Most commonly used nutrient media have a near-neutral pH and a salt content of 1–3 g L⁻¹. This contrasts dramatically to acidic (pH 3.5–5.5) peat water with a mineral salt content of 5–50 mg L⁻¹ and explains why most peat-inhabiting bacteria do not develop on conventional media. Therefore, a more accurate simulation of the peat bog environment in the laboratory requires dilute (1:10–1:100) and acidic (pH 4.0–5.5) media. High potential of this strategy was demonstrated by isolation of novel acidophilic methane-oxidizing bacteria from northern wetlands (Dedysh et al., 1998). Later, a number of peat-inhabiting methanotrophs were isolated on mildly acidic, low-ionic-strength media (Dedysh, 2009; Kip et al., 2011). Though methanogenic archaea are beyond the scope of this review, they have also been successfully retrieved from acidic wetlands using strongly dilute, acidic media (Sizova et al., 2003; Bräuer et al., 2006; Kotsyurbenko et al., 2007; Cadillo-Quiroz et al., 2009). The same strategy was applied for isolation of heterotrophic bacteria from poorly studied phyla. For example, peat-inhabiting members of the Acidobacteria were cultured using low-ionic-strength, low-nutrient media MM, MM1, or 10-fold diluted R2A (Dedysh et al., 2006; Pankratov et al., 2008). One of these media, MM1, does not contain phosphates since they were shown to inhibit growth of some acidobacteria (Pankratov and Dedysh, 2010). In this case, medium pH was adjusted to 4.0–5.0 with alginic acid, which offers a unique possibility of decreasing medium pH without increasing its ionic strength.

Most microbes that thrive in cold and nutrient-poor northern wetlands are slow-growing organisms. Even under optimal growth conditions, isolation of these microbes requires long incubation. Colonies of peat-inhabiting methanotrophs, acidobacteria, planctomycetes, and other fastidious bacteria appear on solid media only after 4–8 weeks of incubation (Dedysh et al., 2000; Pankratov et al., 2008, in press; Kulichevskaya et al., 2009). In the case of the acidotolerant facultative anaerobe Telmatospirillum sibiriense, development of colonies was observed after 5 months of anaerobic incubation (Sizova et al., 2007).

Most of these difficult-to-culture bacteria produce very small (0.1–0.5 mm in diameter) colonies, which can be observed and picked with the use of a dissecting microscope only. The same phenomenon was reported for several rarely cultured groups of soil bacteria, which were most abundant among the mini-colonies that developed after >12 weeks of incubation (Davis et al., 2011). The use of gellan gum (phytagel, gelrite), which produces a very clear medium, allows for easy discrimination of these mini-colonies (Janssen et al., 2002; Stott et al., 2008). This polysaccharide is also free of contaminants, which may inhibit growth of some microbes. Gellan gum was successfully applied for isolation of diverse peat-inhabiting bacteria, including methanotrophs and acidobacteria (Dedysh et al., 2002, in press; Pankratov et al., 2008, in press). Notably, many of these organisms were unable to grow on agar-containing media. It should be kept in mind, however, that polymerization of gellan gum occurs in the presence of bivalent cations, which are normally present in very low concentrations in ombrotrophic wetlands. The possibility that this may negatively effect isolation of some particular bacterial groups cannot be excluded.

As discussed above, most colonies that develop rapidly on the surface of various solid media are formed by bacteria which are not predominant in the peat bog environment. Many of these colonies, however, represent not a pure bacterial culture but a co-culture of two or more microorganisms. Examination of such mixed colonies obtained from acidic peat revealed that many of them contained cells of rarely cultured organisms, such as acidobacteria or planctomycetes (Dedysh et al., 2006). This observation suggested that an isolation approach based on enrichment of mixed cultures or microbial biofilms might in some cases be more efficient for isolating these elusive bacteria than a routine “single-colony pick-up” strategy. The two simplest versions of this alternative approach are given below.

Microbial biofilms consisting of cells of peat-inhabiting bacteria can be obtained by using a simple technique invented by Schlesner (1994) for isolation of planctomycetes. Briefly, the bottom of a Petri dish is covered with a layer of sterilized water agar, which contains cycloheximide to inhibit growth of fungi. Up to 20 sterilized cover slides are placed vertically and partially into the agar layer. These plates are then inoculated with 10–15 ml of peat water, which serves as the only source of nutrients for developing microorganisms, and incubated for 1–2 months. Every 2–3 weeks, a few slides are taken out and examined for the presence of target cells. In our experience, the best screening tool to recognize target cells, as opposed to easily culturable cells, is FISH. It allows direct visualization of the target bacteria within the entire population of cells present in the biofilm (Figure 3). This greatly simplifies all further isolation/purification procedures. Various PCR-based techniques can also be used for screening purposes, but FISH offers a clear advantage of seeing the target. The examples of successful application of this biofilm-mediated enrichment approach include isolation of peat-inhabiting subdivision 3 acidobacterium Bryobacter aggregatus (Kulichevskaya et al., 2010) as well as cultivation of several mildly acidophilic planctomycetes (Kulichevskaya et al., 2007, 2008, 2009, in press).

Careful inspection of impure cultures that are commonly discarded in routine cultivation practice is another way of obtaining an unusual organism. A good example is the isolation of a difficult-to-culture acidobacterium, Telmatobacter bradus (Pankratov et al., in press). This microorganism is a facultative anaerobe which can grow only under reduced oxygen tension. Despite this fact, T. bradus was recovered from aerobically incubated plates where it developed in a co-culture with an unidentified rapidly growing aerobic alphaproteobacterium. The alphaproteobacterium created a microhabitat with reduced oxygen concentration through its rapid growth. Application of FISH with the Acidobacteria-specific probe HoAc1402 was again the main tool to identify one of the organisms in this co-culture as a target for further isolation work.

So far, only a very few anaerobic prokaryotes have been cultivated from acidic wetlands. One of the factors that may have hampered isolation of anaerobes is the proper choice of a reducing agent for anaerobic media preparation. As shown by Sizova et al. (2003), many conventional reducing agents such as Na₂S + l-cysteine-HCl and ascorbic acid did not support growth of methanogenic consortia from Sphagnum-derived peat, but success was achieved with titanium(III)citrate. Later, a similar approach was successfully applied for isolation of several methanogens from acidic peatlands (Bräuer et al., 2006; Cadillo-Quiroz et al., 2009) as well as for isolation and cultivation of some facultatively anaerobic peat-inhabiting bacteria (Sizova et al., 2007; Pankratov et al., in press).

In summary, the major tools for culturing microbes from acidic northern wetlands are readily available, although they need to be carefully adjusted for each particular target bacterium and site. Nevertheless, the bacterial diversity in acidic northern wetlands remains largely unexplored and should benefit from further development of improved cultivation techniques.

Three different physiological groups of alphaproteobacteria commonly occur in acidic wetlands: (i) methanotrophs, (ii) chemo-organotrophs, and (iii) phototrophs. Of these, methanotrophic alphaproteobacteria have received the most research attention due to their important role in reducing methane emission from northern wetlands (reviewed in Dedysh, 2009).

The most typical representatives of this bacterial group in Sphagnum-dominated wetlands are members of the genus Burkholderia (Opelt and Berg, 2004; Belova et al., 2006; Opelt et al., 2007). The species Burkholderia bryophila was described based on a characterization of a number of strains isolated from Sphagnum mosses (Vandamme et al., 2007). These acidotolerant bacteria utilize various carbon substrates including aromatic compounds, fix N₂ and show high antagonistic potential against fungal pathogens.

The Acidobacteria is one of the most diverse groups of prokaryotes in acidic wetlands (Dedysh et al., 2006). Despite this fact, cultured representatives of peat-inhabiting acidobacteria are now available for subdivisions 1 and 3 only. These are the genera Granulicella (Pankratov and Dedysh, 2010), Telmatobacter (Pankratov et al., in press), and Bryocella (Dedysh et al., in press) in subdivision 1 and the genus Bryobacter in subdivision 3 (Kulichevskaya et al., 2010). All of these characterized acidobacteria are mesophilic or psychrotolerant, acidophilic chemo-organotrophs that grow at pH values between 3.0 and 6.5–7.5. Granulicella spp., Bryocella elongata, and B. aggregatus are strict aerobes. T. bradus is a facultative anaerobe that grows under reduced oxygen tension and under anoxic conditions. The preferred growth substrates are sugars though several organic acids and polyalcohols can also be utilized by some strains. Most peat-inhabiting acidobacteria can utilize glucuronic and galacturonic acids, which are released during decomposition of Sphagnum moss. The ability to degrade various plant-derived polymers, such as pectin, xylan, laminarin, lichenan, and starch varies between different species, but only T. bradus is capable of hydrolyzing cellulose. B. elongata possesses an ability to develop in a co-culture with exopolysaccharide-producing acidophilic methanotrophs, presumably by means of feeding on their capsular material. None of the cultured acidobacteria are capable of C1-metabolism or N₂ fixation. However, members of this phylum seem to play an important role in degrading plant-derived polymers in acidic oligotrophic wetlands.

Planctomycetes represent one of the most abundant bacterial groups detectable by FISH in acidic Sphagnum-dominated wetlands (Dedysh et al., 2006; Kulichevskaya et al., 2006a). Several peat-inhabiting, acidophilic planctomycetes were recently isolated and described as members of the novel genera Schlesneria, Singulisphaera, and Zavarzinella (Kulichevskaya et al., 2007, 2008, 2009, in press). These are mesophilic, moderately acidophilic chemo-organotrophs capable of growth at pH values between 4.2 and 7.5. They grow best in aerobic conditions on media containing carbohydrates or N-acetylglucosamine. However, Schlesneria paludicola is also capable of fermentation, while members of the genus Singulisphaera grow well in micro-oxic conditions. All planctomycetes isolated from peat are able to degrade various heteropolysaccharides, but not cellulose or chitin. None of them is capable of fixing N₂. These cultured representatives, however, do not cover all planctomycete diversity detected in acidic wetlands by means of cultivation-independent approaches (Ivanova and Dedysh, unpublished data). The existence of other, still-unknown physiological types of planctomycetes in wetlands cannot be excluded.

Currently available cultures of peat-inhabiting actinobacteria include several species of the genus Mycobacterium (Kazda and Müller, 1979; Kazda, 1980) and a member of the family Micromonosporaceae, Verrucosispora gifhornensis (Rheims et al., 1998). They utilize sugars, several polyalcohols and organic acids as well as some aromatic compounds, but none of them is capable of cellulose degradation.

In contrast to many aquatic and terrestrial habitats, members of the phylum Bacteroidetes are not abundant in acidic wetlands. The species cultured from acidic peat lack the ability to degrade cellulose or chitin. Two examples of such bacteria are Mucilaginibacter spp. (Pankratov et al., 2007) and Chitinophaga arvensicola (Pankratov et al., 2006). They do, however, possess some hydrolytic potential being able to degrade some heteropolysaccharides, such as xylan, laminarin, or pectin, in acidic and cold conditions.

Until recently, possession of a particulate methane monooxygenase enzyme and a well-developed intracytoplasmic membrane system in which pMMO is bound was considered a characteristic feature of all extant aerobic methanotrophs. The first methanotrophic bacterium that did not meet these criteria was isolated from acidic Sphagnum-dominated wetlands and named Methylocella palustris (Dedysh et al., 2000). Two other recognized species of this genus, M. silvestris (Dunfield et al., 2003) and M. tundrae (Dedysh et al., 2004a) were later isolated from an acidic forest soil and tundra wetlands, respectively. All members of the genus Methylocella lack extensive intracytoplasmic membrane structures typical of most other methanotrophs. The absence of pMMO in M. palustris was initially suggested by the failure to detect a pmoA gene, which encodes the 27-kDa polypeptide of pMMO, by PCR or by DNA probing with pmoA from Methylococcus capsulatus Bath (Dedysh et al., 2000). Analysis of M. silvestris BL2^T by SDS-PAGE detected no pMMO-specific polypeptides, and DNA probing with pmoA gene fragments from two close phylogenetic relatives also found no evidence for pMMO (Theisen et al., 2005). Finally, the absence of any pmo gene homologs in M. silvestris BL2^T was conclusively verified by analyzing the genome sequence of this bacterium (Chen et al., 2010).

The absence of pMMO in Methylocella spp. implies that these bacteria cannot be detected using a pmoA-based PCR assay considered universal and specific for all other methanotrophs (Holmes et al., 1995). However, they do possess the mmoX gene encoding the α-subunit of sMMO and can be detected via retrieval of these genes from the environment. A real-time quantitative PCR method was recently developed and validated targeting Methylocella mmoX that allowed detection and quantification of these unusual methanotrophs in a variety of environmental samples (Rahman et al., 2011). Interestingly, Methylocella spp. were detected not only in acidic or neutral habitats, but also in alkaline environments, suggesting that these methanotrophs are not limited to acidic pH in nature.

The diversity of methanotrophs that lack pMMO and use only an sMMO for methane oxidation appears to be not restricted to Methylocella species. Recently, three novel isolates were obtained from Sphagnum-dominated wetlands and an acidic forest soil and were described as representing a novel genus and species, Methyloferula stellata (Vorobev et al., in press). In these bacteria, the mmoX gene could not be amplified with any of the previously known mmoX-targeted primers, which explains why these methanotrophs escaped detection in all previous cultivation-independent studies. The absence of a universal assay for the specific detection of all pMMO-lacking methanotrophs complicates assessment of their abundance and distribution as well as their contribution to the processes of methane oxidation in northern wetlands and other natural environments. We have only recently begun exploring this new group of pMMO-lacking methanotrophs.

For a long time, all methanotrophs were considered to be obligately methylotrophic, i.e., unable to grow on compounds containing C–C bonds. This notion has recently been revised since the ability to grow on methane as well as on some multicarbon substrates, i.e., facultative methanotrophy, was demonstrated in several methanotrophic bacteria. Interestingly, all currently characterized facultative methanotrophs were isolated from acidic environments, such as Sphagnum-dominated wetlands or acidic boreal forest soils. Unusual, pMMO-lacking methanotrophs of the genus Methylocella were the first to be conclusively shown to have a facultative capability (Dedysh et al., 2005a). In addition to methane, they are capable of growth on acetate, pyruvate, succinate, malate, and ethanol. Acetate and methane were used as model substrates to conclusively verify facultative growth in M. silvestris BL2^T. The growth rate and carbon conversion efficiency of M. silvestris BL2^T was higher on acetate than on methane, and when both substrates were provided in excess acetate was preferably used and methane oxidation shut down. Further detailed experiments demonstrated that transcription of the mmo operon in Methylocella was repressed by acetate (Theisen et al., 2005).

We now know that a facultative lifestyle may also occur in pMMO-possessing methanotrophs. Several members of the genera Methylocapsa and Methylocystis were recently shown to be capable of growing on methane as well as on some multicarbon substrates such as acetate or ethanol (Dunfield et al., 2010; Belova et al., 2011; Im et al., 2011). Unlike Methylocella, these methanotrophs prefer to utilize methane, but growth can also occur on acetate and/or ethanol in the absence of methane. Two of these organisms, M. heyeri H2 and Methylocystis sp. strain H2s, were isolated from acidic Sphagnum peat. The latter was shown to represent a numerically abundant methanotroph population in geographically distinct northern wetlands (Belova et al., 2011).

Research on acidic wetlands has changed our perception of the family Beijerinckiaceae. Former knowledge of this bacterial group characterized its representatives as aerobic, acidophilic chemo-heterotrophs with the ability to fix N₂. During the last decade, a novel sub-group of acidophilic serine pathway methano- and methylotrophs that are phylogenetically closely related to the genus Beijerinckia was discovered in acidic Sphagnum peat bogs and forest soils (Dedysh et al., 1998, 2000, 2002; Dedysh et al., 2004a; Dunfield et al., 2003, 2010; Vorob’ev et al., 2009; Vorobev et al., in press). Moreover, methylotrophic autotrophy was found in one earlier described member of the genus Beijerinckia, Beijerinckia mobilis (Dedysh et al., 2005b). Now this family accommodates bacteria with strikingly different lifestyles, such as chemo-heterotrophs, facultative methylotrophs and also facultative and obligate methanotrophs (Figure 4). The 16S rRNA gene sequence similarity values between these phenotypically distinct bacteria range from 96 to 98%, which makes it impossible to predict the phenotype of any novel member in this family based on 16S rRNA gene sequence information alone. Since Beijerinckiaceae is the only bacterial family containing methanotrophic and non-methanotrophic bacteria, it has been chosen for the ongoing comparative genomic study, which should enable elucidating the genetic and metabolic tradeoffs required for a specialized methanotrophic lifestyle compared to more generalist chemoorganotrophic lifestyles.

The Acidobacteria is one of the cosmopolitan but poorly characterized groups of the domain Bacteria. The role of these organisms in natural environments remains poorly understood. Recent analysis of the genomes of three bacteria from the phylum Acidobacteria suggested a potential role of these prokaryotes in degradation of plant, fungal, and insect-derived organic matter (Ward et al., 2009). The genomes were shown to contain the genes encoding candidate cellulases and β-glucosidases, suggesting that acidobacteria are able to degrade cellulose substrates, although experimental confirmation for this ability was missing. Recently, however, the first proofs for cellulose degradation by acidobacteria became available (Pankratov et al., in press). This ability was found in a peat-inhabiting facultative anaerobe Telmatobacter bradus. So far, this is the only characterized cellulolytic acidobacterium, although two other uncharacterized subdivision 1 acidobacteria, strains KBS 83 and CCO287, and subdivision 3 acidobacterium KBS 96 were also recently reported to possess cellulolytic potential (Eichorst et al., 2011; Pankratov et al., 2011). The actual rates of cellulose decomposition by these bacteria are very low and are not comparable to those in well-characterized cellulose degraders. The latter, however, are either absent or present in only very low numbers in cold and acidic northern wetlands. This makes acidobacteria important members of the indigenous hydrolytic microbial community. This may also be true for other acidic water-logged habitats such as a primary tropical peat swamp forest in southern Thailand (Kanokratana et al., 2011). Metagenomic analysis of the microbial community in the surface peat layer revealed a variety of putative genes encoding a range of cellulolytic and hemicellulolytic enzymes from the Acidobacteria, suggesting the key role of these microbes in plant debris degradation.

A specific and highly abundant group of alphaproteobacteria was shown to colonize the hyaline cells of the outer stem cortex and the surface of stem leaves of Sphagnum cuspidatum (Raghoebarsing et al., 2005). Molecular identification of these bacteria showed their distant affiliation to acidophilic methanotrophs of the genera Methylocella and Methylocapsa (93% 16S rRNA gene sequence identity) and, therefore, these bacteria were assumed to be “symbiotic” methanotrophs that live in and on Sphagnum mosses. A number of similar 16S rRNA gene sequences were later retrieved from an acidic peat bog in West Siberia (Dedysh et al., 2006) as well as from other boreal ecosystems. These sequences form a common cluster, which is distinct from the Methylocystaceae and the Beijerinckiaceae, and does not contain cultivated representatives (Figure 2). Interestingly, members of this bacterial group were also detected in cellulolytic peat enrichment cultures, which were incubated without methane (Pankratov et al., 2011). None of the currently available pmoA or mmoX sequence groups retrieved from Sphagnum mosses or Sphagnum-derived peat can be linked to these bacteria. Therefore, the biology of organisms within this cluster remains obscure.

Despite the recent success in cultivating acidobacteria, most members of this phylum that thrive in northern wetlands resist isolation. This is especially true for subdivision 4 and 8 acidobacteria, which seem to be typical for the peat bog environment. At present, subdivision 4 has no characterized members. Subdivision 8 of the Acidobacteria is currently represented by three neutrophilic organisms that display highly contrasting characteristics. Holophaga foetida is a strictly anaerobic, homoacetogenic bacterium that degrades aromatic compounds (Liesack et al., 1994). Geothrix fermentans is also a strict anaerobe that oxidizes acetate as well as several other simple organic acids and long-chain fatty acids with Fe(III) as the electron acceptor (Coates et al., 1999). By contrast, Acanthopleuribacter pedis is a strictly aerobic chemo-organotroph that utilizes only a very limited number of growth substrates including glucose and several amino-acids (Fukunaga et al., 2008). The 16S rRNA gene sequences commonly retrieved from acidic wetlands affiliate with either Holophaga or Geothrix, but display only a very distant relationship to these characterized bacteria. Further cultivation efforts are needed to elucidate the metabolic potentials of the as-yet-uncultivated acidobacteria and their roles in northern wetlands.

Verrucomicrobia-affiliated 16S rRNA gene sequences are abundant in clone libraries obtained from Sphagnum peat (Figure 1). Most of these clones belong to a broad phylogenetic sequence cluster for which cultured representatives have not yet been reported. Therefore, no conclusions can be made concerning the lifestyle of peat-inhabiting verrucomicrobia. Most cultivated members of this phylum are heterotrophs but several recently characterized Verrucomicrobia are thermoacidophilic methanotrophs (Op den Camp et al., 2009). So far, these bacteria were detected in thermal environments only and none of the 16S rRNA gene sequences retrieved from northern wetlands fall close to those of thermoacidophilic methanotrophs. Therefore, the occurrence of methanotrophic Verrucomicrobia in acidic northern wetlands remains a highly intriguing question.

The candidate division OP3 is one of the primary bacterial groups for which no information is currently available. OP3-related 16S rRNA gene sequences have been recovered from many anoxic environments, such as flooded paddy soils, marine sediments, hypersaline deep sea, freshwater lakes, and methanogenic bioreactors (Glöckner et al., 2010). They were also retrieved from Sphagnum-dominated wetlands (Dedysh et al., 2006; Morales et al., 2006; Ivanova and Dedysh, unpublished results). The analysis of metagenomic fosmid libraries constructed from flooded paddy soils revealed that OP3 bacteria share a high proportion of orthologs with members of the Deltaproteobacteria and may possess an anaerobic respiration mode (Glöckner et al., 2010). They may have a facultatively anaerobic lifestyle since OP3-related 16S rRNA gene sequences were found both in oxic and anoxic peat layers.

Designing novel isolation strategies and cultivating uncultured microbes from northern wetlands remain highly challenging tasks given that most microbial inhabitants of these acidic, cold, and nutrient-poor environments are slow-growing, oligotrophic organisms. Hunting for such microbes is perhaps seen as technologically out of step with the increasingly molecular character of modern microbiology. The newly isolated microorganisms, however, are a unique source of novel, unexpected findings, which may revise many old paradigms in our knowledge. They also provide the means to study cell biology, to verify hypotheses emerging from genome sequence data, and to adjust the currently used molecular detection techniques as well as our ideas about the functional role of these microbes in the environment.