Particle-associated diazotrophs drive nitrogen fixation in Arctic subsurface waters

Particle-associated diazotrophs drive nitrogen fixation in Arctic subsurface waters | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Particle-associated diazotrophs drive nitrogen fixation in Arctic subsurface waters Mar Benavides, Arthur Coët, Marta Sebastián, Angela Vogts, Christian Furbo Reeder, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9051103/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Biological dinitrogen (N 2 ) fixation sustains productivity in oligotrophic oceans and is now also thought to contribute substantially to the nitrogen supply in the warming Arctic. Here we demonstrate significant N 2 fixation by particle-associated diazotrophs in subsurface waters of the Barents Sea, the Arctic's front runner of nutrient depletion. As the Arctic becomes nutrient-poor, the dominant nitrogen source to sustain future productivity remains unclear. Comparing our findings with subtropical studies reveals particle-associated non-cyanobacterial diazotrophs as the primary N 2 fixers in subsurface Arctic waters, contrasting with diverse communities in warmer regions. As the Arctic shifts towards oligotrophication, understanding the magnitude and controls of particle-associated N₂ fixation provides critical insights into future nitrogen supply and ecosystem transformation across the rapidly changing Arctic Ocean. Earth and environmental sciences/Ocean sciences/Marine biology Biological sciences/Microbiology/Environmental microbiology/Water microbiology Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Dinitrogen (N 2 ) fixation by diazotrophic microorganisms constitutes the main source of reactive nitrogen in the ocean, where it sustains primary productivity and carbon export 1 , 2 . Traditionally, N 2 fixation was considered largely restricted to oligotrophic, warm tropical and subtropical waters where nutrient limitation favours diazotroph activity. However, several studies have expanded this paradigm, documenting significant N 2 fixation rates at higher latitudes, including temperate and polar regions 1 . The Arctic has warmed three to four times faster than the global average since 1979 due to Arctic ‘amplification’ (ice albedo feedback and heat transport 3 , 4 ) and ‘atlantification’ (warm Atlantic origin water spreading further into the Arctic basin subsurface 5 ). Continued warming leads to glacier and sea-ice melting, which releases nutrients and iron 6 advancing the onset and increasing the magnitude of phytoplankton blooms 7 , 8 . The extent of ice-free waters increases over the year, enhancing nitrogen depletion by phytoplankton and eventually leading to decreased primary productivity and carbon export 9 . The progressively earlier seasonal depletion of nitrogen in the Arctic may expand the niche for diazotrophs 10 , 11 . A preliminary study estimated N 2 fixation supports 17% of the primary productivity in the region 12 , and current ecosystem models predict an increase of diazotrophs in Arctic latitudes towards the end of the 21st century 13 . More recent studies have documented active N 2 fixation across diverse Arctic Ocean regions, including the Chukchi Sea, Central Arctic, Eurasian marginal ice zones, and Fram Strait 14 – 17 . The only specific N 2 fixation measurements available in the region show that the nitroplast (formerly UCYN-A) of the eukaryotic alga Braarudosphaera bigelowii 18 actively fixes N 2 in the Pacific sector of the Arctic at rates similar to those of (sub)tropical waters 17 . Despite the cosmopolitan distribution of this nitroplast across latitudes globally 19 , non-cyanobacterial diazotrophs (NCDs) overwhelmingly dominate nifH gene sequences and expression in the Arctic 14 , 15 , 20 – 23 . Contrary to diazotrophic cyanobacteria, NCDs do not have photosynthetic machinery 24 and based on their genomes, are predicted to rely on other external sources of carbon and energy 25 . Organic particles are rich in carbon and can harbour low oxygen microzones, making them ideal niches for NCDs to fix N 2 26,27 . Recent studies have demonstrated that gammaproteobacterial NCDs fix N₂ in association with particles in the subtropical North Pacific 28 , 29 , but the recurrence and N 2 fixation potential of these associations in the Arctic is unknown. Arctic waters with seasonal ice coverage export almost twice as many particles as ice-free zones 30 , and combined with the Arctic's extremely broad shelf areas, supply coastal waters with substantial volumes of sediment-laden particles rich in potentially bioavailable iron and carbon 31 - key resources enabling NCDs to thrive and perform N 2 fixation 25 , 26 . Here, we aimed to evaluate the N 2 fixation potential and identity of particle-associated NCDs in the Arctic Ocean. Combining single-cell spectrometry, immunofluorescence and molecular methods, we find that particle-associated N₂ fixation in the subsurface waters of the Barents Sea is predominantly driven by NCDs. By comparing our Arctic findings with subtropical studies, we demonstrate that particle-associated NCDs stand out as the dominant diazotrophs in Arctic waters, unlike the more diverse diazotroph communities reported from warmer regions. As a region at the forefront of Arctic oligotrophication 32 , 33 , the magnitude and controls of particle-associated N 2 fixation in the Barents Sea provide hints on the future of reactive nitrogen supply in this globally climate regulating region. Results and Discussion We sampled at fourteen stations aiming at covering the Atlantic and Arctic influenced end-members of the region 34 , including Atlantic Water (AW), warm Atlantic Water (wAW) and Arctic Water (ArW) water masses (Fig. 1 , Table S1 ). Suspended, slow and fast sinking particles (‘susp’, ‘ss’ and ‘fs’, respectively) were collected using a marine snow catcher (MSC) deployed 10 m below the deep chlorophyll maximum 35 , ranging from 22 to 66 m during our cruise (Table S1 ; Fig. S1 ). MSC samples were size-fractionated, allowing to distinguish particle-associated (> 3 µm) from free-living cells ( 0.2 µm) in both DNA and single-cell N 2 fixation measurements (see Methods and Supplementary Information). Bulk measurements across individual MSC fractions revealed that the susp fraction had the highest N₂ fixation rates (up to 1.62 nmol N l − 1 d − 1 ), followed by the fs fraction (up to 0.46 nmol N l − 1 d − 1 ), and the ss fraction (up to 0.22 nmol N l − 1 d − 1 ) (Fig. S2 , Table S2 ). These rates are five to ten times lower than previous measurements in surface waters of the Bering, Chukchi and Beaufort Seas (compiled in ref. 10 ), but in the range of surface water bulk N 2 fixation rates during our cruise in the Barents Sea (Mahaffey et al., in prep.). Based on the bulk N 2 fixation rates observed, a subset of five stations (N00, N06, N08, N10 and N13; Fig. 1 , Table S1 ) was selected for single-cell analysis. Measuring N 2 fixation rates of individual NCDs taxa usually requires combining oligonucleotide probing (catalysed reporter deposition-fluorescence in situ hybridisation or CARD-FISH) with nanoscale secondary ion mass spectrometry (nanoSIMS; e.g., 28,29 ). While this approach resolves relatively broad taxonomic groups, it does not specifically capture the broader NCDs assemblage, which spans multiple taxonomic lineages and is widespread across the global oceans 25 . Here we applied a nitrogenase iron protein (NifH) antibody staining approach able to target virtually all diazotrophs independently of their taxonomy 36 . Based on NifH immunofluorescence counts on 3 and 0.2 µm filters, we found that particle-associated diazotrophs (3 µm filters) comprised 0.25–9.26% of the total microbial community, with the highest proportions observed at stations in wAW waters (Fig. S2 ), situated near the Polar Front 37 . NifH-immunofluorescence-positive cells in the ss fraction were exclusively detected in wAW waters (Fig. S3 ), corresponding to the lowest bulk N₂ fixation rates (Fig. S2 ); consequently, this fraction was excluded from subsequent nanoSIMS analyses. Mapping immunolabeled cells in 3 µm filters for nanoSIMS, we measured particle-associated N 2 fixation rates on 729 individual cells. Most of the rates ranged between 0 and 10 fmol N cell − 1 d − 1 , although values up to 1192 fmol N cell − 1 d − 1 were observed (Table S3 ). Averaging particle-associated N 2 fixation rates per water mass showed an increase from AW (1.83 ± 5.95 fmol N cell − 1 d − 1 , n = 130), through wAW (2.33 ± 5.77 fmol N cell − 1 d − 1 , n = 319), to ArW in the inner Barents Sea (23.2 ± 111 fmol N cell − 1 d − 1 , n = 280), being significantly higher in fs than in susp particles (Kruskal-Wallis test, **p < 0.01) (Fig. 2 ). These rates are higher than those measured on two different B. bigelowii nitroplast lineages in the Bering and Chukchi Seas (7.6 and 13.0 fmol N cell − 1 d − 1 , respectively 17 ), but similar to NCDs’ single-cell N 2 fixation rates measured in the North Pacific Ocean (67–121 fmol N cell − 1 d − 1 29 ). This suggests that particle-associated NCDs exhibit consistent N 2 fixation performance across latitudes. Although the immunolabeling approach used here detects virtually all diazotrophs synthesising the NifH protein, it cannot distinguish between taxa, complicating determination of whether particle-associated diazotrophs were exclusively NCDs or also included diazotrophic cyanobacteria, as the latter are often observed embedded in particles in low latitude waters 2 , 38 , 39 . In this case, cell size can be helpful to discriminate cyanobacteria from NCDs. Although we measured a few outlier cells with diameters of up to 5.24 µm (Table S3 ), 85% of the cells analysed had diameters ranging from 0.4 to 2 µm (Fig. S4 ). Cells within this size range could potentially correspond to the B. bigelowii nitroplast, that is known to fix N 2 in the Arctic 17 and was also identified in surface waters during our cruise (Mahaffey et al., in prep.). However, we can confidently exclude the nitroplast as a contributor to N 2 fixation in subsurface waters as no corresponding nifH gene sequences were detected in our samples (Table S4 ). The sporadic higher rates measured in cells > 2 µm could be attributed to the unicellular cyanobacterium Cyanothece (Amplicon Sequence Variant (ASV)51, Table S3 ), which has been previously observed in Arctic seawater and sea ice brine 40 . Given that most of the active N 2 -fixing cells detected in our study were 4 µm) were only sparsely detected in nifH gene sequences (Table S4 ), we can confidently conclude that particle-associated N 2 fixation in our study was chiefly driven by NCDs. The top ten ASVs belonged to putative motile genera (Fig. 3 ), which would support a particle-associated lifestyle 41 . Azotobacter was mostly detected in the free-living fraction (i.e., 0.2 µm filters) across all water masses sampled (Fig. 3 ). This genus has been previously reported from Arctic melt ponds, water under ice and algal aggregates 22 , and is able to use carbohydrates and organic acids as sources of carbon. Desulfuromonas , which are sulphate-reducers and typical of oxygen deficient waters 42 , were particularly abundant in wAW waters where the MSC deployment depth (i.e., 10 m below the deep chlorophyll maximum) coincided roughly with the oxycline (Figs. S1, S5). Bradyrhizobium was more evenly detected across water masses, while Methylococcus was clearly prevalent in AW (Fig. 3 , Fig. S5 ). Active N 2 -fixing Bradyrhizobia species have been found in driftwood around Svalbard 43 , while Methylococcus are typically found in sediment and wetland ecosystems 44 , 45 . Xanthobacter is a soil methylmercury degrading bacterium, found even in deep bathypelagic waters of the Arctic 46 . This suggests that the samples collected included particles of terrestrial and/or glacial origin, consistent with a recent study 20 and the well-known role of ice-laden sediments to transport microbial communities across the Arctic 47 . Particles in the Arctic Ocean originate from diverse sources, including phytoplankton, macroalgae, and glacial sediment runoff 48 – 50 . However, phytoplankton derived particulate organic carbon export is usually minimum in the boreal summer 51 , when our cruise took place. Over the summer months, as Arctic phytoplankton blooms decline and are exported from surface waters, particles of terrestrial and/or glacial origin likely dominate the water column, as suggested by the diazotroph community composition in our samples (Fig. 3 ), which resembles sediment-associated assemblages. Alternatively, the scarcity of typical pelagic diazotroph communities in our data may reflect respiration of phytoplankton-derived particles by our sampling depth (10 m below the deep chlorophyll maximum, 22–66 m; Table S1 , Fig. S1 ), consistent with particulate organic carbon attenuation horizons as shallow as 30–60 m in the Barents Sea 52 . Comparing this study with studies in the North Pacific 29 and North Atlantic 53 (which employed the same MSC sampling strategy at 10 m below DCM; see Supplementary Information) reveals that tparticle-associated diazotroph communities in the Arctic are remarkably different from those of subtropical regions (Fig. S6 A), showcasing a high degree of endemicity (Fig. S6 B). We find that while gamma- and betaproteobacteria were present in particle-associated diazotroph communities across the North Pacific, North Atlantic and Arctic sites, alphaproteobacterial and Desulfuromonadia were not detected in the North Pacific (Fig. S6 C). Interestingly, cyanobacteria dominated the particle-associated diazotroph community in subsurface waters of both the North Pacific and North Atlantic, whereas they were virtually absent in the Arctic, where NCDs comprised nearly 100% of the particle-associated diazotroph community (Fig. 4 ). This aligns with recent modelling efforts showing that particle-associated NCDs are best suited for fixing N 2 at low temperatures as compared to cyanobacterial diazotrophs 54 . The anticipated shift to an ice-free Barents Sea will extend the oligotrophic summer period, characterised by low algal biomass and consequently reduced vertical particulate organic carbon flux 9 , pointing towards terrestrial and glacial runoff particle-associated NCDs as potential important contributors to nitrogen availability in the future. Methods Particle sampling The DY167 N-Arc GEOTRACES GApr19 cruise took place from 9 July to 13 August 2023 onboard the RRS Discovery. Hydrographic and biogeochemical data acquired included CTD casts, macro- and micronutrients (see Supplementary Information for detailed methods). A marine snow catcher (MSC; OSIL, Havant, Hampshire, UK) was deployed at 10 m below the deep chlorophyll maximum at fourteen stations (Fig. 1 , Table S1 ). After 2 h of settling on deck, the suspended (susp), slow sinking (ss) and fast sinking (fs) particle fractions were subsequently sampled. The susp fraction was sampled in nine 4000 ml polycarbonate bottles, where three bottles were used for N 2 fixation rate measurements (see below), three bottles for DNA sampling, and three bottles for particulate matter natural 15 N atom % enrichment measurements. The ss and fs fractions were sampled in triplicate 500 ml polycarbonate bottles and 40 ml glass flasks, respectively, for N 2 fixation measurements. The content of the other triplicate sets of 500 ml polycarbonate bottles and 100 ml glass flasks was size-fractionated by filtration through 3 and 0.2 µm filters and used for DNA extraction and nifH gene amplification (Supplementary Information). Bulk N fixation rates Each MSC fraction was incubated in triplicate with 10% v/v 15 N 2 -enriched filtered seawater for 48 h in a controlled temperature room adjusted to the temperature of each sampling depth at each station (Table S1 ). Enriched seawater was prepared from 2300 ml seawater, filtered through a 0.2 µm pore size Sartobran PTFE cartridge (Sartorius, Göttingen, Germany), injected with 25 ml 98 atom % 15 N 2 gas (Cambridge Isotope Laboratories, Tewksbury, MA, USA), and stirred on a magnet stirred plate overnight. After incubation, 4000 ml, 500 ml, and 30 ml from the susp, ss and fs fractions, respectively, were filtered onto pre-combusted (450ºC, 6 h) GFF filters (Whatman, Little Chalfont, Buckinghamshire, UK) and stored at -20ºC. Filters were dried at 60ºC for 24 h, pelletised and analysed on an INTEGRA2 Sercon elemental analyser coupled to an isotope ratio mass spectrometer (Crewe, Cheshire, UK) for particulate 15 N/ 14 N isotope ratios, particulate carbon (PC) and particulate nitrogen (PN) concentrations. Subsamples from the incubated susp fraction were taken on 12 ml Exetainer® tubes (Labco, Exeter, UK) and the 15 N atom % enrichment of seawater measured using a membrane inlet mass spectrometer onboard (MIMS, Hiden HPR40, Warrington, UK). Bulk N 2 fixation rates were calculated according to White et al. 55 , applying volume corrections for the different MSC fractions as recommended by Riley et al. 35 . NifH antibody staining, imaging and cell counts Particle-associated single-cell N 2 fixation rates were measured from 500 ml, 50 ml, and 10 ml subsamples taken from each MSC fraction, filtered through 3 µm Isopore polycarbonate filters (Millipore, Tullagreen, Carrightwohill, Ireland). Filters were fixed with 1.6% electron microscopy grade paraformaldehyde (Thermo Fisher, Ward Hill, MA, USA) for 1 h at room temperature, air-dried and stored at -80ºC until further analysis. Samples with significant bulk N 2 fixation rates were selected for nitrogenase immunolabeling 36 and nanoSIMS analyses (stations N00, N06, N08, N10 and N13; see below). Active diazotroph cells were identified and counted by nitrogenase immunolabeling and their contribution to the total bacterial community assessed from comparative DAPI staining. Briefly, cells in filter portions were permeabilised with 0.5% (v/v) dimethyl sulfoxide (DMSO; Sigma-Aldrich, Saint Louis, MO, USA) and washed three time for 5 min in phosphate-buffered saline containing 0.1% Triton X-100 (PBST, Dutscher, Brumath, France), supplemented with 5% (w/v) bovine serum albumin (BSA). Samples were then incubated for 1 h at room temperature in the dark with a polyclonal anti-NifH primary antibody (Agrisera, cat. AS01021A, Vännäs, Sweden) diluted 1:166 (1.2 µl antibody in 198.8 µl PBS). After three rounds of 5 min washes in PBST/BSA, samples were incubated for 45 min at room temperature with a goat anti-Chicken IgY (H + L) secondary antibody conjugated to Alexa Fluor™ 594 (Invitrogen, A-11042), diluted 1:181 (1.1 µl antibody in 198.9 µl PBS). Samples were subsequently washed three times for 5 min in PBST/BSA, and dehydrated in 70% ethanol, and subsequently mounted on slides using 4’,6-diamidino-2-phenylindole (DAPI, ProLong™ glass with DAPI, Thermo Fisher Scientific, Waltham, MA, USA) and stored at -20°C until visualisation. To test the specificity of the antibody, anti-nitrogenase antibody controls were performed (Supplementary Information). Samples were visualised by epifluorescence microscopy using a Zeiss Axio Imager.Z2m microscope (Carl Zeiss MicroImaging, Barcelona, Spain) equipped with an AxioCam MRm camera, at 630× magnification. Images were acquired in black and white using the AxioVision software. A Colibri LED light source was used with the HE-62 multi-filter module, including excitation filters BP 370/40 for DAPI and BP 585/35 for Alexa594, a triple beam splitter (TFT 395/495/605), and emission filters TBP 425/46 + 527/54 + LP615. Exposure times were set to 100 ms for DAPI and 120 ms for the Alexa594-labelled antibody. At least ten fields of view were recorded per filter. Image processing and cell quantification were performed using the automated image analysis software ACMEtool3 (48). Cells were first identified in the DAPI channel and classified as bacterial based on their size (area 18–200 pixels), with an additional signal-to-background ratio (SBR > 2) constraint applied to 0.2 µm filters. NifH-positive cells were subsequently identified within the DAPI mask by detecting an AF594 fluorescence signal associated with a DAPI-identified cell. Cells were classified as NifH⁺ only when an AF594 signal overlapped a DAPI-labelled cell and met channel-specific detection thresholds applied during image analysis. In total, 1,952 fields of view were analysed on 3 µm (particle-associated) filters (95,908 DAPI cells screened, including 3,064 NifH positive cells), and 1,753 fields of view on 0.2 µm (free-living) filters (986,147 DAPI cells screened, including 7,687 NifH positive cells). No NifH positive cells were detected in the ss fraction, therefore, single-cell mass spectrometry analyses were performed only on the susp (susp; n = 402 cells) and fs (fs; n = 327 cells) fractions. Single-cell N fixation rates Samples displaying the strongest and most spatially coherent NifH signal (i.e., NifH positive) on 3 µm filters were selected for nanoSIMS analyses and re-examined by epifluorescence microscopy using an Olympus BX61 microscope equipped with a monochrome Hamamatsu ORCA camera. NifH immunofluorescence was detected using a custom optical cube assembled from Chroma filter components, consisting of an ET599/13x excitation filter, a T612lpxr long-pass dichroic mirror, and an ET632/28m emission filter (Chroma Technology Corp., Bellows Falls, VT, USA), enabling enhanced isolation of the nitrogenase-associated signal and minimising spectral overlap with Chorophyll a fluorescence. DAPI fluorescence was detected using a UB-BP filter set (excitation 360/40 nm, dichroic 400 LP, emission 460/50 nm). Selected regions from the corresponding filters were then subjected to a freeze-transfer procedure to relocate cells and particles onto a solid substrate compatible with high-resolution analyses. Briefly, filter sections were gently rinsed with Milli-Q water, placed upside down onto patterned silicon wafers (1.2 × 1.2 cm, 1 × 1 mm raster; Pelotec SFG12 Finder Grid substrate, Ted Pella Inc., Redding, CA, USA), rapidly frozen at -80°C, and carefully removed while frozen to promote transfer of attached material onto the wafer surface 29 . Following transfer, silicon wafers were re-examined using the same epifluorescence microscope and filter sets to relocate NifH-positive particles. The engraved coordinate grid of the silicon substrate was used to map the position of individual target particles, providing precise spatial references for subsequent NanoSIMS analyses. Prior to NanoSIMS measurements, wafers were coated with a gold layer of approximately 30 nm thickness. The 15 N atom % enrichment of particle-associated diazotroph cells was measured on a nanoSIMS 50L (CAMECA, Gennevilliers, France) at the Leibniz Institute for Baltic Sea Research (IOW, Germany). A 1 pA 16 keV Cesium (Cs + ) primary beam was scanned on a 512 x 512 pixel raster with a raster area of 15 x 15 µm, and a counting time of 250 µs per pixel. Samples were pre-sputtered with 600 pA Cs + current for 2 min in a raster of 30 x 30 µm to remove the gold and surface contaminants and reach the steady state of ion formation. Negative secondary ions 12 C − , 13 C − , 12 C 14 N − , 12 C 15 N − and 31 P − were detected with electron multiplier detectors, and secondary electrons were simultaneously imaged. Sixty serial quantitative secondary ion mass planes were generated, drift corrected and accumulated to the final image. Mass resolving power was > 8000 to resolve isobaric interferences. Data was processed using the Look@nanoSIMS software 56 . Isotope ratio images were generated by dividing the 12 C 15 N − ion count by the 12 C 14 N − ion count pixel by pixel. Individual diazotroph cells were identified from 12 C − and 31 P images. These images were used to define regions of interest (ROIs). For each ROI, the 15 N/ 14 N ratios were calculated based on the ion counts averaged over the ROIs. Additional ROIs were defined in areas without visible ¹⁵N enrichment to estimate internal background ratios, which were centred around natural abundance (~ 0.00367; 15 N atom %). Cells were considered isotopically enriched only when their ¹⁵N/¹⁴N ratios exceeded both natural abundance and the Poisson counting error associated with ion counting statistics. The total number of particles analysed was 729. Single-cell N 2 fixation rates were calculated according to Furbo Reeder et al. 29 . Data analysis, statistics, and plots All data and statistical analyses and plots were done in RStudio v2024.04.2 57 . All plots were done using the ggplot2 58 package, except CTD plots that were done using the oce 59 R package. N 2 fixation values were checked for normality using a Shapiro-Wilk test and significant differences between samples tested with Wilcoxon test using the dplyr package 60 . Amplicon data heatmaps plotted with the ampvis2 R package 61 . Redundancy analysis biplots were calculated and plotted with the phloyseq R package 62 . Missing values in the environmental metadata were imputed prior to redundancy analysis to avoid excessive sample loss. Remaining missing entries in the mixed-type metadata set were imputed using the random forest-based algorithm implemented in the missForest package in R 63 , with a fixed random seed (123) to ensure reproducibility. The imputed metadata matrix (Table S4 ) was subsequently used as the sample-data component when constructing the phyloseq object for ordination analyses. Data availability Source data is provided with this paper. Stainless steel and titanium cast data are available at DOI: 10.5285/48e39c93-2405-99f3-e063-7086abc003fa and DOI: 10.5285/48c1d94c-1647-8717-e063-7086abc0f520 , respectively. nifH gene sequences are available at NCBI under project number PRJNA1412537. Declarations Competing Interests The authors declare that they have no competing interests. Author Contributions MB designed the experiments. MB and AC performed sampling in the Arctic. AC performed immunolabeling assays. AC and MS processed immunolabeling and DAPI staining images at ICM-CSIC (Barcelona, Spain). AC and AV performed nanoSIMS analyses at IOW (Warnemünde, Germany). CFR and EC-G processed MSC nifH data from subtropical cruises for comparison purposes. AC processed Arctic nifH sequencing data. CM, MCL and JEH secured funding for the research cruise. RH, OF, MCL, JEH and CM provided contextual biogeochemical and hydrographic Arctic cruise data. CFR and EC-G contributed to ocean basin intercomparison analyses. MB and AC analysed data and wrote the manuscript with inputs from all co-authors. Acknowledgements The N-ARC project was funded under Natural Environment Research Council (NERC) Discovery grant Nitrogen fixation in the Arctic Ocean, N-ARC (grant numbers NE/T001240/1 and NE/T000570/1) held at the University of Liverpool (CM), and the National Oceanography Centre Liverpool (JEH). Sample analyses were supported by the projects ANITA “Architecture and dyNamics of marIne parTicle colonizAtion” funded by the Excellence Initiative of Aix-Marseille University A*MIDEX, the project PANDA “PArticles as Niches for non-cyanobacterial DiAzotrophs” funded by INSU-EC2CO, and the project MANIOC “iMpact of particle microbial colonisation on Nitrogen cyclIng in the OCean" granted by Aix-Marseille Université Institut Océans to MB. MB was funded by the BIOPOLE National Capability Multicentre Round 2 funding from the Natural Environment Research Council (grant no. NE/W004933/1). Horizon MSCA grant 101150634 to CFR. The NanoSIMS at the Leibniz Institute for Baltic Sea Research (IOW) was funded by the German Federal Ministry of Education and Research (BMBF), grant identifier 03F0626A. The authors warmly thank the captain and crew of the RRS Discovery , as well as NMF staff for assistance with overboard operations. The authors are indebted to L. Wrightson and B. Fisher for MIMS, and L. Norman for nutrient analyses onboard, respectively. We would also like to thank K. Turk-Kubo and J. Magasin for nifH gene annotation, I. Forn for microscopy analyses, and the IBDM platform in Marseille for filter laser marking. Annett Grüttmüller is acknowledged for nanoSIMS routine operation. Finally, we thank S. Giering for providing the MSC devices used in this study. References Zehr JP, Capone DG (2020) Changing perspectives in marine nitrogen fixation. Science 368:eaay9514 Bonnet S et al (2023) Diazotrophs are overlooked contributors to carbon and nitrogen export to the deep ocean. ISME J 17:47–58 Zhou W, Leung LR, Lu J (2024) Steady threefold Arctic amplification of externally forced warming masked by natural variability. Nat Geosci 17:508–515 Rantanen M et al (2022) The Arctic has warmed nearly four times faster than the globe since 1979. Commun Earth Environ 3:1–10 Polyakov IV et al (2017) Greater role for Atlantic inflows on sea-ice loss in the Eurasian Basin of the Arctic Ocean. Science 356:285–291 Wang S, Bailey D, Lindsay K, Moore JK, Holland M (2014) Impact of sea ice on the marine iron cycle and phytoplankton productivity. Biogeosciences 11:4713–4731 Thiele S et al (2023) Seasonality of the bacterial and archaeal community composition of the Northern Barents Sea. Front Microbiol 14:1213718 Lewis KM, van Dijken GL, Arrigo K (2020) R. Changes in phytoplankton concentration now drive increased Arctic Ocean primary production. Science 369:198–202 Kohlbach D et al (2023) Earlier sea-ice melt extends the oligotrophic summer period in the Barents Sea with low algal biomass and associated low vertical flux. Prog Oceanogr 213:103018 von Friesen LW, Riemann L (2020) Nitrogen Fixation in a Changing Arctic Ocean: An Overlooked Source of Nitrogen? Front Microbiol 11:596426 Arrigo KR, van Dijken GL (2015) Continued increases in Arctic Ocean primary production. Prog Oceanogr 136:60–70 Sipler RE et al (2017) Preliminary estimates of the contribution of Arctic nitrogen fixation to the global nitrogen budget. Limnol Oceanogr Lett 2:159–166 Henson SA, Cael BB, Allen SR, Dutkiewicz S (2021) Future phytoplankton diversity in a changing climate. Nat Commun 12:1–8 von Friesen LW et al (2025) Nitrogen fixation under declining Arctic sea ice. Commun Earth Environ 6:811 von Friesen LW et al (2025) From temperate to polar waters: Transition to non-cyanobacterial diazotrophy upon entering the Atlantic gateway of the Arctic Ocean. Limnol Oceanogr. 10.1002/lno.70153 Zander S et al (2025) Contrasting nitrogen fixation between arctic and Atlantic waters in the Fram Strait. Microb Ecol 89:26 Harding K et al (2018) Symbiotic unicellular cyanobacteria fix nitrogen in the Arctic Ocean. Proceedings of the National Academy of Sciences 201813658 10.1073/pnas.1813658115 Coale TH et al (2024) Nitrogen-fixing organelle in a marine alga. Science 384:217–222 Cabello AM et al (2016) Global distribution and vertical patterns of a prymnesiophyte-cyanobacteria obligate symbiosis. ISME J 10:693–706 von Friesen LW, Paulsen ML, Müller O, Gründger F, Riemann L (2023) Glacial meltwater and seasonality influence community composition of diazotrophs in Arctic coastal and open waters. FEMS Microbiol Ecol 99 Blais M et al (2012) Nitrogen fixation and identification of potential diazotrophs in the Canadian Arctic. Global Biogeochem Cycles 26 Fernández-Méndez M et al (2016) Diazotroph diversity in the sea ice, melt ponds, and surface waters of the eurasian basin of the Central Arctic Ocean. Front Microbiol 7:1–18 Fernandez L, Bertilsson S, Peura S (2020) Non-cyanobacterial diazotrophs dominate nitrogen‐fixing communities in permafrost thaw ponds. Limnol Oceanogr 65:S180–S193 Delmont TO et al (2021) Heterotrophic bacterial diazotrophs are more abundant than their cyanobacterial counterparts in metagenomes covering most of the sunlit ocean. ISME J. 10.1038/s41396-021-01135-1 Turk-Kubo KA et al (2022) Non-cyanobacterial diazotrophs: Global diversity, distribution, ecophysiology, and activity in marine waters. FEMS Microbiol Rev. 10.1093/femsre/fuac046 Riemann L et al (2022) Planktonic Aggregates as Hotspots for Heterotrophic Diazotrophy: The Plot Thickens. Front Microbiol 13:875050 Chakraborty S et al (2021) Quantifying nitrogen fixation by heterotrophic bacteria in sinking marine particles. Nat Commun 12:4085 Harding KJ et al (2022) Cell-specific measurements show nitrogen fixation by particle-attached putative non-cyanobacterial diazotrophs in the North Pacific Subtropical Gyre. Nat Commun 13:1–10 Furbo Reeder C et al (2025) Unveiling the contribution of particle-associated non-cyanobacterial diazotrophs to N 2 fixation in the upper mesopelagic North Pacific Gyre. Commun Biol Fadeev E et al (2021) Sea ice presence is linked to higher carbon export and vertical microbial connectivity in the Eurasian Arctic Ocean. Commun Biol 4:1255 Markussen TN, Elberling B, Winter C, Andersen TJ (2016) Flocculated meltwater particles control Arctic land-sea fluxes of labile iron. Sci Rep 6:24033 Slagstad D, Wassmann PFJ, Ellingsen I (2015) Physical constrains and productivity in the future Arctic Ocean. Front Mar Sci 2 Lind S, Ingvaldsen RB, Furevik T (2018) Arctic warming hotspot in the northern Barents Sea linked to declining sea-ice import. Nat Clim Chang 8:634–639 Sundfjord A et al (2020) Suggested water mass definitions for the central and northern Barents Sea, and the adjacent Nansen Basin. Nansen Leg Rep Ser. 10.7557/nlrs.5707 Riley JS et al (2012) The relative contribution of fast and slow sinking particles to ocean carbon export. Global Biogeochem Cycles 26 Geisler E, Bogler A, Rahav E (2019) Bar-Zeev, E. Direct Detection of Heterotrophic Diazotrophs Associated with Planktonic Aggregates. Sci Rep 9:1–9 Barton BI, Lenn Y-D, Lique C (2018) Observed atlantification of the Barents Sea causes the Polar Front to limit the expansion of winter sea ice. J Phys Oceanogr 48:1849–1866 Benavides M et al (2022) Sinking Trichodesmium fixes nitrogen in the dark ocean. ISME J 16:2398–2405 Farnelid H et al (2018) Diverse diazotrophs are present on sinking particles in the North Pacific Subtropical Gyre. ISME J. 10.1038/s41396-018-0259-x Díez B, Bergman B, Pedrós-Alió C, Antó M, Snoeijs P (2012) High cyanobacterial nifH gene diversity in Arctic seawater and sea ice brine. Environ Microbiol Rep 4:360–366 Hallstrøm S et al (2022) Chemotaxis may assist marine heterotrophic bacterial diazotrophs to find microzones suitable for N2 fixation in the pelagic ocean. ISME J 1–10. 10.1038/s41396-022-01299-4 Reeder CF, Löscher CR (2022) Nitrogenases in Oxygen Minimum Zone Waters. Front Mar Sci 9 Mäkipää R et al (2018) Methanotrophs are core members of the diazotroph community in decaying Norway spruce logs. Soil Biol Biochem 120:230–232 Fernandez L et al (2020) Diazotroph genomes and their seasonal dynamics in a stratified humic bog lake. Front Microbiol 11:1500 He R et al (2012) Shifts in identity and activity of methanotrophs in arctic lake sediments in response to temperature changes. Appl Environ Microbiol 78:4715–4723 Sanz-Sáez I et al (2024) Microorganisms involved in methylmercury demethylation and mercury reduction are widely distributed and active in the bathypelagic deep ocean waters. Environ Sci Technol 58:13795–13807 Krumpen T et al (2019) Arctic warming interrupts the Transpolar Drift and affects long-range transport of sea ice and ice-rafted matter. Sci Rep 9:5459 Neukermans G, Reynolds RA, Stramski D (2016) Optical classification and characterization of marine particle assemblages within the western Arctic Ocean. Limnol Oceanogr 61:1472–1494 Hopwood MJ et al (2019) Review Article: How does glacier discharge affect marine biogeochemistry and primary production in the Arctic? Cryosphere Discuss 1–51. 10.5194/tc-2019-136 Swoboda S et al (2024) Release of ballast material during sea-ice melt enhances carbon export in the Arctic Ocean. PNAS Nexus 3:gae081 Dybwad C et al (2022) The influence of sea ice cover and Atlantic water advection on annual particle export north of Svalbard. J Geophys Res C: Oceans 127 Reigstad M, Wexels Riser C, Wassmann P, Ratkova T (2008) Vertical export of particulate organic carbon: Attenuation, composition and loss rates in the northern Barents Sea. Deep Sea Res Part 2 Top Stud Oceanogr 55:2308–2319 Cerdán-García E, Álvarez-Salgado XA, Arístegui J, Martínez-Marrero A, Benavides M (2024) Eddy-driven diazotroph distribution in the subtropical North Atlantic: horizontal variability prevails over particle sinking speed. Commun Biol 7:929 Chakraborty S, Andersen KH, Merico A, Riemann L (2025) Particle-associated N2 fixation by heterotrophic bacteria in the global ocean. Sci Adv 11:eadq4693 White AE et al (2020) A critical review of the 15N2 tracer method to measure diazotrophic production in pelagic ecosystems. Limnol Oceanogr Methods 18:129–147 Polerecky L et al (2012) Look@NanoSIMS - a tool for the analysis of nanoSIMS data in environmental microbiology. Environ Microbiol 14:1009–1023 RStudio Team. RStudio (2023) Wickham H (2009) Ggplot2: Elegant Graphics for Data Analysis . (Springer Science & Business Media. 10.1007/978-0-387-98141-3 Kelley D (2025) R. C. Oce: Analysis of Oceanographic Data Wickham H, François R, Henry L, Müller K, Vaughan D (2023) Dplyr: A Grammar of Data Manipulation Andersen KS, Kirkegaard RH, Karst SM, Albertsen M (2018) Ampvis2: An R Package to Analyse and Visualise 16S rRNA Amplicon Data . 10.1101/299537 McMurdie PJ, Holmes S, Phyloseq (2013) An R Package for Reproducible Interactive Analysis and Graphics of Microbiome Census Data. PLoS ONE 8 Stekhoven DJ, Bühlmann P (2012) MissForest–non-parametric missing value imputation for mixed-type data. Bioinformatics 28:112–118 Additional Declarations There is NO Competing Interest. Supplementary Files FigS4.jpg Fig S4 FigS1.jpg Fig S1 TableS5.xlsx Table S5 NARCMSCSI.docx Supplementary Information FigS7.jpg Fig S7 TableS3.xlsx Table S3 TableS2.xlsx Table S2 FigS2.png Fig S2 TableS1.xlsx Table S1 FigS5.jpg Fig S5 FigS3.jpg Fig S3 FigS6.png Fig S6 TableS4.xlsx Table S4 Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9051103","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":603348783,"identity":"a431d5e2-64d9-4f51-a7b7-3516c2cb177d","order_by":0,"name":"Mar Benavides","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYJACZihpwMBQARdMIFbLGZK0MAC1MLYRoUW3/ewD5oKKO/IGx5k3Pvw5744cfwPzww+MbWk4tZidSTdgnnHmmeGGw2zFxrzbnhlLHGAzlmBsy8Gt5UAaAzNv22HGmc08ZtKM2w4nbgAKAl1YgVvL+WdALf8O2wO1mP/8OQekhf0bfi03QLY0HE7sZ+YxYwAxNjDwgGzB47AbzxgOzzh2OLmfma1YmufYYWOJwzzFEgnn8Hj/fBrj44Kaw7Zt/Ic3fvxRc1iOv71944cPZck4tYDAAVQuKJoS8GoYBaNgFIyCUUAIAABlf1AJWqbKsQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-9502-108X","institution":"National Oceanography Centre","correspondingAuthor":true,"prefix":"","firstName":"Mar","middleName":"","lastName":"Benavides","suffix":""},{"id":603348784,"identity":"01ab3c4c-5c93-482d-af9f-4216bf436d2c","order_by":1,"name":"Arthur Coët","email":"","orcid":"","institution":"Mediterranean Institute of Oceanography","correspondingAuthor":false,"prefix":"","firstName":"Arthur","middleName":"","lastName":"Coët","suffix":""},{"id":603348785,"identity":"12baedc2-c027-4574-8a05-4170a7eb0afe","order_by":2,"name":"Marta Sebastián","email":"","orcid":"https://orcid.org/0000-0001-7175-8941","institution":"Institut de Ciencies del Mar-CSIC","correspondingAuthor":false,"prefix":"","firstName":"Marta","middleName":"","lastName":"Sebastián","suffix":""},{"id":603348786,"identity":"e07e3e12-8152-4cb9-811b-41682ce65dc0","order_by":3,"name":"Angela Vogts","email":"","orcid":"https://orcid.org/0000-0003-4261-3123","institution":"Leibniz Institute for Baltic Sea Research","correspondingAuthor":false,"prefix":"","firstName":"Angela","middleName":"","lastName":"Vogts","suffix":""},{"id":603348787,"identity":"36b85d1b-9432-429d-8d0e-9edacbecf64e","order_by":4,"name":"Christian Furbo Reeder","email":"","orcid":"","institution":"Linnaeus University","correspondingAuthor":false,"prefix":"","firstName":"Christian","middleName":"Furbo","lastName":"Reeder","suffix":""},{"id":603348788,"identity":"808c08b0-15a4-41a8-9227-277627853d9a","order_by":5,"name":"Elena Cerdán-García","email":"","orcid":"https://orcid.org/0000-0002-0409-7722","institution":"ICM-CSIC","correspondingAuthor":false,"prefix":"","firstName":"Elena","middleName":"","lastName":"Cerdán-García","suffix":""},{"id":603348789,"identity":"c815e808-4b54-423e-859b-10a0e414b3e9","order_by":6,"name":"Ruth Hawley","email":"","orcid":"","institution":"University of Southampton","correspondingAuthor":false,"prefix":"","firstName":"Ruth","middleName":"","lastName":"Hawley","suffix":""},{"id":603348790,"identity":"fe1b92c5-933a-4285-ae7c-13691cd84c9d","order_by":7,"name":"Oliver Flanagan","email":"","orcid":"","institution":"University of Southampton","correspondingAuthor":false,"prefix":"","firstName":"Oliver","middleName":"","lastName":"Flanagan","suffix":""},{"id":603348791,"identity":"fae1cd8c-3980-4e07-8b47-602a5c48d15e","order_by":8,"name":"Maeve C Lohan","email":"","orcid":"https://orcid.org/0000-0002-5340-3108","institution":"National Oceanography Centre","correspondingAuthor":false,"prefix":"","firstName":"Maeve","middleName":"C","lastName":"Lohan","suffix":""},{"id":603348792,"identity":"cefd1757-3ca7-4e6d-9e6a-7e6155011210","order_by":9,"name":"Joanne Hopkins","email":"","orcid":"https://orcid.org/0000-0003-1504-3671","institution":"National Oceanography Centre","correspondingAuthor":false,"prefix":"","firstName":"Joanne","middleName":"","lastName":"Hopkins","suffix":""},{"id":603348793,"identity":"da496388-8c8a-493c-a20a-7eba0a27e7dc","order_by":10,"name":"Claire Mahaffey","email":"","orcid":"https://orcid.org/0000-0002-4215-7271","institution":"University of Liverpool","correspondingAuthor":false,"prefix":"","firstName":"Claire","middleName":"","lastName":"Mahaffey","suffix":""}],"badges":[],"createdAt":"2026-03-06 13:26:32","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9051103/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9051103/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108807937,"identity":"d557c299-a84e-456b-9530-1a6703aecce2","added_by":"auto","created_at":"2026-05-08 15:37:37","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":424951,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSampling campaign map. \u003c/strong\u003eMap showing stations sampled for bulk N\u003csub\u003e2\u003c/sub\u003e fixation rates, single-cell N\u003csub\u003e2\u003c/sub\u003e fixation rates and DNA (bulk/single/DNA), bulk N\u003csub\u003e2\u003c/sub\u003e fixation rates and DNA (bulk/DNA) or only DNA, during the DY167 N-Arc cruise.\u003cbr\u003e\n\u003c/p\u003e","description":"","filename":"Fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/63132d78ff93ec3f2d75bc3f.jpg"},{"id":108807927,"identity":"c4dde1ef-9342-4603-bf14-eae6404c40c7","added_by":"auto","created_at":"2026-05-08 15:36:49","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1085787,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eParticle-associated N\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e fixation rates. \u003c/strong\u003e(a) Single-cell N\u003csub\u003e2\u003c/sub\u003e fixation rates (boxplots) overlaid by cell diameter (circles) in the suspended (susp) and fast sinking (fs) marine snow catcher particle fractions, faceted by water mass (AW = Atlantic Water, wAW = warm Atlantic Water, ArW = Arctic Water). Significant differences calculated by Kruskal-Wallis tests are depicted with asterisks (**p \u0026lt; 0.01). The total number of cells analysed was 729 (Table S3) but only rates up to 10 fmol N cell\u003csup\u003e-1\u003c/sup\u003e d\u003csup\u003e-1\u003c/sup\u003e (n = 672) are shown in this plot. Only cells with diameters \u0026lt;2 µm -representing 85% of the cells analysed- are shown in this figure. (b) examples of immunofluorescence microscopy images showing NifH positive cells in pink and DAPI positive cells in green (left) mapped to \u003csup\u003e15\u003c/sup\u003eN-enriched cells analysed by nanoSIMS (right, pointed by arrows) mapped to (right). White cells are the overlap of both NifH and DAPI positive cells, i.e. diazotroph cells actively fixing N\u003csub\u003e2\u003c/sub\u003e (pointed by arrows).\u003c/p\u003e","description":"","filename":"Fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/0928e61a705582609e33747e.jpg"},{"id":108807890,"identity":"ab3b9bf8-eed0-4bc9-bbbd-bf434f9c08f8","added_by":"auto","created_at":"2026-05-08 15:36:27","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":283539,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFree-living and particle-associated diazotroph community composition. \u003c/strong\u003eFree-living (0.2 µm filters) and particle-associated (3 µm filters) \u003cem\u003enifH\u003c/em\u003e gene relative abundance of the top 10 amplicon sequence variants (ASVs) annotated at the class and genus levels, faceted by water mass (AW = Atlantic Water, wAW = warm Atlantic Water, ArW = Arctic Water; see Fig. 1). The full list of annotated ASVs is displayed on Table S3.\u003c/p\u003e","description":"","filename":"Fig3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/a79510eea153c47c51e8c394.jpg"},{"id":108809249,"identity":"766b18c2-e03c-4575-9c36-4919fa46b84a","added_by":"auto","created_at":"2026-05-08 15:51:13","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":143946,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComparison of particle-associated diazotrophs across ocean basins.\u003c/strong\u003e Contribution of particle-associated NCD \u003cem\u003enifH\u003c/em\u003e genes to the total particle-associated diazotroph \u003cem\u003enifH\u003c/em\u003e gene pool in MSC deployments in the subtropical North Pacific \u003csup\u003e29\u003c/sup\u003e and North Atlantic waters \u003csup\u003e53\u003c/sup\u003e, as compared to this study (Arctic).\u003c/p\u003e","description":"","filename":"Fig4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/d356a204b139523c59d677ec.jpg"},{"id":108818374,"identity":"5b9e4039-7a04-4c87-b724-ab9e2b807058","added_by":"auto","created_at":"2026-05-08 16:33:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2326227,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/b4703f78-ca30-4a5b-9267-a7226dd055c0.pdf"},{"id":108809074,"identity":"ed5d0ec6-f687-48b0-9c57-127555de2475","added_by":"auto","created_at":"2026-05-08 15:49:23","extension":"jpg","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":86974,"visible":true,"origin":"","legend":"Fig S4","description":"","filename":"FigS4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/9d9e16e89a2fb92f69f35880.jpg"},{"id":108809228,"identity":"fd2599fe-fe2a-414b-8c63-7ad7d7ffce0b","added_by":"auto","created_at":"2026-05-08 15:51:05","extension":"jpg","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":388915,"visible":true,"origin":"","legend":"Fig S1","description":"","filename":"FigS1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/0872d6ff0c1f5a8897e202e2.jpg"},{"id":108807941,"identity":"59e3b72f-ec3e-402e-bd23-9f0e63453dc9","added_by":"auto","created_at":"2026-05-08 15:37:52","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":84808,"visible":true,"origin":"","legend":"Table S5","description":"","filename":"TableS5.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/506f0da1b02f4e96279374bb.xlsx"},{"id":108807942,"identity":"2291ea7e-3b5a-4b1d-b6c5-a2fa119bc04b","added_by":"auto","created_at":"2026-05-08 15:37:53","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":1485572,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"NARCMSCSI.docx","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/9a3593f5703754853762de85.docx"},{"id":108807932,"identity":"604c3a6b-419c-4f6e-baa0-1af4851b1115","added_by":"auto","created_at":"2026-05-08 15:37:08","extension":"jpg","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":122495,"visible":true,"origin":"","legend":"Fig S7","description":"","filename":"FigS7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/573ea5c046da2cd59cb3b18b.jpg"},{"id":108809492,"identity":"d2f8857e-bd34-4a2f-a2f5-a08992a95793","added_by":"auto","created_at":"2026-05-08 15:53:11","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":56758,"visible":true,"origin":"","legend":"Table S3","description":"","filename":"TableS3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/8cdaee7b9cf4c3f5496aa774.xlsx"},{"id":108809234,"identity":"3ae0dd8f-d283-426d-9848-3789a2fc532a","added_by":"auto","created_at":"2026-05-08 15:51:08","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":9259,"visible":true,"origin":"","legend":"Table S2","description":"","filename":"TableS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/2d31761624ea390d44a80814.xlsx"},{"id":108809495,"identity":"b2e5f270-5fe5-498a-85d5-e64cab1796ef","added_by":"auto","created_at":"2026-05-08 15:53:11","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":144038,"visible":true,"origin":"","legend":"Fig S2","description":"","filename":"FigS2.png","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/ff6cab6ec17d947202f20c3d.png"},{"id":108809276,"identity":"35e4047c-dfe3-4d50-a75e-dce4f9785893","added_by":"auto","created_at":"2026-05-08 15:51:35","extension":"xlsx","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":9395,"visible":true,"origin":"","legend":"Table S1","description":"","filename":"TableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/8f23815be30dcbeb3a4ac6ae.xlsx"},{"id":108807938,"identity":"fa4dd60e-22c3-49b4-b3ee-9b58af064127","added_by":"auto","created_at":"2026-05-08 15:37:37","extension":"jpg","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":335927,"visible":true,"origin":"","legend":"Fig S5","description":"","filename":"FigS5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/c32cfb7a8ac7287732746404.jpg"},{"id":108814276,"identity":"998232c0-1362-4439-8217-9eed8e1f042c","added_by":"auto","created_at":"2026-05-08 16:17:53","extension":"jpg","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":143683,"visible":true,"origin":"","legend":"Fig S3","description":"","filename":"FigS3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/037cda583e1bb2b1d8b415ca.jpg"},{"id":108809369,"identity":"09c9acf1-2de8-44e2-821f-15fee48d300f","added_by":"auto","created_at":"2026-05-08 15:52:49","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":1412144,"visible":true,"origin":"","legend":"Fig S6","description":"","filename":"FigS6.png","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/0144bd0de761c5b45f7bf740.png"},{"id":108809215,"identity":"9a712f68-4e31-4cb8-b558-08cbb52d9af3","added_by":"auto","created_at":"2026-05-08 15:50:59","extension":"xlsx","order_by":13,"title":"","display":"","copyAsset":false,"role":"supplement","size":27765,"visible":true,"origin":"","legend":"Table S4","description":"","filename":"TableS4.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9051103/v1/4a6850bc7a0eea166332d662.xlsx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Particle-associated diazotrophs drive nitrogen fixation in Arctic subsurface waters","fulltext":[{"header":"Introduction","content":"\u003cp\u003eDinitrogen (N\u003csub\u003e2\u003c/sub\u003e) fixation by diazotrophic microorganisms constitutes the main source of reactive nitrogen in the ocean, where it sustains primary productivity and carbon export \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Traditionally, N\u003csub\u003e2\u003c/sub\u003e fixation was considered largely restricted to oligotrophic, warm tropical and subtropical waters where nutrient limitation favours diazotroph activity. However, several studies have expanded this paradigm, documenting significant N\u003csub\u003e2\u003c/sub\u003e fixation rates at higher latitudes, including temperate and polar regions \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. The Arctic has warmed three to four times faster than the global average since 1979 due to Arctic \u0026lsquo;amplification\u0026rsquo; (ice albedo feedback and heat transport \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e) and \u0026lsquo;atlantification\u0026rsquo; (warm Atlantic origin water spreading further into the Arctic basin subsurface \u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e). Continued warming leads to glacier and sea-ice melting, which releases nutrients and iron \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e advancing the onset and increasing the magnitude of phytoplankton blooms \u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. The extent of ice-free waters increases over the year, enhancing nitrogen depletion by phytoplankton and eventually leading to decreased primary productivity and carbon export \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe progressively earlier seasonal depletion of nitrogen in the Arctic may expand the niche for diazotrophs \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. A preliminary study estimated N\u003csub\u003e2\u003c/sub\u003e fixation supports 17% of the primary productivity in the region \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e, and current ecosystem models predict an increase of diazotrophs in Arctic latitudes towards the end of the 21st century \u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. More recent studies have documented active N\u003csub\u003e2\u003c/sub\u003e fixation across diverse Arctic Ocean regions, including the Chukchi Sea, Central Arctic, Eurasian marginal ice zones, and Fram Strait \u003csup\u003e\u003cspan additionalcitationids=\"CR15 CR16\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. The only specific N\u003csub\u003e2\u003c/sub\u003e fixation measurements available in the region show that the nitroplast (formerly UCYN-A) of the eukaryotic alga \u003cem\u003eBraarudosphaera bigelowii\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e actively fixes N\u003csub\u003e2\u003c/sub\u003e in the Pacific sector of the Arctic at rates similar to those of (sub)tropical waters \u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Despite the cosmopolitan distribution of this nitroplast across latitudes globally \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e, non-cyanobacterial diazotrophs (NCDs) overwhelmingly dominate \u003cem\u003enifH\u003c/em\u003e gene sequences and expression in the Arctic \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan additionalcitationids=\"CR21 CR22\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Contrary to diazotrophic cyanobacteria, NCDs do not have photosynthetic machinery \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e and based on their genomes, are predicted to rely on other external sources of carbon and energy \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Organic particles are rich in carbon and can harbour low oxygen microzones, making them ideal niches for NCDs to fix N\u003csub\u003e2\u003c/sub\u003e \u003csup\u003e26,27\u003c/sup\u003e. Recent studies have demonstrated that gammaproteobacterial NCDs fix N₂ in association with particles in the subtropical North Pacific \u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, but the recurrence and N\u003csub\u003e2\u003c/sub\u003e fixation potential of these associations in the Arctic is unknown.\u003c/p\u003e \u003cp\u003eArctic waters with seasonal ice coverage export almost twice as many particles as ice-free zones \u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e, and combined with the Arctic's extremely broad shelf areas, supply coastal waters with substantial volumes of sediment-laden particles rich in potentially bioavailable iron and carbon \u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e - key resources enabling NCDs to thrive and perform N\u003csub\u003e2\u003c/sub\u003e fixation \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Here, we aimed to evaluate the N\u003csub\u003e2\u003c/sub\u003e fixation potential and identity of particle-associated NCDs in the Arctic Ocean. Combining single-cell spectrometry, immunofluorescence and molecular methods, we find that particle-associated N₂ fixation in the subsurface waters of the Barents Sea is predominantly driven by NCDs. By comparing our Arctic findings with subtropical studies, we demonstrate that particle-associated NCDs stand out as the dominant diazotrophs in Arctic waters, unlike the more diverse diazotroph communities reported from warmer regions. As a region at the forefront of Arctic oligotrophication \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e, the magnitude and controls of particle-associated N\u003csub\u003e2\u003c/sub\u003e fixation in the Barents Sea provide hints on the future of reactive nitrogen supply in this globally climate regulating region.\u003c/p\u003e"},{"header":"Results and Discussion","content":"\u003cp\u003eWe sampled at fourteen stations aiming at covering the Atlantic and Arctic influenced end-members of the region \u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e, including Atlantic Water (AW), warm Atlantic Water (wAW) and Arctic Water (ArW) water masses (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Suspended, slow and fast sinking particles (\u0026lsquo;susp\u0026rsquo;, \u0026lsquo;ss\u0026rsquo; and \u0026lsquo;fs\u0026rsquo;, respectively) were collected using a marine snow catcher (MSC) deployed 10 m below the deep chlorophyll maximum \u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e, ranging from 22 to 66 m during our cruise (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). MSC samples were size-fractionated, allowing to distinguish particle-associated (\u0026gt;\u0026thinsp;3 \u0026micro;m) from free-living cells (\u0026lt;\u0026thinsp;3 \u0026micro;m and \u0026gt;\u0026thinsp;0.2 \u0026micro;m) in both DNA and single-cell N\u003csub\u003e2\u003c/sub\u003e fixation measurements (see Methods and Supplementary Information). Bulk measurements across individual MSC fractions revealed that the susp fraction had the highest N₂ fixation rates (up to 1.62 nmol N l\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e d\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e), followed by the fs fraction (up to 0.46 nmol N l\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e d\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e), and the ss fraction (up to 0.22 nmol N l\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e d\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) (Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e, Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). These rates are five to ten times lower than previous measurements in surface waters of the Bering, Chukchi and Beaufort Seas (compiled in ref. \u003csup\u003e10\u003c/sup\u003e), but in the range of surface water bulk N\u003csub\u003e2\u003c/sub\u003e fixation rates during our cruise in the Barents Sea (Mahaffey et al., in prep.). Based on the bulk N\u003csub\u003e2\u003c/sub\u003e fixation rates observed, a subset of five stations (N00, N06, N08, N10 and N13; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) was selected for single-cell analysis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eMeasuring N\u003csub\u003e2\u003c/sub\u003e fixation rates of individual NCDs taxa usually requires combining oligonucleotide probing (catalysed reporter deposition-fluorescence in situ hybridisation or CARD-FISH) with nanoscale secondary ion mass spectrometry (nanoSIMS; e.g., \u003csup\u003e28,29\u003c/sup\u003e). While this approach resolves relatively broad taxonomic groups, it does not specifically capture the broader NCDs assemblage, which spans multiple taxonomic lineages and is widespread across the global oceans \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Here we applied a nitrogenase iron protein (NifH) antibody staining approach able to target virtually all diazotrophs independently of their taxonomy \u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. Based on NifH immunofluorescence counts on 3 and 0.2 \u0026micro;m filters, we found that particle-associated diazotrophs (3 \u0026micro;m filters) comprised 0.25\u0026ndash;9.26% of the total microbial community, with the highest proportions observed at stations in wAW waters (Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e), situated near the Polar Front \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. NifH-immunofluorescence-positive cells in the ss fraction were exclusively detected in wAW waters (Fig. \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e), corresponding to the lowest bulk N₂ fixation rates (Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e); consequently, this fraction was excluded from subsequent nanoSIMS analyses. Mapping immunolabeled cells in 3 \u0026micro;m filters for nanoSIMS, we measured particle-associated N\u003csub\u003e2\u003c/sub\u003e fixation rates on 729 individual cells. Most of the rates ranged between 0 and 10 fmol N cell\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e d\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, although values up to 1192 fmol N cell\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e d\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e were observed (Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e). Averaging particle-associated N\u003csub\u003e2\u003c/sub\u003e fixation rates per water mass showed an increase from AW (1.83\u0026thinsp;\u0026plusmn;\u0026thinsp;5.95 fmol N cell\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e d\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, n\u0026thinsp;=\u0026thinsp;130), through wAW (2.33\u0026thinsp;\u0026plusmn;\u0026thinsp;5.77 fmol N cell\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e d\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, n\u0026thinsp;=\u0026thinsp;319), to ArW in the inner Barents Sea (23.2\u0026thinsp;\u0026plusmn;\u0026thinsp;111 fmol N cell\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e d\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, n\u0026thinsp;=\u0026thinsp;280), being significantly higher in fs than in susp particles (Kruskal-Wallis test, **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These rates are higher than those measured on two different \u003cem\u003eB. bigelowii\u003c/em\u003e nitroplast lineages in the Bering and Chukchi Seas (7.6 and 13.0 fmol N cell\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e d\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, respectively \u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e), but similar to NCDs\u0026rsquo; single-cell N\u003csub\u003e2\u003c/sub\u003e fixation rates measured in the North Pacific Ocean (67\u0026ndash;121 fmol N cell\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e d\u003csup\u003e\u0026minus;\u0026thinsp;1 29\u003c/sup\u003e). This suggests that particle-associated NCDs exhibit consistent N\u003csub\u003e2\u003c/sub\u003e fixation performance across latitudes.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAlthough the immunolabeling approach used here detects virtually all diazotrophs synthesising the NifH protein, it cannot distinguish between taxa, complicating determination of whether particle-associated diazotrophs were exclusively NCDs or also included diazotrophic cyanobacteria, as the latter are often observed embedded in particles in low latitude waters \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. In this case, cell size can be helpful to discriminate cyanobacteria from NCDs. Although we measured a few outlier cells with diameters of up to 5.24 \u0026micro;m (Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e), 85% of the cells analysed had diameters ranging from 0.4 to 2 \u0026micro;m (Fig. \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e). Cells within this size range could potentially correspond to the \u003cem\u003eB. bigelowii\u003c/em\u003e nitroplast, that is known to fix N\u003csub\u003e2\u003c/sub\u003e in the Arctic \u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e and was also identified in surface waters during our cruise (Mahaffey et al., in prep.). However, we can confidently exclude the nitroplast as a contributor to N\u003csub\u003e2\u003c/sub\u003e fixation in subsurface waters as no corresponding \u003cem\u003enifH\u003c/em\u003e gene sequences were detected in our samples (Table \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e). The sporadic higher rates measured in cells\u0026thinsp;\u0026gt;\u0026thinsp;2 \u0026micro;m could be attributed to the unicellular cyanobacterium \u003cem\u003eCyanothece\u003c/em\u003e (Amplicon Sequence Variant (ASV)51, Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e), which has been previously observed in Arctic seawater and sea ice brine \u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. Given that most of the active N\u003csub\u003e2\u003c/sub\u003e-fixing cells detected in our study were \u0026lt;\u0026thinsp;2 \u0026micro;m (Fig. \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e) and diazotrophic cyanobacteria (typically\u0026thinsp;\u0026gt;\u0026thinsp;4 \u0026micro;m) were only sparsely detected in \u003cem\u003enifH\u003c/em\u003e gene sequences (Table \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e), we can confidently conclude that particle-associated N\u003csub\u003e2\u003c/sub\u003e fixation in our study was chiefly driven by NCDs.\u003c/p\u003e \u003cp\u003eThe top ten ASVs belonged to putative motile genera (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), which would support a particle-associated lifestyle \u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. \u003cem\u003eAzotobacter\u003c/em\u003e was mostly detected in the free-living fraction (i.e., 0.2 \u0026micro;m filters) across all water masses sampled (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). This genus has been previously reported from Arctic melt ponds, water under ice and algal aggregates \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, and is able to use carbohydrates and organic acids as sources of carbon. \u003cem\u003eDesulfuromonas\u003c/em\u003e, which are sulphate-reducers and typical of oxygen deficient waters \u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e, were particularly abundant in wAW waters where the MSC deployment depth (i.e., 10 m below the deep chlorophyll maximum) coincided roughly with the oxycline (Figs. S1, S5). \u003cem\u003eBradyrhizobium\u003c/em\u003e was more evenly detected across water masses, while \u003cem\u003eMethylococcus\u003c/em\u003e was clearly prevalent in AW (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Fig. \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003e). Active N\u003csub\u003e2\u003c/sub\u003e-fixing Bradyrhizobia species have been found in driftwood around Svalbard \u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e, while \u003cem\u003eMethylococcus\u003c/em\u003e are typically found in sediment and wetland ecosystems \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e,\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. \u003cem\u003eXanthobacter\u003c/em\u003e is a soil methylmercury degrading bacterium, found even in deep bathypelagic waters of the Arctic \u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. This suggests that the samples collected included particles of terrestrial and/or glacial origin, consistent with a recent study \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e and the well-known role of ice-laden sediments to transport microbial communities across the Arctic \u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eParticles in the Arctic Ocean originate from diverse sources, including phytoplankton, macroalgae, and glacial sediment runoff \u003csup\u003e\u003cspan additionalcitationids=\"CR49\" citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. However, phytoplankton derived particulate organic carbon export is usually minimum in the boreal summer \u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e, when our cruise took place. Over the summer months, as Arctic phytoplankton blooms decline and are exported from surface waters, particles of terrestrial and/or glacial origin likely dominate the water column, as suggested by the diazotroph community composition in our samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), which resembles sediment-associated assemblages. Alternatively, the scarcity of typical pelagic diazotroph communities in our data may reflect respiration of phytoplankton-derived particles by our sampling depth (10 m below the deep chlorophyll maximum, 22\u0026ndash;66 m; Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e), consistent with particulate organic carbon attenuation horizons as shallow as 30\u0026ndash;60 m in the Barents Sea \u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eComparing this study with studies in the North Pacific \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e and North Atlantic \u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e (which employed the same MSC sampling strategy at 10 m below DCM; see Supplementary Information) reveals that tparticle-associated diazotroph communities in the Arctic are remarkably different from those of subtropical regions (Fig. \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003eA), showcasing a high degree of endemicity (Fig. \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003eB). We find that while gamma- and betaproteobacteria were present in particle-associated diazotroph communities across the North Pacific, North Atlantic and Arctic sites, alphaproteobacterial and Desulfuromonadia were not detected in the North Pacific (Fig. \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003eC). Interestingly, cyanobacteria dominated the particle-associated diazotroph community in subsurface waters of both the North Pacific and North Atlantic, whereas they were virtually absent in the Arctic, where NCDs comprised nearly 100% of the particle-associated diazotroph community (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). This aligns with recent modelling efforts showing that particle-associated NCDs are best suited for fixing N\u003csub\u003e2\u003c/sub\u003e at low temperatures as compared to cyanobacterial diazotrophs \u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. The anticipated shift to an ice-free Barents Sea will extend the oligotrophic summer period, characterised by low algal biomass and consequently reduced vertical particulate organic carbon flux \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e, pointing towards terrestrial and glacial runoff particle-associated NCDs as potential important contributors to nitrogen availability in the future.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003cdiv id=\"Sec4\" class=\"Section3\"\u003e \u003ch2\u003eParticle sampling\u003c/h2\u003e \u003cp\u003eThe DY167 N-Arc GEOTRACES GApr19 cruise took place from 9 July to 13 August 2023 onboard the RRS \u003cem\u003eDiscovery.\u003c/em\u003e Hydrographic and biogeochemical data acquired included CTD casts, macro- and micronutrients (see Supplementary Information for detailed methods). A marine snow catcher (MSC; OSIL, Havant, Hampshire, UK) was deployed at 10 m below the deep chlorophyll maximum at fourteen stations (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). After 2 h of settling on deck, the suspended (susp), slow sinking (ss) and fast sinking (fs) particle fractions were subsequently sampled. The susp fraction was sampled in nine 4000 ml polycarbonate bottles, where three bottles were used for N\u003csub\u003e2\u003c/sub\u003e fixation rate measurements (see below), three bottles for DNA sampling, and three bottles for particulate matter natural \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003eN atom % enrichment measurements. The ss and fs fractions were sampled in triplicate 500 ml polycarbonate bottles and 40 ml glass flasks, respectively, for N\u003csub\u003e2\u003c/sub\u003e fixation measurements. The content of the other triplicate sets of 500 ml polycarbonate bottles and 100 ml glass flasks was size-fractionated by filtration through 3 and 0.2 \u0026micro;m filters and used for DNA extraction and \u003cem\u003enifH\u003c/em\u003e gene amplification (Supplementary Information).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e\n\u003ch3\u003eBulk N fixation rates\u003c/h3\u003e\n\u003cp\u003eEach MSC fraction was incubated in triplicate with 10% v/v \u003csup\u003e15\u003c/sup\u003eN\u003csub\u003e2\u003c/sub\u003e-enriched filtered seawater for 48 h in a controlled temperature room adjusted to the temperature of each sampling depth at each station (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Enriched seawater was prepared from 2300 ml seawater, filtered through a 0.2 \u0026micro;m pore size Sartobran PTFE cartridge (Sartorius, G\u0026ouml;ttingen, Germany), injected with 25 ml 98 atom % \u003csup\u003e15\u003c/sup\u003eN\u003csub\u003e2\u003c/sub\u003e gas (Cambridge Isotope Laboratories, Tewksbury, MA, USA), and stirred on a magnet stirred plate overnight. After incubation, 4000 ml, 500 ml, and 30 ml from the susp, ss and fs fractions, respectively, were filtered onto pre-combusted (450\u0026ordm;C, 6 h) GFF filters (Whatman, Little Chalfont, Buckinghamshire, UK) and stored at -20\u0026ordm;C. Filters were dried at 60\u0026ordm;C for 24 h, pelletised and analysed on an INTEGRA2 Sercon elemental analyser coupled to an isotope ratio mass spectrometer (Crewe, Cheshire, UK) for particulate \u003csup\u003e15\u003c/sup\u003eN/\u003csup\u003e14\u003c/sup\u003eN isotope ratios, particulate carbon (PC) and particulate nitrogen (PN) concentrations. Subsamples from the incubated susp fraction were taken on 12 ml Exetainer\u0026reg; tubes (Labco, Exeter, UK) and the \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003eN atom % enrichment of seawater measured using a membrane inlet mass spectrometer onboard (MIMS, Hiden HPR40, Warrington, UK). Bulk N\u003csub\u003e2\u003c/sub\u003e fixation rates were calculated according to White et al. \u003csup\u003e55\u003c/sup\u003e, applying volume corrections for the different MSC fractions as recommended by Riley et al. \u003csup\u003e35\u003c/sup\u003e.\u003c/p\u003e\n\u003ch3\u003eNifH antibody staining, imaging and cell counts\u003c/h3\u003e\n\u003cp\u003eParticle-associated single-cell N\u003csub\u003e2\u003c/sub\u003e fixation rates were measured from 500 ml, 50 ml, and 10 ml subsamples taken from each MSC fraction, filtered through 3 \u0026micro;m Isopore polycarbonate filters (Millipore, Tullagreen, Carrightwohill, Ireland). Filters were fixed with 1.6% electron microscopy grade paraformaldehyde (Thermo Fisher, Ward Hill, MA, USA) for 1 h at room temperature, air-dried and stored at -80\u0026ordm;C until further analysis. Samples with significant bulk N\u003csub\u003e2\u003c/sub\u003e fixation rates were selected for nitrogenase immunolabeling \u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e and nanoSIMS analyses (stations N00, N06, N08, N10 and N13; see below).\u003c/p\u003e \u003cp\u003eActive diazotroph cells were identified and counted by nitrogenase immunolabeling and their contribution to the total bacterial community assessed from comparative DAPI staining. Briefly, cells in filter portions were permeabilised with 0.5% (v/v) dimethyl sulfoxide (DMSO; Sigma-Aldrich, Saint Louis, MO, USA) and washed three time for 5 min in phosphate-buffered saline containing 0.1% Triton X-100 (PBST, Dutscher, Brumath, France), supplemented with 5% (w/v) bovine serum albumin (BSA). Samples were then incubated for 1 h at room temperature in the dark with a polyclonal anti-NifH primary antibody (Agrisera, cat. AS01021A, V\u0026auml;nn\u0026auml;s, Sweden) diluted 1:166 (1.2 \u0026micro;l antibody in 198.8 \u0026micro;l PBS). After three rounds of 5 min washes in PBST/BSA, samples were incubated for 45 min at room temperature with a goat anti-Chicken IgY (H\u0026thinsp;+\u0026thinsp;L) secondary antibody conjugated to Alexa Fluor\u0026trade; 594 (Invitrogen, A-11042), diluted 1:181 (1.1 \u0026micro;l antibody in 198.9 \u0026micro;l PBS). Samples were subsequently washed three times for 5 min in PBST/BSA, and dehydrated in 70% ethanol, and subsequently mounted on slides using 4\u0026rsquo;,6-diamidino-2-phenylindole (DAPI, ProLong\u0026trade; glass with DAPI, Thermo Fisher Scientific, Waltham, MA, USA) and stored at -20\u0026deg;C until visualisation. To test the specificity of the antibody, anti-nitrogenase antibody controls were performed (Supplementary Information).\u003c/p\u003e \u003cp\u003eSamples were visualised by epifluorescence microscopy using a Zeiss Axio Imager.Z2m microscope (Carl Zeiss MicroImaging, Barcelona, Spain) equipped with an AxioCam MRm camera, at 630\u0026times; magnification. Images were acquired in black and white using the AxioVision software. A Colibri LED light source was used with the HE-62 multi-filter module, including excitation filters BP 370/40 for DAPI and BP 585/35 for Alexa594, a triple beam splitter (TFT 395/495/605), and emission filters TBP 425/46\u0026thinsp;+\u0026thinsp;527/54\u0026thinsp;+\u0026thinsp;LP615. Exposure times were set to 100 ms for DAPI and 120 ms for the Alexa594-labelled antibody. At least ten fields of view were recorded per filter.\u003c/p\u003e \u003cp\u003eImage processing and cell quantification were performed using the automated image analysis software ACMEtool3 (48). Cells were first identified in the DAPI channel and classified as bacterial based on their size (area 18\u0026ndash;200 pixels), with an additional signal-to-background ratio (SBR\u0026thinsp;\u0026gt;\u0026thinsp;2) constraint applied to 0.2 \u0026micro;m filters. NifH-positive cells were subsequently identified within the DAPI mask by detecting an AF594 fluorescence signal associated with a DAPI-identified cell. Cells were classified as NifH⁺ only when an AF594 signal overlapped a DAPI-labelled cell and met channel-specific detection thresholds applied during image analysis.\u003c/p\u003e \u003cp\u003eIn total, 1,952 fields of view were analysed on 3 \u0026micro;m (particle-associated) filters (95,908 DAPI cells screened, including 3,064 NifH positive cells), and 1,753 fields of view on 0.2 \u0026micro;m (free-living) filters (986,147 DAPI cells screened, including 7,687 NifH positive cells). No NifH positive cells were detected in the ss fraction, therefore, single-cell mass spectrometry analyses were performed only on the susp (susp; n\u0026thinsp;=\u0026thinsp;402 cells) and fs (fs; n\u0026thinsp;=\u0026thinsp;327 cells) fractions.\u003c/p\u003e\n\u003ch3\u003eSingle-cell N fixation rates\u003c/h3\u003e\n\u003cp\u003eSamples displaying the strongest and most spatially coherent NifH signal (i.e., NifH positive) on 3 \u0026micro;m filters were selected for nanoSIMS analyses and re-examined by epifluorescence microscopy using an Olympus BX61 microscope equipped with a monochrome Hamamatsu ORCA camera. NifH immunofluorescence was detected using a custom optical cube assembled from Chroma filter components, consisting of an ET599/13x excitation filter, a T612lpxr long-pass dichroic mirror, and an ET632/28m emission filter (Chroma Technology Corp., Bellows Falls, VT, USA), enabling enhanced isolation of the nitrogenase-associated signal and minimising spectral overlap with Chorophyll \u003cem\u003ea\u003c/em\u003e fluorescence. DAPI fluorescence was detected using a UB-BP filter set (excitation 360/40 nm, dichroic 400 LP, emission 460/50 nm).\u003c/p\u003e \u003cp\u003eSelected regions from the corresponding filters were then subjected to a freeze-transfer procedure to relocate cells and particles onto a solid substrate compatible with high-resolution analyses. Briefly, filter sections were gently rinsed with Milli-Q water, placed upside down onto patterned silicon wafers (1.2 \u0026times; 1.2 cm, 1 \u0026times; 1 mm raster; Pelotec SFG12 Finder Grid substrate, Ted Pella Inc., Redding, CA, USA), rapidly frozen at -80\u0026deg;C, and carefully removed while frozen to promote transfer of attached material onto the wafer surface \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. Following transfer, silicon wafers were re-examined using the same epifluorescence microscope and filter sets to relocate NifH-positive particles. The engraved coordinate grid of the silicon substrate was used to map the position of individual target particles, providing precise spatial references for subsequent NanoSIMS analyses. Prior to NanoSIMS measurements, wafers were coated with a gold layer of approximately 30 nm thickness.\u003c/p\u003e \u003cp\u003eThe \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003eN atom % enrichment of particle-associated diazotroph cells was measured on a nanoSIMS 50L (CAMECA, Gennevilliers, France) at the Leibniz Institute for Baltic Sea Research (IOW, Germany). A 1 pA 16 keV Cesium (Cs\u003csup\u003e+\u003c/sup\u003e) primary beam was scanned on a 512 x 512 pixel raster with a raster area of 15 x 15 \u0026micro;m, and a counting time of 250 \u0026micro;s per pixel. Samples were pre-sputtered with 600 pA Cs\u003csup\u003e+\u003c/sup\u003e current for 2 min in a raster of 30 x 30 \u0026micro;m to remove the gold and surface contaminants and reach the steady state of ion formation. Negative secondary ions \u003csup\u003e12\u003c/sup\u003eC\u003csup\u003e\u0026minus;\u003c/sup\u003e, \u003csup\u003e13\u003c/sup\u003eC\u003csup\u003e\u0026minus;\u003c/sup\u003e, \u003csup\u003e12\u003c/sup\u003eC\u003csup\u003e14\u003c/sup\u003eN\u003csup\u003e\u0026minus;\u003c/sup\u003e, \u003csup\u003e12\u003c/sup\u003eC\u003csup\u003e15\u003c/sup\u003eN\u003csup\u003e\u0026minus;\u003c/sup\u003e and \u003csup\u003e31\u003c/sup\u003eP\u003csup\u003e\u0026minus;\u003c/sup\u003e were detected with electron multiplier detectors, and secondary electrons were simultaneously imaged. Sixty serial quantitative secondary ion mass planes were generated, drift corrected and accumulated to the final image. Mass resolving power was \u0026gt;\u0026thinsp;8000 to resolve isobaric interferences. Data was processed using the Look@nanoSIMS software \u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. Isotope ratio images were generated by dividing the \u003csup\u003e12\u003c/sup\u003eC\u003csup\u003e15\u003c/sup\u003eN\u003csup\u003e\u0026minus;\u003c/sup\u003e ion count by the \u003csup\u003e12\u003c/sup\u003eC\u003csup\u003e14\u003c/sup\u003eN\u003csup\u003e\u0026minus;\u003c/sup\u003e ion count pixel by pixel. Individual diazotroph cells were identified from \u003csup\u003e12\u003c/sup\u003eC\u003csup\u003e\u0026minus;\u003c/sup\u003e and \u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003eP images. These images were used to define regions of interest (ROIs). For each ROI, the \u003csup\u003e15\u003c/sup\u003eN/\u003csup\u003e14\u003c/sup\u003eN ratios were calculated based on the ion counts averaged over the ROIs. Additional ROIs were defined in areas without visible \u0026sup1;⁵N enrichment to estimate internal background ratios, which were centred around natural abundance (~\u0026thinsp;0.00367; \u003csup\u003e15\u003c/sup\u003eN atom %). Cells were considered isotopically enriched only when their \u0026sup1;⁵N/\u0026sup1;⁴N ratios exceeded both natural abundance and the Poisson counting error associated with ion counting statistics. The total number of particles analysed was 729. Single-cell N\u003csub\u003e2\u003c/sub\u003e fixation rates were calculated according to Furbo Reeder et al. \u003csup\u003e29\u003c/sup\u003e.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eData analysis, statistics, and plots\u003c/h2\u003e \u003cp\u003eAll data and statistical analyses and plots were done in RStudio v2024.04.2 \u003csup\u003e57\u003c/sup\u003e. All plots were done using the \u003cem\u003eggplot2\u003c/em\u003e \u003csup\u003e58\u003c/sup\u003e package, except CTD plots that were done using the \u003cem\u003eoce\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e R package. N\u003csub\u003e2\u003c/sub\u003e fixation values were checked for normality using a Shapiro-Wilk test and significant differences between samples tested with Wilcoxon test using the \u003cem\u003edplyr\u003c/em\u003e package \u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e. Amplicon data heatmaps plotted with the \u003cem\u003eampvis2\u003c/em\u003e R package \u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e. Redundancy analysis biplots were calculated and plotted with the \u003cem\u003ephloyseq\u003c/em\u003e R package \u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. Missing values in the environmental metadata were imputed prior to redundancy analysis to avoid excessive sample loss. Remaining missing entries in the mixed-type metadata set were imputed using the random forest-based algorithm implemented in the \u003cem\u003emissForest\u003c/em\u003e package in R \u003csup\u003e63\u003c/sup\u003e, with a fixed random seed (123) to ensure reproducibility. The imputed metadata matrix (Table \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e) was subsequently used as the sample-data component when constructing the \u003cem\u003ephyloseq\u003c/em\u003e object for ordination analyses.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eData availability\u003c/h3\u003e\n\u003cp\u003eSource data is provided with this paper. Stainless steel and titanium cast data are available at DOI: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5285/48e39c93-2405-99f3-e063-7086abc003fa\u003c/span\u003e\u003cspan address=\"10.5285/48e39c93-2405-99f3-e063-7086abc003fa\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e and DOI: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5285/48c1d94c-1647-8717-e063-7086abc0f520\u003c/span\u003e\u003cspan address=\"10.5285/48c1d94c-1647-8717-e063-7086abc0f520\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, respectively. \u003cem\u003enifH\u003c/em\u003e gene sequences are available at NCBI under project number PRJNA1412537.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eCompeting Interests\u003c/h2\u003e \u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor Contributions\u003c/h2\u003e \u003cp\u003eMB designed the experiments. MB and AC performed sampling in the Arctic. AC performed immunolabeling assays. AC and MS processed immunolabeling and DAPI staining images at ICM-CSIC (Barcelona, Spain). AC and AV performed nanoSIMS analyses at IOW (Warnem\u0026uuml;nde, Germany). CFR and EC-G processed MSC \u003cem\u003enifH\u003c/em\u003e data from subtropical cruises for comparison purposes. AC processed Arctic \u003cem\u003enifH\u003c/em\u003e sequencing data. CM, MCL and JEH secured funding for the research cruise. RH, OF, MCL, JEH and CM provided contextual biogeochemical and hydrographic Arctic cruise data. CFR and EC-G contributed to ocean basin intercomparison analyses. MB and AC analysed data and wrote the manuscript with inputs from all co-authors.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eThe N-ARC project was funded under Natural Environment Research Council (NERC) Discovery grant Nitrogen fixation in the Arctic Ocean, N-ARC (grant numbers NE/T001240/1 and NE/T000570/1) held at the University of Liverpool (CM), and the National Oceanography Centre Liverpool (JEH). Sample analyses were supported by the projects ANITA \u0026ldquo;Architecture and dyNamics of marIne parTicle colonizAtion\u0026rdquo; funded by the Excellence Initiative of Aix-Marseille University A*MIDEX, the project PANDA \u0026ldquo;PArticles as Niches for non-cyanobacterial DiAzotrophs\u0026rdquo; funded by INSU-EC2CO, and the project MANIOC \u0026ldquo;iMpact of particle microbial colonisation on Nitrogen cyclIng in the OCean\" granted by Aix-Marseille Universit\u0026eacute; Institut Oc\u0026eacute;ans to MB. MB was funded by the BIOPOLE National Capability Multicentre Round 2 funding from the Natural Environment Research Council (grant no. NE/W004933/1). Horizon MSCA grant 101150634 to CFR. The NanoSIMS at the Leibniz Institute for Baltic Sea Research (IOW) was funded by the German Federal Ministry of Education and Research (BMBF), grant identifier 03F0626A. The authors warmly thank the captain and crew of the RRS \u003cem\u003eDiscovery\u003c/em\u003e, as well as NMF staff for assistance with overboard operations. The authors are indebted to L. Wrightson and B. Fisher for MIMS, and L. Norman for nutrient analyses onboard, respectively. We would also like to thank K. Turk-Kubo and J. Magasin for \u003cem\u003enifH\u003c/em\u003e gene annotation, I. Forn for microscopy analyses, and the IBDM platform in Marseille for filter laser marking. Annett Gr\u0026uuml;ttm\u0026uuml;ller is acknowledged for nanoSIMS routine operation. Finally, we thank S. Giering for providing the MSC devices used in this study.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eZehr JP, Capone DG (2020) Changing perspectives in marine nitrogen fixation. Science 368:eaay9514\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBonnet S et al (2023) Diazotrophs are overlooked contributors to carbon and nitrogen export to the deep ocean. ISME J 17:47\u0026ndash;58\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou W, Leung LR, Lu J (2024) Steady threefold Arctic amplification of externally forced warming masked by natural variability. Nat Geosci 17:508\u0026ndash;515\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRantanen M et al (2022) The Arctic has warmed nearly four times faster than the globe since 1979. Commun Earth Environ 3:1\u0026ndash;10\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePolyakov IV et al (2017) Greater role for Atlantic inflows on sea-ice loss in the Eurasian Basin of the Arctic Ocean. Science 356:285\u0026ndash;291\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang S, Bailey D, Lindsay K, Moore JK, Holland M (2014) Impact of sea ice on the marine iron cycle and phytoplankton productivity. Biogeosciences 11:4713\u0026ndash;4731\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThiele S et al (2023) Seasonality of the bacterial and archaeal community composition of the Northern Barents Sea. Front Microbiol 14:1213718\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLewis KM, van Dijken GL, Arrigo K (2020) R. Changes in phytoplankton concentration now drive increased Arctic Ocean primary production. Science 369:198\u0026ndash;202\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKohlbach D et al (2023) Earlier sea-ice melt extends the oligotrophic summer period in the Barents Sea with low algal biomass and associated low vertical flux. Prog Oceanogr 213:103018\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evon Friesen LW, Riemann L (2020) Nitrogen Fixation in a Changing Arctic Ocean: An Overlooked Source of Nitrogen? Front Microbiol 11:596426\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArrigo KR, van Dijken GL (2015) Continued increases in Arctic Ocean primary production. Prog Oceanogr 136:60\u0026ndash;70\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSipler RE et al (2017) Preliminary estimates of the contribution of Arctic nitrogen fixation to the global nitrogen budget. Limnol Oceanogr Lett 2:159\u0026ndash;166\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHenson SA, Cael BB, Allen SR, Dutkiewicz S (2021) Future phytoplankton diversity in a changing climate. Nat Commun 12:1\u0026ndash;8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evon Friesen LW et al (2025) Nitrogen fixation under declining Arctic sea ice. Commun Earth Environ 6:811\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evon Friesen LW et al (2025) From temperate to polar waters: Transition to non-cyanobacterial diazotrophy upon entering the Atlantic gateway of the Arctic Ocean. Limnol Oceanogr. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/lno.70153\u003c/span\u003e\u003cspan address=\"10.1002/lno.70153\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZander S et al (2025) Contrasting nitrogen fixation between arctic and Atlantic waters in the Fram Strait. Microb Ecol 89:26\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarding K et al (2018) Symbiotic unicellular cyanobacteria fix nitrogen in the Arctic Ocean. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e 201813658 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1073/pnas.1813658115\u003c/span\u003e\u003cspan address=\"10.1073/pnas.1813658115\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCoale TH et al (2024) Nitrogen-fixing organelle in a marine alga. Science 384:217\u0026ndash;222\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCabello AM et al (2016) Global distribution and vertical patterns of a prymnesiophyte-cyanobacteria obligate symbiosis. ISME J 10:693\u0026ndash;706\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evon Friesen LW, Paulsen ML, M\u0026uuml;ller O, Gr\u0026uuml;ndger F, Riemann L (2023) Glacial meltwater and seasonality influence community composition of diazotrophs in Arctic coastal and open waters. FEMS Microbiol Ecol 99\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlais M et al (2012) Nitrogen fixation and identification of potential diazotrophs in the Canadian Arctic. Global Biogeochem Cycles 26\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFern\u0026aacute;ndez-M\u0026eacute;ndez M et al (2016) Diazotroph diversity in the sea ice, melt ponds, and surface waters of the eurasian basin of the Central Arctic Ocean. Front Microbiol 7:1\u0026ndash;18\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFernandez L, Bertilsson S, Peura S (2020) Non-cyanobacterial diazotrophs dominate nitrogen‐fixing communities in permafrost thaw ponds. Limnol Oceanogr 65:S180\u0026ndash;S193\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDelmont TO et al (2021) Heterotrophic bacterial diazotrophs are more abundant than their cyanobacterial counterparts in metagenomes covering most of the sunlit ocean. ISME J. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41396-021-01135-1\u003c/span\u003e\u003cspan address=\"10.1038/s41396-021-01135-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTurk-Kubo KA et al (2022) Non-cyanobacterial diazotrophs: Global diversity, distribution, ecophysiology, and activity in marine waters. FEMS Microbiol Rev. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/femsre/fuac046\u003c/span\u003e\u003cspan address=\"10.1093/femsre/fuac046\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiemann L et al (2022) Planktonic Aggregates as Hotspots for Heterotrophic Diazotrophy: The Plot Thickens. Front Microbiol 13:875050\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChakraborty S et al (2021) Quantifying nitrogen fixation by heterotrophic bacteria in sinking marine particles. Nat Commun 12:4085\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarding KJ et al (2022) Cell-specific measurements show nitrogen fixation by particle-attached putative non-cyanobacterial diazotrophs in the North Pacific Subtropical Gyre. Nat Commun 13:1\u0026ndash;10\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFurbo Reeder C et al (2025) Unveiling the contribution of particle-associated non-cyanobacterial diazotrophs to N\u003csub\u003e2\u003c/sub\u003e fixation in the upper mesopelagic North Pacific Gyre. Commun Biol\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFadeev E et al (2021) Sea ice presence is linked to higher carbon export and vertical microbial connectivity in the Eurasian Arctic Ocean. Commun Biol 4:1255\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarkussen TN, Elberling B, Winter C, Andersen TJ (2016) Flocculated meltwater particles control Arctic land-sea fluxes of labile iron. Sci Rep 6:24033\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSlagstad D, Wassmann PFJ, Ellingsen I (2015) Physical constrains and productivity in the future Arctic Ocean. Front Mar Sci 2\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLind S, Ingvaldsen RB, Furevik T (2018) Arctic warming hotspot in the northern Barents Sea linked to declining sea-ice import. Nat Clim Chang 8:634\u0026ndash;639\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSundfjord A et al (2020) Suggested water mass definitions for the central and northern Barents Sea, and the adjacent Nansen Basin. Nansen Leg Rep Ser. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7557/nlrs.5707\u003c/span\u003e\u003cspan address=\"10.7557/nlrs.5707\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiley JS et al (2012) The relative contribution of fast and slow sinking particles to ocean carbon export. Global Biogeochem Cycles 26\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeisler E, Bogler A, Rahav E (2019) Bar-Zeev, E. Direct Detection of Heterotrophic Diazotrophs Associated with Planktonic Aggregates. Sci Rep 9:1\u0026ndash;9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarton BI, Lenn Y-D, Lique C (2018) Observed atlantification of the Barents Sea causes the Polar Front to limit the expansion of winter sea ice. J Phys Oceanogr 48:1849\u0026ndash;1866\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBenavides M et al (2022) Sinking Trichodesmium fixes nitrogen in the dark ocean. ISME J 16:2398\u0026ndash;2405\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFarnelid H et al (2018) Diverse diazotrophs are present on sinking particles in the North Pacific Subtropical Gyre. ISME J. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41396-018-0259-x\u003c/span\u003e\u003cspan address=\"10.1038/s41396-018-0259-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eD\u0026iacute;ez B, Bergman B, Pedr\u0026oacute;s-Ali\u0026oacute; C, Ant\u0026oacute; M, Snoeijs P (2012) High cyanobacterial nifH gene diversity in Arctic seawater and sea ice brine. Environ Microbiol Rep 4:360\u0026ndash;366\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHallstr\u0026oslash;m S et al (2022) Chemotaxis may assist marine heterotrophic bacterial diazotrophs to find microzones suitable for N2 fixation in the pelagic ocean. ISME J 1\u0026ndash;10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41396-022-01299-4\u003c/span\u003e\u003cspan address=\"10.1038/s41396-022-01299-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReeder CF, L\u0026ouml;scher CR (2022) Nitrogenases in Oxygen Minimum Zone Waters. Front Mar Sci 9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM\u0026auml;kip\u0026auml;\u0026auml; R et al (2018) Methanotrophs are core members of the diazotroph community in decaying Norway spruce logs. Soil Biol Biochem 120:230\u0026ndash;232\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFernandez L et al (2020) Diazotroph genomes and their seasonal dynamics in a stratified humic bog lake. Front Microbiol 11:1500\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe R et al (2012) Shifts in identity and activity of methanotrophs in arctic lake sediments in response to temperature changes. Appl Environ Microbiol 78:4715\u0026ndash;4723\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSanz-S\u0026aacute;ez I et al (2024) Microorganisms involved in methylmercury demethylation and mercury reduction are widely distributed and active in the bathypelagic deep ocean waters. Environ Sci Technol 58:13795\u0026ndash;13807\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrumpen T et al (2019) Arctic warming interrupts the Transpolar Drift and affects long-range transport of sea ice and ice-rafted matter. Sci Rep 9:5459\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeukermans G, Reynolds RA, Stramski D (2016) Optical classification and characterization of marine particle assemblages within the western Arctic Ocean. Limnol Oceanogr 61:1472\u0026ndash;1494\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHopwood MJ et al (2019) Review Article: How does glacier discharge affect marine biogeochemistry and primary production in the Arctic? Cryosphere Discuss 1\u0026ndash;51. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5194/tc-2019-136\u003c/span\u003e\u003cspan address=\"10.5194/tc-2019-136\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSwoboda S et al (2024) Release of ballast material during sea-ice melt enhances carbon export in the Arctic Ocean. PNAS Nexus 3:gae081\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDybwad C et al (2022) The influence of sea ice cover and Atlantic water advection on annual particle export north of Svalbard. J Geophys Res C: Oceans 127\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReigstad M, Wexels Riser C, Wassmann P, Ratkova T (2008) Vertical export of particulate organic carbon: Attenuation, composition and loss rates in the northern Barents Sea. Deep Sea Res Part 2 Top Stud Oceanogr 55:2308\u0026ndash;2319\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCerd\u0026aacute;n-Garc\u0026iacute;a E, \u0026Aacute;lvarez-Salgado XA, Ar\u0026iacute;stegui J, Mart\u0026iacute;nez-Marrero A, Benavides M (2024) Eddy-driven diazotroph distribution in the subtropical North Atlantic: horizontal variability prevails over particle sinking speed. Commun Biol 7:929\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChakraborty S, Andersen KH, Merico A, Riemann L (2025) Particle-associated N2 fixation by heterotrophic bacteria in the global ocean. Sci Adv 11:eadq4693\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhite AE et al (2020) A critical review of the 15N2 tracer method to measure diazotrophic production in pelagic ecosystems. Limnol Oceanogr Methods 18:129\u0026ndash;147\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePolerecky L et al (2012) Look@NanoSIMS - a tool for the analysis of nanoSIMS data in environmental microbiology. Environ Microbiol 14:1009\u0026ndash;1023\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRStudio Team. RStudio (2023)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickham H (2009) \u003cem\u003eGgplot2: Elegant Graphics for Data Analysis\u003c/em\u003e. (Springer Science \u0026amp; Business Media. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/978-0-387-98141-3\u003c/span\u003e\u003cspan address=\"10.1007/978-0-387-98141-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKelley D (2025) R. C. \u003cem\u003eOce: Analysis of Oceanographic Data\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickham H, Fran\u0026ccedil;ois R, Henry L, M\u0026uuml;ller K, Vaughan D (2023) \u003cem\u003eDplyr: A Grammar of Data Manipulation\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndersen KS, Kirkegaard RH, Karst SM, Albertsen M (2018) \u003cem\u003eAmpvis2: An R Package to Analyse and Visualise 16S rRNA Amplicon Data\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1101/299537\u003c/span\u003e\u003cspan address=\"10.1101/299537\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcMurdie PJ, Holmes S, Phyloseq (2013) An R Package for Reproducible Interactive Analysis and Graphics of Microbiome Census Data. PLoS ONE 8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStekhoven DJ, B\u0026uuml;hlmann P (2012) MissForest\u0026ndash;non-parametric missing value imputation for mixed-type data. Bioinformatics 28:112\u0026ndash;118\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":true,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9051103/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9051103/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBiological dinitrogen (N\u003csub\u003e2\u003c/sub\u003e) fixation sustains productivity in oligotrophic oceans and is now also thought to contribute substantially to the nitrogen supply in the warming Arctic. Here we demonstrate significant N\u003csub\u003e2\u003c/sub\u003e fixation by particle-associated diazotrophs in subsurface waters of the Barents Sea, the Arctic's front runner of nutrient depletion. As the Arctic becomes nutrient-poor, the dominant nitrogen source to sustain future productivity remains unclear. Comparing our findings with subtropical studies reveals particle-associated non-cyanobacterial diazotrophs as the primary N\u003csub\u003e2\u003c/sub\u003e fixers in subsurface Arctic waters, contrasting with diverse communities in warmer regions. As the Arctic shifts towards oligotrophication, understanding the magnitude and controls of particle-associated N₂ fixation provides critical insights into future nitrogen supply and ecosystem transformation across the rapidly changing Arctic Ocean.\u003c/p\u003e","manuscriptTitle":"Particle-associated diazotrophs drive nitrogen fixation in Arctic subsurface waters","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-05 18:16:18","doi":"10.21203/rs.3.rs-9051103/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"1db8f795-5285-4d91-8251-11b0f067a214","owner":[],"postedDate":"May 5th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":64203339,"name":"Earth and environmental sciences/Ocean sciences/Marine biology"},{"id":64203340,"name":"Biological sciences/Microbiology/Environmental microbiology/Water microbiology"}],"tags":[],"updatedAt":"2026-05-05T18:16:18+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-05 18:16:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9051103","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9051103","identity":"rs-9051103","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}