Mycoplasmas are regarded to be useful models for studying the minimum genetic complement required for independent survival of an organism. Mycoplasma bovis is a globally distributed pathogen causing pneumonia, mastitis, arthritis, otitis media and reproductive tract disease, and genome sequences of three strains, the type strain PG45 and two strains isolated in China, have been published. In this study, several Tn4001 based transposon constructs were generated and used to create a M. bovis PG45 insertional mutant library. Direct genome sequencing of 319 independent insertions detected disruptions in 129 genes in M. bovis, 48 of which had homologues in Mycoplasma mycoides subspecies mycoides SC and 99 of which had homologues in Mycoplasma agalactiae. Sixteen genes found to be essential in previous studies on other mycoplasma species were found to be dispensable. Five of these genes have previously been predicted to be part of the core set of 153 essential genes in mycoplasmas. Thus this study has extended the list of non-essential genes of mycoplasmas from that previously generated by studies in other species.
Citation: Sharma S, Markham PF, Browning GF (2014) Genes Found Essential in Other Mycoplasmas Are Dispensable in Mycoplasma bovis. PLoS ONE 9(6): e97100. https://doi.org/10.1371/journal.pone.0097100
Editor: Valerie de Crécy-Lagard, University of Florida, United States of America
Received: January 29, 2014; Accepted: April 14, 2014; Published: June 4, 2014
Copyright: © 2014 Sharma et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: These studies were partially supported by financial support from the Australian Government under the AusAID program. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. No additional external funding was received for this study.
Competing interests: Glenn Browning is an Academic Editor for PLOS ONE. This does not alter the authors' adherence to PLOS ONE Editorial policies and criteria.
Mycoplasmas are a group of obligately parasitic bacteria that evolved from Gram positive organisms by reductive evolution. In the process, they have lost many dispensable genes and are thought to maintain only regulatory systems essential for their survival in vivo –.
The mycoplasmas lack a cell wall and have relatively small genomes (580 to 1380 kbp), but can still perform all the functions required for autonomous life , . Despite their genetic simplicity, many are pathogenic and can persist for very extended periods in their vertebrate hosts. Mycoplasma bovis, a significant pathogen of cattle throughout the world, lies in the hominis phylogenic group, with M. agalactiae, M. fermentans, M. synoviae, M. pulmonis, M. hyopneumoniae, M. arthritidis, M. hominis, M. conjunctivae, M. crocodyli, M. mobile and M. orale , .
The genomes of three strains of M. bovis, the type strain PG45  and two strains isolated in China, Hubei-1  and HB0801 , have been determined. There have been very few functional studies on M. bovis, and its virulence factors and the mechanisms involved in its pathogenicity are largely unknown. However, it is clear that it uses complex strategies to invade and avoid the immune response of the host , .
Only a few tools are available to genetically manipulate mycoplasmas. Transposons have been used to disrupt genes to study their role in virulence and their immunogenicity, to define the minimum genetic complement required for independent survival of an organism –, and as vectors for xenogeneic expression , . Only Tn916 and Tn4001, isolated from Enterococcus faecalis and Staphylococcus aureus, respectively, have been shown to function in mycoplasmas. Tn4001 is smaller (4.7 kbp) than Tn916 (18 kbp) and appears to have a better transformation efficiency , and plasmid pISM2062, carrying the transposon Tn4001 , has been used to introduce this transposon into M. bovis .
In the study described here, a library of M. bovis strain PG45 mutants was created by transformation with Tn4001-based plasmids. The locations of transposon insertions in the genome were identified by genomic sequencing and the catalogue of disruptable genes compared to those generated in other pathogenic mycoplasmas to identify those genes previously thought to be indispensible in mycoplasmas that are dispensable in M. bovis.
Functionality of transposon constructs for M. bovis strain PG45
The series of constructs based on Tn4001 were initially examined for their ability to transform M. gallisepticum strain S6, which was considered a model organism for transformation, as it had been transformed successfully in previous studies in our laboratory , , . Following success in transforming M. gallisepticum, pTn4001complete was used to transform M. bovis strain PG45. Subsequently, M. bovis was transformed with pTn4001single and then with the minitransposons containing either the gentamicin or tetracycline resistance genes. Individual colonies on selective agar plates were selected and cultured in appropriate selective broth and the cultures examined by PCR to confirm the presence of the gentamicin or tetracycline resistance genes.
Randomness of transposon integration
The randomness of transposon integration in the genome was confirmed by direct genomic sequencing of the mutant library (Figure 1), which allowed mapping of the transposon integration site for 319 mutants.
Haystack screening for a xer1 insertion
To identify mutants with a disruption in a specific gene target, transposon insertion sites were initially screened using a PCR-based strategy based on the haystack mutagenesis approach. For each of the four targeted loci, a first round of PCR was performed on each pool using one primer specific for the GOI and a second that would bind to the 5′ or 3′ end of Tn4001. An amplification product was identified in pool 5 using the oligonucleotide primer pair GKxer1 for and IR inverse (Table S1), indicative of a xer1 disruption. The pool contained 29 individual mutants. The second round of PCR was performed on DNA from each the 29 mutants within the pool individually and mutant number 29, which had the xer1 gene disrupted by Tn4001complete, was identified. The PCR yielded an amplification product of around 350 bp, suggesting that the site of insertion of the transposon was expected to be around 350 bp downstream of the start codon of the gene. This was confirmed by cloning the PCR product in pGEM-T and sequencing the insert. Haystack screening did not detect disruptions in p48, oppD or the restriction endonuclease gene, and the absence of these mutations from the library was confirmed by direct genome sequencing.
Non-essential genes in M. bovis
After initial studies using haystack mutagenesis, we used direct sequencing to identify the insertion sites in all the mutants in the library. Of the 319 mutants, 151 were generated using pTn4001single, 125 using pTn4001complete, 40 using pMiniTn4001-gent and 3 using pMiniTn4001-tet. A total of 191 insertions were in annotated ORFs, 38 within predicted intergenic regions, 40 within ICE elements and 50 within transposase genes. Of the 191 insertions in ORFs, 113 were in predicted genes, 56 in genes encoding membrane proteins or lipoproteins and 22 in genes encoding hypothetical proteins. Based upon the criteria for gene disruption, 129 genes had been disrupted, and of these 48 and 99 genes had homologues in M. mycoides subspecies mycoides SC strain PG1 and M. agalactiae strain PG2, respectively (Table 1). There were 21 additional genes that had transposon insertions within the last 15% of the coding sequence and which were therefore not considered to be disrupted, although this may not have been the case if function was located in this region of the protein (Table S2). Several genes were disrupted in multiple mutants. Intergenic regions contain promoters for genes located downstream, so transposon insertions in intergenic regions may have impaired the function of downstream genes or operons, so while these insertion events were not considered gene disruptions, the mutants carrying them (Table S3) may also be important in assessment of gene function. In addition, a total of 90 insertions were observed within integrative conjugative elements (ICE) (Table S4) and transposase genes (Table S5).
A number of notable genes were disrupted, including those annotated as encoding the heat shock proteins ClpB (MBOVPG45_0720) and DnaJ (MBOVPG45_0839), all the genes in the putative nucleotide transporter operon (MBOVPG45_307 to MBOVPG45_311), one gene in the polyamine ABC transporter system operon (MBOVPG45_0135), two genes in the glycerol ABC transporter system operon (MBOVPG45_0748 & MBOVPG45_0749), and in the genes encoding the glycerol kinase (MBOVPG45_0529) and the glycerol uptake facilitator protein (MBOVPG45_0530).
Fewer essential genes in mycoplasmas than in previous studies
In an early study employing transposon mutagenesis, 310 genes were reported to be essential in M. pulmonis . A further study on M. pulmonis found an additional 39 of these 310 genes to be dispensable , and it has been suggested that there are 153 core essential genes in Mycoplasma species . In the study described here on M. bovis, 23 genes considered to be essential in M. pulmonis in the initial study , 16 of which were still found to be essential in the subsequent study , were disrupted (Table 2). Five of these genes, encoding the tRNA modification GTPase TrmE (MBOVPG45_0060), the polyamine ABC transporter permease PotB (MBOVPG45_0135), the methionine adenosyltransferase MetK (MBOVPG45_0227), the chaperone protein DnaJ (MBOVPG45_0839) and the ssrA binding protein SmpB (MBOVPG45_0855), were considered essential in all previous gene essentiality studies in mycoplasmas –,  and have been predicted to form the core set of 153 essential genes in mycoplasmas . Thus our study has demonstrated that mycoplasmas have fewer core essential genes than predicted previously.
Among the other genes considered essential in earlier studies –,  that were disrupted in our library were those coding for the 50S ribosomal protein L34 (RpmH, MBOVPG45_0065), the thiamine biosynthesis protein (ThiI, MBOVPG45_0140), oligosaccharide ABC transporter proteins (MBOVPG45_0307 & 0309), a glycosyltransferase (MBOVPG45_0390), the potassium transporter (KtrA, MBOVPG45_0464), a serine/threonine protein kinase (PknB, MBOVPG45_0629) and a tRNA binding domain containing protein (MBOVPG45_0845). Other genes disrupted in our study that were considered essential in the initial study on M. pulmonis , but that were found to be dispensable in a subsequent study , were those encoding deoxyribose-phosphate aldolase (DeoC, MBOVPG45_0300), the membrane lipoprotein P81 (Mb-mp81, MBOVPG45_0311), deoxyribonuclease IV (Nfo, MBOVPG45_0317), the chromosome segregation protein (Smc, MBOVPG45_0520), a hypothetical protein (MBOVPG45_0534), the drug resistance ABC transporter ATP-binding protein (MBOVPG45_0695), the phosphoglucomutase/phosphomannomutase domain-containing protein (MBOVPG45_0728) and the HemK methyltransferase (MBOVPG45_0849). Another two genes, encoding a phosphate acetyltransferase (MBOVPG45_0153) and a DHH family protein (MBOVPG45_0527), which were reported to be essential in earlier studies, were disrupted in the M. bovis library, but these genes have paralogues in the M. bovis genome and therefore could not be considered to be dispensable based on our study.
Although the genomes of the type strains of M. bovis, M. agalactiae and M. mycoides subspecies mycoides SC, all of which cause disease in ruminants, have been sequenced , gene essentiality data are not available for these species. There has been extensive horizontal gene transfer between these species, with many genes in M. bovis and M. agalactiae probably acquired from the phylogenetically distant M. mycoides cluster ,  during co-infection of the same host . Therefore, genes found to be non-essential in M. bovis are likely to also be non-essential in the other two species. Of the genes disrupted in the M. bovis mutant library, 48 had orthologues in M. mycoides subspecies mycoides SC and 99 had orthologues in M. agalactiae. Six of the 23 essential mycoplasma genes that were found to have transposon insertions in our study have essential orthologues in B. subtilis , .
In our study, there were insertions in 191 predicted ORFs. In earlier studies in M. genitalium 382 genes were found to be indispensable in M. genitalium , while 310 genes were found to be essential in M. pulmonis . A further study in M. pulmonis  found 39 additional genes to be dispensable. Comparison of the data from our study with that obtained for M. pulmonis is of interest as both species have similar genome sizes and lie within the same (hominis) phylogenic group. We found 23 of the 310 genes found to be essential in the initial study on M. pulmonis  were disruptable in M. bovis (Table 2), with 7 of these 23 among those found to be disruptable in the later study on M. pulmonis.
The M. bovis genome has 52 ABC transporter genes, in 14 operons, and nine of these transporter gene ORFs, in four operons, were able to be disrupted (Table 3). Acquisition of nutrients by mycoplasmas appears to predominantly involve ABC transporters, and the low level of redundancy in mycoplasmas suggests that they are likely to be required for nutrient acquisition in vivo, but clearly some are dispensable in the complex media used for culture in vitro.
The dispensability of two genes belonging to the glycerol transport system, along with glycerol kinase and the glycerol uptake facilitator protein, is not surprising considering that there are two predicted glycerol transport systems in M. bovis that may complement each other. Earlier studies have reported that the production of hydrogen peroxide by ruminant mycoplasmas involves the glycerol transport system , , and the transport system appears to be dispensable in M. mycoides subspecies mycoides SC, as European strains, which are less virulent compared to their African counterparts, lack both the gtsB and gtsC glycerol transport genes .
The possibility of gene disruptions in some genes of M. bovis that are essential in other mycoplasmas might be expected, as there are paralogues in M. bovis of the genes encoding phosphate acetyltransferase and the DHH family protein. Similarly, some variations might be expected between different species because of unrecognized redundancy. The serine/threonine protein kinase gene (pknB), which was disrupted in the M. bovis library, has been reported to be essential for growth of M. pulmonis and M. genitalium –, however pknB has been disrupted in M. agalactiae .
The genes MBOVPG45_0307 to MBOVPG45_0311, which are part of a putative nucleotide transporter operon –, were disrupted in our study, and transposon insertions in mslA, the MBOVPG45_0311 homologue in M. gallisepticum, have been reported previously . It has been demonstrated recently that mslA of M. gallisepticum, the MBOVPG45_0311 homologue, binds single and double stranded DNA , suggesting that the mslA may bind and deliver oligonucleotides to the exonuclease, which then processes these oligonucleotides to generate individual nucleotides for transport into the cell via the ABC transporter. The disruption of genes encoding this operon might be tolerated because there are three putative membrane nucleases in the genome of M. bovis strain PG45, MBOVPG45_0089, MBOVPG45_0215 and MBOVPG45_0310.
The dispensability of five genes disrupted in our study, trmE, potB, metK, dnaJ and smpB, which were found to be essential in all previous gene essentiality studies in mycoplasmas, and which were among the predicted set of 153 core mycoplasma genes, could not be explained by predicted redundancy. Although the polyamine transporter system is dispensable in B. subtilis, this may result from complementation by another transport system. It is possible that this may also be the case in M. bovis. The chaperone DnaJ has long been considered to be essential for cellular growth. However, as expression of DnaJ increases in response to cellular stress , it may be dispensable during the optimal growth conditions used for culture in vitro. There are no obvious explanations for the dispensability of rpmH, thiI and ktrA, nor for the genes encoding the glycosyltransferase and the tRNA binding domain-containing protein.
However, it has been pointed out that minimal or core sets of genes are context dependent and it has been suggested that gene persistence is a better indication of the role of specific genes in the long term survival of an organism  and that, in defining the minimal requirements for cellular life, it would be more useful to consider those genes that, while not ubiquitous, were conserved in most genomes. Therefore we have assessed which of the genes that we found to be dispensable (Table 2) are found in most mycoplasma genomes . We have also compared the gene dispensability determined in our study with the persistence and essentiality of orthologues in B. subtilis and E. coli .
The dispensability of rpmH is surprising, as it is conserved in all the fully sequenced mycoplasma genomes , and not only essential in M. pulmonis and M. genitalium, but also in B. subtilis and E. coli . Similarly smc is conserved in all the mycoplasma genomes, as well as in B. subtilis. In recent studies, the rpmH and smc genes have been reported to be borderline persistent , and smc could be disrupted in M. pulmonis  and rpmH in B. subtilis, although the growth of the mutant was affected . The pknB and thiI genes are not highly conserved in the mollicutes, with pknB absent in M. hyorhinis, M. hyopneumoniae, M. conjunctivae and Acholeplasma laidlawii, and thiI not found in M. hyorhinis, M. hyopneumoniae, M. conjunctivae or Ureaplasma urealyticum, its absence being correlated with a mutation in tRNAIle. The gene hemK, which is predicted to code for a methyltransferase, is absent in M. conjunctivae, U. parvum and B. subtilis, while metK, which codes for methionine adenosyltransferase, is conserved in all Mycoplasma species, B. subtilis and E. coli, but is not annotated in U. urealyticum.
Several potential problems with transposon-generated mutant libraries in mycoplasmas were not seen or were addressed by use of differing techniques in our study. In an earlier study  16–86% of colonies growing on selective agar plates lacked a transposon insertion. In this earlier study, it was assumed that these resulted from acquisition of spontaneous resistance, but attempts to decrease the prevalence of pseudotransformants by increasing the concentration of antibiotic in selective agar failed. The problem was overcome in this earlier study by incubation of M. bovis in selective broth for an extended period after transformation, but this may also result in multiplication of mutants and thus increase the prevalence of replicate clones in the final library. However we did not detect any pseudotransformants following transformation with any of our transposon constructs.
Replicative transposition, resulting in multiple insertions in the genome, have been a problem in some studies. We developed several derivatives of Tn4001, including Tn4001single, which lacked one of the IS256 arms, and minitransposons, with the transposase outside the transposon, with the aim of creating transposons that would be incapable of secondary transposition and that would thus generate mutants that could be expected to be genetically stable . That this was desirable was demonstrated by the relatively high frequency of multiple insertion events we saw in mutants created using Tn4001 (data not shown).
The potential presence of insertional hotspots has also been raised as a concern in the use of transposons to generate mutant libraries. The randomness of insertion of Tn4001 and its derivatives was confirmed by genomic sequencing of 319 individual mutants, which demonstrated that insertion events were distributed throughout the genome (Figure 1).
Targeted gene knockout remains a challenge in mycoplasmas. Targeted gene disruption in mycoplasmas has occasionally been achieved through homologous recombination, either employing free DNA or replicable oriC plasmids , , , but the low rate of recombination has necessitated extensive passage to increase the likelihood of acquiring the desired knockout. In many cases recombination with oriC plasimds occurs within the oriC region, or in illegitimate sites, rather than in the desired targets, and if it does occur within the targeted gene it can be difficult to isolate the recombinant clone . Transposon mutagenesis has been the genetic tool most commonly used to manipulate mycoplasmas because of its much greater efficiency, but there have been only limited attempts to identify mutants in libraries with specific phenotypic changes that might be attributable to disruption of specific genes. Mutant libraries have been screened for loss of reactivity with a specific antiserum against LppQ in M. mycoides subspecies mycoides SC , loss of gliding motility in M. pneumoniae  or loss of capacity for growth on cell cultures , . In the absence of a selectable phenotypic trait and to avoid time consuming direct genomic sequencing of all individual mutants, the PCR based haystack mutagenesis approach ,  can be used to identify specific gene knockouts. However, the approach may not be suitable for identification of gene disruptions in large coding regions, and particularly if it occurs in middle of coding regions. In our study the haystack mutagenesis approach was used to identify a xer1 gene disruption. In earlier haystack mutagenesis studies , the transformants were grown in broth as a pool before DNA extraction. This may result in overgrowth of mutants with disruptions in genes not required for optimal growth. Therefore, we picked individual mutants, generated an ordered mutant library, and cultured the mutants to late log phase before creating a series of pools for screening. Instead of using two primer pairs in the Tn4001 region , , a single oligonucleotide primer binding to the IR region of Tn4001 was used, as it could be combined with either a forward or reverse primer flanking the gene of interest to yield a single PCR product in the event of insertion in the desired gene.
Although genome sequences are available for more than 1000 bacterial species, genome-wide essentiality data is available for only 15 species, including three Mycoplasma species, M. genitalium, M. pneumoniae and M. pulmonis –, . A set of 153 core essential mycoplasma genes have been predicted . Some genes expected to be essential were identified as disrupted in an early study , possibly because mutants were not characterised as clonal cultures, but rather as members of a mixed pool, and some genes that were predicted to be non-essential in this initial study appeared to be essential in later studies .
Although the mutant library we have characterised here could not be expected to have included a comprehensive repertoire of mutatable genes as the genome was not saturated with insertions, the lack of insertions in several large genes and transport systems suggests the importance of these genes for optimal growth of M. bovis in vitro. These include two predicted oligopeptide ABC transporter system operons, a predicted carbohydrate uptake ABC transporter system operon and a predicted cobalt ABC transporter system operon. No gene coding for tRNAs or rRNAs, which are considered essential for cell replication, was disrupted. In addition, some large genes that encode membrane proteins or hypothetical proteins were not disrupted, including MBOVPG45_0337 (3419 bp), MBOVPG45_0481 (4547 bp) and MBOVPG45_0710 (8012 bp), and thus these genes may have a role in optimal growth of M. bovis in vitro and may be worthy of further investigation.
One of the largest membrane proteins in M. bovis, MBOVPG45_0710, which is over 8 kbp in length (2670 amino acids) and has full or partial homologues in M. agalactiae MAG6100, M. fermentans MFE_02570, M. crocodyli MCRO_0279, M. synoviae MS53_0328, M. pulmonis MYPU_3130, M. conjunctivae MCJ_003940, M. mobile MMOB4250, M. hyorhinis MYM_0289 and M. hyopneumoniae mhp677 , was not disrupted. Homologues of MBOVPG45_0710 in M. fermentans and M. mobile are predicted to possess lipase activity, and the regions between amino acid residues 90 and 395 of MBOVPG45_0710 had 31% identity to M. hyopneumoniae p65, which has been demonstrated to possess lipase activity . Although the conserved domain is restricted to the amino terminal end of this protein, the large size and lack of disruptions within the gene suggest essentiality of this protein. It may consist of several conserved domains, the functions of which are specific for species related closely to M. bovis.
Thus this study has validated the use of haystack mutagenesis to identify mutants with specific genes disrupted in an ordered mutant library, and has characterised the location of more than 300 transposon insertions in the M. bovis genome, establishing the dispensability of at least 16 genes previously believed to be essential in mycoplasmas. These data will aid in furthering our understanding of the functions of genes and gene products of mycoplasmas.
Bacterial strains and culture conditions
M. bovis type strain PG45 (ATCC 25523) was cultured at 37°C in modified Frey's broth (21 g PPLO, 37 ml yeast extract, 100 ml inactivated swine serum, 4 ml 1.6% phenol red solution, 300 mg penicillin G, 859 ml distilled water, pH adjusted to 7.8) or on mycoplasma agar plates (modified Frey's broth without phenol red with 1% agar added). For the selection of M. bovis transformants, gentamicin (Invitrogen) or tetracycline (Sigma Aldrich) was added to media to a concentration of 50 µg/ml or 5 µg/ml, respectively.
Escherichia coli DH5α cells (Life Technologies) were used for cloning of different transposon constructs and were cultured at 37°C in Luria-Bertani (LB) broth (1% w/v tryptone (Oxoid), 0.5% w/v yeast extract (Oxoid), 0.5% w/v NaCl) with shaking at 200 rpm on an orbital shaker incubator (Ratek) or on LB agar plates (LB broth containing 1% bacteriological agar). Selection of plasmid-transformed E. coli DH5α cells was performed on LB agar containing 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-gal) (Sigma) at 40 µg/ml, isopropyl-β-D-thiogalactopyranoside (IPTG) (Sigma) at 50 µg/ml and an appropriate antibiotic. E. coli DH5α containing plasmid constructs were grown in LB broth or on LB agar plates containing ampicillin (Amresco) at 100 µg/ml, gentamicin at 20 µg/ml or tetracycline at 4 µg/ml.
Agarose gel electrophoresis and plasmid extraction
Polymerase chain reaction (PCR) products and plasmid DNA constructs were analysed using conventional agarose gel electrophoresis in 0.8–2.0% w/v agarose (Scientifix) gels in 1× TBE buffer (89 mM Tris, 89 mM boric acid, 2 mM EDTA, pH 8.0) or 0.5× TPE buffer (1× TPE is 36 mM Tris, 30 mM NaH2PO4, 1 mM EDTA) and stained with ethidium bromide at 0.1 µg/ml. DNA bands were visualised using an ultraviolet transilluminator (Gibco BRL) and imaged using either the Digital Science electrophoresis documentation and analysis system (Kodak) or the Molecular Imager ChemiDoc XRS+ imaging system (Bio-Rad).
PCR products and restriction endonuclease digestion products of plasmids were separated by agarose gel electrophoresis and the DNA in specific bands extracted using the Ultraclean gel spin DNA purification kit (Mo Bio Laboratories) according to the manufacturer's instructions. The Wizard Plus SV Minipreps DNA purification system (Promega) was used to extract up to 2 µg of plasmid DNA from E. coli DH5α cells, whilst for purification of 20 µg or more of plasmid DNA the Qiagen Plasmid Midi kit (Qiagen) was used according to the manufacturer's guidelines.
Amplification of PCR products
The cleavage sites for the restriction endonucleases BglII and NcoI were incorporated into the forward and reverse primers, respectively, used for the amplification of Tn4001 with either a single or both IS256 arms. The same cleavage sites were included in the oligonucleotide primers for the amplification of the gentamicin resistance gene, while SacI and KpnI cleavage sites were included in the forward and reverse primers used for amplification of the transposase (tnp) gene (Table S1). PCR reactions were performed in a thermocycler (iCycler, Bio-Rad) with 50 pg of plasmid DNA as template in a 50 µl reaction containing 5 µl of 10× Mg2+ free HiFi buffer, 2 mM MgSO4, 250 nM of each primer, 200 µM of each deoxyribonucleotide triphosphate (dNTP) and 2.5 U of Platinum HiFi Taq DNA polymerase (Invitrogen).
Development of novel reporter construct
To create a novel transposon from which the antibiotic resistance marker could be excised following transposon insertion in, and disruption of, a specific gene, operator and gene region fragments were designed and then synthesised commercially and cloned in the EcoRV site of pUC57 (GenScript Corporation). The operator region contained an inverted repeat (IR) (39 bp, 5′-gataaagtccgtataattgtgtaaaagtaaaaaggccat-3′) together with the M. bovis tuf promoter (252 bp tuf promoter region located between bases 474270 and 474521 of NCBI Reference Sequence NC_014760.1), a vsp signal sequence (84 bp, gene ID 10014768, predicted protein sequence MKKSKFLLLGSVASLASIPFVAAKCGET) and the FRT sequence (34 bp Flp recognition target, 5′-gaagttcctattctctagaaagtataggaacttc-3′). The gene region included the FRT sequence (34 bp, 5′-gaagttcctattctctagaaagtataggaacttc-3′), an M. bovis codon optimised alkaline phosphatase reporter gene (phoA)  and the IR (39 bp, 5′-atggcctttttacttttacacaattatacggactttatc-3′). The operator and gene segments were digested separately with EcoRI and XhoI and the operator segment ligated to the gene segment in the pUC57 backbone so that the FRT sequences were oriented as direct repeats. The nucleotide sequence of this novel construct, and relevant restriction endonuclease cleavage sites, are shown in Figure S1.
Construction of plasmids carrying transposons
Different Tn4001-based transposon constructs coding for gentamicin or tetracycline resistance and containing a single IS256 arm or both IS256 arms, and minitransposons, were generated (Figure 2). Tn4001 containing either a single or both IS256 arms (Figure S2), including the region coding for the gentamicin resistance gene (aacA-aphD), were amplified from Ptag7  using the primer pairs 1SSIS256 for/2SSISgent rev and 1SSIS256 for/3SSIS256 rev, respectively (Table S1). Each PCR product was ligated to pGEM-T (Promega) and its DNA sequence confirmed by DNA sequencing using ABI PRISM Big Dye 3.1 Terminator chemistry (Life Technologies). Sequencing revealed that use of primer 1SSIS256 for had resulted in amplification of the complete Tn4001, resulting in inclusion of the BglII cleavage site at the 5′ and 3′ ends. Therefore, pGEM-T plasmids containing either a single IS256 arm or the complete Tn4001 were digested with BglII and NcoI or BglII alone, respectively, and ligated between the FRT sites of constructs digested with the same enzymes to generate the pTn4001single and pTn4001complete constructs. To facilitate insertion of the complete Tn4001 the construct was incubated with 150 units of bacterial alkaline phosphatase (BAP, Invitrogen) at 65°C for 1 h to prevent plasmid recircularisation.
Tn4001 with one or both insertion sequences was amplified and inserted in between the FRT sites of the construct to generate pTn4001single (a) and pTn4001complete (b), respectively. The construct pMiniTn4001-gent (c) was developed by amplifying and inserting the gentamicin resistance gene (aacA-aphD) between the two FRT sites of the construct, then the transposase gene (tnp) was amplified and inserted outside of the transposable element (IR, black bar). To generate the plasmid pMiniTn4001-tet (d), a fragment containing the IR, the promoter (p), the signal (s) and an FRT site was ligated to a fragment containing an FRT site, the reporter gene (phoA) and an IR in the pUC57 plasmid backbone. FRT sites have a unique XbaI cleavage site, so ligation of the fragments produced a construct with a single FRT site. The tnp gene was amplified and ligated into the plasmid outside the transposing element, then the tetM resistance gene with its own promoter and terminator was ligated within the construct.
To overcome potential problems associated with subsequent transposition and multiple insertions, Tn4001-based minitransposons containing the genes coding for either gentamicin or tetracycline resistance were developed. For construction of pMiniTn4001-gent (Figure 2), the complete gentamicin resistance gene, with its promoter and terminator sequences, was amplified by PCR from the pTn4001single plasmid construct using the Gmgene for/Gmgene rev primer pair (Table S1), which contained engineered restriction endonuclease cleavage sites. The gentamicin resistance gene was cloned in pGEM-T, released by digestion with BglII and NcoI, and then ligated between the two FRT sites in the novel construct, which had been digested using the same pair of endonucleases. The tnp gene was then amplified from the pTn4001single plasmid using the primer pair Tnp for/Tnp rev, ligated into pGEM-T, excised with SacI and KpnI and then ligated into plasmid that had been cleaved with SacI and KpnI in a site external to the transposing element.
Another minitransposon, pMiniTn4001-tet (Figure 2), which had a single FRT site and encoded the tetracycline resistance gene (tetM), was also generated. In this construct, the M. bovis operator region was substituted with the ltuf promoter and vlhA1.1 signal sequence of M. gallisepticum strain S6 . As the FRT sequences contain a single XbaI cleavage site, ligation of the operator and gene segments after digestion with SacI and XbaI produced a single FRT site (Figure S3) in the construct, with pUC57 as the backbone. The tnp gene was then ligated outside of the transposing element in a site exposed by digestion with SacI and KpnI. Finally, the tetM gene with its own promoter and terminator was released from pMlori  by digestion with SpeI and ligated into the SpeI site in the plasmid containing the tnp gene at the SacI-KpnI site.
Transformation of M. bovis and creation of mutant libraries
Approximately 5 µg of each plasmid construct was used for transformation. The method used was based upon that described by Chopra-Dewasthaly et al. (2005), with some modifications. Briefly, 8 dilutions of a M. bovis culture were made in mycoplasma broth (1∶5, 1∶11.25, 1∶12.2, 1∶13.3, 1∶15, 1∶17.5, 1∶21.65 and 1∶30), and these incubated at 37°C for 16 h (late exponential phase). The cultures were pooled and cells were harvested by centrifugation at 16,000 g for 5 min at room temperature (RT) in a bench-top centrifuge. The cells were washed twice in 250 µl ice-cold HEPES–sucrose buffer (8 mM HEPES, 272 mM sucrose, pH 7.4). The cell pellet was then resuspended in 100 µl HEPES–sucrose buffer containing 5 µg plasmid DNA and transferred to a pre-chilled electroporation cuvette (0.2 cm, Bio-Rad). The mixture was kept on ice for 30 min and then pulsed (2.5 kV, 100 Ω and 25 µF) using a Gene Pulser (Bio-Rad). The cells were immediately resuspended in 1 ml cold mycoplasma broth (4°C), placed on ice for a further 15 min and then incubated at 37°C for 2 h. The transformed culture was then plated onto a selective mycoplasma plate containing 50 µg gentamicin/ml or 5 µg tetracycline/ml. The plates were allowed to dry, then incubated in the dark in an airtight canister at 37°C and examined for colonies after five days. Individual colonies were picked using a Pasteur pipette, inoculated into 500 µl broth containing an appropriate selective antibiotic, and incubated at 37°C until the colour of the medium changed. These cultures were used to create a mutant library of M. bovis, with each clone possessing a transposon insertion created using one of the four different constructs described above.
PCR-based detection of the selectable marker in cloned transformants
To confirm the presence of the transposable element in the genome of the mutants, a screening PCR was performed that targeted the antibiotic resistance determinant. To verify the presence of either antibiotic resistance gene, cells from 100 µl of culture were pelleted by centrifugation at 16,000 g for 5 min at RT, the supernatant discarded and the cell pellet resuspended in 25 µl of distilled water. The resuspended cells were incubated at 100°C for 5 min and used as template for PCR. The PCR assays used 2 µl of DNA template in a 25 µl reaction mixture containing 1.25 U of Gotaq DNA polymerase (Promega) in 1× buffer supplied by the manufacturer, 200 µM of each dNTP, 1.25 mM MgCl2 and 250 nM of each oligonucleotide primer for amplification of the gentamicin (Gm for/Gm rev) or tetracycline (LAtetM for/LBtetM rev) resistance genes (Table S1).
PCR-based screening for specific gene knockouts
The ‘haystack mutagenesis’ approach  was employed to screen the library of transposon mutants for insertions in four targeted genes. To limit the number of PCR reactions, 168 individual transposon-generated mutants were cultured in 1 ml of mycoplasma broth and arranged in seven pools containing 20 to 30 mutants. The genomic DNA was extracted from these pools using the High Pure DNA purification kit (Roche). The insertion of the transposon in the genome could have occurred in either orientation (Figure 3), so the screening PCR was performed using a pairs of primers that included the IR inverse oligonucleotide, which was specific for the transposon but could bind at either end of it, and either a forward or reverse oligonucleotide specific for the gene of interest (GOI) (Table S1) to identify a pool containing the desired GOI-transposon junction. Subsequently, a similar PCR using DNA prepared by boiling a cell pellet suspended in distilled water was performed on all the individual mutants in the positive pool to identify the mutant of interest. The relative position of the transposon insertion within the GOI was estimated from the size of the PCR fragment. To confirm the location of the transposon within the specific gene, the PCR product generated was cloned into pGEM-T and its DNA sequence determined. The location of the transposon in the xer1 gene was further confirmed by direct genome sequencing.
The insertion of the transposable element in a particular gene can occur in two possible orientations. PCR reaction using a primer pair, one based on the 39-bp IR sequence (uppercase) of the transposon and other one being either the forward (in this figure) or reverse primer flanking the gene of interest (GOI) would generate a single PCR product in the event of gene disruption. The relative position of the transposon insertion within the GOI is estimated based on the size of the PCR fragment including the region of binding of forward or reverse primer and primer based on IR region of transposon.
Determination of transposon insertion sites in the genome
After selection from the initial agar plate each mutant was passaged a further two times in selective mycoplasma broth at 37°C to amplify the culture up to a volume of 8–10 ml. The cells were harvested by centrifugation at 11,000 g for 20 min at 4°C, washed twice in phosphate buffered saline (PBS) (140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4), and finally resuspended in 200 µl PBS. Genomic DNA extraction was performed using the High Pure PCR kit (Roche) according to the manufacturer's protocol, except that the initial lysozyme treatment was omitted and the DNA was eluted in 50 µl of elution buffer. DNA sequencing was performed directly on genomic DNA extracted from transposon mutants. The oligonucleotide sequencing primers tuf inverse and T7 universal (Table S1), which bind within the transposon at distances of 42–67 bp and 59–78 bp, respectively, from its insertion site, were used to sequence across the transposon-genomic DNA junction. Each 20 µl reaction contained 2–3 µg of purified genomic DNA, 30 µM of the primer, 4 µl of Big Dye terminator (BDT) v3.1 enzyme mixture and 4 µl of 5× BDT dilution buffer. The sequencing products were purified and their sequence determined. The resultant DNA sequence was then used to identify the location of each transposon in the M. bovis PG45 genome  using BLAST (National Centre for Biotechnology Information, NCBI www.ncbi.nlm.nih.gov). The insertion sites were mapped onto the M. bovis PG45 genome using Geneious Pro 5.1.6 (Biomatters Ltd).
Criteria for gene inactivation
To address the question of which M. bovis genes were dispensable for growth in laboratory media, a gene was considered to be disrupted if the transposon insertion was located after the first three codons and within the first 85% of the protein coding sequence. Global transposon disruption studies – have identified a repertoire of putative essential genes, and a recent study has predicted a set of 153 essential genes for all Mycoplasma species . The dispensable genes in our M. bovis library were compared with the genes defined as essential in these previous studies.
Nucleotide sequence of novel transposon constructs. Relevant restriction endonuclease cleavage sites used to generate the construct are indicated above the sequence. The inverted repeat (IR) regions that act as transposable elements are marked, as well as the tuf promoter (p), the Vsp signal sequence (s), two directly oriented FRT sites and the phoA gene.
Nucleotide sequence of Tn4001 (Ptag7) and deduced amino acid sequences of tnp and aacA-aphD. Relevant primer binding sites are marked above the sequence, while start and stop codons of tnp and aacA-aphD are indicated below the sequence.
Nucleotide sequence of M. gallisepticum based transposon construct and predicted phoA translation. Relevant restriction endonuclease cleavage sites are indicated above the sequence. The transposable element between the inverted repeats (IR) contains the ltuf promoter (p), the vlhA1.1 signal sequence(s), a single FRT site and phoA. The predicted translation of phoA from the ltuf promoter, fused to the vlhA1.1 signal sequence, following expected excision of the resistance marker is shown. The region outside the IRs contains the multicloning sites of the plasmid into which the region was ligated.
Primers used for PCR in this study and their products.
Transposon insertions in M. bovis strain PG45 considered unlikely to disrupt function.
Transposon insertions within predicted intergenic regions in M. bovis strain PG45.
Transposon insertions within integrative conjugative elements (ICEs) in M. bovis strain PG45.
Transposon insertions within transposase genes in M. bovis strain PG45.
Conceived and designed the experiments: SS GFB PFM. Performed the experiments: SS. Analyzed the data: SS PFM GFB. Wrote the paper: SS PFM GFB.
- 1. Caswell JL, Archambault M (2007) Mycoplasma bovis pneumonia in cattle. Animal Health Research Reviews 8: 161–186.
- 2. Dybvig K, Voelker LL (1996) Molecular biology of mycoplasmas. Annual Review of Microbiology 50: 25–57.
- 3. Halbedel S, Stulke J (2007) Tools for the genetic analysis of Mycoplasma. International Journal of Medical Microbiology 297: 37–44.
- 4. Minion FC (2002) Molecular pathogenesis of mycoplasma animal respiratory pathogens. Frontiers in Bioscience 7: d1410–1422.
- 5. Razin S, Yogev D, Naot Y (1998) Molecular biology and pathogenicity of mycoplasmas. Microbiology and Molecular Biology Reviews 62: 1094–1156.
- 6. Rottem S (2003) Interaction of mycoplasmas with host cells. Physiological Reviews 83: 417–432.
- 7. Barre A, de Daruvar A, Blanchard A (2004) MolliGen, a database dedicated to the comparative genomics of Mollicutes. Nucleic Acids Research 32: D307–310.
- 8. Sirand-Pugnet P, Lartigue C, Marenda M, Jacob D, Barre A, et al. (2007) Being pathogenic, plastic, and sexual while living with a nearly minimal bacterial genome. PLoS Genetics 3: e75.
- 9. Wise KS, Calcutt MJ, Foecking MF, Roske K, Madupu R, et al. (2011) Complete genome sequence of Mycoplasma bovis type strain PG45 (ATCC 25523). Infection and Immunity 79: 982–983.
- 10. Li Y, Zheng H, Liu Y, Jiang Y, Xin J, et al. (2011) The complete genome sequence of Mycoplasma bovis strain Hubei-1. PLoS One 6: e20999.
- 11. Qi J, Guo A, Cui P, Chen Y, Mustafa R, et al. (2012) Comparative geno-plasticity analysis of Mycoplasma bovis HB0801 (Chinese Isolate). PLoS One 7: e38239.
- 12. Behrens A, Heller M, Kirchhoff H, Yogev D, Rosengarten R (1994) A family of phase- and size-variant membrane surface lipoprotein antigens (Vsps) of Mycoplasma bovis.. Infection and Immunity 62: 5075–5084.
- 13. Lysnyansky I, Sachse K, Rosenbusch R, Levisohn S, Yogev D (1999) The vsp locus of Mycoplasma bovis: gene organization and structural features. Journal of Bacteriology 181: 5734–5741.
- 14. Dybvig K, Lao P, Jordan DS, Simmons WL (2010) Fewer essential genes in mycoplasmas than previous studies suggest. FEMS Microbiology Letters 311: 51–55.
- 15. French CT, Lao P, Loraine AE, Matthews BT, Yu H, et al. (2008) Large-scale transposon mutagenesis of Mycoplasma pulmonis. Molecular Microbiology 69: 67–76.
- 16. Glass JI, Assad-Garcia N, Alperovich N, Yooseph S, Lewis MR, et al. (2006) Essential genes of a minimal bacterium. Proceedings of the National Academy of Sciences of the United States of America 103: 425–430.
- 17. Whetzel PL, Hnatow LL, Keeler CL Jr, Dohms JE (2003) Transposon mutagenesis of Mycoplasma gallisepticum. Plasmid 49: 34–43.
- 18. Muneta Y, Panicker IS, Kanci A, Craick D, Noormohammadi AH, et al. (2008) Development and immunogenicity of recombinant Mycoplasma gallisepticum vaccine strain ts-11 expressing chicken IFN-gamma. Vaccine 26: 5449–5454.
- 19. Shil PK, Kanci A, Browning GF, Markham PF (2011) Development and immunogenicity of recombinant GapA(+) Mycoplasma gallisepticum vaccine strain ts-11 expressing infectious bronchitis virus-S1 glycoprotein and chicken interleukin-6. Vaccine 29: 3197–3205.
- 20. McNamara PJ (2008) Genetic Manipulation of Staphylococcus aureus. In: Lindsay JA, editor. Staphylococcus: Molecular Genetics. Norfolk, U.K.: Caister Academic Press. pp. 89–130.
- 21. Knudtson KL, Minion FC (1993) Construction of Tn4001lac derivatives to be used as promoter probe vectors in mycoplasmas. Gene 137: 217–222.
- 22. Chopra-Dewasthaly R, Zimmermann M, Rosengarten R, Citti C (2005) First steps towards the genetic manipulation of Mycoplasma agalactiae and Mycoplasma bovis using the transposon Tn4001mod. International Journal of Medical Microbiology 294: 447–453.
- 23. Panicker IS, Kanci A, Chiu CJ, Veith PD, Glew MD, et al. (2012) A novel transposon construct expressing PhoA with potential for studying protein expression and translocation in Mycoplasma gallisepticum. BMC Microbiology 12: 138.
- 24. Tseng CW (2007) Improving mycoplasma vaccines - targets for defined attenuation: School of Veterinary Science, The University of Melbourne.
- 25. Lee SW, Browning GF, Markham PF (2008) Development of a replicable oriC plasmid for Mycoplasma gallisepticum and Mycoplasma imitans, and gene disruption through homologous recombination in M. gallisepticum. Microbiology 154: 2571–2580.
- 26. Lin Y, Zhang RR (2011) Putative essential and core-essential genes in Mycoplasma genomes. Scientific Reports 1: 53.
- 27. Chiu CJ (2006) Protective immune response to antigens expressed by mycoplasma vectors: School of Veterinary Science, The University of Melbourne.
- 28. Hutchison CA, Peterson SN, Gill SR, Cline RT, White O, et al. (1999) Global transposon mutagenesis and a minimal mycoplasma genome. Science 286: 2165–2169.
- 29. Thomas A, Linden A, Mainil J, Bischof DF, Frey J, et al. (2005) Mycoplasma bovis shares insertion sequences with Mycoplasma agalactiae and Mycoplasma mycoides subsp. mycoides SC: Evolutionary and developmental aspects. FEMS Microbiology Letters 245: 249–255.
- 30. Kobayashi K, Ehrlich SD, Albertini A, Amati G, Andersen KK, et al. (2003) Essential Bacillus subtilis genes. Proceedings of the National Academy of Sciences of the United States of America 100: 4678–4683.
- 31. Zhang R, Ou HY, Zhang CT (2004) DEG: a database of essential genes. Nucleic Acids Research 32: D271–272.
- 32. Khan LA, Miles RJ, Nicholas RA (2005) Hydrogen peroxide production by Mycoplasma bovis and Mycoplasma agalactiae and effect of in vitro passage on a Mycoplasma bovis strain producing high levels of H2O2. Veterinary Research Communications 29: 181–188.
- 33. Pilo P, Vilei EM, Peterhans E, Bonvin-Klotz L, Stoffel MH, et al. (2005) A metabolic enzyme as a primary virulence factor of Mycoplasma mycoides subsp. mycoides small colony. Journal of Bacteriology 187: 6824–6831.
- 34. Vilei EM, Frey J (2001) Genetic and biochemical characterization of glycerol uptake in Mycoplasma mycoides subsp. mycoides SC: its impact on H(2)O(2) production and virulence. Clinical and Diagnostic Laboratory Immunology 8: 85–92.
- 35. Skapski A, Hygonenq MC, Sagne E, Guiral S, Citti C, et al. (2011) Genome-scale analysis of Mycoplasma agalactiae loci involved in interaction with host cells. PLoS One 6: e25291.
- 36. Browning GF, Marenda MS, Noormohammadi AH, Markham PF (2011) The central role of lipoproteins in the pathogenesis of mycoplasmoses. Veterinary Microbiology 153: 44–50.
- 37. Hallamaa KM, Browning GF, Tang SL (2006) Lipoprotein multigene families in Mycoplasma pneumoniae. Journal of Bacteriology 188: 5393–5399.
- 38. Schmidt JA, Browning GF, Markham PF (2007) Mycoplasma hyopneumoniae mhp379 is a Ca2+-dependent, sugar-nonspecific exonuclease exposed on the cell surface. Journal of Bacteriology 189: 3414–3424.
- 39. Szczepanek SM, Frasca S Jr, Schumacher VL, Liao X, Padula M, et al. (2010) Identification of lipoprotein MslA as a neoteric virulence factor of Mycoplasma gallisepticum. Infection and Immunity 78: 3475–3483.
- 40. Masukagami Y, Tivendale KA, Mardani K, Ben-Barak I, Markham PF, et al. (2013) The Mycoplasma gallisepticum Virulence Factor Lipoprotein MslA Is a Novel Polynucleotide Binding Protein. Infect Immun 81: 3220–3226.
- 41. De Maio A (1999) Heat shock proteins: facts, thoughts, and dreams. Shock 11: 1–12.
- 42. Acevedo-Rocha CG, Fang G, Schmidt M, Ussery DW, Danchin A (2013) From essential to persistent genes: a functional approach to constructing synthetic life. Trends in Genetics 29: 273–279.
- 43. Liu W, Fang L, Li M, Li S, Guo S, et al. (2012) Comparative genomics of Mycoplasma: analysis of conserved essential genes and diversity of the pan-genome. PLoS One 7: e35698.
- 44. Fang G, Rocha E, Danchin A (2005) How essential are nonessential genes? Molecular Biology and Evolution 22: 2147–2156.
- 45. Danchin A, Fang G, Noria S (2007) The extant core bacterial proteome is an archive of the origin of life. Proteomics 7: 875–889.
- 46. Akanuma G, Nanamiya H, Natori Y, Yano K, Suzuki S, et al. (2012) Inactivation of ribosomal protein genes in Bacillus subtilis reveals importance of each ribosomal protein for cell proliferation and cell differentiation. J Bacteriol 194: 6282–6291.
- 47. Pour-El I, Adams C, Minion FC (2002) Construction of mini-Tn4001tet and its use in Mycoplasma gallisepticum. Plasmid 47: 129–137.
- 48. Lartigue C, Duret S, Garnier M, Renaudin J (2002) New plasmid vectors for specific gene targeting in Spiroplasma citri. Plasmid 48: 149–159.
- 49. Markham PF, Kanci A, Czifra G, Sundquist B, Hains P, et al. (2003) Homologue of macrophage-activating lipoprotein in Mycoplasma gallisepticum is not essential for growth and pathogenicity in tracheal organ cultures. J Bacteriol 185: 2538–2547.
- 50. Cordova CM, Lartigue C, Sirand-Pugnet P, Renaudin J, Cunha RA, et al. (2002) Identification of the origin of replication of the Mycoplasma pulmonis chromosome and its use in oriC replicative plasmids. Journal of Bacteriology 184: 5426–5435.
- 51. Janis C, Bischof D, Gourgues G, Frey J, Blanchard A, et al. (2008) Unmarked insertional mutagenesis in the bovine pathogen Mycoplasma mycoides subsp. mycoides SC: characterization of a lppQ mutant. Microbiology 154: 2427–2436.
- 52. Hasselbring BM, Page CA, Sheppard ES, Krause DC (2006) Transposon mutagenesis identifies genes associated with Mycoplasma pneumoniae gliding motility. Journal of Bacteriology 188: 6335–6345.
- 53. Baranowski E, Guiral S, Sagne E, Skapski A, Citti C (2010) Critical role of dispensable genes in Mycoplasma agalactiae interaction with mammalian cells. Infection and Immunity 78: 1542–1551.
- 54. Schmidt JA, Browning GF, Markham PF (2004) Mycoplasma hyopneumoniae p65 surface lipoprotein is a lipolytic enzyme with a preference for shorter-chain fatty acids. Journal of Bacteriology 186: 5790–5798.