Graphical Abstract graphic file with name fx1.jpg [53]Open in a new tab Highlights * • LELNs-manipulated probiotics protect mice from C. diff infection * • LELNs manipulation modulates gut metabolomics composition * • Cross talk between LGG and STH enhances production of lactic acid and AhR ligands __________________________________________________________________ Nanoparticles; Microbial Metabolism; Dairy Microbiology Introduction Clostridioides difficile (C. diff) is a spore-forming opportunistic pathogen that causes diarrhea and pseudomembranous colitis in humans ([54]Gerding et al., 1995). C. diff infection (CDI) is becoming a significant public health problem, especially in developed countries, associated with increasing morbidity, mortality, and medical costs ([55]Lucado et al., 2006; [56]Dubberke and Olsen, 2012). According to the reports of the Centers of Disease Control and Prevention, there are about 500,000 people who have CDI, which results in over 15,000 directly associated deaths per year in United States alone. Vancomycin and metronidazole are the most used antibiotics to treat CDI, with about 20%–30% recurrence after the antibiotic treatment is stopped ([57]Stevens et al., 2017; [58]Balsells et al., 2019). Fecal microbiota transplant (FMT) has proved to be one of most efficient methods to treat CDI and to also reduce the recurrence of CDI ([59]Bakken et al., 2011). FMT carries increasing risks, including the risk of transmitting other infections such as HIV, hepatitis, and retrovirus; the risk of transferring superbugs with antibiotic resistance; and the risk of microbiota functional divergence in different individuals due to complex and varying composition of fecal microbiota ([60]Wang et al., 2016). All these risks highlight the importance of developing probiotic treatments for CDI. The concept of promoting human health through consumption of beneficial microorganisms has evolved during the last century ([61]George Kerry et al., 2018). Numerous studies have been published under this broad definition, attempting to demonstrate beneficial effects of probiotics to CDI, however, with highly mixed results ([62]Na and Kelly, 2011; [63]Shen et al., 2017; [64]Box et al., 2018). The considerable variation in outcomes between clinical trials may be attributed to multiple factors, including differences in the bacterial strains used among researchers and the survivability of the studied microorganisms in different individuals' gut microenvironment. Therefore, we urgently need to have an alternative approach to develop generalized and efficient probiotics for the prevention and treatment of CDI. Our diet includes a variety of plants, which contain a large quantity of exosome-like nanoparticles (ELNs) of different compositions ([65]Yang et al., 2018; [66]Iravani and Varma, 2019). ELNs play critical roles in inter-bacterial communication ([67]Teng et al., 2018; [68]Sundaram et al., 2019). Therefore the manipulation of probiotics with unlimited resources of ELNs is likely to provide an opportunity to choose personalized ELNs for the optimization of probiotic therapies in general. In this study, as proof of concept, we focused on the critical issue of CDI treatments with current commercially available probiotics formulated in yogurt using in vivo and in vitro experiments. Our focus and design directly characterized the effects of the manipulation of yogurt-derived probiotics with lemon exosome-like nanoparticles (LELNs) on the protection of mice against CDI. In addition, we further explored the cellular and molecular mechanism underlying the effects. Results LELNs Protect Mice from CDI by Enhancing the Survivability of Probiotics Probiotic benefits are prone to be dose dependent ([69]Bertazzoni et al., 2013), and our previous research showed that edible plant-derived nanoparticles are potent factors for increasing beneficial bacteria numbers in the gut ([70]Teng et al., 2018). Here, using LELNs as proof of concept, we developed LELNs that manipulated probiotics in such a way to prevent and serve as a beneficial therapy for CDI. Mice were infected with C. diff spores as shown in [71]Figure S1. To test the ability of the probiotics to prevent CDI, we prepared probiotic yogurt using yogurt starter #2 (YS2) and 11 strains of probiotics (11-SP). Yogurt treatment alone provided no protection to CDI in which 80% mice died in 7 days after infection, whereas LELN-yogurt treatment decreased the mortality to 40% ([72]Figure S2A). Probiotic strains in the yogurt were grouped and treated with LELNs and the mortality evaluated for CDI mice. We confirmed that only Streptococcus thermophilus ST-21 (STH) and Lactobacillus rhamnosus LR-32 (LRH) were essential for protecting mice from CDI ([73]Figures S2B–S2E). We also found that Lactobacillus rhamnosus GG (LGG), which belongs to the same species with LRH, exhibited higher inhibition efficiency to C. diff growth in vitro ([74]Figure S2F); therefore we used LGG to replace LRH in the test probiotic mixture. LELN-pretreated probiotic mixture containing STH and LGG (LELN-LS) further decreased the mortality from 40% to 20%, whereas STH and LGG probiotic mixtures (LS) without the LELN pretreatment provided no protection to CDI ([75]Figure 1A). We also determined colon lengths and gut barrier integrity in infected mice. The colon lengths of both control and the LS group mice were ∼20% shorter than those of non-infected mice, whereas the colon lengths of LELN-LS group mice were comparable with the non-infected mice ([76]Figure 1B). There was obvious mucus damage observed in both control and the LS group of mice, whereas the LELN-LS group mice had similar mucus morphology as non-infected mice ([77]Figure 1C). The gut barrier integrity was also evaluated using fluorescein isothiocyanate-dextran, and a higher fluorescence signal was detected in the serum of PBS control and LS groups of mice compared with the non-infected control mice, whereas the fluorescence signal strength of LELN-LS group mice was comparable with non-infected control mice ([78]Figure 1D). Figure 1. [79]Figure 1 [80]Open in a new tab LELNs-Educated Probiotics Protect Mice Against C. Diff Infection by Increasing Probiotic Survival in the Gut (A) Survival percentage of C. diff-infected mice under different treatments as indicated. PBS treatment was used as a control (n = 10). (B) Colon lengths of C. diff-infected mice (n = 5). (C) Histochemical analysis of colon tissues from C. diff-infected mice by H&E staining (n = 5). (D) Gut permeability analysis measured by the fluorescein isothiocyanate -dextran method (n = 5). (E) Survival percentage of STH under bile challenge in vitro with different treatments as indicated. (F) STH and LELN-STH survivability when passing through the gastrointestinal tract. The significance was analyzed using log rank (Mantel-Cox) tests for survival analyses, t tests for two-group analyses, and ANOVA for multiple group analyses. The significance is shown as ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, and ∗∗∗∗p ≤ 0.0001. p > 0.05 was considered to be not significant (ns). Data are shown as mean ± SD. See also [81]Figure S2. Fruit juice increases lactobacilli survivability in gut, but the cellular and molecular mechanisms in this phenotype are still elusive. We showed that LELNs isolated from lemon extract increased bile resistance of LGG ([82]Figure S2G). Like LGG, bile resistance of STH was also increased over 10-fold due to LELN pretreatment ([83]Figure 1E). The data generated in in vitro test were further supported by the fact that LELN treatment remarkably increased the survivability of STH passing through the gastrointestinal tract as assessed by STH colony-forming unit (CFU) in the feces 24 h after gavage ([84]Figure 1F). LELN-LS Protects Mice from CDI in an AhR-Dependent and AhR-Independent Pathway To further decipher the mechanism of LELN-LS protection to CDI, we performed metabolomics analysis of colon content samples of infected mice. The infected mice were sacrificed at 24 h post infection, and colon content samples were collected for metabolomics analysis. Principal-component and dendrogram analyses showed that both LS and LELN-LS treatment can significantly change colon metabolic composition; however, the changes in LELN-LS mice were more significant ([85]Figures S3A and S3B). Pathway enrichment analysis showed that aromatic amino acid metabolism and aminoacyl-tRNA biosynthesis pathways were most affected due to LS or LELN-LS treatment ([86]Figure S3C). To examine the metabolomics changes in detail, we conducted heatmap cluster analysis and noticed indole-3-carboxaldehyde (I3Ald) and indole-3-lactic acid (I3LA), both of which can be produced by lactobacilli strains and serve as ligands to activate AhR ([87]Zelante et al., 2013; [88]Cervantes-Barragan et al., 2017), were significantly increased in the LELN-LS group compared with the PBS and LS groups ([89]Figure 2A). The concentrations of colonic I3LA and I3Ald were further confirmed by high-performance liquid chromatography (HPLC). The metabolomics analysis was in agreement with the HPLC data indicating that LELN-LS mice had much higher levels of I3Ald and I3LA than PBS controls and LS mice ([90]Figures S3D, S3E, and [91]2B). Interleukin (IL)-22, a cytokine that is secreted by both CD3^+T cells and ILC3 in the gut and serves as a downstream gene of the AhR pathway, was proved to be important in the protection of the host from infection ([92]Parks et al., 2015). Our confocal imaging data also suggest that IL-22 is detected in CD3^+T cells and ILC3 cells ([93]Figures S4A and S4B). qPCR analysis indicates that IL-22 mRNA levels in colon tissue were found to be upregulated over 4-fold in LELN-LS mice and only by 1.5-fold in the LS mice compared with the PBS control mice ([94]Figure 2C). Neither LS nor LELN-LS treatments changed the AhR expression in the colon ([95]Figure 2C). IL-22 protein levels in the small intestine and colon mucus were determined by ELISA. The IL-22 protein levels in the small intestine were shown to be increased from 24 ± 8.5 to 56 ± 6.5 pg/mL and 107 ± 27.6 pg/mL in LS mice and LELN-LS mice, respectively. The colonic IL-22 protein levels were increased from 57 ± 46 to 134 ± 17 pg/mL in LELN-LS mice, but colonic IL-22 protein levels of LS mice were comparable with those of the PBS mice ([96]Figure 2D). To verify the protection effect of I3Ald and I3LA in CDI, I3Ald or I3LA were given to mice via their drinking water at a concentration of 1 mM. Both I3LA and I3Ald increased the survival percentage from 20% to 40% in wild-type C57BL/6 mice ([97]Figure 2E), whereas all the AhR knockout mice (AhR^−/^-), including I3LA- and I3Ald-treated mice died 3 days after being infected ([98]Figure 2F). As both interferon (IFN)-γ and IL-17 were reported to be important to protect against CDI ([99]Chen et al., 2020; [100]Abt et al., 2015), we thus tested whether LELN-LS administration increases IFN-γ and IL-17 expressed in the gut. No significant difference was found in both IFN-γ and IL-17 levels among the groups tested ([101]Figures S4C–S4F). To confirm whether LELN-LS protected mice from CDI only by the AhR pathway, we pretreated AhR^−/^- mice with LELN-LS and conducted a C. diff infection study. All the mice in the PBS control group died within 3 days of being infected, whereas 40% of the LELN-LS group were alive 7 days after being infected ([102]Figure 2F), indicating that an AhR-independent pathway contributes to LELN-LS-mediated protection of mice to CDI. Figure 2. [103]Figure 2 [104]Open in a new tab LELN-LS Protects Mice from C. diff Infection Partially via the AhR Pathway (A) Colon contents metabolomics analysis of C. diff-infected mice with different treatments as indicated. The metabolomics data were normalized by autoscaling (mean-centered and divided by standard deviation), and further transformed with log2. The top 50 most changed metabolites were listed in the heatmap. (B) Confirmation of the concentration of I3Ald and I3LA in the colon content samples using HPLC analysis. (C) Relative mRNA levels of IL-22 and AhR in the colon tissue determined by real-time qPCR; GAPDH were used as internal reference. (D) IL-22 protein levels in the mucus of the small intestine (SI) and colon (LI) determined by ELISA. (E and F) Percent survival of C. diff-infected mice with I3Ald, I3LA, or LELN-LS treatment; both wild-type (E) and AhR knockout (F) C57BL/6 mice were tested. The significance was analyzed using log rank (Mantel-Cox) tests for survival analyses, t tests for two-group analyses, and ANOVA for multiple group analyses. The significance is shown as ∗p ≤ 0.05, ∗∗p ≤ 0.01, and ∗∗∗∗p ≤ 0.0001. p > 0.05 was considered to be not significant (ns). Data are shown as mean ± SD. See also [105]Figure S3. LELN-LS Decreases C. diff Shedding by Altering the Composition of Intestinal Metabolites C. diff shedding in feces is a primary factor in the transmission of CDI and a major issue in preventing the spread. C. diff shedding in the feces of infected mice was determined as a means to study the LELN-LS protection mechanism. Feces were collected from surviving mice on day 7 after infection and CFUs of C. diff were determined by plating on TCCF agar. LS treatment slightly decreased C. diff CFUs compared with the PBS control group, whereas LELN-LS treatment decreased C. diff CFUs by nearly two orders of magnitude ([106]Figure 3A). We then tested whether metabolites from LS directly inhibit C. diff growth in vitro. LS culture supernatant inhibited C. diff growth in a dose-dependent manner ([107]Figure S5A). We also tested whether metabolites from cecum inhibited C. diff growth in vitro. Supernatant from cecum contents were added into the C. diff inoculum (10% by volume). Only the supernatant from the LELN-LS mice cecum content exhibited significant inhibitory effect on C. diff growth ([108]Figure S5B). Both LGG and STH produce lactic acid as an end fermentation product, and lactic acid has been reported to inhibit a large variety of pathogens ([109]Vieco-Saiz et al., 2019). We tested whether lactic acid inhibits C. diff growth in vitro. Lactic acid was found to inhibit C. diff growth efficiently with a minimum inhibitory concentration of 40 mM ([110]Figure S5C). Metabolomics analysis of colon content samples showed lactate levels being significantly increased in the LELN-LS-treated mice ([111]Figure 2A). The lactate concentration in colon content samples was further confirmed using a lactate detection kit. The lactate level in colon content of LELN-LS mice was increased ∼ 4-fold compared with the PBS control mice, whereas there was only a slight but not significant increase in the LS mice ([112]Figure 3B). Figure 3. [113]Figure 3 [114]Open in a new tab LELN-LS Decreases C. diff Shedding in Mouse Feces by Altering the Composition of the Intestinal Metabolites (A) Enumeration of C. diff CFUs in the feces of C. diff-infected mice with different treatments. (B) Lactic acid concentration in the colon contents of C. diff-infected mice. (C and D) (C) Tryptophan and (D) indole concentrations in the colon contents of C. diff-infected mice. (E) C. diff CFUs in the feces of C. diff-infected mice with or without 1 mM indole in drinking water. (F and G) Lactic acid effects on indole production in E. coli under (F) aerobic and (G) anaerobic culture conditions; two different concentrations of lactic acid were added to the E. coli cultures as indicated. (H and I) Lactic acid effect on expression of the indole biosynthesis gene tnaA under (H) aerobic and (I) anaerobic culture conditions, 10 mM of lactic acid were used in the tests. 16s rRNA gene were used as internal reference for relative quantification. The significance was analyzed using t tests for two-group analyses and ANOVA for multiple group analyses. The significance is shown as ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, and ∗∗∗∗p ≤ 0.0001. p > 0.05 was considered to be not significant (ns). Data are shown as mean ± SD. See also [115]Figure S5. We also noticed decreasing levels of indole along with increasing levels of tryptophan in the colon based on metabolomics analysis and confirmed by HPLC ([116]Figures 2A, [117]S3E, [118]3C, and 3D). Recent research indicates that indole levels are increased in patients with CDI, and the authors inferred that indole might play a role in C. diff survival by modulating the gut microbiota ([119]Vieco-Saiz et al., 2019). We propose that LELN-LS can decrease the C. diff burden by inhibiting production of indole in the gut. To test this hypothesis, we challenged mice with a lower dose of 10 C. diff spores to avoid a high mortality. After recovering from an initial diarrhea (day 4 after infection), mice were treated with LELN-LS with or without indole in their drinking water. C. diff numbers in the feces were counted every 2 days for a total of 8 days. C. diff numbers in the LELN-LS control mice decreased significantly over time, whereas C. diff numbers in the indole-treated group persisted at a high level throughout the analysis period ([120]Figure 3E). To further understand the mechanism as to how LELN-LS downregulates indole levels in the gut, we tested the effect of lactic acid on indole biosynthesis in E. coli, which is one of the primary indole producers in the gut ([121]Kumar and Sperandio, 2019). We first tested lactic acid effect on E. coli growth by measuring growth curve. We found that lactic acid only slightly inhibited E. coli growth at a concentration of 10 mM, whereas 1 mM of lactic acid had no effect on E. coli growth ([122]Figure S5D). We then tested lactic acid effect on indole biosynthesis. We found that a lactic acid concentration of 10 mM can decrease indole levels in E. coli culture broth in both aerobic and anaerobic conditions and lactic acid showed stronger inhibition of indole biosynthesis in anaerobic growth conditions ([123]Figures 3F and 3G). We further tested the effect of lactic acid on the expression of the indole biosynthesis gene tnaA. Lactic acid with indicated concentrations were added into E. coli culture when OD600 reaches 0.6, bacteria were collected to isolate total RNA at 2 h after addition of lactic acid. Lactic acid at a concentration of 10 mM decreased tnaA expression ∼ 2-fold in aerobic conditions and over 4-fold in anaerobic conditions ([124]Figures 3H and 3I). To confirm if lactic acid regulation to tnaA expression is specific, we also tested another housekeeping gene rpoD, which is not affected by lactic acid. Collectively, these data suggest that LELN-LS treatment leads to increasing the amount of intestinal lactic acid, which subsequently causes a decrease in the production of indole, which is a survival factor for C. diff. This occurs due to inhibition of the expression of the indole biosynthesis gene tnaA. Co-LS Increases FBP-Mediated LDH Activity and Promotes Production of Lactic Acid and I3LA It is interesting that neither LELNs-pretreated STH nor LELNs-pretreated LGG alone was sufficient to protect mice from CDI ([125]Figure S6A); this indicates a synergetic effect between these two strains. We found that supernatant from in vitro co-cultures of STH and LGG without LELNs (Co-LS) exhibited higher inhibition efficiency to C. diff in vitro compared with pure culture ([126]Figure S6B). We then quantified lactic acid in the broth from LGG, STH, and Co-LS and found Co-LS broth contained higher concentrations of lactic acid than when STH and LGG were cultured separately ([127]Figure 4A). Metabolomics analysis of supernatant from LGG, STH, and Co-LS cultures was performed to analyze the metabolic changes under co-culture conditions. We found that a number of organic acids including lactic acid (LA), p-hydroxyphenyllactic acid (p-HPLA), phenyl-3-lactic acid (P3LA), and I3LA were increased significantly in Co-LS supernatant ([128]Figure 4B). We confirmed the concentration of I3LA in the broth by HPLC, which is consistent with the metabolomics analysis result ([129]Figure 4C). p-HPLA, P3LA, and I3LA have a common biosynthesis pathway in the aryl-lactic acid group, and LA can be biosynthesized by lactate dehydrogenase (LDH) from pyruvic acid ([130]Figure S7). Based on this information we tested whether LDH can convert aryl-pyruvic acid into aryl-lactic acid using an in vitro assay. Using commercially available LDH derived from Lactobacillus leichmannii, we successfully transformed I3PA to I3LA, with NADH as a cofactor ([131]Figure 4D). We also noticed that I3PA is very unstable, especially at a lower pH such as pH4.5, and prone to degrade into I3Ald when LDH is absent ([132]Figure S8A). We thus propose an I3Ald and I3LA biosynthesis pathway in LGG and STH, involving aromatic amino acid aminotransferase (ArAT), LDH, and spontaneous degradation ([133]Figure S8B). Figure 4. [134]Figure 4 [135]Open in a new tab STH Metabolites Activate LGG-LDH and Produce More Lactic Acid and I3LA when in Co-culture (A) Lactic acid concentration in the supernatants of bacterial cultures without LELNs. (B and C) (B) Metabolomics analysis of supernatants from different bacteria cultured as indicated. The metabolomics data were normalized by autoscaling and further transformed with log2. The top 20 most changed metabolites among groups are listed. (C) Confirmation by HPLC of I3LA concentration in culture supernatants. (D) In vitro transformation of I3PA to I3LA by the catalysis of LDH. (E and F) Total LDH activities of (E) LGG and (F) STH under different pH and different concentrations of FBP as indicated. (G) HPLC analyses of intracellular FBP concentration. (H) Statistic analysis of intracellular FBP concentration determined by HPLC. (I) Intracellular FBP concentration of LGG shocked with the supernatant from LGG or STH overnight cultures. (J) Relative mRNA levels of pfk1 and fbp in LGG with indicated treatments, 16S-rRNA was used as internal reference gene for real-time PCR analysis. (K and L) (K) Lactic acid and (L) I3LA production of LGG treated with supernatant from LGG or STH. All in vitro cultures were conducted without LELN. The significance was analyzed using t tests for two-group analyses and ANOVA for multiple group analyses. The significance is shown as ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, and ∗∗∗∗p ≤ 0.0001. p > 0.05 was considered to be not significant (ns). Data are shown as mean ± SD. See also [136]Figures S6 and [137]S8. We then tested whether Co-LS increases the total LDH activity using an in vitro assay. Total LDH activities from bacteria lysates were tested in two different buffer systems, 1X PBS buffer @ pH7.4 and 50 mM acetate buffer @ pH4.5. We found Co-LS exhibited the highest total LDH activity in both buffers tested ([138]Figures S9A and S9B). We also found that the LDH activity of LGG (LGG-LDH) is pH dependent, whereas STH-LDH is pH independent ([139]Figures S9A–S9C). FBP was reported to be an LDH allosteric activator ([140]Garvie, 1980); we thus tested whether FBP activates LGG-LDH and STH-LDH in an in vitro assay. We found that LGG-LDH is remarkably activated by FBP at both concentrations we tested, 1 mM and 10 mM at a pH4.5, whereas STH-LDH was just slightly activated by FBP ([141]Figures 4E and 4F). We then determined the intracellular FBP concentration of LGG, STH, and Co-LS by HPLC and found that the concentration of intracellular FBP in STH is much higher than LGG, whereas intracellular FBP concentration in Co-LS is comparable with STH ([142]Figures 4G and 4H). To test whether metabolites from STH benefit FBP accumulation in LGG, we shocked a stationary-phase LGG culture with STH culture supernatant and noticed that the intracellular concentration of FBP in LGG was increased when compared with being shocked with LGG culture supernatant ([143]Figure 1I). We then tested the expression of two FBP metabolism enzymes, phosphofructokinase-1 (pfk1, LGG_01374) and fructose bisphosphatase (fbp, LGG_02032), to understand the mechanism of STH metabolites benefit. We found the fbp, which converts FBP to fructose-6-phosphate (F6P), is the main rate-limiting enzyme in the gluconeogenesis pathway ([144]Ganapathy et al., 2015), and it was decreased significantly due to STH supernatant shock ([145]Figure 1J), which explains the increase of intracellular FBP of LGG. Finally, we tested whether supernatant from STH increases lactic acid and I3LA productions in LGG. Indeed, we found that STH supernatant treatment increased both lactic acid and I3LA production ∼30% in LGG ([146]Figures 4K and 4L). Discussion In this study we demonstrate that LELNs-manipulated probiotics can inhibit CDI through both AhR-dependent and AhR-independent pathways ([147]Figure 5). We found that maintaining sufficient numbers of probiotics in the intestine is essential for protecting against CDI, and LELNs serve as a potential prebiotic to remarkably increase probiotic survivability in the gut, thus protecting mice from CDI and fecal shedding. Our study opens a new approach to manipulate well-characterized probiotics, such as LGG and STH, with an unlimited resource of edible plant ELNs in a personalized fashion because the gut microenvironment is different in individuals. Figure 5. [148]Figure 5 [149]Open in a new tab Diagram of LELNs-Educated Probiotics to Protect Mice Against C. diff Infection LELNs manipulation increase both LGG and STH survival rate in the gut, which lead to increase production of AhR ligands I3Ald and I3LA, and lactic acid. I3Ald and I3LA induce more IL-22 to protect mice from CDI by activating AhR pathway. On the other hand, lactic acid protected from CDI by directly inhibiting C.diff growth and inhibiting indole biosynthesis. Metabolites from STH can inhibit LGG gluconeogenesis pathway to increase production of I3LA and lactic acid when co-culturing these two strains, thus exhibiting a synergistic effect in protecting against CDI. Various types of harsh gut environmental conditions, particularly in the stomach and small intestine, could be detrimental for the survival of many types of probiotics ([150]Dodoo et al., 2017). In the small intestine, bile acids affect probiotic viability ([151]Wahlstrom et al., 2016). The main functions of bile acids in the lumen are exhibiting antibacterial properties, acting as detergents that disrupt cell membranes, as well as functioning as DNA-damaging agents ([152]Merritt and Donaldson, 2009). Our finding that uptaking LELN leads to increasing survivability of LGG and STH could be due to inducing tolerance to a number of harsh gut environmental conditions inducing bile resistance. Therefore, our finding will provide a foundation for further determining whether LELN-LS also develops resistance to other harsh gut environmental conditions and identifying the molecular mechanism underlying how anti-bile acid of these probiotics is induced by LELNs. Our study shows that LELN-LS treatment increases I3Ald and I3LA levels, whereas decreases the levels of indole in the gut. I3Ald and I3LA act as AhR ligands and activate AhR, which in turn activates IL-22 expression. IL-22 plays critical roles in reducing the severity of many gut infections ([153]Valeri and Raffatellu, 2016). Our research has shown that IL-22 knockout mice exhibit much more severe symptoms upon CDI (data not shown). Meanwhile, LELN-LS treatment also leads to a decrease in indole levels (also an AhR ligand) in the colon; indole is not a high-affinity ligand for AhR in mice ([154]Hubbard et al., 2015). These results provide a foundation for further studying whether genes induced by I3Ald and I3LA are different from the gene(s) induced by indole, which results in the different biological effects on the AhR+ recipient cells. In addition, indole is predominately detected in the large intestine, whereas I3Ald and I3LA are detected in the small and large intestine according to lactobacilli distribution in the gut. This helps explain our data where the fold increase of IL-22 in the colon is lower than the fold increase in the small intestine as a result of LELN-LS treatment. Besides the AhR-mediated pathway that protects mice from CDI, other pathways have been recognized as well for protection of mice from CDI, including innate and adaptive immune responses ([155]Hughes et al., 2016; [156]Solomon, 2013). Our results showed that indole is a survival factor for C. diff and production of bacterial indole is inhibited by lactic acid induced by LELN-LS treatment. In addition, indole is a quorum-sensing molecule ([157]Kumar and Sperandio, 2019; [158]Lee and Lee, 2010) and may have critical impact on gut microbiota composition, resulting in a favorable growth environment for C.diff, whereas LELN-LS treatment changes gut microenvironment to handicap C.diff growth. Although we demonstrated that LELNs renders bile resistance of the probiotics, other properties of LELNs recipient could be also altered, collectively contributing to anti-CDI as we demonstrated in this study. Probiotics are live organisms that confer health benefits on the host when accumulated in the right location in the intestine in adequate amounts, by bringing the microbial balance in the system, and modulating host immune response to inhibit the growth of pathogenic organisms such as C. diff through different mechanisms such as adherence to epithelial cells, modulation of the immune system, and secretion of antimicrobial compounds ([159]Kechagia et al., 2013). Unlike other single factors through which it is almost impossible to alter multiple properties of probiotics, simultaneously, edible plant ELNs consist of proteins, lipids, and nuclear acids, in particular microRNAs, ([160]Sundaram et al., 2019; [161]Teng et al., 2018) and are capable of altering a number of biological activities by targeting multiple factors in ELN recipient probiotics, simultaneously. Hence, our approach will provide a strategy for enhancing the survivability of probiotics like other technology such as encapsulating the probiotics in a colon-targeting polymer, and ELN itself contributes to the beneficial effect on the host by possibly targeting multiple pathways of the probiotics. In addition, unlike polymer, because ELN is derived from the diet we take, there will be much less consideration of the side effects. Therefore, our results presented in this study also open a new avenue to investigate the effect of ELN, in general, on the improvement of probiotic's beneficial properties besides inducing bile acid resistance. It is well recognized that gut microbiota-derived metabolites have huge impact on our overall health ([162]Owyang and Wu, 2014). Our results showed that LELN treatment leads to altering the composition of LELN recipient metabolites; it is conceivable that these altered metabolites may contribute to anti-CDI by modulating not only the activity of gut microbiota near LELN-recipient probiotics but also host cellular activity when they are increased systemically. However, we must be careful about drawing any final conclusions regarding in vivo application based on data generated from an in vitro study. As we demonstrated in this study, in vitro data show that co-culture of LGG and STH leads to the highest level of lactic acid, which inhibits C. diff growth. However, oral administration of LGG and STH (LS) without LELN treatment does not prevent CDI ([163]Figure 1A). This discrepancy is likely due to the probability that the number of LGG and STH probiotic gavage given decreases dramatically while traversing through the intestine ([164]Figure 1F), which limits the amount of beneficial factors that are generated, including lactic acid ([165]Figures 2A and [166]3B). LELNs pretreatment is necessary to enhance the survivability of STH and LGG so that an ample level of lactic acid is produced to prevent C. diff growth in vivo and inhibit indole biosynthesis. Combinations of different probiotics for acquiring better health benefits have been attempted, although little is known about the mechanism of benefit. Our study indicates a synergistic effect between LGG and STH, both of which are essential to protect against CDI. STH is an important industrial strain that is widely used in yogurt and cheese production to accomplish fast acidification ([167]Dandoy et al., 2011). Contrary to LGG-LDH whose activity is pH dependent, our data showed that STH-LDH is pH independent, which is important for allowing the accumulation of lactic acid and fast acidification. LDH may be another factor other than PrtS ([168]Dandoy et al., 2011) that contributes to fast acidification. In addition, we found that the metabolites of STH can inhibit the expression of fbp, the key enzyme in the gluconeogenesis pathway ([169]Ganapathy et al., 2015), and that the reduction of fbp in LGG results in the accumulation of FBP, subsequently activating LGG-LDH. Therefore, our research provides insights not only into the combined usage of probiotics but also into new ideas for future research on the interaction of combined probiotics via a metabolomics approach. Limitations of the Study Although our research showed that LELN increases bile resistance of probiotics LGG and STH, further research needs to be done to dissect out the molecular mechanisms underlying the LELN-mediated cross talk between LGG and STH. Resource Availability Lead Contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Prof. Huang-Ge Zhang ([170]h0zhan17@louisville.edu). Materials Availability This study did not generate new unique reagents. Data and Code Availability This study did not generate datasets/code. Methods All methods can be found in the accompanying [171]Transparent Methods supplemental file. Acknowledgments