Discussion (1zkd)

Discussion

The structural and functional analysis of 1zkd revealed a closer relation to methyltransferases. Proteins with similar structures are e.g. the human methyltransferase 2ex4 and the methyltransferase 1im8 from Haemophilus influenzae. S-adenosylhomocysteine was found in several crystallized proteins, indicating that these methyltransferases belong to the class which uses S-adenosylmethionine as a substrate, leaving s-adenosylhomocysteine as a product after methyl-group transfer to a target.

Under the top ten fold matches taken from the secondary structure matching (SSM) are seven mRNA-Cap Methyltransferases. The ligand binding template matches showed a relation to a protein-l-isoaspartate o-methyltransferase with a structural similarity of 80%.

Although most related proteins found in the analysis got only a sequence identity between 11% and 22,6% this must not be a decline for the significants of the results. S-adenosylmethionine dependent methyltransferases are reported to have a range of five different structural folds which are able to catalyse the transfer from S-adenosylmethionine to different targets and the sequence similarity even within on class of methyltransferases can be as low as 10 %. This shows that a low percentage of sequence similarity must not exclude a query sequence from being a methyltransferase. The structural requirements for catalysing a S-adenosylmethionine dependent methyltransfer appear therefore very flexible (Schubert et al, 2003).

S-Adenosylmethionine dependent methyltransferases are involved in several important functions in the cell like biosynthesis, signal transduction, protein repair, chromatin regulation or gene silencing (Schubert et al, 2003).

Though it remains unclear if our unknown protein and its human and mouse orthologs are modifying other proteins or nucleic acids as RNA. Nevertheless, it seems unlikely that this putative methyltransferase is modifying DNA in the nucleus because of the lack of nuclear localisation signals.

Since the evolutionary context of the study on our protein was structured towards the bacterial lineage that the bacteria Rhodopseudomonas palustris (where our protein sequence was found) comes from, some assumptions can be made.

Most of the bacteria thats are closly associated with our bacterial sequence seem to have a common function which is for nitrogen fixation. As mentioned in the results portion under Clustal X, Rhodopseudomonas palustris, Bradyrhizobium japonicum, Agrobacterium tumefaciens and Azospirillum brasilense all have nitrogen fixation abilities.

However what sets Rhodopseudomonas palustris apart is its ability to encode a vanadium-containing nitrogenase which is able to catalyst the production of about three times as much hydrogen as molybdenum-dependent nitrogenase.

This suggests a reason to further dwell into the genome of Rhodopseudomonas palustris even more as it ideally suited for use as a biocatalyst as it generates ample supplies of ATP from light thus catalyzing reactions that are thermodynamically unfavorable and beyond the potential of chemotrophic organisms. Furthermore, it has been noted that 35% of its genome is still rather unclear in terms of function.

The only logical interpretation that can be developed from the evolutionary tree is that our protein sequence most probably has an almost similar function to its closest protein sequence DUF185 - Rhodopseudomonas palustris CGA009.

The other closest bacteria that can be related to our sequence could be Bradyrhizobium japonicum. Further research into the evolutionary context of this organism could prove fruitful too.

Hence further studies are needed to confirm these results.

Return to Report on 1zkd