Keywords

1 Introduction

Horizontal gene transfer (HGT) is one of the important mechanisms for rapid bacterial evolution and adaptation. HGT proceeds mainly by conjugation (Frost et al. 2005; Heuer and Smalla 2007; Aminov 2011) and is mediated by mobile genetic elements (MGEs), which are DNA segments that can move between bacterial cells (intercellular mobility) (Frost et al. 2005). The elements carry various kinds of genes, such as antibiotic resistance genes, virulence genes, and catabolic genes, and thus, MGEs are important “vehicles” of pathogenically- and environmentally-relevant traits. Plasmids, integrative and conjugative elements (ICEs), and transposons are the important MGEs. Since the 1960s, various bacteria capable of degrading “xenobiotic” compounds have been isolated because of their potential importance in the removal of these compounds from contaminated environments. In this review, we use the term “xenobiotic” compounds in a broad sense to signify compounds that are not natural to the environment, but are rather “guest” chemicals, as defined by Leisinger (1983). The genes involved in catabolic turnover of xenobiotic compounds are sometimes identified on MGEs, especially on plasmids, ICEs, and transposons.

Plasmids are circular or linear extrachromosomal replicons, which are often transmissible by conjugation (Sota and Top 2008; Frost et al. 2005). ICEs are also self-transmissible conjugative elements, but they are generally integrated into the host chromosome (Burrus and Waldor 2004; Wozniak and Waldor 2010). Conjugation can spread genetic elements among bacteria effectively (Guglielmini et al. 2011), and therefore, it is one of the most important mechanisms for rapid evolution and adaptation of bacteria. On the other hand, transposons are genetic elements that are mobilized and transferred between replicons by the activity of a transposase (Mahillon and Chandler 1998). Once transposons integrate into plasmids or ICEs, they can also be transferred into other cells (Frost et al. 2005). Insertion sequences (IS) are a transposons that carry only the transposase gene, and homologous recombination between multiple copies of the same IS element can promote genomic rearrangements (Mahillon and Chandler 1998).

Although many reviews have been published on MGEs that carry catabolic genes for xenobiotic compounds (Tan 1999; Top et al. 2000, 2002; Top and Springael 2003; van der Meer and Sentchilo 2003; Nojiri et al. 2004; Dennis 2005), a large number of new catabolic MGEs have since been reported due to the recent revolution in nucleotide sequencing technology. This chapter summarizes recent studies of major and/or new MGEs that carry catabolic genes, and briefly describes their features.

2 Catabolic Plasmids

Plasmids have been classified into incompatibility (Inc) groups on the basis of their replication and partition systems. When two different plasmids cannot be maintained in the same bacterial cell line, these two plasmids are called “incompatible” and are considered to belong to the same “Inc” group. There are 27 Inc groups for the Enterobacteriaceae (Carattoli 2009), at least 14 groups for the Pseudomonas (Thomas and Haines 2004), and around 18 groups for the gram-positive bacteria (Frost et al. 2005; Sota and Top 2008), although these groupings do not include all the identified plasmids such as plasmids in Sphingomonas. Recently, a new classification of plasmids was proposed, which is based on their transfer systems generally composed of two sets of proteins for mating pair formation (MPF) and mobilization (MOB) (Smillie et al. 2010; Garcillán-Barcia et al. 2009, 2011). Combination of four types of MPFs (MPFF, MPFI, MPFG, and MPFT) and six classes of MOBs (MOBF, MOBH, MOBQ, MOBC, MOBP, and MOBV) enables us to classify a larger number of plasmids whose sequences have been deposited in DNA databases.

Features of major catabolic plasmids, such as host, growth substrate of host, Inc groups, MOB classes and MPF types, and transferability, are listed in Tables 8.1, 8.2, 8.3 and 8.4. They have been identified in bacteria of the phylum Proteobacteria, such as Pseudomonas (γ-proteobacteria), Achromobacter (β-proteobacteria), and Sphingomonas (α-proteobacteria), and in gram-positive bacteria such as Arthrobacter, Flavobacterium, and Rhodococcus, since the 1970s. Because detailed features of IncP-1, IncP-7, and IncP-9 group plasmids have been already described in our previous review (Shintani et al. 2010), we focused especially on the catabolic plasmids in sphingomonads or gram-positive bacteria in this chapter.

Table 8.1 Catabolic plasmids from genus Pseudomonas and those belonging to Pseudomonas incompatibility groupsa
Table 8.2 Catabolic plasmids in sphingomonads
Table 8.3 Catabolic plasmids in other gram-negative bacteria
Table 8.4 Catabolic plasmids in gram-positive bacteria

2.1 Catabolic Plasmids from Genus Pseudomonas and Those Belonging to Pseudomonas Incompatibility Groups

Many catabolic plasmids are classified into the IncP-1, IncP-2, IncP-7, and IncP-9 groups, which carry genes involved in the degradation of various xenobiotic compounds, such as those for toluene/xylene (xyl), (chloro)benzoate (cba), (chloro)aniline (dca), 2,4-dichlorophenoxyacetic acid (2,4-D) (tfd), naphthalene (nah), and carbazole (car), amongst others (Table 8.1). The complete nucleotide sequences of several plasmids in these groups, except for the IncP-2 plasmids, have been determined, and an Inc group-specific plasmid backbone was proposed by comparative analyses (Fig. 8.1). Dennis (2005) compared the genetic organization of IncP-1 plasmids and showed that most catabolic genes (or other genes, such as antibiotic resistance genes) of IncP-1 plasmids were inserted between the trfA and oriV regions and the parA and tra operons (Fig. 8.1a; Dennis 2005). Sota et al. (2007) showed that the structural similarity of IncP-1 plasmids was a result of both the region-specific insertion of transposons and the selective pressure for maintaining transferability and stability of the plasmids. Based on the comparisons of the nucleotide sequences of plasmids, conserved regions of IncP-9 and IncP-7 plasmids (i.e., a plasmid backbone) were also proposed (Fig. 8.1b, c; Sota et al. 2006; Yano et al. 2010). One important difference between these plasmids is their host range. IncP-1 plasmids are known to be broad host range plasmids that can transfer among bacteria belonging to different classes, such as α-, β-, and γ-proteobacteria. Indeed, the host range of IncP-1 catabolic plasmids is broad, as listed in Table 8.1. As for the IncP-7 and IncP-9 plasmids, their host ranges are narrower than that of the IncP-1 plasmids, and most of their hosts belong to γ-proteobacteria, and in particular, to the genus Pseudomonas (Table 8.1).

Fig. 8.1
figure 1figure 1

Proposed IncP-1 (a), IncP-7 (b), and IncP-9 (c) backbones in conjunction with the insertion sites (vertical arrows) of catabolic genes on each plasmid (see Table 8.1). Horizontal white arrows indicate genes for replication and stability of the plasmid, and those in black indicate genes for conjugative transfer. The ssb gene of the IncP-9 backbone is shown in gray. The tra/trh genes of the IncP-7 backbone were deleted in pND6-1 and pWW53

On the other hand, many other catabolic plasmids have been isolated from Pseudomonas. However, the nucleotide sequences of replication or transfer regions for these plasmids are not available, and therefore, it is difficult to classify these plasmids. One exception is pCT14, which carries several genes for a meta cleavage pathway for aromatic rings, including cbzTEXG, bphK, and tdnG (Bramucci et al. 2006). Although the gene encoding its replication protein and the oriV region were proposed, there are no genes of similar sequence in the GenBank/EMBL/DDBJ database; this plasmid is predicted to be of the MOBF class (Table 8.1).

2.2 Catabolic Plasmids of Sphingomonads

Over the past decade, many catabolic plasmids from xenobiotic-degrading sphingomonads (genera Sphingomonas, Sphingobium, Novosphingobium, and Sphingopyxis) belonging to the class α-proteobacteria, have been identified (Table 8.2). pNL1 was isolated from Novosphingobium aromaticivorans DSM 12444 (its previous name was N. aromaticivorans F199), and it is the first catabolic plasmid in sphingomonads whose 184-kb nucleotide sequence has been reported (Romine et al. 1999). Some xenobiotic-degrading sphingomonads carry multiple plasmids in one cell (Basta et al. 2004; Cérémonie et al. 2006; Tabata et al. 2011).The strain DSM 12444 also carries another plasmid of 487 kb, pNL2 (Fredrickson et al. 1991). Basta et al. (2005) compared plasmids from 16 sphingomonad strains that degrade various polycyclic aromatic hydrocarbons (PAHs). Based on Southern blot analyses, a plasmid of the naphthalenesulfonate-degrader Sphingomonas xenophaga BN6 and a plasmid of the dibenzofuran-degrader Sphingomonas sp. HH69 were shown to possess a pNL1-type Rep (replication initiation protein) gene (Basta et al. 2005). Nucleotide sequence comparisons revealed that similar Rep genes were also found in pCAR3, which also carries car genes, in the carbazole-degrader Novosphingobium sp. KA1 (its previous name was Sphingomonas sp. KA1, Shintani et al. 2007), and in pSWIT02, which also carries dxn genes, in the dibenzo-p-dioxin degrader Sphingomonas wittichii RW1 (Miller et al. 2010). The Rep type is classified based on the amino acid sequence identity (>70%) of putative Rep gene products of each sequenced plasmid.

Notably, many plasmids were identified in γ-hexachlorocyclohexane (γ-HCH)-degrading sphingomonads (Table 8.2, Nagata et al. 2007). Sphingobium japonicum UT26 is an archetypal γ-HCH-degrading bacterium, and its whole genome sequence has been determined (Nagata et al. 2010, 2011). This strain has three plasmids, and one of them is the 191-kb pCHQ1, which carries linRDEB (Nagata et al. 2007, 2010, 2011). No Inc groups have been suggested for plasmids from sphingomonads; however, several types of Rep genes are known to be conserved among these bacteria. Indeed, there are other plasmids in sphingomonads that contain genes which show high identities with the Rep gene of pCHQ1 (Table 8.2): pLA1, which was identified in a PAHs-degrader, Novosphingobium pentaromativorans US6-1, and carries bph and xyl genes involved in biphenyl and toluene/xylene degradation (Luo et al. 2012); pSLGP in a lignin-degrader, Sphingobium sp. SK-6 (Masai et al. 2012); and pSPHCH01, in a pentachlorophenol-degrader, Sphingobium chlorophenolicum L-1 (Copley et al. 2012). The last two plasmids, however, do not carry catabolic genes. pLB1 also carries the linB gene, which was identified by performing an exogenous plasmid isolation technique from γ-HCH-contaminated soil using a linB-disrupted UT26 mutant. The original host of pLB1 was unidentifiable, but the plasmid can transfer to Sphingobium japonicum UT26 (Miyazaki et al. 2006). The Rep type of pLB1 is different from that of pCHQ1 because it shows compatibility to pCHQ1 (Miyazaki et al. 2006). Similarly, in addition to pLA1 (pCHQ1-type), N. pentaromativorans US6-1 harbors another plasmid, pLA2, which carries the pLB1-type Rep gene but has no catabolic genes. The conservation of the Rep genes suggests that many plasmids in sphingomonads may be self-transmissible, although this property has been experimentally proved to exist in only a few (Table 8.2).

Plasmids belonging to the same Pseudomonas incompatibility groups always have the same types of genes for conjugative transfer (Table 8.1). In contrast, plasmids in sphingomonads have different types of genes for conjugative transfer, whereas they have the same Rep genes, suggesting that they have “mosaic” genetic structures. While the Rep gene of pSWIT02 is pNL1-like, the genes for plasmid transfer show higher similarity to those of pCHQ1 than to those of pNL1. On the other hand, putative plasmid transfer genes of pLA1 are more similar to those of pNL1 than to those of pCHQ1, while its Rep gene is more similar to that of pCHQ1 (Luo et al. 2012). In addition, several catabolic genes, such as bph on pNL1, car on pCAR1, or lin on pCHQ1 are not organized in a single operon but dispersed on the plasmid or host chromosome in sphingomonads (Romine et al. 1999; Shintani et al. 2007; Nagata et al. 2011). The varied distribution of similar genes and dispersed organization of genes indicate that catabolic plasmids in sphingomonads might have been transferred among the genus, and might have undergone DNA rearrangements with other plasmids and host chromosomes, resulting in the “mosaic” structure.

2.3 Catabolic Plasmids of Other Gram-Negative Bacteria

Catabolic plasmids have also been observed in other gram-negative bacteria belonging to classes α-, β-, and γ-proteobacteria, as listed in Table 8.3, although they have not been investigated in detail. The whole genome sequence of the naphthalene-degrading Polaromonas naphthalenivorans CJ2 has been determined (Jeon et al. 2003, 2006; Yagi et al. 2009). This strain possesses eight plasmids, and at least two of them, pPNAP01 and pPNAP04, carry putative aromatic hydrocarbon-degradative genes (Yagi et al. 2009). The partial sequence of pCMS1, the organophosphate degradative plasmid of Brevundimonas diminuta MG, revealed that its putative transfer genes showed 67–74% identity with those of the IncP-1 plasmid pEST4011 (Pandeeti et al. 2011). This fact implied an evolutionary relationship between pCMS1 and IncP-1 plasmids. Analysis of the nucleotide sequences and identification of open reading frames on these plasmids will be important for elucidating the steps in the evolution of these plasmids in gram-negative bacteria.

2.4 Catabolic Plasmids of Gram-Positive Bacteria

Several plasmids have been identified in xenobiotic-degrading gram-positive bacteria belonging to classes Actinobacteria, Bacilli, and Flavobacteriia (Table 8.4). Some of these bacteria carry circular plasmids and others harbor linear plasmids (Table 8.4). The linear plasmids belong to a class of genetic elements called invertrons, which carry terminal inverted repeats (TIRs) that are covalently bound to terminal proteins at both 5′ termini (Sakaguchi 1990). Linear plasmids have been proposed to have evolved from bacteriophages (Hinnebusch and Tilly 1993). The details of the mechanisms of plasmid transfer between gram-positive bacteria are still unclear (Grohmann et al. 2003).

Rhodococcus is one of the most important genera among gram-positive degraders of alkanes, PCBs, and naphthalene, and many plasmids have been identified in the Rhodococcus species (Table 8.4). pBD2 is a conjugative linear plasmid that carries ipb genes for the catabolism of isopropylbenzene, and it was detected in R. erythropolis BD2 (Dabrock et al. 1994; Stecker et al. 2003). pREL1 and pREC1 were identified in R. erythropolis PR4, an alkane-degrader (Sekine et al. 2006). Several DNA regions in pREL1 and pBD2 are conserved, including genes that encode for terminal protein, lipoproteins, and heavy metal resistance. However, the degradative genes for alkane (pREL1) and for isopropylbenzene (pBD2) are not conserved (Sekine et al. 2006).

R. jostii RHA1 can degrade polychlorinated biphenyls (PCBs) (Seto et al. 1995), and its complete genome sequence has been determined (McLeod et al. 2006). This strain harbors three linear plasmids, pRHL1, pRHL2, and pRHL3 (Shimizu et al. 2001; Masai et al. 1997), and most of the genes involved in the biphenyl degradative pathway are located on the two larger plasmids, pRHL1 and pRHL2 (Shimizu et al. 2001). Notably, many catabolic isozyme genes are distributed throughout the RHA1 genome (Kitagawa et al. 2001; Sakai et al. 2002; McLeod et al. 2006). The four replicons of RHA1, including the three plasmids and its linear chromosome, were suggested to be similar types of linear elements, because their TIRs are highly similar (McLeod et al. 2006).

Arthrobacter utilizes a wide and varied range of xenobiotic compounds and several catabolic plasmids have been identified in this genus (Table 8.4). pAL1 is a linear catabolic plasmid that was detected in the 2-methylquinoline-degrading Arthrobacter nitorguajacolicus Rü61a strain (Parschat et al. 2007; Overhage et al. 2005). The replication region of pAL1 was analyzed in detail, and it revealed that this plasmid carries a novel Rep gene (Kolkenbrock et al. 2010; Wagenknecht and Meinhardt 2011). Parschat et al. (2007) showed that several regions of pAL1 are conserved in pAL1 and the pBD2, pREL1, and pRHL2 plasmids mentioned above, and also in the dibenzofuran-degradative plasmid pDBF1 from Terrabacter sp. DBF63 (Nojiri et al. 2002; Habe et al. 2005). One of the regions includes putative genes for a secretion system possibly involved in conjugation (Parschat et al. 2007). Similarly, 2,3-dihydroxybiphenyl dioxygenase BphC genes are conserved on pLP6 and pTSA421 found in R. globerulus P6 and R. erythropolis TA421 (Kosono et al. 1997).

Other types of catabolic plasmids have also been reported (Table 8.4). pLW1071 is a circular plasmid from Geobacillus thermodentrificans NG80-2 that carries degradative genes for long-chain alkanes (Feng et al. 2007). This plasmid is unique in comparison to other sequenced plasmids, except for a plasmid from Geobacillus sp., G11MC16 (accession no. NZ_ABVH01000017). The putative Rep gene of the plasmid of G11MC16 was similar to that of NG80-2. pGKT2 is a 182-kb circular plasmid carrying xplAB genes found in the hexahydro-1,3,5-trinitro-1,3,5-triazine degrader Gordonia sp. KTR9 (Indest et al. 2010). Gordonia spp. are a metabolically diverse group, with regards to their ability to degrade xenobiotic compounds, and recently, two other catabolic plasmids have been reported in this genus (Table 8.4). Catabolic genes in gram-positive bacteria may also be spread by self-transmissible plasmids (listed in Table 8.4), similar to that observed in the case of gram-negative plasmids, and have an important role in their HGT, although their host range remains unclear.

3 Catabolic Transposons

In some cases, catabolic genes are flanked by two copies of the same or highly-identical insertion sequences (ISs). These elements are known as composite transposons. Tn5280 (van der Meer et al. 1991a), TnHadI (Kawasaki et al. 1985); Sota et al. 2002), and DEH (Weightman et al. 2002) are composite transposons whose transposition ability has been experimentally validated (Table 8.5). As for Tn-DhaI, it encodes pcrABCT which is involved in reductive dechlorination of tetrachloroethene in Desulfitobacterium hafniense TCE1, and detection of the circular form of the transposon strongly indicated that it could transpose (Maillard et al. 2005). As genome sequences of an increasing number of xenobiotic-degrading bacteria are determined, many composite transposon-like genetic structures are being discovered (Table 8.5). Homologous recombination events among several copies of the identical ISs located on regions surrounding catabolic genes possibly increase the plasticity of the genome. There are two kinds of ISs, IS6100 and IS1071, which were frequently associated with various catabolic genes. IS6100 was originally isolated as part of the composite transposon Tn6100 from Mycobacterium fortuitum (Martin et al. 1990), and was found in a wide range of host bacteria, such as Sphingomonas (Dogra et al. 2004), Arthrobacter (Kato et al. 1994), Pseudomonas (Hall et al. 1994), Xanthomonas (Sundin and Bender 1995), Salmonella (Boyd et al. 2000), and Corynebacterium (Tauch et al. 2002). The IS elements were also found in many kinds of xenobiotic-degrading bacteria, and some of them form composite transposon-like structures (Table 8.5). IS6100 was found in many γ-HCH-degrading sphingomonads in the region flanking the lin genes involved in γ-HCH-degradation, suggesting that this IS may have played a key role in the recruitment of the lin genes in these bacteria (Nagata et al. 2011).

Table 8.5 Catabolic transposons

IS1071 was originally identified in a chlorobenzoate-catabolic transposon, Tn5271, from Comamonas testosteroni BR60 (Nakatsu et al. 1991). IS1071 belongs to the class II transposons, which generally carry the genes for their transposition (tnpA, tnpR, and res) and one or more phenotypic traits between their terminal inverted repeats (Grindley 2002). This type of transposon generates a cointegrate of donor and target molecules, and the cointegrate is then resolved at the resolution (res) sites by TnpR (resolvase). This resolution function, however, is lacking in IS1071. The copy number of class II transposons doubles after their transposition by means of a mechanism known as “copy and paste” transposition (Grindley 2002). Many IS1071 sequences have been identified in close proximity to various xenobiotic-degradative genes on self-transmissible plasmids from environmental bacteria (Table 8.5). These data indicate that IS1071 might have been involved in the recruitment of catabolic genes to these plasmids and in the dissemination of these genes among various host strains.

It should be noted that some class II transposons (Grindley 2002) that carry catabolic genes are found in various xenobiotic-degrading bacteria (Table 8.5). In addition to the extensively characterized Tn4651/Tn4653 in the toluene/xylene-degradative plasmid pWW0 (IncP-9) (Tsuda and Iino 1987, 1988; Tsuda et al. 1989), these types of transposons are found in two other toluene/xylene-degradative plasmids, namely pWW53 (IncP-7) and pDK1 (IncP-7), the carbazole degradative plasmid pCAR1 (IncP-7), and the naphthalene degradative plasmid NAH7 (IncP-9). Notably, the transposition function of most of these transposons has been experimentally verified (Table 8.5, Yano et al. 2007, 2010; Shintani et al. 2005, 2011). Although Tn4655 in NAH7 lacks the tnpA gene (Sota et al. 2006), it is able to form a cointegrate when the tnpA gene of Tn4653 is supplied in trans (Tsuda and Iino 1990; Sota et al. 2006). These class II transposons might have been efficiently spread among bacterial replicons via their “copy and paste” transposition, and they can carry longer DNA regions than class I composite transposons can.

4 Catabolic ICEs

ICEs are self-transmissible MGEs that are integrated in the chromosome. These elements carry genes for conjugative transfer and also excision systems to excise from the chromosome (Burrus and Waldor 2004; Wozniak and Waldor 2010). They are replicated as a part of the chromosome, they excise from the chromosome, circularize and then transfer to new hosts, sometimes leading to the integration into these new host chromosomes (Burrus and Waldor 2004; Wozniak and Waldor 2010). ICEs are difficult to identify experimentally, because they are usually physically linked to the host chromosome (Wozniak and Waldor 2010). ICE clc (Ravatn et al. 1998a), bph-sal element (Nishi et al. 2000), and ICEKKS 4677 (Ohtsubo et al. 2003, 2006, 2012) are the ICEs that have been verified experimentally (Table 8.6). Among these, the most in-depth analyses, such as on the mechanisms for excision, transfer, and impact on the host cell, have been performed for ICE clc (Ravatn et al. 1998a, b; Gaillard et al. 2006, 2008, 2010; Sentchilo et al. 2009; Miyazaki and van der Meer 2011a, b).

Table 8.6 Catabolic ICEs

Recently, in silico analyses of complete bacterial genomes have identified putative ICEs in several β- and γ-proteobacteria. Indeed, such analyses of many complete bacterial genomes showed that ICEs are spread among various bacterial subdivisions, and more than 400 putative ICEs are listed in ICEberg (http://db-mml.sjtu.edu.cn/ICEberg/) (Bi et al. 2012). Ryan et al. (2009) reported that an ICETn4731 -related ICE was found in several bacterial genome sequences, and one of them, ICETn4371 6065, carrying the bph gene, was found in a naphthalene degrader, Polaromonas naphthalenivorans CJ2. Interestingly, Bordetella petrii DSM 12804 possesses at least seven large ICEs mostly encoding metabolic functions involved in the degradation of aromatic compounds and detoxification of heavy metals (Lechner et al. 2009). Four of them, ICE-GI1, ICE-GI2, ICE-GI3, and ICE-GI6, are closely related to ICE clc , and the first three carry putative catabolic genes (Table 8.6). It should be noted that their circular intermediates have been detected, and that transmissibility of ICE-GI3 has been confirmed (Lechner et al. 2009). Hichey et al. found a new ICE in the genome of the PAHs-degrader, Delftia sp. Ds1-4, which carries all of the required phenanthrene catabolic genes (Hickey et al. 2012). Because ICEs are not necessarily replicated as circular forms after their integration into the host chromosome, host ranges of ICEs are not dictated by whether the ICEs can be replicated in the host cells. Therefore, their host ranges are likely to be wider than that of other MGEs.

5 Behaviors of Catabolic MGEs

Bioaugmentation by inoculation of highly efficient xenobiotic degraders into polluted sites has been studied as an attractive approach to remove pollutants. However, it is difficult to maintain the high levels of degradative ability of these inoculants, because they are not necessarily able to compete or survive in natural environments (Top et al. 2002). The catabolic MGEs, especially conjugative elements, can be used in alternative bioaugmentation by utilizing the transferability of MGEs into the indigenous bacteria in the polluted sites. In bioaugmentation via inoculation with degraders harboring MGEs, known as “gene bioaugmentation” or “plasmid-mediated bioaugmentation,” the survival of the inoculated degraders is not needed (Bathe 2004; Bathe et al. 2005; Dejonghe et al. 2000; Pepper et al. 2002). There are still, however, large gaps between laboratory conditions and natural systems, and the basic features of MGEs in laboratory conditions do not necessarily reflect their actual behavior in natural systems. Many trials have been conducted to bridge the differences between these conditions by using artificial model environments, which model natural habitats such as soil, plants, and water. While the behaviors of the IncP-1, P-7, and P-9 group plasmids have been summarized recently (Shintani et al. 2010), those of other plasmids, which belong to unknown Inc groups, have been also reported. Detailed analyses have been performed to analyze the effect of conjugative transfer of two kinds of 2,4-D degradative plasmids in soil by using pEMT1 and IncP-1 plasmid pEMT3 in different donors (Top et al. 1995; Dejonghe et al. 2000; Goris et al. 2002). Top et al. (2002) concluded that these catabolic plasmids were most often transferred to, and their genes expressed in, strains that belong to the genera Burkholderia, Ralstonia, and Pseudomonas. Transfer of the plasmid pTOM carrying constitutively transcribed toluene-degradative genes (tom) was shown from Burkholderia cepacia to different endogenous endophytic bacteria in yellow lupine (Barac et al. 2004) or poplar cuttings (Taghavi et al. 2005). Springael et al. reported that ICE clc (B13) of P. putida BN210 was transferred to different bacteria belonging to the class of β-proteobacteria in biofilm reactors under non-sterile conditions (Springael et al. 2002).

These studies, together with those of IncP-1, P-7 and P-9 plasmids, strongly indicate that HGT by means of catabolic MGEs generally occurs in natural environments. Nevertheless, it is still difficult to predict how the catabolic plasmids or their hosts behave in these environments. A more in-depth understanding of HGT of MGEs will be required for practical application of plasmid-mediated bioaugmentation. Behaviors of the MGEs should be analyzed in microbial communities that include uncultivated and non-cultivable bacteria in natural environments. Several cultivation-independent methods to monitor the behavior of environmental bacteria have been reported. Metagenomic analysis combined with reverse-transcriptase real-time PCR analysis revealed the changes in the bacterial community and in abundant functional genes in contaminated environments (Yergeau et al. 2012). Ishii et al. (2011) identified the active N2O reducers in rice paddy soil using stable isotope probing and functional single-cell isolation by micromanipulation. In another study, fluorescence-activated cell sorting (FACS) and micromanipulation enabled the identification and cultivation of independent plasmid transconjugants (Musovic et al. 2006, 2010). The combinations of these cultivation-independent and cultivation-dependent methods will shed light on HGT in microbial communities in various natural environments.

6 Conclusion and Perspectives

As an increasing number of whole genome sequences of bacteria capable of degrading various kinds of xenobiotic compounds are analyzed, a large number of catabolic MGEs have been discovered and studied recently. In silico analyses of the genome sequences of these bacteria enable us to detect new ISs and ICEs; however, experimental confirmation of their ability to mobilize is still required to further our understanding of how they are transmitted among bacteria or replicons. On the other hand, nucleotide sequence information on other Inc group plasmids from Pseudomonas, such as IncP-2 or other plasmids not affiliated to any Inc group (Table 8.1), is also required for further classification of the newly-identified plasmids.

Jones and Marchesi (2007) developed a method for transposon-aided capture of plasmids to discover novel plasmids in various bacterial habitats. This method allowed them to identify plasmids that did not rely on the plasmids’ own replication and transfer systems. Indeed, many novel MGEs have been identified in various sites by the method mentioned above and by metagenomic analyses, such as in activated sludge (Zhang et al. 2011a), river or sea sediments (Elsaied et al. 2011; Kristiansson et al. 2011), wastewater treatment plants (Szczepanowski et al. 2008), human dental plaque (Warburton et al. 2011), and human gut (Jones et al. 2010). These reports suggest that a huge number of unidentified MGEs exist in the environment. Detection and analyses of new catabolic MGEs will help us to understand the mechanism by which MGEs spread and also determine which MGEs are capable of spreading in natural bacterial communities, including those that contain uncultivated and non-cultivable bacteria. These MGEs can possibly be used as new tools for genetic analysis of unidentified bacteria.