Conserved pathway for homarine catabolism in environmental bacteria

Conserved pathway for homarine catabolism in environmental bacteria | 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 Conserved pathway for homarine catabolism in environmental bacteria Frank Ferrer-Gonzalez, Katherine Heal, Joshua Sacks, Yarinet Romero-Maysonet, and 17 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7359689/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 30 Mar, 2026 Read the published version in Nature Microbiology → Version 1 posted You are reading this latest preprint version Abstract Homarine (N-methylpicolinic acid) is a ubiquitous marine metabolite produced by phytoplankton and noted for its bioactivity in marine animals, yet its microbial degradation pathways are uncharacterized. Here, we identify a conserved operon (homABCDER) that mediates homarine catabolism in bacteria using comparative transcriptomics, mutagenesis, and targeted knockouts. Phylogenetic and genomic analyses show this operon distributed across abundant bacterial clades, including coastal copiotrophs (e.g., Rhodobacterales) and open-ocean oligotrophs (e.g., SAR11, SAR116). High-resolution mass spectrometry revealed N-methylglutamic acid and glutamic acid as key metabolic products of homarine in both model and natural systems, with N-methylglutamate dehydrogenase catalyzing their conversion. Metatranscriptomics showed responsive and in situ expression of hom genes aligned with homarine availability. These findings uncover the genetic and metabolic basis of homarine degradation, establish its ecological relevance, and highlight homarine as a versatile growth substrate that feeds into central metabolism via glutamic acid in diverse marine bacteria. Biological sciences/Chemical biology/Natural products Earth and environmental sciences/Ecology/Microbial ecology Earth and environmental sciences/Ecology/Biogeochemistry Biological sciences/Microbiology/Biogeochemistry Biological sciences/Biochemistry/Metabolomics homarine catabolism metabolites marine microbial ecology N-methylglutamic acid Running title: homarine catabolism in environmental bacteria Figures Figure 1 Figure 2 Figure 3 Figure 4 Main Microbial metabolites are central to ocean biogeochemistry, facilitating rapid cycling of labile organic carbon through the microbial loop and driving the annual turnover of tens of petagrams of carbon 1 . Many metabolites also have compound-specific functions like enabling micronutrient acquisition 2 , serving as chemoattractants 3 , or acting as toxins 4 . Advances in environmental metabolomics have uncovered abundant metabolites whose biogeochemical and ecological significance 1 remain unresolved, partly because their metabolic pathways are unknown 5–9 . These unknown pathways hinder genomics, transcriptomics, and proteomics approaches for studying microbial function. Identifying the genes and transformations linked to metabolites enhances our understanding of microbial metabolism and better integrates metabolites into a collective biological, ecological, and biogeochemical framework. Among marine metabolites, homarine ( N -methylpicolinic acid), stands out for both its widespread distribution and multifaceted bioactivity. Homarine, a polar alkaloid with a methylated pyridine ring, was first discovered in lobsters nearly a century ago 10 and is now recognized as ubiquitous in surface seawater 6,7,11 with turnover times ranging from days to weeks 12 . While heterotrophic bacteria are assumed to consume homarine, the metabolic pathway, molecular intermediates, and ecological distribution of its degradation remain uncharacterized. Homarine is produced by abundant phytoplankton groups such as diatoms and Synechococcus at millimolar intracellular concentrations while other groups like dinoflagellates and prasinophytes do not produce it 6,13,14 . Homarine’s pyridine ring suggests specialized enzymatic machinery for its degradation, as seen with other pyridine-containing compounds 15–17 . This, along with its production patterns, may contribute to niche partitioning among phytoplankton and heterotrophs. Homarine also exhibits bioactivity in marine animals, influencing oyster shell morphology 18 , deterring feeding in butterfly fish and sea stars 19,20 , altering larval development in hydroids 21 , and serving as a crab predation deterrent 22 . This study investigates the heterotrophic bacterial degradation of homarine across biochemical, genetic, and ecological dimensions through an integrative approach. Comparative transcriptomics identifies a conserved operon for homarine degradation, while mass spectrometry reveals its biochemical transformation into metabolic products. Comparative genomics maps the global distribution of homologous operons, and metatranscriptomic analyses in natural and amended communities link gene expression to homarine availability, emphasizing its ecological relevance. Results Identification of homarine operon Motivated by homarine’s abundance in marine environments, we sought to understand its bacterial degradation. We isolated a bacterium from seawater at Owen Beach,Tacoma, Washington USA, and demonstrated its ability to use homarine as a primary carbon and nitrogen source (Figure 1A). Genome sequencing identified this Gammaproteobacterium as Cobetia sp. within the order Oceanospirillales , family Halomonadaceae , with high similarity to Cobetia marina (98.51% ANI) and Cobetia pacifica (98.3% ANI, Table S1) 23 . We named this isolate Cobetia sp . OBi1. To uncover genes involved in homarine degradation, we grew Cobetia sp. OBi1 in minimal media with homarine as the sole carbon source and conducted comparative transcriptomics and metabolomics experiments. We compared three conditions: glucose (12 mM C, 0.8 mM NH 4 , control), homarine (12 mM C, no additional NH 4 ), and glucose + homarine (12 mM C, 1 mM C, respectively with 0.8 mM NH 4 ). In both conditions with homarine, 22 genes were enriched compared to the glucose only condition (s-value < 0.05, Figure 1B, Table S2), including three gene clusters. The first cluster encoded a transcriptional regulator, a predicted glycine betaine transporter (GBT), and five putative catabolic genes with unclear function (ACFLL8_11490 to ACFLL8_11520, Figure 1C, Table S3). The second contains a putative sarcosine oxidase complex and the methylenetetrahydrofolate dehydrogenase folD (ACFLL8_07290 to ACFLL8_07310; Figure 1C, Tables S3). The third included motility-associated genes: fliG and flgD (Tables S2, S3). In the homarine-only treatment, we observed broader upregulation of the flagellar and chemotaxis genes, including fliM , fliN , motB , aer , cheR , and cheY 24,25 (Tables S2). These gene enrichments suggest that homarine may induce motility and chemotaxis. To target functional assessments, we identified orthologs of the homarine-induced genes from Cobetia sp. OBi1 in the model marine heterotrophic bacteria Ruegeria pomeroyi DSS-3 using reciprocal best high searches (e-value < 10⁻⁵). Of the seven genes in Cobetia sp. OBi1’s homarine-induced operon with unknown function (ACFLL8_11490–ACFLL8_11520), five were found in a single operon in Ruegeria pomeroyi DSS-3 (SPO3188–SPO3192; Figure 1C, Table S3). Orthologs of ACFLL8_11490 (the first catabolic gene) and ACFLL8_11505 (the glycine betaine transporter) were absent. R. pomeroyi DSS-3 was capable of growth on homarine as the sole carbon and nitrogen source (Figure SX1A). These results support a homarine catabolic operon here termed homABCDER (Figure 1C, Table S3), consisting of four core catabolic genes ( homABCD ), a transcriptional regulator ( homR ), and variably present auxiliary gene ( homE ). To investigate the functions and structures of the proteins encoded by the homABCDER operon we combined domain annotation 26–29 , structural modeling 30,31 and functional prediction 32,33 . In both Cobetia sp. OBi1 and R. pomeroyi DSS-3, HomA contains conserved motifs for a flavin adenine dinucleotide (FAD) binding domain (Table S3). Functional characterization of HomB suggests that it binds cyclic compounds and exhibits catalytic activity (CscoreGo > 0.65, Table S4). AlphaFold modeling yielded high-confidence structures for HomA and HomB (interface predicted template modelling + predicted template modelling: 0.9471 for R. pomeroyi DSS-3; 0.9246 for Cobetia sp. OBi1). Visualization in ChimeraX 34 , revealed hydrogen bonds and electrostatic interactions consistent with heterodimer formation. This resembles the trigonelline catabolic enzymes TgnA and TgnB, which operate as a flavin-supplying and oxygen-activating pair 35 to open the pyridine ring of trigonelline, a structural isomer of homarine. We hypothesize a similar functional relationship for HomA and HomB. HomC is predicted to function as a peptidase (CscoreGO > 0.65, Table S4), while HomD is likely an aldehyde dehydrogenase-like oxidoreductase (CscoreGO > 0.65, Table S3-S4). These findings are consistent with a model in which the operon encodes enzymes involved in distinct but coordinated steps of homarine degradation. Phenotypic analysis of homarine-associated genes We phenotypically characterized the proposed homarine catabolic operon using recombinant techniques and a transposon barcoded mutant library of our model organisms. In Cobetia sp. OBi1, homologous recombination was used to generate a Δ homD mutant (ACFLL8_11515), which was unable to grow on homarine as a sole carbon and nitrogen source while retaining normal growth on glucose (Figure SX1B). Complementation with the wild-type homD gene restored growth on homarine, confirming its role in homarine catabolism (Figure SX1B). Similarly, mutants from a Tn5 transposon barcoded mutant library in R. pomeroyi DSS-3 36 were used to assess gene phenotypes, (Figure SX1A, Table S3 and S5). Knockouts of three catabolic genes ( homB , SPO3189; homC , SPO3190; homD , SPO3191) and the transcriptional regulator ( homR , SPO3192) resulted in the loss of growth on homarine at 4 mM C (Figure 1D, Tables S3 and S5). These results confirm the role of the homarine operon ( homABCDER ) in the degradation of homarine in our model organisms. A knockout strain targeting the transporter gene (ACFLL8_11505) within the hom operon of Cobetia sp. OBi1 (Δ homT ) exhibited reduced growth on homarine compared to the wild-type strain. Plasmid-based complementation with homT restored growth relative to a control expressing red fluorescent protein (RFP) (Figure SX2). Given its location in the hom operon, its upregulation in response to homarine, and the observed growth effects, this gene likely encodes a homarine transporter, which we designate homT . However, Δ homT mutants eventually reached wild-type yields, suggesting functional redundancy in homarine uptake. Such redundancy is common for betaines in marine bacteria 37,38 and competitive inhibition of homarine uptake by other betaines and sulfoniums further supports this in microbial communities 12 . In R. pomeroyi DSS-3, the lack of mutant precluded testing the function of a non-orthologous transporter adjacent to the hom operon (SPO3186). To maintain focus, further characterization of homarine transporters was not pursued. Identification of homarine degradation products using mass spectrometry To investigate the catabolism of homarine, we analyzed the metabolomes of Cobetia sp. OBi1 grown under the glucose, homarine, and glucose + homarine conditions. Metabolites enriched in homarine treatments were identified in intra- and extra-cellular metabolomes (Figure 2A, Figure SX3), calculated relative to “core” metabolites that are relatively constant across a wide range of marine organisms and environmental regimes 11 . Homarine, glutamic acid, and several unidentified compounds were enriched (Figure SX3, Tables S6 and S7). Two compounds were subsequently identified: N -methylglutamic acid (via comparison to standard), and N -methyl glutamine (via MS 1 /MS 2 spectra, Figures SX4 and SX5). To explore conserved degradation products of homarine, we analyzed the metabolomes of R. pomeroyi DSS-3 grown under similar glucose and homarine conditions. We found the enrichment of N- methylglutamic acid, N- methyl glutamine, and compounds with molecular formulas of C 6 H 9 NO 3 , C 6 H 9 NO 4 , and C 7 H 9 NO 5 , when grown with homarine, consistent with Cobetia sp. OBi1 (Figure 2A-C, Tables S8 and S9). Stable-isotope probing was performed to track homarine degradation products in natural marine microbial communities from three locations (Figure 2D). Seawater incubated with isotopically-labeled homarine ( 2 H 3 -homarine for TN412 experiments; 13 C 7 15 N-homarine for RC104 and TN412 experiments, Figure 2D inset) was analyzed for the compounds enriched in our model organisms, with the isotopic labels. In the TN397 experiments, 3 H 3 -labeled N- methylglutamic acid and N- methyl glutamine were enriched, alongside the previously observed unidentified compounds with formulas of C 6 H 9 NO 3 , C 6 H 9 NO 4 , and C 7 H 9 NO 5 (Figure 2E, Table S9). Consistent with the removal of a 2 H 3 - methyl group from N -methyglutamic acid, we did not find 2 H 3 -glutamic acid (Table S9). In the RC104 and TN412 experiments, we identified enrichment of fully 13 C 5-7 , 15 N 1 - N- methylglutamic acid, glutamic acid, N- methyl glutamine, and C 6 H 9 NO 3 , C 6 H 9 NO 4 , C 7 H 9 NO 5 features, confirming them as direct degradation products of homarine. (Figure 2F, Table S9). The consistency in the enriched metabolites in R. pomeroyi DSS-3, Cobetia sp. OBi1, and natural microbial communities support a conserved degradation of homarine to N- methylglutamic acid and subsequently to glutamic acid. Complete isotopic labeling of N -methylglutamic acid and glutamic acid confirms their origins from homarine while the isotopic labeling of 2 H 3 - N- methylglutamic acid highlights the preservation during ring opening and subsequent removal of homarine’s methyl group during degradation to glutamic acid. Untargeted analysis of the 2 H 3 -homarine experiments found no evidence for 2 H 3 -containing compounds beyond N- methylglutamic acid and homarine (Table S10), confirming that homarine’s pyridine ring is opened before demethylation. While earlier studies imply that homarine functions as a methyl donor to form picolinic acid 39 , in these stable isotope experiments we only observed the methylated pyridine ring opening intact during breakdown. Role of N -methyl glutamate dehydrogenase in homarine degradation The discovery of N -methylglutamic acid as a product of homarine degradation prompted us to reexamine upregulated genes in Cobetia sp. OBi1 for involvement with this metabolite. Two gene clusters originally annotated as coding for the heterotetrameric sarcosine oxidase (TSOX) were identified: ACFLL8_02220-02235 and ACFLL8_07290-07305, with the latter showing increased transcription during growth on homarine (Figure 1D, Table S2). Genes encoding TSOX ( soxBDAG ) and N -methyl glutamate dehydrogenase (NMGDH; mgdABCD ) often share high sequence similarity, leading to misannotation by automated pipelines 40,41 . Both enzymes act on a methylated secondary amine, with similar CH 3 NHCH(R)COOH backbones (R = H in sarcosine, R = propanoic acid for N -methylglutamic acid). Phylogenetic analysis revealed that the sox- like genes induced in Cobetia sp. OBi1 (ACFLL8_07290 to ACFLL8_07305) are more closely related to mgd genes from bacteria with demonstrated NMGDH activity than those with TSOX activity (Figures SX6-SX9). In R. pomeroyi DSS-3, distinction between the sox and mgd gene clusters was similarly apparent after phylogenetic analysis (Figure SX6-SX9). Furthermore, growth of mgdA ::Tn5 and mgdD ::Tn5 mutants of R. pomeroyi DSS-3 was significantly reduced when grown on homarine or N -methylglutamic acid compared to the wild type (Figure 1F, Table S5). Together, these results suggest that the sox -like genes upregulated in Cobetia sp. OBi1 correspond to the mgd gene cluster. Previously, the only known route for the formation of N -methylglutamic acid was as an intermediate in glutamate-assisted methylamine utilization, where a methyl group is transferred from methylamines onto glutamic acid, followed by NMGDH-catalyzed demethylation to glutamic acid and formaldehyde or 5,10-methylene-H 4 folate 41,42 . Our field experiments revealed accumulation of 13 C 5 , 15 N-glutamic acid and 13 C 6 , 15 N- N- methyglutamic acid from 13 C 7 , 15 N-homarine, demonstrating a new pathway for the formation of N- methyglutamic acid during the conversion from homarine to glutamic acid (Figure 1C, 2). In 2 H 3 -homarine incubations, 2 H 2 -labeled metabolites were detected (Table S10), consistent with methyl group transfer into the folate cycle. Combined with mgd upregulation in Cobetia sp. OBi1, these results support the role of NMGDH in catalyzing the conversion of homarine-derived N- methyglutamic acid to glutamic acid natural communities (Figure 1E). Glutamic acid can be converted into alpha-keto glutarate for entry into the Krebs cycle, play a central role in nitrogen assimilation via glutamine oxoglutarate aminotransferase (GOGAT), contribute to the biosynthesis of metabolites, peptides, proteins, or function as an osmolyte. This positions homarine as a versatile growth substrate for bacteria equipped with genes to convert homarine into glutamic acid. Homarine-degrading bacteria with conserved homarine catabolic genes To evaluate the distribution of the genetic mechanisms for homarine catabolism, we isolated 26 bacteria strains from coastal and pelagic environments using homarine-enriched agar plates and sequenced their genomes. Sixteen isolates with <99% Average Nucleotide Identity (ANI) were selected for further analysis and screened for axenic growth in oligotrophic seawater supplemented with homarine and phosphate; nine of these were capable of growing on homarine as a sole carbon and nitrogen source. Homarine catabolism was taxonomically restricted to Alphaproteobacteria and Gammaproteobacteria. All homarine-utilizing isolates encoded homABCD , homR, and sometimes homE in a conserved operon (Figure 3A, Table S11). An mgd operon was identified in 8 of 9 genomes with the hom operon, and all mgd -positive isolates also carried soxGADB (Figure 3A, Table S11, Figures SX6-9). Notably, Vibrio isolate PS01 contained the hom operon but lacked mgdABCD, yet still grew on homaine, suggesting genome incompleteness or an alternative pathway for processing N -methylglutamic acid. One isolate, Nitratireductor sp. G4i25, encoded mgdABCD but lacked the hom operon and could not grow on homarine, reinforcing the need of both gene sets for homarine catabolism (Figure 3A, Table S11). Widespread distribution and conservation of hom operon To explore the taxonomic and geographic distribution of bacteria capable of homarine degradation, we searched complete genomes in the NCBI database for operons encoding the homarine catabolic genes ( homA BCD and optionally homE ). Due to sequence similarity between sox and mgd genes, we did not attempt the same search for the mgd operon 40 (Figures SX6-SX9). Across 11,755 genomes containing the hom operon, we identified primarily Alphaproteobacteria (3,148 genomes, 26.8%) and Gammaproteobacteria (8,094 genomes, 68.9%) lineages (Figure 3B, Table S12). Operon organization was highly conserved; homA and homB were adjacent in 97.7% of genomes, consistent with the proposed homodimer structure, while homC and homD were typically nearby (Figure SX10A, Table S12). In contrast, the presence and positioning of homE was variable (Figure SX10A, Table S12). Approximately one-third of the genomes containing homABCD do not include homE . In genomes with homE , it is positioned upstream of homA in 49.9% (like Cobetia sp. OBi1) and downstream of homD in 50.1% (Figure SX10A, Table S12). This variability suggests homE functions as an auxiliary module in the hom operon, but not a core component. Gammaproteobacteria retained homE in 93% of the genomes containing the hom operon (Table S12), suggesting it may extend or enhance catabolic potential. By contrast, 94% of the Alphaproteobacteria genomes lacked homE - including all of the 120 genomes of the highly abundant Pelagibacter clade (Figure SX10A, Table S12). The absence of homE in Alphaproteobacteria, particularly in smaller genomes ( t -test, p = 2.2×10⁻¹⁶, n = 11,755, Figure SX10B), reflects relaxed selective pressure. These observations support a model where homE serves a taxon-specific auxiliary role for homarine metabolism. We identified bacteria encoding the hom operon from a global range of ecosystems, including soil, freshwater, marine, sediments, food (e.g. . cheese, oysters), and human clinical samples ( e.g ... urine, fecal, Figure 3C, Table S12). Although gene presence does not guarantee functionality, detecting the operon in non-aquatic environments suggests homarine may play roles in microbial metabolism beyond marine systems. Among genomes encoding the operon were members of ecologically significant marine heterotrophic clades, including SAR11 (Pelagibacterales) 43 , SAR116 (Puniceispirillales) 44,45 , SAR92 (Cellvibrionales) 45,46 , and Rhodobacterales including members of the Roseobacter clade 47,48 (Figure 3C, Table S12). Notably, SAR11 accounts for ~25% of microbial cells in the ocean 49 , SAR92 represents up to 10% of cells in nearshore waters 50 , and Roseobacters comprise up to 20% of bacterial cells in coastal ecosystems 51 . SAR11’s streamlined genome retains the hom operon, underscoring its importance to Pelagibacterales and suggesting a key role for homarine in their ecological success. The widespread occurrence of the hom operon across diverse regimes and its presence in dominant marine microbes highlights homarine’s broad ecological significance and its pervasive role in microbial metabolism. Expression of hom catabolic genes in marine microbes To determine whether the genomic potential of homarine catabolism translates to metabolic activity in situ , we evaluated gene expression under experimental and environmental conditions. In homarine-amended incubations in Puget Sound, significant transcriptional enrichment was detected for 13 bacterial genes relative to unamended controls (Table S13). Transcripts of homC and homD were enriched in members of the Porticoccaceae (Gammaproteobacteria), and homB was enriched in Oceanospirillaceae (Alphaproteobacteria) (Table S13) . These taxa were identified in our genome search with the genetic capacity for homarine catabolism (Figure 3B, Table S12). Additionally, mgdA, the gene encoding the α -subunit of the NMGDH, was up-regulated in Porticoccaceae (Table S13), reinforcing the role of NMGDH in converting homarine-derived N- methyglutamic acid to glutamic acid and linking the genomic capacity to catabolic activity in natural microbial assemblages. We examined the in situ expression of the hom genes using quantitative metatranscriptomics along a surface ocean transect from the subtropical gyre into the North Pacific Transition Zone (25.87°N to 40.88°N on 158°W). Spatially structured expression of homB and homA was observed, with elevated transcript abundance north of 35°N (Figure 4, Figure SX11). Expression of homB was primarily attributed to Gammaproteobacteria (e.g. Vibrionales, Porticoccaceae) and Alphaproteobacteria (e.g. Pelagibacter , SAR116, Rhodobacterales), alongside unresolved lineages (Figure 4, Table S14). Across stations, homarine concentrations (sum of dissolved and particulate measurements) were significantly correlated with hom gene transcript abundance ( R² = 0.518, p = 0.0052, Spearman’s correlation), supporting coupling between substrate availability and gene expression. Taxa expressing hom genes in situ overlapped with those identified in our perturbation experiments and predicted to encode homarine catabolic pathways from genome-based analyses. These taxa spanned both copiotrophic and oligotrophic lineages, such as Porticoccaceae, Vibrionales, Pelagibacteraceae, SAR116, Oceanospirallaceae, and Rhodobacterales. The detection of transcripts from both fast-growing coastal taxa and genome-streamlined open-ocean bacteria suggests homarine catabolism operates across diverse ecological strategies and niches. These findings highlight homarine degradation as a responsive and ecologically widespread metabolic capacity, with gene expression linked to substrate availability in marine environments. Discussion This study defines a core microbial pathway for the degradation of homarine, a ubiquitous yet previously enigmatic metabolite in marine systems 6,7,11 . Using comparative transcriptomics, transposon mutagenesis, and targeted gene deletions, we identified a conserved operon ( homABCDER ) that mediates homarine catabolism in two model organisms, a newly isolated gammaproteobacteria Cobetia sp . OBi1 and the genetically tractable model marine alphaproteobacteria Ruegeria pomeroyi DSS-3. Metabolomic analyses in both model organisms and natural communities reveal that homarine degradation produces N -methylglutamic acid, which is subsequently converted into glutamic acid by N -methylglutamate dehydrogenase (NMGDH). This refines the understanding of NMGDH, a gene that is frequently misannotated 40,41 , and expands its ecological importance in marine microbial metabolism. As a metabolic product of homarine, glutamic acid plays diverse biochemical roles, including nitrogen assimilation, biosynthesis of metabolites and proteins, and energy metabolism. These findings position homarine as a versatile growth substrate and highlights homarine’s broad impact on microbial growth and resource allocation. The hom operon is broadly distributed across bacterial genomes from diverse environments, including coastal and open-ocean systems, as well as non-aquatic sources. Within bacteria with demonstrated homarine-degradation capabilities, there is a frequent co-occurrence of the hom operon with the mgd gene cluster, which encodes NMGDH and facilitates the conversion of N -methylglutamic acid to glutamic acid. This genomic association underscores the interdependence of the hom and mgd pathways, highlighting their integrated role in nutrient cycling. Genomic mining further revealed that the hom operon is conserved in globally dominant bacterial clades that are key drivers of carbon and nitrogen cycling in marine ecosystems, including SAR11 43 , SAR116 44,45 , and Rhodobacterales 47,48 . Furthermore, variability in the auxiliary homE gene reveals operon modularity and its adaptive tailoring to ecological or genomic constraints. Through perturbation experiments and in situ analyses of gene expression, we show the coupling of homarine availability to homB gene expression in natural marine environments. This demonstrates a clear relationship between substrate availability and microbial metabolic activity, supporting homarine as a regulatory driver of microbial gene function. The responsiveness of these genes to substrate availability suggests that homarine degradation actively shapes microbial community structure and metabolic dynamics in situ . Co-expression of chemotaxis and motility genes alongside the hom operon indicates that bacteria not only sense and metabolize homarine but may actively seek it out in the environment. These findings suggest that homarine degradation contributes to the structuring of microbial communities and influences resource-driven interactions in the ocean. Collectively, this study provides a foundational framework to link metabolic gene function to ecological activity, bridging molecular mechanisms with environmental significance. By elucidating the genetic basis, biochemical transformations, and ecological distribution of homarine catabolism, we expand our understanding of how microbial communities integrate metabolite turnover into broader oceanic biogeochemical cycles. These findings open opportunities for further exploration into how homarine interacts with co-occurring metabolites, impacts microbial interactions, and supports ecosystem functioning in the ocean and beyond. Methods Isolation, growth, and sequencing of Cobetia sp. OBi1 Isolate Owen Beach isolate 1 (OBi1) was obtained by enrichment culturing of seawater collected in Owen Beach (Tacoma, Washington). Seawater was collected in sterile polycarbonate bottles in February 2020. We amended 100 mL seawater samples with 12 mM C homarine and 0.1 mM sodium phosphate and incubated in aerated flasks at 22°C. Culture OD600 was measured daily to monitor growth. After three days, 50 uL samples were streaked on agar plates made with marine broth (Difco 2216) to isolate single colonies. Single colonies were picked and re-streaked twice in the same agar media. The isolated colonies were tested for growth on homarine in seawater from Owen Beach sterilized by filtering through a 0.22 um PES membrane and autoclaving. We supplied the medium with homarine and phosphate as before. OBi1 colonies re-grew on homarine and the isolate was selected for further analysis. We grew OBi1 in seawater medium containing homarine and phosphate in aerated flasks at 25°C to obtain DNA for genome sequencing. Genomic DNA was purified with the NEB Monarch tissue lysis kit using a modified lysis solution containing lysozyme, Proteinase K and RNase A. For whole genome sequencing, we generated a set of Illumina paired-end reads with a depth of 400 Mbp on a NextSeq 2000 and a set of Oxford Nanopore reads at the Microbial Genome Sequencing Center (Pittsburgh, PA). Quality control and adapter trimming was performed with bcl2fastq 52 and porechop 53 for Illumina and Nanopore reads, respectively. A hybrid assembly with both sets of reads was performed with Unicycler using the default parameters 54 . Assembly statistics were recorded with QUAST 55 . The assembly annotation was performed with Prokka 56 . The genome was analyzed with the NCBI prokaryotic annotation and TypeMat tools from the Microbial Genomes Atlas (MiGA) web server 23 to resolve its taxonomic classification and completeness. Transcriptomics of Cobetia sp. OBi1 grown on homarine We generated and compared transcriptomes of OBi1 cultures grown in three conditions with (i) glucose, (ii) glucose+homarine, or (iii) homarine to identify genes involved in homarine catabolism. The growth media for transcriptomics experiments were prepared with sterile-filtered and autoclaved oligotrophic surface seawater from Hawaii to reduce background carbon and nutrients. The medium was amended with PRO99 trace metals and 0.1 mM sodium phosphate. Growth medium (i) was amended with 12 mM carbon as glucose and 0.8 mM nitrogen as ammonium. Growth medium (ii) was the same as medium (i) but was spiked with 1 mM carbon as homarine 30 minutes before RNA extraction. Growth medium (iii) contained 12 mM carbon as homarine and no additional nitrogen source. Before the transcriptomics experiment, we grew 5-mL cultures of OBi1 in each medium (i-iii) to determine the OD600 of cultures in the mid-exponential phase. For the transcriptomic experiment, we pre-grew OBi1 in 5 mL of glucose medium (i), of which 4 mL were pelleted and washed three times with 1 mL of sterile oligotrophic Hawaii seawater and resuspended back in 4 mL of seawater. We used 0.5 mL of washed cells to inoculate 75 mL of each growth medium (i-iii) in duplicate for transcriptomics. Cultures were incubated in aerated flasks at 25°C and 200 rpm for a period of 29-30 hours during which we sub-sampled 0.5-1 mL from each replicate to track growth by OD600. Absorbance measurements were obtained with a Genesys 20 spectrophotometer (Spectronic Instruments). Cells were harvested from 40 mL per replicate when cultures were in mid-exponential phase. Samples were split before RNA extraction with a Zymo Research Quick-RNA™ Fungal Bacterial MiniPrep kit. DNA was digested with DNase I from NEB. RNA samples were repurified with a Zymo Research RNA Clean & Concentrator™-25 kit. The samples obtained were measured in a NanoDrop™ (Thermo Fisher Scientific) and those with > 50 ng/μL concentration were selected for RNA sequencing. RNA samples were sent to the Microbial Genome Sequencing Center for RNA-Seq. DNA samples were treated with RNase free DNase (Invitrogen). Library preparation was performed using Illumina’s Stranded Total RNA Prep Ligation with Ribo-Zero Plus kit and 10bp IDT for Illumina indices. Sequencing was done on a NextSeq2000 giving 2x51bp reads. Demultiplexing, quality control, and adapter trimming was performed with bcl2fastq (v20.20.0.445). RNA-Seq generated > 12 M paired-end reads per sample. RNA-Seq paired-end reads were aligned to the OBi1 genome assembly to quantify the number of reads mapped to each coding gene using Rsubread 57 . DESeq2 was used to normalize gene counts and resolve differentially expressed genes 58 between each pair of OBi1 growth conditions. Significant results were resolved with a log-fold change threshold > (+/-) 1.99 and p <0.05 (Wald test) following adjustment for false discovery. Functions and structures of the homABCDER operon To functionally annotate proteins of the homABCDER operon, we used domain and pathway annotation tools including InterProScan 29 , KEGG 26–28 , and COFACTOR 32,33 . Gene Ontology (GO) terms and enzyme classifications were retrieved from InterPro and COFACTOR, and annotations were retained when COFACTOR CscoreGO values exceeded 0.65. KEGG annotations were obtained by querying the KEGG REST API using gene identifiers for R. pomeroyi DSS-3 (SPO3188–SPO3192). Structural models of HomA and HomB were generated using AlphaFold 30,31 , and the highest-scoring model for each protein was selected based on the combined iptm + ptm confidence score. Predicted structures were visualized in ChimeraX 34 to assess domain organization and potential protein–protein interaction interfaces. Generation of mutants of homarine operon in Cobetia sp. OBi1 All Cobetia sp. experiments were conducted using the OBi1 strain. Plasmids were conjugated into OBi1 using the donor bacterial strain Escherichia coli WM3064, [a derivative of B2155 59 provided by W. Metcalf (University of Illinois, Urbana)], which can grow only in the presence of diaminopimelate (DAP). WM3064 carrying plasmids was grown overnight at 30-37°C, shaking in LB media containing 30-50ug/mL kanamycin final concentration and 0.3 mM DAP final concentration. OBi1 was grown overnight at 25-30°C, shaking in Marine Broth 2216 (MB) media (either Difco or NutriSelect® Plus). To set up conjugations, overnight cultures were mixed in a 1:1 volume ratio, pellet for 1 min at 14000 rpm, resuspended in 1/30th of its initial mixed volume, and spotted onto an MB agar plate containing 0.3 mM DAP. Conjugations were then incubated at room temperature for at least 24 hours. Transconjugant selection was performed by dilution streaking for single colonies from overnight spotted mating on MB agar plates without DAP. The kanamycin concentration of MB agar plates for OBi1 transconjugants is 5 ug/mL for integrating plasmids and 25 ug/mL for replicating plasmids. Strains harboring replicating plasmids for empty vector, fluorescence, or genetic- complementation used the pBVMCS-2 plasmid 60 or the pGingerBK-LacUV5 plasmid 61 . Both plasmid transformations used the conjugation protocol outlined in the previous paragraph. Gene deletion strains were generated using the integrating plasmid pNPTS138 (2) for homD (protein accession ACFLL8_11515) and a modified pNPTS138 for homT (protein accession ACFLL8_11505). pNPTS138 transformation and integration occurs at a chromosomal site homologous to the insertion sequence in pNPTS138. The modified pNPTS138 replaced the aph(3’)-Ia kanamycin resistance gene with aph(3’)-IIa, added the lacI gene, and replaced the sacB gene promoter with the lac-operator. Resistance cassette, lacI, and lac-operator were sourced from 62 ; henceforth, the modified pNPTS138 plasmid will be called pMZT1. Single colony transconjugants were inoculated into liquid MB media at 25-30°C shaking for 48-72 hours for non- selective growth. Nonselective liquid growth allows for a second recombination event to occur, which either restores the native locus or replaces the native locus with the insertion sequence that was engineered into pNPTS138. Counter-selection for the second recombination of pNPTS138 excision was carried out by passing non-selective cultures (1:100 dilution) into new liquid MB media with 10-20% (w/v) final sucrose concentration for 6-8 hours. For counter-selections with pMZT1 a final concentration of 0.5 mM IPTG was added. Serial dilutions of counter-selection cultures were plated on MB agar with 10-20 % sucrose, and for counter-selection with the pMZT1 plasmid, the sucrose agar contained 0.5mM IPTG. Colonies were subjected to PCR genotyping and sequencing to confirm allele replacement. For homD deletion analysis, primers homD_KO check F/R were utilized, while primers homT_comp F and homT_KO check R validated the deletion of homT (Table S16). Plasmid construction involved various combinations of restriction enzyme digestion, PCR amplification, and Gibson Assembly. The plasmid pNPTS138- homD -KO was generated by double digestion of pNPTS138 with NheI and HindIII followed by assembly with a synthesized in-frame knockout (KO) allele of homD from GeneWiz, then transformed into chemically competent WM3064 cells. Similarly, pMZT1- homT -KO was constructed by double digestion of pMZT1 with NheI and HindIII and assembly with a synthesized KO allele of homT from GeneWiz, followed by Gibson Assembly and transformation into WM3064 cells. The pMZT1 plasmid itself was created using two PCR fragments—one amplified from pGinger-LacUV (2.6 kb) and another from pNPTS138 (4 kb)—assembled via Gibson Assembly and transformed into Zymo Mix and Go DH5-alpha chemically competent cells. For pBVMCS- homD , pBVMCS-2 was linearized via double digestion with EcoRI and NdeI and assembled with a PCR fragment containing the native homD promoter and coding sequence amplified from OBi1 genomic DNA (gDNA), followed by transformation into WM3064 cells. Construction of pLacUV-homT involved linearization of pGinger-LacUV with EcoRI and BamHI, assembly with a PCR fragment containing the homT coding sequence amplified from OBi1 gDNA, and placement of homT under an IPTG-inducible promoter. The final construct was assembled with Gibson Assembly and transformed into WM3064 cells. Phenotype experiments with Cobetia sp. OBi1 knockout mutants To determine OBi1 requires homD to grow on homarine, we grew the wild type and Δ homD strains in seawater medium containing homarine at 12 mM carbon concentration. We used a medium with glucose at 12 mM carbon and 1.7 mM ammonium as a positive control for growth. The base seawater medium contained PRO99 trace metals and 0.2 mM phosphate to support growth. Wild type and Δ homD strains were grown in ½ YTSS medium without antibiotics. The ΔhomD complementation strain (Δ homD +pBVMCS- homD ) and its empty vector control strain (Δ homD +pBVMCS) were grown with kanamycin (50 µg/mL) selection. Strains were grown from a single colony in ½ YTSS overnight to an OD600 of 0.8. Culture aliquots of 1 mL were pelleted at 3000 x g to harvest cells. Cells were washed five times with base seawater medium and starved overnight to reduce background growth. Washed cells were diluted 100-fold in seawater and 10 uL were arrayed into 96-well plates containing 0.3 ml per well of the homarine or glucose test media. The plates were incubated on a shaker at 22°C removed from light. Discrete OD600 measurements were obtained on a SpectraMax Plus microplate reader (Molecular Devices). The background seawater absorbance was subtracted from all measurements to quantify growth. To compare the growth of wild-type OBi1 and Δ homT , cultures were pre-grown in ½ YTSS incubated overnight at 200 RPM and 25°C. Overnight cultures of Δ homT plasmid complementation strains incubated with 0.5 mM IPTG to induce RFP or HomT expression. Cell pellets of 1 mL of culture were washed five times in base seawater medium. Washed cells were diluted twenty-fold in seawater and 10 uL were arrayed into 96-well plates containing 0.3 ml per well of the homarine or glucose test media. Plasmid complementation strains incubated with varying concentrations of IPTG (0-0.5 mM). The plates were covered with breathable film to reduce evaporative water loss and incubated at 25°C in the microplate reader to record absorbance and RFP fluorescence (excitation and emission). A logistic growth model was implemented in R with the logit and non-linear least squares (nls) function in the car package to estimate the growth rate parameter. A one-way ANOVA and the median test for multiple comparisons from the agricolae package were implemented in R to test for significant differences in growth rates. Phenotype experiments with mutants of Ruegeria pomeryoi DSS-3 Gene clusters for Cobetia sp. OBi1; ACFLL8_11490 ( homE ), ACFLL8_11495 ( homA ), ACFLL8_11500 ( homB ), ACFLL8_11510 ( homC ), ACFLL8_11515 ( homD ), ACFLL8_11520 ( homR ) were pairwise aligned to Ruegeria pomeroyi using blastp with a BLOSUM62 matrix 63 , Table S3). To confirm the involvement of homarine catabolic genes, we utilized RB-TnSeq mutants of Ruegeria pomeroyi DSS-3. Detailed methods for generating and arraying the R. pomeroyi DSS-3barcoded mutant library are provided and summarized in previous work 36 64 . Mutant cultures were pre-grown overnight in ½ YTSS broth containing 50 μg/ml kanamycin. Screens were performed in L1 minimal medium amended with Basal Medium vitamins 65,66 , 0.8 mM ammonium, 50 μg/ml kanamycin, and 100 μM phosphorus as PO 4 3- . Washed (3x) overnight cultures of individual mutants were inoculated at an OD600 of 0.01 into modified L1 medium with a single substrate as the sole carbon source at 4 mM carbon, n = 6. Plates were incubated at 25°C with shaking at 425 rpm for 72 hours, and optical density (OD600) was measured at 5-minute intervals using a Synergy H1 microplate reader (BioTek Instruments, Inc., Vermont, USA), corrected to a pathlength of 1 cm, assuming a volume of 200 μl. As a positive control, the same medium was inoculated with washed overnight cultures of wild-type R. pomeroyi DSS-3, n = 3. Wild-type R. pomeroyi DSS-3 was also grown in kanamycin to test the effectiveness of the antibiotic, n =3. For media control, wells filled with 200 μl of medium without inoculum were incubated, n = 3. Preparation, mass spectrometry, and analysis of metabolomics experiment with Cobetia sp. OBi1 Cobetia sp. OBi1 was grown in three conditions, as described for comparative transcriptomic analyses: homarine, glucose, and glucose+homarine. Overnight cultures of Cobetia sp. OBi1 (100 mL) were grown at 25°C grown in glucose-amended seawater media as described for Cobetia sp. OBi1 transcriptomics experiments. Next, 10 mL subsamples were taken from overnight culture and centrifuged for 15 minutes at 2800 g, and resuspended in the experimental growth media homarine, glucose, or glucose+homarine, in triplicate (again, as described for Cobetia sp. OBi1 transcriptomics experiments). These samples were incubated for 1 hr at 25°C in a dark shaker before harvesting. Cells were harvested by centrifugation for 15 minutes at 2800 g, and supernatant was collected and filtered through a 0.22 um PES membrane filter. Cell pellets and supernatant samples were stored at -80°C and -20°C, respectively. For particulate metabolomics, cell pellets were extracted using a combination of mechanical and chemical disruption techniques as described in previous work 67 . Metabolites from the supernatant were extracted using a cation-exchange-based solid phase extraction technique as described previously 7 , with 1 mL of supernatant diluted into 10 mL of HPLC grade water. Isotopically-labeled internal standards were added for normalization purposes, as reported in Table S17. Metabolomics data were acquired by liquid chromatography paired with high resolution mass spectrometry (LC-MS) on a ThermoOrbitrap Q-Exactive HF Mass Spectrometer (QE). Particulate analyses were performed as reported 67 with modifications reported previously 68 . In short, samples were introduced via Hydrophilic interaction liquid chromatography (HILIC) in both positive and negative modes using polarity switching. All samples were introduced in full scan mode. In addition to the full scan analyses, samples were pooled and monitored in data-dependent acquisition (DDA) mode to acquire MS 2 data. These data were acquired at three different collision energies (20, 35, and 50 V), with separate injections for each collision energy and ionization mode (positive or negative). Pooled samples were injected at full strength and also diluted 1:1 with water to aid in normalization, as previously described 67 . Data from the QE (.raw files) were converted to the open source .mzML file format using MSConvert 68 . For targeted analyses, peaks were integrated in Skyline 69 using a template of known compounds for which we have authentic standards. Peaks were identified by comparing to standards run in reconstitution solvent as well as spiked into a pooled sample. Further quality control was performed to remove small and low-quality peaks using the quality control procedure as described ( 67 . Quality control parameters for dissolved and particulate data from HILIC chromatography included a signal to noise ratio of at least 3, a mass within 6 ppm of the standards’ calculated mass, a retention time within 2.5 minutes of the standards’ retention time, a minimum area of 40000, and a ratio of signal over signal from blank of at least 3. Finally, peaks were normalized to reduce variability introduced during data acquisition using best-matched internal standard (B-MIS 67 . For untargeted analyses, we used MS-Dial v4.9 70 to extract mass features using the parameters supplied in Table S18. This resulted in a list of mass features ( m/z and retention time features) that could be compared quantitatively between samples, regardless of the ability to identify these compounds. We dereplicated mass features from MSDial by identifying and annotating adducts (including between positive and negative polarity) and isotopes based and consolidated mass features for downstream statistical analyses. Mass features that correspond to our targeted compounds were annotated accordingly. For instances where a mass feature was not detected in all samples of an experiment, we replaced missing values with 0.02 times the minimum detected value for that mass feature within the same experiment, ensuring all features have a baseline value to facilitate calculations and comparisons in downstream analyses. Next, we performed an enrichment analysis to calculate the relative enrichment of individual metabolites or mass features between treatments when compared to core metabolites. This is similar to normalizing mass feature area to biomass, a common normalization technique, but does not rely on a separate biomass measurement. Instead, the enrichment analysis calculates a scaled area for each compound or mass feature relative to each ‘core metabolite’ to assess its enrichment under different treatment conditions. Core metabolites were decided based on previous work 6 that identified metabolites that follow biomass trends by depth and latitudinally and are observable in most phytoplankton when analyzed under our analytical conditions - these are reported in Table S9. These scaled areas are used to compute fold changes and perform statistical analyses. To account for the fact that some of the ‘core metabolites’ may also be affected by treatments, we performed an outlier detection step, where Grubbs' test 71 is applied to identify extreme values in fold changes for core metabolites, and those core metabolites detected as outliers were excluded from further analysis. After filtering, the final enrichment results are summarized by calculating the median p -values (after applying a Benjamini Hochberge false discovery rate correction 72 and fold changes for each metabolite across sample treatments. Preparation, mass spectrometry, and analysis of metabolomics experiment with Ruegeria pomeryoi DSS3 Cultures of Ruegeria pomeroyi DSS-3 were revived from cryostocks onto ½ YTSS agar plates and incubated at 30 °C for 6 days. Single colonies were inoculated into 11 mL of glucose minimal media (GMM) and grown overnight at 30 °C with shaking at 200 rpm. GMM was prepared using a modified L1 minimal medium with glucose 12 mM C as the sole carbon source. All cultures were maintained in sterile 15 mL assay tubes. Overnight cultures were diluted to an optical density of 0.1 at 600 nm (OD₆₀₀) in fresh GMM, incubated for 11–12 h, and amended with glucose (4 mM C, 0.8 mM NH 4 , control), homarine (500 nM C, no additional NH 4 ), or glucose + homarine (1 mM C, 2 mM C, respectively with 0.8 mM NH 4 ). Homarine additions were staggered across time points to ensure consistent incubation durations. At each time point, samples were collected for cell counts and particulate metabolites. For cell counts, 1 mL of culture was fixed with glutaraldehyde (final concentration 1%) in labeled cryovials, held at 4 °C for 20 minutes, and then stored at –80 °C. Particulate metabolites were collected by filtering cultures through combusted glass fiber filters using a vacuum manifold set to 8 psi; filters were wrapped in combusted foil and flash-frozen in liquid nitrogen. The experiment cultures were sampled after two hours in biological triplicate, as well as the three control conditions: glucose-only controls, glucose plus homarine controls (uninoculated), and a filter blank. For metabolite extractions, a one phase extraction was performed with 40:40:20:0.01 methanol:acetonitrile:water:formic acid solution as the extraction solvent 73 . Filters were placed in 15 mL Teflon tubes with pre-chilled extraction solvent, incubated at -20 ºC for 10 minutes, bead beaten with silica beads, and centrifuged. The solvent was then collected, transferred into glass tubes, and the procedure was repeated three times while keeping samples cold as much as possible. Samples were dried down under nitrogen gas, reconstituted in 400 uL of H2O, and stored at -80 ºC until analysis by LC-MS. Isotopically-labeled internal standards were added for normalization purposes, as reported in Table S17. Samples were filtered, extracted, and run on the instrument as reported for the metabolomics experiment for Cobetia sp. OBi, but only particulate samples were acquired and run on the HILIC method for LCMS, as these data were the most promising for identification of intermediates of homarine based on the Cobetia sp. OBi experiment. We used Skyline 69 to mine the Ruegeria pomeryoi DSS-3 metabolomics data for features that showed enrichment in Cobetia sp. OBi under homarine conditions, and all the core metabolites, as reported in Table S9. These features underwent the same normalization and calculations for enrichment as the Cobetia sp. OBi data as reported above. Field experiments using isotopically labeled homarine Experiments using 2 H 3 -homarine were performed on research cruise TN397 in the Fall of 2021 at two different stations in the North Pacific (described in Table S8 and displayed in Figure 2). 2 H 3 -homarine was purchased from Toronto Research Chemicals and was injected onto a Q-Exactive HF Orbitrap Mass Spectrometer (QE-HF) to confirm the mass of the deuterium label (141.0743 m/z) and the retention time (6.4 minutes, same as non-labeled homarine). Seawater was collected into 21 acid washed 10 L polycarbonate carboys from a trace metal clean stayfish system suspended at a depth of 8 m and prefiltered through 100 µm nylon mesh. Three unamended samples were collected immediately after seawater collection to provide samples of the starting community. Nine treatment bottles were spiked with 500 nM 2 H 3 -homarine with nine control bottles receiving no additions. Bottles were incubated in blue-shaded temperature and light-controlled incubators designed to mimic mixed-layer conditions of the sampling location. Triplicate bottles with and without homarine addition were harvested at 2, 24, and 48 hours. All particulate samples (4 L) were collected using peristaltic pumps onto Durapore® 0.22 μm, 47 mm, hydrophilic PVDF membrane filters, flash frozen in liquid nitrogen, and stored at -80°C. Additional stable isotope experiments with 13 C 7 - 15 N-labeled homarine were performed on two different cruises: TN412 (Winter 2023) in the North Pacific, and RC104 (Summer 2023) in Puget Sound (described in Table S8 and displayed in Figure 2). The preparation of the 13 C 7 - 15 N-labeled homarine is described below. For TN412, the experiments were performed at two different stations; for RC104, the experiment was performed at one station. For all TN412 and RC104 experiments, seawater was collected through a trace metal clean stayfish system suspended at a depth of 8 m prefiltered through 100 µm nylon mesh. Triplicate samples were collected into acid washed 2 L polycarbonate bottles, spiked with 90 nM of 13 C 7 - 15 N-labeled homarine, and incubated in temperature and light-controlled incubators for 5 different timepoints (6, 12, 24, 48 and 96 hours). Triplicates of spiked and unspiked samples were filtered as quickly as possible, no more than 30 minutes (T0, and unamended control samples, respectively). Particulate samples were collected identically as the TN397 experiment described above. To prepare 13 C and 15 N-labeled homarine, Synechococcus sp. WH8102 was grown in Pro99 media 74 prepared with 15 N sodium nitrate and 13 C sodium bicarbonate and no HEPES buffer. Synechococcus sp. WH8102 has been previously shown to produce a large amount of homarine 6 . Cultures were grown in 250 mL glass media bottles with minimal headspace. Cells were collected by filtration onto 0.2 µm Durapore filters using combusted borosilicate filter towers. All filters were stored at -80°C until extraction. Filters were extracted using 40:40:20:0.01 methanol:acetonitrile:water:formic acid solution as the extraction solvent in 15 mL Teflon tubes. The solvent was dried under N 2 gas, reconstituted with H 2 O, and injected onto a Q-Exactive HF Orbitrap Mass Spectrometer (QE-HF) to confirm the production of 3 C- 15 N-labeled homarine (146.0754 m/z). After confirming the 13 C- 15 N-labeling of homarine, Synechococcus sp. WH8102 was grown in approximately 10 L, collected after 7 days, and extracted as previously described. Next, this extract was purified for spiking into environmental samples for stable isotope experiments. Purification was done using cation exchange chromatography and a Supelcosil LC-SCX column (25cm x 4.6mm, 5µm particle size) with a Thermo Scientific Vanquish UHPLC, fraction collector, and diode array detector. Solvent A was water with 2% formic acid and solvent B was water with 1% formic acid and 100 mM ammonium formate. The column was held at 0% B for 2 minutes, ramped to 20% B over 10 minutes, ramped to 100% B over 8 minutes, held at 100% B for 5 minutes, and equilibrated back to 0% B for 5 minutes (total run time is 30 minutes). The column temperature was maintained at 25°C and the flow rate was 1.0 mL/min. Purification of labeled homarine was achieved by UV based collection of a peak at 270 nm wavelength and a retention time of 4.3 minutes, which was confirmed by a homarine standard. Fractions containing labeled homarine were pooled together, dried under N 2 gas, and injected into the QE-HF to confirm the purity of the labeled homarine (any interfering compounds were confirmed to be less than 10%). Samples were extracted, and data were acquired as reported for the metabolomics experiment for Cobetia sp. OBi1 above. To prevent confusion using the isotope labels, we used a subset of isotopically-labeled internal standards, as reported in Table S17. Using the approach described above for the Ruegeria pomeryoi DSS-3 metabolomics data, we mined these data for the core metabolites as well as the features that showed enrichment in Cobetia sp. OBi under homarine conditions, both monoisotopic and expected isotopologues, as reported in Table S9. Again, these features underwent the same normalization and calculations for enrichment as the Cobetia sp. OBi data as reported above, using unamended control samples for comparison. For the TN397 with 2 H 3 -homarine amendments, we also performed an untargeted analysis in an attempt to find additional features that would correspond to a methyl transfer. This approach is analogous to previous studies that utilized natural abundance isotopes to track sulfur and iron-containing metabolites 13,75 . We used CoreMS 76 to identify mass features using a persistent homology approach with liberal thresholds, yielding thousands of features per sample (of which many are expected to be false positives or very low abundance peaks). For each samples’ features, we identified pairs of features that could correspond to 2 H 3 , 2 H 2 , and 13 C isotopologues - these pairs needed to be within 0.1 minutes of one another, with a mass error of expected isotopologues of no more than 4 ppm. From there, we collected features with putative 2 H 3 , 2 H 2 isotopologues within all the samples into retention time and m/z groups and filtered out any features that were present in control samples or a single sample. We visually inspected these mass feature pairs and discarded pairs that did not have consistent peak shapes, resulting in 7 quality mass features (Table S10). Note that this approach will only yield features that have an observable level of the monoisotopic version of a compound in the samples. Isolation, preparation, and genome sequencing of additional marine isolates Additional homarine degrading bacteria were obtained by culturing bacteria in seawater samples on solid media made with sterile oligotrophic seawater, 1.5% ultrapure agarose (SeaKem), 12 mM carbon as homarine and 0.1 mM sodium phosphate. Seawater samples were collected at various nearshore locations in Puget Sound near Tacoma, Washington and in the equatorial North Pacific during Gradients IV cruise TN397. Bacteria were concentrated by vacuum filtration onto a 0.22 µm Supor membrane, resuspended in 1 mL of sterile oligotrophic seawater, plated on homarine challenge plates, and incubated 1-2 weeks at 25°C. Single colonies were re-streaked in marine broth agar plates to obtain axenic strains. The isolate collection was re-screened in oligotrophic seawater media containing homarine and phosphate to confirm growth on homarine. Two colonies per isolate were tested in 24-well plates containing 2 mL per well of media containing homarine or control media lacking homarine. Growth (OD600) was measured on a SpectraMax Plus microplate reader (Molecular Devices) after 4 and 8 days of incubation. Isolates with confirmed growth on homarine were regrown in 5 mL of marine broth medium for 1-2 days. Cultures were pelleted and stored at -20°C before DNA extraction. Bacterial cell pellets were rinsed in 5 mL of 0.9% saline solution by vortexing, 1 mL was transferred to sterile microtubes and centrifuged at 3500 x g for 5 minutes and the supernatant was discarded. DNA was extracted with a Monarch® Genomic DNA Purification Kit using the tissue lysis solution supplemented with lysozyme followed by treatment with Proteinase K and RNase A. Purified DNA was analyzed on a Nanodrop spectrophotometer. Samples with >20 ng/µL were sent to SeqCenter (Pittsburgh, PA) for Illumina whole genome sequencing. Sample libraries were prepared using the Illumina DNA Prep kit and IDT 10bp UDI indices, and sequenced on an Illumina NextSeq 2000, producing 2x151bp reads. Illumina reads were trimmed and assembled with Trim Galore 77 , DOI:10.5281/zenodo.5127898; DOI:10.14806/ej.17.1.200) and the Unicycler SPAdes-optimiser 54 , respectively. We annotated assembly contigs with Prokka 56 . Identification of the homarine operon and sox and mgd genes in isolates To identify genes involved in homarine catabolism, we conducted reciprocal best BLAST hit (RBH) analyses across isolate genomes. Protein sequences from all genomes were used to generate BLASTP databases (NCBI BLAST+), and an all-vs-all BLASTP search was performed using a parallelized script. Gene pairs were considered RBHs if they were each other’s top-scoring match by bitscore. RBH results were screened for homologs of the homABCDER operon from Ruegeria pomeroyi DSS-3 (e.g., SPO3186–SPO3192) and Cobetia sp . OBi1 (ACFLL8_11490–ACFLL8_11520), as well as for subunits of sarcosine oxidase ( sox ) and N -methylglutamate dehydrogenase ( mgd ). Reference sox subunits from Stenotrophomonas maltophilia (AAY34150.1 78 ) and Corynebacterium sp. (P40875.2 79 ) and mgd subunits from Cobetia sp. OBi1 (ACFLL8_07290-07305), R. pomeroyi DSS-3 (SPO1585-1588), Methylocella silvestris (WP_008064110.1 40 ) and Methyloversatilis universalis (WP_012591629 41 ) Filter hits were concatenated into a non-redundant set, corresponding protein sequences were extracted from genome-wide protein files and annotated by gene identity and genome of origin. To guide alignment strategies, sequence variability was first assessed by calculating the average pairwise identity, average identities: soxA/mgdC : 39.64%, soxG/mgdD : 31.02%, soxB/mgdA : 46.18%, and soxD/mgdB : 34.62%. Due to the high sequence divergence observed across gene pairs, alignments were performed using MAFFT 80 with the E-INS-i algorithm and a BLOSUM62 matrix, to optimize for sequences with multiple conserved domains and long gaps. The alignments were subjected to model selection using ProtTest v3 81 . Best-fit substitution models were chosen based on AIC/BIC criteria and applied per gene pair as follows: soxB / mgdA and soxA / mgdC : LG+I+G, soxD / mgdB : WAG+I+G, and soxG / mgdD : WAG+G. Maximum likelihood (ML) phylogenetic trees were generated using RAxML with bootstrapping (1,000 replicates) on each gene pair alignment using the chosen substitution models, automated ML inference, bootstrap generation, and support value mapping. Tree diagrams were generated using iTol 82 . Identification of the homarine operon in NCBI genomes We conducted a broad survey to identify and characterize homarine operon orthologs. First, we downloaded and processed the NCBI ClusteredNR BLAST database 83 . We ran BLAST searches to identify homologs of the homarine catabolic genes ( homA, homB, homC, homD, and homE ). We retrieved cluster members of relevant hits using the ClusteredNR scripts 84 and downloaded their corresponding nucleotide coordinates from NCBI with Entrez Direct 85 to identify candidate genomic loci with at least three of the six conserved genes co-located within 10,000 bp. To ensure completeness of our initial survey, we downloaded the genomes in which we found candidate homarine catabolic operons to perform a second round of BLAST searches and selected loci that contained at least one ortholog of homA , homB , homC , and homD co-located within 10,000 bp. The resulting genomic loci data are reported in Table S12. We retrieved BioSample metadata from NCBI for each of the genomes where we found at least homA , homB , homC , and homD co-located within 10,000 bp. Metadata were parsed and mined to retrieve key biosample attributes such as isolation source, environmental context, geographic location, and additional identifiers. Next, we cleaned and organized the biosample metadata for comparison. This involved categorizing isolation sources into environmental and non-environmental types and further classifying environmental sources by specific habitats; latitude and longitude data were parsed directly from metadata or inferred from reported geographic location. Note that many of the biosample records accompanying the NCBI genomes lacked sufficient metadata to parse the environmental context or geographical location of the source. Resulting mapping data is reported in Table S12. Environmental transcriptomics perturbation experiment To identify microorganisms and genes responsive to homarine in Puget Sound, we conducted short ship-board incubations with seawater amended with homarine and collected total RNA for metatranscriptomics. Seawater was sampled with the RV Carson’s CTD from 4 m depth during RC-0078 (Station 2) in the Salish Sea near the Strait of Juan de Fuca on June 5, 2022. Seawater was transferred to acid-clean, blue-tinted, 5-Liter polycarbonate carboys for acclimation in a deck incubator with flow-through surface seawater. A set of three carboys were supplied with 0.1 mM carbon as homarine and three other unamended carboys served as control group. We incubated samples for 3.5 hours and proceeded to concentrate the microbial fraction by filtering each through a 47 mm-diameter, 0.22 µm pore-size polycarbonate membrane with a peristaltic pump. The filters were flash-frozen in liquid nitrogen at sea and were transferred to -80°C storage in the lab. Total RNA was extracted from filters using a ZymoBIOMICS RNA miniprep kit following the manufacturer protocol. RNA was quantified on a NanoDrop spectrophotometer. Samples with >50 ng/uL were sent to SeqCenter (Pittsburgh, PA, USA) for RNA-Seq. RNA samples were DNAse treated with Invitrogen DNAse (RNAse free). Library preparation was performed using Illumina’s Stranded Total RNA Prep Ligation with Ribo-Zero Plus kit and 10bp IDT for Illumina indices. Sequencing was done on a NovaSeq 6000 giving 2x51bp reads. Demultiplexing, quality control, and adapter trimming was performed with bcl-convert (v4.0.3). For metatranscriptomic analysis, we first parsed rRNA from non-rRNA reads with SortMeRNA. Non-rRNA reads (mRNA fragments) were assembled with PLASS 86 and mapped back to the 9,574,265 assembled proteins of length >100 amino acids with MMSeqs2 87 with the sensitivity parameter and maximum accepted targets set to 1. To quantify mapped reads, we retained only the best alignment for each unique read and reads with 100% identity to an assembled protein resulting in 42,155,143 reads and 1,367,930 proteins. We used DESeq2 58 and the tabulated read count data to detect differentially expressed genes between seawater samples supplied with homarine and unamended seawater. DESeq2 was implemented with a log 2 -fold change threshold of 1, the alternative hypothesis statement set to greater Abs, the significance level set to 0.05, and the default false discovery rate adjustment. We annotated 49 differentially expressed genes with the NCBI non-redundant database using BLASTP (excluding environmental samples 63 and the eggNOG 5.0.2 protein database using eggNOG-mapper v2.1.9 88,89 in ultra-sensitive mode. In-situ Homarine Concentration Particulate and dissolved metabolites were sampled and measured as described in previous work 6,7 . Briefly, water was collected from the shipboard flow-through underway sampling system and particulate metabolites were sampled by filtering the seawater using peristaltic pumps onto 142 mm diameter, 0.2 um pore size PTFE Omnipore filters, flash frozen in liquid nitrogen, and stored at -80 C until analysis. Dissolved metabolites were sampled by collecting the filtrate in 50 mL acid washed polypropylene Falcon tubes. Particulate metabolites were extracted using a modified Bligh and Dyer extraction as described 67 . Dissolved metabolites were extracted using cation-exchange solid phase extraction as reported 7 . Following extraction, metabolites were dried down under N 2 gas, reconstituted in water with isotope labeled internal standards, and analyzed using liquid chromatography mass spectrometry using the same acquisition approach as described for Cobetia sp. OBi1 samples . Homarine concentrations were quantified by comparison to a 2 H 3 -homarine internal standard. In-situ Metatranscriptomes Metatranscriptome samples were collected during the Gradients 3 cruise (KM1906) in the North Pacific in 2019. Samples were collected from the ship’s underway system by subsampling approximately 5-10 L of seawater into 10 L polycarbonate carboys, pre-filtered through a 100 μm mesh filter, and sequentially filtered through 3 μm and 0.2 μm polycarbonate filters. Filters were flash frozen in liquid nitrogen and stored at -80°C before processing. Metatranscriptomes were generated by extracting total RNA from the filters using the Direct-zol MiniPrep Plus kit (Zymo Research R2720). During extraction, a set of 14 synthetic spike-in mRNA standards was added as previously described 90,91 to enable quantification of absolute transcript abundances across samples. The extracted RNA was quantified with Qubit fluorometer 1.0 (ThermoFisher) and quality controlled using a Bioalanyzer 2100 (Agilent). Total community mRNA was generated by rRNA depletion using the ThermoFisher RiboMinus (Yeast+Bacteria; K15503, K15504) kit. Samples were sequenced on the Illumina NovaSeq6000 platform using S2 flow cell with 300 cycles. Library prep and sequencing was done at the Northwest Genomics Center (University of Washington, Seattle). Raw Illumina reads from rRNA depleted samples were quality controlled with Trimmomatic v0.36 (MAXINFO:135:0.5, LEADING:3, TRAILING:3, MINLEN:60, and AVGQUAL:20) 92 . Short reads were assembled into longer contigs using Trinity de novo assembler v2.3.2 (--normalize_reads --min_kmer_cov 2 --min_contig_length 300) 93 . All assemblies were performed on short reads from combined replicates for each sample. Assembled contigs were translated into amino acid space in six frames with transeq 94 vEMBOSS:6.6.0.059 using Standard Genetic Code, and the longest coding frame was retained for future analysis. The best frame translated assemblies were clustered at 99% amino acid identity with MMseqs2 87 . To quantify assembled contig abundances and obtain raw transcript counts, unassembled short reads were mapped to the assembled contigs using kallisto v0.46.1 95 . The raw transcript counts for each assembled contig were normalized to total transcripts per liter using normalization factors based on the recovery of 14 synthetic internal mRNA 13,91,96 . Standard counts were recovered using the Bowtie 2 aligner v.2.5.2 using the standard pipeline for paired-end reads 90 . Taxonomic annotation of the assembled contigs was done with DIAMOND alignment to a combined reference database containing MarFeRReT v1.1 97 and MARMICRODB v.1.0 98 reference databases. Functional annotation of the assembled contigs was done using HMMER 3.3 99 against the Kyoto Encyclopedia of Genes and Genomes (KEGG) KOfam Hidden Markov Models (HMMs) database 26 with best KOfam predictions with a threshold score > 30 retained for each contig, and against Pfam 35.0 100 HMMs where the ‘trusteed cutoff’ assigned by Pfam to each hmm profile was used as a minimum score threshold. To estimate abundances of selected HMM profiles in total community metatranscriptomes, assembled and annotated contigs of non-selected transcripts from prokaryotes and eukaryotes from the 3 μm and 0.2 μm fractions of each sample replicate were combined. Only contigs taxonomically annotated as “Bacteria” at the kingdom level were retained during the analysis. Hidden Markov Models profiles for genes in hom operon Homarine catabolic protein sequences for each gene were aligned using MAFFT with a BLOSUM62 substitution matrix. Alignments were manually curated to remove poorly conserved N- and C-terminal regions and internal gaps. Hidden Markov Models (HMMs) were then built from the curated alignments using HMMER v3.3.2 101 . HMM profile specificity was tested against a curated genome database of isolates capable of growing in homarine with co-located hom catabolic genes. Profiles for homA and homB showed strong specificity with minimal off-target hits. However, homC returned several false positives, and homD produced numerous non-specific matches, rendering these profiles unsuitable for further analysis. Each HMM profile was used to search for homologs in environmental metatranscriptomes collected on the Gradients 3 cruise using hmmsearch (HMMER v3.3.2) with the following parameters: --cpu 4 -T 30 --incT 30. The results were parsed and filtered, retaining only high-confidence hits based on alignment coverage, score, and contig completeness. Hits were manually curated against the reference HMM profile, and sequences that did not align properly or did not match the expected profile length were discarded. Finally, filtered gene hits were cross-referenced with sample metadata. We focused our analysis on the homB results, which showed the highest significance (e-values ranging from 3.8E-43 to 3.9E-240), while homA still significant, had comparatively weaker support (e-values ranging from 1.5E-06 to 9.5E-57). Declarations DATA AVAILABILITY All datasets generated or analyzed during this study are publicly available or will be made available upon publication. Genome assemblies for Cobetia sp. OBi1 and additional homarine-degrading isolates, and transcriptomic data from Cobetia sp. OBi1 grown under homarine, glucose + homarine, and glucose-only conditions are available in NCBI BioProject PRJNA862506. Metabolomic datasets for Cobetia sp. OBi1 and Ruegeria pomeroyi DSS-3, including intracellular and dissolved profiles and standard-matched metabolites, are deposited to Metabolomics Workbench repository under study ID (TBD). Metabolomics data from environmental seawater incubations with 2 H 3 or 13 C 7 15 N-labeled homarine are deposited to Metabolomics Workbench repository under study ID (TBD). Environmental data from the Gradients North Pacific transect are available through Simons CMAP (https://simonscmap.com/catalog/datasets/Gradients3_KM1906_Optics_LISST_ACS_ECO). Environmental metatranscriptome data from the Gradients 3 North Pacific transect are available through NCBI SRA under BioProject PRJNA1091352. Environmental metabolomics data from the Gradients 3 North Pacific transect are available through Metabolomics Workbench repository under study ID TBD. Custom analysis scripts used in this study are available on GitHub at (https://github.com/IngallsLabUW/Homarine_Catabolism_MS). ACKNOWLEDGEMENTS We thank the captains, crews, and science parties of the R/V Kilo Moana on cruise KM1906, of the R/V Thomas G. Thompson on cruises TN397 and TN412, and R/V Rachel Carson on cruises RC078 and RC104. We thank Lidimarie Trujillo Rodriguez and Susan Garcia for assistance with laboratory analyses; Sacha N. Coesel, Shiri Graff Van Creveld, Stephen Blaskowski, David Weidner, Ben Grodner, and Jeremy E. Schrier for valuable conversations and assistance with data analysis; Mary Ann Moran and Christopher R. Reisch for access to the Ruegeria pomeroyi DSS-3 arrayed BarSeq library. This work was supported by grants from the Simons Foundation (LS award ID: 385428, A.E.I.; SCOPE award ID 329108, A.E.I.; award ID 916384, B.P.D.), the National Science Foundation (OCE-2125886 to A.E.I. and K.R.H.; OCE-2124712 to O.A.S.), the University of Florida (Research Opportunity Seed Fund award ID 00131632 to B.P.D.), and the Center for Chemical Currencies of a Microbial Planet (C-CoMP; award ID 2019589 to F.X.F-G). References Moran, M. A. et al. The Ocean’s labile DOC supply chain. Limnol. Oceanogr. 67 , 1007–1021 (2022). Schalk, I. J. Bacterial siderophores: diversity, uptake pathways and applications. Nat. Rev. Microbiol. 23 , 24–40 (2025). Seymour, J. R., Simó, R., Ahmed, T. & Stocker, R. Chemoattraction to Dimethylsulfoniopropionate Throughout the Marine Microbial Food Web. Science 329 , 342–345 (2010). Harvey, E. L. et al. A Bacterial Quorum-Sensing Precursor Induces Mortality in the Marine Coccolithophore, Emiliania huxleyi . Front. Microbiol. 7 , (2016). Boysen, A. K. et al. Particulate Metabolites and Transcripts Reflect Diel Oscillations of Microbial Activity in the Surface Ocean. mSystems 6 , (2021). Heal, K. R. et al. Marine Community Metabolomes Carry Fingerprints of Phytoplankton Community Composition. mSystems 6 , (2021). Sacks, J. S., Heal, K. R., Boysen, A. K., Carlson, L. T. & Ingalls, A. E. Quantification of dissolved metabolites in environmental samples through cation‐exchange solid‐phase extraction paired with liquid chromatography–mass spectrometry. Limnol. Oceanogr. Methods 20 , 683–700 (2022). Widner, B., Kido Soule, M. C., Ferrer-González, F. X., Moran, M. A. & Kujawinski, E. B. Quantification of Amine- and Alcohol-Containing Metabolites in Saline Samples Using Pre-extraction Benzoyl Chloride Derivatization and Ultrahigh Performance Liquid Chromatography Tandem Mass Spectrometry (UHPLC MS/MS). Anal. Chem. 93 , 4809–4817 (2021). Holderman, N. R., Ferrer‐González, F. X., Glushka, J., Moran, M. A. & Edison, A. S. Dissolved organic metabolite extraction from high‐salt media. NMR Biomed. 36 , (2023). Hoppe-Seyler, F. A. Über das Homarin, eine bisher unbekannte tierische Base. Z. Für Physiol. Chem. 222 , 105–115 (1933). Kumler, W. et al. Metabolites reflect variability introduced by mesoscale eddies in the North Pacific Subtropical Gyre. Front. Mar. Sci. 11 , (2024). Sacks, J. S. et al. The cycling of glycine betaine and homarine in marine microbial communities: quantitative flux measurements and the role of competitive uptake inhibition. Limnol. Oceanogr. Lett. (In revision.). Durham, B. P. et al. Sulfonate-based networks between eukaryotic phytoplankton and heterotrophic bacteria in the surface ocean. Nat. Microbiol. 4 , 1706–1715 (2019). Dawson, H. M. et al. Microbial metabolomic responses to changes in temperature and salinity along the western Antarctic Peninsula. ISME J. 17 , 2035–2046 (2023). Zhang, Y. et al. Biodegradation of Picolinic Acid by Rhodococcus sp. PA18. Appl. Sci. 9 , 1006 (2019). Yoshida, T. & Nagasawa, T. Enzymatic functionalization of aromatic N -heterocycles: Hydroxylation and carboxylation. J. Biosci. Bioeng. 89 , 111–118 (2000). Qiu, J. et al. A Novel Degradation Mechanism for Pyridine Derivatives in Alcaligenes faecalis JQ135. Appl. Environ. Microbiol. 84 , e00910-18 (2018). Roney, S. H. et al. Common fear molecules induce defensive responses in marine prey across trophic levels. Oecologia 202 , 655–667 (2023). Maldonado, A. et al. Hard coral ( Porites lobata ) extracts and homarine on cytochrome P450 expression in Hawaiian butterflyfishes with different feeding strategies. Comp. Biochem. Physiol. Part C Toxicol. Pharmacol. 179 , 57–63 (2016). McClintock, J. B. et al. Homarine as a feeding deterrent in common shallow-water antarctic lamellarian gastropod Marseniopsis mollis : A rare example of chemical defense in a marine prosobranch. J. Chem. Ecol. 20 , 2539–2549 (1994). Berking, S. Homarine ( N -methylpicolinic acid) and trigonelline ( N -methylnicotinic acid) appear to be involved in pattern control in a marine hydroid. Development 99 , 211–220 (1987). Poulin, R. X. et al. Chemical encoding of risk perception and predator detection among estuarine invertebrates. Proc. Natl. Acad. Sci. 115 , 662–667 (2018). Rodriguez-R, L. M. et al. The Microbial Genomes Atlas (MiGA) webserver: taxonomic and gene diversity analysis of Archaea and Bacteria at the whole genome level. Nucleic Acids Res. 46 , W282–W288 (2018). Sarkar, M. K., Paul, K. & Blair, D. Chemotaxis signaling protein CheY binds to the rotor protein FliN to control the direction of flagellar rotation in Escherichia coli . Proc. Natl. Acad. Sci. 107 , 9370–9375 (2010). Spohn, G. & Scarlato, V. Motility, Chemotaxis, and Flagella. in Helicobacter pylori: Physiology and Genetics. 1–23 (ASM Press, 2001). Kanehisa, M. & Goto, S. KEGG: Kyoto Encyclopedia of Genes and Genomes. 28 , 27–30 (2000). Kanehisa, M. Toward understanding the origin and evolution of cellular organisms. Protein Sci. 28 , 1947–1951 (2019). Kanehisa, M., Furumichi, M., Sato, Y., Matsuura, Y. & Ishiguro-Watanabe, M. KEGG: biological systems database as a model of the real world. (2025). Blum, M. et al. InterPro: the protein sequence classification resource in 2025. Nucleic Acids Res. 53 , D444–D456 (2025). Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596 , 583–589 (2021). Cianfrocco, M. A., Wong-Barnum, M., Youn, C., Wagner, R. & Leschziner, A. COSMIC2: A Science Gateway for Cryo-Electron Microscopy Structure Determination. in Practice and Experience in Advanced Research Computing 2017: Sustainability, Success and Impact 1–5 (ACM, New Orleans LA USA, 2017). doi:10.1145/3093338.3093390. Roy, A., Yang, J. & Zhang, Y. COFACTOR: an accurate comparative algorithm for structure-based protein function annotation. Nucleic Acids Res. 40 , W471–W477 (2012). Zhang, C., Freddolino, L. & Zhang, Y. COFACTOR: improved protein function prediction by combining structure, sequence and protein–protein interaction information. Nucleic Acids Res. 45 , W291–W299 (2017). Pettersen, E. F. et al. UCSF ChimeraX : Structure visualization for researchers, educators, and developers. Protein Sci. 30 , 70–82 (2021). Perchat, N. et al. Elucidation of the trigonelline degradation pathway reveals previously undescribed enzymes and metabolites. Proc. Natl. Acad. Sci. 115 , (2018). Mejia, C. et al. An Arrayed Transposon Library of Ruegeria pomeroyi DSS-3. Preprint at https://doi.org/10.1101/2022.09.11.507510 (2022). Bürgmann, H. et al. Transcriptional response of Silicibacter pomeroyi DSS‐3 to dimethylsulfoniopropionate (DMSP). Environ. Microbiol. 9 , 2742–2755 (2007). Noell, S. E. & Giovannoni, S. J. SAR11 bacteria have a high affinity and multifunctional glycine betaine transporter. Environ. Microbiol. 21 , 2559–2575 (2019). James C Netherton III, J. C. & Gurin, S. Biosynthesis and physiological role of homarine in marine shrimp. J. Biol. Chem. 257 , 11971–11975 (1982). Chen, Y. et al. γ-Glutamylmethylamide Is an Essential Intermediate in the Metabolism of Methylamine by Methylocella silvestris . Appl. Environ. Microbiol. 76 , 4530–4537 (2010). Latypova, E. et al. Genetics of the glutamate‐mediated methylamine utilization pathway in the facultative methylotrophic beta‐proteobacterium Methyloversatilis universalis FAM5. Mol. Microbiol. 75 , 426–439 (2010). Hersh, L. B., Peterson, J. A. & Thompson, A. A. An N-methyl glutamate dehydrogenase from Pseudomonas M.A. Arch. Biochem. Biophys. 145 , 115–120 (1971). Giovannoni, S. J. SAR11 Bacteria: The Most Abundant Plankton in the Oceans. Annu. Rev. Mar. Sci. 9 , 231–255 (2017). Rappé, M. S., Kemp, P. F. & Giovannoni, S. J. Phylogenetic diversity of marine coastal picoplankton 16S rRNA genes cloned from the continental shelf off Cape Hatteras, North Carolina. Limnol. Oceanogr. 42 , 811–826 (1997). Swan, B. K. et al. Prevalent genome streamlining and latitudinal divergence of planktonic bacteria in the surface ocean. Proc. Natl. Acad. Sci. 110 , 11463–11468 (2013). Stingl, U., Desiderio, R. A., Cho, J.-C., Vergin, K. L. & Giovannoni, S. J. The SAR92 Clade: an Abundant Coastal Clade of Culturable Marine Bacteria Possessing Proteorhodopsin. Appl. Environ. Microbiol. 73 , 2290–2296 (2007). Luo, H. & Moran, M. A. Evolutionary Ecology of the Marine Roseobacter Clade. Microbiol. Mol. Biol. Rev. 78 , 573–587 (2014). Simon, M. et al. Phylogenomics of Rhodobacteraceae reveals evolutionary adaptation to marine and non-marine habitats. ISME J. 11 , 1483–1499 (2017). Giovannoni, S. J. et al. Genome Streamlining in a Cosmopolitan Oceanic Bacterium. Science 309 , 1242–1245 (2005). He, X.-Y. et al. SAR92 clade bacteria are potentially important DMSP degraders and sources of climate-active gases in marine environments. mBio 14 , e01467-23 (2023). Moran, M. A. et al. Ecological Genomics of Marine Roseobacters. Appl. Environ. Microbiol. 73 , 4559–4569 (2007). Holtgrewe, M., Messerschmidt, C., Nieminen, M. & Beule, D. DigestiFlow: from BCL to FASTQ with ease. Bioinformatics 36 , 1983–1985 (2020). Bonenfant, Q., Noé, L. & Touzet, H. Porechop_ABI: discovering unknown adapters in Oxford Nanopore Technology sequencing reads for downstream trimming. Bioinforma. Adv. 3 , vbac085 (2023). Wick, R. R., Judd, L. M., Gorrie, C. L. & Holt, K. E. Unicycler: Resolving bacterial genome assemblies from short and long sequencing reads. PLOS Comput. Biol. 13 , e1005595 (2017). Gurevich, A., Saveliev, V., Vyahhi, N. & Tesler, G. QUAST: quality assessment tool for genome assemblies. Bioinformatics 29 , 1072–1075 (2013). Seemann, T. Prokka: rapid prokaryotic genome annotation. Bioinformatics 30 , 2068–2069 (2014). Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30 , 923–930 (2014). Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15 , 550 (2014). Dehio, C. & Meyer, M. Maintenance of broad-host-range incompatibility group P and group Q plasmids and transposition of Tn5 in Bartonella henselae following conjugal plasmid transfer from Escherichia coli. J. Bacteriol. 179 , 538–540 (1997). Thanbichler, M., Iniesta, A. A. & Shapiro, L. A comprehensive set of plasmids for vanillate- and xylose-inducible gene expression in Caulobacter crescentus. Nucleic Acids Res. 35 , e137–e137 (2007). Ried & Collmer. An nptI-sacR cartridge for constructing directed, unmarked mutations in Gram negative bacteria by marker exchange eviction mutagenesis. Gene 57 , 239–246 (1987). Pearson, A. N. et al. The pGinger Family of Expression Plasmids. Microbiol. Spectr. 11 , e00373-23 (2023). Camacho, C. et al. BLAST+: architecture and applications. BMC Bioinformatics 10 , 421 (2009). Schroer, W. F. et al. Functional annotation and importance of marine bacterial transporters of plankton exometabolites. ISME Commun. 3 , 37 (2023). Baumann, P. & Baumann, L. The Marine Gram-Negative Eubacteria: Genera Photobacteri in Beneckea, Alteromonas, Pseudomonas, and Alcaligenes . The Prokaryotes 2 , 1302–1331 (1981). Guillard, R. R. L. & Hargraves, P. E. Stichochrysis immobilis is a diatom, not a chrysophyte. Phycologia 32 , 234–236 (1992). Boysen, A. K., Heal, K. R., Carlson, L. T. & Ingalls, A. E. Best-Matched Internal Standard Normalization in Liquid Chromatography–Mass Spectrometry Metabolomics Applied to Environmental Samples. Anal. Chem. 90 , 1363–1369 (2018). Boysen, A. K. et al. Glycine betaine uptake and metabolism in marine microbial communities. Environ. Microbiol. 24 , 2380–2403 (2022). Adams, K. J. et al. Skyline for Small Molecules: A Unifying Software Package for Quantitative Metabolomics. J. Proteome Res. 19 , 1447–1458 (2020). Tsugawa, H. et al. A lipidome atlas in MS-DIAL 4. Nat. Biotechnol. 38 , 1159–1163 (2020). Grubbs, F. E. Procedures for Detecting Outlying Observations in Samples. Technometrics 11 , 1–21 (1969). Benjamini, Y. & Hochberg, Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J. R. Stat. Soc. Ser. B Methodol. 57 , 289–300 (1995). Canelas, A. B. et al. Quantitative evaluation of intracellular metabolite extraction techniques for yeast metabolomics. Anal. Chem. 81 , 7379–7389 (2009). Durham, B. P. et al. Chemotaxonomic patterns in intracellular metabolites of marine microbial plankton. Front. Mar. Sci. 9 , 864796 (2022). Boiteau, R. M. & Repeta, D. J. An extended siderophore suite from Synechococcus sp. PCC 7002 revealed by LC-ICPMS-ESIMS. Metallomics 7 , 877–884 (2015). Corilo, Y. E., Kew, W. R., McCue, L. A., Heal, K. & Carr, J. C. EMSL-Computing/CoreMS: CoreMS 3.0.0. Zenodo https://doi.org/10.5281/zenodo.14009575 (2024). Krueger, F. Trim Galore!: A wrapper around Cutadapt and FastQC to consistently apply adapter and quality trimming to FastQ files, with extra functionality for RRBS data. Babraham Inst. (2015). Hassan-Abdallah, A. et al. Cloning, expression and crystallization of heterotetrameric sarcosine oxidase from Pseudomonas maltophilia . Protein Expr. Purif. 43 , 33–43 (2005). Suzuki, M. Purification and Some Properties of Sarcosine Oxidase from Corynebacterium sp. U-96. J. Biochem. (Tokyo) 89 , 599–607 (1981). Katoh, K. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 30 , 3059–3066 (2002). Abascal, F., Zardoya, R. & Posada, D. ProtTest: selection of best-fit models of protein evolution. Bioinformatics 21 , 2104–2105 (2005). Letunic, I. & Bork, P. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res. 49 , W293–W296 (2021). Sayers, E. W. et al. Database resources of the National Center for Biotechnology Information in 2023. Nucleic Acids Res. 51 , D29–D38 (2023). Celebi, F. M., Eisen, J. A., Hochstrasser, M. L. & Reiter, T. Clustering the NCBI nr database to reduce database size and enable faster BLAST searches. (2023). Kans, J. Entrez direct: E-utilities on the UNIX command line. (2024). Steinegger, M., Mirdita, M. & Söding, J. Protein-level assembly increases protein sequence recovery from metagenomic samples manyfold. Nat. Methods 16 , 603–606 (2019). Steinegger, M. & Söding, J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat. Biotechnol. 35 , 1026–1028 (2017). Cantalapiedra, C. P., Hernández-Plaza, A., Letunic, I., Bork, P. & Huerta-Cepas, J. eggNOG-mapper v2: Functional Annotation, Orthology Assignments, and Domain Prediction at the Metagenomic Scale. Mol. Biol. Evol. 38 , 5825–5829 (2021). Huerta-Cepas, J. et al. eggNOG 5.0: a hierarchical, functionally and phylogenetically annotated orthology resource based on 5090 organisms and 2502 viruses. Nucleic Acids Res. 47 , D309–D314 (2019). Groussman, R. D., Coesel, S. N., Durham, B. P., Schatz, M. J. & Armbrust, E. V. The North Pacific Eukaryotic Gene Catalog of metatranscriptome assemblies and annotations. Sci. Data 11 , 1161 (2024). Satinsky, B. M., Gifford, S. M., Crump, B. C. & Moran, M. A. Use of Internal Standards for Quantitative Metatranscriptome and Metagenome Analysis. in Methods in Enzymology vol. 531 237–250 (Elsevier, 2013). Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30 , 2114–2120 (2014). Grabherr, M. G. et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat. Biotechnol. 29 , 644–652 (2011). Rice, P., Longden, I. & Bleasby, A. EMBOSS: the European molecular biology open software suite. Trends Genet. 16 , 276–277 (2000). Bray, N. L., Pimentel, H., Melsted, P. & Pachter, L. Near-optimal probabilistic RNA-seq quantification. Nat. Biotechnol. 34 , 525–527 (2016). Park, J. et al. Siderophore production and utilization by marine bacteria in the North Pacific Ocean. Limnol. Oceanogr. 68 , 1636–1653 (2023). Groussman, R. D., Blaskowski, S., Coesel, S. N. & Armbrust, E. V. MarFERReT, an open-source, version-controlled reference library of marine microbial eukaryote functional genes. Sci. Data 10 , 926 (2023). Hogle, S. L. MARMICRODB database for taxonomic classification of (marine) metagenomes (1.0.0). https://doi.org/10.5281/zenodo.3520509 (2019). Eddy, S. R. Accelerated Profile HMM Searches. PLoS Comput. Biol. 7 , e1002195 (2011). Mistry, J. et al. Pfam: The protein families database in 2021. Nucleic Acids Res. 49 , D412–D419 (2021). Potter, S. C. et al. HMMER web server: 2018 update. Nucleic Acids Res. 46 , W200–W204 (2018). Schymanski, E. L. et al. Identifying Small Molecules via High Resolution Mass Spectrometry: Communicating Confidence. Environ. Sci. Technol. 48 , 2097–2098 (2014). Additional Declarations There is NO Competing Interest. Supplementary Files Supptables20250617.xlsx Supplemental Table S1: Cobetia sp. OBi1 genome sequencing and assembly summary Supplemental Table S2: Differential gene expression results from transcriptomics experiment of Cobetia sp. OBi1 grown in glucose + homarine vs glucose and homarine vs glucose. Supplemental Table S3: Upregulated genes of OBi1 grown in homarine and ortholog genes between Cobetia sp. OBi1 and R. pomeroyi DSS-3. Supplemental Table S4: Predicted functional characterization of Hom proteins Supplemental Table S5: R. pomeroyi DSS-3 mutant growth phenotypes Supplemental Table S6: Results from metabolomics experiments with Cobetia sp. OBi1, particulate samples only. Supplemental Table S7: Results from metabolomics experiments with Cobetia sp. OBi1, dissolved samples only. Supplemental Table S8: Description of samples collected for metabolomics analyses. Supplemental Table S9: Results from targeted search of metabolomics data, see sample descriptions in Table S8. Supplemental Table S10 - Results of untargeted search for 2 H 3 and 2 H 2 isotopologues in the TN379 experiments. Mass features were detected at both the monoisotopic and isotopologue masses in only +homarine treatments (not controls) and were visually inspected to ensure quality peaks. All detected resulting mass features were identified and are reported here. Supplemental Table S11: Summary of isolates tested in this study, including the ability to grow on homarine; the NCBI-derived taxonomy and associated accession numbers; as well as the locations of the genes identified as hom, sox, or mgd genes. Supplemental Table S12: Full search results for hom operon in NCBI genomes. Supplemental Table S13: Upregulated bacterial genes in the homarine pertubation experiment in the environment. Supplemental Table S14: In-situ expression of hom A and hom B catabolic genes. Supplemental Table S15: Homarine concentrations concurrent with in-situ expression of hom A and hom B. Supplemental Table S16: Primers used for Cobetia sp. OBi1 mutants Supplemental Table S17: Isotopically-labeled internal standards used in the metabolomics analyses. Supplemental Table S18: Parameters used in MSD SupplementaryFigures.docx Cite Share Download PDF Status: Published Journal Publication published 30 Mar, 2026 Read the published version in Nature Microbiology → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7359689","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":502743303,"identity":"37de019c-d5cf-407b-86c3-637584f071a6","order_by":0,"name":"Frank Ferrer-Gonzalez","email":"","orcid":"https://orcid.org/0000-0003-0570-5168","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Frank","middleName":"","lastName":"Ferrer-Gonzalez","suffix":""},{"id":502743304,"identity":"b461e2ce-9d60-4756-8c9a-0cfded816d56","order_by":1,"name":"Katherine Heal","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Katherine","middleName":"","lastName":"Heal","suffix":""},{"id":502743305,"identity":"eed32148-22d9-41db-bab2-409745602b3a","order_by":2,"name":"Joshua Sacks","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Joshua","middleName":"","lastName":"Sacks","suffix":""},{"id":502743306,"identity":"9adcde5e-f7c2-4035-aee9-084541e58b78","order_by":3,"name":"Yarinet Romero-Maysonet","email":"","orcid":"","institution":"Ana G. Méndez University","correspondingAuthor":false,"prefix":"","firstName":"Yarinet","middleName":"","lastName":"Romero-Maysonet","suffix":""},{"id":502743307,"identity":"69469f22-842e-4d98-9131-21d3c35c9971","order_by":4,"name":"Anna Finch","email":"","orcid":"https://orcid.org/0009-0007-2160-7177","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"","lastName":"Finch","suffix":""},{"id":502743308,"identity":"1f6b7c91-dadd-4ff4-b7bc-b64401f0a304","order_by":5,"name":"Laura Carlson","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Laura","middleName":"","lastName":"Carlson","suffix":""},{"id":502743309,"identity":"48e53cdb-d615-4459-bf0e-fc393d024b35","order_by":6,"name":"Lisa Coe","email":"","orcid":"https://orcid.org/0000-0002-2244-9058","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Lisa","middleName":"","lastName":"Coe","suffix":""},{"id":502743310,"identity":"84dc182f-a251-4194-8f7d-8857b4ed3383","order_by":7,"name":"Zinka Bartolek","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Zinka","middleName":"","lastName":"Bartolek","suffix":""},{"id":502743311,"identity":"7c84bf24-8f12-46a5-a9dc-053abe677af1","order_by":8,"name":"Claudia Luthy","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Claudia","middleName":"","lastName":"Luthy","suffix":""},{"id":502743312,"identity":"f4db067a-5c42-4e95-af35-d6ae97499496","order_by":9,"name":"Moira Gaffney","email":"","orcid":"","institution":"University of Puget Sound","correspondingAuthor":false,"prefix":"","firstName":"Moira","middleName":"","lastName":"Gaffney","suffix":""},{"id":502743313,"identity":"1f18174b-bd02-44ba-be0e-8d4cbcaff89e","order_by":10,"name":"Sabine Angier","email":"","orcid":"","institution":"University of Rhode Island Graduate School of Oceanography","correspondingAuthor":false,"prefix":"","firstName":"Sabine","middleName":"","lastName":"Angier","suffix":""},{"id":502743314,"identity":"f260e53f-8fc2-483e-8b84-d8b29efeabf2","order_by":11,"name":"Samantha Flynn","email":"","orcid":"","institution":"Harvard Medical School","correspondingAuthor":false,"prefix":"","firstName":"Samantha","middleName":"","lastName":"Flynn","suffix":""},{"id":502743315,"identity":"73de87af-8d50-45ec-a086-768be5adece3","order_by":12,"name":"Chiara Bachmann Gómez","email":"","orcid":"https://orcid.org/0009-0004-3262-0527","institution":"Whitman College","correspondingAuthor":false,"prefix":"","firstName":"Chiara","middleName":"Bachmann","lastName":"Gómez","suffix":""},{"id":502743316,"identity":"7b5b6638-ddab-48d7-af9b-2ec46152bc35","order_by":13,"name":"Jensen Dunn","email":"","orcid":"","institution":"Whitman College","correspondingAuthor":false,"prefix":"","firstName":"Jensen","middleName":"","lastName":"Dunn","suffix":""},{"id":502743317,"identity":"1cd7c6de-d9c1-47bc-9889-d441f52cb105","order_by":14,"name":"Kenzie Bay","email":"","orcid":"","institution":"Whitman College","correspondingAuthor":false,"prefix":"","firstName":"Kenzie","middleName":"","lastName":"Bay","suffix":""},{"id":502743318,"identity":"510c9980-edbf-4871-b817-f55d1f64476f","order_by":15,"name":"Lia Yamamoto","email":"","orcid":"","institution":"Whitman College","correspondingAuthor":false,"prefix":"","firstName":"Lia","middleName":"","lastName":"Yamamoto","suffix":""},{"id":502743319,"identity":"5bd4f023-d832-471a-bb60-5bd432559cff","order_by":16,"name":"Matthew Tien","email":"","orcid":"","institution":"Whitman College","correspondingAuthor":false,"prefix":"","firstName":"Matthew","middleName":"","lastName":"Tien","suffix":""},{"id":502743320,"identity":"a7dfc516-858b-46c8-83cc-54fb688745a8","order_by":17,"name":"E. Virginia Armbrust","email":"","orcid":"https://orcid.org/0000-0001-7865-5101","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"E.","middleName":"Virginia","lastName":"Armbrust","suffix":""},{"id":502743321,"identity":"08f6e2d4-397d-49bf-a7cc-676039a19479","order_by":18,"name":"Bryndan Durham","email":"","orcid":"https://orcid.org/0000-0002-2253-157X","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Bryndan","middleName":"","lastName":"Durham","suffix":""},{"id":502743322,"identity":"b59d78d2-24a2-4065-a19b-355a694ef3cb","order_by":19,"name":"Oscar Sosa","email":"","orcid":"https://orcid.org/0000-0003-4235-9962","institution":"University of Puget Sound","correspondingAuthor":false,"prefix":"","firstName":"Oscar","middleName":"","lastName":"Sosa","suffix":""},{"id":502743302,"identity":"45b4ddc5-1f4e-4d45-9970-e2f42910808f","order_by":20,"name":"Anitra Ingalls","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvklEQVRIiWNgGAWjYDACZhBRYMHAD+axEa3FQIJBsoFoLQxQLQYHiNXC3857+MUPAwl74xs5Bgwfyg4T1iJxmC/NssdAInEbUAvjjHNEaGE4zGNmwGMgkWB2I3cDM28bEVrkgVoM/4AcNgOo5S8xWgwO8xg/BtrCuEECqIWRGC2GQFuYZYB+mXHm/YeDPefSCWuRO3/G+OObCht7/va0xAc/yqwJawECNgkY6wBR6oGA+QOxKkfBKBgFo2CEAgC8fjXflivh6gAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0003-1953-7329","institution":"University of Washington","correspondingAuthor":true,"prefix":"","firstName":"Anitra","middleName":"","lastName":"Ingalls","suffix":""}],"badges":[],"createdAt":"2025-08-13 01:10:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7359689/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7359689/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41564-026-02313-7","type":"published","date":"2026-03-30T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":89530362,"identity":"391943dc-7bd1-496e-8c88-42fb617786d5","added_by":"auto","created_at":"2025-08-21 03:46:35","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":377591,"visible":true,"origin":"","legend":"\u003cp\u003eA conserved gene operon for homarine catabolism in marine bacteria. (A) \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 growth on glucose (12 mM C), glucose with homarine (1 mM C) or homarine (12 mM C) for comparative transcriptomics. Gray vertical lines indicate sampling times. (B) Differential gene expression fold change (FC) of \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 in homarine vs. glucose. (C) Differential gene expression of \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 in glucose with homarine vs. glucose. Black vertical lines indicate significantly upregulated genes (D) \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 genes (arrows) upregulated in homarine relative to glucose and homologous gene clusters in \u003cem\u003eR. pomeroyi\u003c/em\u003e DSS-3. Arrow fill color corresponds to the type of function performed by each gene. Arrow outline color indicates the results of growth tests with in-frame deletion mutants of OBi1 or transposon insertion Kan\u003csup\u003eR\u003c/sup\u003e mutants of \u003cem\u003eR. pomeroyi\u003c/em\u003e DSS-3.\u0026nbsp; Glycine betaine transporter (GBT). Gray stars * indicate genes upregulated in glucose + homarine compared to glucose alone; black stars * indicate genes upregulated in homarine compared to glucose. Full descriptions of genes found in Table S3.\u0026nbsp; (E) Proposed homarine catabolic pathway inferred from omics data.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7359689/v1/2e0104fea68069c8b8028fe8.png"},{"id":89530360,"identity":"556eceb1-0007-4c47-b2ff-1557e1e2748a","added_by":"auto","created_at":"2025-08-21 03:46:34","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":476415,"visible":true,"origin":"","legend":"\u003cp\u003eResults of metabolomics experiments in lab and field. (A) Enrichment factors of\u003cem\u003e \u003c/em\u003eselect metabolites in \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 and \u003cem\u003eR. pomeroyi \u003c/em\u003eDSS-3 during growth on homarine and glucose + homarine growth as compared to glucose-only growth. “Core metabs” show the maximum (max), median (med), and minimum (min) calculated enrichment factor for the subset of metabolites used as normalizing factor for enrichment calculations for comparison. Molecular formulas are reported for identified compounds. Compounds without identification are described with the putative molecular formula, ionization mode (- or +) and retention time; assuming [M+H]+ or [M-H]- forms. Asterisks designate enrichments found to be significant (fdr-adjusted \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05). (B) Extracted ion chromatograms of \u003cem\u003eN\u003c/em\u003e-methylglutamic acid in \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 and \u003cem\u003eR. pomeroyi\u003c/em\u003e DSS-3 during growth on homarine and glucose, with structure (inset).\u0026nbsp; (C) Same as (B), but for \u003cem\u003eN\u003c/em\u003e-methylglutamine. (D) Map of stable isotope experiments and structure of homarine (inset), with atoms highlighted that were subject to stable isotope probing. (E) Enrichment factors of\u003cem\u003e \u003c/em\u003eselect metabolites in stable isotope experiment for TN397 where \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-labeled homarine was added (Hs on the *C in (C) are labeled). The selected metabolites were all detected with \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e stable isotope.\u0026nbsp; Core metabolites, scale, and labels are the same as in (A).\u0026nbsp; (F) Enrichment factors of\u003cem\u003e \u003c/em\u003eselect metabolites in stable isotope experiment for RC104 and TN412 where \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e7\u003c/sub\u003e,\u003csup\u003e15\u003c/sup\u003eN-homarine was added (blue and red highlighted in in (C) are labeled).\u0026nbsp; The selected metabolites were all detected with all Cs as \u003csup\u003e13\u003c/sup\u003eC and 1 N as \u003csup\u003e15\u003c/sup\u003eN isotope.\u0026nbsp; Core metabolites, scale, and labels are the same as in (A). Detailed data found in Tables S6, S8 and S9.\u0026nbsp;\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7359689/v1/1b8ab273efe71efa610608d4.png"},{"id":89530365,"identity":"e77def93-8e57-4d05-a388-2b2d4d534eb1","added_by":"auto","created_at":"2025-08-21 03:46:35","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":334021,"visible":true,"origin":"","legend":"\u003cp\u003eConservation of the \u003cem\u003ehom \u003c/em\u003eoperon in isolates and publicly available genomes. (A) Isolates tested in this study, with + showing growth on homarine as primary source of carbon and nitrogen from three independent replicates, based on optical density (OD600); checkmarks note the presence of \u003cem\u003ehom\u003c/em\u003e genes, the presence of \u003cem\u003emgd\u003c/em\u003e genes, the presence of \u003cem\u003esox\u003c/em\u003e genes (more details in Figures SX6-SX9). The class and family of each isolate are color coded. (B) Sunburst plot highlighting taxonomy of genomes with identified \u003cem\u003ehom\u003c/em\u003e operon in NCBI genomes. (C) Geographic distribution and environment type of NCBI genomes from environmental samples (circles) with stars showing location of additional marine isolates from panel A, see Tables S11 and S12 for details.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7359689/v1/d1221c2d354a262ba6338be0.png"},{"id":89531019,"identity":"0443110d-8c39-4aa7-ab57-215a71f36030","added_by":"auto","created_at":"2025-08-21 03:54:34","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":137745,"visible":true,"origin":"","legend":"\u003cp\u003eIn-situ expression of \u003cem\u003ehomB \u003c/em\u003ealigns with homarine abundance in the environment. Total homarine (particulate and dissolved combined) concentrations (blue circles), transcript abundances of \u003cem\u003ehomB \u003c/em\u003ein heterotrophic bacteria (colored bars), and chlorophyll \u003cem\u003ea\u003c/em\u003e (μg/L) concentrations (green line) across the latitude 25.87° to 40.88°. Detailed data found in Tables S14 and S15.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7359689/v1/78a072862596bb757b2d62ca.png"},{"id":105789259,"identity":"97e4c6c7-3177-4ac6-9ae8-820fbde6571a","added_by":"auto","created_at":"2026-03-31 07:14:37","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2639442,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7359689/v1/3f167ef7-de18-487c-be86-c47e9e851bef.pdf"},{"id":89530364,"identity":"12be3212-ba2d-4e36-b806-74cf47bec081","added_by":"auto","created_at":"2025-08-21 03:46:35","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3395296,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplemental Table S1: \u003c/strong\u003e\u003cem\u003eCobetia sp.\u003c/em\u003e OBi1 genome sequencing and assembly summary\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S2: \u003c/strong\u003eDifferential gene expression results from transcriptomics experiment of \u003cem\u003eCobetia sp.\u003c/em\u003e OBi1 grown in glucose + homarine vs glucose and homarine vs glucose.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S3:\u003c/strong\u003e Upregulated genes of OBi1 grown in homarine and ortholog genes between \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 and \u003cem\u003eR. pomeroyi \u003c/em\u003eDSS-3.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S4\u003c/strong\u003e:\u003cem\u003e \u003c/em\u003ePredicted functional characterization of Hom proteins\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S5\u003c/strong\u003e:\u003cem\u003e R. pomeroyi \u003c/em\u003eDSS-3 mutant growth phenotypes\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S6\u003c/strong\u003e: Results from metabolomics experiments with \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1, particulate samples only.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S7\u003c/strong\u003e: Results from metabolomics experiments with \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1, dissolved samples only.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S8\u003c/strong\u003e: Description of samples collected for metabolomics analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S9\u003c/strong\u003e: Results from targeted search of metabolomics data, see sample descriptions in Table S8.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S10 - \u003c/strong\u003eResults of untargeted search for \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e and \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e2 \u003c/sub\u003eisotopologues in the TN379 experiments.\u0026nbsp; Mass features were detected at both the monoisotopic and isotopologue masses in only +homarine treatments (not controls) and were visually inspected to ensure quality peaks.\u0026nbsp; All detected resulting mass features were identified and are reported here.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S11\u003c/strong\u003e: Summary of isolates tested in this study, including the ability to grow on homarine; the NCBI-derived taxonomy and associated accession numbers; as well as the locations of the genes identified as \u003cem\u003ehom, sox, or mgd \u003c/em\u003egenes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S12\u003c/strong\u003e: Full search results for \u003cem\u003ehom\u003c/em\u003e operon in NCBI genomes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S13: \u003c/strong\u003eUpregulated bacterial genes in the homarine pertubation experiment in the environment.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S14\u003c/strong\u003e: \u003cem\u003eIn-situ \u003c/em\u003eexpression of \u003cem\u003ehom\u003c/em\u003eA and \u003cem\u003ehom\u003c/em\u003eB catabolic genes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S15\u003c/strong\u003e: Homarine concentrations concurrent with \u003cem\u003ein-situ \u003c/em\u003eexpression of \u003cem\u003ehom\u003c/em\u003eA and \u003cem\u003ehom\u003c/em\u003eB.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S16\u003c/strong\u003e: Primers used for \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 mutants\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S17\u003c/strong\u003e: Isotopically-labeled internal standards used in the metabolomics analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table S18\u003c/strong\u003e: Parameters used in MSD\u003c/p\u003e","description":"","filename":"Supptables20250617.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7359689/v1/ba78013a87c709021f125c92.xlsx"},{"id":89530363,"identity":"f1b11d4c-37a1-42bc-b9f1-08c0867ed963","added_by":"auto","created_at":"2025-08-21 03:46:35","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":1722858,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-7359689/v1/2b5802b6260b7d38948c3139.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Conserved pathway for homarine catabolism in environmental bacteria","fulltext":[{"header":"Main","content":"\u003cp\u003eMicrobial metabolites are central to ocean biogeochemistry, facilitating rapid cycling of labile organic carbon through the microbial loop and driving the annual turnover of tens of petagrams of carbon\u003csup\u003e1\u003c/sup\u003e. Many metabolites also have compound-specific functions like enabling micronutrient acquisition\u003csup\u003e2\u003c/sup\u003e, serving as chemoattractants\u003csup\u003e3\u003c/sup\u003e, or acting as toxins\u003csup\u003e4\u003c/sup\u003e. Advances in environmental metabolomics have uncovered abundant metabolites whose biogeochemical and ecological significance\u003csup\u003e1\u003c/sup\u003e remain unresolved, partly because their metabolic pathways are unknown\u003csup\u003e5\u0026ndash;9\u003c/sup\u003e. These unknown pathways hinder genomics, transcriptomics, and proteomics approaches for studying microbial function. Identifying the genes and transformations linked to metabolites enhances our understanding of microbial metabolism and better integrates metabolites into a collective biological, ecological, and biogeochemical framework.\u003c/p\u003e\n\u003cp\u003eAmong marine metabolites, homarine (\u003cem\u003eN\u003c/em\u003e-methylpicolinic acid), stands out for both its widespread distribution and multifaceted bioactivity. Homarine, a polar alkaloid with a methylated pyridine ring, was first discovered in lobsters nearly a century ago\u003csup\u003e10\u003c/sup\u003e and is now recognized as ubiquitous in surface seawater\u003csup\u003e6,7,11\u003c/sup\u003e with turnover times ranging from days to weeks\u003csup\u003e12\u003c/sup\u003e. While heterotrophic bacteria are assumed to consume homarine, the metabolic pathway, molecular intermediates, and ecological distribution of its degradation remain uncharacterized. Homarine is produced by abundant phytoplankton groups such as diatoms and \u003cem\u003eSynechococcus \u003c/em\u003eat millimolar intracellular concentrations while other groups like dinoflagellates and prasinophytes do not produce it\u003csup\u003e6,13,14\u003c/sup\u003e. Homarine\u0026rsquo;s pyridine ring suggests specialized enzymatic machinery for its degradation, as seen with other pyridine-containing compounds\u003csup\u003e15\u0026ndash;17\u003c/sup\u003e. This, along with its production patterns, may contribute to niche partitioning among phytoplankton and heterotrophs. Homarine also exhibits bioactivity in marine animals, influencing oyster shell morphology\u003csup\u003e18\u003c/sup\u003e, deterring feeding in butterfly fish and sea stars\u003csup\u003e19,20\u003c/sup\u003e, altering larval development in hydroids\u003csup\u003e21\u003c/sup\u003e, and serving as a crab predation deterrent\u003csup\u003e22\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThis study investigates the heterotrophic bacterial degradation of homarine across biochemical, genetic, and ecological dimensions through an integrative approach. Comparative transcriptomics identifies a conserved operon for homarine degradation, while mass spectrometry reveals its biochemical transformation into metabolic products. Comparative genomics maps the global distribution of homologous operons, and metatranscriptomic analyses in natural and amended communities link gene expression to homarine availability, emphasizing its ecological relevance.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003e\u003cem\u003eIdentification of homarine operon\u003c/em\u003e\u003c/h2\u003e\n\u003cp\u003eMotivated by homarine\u0026rsquo;s abundance in marine environments, we sought to understand its bacterial degradation. We isolated a bacterium from seawater at Owen Beach,Tacoma, Washington USA, and demonstrated its ability to use homarine as a primary carbon and nitrogen source (Figure 1A). Genome sequencing identified this Gammaproteobacterium as \u003cem\u003eCobetia sp.\u003c/em\u003e within the order \u003cem\u003eOceanospirillales\u003c/em\u003e, family \u003cem\u003eHalomonadaceae\u003c/em\u003e, with high similarity to \u003cem\u003eCobetia marina\u003c/em\u003e (98.51% ANI) and \u003cem\u003eCobetia pacifica\u003c/em\u003e (98.3% ANI, Table S1)\u003csup\u003e23\u003c/sup\u003e. We named this isolate\u003cem\u003e Cobetia sp\u003c/em\u003e. OBi1.\u003c/p\u003e\n\u003cp\u003eTo uncover genes involved in homarine degradation, we grew \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 in minimal media with homarine as the sole carbon source and conducted comparative transcriptomics and metabolomics experiments. We compared three conditions: \u003cem\u003eglucose \u003c/em\u003e(12 mM C, 0.8 mM NH\u003csub\u003e4\u003c/sub\u003e, control), \u003cem\u003ehomarine \u003c/em\u003e(12 mM C, no additional NH\u003csub\u003e4\u003c/sub\u003e), and \u003cem\u003eglucose \u003c/em\u003e+ \u003cem\u003ehomarine\u003c/em\u003e (12 mM C, 1 mM C, respectively with 0.8 mM NH\u003csub\u003e4\u003c/sub\u003e). In both conditions with homarine, 22 genes were enriched compared to the glucose only condition (s-value \u0026lt; 0.05, Figure 1B, Table S2), including three gene clusters. The first cluster encoded a transcriptional regulator, a predicted glycine betaine transporter (GBT), and five putative catabolic genes with unclear function (ACFLL8_11490 to ACFLL8_11520, Figure 1C, Table S3). The second contains a putative sarcosine oxidase complex and the methylenetetrahydrofolate dehydrogenase \u003cem\u003efolD \u003c/em\u003e(ACFLL8_07290 to ACFLL8_07310; Figure 1C, Tables S3). The third included motility-associated genes: \u003cem\u003efliG \u003c/em\u003eand \u003cem\u003eflgD \u003c/em\u003e(Tables S2, S3). In the homarine-only treatment, we observed broader upregulation of the flagellar and chemotaxis genes, including \u003cem\u003efliM\u003c/em\u003e, \u003cem\u003efliN\u003c/em\u003e, \u003cem\u003emotB\u003c/em\u003e, \u003cem\u003eaer\u003c/em\u003e, \u003cem\u003echeR\u003c/em\u003e, and \u003cem\u003echeY\u003c/em\u003e\u003csup\u003e24,25\u003c/sup\u003e (Tables S2). These gene enrichments suggest that homarine may induce motility and chemotaxis.\u003c/p\u003e\n\u003cp\u003eTo target functional assessments, we identified orthologs of the homarine-induced genes from \u003cem\u003eCobetia\u003c/em\u003e sp. OBi1 in the model marine heterotrophic bacteria \u003cem\u003eRuegeria pomeroyi\u003c/em\u003e DSS-3 using reciprocal best high searches (e-value \u0026lt; 10⁻⁵). Of the seven genes in \u003cem\u003eCobetia sp.\u003c/em\u003e OBi1\u0026rsquo;s homarine-induced operon with unknown function (ACFLL8_11490\u0026ndash;ACFLL8_11520), five were found in a single operon in \u003cem\u003eRuegeria pomeroyi\u003c/em\u003e DSS-3 (SPO3188\u0026ndash;SPO3192; Figure 1C, Table S3). Orthologs of ACFLL8_11490 (the first catabolic gene) and ACFLL8_11505 (the glycine betaine transporter) were absent. \u003cem\u003eR. pomeroyi\u003c/em\u003e DSS-3 was capable of growth on homarine as the sole carbon and nitrogen source (Figure SX1A). These results support a homarine catabolic operon here termed \u003cem\u003ehomABCDER \u003c/em\u003e(Figure 1C, Table S3), consisting of four core catabolic genes (\u003cem\u003ehomABCD\u003c/em\u003e), a transcriptional regulator (\u003cem\u003ehomR\u003c/em\u003e), and variably present auxiliary gene (\u003cem\u003ehomE\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003eTo investigate the functions and structures of the proteins encoded by the \u003cem\u003ehomABCDER \u003c/em\u003eoperon we combined domain annotation\u003csup\u003e26\u0026ndash;29\u003c/sup\u003e, structural modeling\u003csup\u003e30,31\u003c/sup\u003e and functional prediction\u003csup\u003e32,33\u003c/sup\u003e. In both \u003cem\u003eCobetia \u003c/em\u003esp. OBi1 and \u003cem\u003eR. pomeroyi \u003c/em\u003eDSS-3, HomA contains conserved motifs for a flavin adenine dinucleotide (FAD) binding domain (Table S3). Functional characterization of HomB suggests that it binds cyclic compounds and exhibits catalytic activity (CscoreGo \u0026gt; 0.65, Table S4). AlphaFold modeling yielded high-confidence structures for HomA and HomB (interface predicted template modelling + predicted template modelling: 0.9471 for \u003cem\u003eR. pomeroyi\u003c/em\u003e DSS-3; 0.9246 for \u003cem\u003eCobetia \u003c/em\u003esp. OBi1). Visualization in ChimeraX\u003csup\u003e34\u003c/sup\u003e, revealed hydrogen bonds and electrostatic interactions consistent with heterodimer formation. This resembles the trigonelline catabolic enzymes TgnA and TgnB, which operate as a flavin-supplying and oxygen-activating pair\u003csup\u003e35\u003c/sup\u003e to open the pyridine ring of trigonelline, a structural isomer of homarine. We hypothesize a similar functional relationship for HomA and HomB. HomC is predicted to function as a peptidase (CscoreGO \u0026gt; 0.65, Table S4), while HomD is likely an aldehyde dehydrogenase-like oxidoreductase (CscoreGO \u0026gt; 0.65, Table S3-S4). These findings are consistent with a model in which the operon encodes enzymes involved in distinct but coordinated steps of homarine degradation.\u003c/p\u003e\n\u003ch2\u003e\u003cem\u003ePhenotypic analysis of homarine-associated genes\u003c/em\u003e\u003c/h2\u003e\n\u003cp\u003eWe phenotypically characterized the proposed homarine catabolic operon using recombinant techniques and a transposon barcoded mutant library of our model organisms. In \u003cem\u003eCobetia\u003c/em\u003e sp. OBi1, homologous recombination was used to generate a \u0026Delta;\u003cem\u003ehomD \u003c/em\u003emutant (ACFLL8_11515), which was unable to grow on homarine as a sole carbon and nitrogen source while retaining normal growth on glucose (Figure SX1B). Complementation with the wild-type \u003cem\u003ehomD\u003c/em\u003e gene restored growth on homarine, confirming its role in homarine catabolism (Figure SX1B). Similarly, mutants from a Tn5 transposon barcoded mutant library in\u003cem\u003e R. pomeroyi \u003c/em\u003eDSS-3\u003csup\u003e36\u003c/sup\u003e were used to assess gene phenotypes, (Figure SX1A, Table S3 and S5). Knockouts of three catabolic genes (\u003cem\u003ehomB\u003c/em\u003e, SPO3189; \u003cem\u003ehomC\u003c/em\u003e, SPO3190; \u003cem\u003ehomD\u003c/em\u003e, SPO3191) and the transcriptional regulator (\u003cem\u003ehomR\u003c/em\u003e, SPO3192) resulted in the loss of growth on homarine at 4 mM C (Figure 1D, Tables S3 and S5). These results confirm the role of the homarine operon (\u003cem\u003ehomABCDER\u003c/em\u003e) in the degradation of homarine in our model organisms.\u003c/p\u003e\n\u003cp\u003eA knockout strain targeting the transporter gene (ACFLL8_11505) within the \u003cem\u003ehom \u003c/em\u003eoperon of \u003cem\u003eCobetia\u003c/em\u003e sp. OBi1 (\u0026Delta;\u003cem\u003ehomT\u003c/em\u003e) exhibited reduced growth on homarine compared to the wild-type strain. Plasmid-based complementation with \u003cem\u003ehomT \u003c/em\u003erestored growth relative to a control expressing red fluorescent protein (RFP) (Figure SX2). Given its location in the \u003cem\u003ehom\u003c/em\u003e operon, its upregulation in response to homarine, and the observed growth effects, this gene likely encodes a homarine transporter, which we designate \u003cem\u003ehomT\u003c/em\u003e. However, \u0026Delta;\u003cem\u003ehomT\u003c/em\u003e mutants eventually reached wild-type yields, suggesting functional redundancy in homarine uptake. Such redundancy is common for betaines in marine bacteria\u003csup\u003e37,38\u003c/sup\u003e and competitive inhibition of homarine uptake by other betaines and sulfoniums further supports this in microbial communities\u003csup\u003e12\u003c/sup\u003e. In \u003cem\u003eR. pomeroyi\u003c/em\u003e DSS-3, the lack of mutant precluded testing the function of a non-orthologous transporter adjacent to the \u003cem\u003ehom\u003c/em\u003e operon (SPO3186). To maintain focus, further characterization of homarine transporters was not pursued.\u003c/p\u003e\n\u003ch2\u003e\u003cem\u003eIdentification of homarine degradation products using mass spectrometry\u003c/em\u003e\u003c/h2\u003e\n\u003cp\u003eTo investigate the catabolism of homarine, we analyzed the metabolomes of \u003cem\u003eCobetia \u003c/em\u003esp. OBi1 grown under the glucose, homarine, and glucose + homarine conditions. Metabolites enriched in homarine treatments were identified in intra- and extra-cellular metabolomes (Figure 2A, Figure SX3), calculated relative to \u0026ldquo;core\u0026rdquo; metabolites that are relatively constant across a wide range of marine organisms and environmental regimes\u003csup\u003e11\u003c/sup\u003e. Homarine, glutamic acid, and several unidentified compounds were enriched (Figure SX3, Tables S6 and S7). Two compounds were subsequently identified: \u003cem\u003eN\u003c/em\u003e-methylglutamic acid (via comparison to standard), and \u003cem\u003eN\u003c/em\u003e-methyl glutamine (via MS\u003csup\u003e1\u003c/sup\u003e/MS\u003csup\u003e2\u003c/sup\u003e spectra, Figures SX4 and SX5). To explore conserved degradation products of homarine, we analyzed the metabolomes of \u003cem\u003eR. pomeroyi\u003c/em\u003e DSS-3 grown under similar glucose and homarine conditions. We found the enrichment of \u003cem\u003eN-\u003c/em\u003emethylglutamic acid, \u003cem\u003eN-\u003c/em\u003emethyl glutamine, and compounds with molecular formulas of C\u003csub\u003e6\u003c/sub\u003eH\u003csub\u003e9\u003c/sub\u003eNO\u003csub\u003e3\u003c/sub\u003e, C\u003csub\u003e6\u003c/sub\u003eH\u003csub\u003e9\u003c/sub\u003eNO\u003csub\u003e4\u003c/sub\u003e, and C\u003csub\u003e7\u003c/sub\u003eH\u003csub\u003e9\u003c/sub\u003eNO\u003csub\u003e5\u003c/sub\u003e, when grown with homarine, consistent with \u003cem\u003eCobetia \u003c/em\u003esp. OBi1 (Figure 2A-C, Tables S8 and S9).\u003c/p\u003e\n\u003cp\u003eStable-isotope probing was performed to track homarine degradation products in natural marine microbial communities from three locations (Figure 2D). Seawater incubated with isotopically-labeled homarine (\u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-homarine for TN412 experiments; \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e7\u003c/sub\u003e\u003csup\u003e15\u003c/sup\u003eN-homarine for RC104 and TN412 experiments, Figure 2D inset) was analyzed for the compounds enriched in our model organisms, with the isotopic labels. In the TN397 experiments, \u003csup\u003e3\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-labeled \u003cem\u003eN-\u003c/em\u003emethylglutamic acid and \u003cem\u003eN-\u003c/em\u003emethyl glutamine were enriched, alongside the previously observed unidentified compounds with formulas of C\u003csub\u003e6\u003c/sub\u003eH\u003csub\u003e9\u003c/sub\u003eNO\u003csub\u003e3\u003c/sub\u003e, C\u003csub\u003e6\u003c/sub\u003eH\u003csub\u003e9\u003c/sub\u003eNO\u003csub\u003e4\u003c/sub\u003e, and C\u003csub\u003e7\u003c/sub\u003eH\u003csub\u003e9\u003c/sub\u003eNO\u003csub\u003e5\u003c/sub\u003e (Figure 2E, Table S9). Consistent with the removal of a \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e- methyl group from \u003cem\u003eN\u003c/em\u003e-methyglutamic acid, we did not find \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-glutamic acid (Table S9). In the RC104 and TN412 experiments, we identified enrichment of fully \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e5-7\u003c/sub\u003e,\u003csup\u003e15\u003c/sup\u003eN\u003csub\u003e1\u003c/sub\u003e- \u003cem\u003eN-\u003c/em\u003emethylglutamic acid, glutamic acid, \u003cem\u003eN-\u003c/em\u003emethyl glutamine, and C\u003csub\u003e6\u003c/sub\u003eH\u003csub\u003e9\u003c/sub\u003eNO\u003csub\u003e3\u003c/sub\u003e, C\u003csub\u003e6\u003c/sub\u003eH\u003csub\u003e9\u003c/sub\u003eNO\u003csub\u003e4\u003c/sub\u003e, C\u003csub\u003e7\u003c/sub\u003eH\u003csub\u003e9\u003c/sub\u003eNO\u003csub\u003e5 \u003c/sub\u003efeatures, confirming them as direct degradation products of homarine. (Figure 2F, Table S9).\u003c/p\u003e\n\u003cp\u003eThe consistency in the enriched metabolites in \u003cem\u003eR. pomeroyi \u003c/em\u003eDSS-3, \u003cem\u003eCobetia \u003c/em\u003esp. OBi1, and natural microbial communities support a conserved degradation of homarine to \u003cem\u003eN-\u003c/em\u003emethylglutamic acid and subsequently to glutamic acid. Complete isotopic labeling of \u003cem\u003eN\u003c/em\u003e-methylglutamic acid and glutamic acid confirms their origins from homarine while the isotopic labeling of \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-\u003cem\u003eN-\u003c/em\u003emethylglutamic acid highlights the preservation during ring opening and subsequent removal of homarine\u0026rsquo;s methyl group during degradation to glutamic acid. Untargeted analysis of the \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-homarine experiments found no evidence for \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-containing compounds beyond \u003cem\u003eN-\u003c/em\u003emethylglutamic acid and homarine (Table S10), confirming that homarine\u0026rsquo;s pyridine ring is opened before demethylation. While earlier studies imply that homarine functions as a methyl donor to form picolinic acid\u003csup\u003e39\u003c/sup\u003e, in these stable isotope experiments we only observed the methylated pyridine ring opening intact during breakdown.\u003c/p\u003e\n\u003ch2\u003e\u003cem\u003eRole of \u003c/em\u003eN\u003cem\u003e-methyl glutamate dehydrogenase in homarine degradation\u003c/em\u003e\u003c/h2\u003e\n\u003cp\u003eThe discovery of \u003cem\u003eN\u003c/em\u003e-methylglutamic acid as a product of homarine degradation prompted us to reexamine upregulated genes in \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 for involvement with this metabolite. Two gene clusters originally annotated as coding for the heterotetrameric sarcosine oxidase (TSOX) were identified: ACFLL8_02220-02235 and ACFLL8_07290-07305, with the latter showing increased transcription during growth on homarine (Figure 1D, Table S2). Genes encoding TSOX (\u003cem\u003esoxBDAG\u003c/em\u003e) and \u003cem\u003eN\u003c/em\u003e-methyl glutamate dehydrogenase (NMGDH; \u003cem\u003emgdABCD\u003c/em\u003e) often share high sequence similarity, leading to misannotation by automated pipelines\u003csup\u003e40,41\u003c/sup\u003e. Both enzymes act on a methylated secondary amine, with similar CH\u003csub\u003e3\u003c/sub\u003eNHCH(R)COOH backbones (R = H in sarcosine, R = propanoic acid for \u003cem\u003eN\u003c/em\u003e-methylglutamic acid). Phylogenetic analysis revealed that the \u003cem\u003esox-\u003c/em\u003elike genes induced in \u003cem\u003eCobetia \u003c/em\u003esp. OBi1 (ACFLL8_07290 to ACFLL8_07305) are more closely related to \u003cem\u003emgd \u003c/em\u003egenes from bacteria with demonstrated NMGDH activity than those with TSOX activity (Figures SX6-SX9). In \u003cem\u003eR. pomeroyi \u003c/em\u003eDSS-3, distinction between the \u003cem\u003esox\u003c/em\u003e and \u003cem\u003emgd \u003c/em\u003egene clusters was similarly apparent after phylogenetic analysis (Figure SX6-SX9). Furthermore, growth of \u003cem\u003emgdA\u003c/em\u003e::Tn5 and \u003cem\u003emgdD\u003c/em\u003e::Tn5 mutants of \u003cem\u003eR. pomeroyi \u003c/em\u003eDSS-3 was significantly reduced when grown on homarine or \u003cem\u003eN\u003c/em\u003e-methylglutamic acid compared to the wild type (Figure 1F, Table S5). Together, these results suggest that the \u003cem\u003esox\u003c/em\u003e-like genes upregulated in \u003cem\u003eCobetia sp.\u003c/em\u003e OBi1 correspond to the \u003cem\u003emgd \u003c/em\u003egene cluster.\u003c/p\u003e\n\u003cp\u003ePreviously, the only known route for the formation of \u003cem\u003eN\u003c/em\u003e-methylglutamic acid was as an\u003c/p\u003e\n\u003cp\u003eintermediate in glutamate-assisted methylamine utilization, where a methyl group is transferred from methylamines onto glutamic acid, followed by NMGDH-catalyzed demethylation to glutamic acid and formaldehyde or 5,10-methylene-H\u003csub\u003e4\u003c/sub\u003efolate\u003csup\u003e41,42\u003c/sup\u003e. Our field experiments revealed accumulation of \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e5\u003c/sub\u003e,\u003csup\u003e15\u003c/sup\u003eN-glutamic acid and \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e6\u003c/sub\u003e,\u003csup\u003e15\u003c/sup\u003eN-\u003cem\u003eN-\u003c/em\u003emethyglutamic acid from \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e7\u003c/sub\u003e,\u003csup\u003e15\u003c/sup\u003eN-homarine, demonstrating a new pathway for the formation of \u003cem\u003eN-\u003c/em\u003emethyglutamic acid during the conversion from homarine to glutamic acid (Figure 1C, 2). In \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-homarine incubations, \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e2\u003c/sub\u003e-labeled metabolites were detected (Table S10), consistent with methyl group transfer into the folate cycle. Combined with \u003cem\u003emgd \u003c/em\u003eupregulation in \u003cem\u003eCobetia \u003c/em\u003esp. OBi1, these results support the role of NMGDH in catalyzing the conversion of homarine-derived \u003cem\u003eN-\u003c/em\u003emethyglutamic acid to glutamic acid natural communities (Figure 1E). Glutamic acid can be converted into alpha-keto glutarate for entry into the Krebs cycle, play a central role in nitrogen assimilation via glutamine oxoglutarate aminotransferase (GOGAT), contribute to the biosynthesis of metabolites, peptides, proteins, or function as an osmolyte. This positions homarine as a versatile growth substrate for bacteria equipped with genes to convert homarine into glutamic acid.\u003c/p\u003e\n\u003ch2\u003e\u003cem\u003eHomarine-degrading bacteria with conserved homarine catabolic genes\u003c/em\u003e\u003c/h2\u003e\n\u003cp\u003eTo evaluate the distribution of the genetic mechanisms for homarine catabolism, we isolated 26 bacteria strains from coastal and pelagic environments using homarine-enriched agar plates and sequenced their genomes. Sixteen isolates with \u0026lt;99% Average Nucleotide Identity (ANI) were selected for further analysis and screened for axenic growth in oligotrophic seawater supplemented with homarine and phosphate; nine of these were capable of growing on homarine as a sole carbon and nitrogen source. Homarine catabolism was taxonomically restricted to Alphaproteobacteria and Gammaproteobacteria. All homarine-utilizing isolates encoded \u003cem\u003ehomABCD\u003c/em\u003e, \u003cem\u003ehomR, \u003c/em\u003eand sometimes \u003cem\u003ehomE\u003c/em\u003e in a conserved operon (Figure 3A, Table S11). An \u003cem\u003emgd\u003c/em\u003e operon was identified in 8 of 9 genomes with the \u003cem\u003ehom\u003c/em\u003e operon, and all \u003cem\u003emgd\u003c/em\u003e-positive isolates also carried \u003cem\u003esoxGADB \u003c/em\u003e(Figure 3A, Table S11, Figures SX6-9). Notably, \u003cem\u003eVibrio\u003c/em\u003e isolate PS01 contained the \u003cem\u003ehom\u003c/em\u003e operon but lacked \u003cem\u003emgdABCD,\u003c/em\u003e yet still grew on homaine, suggesting genome incompleteness or an alternative pathway for processing \u003cem\u003eN\u003c/em\u003e-methylglutamic acid. One isolate, \u003cem\u003eNitratireductor\u003c/em\u003e sp. G4i25, encoded \u003cem\u003emgdABCD \u003c/em\u003ebut lacked the \u003cem\u003ehom \u003c/em\u003eoperon and could not grow on homarine, reinforcing the need of both gene sets for homarine catabolism (Figure 3A, Table S11).\u003c/p\u003e\n\u003ch2\u003e\u003cem\u003eWidespread distribution and conservation of \u003c/em\u003ehom\u003cem\u003e operon\u003c/em\u003e\u003c/h2\u003e\n\u003cp\u003eTo explore the taxonomic and geographic distribution of bacteria capable of homarine degradation, we searched complete genomes in the NCBI database for operons encoding the homarine catabolic genes (\u003cem\u003ehomA\u003c/em\u003eBCD and optionally \u003cem\u003ehomE\u003c/em\u003e). Due to sequence similarity between \u003cem\u003esox \u003c/em\u003eand \u003cem\u003emgd \u003c/em\u003egenes, we did not attempt the same search for the \u003cem\u003emgd operon\u003c/em\u003e\u003csup\u003e40\u003c/sup\u003e (Figures SX6-SX9). Across 11,755 genomes containing the \u003cem\u003ehom \u003c/em\u003eoperon, we identified primarily Alphaproteobacteria (3,148 genomes, 26.8%) and Gammaproteobacteria (8,094 genomes, 68.9%) lineages (Figure 3B, Table S12). Operon organization was highly conserved; \u003cem\u003ehomA\u003c/em\u003e and \u003cem\u003ehomB\u003c/em\u003e were adjacent in 97.7% of genomes, consistent with the proposed homodimer structure, while \u003cem\u003ehomC\u003c/em\u003e and \u003cem\u003ehomD \u003c/em\u003ewere typically nearby (Figure SX10A, Table S12). In contrast, the presence and positioning of \u003cem\u003ehomE\u003c/em\u003e was variable (Figure SX10A, Table S12). Approximately one-third of the genomes containing\u003cem\u003e homABCD\u003c/em\u003e do not include \u003cem\u003ehomE\u003c/em\u003e. In genomes with \u003cem\u003ehomE\u003c/em\u003e, it is positioned upstream of \u003cem\u003ehomA \u003c/em\u003ein 49.9% (like \u003cem\u003eCobetia \u003c/em\u003esp. OBi1) and downstream of \u003cem\u003ehomD \u003c/em\u003ein 50.1% (Figure SX10A, Table S12). This variability suggests \u003cem\u003ehomE\u003c/em\u003e functions as an auxiliary module in the \u003cem\u003ehom \u003c/em\u003eoperon, but not a core component. Gammaproteobacteria retained \u003cem\u003ehomE\u003c/em\u003e in 93% of the genomes containing the \u003cem\u003ehom\u003c/em\u003e operon (Table S12), suggesting it may extend or enhance catabolic potential. By contrast, 94% of the Alphaproteobacteria genomes lacked \u003cem\u003ehomE\u003c/em\u003e - including all of the 120 genomes of the highly abundant\u003cem\u003e Pelagibacter \u003c/em\u003eclade (Figure SX10A, Table S12). The absence of \u003cem\u003ehomE \u003c/em\u003ein Alphaproteobacteria, particularly in smaller genomes (\u003cem\u003et\u003c/em\u003e-test,\u003cem\u003e p\u003c/em\u003e = 2.2\u0026times;10⁻\u0026sup1;⁶, \u003cem\u003en\u003c/em\u003e = 11,755, Figure SX10B), reflects relaxed selective pressure. These observations support a model where\u003cem\u003e homE \u003c/em\u003eserves a taxon-specific auxiliary role for homarine metabolism.\u003c/p\u003e\n\u003cp\u003eWe identified bacteria encoding the \u003cem\u003ehom\u003c/em\u003e operon from a global range of ecosystems, including soil, freshwater, marine, sediments, food \u003cem\u003e(e.g.\u003c/em\u003e. cheese, oysters), and human clinical samples (\u003cem\u003ee.g ...\u003c/em\u003e urine, fecal, Figure 3C, Table S12). Although gene presence does not guarantee functionality, detecting the operon in non-aquatic environments suggests homarine may play roles in microbial metabolism beyond marine systems. Among genomes encoding the operon were members of ecologically significant marine heterotrophic clades, including SAR11 (Pelagibacterales)\u003csup\u003e43\u003c/sup\u003e\u003cem\u003e, \u003c/em\u003eSAR116 (Puniceispirillales)\u003csup\u003e44,45\u003c/sup\u003e, SAR92 (Cellvibrionales)\u003csup\u003e45,46\u003c/sup\u003e, and Rhodobacterales including members of the Roseobacter clade\u003csup\u003e47,48\u003c/sup\u003e (Figure 3C, Table S12). Notably, SAR11 accounts for ~25% of microbial cells in the ocean\u003csup\u003e49\u003c/sup\u003e, SAR92 represents up to 10% of cells in nearshore waters\u003csup\u003e50\u003c/sup\u003e, and Roseobacters comprise up to 20% of bacterial cells in coastal ecosystems\u003csup\u003e51\u003c/sup\u003e. SAR11\u0026rsquo;s streamlined genome retains the \u003cem\u003ehom\u003c/em\u003e operon, underscoring its importance to Pelagibacterales and suggesting a key role for homarine in their ecological success. The widespread occurrence of the \u003cem\u003ehom \u003c/em\u003eoperon across diverse regimes and its presence in dominant marine microbes highlights homarine\u0026rsquo;s broad ecological significance and its pervasive role in microbial metabolism.\u003c/p\u003e\n\u003ch2\u003e\u003cem\u003eExpression of \u003c/em\u003ehom\u003cem\u003e catabolic genes in marine microbes\u003c/em\u003e\u003c/h2\u003e\n\u003cp\u003eTo determine whether the genomic potential of homarine catabolism translates to metabolic activity \u003cem\u003ein situ\u003c/em\u003e, we evaluated gene expression under experimental and environmental conditions. In homarine-amended incubations in Puget Sound, significant transcriptional enrichment was detected for 13 bacterial genes relative to unamended controls (Table S13). Transcripts of \u003cem\u003ehomC\u003c/em\u003e and \u003cem\u003ehomD\u003c/em\u003e were enriched in members of the Porticoccaceae (Gammaproteobacteria), and \u003cem\u003ehomB\u003c/em\u003e was enriched in Oceanospirillaceae (Alphaproteobacteria) (Table S13)\u003cem\u003e. \u003c/em\u003eThese taxa were identified in our genome search with the genetic capacity for homarine catabolism (Figure 3B, Table S12). Additionally, \u003cem\u003emgdA, \u003c/em\u003ethe gene encoding the \u003cem\u003e\u0026alpha;\u003c/em\u003e-subunit of the NMGDH, was up-regulated in Porticoccaceae (Table S13), reinforcing the role of NMGDH in converting homarine-derived \u003cem\u003eN-\u003c/em\u003emethyglutamic acid to glutamic acid and linking the genomic capacity to catabolic activity in natural microbial assemblages.\u003c/p\u003e\n\u003cp\u003eWe examined the in situ expression of the \u003cem\u003ehom \u003c/em\u003egenes using quantitative metatranscriptomics along a surface ocean transect from the subtropical gyre into the North Pacific Transition Zone (25.87\u0026deg;N to 40.88\u0026deg;N on 158\u0026deg;W). Spatially structured expression of \u003cem\u003ehomB\u003c/em\u003e and \u003cem\u003ehomA \u003c/em\u003ewas observed, with elevated transcript abundance north of 35\u0026deg;N (Figure 4, Figure SX11). Expression of \u003cem\u003ehomB \u003c/em\u003ewas primarily attributed to Gammaproteobacteria (e.g. Vibrionales, Porticoccaceae) and Alphaproteobacteria (e.g. \u003cem\u003ePelagibacter\u003c/em\u003e, SAR116, Rhodobacterales), alongside unresolved lineages (Figure 4, Table S14). Across stations, homarine concentrations (sum of dissolved and particulate measurements) were significantly correlated with \u003cem\u003ehom\u003c/em\u003e gene transcript abundance (\u003cem\u003eR\u0026sup2;\u003c/em\u003e = 0.518, \u003cem\u003ep\u003c/em\u003e = 0.0052, Spearman\u0026rsquo;s correlation), supporting coupling between substrate availability and gene expression.\u003c/p\u003e\n\u003cp\u003eTaxa expressing \u003cem\u003ehom\u003c/em\u003e genes \u003cem\u003ein situ\u003c/em\u003e overlapped with those identified in our perturbation experiments and predicted to encode homarine catabolic pathways from genome-based analyses. These taxa spanned both copiotrophic and oligotrophic lineages, such as Porticoccaceae, Vibrionales, Pelagibacteraceae, SAR116, Oceanospirallaceae, and Rhodobacterales. The detection of transcripts from both fast-growing coastal taxa and genome-streamlined open-ocean bacteria suggests homarine catabolism operates across diverse ecological strategies and niches. These findings highlight homarine degradation as a responsive and ecologically widespread metabolic capacity, with gene expression linked to substrate availability in marine environments.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study defines a core microbial pathway for the degradation of homarine, a ubiquitous yet previously enigmatic metabolite in marine systems\u003csup\u003e6,7,11\u003c/sup\u003e. Using comparative transcriptomics, transposon mutagenesis, and targeted gene deletions, we identified a conserved operon (\u003cem\u003ehomABCDER\u003c/em\u003e) that mediates homarine catabolism in two model organisms, a newly isolated gammaproteobacteria \u003cem\u003eCobetia sp\u003c/em\u003e. OBi1 and the genetically tractable model marine alphaproteobacteria \u003cem\u003eRuegeria pomeroyi \u003c/em\u003eDSS-3.\u003c/p\u003e\n\u003cp\u003eMetabolomic analyses in both model organisms and natural communities reveal that homarine degradation produces \u003cem\u003eN\u003c/em\u003e-methylglutamic acid, which is subsequently converted into glutamic acid by \u003cem\u003eN\u003c/em\u003e-methylglutamate dehydrogenase (NMGDH). This refines the understanding of NMGDH, a gene that is frequently misannotated\u003csup\u003e40,41\u003c/sup\u003e, and expands its ecological importance in marine microbial metabolism. As a metabolic product of homarine, glutamic acid plays diverse biochemical roles, including nitrogen assimilation, biosynthesis of metabolites and proteins, and energy metabolism. These findings position homarine as a versatile growth substrate and highlights homarine\u0026rsquo;s broad impact on microbial growth and resource allocation.\u003c/p\u003e\n\u003cp\u003eThe \u003cem\u003ehom\u003c/em\u003e operon is broadly distributed across bacterial genomes from diverse environments, including coastal and open-ocean systems, as well as non-aquatic sources. Within bacteria with demonstrated homarine-degradation capabilities, there is a frequent co-occurrence of the \u003cem\u003ehom\u003c/em\u003e operon with the \u003cem\u003emgd\u003c/em\u003e gene cluster, which encodes NMGDH and facilitates the conversion of \u003cem\u003eN\u003c/em\u003e-methylglutamic acid to glutamic acid. This genomic association underscores the interdependence of the \u003cem\u003ehom\u003c/em\u003e and \u003cem\u003emgd\u003c/em\u003e pathways, highlighting their integrated role in nutrient cycling. Genomic mining further revealed that the \u003cem\u003ehom \u003c/em\u003eoperon is conserved in globally dominant bacterial clades that are key drivers of carbon and nitrogen cycling in marine ecosystems, including SAR11\u003csup\u003e43\u003c/sup\u003e, SAR116\u003csup\u003e44,45\u003c/sup\u003e, and Rhodobacterales\u003csup\u003e47,48\u003c/sup\u003e. Furthermore, variability in the auxiliary \u003cem\u003ehomE\u003c/em\u003e gene reveals operon modularity and its adaptive tailoring to ecological or genomic constraints.\u003c/p\u003e\n\u003cp\u003eThrough perturbation experiments and \u003cem\u003ein situ\u003c/em\u003e analyses of gene expression, we show the coupling of homarine availability to \u003cem\u003ehomB\u003c/em\u003e gene expression in natural marine environments. This demonstrates a clear relationship between substrate availability and microbial metabolic activity, supporting homarine as a regulatory driver of microbial gene function. The responsiveness of these genes to substrate availability suggests that homarine degradation actively shapes microbial community structure and metabolic dynamics \u003cem\u003ein situ\u003c/em\u003e. Co-expression of chemotaxis and motility genes alongside the \u003cem\u003ehom\u003c/em\u003e operon indicates that bacteria not only sense and metabolize homarine but may actively seek it out in the environment. These findings suggest that homarine degradation contributes to the structuring of microbial communities and influences resource-driven interactions in the ocean.\u003c/p\u003e\n\u003cp\u003eCollectively, this study provides a foundational framework to link metabolic gene function to ecological activity, bridging molecular mechanisms with environmental significance. By elucidating the genetic basis, biochemical transformations, and ecological distribution of homarine catabolism, we expand our understanding of how microbial communities integrate metabolite turnover into broader oceanic biogeochemical cycles. These findings open opportunities for further exploration into how homarine interacts with co-occurring metabolites, impacts microbial interactions, and supports ecosystem functioning in the ocean and beyond.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cem\u003e\u003cu\u003eIsolation, growth, and sequencing of \u003c/u\u003e\u003c/em\u003e\u003cu\u003eCobetia sp. \u003cem\u003eOBi1 \u003c/em\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eIsolate Owen Beach isolate 1 (OBi1) was obtained by enrichment culturing of seawater collected in Owen Beach (Tacoma, Washington). Seawater was collected in sterile polycarbonate bottles in February 2020. We amended 100 mL seawater samples with 12 mM C homarine and 0.1 mM sodium phosphate and incubated in aerated flasks at 22\u0026deg;C. Culture OD600 was measured daily to monitor growth. After three days, 50 uL samples were streaked on agar plates made with marine broth (Difco 2216) to isolate single colonies. Single colonies were picked and re-streaked twice in the same agar media. The isolated colonies were tested for growth on homarine in seawater from Owen Beach sterilized by filtering through a 0.22 um PES membrane and autoclaving. We supplied the medium with homarine and phosphate as before. OBi1 colonies re-grew on homarine and the isolate was selected for further analysis.\u003c/p\u003e\n\u003cp\u003eWe grew OBi1 in seawater medium containing homarine and phosphate in aerated flasks at 25\u0026deg;C to obtain DNA for genome sequencing. Genomic DNA was purified with the NEB Monarch tissue lysis kit using a modified lysis solution containing lysozyme, Proteinase K and RNase A. For whole genome sequencing, we generated a set of Illumina paired-end reads with a depth of 400 Mbp on a NextSeq 2000 and a set of Oxford Nanopore reads at the Microbial Genome Sequencing Center (Pittsburgh, PA). Quality control and adapter trimming was performed with bcl2fastq\u003csup\u003e52\u003c/sup\u003e and porechop\u003csup\u003e53\u003c/sup\u003e for Illumina and Nanopore reads, respectively. A hybrid assembly with both sets of reads was performed with Unicycler using the default parameters\u003csup\u003e54\u003c/sup\u003e. Assembly statistics were recorded with QUAST\u003csup\u003e55\u003c/sup\u003e. The assembly annotation was performed with Prokka\u003csup\u003e56\u003c/sup\u003e. The genome was analyzed with the NCBI prokaryotic annotation and TypeMat tools from the Microbial Genomes Atlas (MiGA) web server\u003csup\u003e23\u003c/sup\u003e to resolve its taxonomic classification and completeness.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003eTranscriptomics of \u003c/u\u003e\u003c/em\u003e\u003cu\u003eCobetia sp.\u003cem\u003e OBi1 grown on homarine\u003c/em\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eWe generated and compared transcriptomes of OBi1 cultures grown in three conditions with (i) glucose, (ii) glucose+homarine, or (iii) homarine to identify genes involved in homarine catabolism. The growth media for transcriptomics experiments were prepared with sterile-filtered and autoclaved oligotrophic surface seawater from Hawaii to reduce background carbon and nutrients. The medium was amended with PRO99 trace metals and 0.1 mM sodium phosphate. Growth medium (i) was amended with 12 mM carbon as glucose and 0.8 mM nitrogen as ammonium. Growth medium (ii) was the same as medium (i) but was spiked with 1 mM carbon as homarine 30 minutes before RNA extraction. Growth medium (iii) contained 12 mM carbon as homarine and no additional nitrogen source. Before the transcriptomics experiment, we grew 5-mL cultures of OBi1 in each medium (i-iii) to determine the OD600 of cultures in the mid-exponential phase. For the transcriptomic experiment, we pre-grew OBi1 in 5 mL of glucose medium (i), of which 4 mL were pelleted and washed three times with 1 mL of sterile oligotrophic Hawaii seawater and resuspended back in 4 mL of seawater. We used 0.5 mL of washed cells to inoculate 75 mL of each growth medium (i-iii) in duplicate for transcriptomics. Cultures were incubated in aerated flasks at 25\u0026deg;C and 200 rpm for a period of 29-30 hours during which we sub-sampled 0.5-1 mL from each replicate to track growth by OD600. Absorbance measurements were obtained with a Genesys 20 spectrophotometer (Spectronic Instruments). Cells were harvested from 40 mL per replicate when cultures were in mid-exponential phase. Samples were split before RNA extraction with a Zymo Research Quick-RNA\u0026trade; Fungal Bacterial MiniPrep kit. DNA was digested with DNase I from NEB. RNA samples were repurified with a Zymo Research RNA Clean \u0026amp; Concentrator\u0026trade;-25 kit. The samples obtained were measured in a NanoDrop\u0026trade; (Thermo Fisher Scientific) and those with \u0026gt; 50 ng/\u0026mu;L concentration were selected for RNA sequencing. RNA samples were sent to the Microbial Genome Sequencing Center for RNA-Seq. DNA samples were treated with RNase free DNase (Invitrogen). Library preparation was performed using Illumina\u0026rsquo;s Stranded Total RNA Prep Ligation with Ribo-Zero Plus kit and 10bp IDT for Illumina indices. Sequencing was done on a NextSeq2000 giving 2x51bp reads. Demultiplexing, quality control, and adapter trimming was performed with bcl2fastq (v20.20.0.445). RNA-Seq generated \u0026gt; 12 M paired-end reads per sample.\u003c/p\u003e\n\u003cp\u003eRNA-Seq paired-end reads were aligned to the OBi1 genome assembly to quantify the number of reads mapped to each coding gene using \u003cem\u003eRsubread\u003c/em\u003e\u003csup\u003e57\u003c/sup\u003e. DESeq2 was used to normalize gene counts and resolve differentially expressed genes\u003csup\u003e58\u003c/sup\u003e between each pair of OBi1 growth conditions. Significant results were resolved with a log-fold change threshold \u0026gt; (+/-) 1.99 and \u003cem\u003ep\u003c/em\u003e\u0026lt;0.05 (Wald test) following adjustment for false discovery.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003eFunctions and structures of the homABCDER operon\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo functionally annotate proteins of the \u003cem\u003ehomABCDER\u003c/em\u003e operon, we used domain and pathway annotation tools including InterProScan\u003csup\u003e29\u003c/sup\u003e, KEGG\u003csup\u003e26\u0026ndash;28\u003c/sup\u003e, and COFACTOR\u003csup\u003e32,33\u003c/sup\u003e. Gene Ontology (GO) terms and enzyme classifications were retrieved from InterPro and COFACTOR, and annotations were retained when COFACTOR CscoreGO values exceeded 0.65. KEGG annotations were obtained by querying the KEGG REST API using gene identifiers for \u003cem\u003eR. pomeroyi\u003c/em\u003e DSS-3 (SPO3188\u0026ndash;SPO3192). Structural models of HomA and HomB were generated using AlphaFold\u003csup\u003e30,31\u003c/sup\u003e, and the highest-scoring model for each protein was selected based on the combined iptm + ptm confidence score. Predicted structures were visualized in ChimeraX\u003csup\u003e34\u003c/sup\u003e to assess domain organization and potential protein\u0026ndash;protein interaction interfaces.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003eGeneration of mutants of homarine operon in \u003c/u\u003e\u003c/em\u003e\u003cu\u003eCobetia sp.\u003cem\u003e OBi1 \u003c/em\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eAll \u003cem\u003eCobetia sp. \u003c/em\u003eexperiments were conducted using the OBi1 strain. Plasmids were conjugated into OBi1 using the donor bacterial strain \u003cem\u003eEscherichia coli\u003c/em\u003e WM3064, [a derivative of B2155\u003csup\u003e59\u003c/sup\u003e provided by W. Metcalf (University of Illinois, Urbana)], which can grow only in the presence of diaminopimelate (DAP). WM3064 carrying plasmids was grown overnight at 30-37\u0026deg;C, shaking in LB media containing 30-50ug/mL kanamycin final concentration and 0.3 mM DAP final concentration. OBi1 was grown overnight at 25-30\u0026deg;C, shaking in Marine Broth 2216 (MB) media (either Difco or NutriSelect\u0026reg; Plus). To set up conjugations, overnight cultures were mixed in a 1:1 volume ratio, pellet for 1 min at 14000 rpm, resuspended in 1/30th of its initial mixed volume, and spotted onto an MB agar plate containing 0.3 mM DAP. Conjugations were then incubated at room temperature for at least 24 hours. Transconjugant selection was performed by dilution streaking for single colonies from overnight spotted mating on MB agar plates without DAP. The kanamycin concentration of MB agar plates for OBi1 transconjugants is 5 ug/mL for integrating plasmids and 25 ug/mL for replicating plasmids.\u003c/p\u003e\n\u003cp\u003eStrains harboring replicating plasmids for empty vector, fluorescence, or genetic- complementation used the pBVMCS-2 plasmid\u003csup\u003e60\u003c/sup\u003e or the pGingerBK-LacUV5 plasmid\u003csup\u003e61\u003c/sup\u003e. Both plasmid transformations used the conjugation protocol outlined in the previous paragraph. Gene deletion strains were generated using the integrating plasmid pNPTS138 (2) for \u003cem\u003ehomD\u003c/em\u003e (protein accession ACFLL8_11515) and a modified pNPTS138 for \u003cem\u003ehomT\u003c/em\u003e (protein accession ACFLL8_11505). pNPTS138 transformation and integration occurs at a chromosomal site homologous to the insertion sequence in pNPTS138. The modified pNPTS138 replaced the aph(3\u0026rsquo;)-Ia kanamycin resistance gene with aph(3\u0026rsquo;)-IIa, added the lacI gene, and replaced the sacB gene promoter with the lac-operator. Resistance cassette, lacI, and lac-operator were sourced from\u003csup\u003e62\u003c/sup\u003e; henceforth, the modified pNPTS138 plasmid will be called pMZT1.\u003c/p\u003e\n\u003cp\u003eSingle colony transconjugants were inoculated into liquid MB media at 25-30\u0026deg;C shaking for 48-72 hours for non- selective growth. Nonselective liquid growth allows for a second recombination event to occur, which either restores the native locus or replaces the native locus with the insertion sequence that was engineered into pNPTS138. Counter-selection for the second recombination of pNPTS138 excision was carried out by passing non-selective cultures (1:100 dilution) into new liquid MB media with 10-20% (w/v) final sucrose concentration for 6-8 hours. For counter-selections with pMZT1 a final concentration of 0.5 mM IPTG was added. Serial dilutions of counter-selection cultures were plated on MB agar with 10-20 % sucrose, and for counter-selection with the pMZT1 plasmid, the sucrose agar contained 0.5mM IPTG.\u003c/p\u003e\n\u003cp\u003eColonies were subjected to PCR genotyping and sequencing to confirm allele replacement. For \u003cem\u003ehomD \u003c/em\u003edeletion analysis, primers homD_KO check F/R were utilized, while primers homT_comp F and homT_KO check R validated the deletion of \u003cem\u003ehomT\u003c/em\u003e (Table S16). Plasmid construction involved various combinations of restriction enzyme digestion, PCR amplification, and Gibson Assembly. The plasmid pNPTS138-\u003cem\u003ehomD\u003c/em\u003e-KO was generated by double digestion of pNPTS138 with NheI and HindIII followed by assembly with a synthesized in-frame knockout (KO) allele of \u003cem\u003ehomD\u003c/em\u003e from GeneWiz, then transformed into chemically competent WM3064 cells. Similarly, pMZT1-\u003cem\u003ehomT\u003c/em\u003e-KO was constructed by double digestion of pMZT1 with NheI and HindIII and assembly with a synthesized KO allele of \u003cem\u003ehomT\u003c/em\u003e from GeneWiz, followed by Gibson Assembly and transformation into WM3064 cells. The pMZT1 plasmid itself was created using two PCR fragments\u0026mdash;one amplified from pGinger-LacUV (2.6 kb) and another from pNPTS138 (4 kb)\u0026mdash;assembled via Gibson Assembly and transformed into Zymo Mix and Go DH5-alpha chemically competent cells. For pBVMCS-\u003cem\u003ehomD\u003c/em\u003e, pBVMCS-2 was linearized via double digestion with EcoRI and NdeI and assembled with a PCR fragment containing the native \u003cem\u003ehomD\u003c/em\u003e promoter and coding sequence amplified from OBi1 genomic DNA (gDNA), followed by transformation into WM3064 cells. Construction of pLacUV-homT involved linearization of pGinger-LacUV with EcoRI and BamHI, assembly with a PCR fragment containing the homT coding sequence amplified from OBi1 gDNA, and placement of \u003cem\u003ehomT\u003c/em\u003e under an IPTG-inducible promoter. The final construct was assembled with Gibson Assembly and transformed into WM3064 cells.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003ePhenotype experiments with \u003c/u\u003e\u003c/em\u003e\u003cu\u003eCobetia sp.\u003cem\u003e OBi1 knockout mutants\u003c/em\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eTo determine OBi1 requires \u003cem\u003ehomD \u003c/em\u003eto grow on homarine, we grew the wild type and \u0026Delta;\u003cem\u003ehomD \u003c/em\u003estrains in seawater medium containing homarine at 12 mM carbon concentration. We used a medium with glucose at 12 mM carbon and 1.7 mM ammonium as a positive control for growth. The base seawater medium contained PRO99 trace metals and 0.2 mM phosphate to support growth. Wild type and \u0026Delta;\u003cem\u003ehomD\u003c/em\u003e strains were grown in \u0026frac12; YTSS medium without antibiotics. The \u003cem\u003e\u0026Delta;homD \u003c/em\u003ecomplementation strain (\u0026Delta;\u003cem\u003ehomD\u003c/em\u003e+pBVMCS-\u003cem\u003ehomD\u003c/em\u003e) and its empty vector control strain (\u0026Delta;\u003cem\u003ehomD\u003c/em\u003e+pBVMCS) were grown with kanamycin (50 \u0026micro;g/mL) selection. Strains were grown from a single colony in \u0026frac12; YTSS overnight to an OD600 of 0.8. Culture aliquots of 1 mL were pelleted at 3000 x g to harvest cells. Cells were washed five times with base seawater medium and starved overnight to reduce background growth. Washed cells were diluted 100-fold in seawater and 10 uL were arrayed into 96-well plates containing 0.3 ml per well of the homarine or glucose test media. The plates were incubated on a shaker at 22\u0026deg;C removed from light. Discrete OD600 measurements were obtained on a SpectraMax Plus microplate reader (Molecular Devices). The background seawater absorbance was subtracted from all measurements to quantify growth.\u003c/p\u003e\n\u003cp\u003eTo compare the growth of wild-type OBi1 and \u0026Delta;\u003cem\u003ehomT\u003c/em\u003e, cultures were pre-grown in \u0026frac12; YTSS incubated overnight at 200 RPM and 25\u0026deg;C. Overnight cultures of \u0026Delta;\u003cem\u003ehomT\u003c/em\u003e plasmid complementation strains incubated with 0.5 mM IPTG to induce RFP or HomT expression. Cell pellets of 1 mL of culture were washed five times in base seawater medium. Washed cells were diluted twenty-fold in seawater and 10 uL were arrayed into 96-well plates containing 0.3 ml per well of the homarine or glucose test media. Plasmid complementation strains incubated with varying concentrations of IPTG (0-0.5 mM). The plates were covered with breathable film to reduce evaporative water loss and incubated at 25\u0026deg;C in the microplate reader to record absorbance and RFP fluorescence (excitation and emission). A logistic growth model was implemented in R with the logit and non-linear least squares (nls) function in the \u003cem\u003ecar\u003c/em\u003e package to estimate the growth rate parameter. A one-way ANOVA and the median test for multiple comparisons from the \u003cem\u003eagricolae\u003c/em\u003e package were implemented in R to test for significant differences in growth rates.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003ePhenotype experiments with mutants of Ruegeria pomeryoi DSS-3\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eGene clusters for \u003cem\u003eCobetia \u003c/em\u003esp. OBi1; ACFLL8_11490 (\u003cem\u003ehomE\u003c/em\u003e), ACFLL8_11495 (\u003cem\u003ehomA\u003c/em\u003e), ACFLL8_11500 (\u003cem\u003ehomB\u003c/em\u003e), ACFLL8_11510 (\u003cem\u003ehomC\u003c/em\u003e), ACFLL8_11515 (\u003cem\u003ehomD\u003c/em\u003e), ACFLL8_11520 (\u003cem\u003ehomR\u003c/em\u003e) were pairwise aligned to \u003cem\u003eRuegeria pomeroyi\u003c/em\u003e using blastp with a BLOSUM62 matrix\u003csup\u003e63\u003c/sup\u003e, Table S3). To confirm the involvement of homarine catabolic genes, we utilized RB-TnSeq mutants of \u003cem\u003eRuegeria pomeroyi \u003c/em\u003eDSS-3. Detailed methods for generating and arraying the \u003cem\u003eR. pomeroyi\u003c/em\u003e DSS-3barcoded mutant library are provided and summarized in previous work\u003csup\u003e36\u003c/sup\u003e\u003csup\u003e64\u003c/sup\u003e. Mutant cultures were pre-grown overnight in \u0026frac12; YTSS broth containing 50 \u0026mu;g/ml kanamycin. Screens were performed in L1 minimal medium amended with Basal Medium vitamins\u003csup\u003e65,66\u003c/sup\u003e, 0.8 mM ammonium, 50 \u0026mu;g/ml kanamycin, and 100 \u0026mu;M phosphorus as PO\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e3-\u003c/sup\u003e. Washed (3x) overnight cultures of individual mutants were inoculated at an OD600 of 0.01 into modified L1 medium with a single substrate as the sole carbon source at 4 mM carbon, \u003cem\u003en\u003c/em\u003e = 6. Plates were incubated at 25\u0026deg;C with shaking at 425 rpm for 72 hours, and optical density (OD600) was measured at 5-minute intervals using a Synergy H1 microplate reader (BioTek Instruments, Inc., Vermont, USA), corrected to a pathlength of 1 cm, assuming a volume of 200 \u0026mu;l. As a positive control, the same medium was inoculated with washed overnight cultures of wild-type \u003cem\u003eR. pomeroyi \u003c/em\u003eDSS-3, \u003cem\u003en\u003c/em\u003e = 3. Wild-type \u003cem\u003eR. pomeroyi\u003c/em\u003e DSS-3 was also grown in kanamycin to test the effectiveness of the antibiotic, \u003cem\u003en\u003c/em\u003e =3. For media control, wells filled with 200 \u0026mu;l of medium without inoculum were incubated, \u003cem\u003en\u003c/em\u003e = 3.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003ePreparation, mass spectrometry, and analysis of metabolomics experiment with \u003c/u\u003e\u003c/em\u003e\u003cu\u003eCobetia sp. \u003cem\u003eOBi1 \u003c/em\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCobetia \u003c/em\u003esp. OBi1 was grown in three conditions, as described for comparative transcriptomic analyses: homarine, glucose, and glucose+homarine. Overnight cultures of \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 (100 mL) were grown at 25\u0026deg;C grown in glucose-amended seawater media as described for\u003cem\u003e Cobetia sp. \u003c/em\u003eOBi1 transcriptomics experiments. Next, 10 mL subsamples were taken from overnight culture and centrifuged for 15 minutes at 2800 g, and resuspended in the experimental growth media homarine, glucose, or glucose+homarine, in triplicate (again, as described for \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 transcriptomics experiments). These samples were incubated for 1 hr at 25\u0026deg;C in a dark shaker before harvesting. Cells were harvested by centrifugation for 15 minutes at 2800 g, and supernatant was collected and filtered through a 0.22 um PES membrane filter. Cell pellets and supernatant samples were stored at -80\u0026deg;C and -20\u0026deg;C, respectively.\u003c/p\u003e\n\u003cp\u003eFor particulate metabolomics, cell pellets were extracted using a combination of mechanical and chemical disruption techniques as described in previous work\u003csup\u003e67\u003c/sup\u003e. Metabolites from the supernatant were extracted using a cation-exchange-based solid phase extraction technique as described previously\u003csup\u003e7\u003c/sup\u003e, with 1 mL of supernatant diluted into 10 mL of HPLC grade water. Isotopically-labeled internal standards were added for normalization purposes, as reported in Table S17.\u003c/p\u003e\n\u003cp\u003eMetabolomics data were acquired by liquid chromatography paired with high resolution mass spectrometry (LC-MS) on a ThermoOrbitrap Q-Exactive HF Mass Spectrometer (QE). Particulate analyses were performed as reported\u003csup\u003e67\u003c/sup\u003e with modifications reported previously\u003csup\u003e68\u003c/sup\u003e. In short, samples were introduced via Hydrophilic interaction liquid chromatography (HILIC) in both positive and negative modes using polarity switching. All samples were introduced in full scan mode. In addition to the full scan analyses, samples were pooled and monitored in data-dependent acquisition (DDA) mode to acquire MS\u003csup\u003e2\u003c/sup\u003e data. These data were acquired at three different collision energies (20, 35, and 50 V), with separate injections for each collision energy and ionization mode (positive or negative). Pooled samples were injected at full strength and also diluted 1:1 with water to aid in normalization, as previously described\u003csup\u003e67\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eData from the QE (.raw files) were converted to the open source .mzML file format using MSConvert\u003csup\u003e68\u003c/sup\u003e. For targeted analyses, peaks were integrated in Skyline\u003csup\u003e69\u003c/sup\u003e using a template of known compounds for which we have authentic standards. Peaks were identified by comparing to standards run in reconstitution solvent as well as spiked into a pooled sample. Further quality control was performed to remove small and low-quality peaks using the quality control procedure as described (\u003csup\u003e67\u003c/sup\u003e. Quality control parameters for dissolved and particulate data from HILIC chromatography included a signal to noise ratio of at least 3, a mass within 6 ppm of the standards\u0026rsquo; calculated mass, a retention time within 2.5 minutes of the standards\u0026rsquo; retention time, a minimum area of 40000, and a ratio of signal over signal from blank of at least 3. Finally, peaks were normalized to reduce variability introduced during data acquisition using best-matched internal standard (B-MIS\u003csup\u003e67\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eFor untargeted analyses, we used MS-Dial v4.9\u003csup\u003e70\u003c/sup\u003e to extract mass features using the parameters supplied in Table S18. This resulted in a list of mass features (\u003cem\u003em/z\u003c/em\u003e and retention time features) that could be compared quantitatively between samples, regardless of the ability to identify these compounds. We dereplicated mass features from MSDial by identifying and annotating adducts (including between positive and negative polarity) and isotopes based and consolidated mass features for downstream statistical analyses. Mass features that correspond to our targeted compounds were annotated accordingly. For instances where a mass feature was not detected in all samples of an experiment, we replaced missing values with 0.02 times the minimum detected value for that mass feature within the same experiment, ensuring all features have a baseline value to facilitate calculations and comparisons in downstream analyses.\u003c/p\u003e\n\u003cp\u003eNext, we performed an enrichment analysis to calculate the relative enrichment of individual metabolites or mass features between treatments when compared to core metabolites. This is similar to normalizing mass feature area to biomass, a common normalization technique, but does not rely on a separate biomass measurement. Instead, the enrichment analysis calculates a scaled area for each compound or mass feature relative to each \u0026lsquo;core metabolite\u0026rsquo; to assess its enrichment under different treatment conditions. Core metabolites were decided based on previous work\u003csup\u003e6\u003c/sup\u003e that identified metabolites that follow biomass trends by depth and latitudinally and are observable in most phytoplankton when analyzed under our analytical conditions - these are reported in Table S9. These scaled areas are used to compute fold changes and perform statistical analyses. To account for the fact that some of the \u0026lsquo;core metabolites\u0026rsquo; may also be affected by treatments, we performed an outlier detection step, where Grubbs' test\u003csup\u003e71\u003c/sup\u003e is applied to identify extreme values in fold changes for core metabolites, and those core metabolites detected as outliers were excluded from further analysis. After filtering, the final enrichment results are summarized by calculating the median \u003cem\u003ep\u003c/em\u003e-values (after applying a Benjamini Hochberge false discovery rate correction\u003csup\u003e72\u003c/sup\u003e and fold changes for each metabolite across sample treatments.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003ePreparation, mass spectrometry, and analysis of metabolomics experiment with \u003c/u\u003e\u003c/em\u003e\u003cu\u003eRuegeria pomeryoi \u003cem\u003eDSS3 \u003c/em\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eCultures of \u003cem\u003eRuegeria pomeroyi \u003c/em\u003eDSS-3 were revived from cryostocks onto \u0026frac12; YTSS agar plates and incubated at 30 \u0026deg;C for 6 days. Single colonies were inoculated into 11 mL of glucose minimal media (GMM) and grown overnight at 30 \u0026deg;C with shaking at 200 rpm. GMM was prepared using a modified L1 minimal medium with glucose 12 mM C as the sole carbon source. All cultures were maintained in sterile 15 mL assay tubes. Overnight cultures were diluted to an optical density of 0.1 at 600 nm (OD₆₀₀) in fresh GMM, incubated for 11\u0026ndash;12 h, and amended with \u003cem\u003eglucose \u003c/em\u003e(4 mM C, 0.8 mM NH\u003csub\u003e4\u003c/sub\u003e, control), \u003cem\u003ehomarine \u003c/em\u003e(500 nM C, no additional NH\u003csub\u003e4\u003c/sub\u003e), or \u003cem\u003eglucose \u003c/em\u003e+ \u003cem\u003ehomarine\u003c/em\u003e (1 mM C, 2 mM C, respectively with 0.8 mM NH\u003csub\u003e4\u003c/sub\u003e). Homarine additions were staggered across time points to ensure consistent incubation durations. At each time point, samples were collected for cell counts and particulate metabolites. For cell counts, 1 mL of culture was fixed with glutaraldehyde (final concentration 1%) in labeled cryovials, held at 4 \u0026deg;C for 20 minutes, and then stored at \u0026ndash;80 \u0026deg;C. Particulate metabolites were collected by filtering cultures through combusted glass fiber filters using a vacuum manifold set to 8 psi; filters were wrapped in combusted foil and flash-frozen in liquid nitrogen. The experiment cultures were sampled after two hours in biological triplicate, as well as the three control conditions: glucose-only controls, glucose plus homarine controls (uninoculated), and a filter blank.\u003c/p\u003e\n\u003cp\u003eFor metabolite extractions, a one phase extraction was performed with 40:40:20:0.01 methanol:acetonitrile:water:formic acid solution as the extraction solvent\u003csup\u003e73\u003c/sup\u003e. Filters were placed in 15 mL Teflon tubes with pre-chilled extraction solvent, incubated at -20 \u0026ordm;C for 10 minutes, bead beaten with silica beads, and centrifuged. The solvent was then collected, transferred into glass tubes, and the procedure was repeated three times while keeping samples cold as much as possible. Samples were dried down under nitrogen gas, reconstituted in 400 uL of H2O, and stored at -80 \u0026ordm;C until analysis by LC-MS. Isotopically-labeled internal standards were added for normalization purposes, as reported in Table S17.\u003c/p\u003e\n\u003cp\u003eSamples were filtered, extracted, and run on the instrument as reported for the metabolomics experiment for \u003cem\u003eCobetia sp. \u003c/em\u003eOBi, but only particulate samples were acquired and run on the HILIC method for LCMS, as these data were the most promising for identification of intermediates of homarine based on the \u003cem\u003eCobetia sp. \u003c/em\u003eOBi experiment. We used Skyline\u003csup\u003e69\u003c/sup\u003e to mine the \u003cem\u003eRuegeria pomeryoi \u003c/em\u003eDSS-3 metabolomics data for features that showed enrichment in \u003cem\u003eCobetia sp. \u003c/em\u003eOBi under homarine conditions, and all the core metabolites, as reported in Table S9. These features underwent the same normalization and calculations for enrichment as the \u003cem\u003eCobetia sp. \u003c/em\u003eOBi data as reported above.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003eField experiments using isotopically labeled homarine\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eExperiments using \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-homarine were performed on research cruise TN397 in the Fall of 2021 at two different stations in the North Pacific (described in Table S8 and displayed in Figure 2). \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-homarine was purchased from Toronto Research Chemicals and was injected onto a Q-Exactive HF Orbitrap Mass Spectrometer (QE-HF) to confirm the mass of the deuterium label (141.0743 m/z) and the retention time (6.4 minutes, same as non-labeled homarine). Seawater was collected into 21 acid washed 10 L polycarbonate carboys from a trace metal clean stayfish system suspended at a depth of 8 m and prefiltered through 100 \u0026micro;m nylon mesh. Three unamended samples were collected immediately after seawater collection to provide samples of the starting community. Nine treatment bottles were spiked with 500 nM \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-homarine with nine control bottles receiving no additions. Bottles were incubated in blue-shaded temperature and light-controlled incubators designed to mimic mixed-layer conditions of the sampling location. Triplicate bottles with and without homarine addition were harvested at 2, 24, and 48 hours. All particulate samples (4 L) were collected using peristaltic pumps onto Durapore\u0026reg; 0.22 \u0026mu;m, 47 mm, hydrophilic PVDF membrane filters, flash frozen in liquid nitrogen, and stored at -80\u0026deg;C.\u003cbr /\u003e Additional stable isotope experiments with \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e7\u003c/sub\u003e-\u003csup\u003e15\u003c/sup\u003eN-labeled homarine were performed on two different cruises: TN412 (Winter 2023) in the North Pacific, and RC104 (Summer 2023) in Puget Sound (described in Table S8 and displayed in Figure 2). The preparation of the \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e7\u003c/sub\u003e-\u003csup\u003e15\u003c/sup\u003eN-labeled homarine is described below. For TN412, the experiments were performed at two different stations; for RC104, the experiment was performed at one station. For all TN412 and RC104 experiments, seawater was collected through a trace metal clean stayfish system suspended at a depth of 8 m prefiltered through 100 \u0026micro;m nylon mesh. Triplicate samples were collected into acid washed 2 L polycarbonate bottles, spiked with 90 nM of \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e7\u003c/sub\u003e-\u003csup\u003e15\u003c/sup\u003eN-labeled homarine, and incubated in temperature and light-controlled incubators for 5 different timepoints (6, 12, 24, 48 and 96 hours). Triplicates of spiked and unspiked samples were filtered as quickly as possible, no more than 30 minutes (T0, and unamended control samples, respectively). Particulate samples were collected identically as the TN397 experiment described above.\u003cbr /\u003e To prepare \u003csup\u003e13\u003c/sup\u003eC and \u003csup\u003e15\u003c/sup\u003eN-labeled homarine, \u003cem\u003eSynechococcus \u003c/em\u003esp. WH8102 was grown in Pro99 media\u003csup\u003e74\u003c/sup\u003e prepared with \u003csup\u003e15\u003c/sup\u003eN sodium nitrate and \u003csup\u003e13\u003c/sup\u003eC sodium bicarbonate and no HEPES buffer. \u003cem\u003eSynechococcus sp. \u003c/em\u003eWH8102 has been previously shown to produce a large amount of homarine\u003csup\u003e6\u003c/sup\u003e. Cultures were grown in 250 mL glass media bottles with minimal headspace. Cells were collected by filtration onto 0.2 \u0026micro;m Durapore filters using combusted borosilicate filter towers. All filters were stored at -80\u0026deg;C until extraction. Filters were extracted using 40:40:20:0.01 methanol:acetonitrile:water:formic acid solution as the extraction solvent in 15 mL Teflon tubes. The solvent was dried under N\u003csub\u003e2\u003c/sub\u003e gas, reconstituted with H\u003csub\u003e2\u003c/sub\u003eO, and injected onto a Q-Exactive HF Orbitrap Mass Spectrometer (QE-HF) to confirm the production of \u003csup\u003e3\u003c/sup\u003eC-\u003csup\u003e15\u003c/sup\u003eN-labeled homarine (146.0754 m/z).\u003cbr /\u003e After confirming the \u003csup\u003e13\u003c/sup\u003eC-\u003csup\u003e15\u003c/sup\u003eN-labeling of homarine, \u003cem\u003eSynechococcus \u003c/em\u003esp. WH8102 was grown in approximately 10 L, collected after 7 days, and extracted as previously described. Next, this extract was purified for spiking into environmental samples for stable isotope experiments. Purification was done using cation exchange chromatography and a Supelcosil LC-SCX column (25cm x 4.6mm, 5\u0026micro;m particle size) with a Thermo Scientific Vanquish UHPLC, fraction collector, and diode array detector. Solvent A was water with 2% formic acid and solvent B was water with 1% formic acid and 100 mM ammonium formate. The column was held at 0% B for 2 minutes, ramped to 20% B over 10 minutes, ramped to 100% B over 8 minutes, held at 100% B for 5 minutes, and equilibrated back to 0% B for 5 minutes (total run time is 30 minutes). The column temperature was maintained at 25\u0026deg;C and the flow rate was 1.0 mL/min. Purification of labeled homarine was achieved by UV based collection of a peak at 270 nm wavelength and a retention time of 4.3 minutes, which was confirmed by a homarine standard. Fractions containing labeled homarine were pooled together, dried under N\u003csub\u003e2\u003c/sub\u003e gas, and injected into the QE-HF to confirm the purity of the labeled homarine (any interfering compounds were confirmed to be less than 10%).\u003cbr /\u003e Samples were extracted, and data were acquired as reported for the metabolomics experiment for \u003cem\u003eCobetia \u003c/em\u003esp. OBi1 above. To prevent confusion using the isotope labels, we used a subset of isotopically-labeled internal standards, as reported in Table S17. Using the approach described above for the \u003cem\u003eRuegeria pomeryoi \u003c/em\u003eDSS-3 metabolomics data, we mined these data for the core metabolites as well as the features that showed enrichment in \u003cem\u003eCobetia sp. \u003c/em\u003eOBi under homarine conditions, both monoisotopic and expected isotopologues, as reported in Table S9. Again, these features underwent the same normalization and calculations for enrichment as the \u003cem\u003eCobetia sp. \u003c/em\u003eOBi data as reported above, using unamended control samples for comparison.\u003c/p\u003e\n\u003cp\u003eFor the TN397 with \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-homarine amendments, we also performed an untargeted analysis in an attempt to find additional features that would correspond to a methyl transfer. This approach is analogous to previous studies that utilized natural abundance isotopes to track sulfur and iron-containing metabolites\u003csup\u003e13,75\u003c/sup\u003e. We used CoreMS\u003csup\u003e76\u003c/sup\u003e to identify mass features using a persistent homology approach with liberal thresholds, yielding thousands of features per sample (of which many are expected to be false positives or very low abundance peaks). For each samples\u0026rsquo; features, we identified pairs of features that could correspond to \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e, \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e2\u003c/sub\u003e, and \u003csup\u003e13\u003c/sup\u003eC isotopologues - these pairs needed to be within 0.1 minutes of one another, with a mass error of expected isotopologues of no more than 4 ppm. From there, we collected features with putative \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e, \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e2 \u003c/sub\u003eisotopologues within all the samples into retention time and m/z groups and filtered out any features that were present in control samples or a single sample. We visually inspected these mass feature pairs and discarded pairs that did not have consistent peak shapes, resulting in 7 quality mass features (Table S10). Note that this approach will only yield features that have an observable level of the monoisotopic version of a compound in the samples.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003eIsolation, preparation, and genome sequencing of additional marine isolates\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAdditional homarine degrading bacteria were obtained by culturing bacteria in seawater samples on solid media made with sterile oligotrophic seawater, 1.5% ultrapure agarose (SeaKem), 12 mM carbon as homarine and 0.1 mM sodium phosphate. Seawater samples were collected at various nearshore locations in Puget Sound near Tacoma, Washington and in the equatorial North Pacific during Gradients IV cruise TN397. Bacteria were concentrated by vacuum filtration onto a 0.22 \u0026micro;m Supor membrane, resuspended in 1 mL of sterile oligotrophic seawater, plated on homarine challenge plates, and incubated 1-2 weeks at 25\u0026deg;C. Single colonies were re-streaked in marine broth agar plates to obtain axenic strains. The isolate collection was re-screened in oligotrophic seawater media containing homarine and phosphate to confirm growth on homarine. Two colonies per isolate were tested in 24-well plates containing 2 mL per well of media containing homarine or control media lacking homarine. Growth (OD600) was measured on a SpectraMax Plus microplate reader (Molecular Devices) after 4 and 8 days of incubation. Isolates with confirmed growth on homarine were regrown in 5 mL of marine broth medium for 1-2 days. Cultures were pelleted and stored at -20\u0026deg;C before DNA extraction. Bacterial cell pellets were rinsed in 5 mL of 0.9% saline solution by vortexing, 1 mL was transferred to sterile microtubes and centrifuged at 3500 x \u003cem\u003eg\u003c/em\u003e for 5 minutes and the supernatant was discarded. DNA was extracted with a Monarch\u0026reg; Genomic DNA Purification Kit using the tissue lysis solution supplemented with lysozyme followed by treatment with Proteinase K and RNase A. Purified DNA was analyzed on a Nanodrop spectrophotometer. Samples with \u0026gt;20 ng/\u0026micro;L were sent to SeqCenter (Pittsburgh, PA) for Illumina whole genome sequencing. Sample libraries were prepared using the Illumina DNA Prep kit and IDT 10bp UDI indices, and sequenced on an Illumina NextSeq 2000, producing 2x151bp reads. Illumina reads were trimmed and assembled with Trim Galore\u003csup\u003e77\u003c/sup\u003e, DOI:10.5281/zenodo.5127898; DOI:10.14806/ej.17.1.200) and the Unicycler SPAdes-optimiser\u003csup\u003e54\u003c/sup\u003e, respectively. We annotated assembly contigs with Prokka\u003csup\u003e56\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003eIdentification of the homarine operon and sox and mgd genes in isolates\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo identify genes involved in homarine catabolism, we conducted reciprocal best BLAST hit (RBH) analyses across isolate genomes. Protein sequences from all genomes were used to generate BLASTP databases (NCBI BLAST+), and an all-vs-all BLASTP search was performed using a parallelized script. Gene pairs were considered RBHs if they were each other\u0026rsquo;s top-scoring match by bitscore.\u003c/p\u003e\n\u003cp\u003eRBH results were screened for homologs of the \u003cem\u003ehomABCDER\u003c/em\u003e operon from \u003cem\u003eRuegeria pomeroyi\u003c/em\u003e DSS-3 (e.g., SPO3186\u0026ndash;SPO3192) and \u003cem\u003eCobetia\u003c/em\u003e\u003cem\u003esp\u003c/em\u003e. OBi1 (ACFLL8_11490\u0026ndash;ACFLL8_11520), as well as for subunits of sarcosine oxidase (\u003cem\u003esox\u003c/em\u003e) and \u003cem\u003eN\u003c/em\u003e-methylglutamate dehydrogenase (\u003cem\u003emgd\u003c/em\u003e). Reference \u003cem\u003esox \u003c/em\u003esubunits from \u003cem\u003eStenotrophomonas maltophilia\u003c/em\u003e (AAY34150.1\u003csup\u003e78\u003c/sup\u003e) and \u003cem\u003eCorynebacterium\u003c/em\u003e sp. (P40875.2\u003csup\u003e79\u003c/sup\u003e) and \u003cem\u003emgd\u003c/em\u003e subunits from \u003cem\u003eCobetia\u003c/em\u003e sp. OBi1 (ACFLL8_07290-07305), \u003cem\u003eR. pomeroyi\u003c/em\u003e DSS-3 (SPO1585-1588), \u003cem\u003eMethylocella silvestris\u003c/em\u003e (WP_008064110.1\u003csup\u003e40\u003c/sup\u003e) and \u003cem\u003eMethyloversatilis universalis \u003c/em\u003e(WP_012591629\u003csup\u003e41\u003c/sup\u003e)\u003c/p\u003e\n\u003cp\u003eFilter hits were concatenated into a non-redundant set, corresponding protein sequences were extracted from genome-wide protein files and annotated by gene identity and genome of origin. To guide alignment strategies, sequence variability was first assessed by calculating the average pairwise identity, average identities: \u003cem\u003esoxA/mgdC\u003c/em\u003e: 39.64%, \u003cem\u003esoxG/mgdD\u003c/em\u003e: 31.02%, \u003cem\u003esoxB/mgdA\u003c/em\u003e: 46.18%, and \u003cem\u003esoxD/mgdB\u003c/em\u003e: 34.62%. Due to the high sequence divergence observed across gene pairs, alignments were performed using MAFFT\u003csup\u003e80\u003c/sup\u003e with the E-INS-i algorithm and a BLOSUM62 matrix, to optimize for sequences with multiple conserved domains and long gaps. The alignments were subjected to model selection using ProtTest v3\u003csup\u003e81\u003c/sup\u003e. Best-fit substitution models were chosen based on AIC/BIC criteria and applied per gene pair as follows: \u003cem\u003esoxB\u003c/em\u003e/\u003cem\u003emgdA\u003c/em\u003e and \u003cem\u003esoxA\u003c/em\u003e/\u003cem\u003emgdC\u003c/em\u003e: LG+I+G, \u003cem\u003esoxD\u003c/em\u003e/\u003cem\u003emgdB\u003c/em\u003e: WAG+I+G, and \u003cem\u003esoxG\u003c/em\u003e/\u003cem\u003emgdD\u003c/em\u003e: WAG+G. Maximum likelihood (ML) phylogenetic trees were generated using RAxML with bootstrapping (1,000 replicates) on each gene pair alignment using the chosen substitution models, automated ML inference, bootstrap generation, and support value mapping. Tree diagrams were generated using iTol\u003csup\u003e82\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003eIdentification of the homarine operon in NCBI genomes\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted a broad survey to identify and characterize homarine operon orthologs. First, we downloaded and processed the NCBI ClusteredNR BLAST database\u003csup\u003e83\u003c/sup\u003e. We ran BLAST searches to identify homologs of the homarine catabolic genes (\u003cem\u003ehomA, homB, homC, homD, \u003c/em\u003eand\u003cem\u003e homE\u003c/em\u003e). We retrieved cluster members of relevant hits using the ClusteredNR scripts\u003csup\u003e84\u003c/sup\u003e and downloaded their corresponding nucleotide coordinates from NCBI with Entrez Direct\u003csup\u003e85\u003c/sup\u003e to identify candidate genomic loci with at least three of the six conserved genes co-located within 10,000 bp. To ensure completeness of our initial survey, we downloaded the genomes in which we found candidate homarine catabolic operons to perform a second round of BLAST searches and selected loci that contained at least one ortholog of \u003cem\u003ehomA\u003c/em\u003e, \u003cem\u003ehomB\u003c/em\u003e, \u003cem\u003ehomC\u003c/em\u003e, and \u003cem\u003ehomD \u003c/em\u003eco-located within 10,000 bp. The resulting genomic loci data are reported in Table S12.\u003c/p\u003e\n\u003cp\u003eWe retrieved BioSample metadata from NCBI for each of the genomes where we found at least \u003cem\u003ehomA\u003c/em\u003e, \u003cem\u003ehomB\u003c/em\u003e, \u003cem\u003ehomC\u003c/em\u003e, and \u003cem\u003ehomD \u003c/em\u003eco-located within 10,000 bp. Metadata were parsed and mined to retrieve key biosample attributes such as isolation source, environmental context, geographic location, and additional identifiers. Next, we cleaned and organized the biosample metadata for comparison. This involved categorizing isolation sources into environmental and non-environmental types and further classifying environmental sources by specific habitats; latitude and longitude data were parsed directly from metadata or inferred from reported geographic location. Note that many of the biosample records accompanying the NCBI genomes lacked sufficient metadata to parse the environmental context or geographical location of the source. Resulting mapping data is reported in Table S12.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003eEnvironmental transcriptomics perturbation experiment\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo identify microorganisms and genes responsive to homarine in Puget Sound, we conducted short ship-board incubations with seawater amended with homarine and collected total RNA for metatranscriptomics. Seawater was sampled with the RV Carson\u0026rsquo;s CTD from 4 m depth during RC-0078 (Station 2) in the Salish Sea near the Strait of Juan de Fuca on June 5, 2022. Seawater was transferred to acid-clean, blue-tinted, 5-Liter polycarbonate carboys for acclimation in a deck incubator with flow-through surface seawater. A set of three carboys were supplied with 0.1 mM carbon as homarine and three other unamended carboys served as control group. We incubated samples for 3.5 hours and proceeded to concentrate the microbial fraction by filtering each through a 47 mm-diameter, 0.22 \u0026micro;m pore-size polycarbonate membrane with a peristaltic pump. The filters were flash-frozen in liquid nitrogen at sea and were transferred to -80\u0026deg;C storage in the lab. Total RNA was extracted from filters using a ZymoBIOMICS RNA miniprep kit following the manufacturer protocol. RNA was quantified on a NanoDrop spectrophotometer. Samples with \u0026gt;50 ng/uL were sent to SeqCenter (Pittsburgh, PA, USA) for RNA-Seq. RNA samples were DNAse treated with Invitrogen DNAse (RNAse free). Library preparation was performed using Illumina\u0026rsquo;s Stranded Total RNA Prep Ligation with Ribo-Zero Plus kit and 10bp IDT for Illumina indices. Sequencing was done on a NovaSeq 6000 giving 2x51bp reads. Demultiplexing, quality control, and adapter trimming was performed with bcl-convert (v4.0.3). For metatranscriptomic analysis, we first parsed rRNA from non-rRNA reads with SortMeRNA. Non-rRNA reads (mRNA fragments) were assembled with PLASS\u003csup\u003e86\u003c/sup\u003e and mapped back to the 9,574,265 assembled proteins of length \u0026gt;100 amino acids with MMSeqs2\u003csup\u003e87\u003c/sup\u003e with the sensitivity parameter and maximum accepted targets set to 1. To quantify mapped reads, we retained only the best alignment for each unique read and reads with 100% identity to an assembled protein resulting in 42,155,143 reads and 1,367,930 proteins. We used DESeq2\u003csup\u003e58\u003c/sup\u003e and the tabulated read count data to detect differentially expressed genes between seawater samples supplied with homarine and unamended seawater. DESeq2 was implemented with a log\u003csub\u003e2\u003c/sub\u003e-fold change threshold of 1, the alternative hypothesis statement set to greater Abs, the significance level set to 0.05, and the default false discovery rate adjustment. We annotated 49 differentially expressed genes with the NCBI non-redundant database using BLASTP (excluding environmental samples\u003csup\u003e63\u003c/sup\u003e and the eggNOG 5.0.2 protein database using eggNOG-mapper v2.1.9\u003csup\u003e88,89\u003c/sup\u003e in ultra-sensitive mode.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eIn-situ\u003cem\u003e Homarine Concentration\u003c/em\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eParticulate and dissolved metabolites were sampled and measured as described in previous work\u003csup\u003e6,7\u003c/sup\u003e. Briefly, water was collected from the shipboard flow-through underway sampling system and particulate metabolites were sampled by filtering the seawater using peristaltic pumps onto 142 mm diameter, 0.2 um pore size PTFE Omnipore filters, flash frozen in liquid nitrogen, and stored at -80 C until analysis. Dissolved metabolites were sampled by collecting the filtrate in 50 mL acid washed polypropylene Falcon tubes. Particulate metabolites were extracted using a modified Bligh and Dyer extraction as described\u003csup\u003e67\u003c/sup\u003e. Dissolved metabolites were extracted using cation-exchange solid phase extraction as reported\u003csup\u003e7\u003c/sup\u003e. Following extraction, metabolites were dried down under N\u003csub\u003e2\u003c/sub\u003e gas, reconstituted in water with isotope labeled internal standards, and analyzed using liquid chromatography mass spectrometry using the same acquisition approach as described for \u003cem\u003eCobetia sp. \u003c/em\u003eOBi1 samples\u003cem\u003e.\u003c/em\u003e Homarine concentrations were quantified by comparison to a \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3\u003c/sub\u003e-homarine internal standard.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eIn-situ\u003cem\u003e Metatranscriptomes\u003c/em\u003e\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eMetatranscriptome samples were collected during the Gradients 3 cruise (KM1906) in the North Pacific in 2019. Samples were collected from the ship\u0026rsquo;s underway system by subsampling approximately 5-10 L of seawater into 10 L polycarbonate carboys, pre-filtered through a 100 \u0026mu;m mesh filter, and sequentially filtered through 3 \u0026mu;m and 0.2 \u0026mu;m polycarbonate filters. Filters were flash frozen in liquid nitrogen and stored at -80\u0026deg;C before processing. Metatranscriptomes were generated by extracting total RNA from the filters using the Direct-zol MiniPrep Plus kit (Zymo Research R2720). During extraction, a set of 14 synthetic spike-in mRNA standards was added as previously described\u003csup\u003e90,91\u003c/sup\u003e to enable quantification of absolute transcript abundances across samples. The extracted RNA was quantified with Qubit fluorometer 1.0 (ThermoFisher) and quality controlled using a Bioalanyzer 2100 (Agilent). Total community mRNA was generated by rRNA depletion using the ThermoFisher RiboMinus (Yeast+Bacteria; K15503, K15504) kit. Samples were sequenced on the Illumina NovaSeq6000 platform using S2 flow cell with 300 cycles. Library prep and sequencing was done at the Northwest Genomics Center (University of Washington, Seattle).\u003c/p\u003e\n\u003cp\u003eRaw Illumina reads from rRNA depleted samples were quality controlled with Trimmomatic v0.36 (MAXINFO:135:0.5, LEADING:3, TRAILING:3, MINLEN:60, and AVGQUAL:20)\u003csup\u003e92\u003c/sup\u003e. Short reads were assembled into longer contigs using Trinity de novo assembler v2.3.2 (--normalize_reads --min_kmer_cov 2 --min_contig_length 300)\u003csup\u003e93\u003c/sup\u003e. All assemblies were performed on short reads from combined replicates for each sample. Assembled contigs were translated into amino acid space in six frames with transeq\u003csup\u003e94\u003c/sup\u003e vEMBOSS:6.6.0.059 using Standard Genetic Code, and the longest coding frame was retained for future analysis. The best frame translated assemblies were clustered at 99% amino acid identity with MMseqs2\u003csup\u003e87\u003c/sup\u003e. To quantify assembled contig abundances and obtain raw transcript counts, unassembled short reads were mapped to the assembled contigs using kallisto v0.46.1\u003csup\u003e95\u003c/sup\u003e. The raw transcript counts for each assembled contig were normalized to total transcripts per liter using normalization factors based on the recovery of 14 synthetic internal mRNA\u003csup\u003e13,91,96\u003c/sup\u003e. Standard counts were recovered using the Bowtie 2 aligner v.2.5.2 using the standard pipeline for paired-end reads\u003csup\u003e90\u003c/sup\u003e. Taxonomic annotation of the assembled contigs was done with DIAMOND alignment to a combined reference database containing MarFeRReT v1.1\u003csup\u003e97\u003c/sup\u003e and MARMICRODB v.1.0\u003csup\u003e98\u003c/sup\u003e reference databases. Functional annotation of the assembled contigs was done using HMMER 3.3\u003csup\u003e99\u003c/sup\u003e against the Kyoto Encyclopedia of Genes and Genomes (KEGG) KOfam Hidden Markov Models (HMMs) database\u003csup\u003e26\u003c/sup\u003e with best KOfam predictions with a threshold score \u0026gt; 30 retained for each contig, and against Pfam 35.0\u003csup\u003e100\u003c/sup\u003e HMMs where the \u0026lsquo;trusteed cutoff\u0026rsquo; assigned by Pfam to each hmm profile was used as a minimum score threshold.\u003c/p\u003e\n\u003cp\u003eTo estimate abundances of selected HMM profiles in total community metatranscriptomes, assembled and annotated contigs of non-selected transcripts from prokaryotes and eukaryotes from the 3 \u0026mu;m and 0.2 \u0026mu;m fractions of each sample replicate were combined. Only contigs taxonomically annotated as \u0026ldquo;Bacteria\u0026rdquo; at the kingdom level were retained during the analysis.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cu\u003eHidden Markov Models profiles for genes in hom operon\u003c/u\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eHomarine catabolic protein sequences for each gene were aligned using MAFFT with a BLOSUM62 substitution matrix. Alignments were manually curated to remove poorly conserved N- and C-terminal regions and internal gaps. Hidden Markov Models (HMMs) were then built from the curated alignments using HMMER v3.3.2\u003csup\u003e101\u003c/sup\u003e. HMM profile specificity was tested against a curated genome database of isolates capable of growing in homarine with co-located \u003cem\u003ehom \u003c/em\u003ecatabolic genes. Profiles for \u003cem\u003ehomA\u003c/em\u003e and \u003cem\u003ehomB\u003c/em\u003e showed strong specificity with minimal off-target hits. However, \u003cem\u003ehomC\u003c/em\u003e returned several false positives, and \u003cem\u003ehomD\u003c/em\u003e produced numerous non-specific matches, rendering these profiles unsuitable for further analysis.\u003c/p\u003e\n\u003cp\u003eEach HMM profile was used to search for homologs in environmental metatranscriptomes collected on the Gradients 3 cruise using hmmsearch (HMMER v3.3.2) with the following parameters: --cpu 4 -T 30 --incT 30. The results were parsed and filtered, retaining only high-confidence hits based on alignment coverage, score, and contig completeness. Hits were manually curated against the reference HMM profile, and sequences that did not align properly or did not match the expected profile length were discarded. Finally, filtered gene hits were cross-referenced with sample metadata. We focused our analysis on the \u003cem\u003ehomB \u003c/em\u003eresults, which showed the highest significance (e-values ranging from 3.8E-43 to 3.9E-240), while \u003cem\u003ehomA \u003c/em\u003estill significant, had comparatively weaker support (e-values ranging from 1.5E-06 to 9.5E-57).\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eDATA AVAILABILITY\u003c/h2\u003e\n\u003cp\u003eAll datasets generated or analyzed during this study are publicly available or will be made available upon publication. Genome assemblies for \u003cem\u003eCobetia\u003c/em\u003e sp. OBi1 and additional homarine-degrading isolates, and transcriptomic data from \u003cem\u003eCobetia\u003c/em\u003e sp. OBi1 grown under homarine, glucose + homarine, and glucose-only conditions are available in NCBI BioProject PRJNA862506. Metabolomic datasets for \u003cem\u003eCobetia\u003c/em\u003e sp. OBi1 and \u003cem\u003eRuegeria pomeroyi\u003c/em\u003e DSS-3, including intracellular and dissolved profiles and standard-matched metabolites, are deposited to Metabolomics Workbench repository under study ID (TBD). Metabolomics data from environmental seawater incubations with \u003csup\u003e2\u003c/sup\u003eH\u003csub\u003e3 \u003c/sub\u003eor \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e7\u003c/sub\u003e\u003csup\u003e15\u003c/sup\u003eN-labeled homarine are deposited to Metabolomics Workbench repository under study ID (TBD). Environmental data from the Gradients North Pacific transect are available through Simons CMAP (https://simonscmap.com/catalog/datasets/Gradients3_KM1906_Optics_LISST_ACS_ECO). Environmental metatranscriptome data from the Gradients 3 North Pacific transect are available through NCBI SRA under BioProject PRJNA1091352. Environmental metabolomics data from the Gradients 3 North Pacific transect are available through Metabolomics Workbench repository under study ID TBD. Custom analysis scripts used in this study are available on GitHub at (https://github.com/IngallsLabUW/Homarine_Catabolism_MS).\u003c/p\u003e\n\u003ch2\u003eACKNOWLEDGEMENTS\u003c/h2\u003e\n\u003cp\u003eWe thank the captains, crews, and science parties of the \u003cem\u003eR/V Kilo Moana\u003c/em\u003e on cruise KM1906, of the \u003cem\u003eR/V Thomas G. Thompson \u003c/em\u003eon cruises TN397 and TN412, and \u003cem\u003eR/V Rachel Carson \u003c/em\u003eon cruises RC078 and RC104. We thank Lidimarie Trujillo Rodriguez and Susan Garcia for assistance with laboratory analyses; Sacha N. Coesel, Shiri Graff Van Creveld, Stephen Blaskowski, David Weidner, Ben Grodner, and Jeremy E. Schrier for valuable conversations and assistance with data analysis; Mary Ann Moran and Christopher R. Reisch for access to the \u003cem\u003eRuegeria pomeroyi \u003c/em\u003eDSS-3 arrayed BarSeq library. This work was supported by grants from the Simons Foundation (LS award ID: 385428, A.E.I.; SCOPE award ID 329108, A.E.I.; award ID 916384, B.P.D.), the National Science Foundation (OCE-2125886 to A.E.I. and K.R.H.; OCE-2124712 to O.A.S.), the University of Florida (Research Opportunity Seed Fund award ID 00131632 to B.P.D.), and the Center for Chemical Currencies of a Microbial Planet (C-CoMP; award ID 2019589 to F.X.F-G).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMoran, M. A. \u003cem\u003eet al.\u003c/em\u003e The Ocean\u0026rsquo;s labile DOC supply chain. \u003cem\u003eLimnol. Oceanogr. \u003c/em\u003e\u003cstrong\u003e67\u003c/strong\u003e, 1007\u0026ndash;1021 (2022).\u003c/li\u003e\n\u003cli\u003eSchalk, I. J. Bacterial siderophores: diversity, uptake pathways and applications. \u003cem\u003eNat. Rev. Microbiol. \u003c/em\u003e\u003cstrong\u003e23\u003c/strong\u003e, 24\u0026ndash;40 (2025).\u003c/li\u003e\n\u003cli\u003eSeymour, J. R., Sim\u0026oacute;, R., Ahmed, T. \u0026amp; Stocker, R. Chemoattraction to Dimethylsulfoniopropionate Throughout the Marine Microbial Food Web. \u003cem\u003eScience \u003c/em\u003e\u003cstrong\u003e329\u003c/strong\u003e, 342\u0026ndash;345 (2010).\u003c/li\u003e\n\u003cli\u003eHarvey, E. L. \u003cem\u003eet al.\u003c/em\u003e A Bacterial Quorum-Sensing Precursor Induces Mortality in the Marine Coccolithophore, \u003cem\u003eEmiliania huxleyi\u003c/em\u003e. \u003cem\u003eFront. Microbiol. \u003c/em\u003e\u003cstrong\u003e7\u003c/strong\u003e, (2016).\u003c/li\u003e\n\u003cli\u003eBoysen, A. K. \u003cem\u003eet al.\u003c/em\u003e Particulate Metabolites and Transcripts Reflect Diel Oscillations of Microbial Activity in the Surface Ocean. \u003cem\u003emSystems \u003c/em\u003e\u003cstrong\u003e6\u003c/strong\u003e, (2021).\u003c/li\u003e\n\u003cli\u003eHeal, K. R. \u003cem\u003eet al.\u003c/em\u003e Marine Community Metabolomes Carry Fingerprints of Phytoplankton Community Composition. \u003cem\u003emSystems \u003c/em\u003e\u003cstrong\u003e6\u003c/strong\u003e, (2021).\u003c/li\u003e\n\u003cli\u003eSacks, J. S., Heal, K. R., Boysen, A. K., Carlson, L. T. \u0026amp; Ingalls, A. E. Quantification of dissolved metabolites in environmental samples through cation‐exchange solid‐phase extraction paired with liquid chromatography\u0026ndash;mass spectrometry. \u003cem\u003eLimnol. Oceanogr. Methods \u003c/em\u003e\u003cstrong\u003e20\u003c/strong\u003e, 683\u0026ndash;700 (2022).\u003c/li\u003e\n\u003cli\u003eWidner, B., Kido Soule, M. C., Ferrer-Gonz\u0026aacute;lez, F. X., Moran, M. A. \u0026amp; Kujawinski, E. B. Quantification of Amine- and Alcohol-Containing Metabolites in Saline Samples Using Pre-extraction Benzoyl Chloride Derivatization and Ultrahigh Performance Liquid Chromatography Tandem Mass Spectrometry (UHPLC MS/MS). \u003cem\u003eAnal. Chem. \u003c/em\u003e\u003cstrong\u003e93\u003c/strong\u003e, 4809\u0026ndash;4817 (2021).\u003c/li\u003e\n\u003cli\u003eHolderman, N. R., Ferrer‐Gonz\u0026aacute;lez, F. X., Glushka, J., Moran, M. A. \u0026amp; Edison, A. S. Dissolved organic metabolite extraction from high‐salt media. \u003cem\u003eNMR Biomed. \u003c/em\u003e\u003cstrong\u003e36\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eHoppe-Seyler, F. A. \u0026Uuml;ber das Homarin, eine bisher unbekannte tierische Base. \u003cem\u003eZ. F\u0026uuml;r Physiol. Chem. \u003c/em\u003e\u003cstrong\u003e222\u003c/strong\u003e, 105\u0026ndash;115 (1933).\u003c/li\u003e\n\u003cli\u003eKumler, W. \u003cem\u003eet al.\u003c/em\u003e Metabolites reflect variability introduced by mesoscale eddies in the North Pacific Subtropical Gyre. \u003cem\u003eFront. Mar. Sci. \u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eSacks, J. S. \u003cem\u003eet al.\u003c/em\u003e The cycling of glycine betaine and homarine in marine microbial communities: quantitative flux measurements and the role of competitive uptake inhibition. \u003cem\u003eLimnol. Oceanogr. Lett.\u003c/em\u003e (In revision.).\u003c/li\u003e\n\u003cli\u003eDurham, B. P. \u003cem\u003eet al.\u003c/em\u003e Sulfonate-based networks between eukaryotic phytoplankton and heterotrophic bacteria in the surface ocean. \u003cem\u003eNat. Microbiol. \u003c/em\u003e\u003cstrong\u003e4\u003c/strong\u003e, 1706\u0026ndash;1715 (2019).\u003c/li\u003e\n\u003cli\u003eDawson, H. M. \u003cem\u003eet al.\u003c/em\u003e Microbial metabolomic responses to changes in temperature and salinity along the western Antarctic Peninsula. \u003cem\u003eISME J. \u003c/em\u003e\u003cstrong\u003e17\u003c/strong\u003e, 2035\u0026ndash;2046 (2023).\u003c/li\u003e\n\u003cli\u003eZhang, Y. \u003cem\u003eet al.\u003c/em\u003e Biodegradation of Picolinic Acid by \u003cem\u003eRhodococcus sp.\u003c/em\u003ePA18. \u003cem\u003eAppl. Sci. \u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 1006 (2019).\u003c/li\u003e\n\u003cli\u003eYoshida, T. \u0026amp; Nagasawa, T. Enzymatic functionalization of aromatic \u003cem\u003eN\u003c/em\u003e-heterocycles: Hydroxylation and carboxylation. \u003cem\u003eJ. Biosci. Bioeng. \u003c/em\u003e\u003cstrong\u003e89\u003c/strong\u003e, 111\u0026ndash;118 (2000).\u003c/li\u003e\n\u003cli\u003eQiu, J. \u003cem\u003eet al.\u003c/em\u003e A Novel Degradation Mechanism for Pyridine Derivatives in \u003cem\u003eAlcaligenes faecalis \u003c/em\u003eJQ135. \u003cem\u003eAppl. Environ. Microbiol. \u003c/em\u003e\u003cstrong\u003e84\u003c/strong\u003e, e00910-18 (2018).\u003c/li\u003e\n\u003cli\u003eRoney, S. H. \u003cem\u003eet al.\u003c/em\u003e Common fear molecules induce defensive responses in marine prey across trophic levels. \u003cem\u003eOecologia \u003c/em\u003e\u003cstrong\u003e202\u003c/strong\u003e, 655\u0026ndash;667 (2023).\u003c/li\u003e\n\u003cli\u003eMaldonado, A. \u003cem\u003eet al.\u003c/em\u003e Hard coral (\u003cem\u003ePorites lobata\u003c/em\u003e) extracts and homarine on cytochrome P450 expression in Hawaiian butterflyfishes with different feeding strategies. \u003cem\u003eComp. Biochem. Physiol. Part C Toxicol. Pharmacol. \u003c/em\u003e\u003cstrong\u003e179\u003c/strong\u003e, 57\u0026ndash;63 (2016).\u003c/li\u003e\n\u003cli\u003eMcClintock, J. B. \u003cem\u003eet al.\u003c/em\u003e Homarine as a feeding deterrent in common shallow-water antarctic lamellarian gastropod \u003cem\u003eMarseniopsis mollis\u003c/em\u003e: A rare example of chemical defense in a marine prosobranch. \u003cem\u003eJ. Chem. Ecol. \u003c/em\u003e\u003cstrong\u003e20\u003c/strong\u003e, 2539\u0026ndash;2549 (1994).\u003c/li\u003e\n\u003cli\u003eBerking, S. Homarine (\u003cem\u003eN\u003c/em\u003e-methylpicolinic acid) and trigonelline (\u003cem\u003eN\u003c/em\u003e-methylnicotinic acid) appear to be involved in pattern control in a marine hydroid. \u003cem\u003eDevelopment \u003c/em\u003e\u003cstrong\u003e99\u003c/strong\u003e, 211\u0026ndash;220 (1987).\u003c/li\u003e\n\u003cli\u003ePoulin, R. X. \u003cem\u003eet al.\u003c/em\u003e Chemical encoding of risk perception and predator detection among estuarine invertebrates. \u003cem\u003eProc. Natl. Acad. Sci. \u003c/em\u003e\u003cstrong\u003e115\u003c/strong\u003e, 662\u0026ndash;667 (2018).\u003c/li\u003e\n\u003cli\u003eRodriguez-R, L. M. \u003cem\u003eet al.\u003c/em\u003e The Microbial Genomes Atlas (MiGA) webserver: taxonomic and gene diversity analysis of Archaea and Bacteria at the whole genome level. \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e46\u003c/strong\u003e, W282\u0026ndash;W288 (2018).\u003c/li\u003e\n\u003cli\u003eSarkar, M. K., Paul, K. \u0026amp; Blair, D. Chemotaxis signaling protein CheY binds to the rotor protein FliN to control the direction of flagellar rotation in\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u003cem\u003eEscherichia coli\u003c/em\u003e. \u003cem\u003eProc. Natl. Acad. Sci. \u003c/em\u003e\u003cstrong\u003e107\u003c/strong\u003e, 9370\u0026ndash;9375 (2010).\u003c/li\u003e\n\u003cli\u003eSpohn, G. \u0026amp; Scarlato, V. Motility, Chemotaxis, and Flagella. in \u003cem\u003eHelicobacter pylori: Physiology and Genetics.\u003c/em\u003e 1\u0026ndash;23 (ASM Press, 2001).\u003c/li\u003e\n\u003cli\u003eKanehisa, M. \u0026amp; Goto, S. KEGG: Kyoto Encyclopedia of Genes and Genomes. \u003cstrong\u003e28\u003c/strong\u003e, 27\u0026ndash;30 (2000).\u003c/li\u003e\n\u003cli\u003eKanehisa, M. Toward understanding the origin and evolution of cellular organisms. \u003cem\u003eProtein Sci. \u003c/em\u003e\u003cstrong\u003e28\u003c/strong\u003e, 1947\u0026ndash;1951 (2019).\u003c/li\u003e\n\u003cli\u003eKanehisa, M., Furumichi, M., Sato, Y., Matsuura, Y. \u0026amp; Ishiguro-Watanabe, M. KEGG: biological systems database as a model of the real world. (2025).\u003c/li\u003e\n\u003cli\u003eBlum, M. \u003cem\u003eet al.\u003c/em\u003e InterPro: the protein sequence classification resource in 2025. \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e53\u003c/strong\u003e, D444\u0026ndash;D456 (2025).\u003c/li\u003e\n\u003cli\u003eJumper, J. \u003cem\u003eet al.\u003c/em\u003e Highly accurate protein structure prediction with AlphaFold. \u003cem\u003eNature \u003c/em\u003e\u003cstrong\u003e596\u003c/strong\u003e, 583\u0026ndash;589 (2021).\u003c/li\u003e\n\u003cli\u003eCianfrocco, M. A., Wong-Barnum, M., Youn, C., Wagner, R. \u0026amp; Leschziner, A. COSMIC2: A Science Gateway for Cryo-Electron Microscopy Structure Determination. in \u003cem\u003ePractice and Experience in Advanced Research Computing 2017: Sustainability, Success and Impact\u003c/em\u003e 1\u0026ndash;5 (ACM, New Orleans LA USA, 2017). doi:10.1145/3093338.3093390.\u003c/li\u003e\n\u003cli\u003eRoy, A., Yang, J. \u0026amp; Zhang, Y. COFACTOR: an accurate comparative algorithm for structure-based protein function annotation. \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e40\u003c/strong\u003e, W471\u0026ndash;W477 (2012).\u003c/li\u003e\n\u003cli\u003eZhang, C., Freddolino, L. \u0026amp; Zhang, Y. COFACTOR: improved protein function prediction by combining structure, sequence and protein\u0026ndash;protein interaction information. \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e45\u003c/strong\u003e, W291\u0026ndash;W299 (2017).\u003c/li\u003e\n\u003cli\u003ePettersen, E. F. \u003cem\u003eet al. \u003c/em\u003eUCSF ChimeraX : Structure visualization for researchers, educators, and developers. \u003cem\u003eProtein Sci. \u003c/em\u003e\u003cstrong\u003e30\u003c/strong\u003e, 70\u0026ndash;82 (2021).\u003c/li\u003e\n\u003cli\u003ePerchat, N. \u003cem\u003eet al.\u003c/em\u003e Elucidation of the trigonelline degradation pathway reveals previously undescribed enzymes and metabolites. \u003cem\u003eProc. Natl. Acad. Sci. \u003c/em\u003e\u003cstrong\u003e115\u003c/strong\u003e, (2018).\u003c/li\u003e\n\u003cli\u003eMejia, C. \u003cem\u003eet al.\u003c/em\u003e An Arrayed Transposon Library of \u003cem\u003eRuegeria pomeroyi\u003c/em\u003e DSS-3. Preprint at https://doi.org/10.1101/2022.09.11.507510 (2022).\u003c/li\u003e\n\u003cli\u003eB\u0026uuml;rgmann, H. \u003cem\u003eet al.\u003c/em\u003e Transcriptional response of \u003cem\u003eSilicibacter pomeroyi\u003c/em\u003e DSS‐3 to dimethylsulfoniopropionate (DMSP). \u003cem\u003eEnviron. Microbiol. \u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 2742\u0026ndash;2755 (2007).\u003c/li\u003e\n\u003cli\u003eNoell, S. E. \u0026amp; Giovannoni, S. J. SAR11 bacteria have a high affinity and multifunctional glycine betaine transporter. \u003cem\u003eEnviron. Microbiol. \u003c/em\u003e\u003cstrong\u003e21\u003c/strong\u003e, 2559\u0026ndash;2575 (2019).\u003c/li\u003e\n\u003cli\u003eJames C Netherton III, J. C. \u0026amp; Gurin, S. Biosynthesis and physiological role of homarine in marine shrimp. \u003cem\u003eJ. Biol. Chem. \u003c/em\u003e\u003cstrong\u003e257\u003c/strong\u003e, 11971\u0026ndash;11975 (1982).\u003c/li\u003e\n\u003cli\u003eChen, Y. \u003cem\u003eet al.\u003c/em\u003e \u0026gamma;-Glutamylmethylamide Is an Essential Intermediate in the Metabolism of Methylamine by \u003cem\u003eMethylocella silvestris\u003c/em\u003e. \u003cem\u003eAppl. Environ. Microbiol. \u003c/em\u003e\u003cstrong\u003e76\u003c/strong\u003e, 4530\u0026ndash;4537 (2010).\u003c/li\u003e\n\u003cli\u003eLatypova, E. \u003cem\u003eet al.\u003c/em\u003e Genetics of the glutamate‐mediated methylamine utilization pathway in the facultative methylotrophic beta‐proteobacterium \u003cem\u003eMethyloversatilis universalis\u003c/em\u003e FAM5. \u003cem\u003eMol. Microbiol. \u003c/em\u003e\u003cstrong\u003e75\u003c/strong\u003e, 426\u0026ndash;439 (2010).\u003c/li\u003e\n\u003cli\u003eHersh, L. B., Peterson, J. A. \u0026amp; Thompson, A. A. An N-methyl glutamate dehydrogenase from \u003cem\u003ePseudomonas\u003c/em\u003e M.A. \u003cem\u003eArch. Biochem. Biophys. \u003c/em\u003e\u003cstrong\u003e145\u003c/strong\u003e, 115\u0026ndash;120 (1971).\u003c/li\u003e\n\u003cli\u003eGiovannoni, S. J. SAR11 Bacteria: The Most Abundant Plankton in the Oceans. \u003cem\u003eAnnu. Rev. Mar. Sci. \u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 231\u0026ndash;255 (2017).\u003c/li\u003e\n\u003cli\u003eRapp\u0026eacute;, M. S., Kemp, P. F. \u0026amp; Giovannoni, S. J. Phylogenetic diversity of marine coastal picoplankton 16S rRNA genes cloned from the continental shelf off Cape Hatteras, North Carolina. \u003cem\u003eLimnol. Oceanogr. \u003c/em\u003e\u003cstrong\u003e42\u003c/strong\u003e, 811\u0026ndash;826 (1997).\u003c/li\u003e\n\u003cli\u003eSwan, B. K. \u003cem\u003eet al.\u003c/em\u003e Prevalent genome streamlining and latitudinal divergence of planktonic bacteria in the surface ocean. \u003cem\u003eProc. Natl. Acad. Sci. \u003c/em\u003e\u003cstrong\u003e110\u003c/strong\u003e, 11463\u0026ndash;11468 (2013).\u003c/li\u003e\n\u003cli\u003eStingl, U., Desiderio, R. A., Cho, J.-C., Vergin, K. L. \u0026amp; Giovannoni, S. J. The SAR92 Clade: an Abundant Coastal Clade of Culturable Marine Bacteria Possessing Proteorhodopsin. \u003cem\u003eAppl. Environ. Microbiol. \u003c/em\u003e\u003cstrong\u003e73\u003c/strong\u003e, 2290\u0026ndash;2296 (2007).\u003c/li\u003e\n\u003cli\u003eLuo, H. \u0026amp; Moran, M. A. Evolutionary Ecology of the Marine Roseobacter Clade. \u003cem\u003eMicrobiol. Mol. Biol. Rev. \u003c/em\u003e\u003cstrong\u003e78\u003c/strong\u003e, 573\u0026ndash;587 (2014).\u003c/li\u003e\n\u003cli\u003eSimon, M. \u003cem\u003eet al.\u003c/em\u003e Phylogenomics of \u003cem\u003eRhodobacteraceae\u003c/em\u003e reveals evolutionary adaptation to marine and non-marine habitats. \u003cem\u003eISME J. \u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, 1483\u0026ndash;1499 (2017).\u003c/li\u003e\n\u003cli\u003eGiovannoni, S. J. \u003cem\u003eet al.\u003c/em\u003e Genome Streamlining in a Cosmopolitan Oceanic Bacterium. \u003cem\u003eScience \u003c/em\u003e\u003cstrong\u003e309\u003c/strong\u003e, 1242\u0026ndash;1245 (2005).\u003c/li\u003e\n\u003cli\u003eHe, X.-Y. \u003cem\u003eet al.\u003c/em\u003e SAR92 clade bacteria are potentially important DMSP degraders and sources of climate-active gases in marine environments. \u003cem\u003emBio \u003c/em\u003e\u003cstrong\u003e14\u003c/strong\u003e, e01467-23 (2023).\u003c/li\u003e\n\u003cli\u003eMoran, M. A. \u003cem\u003eet al.\u003c/em\u003e Ecological Genomics of Marine Roseobacters. \u003cem\u003eAppl. Environ. Microbiol. \u003c/em\u003e\u003cstrong\u003e73\u003c/strong\u003e, 4559\u0026ndash;4569 (2007).\u003c/li\u003e\n\u003cli\u003eHoltgrewe, M., Messerschmidt, C., Nieminen, M. \u0026amp; Beule, D. DigestiFlow: from BCL to FASTQ with ease. \u003cem\u003eBioinformatics \u003c/em\u003e\u003cstrong\u003e36\u003c/strong\u003e, 1983\u0026ndash;1985 (2020).\u003c/li\u003e\n\u003cli\u003eBonenfant, Q., No\u0026eacute;, L. \u0026amp; Touzet, H. Porechop_ABI: discovering unknown adapters in Oxford Nanopore Technology sequencing reads for downstream trimming. \u003cem\u003eBioinforma. Adv. \u003c/em\u003e\u003cstrong\u003e3\u003c/strong\u003e, vbac085 (2023).\u003c/li\u003e\n\u003cli\u003eWick, R. R., Judd, L. M., Gorrie, C. L. \u0026amp; Holt, K. E. Unicycler: Resolving bacterial genome assemblies from short and long sequencing reads. \u003cem\u003ePLOS Comput. Biol. \u003c/em\u003e\u003cstrong\u003e13\u003c/strong\u003e, e1005595 (2017).\u003c/li\u003e\n\u003cli\u003eGurevich, A., Saveliev, V., Vyahhi, N. \u0026amp; Tesler, G. QUAST: quality assessment tool for genome assemblies. \u003cem\u003eBioinformatics \u003c/em\u003e\u003cstrong\u003e29\u003c/strong\u003e, 1072\u0026ndash;1075 (2013).\u003c/li\u003e\n\u003cli\u003eSeemann, T. Prokka: rapid prokaryotic genome annotation. \u003cem\u003eBioinformatics \u003c/em\u003e\u003cstrong\u003e30\u003c/strong\u003e, 2068\u0026ndash;2069 (2014).\u003c/li\u003e\n\u003cli\u003eLiao, Y., Smyth, G. K. \u0026amp; Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. \u003cem\u003eBioinformatics \u003c/em\u003e\u003cstrong\u003e30\u003c/strong\u003e, 923\u0026ndash;930 (2014).\u003c/li\u003e\n\u003cli\u003eLove, M. I., Huber, W. \u0026amp; Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. \u003cem\u003eGenome Biol. \u003c/em\u003e\u003cstrong\u003e15\u003c/strong\u003e, 550 (2014).\u003c/li\u003e\n\u003cli\u003eDehio, C. \u0026amp; Meyer, M. Maintenance of broad-host-range incompatibility group P and group Q plasmids and transposition of Tn5 in Bartonella henselae following conjugal plasmid transfer from Escherichia coli. \u003cem\u003eJ. Bacteriol. \u003c/em\u003e\u003cstrong\u003e179\u003c/strong\u003e, 538\u0026ndash;540 (1997).\u003c/li\u003e\n\u003cli\u003eThanbichler, M., Iniesta, A. A. \u0026amp; Shapiro, L. A comprehensive set of plasmids for vanillate- and xylose-inducible gene expression in Caulobacter crescentus. \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e35\u003c/strong\u003e, e137\u0026ndash;e137 (2007).\u003c/li\u003e\n\u003cli\u003eRied \u0026amp; Collmer. An nptI-sacR cartridge for constructing directed, unmarked mutations in Gram negative bacteria by marker exchange eviction mutagenesis. \u003cem\u003eGene \u003c/em\u003e\u003cstrong\u003e57\u003c/strong\u003e, 239\u0026ndash;246 (1987).\u003c/li\u003e\n\u003cli\u003ePearson, A. N. \u003cem\u003eet al.\u003c/em\u003e The pGinger Family of Expression Plasmids. \u003cem\u003eMicrobiol. Spectr. \u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, e00373-23 (2023).\u003c/li\u003e\n\u003cli\u003eCamacho, C. \u003cem\u003eet al.\u003c/em\u003e BLAST+: architecture and applications. \u003cem\u003eBMC Bioinformatics \u003c/em\u003e\u003cstrong\u003e10\u003c/strong\u003e, 421 (2009).\u003c/li\u003e\n\u003cli\u003eSchroer, W. F. \u003cem\u003eet al.\u003c/em\u003e Functional annotation and importance of marine bacterial transporters of plankton exometabolites. \u003cem\u003eISME Commun. \u003c/em\u003e\u003cstrong\u003e3\u003c/strong\u003e, 37 (2023).\u003c/li\u003e\n\u003cli\u003eBaumann, P. \u0026amp; Baumann, L. The Marine Gram-Negative Eubacteria: Genera Photobacteri in \u003cem\u003eBeneckea, Alteromonas, Pseudomonas, \u003c/em\u003eand \u003cem\u003eAlcaligenes\u003c/em\u003e. \u003cem\u003eThe Prokaryotes \u003c/em\u003e\u003cstrong\u003e2\u003c/strong\u003e, 1302\u0026ndash;1331 (1981).\u003c/li\u003e\n\u003cli\u003eGuillard, R. R. L. \u0026amp; Hargraves, P. E. \u003cem\u003eStichochrysis immobilis \u003c/em\u003eis a diatom, not a chrysophyte. \u003cem\u003ePhycologia \u003c/em\u003e\u003cstrong\u003e32\u003c/strong\u003e, 234\u0026ndash;236 (1992).\u003c/li\u003e\n\u003cli\u003eBoysen, A. K., Heal, K. R., Carlson, L. T. \u0026amp; Ingalls, A. E. Best-Matched Internal Standard Normalization in Liquid Chromatography\u0026ndash;Mass Spectrometry Metabolomics Applied to Environmental Samples. \u003cem\u003eAnal. Chem. \u003c/em\u003e\u003cstrong\u003e90\u003c/strong\u003e, 1363\u0026ndash;1369 (2018).\u003c/li\u003e\n\u003cli\u003eBoysen, A. K. \u003cem\u003eet al.\u003c/em\u003e Glycine betaine uptake and metabolism in marine microbial communities. \u003cem\u003eEnviron. Microbiol. \u003c/em\u003e\u003cstrong\u003e24\u003c/strong\u003e, 2380\u0026ndash;2403 (2022).\u003c/li\u003e\n\u003cli\u003eAdams, K. J. \u003cem\u003eet al.\u003c/em\u003e Skyline for Small Molecules: A Unifying Software Package for Quantitative Metabolomics. \u003cem\u003eJ. Proteome Res. \u003c/em\u003e\u003cstrong\u003e19\u003c/strong\u003e, 1447\u0026ndash;1458 (2020).\u003c/li\u003e\n\u003cli\u003eTsugawa, H. \u003cem\u003eet al.\u003c/em\u003e A lipidome atlas in MS-DIAL 4. \u003cem\u003eNat. Biotechnol. \u003c/em\u003e\u003cstrong\u003e38\u003c/strong\u003e, 1159\u0026ndash;1163 (2020).\u003c/li\u003e\n\u003cli\u003eGrubbs, F. E. Procedures for Detecting Outlying Observations in Samples. \u003cem\u003eTechnometrics \u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, 1\u0026ndash;21 (1969).\u003c/li\u003e\n\u003cli\u003eBenjamini, Y. \u0026amp; Hochberg, Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. \u003cem\u003eJ. R. Stat. Soc. Ser. B Methodol. \u003c/em\u003e\u003cstrong\u003e57\u003c/strong\u003e, 289\u0026ndash;300 (1995).\u003c/li\u003e\n\u003cli\u003eCanelas, A. B. \u003cem\u003eet al.\u003c/em\u003e Quantitative evaluation of intracellular metabolite extraction techniques for yeast metabolomics. \u003cem\u003eAnal. Chem. \u003c/em\u003e\u003cstrong\u003e81\u003c/strong\u003e, 7379\u0026ndash;7389 (2009).\u003c/li\u003e\n\u003cli\u003eDurham, B. P. \u003cem\u003eet al.\u003c/em\u003e Chemotaxonomic patterns in intracellular metabolites of marine microbial plankton. \u003cem\u003eFront. Mar. Sci. \u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 864796 (2022).\u003c/li\u003e\n\u003cli\u003eBoiteau, R. M. \u0026amp; Repeta, D. J. An extended siderophore suite from \u003cem\u003eSynechococcus sp.\u003c/em\u003e PCC 7002 revealed by LC-ICPMS-ESIMS. \u003cem\u003eMetallomics \u003c/em\u003e\u003cstrong\u003e7\u003c/strong\u003e, 877\u0026ndash;884 (2015).\u003c/li\u003e\n\u003cli\u003eCorilo, Y. E., Kew, W. R., McCue, L. A., Heal, K. \u0026amp; Carr, J. C. EMSL-Computing/CoreMS: CoreMS 3.0.0. Zenodo https://doi.org/10.5281/zenodo.14009575 (2024).\u003c/li\u003e\n\u003cli\u003eKrueger, F. Trim Galore!: A wrapper around Cutadapt and FastQC to consistently apply adapter and quality trimming to FastQ files, with extra functionality for RRBS data. \u003cem\u003eBabraham Inst.\u003c/em\u003e (2015).\u003c/li\u003e\n\u003cli\u003eHassan-Abdallah, A. \u003cem\u003eet al.\u003c/em\u003e Cloning, expression and crystallization of heterotetrameric sarcosine oxidase from \u003cem\u003ePseudomonas maltophilia\u003c/em\u003e. \u003cem\u003eProtein Expr. Purif. \u003c/em\u003e\u003cstrong\u003e43\u003c/strong\u003e, 33\u0026ndash;43 (2005).\u003c/li\u003e\n\u003cli\u003eSuzuki, M. Purification and Some Properties of Sarcosine Oxidase from \u003cem\u003eCorynebacterium sp. \u003c/em\u003eU-96. \u003cem\u003eJ. Biochem. (Tokyo) \u003c/em\u003e\u003cstrong\u003e89\u003c/strong\u003e, 599\u0026ndash;607 (1981).\u003c/li\u003e\n\u003cli\u003eKatoh, K. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e30\u003c/strong\u003e, 3059\u0026ndash;3066 (2002).\u003c/li\u003e\n\u003cli\u003eAbascal, F., Zardoya, R. \u0026amp; Posada, D. ProtTest: selection of best-fit models of protein evolution. \u003cem\u003eBioinformatics \u003c/em\u003e\u003cstrong\u003e21\u003c/strong\u003e, 2104\u0026ndash;2105 (2005).\u003c/li\u003e\n\u003cli\u003eLetunic, I. \u0026amp; Bork, P. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e49\u003c/strong\u003e, W293\u0026ndash;W296 (2021).\u003c/li\u003e\n\u003cli\u003eSayers, E. W. \u003cem\u003eet al.\u003c/em\u003e Database resources of the National Center for Biotechnology Information in 2023. \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e51\u003c/strong\u003e, D29\u0026ndash;D38 (2023).\u003c/li\u003e\n\u003cli\u003eCelebi, F. M., Eisen, J. A., Hochstrasser, M. L. \u0026amp; Reiter, T. Clustering the NCBI nr database to reduce database size and enable faster BLAST searches. (2023).\u003c/li\u003e\n\u003cli\u003eKans, J. Entrez direct: E-utilities on the UNIX command line. (2024).\u003c/li\u003e\n\u003cli\u003eSteinegger, M., Mirdita, M. \u0026amp; S\u0026ouml;ding, J. Protein-level assembly increases protein sequence recovery from metagenomic samples manyfold. \u003cem\u003eNat. Methods \u003c/em\u003e\u003cstrong\u003e16\u003c/strong\u003e, 603\u0026ndash;606 (2019).\u003c/li\u003e\n\u003cli\u003eSteinegger, M. \u0026amp; S\u0026ouml;ding, J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. \u003cem\u003eNat. Biotechnol. \u003c/em\u003e\u003cstrong\u003e35\u003c/strong\u003e, 1026\u0026ndash;1028 (2017).\u003c/li\u003e\n\u003cli\u003eCantalapiedra, C. P., Hern\u0026aacute;ndez-Plaza, A., Letunic, I., Bork, P. \u0026amp; Huerta-Cepas, J. eggNOG-mapper v2: Functional Annotation, Orthology Assignments, and Domain Prediction at the Metagenomic Scale. \u003cem\u003eMol. Biol. Evol. \u003c/em\u003e\u003cstrong\u003e38\u003c/strong\u003e, 5825\u0026ndash;5829 (2021).\u003c/li\u003e\n\u003cli\u003eHuerta-Cepas, J. \u003cem\u003eet al.\u003c/em\u003e eggNOG 5.0: a hierarchical, functionally and phylogenetically annotated orthology resource based on 5090 organisms and 2502 viruses. \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e47\u003c/strong\u003e, D309\u0026ndash;D314 (2019).\u003c/li\u003e\n\u003cli\u003eGroussman, R. D., Coesel, S. N., Durham, B. P., Schatz, M. J. \u0026amp; Armbrust, E. V. The North Pacific Eukaryotic Gene Catalog of metatranscriptome assemblies and annotations. \u003cem\u003eSci. Data \u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, 1161 (2024).\u003c/li\u003e\n\u003cli\u003eSatinsky, B. M., Gifford, S. M., Crump, B. C. \u0026amp; Moran, M. A. Use of Internal Standards for Quantitative Metatranscriptome and Metagenome Analysis. in \u003cem\u003eMethods in Enzymology\u003c/em\u003e vol. 531 237\u0026ndash;250 (Elsevier, 2013).\u003c/li\u003e\n\u003cli\u003eBolger, A. M., Lohse, M. \u0026amp; Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. \u003cem\u003eBioinformatics \u003c/em\u003e\u003cstrong\u003e30\u003c/strong\u003e, 2114\u0026ndash;2120 (2014).\u003c/li\u003e\n\u003cli\u003eGrabherr, M. G. \u003cem\u003eet al.\u003c/em\u003e Full-length transcriptome assembly from RNA-Seq data without a reference genome. \u003cem\u003eNat. Biotechnol. \u003c/em\u003e\u003cstrong\u003e29\u003c/strong\u003e, 644\u0026ndash;652 (2011).\u003c/li\u003e\n\u003cli\u003eRice, P., Longden, I. \u0026amp; Bleasby, A. EMBOSS: the European molecular biology open software suite. \u003cem\u003eTrends Genet. \u003c/em\u003e\u003cstrong\u003e16\u003c/strong\u003e, 276\u0026ndash;277 (2000).\u003c/li\u003e\n\u003cli\u003eBray, N. L., Pimentel, H., Melsted, P. \u0026amp; Pachter, L. Near-optimal probabilistic RNA-seq quantification. \u003cem\u003eNat. Biotechnol. \u003c/em\u003e\u003cstrong\u003e34\u003c/strong\u003e, 525\u0026ndash;527 (2016).\u003c/li\u003e\n\u003cli\u003ePark, J. \u003cem\u003eet al.\u003c/em\u003e Siderophore production and utilization by marine bacteria in the North Pacific Ocean. \u003cem\u003eLimnol. Oceanogr. \u003c/em\u003e\u003cstrong\u003e68\u003c/strong\u003e, 1636\u0026ndash;1653 (2023).\u003c/li\u003e\n\u003cli\u003eGroussman, R. D., Blaskowski, S., Coesel, S. N. \u0026amp; Armbrust, E. V. MarFERReT, an open-source, version-controlled reference library of marine microbial eukaryote functional genes. \u003cem\u003eSci. Data \u003c/em\u003e\u003cstrong\u003e10\u003c/strong\u003e, 926 (2023).\u003c/li\u003e\n\u003cli\u003eHogle, S. L. MARMICRODB database for taxonomic classification of (marine) metagenomes (1.0.0). https://doi.org/10.5281/zenodo.3520509 (2019).\u003c/li\u003e\n\u003cli\u003eEddy, S. R. Accelerated Profile HMM Searches. \u003cem\u003ePLoS Comput. Biol. \u003c/em\u003e\u003cstrong\u003e7\u003c/strong\u003e, e1002195 (2011).\u003c/li\u003e\n\u003cli\u003eMistry, J. \u003cem\u003eet al.\u003c/em\u003e Pfam: The protein families database in 2021. \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e49\u003c/strong\u003e, D412\u0026ndash;D419 (2021).\u003c/li\u003e\n\u003cli\u003ePotter, S. C. \u003cem\u003eet al.\u003c/em\u003e HMMER web server: 2018 update. \u003cem\u003eNucleic Acids Res. \u003c/em\u003e\u003cstrong\u003e46\u003c/strong\u003e, W200\u0026ndash;W204 (2018).\u003c/li\u003e\n\u003cli\u003eSchymanski, E. L. \u003cem\u003eet al.\u003c/em\u003e Identifying Small Molecules via High Resolution Mass Spectrometry: Communicating Confidence. \u003cem\u003eEnviron. Sci. Technol. \u003c/em\u003e\u003cstrong\u003e48\u003c/strong\u003e, 2097\u0026ndash;2098 (2014).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"homarine, catabolism, metabolites, marine microbial ecology, N-methylglutamic acid Running title: homarine catabolism in environmental bacteria","lastPublishedDoi":"10.21203/rs.3.rs-7359689/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7359689/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Homarine (N-methylpicolinic acid) is a ubiquitous marine metabolite produced by phytoplankton and noted for its bioactivity in marine animals, yet its microbial degradation pathways are uncharacterized. Here, we identify a conserved operon (homABCDER) that mediates homarine catabolism in bacteria using comparative transcriptomics, mutagenesis, and targeted knockouts. Phylogenetic and genomic analyses show this operon distributed across abundant bacterial clades, including coastal copiotrophs (e.g., Rhodobacterales) and open-ocean oligotrophs (e.g., SAR11, SAR116). High-resolution mass spectrometry revealed N-methylglutamic acid and glutamic acid as key metabolic products of homarine in both model and natural systems, with N-methylglutamate dehydrogenase catalyzing their conversion. Metatranscriptomics showed responsive and in situ expression of hom genes aligned with homarine availability. These findings uncover the genetic and metabolic basis of homarine degradation, establish its ecological relevance, and highlight homarine as a versatile growth substrate that feeds into central metabolism via glutamic acid in diverse marine bacteria.","manuscriptTitle":"Conserved pathway for homarine catabolism in environmental bacteria","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-21 03:46:30","doi":"10.21203/rs.3.rs-7359689/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-microbiology","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"nmicrobiol","sideBox":"Learn more about [Nature Microbiology](http://www.nature.com/nmicrobiol/)","snPcode":"","submissionUrl":"","title":"Nature Microbiology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d8367169-0a7d-410b-8659-ccc7c0351b56","owner":[],"postedDate":"August 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":53405309,"name":"Biological sciences/Chemical biology/Natural products"},{"id":53405310,"name":"Earth and environmental sciences/Ecology/Microbial ecology"},{"id":53405311,"name":"Earth and environmental sciences/Ecology/Biogeochemistry"},{"id":53405312,"name":"Biological sciences/Microbiology/Biogeochemistry"},{"id":53405313,"name":"Biological sciences/Biochemistry/Metabolomics"}],"tags":[],"updatedAt":"2026-03-31T07:14:31+00:00","versionOfRecord":{"articleIdentity":"rs-7359689","link":"https://doi.org/10.1038/s41564-026-02313-7","journal":{"identity":"nature-microbiology","isVorOnly":false,"title":"Nature Microbiology"},"publishedOn":"2026-03-30 04:00:00","publishedOnDateReadable":"March 30th, 2026"},"versionCreatedAt":"2025-08-21 03:46:30","video":"","vorDoi":"10.1038/s41564-026-02313-7","vorDoiUrl":"https://doi.org/10.1038/s41564-026-02313-7","workflowStages":[]},"version":"v1","identity":"rs-7359689","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7359689","identity":"rs-7359689","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}