Bacterial vitamin B6 is required for post-embryonic development in C. elegans

Nutritional intake influences animal growth, reproductive capacity, and survival of animals. Under nutrition deficiency, animal developmental arrest occurs as an adaptive strategy to survive. However, the nutritional basis and the underlying nutrient sensing mechanism essential for animal regrowth after developmental arrest remain to be explored. In Caenorhabditis elegans, larvae undergo early developmental arrest are stress resistant, and they require certain nutrients to recover postembryonic development. Here, we investigated the developmental arrest in C. elegans feeding on Lactiplantibacillus plantarum, and the rescue of the diapause state with trace supplementation of Escherichia coli. We performed a genome-wide screen using 3983 individual gene deletion E. coli mutants and identified E. coli genes that are indispensable for C. elegans larval growth on originally not nutritionally sufficient bacteria L. plantarum. Among these crucial genes, we confirmed E. coli pdxH, and the downstream metabolite pyridoxal 5-P (PLP, Vitamin B6) as important nutritional factors for C. elegans postembryonic development. Transcriptome results suggest that bacterial pdxH affects host development by coordinating host metabolic processes and PLP binding. Additionally, the developmental arrest induced by the L. plantarum diet in worm does not depend on the activation of FoxO/DAF-16. Altogether, these results highlight the role of microbial metabolite PLP as a crucial cofactor to restore postembryonic development in C. elegans.

https://doi.org/10.1038/s42003-024-05992-2Bacterial vitamin B6 is required for postembryonic development in C. elegans Check for updates Min Feng 1 , Baizhen Gao1 , Daniela Ruiz 1 , Luis Rene Garcia2 & Qing Sun 1 Nutritional intake influences animal growth, reproductive capacity, and survival of animals.Under nutrition deficiency, animal developmental arrest occurs as an adaptive strategy to survive.However, the nutritional basis and the underlying nutrient sensing mechanism essential for animal regrowth after developmental arrest remain to be explored.In Caenorhabditis elegans, larvae undergo early developmental arrest are stress resistant, and they require certain nutrients to recover postembryonic development.Here, we investigated the developmental arrest in C. elegans feeding on Lactiplantibacillus plantarum, and the rescue of the diapause state with trace supplementation of Escherichia coli.We performed a genome-wide screen using 3983 individual gene deletion E. coli mutants and identified E. coli genes that are indispensable for C. elegans larval growth on originally not nutritionally sufficient bacteria L. plantarum.Among these crucial genes, we confirmed E. coli pdxH, and the downstream metabolite pyridoxal 5-P (PLP, Vitamin B6) as important nutritional factors for C. elegans postembryonic development.Transcriptome results suggest that bacterial pdxH affects host development by coordinating host metabolic processes and PLP binding.Additionally, the developmental arrest induced by the L. plantarum diet in worm does not depend on the activation of FoxO/DAF-16.Altogether, these results highlight the role of microbial metabolite PLP as a crucial cofactor to restore postembryonic development in C. elegans.
A well-balanced diet contains macronutrients such as carbohydrates, fats, and proteins, as well as micronutrients such as vitamins and minerals to sustain basal metabolism and tissue function.Lack of certain nutrients leads to alterations in metabolism and causes severe deficiency in animal growth and reproduction.Therefore, understanding the role of individual nutrients is critical for revealing regulatory mechanisms of animal development.
The nematode Caenorhabditis elegans is a good model to study the effects of diet on life history traits such as development, reproduction, and aging 1 .As a bacterivore, C. elegans consumes a wide range of bacteria, including bacteria found in human microbiota such as E. coli.Multiple genetic engineering techniques have been developed to modify the genome of both C. elegans and its diet bacteria, making this model suitable for investigating the effects of bacteria-derived nutrients on host development 2 .Nutrient availability is a determinant of larval development in C. elegans.To survive unfavorable nutritional environments, the larvae can alter their developmental progression 3 .For example, in the absence of nutrients, newly hatched L1 larvae immediately arrest their postembryonic development 4 .Therefore, the C. elegans developmental arrest model provides an opportunity to study how the animal adapts to fluctuations in nutrient availability.Animals obtain essential nutrients not only from food intake but also from the synthesis of co-existing symbiosis microbiota.Previous studies emphasized the importance of metabolites derived from bacteria on larvae development in C. elegans.For example, vitamin B12, a metabolite of Comamonas aquatica D1877, accelerates host development via the methionine/S-Adenosylmethionine (SAM) cycle 5 .In addition, bacteriasecreted enterobactin facilitates mitochondrial iron uptake and promotes worm development by binding to the ATP synthase 6 .Additionally, vitamin B2 provided by bacteria regulates worm protease activities and development 7 .However, the effects of microbial metabolites/nutrients on postembryonic development in animals are missing, which is essential for understanding both the nutritional requirements and the intricate mechanism involved in animal growth.
In this study, an interspecies systems biology approach with C. elegans and its microbial diets, E. coli and Lactiplantibacillus plantarum, were used to identify bacterial metabolites that affect the animal's earlystage development and gene expression.L. plantarum has been shown to have probiotic effects in C. elegans when fed to adults [8][9][10] .We observed that hatched L1 C. elegans developmentally arrest or stalled when feeding on L. plantarum diet; however transient feeding with trace E. coli can promote larval development and further feeding on L. plantarum, which was originally not nutritionally sufficient.Through a systematic screen on ~4000 single gene deletion strains, we identified 29 E. coli genes that are essential for E. coli to promote C. elegans growth on an L. plantarum diet.The E. coli pdxH gene and the downstream metabolites vitamin B6 were further confirmed to coordinate host metabolic processes and PLPbinding activity thereby contributing to the host development.

L. plantarum induces developmental arrest in C. elegans
The nutritional composition of bacterial food can induce larval developmental delay in C. elegans.To investigate the effect of two different food sources, a commonly consumed probiotic Gram-positive bacterium L. plantarum and the standard laboratory food E. coli OP50, we fed synchronized L1 worms with L. plantarum and E. coli OP50 individually and measured worm development.Synchronized L1 worms were fed with L. plantarum or E. coli OP50.We found that ~95% of worms fed on OP50 developed to adults within 3 days at 21-22 °C, whereas worms fed solely on L. plantarum developmentally arrested at the early larval stage (Fig. 1a), suggesting that L. plantarum may not provide sufficient nutrients for C. elegans larvae development.To further investigate the food preference of C. elegans, two behavior assays to assess food dwelling and food choice assay were performed.Briefly, in the food-dwelling assay, we placed worms on a bacterial lawn and observed their location after a 24-h period.For the food choice assay, we set a plate with OP50 lawn on one side and L. plantarum lawn on the opposite side.The percentages of worms found on each bacterial lawn were monitored at specific time intervals to assess their food preference.The results showed worms tend to stay on OP50 as compared to the L. plantarum lawn (Fig. 1b and c), indicating L. plantarum is a less favorable food source.Interestingly, in the food choice assay, after OP50 was consumed on day 4, ~25% of worms switched to the L. plantarum lawn and consumed it (Fig. 1c), suggesting that worms can utilize L. plantarum food for growth if they previously ate OP50.
To further elucidate the stage at which developmental arrest occurred in the presence of L. plantarum, we assessed M lineage patterning as an indicator of worm development progression.In postembryonic development, a single mesodermal blast cell (M) undergoes division to produce a small number of additional mesodermal cells 11,12 .In hermaphrodites, the M divisions occurring in early larval development result in the formation of 14 striated body wall muscles, two sex myoblasts (SMs), and two coelomocytes 13 .By the L4 stage, the SMs undergo division, resulting in 16 SM descendants located near the vulval opening 11 .Here, we used hlh-8::gfp reporter expressed in the M-cell as the indicator.We observed that the GFP signal displayed the pattern characteristic of the stage between the late L1 stage to L2 stage in worms fed L. plantarum after 3 days, while worms fed OP50 exhibited a GFP signal pattern resembling the L4 stage (Fig. 1d).This observation strongly suggests that the development of worms fed with L. plantarum arrested or stalled at the late L1 stage to L2 stage.

L. plantarum-induced development arrest is not due to bacterial avoidance
As C. elegans pharyngeal pumping filters bacteria and transports them from the stoma to the isthmus 14 , we then asked if the size of L. plantarum (0.9-1.2 µm wide and 3-8 µm long 15 ) is too large for L1 worms to feed.In order to see if L. plantarum can pass the worm pharynx and reach the gut, we used a green fluorescent dye carboxyfluorescein diacetate succinimidyl ester (CFSE) to stain L. plantarum and fed the fluorescent L. plantarum to C. elegans.After 2 h of incubation, we found L1 worms fed CFSE stained L. plantarum had green fluorescence signal both in the pharynx and the gut (Fig. 2a), indicating that the dietary effect from L. plantarum on worm development was not due to deficiency of bacteria uptake.
Since starvation causes animal physiological and behavioral changes, we wondered whether these changes are responsible for the developmental arrest in worms fed on L. plantarum.To test this hypothesis, newly isolated eggs were placed on the L. plantarum lawn to avoid starvation.However, the worms stayed at the L1/L2 stage 5 days after being placed on the L. plantarum lawn (Fig. 2b), suggesting the developmental arrest induced by L. plantarum is not because the worm avoided L. plantarum after hatching.To explore whether this developmental arrest effect is reversible, OP50 was added to worms exposed to L. plantarum for 6 days.After adding OP50, developmental arrested worms recovered to adults with viable progenies within 3 days (Fig. 2c).This suggests that the nutrient deficiency of L. plantarum induced a reversible developmental arrest to the host, which is a similar protective response to starvation.We next asked if L. plantarum is also a nutrient-deficient food source for another typical development arrested stage, dauer larvae.To test this, dauer larvae were isolated and fed either OP50 or L. plantarum.We found that dauer worms grew normally into adults when fed L. plantarum food (Fig. 2d).These findings suggest that L. plantarum lacks certain nutrients that are essential for the growth of early larval stage worms, but it provides sufficient nutrients for exit from the dauer stage to adulthood.
A trace amount of E. coli can support the development of worms fed on L. plantarum Previous studies have shown that when larvae are provided with inadequate food, larval growth can be restored by supplementing a trace amount (0.2 µl, OD 600 = 1, 1.6 × 108 c.f.u.) of live E. coli 6 .To determine whether E. coli metabolites can compensate for some L. plantarum nutritional deficiency, L1 worms first were fed the same trace amount of live E. coli OP50 or BW25113.The optical refractivity of worms fed with 0.2 µl of bacterial culture changed so that they were easier to visualize, but they failed to develop substantial biomass (Fig. 3a).
Next, we fed the worms abundant L. plantarum together with trace live E. coli.We found that larval growth was recovered with the supplemented E. coli (Fig. 3b), and the progeny from those worms were also able to grow on L. plantarum, suggesting that factors from the trace amount of E. coli stimulate C. elegans surpass L1/L2 stage so they can use L. plantarum as nutrients.To characterize further the active compounds from E. coli, we supplemented L. plantarum with heat-treated, UV-treated E. coli or E. coli-conditioned media.We observed that while live E. coli promoted C. elegans growth of worms on L. plantarum, none of the above conditions were as effective (Fig. 3c).Thus, E. coli factors that rendered the L. plantarum edible are not secreted, heat nor UV stable.Increasing evidence suggests that various bioactive dietary factors may modify the epigenome and could thus be incorporated into an epigenetic diet 16 .Here we sought to determine whether a diet supplemented with L. plantarum could render the diet edible for their offspring.To test this, gravid worms were fed trace amounts of E. coli with an abundant supply of L. plantarum, then bleached to obtain L1 worms, which were subsequently placed on an L. plantarum lawn.As a control, L1 worms from mothers fed only E. coli were used.After 5 days on the plate, the worms remained in the early developmental stage, similar to the control group (Fig. 3d).These results suggest that potential epigenetic modifications resulting from the L. plantarum diet were insufficient to render the diet edible for their offspring.
A high-throughput screen of E. coli single gene deletions identifies the genes required for C. elegans to fed and develop on L. plantarum To identify systematically the bacterial components that act as nutrients to recover worm development, we performed a genome-wide screen using the Keio collection library.This library contains 3985 single gene knockout E. coli strains, with individual genes replaced with a kanamycin cassette in the K12 BW25113 background 17 .To conduct the screen, ~100 synchronized L1 worms were placed onto the plates with a trace amount of E. coli mutants from the Keio collection, together with L. plantarum (Fig. 4a).We then monitored the worms' development for 4-6 days.The wild-type parental strain E. coli BW25113 was used as the positive control.In the primary screen, we selected E. coli candidates that potentially delayed worm development.We performed a secondary screen with biological triplicates to narrow the candidates to 421 E. coli mutants (Fig. 4a).After a third screening, we confirmed 29 mutant E. coli strains that failed to support worm growth (Fig. 4b, Table 1).

E. coli vitamin B6 is essential for post-embryonic development in C. elegans
To classify the function of these 29 genes essential for recovering C. elegans early larval stage development, we performed a KEGG pathway enrichment analysis.We found that the genes are involved in a variety of metabolic pathways, including vitamin B6 metabolism, ferric enterobactin transports, biosynthesis of siderophore group non-ribosomal peptides, cysteine and methionine metabolism, pantothenate and CoA biosynthesis, and some unclassified pathways (Table 1).Out of these 29 genes, 5 of them are involved in the E. coli vitamin B6 biosynthesis pathway (Table 1).In contrast to eukaryotes, E. coli is capable of synthesizing the active form of vitamin B6, pyridoxal 5′-phosphate (PLP), through both de novo and salvage biosynthesis pathways (Fig. 5a).In order to investigate the role of these two pathways in regulating the growth of worms fed with L. plantarum, we fed the worms with various E. coli mutants and evaluated their growth by measuring worm size.Our results indicate that the development of worms arrested at early stage when fed with a trace amount of E. coli mutants with disrupted vitamin B6 de novo biosynthesis (Fig. 5b).This suggests that E. coli vitamin B6 de novo biosynthesis pathway plays an important role in recovering earlystage larval growth in the context of L. plantarum diet.To eliminate the possibility of contamination of these E. coli mutant strains, we performed PCR using gene-specific primers and verified the correct gene deletion in the knockout mutants (Supplementary Fig. 1).
To further investigate the E. coli metabolites involved in vitamin B6 metabolism-regulating arrested larvae development, the intermediate pyridoxine (PN) and final metabolite pyridoxal 5′-phosphate (PLP) were supplemented to the worms fed on trace E. coli pdxJ or E. coli pdxH mutants together with plenty of L. plantarum.We found the worm development completely recovered by dietary supplementation of the pdxH downstream metabolite PLP (Fig. 5c, Supplementary Fig. 2).Meanwhile, supplementing with intermediate PN rescued worm growth on pdxJ mutant but not pdxH mutant (Fig. 5d, Supplementary Fig. 2), indicating that the final product PLP is essential for C. elegans larval development.To determine whether the growth effect is restricted to a specific bacterial background, we added PLP only, PLP supplemented with the cofactor Nicotinamide, and PLP with trace amounts of UV-killed E. coli pdxH mutant to the worms that were fed on L. plantarum, individually.We found that supplementing 1 mM PLP without live pdxH mutant did not promote the worm growth (Fig. 5e), suggesting that the beneficial role of PLP for larval growth either requires some live E. coli molecules to deliver the PLP to C. elegans, or C. elegans is dependent on some downstream bacterial usage of PLP.As trace amounts of E. coli pdxH mutant cannot support the growth of worms fed on L. plantarum, we investigated whether a plentiful amount of the pdxH mutant alone could support worm development.We placed worms on an E. coli pdxH mutant lawn and used BW25113 lawn as the control.The results show that the lack of pdxH does not affect worm growth (Fig. 5f).Next, we explored whether other components from E. coli could compensate for the function of PLP and support the growth of worms fed on L. plantarum.To investigate this, we supplemented worms with an L. plantarum diet with varying concentrations of the pdxH mutant.We found that supplementation of no less than 5 µl of pdxH mutant (OD600 = 1) could partially recover the growth of the worms (Fig. 5g).This suggests that the function of pdxH can be compensated for by increasing the amount of the E. coli mutant.The pdxH mutant was grown in rich nutrient-containing yeast extract, which provided the vitamins required for the mutant's enzyme functions.Taken together, these results demonstrate that E. coli pdxH mutant provides limited nutrition, with vitamin B6 deriving from its prior culturing conditions, supporting worm growth.

Bacterial pdxH affects host development by coordinating host metabolic processes and PLP binding activity
To investigate the signaling mechanism that leads to the developmental arrest of C. elegans fed on L. plantarum due to the absence of pdxH in E. coli, we carried out RNA sequencing and expression analysis to identify the host pathways regulated by E. coli pdxH activity.Specifically, we compared the RNA-seq results of worms fed an L. plantarum diet with either a trace amount of E. coli pdxH mutant or wild-type E. coli BW25113.We identified 120 genes with increased expression and 830 genes with reduced expression with a Gfold (0.01) value (which indicates the fold change in gene expression) over 0.5 (Supplementary Data 1), suggesting widespread effects of the lack of pdxH in E. coli diet on host gene expression in C. elegans.Gene ontology (GO) analysis showed that upregulated genes were mainly involved in the metabolic process and oxidation-reduction process (Fig. 6a), while downregulated genes were mainly involved in development and growth (Fig. 6b), consistent with the developmental arrest phenotype.To gain insight into the crosstalk of vitamin B6 biosynthesis in E. coli with host vitamin B6 metabolism, we assessed the expression level of genes involved in  host vitamin B6 metabolism.The RNA-seq results showed fluctuated gene transcription in host PLP binding (Fig. 6c).Further qPCR analysis confirmed the relative mRNA level changes of two representative genes eppl-1 and F26H9.5, compared to the internal reference gene act-1 (Fig. 6d).These results suggest a fundamental role of E. coli vitamin B6 in regulating host PLP binding activity and restoring worm development with a diet of L. plantarum.

The developmental arrest induced by the L. plantarum diet in worms does not rely on FoxO/DAF-16 activation
The insulin-like signaling (IIS) pathway plays a pivotal role in regulating developmental arrest and gene expression in C. elegans when confronted with adverse conditions such as nutrient depletion and oxidative stress 18,19 .The FOXO transcription factor DAF-16, a downstream effector of the IIS pathway, is responsible for integrating stress signals and translocating from the cytoplasm to the nucleus 20,21 .To shed light on the participation of DAF-16 in worm developmental arrest mediated by L. plantarum, we investigated the subcellular localization of DAF-16 in worms exposed to various conditions, including those exclusively fed L. plantarum, L. plantarum in combination with 0.2 µl of BW25113, L. plantarum with 0.2 µl of the E. coli pdxH mutant, and L. plantarum with 0.2 µl of the E. coli pdxH mutant in addition to 1 mM PLP.We found that worms fed L. plantarum only or L. plantarum with 0.2 µl of the E. coli pdxH mutant exhibited a significant increase in the nuclear localization of DAF-16 after 4 days incubation when compared to worms fed L. plantarum with 0.2 µl of wild-type E. coli BW25113 (Fig. 6e).In addition, PLP supplementation reversed the DAF-16 nuclear translocation (Fig. 6e), indicating the nuclear translocation of DAF-16 in worms fed L. plantarum with 0.2 µl of the E. coli pdxH mutant is likely due to stress associated with deficiency in PLP or a decrease in a downstream PLP-regulated process.To further clarify the role of daf-16, we asked whether daf-16(mu86) null mutants developmentally arrest or can develop on L. plantarum.When this mutant was exclusively fed L. plantarum, the worms exhibited developmental arrest (Fig. 6f).This suggests that processes that lead to developmental arrest do not rely on daf-16 for activation.

Discussion
Animals require essential nutrients to support their development.Microbes, either through commensal interactions or as food sources, provide certain essential nutrients that cannot be synthesized by the animal.Our study shows that L. plantarum, when provided as a nutrient-deficient food source for C. elegans, can induce reversible developmental arrest.This finding enhances our understanding that larvae hatching in the presence of food but lacking essential nutrients fail to develop.Notably, a trace amount of E. coli enables worms to overcome this developmental arrest and complements the nutrient deficiency in L. plantarum.With this L. plantarum-induced developmental arrest model, we screened an E. coli single deletion library and found that the alteration of bacterial genetic composition influences C. elegans recovery from developmental arrest.As a result, the bacterial pdxH is not nutritionally adequate for the development of worms fed on L. plantarum.However, the supplementation of the pdxH downstream metabolite PLP compensates for the nutrient deficiency and recovers developmental arrested worm growth.These results emphasized the importance of coexisting E. coli acts synergistically with L. plantarum and provides essential nutrients to larval development in C. elegans.
Interestingly, our RNA-seq analysis revealed multiple PLP binding genes are decreased in larval C. elegans fed with L. plantarum (Fig. 6c), one of which is tatn-1 (Fig. 6c).tatn-1 gene encodes a tyrosine aminotransferase (TAT), previous studies have shown that TAT is a PLP-dependent enzyme catalyzes the conversion of tyrosine to 4-hydroxyphenylpyruvate, which is a rate-limiting step to metabolize tyrosine to fumarate and acetoacetate 22,23 .RNAi knockdown of tatn-1 gene expression can stall larval development 24,25 .Here, we hypothesize that the PLP from E. coli functions as a coenzyme of worm TAT, the absence of PLP from the dietary bacteria leads to a reduced expression of tatn-1 (Fig. 6c), subsequently leading to downstream deficiency in fumarate and acetoacetate or detrimental buildup of tyrosine and the developmental arrest.It is possible that E. coli is responsible for delivering vitamin B6 to C. elegans TATN-1, either through a bacterial conjugate (such as a PLP-binding bacterial protein) or it is equally possible that C. elegans does not directly require vitamin B6, but instead, it relies on specific rate-limiting molecules from E. coli that require vitamin B6 for their synthesis (Fig. 6g).
In most animals, including C elegans, the PLP biosynthetic pathway is dysfunctional, requiring them to obtain vitamin B6 from their diet or microbiome 26 .Some species of microbiota including Lactobacillus lack the ability to biosynthesize vitamin B6 27 , which makes synergistic in providing nutrients to the host important between microbes.Colonization of bacteria that has genomic diversity in vitamin B6 synthesis in C. elegans is beneficial to the health and life traits of the host 28 .In a chemically defined medium, C. elegans maintenance medium (CeMM), the active form of bacterial vitamin B6, PLP, is indispensable for worm population growth and health 29 .Lack of autonomous E. coli synthesis of PLP does not affect worm growth as the diet alone, provided sufficient live mutant bacteria is fed to the worms (Fig. 5f).This observation suggests that PLP-binding enzymes in living pdxH mutants can either (i) serve as a delivery vehicle for the vitamin to C. elegans enzymes, such as tyrosine aminotransferase, or (ii) vitamin B6 might be required for E. coli PLP-dependent enzymes to synthesize essential molecules, such as amino acids, that might be deficient in the context of a L. plantarum diet for larval C. elegans 30 .When pdxH mutant is cultivated in nutritionally rich LB broth (usually containing yeast extract), it likely provides vitamin B6 along with all of the amino acids that are synthesized by vitamin B6 binding enzymes.It's possible that the essential enzymes in the pdxH mutant diet are loaded with vitamin B6 from its yeast extract containing growth media.Consequently, the greater the quantity of pdxH mutant added to C. elegans, the more indirectly yeast-derived vitamin B6 is introduced, thereby supporting worm growth.The ability of organisms to sense their nutritional environment enables them to alter their growth and metabolism accordingly.Previous studies on larval arrest have shown that the insulin/insulin-like growth factor signaling (IIS) pathway plays a crucial role in sensing the nutritional environment and regulating entry into arrest 31 .The IIS pathway is the principal regulator linking nutrient levels to metabolism and development in C. elegans.DAF-2, which is an ortholog of the insulin/IGF-1 transmembrane receptor, plays a key role in regulating worm development 32 .DAF-16 is a widely expressed transcription factor belonging to the Forkhead family, it regulates the expression of various genes involved in development, longevity, and stress resistance 32 .DAF-2 functions primarily through the regulation of DAF-16 to control worm developmental arrest 33 .In this study, although DAF-16 does translocate into the nucleus upon L. plantarum feeding, it does not seem to be the primary cause of developmental arrest or stalling.The developmental defect from the L. plantarum diet is possibly due to a lack of rate-limiting factors crucial for robust developmental progression.DAF-16 might just promote a coping mechanism to sustain worm survival until an adequate amount of rate-limiting substance is acquired.This rate-limiting substance could potentially be vitamin B6 delivered via a bacteria-derived vehicle or other E. coli factors that rely on vitamin B6 for their synthesis.As we gain a deeper understanding of the regulatory network controlling postembryonic development, we may discover additional signals and signaling centers in the future.Given that Lactiplantibacillus is also a commensal bacterium in many other metazoans, including humans, investigating the interaction between Lactiplantibacillus and C. elegans may offer new insights into evolutionarily conserved probiotic effects.

Methods
Nematode and bacterial strains Wild type N2 Bristol, PD4667 hlh-8::gfp + dpy-20(+), TJ357 daf-16p::daf-16a/b::gfp + rol-6(su1006), and CF1037 daf-16(mu86) I was obtained from the Caenorhabditis Genetics Center (CGC) and maintained on nematode growth medium (NGM) 6 cm Petri dish plates at room temperature.All experiments were performed with synchronized L1 stage hermaphrodite animals.E. coli OP50 and the Keio collection parent strain E. coli BW25113 were grown at 37 °C in Lysogeny broth (LB) Miller's version.E. coli deletion strains were grown at 37 °C in LB with 50 µg/ml kanamycin.Lactiplantibacillus plantarum ATCC8014 was grown at 37 °C in Lactiplantibacillus MRS broth for 24 h.Bacterial cultures were seeded on NGM plates and dried for 1 h at room temperature prior to use in experiments.

CFSE staining
The green fluorescent dye 5(6)-Carboxyfluorescein diacetate succinimidyl ester (CFSE, MedChemExpress) was used to track the bacteria.1 ml of L. plantarum cultured 24 h was harvested and centrifuged at 4000 × g for 5 min.The resulting pellet was resuspended in 1 ml of 10 µM CFSE in PBS and incubated in the shaker at 37 °C at 200 rpm for 30 min while avoiding light.After staining, the bacteria were washed three times with PBS and resuspended in PBS.

Dauer larvae isolation
The N2 hermaphrodites were grown with OP50 at room temperature for ~10 days, during which there was an abundance of dauer larvae on the plate.The worms were then washed off the plate, and the worm pellet was resuspended in 1% SDS (sodium dodecyl sulfate).The worms were incubated in 1% SDS for 30 min with gentle agitation.Then, the worms were washed 1-5 times with M9 buffer to remove all SDS.To remove the carcasses of non-dauer larvae, 6 ml of 30% ice-cold sucrose was added to the worm pellet and then centrifuged at 4 °C for 5 min at a speed of 3000 rpm.The floating worms were aspirated, and the pellet was washed twice with M9 buffer to remove excess sucrose.To eliminate the potential existing OP50 in the gut, the isolated dauer worms were washed twice with 100 µg/ml gentamycin, each wash lasting 10 min.After washing three times with M9 to remove gentamycin, the worms were suspended in the M9 buffer.

E. coli Keio collection screen
The MRS broth-cultured L. plantarum was concentrated to OD600 = 10 and resuspended in an M9 buffer.Subsequently, 150 µl of L. plantarum suspension (1.2 × 10 8 c.f.u.) was added to one side of the peptone-free NGM plate.The E. coli mutants were cultured overnight at 37 °C in LB medium with 10 mg/ml kanamycin.Then, 0.2 µl of each mutant culture (OD600 = 1, 1.6 × 10 5 c.f.u.) was seeded to the other side of L. plantarum plate.Approximately 300 synchronized L1 worms were added to the screen plate and placed at room temperature.Then worm sizes were scored on days 3 and 4. Each mutant in the library was screened once for the primary screen.For the secondary screening, 421 candidate mutants were screened in triplicate to confirm the slow-growth phenotype.

PCR verification of deletions
Single gene deletion E. coli strains were streaked onto LB agar plates containing 50 µg/ml kanamycin.Subsequently, PCR was performed on individual colonies using genomic and kanamycin-cassette-specific primers.Genomic primers were designed specifically for each strain at a distance of 100-300 bases upstream of the start codon of the gene, and 100-300 bases downstream of the stop codon.The primers used are listed in Supplementary Table 1.The PCR products were then analyzed for the correct size by agarose gel electrophoresis.

Metabolites supplementation
Pyridoxine (PN, Fisher BioReagents, 1 mM) and pyridoxal 5′-phosphate (PLP, Thermo Scientific, 1 mM) were dissolved in water and added to E. coli mutants and then spotted onto the plates.
Total RNA extraction C. elegans at the same stage grown on L. plantarum together with a trace amount (0.2 µl, 1.6 × 10 5 c.f.u.) of wild-type BW25113 or pdxH mutant were collected from peptone-free NGM plates and washed three times with M9 buffer.Then 250 µl of TRIzol reagent (Invitrogen) was added to the worm elegans.a The vitamin B6 synthesis pathway in E. coli with genes essential for worm growth from the screening labeled in black.b Bar graph showing the development of synchronized L1 worms fed 0.2 µl E. coli mutants with disrupted vitamin B6 biosynthesis together with 150 µl L. plantarum.The volume of the worms was measured 3 days after the larvae were placed on the plates.Number of worms scored in each group are 72, 31, 25, 26, 16, 31, 37, 30.c The final metabolite pyridoxal 5′-phosphate (PLP) was supplemented to worms fed on 0.2 µl E. coli pdxJ or E. coli pdxH mutants together with 150 µl L. plantarum.The volume of the worms was measured 3 days after the larvae were placed on the plates.d The intermediate pyridoxine (PN) was supplemented to the worms fed on 0.2 µl E. coli pdxJ or E. coli pdxH mutants together with 150 µl L. plantarum.The volume of the worms was measured 3 days after the larvae were placed on the plates.e Worms were fed on L. plantarum together with PLP only, PLP supplemented with the cofactor Nicotinamide, and PLP with trace amounts of UV-killed E. coli pdxH mutant, respectively.f Worms were placed on an E. coli pdxH mutant lawn and a BW25113 lawn as the control.Worm growth was measured 2 days after L1 was placed on the plates.Number of worms scored in each group are 50 and 44.g Worms were fed with an L. plantarum diet with varying concentrations of the pdxH mutant.Worm growth was measured 4 days after L1 was placed on the plates.All data are representative of at least three independent experiments.Scale bar, 1000 μm.n = number of worms scored.Data are represented as mean ± SEM.Significance determined by unpaired two-tailed Student's t-test.* Indicates P-value < 0.05, ** indicates P-value < 0.01, N.S. indicates non-significant difference.pellet and worms were homogenized for 5 min.RNA was isolated by adding 50 µl of chloroform, followed by centrifugation to separate the aqueous phase.The aqueous phase was transferred to a new tube and mixed with isopropanol (125 µl) to precipitate RNA.The resulting RNA pellet was washed twice with 70% ethanol (250 µl).RNA pellets were air-dried and then re-suspended in RNase-free water.Any potential genomic DNA contamination was removed by DNaseI treatment using the TURBO DNAfree Kit (Invitrogen).
Reverse transcription-quantitative PCR (RT-qPCR) RT-qPCR was performed using the CFX Opus 96 Real-Time PCR System (Bio-rad).To begin, total RNA samples were reverse transcripted into firststrand cDNA using the iScript Select cDNA Synthesis kit (Bio-rad). 1 µl of a 10X diluted cDNA sample was then used in the qPCR reaction with SsoAdvanced Universal SYBR Green Supermix (Bio-rad).Primers were designed using Primer3 software (v4.1.0).Relative mRNA expression levels were calculated using the 2 −△△Ct method, and the reference gene act-1 was used to normalize the gene expression data.

RNA-seq
Purified RNA was quantified and assessed for quality using an Epoch 2 microplate spectrophotometer (BioTek).cDNA libraries were prepared from 500 ng RNA per sample.Libraries were then sequenced on an Illumina HiSeq 2000 sequencing machine using paired-end sequencing with a read length of 100 nucleotides.Adaptor sequences and low-quality reads were removed using Trimmomatic, and the RNA-Seq reads were then mapped to the C. elegans genome using STAR 2.5.3a with default settings.Transcript abundance, measured as read counts per gene, was extracted using HTSeq.Differential gene expression analysis was performed using GFold 34 .Gene ontology (GO) analysis was conducted using DAVID.

DAF-16 localization assay
The strain TJ356 (daf-16::gfp) was used to detect the intracellular DAF-16 localization.Synchronized L1 worms were transferred to peptone-free NGM plates covered with the lawns of plenty of L. plantarum only, L. plantarum with 0.2 µl of BW25113, L. plantarum with 0.2 µl of E. coli pdxH mutant, and L. plantarum with 0.2 µl of E. coli pdxH mutant as well as 1 mM PLP.The worms were placed at room temperature for 3 days, then collected and washed three times by M9 buffer.The worms were fixed on the 2% agarose pad of the slides, anesthetized with 0.25 mM levamisole, and covered with coverslips.The worms were imaged using a confocal scanning laser microscope.The percentages of DAF-16 localization (cytosolic, nuclear, and both) were calculated.

Statistics and reproducibility
The statistical analyses were conducted using GraphPad Prism and Excel.Number of worms scored is indicated as n.The results are presented as the mean ± SEM, and the data were evaluated using an unpaired twotailed Student's t-test.Data are represented as mean ± SEM. * Indicates P-value < 0.05, ** indicates P-value < 0.01, N.S. indicates non-significant difference.All data are representative of at least three independent experiments.

Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Fig. 1 |
Fig. 1 | L. plantarum induces early-stage larval arrest in C. elegans.a Microscope image and bar graph showing that worms fed with L. plantarum were arrested at the early larvae stage three days after hatching.Scale bar, 1000 μm.b Schematic drawing and quantitative data of the food dwelling assay.Circles indicate the bacterial lawn of OP50 (upper) and L. plantarum (bottom), respectively.The animals were scored 24 h after L1 worms were placed on the food spot.c Schematic drawing, microscope images and quantitative data of the food choice assay.Synchronized L1 worms were placed in the center spot (origin).OP50 (upper) and L. plantarum (bottom) bacteria were placed on opposite sides of the plate.The percentages of worms on each bacterial lawn were calculated at the indicated time.Worms moved to L. plantarum lawn and ate it after 4 days when OP50 was totally consumed.Scale bar, 1000 μm.d Synchronized L1 worms with hlh-8::gfp reporter expressed in the M-cell were fed with either L. plantarum or OP50 for 3 days.Fluorescent microscope images showing the GFP signal distribution.Scale bar, 50 μm.All data are representative of at least three independent experiments.n = number of worms scored.Data are represented as mean ± SEM.Significance determined by unpaired two-tailed Student's t-test.* indicates P-value < 0.05, ** indicates P-value < 0.01, N.S. indicates non-significant difference.

Fig. 2 |
Fig. 2 | L. plantarum-induced development arrest is not due to bacterial avoidance.a Microscope images showing the distribution of GFP signal in the newly hatched L1 worms fed either unstained L. plantarum or CFSE-labeled fluorescent L. plantarum, 2 h after feeding.Scale bar, 50 μm.b Microscope images and bar graph demonstrating that newly hatched worms fed on L. plantarum without starvation were arrested at the early larval stage three days after being placed on the bacteria.Scale bar, 1000 μm.c Worms grown on L. plantarum for 6 days can recover to adults with viable progeny 4 days after switching to OP50 food.Scale bar, 1000 μm.d Microscope images and bar graph showing that dauer larvae fed L. plantarum had a similar percentage of adult worms to those fed on OP50, 2 days after being placed on the bacteria.Scale bar, 1000 μm.All data are representative of at least three independent experiments.n = number of worms scored.Data are represented as mean ± SEM.Significance determined by unpaired two-tailed Student's t-test.* Indicates P-value < 0.05, ** indicates P-value < 0.01, N.S. indicates non-significant difference.

Fig. 3 |
Fig. 3 | A trace amount of E. coli can support the development of worms fed on L. plantarum.a Microscope image and bar graph showing that synchronized L1 worms fed on inadequate amounts of E. coli OP50 or BW25113 alone remained small.The volume of the worms was measured 4 days after the larvae were placed on the plates.Number of worms scored in the OP50 diet is 42, 32, 46, 46, 46, 46; Number of worms scored in the BW25113 diet are 46, 46, 40, 46, 46, 46.b Schematic drawing, microscope images, and bar graphs showing that a trace amount of E. coli OP50 supports the postembryonic growth of worms fed L. plantarum.The volume of the worms was measured 4 days after the larvae were placed on the plates.c Representative images and bar graph showing synchronized worms fed on 150 µl L. plantarum together with 0.2 µl live E. coli, 0.2 µl heat-killed E. coli, 0.2 µl UV-killed E. coli, or 150 µl E. coli conditioned media for 4 days.d L1 worms from mothers fed only E. coli or from mothers fed trace amounts of E. coli with an abundant supply of L. plantarum were placed on a L. plantarum lawn, respectively.Worm growth was measured 5 days after larvae were placed on the plates.All data are representative of at least three independent experiments.Scale bar, 1000 μm.n = number of worms scored.Data are represented as mean ± SEM.Significance determined by unpaired two-tailed Student's t-test.* Indicates P-value < 0.05, ** indicates P-value < 0.01, N.S. indicates non-significant difference.

Fig. 5 |
Fig. 5 | E. coli vitamin B6 is essential for post-embryonic development in C. elegans.a The vitamin B6 synthesis pathway in E. coli with genes essential for worm growth from the screening labeled in black.b Bar graph showing the development of synchronized L1 worms fed 0.2 µl E. coli mutants with disrupted vitamin B6 biosynthesis together with 150 µl L. plantarum.The volume of the worms was measured 3 days after the larvae were placed on the plates.Number of worms scored in each group are 72, 31, 25, 26, 16, 31, 37, 30.c The final metabolite pyridoxal 5′-phosphate (PLP) was supplemented to worms fed on 0.2 µl E. coli pdxJ or E. coli pdxH mutants together with 150 µl L. plantarum.The volume of the worms was measured 3 days after the larvae were placed on the plates.d The intermediate pyridoxine (PN) was supplemented to the worms fed on 0.2 µl E. coli pdxJ or E. coli pdxH mutants together with 150 µl L. plantarum.The volume of the worms was measured 3 days after the

Fig. 6 |
Fig. 6 | The function of E. coli pdxH relies on its ability to coordinate host metabolic processes and PLP-binding activity.a The most enriched gene ontology (GO) terms in upregulated genes in worms fed on E. coli pdxH vs WT E. coli together with 150 µl L. plantarum.b The most enriched gene ontology (GO) terms in downregulated genes in worms fed on 0.2 µl E. coli pdxH vs. WT E. coli together with 150 µl L. plantarum.c RNA-seq results reveal the expression levels of genes involved in host vitamin B6 metabolism.Down-regulated genes are indicated in blue, and upregulated genes are indicated in red.d qPCR analysis shows the relative changes in the mRNA level of two representative genes, eppl-1 and F26H9.5, compared to the internal reference gene act-1.e Bar graph demonstrating the activation of DAF-16::GFP nuclear translocation in worms fed L. plantarum only or L. plantarum together with 0.2 µl E. coli pdxH compared to the worms fed L. plantarum together with 0.2 µl E. coli pdxH or wt E. coli BW25113.f Representative images and bar graphs showing daf-16 null worms were fed either L. plantarum or OP50.Scale bar, 1000 μm.Data are presented as mean ± SEM in three independent experiments; Significance is determined by unpaired two-tailed Student's t-test.* Indicates Pvalue < 0.05, ** indicates P-value < 0.01, N.S. indicates non-significant difference.g A proposed model suggests that bacterial PLP acts as a cofactor of bacterial-derived molecules, promoting L. plantarum consumption and worm growth.

Table 1 |
Gene annotation of 29 identified E. coli genes from the screening