Droplet microfluidic iX-seq platform enables discovery of gut bacterial cross-feeding between Phascolarctobacterium faecium and Eubacterium limosum

Droplet microfluidic iX-seq platform enables discovery of gut bacterial cross-feeding between Phascolarctobacterium faecium and Eubacterium limosum | 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 Droplet microfluidic iX-seq platform enables discovery of gut bacterial cross-feeding between Phascolarctobacterium faecium and Eubacterium limosum Kazuki Tanaka, Isaiah Song, Naoki Tanigawa, Mitsuko Komatsu, Chiharu Ishii, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8817168/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Background The human gut microbiota comprises hundreds of bacterial species that form a dynamic metabolic network, in which the exchange of microbial metabolites, known as cross-feeding, is integral to community function and host health. However, defining and isolating specific cross-feeding interactions remains challenging due to the limitations of conventional culture-based approaches. Droplet microfluidics enables high-throughput screening and isolation of bacteria from complex communities by encapsulating cells in water-in-oil droplets, allowing for parallel cultivation of thousands of stochastically contained microcultures. In this study, we developed a droplet microfluidics-based screening platform integrated with 16S rRNA gene amplicon sequencing, collectively termed iX-seq, to identify novel cross-feeding interactions directly from a human fecal sample. Results Using iX-seq, we identified Phascolarctobacterium faecium and Eubacterium limosum as potential unidirectional cross-feeders, wherein the growth of P. faecium was obligately supported by E. limosum in droplet co-cultures. The growth of P. faecium was modestly affected when co-cultured in vitro but notably required E. limosum co-inoculation for intestinal colonization in germ-free mice. Metabolomic profiling implicated 2-oxoglutarate (2-OG), an intermediate in P. faecium ’s succinate metabolic pathway, as the key differentiating metabolite between E. limosum -inoculated mouse and control mouse feces. Supplementation of 2-OG in monoculture increased P. faecium growth in a concentration-dependent manner, confirming its supportive role. Conclusions Our results reveal a previously unrecognized, potential 2-OG-mediated cross-feeding relationship between P. faecium and E. limosum , offering new insight into gut microbial metabolic dependencies. By enabling high-throughput, culture-based screening directly from fecal samples, the iX-seq approach demonstrates a practical framework for studying cross-feeding interactions within the gut microbiota and represents a potential paradigm shift in how complex microbial communities are experimentally investigated. Biological sciences/Biological techniques Biological sciences/Biotechnology Biological sciences/Microbiology droplet microfluidics gut microbiota cross-feeding Phascolarctobacterium faecium Eubacterium limosum 2-oxoglutarate Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Background The human gut microbiota is a dense, complex network of trillions of microorganisms, largely dominated by bacteria. It has been implicated in numerous facets of human health, including nutrient absorption, immune system development, and mental function 1 . The biological importance of the gut microbiota has thus been likened to the extent of an organ of the human body 2 . Furthermore, abnormalities in the balance of gut microbial species, referred to as dysbiosis, have been linked to disease states such as those associated with inflammatory bowel disease (IBD) 3 – 5 , type 2 diabetes 6 , 7 , atherosclerosis 8 , 9 , and colorectal cancer 10 , 11 . The increasingly documented benefits of maintaining a healthy gut microbiota and the physiological repercussions of dysbiosis emphasize the need to better understand our microscopic companions and how we can ensure a mutually beneficial relationship. The composition of the gut microbiota is both robust and dynamic, adapting to external influences such as diet and disease while preserving a relatively stable group of core taxa unique to each individual 12 , 13 . Microbial communities in the gut are characterized by diverse interactions that collectively form a complex ecological network 13 . This ecological balance is critical for the maintenance of a healthy gut microbial community. One of the most noteworthy interactions between gut bacteria is referred to as cross-feeding, in which metabolic byproducts produced by a bacterium are utilized as a nutrient source for one or more others. Cross-feeding is a common and well-established phenomenon in microbial communities, as evidenced by the syntrophy observed in defined groups of intestinal bacteria 14 . As an example, in one co-culture study, Anaerostipes caccae DSM 14662 could grow by utilizing the fructose produced by Bifidobacterium longum BB536 in an otherwise nutritionally incompatible environment 15 . Moreover, the presence of bacterial genes for metabolizing substrates absent from the environment, along with adaptive shifts in their expression during cultivation in mixed bacterial cultures, have led researchers to infer cross-feeding in informatics-based studies of microbial communities 16 – 18 . However, due to factors such as nutritional preferences, secretion and metabolism of both waste products and intermediate substrates 19 , and shifts in macronutrient bioavailability due to fluctuations in host diet 20 , 21 , the cross-feeding network is deceptively complex. This complexity poses a significant challenge to establishing a holistic framework for understanding the metabolic functionalities of a gut microbial community, both within the community itself and in its net output. Such characteristics profoundly impact host health and are promising targets for biomedical research. While attempts have been made to use in silico modeling of metabolic networks with metagenomic and metabolomic data to predict large-scale metabolic functionalities or disease associations within the gut microbiota 22 – 27 , further research is needed. Specifically, a reductionist approach exploring relationships between specific species coupled with wet-lab validation of in silico results will be necessary before these frameworks can reliably contextualize the complexity of gut microbial interactions. In feces, which are commonly used as a proxy for the gut microbiota, investigating the interactions between thousands of gut bacterial species using conventional low-throughput methods is conceivably time-consuming and inefficient. Data-driven approaches have therefore been revolutionary in developing our understanding of the gut microbiota, allowing researchers to predict functional potential, composition, and other characteristics of microbial populations based on metagenomics and other “-omics” data. By studying nucleic acids, metabolites, and other molecules associated with bacterial life, these approaches can even circumvent common wet-lab hurdles such as “unculturable” bacteria. Yet, this information fails to provide mechanistic evidence of biological phenomena such as cross-feeding, and interpretation of bioinformatics data largely relies on the painstakingly accumulated results of wet-lab research. Innovative strategies are thus necessary to bridge the gap between these approaches. A promising approach developed in recent years involves the utilization of droplet microfluidics, in which bacterial cells are randomly encapsulated in oil droplets to form water-in-oil microcultures containing as few as one cell per droplet. This approach creates a stochastically contained environment within each droplet, enabling the parallel cultivation of hundreds or even thousands of bacterial microcultures. Several such droplet-based techniques for cell cultivation have been described in the literature 28 – 30 and have been successfully utilized for the screening and analysis of complex microbial populations in recent years 31 – 33 . Recognizing the potential of this approach, we decided to utilize this technology in our exploration of gut microbial cross-feeding, henceforth referred to as iX-seq. Our goal was to identify and isolate novel cross-feeding relationships between gut bacteria using this droplet-based method of anaerobic bacterial cultivation (Fig. 1 ). To establish proof of concept that iX-seq can be used to screen for potential cross-feeders in a complex microbial community, we aimed to identify potential cross-feeding bacterial pairs from fecal samples. These pairs would consist of a “receiver” strain, whose survival depends on metabolic substrates supplied by another strain, and a “sender” strain, capable of providing said substrates. We optimized our process to maximize 1-to-2-cell cultures, wherein a receiver strain would be identified by its ability to proliferate in the presence of a sender strain and failure to survive on its own. Identified sender-receiver pairs would then be subjected to further analyses, including metabolomics analysis and in vivo transplantation, to confirm cross-feeding relationships and potential mechanisms. This unidirectional cross-feeding approach provided a basic framework for testing our methodology and evaluating its feasibility in investigating the diverse mechanisms of cross-feeding within the gut microbiota. Results Droplet microculture screening for unidirectional cross-feeding in the human gut microbiota Fresh human feces were collected from a healthy Japanese volunteer and processed for microfluidic droplet preparation, maximizing 1-2-cell microcultures. Representative fluorescence images of bacterial cell growth are shown in Fig. 2 B. The premise of our investigation was as follows: if a given bacterium was undetectable in one-species droplets but detectable in two-species droplets—in which the accompanying species was also detectable in one-species droplets—this was interpreted as indicative of a potential unidirectional cross-feeding relationship. Droplets that contained no cells or more than two species were excluded from the analysis due to the complexity of interpreting relationships in higher-order combinations of senders and receivers. By analyzing the droplet species compositions, we found that our droplet protocol was successful in focusing droplet compositions to 1 ~ 4 species (Fig. 2 C). After excluding the droplets that did not meet the aforementioned criteria, approximately ~ 65% of droplets remained as viable candidates for analysis. Surprisingly, we found that the one-species droplets consisted of only three different species: unidentified members of Eubacterium (809/932 one-species droplets), Escherichia - Shigella (122/932), and Ruminococcaceae UBA1819 (1/932). In all, 22 different bacteria were identified in the mono- and two-species droplets (Fig. 2 B). Using an observation rate threshold of 0.9 and requiring presence in at least three distinct droplets, a total of twelve potential receivers were identified (Table 1). Of these, we decided to focus on the pairing of two bacteria belonging to the genera Phascolarctobacterium and Eubacterium , for which a 16S rRNA sequence search using BLASTn identified them as Phascolarctobacterium faecium ACM 3679 (99.70% nt sequence identity, E-value = 7e-173) and Eubacterium limosum JCM 6421 (99.09% nt sequence identity, E-value = 6e-169). According to the literature, Phascolarctobacterium species can only utilize a narrow range of carbon sources and display a nutritional preference for succinate, which leads to the production of propionate 34 – 37 . There is medical interest in this taxon due to its reported ability to suppress growth of C. difficile in the human gut by reducing the luminal availability of succinate, which is also a utilizable energy source for C. difficile 35 . Receiver Sender Rel. freq. No. of droplets Enterobacter unclassified Escherichia / Shigella unclassified 1 247 Eggerthella uncultured Eubacterium unclassified 0.9885 86 Catabacter hongkongensis Eubacterium unclassified 1 62 Phascolarctobacterium Eubacterium unclassified 1 43 Bacteroides cellulosilyticus CL02T12C19 Eubacterium unclassified 1 31 Bacteroides fragilis Eubacterium unclassified 0.9355 29 Raoultibacter timonensis Eubacterium unclassified 1 20 Enterobacteriaceae unclassified Escherichia / Shigella unclassified 1 19 Bacteroides thetaiotaomicron Eubacterium unclassified 0.9375 15 Alistepes uncultured Eubacterium unclassified 1 10 Lachnospiraceae NK4A136 uncultured Eubacterium unclassified 1 3 Bacteroides unclassified Eubacterium unclassified 1 3 Going further, the Phascolarctobacterium faecium species specifically was reported to utilize succinate produced by Bacteroides thetaiotamicron 38 , implying that there is an established cross-feeding relationship between these two gut inhabitants. Coincidentally, many Eubacterium species are also known to produce succinate as a metabolic byproduct 39 , so we hypothesized that a similar metabolic exchange was occurring in the pairing of E. limosum and P. faecium. For these reasons, we elected to test these species for the presence of cross-feeding and validate the iX-seq methodology. In vitro examination of P. faecium and E. limosum as possible cross-feeding bacteria Our initial focus was to determine whether the observed cross-feeding phenomenon in the droplet cultures could be reproduced in vitro . We measured the growth of P. faecium co-cultured with E. limosum in mYCFA medium over a two-day time course. As a negative control, P. faecium was cultivated in monoculture with no additional substrates, while 1% (w/v) succinate-supplemented growth medium was also tested in recognition of the species’ aforementioned ability to metabolize succinate 35 , 37 . To measure growth, we recorded CFU/mL and DNA copy number. The CFU count represents the viable cell count (Fig. 3 A), while the DNA copy number estimates relative cell biomass (Fig. 3 B). The control P. faecium monoculture exhibited a steady decline in CFU count and almost no increase in DNA copy number over time, indicative of limited cell proliferation and gradual cell death. In contrast, the succinate group showed a marked increase in CFUs and biomass by Day 1, followed by a sharp decrease in both metrics by Day 2. This growth-collapse trajectory likely reflects nutrient depletion or rapid accumulation of toxic metabolic byproducts, both of which would result in the precipitous loss of viability observed on Day 2. The timeline aligns with a previous study of P. faecium ’s growth kinetics, reporting that OD 600 peaked between approximately 24 and 42 hours after inoculation in GAM supplemented with 1% (w/v) succinate, during which the succinate was also fully depleted 38 . Notably, that study did not observe a collapse, but rather a stationary phase after reaching the peak by 42 hours. This may have been due to the presence of alternative growth substrates or protective cofactors in GAM that mitigated the metabolic stress caused by succinate depletion or toxic metabolite buildup. This would not be possible in the less-complex mYCFA medium, as it lacks many of the cell-derived extracts and complex nutrients found in the richer GAM. Overall, we hypothesize that the observations in the succinate-supplemented cultures were indicative of rapid growth, given that succinate serves as a preferred energy source for P. faecium 35 , 37 . Interestingly, the P. faecium and E. limosum co-cultures did not show an increase in CFU count over time as we expected. The DNA copy number did not differ significantly from the negative control on Days 1 and 2 either, though it was modestly higher. However, there was a statistically significant attenuation of CFU reduction between the co-culture and negative control, suggesting that while E. limosum could not effectively sustain growth of P. faecium by itself, it attenuated the decrease in viable cell count as a result of succinate deficiency in the growth medium. From these results, it appeared that E. limosum was able to provide some metabolic benefit to P. faecium , though it is unclear whether succinate production was the causal mechanism. Mouse intestinal colonization model of P. faecium and E. limosum cross-feeding It is widely recognized that the intestinal environment, where gut microbes colonize, produces vastly different outcomes compared to in vitro research conditions. Nutrient availability, shifts in pH, abundance of host-derived molecules, and countless other factors create an environment that cannot easily be replicated outside of the body. Since cross-feeding occurs within the context of the intestines, the next logical step in our study was to validate the predicted cross-feeding relationship through transplantation into an in vivo mouse model. Germ-free (GF) BALB/c mice were initially inoculated with either sterile PBS or E. limosum and were observed for two weeks (Fig. 4 A-B). Then, both groups were further inoculated with P. faecium. Fecal samples from each mouse were collected each week and measured for CFU/g feces of P. faecium and E. limosum . Interestingly, it was observed that P. faecium was only able to successfully colonize the mouse intestines when co-inoculated with E. limosum . At Week 3, the E. limosum group showed high E. limosum and P. faecium viable cell counts at about 10 10 and 10 6 CFUs per gram of feces, respectively (Fig. 4 B). These numbers remained consistent throughout the fourth and final week of measurement. It was apparent that some characteristic(s) of E. limosum conferred P. faecium the ability to colonize the mouse intestines. Hypothesizing that cross-feeding underlies this phenomenon, we performed metabolome analysis of fecal samples from both PBS and E. limosum -inoculated mouse groups at Week 2 (prior to P. faecium inoculation) using principal component analysis (PCA) (Fig. 4 C-D). As expected, we found that the metabolome profiles of each group were compositionally distinct. A number of distinguishing features were identified, including several amino acids. However, the metabolite with the greatest influence on group separation was 2-oxoglutarate (2-OG), a molecule that is involved in the succinate metabolic pathway as a precursor to succinyl-CoA. Specifically, glutamate is hypothesized to be an alternative substrate to succinate in that it is intracellularly converted to 2-OG and subsequently to succinyl-CoA, a common intermediate predicted to be shared between succinate and glutamate in the succinate metabolic pathway of P. faecium38 (Fig. 5A). Indeed, 2-OG was found in high quantities in the feces of E. limosum-inoculated mice, while mice without E. limosum did not excrete any detectable 2-OG (Fig. 4E). As we know that P. faecium can metabolize succinate, presumably by this pathway, we predicted that 2-OG was also implicated in the ability of P. faecium to colonize the mouse through metabolic cross-feeding from E. limosum. Analysis of 2-OG as a cross-fed metabolite promoting P. faecium growth and colonization We next performed in vitro growth experiments using P. faecium and 2-OG in order to investigate what effects it would have on P. faecium growth in defined conditions. 2-OG showed a significant increase in P. faecium growth that scaled with higher concentrations, reflected in both increased DNA copy numbers and CFUs by Day 2 (Fig. 5 B-C). As 2-OG supplementation successfully promoted P. faecium growth, the data suggests that 2-OG production facilitating cross-feeding by E. limosum is at least one mechanism that may have contributed to colonization of P. faecium in the mouse intestines. Discussion In this study, we demonstrated the effectiveness of using iX-seq to supplement traditional methods of streak plating and colony isolation for screening of bacteria in complex environmental samples such as fecal samples. By applying this high-throughput methodology to bacterial communities, researchers can efficiently and rapidly identify bacteria of interest for their purposes. As shown in the results, we were able to identify a number of potential cross-feeding organisms in fecal samples by virtue of the ability of the benefactor “sender” strain to support the growth of a beneficiary “receiver” strain. Furthermore, we were able to support the existence of this interaction through in vitro culturing and further evidenced their relationship in a mouse model. This culminated in the identification of what we believe to be a previously unreported cross-feeding relationship between P. faecium and E. limosum , thereby demonstrating the potential of iX-seq in exploring the relationships of complex microbial communities. Through the progression of high-throughput screening, 16S rRNA gene sequencing, in vitro culture, in vivo colonization, and metabolomic analysis, we were able to discover the pairing of P. faecium and E. limosum as cross-feeding members of the gut microbiota that are hypothesized to exert growth-supportive effects through the production of 2-OG. However, it should be noted that the involvement of 2-OG is largely speculative, as we have not yet explored the mechanisms of 2-OG production and whether the metabolite was directly involved in P. faecium colonization of E. limosum -colonized GF mice. Nutrient supplementation via cross-feeding is only one of several possible explanations, as E. limosum may have modified the gut environment to favor P. faecium through alterations in luminal pH, redox potential, modulation of host-derived compounds, or other adjustment of physiochemical conditions. Furthermore, while E. limosum did appear to support P. faecium growth in droplet cultures, E. limosum ’s growth-supportive effects were comparatively limited when cultured in larger culture volumes in vitro , raising further questions as to what conditions are necessary for P. faecium to receive benefits from E. limosum . To elucidate the exact nature of their relationship, further investigation and mechanistic studies are necessary. However, as the initial goal of this study was to test the viability of iX-seq for research on cross feeding, further investigation into the mechanisms underlying this predicted cross-feeding relationship is a topic of future research. One obvious limitation of iX-seq that we realized was the limited diversity of bacteria that could be cultivated in single-species droplets, as well as the overall diversity of cultured bacteria. A key advantage of droplet microfluidics is believed to be the ability to isolate low-abundance bacteria from complex communities 40 . However, with only 22 species identified and very few able to grow in single-species droplets in our case, further development of our methodology is necessary to maximize species diversity and cell viability. As applications of this technology to microbiology are still in their infancy, there are various factors to consider such as medium composition, droplet stability, anaerobicity, and sample handling, for which there are no universal standards. While droplet microfluidics is a powerful and increasingly popular technique in microbial ecology, the wide range of available microfluidic devices and methods necessitates methodological development on a case-by-case basis 41 . And regrettably, as with traditional isolation methods, we must concede that certain bacteria may be unculturable, condemning these species to the confines of metagenomic data until advancements in bacterial culturing techniques facilitate their cultivation. Nevertheless, we aim to continue refining our own methodology to maximize the diversity of cultivable species and enable the observation of unknown cross-feeding interactions within the gut microbiota. In order to narrow our focus to the most straightforward approach to droplet microfluidics-based screening, we only conducted tests to identify unidirectional cross-feeding bacteria and excluded other possibilities in this study. However, droplet microfluidics may theoretically be applied to various cross-feeding scenarios as we continue to develop our methodology. At higher levels of complexity, cross-feeding can also be bidirectional (both parties provide substrates to each other) or multidirectional (multiple parties and exchanges of substrates are involved), illustrating the ecological interconnectedness of the gut metabolic network. There are also a number of possible mechanisms for cross-feeding, such as the secretion of small molecules and degradation of complex molecules by the sender strain 20 . While technically more complicated, exploration of such interactions may be achievable by using innovative approaches such as the development of a metabolite-detection pipeline integrated into the iX-seq method. Conclusion High-throughput methods such as iX-seq are invaluable for studying the complex microbial interactions within the human gut, enabling the evaluation of the thousands of species that constitute the gut microbiota. Using the iX-seq platform, we were able to isolate and provide evidence for a previously unknown cross-feeding relationship between P. faecium and E. limosum , also demonstrating the utility and potential of droplet microfluidics. Due to the increasing accessibility and efficiency of metagenomic sequencing, the mechanistic insights needed to contextualize the vast amounts of generated data are increasingly necessary, particularly for translating gut microbiota research into applications beyond the lab, such as in health and medicine. Methods Culture conditions Modified YCFA (mYCFA) medium was prepared as described in a previous study 42 . The medium was sparged with an anoxic gas mixture of 85% nitrogen, 5% hydrogen, and 10% carbon dioxide while heating, supplemented with L-cysteine hydrochloride monohydrate, and dispensed into 30 mL serum bottles at 10 mL volumes under anoxic gas flow. The bottles were sealed with butyl rubber stoppers and aluminum crimp seals to preserve anaerobicity and autoclaved at 121°C for 20 minutes. To initiate bacterial cultures, bottles were inoculated with the respective inocula via syringe and incubated at 37 ˚C without agitation. In the case of well plate incubation, well plates were incubated partially uncovered in a vinyl anaerobic chamber (Coy Laboratory Products, Inc., Grass Lake, MI, USA) at 37 ˚C. The chamber contained the same anoxic gas mixture as used for medium preparation. Fecal sample collection and preparation A fecal sample from a healthy Japanese volunteer was collected on the day of testing. Subjects with intestinal diseases, skin diseases, those who had taken antibiotics, or were allergic to the specified diet were excluded from consideration. All steps following fecal sample collection were performed in an anaerobic chamber containing an anoxic gas mixture of 85% nitrogen, 5% hydrogen, and 10% carbon dioxide. The fecal sample was homogenized in mYCFA as described in a previous study 42 at a concentration of 10% (w/v), filtered through a 40 µM cell strainer to remove large debris, and further diluted in mYCFA medium supplemented with 0.5% SeaPlaque agarose, resulting in a final concentration of 0.01% (v/v). To prevent the agarose from solidifying, the diluted sample was maintained at 37 ˚C using a heat block. Microfluidic droplet generation A Droplet Generator (On-chip Biotechnologies Co., Ltd, Tokyo, Japan) was used to generate water-in-oil microfluidic droplets from the fecal suspension with a dispensation rate of 0 ~ 9 cells per droplet. The oil used was a 5% 008-FluoroSurfactant formulation (008-FluoroSurfactant-5wtH-10mL (2003001); On-chip Biotechnologies Co., Ltd.). Each droplet had a diameter of approximately 60 µm and was generated following the manufacturer’s instructions. The droplets were collected in 1.5 mL microcentrifuge tubes and incubated at 37°C overnight in an anaerobic chamber. Following incubation, an equal volume of oil was added to each sample, and the tubes were cooled at 4°C for 15 minutes to allow the agarose to solidify. Subsequently, an equal volume of 1H,1H,2H,2H-perfluoro-1-octanol was added and mixed with the suspension to disrupt the oil layer. Then, 50 µL of T buffer (On-chip Biotechnologies Co., Ltd.) was added, and the mixture was centrifuged at 200 × g for 3 minutes at room temperature. The supernatant was collected, and the resulting gel beads (GBs) were recovered. These GBs were then dispensed into 384-well plates containing 100 µL of mYCFA per well using the Single-Particle Isolation and Sequencing (SPiS) instrument (On-chip Biotechnologies Co., Ltd.) and once more incubated overnight at 37°C in an anaerobic chamber. DNA barcoding and 16S rRNA gene sequencing After cultivation in the 384-well plates, 20 µL of each well culture were transferred to a new well plate, mixed with an equal volume of 20% (v/v) glycerol in PBS for a final concentration of 10% (v/v) glycerol, and stored at -80°C for future analysis. The remaining culture volume was used for unique 16S rRNA gene barcoding of bacterial DNA using the Echo 525 Liquid Handler. The 16S rRNA gene library was then sequenced using the Illumina MiSeq platform, and bacterial amplicon sequence variants (ASVs) were assigned using QIIME2. Droplet composition data was manually analyzed using R. Strain isolation from droplets The bacteria of interest were collected from their wells in the 384-well plate 10% (v/v) glycerol stocks and revitalized in liquid mYCFA medium. The bacteria were then cultured for two days at 37°C in an anaerobic chamber and subsequently passaged onto Gifu Anaerobic Medium (GAM; Nissui Pharmaceutical Co. Ltd., Tokyo, Japan) agar plates with or without vancomycin (20 µg/mL) to culture for four additional days until distinct colonies were visible. The purpose of vancomycin was to isolate P. faecium from two-species droplets, utilizing E. limosum ’s inherent sensitivity to the antibiotic and P. faecium ’s resistance. Individual colonies were then isolated and cultured in liquid mYCFA medium for an additional two days before mixing with an equal volume of 20% (v/v) glycerol in PBS for storage at -80°C at a final glycerol concentration of 10% (v/v). Mouse experiments Germ-free (GF) BALB/c mice (female, 6–20 weeks of age) were housed in gnotobiotic isolators and fed an autoclaved, standard rodent chow diet (CMF, Oriental Yeast Co., Ltd., Tokyo, Japan). P. faecium was cultured in GAM medium supplemented with 1% (w/v) succinate for 48 hours. After cultivation, the cultures were centrifuged at 10,000 × g for 5 minutes at room temperature, and the supernatant was discarded. The resulting cell pellets were resuspended in freshly prepared GAM medium. E. limosum was cultured in GAM medium for 24 hours and was processed in the same manner as P. faecium . The mice were inoculated with P. faecium or E. limosum cells by oral gavage of 200 µL of 1×10 8 colony-forming units (CFUs). During the 4-week experiment, feces were collected at weekly intervals for additional testing. All experiments were conducted in accordance with protocols approved by the University of Tsukuba Animal Experiment Committee. CE-TOFMS-based metabolome analysis Fecal metabolites were extracted from the samples and detected by capillary electrophoresis time-of-flight mass spectrometry (CE-TOFMS) using methods described in previous studies 43 , 44 . Briefly, fecal metabolites were extracted from 10 mg of freeze-dried feces by adding 500 µL of methanol containing 20 µM each of methionine sulfone (A17027; Alfa Aesar, Ward Hill, MA, USA) and D-camphol-10-sulfonic acid (CSA) (4987481429680; FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) as internal standards. and shaking vigorously with 100 mg of 0.1-mm zirconia/silica beads (BioSpec Products, Inc., Bartlesville, OK, USA) for 5 min at 1500 rpm using a Shake Master Neo (Bio Medical Science, Inc., Tokyo, Japan). Next, this mixture was mixed with 200 µL of ultrapure water and 500 µL of chloroform and shaken again. The suspension was centrifuged at 4,600 × g for 15 min at 4°C, and the resulting supernatant was transferred to a 5-kDa-cutoff filter column (Ultrafree MC-PLHCC 250/pk for Metabolome Analysis; Human Metabolome Technologies, Tsuruoka, Japan). The flow-through was dried under a vacuum, and the residue was then dissolved in 50 µL of Milli-Q water containing reference compounds (200 µM each of 3-aminopyrrolidine and trimesate). The levels of extracted metabolites were measured in both positive and negative modes by CE-TOFMS as described previously 45 . All CE-TOFMS experiments were performed using an Agilent capillary electrophoresis system (Agilent Technologies, Santa Clara, CA, USA). Raw data was analyzed using our proprietary automatic integration software MasterHands (ver. 2.16.0.15) 45 . Principal component analysis (PCA) was performed using the software MetaboAnalyst 6.0 46,47 . Metabolome data was normalized by sum, log 10 -transformed, and Pareto-scaled prior to analysis. Declarations Ethics approval and consent to participate This study was approved by the Ethical Committees of Keio University Shonan Fujisawa Campus (No. 355). All subjects were informed of the purpose of this study, and written consent was obtained from all subjects. This study was conducted with strict consideration for privacy. Consent for publication Not applicable. Availability of data and material Microbiome data has been deposited to DDBJ (accession number PRJDB37686), and metabolome data has been deposited to MetaboBank (accession number MTBKS266). Competing interests The authors declare no competing interests. Funding JSPS KAKENHI (22H03541 to SF), AMED-CREST (JP23gm1010009 to SF), JST ERATO (JPMJER1902 to SF), the Food Science Institute Foundation (to SF). Authors’ contributions SF, KT, and IS conceived the study and designed the experiments. KT, NT, and NO conducted animal experiments. NT and MK performed microfluidics experiments, metagenome and metabolome analyses. IS and NT conducted informatics analysis. SF, NO, JY, GN, CI, and KT provided critical review and commentary of the manuscript. All authors contributed to the article and approved the submitted version. Acknowledgements We wish to thank Noriko Kagata and Tatsuji Takahashi for their technical assistance. Figures were created or modified using BioRender.com. References Hou, K. et al. Microbiota in health and diseases. Signal Transduct Target Ther 7, 135 (2022). O’Hara, A. M. & Shanahan, F. The gut flora as a forgotten organ. EMBO Rep 7, 688–693 (2006). Nishida, A. et al. Gut microbiota in the pathogenesis of inflammatory bowel disease. Clin J Gastroenterol 11, 1–10 (2018). Ni, J., Wu, G. D., Albenberg, L. & Tomov, V. T. Gut microbiota and IBD: causation or correlation? Nat Rev Gastroenterol Hepatol 14, 573–584 (2017). Qiu, P. et al. The Gut Microbiota in Inflammatory Bowel Disease. Front Cell Infect Microbiol Volume 12-2022, (2022). Cunningham, A. L., Stephens, J. W. & Harris, D. A. Gut microbiota influence in type 2 diabetes mellitus (T2DM). Gut Pathog 13, 50 (2021). Zhou, Z., Sun, B., Yu, D. & Zhu, C. Gut Microbiota: An Important Player in Type 2 Diabetes Mellitus. Front Cell Infect Microbiol Volume 12-2022, (2022). Fusco, W. et al. Gut microbiota and atherosclerosis. Gut gutjnl-2025-335610 (2025) doi: 10.1136/gutjnl-2025-335610 . Jonsson, A. L. & Bäckhed, F. Role of gut microbiota in atherosclerosis. Nat Rev Cardiol 14, 79–87 (2017). Wong, C. C. & Yu, J. Gut microbiota in colorectal cancer development and therapy. Nat Rev Clin Oncol 20, 429–452 (2023). Kim, J. & Lee, H. K. Potential Role of the Gut Microbiome In Colorectal Cancer Progression. Front Immunol Volume 12-2021, (2022). Rinninella, E. et al. What is the Healthy Gut Microbiota Composition? A Changing Ecosystem across Age, Environment, Diet, and Diseases. Microorganisms 7, 14 (2019). Coyte, K. Z. & Rakoff-Nahoum, S. Understanding Competition and Cooperation within the Mammalian Gut Microbiome. Current Biology 29, R538–R544 (2019). Culp, E. J. & Goodman, A. L. Cross-feeding in the gut microbiome: Ecology and mechanisms. Cell Host Microbe 31, 485–499 (2023). Gwen, F., Angeliki, V., Kristof, V. & De, V. L. Cross-Feeding between Bifidobacterium longum BB536 and Acetate-Converting, Butyrate-Producing Colon Bacteria during Growth on Oligofructose. Appl Environ Microbiol 72, 7835–7841 (2006). Sander, S. et al. Mixed-Culture Transcriptome Analysis Reveals the Molecular Basis of Mixed-Culture Growth in Streptococcus thermophilus and Lactobacillus bulgaricus. Appl Environ Microbiol 76, 7775–7784 (2010). Garcia, S. L. et al. Auxotrophy and intrapopulation complementary in the ‘interactome’ of a cultivated freshwater model community. Mol Ecol 24, 4449–4459 (2015). Woyke, T. et al. Symbiosis insights through metagenomic analysis of a microbial consortium. Nature 443, 950–955 (2006). C, C. C. et al. Ecological Importance of Cross-Feeding of the Intermediate Metabolite 1,2-Propanediol between Bacterial Gut Symbionts. Appl Environ Microbiol 86, e00190-20 (2020). Smith, N. W., Shorten, P. R., Altermann, E., Roy, N. C. & McNabb, W. C. The Classification and Evolution of Bacterial Cross-Feeding. Front Ecol Evol 7, (2019). Oliphant, K. & Allen-Vercoe, E. Macronutrient metabolism by the human gut microbiome: major fermentation by-products and their impact on host health. Microbiome 7, 91 (2019). Hoek, M. J. A. van & Merks, R. M. H. Emergence of microbial diversity due to cross-feeding interactions in a spatial model of gut microbial metabolism. BMC Syst Biol 11, 56 (2017). Sung, J. et al. Global metabolic interaction network of the human gut microbiota for context-specific community-scale analysis. Nat Commun 8, 15393 (2017). Goyal, A., Wang, T., Dubinkina, V. & Maslov, S. Ecology-guided prediction of cross-feeding interactions in the human gut microbiome. Nat Commun 12, 1335 (2021). Proffitt, C. et al. Genome-scale metabolic modelling of the human gut microbiome reveals changes in the glyoxylate and dicarboxylate metabolism in metabolic disorders. iScience 25, 104513 (2022). Claudia, S.-A., María, R.-F., Daniel, G. & M, M. A. J. Using metabolic networks to predict cross-feeding and competition interactions between microorganisms. Microbiol Spectr 12, e02287-23 (2024). Goyal, A., Wang, T., Dubinkina, V. & Maslov, S. Ecology-guided prediction of cross-feeding interactions in the human gut microbiome. Nat Commun 12, 1335 (2021). Martin, K. et al. Generation of larger numbers of separated microbial populations by cultivation in segmented-flow microdevices. Lab Chip 3, 202–207 (2003). Clausell-Tormos, J. et al. Droplet-Based Microfluidic Platforms for the Encapsulation and Screening of Mammalian Cells and Multicellular Organisms. Chem Biol 15, 427–437 (2008). Liu, W., Kim, H. J., Lucchetta, E. M., Du, W. & Ismagilov, R. F. Isolation, incubation, and parallel functional testing and identification by FISH of rare microbial single-copy cells from multi-species mixtures using the combination of chemistrode and stochastic confinement. Lab Chip 9, 2153–2162 (2009). M, V. M. et al. Interindividual Variation in Dietary Carbohydrate Metabolism by Gut Bacteria Revealed with Droplet Microfluidic Culture. mSystems 5, 10.1128/msystems.00864 – 19 (2020). Watterson, W. J. et al. Droplet-based high-throughput cultivation for accurate screening of antibiotic resistant gut microbes. Elife 9, e56998 (2020). Hsu, R. H. et al. Microbial Interaction Network Inference in Microfluidic Droplets. Cell Syst 9, 229–242.e4 (2019). Yohei, W., Fumiko, N. & Masami, M. Characterization of Phascolarctobacterium succinatutens sp. nov., an Asaccharolytic, Succinate-Utilizing Bacterium Isolated from Human Feces. Appl Environ Microbiol 78, 511–518 (2012). Nagao-Kitamoto, H. et al. Interleukin-22-mediated host glycosylation prevents Clostridioides difficile infection by modulating the metabolic activity of the gut microbiota. Nat Med 26, 608–617 (2020). Stackebrandt, E. & Osawa, R. Phascolarctobacterium. in Bergey’s Manual of Systematics of Archaea and Bacteria 1–4 (2015). doi: https://doi.org/10.1002/9781118960608.gbm00700 . Del Dot, T., Osawa, R. & Stackebrandt, E. Phascolarctobacterium faecium gen. nov, spec. nov., a Novel Taxon of the Sporomusa Group of Bacteria. Syst Appl Microbiol 16, 380–384 (1993). Ikeyama, N. et al. Microbial interaction between the succinate-utilizing bacterium Phascolarctobacterium faecium and the gut commensal Bacteroides thetaiotaomicron. Microbiologyopen 9, e1111 (2020). Wade, W. G. Eubacterium. in Bergey’s Manual of Systematics of Archaea and Bacteria 1–36 (2015). doi: https://doi.org/10.1002/9781118960608.gbm00629 . Yin, J. et al. A droplet-based microfluidic approach to isolating functional bacteria from gut microbiota. Front Cell Infect Microbiol 12, (2022). He, Z., Wu, H., Yan, X. & Liu, W. Recent advances in droplet microfluidics for microbiology. Chinese Chemical Letters 33, 1729–1742 (2022). Nakanishi, Y. et al. Dynamic Omics Approach Identifies Nutrition-Mediated Microbial Interactions. J Proteome Res 10, 824–836 (2011). Mishima, E. et al. Evaluation of the impact of gut microbiota on uremic solute accumulation by a CE-TOFMS–based metabolomics approach. Kidney Int 92, 634–645 (2017). Soga, T. et al. Quantitative Metabolome Analysis Using Capillary Electrophoresis Mass Spectrometry. J Proteome Res 2, 488–494 (2003). Sugimoto, M., Wong, D. T., Hirayama, A., Soga, T. & Tomita, M. Capillary electrophoresis mass spectrometry-based saliva metabolomics identified oral, breast and pancreatic cancer-specific profiles. Metabolomics 6, 78–95 (2010). Pang, Z. et al. MetaboAnalyst 5.0: narrowing the gap between raw spectra and functional insights. Nucleic Acids Res 49, W388–W396 (2021). Xia, J. & Wishart, D. S. Metabolomic Data Processing, Analysis, and Interpretation Using MetaboAnalyst. Curr Protoc Bioinformatics 34, 14.10.1–14.10.48 (2011). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 29 Apr, 2026 Reviews received at journal 22 Mar, 2026 Reviewers agreed at journal 04 Mar, 2026 Reviewers invited by journal 04 Mar, 2026 Editor assigned by journal 19 Feb, 2026 Submission checks completed at journal 14 Feb, 2026 First submitted to journal 07 Feb, 2026 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-8817168","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":601671431,"identity":"dec45cc4-43b6-4596-bb3c-61201bb79dc9","order_by":0,"name":"Kazuki Tanaka","email":"","orcid":"","institution":"Keio University","correspondingAuthor":false,"prefix":"","firstName":"Kazuki","middleName":"","lastName":"Tanaka","suffix":""},{"id":601671432,"identity":"1a421391-fa12-4f3d-a505-6ff9a0c09ff0","order_by":1,"name":"Isaiah Song","email":"","orcid":"","institution":"Keio University","correspondingAuthor":false,"prefix":"","firstName":"Isaiah","middleName":"","lastName":"Song","suffix":""},{"id":601671435,"identity":"57546655-2657-43bd-80ac-bebc396e9095","order_by":2,"name":"Naoki Tanigawa","email":"","orcid":"","institution":"Keio University","correspondingAuthor":false,"prefix":"","firstName":"Naoki","middleName":"","lastName":"Tanigawa","suffix":""},{"id":601671436,"identity":"26749cbc-c5fb-4b62-ba6b-b1fddd10c951","order_by":3,"name":"Mitsuko Komatsu","email":"","orcid":"","institution":"Keio University","correspondingAuthor":false,"prefix":"","firstName":"Mitsuko","middleName":"","lastName":"Komatsu","suffix":""},{"id":601671437,"identity":"9f8430e9-ef0f-4381-9e85-121cd9c3a93c","order_by":4,"name":"Chiharu Ishii","email":"","orcid":"","institution":"Keio University","correspondingAuthor":false,"prefix":"","firstName":"Chiharu","middleName":"","lastName":"Ishii","suffix":""},{"id":601671438,"identity":"9f752fba-81f6-4bb2-81ad-818b76382e09","order_by":5,"name":"Gaku Nakato","email":"","orcid":"","institution":"Kanagawa Institute of Industrial Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Gaku","middleName":"","lastName":"Nakato","suffix":""},{"id":601671439,"identity":"421535de-4fbe-43dd-829b-f3f9a5b6fa1c","order_by":6,"name":"Jiayue Yang","email":"","orcid":"","institution":"Keio University","correspondingAuthor":false,"prefix":"","firstName":"Jiayue","middleName":"","lastName":"Yang","suffix":""},{"id":601671440,"identity":"95a8a7ad-cedd-4743-bb5e-03a851e61f7b","order_by":7,"name":"Nozomu Obana","email":"","orcid":"","institution":"University of Tsukuba","correspondingAuthor":false,"prefix":"","firstName":"Nozomu","middleName":"","lastName":"Obana","suffix":""},{"id":601671441,"identity":"5973434b-fd4c-40f9-8710-2606e7924870","order_by":8,"name":"Shinji Fukuda","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYDAC5gPMDAxsDHJABgNjA4MERFQCnxa2BLAWYyCDRC2JDRAtRAD5Nh5jY54yu/T5bcwPGGdUWMgxsB9+wGC5A7cWg2M8xsk855JzNxxjM2DccEbCmIEnzYBB8gweLfI9xod525hzN8j3MDA+bJNIbGDIYWCQbMPvMKCW+nQgA6jlH1AL/xv8WhhADuNtO5wAZDAwbmwAapEgYIvBMbZiwznnjhuC/HJwxjEJYzaJZwYH8PlFvo15s8Sbsmp5IOPhw56aOjl+/uSHjyXxhBgKOAAi2ID4sGQDkVrggPEjyVpGwSgYBaNgGAMAqdpHCdlWNgkAAAAASUVORK5CYII=","orcid":"","institution":"Keio University","correspondingAuthor":true,"prefix":"","firstName":"Shinji","middleName":"","lastName":"Fukuda","suffix":""}],"badges":[],"createdAt":"2026-02-07 17:24:42","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8817168/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8817168/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104212707,"identity":"46924eba-2707-4dd4-9490-e211ff21f949","added_by":"auto","created_at":"2026-03-09 08:14:05","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":126768,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOverview of iX-seq. iX-seq enables high-throughput screening of bacterial interactions by combining anaerobic co-culture, single-droplet sorting, and sequence-based identification\u003c/strong\u003e. Only droplets containing one or two species were retained for analysis in this study.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8817168/v1/e55792aa40a85b3efcd8c1d5.jpg"},{"id":104405171,"identity":"7ec7dcf5-bd56-45a6-83c8-a0676d9272c0","added_by":"auto","created_at":"2026-03-11 12:21:59","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":172589,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBacterial compositions in droplet microcultures\u003c/strong\u003e. \u003cstrong\u003e(A)\u003c/strong\u003e Fluorescence visualization of bacterial cells stained with FM4-64 dye after 0 and 2 days of incubation in microfluidic droplets. \u003cstrong\u003e(B)\u003c/strong\u003eHeatmap showing observed bacterial compositions in two-species droplet microcultures. For each \u003cem\u003ey\u003c/em\u003e-axis species, the heatmap shows (1) the frequency of co-occurrence with \u003cem\u003ex\u003c/em\u003e-axis species as a proportion of total droplets containing the \u003cem\u003ey\u003c/em\u003e-axis species, and (2) the absolute number of droplets in which each pair was observed. \u003cstrong\u003e(C)\u003c/strong\u003eRelative frequency distribution of species counts per droplet (2707 total droplets).\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8817168/v1/c444b634cb17931a4a18cb58.jpg"},{"id":104212710,"identity":"fc4af174-b22d-48c4-a997-41a5fce82bd8","added_by":"auto","created_at":"2026-03-09 08:14:06","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":71359,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eIn vitro\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e growth of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eP. faecium\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e in succinate-supplemented medium or co-cultured with \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eE. limosum\u003c/strong\u003e\u003c/em\u003e.\u0026nbsp; Bar plots represent \u003cstrong\u003e(A)\u003c/strong\u003e log\u003csub\u003e10\u003c/sub\u003e CFU/mL and \u003cstrong\u003e(B)\u003c/strong\u003e DNA copy number of \u003cem\u003eP. faecium\u003c/em\u003e cultures over a 2-day period under different conditions: Control (non-supplemented \u003cem\u003eP. faecium\u003c/em\u003e monoculture), Succinate (1% (w/v) succinate-supplemented \u003cem\u003eP. faecium\u003c/em\u003e monoculture), \u003cem\u003eand E. limosum\u003c/em\u003e (\u003cem\u003eP. faecium\u003c/em\u003e + \u003cem\u003eE. limosum\u003c/em\u003e co-culture).\u0026nbsp; Data shows mean ± standard deviation (\u003cem\u003en\u003c/em\u003e = 3).\u0026nbsp; Statistical significance was evaluated using Tukey-Kramer test (**\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01).\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8817168/v1/b12e27d04ce17aa8014b255e.jpg"},{"id":104212708,"identity":"8ea11088-7497-4db0-a5b8-cbfed35802f5","added_by":"auto","created_at":"2026-03-09 08:14:05","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":102928,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eIn vivo \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003egrowth of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eP. faecium \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003ein GF mice co-inoculated with \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eE. limosum\u003c/strong\u003e\u003c/em\u003e.\u0026nbsp; Overview of 4-week mouse experiment and log\u003csub\u003e10\u003c/sub\u003e CFU/g feces of \u003cem\u003eE. limosum \u003c/em\u003eand \u003cem\u003eP. faecium\u003c/em\u003e.\u0026nbsp; Germ-free (GF) BALB/c mice were initially inoculated with either \u003cstrong\u003e(A)\u003c/strong\u003e PBS or \u003cstrong\u003e(B)\u003c/strong\u003e \u003cem\u003eE. limosum\u003c/em\u003e at Week 0 and then were subsequently inoculated with \u003cem\u003eP. faecium\u003c/em\u003e after 2 weeks (\u003cem\u003en \u003c/em\u003e= 5).\u0026nbsp; \u003cstrong\u003e(A-B)\u003c/strong\u003e CFUs of \u003cem\u003eE. limosum \u003c/em\u003eand \u003cem\u003eP. faecium \u003c/em\u003ein feces were measured weekly.\u0026nbsp; \u003cstrong\u003e(C-D)\u003c/strong\u003e PCA of fecal metabolome data. \u003cstrong\u003e\u0026nbsp;(C)\u003c/strong\u003e Scores plot showing separation of metabolic profiles in feces from \u003cem\u003eE. limosum\u003c/em\u003e-inoculated GF mice and PBS-inoculated GF mice after two weeks and prior to \u003cem\u003eP. faecium\u003c/em\u003e inoculation.\u0026nbsp; \u003cstrong\u003e(D)\u003c/strong\u003e Loadings plot illustrating the contribution of individual metabolites to the principal components.\u0026nbsp; The most strongly discriminating metabolites between groups are located toward the left and right ends of the plot, and the top-4 discriminatory metabolites on each side are labeled. \u003cstrong\u003e\u0026nbsp;(E) \u003c/strong\u003eConcentrations of 2-oxoglutarate in feces from control and treatment group.\u0026nbsp; Statistical significance was evaluated using Tukey-Kramer test (***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001).\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8817168/v1/65c3c6e45980ec40cd729a9f.jpg"},{"id":104212709,"identity":"00c7e64b-0a92-4e66-bb67-dfbd4dbc9b02","added_by":"auto","created_at":"2026-03-09 08:14:05","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":69816,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eIn vitro P. faecium\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e growth in 2-oxoglutarate (2-OG)-supplemented medium\u003c/strong\u003e.\u0026nbsp; Bar plots show \u003cstrong\u003e(A)\u003c/strong\u003e log\u003csub\u003e10\u003c/sub\u003e CFU/mL and \u003cstrong\u003e(B)\u003c/strong\u003e DNA copy number of \u003cem\u003eP. faecium \u003c/em\u003ein cultures supplemented with increasing concentrations of 2-OG.\u0026nbsp; Statistical significance was evaluated using Tukey-Kramer test (**p \u0026lt; 0.01).\u0026nbsp; \u003cstrong\u003e(C) \u003c/strong\u003eSchematic overview of the succinate metabolic pathway in \u003cem\u003eP. faecium \u003c/em\u003eand the predicted involvement of 2-OG.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8817168/v1/e8febb26ea8cf656ce589b75.jpg"},{"id":104409064,"identity":"c9a25b6d-7d92-4df6-b211-89d5837b180b","added_by":"auto","created_at":"2026-03-11 12:44:06","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1672611,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8817168/v1/ad30f067-bda6-4d79-acce-d8ebd3ad37dd.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Droplet microfluidic iX-seq platform enables discovery of gut bacterial cross-feeding between Phascolarctobacterium faecium and Eubacterium limosum","fulltext":[{"header":"Background","content":"\u003cp\u003eThe human gut microbiota is a dense, complex network of trillions of microorganisms, largely dominated by bacteria. It has been implicated in numerous facets of human health, including nutrient absorption, immune system development, and mental function\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. The biological importance of the gut microbiota has thus been likened to the extent of an organ of the human body\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Furthermore, abnormalities in the balance of gut microbial species, referred to as dysbiosis, have been linked to disease states such as those associated with inflammatory bowel disease (IBD)\u003csup\u003e\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e, type 2 diabetes\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e, atherosclerosis\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e, and colorectal cancer\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. The increasingly documented benefits of maintaining a healthy gut microbiota and the physiological repercussions of dysbiosis emphasize the need to better understand our microscopic companions and how we can ensure a mutually beneficial relationship.\u003c/p\u003e \u003cp\u003eThe composition of the gut microbiota is both robust and dynamic, adapting to external influences such as diet and disease while preserving a relatively stable group of core taxa unique to each individual\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. Microbial communities in the gut are characterized by diverse interactions that collectively form a complex ecological network \u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. This ecological balance is critical for the maintenance of a healthy gut microbial community. One of the most noteworthy interactions between gut bacteria is referred to as cross-feeding, in which metabolic byproducts produced by a bacterium are utilized as a nutrient source for one or more others. Cross-feeding is a common and well-established phenomenon in microbial communities, as evidenced by the syntrophy observed in defined groups of intestinal bacteria\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. As an example, in one co-culture study, \u003cem\u003eAnaerostipes caccae\u003c/em\u003e DSM 14662 could grow by utilizing the fructose produced by \u003cem\u003eBifidobacterium longum\u003c/em\u003e BB536 in an otherwise nutritionally incompatible environment\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. Moreover, the presence of bacterial genes for metabolizing substrates absent from the environment, along with adaptive shifts in their expression during cultivation in mixed bacterial cultures, have led researchers to infer cross-feeding in informatics-based studies of microbial communities\u003csup\u003e\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. However, due to factors such as nutritional preferences, secretion and metabolism of both waste products and intermediate substrates\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e, and shifts in macronutrient bioavailability due to fluctuations in host diet\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e, the cross-feeding network is deceptively complex. This complexity poses a significant challenge to establishing a holistic framework for understanding the metabolic functionalities of a gut microbial community, both within the community itself and in its net output. Such characteristics profoundly impact host health and are promising targets for biomedical research. While attempts have been made to use \u003cem\u003ein silico\u003c/em\u003e modeling of metabolic networks with metagenomic and metabolomic data to predict large-scale metabolic functionalities or disease associations within the gut microbiota\u003csup\u003e\u003cspan additionalcitationids=\"CR23 CR24 CR25 CR26\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e, further research is needed. Specifically, a reductionist approach exploring relationships between specific species coupled with wet-lab validation of \u003cem\u003ein silico\u003c/em\u003e results will be necessary before these frameworks can reliably contextualize the complexity of gut microbial interactions.\u003c/p\u003e \u003cp\u003eIn feces, which are commonly used as a proxy for the gut microbiota, investigating the interactions between thousands of gut bacterial species using conventional low-throughput methods is conceivably time-consuming and inefficient. Data-driven approaches have therefore been revolutionary in developing our understanding of the gut microbiota, allowing researchers to predict functional potential, composition, and other characteristics of microbial populations based on metagenomics and other \u0026ldquo;-omics\u0026rdquo; data. By studying nucleic acids, metabolites, and other molecules associated with bacterial life, these approaches can even circumvent common wet-lab hurdles such as \u0026ldquo;unculturable\u0026rdquo; bacteria. Yet, this information fails to provide mechanistic evidence of biological phenomena such as cross-feeding, and interpretation of bioinformatics data largely relies on the painstakingly accumulated results of wet-lab research. Innovative strategies are thus necessary to bridge the gap between these approaches.\u003c/p\u003e \u003cp\u003eA promising approach developed in recent years involves the utilization of droplet microfluidics, in which bacterial cells are randomly encapsulated in oil droplets to form water-in-oil microcultures containing as few as one cell per droplet. This approach creates a stochastically contained environment within each droplet, enabling the parallel cultivation of hundreds or even thousands of bacterial microcultures. Several such droplet-based techniques for cell cultivation have been described in the literature\u003csup\u003e\u003cspan additionalcitationids=\"CR29\" citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e and have been successfully utilized for the screening and analysis of complex microbial populations in recent years\u003csup\u003e\u003cspan additionalcitationids=\"CR32\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. Recognizing the potential of this approach, we decided to utilize this technology in our exploration of gut microbial cross-feeding, henceforth referred to as iX-seq.\u0026nbsp;Our goal was to identify and isolate novel cross-feeding relationships between gut bacteria using this droplet-based method of anaerobic bacterial cultivation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo establish proof of concept that iX-seq can be used to screen for potential cross-feeders in a complex microbial community, we aimed to identify potential cross-feeding bacterial pairs from fecal samples. These pairs would consist of a \u0026ldquo;receiver\u0026rdquo; strain, whose survival depends on metabolic substrates supplied by another strain, and a \u0026ldquo;sender\u0026rdquo; strain, capable of providing said substrates. We optimized our process to maximize 1-to-2-cell cultures, wherein a receiver strain would be identified by its ability to proliferate in the presence of a sender strain and failure to survive on its own. Identified sender-receiver pairs would then be subjected to further analyses, including metabolomics analysis and \u003cem\u003ein vivo\u003c/em\u003e transplantation, to confirm cross-feeding relationships and potential mechanisms. This unidirectional cross-feeding approach provided a basic framework for testing our methodology and evaluating its feasibility in investigating the diverse mechanisms of cross-feeding within the gut microbiota.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eDroplet microculture screening for unidirectional cross-feeding in the human gut microbiota\u003c/h2\u003e \u003cp\u003eFresh human feces were collected from a healthy Japanese volunteer and processed for microfluidic droplet preparation, maximizing 1-2-cell microcultures. Representative fluorescence images of bacterial cell growth are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB. The premise of our investigation was as follows: if a given bacterium was undetectable in one-species droplets but detectable in two-species droplets\u0026mdash;in which the accompanying species was also detectable in one-species droplets\u0026mdash;this was interpreted as indicative of a potential unidirectional cross-feeding relationship. Droplets that contained no cells or more than two species were excluded from the analysis due to the complexity of interpreting relationships in higher-order combinations of senders and receivers.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eBy analyzing the droplet species compositions, we found that our droplet protocol was successful in focusing droplet compositions to 1\u0026thinsp;~\u0026thinsp;4 species (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). After excluding the droplets that did not meet the aforementioned criteria, approximately\u0026thinsp;~\u0026thinsp;65% of droplets remained as viable candidates for analysis.\u003c/p\u003e \u003cp\u003eSurprisingly, we found that the one-species droplets consisted of only three different species: unidentified members of \u003cem\u003eEubacterium\u003c/em\u003e (809/932 one-species droplets), \u003cem\u003eEscherichia\u003c/em\u003e-\u003cem\u003eShigella\u003c/em\u003e (122/932), and Ruminococcaceae UBA1819 (1/932). In all, 22 different bacteria were identified in the mono- and two-species droplets (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Using an observation rate threshold of 0.9 and requiring presence in at least three distinct droplets, a total of twelve potential receivers were identified (Table\u0026nbsp;1). Of these, we decided to focus on the pairing of two bacteria belonging to the genera \u003cem\u003ePhascolarctobacterium\u003c/em\u003e and \u003cem\u003eEubacterium\u003c/em\u003e, for which a 16S rRNA sequence search using BLASTn identified them as \u003cem\u003ePhascolarctobacterium faecium\u003c/em\u003e ACM 3679 (99.70% nt sequence identity, E-value\u0026thinsp;=\u0026thinsp;7e-173) and \u003cem\u003eEubacterium limosum\u003c/em\u003e JCM 6421 (99.09% nt sequence identity, E-value\u0026thinsp;=\u0026thinsp;6e-169). According to the literature, \u003cem\u003ePhascolarctobacterium\u003c/em\u003e species can only utilize a narrow range of carbon sources and display a nutritional preference for succinate, which leads to the production of propionate\u003csup\u003e\u003cspan additionalcitationids=\"CR35 CR36\" citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. There is medical interest in this taxon due to its reported ability to suppress growth of \u003cem\u003eC. difficile\u003c/em\u003e in the human gut by reducing the luminal availability of succinate, which is also a utilizable energy source for \u003cem\u003eC. difficile\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReceiver\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSender\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRel. freq.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNo. of droplets\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eEnterobacter\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEscherichia\u003c/em\u003e/\u003cem\u003eShigella\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e247\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eEggerthella\u003c/em\u003e uncultured\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEubacterium\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.9885\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e86\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCatabacter hongkongensis\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEubacterium\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e62\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePhascolarctobacterium\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEubacterium\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e43\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eBacteroides cellulosilyticus\u003c/em\u003e CL02T12C19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEubacterium\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eBacteroides fragilis\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEubacterium\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.9355\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e29\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eRaoultibacter timonensis\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEubacterium\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEnterobacteriaceae unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEscherichia\u003c/em\u003e/\u003cem\u003eShigella\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eBacteroides thetaiotaomicron\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEubacterium\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.9375\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAlistepes\u003c/em\u003e uncultured\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEubacterium\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLachnospiraceae NK4A136 uncultured\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEubacterium\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eBacteroides\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eEubacterium\u003c/em\u003e unclassified\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eGoing further, the \u003cem\u003ePhascolarctobacterium faecium\u003c/em\u003e species specifically was reported to utilize succinate produced by \u003cem\u003eBacteroides thetaiotamicron\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e, implying that there is an established cross-feeding relationship between these two gut inhabitants. Coincidentally, many \u003cem\u003eEubacterium\u003c/em\u003e species are also known to produce succinate as a metabolic byproduct\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e, so we hypothesized that a similar metabolic exchange was occurring in the pairing of \u003cem\u003eE. limosum\u003c/em\u003e and \u003cem\u003eP. faecium.\u003c/em\u003e For these reasons, we elected to test these species for the presence of cross-feeding and validate the iX-seq methodology.\u003c/p\u003e \u003cp\u003e \u003cb\u003eIn vitro\u003c/b\u003e \u003cb\u003eexamination of\u003c/b\u003e \u003cb\u003eP. faecium\u003c/b\u003e \u003cb\u003eand\u003c/b\u003e \u003cb\u003eE. limosum\u003c/b\u003e \u003cb\u003eas possible cross-feeding bacteria\u003c/b\u003e\u003c/p\u003e \u003cp\u003eOur initial focus was to determine whether the observed cross-feeding phenomenon in the droplet cultures could be reproduced \u003cem\u003ein vitro\u003c/em\u003e. We measured the growth of \u003cem\u003eP. faecium\u003c/em\u003e co-cultured with \u003cem\u003eE. limosum\u003c/em\u003e in mYCFA medium over a two-day time course. As a negative control, \u003cem\u003eP. faecium\u003c/em\u003e was cultivated in monoculture with no additional substrates, while 1% (w/v) succinate-supplemented growth medium was also tested in recognition of the species\u0026rsquo; aforementioned ability to metabolize succinate\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTo measure growth, we recorded CFU/mL and DNA copy number. The CFU count represents the viable cell count (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA), while the DNA copy number estimates relative cell biomass (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). The control \u003cem\u003eP. faecium\u003c/em\u003e monoculture exhibited a steady decline in CFU count and almost no increase in DNA copy number over time, indicative of limited cell proliferation and gradual cell death. In contrast, the succinate group showed a marked increase in CFUs and biomass by Day 1, followed by a sharp decrease in both metrics by Day 2. This growth-collapse trajectory likely reflects nutrient depletion or rapid accumulation of toxic metabolic byproducts, both of which would result in the precipitous loss of viability observed on Day 2. The timeline aligns with a previous study of \u003cem\u003eP. faecium\u003c/em\u003e\u0026rsquo;s growth kinetics, reporting that OD\u003csub\u003e600\u003c/sub\u003e peaked between approximately 24 and 42 hours after inoculation in GAM supplemented with 1% (w/v) succinate, during which the succinate was also fully depleted\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Notably, that study did not observe a collapse, but rather a stationary phase after reaching the peak by 42 hours. This may have been due to the presence of alternative growth substrates or protective cofactors in GAM that mitigated the metabolic stress caused by succinate depletion or toxic metabolite buildup. This would not be possible in the less-complex mYCFA medium, as it lacks many of the cell-derived extracts and complex nutrients found in the richer GAM. Overall, we hypothesize that the observations in the succinate-supplemented cultures were indicative of rapid growth, given that succinate serves as a preferred energy source for \u003cem\u003eP. faecium\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/em\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eInterestingly, the \u003cem\u003eP. faecium\u003c/em\u003e and \u003cem\u003eE. limosum\u003c/em\u003e co-cultures did not show an increase in CFU count over time as we expected. The DNA copy number did not differ significantly from the negative control on Days 1 and 2 either, though it was modestly higher. However, there was a statistically significant attenuation of CFU reduction between the co-culture and negative control, suggesting that while \u003cem\u003eE. limosum\u003c/em\u003e could not effectively sustain growth of \u003cem\u003eP. faecium\u003c/em\u003e by itself, it attenuated the decrease in viable cell count as a result of succinate deficiency in the growth medium. From these results, it appeared that \u003cem\u003eE. limosum\u003c/em\u003e was able to provide some metabolic benefit to \u003cem\u003eP. faecium\u003c/em\u003e, though it is unclear whether succinate production was the causal mechanism.\u003c/p\u003e \u003cp\u003e \u003cb\u003eMouse intestinal colonization model of\u003c/b\u003e \u003cb\u003eP. faecium\u003c/b\u003e \u003cb\u003eand\u003c/b\u003e \u003cb\u003eE. limosum\u003c/b\u003e \u003cb\u003ecross-feeding\u003c/b\u003e\u003c/p\u003e \u003cp\u003eIt is widely recognized that the intestinal environment, where gut microbes colonize, produces vastly different outcomes compared to \u003cem\u003ein vitro\u003c/em\u003e research conditions. Nutrient availability, shifts in pH, abundance of host-derived molecules, and countless other factors create an environment that cannot easily be replicated outside of the body. Since cross-feeding occurs within the context of the intestines, the next logical step in our study was to validate the predicted cross-feeding relationship through transplantation into an \u003cem\u003ein vivo\u003c/em\u003e mouse model.\u003c/p\u003e \u003cp\u003eGerm-free (GF) BALB/c mice were initially inoculated with either sterile PBS or \u003cem\u003eE. limosum\u003c/em\u003e and were observed for two weeks (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003eA-B). Then, both groups were further inoculated with \u003cem\u003eP. faecium.\u003c/em\u003e Fecal samples from each mouse were collected each week and measured for CFU/g feces of \u003cem\u003eP. faecium\u003c/em\u003e and \u003cem\u003eE. limosum\u003c/em\u003e. Interestingly, it was observed that \u003cem\u003eP. faecium\u003c/em\u003e was only able to successfully colonize the mouse intestines when co-inoculated with \u003cem\u003eE. limosum\u003c/em\u003e. At Week 3, the \u003cem\u003eE. limosum\u003c/em\u003e group showed high \u003cem\u003eE. limosum\u003c/em\u003e and \u003cem\u003eP. faecium\u003c/em\u003e viable cell counts at about 10\u003csup\u003e10\u003c/sup\u003e and 10\u003csup\u003e6\u003c/sup\u003e CFUs per gram of feces, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). These numbers remained consistent throughout the fourth and final week of measurement. It was apparent that some characteristic(s) of \u003cem\u003eE. limosum\u003c/em\u003e conferred \u003cem\u003eP. faecium\u003c/em\u003e the ability to colonize the mouse intestines.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eHypothesizing that cross-feeding underlies this phenomenon, we performed metabolome analysis of fecal samples from both PBS and \u003cem\u003eE. limosum\u003c/em\u003e-inoculated mouse groups at Week 2 (prior to \u003cem\u003eP. faecium\u003c/em\u003e inoculation) using principal component analysis (PCA) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003eC-D). As expected, we found that the metabolome profiles of each group were compositionally distinct. A number of distinguishing features were identified, including several amino acids.\u003c/p\u003e \u003cp\u003eHowever, the metabolite with the greatest influence on group separation was 2-oxoglutarate (2-OG), a molecule that is involved in the succinate metabolic pathway as a precursor to succinyl-CoA. Specifically, glutamate is hypothesized to be an alternative substrate to succinate in that it is intracellularly converted to 2-OG and subsequently to succinyl-CoA, a common intermediate predicted to be shared between succinate and glutamate in the succinate metabolic pathway of P. faecium38 (Fig. 5A). Indeed, 2-OG was found in high quantities in the feces of E. limosum-inoculated mice, while mice without E. limosum did not excrete any detectable 2-OG (Fig. 4E). As we know that P. faecium can metabolize succinate, presumably by this pathway, we predicted that 2-OG was also implicated in the ability of P. faecium to colonize the mouse through metabolic cross-feeding from E. limosum.\u003c/p\u003e \u003cp\u003e \u003cb\u003eAnalysis of 2-OG as a cross-fed metabolite promoting\u003c/b\u003e \u003cb\u003eP. faecium\u003c/b\u003e \u003cb\u003egrowth and colonization\u003c/b\u003e\u003c/p\u003e \u003cp\u003eWe next performed \u003cem\u003ein vitro\u003c/em\u003e growth experiments using \u003cem\u003eP. faecium\u003c/em\u003e and 2-OG in order to investigate what effects it would have on \u003cem\u003eP. faecium\u003c/em\u003e growth in defined conditions. 2-OG showed a significant increase in \u003cem\u003eP. faecium\u003c/em\u003e growth that scaled with higher concentrations, reflected in both increased DNA copy numbers and CFUs by Day 2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e5\u003c/span\u003eB-C). As 2-OG supplementation successfully promoted \u003cem\u003eP. faecium\u003c/em\u003e growth, the data suggests that 2-OG production facilitating cross-feeding by \u003cem\u003eE. limosum\u003c/em\u003e is at least one mechanism that may have contributed to colonization of \u003cem\u003eP. faecium\u003c/em\u003e in the mouse intestines.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we demonstrated the effectiveness of using iX-seq to supplement traditional methods of streak plating and colony isolation for screening of bacteria in complex environmental samples such as fecal samples. By applying this high-throughput methodology to bacterial communities, researchers can efficiently and rapidly identify bacteria of interest for their purposes.\u003c/p\u003e \u003cp\u003eAs shown in the results, we were able to identify a number of potential cross-feeding organisms in fecal samples by virtue of the ability of the benefactor \u0026ldquo;sender\u0026rdquo; strain to support the growth of a beneficiary \u0026ldquo;receiver\u0026rdquo; strain. Furthermore, we were able to support the existence of this interaction through \u003cem\u003ein vitro\u003c/em\u003e culturing and further evidenced their relationship in a mouse model. This culminated in the identification of what we believe to be a previously unreported cross-feeding relationship between \u003cem\u003eP. faecium\u003c/em\u003e and \u003cem\u003eE. limosum\u003c/em\u003e, thereby demonstrating the potential of iX-seq in exploring the relationships of complex microbial communities. Through the progression of high-throughput screening, 16S rRNA gene sequencing, \u003cem\u003ein vitro\u003c/em\u003e culture, \u003cem\u003ein vivo\u003c/em\u003e colonization, and metabolomic analysis, we were able to discover the pairing of \u003cem\u003eP. faecium\u003c/em\u003e and \u003cem\u003eE. limosum\u003c/em\u003e as cross-feeding members of the gut microbiota that are hypothesized to exert growth-supportive effects through the production of 2-OG.\u003c/p\u003e \u003cp\u003eHowever, it should be noted that the involvement of 2-OG is largely speculative, as we have not yet explored the mechanisms of 2-OG production and whether the metabolite was directly involved in \u003cem\u003eP. faecium\u003c/em\u003e colonization of \u003cem\u003eE. limosum\u003c/em\u003e-colonized GF mice. Nutrient supplementation via cross-feeding is only one of several possible explanations, as \u003cem\u003eE. limosum\u003c/em\u003e may have modified the gut environment to favor \u003cem\u003eP. faecium\u003c/em\u003e through alterations in luminal pH, redox potential, modulation of host-derived compounds, or other adjustment of physiochemical conditions. Furthermore, while \u003cem\u003eE. limosum\u003c/em\u003e did appear to support \u003cem\u003eP. faecium\u003c/em\u003e growth in droplet cultures, \u003cem\u003eE. limosum\u003c/em\u003e\u0026rsquo;s growth-supportive effects were comparatively limited when cultured in larger culture volumes \u003cem\u003ein vitro\u003c/em\u003e, raising further questions as to what conditions are necessary for \u003cem\u003eP. faecium\u003c/em\u003e to receive benefits from \u003cem\u003eE. limosum\u003c/em\u003e. To elucidate the exact nature of their relationship, further investigation and mechanistic studies are necessary. However, as the initial goal of this study was to test the viability of iX-seq for research on cross feeding, further investigation into the mechanisms underlying this predicted cross-feeding relationship is a topic of future research.\u003c/p\u003e \u003cp\u003eOne obvious limitation of iX-seq that we realized was the limited diversity of bacteria that could be cultivated in single-species droplets, as well as the overall diversity of cultured bacteria. A key advantage of droplet microfluidics is believed to be the ability to isolate low-abundance bacteria from complex communities\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. However, with only 22 species identified and very few able to grow in single-species droplets in our case, further development of our methodology is necessary to maximize species diversity and cell viability. As applications of this technology to microbiology are still in their infancy, there are various factors to consider such as medium composition, droplet stability, anaerobicity, and sample handling, for which there are no universal standards. While droplet microfluidics is a powerful and increasingly popular technique in microbial ecology, the wide range of available microfluidic devices and methods necessitates methodological development on a case-by-case basis\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. And regrettably, as with traditional isolation methods, we must concede that certain bacteria may be unculturable, condemning these species to the confines of metagenomic data until advancements in bacterial culturing techniques facilitate their cultivation. Nevertheless, we aim to continue refining our own methodology to maximize the diversity of cultivable species and enable the observation of unknown cross-feeding interactions within the gut microbiota.\u003c/p\u003e \u003cp\u003eIn order to narrow our focus to the most straightforward approach to droplet microfluidics-based screening, we only conducted tests to identify unidirectional cross-feeding bacteria and excluded other possibilities in this study. However, droplet microfluidics may theoretically be applied to various cross-feeding scenarios as we continue to develop our methodology. At higher levels of complexity, cross-feeding can also be bidirectional (both parties provide substrates to each other) or multidirectional (multiple parties and exchanges of substrates are involved), illustrating the ecological interconnectedness of the gut metabolic network. There are also a number of possible mechanisms for cross-feeding, such as the secretion of small molecules and degradation of complex molecules by the sender strain\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. While technically more complicated, exploration of such interactions may be achievable by using innovative approaches such as the development of a metabolite-detection pipeline integrated into the iX-seq method.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eHigh-throughput methods such as iX-seq are invaluable for studying the complex microbial interactions within the human gut, enabling the evaluation of the thousands of species that constitute the gut microbiota. Using the iX-seq platform, we were able to isolate and provide evidence for a previously unknown cross-feeding relationship between \u003cem\u003eP. faecium\u003c/em\u003e and \u003cem\u003eE. limosum\u003c/em\u003e, also demonstrating the utility and potential of droplet microfluidics. Due to the increasing accessibility and efficiency of metagenomic sequencing, the mechanistic insights needed to contextualize the vast amounts of generated data are increasingly necessary, particularly for translating gut microbiota research into applications beyond the lab, such as in health and medicine.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eCulture conditions\u003c/h2\u003e \u003cp\u003eModified YCFA (mYCFA) medium was prepared as described in a previous study\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. The medium was sparged with an anoxic gas mixture of 85% nitrogen, 5% hydrogen, and 10% carbon dioxide while heating, supplemented with L-cysteine hydrochloride monohydrate, and dispensed into 30 mL serum bottles at 10 mL volumes under anoxic gas flow. The bottles were sealed with butyl rubber stoppers and aluminum crimp seals to preserve anaerobicity and autoclaved at 121\u0026deg;C for 20 minutes. To initiate bacterial cultures, bottles were inoculated with the respective inocula via syringe and incubated at 37 ˚C without agitation. In the case of well plate incubation, well plates were incubated partially uncovered in a vinyl anaerobic chamber (Coy Laboratory Products, Inc., Grass Lake, MI, USA) at 37 ˚C. The chamber contained the same anoxic gas mixture as used for medium preparation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eFecal sample collection and preparation\u003c/h2\u003e \u003cp\u003eA fecal sample from a healthy Japanese volunteer was collected on the day of testing. Subjects with intestinal diseases, skin diseases, those who had taken antibiotics, or were allergic to the specified diet were excluded from consideration. All steps following fecal sample collection were performed in an anaerobic chamber containing an anoxic gas mixture of 85% nitrogen, 5% hydrogen, and 10% carbon dioxide. The fecal sample was homogenized in mYCFA as described in a previous study\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e at a concentration of 10% (w/v), filtered through a 40 \u0026micro;M cell strainer to remove large debris, and further diluted in mYCFA medium supplemented with 0.5% SeaPlaque agarose, resulting in a final concentration of 0.01% (v/v). To prevent the agarose from solidifying, the diluted sample was maintained at 37 ˚C using a heat block.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMicrofluidic droplet generation\u003c/h3\u003e\n\u003cp\u003eA Droplet Generator (On-chip Biotechnologies Co., Ltd, Tokyo, Japan) was used to generate water-in-oil microfluidic droplets from the fecal suspension with a dispensation rate of 0\u0026thinsp;~\u0026thinsp;9 cells per droplet. The oil used was a 5% 008-FluoroSurfactant formulation (008-FluoroSurfactant-5wtH-10mL (2003001); On-chip Biotechnologies Co., Ltd.). Each droplet had a diameter of approximately 60 \u0026micro;m and was generated following the manufacturer\u0026rsquo;s instructions. The droplets were collected in 1.5 mL microcentrifuge tubes and incubated at 37\u0026deg;C overnight in an anaerobic chamber. Following incubation, an equal volume of oil was added to each sample, and the tubes were cooled at 4\u0026deg;C for 15 minutes to allow the agarose to solidify. Subsequently, an equal volume of 1H,1H,2H,2H-perfluoro-1-octanol was added and mixed with the suspension to disrupt the oil layer. Then, 50 \u0026micro;L of T buffer (On-chip Biotechnologies Co., Ltd.) was added, and the mixture was centrifuged at 200 \u0026times; g for 3 minutes at room temperature. The supernatant was collected, and the resulting gel beads (GBs) were recovered. These GBs were then dispensed into 384-well plates containing 100 \u0026micro;L of mYCFA per well using the Single-Particle Isolation and Sequencing (SPiS) instrument (On-chip Biotechnologies Co., Ltd.) and once more incubated overnight at 37\u0026deg;C in an anaerobic chamber.\u003c/p\u003e\n\u003ch3\u003eDNA barcoding and 16S rRNA gene sequencing\u003c/h3\u003e\n\u003cp\u003eAfter cultivation in the 384-well plates, 20 \u0026micro;L of each well culture were transferred to a new well plate, mixed with an equal volume of 20% (v/v) glycerol in PBS for a final concentration of 10% (v/v) glycerol, and stored at -80\u0026deg;C for future analysis. The remaining culture volume was used for unique 16S rRNA gene barcoding of bacterial DNA using the Echo 525 Liquid Handler. The 16S rRNA gene library was then sequenced using the Illumina MiSeq platform, and bacterial amplicon sequence variants (ASVs) were assigned using QIIME2. Droplet composition data was manually analyzed using R.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eStrain isolation from droplets\u003c/h2\u003e \u003cp\u003eThe bacteria of interest were collected from their wells in the 384-well plate 10% (v/v) glycerol stocks and revitalized in liquid mYCFA medium. The bacteria were then cultured for two days at 37\u0026deg;C in an anaerobic chamber and subsequently passaged onto Gifu Anaerobic Medium (GAM; Nissui Pharmaceutical Co. Ltd., Tokyo, Japan) agar plates with or without vancomycin (20 \u0026micro;g/mL) to culture for four additional days until distinct colonies were visible. The purpose of vancomycin was to isolate \u003cem\u003eP. faecium\u003c/em\u003e from two-species droplets, utilizing \u003cem\u003eE. limosum\u003c/em\u003e\u0026rsquo;s inherent sensitivity to the antibiotic and \u003cem\u003eP. faecium\u003c/em\u003e\u0026rsquo;s resistance. Individual colonies were then isolated and cultured in liquid mYCFA medium for an additional two days before mixing with an equal volume of 20% (v/v) glycerol in PBS for storage at -80\u0026deg;C at a final glycerol concentration of 10% (v/v).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eMouse experiments\u003c/h2\u003e \u003cp\u003eGerm-free (GF) BALB/c mice (female, 6\u0026ndash;20 weeks of age) were housed in gnotobiotic isolators and fed an autoclaved, standard rodent chow diet (CMF, Oriental Yeast Co., Ltd., Tokyo, Japan). \u003cem\u003eP. faecium\u003c/em\u003e was cultured in GAM medium supplemented with 1% (w/v) succinate for 48 hours. After cultivation, the cultures were centrifuged at 10,000 \u0026times; g for 5 minutes at room temperature, and the supernatant was discarded. The resulting cell pellets were resuspended in freshly prepared GAM medium. \u003cem\u003eE. limosum\u003c/em\u003e was cultured in GAM medium for 24 hours and was processed in the same manner as \u003cem\u003eP. faecium\u003c/em\u003e. The mice were inoculated with \u003cem\u003eP. faecium\u003c/em\u003e or \u003cem\u003eE. limosum\u003c/em\u003e cells by oral gavage of 200 \u0026micro;L of 1\u0026times;10\u003csup\u003e8\u003c/sup\u003e colony-forming units (CFUs). During the 4-week experiment, feces were collected at weekly intervals for additional testing. All experiments were conducted in accordance with protocols approved by the University of Tsukuba Animal Experiment Committee.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eCE-TOFMS-based metabolome analysis\u003c/h2\u003e \u003cp\u003eFecal metabolites were extracted from the samples and detected by capillary electrophoresis time-of-flight mass spectrometry (CE-TOFMS) using methods described in previous studies\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. Briefly, fecal metabolites were extracted from 10 mg of freeze-dried feces by adding 500 \u0026micro;L of methanol containing 20 \u0026micro;M each of methionine sulfone (A17027; Alfa Aesar, Ward Hill, MA, USA) and D-camphol-10-sulfonic acid (CSA) (4987481429680; FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) as internal standards. and shaking vigorously with 100 mg of 0.1-mm zirconia/silica beads (BioSpec Products, Inc., Bartlesville, OK, USA) for 5 min at 1500 rpm using a Shake Master Neo (Bio Medical Science, Inc., Tokyo, Japan). Next, this mixture was mixed with 200 \u0026micro;L of ultrapure water and 500 \u0026micro;L of chloroform and shaken again. The suspension was centrifuged at 4,600 \u0026times; g for 15 min at 4\u0026deg;C, and the resulting supernatant was transferred to a 5-kDa-cutoff filter column (Ultrafree MC-PLHCC 250/pk for Metabolome Analysis; Human Metabolome Technologies, Tsuruoka, Japan). The flow-through was dried under a vacuum, and the residue was then dissolved in 50 \u0026micro;L of Milli-Q water containing reference compounds (200 \u0026micro;M each of 3-aminopyrrolidine and trimesate). The levels of extracted metabolites were measured in both positive and negative modes by CE-TOFMS as described previously\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. All CE-TOFMS experiments were performed using an Agilent capillary electrophoresis system (Agilent Technologies, Santa Clara, CA, USA). Raw data was analyzed using our proprietary automatic integration software MasterHands (ver. 2.16.0.15)\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. Principal component analysis (PCA) was performed using the software MetaboAnalyst 6.0\u003csup\u003e46,47\u003c/sup\u003e. Metabolome data was normalized by sum, log\u003csub\u003e10\u003c/sub\u003e-transformed, and Pareto-scaled prior to analysis.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by the Ethical Committees of Keio University Shonan Fujisawa Campus (No. 355). \u0026nbsp;All subjects were informed of the purpose of this study, and written consent was obtained from all subjects. \u0026nbsp;This study was conducted with strict consideration for privacy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and material\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMicrobiome data has been deposited to DDBJ (accession number PRJDB37686), and metabolome data has been deposited to MetaboBank (accession number MTBKS266).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJSPS KAKENHI (22H03541 to SF), AMED-CREST (JP23gm1010009 to SF), JST ERATO (JPMJER1902 to SF), the Food Science Institute Foundation (to SF). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors’ contributions\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSF, KT, and IS conceived the study and designed the experiments. \u0026nbsp;KT, NT, and NO conducted animal experiments. \u0026nbsp;NT and MK performed microfluidics experiments, metagenome and metabolome analyses. \u0026nbsp; IS and NT conducted informatics analysis. \u0026nbsp;SF, NO, JY, GN, CI, and KT provided critical review and commentary of the manuscript. \u0026nbsp;All authors contributed to the article and approved the submitted version.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe wish to thank Noriko Kagata and Tatsuji Takahashi for their technical assistance. \u0026nbsp; Figures were created or modified using BioRender.com. \u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eHou, K. \u003cem\u003eet al.\u003c/em\u003e Microbiota in health and diseases. \u003cem\u003eSignal Transduct Target Ther\u003c/em\u003e 7, 135 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO\u0026rsquo;Hara, A. M. \u0026amp; Shanahan, F. The gut flora as a forgotten organ. \u003cem\u003eEMBO Rep\u003c/em\u003e 7, 688\u0026ndash;693 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNishida, A. \u003cem\u003eet al.\u003c/em\u003e Gut microbiota in the pathogenesis of inflammatory bowel disease. \u003cem\u003eClin J Gastroenterol\u003c/em\u003e 11, 1\u0026ndash;10 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNi, J., Wu, G. D., Albenberg, L. \u0026amp; Tomov, V. T. Gut microbiota and IBD: causation or correlation? \u003cem\u003eNat Rev Gastroenterol Hepatol\u003c/em\u003e 14, 573\u0026ndash;584 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQiu, P. \u003cem\u003eet al.\u003c/em\u003e The Gut Microbiota in Inflammatory Bowel Disease. \u003cem\u003eFront Cell Infect Microbiol\u003c/em\u003e Volume 12-2022, (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCunningham, A. L., Stephens, J. W. \u0026amp; Harris, D. A. Gut microbiota influence in type 2 diabetes mellitus (T2DM). \u003cem\u003eGut Pathog\u003c/em\u003e 13, 50 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, Z., Sun, B., Yu, D. \u0026amp; Zhu, C. Gut Microbiota: An Important Player in Type 2 Diabetes Mellitus. \u003cem\u003eFront Cell Infect Microbiol\u003c/em\u003e Volume 12-2022, (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFusco, W. \u003cem\u003eet al.\u003c/em\u003e Gut microbiota and atherosclerosis. \u003cem\u003eGut\u003c/em\u003e gutjnl-2025-335610 (2025) doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1136/gutjnl-2025-335610\u003c/span\u003e\u003cspan address=\"10.1136/gutjnl-2025-335610\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJonsson, A. L. \u0026amp; B\u0026auml;ckhed, F. Role of gut microbiota in atherosclerosis. \u003cem\u003eNat Rev Cardiol\u003c/em\u003e 14, 79\u0026ndash;87 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWong, C. C. \u0026amp; Yu, J. Gut microbiota in colorectal cancer development and therapy. \u003cem\u003eNat Rev Clin Oncol\u003c/em\u003e 20, 429\u0026ndash;452 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim, J. \u0026amp; Lee, H. K. Potential Role of the Gut Microbiome In Colorectal Cancer Progression. \u003cem\u003eFront Immunol\u003c/em\u003e Volume 12-2021, (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRinninella, E. \u003cem\u003eet al.\u003c/em\u003e What is the Healthy Gut Microbiota Composition? A Changing Ecosystem across Age, Environment, Diet, and Diseases. \u003cem\u003eMicroorganisms\u003c/em\u003e 7, 14 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCoyte, K. Z. \u0026amp; Rakoff-Nahoum, S. Understanding Competition and Cooperation within the Mammalian Gut Microbiome. \u003cem\u003eCurrent Biology\u003c/em\u003e 29, R538\u0026ndash;R544 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCulp, E. J. \u0026amp; Goodman, A. L. Cross-feeding in the gut microbiome: Ecology and mechanisms. \u003cem\u003eCell Host Microbe\u003c/em\u003e 31, 485\u0026ndash;499 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGwen, F., Angeliki, V., Kristof, V. \u0026amp; De, V. L. Cross-Feeding between Bifidobacterium longum BB536 and Acetate-Converting, Butyrate-Producing Colon Bacteria during Growth on Oligofructose. \u003cem\u003eAppl Environ Microbiol\u003c/em\u003e 72, 7835\u0026ndash;7841 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSander, S. \u003cem\u003eet al.\u003c/em\u003e Mixed-Culture Transcriptome Analysis Reveals the Molecular Basis of Mixed-Culture Growth in Streptococcus thermophilus and Lactobacillus bulgaricus. \u003cem\u003eAppl Environ Microbiol\u003c/em\u003e 76, 7775\u0026ndash;7784 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarcia, S. L. \u003cem\u003eet al.\u003c/em\u003e Auxotrophy and intrapopulation complementary in the \u0026lsquo;interactome\u0026rsquo; of a cultivated freshwater model community. \u003cem\u003eMol Ecol\u003c/em\u003e 24, 4449\u0026ndash;4459 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWoyke, T. \u003cem\u003eet al.\u003c/em\u003e Symbiosis insights through metagenomic analysis of a microbial consortium. \u003cem\u003eNature\u003c/em\u003e 443, 950\u0026ndash;955 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eC, C. C. \u003cem\u003eet al.\u003c/em\u003e Ecological Importance of Cross-Feeding of the Intermediate Metabolite 1,2-Propanediol between Bacterial Gut Symbionts. \u003cem\u003eAppl Environ Microbiol\u003c/em\u003e 86, e00190-20 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmith, N. W., Shorten, P. R., Altermann, E., Roy, N. C. \u0026amp; McNabb, W. C. The Classification and Evolution of Bacterial Cross-Feeding. \u003cem\u003eFront Ecol Evol\u003c/em\u003e 7, (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOliphant, K. \u0026amp; Allen-Vercoe, E. Macronutrient metabolism by the human gut microbiome: major fermentation by-products and their impact on host health. \u003cem\u003eMicrobiome\u003c/em\u003e 7, 91 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoek, M. J. A. van \u0026amp; Merks, R. M. H. Emergence of microbial diversity due to cross-feeding interactions in a spatial model of gut microbial metabolism. \u003cem\u003eBMC Syst Biol\u003c/em\u003e 11, 56 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSung, J. \u003cem\u003eet al.\u003c/em\u003e Global metabolic interaction network of the human gut microbiota for context-specific community-scale analysis. \u003cem\u003eNat Commun\u003c/em\u003e 8, 15393 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoyal, A., Wang, T., Dubinkina, V. \u0026amp; Maslov, S. Ecology-guided prediction of cross-feeding interactions in the human gut microbiome. \u003cem\u003eNat Commun\u003c/em\u003e 12, 1335 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eProffitt, C. \u003cem\u003eet al.\u003c/em\u003e Genome-scale metabolic modelling of the human gut microbiome reveals changes in the glyoxylate and dicarboxylate metabolism in metabolic disorders. \u003cem\u003eiScience\u003c/em\u003e 25, 104513 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClaudia, S.-A., Mar\u0026iacute;a, R.-F., Daniel, G. \u0026amp; M, M. A. J. Using metabolic networks to predict cross-feeding and competition interactions between microorganisms. \u003cem\u003eMicrobiol Spectr\u003c/em\u003e 12, e02287-23 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoyal, A., Wang, T., Dubinkina, V. \u0026amp; Maslov, S. Ecology-guided prediction of cross-feeding interactions in the human gut microbiome. \u003cem\u003eNat Commun\u003c/em\u003e 12, 1335 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartin, K. \u003cem\u003eet al.\u003c/em\u003e Generation of larger numbers of separated microbial populations by cultivation in segmented-flow microdevices. \u003cem\u003eLab Chip\u003c/em\u003e 3, 202\u0026ndash;207 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClausell-Tormos, J. \u003cem\u003eet al.\u003c/em\u003e Droplet-Based Microfluidic Platforms for the Encapsulation and Screening of Mammalian Cells and Multicellular Organisms. \u003cem\u003eChem Biol\u003c/em\u003e 15, 427\u0026ndash;437 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, W., Kim, H. J., Lucchetta, E. M., Du, W. \u0026amp; Ismagilov, R. F. Isolation, incubation, and parallel functional testing and identification by FISH of rare microbial single-copy cells from multi-species mixtures using the combination of chemistrode and stochastic confinement. \u003cem\u003eLab Chip\u003c/em\u003e 9, 2153\u0026ndash;2162 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM, V. M. \u003cem\u003eet al.\u003c/em\u003e Interindividual Variation in Dietary Carbohydrate Metabolism by Gut Bacteria Revealed with Droplet Microfluidic Culture. \u003cem\u003emSystems\u003c/em\u003e 5, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1128/msystems.00864\u0026thinsp;\u0026ndash;\u0026thinsp;19\u003c/span\u003e\u003cspan address=\"10.1128/msystems.00864\u0026thinsp;\u0026ndash;\u0026thinsp;19\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\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 9, e56998 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHsu, R. H. \u003cem\u003eet al.\u003c/em\u003e Microbial Interaction Network Inference in Microfluidic Droplets. \u003cem\u003eCell Syst\u003c/em\u003e 9, 229\u0026ndash;242.e4 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYohei, W., Fumiko, N. \u0026amp; Masami, M. Characterization of Phascolarctobacterium succinatutens sp. nov., an Asaccharolytic, Succinate-Utilizing Bacterium Isolated from Human Feces. \u003cem\u003eAppl Environ Microbiol\u003c/em\u003e 78, 511\u0026ndash;518 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNagao-Kitamoto, H. \u003cem\u003eet al.\u003c/em\u003e Interleukin-22-mediated host glycosylation prevents Clostridioides difficile infection by modulating the metabolic activity of the gut microbiota. \u003cem\u003eNat Med\u003c/em\u003e 26, 608\u0026ndash;617 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStackebrandt, E. \u0026amp; Osawa, R. Phascolarctobacterium. in \u003cem\u003eBergey\u0026rsquo;s Manual of Systematics of Archaea and Bacteria\u003c/em\u003e 1\u0026ndash;4 (2015). doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/9781118960608.gbm00700\u003c/span\u003e\u003cspan address=\"10.1002/9781118960608.gbm00700\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDel Dot, T., Osawa, R. \u0026amp; Stackebrandt, E. Phascolarctobacterium faecium gen. nov, spec. nov., a Novel Taxon of the Sporomusa Group of Bacteria. \u003cem\u003eSyst Appl Microbiol\u003c/em\u003e 16, 380\u0026ndash;384 (1993).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIkeyama, N. \u003cem\u003eet al.\u003c/em\u003e Microbial interaction between the succinate-utilizing bacterium Phascolarctobacterium faecium and the gut commensal Bacteroides thetaiotaomicron. \u003cem\u003eMicrobiologyopen\u003c/em\u003e 9, e1111 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWade, W. G. Eubacterium. in \u003cem\u003eBergey\u0026rsquo;s Manual of Systematics of Archaea and Bacteria\u003c/em\u003e 1\u0026ndash;36 (2015). doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/9781118960608.gbm00629\u003c/span\u003e\u003cspan address=\"10.1002/9781118960608.gbm00629\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYin, J. \u003cem\u003eet al.\u003c/em\u003e A droplet-based microfluidic approach to isolating functional bacteria from gut microbiota. \u003cem\u003eFront Cell Infect Microbiol\u003c/em\u003e 12, (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe, Z., Wu, H., Yan, X. \u0026amp; Liu, W. Recent advances in droplet microfluidics for microbiology. \u003cem\u003eChinese Chemical Letters\u003c/em\u003e 33, 1729\u0026ndash;1742 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNakanishi, Y. \u003cem\u003eet al.\u003c/em\u003e Dynamic Omics Approach Identifies Nutrition-Mediated Microbial Interactions. \u003cem\u003eJ Proteome Res\u003c/em\u003e 10, 824\u0026ndash;836 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMishima, E. \u003cem\u003eet al.\u003c/em\u003e Evaluation of the impact of gut microbiota on uremic solute accumulation by a CE-TOFMS\u0026ndash;based metabolomics approach. \u003cem\u003eKidney Int\u003c/em\u003e 92, 634\u0026ndash;645 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoga, T. \u003cem\u003eet al.\u003c/em\u003e Quantitative Metabolome Analysis Using Capillary Electrophoresis Mass Spectrometry. \u003cem\u003eJ Proteome Res\u003c/em\u003e 2, 488\u0026ndash;494 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSugimoto, M., Wong, D. T., Hirayama, A., Soga, T. \u0026amp; Tomita, M. Capillary electrophoresis mass spectrometry-based saliva metabolomics identified oral, breast and pancreatic cancer-specific profiles. \u003cem\u003eMetabolomics\u003c/em\u003e 6, 78\u0026ndash;95 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePang, Z. \u003cem\u003eet al.\u003c/em\u003e MetaboAnalyst 5.0: narrowing the gap between raw spectra and functional insights. \u003cem\u003eNucleic Acids Res\u003c/em\u003e 49, W388\u0026ndash;W396 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXia, J. \u0026amp; Wishart, D. S. Metabolomic Data Processing, Analysis, and Interpretation Using MetaboAnalyst. \u003cem\u003eCurr Protoc Bioinformatics\u003c/em\u003e 34, 14.10.1\u0026ndash;14.10.48 (2011).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-biofilms-and-microbiomes","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjbiofilms","sideBox":"Learn more about [npj Biofilms and Microbiomes](http://www.nature.com/npjbiofilms/)","snPcode":"41522","submissionUrl":"https://submission.springernature.com/new-submission/41522/3","title":"npj Biofilms and Microbiomes","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"droplet microfluidics, gut microbiota, cross-feeding, Phascolarctobacterium faecium, Eubacterium limosum, 2-oxoglutarate","lastPublishedDoi":"10.21203/rs.3.rs-8817168/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8817168/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eThe human gut microbiota comprises hundreds of bacterial species that form a dynamic metabolic network, in which the exchange of microbial metabolites, known as cross-feeding, is integral to community function and host health. However, defining and isolating specific cross-feeding interactions remains challenging due to the limitations of conventional culture-based approaches. Droplet microfluidics enables high-throughput screening and isolation of bacteria from complex communities by encapsulating cells in water-in-oil droplets, allowing for parallel cultivation of thousands of stochastically contained microcultures. In this study, we developed a droplet microfluidics-based screening platform integrated with 16S rRNA gene amplicon sequencing, collectively termed iX-seq, to identify novel cross-feeding interactions directly from a human fecal sample.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eUsing iX-seq, we identified \u003cem\u003ePhascolarctobacterium faecium\u003c/em\u003e and \u003cem\u003eEubacterium limosum\u003c/em\u003e as potential unidirectional cross-feeders, wherein the growth of \u003cem\u003eP. faecium\u003c/em\u003e was obligately supported by \u003cem\u003eE. limosum\u003c/em\u003e in droplet co-cultures. The growth of \u003cem\u003eP. faecium\u003c/em\u003e was modestly affected when co-cultured in vitro but notably required \u003cem\u003eE. limosum\u003c/em\u003e co-inoculation for intestinal colonization in germ-free mice. Metabolomic profiling implicated 2-oxoglutarate (2-OG), an intermediate in \u003cem\u003eP. faecium\u003c/em\u003e\u0026rsquo;s succinate metabolic pathway, as the key differentiating metabolite between \u003cem\u003eE. limosum\u003c/em\u003e-inoculated mouse and control mouse feces. Supplementation of 2-OG in monoculture increased \u003cem\u003eP. faecium\u003c/em\u003e growth in a concentration-dependent manner, confirming its supportive role.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eOur results reveal a previously unrecognized, potential 2-OG-mediated cross-feeding relationship between \u003cem\u003eP. faecium\u003c/em\u003e and \u003cem\u003eE. limosum\u003c/em\u003e, offering new insight into gut microbial metabolic dependencies. By enabling high-throughput, culture-based screening directly from fecal samples, the iX-seq approach demonstrates a practical framework for studying cross-feeding interactions within the gut microbiota and represents a potential paradigm shift in how complex microbial communities are experimentally investigated.\u003c/p\u003e","manuscriptTitle":"Droplet microfluidic iX-seq platform enables discovery of gut bacterial cross-feeding between Phascolarctobacterium faecium and Eubacterium limosum","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-09 08:13:58","doi":"10.21203/rs.3.rs-8817168/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"335081878892497728184743941847194400303","date":"2026-04-29T12:58:23+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-22T14:55:41+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"296942140910696441896053555309326302714","date":"2026-03-05T03:21:05+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-04T06:43:18+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-19T08:48:05+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-14T13:14:32+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Biofilms and Microbiomes","date":"2026-02-07T17:11:35+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-biofilms-and-microbiomes","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjbiofilms","sideBox":"Learn more about [npj Biofilms and Microbiomes](http://www.nature.com/npjbiofilms/)","snPcode":"41522","submissionUrl":"https://submission.springernature.com/new-submission/41522/3","title":"npj Biofilms and Microbiomes","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"1603ffcf-38d0-47eb-bc99-8dc60caeb2d3","owner":[],"postedDate":"March 9th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":64030591,"name":"Biological sciences/Biological techniques"},{"id":64030592,"name":"Biological sciences/Biotechnology"},{"id":64030593,"name":"Biological sciences/Microbiology"}],"tags":[],"updatedAt":"2026-03-09T08:13:59+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-09 08:13:58","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8817168","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8817168","identity":"rs-8817168","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}