Distribution Characteristics and Driving Factors of the Bacterial Community Structure in the Soil Profile of a Discontinuous Permafrost Region

Abstract

Global warming leads to the melting of permafrost, affects changes in soil microbial community structures and related functions, and contributes to the soil carbon cycle in permafrost areas. Located at the southern edge of Eurasia’s permafrost region, the Greater Khingan Mountains are very sensitive to climate change. Therefore, by analyzing the bacterial community structure, diversity characteristics, and driving factors of soil profiles (active surface layer, active deep layer, transition layer, and permafrost layer) in this discontinuous permafrost region, this research provides support for the study of the carbon cycling process in permafrost regions. The results show that the microbial diversity (Shannon index (4.81)) was the highest at 0–20 cm, and the Shannon index of the surface soil of the active layer was significantly higher than that of the other soil layers. Acidobacteria and Proteobacteria were the dominant bacteria in the active layer soil of the permafrost area, and Chloroflexi, Actinobacteria, and Firmicutes were the dominant bacteria in the permafrost layer. Chloroflexi made the greatest contribution to the bacterial community in the permafrost soil, and Bacteroidota made the greatest contribution to the bacterial community in the active surface soil. The structure, richness, and diversity of the soil bacterial community significantly differed between the active layer and the permafrost layer. The number of bacterial species was the highest in the active layer surface soil and the active layer bottom soil. The difference in the structure of the bacterial community in the permafrost soil was mainly caused by changes in electrical conductivity and soil–water content, while that in the active layer soil was mainly affected by pH and soil nutrient indices. Soil temperature, NO₃⁻-N, and pH had significant effects on the structure of the bacterial community. The active layer and permafrost soils were susceptible to environmental information processing and genetic information processing. Infectious disease: the number of bacterial species was significantly higher in the surface and permafrost layers than in the other layers of the soil. In conclusion, changes in the microbial community structure in soil profiles in discontinuous permafrost areas sensitive to climate change are the key to soil carbon cycle research.

1. Introduction

The permafrost region, as an important ecosystem in northern China, plays an important role in carbon sequestration [1]. Permafrost contains approximately 1300 Pg of carbon, equivalent to half of the organic carbon in the world [2]. Permafrost is degrading due to a warming climate, and it is estimated that global permafrost will release 200 billion tons of carbon into the atmosphere over the next 300 years [3]. The soil depth profile represents different degrees of permafrost melting, including the active layer, transition layer, and permafrost layer [4]. The active layer represents the annual thawing and refreezing of the surface layer, which is greatly disturbed by the environment. The transition layer is frozen, but occasionally thaws during warm summers. The permafrost layer is an extreme environment that maintains low annual freezing temperatures and nutrient availability, but it is a relatively stable habitat for microbial communities [5]. Due to the large temperature difference between the four seasons in the permafrost area, the difference in environmental disturbance between different soil layers is large. With the increase in the temperature in the northern hemisphere, there is an urgent need to understand the response of different soil layers in the permafrost area to environmental changes [6].

Soil microorganisms are important participants in the carbon cycle of underground ecosystems, and they are easily affected by changes in the soil environment [7,8]. The degradation of soil organic carbon (SOC) is mainly driven by microorganisms [9], and soil nutrients and water also control the abundance and structure of microbial communities [10]. As a result of their low metabolic rate in frozen soil, unstable carbon is not degraded by microorganisms [11]. However, thawing permafrost under a warming climate greatly activates a variety of oligotrophic and symbiotic bacteria and enriches carbohydrate transporters and metabolism-related genes [12]. This leads to an increase in the abundance of bacteria, and microbial transformation results in soil carbon dominated by aliphatic carbon [13]. The characteristics of soil profiles significantly affected by environmental changes in discontinuous permafrost areas aid in microbial species replacement and richness studies by clarifying the causes of changes in microbial diversity. Therefore, understanding the structure of the soil microbial community and the diversity of different soil layers in permafrost regions is a key issue for understanding the changes in organic carbon storage in permafrost regions under the backdrop of climate warming.

The Arctic Village of the Greater Khingan Mountains is located on the southern edge of the Eurasian permafrost area, which is a discontinuous permafrost distribution area and is very sensitive to climate change [14]. The thawing of the permafrost in this area leads to an increase in the depth of the active layer [15], as well as changes in soil hydrothermal status and nutrients [16,17]. Therefore, by analyzing the characteristics and driving factors of the structure and diversity of the bacterial community in soil profiles (active surface layer, active deep layer, transition layer, and permafrost layer) in permafrost areas, this study hypothesizes that the composition and diversity of the bacterial community are affected by vertical differences in the soil, and it aims to (1) predict the contribution and function of the bacterial communities in different soil layers (2) and determine the driving factors that affect the composition of these communities. The bacterial community structure, diversity characteristics, and driving factors of the soil profile in the permafrost region are studied in order to provide a basis for understanding soil carbon cycling in the permafrost region under climate warming.

2. Materials and Methods

2.1. Research Area and Collection of Soil Samples

The study area is located in the Arctic Village in the permafrost region of the Greater Khingan Mountains, China, at a high latitude above 53° N (Figure 1a). The annual average temperature is about −5 °C, and the extreme minimum temperature in winter can drop to −50 °C. Larix gmelinii-sphagnum peatland is widely developed in Beiji Village. The soil is peat, the peat layer is more than 60 cm, and the permafrost layer is less than 60 cm. The peat layer is divided into two layers, with 0–20 cm being sphagnum peat and 20–60 cm being woody and herbaceous peat, and woody and herbaceous peat continues below the 60 cm frozen layer. Larix gmelinii is the single dominant species in the tree layer. There are many species of Sphagnum moss, including Sphagnum magellanium, Polytrichum commune, and Aulacomiumpalustre.

In October 2022, soil profiles were dug from the Larix gmelinii-sphagnum peatland, and soil samples from different soil layers (surface active layer (0–20 cm), deep active layer (20–60 cm), transition layer (60–80 cm), and permafrost layer (80–120 cm)) were collected (Figure 1b). A sterile serrated knife was used to collect the soil samples from the different layers, and they were placed on a clean plastic tarp. The next sample was collected after cleaning with 70% ethanol. Three parallel samples were collected from each soil layer and placed in aseptic bags, all of which were cryopreserved during transport and immediately transported to the laboratory, where one part was stored at 4 °C for a soil properties analysis, and the other part was stored at −80 °C for a microbiological analysis.

2.2. Soil Sample Testing

The total organic carbon (TOC) of the soil was determined using a Multi N/C 3100 TOC (Analytik-Jena, Llmenau OT Lanwegen, Germany). The total nitrogen (TN), total phosphorus (TP), soil ammonium nitrogen (NH₄⁺-N), and nitrate nitrogen (NH₃⁻-N) were determined using a continuous flow analyzer (Skalar, Breda, The Netherlands). The pH value of the soil, which had a wet soil ratio of 1:10, was measured with a PHSJ-3F pH meter. Soil conductivity was measured using a portable conductivity meter (DDBJ-350, INESA Scientific Instrument Co., Ltd., Shanghai, China). The soil temperature and moisture content were measured with a portable soil velocimeter (M129310).

2.3. DNA Extraction, Amplification, and Sequencing

Genomic DNA from each soil sample was isolated using a PowerSoil DNA Separation Kit (MO BIO, Carlsbad, CA, USA). Genomic DNA was extracted using 1% agarose gel electrophoresis. Concentrations were determined using a Nanodrop 2000 (Thermo, Waltham, MA, USA). To amplify the hypervariable region V4-V5 of the 16S rRNA gene, the primers 515F(5′-GTGCCAGCMGCCGCGG-3′) and 907R(5′-CCGTCAATTCMTTTRAGTTT-3) were used. Specific primers with barcodes were synthesized according to the specified sequencing region. The PCR products of the same sample were mixed and tested using 2% agarose gel electrophoresis, with 3 replicates per sample. The PCR products were recovered using an AxyPrepDNA gel recovery kit (Axygen Biosciences, Union City, CA, USA), then eluted with Tris-HCl buffer, and tested using 2% agarose gel electrophoresis. The PCR products were quantified using the QuantiFluorTM-ST Blue fluorescence quantification system (Promega, Madison, WI, USA) based on the initial quantitative results of electrophoresis.

2.4. Sequencing of Data Processing

The PE reads obtained using MiSeq sequencing were first spliced according to overlap, and then an OTU cluster analysis and a species taxonomic analysis were performed. A bioinformatics statistical analysis was performed for OTUs at a 97% similarity level. The Bayes RDP classifier algorithm was used to classify representative OTU sequences with a 97% similarity level. The richness and diversity of the microbial community were determined using a single-sample diversity analysis (alpha diversity). The PCoA (principal coordinate analysis) data dimensionality reduction analysis method can be used to study the similarity or difference in sample community composition. LEfSe (linear discriminant analysis effect size) can be used to identify the species characteristics that best explain the differences between two or more groups of samples and the extent to which these characteristics affect the differences between groups. A co-occurrence network analysis can be used to show the distribution between samples and species. Through a correlation analysis of species abundance information among different samples, the co-existence relationship of species in environmental samples can be obtained, and the similarity and difference between samples can be highlighted. The KEGG database (Kyoto Encyclopedia of Genes and Genomes, http://www.genome.jp/kegg/, accessed on 21 July 2024) was used to predict gene function.

2.5. Statistical Analysis

An ANOVA (single-factor analysis of variance) was conducted to analyze differences in the soil properties, soil bacterial community richness and diversity index, relative abundance of the main microbial phyla and genera, and functional groups of functional gene levels 1 and 2 between four different soil layers. The correlation between the main bacterial phyla and environmental factors was analyzed across different soil layers. All the above analyses were performed using the SPSS 20.0 software platform (IBM, Armonk, NY, USA) and Tukey’s post hoc HSD (Honest Significant Difference) test. Venn diagrams of graphical descriptions of unique and shared bacterial genera between different soil layers were created using the “VennDiagram” package in R (Team RDC 2016). The PLS-PM (partial least squares path model) was used to analyze the influence of the soil layer on soil bacterial metabolism in the permafrost region.

3. Results

3.1. Physical and Chemical Properties of Different Soil Layers

The pH of the surface soil in the active layer was significantly higher than that of the other soil layers (p < 0.05) (

Table 1. Physical and chemical properties of different soil layers.

Sample	pH	Cond (S/m)	ST (°C)	SWC (%)	TOC (g/kg)	TN (g/kg)	TP (g/kg)	HN4+-N (mg/kg)	HN3−-N (mg/kg)
Upper active layer	5.41 ± 0.03 a	5.01 ± 0.24 c	0.07 ± 0.05 a	19.57 ± 1.48 b	33.82 ± 0.61 b	2.13 ± 0.09 a	828.33 ± 4.43 b	12.33 ± 0.19 a	4.33 ± 0.09 a
Lower active layer	5.28 ± 0.01 b	6.32 ± 0.09 b	0.30 ± 0.08 a	22.27 ± 0.71 a	43.59 ± 1.44 a	1.79 ± 0.02 b	1115.50 ± 15.24 a	10.90 ± 0.28 b	2.08 ± 0.09 b
Transition layer	5.01 ± 0.04 c	5.71 ± 0.12 b	0.17 ± 0.34 a	22.8 ± 0.94 a	20.31 ± 0.64 c	1.41 ± 0.04 c	691.4 ± 22.62 c	8.14 ± 0.09 d	1.14 ± 0.08 c
Permafrost layer	4.91 ± 0.05 c	7.85 ± 0.71 a	−1.97 ± 0.48 b	21.97 ± 0.09 a	15.99 ± 0.43 d	1.52 ± 0.02 c	507.44 ± 11 d	10.20 ± 0.14 c	1.33 ± 0.03 c

Note: Different letters indicate a significant difference among soil depths, determined using one-way ANOVA (LSD, p < 0.05). Data shown are mean values ± standard deviation (n = 3). Abbreviations: cond: electrical conductivity; ST: soil temperature; NH₄⁺-N, ammonium nitrogen; NO₃⁻-N, nitrate nitrogen; SWC, soil–water content; TOC, total organic carbon; TN, total nitrogen; TP, total phosphorus.

); the cond showed the opposite trend, being significantly higher in the permafrost layer than in the other soil layers (p < 0.05). The ST of the active and transition layers was significantly higher than that of the permafrost layer (p < 0.05); the SWC showed the opposite trend, being significantly higher in the deep soil than in the surface soil of the active layer (p < 0.05). The TOC and TP were the highest at the bottom of the active layer, and the differences between the four soil layers were significant (p < 0.05). The TN, HN₄⁺-N, and HN₃⁻-N values were significantly higher in the active layer than in the other layers (p < 0.05).

3.2. Richness and Diversity of Soil Bacterial Communities along Soil Profiles in Permafrost Regions

A total of 388 bacterial soil OTUs were identified in the four layers of soil samples, of which 242 OTUs were shared, indicating that the composition of the soil OTUs varied with the soil depth. The number of unique bacterial OTUs was the highest at 0–20 cm and the lowest at 60–80 cm (Figure 2). The Shannon index (4.81) was the highest at 0–20 cm, and the Shannon index of the surface soil of the active layer was significantly higher than that of the other soil layers. The Simpson index of the transition layer was significantly higher than that of the other soil layers (Table 1). The ACE (839) and Chao1 (854.42) indices peaked at 20–60 cm, and they were significantly higher in the deep soil of the active layer than in the other soil layers (

Table 2. Index of richness and diversity of the soil bacterial community in different soil layers.

Sample	Shannon	Simpson	ACE	Chao1	Good’s Coverage (%)
Upper active layer	4.81 ± 0.06 a	0.021 ± 0.002 b	811.94 ± 44.33 ab	801.42 ± 31.74 ab	99.43 ± 0.13
Lower active layer	4.44 ± 0.01 b	0.037 ± 0.002 ab	839.00 ± 4.42 a	856.42 ± 17.47 a	99.28 ± 0.04
Transition layer	4.20 ± 0.15 b	0.061 ± 0.02 a	725.75 ± 13.66 b	731.45 ± 14.12 b	99.38 ± 0.03
Permafrost layer	4.45 ± 0.09 b	0.028 ± 0.005 b	751.90 ± 11.95 b	759.87 ± 15.4 b	99.51 ± 0.14

Good’s coverage: sequencing depth index. a, b: If there is one identical marking letter, the difference is not significant, and if there is different marking letter, the difference is significant.

). A PCoA (based on Bray–Curtis calculations) showed that the cumulative contribution of PCo1 and PCo2 was 69.89%, with the structure of the bacterial community in the active layer (0–80 cm) and the permafrost layer (80–120 cm) forming two distinct clusters along the PCo1 axis (Figure 3).

3.3. Soil Bacterial Community Structure along Soil Profiles in Permafrost Regions

At the phylum level, Acidobacteria (20%–89% to 28.48%), Proteobacteria (14.95%–34.24%), Chloroflexi (10.56%–25.4%), Actinobacteria (12.3%–15.39%), Bacteroidota (2.87%–6.21%), and Firmicutes (2.1%–8.38%) dominated in the soil samples taken from the four depths (Figure 4). However, their relative abundance varied greatly between the depths (Figure 5). The relative abundance of Proteobacteria was significantly higher in the 60–80 cm soil than in the permafrost soil. The relative abundance of Chloroflexi, Firmicutes, and Gemmatimonadota was significantly higher in the permafrost layer than in the active layer. The relative abundance of Bacteroidota and Elusimicrobiota was significantly higher in the active layer than in the permafrost layer. The relative abundance of Myxococcota was significantly higher in the topsoil than in the active layer and permafrost soils. The relative abundance of MBNT15 was significantly higher in the 0–20 cm and 20–60 cm soils than in the 60–80 cm soil.

At the genus level, Gallionellaceae (4.27%–20.21%), Subgroup_7 (5.7%–10.76%), Oryzhumus (6.82%–9.87%), and Acidobacteriales (4.2%–9.87%) dominated the soil samples taken from the four depths (Figure 4). Their relative abundance varied considerably between the depths (Figure 5). The relative abundance of Gallionellaceae and Vicinamibacterales was significantly higher in the 60–80 cm soil than in the permafrost and topsoil. The relative abundance of Acidobacteriales, unclassified_f__Gallionellaceae, Pseudolabrys, and BSV26 was significantly higher in the active layer than in the permafrost soil. The relative abundance of RGB-13-54-9, Gemmatimonadaceae, Bacteria, and Alicyclobacillus was significantly higher in the permafrost soil than in the active soil. The relative abundance of Candidatus_Koribacter was significantly higher in the topsoil than in the active layer and permafrost soils.

The LEfSe multilevel species difference discriminant analysis (LDA) revealed a total of 38 taxa (from phylum to genus) with effect values >4 in the different soil layers, among which 22 taxa were in the permafrost layer (Supplementary Figure S1). The 0–20 cm soil was typically enriched in Acidobacteriae, Candidatus Koribacter, Koribacteraceae, and Bacteroidota (Supplementary Figure S2). The 60–80 cm soil was typically enriched in Gammaproteobacteria, Burkholderiales, and Gallionellaceae. Chloroflexi, Anaerolineae, RBG-13-54-9, Ktedonobacteria, Ktedonobacteraceae, Ktedonobacterales, Gemmatimonadetes, Gemmatimonadaceae, Gemmatimonadota, and Gemmatimonadales were generally enriched in the 80–120 cm soil. The positive correlation of 14 gates was greater than the negative correlation (p < 0.05) (Supplementary Figure S3). Some groups, such as Bacteroidata, Chloroflexi, Actinobacteriota, Acidobacteria, and Proteobacteria, were found to play a central role (high connectivity) in the relatively abundant network, suggesting that they are the dominant phyla in the bacterial community. Although Gemmatimonadota had high connectivity, it was not a dominant species.

3.4. Relationship between the Major Bacterial Phyla and Environmental Factors along the Soil Profiles in Permafrost Regions

The results of the correlation analysis of the soil physical and chemical properties in the permafrost areas and the relative abundance of the major bacterial phyla showed different correlations and significance levels (Figure 6). Bacteroidota was significantly correlated with pH, ST, TOC, and TP (p < 0.01) and negatively correlated with cond (p < 0.01). Firmicutes and Gemmatimonadota were significantly correlated with cond (p < 0.01) and negatively correlated with pH, ST, TOC, and TP (p < 0.01). Myxococcota was significantly correlated with TN, NH₄⁺-N, and NO₃⁻-N (p < 0.01). RCP2-54 was negatively correlated with TN, NH₄⁺-N, and NO₃^—N (p < 0.01). Elusimicrobiota was significantly correlated with pH (p < 0.01) and negatively correlated with cond (p < 0.01). MBNT15 was negatively correlated with TOC (p < 0.01). Chloroflexi was negatively correlated with ST (p < 0.01).

3.5. Prediction and Analysis of Soil Bacterial Community Function along Soil Profiles in Permafrost Regions

A total of 6 first-grade functional groups and 18 second-grade functional groups with a relative abundance >1% were functionally labeled using Tax4Fun (Figure 7). Genes with a high absolute abundance in grade 1 functional groups included those related to metabolism, environmental information processing, and genetic information processing. Environmental information processing was significantly higher in the 0–20 cm, 60–80 cm, and 80–120 cm soils than in the 20–60 cm soil (p < 0.05). Genetic information processing was significantly higher in the 0–20 cm, 20–60 cm, and 80–120 cm soils than in the 60–80 cm soil (p < 0.05). Cellular processes were significantly higher in the active layer soil than in the permafrost soil (p < 0.05). Human disease was significantly higher in the surface and permafrost layers than in the active layer soil (p < 0.05). Genes with a high absolute abundance in the second-order functional group were those related to carbohydrate metabolism, amino acid metabolism, and membrane transport. Carbohydrate metabolism was significantly higher in the 20–60 cm, 60–80 cm, and 80–120 cm soils than in the 0–20 cm soil (p < 0.05). Infectious disease: the size of the bacterial community was significantly higher in the surface and permafrost layers than in the active layer soil (p < 0.05). Membrane transport was significantly higher in the 0–20 cm and 20–60 cm soils than in the 60–80 cm and 80–120 cm soils (p < 0.05). The following were significantly higher in the 0–20 cm, 20–60 cm, and 80–120 cm soils than in the 60–80 cm soil: the metabolism of cofactors and vitamins; replication and repair; folding, sorting, and degradation (p < 0.05). Nucleotide metabolism was significantly higher in the 0–20 cm soil than in the 20–60 cm, 80–120 cm, and 60–80 cm soils (p < 0.05). The metabolism of terpenoids and polyketides was significantly higher in the 20–60 cm soil than in the 0–20 cm, 80–120 cm, and 60–80 cm soils (p < 0.05). The metabolism of other amino acids was significantly higher in the active layer soil than in the permafrost soil (p < 0.05).

3.6. Soil Profiles Affect Soil Bacterial Metabolic Pathways in Permafrost Regions

PLS-PM explained 94% of the total variance affecting the soil bacterial metabolic pathways (Figure 8). There was a significant negative correlation between the soil layer and soil bacterial metabolic pathways (p < 0.05), and there was a positive correlation between the soil layer, soil physical and chemical properties, and soil bacterial community composition. The soil layer was significantly negatively correlated with the physical properties (p < 0.001), and the soil bacterial community significantly affected the soil bacterial metabolic pathways (p < 0.01). The soil chemical properties significantly affected the soil bacterial diversity (p < 0.05). The soil layer in the permafrost area indirectly affected the composition of the soil bacterial community by influencing the chemical properties of the soil. In conclusion, the soil layer in the permafrost area both directly and indirectly affected the metabolic pathways of the soil bacteria through the physical properties and bacterial communities of the soil.

4. Discussion

4.1. Differences in Soil Profiles Affect the Composition of the Microbial Community in Permafrost Areas

Significant differences were found in the composition of the soil microbial communities in different soil layers in the permafrost area of the Greater Hingingan Mountains. The main bacterial groups were Acidobacteria, Proteobacteria, Chloroflexi, and Actinobacteria. These results are consistent with those of previous studies in Alaska, the Arctic, and the Tibetan Plateau [4,18,19]. Bacteroidota, Elusimicrobiota, Myxococcota, and Proteobacteria live in nutrient-rich environments, where their abundance is significantly higher than in permafrost soil. The abundance of Proteobacteria was significantly higher at the bottom of the active layer than in the permafrost layer, which is due to the abundance of deep-rooted vegetation in the peatland [20]. At the genetic level, Gallionellaceae also showed the same pattern of change. Consistently with other studies, higher levels of nutrient input increased the abundance of Proteobacteria in the soil [5,21]. Chloroflexi, Firmicutes, and Gemmatimonadota exist in anaerobic or nutritionally poor soil environments, and their ability to maintain metabolic activity and DNA repair mechanisms at low temperatures has been confirmed by studies in permafrost in Iceland and the Tibetan Plateau [22,23,24]. At the gene level, RBG-13-54-9, Gemmatimonadaceae, and Alicyclobacillus also showed the same pattern of change, and their relative abundance was significantly higher in the permafrost layer than in the active layer. Although Actinobacteria did not differ significantly between the different soil layers, they were still the dominant phylum, including Oryzhumus at the genetic level. Actinobacteria can cope with nutrient limitations at low temperatures by hydrolyzing complex organic compounds [25]. The dominant bacteria in the active layer soil of the permafrost area were Acidobacteria and Proteobacteria, and the dominant bacteria in the permafrost layer were Chloroflexi, Actinobacteria, and Firmicutes. Studies have shown that the increasing ratio of Actinobacteria to Proteobacteria can be used as an indicator of permafrost degradation [20]. Permafrost degradation causes the migration of soil microorganisms in the active and permafrost layers, increases the decomposition of soil organic matter, and increases the loss of carbon in permafrost areas in the short term [24,26]. An LEfSe analysis showed that Chloroflexi made the greatest contribution to the bacterial community in the permafrost soil and that Bacteroidota made the greatest contribution to the bacterial community in the active surface soil. Some groups, such as Bacteroidata, Chloroflexi, Actinobacteriota, Acidobacteria, and Proteobacteria, were found to play a central role (high connectivity) in the relatively abundant network, suggesting that these species can adapt to multiple environments. The positive correlation of 14 phyla was higher than the negative correlation (p < 0.05), indicating cooperative metabolism and resource sharing among species. Gemmatimonadota, Verrucomicrobiota, and Desulfobacterota were not the dominant species, but they were found to play an important role in the network.

4.2. Differences in Soil Profiles Affect Bacterial Richness and Diversity in Permafrost Areas

There were 242 soil bacterial OTUs shared among the different soil layers in the permafrost area, and the number of unique bacterial OTUs differed across the soil layers. The richness and diversity of soil bacteria in permafrost areas vary with the soil layer. Significant differences were found in the soil bacterial richness and diversity between the different soil layers in the permafrost areas. The number of microbial species was the highest in the active layer surface soil and the active layer bottom soil. This is consistent with the findings obtained in the Tibetan Plateau and Arctic regions [18,24,27]. Because the soil in the active layer of the permafrost area was greatly disturbed by soil texture and environmental changes, the soil bacterial richness and diversity index first increased and then decreased with the increase in the soil layer. The lower microbial diversity in the permafrost may be due to the fact that permafrost habitats are generally nutrient-poor, with temperatures below freezing and low water availability [28]. The soil bacterial richness and diversity index was the lowest at the bottom of the active layer but increased in the permafrost layer, which was caused by the decrease in soil moisture and nutrients with the increasing soil depth. Due to the unique bacterial community in the permafrost, the richness and diversity index was higher than that of the soil at the bottom of the active layer. The PCoA results showed that the bacterial communities in the active layer and the permafrost soils were obviously separated and that the structures of the bacterial communities in these soils were significantly different.

4.3. Effects of the Environment on the Soil Bacterial Community in the Permafrost Area

The difference in the bacterial community structure in the permafrost soil was mainly caused by changes in electrical conductivity and soil–water content, while that in the active layer soil was mainly affected by the pH and soil nutrient indices. This is consistent with the results of studies on the effects of permafrost collapse on soil bacterial communities on the Qinghai–Tibet Plateau [23]. Since Proteobacteria live in nutrient-rich environments, they are positively correlated with soil temperature. Firmicutes, Gemmatimonadota, and RCP2-54 live in anaerobic or nutritionally poor soil environments, and they are positively correlated with SWC and electrical conductivity. In permafrost areas, with the deepening of the active soil layer, the physical and chemical properties of the soil change due to the influence of freeze–thaw cycles and soil texture, which affects the composition and diversity of the soil microbial community. In terms of the effects of the soil environment on the microbial communities in the different soil layers, it was found that soil temperature, NO₃⁻-N, and pH had great effects on the structure of the bacterial community, and the NH₄⁺-N and NO₃⁻-N contents in the active soil decreased with depth. This is consistent with the results of other studies in permafrost areas. NH₄⁺-N is the main form of inorganic nitrogen in boreal forest soil, and it affects the nitrogen fixation capacity of forest soil [17]. Recurring freeze–thaw cycles in the active layer can have significant effects on microbial communities. In addition, soil structure, moisture, and substrate availability vary with the freeze–thaw cycle; thus, the mechanism underlying the effects of freeze–thaw cycles on microbial communities will be further investigated in the future [29,30,31].

4.4. Analysis of the Prediction Function of Bacterial Communities along Soil Profiles in Permafrost Regions

Closely phylogenetically related microorganisms have similar habitat associations, and potential changes in community function can be inferred [28]. Some studies have suggested that community function may be more sensitive to thawing permafrost than community composition [18]. In this study, the genes with the highest absolute abundance of grade 1 functional groups were those related to metabolism, environmental information processing, and genetic information processing. The active layer topsoil and permafrost soil were found to be susceptible to environmental information processing and genetic information processing. This is consistent with the functional predictions in other studies [20]. Cellular processes were significantly higher in the active layer soils than in the permafrost soils. The genes with a high absolute abundance in the second-order functional group were those related to carbohydrate metabolism, amino acid metabolism, and membrane transport. Carbohydrate metabolism was significantly higher in the deep active layer soil and the permafrost soil than in the surface soil. Nucleotide metabolism was significantly higher in the active layer surface soil than in the deep soil and the permafrost soil. The metabolism of other amino acids was significantly higher in the active layer soil than in the permafrost soil. Human diseases were significantly higher in the surface layer and permafrost soils than in the active layer soil. Infectious disease: the number of bacterial species was significantly higher in the surface layer and permafrost layer than in the other layers.

5. Conclusions

The dominant bacteria in the active layer soil of the permafrost area were Acidobacteria and Proteobacteria, and the dominant bacteria in the permafrost layer were Chloroflexi, Actinobacteria, and Firmicutes. Chloroflexi made the greatest contribution to the bacterial community in the permafrost soil, and Bacteroidota made the greatest contribution to the bacterial community in the active surface soil. The number of microbial species was the highest in the active layer surface soil and the active layer bottom soil. The bacterial communities in the active layer soil and permafrost soil were obviously separated, and the bacterial community structures of these soils were significantly different. The difference in the bacterial community structure in the permafrost soil was mainly caused by changes in electrical conductivity and soil–water content, while that in the active layer soil was mainly affected by pH and soil nutrient indices.