Zinc finger (ZF) protein Kaiso mediates the transcription repression by binding with methylated DNA through ZF domains and recruiting the nuclear receptor co-repressor (NCoR) complex via its BTB/POZ (Broad complex, Tramtrack, Bric-à-brac/Pox virus and Zinc finger) domain. Investigating the molecular mechanism of interactions of Kaiso with the NCoR protein is essential to understand the role of Kaiso in the transcription repression process. A detailed study on the binding mechanism of Kaiso with the NCoR complex is still lacking due to the intrinsically disordered nature of the NCoR protein. In this work, we employed molecular modeling, docking, and molecular dynamics simulation to investigate the formation of the Kaiso–NCoR complex. We modeled the complex and predicted the key interacting residues as well as the interfacial interaction involved in the binding of Kaiso with NCoR. Our results reveal that various inter-protein interactions, such as salt bridges, hydrogen bonds, and hydrophobic interactions between the interfacial residues, play crucial roles in forming and stabilizing the Kaiso–NCoR complex. Our investigations provide molecular insights into how Kaiso recruits the NCoR complex via its BTB/POZ domain and mediates transcription repression.

DNA methylation is a covalent modification of the cytosine base in DNA due to the addition of methyl group to the fifth carbon of the cytosine in CpG dinucleotides.1,2 It is a ubiquitous epigenetic modification that regulates various cellular processes, such as gene expression regulation, X-chromosome inactivation, genomic imprinting, genomic stability, and cellular differentiation and development.3–7 DNA methylation attracts specific proteins, known as methyl CpG binding proteins (MBPs), which contain the specific DNA binding domain.8 These proteins recognize and bind to the methylated CpG (mCpG) sites in the DNA and mediate to translate the DNA methylation signal into transcriptional outcomes. They recruit nuclear co-repressor complexes that modulate the chromatin structures, thereby repressing the transcription.8–10 Three classes of MBPs have been identified: methyl-CpG binding domain (MBD) proteins,10 SET and RING finger-associated (SRA) proteins,11 and C2H2 zinc finger (ZF) proteins.12,13 Kaiso is the archetypal member of C2H2 zinc finger (ZF) proteins and belongs to the BTB/POZ (Broad complex, Tramtrack, Bric-à-brac/Pox virus and Zinc finger) subfamily of transcription factors.14 It contains three classical C2H2 ZF domains in the C-terminal region that are involved in DNA recognition. As a methyl CpG binding protein (MBP), it recognizes the methylated DNA and binds with the consecutive pairs of symmetrically methylated CpG sites in DNA.15–18 Kaiso exhibits dual specificity for DNA binding: it binds with methylated CpG sites in DNA and has a longer consensus sequence of the form TCCTGCNA, known as the Kaiso binding site (KBS).15,16 Kaiso facilitates transcription repression by recruiting the nuclear co-repressor (NCoR) complex via the N-terminal BTB/POZ domain, which modulates the chromatin structure into a transcriptionally inactive state.19 It regulates the several genes involved in development and cancer. Its direct role in tumorigenesis is highlighted by its overexpression in different human cancers, such as breast,20 colorectal,21 lung,22 colon,23 and prostate24 cancers.

Nuclear receptor co-repressor (NCoR) protein exists in a large protein complex of the approximate size of 1.5–2 MDA, which consists of HDAC3, WD-40 repeat histone binding proteins TBL1 and TBLR1, cellular signaling protein GPS2, and Kaiso.19,25,26 Another essential co-repressor protein, SMRT, also known as NCoR2, shows a high level of similarity with NCoR in terms of structure and function.25,27 SMRT and NCoR mediate the transcription repression via unliganded nuclear receptors.27 NCoR plays an essential role in various cellular processes, such as cellular differentiation, development, and tumorigenesis, by mediating transcription repression by various transcription factors, such as Kaiso.19,27–30 Kaiso, as the component of the NCoR complex, is involved in the transcription repression of hypermethylated and Kaiso binding site containing genes. Previous studies have highlighted how Kaiso brings about transcription repression.19 It recruits the nuclear repressor complex to the methylated site in DNA, which modulates the chromatin structure, making it inaccessible for transcription. HDAC3 is shown to be mainly responsible for carrying out the repression function of these complexes. This process involves the histone deacetylation and H3–K9 methylation.19 

Because of the presence of flexible and intrinsically disordered domains in the NCoR protein, the structural characterization of the molecular interaction of Kaiso with NCoR is challenging and has yet to be carried out. Repression domain 1 (RD1) of NCoR is shown to interact with the BTB/POZ domain of Kaiso;19 however, specific binding sites, key residues, and major interactions involved in forming the stable complex are unknown. Identification of key residues and their role in stable complex formation is essential to understand the binding mechanism of Kaiso with NCoR, via which it recruits a co-repressor complex to mediate transcription repression.

In this work, we have studied the molecular mechanism of complex formation and stabilization between Kaiso and NCoR using a computational approach. The molecular docking technique is routinely used to predict the binding interface of the interacting proteins. Molecular dynamics (MD) simulation is widely used to investigate the atomic-level detailed dynamics of interactions between proteins and their binding partners at the molecular level.31 Here, we first modeled the structure of the Kaiso–NCoR complex using molecular modeling and docking and then performed MD simulations to investigate the structural stability, interactions, and dynamics of the best predicted complex. We identified the specific binding site and the crucial residues and estimated the key parameters related to the structural stability and binding affinity of the complex. Our results show that the three helices at the end of the RD1 region, consisting of residues 300–373, are responsible for forming a stable complex with the BTB/POZ domain of Kaiso. Since the POZ domain in Kaiso is highly hydrophobic in nature, hydrophobic interactions play an important role in the binding of Kaiso with NCoR. In addition, other non-covalent interactions, such as salt bridges, hydrogen bonds, electrostatic and van der Waals interactions, are involved in forming and stabilizing the complex. Specifically, the salt bridges ASP33(Kaiso)–ARG339(NCoR), LYS42(Kaiso)–GLU348(NCoR), and ARG44(Kaiso)–GLU348(NCoR) are the major interactions in stabilizing the complex. To the best of our knowledge, this is the first computational study to investigate the molecular mechanism of interactions of Kaiso with NCoR protein, incorporating structural fluctuation and dynamics.

The entire structure of the NCoR protein was modeled using the RoseTTAFold.32 As the RD1 domain of NCoR is known to interact with the BTB/POZ domain of Kaiso,19 we selected it as the input structure of NCoR for the molecular docking. The molecular structure for the BTB/POZ domain of Kaiso was taken from the Protein Data Bank (PDB ID: 3M4T). Molecular docking of these two protein structures was carried out using the HDOCK protein docking server.33 The best predicted Kaiso–NCoR complex was used as the input structure for molecular dynamics simulation and further analysis.

The system input files for the MD simulation were prepared using the solution builder package of the CHARMM-GUI web server.34 The system was solvated using TIP3P water in a cubical box with a 10 Å padding around the complex and neutralized by adding 0.15M of NaCl, which resulted in the system containing 59 480 atoms in the box with dimensions of 86 × 86 × 86 Å3. All-atom MD simulations were performed with the NAMD 2.14 package35 using the CHARMM36m force field.36 The energy minimization of the system was carried out for 10 000 steps using the conjugate gradient and line search algorithm to remove any steric hindrance and to bring the system to the minimum potential energy. The equilibration run was performed for 125 ps with a 1 fs integration time step at 300 K to achieve the state of thermodynamic equilibrium before the production run. During the equilibration run, the protein’s heavy atoms were harmonically restrained with a force constant of 1.0 kcal/mol/Å for the backbone and 0.1 kcal/mol/Å for the side chain. The Particle Mesh Ewald (PME) method37 was employed to calculate the long-range interactions, and the covalent bonds were constrained using the SHAKE algorithm. The cutoff distance of 12 Å was taken for non-bonded interactions. The Nosé–Hoover Langevin method with a piston period of 50 fs and a decay of 25 fs was used to control the pressure. Similarly, the temperature was controlled by employing the Langevin temperature coupling with a friction coefficient of 1 ps−1. Two independent production runs of 600 ns were carried out under the NPT condition at 300 K and 1 atm pressure, taking a 2 fs time step.

Visual molecular dynamics (VMD)38 was used for analyzing the MD simulation trajectories as well as for visualization and image rendering purposes. The hydrogen bonds were calculated using the hydrogen bond plugin in VMD, with a heavy atom distance and a bond angle cutoff of 3.5 Å and 30°, respectively. Similarly, the solvent-accessible surface area (SASA) of the proteins and complex was estimated using VMD. The non-bonded interaction energies were calculated using the NAMD energy plugin in VMD.

We performed molecular modeling, docking, and MD simulations to study the complex formation and interaction of Kaiso with the NCoR protein. The optimum complex of Kaiso and NCoR was predicted using molecular modeling and docking. We carried out an MD simulation to assess the structural stability of the predicted complex and identify the key residues and significant non-covalent interactions involved in the complex formation and stabilization. Furthermore, we estimated the contact surface area and binding free energy of the Kaiso–NCoR complex.

The complete structures of NCoR and Kaiso proteins are shown in Figs. 1(a) and 1(b). The NCoR protein has a highly disordered structure with several flexible and unstructured domains. It is a large protein consisting of 2484 amino acid residues. The RD1 domain of NCoR, consisting of residues 1–373, also consists of several alpha helices and flexible loops [Fig. 1(c)]. In contrast, the structure of Kaiso is more defined with a highly stable N-terminal BTB/POZ domain and C-terminal ZF domains. Figure 1(d) shows the crystal structure of the BTB/POZ domain (1–117 residues) of Kaiso. Previous experimental work by Yoon et al. reported that NCoR binds with the BTB/POZ domain of Kaiso via its RD1 domain.19 Based on this information, we performed the protein–protein docking of the RD1 domain of NCoR and the BTB/POZ domain of Kaiso using the HDOCK docking server.33 Out of the top ten predicted models from HDOCK, we selected the model of rank one, with a minimum docking score (−324.75) and a maximum confidence score (0.975), as the best predicted model for the Kaiso–NCoR binding. The complex with the lowest docking score and highest confidence score refers to the most possible binding mode of the proteins to form a complex.33 The structure of the Kaiso–NCoR complex predicted by HDOCK was confirmed by independent docking with PatchDock39 and FireDock.40  Figure 1(e) shows the best-predicted complex obtained from HDOCK. As the all-atom MD simulation of the entire complex is computationally expensive, we truncated the complex shown in Fig. 1(e) by removing the non-interacting region of both proteins. The complex chosen for MD simulation consists of 299–371 residues of NCoR and 9–117 residues of Kaiso, as shown in Fig. 1(f).

FIG. 1.

(a) The complete structure of the NCoR protein showing the RD1 domain (in cyan color). (b) The entire structure of the Kaiso protein showing the BTB/POZ domain (in orange color). (c) RD1 domain of NCoR. (d) BTB/POZ domain of Kaiso (PBD ID: 34MT). (e) Best predicted Kaiso–NCoR complex via molecular docking. (f) Kaiso–NCoR complex used for the all-atom MD simulation.

FIG. 1.

(a) The complete structure of the NCoR protein showing the RD1 domain (in cyan color). (b) The entire structure of the Kaiso protein showing the BTB/POZ domain (in orange color). (c) RD1 domain of NCoR. (d) BTB/POZ domain of Kaiso (PBD ID: 34MT). (e) Best predicted Kaiso–NCoR complex via molecular docking. (f) Kaiso–NCoR complex used for the all-atom MD simulation.

Close modal

Since proteins undergo structural reorganization upon interaction, the detailed study of the complex formation and stabilization dynamics can be realized by the MD simulation. Figure S1(a) shows the structure of the complex at the start of the simulation, which reveals several inter-protein interactions. As the simulation progresses, the proteins undergo structural reorganization, and the interactions become more stable. The Kaiso–NCoR complex was stable throughout the 600 ns of simulation, as shown in movie S1. Figure S1(b) shows the stable structure of the complex at the end of the 600 ns simulation, which is more compact and stable than the starting structure.

To assess the structural integrity of the Kaiso–NCoR complex, we estimated the root mean square deviation (RMSD) using the initial structure as the reference. The protein backbone atoms were taken for RMSD calculations. After initial structural reorganization, the complex becomes properly stabilized after 300 ns of the simulation, as shown by the RMSD curve in Fig. 2(a). As for individual protein components, the NCoR undergoes greater conformational rearrangement corresponding to the large fluctuations in RMSD between 0 and 300 ns, whereas the structure of Kaiso remains conserved after 150 ns, as shown in Fig. S2 of the supplementary material. Similarly, the total energy of the complex is stable in Fig. 2(b), ensuring the stable complex. Since the complex is properly stabilized after 300 ns, we have estimated all average quantities, such as RMSF, hydrogen bonds, contact area, and binding free energy from the last 300 ns (300–600 ns) of the simulation trajectories.

FIG. 2.

Time evolution of the (a) RMSD measurement and (b) total interaction energy of the Kaiso–NCoR complex during simulation. RMSF measurements of (c) Kaiso residues and (d) NCoR residues in the Kaiso–NCoR complex.

FIG. 2.

Time evolution of the (a) RMSD measurement and (b) total interaction energy of the Kaiso–NCoR complex during simulation. RMSF measurements of (c) Kaiso residues and (d) NCoR residues in the Kaiso–NCoR complex.

Close modal

We further evaluated the flexibility of the individual residues of proteins in the Kaiso–NCoR complex by calculating the root mean square fluctuation (RMSF) of the backbone Cα atoms. As shown in Figs. 2(c) and 2(d), the residues in the extremities in both proteins show significant fluctuation, as they are not involved in binding. As the BTB/POZ is the structured domain, the Kaiso residues are more stable than the NCoR residues in the complex. However, the Kaiso residues in the range 88–96, which correspond to the loop between two α-helices, are relatively more flexible. In NCoR, the residues corresponding to the multiple dips in the RMSF plot have stable interactions with Kaiso and play an important role in complex formation. In contrast, the residues 326–334, which are away from the binding region and are not involved in the interactions, have higher values of RMSF.

To understand the mechanism of the interaction of Kaiso with NCoR in the complex, we investigated various inter-protein interactions and explored their dynamics during the simulation. To identify the inter-protein interactions and residue–residue contacts, we calculated the contact map of the Kaiso–NCoR complex. The contact map represents the distance between all possible residue pairs in the complex. Figure S3 shows the representative contact map of the Kaiso–NCoR complex calculated from ProteinTools.41 The contact distances are displayed as a color-coded matrix, where lighter colors indicate residue pairs that are close to each other and darker colors indicate residue pairs that are distant from each other. The light-colored (<5 Å) cells distributed in the upper right (or lower left) block of the contact map, as shown in Fig. S3 of the supplementary material, represent the inter-protein contacts within a 5 Å distance. This work focuses on the major residues involved in the salt bridges, hydrogen bonds, and hydrophobic interactions in forming and stabilizing the Kaiso–NCoR complex. Figure 3 reveals information about the salt bridge interactions in the Kaiso–NCoR complex during MD simulation. During the simulation, the complex undergoes relaxation and structural rearrangement, resulting in the formation of salt bridges absent in the starting structure, as shown in Figs. 3(a) and 3(b). As shown in Fig. 3(c), the major salt bridges that are important to stabilize the complex are ASP33(Kaiso)–ARG339(NCoR), LYS42(Kaiso)–GLU348(NCoR), and ARG44(Kaiso)–GLU348(NCoR). The salt bridges become more stable as the simulation evolves. As shown in Fig. 3(d), ARG44–GLU348 and LYS42–GLU348 are formed within 150 ns of the simulation, whereas ASP33–ARG339 is formed only after 350 ns. Among these three salt bridges, ASP33–ARG339 is most stable after 400 ns. ASP33–ARG339 and LYS42–GLU348 are stronger than ARG44–GLU348, as indicated by the smaller distance between the interacting residue pair. Our result is consistent with the previous experimental work by Park et al.,42 which reported the important role of charged residues such as ASP33 in the interaction of Kaiso with NCoR protein and in carrying out the repression. Similarly, LYS42 at the interaction site plays an important role in the transcription repression mediated by Kaiso. The SUMOylation at LYS42 switches Kaiso from a transcription repressor to an activator.43 

FIG. 3.

The molecular complex of Kaiso (orange) and NCoR (cyan) (a) at the start of the simulation (0 ns) and (b) at the end of the simulation (600 ns). The interfacial residues involved in the inter-protein salt bridges are shown in the vdW representation. (c) Atomic-level detailed representation of the salt bridge interactions in the Kaiso–NCoR complex. (d) Time evolution of the ion pair distance for the three major salt bridges formed between NCoR and Kaiso.

FIG. 3.

The molecular complex of Kaiso (orange) and NCoR (cyan) (a) at the start of the simulation (0 ns) and (b) at the end of the simulation (600 ns). The interfacial residues involved in the inter-protein salt bridges are shown in the vdW representation. (c) Atomic-level detailed representation of the salt bridge interactions in the Kaiso–NCoR complex. (d) Time evolution of the ion pair distance for the three major salt bridges formed between NCoR and Kaiso.

Close modal

In addition, we analyzed the hydrogen bonding interactions between the interfacial residues in Kaiso and NCoR. Hydrogen bonds are crucial interactions involved in forming and stabilizing the biomolecular complexes. Figure 4(a) features the inter-protein interactions in the Kaiso–NCoR complex, including several hydrogen bonds. The total number of hydrogen bonds formed between the interfacial residues in Kaiso and NCoR as a function of simulation time is shown in Fig. 4(b), which is relatively stable throughout the simulation after 100 ns. A significant number of hydrogen bonds are formed between the interfacial residues in the Kaiso–NCoR complex. Figure 4(c) shows the percentage occupancy of the important (occupancy > 50%) hydrogen bonds for the last 300 ns of the simulation.

FIG. 4.

(a) Representative snapshot from the simulation trajectory, showing the major interactions involved in complex formation. (b) Time evolution of the total interfacial hydrogen bonds formed between Kaiso and NCoR. (c) Percentage occupancy of the major hydrogen bonds involved in the binding of Kaiso with NCoR. (d) Time evolution of the distance between the heavy atoms involved in the inter-protein hydrogen bonds in the Kaiso–NCoR complex.

FIG. 4.

(a) Representative snapshot from the simulation trajectory, showing the major interactions involved in complex formation. (b) Time evolution of the total interfacial hydrogen bonds formed between Kaiso and NCoR. (c) Percentage occupancy of the major hydrogen bonds involved in the binding of Kaiso with NCoR. (d) Time evolution of the distance between the heavy atoms involved in the inter-protein hydrogen bonds in the Kaiso–NCoR complex.

Close modal

The major residue pairs involved in the inter-protein hydrogen bond in the complex are ASP33(Kaiso)–ARG339(NCoR), LYS42(Kaiso)–GLU348(NCoR), THR35(Kaiso)–TYR342(NCoR), SER53(Kaiso)–TRP318 (NCoR), and ARG44(Kaiso)–GLU348(NCoR). Hydrogen bonds with less than 50% occupancy are not shown here. As shown in the distance plot of Fig. 4(d), the hydrogen bonds get stronger and more stable as the simulation progresses. The THR35(Kaiso)–TYR342(NCoR) bond is formed around 150 ns and remains significantly stable throughout the simulation. In addition, the SER53(Kaiso)–TRP318 (NCoR) hydrogen bond is formed toward the start of the simulation and remains stable. These major salt bridges and hydrogen bonds are also observed in another independent simulation of 600 ns (Figs. S5 and S6 and Table S1of the supplementary material), confirming the consistency of our prediction regarding their role in complex formation and stabilization.

Since the BTB/POZ domain of Kaiso is highly hydrophobic, it interacts with other proteins mainly through hydrophobic interactions. The hydrophobic interaction is one of the major interactions responsible for the stability of the protein–protein complex. Our results show that the binding interface of Kaiso and NCoR protein contains several hydrophobic residues that help us form and stabilize the complex, as shown in Fig. 5(a). Specifically, the hydrophobic residues in NCoR, ILE307, LEU314, TRP318, VAL322, PHE346, ILE349, and LEU369, present in the binding interface make hydrophobic interactions with LEU116, PHE112, ILE85, ILE49, VAL67, LEU30, and PHE31 of Kaiso. The ILE307 in NCoR makes hydrophobic contact with the LEU116 of Kaiso. Similarly, the PHE112 in Kaiso interacts with the LEU314 of NCoR. In addition, TRP318 and VAL322 of NCoR are inserted into the hydrophobic pocket created by the ILE49 and ILE85 of Kaiso. Moreover, the PHE346 in NCoR and VAL67 in Kaiso are involved in hydrophobic interactions. The ILE349 in NCoR interacts with the LEU30 of Kaiso. Furthermore, the LUE369 in NCoR occasionally contacts the LEU30 and PHE31 of Kaiso. These hydrophobic interactions are important for forming and stabilizing the complex between the BTB/POZ domain of Kaiso and the NCoR protein.

FIG. 5.

(a) Graphical representation of the hydrophobic interactions in the Kaiso–NCoR complex. (b) Time evolution of the hydrophobic contact area of the Kaiso–NCoR complex during simulation.

FIG. 5.

(a) Graphical representation of the hydrophobic interactions in the Kaiso–NCoR complex. (b) Time evolution of the hydrophobic contact area of the Kaiso–NCoR complex during simulation.

Close modal
To estimate the contribution of hydrophobic interactions to the stability of the complex, we have calculated the hydrophobic contact area of the Kaiso–NCoR complex. The contact area is the surface area buried at the binding interface between two proteins, which is unavailable for interaction with the solvent molecules. Hydrophobic interaction at the binding interface reduces the solvent-accessible surface area of the complex. It increases the contact surface area between two proteins, leading to greater stability of the complex.44,45 We have estimated the hydrophobic contact area using the following equation:44 
Contact Area=SAKaiso+SANCoRSAComplex/2,

where SAKaiso, SANCoR, and SAcomplex are the hydrophobic solvent-accessible surface area (SASA) of the Kaiso, NCoR, and Kaiso–NCoR complex, respectively.

Figure 5(b) shows the time evolution of the hydrophobic contact area at the Kaiso–NCoR interface. The average contact area for the last 300 ns of the MD simulation trajectory is 403 ± 66 Å2. Previous theoretical and experimental studies suggested that burying 1 Å2 of the hydrophobic surface at the protein–protein interface enhances the complex stabilization free energy by −15 ± 1.2 cal/mol.44,45 Therefore, in the case of the Kaiso–NCoR complex, the burial of the 403 Å2 of hydrophobic surface area at the Kaiso–NCoR interface contributes nearly −6 kcal/mol to its stability. This implies that hydrophobic interactions play a crucial role in forming and stabilizing the Kaiso–NCoR complex.

We further estimated the electrostatic and van der Waals interactions involved in the binding of the complex. As shown in Fig. 6, the electrostatic interaction energy is stable after 200 ns and has an average value of −386 ± 84 kcal/mol in the last 300 ns of the simulation. The van der Waals interaction energy is stable throughout the simulation with an average value of −103 ± 11 kcal/mol. Both non-bonded interaction energies contribute to the stable binding of the Kaiso–NCoR complex.

FIG. 6.

Time evolution of the non-bonded (electrostatic and vdW) interaction energy in the Kaiso–NCoR complex.

FIG. 6.

Time evolution of the non-bonded (electrostatic and vdW) interaction energy in the Kaiso–NCoR complex.

Close modal

To estimate the binding affinity of Kaiso with NCoR, we calculated the binding free energy of the complex using molecular mechanics with a generalized Born and surface area solvation (MM/GBSA) approach.46 Although MM/GBSA overestimates the value of binding energy, it is an efficient and computationally less expensive way to estimate the binding free energy of the complex.17,47,48 We have calculated the binding free energy of the Kaiso–NCoR complex from the last 300 ns of the MD simulation trajectory. The MM/GBSA binding free energy of the Kaiso–NCoR complex is −79.2 ± 2.1 kcal/mol, which is a significant binding free energy for a stable protein–protein complex.47 This observation indicates that Kaiso forms a stable complex with NCoR via various non-bonded interactions.

C2H2 zinc finger protein Kaiso binds with the methylated DNA and brings about transcription repression via recruiting the nuclear co-repressor complex through its N-terminal POZ/BTB domain in a methylation-dependent manner. In this work, we investigated the molecular mechanism of complex formation and stabilization of Kaiso with NCoR protein, employing structure modeling, molecular docking, and MD simulations. We obtained the Kaiso–NCoR complex from molecular docking and assessed its stability and other structural features using MD simulations. Our results show that the BTB/POZ domain in Kaiso interacts with NCoR mainly through a region containing three alpha helices with residues 300–370 at the C-terminal end of the RD1 domain. We predicted the major interacting residues and the crucial non-covalent interactions in stabilizing the complex between two proteins. Our results show that three salt bridges between the interfacial charged residues are the major interactions involved in the binding of Kaiso with the NCoR protein. In addition, several hydrogen bonding interactions between the polar residues at the binding interface are also involved in the binding. Furthermore, hydrophobic interactions play a significant role in forming and stabilizing the Kaiso–NCoR complex because of the highly hydrophobic nature of the BTB/POZ domain of Kaiso. Our results reveal that Kaiso interacts strongly with NCoR, forming a stable complex via various non-covalent interactions. Our first structural characterization of the Kaiso–NCoR complex outlines the molecular basis of the interaction of Kaiso with NCoR, which offers valuable insights into the recruitment mechanism of the NCoR complex for the transcription repression mediated by Kaiso.

(1) Figure S1 shows the structure of the Kaiso–NCoR complex at the start and end of the simulation. (2) Figure S2 shows the time evolution of the RMSD of Kaiso and NCoR proteins during simulation. (3) Figure S3 shows the contact map of the Kaiso–NCoR complex. (4) Figures S4–S6 show the time evolution of the RMSD, salt bridges, and hydrogen bond distances in the repeat run of the Kaiso–NCoR complex. (5) Table S1 compares the occupancy percentages of the primary hydrogen bonds in two independent runs. (6) A movie file showing the structural dynamics and inter-protein interactions in the Kaiso–NCoR complex.

The authors acknowledge the support from the Research Coordination and Development Council (RCDC) of Tribhuvan University under Grant No. TU-NPAR-077/78-ERG 14. B.T. acknowledges the PhD Fellowship and Research Support Grant (Award No. PhD-78/79-S&T-15) from the University Grants Commission (UGC), Nepal.

The authors have no conflicts to disclose.

B.T. performed the modeling, docking, and MD simulation, analyzed the data, and wrote the manuscript. N.P.A. supervised the work and contributed to data analysis and interpretation.

Bidhya Thapa: Data curation (lead); Formal analysis (lead); Methodology (lead); Validation (lead); Visualization (lead); Writing – original draft (lead); Writing – review & editing (equal). Narayan P. Adhikari: Conceptualization (equal); Resources (equal); Supervision (equal); Writing – review & editing (equal).

The data supporting the findings of this study are available from the corresponding author upon reasonable request.

1.
A.
Bird
, “
The essentials of DNA methylation
,”
Cell
70
(
1
),
5
8
(
1992
).
2.
A.
Rottach
,
H.
Leonhardt
, and
F.
Spada
, “
DNA methylation-mediated epigenetic control
,”
J. Cell. Biochem.
108
(
1
),
43
51
(
2009
).
3.
F.
Blasi
and
D.
Toniolo
, “
DNA methylation and X-chromosome inactivation
,”
Mol. Biol. Med.
1
(
2
),
271
274
(
1983
).
4.
E.
Li
,
C.
Beard
, and
R.
Jaenisch
, “
Role for DNA methylation in genomic imprinting
,”
Nature
366
(
6453
),
362
365
(
1993
).
5.
P. A.
Jones
and
D.
Takai
, “
The role of DNA methylation in mammalian epigenetics
,”
Science
293
(
5532
),
1068
1070
(
2001
).
6.
T. B.
Miranda
and
P. A.
Jones
, “
DNA methylation: The nuts and bolts of repression
,”
J. Cell. Physiol.
213
(
2
),
384
390
(
2007
).
7.
A.
Bird
, “
DNA methylation patterns and epigenetic memory
,”
Genes Dev.
16
(
1
),
6
21
(
2002
).
8.
P. A.
Wade
, “
Methyl CpG-binding proteins and transcriptional repression
,”
Bioessays
23
(
12
),
1131
1137
(
2001
).
9.
S. U.
Kass
,
D.
Pruss
, and
A. P.
Wolffe
, “
How does DNA methylation repress transcription?
,”
Trends Genet.
13
(
11
),
444
449
(
1997
).
10.
B.
Hendrich
and
A.
Bird
, “
Identification and characterization of a family of mammalian methyl CpG-binding proteins
,”
Genet. Res.
72
(
1
),
59
72
(
1998
).
11.
M.
Bostick
,
J. K.
Kim
,
P. O.
Estève
,
A.
Clark
,
S.
Pradhan
, and
S. E.
Jacobsen
, “
UHRF1 plays a role in maintaining DNA methylation in mammalian cells
,”
Science
317
(
5845
),
1760
1764
(
2007
).
12.
A.
Klug
, “
The discovery of zinc fingers and their applications in gene regulation and genome manipulation
,”
Annu. Rev. Biochem.
79
,
213
231
(
2010
).
13.
N. O.
Hudson
and
B. A.
Buck-Koehntop
, “
Zinc finger readers of methylated DNA
,”
Molecules
23
(
10
),
2555
(
2018
).
14.
J. M.
Daniel
and
A. B.
Reynolds
, “
The catenin p120(ctn) interacts with Kaiso, a novel BTB/POZ domain zinc finger transcription factor
,”
Mol. Cell. Biol.
19
(
5
),
3614
3623
(
1999
).
15.
J. M.
Daniel
,
C. M.
Spring
,
H. C.
Crawford
,
A. B.
Reynolds
, and
A.
Baig
, “
The p120ctn-binding partner Kaiso is a bi-modal DNA-binding protein that recognizes both a sequence-specific consensus and methylated CpG dinucleotides
,”
Nucleic Acids Res.
30
(
13
),
2911
2919
(
2002
).
16.
A.
Prokhortchouk
,
B.
Hendrich
,
H.
Jørgensen
,
A.
Ruzov
,
M.
Wilm
,
G.
Georgiev
,
A.
Bird
, and
E.
Prokhortchouk
, “
The p120 catenin partner Kaiso is a DNA methylation-dependent transcriptional repressor
,”
Genes Dev.
15
(
13
),
1613
1618
(
2001
).
17.
B.
Thapa
,
N. P.
Adhikari
,
P. B.
Tiwari
, and
P. P.
Chapagain
, “
A 5′-flanking C/G pair at the core region enhances the recognition and binding of Kaiso to methylated DNA
,”
J. Chem. Inf. Model.
63
(
7
),
2095
2103
(
2022
).
18.
B. A.
Buck-Koehntop
,
R. L.
Stanfield
,
D. C.
Ekiert
,
M. A.
Martinez-Yamout
,
H. J.
Dyson
,
I. A.
Wilson
, and
P. E.
Wright
, “
Molecular basis for recognition of methylated and specific DNA sequences by the zinc finger protein Kaiso
,”
Proc. Natl. Acad. Sci. U. S. A.
109
(
38
),
15229
15234
(
2012
).
19.
H.-G.
Yoon
,
D. W.
Chan
,
A. B.
Reynolds
,
J.
Qin
, and
J.
Wong
, “
N-CoR mediates DNA methylation-dependent repression through a methyl CpG binding protein Kaiso
,”
Mol. Cell
12
(
3
),
723
734
(
2003
).
20.
B. I.
Bassey-Archibong
,
L. G.
Rayner
,
S. M.
Hercules
,
C. W.
Aarts
,
A.
Dvorkin-Gheva
,
J. L.
Bramson
,
J. A.
Hassell
, and
J. M.
Daniel
, “
Kaiso depletion attenuates the growth and survival of triple negative breast cancer cells
,”
Cell Death Dis.
8
(
3
),
e2689
(
2017
).
21.
C. C.
Pierre
,
J.
Longo
,
M.
Mavor
,
S. B.
Milosavljevic
,
R.
Chaudhary
,
E.
Gilbreath
,
C.
Yates
, and
J. M.
Daniel
, “
Kaiso overexpression promotes intestinal inflammation and potentiates intestinal tumorigenesis in Apc(Min/+) mice
,”
Biochim. Biophys. Acta, Mol. Basis Dis.
1852
(
9
),
1846
1855
(
2015
).
22.
S.-D.
Dai
,
Y.
Wang
,
Y.
Miao
,
Y.
Zhao
,
Y.
Zhang
,
G.-Y.
Jiang
,
P.-X.
Zhang
,
Z.-Q.
Yang
, and
E.-H.
Wang
, “
Cytoplasmic Kaiso is associated with poor prognosis in non-small cell lung cancer
,”
BMC Cancer
9
,
178
(
2009
).
23.
E. C.
Lopes
,
E.
Valls
,
M. E.
Figueroa
,
A.
Mazur
,
F.-G.
Meng
,
G.
Chiosis
,
P. W.
Laird
,
N.
Schreiber-Agus
,
J. M.
Greally
,
E.
Prokhortchouk
, and
A.
Melnick
, “
Kaiso contributes to DNA methylation-dependent silencing of tumor suppressor genes in colon cancer cell lines
,”
Cancer Res.
68
(
18
),
7258
7263
(
2008
).
24.
H.
Wang
,
W.
Liu
,
S.
Black
,
O.
Turner
,
J. M.
Daniel
,
W.
Dean-Colomb
,
Q. P.
He
,
M.
Davis
, and
C.
Yates
, “
Kaiso, a transcriptional repressor, promotes cell migration and invasion of prostate cancer cells through regulation of miR-31 expression
,”
Oncotarget
7
(
5
),
5677
(
2016
).
25.
J.
Li
,
J.
Wang
,
J.
Wang
,
Z.
Nawaz
,
J. M.
Liu
,
J.
Qin
, and
J.
Wong
, “
Both corepressor proteins SMRT and N-CoR exist in large protein complexes containing HDAC3
,”
EMBO J.
19
(
16
),
4342
4350
(
2000
).
26.
H. G.
Yoon
,
D. W.
Chan
,
Z. Q.
Huang
,
J.
Li
,
J. D.
Fondell
,
J.
Qin
, and
J.
Wong
, “
Purification and functional characterization of the human N-CoR complex: The roles of HDAC3, TBL1 and TBLR1
,”
EMBO J.
22
,
1336
(
2003
).
27.
P. J.
Watson
,
L.
Fairall
, and
J. W.
Schwabe
, “
Nuclear hormone receptor co-repressors: Structure and function
,”
Mol. Cell. Endocrinol.
348
(
2
),
440
449
(
2012
).
28.
J.
Wang
,
T.
Hoshino
,
R. L.
Redner
,
S.
Kajigaya
, and
J. M.
Liu
, “
ETO, fusion partner in t(8;21) acute myeloid leukemia, represses transcription by interaction with the human N-CoR/mSin3/HDAC1 complex
,”
Proc. Natl. Acad. Sci. U. S. A.
95
(
18
),
10860
10865
(
1998
).
29.
S.
Wang
,
G.
Fusaro
,
J.
Padmanabhan
, and
S. P.
Chellappan
, “
Prohibitin co-localizes with Rb in the nucleus and recruits N-CoR and HDAC1 for transcriptional repression
,”
Oncogene
21
(
55
),
8388
8396
(
2002
).
30.
S. K.
Raghav
,
S. M.
Waszak
,
I.
Krier
,
C.
Gubelmann
,
A.
Isakova
,
T. S.
Mikkelsen
, and
B.
Deplancke
, “
Integrative genomics identifies the corepressor SMRT as a gatekeeper of adipogenesis through the transcription factors C/EBPβ and KAISO
,”
Mol. Cell
46
(
3
),
335
350
(
2012
).
31.
M.
Karplus
and
J. A.
McCammon
, “
Molecular dynamics simulations of biomolecules
,”
Nat. Struct. Biol.
9
(
9
),
646
652
(
2002
).
32.
M.
Baek
,
F.
DiMaio
,
I.
Anishchenko
,
J.
Dauparas
,
S.
Ovchinnikov
,
G. R.
Lee
,
J.
Wang
,
Q.
Cong
,
L. N.
Kinch
,
R. D.
Schaeffer
et al, “
Accurate prediction of protein structures and interactions using a three-track neural network
,”
Science
373
(
6557
),
871
876
(
2021
).
33.
Y.
Yan
,
D.
Zhang
,
P.
Zhou
,
B.
Li
, and
S.-Y.
Huang
, “
HDOCK: A web server for protein–protein and protein–DNA/RNA docking based on a hybrid strategy
,”
Nucleic Acids Res.
45
(
W1
),
W365
W373
(
2017
).
34.
J.
Lee
,
X.
Cheng
,
S.
Jo
,
A. D.
MacKerell
,
J. B.
Klauda
, and
W.
Im
, “
CHARMM-GUI input generator for NAMD, GROMACS, AMBER, OpenMM, and CHARMM/OpenMM simulations using the CHARMM36 additive force field
,”
Biophys. J.
110
(
3
),
641a
(
2016
).
35.
J. C.
Phillips
,
R.
Braun
,
W.
Wang
,
J.
Gumbart
,
E.
Tajkhorshid
,
E.
Villa
,
C.
Chipot
,
R. D.
Skeel
,
L.
Kale
, and
K.
Schulten
, “
Scalable molecular dynamics with NAMD
,”
J. Comput. Chem.
26
(
16
),
1781
1802
(
2005
).
36.
J.
Huang
and
A. D.
MacKerell
, Jr.
, “
CHARMM36 all-atom additive protein force field: Validation based on comparison to NMR data
,”
J. Comput. Chem.
34
(
25
),
2135
2145
(
2013
).
37.
U.
Essmann
,
L.
Perera
,
M. L.
Berkowitz
,
T.
Darden
,
H.
Lee
, and
L. G.
Pedersen
, “
A smooth particle mesh Ewald method
,”
J. Chem. Phys.
103
(
19
),
8577
8593
(
1995
).
38.
W.
Humphrey
,
A.
Dalke
, and
K.
Schulten
, “
VMD: Visual molecular dynamics
,”
J. Mol. Graphics
14
(
1
),
33
38
(
1996
).
39.
D.
Schneidman-Duhovny
,
Y.
Inbar
,
R.
Nussinov
, and
H. J.
Wolfson
, “
PatchDock and SymmDock: Servers for rigid and symmetric docking
,”
Nucleic Acids Res.
33
(
suppl_2
),
W363
W367
(
2005
).
40.
E.
Mashiach
,
D.
Schneidman-Duhovny
,
N.
Andrusier
,
R.
Nussinov
, and
H. J.
Wolfson
, “
FireDock: A web server for fast interaction refinement in molecular docking
,”
Nucleic Acids Res.
36
(
suppl_2
),
W229
W232
(
2008
).
41.
N.
Ferruz
,
S.
Schmidt
, and
B.
Höcker
, “
ProteinTools: A toolkit to analyze protein structures
,”
Nucleic Acids Res.
49
(
W1
),
W559
W566
(
2021
).
42.
J.-I.
Park
,
S. W.
Kim
,
J. P.
Lyons
,
H.
Ji
,
T. T.
Nguyen
,
K.
Cho
,
M. C.
Barton
,
T.
Deroo
,
K.
Vleminckx
, and
P. D.
McCrea
, “
Kaiso/p120-Catenin and TCF/beta-catenin complexes coordinately regulate canonical Wnt gene targets
,”
Dev. Cell
8
(
6
),
843
854
(
2005
).
43.
S.
Zhenilo
,
I.
Deyev
,
E.
Litvinova
,
N.
Zhigalova
,
D.
Kaplun
,
A.
Sokolov
,
A.
Mazur
, and
E.
Prokhortchouk
, “
DeSUMOylation switches Kaiso from activator to repressor upon hyperosmotic stress
,”
Cell Death Differ.
25
(
11
),
1938
1951
(
2018
).
44.
X.
Zou
,
W.
Ma
,
I. A.
Solov’yov
,
C.
Chipot
, and
K.
Schulten
, “
Recognition of methylated DNA through methyl-CpG binding domain proteins
,”
Nucleic Acids Res.
40
(
6
),
2747
2758
(
2012
).
45.
B.
Vallone
,
A. E.
Miele
,
P.
Vecchini
,
E.
Chiancone
, and
M.
Brunori
, “
Free energy of burying hydrophobic residues in the interface between protein subunits
,”
Proc. Natl. Acad. Sci. U. S. A.
95
(
11
),
6103
6107
(
1998
).
46.
E.
Wang
,
H.
Sun
,
J.
Wang
,
Z.
Wang
,
H.
Liu
,
J. Z.
Zhang
, and
T.
Hou
, “
End-point binding free energy calculation with MM/PBSA and MM/GBSA: Strategies and applications in drug design
,”
Chem. Rev.
119
(
16
),
9478
9508
(
2019
).
47.
L. S.
Dhami
,
P.
Dahal
,
B.
Thapa
,
N.
Gautam
,
N.
Pantha
,
R.
Adhikari
, and
N. P.
Adhikari
, “
Insights from in silico study of receptor energetics of SARS-CoV-2 variants
,”
Phys. Chem. Chem. Phys.
26
,
8794
(
2024
).
48.
C.
Mulakala
and
V. N.
Viswanadhan
, “
Could MM-GBSA be accurate enough for calculation of absolute protein/ligand binding free energies?
,”
J. Mol. Graphics Modell.
46
,
41
51
(
2013
).