Molecular biology of mycoplasmas : from the minimum cell concept to the arti fi cial cell

Mycoplasmas are a large group of bacteria, sorted into different genera in the Mollicutes class, whose main characteristic in common, besides the small genome, is the absence of cell wall. They are considered cellular and molecular biology study models. We present an updated review of the molecular biology of these model microorganisms and the development of replicative vectors for the transformation of mycoplasmas. Synthetic biology studies inspired by these pioneering works became possible and won the attention of the mainstream media. For the fi rst time, an artifi cial genome was synthesized (a minimal genome produced from consensus sequences obtained from mycoplasmas). For the fi rst time, a functional artifi cial cell has been constructed by introducing a genome completely synthesized within a cell envelope of a mycoplasma obtained by transformation techniques. Therefore, this article offers an updated insight to the state of the art of these peculiar organisms’ molecular biology.


INTRODUCTION
It is believed that mycoplasmas originated through degenerative evolution from Gram-positive ancestors, with the loss of non-essential genes (Dybvig and Voelker 1996).The term mycoplasma is often used to refer to all microorganisms in the Mollicutes class, regardless of which genus they belong to.In order to prevent confusion with microorganisms specifi c to the Mycoplasma genus, the term mollicutes has come into use as a generic name for all microorganisms in the class (Razin et al. 1998).The criteria for defi ning the taxonomy of mollicutes are established by the Subcommittee on the Taxonomy of Mollicutes, linked to the International Committee on Systematics of Prokaryotes, and can be found at its website: <http://icsp.org/subcommittee/mollicutes/>.

MOLECULAR BIOLOGY OF MOLLICUTES
Aside from the fundamental differences between these and other prokaryotes, the molecular biology CAIO M.M. CORDOVA et al.
of mollicutes is very typical in some respects.Having evolved from Gram-positive bacteria, gene expression and other genetic characteristics of mollicutes are similar to those of their ancestors.On the other hand, because of the limited capacity of their genome, mollicutes do not have many of the enzymatic pathways characteristics of the majority of bacteria (Dybvig and Voelker 1996).These characteristics make mollicutes particularly interesting in molecular and cellular biology studies, as, even with these limitations, they are capable of independent replication and parasitism, causing diseases that lead to losses in agriculture as well as in human health.

GENOME STRUCTURE AND ORGANIZATION
Mollicutes have a chromosome made up of circular double stranded DNA, which can vary between 580 and 1,800 kb (Razin et al. 1998).The 580 kb chromosome of the human pathogen M. genitalium is currently the smallest known genome (Lucier et al. 1994).The GC content of the majority of mollicutes' chromosomes is around 35%, with 25% being the smallest (Glass et al. 2000).Even though they have such small genomes, some mollicutes have repeated sequences of DNA in their chromosomes.The repeated elements found in mollicutes are homologous to members of the IS3 Family Insertion Sequence (Westberg et al. 2002, Pilo et al. 2003, Ditty et al. 2003, Thomas et al. 2005, Lysnyansky et al. 2009).
With the exception of spiroplasmas, extrachromosomal DNA is rarely found in mollicutes.Currently, only a few plasmids have been isolated from M. mycoïdes subsp.mycoïdes and from S. citri (Bergemann et al. 1989).On the other hand, numerous replicative forms of viral genomes (phages) have been isolated from Acholeplasma and other species of Mycoplasma and Spiroplasma (Maniloff 1992).
In recent years, the greatest advances in the understanding of mollicute biology were obtained with the sequencing of the complete genome of various species.Currently, the complete genome sequences of 46 species and strains have been published.For a detailed analysis, we suggest visiting MolliGen.org.MolliGen is a database dedicated to the comparative genomics of bacteria belonging to the Mollicutes class.It offers sequence and analysis of data collected on 54 genomes belonging to 37 species.The MolliGen database provides pages with genome and genetic elements, homology relationships defined by various methods, precompiled statistical data, metabolic pathways, and a large collection of tools to explore and compare genomes (Barré et al. 2004).Currently there is an effort, sponsored by the International Organization for Mycoplasmology (IOM), to encourage genome sequencing of all species of mollicutes.

DNA REPLICATION AND REPAIR
Studies in this area have long maintained the idea that mollicutes were very limited with respect to their DNA repair mechanisms.However, at least in M. pulmonis, transposon induced mutagenesis in the recA, ruvA, ruvB, and ruvC genes produced defi cient phenotypes (French et al. 2008).Through in silico analysis of sequenced genomes, the only complete repair mechanism found in mycoplasmas is that of repair by nucleotide excision repair (NER), probably complemented by enzymes from the base excision repair (BER) mechanism and the recombination mechanism (Carvalho et al. 2005).The NER mechanism was also identified in M. genitalium, in which the RecA protein has an active role in the antigenic variation mechanism but not in DNA repair (Burgos et al. 2012).A histone-like protein that recognizes nucleotide incorporation errors in DNA was identifi ed in M. gallisepticum.However, this does not appear to be a rule among mycoplasmas, which, during their evolution, lost the DNA mismatch repair mechanism (MMR) (Kamashev et al. 2011).Nevertheless, despite their limitations, it is evident that mycoplasmas possess an adequate set of genes and gene products, even if they have not yet been completely identifi ed, to promote DNA repair in a way that makes their development possible, causing signifi cant diseases in a wide range of hosts (Gorbachev et al. 2013).

GENETIC RECOMBINATION
Until recently, there was no evidence that homologous recombination could occur in mollicutes' genomes.In M. agalactiae, the necessary genes for homologous recombination appear to be reduced, showing only the components involved in the formation of the presynaptic complex (recD, recO, and recR), strand exchange (recA), branch migration (ruvA and ruvB), and the Holliday junction resolution (recU and yqgF); however, it has already been demonstrated in other bacteria that, with the absence of complete sets of genes, homologous recombination can occur (Rocha et al. 2005).Gene disruption by homologous recombination has already been proven in M. genitalium (Dhandayuthapani et al. 1999) and in S. citri (Duret et al. 1999), and in Mycoplasma mycoïdes subsp.capri, inclusion of recA from E. coli in a suicide plasmid favored the occurrence of mutants (Allam et al. 2010).Apparently there are various mechanisms that can be explored to increase the effectiveness of this approach, which may be species-dependent.

GENETIC MATERIAL TRANSFER
Because of the absence of a cellular wall, it can be expected that the introduction of exogenous genetic material to mollicute cells happens without much diffi culty.In fact, the chromosomic DNA exchange occurring during direct contact between mollicute cells (Barroso and Labarere 1988), and the spontaneous conjugal transfer of Tn916 between Enterococcus faecalis and M. hominis (Roberts and Kenny 1987) is probably due to the momentary fusion of cellular membranes in the contact region.
Nevertheless, the efficiency of DNA transfer in these cases was low.

VECTORS
Transfection experiments of S. citri with the SpV1 virus containing an inserted gene that codes the M. pulmonis P1 adhesin resulted in an expression of this spiroplasma fragment.However, this system is inefficient, since there is a rapid loss of the cloned DNA fragment.Transposons have also been used in the transformation of A. laidlawii and M. pulmonis in antibiotic-resistant mollicutes using the pAM120 plasmid, containing the Tn916 streptococcus transposon, which confers resistance to tetracycline through the expression of the tetM gene (Dybvig and Cassell 1987).Other transposons have also been used with various species of mollicutes since then, such as Tn1545 and Tn4001 (Cao et al. 1994, de Barbeyrac et al. 1996, Foissac et al. 1997, Kapke et al. 1994).The transposons undergo excision from the plasmid that contains them, and integrate with the mollicute chromosome in random positions (Foissac et al. 1997).Mutagenesis by transposon is done by inserting it into genes, causing the inactivation of transcription or the formation of truncated products.The genes containing the transposon are then mapped and identifi ed, and the defi nitive proof of its function is given by the complementation of the mutant with the wild-type gene.This strategy has been used in M. pneumoniae (Fisseha et al. 1999), M. genitalium (Reddy et al. 1996) and S. citri (Jacob et al. 1997).However, in many cases, these elements transpose actively, changing location in the chromosome frequently (Foissac et al. 1997).
Another type of vector was constructed by inserting the oriC region of S. citri in the commercial plasmid pBS (Stratagene, La Jolla, Ca, USA), and the resulting recombinant plasmid was effi ciently introduced into the spiroplasma cells by electroporation, where it was able to replicate and maintain stability as an extrachromosomal element (Renaudin et al. 1995).Vectors derived from this plasmid show themselves to be capable of inactivating specific S. citri genes (Duret et al. 1999).The great advantage of these types of replicative plasmids is the possibility to direct the insertion of the vector to a specifi c region of the chromosome.In this way interference from the random insertion of the vector in other mollicute genes can be avoided, which can happen with viral particles and transposons.This method also makes it possible to promote complementation of the mollicute genes in mutant strains (Gaurivaud et al. 2000), definitively proving its function.Vectors derived from the pKMK1 plasmid were also used in the transformation of M. mycoïdes and M. capricolum (King and Dybvig 1993).However, these plasmids produce few copies, they are incapable of transforming other species of mollicutes, and the stability of the inserts in these vectors does not appear to be adequate (Dybvig and Voelker 1996).
The cloning of the fragment containing the DnaA boxes located near the M. capricolum subsp.capricolum dnaA gene, through the pSRT2 vector, initially developed to transform S. citri (Lartigue et al. 2002), made possible the construction of two new plasmids, pMCO2 and pMCO3.The pMCO2 plasmid was constructed using the oriC region of M. capricolum inserted at the BamHI multiple cloning sites oriented in the same direction as the tetM gene.Construction of the pMCO3 plasmid was done in a similar manner to that of pMCO2, although the cloned fragment was oriented in the opposite direction of the tetM gene.By cloning the amplifi ed fragments of the spoJ-dnaA and dnaN-spoJ intergenic regions of M. genitalium in the pSRT2 vector, it was possible to obtain the pMGO1 and pMGO2 plasmids.In the same way, cloning the fragments containing the putative M. pulmonis UAB CTIP DnaA boxes in the pSRT2 vector made it possible to create the new plasmid named pMPO1.This plasmid contains the M. pulmonis oriC region inserted at the BamHI site of the multiple cloning regions in the same orientation as the tetM gene.By amplifying the 327 bp fragment containing the DnaA boxes located in the dnaA-dnaN intergenic regions of M. pulmonis and cloning it in the pSRT2 vector, the pMPO2 plasmid was constructed.In the same way, by cloning a 262 bp fragment containing the DnaA boxes located in the recD-dnaA intergenic region, the pMPO3 plasmid was constructed.The pMPO4 plasmid was constructed cloning both of these fragments side by side at the BamHI site of the pSRT2 vector.The pMPO5 plasmid was constructed cloning these two fragments with the tetM gene in between (Cordova et al. 2002).Further studies using the results of this work were able to produce the specifi c oriC plasmids capable of transforming the M. mycoïdes subsp.mycoïdes SC and M. mycoïdes subsp.capri (Lartigue et al. 2003).Suicide plasmids were constructed by amplifying a 717 bp internal fragment of the M. pulmonis hemolysin A gene and inserting it in the pSRT2, pMPO1 and pMPO5 plasmids, by which disruption vectors were obtained, with the names pSRT2-∆hlyA, pMPO1-∆hlyA and pMPO5-∆hlyA, respectively (Cordova et al. 2002).
Plasmids with the M. capricolum subsp.capricolum oriC region are not capable of replicating in M. mycoïdes subsp.mycoïdes SC or M. mycoïdes subsp.capri, all phylogenetically related and belonging to the Mycoïdes group.On the other hand, plasmids with the oriC region from M. mycoïdes subsp.mycoïdes SC, M. mycoïdes subsp.capri or even from S. citri are capable of transforming and replicating in M. capricolum.M. mycoïdes subsp.mycoïdes SC and M. mycoïdes subsp.capri are capable of being transformed with each other's plasmids (Lartigue et al. 2003).The use of plasmids with the M. mycoïdes subsp.capricolum oriC region was shown to be viable for heterologous gene expression in this mollicute, but gene disruption by homologous recombination was only possible with plasmids containing the S. citri oriC region (Janis et al. 2005).

PRESENCE OF DNA RESTRICTION-MODIFICATION SYSTEMS
Restriction-modifi cation systems are very common in mollicutes and can obstruct gene transfer.Some studies have shown that the transformation of S. citri with isolated E. coli plasmids are only successful when the E. coli strain used had type I EcoK methyltransferase, which suggests that the methylation by EcoK is necessary to protect plasmid DNA from the S. citri restriction system (Bové 1993).Because of this, transformation errors of certain mollicutes with certain vectors can be due to restriction enzyme activity (Dybvig andVoelker 1996, Sirand-Pugnet et al. 2007).

HORIZONTAL GENE TRANSFER
The possibility of horizontal gene transfer between different mollicutes, at least for more phylogenetically related species, has long been investigated, considering the small genome and probable degenerative evolution mechanism, which in theory doesn't contribute to the acquisition of new genes.One of the fi rst studies to demonstrate horizontal gene transfer was between different serotypes of Ureaplasma sp., isolated from clinical samples (Xiao et al. 2011).These studies explained the persistent difficulty in relating different levels of pathogenicity and thus different clinical presentations to different microorganism serotypes.In this research, it was demonstrated that not less than 40% of the isolated U. urealyticum/ parvum strains were genetic mosaics.Evidence of horizontal gene transfer had already been reported in species located in the same ecological site (Sirand-Pugnet et al 2007) and in mycoplasmas of the same species (Teachman et al. 2002).It was also demonstrated that certain strains of Ureaplasma sp. are capable of forming biofi lms, just as various mollicutes that infect animals, such as M. pulmonis, a characteristic acquired from other bacterial species (Garcia-Castillo et al. 2008).In addition to biofi lms, the presence of an integrative conjugative element (ICE) in M. agalactiae, as multiple copies in the chromosome or in free circular form, is worthy of attention because it is a microorganism responsible for a disease that can also be caused by M. capricolum subsp.capricolum and M. mycoïdes subsp.capri (Marenda et al. 2006).ICEs can also be called CONSTINS or conjugative transposons and are capable of conferring various metabolic or resistance characteristics to their hosts.They are self-transmissible elements, which carry the genes necessary for excision, conjugative transfer, and integration with the host (Burrus and Waldor 2004).There are indications that at least 18% of the M. agalactiae genome was transferred horizontally from mycoplasmas of the Mycoïdes group, since the former, even though it can inhabit the same hosts and ecological sites, is phylogenetically located in the Hominis group (Sirand-Pugnet et al. 2007).Nevertheless, half of these genes' functions are still unknown, and the mass transfer of genes is not believed to have had a very signifi cant role in the evolution of mollicutes.However, this data indicates that, in addition to being able to survive against the host's defenses, mollicutes are also capable of promoting chromosomic exchange to a certain degree, which avoids genome stagnation, making possible the continued adaptation to new sites and even new hosts (Sirand-Pugnet et al. 2007).Nevertheless, the mechanism by which gene transfer occurs between mollicute cells is still not well documented (Dordet-Frisoni et al. 2014).

THE MINIMAL GENOME CONCEPT
Because M. genitalium has the smallest bacterial genome currently known (Fraser et al. 1995), some authors have investigated the hypothesis that its genome represents the minimum gene set suffi cient for life in a microorganism capable of self-replication.Mushegian and Koonin (1996) CAIO M.M. CORDOVA et al. compared this mycoplasma's genome with that of Haemophilus influenzae, identifying 240 common genes.Adding 22 genes related to essential metabolic pathways not identifi ed in these microorganisms, and subtracting 6 redundant genes specifi c to parasites, the authors came to a total of 256 genes, which they classifi ed as the approximate minimum number of genes necessary for the functioning of a cell.However, M. genitalium has double this number of genes.Maniloff (1996) suggests that the difference is probably refl ected in that the comparison was made between a minimal genome produced by billions of years of evolution and a minimal gene set produced by a computer.Other authors attempted to determine the minimum number of essential genes through mutagenesis experiments with M. genitalium (Hutchinson et al. 1999).In a study involving global transposon mutagenesis, the authors determined that 387 genes is the minimum set necessary for life (Glass et al. 2006).In a comparison of 186 genomes, including M. genitalium, other authors determined that 702 is the minimum number for Gram-negative bacteria and 959 for Gram-positive bacteria.Naturally, the necessity for genes in mycoplasma cells is lower, since they maintain a parasitic life in their hosts (Huang et al. 2013).In M. pneumoniae, phylogenetically related to M. genitalium, 33% of its 816 kpb genome, including 0.9% of intergenic regions, were considered essential (Lluch-Senar et al. 2015).In M. mycoïdes, using gene cloning in yeast cells and carrying out transposon mutagenesis to reduce the genome, making use of yeast's sexual cycle, 368 genes were determined as non-essential gene candidates.From these, 11 of 14 genes were of the DNA restriction-modifi cation system, 3 of 5 were vlc genes, and 7 of 22 were integrative conjugative elements (Suzuki et al. 2015).

CONSTRUCTION OF AN ARTIFICIAL CELL
With the relative success of the studies attempting to determine the minimal gene set for necessary for life, at least for a bacterium, the following dream began taking shape, making use of synthetic biology: to construct a cell from the available genetic information.Creating life has not yet been accomplished, as one of the major obstacles is completely synthesizing a viable cell envelope.Nature is still very mysterious.However, constructing a new cell from the genome of a different bacterium has been done, by inserting the genome in the cell envelope (membrane) of the other.It all began with cloning experiments and transposing the entire M. mycoïdes genome into a yeast cell (from which the original DNA was removed) (Benders et al. 2010).Keeping this cloned genome in a plasmid has also been done (Karas et al. 2013), and later transposing it again into the cell membrane of another mycoplasma species (M.capricolum) (Lartigue et al. 2009).This mycoplasma to yeast genome transfection technology has also made possible new methods of reducing the genome, in the continuous search for the minimal essential gene set.In this way, it was possible to create strains with reduced genomes but normal growth, without defective phenotypes (Suzuki et al. 2015).The recombination of mycoplasma genomic elements in yeasts may make possible not only the arrangement of deleted regions (non-essential), but also a combination of deletions, insertions and substitutions of genetic elements.It is possible to recombine two entire synthetic genomes and recombine gene cassettes between them, or to recombine synthetic genome segments through homologous recombination (Suzuki et al. 2015).
A major step towards the creation of life was the synthesis of a complete genome from the M. genitalium sequence information in genetic databases, fi rst cloned in E. coli and Saccharomyces cerevisiae (Gibson et al. 2008).The next step, labeled genome synthesis technology, was to construct a new genome from the M. mycoïdes sequence, including 'watermark' sequences, dele-tions, polymorphisms and mutations, and then inserting this new genome into the cell envelope of M. capricolum.This new cell was shown to be completely controlled by the new synthetic genome (Gibson et al. 2010).This created a new microorganism, neither M. mycoïdes nor M. capricolum, which became widely known as Mycoplasma laboratorium.
The cloning of genomes in yeast, in addition to having been recreated with other bacteria, was also demonstrated with eukaryotic chromosomes (Karas et al. 2013).
If, or better yet, when it becomes truly possible to create life in a laboratory from the synthesis of complex molecules using only chemical precursors based on the available genetic information, uniting it all in a synthetic cell envelope, many paradigms will be broken, not only in science, but also in philosophy, theology, ethics and sociology.