An even more striking difference is
An even more striking difference is observed when comparing active site residues, which are evolutionarily very well conserved in other soluble PPases 18, 19. In 31 known PPase sequences, 13 functionally important active site residues are conserved in all sequences, but only two of them (D70 and D97 in E. coli PPase numbering) are conserved in B. subtilis PPase (Fig. 2).
In addition to soluble PPases, plants and certain bacteria have a membrane-bound PPase, which works as a reversible proton pump. Membrane-bound PPases differ in many respects from soluble PPases – they are much larger (660–770 amino what is c2 residues per monomer) and do not have any sequence similarity to soluble PPases 29, 30, 31. The B. subtilis PPase described here is clearly a soluble PPase, and it does not have any sequence similarity to membrane-bound PPases.
A search through the GenBank indicated the following four proteins having the highest identity (>30%) to the primary structure of B. subtilis PPase: `hypothetical exopolyphosphatases' of Streptococcus mutans, Methanococcus jannaschii, Archaeoglobus fulgidus and intrageneric coaggregation-relevant adhesin protein of Streptococcus gordonii. These four proteins have 307–311 amino acids/subunit, which is very close to the value of 309 for B. subtilis PPase. Identities of these four proteins are 57, 44, 40 and 57%, respectively, vs. B. subtilis PPase and only 13, 16, 18 and 15%, respectively, vs. E. coli exopolyphosphatase . This suggests that the four proteins mentioned above are in fact B. subtilis type PPases. This fits very well with the whole genome analysis showing that B. subtilis, M. jannaschii34, 35and A. fulgidusseem to have no typical ppa gene for the soluble PPase. On the other hand, B. subtilis PPase is clearly not an exopolyphosphatase, which is typically larger by size (396–513 amino acids/subunit) 32, 37and does not degrade PPi at all , or degrades it at a very low rate .
The 13 functionally important conserved active site residues of the known soluble PPases include five Asp, two Glu, two Lys, one Arg and two Tyr residues (Fig. 2). Sequence alignment of B. subtilis PPase and its four homologues, putative members of a new family of PPases, showed nine Asp, three Glu, four Lys and one Arg residues which are conserved in all these five proteins (Fig. 3) – these are obvious candidates for the active site residues, if the same types of residues are responsible for catalysis in both families of soluble PPases. Furthermore, there are several conserved polar residues of other types (two Asn, three Gln, four His, four Thr and five Ser), which may also have functional roles (Fig. 3). Expression and characterization of these four hypothetical PPases will eventually provide a test for these predictions. There are also several highly conserved regions, the most prominent of which is an eight residue long sequence near the C-terminus (294-SRKKQVVP-301). A search through the protein sequence database has indicated that only the five proteins shown in Fig. 3 have a perfect match with this octapeptide, suggesting this sequence to be a fingerprint for the B. subtilis type of PPases. It will be interesting to see what is the role of this highly conserved region in PPase structure and functioning.
In summary, B. subtilis has a unique type of soluble PPase, which is completely different from the closely related PPases studied so far 18, 19. Accordingly, B. subtilis PPase is a first example of a new family of soluble PPases, which we name Bs family. Putative representatives of the Bs family are also observed in four other bacterial strains, at least two of which, like B. subtilis, do not seem to have a typical ppa gene in their genomes.
Introduction Nucleotides function as regulatory molecules for various cellular metabolic pathways and as precursors for DNA and RNA synthesis. As reviewed in other sections of this issue , , nucleotides are at high risk for damage by reactive molecules, such as reactive oxygen species, which are generated by normal metabolism or by the exposure to ionizing radiation and chemicals. Some of the damaged nucleotides are thought to cause deleterious effects on human health by disturbing physiological and essential functions of canonical nucleotides in various metabolic pathways including DNA and RNA synthesis , . In these processes, damaged nucleotides may be recognized by certain cellular proteins. If these cellular proteins bind to the damaged nucleotides more stably than to canonical nucleotides, or if the damaged nucleotides exist in excess, functions of the proteins may be perturbed. On the other hand, organisms express enzymes that hydrolyze deleterious damaged nucleotides to eliminate them from the nucleotide pools , , . For example, MTH1, hydrolyzes oxidized purine nucleoside triphosphates, such as 8-oxo-dGTP, to the corresponding monophosphate forms, thus sanitizing the nucleotide pools .