Abstract

   BCL2 is a critical regulator of intrinsic and extrinsic pathways of
   apoptosis that have been implicated in cancer progression and
   therapeutic resistance. In this study, the protein–protein interactions
   (PPIs) of BCL2 with potential binding partners and their role in cancer
   was investigated. A comprehensive PPI network for BCL2 has been
   generated by using the Protein Interactions Network Analysis (PINA)
   platform to identify key interactors. To further investigate the
   network, Molecular Operating Environment (MOE), Search Tool for the
   Retrieval of Interacting Genes (STRING), Residue Interaction Network
   Generation (RING), and the gProfiler server were used. Docking and
   Molecular Dynamics (MD) simulations were performed by using HDOCK and
   Gromacs to analyze the binding dynamics and stability of protein
   complexes. The BCL2 interactome revealed that three key interactors
   (p53, RAF1, and MAPK1) are involved in cancer-related processes.
   Docking studies highlighted BCL2 residues such as ASP111, ASP140,
   ARG107, and ARG146 that were predominantly involved in multiple
   hydrogen bonds, ionic interactions, and van der Waals contacts,
   highlighting conserved binding sites that play critical roles in the
   stability and specificity of protein–protein interactions. MD
   simulations (200 ns) of the BCL2-p53 complex showed that the RMSD was
   increased, suggesting the suppression of BCL2’s anti-apoptotic activity
   by p53. The RMSD for BCL2-RAF1 was also increased, showing protein
   domain structural rearrangements that enhance BCL2 anti-apoptotic
   activity. The BCL2-MAPK1 complex revealed structural, distinct
   flexibility patterns and dynamic hydrogen bonding interactions. These
   findings provide valuable insights into the molecular dynamics by which
   BCL2 modulates apoptosis and its potential as a promising therapeutic
   in cancer and apoptosis-related diseases.

1. Introduction

   The B-cell lymphoma 2 (BCL2) protein family, located on human
   chromosome 18, plays a crucial role in regulating cell death by
   apoptosis, a process essential for maintaining cellular homeostasis and
   preventing cancerous transformations. This family includes pro-survival
   and anti-apoptotic proteins that control the mitochondrial outer
   membrane (MOMP), which is responsible for releasing cytochrome c, a
   critical event in apoptosis [[26]1,[27]2]. Dysregulation of apoptotic
   regulators is frequently observed in various cancers, contributing to
   uncontrolled cell growth, differentiation, survival, and resistance to
   therapeutic treatments. BCL2 is often overexpressed during tumor
   development, indicating its importance as a survival factor. It
   inhibits intrinsic apoptosis pathways by controlling the mitochondrial
   membrane’s permeability, inhibiting the release of cytochrome c, or by
   preventing caspase activation by binding to apoptosis-inducing factor
   (AIF-1) [[28]1]. Additionally, BCL2 reduces inflammation by impairing
   NLRP1 inflammasome activation, leading to reduced CASP1 activation and
   IL1B production, which contribute to modulating the immune responses
   [[29]3].

   Apoptotic regulators can be categorized into three groups: pro-survival
   (BCL2-like proteins), pro-apoptotic BH3-only proteins, and
   pro-apoptotic effector proteins. BCL2 proteins contain at least one or
   all four conserved homology domains (BH1-4) and display two central
   hydrophobic α-helices surrounded by six or seven amphipathic α-helices
   of varying lengths [[30]4]. Pro-survival members such as BCL2, BCL-XL,
   BCLW, MCL-1, and A1 have a hydrophobic groove on their surface, which
   facilitates binding to various pro-apoptotic proteins such as BAX, BAK,
   and BID [[31]5,[32]6,[33]7,[34]8]. Only the BH3 region of pro-apoptotic
   proteins interacts with the pro-survival members. The multi-domain
   homologs such as BAK and BAX promote apoptosis, whereas others such as
   BCL2 and BCL-XL protect against apoptosis [[35]2]. The BH3-only protein
   contains a single domain responsible for interacting and regulating the
   function of the multi-domain homologs. BH3 members are pro-apoptotic in
   their behavior and are responsible for controlling the system that
   promotes the pro-apoptotic effects of BAK and BAX [[36]9]. Different
   binding affinities to multi-domain proteins have been observed in
   BH3-only proteins. This feature was associated with the sequences
   exhibited in the surface groove. BCL2 binding to BAX blocks the release
   of cytochrome c from the mitochondria.

   BCL2 regulates apoptosis through interactions with various proteins. It
   forms homo/heterodimers with pro-apoptotic proteins (BAX, BAD, BAK)
   being essential for its anti-apoptotic function [[37]10]. It forms a
   complex with proteins such as EI24, APAF1, BBC3, TP53BP2, and FKBP8
   modulating its apoptotic functions [[38]3]. Other interactions,
   including with BAG1, RAF1, and EGLN3, further regulate its
   anti-apoptotic activity by disrupting the BAX-BCL2 complex. These
   diverse interactions highlight BCL2’s role in balancing apoptosis and
   autophagy [[39]11].

   Cellular survival is governed by signaling pathways such as PI3K/AKT,
   JAK-STAT, and ERK1/2. AKT activation in the PI3K pathway phosphorylates
   and inhibits BCL2 family members promoting cell survival. Meanwhile,
   the JAK/STAT pathway induces BCL2 proteins and ERK1/2 signaling that
   enhances the transcription of BCL2 and BCL-XL through CREB
   phosphorylation. This intricate interplay between these signaling
   pathways and the BCL2 protein family ensures the regulation of
   apoptosis and cellular integrity. As the BCL2 protein family is
   involved in various signaling pathways, its dysregulation facilitates
   uncontrolled cell proliferation and contributes to therapy resistance.

   The aim of this study is to explore the protein–protein interaction
   (PPI) of BCL2 focusing on cancer-related proteins and their potential
   as a therapeutic target. To achieve this goal, a PPI network using PINA
   platform was generated to identify key interactors of BCL2 within the
   context of cancer biology. MOE, STRING, and gProfiler were utilized to
   investigate the interactions, and functional and biological
   annotations. Docking studies were performed to elucidate the novel
   interactions due to unavailable experimental structures of BCL2-MAPK1,
   and BCL2-RAF1. MD simulations (200 ns) of the key identified proteins
   were conducted to analyze the stability and conformational changes of
   protein complexes over time. By integrating the multi-dimensional
   computational approaches, a deeper insight into the molecular
   mechanisms of BCL2-associated partner proteins and their role in cancer
   can be identified.

2. Materials and Methods

   The workflow of this study is presented in [40]Figure 1. The
   methodology consisted of data compilation, data cleaning,
   protein–protein interaction (PPI) analysis, hub genes identification,
   molecular docking, and MD simulation.

Figure 1.

   [41]Figure 1
   [42]Open in a new tab

   The workflow used in the study.

2.1. Data Collection and Cleaning

   The protein–protein interaction (PPI) network of BCL2 was generated by
   the PINA (v3.0) “URL
   [43]https://omics.bjcancer.org/pina/queryProteinSet.action platform
   (accessed on 7 November 2024)”. The initial query for BCL2 (Uniprot:
   [44]P10415) yielded 133 unique protein interactors retrieved from all
   sources and large-scale studies available in the PINA repository. The
   PPI dataset was filtered to include only direct interactions, focusing
   on an experimentally validated interactome. This filtering further
   reduced the total number of interactions from 133 to 59 direct
   interactors. Direct interactions were defined as those where direct
   molecular binding/association between two proteins had been confirmed
   by experimentally verified sources. Duplicate interactors were removed
   to create a non-redundant list.

2.2. Cancer Drivers and Drug Target Selection

   The list of direct interactors was further filtered to focus on
   proteins relevant to cancer to include only those known to be cancer
   drivers and cancer drug targets. Cancer drivers and cancer drug targets
   were selected, and the common genes were identified. Subsequently, the
   interactors for each of the identified genes were queried using PINA.
   The results for each gene yielded a number of interactors that were
   filtered based on publication references. Data were cleaned to remove