Sequencing of the complete genome of Mycobacterium tuberculosis, combined with the rapidly increasing need to improve tuberculosis management through better drugs and vaccines, has initiated extensive research on several key proteins from the pathogen. RecA, an ubiquitous multifunctional protein, is a key component of the processes of homologous genetic recombination and DNA repair. Structural knowledge of MtRecA is imperative for a full understanding of both these activities and any ensuing application.  The crystal structure of MtRecA, presented here, has six molecules in the unit cell forming a 61 helical filament, with a deep groove capable of binding DNA.  The observed weakening in the higher order aggregation of filaments into bundles may have implications for recombination in mycobacteria.  The structure of the complex reveals the atomic interactions of ADP-AlF4, an ATP analogue with the P-loop-containing binding pocket. The structures explain reduced levels of interactions of MtRecA with ATP, despite sharing the same fold, topology and high sequence similarity with EcRecA.  The formation of a helical filament with a deep groove appears to be an inherent property of MtRecA. The histidine in loop L1 appears to be positioned appropriately for DNA interaction.

Click here to get co-ordinates of MtRecA apo structureand MtRecA-ADP-AlF4 complex
                                                    (PDB1G19.ENT)                                    (PDB1G18.ENT)

Space group P61 P61
Unit cell dimension a=108.1A,c=72.8A a=107.9A,c=72.0A
Resolution 3.0 3.8?
R-merge 10.7 14.5
Total number of unique reflections 9654 4738
Completeness Overall 98.4 97.8
I/Sig(I) Overall 17.2 14.4
Number of protein atoms 2276 2204
Number of ligand atoms 5 32
Number of solvent atoms 65 Null
R-factor 21.8%  22.0%
R-free 27.0% 28.0%

    Filaments and Bundles
 Click here to get co-ordinates of filaments
  MtRecA crystal (72 A)
  EcRecA crystal (82 A)
  MtRecA model (95 A)
  EcRecA model (95 A)
MtRecA aggregates into filaments which further aggregate into bundles. The picture here shows filaments of both MtRecA (first from left) and EcRecA (second from left) as seen in the crystal structure. The picture below shows what they look like when viewed through their helix axes. DNA is expected to bind through the central cavity. The active nucleoprotein filament however  is expected to have a longer helical pitch of 95A as against the 72 or 82 A pitch for MtRecA and EcRecA seen in the crystal structures. These 95 A filaments for both MtRecA and EcRecA have been modelled as shown here. The highly negative charge on the surface of MtRecA as compared to that of EcRecA can be clearlty seen from the picture.

The filaments so formed further aggregate into bundles.The bundles formed in MtRecA as seen in the crystal structure is shown here. Each filamnet is shown in a different colour and an orthogonal view is shown in the lower panel. They differ from those formed in EcRecA by having much fewer and weaker interactions between the filaments.

  Nucleotide binding and hydrolysis


Electron density for ADp-AlF4 as seen in the crystal structure of the complex is shown in the picture on the left.  The ATP binding region of MtRecA is widened compared to that in the prototype EcRecA,  there by explaining the suboptimal ATP binding and hydrolysis.  The picture above shows the surface representation of the three regions P, B and S of the binding site, responsible for phosphate, base and sugar binding respectively.


The crystal structure of MtRecA confirms that the protein in the crystals polymerizes into a filament, closely resembling that observed by electron microscopy for  EcRecA. Given its need to bind long stretches of both single-stranded and double-stranded DNA, the filament formation resulting in a deep groove capable of binding DNA, appears to be essential for its function. It is indeed fortunate that the protein crystallizes in the space group P61, compatible with the natural polymerization mode. The structure of this aggregated state, therefore, is far more informative than that of the individual molecule. The filamentous structures of RecA further aggregate into bundles, mainly as a form of storage and as a mechanism to keep RecA inactive when not needed, which may have implications for the efficiency of recombination and random integration. It is also known that unlike in yeast and many bacteria, mycobacteria promotes a higher degree of illegitimate recombination.  Modelling studies indicate the possibility of the formation of active filaments without any substantial change in molecular conformation.   The crystal structure of MtRecA complexed to ADP-AlF4 reveals a widening of the P-loop binding pocket resulting in weaker binding.  The structure  coupled with modelling studies shows how the protein can adapt itself to result in weaker nucleotide-binding and sub-optimal hydrolysis, despite all the residues in the binding site being invariant.  This is achieved through variations in tertiary interactions of the binding site residues with their neighbourhoods.  It is possible that such sub-optimal levels of binding and enzyme activity are a general feature in mycobacteria and may have some advantage for its survival in macrophages.  The crystal structure of MtRecA reported here allows its exploitation as a drug target, especially because it has a central role in several processes crucial for bacterial survival. Finally, the crystal structure throws additional light on the biology of recombination in mycobacteria, which is crucial to design attenuated strains of the tubercle bacilli, required to contrive an improved or robust vaccine.