Abstract
The use of beneficial bacteria to promote gastrointestinal heath is widely practiced, however, the mechanisms whereby many of these microbes elicit their beneficial effects remain elusive. Previously, we conducted a screen for the discovery of novel beneficial microbes and identified the potent cytoprotective effects of a strain of Lactococcus lactis subsp. cremoris. Here, we show that dietary supplementation with L. lactis subsp. cremoris induced transcript enrichment of a set of genes within the colon whose functions are associated with host cell and microbe interactions. Specifically, L. lactis subsp. cremoris induced the expression of tlr2, which we show was required for L. lactis subsp. cremoris to elicit its beneficial effects on the intestine. L. lactis subsp. cremoris did not confer beneficial effects in mice deficient in TLR-2, or deficient in its adaptor protein Myd88 in chronic gut injury models. In addition to cytoprotection, culture supernatant from L. lactis subsp. cremoris accelerated epithelial migration in a cultured epithelial cell scratch wound assay; and effect that was abrogated by a TLR-2 antagonist. Furthermore, L. lactis subsp. cremoris accelerated epithelial tissue restitution following the infliction of a colonic wound biopsy in a TLR-2 and Myd88-dependent manner. Within colonic wounds, L. lactis subsp. cremoris induced the activation of signaling pathways that function in tissue restitution following injury, including the ERK signaling pathway, and of focal adhesion complex (FAC) proteins. Together, these data demonstrate that L. lactis subsp. cremoris signals via the TLR2/MyD88-axis to confer cytoprotection and accelarated tissue restituion in the gut epithelium. These data point to evolving adaptations where beneficial gut microbes moduate innate immune signaling to excert positive influnces on host physiology.
Introduction
Supplementation of the diet with live bacteria, termed probiotic therapy, is a continuously expanding market, with recent reports recording that up to 4 million adults consume probiotics in the USA, and with up to 60% of healthcare providers prescribing probiotics to patients [1, 2]. Consumption of probiotics has been reported to improve resistance against pathogens [3], prevent cardio-metabolic disease [4, 5], help mental and behavioral wellbeing [6], ease gastrointestinal (GI) distress [7, 8], and priming of the immune system [9, 10]. In spite of these reports, evidence of probiotic efficacy remains sparse, especially in humans [11, 12], and the efficacies of some probiotics for existing conditions is highly debated [13]. Recent reports have examined the extent to which probiotics colonize the human GI tract, and in particular the mucosa-associated surfaces [14]. In addition, the modulatory effects of probiotics on community structure of the resident microbiome in people have not been thoroughly examined, with a recent review reporting no effect of probiotics on fecal microbiome composition [15], and other studies observing modifications in the fecal microbial community structure in patients that ingested probiotics [16–18]. Due to the shortage of conclusive empirical evidence, further investigations focused on the effects of probiotics on the host are required for clinicians and consumers to be able to make full and educated choices regarding therapies.
Inflammatory bowel disease (IBD) is caused by defects that interact across a genetic, immunologic and environmental axis, creating symptoms that are often difficult to treat [19]. Disease of the intestine including Crohn’s disease (CD) and Ulcerative colitis (UC), have at their source, uncontrolled gut mucosal inflammation [20, 21]. The intestine is lined by epithelial cells that form a strong yet selectively permeable barrier to protect the underlying lamina propria from luminal contents. However, sustained inflammation experienced during IBD often leads to increased gut permeability and breaches in the epithelial barrier that can often exacerbate damaged mucosa [21, 22]. A delay in promoting intestinal barrier restitution can cause further tissue damage that increase wounding and leads to chronic disease [20, 21, 23]. The chronic inflammation experienced by IBD patients often disrupts the normal healing process and creates defects in wound repair [24]. A decrease in inflammation is a primary therapeutic goal for most IBD patients, and generally a first line treatment. Since the intestinal epithelium is paramount to wound healing, augmenting the signaling pathways that drive reorganization of the epithelium and promote the growth and repair necessary to rebuild the intestinal mucosa is critical.
The microbial composition in IBD patients becomes aberrant with normal microflora such as the genera Bifidobacterium and Lactobacillus decreasing, while pathogenic or potential harmful bacteria are increased [25, 26]. Critically, dietary supplementation with beneficial microbes may balance the indigenous microbiome in IBD patients, and have a therapeutic effect on IBD [27, 28]. Indeed, beneficial microbes, that induce an advantage to health [29, 30] can prompt a change in the diversity and metabolic activity of the gut microbiota [31], modulate innate and adaptive immune responses [32, 33], or significantly improve epithelial barrier function [34, 35]. Despite this knowledge, the lack of empirical data to substantiate claims made by marketed microbial products poses a considerable challenge to consumers and caregivers when selecting a suitable microbe as a potential therapeutic intervention to treat a given ailment, including IBD. Thus, there is a pressing need to expand primary scientific literature that experimentally substantiates the proposed uses of beneficial microbes.
Our group has reported on the conserved molecular mechanisms of host cell and beneficial microbe interactions in the intestine of Drosophila and mice [36–38]. Using assays we developed to show the mechanism of the beneficial effects of microbes in the Drosophila gut, we initiated a screen to discover new and potentially more efficacious beneficial bacteria. To this end, we successfully identified a previously uncharacterized beneficial bacterium, namely L. lactis subsp. cremoris ATCC19257 that elicited highly efficacious protective influences on the Drosophila and mouse intestine [39]. The objective of this study is to identify the key molecular components and functional elements that mediate L. lactis subsp. cremoris capacity to promote the host’s ability to restrain intestinal inflammation and accelerate restitution of intestinal injury and wound repair. These studies will contribute to inform potential ‘Early Clinical Trial with Live Bio-therapeutic Products’, using L. lactis ATCC19257 to treat intestinal diseases. To identify the molecular mechanisms whereby L. lactis subsp. cremoris elicits its beneficial effects on the host, we characterized colonic gene expression induced by L. lactis subsp. cremoris, the host sentinel elements that recognize L. lactis subsp. cremoris, and the downstream host signaling pathways that are activated to elicit beneficial influences.
Herein, we show that L. lactis subsp. cremoris induced the expression of a set of genes in the colon that are associated with host tissue homeostasis and response to injury, including the JAK-STAT signaling pathway. Kegg pathway analysis also identified TLR2 and Myd88 as potential mediators of host cell and L. lactis subsp. cremoris signaling. Indeed, we show that L. lactis subsp. cremoris also elicited cytoprotection against DSS experimentally induced chronic colitis injury in mice in a Myd88 and TLR2-dependent manner. Furthermore, the culture supernatant of L. lactis subsp. cremoris induced cell proliferation in both cultured human epithelia cells, and in ex vivo organoids and TLR2-dependent manner. The capacity to accelerate wound restitution is a valuable property of putative beneficial bacteria. We show that the filtered supernatant from cultures of L. lactis subsp. cremoris was sufficient to accelerate restitution of wounds inflicted on cultured cells, and also show that supplementation of L. lactis subsp. cremoris accelerated the restitution of or on biopsy wound inflicted on colonic tissue. In both cultured cells and colonic tissue, L. lactis subsp. cremoris activated the expression of Focal adhesion complex (FAC) proteins which function in cell movement on the leading edge of healing wounds in a TLR2 - dependent manner. These data show that L. lactis subsp. cremoris signals via the TLR2/MyD88-axis to confer cytoprotection and accelarated tissue restituion in the gut epithelium. These data show that beneficial L. lactis subsp. cremoris can moduate innate immune signaling to confer beneficial influnces on host physiology.
Results
L. lactis subsp. cremoris dietary supplementation induces transcript enrichment within the colon
To assess the extent to which L. lactis subsp. cremoris activates gene expression in the colon, germ-free C57BL/6 mice were fed 1×108 CFU L. lactis subsp. cremoris by oral gavage. 4 hours after ingestion, the mice were sacrificed, colons were harvested, and transcript enrichment measured by RNAseq analysis. Principle component analysis (PCA) analysis showed that the transcriptome of mice fed L. lactis subsp. cremoris clustered to a tight and separate group from control fed vehicle mice (Figure 1A). Representation of the transcriptome of these mice by volcano plot analysis identified the most statistical significant and highest fold change genes activated in the colon by L. lactis subsp. cremoris (Figure 1B). Visualizing the expression of genes across the samples by heat map analysis identified the top differentially expressed genes in the colon activated by L. lactis subsp. cremoris, including the upregulation of tlr2 transcript levels (Figure 1C). In addition, KEGG pathways based on all differentially expressed genes revealed significantly enriched activation of cell signaling pathways that function in host tissue homeostasis, including the JAK-STAT pathway (Figure 1D and 1E). These data are consistent with our previous report where we showed JAK-STAT pathways signals can be initiated by probiotic lactobacilli in the gut lumen to sub-epithelial compartments [40]. To establish how these pathways may be increased by L. lactis subsp. cremoris, we performed string pathway analysis using the leading edge genes identified in figure 1E and identified proteins that function in host cell and microbe interactions, including Myd88 and TLR2 (Figure 1F). These data show that L. lactis subsp. cremoris activates a specific genetic signature through host cell microbial sentinel receptors, which may mediate its beneficial influences on host health and disease.
L. lactis subsp. cremoris dietary supplementation protects against gut injury induced by Dextran Sodium Sulfate (DSS) in a Myd88-dependent manner
A key mediator of host cell and microbe interaction is the pattern recognition receptor (PRR) adaptor protein Myd88. Because string pathway analysis pointed to a potential role of innate singling, we investigated if Myd88 is required for L. lactis subsp. cremoris to elicit its beneficial influences on host health and disease. We supplemented the diet of C57BL/6 wild-type and Myd88-/- mice daily for 2 weeks with 1×108 CFU L. lactis subsp. cremoris or vehicle control by oral gavage. Mice were then subjected to multiple rounds of 3% DSS in their disease activity recorded. We first show that C57BL/6 mice that had been supplemented with L. lactis subsp. cremoris exhibited significantly less disease activity index compared to C57BL/6 that were treated with vehicle control (Figure 2A). L. lactis subsp. cremoris supplementation significantly lowered diseases parameters in wild type mice, but not in Myd88-/- mice up to day 13 following inclusion of DSS in the drinking water (Figure 2B). The protective influences of L. lactis subsp. cremoris were most pronounces at day 5 following inclusion of DSS in the drinking water (Figure 2C). We repeated the experiment as described in figure 2A but sacrificed the experimental groups at day 5 following after inclusion of DSS in the drinking water. L. lactis subsp. cremoris supplementation significantly protected against DSS-induced colon reduced length in wild type mice, but not in Myd88-/- mice (Figure 2D and E). Before mice were sacrificed, they were subjected to colonoscopy analysis. L. lactis subsp. cremoris supplementation markedly protected wild type mice, but not Myd88-/- mice from DSS-induced tissue injury, as detected by visibly less tissue injury in the colonoscope view, and by histological analysis (Figure 2D and E). These data indicate that L. lactis subsp. cremoris supplementation protects against acute gut injury induced by acute Dextran Sodium Sulfate (DSS) in a Myd88-dependent manner.
L. lactis subsp. cremoris supplementation protects against chronic DSS colitis and improves rate of healing in a Myd88 dependent manner
As an alternative to acute the DSS mouse injury model outlined in figure 2, we also assessed the extent to which L. lactis subsp. cremoris supplementation protects against the milder injury inducing chronic DSS colitis mouse model. Six-week-old C57BL/6 WT or Myd88-/- and mice were administered either 1 × 109 CFU total L. lactis subsp. cremoris or HBSS (vehicle control) for 7 days before 2% DSS was included in the drinking water of experimental groups for repeated rounds over 28 days (Figure 3A). L. lactis subsp. cremoris supplementated mice had significantly lower disease activity in the colonic tissues in a Myd88-dependent manner over the entire study period (Figure 3A-E). At sacrifice at day 28, no differences were detected in spleen weight or colon weight between groups (Figure 3F and G). However, L. lactis subsp. cremoris supplementation significantly protected against DSS-induced reduced colon length, preserving colon length in wild type mice, but not in Myd88-/- mice at day 28 following the repeated DSS cycles. Furthermore, histological analysis of colonic tissue harvested at day 28 following the repeated DSS cycles confirmed that L. lactis subsp. cremoris supplementation mice significantly reduced colonic injury in a Myd88-dependent manner (Figure 3I-M). Together, these data show that L. lactis subsp. cremoris supplementation protects against chronic DSS colitis and improves healing rate following DSS injury of in a Myd88 dependent manner.
L. lactis subsp. cremoris promotes healing of biopsy wounds inflicted on the colon of in a Myd88-dependent manner
To confirm our observation of accelerated recovery following DSS-induced injury and faster wound healing in DSS injured mice, we inflicted repeated biopsies on the colon of individual mice and quantitated healing rates by colonoscopy. 6-week-old C57BL/6 WT or Myd88-/- mice were supplemented orally with either 1×108 CFU total L. lactis subsp. cremoris, or with the non probiotic B. cereus, or vehicle control for 7 days. Mice were then inflicted with 4 wounds at distances of 1, 2, 3 and 4cm into the colon using biopsy forceps. Measurement of the rate of healing of biopsy was assessed by colonoscopy at 48 hours and 96 hours after initial wounding revealed that mice supplemented with L. lactis subsp. cremoris had significantly faster healing rates compared to mice supplemented with B. cereus or vehicle control (Figure 4A-C).
Importantly, no differences in the healing rates of wounds were detected in Myd88-/- mice supplemented with L. lactis subsp. cremoris (Figure 4A-C), showing that L. lactis subsp. cremoris supplementation promoted healing of biopsy wounds inflicted on the colon in a Myd88- dependent manner.
L. lactis subsp. cremoris culture supernatant promotes cytoprotection and induces cell proliferation in cultured epithelial cells and in ex vivo cultured organoids in a TLR2-dependent manner
Dietary supplementation with L. lactis subsp. cremoris elicited beneficial influences on host colonic tissue. The colonic mucosa is separated from the luminal bacteria in the fecal stream by a thick mucus layer. Therefore, the beneficial bacteria do not come in direct physical contact with colonocytes, and therefore must exert their effects by the release of factors that penetrate the mucus layer. Upstream of Myd88 are the toll-like receptors (TLRs). Indeed, transcriptome analysis by RNAseq of colonic tissue of mice fed L. lactis subsp. cremoris detected an increase in expression of tlr2 (Figure 1C). We confirmed this observation by quantitative (q)RT-PCR and show significantly enriched tlr2 transcript levels in cultured T84 cells overlayed with media containing a 1 in 100 dilution of a filter sterilized culture supernatant collected from centrifuged cultures of L. lactis subsp. cremoris compared to media control (Figure 5A). To assess the extent to which L. lactis subsp. cremoris culture supernatant promotes wound healing, we inflicted a scratch wound on cultured T84 human epithelial cells, and then overlaid the cells with the filter sterilized culture supernatant of L. lactis subsp. cremoris or B. cereus or vehicle control at a 1:100 dilution factor for up to 72 hours and monitored healing rates. Culture supernatant of L. lactis subsp. cremoris significantly accelerated wound closure compared to B. cereus or vehicle control - a response that was abrogated by the presence of a TLR2 antagonist in the culture media (Figure 5B-E). Cell movement is a critical aspect of wound healing that involves the migration of cells. Cell signaling networks including the ERK signaling pathway, and the composite of proteins that comprise the focal adhesion complexes (FAC) are mediators some of cell migration. We detected markedly increased levels of phosphorylated ERK in cultured cells overlaid with filter sterilized culture supernatant of L. lactis subsp. cremoris, compared to filtered media of B. cereus or vehicle control, or of filtered media from a culture of Lactobacillus rhamnosus GG that has been shown to activate ERK pathway signaling[41, 42] (Figure 5F and 5G). We also detected markedly increased levels of phosphorylated focal adhesion kinase (p-FAK), a marker for FAC activity, in cultured cells overlaid with culture supernatant of L. lactis subsp. cremoris - a response that was abrogated by the presence of a TLR2 antagonist in the culture media (Figure 5H). Together, these data show that culture supernatants promote healing of scratch wounds inflicted on cultured epithelial cells and activate cell signalling pathways that function in cell migration in a TLR2-dependent manner.
L. lactis subsp. cremoris supplementation induces TLR2 dependent activation of ERK and FAK to exert pro-restitutive effects on the colon
To investigate if TLR2 is required for L. lactis subsp. cremoris to elicit its beneficial influences on host health and disease, we supplemented the diet of C57BL/6 WT or TLR2-/- mice daily for 2 weeks with 1×108 CFU L. lactis subsp. cremoris or vehicle control by oral gavage before repeated rounds of 2% DSS included in the drinking water of experimental groups for 28 days (Figure 6A). L. lactis subsp. cremoris supplementation mice significantly lowered disease activity in the colonic tissues of mice in a TLR2-dependent manner over the entire study period (Figure 6B-E). These data show that L. lactis subsp. cremoris dietary supplementation protects against gut injury induced by chronic Dextran Sodium Sulfate (DSS) in a TLR2-dependent manner. To confirm our observation of accelerated recovery following DSS-induced injury wound healing in DSS injured mice, we inflicted repeated biopsies on the colon of individual mice and quantitated healing rates by colonoscopy. 6-week-old C57BL/6 WT or TLR2-/- mice were supplemented with either 1×108 CFU total L. lactis subsp. cremoris, Bacillus cereus (BC) or vehicle control for 7 days. Mice were then inflicted with 4 wounds at distances of 1, 2, 3 and 4cm into the colon using biopsy forceps (Figure 6F). Measurement of the rate of healing of biopsy was assessed by colonoscopy at 48 hours and 96 hours after initial wounding revealed that mice supplemented with L. lactis subsp. cremoris had significantly accelerated healing rates in a TLR2-depenendet manner (Figure 6G-H). To confirm our findings of L. lactis subsp. cremoris-induced activation of ERK and FAK signalling in cultured cells, we also assessed pathway signalling within biopsy wounds. Twelve-week-old WT C57BL/6 or TLR2-/- mice were given four wounds at distances of 1, 2, 3 and 4cm into the colon using biopsy forceps, then subjected to rectal gavage with 1×109 CFU of L. lactis subsp. cremoris, or media control. After 30 mins, mice were sacrificed, and wound beds mounted in OCT for immunostaining. We detected markedly elevated levels of phospho-FAK within the wound beds of WT C57BL/6 supplemented with L. lactis subsp. cremoris in a TLR2-dependent manner (Figure 6I). These data show that L. lactis subsp. cremoris supplementation promoted the activation of pathways that function in healing of biopsy wounds including the activation of FAK phosphorylation within biopsy wounds in a TLR2-dependent manner.
L. lactis subsp. cremoris supernatants elicits improved survival in vitro and promotes proliferation and increased stemness in ex vivo enteroid cultures
TLR2 functions in the sensing of bacterial cell-wall components, such as lipoproteins from Gram-negative and Gram-positive bacteria, and lipoteichoic acid from Gram-positive bacteria. We showed that L. lactis subsp. cremoris does not protect Myd88-/- or TLR2-/- mice from gut injury. To further test the extent to which a factor released by L. lactis subsp. cremoris can elicit beneficial effects on cells, we collected filter sterilized supernatant from centrifuged cultures of this bacterium and included it in the growth media of cultured human T84 colonic epithelial cells. Measurements of lactate dehydrogenase (LDH) levels, which is is rapidly released into the cell culture supernatant when the plasma membrane is damaged as cells undergo apoptosis, necrosis, or other forms of cellular damage revealed significantly lower levels of LDH released by cultured T84 cells overlaid with L. lactis subsp. cremoris culture supernatant compared to controls. (Figure 7A). These data are evidence that L. lactis subsp. cremoris does not induce any significant toxicity on cultured cells. In addition, we assessed the capacity of L. lactis subsp. cremoris filtered culture supernatant to promote tissue development. We generated ex vivo enteroids from C57BL/6 WT and TLR2-/- mice. At day 1 of growth, enteroids media was supplemented with filter sterilized cultured supernatants of L. lactis subsp. cremoris, B. cereus, or MRS bacterial culture media. Analysis of development by quantification of enteroid budding showed that enteroids overlayed with L. lactis subsp. cremoris supernatant had significantly increased growth (budding) after 6 days of enteroid growth compared to MRS control in a TLR2- dependent manner (Figure 7B-D). Of note, TLR2-/- enteroid budding was significantly delayed in each experimental group irrespective of treatment. Finally, consistent with our results in cultured cells and in biopsy wounds, we also show that L. lactis subsp. cremoris filter sterilized culture supernatant markedly activated phospho-FAK and phospho-ERK in situ in enteroids after 6 days of growth (Figure 7E). Together, these data show that L. lactis subsp. cremoris supernatants elicits no toxic influences in T84 cells and promotes proliferation and increased stemness in ex vivo enteroid cultures in a TLR2-dependent manner.
Discussion
An ever-expanding body of evidence validates the concept that beneficial bacteria within the intestine positively influences health by stimulating gut homeostasis and promoting tissue restitution following injury. However, few of the molecular mechanisms whereby beneficial bacteria signal to sub-epithelial compartments to elicit their influence on the host have been recognized. In this study, we identified a set of genes induced by L. lactis subsp. cremoris in the colon that are associated with host tissue homeostasis and response to injury. We showed that L. lactis subsp. cremoris also elicited cytoprotection from colitis injury in mice in a Myd88 and TLR2-dependent mechanism. We showed that the culture supernatant from L. lactis subsp. cremoris was sufficient to induce accelerated wound healing and activation of ErK and FAK siganling in a TLR2-dependent manner. These data define a mechanimsm whereby L. lactis subsp. cremoris signals via the TLR2/MyD88-axis to confer beneficial infleucnes on the the gut epithelium.
These data contribute to the gradually expanding literature describing the effects of beneficial bacteria on host tissue. The focus of previous investigations from our group was the most widely studied strain of probiotic bacteria, namely Lactobacillus rhamnosus GG. We previously reported that L. rhamnosus GG facilitates intestinal stem cell turnover [36], wound restitution in the gut [43], mediates cytoprotective effects in the gut [38], as well as cell migration [42, 44]. We also showed that L. rhamnosus GG can prevent sex steroid depletion-induced bone loss resulting in decreased gut permeability and intestinal and systemic inflammation [45], as well as showing the role L. rhamnosus GG plays in stimulating bone formation by expanding the pool of Treg cells in the gut and the bone marrow [46]. Other reported mechanism of action of LGG are via the generation of the P40 and P75 proteins that belong to a family of secreted cell wall proteins that contain a carboxy(C)-terminal CHAP or NlpC/P60 superfamily domains. P40 was shown to prevent cytokine-induced epithelial damage and apoptosis [47–51]. Another very well-studied beneficial bacterium is Bacteroides fragilis, which released a bacterial polysaccharide (PSA) directs the cellular and physical maturation of the developing immune system [52]. B. fragilis was shown to be equally effective in preventing colitis and allergic encephalomyelitis (EAE) in murine models[53–55]. The immunomodulatory activities of PSA was shown to induce regulatory T cells to secrete IL-10, a potent anti-inflammatory cytokine that restrains pathogenic inflammation in the gut, as well as systemically in the brain [54, 56–58].
The limitations of our previous report on L. lactis subsp. cremoris was that the molecular mechanism whereby L. lactis subsp. cremoris confers its beneficial effects remained unknown [39]. We reported on the mechanisms whereby Lactobacillus rhamnosus GG elicits its beneficial response via the Nox1-dependent cellular generation of reactive oxygen species (ROS), which acts as a signaling molecule to transduce bacterial-induced signals to sub epithelial compartments [36, 38, 42, 43, 59–61]. Furthermore, L. rhamnosus GG is reported to be sensed by host epithelial cells via sensing by formyl peptide receptors (FPRs), which are G protein-coupled receptors and were first characterized in the transduction of chemotactic signals in phagocytes, and the mediation of host-defense and inflammatory responses by superoxide production [42, 43]. Because contact of L. lactis subsp. cremoris with epithelial cells does not activate detectable cellular ROS generation [39], we speculated that L. lactis subsp. cremoris elicited its beneficial effects by a different signaling mechanism to L. rhamnosus GG. The employment of methodologies such as RNAseq which measures transcript enrichment in tissues was instrumental in elucidating candidate functional elements that transduce the beneficial effects of L. lactis subsp. cremoris to the colonic tissue. The function of these elements, namely TLR2 and Myd88 were are confirmed using mice with knockout mutations that disrupt their function. These data highlight a contrast between the mechanism whereby two lactic acid bacteria, namely L. lactis subsp. cremoris and L. rhamnosus GG are sensed by host cells to elicit their beneficial effects.
Existing treatments for gut inflammation include oral dosing of corticosteroids, antibiotics, and immunomodulators by subcutaneous or intravenous administration [62]. The concept of employing microbial-based platforms as treatment to treat intestinal injury is an progressively accepted as a feasible approach [63]. Most existing beneficial bacteria are lactic acid bacteria from the lactobacilli genus [64–67]. Importantly, the characterization of L. lactis subsp. cremoris as an effective bacterium for limiting gut injury, identifies a beneficial lactococcus species with necessary features for therapy. L. lactis strains have been widely used in the production of dairy products, and consequently have a generally recognized as safe (GRAS) status. These desirable properties positively contribute to the prospect of using L. lactis subsp. cremoris as a therapeutic to treat intestinal ailments since potential toxological effects may be unlikely. Together, the innovative approaches of identifying a new and substantially more efficacious beneficial bacteria will yielded a considerable leap in improving the concept behind, and the usage of beneficial bacteria. Ultimately, these discoveries will exert a sustained and powerful influence by directly informing decisions related to data for product formulation, doses and timing for an Early Clinical Trial to substantiate the use of L. lactis ATCC 19257 as a therapeutic modality to accelerate recovery from, preserve remission, or prevent IBD.
Materials and methods
Experimental Mice
All experiments were done using C57BL/6 mice, TLR2 full body knockout mice (Stock No: 004650) and Myd88 knockout mice (Stock No: 009088) purchased from The Jackson Laboratories. Animal procedures were approved by the Institutional Animal Care and Use Committee of Emory University.
Bacterial strains and culture preparation and administration of Bacteria
The following bacteria were purchased from the American Type Culture Collection (ATCC) (Manasas, VA): Lactobacillus rhamnosus GG ATCC 53103 and Lactococcus lactis subspecies cremoris ATCC 19257. All media was propagated according to instructions provided by the ATCC. Bacillus cereus was isolated by our research group [36]. Mice were gavaged with commensals as described previously [39]. Briefly Lactococcus lactis subspecies cremoris ATCC 19257, Bacillus cereus, or vehicle control, HBSS, were orally gavaged to mice every day in the mid-afternoon at a dose of 2×108 CFUs at a volume of 100uL with control mice receiving 100uL of HBSS.
Real-time PCR
Mouse colonic and intestinal tissue was frozen in liquid nitrogen and stored at −80°C until use. Tissues were homogenized for RNA extraction using TRIzol reagent (Invitrogen, Carlsbad, California, USA), and cleaned using RNeasy kit (Qiagen) total RNA clean up protocol. cDNA of these samples was synthesized from 1μg of total RNA using iScript Reverse Transcription Kit, and relative transcript levels were measured for each sample in triplicate by quantitative qRT-PCR using SYBR Green PCR Kit (Biorad), on a MyiQ™ Real time PCR system (Biorad). The data generated by qPCR assays were normalized using the average value of the HBSS treatment control group.
Scratch Wound Assay
Wound healing assays were conducted on T84 cells using a modified protocol as previously described[68]. Briefly, T84 cells were grown as confluent monolayers in 12-well plates and scratched using a P200 tip. Cellular debris from the wound was removed by washing twice in sterile PBS. Cell culture media containing the TLR2i, C29 was included in the culture medium at 10uM final concentration. Wounds were photographed hourly for 24 h using a Zeiss Axiovert 200 M microscope with a Zeiss Incubator XL-3 489 attachment, maintaining cells at 37 °C/5% CO2. Cellular migration was measured using AxioVision v4.9.1 software (Carl Zeiss, Oberkochen, Germany) by measuring the size of the wound at each time point. Three individual measurements were made on each wound and averaged. Three independent experiments with a minimum of three replicates per experiment were pooled. Wound closure was calculated as percentage wound closure from t = 0.
DSS-colitis model in mice
Groups of C57BL/6 mice were purchased from Jackson laboratory and maintained the Emory University Department of Animal Resources. Pure cultures of beneficial bacteria, including, or L. lactis cremoris, or B. cereus (2×109 cfu total) were administered by oral gavage daily for indicated time periods, typicall between 7 and 14 days. Dextran Sodium Sulfate (DSS) was included in the drinking water of some groups of mice to a final concentration of 3% for acute DSS injury, or 2% for the chronic injury model. Body weights, and disease activity index were monitored daily until mice had lost 80% body weights (seven days), whereupon DSS was removed from the drinking water. Murine body weights, and disease activity index (DAI) were monitored daily for a further three days before mice were sacrificed and histological sections of the colon were prepared for disease activity. All animal experiments were approved by the Emory University institutional ethical committee and performed according to the legal requirements.
Endoscope
All procedures and post-anesthesia care were performed in accordance within IACUC guidelines. Each mouse was anesthetized with 3% isoflurane and maintained with 1.5% isoflurane in an oxygen/air mixture by using a gas anesthesia mask. Body temperature was maintained during the procedure at 37°C with a homeothermic blanket (Harvard Apparatus, Holliston, MA). For monitoring of tumor burden, a high-resolution mouse video endoscopic system was utilized (Karl Storz Endoskope, Tuttlingen, Germany). The system consists of a miniature endoscope (1.9 mm outer diameter), a xenon light source, a triple chip camera, and an air pump for regulated inflation of the mouse colon (Karl Storz, Tuttlingen, Germany). The endoscopic procedure was viewed on a color monitor and digitally recorded for post-procedure review.
Enteroid cultures
After washing the mouse colons and intestines (approx. 15cm of Jejunum) 15 times with cold phosphate buffered saline (PBS) without Ca2+/Mg2+, the isolated intestinal tissue was cut into ∼3-to 5-mm pieces and incubated in Gentle Cell Dissociation Reagent (Stem Cell Technologies, Vancouver, Canada). After removal of the Gentle Cell Dissociation Reagent, and resuspension with PBS with 0.1% BSA, the villi and mucus were removed with a 70-μm cell strainer (BD Biosciences, NJ). After being centrifuged at 8°C at 290 x g for 5 minutes, the collected crypts were resuspended in cold DMEM / F12 with 15mM HEPES and counted. 150 enteroid or colonoid crypts were plated in each well in fresh Matrigel at a ratio of 1:1. Passage of enteroids or colonoids derived from single cells was performed every 7-8 days. Post-passaging, half of the colonoids were stimulated with 100ng/ml of PEGylated leptin (Peptides International, KY) added to media 3 times weekly. Light microscopy was performed on cultures 2 days post passaging, with representative images selected after counting 6 images at 100x magnification from each group. Organoid counts were performed 2, 3- and 7-days pre-passaging, and at 7 days post-passaging, with small spheres consisting of enclosed sphere structures, large spheres having a defined ‘lumen”, budding spheres having one bud from the main structure and enteroids or colonoids having multiple buds.
Antibodies and reagents
Colon samples for immunofluorescence were swiss-rolled and fixed in 10% formalin and then processed and embedded in paraffin. 5-micron sections were cut and underwent a deparaffination process followed by a citrate antigen retrieval step before being permeabilized in 0.5% triton-X 100 in PBS and blocked with 5% normal goat serum. Primary antibodies include, from Cell Signaling Technology (Danvers, MA), anti-Phospho-p44/42 MAPK cat#4370, anti-phospho-FAK (Tyr397) cat#3283. Samples were incubated in primary antibody overnight at 4ºC. Samples were then washed three times and incubated for 1 hour in fluorochrome-conjugated secondary antibody. Samples were again washed and incubated with DAPI at a concentration of 1:10000 in PBS for 5 minutes. Samples were mounted using Prolong diamond antifade and imaged on an Olympus FV1000 confocal microscope at 20X. At least 3 sections and three different images were taken for each sample. For immunoblot analysis, immunoreactive species were detected using the aforementioned anti-Phospho-p44/42 MAPK cat#4370, anti-phospho-FAK (Tyr397) cat#3283, followed by anti-rabbit HRP or anti-mouse HRP followed by visualization with ECL chemiluminescence detection reagent (GE Healthcare Biosciences Piscataway NJ). Experiments were conducted in triplicate and the average density of the band quantified using ImageJ software.
Statistical Analysis
One-way ANOVA, with Bonferroni’s multiple comparisons test, was used for analysis of western blot flow cytometry and NTA data. Wound healing assay and invasion assay data were analysed using a Kruskal-Wallis test with Dunn’s multiple comparisons. Two-way ANOVA with Bonferroni’s multiple comparisons was used for analysis of proliferation data. Differences were considered statistically significant for P-values < 0.05. All analyses were performed using GraphPad Prism v6.0.4 software (GraphPad, San Diego, CA, USA). Data are presented as the mean ± standard error of the mean (SEM).
Author Contributions
RMJ, CRN and JAO conceived and designed the experiments. CRN and RNK performed all experiments on enteroids. CRN, JAO, LL, and MB performed experiments on the murine animal model. RMJ, CRN and JAO wrote the manuscript.
Declaration of Interests
The authors declare no competing interests.
Acknowledgments
CRN is supported by an American Heart Association fellowship 19POST34370006. JAO is supported by the NIH through F31CA247415. RMJ is supported, in part, by a startup fund from The Emory University School of Medicine.