Abstract Background: Hypoxia, a condition where there is a lack of oxygen, is known to play a role in cancer progression. Objective: This study investigates the correlation between HIF1A gene-altered expression and hypoxia in Bangladeshi breast cancer (BC) cases and TCGA_BC datasets. Design: This case-control study compares BC cases to healthy controls to understand the relationship between gene changes and cancer. Method: This study used advanced analysis methods to examine the transcriptional landscape of BC, and quantitatively assessed its correlation using integrated multi-omics analysis. Results: In Bangladeshi BC cases, the T allele of HIF1A rs1154946 correlates notably (P-value < .001) with BC incidence. ELISA results confirmed a significant association (P-value < .005) between elevated HIF1A expression and BC-related hypoxia. Bioinformatics eQTL analysis validated the correlation between increased HIF1A expression and rs11549465 T allele (P-value < .01). Structural analyses suggested that rs11549465 (P582S) mutation may decrease protein stability (ΔΔG-value: −1.24 kcal/mole), potentially affecting HIF1A function. HIF1A enrichment analysis in BC underscores strong associations with oxygen levels, hypoxia, metabolic processes, apoptosis, and programed cell death (P-value < .001). Transcriptomic data demonstrated a robust correlation (P-value < .0001) between HIF1A expression and copy-number alterations, mutations, and abnormal methylation. Altered HIF1A expression showed strong negative correlations (P-value < .00001) with methylation and the expression of the ER (ESR1), in Whites. Survival analysis revealed marked differences in overall survival linked to high and low HIF1A expression (P-value < .00001). Furthermore, HIF1A expression significantly correlated (P-value < .000001) with hypoxia, TMB, MSI, and immune infiltration by CD8+ T cells, neutrophils, dendritic, and macrophages, providing deeper insights into the BC microenvironment. Conclusion: Thus, the HIF1A gene could serve as a promising biomarker for breast cancer progression, control, and survival across ethnicities, emphasizing its role in disease development and regulation. Keywords: Hypoxia, HIF1A, inflammation, disease, rs11549465, SNP, polymorphism Plain Language Summary Hypoxia is a known indicator of cancer progression. This study investigates the correlation between HIF1A gene-altered expression and hypoxia in Bangladeshi breast cancer (BC) cases and TCGA_BC datasets. This case-control study examined the transcriptional landscape of breast cancer, and quantitatively assessed its correlation using integrated multi-omics analysis. In Bangladeshi BC cases, the T allele of HIF1A rs1154946 correlates notably ( P-value <.001) with BC incidence. ELISA results confirmed a significant association (P-value < .005) between elevated HIF1A expression and BC-related hypoxia. Bioinformatics eQTL analysis validated the correlation between increased HIF1A expression and rs11549465 T allele (P-value < .01). Structural analyses suggested that rs11549465 (P582S) mutation may decrease protein stability (ΔΔG-value: –1.24 kcal/mole), potentially affecting HIF1A function. HIF1A enrichment analysis in BC underscores strong associations with oxygen levels, hypoxia, metabolic processes, apoptosis, and programed cell death ( P-value <.001). Transcriptomic data demonstrated a robust correlation ( P-value < .0001) between HIF1A expression and copy-number alterations, mutations, and abnormal methylation. Altered HIF1A expression showed strong negative correlations ( P-value < .00001) with methylation and the expression of the ER ( ESR1), in Whites. Survival analysis revealed marked differences in overall survival linked to high and low HIF1A expression ( P-value < .00001). Furthermore, HIF1A expression significantly correlated ( P-value < .000001) with hypoxia, TMB, MSI, and immune infiltration by CD8+ T cells, neutrophils, dendritic, and macrophages, providing deeper insights into the BC microenvironment. Thus, the HIF1A gene could serve as a promising biomarker for breast cancer progression, control, and survival across ethnicities, emphasizing its role in disease development and regulation. Introduction For women across the globe, Breast Cancer (BC) is the second leading cause of death.^[27]1,[28]2 The most common risk factors associated with breast cancer are inheritance, genetics, obesity and overweight, alcohol consumption, smoking, menopause, prolonged hormone replacement therapy, sedentary lifestyle, and reproductive issues.^[29]3-[30]5 In cancer biology, hypoxia is the well-known index in association with tumor development, regulation and metastasis.^ [31]6 The Hypoxia-Inducible Factor 1 (HIF1) plays a vital role in oxygen sensing under normal conditions (normoxia) and/or in tumor development and metastasis (hypoxia) by regulating the expression of over 1000 target genes that are involved in numerous cellular functions like glucose metabolism, cell survival, cell proliferation, cell apoptosis, erythropoiesis, iron metabolism and especially in angiogenesis.^[32]7-[33]11 HIF1 is a heterodimer basic helix-loop-helix protein containing the α-subunit (HIF1A) and β subunit (HIF1B). Under normal conditions (normoxia), only the α-subunit (HIF1A) expresses. This transcription factor after sensing the levels of oxygen degrades rapidly through a proteolytic degradation process and the β-subunit (HIF1B) never expresses. In this process, at the N-terminal oxygen-sensitive degradation domain of HIF1A, 2 proline residues (P402 and P564) undergo hydroxylation which then binds with the Von Hippel–Lindau (VHL) protein. This VHL conjugate is then a binding target of Ubiquitin molecules and subsequently, the ubiquitinylation process triggers proteasomal degradation of the HIF1A.^[34]12-[35]14 Whereas, under hypoxic conditions, these 2 proline residues at the oxygen-sensitive degradation domain of HIF1A do not go through hydroxylation and as a consequence do not undergo proteasomal degradation. The HIF1A is stabilized and then moves from the cytosol to the nucleus and constitutively the HIF1B, which is also known as Aryl hydrocarbon Receptor Nuclear Translocator (ARNT) expressed and binds with HIF1A subunit forming the heterodimer and then regulates transcription of various genes controlling erythropoiesis, tumor proliferation, angiogenesis and many more.^[36]15-[37]20 Thus the HIF1A is a crucial transcription factor that plays an important role in the development of different diseases or disorders in the human body by regulating the functions of some genes at low oxygen concentrations.^ [38]21 However, direct evidence of the levels of varied expression of HIF1A, their roles, and exact mechanism, correlating with various diseases remain obscured. In recent years, the availability of high-throughput multi-Omics data provided the opportunity to unfold the molecular basis of cancer by analyzing genetic alterations like copy number alterations, mutations, and epigenetic regulations in tumors. Now, genetic polymorphisms like Single Nucleotide polymorphisms (SNPs) are considered the most valuable biomarkers in association with human diseases and/or disorders and cancers.^[39]22-[40]24 The human HIF1A gene has 35 SNPs.^ [41]4 From the literature, the involvement of the 2 SNPs rs11549465 (C/T), and rs11549467 (G/A) of the HIF1A gene was mostly reported in association with different diseases and cancer.^[42]22-[43]24 These 2 SNPs (rs11549465 and re11549467) are located within the oxygen-dependent degradation (ODD) domain of exon 12 of the HIF1A gene.^[44]25,[45]26 They are responsible for amino acid substitution from proline to serine (P582S, rs11549465) and alanine to threonine (A588T, rs11549467) respectively.^[46]25,[47]26 Involvement of these 2 SNPs (P582S, A588T) was reported with increased levels of expression as well as transcriptional activity in the progression of different cancers including oral, colorectal, head and neck, renal, and hepatic cancer.^[48]3,[49]27-[50]29 Breast cancer biology is complex, and yet to fully comprehend. In recent years, the availability of TCGA (The Cancer Genome Atlas) datasets and resources like cBioportal allow us to integrate, validate and compare various levels of multi-omics information across different races and to understand complex diseases like cancer better and move a step forward toward precision medicine.^ [51]30 This study aimed to systematically evaluate the expression, function, and prognostic value of the HIF1A gene in breast invasive carcinoma (BC), interrelate the link between hypoxia and breast cancer incidences among different races and subtypes, and also to assess the prognostic relevance of HIF1A expression in overall survival. The main target was to quantitatively measure hypoxia in Bangladeshi breast cancer patients and evaluate their clinical and demographic factors. In addition, integration of literature mining and case-control datasets from The Cancer Genome Atlas (TCGA) was employed to assess HIFA transcripts and their genetic modifications in BC. To evaluate hypoxia in BC, we compared the levels of HIF1A expression in breast cancer patients of different races (Caucasian, African, and Asian) through multi-omics approaches using cBioPortal.^[52]30-[53]36 This study investigated the structure-function relationship of HIF1A and examined the impact of SNPs in association with hypoxia and BC. Enrichment analyses and integrated network construction were utilized to establish correlations among interactions and regulatory pathways associated with BC. Here, we targeted to evaluate the possibility of HIF1A, as a potential biomarker in BC screening, by interrelating breast cancer genomic data with clinical features. Thus, this study aims to comprehensively evaluate the significance of HIF1A in breast cancer, with the ultimate goal of enhancing screening, diagnosis, and treatment for Bangladeshi breast cancer patients. Methods Bangladeshi breast cancer and healthy control study subjects’ inclusion-exclusion criteria A total of 29 subjects including 12 women who were diagnosed with breast cancer (cases) and 17 healthy adult Bangladeshi women (controls), who were willing to participate, were enrolled for this study. BC patients were screened through histopathological evaluation and samples were collected from the National Institute of Cancer Research and Hospital (NICRH), Mohakhali, Dhaka from January 2023 to March 2023. The study participants’ inclusion criteria were: adult Bangladeshi women between 25 and 65 years of age, not pregnant, with no major comorbidities, and willing to participate. The exclusion criteria were: pregnant women, patients with severe debilitating conditions and unwilling to participate in the study. The research goals and procedures were properly explained to all participants before enrolling and their verbal and written consent were collected before inclusion and taking samples. The study was conducted in full compliance with the Declaration of Helsinki guidelines, ensuring the privacy and anonymity of all participants throughout the process. Intravenous blood sampling was conducted by trained professionals. This study was approved by the Ethics Committee of the National Institute of Cancer Research and Hospital (NICRH), Mohakhali, Dhaka, Bangladesh (Reference: NICRH/IRB/2023/28). Bangladeshi BC cases genomic data analysis Chemicals and reagent kits used: K[2]-EDTA vacuum blood collection tube (BD, Franklin lakes, USA), ReliaPrep™ Blood gDNA Miniprep System (Promega, Madison, USA), Agarose powder (Sigma Aldrich, Germany), Tris-HCl, EDTA, Ethidium Bromide, 6X loading dye (Promega), GoTaq^® G2 Master Mix (Promega, Madison, USA), 100 bp DNA ladder (Invitrogen™, Thermo Fisher Scientific, Austria), Wizard^® SV Gel and PCR Clean-Up System (Promega, Madison, USA), HIF1A gene primers (IDT, Singapore), HIF1A protein ELISA Kit (Invitrogen™, Thermo Fisher Scientific, Austria), Ethanol (Merck, Germany) and so on. All these chemicals and reagents were of analytical grade and available in the market. Genotyping of the breast cancer cases and healthy controls: The whole human genomic DNA (gDNA) of both cases and the control subjects were isolated from the buffy coat of collected blood samples using the ReliaPrep™ Blood gDNA Miniprep System (Promega, Madison, USA). All isolated gDNA samples were qualitatively analyzed using 0.8% (w/v) agarose gel electrophoresis and quantitatively assayed using a Nanodrop 2000 spectrophotometer (Thermo Scientific, USA) and the DNA purity index of the extracted genomic samples was measured. The isolated gDNA of both cases and controls were used as a template to amplify the HIF1A gene, using synthetic oligonucleotide primers (forward: 5ʹ-AAGGTGTGGCCATTGTAAAAACTCATG-3ʹ and reverse: 5ʹ-GTGAGGCTGTCCGACTTTGAGTATCTC-3ʹ; from Integrated DNA Technology [IDT], Singapore), GoTaq^® G2 Master Mix (Promega, Madison, USA), and a Thermocycler (Esco Swift™ MaxPro, USA) by polymerase chain reactions (PCR). The PCR amplified products were evaluated using agarose gel electrophoresis and were further purified and extracted using the Wizard^® SV Gel and PCR Clean-Up System (Promega, Madison, USA). Ensuring the quality and purity of the amplified samples (both breast cancer cases and healthy controls), DNA samples were sent for sequencing. All amplified samples were sequenced using both the HIF1A forward and reverse primers. The sequencing analysis was conducted by the 1st Base Life Technologies (Malaysia) using the ABI-prism Big Dye Terminator Cycle sequencing ready reaction kit and sequencing reactions were analyzed using an ABI 3500 automated genetic analyzer (ABI Prism, USA). Evaluating the DNA sequenced data of HIF1A gene rs11549465 and rs11549467 polymorphisms: The DNA sequenced data of both cases and controls of the HIF1A rs11549465 (C/T) and rs11549467 (G/A) SNPs were carefully aligned, matched and analyzed using BLAST (the Basic Local Alignment Search Tool; [54]https://blast.ncbi.nlm.nih.gov/Blast.cgi) and mutations were evaluated using dbSNP ([55]https://www.ncbi.nlm.nih.gov/snp/). The 1000 Genomes Project, Phase 3 dataset ([56]http://asia.ensembl.org/Homo_sapiens/Variation/Population?db=core; r=14:61740339-61741339;v=rs11549465), was used for Linkage Disequilibrium (LD) and allele frequency distribution analysis of the HIF1A rs11549465 and rs11549467 SNPs among different Asian countries including Bangladesh. The risk frequency of the HIF1A rs11549465 polymorphism in breast and other tissues was compared and analyzed through multi-tissue eQTL regression analysis using the GTex portal ([57]https://gtexportal.org/home/snp/). Further, these rs11549465 (C > T) and rs11549467 (G > A) polymorphisms were evaluated for their mutagenic properties, and structural effects through several bioinformatics tools: like I-Mutant ([58]https://folding.biofold.org/i-mutant/i-mutant2.0.html): for determining the protein stability due to the amino acid substitution at a particular position; Consurf ([59]https://consurf.tau.ac.il/consurfindex.php): for detecting the protein’s buried or exposed position after the amino acid alteration using the conservation score; SNAP2 ([60]https://rostlab.org/services/snap2web/): to evaluate the functional effect of the altered protein, and NetSurfP-3 ([61]https://dtu.biolib.com/NetSurfP-3/): was used for evaluating the surface accessibility prediction of the HIF1A polymorphisms (rs11549465 and rs11549467). Structural-functional analysis of the HIF1A wild-type and mutant proteins: To comprehend the functional associations and pathway regulation of HIF1A, a protein-protein interactions network was generated using the STRING database ([62]https://string-db.org/). The topological degree and betweeness of the hub proteins in the generated PPI network were analyzed, and the protein class abundance interactome of HIF1A was evaluated using the Konnect2prot (k2p; [63]https://konnect2prot.thsti.in/) data portal. The enrichment analysis of the HIF1A-associated network proteins was conducted through STRING, and ShinyGO 0.77 ([64]http://bioinformatics.sdstate.edu/go/), Also, the correlation networks among different hub proteins with signaling pathways and disease hallmarks were analyzed through k2p. 3D structural model of the HIF1A wild-type and P582S mutant proteins prediction and evaluation through homology modeling and Molecular dynamics (MD) simulations: The human hypoxia-inducible factor 1 subunit alpha (HIF1A) protein comprising of 826 residues, sequence data were retrieved from the UniProt ([65]https://www.uniprot.org/uniprotkb/Q16665), and RefSeq ([66]https://www.ncbi.nlm.nih.gov/gene/3091) databases. The 3D domain model was predicted using the ab initio protein structure prediction server Robetta ([67]https://robetta.bakerlab.org/). To understand the structural variations in the mutant protein compared to the wild-type structure, the predicted 3D coordinated models were validated through comparative protein domains analysis, energy minimization, and molecular dynamics simulations (MDS) using Robetta. The 3D structure modeling and analysis were performed using SWISS-MODEL ([68]https://swissmodel.expasy.org/), I-TASSER ([69]https://zhanggroup.org/I-TASSER/), Discovery Studio ([70]https://discover.3ds.com/discovery-studio-visualizer), Swiss-PDB Viewer ([71]https://spdbv.unil.ch/) and TM-align ([72]https://seq2fun.dcmb.med.umich.edu//TM-align/). The HIF1A wild-type and Mutant (P582S) proteins 3D models were super-imposed based on analyzing the TM-score (template modeling score) and RMSD (root mean square deviation) values. A TM-score ranging from 0 (unrelated protein) to 1 (perfect match) was used for model evaluation. The 3D models of WT_HIF1A and MT_HIF1A were subjected to MD simulations in water using WebGro ([73]https://simlab.uams.edu/) to assess the structural impact of the mutation through energy minimization, equilibration, molecular dynamics (MD), and trajectory analysis. The MD simulations, focusing on structural dynamics and thermodynamics, utilized the GROMOS9643A1 force field and the SPC water model. Energy minimization was performed using a steepest-descent integrator with 5000 steps. System equilibration and MD runs were conducted with NVD/NPT conditions, at 300 K temperature, 1 atm pressure, using a Leap-frog MD integrator, with a simulation duration of 50 000 ps (50 ns) and approximately 1000 frames per simulation. Root Mean Square Deviation (RMSD) and Root Mean Square Fluctuation (RMSF) of the protein residues were analyzed, along with calculations of the Radius of Gyration (Rg) for structural compactness, Solvent-Accessible Surface Area (SASA), and the average number of hydrogen bonds per frame over time. Additionally, the collective motions of the predicted protein models were examined through Normal Mode Analysis (NMA) in torsional space using the iMods server ([74]https://imods.iqf.csic.es/), which maintained stereochemistry, performed vibrational analysis, and analyzed morphing trajectories to rapidly characterize potential conformational changes. Furthermore, the WAXSiS automated server ([75]https://waxsis.uni-saarland.de/) was used to compute Small- and Wide-Angle X-ray Scattering (SAXS/WAXS, or SWAXS) curves for the 3D protein models in solution. These calculations were based on explicit-solvent all-atom MD simulations conducted with YASARA, typically lasting between 15 and 250 picoseconds. The protein model was placed in a cuboid simulation box filled with explicit water and counter ions, with protein atoms described by the AMBER03 force field and the TIP3P water model. Following energy minimization, a short MD simulation was carried out. After the simulation, a spatial envelope including the biomolecule and its solvation shell within 7 Å of the biomolecule atoms was constructed. The excluded-solvent scattering was computed based on a prior pure-water simulation, and all frames were used for the subsequent SAXS/WAXS curve calculation. Measuring the HIF1A protein levels in Bangladeshi BC patients using ELISA Blood samples from both BC patients and healthy controls were collected into K[2]-EDTA vacuum tubes (BD, Franklin Lakes, USA) and plasma was separated after overnight incubation. The HIF1A protein expression in both cases and controls plasma was evaluated using the HIF1A ELISA (Enzyme-Linked Immunosorbent Assay) kit (Invitrogen™, Thermo Fisher Scientific, Austria). All the reagents and samples were incubated at room temperature for an hour. Then 50 μl of the samples/ standard was added into each antibody pre-coated well of the ELISA plate (except the blank), covered carefully with the supplied adhesive plate covers and incubated for 2 hours. After the incubation period, the wells were washed carefully using wash buffer 3 times and then 50 μl of Biotinylated antibody reagent was added into each well and further incubated for an hour. Then following washing each well properly, 50 μl of Streptavidin HRP solution was added and incubated for 30 minutes. Again, after washing the wells properly, 100 μl of TMB substrate was added into each well, and incubated for 30 minutes, followed by adding 100 μl of stop solutions into each well. Then the absorbance were taken at 450 nm using a multi-plate reader spectrophotometer (Multiskan FC photometer, Thermo Fisher Scientific, UK) and the HIF1A protein levels were evaluated using a standard curve. Breast cancer genomic analysis using TCGA breast invasive carcinoma dataset Breast cancer dataset selection and pre-processing: In this study, The Cancer Genome Atlas (TCGA) breast invasive carcinoma dataset (TCGA, PanCancer Atlas) containing 1084 samples was selected for further investigation. The multidimensional breast cancer genomics data of the selected study was retrieved using the cBio Cancer Genomics Portal (cBioportal; [76]http://www.cbioportal.org).^[77]30-[78]38 The inclusion criteria of the selected TCGA BC dataset were: a case-control study including females, availability of patients’ ethnicity, clinical, and subtypes data. The 1084 patient/case dataset of the selected study was then screened based on altered/varied expression of the Hugo gene symbol HIF1A. The responsive dataset was further screened based on having genomic altered expressions (mutations, structural variants, and putative copy number alterations) from GISTIC (Genomic Identification of Significant Targets in Cancer) and only HIF1A mRNA expression samples with z-scores (threshold ± 2.0) of tumor samples in comparison to adjacent normal samples (RNA Seq V2 RSEM), and protein expression in the cohort, measured by reverse-phase protein array (RPPA) with a z-score threshold of ±2, was retrieved. Only the responsive dataset with a minimum of 2-fold up or down-regulation values of the HIF1A expression, in comparison to controls, was then selected.^[79]30-[80]38 Thus, finally, only 37% of the patients’ data of the selected study (TCGA, PanCancer Atlas), which met all the screening criteria, were used for further genomic profiling and correlation analyses. Assessment of changes in Transcriptomic mRNA expression of HIF1A in BC: The HIF1A gene altered transcript (mRNA) expression, for breast cancer patients and normal samples were extracted and analyzed using cBioPortal ([81]http://www.cbioportal.org) and ULCAN ([82]https://ualcan.path.uab.edu/analysis.html) data portals.^[83]32-[84]36,[85]39 The multi-omics data of mutations, structural variants, fraction genome alterations, copy number variations, co-expressions, protein expression, comparison, enrichment, and survival analysis were extracted and analyzed using cBioPortal. The different mutation types of HIF1A in TCGA breast cancer samples were detected through the DepMap ([86]https://depmap.org/portal/) portal.^ [87]40 The BC patients’ data stratified based on race categories, and breast cancer subtypes of the HIF1A gene were extracted from cBioPortal. The correlation between the 2 variables was analyzed using Spearman’s or Pearson’s test and a P-value < .05 was considered significant. All analysis was conducted and evaluated strictly following the cBioPortal guidelines. The expression variations of HIF1A among different race categories and subtypes were analyzed using UALCAN data portal. The GTex portal ([88]https://gtexportal.org/home/snp/) was used to determine the risk frequency of HIF1A polymorphisms in different tissues.^ [89]41 For the selected TCGA BC patient clinical data analysis, like patient’s age, sex, tumor stage, breast cancer subtype, and race was evaluated using the OncoPrint in cBioPortal. Co-expression between the HIF1A gene in TCGA BC samples and tumor markers like ER (Estrogen receptor), PR (Progesterone Receptor), and HER2 (Human Epidermal growth factor Receptor 2) was conducted using the GTEx, GEPIA2^ [90]42 and cBioPortal. Pearson’s and Spearman’s correlation coefficients were calculated against all other genes and a P-value < .05 indicated the statistical significance. Also, mutual exclusivity analyses were performed for ER, and HIF1A, where Fisher’s exact test was used to analyze whether alterations are significantly mutually exclusive or co-occurring between every pair of query genes, and a P-value < .05 was considered significant using cBioPortal.^[91]30-[92]36 Gene Ontology (GO) enrichment and gene set enrichment analysis (GSEA) were performed to explore the relevant pathways involved with HIF1A in BC. The ClinOmics Trail database ([93]https://clinomicstrail.bioinf.unisb.de/) was used to conduct enrichment analyses of the pathway activities by the HIF1A gene in the selected TCGA breast cancer dataset.^ [94]43 GO enrichment contains molecular function (MF), cellular component (CC), and biological process (BP). P < .05, and FDR < 0.25 are defined as significantly enriched.^ [95]43 In this study, the Kaplan-Meier (KM) plot, log-rank test, and Cox regression analysis were used for survival analyses.^ [96]44 The overall survival analysis was conducted based on the low and high HIF1A mRNA expression of breast cancer patients using KM-Plotter ([97]https://kmplot.com/analysis/).^ [98]44 The KM Plotter was used for the validation of the HIF1A as a survival biomarker. The statistical significance between the survival curves was analyzed using the log-rank test, and a P-value < .05 was considered statistically significant.^ [99]44 Analysis of HIF1A gene mutation and methylation level in TCGA BC: The DNA methylation pattern of the selected TCGA breast cancer patients was extracted and evaluated using cBioPortal ([100]http://www.cbioportal.org).^[101]30-[102]33 The mean methylation pattern of HIF1A in normal, as well as tumor samples of the selected TCGA Breast invasive carcinoma, was evaluated using different probes of TCGA-Wanderer ([103]http://maplab.imppc.org/wanderer/).^ [104]45 TCGA-Wanderer provided the level 3 TCGA methylation arrays (450k Infinium chip) and expression data (Illumina HiSeq RNAseq, summarized by exons and genes) to analyze the HIF1A expression and levels of methylation of the breast cancer samples.^ [105]44 The adjusted P-value threshold < .05 indicated was considered statistically significant. Also, MethSurv ([106]https://biit.cs.ut.ee/methsurv/) was used to perform multivariable survival analysis using methylation data of TCGA breast cancer.^ [107]46 The statistical significance of the survival curves was evaluated using the log-rank test, and a P-value < .05 was considered statistically significant. Correlation between hypoxia and HIF1A expression in TCGA BC patients: The differential expression of the HIF1A gene between cases compared to controls of the TCGA BC dataset was evaluated to measure hypoxia using 3 different score-based approaches: including Buffo hypoxia, Ragnum hypoxia, and Winter hypoxia^[108]47-[109]49 from cBioportal, ([110]http://www.cbioportal.org/). Pearson correlation coefficient was used for correlation analysis. A 2-sided P-value < .05 was considered as statistically significant. The levels of HIF1A expression in BC cases were further stratified based on race and BC subtype categories were evaluated using the Buffo hypoxic score.^ [111]47 Group differences analysis was performed using Wilcoxon test or Kruskal–Wallis test, and expressed as means ± standard deviation (SD). Also, Pearson correlation analysis between HIF1A expression and methylation of the TCGA BC dataset was evaluated using Buffo hypoxia score using cBioPortal.^[112]30-[113]33,47 Correlation of HIF1A expression and immune regulators TMB and MSI: The Tumor Mutational Burden (TMB) and Micro Satellite Instability (MSI) are 2 emerging promising, predictive biomarkers for immunotherapy in cancer treatment.^ [114]50 Correlation between altered HIF1A gene expression and mutation in BC and immune regulators TMB and MSI evaluation were conducted using the cBioPortal.^[115]30-[116]33,[117]50-[118]53 TMB is considered high if it exceeds a predetermined threshold, this cut-off may widely vary based on differing factors, including tumor type and usually 10 muts/Mb cut-off where used when analyzing different solid tumors.^[119]53,[120]54 In cases of high values, TMB is a predictive biomarker, potentially indicating a high rate of response to immunotherapy.^ [121]55 In BC patients’ correlation between HIF1A altered expression and TMB, MSI was evaluated together with hypoxia score using cBioPortal.^[122]30-[123]33,47,[124]50-[125]53 Spearman and Pearson correlation analysis was conducted to investigate the correlation between HIF1A expression in TCGA BC patients and TMB, MSI, a P-value < .05 indicated the statistical significance. Association of HIF1A expression and Immune cell Infiltration in Breast cancer: The relationship between differences in HIF1A expression in TCGA breast cancer patients and levels of immune infiltration was analyzed through the Tumor Immune Estimation Resource (TIMER, [126]https://cistrome.shinyapps.io/timer/).^[127]56,[128]57 The differential somatic copy number alterations (SCNA) of HIF1A gene in breast tumor and adjacent normal tissues was compared and the association between SCNA and immune infiltrates was evaluated as defined by GISTIC 2.0 using TIMER.^[129]34-[130]36,56,57 TIMER uses the 2-sided Wilcoxon test to statistically evaluate, and a P-value < .05 was considered significant.^[131]56,[132]57 Statistical analysis All data were analyzed using SPSS Statistics 20 (Version 27.0. Armonk, NY: IBM Corp.2020), and Microsoft Excel (2013). The number of participants was presented as a whole number, percentage, and average ± standard deviation. Chi-square test and ANOVA were performed and a P-value less than .05 was considered statistically significant. A 2-sided P-value of <.05 was considered statistically significant. Results Genotyping and evaluating Bangladeshi BC study participants through statistical and Bioinformatics analysis The DNA sequencing data for the HIF1A gene polymorphisms, rs11549465 (C > T) and rs11549467 (G > A), in Bangladeshi BC patients (cases) and healthy controls were analyzed through the BLAST sequence alignment tool ([133]https://blast.ncbi.nlm.nih.gov/Blast.cgi). Mutations were assessed using the dbSNP database ([134]https://www.ncbi.nlm.nih.gov/snp/; [135]Figure 1). The genotyping and distribution of allele frequencies for rs11549465 and rs11549467 single nucleotide polymorphisms (SNPs) in Bangladeshi BC cases and healthy controls were statistically evaluated using the chi-square test ([136]Table 1A). Additionally, a regression analysis was performed to assess genetic variations across different models. The analyzed data for rs11549465 (C > T) polymorphism revealed a significant association of the rs11549465 (C > T) SNP across all comparison models: Dominant (CT + TT vs CC: OR 3.30, 95% CI = 1.73-6.30; P-value < .001); Recessive (TT vs CC + CT: OR 2.24, 95% CI = 1.16-4.34; P-value .016); Allele (T vs C: OR 2.79, 95% CI = 1.46-5.35; P-value .002); Homozygote (TT vs CC: OR 2.59, 95% CI = 1.33-5.04; P-value .005); and Heterozygote (CT vs CC: OR 14.07, 95% CI = 1.75-112.9; P-value .013; [137]Table 1B). Hence, the analyzed data indicated a robust connection between the variant T allele and a heightened susceptibility to BC ([138]Table 1). However, no such association was observed for the HIF1A rs11549467 (G > A) polymorphism ([139]Table 1). Figure 1. [140]Figure 1. [141]Open in a new tab Genotyping the HIF1A gene SNP_rs11549465 DNA sequence and Bioinformatics analysis of the Bangladeshi BC cases. (A) PCR amplification. M: 100 bp PCR marker; S1-S7: PCR products of HIF1A gene amplification for BC samples. (B) DNA sequencing analysis; (C) BLASTn analysis of the sequenced BC sample DNA. Table 1. Genotyping and allele frequency distribution of the HIF1A gene SNPs (rs11549465 and rs11549467) analysis data of Bangladeshi BC cases. A. SNP analysis rs11549465 (C > T) polymorphism rs11549467 (G > A) polymorphism Genotypes Cases (%) Controls (%) Genotypes Cases (%) Controls (%) CC 07 (58) 14 (82) GG 12 (100) 17 (100) CT 01 (09) 00 (00) GA 00 (00) 00 (00) TT 04 (33) 03 (18) AA 00 (00) 00 (00) CC + CT 08 (67) 14 (82) GG + GA 12 (100) 17 (100) CT + TT 05 (42) 03 (18) GA + AA 00 (00) 00 (00) C allele 15 (62) 28 (82) G allele 24 (100) 34 (100) T allele 09 (38) 06 (18) A allele – – B. Genetic comparisons rs11549465 (C > T) polymorphism Odds ratio (95% confidence interval) P-value  Homozygous model (TT vs CC) 2.59 (1.33-5.04) .005  Heterozygous model (CT vs CC) 14.07 (1.75-112.90) .013  Dominant model (CT + TT vs CC) 3.30 (1.73-6.30) <.001  Recessive model (TT vs CC + CT) 2.24 (1.16-4.34) .016  Allele model (T vs C) 2.79 (1.46-5.35) .002 rs11549467 (G > A) polymorphism  Homozygous model (AA vs GG) 1.00 (0.06-16.21) 1.000  Heterozygous model (GA vs GG) 1.00 (0.06-16.21) 1.000  Dominant model (GA + AA vs GG) 1.00 (0.06-16.21) 1.000  Recessive model (AA vs GG + GA) 1.00 (0.06-16.21) 1.000  Allele model (A vs G) 1.00 (0.06-16.21) 1.000 [142]Open in a new tab Data are expressed as numbers (%), P-value was determined by the regression test. The odds ratios were calculated by taking the percentage values of the case and control number. The zero was adjusted. The P-value < .05 is considered significant. Since the participant count in our case-control study on BC in Bangladesh was relatively low, we analyzed linkage disequilibrium (LD), and examined the allele frequencies of the HIF1A gene variants rs11549465 and rs11549467 in various populations sourced from the 1000 Genomes Phase 3 dataset ([143]Table 2; [144]Figure 2). Bioinformatics analysis of the HIF1A rs11549465 C/T and rs11549467 G/A polymorphisms were conducted for allele frequency and linkage disequilibrium (LD) data. The correlation (r^2) and LD coefficient values revealed that these 2 missense mutations exhibited complete LD and had not undergone recombination ([145]Table 2). In the case of the rs11549465 polymorphism, alterations in motifs (CDP, Pou5f1) were observed, while no such effects were noted for the rs11549467 polymorphism ([146]Table 2). Additionally, we conducted a comparison between the assessed Minor Allele Frequency (MAF) of the HIF1A rs11549465 polymorphism in Bangladeshi breast cancer cases and controls with allele frequency information sourced from the 1000 Genomes Phase 3 ([147]Figure 2). The allele frequency data for the HIF1A gene rs11549465 polymorphism were analyzed across various race categories (Asian, Black, and White), Asian sub-categories (Asian, South Asians, and East Asians), and specific Asian countries (Bangladesh, India, Pakistan, Sri Lanka, China, Japan, and Vietnam), extracted from the 1000 Genomes Project. Examination of the rs11549465 SNP revealed that the minor T allele frequencies in White, Black, and Asian populations were 10%, 6.2%, and 8.2%, respectively ([148]Figure 2A). In the breakdown of Asian subcategories, South Asians exhibited a higher Minor Allele Frequency (MAF) at 11.9%, surpassing the 4.6% observed in East Asians ([149]Figure 2B). Table 2. Linkage disequilibrium (LD) analysis of the HIF1A gene rs11549465 and rs11549467 polymorphisms from 1000 Genome project phase 3. Variant Chr: Pos LD LD Allele Race category[150]^* Allele frequency Motifs changed Function annotation (hg38) (r^2) (D’) Ref Alt AFR AMR ASN EUR (dbSNP) rs11549465 146: 1740839 1 1 C T 0.04 0.11 0.05 0.10 CDP, Pou5f1 Missense rs11549467 14: 61740857 1 1 G A 0.00 0.00 0.03 0.01 – Missense [151]Open in a new tab ^* Race category: AFR: African; AMR: American; ASN: Asian. Figure 2. [152]Figure 2. [153]Open in a new tab The Minor Allele Frequency (MAF) for the HIF1A rs11549465 polymorphism from the 1000 Genomes Phase 3 dataset. (A) Race wise. (B) Asian sub-category wise. (C) Asian countrywide. *The frequency of the C allele or T allele both in BD BC patients and control groups of Bangladeshi women from this study. (D) GTEx multi-tissue eQTLs meta-analyses data showing the association between the risk frequency of rs11549465C T allele and HIF1A in breast mammary tissue from the TCGA database. When examining the 1000 Genomes data across Asian countries, the MAF values for India (12.9%), Pakistan (12.0%), Sri Lanka (9.8%), and Bangladesh (9.3%) were notably higher than those for Vietnam (7.1%), China (4.0%), and Japan (3.9%; [154]Figure 2C). For a more thorough investigation, we compared the Minor Allele Frequency (MAF) data obtained from this study for Bangladeshi breast cancer cases and controls with the information derived from the 1000 Genomes sequencing. This analysis revealed an altered allele frequency of 38.0% for Bangladeshi breast cancer cases and 18.0% for controls ([155]Figure 2C). The analysis of data from the 1000 Genomes project revealed variations in the frequency of the minor T allele of rs11549465 across different race categories. Furthermore, a significant disparity in the T allele frequency was evident between South and East Asians. When we compared our evaluated Bangladeshi BC patients with healthy controls, the association of the minor T allele frequency with BC was markedly more pronounced. Collectively, the comparison data of mutant T allele frequencies suggested notable variations in the HIF1A rs11549465 polymorphism among diverse global populations, emphasizing the need for further assessment to establish its correlation with breast cancer susceptibility ([156]Figure 2). To assess the global impact of the rs11549465 (C/T) polymorphism on HIF1A expression and its potential link to breast cancer risk, we conducted a comprehensive analysis using data from the Genotype-Tissue Expression (GTEx; [157]https://gtexportal.org/home/; [158]Figure 2D). The resulting plot illustrates the slope of the linear regression derived from normalized expression data in individual tissues’ expression quantitative trait locus (eQTL) analyses. Our findings revealed a robust correlation between the rs11549465(C/T) polymorphism and HIF1A expression across diverse tissues (Meta-Analysis RE2 P-value: 2.85e-35). Specifically, the minor T allele of the rs11549465 polymorphism exhibited significantly elevated HIF1A expression in breast mammary tissue (P-value .009; [159]Figure 2D). This observation implies a potential association of rs11549465(C > T) with the tissue-specific regulation of HIF1A expression. Bioinformatics tools I-Mutant, Consurf, SNAP2, and NetSurfP-3, were utilized to assess the structural and functional consequences of the mutations at HIF1A rs11549465 (C/T) and rs11549467 (G/A). I-Mutant ([160]https://folding.biofold.org/i-mutant/i-mutant2.0.html) analyzed data shown, the changes in free energy (DDG value) for the mutations at rs11549465 (P582S) and rs11549467 (A588T) were −1.24 kcal/mole and -0.37 Kcal/mole, respectively. This implies that both variations are likely to result in a decrease in protein stability ([161]Table 3). The data analyzed using Consurf ([162]https://consurf.tau.ac.il/consurfindex.php) yielded conservation scores of 4 and 2 for the rs11549465 and rs11549467 polymorphisms, respectively. This indicated that both mutants are likely to be structurally exposed at their respective positions ([163]Table 3). The functional scores assigned by the SNAP2 tool ([164]https://rostlab.org/services/snap2web/) for the rs11549465 and rs11549467 polymorphisms were 13 and −54, respectively. These scores suggest that the rs11549465(C/T) polymorphism is likely to induce a significant alteration, while the rs11549467 (G/A) variant is expected to have a neutral effect ([165]Table 3). Additionally, we assessed the surface accessibility of these 2 polymorphisms using NetSurfP3 ([166]https://dtu.biolib.com/NetSurfP-3/), and the findings suggested that both variants led to overall structural disorders (97% and 99%; [167]Table 4). Nevertheless, the relative surface accessibility (RSA) for the rs11549465 and rs11549467 polymorphisms was 62% and 59%, respectively, and the absolute surface accessibility (ASA) evaluation yielded values of 98% and 76%. This indicates that the rs11549465 variant was more exposed than the rs11549467 variant ([168]Table 4). In summary, the bioinformatics analyses strongly suggest that the rs11549465(C > T) variant induces a more significant structural alterations, affecting protein stability and potentially leading to functional impairment of the HIF1A gene ([169]Table 4). Table 3. Bioinformatics analysis of the HIF1A SNPs (rs11549465, and rs11549467) structural-functional properties. Amino acid Mutation I- Mutant Consurf SNAP2 DDG value kcal/mole Stability* Conservation score Buried/Exposed Functional score Effect P582S (rs11549465) − 1.24 Decrease 4 Exposed 13 Effective A588T (rs11549467) − 0.37 Decrease 2 Exposed −54 Neutral [170]Open in a new tab ^* DDG < −0.5: large decrease of stability. Table 4. Surface accessibility prediction of HIF1A variants. Amino acid alteration NetSurf 3 Class assignment Relative surface accessibility (RSA)[171]^* Absolute Surface Accessibility (ASA) A[172]^* P[disorder (%)] P582S (rs11549465) Exposed 62% (0.616284) 98 (97.98912) 97 A588T(rs11549467) Exposed 59% (0.589445) 76 (76.03836) 99 [173]Open in a new tab ^* At pH 7.0 and 25°C. Functional analysis of the HIF1A protein through Protein-protein interactions (PPI) network To understand the functional connections and regulatory pathways of HIF1A expression, we generated a protein-protein interactions (PPI) network using the STRING database ([174]https://string-db.org/). We examined the shortest paths of connections from hypoxia-related proteins to BC progression and pinpointed the central hub proteins responsible for spreading throughout the protein-protein interactome (PPI) network. Furthermore, we identified the top spreaders, which are highly connected hub proteins interacting with HIF1A (like STAT3, MDM2, SMAD3, MAPK3, CDK1, RELA, AKT1, TP53, HIF1AN, EP300, CREBBP, VHL, TCEB2, SRC, MYC, GSK3B, SP1, PRKACA, ESR1, MAPK1 etc.; [175]Figure 3A). The assessment of the topological degree and betweenness of hub proteins within the created PPI network was conducted using Konnect2prot (k2p; [176]https://konnect2prot.thsti.in/; [177]Figure 3B; [178]Supplemental File 1, Table T1). Through the analysis of shortest paths, several routes for the dissemination of metastasis and angiogenesis from hypoxic proteins were identified. Collectively, a PPI network comprising functional partners of HIF1A was established, featuring a high confidence score of 0.999. The average node degree was 7.09, the average local clustering coefficient was 0.813, and the PPI enrichment P-value was <.001 ([179]Figure 3A). To gain a deeper insight into the regulation and biological functions of primary interacting partners, as well as to pinpoint key mutation genes, we conducted functional pathways enrichment analyses ([180]Figure 3C). The P-values for the enrichment analysis were derived from the hypergeometric distribution. After applying the Benjamini-Hochberg correction for multiple testing, the most significantly enriched signaling pathways, based on log10(FDR) fold enrichment, encompassed HIF-1, resistance to EGFR tyrosine kinase inhibitors, insulin resistance, focal adhesion, adherens junction, thyroid hormone, estrogen, AGE-RAGE, JAK-STAT, Wnt, and PI3K-Akt ([181]Figure 3C; [182]Supplemental File 1, Table T2). Notably, the top-fold enrichment in disease pathways encompassed various cancer types such as prostate, breast, endometrial, colorectal, gastric, and melanogenesis. Additionally, it included other diseases like diabetic cardiomyopathy, alcoholic liver disease, Cushing syndrome, measles, atherosclerosis, hepatitis, and pathways associated with neurodegeneration ([183]Figure 3C). Moreover, an examination of correlation networks was conducted to analyze the connections among various hub proteins (notable spreaders), their association with signaling pathways, and disease hallmarks ([184]Figure 3D and [185]E; [186]Supplemental File 1, Table T2). Additionally, the abundance interactome of HIF1A’s protein class was assessed using the k2p data portal ([187]Figure 3F). The resulting interactome network unveiled high-confidence interactions and regulations among hypoxic and metastatic proteins. Figure 3. [188]Figure 3. [189]Open in a new tab Functional analysis of the HIF1A protein using bioinformatics. (A) Protein-protein interaction (PPI) network of HIF1A. (B and C) Topological degree versus betweenness of the hubs of the predicted Protein-protein interaction (PPI) network; and Pathway Enrichment analysis exhibiting the Benjamini-Hochberg False Discovery Rate (FDR)-adjusted threshold of 0.01. (D-F) Correlation networks prediction between important hub proteins with the disease hallmarks, signaling pathways, and protein class abundance of the interactome. Structural analysis of the HIF1A wild-type and P582S mutant proteins HIF1A ([190]Q16665) exhibits a high degree of conservation across various species, and its conserved domains are: the HLH (Helix-loop-helix) domain, PAS domain, HIF-1 domain, and CTAD (C-terminal transactivation domain; [191]Figure 4A). The structured N-terminal domain, retaining the active site, facilitates DNA binding and dimerization. The PAS domain plays a vital role in ligand binding and functions as a sensor for light and oxygen in the signal transduction pathway. The HIF-1 domain, positioned at 549 to 582, regulates cellular homeostatic responses to changes in oxygen availability. In the presence of oxygen, HIF1A undergoes ubiquitination, leading to targeted proteasomal degradation. The CTAD domain, spanning 40 residues, exhibits a high degree of conservation ([192]Supplemental File 1, Figure S1; [193]Table 5). Variations in single nucleotide polymorphisms (SNPs) often have a notable impact on the structure and functionality of proteins. In the HIF1A protein, the missense mutation rs11549465, located at position 582, resulted in the substitution of a nonpolar, aliphatic proline residue (Pro, P) with a polar residue containing a reactive hydroxyl group, namely serine (Ser, S; [194]Figure 4A). Although the Protein Data Bank (PDB) contains several X-ray crystallography and NMR structures primarily covering the N-terminal half of the HIF1A protein, there is a lack of structures for the C-terminal domain. Therefore, this study employed ab initio, de novo 3D modeling through Robetta. Figure 4. [195]Figure 4. [196]Open in a new tab Three dimensional homology model prediction of the HIF1A wild-type (WT) and P582S mutant (MT) proteins. (A) The domain architecture of HIF1A (RefSeq 3091; UniProt [197]Q16665); (B and C) Ramachandran plot of the WT_HIF1A and MT_HIF1A proteins; (D and E) 3D homology models of the WT_HIF1A and MT_HIF1A; (F and G) 3D model of WT_HIF1A and the mutant showing all Pro (P) residues including the P582 position at the MT_HIF1A model, showing the amino-terminal (N-tr.) residues in yellow and the carboxyl-terminal (C-tr) in blue. (H) Superposition of the 3D structural models of the WT_HIF1A with MT_HIF1A prediction based on pair-wise structure comparison. Table 5. Conserved domains prediction of HIF1A protein ([198]NP_001521) for 3D modeling. Name Accession Description Interval E-value bHLH-PAS_HIF1a_PASD8 Cd19727 basic helix-loop-helix-Per-ARNT-Sim (bHLH-PAS) domain found in hypoxia-inducible factor . . . 13-83 4.01e-42 PAS_3 pfam08447 PAS fold; The PAS fold corresponds to the structural domain that has previously been defined . . . 254-339 5.44e-21 HIF-1a_CTAD pfam08778 HIF-1 alpha C terminal transactivation domain; Hypoxia inducible factor-1 alpha (HIF-1 alpha) . . . 789-825 9.16e-17 HIF-1 pfam11413 Hypoxia-inducible factor-1; HIF-1 is a transcriptional complex and controls cellular systemic . . . 551-581 2.96e-11 PAS smart00091 PAS domain; PAS motifs appear in archaea, eubacteria and eukarya. Probably the most surprising . . . 93-148 1.47e-10 [199]Open in a new tab 3D homology modeling and Molecular dynamic (MD) simulation of HIF1A wild-type (WT) and P582S mutant (MT) proteins The 3D structures of both wild-type HIF1A (WT_HIF1A) and P582S mutant (MT_HIF1A) proteins were predicted and analyzed using Robetta ([200]https://robetta.bakerlab.org/) and I-TESSER ([201]https://zhanggroup.org/I-TASSER/) servers ([202]Figure 4). Model visualization and representation were carried out using the Discovery Studio ([203]https://discover.3ds.com/discovery-studio-visualizer) and Swiss-PdbViewer (SPDV; [204]https://spdbv.unil.ch/; [205]Figure 4). Robetta was employed for the prediction of 3D coordinate models, which were validated via comparative analysis of protein domains, energy minimization, and simulations using molecular dynamics (MDS; [206]Table 5; [207]Figure 4). The Ramachandran analysis was carried out for the WT and MT_HIF1A proteins ([208]Figure 4B and [209]C). The analysis of Ramachandran plots revealed that the residues in the wild-type protein predominantly occupied the favored and allowed regions, indicating a higher level of structural integrity in WT_HIF1A exhibited the 3D protein models of the wild-type (WT_HIF1A; [210]Figure 4D and [211]F) and the mutant P582S (MT_HIF1A; [212]Figure 4E and [213]G). The substitution at position P582S, replaced the nonpolar, cyclic bend-forming amino acid at this site with a polar, nucleophilic functional group ([214]Figure 4G). [215]Figure 4F and [216]G exhibited the WT and MT_HIF1A 3D models showing all Pro (P) residues, and emphasizing the regions affected by the P582 substitution. The quality and accuracy of the predicted wild-type (WT_HIF1A) and P582S mutant (MT_HIF1A) protein models were assessed by I-TESSER ([217]Table 6) mainly focusing on identifying differences in hydrophobicity and size between the native (wild-type) and modified residue. These variations could potentially disrupt interactions with neighboring residues, ultimately influencing the overall structure and function of the proteins ([218]Table 6; [219]Supplemental File 1, Table T3-T5). The MT_HIF1A model presented structural drifting, confirmed by energy refinement evaluation ([220]Figure 4E and [221]G). We calculated root-mean-square deviation (RMSD) values for the MT model ([222]Table 6). The value means the average distance of α-carbon backbones between mutant and wild-type models. The generated WT_HIF1A and MT_HIF1A models with an estimated RMSD values of (9.1 ± 4.6 Å) and (12.4 ± 4.3 Å) and a TM score of (0.68 ± 0.12), and (0.52 ± 0.15) respectively, indicated models were of correct topology ([223]Figure 4B-G; [224]Table 6). For further evaluation, the modeled mutant protein (MT_HIF1A) was superimposed on the wild-type (WT_HIF1A) protein using TM-align ([225]https://seq2fun.dcmb.med.umich.edu//TM-align/; [226]Figure 4H; [227]Supplemental File 1: Table T6). The estimated RMSD values of 8.87 Å and TM score of 0.21 indicate random structural similarity ([228]Table 7; [229]Figure 4H). The data alignment revealed notable structural shifts within the ODD domain housing the P582S mutation. Consequently, the variance in structure between the mutant and wild-type proteins exhibited a positive correlation with the RMSD value. The presence of the P582S mutation may led to a reduction in hydrophobicity, potentially impacting the stability and intra-molecular interactions of the HIF1A protein. This, in turn, could affect the functionality of the HIF1A protein and contribute to the onset of diseases. Table 6. The quality and accuracy assessment of the wild-type (WT) and P582S mutant (MT) structure models of HIF1A by I-TESSER. Model C-score Estimated RMSD value Estimated TM-score[230]^* WT_HIF1A −0.26 9.1 ± 4.6 Å 0.68 ± 0.12 MT_HIF1A −1.60 12.4 ± 4.3 Å 0.52 ± 0.15 [231]Open in a new tab ^* A TM-score > 0.5 indicates a model of correct topology. Table 7. Superposition of the predicted 3D structure models of the HIF1A wild-type (WT) and P582S mutant (MT) prediction based on pair-wise structural homology analysis using the TM-align algorithm. Model No. of residues Superposition of WT & MT_HIF1A models Estimated RMSD value Estimated TM-score[232]^* WT_HIF1A 826 8.87 0.21 MT_HIF1A 826 [233]Open in a new tab ^* 0.0  T) polymorphism and the risk of developing BC. This connection was consistently observed across different comparison models, highlighting a strong association between the T variant of this polymorphism and increased susceptibility to BC. From the 1000 Genomes phase 3, the minor allele frequency (MAF) data exhibited the T allele of the HIF1A rs11549465 polymorphism was more prevalent in Asians, particularly South Asians, compared to other race categories. The GTEx Data indicated that the T allele of this polymorphism was significantly correlated with higher expression of HIF1A in breast mammary tissue (P-value .009), suggesting that the T allele may influence the regulation of HIF1A expression in a tissue-specific manner. In this study, the frequency of the minor T allele was 38% among Bangladeshi BC patients, which was more than twice the frequency found in the control group (18%). This substantial difference underscores the importance of further investigation to firmly establish the T allele’s correlation with increasing breast cancer risk. The bioinformatics analysis of the rs11549465 SNP protein mutation (P582S) reveals significant impacts on the structural stability of the HIF1A protein. The mutation results in a DDG value of −1.24 kcal/mol, indicating a decrease in protein stability. Structural exposure and surface accessibility analysis revealed that the P582S mutation causes to be structurally exposed, with an absolute surface accessibility of 98%, indicating that this region of the mutant protein is more exposed to the solvent and leads to a 97% structural disorder, suggesting significant deviations from the wild-type structure, which further may affect protein stability and function. Functional investigation through the protein-protein interaction (PPI) network and pathway enrichment analysis revealed significant impacts (−log10 [FDR] fold) on various signaling pathways, including HIF-1, resistance to EGFR tyrosine kinase inhibitors, insulin resistance, focal adhesion, adherens junction, thyroid hormone, estrogen, AGE-RAGE, JAK-STAT, Wnt, and PI3K-Akt pathways. These pathways are crucial for cell signaling, growth, and metabolism. The top-fold enrichment in disease hallmarks indicated that the P582S mutation is associated with a range of diseases, particularly various cancer types, diabetic cardiomyopathy, alcoholic liver disease, Cushing syndrome, measles, atherosclerosis, hepatitis, and neurodegenerative diseases. The P582S mutation replaces a nonpolar, aliphatic side chain with a polar, reactive hydroxyl group, significantly altering the protein’s structure. 3D homology models (of WT_HIF1A and MT_HIF1A proteins) and molecular dynamics (MD) simulations suggested altered topology and structural shifts, particularly in the oxygen-dependent degradation (ODD) domain where the mutation occurs. This mutation reduces hydrophobicity, potentially impacting stability and intramolecular interactions, thereby increasing the risk of functional impairments and contributing to the development of various diseases, particularly cancer. This comprehensive analysis underscores the critical role of the P582S mutation in altering the structure and function of the HIF1A protein, with significant implications for disease mechanisms and potential therapeutic targets. This thorough analysis highlights the crucial role of the P582S mutation in altering the structure and function of the HIF1A protein, which has significant implications for disease mechanisms and potential therapeutic targets. This detailed MD simulations analysis of the WT and MT forms of HIF1A in water, analyzing their structural stability, flexibility, and compactness, and investigating their dynamic behavior through normal mode analysis and SAXS/WAXS scattering curves helps to understand how the mutation affects the protein’s structure and dynamics, potentially providing insights into its functional implications. Multi-omics data integration approaches combine various types of biological data (such as genomics, transcriptomic, and proteomics) to provide a more thorough and detailed understanding of disease outcomes and biomarker identification.^[344]59-[345]62 This integration offers enhanced insights and predictions compared to analyzing each data type individually. In this study, we integrated multi-omics data to examine the altered expression of HIF1A in TCGA BC case-control samples. A significant alteration (P-value 4.32e-05) was observed in BC cases compared to controls. Correlation analysis revealed a significant association (P-value 7.2e-25) between the fraction genome altered and mutational counts with HIF1A expression in BC samples. Predominantly, amplification or up-regulation (P-value < .05) mutation was found in samples across various races. Race category-stratified data indicated most significant changes in HIF1A expression were observed in Whites (P-value 3.0e-06). The frequency of genomic alterations and associated regulatory pathways are good indicators of BC progression and severity. Pathway enrichment analysis, based on differential expression scores, identified the most significant signaling pathways (P-value < .005) related to HIF1A expression: MAPK, Wnt, Notch, PI3-Akt, TGF-beta, and HIF-1. These pathways regulate various aspects of the BC microenvironment, including motility, invasion, modulation, and plasticity. The KM plots of survival analysis showed significant differences (P-value 1.9e-06) in survival times between patients with high and low HIF1A expression levels in TCGA BC cases. To understand the regulation of HIF1A gene expression, we assessed the DNA methylation pattern, and an aberrant DNA methylation pattern (P-value 4.7e-10) was associated with altered HIF1A expression in BC samples. Stratifying the methylation data by ethnicity and BC stage revealed a strong association between altered methylation patterns and Caucasians (P-value 1.5e-07) as well as stages 2 and 3 of BC (P-values < 5e-05). Multivariable survival analysis indicated that higher levels of methylation significantly impacted survival outcomes (P-value < .005). In summary, this study utilized multi-omics data to explore the role of HIF1A in breast cancer, uncovering significant genetic and epigenetic alterations, pathway involvements, and their associations with disease progression and patient survival, with specific insights into racial and stage-related differences. The study investigated the levels of HIF1A protein in Bangladeshi BC patients and healthy controls using ELISA, and more than 2-fold elevated expression (P-value .003) levels were found in BC patients compared to healthy controls. Manhattan plot enrichment analysis identified the most significant (FDR q-value < 0.001) biological processes related to HIF1A levels, including responses to various stresses (endogenous and abiotic stimuli), reactions to oxygen levels and hypoxia, positive regulation of metabolic processes, and regulation of apoptotic processes and programed cell death. Also, enrichment analysis of the most significant pathways associated with HIF1A protein levels was identified with a Benjamini-Hochberg FDR < 0.001 including HIF-1 signaling, pathways associated with different cancers (breast, renal, prostate, pancreatic, and thyroid), estrogen signaling pathways, and endocrine resistance pathways. This study highlighted the most curated interactions between HIF1A and small molecules, notably oxygen and glucose ranked among the top interactions, suggesting a strong correlation between elevated HIF1A levels in plasma and sensing of hypoxia. In summary, the study demonstrates a significant elevation of HIF1A protein levels in breast cancer patients, correlates with hypoxia-related processes, and suggests the potential of HIF1A as a biomarker for breast cancer diagnosis. The elevated levels of circulating HIF1A could potentially serve as an indicator for breast cancer and the ELISA procedure might become a promising alternative for measuring hypoxia in the future. To evaluate hypoxia, we used integrated multi-omics approaches to investigate the relationship between altered levels of HIF1A expression and hypoxia in TCGA BC cases. We employed Buffo, Ragnum, and Winter hypoxia scoring methods, and the analyzed data demonstrated a robust correlation (P-value < 5e-18). Ethnicity-stratified hypoxia scores indicated that Black and Asian race categories exhibited the most significant alterations and highest levels of hypoxia compared to Whites. In cancer, defective DNA repair leads to an elevated TMB, which is an indicator of the overall number of nonsynonymous mutations per megabase of coding regions within a tumor genome.^[346]50,[347]52 Regression analysis revealed a strong correlation between mutation count and TMB in TCGA BC cases. A linear correlation was also observed between the count of HIF1A mutations and TMB scores in TCGA BC patients (P-value .00), with the highest mutational burden observed in Caucasians. Subtype-stratified data revealed a higher incidence of estrogen receptor-positive (ER+) cases in TCGA BC samples. HIF1A expression covariate-adjusted methylation data showed a significant increase in the ER+ group, with the highest methylation levels in Whites compared to other race categories. Comparing HIF1A altered expression levels between ER+ and ER- groups revealed significant differences (P-value < 1.0e-10), and the ER- group (i.e. HER2 and triple-negative breast cancer [TNBC] subtypes) exhibited higher levels of hypoxia and TMB scores. A negative correlation (P-value 7.9e-29) was observed between HIF1A and ESR1 gene expressions. This study demonstrated a higher incidence of HIF1A genomic alterations and elevated hypoxia and TMB scores in ER- BC cases compared to ER+ cases. Additionally, correlation analysis between levels of tumor-infiltrating immune cells and differential HIF1A expression in TCGA BC cases revealed a strong positive correlation with CD8+ T cells, neutrophils, dendritic cells, and macrophages (P-value ⩽ 9.9e-20) in the microenvironment of BC. Thus, our results indicated that in BC, differential HIF1A expression was linked not only to immune cell infiltration but also to immune checkpoint genes, TMB, and microsatellite instability (MSI) of immune regulators within the breast tumor microenvironment which may be crucial in influencing tumor growth, invasion, and metastasis. However, the primary limitation of this study was the small dataset of BD breast cancer cases, necessitating further research to reach definitive conclusions. Future studies should include long-term follow-ups and work with a larger sample cohort to enhance the robustness of the findings. Another limitation was the racial composition of the case-control samples in the TCGA BC datasets, which predominantly consisted of individuals of the White race category. There are limited datasets available for African and Asian populations, particularly South Asians. The availability of more BC genomics data across all racial categories in the future would help confirm these findings. Conclusion The HIF1A transcription factor acts as a key regulator, sensing oxygen levels to meet cellular metabolic needs, ensuring protective responses to hypoxia, and maintaining oxygen homeostasis. In tumors, hypoxia arises from altered metabolism or inadequate vascularization.[348]63-[349]65 In this study, we used a novel multi-omics approach utilizing case-control data to analyze the transcriptional landscape of BC, specifically investigating the correlation between HIF1A expression levels and hypoxia. In this study, we assessed hypoxia status in both Bangladeshi BC and TCGA BC cases. DNA sequencing of the HIF1A gene rs11549465 polymorphism in Bangladeshi cases and controls revealed a strong association between the altered T allele and breast cancer susceptibility. 1000 Genomes analyzed data indicated a higher frequency of the altered T allele among Whites and Asians, with South Asians showing a higher frequency than East Asians. Missense mutations and up-regulation of HIF1A transcript were more prevalent in BC cases than controls. Through eQTL analysis, a significant correlation was found between the minor T allele and elevated HIF1A in BC, indicating an increased risk associated with rs11549465 C/T polymorphism. The structural analysis showed that the rs11549465 (C>T) mutation significantly alters the structure of the HIF1A protein, potentially impacting its stability and function. Additionally, when the oxygen-sensing domain of the wild-type HIF1A and the rs11549465 mutant were modeled in 3D, notable structural shifts were observed in the ODD domain, suggesting functional impairment. In this study, we focused on identifying key proteins and regulatory pathways involved in the crosstalk between hypoxia and BC progression. Enrichment analysis revealed responses to various stimuli, oxygen levels, hypoxia, positive regulation of metabolic processes, as well as the regulation of apoptotic processes and programed cell death with utmost significance. Bangladeshi BC cases showed increased HIF1A expression, strongly linked to hypoxia. The TCGA BC dataset highlighted notable differences in HIF1A gene expressions between cases and controls. Missense mutations and elevated HIF1A expression were significantly linked to BC. Correlation analysis across different race categories revealed significant variations in the fraction of altered genome and mutation count in BC samples. The overall survival analysis of the TCGA BC patients revealed noteworthy differences in survival time between high and low levels of HIF1A expression. Notably, a significant negative correlation between HIF1A expression levels and altered methylation was most prominently associated with Caucasians. The ethnicity-stratified analysis revealed greater genome alterations and hypoxia in Asians and Blacks compared to Whites. Moreover, a correlation analysis of HIF1A and ESR1 Expression in BC data revealed significant negative correlations. Covariate-adjusted methylation profile analysis of HIF1A expression showed increased DNA methylation in the ER+ group compared to the ER- group. In the race-categorized data, BC cases of White ethnicity exhibited the highest methylation levels among different racial categories. The multivariable survival analysis revealed significant alteration in survival time associated with high methylation. Significant alterations in Buffo hypoxia and TMB scores were observed in HIF1A fraction genome alteration levels between ER+ and ER- subtypes of BC. The ER- category showed heightened levels of hypoxia. The correlation analysis of HIF1A expression with immune regulators such as tumor mutational burden (TMB) and microsatellite instability score (MSI) showed significant associations in BC. HIF1A expression exhibited a robust positive correlation with immune cell infiltration, including CD8+ T cells, neutrophils, dendritic cells, and macrophages, indicating the BC microenvironment. Early detection and diagnosis can render most cancers treatable before spreading to secondary tissues. This study established a significant correlation between altered expression of HIF1A and hypoxia in BC patients across diverse racial groups. The observed ethnicity-based link underscores the need for further genomic research in Asian and African populations. Consequently, HIF1A may serve as a valuable predictive biomarker for BC assessment across all races. Supplemental Material sj-pdf-1-bmi-10.1177_11772719241278176 – Supplemental material for Exploring the Correlation Between Hypoxia, HIF1A Variants, and Breast Cancer in Different Ethnicities, and Bangladeshi Women: Through ELISA and Integrative Multi-Omics Analysis [350]sj-pdf-1-bmi-10.1177_11772719241278176.pdf^ (1.3MB, pdf) Supplemental material, sj-pdf-1-bmi-10.1177_11772719241278176 for Exploring the Correlation Between Hypoxia, HIF1A Variants, and Breast Cancer in Different Ethnicities, and Bangladeshi Women: Through ELISA and Integrative Multi-Omics Analysis by Md. Shihabul Islam and Jesmin in Biomarker Insights sj-pdf-2-bmi-10.1177_11772719241278176 – Supplemental material for Exploring the Correlation Between Hypoxia, HIF1A Variants, and Breast Cancer in Different Ethnicities, and Bangladeshi Women: Through ELISA and Integrative Multi-Omics Analysis [351]sj-pdf-2-bmi-10.1177_11772719241278176.pdf^ (4.1MB, pdf) Supplemental material, sj-pdf-2-bmi-10.1177_11772719241278176 for Exploring the Correlation Between Hypoxia, HIF1A Variants, and Breast Cancer in Different Ethnicities, and Bangladeshi Women: Through ELISA and Integrative Multi-Omics Analysis by Md. Shihabul Islam and Jesmin in Biomarker Insights Acknowledgments