Introduction

An estimated 25 % of Earth’s terrestrial surface is underlain by permafrost. The classic definition of permafrost is ground that remains at or below 0 °C continuously for at least 2 years (Jansson and Taş 2014). The areal extent of permafrost in China ranks third in world after Russia and Canada, and elevationally controlled permafrost, which is called altitudinal or high-altitude permafrost, ranks first (Ran et al. 2012). High-altitude permafrost mainly distribute in mountainous west China, such as the Chinese Altai Mountains, the Chinese Tianshan Mountains, the Qilian Mountains, and the Qinghai-Tibet Plateau (Zhao et al. 2014). A latest map of frozen ground in China, including both permafrost and seasonally frozen ground, shows that the total area of permafrost is estimated at ~1.59 × 106 km2 (Ran et al. 2012; Fig. S1).

The extreme characteristics of permafrost lead to early conclusion that permafrost soils were completely devoid of biological entities (Gilichinsky 2002). However, current evidence reveals that many microorganisms are able to adapt and even active at subzero temperatures in permafrost as “a community of survivors” (Friedmann 1994). The microbial ecology of permafrost has recently become the focus of intense research efforts owing to the emerging concerns about the impacts of climate change and possibly subsequent permafrost thaw on the microbial degradation of trapped organic matter, with the increased potential for the release of the greenhouse gases as a consequence (Mackelprang et al. 2011; Jansson and Taş 2014).

Compared with latitudinal permafrost in polar regions, altitudinal permafrost, especially permafrost on the Qinghai-Tibet Plateau, is more sensitive to the combined influence of climatic warming and surface conditions (Jin et al. 2000). Recent investigations indicated that most permafrost regions of China were being affected by climatic warming and significant permafrost degradation had occurred (Li et al. 2008b). Description of abundance, activity, diversity, and distribution of microorganisms in permafrost is crucial to our understanding of how microorganisms survive in permafrost, and how they respond to current climate change and subsequent permafrost thawing. In fact, microbes in permafrost in China have been the subject of meticulous studies over the last decade (Table 1). Microbiologists focus on assembling an inventory, which would cover the taxonomic diversity of microorganisms inhabiting permafrost of various locations in China. In this review, we summarize our current knowledge of the microbial ecology and biodiversity of permafrost in China. To make the picture complete, information on microorganisms from the overlying active layer soils, seasonally frozen ground, and soils affected by discontinuous permafrost is also included.

Table 1 Summary of studies of microbial biodiversity in permafrost soils in China by culture-dependent and culture-independent methods
Full size table

Microbial abundance of permafrost in China

A large number (up to 1010 cells g−1) of microbial cells was detected in permafrost in China but varied over a large range among different permafrost environments (Table 1). Microbial direct counts as determined by epifluorescence microscopy with 4′,6-diamino-2-phenylindole (DAPI) staining ranged from 107 to 109 cells/g of soil or sediment, which were consistent with permafrost samples taken from Arctic, but were relatively higher than Siberian and Antarctic samples (Fig. 1a). Furthermore, the number of aerobic bacterial CFU reached 6 × 107/g of soil or sediment, which was higher than the quantities of Siberian, Arctic, and Antarctic permafrost (Fig. 1b). The variation in the abundance of the microbes in different permafrost environments may reflect the water or nutrient content of the habitat (Shivaji and Reddy 2010 and references therein).

Fig. 1
figure 1

Boxplots of microbial abundance in various permafrost environments. a Total direct counts in permafrost in China (107–109 cells/g), Arctic (107–109; Hansen et al. 2007; Steven et al. 2007, 2008), Siberia (103–108; Gilichinsky 2002; Vishnivetskaya et al. 2006), and Antarctic (103–108; Aislabie et al. 2006; Gilichinsky et al. 2007), respectively. b Viable bacteria counts in permafrost in China (0–107 CFU/g), Arctic (0–106; Hansen et al. 2007; Steven et al. 2007, 2008), Siberia (0–107; Shi et al. 1997; Vishnivetskaya et al. 2000; Gilichinsky 2002; Vishnivetskaya et al. 2006) and Antarctic (0–105; Vorobyova et al., 1997; Aislabie et al. 2006; Gilichinsky et al. 2007), respectively. Components of the boxplot are: top of the box, upper quartile; midline of box, median; bottom of box, lower quartile; bars, 1.5 times length of box (1.5 times the horizontal spread); dots, values that are > or <1.5 times the horizontal spread of the distribution, plus the upper or lower quartile

Full size image

In general, only a small percentage of total cells are represented by cultured isolates. For example, in Beilu River (on the Qinghai-Tibet Plateau) permafrost soils, viable cell counts were 102–106 CFU/g of soil, while total counts were in the range of 107–109 cells/g (Feng et al. 2004). Similar results were also found by Wang et al. (2011), Li et al. (2012a), and Zhang et al. (2013b). The lack of cultured organisms might be owing to non-culturable nature of microorganisms (e.g., “dwarf” cells), inappropriate pretreatments of samples (e.g., thawing regimes) or particular microbes requiring specific culture conditions (Steven et al. 2006; Kochkina et al. 2012). On the other hand, direct microbial counts as determined by quantitative PCR or epifluorescence microscopy tend to overestimate total cell counts because dead microbial cells or naked DNA may remain well preserved for long periods of time at constant subzero temperatures in permafrost (Willerslev et al. 2004).

Present data show that abundance of microbial community seems to be influenced by specific abiotic factors. For example, with increasing depth or age of permafrost, both the ability to recover viable cells from permafrost and the number of viable and total cells decreased; however, the diversity of bacterial isolates seemed to be independent from permafrost depth (Feng et al. 2004; Zhang et al. 2007a; Wang et al. 2011; Ollivier et al. 2013; Tai et al. 2014). Besides, Zhang et al. (2009) found that archaeal amoA abundance in alpine and permafrost soils significantly decreased with altitude on Mount Everest (Tibet Plateau) and archaeal ammonia oxidizers were more abundant than bacterial ones at altitudes below 5400 m above sea level, while the situation was reversed at higher altitudes. Viable microbial numbers were also different among sampling months (Chen et al. 2011) and vegetation types (Yu and Shi 2011; Li et al. 2012a). Furthermore, microbial abundance was closely correlated with soil physicochemical properties (e.g., soil moisture, pH, organic carbon, and total nitrogen content) (Wang et al. 2011; Zhang et al. 2013b, 2014c; Tai et al. 2014). Therefore, the origin, age, and physiochemical characteristics of permafrost combined with other factors probably determine the abundance of microbial community (also see Steven et al. 2009; Hu et al. 2012).

Diversity of viable microorganisms

A considerable diversity of viable microorganisms, including bacteria, archaea, yeasts, filamentous fungi, and microalgae have been detected in permafrost and associated environments in China. A list of microbial generic names recorded as culture from permafrost in China is provided (Table S1).

Phylogenetic groups of the bacterial isolates from Chinese permafrost generally fall into six categories: Actinobacteria, Firmicutes, α-Proteobacteria, β-Proteobacteria, γ-Proteobacteria, and Cytophaga-Flavobacterium-Bacteroides (CFB) group (Table S1). This was consistent with previous studies which indicated that these bacterial phyla were also dominant in Arctic permafrost (Steven et al. 2009 and references therein). However, in Antarctic soil samples, a few have been identified to be associated with Deinococcus-Thermus and Spirochaetes except for the aforementioned dominating groups (Shivaji and Reddy 2010). Both Gram-positive and Gram-negative isolates are represented, and spore-forming Bacteria are also commonly isolated, although the abundance of spore-forming Bacteria varies largely among geographically separated permafrost samples. For example, spore-forming genera represented 53 % of culturable bacteria in permafrost samples collected along the Qinghai-Tibet Railway (Liu et al. 2008), but only 18, 6, and 1 % in the Qinghai-Tibet Plateau (Li et al. 2012a), Beilu River basin (Zhang et al. 2007a), and Tianshan Mountains (Bai et al. 2006) samples, respectively. Actinobacteria and Proteobacteria always represented a high proportion of the permafrost bacterial community, accounting for approximately 80 % of alpine permafrost isolates (Bai et al. 2006), 82 and 79 % of plateau permafrost isolates recovered from Beilu River basin (Zhang et al. 2007a) and Tuotuohe (Yang et al. 2012a), respectively. Bacterial isolates from Chinese permafrost are almost all aerobic heterotrophs and could be assigned to at least 83 genera (Table S1), which are comparable to the number of previously described genera (at least 70) from different permafrost regions in the world (Steven et al. 2009; Margesin and Miteva 2011). Among them, the genera Arthrobacter, Bacillus, Brevundimonas, Flavobacterium, Microbacterium, Paenibacillus, Planococcus, Planomicrobium, Pseudomonas, Psychrobacter, Sphingomonas, and Streptomyces were frequently isolated.

Previous descriptions of viable Archaea in polar permafrost mainly focused on diversity and abundance of methanogenic Archaea and their methanogenic activity (Rivkina et al. 2002, 2007). However, investigation of viable archaea in permafrost in China remains limited. Isolates related to the genera Methanolobus and Methanomethylovorans were recovered from frozen ground of Zoige wetland on the Qinghai-Tibet Plateau (Zhang et al. 2008a). Moreover, the study reported a novel psychrophilic methanogen, Methanolobus psychrophilus sp. nov., which showed the highest methanogenesis rate from methanol and large number (about 17 % of the total archaea) in situ soil temperatures, suggesting its potential role in methane emission of the wetland.

Diversity of viable fungi has been extensively investigated in polar permafrost regions. A review by Ozerskaya et al. (2009) summarized hyphal fungi detected in Arctic permafrost by culturing, and showed that species of the genera Penicillium, Aspergillus, Geomyces, and Cladosporium were very common. In Antarctic and sub-Antarctic soil systems, at least 400 taxonomically distinct genera were recorded by isolation, suggesting that the fungi might be the most diverse biota in the Antarctic (Bridge and Spooner 2012). As of now, our knowledge of viable Fungi community in Chinese permafrost were obtained from sporadic reports. In addition, previous studies have indicated that there was no relation between the amount of fungi and the depth of permafrost (Gilichinsky et al. 2007; Kochkina et al. 2012). Hu et al. (2014) found that culturable fungi occurred in low number (0–103 CFU g−1) in active layer soils collected from the Kunlun Mountain Pass, however, no cells were detected in deeper horizons (≥200 cm of depth intervals). The recovered isolates distributed among five different genera: Geomyces, Cladosporium, Alternaria, Rhodotorula, and Cryptococcus representing two phylogenetic groups of Ascomycota and Basidiomycota.

Recent researches also suggested that permafrost and associated environments on the Qinghai-Tibet Plateau could harbor a wide diversity of oleaginous microorganisms, including bacteria, yeasts, filamentous fungi, and microalgae (Li et al. 2012c; Liu et al. 2014). The lipid content of these isolates was high and varied from 7.3 to 40.7 %. Molecular identity analysis showed that they belong to 17 different genera, including Ochrobactrum, Agrobacterium, Cryptococcus, Rhodotorula, Fusarium, Botryococcus, and Nannochloris. To our knowledge, very few data are available on the functional and anaerobic bacteria, viruses or phages in permafrost in China.

Phenotypic traits of permafrost isolates

Permafrost is considered as an extreme environment because indigenous microorganisms must survive prolonged exposure to subzero temperatures and background radiation for geological time scales in a habitat with low water activity and extremely low rates of nutrient and metabolite transfer (Steven et al. 2006; Margesin and Miteva 2011). During the long evolutionary period, these organisms adopt a variety of physiological adaptations that have allowed them to survive in permafrost. The bacterial communities isolated from permafrost in China were primarily rod shaped together with a few cocci (for example, Fig. S2) and always formed pigmented colonies (Bai et al. 2006; Zhang et al. 2007a, b). The synthesis of pigments is normally regarded as an effective protection strategy against the harmful radiation (Rothschild and Mancinelli 2001).

The recovery of bacterial isolates from Chinese permafrost samples is facilitated using oligotrophic media on which high colony recovery efficiency and abundance were observed after plating (e.g., Bai et al. 2006; Zhang et al. 2007a, b; Chen et al. 2011), suggesting that indigenous viable communities are mainly oligotrophic. Such a characteristic was also observed in Arctic and Antarctic permafrost isolates (Steven et al. 2009). Microbial abundance in subsurface soils is associated with soil porosity, as soil pores are required for movement of groundwater and therefore larger pore sizes correlate with an increased supply of organic compounds (Kaiser and Bollag 1990). In general, the solid phase (ice) makes up 92–97 % of total water volume in permafrost (Gilichinsky 2002). At such high ice contents, the permafrost soil pores are completely filled with ice, obviously limiting the availability of organic matters and therefore selecting for oligotrophic microbial communities.

Consistent with previous reports in other permafrost environments (Gilichinsky 2002; Steven et al. 2006), the microbial communities isolated from permafrost in China are primarily psychrotolerant rather than psychrophilic. For example, only 2 out of 68 bacterial strains isolated from alpine permafrost in the Tianshan Mountains were psychrophilic (Bai et al. 2006), with optimal growth occurring at about 12 °C and the upper limit being 20 °C. Similar results were also reported by Bai et al. (2005) and Zhang et al. (2007a), indicating that the indigenous microorganisms could adapt the ambient subzero temperatures.

Furthermore, Chinese permafrost isolates tend to be halotolerant, which also agrees with the previous researches (Gilichinsky 2002; Steven et al. 2008). For example, Zhang et al. (2007b) found that all the bacterial strains isolated from the Qinghai-Tibet Plateau could grow on PYGV agar supplemented with 5 % NaCl and several strains could even grow with NaCl concentration up to 20 %. In general, microbial survival in permafrost is restricted to small amounts of unfrozen water inside soil, ice, or brine veins. Such environments generally contain high concentration of salts (D’Amico et al. 2006). Thus, there may be a link between halotolerance and microbial survival under the condition of desiccation. Moreover, halotolerant microorganisms are capable of coping with osmotic stress. Responses to osmotic stress in high-salt environments are similar to responses of a microbial cell during desiccation when compatible solutes such as K+ and sucrose accumulate and hence stabilize cells (Rothschild and Mancinelli 2001).

The producing of cold-active enzymes, including protease, cellulose, amylase, lipase, β-mannanase and β-xylanase, is frequently detected in the microbial communities isolated from permafrost in China (Bai et al. 2005; Zhang et al. 2007b, 2008c). In general, cold-active enzymes increase the flexibility of their structure in order to compensate for the freezing effect of their cold habitats (Feller 2007). The abilities of these microorganisms to produce cold-active enzymes facilitate their survival at low temperatures and also suggest that they might be of potential value for biotechnological exploitation.

Furthermore, lipid biosynthesis in oleaginous microorganisms isolated from permafrost and associated environments in China could be crucial to their survival in cold environments (Li et al. 2012c; Liu et al. 2014). Generally, the lipid/protein ratio increases significantly at a low temperature in many eukaryotes and prokaryotes, due to the relative increase in lipid biosynthesis (Guschina and Harwood 2006). Also oleaginous microorganisms can accumulate high amounts of neutral storage lipids under appropriate nutrition conditions (Amaretti et al. 2010). The major neutral lipids accumulate in hydrophobic lipid particles and serve as storage of the building blocks for membrane lipids synthesis (Rossi et al. 2009). At low temperature, it is well known that some membrane lipid modifications and fatty acid changes require de novo biosynthesis to maintain membrane fluidity and normal function (Margesin 2012). Accordingly, the accumulation of lipid might be closely associated with the cold adaption in oleaginous microorganisms (Rossi et al. 2009; Amaretti et al. 2010).

Recent genomic studies revealed that genomes of cold-adapted microorganisms have been shown to possess and express genes encoding several features that are required for surviving in permafrost (Jansson and Taş 2014). Future wider application of state-of-the-art ‘omics’ technologies will lead to a better understanding of cold adaptation features and the ecological roles of permafrost microorganisms.

Novel microbial taxa

Most of the microorganisms isolated from Chinese permafrost were identified as known species (using molecular methods); however, some novel, possibly endemic, taxa (at genus or species level) were also isolated. A list of recently reported novel bacterial, archaeal, and fungal genera and species from frozen soils in China was showed in Table S2. At least thirty-three novel taxa were described, among which Moheibacter sediminis gen. nov., sp. nov., was isolated from the permafrost sediment of Mo-he basin in northeast China (Zhang et al. 2014b). In addition to bacteria, Methanospirillum psychrodurum sp. nov., a strictly anaerobic, hydrogenotrophic, methanogenic archaeon, was isolated from a frozen ground-affected soil of Madoi wetland on the Qinghai-Tibet Plateau (Zhou et al. 2014). Furthermore, a majority of these novel strains are psychrophilic or psychrotolerant organisms. These studies further indicate that terrestrial cold environments in China represent a specific ecological niche for prolonging survival of original microbial lineages.

Culture-independent microbial diversity and community composition in permafrost in China

Culture-independent methodologies adopt molecular-based tools to analyze nucleic acids isolated directly from soil samples with no need to culture, and lead to the realization that the microbial diversity is far greater than that detected by cultivable approach. Recent molecular studies revealed that permafrost and associated environments in China host a large diversity of microbial communities, although there is considerable variability in the microbial composition of permafrost in different geographical locations (Fig. 2). These differences might be attributed to differences in sample origin as well as differences in techniques that were used.

Fig. 2
figure 2

Microbial composition of permafrost from different geographical locations in China. a–j Microbial community composition in different permafrost environments of China (Yang et al. 2008; Han et al. 2011; Zhang et al. 2013b; Dan et al. 2014; Hu et al. 2014; Yun et al. 2014). Pie charts represent relative abundances of different phyla in a sample set. The bacterial and archaeal communities in the Kunlun Mountain Pass are derived from unpublished data (h, i). k Community structure of methanotrophic bacteria in permafrost soils sampled from a littoral wetland of Lake Namco (Yun et al. 2014). l and m Composition of functional bacterial groups in the Tianshan Mountains (Wu et al. 2012a) and Beilu River permafrost region (Zhang et al. 2013a), respectively

Full size image

Bacteria

As mentioned above, bacterial isolates recovered from permafrost in China belong to four phylogenetic groups: Actinobacteria, Proteobacteria, Firmicutes, and CFB. However, Yang et al. (2008) detected 7 Bacteria phyla in permafrost sediments from the Tianshan Mountains using denaturing gradient gel electrophoresis, with predominance of Proteobacteria and Actinobacteria (Fig. 2a). The same situation was reported for Kunlun Mountain Pass (Fig. 2h; unpublished data), Beilu River (Zhang et al. 2013b), and latitudinal permafrost samples (Yang et al. 2012b). Furthermore, a diverse Bacteria community that consisted of 15 bacterial phyla, containing two candidate phyla (TM7 and OP11), was detected in Mo-he permafrost 16S rDNA clone libraries, among which Acidobacteria was the dominant group (Fig. 1e; Dan et al. 2014). Acidobacteria-related sequences were also found to be prevalent in an alluvial permafrost wetland on the Sanjiang Plain in northeast China (Yun et al. 2014). A previous study showed that Acidobacteria was abundant in acidic soils and their abundance decreased with increasing pH scale (Lauber et al. 2009). Thus, frequent detection of Acidobacteria in the Mo-he and Sanjiang Plain permafrost might be due to their slightly acidic pH conditions (pH 6.0 and 5.36–5.50, respectively). An interesting comparative assessment of bacterial communities in both gas hydrate and non-gas hydrate samples collected from the DK-2 borehole in the Qilian Mountains permafrost region revealed that Proteobacteria and Bacteroidetes were the major groups in non-gas hydrate sample, while sequences related to Deinococcus-Thermus were found mostly in gas hydrate clone library (Fig. 2c, d; Han et al. 2011). This dissimilarity could be partly attributed to lower moisture content in the gas hydrate sample than the non-gas hydrate one (Han et al. 2011). Members of the phylum Deinococcus-Thermus are typically tolerant to both radiation and desiccation, with genes involved in resistance to radiation also can contribute to desiccation toleration (Mattimore and Battista 1996). These features are likely to help organisms survive in permafrost soils where they are exposed to low water availability and constant background radiation. Yun et al. (2014) identified bacterial communities in permafrost soils sampled from a littoral wetland of Lake Namco on the Qinghai-Tibet Plateau. Differently, the authors found that Chloroflexi-related sequences accounted for a large proportion (~21 %) of bacterial communities though Proteobacteria and Actinobacteria were the most abundant phyla in permafrost wetland soils (Fig. 2f). Since a previous study has already revealed higher number of Chloroflexi in water-saturated alpine tundra meadow soils than in drier tundra soils (Costello and Schmidt 2006). Chloroflexi may be abundant in wet subsurface soil, as for the permafrost wetland of Lake Namco, that is water saturated almost the year round.

In recent years, data based on high-throughput sequencing of 16S rRNA genes are becoming increasingly available for a range of permafrost and active layer samples and reveal an immense bacterial diversity than previously described (see Wu et al. 2012a; Yang et al. 2012b; Zhang et al. 2013a, b, 2014c). For example, a regional investigation in bacterial communities of permafrost-affected soil collected from 19 sites along the Qinghai-Tibet highway found 33 bacterial phyla, and these communities differed significantly among vegetation types and were mainly influenced by variations in soil carbon/nitrogen ratio (Zhang et al. 2014c).

Overall, the bacterial diversity in Chinese permafrost is generally higher than the diversity of fungi or archaea. Frequently detected bacterial phyla in 16S rDNA datasets include Proteobacteria, Actinobacteria, Firmicutes, Acidobacteria, and Bacteroidetes, as well as several uncharacterized or candidate phyla. These groups also commonly occur in Arctic permafrost (Jansson and Taş 2014). However, in Antarctic soil systems, in addition to these bacterial groups, sequences corresponding to Deinococcus-Thermus and Cyanobacteria were also found at a high level (Cowan et al. 2010 and references therein). As mentioned above, members of the Deinococcus-Thermus group are known for their ability in resisting to both radiation and desiccation to ensure their survival in extreme Antarctic soils. The cyanobacteria also played a pivotal role in Antarctic terrestrial ecosystem due to their participations in the C and N acquisition and the development of other biological communities, because environmental harshness in Antarctica generally precluded the survival of higher eukaryotic phototrophs (de la Torre et al. 2003). Furthermore, consistent with previous studies in acidic Mo-he permafrost (Dan et al. 2014) and an acidic permafrost wetland on the Sanjiang Plain (Yun et al. 2014); Wilhelm et al. (2011) also found that sequences related to Acidobacteria dominated both active layer and permafrost clone libraries in an acidic wetland from the Canadian High Arctic, which further supported the viewpoint that the phylum Acidobacteria was abundant in acidic environments. Some recent reports have also highlighted the representatives of uncharacterized or novel phyla in Chinese permafrost. For example, Yun et al. (2014) found that uncharacterized representatives of Chloroflexi were abundant in permafrost wetland soils. Also, several previous studies have indicated that Chloroflexi-related sequences are ubiquitous in Arctic permafrost soils (Mackelprang et al. 2011; Wilhelm et al. 2011). However, there are few cultured representatives of Chloroflexi and none of these has close sequence similarity to sequences that were found in permafrost. It is therefore that permafrost from different geographical locations shared a similar core of bacterial taxa, but also revealed differences in composition of the bacterial community. These differences may reflect the unique and extreme conditions of the permafrost environments.

Fungi

Although fungi are known to be important in soil ecosystems, diversity, distribution, and the function of fungi in Chinese permafrost are largely unknown. To our best knowledge, three molecular studies of fungal communities in permafrost in China have been published so far. Hu et al. (2014) investigated the fungal diversity and community structure by cloning-RFLP analysis of sediments through a 10-m-long permafrost core from the Kunlun Mountain Pass (Fig. 2j). A large proportion of phylotypes (25/62) distantly related to known fungal species was detected, possibly belonging to new taxa, and Ascomycota were predominant. The results further indicated that the community composition of fungi varied with depth, while these communities largely distributed according to core layers. The study also detected some fungal taxa related to the genera Geomyces, Phoma, Mortierella, Thelebolus, Cryptococcus, Leucosporidiella, Rhodotorula, Dioszegia, Penicillium, Alternaria, and Cladosporium which occurred frequently in extreme cryo-environment, suggesting their cosmopolitan psychrophilic or psychrotolerant nature (Ozerskaya et al. 2009; Bridge and Spooner 2012). Therefore, this study underlined that both cosmopolitan and possibly endemic fungal species occur in the Kunlun Mountains permafrost. Furthermore, fungal communities detected in the Qinghai-Tibet Plateau permafrost associated with different vegetation types encompassed the phylogenetic groups Ascomycota, Basidiomycota, Mucoromycotina, Chytridiomycota, Microsporidia, Glomeromycota, Kickxellomycotina, and Blastocladiomycota, with Ascomycota, Basidiomycota, and Mucoromycotina were the most abundant phyla (Fig. 2g; Zhang et al. 2013b, 2014c). Consistent with these results, fungal groups belonging to the Ascomycota, Basidiomycota, and Mucoromycotina were also major forms of fungi in polar permafrost (Ozerskaya et al. 2009; Kochkina et al. 2012). However, as an exception, recent researches in high alpine ecosystems showed that snow bed soils were dominated by groups of previously undescribed chytrids (Freeman et al. 2009; Schmidt et al. 2012).

Archaea

The present data show that archaeal sequences detected in permafrost in China include those from the Euryarchaeota, Crenarchaeota, and Thaumarchaeota. In the Qinghai-Tibet Plateau permafrost, Archaea communities were dominated by OTUs affiliated to Group1.3b/MCG-A within Crenarchaeota and methanogenic and unclassified groups within Euryarchaeota (Wei et al. 2014). However, our recent studies revealed that sequences related to Crenarchaeota and methanogenic archaea were not detected in the Kunlun Mountains permafrost clone libraries, instead, the archaeal communities primarily comprised sequences belonging to the ammonia-oxidizing Thaumarchaeota Group 1.1b (Fig. 2i; unpublished data). An interesting result of the molecular characterization of archaeal communities in Tianshan Mountains permafrost was the detection of a large number of sequences related to the halophilic Euryarchaeota Group I (Fig. 2b; Yang et al. 2008). Detection of halophilic archaea in alpine permafrost also supports the above-mentioned viewpoint that the primary habitat for microbial life in permafrost exists as lines of liquid water that have high-salt content within ice or surrounding soil microaggregates (Jansson and Taş 2014). Furthermore, archaeal communities are also extensively characterized in natural wetlands in cold area, like permafrost and tundra, which contribute a large proportion of global methane emission (Cavicchioli 2006). For instance, Zhang et al. (2008b) found that both of the phyla Euryarchaeota and Crenarchaeota were detected in frozen soils of Zoige wetland on the Qinghai-Tibet Plateau, with a novel uncultured methanogen cluster, Zoige cluster I affiliated to Methanosarcinales, being dominant. In another study by Tian et al. (2012), the relationship between archaeal community dynamics and vegetation type was investigated during growing season in the Zoige peatlands. The results suggested the prevalence of acetoclastic methanogen families Methanosarcinaceae and Methanosaetaceae, together with the archaeal community composition was related to vegetation type.

Previous researches generally revealed that the phyla Crenarchaeota and Euryarchaeota dominated archaeal communities in permafrost soils (for example, Rivkina et al. 2007; Steven et al. 2007, 2008; Mackelprang et al. 2011). However, we recently found that the phylum Thaumarchaeota was highly abundant in the Kunlun Mountains permafrost clone libraries (unpublished data). Similarly, the archaeal community in ice wedges from the Canadian High Arctic permafrost had low diversity and mainly comprised Thaumarchaeota-related OTUs (Wilhelm et al. 2012). It is necessary to note that the initial assignment of thaumarchaeota was to the Crenarchaeota (Brochier-Armanet et al. 2008), and as a result it is possible that some 16S rDNA sequences identified as crenarchaeota in previously published work were actually affiliated with the Thaumarchaeota. For example, Yang et al. (2008) detected two low-temperature Crenarchaeota-related sequences that clustered with Candidatus Nitrosopumilus maritimus SCM1 from the Tianshan Mountains permafrost (see their Fig. 4), which was now thought belonged to the Thaumarchaeota. In addition, a study in Antarctic soils of the Ross Sea region reported by Ayton et al. (2010) revealed that more than 99 % of the archaeal soil clones belonged to Group 1.1b Crenarchaeota and clustered together with 16S rRNA genes from fosmid 54d9 (AJ496176; see their Fig. 2), which was also thought to be thaumarchaeota now. Thus, archaea associated with the phylum Thaumarchaeota might be ubiquitous in permafrost from different geographical locations.

The ecology of microbial functional groups in Chinese permafrost ecosystem

The 16S rRNA gene-sequence data have shown that Chinese permafrost contains microbial taxa from different functional guilds (see Zhang et al. 2009, 2013a, 2014c; Wu et al. 2012a; Yun et al. 2014), including the ammonia-oxidizing bacteria and archaea, methane-oxidizing bacteria, nitrifying bacteria, nitrogen-fixing bacteria, and sulfur- and sulfate-reducing bacteria (Fig. 2k–m). These groups may play a central role in the conceptual feedback loop of key biogeochemical cycling that could be induced by climate change, resulting in the potential for emissions of the greenhouse gases CO2, CH4, and N2O (Mackelprang et al. 2011). However, the long-term consequences of probable microbial responses to climate change in permafrost environments are not well understood (Jansson and Taş 2014), although a previous study has shown that 3 years of experimental warming increased microbial biomass and affected the soil microbial community composition in the Tibetan alpine grasslands (Zhang et al. 2014a). Furthermore, what remains unclear is whether these organisms represent historical remnants from previous microbial communities, or whether they are viable or even active in permafrost. As these studies assess microbial communities using the DNA-based methods, the occurrence of these certain groups in total community does not always mean their ecological or functional potential. Consequently, future studies using novel techniques such as stable isotope probing (SIP) or RNA-based methods are encouraged to identify active participants in biogeochemical cycling processes in permafrost ecosystem. Enzyme activity is a useful indicator of potential microbial activity and may provide some evidence of the metabolic range of the permafrost (Wu et al. 2012b). Moreover, a study by Geng et al. (2012) focused on soil respiration in the Tibetan alpine grasslands suggested that belowground biomass and soil moisture may be important drivers of Tibet Plateau soil respiration.

A metagenomic study by Guan et al. (2013) evaluated the bacterial community structure and function as well as their correlation with environmental factors in the Qinghai-Tibet Plateau major ecosystems. The taxonomic and functional composition of bacterial communities were more dissimilar within alpine meadow samples than among farmland samples, and the same pattern was observed in elements cycles and pathways associated with adaption to environment and land use types. Additionally, bacterial communities were significantly correlated with geogenic variables. Specifically, the root-nodule bacteria were negatively correlated with the soil moisture and pH, while Thiobacillus associated with sulfur cycles showed potential responses to low temperature and intense UV radiation.

Conclusions and perspectives

Both culture-dependent and culture-independent investigations have revealed that permafrost in China harbored diverse and novel microbial communities (including representatives of Gram-positive and Gram-negative bacteria, archaea, yeasts, filamentous fungi, and microalgae). Some microbial taxa are ubiquitous in permafrost from different geographical locations, suggesting their cosmopolitan psychrophilic or psychrotolerant nature. The future studies are encouraged to address the probable responses of permafrost microorganisms to climate change in regard to the increased metabolic rates associated with higher temperatures and nutrient availability due to the melting of permafrost. By applying SIP or RNA-based methods combined with high-throughput sequencing techniques and other ‘omics’ approaches, we could define active members of the permafrost microbial community and their functional potential and how they potentially respond to climate change. More importantly, the geographical separation of permafrost ecosystem in China provides an ideal model to improve biogeographical studies (Fig. S1). Therefore, future work should also focus on the geographic or spatial patterns of permafrost microbial community and the influence of ecological processes.