SciELO - Scientific Electronic Library Online

vol.30 issue1  suppl.Differential metabolism of Mycoplasma species as revealed by their genomesABC transporters in Mycoplasma hyopneumoniae and Mycoplasma synoviae: insights into evolution and pathogenicity author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Genetics and Molecular Biology

Print version ISSN 1415-4757On-line version ISSN 1678-4685

Genet. Mol. Biol. vol.30 no.1 suppl.0 São Paulo  2007 



Purine and pyrimidine nucleotide metabolism in Mollicutes



Cristiano Valim Bizarro; Desirée Cigaran Schuck

Laboratório de Genômica Estrutural e Funcional, Centro de Biotecnologia, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil

Send correspondence to




Several mollicute genome projects are underway, offering unique opportunities to study genes and metabolic pathways on a genome-wide scale. Here, we have analyzed the conservation and diversity of purine and pyrimidine metabolism in mycoplasmas. An evaluation of discrepancies between genomic analysis and enzymatic data revealed interesting aspects about these organisms. We found important examples in which enzyme activity was reported without the annotation of a corresponding gene. An interesting example concerns phosphopentomutase. In Mollicutes, we have identified CDSs orthologous to sequences recently identified as new phosphopentomutases in archaeobacteria that are structurally related to phosphomannomutases. It is suggested that these sequences could replace the function of phosphopentomutases in mollicutes lacking the canonical phosphopentomutase gene (deoB). Also, the activity of 5'-nucleotidase was reported in mollicutes that do not possess any CDS related to ushA. Hypothetical proteins exhibiting domains similar to newly characterized 5' nucleotidases in Escherichia coli are proposed as possible CDSs related to this enzymatic activity in Mollicutes. Based on our analysis, the reductive genome evolution of Mollicutes does not appear to result in a minimum set of genes nor a minimum set of metabolic functions shared by all mollicute species.

Key words: mollicutes, purine, pyrimidine, metabolism, metabolic pathways.




Mollicutes are wall-less bacteria found widespread in nature. According to the International Committee on Systematics of Prokaryotes (ICSP), there are 204 described mollicute species, 119 of which belong to the genus Mycoplasma (ICSP Subcommittee on the Taxonomy of Mollicutes), which includes pathogens of medical and veterinary importance (Razin et al., 1998). These bacteria, together with Buchnera species (Gil et al., 2002) and the archaeobacterium Nanoarchaeum equitans (Waters et al., 2003), are among the smallest independently self-replicating cellular organisms, with reduced genome sizes and low GC content. Mycoplasmas present tissue and host-specificity, and are found as free-living or possibly intracellular parasites (Lo et al., 1993; Baseman et al., 1995; Dallo and Baseman, 2000; Momynaliev et al., 2000).

Mollicutes evolved from Gram-positive bacterial ancestors (Woese, 1987) through a genome reduction process, similar to what occurred in Wolbachia and Buchnera genera (van Ham et al., 2003; Wu et al., 2004). In recent years, we have witnessed the appearance of many mollicute genome projects. There are 14 completely sequenced genomes from 12 different mollicute species deposited in GenBank and another 13 genome sequencing projects from 11 new species in progress (NCBI Microbial Genomes). Thus, this group offers unique opportunities to study genes and metabolic pathways comparatively on a genome-wide scale.

In this work, we studied the purine and pyrimidine metabolism of these bacteria, which is an important topic of research in mollicute biochemistry. Nutritional studies have been hampered by the fastidious nature of these cells and the associated difficulty in developing defined media that enable optimal growth (Pollack, 2002). Nevertheless, defined growth media have been developed for both Mycoplasma mycoides subsp. mycoides (Rodwell, 1960) and Acholeplasma laidlawii (Rodwell and Mitchell, 1979), which became biochemical research models for Mollicutes (Pollack et al., 1997). These studies revealed the general metabolic capabilities of these organisms and some exciting new findings, such as PPi-dependent nucleoside kinase activities, never described previously in any living organism (Tryon and Pollack, 1984; Wang et al., 2001).

Much of this effort was made to try to define the minimum metabolic activities sufficient to support a living cell. More recently, this question was posed in a genomic context, as the minimal genome content necessary to support a living cell. Mushegian and Koonin carried out the first comparative analysis of Mycoplasma genitalium and Haemophilus influenzae (the only bacterial genomes available at that time) and tried to define the minimal gene set (Mushegian and Koonin, 1996). Later, minimal cell models were developed specifically for purine and pyrimidine tranpsort and metabolism (Castellanos et al., 2004), based on experimentally confirmed essential genes in M. genitalium and Mycoplasma pneumoniae (Hutchison et al., 1999) and comparative analysis. This kind of approach would benefit considerably from a more comprehensive view of genome and metabolic diversity within mycoplasmas. As we have previously shown (Vasconcelos et al., 2005), the genome reduction process that occurred during the evolution of mycoplasmas has led to the retention of alternate redundant biochemical pathways and not to a single minimal metabolism-related gene set, a finding with important implications for the minimal cell concept. Here, we focus on the conservation and diversity of purine and pyrimidine metabolism in mycoplasmas and we also outline some intriguing and currently unanswered questions about nucleotide metabolic pathways in Mollicutes.



Using the System for Automated Bacterial Integrated Annotation (SABIÁ) (Almeida et al., 2004), we retrieved clusters of orthologous sequences shared by genome-sequenced mollicutes containing CDSs related to purine and pyrimidine nucleotide metabolism. The annotation of each CDS obtained was confirmed by BLAST similarity searches (National Center for Biotechnology Information - NCBI). The individual confirmation of each sequence retrieved using the SABIÁ system allowed us to identify and discard sequences that clustered together by using the presence of domains unrelated to the enzyme considered. Sequences possessing an e-value greater than E-10 were maintained in the dataset only if a protein domain related to the particular enzyme considered was detected using the CD-search engine (Marchler-Bauer and Bryant, 2004). The set of enzyme activities analyzed in this study was constructed based on BRENDA enzyme database, International Union for Biochemistry and Molecular Biology (IUBMB) Nomenclature Committee Recommendations, the Kyoto Encyclopedia of Genes and Genomes (KEGG) (Kanehisa, 1997; Kanehisa and Goto, 2000) and on an extensive review of biochemical studies in Mollicutes. The lists of annotated CDSs for all sequenced mollicutes (NCBI Microbial Genomes) were inspected for annotated genes related to purine and pyrimidine nucleotide metabolism not included in our primary database. The annotation of each CDS identified by this strategy was confirmed by BLAST similarity searches as mentioned above. Finally, we reconfirmed our ensemble of CDSs retrieved using the Molligen database (Barre et al., 2004).

We systematically analyzed the biochemical studies in which enzyme activities related to purine or pyrimidine nucleotide metabolism were reported for the currently genome-sequenced mollicutes, which includes organisms from the Hominis group (Mycoplasma hyopneumoniae, Mycoplasma mobile, Mycoplasma pulmonis and Mycoplasma synoviae), Pneumoniae group (Mycoplasma gallisepticum, M. genitalium, M. pneumoniae, Mycoplasma penetrans and Ureaplasma urealyticum), M. mycoides subsp. mycoides, Mesoplasma florum, and Onion Yellow Phytoplasma. For each enzyme activity/species combination, a decision was made among the following options: activity detected, no activity detected, not studied, activity suggested, no activity suggested, and structure determined. Proteomic data were also included as a way to validate gene annotations. CDSs with confirmed expression by proteomic studies were assigned as expression was detected.

Discrepancies in which an enzyme activity was reported and no gene annotation was found were selected for detailed analysis. We screened mollicute genomes for domains related to the enzyme activity of interest using the Conserved Domain Architecture Retrieval Tool (CDART) (Geer et al., 2002), which finds protein similarities using sensitive protein domain profiles. Using the same approach, sequences included in the same COG or sharing a pfam entry in mollicute genomes were retrieved and analyzed. We also reviewed the literature to find cases of other organisms in which a protein belonging to a structurally novel class was involved in the same reaction pathway. When these cases were encountered, we searched for orthologous sequences in Mollicutes.

The combined and curated data were used to construct models for the purine and pyrimidine nucleotide pathways possibly present in each studied organism. These graphical representations of metabolic pathways were used to evaluate the differences and similarities presented by Mollicutes in the purine and pyrimidine nucleotide metabolism. These data were assembled to create a general representation of reaction pathways possibly present in at least one completely sequenced mollicute.


Results and Discussion

Nucleotide precursor's uptake

Most mollicutes are unable to synthesize de novo purine and pyrimidine bases (Mitchell and Finch, 1977). A possible exception is M. penetrans, which has an orotate-related pathway for converting carbamoyl-phosphate to UMP (Sasaki et al., 2002). This limited metabolic capability made mycoplasma cells dependent on environmentally-derived nucleotide precursors. However, there are fewer transporters in mycoplasmas than in most bacteria (Fraser et al., 2000). It was suggested that the reduction in the number of transporters has been compensated by the presence of transporters with broad substrate specificity (Saurin and Dassa, 1996). No nucleobase or nucleoside transporter was found in M. genitalium and M. pneumoniae genomes (Paulsen et al., 2000). It was suggested that transporters with a wide variety of substrates identified in both species could be involved in nucleic acid precursor import, including 11 ATP-binding cassette (ABC) and one Major Facilitator Superfamily (MFS) primary active transporter (Pollack, 2002).

External membrane-associated nuclease activities may be important for nucleotide precursor uptake, and were found in all 20 mycoplasma species tested (Minion et al., 1993). Later, the first membrane nuclease gene, mnuA, was cloned and isolated in M. pulmonis (Jarvill-Taylor et al., 1999). We found mnuA orthologous sequences in M. hyopneumoniae (232, J and 7448 strains), M. gallisepticum, M. pneumoniae, M. penetrans and U. urealyticum (Table S1 in supplementary online material - see Internet Resources). Besides a possible involvement in the pathogenic process (Bendjennat et al., 1999), the host DNA and RNA degradation mediated by mycoplasma membrane nucleases perhaps provided the bacterial cell with small oligonucleotides and free bases, suggested as important routes for nucleotide precursor uptake (Finch and Mitchell, 1992). Extracellular dephosphorylation of medium nucleotide monophosphates was perhaps achieved by ecto 5'nucleotidases, followed possibly by deribosylation of nucleosides.

Purine and pyrimidine nucleotide pathways in Mollicutes

Based on a previous nutritional study showing that guanine was the unique purine precursor required for M. mycoides subsp. mycoides growth (Rodwell, 1960) and on experiments of incorporation of labeled nucleotide precursors into RNA, Mitchell and Finch (1977) proposed pathways for purine nucleotide biosynthesis. Later studies have offered a more complete picture of purine nucleotide interconversions in Mollicutes (Pollack et al., 1997). A schematic diagram of proposed purine and pyrimidine pathways in Mollicutes is presented in Figures 1 and 2. These pathway representations include any activity related to purine or pyrimidine metabolism described in at least one genome-sequenced mollicute plus genome annotation data. In Tables S1 and S2 (supplementary online material - see Internet Resources), each reaction path is associated with gene annotation and references for enzyme activity or expression data from proteomic studies for all genome-sequenced mollicutes.

Nucleobases, ribo- and deoxyribonucleosides are imported by the bacterial cell, where purine and pyrimidine nucleosides are deribosylated to nucleobases by the purine nucleoside phosphorylase activity (PNP) or by the pyrimidine nucleoside phosphorylase (thymidine phosphorylase) activity, respectively. Deoxynucleosides can also be directly phosphorylated by deoxyribonucleoside kinases, generating the corresponding deoxyribonucleoside monophosphates (dNMPs). Cytoplasmic nucleobases can be converted to the corresponding nucleoside monophosphates by the activities of adenine phosphoribosyltransferase (APRT), hypoxanthine-guanine phosphoribosyl transferase (HGPRT) and uracil phosphoribosyltransferase (UPRT). Conversion of (d)NMPs to (d)NDPs is achieved by nucleoside monophosphate (NMP) kinases. The ultimate step, phosphorylation of (d)NDPs to (d)NTPs, constitutes a major gap in the understanding of mollicute nucleotide metabolism, as no nucleoside diphosphate kinase (ndk) gene was identified in any genome from Mollicutes. It is possible that, in the absence of a genuine NDP kinase activity, low substrate-specificity kinases could provide the necessary (d)NTPs required for normal growth and reproduction (Pollack et al., 2002). Interestingly, adenylate kinase was identified as the enzyme responsible for the NDP kinase activity detected in Escherichia coli ndk mutants (Lu and Inouye, 1996). Later, it was found that adenylate kinase from Mycobacterium tuberculosis also has NDP kinase activity (Meena et al., 2003). It would be interesting to functionally assay adenylate kinase from Mollicutes for NDP kinase activity.

The conversion of ribonucleotide precursors to the corresponding deoxyribonucleotides could be achieved by the activity of nucleoside diphosphate reductase (NDR). M. penetrans also contains a gene coding for a ribonucleotide-triphosphate reductase (RTR) and an RTR activating protein NrdG. Importantly, a novel route for deoxynucleotide synthesis was proposed based on enzymatic studies of purine and pyrimidine nucleoside phosphorylases, PNP and PyNP, respectively (McElwain and Pollack, 1987). PNP and PyNP can interconvert nucleosides and nucleobases. When converting nucleobases into nucleosides, these enzymes can accept a deoxyribose-1-phosphate instead of ribose-1-phosphate, generating deoxyribonucleosides. It was also found that mycoplasmas have the ability to phosphorylate all natural deoxynucleosides via deoxynucleoside kinases with ATP-dependent activities but also, in some species, NMP-dependent (phosphotransferases) and PPi-dependent activities (Wang et al., 2001).

Combining genomic and enzymatic data: A powerful tool for data mining in purine and pyrimidine nucleotide metabolism of Mollicutes

A more reliable model of the metabolic potential of an organism can be achieved by combining genomic analysis with enzymatic data. A systematic evaluation of discrepancies between the two datasets will reveal interesting aspects about these organisms and prompt further research. The differences found between enzymatic studies reported in the literature and gene annotation for the 12 mollicutes with sequenced genomes are found in Tables 1 and 2 for purine and pyrimidine nucleotide pathways, respectively. Tables S1 and S2 (supplementary online material - see Internet Resources) depict the complete datasets constructed in this study, integrating gene annotation, enzyme activity studies and proteomic analysis.

An interesting example concerns phosphopentomutase, an enzyme that interconverts deoxyribose-1-phosphate (dR1P) and deoxyribose-5-phosphate (dR5P) (Figures 1 and 2), and which is considered crucial to the development of an acceptable integrated scheme linking RNA, DNA, glycolysis and the pentose phosphate pathway (Pollack, 2001). A gene coding for this enzyme (deoB) was found in all sequenced mycoplasmas from Hominis group (M. hyopneumoniae, M. mobile, M. pulmonis and M. synoviae) but it was not found in any other mollicute (Table 3). However, phosphopentomutase activity was reported in both U. urealyticum and M. mycoides subsp. mycoides extracts (Cocks et al., 1985), which do not possess any gene related to deoB. Phosphopentomutases (COG1015) are included in a domain family together with the glycolytic enzyme 2,3-bisphosphoglycerate-independent phosphoglycerate mutase (pfam01676). We searched for pfam01676-containing CDSs in all mollicute genomes. However, all the sequences retrieved after a BLAST search using CDART (see Methods) were part of the already annotated deoB or pgm genes, coding for phosphopentomutase and phosphoglycerate mutase (COG0696), respectively.

Interestingly, a structurally novel phosphopentomutase was characterized in Thermococcus kodakaraensis, which is similar to phosphomannomutases within COG1109 (Rashid et al., 2004). We searched for COG1109-related sequences in genome-sequenced mollicutes and analyzed the genomic position of each retrieved sequence. The results are summarized in Table 3. Within the four sequenced mycoplasmas from Hominis group, all containing a deoB gene, M. hyopneumoniae and M. synoviae do not possess any COG1109-related sequence while M. pulmonis and M. mobile contain two and three genes, respectively. These sequences are not colinear with any gene related to the metabolism of (d)R1P or (d)R5P. Nevertheless, M. mycoides, M. florum and all the mycoplasmas from the Pneumoniae group, which do not contain the phosphopentomutase deoB gene, possess only one sequence related to COG1109. This sequence is located between deoA and deoC genes in both M. mycoides and M. florum, while in the Pneumoniae group it is found adjacent to the cdd gene. The structural organization of the region containing the COG1109-related sequence, cdd, deoA, deoC and deoD genes in Pneumoniae group suggest the presence of an operon. It should be noted that deoA, deoC and deoD gene products are metabolically linked to phosphopentomutase, as the corresponding reactions performed by thymidine phosphorylase (deoA), deoxyribose-5-phosphate aldolase (deoC) and purine nucleoside phosphorylase (deoD) involve (d)R1P or (d)R5P (Figures 1 and 2). We propose that sequences containing COG1109 in M. mycoides, M. florum and the Pneumoniae group code for phosphopentomutases and represent a new example of non-orthologous gene displacement (NOD). A functional analysis of the corresponding sequences will be fundamental to confirm this hypothesis.

Another important disagreement between activity studies and gene annotation found in this work involves 5'-nucleotidase. Despite its importance in nucleotide metabolism, the prokaryotic 5'-nucleotidases are poorly characterized. It is generally assumed that the ushA gene product is responsible for the 5'-nucleotidase activity found in bacteria, including the intracellular interconversion of all (d)NMPs to the corresponding (deoxy)nucleosides. However, E. coli UshA is a periplasmic protein that also possesses UDP-sugar hydrolase activity (Neu, 1967) whose major physiological role is the degradation of exogenous UDP-glucose and 5'-nucleotides for internal utilization of reaction products (Glaser et al., 1967). We found that the 5 genome-sequenced mollicute species for which a nucleotidase activity was reported do not possess any CDS related to ushA (Figure 1). Also, there are reports of nucleotidase activity detection towards some (d)NMPs but not others, indicating different substrate specificities (Tryon and Pollack, 1985).

Proudfoot et al. (2004) identified three uncharacterized E. coli proteins containing nucleotidase activity, SurE, YfbR, and YjjG, which exhibit different substrate specificities. These new nucleotidases belong to different enzyme superfamilies, possibly involving distinct catalytic mechanisms: SurE-like family (SurE), HD domain family (YfbR), and haloacid dehalogenase (HAD)-like superfamily (YjjG). We did not find orthologous sequences to these E.coli genes in Mollicutes. However, we found sequences containing HD domain or (HAD)-like domains (but not SurE-like) in different domain architectures (data not shown). These sequences could be considered as candidates for novel 5'-nucleotidases and could be experimentally evaluated in further efforts to characterize the proteins exhibiting nucleotidase activity in Mollicutes.

Uridine phosphorylase activity was detected in 5 genome-sequenced mycoplasmas, but no gene related to udp was annotated in any mollicute (Table 2). It is possible that thymidine phosphorylase (deoA) is responsible for the uridine phosphorylase activity detected, as it was found for Giardia lamblia, in which thymidine, uridine and deoxyuridine phosphorylase activities remained associated throughout the enzyme purification process (Lee et al., 1988).

Comparative analysis of purine and pyrimidine pathways in Mollicutes

For each genome-sequenced mollicute, we constructed schematic representations of purine and pyrimidine nucleotide pathways by integrating data from gene annotation and enzymatic studies (data not shown). In an effort to represent potential metabolic pathways presented by these mollicutes, an inclusive approach was adopted. A reaction was considered to be present if we detected an annotated gene coding for a protein orthologous to enzymes known to catalyze the corresponding reaction in other organisms, or if the reaction itself was experimentally validated in this organism. However, even if a particular reaction was not studied and no gene annotation was found, there still remains the possibility that the product of an uncharacterized CDS could possess the corresponding enzyme activity. As shown in Tables 1 and 2, we found important examples in which enzyme activity was reported without the annotation of a corresponding gene. These unknown enzymes may have orthologs in related mycoplasmas for which no enzyme activity study was carried out. This possibility was included in each pathway model as an undefined reaction.

It appears that M. hyopneumoniae, M. mycoides and U. urealyticum are not able to synthesize thymidylate derivatives using a thyA-coded thymidylate synthase. As we were not able to find any sequence related to the alternate ThyX thymidylate synthase (data not shown), the dTMP pools would then be dependent on direct uptake, as already characterized for M. mycoides (Neale et al., 1984), or by the concerted action of thymidine phosphorylase and thymidine kinase activities. In this scenario, phosphopentomutase activity would be crucial in the dTMP metabolism of these three species, a possibility that could be experimentally evaluated. Conversely, Onyon Yellow Phytoplasma seems to rely exclusively on thymidylate synthase activity to regulate the dTMP pool. There is no gene annotated for either thymidine phosphorylase or phosphopentomutase. Moreover, no COG1109-related sequence was found in this species, as was the case for the other mollicutes for which no phosphopentomutase gene was found (Table 3). A gene coding for thymidine kinase is present in Onyon Yellow Phytoplasma. However, its product is probably a bifunctional enzyme, exhibiting both thymidine and deoxyuridine kinase activities. The remaining mollicutes studied appear to possess both pathways for dTMP biosynthesis.

There are different pathways for interconversion of cytidine and uridine derivatives in Mollicutes in both pyrimidine ribonucleotide and deoxyribonucleotide metabolism. In fact, M. mycoides, M. florum and Onyon Yellow Phytoplasma possess a gene coding for cytosine deaminase, an enzyme that directly converts cytosine into uracil nucleobase. The comparative analysis of routes for both CTP and UTP nucleotide biosynthesis in Mollicutes reveal an intricate pattern of alternate retention of redundant metabolic pathways (Figure 3). UTP biosynthesis can proceed from uracil and PRPP by the activity of uracil phosphoribosyltransferase (UPRT), followed by nucleoside monophosphate kinase (NMK) and nucleoside diphosphate kinase (NDK) activities. CTP may be produced from UTP by the activity of CTP synthase. However, both pyrimidine nucleotides may be produced by alternate routes from the corresponding nucleosides using a bifunctional enzyme, cytidine/uridine kinase. Genome-sequenced mollicutes can be divided into 5 categories according to the presence of these three enzymes (Figure 3). M. gallisepticum, U. urealyticum, M. mycoides and M. florum belong to the first category, presumably possessing the three enzyme activities. These organisms exhibit redundant metabolic pathways for both CTP and UTP nucleotide biosynthesis. The phylogenetically related M. genitalium and M. pneumoniae possess only UPRT and cytidine/uridine kinase enzymes. These organisms retained alternate ways to produce UTP but rely only on cytidine kinase activity to produce CTP. However, interconversion of cytidine and uridine precursors is also occurring, as both mycoplasmas possess a cytidine deaminase gene. In the third category, Onion Yellow Phytoplasma has lost the UPRT gene, exhibiting only one route for UMP synthesis but retaining redundancy in the cytidine nucleotide pathway. Four species (M. penetrans, M. hyopneumoniae, M. penetrans and M. hominis) do not exhibit alternate ways to produce both UTP and CTP nucleotides. These mollicutes do not possess cytidine/uridine kinase but have UPRT and CTP synthase enzymes. M. synoviae was placed in an isolated category as this organism does not possess any gene related to cytidine/uridine kinase or CTP synthase. It is currently not known how M. synoviae could obtain cytidylates, except for a direct uptake from the environment.

Differences in the ability to interconvert guanine, inosine and adenine nucleotides are interesting to consider when devising defined media for Mollicutes. Based on nutritional studies using M. mycoides, it was proposed that adenine nucleotides could be formed from guanine nucleotides, but that GMP could not be formed from IMP (Mitchell and Finch, 1977). Taking into account these studies on M. mycoides, guanine nucleotide pathways were considered the "Achilles heel of the Mollicutes" because of the supposed dependence on phosphoribosyltransferase activity or transport of preformed guanine derivatives (Pollack, 2002). The genomic data from M. mycoides corroborated nutritional studies. This organism possesses a GMP reductase, which converts GMP into IMP, adenylosuccinate synthase and adenylosuccinate lyase, both involved in the formation of AMP from IMP. Moreover, no genomic sequence was found related to either GMP synthase or inosine 5' -monophosphate dehydrogenase, the nucleotide pathway from IMP to GMP through an XMP intermediate. However, a more comprehensive analysis of combined genomic and enzymatic data indicates that generalizations based on M. mycoides studies can be misleading when applied to other mollicutes.

From the 12 genome-sequenced mollicutes, no organism other than M. mycoides possesses the same set of enzymes involved in GMP to AMP conversion presented by M. mycoides. All the species from the Hominis group plus U. urealyticum, M. pneumoniae, M. genitalium and Onion Yellow Phytoplasma do not possess any enzyme involved in interconversion of GMP and IMP or conversion of the latter into AMP. A predictable consequence of this observation is that these organisms would not be able to grow in a medium containing only guanine as a preformed purine nucleobase, as is the case for M. mycoides (Rodwell, 1960). M. gallisepticum is presumably unable to interconvert GMP and IMP, but can generate AMP from IMP. Adenylosuccinate synthase and adenylosuccinate lyase activities were reported for M. gallisepticum (Tryon and Pollack, 1985) but no related gene was found. Further work will be required to validate these experimental data. Curiously, there is a GMP synthase annotated but we have not found any sequence related to IMP dehydrogenase in this organism. It is possible that the GMP synthase gene from M. gallisepticum represents a relict from an ancestral pathway no longer present in this species. M. penetrans and M. florum possess the complete set of enzymes related to interconvertion of GMP to IMP and convertion of IMP to AMP.

Our data is consistent with the view that a reduction in the metabolic repertoire accompanied the reductive evolution of mycoplasmas' genomes. However, in our view, this reduction process, which occurred in parallel in many lineages, does not seem to lead to an unequivocal minimum set of genes or even to a minimum set of metabolic functions. It is apparent from the examples described in this work that the metabolic repertoire displayed by M. genitalium, the smallest known mollicute, does not represent a minimum set of enzyme activities shared by the other mollicutes. Instead, different minimum sets of enzymatic functions could be generated through reductive evolution from an ancestral organism displaying redundant activities. We think it would be worth considering this hypothesis in further studies.


Concluding Remarks

In this work, we have systematically analyzed enzyme activity reports on purine and pyrimidine nucleotide metabolic pathways from the currently completely sequenced mollicutes. These data were combined with the corresponding gene annotation to generate a more complete description of nucleotide metabolism in Mollicutes. We found that members of a structurally novel family of phosphopentomutases recently characterized in the archaeobacterium Thermococcus kodakaraensis seem to be present in Mollicutes lacking the conventional deoB gene but displaying the corresponding enzyme activity.

The predicted purine and pyrimidine nucleotide pathways for each species were analyzed comparatively, and revealed the extent of conservation and diversity in the nucleotide metabolism of Mollicutes. Based on the comparative analysis, it is suggested that our ability to offer generalizations about mollicute biochemistry based on well-studied species, like M. mycoides and A. laidlawii, is rather limited. A complex pattern of redundancy and alternate retention of redundant pathways seems to emerge and the limitations of the minimum genome concept were discussed in this context.



C.V.B. is recipient of a CNPq DTI fellowship. D.C.S. is recipient of the Vallée S.A. fellowship. Supported by MCT/CNPq and FAPERGS.



Almeida LG, Paixão R, Souza RC, Costa GC, Barrientos FJ, Santos MT, Almeida DF and Vasconcelos AT (2004) A system for automated bacterial (genome) integrated annotation -SABIA. Bioinformatics 20:2832-2833.        [ Links ]

Barre A, de Daruvar A and Blanchard A (2004) MolliGen, a database dedicated to the comparative genomics of Mollicutes. Nucleic Acids Res 32:D307-D310.        [ Links ]

Baseman JB, Lange M, Criscimagna NL, Giron JA and Thomas CA (1995) Interplay between mycoplasmas and host target cells. Microb Pathog 19:105-116.        [ Links ]

Bendjennat M, Blanchard A, Loutfi M, Montagnier L and Bahraoul E (1999) Role of Mycoplasma penetrans endonuclease P40 as a potential pathogenic determinant. Infect Immun 67:4456-4462.        [ Links ]

Carnrot C, Wehelie R, Eriksson S, Bolske G and Wang L (2003) Molecular characterization of thymidine kinase from Ureaplasma urealyticum: Nucleoside analogues as potent inhibitors of mycoplasma growth. Mol Microbiol 50:771-780.        [ Links ]

Castellanos M, Wilson DB and Shuler ML (2004) A modular minimal cell model: Purine and pyrimidine transport and metabolism. Proc Natl Acad Sci USA 101:6681-6686.        [ Links ]

Cocks BG, Brake FA, Mitchell A and Finch LR (1985) Enzymes of intermediary carbohydrate metabolism in Ureaplasma urealyticum and Mycoplasma mycoides subsp. mycoides. J Gen Microbiol 131:2129-2135.        [ Links ]

Cocks BG, Youil R and Inch LR (1988) Comparison of enzymes of nucleotide metabolism in two members of the Mycoplasmataceae family. Int J Syst Bacteriol 38:273-278.        [ Links ]

Dallo SF and Baseman JB (2000) Intracellular DNA replication and long-term survival of pathogenic mycoplasmas. Microb Pathog 29:301-309.        [ Links ]

Davis JW, Nelson P and Ranglin R (1984) Enzyme activities contributing to hypoxanthine production in Ureaplasma. Isr J Med Sci 20:946-949.        [ Links ]

Finch IR and Mitchell A (1992) Sources of nucleotides. In: Maniloff I, McElhancy RN, Finch LR and Baseman JB (eds) Mycoplasmas: Molecular Biology and Pathogenesis. American Society for Microbiology, Washington, D.C., pp 211-230.        [ Links ]

Fraser CM, Eisen J, Fleischmann RD, Ketchum KA and Peterson S (2000) Comparative genomics and understanding of microbial biology. Emerg Infect Dis 6:505-512.        [ Links ]

Geer LY, Domrachev M, Lipman DJ and Bryant SH (2002) CDART: Protein homology by domain architecture. Genome Res 12:1619-1623.        [ Links ]

Gil R, Sabater-Munoz B, Latorre A, Silva FJ and Moya A (2002) Extreme genome reduction in Buchnera spp.: Toward the minimal genome needed for symbiotic life. Proc Natl Acad Sci USA 99:4454-4458.        [ Links ]

Glaser L, Melo A and Paul R (1967) Uridine diphosphate sugar hydrolase. Purification of enzyme and protein inhibitor. J Biol Chem 242:1944-1954.        [ Links ]

Hamet M, Bonissol C and Cartier P (1980) Enzymatic activities on purine and pyrimidine metabolism in nine mycoplasma species contaminating cell cultures. Clin Chim Acta 103:15-22.        [ Links ]

Hutchison CA, Peterson SN, Gill SR, Cline RT, White O, Fraser CM, Smith HO and Venter JC (1999) Global transposon mutagenesis and a minimal Mycoplasma genome. Science 286:2165-2169.        [ Links ]

Jarvill-Taylor KJ, VanDyk C and Minion FC (1999) Cloning of mnuA, a membrane nuclease gene of Mycoplasma pulmonis, and analysis of its expression in Escherichia coli. J Bacteriol 181:1853-1860.        [ Links ]

Johnson S and Pitcher D (2000) Distribution of ecto 5'-nucleotidase on Mycoplasma species associated with arthritis. FEMS Microbiol Lett 192:59-65.        [ Links ]

Kanehisa M (1997) A database for post-genome analysis. Trends Genet 13:375-376.        [ Links ]

Kanehisa M and Goto S (2000) KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res 28:27-30.        [ Links ]

Lee CS, Jimenez BM and O'Sullivan WJ (1988) Purification and characterization of uridine (thymidine) phosphorylase from Giardia lamblia. Mol Biochem Parasitol 30:271-277.        [ Links ]

Lo SC, Hayes MM, Kotani H, Pierce PF, Wear DJ, Newton PB 3rd, Tully JG and Shih JW (1993) Adhesion onto and invasion into mammalian cells by Mycoplasma penetrans: A newly isolated mycoplasma from patients with AIDS. Mol Pathol 6:276-280.        [ Links ]

Lu Q and Inouye M (1996) Adenylate kinase complements nucleoside diphosphate kinase deficiency in nucleotide metabolism. Proc Natl Acad Sci USA 93:5720-5725.        [ Links ]

Marchler-Bauer A and Bryant SH (2004) CD-Search: Protein domain annotations on the fly. Nucl Acids Res 32:W327-W331.        [ Links ]

McElwain MC and Pollack JD (1987) Synthesis of deoxyribomononucleotides in Mollicutes: Dependence on deoxyribose-1-phosphate and Ppi. J Bacteriol 169:3647-3653.        [ Links ]

McElwain MC, Chandler DFK, Barile MF, Young TF, Tryon VV, Davis Jr JW, Petzel JP, Chang CJ, Williams MV and Pollack JD (1988) Purine and pyrimidine metabolism in Mollicutes species. Int J Syst Bacteriol 38:417-423.        [ Links ]

McGarrity GJ, Gamon L, Steiner T, Tully J and Kotani H (1985) Uridine phosphorylase activity among the class Mollicutes. Curr Microbiol 12:107-112.        [ Links ]

Meena LS, Chopra P, Bedwal RS and Singh Y (2003) Nucleoside diphosphate kinase-like activity in adenylate kinase of Mycobacterium tuberculosis. Biotechnol Appl Biochem 38:169-174.        [ Links ]

Minion FC, Jarvill-Taylor KJ, Billings DE and Tigges E (1993) Membrane-associated nuclease activities in mycoplasmas. J Bacteriol 175:7842-7847.        [ Links ]

Mitchell A and Finch LR (1977) Pathways of nucleotide biosynthesis in Mycoplasma mycoides subsp. mycoides. J Bacteriol 130:1047-1054.        [ Links ]

Mitchell A and Finch LR (1979) Enzymes of pyrimidine metabolism in Mycoplasma mycoides subsp. mycoides. J Bacteriol 137:1073-1080.        [ Links ]

Mitchell A, Sin IL and Finch LR (1978) Enzymes of purine metabolism in Mycoplasma mycoides subsp. mycoides. J Bacteriol 134:706-707.        [ Links ]

Momynaliev KT, Orlova VF, Zhukotskii AV, Kogan EM and Govorun VM (2000) The use of in situ polymerase chain reaction for detection the intracellular localization of Mycoplasma hominis in cultured HeLa cells. Tsitologiia 42:202-208.        [ Links ]

Mushegian AR and Koonin EV (1996) A minimal gene set for cellular life derived by comparison of complete bacterial genomes. Proc Natl Acad Sci USA 93:10268-10273.        [ Links ]

Neale GA, Mitchell A and Finch LR (1983a) Pathways of pyrimidine deoxyribonucleotide biosynthesis in Mycoplasma mycoides subsp. mycoides. J Bacteriol 154:17-22.        [ Links ]

Neale GA, Mitchell A and Finch LR (1983b) Enzymes of pyrimidine deoxyribonucleotide metabolism in Mycoplasma mycoides subsp. mycoides. J Bacteriol 156:1001-1005.        [ Links ]

Neale GA, Mitchell A and Finch LR (1984) Uptake and utilization of deoxynucleoside 5'-monophosphates by Mycoplasma mycoides subsp. mycoides. J Bacteriol 158:943-947.        [ Links ]

Neu HC (1967) The 5'-nucleotidase of Escherichia coli. II. Surface localization and purification of the Escherichia coli 5'-nucleotidase inhibitor. J Biol Chem 242:3896-3904.        [ Links ]

Paulsen I, Nguyen L, Sliwinski MK, Rabus R and Saier M Jr. (2000) Microbial genome analyses: Comparative transport capabilities in eighteen prokaryotes. J Mol Biol 301:75-100.        [ Links ]

Pollack JD (2001) Ureaplasma urealyticum: An opportunity for combinatorial genomics. Trends Microbiol 9:169-175.        [ Links ]

Pollack JD (2002) The necessity of combining genomic and enzymatic data to infer metabolic function and pathways in the smallest bacteria: Amino acid, purine and pyrimidine metabolism in Mollicutes. Front Biosci 7:d1762-d1781.        [ Links ]

Pollack JD, Williams MV, Banzon J, Jones MA, Harvey L and Tully JG (1996) Comparative metabolism of Mesoplasma, Entomoplasma, Mycoplasma, and Acholeplasma. Int J Syst Bacteriol 46:885-890.        [ Links ]

Pollack JD, Williams MV and McElhaney RN (1997) The comparative metabolism of the Mollicutes (Mycoplasmas): The utility for taxonomic classification and the relationship of putative gene annotation and phylogeny to enzymatic function in the smallest free-living cells. Crit Rev Microbiol 23:269-354.        [ Links ]

Pollack JD, Myers MA, Dandekar T and Herrmann R (2002) Suspected utility of enzymes with multiple activities in the small genome Mycoplasma species: The replacement of the missing "household" nucleoside diphosphate kinase gene and activity by glycolytic kinases. OMICS 6:247-258.        [ Links ]

Proudfoot M, Kuznetsova E, Brown G, Rao NN, Kitagawa M, Mori H, Savchenko A and Yakunin AF (2004) General enzymatic screens identify three new nucleotidases in Escherichia coli. Biochemical characterization of SurE, YfbR, and YjjG. J Biol Chem 279:54687-54694.        [ Links ]

Rashid N, Imanaka H, Fukui T, Atomi H and Imanaka T (2004) Presence of a novel phosphopentomutase and a 2-deoxyribose 5-phosphate aldolase reveals a metabolic link between pentoses and central carbon metabolism in the hyperthermophilic archaeon Thermococcus kodakaraensis. J Bacteriol 186:4185-4191.        [ Links ]

Razin S, Yogev D and Naot Y (1998) Molecular biology and pathogenicity of mycoplasmas. Microbiol Mol Biol Rev 62:1094-1156.        [ Links ]

Rodwell AW (1960) Nutrition and metabolism of Mycoplasma mycoides var. mycoides. Ann NY Acad Sci 79:499-507.        [ Links ]

Rodwell AW and Mitchell A (1979) Nutrition, growth and reproduction. In: Barile MF and Razin S (eds) The Mycoplasmas. V. I. Academic Press, New York, pp 103-139.        [ Links ]

Sasaki Y, Ishikawa J, Yamashita A, Oshima K, Kenri T, Furuya K, Yoshino C, Horino A, Shiba T, Sasaki T and Hattori M (2002) The complete genomic sequence of Mycoplasma penetrans, an intracellular bacterial pathogen in humans. Nucleic Acids Res 30:5293-5300.        [ Links ]

Saurin W and Dassa E (1996) In the search of Mycoplasma genitalium lost substrate-binding proteins: Sequence divergence could be the result of a broader substrate specificity. Mol Microbiol 22:389-390.        [ Links ]

Tryon VV and Pollack D (1984) Purine metabolism in Acholeplasma laidlawii B: Novel PPi-dependent nucleoside kinase activity. J Bacteriol 159:265-270.        [ Links ]

Tryon VV and Pollack JD (1985) Distinctions in Mollicutes purine metabolism: Pyrophosphate-dependent nucleoside kinase and dependence on guanylate salvage. Int J Syst Bacteriol 35:497-501.        [ Links ]

van Ham RC, Kamerbeek J, Palacios C, Rausell C, Abascal F, Bastolla U, Fernandez JM, Jimenez L, Postigo M, Silva FJ, Tamames J, Viguera E, Latorre A, Valencia A, Moran F and Moya A (2003) Reductive genome evolution in Buchnera aphidicola. Proc Natl Acad Sci USA 100:581-586.        [ Links ]

Vasconcelos AT, Ferreira HB, Bizarro CV, Bonatto SL, Carvalho MO, Pinto PM, Almeida DF, Almeida LG, Almeida R, Alves-Filho L, et al. (2005) Swine and poultry pathogens: The complete genome sequences of two strains of Mycoplasma hyopneumoniae and a strain of Mycoplasma synoviae. J Bacteriol 187:5568-5577.        [ Links ]

Wang L, Westberg J, Bölske G and Eriksson S (2001) Novel deoxynucleoside-phosphorylating enzymes in mycoplasmas: Evidence for efficient utilization of deoxynucleosides. Mol Microbiol 42:1065-1073.        [ Links ]

Waters E, Hohn MJ, Ahel I, Graham DE, Adams MD, Barnstead M, Beeson KY, Bibbs L, Bolanos R, Keller M, Kretz K, Lin X, Mathur E, Ni J, Podar M, Richardson T, Sutton GG, Simon M, Soll D, Stetter KO, Short JM and Noordewier M (2003) The genome of Nanoarchaeum equitans: Insights into early archaeal evolution and derived parasitism. Proc Natl Acad Sci U S A 100:12984-12988.        [ Links ]

Williams MV and Pollack JD (1984) Purification and characterization of a dUTPase from Acholeplasma laidlawii B-PG9. J Bacteriol 159:278-282.        [ Links ]

Williams MV and Pollack JD (1985) Pyrimidine deoxynucleotide metabolism in members of the class Mollicutes. Int J Syst Bacteriol 35:227-230.        [ Links ]

Williams MV and Pollack JD (1990) The importance of differences in pyrimidine metabolism of the mollicutes. Zentralbl Bakteriol Suppl 20:163-171.        [ Links ]

Woese CR (1987) Bacterial evolution. Microbiol Rev 51:221-271.        [ Links ]

Wu M, Sun LV, Vamathevan J, Riegler M, Deboy R, Brownlie JC, McGraw EA, Martin W, Esser C, Ahmadinejad N, et al. (2004) Phylogenomics of the reproductive parasite Wolbachia pipientis wMel: A streamlined genome overrun by mobile genetic elements. PLoS Biol 2:E69.        [ Links ]


Internet Resources

BRaunschweig ENzyme DAtabase (BRENDA), (2/10/2006).

International Committee on Systematics of Prokaryotes (ICSP) Subcommittee on the Taxonomy of Mollicutes, (2/13/2006).

International Union for Biochemistry and Molecular Biology (IUBMB) Nomenclature Committee Recommendations, (2/12/2006)

National Center for Biotechnology Information (NCBI),

National Center for Biotechnology Information (NCBI) Microbial Genomes, (1/26/2006).


Supplementary Online Material

Tables S1 and S2: This material is part of the electronic version at: and in (



Send correspondence to
Cristiano Valim Bizarro
Laboratório de Genômica Estrutural e Funcional, Centro de Biotecnologia
Universidade Federal do Rio Grande do Sul
Caixa Postal 15005, 91501-970 Porto Alegre, RS, Brazil

Received: April 12, 2006; Accepted: October 16, 2006.



Associate Editor: Ana Tereza Vasconcelos

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License