Abstract
The Tibetan Plateau (TP) glaciers are influenced by monsoon and westerlies. They are highly sensitive to climate change, with atmospheric nitrogen deposition significantly impacting microbial communities and functions. However, key uncertainties persist regarding biogeography and drivers of genes and microbial communities involved in nitrogen cycling. Here, we investigate the diversity and transcriptional activity of microbial communities and nitrogen-cycling genes using 85 metagenomes and 28 metatranscriptomes from the ablation zone of 21 TP glaciers. Our results show that over 90% of the glacial taxa possess the potential for nitrogen metabolism, with ~33% exhibiting transcriptional activity. Moreover, monsoon-dominated glaciers present greater microbial diversity and higher prevalence of nitrogen-fixing genes than westerlies-dominated glaciers, linked to higher temperatures. Comparatively, the latter show elevated genomic potential for nitrous oxide emissions, likely due to higher nitrate concentrations. These findings establish temperature-nitrogen co-regulation of microbial nitrogen transformations, critical for predicting climate feedback in the extreme environment.
Similar content being viewed by others
Introduction
Atmospheric nitrogen deposition levels have significantly increased globally since the Industrial Revolution, triggering biodiversity loss, soil acidification, and water security risks1,2,3. The Tibetan Plateau (TP), often referred to as the Asian Water Tower supporting billions of people downstream4, is under pressure from increased nitrogen inputs driven by atmospheric circulations and intensified human activities5. Nitrogen inputs potentially alter the composition and function of microbial communities in fragile glacial ecosystems through nitrogen bioavailability shifts6. Specifically, insufficient nitrogen limits microbial growth and activity, while excess nitrogen leads to increased N2O emissions, which intensifies climate warming feedback7,8. Moreover, glacial ecosystems function as biogeochemical hubs, transforming atmospherically deposited nitrogen and regulating its release to downstream areas via meltwater-mediated transport9,10,11,12,13. However, the limited understanding of nitrogen cycling mechanisms across TP glaciers restricts our ability to predict climate resilience and ecological changes in these climate-sensitive regions.
Glacier surfaces, particularly in the ablation zone, host diverse microbial communities whose activities significantly influence nitrogen transformations14,15. Microbial nitrogen transformations encompass six distinct pathways: nitrogen fixation, nitrification, anaerobic ammonium oxidation (anammox), denitrification, assimilatory nitrate reduction (ANRA), and dissimilatory nitrate reduction (DNRA)16. Abundant denitrifying and nitrogen-assimilating microorganisms, along with their associated genes, have been reported in Arctic snow10, TP snow17, Antarctic glacier ice9, and global cryoconite12. In contrast, nitrogen fixation and nitrification were less prevalent on glacier surfaces. Segawa et al. reported the low expression of amoA gene for nitrification in an Asia glacier cryoconite by qPCR and metatranscriptomic sequencing18,19. Comparatively, Telling et al. noted that nitrogen inputs from nitrogen fixation on glacier surfaces are more than two orders of magnitude lower than the combined nitrogen inputs from snowmelt and rain in Svalbard11. Current understanding of microbial nitrogen cycling on TP glacier surfaces remain limited, particularly in snow and ice habitats compared to cryoconite. Cryoconite is a microbial hotspot for nitrogen cycling18,20. In comparison, snow is a transient reservoir of microorganisms and nutrients from atmospheric deposition, whereas ice preserves microbial communities shaped by post-depositional environmental selection21,22. These supraglacial habitats host distinct microbial communities and exhibit different contributions to nitrogen cycling within glacier ablation zones. Furthermore, ablation zones serve as microbial hotspots and are particularly sensitive to climate change23,24,25. Characterized by pronounced temperature fluctuations, elevated meltwater inputs, and enhanced nutrient availability, ablation zones contrast sharply with accumulation zones or englacial environments21,26. Therefore, comprehensive investigations across distinct supraglacial habitats facilitate accurate prediction of nitrogen cycling dynamics on TP glacier ablation zones under climate change.
Owing to the influence of different atmospheric circulations (monsoon and westerlies), significant spatial heterogeneity has been documented on TP glacier surfaces, including temperature, nutrients, and microbial community composition27,28,29,30. Atmospheric circulation not only directly affects the dispersal of microorganisms, but also affects microbial communities by controlling climatic conditions and nutrient inputs27,28. Previous studies have revealed a higher diversity of airborne bacteria and antibiotic resistance genes in the monsoon-dominated glaciers than those in the westerlies-dominated glaciers29,30. In addition, glaciers in monsoon-dominated regions exhibited a greater rate and magnitude of warming, while those in regions influenced by westerlies demonstrated higher nitrogen deposition27,28. However, while existing research has compared microbial nitrogen cycling between TP and polar glaciers12, critical intra-Plateau contrasts remain unexplored. This knowledge gap hinders predictions of climate-sensitive nitrogen dynamics, including potential N2O flux variations and meltwater-mediated nitrogen transport. Addressing this gap is essential for understanding regional climate feedback and informing water resource management across Asia’s headwaters.
The taxonomy, distribution patterns, and environmental controls of nitrogen-cycling microorganisms remain poorly explored across TP glaciers. This study combines metagenomic, metatranscriptomic, and biogeochemical analyses to examine the connections between nitrogen-cycling microbial communities, functional genes, and regional environmental gradients. Our results demonstrate that temperature and nitrate concentrations jointly shape the spatial distribution of genes and microbial communities involved in nitrogen cycling. This relationship influences ecosystem-level nitrogen retention and export patterns, with implications for understanding future climate change impacts on glacial nutrient dynamics. By clarifying these connections, this work contributes to improved projections of nitrogen-related processes (including N2O fluxes and meltwater nutrient transport) and supports water resource management in Asia’s glacier-fed river systems. More broadly, these findings highlight glaciers as active biogeochemical reactors rather than passive ice reservoirs, offering mechanistic insights into cryosphere-mediated nitrogen cascades in a warming world.
Results
Glacial taxa diversity and nitrogen-cycling potential
From 85 metagenomic samples (cryoconite, ice, and snow) across the ablation zone of 21 TP glacier surfaces31 (Supplementary Fig. S1 and Supplementary Data 2), 758 medium-high quality species-level bins (SGBs) with an average mapping rate of nearly 50% against 12,022,728,908 clean reads were obtained. These SGBs spanned 25 known bacterial phyla and three archaeal phyla (i.e., Halobacteriota, Thermoplasmatota and Thermoproteota) (Fig. 1A; and Supplementary Data 3). Predominant phyla included Proteobacteria (301 SGBs, 40%), Actinobacteriota (114 SGBs, 15%) and Bacteroidota (106 SGBs, 14%).
A Phylogenetic tree was generated using 758 SGBs. The outer ring color strip represents the regions (westerlies and monsoon) and habitats (cryoconite, ice, and snow) from which the SGBs were obtained. Bootstraps are shown in the pink triangle and are based on 1000 replicated trees. Bootstrap values are indicated by pink triangles and are based on 1000 replicated trees. B Potential nitrogen-cycling species associated with six nitrogen-cycling pathways. Each pathway is labeled with the key gene(s) involved and the number of SGBs identified. C Taxonomic composition of SGBs containing key nitrogen-cycling genes. D Simplified upset plot showing the overlap in SGB numbers across different nitrogen-cycling pathways. The figure at the top displays the number of SGBs associated with a single nitrogen-cycling pathway, and the figure on the right shows the number of SGBs with the potential to participate in multiple nitrogen-cycling pathways. Abbreviations: SGB, species-level genome bin.
To reveal microbial nitrogen-cycling process, we retrieved the nitrogen-cycling genes (NCGs) from these SGBs. Only nine out of 68 nitrogen-cycling genes were absent across the identified SGBs, including bacterial amoA, archaeal amoAB genes for ammoxidation, nxrA and nxrB genes for nitrite oxidation, hdh and hzo genes for hydrazine oxidation into nitrogen, pmoA and pmoB genes. In addition to genes involved in organic nitrogen degradation and synthesis (e.g., glutamate dehydrogenase and urease), we further investigated 15 key genes from six inorganic nitrogen transformation pathways (see “Methods”). Our result showed that most of SGBs (708, 93.4%) contained at least one NCG (Fig. 1B). Only 51 SGBs did not detect any NCG, most of which belonged to Patescibacteria (14) that are usually associated with parasitic lifestyle32. Most SGBs (84.6%) contained NCGs involved in the assimilatory nitrate reduction to ammonia (ANRA) (i.e., nasA and nirA gene), followed by those containing denitrification genes nirK (58.9%) and norB (47.0%). However, only ten SGBs process the genes of nitrification and anammox (Fig.1B). These results revealed diverse and abundant microbial communities mediating nitrate reduction, whereas the genes associated with nitrification and anammox were minimal on the TP glacier ablation zones.
The SGBs with potential to catalyze nitrogen transformation distributed in diverse taxonomic lineage (Fig. 1C). SGBs associated with ANRA and denitrification mainly belong to the phylum Proteobacteria (mean proportion 45%), Actinobacteriota (16%) and Bacteroidota (12%). In genus level, ten genera with the proportion over 5% were found, including six genera (Janthinobacterium, Pseudomonas_E, Arenimonas, Massilia, Rhodoferax and Undibacterium) of phylum Proteobacteria and four genera (UBA4719, JACMOY01, Aeromicrobium and Cryobacterium) of phylum Actinobacteriota (Supplementary Fig. S2 and Supplementary Data 3). For nitrogen fixation, domain linage also belonged to Proteobacteria (43%), followed by Cyanobacteria (13%) and Chloroflexota (9%). The genera LMDS01 (member of Burkholderiaceae family), CAHJXG01 (member of Acetobacteraceae family) and Phormidesmis_A were abundant (Supplementary Fig. S2 and Supplementary Data 3). Especially, taxonomy composition of SGBs involve the DNRA (nrfA gene) was dominated by Bacteroidota (25.6%) and Myxococcota (18.6%), which were different from other nitrogen-cycling pathways (Fig. 1C). For nitrification, only seven SGBs harbored the hao gene encoding hydroxylamine oxidoreductase, belonging to seven different phyla (Verrucomicrobiota, Planctomycetota, Myxococcota, Acidobacteriota, Desulfobacterota, Chloroflexota, Armatimonadota). Additionally, three SGBs contained the hzsA gene involved anammox, belonging to the families Roseiflexaceae, Polyangiaceae, and CAJBBX01 (member of the phylum Armatimonadota). Meanwhile, 71.5% of SGBs could participate different nitrogen-cycling pathways (Fig. 1D). Overall, the microorganisms that catalyze nitrate reduction and assimilation were prevalent on the glacier ablation zones, reflecting nitrogen removal potential in supraglacial ecosystems.
Transcriptionally active SGB and nitrogen-cycling genes
Based on metatranscriptomic data, we found that 33% of SGBs with maximum coverage over 0.3 had transcriptional activity (hereafter “active SGB”) on the glacier ablation zones, which accounted for 48.6% of the community by relative abundance. These active SGBs showed the highest relative abundance in cryoconite from the monsoon-dominated region (74%) and lowest in the ice from westerlies-dominated region (27%) (Supplementary Fig. S3). Active SGBs spanned into 17 phyla, most belonging to lineages including Proteobacteria (96), Actinobacteriota (48), Bacteroidota (32) and Cyanobacteria (19). Notably, two active SGBs of methanogenic archaea belonging to the genera Methanosarcina and Methanosphaerula were also active.
To ascertain whether active SGBs are involve in nitrogen-cycling processes, we examined the transcriptional activity of NCGs in active SGBs (Fig. 2A; and Supplementary Data 4). In total, 1,018 NCGs were detected in active SGBs, of which 943 (93%) exhibited transcriptional activity. The proportion of active NCGs exceeded 85% for SGBs in all phyla, except for the phylum Firmicutes. Only one active SGBs belonging to the Firmicutes phylum demonstrated the potential for nitrous oxide reduction, but no transcriptional activity of the nosZ gene was observed. The active SGBs belonging to the Proteobacteria exhibited the highest potential and transcriptional activity of the nasA gene. However, few active NCGs were detected in non-active SGBs (Fig. 2A). A total of 2,209 NCGs were identified in these SGBs, but only 707 (32%) showed transcriptional activity.
A Heatmap showing potential and transcribed nitrogen-cycling genes in active and non-active SGBs. The bar plot at the top indicates the number of genes for each nitrogen-cycling pathway, and the stacked diagram on the right shows the proportion of transcriptionally active genes across phyla. B Transcriptionally active SGBs associated with six nitrogen-cycling pathways. Each pathway is labeled with the key gene(s), the number of transcriptionally active SGBs, and their proportion (ratio of active SGBs containing the gene to all SGBs containing the gene). C Differences between transcriptionally active and non-active SGBs in optimal growth temperature (OGT), relative abundance, and correlations with temperature. Abbreviations: SGB, species-level genome bin.
Active SGBs associated with the ANRA were the most abundant (nasA: 153 SGBs; nirA: 154 SGBs; Fig. 2B). More than half (53%) of the identified nitrogen-fixing SGBs were active, while only 12% of the DNRA associated SGBs were active (12%; Fig. 2B). For the anammox and nitrification, one active SGB affiliated with the phylum Armatimonadota contained the hzsA gene, and three active SGBs belonging to the phyla Acidobacteriota, Chloroflexota and Armatimonadota contained hao gene (Supplementary Data 4). Taxonomic composition of active and non-active SGBs catalyzed different nitrogen-cycling processes showed no apparent regional and habitat differences (Supplementary Fig. S4). Moreover, active and non-active SGBs involved with ANRA and denitrification (i.e., nasA, nirA, nirK and nosZ) were abundant across habitats and regions (Supplementary Fig. S4).
Distinct genome features (e.g., genome size, GC content, OGT: optimal growth temperature) were identified between the active and non-active SGBs. The mean OGT for all SGBs was ~29 °C (ranging from 5 ~ 103 °C). The highest OGT was found in an archaeal SGB of the genus Pyrobaculum, known for its thermophilic properties33. Conversely, the lowest OGT was found in bacterial SGBs of genus CAILKP01 (member of Phycisphaerales order). Since only two archaeal SGBs were active, we only compared genome features between active and non-active SGBs in bacterial communities. Although the genome size showed no significant difference, our results revealed a significantly higher GC content and OGT in active bacterial SGBs (Two-sided Wilcox test; p < 0.05; Supplementary Fig. S5). Moreover, the relative abundance of active SGBs was higher than that of non-active SGB, with the former also exhibited a stronger correlation with temperature (Fig. 2C; and Supplementary Fig. S6). These results indicate significant genetic and physiological differences between active and non-active SGBs, despite their similar nitrogen-cycling capabilities.
Biogeographic patterns of taxa and genes involved in nitrogen cycling
Our results revealed significant variations in alpha diversity and composition of microbial community between the monsoon- and westerlies-dominated glacier ablation zones. A significant higher taxonomic α-diversity was found in ice and snow of the monsoon-dominated glaciers than that of the westerlies-dominated glaciers (Two-sided Wilcox test; both p < 0.05; Fig. 3A). Significant regional differences in community structure were also found in all habitats (PERMANOVA, Cryoconite: adonis R2 = 0.219, p = 0.001; Ice: adonis R2 = 0.104, p = 0.001; Snow: adonis R2 = 0.121, p = 0.001; Supplementary Fig. S7). However, habitats differences explained a higher proportion of community variations than regional differences (Habitat: adonis R2 = 0.138, p < 0.001; Region: adonis R2 = 0.057, p < 0.001; Fig. 3B). Furthermore, the metagenomic relative abundance of SGBs associated with different nitrogen-cycling pathways showed significant regional differences (Fig. 3C). In the cryoconite, monsoon-dominated glaciers exhibited significantly higher relative abundance of SGBs involved in ANRA (nasA), nitrogen fixation (nifH) and anammox (hzsA) compared to westerlies-dominated glaciers (Two-sided Wilcox test, both p < 0.05). In the ice, westerlies-dominated glaciers showed higher relative abundance of SGBs containing N2O production gene (norB) but lower relative abundance of SGBs associated with nitrate reduction (napA) than monsoon-dominated glaciers (Two-sided Wilcox test, both p < 0.05). In the snow, westerlies-dominated glaciers showed higher relative abundance of SGBs contained norB and nirK genes than monsoon-dominated glaciers (Two-sided Wilcox test, both p < 0.05).
A Differences in alpha diversity of the SGB community between regions across habitats (Two-sided Wilcox test). B NMDS plot showing differences in community composition across habitats (indicated by color) and regions (indicated by shape). C Differences in the relative abundance of SGBs containing key nitrogen-cycling genes between regions across habitats (Two-sided Wilcox test). The dotted brown line indicates the 1% scale. In panels A, C the data are represented as boxplots in which the middle line is the median, the lower and upper hinges correspond to the first and third quartiles, the upper whisker extends from the hinge to the highest value no further than 1.5 × interquartile range (IQR) from the hinge and the lower whisker extends from the hinge to the lowest value no further than 1.5 × IQR of the hinge. Samples sizes for each group are shown below: Cryoconite: n = 9 for Monsoon group and n = 11 for Westerlies group; Ice: n = 25 for Monsoon group and n = 14 for Westerlies group; Snow: n = 16 for Monsoon group and n = 10 for Westerlies group. Asterisks represent different significance levels obtained after a Two-sided Wilcox test. Significance levels: **** (p ≤ 0.0001); *** (p ≤ 0.001); ** (p ≤ 0.01); * (p < 0.05); ns denotes p > 0.05. Abbreviations: NMDS, non-metric multidimensional scaling; SGB, species-level genome bin.
Beside the regional differences in the SGB community, we further examine biogeographic pattern of the all nitrogen-cycling genes based on open reading frame (ORF) annotation. A total of 86,316 nonredundant ORFs were annotated into 62 nitrogen cycling genes. In addition, 43.5% of non-redundancy ORFs were considered as transcriptional active, while no RNA transcript for ORFs associated with hzo, anfG, and narW genes were detected. For active genes, the glnA gene, which encodes the glutamine synthetase had highest average TPM (94,926) in the metatranscriptomes (Fig. 4A). Beside genes of organic nitrogen degradation and synthesis, the nirK gene showed highest expression with average TPM 72,618, followed by the nasA gene (60,204). For DNRA, the nirB gene showed highest average TPM (46,705) than other genes of DNRA. For nitrogen fixation, the expression of nifD gene (32,612) encoding MoFe nitrogenase was higher than the nifH gene encoding Fe nitrogenase (8634). As for the genes of nitrification, the bacterial amoC gene had highest expression with 163 average TPM. Overall, our results showed that genes of denitrification and ANRA were highly transcribed (Fig. 4A).
A Expression levels of nitrogen-cycling genes in metatranscriptomes. B Regional differences in gene expression based on partial least squares-discriminant analysis (PLS-DA) in the cryoconite, ice (C) and snow (D).
Alpha diversity patterns of the active NCGs mirrored those of the SGB community (Supplementary Fig. S8). Ice from the monsoon-dominated glaciers exhibited significantly higher Shannon index compared to the westerlies-dominated glaciers (Two-sided Wilcox test, p < 0.001). Partial least squares-discriminant analysis (PLS-DA) conducted separately on cryoconite, ice and snow samples confirmed that regional difference strongly influenced NCGs composition (the goodness of prediction of PLS-DA model, Q2 = 0.58 for cryoconite, Q2 = 0.656 for ice, Q2 = 0.415 for snow; Fig. 4B–C). The PLS-DA model identified 24, 24, and 19 regionals differentially NCGs with VIP > 1 (Variable Importance for the Projection) for the cryoconite, ice, and snow, respectively (Supplementary Fig. S9 and Supplementary Data 5). Importantly, besides NCGs associated with organic nitrogen degradation and synthesis, the nosZ and NR genes in the cryoconite were most prominent in the westerlies-dominated glaciers and monsoon-dominated glaciers, respectively (Supplementary Fig. S9A). In the ice, the norB and nasA genes dominated in the westerlies-dominated glaciers and monsoon-dominated glaciers, respectively (Supplementary Fig. S9B). In the snow, the pmoB and nirB genes had the greatest contribution to the westerlies-dominated glaciers and monsoon-dominated glaciers, respectively (Supplementary Fig. S9C). Additionally, enrichment patterns of nitrogen-cycling genes in the community level were mirrored those observed in the SGBs. For instance, the nirK and norB genes of the snow were enriched in the westerlies-dominated glaciers. The napA gene of ice was enriched in the monsoon-dominated glaciers and the norB gene of ice was enriched in the westerlies-dominated glaciers. In the cryoconite, the nasA, hzs and nifH genes were enriched in the monsoon-dominated glaciers.
The impact of meteorological factors on biogeographic patterns
Meteorological factors (i.e., temperature, precipitation, solar radiation, wind speed) showed significant difference between TP glaciers in monsoon- and westerlies-dominated regions. Specifically, the monsoon-dominated glaciers exhibited significantly higher temperature and precipitation but significantly lower wind speed and solar radiation compared to the westerlies-dominated glaciers (Two-sided Wilcox test, all p < 0.05; Supplementary Fig. S10). These differences contributed to significant variations in alpha diversity (Shannon and PD: phylogenetic diversity) of the microbial communities and nitrogen-cycling genes across habitats and regions (Supplementary Fig. S11 and S12).
To quantify the impact of meteorological factors on community diversity across habitat types, we applied linear mixed-effects models with alpha diversity of microbial communities as a response variable and habitat type as a random effect (detail see Methods). Temperature emerged as a pivotal driver, exhibiting a greater influence on both the Shannon index and phylogenetic diversity of microbial communities compared to other meteorological factors (Supplementary Data 6). Additionally, solar radiation and wind speed also exerted notable effects, contributing to variations in Shannon index across habitats. For the Shannon index, the optimal model showed temperature, solar radiation, and wind speed were significant influence factors, together explaining 40% of the variations. Meanwhile, the optimal model of phylogenetic diversity showed that temperature and solar radiation explain 39% of the variations. When linear fitting was applied between meteorological factors and community diversity across habitats, significant correlations were also observed: temperature was positively correlated with community diversity in the ice (Shannon, R2 = 0.43, p < 0.001; PD, R2 = 0.30, p < 0.001), while solar radiation was positively correlated in the cryoconite (Shannon, R2 = 0.64, p < 0.001; PD, R2 = 0.23, p < 0.001; Supplementary Fig. S11). Wind speed showed contrasting effects between ice and cryoconite habitats, being negatively correlated with Shannon diversity in the ice (R2 = 0.24, p = 0.001) but positively correlated in the cryoconite (R2 = 0.32, p = 0.009; Supplementary Fig. S11). Notably, no significant correlations were found in the snow (Supplementary Fig. S11; Supplementary Result). These findings underscore habitat-specific responses of microbial communities to meteorological factors. Moreover, these correlation patterns were also observed at the community level. The Shannon index of nitrogen-cycling genes was significantly positively correlated with temperature in the ice (R2 = 0.38, p < 0.001; Supplementary Fig. S12). Spatial factors such as latitude and altitude could also influence microbial communities. However, our results showed that temperature explained more variation than latitude in the Shannon diversity of microbial communities (Temperature, R2 = 0.43; Latitude, R2 = 0.39, both p < 0.001; Supplementary Fig. S13). Our findings indicate that temperature is a primary driver of diversity in genes and microbial communities involved in nitrogen cycling, with different habitats exhibiting specific responses to meteorological factors.
To decipher the underlying factors contributing to the differences in microbial diversity across habitats, we conducted an in-depth examination of the microbial community assembly process using the neutral community model (NCM), Nearest taxon index (NTI) and normalized stochasticity ratio (NST) analysis. Different approaches showed similar results that deterministic process shaped microbial communities on the glacier surface (Supplementary Result and Fig. S14). The neutral community model described the relationship between the occurrence frequency of SGBs and their relative abundance in different habitats (Supplementary Fig. S14A-C). The relative contribution of stochastic processes was the lowest in the cryoconite (NCM R2 = 0.277) and the highest in the snow (NCM R2 = 0.593). The NTI showed a phylogenetic clustered distribution in all habitats (NTI > 2; Supplementary Fig. S14D). In addition, the NST ratio indicated that the microbial communities in cryoconite and ice were predominantly influenced by deterministic assembly processes (NST < 50%), with cryoconite exhibiting a significantly higher influence from deterministic processes (lowest NST ratio) than other habitats (Two-sided Wilcox test; p < 0.001; Supplementary Fig. S14E).
Microbial nitrous oxide emission potential is influenced by temperature and nitrate concentrations
Given the influence of temperature on the alpha diversity of the microbial community in the ablation zone ice, we further investigate the microbial feedbacks (e.g. microbial nitrous oxide emission potential) to climate warming involving nitrogen cycle. Nitrous oxide, a potent greenhouse gas, can be both produced (norB gene) and consumed (nosZ gene) by microbes. The abundance ratio of transcriptionally active norB and nosZ genes was used to assess the genomic potential for N2O emission, as these genes are well-established markers for denitrification-derived N2O fluxes34. A ratio greater than 1 indicates N2O emission, while less than 1 indicates N2O consumption. Our results revealed a significantly higher potential for microbial N2O emission in westerlies-dominated glaciers (mean: 1.3) compared to monsoon-dominated glaciers (mean: 0.64; Two-sided Wilcox test, p < 0.001; Fig. 5A). Moreover, temperature positively correlated with the microbial N2O emission potential (monsoon: R2 = 0.25, p = 0.011; westerlies: R2 = 0.44, p = 0.01; Fig. 5B). Across the two regions, microbial N2O emission potential was associated with the nitrate concentrations, which ranged from 0.005 to 0.22 mg/L in glacier ice (Supplementary Data 2). Linear fitting revealed that the microbial N2O emission potential was positively correlated with nitrate concentrations (R2 = 0.20, p < 0.001; Fig. 5C).
A Differences in microbial nitrous oxide (N2O) emission potential (indicated by the relative abundance ratio of norB and nosZ genes) between regions (n = 25 for Monsoon group; n = 14 for Westerlies group; Two-sided Wilcox test). Data are represented as boxplots in which the middle line is the median, the lower and upper hinges correspond to the first and third quartiles, the upper whisker extends from the hinge to the highest value no further than 1.5 × interquartile range (IQR) from the hinge and the lower whisker extends from the hinge to the lowest value no further than 1.5 × IQR of the hinge. Asterisks represent different significance levels obtained after a Two-sided Wilcox test. Significance levels: **** (p ≤ 0.0001); *** (p ≤ 0.001); ** (p ≤ 0.01); * (p < 0.05). B, C Linear fitting between N2O emission potential and temperature (B) and nitrate concentrations (C). The orange points indicate samples from westerlies-dominated region and bule points indicated samples from monsoon-dominated region. Blue lines represent linear regression lines, and shaded areas indicate 95% confidence intervals. P values are calculated by two-sided ordinary least squares linear regression (p < 0.05 indicates a significant correlation). D N2O emission potential of Tibetan Plateau glaciers. The map shows the five-year average temperature of the Tibetan Plateau (source: ERA5 dataset). Glaciers are colored based on the ratio of norB to nosZ genes. Maps were created in ArcMap 10.2. Pi charts show the glacier number in different N2O emission potential ranks in monsoon- and westerlies-dominated regions.
We assessed the source-sink dynamics of microbial N2O across TP glaciers based on the temperature-N2O emission relationship. Our results indicated that 95% of glaciers in the monsoon-dominated region acted as potential microbial N2O sinks, while 58% of glaciers in the westerlies-dominated region were potential microbial N2O sources (Fig. 5D). Especially, 6% of glaciers in the monsoon-dominated region were near to the critical transition point (0.9 <norB/nosZ ≤1), suggesting a potential shift from N2O sink to source with future warming. Notably, in the monsoon-dominated region, the potential for N2O emissions from glaciers increases from west to east, corresponding to increasing human activity.
In addition to denitrification for N2O reduction (N2O → N2), biological assimilation of N2O into biomass (N2O → NH3) has been observed in laboratory experiments and field studies35,36. While this hypothesized mechanism has similarities with nitrogen fixation, theoretical models suggest that N2O assimilation may require less energy, which could provide ecological advantages in energy-limited cold environments37. Since no genes for N2O assimilation are currently known, we analyzed the regional difference in nitrogen fixation to infer possible regional differences in putative N2O assimilation. Our results showed a significantly higher relative abundance of nifH gene in the monsoon-dominated glaciers than in westerlies-dominated glaciers (Two-sided Wilcox test, p < 0.05; Supplementary Fig. S14A), with the differences in gene abundance mainly attributed to the variations in nitrate and temperature. The relative abundance of active nifH gene positively correlated with the temperature (R2 = 0.31, p < 0.001) and negatively correlated with nitrate concentrations (R2 = 0.20, p = 0.031; Supplementary. S15B-C). Collectively, these results highlight spatially heterogeneous interactions among temperature, nitrate availability, and microbial nitrogen-cycling functions (e.g., nitrogen fixation and denitrification) on the TP glacier ablation zones.
Discussion
In this study, we characterized the diversity and transcriptional activity of microbial communities and nitrogen-cycling genes on the TP glacier ablation zones, revealing geographic patterns of microbial nitrogen-cycling driven by temperature and nitrate concentrations. These findings establish temperature-nitrate co-regulation of microbial nitrogen transformations in high-altitude supraglacial ecosystems, with potential applications for predicting climate feedbacks and greenhouse gas emissions. Our findings indicate that 93.4% of the identified SGBs are potential nitrogen-cycling taxa, with 33% of them being transcriptionally active, indicating the existence of active and diverse nitrogen-cycling taxa on the TP glacier ablation zones. This observation suggests that the capacity of microorganisms to engage in nitrogen cycling may be essential for fulfilling community needs in oligotrophic environments. Similarly, nearly 40% of genomes identified within groundwater ecosystems were associated with non-assimilatory redox reactions38. Additionally, the proportion of active SGBs observed in our study is lower than the 50% reported on the glacier surfaces of Iceland and Greenland14. However, it falls within the range observed in soil studies, where 10-40% of microorganisms were classified as potentially active39. Variations in these activity proportions may arise from methodological or regional biases. Nonetheless, these microorganisms are potentially active and significantly contribute to the biogeochemical processes occurring on glacier surfaces.
In this study, nitrogen removal taxa and genes (e.g., denitrification and nitrate assimilation) were prevalent but anammox and nitrification taxa and genes were rare on the TP glacier ablation zones (Fig. 1C; and Supplementary Results and Fig. S16). However, previous studies have suggested the presence of nitrifiers in supraglacial, englacial, and subglacial ecosystems19,40,41,42. Nevertheless, the rate of nitrification has only been estimated in the englacial ecosystem (0.74 nmol N per g per year), which is thousands of times lower than that in soils40,43. Thus, nitrification may occur on glacier surfaces, but the activity could be extremely low. Future studies integrating multi-compartment sampling with metabolic rate assays are needed to unravel glacier-scale nitrogen cycling dynamics.
Transcriptionally active SGBs were found to have higher OGT, GC content and relative abundance in the community compared to non-active SGBs (Fig. 2C), suggesting that temperature is a potent factor in influencing microbial activity and abundance. Typically, microbes with higher OGT and GC content are adapted to warmer environments44,45. Although the mean annual air temperature on glaciers is below 0 °C, seasonal and diurnal temperature fluctuations can provide potential windows for microbial activity. Given that microbial metabolic rate increased with temperature46, it is reasonable that active SGBs demonstrate stronger correlations with temperatures, which further explains their higher relative abundance compared to non-active SGBs (Supplementary Fig. S6). These findings indicate that glaciers might act as dynamic collectors and filters for allochthonous taxa rather than merely serving as shelters for autochthonous psychrophiles.
The complex interplay of atmospheric circulations in the TP glaciers creates significant spatial heterogeneity in temperature and nutrient availability27,28. Building on this foundation, we observed significant regional differences in genes and microbial communities involved in nitrogen cycling (Fig. 3C and Supplementary Fig. S8). In the monsoon-dominated region, the genes and SGBs associated with nitrate assimilation, reduction, and nitrogen fixation were more abundant in ice and cryoconite habitats than those in the westerlies-dominated region (Fig. 3C; Fig. 6). The lower nitrate concentrations in this region may explain the enrichment of genes associated with nitrate assimilation and nitrogen fixation functions, which support microbial growth under nitrogen-limited conditions. Additionally, the high relative abundance of Cyanobacteria in cryoconite may account for the observed enrichment in these processes, which have been documented for their roles in nitrate assimilation and nitrogen fixation47,48. Moreover, SGBs containing the napA gene, which encodes periplasmic nitrate reductase in aerobic denitrifiers49, were more abundant in the ice from the monsoon-dominated region compared with that from the westerlies-dominated region. Given that melted ice has a higher oxygen saturation than frozen ice50, we propose that accelerated melting in monsoon-dominated glaciers (relative to westerlies-dominated glaciers) generates oxygen-enriched conditions51, thereby favoring napA-mediated denitrification. This finding indicates that aerobic denitrifiers could be stimulated by increasing oxygen availability during glacier melting. In contrast, in the westerlies-dominated region, denitrification genes (norB and nirK) and SGBs were found to be more abundant in snow and ice habitats than those in the monsoon-dominated region (Fig. 3C; Fig. 6). The observed enrichment of these genes and SGBs may be attributed to higher nitrogen deposition in westerlies-dominated regions28,52, where excess nitrogen could stimulate denitrification. Overall, the observed enrichment of taxa and genes across different regions is significantly influenced by nitrogen availability, indicating that nitrogen deposition driven by atmospheric circulation may play an important role in microbial nitrogen cycling on the TP glacier.
Nitrogen cycle pathways and key genes are shown in cycle maps, with dotted lines indicating pathways where the proportion of nitrogen-cycling species is less than 1%, and thicker solid lines indicating higher relative abundance of species with the gene. Enriched SGBs (based on Fig. 3C) are colored by habitat and shaped by region. Rectangular interior background color gradients indicate regional differences in alpha diversity (based on Fig. 3A). Right triangles at the edges indicate high and low variations in diversity and deterministic processes between habitats, and regional variations in temperature and nitrate concentrations.
In addition to nitrogen availability, our results indicate that temperature also significantly influences the alpha diversity of microbial communities (Supplementary Fig. S11). Previous research has established that temperature exerts a considerable impact on microbial communities in various ecosystems, including forest soil53 and marine environments54. For example, microbial communities in forest soils show a positive correlation with temperature53, while species richness tends to increase notably with warming in marine ecosystems54. Moreover, gene abundances for key nitrogen-cycling genes (e.g., nirS and pmoA) increase alongside increasing mean annual temperatures in paddy soil55. This linear correlation between temperature and microbial diversity provides a quantitative basis for predicting how climate warming may affect biodiversity and ecosystem processes in cold environments. However, it is important to consider potential temperature threshold effects that could hinder further increases in diversity56.
Our findings indicate that temperature and nitrate concentration jointly influence microbial N2O emission potential and nitrogen fixation on the TP glacier ablation zones (Fig. 5A–C and Supplementary Fig. S15B–C). Interestingly, glaciers in the westerlies-dominated region exhibit lower temperatures yet show a higher potential for microbial N2O emission compared to those in the monsoon-dominated region (Fig. 5A). In monsoon-dominated glaciers, the limited total nitrogen and nitrate concentrations favor microbes with processes such as nitrogen fixation and N2O assimilation to survive under nitrogen-limited conditions (Supplementary Fig. S15B–C). The elevated temperatures in this region further facilitate these energy-intensive microbial processes. Although the enzymatic basis of N2O assimilation and its competition with nitrogen fixation remain unresolved, a prior study in the lake ecosystem has reported temperature- and substrate-dependent variations in biological N2O assimilation and nitrogen fixation potentials37. In contrast, glaciers in the westerlies-dominated region demonstrate a higher potential for microbial N2O emission, associated with higher nitrate concentrations (Fig. 5C). This aligns with soil studies showing that N2O emissions increase with elevated nitrogen inputs57,58. However, lower temperature and microbial activity in the westerlies-dominated region may limit the actual contribution of microbial N2O emissions to climate warming (Supplementary Fig. S3 and S10). While nitrate concentrations exhibit a positive correlation with microbial N2O emission potential (Fig. 5C), this relationship may not directly reflect the impact of depositional nitrogen inputs. Microbial assimilation and denitrification likely modify nitrate pools (Supplementary Results and Fig. S17), decoupling ambient nitrate concentration from external nitrogen deposition. Thus, observed correlations may reflect the joint effects of in situ microbial processing and deposition-driven enrichment, with their relative contributions modulated by region-specific microbial activity. To disentangle these mechanisms, future studies should integrate multi-species nitrogen profiling (NH4⁺, DON, PN) with process-specific isotopic tracers (e.g., δ¹⁵N/δ¹⁸O in NO3⁻), enabling explicit discrimination between depositional fluxes and microbial transformations.
N2O is a potent greenhouse gas contributing to ~6% of global warming, yet demonstrating a longer atmospheric lifetime and global warming potential than CO259. Early studies have demonstrated the significant influence of climate change on N2O emissions in the cryosphere60, but little is known regarding the N2O fluxes in glaciers. A study in the Greenland ice sheet revealed that 4.4 to 28 millimoles of CH4 per square meter per day will be released61. Thus, glaciers may contribute to the N2O budget under climate change. Our predictive modeling indicates that TP glacier ablation zones, particularly those situated in the westerly-dominated region, were potentially microbial N2O sources (Fig. 5D). Moreover, our results indicate that climate warming may lead to a potential shift in the role of glaciers in N2O dynamics, with ~8% transitioning from being a sink to a source of N2O in monsoon-dominated regions (Fig. 5D). This change is particularly evident in glaciers located in the southeastern Tibetan Plateau, where stronger anthropogenic nitrogen deposition exacerbates the situation62. Such shifts suggest the potential influence of supraglacial microbial communities on greenhouse gas dynamics.
Pronounced regional disparities in microbial community structure and nitrogen-cycling gene distribution were identified across the ablation zone ice, demonstrating significant correlations with temperature and nitrate gradients (Supplementary Figs. S11, S12). However, post-depositional alterations in the ablation zone ice, mediated by microbial succession, chemical migration, and physical recrystallization during glacial transport, also influence microbial community dynamics22. The significant divergence between ablation zone ice and fresh snow microbiomes supports the impacts of post-depositional processes (Supplementary Fig. S18A). Nevertheless, regional differences in the ice (monsoon vs. westerlies) explained 3.2-fold greater variance than the differences between the snow and ice (Supplementary Fig. S18B), reflecting temperature and nitrate availability variation across regions as primary drivers. Thus, historical depositional processes determine the initial microbial communities in the ablation zone ice, while post-depositional effects modify community composition and nitrogen-cycling potentials, but have weaker influence than the regional effects. Given the inherent microbial and metabolic heterogeneity across entire glacial systems, compounded by glacier dynamic and habitat stratification21,63, the extrapolation of our findings to unstudied glacial habitats or regional systems requires rigorous validation. Future investigations should prioritize glacier-scale transects that explicitly contrast ablation versus accumulation zones and supraglacial versus englacial habitats to elucidate glacier-scale nitrogen cycling processes.
Taken together, these results demonstrate that temperature and nitrogen availability gradients between monsoon- and westerlies-dominated regions drive distinct biogeographic patterns of genes and microbial communities involved in nitrogen cycling. Temperature emerges as the dominant regulator of microbial community activity and diversity, whereas nitrate concentrations modulate the partitioning of nitrogen transformation pathways, thereby governing regional N2O emission potential. These findings revealed previously unrecognized linkages between atmospheric circulation regimes and microbial nitrogen cycling on the ablation zone of glacier surfaces. Critically, this study suggests that warming and nitrogen deposition jointly regulate ecosystem functioning in cryospheric biomes, extending a framework previously established in soils64,65. By quantifying temperature-microbe-nitrate interactions, this work advances the predictive capacity for microbial nitrogen cycling under anthropogenic forcing, including warming and atmospheric nitrogen deposition. Such insights are critical for refining Earth system models and informing adaptive strategies to mitigate cascading impact.
Methods
Metatranscriptomic samples and metagenomic data collection
Metatranscriptomic samples were collected in the same glaciers with metagenomic data of TG2G31, but only 28 metatranscriptomic data were available due to low biomass, including 18 cryoconite, 5 ice, and 5 snow (Supplementary Fig. S1; Supplementary Data 1 and 2). The sample collection process was the same with TG2G samples. All samples were collected from the ablation zone of the glacier surfaces. Snow samples were collected within 48 h after precipitation (snow fall) to minimize post-depositional changes. Ice samples were obtained from supraglacial layers (0–20 cm depth). Briefly, snow and ice samples were collected within a square meter area using a pre-cleaned steel scoop and ice axe and then 2 mm of the sample surface was scraped with a sterile scalpel for decontamination, finally placed into 2 L sterile Whirl-Pak bags (Nasco, Fort Atkinson, USA). The cryoconite samples were collected from holes using a stainless-steel scoop and placed in pre-cleaned 1-L polycarbonate bottles (Nalgene, Thermo Fisher, USA). All samples were stored at −20 °C during the transportation from the field to the laboratory in Lhasa City (Tibet, China). Snow and ice samples were de-frozen at 4 °C at dark and then filtered through 0.2 µm polycarbonate filters (Millipore) immediately. All samples were stored at −80 °C until total RNA extraction.
Monthly averaged meteorological data of 21 glaciers from the ERA5 dataset was downloaded, which is a global atmospheric reanalysis product produced by the ECMWF (European Center for Medium-Range Weather Forecasts)66. These glaciers were distributed over the TP and can be loosely classified into two climate domains: the monsoon-dominated region is influenced by Indian monsoon and the westerlies-dominated region is influenced by westerlies51. According the location of 21 glaciers and pervious study28,30, we used the 32°N as the boundary for the monsoon-dominated glaciers and westerlies-dominated glaciers. Then we selected mainly meteorological data that could influence glaciers, including 2-meter temperature (Kelvins, K), 10-meter wind speed (meters per second, m/s), total sky direct solar radiation at the surface (joules per square meter, J/\({m}^{2}\)), total precipitation (meter of water equivalent) and evaporation (meter of water equivalent) data of each sampling site from 2016 to 2020. The NetCDF (Network Common Data Form) format dataset was downloaded from the website (https://cds.climate.copernicus.eu/datasets) and transformed to table format based on geographic coordinates by R package raster (v3.5-15)67. Finally, these meteorological data were converted to a five-year average value for each sampling point for downstream analysis. In addition, the nitrate concentrations of glacier ice and snow were measured using a Dionex ion Chromatograph System 2000 (Dionex Corp, USA) as previously described28. Detailed descriptions of the sampling sites, laboratory measurements, meteorological data and sequencing information for the metagenomic data can be found in Supplementary Data 2 and our previous work28,31,68.
RNA extraction and sequencing
Total RNA was extracted from all samples using the RNAeasy PowerSoil Total RNA Kit (QIAGEN, Hilden, Germany) according to the manufacturer’s instructions and eluted in 50 µL of RNase-free water. Residual DNA removal and RNA cleaning were conducted using DNase I (Zymo, Irvine, USA) and the RNA Clean & Concentrator −5 (Zymo, Irvine, USA), respectively. The quantity and quality of total RNA were measured using the RNA HS Assay Kit (Thermo Fisher Scientific, Waltham, USA) with a Qubit 3 (Invitrogen, Carlsbad, USA) and a Bioanalyzer 2100 (Agilent, Santa Clara, USA). Ribosomal RNA was removed using the Ribo-off rRNA Depletion Kit V2 (Vazyme, Nanjing, China), and libraries were prepared using the VAHTS Universal V6 RNA-seq Library Prep Kit (Vazyme, Nanjing, China) for Illumina, following the manufacturer’s recommendations. Then, the library was sequenced on the Illumina Novaseq 6000 platform in paired-end 2×150 mode by Guangdong Magigene Biotechnology Co.
Metatranscriptome and metagenome analysis
In total, 1.48 TB raw metatranscriptomic bases were obtained, with an average of 353.5 million reads for each sample. Raw reads were trimmed with Trimgalore (v0.39) with default parameters69. The remaining ribosome RNA reads were removed from total RNA with sortmeRNA (v4.3.6), which yielded 0.84 TB mRNA reads70. An average of 205.6 million mRNA reads were obtained for each sample (Supplementary Data 1).
Raw reads, assembled contigs and genome bins of 85 metagenomic data were obtained TG2G dataset. Raw metagenomic reads were also trimmed by Trimgalore (v0.39) with default parameters69. Assembled contigs with length >1000 bp were used to predict open reading frames (ORFs) by Prodigal (v2.6.3)71. Raw 2358 medium-high quality bins cluster into species level by dRep (v3.2.2) with parament: ANI 95, which yielded 758 species-level genome bins (SGBs)72. Taxonomy classification of each SGB was performed by GTDB-tk (v2.1.1) with the GTDB database(r207)73,74. A phylogenetic tree for the SGBs was reconstructed using the protein sequences of 43 universal single-copy genes by checkM75. The iTOL website was used to better visualize the phylogenetic tree76. We estimated the optimal growth temperature for 758 SGBs using Growthpred77. SGBs with fewer than 10 highly expressed genes were excluded. The optimal growth temperature was calculated with the script growthpred-v1.08.py, applying parameters (-r -t -c 0).
Functional annotation of SGBs and ORFs
The protein sequence of each SGB was predicted by Prokka (v1.14.6) and used to annotate the nitrogen-cycling gene by Diamond (v2.0.15) against the NCyc database78. NCyc database contains 68 nitrogen cycling genes belonging to seven pathways, including Anammox, Assimilatory nitrate reduction to ammonia (ANRA), Denitrification, Dissimilatory nitrate reduction to ammonia (DNRA), Nitrification, Nitrogen fixation, and Organic degradation and synthesis78. We used 15 key nitrogen-cycling genes to represent six nitrogen-cycling pathways expected to Organic degradation and synthesis. For instance, nitrogen fixation (nifH), nitrification (amoA, hao and nxrA), denitrification (narG, napA, nirK, nirS, norB and nosZ), ANRA (nasA and nirA), DNRA (nrfA) and anammox (hzo and hzsA). These key nitrogen-cycling genes also be used in previous studies of global cryoconite12 and Antarctic coastal79.
To unveil the diversity of nitrogen-cycling gene at the community level, only the protein sequence of ORFs with length > 100 bp were kept for annotation by Diamond (v2.0.15) against the NCyc database. A total of 113,926 ORFs were considered as potential NCGs and then dereplicated into 86,316 nonredundant NCGs by CD-hit (v4.8.1) with 100% identity80.
Estimated metagenomic abundance and transcriptional activity
To calculate the abundance of SGBs and NCGs, metagenomic clean reads were mapped to 758 SGBs by CoverM (v0.6.0; https://github.com/wwood/CoverM) and 86,316 nonredundant NCGs by Salmon (v0.13.1) with default parameters to obtain transcripts per million (TPM) abundance81. Metatranscriptomic data were used to identify the transcriptional activity of genomes and genes in pervious study38. In this study, metatranscriptomic reads were mapped to SGBs and nonredundant NCGs by bowtie2 (v2.2.5)82. The mapped resulting SAM files were converted to BAM files by Samtools (v1.15.1) with -q 30 option to keep high-quality mappings83, and then CoverM (v0.6.0) was used to calculate coverage and TPM with the threshold (--min-read-percent-identity 95 --min-read-aligned-percent 75)84. Only maximum coverage not below 30% was considered as active SGBs or NCGs85.
Community diversity and assembly process
To assess the alpha diversity, Shannon indexes was evaluated based on TPM abundance of SGBs and NCGs. Phylogenetic diversity was evaluated based on TPM abundance and phylogenic tree of SGBs. The α-diversity of microbial communities was assessed using the R package vegan (v2.5.6)86. The β-diversity of the microbial community compositions was estimated by calculating the Bray–Curtis dissimilarity. Compositional differences between the regions or habitat were assessed using permutational multivariate analysis of variance (PERMANOVA) by the adonis function in the vegan package (v2.5.6) and visualized using non-metric multidimensional scaling (NMDS)86. Partial least squares-discriminant analysis (PLS-DA) was employed to identify the distinctions in nitrogen-cycling genes between regions across habitats using the mixOmics (v6.14.1) package87, followed by variable importance in projection (VIP) analysis was conducted by ropls (v1.22.0) packages.
The neutral community model (NCM) was used to assess the influence of stochastic processes on microbial community assembly across habitats88. This model fitted the relationship between the occurrence frequency of SGBs and their relative abundance. Fitting statistics were computed with 95% confidence intervals using bootstrapping with 1,000 replicates. The parameter R2 indicates the degree to which the neutral model fits the data; an R2 value close to 1 suggests a strong alignment with stochastic processes, whereas a value near 0 indicates a predominance of deterministic factors.
The nearest taxon index (NTI) and normalized stochasticity ratios (NSTs) were calculated to validate the results of the neutral community model (NCM). NTI was used to assess the degree of phylogenetic clustering. First, the mean nearest taxon distance (MNTD) was computed using the picante package (v1.8.2)89. The NTI values correspond to the negative standardized effect size of the observed MNTD distances. NTI values greater than 2 indicate phylogenetic clustering, while NTI values less than −2 signify phylogenetic over-dispersion. The normalized stochasticity ratios for microbial communities were quantified using the NST package (v3.0.6)90. This index measures the relative influence of stochasticity versus determinism, considering scenarios where deterministic factors drive communities to be either more similar or more dissimilar than expected under random patterns. An NST value less than 0.5 indicates a more deterministic assembly, whereas a value greater than 0.5 suggests a more stochastic community assembly.
Linear mixed-effects model
To investigate the relationships between meteorological factors (such as temperature, precipitation, solar radiation, and wind speed) and alpha diversity across habitats, we modeled these relationships with linear mixed-effects models using the lmer function in the lme4 package (v1.1-29)91. We used habitats (i.e., cryoconite, ice and snow) as random effects, allowing the intercepts and slopes of the regression to vary across habitats if supported by model selection. To obtain the optimal model, we remove the autocorrelation variable based on the variance inflation factors (VIF) thresholds 10. We also performed a multimodel inference (using the function anova) to select the simplest models that explained the most variation based on Akaike’s information criterion (AIC). The marginal coefficient of determination (pseudo-R2) was calculated using the function ‘r.squaredGLMM’ of the MuMIn package (v1.46.0) to represent the variance explained by the fixed effect in the linear mixed-effects model. The optimal model for Shannon index and Phylogenetic diversity can be found in Supplementary Data 6.
Linear regression between meteorological factors and alpha diversity was plotted by ggplot2 package (v3.5.0), and then p values and R2 values were labeled by stat_poly_eq function ggpmisc package (v0.4.6)92. Based on the linear regression between temperature and microbial N2O emission potential, we predicted microbial N2O emission potential on the TP glacier ablation zones. The NetCDF format dataset of temperature was downloaded from the website (https://cds.climate.copernicus.eu/datasets) and glacier data was downloaded in pervious study93. The map visualization was conducted using ArcMap (v10.2).
Statistical analysis
The significance of the comparisons was determined using the Two-sided Wilcox test in the ggpubr (v0.4.0) package94. Stars are used to define significance level as follow: **** (p ≤ 0.0001); *** (p ≤ 0.001); ** (p ≤ 0.01); * (p < 0.05). The correlation between SGBs abundance and meteorological factor was canulated by Spearman correlation test by rcorr function of R package correlation (v0.8.1). All statistical analyses were performed using R version 4.0.3.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
The 28 metatranscriptomic data generated in this study have been deposited in the NCBI SRA database under BioProject accession number PRJNA1115801. Raw 85 metagenomic samples were available in TG2G31 dataset (BioProject accession number PRJNA813429). The analytical data including, non-redundancy ORFs and SGB genomes, are accessible on the Figshare (https://doi.org/10.6084/m9.figshare.26860918).
Code availability
Analysis code and scripts are available on GitHub (https://github.com/Hame1N/TPGMN)
References
Galloway, J. N. et al. Transformation of the nitrogen cycle: recent trends, questions, and potential solutions. Science 320, 889–892 (2008).
Liu, X. et al. Enhanced nitrogen deposition over China. Nature 494, 459–462 (2013).
Bowman, W. D., Cleveland, C. C., Halada, Ĺ, Hreško, J. & Baron, J. S. Negative impact of nitrogen deposition on soil buffering capacity. Nat. Geosci. 1, 767–770 (2008).
Yao, T. et al. The imbalance of the Asian water tower. Nat. Rev. Earth Env. 3, 618–632 (2022).
An, Y. Q. et al. Concentrations, compositions, and deposition rates of dissolved nitrogen in western China: insights from snow records. Front. Environ. Sci. 9, 9 (2022).
Vitousek, P. M., Hättenschwiler, S., Olander, L. P. & Allison, S. D. Nitrogen and nature. AMBIO 31, 97–101 (2002).
Harris, E. et al. Warming and redistribution of nitrogen inputs drive an increase in terrestrial nitrous oxide emission factor. Nat. Commun. 13, 4310 (2022).
Schmidt, S. K. et al. Microbial biogeochemistry and phosphorus limitation in cryoconite holes on glaciers across the Taylor Valley, McMurdo dry valleys, Antarctica. Biogeochemistry 158, 313–326 (2022).
Cabezas, A. et al. Ubiquity and diversity of cold adapted denitrifying bacteria isolated from diverse Antarctic ecosystems. Front. Microbiol. 13, 14 (2022).
Larose, C., Dommergue, A. & Vogel, T. M. Microbial nitrogen cycling in Arctic snowpacks. Environ. Res. Lett. 8, 035004 (2013).
Telling, J. et al. Nitrogen fixation on Arctic glaciers, Svalbard. J. Geophys. Res. Biogeosci. 116, 8 (2011).
Murakami, T. et al. Metagenomics reveals global-scale contrasts in nitrogen cycling and cyanobacterial light-harvesting mechanisms in glacier cryoconite. Microbiome 10, 14 (2022).
Gao, T. et al. Characterization, sources and transport of dissolved organic carbon and nitrogen from a glacier in the Central Asia. Sci. Total Environ. 725, 138346 (2020).
Bradley, J. A. et al. Active and dormant microorganisms on glacier surfaces. Geobiology 21, 244–261 (2023).
Stibal, M., Šabacká, M. & Žárský, J. Biological processes on glacier and ice sheet surfaces. Nat. Geosci. 5, 771–774 (2012).
Kuypers, M. M. M., Marchant, H. K. & Kartal, B. The microbial nitrogen-cycling network. Nat. Rev. Microbiol. 16, 263–276 (2018).
Yan, P. Y. et al. Diversity of snow bacteria from the Zangser Kangri glacier in the Tibetan Plateau environment. Geomicrobiol. J. 34, 37–44 (2017).
Segawa, T. et al. The nitrogen cycle in cryoconites: naturally occurring nitrification-denitrification granules on a glacier. Environ. Microbiol. 16, 3250–3262 (2014).
Segawa, T. et al. Redox stratification within cryoconite granules influences the nitrogen cycle on glaciers. FEMS Microbiol. Ecol. 96, 13 (2020).
Cook, J., Edwards, A., Takeuchi, N. & Irvine-Fynn, T. Cryoconite: the dark biological secret of the cryosphere. Prog. Phys. Geog. 40, 66–111 (2016).
Hodson, A. et al. Glacial ecosystems. Ecol. Monogr. 78, 41–67 (2008).
Xiang, S., Shang, T., Chen, Y. & Yao, T. Deposition and postdeposition mechanisms as possible drivers of microbial population variability in glacier ice. FEMS Microbiol. Ecol. 70, 165–176 (2009).
Smith, H. J. et al. Biofilms on glacial surfaces: hotspots for biological activity. NPJ Biofilms Microbiomes 2, 16008 (2016).
Halbach, L. et al. Single-cell imaging reveals efficient nutrient uptake and growth of microalgae darkening the Greenland ice sheet. Nat. Commun. 16, 1521 (2025).
Noël, B., van de Berg, W. J., Lhermitte, S. & van den Broeke, M. R. Rapid ablation zone expansion amplifies north Greenland mass loss. Sci. Adv. 5, eaaw0123 (2019).
Margesin, R. & Collins, T. Microbial ecology of the cryosphere (glacial and permafrost habitats): current knowledge. Appl. Microbiol. Biotechnol. 103, 2537–2549 (2019).
Ren, S. et al. Changes in glacier surface temperature across the Third Pole from 2000 to 2021. Remote Sens. Environ. 305, 114076 (2024).
Liu, Y. et al. A comprehensive dataset of microbial abundance, dissolved organic carbon, and nitrogen in Tibetan Plateau glaciers. Earth Syst. Sci. Data 14, 2303–2314 (2022).
Qi, J. et al. Effect of Indian monsoon on the glacial airborne bacteria over the Tibetan Plateau. Sci. Total Environ. 831, 154980 (2022).
Mao, G. et al. Monsoon affects the distribution of antibiotic resistome in Tibetan glaciers. Environ. Pollut. 317, 120809 (2023).
Liu, Y. et al. A genome and gene catalog of glacier microbiomes. Nat. Biotechnol. 40, 1341–1348 (2022).
Hu, H. et al. Global abundance patterns, diversity, and ecology of Patescibacteria in wastewater treatment plants. Microbiome 12, 55 (2024).
Fernandes, A. T. et al. The multicopper oxidase from the archaeon Pyrobaculum aerophilum shows nitrous oxide reductase activity. FEBS J. 277, 3176–3189 (2010).
Zhang, L. et al. Unexpectedly minor nitrous oxide emissions from fluvial networks draining permafrost catchments of the East Qinghai-Tibet Plateau. Nat. Commun. 13, 950 (2022).
Farías, L. et al. Biological N2O fixation in the eastern South Pacific Ocean and marine cyanobacterial cultures. PLoS One 8, e63956 (2013).
Cornejo, M., Murillo, A. A. & Farías, L. An unaccounted for N2O sink in the surface water of the eastern subtropical South Pacific: physical versus biological mechanisms. Prog. Oceanogr. 137, 12–23 (2015).
Si, Y. et al. Direct biological fixation provides a freshwater sink for N2O. Nat. Commun. 14, 6775 (2023).
Mosley, O. E. et al. Nitrogen cycling and microbial cooperation in the terrestrial subsurface. ISME J. 16, 2561–2573 (2022).
Blagodatskaya, E. & Kuzyakov, Y. Active microorganisms in soil: critical review of estimation criteria and approaches. Soil Biol. Biochem. 67, 192–211 (2013).
Hattori, S., Li, Z., Yoshida, N. & Takeuchi, N. Isotopic evidence for microbial nitrogen cycling in a glacier interior of high-mountain Asia. Environ. Sci. Technol. 57, 15026–15036 (2023).
Wynn, P. M., Hodson, A. J., Heaton, T. H. E. & Chenery, S. R. Nitrate production beneath a High Arctic Glacier, Svalbard. Chem. Geol. 244, 88–102 (2007).
Wang, F. et al. High proportional nitrification-derived nitrate in glacier runoffs on the Tibetan Plateau indicated by triple oxygen isotopes of nitrate. J. Hydrol. 614, 128632 (2022).
Brankatschk, R., Töwe, S., Kleineidam, K., Schloter, M. & Zeyer, J. Abundances and potential activities of nitrogen cycling microbial communities along a chronosequence of a glacier forefield. ISME J. 5, 1025–1037 (2011).
Leamy, K. A., Yennawar, N. H. & Bevilacqua, P. C. Molecular mechanism for folding cooperativity of functional RNAs in living organisms. Biochemistry 57, 2994–3002 (2018).
Hocher, A. et al. Growth temperature and chromatinization in archaea. Nat. Microbiol. 7, 1932–1942 (2022).
Brown, J. H., Gillooly, J. F., Allen, A. P., Savage, V. M. & West, G. B. Toward a metabolic theory of ecology. Ecology 85, 1771–1789 (2004).
Ohashi, Y. et al. Regulation of nitrate assimilation in cyanobacteria. J. Exp. Bot. 62, 1411–1424 (2011).
Esteves-Ferreira, A. A., Inaba, M., Fort, A., Araújo, W. L. & Sulpice, R. Nitrogen metabolism in cyanobacteria: metabolic and molecular control, growth consequences and biotechnological applications. Crit. Rev. Microbiol. 44, 541–560 (2018).
Sparacino-Watkins, C., Stolz, J. F. & Basu, P. Nitrate and periplasmic nitrate reductases. Chem. Soc. Rev. 43, 676–706 (2014).
Brown, G. H., Tranter, M., Sharp, M. J., Davies, T. D. & Tsiouris, S. Dissolved oxygen variations in alpine glacial meltwaters. Earth Surf. Process. Landf. 19, 247–253 (1994).
Yao, T. et al. Different glacier status with atmospheric circulations in Tibetan Plateau and surroundings. Nat. Clim. Change 2, 663–667 (2012).
Liu, Y., Xu, R., Wang, Y., Pan, Y. & Piao, S. Wet deposition of atmospheric inorganic nitrogen at five remote sites in the Tibetan Plateau. Atmos. Chem. Phys. 15, 11683–11700 (2015).
Zhou, J. et al. Temperature mediates continental-scale diversity of microbes in forest soils. Nat. Commun. 7, 12083 (2016).
Antão, L. H. et al. Temperature-related biodiversity change across temperate marine and terrestrial systems. Nat. Ecol. Evol. 4, 927–933 (2020).
Chen, K.-H. et al. Metabolic coupling between soil aerobic methanotrophs and denitrifiers in rice paddy fields. Nat. Commun. 15, 3471 (2024).
Wang, B. et al. Temperature thresholds drive the biogeographic pattern of root endophytic fungal diversity in the Qinghai-Tibet Plateau. Sci. Total Environ. 889, 164270 (2023).
Shcherbak, I., Millar, N. & Robertson, G. P. Global metaanalysis of the nonlinear response of soil nitrous oxide (N2O) emissions to fertilizer nitrogen. Proc. Natl. Acad. Sci. USA 111, 9199–9204 (2014).
Harter, J. et al. Linking N2O emissions from biochar-amended soil to the structure and function of the N-cycling microbial community. ISME J. 8, 660–674 (2014).
Ravishankara, A. R., Daniel, J. S. & Portmann, R. W. Nitrous oxide (N2O): the dominant ozone-depleting substance emitted in the 21st century. Science 326, 123–125 (2009).
Zhao, C. et al. Increased global warming potential during freeze-thaw cycle is primarily due to the contribution of N2O rather than CO2. Sci. Total Environ. 954, 176232 (2024).
Lamarche-Gagnon, G. et al. Greenland melt drives continuous export of methane from the ice-sheet bed. Nature 565, 73–77 (2019).
Wang, W. et al. Characteristics of atmospheric reactive nitrogen deposition in Nyingchi city. Sci. Rep. 9, 4645 (2019).
Anesio, A. M., Lutz, S., Chrismas, N. A. M. & Benning, L. G. The microbiome of glaciers and ice sheets. NPJ Biofilms Microbiomes 3, 10 (2017).
Liang, C. & Balser, T. C. Warming and nitrogen deposition lessen microbial residue contribution to soil carbon pool. Nat. Commun. 3, 1222 (2012).
Knorr, M. A. et al. Unexpected sustained soil carbon flux in response to simultaneous warming and nitrogen enrichment compared with single factors alone. Nat. Ecol. Evol. 8, 2277–2285 (2024).
Hersbach, H. et al. The ERA5 global reanalysis. Q. J. R. Meteorol. Soc. 146, 1999–2049 (2020).
Hijmans, R. J. et al. Raster: geographic data analysis and modeling. https://CRAN.R-project.org/package=terra (2015).
Zhang, Z., Liu, Y., Zhao, W. & Ji, M. Radiation impacts gene redundancy and biofilm regulation of cryoconite microbiomes in Northern Hemisphere glaciers. Microbiome 11, 228 (2023).
Krueger F., James F., Ewels P., Afyounian E. & Schuster-Boeckler B. FelixKrueger/TrimGalore: v0.6.7. https://doi.org/10.5281/zenodo.5127899 (2021).
Kopylova, E., Noé, L. & Touzet, H. SortMeRNA: fast and accurate filtering of ribosomal RNAs in metatranscriptomic data. Bioinformatics 28, 3211–3217 (2012).
Hyatt, D. et al. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinf 11, 119 (2010).
Olm, M. R., Brown, C. T., Brooks, B. & Banfield, J. F. dRep: a tool for fast and accurate genomic comparisons that enables improved genome recovery from metagenomes through de-replication. ISME J. 11, 2864–2868 (2017).
Parks, D. H. et al. A standardized bacterial taxonomy based on genome phylogeny substantially revises the tree of life. Nat. Biotechnol. 36, 996–1004 (2018).
Chaumeil, P.-A., Mussig, A. J., Hugenholtz, P. & Parks, D. H. GTDB-Tk: a toolkit to classify genomes with the Genome Taxonomy Database. Bioinformatics 36, 1925–1927 (2020).
Parks, D. H., Imelfort, M., Skennerton, C. T., Hugenholtz, P. & Tyson, G. W. CheckM: assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Res 25, 1043–1055 (2015).
Letunic, I. B. P. Interactive tree of life (iTOL) v3: an online tool for the display and annotation of phylogenetic and other trees. Nucleic Acids Res 44, W242–W245 (2016).
Vieira-Silva, S. & Rocha, E. P. The systemic imprint of growth and its uses in ecological (meta)genomics. PLoS Genet 6, e1000808 (2010).
Tu, Q., Lin, L., Cheng, L., Deng, Y. & He, Z. NCycDB: a curated integrative database for fast and accurate metagenomic profiling of nitrogen cycling genes. Bioinformatics 35, 1040–1048 (2019).
Han, P. et al. Unveiling unique microbial nitrogen cycling and nitrification driver in coastal Antarctica. Nat. Commun. 15, 3143 (2024).
Li, W. & Godzik, A. CD-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics 22, 1658–1659 (2006).
Patro, R., Duggal, G., Love, M. I., Irizarry, R. A. & Kingsford, C. Salmon provides fast and bias-aware quantification of transcript expression. Nat. Methods 14, 417–419 (2017).
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).
Li, H. et al. The sequence alignment/map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).
Seyler, L. M., Trembath-Reichert, E., Tully, B. J. & Huber, J. A. Time-series transcriptomics from cold, oxic subseafloor crustal fluids reveals a motile, mixotrophic microbial community. ISME J. 15, 1192–1206 (2021).
Giordano, N. et al. Genome-scale community modelling reveals conserved metabolic cross-feedings in epipelagic bacterioplankton communities. Nat. Commun. 15, 2721 (2024).
Oksanen J. et al. vegan: community ecology package. https://CRAN.R-project.org/package=vegan (2015).
Rohart, F., Gautier, B., Singh, A. & KA, L. C. mixOmics: an R package for ‘omics feature selection and multiple data integration. PLoS Comput. Biol. 13, e1005752 (2017).
Sloan, W. et al. Quantifying the roles of immigration and chance in shaping prokaryote community structure. Environ. Microbiol. 8, 732–740 (2006).
Kembel, S. et al. Picante: R tools for integrating phylogenies and ecology. Bioinformatics 26, 1463–1464 (2010).
Ning, D., Deng, Y., Tiedje, J. M. & Zhou, J. A general framework for quantitatively assessing ecological stochasticity. Proc. Natl. Acad. Sci. USA 116, 16892–16898 (2019).
Bates, D., Mächler, M., Bolker, B. & Walker, S. Fitting linear mixed-effects models using lme4. J. Stat. Softw. 67, 1–48 (2015).
Aphalo P., Slowikowski K. ggpmisc: miscellaneous Extensions to “ggplot2”. https://CRAN.R-project.org/package=ggpmisc (2021).
Liu, S. et al. The contemporary glaciers in China based on the Second Chinese Glacier Inventory. Acta Geographica Sin. 70, 3–16 (2015).
Kassambara, A. ggpubr: “ggplot2” based publication ready plots. https://rpkgs.datanovia.com/ggpubr/ (2020).
Acknowledgements
This study was financially supported by the following funding: Natural Science Foundation of China (Grant No. 42421001, 42330410 and U21A20176 from Y.Q.L.). This study also supported by the Global Ocean Negative Carbon Emissions (Global ONCE) Program. We thank the Second Tibetan Plateau Scientific Expedition and Research Program (STEP) for conducting sampling on the Tibetan glaciers. For this study, sampling permissions were obtained from STEP, which facilitated access to the sampling location. We thank Prof. Jianjun Wang and Prof. Peng Xing for helpful discussions and proofreading of the manuscript. Many thanks to Dr. Jing Qi. We also acknowledge the Supercomputing Center of Lanzhou University for computing resources.
Author information
Authors and Affiliations
Contributions
Y.Q.L. and M.K.J. designed the study, and Z.H.Z. analyzed data, and wrote the draft of the manuscript. K.S.L. and Y.Y.C. analyzed data and revised the word. F.W. and G.N.M. revised the word. W.S.Z. revised the word, discussed the results and edited the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks Zhongjun Jia and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Zhang, Z., Liu, Y., Zhao, W. et al. Distinct genes and microbial communities involved in nitrogen cycling between monsoon- and westerlies-dominated Tibetan glaciers. Nat Commun 16, 5926 (2025). https://doi.org/10.1038/s41467-025-61002-x
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41467-025-61002-x
