The bacteria of the Planctomycetes, Verrucomicrobia, Chlamydiae, and Lentisphaerae phyla along with the Candidate Poribacteria, Candidate phylum OP3 and Candidate division WWE2 are collectively grouped and referred to as the PVC superphylum or the PVC clade (Wagner and Horn, 2006). The PVC group is comprised of species that are of much importance due to their characteristics and the roles they play in many areas of life. Species of the Chlamydiae phylum are one of the most widely studied microorganisms due to their pathogenic capacities in humans and in animals. They are responsible for many human illnesses including sexually transmitted urinary tract infections, trachoma, and pneumonia (Sachse et al., 2009). Species of the phylum Planctomycetes are renowned for their unusual cellular features such as internal compartmentalization, sterol biosynthesis, and endocytosis-analogous pathways that are generally associated with the eukaryotes (Fuerst and Webb, 1991; Lindsay et al., 1997; Pearson et al., 2003; Ward et al., 2006; Lonhienne et al., 2010; Fuerst and Sagulenko, 2011; McInerney et al., 2011). This phylum also harbors a group of anaerobic chemoautotrophic “anammox” (anaerobic ammonium oxidation) organisms (van de Graaf et al., 1995; Strous et al., 1999). These anammox species can oxidize ammonium to dinitrogen and are therefore quite useful in decontamination of wastewater rich in ammonia (Dalsgaard et al., 2003). Their importance is underscored by estimates which suggest that anammox bacteria may contribute up to 50% of the atmospheric nitrogen (Devol, 2003). The species from the phylum Verrucomicrobia are abundant in soil based environments with estimates proposing that up to 10% of all bacteria in the soil belong to this phylum (Sangwan et al., 2005). These bacteria are also found in aquatic environments (Martiny et al., 2005; Haukka et al., 2006) and known to associate with eukaryotic species as indicated by their presence in termite guts, human intestines, nematodes, and some ciliate protozoa (Petroni et al., 2000; Vandekerckhove et al., 2002; Shinzato et al., 2005; Wang et al., 2005). Some members of the Verrucomicrobiae are known to exist in ultramicrobial sizes, others to possess extensions of the cellular membrane termed the prosthecae and some also exist in acidophilic environments (Hedlund et al., 1997; Janssen et al., 1997; Pol et al., 2007). Thus, the species of the PVC phylum are important in our quest to better understand prokaryotic evolution, microbial ecology, and physiology.

Though much diversity exists among the bacteria of different phyla that comprises this superphylum, a close relationship among them has been suggested by the 16S rRNA trees and number of other phylogenetic studies employing single gene and multi-gene analyses of protein sequences (Cho et al., 2004; Wagner and Horn, 2006; Hou et al., 2008; Pilhofer et al., 2008; Glockner et al., 2010; Siegl et al., 2011). Among the members of this clade, the Planctomycetes and Chlamydiae were observed to be phylogenetically related as early as 1986 based on 16S rRNA secondary structures and phylogenetic trees (Weisburg et al., 1986; Woese, 1987; Fuerst, 1995). A close relationship of the Verrucomicrobia to the Chlamydiae and Planctomycetes was first observed by Hedlund et al. (1996) and the “sister-taxon” grouping of the Lentisphaerae to the Verrucomicrobia was recognized with the isolation of the first Lentisphaerae organism Victivallis vadensis (Zoetendal et al., 2003; Cho et al., 2004). The taxonomic entity labeled as the PVC superphylum was proposed in 2006, based on 16S ribosomal data, by Wagner and Horn (2006) to encompass the monophyletic group comprised of the above four phyla along with the recently discovered Candidate Poribacteria, Candidate phylum OP3 and Candidate phylum WWE2 (Hugenholtz et al., 1998; Fieseler et al., 2004; Chouari et al., 2005; Wagner and Horn, 2006). However, a monophyletic grouping of the different bacteria belonging to these phyla has also been disputed by other phylogenetic studies based upon 16S rRNA as well as several single gene and concatenated protein phylogenies (Ward et al., 2000; Jenkins and Fuerst, 2001; Ciccarelli et al., 2006; Griffiths and Gupta, 2007; Santarella-Mellwig et al., 2010).

Apart from their linkages in phylogenetic trees, little evidence exists to group the different phyla that are part of the PVC clade into a single large group. Nevertheless, some uncommon features are seen to be shared by multiple phyla of the group. The Verrucomicrobia along with the Poribacteria and Lentisphaerae share a similar intracellular structural plan with the Planctomycetes in having membranous borders dividing the cell into compartments (Fieseler et al., 2004; Lee et al., 2009; Fuerst and Sagulenko, 2011). Planctomycetes and Chlamydiae lack peptidoglycan in their cell walls (Konig et al., 1984; Liesack et al., 1986; Fox et al., 1990; Staley et al., 1992; Ward et al., 2006; Fuerst and Sagulenko, 2011). Also common among the Chlamydiae and Planctomycetes is the lack of FtsZ-based cell division (Bernander and Ettema, 2010; Fuerst and Sagulenko, 2011). However, as these features are not exclusive to the members of the PVC group and not found in all species of the phyla comprising the PVC group, they do not provide much clarity in the debate concerning the grouping of these phyla into a superphylum.

Due to the advent of rapid genomic sequencing techniques and availability of genomic sequences, comparative genomics provide powerful means for answering a variety of questions related to bacterial evolution. Using genome sequences, many approaches are being used to understand the evolutionary relationships among bacteria. While some approaches using whole genome alignments have been most used (or are mainly applicable) for studying closely related organisms (Angiuoli and Salzberg, 2011; Agren et al., 2012; Sahl et al., 2012), other comparative genomic approaches involving identification of molecular markers in the forms of either conserved signature inserts or deletions (CSIs) or conserved signature proteins (CSPs) have been extensively used to define taxonomic clades of different phylogenetic ranks in molecular terms (Gupta, 1998, 2010; Gupta and Griffiths, 2002; Dutilh et al., 2008; Gao and Gupta, 2012). The applications of these approaches previously to the Chlamydiae species have led to identification of numerous CSIs and CSPs that are specific for the species from this phylum or a number of its subclades (Griffiths et al., 2005, 2006; Gupta and Griffiths, 2006). Some interesting cases of lateral gene transfers (LGTs) between Actinobacteria and Chlamydiae were also identified by these studies (Griffiths and Gupta, 2006). Additionally, our work using these approaches also indicated that the phyla Chlamydiae and Verrucomicrobia are specifically related and they shared a common ancestor exclusive of the Planctomycetes (Griffiths and Gupta, 2007). However, thus far no molecular markers have been identified that are specific for the Planctomycetes and/or Verrucomicrobia phyla or those linking all members of the PVC group. In the present work, we describe the results of comparative genomic analysis aimed at identifying molecular markers that are uniquely shared by either the Planctomycetes or Verrucomicrobia phyla or those that are commonly shared by different main groups of the PVC superphylum. Additionally, we also report phylogenetic studies based upon concatenated protein sequences to evaluate the relationships among the PVC clade of bacteria.

Complete or partial genomic sequences are now available for 37 species/strains belonging to the PVC group (see Table 1). For phylogenetic analyses, sequences for 16 housekeeping and ribosomal proteins (ArgRS, EF-G, EF-Tu, GyrA, GyrB, DnaK, IleRS, RecA, RpoB, RpoC, TrpRS, UvrD, ValRS along with ribosomal proteins L1, L5, and S12) were utilized. The protein sequences for various species of the PVC group and for species from some other bacterial phyla were retrieved from the NCBI protein database and their alignments were constructed using the ClustalX 1.83 program (Jeanmougin et al., 1998; NCBI protein database, 2012). After concatenation of all of these sequence alignments into a single file, the poorly aligned regions were removed using the Gblocks_0.91b program (Castresana, 2000). The remaining 7016 aligned and homologous characters were employed for construction of phylogenetic trees using the neighbor-joining (NJ) and maximum likelihood (ML) algorithms as described in our earlier work (Gupta and Mok, 2007; Gupta and Bhandari, 2011; Naushad and Gupta, 2012).

Identification of CSIs that are specific for the PVC group of species was carried out using similar procedures as described in our earlier work (Griffiths et al., 2005; Gupta and Bhandari, 2011; Naushad and Gupta, 2012). Briefly, BlastP searches were initially conducted on various proteins from the genomes of Opitutus terrae (van Passel et al., 2011a) and Pirellula staleyi (Clum et al., 2009) and sequences for 10–12 species that included assorted species from the PVC group and some from other phyla were retrieved. Sequence alignments for these proteins were created and manually examined for inserts or deletions that were flanked on both sides by conserved regions (Gupta and Griffiths, 2002; Gupta and Bhandari, 2011; Naushad and Gupta, 2012). A second, more detailed BlastP search was then carried out on the identified sequence consisting of the indel and the conserved flanking region. The indels that were specific for the members of the PVC group were formatted into signature files showing the sequence alignments and GenBank identifier (GI) numbers of various proteins.

The PVC superphylum is proposed to be composed of numerous species that are part of four phyla and three candidate phyla. With several cellular features unique to members of this group of bacteria as well as the important pathogenic organisms present within this group, the relationships that these bacteria share with other prokaryotes and with each other is of great evolutionary interest (Devol, 2003; Sachse et al., 2009; Fuerst and Sagulenko, 2011; McInerney et al., 2011). However, elucidation of the relationships among the PVC group of bacteria has thus far proven difficult and led to contradictory results by phylogenetic means. In this work, we report for the first time identification of molecular markers in the form of CSIs and CSPs that are unique and distinctive characteristics of species from the phyla Verrucomicrobia and Planctomycetes and others that provide independent support for the grouping of species from the phyla Planctomycetes, Verrucomicrobia, Chlamydiae, and Lentisphaerae into larger clades. Large numbers of CSIs and CSPs for the Chlamydiae species were identified in our earlier work (Griffiths et al., 2005, 2006; Gupta and Griffiths, 2006). Based upon the species distribution patterns of these markers, the evolutionary stages where the genetic changes responsible for them have likely occurred are depicted in Figure 7.

Grounded upon the identified markers, it is now possible to clearly distinguish species from each of the three main phyla (viz. Planctomycetes, Verrucomicrobia, and Chlamydiae) that comprise the PVC clade of bacteria in molecular terms. The specificities of these markers for the species from these clades provide independent evidence for the monophyly of these clades. Additionally, based upon these molecular markers a number of relationships within these bacterial phyla can also be consolidated. Within Verrucomicrobia, newly identified CSIs allow the species from the class Opitutae and family Opitutaceae to be distinguished in molecular terms. The species distribution of these CSIs strongly indicate that the species V. bacterium DG1235, which is currently a part of the class Verrucomicrobiae, should in fact be transferred to the class Opitutae. A number of CSIs also provide evidence that the two unclassified species belonging to the family Opitutaceae viz. O. bacterium TAV5 and TAV1 are closely related to D. colitermitum and they should perhaps be assigned to the genus Diplosphaera. Within Planctomycetes, the species distribution pattern of the identified CSIs strongly indicates that the anammox species K. stuttgartiensis constitutes the deepest branching lineage of this phylum, which is consistent with its branching in the phylogenetic tree. However, this inference is at variance with the current assignment of K. stuttgartiensis to the class Planctomycetia, whereas the species Ph. mikurensis which branches less deeply than K. stuttgartiensis is part of a separate class (Phycisphaerae). The anammox organisms such as K. stuttgartiensis possess a number of distinctive features such as the presence of an ammonium oxidizing organelle called the anammoxosome and cell division by constrictive binary fission, which differentiate them from other members of the class Planctomycetia (van Niftrik et al., 2009).

More importantly, in the present work, we have also identified some signatures that are helpful in clarifying how the species from the PVC phyla of bacteria are related and providing some evidence supporting their amalgamation into larger clades. However, only a couple of signatures that are helpful in this regard were identified. The most significant of these signatures is a 3 aa long insert in the RpoB protein that is commonly and uniquely shared by all of the sequenced Chlamydiae, Verrucomicrobia, and Lentisphaerae species but not found in any other bacteria. The observed species specificity of this signature, in this important protein, strongly indicates that the species from these three phyla shared a common ancestor exclusive of all other bacteria. The RpoB protein also contains a number of other CSIs in other regions of the protein that are specific for other groups/phyla of bacteria (Griffiths and Gupta, 2007; Gupta and Mok, 2007; Gao et al., 2009; Gupta and Bhandari, 2011). The high degree of specificity of these CSIs for different groups/phyla of bacteria provides evidence that the gene for RpoB has not been laterally transferred among different bacterial groups. An other signature that is informative in this regard consists of a small protein of unknown function that is specifically found in all of the species from the above three phyla of bacteria and also in the Planctomycetes. The observed species specificity of this protein suggests that the gene for this protein very likely originated in a common ancestor of the PVC clade of bacteria. However, in this case other possibilities to account for the species distribution of this protein cannot be entirely excluded. Nonetheless, the unique shared presence of this protein by various species that are part of the PVC clade provide evidence supporting their grouping into a large clade.

The molecular markers described in the present work, in addition to their usefulness for evolutionary and taxonomic studies, also provide novel and valuable tools for the identification of these organisms in different environments. In view of the presence of the identified CSIs in conserved regions of various proteins, degenerate primers based upon conserved regions in them can be designed for selective amplification (detection) of sequences from various species from these groups. Additionally, blast searches with the sequence queries based upon these proteins also provide useful identification tools for detection of both known and unknown species from these phyla in metagenomic sequences. Finally, the identified CSIs and CSP provide novel tools for genetic and biochemical studies and functional studies on them could lead to discovery of novel biochemical and/or physiochemical properties that are commonly shared by these phyla or the PVC clade of bacteria.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.