BIBECHANA Vol. 22, No. 2, August 2025, 108-115 ISSN 2091-0762 (Print), 2382-5340 (Online) Journal homepage: http://nepjol.info/index.php/BIBECHANA Publisher:Dept. of Phys., Mahendra Morang A. M. Campus (Tribhuvan University)Biratnagar Structural characterization of HDAC2-MTA1 complex Narayan Gautam1,2, Narayan P. Adhikari1 1Central Department of Physics, Tribhuvan University, Kirtipur, Nepal 2Tri-Chandra Multiple Campus, Tribhuvan University, Kathmandu 44613, Nepal ∗Corresponding author. Email:narayan.adhikari@cdp.tu.edu.np Abstract Histone deacetylases are recruited to specific transcriptional repression complexes through in- teractions with corepressor proteins. This recruitment leads to chromatin condensation and transcriptional silencing. In this study, we modeled the complex structure of HDAC2 with MTA1 and investigated the HDAC2 and MTA1 interactions using all-atom molecular dy- namics (MD) simulation. Our results show that the ELM2-SANT domains of MTA1 wrap completely around HDAC2. We identified the different types of interactions such as hydrogen bonds, salt bridges, and hydrophobic interactions. Specifically, GLU186, GLU147, GLU163, LYS166, and LYS124 amino acid residues of HDAC2 form both hydrogen bond and salt bridge interactions with ARG168, ARG189, ASP187, and GLU195 amino acid residues of MTA1. Additionally, TYR15 amino acid of HDAC2 form hydrogen bonds with GLU195 amino acid of MTA1. In addition to hydrogen bond and salt bridge interactions, we also analyzed hy- drophobic interaction. Keywords HDAC2, MTA1, Molecular dynamics simulation, Salt bridge, Hydrogen bonding, Hydrophobic in- teractions, Solvent-accessible surface area, MM-GBSA Article information Manuscript received: January 19, 2025; Revised: February 6, 2025; Accepted: February 10, 2025 DOI https://doi.org/10.3126/bibechana.v22i2.74254 This work is licensed under the Creative Commons CC BY-NC License. https://creativecommons. org/licenses/by-nc/4.0/ 1 Introduction Nucleosomes are the basic units of chromatin that consist of histone and non-histone proteins in eu- karyotic cells. DNA is wrapped around histones that help the DNA fit into the nucleus. Transcrip- tion factors are proteins that interact with the DNA and help to remodel the chromatin structure by re- pressing or expressing the transcription. When the lysine residues on the amino-terminal of histone are acetylated, the interaction between the histones and DNA is weakened, and the chromatin is accessible for the transcription. The acetylation is possible due to the enzyme HATs. When the acetyl group is removed with the help of HDACs from the lysine of histones on the amino-terminal, the interaction of the histone with DNA is strengthened. At that time, the chromatin is inaccessible for the transcrip- tion. Many studies have predicted that there is an im- portant link between HDACs and cancer. HDAC- related interactions are known to be associated with transcriptional repression of some genes, and abnor- mal recruitment of HDACs to their promoter is con- 108 http://nepjol.info/index.php/BIBECHANA narayan.adhikari@cdp.tu.edu.np https://doi.org/10.3126/bibechana.v22i2.74254 https://creativecommons.org/licenses/by-nc/4.0/ https://creativecommons.org/licenses/by-nc/4.0/ Narayan Gautam and Narayan P. Adhikari/ BIBECHANA 22 (2025) 108-115 109 tributed to the tumorgenesis. HDAC (NuRD com- plex) with that of MTA1 have indicated a possible function of HDACs in tumor progression. MTA1 also physically interacts with HDAC1. MTA1 is a target of heregulin (HRG) and plays a significant role in the repression of estrogen receptor (ER)- mediated transcription by recruiting HDACs [1]. The metastatic is regulated by different genes. Out of these genes, MTA proteins play crucial roles in the development and spread of cancer cells [2, 3]. Histone modification is one of the important epi- genetic modifications that is essential for the reg- ulation of gene expression. Histone deacetylases (HDACs) are enzymatic proteins that remove acetyl groups from histone and non-histone proteins which repress the gene expression [4,5]. HDACs are asso- ciated with different biological processes such as cel- lular proliferation, apoptosis, and signal transduc- tion [6, 7]. Overexpression and abnormal expres- sion of HDACs to their promoter causes different types of diseases such as cancer, neurodegenerative disorders, HIV, diabetes, inflammation and cardiac diseases [8]. The inhibition of histone deacety- lases (HDACs) results in increased acetylation of histones and various non-histone proteins, encom- passing signaling mediators [9, 10], tumor suppres- sors [11], transcription factors [12], and DNA repair proteins [13]. There are 18 members in the HDAC family, divided into 4 classes based on their struc- ture and function [14]. Class I includes HDAC1-3 and 8; class II includes HDAC4-7, 9, and 10; class III includes sirtuin enzymes (SIRT 1-7); and class IV comprises only HDAC11, which is structurally different from the others. Within class II, there are further subdivisions: class IIa (HDAC4, 5, 7, and 9) and class IIb (HDAC6 and 10), based on their sub- cellular localization and expression patterns. Hi- stone deacetylases 1 and 2 are required for brain development [15] and heart development [16]. The different types of MTA proteins are MTA1, MTA2, and MTA3 [17, 18]. MTA1 and MTA2 are expressed ubiquitously but the expression pattern of MTA3 is restricted. MTA1 and MTA2 are mainly found in the nucleus [19]. MTA1 was originally identified through from rat metastatic breast tu- mors [20] after that it was found that MTA1 is widely up-regulated in human tumors [21]. MTA1, MTA2, including HDAC1, HDAC2 are the sub- units of NuRD complex that provide the essential role of MTA and HDACs proteins to the chromatin modification [22, 23]. There are different domains in the MTA1 protein. Bromo-adjacent-homology (BAH) domain, the EGL-27 and MTA1 homol- ogy 2 (ELM2) domain, and the SWI3, ADA2, N- CoR, and TFIII-B (SANT) domains are the im- portant domains [18]. After recruiting HDACs, the ELM2 domain functions as a transcriptional repres- sion domain [24]. The ELM2 and SANT domain of the MTA1 establish numerous interactions with the HDAC [1,25]. The structure of HDAC1 in com- plex with MTA1 from the NuRD complex was re- ported by Millard and coauthors in 2013. They reported that ELM2-SANT domains from MTA1 wrap completely around HDAC1 [25]. HDAC2 con- tains many domains with defined functions. The catalytic domain common to all class I HDACs is formed by a stretch of more than 300 amino acids constituting a large portion of the protein [26]. A nuclear localization signal (NLS) at the C- terminus can only be found in HDAC1 [27], whereas a C-terminal coiled-coil domain (possibly enabling additional protein-protein associations) seems to be specific for HDAC2 [6]. Despite previous re- search into interactions between HDAC2 and MTA1 [25], the precise nature of the HDAC2-MTA1 func- tional complex remains unknown. We used all-atom molecular dynamics (MD) simulations to investi- gate the interactions of HDAC2 and MTA1. MD simulations and in silico studies have successfully been used to investigate the molecular interactions, including HDACs [28, 29]. First, we predicted the structure of the HDAC2-MTA1 complex and per- formed MD simulations to investigate the HDAC2- MTA1 interactions. We identified the binding sites and crucial residues and estimated the key param- eters related to the structural stability and binding affinity of the complex. 2 Methods 2.1 System preparation The structure of the HDAC2-MTA1 complex was predicted with AlphaFold2 Colab [30] using the FASTA sequences of HDAC2 (UniProt [31] acces- sion ID Q92769) and MTA1 (UniProt [31] accession ID Q13330). HDAC2 and MTA1 proteins have 488 and 715 amino acids, respectively. The AlphaFold2 prediction of the HDAC2-MTA1 complex was ob- tained with full-length sequences for HDAC2 and MTA1. 2.2 MD simulations We utilized the NAMD software [32] and the CHARMM36m force field [33] to conduct all-atom MD simulations of the HDAC2-MTA1 complexes, as employed in previous studies [28, 29, 34]. The simulation input files were prepared using the CHARMM-GUI web server [35]. In cubic boxes, the systems were solvated with a TIP3 water model and 150 mM NaCl. The HDAC2-MTA1 complex, after being solvated and ion-neutralized, contained a total of 139681 atoms. The system was minimized for 10,000 steps with heavy atoms being harmoni- cally restrained, followed by an equilibration of 100 Narayan Gautam and Narayan P. Adhikari/ BIBECHANA 22 (2025) 108-115 110 picoseconds using the NVT ensemble. The produc- tion runs were performed using a time step of 2 femtoseconds. 2.3 Data analysis The VMD software [36] was utilized to analyze sim- ulation trajectories. ChimeraX [37] and VMD were utilized to effectively visualize protein structures and generate high-quality figures. The binding free energy was calculated using the MM/GBSA ap- proach with NAMD, as described in previous pub- lications [28,29,34]. The equation used to estimate binding free energy for the complex is as followed with the entropy term ignored [38]: ∆G = GHDAC2−MTA1 −GHDAC2 −GMTA1 (1) where, GHDAC2−MTA1 or GHDAC2 or GMTA1 is the sum of energy contributions due to inter- nal (bond, angle, dihedral, and improper), elec- trostatic, van der Waals, and solvation (polar and non-polar contributions) energies (G = Einternal + Eelectrostatic + Evdw + Esolvation). A contact dis- tance cutoff of 3.5 Å and an angle cutoff of 30 de- grees were used to analyze hydrogen bonding, and the same atomic distance cutoff was used to analyze salt bridges. Contact surface area was calculated using the formula [39]: ContactArea = SASAHDAC2 + SASAMTA1 − SASAHDAC2−MTA1 2 (2) where, SASAHDAC2 corresponds to solvent- accessible surface area (SASA) of HDAC2 at any time, SASAMTA1 to SASA of MTA1, and SASAHDAC2−MTA1 to SASA of the complex. VMD was used to measure the SASA values using the simulation trajectories. 3 Results and Discussion We used AlphaFold2 to predict the molecular struc- ture of the HDAC2-MTA1 complex and performed MD simulations to study into how the complex formed between HDAC2 and the MTA1 protein. To evaluate the structural stability of the predicted complex and identify the crucial residues that con- tribute to the formation, we performed an MD sim- ulation. Additionally, we calculated the binding free energy and contact surface area of the HDAC2- MTA1 complex. 3.1 Prediction of the HDAC2-MTA1 com- plex Figures 1(a) and 1(b) illustrate the individual com- plete structures of HDAC2 and MTA1 proteins predicted from AlphaFold2 respectively. Although the MTA1 protein exhibits a highly disordered structure with unstructured loops, HDAC2 has a more stable structure. Figure 1(c) illustrates the HDAC2-MTA1 complex structures predicted by Al- phaFold2. This predicted model’s interface pre- dicted template modeling (ipTM) score is 0.76. Model confidence is typically associated with an ipTM score of 0.75 or higher [40]. To the best of our knowledge, the specific HDAC2 amino acids that interact with MTA1 have not yet reported. We were able to predict the structure of the HDAC2- MTA1 complex by giving Alphafold2 a full-length sequence of both HDAC2 and MTA1. The HDAC2- MTA1 complex included a full-length HDAC2 and MTA1 sequence, as illustrated in Figure 1(c). A robust cutoff of 5 Å is effective in establishing solid foundations of the Protein Structure Networks ap- proach [41]. The amino acids in the MTA1 protein strongly interact with HDAC2 in several areas, in- cluding the helix, the beta sheet, and the unstruc- tured loops. To make the system as small as pos- sible, we removed unstructured loops and helices that were away from more than 5 Angstroms from where the HDAC2 and MTA1 meets as shown in figure 1(d). This truncation significantly reduces the size of the system as well as the computational cost. Subsequently, we performed a 500 ns MD sim- ulation of the complex that included all atoms. Figure 1: (a) Structure of HDAC2 (b) Structure of MTA1(c) Modeled Structure of HDAC2-MTA1 Complex; (d) Structure of the HDAC2-MTA1 com- plex after removing the loops beyond five angstroms form each other. Narayan Gautam and Narayan P. Adhikari/ BIBECHANA 22 (2025) 108-115 111 3.2 Structural stability of the HDAC2- MTA1 complex We monitored the structural stability of the HDAC2-MTA1 complex using RMSD measure- ments as shown in figure 2(a) and used simulation results to confirm HDAC2-MTA1 complex forma- tion. RMSD plot shows the initial reorganizations until 280 ns, but after that the structural stability of the complex was maintained with relatively sta- ble RMSD measurements. Figure 2(b) shows the structure of the complex after 500 ns of simulation of the HDAC2-MTA1. Figure 2: (a) RMSD of the HDAC2-MTA1 complex from the 500 ns simulation trajectory. (b) HDAC2- MTA1 complex structure after 500 ns of simulation time. 3.3 Major interaction involved in the com- plex formation and stabilization To understand the mechanism of the interaction of HDAC2 with MTA1, we explored their dynamics during the simulation of the complex. To iden- tify the inter-protein interactions, we evaluated Al- phaFold protein-protein interaction with ChimeraX in the predicted complex structure of the HDAC2 and MTA1 that lie within a 5 Angstrom distance in the HDAC2-MTA1 complex. Figure 1(d) shows the structure of the HDAC2-MTA1 complex, in which both proteins are within 5 Angstrom to each other. Alphafold believes that blue color lines have high confidence, which means residue pairs that are close to each other, and red color lines indicate that the residue pairs have low confidence. 3.3.1 Hydrogen bonding between HDAC2 and MTA1 To predict the hydrogen bonds formed between HDAC2 and MTA1, we analyzed the data from the simulation trajectories. This was carried out us- ing the analysis protocol outlined in the methods section. Frequently, a specific amino acid in one protein would form hydrogen bonds with another protein via different atoms. In our simulation, we only took into account individual bonds that were occupying more than 25%. Table 1 displays the occupancies of hydrogen bonds. For each pair of residues, the occupancies presented in Table 1 rep- resent the sum of hydrogen bonding contributions from various atoms within the same residue. The time evolution of each hydrogen bond established by each atom pair from the analysis of the trajec- tory are shown in Figure 3(a). The types of atoms are shown inside parentheses. Figure 3: (a) Distance of hydrogen bonds formed between the residues in HDAC2 and MTA1. The type of atoms that are predicted to establish the contacts are provided inside parentheses of each residue. (b) Orientation of residues in both HDAC2 and MTA1 that form the hydrogen bonds. The crucial residues are shown in licorice representation and the black dotted lines represent the hydrogen bonds. Table 1: Hydrogen bond occupancies obtained from the analysis of trajectories of HDAC2-MTA1 sim- ulation. Except TYR15-GLU195 residues pair, all other residues pairs are involved in the formation of salt bridges as well as hydrogen bonds in the for- mation of HDAC2-MTA1 protein complex. HDAC2 MTA1 Occupancy GLU186 ARG168 156.84% GLU147 ARG168 137.00% GLU163 ARG189 130.28% TYR15 GLU195 63.64% LYS166 ASP187 58.24% LYS124 GLU195 46.00% Narayan Gautam and Narayan P. Adhikari/ BIBECHANA 22 (2025) 108-115 112 Table 2: Hydrogen bond length between the atoms of the residues of HDAC2 and MTA1 of the final frame obtained from the 500 ns of the simulation of HDAC2-MTA1 complex HDAC2 residues MTA1 residues Bond-length (Å) (Mean±S.D.) GLU186(OE2) ARG168(NE) 2.9±0.17 GLU186(OE1) ARG168(NH2) 2.9±0.19 GLU147(OE2) ARG168(NH2) 3.6±0.90 GLU147(OE2) ARG168(NH1) 3.4±0.79 GLU147(OE1) ARG168(NH1) 3.3±0.73 GLU147(OE1) ARG168(NH2) 3.5±0.87 GLU163(OE2) ARG189(NE) 3.4±0.49 GLU163(OE1) ARG189(NH2) 3.1±0.38 GLU163(OE2) ARG189(NH2) 3.2±0.48 TYR15(OH) GLU195(OE1) 3.2±0.80 LYS166(NZ) ASP187(OD1) 3.4±1.31 LYS124(NZ) GLU195(OE2) 2.8±0.26 3.3.2 Contribution of a salt bridge for the formation of the HDAC7-MEF2A complex Figure 4: (a) Time evolution of formation of salt bridges between HDAC2 and MTA1 during the complex formation. (b),(c) and (d) Orientation of the residues that formed salt bridges in licorice form. LYS124 of HDAC2 and GLU195 of MTA1 form a strong salt bridge, despite the hydro- gen bond occupancy being relatively low at only 46.00%. This salt bridge is crucial for estab- lishing the linkage between HDAC2 and MTA1, as shown in the figure 4(a) and 4(b) respec- tively. Three additional salt bridges that we pre- dicted are: GLU147(HDAC2)-ARG168(MTA1), GLU163(HDAC2)-ARG189(MTA1), and GLU186(HDAC2)−ARG168(MTA1), performed nearly equal contributions to the formation of the HDAC2-MTA1 complex. The salt bridges ex- hibit stability throughout the 500 ns simulation. The remaining salt bridge that we predicted is LYS166(HDAC2)-ASP187(MTA1) and the stabil- ity of this salt bridge remains constant throughout the simulation, except for the time period between 380 ns and 410 ns. Figure 4(b) shows the pre- cise orientation of all these residues that form the HDAC2-MTA1 complex within each protein. 3.3.3 Contribution due to hydrophobic in- teractions Figure 5: (a) Regions of hydrophobic residues that form hydrophobic interaction between HDAC2 and MTA1, and (b) Hydrophobic residues in HDAC2 (magenta texts) that establish hydrophobic interac- tions with MTA1 residues (orange red texts) with surface and licorice representation. The hydrophobic interactions play a crucial role in the formation and stabilization of the complex between HDAC2 and MTA1. Our results show that there are a number of hydrophobic inter- actions between residues in HDAC2 and MTA1 across the interface. We predicted that hydropho- bic residues ILE54, VAL119, ALA120, VAL123, LEU162, LEU165, ALA187, PHE188, ILE408, ALA410 of HDAC2 make hydrophobic interactions with ALA175, ILE177, LEU180, LEU181, LEU194, VAL198, TRP199 of MTA1. Although there are many hydrophobic interactions in the formation of HDAC2-MTA1 complex, the crucial residues that form hydrophobic interaction is shown in figure 5(a). The zoom viewed of figure 5(a) is shown in figure 5(b) in Surface and Licorice representation with Label. Narayan Gautam and Narayan P. Adhikari/ BIBECHANA 22 (2025) 108-115 113 Table 3: The hydrophobic interactions between the different amino acid residues of HDAC2 and MTA1. HDAC2 MTA1 ILE54 VAL198, TRP199 VAL119 LEU194 ALA120 LEU194 VAL123 LEU194 LEU162 ALA175, ILE177 LEU165 ILE177, LEU180 ALA187 ALA175 PHE188 ILE177 ILE408 LEU180 ALA410 LEU180, LEU181 As shown in figure 5(b), the hydrophobic residues ILE54 of HDAC2 binds strongly with VAL198, TRP199 of MTA1. Three amino acid residues VAL119, ALA120 & VAL123 of HDAC2 make hydrophobic cavity. In that cavity, LEU194 of MTA1 deeply inserted and make strong interaction. Also, IILE177 and ALA175 of MTA1 make hy- drophobic interactions with LEU162, ALA187, and PHE188 of HDAC2. Similarly, LEU165, ILE408, and ALA410 of HDAC2 interact with LEU180 and LEU181 of MTA1. We have calculated the hy- drophobic contact area of the HDAC2-MTA1 com- plex by using the equation as described in the method section. Figure 6(a) shows the contact area measurements for the HDAC2-MTA1 com- plex. We calculated the average and standard de- viation (S.D.) of all contact area values (values for each frame of entire 500 ns simulations) as shown in figure 6(a) and obtained 243.89±23.44 Å2 (Mean±S.D.). The time evolution of hydrophobic contact area was monitored over a 500 ns molecular dynamics simulation, as shown in figure. At the initial phase, the contact area is stable and is predominantly fluc- tuated between 200 and 250 Å2, which represent the relatively stable interactions with minor fluc- tuations. After 350-500 ns, the contact area is in- creased, with peak values reaching approximately about 340 Å2. The higher contact area after 350 ns is due to the formation of new hydrophobic con- tacts between ILE408 and ALA410 of HDAC2 with LEU180 and LEU181 of MTA1 in the rearrange- ment of the HDAC2-MTA1 complex. Prior theoret- ical and experimental research has indicated that burying 1 Å2 of the hydrophobic surface at the protein-protein interface increases the complex sta- bilization free energy by -15 ± 1.2 cal/mol. Thus, in the case of the HDAC2-MTA1 complex, the hy- drophobic surface area at the HDAC2-MTA1 inter- face, which measures 243.89 Å2, plays a significant role in contributing approximately -3.66 kcal/mol to its stability. This suggests that hydrophobic in- teractions are essential in the formation and stabi- lization of the HDAC2–MTA1 complex. Figure 6: (a) Contact surface area measurements HDAC2 and MTA1 in HDAC2-MTA1 complex, and (b) Time evolution of the non-bonded interaction energy in the HDAC2-MTA1 complex. We also analyzed two interactions, electrostatic and van der Waals, that help to stabilize HDAC2- MTA1 complex. Electrostatic interactions are due to the charged residues in the proteins. Our result estimated that these interactions are stable after 280 ns of simulation and its value was -1220.25 ± 126.65 kcal/mol. Van der Waals interactions are relatively weaker interactions. These interactions remained stable from the beginning to the end of our simulation. The average strength of the van der Waals interactions was -237.55 ± 20.53 kcal/mol. These negative values of both electrostatic and Van der Waals interaction energy is favorable for the interaction of HDAC2-MTA1. Both non-bonded interaction energies contribute to the stable bind- ing of the HDAC2–MTA1 complex. This observa- tion suggests that HDAC2 and MTA1 form a sta- ble complex through multiple non-covalent interac- tions. 3.3.4 Binding free energy of the complex In order to determine the strength of the in- teraction between HDAC2 and MTA1, we em- ployed the molecular mechanics with a general- ized Born and surface area solvation (MM/GBSA) method to calculate the binding free energy of the complex. We have determined the binding free energy of the HDAC2–MTA1 complex us- ing the molecular dynamics (MD) simulation tra- jectory. The MM/GBSA binding free energy of the HDAC2–MTA1 complex is -223.14 ± 17.71 kcal/mol, indicating a substantial binding free en- ergy for a stable protein–protein complex. 4 Conclusions The HDAC (NuRD) complex with that of MTA1 have indicated a possible function of HDACs in tu- mor progression. MTA1 is a target of heregulin (HRG) and plays a significant role in the repres- sion of estrogen receptor (ER)-mediated transcrip- Narayan Gautam and Narayan P. Adhikari/ BIBECHANA 22 (2025) 108-115 114 tion by recruiting HDACs. In this work, we inves- tigated the molecular mechanism of complex for- mation of HDAC2 with MTA1 protein, using struc- ture predicting tool AlphaFold2 and performed MD simulations. We obtained the HDAC2-MTA1 com- plex from molecular modeling and assessed its sta- bility using MD simulations. Our results show that HDAC2 interacts with MTA1 through ELM2- SANT domains as in HDAC1-MTA1 interaction. Several hydrogen bonding interactions between the residues at the binding interface are also involved in the binding of HDAC2 and MTA1. In addi- tion to hydrogen bonding, salt bridges between the interfacial charged residues are the also the ma- jor interactions involved in the binding. Further- more, hydrophobic interactions play a significant role in forming and stabilizing the HDAC2-MTA1 complex. Our results reveal that HDAC2 strongly wrapped by MTA1, forming a stable complex via different interactions. Our first structural charac- terization of the HDAC2–MTA1 complex outlines the molecular basis of the interaction of HDAC2 with MTA1, which offers valuable insights into the recruitment mechanism of the HDAC complex for the transcription repression mediated by MTA1. Disclosure statement There are no conflicts to declare. Author contributions NPA supervised the project. NG performed the MD simulations and analyzed the data. NG and NPA contributed to the data analysis and interpre- tation. NG wrote the manuscript NPA contributed to manuscript editing. References [1] A. Mazumdar et al. Transcriptional repression of oestrogen receptor by metastasis-associated protein 1 corepressor. 3(1):30–37, 2001. [2] D. Hanahan and R.A. Weinberg. The hall- marks of cancer. 100(1):57–70, 2000. [3] N. Sen et al. Role of mta1 in cancer progression and metastasis. 33:879–889, 2014. [4] P.A. Marks et al. Histone deacetylases and can- cer: causes and therapies. 1(3):194–202, 2001. [5] B. Weinhold. Epigenetics: the science of change. National Institute of Environmental Health Sciences, 2006. [6] I. Gregoretti, Y.-M. Lee, and H.V. Goodson. Molecular evolution of the histone deacetylase family: functional implications of phylogenetic analysis. 338(1):17–31, 2004. [7] G. Brosch, P. Loidl, and S. Graessle. Histone modifications and chromatin dynamics: a fo- cus on filamentous fungi. 32(3):409–439, 2008. [8] P. Gediya et al. Histone deacetylase 2: A po- tential therapeutic target for cancer and neu- rodegenerative disorders. 216:113332, 2021. [9] Z.-l. Yuan et al. Stat3 dimerization regu- lated by reversible acetylation of a single lysine residue. 307(5707):269–273, 2005. [10] H. Kook et al. Cardiac hypertrophy and histone deacetylase–dependent transcriptional repression mediated by the atypical home- odomain protein hop. 112(6):863–871, 2003. [11] A. Ito et al. Mdm2–hdac1-mediated deacety- lation of p53 is required for its degradation. 2002. [12] J. Luo et al. Acetylation of p53 augments its site-specific dna binding both in vitro and in vivo. 101(8):2259–2264, 2004. [13] G. Marzio et al. E2f family members are dif- ferentially regulated by reversible acetylation. 275(15):10887–10892, 2000. [14] N. Sengupta and E. Seto. Regulation of histone deacetylase activities. 93(1):57–67, 2004. [15] J. Jaworska, M. Ziemka-Nalecz, and T. Za- lewska. Histone deacetylases 1 and 2 are re- quired for brain development. 59(4-6):171–177, 2015. [16] R.L. Montgomery et al. Histone deacety- lases 1 and 2 redundantly regulate car- diac morphogenesis, growth, and contractility. 21(14):1790–1802, 2007. [17] N.J. Bowen et al. Mi-2/nurd: multiple com- plexes for many purposes. 1677(1-3):52–57, 2004. [18] B. Manavathi and R. Kumar. Metastasis tu- mor antigens, an emerging family of multi- faceted master coregulators. 282(3):1529–1533, 2007. [19] A. Simpson et al. Differential expression and subcellular distribution of the mouse metastasis-associated proteins mta1 and mta3. 273(1):29–39, 2001. [20] Y. Toh, S.D. Pencil, and G.L. Nicol- son. A novel candidate metastasis-associated gene, mta1, differentially expressed in highly metastatic mammary adenocarcinoma cell lines. cdna cloning, expression, and protein analyses. 269(37):22958–22963, 1994. Narayan Gautam and Narayan P. Adhikari/ BIBECHANA 22 (2025) 108-115 115 [21] M. Esteller. DNA methylation, epigenetics and metastasis, volume 7. Springer Science & Busi- ness Media, 2005. [22] Y. Xue et al. Nurd, a novel complex with both atp-dependent chromatin-remodeling and his- tone deacetylase activities. 2(6):851–861, 1998. [23] Y. Zhang et al. Analysis of the nurd sub- units reveals a histone deacetylase core com- plex and a connection with dna methylation. 13(15):1924–1935, 1999. [24] L. Wang et al. Atrophin recruits hdac1/2 and g9a to modify histone h3k9 and to determine cell fates. 9(6):555–562, 2008. [25] C.J. Millard et al. Class i hdacs share a com- mon mechanism of regulation by inositol phos- phates. 51(1):57–67, 2013. [26] A.J. Ruijter et al. Histone deacetylases (hdacs): characterization of the classical hdac family. 370(3):737–749, 2003. [27] J. Taplick et al. Homo-oligomerisation and nu- clear localisation of mouse histone deacetylase 1. 308(1):27–38, 2001. [28] N. Gautam et al. Characterization of molecular interactions between hdac7 and mef2a. Jour- nal of Biomolecular Structure and Dynamics, pages 1–10, 2024. [29] L.S. Dhami et al. Insights from in silico study of receptor energetics of sars-cov-2 variants. 26(11):8794–8806, 2024. [30] J. Jumper et al. Highly accurate protein struc- ture prediction with alphafold. 596(7873):583– 589, 2021. [31] UniProt Consortium. Uniprot: the universal protein knowledgebase in 2023. Nucleic Acids Research, 51(D1):D523–D531, 2022. [32] J.C. Phillips et al. Scalable molecular dynam- ics with namd. 26(16):1781–1802, 2005. [33] J. Huang et al. Charmm36m: an improved force field for folded and intrinsically disor- dered proteins. 14(1):71–73, 2017. [34] P.B. Tiwari et al. Covalent complex of dna and bacterial topoisomerase: implications in an- tibacterial drug development. 15(7):623–631, 2020. [35] J. Lee et al. Charmm-gui input gener- ator for namd, gromacs, amber, openmm, and charmm/openmm simulations using the charmm36 additive force field. 110(3):641a, 2016. [36] W. Humphrey, A. Dalke, and K. Schulten. Vmd: visual molecular dynamics. Journal of Molecular Graphics, 14(1):33–38, 1996. [37] E.F. Pettersen et al. Ucsf chimerax: Struc- ture visualization for researchers, educators, and developers. 30(1):70–82, 2021. [38] X. Zhang, H. Perez-Sanchez, and F.C. Light- stone. A comprehensive docking and mm/gbsa rescoring study of ligand recognition upon binding antithrombin. 17(14):1631–1639, 2017. [39] X. Zou et al. Recognition of methylated dna through methyl-cpg binding domain proteins. 40(6):2747–2758, 2012. [40] R. Yin et al. Benchmarking alphafold for pro- tein complex modeling reveals accuracy deter- minants. 31(8):e4379, 2022. [41] J. Salamanca Viloria et al. An optimal dis- tance cutoff for contact-based protein struc- ture networks using side-chain centers of mass. 7(1):2838, 2017. Introduction Methods System preparation MD simulations Data analysis Results and Discussion Prediction of the HDAC2-MTA1 complex Structural stability of the HDAC2-MTA1 complex Major interaction involved in the complex formation and stabilization Hydrogen bonding between HDAC2 and MTA1 Contribution of a salt bridge for the formation of the HDAC7-MEF2A complex Contribution due to hydrophobic interactions Binding free energy of the complex Conclusions