Microfluidic droplet cultivation under extreme pressure enables isolation and characterization of distinct deep-sea microbial dark matter

Microfluidic droplet cultivation under extreme pressure enables isolation and characterization of distinct deep-sea microbial dark matter | 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 Microfluidic droplet cultivation under extreme pressure enables isolation and characterization of distinct deep-sea microbial dark matter Wenbin Du, Zhiyi Wang, Tong Yu, Linfeng Gong, Haibin Qi, Beiyu Hu, and 14 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7227821/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Deep-sea microorganisms comprise the Earth's largest and least explored microbiome, yet the vast majority remain uncultivated due to challenges of preserving in situ high hydrostatic pressure and preventing loss of viability and diversity during recovery, which limits our ability to explore their ecological functions and adaptive strategies. Here, we introduce DeepDrop, a microfluidics platform that enables high-throughput single-cell cultivation under pressures spanning the full ocean depth directly aboard research vessels, following direct colony formation via pipette-generated double emulsions. Applying to hadal samples, DeepDrop recovered >50% more microbial diversity than conventional high-pressure bulk cultivation, including rare taxa with streamlined genomes and distinctive genetic features associated with pressure adaptation. Combined metagenomic and transcriptomic analyses revealed that DeepDrop enriched pressure-adapted taxa carrying key stress-related genes and induced coordinated transcriptional reprogramming, characterized by upregulation of stress pathways and repression of motility. By integrating shipboard deployment, pressure-stable droplet cultivation, and efficient recovery, DeepDrop offers a powerful platform for accessing deep-sea microbial dark matter and illuminating microbial life strategies under extreme environmental constraints. Biological sciences/Biological techniques/Microbiology techniques Biological sciences/Biological techniques/Lab-on-a-chip Biological sciences/Microbiology/Environmental microbiology/Water microbiology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Deep-sea microorganisms thrive under high hydrostatic pressure (HHP), harbor diverse genetic and metabolic potential, and play pivotal roles in global biogeochemical cycling 1 , ecosystem functioning 2 , and long-term carbon storage 3 . Considerable efforts have focused on uncovering these microbes due to their promising biotechnological applications, including novel drug discovery, extreme-environment enzyme development, and bioremediation 3–5 . Although metagenomic surveys have significantly advanced our understanding of deep-sea microbial ecology and adaptive mechanisms 2,6 , more than 99% of deep-sea microbial taxa remain uncultivated, largely due to methodological limitations 7,8 . For example, depressurization during retrieval and delays in post-sampling handling frequently result in substantial loss of microbial viability and biodiversity 9 , underscoring the need for prompt shipboard cultivation strategies that mimic deep-sea conditions. Traditional cultivation methods, such as plate spreading and dilution-to-extinction, fail to replicate extreme in situ conditions and disproportionately favor fast-growing taxa. As a result, the number of novel taxa among isolates decreases with prolonged cultivation, and slow-growing or stress-tolerant species are consistently excluded from subsequent analyses 10 . Rare taxa are especially vulnerable to competitive suppression, as dominant microbes can rapidly deplete nutrients or secrete inhibitors that suppress neighboring cells, ultimately limiting the diversity that can be recovered. To address these barriers, several technical advances have emerged, including diffusion chambers that utilize in situ conditions 7 , pressure-retaining bioreactors that simulate deep-ocean pressures 11 , and microfluidic technologies capable of isolating single microbial cells in controlled microscale environments 12,13 . Droplet microfluidics has proven to be a particularly transformative technique, enabling the growth of dormant or previously uncultivated taxa, even with standard media, by eliminating interspecific competition 14–16 . This approach has been used to recover a broader diversity of microbes from various environments 17–20 . However, several key limitations have prevented droplet-based methods from being effectively applied for deep-sea microbial cultivation. Most importantly, the joint effects of deep-sea pressure and microscale confinement on microbial growth, viability, and metabolic activation—factors critical for recovering greater microbial diversity from the deep sea—remain unclear. Existing droplet systems are restricted to laboratory use, lack compatibility with high-pressure environments, and often rely on chemical demulsification, which may damage cells and reduce recovery efficiency 21 . These constraints not only limit our ability to simulate in situ conditions but also hinder the retrieval of rare or pressure-adapted microbes. Thus, an integrated platform that maintains droplet integrity under deep-sea pressure, allows straightforward cell recovery, and can be deployed on shipboard to unlock deep-sea microbial dark matter is needed. Here, we present DeepDrop, a droplet-based method designed for shipboard high-throughput single-cell cultivation of microbes under HHP. DeepDrop enables rapid shipboard processing while maintaining incubation pressures up to 110 MPa, effectively simulating deep-sea conditions. We validated DeepDrop by demonstrating droplet stability at 110 MPa, as well as the selective enrichment and isolation of pressure-adapted taxa from a model microbial community. When applied to deep-sea samples, DeepDrop yielded a more diverse microbial community with a distinct composition compared to conventional cultivation methods, as revealed by metagenomic analysis. Transcriptomic analyses of the most abundant species further revealed key adaptive mechanisms specific to DeepDrop. Finally, we demonstrated DeepDrop's ability to isolate rare and novel microbial species, underscoring its potential to cultivate previously uncultivated deep-sea taxa. Results The DeepDrop workflow for shipboard cultivation To bridge the gap between deep-sea and shipboard cultivation conditions, we developed DeepDrop, an integrated method that combines droplet microfluidics and HHP incubation (Fig. 1a). The method involves a rapid shipboard workflow: deep-sea samples are pretreated and converted into cell suspensions within 20 minutes after sample collection. A portable, custom-built microfluidic instrument equipped with a flow-focusing chip (Supplementary Fig. 1) is subsequently used to encapsulate individual cells into picoliter droplets at 5×10 5 drops/min. These droplets are collected in syringes and transferred to a titanium alloy pressure vessel, where hydrostatic pressure is applied via syringe pistons. The pressure transmission within the system is verified using compressible foam, which deforms in response to the applied HHP (Supplementary Fig. 2a,b). Notably, the droplets remain stable without fusion even at hydrostatic pressures of up to 110 MPa, as validated in the laboratory to simulate the extreme conditions of the Mariana Trench (11,000 meters) (Supplementary Fig. 2c). Following HHP incubation, pipette-driven double emulsification is performed to convert droplets into water-in-oil-in-water emulsions, enabling linear dilution and direct plating of droplets onto diverse agar media without demulsification (Supplementary Fig. 3). This end-to-end workflow maintains microbial viability by preserving in situ pressure levels, facilitating high-throughput isolation of novel pressure-adapted taxa. To evaluate the efficacy of DeepDrop, we compared microbial enrichment in cultures under various conditions: droplets at in situ pressure, droplets at half in situ pressure, droplets at atmospheric pressure, and bulk cultures under HHP conditions (Fig. 1b). After incubation, droplets were reemulsified into double emulsions and plated onto nutrient-diluted agar for strain isolation. Metagenomic sequencing was performed on all samples in parallel to analyze microbial composition and functional gene enrichment. This integrated approach establishes a direct link between cultivation environment, strain recovery, and functional adaptation, validating DeepDrop as a robust platform for accessing the deep-sea microbiome. Droplet stability under HHP and its suitability for deep-sea microbe cultivation To assess the structural robustness of droplets under prolonged high-pressure incubation, we tested droplets prepared with 2216E medium at simulated deep-sea pressures corresponding to depths of 4,000–8,000 meters (40–80 MPa), with atmospheric pressure (0.1 MPa) as a control. During a 7-day incubation onboard a research vessel, the pressure chambers were exposed to continuous ship-induced shaking. Droplet diameters remained unchanged across all conditions, confirming excellent structural stability under both high pressure and mechanical disturbance (Supplementary Fig. 2d, e). These results validate the suitability of DeepDrop for long-term deep-sea microbial cultivation under realistic field conditions. We next assessed DeepDrop's ability to selectively enrich pressure-adapted microbes using a mock community composed of the pressure-tolerant isolate Marinobacter profundi 22 and pressure-sensitive Escherichia coli (Fig. 2a). In pure cultures diluted to single cells, both strains proliferated under atmospheric pressure. In contrast, only M. profundi exhibited robust growth at 50 MPa (Fig. 2b). In droplets containing the mixed community, 16S rRNA gene sequencing revealed that E. coli dominated at 0.1 MPa, with M. profundi comprising just 16.1% of the total population. At 50 MPa, however, DeepDrop selectively enriched M. profundi , increasing its relative abundance to 91.6% (Fig. 2c, d). Colony profiling using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) confirmed that significantly more M. profundi colonies were recovered from DeepDrop cultures incubated at 50 MPa than from those cultivated at 0.1 MPa (Fig. 2e, Supplementary Fig. 4). At atmospheric pressure, E. coli overgrowth suppressed M. profundi isolation, whereas high-pressure cultivation enabled successful recovery of this pressure-tolerant strain. These results establish DeepDrop as a robust method for maintaining high-pressure integrity and selectively recovering HHP-adapted microbes from competitive communities. High pressure and microscale confinement synergistically enhance taxonomic richness and evenness Previous droplet-based cultivation studies have shown increased microbial richness compared to conventional methods under atmospheric conditions. However, it remains unknown whether combining this with in situ high pressure can further improve deep-sea microbial recovery. To address this, we applied DeepDrop to a near-bottom seawater sample collected from 5,500 m depth (S1, Supplementary Table 1), cultivating it under four conditions: DeepDrop 55 MPa ( in situ pressure), DeepDrop 28 MPa (half of the in situ pressure), droplet 0.1 MPa (atmospheric pressure), and conventional bulk culture at 55 MPa. All samples underwent metagenomic sequencing after cultivation. After cultivation, the samples were demulsified and pooled for metagenomic sequencing to reveal enriched microbial diversity. DeepDrop 55 MPa outperformed bulk HHP cultivation in microbial richness, as measured by the Chao1 index, which estimates species richness (Fig. 3a). Concerning community structure, DeepDrop 55 MPa resulted in greater evenness than lower-pressure conditions did, as indicated by the Simpson index (Fig. 3b). Principal coordinate analysis (PCoA) further revealed distinct clustering of communities under DeepDrop 55 MPa conditions compared with other conditions (Fig. 3c). Notably, DeepDrop outperformed bulk HHP cultivation, recovering 135 species compared to the 89 species recovered with bulk culture; notably, 51 unique species were identified with DeepDrop, while only five unique species were identified with bulk culture (Fig. 3d, e). DeepDrop thus captured approximately 94.4% of the taxa recovered by the bulk methods while accessing additional unique species (Fig. 3d). At the genus level, DeepDrop at 55 MPa recovered 79 genera, representing a nearly 50% increase over the 54 genera identified by bulk cultivation at the same pressure, with 30 genera unique to DeepDrop compared with only five unique to the bulk (Extended Data Fig. 1a, Supplementary Table 2). More specifically, the predominant microbial taxa under DeepDrop 55 MPa were distinct from those under the other conditions (Fig. 3f), highlighting the synergistic effect of HHP and microscale confinement in accessing unique deep-sea microbial species. DeepDrop selectively enriches distinct microbial taxa from deep-sea samples Microbial species can exhibit different sensitivities to varying hydrostatic pressures 23 . Our study revealed marked shifts in species abundances along a pressure gradient from 0.1 MPa to 28 MPa and 55 MPa (Fig. 3g). Notably, Aequorivita SGB2713, Psychrobacter oceani , and Halomonas meridiana presented progressively increased relative abundance with rising pressure. Linear discriminant analysis effect size (LEfSe) further identified these species as the most significantly enriched in the droplet environment under HHP, highlighting them as potential indicator taxa of pressure-adapted communities (Supplementary Fig. 5a, b). Conversely, the abundances of Alteromonas macleodii and Thalassospira indica decreased markedly, indicating pressure sensitivity or inhibition of proliferation. Meanwhile, species like Pseudomonas stutzeri displayed resilience to pressure variation, maintaining consistent abundances across conditions without significant shifts. Collectively, these findings demonstrate that integrating HHP with microscale confinement in DeepDrop enhances the recovery of unique pressure-adapted taxa, substantially expanding the recovery of deep-sea microbiota compared to conventional methods. To further dissect the effects of microscale confinement, we compared the relative abundance of each species in DeepDrop and bulk cultivation (Fig. 3h). Certain taxa, notably Alcanivorax xenomutans, Aequorivita SGB2713, and Psychrobacter oceani, showed substantial enrichment in DeepDrop, increasing by 91-, 20-, and 4.3-fold, respectively. Additionally, the 51 species exclusively detected in DeepDrop accounted for approximately 2.8% of the total community abundance, emphasizing the unique selective pressures imposed by confinement (Fig. 3d, e). This pattern was further supported by LEfSe analysis (Supplementary Fig. 5c, d). In contrast, some species thrived better in bulk 55 MPa, with Alteromonas abrolhosensis and Salipiger bermudensis showing the greatest reductions in DeepDrop 55 MPa, by 39- and 24-fold, respectively. These findings indicate that while confinement does not universally benefit all microbial taxa, it creates a distinct environment that selectively favors certain species, enabling the enrichment of microbes that would otherwise be unrecoverable in bulk cultures. Finally, analysis of the top 10 genera, showing differential abundance across all cultivation conditions (Extended Data Fig. 1b), revealed a pronounced synergistic effect of high pressure and droplet confinement. Genera such as Aequorivita SGB2713, Psychrobacter oceani , and H. meridiana , individually responsive to either pressure or confinement, exhibited markedly greater relative abundances under the combined DeepDrop 55 MPa condition. These findings collectively demonstrate that DeepDrop, by integrating extreme hydrostatic pressure and microscale confinement, effectively enriches distinct microbial taxa from hadal samples, significantly expanding access to previously uncultivated deep-sea microbiota. DeepDrop enriched deep-sea taxa exhibit unique functional adaptations To elucidate the genetic mechanisms underlying microbial adaptation within DeepDrop, we further analyzed the metagenomic data of S1 across various cultivation conditions. KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway enrichment analysis revealed that genes associated with ATP-binding cassette (ABC) transporters and quorum sensing (QS) were uniquely and significantly enriched under DeepDrop 55 MPa (Fig. 4a). Additionally, species-specific analyses linked these enriched pathways to microbial taxa predominating under DeepDrop cultivation conditions (Supplementary Figs. 6–7), suggesting that these genes potentially contribute to microbial survival under DeepDrop cultivation conditions. To investigate the genetic basis for microbial adaptations under DeepDrop cultivation, we performed gene-level enrichment analysis on metagenomic data from cultures grown under various conditions. Genes encoding ABC transporters involved in osmoprotectant transport (e.g., betaine, glycerol, glycine) and membrane component trafficking (e.g., lipopolysaccharides, phospholipids) were particularly enriched under DeepDrop conditions (Fig. 4b). Additionally, genes associated with quorum sensing (QS) pathways, particularly those involved in inducer production and receptor synthesis common in marine bacteria, were significantly enriched in DeepDrop cultures at 55 MPa (Fig. 4b). In contrast, genes linked to chemotaxis, flagellar assembly, and biofilm formation pathways were more enriched in the bulk HHP cultures compared to DeepDrop 55 MPa and other droplet conditions (Fig. 4a). Specifically, genes encoding aerotaxis-related chemotaxis, flagellar stator components, and polysaccharide biosynthesis were more enriched in bulk HHP cultures (Fig. 4c), reflecting a selective advantage for motility and surface colonization under bulk conditions, likely driven by intensified interspecies competition and chemical gradients. Conversely, the spatial isolation and homogeneous nutrient conditions within picoliter droplets likely diminish the need for motility and environmental navigation, resulting in diminished investment in flagellar systems and chemotactic responses. These findings highlight distinct microbial adaptation strategies driven by the combined selective pressures of high hydrostatic pressure and microscale confinement in DeepDrop. Transcriptomic response of representative DeepDrop taxa under various microenvironments To investigate the physiological adaptations of representative taxa under various conditions, we conducted transcriptomic profiling of H. meridiana CR14, a dominant strain significantly enriched in DeepDrop 55 MPa compared to other conditions (Fig. 3f). Cultures were grown under four conditions: DeepDrop 55 MPa, bulk 55 MPa, droplet 0.1 MPa, and bulk 0.1 MPa. PCoA revealed a distinct transcriptomic profile for DeepDrop 55 MPa, indicating the combined influence of HHP and single-cell confinement (Fig. 4d). To investigate gene expression changes driven by HHP, we compared the transcriptomic profiles of H. meridiana CR14 cultured under DeepDrop 55 MPa and droplet 0.1 MPa conditions. This comparison revealed 594 upregulated and 565 downregulated genes, with a significant upregulation (P < 1e-50) in genes involved in proteostasis (e.g., chaperones, proteases), ROS detoxification (e.g., peroxidases), membrane stabilization (e.g., lipopolysaccharide transporters), and osmoregulation. These changes suggest a comprehensive stress response to HHP (Fig. 4e). Conversely, genes associated with flagellar assembly were significantly downregulated, indicating a shift away from energy-intensive motility processes under HHP (Extended Data Fig. 2a, c). Further investigation of the combined effects of HHP and microscale confinement was conducted by comparing the transcriptomic profiles of H. meridiana CR14 grown under DeepDrop 55 MPa and bulk 55 MPa conditions. This comparison revealed 740 upregulated and 600 downregulated genes. The droplet condition showed reduced expression of ribosomal genes and increased transcription of genes related to alternative carbon metabolism, suggesting resource limitation within picoliter droplets (Fig. 4f, Extended Data Fig. 2b). Similar transcriptional shifts were also observed when comparing 0.1 MPa droplet conditions with bulk 0.1 MPa cultures (Supplementary Fig. 8), indicating that spatial confinement alone can drive significant changes in microbial metabolism. Together, these results demonstrate that deep-sea microbial taxa, such as H. meridiana CR14, adapt to DeepDrop conditions by upregulating genes involved in proteostasis, oxidative stress mitigation, and the maintenance of membrane and osmotic stability, while downregulating genes related to energy-consuming motility systems (Fig. 4g). These transcriptomic changes correlate with the community-level trends observed in the metagenomics, further supporting the mechanistic basis of DeepDrop's ability to selectively enrich pressure-adapted taxa. DeepDrop enables the recovery of novel lineages through genome-resolved and culture-based approaches To assess the capacity of DeepDrop to recover novel microbial lineages, we employed both genome-resolved metagenomics and culture-based isolation following high-pressure droplet cultivation. These complementary strategies enabled us to assess how the system facilitates the discovery of previously inaccessible taxa from deep-sea environments. Metagenomic binning across all cultivation conditions yielded 115 metagenome-assembled genomes (MAGs) with >50% completeness and <10% contamination (Supplementary Table S3). Among these MAGs, 53 met high-quality standards (≥80% completeness, ≤5% contamination) and were used for comparative analysis (Fig. 5a). DeepDrop at 55 MPa recovered 16 high-quality MAGs—more than those obtained from bulk HHP cultivation (n = 9) and slightly exceeding DeepDrop at 28 MPa (n = 15) and droplet cultures at atmospheric pressure (n = 13). Among the DeepDrop 55 MPa MAGs, three lacked genus-level classification based on Genome Taxonomy Database (GTDB) 24 , suggesting considerable novelty. Although total MAG counts were comparable across conditions, the partially distinct MAG profiles recovered via DeepDrop point to the selective enrichment of unique genomic lineages not captured by bulk approaches. To explore functional adaptations, we screened the high-quality MAGs for the presence or absence of key gene modules that had been previously identified as enriched at the community level. A subset of MAGs, particularly those from droplet-based cultivation, lacked chemotaxis and flagellar biosynthesis genes, consistent with the observed decrease in the enrichment of motility-related pathways in metagenomes from DeepDrop cultures (Fig. 4a, Fig. 5a). These MAGs presented highly streamlined genomes, a pattern further supported by the disproportionately small genome sizes observed in droplet-based cultures (Supplementary Fig. 9). Together, these findings support the hypothesis that spatial confinement reduces selective pressure for motility, favoring stress adaptation. To test whether metagenomically enriched taxa could also be isolated in culture, we developed a droplet-to-colony pipeline that utilizes a pipette-driven double-emulsification step to directly plate incubated droplets. This simple approach enabled successful colony formation from droplets containing tens to hundreds of cells—a critical feature for recovering rare or slow-growing microbes. DeepDrop cultivation of sample S1 at 55 MPa followed by plating on five marine media yielded 176 isolates across 24 genera (Fig. 5b). Remarkably, 83% of these genera (20/24) were rare taxa (<1% relative abundance) in the original community. Several, including Aequorivita and Psychrobacter , harbored novel species previously shown to be enriched in DeepDrop metagenomes (Extended Data Fig. 1b). These results demonstrate a strong correspondence between taxa enriched at the community level and those recoverable via cultivation, highlighting DeepDrop's capacity to bridge metagenomic discovery and culture recovery. The DeepDrop isolation protocol was applied to samples S2 and S3, in addition to sample S1, resulting in a total of 403 isolates. From these isolates, 346 high-quality, near-full-length 16S rRNA gene sequences were obtained, enabling the identification of 70 species spanning 33 genera and four phyla (Fig. 5c, Extended Data Fig. 3). Notably, four novel species (marked with red stars) were identified and characterized through morphological examination and phylogenetic analysis (Supplementary Figs. 10). Furthermore, 31 species isolated from S1 represented rare taxa, each with a relative abundance of less than 1% in the original community. These findings underscore DeepDrop's capacity to recover previously uncultivated, rare species from deep-sea environments, highlighting its potential for exploring and understanding microbial diversity in the deep sea." To increase recovery diversity, we employed five commonly used marine and laboratory culture media—2216E, 2216E+ (with 3.9% NaCl), R2A, YPD, and LB. These media were diluted to one-tenth strength in artificial seawater, offering a broad range of nutrients to better simulate marine conditions. The medium composition markedly influenced taxonomic richness, with R2A agar supporting the highest diversity, including 19 species recovered exclusively from this medium (Fig. 5d). Habitat-specific trends were also apparent: H. meridiana , Halopseudomonas aestusnigri , and Pseudomonas aeruginosa were dominant in samples S1, S2, and S3, respectively. These patterns might be influenced by the different depths, sample types, and collection methods, which together created varying environmental conditions across the samples. Finally, representative isolates such as H. meridiana CR14, Aequorivita sp. CE311, Psychrobacter sp. CR212, and Alcanivorax xenomutans CRS8 displayed robust growth at both 0.1 and 55 MPa, confirming their HHP-tolerance (Supplementary Fig. 10e). These results illustrate the impact of media composition and environmental variability on microbial recovery, highlighting DeepDrop's capacity to access a broader diversity of deep-sea microbes across varied conditions. Discussion Microbial communities in extreme environments, such as deep-sea ecosystems, present significant challenges for cultivation due to the high pressures and unique conditions that limit microbial growth and survival. 7–9,25,26 As an advanced platform that combines high-pressure incubation with droplet microfluidics, DeepDrop offers a robust solution for scaling and deploying microbial cultivation under extreme conditions. It enables the generation of picoliter-scale droplets capable of withstanding hydrostatic pressures up to 110 MPa, replicating the extreme conditions found in the deepest ocean 27 . By isolating individual cells within droplets, DeepDrop prevents overgrowth and interspecific competition, allowing slow-growing or dormant microbes to be cultivated effectively. Additionally, by directly plating double-emulsified droplets containing high-density cells grown from single cells, the system improves recovery efficiency while preventing the cytotoxicity associated with chemical demulsification. Thus, DeepDrop provides a versatile and efficient tool for cultivating and studying deep-sea microorganisms, offering valuable insights into their diversity and functional capabilities. To evaluate the platform's effectiveness in practice, we compared DeepDrop to conventional bulk high-pressure cultivation. We found that it markedly improves the recovery of microbial diversity from deep-sea samples. For example, cultivation of S1 at 55 MPa yielded 135 species under DeepDrop, substantially exceeding the 89 species retrieved using bulk methods. More importantly, the taxonomic profiles differed significantly between formats, suggesting that the unique interplay of hydrostatic pressure and microscale confinement fosters the selective enrichment of otherwise inaccessible microbial lineages. This trend was reinforced by pressure‒gradient experiments, which revealed differential species responses shaped by confinement and pressure. Genome-resolved analyses further supported these findings: DeepDrop 55 MPa recovered the highest number of high-quality MAGs, including three with no taxonomic assignment in the GTDB database. The limited overlap between novel MAGs recovered via DeepDrop and those from bulk HHP cultivation underscores the platform's capacity to uncover hidden phylogenetic and functional diversity. These findings build upon and extend earlier studies in high-pressure microbiology by demonstrating that physical compartmentalization 18,28 , when coupled with environmental simulation, enhances both the breadth and depth of microbial recovery. Beyond expanding taxonomic breadth, DeepDrop offers mechanistic insight into the functional adaptations that enable microbial survival under high-pressure microscale confinement. Metagenomic analysis revealed enrichment of genes involved in osmoprotectant transport, membrane component stabilization, and quorum sensing, alongside depletion of genes linked to flagellar assembly, chemotaxis, and biofilm formation. These shifts reflect an ecological transition from motility-driven competition to survival-oriented strategies, favoring nonmotile phenotypes adapted to the constrained droplet environment. Traits such as osmoprotectant accumulation and phospholipid transport help maintain cellular homeostasis and membrane integrity under HHP 29–32 , while confined geometries likely accelerate quorum sensing activation by concentrating signaling molecules 33 . Supporting this interpretation, genome-level annotations of high-quality MAGs from droplet cultivation revealed a distinct subset lacking motility-related genes, consistent with reduced selection for active dispersal. To further explore this adaptation at the transcriptional level, we profiled H. meridiana CR14, a dominant DeepDrop-enriched species. Transcriptomic analysis showed coordinated upregulation of stress response pathways, including proteostasis 34 , oxidative stress defense 35 , and membrane stabilization 36 . These changes were accompanied by suppression of motility gene expression 37,38 and reduced ribosomal activity, along with increased expression of pathways associated with alternative carbon metabolism 35 . These features are characteristic of a resource-efficient, low-growth physiological state and are consistent with previously reported adaptations in γ-Proteobacteria under pressure or nutrient limitation 39 . Together, these multiomic data demonstrate that DeepDrop selectively cultivates stress-resilient, nonmotile microbes whose transcriptional programs are tuned for energy conservation and environmental resilience under the dual constraints of pressure and spatial confinement (Fig. 4g). A distinguishing innovation of DeepDrop is its pipette-based double-emulsion plating technique, which converts primary water-in-oil droplets into stable water-in-oil-in-water emulsions. This process prevents droplet coalescence, allows precise dilution, and enables even distribution on agar surfaces without the need for chemical demulsifiers. By preserving droplet integrity and enabling gentle transfer to solid media, this approach enhances the viability and recovery of rare or slow-growing strains. Applying this workflow, we successfully isolated 70 species from 346 colonies cultivated on standard marine media, including 12 isolates belonging to four novel species. Notably, many of these isolates originated from genera that were preferentially enriched under DeepDrop conditions, highlighting the platform's capacity to link environmental selection with strain-level cultivation. This outcome reflects a key advantage of plating droplets containing high-density outgrowths derived from single cells. Even when individual cells have a low probability of forming colonies, the cumulative probability within droplets is higher, and direct competition between species is minimized. As a result, slow-growing or low-abundance taxa have an improved chance of recovery. Together, these findings demonstrate that DeepDrop not only enables access to previously underrepresented deep-sea lineages but also serves as a practical tool for cultivating microbes that have long evaded conventional isolation efforts. While DeepDrop effectively enhances the recovery of diverse and previously uncultivated microbial taxa, several limitations remain. The use of standard marine media likely constrained the diversity and novelty of isolates, suggesting that customized media formulations tailored to deep-sea nutritional and physicochemical conditions could further expand recovery 8,40 . Additionally, the current cultivation protocol employed uniform incubation conditions, including fixed temperature, oxygen levels, and duration, which may not capture the full range of physiological adaptations across deep-sea niches. Introducing variable temperature regimens 41 , extended cultivation periods 42 , and anaerobic conditions 43 may improve the isolation of specialized microbes. A notable challenge is the recovery of obligate piezophilic species 44 . Integrating pressure-retaining samplers and continuous high-pressure cultivation modules could address this gap 45 . Despite these constraints, DeepDrop offers a transformative platform for deep-sea microbiology, recreating high-pressure environments at single-cell resolution while enabling practical, shipboard-compatible workflows. Continued refinement of the system, including media optimization and pressure-retaining capabilities, will expand its utility for cultivating microbial "dark matter." Ultimately, DeepDrop opens new avenues for exploring the ecological roles and biotechnological potential of microorganisms adapted to Earth's most extreme environments. Methods Sample collection and preparation. Three deep-sea samples were collected from the southern Mid-Atlantic Ridge on the Deep Sea No. 1 research vessel in February 2024 using specialized equipment (Supplementary Table 1). Near-bottom seawater (S1, 5500-meter depth) was collected with a CTD Rosette sampler, surface sediment (S2, 3300-meter depth) was obtained using a push core sampler deployed by the Jiaolong submersible 46 , and a 0.22 μm pore-size filter membrane (Millipore, Burlington, MA) with concentrated bacterial communities (S3, 2000-meter depth) was retrieved using an custom-built in situ seawater filtration system. For pretreatment, microbial cells from S1 were concentrated using a large volume concentration kit (Cat. No. CC01116, InnovaPrep, Drexel, MO). The sediment from S2 was suspended in sterile near-bottom seawater and shaken; the supernatant was then collected. In contrast, microbes from S3 were eluted from filter membranes in sterile seawater containing 10% glycerol. All samples were adjusted to ~10 7 cells/mL in 2216E medium before droplet generation, ensuring efficient single-cell encapsulation (~0.3 cells/droplet) as predicted by Poisson statistics. Microfluidic chip fabrication. The droplet-generation microfluidic chip was modified from a previous design to produce 30 pL droplets 47 (Supplementary Fig. 2). Soft lithography was used to fabricate chips with channel heights of 40 μm 48 . To render the channels hydrophobic, filtered Aquapel (PGW Auto Glass, Cranberry Twp, PA) was applied and incubated for 3 minutes, followed by flushing with fluorinated oil (Novec 7500, 3M, St. Paul, MN) and drying at 120°C for 10 minutes. Droplet cultivation under high hydrostatic pressures. Using the DeepDrop method, microbial cells were encapsulated into 30 pL droplets with fluorinated droplet-generation oil (Cat. No. 186-4006; Bio-Rad, Hercules, CA) at flow rates of 1600 μL/h (oil) and 800 μL/h (aqueous). Approximately 300 μL droplets were collected in 1 mL disposable plastic syringes and overlaid with mineral oil (Cat. No. M175243, Mreda, Beijing, China) to prevent coalescence during pressurization. The syringe needle was sealed with a butyl rubber plug and placed into a water-filled titanium alloy pressure vessel to simulate HHP conditions. For the bulk HHP culture controls, 300 μL of microbial suspension was directly added to syringes without droplet encapsulation. Atmospheric droplet cultures followed the same workflow without the application of pressure. All the cultures were incubated at 25°C. Notably, owing to the extraordinary stability of the carbon–fluorine bond, fluorinated oils are recalcitrant to microbial degradation and cannot serve as a carbon source. Double emulsification and microbial isolation. Following HHP incubation, the droplets were transferred to a centrifuge tube containing 1% Tween-80 in PBS (pH 7.4). Using a 1-mL pipette, the droplets were gently aspirated and dispensed at a rate of 20 cycles min -1 , with microscopy examination ensuring completeness. The double emulsions were then diluted and spread onto various agar media, including 1/10 strength 2216E, R2A, LB, YPD, and 2216E supplemented with 3.9% NaCl (2216E+). The plates were incubated at 30°C to isolate microbial colonies. DeepDrop for mock community construction and validation of deep-sea strains. The Mock community was constructed using pressure-adapted Marinobacter profundi PWS21 T isolated from a deep-sea sediment sample collected from the New Britain Trench 22 and pressure-sensitive Escherichia coli ATCC 25922 obtained from the American Type Culture Collection (ATCC). Both strains were cultured in 2216E broth at 30 °C. Cells were stained with 10 μM SYTO9 fluorescent dye (Cat. No. FL0260, Toyephon, Beijing, China), and cell concentrations were quantified using a customized cell counting chip 49 . The two strains were then adjusted to 4 × 10 6 cells/mL, mixed, and encapsulated into picoliter droplets for cultivation under two pressure conditions: 50 MPa and atmospheric pressure (0.1 MPa). After a 3-day incubation, emulsions from each condition, along with the initial microbial mixture, were demulsified for high-throughput 16S rRNA gene amplicon analysis using next-generation sequencing to profile community composition. In parallel, emulsions from each condition were re-emulsified by pipetting to generate double emulsions, which were subsequently plated on 2216E agar to assess strain recovery. Each condition was tested in triplicate, and 30 strains were randomly selected from each replicate for identification. Identification of strains isolated via DeepDrop. For mock community isolates, bacterial identification was performed using an EXS 2600 MALDI-TOF MS system (Zybio, Chongqing, China) according to the manufacturer's protocol. For isolates from deep-sea samples, genomic DNA was extracted from single colonies via the DeepDrop single-cell cultivation method. The bacterial 16S rRNA gene was amplified via the universal primers 27F (5'-AGAGTTTGATCCTGGCTCAG-3') and 1492R (5'-GGTTACCTTGTTACGACTT-3'), purified, and sequenced on a 3730xl DNA Analyzer (Thermo Fisher Scientific, Wilmington, DE). The assembled sequences were analyzed using DNASTAR software and subjected to BLAST searches against the EzBioCloud database (http://www.ezbiocloud.net/). Isolates with <98.7% 16S rRNA sequence similarity and <95% ANI similarity to the closest reference genome were designated novel species 50 . Pooled 16S rRNA sequencing of cultured communities. Genomic DNA was extracted from DeepDrop cultures, atmospheric droplet cultures, and mock community samples using the FastPure Soil DNA Isolation Kit (MJYH, Shanghai, China). The V3-V4 region of the 16S rRNA gene was amplified with the primers 338F and 806R, prepared into libraries using the NEXTFLEX Rapid DNA-Seq Kit (Revvity, Waltham, MA), and sequenced on the Illumina NextSeq2000 platform (Illumina, San Diego, CA). The raw reads were quality-filtered with fastp 51 (v0.23.2) and merged using Flash 52 (v1.2.11). OTUs were clustered at 97% similarity using UPARSE 53 (v7.1) and taxonomically annotated using the Silva 16S rRNA database (v138) after rarefaction. Community structure differences were assessed via nonmetric multidimensional scaling (NMDS) based on Euclidean distance. Metagenomic profiling of deep-sea water samples under diverse culture conditions. Genomic DNA was extracted by Guangdong MagiGene Biotechnology Co., Ltd. (Guangzhou, China) and used to construct sequencing libraries with the ALFA-SEQ DNA Library Prep Kit. Paired-end 150 bp sequencing was performed on the Illumina platform. Quality control of the raw reads was conducted using fastp (v0.23.2) to remove low-quality reads and adapters, resulting in high-quality clean reads. These reads were assembled into scaffolds using MEGAHIT 54 (v1.2.9), followed by fragmentation into scaftigs (≥500 bp) by removing ambiguous regions containing "N". Open reading frames (ORFs) were predicted with Prodigal 55 (v2.6.3), and nonredundant unigenes were clustered with MMseqs2 56 at 95% sequence identity. Gene abundance was quantified by mapping clean reads to the unigene catalog using BBMap. Taxonomic annotations were performed using MetaPhlAn 57 , leveraging species-specific marker genes, while functional annotations were conducted against the KEGG database using DIAMOND 58 . For MAG recovery, high-quality reads were assembled using SPAdes 59 (v3.15.5) using the '--meta' option. The MAGs were then binned using MetaWRAP 60 (v1.3.2), and their completeness and contamination were estimated using CheckM 61 (v1.2.2). Only MAGs with completeness ≥50% and contamination ≤ 10% were retained. Taxonomic classification of each MAG was performed with GTDB-Tk 24 , (v2.4.0) based on the GTDB taxonomy release 220, while protein-coding sequences (CDSs) were predicted using PROKKA 62 (v1.14.6). Finally, all the predicted proteins were annotated using the KEGG database with DIAMOND BLASTP (E-value =1e-5). Whole-genome sequencing and analysis of the enriched isolates. Genomic DNA was isolated from bacterial cultures using a magnetic bead-based bacterial/fungal DNA extraction kit (Majorbio, Shanghai, China). DNA quality was evaluated with a Nanodrop spectrophotometer (ND-1000, Thermo Fisher), and gel electrophoresis was performed. Paired-end DNA libraries (PE150) were constructed using the NEXTFLEX Rapid DNA-Seq Kit (Revvity, Waltham, MA) and sequenced on an Illumina platform. Quality control of the raw reads was conducted with Fastp. Genome assembly was performed using SOAPdenovo (v2.04) 63 , and residual gaps in the scaffolds were closed with GapCloser (v1.12). Genes were predicted using Prodigal (v2.6.3). Transcriptomic analysis of H. meridiana CR14 under different culture conditions. Total RNA from H. meridiana CR14 cultivated under different conditions was extracted by Guangdong MagiGene Biotechnology Co., Ltd. (Guangzhou, China), and its quality was assessed by agarose gel electrophoresis and a Qubit 3.0, with further validation using the Agilent 4200 system. RNA libraries were prepared with the ALFA-SEQ RNA Library Prep Kit II (MagiGene) and sequenced on the Illumina platform to generate paired-end 150 bp reads. Raw reads were processed with Fastp for quality control, and ribosomal RNA was filtered using Bowtie2 64 (v2.4.5) against the NCBI Rfam database. High-quality reads were mapped to the H. meridiana CR14 genome with Bowtie2, and transcript abundance was calculated as TPM values using RSEM 65 (v1.3.3). Differentially expressed genes (DEGs) were identified with DESeq2 66 (v1.34.0) using thresholds of FDR ≤ 0.05 and |log2(fold change)| ≥ 1. KEGG pathway analyses of the DEGs were conducted using clusterProfiler 67 (v4.2.2). Statistical analyses were carried out on the MagiGene Cloud Platform (http://cloud.magigene.com). Declarations Data availability The data that support the findings of this study are available from the corresponding author upon request. All sequencing data, including metagenomic, transcriptomic, 16S rRNA amplicon sequencing, and strain genome sequences, have been submitted to the National Microbial Data Center (NMDC) and are accessible under BioProject ID NMDC10019562 (https://nmdc.cncb.ac.cn/). Acknowledgments This work was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB0810000) and the National Natural Science Foundation of China (92251302, 42376238). We appreciate the support from the China Ocean Mineral Resources R&D Association (COMRA) and the National Deep Sea Center for organizing and supporting the deep-sea expedition (DY83). We thank the scientists and crew of the research vessel Deep Sea No. 1 , as well as the pilots of the JiaoLong human-occupied vehicle (HOV),for their invaluable assistance in collecting the samples. Author contributions Z.W.: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing - original draft. T.Y.: Conceptualization, Investigation. L.G.: Investigation. H.Q.: Resources. B.H.: Conceptualization, Visualization. X.D.: Investigation. X.Z.: Investigation. C.L.: Visualization. S.W.: Investigation. X.W.: Formal analysis. Y.S.: Formal analysis. Z.W.: Investigation. G.L.: Resources. H.D.: Resources. L.J.: Formal analysis. X.X.: Resources, Writing - review & editing. L.H.: Writing - review & editing. H.W.: Conceptualization, Resources, Writing - review & editing. Z.S.: Conceptualization, Resources, Writing - review & editing. W.D.: Conceptualization, Formal analysis, Funding acquisition, Project administration, Supervision, Writing - review & editing. Competing interests The authors declare no competing interests. Additional information Supplementary information is available for this paper at https://doi.org/10.1038/XXXXX-X. Correspondence and requests for materials should be addressed to Wenbin Du. References Zeng, X., Alain, K. & Shao, Z. Microorganisms from deep-sea hydrothermal vents. Mar. Life Sci. Technol. 3 , 204–230 (2021). Dong, X. et al. Evolutionary ecology of microbial populations inhabiting deep sea sediments associated with cold seeps. Nat. Commun. 14 , (2023). Jiao, N. et al. The microbial carbon pump and climate change. Nat. Rev. Microbiol. 22 , 408–419 (2024). Zhou, Z., St. John, E., Anantharaman, K. & Reysenbach, A.-L. Global patterns of diversity and metabolism of microbial communities in deep-sea hydrothermal vent deposits. Microbiome 10 , 241 (2022). Jørgensen, B. B. & Boetius, A. Feast and famine — microbial life in the deep-sea bed. Nat. Rev. Microbiol. 5 , 770–781 (2007). Ansorge, R. et al. Functional diversity enables multiple symbiont strains to coexist in deep-sea mussels. Nat. Microbiol. 4 , 2487–2497 (2019). Kaeberlein, T., Lewis, K. & Epstein, S. S. Isolating ‘Uncultivable’ Microorganisms in Pure Culture in a Simulated Natural Environment. Science 296 , 1127–1129 (2002). Foustoukos, D. I. et al. Cultivation of uncultured marine microorganisms. Mar Life Sci Technol 3 , 117–120 (2021). La Cono, V. et al. Shifts in the meso‐ and bathypelagic archaea communities composition during recovery and short‐term handling of decompressed deep‐sea samples. Environ. Microbiol. Rep. 7 , 450–459 (2015). Rinke, C. et al. Insights into the phylogeny and coding potential of microbial dark matter. Nature 499 , 431–437 (2013). Zeng, X. et al. Pyrococcus CH1, an obligate piezophilic hyperthermophile: extending the upper pressure-temperature limits for life. ISME J. 3 , 873–876 (2009). Ma, L. et al. Gene-targeted microfluidic cultivation validated by isolation of a gut bacterium listed in Human Microbiome Project’s Most Wanted taxa. Proc. Natl. Acad. Sci. U.S.A. 111 , 9768–9773 (2014). Hu, B. et al. One cell at a time: droplet-based microbial cultivation, screening and sequencing. Mar. Life Sci. Technol. 3 , 169–188 (2021). Carnes, E. C. et al. Confinement-induced quorum sensing of individual Staphylococcus aureus bacteria. Nat. Chem. Biol. 6 , 41–45 (2010). Weitz, M. et al. Communication and computation by bacteria compartmentalized within microemulsion droplets. J. Am. Chem. Soc. 136 , 72–75 (2014). Orevi, T., Sørensen, S. J. & Kashtan, N. Droplet size and surface hydrophobicity enhance bacterial plasmid transfer rates in microscopic surface wetness. ISME Commun. 2 , 72 (2022). Watterson, W. J. et al. Droplet-based high-throughput cultivation for accurate screening of antibiotic resistant gut microbes. eLife 9 , e56998 (2020). Hu, B. et al. High-throughput single-cell cultivation reveals the underexplored rare biosphere in deep-sea sediments along the Southwest Indian Ridge. Lab Chip 20 , 363–372 (2020). Jiang, M.-Z. et al. Droplet microfluidics-based high-throughput bacterial cultivation for validation of taxon pairs in microbial co-occurrence networks. Sci. Rep. 12 , 18145 (2022). Dai, J. et al. Microfluidic droplets with amended culture media cultivate a greater diversity of soil microorganisms. Appl. Environ. Microbiol. 91 , e01794-24 (2025). Karbaschi, M., Shahi, P. & Abate, A. R. Rapid, chemical-free breaking of microfluidic emulsions with a hand-held antistatic gun. Biomicrofluidics 11 , 044107 (2017). Cao, J. et al. Marinobacter profundi sp. nov., a slightly halophilic bacterium isolated from a deep-sea sediment sample of the New Britain Trench. Antonie van Leeuwenhoek 79 , 5315–5316 (2022). Dai, C. et al. Limited carbon cycling due to high-pressure effects on the deep-sea microbiome. Nat. Geosci. 15 , 1041–1047 (2022). Chaumeil, P.-A., Mussig, A. J., Hugenholtz, P. & Parks, D. H. GTDB-Tk v2: memory friendly classification with the genome taxonomy database. Bioinformatics 38 , 5315–5316 (2022). Yang, Z. et al. Cultivation strategies for prokaryotes from extreme environments. iMeta 2 , (2023). Shu, W.-S. & Huang, L.-N. Microbial diversity in extreme environments. Nat. Geosci. 20 , 219–235 (2022). Xiao, X. et al. Microbial ecosystems and ecological driving forces in the deepest ocean sediments. Cell 188 , 1363-1377.e9 (2025). Sun, Y. et al. A programmable droplet-based microfluidic device applied to multiparameter analysis of single microbes and microbial communities. Proc. Natl. Acad. Sci. U.S.A. 109 , 7665–7670 (2012). Coves, X. et al. The Mla system and its role in maintaining outer membrane barrier function in Stenotrophomonas maltophilia . Front. Cell. Infect. Microbiol. 14 , 1346565 (2024). Zheng, R., Wang, C., Cai, R., Shan, Y. & Sun, C. Mechanisms of nucleic acid degradation and high hydrostatic pressure tolerance of a novel deep-sea wall-less bacterium. mBio 14 , e00958-23 (2023). Wang, H., Zhang, Y., Bartlett, D. H. & Xiao, X. Transcriptomic analysis reveals common adaptation mechanisms under different stresses for moderately piezophilic bacteria. Microb. Ecol. 81 , 617–629 (2021). Qiu, X. et al. Transcriptomic Analysis Reveals that Changes in Gene Expression Contribute to Microbacterium sediminis YLB-01 Adaptation at Low Temperature Under High Hydrostatic Pressure. Curr. Microbiol. 38 , 5315–5316 (2022). Boedicker, J. Q., Vincent, M. E. & Ismagilov, R. F. Microfluidic confinement of single cells of bacteria in small volumes initiates high‐density behavior of quorum sensing and growth and reveals its variability. Angew. Chem., Int. Ed. 48 , 5908–5911 (2009). Collins, F. W. J. et al. The microbiome of deep-sea fish reveals new microbial species and a sparsity of antibiotic resistance genes. Gut Microbes 13 , 1–13 (2021). Li, J., Xiao, X., Zhou, M. & Zhang, Y. Strategy for the Adaptation to Stressful Conditions of the Novel Isolated Conditional Piezophilic Strain Halomonas titanicae ANRCS81. Appl. Environ. Microbiol. 89 , (2023). Qiu, X. et al. Metabolic adaptations of Microbacterium sediminis YLB-01 in deep-sea high-pressure environments. Appl. Microbiol. Biotechnol. 108 , 170 (2024). Mullane, K. K., Nishiyama, M., Kurihara, T. & Bartlett, D. H. Compounding deep sea physical impacts on marine microbial motility. Front. Mar. Sci. 10 , (2023). Smedile, F. et al. Adaptations to high pressure of Nautilia sp. strain PV-1, a piezophilic Campylobacterium (aka Epsilonproteobacterium) isolated from a deep-sea hydrothermal vent. Environ. Microbiol. 24 , 6164–6183 (2022). Ferreira, J. L. et al. γ-proteobacteria eject their polar flagella under nutrient depletion, retaining flagellar motor relic structures. PLOS Biol. 17 , e3000165 (2019). Sun, Y., Liu, Y., Pan, J., Wang, F. & Li, M. Perspectives on Cultivation Strategies of Archaea. Microb. Ecol. 109 , 7665–7670 (2020). Yang, Y., Zhao, W. & Xiao, X. The upper temperature limit of life under high hydrostatic pressure in the deep biosphere. Antonie van Leeuwenhoek 176 , 103604 (2021). Imachi, H. et al. Cultivation of previously uncultured microorganisms with a continuous-flow down-flow hanging sponge (DHS) bioreactor, using a syntrophic archaeon culture obtained from deep marine sediment as a case study. Nat. Protoc. 17 , 2784–2814 (2022). Zhuang, Y. et al. Paralabilibaculum antarcticum gen. nov., sp. nov., an anaerobic marine bacterium of the family Marinifilaceae isolated from Antarctica sea ice. Antonie van Leeuwenhoek 112 , 425–434 (2019). Zhang, Y. et al. Current developments in marine microbiology: high-pressure biotechnology and the genetic engineering of piezophiles. Mar. Technol. Soc. J. 33 , 157–164 (2015). Foustoukos, D. I. et al. High-pressure continuous culturing: life at the extreme. Appl. Environ. Microbiol. 91 , (2025). Cui, W. Development of the Jiaolong Deep Manned Submersible. Antonie van Leeuwenhoek 47 , 37–54 (2013). Zilionis, R. et al. Single-cell barcoding and sequencing using droplet microfluidics. Nat. Protoc. 12 , 44–73 (2017). Duffy, D. C., McDonald, J. C., Schueller, O. J. A. & Whitesides, G. M. Rapid prototyping of microfluidic systems in poly(dimethylsiloxane). Anal. Chem. 70 , 4974–4984 (1998). Wu, X. L., Xie, B. L., Qiao, Y. X., Yuan, S. & Du, W. B. μMET: A Novel Reusable Microfluidic Chip for Precision Microbial Enumeration Tests. Anal. Chem. 96 , 630–635 (2024). Kim, M., Oh, H.-S., Park, S.-C. & Chun, J. Towards a taxonomic coherence between average nucleotide identity and 16S rRNA gene sequence similarity for species demarcation of prokaryotes. Int. J. Syst. Evol. Microbiol. 64 , 346–351 (2014). Chen, S., Zhou, Y., Chen, Y. & Gu, J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics 34 , i884–i890 (2018). Magoč, T. & Salzberg, S. L. FLASH: fast length adjustment of short reads to improve genome assemblies. Bioinformatics 27 , 2957–2963 (2011). Edgar, R. C. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat. Methods 10 , 996–998 (2013). Li, D., Liu, C.-M., Luo, R., Sadakane, K. & Lam, T.-W. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics 31 , 1674–1676 (2015). Hyatt, D. et al. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinf. 11 , 119 (2010). Steinegger, M. & Söding, J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat. Biotechnol. 35 , 1026–1028 (2017). Segata, N. et al. Metagenomic microbial community profiling using unique clade-specific marker genes. Nat. Methods 9 , 811–814 (2012). Buchfink, B., Xie, C. & Huson, D. H. Fast and sensitive protein alignment using DIAMOND. Nat. Methods 12 , 59–60 (2015). Nurk, S., Meleshko, D., Korobeynikov, A. & Pevzner, P. A. metaSPAdes: a new versatile metagenomic assembler. Genome Res. 27 , 824–834 (2017). Uritskiy, G. V., DiRuggiero, J. & Taylor, J. MetaWRAP—a flexible pipeline for genome-resolved metagenomic data analysis. Microbiome 6 , 158 (2018). Parks, D. H., Imelfort, M., Skennerton, C. T., Hugenholtz, P. & Tyson, G. W. CheckM: assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Res. 25 , 1043–1055 (2015). Seemann, T. Prokka: rapid prokaryotic genome annotation. Bioinformatics 30 , 2068–2069 (2014). Li, R. et al. De novo assembly of human genomes with massively parallel short read sequencing. Genome Res. 20 , 265–272 (2010). Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9 , 357–359 (2012). Li, B. & Dewey, C. N. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. Antonie van Leeuwenhoek 12 , 323 (2011). Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15 , 550 (2014). Wu, T. et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation 2 , 100141 (2021). Additional Declarations There is NO Competing Interest. Supplementary Files DeepDropSOM.pdf Supplemantary Information ExtendedFigures.docx Cite Share Download PDF Status: Posted 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-7227821","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":512938192,"identity":"bcaaae47-14e2-4782-9e4e-990087679143","order_by":0,"name":"Wenbin Du","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAqElEQVRIiWNgGAWjYHACNoYEBhsekrWkkaqFgeEwCeoNjvcee/Bwx3kZ3Qbmhx8Yau4QoeXMuXSDxDO3ecwOsBlLMBx7RliL2Y0cM4nENpAWBjMGxgYiXGh2/w1IyzmgFvZvRGq5wQPScgCohYdIW+zP5JgbJLYl85gd5imWSDhGhBbJ9jNmD3+22dmbHW/f+OFDDSmhzcAMxAmkaBgFo2AUjIJRgBsAAO3MNwbtAMVhAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-7401-1410","institution":"Institute of Microbiology, Chinese Academy of Sciences","correspondingAuthor":true,"prefix":"","firstName":"Wenbin","middleName":"","lastName":"Du","suffix":""},{"id":512938193,"identity":"52f76e00-7140-4d9d-b448-0cf6b86b0e5e","order_by":1,"name":"Zhiyi Wang","email":"","orcid":"","institution":"Institute of Microbiology, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Zhiyi","middleName":"","lastName":"Wang","suffix":""},{"id":512938194,"identity":"5843ed81-5a37-45e9-8ae3-383da6d630ad","order_by":2,"name":"Tong Yu","email":"","orcid":"","institution":"Third Institute of Oceanography, Ministry of Natural Resources","correspondingAuthor":false,"prefix":"","firstName":"Tong","middleName":"","lastName":"Yu","suffix":""},{"id":512938195,"identity":"b8eb0f87-0d39-4230-a78a-166733c5bf6d","order_by":3,"name":"Linfeng Gong","email":"","orcid":"https://orcid.org/0000-0002-2561-9550","institution":"Third Institute of Oceanography, Ministry of Natural Resources","correspondingAuthor":false,"prefix":"","firstName":"Linfeng","middleName":"","lastName":"Gong","suffix":""},{"id":512938196,"identity":"5f8f53e6-4f5a-4a41-a22f-50304d0dc6fa","order_by":4,"name":"Haibin Qi","email":"","orcid":"","institution":"National Deep Sea Center, Ministry of Natural Resources","correspondingAuthor":false,"prefix":"","firstName":"Haibin","middleName":"","lastName":"Qi","suffix":""},{"id":512938197,"identity":"37035ca5-7fc6-43a6-8c2b-818afd64aa90","order_by":5,"name":"Beiyu Hu","email":"","orcid":"","institution":"Institute of Zoology, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Beiyu","middleName":"","lastName":"Hu","suffix":""},{"id":512938198,"identity":"37c9a192-ec3a-4936-868a-7c43f3e3e7d9","order_by":6,"name":"Xin Dai","email":"","orcid":"","institution":"Institute of Microbiology，Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Xin","middleName":"","lastName":"Dai","suffix":""},{"id":512938199,"identity":"e9e604a5-ef03-4cbb-8984-bd3c39e0ba03","order_by":7,"name":"Xiaowei Zheng","email":"","orcid":"https://orcid.org/0000-0001-9731-4476","institution":"Institute of Microbiology, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Xiaowei","middleName":"","lastName":"Zheng","suffix":""},{"id":512938200,"identity":"8d0d98f9-191d-4e59-9ba5-c180129e0324","order_by":8,"name":"Caiming Li","email":"","orcid":"","institution":"Institute of Microbiology, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Caiming","middleName":"","lastName":"Li","suffix":""},{"id":512938201,"identity":"3f3e7386-c799-4435-be33-39004a837feb","order_by":9,"name":"Sha Wang","email":"","orcid":"","institution":"Third Institute of Oceanography, Ministry of Natural Resources","correspondingAuthor":false,"prefix":"","firstName":"Sha","middleName":"","lastName":"Wang","suffix":""},{"id":512938202,"identity":"d622ec5b-e2b6-42d3-99c0-fcdeae51cd37","order_by":10,"name":"Xiaolin Wu","email":"","orcid":"","institution":"Institute of Microbiology, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Xiaolin","middleName":"","lastName":"Wu","suffix":""},{"id":512938203,"identity":"442eb27c-be58-4851-9d75-a137b5a5e511","order_by":11,"name":"Yan Sun","email":"","orcid":"","institution":"Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Yan","middleName":"","lastName":"Sun","suffix":""},{"id":512938204,"identity":"d0e6bc99-b32c-4dc1-b9c2-642b546e89bc","order_by":12,"name":"Zerui Wang","email":"","orcid":"","institution":"Institute of Microbiology, Institute of Microbiology","correspondingAuthor":false,"prefix":"","firstName":"Zerui","middleName":"","lastName":"Wang","suffix":""},{"id":512938205,"identity":"f17dfeba-e58e-4752-a75d-c98775197bb5","order_by":13,"name":"Gaoyuan Li","email":"","orcid":"","institution":"State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences","correspondingAuthor":false,"prefix":"","firstName":"Gaoyuan","middleName":"","lastName":"Li","suffix":""},{"id":512938206,"identity":"64c81b85-9846-4b01-a3ee-dc158523e461","order_by":14,"name":"Hailiang Dong","email":"","orcid":"https://orcid.org/0000-0002-7468-1350","institution":"China University of Geosciences, Beijing","correspondingAuthor":false,"prefix":"","firstName":"Hailiang","middleName":"","lastName":"Dong","suffix":""},{"id":512938207,"identity":"bc87c5df-ec02-4c7c-af09-4e174da6294a","order_by":15,"name":"Li Jiang","email":"","orcid":"https://orcid.org/0000-0002-3771-6518","institution":"Third Institute of Oceanography,Ministry of Natural Resources","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Jiang","suffix":""},{"id":512938208,"identity":"939d16d4-5230-43f0-aa99-970c9801e8f3","order_by":16,"name":"Xiang Xiao","email":"","orcid":"https://orcid.org/0000-0001-6785-7728","institution":"Shanghai Jiao Tong University","correspondingAuthor":false,"prefix":"","firstName":"Xiang","middleName":"","lastName":"Xiao","suffix":""},{"id":512938209,"identity":"665cf6a9-48df-454e-af88-4ad9a6d8b8f8","order_by":17,"name":"Li Huang","email":"","orcid":"","institution":"Institute of Microbiology","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Huang","suffix":""},{"id":512938210,"identity":"31d59856-6a23-44ea-b97f-d64ea0620d76","order_by":18,"name":"Hongliang Wang","email":"","orcid":"","institution":"National Deep Sea Center","correspondingAuthor":false,"prefix":"","firstName":"Hongliang","middleName":"","lastName":"Wang","suffix":""},{"id":512938211,"identity":"ed7ea187-e19e-48c5-8501-f95a97b5852f","order_by":19,"name":"Zongze Shao","email":"","orcid":"","institution":"Third Institute of Oceanography, Ministry of Natural Resources","correspondingAuthor":false,"prefix":"","firstName":"Zongze","middleName":"","lastName":"Shao","suffix":""}],"badges":[],"createdAt":"2025-07-27 18:10:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7227821/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7227821/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103368962,"identity":"61fe4417-5003-41e2-8ea1-1e239976a991","added_by":"auto","created_at":"2026-02-25 01:19:11","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1785422,"visible":true,"origin":"","legend":"\u003cp\u003eDeepDrop workflow for shipboard cultivation of deep-sea samples under high hydrostatic pressure (HHP). a, Deep-sea water and sediment samples were pretreated and encapsulated into droplets using a custom-built microfluidic device. The droplets were collected in syringes and subsequently pressurized in titanium vessels for single-cell cultivation under HHP. The middle panel shows representative microphotographs of droplets before and after 1 hour of incubation under 110 MPa (scale bar, 50 μm). b, Comparison of DeepDrop methods with bulk culture at HHP and droplet cultivation at atmospheric pressure. DeepDrop cultivations were reemulsified by pipetting and plated on different agar media; isolates were identified via 16S rRNA sequencing. Biomass under all cultivation conditions (DeepDrop 55 MPa, DeepDrop 28 MPa, Droplet 0.1 MPa, and Bulk 55 MPa) was demulsified for metagenomic sequencing, allowing for connections between pressure/confinement conditions, community composition, isolate recovery, and functional potential. The inset microphotograph shows the uniformity of the reemulsified double emulsions achieved by pipetting.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7227821/v1/9d7da4978c108e9c2dce7bea.png"},{"id":103506186,"identity":"d548a09b-94b0-4bdc-99c6-a8be3f4a051c","added_by":"auto","created_at":"2026-02-26 13:34:28","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":3016821,"visible":true,"origin":"","legend":"\u003cp\u003eValidation of DeepDrop's capacity for the selective enrichment of pressure-tolerant microbes.\u003cstrong\u003e \u003c/strong\u003ea,\u003cstrong\u003e \u003c/strong\u003eSchematic of DeepDrop's selection mechanism. A mock community containing the pressure-sensitive \u003cem\u003eE. coli\u003c/em\u003e (Strain A) and the pressure-tolerant \u003cem\u003eMarinobacter profundi\u003c/em\u003e (Strain B) was coencapsulated at the single-cell level and cultured under either 0.1 MPa or 50 MPa pressure. Changes in strain abundance and colony recovery were assessed to evaluate the effectiveness of selective enrichment. b, Representative droplet images show inhibited growth of E. coli and robust expansion of M. profundi under 50 MPa compared to 0.1 MPa and day 0 controls (scale bar, 50 µm). c, Non-metric multidimensional scaling (NMDS) plot based on Bray-Curtis dissimilarities of 16S rRNA gene profiles from three biological replicates of the original mixture and post-incubation droplets at 0.1 MPa and 50 MPa, indicating distinct community shifts under HHP. \u003cstrong\u003ed,\u003c/strong\u003e Bar plots of 16S rRNA-based relative abundance show \u003cem\u003eM. profundi\u003c/em\u003e dominance in 50 MPa cultures, in contrast to \u003cem\u003eE. coli\u003c/em\u003e overgrowth at 0.1 MPa. e, MALDI-TOF MS identification of isolates recovered from each condition confirms preferential isolation of \u003cem\u003eM. profundi\u003c/em\u003e at 50 MPa (n = 3 replicates).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7227821/v1/ef79481cb62e94eaf0b33708.png"},{"id":103507095,"identity":"d6de0cec-2b6a-4e95-afe8-50e4f0291ed3","added_by":"auto","created_at":"2026-02-26 13:40:24","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1771028,"visible":true,"origin":"","legend":"\u003cp\u003ePressure and confinement drive the selective enrichment of distinct microbial communities. a, Chao1 index comparing species richness (*, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05; ns, not significant). b, Simpson index comparing species evenness (**, \u003cem\u003ep\u003c/em\u003e\u0026lt; 0.01). c, Principal coordinate analysis (PCoA) showing significant clustering differences in microbial community compositions between DeepDrop and other conditions. d, Venn diagram illustrating the number of unique species identified under each cultivation condition. e, Bar chart showing the community composition of species uniquely present under 55 MPa bulk and 55 MPa DeepDrop conditions. f, Heatmap of the top 50 microbial taxa detected under four cultivation conditions: DeepDrop at 55 MPa (mimicking \u003cem\u003ein situ\u003c/em\u003e pressure), 28 MPa (half-pressure), 0.1 MPa (atmospheric), and bulk culture at 55 MPa. The color gradient indicates relative abundance, illustrating how HHP and microscale confinement drive species enrichment. g, Representative taxa that potentially respond to hydrostatic pressure. h, Confinement-induced shifts in the relative abundance of species enriched under DeepDrop conditions.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7227821/v1/91111b846bf78a71ba91e4bb.png"},{"id":103507185,"identity":"5cd88c4c-f4b2-4bb4-8095-baa91cff4106","added_by":"auto","created_at":"2026-02-26 13:40:40","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2938998,"visible":true,"origin":"","legend":"\u003cp\u003eFunctional adaptations of deep-sea microbes under DeepDrop culture conditions. a, KEGG pathway enrichment from metagenomes under four conditions (DeepDrop 55 MPa, Bulk 55 MPa, DeepDrop 28 MPa, Droplet 0.1 MPa) highlights significant shifts under DeepDrop 55 MPa (red dots). b, DeepDrop 55 MPa enriched genes related to osmoprotectant transport, membrane component transport, and autoinducer biosynthesis. c, Bulk 55 MPa cultures showed higher abundance of genes involved in aerotaxis, flagellar stator assembly, and polysaccharide biosynthesis.Statistical comparisons were performed via two-sided t tests (**p \u0026lt; 0.01; ***p \u0026lt; 0.001; ****p \u0026lt; 0.0001). d, PCoA of \u003cem\u003eHalomonas meridiana\u003c/em\u003eCR14 transcriptomes revealed distinct expression profiles under the four cultivation conditions. e,f, Volcano plots of DEGs in CR14 comparing DeepDrop 55 MPa with droplet 0.1 MPa (e) and DeepDrop 55 MPa with Bulk 55 MPa (f) show upregulation of genes involved in proteostasis, ROS removal, membrane repair, and osmoregulation. g, Schematic of five inferred adaptive strategies under DeepDrop high-pressure microenvironments: membrane stabilization, ROS detoxification, proteostasis, osmoregulation, and motility repression.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7227821/v1/d6e180a0a42a0afe7e8b6768.png"},{"id":103368968,"identity":"33e3cc42-4a7a-4e44-bb2f-a635bbcc250e","added_by":"auto","created_at":"2026-02-25 01:19:12","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":6941587,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDeepDrop enables recovery of rare and novel microbial taxa. a, \u003c/strong\u003ePhylogenetic tree and completeness heatmap of 54 metagenome-assembled genomes (MAGs; \u0026gt;80%) recovered across cultivation conditions. Blue squares indicate species-level MAGs (ANI ≥95%); purple squares denote novel MAGs (ANI \u0026lt;95%); white squares indicate absence. The functional gene matrix indicates the presence or absence of genes involved in osmoregulation, proteostasis, ROS detoxification, quorum sensing, motility, and chemotaxis. \u003cstrong\u003eb,\u003c/strong\u003e Phylogenetic tree and enrichment summary of 24 cultured genera from S1. Heatmaps show genus-level abundance in the original community and DeepDrop 55 MPa cultures. Novel genera are marked with red circles. \u003cstrong\u003ec,\u003c/strong\u003e Phylogenetic tree of 70 isolated species from samples S1–S3. Leaf colors denote phylum; outer tracks indicate sample of origin and number of isolates. Yellow circles mark rare taxa (\u0026lt;1% in S1); red stars indicate novel species. \u003cstrong\u003ed, \u003c/strong\u003eVenn diagram showing species distribution across five marine agar media types (2216E, 2216E+, R2A, YPD, LB; all diluted 1:10 in seawater).\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7227821/v1/8d74feb1023cf0dc26d52f5d.png"},{"id":103511325,"identity":"31ab2f4f-5627-4681-b952-4ee67f4f4df9","added_by":"auto","created_at":"2026-02-26 14:09:11","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":18057955,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7227821/v1/210e064e-4520-4aa2-a041-b153f7f4ec10.pdf"},{"id":103368967,"identity":"91ba0431-5c06-40a9-90c1-69e144cf7b39","added_by":"auto","created_at":"2026-02-25 01:19:12","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":8462070,"visible":true,"origin":"","legend":"Supplemantary Information","description":"","filename":"DeepDropSOM.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7227821/v1/4159f60398844e7e79e3c86e.pdf"},{"id":103368965,"identity":"98f7ac60-ddda-4ee8-8742-5ecd57aa1844","added_by":"auto","created_at":"2026-02-25 01:19:11","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":2703165,"visible":true,"origin":"","legend":"","description":"","filename":"ExtendedFigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-7227821/v1/4ea9d0591836da37dd5a3fac.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Microfluidic droplet cultivation under extreme pressure enables isolation and characterization of distinct deep-sea microbial dark matter","fulltext":[{"header":"Introduction","content":"\u003cp\u003eDeep-sea microorganisms thrive under high hydrostatic pressure (HHP), harbor diverse genetic and metabolic potential, and play pivotal roles in global biogeochemical cycling\u003csup\u003e1\u003c/sup\u003e, ecosystem functioning\u003csup\u003e2\u003c/sup\u003e, and long-term carbon storage\u003csup\u003e3\u003c/sup\u003e. Considerable efforts have focused on uncovering these microbes due to their promising biotechnological applications, including novel drug discovery, extreme-environment enzyme development, and bioremediation\u003csup\u003e3–5\u003c/sup\u003e. Although metagenomic surveys have significantly advanced our understanding of deep-sea microbial ecology and adaptive mechanisms\u003csup\u003e2,6\u003c/sup\u003e, more than 99% of deep-sea microbial taxa remain uncultivated, largely due to methodological limitations\u003csup\u003e7,8\u003c/sup\u003e. For example, depressurization during retrieval and delays in post-sampling handling frequently result in substantial loss of microbial viability and biodiversity\u003csup\u003e9\u003c/sup\u003e, underscoring the need for prompt shipboard cultivation strategies that mimic deep-sea conditions.\u003c/p\u003e\n\u003cp\u003eTraditional cultivation methods, such as plate spreading and dilution-to-extinction, fail to replicate extreme \u003cem\u003ein situ\u0026nbsp;\u003c/em\u003econditions and disproportionately favor fast-growing taxa. As a result, the number of novel taxa among isolates decreases with prolonged cultivation, and slow-growing or stress-tolerant species are consistently excluded from subsequent analyses\u003csup\u003e10\u003c/sup\u003e. Rare taxa are especially vulnerable to competitive suppression, as dominant microbes can rapidly deplete nutrients or secrete inhibitors that suppress neighboring cells, ultimately limiting the diversity that can be recovered. To address these barriers, several technical advances have emerged, including diffusion chambers that utilize \u003cem\u003ein situ\u003c/em\u003e conditions\u003csup\u003e7\u003c/sup\u003e, pressure-retaining bioreactors that simulate deep-ocean pressures\u003csup\u003e11\u003c/sup\u003e, and microfluidic technologies capable of isolating single microbial cells in controlled microscale environments\u003csup\u003e12,13\u003c/sup\u003e. Droplet microfluidics has proven to be a particularly transformative technique, enabling the growth of dormant or previously uncultivated taxa, even with standard media, by eliminating interspecific competition\u003csup\u003e14–16\u003c/sup\u003e. This approach has been used to recover a broader diversity of microbes from various environments\u003csup\u003e17–20\u003c/sup\u003e. However, several key limitations have prevented droplet-based methods from being effectively applied for deep-sea microbial cultivation.\u003c/p\u003e\n\u003cp\u003eMost importantly, the joint effects of deep-sea pressure\u0026nbsp;and microscale confinement on microbial growth, viability, and metabolic activation—factors critical for recovering\u0026nbsp;greater microbial diversity from the deep sea—remain unclear. Existing droplet systems are restricted to laboratory use, lack compatibility with high-pressure environments, and often rely on chemical demulsification, which may damage cells and reduce recovery efficiency\u003csup\u003e21\u003c/sup\u003e. These constraints not only limit our ability to simulate \u003cem\u003ein situ\u003c/em\u003e conditions but also hinder the retrieval of rare or pressure-adapted microbes. Thus, an integrated platform that maintains droplet integrity under deep-sea pressure, allows straightforward cell recovery, and can be deployed on shipboard to unlock deep-sea microbial dark matter is needed.\u003c/p\u003e\n\u003cp\u003eHere, we present DeepDrop, a droplet-based method designed for shipboard high-throughput single-cell cultivation of microbes under HHP. DeepDrop enables rapid shipboard processing while maintaining incubation pressures up to 110 MPa, effectively simulating deep-sea conditions. We validated DeepDrop by demonstrating droplet stability at 110 MPa, as well as the selective enrichment and isolation of pressure-adapted taxa from a model microbial community. When applied to deep-sea samples, DeepDrop yielded a more diverse microbial community with a distinct composition compared to conventional cultivation methods, as revealed by metagenomic analysis. Transcriptomic analyses of the most abundant species further revealed key adaptive mechanisms specific to DeepDrop. Finally, we demonstrated DeepDrop's ability to isolate rare and novel microbial species, underscoring its potential to cultivate previously uncultivated deep-sea taxa.\u003c/p\u003e"},{"header":"Results","content":"\u003ch3\u003eThe DeepDrop workflow for shipboard cultivation\u003c/h3\u003e\n\u003cp\u003eTo bridge the gap between deep-sea and shipboard cultivation conditions, we developed DeepDrop, an integrated method that combines droplet microfluidics and HHP incubation (Fig. 1a). The method involves a rapid shipboard workflow: deep-sea samples are pretreated and converted into cell suspensions within 20 minutes after sample collection. A portable, custom-built microfluidic instrument equipped with a flow-focusing chip (Supplementary Fig. 1) is subsequently used to encapsulate individual cells into picoliter droplets at 5×10\u003csup\u003e5\u003c/sup\u003e drops/min. These droplets are collected in syringes and transferred to a titanium alloy pressure vessel, where hydrostatic pressure is applied via syringe pistons. The pressure transmission within the system is verified using compressible foam, which deforms in response to the applied HHP (Supplementary Fig. 2a,b). Notably, the droplets remain stable without fusion even at hydrostatic pressures of up to 110 MPa, as validated in the laboratory to simulate the extreme conditions of the Mariana Trench (11,000 meters) (Supplementary Fig.\u0026nbsp;2c). Following HHP incubation, pipette-driven double emulsification is performed to convert droplets into water-in-oil-in-water emulsions, enabling linear dilution and direct plating of droplets onto diverse agar media without demulsification (Supplementary Fig.\u0026nbsp;3). This end-to-end workflow maintains microbial viability by preserving \u003cem\u003ein situ\u003c/em\u003e pressure levels, facilitating high-throughput isolation of novel pressure-adapted taxa.\u003c/p\u003e\n\u003cp\u003eTo evaluate the efficacy of DeepDrop, we compared microbial enrichment in cultures under various conditions: droplets at \u003cem\u003ein situ\u003c/em\u003e pressure, droplets at half \u003cem\u003ein situ\u0026nbsp;\u003c/em\u003epressure, droplets at atmospheric pressure, and bulk cultures under HHP conditions (Fig. 1b). After incubation, droplets were reemulsified into double emulsions and plated onto nutrient-diluted agar for strain isolation. Metagenomic sequencing was performed on all samples in parallel to analyze microbial composition and functional gene enrichment. This integrated approach establishes a direct link between cultivation environment, strain recovery, and functional adaptation, validating DeepDrop as a robust platform for accessing the deep-sea microbiome.\u003c/p\u003e\n\u003ch3\u003eDroplet\u0026nbsp;stability\u0026nbsp;under HHP\u0026nbsp;and\u0026nbsp;its\u0026nbsp;suitability for deep-sea microbe cultivation\u003c/h3\u003e\n\u003cp\u003eTo assess the structural robustness of droplets under prolonged high-pressure incubation, we tested droplets prepared with 2216E medium at simulated deep-sea pressures corresponding to depths of 4,000–8,000 meters (40–80 MPa), with atmospheric pressure (0.1 MPa) as a control. During a 7-day incubation onboard a research vessel, the pressure chambers were exposed to continuous ship-induced shaking. Droplet diameters remained unchanged across all conditions, confirming excellent structural stability under both high pressure and mechanical disturbance (Supplementary Fig. 2d, e). These results validate the suitability of DeepDrop for long-term deep-sea microbial cultivation under realistic field conditions.\u003c/p\u003e\n\u003cp\u003eWe next assessed DeepDrop's ability to selectively enrich pressure-adapted microbes using a mock community composed of the pressure-tolerant isolate \u003cem\u003eMarinobacter profundi\u003c/em\u003e\u003csup\u003e22\u003c/sup\u003e and pressure-sensitive \u003cem\u003eEscherichia coli\u003c/em\u003e (Fig. 2a). In pure cultures diluted to single cells, both strains proliferated under atmospheric pressure. In contrast, only \u003cem\u003eM. profundi\u003c/em\u003e exhibited robust growth at 50 MPa (Fig. 2b). In droplets containing the mixed community, 16S rRNA gene sequencing revealed that \u003cem\u003eE. coli\u003c/em\u003e dominated at 0.1 MPa, with \u003cem\u003eM. profundi\u003c/em\u003e comprising just 16.1% of the total population. At 50 MPa, however, DeepDrop selectively enriched \u003cem\u003eM. profundi\u003c/em\u003e, increasing its relative abundance to 91.6% (Fig. 2c,\u0026nbsp;d). Colony profiling using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) confirmed that significantly more \u003cem\u003eM. profundi\u003c/em\u003e colonies were recovered from DeepDrop cultures incubated at 50 MPa than from those cultivated at 0.1 MPa (Fig. 2e,\u0026nbsp;Supplementary Fig.\u0026nbsp;4). At atmospheric pressure, \u003cem\u003eE. coli\u003c/em\u003e overgrowth suppressed \u003cem\u003eM. profundi\u003c/em\u003e isolation, whereas high-pressure cultivation enabled successful recovery of this\u0026nbsp;pressure-tolerant strain. These results establish DeepDrop as a robust method for maintaining high-pressure integrity and selectively recovering HHP-adapted microbes from competitive communities.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHigh pressure and microscale confinement synergistically enhance taxonomic richness and evenness\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePrevious droplet-based cultivation studies have shown increased microbial richness compared to conventional methods under atmospheric conditions. However, it remains unknown whether combining this with in situ high pressure can further improve deep-sea microbial recovery. To address this, we applied DeepDrop to a near-bottom seawater sample collected from 5,500 m depth (S1,\u0026nbsp;Supplementary Table 1), cultivating it under four conditions: DeepDrop 55 MPa (\u003cem\u003ein situ\u003c/em\u003e pressure), DeepDrop 28 MPa (half of the \u003cem\u003ein situ\u003c/em\u003e pressure), droplet 0.1 MPa (atmospheric pressure), and conventional bulk culture at 55 MPa. All samples underwent metagenomic sequencing after cultivation. After cultivation, the samples were demulsified and pooled for metagenomic sequencing to reveal enriched microbial diversity.\u003c/p\u003e\n\u003cp\u003eDeepDrop 55 MPa outperformed bulk HHP cultivation in microbial richness, as measured by the Chao1 index, which estimates species richness (Fig. 3a). Concerning community structure, DeepDrop 55 MPa resulted in greater evenness than lower-pressure conditions did, as indicated by the Simpson index (Fig. 3b). Principal coordinate analysis (PCoA) further revealed distinct clustering of communities under DeepDrop 55 MPa conditions compared with other conditions (Fig. 3c). Notably, DeepDrop outperformed bulk HHP cultivation, recovering 135 species compared to the 89 species recovered with bulk culture; notably, 51 unique species were identified with DeepDrop, while only five unique species were identified with bulk culture (Fig. 3d, e). DeepDrop thus captured approximately 94.4% of the taxa recovered by the bulk methods while accessing additional unique species (Fig. 3d). At the genus level, DeepDrop at 55 MPa recovered 79 genera, representing a nearly 50% increase over the 54 genera identified by bulk cultivation at the same pressure, with 30 genera unique to DeepDrop compared with only five unique to the bulk (Extended Data Fig. 1a, Supplementary Table 2). More specifically, the predominant microbial taxa under DeepDrop 55 MPa were distinct from those under the other conditions (Fig. 3f), highlighting the synergistic effect of HHP and microscale confinement in accessing unique deep-sea microbial species.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeepDrop selectively enriches distinct microbial taxa from deep-sea samples\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMicrobial species can exhibit different sensitivities to varying hydrostatic pressures\u003csup\u003e23\u003c/sup\u003e. Our study revealed marked shifts in species abundances along a pressure gradient from 0.1 MPa to 28 MPa and 55 MPa (Fig. 3g). Notably, \u003cem\u003eAequorivita\u0026nbsp;\u003c/em\u003eSGB2713, \u003cem\u003ePsychrobacter oceani\u003c/em\u003e, and \u003cem\u003eHalomonas meridiana\u003c/em\u003e presented progressively increased relative abundance with rising pressure. Linear discriminant analysis effect size (LEfSe) further identified these species as the most significantly enriched in the droplet environment under HHP, highlighting them as potential indicator taxa of pressure-adapted communities (Supplementary Fig. 5a, b). Conversely, the abundances of \u003cem\u003eAlteromonas macleodii\u003c/em\u003e and \u003cem\u003eThalassospira indica\u003c/em\u003e decreased markedly, indicating pressure sensitivity or inhibition of proliferation. Meanwhile, species like \u003cem\u003ePseudomonas stutzeri\u003c/em\u003e displayed resilience to pressure variation, maintaining consistent abundances across conditions without significant shifts. Collectively, these findings demonstrate that integrating HHP with microscale confinement in DeepDrop enhances the recovery of unique pressure-adapted taxa, substantially expanding the recovery of deep-sea microbiota compared to conventional methods.\u003c/p\u003e\n\u003cp\u003eTo further dissect the effects of microscale confinement, we compared the relative abundance of each species in DeepDrop and bulk cultivation (Fig. 3h). Certain taxa, notably Alcanivorax xenomutans, Aequorivita SGB2713, and Psychrobacter oceani, showed substantial enrichment in DeepDrop, increasing by 91-, 20-, and 4.3-fold, respectively. Additionally, the 51 species exclusively detected in DeepDrop accounted for approximately 2.8% of the total community abundance, emphasizing the unique selective pressures imposed by confinement (Fig. 3d, e). This pattern was further supported by LEfSe analysis (Supplementary Fig. 5c, d). In contrast, some species thrived better in bulk 55 MPa, with \u003cem\u003eAlteromonas abrolhosensis\u003c/em\u003e and \u003cem\u003eSalipiger bermudensis\u003c/em\u003e showing the greatest reductions in DeepDrop 55 MPa, by 39- and 24-fold, respectively. These findings indicate that while confinement does not universally benefit all microbial taxa, it creates a distinct environment that selectively favors certain species, enabling the enrichment of microbes that would otherwise be unrecoverable in bulk cultures.\u003c/p\u003e\n\u003cp\u003eFinally, analysis of the top 10 genera, showing differential abundance across all cultivation conditions (Extended Data Fig. 1b), revealed a pronounced synergistic effect of high pressure and droplet confinement. Genera such as \u003cem\u003eAequorivita\u0026nbsp;\u003c/em\u003eSGB2713, \u003cem\u003ePsychrobacter oceani\u003c/em\u003e, and \u003cem\u003eH. meridiana\u003c/em\u003e, individually responsive to either pressure or confinement, exhibited markedly greater relative abundances under the combined DeepDrop 55 MPa condition. These findings collectively demonstrate that DeepDrop, by integrating extreme hydrostatic pressure and microscale confinement, effectively enriches distinct microbial taxa from hadal samples, significantly expanding access to previously uncultivated deep-sea microbiota.\u003c/p\u003e\n\u003ch3\u003eDeepDrop enriched deep-sea taxa exhibit unique functional adaptations\u003c/h3\u003e\n\u003cp\u003eTo elucidate the genetic mechanisms underlying microbial adaptation within DeepDrop, we further analyzed the metagenomic data of S1 across various cultivation conditions. KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway enrichment analysis revealed that genes associated with ATP-binding cassette (ABC) transporters and quorum sensing (QS) were uniquely and significantly enriched under DeepDrop 55 MPa (Fig. 4a). Additionally, species-specific analyses linked these enriched pathways to microbial taxa predominating under DeepDrop cultivation conditions (Supplementary Figs. 6–7), suggesting that these genes potentially contribute to microbial survival under DeepDrop cultivation conditions.\u003c/p\u003e\n\u003cp\u003eTo investigate the genetic basis for microbial adaptations under DeepDrop cultivation, we performed gene-level enrichment analysis on metagenomic data from cultures grown under various conditions. Genes encoding ABC transporters involved in osmoprotectant transport (e.g., betaine, glycerol, glycine) and membrane component trafficking (e.g., lipopolysaccharides, phospholipids) were particularly enriched under DeepDrop conditions (Fig. 4b). Additionally, genes associated with quorum sensing (QS) pathways, particularly those involved in inducer production and receptor synthesis common in marine bacteria, were significantly enriched in DeepDrop cultures at 55 MPa (Fig. 4b). In contrast, genes linked to chemotaxis, flagellar assembly, and biofilm formation pathways were more enriched in the bulk HHP cultures compared to DeepDrop 55 MPa and other droplet conditions (Fig. 4a). Specifically, genes encoding aerotaxis-related chemotaxis, flagellar stator components, and polysaccharide biosynthesis were more enriched in bulk HHP cultures (Fig. 4c), reflecting a selective advantage for motility and surface colonization under bulk conditions, likely driven by intensified interspecies competition and chemical gradients. Conversely, the spatial isolation and homogeneous nutrient conditions within picoliter droplets likely diminish the need for motility and environmental navigation, resulting in diminished investment in flagellar systems and chemotactic responses. These findings highlight distinct microbial adaptation strategies driven by the combined selective pressures of high hydrostatic pressure and microscale confinement in DeepDrop.\u003c/p\u003e\n\u003ch3\u003eTranscriptomic response of representative DeepDrop taxa under various microenvironments\u003c/h3\u003e\n\u003cp\u003eTo investigate the physiological adaptations of representative taxa under various conditions, we conducted transcriptomic profiling of \u003cem\u003eH. meridiana\u003c/em\u003e CR14, a dominant strain significantly enriched in DeepDrop 55 MPa compared to other conditions (Fig. 3f). Cultures were grown under four conditions: DeepDrop 55 MPa, bulk 55 MPa, droplet 0.1 MPa, and bulk 0.1 MPa. PCoA revealed a distinct transcriptomic profile for DeepDrop 55 MPa, indicating the combined influence of HHP and single-cell confinement (Fig. 4d).\u003c/p\u003e\n\u003cp\u003eTo investigate gene expression changes driven by HHP, we compared the transcriptomic profiles of H. meridiana CR14 cultured under DeepDrop 55 MPa and droplet 0.1 MPa conditions. This comparison revealed 594 upregulated and 565 downregulated genes, with a significant upregulation (P \u0026lt; 1e-50) in genes involved in proteostasis (e.g., chaperones, proteases), ROS detoxification (e.g., peroxidases), membrane stabilization (e.g., lipopolysaccharide transporters), and osmoregulation. These changes suggest a comprehensive stress response to HHP (Fig. 4e). Conversely, genes associated with flagellar assembly were significantly downregulated, indicating a shift away from energy-intensive motility processes under HHP (Extended Data Fig. 2a, c).\u003c/p\u003e\n\u003cp\u003eFurther investigation of the combined effects of HHP and microscale confinement was conducted by comparing the transcriptomic profiles of \u003cem\u003eH. meridiana\u003c/em\u003e CR14 grown under DeepDrop 55 MPa and bulk 55 MPa conditions. This comparison revealed 740 upregulated and 600 downregulated genes. The droplet condition showed reduced expression of ribosomal genes and increased transcription of genes related to alternative carbon metabolism, suggesting resource limitation within picoliter droplets (Fig. 4f, Extended Data Fig. 2b). Similar transcriptional shifts were also observed when comparing 0.1 MPa droplet conditions with bulk 0.1 MPa cultures (Supplementary Fig. 8), indicating that spatial confinement alone can drive significant changes in microbial metabolism.\u003c/p\u003e\n\u003cp\u003eTogether, these results demonstrate that deep-sea microbial taxa, such as \u003cem\u003eH. meridiana\u003c/em\u003e CR14, adapt to DeepDrop conditions by upregulating genes involved in proteostasis, oxidative stress mitigation, and the maintenance of membrane and osmotic stability, while downregulating genes related to energy-consuming motility systems (Fig. 4g). These transcriptomic changes correlate with the community-level trends observed in the metagenomics, further supporting the mechanistic basis of DeepDrop's ability to selectively enrich pressure-adapted taxa.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeepDrop enables the recovery of novel lineages through genome-resolved and culture-based approaches\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo assess the capacity of DeepDrop to recover novel microbial lineages, we employed both genome-resolved metagenomics and culture-based isolation following high-pressure droplet cultivation. These complementary strategies enabled us to assess how the system facilitates the discovery of previously inaccessible taxa from deep-sea environments.\u003c/p\u003e\n\u003cp\u003eMetagenomic binning across all cultivation conditions yielded 115 metagenome-assembled genomes (MAGs) with \u0026gt;50% completeness and \u0026lt;10% contamination (Supplementary Table S3). Among these MAGs, 53 met high-quality standards (≥80% completeness, ≤5% contamination) and were used for comparative analysis (Fig. 5a). DeepDrop at 55 MPa recovered 16 high-quality MAGs—more than those obtained from bulk HHP cultivation (n = 9) and slightly exceeding DeepDrop at 28 MPa (n = 15) and droplet cultures at atmospheric pressure (n = 13). Among the DeepDrop 55 MPa MAGs, three lacked genus-level classification based on Genome Taxonomy Database (GTDB)\u003csup\u003e24\u003c/sup\u003e, suggesting considerable novelty. Although total MAG counts were comparable across conditions, the partially distinct MAG profiles recovered via DeepDrop point to the selective enrichment of unique genomic lineages not captured by bulk approaches.\u003c/p\u003e\n\u003cp\u003eTo explore functional adaptations, we screened the high-quality MAGs for the presence or absence of key gene modules that had been previously identified as enriched at the community level. A subset of MAGs, particularly those from droplet-based cultivation, lacked chemotaxis and flagellar biosynthesis genes, consistent with the observed decrease in the enrichment of motility-related pathways in metagenomes from DeepDrop cultures (Fig. 4a,\u0026nbsp;Fig. 5a). These MAGs presented highly streamlined genomes, a pattern further supported by the disproportionately small genome sizes observed in droplet-based cultures (Supplementary Fig. 9). Together, these findings support the hypothesis that spatial confinement reduces selective pressure for motility, favoring stress adaptation.\u003c/p\u003e\n\u003cp\u003eTo test whether metagenomically enriched taxa could also be isolated in culture, we developed a droplet-to-colony pipeline that utilizes a pipette-driven double-emulsification step to directly plate incubated droplets. This simple approach enabled successful colony formation from droplets containing tens to hundreds of cells—a critical feature for recovering rare or slow-growing microbes. DeepDrop cultivation of sample S1 at 55 MPa followed by plating on five marine media yielded 176 isolates across 24 genera (Fig. 5b). Remarkably, 83% of these genera (20/24) were rare taxa (\u0026lt;1% relative abundance) in the original community. Several, including \u003cem\u003eAequorivita\u003c/em\u003e and \u003cem\u003ePsychrobacter\u003c/em\u003e, harbored novel species previously shown to be enriched in DeepDrop metagenomes (Extended Data Fig. 1b). These results demonstrate a strong correspondence between taxa enriched at the community level and those recoverable via cultivation, highlighting DeepDrop's capacity to bridge metagenomic discovery and culture recovery.\u003c/p\u003e\n\u003cp\u003eThe DeepDrop isolation protocol was applied to samples S2 and S3, in addition to sample S1, resulting in a total of 403 isolates. From these isolates, 346 high-quality, near-full-length 16S rRNA gene sequences were obtained, enabling the identification of 70 species spanning 33 genera and four phyla (Fig. 5c, Extended Data Fig. 3). Notably, four novel species (marked with red stars) were identified and characterized through morphological examination and phylogenetic analysis (Supplementary Figs. 10). Furthermore, 31 species isolated from S1 represented rare taxa, each with a relative abundance of less than 1% in the original community. These findings underscore DeepDrop's capacity to recover previously uncultivated, rare species from deep-sea environments, highlighting its potential for exploring and understanding microbial diversity in the deep sea.\"\u003c/p\u003e\n\u003cp\u003eTo increase recovery diversity, we employed five commonly used marine and laboratory culture media—2216E, 2216E+ (with 3.9% NaCl), R2A, YPD, and LB. These media were diluted to one-tenth strength in artificial seawater, offering a broad range of nutrients to better simulate marine conditions. The medium composition markedly influenced taxonomic richness, with R2A agar supporting the highest diversity, including 19 species recovered exclusively from this medium (Fig. 5d). Habitat-specific trends were also apparent: \u003cem\u003eH. meridiana\u003c/em\u003e, \u003cem\u003eHalopseudomonas aestusnigri\u003c/em\u003e, and \u003cem\u003ePseudomonas aeruginosa\u003c/em\u003e were dominant in samples S1, S2, and S3, respectively. These patterns might be influenced by the different depths, sample types, and collection methods, which together created varying environmental conditions across the samples. Finally, representative isolates such as \u003cem\u003eH. meridiana\u003c/em\u003e CR14, \u003cem\u003eAequorivita\u003c/em\u003e sp. CE311, \u003cem\u003ePsychrobacter\u003c/em\u003e sp. CR212, and \u003cem\u003eAlcanivorax xenomutans\u003c/em\u003e CRS8 displayed robust growth at both 0.1 and 55 MPa, confirming their HHP-tolerance (Supplementary Fig. 10e). These results illustrate the impact of media composition and environmental variability on microbial recovery, highlighting DeepDrop's capacity to access a broader diversity of deep-sea microbes across varied conditions.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eMicrobial communities in extreme environments, such as deep-sea ecosystems, present significant challenges for cultivation due to the high pressures and unique conditions that limit microbial growth and survival.\u003csup\u003e7–9,25,26\u003c/sup\u003e As an advanced platform that combines high-pressure incubation with droplet microfluidics, DeepDrop offers a robust solution for scaling and deploying microbial cultivation under extreme conditions. It enables the generation of picoliter-scale droplets capable of withstanding hydrostatic pressures up to 110 MPa, replicating the extreme conditions found in the deepest ocean\u003csup\u003e27\u003c/sup\u003e. By isolating individual cells within droplets, DeepDrop prevents overgrowth and interspecific competition, allowing slow-growing or dormant microbes to be cultivated effectively. Additionally, by directly plating double-emulsified droplets containing high-density cells grown from single cells, the system improves recovery efficiency while preventing the cytotoxicity associated with chemical demulsification. Thus, DeepDrop provides a versatile and efficient tool for cultivating and studying deep-sea microorganisms, offering valuable insights into their diversity and functional capabilities.\u003c/p\u003e\n\u003cp\u003eTo evaluate the platform's effectiveness in practice, we compared DeepDrop to conventional bulk high-pressure cultivation. We found that it markedly improves the recovery of microbial diversity from deep-sea samples. For example, cultivation of S1 at 55 MPa yielded 135 species under DeepDrop, substantially exceeding the 89 species retrieved using bulk methods. More importantly, the taxonomic profiles differed significantly between formats, suggesting that the unique interplay of hydrostatic pressure and microscale confinement fosters the selective enrichment of otherwise inaccessible microbial lineages. This trend was reinforced by pressure‒gradient experiments, which revealed differential species responses shaped by confinement and pressure. Genome-resolved analyses further supported these findings: DeepDrop 55 MPa recovered the highest number of high-quality MAGs, including three with no taxonomic assignment in the GTDB database. The limited overlap between novel MAGs recovered via DeepDrop and those from bulk HHP cultivation underscores the platform's capacity to uncover hidden phylogenetic and functional diversity. These findings build upon and extend earlier studies in high-pressure microbiology by demonstrating that physical compartmentalization\u003csup\u003e18,28\u003c/sup\u003e, when coupled with environmental simulation, enhances both the breadth and depth of microbial recovery.\u003c/p\u003e\n\u003cp\u003eBeyond expanding taxonomic breadth, DeepDrop offers mechanistic insight into the functional adaptations that enable microbial survival under high-pressure microscale confinement. Metagenomic analysis revealed enrichment of genes involved in osmoprotectant transport, membrane component stabilization, and quorum sensing, alongside depletion of genes linked to flagellar assembly, chemotaxis, and biofilm formation. These shifts reflect an ecological transition from motility-driven competition to survival-oriented strategies, favoring nonmotile phenotypes adapted to the constrained droplet environment. Traits such as osmoprotectant accumulation and phospholipid transport help maintain cellular homeostasis and membrane integrity under HHP\u003csup\u003e29–32\u003c/sup\u003e, while confined geometries likely accelerate quorum sensing activation by concentrating signaling molecules\u003csup\u003e33\u003c/sup\u003e. Supporting this interpretation, genome-level annotations of high-quality MAGs from droplet cultivation revealed a distinct subset lacking motility-related genes, consistent with reduced selection for active dispersal. To further explore this adaptation at the transcriptional level, we profiled \u003cem\u003eH. meridiana\u003c/em\u003e CR14, a dominant DeepDrop-enriched species. Transcriptomic analysis showed coordinated upregulation of stress response pathways, including proteostasis\u003csup\u003e34\u003c/sup\u003e, oxidative stress defense\u003csup\u003e35\u003c/sup\u003e, and membrane stabilization\u003csup\u003e36\u003c/sup\u003e. These changes were accompanied by suppression of motility gene expression\u003csup\u003e37,38\u003c/sup\u003e and reduced ribosomal activity, along with increased expression of pathways associated with alternative carbon metabolism\u003csup\u003e35\u003c/sup\u003e. These features are characteristic of a resource-efficient, low-growth physiological state and are consistent with previously reported adaptations in γ-Proteobacteria under pressure or nutrient limitation\u003csup\u003e39\u003c/sup\u003e. Together, these multiomic data demonstrate that DeepDrop selectively cultivates stress-resilient, nonmotile microbes whose transcriptional programs are tuned for energy conservation and environmental resilience under the dual constraints of pressure and spatial confinement (Fig. 4g).\u003c/p\u003e\n\u003cp\u003eA distinguishing innovation of DeepDrop is its pipette-based double-emulsion plating technique, which converts primary water-in-oil droplets into stable water-in-oil-in-water emulsions. This process prevents droplet coalescence, allows precise dilution, and enables even distribution on agar surfaces without the need for chemical demulsifiers. By preserving droplet integrity and enabling gentle transfer to solid media, this approach enhances the viability and recovery of rare or slow-growing strains. Applying this workflow, we successfully isolated 70 species from 346 colonies cultivated on standard marine media, including 12 isolates belonging to four novel species. Notably, many of these isolates originated from genera that were preferentially enriched under DeepDrop conditions, highlighting the platform's capacity to link environmental selection with strain-level cultivation. This outcome reflects a key advantage of plating droplets containing high-density outgrowths derived from single cells. Even when individual cells have a low probability of forming colonies, the cumulative probability within droplets is higher, and direct competition between species is minimized. As a result, slow-growing or low-abundance taxa have an improved chance of recovery. Together, these findings demonstrate that DeepDrop not only enables access to previously underrepresented deep-sea lineages but also serves as a practical tool for cultivating microbes that have long evaded conventional isolation efforts.\u003c/p\u003e\n\u003cp\u003eWhile DeepDrop effectively enhances the recovery of diverse and previously uncultivated microbial taxa, several limitations remain. The use of standard marine media likely constrained the diversity and novelty of isolates, suggesting that customized media formulations tailored to deep-sea nutritional and physicochemical conditions could further expand recovery\u003csup\u003e8,40\u003c/sup\u003e. Additionally, the current cultivation protocol employed uniform incubation conditions, including fixed temperature, oxygen levels, and duration, which may not capture the full range of physiological adaptations across deep-sea niches. Introducing variable temperature regimens\u003csup\u003e41\u003c/sup\u003e, extended cultivation periods\u003csup\u003e42\u003c/sup\u003e, and anaerobic conditions\u003csup\u003e43\u003c/sup\u003e may improve the isolation of specialized microbes. A notable challenge is the recovery of obligate piezophilic species\u003csup\u003e44\u003c/sup\u003e. Integrating pressure-retaining samplers and continuous high-pressure cultivation modules could address this gap\u003csup\u003e45\u003c/sup\u003e. Despite these constraints, DeepDrop offers a transformative platform for deep-sea microbiology, recreating high-pressure environments at single-cell resolution while enabling practical, shipboard-compatible workflows. Continued refinement of the system, including media optimization and pressure-retaining capabilities, will expand its utility for cultivating microbial \"dark matter.\" Ultimately, DeepDrop opens new avenues for exploring the ecological roles and biotechnological potential of microorganisms adapted to Earth's most extreme environments.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch3\u003eSample collection and preparation.\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThree deep-sea samples were collected from\u0026nbsp;the southern Mid-Atlantic Ridge on the Deep Sea No. 1 research vessel\u0026nbsp;in February 2024 using specialized equipment\u0026nbsp;(Supplementary Table 1). Near-bottom seawater (S1, 5500-meter depth) was collected with a CTD Rosette sampler, surface sediment (S2, 3300-meter depth) was obtained using a push core sampler deployed by the \u003cem\u003eJiaolong\u003c/em\u003e submersible\u003csup\u003e46\u003c/sup\u003e, and a 0.22 μm pore-size filter membrane (Millipore, Burlington, MA) with concentrated bacterial communities (S3, 2000-meter depth) was retrieved using an custom-built \u003cem\u003ein\u003c/em\u003e \u003cem\u003esitu\u003c/em\u003e seawater filtration system. For pretreatment, microbial cells from S1 were concentrated using a large volume concentration kit (Cat. No. CC01116, InnovaPrep, Drexel, MO). The sediment from S2 was suspended in sterile near-bottom seawater and shaken; the supernatant was then collected. In contrast, microbes from S3 were eluted from filter membranes in sterile seawater containing 10% glycerol. All samples were adjusted to ~10\u003csup\u003e7\u003c/sup\u003e cells/mL in 2216E medium before droplet generation, ensuring efficient single-cell encapsulation (~0.3 cells/droplet) as predicted by Poisson statistics.\u003c/p\u003e\n\u003ch3\u003eMicrofluidic chip fabrication.\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThe droplet-generation microfluidic chip was modified from a previous design to produce 30 pL droplets\u003csup\u003e47\u003c/sup\u003e (Supplementary Fig. 2). Soft lithography was used to fabricate chips with channel heights of 40 μm\u003csup\u003e48\u003c/sup\u003e. To render the channels hydrophobic, filtered Aquapel (PGW Auto Glass, Cranberry Twp, PA) was applied and incubated for 3 minutes, followed by flushing with fluorinated oil (Novec 7500, 3M, St. Paul, MN) and drying at 120°C for 10 minutes.\u003c/p\u003e\n\u003ch3\u003eDroplet cultivation under high hydrostatic pressures.\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eUsing the DeepDrop method, microbial cells were encapsulated into 30 pL droplets with fluorinated droplet-generation oil (Cat. No. 186-4006; Bio-Rad, Hercules, CA) at flow rates of 1600 μL/h (oil) and 800 μL/h (aqueous). Approximately 300 μL droplets were collected in 1 mL disposable plastic syringes and overlaid with mineral oil (Cat. No. M175243, Mreda, Beijing, China) to prevent coalescence during pressurization. The syringe needle was sealed with a butyl rubber plug and placed into a water-filled titanium alloy pressure vessel to simulate HHP conditions. For the bulk HHP culture controls, 300 μL of microbial suspension was directly added to syringes without droplet encapsulation. Atmospheric droplet cultures followed the same workflow without the application of pressure. All the cultures were incubated at 25°C. Notably, owing to the extraordinary stability of the carbon–fluorine bond, fluorinated oils are recalcitrant to microbial degradation and cannot serve as a carbon source.\u003c/p\u003e\n\u003ch3\u003eDouble emulsification and microbial isolation.\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eFollowing HHP incubation, the droplets were transferred to a centrifuge tube containing 1% Tween-80 in PBS (pH 7.4). Using a 1-mL pipette, the droplets were gently aspirated and dispensed at a rate of 20 cycles min\u003csup\u003e-1\u003c/sup\u003e, with microscopy examination ensuring completeness. The double emulsions were then diluted and spread onto various agar media, including 1/10 strength 2216E, R2A, LB, YPD, and 2216E supplemented with 3.9% NaCl (2216E+). The plates were incubated at 30°C to isolate microbial colonies.\u003c/p\u003e\n\u003ch3\u003eDeepDrop for mock community construction and validation of deep-sea strains.\u0026nbsp;\u003c/h3\u003e\n\u003ch3\u003eThe Mock community was constructed using pressure-adapted \u003cem\u003eMarinobacter profundi\u003c/em\u003e PWS21\u003csup\u003eT\u003c/sup\u003e isolated from a deep-sea sediment sample collected from the New Britain Trench\u003csup\u003e22\u003c/sup\u003e and pressure-sensitive \u003cem\u003eEscherichia coli\u003c/em\u003e ATCC 25922\u0026nbsp;obtained from the American Type Culture Collection (ATCC).\u0026nbsp;Both strains were cultured in 2216E broth at 30 °C. Cells were stained with 10 μM SYTO9 fluorescent dye\u0026nbsp;(Cat. No. FL0260, Toyephon, Beijing, China), and cell concentrations were quantified using a customized cell counting chip\u003csup\u003e49\u003c/sup\u003e. The two strains were then adjusted to 4 × 10\u003csup\u003e6\u003c/sup\u003e cells/mL, mixed, and encapsulated into picoliter droplets for cultivation under two pressure conditions: 50 MPa and atmospheric pressure (0.1 MPa). After a 3-day incubation, emulsions from each condition, along with the initial microbial mixture, were demulsified for high-throughput 16S rRNA gene amplicon analysis using next-generation sequencing to profile community composition. In parallel, emulsions from each condition were re-emulsified by pipetting to generate double emulsions, which were subsequently plated on 2216E agar to assess strain recovery. Each condition was tested in triplicate, and 30 strains were randomly selected from each replicate for identification.\u003c/h3\u003e\n\u003ch3\u003eIdentification of strains isolated via DeepDrop.\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eFor mock community isolates, bacterial identification was performed using an EXS 2600 MALDI-TOF MS system (Zybio, Chongqing, China) according to the manufacturer's protocol. For isolates from deep-sea samples, genomic DNA was extracted from single colonies via the DeepDrop single-cell cultivation method. The bacterial 16S rRNA gene was amplified via the universal primers 27F (5'-AGAGTTTGATCCTGGCTCAG-3') and 1492R (5'-GGTTACCTTGTTACGACTT-3'), purified, and sequenced on a 3730xl DNA Analyzer (Thermo Fisher Scientific, Wilmington, DE). The assembled sequences were analyzed using DNASTAR software and subjected to BLAST searches against the EzBioCloud database (http://www.ezbiocloud.net/). Isolates with \u0026lt;98.7% 16S rRNA sequence similarity and \u0026lt;95% ANI similarity to the closest reference genome were designated novel species\u003csup\u003e50\u003c/sup\u003e.\u003c/p\u003e\n\u003ch3\u003ePooled 16S rRNA sequencing of cultured communities.\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eGenomic DNA was extracted from DeepDrop cultures, atmospheric droplet cultures, and mock community samples using the FastPure Soil DNA Isolation Kit (MJYH, Shanghai, China). The V3-V4 region of the 16S rRNA gene was amplified with the primers 338F and 806R, prepared into libraries using the NEXTFLEX Rapid DNA-Seq Kit (Revvity, Waltham, MA), and sequenced on the Illumina NextSeq2000 platform (Illumina, San Diego, CA). The raw reads were quality-filtered with fastp\u003csup\u003e51\u003c/sup\u003e (v0.23.2) and merged using Flash\u003csup\u003e52\u003c/sup\u003e (v1.2.11). OTUs were clustered at 97% similarity using UPARSE\u003csup\u003e53\u003c/sup\u003e (v7.1) and taxonomically annotated using the Silva 16S rRNA database (v138) after rarefaction. Community structure differences were assessed via nonmetric multidimensional scaling (NMDS) based on Euclidean distance.\u003c/p\u003e\n\u003ch3\u003eMetagenomic profiling of deep-sea water samples under diverse culture conditions.\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eGenomic DNA was extracted by Guangdong MagiGene Biotechnology Co., Ltd. (Guangzhou, China) and used to construct sequencing libraries with the ALFA-SEQ DNA Library Prep Kit. Paired-end 150 bp sequencing was performed on the Illumina platform. Quality control of the raw reads was conducted using fastp (v0.23.2) to remove low-quality reads and adapters, resulting in high-quality clean reads. These reads were assembled into scaffolds using MEGAHIT\u003csup\u003e54\u003c/sup\u003e (v1.2.9), followed by fragmentation into scaftigs (≥500 bp) by removing ambiguous regions containing \"N\". Open reading frames (ORFs) were predicted with Prodigal\u003csup\u003e55\u003c/sup\u003e (v2.6.3), and nonredundant unigenes were clustered with MMseqs2\u003csup\u003e56\u003c/sup\u003e at 95% sequence identity. Gene abundance was quantified by mapping clean reads to the unigene catalog using BBMap. Taxonomic annotations were performed using MetaPhlAn\u003csup\u003e57\u003c/sup\u003e, leveraging species-specific marker genes, while functional annotations were conducted against the KEGG database using DIAMOND\u003csup\u003e58\u003c/sup\u003e. For MAG recovery, high-quality reads were assembled using SPAdes\u003csup\u003e59\u003c/sup\u003e (v3.15.5) using the '--meta' option. The MAGs were then binned using MetaWRAP\u003csup\u003e60\u003c/sup\u003e (v1.3.2), and their completeness and contamination were estimated using CheckM\u003csup\u003e61\u003c/sup\u003e (v1.2.2). Only MAGs with completeness ≥50% and contamination ≤ 10% were retained. Taxonomic classification of each MAG was performed with GTDB-Tk\u003csup\u003e24\u003c/sup\u003e, (v2.4.0) based on the GTDB taxonomy release 220, while protein-coding sequences (CDSs) were predicted using PROKKA\u003csup\u003e62\u003c/sup\u003e (v1.14.6). Finally, all the predicted proteins were annotated using the KEGG database with DIAMOND BLASTP (E-value =1e-5).\u003c/p\u003e\n\u003ch3\u003eWhole-genome sequencing and analysis of the enriched isolates.\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eGenomic DNA was isolated from bacterial cultures using a magnetic bead-based bacterial/fungal DNA extraction kit (Majorbio, Shanghai, China). DNA quality was evaluated with a Nanodrop spectrophotometer (ND-1000, Thermo Fisher), and gel electrophoresis was performed. Paired-end DNA libraries (PE150) were constructed using the NEXTFLEX Rapid DNA-Seq Kit (Revvity, Waltham, MA) and sequenced on an Illumina platform. Quality control of the raw reads was conducted with Fastp. Genome assembly was performed using SOAPdenovo (v2.04)\u003csup\u003e63\u003c/sup\u003e, and residual gaps in the scaffolds were closed with GapCloser (v1.12). Genes were predicted using Prodigal (v2.6.3).\u003c/p\u003e\n\u003ch3\u003eTranscriptomic analysis of \u003cem\u003eH. meridiana\u003c/em\u003e CR14 under different culture conditions.\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eTotal RNA from H. meridiana CR14 cultivated under different conditions was extracted by Guangdong MagiGene Biotechnology Co., Ltd. (Guangzhou, China), and its quality was assessed by agarose gel electrophoresis and a Qubit 3.0, with further validation using the Agilent 4200 system. RNA libraries were prepared with the ALFA-SEQ RNA Library Prep Kit II (MagiGene) and sequenced on the Illumina platform to generate paired-end 150 bp reads. Raw reads were processed with Fastp for quality control, and ribosomal RNA was filtered using Bowtie2\u003csup\u003e64\u003c/sup\u003e (v2.4.5) against the NCBI Rfam database. High-quality reads were mapped to the \u003cem\u003eH. meridiana\u003c/em\u003e CR14 genome with Bowtie2, and transcript abundance was calculated as TPM values using RSEM\u003csup\u003e65\u003c/sup\u003e (v1.3.3). Differentially expressed genes (DEGs) were identified with DESeq2\u003csup\u003e66\u003c/sup\u003e (v1.34.0) using thresholds of FDR ≤ 0.05 and |log2(fold change)| ≥ 1. KEGG pathway analyses of the DEGs were conducted using clusterProfiler\u003csup\u003e67\u003c/sup\u003e (v4.2.2). Statistical analyses were carried out on the MagiGene Cloud Platform (http://cloud.magigene.com).\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eData availability\u003c/h2\u003e\n\u003cp\u003eThe data that support the findings of this study are available from the corresponding author upon request. All sequencing data, including metagenomic, transcriptomic, 16S rRNA amplicon sequencing, and strain genome sequences, have been submitted to the National Microbial Data Center (NMDC) and are accessible under BioProject ID NMDC10019562 (https://nmdc.cncb.ac.cn/).\u003c/p\u003e\u003ch2\u003eAcknowledgments\u003c/h2\u003e\n\u003cp\u003eThis work was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB0810000) and the National Natural Science Foundation of China (92251302, 42376238).\u0026nbsp;We appreciate the support from the China Ocean Mineral Resources R\u0026amp;D Association (COMRA) and the National Deep Sea Center for organizing and supporting the deep-sea expedition (DY83). We thank the scientists and crew of the research vessel \u003cem\u003eDeep Sea No. 1\u003c/em\u003e, as well as the pilots of the \u003cem\u003eJiaoLong\u003c/em\u003e human-occupied vehicle (HOV),for their invaluable assistance in collecting the samples.\u003c/p\u003e\n\u003ch2\u003eAuthor contributions\u003c/h2\u003e\n\u003cp\u003eZ.W.: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing - original draft. T.Y.: Conceptualization, Investigation. L.G.: Investigation. H.Q.: Resources. B.H.: Conceptualization, Visualization. X.D.: Investigation. X.Z.: Investigation. C.L.: Visualization. S.W.: Investigation. X.W.: Formal analysis. Y.S.: Formal analysis. Z.W.: Investigation. G.L.: Resources. H.D.: Resources. L.J.: Formal analysis. X.X.: Resources, Writing - review \u0026amp; editing. L.H.: Writing - review \u0026amp; editing. H.W.: Conceptualization, Resources, Writing - review \u0026amp; editing. Z.S.: Conceptualization, Resources, Writing - review \u0026amp; editing. W.D.: Conceptualization, Formal analysis, Funding acquisition, Project administration, Supervision, Writing - review \u0026amp; editing.\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003ch2\u003eAdditional information\u003c/h2\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary information\u0026nbsp;\u003c/strong\u003eis available for this paper at https://doi.org/10.1038/XXXXX-X.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorrespondence and requests for materials\u003c/strong\u003e should be addressed to Wenbin Du.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eZeng, X., Alain, K. \u0026amp; Shao, Z. Microorganisms from deep-sea hydrothermal vents. \u003cem\u003eMar. Life Sci. Technol.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 204\u0026ndash;230 (2021).\u003c/li\u003e\n\u003cli\u003eDong, X. \u003cem\u003eet al.\u003c/em\u003e Evolutionary ecology of microbial populations inhabiting deep sea sediments associated with cold seeps. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eJiao, N. \u003cem\u003eet al.\u003c/em\u003e The microbial carbon pump and climate change. \u003cem\u003eNat. Rev. Microbiol.\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 408\u0026ndash;419 (2024).\u003c/li\u003e\n\u003cli\u003eZhou, Z., St. John, E., Anantharaman, K. \u0026amp; Reysenbach, A.-L. Global patterns of diversity and metabolism of microbial communities in deep-sea hydrothermal vent deposits. \u003cem\u003eMicrobiome\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 241 (2022).\u003c/li\u003e\n\u003cli\u003eJ\u0026oslash;rgensen, B. B. \u0026amp; Boetius, A. Feast and famine \u0026mdash; microbial life in the deep-sea bed. \u003cem\u003eNat. Rev. Microbiol.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 770\u0026ndash;781 (2007).\u003c/li\u003e\n\u003cli\u003eAnsorge, R. \u003cem\u003eet al.\u003c/em\u003e Functional diversity enables multiple symbiont strains to coexist in deep-sea mussels. \u003cem\u003eNat. Microbiol.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 2487\u0026ndash;2497 (2019).\u003c/li\u003e\n\u003cli\u003eKaeberlein, T., Lewis, K. \u0026amp; Epstein, S. S. Isolating \u0026lsquo;Uncultivable\u0026rsquo; Microorganisms in Pure Culture in a Simulated Natural Environment. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e296\u003c/strong\u003e, 1127\u0026ndash;1129 (2002).\u003c/li\u003e\n\u003cli\u003eFoustoukos, D. I. \u003cem\u003eet al.\u003c/em\u003e Cultivation of uncultured marine microorganisms. \u003cem\u003eMar Life Sci Technol\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 117\u0026ndash;120 (2021).\u003c/li\u003e\n\u003cli\u003eLa Cono, V. \u003cem\u003eet al.\u003c/em\u003e Shifts in the meso‐ and bathypelagic archaea communities composition during recovery and short‐term handling of decompressed deep‐sea samples. \u003cem\u003eEnviron. Microbiol. Rep.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 450\u0026ndash;459 (2015).\u003c/li\u003e\n\u003cli\u003eRinke, C. \u003cem\u003eet al.\u003c/em\u003e Insights into the phylogeny and coding potential of microbial dark matter. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e499\u003c/strong\u003e, 431\u0026ndash;437 (2013).\u003c/li\u003e\n\u003cli\u003eZeng, X. \u003cem\u003eet al.\u003c/em\u003e \u003cem\u003ePyrococcus\u003c/em\u003eCH1, an obligate piezophilic hyperthermophile: extending the upper pressure-temperature limits for life. \u003cem\u003eISME J.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 873\u0026ndash;876 (2009).\u003c/li\u003e\n\u003cli\u003eMa, L. \u003cem\u003eet al.\u003c/em\u003e Gene-targeted microfluidic cultivation validated by isolation of a gut bacterium listed in Human Microbiome Project\u0026rsquo;s Most Wanted taxa. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e111\u003c/strong\u003e, 9768\u0026ndash;9773 (2014).\u003c/li\u003e\n\u003cli\u003eHu, B. \u003cem\u003eet al.\u003c/em\u003e One cell at a time: droplet-based microbial cultivation, screening and sequencing. \u003cem\u003eMar. Life Sci. Technol.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 169\u0026ndash;188 (2021).\u003c/li\u003e\n\u003cli\u003eCarnes, E. C. \u003cem\u003eet al.\u003c/em\u003e Confinement-induced quorum sensing of individual Staphylococcus aureus bacteria. \u003cem\u003eNat. Chem. Biol.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 41\u0026ndash;45 (2010).\u003c/li\u003e\n\u003cli\u003eWeitz, M. \u003cem\u003eet al.\u003c/em\u003e Communication and computation by bacteria compartmentalized within microemulsion droplets. \u003cem\u003eJ. Am. Chem. Soc.\u003c/em\u003e \u003cstrong\u003e136\u003c/strong\u003e, 72\u0026ndash;75 (2014).\u003c/li\u003e\n\u003cli\u003eOrevi, T., S\u0026oslash;rensen, S. J. \u0026amp; Kashtan, N. Droplet size and surface hydrophobicity enhance bacterial plasmid transfer rates in microscopic surface wetness. \u003cem\u003eISME Commun.\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 72 (2022).\u003c/li\u003e\n\u003cli\u003eWatterson, W. J. \u003cem\u003eet al.\u003c/em\u003e Droplet-based high-throughput cultivation for accurate screening of antibiotic resistant gut microbes. \u003cem\u003eeLife\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, e56998 (2020).\u003c/li\u003e\n\u003cli\u003eHu, B. \u003cem\u003eet al.\u003c/em\u003e High-throughput single-cell cultivation reveals the underexplored rare biosphere in deep-sea sediments along the Southwest Indian Ridge. \u003cem\u003eLab Chip\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 363\u0026ndash;372 (2020).\u003c/li\u003e\n\u003cli\u003eJiang, M.-Z. \u003cem\u003eet al.\u003c/em\u003e Droplet microfluidics-based high-throughput bacterial cultivation for validation of taxon pairs in microbial co-occurrence networks. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 18145 (2022).\u003c/li\u003e\n\u003cli\u003eDai, J. \u003cem\u003eet al.\u003c/em\u003e Microfluidic droplets with amended culture media cultivate a greater diversity of soil microorganisms. \u003cem\u003eAppl. Environ. Microbiol.\u003c/em\u003e \u003cstrong\u003e91\u003c/strong\u003e, e01794-24 (2025).\u003c/li\u003e\n\u003cli\u003eKarbaschi, M., Shahi, P. \u0026amp; Abate, A. R. Rapid, chemical-free breaking of microfluidic emulsions with a hand-held antistatic gun. \u003cem\u003eBiomicrofluidics\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 044107 (2017).\u003c/li\u003e\n\u003cli\u003eCao, J. \u003cem\u003eet al.\u003c/em\u003e Marinobacter profundi sp. nov., a slightly halophilic bacterium isolated from a deep-sea sediment sample of the New Britain Trench. \u003cem\u003eAntonie van Leeuwenhoek\u003c/em\u003e \u003cstrong\u003e79\u003c/strong\u003e, 5315\u0026ndash;5316 (2022).\u003c/li\u003e\n\u003cli\u003eDai, C. \u003cem\u003eet al.\u003c/em\u003e Limited carbon cycling due to high-pressure effects on the deep-sea microbiome. \u003cem\u003eNat. Geosci.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1041\u0026ndash;1047 (2022).\u003c/li\u003e\n\u003cli\u003eChaumeil, P.-A., Mussig, A. J., Hugenholtz, P. \u0026amp; Parks, D. H. GTDB-Tk v2: memory friendly classification with the genome taxonomy database. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 5315\u0026ndash;5316 (2022).\u003c/li\u003e\n\u003cli\u003eYang, Z. \u003cem\u003eet al.\u003c/em\u003e Cultivation strategies for prokaryotes from extreme environments. \u003cem\u003eiMeta\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eShu, W.-S. \u0026amp; Huang, L.-N. Microbial diversity in extreme environments. \u003cem\u003eNat. Geosci.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 219\u0026ndash;235 (2022).\u003c/li\u003e\n\u003cli\u003eXiao, X. \u003cem\u003eet al.\u003c/em\u003e Microbial ecosystems and ecological driving forces in the deepest ocean sediments. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e188\u003c/strong\u003e, 1363-1377.e9 (2025).\u003c/li\u003e\n\u003cli\u003eSun, Y. \u003cem\u003eet al.\u003c/em\u003e A programmable droplet-based microfluidic device applied to multiparameter analysis of single microbes and microbial communities. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e109\u003c/strong\u003e, 7665\u0026ndash;7670 (2012).\u003c/li\u003e\n\u003cli\u003eCoves, X. \u003cem\u003eet al.\u003c/em\u003e The Mla system and its role in maintaining outer membrane barrier function in \u003cem\u003eStenotrophomonas maltophilia\u003c/em\u003e. \u003cem\u003eFront. Cell. Infect. Microbiol.\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 1346565 (2024).\u003c/li\u003e\n\u003cli\u003eZheng, R., Wang, C., Cai, R., Shan, Y. \u0026amp; Sun, C. Mechanisms of nucleic acid degradation and high hydrostatic pressure tolerance of a novel deep-sea wall-less bacterium. \u003cem\u003emBio\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, e00958-23 (2023).\u003c/li\u003e\n\u003cli\u003eWang, H., Zhang, Y., Bartlett, D. H. \u0026amp; Xiao, X. Transcriptomic analysis reveals common adaptation mechanisms under different stresses for moderately piezophilic bacteria. \u003cem\u003eMicrob. Ecol.\u003c/em\u003e \u003cstrong\u003e81\u003c/strong\u003e, 617\u0026ndash;629 (2021).\u003c/li\u003e\n\u003cli\u003eQiu, X. \u003cem\u003eet al.\u003c/em\u003e Transcriptomic Analysis Reveals that Changes in Gene Expression Contribute to Microbacterium sediminis YLB-01 Adaptation at Low Temperature Under High Hydrostatic Pressure. \u003cem\u003eCurr. Microbiol.\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 5315\u0026ndash;5316 (2022).\u003c/li\u003e\n\u003cli\u003eBoedicker, J. Q., Vincent, M. E. \u0026amp; Ismagilov, R. F. Microfluidic confinement of single cells of bacteria in small volumes initiates high‐density behavior of quorum sensing and growth and reveals its variability. \u003cem\u003eAngew. Chem., Int. Ed.\u003c/em\u003e \u003cstrong\u003e48\u003c/strong\u003e, 5908\u0026ndash;5911 (2009).\u003c/li\u003e\n\u003cli\u003eCollins, F. W. J. \u003cem\u003eet al.\u003c/em\u003e The microbiome of deep-sea fish reveals new microbial species and a sparsity of antibiotic resistance genes. \u003cem\u003eGut Microbes\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 1\u0026ndash;13 (2021).\u003c/li\u003e\n\u003cli\u003eLi, J., Xiao, X., Zhou, M. \u0026amp; Zhang, Y. Strategy for the Adaptation to Stressful Conditions of the Novel Isolated Conditional Piezophilic Strain \u003cem\u003eHalomonas titanicae\u003c/em\u003e ANRCS81. \u003cem\u003eAppl. Environ. Microbiol.\u003c/em\u003e \u003cstrong\u003e89\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eQiu, X. \u003cem\u003eet al.\u003c/em\u003e Metabolic adaptations of Microbacterium sediminis YLB-01 in deep-sea high-pressure environments. \u003cem\u003eAppl. Microbiol. Biotechnol.\u003c/em\u003e \u003cstrong\u003e108\u003c/strong\u003e, 170 (2024).\u003c/li\u003e\n\u003cli\u003eMullane, K. K., Nishiyama, M., Kurihara, T. \u0026amp; Bartlett, D. H. Compounding deep sea physical impacts on marine microbial motility. \u003cem\u003eFront. Mar. Sci.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eSmedile, F. \u003cem\u003eet al.\u003c/em\u003e Adaptations to high pressure of \u003cem\u003eNautilia\u003c/em\u003e sp. strain PV-1, a piezophilic Campylobacterium (aka Epsilonproteobacterium) isolated from a deep-sea hydrothermal vent. \u003cem\u003eEnviron. Microbiol.\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 6164\u0026ndash;6183 (2022).\u003c/li\u003e\n\u003cli\u003eFerreira, J. L. \u003cem\u003eet al.\u003c/em\u003e \u0026gamma;-proteobacteria eject their polar flagella under nutrient depletion, retaining flagellar motor relic structures. \u003cem\u003ePLOS Biol.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, e3000165 (2019).\u003c/li\u003e\n\u003cli\u003eSun, Y., Liu, Y., Pan, J., Wang, F. \u0026amp; Li, M. Perspectives on Cultivation Strategies of Archaea. \u003cem\u003eMicrob. Ecol.\u003c/em\u003e \u003cstrong\u003e109\u003c/strong\u003e, 7665\u0026ndash;7670 (2020).\u003c/li\u003e\n\u003cli\u003eYang, Y., Zhao, W. \u0026amp; Xiao, X. The upper temperature limit of life under high hydrostatic pressure in the deep biosphere. \u003cem\u003eAntonie van Leeuwenhoek\u003c/em\u003e \u003cstrong\u003e176\u003c/strong\u003e, 103604 (2021).\u003c/li\u003e\n\u003cli\u003eImachi, H. \u003cem\u003eet al.\u003c/em\u003e Cultivation of previously uncultured microorganisms with a continuous-flow down-flow hanging sponge (DHS) bioreactor, using a syntrophic archaeon culture obtained from deep marine sediment as a case study. \u003cem\u003eNat. Protoc.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 2784\u0026ndash;2814 (2022).\u003c/li\u003e\n\u003cli\u003eZhuang, Y. \u003cem\u003eet al.\u003c/em\u003e Paralabilibaculum antarcticum gen. nov., sp. nov., an anaerobic marine bacterium of the family Marinifilaceae isolated from Antarctica sea ice. \u003cem\u003eAntonie van Leeuwenhoek\u003c/em\u003e \u003cstrong\u003e112\u003c/strong\u003e, 425\u0026ndash;434 (2019).\u003c/li\u003e\n\u003cli\u003eZhang, Y. \u003cem\u003eet al.\u003c/em\u003e Current developments in marine microbiology: high-pressure biotechnology and the genetic engineering of piezophiles. \u003cem\u003eMar. Technol. Soc. J.\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 157\u0026ndash;164 (2015).\u003c/li\u003e\n\u003cli\u003eFoustoukos, D. I. \u003cem\u003eet al.\u003c/em\u003e High-pressure continuous culturing: life at the extreme. \u003cem\u003eAppl. Environ. Microbiol.\u003c/em\u003e \u003cstrong\u003e91\u003c/strong\u003e, (2025).\u003c/li\u003e\n\u003cli\u003eCui, W. Development of the \u003cem\u003eJiaolong\u003c/em\u003e Deep Manned Submersible. \u003cem\u003eAntonie van Leeuwenhoek\u003c/em\u003e \u003cstrong\u003e47\u003c/strong\u003e, 37\u0026ndash;54 (2013).\u003c/li\u003e\n\u003cli\u003eZilionis, R. \u003cem\u003eet al.\u003c/em\u003e Single-cell barcoding and sequencing using droplet microfluidics. \u003cem\u003eNat. Protoc.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 44\u0026ndash;73 (2017).\u003c/li\u003e\n\u003cli\u003eDuffy, D. C., McDonald, J. C., Schueller, O. J. A. \u0026amp; Whitesides, G. M. Rapid prototyping of microfluidic systems in poly(dimethylsiloxane). \u003cem\u003eAnal. Chem.\u003c/em\u003e \u003cstrong\u003e70\u003c/strong\u003e, 4974\u0026ndash;4984 (1998).\u003c/li\u003e\n\u003cli\u003eWu, X. L., Xie, B. L., Qiao, Y. X., Yuan, S. \u0026amp; Du, W. B. \u0026mu;MET: A Novel Reusable Microfluidic Chip for Precision Microbial Enumeration Tests. \u003cem\u003eAnal. Chem.\u003c/em\u003e \u003cstrong\u003e96\u003c/strong\u003e, 630\u0026ndash;635 (2024).\u003c/li\u003e\n\u003cli\u003eKim, M., Oh, H.-S., Park, S.-C. \u0026amp; Chun, J. Towards a taxonomic coherence between average nucleotide identity and 16S rRNA gene sequence similarity for species demarcation of prokaryotes. \u003cem\u003eInt. J. Syst. Evol. Microbiol.\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 346\u0026ndash;351 (2014).\u003c/li\u003e\n\u003cli\u003eChen, S., Zhou, Y., Chen, Y. \u0026amp; Gu, J. fastp: an ultra-fast all-in-one FASTQ preprocessor. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e, i884\u0026ndash;i890 (2018).\u003c/li\u003e\n\u003cli\u003eMagoč, T. \u0026amp; Salzberg, S. L. FLASH: fast length adjustment of short reads to improve genome assemblies. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 2957\u0026ndash;2963 (2011).\u003c/li\u003e\n\u003cli\u003eEdgar, R. C. UPARSE: highly accurate OTU sequences from microbial amplicon reads. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 996\u0026ndash;998 (2013).\u003c/li\u003e\n\u003cli\u003eLi, D., Liu, C.-M., Luo, R., Sadakane, K. \u0026amp; Lam, T.-W. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct \u003cem\u003ede Bruijn\u003c/em\u003e graph. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 1674\u0026ndash;1676 (2015).\u003c/li\u003e\n\u003cli\u003eHyatt, D. \u003cem\u003eet al.\u003c/em\u003e Prodigal: prokaryotic gene recognition and translation initiation site identification. \u003cem\u003eBMC Bioinf.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 119 (2010).\u003c/li\u003e\n\u003cli\u003eSteinegger, M. \u0026amp; S\u0026ouml;ding, J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 1026\u0026ndash;1028 (2017).\u003c/li\u003e\n\u003cli\u003eSegata, N. \u003cem\u003eet al.\u003c/em\u003e Metagenomic microbial community profiling using unique clade-specific marker genes. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 811\u0026ndash;814 (2012).\u003c/li\u003e\n\u003cli\u003eBuchfink, B., Xie, C. \u0026amp; Huson, D. H. Fast and sensitive protein alignment using DIAMOND. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 59\u0026ndash;60 (2015).\u003c/li\u003e\n\u003cli\u003eNurk, S., Meleshko, D., Korobeynikov, A. \u0026amp; Pevzner, P. A. metaSPAdes: a new versatile metagenomic assembler. \u003cem\u003eGenome Res.\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 824\u0026ndash;834 (2017).\u003c/li\u003e\n\u003cli\u003eUritskiy, G. V., DiRuggiero, J. \u0026amp; Taylor, J. MetaWRAP\u0026mdash;a flexible pipeline for genome-resolved metagenomic data analysis. \u003cem\u003eMicrobiome\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 158 (2018).\u003c/li\u003e\n\u003cli\u003eParks, D. H., Imelfort, M., Skennerton, C. T., Hugenholtz, P. \u0026amp; Tyson, G. W. CheckM: assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. \u003cem\u003eGenome Res.\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 1043\u0026ndash;1055 (2015).\u003c/li\u003e\n\u003cli\u003eSeemann, T. Prokka: rapid prokaryotic genome annotation. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 2068\u0026ndash;2069 (2014).\u003c/li\u003e\n\u003cli\u003eLi, R. \u003cem\u003eet al.\u003c/em\u003e De novo assembly of human genomes with massively parallel short read sequencing. \u003cem\u003eGenome Res.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 265\u0026ndash;272 (2010).\u003c/li\u003e\n\u003cli\u003eLangmead, B. \u0026amp; Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 357\u0026ndash;359 (2012).\u003c/li\u003e\n\u003cli\u003eLi, B. \u0026amp; Dewey, C. N. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. \u003cem\u003eAntonie van Leeuwenhoek\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 323 (2011).\u003c/li\u003e\n\u003cli\u003eLove, M. I., Huber, W. \u0026amp; Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. \u003cem\u003eGenome Biol.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 550 (2014).\u003c/li\u003e\n\u003cli\u003eWu, T. \u003cem\u003eet al.\u003c/em\u003e clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. \u003cem\u003eInnovation\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 100141 (2021).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7227821/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7227821/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Deep-sea microorganisms comprise the Earth's largest and least explored microbiome, yet the vast majority remain uncultivated due to challenges of preserving in situ high hydrostatic pressure and preventing loss of viability and diversity during recovery, which limits our ability to explore their ecological functions and adaptive strategies. Here, we introduce DeepDrop, a microfluidics platform that enables high-throughput single-cell cultivation under pressures spanning the full ocean depth directly aboard research vessels, following direct colony formation via pipette-generated double emulsions. Applying to hadal samples, DeepDrop recovered \u003e50% more microbial diversity than conventional high-pressure bulk cultivation, including rare taxa with streamlined genomes and distinctive genetic features associated with pressure adaptation. Combined metagenomic and transcriptomic analyses revealed that DeepDrop enriched pressure-adapted taxa carrying key stress-related genes and induced coordinated transcriptional reprogramming, characterized by upregulation of stress pathways and repression of motility. By integrating shipboard deployment, pressure-stable droplet cultivation, and efficient recovery, DeepDrop offers a powerful platform for accessing deep-sea microbial dark matter and illuminating microbial life strategies under extreme environmental constraints.","manuscriptTitle":"Microfluidic droplet cultivation under extreme pressure enables isolation and characterization of distinct deep-sea microbial dark matter","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-25 01:19:07","doi":"10.21203/rs.3.rs-7227821/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7c8b10e5-340f-4021-9ed5-afa3b3571291","owner":[],"postedDate":"February 25th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":54484334,"name":"Biological sciences/Biological techniques/Microbiology techniques"},{"id":54484335,"name":"Biological sciences/Biological techniques/Lab-on-a-chip"},{"id":54484336,"name":"Biological sciences/Microbiology/Environmental microbiology/Water microbiology"}],"tags":[],"updatedAt":"2026-02-25T01:19:07+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-25 01:19:07","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7227821","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7227821","identity":"rs-7227821","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}