|Home | About | Journals | Submit | Contact Us | Français|
DNA mimicry is a direct and effective strategy by which the mimic competes with DNA for the DNA binding sites on other proteins. Until now, only about a dozen proteins have been shown to function via this strategy, including the DNA mimic protein DMP19 from Neisseria meningitides. We have shown previously that DMP19 dimer prevents the operator DNA from binding to the transcription factor NHTF. Here, we provide new evidence that DMP19 monomer can also interact with the Neisseria nucleoid-associated protein HU. Using BS3 crosslinking, gel filtration and isothermal titration calorimetry assays, we found that DMP19 uses its monomeric form to interact with the Neisseria HU dimer. Crosslinking conjugated mass spectrometry was used to investigate the binding mode of DMP19 monomer and HU dimer. Finally, an electrophoretic mobility shift assay (EMSA) confirmed that the DNA binding affinity of HU is affected by DMP19. These results showed that DMP19 is bifunctional in the gene regulation of Neisseria through its variable oligomeric forms.
Protein-DNA interactions are very important to the control of cellular mechanisms such as gene regulation, DNA repair and recombination . DNA mimic proteins are one of the emerging control factors that affect these interactions by occupying the DNA binding sites on the proteins [1–3]. Control factors of this category usually present with DNA phosphate backbone-like negative surface charge distributions and DNA double helical groove-like shapes to deceive the DNA binding proteins [1–3]. However, fewer than 20 DNA mimic proteins have been reported so far, and much more basic research is still needed to fully understand the novel mechanisms of these control factors .
In our previous attempts at discovering novel DNA mimic proteins [3–9], Neisseria meningitides DMP19 (DNA mimic protein 19 kDa) was one of the DNA mimic proteins that we identified . The crystal structure of the DMP19 dimer has a B-form DNA-like negative charge distribution on one side of its surface. This specialized pattern of negative charges is also complementary to the positive charge distribution of the DNA-binding region of the Neisseria transcription factor NHTF . Functionally, the DMP19 dimer prevents NHTF from binding to DNA by directly occupying its DNA-binding site . In the present study, we now show that DMP19 in its monomeric form can also interact with the Neisseria nucleoid-associated protein HU dimer. We first observed this interaction in a BS3 crosslinking assay, and the complex formed by the DMP19 monomer and HU dimer was then identified by gel filtration. The binding affinity between these two proteins was determined using isothermal titration calorimetry (ITC), and the calculated binding stoichiometry suggested that the ratio of DMP19 monomer to HU dimer was 1:1 in this complex. Cross-linking coupled mass spectrometry was then used to predict the interface of the binding. Finally, an electrophoretic mobility shift assay (EMSA) confirmed that the DNA binding affinity of HU was attenuated by DMP19. All these results suggest that DMP19 monomer is a competitive inhibitor of the Neisseria HU protein. We therefore conclude that DMP19 is a bi-functional DNA mimic protein that is involved in controlling nucleoid formation as well as gene regulation.
The purification of DMP19 and Neisseria HU was carried out as described previously [5–6]. Briefly, both proteins were expressed at 16°C for 16 hours after the addition of 1 mM isopropyl-β-D-thiogalactopyranoside (IPTG). Immobilized metal-ion chromatography (Ni-NTA) and gel filtration (Superdex 200 pg) were used to purify the soluble proteins. The proteinase FactorXa was used to remove the N-terminal His10-tag of DMP19.
For the cross-linking assay, C-terminal His6-tagged Neisseria HU and DMP19 were both dialyzed to 1X PBS buffer. Subsequently, 30 μM purified C-terminal His6-tagged Neisseria HU protein was mixed with 30 μM DMP19 in 100 μl binding buffer (1x PBS), and the reaction mixture was incubated for 15 minutes at 25°C. Following the addition of BS3 (Sigma-Aldrich) to a final concentration of 0.5 mM, incubation was then continued for another 30 min. Each reaction was stopped by 50 mM Tris-HCl. Finally, 6 μl of each reaction mixture were analyzed by SDS-PAGE.
A His-pulldown assay was used to confirm the interaction between DMP19 and Neisseria HU. Briefly, 30 μM C-terminal His6-tagged Neisseria HU and 30 μM DMP19 were mixed and incubated in 100 μl binding buffer (1x PBS; 20mM imidazole) for 15 minutes at 25°C. After adding 20 μl magnetic Ni-NTA beads (Genscript), the mixture was incubated for another 15 minutes. Subsequently, the beads were collected by a magnetic stand and washed with 1 ml binding buffer three times. Finally, the proteins that bound to the beads were eluted by elution buffer (20 mM Tris base pH8.0, 100 mM NaCl and 500 mM imidazole) and analyzed by SDS-PAGE.
A Superose 12 gel filtration column (GE Healthcare) was used to evaluate the approximate molecular weight of the DMP19 and HU complex. The buffer used in this assay was 20 mM Tris pH 7.4 and 100 mM NaCl. For the sample loading, either 100 μl of sample buffer containing 750 μM C-terminal His6-tagged HU or 1500 μM tag-free DMP19 were used, or both 750 μM C-terminal His6-tagged Neisseria HU and 1500 μM tag-free DMP19 together. Four standard proteins Conalbumin (C; 75 kDa) ovalbumin (OA; 43 kDa), carbonic anhydrase (CA; 29 kDa), and RNase A (RA; 13.7 kDa) were also analyzed by the same column to obtain a plot of Kav against log MW. For each protein, the logarithm of molecular weight (log MW) was plotted against Kav, which was calculated as follows: Kav = (Ve -Vo)/(Vt -Vo), where Ve is the elution volume, Vo is the column void volume using blue dextran 2000, and Vt is the total column bed volume (24 ml for the Superose 12 10/30 gel filtration column).
ITC200 (GE Healthcare) was used to measure the binding affinity of HU and DMP19. Before the titration, both HU and DMP19 were dialyzed against an ITC buffer of 20 mM Tris pH 7.4 and 50 mM NaCl. After the buffer change, the C-terminal His6-tagged Neisseria HU dimer (300 μM) and DMP19 monomer (30 μM) were loaded into the syringe and sample cell, respectively. Two μl of the HU was injected every 3 min until the DMP19 was saturated (a total of 20 injections) at 25°C. Except for the first injection point, data from the entire titration curve were used to calculate the thermodynamic parameters. The program ORIGIN 7 (GE Healthcare) was used to calculate enthalpy (ΔH), binding entropy (ΔS), the equilibrium constant (1/Kd) and the stoichiometric ratio (N). Gibbs free energy (ΔG) was calculated using the equation: ΔG = ΔH − TΔS.
The experimental procedure was modified from a crosslinking based LC/MS study . Briefly, the bands containing cosslinked DMP19 and HU proteins were excised from the SDS PAGE, followed by in-gel digestion with trypsin and chymotrypsin. Subsequently, the digested peptide mixtures were subjected to a NanoLC-nanoESI-MS/MS analysis using the standard protocol of the Academia Sinica Common Mass Spectrometry Facilities of Institute of Biological Chemistry [11–12]. Finally, all data was analyzed by using the Massmatrix software .
The crystal structure of the DMP19 monomer (PDB ID: 3WUR) and a protein model of the Neisseria HU dimer were used to make the DMP19-HU complex. The three-dimensional structure of Neisseria HU dimer was predicted using SWISS-MODEL Web server (http://swissmodel.expasy.org/) with a HU dimer of Anabaena HU-DNA cocrystal structure (PDB ID: 1P78) as the modelling template. The DNA binding region of Neisseria HU was also proposed by analogy to the Anabaena HU-DNA cocrystal structure (S1 Fig). Subsequently, the crystal structure of DMP19 monomer and the predicted model of Neisseria HU dimer were uploaded to the ZDOCK server (http://zdock.uma2ssmed.edu)  to generate potential DMP19-Neisseria HU complexes. A complex that satisfied the cross-linking distance from the output complexes was selected for further analysis. The graphic software Chimera was used to produce the figures in this study .
To observe the HU/DNA binding, 8 μM C-terminal His6-tagged Neisseria HU protein dimer (16 μM monomer) was mixed with 0.5 μM double-stranded DNA substrate (48 bp; 5’- CTC GCG AGC TCG CGA GCT CGC GAG CTC GCG AGC TCG CGA GCT CGT CGC GAG—3’) in 20 μl reaction buffer (20 mM Tris pH 7.4 and 100 mM NaCl) and incubated at 25°C for 15 min. For the competitive binding assays, 8 μM His6-tagged HU dimer was mixed with 8 or 16 μM DMP19 in 14 μl reaction buffer and pre-incubated at 25°C for 30 min. Subsequently, 1 μl of binding buffer containing the DNA substrate was added, and the incubation continued for 15 min. The reaction products were analyzed by a 4% agarose gel and stained with SYBR Green I (Sigma-Aldrich).
When we used BS3 crosslinking assays to search for new interactions between DNA mimic proteins and DNA binding proteins, we found that DMP19 could interact with Neisseria HU (Fig 1A). Shifted bands that indicated the cross-linking of proteins were observed only in the Neisseria HU/DMP19 reaction (Fig 1; arrows). An interaction between Neisseria HU and DMP19 was further confirmed by a His-pulldown assay (Fig 1B). We next used gel filtration to estimate the native molecular weights of DMP19, Neisseria HU, and the complex (Fig 2; Table 1). Consistent with our previous reports [5–6], the observed molecular weights of both the DMP19 and Neisseria HU proteins were close to the theoretical molecular weights of their dimeric forms in the gel filtration analysis (Table 1). However, the observed molecular weight of the DMP19/Neisseria HU complex was about 45.7 kDa, which is close to both of the possible monomer/dimer interactions but does not match either the monomer/monomer or dimer/dimer interaction (Fig 2; Table 1). This suggests the monomeric form of one of the two proteins interacts with the dimeric form of the other protein.
Isothermal titration calorimetry (ITC) was used to determine both binding affinity as well as stoichiometry of Neisseria HU and DMP19. When 300 μM HU dimer was titrated into 30 μM DMP19 monomer, a 1:1 binding ITC curve could be seen (stoichiometry; N = 1) (Fig 3; Table 2). This result was consistent with a Neisseria HU dimer/DMP19 monomer binding. The binding constant (Kd) is 0.50 ± 0.09 μM. Compared to the binding of Neisseria HU dimer to 8-mer DNA (Kd = 0.72 ± 0.06 μM, which was also determined by ITC; reference 6), DMP19 monomer has a relatively higher affinity to the HU dimer.
Cross-linking coupled mass spectrometry was used to study the binding mode between DMP19 monomer and HU dimer. The crosslinked DMP19/Neisseria HU proteins in the shifted bands (Fig 1; arrows) were extracted and analyzed by nanoLC-nanoESI-MS/MS. The results showed four crosslinked pairs between DMP19 lysine residues of DMP19 (Lys 91, Lys 95 and Lys 99) and HU (Lys 64 and Lys 78; Fig 4; Table 3). We next constructed a binding model of the DMP19 monomer and HU dimer based on these observations. Given that the end-to-end distance of the crosslinker BS3 is 11.4 Å, three lysine residues on DMP19 (Lys 91, Lys 95 and Lys 99) can be connected with the Lys 64 of one of the monomers in the the HU dimer, while DMP19 Lys 99 can also connect with the Lys 78 of the same HU monomer. By adjusting the orientations and positions of the DMP19 monomer and Neisseria HU dimer so that the distances between the connected Lys residues are all close to 11.4 Å, we constructed the three-dimensional model shown in Fig 5. In this model, the DMP19 monomer directly occupies the DNA-binding region of Neisseria HU dimer and thus prevents its binding to DNA.
The above results suggest that DMP19 monomer is likely to be a competitive inhibitor of Neisseria HU, and we next therefore used an electrophoresis mobility shift assay (EMSA) to investigate if DMP19 monomer could indeed prevent DNA from binding to Neisseria HU dimer. As shown in Fig 6, the DNA band shifting, which indicates the Neisseria HU-DNA binding, was significantly decreased by adding DMP19 monomer into the reaction. By contrast, neither of the negatively charged control proteins, BSA (Bovine serum albumin) and ICP11 (another DNA mimic that binds histone), caused any change in the EMSA band position of the HU-bound DNA. This result supports the above model in which the DMP19 monomer directly occupies the DNA-binding region of HU dimer, and it confirmed the strong interaction between the two proteins.
Neisseria DMP19 is one of the DNA mimic proteins that we recently identified . In the first report of Neisseria DMP19, this DNA mimic was proposed as a transcriptional regulator . The crystal structure showed that DMP19 dimer has a surface charge distribution that is complementary to the DNA binding surface of Neisseria NHTF transcription factor, and thus affects the binding of NHTF to its respective operator DNA . Functionally, NHTF belongs to the XRE (Xenobiotic Response Element) transcription factor family and acts as an auto-feedback repressor of the down-stream genes including NHTF itself and the Protein-PII uridylyltransferase gene glnD . The binding of DMP19 dimer to NTHF interferes with this negative gene control, and restores the expression of the affected genes .
In a subsequently determined DMP19 structure, a new conformation was revealed when crown ether was present in the crystal droplet . In this new DMP19 structure, helices α1 and α2 were rotated by ~180 degrees relative to the original structure (S2 Fig). Although two DMP19 monomers were found in an asymmetric unit of the crystal, two intercalating crown ether molecules prevented formation of the core DMP19 dimer interface (S2 Fig; right). Furthermore, the rotated helices α1 and α2 prevented these two monomers from coming close enough together to assemble into a dimer. These observations suggest that DMP19 is monomeric in this new crystal form. Interestingly, when we tried to dock DMP19 monomer and Neisseria HU dimer, helices α1 and α2 of the DMP19 monomer from the first crystal of dimeric DMP19 produced a steric clash with the HU protein, and only the DMP19 monomer from the new crystal could be successfully docked onto the HU dimer (Fig 5). All of these observations suggest that DMP19 has two conformations as well as two oligomeric states and that these are used to perform different functions. Thus in addition to binding to NHTF as a dimer , here we provide evidence that DMP19 has another functional role as a monomer, specifically, that it could act as a competitive regulator of Neisseria HU.
The highly-conserved bacterial nucleoid-associated HU proteins are involved in many important functions such as nucleoid compaction, gene regulation and DNA replication initiation . HU proteins usually function as a dimer. They induce DNA bending that results in a consistent negative supercoiling via non-sequence-specific DNA binding [18–19]. In addition to DNA, HU is also capable of binding to RNA as well as DNA-RNA hybrids . By binding to rpoS mRNA, Escherichia HU is able to regulate rpoS translation . Furthermore, it has recently been confirmed that HU can interact with noncoding RNAs in E. coli. [22–23], and that its homodimeric and heterodimeric forms (i.e. HUα/HUα and HUα/HUβ) control nucleoid compaction in E.coli . Although the nucleic acid binding affinity of HU protein has been investigated in detail, only a limited number of reports have studied the protein network between HU and its control proteins. Nevertheless, protein-protein interactions with HU are clearly important. For example, an interaction between DnaA and HU is critical for initiation at the replication origin of the E. coli chromosome .
In addition to DMP19, two other DNA mimic proteins, Neisseria DMP12 and Haemophilus influenzae HI1450, have also been shown to bind to HU proteins [6, 26–27]. All three of these DNA mimic proteins were found to bind to HU protein in their monomeric form, but since the amino acids and structures of DMP12, DMP19 and HI1450 are all different to each other (Fig 7A and 7B), it is surprising that these three DNA mimic proteins should be able to bind to the same target. However, we note that they all share a similar B form DNA-like negative charge distribution on their surface (Fig 7C), and these specialized negative charge distributions may play an important role in the recognition of HU proteins. The HU binding activities of these three DNA mimic proteins have all been confirmed by ITC assays [6, 27]. Compared to DMP12 and HI1450, DMP19 has the highest binding affinity to HU proteins (Kd = 2.8 and 0.5 μM for DMP12 and DMP19 monomer to Neisseria HU binding; 3.0 μM for H1450 to Haemophilus influenzae HU; all measured by ITC). Among these three DNA mimics, only DMP19 monomer has a higher binding affinity to Neisseria HU than 8bp dsDNA (Kd = 0.7 μM; reference 6), suggesting DMP19 monomer can serve as a potent inhibitor of Neisseria HU.
The next question to consider is how DMP19 might use its bi-functional activity to carry out transcriptional control. We hypothesize that DMP19 is an effective gene enhancer in Neisseria spps. By preventing HU from binding, DMP19 monomer may first release selected gene regions in chromosomal DNA. Subsequently, if DMP19 dimer disrupts the gene repressor activity of NHTF, this would ensure the continued expression of NHTF-controlled genes. Thus, for example, glnD would still be expressed in response to changes in the environment. However, this hypothesis still needs to be confirmed. Meanwhile, the results presented here provide strong evidence that DMP19 monomer is a competitive inhibitor of Neisseria HU, and that DMP19 therefore possesses a bi-functionality that depends on its different oligomeric states. This ability is unique among the few known DNA mimic proteins.
The Lys 78 residues of the Neisseria HU dimer are located at the center of the DNA binding region. Spatially, the Lys 64 residues are located at the ends of the flexible arms. All of these features are important for HU-DNA binding.
The conformational differences shown here may affect the oligomerization and thus the functionality of DMP19.
This work was supported financially by the Ministry of Science and Technology (Grant MOST 103-2311-B-038–005 and MOST 106-2311-B-038–002). LTQ-Orbitrap data and additional technical assistance were provided by Dr Shu-Yu Lin of the Academia Sinica Common Mass Spectrometry Facilities located at the Institute of Biological Chemistry. We also thank Dr Shu-Chuan Jao of the Biophysics Core Facility, Department of Academic Affairs and Instrument Service at Academia Sinica for her help in ITC analysis.
This work was supported financially by Ministry of Science and Technology (Grant MOST 103-2311-B-038 -005 and MOST 106-2311-B-038 -002). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
All relevant data are within the paper and its Supporting Information files.