/journal.pone.0318079
Editor: Luis Ramos, Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubiran: Instituto Nacional de Ciencias Medicas y Nutricion Salvador Zubiran, MEXICO
Received: April 12, 2024; Accepted: January 9, 2025; Published: June 9, 2025
Copyright: © 2025 Shaon et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The corresponding author, Md. Zakir Hossain Howlader (ZHH) acknowledges the partial funding of this work from the Integrated Health Research and Development Fund, Ministry of Health and Family Welfare, Bangladesh. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: AKR1C3, Aldo-ketoreductase family 1 member C3; BC, Breast cancer; SNP, Single nucleotide polymorphism; DHE, 17β-dihydroexemstane; ESR, Erythrocyte sedimentation rate; PR, Progesterone receptor; ER, Estrogen receptor; HER2, Human epidermal growth factor receptor type 2; EDTA, Ethylenediaminetetraacetic Acid; PCR-RFLP, Polymerase chain reaction-restriction fragment length polymorphism; rpm, revolution per minute; RMSD, Root mean square deviation; RMSF, Root mean square deviation; Rg, Radius of gyration; SASA, Solvent accessible surface area; BMI, Body mass index; AIC, Akaike information criterion; BIC, Bayesian information criterion; OR, Odds ratio; 95% CI, 95% confidence interval; Ref., Reference; ns, not significant; s, significant
1. Introduction
Cancer is characterized by uncontrolled and continuous cell proliferation that spreads to different body parts, infecting healthy tissues and organs [1]. Among women, breast carcinoma is the most frequently diagnosed form of cancer. It is considered a critical global health concern and stands as the leading cause of cancer-related mortality among women [2]. The majority of breast tumors arise from the cells lining the ducts and are referred to as ductal carcinomas. In addition, lobular malignancies develop in the cells lining the lobules, while a small percentage grow in other tissues [3]. Global Cancer Observatory (GLOBOCAN) 2022 reported that breast cancer was the most prevalent form of cancer for women (https://gco.iarc.who.int/media/globocan/factsheets/populations/900-world-fact-sheet.pdf) representing 23.8% of newly identified cases of female cancer worldwide. Moreover, according to GLOBOCAN 2022 data, Bangladesh reported 12,989 new breast cancer cases and 6,162 deaths, placing the disease fourth in terms of incidence and sixth in mortality within the country (https://gco.iarc.who.int/media/globocan/factsheets/populations/50-bangladesh-fact-sheet.pdf).
Understanding the risk factors for breast cancer is essential for early screening, which significantly enhances the likelihood of successful treatment outcomes. Breast cancer is more prevalent in developed nations due to factors such as early onset of menstruation, delayed age at first childbirth, absence of children, obesity, alcohol consumption, insufficient physical activity, and limited breastfeeding [4]. The variation in the proteins involved in their activation or inactivation plays a key role in assessing an individual’s risk of developing cancer [5]. Furthermore, it is crucial to guide therapeutic strategies by locating genetic markers associated with breast cancer development, as personalized medication is possible with the help of genetic risk profiles [6].
In the exploration of the genetic landscape of breast cancer [7,8], it has become evident that a hereditary component plays a pivotal role in the disease etiology, accounting for 5-10% of all cases [9]. Central to this genetic framework are the BRCA1 and BRCA2 genes, in which pathogenic variants significantly increase breast cancer risk [10], and other genes, such as PALB2, TP53, PTEN, CDH1, CHEK2, and ATM, each contributing to the multifaceted nature of breast cancer susceptibility [11,12]. These genetic factors highlight the autosomal dominant inheritance patterns in most hereditary breast cancer cases and the intricate interplay between high-penetrance genes and numerous moderate-to-low penetrance genes across various populations. Furthermore, alterations in genes such as MYC, ERBB2, FGFR1, GATA3 [13], and AKR1C3 [14,15] in tumor cells signal their critical roles in the progression of early breast malignancies. Our target gene, the Aldo-keto reductase family 1 member C3 (AKR1C3) gene, has been reported to be highly expressed in breast malignancies and is associated with tumor invasiveness, aggression, and resistance to chemotherapy [16,17].
The Aldo-keto reductase 1C (AKR1C) superfamily gene comprises four closely related isoforms (AKR1C1-AKR1C4) [18,19]. The AKR1C3 enzyme is translated from the AKR1C3 gene located on chromosome 10 at position p15-p14. This gene spans approximately 13 kb and has nine exons. Transcripts typically range in size, with the most common being approximately 1.4 kb [20]. The AKR1C3 protein is approximately 37 kilodaltons in molecular mass and consists of 323 amino acids. The three-dimensional form of the protein has been elucidated, revealing a typical (α/β)8-barrel crease, a characteristic of the Aldo-keto reductase superfamily. The catalytic center contains a deeply buried NADP(H) cofactor binding region, which is essential for the enzyme’s reduction activity. The structures also provide insights into the substrate specificity and catalytic mechanism of these enzymes [21].
The AKR1C3 enzyme metabolizes a wide range of substrates [22]. It mainly acts as a 17-ketoreductase that changes the attraction of hormone molecules to their specific receptors. For instance, AKR1C3 transforms 5α-androstane-3,17-dione and Δ4-androstene-3,17-dione into 5α-dihydrotestosterone and testosterone, which increases their attraction to androgen receptors [23]. AKR1C3 converts progesterone and estrone into 20α-hydroxyprogesterone and 17β-estradiol, respectively, controlling their ability to bind to estrogen receptors and progesterone receptors [24,25]. AKR1C3 controls cell growth and specialization by acting as a prostaglandin F2 (PGF2) synthase [26]. The conversion of hormones by AKR1C3 can enhance the development of breast cancer cells. Initially, two AKR1C3 metabolites, 17β-estradiol and 20α-OHP, support the growth of breast tumor cells [27]. AKR1C3 also promoted the production of PGF2 products, which in turn supported the proliferation of breast tumor cells [26]. Furthermore, AKR1C3 is involved in reducing endogenous reactive aldehydic or ketonic chemicals and aiding in their removal via the phase I biotransformation activity of other enzymes [28]. AKR1C3 can also process various external substances, such as numerous chemotherapeutic agents, utilizing carbonyl reductase action, leading to detoxification and resistance to chemotherapy [19,29].
Many single nucleotide polymorphisms (SNPs) have been found in the coding and noncoding regions of the AKR1C3 gene. Molecular epidemiology studies have established correlations among many of these SNPs and different illnesses. Nevertheless, these connections are present, and not all SNPs have functional consequences [30]. One of these SNPs is rs12529 which appears disproportionately in different ethnic backgrounds [31]. This evidence prompted the idea that the rs12529 variant may influence the advancement, intensity, or reaction to therapy differently in different populations in the case of illnesses where AKR1C3 has a crucial role [32–36]. Shiota et al. [33] reported that the rs12529 variant did not hinder the enzyme’s operation but may impact its physiological function by causing changes in gene expression, posttranslational modifications, or intracellular enzyme degradation. Again, a meta-analysis suggested that the altered allele (G) of rs12529 might be attributed to increased cancer risk in Asians [37]. In studies of the relationship between the rs12529 polymorphism and cancer, Lan Q et al. [35] reported that Chinese people who carry the rs12529 homozygous GG genotype had significantly increased susceptibility to lung cancer. Again, the occurrence of the rs12529 mutation was linked to alterations in physical well-being of life observed with androgen deprivation therapy (ADT) in prostate cancer, where the minor allele elicited a greater response. However, the impact of rs12529 (H5Q) on the process of androgen production still needs to be identified [36]. The rs12529 variant is a significant predictive factor when AKR1C3 inhibitors are created for clinical use [35]. The functional analysis showed that rs12529 (H5Q) had a kcat/Km reduction ranging from 40 to 200 times for the 17-ketosteroid reduction of exemestane. Reduced production of the active metabolite 17β-dihydroexemstane (DHE) leads to increased resistance to exemestane drugs in the case of breast cancer [33].
The primary goal of this investigation was to compare a small group of healthy females to breast cancer patients in order to evaluate the importance of the C15G (rs12529) variant as a potential risk factor. Additionally, the study explored the structural and functional changes in the AKR1C3 enzyme caused by this polymorphism (rs12529) using computational techniques such as molecular dynamics simulation and molecular docking.
2. Methods and materials
2.1 Study participants
This was a population-based case-control study in which breast cancer patients were considered cases. Participants with a previous history of any cancer were excluded from the study, as were individuals currently undergoing chemotherapy or any other cancer treatment. Additionally, participants with severe comorbidities such as cardiovascular diseases, autoimmune disorders, or conditions that could significantly alter genetic expression or immune responses were excluded. The control group consisted of age-matched healthy individuals without any history of chronic diseases, ensuring that the study focused on genetic associations without the confounding effects of pre-existing conditions or treatments. This study included a total of 620 participants, comprising 310 individuals diagnosed with breast cancer (cases) and 310 age-matched healthy individuals as controls. The research received approval from the Ethical Review Committee of the Department of Biochemistry and Molecular Biology at the University of Dhaka (Ref. No. BMBDU-ERC/EC/23/014). Furthermore, we affirm that all procedures conducted during this study adhered to the applicable guidelines and regulations.
Participants were selected from the National Institute of Cancer Research & Hospital (NICRH) in Dhaka, Bangladesh, during the period from July 1, 2023, to December 30, 2023. They were diagnosed with breast carcinoma using mammography, breast ultrasound, biopsy, or magnetic resonance imaging (MRI). All participants were female. During the same timeframe, control subjects were recruited from the National Institute of Ear, Nose, and Throat (NIENT) in Dhaka, Bangladesh. These individuals, all female, had no history or signs of cancer. Both the case and the control groups were recruited from Dhaka and matched for age and sex to minimize bias. As Dhaka is the capital and the primary hub for medical treatment in Bangladesh, attracting patients from all regions of the country [37], and predictably, we have found people from all divisions of Bangladesh. Additionally, the socio-economic status of the controls closely mirrored that of the breast cancer patients, with the majority of both groups coming from lower-income backgrounds.
Informed written consent was obtained from all participants prior to sample collection. Participants were fully briefed on the study’s objectives, procedures, and potential risks and were informed of their right to withdraw without consequence. All data were anonymized to ensure confidentiality by removing personal identifiers and assigning unique study codes. Electronic data were stored in a password-protected file, while paper records were kept in a locked file cabinet. Data analysis was performed in aggregate to safeguard participants’ privacy further. Data on sociodemographic factors, such as age, height, weight, family income, living area, educational background, medical history, menstrual and reproductive history, and family history of cancer, were collected using a structured questionnaire. Participants were also asked to provide comprehensive details about their breast cancer history, including pathological tumor grade, tumor size, age at diagnosis, prescribed medications, total white blood cell count (WBC), erythrocyte sedimentation rate (ESR), and biomarker status for progesterone receptor (PR), estrogen receptor (ER), and human epidermal growth factor receptor 2 (HER2).
2.2 Sample collection
Trained phlebotomists drew 5.0 mL of venous blood from each participant using disposable syringes under strict aseptic conditions. The collected blood was transferred into vacutainer tubes containing ethylene diamine tetra acetic acid (EDTA). Plasma was separated by centrifuging the samples at 3,000 rpm for 15 minutes. Both the plasma and cellular components were stored at -20°C for future analysis.
2.3 Genotyping of rs12529 (C15G)
DNA was isolated from the fractions of cells using the organic extraction method. Genotyping of rs12529 was conducted through polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). The PCR reactions were set up in a total volume of 15 µ L within a PCR tube containing 7.5 µ L of GoTaq G2 Green Master Mix (Promega Corporation), 5.05 µ L of nuclease-free water, 0.45 µ L of DMSO, 0.5 µ L each of forward and reverse primers, and 1 µ L of extracted DNA. The forward primer used was F (5′-TGCAATTTTCTCCACAGACCA-3′), and the reverse primer was R (5′-AAGCAGTACGTGACCATAGGA-3′). Details of the primer sequences are presented in S1 Table, while the PCR conditions are outlined in S2 Table.
The 450 bp PCR product was subjected to digestion using the AleI v2 restriction enzyme. The reaction, performed in a 15 μL volume at 37°C for 16 hours, produced two fragments of 189 bp and 261 bp in the presence of the reference C allele. For the mutant homozygous G/G genotype, no cleavage occurred, resulting in a single 450 bp fragment. The heterozygous C/G genotype yielded three fragments, measuring 450 bp, 261 bp, and 189 bp, respectively. The wild-type homozygous C/C genotype generated two fragments of 261 bp and 189 bp, as illustrated in Fig 1. To ensure reliable genotyping results, multiple quality control measures were implemented. Negative (no-template) controls were included in each PCR batch to detect contamination. Minimal PCR failures were reprocessed as needed. Duplicate PCRs were conducted for a subset of samples, and inconsistent replicates were re-analyzed to confirm results. PCR products were separated on 2% agarose gels and visualized under UV light. Two researchers independently performed genotyping results, with ambiguous results reviewed and retested for accuracy. Only samples with clear, reproducible results were included in the final analysis.
The presence of 261 bp and 189 bp on (from left) the 7th, 12th, and 13th wells indicates the existence of the homozygous wild-type C/C genotype. The 450 bp, 261 bp, and 189 bp in lanes 2nd, 4th, 8th, and 15th indicate the existence of a heterozygous mutant C/G genotype. In comparison, the 450 bp on lanes 3rd, 5th, 6th, 9th, 10th, 11th, 14th, 16th and 17th indicate the homozygous mutant G/G genotype. The first lane (from left) contains a 100 bp DNA ladder.
2.4 False-positive report probability (FPRP) analyses
The FPRP was determined to evaluate the reliability of the results [38]. A threshold of 0.2 was established for the FPRP, and prior probabilities of 0.25, 0.01, 0.001, and 0.0001 were used to identify an odds ratio (OR) of 1.5 associated with cancer risk. Results were deemed significant if their FPRP values were less than 0.2. The choice of an OR of 1.5 was based on standard practice in genetic epidemiology, where an OR of 1.5 is often used as a threshold for detecting moderate effects that are biologically plausible [39]. While this decision was not directly influenced by prior literature in this population, it aligns with typical thresholds for evaluating the significance of genetic associations [40].
2.5 Statistical analyses
We calculated the sample size using the G * Power program version 3.1.9.7 [41], [Type I error, α= 5% and predicted power, (1-β) = 80%]. The calculated sample size stated a number of at least 308 for each case and control group. Data analysis was conducted using GraphPad Prism (version 10.1.2) and the R programming language (version 4.1.2). Quantitative variables, such as age and BMI, were represented as the mean ± standard deviation (SD), while categorical variables were described as percentages (%). Differences in the means of continuous variables were evaluated using unpaired t-tests, and relationships between categorical variables were analyzed with two-tailed chi-square tests. The link of AKR1C3 gene rs12529 polymorphism with breast cancer was estimated by logistic regression according to four genotypic models (additive model 1, additive model 2, dominant model, and recessive model) and allele model using OR with 95% confidence intervals (CI). The Akaike information criterion (AIC) and Bayesian information criterion (BIC) are used to compare model fit while accounting for complexity. AIC rewards fit but penalizes excessive parameters to avoid overfitting [42], while BIC applies a more substantial penalty for complexity, particularly with larger sample sizes [43]. Based on these AIC and BIC values, the best-fit model was selected. The OR at the 95% CI was adjusted with age at menarche. A p-value of less than 0.05 indicated statistical significance in all analyses. The Bonferroni correction was applied in this study to control the risk of Type I errors due to multiple comparisons, the newly calculated p-value = (observed p-value × number of observations). Adjusting the observed p-value enhances the reliability of results in analyses with numerous tests, such as genotypic models and subgroup analyses. This p-value correction method maintains the integrity of the findings by controlling the family-wise error rate (FWER) and ensuring that associations are not due to random [44]. The FPRP analyses were performed using the gap package of R programming language [45]. Hardy-Weinburg equilibrium (HWE) was made using the “SHEsisPlus” (http://shesisplus.bio-x.cn/SHEsis.html) web tool [46,47].
2.6 In silico analysis
Various in-silico analysis tools were utilized following the methodologies outlined in our previously published research [48–50]. For example, SIFT (Sorting intolerant from tolerant) (https://sift.bii.a-star.edu.sg) [51] and Polyphen-2 (Polymorphism phenotyping v2) (http://genetics.bwh.harvard.edu/pph2) [52] were utilized for the prediction of the effect of the SNPs on the function of the protein. PredictSNP (https://loschmidt.chemi.muni.cz/predictsnp/) [53], SNAP (Semi-HMM-based nucleic acid parser) [54], PhD-SNP (Predictor of human deleterious single nucleotide polymorphisms) (https://snps.biofold.org/phd-snp/phd-snp.html) [55], and MAPP (Multivariate analysis of protein polymorphisms) (http://mendel.stanford.edu/SidowLab/downloads/MAPP/index.html) [56] were used to estimate disease associations with the SNPs. MUpro (https://mupro.proteomics.ics.uci.edu/) [57] and INPS-MD (Impact of nonsynonymous mutations on protein stability - multidimension) (https://inpsmd.biocomp.unibo.it/inpsSuite/) [58] were used to predict protein stability due to polymorphisms. The HOPE (https://www3.cmbi.umcn.nl/hope/) database was utilized to analyze changes in proteins due to amino acid substitution [59]. SWISS-MODEL (https://swissmodel.expasy.org/) [60] was used to construct a 3D model of the mutant protein using the Q95JH6.1 template from the AlphaFold database (https://alphafold.ebi.ac.uk/) [61,62]. Swiss model assessment, PROSA (https://prosa.services.came.sbg.ac.at/prosa.php) [63], and ERRAT (https://saves.mbi.ucla.edu/) [64] were used to validate the 3D model. Using AutoDock Vina v1.2.0 [65], the binding energy of protein-ligand docking was calculated.
2.7 Molecular dynamics (MD) simulations
The GROningen MAchine for Chemical Simulations (GROMACS) (version 2020.6) was utilized for a 100 nanosecond molecular dynamics simulation [66] for the ligand-protein complexes. The simulated scenario employs the CHARMM36m force field. A water box was created around the protein surface with corners located 1 nm away utilizing the TIP3 water model. The necessary ions were utilized to balance the systems. A simulation lasting 100 ns was conducted under periodic boundary conditions and a temporal integration step of 2 fs after energy minimization, isothermal isochoric (NVT) equilibration, and system isobaric (NPT) equilibration. The trajectory data were analyzed using a snapshot interval of 100 picoseconds. The GROMACS program’s embedded rms, rmsf, gyrate, sasa, and h-bond packages were utilized to analyze the simulation’s root mean square deviation (RMSD), root mean square fluctuation (RMSF), radius of gyration (Rg), and solvent accessible surface area (SASA). Plots for the investigations as mentioned earlier were produced utilizing the ggplot2 program in RStudio. At the Bioinformatics Division of the National Institute of Biotechnology, high-performance simulation stations utilizing the Ubuntu 20.04.4 LTS operating system were used for all MD simulations.
3. Results
3.1 Patients’ demographic characteristics
Among the 310 patients included in this study, 80% had no previous family history of cancer. Most of the patients (84.20%) were housewives by profession. The majority of the patients (64.20%) resided in rural areas of Bangladesh, which is reflected by their family income. 79.03% of patients had a family income lower than 20,000 BDT per month, and 27.42% had no formal education, with 17.10% and 35.48% with primary and secondary education, respectively. The demographic characteristics are shown in Fig 2.
(A) shows the patients’ family history of cancer, (B) shows the occupation of the patients, (C) represents the area of residence, (D) presents the range of family income of the patients, and (E) demonstrates the educational status of the patients.
3.2 Baseline characteristics of the participants
Table 1 presents the study subjects’ baseline characteristics according to age, BMI, menstrual history, age at first menstruation, and pregnancy count. The results revealed significant differences between BC patients and healthy controls regarding age at menarche but not age, BMI, menstrual status, or number of pregnancies.
3.3 Clinicopathological data of breast cancer patients
As shown in Table 2, the clinicopathological data showed that 52.58% of patients were diagnosed with cancer after 40 years of age. In this study, nearly all patients (98.06%) were diagnosed with invasive ductal carcinoma (IDC), while only 6 patients had invasive lobular carcinoma (ILC). For hormone receptor status, all patients (100%) were ER + and PR + , with 67.42% being HER2 + . Furthermore, 67.42% of patients had a tumor with a diameter greater than 2 cm (T2 and T3), and 64.84% of patients had grade 2 (G2) tumors.
3.4 Genotypic distribution of AKR1C3 rs12529 polymorphism and the risk of developing breast cancer
The genotype frequencies among control subjects were 19.40%, 44.80%, and 35.80% for homozygous wild type (Hz wild, CC), heterozygous variant (Ht variant, CG), and homozygous variant (Hz variant, GG), respectively. Among the patients, the frequencies were 14.50%, 33.50%, and 51.90% for Hz wild type (CC), Ht variant (CG), and Hz variant (GG), respectively.
The association analysis according to the additive model 1, additive model 2, dominant model, and recessive model is shown in Table 3. Additive model 2 was found to increase the statistically significant risk of breast cancer development (p = 0.016) in the case of the homozygous variant GG. The homozygous variant GG significantly increased the risk of breast cancer by 1.97 times (p = 0.016, OR = 1.97, 95% CI = 1.22 to 3.16). The recessive model showed a 1.95-fold increase in the risk of developing breast cancer, and the risk was found to be statistically significant (p = 0.0004, OR = 1.95, 95% CI = 1.40 to 2.73). The recessive model was identified as the best-fit model based on its lowest AIC (314) and BIC (322.86) values, indicating it provided the most parsimonious explanation of the data. These values were lower than those of the other models, which provided a statistically significant association, suggesting the recessive model best represents the relationship between AKR1C3 rs12529 genotype and breast cancer risk in our cohort. These findings support including the recessive model as the most appropriate for further analysis.
The allele frequencies were 0.4177 and 0.5823 for wild-type (C) and altered (G) alleles, respectively, among the control subjects. On the other hand, the frequencies were 0.3129 and 0.6871 for the wild-type and altered alleles, respectively, among the patients. The allele frequency analysis revealed that the risk of developing breast cancer was 1.58 times greater in G allele carriers (p = 0.0001, OR = 1.58, 95% CI = 1.25 to 1.99) than in those with the reference C allele.
The genotype distribution of rs12529 in the study participants is shown in Fig 3.
The GG genotype was dominant in the case (51.90%) group. The least frequently observed genotype was CC for both the case (14.50%) and control (19.40%) groups.
3.5 False-positive report probability (FPRP) results
Table 4 presents the FPRP values along with the statistical power for findings related to the AKR1C3 rs12529 polymorphism. Based on the FPRP analysis, the additive model 2, recessive model, and allelic model demonstrated a statistically significant association of rs12529 with breast cancer, with FPRP values of 0.10, 0.005, and 0.0001, respectively, at a prior probability threshold of 0.25.
3.6 Frequency distribution of rs12529 (C15G) genotypes in the study subjects according to menopausal status
Table 5 presents the distribution of the AKR1C3 rs12529 genotype in patients and control subjects according to menopausal status.
In the case of premenopausal women, the GG genotype carriers showed a 1.92 times increasing risk of the development of breast malignancy (p = 0.036, OR = 1.92, 95% CI = 1.18 to 3.09). The other models (additive model 1, additive model 2, and dominant model) didn’t show any statistically significant association with the risk of breast cancer development. Again, while analyzing the data of postmenopausal women, the same GG genotype carriers showed a 1.95 times enhanced risk of breast malignancy development (p = 0.008, OR = 1.95, 95% CI = 1.28 to 3.02). The rest of the three models (additive model 1, additive model 2, and dominant model) didn’t present any statistically significant association with the risk of breast cancer development.
3.7 Association of rs12529 (C15G) with tumor grade and size in breast cancer patients
In the patient group, polymorphism associations with tumor size and grade were analyzed. The results are presented in Tables 6 and 7. The alternative allele (G) showed no significant association with tumor size or grade.
3.8 Assessment of constancy in genotype frequency
Table 8 shows the results of the Hardy‒Weinberg equilibrium (HWE) test, which was used to evaluate the consistency of genetic frequency in the study subjects. According to the p-values, the genotypes of rs12529 were in the HWE for the healthy controls (p = 0.397), but in the case of breast cancer patients, the genotypes of the rs12529 deviated from the HWE.
3.9 In silico analysis
3.9.1 Analysis of functional consequences of rs12529 (H5Q).
Various in-silico tools were utilized to assess the impact of mutations on protein function. All tools classified the H5Q mutation as either tolerated or neutral. However, MUpro and INPS-MD predicted a reduction in the stability of the associated protein due to this mutation. The detailed predictions are summarized in Table 9, while the scores from different web-based tools are provided in S3 Table.
Additionally, the HOPE server was used to analyze several characteristics impacted by amino acid substitution. The H5Q mutation introduced an amino acid with distinct differences in size, charge, and hydrophobic properties. This alteration in amino acid composition led to a decrease in interactions and the disruption of hydrogen bonding. A summary of the key findings from the HOPE analysis can be found in S4 Table.
3.9.2 Homology modeling.
The complete three-dimensional structure of the human AKR1C3 protein was obtained from the AlphaFold database. The FASTA sequence (UniProt ID: P42330) was then utilized to model the 3D structure of the mutant AKR1C3 protein (H5Q) using Q95JH6.1.A (AlphaFold ID) as the template. The predicted structure was assessed using the SWISS-MODEL structure evaluation tool, ProSA-Web, and ERRAT, all of which indicated satisfactory results and high-quality models. The evaluation scores from these tools are detailed in S5 Table.
3.9.3 Protein-ligand docking.
The affinity between the native form of the protein-ligand complex and its mutant variant varies. Specifically, the interaction between the wild-type AKR1C3 protein and its ligand NADPH had a binding energy of -10.8 kcal/mol. This energy shifts to -9.2 kcal/mol when examining the interaction of the mutant H5Q AKR1C3 with NADPH.
3.9.4 Molecular dynamics (MD) simulation.
3.9.4.1 Root mean square deviation (RMSD) of the wild type and mutant (H5Q) AKR1C3 protein.
The stability of the protein-ligand complex was assessed through RMSD analysis. Variations in RMSD values indicate the extent of conformational adjustments in the protein triggered by the interaction with a ligand. In this context, the RMSD value for the native form of AKR1C3, in complex with NADPH as the ligand, was approximately 0.15 nm. Conversely, the RMSD for the mutant variant of AKR1C3, which also complexed with NADPH, was approximately 0.2 nm. This finding demonstrated a distinct pattern of conformational changes between the wild-type and mutant proteins. The wild-type AKR1C3-NADPH complex (green line) and mutant AKR1C3-NADPH complex (purple line) are depicted in Fig 4.
The pattern of conformational change in wild-type AKR1C3 (green line) was significantly different from that in mutant AKR1C3 (purple line).
3.9.4.2 Root mean square fluctuation (RMSF) of the wild-type and mutant (H5Q) AKR1C3 protein.
The RMSF graph shows the mobility of wild-type (green line) AKR1C3 (complexed with the ligand, NADPH) and mutant (purple line) AKR1C3 (complexed with the ligand, NADPH). From the 1st amino acid to the 100th amino acid of wild-type AKR1C3 (complexed with the ligand, NADPH), two peaks corresponding to RMSF were observed, whereas no peak corresponding to mutant AKR1C3 (complexed with the ligand, NADPH) was detected. Five more peaks were detected from the 100th amino acid to the end of the mutant AKR1C3 protein than those in the wild-type AKR1C3 protein (Fig 5).
The mutant version exhibited distinct RMSF patterns, with two major peaks in the first 100 amino acids absent in the mutant and five significant peaks beyond the 100th amino acid that were higher than those observed in the wild type.
3.9.4.3 The radius of gyration of wild and mutant (H5Q) AKR1C3 complexed with the ligand NADPH.
The Rg of the mutant (purple line) AKR1C3 (complexed with the ligand NADPH) was relatively unstable. In contrast, the Rg of the wild type (green line) AKR1C3 (complexed with the ligand NADPH) was relatively stable during the simulation (Fig 6).
The Rg value of wild-type (green line) AKR1C3 was relatively stable during the simulation.
3.9.4.4 Solvent accessible surface area (SASA) of wild-type and mutant (H5Q) AKR1C3 complexed with the ligand NADPH.
The SASA values of wild-type (green line) AKR1C3 (complexed with the ligand, NADPH) and mutant (purple line) AKR1C3 (complexed with the ligand, NADPH) significantly differed throughout the simulation (Fig 7). The wild-type variant had a lower SASA value than the mutant type.
4. Discussion
Breast cancer is the most prevalent form of cancer affecting women. It is relatively common in countries with substantial economies; although it is currently modest, the rate is increasing in emerging nations. This disease is the result of a mix of hereditary and environmental causes. Although the precise mechanism of breast cancer development is not yet known, several proven risk factors, including early menarche, late menopause, age at first childbirth, infertility, and genealogy, have been previously identified [67]. Furthermore, exposure to environmental variables such as ionizing radiation and carcinogens from chemicals has been linked to increased chances of developing breast cancer [68].
AKR1C3 displays pluripotency through varying levels of 3-keto-, 17-keto-, and 20-ketosteroid reductase activity [69]. Its 17-ketosteroid reductase function distinguishes the enzyme as 17β-HSD (17β-hydroxysteroid dehydrogenase) type 5, which is responsible for generating testosterone through an intracrine process and is different from the androgenic 17β-HSD (type 3 17β-hydroxysteroid dehydrogenase) [70] and the estrogenic 17β-HSD (type 1 17β-hydroxysteroid dehydrogenase) [71,72]. Therefore, any nonsynonymous mutation that increases the activity of the 17β-HSD type 5 form of AKR1C3 can increase testosterone availability. This excessive testosterone can further be converted into more potent estradiol by aromatase, which can initiate more growth responses in breast cells. In addition, this mutation can also alter the amount of other androgens and estrogens by affecting the structural and functional activities of androgenic 17β-HSD (type 3 17β-hydroxysteroid dehydrogenase) and estrogenic 17β-HSD (type 1 17β-hydroxysteroid dehydrogenase).
The 3-ketosteroid reductase function of AKR1C3 is represented by the enzyme 3α-HSD type 2, whose function is to convert the potent steroid hormone 5α-dihydrotestosterone (5α-DHT) into weak 5α-androstane-3α,17β-diol [18,69,73]. Thus, it reduces the amount of testosterone and its potent derivative 5α-dihydrotestosterone (5α-DHT). Therefore, any kind of nonsynonymous mutation that decreases the activity of the 3α-HSD type 2 form of AKR1C3 can inhibit the conversion of 5α-dihydrotestosterone (5α-DHT) into the weak 5α-androstane-3α,17β-diol. Therefore, more 5α-dihydrotestosterone (5α-DHT) and its immediate previous form, testosterone, can be further converted into the more potent estradiol by aromatase, which can initiate more growth responses in breast cells. AKR1C3 also acts as a prostaglandin (PG) F2α synthase and transforms PGH2 to prostaglandin F2α (PGF2α) and prostaglandin D2 (PGD2) to 11β-prostaglandin F2α (11β-PGF2α) [74,75]. PGF2α epimers function as ligands for the FP receptor, stimulating cell proliferation pathways [20]. Therefore, any gain-of-function mutation can promote the production of 11β-PGF2α and result in excessive breast growth and other cells.
This study aimed to evaluate the relationship between the rs12529 polymorphism in the AKR1C3 gene and the risk of breast cancer within the Bangladeshi population. Furthermore, the effects of this polymorphism on the AKR1C3 protein were analyzed using various in-silico tools and molecular dynamics simulations. The rs12529 (H5Q) variant shows varying frequencies across different ethnic groups [31]. This polymorphism has been well studied for its relation with various types of malignancies, e.g., prostate cancer [32], lung cancer [31], and bladder cancer [76], in different populations. However, rs12529 is comparatively less studied for its association with breast cancer. Its association with breast cancer risk has been studied in only 2 populations, i.e., the French and Canadian populations, and there was no significant association between rs12529 and breast cancer risk [77]. In South Asia, the association between rs12529 and breast cancer has not yet been studied. In Bangladesh, various studies have explored the associations between breast cancer and different gene polymorphisms, including rs13181 of ERCC2, rs2276466 of ERCC4, rs80357713 and rs80357906 of BRCA1, rs11571653 of BRCA2, rs1136201 of HER2, rs1042522 of TP53, rs16260 of CDH1, rs25487 of XRCC1, rs861539 of XRCC3, as well as rs1219648, rs2420946, and rs2981582 of the FGFR2 gene [78–82]. Consequently, this study represents the first to investigate the link between rs12529 of the AKR1C3 gene and breast cancer across the entire South Asian population.
This study examined the relationship between the AKR1C3 gene rs12529 (H5Q) polymorphism and breast cancer in a specific group of people from the Bangladeshi community. We analyzed the AKR1C3 genotypes of 620 individuals, including 310 healthy individuals and 310 breast cancer patients. Both groups were similar in age, BMI, menstrual status, and number of pregnancies but were significantly different in age at menarche (Table 1). This study revealed a positive correlation between this genetic variation and the development of breast cancer in Bangladeshi individuals.
In our study, the AKR1C3 rs12529 genotypes and genotype frequencies varied significantly between the two groups, and the frequency of the G allele was greater in BC patients than in healthy individuals (Table 3), although our results are not consistent with the results of Plourde et al. (2009) [77]. We observed a significantly increased risk of breast cancer due to rs12529 in additive model 2 (GG vs CC) and recessive model (GG vs CC+CG). The risk of developing breast cancer was 1.97-fold (OR = 1.97; 95% CI = 1.22 to 3.16; p = 0.004) greater in breast cancer patients with the GG genotype (in the case of additive model 2). Women with the GG genotype (recessive model) were 1.95 times more likely to develop breast cancer, and this difference was statistically significant (OR = 1.95; 95% CI = 1.40 to 2.73; p = 0.0004). When stratified against menopausal status, we found a statistically significant association between rs12529 and the risk of breast cancer development in both pre-menopausal women and post-menopausal women. The GG genotype (recessive model) significantly increased breast cancer risk by 1.92- and 1.95-fold, respectively, in the case of premenopausal and postmenopausal women.
This association between the AKR1C3 polymorphism (rs12529) and breast cancer risk in the Bangladeshi population is consistent with findings from other Asian populations, where the G allele of rs12529 has been found to be linked to an increased risk of cancer [35]. However, studies in European populations, such as those in France and Canada, have not found significant associations between rs12529 and breast cancer risk [77]. These discrepancies may be due to differences in allele frequencies, genetic backgrounds, or environmental factors across ethnic groups. In South Asia, the G allele appears to be more prevalent, suggesting that this polymorphism may have a more substantial effect in these populations. Comparing these results with global studies, especially considering the varying allele frequencies, will enhance our understanding of the genetic risk factors for breast cancer and may inform personalized approaches to prevention and treatment.
In this study, we found no significant association between the rs12529 (H5Q) polymorphism and tumor grade or size in breast cancer patients. This suggests that while rs12529 may influence breast cancer susceptibility, it may not directly affect tumor aggressiveness or other clinicopathological features. These findings highlight the complex nature of breast cancer biology, where genetic, environmental, and hormonal factors shape tumor progression. Previous studies [35,77] have not explored the potential link between rs12529 (H5Q) and tumor grade or size in breast cancer.
In the case of HWE test results, the p-value of 0.397 indicates that the genotype frequencies are consistent with HWE for the healthy control subjects. This suggests that, in the control group, the alleles at the rs12529 locus segregate randomly and that no factors such as selection, mutation, or non-random mating influence these frequencies significantly. In contrast, the breast cancer patients exhibit a significant deviation from HWE, as evidenced by a p-value of 0.0003 and a high chi-square statistic (χ² = 16.08). This deviation could indicate that one or more evolutionary forces are acting on the rs12529 locus within this group. One potential explanation for the deviation from HWE in cases could be population stratification or substructure within our study population. While we attempted to recruit participants from a homogenous population, subtle genetic substructures might still exist. Another factor that could contribute to the deviation from HWE is genotyping errors. Moreover, the possibility of sampling bias due to the relatively small sample size may contribute to the deviation from HWE. Overall, deviations from HWE may occur due to natural selection related to breast cancer, population stratification, or genotyping errors. However, the latter is less likely, given that the control group is in equilibrium.
The utilization of various in-silico tools to assess the effect of the H5Q mutation on protein function yielded insights into its tolerance or neutral nature, with MUpro and INPS-MD specifically revealing a decrease in protein stability. The HOPE server analysis further delineated the impact of the mutation, emphasizing alterations in amino acid properties that led to disrupted interactions and hydrogen bonds. Notably, the mutation affected the binding energy of the ligand NADPH, indicating an alteration in affinity for the mutant AKR1C3. These findings collectively suggest that while the H5Q mutation is biochemically tolerated, it induces subtle yet significant changes in the structural and functional landscape of the protein, potentially influencing its biological role and therapeutic targeting strategies. These findings are consistent with the results reported by Shiota et al. [34], who concluded that the rs12529 variation could impact the physiological roles of AKR1C3 by modifying its transcriptional process, posttranslational changes, or intracellular degradation of proteins.
Molecular dynamics simulation
A molecular dynamics simulation study on rs12529 revealed notable differences in the behavior of the AKR1C3 enzyme between the wild type and its mutant form when both were bound to NADPH. The RMSD analysis showed that the wild-type AKR1C3-NADPH complex had lower RMSD values, suggesting minor conformational shifts relative to the mutant form. This observation indicated that ligand binding induces more pronounced structural changes in the mutant protein. Furthermore, RMSF data indicated altered flexibility patterns in the mutant AKR1C3, potentially affecting its function or structure. The introduction of the H5Q mutation was found to destabilize AKR1C3, as evidenced by fluctuating Rg values, suggesting conformational changes and decreased protein compactness. Additionally, the mutant exhibited a greater SASA, indicating greater surface exposure, which could have implications for enzymatic activity or interaction capabilities, highlighting the structural and functional impact of the rs12529 mutation on AKR1C3.
Molecular dynamics simulations and in-silico analyses provided key insights into the functional impact of the rs12529 (H5Q) mutation in AKR1C3. The mutation was found to reduce protein stability, increase structural fluctuations, and alter ligand binding affinity. These changes may impair the enzyme’s ability to metabolize key steroids, such as testosterone and dihydrotestosterone (DHT), which are involved in estrogen biosynthesis via the aromatase pathway. Given estrogen’s role in breast cancer proliferation, this alteration could enhance the production of potent estrogens, such as estradiol, promoting cancer progression. Additionally, the mutation may affect AKR1C3’s role in prostaglandin synthesis, further contributing to tumorigenesis. Previous functional analyses also showed that rs12529 (H5Q) decreases the kcat/Km for 17-ketosteroid reduction by AKR1C3, supporting its potential impact on breast cancer [33]. Our findings highlight AKR1C3 as a potential therapeutic target in breast cancer, particularly for individuals with H5Q mutation. Targeting AKR1C3 with enzyme inhibitors or steroid metabolism modulators could offer personalized treatment strategies. Further research on AKR1C3 variants and their impact on estrogen synthesis and prostaglandin pathways is essential for understanding their role in breast cancer pathology.
Our investigation revealed limitations that must be considered when interpreting our results. First, the sample size was not large. Furthermore, we focused on only a single variant; considering the multitude of mutations that can impact the way a gene functions, it seems insufficient to make conclusive findings. Future studies require larger sample sizes, different populations, and additional genetic variations. There is a significant opportunity for further investigations into the expression rates of AKR1C3 variants in BC patients and for examining the specific underlying biological mechanisms. Consequently, future investigations should consider these findings to uncover this connection.
Overall, our study identified a genetic variant that is significantly associated with breast cancer risk. This finding supports the role of genetic predisposition in the pathogenesis of breast cancer, highlighting specific gene variants and pathways that may be involved in the initiation and advancement of the illness. Moreover, our findings have important implications for developing personalized medicine approaches to breast cancer. By identifying genetic markers associated with increased risk, we can improve risk stratification and early detection strategies. This can lead to more tailored screening programs and potentially earlier diagnosis, which is crucial for improving patient outcomes. Additionally, understanding the role of these genes in breast cancer pathogenesis opens up possibilities for the development of targeted therapies aimed at specific molecular pathways, which could enhance treatment efficacy and reduce side effects.
In conclusion, we found an association between an AKR1C3 gene polymorphism (rs12529) and breast malignancy in Bangladeshi individuals. The frequency of the G allele was notably greater in breast cancer individuals than in control subjects. Thus, the GG genotype is considered a risk factor, and the CC genotype is a molecular marker for reducing breast cancer incidence. In-silico analyses and MD simulations suggested that the H5Q mutation in AKR1C3 results in decreased stability and increased structural fluctuations, indicating a potentially tolerable or neutral effect on function despite significant conformational changes compared to those of the wild type. Therefore, genotyping of the AKR1C3 (rs12529) gene could be a biomarker for the early diagnosis of breast cancer and could aid in personalized medicine as a novel therapy for breast carcinoma.
Supporting information
S1 Table. Primers used to amplify target regions.
https://doi.org/10.1371/journal.pone.0318079.s002
(PDF)
S2 Table. PCR conditions used to amplify the target regions.
https://doi.org/10.1371/journal.pone.0318079.s003
(PDF)
S3 Table. Scores from different tools for the analysis of functional consequences of the target SNPs.
https://doi.org/10.1371/journal.pone.0318079.s004
(PDF)
S4 Table. A depiction of the highlights of the HOPE predictions.
https://doi.org/10.1371/journal.pone.0318079.s005
(PDF)
S5 Table. Evaluation scores of the predicted protein models.
https://doi.org/10.1371/journal.pone.0318079.s006
(PDF)
References
- 1. Sarkar S, Horn G, Moulton K, Oza A, Byler S, Kokolus S, et al. Cancer development, progression, and therapy: an epigenetic overview. Int J Mol Sci. 2013;14(10):21087–113. pmid:24152442
- 2. Omran MH, Fotouh BE, Shousha WG, Ismail A, Ibrahim NE, Ramadan SS. Strong Correlation of MTHFR gene polymorphisms with breast cancer and its prognostic clinical factors among Egyptian females. Asian Pac J Cancer Prev. 2021;22(2):617–26. pmid:33639682
- 3. Lei S, Zheng R, Zhang S, Wang S, Chen R, Sun K, et al. Global patterns of breast cancer incidence and mortality: A population-based cancer registry data analysis from 2000 to 2020. Cancer Commun (Lond). 2021;41(11):1183–94. pmid:34399040
- 4. Alam NF, Ahmed R, Mahmud Z, Tamanna S, Shaon MA, Howlader MZH. Genetic association and computational analysis of MTHFR gene polymorphisms rs1801131 and rs1801133 with breast cancer in the Bangladeshi population. Sci Rep. 2024;14(1):24232. pmid:39414907
- 5. Nabil IK, Mahmud Z, Tamanna S, Rahman MM, Rashid G, Hossain MS, et al. Vitamin D deficiency and the vitamin D receptor (VDR) gene polymorphism rs2228570 (FokI) are associated with an increased susceptibility to hypertension among the Bangladeshi population. PLoS One. 2024;19(3):e0297138. pmid:38483874
- 6. Saba AA, Mahmud Z, Ansari F, Ahmed R, Nur J, Alam MS, et al. Single nucleotide variants rs7975232 and rs2228570 within vitamin D receptor gene confers protection against severity of COVID-19 infection in Bangladeshi population. Gene Reports. 2024;36:101981.
- 7. Nila NN, Tamanna S, Mahmud Z, Washif M, Ahmad T, Rahman ATMA, et al. The IKAROS transcription factor gene IKZF1 as a critical regulator in the pathogenesis of childhood acute lymphoblastic leukemia: insights from a Bangladeshi population. Ind J Clin Biochem. 2024.
- 8. Tamanna S, Jannat T, Devi S, Mahmud Z, Mahmood MS, Faisal MF, et al. Exploring the Associations of maternal and neonatal eNOS gene variant rs2070744 with nitric oxide levels, oxidative stress and adverse outcomes in preeclampsia: a study in the Bangladeshi Population. Ind J Clin Biochem. 2024.
- 9. Apostolou P, Fostira F. Hereditary breast cancer: the era of new susceptibility genes. Biomed Res Int. 2013;2013747318. pmid:23586058
- 10. Lee A, Moon BI, Kim TH. BRCA1/BRCA2 pathogenic variant breast cancer: treatment and prevention strategies. Ann Lab Med. 2020;40(2):114–21. pmid:31650727
- 11. Harbeck N, Penault-Llorca F, Cortes J, Gnant M, Houssami N, Poortmans P, et al. Breast cancer. Nat Rev Dis Primers. 2019;5(1):66. pmid:31548545
- 12. Taylor A, Brady AF, Frayling IM, Hanson H, Tischkowitz M, Turnbull C, et al. Consensus for genes to be included on cancer panel tests offered by UK genetics services: guidelines of the UK Cancer Genetics Group. J Med Genet. 2018;55(6):372–7. pmid:29661970
- 13. Nik-Zainal S, Davies H, Staaf J, Ramakrishna M, Glodzik D, Zou X, et al. Landscape of somatic mutations in 560 breast cancer whole-genome sequences. Nature. 2016;534(7605):47–54. pmid:27135926
- 14. Zeng C-M, Chang L-L, Ying M-D, Cao J, He Q-J, Zhu H, et al. Aldo-Keto Reductase AKR1C1-AKR1C4: Functions, Regulation, and Intervention for Anti-cancer Therapy. Front Pharmacol. 2017;8119. pmid:28352233
- 15. Bukum N, Novotna E, Morell A, Hofman J, Wsol V. Buparlisib is a novel inhibitor of daunorubicin reduction mediated by aldo-keto reductase 1C3. Chem Biol Interact. 2019;302101–7. pmid:30703376
- 16. Penning TM. AKR1C3 (type 5 17β-hydroxysteroid dehydrogenase/prostaglandin F synthase): Roles in malignancy and endocrine disorders. Mol Cell Endocrinol. 2019;48982–91. pmid:30012349
- 17. Khanna M, Qin KN, Wang RW, Cheng KC. Substrate specificity, gene structure, and tissue-specific distribution of multiple human 3 alpha-hydroxysteroid dehydrogenases. J Biol Chem. 1995;270(34):20162–8. pmid:7650035
- 18.
Lin H-K, Jez JM, Schlegel BP, Peehl DM, Pachter JA, Penning TM. Expression and characterization of recombinant type 2 3-hydroxysteroid dehydrogenase (HSD) from human prostate: Demonstration of bifunctional 3/17-HSD activity and cellular distribution. 1971.
- 19. Hofman J, Malcekova B, Skarka A, Novotna E, Wsol V. Anthracycline resistance mediated by reductive metabolism in cancer cells: the role of aldo-keto reductase 1C3. Toxicol Appl Pharmacol. 2014;278(3):238–48. pmid:24832494
- 20. Penning TM, Steckelbroeck S, Bauman DR, Miller MW, Jin Y, Peehl DM, et al. Aldo-keto reductase (AKR) 1C3: role in prostate disease and the development of specific inhibitors. Mol Cell Endocrinol. 2006;248(1–2):182–91. pmid:16417966
- 21. Penning TM, et al. Human 3α-hydroxysteroid dehydrogenase isoforms (AKR1C1-AKR1C4) of the aldo-keto reductase superfamily: functional plasticity and tissue distribution reveals roles in the inactivation and formation of male and female sex hormones. Journal Name. 2000;Volume Number(Issue Number):Page Range.
- 22. Byrns MC, Steckelbroeck S, Penning TM. An indomethacin analogue, N-(4-chlorobenzoyl)-melatonin, is a selective inhibitor of aldo-keto reductase 1C3 (type 2 3alpha-HSD, type 5 17beta-HSD, and prostaglandin F synthase), a potential target for the treatment of hormone dependent and hormone independent malignancies. Biochem Pharmacol. 2008;75(2):484–93. pmid:17950253
- 23. Penning TM. Aldo-Keto Reductase (AKR) 1C3 inhibitors: a patent review. Expert Opin Ther Pat. 2017;27(12):1329–40. pmid:28895472
- 24. Chewchuk S, Guo B, Parissenti AM. Alterations in estrogen signalling pathways upon acquisition of anthracycline resistance in breast tumor cells. PLoS One. 2017;12(2):e0172244. pmid:28196134
- 25. Mantel A, Newsome A, Thekkudan T, Frazier R, Katdare M. The role of aldo-keto reductase 1C3 (AKR1C3)-mediated prostaglandin D2 (PGD2) metabolism in keloids. Exp Dermatol. 2016;25(1):38–43. pmid:26308156
- 26. Zhong T, Xu F, Xu J, Liu L, Chen Y. Aldo-keto reductase 1C3 (AKR1C3) is associated with the doxorubicin resistance in human breast cancer via PTEN loss. Biomed Pharmacother. 2015;69317–25. pmid:25661377
- 27. Clarke L, Zheng-Bradley X, Smith R, Kulesha E, Xiao C, Toneva I, et al. The 1000 Genomes Project: data management and community access. Nat Methods. 2012;9(5):459–62. pmid:22543379
- 28. Figueroa JD, Malats N, García-Closas M, Real FX, Silverman D, Kogevinas M, et al. Bladder cancer risk and genetic variation in AKR1C3 and other metabolizing genes. Carcinogenesis. 2008;29(10):1955–62. pmid:18632753
- 29. Cunningham JM, Hebbring SJ, McDonnell SK, Cicek MS, Christensen GB, Wang L, et al. Evaluation of genetic variations in the androgen and estrogen metabolic pathways as risk factors for sporadic and familial prostate cancer. Cancer Epidemiol Biomarkers Prev. 2007;16(5):969–78. pmid:17507624
- 30. Liu C, Hsu Y-H, Pan P-C, Wu M-T, Ho C-K, Su L, et al. Maternal and offspring genetic variants of AKR1C3 and the risk of childhood leukemia. Carcinogenesis. 2008;29(5):984–90. pmid:18339682
- 31. Lan Q, Mumford JL, Shen M, Demarini DM, Bonner MR, He X, et al. Oxidative damage-related genes AKR1C3 and OGG1 modulate risks for lung cancer due to exposure to PAH-rich coal combustion emissions. Carcinogenesis. 2004;25(11):2177–81. pmid:15284179
- 32. Karunasinghe N, Zhu Y, Han DY, Lange K, Zhu S, Wang A, et al. Quality of life effects of androgen deprivation therapy in a prostate cancer cohort in New Zealand: can we minimize effects using a stratification based on the aldo-keto reductase family 1, member C3 rs12529 gene polymorphism?. BMC Urol. 2016;16(1):48. pmid:27485119
- 33. Platt A, Xia Z, Liu Y, Chen G, Lazarus P. Impact of nonsynonymous single nucleotide polymorphisms on in-vitro metabolism of exemestane by hepatic cytosolic reductases. Pharmacogenet Genomics. 2016;26(8):370–80. pmid:27111237
- 34. Shiota M, Endo S, Fujimoto N, Tsukahara S, Ushijima M, Kashiwagi E, et al. Polymorphisms in androgen metabolism genes with serum testosterone levels and prognosis in androgen-deprivation therapy. Urol Oncol. 2020;38(11):849.e11-849.e18. pmid:32712140
- 35. Zhu G, Li X, Wang Y. Association of the AKR1C3 polymorphism with cancer risk: a meta-analysis and systematic review. Int J Clin Exp Med. 2017;10(5):1234–45.
- 36. Fiandalo MV, Gewirth DT, Mohler JL. Potential impact of combined inhibition of 3α-oxidoreductases and 5α-reductases on prostate cancer. Asian J Urol. 2019;6(1):50–6. pmid:30775248
- 37. Fatima P, Hossain MM, Rahman D, Hossain HB, Hossain HN, Hossain FA, et al. Treatment Seeking Behavior and the Profile of Infertile Patients Attending the Tertiary Infertility Center at Dhaka. J Bangladesh Coll Phys. 2017;34(3):140–4.
- 38. Wacholder S, Chanock S, Garcia-Closas M, El Ghormli L, Rothman N. Assessing the probability that a positive report is false: an approach for molecular epidemiology studies. J Natl Cancer Inst. 2004;96(6):434–42. pmid:15026468
- 39. Peddle L, Rahman P. Genetic epidemiology of complex phenotypes. Methods Mol Biol. 2009;473187–201. pmid:19160739
- 40. Caliebe A, Nothnagel M. Special issue on “Genetic epidemiology of complex diseases: impact of population history and modelling assumptions”. Hum Genet. 2020;139(1):1–3. pmid:31664516
- 41. Author Last Name AFN. Statistics textbooks in the [Complete Title Needed]. 2007.
- 42. Burbine A, Fryer D, Sturtevant J. Akaike information criterion to select well-fit resist models. SPIE Proceedings. 2015;942794270J.
- 43. Gao C, Wei P, Xie Z, Zhang P, Tahir MM, Toktonazarovich TK, et al. Genomic identification of the NF-Y gene family in apple and functional analysis of MdNF-YB18 involved in flowering transition. Mol Breed. 2024;45(1):4. pmid:39726978
- 44. Tyler CW. Rational Approaches to Correcting for Multiple Tests. ei. 2018;30(14):1–8.
- 45. Zhao JH. gap: Genetic Analysis Package. J Stat Soft. 2007;23(8):.
- 46. Shen J, Li Z, Chen J, Song Z, Zhou Z, Shi Y. SHEsisPlus, a toolset for genetic studies on polyploid species. Sci Rep. 2016;624095. pmid:27048905
- 47. Sim N-L, Kumar P, Hu J, Henikoff S, Schneider G, Ng PC. SIFT web server: predicting effects of amino acid substitutions on proteins. Nucleic Acids Res. 2012;40(Web Server issue):W452-7. pmid:22689647
- 48. Nila NN, Mahmud Z, Paul A, Rahman T, Hossain Howlader MZ, Hosen MI. Investigating the structural and functional consequences of germline single nucleotide polymorphisms located in the genes of the alternative lengthening of telomere (ALT) pathway. Heliyon. 2024;10(12):e33110. pmid:39021990
- 49. Tanshee RR, Mahmud Z, Nabi AHMN, Sayem M. A comprehensive in silico investigation into the pathogenic SNPs in the RTEL1 gene and their biological consequences. PLoS One. 2024;19(9):e0309713. pmid:39240887
- 50. Ahmed HU, Paul A, Mahmud Z, Rahman T, Hosen MI. Comprehensive characterization of the single nucleotide polymorphisms located in the isocitrate dehydrogenase isoform 1 and 2 genes using in silico approach. Gene Reports. 2021;24101259.
- 51. Shi Y, Yong Y. Briefings SHEsis, a powerful software platform for analyses of linkage disequilibrium, haplotype construction, and genetic association at polymorphism loci. [Online]. Available from: www.cell-research.com
- 52. Adzhubei I, Jordan DM, Sunyaev SR. Predicting functional effect of human missense mutations using PolyPhen-2. Curr Protoc Hum Genet. 2013;Chapter 7Unit7.20. pmid:23315928
- 53. Bendl J, Stourac J, Salanda O, Pavelka A, Wieben ED, Zendulka J, et al. PredictSNP: robust and accurate consensus classifier for prediction of disease-related mutations. PLoS Comput Biol. 2014;10(1):e1003440. pmid:24453961
- 54. Johnson AD, Handsaker RE, Pulit SL, Nizzari MM, O’Donnell CJ, de Bakker PIW. SNAP: a web-based tool for identification and annotation of proxy SNPs using HapMap. Bioinformatics. 2008;24(24):2938–9. pmid:18974171
- 55. Capriotti E, Fariselli P. PhD-SNPg: a webserver and lightweight tool for scoring single nucleotide variants. Nucleic Acids Res. 2017;45(W1):W247–52. pmid:28482034
- 56. Stone EA, Sidow A. Physicochemical constraint violation by missense substitutions mediates impairment of protein function and disease severity. Genome Res. 2005;15(7):978–86. pmid:15965030
- 57. Cheng J, Randall A, Baldi P. Prediction of protein stability changes for single-site mutations using support vector machines. Proteins. 2006;62(4):1125–32. pmid:16372356
- 58. Doe J. Understanding the Unknown. Journal of Unexplained Phenomena. 2023;12(3):45–67.
- 59. Venselaar H, Te Beek TAH, Kuipers RKP, Hekkelman ML, Vriend G. Protein structure analysis of mutations causing inheritable diseases. An e-Science approach with life scientist friendly interfaces. BMC Bioinformatics. 2010;11548. pmid:21059217
- 60. Waterhouse A, Bertoni M, Bienert S, Studer G, Tauriello G, Gumienny R, et al. SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res. 2018;46(W1):W296–303. pmid:29788355
- 61. Varadi M, Anyango S, Deshpande M, Nair S, Natassia C, Yordanova G, et al. AlphaFold Protein Structure Database: massively expanding the structural coverage of protein-sequence space with high-accuracy models. Nucleic Acids Res. 2022;50(D1):D439–44. pmid:34791371
- 62. Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, et al. Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596(7873):583–9. pmid:34265844
- 63. Wiederstein M, Sippl MJ. ProSA-web: interactive web service for the recognition of errors in three-dimensional structures of proteins. Nucleic Acids Res. 2007;35(Web Server issue):W407-10. pmid:17517781
- 64. Doe J. Understanding the Challenges of Citation. Journal of Citation Studies. 2023;12(3):45–67.
- 65. Eberhardt J, Santos-Martins D, Tillack AF, Forli S. AutoDock Vina 1.2.0: New Docking Methods, Expanded Force Field, and Python Bindings. J Chem Inf Model. 2021;61(8):3891–8. pmid:34278794
- 66. Abraham MJ, Murtola T, Schulz R, Páll S, Smith JC, Hess B, et al. GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX. 2015;1–219–25.
- 67.
RG. Dumitrescu and I. Cotarla, “Understanding breast cancer risk-where do we stand in 2005? Special Article • Introduction • Descriptive epidemiology of breast cancer-Age-Geographic location • Breast cancer risk factors epidemiology-Reproductive factors-Exogenous hormones-Lifestyle risk factors-Familial history of breast cancer-Other breast cancer risk factors • Molecular epidemiology of breast cancer • Summary and conclusions,” 2005.
- 68.
Perspectives on the chemical etiology BC and carcinogens.
- 69. Penning TM, et al. Human 3α-hydroxysteroid dehydrogenase isoforms (AKR1C1-AKR1C4) of the aldo-keto reductase superfamily: functional plasticity and tissue distribution reveals roles in the inactivation and formation of male and female sex hormones [Online]. 2000. Available: www.med.upenn.edu
- 70. Andersson S, Geissler WM, Patel S, Wu L. The molecular biology of androgenic 17 beta-hydroxysteroid dehydrogenases. J Steroid Biochem Mol Biol. 1995;53(1–6):37–9. pmid:7626483
- 71. Jin J-Z, Lin S-X. Human estrogenic 17-hydroxysteroid dehydrogenase: Predominance of estrone reduction and its induction by NADPH [Online]. 1999. Available from: http://www.idealibrary.com
- 72. Marchais-Oberwinkler S, Henn C, Möller G, Klein T, Negri M, Oster A, et al. 17β-Hydroxysteroid dehydrogenases (17β-HSDs) as therapeutic targets: protein structures, functions, and recent progress in inhibitor development. J Steroid Biochem Mol Biol. 2011;125(1–2):66–82. pmid:21193039
- 73.
Liao S, Liang T, Fang S, Castaeda E, Shao T-C. steroid structure and androgenic activity specificities involved in the receptor binding and nuclear retention of various androgens. 1973.
- 74. Komoto J, Yamada T, Watanabe K, Takusagawa F. Crystal structure of human prostaglandin F synthase (AKR1C3). Biochemistry. 2004;43(8):2188–98. pmid:14979715
- 75.
Suzuki-Yamamoto T, et al. cDNA cloning, expression and characterization of human prostaglandin F synthase 1.
- 76. Tiryakioğlu NO, Tunalı NE. Association of AKR1C3 polymorphisms with bladder cancer. Urological Oncol. 2023.
- 77. Plourde M, Ferland A, Soucy P, Hamdi Y, Tranchant M, Durocher F, et al. Analysis of 17beta-hydroxysteroid dehydrogenase types 5, 7, and 12 genetic sequence variants in breast cancer cases from French Canadian Families with high risk of breast and ovarian cancer. J Steroid Biochem Mol Biol. 2009;116(3–5):134–53. pmid:19460435
- 78. Sahaba SA, Rashid MA, Islam MS, Nahid NA, Apu MNH, Sultana TN, et al. The link of ERCC2 rs13181 and ERCC4 rs2276466 polymorphisms with breast cancer in the Bangladeshi population. Mol Biol Rep. 2022;49(3):1847–56. pmid:34837148
- 79. Parvin S, Islam MS, Al-Mamun MMA, Islam MS, Ahmed MU, Kabir ER, et al. Association of BRCA1, BRCA2, RAD51, and HER2 gene polymorphisms with the breast cancer risk in the Bangladeshi population. Breast Cancer. 2017;24(2):229–37. pmid:27068824
- 80. Jahan N, Begum M, Barek MA, Aziz MA, Hossen MS, Bhowmik KK, et al. Evaluation of the association between FGFR2 gene polymorphisms and breast cancer risk in the Bangladeshi population. Genes (Basel). 2023;14(4):819. pmid:37107577
- 81. Shabnaz S, Ahmed MU, Islam MS, Islam MR, Al-Mamun MMA, Islam MS, et al. Breast cancer risk in relation to TP53 codon 72 and CDH1 gene polymorphisms in the Bangladeshi women. Tumour Biol. 2016;37(6):7229–37. pmid:26666818
- 82. Howlader NR, Rahman MM, Hossain MA, Sultana R, Hossain SM, Mazid MA, et al. Genetic polymorphisms in DNA repair genes XRCC1 and 3 are associated with increased risk of breast cancer in Bangladeshi population. Breast Cancer Res Treat. 2020;182(3):739–50. pmid:32562117