Comprehensive analysis of the BID gene to uncover the role of novel alternative splicing isoforms in colon adenocarcinoma: a systematic review of literature and bioinformatics analysis

Abstract

Background

Colon Adenocarcinoma (COAD) is the second leading cause of cancer-related death worldwide, with a rising incidence. Apoptosis is a key contributor to cancer, serving as a prognostic and diagnostic marker. BID, as a link between extrinsic and intrinsic apoptosis pathways, may play an important role in COAD development.

Materials and methods

A systematic review was conducted to assess the role of the BID gene in COAD. A systematic review was conducted using PubMed, Web of Science, and Scopus up to December 14, 2023. Two reviewers screened articles according to the PRISMA 2020 guidelines, and the risk of bias was assessed using the ROBINS-I tool. The study protocol was registered in PROSPERO (CRD42023455350). A pan-cancer analysis of BID was performed using TCGA and HPA data. In COAD, BID expression, methylation, isoforms, and correlations with immune cell infiltration were analyzed in conjunction with clinical data. miRNAs regulating BID were investigated, and gene set enrichment analysis identified associated GO terms and KEGG pathways. Phylogenetic analysis of BID isoforms and variants was conducted using sequences from GenBank and Ensembl, aligned with Clustal Omega and NCBI tools to assess conservation and similarity.

Results

From 661 articles identified across PubMed, Web of Science, and Scopus, 205 studies were included in the review, of which 17 human studies analyzed BID in COAD. BID mediates apoptosis in COAD through the Fas/FasL and Apo2/TRAIL pathways, activating DR4 and DR5, which leads to the recruitment of FADD and the activation of CASP-8 and CASP-10. CASP-8 cleaves BID into tBID, which activates BAX and BAK, causing mitochondrial outer membrane permeabilization, cytochrome c release, apoptosome formation, CASP-9 activation, and apoptosis. Apoptosis is modulated by ER stress/UPR, DPP3/CDK1, PIDD/CASP-2, NF-κB/Nur77, APPL/DCC, PI3K/AKT, HIF-1, and miR-20, MCL-1, BCL-2/BCL-xL. In COAD, BID is overexpressed, particularly in isoforms BID-L, BID-EL, and BID-S, whereas BID-Si6 remains unchanged. BID expression is lower in patients with advanced-stage and N2 disease, suggesting potential prognostic value. BID expression inversely correlates with the infiltration of B cells, CD4+ T cells, and regulatory T cells. The hsa-miR-194-3p and hsa-miR-149-5p were found to be correlated with BID expression in COAD. Phylogenetic analysis reveals that BID is evolutionarily conserved, underscoring its functional importance.

Conclusions

BID, specifically the BID-EL isoform, serves as a prognostic and diagnostic biomarker in COAD, highlighting the need for further research on its potential as a therapeutic target in this disease.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12935-025-04031-2.

Introduction

Colorectal cancer (CRC), including colon adenocarcinoma (COAD), ranks as the third most prevalent cancer globally, representing about 10% of all cancer diagnoses, and stands as the second leading cause of cancer-related mortality worldwide [1]. In 2020, there were an estimated 1.9 million new cases and 930,000 deaths, with the highest incidence and mortality in high-HDI regions (Australia/New Zealand, Europe) and the lowest in parts of Africa and Southern Asia. By 2040, the burden is projected to rise to 3.2 million cases and 1.6 million deaths, largely concentrated in high and very high Human Development Index (HDI) countries [2]. Major environmental factors contributing to colorectal cancer include diets high in red and processed meats, low intake of fruits and fiber, obesity, physical inactivity, smoking, and excessive alcohol consumption. Also, important genetic factors comprise inherited syndromes such as Lynch syndrome (hereditary non-polyposis colorectal cancer) and familial adenomatous polyposis, along with mutations in DNA mismatch repair genes and other recognized loci [3–5].

The inhibition or evasion of apoptosis plays a crucial role in colorectal tumorigenesis, as it enables abnormal cells to escape programmed cell death, persist, and ultimately form tumors. Chemotherapeutic drugs typically rely on activating apoptotic pathways, so defects in these pathways, such as p53 mutations or altered expression of BCL-2 family proteins, can lead to drug resistance. These dysfunctions underscore the importance of apoptosis in both COAD development and treatment resistance [1, 6]. The BCL-2 family of proteins represents a central regulator of apoptosis, encompassing both pro- and anti-apoptotic members that determine cell fate. Dysregulation of this family, particularly the overexpression of anti-apoptotic proteins such as BCL-2 or BCL-XL, has been implicated in tumorigenesis and resistance to therapy in COAD [7–10]. Consequently, alterations in the balance of BCL-2 family members highlight the pivotal role of apoptosis control in colorectal cancer biology. Given the importance of this, further exploration of additional apoptotic regulators and signaling pathways in COAD is essential to understand the mechanisms of tumor progression and therapeutic resistance.

The BH3-interacting domain death agonist (BID) is a pro-apoptotic Bcl-2 family protein essential for apoptosis, activated by proteolytic cleavage to expose its BH3 domain and trigger mitochondrial cell death pathways, with multiple naturally occurring isoforms—Bid(EL), Bid(S), and Bid(ES), generated by alternative splicing of the Bid gene. Bid(EL) corresponds to full-length Bid with additional N-terminal sequence and promotes apoptosis, Bid(S) contains the N-terminal domains but lacks the BH3 domain and inhibits apoptotic responses by blocking the action of pro-apoptotic Bid, and Bid (ES) has only the C-terminal Bid sequence and can induce apoptosis but also partially inhibit truncated Bid's effects, with each isoform exhibiting unique expression, localization, and impact on cell fate, particularly during granulocyte maturation and in response to apoptotic stimuli [11, 12]. BID functions as a connecting link between death receptor signaling and the mitochondrial pathway of apoptosis [13]. The BID gene encodes a death agonist capable of forming heterodimers with either the agonist BAX or the BCL2 [14]. Activation of the BID is a significant step in the apoptosis pathway, ultimately leading to the release of apoptotic factors and programmed cell death [15, 16]. BID is involved in the occurrence and progression of various tumors; however, under certain conditions, it can also act as a tumor suppressor by preventing the transition of premalignant cells to malignancy. Evidence suggests that BID expression increases during tumor progression in several cancers, such as COAD. Under hypoxic conditions, HIF-1 regulates the downregulation of BID. This downregulation contributes to decreased drug-induced apoptosis and clonogenic resistance, further complicating treatment efficacy in tumors [17–20].

Given the pivotal role of the BID gene in COAD and the current lack of information regarding its variants, we conducted a systematic review of the literature to elucidate the multifaceted significance of BID in various aspects of COAD, encompassing its role in disease development, prognostic implications, and potential as a therapeutic target. Furthermore, we comprehensively analyzed BID's predictive and prognostic value in COAD utilizing data from The Cancer Genome Atlas (TCGA) cohort, which incorporated gene mutation, expression, and methylation profiles. Additionally, we quantified the expression levels of five BID isoforms (BID-L, BID-EL, BID-Si6, BID-S, and BID-ES-1) in COAD specimens to provide a more refined understanding of BID's molecular landscape in this malignancy (Fig. 1).

Fig. 1.

Study flowchart and BID analysis in COAD. This systematic review screened 205 articles and identified 17 human studies on BID. Mechanistically, caspase-8 cleaves BID to tBID, which activates BAX/BAK, induces mitochondrial MOMP, and triggers apoptosis. In silico analyses showed BID overexpression in COAD, differential expression of four isoforms, correlations with hsa-miR-194-3p and hsa-miR-149-5p, and inverse associations with immune cell infiltration. Clinical data confirmed BID-EL upregulation in COAD

Materials and methods

Systematic review

Literature search methodology

This systematic review examines the role of the BID gene in COAD. A search for relevant articles was conducted until December 14, 2023, using PubMed, Web of Science, and Scopus databases. The search algorithm included key terms related to the BID gene from the BCL-2 family and its effect on COAD, including “BH3-interacting domain”, “Colorectal Carcinoma”, “Truncated BID Protein”, and “genes affecting BID”. While reviewing sources related to this research, meta-analyses, correspondence, reviews, case reports, and personal opinion studies were excluded. More information about the articles of these three databases and the search strategy is provided in Table 1.

Table 1.

Search queries and databases

Database	Search query	Number of extracted studies
PubMed	(“BID gene” OR “BID Protein” OR “tBID” OR “Truncated BID” OR “BH3-interacting domain” OR “BH3 Interacting Domain Death Agonist Protein”) AND (“Colorectal cancer” OR “Colorectal tumor*” OR “Colorectal carcinoma” OR “Colorectal adenocarcinoma” OR “Colorectal neoplasm*” OR “Rectal neoplasm*” OR “Sigmoid neoplasm*”)	322
Web of Science	#1: (“BID gene” OR “BID Protein” OR “tBID Protein” OR “Truncated BID Protein” OR “BH3-interacting domain” OR “BH3 Interacting Domain Death Agonist Protein”) #2: (“Colorectal cancer” OR “Colorectal tumor*” OR “Colorectal carcinoma” OR “Colorectal adenocarcinoma” OR “Colorectal neoplasm*” OR “Rectal neoplasm*” OR “Sigmoid neoplasm*”) #1 AND #2	130
Scopus	TITLE-ABS-KEY (“BID gene” OR “BID Protein” OR “tBID Protein” OR “Truncated BID Protein” OR “BH3-interacting domain” OR “BH3 Interacting Domain Death Agonist Protein”) AND TITLE-ABS-KEY (“Colorectal cancer” OR “Colorectal tumor*” OR “Colorectal carcinoma” OR “Colorectal adenocarcinoma” OR “Colorectal neoplasm*” OR “Rectal neoplasm*” OR “Sigmoid neoplasm*”)	209

Eligibility criteria for inclusion and exclusion

During the first screening stage, the initially identified articles, duplicate articles, and those published prior to 2000 were excluded. In addition, all articles in non-English languages were excluded. Two experts (ZA and HGH) independently reviewed information, including demographics, article content, research methods, and the results of each article, to identify studies that may have been relevant to this research. The criteria for data extraction from the selected articles included the year of publication, the objectives of the article, the type of study, the intervention performed,‏ the inquiry for BID gene expression, and the drugs used during the interventions. The applied principles to find appropriate articles and methods were continuously compared with the content of the PRISMA 2020 statement to ensure comprehensive coverage of relevant articles [21]. The protocol for this systematic review was registered with the PROSPERO website (CRD42023455350, https://www.crd.york.ac.uk/prospero/display_record.php).

Risk of bias

Two researchers, ZA and HGH, performed their assessments independently to assess potential bias. A third rater (MP and ZS) is consulted in disagreement between the two raters. The assessment of bias used the Risk of Bias in Interventions-Nonrandomized Studies (ROBINS-I) instrument, which includes eight different domains: confounding bias, outcome measurement bias, participant selection bias, classification bias, bias due to deviations in the intended interventions, bias due to excluded data, bias in the selection of reported outcomes, and general bias. This rigorous approach ensured a comprehensive assessment of potential biases in the included studies [22].

Bioinformatics analysis

Mutation, gene expression, and dna methylation analysis

To study Cancer Genomics, cBioPortal (v5.1.10; http://www.cbioportal.org) was used to retrieve the mutational profile of the BID gene in the TCGA-COAD cohort [23, 24]. Additionally, cBioPortal was utilized to investigate the relationship between BID gene expression, copy number alterations (CNAs), and DNA methylation. The downloaded genomic datasets were derived from TCGA pipelines, which utilize RSEM (RNA-Seq by Expectation–Maximization) normalized expression values, GISTIC2.0 for CNAs, and the Illumina Infinium HumanMethylation450 BeadChip array for DNA methylation.

To examine mRNA expression levels, we used GEPIA2 (http://gepia2.cancer-pku.cn) and UALCAN (http://ualcan.path.uab.edu/) [25, 26]. GEPIA2 integrates TCGA and GTEx RNA-seq datasets, applying log2(TPM + 1) transformed values, quantile normalization, and differential expression analysis using one-way ANOVA. UALCAN, which uses TCGA Level 3 RNA-seq data, was additionally employed to compare BID expression between COAD tumor tissues and normal controls, as well as across clinical subgroups stratified by tumor stage, grade, and other clinicopathological variables.

The prognostic relevance of BID was assessed using both GEPIA2 and UALCAN. Kaplan–Meier survival plots were generated, and statistical significance was calculated by log-rank tests, with hazard ratios derived from Cox proportional hazards regression.

The relationship between BID expression and immune infiltration was analyzed using TIMER2 (http://timer.cistrome.org) and GEPIA2021 (http://gepia2021.cancer-pku.cn/) [25, 27]. TIMER2 utilizes deconvolution-based algorithms, such as CIBERSORT, quanTIseq, and EPIC, to estimate the abundance of tumor-infiltrating immune subsets. In contrast, GEPIA2021 enables correlation analysis between gene expression and immune markers using Spearman correlation coefficients.

For isoform-level expression, we used multiple platforms to ensure robustness. GEPIA2 was applied to retrieve the pan-cancer expression of BID isoforms and to identify differentially expressed isoforms in TCGA-COAD. The UCSC Xena Browser (https://xenabrowser.net/) [28], which hosts harmonized TCGA/GTEx RNA-seq data processed via the Toil pipeline with upper-quartile normalized TPM values, was used to directly compare expression levels of BID isoforms across COAD tissues, adjacent normal tissues, and normal colon tissues. P-values were adjusted for multiple testing using the Benjamini–Hochberg false discovery rate (FDR) method.

The epigenetic regulation of BID was further evaluated using UALCAN to extract promoter methylation profiles based on TCGA Illumina 450 K methylation array data. Comparisons between tumor and normal samples, as well as across clinical subgroups, were performed using Student’s t-test as implemented in the UALCAN platform.

To identify post-transcriptional regulators, we queried the DIANA-microT-CDS database (v5.0; http://diana.imis.athena-innovation.gr), which predicts miRNA–mRNA interactions based on binding site conservation and context scores. Candidate regulatory miRNAs were validated by cross-referencing with experimentally supported interactions in miRTarBase and confirmed in the TCGA-COAD dataset. Only common miRNAs across both databases were retained as potential BID regulators [29] Subsequently, the OncomiR database (http://www.oncomir.org/) [30] was used to assess the prognostic impact of these candidate miRNAs on the overall survival (OS) of COAD patients.

Finally, protein-level expression was investigated using the Human Protein Atlas (HPA; http://www.proteinatlas.org) [31]. Immunohistochemistry (IHC) images of BID in COAD and normal colon tissues were retrieved. Protein expression levels in HPA are semi-quantitatively scored based on staining intensity (strong, moderate, weak, or negative) and the fraction of stained cells (> 75%, 25–75%, or < 25%). These two parameters are combined to classify expression into high (strong with > 25%), medium (strong with 25%), low (moderate with 25%), or not detected (weak/negative with < 25%).

Functional enrichment analysis

The top 100 positive or negative BID-correlated genes from TCGA-COAD normal and tumor samples were retrieved from GEPIA2. The top 50 positively and negatively BID-correlated co-expression genes in TCGA-COAD tumor samples were downloaded from LinkedOmics (http://www.linkedomics.org) [32]. The protein–protein interaction (PPI) network between BID and other proteins was retrieved from STRING (http://string-db.org) [33]. To further explore the biological significance of these genes, we performed Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses using the Enrichr platform (https://maayanlab.cloud/Enrichr/) [34, 35].

BID sequence-based phylogenetic tree

Phylogenetic analyses were performed using nucleotide sequences of BID isoforms, variants, and transcripts obtained from public databases. The sequences were retrieved from GenBank and Ensembl databases. Multiple sequence alignment was conducted using Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/) and additional alignment tools available on the NCBI (National Center for Biotechnology Information) website. Sequence identity and query coverage parameters were evaluated to assess the similarity among the aligned sequences.

Results

Systematic review

The selection of included studies

Figure 2 is a flowchart illustrating the process of searching, screening, and selecting references from the literature. According to our research strategy, 661 published papers were found in all available databases. Two hundred twenty-eight duplicate articles were omitted, and afterward, two separate teams, each consisting of two members, screened 433 titles and abstracts, during which 215 papers were eliminated. Moreover, the remaining 219 articles were thoroughly evaluated, and 14 were crossed off during this process. All full-text original articles were obtained via open access or institutional subscription, and all papers were in English. Finally, a total of 205 papers were gathered for the systematic review. Supplementary Table 1 provides the details.

Fig. 2.

The Study PRISMA flowchart

Study characteristics

Of the 17 BID-associated studies, the majority were conducted in the United States and China, with four studies each, followed by South Korea, Iran, and the United Kingdom, with two studies each. Single studies were conducted in Hong Kong, Japan, and Germany.

Regarding study materials, 10 studies focused exclusively on human tissue samples, 7 studies included both human samples and cell lines, and 5 studies investigated a combination of human samples, cell lines, and animal models.

Concerning experimental methods, 11 studies examined the BID gene in human samples: 5 used RT-PCR, 3 employed immunohistochemistry (IHC), 1 conducted proteomic analysis, 1 investigated genetic polymorphisms, and 1 utilized microarray analysis. In 6 studies, the expression level of the BID gene was not assessed in human samples (Supplementary Table 2).

Apoptosis pathways

Apoptosis, or programmed cell death, is an essential biological process by which damaged, unwanted, or potentially injurious cells are removed. It is a critical constituent in cellular homeostasis and several physiological processes, including development, immune response, and tissue remodeling. While necrosis refers to uncontrolled cell death resulting from injury and disease, apoptosis is a regulated and orderly process characterized by appropriate morphological and biochemical changes [36]. Accordingly, apoptosis is primarily initiated through two pathways—extrinsic and intrinsic—which ultimately converge on a common caspase-execution phase [37].

Extrinsic pathway of apoptosis

The extrinsic apoptosis pathway initiates with the activation of death receptors from the tumor necrosis factor (TNF) family, including TNF receptor 1 (TNF-R1), CD95 (also known as APO-1 or Fas), TRAIL receptors (TRAIL-R1/DR4 and TRAIL-R2/DR5), DR3, and DR6 [38]. The Fas/FasL signaling pathway represents a critical regulatory mechanism in the induction of apoptosis. Activation of this pathway begins with Fas ligand (FasL) binding to Fas receptors, which triggers the formation of a Fas trimer. This trimerization facilitates clustering of the Fas receptor's death domains (DD), which subsequently recruits the Fas-associated death domain (FADD) adaptor protein. FADD then interacts with procaspases −8 and −10, forming the death-inducing signaling complex (DISC) at the cell membrane. The DISC assembly initiates the activation of caspases through proteolytic cleavage, triggering a cascade of enzymatic reactions that ultimately leads to apoptotic cell death [39, 40]. The activation of caspase (CASP)−8 and CASP-10 from their inactive precursors, procaspase-8 and procaspase-10, following the trimerization of FADD, serves as a pivotal step in propagating the signaling cascade of the extrinsic apoptosis pathway [41, 42].

Active CASP-8 initiates apoptosis by cleaving and activating CASP-3 and CASP-7, which are crucial in both extrinsic and intrinsic apoptosis pathways. CASP-3 drives DNA fragmentation and morphological changes typical of apoptosis, while CASP-7 aids in cell detachment from the extracellular matrix during post-apoptotic clearance. Although CASP-3 is dominant in apoptosis, CASP-7 is essential for cellular viability in later stages. Their coordinated activities ensure effective cell death and resolution of apoptosis [43–45].

Additionally, CASP-10 can signal to trigger apoptosis, even when CASP-8, the primary initiator, is absent. However, this signal is weaker in the absence of CASP-8. These findings indicate that while CASP-10 is not as effective as CASP-8, it can still contribute to ensuring apoptosis occurs, serving as a backup mechanism [46]. Additionally, CASP-8 can cleave full-length BID into its active form, tBID, which then translocates to the mitochondrial outer membrane [47]. This activation links the extrinsic and intrinsic apoptotic pathways.

Intrinsic pathway of apoptosis

The intrinsic apoptosis pathway is initiated by various cellular stressors or damage, which activate members of the BH3-only protein family. These proteins promote apoptosis by directly activating BAX and BAK, leading to mitochondrial outer membrane permeabilization (MOMP). This permeabilization forms pores that allow the release of intermembrane space proteins, such as cytochrome c (Cyt c). Cyt c release is a critical event, as it interacts with apoptotic protease-activating factor 1 (APAF1) in the cytoplasm, forming the apoptosome. The apoptosome activates caspase-9, which subsequently activates executioner caspases, including caspase-3 and caspase-7, ultimately leading to cell death. Additionally, CASP-8 can indirectly contribute to mitochondrial membrane permeabilization by cleaving the BH3-only protein BID, as previously mentioned. tBID relocates to the mitochondria, where it activates the oligomerization of BAX or BAK, leading to the release of cytochrome c [42, 48–51].

BID structure

The BCL-2 protein family members share structural similarities characterized by one to four conserved regions known as BCL-2 homology (BH1–BH4) domains (Fig. 3A). These domains comprise eight α-helical segments linked together and exhibit significant sequence homology with BCL-2 [8, 52]. Typically, BCL-2 family proteins contain four conserved BH domains, BH1, BH2, BH3, and BH4, all formed by these interconnected α-helices [52]. These domains are critical for mediating functional interactions between family members, allowing them to regulate apoptotic and anti-apoptotic signaling pathways.

Fig. 3.

BID Isoforms and Mechanism in Apoptosis. (A) The BID molecule has four isoforms: BID-L (22 kDa), BID-EL (26.8 kDa), BID-S (14.8 kDa), and BID-ES (11.3 kDa). The locations of the pro-apoptotic BH3 (purple box) and inhibitory BH3B (green box) domains are shown, along with the cleavage site for caspase-8 (red vertical line). (B) The apoptotic pathway shows how Fas/FasL and TRAIL activate DR4/DR5, leading to caspase activation, BID cleavage to tBID, mitochondrial MOMP, and apoptosis.

However, BH3-only proteins, particularly BID (BH3 Interacting Domain Death Agonist), carry only the BH3 domain and lack the BH4 domain. BID is critical in cell death pathways activated by proteases such as caspases. As a key mediator, BID acts as a sentinel, linking protease-mediated death signals to the core apoptotic mechanism [53, 54]. Cleaved BID triggers multiple mitochondrial alterations, such as the release of proteins from the intermembrane space, reorganization of cristae, loss of membrane potential, increased membrane permeability, and elevated production of reactive oxygen species [55].

The human BID gene is located on chromosome 22 at position 22q11.21. According to the NCBI Gene database, it comprises 8 exons, including five conserved coding exons and three additional exons that may be coding or non-coding. This gene is positioned in the GRCh38.p14 reference assembly within the nucleotide range of 17,734,138 to 17,774,665 on the complementary strand of chromosome 22, spanning a genomic region of approximately 40 kilobases [11, 55, 56].

Alternative splicing of the human BID gene produces several distinct isoforms: BID-EL, BID-S,BidSi6 and BID-ES [11] (Fig. 3A). BID-EL, the full-length isoform, consists of 241 amino acids, while BID-S and BID-ES are shorter forms with specific functional distinctions. BID-ES contains 99 amino acids and retains the sequence downstream of the BH3 domain, whereas BID-S comprises 137 amino acids and includes the N-terminal regulatory domain, excluding the BH3 domain. Functionally, BID-EL promotes apoptosis, whereas BID-S inhibits Fas-mediated apoptosis and counteracts the pro-apoptotic activity of cleaved BID (tBID). BidSi6 is a variably sized, short BID isoform that retains a unique exonic insert specific to Si6, displaying strong inhibitory effects on Fas-mediated apoptosis and tBID-induced cell death—activities it shares with other short BID isoforms but potentially with greater effectiveness. BID-ES, similarly, can partially attenuate tBID's pro-apoptotic effects. Importantly, these three endogenously expressed BID isoforms show distinct expression patterns, subcellular localizations, and roles in cell death [11, 57, 58]. The differential expression of these BID variants likely suggests unique functional roles within cellular apoptotic regulation [59].

BID cellular homeostasis

CASP-8 cleaves BID at Asp59 within its N-terminal region during apoptosis [60]. This cleavage produces tBID, which translocates to the mitochondria, promoting the oligomerization of BAX or BAK and releasing cytochrome c [51]. Cytochrome c subsequently activates the apoptotic cascade by triggering the activation of CASP-9 and CASP-3, ultimately leading to cell death [13, 60]. Initially, CASP-8 was identified as the enzyme responsible for cleaving and activating BID following the activation of death receptors. Later studies revealed that other proteases, such as granzyme B, calpains, lysosomal enzymes, phosphorylated kinases, and additional caspases, can also cleave, activate, or modify BID through phosphorylation. CASP-10, a homolog of CASP-8, can activate BID in the absence of CASP-8 [62]. CASP-2, an upstream caspase, may be required for successful BID cleavage in some type II cells stimulated by TRAIL (tumor necrosis factor-related apoptosis-inducing ligand). Additionally, the BID can be cleaved by the effector CASP-3 but not by CASP-7 or CASP-6 [63–66].

Calcium-dependent proteases, specifically type I and type II calpains, can also cleave human BID at the Gly70–Arg71 site within the loop region, facilitating its translocation to the mitochondria and subsequent induction of cytochrome c release in vitro [67, 68]. Granzyme B, a serine protease, induces mitochondrial dysfunction and cytochrome c release by cleaving BID within the loop region downstream of the CASP-8 cleavage site [69–71]. Lysosomal enzymes, encompassing a wide range of catabolic enzymes, also participate in the apoptotic process by cleaving BID within the loop region, activating mitochondrial pathways [72–74]. Casein Kinase I and II phosphorylate human BID at T59, preventing its cleavage by CASP-8 [75, 76]. Inhibition of protein kinase CK2 has been shown to accelerate CASP-8 activation and BID cleavage, indicating that CK2 signaling plays a crucial role in regulating this process [77].

Although few studies have investigated the role of BID transcription in regulatory processes, Sax et al. and Mandal et al. identified p53 as a positive regulator of BID [78, 79]. They demonstrated that overexpression of p53 increases BID mRNA levels in both cell lines and mouse models. BID mRNA levels rise in a p53-dependent manner both in vitro and in vivo, with strong expression observed in the splenic red pulp and colonic epithelium of γ-irradiated mice. The human and mouse BID genes contain p53-binding DNA response elements that mediate the p53-dependent transactivation of a reporter gene. Additionally, BID-null mouse embryonic fibroblasts are more resistant to DNA-damaging agents, such as adriamycin and 5-fluorouracil, and stabilize endogenous p53. Therefore, BID is a p53-responsive "chemosensitivity gene" that may enhance the cell death response to chemotherapy [78].

BID lacks the hydrophobic carboxy-terminal transmembrane (TM) domain responsible for its membrane localization. However, it shares sequence homology with other BCL-2 family proteins in its BH3 domain, essential for their interactions and pro-apoptotic function. Since Bak is constitutively localized to the mitochondria, tBID can engage with Bak through their BH3 domains, leading to Bak oligomerization and the subsequent release of cytochrome c [80, 81]. BAX is typically cytosolic but also translocates to the mitochondria upon apoptotic stimulation. Research indicates that tBID interacts strongly with Bax through its BH3 domain, facilitating BAX insertion into the mitochondrial membrane and promoting its oligomerization [82, 83]. Anti-apoptotic proteins, such as BCL-2 or BCL-xL, can sequester BID through protein–protein interactions, preventing its engagement with downstream effectors, Bax or Bak, and thereby inhibiting its pro-apoptotic activity [14, 84].

Humanin, an endogenous peptide composed of 24 amino acids, inhibits the pro-apoptotic protein Bax and engages with BID to obstruct the release of cytochrome c and SMAC. The protein MTCH2/MIMP, which interacts with tBID and is located in the outer mitochondrial membrane, facilitates the recruitment of tBID to the mitochondria. This process promotes the activation of Bax and Bak, leading to the permeabilization of the mitochondrial outer membrane and, subsequently, apoptosis [85, 86].

Ubiquitin-mediated, proteasome-dependent proteolysis regulates protein stability by tagging specific protein domains for degradation through a multistep process involving ubiquitin-activating (E1), conjugating (E2), and ligating (E3) enzymes; for BID, while the full-length form is generally stable, the cleaved N-terminal fragment (tBid-N) is marked for unconventional ubiquitination primarily within helix 1 (spanning residues 14–15), which enables its recognition and rapid degradation by the 26S proteasome, thereby facilitating the pro-apoptotic function of the C-terminal tBid fragment (tBid-C). Mutation or deletion of these key N-terminal residues in tBid-N disrupts ubiquitination, increases its stability, and inhibits the release of BH3-mediated apoptotic activity, while proteasome inhibitors such as MG-132 and lactacystin stabilize tBID and enhance apoptosis by preventing this targeted degradation. Thus, the proteolytic regulation is highly domain-specific: the N-terminal domain of cleaved BID (not the full-length protein or tBid-C) is the principal site of ubiquitination and proteasomal degradation following apoptotic stimuli [87, 88]. In contrast, inhibitors specific to caspases or lysosomes do not affect the stability of tBID. Mutations in the putative ubiquitin acceptor sites within tBID stabilize the protein and enhance its mitochondrial toxicity [88]. Furthermore, the proteolytic cleavage of BID results in the release of the N-terminal tBID fragment (tBID-N), which becomes ubiquitinated and subsequently degraded by the proteasome. This process liberates the C-terminal BH3 domain of tBID (tBID-C), thus promoting apoptosis [87].

From BID-induced apoptosis to COAD

Abnormalities in apoptotic function play a critical role in the pathogenesis of COAD and its resistance to chemotherapy and radiotherapy, both of which rely on inducing cancer cell death. BID, a pro-apoptotic member of the BCL-2 family, is essential for death receptor-mediated apoptosis across various cellular systems. Its apoptotic function involves the cleavage of BID into its truncated form, tBID, which subsequently translocates to the mitochondria, promoting the release of apoptogenic proteins such as cytochrome c. This process is integral to the activation of the intrinsic apoptotic pathway, highlighting BID’s pivotal role in COAD progression and therapeutic resistance [89, 90] (Fig. 3B).

BID expression was evaluated in liver diseases, including hepatocellular carcinoma (HCC), liver metastases from COAD, chronic hepatitis, and liver cirrhosis. BID levels were lower in HCC tumor tissues compared to adjacent non-tumorous tissues, except in poorly differentiated HCC, where strong BID staining suggested apoptosis or necrosis. In chronic hepatitis and liver cirrhosis, BID expression increased from the core to the periphery, also seen in non-tumorous HCC tissues, indicating its potential as an early marker for HCC development. In liver metastases, BID was expressed at higher levels in epithelial cells than in non-tumorous liver tissues, with heavy nuclear staining in metastatic cells, which remained mitotically active despite BID positivity, suggesting proliferation rather than apoptosis. Overall, BID downregulation in HCC and its gradient increase in chronic liver diseases may serve as a pathological marker for tumor progression and carcinogenesis [91]. Further studies have demonstrated that the regulation of BID is intricately linked to various cellular mechanisms and external stimuli, including a complex network of signaling pathways, transcription factors, and miRNAs, which influence its role in apoptosis and COAD progression.

Kim et al. studied the molecular genetics of the serrated pathway in colorectal carcinogenesis, with a focus on serrated polyp formation. Their microarray analysis of serrated adenoma (SA) and normal adjacent tissues revealed 124 genes with significant expression differences, including 73 upregulated and 51 downregulated in SA. The upregulated genes were involved in cell proliferation, apoptosis, glandular formation, and metabolic pathways, while downregulated genes were linked to cell proliferation, differentiation, and signal transduction. Notably, CASP8 and BID were among the upregulated genes; previous studies have also shown that BID is associated with the progression from adenoma to carcinoma in COAD [92]. Further, Forouzesh et al. examined the expression patterns of two isoforms of the BID gene (BID-Si6 and BID-EL) in colorectal adenomatous polyps to assess their potential as biomarkers. Their results indicated a significant overexpression of both BID-Si6 and BID-EL isoforms in adenomatous polyps as well as in adjacent non-polyp tissues from the same patients compared to normal colon tissues. However, no significant differences were observed in expression levels between polyps and adjacent non-polyp tissues. Furthermore, the study found no significant correlations between the expression of these isoforms and other clinicopathological features [58].

Apo2L/TRAIL induces apoptosis through death receptors DR4 and DR5, forming both homomeric and heteromeric complexes that recruit FADD and activate caspase-8 [93]. Engagement of death receptors with Apo2L/TRAIL triggers tumor-cell death via caspase-8-mediated cleavage of BID, activating BAK or BAX and caspase-9. Ravi et al. explored the role of casein kinase II (CK2) in protecting COAD cells from Apo2L/TRAIL-induced apoptosis. They found that CK2 inhibits caspase-8-mediated cleavage of BID, reducing active tBID formation. CK2 also promotes NF-κB-mediated BCL-xL expression, which sequesters tBID and limits its activation of BAX. COAD cells with constitutive CK2 activation have a high BCL-xL/tBID ratio and resistance to Apo2L/TRAIL-induced apoptosis. Inhibition of CK2 reduces this ratio, making COAD cells sensitive to Apo2L/TRAIL-induced caspase-9 activation and apoptosis. CK2 inhibitors enhance tumor cell death through a BAX-dependent mechanism, independent of p53 [95].

Hypoxia-induced ER stress activates the UPR via PERK, resulting in eIF2α phosphorylation, global protein synthesis inhibition, and selective translation of ATF4 for adaptive signaling. However, prolonged stress leads to apoptosis through calcium leakage from the ER, activation of death effectors, and ATF4-CHOP-mediated induction of pro-apoptotic genes, while suppressing anti-apoptotic BCL-2 proteins. Therefore, the PERK/eIF2α/ATF4/CHOP signaling pathway plays a crucial role in tumor suppression during ER stress [96]. Colorectal tumorigenesis often begins with adenomatous precursor lesions, driven by alterations in the Adenomatous Polyposis Coli (APC) tumor suppressor pathway. This leads to aberrant Wnt signaling, β-catenin accumulation, and the activation of oncogenes, such as c-Myc. Non-steroidal anti-inflammatory drugs (NSAIDs) are linked to a reduced risk of COAD. Fletcher et al. showed that NSAIDs upregulate DR5, activating the extrinsic apoptotic pathway in COAD cells. NSAIDs induce endoplasmic reticulum (ER) stress, which upregulates the PERK/eIF2α/ATF4/CHOP axis, leading to increased DR5 expression. This triggers caspase-8 activation and BID cleavage, enhancing cell-surface calreticulin (CRT) translocation, which promotes phagocytosis by dendritic cells (DCs) and elevates tumor-infiltrating lymphocytes (TILs) in APC^Min/+ mouse small intestine and colon polyps through caspase-8 and BID cleavage activation [97]. Additionally, Seenath et al. found that HIF-1 suppresses the transcription of the pro-apoptotic gene BID in COAD cells, as demonstrated by in vitro experiments. They observed a notable decrease in BID expression in areas exhibiting high nuclear positivity for HIF-1alpha (HIF-1α). Given the pivotal role of BID in drug-induced apoptosis, these findings emphasize the potential of targeting HIF-1 as a therapeutic approach to improve treatment outcomes in COAD [98].

Cdk1/cyclin B1 inhibits caspase-8 activation by phosphorylating procaspase-8 at Ser-387, preventing its processing. This mechanism is particularly relevant in conditions such as breast cancer and proliferating T cells, where Cdk1/cyclin B1 activity is increased, leading to higher phosphorylation of procaspase-8 and reduced caspase-8 activation [18]. Tong et al. demonstrated through Immunohistochemistry (IHC) analysis that DPP3 levels were elevated in COAD tissues compared to normal tissues. They found a positive correlation between DPP3 expression and lymphatic metastasis, the number of positive lymph nodes, and pathological stage in COAD patients, with higher DPP3 expression linked to poor prognosis. In vitro, DPP3 downregulation significantly hindered cell proliferation, colony formation, and migration, while promoting apoptosis. Mechanistically, DPP3 depletion upregulated apoptotic regulators such as BID, BIM, Caspase3, Caspase8, HSP60, p21, p27, p53, and SMAC. In vivo, reducing DPP3 levels decreased COAD tumorigenicity. Investigations showed that DPP3 exerts these effects by targeting CDK1 [99].

In response to cellular stress, PIDD activates caspase-2, initiating pathways for apoptosis and cell cycle arrest. Active caspase-2 cleaves BID, generating tBID, which induces mitochondrial outer membrane permeabilization, leading to apoptosis. Caspase-2 also cleaves MDM2, forming MDM2 p60, which binds to and stabilizes p53, resulting in cell cycle arrest. Stabilized p53 enhances PIDD expression, creating a feed-forward loop where increased PIDD amplifies caspase-2 activation, reinforcing the apoptotic and cell cycle arrest responses to stress [100].

Chromosomal instability (CIN) contributes significantly to cancer progression, intratumor variability, and treatment resistance, arising from errors in chromosome segregation and a mechanism that allows survival of cells with abnormal chromosome numbers. A genomic analysis of COAD tissues and cell lines revealed frequent loss of heterozygosity and mutations in the BCL9L gene in tumors exhibiting chromosomal instability (CIN). Deficiency of BCL9L was shown to promote tolerance of chromosome segregation errors, maintain aneuploidy, and enhance genetic heterogeneity, potentially through modulation of the Wnt signaling pathway. BCL9L dysfunction induces aneuploidy tolerance in both TP53 wild-type and mutant cells by reducing caspase-2 levels and inhibiting MDM2 and BID cleavage. These findings suggest that targeting the BCL9L/caspase-2/BID axis and aneuploidy tolerance mechanisms could limit COAD cell diversity and evolution, offering a promising strategy to improve treatment outcomes for patients with CIN-associated COAD [101].

Previous studies have indicated that disruption of circadian rhythms contributes to the initiation and progression of COADs. Wang et al. investigated the expression patterns of the human Clock (hClock), a central gene in the circadian gene family, in COADs to explore its potential effects. They found higher levels of hClock expression in COAD tissues than adjacent non-cancerous tissues, particularly in poorly differentiated or late-stage Dukes' grade tumors and in cases with lymph node metastasis. The study also highlighted the stable expression of hClock in colorectal mucosa and its role in regulating downstream clock-controlled genes. Additionally, the findings suggest a potential interaction between hClock and HIF-1α/ARNT, which activates VEGF and promotes tumor angiogenesis and metastasis in COAD. However, no significant correlation was observed between hClock and other genes, including Bak, Bax, BID, tumor necrosis factor receptor 1 (TNFR1), and TNFR2 [102, 103]. Zhu et al. conducted a study confirming elevated levels of LDB1 expression in COAD tumor tissues, revealing a significant association between LDB1 upregulation and poor prognosis in COAD patients. They found that increased LDB1 levels correlated positively with CCNA1, BCL2, and BCLW expression while showing negative correlations with pro-apoptotic signals such as BID, BAX, and BAK. Furthermore, silencing LDB1 expression led to substantial inhibition of COAD cell growth in vitro, and COAD cells with reduced LDB1 levels exhibited decreased tumorigenesis rates in tumor-bearing nude mice. Importantly, experimental evidence highlighted that LDB1 enhanced resistance to oxaliplatin in COAD cells. Additionally, LDB1 suppression enhances oxaliplatin's anti-tumor effects, while its expression increases in oxaliplatin-resistant COAD cells. Targeting LDB1 may offer a potential therapeutic strategy for COAD treatment [104].

Mcl-1 is crucial in inhibiting apoptosis in well-differentiated cells by sequestering pro-apoptotic molecules, such as BAD, BID, and BAX, alongside other apoptotic factors. Additionally, pAKT interferes with apoptosis by promoting the interaction between BAD and BCL-XL. Previous investigations have highlighted the upregulation of pAKT and Mcl-1 expression in COAD. Henderson-Jackson et al. observed a correlation between elevated Mcl-1 protein levels and advanced grade and stage in COAD, as well as a positive association with pAKT expression. Furthermore, they documented the upregulation of pAKT during the transition from normal colon mucosa to COAD [105].

Additionally, Huang and colleagues found elevated miR-20a levels in the serum, tumor tissues, and cell lines of COAD patients. They also determined that miR-20 could sponge BID mRNA, so inhibiting miR-20a enhanced TRAIL's anti-tumor effect via the caspase-8-dependent pathway by sequestering BID. Knockdown of miR-20a increases tBID translocation to mitochondria, initiating apoptosis [106].

Hu et al. studied the effects of hydrophobic bile acids (BAs) on COAD cells, revealing a cascade of events leading to oxidative stress and cell death. They found that hydrophobic BAs generate reactive oxygen species (ROS), activating TNFα-mediated signaling pathways. This activation triggers the upregulation of AP1 (c-FOS/c-JUN) and NF-κB, which enhances the expression of Nur77, a nuclear receptor family member. Nur77 upregulates BID expression, contributing to BA-induced cell death. Moreover, the nuclear export of Nur77 is regulated by a complex with RXRα and RARβ. In the cytosol, Nur77 interacts with BCL2, converting it into a pro-apoptotic molecule, thus inducing apoptosis in COAD cells [103].

BID as a prognostic biomarker and therapeutic target in COAD

The exploration of BID as a biomarker in colorectal polyps, drug resistance, and patient outcomes has emerged as a pivotal area of investigation in COAD research. Proapoptotic BH3-only proteins, such as BAD and BID, initiate apoptosis by binding to regulatory sites on prosurvival BCL-2 proteins, thereby neutralizing their function. FA Sinicrope et al. demonstrated that high expression of BAD protein in the cytoplasm of tumor cells correlated significantly with more favorable overall survival (OS) rates in a univariate analysis. Moreover, the combined BAD and BID variable was a prognostic indicator for disease-free survival (DFS) and OS. Stage and histologic grade, but not DNA mismatch repair status, were also found to be prognostic for OS. Multivariate Cox analysis revealed that high expression of BAD and BID was an independent predictor of OS, even after adjusting for stage, grade, age, treatment, and study. Moreover, the combined BAD-BID variable independently predicted both DFS and OS, demonstrating a stronger prognostic value than either gene alone. Hence, proapoptotic BAD and BID proteins emerge as independent prognostic factors in colon cancer patients undergoing adjuvant treatment. Validation of BAD and BID expression may offer valuable insights for risk stratification and patient selection for adjuvant chemotherapy [107].

JG Kim demonstrated that the Prostaglandin synthase 2/cyclooxygenase 2 (PTGS2/COX2) 8473 T > C polymorphism is associated with prognosis for patients with recurrent or metastatic COAD treated with capecitabine and oxaliplatin. However, single-nucleotide polymorphisms (SNPs) of 14 other apoptosis-related genes (BID, FAS, FASL, TNFRSF10B, AKT1, TP53, caspase 3, caspase 6–10, BCL2L, and RIPK1) did not show a significant association with OS [108].

Aslam MN and colleagues explored the effects of Aquamin, a red algae-derived natural product rich in calcium, magnesium, and trace elements, on protein expression in the colon. Through proteomic analysis, they found that Aquamin supplementation resulted in the upregulation of various proteins involved in apoptosis, including BID and proteins associated with cell–cell adhesion and cytokeratins—conversely, Aquamin downregulated proteins related to proliferation and nucleic acid metabolism. Notably, Aquamin induced the expression of pro-apoptotic proteins such as CARD16, BIRC6, MIEN1, and NOL3, while inhibiting others like IRF3, PPM1F, CYLD, CASP14, BAD, RTN3, PAWR, MAP3K7, and MRPL41. This dual effect suggests a mechanism by which Aquamin may promote apoptosis while concurrently suppressing processes associated with cell proliferation and nucleic acid metabolism [110].

Physical activity is recognized as a protective factor against cancer, with resistance exercise methods, such as whole-body electromyostimulation (WB-EMS), showing significant anti-cancer effects. Schwappacher and colleagues investigated serum from advanced prostate cancer and COAD patients after a 12-week WB-EMS training, a tolerable resistance exercise for physically weakened patients. They found that serum from these patients inhibited the proliferation and induced apoptosis in human prostate and colon cancer cells. Additionally, the electric pulse stimulation-conditioned myotube medium also impaired the viability of cancer cells. The study revealed an increase in the expression of the BID gene, a key regulator of apoptosis, following treatment with post-WB-EMS serum. This suggests a potential mechanism through which exercise modulates apoptosis in cancer cells, shedding light on the anti-cancer effects of exercise [111].

Carcinogenesis and chemotherapy resistance may result from variations in the expression of genes that regulate apoptosis. Changes in gene expression during apoptosis can contribute to the development of COAD and resistance to 5-Fluorouracil (5-FU) treatment. Manoochehri et al. conducted research revealing the downregulation of proapoptotic genes BAX and BID and the upregulation of antiapoptotic genes CIAP1 and XIAP in 5-FU-resistant COAD cells compared to wild-type cells. Additionally, BAX and FAS genes were downregulated in tumor samples compared to adjacent normal tissues. These findings suggest that BAX downregulation may play a crucial role in colorectal carcinogenesis and resistance to 5-FU treatment [112]. These findings elucidate the pivotal role of BID as a biomarker across different stages of COAD development and therapeutic interventions, highlighting its relevance in clinical practice and research endeavors.

Bioinformatics analysis

A comprehensive review of BID’s role in COAD has highlighted its critical involvement in regulating apoptosis, promoting tumor progression, and contributing to treatment resistance. Given its intricate network of interactions with key apoptotic regulators, the dysregulation of BID and its isoforms may serve as a potential biomarker for prognosis and a therapeutic target in COAD. However, despite the accumulating evidence, a deeper understanding of BID's molecular landscape, including its expression patterns, mutational profile, and regulatory mechanisms, remains essential. To bridge this knowledge gap, we conducted a bioinformatics analysis using data from TCGA to systematically investigate BID gene expression, mutational variations, DNA methylation status, and isoform-specific expression patterns in COAD. This analysis will provide a comprehensive molecular characterization of BID and its isoforms, offering novel insights into its clinical relevance in COAD.

BID gene missense and splice mutations

To assess the mutational profile of BID across different cancers, we downloaded the alteration profile of the BID gene from cBioPortal in the TCGA Pan-Cancer Atlas. We found that mutation of BID was present in several cancer types, of which Uterine Corpus Endometrial Carcinoma, Breast Invasive Carcinoma, Head and Neck Squamous Cell Carcinoma, and COAD exhibited the highest frequency of mutation with 11, 6, 4, and 3 mutated samples, respectively (Fig. 4A). In addition, we examined the mutation types of the BID gene in COAD. Of the three aforementioned mutations, two were missense mutations, and one was a splice mutation. Further details on the locations of these mutations, as well as their functional implications, are comprehensively discussed in Supplementary Table 3.

Fig. 4.

Multiomics Analysis of BID in COAD. (A) Frequency of BID genetic alterations across various cancer types from cBioPortal. (B) BID expression levels in pancancers compared to control samples. (C) BID expression in COAD patients versus normal patients, and (D) adjacent normal tissue. (E) Comparative analysis of BID isoform transcripts in COAD and normal colon tissues, sourced from the UCSC Xena browser. (F-I) Differential BID gene expression associated with cancer stage (F), histology (G), nodal metastasis (H), and TP53 mutation status (I). (J) BID promoter methylation levels in tumor versus normal samples. (K-M) Immunohistochemistry IHC images showing BID protein staining in COAD samples, retrieved from the HPA database. (*: P-value < 0.05, **: P-value < 0.01, ***: P-value < 0.001)

BID gene and isoforms expression

We examined the expression profile of the BID gene in pan-cancers using GEPIA2 (Fig. 4B). We found that the expressions of the BID gene were significantly dysregulated in various cancers, such as BLCA, CESC, ESCA, HNSC, KIRC, LGG, LUSC, OV, PAAD, READ, SKCM, STAD, TGCT, THCA, UCEC, and UCS. The expression levels of the BID gene in COAD patients were also higher than in both normal patients and adjacent normal samples (P-value < 0.05) (Fig. 4C and D).

We proceeded to examine the differentially expressed isoforms in COAD to assess whether any of the BID isoforms exhibited differential expression in COAD patients. We found that BID-001 (BID-205 (coding)), 006, 009 (BID-L), and 202 (BID-Es1) isoforms had significantly higher expressions in tumor samples compared with normal tissues (Supplementary Table 4). Moreover, in the examination and comparison of the data retrieved from USCS Xena regarding the transcripts of BID isoforms in COAD patients and healthy colon tissue, it was discovered that the highest expression and density in cancerous tissue among the studied isoforms belong to BID-L. Furthermore, a higher density of transcripts in COAD data compared to healthy tissue has also been documented regarding the BID-EL and BID-S isoforms. In contrast, in BID-Si6, the results obtained from cancerous and healthy data were identical, and no noteworthy difference was observed (Fig. 4E).

We further studied the BID gene expression in TCGA-COAD based on clinical characteristics, including stage, histological subtypes, and nodal metastasis. The expression of BID is higher at all stages compared to normal samples. Interestingly, patients at stages I and II showed higher BID expression compared to those at stage III (P-values = 1.278e-02 and 3.584e-02, respectively; Fig. 4F). However, this difference among other stages was not statistically significant. We found that expression of BID is significantly higher in patients with colon adenocarcinoma compared to those with colon mucinous adenocarcinoma (P-value = 1.163e-09; Fig. 4G).

Moreover, we found that N2 patients exhibited a significantly lower level of BID expression compared with N0 patients (P-value = 2.637e-03; Fig. 4H). However, no significant difference was observed between the other groups with nodal metastasis. The expression of BID in TP53 mutant and non-mutant samples did not differ significantly (Fig. 4I). We also studied the differences in BID expression based on demographic features, including race, gender, weight, and age. Although all groups had higher expression levels of BID compared to normal samples, no significant in-group differences were observed in these clinical data (Supplementary Fig. 1A-D).

Further, we analyzed the predictive potential of BID expression on the prognosis of COAD patients. We found that the increased expression of BID does not account for altered OS, even after adjusting for BMI, race, and sex (P-value > 0.05; Supplementary Fig. 1E-H).

Promoter methylation of the BID gene

Analysis of BID methylation data showed that out of 16 studied probes, 11 had significant hypomethylation in COAD compared to normal patients (Fig. 4J and Supplementary Table 5). To assess the relationship between BID methylation and its expression profile, we analyzed data from the cBioPortal database. We found that there seems to be no significant correlation between BID expression and methylation level (Supplementary Fig. 2 A and B).

The protein expression of the BID gene

BID expression at the protein level was analyzed across all cancers using immunohistochemistry (IHC) data from the HPA database. Moderate to strong cytoplasmic staining was observed in most malignancies, including COAD, endometrial cancer, head and neck cancer, malignant lymphomas, liver and urothelial cancer, as well as several cases of malignant melanoma, testicular and pancreatic cancer (Supplementary Fig. 3). To conduct a more detailed analysis of BID protein levels in COAD, IHC data were collected from both healthy and cancerous COAD samples. The majority of COAD tissues exhibited moderate intensity of BID staining (6 out of 10). One tissue exhibited high intensity, two showed low intensity, and one showed no detection of BID protein intensity (Fig. 4K-M). In contrast, all normal tissues displayed moderate intensity of BID protein staining.

Identification of miRNAs targeting the BID gene

Data extracted from DIANA revealed the presence of 14 potential BID-regulating miRNAs. Among these miRNAs, hsa-miR-26b-5p and hsa-miR-26a-5p, with a prediction score > 0.9, were introduced as the main regulators of BID expression. Based on the results, they had negative regulatory effects on BID mRNA expression. The putative binding site of BID 3' UTR by hsa-miR-26b-5p and hsa-miR-26a-5p retrieved from DIANA-microT is shown in Fig. 5A.

Fig. 5.

Exploration for miRNAs on BID overexpression in COAD. (A) The putative binding site of BID 3’ UTR by hsa-miR-26b-5p and hsa-miR-26a-5p. (B and C) Kaplan-Meier curves of OS for COAD patients with high and low hsa-miR-194-3p (B) and hsa-miR-149-5p (C) expression levels. (D) GI cancers survey table, correlating BID gene expression level with the infiltration of innate (Neutrophils, Monocytes, NK cells, DC cells, Eosinophils) and adaptive immune cells (CD4+ T cells, CD8+ T cells, T regulatory cells, B cells) in cancer tissue. (E) Correlations between BID mRNA expression and immune cell infiltration in COAD tissues.

The survival predictive potential of the 14 identified miRNAs was investigated by plotting Kaplan–Meier curves. The analysis revealed a significant association between hsa-miR-194-3p, hsa-miR-149-5p, and the overall survival of COAD patients (Fig. 5B and C).

Immune signature of the BID gene

We investigated the correlation between the BID gene expression and the infiltration of various immune cells in gastrointestinal cancers. As shown in Fig. 5D, the BID gene expression exhibits the highest positive correlation with CD56 cells and the highest negative correlation with Tem CD4⁺ T cell lymphocytes in cancer tissues. Also, regarding the investigated cancer type, the highest correlation between BID gene expression and immune cell proliferation was seen in cholangiocarcinoma (CHOL).

Then, the correlation between immune cell infiltration and the BID gene expression was assessed in COAD tissues. There was a significant negative correlation between BID expression and the level of immune infiltration of B cells (Pearson's r = −0.138, P-value < 0.05), CD4⁺ T cells (Pearson's r = −0.167, P-value < 0.01), and T regulatory cells (Pearson's r = −0.149, P-value < 0.05; Fig. 5E). However, the correlations between BID expression and the infiltration level of CD8⁺ T cells, neutrophils, macrophages, NK cells, and dendritic cells were not significant (P-value > 0.05; Supplementary Fig. 4). 

Co-expression genes and protein–protein interactions (PPI) network of the BID gene

The analysis revealed the top 50 genes that exhibit both positive and negative co-expression with BID (Fig. 6A and B). Notably, DRG1, TOMM22, and PHF5A demonstrated the highest positive correlation, while MLL, OTUD7B, and PLEKHM3 displayed the strongest negative correlation with BID. We have also identified the 50 top BID-binding proteins and subsequently visualized the protein–protein network (Fig. 6C). We analyzed these proteins to identify common genes among the set of 50 proteins, and the top 100 genes correlated with BID. SIVA1 was identified as a shared gene (Fig. 6D).

Fig. 6.

Functional analysis and Co-expressed genes with BID. (A and B) Top 50 positively (A) and negatively (B) correlated genes with Bid in COAD. (C) Protein-protein interaction network of 50 Bid binding proteins. (D) Intersection analysis of BID correlated and interacting proteins. (E-H) Functional enrichment analysis of Bid in COAD. (E) GO biological process analysis. (F) GO molecular function analysis. (G) GO cellular component analysis. (H) KEGG pathway enrichment analysis.

Functional enrichment analysis of the BID gene

Gene Ontology (GO) and KEGG pathway enrichment analyses were performed by combining 100 BID-related genes and 50 BID-binding proteins. GO biological process analysis showed that most of these genes are related to the pathways or cell biology of mitotic spindle organization, anaphase-promoting complex-dependent catabolic process, microtubule cytoskeleton organization involved in mitosis, and DNA replication (Fig. 6E). GO molecular function analysis also illustrated that most of these genes are related to various molecular functions such as single-stranded DNA helicase activity, RNA binding, 3'−5' DNA helicase activity, and histone kinase activity (Fig. 6F). Nuclear chromosome, nucleus, U1snRNP, and U2 snRN are the cellular components which were found most related cellular components to these genes in GO cellular components analysis (Fig. 6G). Moreover, KEGG analysis showed that "Spliceosome", "Cell cycle", "DNA replication", "p53 signaling pathway", and " Pyrimidine metabolism" pathways may be involved in the effect of BID on COAD pathogenesis (Fig. 6H).

Phylogenetic tree analysis

To investigate the evolutionary relationships and sequence divergence within the BID gene, we constructed a comprehensive phylogenetic tree using multiple sequence alignment. Phylogenetic analysis provides insights into the conservation and divergence of gene variants, transcripts, and isoforms, highlighting potentially distinct functional subgroups. The resulting tree (Fig. 7A) illustrates the clustering of BID isoforms, variants, and transcripts, with bootstrap values indicating the robustness of each branch.

Fig. 7.

Phylogenetic tree analysis of BID genes using multiple sequence alignment (A) The Bid isoforms and different variants phylogenetic tree. Consensus tree based on alignment of the Bid isoforms and Bid variants sequences. Trees were made using RaxML v8.2.11 and bootstrap values are found on each node. (B) The Bid variants phylogenetic tree. Consensus tree based on alignment of the Bid variants sequences. Trees were made using RaxML v8.2.11 and bootstrap values are found on each node. (C) The Bid transcripts phylogenetic tree. Consensus tree based on alignment of the Bid transcripts sequences. Trees were made using RaxML v8.2.11 and bootstrap values are found on each node. (D) The Bid isoforms phylogenetic tree. Consensus tree based on alignment of the Bid isoforms sequences. Trees were made using RaxML v8.2.11 and bootstrap values are found on each node.

To further examine the sequence divergence among BID mRNA reference sequences, a separate tree was generated using eight BID variants: variant1 (NM_197966.2) through variant7 (NM_001244572.1), and variantX1 (XM_017028906.1). One variant clustered with variantX1, forming a branch distinct from the other variants, suggesting a unique evolutionary relationship (Fig. 7B).

A phylogenetic tree of BID transcripts (Bid-201 to Bid-211, Bid-205–2, and Bid-211–2) revealed a similar pattern: one transcript clustered closely with variantX1, separating it from the remaining transcripts and indicating a unique subgroup (Fig. 7C). Bid-201 (ENST00000317361_11), Bid-202 (ENST00000342111_9), Bid-203 (ENST00000399765_5), Bid-204 (ENST00000399767_6), Bid-205 (ENST00000473439_5), Bid-206 (ENST00000494097_5), Bid-207 (ENST00000550946_5), Bid-208 (ENST00000551952_5), Bid-209 (ENST00000552886_1), Bid-210 (ENST00000614949_4), Bid-211 (ENST00000622694_5), Bid-205–2 (ENST00000399774.7), Bid-211–2 (ENST00000611040.1).

Finally, A phylogenetic tree was constructed by using the multiple sequence alignment of Bid isoforms (BidL, BidEL, BidES_1b, BidS, andBidSi6) consistent with the patterns observed at the mRNA level (Fig. 7D). BidL (EU678294), BidEL (AF250233), BidES_1b (EU678293), BidS (AY005151) and BidSi6 (EU678292, MG957990, MG957991, and MH121045).

Discussion

Apoptosis is a physiologically programmed cell death that plays a vital role in the processes of cancer development [113–117]. Strategies to induce apoptosis in tumor cells include directly activating pro-apoptotic molecules, modulating anti-apoptotic proteins, or restoring the function of tumor-suppressor genes [118].

The BCL-2 family of proteins is a pivotal regulator of apoptosis, characterized by conserved BCL-2 homology (BH) domains. These proteins are classified into three groups based on their structure and function. The anti-apoptotic proteins, such as BCL-2 and BCL-XL, possess all four BH domains (BH1–4) and inhibit apoptosis by sequestering inactive caspases within the apoptosome or preventing the release of mitochondrial apoptogenic factors, including cytochrome c and apoptosis-inducing factor (AIF), into the cytoplasm [7, 8]. The pro-apoptotic effector proteins, including BAX, BAK, and BOK, contain three BH domains (BH1–3) and promote apoptosis by permeabilizing the mitochondrial membrane, facilitating the release of cytochrome c, and initiating the apoptotic cascade. The third group, the BH3-only proteins, includes members such as BID, BIM, BIK, BAD, PUMA, NOXA, and HRK, which contain a single BH3 domain [9]. BH3-only proteins detect stress signals and drive apoptosis by blocking anti-apoptotic BCL-2 proteins, activating Bax/Bak [10]. The involvement of BCL-2 family members in tumorigenesis is primarily attributed to their regulation of apoptosis. Malignancies often exhibit abnormal overexpression of anti-apoptotic BCL-2 family members or a significant reduction of pro-apoptotic BCL-2 family proteins, inhibiting apoptosis [8]. Thus, to overcome tumor treatment resistance, apoptosis may be promoted by targeting anti-apoptotic proteins of the BCL-2 family. Venetoclax, a selective BCL-2 inhibitor, and Navitoclax, a BCL-xL/BCL-2 inhibitor, are orally bioavailable drugs that induce apoptosis by directly targeting and inhibiting pro-survival BCL-2 family proteins [119–122]. BID serves a pivotal function in apoptosis by acting as a molecular bridge between the extrinsic (death receptor) and intrinsic (mitochondrial) apoptotic pathways. Once activated by death receptor signals—such as Fas or TNFα—BID is cleaved by caspase-8, producing truncated BID (tBID). This tBID translocates to mitochondria, promoting the release of cytochrome c, which then activates downstream caspases and the intrinsic pathway, amplifying the apoptotic response. Functionally, this crosstalk orchestrated by BID ensures that death receptor stimulation can efficiently trigger mitochondrial-dependent apoptosis, especially in cell types where direct caspase activation is insufficient for cell death [117, 123, 124].

Elevated BID expression has been declared in some tumors, such as gliomas and prostate cancers [125–127]. Given the pivotal role of the BID gene in regulating apoptosis, we conducted a systematic review of the literature on this gene and its isoforms. Furthermore, our bioinformatics analysis assessed various aspects, including the examination of gene expression levels, mutations, methylation patterns, transcript variations, and protein expression, as well as their implications for immune cell function. Our study has revealed a significant overexpression of the BID gene and BID-EL in COAD patients compared to healthy individuals. Since BID-EL is considered a pro-apoptotic and apoptosis-inducing isoform [128], its increase in COAD tissues may indicate the sensitivity of tumor cells to external apoptotic stimuli, including chemotherapy agents and radiotherapy, and is promising to achieve effective treatment in COAD patients.

The main BID isoforms are BID-S, BID-EL, and BID-ES. BID-S lacks the BH3 domain and suppresses tBID’s pro-apoptotic effects, while BID-EL promotes apoptosis. BID-ES includes the sequence downstream of the BH3 domain. Functional studies show that BID-EL induces apoptosis, while BID-S inhibits Fas-mediated apoptosis. These isoforms differ in expression patterns, cellular localization, and apoptosis induction, regulating BID function and influencing cell fate. BID isoforms hold promise as biomarkers and therapeutic targets, particularly in cancer, due to their differential expression and roles in apoptosis regulation. Certain isoforms, such as BidSi6 and BidEL, are overexpressed in cancerous tissues like colorectal adenomatous polyps and can be detected using molecular techniques like real-time PCR, aiding early diagnosis and prognosis. Their distinct functional properties further underscore their therapeutic potential, with targeted interventions leveraging specific BID isoform activity in oncology [11, 58, 128]. BID isoforms exhibit distinct subcellular localization patterns, which are closely linked to their structure and pro- or anti-apoptotic functions. For instance, full-length BID (BID-EL) is mainly cytosolic under normal conditions but can translocate to mitochondrial membranes upon apoptotic signaling, whereas truncated BID (tBID) efficiently localizes to the mitochondrial outer membrane to execute its apoptotic role. Other splice variants, such as BID-S and BID-ES, display unique intracellular distributions: some are retained in the cytoplasm, others may be found associated with organelle membranes, and evidence suggests that a fraction of BID can even localize to the nucleus where it contributes to DNA damage responses [11, 12, 129, 130]. Additionally, bioinformatic tools including PSORT, LOCATE, DeepLoc, and PLoc-Deep-mHum—can predict the subcellular localization of BID isoforms by analyzing amino acid sequences and evolutionary features, thus providing valuable complementary insights for isoforms with poorly characterized distributions and supporting extended in silico evaluation in future research [131–134].

In this study, it was hypothesized that BID gene mutations would lead to different outcomes. The results revealed splice mutations in COAD and missense mutations in COAD/READ. BID gene expression was examined in relation to clinical characteristics such as stage, histological subtypes, and nodal metastasis. BID expression was elevated in all stages compared to normal samples, with higher expression in stages I and II compared to stage III. Additionally, BID expression was significantly higher in N0 patients than in N2 patients. These findings suggest that higher BID expression in early-stage COAD may indicate an active apoptotic response, limiting tumor growth, whereas its reduction in advanced stages and lymph node-positive cases may facilitate tumor survival, metastasis, and resistance to apoptosis. The results support BID as a potential prognostic biomarker, with its downregulation correlating with disease progression and metastatic potential. Although BID expression was elevated across all demographic groups (race, gender, weight, age, TP53 mutation status), no significant within-group differences were found. Further investigation into BID isoform-specific expression and therapeutic reactivation strategies could provide valuable insights for improving COAD treatment.

DNA methylation, a key epigenetic modification, plays a crucial role in gene regulation, with aberrant methylation contributing to the pathogenesis of COAD and chromosomal instability. The observed differential methylation in 11 out of 16 BID probes suggests a strong association between BID epigenetic regulation and the progression of COAD. Specifically, the downregulation of BID promoter methylation in COAD, aligning with its expression profile, indicates a potential disruption in apoptosis regulation. Hypomethylation of BID may lead to altered transcriptional control, potentially resulting in dysregulated apoptotic signaling, a hallmark of cancer development. Given BID’s role in tumor suppression, its epigenetic deregulation may facilitate tumor survival and progression by weakening apoptosis-mediated cell death mechanisms.

Further, miRNAs are the most well-studied epigenetic regulators in COAD and have been reported to be associated with the progression and prognosis of COAD. The analysis identified 14 potential miRNAs regulating BID. Among them, hsa-miR-26b-5p and hsa-miR-26a-5p emerged as the strongest candidates for directly regulating BID expression. However, neither of these showed a significant association with the OS of patients with COAD. In contrast, two other BID-related miRNAs, hsa-miR-194-3p and hsa-miR-149-5p, although not the primary regulators of BID, demonstrated significant correlations with patient OS, suggesting their potential prognostic value [135, 136]. In recent studies, miR-149-5p has been demonstrated to function as a tumor suppressor in the progression and chemoresistance of various tumors, including COAD [136].

To further elucidate the mechanisms underlying BID in carcinogenesis and the development of COAD, we investigated the relationship between BID expression and immune cell infiltration. The negative correlation between BID expression and immune cell infiltration in COAD suggests a potential role for BID in modulating the tumor immune microenvironment. Higher BID expression, associated with reduced B cell, CD4 + T cell, and regulatory T cell infiltration, may indicate an apoptosis-driven limitation on the survival of immune cells within tumors. Conversely, lower BID expression could facilitate immune tolerance and tumor immune escape. Given the significance of immune infiltration in tumor progression and therapy response, these findings suggest BID as a potential immunomodulatory factor, warranting further investigation into its role in immune evasion and therapeutic targeting in COAD.

Correlated genes, PPI network, and BID functional enrichment analyses were also investigated in COAD. We identified 100 BID-related genes and 50 BID-binding proteins. It was found that the DRG1, TOMM22, and PHF5A genes have the highest positive correlation with the BID gene, while the MLL, OTUD7B, and PLEKHM3 genes are the top three genes with a negative correlation with BID. Regarding the proposed genes with the highest positive correlation with the BID gene, recent studies in COAD have shown that TOMM22 and PHF5A, which are significantly overexpressed in tumor samples, are associated with a poor prognosis for COAD patients [137, 138]. TOMM22 is a central receptor component of the outer mitochondrial membrane translocase, responsible for recognizing and translocating cytosolically synthesized mitochondrial preproteins [138]. PHF5A plays a vital role in regulating the cell cycle and maintaining the pluripotency and differentiation of stem cells [137]. Meanwhile, DRG1, the other positively correlated gene, likely plays a crucial role in preventing liver metastasis in COAD by inducing cell differentiation [139]. In addition, the functional enrichment analysis of the BID gene with 100 related genes revealed that the "Spliceosome", "Cell cycle", “DNA replication", and "p53 signaling pathway" processes may be involved in the BID gene’s mechanism of action. Pieces of evidence obtained from laboratories and clinical trials have demonstrated that dysregulation of the p53 tumor suppressor gene is one of the most common events implicated in the transformation of COAD, as well as its invasive and metastatic characteristics, in COAD [140]. Regarding the other influential pathways, it has been demonstrated that Spliceosome and DNA replication genes are misregulated in COAD patients, promoting tumorigenesis [141, 142]. Taken together, these findings highlight BID and its associated genes as key contributors to COAD tumorigenesis through their involvement in critical molecular pathways driving disease initiation and progression.

Future prospect

Future research on BID and its alternative splicing isoforms in COAD holds significant promise for advancing both diagnostic and therapeutic strategies. Our comprehensive analysis highlights the potential of BID as a prognostic biomarker and its involvement in key regulatory mechanisms, including promoter methylation, miRNA interactions, immune modulation, and co-expression networks. Further experimental validation of BID isoform expression and its functional impact on tumor progression, apoptosis resistance, and immune response could provide deeper mechanistic insights. Additionally, targeting BID through epigenetic modulators, RNA-based therapies, or small-molecule inhibitors may offer novel therapeutic avenues. Integrating BID expression profiles with multi-omics data and clinical outcomes could refine personalized treatment strategies, particularly in patients exhibiting distinct immune or molecular subtypes of COAD. Nevertheless, interpretation of multi-omic public datasets is inherently challenging, and findings may be influenced by inter-study heterogeneity and potential bias in literature synthesis. Addressing these limitations through rigorous validation in independent cohorts and standardized analytic frameworks will be crucial for translating insights into clinical applications. Considering the current limitations in available databases, future studies should investigate additional BID isoforms to provide a more comprehensive characterization of their biological and clinical significance. Future studies should also explore BID's role in therapy resistance, potentially paving the way for combination approaches that enhance the efficacy of current COAD treatments. Similarly, insights from studies in mantle cell lymphoma (MCL) reveal that O-GlcNAcylation enhances MCL sensitivity to bortezomib by stabilizing tBID and preventing its degradation. OGA inhibition using ketoconazole amplifies bortezomib-induced apoptosis, even in resistant MCL cells, while sparing normal cells [143]. These findings suggest that OGA inhibitors, such as ketoconazole, may serve as promising adjuvant therapies for drug-resistant MCL. This strategy could also inspire approaches targeting BID in COAD, where enhancing tBID stability through O-GlcNAcylation could potentiate treatment responses, particularly in cases of drug resistance.

Conclusion

In conclusion, we found that BID and, more specifically, the BID-EL isoform, are significantly upregulated in patients with COAD. The BID gene and its isoforms play a significant role in apoptosis signaling and are crucial to the pathogenesis and development of COAD. Meanwhile, considering the BID-EL function, which is a pro-apoptotic isoform, its significant over-expression in COAD patients may indicate the sensitivity of cancerous cells to external apoptotic stimuli, including chemotherapy agents and radiotherapy.

Abbreviations

HDI Human

Development index

BCL-2

B-cell lymphoma 2

COAD

Colon adenocarcinoma

BCL-XL

B-cell lymphoma-extra large

BID

BH3-interacting domain death agonist

BAX

BCL-2 associated X-protein

HIF-1

Hypoxia-inducible factor 1

TCGA

The cancer genome atlas

ROBINS-I

Risk of bias in interventions-nonrandomized studies

CNAs

Copy number alterations

RSEM

RNA-seq by expectation-maximization

GTEx

Genotype tissue expression

GEPIA

Gene expression profiling interactive analyses

ANOVA

Analysis of variance

UALCAN

The University of Alabama at Birmingham cancer data analysis

UCSC

University of California, Santa Cruz

FDR

False discovery rate

OS

Overall survival

HPA

Human Protein Atlas

IHC

Immunohistochemistry

PPI

Protein-protein interaction

GO

Gene ontology

KEGG

Kyoto Encyclopedia of genes and genomes

NCBI

National center for Biotechnology information

RT-PCR

Reverse transcription by polymerase chain reaction

TRAIL

Tumor necrosis factor-related apoptosis-inducing ligand

DD

Death domains

FADD

Fas-associated death domain

DISC

Death-inducing signaling complex

BAK

BCL-2-antagonist/killer 1

MOMP

Mitochondrial outer membrane permeabilization

Cyt c

Cytochrome c

APAF1

Apoptotic protease-activating factor 1

BH

BCL-2 homology

TM

Terminal transmembrane

SMAC

Second mitochondria-derived activator of caspases

MIMP/MTCH2

Mitochondrial carrier Homolog 2 protein

HCC

Hepatocellular carcinoma

SA

Serrated adenoma

CK2

Casein kinase II

ER

Endoplasmic reticulum

UPR

Unfolded protein response

CHOP

C/EBP homologous protein

PERK

Protein kinase R (PKR)-like endoplasmic reticulum kinase

eIF2α

Eukaryotic translation initiation factor 2A

ATF4

Activating transcription factor 4

APC

Adenomatous Polyposis Coli

CRT

Cell-surface calreticulin

DCs

Dendritic cells

TILs

Tumor-infiltrating lymphocytes

DPP3

Dipeptidyl peptidase 3

BIM/BCL2L11

BCL-2 Interacting Mediator of cell death/BCL-2-like 11

PIDD

P53-induced protein with a death domain

MDM2

Mouse double minute 2 homolog

CIN

Chromosomal instability

BCL9L

B-cell CLL/lymphoma 9 like

hClock

Human Clock

ARNT

Aryl hydrocarbon receptor nuclear translocator

HSP60

60 KDa heat shock protein

VEGF

Vascular endothelial growth factor

TNFR

Tumor necrosis factor receptor

LDB1

LIM Domain Binding 1

CCNA1

Cyclin A1

Mcl-1

Myeloid cell leukemia-1

BAD

BCL-2 Associated Agonist Of Cell Death

Bas

Bile acids

ROS

Reactive oxygen species

AP1

Activating protein-1

RARβ

Retinoic acid receptor beta

DFS

Disease-free survival

PTGS2/COX2

Prostaglandin-Endoperoxide Synthase 2/ Cyclooxygenase-2

TNFRSF10B

Tumor necrosis factor receptor superfamily member 10B

BCL2L

BCL-2-like protein 1

RIPK1

Receptor-interacting serine/threonine-protein kinase 1

CARD16

Caspase recruitment domain family member 16

BIRC6

Baculoviral inhibitor of apoptosis repeat-containing 6

MIEN1

Migration and invasion enhancer 1

NOL3

Nucleolar protein 3

IRF3

Interferon regulatory factor 3

PPM1F

Protein phosphatase, Mg2+/Mn2+ Dependent 1F

CYLD

Cylindromatosis lysine 63 deubiquitinase

RTN3

Reticulon 3

PAWR

Pro-apoptotic WT1 regulator

MAP3K7

Mitogen-activated protein Kinase Kinase Kinase 7

MRPL41

Mitochondrial Ribosomal Protein L41

WB-EMS

Whole-body electromyostimulation

5-FU

5-Fluorouracil

CIAP1

Cellular inhibitor of apoptosis protein 1

XIAP

X-linked inhibitor of apoptosis protein

BLCA

Bladder carcinoma

CESC

Cervical squamous cell carcinoma

ESCA

Esophageal cancer

HNSC

Head and neck squamous cell carcinomas

LUSC

Lung squamous cell carcinoma

OV

Ovarian cancer

PAAD

Pancreatic adenocarcinoma

READ

Rectum Adenocarcinoma

SKCM

Skin cutaneous melanoma

STAD

Stomach adenocarcinoma

TGCT

Tenosynovial giant cell tumor

UCEC

Uterine corpus endometrial carcinoma

UCS

Uterine carcinosarcoma

THYM

Thymoma

THCA

Thyroid carcinoma

KIRP

Kidney renal papillary carcinoma

KIRC

Kidney renal clear cell carcinoma

KICH

Kidney chromophobe

LGG

Brain lower-grade glioma

DLBC

Diffuse large B-cell lymphoma

DRG1

Developmentally regulated GTP binding protein 1

TOMM22

Translocase of outer mitochondrial membrane 22

PHF5A

PHD finger protein 5A

MLL

Mixed lineage leukemia

OTUD7B

OTU Deubiquitinase 7B

PLEKHM3

Pleckstrin homology domain containing M3

AIF

Apoptosis-inducing factor

Bok

BCL-2 related ovarian killer

BIK

BCL-2 interacting killer

PUMA

P53-upregulated modulator of apoptosis

PMAIP1

Phorbol-12-myristate-13-acetate-induced protein 1

HRK

Harakiri, BCL-2 interacting protein

MCL

Mantle cell lymphoma

OGA

O-GlcNAcase

BMI

Body mass index

Author contributions

Conceptualization: ZS; Methodology – Systematic search: ZS, MP, AH, LSA; Methodology – Data extraction: MP, HGA, ZA, AS, and KQ; Methodology – in silico analysis: ZS, RM, and AA; Writing – original draft: AZ, MP, RM, AA, and AS; Writing – review & editing: Z.S, E.N.M, AZ, and MP; Concept map creation and Illustration and concept map: AZ, ZE, and ZS; Supervision: Z.S.

Funding

The author(s) declare that no financial support was received for the research, authorship, and/or publication of this article.

Data availability

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

Declarations

Competing interests

The authors declare no competing interests.