Neurodegenerative diseases are often associated with oxidative stress, and while probiotics may influence neuronal health, the underlying mechanisms remain poorly understood. Using the sod-1 A4V^M amyotrophic lateral sclerosis (ALS) model in Caenorhabditis elegans, we investigated the protective effects of the probiotic Enterococcus faecium against oxidative stress-induced neurodegeneration. Animals fed E. faecium showed reduced motor neuron degeneration under oxidative stress compared to those maintained on a standard Escherichia coli diet. Transcriptome analysis revealed a significant enrichment of oxidoreductase genes, including cytochrome P450 (cyp) genes. RNAi-mediated knockdown of cyp genes impaired E. faecium-mediated neuroprotection, and this loss correlated with increased reactive oxygen species (ROS) levels. We identified the conserved nuclear hormone receptor NHR-86 as a key regulator of cyp gene expression and neuroprotection. Loss of nhr-86 abolished the probiotic’s protective benefits, while transgenic expression of nhr-86 restored cyp induction and neuronal resilience. Importantly, intestinal expression of NHR-86 was sufficient to restore CYP induction and neuronal resilience, whereas neuronal knockdown had no effect, indicating that gut NHR-86 activity is essential for this protective pathway. These findings reveal a previously uncharacterized NHR-CYP regulatory axis activated by an intestinal probiotic, highlighting a mechanistic link between microbial signals and host neuroprotection.

E. faecium protects C. elegans against motor neuron degeneration under oxidative stress

The SOD-1 A4V mutation is the most prevalent disease-associated SOD-1 allele among ALS patients in North America [6]. Patients carrying this mutation typically experience selective degeneration and death of cholinergic spinal motor neurons, while glutamatergic neurons remain functional. Similarly, in C. elegans, the sod-1 A4V^M model exhibits cholinergic motor neuron loss under oxidative stress, with glutamatergic neurons unaffected. Moreover, single-copy sod-1 A4V^M knock-in animals in sod-1(-) background show significantly greater susceptibility to oxidative stress compared to sod-1 WT^M knock-in animals in the same genetic background [22], exhibiting an accelerated neurodegenerative phenotype, and providing a reliable model for detecting motor neuron degeneration. Under basal conditions, the single-copy SOD-1 A4V^M model does not show cholinergic motor neuron loss or functional defects in young adult animals, indicating that neurodegeneration in this model is primarily stress-dependent [22]. Therefore, we selected the ALS sod-1 A4V^M model to investigate the effects of probiotic E. faecium on the integrity of cholinergic motor neurons under oxidative stress, which was induced using paraquat [23]. Following paraquat exposure, some sod-1 A4V^M animals displayed defects in cholinergic motor neurons (Figs 1A and S1). To test whether E. faecium protects C. elegans against cholinergic motor neuron degeneration under oxidative stress, animals were pretreated with E. faecium for 24 hours before being transferred to paraquat plates (Fig 1B). Escherichia coli, a bacterial food for laboratory maintenance of C. elegans, was used as a control. The cholinergic motor neuron integrity was studied using GFP retention/loss in animals carrying the unc-17p::GFP gene, which expresses fluorescent proteins in cholinergic neurons. Less than 10 % of sod-1 WT^M animals showed motor neuron degeneration when pretreated with either E. coli or E. faecium, with no significant difference between the two treatments (Figs 1C, S1, and S2). Fifty percent of sod-1 A4V^M animals showed motor neuron degeneration when pretreated with E. coli, while only 20 % of sod-1 A4V^M animals pretreated with E. faecium showed motor neuron degeneration (Figs 1C and S2). Importantly, we found that both UV-killed and heat-killed E. faecium failed to provide neuroprotection (S3A and S3B Fig), indicating that live E. faecium is required for this protective effect.

We also assessed whether E. faecium-mediated neuroprotection promotes survival during oxidative stress. Animals treated with E. faecium survived significantly better than those treated with E. coli (Fig 1D). Since E. faecium protects motor neurons against neurodegeneration, we investigated whether E. faecium treatment affects locomotion. Animals were pretreated with E. faecium followed by paraquat-induced oxidative stress. E. faecium rescued the oxidative stress-induced locomotion defect (Fig 1E). These results indicate that probiotic E. faecium preserves cholinergic motor neuron integrity and locomotor function under oxidative stress.

Because ALS patients with the SOD1 A4V mutation exhibit loss of glutamatergic neurons, we next examined whether E. faecium protects these neurons. Using a sod-1 G85R^M model of glutamatergic neurodegeneration [22], we studied the sensory glutamatergic neurons PHA and PHB, which are susceptible to degeneration under paraquat-induced oxidative stress (S4A Fig). E. faecium treatment reduced degeneration in these neurons (S4B Fig), indicating that its neuroprotective effects extend beyond cholinergic motor neurons.

Classic immune pathways are not required in probiotic-mediated neuroprotection

To identify genes and pathways required in E. faecium-mediated neuroprotection, we performed transcriptome analysis comparing E. faecium-treated and E. coli-treated sod-1 A4V^M animals. Of the 302 differentially regulated genes, 227 were up-regulated, and 75 were down-regulated upon exposure to E. faecium (S1 Table). To identify related gene groups directly or indirectly regulated by E. faecium exposure, we performed an unbiased gene enrichment analysis using the Database for Annotation, Visualization, and Integrated Discovery (DAVID; http://david.abcc.ncifcrf.gov/) [24]. The top gene ontology (GO) clusters with the highest DAVID enrichment scores are shown in Fig 2A and 2B and S2 Table. For the down-regulated gene subset responding to E. faecium exposure, the most highly enriched clusters were related to innate immune response and transmembrane transport, followed by lipid transport, S-adenosylmethionine biosynthesis, and one-carbon metabolic process (Fig 2A). For the up-regulated gene set, the highest-scoring ontology clusters included oxidoreductase activity and hydrolase activity, followed by genes related to iron ion binding and structure constituent of the cuticle, among others (Fig 2B). As expected, a similar enrichment was also observed using a Wormbase enrichment analysis tool (https://wormbase.org/tools/enrichment/tea/tea.cgi) [25,26], which is specific for C. elegans gene data analysis. Innate immune response genes and oxidoreductase activity genes were also highly enriched among down-regulated and up-regulated genes, respectively (S5 Fig and S2 Table). Additionally, we compared our enrichment analysis with a similar study that examined gene expression differences between E. faecium and E. coli treated wild-type animals. We found comparable results, even though their comparison was between E. faecium and heat-killed E. coli (S2 Table) [27].

We validated gene expression using qRT-PCR, selecting candidate genes from two key clusters: the innate immune response cluster (top down-regulated) and oxidoreductase activity cluster (top up-regulated). These genes were assessed under four conditions: sod-1 WT^M and sod-1 A4V^M animals exposed to either E. coli or E. faecium. Innate immune genes were consistently down-regulated in an E. faecium–dependent manner (Fig 2C), while oxidoreductase activity genes were up-regulated (Fig 2D). The qRT-PCR results confirmed the RNA-seq findings, demonstrating a positive correlation and reinforcing that E. faecium specifically drives these gene expression changes.

Since the innate immune response cluster was among the most strongly down-regulated by E. faecium exposure, we hypothesized that activation of canonical immune pathways may contribute to motor neuron degeneration and that their suppression could be protective. To test this, we used RNAi to knock down classic immune pathways, including PMK-1 [28–30], DAF-16 [31–33], ELT-2 [34,35], and SKN-1 [36,37], in animals subjected to paraquat-induced oxidative stress in the presence and absence of E. faecium (Fig 3A). Because RNAi by feeding is largely ineffective in most neurons, these experiments primarily reflect the roles of these genes in the intestine and other non-neuronal tissues. As shown in Figs 3B and S6, RNAi of pmk-1, daf-16, elt-2, or skn-1 did not prevent motor neuron degeneration in animals fed E. coli and did not alter the neuroprotective effect of E. faecium treatment, indicating that modulation of these canonical immune pathways is not a determining factor in neuroprotection under oxidative stress.

Cytochrome P450 genes protect against motor neuron degeneration during oxidative stress

Our transcriptomic analysis of E. faecium-treated animals revealed a strong enrichment of oxidoreductase genes (Fig 2B), suggesting that these genes may play a role in neuroprotection. Notably, 10 out of the 23 genes in the oxidoreductase activity cluster belong to the cytochrome P450 (CYP) family (S2 Table), a group previously linked to neurodegenerative processes [17,38]. qRT-PCR analysis confirmed that multiple cyp genes, including cyp-35A3, cyp-35A1, cyp-35C1, cpy-33C8, cpy-34A9, cyp-35D1, cpy-13A7, cpy-33C9, cyp-35A5, and cyp-13A5, were up-regulated upon E. faecium exposure in sod-1 WT^M and sod-1 A4V^M animals (S7 Fig).

Some CYP enzymes contribute to ROS generation [39,40], while others help defense against oxidative stress [41–43]. To test whether CYPs are required in E. faecium-mediated neuroprotection, we knocked down several cyp genes up-regulated by E. faecium. Due to sequence similarity and redundancy, cyp-35A12345 can be knocked down by mixed RNAi, which targets all five genes [44]. Knock down of cyp-35A12345 decreased E. faecium-mediated protection against motor neuron degeneration (Figs 4A and S8). We also detected the survival of animals and locomotion with cyp-35A12345 knockdown and found that inhibition of cyp-35A12345 also decreased animal survival (Fig 4B) and locomotion (Fig 4C) protection conferred by E. faecium exposure.

To explore the mechanism of CYP-mediated neuroprotection during oxidative stress, we measured intracellular ROS levels using the fluorescent probes 2′,7′-dichlorofluorescein diacetate (H2DCFDA) [45]. Animals fed on E. coli followed by paraquat treatment showed high ROS levels, while E. faecium-treated animals showed low ROS levels when exposed to paraquat. Knocking down cyp-35A12345 decreased the reduction of ROS levels conferred by E. faecium exposure (Fig 4D and 4E). Together, these results demonstrate that CYPs are required in E. faecium-mediated neuroprotection and function, at least in part, by reducing intracellular ROS levels during oxidative stress.

NHR-86 is required in E. faecium-mediated neuroprotection

CYP enzymes are central to xenobiotic metabolism and are tightly regulated by NHRs in mammals and C. elegans [46–49]. In response to xenobiotic exposure, NHRs function as transcription factors that activate genes encoding metabolic enzymes and components of multi-drug efflux pumps [50]. To determine whether NHRs mediate E. faecium-induced expression of cyp genes, we first identified candidate nhr genes involved in this regulatory pathway.

A previous transcriptomic study found that both cyp and nhr genes were up-regulated in C. elegans exposed to genotoxic compounds, suggesting coordinated regulation of these gene families [48,51]. We performed a candidate mutant screen targeting 13 nhr genes responsible for regulating cyp genes (nhr-62, nhr-130, nhr-201, nhr-203, nhr-237, nhr-86, nhr-207, nhr-235, nhr-11, nhr-12, nhr-106, nhr-196, and nhr-205) using cyp-35A1, cyp-35A3, and cyp-35A5 as transcriptional readouts. Treatment with E. faecium in wild-type animals led to robust induction of cyp-35A1 expression. This induction was significantly reduced by RNAi targeting either nhr-203 or nhr-86 (Fig 5A), indicating that both nuclear receptors contribute to the regulation of cyp-35A1. In contrast, the expression of cyp-35A3 and cyp-35A5 in response to E. faecium depended solely on nhr-86, as their mRNA levels were unaffected by nhr-203 knockdown (S9 Fig).

We next tested whether nhr-86 or nhr-203 was required on E. faecium-mediated neuroprotection. Knocking down nhr-86, but not nhr-203, impaired E. faecium-mediated protection against motor neuron degeneration (Figs 5B and S10) and reduced the suppression of ROS levels (Figs 5C and S11). These findings were further supported by large particle flow cytometry, which showed that E. faecium-induced reductions in intracellular ROS were dependent on nhr-86 (Fig 5D and 5E), but not on nhr-203 (S12 Fig). Additionally, nhr-86 knockdown abolished the protective effects of E. faecium on oxidative stress-induced locomotion and survival (Fig 5F and 5G). Together, these data indicate that nhr-86, but not nhr-203, is essential for E. faecium-mediated neuroprotection and functions upstream of CYP gene activation, while nhr-86 expression itself is not regulated by E. faecium, as shown by RNA-seq (S1 Table) and qRT-PCR (S13 Fig).

To further assess the role of NHR-86 in the transcriptional program activated by E. faecium, we compared the E. faecium–regulated transcriptome with the published NHR-86 ChIP-seq dataset [52]. This analysis revealed an overlap between E. faecium-induced genes and NHR-86-bound targets, including CYP genes such as cyp-35A1 and cyp-35A5, as well as numerous other genes (S1 Table). These findings indicate that NHR-86 acts as one of several transcriptional effectors mediating the host response to E. faecium.

To confirm that nhr-86 is required for this protective pathway (Fig 5B–5G), we tested whether expression of nhr-86 under its endogenous promoter could rescue the reduced neuroprotection in sod-1 A4V^M animals. These animals were generated by crossing nhr-86(tm2590);nhr-86p::nhr-86 animals with sod-1 A4V^M animals (Fig 6A). Expression of NHR-86 fully restored E. faecium-mediated protection against cholinergic motor neuron degeneration in sod-1 A4V^M animals (Fig 6B), and rescued the mRNA levels of cyp-35A1, cyp-35A3, and cyp-35A5 upon E. faecium exposure in sod-1 A4V^M;nhr-86(tm2590) animals (Figs 6C and S14), and restored the protective effects of E. faecium on animal survival (Fig 6D–6F) and locomotion (Fig 6G). These results indicate that the expression of nhr-86 under its own promoter is sufficient to reestablish CYP activation and confer neuroprotection in response to E. faecium. Notably, overexpression of NHR-86 alone did not confer neuroprotection or induce cyp expression on an E. coli diet (Fig 6B and 6C), indicating that NHR-86 activation depends on specific signals provided by E. faecium.

The intestinal function of NHR-86 is required in E. faecium-mediated neuroprotection

NHR-86 is expressed in neurons, the intestine, and hypodermis [53]; however, because the CYP genes it regulates are primarily intestinal [54] and neurodegeneration occurs in neurons, we focused on its neuronal and intestinal roles.

To determine whether the neuronal NHR-86 function is required for neuroprotection during oxidative stress, we performed a neuron-specific knockdown of nhr-86 in sod-1 A4V^M animals. We generated a neuron-specific RNAi ALS model by crossing sod-1 A4V^M animals with the neuron-specific RNAi strain MAH677. Neuron-specific knockdown of nhr-86 had no effect on E. faecium-mediated neuroprotection (S15A and S15B Fig). dnc-1 served as a positive control, as its knockdown causes cholinergic motor neuron degeneration [55]. smn-1 was used as a negative control, as its neuron-specific RNAi affects only GABAergic motor neurons but does not impact cholinergic motor neurons [56]. Together, these results indicate that the neuronal function of NHR-86 is not required for the neuroprotective effect of E. faecium.

To determine whether the intestinal function of NHR-86 is required for neuroprotection during oxidative stress, we expressed NHR-86 specifically in the intestine of sod-1 A4V^M animals (Fig 7A). For intestine-specific expression, nhr-86 was placed under the control of vha-6 promoter, which is active in intestinal cells [57]. Intestine-specific expression of NHR-86 fully restored E. faecium-mediated protection against motor neuron degeneration in sod-1 A4V^M animals (Fig 7B), and rescued the mRNA levels of cyp-35A1, cyp-35A3, and cyp-35A5 in sod-1 A4V^M;nhr-86(tm2590) animals (Figs 7C and S16). It also restored the protective effect of E. faecium on animal survival (Figs 6D, 6E, and 7D) and locomotion (Fig 7E). Together, these results indicate that the intestinal expression of NHR-86 is sufficient to reestablish CYP activation and confer neuroprotection upon exposure to E. faecium.

Fig 8. Model of Enterococcus faecium-mediated neuroprotection under oxidative stress.

The intestinal probiotic E. faecium protects motor neurons from degeneration during paraquat-induced oxidative stress by inducing cytochrome P450 (CYP). This induction requires the nuclear hormone receptor NHR-86. CYP enzymes detoxify reactive oxygen species (ROS) generated during oxidative stress, reducing ROS burden and protecting motor neurons. This model highlights an NHR-CYP regulatory axis through which gut microbial signals enhance neuronal resilience under environmental stress. Figure created using BioRender. Sang, Y. (2026) https://BioRender.com/fg7i4lx.

https://doi.org/10.1371/journal.pbio.3003627.g008

Bacterial strains

The following bacterial strains were used: Escherichia coli OP50, E. coli HT115(DE3), Enterococcus faecium NCTC 7171. E. faecium was grown in brain–heart infusion (BHI) medium at 37 °C, and the cultures of other bacteria were grown in Luria-Bertani (LB) broth at 37 °C.

C. elegans strains

All C. elegans strains were maintained on nematode growth medium (NGM) and fed E. coli strain OP50. The C. elegans strains wild-type N2 Bristol, nhr-62(gk478), nhr-86(tm2590), nhr-130(gk710), nhr-235(gk1085), nhr-178(gk1005), nhr-178(gk1158), nhr-201(gk1231), nhr-203(gk660), nhr-207(gk632) and nhr-86(tm2590);nhr-86p::nhr-86(ORF)::GFP used were obtained from the Caenorhabditis Genetics Center (University of Minnesota, Minneapolis, MN). nhr-142(gk826), nhr-237(gk1050), nhr-11(tm1679), nhr-12(tm1038), nhr-106(tm1663), nhr-176(tm954), nhr-196(tm1042), and nhr-205(tm1185) animals were obtained from National Bioresource Project (NBRP), Japan.HA2630 sod-1 WT^M and HA2631 sod-1 A4V^M strain were gifts from Dr. Anne Hart [22]. Detailed strain information is listed in S3 Table.

Probiotic E. faecium pretreatment and paraquat exposure

Synchronized young adult animals were washed three times in M9 buffer and transferred to E. faecium lawns grown on 2% agar BHI plates for colonization for 1 day at 25 °C. The worms were then washed and transferred to paraquat agar plates containing OP50 lawn. A 250 mM stock solution of paraquat (Sigma-Aldrich 856177) was prepared in water and added to NGM media to achieve the 5 mM paraquat concentration. E. coli OP50 lawns were used as pretreatment control.

Fluorescent microscopy and cholinergic motor neurons integrity assessment

To assess the integrity of cholinergic motor neurons in C. elegans following paraquat exposure, we exposed HA2630 or HA2631 animals expressing unc-17p::GFP to 5 mM paraquat according to the specified protocol. After 24 hours, the animals were washed three times with M9 buffer, immobilized using 30 mM sodium azide (Sigma) in M9 buffer, mounted on a 2% agarose pad, and visualized using a Leica M165 FC fluorescence stereomicroscope. Defective cholinergic motor neurons were identified by scoring for the absence of at least two cholinergic motor neurons posterior to the vulva. All image acquisition and manual scoring were performed with researchers blinded to genotype and treatment.

Survival assay

Following 24 hours of paraquat exposure, the survival of sod-1 A4V^M animals pretreated on E. faecium or E. coli lawns was recorded. Time point 0 was defined as the completion of the 24-hour paraquat treatment. Animals were transferred to new paraquat plates daily until all animals stopped laying eggs, and survival was scored.

Locomotion phenotype analysis

An image-based assay was used to assess the locomotion speed of C. elegans. Animals were transferred to a plate with a fresh bacterial lawn to trace movement tracks. Upon release, worms exhibited a maximum movement response for a short duration. A 15-s movie was recorded for each worm, and a picture was taken at the 15th second. The ratio of movement distance to body length, measured using NIH ImageJ software, was used as a movement index [73]. Measurements were taken 48 hours after transferring to 5 mM paraquat.

Glutamatergic neuron degeneration

Day 1 adult animals, pretreated with E. faecium and subsequently exposed to paraquat, were washed off the plates with M9 buffer and incubated with DiD (Fisher DiIC18(5), D307) in a microcentrifuge tube. After 1 hour, animals were transferred to a regular NGM plate. After another hour, animals were immobilized with 20 mM sodium azide in M9 buffer, mounted on a 2% agar pad, and covered with a glass coverslip. Fluorescent neuronal cell bodies were visualized and scored for lack of dye uptake under a Leica M165 FC fluorescence microscope.

RNA interference

RNA interference (RNAi) was used to generate loss-of-function phenotypes by feeding nematodes with E. coli strain HT115(DE3) expressing double-stranded RNA homologous to a target gene [74,75]. E. coli with the appropriate vectors were grown overnight at 37 °C in LB broth containing ampicillin (100 mg/mL) and tetracycline (12.5 mg/mL), then plated onto NGM plates containing 100 mg/mL ampicillin and 3 mM isopropyl β-D-thiogalactoside (IPTG) (RNAi plates). RNAi-expressing bacteria were allowed to grow overnight at 37 °C. Gravid adults were transferred to RNAi-expressing bacterial lawns and allowed to lay eggs for 2 hours. The gravid adults were then removed, and the eggs were allowed to develop at 20 °C in young adults for subsequent assays. First-generation RNAi was performed from eggs to young adults for all genes, except for elt-2 RNAi, which was conducted after animals reached the L4 stage on nematode growth medium.unc-22 RNAi was included as a positive control to account for RNAi efficiency. All RNAi clones were from the Ahringer RNAi library. Note: Neuronal RNAi is inefficient by feeding; the results here indicate that classic immune pathways in non-neuronal tissues are not required for E. faecium-mediated neuroprotection.

RNA sequencing and analysis

Approximately 100 gravid animals were placed on 10-cm NGM plates seeded with E. coli OP50 for 3 hours to obtain a synchronized population, which developed to the young adult stage at 20 °C. The synchronized young adult worms were washed three times in M9 buffer and transferred to E. faecium lawns grown on 2% agar BHI plates for colonization for 24 hours at 25 °C. The animals were then washed off the plates with M9, frozen in QIAzol using ethanol/dry ice, and stored at −80 °C prior to RNA extraction. Total RNA was extracted using the RNeasy Plus Universal Kit (Qiagen, the Netherlands), and residual genomic DNA was removed using TURBO DNase (Life Technologies, Carlsbad, CA). A total of 6 μg of RNA was reverse-transcribed with random primers using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA).

Library construction and RNA sequencing on the BGISEQ-500 platform were performed according to established protocols, and paired-end reads of 100 bp were obtained for subsequent data analysis. The RNA sequence data were analyzed using a workflow constructed for Galaxy (https://usegalaxy.org). The RNA reads were aligned to the C. elegans genome (WS230) using the aligner STAR. Counts were normalized for sequencing depth and RNA composition across all samples. Differential gene expression analysis was performed using normalized samples, and genes exhibiting at least a 2-fold change were considered differentially expressed. The differentially expressed genes were subjected to SimpleMine tools from WormBase to obtain information such as WormBase IDs and gene names for further analyses. GO analysis was performed using the WormBase IDs in the DAVID Bioinformatics Database and a C. elegans data enrichment analysis tool.

RNA isolation and quantitative reverse transcription PCR (qRT-PCR)

Animals were synchronized, and total RNA was extracted as described. qRT-PCR was performed using the Applied Biosystems One-Step Real-Time PCR protocol with SYBR Green fluorescence on an Applied Biosystems 7900HT machine. Reactions were analyzed per the manufacturer’s instructions. Relative fold changes were calculated using the comparative CT (2^−ΔΔCT) method and normalized to pan-actin (act-1, -3, -4). Cycle thresholds were determined using StepOnePlus Real-Time PCR System Software v2.3. All results are based on three independent experiments, each performed in triplicate, with primer sequences listed in S4 Table.

Measurement of reactive oxygen species (ROS) in C. elegans

Worms were pretreated with E. faecium and exposed to paraquat, then subjected to ROS assays. ROS formation was measured using the fluorescent probes 2′,7′-dichlorofluorescein diacetate (H2DCFDA) (Sigma-Aldrich D6883). Animals with 24 hours E. faecium pretreatment and 24 hours of paraquat treatment were collected into 1 mL PBS, washed three times, and incubated with 50 μM H2DCFDA for 60 min at room temperature. Controls were incubated with DMSO. After washing three times with PBS, animals were immobilized using 30 mM sodium azide (Sigma) in M9 buffer, mounted on a 2% agarose pad, and visualized using a Leica M165 FC fluorescence stereomicroscope. DCF fluorescence intensity was measured at excitation/emission wavelengths of 488/510–560 nm using laser scanning confocal microscopy. The assay was performed in three independent trials with at least 25 animals per assay.

COPAS biosorter GFP analysis

DCF fluorescence intensity was analyzed using the COPAS Biosort flow cytometer (Union Biometrica). Worms pretreated with E. faecium and exposed to paraquat were washed three times and incubated with 50 μM H2DCF-DA before analysis. Fluorescence levels were measured using a 405-nm excitation laser and a 405-nm band-pass excitation filter with a neutral density of 0.6. The emission data were collected using a 512/25-nm band-pass emission filter (green channel). The acquired data were exported into Excel files and analyzed using Prism 9 (GraphPad).

Intestinal rescue of NHR-86

The vha-6 promoter and nhr-86 coding regions were PCR-amplified separately from Bristol N2 C. elegans genomic DNA using the primers listed in S4 Table. The two fragments were fused by overlapping PCR to generate the vha-6p::nhr-86 sequence. This sequence and the backbone plasmid pMS04 (Addgene #168172) were digested with the Kpn I and BgI II restriction enzymes. The fused vha-6p::nhr-86 fragment was then ligated into plasmid pMS04 between the Kpn I and BgI II sites to create the expression construct pMS04_vha-6p::nhr-86::mCherry. The final plasmid was purified and verified by sequencing. Young adult hermaphrodite sod-1(tm776) II; unc-119(+) III; rtSi008 [sod-1p::sod-1A4V M::sod-1 3’UTR + Cbr-unc-119(+)] IV; vsIs48 [unc-17p::GFP]; nhr-86(tm2590) animals were transformed by microinjection of plasmids into the gonads. A mixture containing the pMS04_vha-6p::nhr-86::mCherry plasmids (25 ng/µL) and pRF4_ rol-6(su1006) (25 ng/µL) as a transformation marker was injected into the worm. Successful transformation was determined by the identification of the selection marker as a dominant roller. At least three independent lines carrying extrachromosomal arrays were obtained for each construct.

Statistical analysis

Data were analyzed using a two-tailed Student t test for independent samples. For comparisons involving more than two groups, one-way ANOVA followed by post hoc analysis was employed. All experiments were conducted at least three times, with error bars representing the standard deviation unless otherwise noted. Statistical significance was defined as P < 0.05. The number of animals per experiment is denoted by “n”. “NS” indicates non-significant results, while asterisks denote significance levels: * P < 0.05, ** P < 0.01, *** P < 0.001. Survival fractions were calculated using the Kaplan–Meier method, and differences between survival curves were assessed with the log-rank test (S5 Table).

S1 Fig. Motor neuron degeneration under paraquat-induced oxidative stress.

Representative images show animals lacking (A) zero, (B) one, (C) two, and (D) more than 2 motor neurons posterior to the vulva. Animals missing at least two neurons were scored as defective. Missing neurons are marked with red circles.

https://doi.org/10.1371/journal.pbio.3003627.s001

(PDF)

S2 Fig. Motor neuron degeneration with or without Enterococcus faecium pretreatment.

Motor neuron degeneration in sod-1 WT^M and in sod-1 A4V^M animals was assessed under paraquat-induced oxidative stress, with or without E. faecium pretreatment. Animals missing at least two neurons were scored as defective. A two-tailed Student t test was performed to compare data between different treatments within the same group.

https://doi.org/10.1371/journal.pbio.3003627.s002

(PDF)

S3 Fig. Live Enterococcus faecium is required for neuroprotection.

(A) sod-1 A4V^M animals were pretreated for 24 hours with live, UV-killed, or heat-killed E. faecium before exposure to paraquat-induced oxidative stress. Cholinergic motor neuron integrity was assessed. Data are represented as mean ± SD from three independent experiments. Statistical significance was determined using one-way ANOVA with Tukey’s multiple comparison test (***p < 0.001). (B) Summary of motor neuron defects from (A) across three independent experiments.

https://doi.org/10.1371/journal.pbio.3003627.s003

(PDF)

S4 Fig. Oxidative stress-induced glutamatergic neuron degeneration in sod-1 G85R^C animals.

(A) Representative images of sod-1 G85R^C animals showing intact glutamatergic sensory neurons in the tail labeled with DiD. After paraquat treatment, PHA and PHB neurons fail to take up the dye, indicating degeneration or neuronal loss. Scale bar represents 10 μm. (B) Percentage of defective PHA and PHB neurons in sod-1 G85R^C ALS model animals after paraquat treatment. Results from three independent trials are shown. Error bars indicate ±SD. Statistical significance was determined using one-way ANOVA with Tukey’s multiple comparison test (***p < 0.001).

https://doi.org/10.1371/journal.pbio.3003627.s004

(PDF)

S6 Fig. Motor neuron degeneration under different RNAi knockdowns.

sod-1 A4V^M animals were fed control, pmk-1, elt-2, daf-16, or skn-1 RNAi. Motor neuron degeneration was assessed under paraquat-induced oxidative stress, with or without Enterococcus faecium pretreatment. Animals missing at least two neurons were scored as defective. A two-tailed Student t test was performed to compare data between different treatments within the same group.

https://doi.org/10.1371/journal.pbio.3003627.s006

(PDF)

S7 Fig. mRNA levels of cytochrome P450 genes.

qRT-PCR validation of cyp genes that were up-regulated with Enterococcus faecium pretreatment (N = 3 biological replicates). Values are expressed as the fold difference compared with sod-1 WT^M animals fed on Escherichia coli ± SD by one-way ANOVA with Tukey’s multiple comparison test.

https://doi.org/10.1371/journal.pbio.3003627.s007

(PDF)

S8 Fig. Motor neuron degeneration under cyp-35A12345 RNAi.

sod-1 A4V^M animals were fed control or cyp-35A12345 RNAi. Motor neuron degeneration was assessed under paraquat-induced oxidative stress, with or without Enterococcus faecium pretreatment. Animals missing at least two neurons were scored as defective. One-way ANOVA was performed to compare the data. P values: Empty vector, Ec-paraquat versus Empty vector, Ef-paraquat, P < 0.0001; Empty vector, Ec-paraquat versus cyp-35A12345 RNAi, Ec-paraquat, P = 0.9997; Empty vector, Ef-paraquat versus cyp-35A12345 RNAi, Ef-paraquat, P < 0.0001; cyp-35A12345 RNAi, Ec-paraquat versus cyp-35A12345 RNAi, Ef-paraquat, P = 0.0022.

https://doi.org/10.1371/journal.pbio.3003627.s008

(PDF)

S9 Fig. Screen of NHRs required for cyp activation by Enterococcus faecium.

(A) Screening of nhr genes using (A) cyp-35A3 and (B) cyp-35A5 as reporters. Young adult nhr mutant animals pretreated on E. faecium or Escherichia coli lawns were harvested for RNA extraction and qRT-PCR assay. (N = 3 biological replicates). Values represent fold differences relative to animals fed on E. coli, analyzed by one-way ANOVA with Tukey’s multiple comparisons test, * P < 0.05, ** P < 0.01, *** P < 0.001.

https://doi.org/10.1371/journal.pbio.3003627.s009

(PDF)

S10 Fig. Motor neuron degeneration under nhr-86 or nhr-203 RNAi.

sod-1 A4V^M animals were fed control, nhr-86 or nhr-203 RNAi. Motor neuron degeneration was assessed under paraquat-induced oxidative stress, with or without Enterococcus faecium pretreatment. Animals missing at least two neurons were scored as defective. One-way ANOVA was performed to compare the mean of nhr-86 RNAi and nhr-203 RNAi with the mean of empty vector pretreated with E. faecium.

https://doi.org/10.1371/journal.pbio.3003627.s010

(PDF)

S11 Fig. Quantification of DCF fluorescence intensity.

Intracellular ROS analysis of sod-1 A4V^M animals after cyp-35A12345 knockdown, following Enterococcus faecium pretreatment and paraquat exposure. n = 25. The “n” represents the number of animals in each experiment. N = 3 biological replicates.

https://doi.org/10.1371/journal.pbio.3003627.s011

(PDF)

S12 Fig. Intracellular ROS analysis of sod-1 A4V^M animals.

(A) Dot-plot representation of green fluorescence intensity versus TOF of empty vector and nhr-203 RNAi animals exposed to Enterococcus faecium or Escherichia coli followed by paraquat exposure (n = 300). (E) Frequency distribution of green fluorescence in vector or nhr-203 RNAi animals exposed to E. faecium or E. coli followed by paraquat exposure. Three independent experiments were conducted. Empty vector, Ec-paraquat versus Empty vector, Ef-paraquat, P = NS; nhr-203 RNAi, Ec-paraquat versus nhr-203 RNAi, Ef-paraquat, P = NS.

https://doi.org/10.1371/journal.pbio.3003627.s012

(PDF)

S13 Fig. mRNA levels of nhr-86.

qRT-PCR test nhr-86 expression Enterococcus faecium pretreatment (N = 3 biological replicates). Values are expressed as the fold difference compared with sod-1 WT^M animals fed on Escherichia coli ± SD by one-way ANOVA with Tukey’s multiple.

https://doi.org/10.1371/journal.pbio.3003627.s013

(PDF)

S14 Fig. mRNA levels of cyp genes with nhr-86 rescue.

mRNA level of (A) cyp-35A3 and (B) cyp-35A5 in sod-1 A4V^M, sod-1 A4V^M;nhr-86(tm2590), and sod-1 A4V^M;nhr-86(tm2590);nhr-86p::nhr-86 animals. (N = 3 biological replicates). Values are expressed as the fold difference compared with sod-1 A4V^M animals treated by Escherichia coli ± SD by one-way ANOVA with Tukey’s multiple comparisons test, * P < 0.05, ** P < 0.01, *** P < 0.001.

https://doi.org/10.1371/journal.pbio.3003627.s014

(PDF)

S15 Fig. Motor neuron degeneration under different RNAi knockdowns.

(A) Oxidative stress induced cholinergic motor neuron degeneration in sod-1 A4V^M animals. sod-1 A4V^M;sid-1(qt9); rgef-1p::GFP;rgef-1p::sid-1 animals (AA258) were fed control, nhr-86, dnc-1, or smn-1 RNAi. After growing to young adults on RNAi plates, the animals were treated with Enterococcus faecium or Escherichia coli for 24 hours before a 24-hour paraquat exposure. Cholinergic motor neuron integrity was accessed by scoring for the absence of at least two cholinergic motor neurons posterior to the vulva. * P < 0.05, ** P < 0.01, *** P < 0.001, two-way ANOVA. (B) Summary of motor neuron defects from (A) across three independent experiments.

https://doi.org/10.1371/journal.pbio.3003627.s015

(PDF)

S16 Fig. mRNA levels of cyp genes with nhr-86 rescue in intestine.

mRNA level of (A) cyp-35A3 and (B) cyp-35A5 in sod-1 A4V^M, sod-1 A4V^M;nhr-86(tm2590), and sod-1 A4V^M;nhr-86(tm2590);vha-6p::nhr-86 animals. (N = 3 biological replicates). Values are expressed as the fold difference compared with sod-1 A4V^M animals treated by Escherichia. coli ± SD by one-way ANOVA with Tukey’s multiple comparisons test, * P < 0.05, ** P < 0.01, *** P < 0.001.

https://doi.org/10.1371/journal.pbio.3003627.s016

(PDF)