Distinct patterns of soil bacterial and fungal communities in the treeline ecotone

Highlights

	● Homogeneous selection and dispersal limitation played dominant roles in shaping bacterial and fungal communities, respectively.
	● Keystone bacteria were more critical for maintaining network stability above the treeline, while fungi were the keystone taxa for network stability below the treeline.
	● Oligotrophic species were predominantly enriched above the treeline, whereas copiotrophic species were more abundant below the treeline.
	● Microbial communities responded greatly to treeline shift than slope aspect.

Abstract

The upward shift of the alpine treeline driven by global climate change has been extensively observed across many mountain ecosystems worldwide. However, variations in belowground microbial communities in the treeline ecotone, as well as the influence of microtopographic factors (e.g., slope aspect) on these changes, remain unclear. Here, we collected soil samples from different aspects above or below the treeline and analyzed the microbial communities using high-throughput sequencing. Our study revealed distinct community characteristics, co-occurrence patterns, and assembly processes between bacterial and fungal communities. Especially, homogeneous selection and dispersal limitation played dominant roles in shaping bacterial and fungal communities, respectively. Keystone bacteria were more critical for maintaining network stability above the treeline, while fungi were the keystone taxa for network stability below the treeline. We also found that oligotrophic species such as Acidobacteriota, Chloroflexi, Verrucomicrobiota, and Ascomycota were predominantly enriched above the treeline, whereas copiotrophic species like Proteobacteria, Gemmatimonadota, Actinobacteriota, and Firmicutes were more abundant below the treeline. Our results uncovered that microbial communities responded greatly to treeline shift than slope aspect, and also imply that the upward shift of the alpine treeline may increase the stochasticity of microbial communities. These findings facilitate our understanding of how microbial communities in the treeline transition zones of alpine ecosystems respond to global warming and their potential effects on soil carbon dynamics.

Graphical abstract

Keywords

alpine treeline / slope aspect / r/K strategy / keystone species / community assembly / Qinghai-Tibet Plateau

1 Introduction

The alpine treeline is an ecological transition zone between forest and shrubland (Körner and Paulsen, 2004). As an important ecosystem boundary within vertical natural zones, it is characterized by its ecological fragility and sensitivity while being minimally affected by human activities, making it an ideal natural monitor for climate change (Fajardo and McIntire, 2012; Aakala et al., 2014). Under the background of global warming, nearly 90% of treeline in the northern hemisphere are experiencing treeline rise to varying degrees (Lu et al., 2021). Warming-induced upward shift of the alpine treeline increases the biomass of forest lands for photosynthetic carbon sequestration in high mountain areas (Devi et al., 2008). However, replacement of tundra by forest may decrease the surface albedo and increase the possibility of wild fire, all of which could, in turn, amplify the effects of climate warming (Dial et al., 2024). Additionally, changes in the position of the treeline can alter the chemical composition and indigenous microbial composition and functions in alpine lakes, exacerbating the risk of biodiversity loss and ecosystem degradation (Wang et al., 2022; Catalán et al., 2024). Therefore, understanding the ecological impacts of treeline shifts in the context of climate warming is of significant importance. Currently, numerous studies have concerned the changes in plant communities and soil nutrient cycling at the treeline, yet little is known about the responses of soil microbial communities to climate change within this critical area (Sigdel et al., 2018; Fetzer et al., 2024).

Soil microbial communities play a vital role in organic matter decomposition and nutrient cycling, and also closely connect with aboveground plants (van der Heijden et al., 2008; Sokol et al., 2022). Changes in vegetation types in the treeline ecotone can directly impact microbial communities through different root deposits and exudates, while also indirectly altering microbial composition by influencing soil moisture, temperature, and nutrient availability (Huang et al., 2024). Limited studies have shown that the shrub-dominated upper limit of the treeline typically exhibits higher soil temperatures, abundant moisture, and nutrient availability, all of which lead to significant changes in community composition and increased microbial diversity (Ding et al., 2015; Yang et al., 2024). Currently, researches in microbial ecology have shifted from merely describing community composition and distribution to investigating complex co-occurrence networks and the underlying mechanisms of community assembly (Ye et al., 2024). Warming enhances microbial activity, and increase soil nutrients boost microbial abundance, both of which can drive microbial networks toward greater complexity (Zhou et al., 2020). Moreover, the significantly higher soil temperature, acting as a deterministic force, enhanced the role of homogeneous selection in community assembly (Li et al., 2024a). Therefore, it is expected that in shrubland above the treeline, microbial co-occurrence networks will be more complex, and deterministic processes play a dominant role. However, how microbial species interactions respond to treeline shift remains unclear, as well as their community assembly processes.

The Qinghai-Tibet Plateau has the highest natural treeline in the Northern Hemisphere (Miehe et al., 2007). Sygera Mountain, a typical alpine region located on the Qinghai-Tibet Plateau, has been observed experiencing an upward shift in the treeline over the past century due to the changing climate conditions (Liang et al., 2016; Mayor et al., 2017). In this study, we investigated microbial community characteristics, co-occurrence patterns and assembly processes in soil samples collected from different aspects above or below the treeline on Sygera Mountain. Slope aspect is an unneglectable factor related to light intensity and hydrological conditions, and it has been reported that slope aspect significantly impacted microbial communities in the southeastern Qinghai-Tibet Plateau (Li et al., 2024b). Thus, we considered this microtopography factor to test whether its effects on microbial communities are larger than treeline shift or not. We hypothesized that: (1) In the light of vegetation transition during treeline upward shifts, microbial communities would respond greatly to treeline shift than slope aspect. (2) As plant forms symbiotic associations with fungi, bacterial and fungal communities would have distinct diversity and co-occurrence patterns and assembly processes.

2 Materials and methods

2.1 Study site and soil sampling

We conducted this research in Sygera Mountain (29.61° N, 94.60° E), which is located in southeastern of the Qinghai-Tibet Plateau, China. This area is characterized by a cold temperate montane monsoon climate, with distinct vertical zonation distributions in vegetation along the elevation gradient. Vegetation types and soils were investigated surrounding the treeline ecotone on Sygera Mountain, with different aspects (sunny or shady slope). On the sunny slope, dominant plant species above the treeline were Salix oritrepha and Rhododendron nyingchiense, and Sabina saltuaria was dominant below the treeline. While on the shady slope, Rhododendron nivale and Abies georgii were prevalent above and below the treeline, respectively.

Soil surface samples in the treeline ecotone from both the sunny and shady slopes were collected in August 2020. Totally, we collected 24 soil samples (6 replicates×2 slopes×2 treeline positions) from the established 1 m×1 m quadrats, with the spacing between the six replicates being approximately 15 meters. At each quadrat, five subsamples of 0−15 cm topsoil were collected after removing surface stones and litter, and mixed thoroughly to form a single sample. The collected soil samples were promptly sealed in sterile bags and transported to the laboratory under refrigeration. Upon arrival, soil samples were sieved through a 2 mm mesh to remove visible residues (e.g., roots, stones, and organic debris), then divided into two portions for subsequent experiments: one was stored at 4 °C for the analysis of soil physicochemical properties and soil respiration, and the other was stored at −40 °C for subsequent DNA extraction.

2.2 Soil properties and respiration analysis

Soil pH was determined using a soil-to-water ratio of 1:5 (w/v) by a pH meter (Mettler Toledo, Zurich, Switzerland). Soil temperature (ST) was measured at a depth of 12 cm using a portable soil moisture, temperature, and electrical conductivity rapid measurement device (SPECTRUM TDR150, Amreica). Soil moisture (SM) was determined by the gravimetric method after drying soil samples to a constant weight. Total carbon (TC) and total nitrogen (TN) contents were measured by an elemental analyzer (Vario EL III, Elementar, Germany) after the soil samples were fully combusted at a high temperature of 1150 °C. Soil ammonium nitrogen (${\text{NH}}_{\text{4}}^{+}$-N) and nitrate nitrogen (${\text{NO}}_{\text{3}}^{-}$-N) were analyzed using a continuous flow analyzer (TRAACS 2000, Norderstedt, Germany).

The carbon dioxide concentrations for measuring soil basal respiration (SBR) and substrate induced respiration (SIR) were determined using the Li-820 infrared gas analyzer (Li-Cor, New York, USA). Subsequently, the Soil Carbon Availability Index (CAI) was calculated by dividing soil basal respiration by substrate induced respiration.

2.3 DNA extraction, amplicon sequencing, and data processing

Total DNA was extracted from 0.5 g soil samples using the MoBio Power Soil Extraction Kit (MoBio Laboratories, Carlsbad, CA, USA) following the manufacturer’s protocol. The purity of the extracted DNA was measured using a NanoDrop ND-1000 Spectrophotometer (NanoDrop, Wilmington, USA). Universal primer sets, 515F (5′-GTGCCAGCMGCCGCGGTAA-3′)/806R (5′-GGACTACHVGGGTWTCTAAT-3′) and ITS1F (5′-CTTGGTCATTTAGAGGAAGTAA-3′)/ITS2 (5′-GCTGCGTTCTTCATCGATGC-3′) were used to amplify the V4 region of bacterial 16S rRNA gene and the internal transcribed spacer 1 region (ITS1) through triplicate polymerase chain reactions (PCRs). Finally, high-throughput sequencing was conducted using the Illumina NovaSeq PE250 sequencing platform (Illumina Inc., CA, USA).

The sequencing data were processed using the Quantitative Insights Into Microbial Ecology 2 (QIIME2, version 2022.2) pipeline. Cutadapt was employed to trim primer sequences, followed by the Divisive Amplicon Denoising Algorithm 2 method (DADA2) for denoising and joining paired-end reads (Martin, 2011; Callahan et al., 2016). Amplicon sequence variants (ASVs) were generated by clustering sequences at 100% similarity (Callahan et al., 2017). Taxonomic annotation of ASVs were assigned using a Naive Bayes classifier trained on the SILVA 132 database (see the website of SILVA) for bacterial 16S sequences and the UNITE database (see the website of National Genomics Data Center) for fungal ITS sequences. Sequence alignment was conducted using mafft, and phylogenetic trees for bacteria and fungi were constructed using FastTree (see the website of microbesonline.org/fasttree/). To comprehensively assess the impact of rare species on the community, singletons were retained, while non-bacterial ASVs were excluded from the 16S sequences. To normalize samples to the same total reads, bacterial and fungal sequences were rarefied to 76419 and 75749 for downstream community analysis.

It is noteworthy that though the ITS region has been shown to be less phylogenetically informative, ITS is considered reliable for family or genus level analyses (Fouquier et al., 2016; Tedersoo et al., 2018). A recent study also demonstrated that iCAMP analysis based on both the original ITS tree and the hierarchical classification tree provided nearly identical results (Osburn et al., 2021). This suggests that while ITS trees may have some limitations, particularly with tips showing long phylogenetic distances, iCAMP analysis focus on highly related taxa may reduce the impact of these issues (Fan et al., 2024). To mitigate the potential biases that may arise from adjusting phylogenetic trees, we used an identical approach to construct both the bacterial and fungal phylogenetic trees for the subsequent analysis. However, for verification purposes, we also constructed a hierarchical classification tree derived from the UNITE taxonomy for our fungal ASVs (Tedersoo et al., 2018).

2.4 Co-occurrence network analysis

Amplicon sequence variants (ASVs) for bacteria and fungi with relative abundance ranking in the top 50 were selected for subsequent analysis. Thereafter, the Spearman correlation coefficients between each ASV were calculated using the WGCNA package, and the significant relationships (|r|>0.8; p<0.05) were used for the construction of microbial co-occurrence networks. To understand the interactions of keystone ASVs within networks, the within-module connectivity (Zi) and among-module connectivity (Pi) for each node was calculated. These nodes can be categorized into four types: network hubs (Zi>2.5 and Pi>0.62), module hubs (Zi>2.5 and Pi<0.62), connectors (Zi<2.5 and Pi>0.62), and peripherals (Zi<2.5 and Pi<0.62), with the first three being indefined as keystone taxa (Yuan et al., 2021). The co-occurrence networks of keystone nodes were constructed and visualized using the Gephi (see the website of gephi.github.io) platform.

2.5 Community assembly

We employed the Infer Community Assembly Mechanisms by Phylogenetic-bin-based null model (iCAMP), adapted from the framework developed by Stegen et al., (2013) to analyze and quantify the community assembly processes of microorganisms (Ning et al., 2020). All ASVs were initially divided into distinct bins based on their phylogenetic relationships. The “bin. Size. Limit” was set to 32 and 28 for bacterial and fungal communities, respecitively. Subsequently, ecological processes for both individual bins and the entire community were determined using the beta Net Relatedness Index (βNRI) and the modified Raup–Crick metric (RC). Specifically, values of βNRI<−1.96 and βNRI>1.96 are considered as homogeneous selection (HoS) and heterogeneous selection (HeS), respectively. For cases where |βNRI|≤1.96, further differentiation is performed using the RC index: RC<−0.95 indicates homogeneous dispersal (HD); RC>0.95 points to dispersal limitation (DL); and |RC|≤0.95 suggests the role of drift (DR). In iCAMP, homogeneous selection and heterogeneous selection constitute deterministic processes, while dispersal limitation, homogeneous dispersal, and drift constitute stochastic processes. The above analysis was conducted within the iCAMP (version 1.5.12) package.

2.6 Statistical analysis

All analyses were conducted using R software (version 4.3.1). The alpha diversity at the ASV and Phylun level, including richness and Pielou index, were estimated using the vegan package. Differences in the microbial alpha diversity indices above and below the treeline were tested using the Wilcoxon rank-sum test. To examine the compositional dissimilarities in bacterial and fungal communities among samples, principal coordinates analysis (PCoA) analysis based on Bray−Curtis dissimilarity was conducted using the vegan package. Three statistical analysis methods including Analysis of dissimilarities (ADONIS), Analysis of similarities (ANOSIM), and Multi Response Permutation Procedure (MRPP) were carried out to test the compositional differences between above and below the treeline positions. The partial Mantel test and distance-based redundancy analysis (db-RDA) were performed to determine the effects of environmental factors on the composition of microbial communities. Further, the contribution of individual factors to overall community variation was quantified using the rdacca.hp package (Lai et al., 2022). To explore the distribution of bacteria with different nutrition strategies, the likely r-strategists (e.g., Proteobacteria, Gemmatimonadota, Actinobacteriota, and Firmicutes) and K-strategists (e.g., Acidobacteriota, Chloroflexi, Verrucomicrobiota, and Planctomycetota) were summed up, and the bacterial r/K strategy ratio was calculated. Linear discriminant analysis (LDA) effect size (LEfSe) was used to identify the biomarkers enriched above or below the treeline. The average value of environmental factors in each location was calculated, and then Pearson correlations between ecological processes and environmental factors were determined to explore the trends in the variation of assembly processes.

3 Results

3.1 Microbial diversity and composition

Based on the richness and Pielou index, we found that there was no significant difference for bacterial and fungal communities between above and below the treeline, either in sunny or shady slopes (Fig.1−Fig.1). But when we see the diversity of specific phyla, Actinobacteriota, Bacteroidota, and Gemmatimonadota had significantly higher richness below the treeline on the sunny slope, while the RCP2−54 had higher richness above the treeline on the shady slope (p<0.05, Fig.1). Bacterial diversity was higher in the sunny slope than that in shady slope, while fungal diversity responded weakly to slope.

Fig.1 Richness and Pielou index of the bacterial (A, B) and fungal (C, D) communities above and below the treeline. (E) Microbial richness at the top ten bacterial phyla level and top two fungal phyla level in the treeline ecotone on both the sunny and shady slopes. (F) Relative abundance of r-strategy bacteria in different treeline positions. (G) Relative abundance of K-strategy bacteria in different treeline positions. (H) Bacterial r-strategy/K-strategy ratio in different treeline positions. Only significant differences at p<0.05 in the panel are shown above the boxes. Significance levels are as follows: *p<0.05, **p<0.01, *** p<0.001, ns: not significant.

Full size|PPT slide

PCoA analysis showed that, except for the fungal community on the shady slope, the compositional dissimilarities of both bacterial and fungal community were significant in the treeline ecotone (Fig.2 and Fig.2). Specifically, the compositional variation induced by treeline effect was greater on sunny slope for bacterial community, but on shady slope for fungal community (Tables S2 and S3).

Fig.2 Principal coordinate analysis (PCoA) based on Bray–Curtis distance for the bacterial (A) and fungal (D) communities in different treeline positions. Distance-based redundancy analysis (db-RDA) of the bacterial (B, C) and fungal (E, F) communities on both the sunny (the second column) and shady (the third column) slopes. Axes legends represent the percentages of variation explained by the axis. ST: soil temperature, SM: soil moisture.

Full size|PPT slide

Our research on the distribution of soil bacteria found that r-strategists had significantly higher relative abundance below the treeline, and the ratio of r/K strategy bacteria was also higher at the lower treeline limit on the sunny slope (Fig.1 and Fig.1). LEFSe analysis (p<0.05 and LDA≥3) was further conducted to identify potential biomarkers of the treeline ecotone. A total of 35 bacterial ASVs and 17 fungal ASVs were identified, which can be used to characterize microbial enrichment in the treeline ecotone. Our findings revealed that the enriched bacterial ASVs on the sunny slope were predominantly associated with Proteobacteria, while on the shady slope, they were mainly affiliated with Acidobacteriota (Fig.3 and Fig.3). The selected fungal ASVs were almost entirely from Ascomycota (Fig.3 and Fig.3).

Fig.3 LEfSe analysis for the biomarkers of bacterial (A, B) and fungal (C, D) communities in the treeline ecotone on the sunny (left) and shady (right) slopes. A positive Linear Discriminant Analysis (LDA) value means that the biomarkers were enriched above the treeline, and a negative LDA value means that they were enriched below the treeline.

Full size|PPT slide

The results of partial Mantel test indicated that the microbial community composition was significantly correlated with ST, SM, TC, and soil nitrogen content (i.e., TN, ${\text{NH}}_{\text{4}}^{+}$-N, and ${\text{NO}}_{\text{3}}^{-}$-N) (Tables S4 and S5). Briefly, soil physicochemical factors accounted for 64.57% and 34.34% of the variance in bacterial communities on the sunny and shady slopes, respectively, but only explained 20.53% and 12.93% of the differences in fungal communities (Tables S6 and S7). In particular, changes in microbial communities were closely related to variations in soil nitrogen content (i.e., ${\text{NH}}_{\text{4}}^{+}$-N and ${\text{NO}}_{\text{3}}^{-}$-N) (Fig.2, Fig.2, Fig.2, and Fig.2; Tables S6 and S7). Additionally, bacterial communities were also affected by ST, while variations in fungal communities were driven by soil pH as well (Fig.2, Fig.2, Fig.2, and Fig.2; Tables S6 and S7).

3.2 Bacterial-fungal co-occurrence patterns

The keystone taxa were identified via bacterial-fungal co-occurrence networks. The results showed that, above the treeline most keystone taxa were from bacterial communities, whereas below the treeline fungal keystone taxa dominated (Fig.4−Fig.4). This indicates that treeline shift may change the co-occurrence patterns and bacterial-fungal interactions. Meanwhile, we also noticed that the correlations within fungal keystone taxa were positive, in contrast, most of the correlations between bacterial and fungal taxa were negative.

Fig.4 The interkingdom networks in the treeline ecotone for keystone bacteria and fungi on both the sunny and shady slopes. SUA: above the treeline on the sunny slope, SUB: below the treeline on the sunny slope, SHA: above the treeline on the shady slope, SHB: below the treeline on the shady slope.

Full size|PPT slide

3.3 Assembly processes of microbial communities

Homogeneous selection (58.01%−75.32%) played a predominant role in bacterial community assembly, while dispersal limitation (36.38%−67.28%) was the most significant ecological process for fungal community assembly (Fig.5 and Fig.5). Despite the different dominant assembly processes in bacterial and fungal communities, both of which showed higher dispersal limitation and lower homogeneous selection below the treeline. This suggests that the upward shift of the treeline will increase the stochasticity of microbial community.

Fig.5 The relative importance of different ecological processes in bacterial (A) and fungal (B) communities. Relationship between microbial aseembly processes and soil properties (C, D). Significant Pearson correlation coefficients are noted by asterisks. Significance levels are as follows: *p<0.05; **p<0.01; ***p<0.001. HoS: homogeneous selection, HeS: heterogeneous selection, HD: homogenizing dispersal, DL: dispersal limitation, DR: drift, ST: soil temperature, SM: soil moisture, TC: total carbon, TN: total nitrogen, A vs B: ecological processes between samples from above and below the treeline.

Full size|PPT slide

For bacterial communities below the treeline, the decrease in homogeneous selection was primarily associated with Acidobacteriota (Bin 150 on both slopes; Bin 152 on the shady slope) and Chloroflexi (Bin 62 on the sunny slope). The increase of dispersal limitation was mainly attributed to the responses of bins in Proteobacteria (Bin 106 and Bin 105 on the sunny and shady slopes, respectively) and Bacteroidota (Bin 61 and Bin 57 on the sunny and shady slopes, respectively) (Fig.6 and Fig.6). Among the observed fungal ASVs, changes in different ecological processes were largely influenced by Ascomycota, but could be attributed to different bins. The decrease in homogeneous selection on both slopes was mainly associated with Bin 73 and Bin 74, while the increase in dispersal limitation, could also be attributed to Bin 6 and Bin 71 dominated by Basidiomycota, and Bin 79 dominated by Mortierellomycota (Fig.6 and Fig.6).

Fig.6 Bacterial (A, B) and fungal (C, D) community assembly mechanisms within different bins in the treeline ecotone on the sunny (left) and shady (right) slopes. Phylogenetic tree (centre), relative importance of different ecological processes in each bin (stacked bars in the first annulus), relative abundance of each bin (2nd annulus), position-induced changes of HoS, DL, and DR were shown in the third, forth, and fifth annulus, respectively. HoS: homogeneous selection, HeS: heterogeneous selection, HD: homogenizing dispersal, DL: dispersal limitation, DR: drift.

Full size|PPT slide

Pearson correlation analysis further showed the influence of environmental factors on microbial community assembly processes. Bacterial communities were primarily affected by soil nitrogen content, with ${\text{NH}}_{\text{4}}^{+}$-N showing significant positive or negative correlations with homogeneous selection and dispersal limitation, respectively. Fungi were primarily influenced by soil carbon content and physical properties, with homogeneous selection positively correlated with TC and SM, and dispersal limitation negatively correlated with ST and SM (p<0.05, Fig.5 and Fig.5).

4 Discussion

4.1 Compositional differences in the treeline ecotone and their potential ecological consequences

In line with our first hypothesis, we found that compositional differences between above and below the treeline were greater than that between different slopes. A study conducted on Baima Mountain indicated that the variations of soil physicochemical properties caused by aspect were much weaker than elevation, leading to a consistent pattern of microbial distribution across different aspects (Zhao et al., 2024). Slope aspect, as a microtopographic factor on a local scale, can create distinct hydrothermal and nutrient conditions. However, in the light of vegetation transition during treeline upward shifts, microbial communities would respond greatly to treeline shift than slope aspect. This suggests that the microbial ecological effects induced by treeline shifts due to climate warming are far more significant than those caused by slope aspect.

When we focused on the specific taxa, we found that Acidobacteriota, Chloroflexi, Verrucomicrobiota, and Ascomycota were predominantly enriched above the treeline, whereas Proteobacteria, Gemmatimonadota, Actinobacteriota, and Firmicutes were more abundant below the treeline (Fig.4). Interestingly, microorganisms enriched above the treeline were categorized as K-strategists (oligotrophic species), whereas those below the treeline primarily survived with r strategy (copiotrophic species) (Razanamalala et al., 2018; Chen et al., 2022; Dai et al., 2022). Here, we used the CAI index (calculated by dividing soil basal respiration by substrate induced respiration) to quantify the degree of carbon starvation of microorganisms, with a higher CAI index indicating a richer carbon status and relatively better soil nutrient conditions (Cheng et al., 1996; Parsapour et al., 2018). The results showed that CAI values were higher below the treeline (Table S1), suggesting that microorganisms in this area had access to more available carbon, which favored the enrichment of r-strategists. It is also worth considering that below-treeline communities might be enriched with rhizospheric microbial communities, which tend to grow faster (copiotrophs) in the presence of labile root exudates. A study in tropical rainforests found that the addition of labile carbon to the soil led to a rapid response from γ-Proteobacteria and Firmicutes, replacing Acidobacteriota as the dominant group and enhancing soil respiration (Cleveland et al., 2007). Singh et al. also showed that soil dominated by oligotrophic microorganisms had a lower carbon turnover rate, suggesting their role in long-term carbon storage (Singh et al., 2010). Under the background of global warming, the upward shift of alpine treeline has been widely observed, which increases the area of forest land for photosynthetic carbon fixation and effectively mitigates climate change (Devi et al., 2008; Franke et al., 2017). Our study points out that after warming, the coniferous forest is likely to move upward into the shrubland area, where copiotrophic microorganisms are expected to replace the original oligotrophic groups as the dominant species. Oligotrophic communities, with lower microbial biomass, might contribute to a smaller mineral-associated organic matter (MAOM) pool, which is a slow-cycling and stable soil organic carbon (SOC) pool (Sokol et al., 2019). Conversely, copiotrophic communities, with higher microbial biomass and turnover, might lead to a larger MAOM pool. While a larger MAOM pool might increase the soil’s capacity for long-term carbon storage, the rapid turnover associated with copiotrophic dominance may simultaneously intensify carbon fluxes, potentially offsetting the stabilization effect of MAOM. Therefore, we speculate that this shift in dominant species is projected to accelerate the soil carbon turnover rate, leading to increased carbon dioxide production and potentially affecting the carbon source-sink balance in alpine regions.

4.2 Distinct bacterial-fungal co-occurrence patterns in the treeline ecotone

We found that keystone bacteria were more critical for maintaining network stability above the treeline, while fungi were the keystone taxa for network stability below the treeline (Fig.4−Fig.4). Studies have suggested that bacteria are typically influenced by the availability of environmental resources, while fungi have closer associations with plants (Genre et al., 2020; Piton et al., 2023). For instance, a research conducted in African tropical forests has shown that vegetation characteristics, rather than soil properties, were the primary factors influencing fungal community composition (Peay et al., 2013). Therefore, in shrublands with relatively better nutrient conditions, the warm and fertile environment enhances bacterial metabolic processes, making their contribution to network stability indispensable (Brown et al., 2004). Soil fungi are considered as K-strategists and are capable of decomposing complex components of litter. Compared to shrublands, forests exhibit higher plant diversity and greater litter production, which correspondingly increases the importance of fungi below the treeline (Ding et al., 2015; Gaudel et al., 2024; Zhang et al., 2024). What’more, in the treeline ecotone of Changbai Mountain, fungi, in particular, EcM fungi, played a central role connecting the multi-kingdom communities and co-occurrence network (Yang et al., 2022). Therefore, compared with shrublands, soil fungi in alpine forests should be more crucial for maintaining the stability of interkingdom network here.

In addition, We found that in interkingdom networks, fungi predominantly exhibited positive correlations with each other, indicating that most keystone fungi utilize resources cooperatively or share similar ecological niches (Deng et al., 2016). Although some studies have suggested that fungi tend to break down the complex structures of recalcitrant organic compounds, such as lignin and cellulose, by secreting specific enzymes, thereby creating conditions for bacteria to further hydrolyze and utilize these substances (de Boer et al., 2005). Our research revealed that competition was prevalent between keystone bacteria and fungi. A previous study also demonstrated that microbes tend to exhibit positive correlations within kingdoms but negative correlations between kingdoms (Agler et al., 2016). The interkingdom competition between bacteria and fungi may be attributed to their competition for resources or antagonistic effects caused by antimicrobial compounds (Hassani et al., 2018). Existing research has already shown that networks dominated by competition rather than cooperation exhibit higher stability (Coyte et al., 2015). We reasonably speculate that keystone bacteria and fungi maintain the stability of keystone networks through competitive relationships, thereby enabling the overall microbial community to better adapt to the variable natural environment.

4.3 Assembly mechanisms of microbial communities in the treeline ecotone

Our study supported the second hypothesis that the dominant assembly processes in bacterial and fungal communities were different. The bacterial community tended to follow a deterministic distribution pattern, with homogeneous selection being predominant. In contrast, the fungal community exhibited a more stochastic distribution pattern, with dispersal limitation being the dominant factor. Soil pH affects microbial assembly processes on multiple dimensions, with extreme soil pH tending to impose strong environmental selection pressures, while relatively neutral environments (e.g., pH ranged from 6.5 to 7.5 for bacteria) lead to less phylogenetic clustering and greater stochasticity (Tripathi et al., 2018; Luan et al., 2023). In the area of this study, pH values of the treeline ecotone ranged from 5.0 to 5.5, which was optimal for the growth of fungi but not bacteria, thus resulting in higher stochasticity in fungal assembly processes. Furthermore, fungi heavily rely on spores for reproduction and dispersal, which are subject to external media (e.g., wind and water), leading to more significant dispersal limitations than bacteria (Hassett et al., 2015).

Despite the different dominant assembly processes in bacterial and fungal communities, we found that the effects of treeline shift were uniform, both increasing the stochasticity of microbial communities. Analyzing the relationship between community assembly processes and environmental factors revealed that increasing soil nutrient levels (${\text{NH}}_{\text{4}}^{+}$-N) significantly enhanced the proportion of homogeneous selection in bacterial community. Shrublands above the treeline possessed relatively abundant soil nutrients, which created a more uniform environment for microbial survival and led to a homogenization effect. This phenomenon is consistent with reports on microbial assembly mechanisms in lakes and bioreactors, where homogeneous selection increased with higher concentrations of dissolved organic matter (Li et al., 2022, 2024c). Additionally, the significantly higher soil temperatures in shrublands, acting as a deterministic force, enhanced the role of deterministic processes (homogeneous selection) in community assembly and reduced the impact of stochastic processes (Zhou et al., 2014; Ning et al., 2020; Li et al., 2024a). In this study, compared to coniferous forests, shrublands had higher soil moisture content, which not only softened the soil texture and enhanced the soil porosity, but also served as a medium that promoted the dispersal capability of fungal spores and mycelia (Fu et al., 2023). Coupled with the increase in soil temperature, which boosted microbial motility, these factors collectively contributed to the lower influence of dispersal limitation on the community assembly process at the upper limit of the treeline. Some studies also indicated that the ecological processes of dispersal limitation could make communities more opportunistic, shifting from K-strategists to r-strategists (Telford et al., 2006; Ke et al., 2023). This finding aligns with our observations, where the lower treeline limit with higher dispersal limitation accumulated more r-strategists.

5 Conslusions

Our study revealed the remarkable responses of soil microbial communities in the treeline ecotone. Such responses were different between bacterial and fungal communities. Compared to the slope aspect, microbal community responded greatly to the treeline shift (i.e., between above and below the treeline). These results suggest that the microbial ecological effects induced by treeline shift due to climate warming are far more significant than those caused by slope aspect, indicating that microtopography may be a negligible factor when investigating microbial distribution patterns in different vegetation types. Our research enhances the understanding of microbial ecology in treeline ecosystems and provides indicative insights into how microbes in alpine ecosystems might respond to climate change and the potential ecological consequences that may ensue.

References

[1] Aakala, T., Hari, P., Dengel, S., Newberry, S.L., Mizunuma, T., Grace, J., 2014. A prominent stepwise advance of the tree line in north‐east Finland. Journal of Ecology102, 1582–1591.

[2] Agler, M.T., Ruhe, J., Kroll, S., Morhenn, C., Kim, S.T., Weigel, D., Kemen, E.M., 2016. Microbial hub taxa link host and abiotic factors to plant microbiome variation. PLoS Biology14, e1002352.

[3] Brown, J.H., Gillooly, J.F., Allen, A.P., Savage, V.M., West, G.B., 2004. Toward a metabolic theory of ecology. Ecology85, 1771–1789.

[4] Callahan, B.J., McMurdie, P.J., Holmes, S.P., 2017. Exact sequence variants should replace operational taxonomic units in marker-gene data analysis. The ISME Journal11, 2639–2643.

[5] Callahan, B.J., McMurdie, P.J., Rosen, M.J., Han, A.W., Johnson, A.J.A., Holmes, S.P., 2016. DADA2: high-resolution sample inference from Illumina amplicon data. Nature Methods13, 581–583.

[6] Catalán, N., Rofner, C., Verpoorter, C., Pérez, M.T., Dittmar, T., Tranvik, L., Sommaruga, R., Peter, H., 2024. Treeline displacement may affect lake dissolved organic matter processing at high latitudes and altitudes. Nature Communications15, 2640.

[7] Chen, H.Y., Jing, Q.F., Liu, X., Zhou, X.H., Fang, C.M., Li, B., Zhou, S.R., Nie, M., 2022. Microbial respiratory thermal adaptation is regulated by r-/K-strategy dominance. Ecology Letters25, 2489–2499.

[8] Cheng, W.X., Zhang, Q.L., Coleman, D.C., Ronald Carroll, C., Hoffman, C.A., 1996. Is available carbon limiting microbial respiration in the rhizosphere? Soil Biology and Biochemistry 28, 1283–1288.

[9] Cleveland, C.C., Nemergut, D.R., Schmidt, S.K., Townsend, A.R., 2007. Increases in soil respiration following labile carbon additions linked to rapid shifts in soil microbial community composition. Biogeochemistry82, 229–240.

[10] Coyte, K.Z., Schluter, J., Foster, K.R., 2015. The ecology of the microbiome: networks, competition, and stability. Science350, 663–666.

[11] Dai, T.J., Wen, D.H., Bates, C.T., Wu, L.W., Guo, X., Liu, S., Su, Y.F., Lei, J.S., Zhou, J.Z., Yang, Y.F., 2022. Nutrient supply controls the linkage between species abundance and ecological interactions in marine bacterial communities. Nature Communications13, 175.

[12] de Boer, W., Folman, L.B., Summerbell, R.C., Boddy, L., 2005. Living in a fungal world: impact of fungi on soil bacterial niche development. FEMS Microbiology Reviews29, 795–811.

[13] Deng, Y., Zhang, P., Qin, Y.J., Tu, Q.C., Yang, Y.F., He, Z.L., Schadt, C.W., Zhou, J.Z., 2016. Network succession reveals the importance of competition in response to emulsified vegetable oil amendment for uranium bioremediation. Environmental Microbiology18, 205–218.

[14] Devi, N., Hagedorn, F., Moiseev, P., Bugmann, H., Shiyatov, S., Mazepa, V., Rigling, A., 2008. Expanding forests and changing growth forms of Siberian larch at the Polar Urals treeline during the 20th century. Global Change Biology14, 1581–1591.

[15] Dial, R.J., Maher, C.T., Hewitt, R.E., Wockenfuss, A.M., Wong, R.E., Crawford, D.J., Zietlow, M.G., Sullivan, P.F., 2024. Arctic sea ice retreat fuels boreal forest advance. Science383, 877–884.

[16] Ding, J.J., Zhang, Y.G., Deng, Y., Cong, J., Lu, H., Sun, X., Yang, C.Y., Yuan, T., Van Nostrand, J.D., Li, D.Q., Zhou, J.Z., Yang, Y.F., 2015. Integrated metagenomics and network analysis of soil microbial community of the forest timberline. Scientific Reports5, 7994.

[17] Fajardo, A., McIntire, E.J.B., 2012. Reversal of multicentury tree growth improvements and loss of synchrony at mountain tree lines point to changes in key drivers. Journal of Ecology100, 782–794.

[18] Fan, Q.P., Liu, K.F., Wang, Z.L., Liu, D., Li, T., Hou, H.Y., Zhang, Z.J., Chen, D.H., Zhang, S., Yu, A.L., Deng, Y.C., Cui, X.Y., Che, R.X., 2024. Soil microbial subcommunity assembly mechanisms are highly variable and intimately linked to their ecological and functional traits. Molecular Ecology33, e17302.

[19] Fetzer, J., Moiseev, P., Frossard, E., Kaiser, K., Mayer, M., Gavazov, K., Hagedorn, F., 2024. Plant–soil interactions alter nitrogen and phosphorus dynamics in an advancing subarctic treeline. Global Change Biology30, e17200.

[20] Fouquier, J., Rideout, J.R., Bolyen, E., Chase, J., Shiffer, A., McDonald, D., Knight, R., Caporaso, J.G., Kelley, S.T., 2016. ghost-tree: creating hybrid-gene phylogenetic trees for diversity analyses. Microbiome4, 11.

[21] Franke, A.K., Bräuning, A., Timonen, M., Rautio, P., 2017. Growth response of Scots pines in polar-alpine tree-line to a warming climate. Forest Ecology and Management399, 94–107.

[22] Fu, F.W., Li, J.R., Li, Y.Y., Chen, W.S., Ding, H.H., Xiao, S.Y., 2023. Simulating the effect of climate change on soil microbial community in an Abies georgei var. smithii forest. Frontiers in Microbiology14, 1189859.

[23] Gaudel, G., Xing, L., Shrestha, S., Poudel, M., Sherpa, P., Raseduzzaman, M., Zhang, X.F., 2024. Microbial mechanisms regulate soil organic carbon mineralization under carbon with varying levels of nitrogen addition in the above-treeline ecosystem. Science of the Total Environment917, 170497.

[24] Genre, A., Lanfranco, L., Perotto, S., Bonfante, P., 2020. Unique and common traits in mycorrhizal symbioses. Nature Reviews Microbiology18, 649–660.

[25] Hassani, M.A., Durán, P., Hacquard, S., 2018. Microbial interactions within the plant holobiont. Microbiome6, 58.

[26] Hassett, M.O., Fischer, M.W.F., Money, N.P., 2015. Mushrooms as rainmakers: how spores act as nuclei for raindrops. PLoS One10, e0140407.

[27] Huang, Q., Wang, B.R., Shen, J.K., Xu, F.J., Li, N., Jia, P.H., Jia, Y.J., An, S.S., Amoah, I.D., Huang, Y.M., 2024. Shifts in C-degradation genes and microbial metabolic activity with vegetation types affected the surface soil organic carbon pool. Soil Biology and Biochemistry192, 109371.

[28] Ke, W.S., Li, C.X., Zhu, F., Luo, X.H., Li, X., Wu, C., Hartley, W., Xue, S.G., 2023. The assembly process and co-occurrence patterns of soil microbial communities at a lead smelting site. Science of the Total Environment894, 164932.

[29] Körner, C., Paulsen, J., 2004. A world-wide study of high altitude treeline temperatures. Journal of Biogeography31, 713–732.

[30] Lai, J.S., Zou, Y., Zhang, J.L., Peres‐Neto, P.R., 2022. Generalizing hierarchical and variation partitioning in multiple regression and canonical analyses using the rdacca. hp R package. Methods in Ecology and Evolution13, 782–788.

[31] Li, J.J., Fan, M.C., Wei, Z.H., Zhang, K., Ma, X., Shangguan, Z.P., 2024a. Broad environmental adaptation of abundant microbial taxa in Robinia pseudoacacia forests during long-term vegetation restoration. Environmental Research242, 117720.

[32] Li, S.D., Fang, J., Zhu, X.S., Spencer, R.G.M., Álvarez -Salgado, X.A., Deng, Y.C., Huang, T., Yang, H., Huang, C.C., 2022. Properties of sediment dissolved organic matter respond to eutrophication and interact with bacterial communities in a plateau lake. Environmental Pollution301, 118996.

[33] Li, X.W., Li, X.L., Shi, Y., Zhao, S.J., Liu, J.L., Lin, Y.Y., Li, C.L., Zhang, C.H., 2024b. Effects of microtopography on soil microbial communities in alpine meadows on the Qinghai-Tibetan Plateau. CATENA239, 107945.

[34] Li, W.J., Xia, Y., Li, N., Chang, J., Liu, J., Wang, P., He, X.W., 2024c. Temporal assembly patterns of microbial communities in three parallel bioreactors treating low-concentration coking wastewater with differing carbon source concentrations. Journal of Environmental Sciences137, 455–468.

[35] Liang, E.Y., Wang, Y.F., Piao, S.L., Lu, X.M., Camarero, J.J., Zhu, H.F., Zhu, L.P., Ellison, A.M., Ciais, P., Peñuelas, J., 2016. Species interactions slow warming-induced upward shifts of treelines on the Tibetan Plateau. Proceedings of the National Academy of Sciences of the United States of America113, 4380–4385.

[36] Lu, X.M., Liang, E.Y., Wang, Y.F., Babst, F., Camarero, J.J., 2021. Mountain treelines climb slowly despite rapid climate warming. Global Ecology and Biogeography30, 305–315.

[37] Luan, L., Shi, G.P., Zhu, G.F., Zheng, J., Fan, J.B., Dini-Andreote, F., Sun, B., Jiang, Y.J., 2023. Biogeographical patterns of abundant and rare bacterial biospheres in paddy soils across East Asia. Environmental Microbiology25, 294–305.

[38] Martin, M., 2011. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet Journal17, 10–12.

[39] Mayor, J.R., Sanders, N.J., Classen, A.T., Bardgett, R.D., Clément, J.C., Fajardo, A., Lavorel, S., Sundqvist, M.K., Bahn, M., Chisholm, C., Cieraad, E., Gedalof, Z., Grigulis, K., Kudo, G., Oberski, D.L., Wardle, D.A., 2017. Elevation alters ecosystem properties across temperate treelines globally. Nature542, 91–95.

[40] Miehe, G., Miehe, S., Vogel, J., Co, S., La, D., 2007. Highest treeline in the northern hemisphere found in southern Tibet. Mountain Research and Development27, 169–173.

[41] Ning, D.L., Yuan, M.T., Wu, L.W., Zhang, Y., Guo, X., Zhou, X.S., Yang, Y.F., Arkin, A.P., Firestone, M.K., Zhou, J.Z., 2020. A quantitative framework reveals ecological drivers of grassland microbial community assembly in response to warming. Nature Communications11, 4717.

[42] Osburn, E.D., Aylward, F.O., Barrett, J.E., 2021. Historical land use has long-term effects on microbial community assembly processes in forest soils. ISME Communications1, 48.

[43] Parsapour, M.K., Kooch, Y., Hosseini, S.M., Alavi, S.J., 2018. C and N cycle monitoring under Quercus castaneifolia plantation. Forest Ecology and Management427, 26–36.

[44] Peay, K.G., Baraloto, C., Fine, P.V.A., 2013. Strong coupling of plant and fungal community structure across western Amazonian rainforests. The ISME Journal7, 1852–1861.

[45] Piton, G., Allison, S.D., Bahram, M., Hildebrand, F., Martiny, J.B.H., Treseder, K.K., Martiny, A.C., 2023. Life history strategies of soil bacterial communities across global terrestrial biomes. Nature Microbiology8, 2093–2102.

[46] Razanamalala, K., Razafimbelo, T., Maron, P.A., Ranjard, L., Chemidlin, N., Lelièvre, M., Dequiedt, S., Ramaroson, V.H., Marsden, C., Becquer, T., Trap, J., Blanchart, E., Bernard, L., 2018. Soil microbial diversity drives the priming effect along climate gradients: a case study in Madagascar. The ISME Journal12, 451–462.

[47] Sigdel, S.R., Wang, Y.F., Camarero, J.J., Zhu, H.F., Liang, E.Y., Peñuelas, J., 2018. Moisture-mediated responsiveness of treeline shifts to global warming in the Himalayas. Global Change Biology24, 5549–5559.

[48] Singh, B.K., Bardgett, R.D., Smith, P., Reay, D.S., 2010. Microorganisms and climate change: terrestrial feedbacks and mitigation options. Nature Reviews Microbiology8, 779–790.

[49] Sokol, N.W., Sanderman, J., Bradford, M.A., 2019. Pathways of mineral‐associated soil organic matter formation: Integrating the role of plant carbon source, chemistry, and point of entry. Global Change Biology25, 12–24.

[50] Sokol, N.W., Slessarev, E., Marschmann, G.L., Nicolas, A., Blazewicz, S.J., Brodie, E.L., Firestone, M.K., Foley, M.M., Hestrin, R., Hungate, B.A., Koch, B.J., Stone, B.W., Sullivan, M.B., Zablocki, O., LLNL Soil Microbiome Consortium, Pett-Ridge, J., 2022. Life and death in the soil microbiome: how ecological processes influence biogeochemistry. Nature Reviews Microbiology20, 415–430.

[51] Stegen, J.C., Lin, X.J., Fredrickson, J.K., Chen, X.Y., Kennedy, D.W., Murray, C.J., Rockhold, M.L., Konopka, A., 2013. Quantifying community assembly processes and identifying features that impose them. The ISME Journal7, 2069–2079.

[52] Tedersoo, L., Sánchez-Ramírez, S., Kõljalg, U., Bahram, M., Döring, M., Schigel, D., May, T., Ryberg, M., Abarenkov, K., 2018. High-level classification of the Fungi and a tool for evolutionary ecological analyses. Fungal Diversity90, 135–159.

[53] Telford, R.J., Vandvik, V., Birks, H.J.B., 2006. Dispersal limitations matter for microbial morphospecies. Science312, 1015.

[54] Tripathi, B.M., Stegen, J.C., Kim, M., Dong, K., Adams, J.M., Lee, Y.K., 2018. Soil pH mediates the balance between stochastic and deterministic assembly of bacteria. The ISME Journal12, 1072–1083.

[55] van der Heijden, M.G.A., Bardgett, R.D., van Straalen, N.M., 2008. The unseen majority: soil microbes as drivers of plant diversity and productivity in terrestrial ecosystems. Ecology Letters11, 296–310.

[56] Wang, X.Y., Wang, T., Xu, J.F., Shen, Z.H., Yang, Y.P., Chen, A.P., Wang, S.P., Liang, E.Y., Piao, S.L., 2022. Enhanced habitat loss of the Himalayan endemic flora driven by warming-forced upslope tree expansion. Nature Ecology & Evolution6, 890–899.

[57] Yang, Y.Y., Chen, Q.Q., Zhou, Y., Yu, W., Shi, Z., 2024. Soil bacterial community composition and function play roles in soil carbon balance in alpine timberline ecosystems. Journal of Soils and Sediments24, 323–336.

[58] Yang, T., Tedersoo, L., Liu, X., Gao, G.F., Dong, K., Adams, J.M., Chu, H.Y., 2022. Fungi stabilize multi-kingdom community in a high elevation timberline ecosystem. iMeta1, e49.

[59] Ye, G.P., Wang, Y.Y., Cui, X.D., Jin, Y., Hu, H.W., Liu, J., Guo, Z.Y., Lin, Y.X., 2024. High stochasticity in rare bacterial community assembly in rice-wheat rotation soils at a regional scale. Soil Biology and Biochemistry195, 109479.

[60] Yuan, M.M., Guo, X., Wu, L.W., Zhang, Y., Xiao, N.J., Ning, D.L., Shi, Z., Zhou, X.S., Wu, L.Y., Yang, Y.F., Tiedje, J.M., Zhou, J.Z., 2021. Climate warming enhances microbial network complexity and stability. Nature Climate Change11, 343–348.

[61] Zhang, C.F., de Pasquale, S., Hartman, K., Stanley, C.E., Berendsen, R.L., van der Heijden, M.G.A., 2024. The microbial contribution to litter decomposition and plant growth. Environmental Microbiology Reports16, e13205.

[62] Zhao, D.H., Shen, C.C., Zhang, Z.M., Wang, J.C., Zhang, L.M., Chen, B.D., Sun, G.X., Ge, Y., 2024. Interactions between soil bacterial communities, assembly processes and microbial functions along the elevational gradient. CATENA235, 107698.

[63] Zhou, J.Z., Deng, Y., Zhang, P., Xue, K., Liang, Y.T., Van Nostrand, J.D., Yang, Y.F., He, Z.L., Wu, L.Y., Stahl, D.A., Hazen, T.C., Tiedje, J.M., Arkin, A.P., 2014. Stochasticity, succession, and environmental perturbations in a fluidic ecosystem. Proceedings of the National Academy of Sciences of the United States of America111, E836–E845.

[64] Zhou, H., Gao, Y., Jia, X.H., Wang, M.M., Ding, J.J., Cheng, L., Bao, F., Wu, B., 2020. Network analysis reveals the strengthening of microbial interaction in biological soil crust development in the Mu Us Sandy Land, northwestern China. Soil Biology and Biochemistry144, 107782.

Author contributions

Huijun Xu: Methodology, Investigation, Formal analysis, Data curation, Visualization, Writing – original draft. Congcong Shen: Conceptualization, Methodology, Investigation, Data curation, Writing – review & editing. Jiang Wang: Writing – review & editing. Yuan Ge: Conceptualization, Supervision, Project administration, Writing – review & editing, Funding acquisition.

Data availability

Sequences of bacterial 16S rRNA genes and fungal ITS region are available in the Science Data Bank (ScienceDB) with the data DOI: 10.57760/sciencedb.17951.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant Nos. 42177274, 42377119), and the Second Tibetan Plateau Scientific Expedition and Research Program (Grant Nos. 2019QZKK0308, 2019QZKK0306).

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Electronic supplementary material

Supplementary material is available in the online version of this article at https://doi.org/10.1007/s42832-024-0287-6 and is accessible for authorized users.

RIGHTS & PERMISSIONS

2024 Higher Education Press