Philippa Kaur

Effects of seasonality on physiochemical properties of waterThe characteristics of the physical and chemical factors of the river are shown in Table 1. Variance analysis showed that the environmental conditions during the four seasons were significantly different (P  More

Microbial community samplingHippo gut microbiomeWe characterized the microbial communities in the hippo gut by collecting ten samples of fresh hippo feces adjacent to four hippo pools early in the morning (prior to desiccation by the sun) in September 2017. We collected feces from different pools and locations adjacent to the pool to include the feces of different individuals so we could estimate the similarity of the gut microbiome among individuals and across the landscape. The four hippo pools are sufficiently far apart that there was likely no intermixing of hippos among them.Each individual hippo feces sample was gently homogenized by hand and then the liquid was gently squeezed from the coarse particulate organic matter. A portion of the liquid (approximately 10 mL) was vacuum filtered through a Supor polysulfone membrane (0.2-µm pore size; Pall, Port Washington, NY, USA). After approximately 10 mL had filtered through and the filter was dry, 15 mL of RNALater was gently poured onto the filter and allowed to contact the collected biomass on the filter for 15 min before being removed by filtration. The filter was stored dry in a sterile petri dish and transferred to a refrigerator within several hours, then to a − 20 °C freezer for storage within several days.During the July 2016 survey of hippo pools, we collected an additional two samples of fresh hippo feces near a high-subsidy hippo pool and filtered approximately 10 mL of the liquid portion after homogenization as detailed above. The filter was then folded twice to preserve the biomass on the filter and stored in 14 mL of RNALater.Aquatic ecosystemWe characterized the microbial communities in the water column of hippo pools across a gradient of hippo subsidy (July 2016, N = 12 pools). We collected samples from the upstream, downstream, surface, and bottom of both pools containing hippos and pools that lacked hippos. Subsamples were also analyzed for biogeochemical variables (details provided below). We also collected water samples in four of the high-subsidy hippo pools every 2–3 days starting immediately after a flushing event until the next flushing event (August and September 2017, Supplementary Fig. S1)23. The number of hippos, discharge and volume for each pool are presented in Dutton et al (2020)23.We sampled the aquatic microbial community and biogeochemical variables along a longitudinal transect down both the Mara and Talek rivers (Supplementary Fig. S1, Supplementary Table S2). For the Mara River, we sampled an approximately 100-km transect along a gradient of hippo numbers (N = 10 locations, from 0 to ~ 4000 hippos). For the Talek River, we sampled an approximately 30-km transect to the confluence with the Mara (N = 8 locations, from 0 to 700 hippos). Mara River sites 9 and 10 are downstream of the confluence with the Talek River. Water samples were collected from each site in a well-mixed flowing section away from any hippo pools.Aquatic microbial samples were collected by filtering water samples through a Supor polysulfone filter (0.2-µm pore size; Pall, Port Washington, NY, USA) and then preserving the filter in RNALater Stabilization Solution (Ambion, Inc., Austin, TX, USA). In 2017, the filters were preserved with RNA Later and then frozen for analysis.Mesocosm experimentWe collected river water from the Mara River upstream of the distribution of hippos and placed it in 45 1-L bottles in a large water basin covered by a dark tarp to help regulate temperature and prevent algal production. Bottles were randomly assigned to the control, bacteria, and bacteria + virus treatments. We collected fresh hippo feces from multiple locations adjacent to the Mara River. After homogenization, half of the hippo feces was sterilized in a pressure cooker, which testing confirmed had similar sterilization results as an autoclave53 (see Supplementary Materials). Five grams of sterilized hippo feces was placed into each bottle to provide an organic matter substrate without viable bacteria or viruses. The unsterilized hippo feces was expressed, and the resulting liquid was filtered through 0.7-µm GF/F filters (0.7-µm pore size; Whatman, GE Healthcare Life Sciences, Pittsburgh, PA, USA) and 0.2-µm Supor filters to physically separate the bacteria (on the filter papers) from the viruses (in the filtrate). Half the filtrate was then sterilized with a UV light treatment (Supplementary Fig. S4). The UV light treatment did not significantly alter DOC quality (see Supplementary Materials).We prepared 15 bottles for each of three treatments—control, bacteria, and bacteria + virus—as follows: Control Unfiltered river water, 5 g wet weight sterilized hippo feces, and two blank Supor filters; Bacteria Unfiltered river water, 5 g wet weight sterilized hippo feces, two Supor filters containing bacteria, and 4 mL sterilized filtrate; Virus Unfiltered river water, 5 g wet weight sterilized hippo feces, two Supor filters containing bacteria, and 4 mL unsterilized filtrate containing viruses.We conducted the experiment for 27 days from September to October 2017. We terminated the experiment after 27 days because we were trying to replicate the microbial communities in hippo pools as best as we could and the hippo pools rarely go more than 1 month before they are flushed out by a flood25. Initial microbial samples of the river water, hippo feces bacteria and hippo fecal liquid filtrate were taken on day 0, and three replicate samples per treatment were destructively sampled on day 3, 9, 15, 21, and 27. During each time step, the microbial communities were sampled using the methods detailed above, and chemical analyses were done on the water samples as described below. We also measured chlorophyll a, dissolved oxygen, temperature, conductivity, total dissolved solids, turbidity, and pH with a Manta2 water quality sonde (Eureka Water Probes, Austin, TX, USA).Microbial community characterizationWe used 16S rRNA sequencing to characterize the active microbial communities. We extracted both DNA and RNA from our preserved samples, then used RNA to synthesize cDNA to represent the “active” microbial community and the total DNA in the sample to represent the “total” microbes present, including those that may not be actively replicating54. Due to the continual loading of hippo feces into pools and the long half-life of DNA, we would expect there to be significant quantities of microbial DNA derived from hippo feces within the pools. However, there would be less accumulation of RNA because of RNA’s shorter half-life. The active communities identified through this RNA-based approach are the ones that would potentially contribute to ecosystem function55 as indicated by the protein synthesis potential, although relationships between activity and rRNA concentrations in individual taxa within mixed communities can vary56. Nevertheless, this method provides an overall characterization of the microbial community’s potential activity.We used the Qiagen RNeasy Powerwater Kit (Qiagen, Hilden, Germany) to extract the DNA and RNA from the material on the filter using a slightly modified manufacturer’s protocol to allow for the extraction of both DNA and RNA. After extraction, we split the total extracted volume (100 µL per sample) into two groups. We treated one group with the DNase Max Kit (Qiagen, Hilden, Germany) to remove all DNA and serve as the RNA group of samples.We used the RNA group of samples to synthesize cDNA using the SuperScript III First Strand Synthesis Kit (Invitrogen, Carlsbad, CA, USA). DNA and cDNA were quantified using the PicoGreen dsDNA Assay Kit (Molecular Probes, Eugene, OR, USA) then normalized to 5 ng/µL. Amplicon library preparation was done using a dual-index paired-end approach57. We amplified the V4 region of the 16S rRNA gene using dual-index primers (F515/R805) and AccuPrime Pfx SuperMix (Invitrogen, Carlsbad, CA, USA) in duplicate for each sample using the manufacturer’s recommended thermocycling routine.Samples were then pooled, purified and normalized using the SequelPrep Normalization Plate Kit (Invitrogen, Carlsbad, CA, USA). Barcoded amplicon libraries were then sequenced at the Yale Center for Genome Analysis (New Haven, CT, USA) using an Illumina Miseq v2 reagent kit (Illumina, San Diego, CA, USA) to generate 2 × 250 base pair paired-end reads.Sampling took place in 2016 and 2017 and involved two separate sequencing runs. The first sequencing run included negative controls and a mock community (D6306, Zymo Research, Irvin, CA, USA). The second sequencing run included negative controls, a mock community (D6306), and a single E. coli strain. In both runs, the mock community and single E. coli strain were well reconstructed from the sequences, and there was minimal contamination in the negative controls, mock community and E. coli strain.From those two sequencing campaigns, we received over 2 million raw sequences from the first campaign and over 7 million raw sequences for the second campaign. For the microbial community analyses, only samples collected and sequenced during the same campaign are analyzed together to prevent preservation or sequencing biases. However, samples within the two separate campaigns were preserved and sequenced using identical methods with only a minor modification (mentioned above) to increase the preservation of genetic material.We de-multiplexed sequenced reads then removed barcodes, indexes, and primers using QIIME258. We used DADA2 with a standard workflow in R59 to infer exact sequence variants (ESV) for each sample60. We assigned taxonomy using a naïve Bayesian classifier and the SILVA training set v. 128 database61,62. We removed potential contamination in samples from both campaigns by using the statistical technique in the R package, decontam63. We used Phyloseq to characterize, ordinate, and compare microbial communities64 with their standard workflow59.Chemical analysesAll water samples collected in the field and in the experiment were analyzed for dissolved ferrous iron (Fe(II)), hydrogen sulfide (H2S), dissolved organic carbon (DOC), inorganic nutrients, major ions, dissolved gases, and biochemical oxygen demand following the standard methods provided in detail in Dutton et al (2020)23.Statistical analysesWe computed all statistical analyses in the R 4.1.1 statistical language in RStudio 2021.09.0 using α = 0.05 to determine significance65,66. Error bars in the figures represent standard deviation of the means. All data and R code for the statistics and data treatments are provided in the Mendeley Data Online Repository67.We used the Bray–Curtis dissimilarity matrix followed by ordination with NMDS to examine differences between individual hippo gut microbiomes; between low-, medium-, and high-subsidy hippo pools; and between a gradient of hippo pools and the environment. We used a CCA to test for the influence of biogeochemical drivers on microbial community composition using biogeochemical data that were previously published but collected concurrently with this study23. We constrained the CCA ordination by soluble reactive phosphorus, nitrate, methane, BOD, and sulfate, which were all previously shown to be important drivers in the variation between pools23. We used PERMANOVA and PERMDISP to test for significant differences between groups68.
We compared aquatic microbial communities from the bottom of high-subsidy hippo pools (N = 15), from hippo feces (N = 10, the hippo gut microbiome) and upstream of high-subsidy hippo pools (N = 15, free of hippo gut microbiome influence) using the Bray–Curtis dissimilarity matrix on the relative abundances for the active aquatic microbial communities collected from the different sample types followed by ordination with NMDS. 95% confidence ellipses were generated. We then determined the active taxa that were shared between the hippo gut microbiome (hippo feces) and the bottom of the high-subsidy hippo pools and not present in the upstream samples from high-subsidy hippo pools.We used LEfSe to calculate the differential abundance of microbial taxa between upstream (N = 14), downstream (N = 16), at the surface (N = 17) and at the bottom (N = 14) of hippo pools and calculated their effect size69. We then calculated the correlation of microbial taxa to the measured biogeochemistry using Pearson’s correlation coefficient with a false discovery rate corrected p-value in the microeco R package70.
We used SourceTracker to quantify the contribution of the hippo gut, upstream waters, or unknown sources to the active aquatic microbial communities in the bottom waters of three of the high-subsidy hippo pools between flushing flows71. We also used the Bray–Curtis dissimilarity matrix followed by ordination with NMDS to examine changes in the active aquatic microbial communities in one of the high subsidy hippo pools through time after flushing flows.For the experiment, we calculated the Bray–Curtis dissimilatory matrix followed by ordination with NMDS for the active aquatic microbial communities over time in each of the three experimental treatments. We used SourceTracker to determine the proportion of the active aquatic microbial community in each treatment that originated from the hippo gut, the river water, or unknown sources71. We analyzed the biogeochemical differences among experimental treatments by fitting a linear mixed effects model for each of the biogeochemical variables throughout the experiment with the nlme package in R72. We fit the model with the restricted maximum likelihood method and a continuous autoregressive temporal correlation structure with sample day as the repeated factor. Treatment and time were fixed effects and individual bottles were treated as random effects. We conducted a pairwise post-hoc test with an ANOVA and the emmeans package in R to estimate marginal means with a Tukey adjusted p-value for multiple comparisons73,74. More

MPs abundanceIn Table 1 MP abundance (mean value ± standard deviation) values are presented by shape, size range, color and polymer type categories for each sampling site. MP were found in all analyzed salt samples including pellets, fibers, fragments, films and lines (Fig. 3). MP total abundance values per site ranged from 74.7 to 136.7 particles kg−1 in the following order of increasing abundance: S3  black (17%)  > blue (15%)  > green and transparent (10% each)  > pink (6%)  > colorless (5%). In terms of size, most particles were in the category 500–1000 µm, except for S3 (1000–5000 µm) (Table 1). The distribution of MP particles based on size range was: 500–1000 µm (40%)  > 1000–5000 µm (34%)  > 250–500 µm (26%). For salts from the Atlantic and the Pacific Ocean, originating from Brazil, the United Kingdom, and the USA, Kim et al.12 reported a higher abundance of MP in size range 100–1000 µm while sizes in the range 100–5000 µm were reported for salt samples from the Black Sea. Seth and Shriwastay20 found that 80% of fibers found in salt samples from the Indian Sea were smaller than 2000 μm in length. MP size range differences among the various studies are suggested to depend on the degree of weathering for a given environment30, different climatic conditions such as wind, rain, temperature, salinity, and waves influencing size range composition. Also, for runoff, rivers, and atmospheric fallout transportation, smaller MP size ranges can be expected to be associated with a longer range from the initial plastic sources31,32,33. Nevertheless, more detailed information about MP polymer/color features within the size ranges are needed to achieve stronger conclusions about potential long/short-range sources.Figure 6Microplastics abundance (particles kg−1) by color in sea salt samples from stations S1 to S8.Full size imageFigure 7Microplastics abundance (particles kg−1) by size range in sea salt samples from stations S1 to S8.Full size imageMP polymer compositionFour types of polymer, namely polypropylene (PP), polystyrene (PS), polyethylene (PE), and polyethylene terephthalate (PET), were identified with FT-MIR-NIR (Supplementary Figure S1). These results are in accordance with those reported for salt samples in other studies worldwide (Table 1). These polymer types are widely used in daily life products, packaging, single-use plastics, and clothes, contributing to plastic pollution worldwide21. PET presented the highest contribution at all sampling sites, at ~ 48%, whereas PS was found to be least, at ~ 15% (Fig. 8, Table 1). Iñiguez et al.34 also reported PET predominance (83.3%) in Spanish table salt samples. PET predominance could be explained by its high density (1.30 g cm−3), making particles prone to sedimentation during the salt crystallization process19. PE (0.94 g cm−3), PP (0.90 g cm−3), and PS (1.05 cm−3) presented lower or similar densities to seawater (~ 1.02 g cm−3), making these more prone to flotation and possible loss due to wind during desiccation.Figure 8Microplastics abundances (particles kg−1) by polymer composition in sea salt samples from stations S1 to S8.Full size imageRisks assessmentDuring degradation, MP tends to emit monomers and different types of additives, these having the potential to cause harm to ecological systems and health18, 35. Results for the polymeric risks indices are presented in Fig. 9. According to polymer risk classification, all salts samples showed low risks, being similar to the entire study area. To date, none of the published studies have applied chemometric models in evaluating MP pollution in salts, posing difficulties when comparing our results. Information on the hazards of MP from ingestion to human health is still highly unclear. Other than exposure, the destiny and transit of ingested MP in the human body, including intestinal digestion and biliary discharge, have not been determined in previous research and remained largely unclear36. Some studies conducted impact assessments based on in vitro models37,38. However, whether the exposure concentrations used in such studies indicate the MP consumed and collected in humans is inconclusive. Previous studies found that toxicity, oxidative stress, and inflammation could result from MP exposure, including immune disruption and neurotoxicity effects, among others39. Therefore, an immediate effort is required to assess the health consequences of these MP when they reach the human body.Figure 9Polymeric risk indices for MP types in salts from stations S1 to S8.Full size image More

AffiliationsDepartment of Ocean Sciences, University of California, Santa Cruz, CA, USARachel A. FosterDepartment of Biogeochemistry, Max Planck Institute for Marine Microbiology, Bremen, GermanyRachel A. Foster, Daniela Tienken, Sten Littmann & Marcel M. M. KuypersDepartment of Ecology, Environment and Plant Sciences, Stockholm University, Stockholm, SwedenRachel A. FosterDepartment of Geosciences, Swedish Museum of Natural History, Stockholm, SwedenMartin J. WhitehouseDepartment of Oceanography, University of Hawai’i at Mānoa, Honolulu, HI, USAAngelicque E. WhiteAuthorsRachel A. FosterDaniela TienkenSten LittmannMartin J. WhitehouseMarcel M. M. KuypersAngelicque E. WhiteCorresponding authorCorrespondence to
Rachel A. Foster. More

Coral collection and laboratory acclimationCorals were obtained from the nursery at Mote Marine Laboratory’s Elizabeth Moore International Center for Coral Reef Research & Restoration. Acropora cervicornis fragments (apical tips ~ 5–6 cm in length) were obtained from Mote’s in situ nursery and mounted to a plastic base with marine epoxy (Instant Ocean Holdfast). The A. cervicornis genotypes used in this study were Mote Marine Laboratory genotype numbers 7, 31, 50, 57, and 70 (n = 12 ramets from each of the 5 genets). The coral O. faveolata was obtained from Mote’s ex situ nursery as fragments (~ 3–4 cm2) attached to individual carbonate plugs. Six genotypes of O. faveolata were used in deoxygenation treatments from genotype numbers F3A, F8, F27, F61, F125, and F132 (n = 12 ramets from each of the 6 genets). The immediate environmental conditions at the ex situ coral nursery are not known in detail, however the corals were maintained in an open, flow-through system with high flow rates and continuous bubbling with air stones.Corals were transported to the wet lab facilities at the Smithsonian Marine Station (SMS) in Fort Pierce, FL and experiments were conducted between October and December 2019. Corals were acclimated to laboratory conditions for 4–6 weeks before exposure to deoxygenation treatments. During this period, corals were kept in an indoor, closed seawater system. Holding tanks contained ~ 570 L of seawater that was recirculated through a sump containing rigorous aeration (ambient air) and a heater/chiller that maintained the temperature at ~ 27 ºC (see Table S3 for all parameters). Water was pumped from the sump through a UV sterilizer (Coralife) and then back into the holding tank. Water flow and circulation was maintained by two aquarium pumps (AquaTop MaxFlow MCP-5), and yielded a full water exchange through the sump ~ 8.5 times per hour. Light was provided by LED aquarium lights (HQD) with a maximum photosynthetically active radiation (PAR) of ~ 300 µmol photon m−2 s−1. Full water changes were conducted weekly, and non-living surfaces on the coral fragments (e.g., plastic or carbonate bases) were cleaned at least once a week to reduce algal growth. Conditions within the acclimation tanks closely matched the ambient (i.e., normoxic) conditions at the collection locations and in the oxygen treatments (Table S3, Fig. S4).Field dissolved oxygen conditionsOxygen concentrations were monitored at Mote Marine Laboratory’s in situ nursery (24.56 latitude, − 81.40 longitude) at the depth of coral outplants (~ 5–6 m) to parameterize oxygen conditions used in laboratory experiments. A dissolved oxygen sensor (miniDot, PMEL) was calibrated according to manufacturer protocols and attached to the benthos adjacent to coral outplants. Dissolved oxygen and temperature were logged every ten minutes for the duration of deployment. Hourly and daily average DO concentrations are presented for one month encompassing the approximate time of coral retrievals (Fig. S4). The 6.25 mg L−1 treatment in the laboratory experiments simulates the average normoxic conditions at the site of collection (6.18 ± 0.10, SD), while the most severe deoxygenation treatment (1.00 mg L−1) simulated extreme oxygen depletion that has occurred during acute deoxygenation events on coral reefs21,22,52 (Fig. S4). The two intermediate deoxygenation treatments represented incremental increases by ~ 2 mg L−1 to capture coral responses over a full range of oxygen conditions (Fig. S4).Experimental designDeoxygenation experiments were conducted in the wet lab facilities at SMS using a mesocosm array consisting of 12 tanks (50 L, AquaLogic Systems), with each tank functioning as a closed system with fully independent temperature and oxygen control. Targeted oxygen levels for treatments were 1.00, 2.25, 4.25, and 6.25 mg L−1 DO, with three independent tank replicates for each oxygen level. Oxygen treatments are referred to by the targeted DO concentrations. Species were run in separate, sequential experiments, and genotypes within a species were co-mingled in treatment tanks with one replicate of each genotype per tank.Oxygen concentrations were maintained in each independent tank by bubbling seawater with nitrogen gas and ambient air, with gas injection controlled through a DO feedback and solenoid valves (Neptune Systems). Oxygen levels were monitored every 15 s in each tank by an OxyGuard DO probe connected to an aquarium controller (Neptune Systems, Apex Aquacontroller), which opened or closed the respective solenoid valves to add nitrogen or air in order to maintain the programmed treatment conditions. Oxygen probes were calibrated at the start of experiments, and then again after five days, following the manufacturer’s protocol.Temperature control was maintained in treatment aquaria through an independent heating/chilling loop in each tank and monitored by AquaLogic temperature probes (sensu53). The average temperature at the collection site in the month prior to collection was 29.01 ± 0.62 ℃ (SD) (Fig. S4), which represents seasonally warm temperatures of the Florida Keys54,55. To reduce the potential confounding effects of temperature on responses to deoxygenation, we acclimated corals to 27 ℃ during the holding period, and conducted experiments at ~ 27 ℃. This represents the average annual temperature at the collection site in the Florida Keys, is the temperature that O. faveolata fragments were maintained at in the ex situ nursery prior to collection (27.0 ± 0.62 ℃), and is commonly used as the ambient temperature for corals from this location54,56. Each tank contained an additional probe (Neptune Systems) that logged temperature every 10 s for the duration of the experiment. In addition to continuous monitoring of DO and temperature, discrete measurements were taken each day for DO, temperature, pH, and salinity with a handheld multi-parameter water quality meter (YSI ProDSS with optical DO sensor) (Tables 1, 2, Tables S1, S2). The YSI DO probe was calibrated at the start of every day following manufacturer protocols and used to validate DO measurements from OxyGuard probes and to adjust programmed values to maintain treatment targets when necessary.Individual tanks were supplied with a 7-color LED aquarium light (Aquaillumination, Hydra 52), programmed to simulate a diel light cycle over a 12:12 h photoperiod, and set to maximum irradiance of ~ 300 µmol photon m−2 s−1 (PAR). This maximum intensity is likely lower than what corals may experience in situ during peak midday irradiances, however, it matches the irradiance levels these corals were maintained at in the ex situ nursery (320 ± 182 SD, n = 45) and is sufficient to stimulate maximal photosynthesis without causing light stress45. Light levels were measured with a light meter (Licor, LI-1400) and an underwater spherical quantum sensor (LI-193SA) submerged at the center of each tank at the start of each experiment and again after five days. Oxygen, temperature, and light conditions were effectively maintained at or near targeted levels for the duration of each experiment (Fig. 1, Table 1), with minimal differences between replicate tanks within a treatment (Tables S1, S2). We did not simultaneously manipulate pH in treatment tanks, which resulted in differences in pH between treatments due to bubbling with nitrogen gas (Tables 1, 2).Coral conditionCorals were monitored daily during the experiments for changes in live tissue cover and photophysiology, and then destructively sampled at the end of the experiment for symbiont densities. Fragments were visually evaluated at each time point for evidence of bleaching, tissue loss, and mortality. Tissue loss was quantified as the percent of each fragment with exposed skeleton, resulting from the sloughing of tissue. Fragments were categorized as dead when all living tissue was lost.The A. cervicornis experiment ended after five days, at which point 50% of corals in the lowest DO treatment were dead or displayed partial mortality. We ended the O. faveolata experiment after 11 days, more than twice the duration of the A. cervicornis experiment (Fig. 1), at which point no observable signs of deterioration were detected in corals from any treatment (Fig. 2a).PAM fluorometryPAM fluorometry is a non-destructive method that directly measures chlorophyll fluorescence and the activity of photosystem II (PSII)57, and the quantum yield can be used as a proxy for coral stress36,58. For all PAM readings, one measurement was taken from the same position on each fragment with the probe held at a 90º angle ~ 0.5 cm from the coral surface (angle and distance were maintained with a Walz probe holder). To optimize initial fluorescence (F0) to between 300–500, the following PAM settings were used: gain = 2, damp = 2, saturation intensity = 8, saturation width = 0.8, and measuring light intensity = 10 for A. cervicornis and 6 for O. faveolata. PAM measurements were taken pre-dawn daily, at 0500–0600 h. The position on A. cervicornis fragments was shifted if tissue sloughing occurred so that measurements were taken on living tissue. After the final PAM measurements, corals were snap frozen in liquid nitrogen and stored at − 80 ºC for subsequent analyses.Symbiont densitySymbiont density analyses were conducted at the University of Florida in Gainesville, FL following standard protocols (sensu59). In brief, tissue was stripped from the coral skeleton with an air brush (Master Airbrush S68) and filtered seawater. The tissue slurry was then homogenized with a handheld electric tissue homogenizer (Tissue-Tearor) and subsampled for symbiont counts. Symbiont cells were counted with a hemocytometer, with six replicate counts per fragment. Replicate counts were averaged per fragment for all analyses.Surface area of A. cervicornis fragments was determined by wax dipping60 and the surface area of O. faveolata fragments was determined through image analysis in ImageJ. For A. cervicornis, surface area was then corrected by the percent tissue loss recorded for each fragment. Symbiont densities were normalized to fragment live tissue surface area and expressed as cells per cm−2.Statistical analysesAll analyses were conducted in R (v 4.0.2)61. The effects of fixed and random factors were evaluated with linear mixed effects models using the package lme462. Normality and homoscedasticity of variances of response variables were evaluated by visual inspection of residuals and Levene’s tests, respectively. All variables met model assumptions, and results are reported for full models that included all fixed and random effects.For maximum quantum yield and tissue loss, the factor corresponding to measurements repeated daily from the start to end of the experiment (i.e., Day 1, 2, 3, etc.) is referred to as “day”. Day, genotype, and treatment were treated as fixed factors in the analysis of maximum quantum yield and tissue loss, with individual included as a random factor to account for repeated measures, and tank as random factor to account for fragments of different genotypes within a tank (Tables 2, 3). For symbiont density, genotype and treatment were analyzed as fixed factors, and tank as a random factor (Table 4). The significance of fixed effects was evaluated with type II ANOVA tables using Satterthwaite’s method, and Tukey’s post-hoc tests were used where necessary to determine significant differences between levels of a factor using the package emmeans63. Data in main figures in which genotype was not a significant effect (i.e., all response variables) are presented as treatment averages (± SE), pooled across genotypes. More

1.Minter, N. J., Franks, N. R. & Brown, K. A. R. Morphogenesis of an extended phenotype: Four-dimensional ant nest architecture. J. R. Soc. Interface 9, 586–595 (2012).PubMed 

Google Scholar 
2.Dawkins, R. The Extended Phenotype: The Long Reach of the Gene (Oxford University Press, 2016).
Google Scholar 
3.Tschinkel, W. R. The architecture of subterranean ant nests: Beauty and mystery underfoot. J. Bioecon. 17, 271–291 (2015).
Google Scholar 
4.Brian, M. V. & Brian, M. V. Production Ecology of Ants and Termites (Cambridge University Press, 1978).
Google Scholar 
5.De Bruyn, L. A. L. & Conacher, A. J. The role of termites and ants in soil modification: A review. Soil Res. 28, 55–93 (1990).
Google Scholar 
6.Sankovitz, M. A. & Breed, M. D. Effects of Formica podzolica ant colonies on soil moisture, nitrogen, and plant communities near nests. Ecol. Entomol. 44, 71–80 (2019).
Google Scholar 
7.Tschinkel, W. R. Subterranean ant nests: Trace fossils past and future?. Palaeogeogr. Palaeoclimatol. Palaeoecol. 192, 321–333 (2003).
Google Scholar 
8.Pinter-Wollman, N. Nest architecture shapes the collective behaviour of harvester ants. Biol. Lett. 11, 20150695 (2015).PubMed 
PubMed Central 

Google Scholar 
9.Rosengren, R., Fortelius, W., Lindström, K. & Luther, A. Phenology and causation of nest heating and thermoregulation in red wood ants of the Formica rufa group studied in coniferous forest habitats in southern Finland. Ann. Zool. Fennici 24, 147–155 (1987).
Google Scholar 
10.Hölldobler, B. & Wilson, E. O. The Ants (Harvard University Press, 1990).
Google Scholar 
11.Savolainen, R. & Vepsäläinen, K. A competition hierarchy among boreal ants: Impact on resource partitioning and community structure. Oikos 51, 135–155 (1988).
Google Scholar 
12.Frouz, J., Jílková, V. & Sorvari, J. Contribution of wood ants to nutrient cycling and ecosystem function. Wood Ant Ecol. Conserv. https://doi.org/10.1017/CBO9781107261402.010 (2016).Article 

Google Scholar 
13.Seeley, T. & Heinrich, B. Regulation of Temperature in the Nests of Social Insects (FAO, 1981).
Google Scholar 
14.Hillel, D. Environmental Soil Physics: Fundamentals, Applications, and Environmental Considerations (Elsevier, 1998).
Google Scholar 
15.Blomqvist, M. M., Olff, H., Blaauw, M. B., Bongers, T. & Van Der Putten, W. H. Interactions between above- and belowground biota: Importance for small-scale vegetation mosaics in a grassland ecosystem. Oikos 90, 582–598 (2000).
Google Scholar 
16.MacMahon, J. A., Mull, J. F. & Crist, T. O. Harvester ants (Pogonomyrmex spp.): Their community and ecosystem influences. Annu. Rev. Ecol. Syst. 31, 265–291 (2000).
Google Scholar 
17.Jones, C. G., Lawton, J. H. & Shachak, M. Organisms as ecosystem engineers. In Ecosystem Management 130–147 (Springer, 1994).
Google Scholar 
18.Jouquet, P., Dauber, J., Lagerlöf, J., Lavelle, P. & Lepage, M. Soil invertebrates as ecosystem engineers: Intended and accidental effects on soil and feedback loops. Appl. Soil Ecol. 32, 153–164 (2006).
Google Scholar 
19.Khuong, A. et al. Stigmergic construction and topochemical information shape ant nest architecture. Proc. Natl. Acad. Sci. U. S. A. 113, 1303–1308 (2016).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
20.Bishop, T. R. et al. Thermoregulatory traits combine with range shifts to alter the future of montane ant assemblages. Glob. Change Biol. 25, 2162–2173 (2019).ADS 

Google Scholar 
21.Sala, O. E. et al. Global biodiversity scenarios for the year 2100. Science 287, 1770–1774 (2000).CAS 
PubMed 

Google Scholar 
22.Braschler, B. et al. Realised rather than fundamental thermal niches predict site occupancy: Implications for climate change forecasting. J. Anim. Ecol. 89, 2863–2875 (2020).PubMed 

Google Scholar 
23.Roeder, K. A., Bujan, J., Beurs, K. M., Weiser, M. D. & Kaspari, M. Thermal traits predict the winners and losers under climate change: An example from North American ant communities. Ecosphere 12, e03645 (2021).
Google Scholar 
24.Deutsch, C. A. et al. Impacts of climate warming on terrestrial ectotherms across latitude. Proc. Natl. Acad. Sci. U. S. A. 105, 6668–6672 (2008).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
25.Diamond, S. E., Sorger, D. M., Hulcr, J. & Pelini, S. L. Who likes it hot? A global analysis of the climatic, ecological, and evolutionary determinants of warming tolerance in ants. Glob. Change Biol. 18, 448–456 (2012).ADS 

Google Scholar 
26.Hoffmann, A. A., Chown, S. L. & Clusella-Trullas, S. Upper thermal limits in terrestrial ectotherms: How constrained are they?. Funct. Ecol. 27, 934–949 (2013).
Google Scholar 
27.Wilson, E. O. The effects of complex social life on evolution and biodiversity. Oikos 63, 13–18 (1992).CAS 

Google Scholar 
28.Huey, R. B. & Stevenson, R. D. Integrating thermal physiology and ecology of ectotherms: A discussion of approaches. Am. Zool. 19, 357–366 (1979).
Google Scholar 
29.Deslippe, R. J. & Savolainen, R. Colony foundation and polygyny in the ant Formica podzolica. Behav. Ecol. Sociobiol. 37, 1–6 (1995).
Google Scholar 
30.Schneider, C. A., Rasband, W. S. & Eliceiri, K. W. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675 (2012).CAS 
PubMed 
PubMed Central 

Google Scholar 
31.Chambers, J. M., Freeny, A. & Heiberger, R. M. Analysis of variance; designed experiments. Stat. Models S 5, 145–193 (1992).
Google Scholar 
32.Fox, J. Applied Regression Analysis and Generalized Linear Models (SAGE Publications, 2015).
Google Scholar 
33.Mikheyev, A. S. & Tschinkel, W. R. Nest architecture of the ant Formica pallidefulva: Structure, costs and rules of excavation. Insectes Soc. 51, 30–36 (2004).
Google Scholar 
34.Coppernoll-Houston, D. & Potter, C. Field measurements and satellite remote sensing of daily soil surface temperature variations in the lower Colorado desert of California. Climate 6, 94 (2018).
Google Scholar 
35.Jílková, V., Cajthaml, T. & Frouz, J. Respiration in wood ant (Formica aquilonia) nests as affected by altitudinal and seasonal changes in temperature. Soil Biol. Biochem. 86, 50–57 (2015).
Google Scholar 
36.Kadochová, Š, Frouz, J. & Roces, F. Sun basking in red wood ants Formica polyctena (Hymenoptera, Formicidae): Individual behaviour and temperature-dependent respiration rates. PLoS ONE 12, e0170570 (2017).PubMed 
PubMed Central 

Google Scholar 
37.Bollazzi, M., Kronenbitter, J. & Roces, F. Soil temperature, digging behaviour, and the adaptive value of nest depth in South American species of Acromyrmex leaf-cutting ants. Oecologia 158, 165–175 (2008).ADS 
PubMed 

Google Scholar 
38.Stockan, J. A. & Robinson, E. J. H. Wood Ant Ecology and Conservation (Cambridge University Press, 2016).
Google Scholar 
39.Porter, S. D. Impact of temperature on colony growth and developmental rates of the ant, Solenopsis invicta. J. Insect Physiol. 34, 1127–1133 (1988).
Google Scholar 
40.Lapointe, S. L., Serrano, M. S. & Jones, P. G. Microgeographic and vertical distribution of Acromynnex landolti (Hymenoptera: Formicidae) nests in a Neotropical Savanna. Environ. Entomol. 27, 636–641 (1998).
Google Scholar 
41.Fowler, H. G. Leaf-cuttings ants of the genera Atta and Acromyrmex of Paraguay (Hymenoptera: Formicidae). Mmitt. Mus. Naturkunde Berl. Dtsch. Entomol. Z. 32, 19–34 (2008).
Google Scholar 
42.Hansell, M. & Hansell, M. H. Animal Architecture (OUP, 2005).
Google Scholar 
43.Shik, J. Z., Arnan, X., Oms, C. S., Cerdá, X. & Boulay, R. Evidence for locally adaptive metabolic rates among ant populations along an elevational gradient. J. Anim. Ecol. 88, 1240–1249 (2019).PubMed 

Google Scholar 
44.Cerda, X., Retana, J. & Cros, S. Critical thermal limits in Mediterranean ant species: Trade-off between mortality risk and foraging performance. Funct. Ecol. 12, 45–55 (1998).
Google Scholar 
45.Kaspari, M., Clay, N. A., Lucas, J., Yanoviak, S. P. & Kay, A. Thermal adaptation generates a diversity of thermal limits in a rainforest ant community. Glob. Change Biol. 21, 1092–1102 (2015).ADS 

Google Scholar 
46.Talbot, M. Distribution of ant species in the Chicago region with reference to ecological factors and physiological toleration. Ecology 15, 416–439 (1934).
Google Scholar 
47.Arnan, X. & Blüthgen, N. Using ecophysiological traits to predict climatic and activity niches: Lethal temperature and water loss in Mediterranean ants: Using physiology to predict niches. Glob. Ecol. Biogeogr. 24, 1454–1464 (2015).
Google Scholar 
48.Arnan, X., Blüthgen, N., Molowny-Horas, R. & Retana, J. Thermal characterization of European ant communities along thermal gradients and its implications for community resilience to temperature variability. Front. Ecol. Evol. 3, 138 (2015).
Google Scholar 
49.Baudier, K. M. & O’Donnell, S. Structure and thermal biology of subterranean army ant bivouacs in tropical montane forests. Insectes Soc. 63, 467–476 (2016).
Google Scholar 
50.Penick, C. A. & Tschinkel, W. R. Thermoregulatory brood transport in the fire ant, Solenopsis invicta. Insectes Soc. 55, 176–182 (2008).
Google Scholar 
51.Penick, C. A., Diamond, S. E., Sanders, N. J. & Dunn, R. R. Beyond thermal limits: Comprehensive metrics of performance identify key axes of thermal adaptation in ants. Funct. Ecol. 31, 1091–1100 (2017).
Google Scholar 
52.Kearney, M., Shine, R. & Porter, W. P. The potential for behavioral thermoregulation to buffer ‘cold-blooded’ animals against climate warming. Proc. Natl. Acad. Sci. U. S. A. 106, 3835–3840 (2009).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
53.Ghalambor, C. K., Huey, R. B., Martin, P. R., Tewksbury, J. J. & Wang, G. Are mountain passes higher in the tropics? Janzen’s hypothesis revisited. Integr. Comp. Biol. 46, 5–17 (2006).PubMed 

Google Scholar  More

1.Limpricht KG. Die laubmoose. In: Rabenhorst L (ed). Kryptogamen-Flora von Deutschland, Oesterreich und der Schweiz, Zweite Auflage. 1890. Kummer, Leipzig.2.Basilier K. Fixation and uptake of nitrogen in Sphagnum blue-green algal associations. Oikos. 1980;34:239.CAS 

Google Scholar 
3.Granhall U, Selander H. Nitrogen fixation in a subarctic mire. Oikos. 1973;24:8.
Google Scholar 
4.Basilier K, Granhall U, Stenström T-A. Nitrogen fixation in wet minerotrophic moss communities of a subarctic mire. Oikos. 1978;31:236.CAS 

Google Scholar 
5.Basilier K. Moss-associated nitrogen fixation in some mire and coniferous forest environments around Uppsala, Sweden. Lindbergia. 1979;5:84–88.CAS 

Google Scholar 
6.Meeks JC. Physiological adaptations in nitrogen-fixing Nostoc–plant symbiotic associations. In: Pawlowski K (ed). Prokaryotic symbionts in plants. 2007. Springer, Berlin, Heidelberg, pp 181–205.7.Adams DG. Cyanobacteria in symbiosis with hornworts and liverworts. Cyanobacteria in symbiosis. 2002. Springer, Dordrecht, pp 117-35.8.Meeks JC, Elhai J. Regulation of cellular differentiation in filamentous cyanobacteria in free-living and plant-associated symbiotic growth states. Microbiol Mol Biol Rev. 2002;66:94–121.CAS 
PubMed 
PubMed Central 

Google Scholar 
9.Kostka JE, Weston DJ, Glass JB, Lilleskov EA, Shaw AJ, Turetsky MR. The Sphagnum microbiome: new insights from an ancient plant lineage. N Phytol. 2016;211:57–64.CAS 

Google Scholar 
10.Granhall U, Hofsten AV. Nitrogenase activity in relation to intracellular organisms in Sphagnum mosses. Physiol Plant. 1976;36:88–94.
Google Scholar 
11.van den Elzen E, Kox MAR, Harpenslager SF, Hensgens G, Fritz C, Jetten MSM, et al. Symbiosis revisited: phosphorus and acid buffering stimulate N2 fixation but not Sphagnum growth. Biogeosciences. 2017;14:1111–22.
Google Scholar 
12.Yu Z, Loisel J, Brosseau DP, Beilman DW, Hunt SJ. Global peatland dynamics since the last glacial maximum. Geophys Res Lett. 2010;37:1–5.
Google Scholar 
13.Lindo Z, Nilsson MC, Gundale MJ. Bryophyte-cyanobacteria associations as regulators of the northern latitude carbon balance in response to global change. Glob Chang Biol. 2013;19:2022–35.PubMed 

Google Scholar 
14.Carrell AA, Kolton M, Glass JB, Pelletier DA, Warren MJ, Kostka JE, et al. Experimental warming alters the community composition, diversity, and N2 fixation activity of peat moss (Sphagnum fallax) microbiomes. Glob Chang Biol. 2019;25:2993–3004.PubMed 
PubMed Central 

Google Scholar 
15.Rai AN, Söderbäck E, Bergman B. Tansley Review No. 116. N Phytol. 2000;147:449–81.CAS 

Google Scholar 
16.Adams DG, Duggan PS. Cyanobacteria-bryophyte symbioses. J Exp Bot. 2008;59:1047–58.CAS 
PubMed 

Google Scholar 
17.Bay G, Nahar N, Oubre M, Whitehouse MJ, Wardle DA, Zackrisson O, et al. Boreal feather mosses secrete chemical signals to gain nitrogen. N Phytol. 2013;200:54–60.CAS 

Google Scholar 
18.Warshan D, Espinoza JL, Stuart RK, Richter RA, Kim S-Y, Shapiro N, et al. Feathermoss and epiphytic Nostoc cooperate differently: expanding the spectrum of plant–cyanobacteria symbiosis. ISME J. 2017;12:1–13.
Google Scholar 
19.Stuart RK, Pederson ERA, Weyman PD, Weber PK, Rassmussen U, Dupont CL. Bidirectional C and N transfer and a potential role for sulfur in an epiphytic diazotrophic mutualism. ISME J. 2020;14:3068–78.CAS 
PubMed 
PubMed Central 

Google Scholar 
20.Veličković D, Chu RK, Carrell AA, Thomas M, Paša-Tolić L, Weston DJ, et al. Multimodal MSI in conjunction with broad coverage spatially resolved MS2 increases confidence in both molecular identification and localization. Anal Chem. 2018;90:702–7.PubMed 

Google Scholar 
21.Nagy G, Veličković D, Chu RK, Carrell AA, Weston DJ, Ibrahim YM, et al. Towards resolving the spatial metabolome with unambiguous molecular annotations in complex biological systems by coupling mass spectrometry imaging with structures for lossless ion manipulations. Chem Commun. 2019;55:306–9.CAS 

Google Scholar 
22.Warshan D, Liaimer A, Pederson E, Kim S-Y, Shapiro N, Woyke T, et al. Genomic changes associated with the evolutionary transitions of Nostoc to a plant symbiont. Mol Biol Evol. 2018;35:1160–75.CAS 
PubMed 
PubMed Central 

Google Scholar 
23.Rippka R, Deruelles J, Waterbury JB, Herdman M, Stanier RY. Generic assignments, strain histories and properties of pure cultures of cyanobacteria. J Gen Microbiol. 1979;111:1–61.
Google Scholar 
24.Hanson PJ, Riggs JS, Robert Nettles W, Phillips JR, Krassovski MB, Hook LA, et al. Attaining whole-ecosystem warming using air and deep-soil heating methods with an elevated CO2 atmosphere. Biogeosciences. 2017;14:861–83.CAS 

Google Scholar 
25.Frank W, Decker EL, Reski R. Molecular tools to study Physcomitrella patens. Plant Biol. 2005;7:220–7.CAS 
PubMed 

Google Scholar 
26.Yao Y, Sun T, Wang T, Ruebel O, Northen T, Bowen BP. Analysis of metabolomics datasets with high-performance computing and metabolite atlases. Metabolites. 2015;5:431–2.CAS 
PubMed 
PubMed Central 

Google Scholar 
27.Sumner LW, Amberg A, Barrett D, Beale MH, Beger R, Daykin CA, et al. Proposed minimum reporting standards for chemical analysis. Metabolomics. 2007;3:211–21.CAS 
PubMed 
PubMed Central 

Google Scholar 
28.Anders S, Pyl PT, Huber W. HTSeq—a Python framework to work with high-throughput sequencing data. Bioinformatics. 2015;31:166–9.CAS 

Google Scholar 
29.Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, et al. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol. 2012;19:455–77.CAS 
PubMed 
PubMed Central 

Google Scholar 
30.Gurevich A, Saveliev V, Vyahhi N, Tesler G. QUAST: quality assessment tool for genome assemblies. Bioinformatics. 2013;29:1072–5.CAS 
PubMed 
PubMed Central 

Google Scholar 
31.Seemann T. Genome analysis Prokka: rapid prokaryotic genome annotation. Bioinformatics. 2014;30:2068–9.CAS 

Google Scholar 
32.Finn RD, Clements J, Eddy SR. HMMER web server: Interactive sequence similarity searching. Nucleic Acids Res. 2011;39:W29–W37.CAS 
PubMed 
PubMed Central 

Google Scholar 
33.Schwacke R, Ponce-Soto GY, Krause K, Bolger AM, Arsova B, Hallab A, et al. MapMan4: a refined protein classification and annotation framework applicable to multi-omics data analysis. Mol Plant. 2019;12:879–92.CAS 
PubMed 

Google Scholar 
34.Black K, Osborne B. An assessment of photosynthetic downregulation in cyanobacteria from the Gunnera-Nostoc symbiosis. N Phytol. 2004;162:125–32.CAS 

Google Scholar 
35.Santi C, Bogusz D, Franche C. Biological nitrogen fixation in non-legume plants. Ann Bot. 2013;111:743–67.CAS 
PubMed 
PubMed Central 

Google Scholar 
36.Clymo RS. The growth of Sphagnum: some effects of environment. Br Ecol Soc. 1973;61:849–69.
Google Scholar 
37.Lamers LPM, Farhoush C, Van Groenendael JM, Roelofs JGM. Calcareous groundwater raises bogs; the concept of ombrotrophy revisited. J Ecol. 1999;87:639–48.
Google Scholar 
38.Nayak S, Prasanna R. Soil pH and its role in cyanobacterial abundance and diversity in rice field soils. Appl Ecol Environ Res. 2007;5:103–13.
Google Scholar 
39.Jassey VEJ, Meyer C, Dupuy C, Bernard N, Mitchell EAD, Toussaint ML, et al. To what extent do food preferences explain the trophic position of heterotrophic and mixotrophic microbial consumers in a Sphagnum peatland? Micro Ecol. 2013;66:571–80.
Google Scholar 
40.Meeks JC. Symbiosis between nitrogen- fixing cyanobacteria and plants. Symbiosis. 1998;48:266–76.
Google Scholar 
41.Pate S, Lindblad P, Atkins A. Planta in coralloid roots of cycad-Nostoc symbioses. Planta. 1988;176:461–71.CAS 
PubMed 

Google Scholar 
42.Xie B, Chen DS, Zhou K, Xie YQ, Li YG, Hu GY, et al. Symbiotic abilities of Sinorhizobium fredii with modified expression of purL. Appl Microbiol Biotechnol. 2006;71:505–14.CAS 
PubMed 

Google Scholar 
43.Xie B, Chen D, Cheng G, Ying Z, Xie F, Li Y, et al. Effects of the purl gene expression level on the competitive nodulation ability of Sinorhizobium fredii. Curr Microbiol. 2009;59:193–8.CAS 
PubMed 

Google Scholar 
44.Giraud E, Moulin L, Vallenet D, Barbe V, Cytryn E, Avarre JC, et al. Legumes symbioses: absence of Nod genes in photosynthetic bradyrhizobia. Science. 2007;316:1307–12.PubMed 

Google Scholar 
45.Kim JK, Jang HA, Won YJ, Kikuchi Y, Heum Han S, Kim CH, et al. Purine biosynthesis-deficient Burkholderia mutants are incapable of symbiotic accommodation in the stinkbug. ISME J. 2014;8:552–63.CAS 
PubMed 

Google Scholar 
46.An R, Grewal PS. Molecular mechanisms of persistence of mutualistic bacteria Photorhabdus in the entomopathogenic nematode host. PLoS ONE. 2010;5:e13154.PubMed 
PubMed Central 

Google Scholar 
47.Zrenner R, Stitt M, Sonnewald U, Boldt R. Pyrimidine and purine biosynthesis and degradation in plants. Annu Rev Plant Biol. 2006;57:805–36.CAS 
PubMed 

Google Scholar 
48.Atkins CA, Smith PMC. Translocation in legumes: assimilates, nutrients, and signaling molecules. Plant Physiol. 2007;144:550–61.49.Ueda S, Ikeda M, Yamakawa T. Provision of carbon skeletons for amide synthesis in non-nodulated soybean and pea roots in response to the source of nitrogen supply. Soil Sci Plant Nutr. 2008;54:732–7.CAS 

Google Scholar 
50.Kaur H, Chowrasia S, Gaur VS, Mondal TK. Allantoin: emerging role in plant abiotic stress tolerance. Plant Mol Biol Rep. 2021;39:648–61.
Google Scholar 
51.Paul MJ, Primavesi LF, Jhurreea D, Zhang Y. Trehalose metabolism and signaling. Annu Rev Plant Biol. 2008;59:417–41.CAS 
PubMed 

Google Scholar 
52.Iturriaga G, Suárez R, Nova-Franco B. Trehalose metabolism: from osmoprotection to signaling. Int J Mol Sci. 2009;10:3793–810.CAS 
PubMed 
PubMed Central 

Google Scholar 
53.John R, Raja V, Ahmad M, Jan N, Majeed U, Ahmad S, et al. Trehalose: metabolism and role in stress signaling in plants. Stress signaling in plants: genomics and proteomics perspective, Volume 2. 2016. Springer International Publishing, pp 261–75.54.Sharma K, Palatinszky M, Nikolov G, Berry D, Shank EA. Transparent soil microcosms for live-cell imaging and non-destructive stable isotope probing of soil microorganisms. Elife. 2020;9:1–28.
Google Scholar 
55.Streeter JG. Effect of trehalose on survival of Bradyrhizobium japonicum during desiccation. J Appl Microbiol. 2003;95:484–91.CAS 
PubMed 

Google Scholar 
56.Sugawara M, Cytryn EJ, Sadowsky MJ. Functional role of Bradyrhizobium japonicum trehalose biosynthesis and metabolism genes during physiological stress and nodulation. Appl Environ Microbiol. 2010;76:1071–81.CAS 
PubMed 

Google Scholar 
57.Suárez R, Wong A, Ramírez M, Barraza A, Orozco MDC, Cevallos MA, et al. Improvement of drought tolerance and grain yield in common bean by overexpressing trehalose-6-phosphate synthase in rhizobia. Mol Plant-Microbe Interact. 2008;21:958–66.PubMed 

Google Scholar 
58.Mackay MA, Norton RS, Borowitzka LJ. Organic osmoregulatory solutes in cyanobacteria. J Gen Microbiol. 1984;130:2177–91.CAS 

Google Scholar 
59.Reed RH, Richardson DL, Warr SRC, Stewart WDP. Carbohydrate accumulation and osmotic stress in cyanobacteria. J Gen Microbiol. 1984;130:1–4.CAS 

Google Scholar 
60.Csonka LN. Physiological and genetic responses of bacteria to osmotic stress. Microbiol Rev. 1989,53:121–47.61.Moore EK, Hopmans EC, Rijpstra WIC, Villanueva L, Dedysh SN, Kulichevskaya IS, et al. Novel mono-, di-, and trimethylornithine membrane lipids in northern wetland planctomycetes. Appl Environ Microbiol. 2013;79:6874–84.CAS 
PubMed 
PubMed Central 

Google Scholar 
62.Clifford EL, Varela MM, De Corte D, Bode A, Ortiz V, Herndl GJ, et al. Taurine is a major carbon and energy source for marine prokaryotes in the North Atlantic ocean off the Iberian Peninsula. Micro Ecol. 2019;78:299–312.CAS 

Google Scholar 
63.Amin SA, Hmelo LR, van Tol HM, Durham BP, Carlson LT, Heal KR, et al. Interaction and signalling between a cosmopolitan phytoplankton and associated bacteria. Nature. 2015;522:98–101.CAS 
PubMed 

Google Scholar 
64.Dixon RA, Paiva NL. Stress-induced phenylpropanoid metabolism. Plant Cell. 1995,7:1085–97.65.Payyavula RS, Navarre DA, Kuhl JC, Pantoja A, Pillai SS. Differential effects of environment on potato phenylpropanoid and carotenoid expression. BMC Plant Biol. 2012;12:39.CAS 
PubMed 
PubMed Central 

Google Scholar 
66.Huang J, Gu M, Lai Z, Fan B, Shi K, Zhou YH, et al. Functional analysis of the Arabidopsis PAL gene family in plant growth, development, and response to environmental stress. Plant Physiol. 2010;153:1526–38.CAS 
PubMed 
PubMed Central 

Google Scholar 
67.Kouchi H. Large-scale analysis of gene expression profiles during early stages of root nodule formation in a model legume, Lotus japonicus. DNA Res. 2004;11:263–74.CAS 
PubMed 

Google Scholar 
68.Chen Y, Li F, Tian L, Huang M, Deng R, Li X, et al. The phenylalanine ammonia lyase gene LjPAL1 is involved in plant defense responses to pathogens and plays diverse roles in Lotus japonicus -rhizobium symbioses. Mol Plant-Microbe Interact. 2017;30:739–53.CAS 
PubMed 

Google Scholar 
69.Bragina A, Berg C, Cardinale M, Shcherbakov A, Chebotar V, Berg G. Sphagnum mosses harbour highly specific bacterial diversity during their whole lifecycle. ISME J. 2012;6:802–13.CAS 
PubMed 

Google Scholar 
70.Bragina A, Berg C, Müller H, Moser D, Berg G. Insights into functional bacterial diversity and its effects on Alpine bog ecosystem functioning. Sci Rep. 2013;3:1955.PubMed 
PubMed Central 

Google Scholar  More

Data for all analysed radionuclides are presented in the “Supplementary Material”. Cryoconite samples were collected on Nalli Glacier (Supplementary Fig. S1) on Sept. 25, 2017 (samples 1701–1714) and on Sept. 10, 2018 (samples 1801–1814) at 28 spots (Fig. 2, Supplementary Table S1). Gamma spectrometric analysis of samples showed the presence of anthropogenic radionuclides 137Cs, 241Am, and 207Bi. All quoted radioactivity values were recalculated for the sampling date, except those for 241Am since the concentration of the parent 241Pu isotope is unknown. However, for this isotope, the correction for decay is negligible. The activity of 137Cs reached 8093 (± 69) Bq kg−1 of dry weight, that of 241Am reached 58.3 (± 2.3) Bq kg−1 and that of 207Bi reached 6.3 (± 0.6) Bq kg−1. The natural radionuclides 210Pb and 7Be were also present in all samples. The activity of 210Pb varied in the range of 1280–9750 Bq kg−1. In addition, in the investigated samples, a significant amount of short-lived cosmogenic radionuclide 7Be was found, whose specific activity reached 2418 (± 76) Bq kg−1 (Fig. 3, Supplementary Table S2). To evaluate the contribution of atmospheric components to the total 210Pb activity, 226Ra activity was determined and found to be 17–27 Bq kg−1 (Supplementary Table S2). Based on the 210Pb/226Ra ratio, we conclude that more than 98% of 210Pb was of atmospheric provenance.Figure 2Location of sampling points on Nalli Glacier. A—137Cs activity zone  95%) of corresponding rocks and numerous outcrops likely promoted entrapment of these elements into explosion clouds, and their subsequent precipitation with radionuclides. This feature of the geological structure of the area explains the extremely high enrichment of surface waters in elements such as Zn, Pb, Sr, Ni, As, Cr, Co, Se, Te, Cd, W, Cu, Sb, and Sn; for many of them, the excess reaches 10-fold with respect to the Clrake values51. This hypothesis is supported by obvious correlations between the concentrations of Bi, Ag, Sn, Sb, Pb, Cd, W, and Cu and the activity of anthropogenic radionuclides 137Cs, 241Am and 207Bi. This relationship is obviously related to the simultaneous release of elements and radionuclides from the contaminated ice layer and their entrapment in cryoconite holes. More