1. Microbiology and Infectious Disease

Lactococcus lactis NCDO2118 exerts visceral antinociceptive properties in rat via GABA production in the gastro-intestinal tract

  1. Valérie Laroute
  2. Catherine Beaufrand
  3. Pedro Gomes
  4. Sébastien Nouaille
  5. Valérie Tondereau
  6. Marie-Line Daveran-Mingot
  7. Vassilia Theodorou
  8. Hélène Eutamene  Is a corresponding author
  9. Muriel Mercier-Bonin
  10. Muriel Cocaign-Bousquet  Is a corresponding author
  1. Toulouse Biotechnology Institute (TBI), Université de Toulouse, CNRS, INRAE, INSA, France
  2. Toxalim (Research Centre in Food Toxicology), Université de Toulouse, INRAE, ENVT, INP-Purpan, UPS, France
https://doi.org/10.7554/eLife.77100
Share this article
Cite this article
  1. Valérie Laroute
  2. Catherine Beaufrand
  3. Pedro Gomes
  4. Sébastien Nouaille
  5. Valérie Tondereau
  6. Marie-Line Daveran-Mingot
  7. Vassilia Theodorou
  8. Hélène Eutamene
  9. Muriel Mercier-Bonin
  10. Muriel Cocaign-Bousquet
(2022)
Lactococcus lactis NCDO2118 exerts visceral antinociceptive properties in rat via GABA production in the gastro-intestinal tract
eLife 11:e77100.
https://doi.org/10.7554/eLife.77100
  2. Download BibTeX
  3. Download .RIS

Abstract

Gut disorders associated to irritable bowel syndrome (IBS) are combined with anxiety and depression. Evidence suggests that microbially produced neuroactive molecules, like γ-aminobutyric acid (GABA), can modulate the gut-brain axis. Two natural strains of Lactococcus lactis and one mutant were characterized in vitro for their GABA production and tested in vivo in rat by oral gavage for their antinociceptive properties. L. lactis NCDO2118 significantly reduced visceral hypersensitivity induced by stress due to its glutamate decarboxylase (GAD) activity. L. lactis NCDO2727 with similar genes for GABA metabolism but no detectable GAD activity had no in vivo effect, as well as the NCDO2118 ΔgadB mutant. The antinociceptive effect observed for the NCDO2118 strain was mediated by the production of GABA in the gastro-intestinal tract and blocked by GABAB receptor antagonist. Only minor changes in the faecal microbiota composition were observed after the L. lactis NCDO2118 treatment. These findings reveal the crucial role of the microbial GAD activity of L. lactis NCDO2118 to deliver GABA into the gastro-intestinal tract for exerting antinociceptive properties in vivo and open avenues for this GRAS (Generally Recognized As safe) bacterium in the management of visceral pain and anxious profile of IBS patients.

Editor's evaluation

Gut microbes produce neuromodulators including the neurotransmitter γ-aminobutyric acid (GABA), which could regulate pain and other neurological outcomes. This study examines how oral administration of the probiotic Lactococcus lactis affects visceral pain in rats, finding that a specific strain of L. lactis (NCD02118) suppresses stress induced pain in a manner dependent on bacterial GadB, the enzyme mediates GABA synthesis.

https://doi.org/10.7554/eLife.77100.sa0

Introduction

As defined by the Rome IV criteria, irritable bowel syndrome (IBS) is a functional gastro-intestinal disorder affecting 3–5% of adults in industrial countries. IBS results in various symptoms that strongly affect the patients’ quality of life (Dean et al., 2005) and a significant socioeconomic burden (Peery et al., 2012). Abdominal pain is a prevalent IBS symptom, associated with changes in bowel habits (diarrhoea and/or constipation; Camilleri et al., 2012). Even though IBS pathophysiology remains not completely understood, IBS symptoms may originate from peripheral and/or central mechanisms resulting in a dysfunctional gut-brain axis (Fichna and Storr, 2012). Anxiety and stressful psychological life events lead to central nervous system (CNS) disorder triggering in turn gut motility dysfunctions (Mönnikes et al., 2001) and strengthen visceral sensitivity (Greenwood-Van Meerveld et al., 2016). For a better understanding of the IBS pathophysiology, stress animal models have been developed mimicking IBS features, such as changes in visceral sensitivity and gut transit time (Gué et al., 1997).

Related to the complex pathophysiology of IBS and the diversity of IBS patients’ profiles, the existing therapeutic strategies promote solutions usually limited to treat symptoms (motor/sensory disturbances). In addition, in several cases with psychological comorbidity, psychopharmacological drugs are used (Dekel et al., 2013). As an effective treatment, alternative probiotics were used to modulate visceral pain in IBS patients (Moayyedi et al., 2010; Hungin et al., 2013; Ford et al., 2014; Didari et al., 2015). However, no consensus results were obtained from human clinical studies (Mazurak et al., 2015) and mechanisms by which microbes signal from the gut lumen to the CNS, thus influencing pain perception, remain to be elucidated.

In recent years, mounting evidence has suggested that microbially produced neuroactive molecules can modulate the gut-brain axis communication (Cryan and O’Mahony, 2011; Lyte, 2011; Reid, 2011; Cryan and Dinan, 2012). For example, ingestion of a Lactobacillus strain, i.e., L. rhamnosus (JB-1) is able to regulate emotional behaviour and central γ-aminobutyric acid (GABA)ergic system in mouse via vagus nerve pathway (Bravo et al., 2011). Recently, GABA-modulating bacteria of the human gut microbiota have been associated to brain signatures related to depression (Strandwitz et al., 2019). In fact, GABA is the main inhibitory CNS neurotransmitter in mammals (Wong et al., 2003), and several important physiological functions have been characterized, such as neurotransmission, relaxing, and tranquilizer effects (Hayakawa et al., 2007; Li and Cao, 2010). GABA exerts these major functional effects via two GABA receptor subtypes, i.e., GABAA and GABAB (Hyland and Cryan, 2010) doing these receptors important pharmacological targets for clinically relevant anti-anxiety agents (Foster and Kemp, 2006).

In the human diet, glutamate constitutes up to 8–10% of amino acids; it is involved in gut protein metabolism and is the precursor of different important molecules produced within the intestinal mucosa, like GABA. GABA is synthetized by a pyridoxal-5’-phosphate (PLP)-dependent enzyme glutamate decarboxylase (GAD; EC 4.1.1.15) by irreversible α-decarboxylation of L-glutamate and consumption of one cytoplasmic proton (Ueno, 2000). Prokaryotic and eukaryotic cells are able to synthesize GABA, through the decarboxylation of glutamate and genes encoding GAD are present in the gut microbiota (Mazzoli and Pessione, 2016). Glutamate import and GABA export generally occur in bacteria simultaneously via a specific glutamate/γ-amino butyrate antiporter (Ma et al., 2012). The glutamate-dependent system is associated with acid resistance in many bacteria (Feehily and Karatzas, 2013).

Due to this large repertoire of beneficial effects, GABA has been classified as a health-promoting bioactive component in foods and pharmaceuticals (Li and Cao, 2010). Among bacteria, lactic acid bacteria (LAB) are generally recognized as safe sources to produce GABA at high levels and in an eco-friendly way (Dhakal et al., 2012). Numerous studies revealed that food-derived lactobacilli and gut-derived Lactobacillus and Bifidobacterium species are able to produce GABA in vitro (Li and Cao, 2010; Yunes et al., 2016). Interestingly, Linares et al., 2016 reported the use of a novel Streptococcus thermophilus strain for the production of a naturally GABA-enriched yogurt. However, the link between bacterial GABA production in vitro and neuromodulatory activity in vivo remains poorly investigated. To date, only the human gut commensal GABA-producing Bifidobacterium dentium was shown in vivo to modulate sensory neuron activity in a rat faecal retention model of visceral hypersensitivity (Pokusaeva et al., 2017). Here, we focused on the transient food-borne LAB Lactococcus lactis. Frequently encountered in dairy products, L. lactis is one of the most ingested bacteria (Mills et al., 2010; Laroute et al., 2017). Although not considered as a commensal bacterium, L. lactis was found to persist transiently in the gut, depending on the strain under study (Wang et al., 2011; Radziwill-Bienkowska et al., 2016; Zhang et al., 2016). It was also demonstrated that L. lactis may exert in vivo a potent anti-inflammatory activity attenuating colitis (Nishitani et al., 2009; Luerce et al., 2014; Ballal et al., 2015; Berlec et al., 2017; Nomura et al., 1999). L. lactis is also able to produce GABA in vitro (Nomura et al., 1999) and the well-characterized NCDO2118 strain besides its probiotic properties (Cordeiro et al., 2021), is considered as an efficient GABA producer among the species (Oliveira et al., 2017; Mazzoli et al., 2010; Oliveira et al., 2014; Laroute et al., 2016; Laroute et al., 2021).

Here, we determined whether L. lactis NCDO2118 is able to deliver GABA in vivo via glutamate decarboxylation (GAD activity) and to exert GABAergic signalling-dependent antinociceptive properties in a stress-induced visceral hypersensitivity rat model.

Results

High in vitro GABA production by L. lactis NCDO2118 is related to its GAD activity

The NCDO2118 strain previously demonstrated to produce GABA in chemically defined medium (Laroute et al., 2016), was grown in batch bioreactor with complex medium allowing high production of biomass and GABA (see Experimental section). The biomass and GABA concentrations were measured all along the culture (Figure 1A and B). In these conditions, the maximal concentrations of GABA and biomass reached, respectively, 40 mM at 24 hr and around 2.0 g/L at 8 hr, which are higher than the previously reported values in chemically defined medium (Laroute et al., 2016).

Figure 1
Download asset Open asset
Growth and GABA production of the different Lactococcus lactis strains.

(A) γ-Aminobutyric acid (GABA) production (mM); (B) evolution of biomass (g/L) during growth of Lactococcus lactis NCDO2118 (□, ■) or NCDO2727 (○, ●) in M17 supplemented with glutamate (8 g/L), arginine (5 g/L), glucose (45 g/L), and NaCl (300 mM). Two independent duplicates were performed.

Figure 1—source data 1

Table of biomass and GABA concentrations for L. lactis NCDO2118 and NCDO2727 strains.

https://cdn.elifesciences.org/articles/77100/elife-77100-fig1-data1-v1.xlsx

The NCDO2727 strain was demonstrated here to have the gad operon involved in GABA biosynthesis similar to NCDO2118, in terms of genetic organization (gene sequences and promoter). gadR, gadC, and gadB, the three genes of the gad operon are located between kefA and rnhB like in the NCDO2118 strain and the sequences of the corresponding proteins, GadR, C, and B, are highly similar with more than 99% of identity. Only two single nucleotide polymorphisms and a deletion of one bp were identified in the gadR and gadCB promoters, respectively, compared to NCDO2118. However, when tested for its GABA-producing performances under similar culture conditions, GABA production was very weak (GABA concentration <0.2 mM, Figure 1). The biomass production was lower (maximal biomass around 0.5 g/L at 24 hr) than for the NCDO2118 strain (Figure 1). The GABA concentration did not increase when the biomass production increased during growth of NCDO2727 strain in medium with yeast extract and glutamate (data not shown). For in vivo experiments, the bacterial amount to be orally administrated was adjusted to treat animals with similar viable bacterial cell number for the two strains (i.e. 109 CFU (Colony-Forming Units) per day; see below).

Interestingly, a high intracellular GAD activity was obtained at 7 hr in the NCDO2118 strain (45.3 ± 4.7 µmol/min.mg) consistently with its high GABA production ability, while in the NCDO2727 strain, no activity was detected. The difference of GAD activity between the two strains was confirmed at 24 hr of culture (Supplementary file 1). This comparison demonstrated that in vitro GABA production could not be solely associated to GABA genetic determinants in L. lactis.

Only GABA-producing L. lactis NCDO2118 prevents visceral hypersensitivity induced by stress in response to colorectal distension

The influence of a 10-day chronic treatment by L. lactis NCDO2118 or NCDO2727 on stress-induced visceral hypersensitivity to colorectal distension (CRD) is shown in Figure 2. Rats were given L. lactis strains by oral gavage once daily with washed bacterial cells at the same amount (109 CFU per day) whatever the strain under study. We first verified that GABA-producing L. lactis NCDO2118 in presence of glutamate had no impact on basal CRD sensitivity (Figure 2—figure supplement 1). Then, in vehicle-treated rats, a 2 hr of partial restraint stress (PRS) significantly increased the number of abdominal contractions compared to basal conditions for all the pressures of distention applied from 30 mmHg (p<0.001, Figure 2). In presence of glutamate, oral administration of L. lactis NCDO2118 suppressed the PRS-induced enhancement of abdominal contractions (p<0.01 for all CRD pressures applied, Figure 2A), restoring a quasi-basal sensitivity to CRD. However, in absence of glutamate, the NCDO2118 strain did not exert antinociceptive properties (Figure 2A). Glutamate alone, at the concentration used (0.2% [w/v]), had no impact on basal CRD sensitivity and on visceral hypersensitivity response (Figure 2—figure supplement 2). Oral administration of the low GABA-producing L. lactis NCDO2727 in presence of glutamate failed to reduce stress-induced visceral hypersensitivity (Figure 2B).

Figure 2 with 2 supplements see all
Download asset Open asset
L. lactis effect on visceral hypersensitivity induced by stress in response to colorectal distension.

(A) Effect of 10-day oral administration of γ-aminobutyric acid (GABA)-producing Lactococcus lactis NCDO2118 (109 CFU per day) in presence or absence of glutamate on PRS-induced visceral hypersensitivity at all the distension pressures of colorectal distension (CRD; from 15 to 60 mmHg). Data are expressed as means ± SEM (n=9 for the ‘vehicle’ group [] and ‘PRS + vehicle’ group []; n=7 for ‘PRS + NCDO2118’ group []; n=12 for the ‘PRS + NCDO2118 + glutamate’ group []); (B) Effect of 10-day oral administration of GABA-producing L. lactis NCDO2118 and low GABA-producing L. lactis NCDO2727 (109 CFU per day) in presence of glutamate on PRS-induced visceral hypersensitivity at all the distension pressures of CRD. Data are expressed as means ± SEM (n=9 for the ‘vehicle’ group [] and ‘PRS + vehicle’ group []; n=12 for the ‘PRS + NCDO2118 + glutamate’ group []; n=8 for the ‘PRS + NCDO2727 + glutamate’ group []). *p<0.05, **p<0.01, ***p<0.001 vs. basal values for animals treated with vehicle. +p<0.05, ++p<0.01 vs. values for stressed animals treated with vehicle.

Figure 2—source data 1

Effect of 10-day oral administration of GABA-producing L. lactis NCDO2118 on PRS-induced visceral hypersensitivity.

https://cdn.elifesciences.org/articles/77100/elife-77100-fig2-data1-v1.txt
Figure 2—source data 2

Effect of 10-day oral administration of GABA-producing L. lactis NCDO2118 and low GABA-producing L. lactis NCDO2727 on PRS-induced visceral hypersensitivity.

https://cdn.elifesciences.org/articles/77100/elife-77100-fig2-data2-v1.txt
Figure 2—source data 3

Incidence of 10-day oral administration of L. lactis NCDO2118 in presence of glutamate on basal visceral sensitivity.

https://cdn.elifesciences.org/articles/77100/elife-77100-fig2-data3-v1.txt
Figure 2—source data 4

Incidence of 10-day oral administration of glutamate on basal visceral sensitivity and PRS-induced visceral hypersensitivity.

https://cdn.elifesciences.org/articles/77100/elife-77100-fig2-data4-v1.txt

The antinociceptive effect of L. lactis NCDO2118 is due to its ability to deliver GABA in vivo and is mediated by the activation of GABAB receptors

The growth of the NCDO2118 ΔgadB mutant strain, in which GABA pathway was interrupted (see Experimental section for strain construction details), was close to that of the NCDO2118 strain under same culture conditions (Figure 3—figure supplement 1). However, this strain did not produce GABA (GABA concentration in the bioreactor <0.3 mM Figure 3—figure supplement 1) and had no detectable GAD activity. There was no impact of this strain on the visceral hypersensitivity response induced by stress in response to CRD in presence of glutamate (Figure 3A, p<0.05 NCDO2118 ΔgadB- vs. NCDO2118-treated animals). The beneficial effect of L. lactis NCDO2118 treatment on stress-induced visceral hypersensitivity (P<0.05) was completely abolished when animals received the GABAB receptor antagonist SCH-50911 in presence of glutamate (Figure 3B).

Figure 3 with 1 supplement see all
Download asset Open asset
The anti-nociceptive effect of L. lactis NCDO2118.

(A) Effect of 10-day oral administration of γ-aminobutyric acid (GABA)-producing Lactococcus lactis NCDO2118 and non-GABA producing NCDO2118 ΔgadB mutant (109 CFU per day) on PRS-induced visceral hypersensitivity at all the distension pressures of colorectal distension (CRD; from 15 to 60 mmHg). Data are expressed as means ± SEM (n=9 for the ‘vehicle’ group [] and ‘PRS + vehicle’ group []; n=12 for the ‘PRS + NCDO2118 + glutamate’ group []; n=13 for the ‘PRS + NCDO2118 ΔgadB + glutamate’ group []); (B) Effect of 10-day oral administration of GABA-producing L. lactis NCDO2118 (109 CFU per day) on PRS-induced visceral hypersensitivity at all the distension pressures of CRD in presence or not of GABAB receptor antagonist SCH-50911 (3 mg/kg bw, IP). Data are expressed as means ± SEM (n=9 for the ‘vehicle’ group [] and ‘PRS + vehicle’ group []; n=12 for the ‘PRS + NCDO2118 + glutamate’ group []; n=7 for the ‘PRS + NCDO2118 + glutamate + SCH-50911’ group []). *p<0.05, **p<0.01, ****p<0.0001 vs. basal values for animals treated with vehicle. +p<0.05, +++p<0.001 vs. values for stressed animals treated with vehicle. $p<0.05 vs. for stressed animals treated with NCDO2118 + glutamate.

Figure 3—source data 1

Effect of 10-day oral administration of GABA-producing L. lactis NCDO2118 and non-GABA producing NCDO2118 delta-gadB mutant on PRS-induced visceral hypersensitivity.

https://cdn.elifesciences.org/articles/77100/elife-77100-fig3-data1-v1.txt
Figure 3—source data 2

Effect of 10-day oral administration of GABA-producing L. lactis NCDO2118 on PRS-induced visceral hypersensitivity in presence or not of GABAB receptor antagonist SCH-50911.

https://cdn.elifesciences.org/articles/77100/elife-77100-fig3-data2-v1.txt
Figure 3—source data 3

Table of biomass and GABA concentrations for L. lactis NCDO2118 and its ΔgadB mutant.

https://cdn.elifesciences.org/articles/77100/elife-77100-fig3-data3-v1.xlsx

GABA is produced by L. lactis NCDO2118 under ‘stomach-like’ conditions in vitro and in vivo in the stomach

Experiments were performed in vivo with rats fed L. lactis NCDO2118, vehicle or L. lactis NCDO2727, in the presence of glutamate for 10 days. For the NCDO2118 strain, the GABA level was high in the stomach and then tended to decrease in the other gut compartments (p<0.05 in caecum vs stomach; Figure 4A). In contrast, in vehicle- or NCDO2727-treated animals, the GABA levels were similar all along the gastro-intestinal tract (Figure 4B). Consistently, in the stomach, the GABA production by L. lactis NCDO2118 was higher than that measured in vehicle or NCDO2727 strain (Figure 4C). To reinforce our findings on the potential role of the gastric region, we performed further experiments in vitro under ‘stomach-like’ conditions in the presence of glutamate (at pH = 4.6 in acetate buffer or gastric juice sampled from naive rats). In particular in rat gastric juice, we found that the rate of GABA production by the NCDO2727 strain was similar to the vehicle but the rate of GABA production by the NCDO2118 strain was fourfold increased (Supplementary file 2).

Figure 4
Download asset Open asset
L. lactis effect on the GABA production all along the gastro-intestinal tract.

(A) Variation of the γ-aminobutyric acid (GABA) concentration all along the gastro-intestinal tract for the group ‘NCDO2118 + glutamate’ () vs. the group ‘vehicle’ () after a 10-day daily oral administration at 109 CFU per day. (B) Variation of the GABA concentration all along the gastro-intestinal tract for the group ‘NCDO22727 + glutamate’ () vs. group ‘vehicle’ () after a 10-day daily oral administration at 109 CFU per day. (C) Variation of the GABA concentration in the gastric content for the group ‘vehicle’ (), the group ‘NCDO2118 + glutamate’ (), and the group ‘NCDO2727 + glutamate’ () after a 10-day daily oral administration at 109 CFU per day. Data are expressed as means ± SEM. Non-parametric Kruskal-Wallis test, supplemented by Dunn’s multiple comparison test, revealed no statistical differences between samples except caecum vs. stomach for the NCDO2118 strain (*p<0.05).

Figure 4—source data 1

Variation of the GABA concentration all along the gastro-intestinal tract for NCDO2118+glutamate.

https://cdn.elifesciences.org/articles/77100/elife-77100-fig4-data1-v1.txt

Oral treatment with L. lactis NCDO2118 has minor influence on faecal microbiota abundance, composition, and diversity but influences specific genera

The composition of the faecal microbiota was not globally impacted by the L. lactis NCDO2118 treatment (Figure 5). Particularly, no changes in α and β diversities were observed (Figure 5A and C). Besides, the non-parametric Kruskal-Wallis test on each of the 150 identified genera present in samples revealed few differences in taxon abundances in faecal microbiota, in particular significantly for three genera, Clostridium (p=0.023), Frisingicoccus, and Papillibacter (p=0.041), and nearly significantly for Escherichia genus (p=0.069), after L. lactis NCDO2118 treatment (Supplementary file 3).

Figure 5
Download asset Open asset
Overview of the faecal microbiota after a 10-day daily oral administration of γ-aminobutyric acid (GABA)-producing L. lactis NCDO2118 (109 CFU per day) in presence of glutamate vs. ‘vehicle’ group.

(A) Alpha diversity within faecal samples. Two measures are shown: mean observed number of OTUs per sample, an estimate of richness (left panel), and Shannon Index, indicating the evenness of the sample (right sample). One-way ANOVA followed by the post hoc pairwise Tukey’s test revealed no statistical differences between samples (p>0.05). (B) Top 15 dominant bacterial genera in faecal samples; (C) MDS ordination plot of the Bray-Curtis distance between samples, a representation of phylogenetic similarity. PERMANOVA revealed no statistical differences between samples (p>0.05).

Discussion

For the first time, we demonstrated that a transient food-borne LAB L. lactis (strain NCDO2118) exerts, in an animal model of acute stress, visceral anti-hypersensitivity effects via its ability to produce significant levels of GABA in vivo, due to an active GAD enzymatic potential and not solely to the presence of gadB gene. Indeed, the low GABA producer NCDO2727 strain and NCDO2118 ΔgadB mutant strain unable to produce GABA, fail to exert such antinociceptive effect. This effect is mediated by the metabolically active NCDO2118 which delivers GABA in the gastro-intestinal tract, that in turn influences host GABAergic host physiological system.

L. lactis is a very common bacterium widely used in food industry with a GRAS status. Bacteria can metabolize, via decarboxylation mechanisms, several amino acids to synthetize active amino compounds. In particular, GAD catalyses the irreversible α-decarboxylation of glutamate into GABA. Even though genes encoding GAD are found in numerous bacteria present in different environments and ecosystems, i.e., native soil, natural landscape as well as in the gastro-intestinal tract, the ability of these bacteria to produce GABA has not systematically been evidenced. Interestingly, among L. lactis bacteria studied here, NCDO2727 was qualified as a low producer of GABA and NCDO2118 as a high producer. Although these two strains harbour similar GAD encoding gadB gene, they differed significantly at the level of their GAD activity. Because the promoters of gad genes are nearly identical in the two strains, transcriptional regulations should not explain the observed differences. The growth defect of NCDO2727 should also not be involved since higher growth of the strain in another culture medium did not restore a high GABA production. Herein, we demonstrate in vivo that only a 10-day chronic oral administration of the high producer NCDO2118 strain displaying a high GAD activity is able to reduce visceral hypersensitivity induced by stress. This effect was accompanied by an increased GABA production in the gastric compartment. Taken together, all these observations suggest that transient NCDO2118 strain is metabolically active in vivo to deliver GABA in the gastro-intestinal tract lumen. We evaluated the impact of the NCDO2118 treatment on the faecal microbiota and observed no changes in α or β diversity. This result is consistent with observations in many other studies on IBS that have shown that probiotic treatment did not shift the overall diversity of the intestinal microbiota of mice, rats, or humans (Cussotto et al., 2021; Grazul et al., 2016; Jeffery et al., 2020; Labus et al., 2017; Tap et al., 2017). Indeed, in ‘healthy microbiota’ the addition of a probiotic does not promptly impact the resident microbial populations but rather could have a significant impact on reshaping the microbiota in an already existing dysbiosis humans (Alander et al., 1999; Eloe-Fadrosh et al., 2015; Lee et al., 2004). We next compared genera abundancies after treatment using non-parametric tests and found an enrichment of the Clostridium genus after L. lactis NCDO2118 treatment compared to the vehicle group. This genus was shown in silico as modulating GABA (i.e. GABA producer and consumer) in humans (Strandwitz et al., 2019). An enrichment of the Escherichia genus was also observed. Interestingly, in the study on the E. coli Nissle 1917 strain, Pérez-Berezo and colleagues showed that GABA was not able to directly cross the epithelial barrier (Pérez-Berezo et al., 2017), requiring the C12AsnGABAOH lipopeptide for functionalization and subsequent translocation. It is then possible that the enrichment of the genus Escherichia in the faecal microbiota of L. lactis-treated animals could help reduce visceral pain via a lipopeptide-mediated GABA functionalization pathway. Of note, in our study, using the same method as previously described (Pérez-Berezo et al., 2017), we were not able to detect any C12AsnGABAOH lipopeptide in L. lactis NCDO2118 (results not shown). Other yet unidentified lipopeptides could be at play, with the possible contribution of other bacteria capable of producing such functional GABA. Altogether, further investigations are needed for a better understanding of the site(s) of interaction between luminal GABA delivery by L. lactis NCDO2118 and the gastro-intestinal barrier and its environment.

GABA is the major inhibitory neuromediator, mainly involved in regulating physiological and psychological responses. Today, indirect evidences point out the effect of gut microbiota as well as probiotic strains (Cryan and Dinan, 2012) on CNS by modulating the GABAergic system. Moreover, alterations of this GABAergic system have a key role in the development of stress-related psychiatric disorders (Schür et al., 2016). Converging evidence, using genetic and pharmacological approaches, illustrates the important role of GABAB receptors in anxiety disorders. Using a specific antagonist (SCH-50911), we demonstrated the involvement of GABAB receptor in the visceral antinociceptive effect mediated by the delivery of GABA produced by the NCDO2118 strain. GABAB receptors are strongly expressed in the gastro-intestinal tract (Hyland and Cryan, 2010) and widely distributed from the stomach to the ileum in the enteric nervous system (ENS; Auteri et al., 2015). In rodents, previous studies established the presence of autocrine and paracrine GABA signalling mechanisms in gastro-intestinal epithelial cells via GABA receptor activation. In rat, GABAB receptors have also been identified on the mucosal gland cells of the stomach (Castelli et al., 1999; Erdo and Bowery, 1986; Erdö et al., 1990; Krantis et al., 1994). Actually, the GABAB metabotropic receptors in the gastro-intestinal tract are known to regulate several gut functions and gut-to-brain signalling pathway (Hyland and Cryan, 2010). In mice, selective stimulation of GABAB receptors increases gastric acid secretion both through vagal cholinergic and gastrin-dependent mechanisms (Piqueras and Martinez, 2004). These combined central and peripheral mechanisms result in an increase in the luminal acidity, which provides a supportive environment for the GAD activity of L. lactis NCDO2118. Accordingly, and despite a limited number of animals, we observed herein a significant level of GABA production measured after 10-day administration of L. lactis NCDO2118 in the stomach, which was supported by an in vitro GABA production in gastric juice sampled from naive rats and supplemented with glutamate; in this way, a dynamic ‘virtuous circle’ would be generated between L. lactis NCDO2118 and the host. Furthermore, vagal and splanchnic afferents of GABAB receptors are involved in the modulation of sensitivity (Hyland and Cryan, 2010). Probiotic strains have been described to modulate brain functions via a dependent vagus activation pathway (Bravo et al., 2011). This study was the first to demonstrate the ability of Lactobacillus to modify central levels of GABA through the stimulation of the vagus nerve. This central GABA release goes hand with beneficial effect on emotional behaviour impairments induced by stress. However, the primary molecular mechanisms underlying how Lactobacillus stimulated vagal afferents were not resolved. Here, we could firmly demonstrate that a neurotransmitter, i.e., GABA, directly produced by L. lactis NCDO2118 into the gastro-intestinal tract lumen, exerted beneficial effect on stress-induced gut visceral hypersensitivity via GABAB receptor stimulation. Even though we do not investigate the activation of the ENS or the CNS herein, our results associated with previous published data bring strengths on the table in the functional communication between bacteria, gut, and brain.

In conclusion, our data strongly suggest that, according to its ability to deliver into the gastro-intestinal tract lumen neurometabolites like GABA in significant levels, L. lactis NCDO2118 could be considered as a helpful candidate in the management of functional gastro-intestinal disorders associated with stressful situations. Regarding our results, GAD activity rather than genetic organization of gad genes appears as a key determinant for in vivo antinociceptive properties of L. lactis. Therefore, this work opens perspectives for GRAS L. lactis strains as future therapeutic agents for the management of visceral pain and the anxious profile of IBS patients.

Materials and methods

Bacterial strains, medium, and culture conditions

Request a detailed protocol

The strains NCDO2118 and NCDO2727, two L. lactis subsp. lactis from vegetable origin, were used throughout this study. L. lactis NCDO2118 ΔgadB mutant was constructed by double crossing over in the chromosome as previously described (Maguin et al., 1996). Briefly, two fragments upstream and downstream of the gadB coding sequence were PCR amplified, fused by overlapping PCR, and cloned in the pGhost9 vector using Gibson assembly method (New Englang Biolabs). Primers are listed in Supplementary file 4.

Bacterial cultures were performed in duplicate in 2 L Biostat B-plus bioreactor (Sartorius, Melsungen, Germany) in M17 supplemented with 55 mM (8 g/L) glutamate, 29 mM (5 g/L) arginine, 250 mM (45 g/L) glucose, and 300 mM NaCl. Cultures were incubated at 30°C. Fermentations were carried out under oxygen-limiting conditions. pH was maintained at 6.6 by KOH addition for 8 hr, then pH was dropped and regulated at 4.6. Culture was inoculated with cells from pre-cultures grown in Erlenmeyer flask on similar medium, harvested during the exponential phase, and concentrated in order to obtain an initial optical density at 580 nm of 0.25 in the fermenter. Bacterial growth was estimated by spectrophotometric measurements at 580 nm (Libra S11, Biochrom, BIOSERV, Massy, France; 1 Unity of absorbance is equivalent to 0.3 g dry weight/L). Samples were collected every 30 min for HPLC (High Performance Liquid Chromatography) measurement of GABA concentration in the growth medium. For in vivo assays and measurement of bacterial strain activity in the gastric juice in vitro, cells were harvested before the pH modification (i.e. at 7 hr for NCDO2118 and NCDO2727). The culture volume required for approximately 3 × 1011 CFU (colony-forming units) was centrifuged to pellet the bacterial cells. Then, cells were washed and suspended in 0.9% (m/v) NaCl containing 15% glycerol (v/v) to a final concentration of 109 CFU/mL. For GAD activity measurements, 150 mg of bacterial cells were harvested at 7 hr and 24 hr of culture.

Sequence analysis of gad operon

Request a detailed protocol

In the NCDO2118 strain, the gad operon sequence was extracted from the chromosome sequence deposited in NCBI-GenBank database under the accession number CP009054. The gad operon in NCDO2727 strain was sequenced and deposited in NCBI-GenBank database under the accession number MK225577. Briefly, DNA was amplified using two primers, GadSeq_F (5’ –TCCAGAAATAACAGCTACATTGACATAATG –3’) and GadSeq_R (5’– TAACAGCCCCATTATCTAAGATTACTCC –3’) and the Q5 High-Fidelity DNA polymerase (NEB) and sequenced by Eurofins Genomics (primers listed in Supplementary file 4). Sequences of gad operon were compared by Blast alignment algorithm.

Animals and surgical procedure

Request a detailed protocol

Adult female Wistar rats (200–225 g) were purchased from Janvier Labs (Le Genest St Isle, France) and individually housed in polypropylene cages under standard conditions (temperature 22 ± 2°C and a 12 hr light/dark cycle) with free access to water and food (standard pellets 2016, Envigo RMS SARL, Gannat, France). All experiments were approved by the Local Animal Care and Use Committee (APAFiS#5577–201606061639777 v3, see Colorectal and Under PRS conditions sections, and APAFiS#14898–2018043016031426, see Without PRS) in compliance with European directive 2010/63/UE.

Colorectal distension procedure and acute stress procedure

Request a detailed protocol

Under general anaesthesia by intraperitoneal administration of 0.6 mg/kg acepromazine (calmivet, Vetoquinol, Lure, France) and 120 mg/kg ketamine (Imalgene 1000, Merial, Lyon, France), rats were equipped with NiCr wire electrodes implanted in the abdominal striated muscle for EMG recording (Morteau et al., 1994). Animals were then accustomed to be in polypropylene tunnels for several days before CRD. A 4-cm long latex balloon, fixed on rigid catheter was used. CRD was performed after insertion of the balloon in the rectum at 1 cm from the anus. The tube was fixed at the basis of the tail. Isobaric distensions of the colon were performed from 0 to 60 mmHg using a Distender Series IIR Barostat (G&J Electronics Inc, Toronto, Canada) with each distension step lasting 5 min. The striated muscle spike bursts, related to abdominal cramps, were recorded on an electroencephalograph machine (Mini VIII, Alvar, Paris, France).

PRS, a relatively mild non-ulcerogenic model of stress, was performed as previously described (Williams et al., 1988). Briefly, rats were sedated with diethyl-ether and their fore shoulders, upper forelimbs and thoracic trunk were wrapped in a confining harness of paper tape to restrict, but not prevent, body movements. Rats were then placed in their home cage for 2 hr.

Experimental protocol for in vivo assays

Under PRS conditions

Request a detailed protocol

Series of experiments, based on a 10-day treatment by oral gavage (as detailed below), were conducted using, for each series, three groups of 7–12 female rats equipped for EMG. For all oral treatments used, basal and post-PRS (20 min after the 2 h PRS session) abdominal responses to CRD were recorded on day 9 and day 10, respectively.

In the first series of experiments, groups of rats were orally treated for 10 days (1 mL/rat) with L. lactis NCDO2118 (109 CFU/day) plus glutamate (0.2% [w/v]) or not, L. lactis NCDO2727 plus glutamate (0.2% [w/v]) or vehicle (NaCl 0.9% [w/v] + glycerol 15% [v/v]). In the second series of experiments, rats were divided into three groups and orally treated for 10 days (1 mL/rat) with L. lactis NCDO2118 (109 CFU/day) plus glutamate (0.2% [w/v]), L. lactis NCDO2118 plus glutamate (0.2% [w/v]), and GABAB receptor antagonist (2 S) (+)–5,5-dimethyl-2-morpholineacetic acid (SCH-50911, Sigma-Aldrich SML1040; 3 mg/kg body weight, IP 20 min before the PRS session) or vehicle (NaCl 0.9% [w/v] + glycerol 15% [v/v]). In the last series of experiments, rats were divided into three groups and orally treated for 10 days (1 mL/rat) with L. lactis NCDO2118 (109 CFU/day) plus glutamate (0.2% [w/v]), L. lactis NCDO2118 ΔgadB (109 CFU/day) plus glutamate (0.2% [w/v]) or vehicle (NaCl 0.9% [w/v] + glycerol 15% [v/v]).

Without PRS

Request a detailed protocol

In order to determine GABA levels at different locations of the gastro-intestinal tract, further experiments without PRS application were performed as follows: three groups of rats (n=5 per group) were orally treated for 10 days (1 mL/rat) with L. lactis NCDO2118 (109 CFU/day) plus glutamate (0.2% [w/v]), L. lactis NCDO2727 plus glutamate (0.2% [w/v]), or vehicle (NaCl 0.9% [w/v] + glycerol 15% [v/v]). After the last administration, deep inhalation anaesthesia with Isoflurane, followed by terminal aortic blood sampling, was performed. After the sacrifice, gastric, ileal, caecal, and colonic contents, as well as faeces, were collected for GABA measurements (see GABA extraction and quantification).

GAD activity and GABA production by L. lactis cells

Request a detailed protocol

For GAD activity measurement, 150 mg bacterial cells were washed twice with 0.2% KCl (w/v) and suspended in 3 mL sodium acetate buffer (100 mM, pH 4.6) containing 4.5 mM MgCl2, 22% (v/v) glycerol, and 1.5 mM dithiothreitol. This mixture was distributed into three tubes containing 6 mg of glass beads. Then, cells were disrupted in a FastPrep-24 homogenizer (MP Biomedicals, Illkirch, France) using six cycles of 30 s at 6.5 m/s interrupted by 1 min incubation on ice. Cell debris was removed by centrifugation for 15 min at 10,000 g and 4°C. The supernatant was used for enzyme assays, and the protein concentration of the extract was determined by the Bradford method. Enzyme assay was realized with 0.5 mL of substrate solution, consisting of 20 mM sodium glutamate, 2 mM PLP incubated at 30°C then mixed with 0.5 mL supernatant. Every 30 min until 4 hr, 100 µL were sampled and inactivated by boiling for 5 min to stop the decarboxylation reaction. Reaction mixtures were subsequently analysed for the presence of GABA using HPLC.

The kinetics of GABA production by L. lactis NCDO2118 and L. lactis NCDO2727 was monitored in vitro at 37°C as follows: 0.5 mL of each strain (109 CFU/mL) or the NaCl/glycerol vehicle was added to 1 mL of 100 mM acetate buffer pH = 4.6 or gastric content from naive rat and 0.5 mL glutamate 0.4% (w/v). 100 µL of assay mixture were sampled at regular time intervals (until 60 min) and inactivated by boiling to stop the decarboxylation reaction. Reaction mixtures were subsequently analysed using HPLC. The initial rate was determined as the amount glutamate converted into GABA (µmol) per minute.

GABA extraction and quantification

Request a detailed protocol

GABA concentration in culture supernatant or in reaction mixtures, associated to assays for GAD activity and GABA production, was measured by HPLC (Agilent Technologies 1200 Series, Waldbronn, Germany) as previously described (Laroute et al., 2016).

GABA measurements were realised in samples from different digestive tract locations (gastric, ileal, caecal, and colonic contents and faeces). GABA was extracted from 200 mg of contents with methanol (three times with 3 mL) at room temperature. The mixture was centrifuged and the supernatant was transferred into tube and concentrated to dryness. Dried samples were resuspended in 100 µL methanol then analysed by HPLC as described above.

Faecal microbiota composition using 16S rRNA gene sequencing

Request a detailed protocol

Faeces were collected at the end of the 10-day oral treatment, just before the 2 hr PRS session (see Under PRS conditions). Faecal samples were stored at –80°C until DNA was extracted using the ZymoBIOMICS DNA Miniprep Kit (D4300, Zymo Research) following manufacturer’s instructions. The 314 F/805 R primers (5′ GACTACHVGGGTATCTAATCC-Forward primer), 5′ GACTACHVGGGTATCTAATCC-Reverse primer) were used to amplify the V3-V4 variable regions of the 16 S rRNA gene. Forward primer and reverse primer carried overhang adapters (5′ CTTTCCCTACACGACGCTCTTCCGATCT-Forward primer, (5′ GGAGTTCAGACGTGTGCTCTTCCGATCT-Reverse primer) for Illumina index and sequencing adapters. First round of PCR was carried in a single-step 30 cycles using the MTP Taq DNA Polymerase Kit (D7442, Sigma-Aldrich) under the following conditions: 94°C for 1 min, followed by 30 cycles of 94°C for 1 min, 65°C for 1 min, and 72°C for 1 min, after which a final elongation step at 72°C for 10 min was performed. Second-round amplicons libraries and sequencing were performed at the Sequencing Platform of Toulouse (GeT-Biopuces) on an Illumina-MiSeq following the manufacturer’s guidelines.

Bioinformatic treatments of all samples were performed with the Find Rapidly OTUs with Galaxy Solution pipeline (Escudié et al., 2018). Briefly, reads were contigued with the VSEARCH software and sequences with sizes inferior to 300 bp or superior to 700 bp were eliminated. A clustering at 97% similarity was used to define OTUs with the SWARM algorithm and chimeric sequences were also removed. Sequences were then filtered using the phiX contaminant databank and the OTUs presenting a frequency inferior to 0.005% on all samples and present in less than three samples were also removed (Bokulich et al., 2013). The R package ANOMALY (Fisch et al., 2022) was used for the assignment of the taxonomic classification of the representative sequences of each OTU. For that, the algorithm IDTAXA and the reference databases SILVA 138 16 S and GTDB bac120_arc22 were used (Parks et al., 2020; Quast et al., 2013). Alpha (α) and beta (β) diversities and differential abundancies analyses were performed in R using the Phyloseq package (McMurdie et al., 2013). Alpha diversity metrics (i.e. observed OTUs and Shannon Index) were calculated based on the genus level. One-way ANOVA was used to test α diversity metric dissimilarities with post hoc Tukey’s test. The Bray-Curtis distance was used to measure β diversity metrics. We explored the community structure of the samples with PERMANOVA (Kelly et al., 2015).

Statistical methods

Request a detailed protocol

The software GraphPad Prism 9.1 (GraphPad, San Diego, CA) was used for other statistical analyses. For animal experiments, data are reported as the means ± SEM. One-way ANOVA, followed by Tukey’s Multiple Comparison test, was performed to compare data between the different groups of animals. For GABA measurements, data are reported as the means ± SEM and the non-parametric Kruskal-Wallis test, supplemented by Dunn’s multiple comparison test, was used. Statistical significance was accepted at p<0.05.

Data availability

Relevant additional data files are provided as source data files. For figures 1A 1B and S3: Excel file. For figures 2A 2B, 3A 3B, 4, S1 and S2: txt format. For figure 5: All data (raw and treated) can be found in this link: https://forgemia.inra.fr/umrf/exploremetabar.

References

    1. Dean BB
    2. Aguilar D
    3. Barghout V
    (2005)
    Impairment in work productivity and health-related quality of life in patients with IBS
    The American Journal of Managed Care 11:S17–S26.
    1. Reid G
    (2011) Neuroactive probiotics
    BioEssays: News and Reviews in Molecular, Cellular and Developmental Biology 33:562.
    https://doi.org/10.1002/bies.201100074

Decision letter

  1. Gisela Storz
    Senior and Reviewing Editor; National Institute of Child Health and Human Development, United States

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting your work entitled "GABA delivery by Lactococcus lactis counteracts stress-induced gut hypersensitivity via GABAB receptor activation in rat" for consideration by eLife. Your article has been reviewed by two peer reviewers, and the evaluation has been overseen by a Senior Editor. The reviewers opted to remain anonymous.

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered further for publication in eLife.

Overall there was appreciation for your study of the neuromodulatory effects of Lactococcus lactis. However, both reviewers thought additional experiments were required to raise the manuscript to eLife standards. A significantly revised manuscript that addresses the reviewers comments below and further dissects the mechanisms by which L. lactis affects the neuronal functions could be of interest to a broad audience.

Reviewer #1:

The manuscript is original, well discussed and it touches a topic poorly explored so far which is the neuromodulatory effects of L.lactis and other lactic acid bacteria in vivo. However, as it stands, the work will need additional results to be suitable for publication by eLife (considering the journal standards). There are already reports on the neuronal effects of GABA secreted by acid lactic bacteria (such as Lactobacillus and Streptococci) and Bifidobacterium. The novelty of this present study would be to dissect the mechanisms by which luminal bacteria would affect neuronal functions in the GI tract.

1) The results of GABA measurements in the fecal content clearly show that feeding glutamate alone results in GABA increase in the feces. It is possible that lactic bacteria in the gut microbiota play a role in GABA production. L.lactis NCDO2118, but not L.lactis NCDO2727, may have an impact in gut microbiota composition changing the abundance of bacteria that help the conversion of glutamate into GABA. This distinctive effect of NCDO2118 strain may be related to its high intracellular GAD activity (as compared to NCDO2727) and the role of high amounts of GABA (produced by this strain) in the expansion of other GABA-producing strains in microbiota. This would be compatible with the result obtained with the deficient-GABAB mutant strain of L.lactis NCDO2118. The role of microbiota in the effects reported by the manuscript needs to be investigated.

2) There is no data in the manuscript on the mechanisms by which GABA exerted its visceral anti-hypersensitivity effect during IBS model. Does bacteria-derived GABA access the enteric nervous system (ENS)? How GABA crosses the epithelial barrier? Is GABA binding to GABARB receptors in the neurons or the epithelia (neuroendocrine cells)? GABARB receptors have been identified in both myenteric and submucosal neurons. In the rat mucosal epithelium, positive cells for GABARB have been also detected by immunohistochemistry studies along the length of the GI tract (from the gastric mucosa to the colon), but it has been proposed that the effects of GABA in these receptors differ since neuroendocrine cells are also producers of somatotastin in the gastric epithelium, but GABAR+ positive cells in other parts of the GI tract (such as the duodenum) produced serotonin as well. The novelty of the study would be to shed some light on the mechanisms behind the antinociceptive effects of GABA secreted by luminal L. lactis.

Reviewer #2:

The aim of this manuscript is to demonstrate that a strain delivering GABA (L. lactis NCDO2118) in the gut is able exert to anti-nociceptive properties in a stress-induced visceral hypersensitivity rat model. This study is well-conducted, based on strong in vivo evidences. They particularly managed to established that the GAD activity from L. lactis and the subsequent GABAergic signaling is a key mechanism in the control of visceral hypersensitivity induced by a stress. Nevertheless, I would have 2 majors comments to make (more detailed above) about (1) the impact on anxiety-like behaviors induced by the stress and (2) the bacteria culture conditions and bacteria growth.

(1) The authors present in their manuscript behavior defects associated with abdominal pains and IBS, is it possible to mimic such symptoms (anxiety, depression,…) using their stress model and if yes, does the NCDO2118 strain can reverse such model induced behavior defects in rats?

(2) The authors present data demonstrating that the NCDO2118 strain produced more GABA in certain culture conditions (pH 4,6) in comparison to the NCDO2727 strain (Figure 1A), but my concern is about the biomass corresponding to the bacterial growth. In fact, when we look at the Figure 1B, we can see that only the NCDO2118 strain present an increase of the biomass, which is almost not the case for the NCDO2727 strain. The authors should comment on this. Does such growth defect could induce a bias in the in vivo observed data using such bacteria culture? In addition, the authors should also give similar data about the NCDO2118ΔgadB strain, just to make sure that such deletion did not induce bacterial growth defect as the NCDO2727 strain. Finally, about the in vivo experiments, I would be interested to see what would be the results on colonic sensitivity using killed bacteria (heat killed for example).

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "Lactococcus lactis NCDO2118 exerts visceral antinociceptive properties in rat: increase in GABA concentration and activation of GABAB receptor in the gastro-intestinal tract" for further consideration by eLife. Your revised article has been evaluated by Gisela Storz (Senior Editor) and a Reviewing Editor.

The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

Essential revisions:

1) Incorporate additional controls as outlined in detail in the individual reports are critical.

2) Discuss the significance/lack of significance of data.

3) Identify which cell types are involved in the beneficial effect on visceral hypersensitivity.

4) Determine why only 2118 but not 2727 other natural L. lactis strains can produce GABA.

Reviewer #1

Laroute et al. investigated the potential anti-hypersensitive properties of a Generally Recognized As Safe (GRAS) lactic acid bacterium, the Lactococcus lactis food-borne bacterium. They have first demonstrated that, among the L. lactis bacteria, only some of them are able to produce active GABA, even if the encoding genes are present in the genome. In addition, those GABA-producing bacteria are able to relieve stress-induced visceral hypersensitivity. Finally, this anti-hypersensitive effect is dependent on the bacteria-produced GABA since genetic and pharmacological inhibition of the GABA leads to a loss of its ability to relieve the stress-induced visceral hypersensitivity.

In general, the presented data are really convincing, and I would need some more controls, in particular in in vivo experiments:

1. The in vitro data presented in figure 1A nicely demonstrate the difference in the GABA production between 2 strains of L. lactis (NCDO2118 and NCDO2727). I just have a concern about the figure 1B since it seems the non GABA-producing bacteria (NCD02727) present a defect in its growth in comparison to the GABA-producing L. lactis (NCDO2118). May it be a reason why the NCDO2727 did not produce GABA (different growth stage in the bacteria culture) and consequently it loses its in vivo properties on visceral hypersensitivity?

2. The in vivo results are beautiful and really convincing to me (Figures 2 and 3). I just have the feeling some controls are missing to reinforce the results. On figure 2, I would like to see the effect of the NCDO2118 strain + glutamate in control animals to make sure it will not reduce the basal visceral hypersensitivity? On the figure 3, what happened in PRS + Vehicle + inhibitor, to make sure that the inhibitor does not affect sensitivity by itself?

Finally, I feel that the weakness of the study is in the last figures 4 and 5, since, in my point of view, it did not present as strong data as on the figures 1-3. On the Figure 4, I would appreciate if the authors could increase the n per group since we ca not really see significant differences for the GABA concentration along the GI tract, and it should be more detailed in the discussion. On the figure 5 about microbiota analysis, the authors discussed some differences that are not "significant" as for Escherichia genus since others are not discussed (despite the significance).

Reviewer #2:

Lactococcus lactis is a probiotic bacteria that has shown to have multiple beneficial effects on the gut, including anti-inflammatory effects in models of inflammatory bowel diseases. Irritable bowel syndrome (IBS) is characterized by visceral pain. It would be important to define ways to treat IBS and other gut-related disease using gut microbes. The authors have previously found that the L. lactis strain NCDO02118 could synthesize GABA, depending on culture conditions (Laroute et al., Microorganisms 2021; Laroute et al., Front. Microbiol 2016), but the effects in vivo in animal models of visceral pain had not been well studied. This study tested two strains of Lactococcus lactis for their ability to produce GABA, and if oral gavage of the strains could modulate visceral hypersensitivity in rats. The authors showed that the strain NCDO02118 could produce GABA in vitro and reduce visceral hypersensitivity (measured by colorectal distension) induce by restraint stress. The other strain NCDO02727 did not produce GABA in vitro and had no effect on visceral hypersensitivity. The antinociceptive effect from NCDO02118 depended on expression of GadB. Furthermore, the blockade of visceral hypersensitivity could be blocked with a GABAB receptor antagonist, indicating that this receptor mediated pain blockade. GABA levels were also measured in different segments of the gut following oral administration. NCDO02118 administration did not have major impacts on fecal microbiome composition. This study provides evidence that a probiotic delivered into GI tract can alleviate visceral pain and that this could be related to production of GABA.

Strengths:

This study is one of the first to show that deficiency in glutamate dehydroxylase (GAD) component (GadB) in a bacterial strain regulates GABA production and that this has a physiological effect on visceral pain phenotypes in rats. The result that the NCDO02118 strain has antinociceptive properties that are dependent on the GABAB receptors are interesting and has therapeutic implications for IBS and other diseases.

Weaknesses:

The authors suggest that the NCDO02118 strain produces GABA in vivo, however, the data from this experiment was not significantly different when GABA levels were measured. The host mechanisms involved blockade of pain in vivo are also unclear. For example, the cell types that express GABA receptors can be better characterized and whether GABA signals in specific ways to block the perception of pain. This study could also be strengthened by characterizing why only 2118 but not 2727 or other natural L. lactis strains that can produce GABA, although both of them carry genes that are responsible for GABA metabolism.

Recommendations for the authors:

1. The authors state the NCDO2118 strain increased GABA in vivo, however the data in Figure 4 shows a significant decrease of GABA in the cecum and no significant differences along the rest of the GI tract in NCD02118+glutamate vs. vehicle groups (Figure 4A-C). The authors also state that the NDCO2118 strain has a higher concentration of GABA in the stomach in Figure 4C, but this is not a significant difference either. How does this lack of differences correlate with visceral pain blockade? Is it due to poor colonization of NCDO02118 after gavage or improper microenvironment?

2. The authors observed higher GABA in the stomach compared with cecum (Figure 4A). However, the colorectal distension assay was performed in the colon and rectum. How does local GABA upregulation in stomach contribute to the alleviation of visceral pain in the colon? Are circulating GABA levels changed?

3. The authors show that GABAB receptor antagonist blocks the beneficial effect of the NCDO02118 strain on visceral hypersensitivity, but the authors do not try to identify which cell types are involved in this mechanism. Enteric, vagal, and spinal neurons that innervate the gut all have GABA receptors and identifying which subtype is involved would strengthen the study.

4. The authors show an interesting phenotype that the NCDO02727 strain does not produce GABA even though it has the proper machinery. This study does not explore why this strain is not able to make GABA. Can the authors reveal some mechanistic insight into the differences between NCDO02727 vs. NCDO2118?

5. Do gut anti-inflammatory properties of NCDO02118 contribute to their antinociceptive effect in response to PRS? NCDO02118 admininstration induces the upregulation of IL-10 (DOI: 10.3389/fmicb.2021.623920), which can block pain. Is IL-10 also induced in these mice?

6. The authors suggest that L. lactis may be a treatment option for visceral pain and anxiety associated with irritable bowel syndrome (IBS). However, the study does not test whether L. lactis has any effect on anxiety behavior. Can the authors test anxiety behaviors?

https://doi.org/10.7554/eLife.77100.sa1

Author response

[Editors’ note: the authors resubmitted a revised version of the paper for consideration. What follows is the authors’ response to the first round of review.]

Reviewer #1:

The manuscript is original, well discussed and it touches a topic poorly explored so far which is the neuromodulatory effects of L.lactis and other lactic acid bacteria in vivo. However, as it stands, the work will need additional results to be suitable for publication by eLife (considering the journal standards). There are already reports on the neuronal effects of GABA secreted by acid lactic bacteria (such as Lactobacillus and Streptococci) and Bifidobacterium. The novelty of this present study would be to dissect the mechanisms by which luminal bacteria would affect neuronal functions in the GI tract.

1) The results of GABA measurements in the fecal content clearly show that feeding glutamate alone results in GABA increase in the feces. It is possible that lactic bacteria in the gut microbiota play a role in GABA production. L.lactis NCDO2118, but not L.lactis NCDO2727, may have an impact in gut microbiota composition changing the abundance of bacteria that help the conversion of glutamate into GABA. This distinctive effect of NCDO2118 strain may be related to its high intracellular GAD activity (as compared to NCDO2727) and the role of high amounts of GABA (produced by this strain) in the expansion of other GABA-producing strains in microbiota. This would be compatible with the result obtained with the deficient-GABAB mutant strain of L.lactis NCDO2118. The role of microbiota in the effects reported by the manuscript needs to be investigated.

We have performed faecal microbiota analyses after L. lactis NCDO2118 treatment and its control using 16S rRNA gene sequencing. The microbiota composition was not globally impacted by the L. lactis NCDO2118 treatment. No changes in the and diversities were observed but few changes at the genus level were detected. These results allowed us to give some additional hypothesis on the potential mechanisms involved, which are discussed in the new version of the manuscript.

2) There is no data in the manuscript on the mechanisms by which GABA exerted its visceral anti-hypersensitivity effect during IBS model. Does bacteria-derived GABA access the enteric nervous system (ENS)? How GABA crosses the epithelial barrier? Is GABA binding to GABARB receptors in the neurons or the epithelia (neuroendocrine cells)? GABARB receptors have been identified in both myenteric and submucosal neurons. In the rat mucosal epithelium, positive cells for GABARB have been also detected by immunohistochemistry studies along the length of the GI tract (from the gastric mucosa to the colon), but it has been proposed that the effects of GABA in these receptors differ since neuroendocrine cells are also producers of somatotastin in the gastric epithelium, but GABAR+ positive cells in other parts of the GI tract (such as the duodenum) produced serotonin as well. The novelty of the study would be to shed some light on the mechanisms behind the antinociceptive effects of GABA secreted by luminal L. lactis.

Concerning the GABA delivery by L. lactis NCDO2118, we measured the GABA level in different compartments of the upper and lower gastrointestinal tract and observed an increase in GABA level in the stomach. Based on this very interesting observation, we introduced in the discussion a new concept of a dynamic ‘virtuous circle’ between L. lactis and its host as described below.

In rodents, previous studies established the presence of GABAB receptors on the mucosal gland cells of the stomach (Castelli et al. 1999; Erdo and Bowery 1986; Erdo et al. 1990; Krantis et al. 1994), which when stimulated increase gastric acid secretion both through vagal cholinergic and gastrin-dependent mechanisms (Piqueras et al. 2004). These combined central and peripheral mechanisms result in an increase in the luminal acidity, which provides a supportive environment for the GAD activity of L. lactis NCDO2118.

Reviewer #2:

The aim of this manuscript is to demonstrate that a strain delivering GABA (L. lactis NCDO2118) in the gut is able exert to anti-nociceptive properties in a stress-induced visceral hypersensitivity rat model. This study is well-conducted, based on strong in vivo evidences. They particularly managed to established that the GAD activity from L. lactis and the subsequent GABAergic signaling is a key mechanism in the control of visceral hypersensitivity induced by a stress. Nevertheless, I would have 2 comments to make (more detailed above) about (1) the impact on anxiety-like behaviors induced by the stress and (2) the bacteria culture conditions and bacteria growth.

(1) The authors present in their manuscript behavior defects associated with abdominal pains and IBS, is it possible to mimic such symptoms (anxiety, depression,…) using their stress model and if yes, does the NCDO2118 strain can reverse such model induced behavior defects in rats?

(2) The authors present data demonstrating that the NCDO2118 strain produced more GABA in certain culture conditions (pH 4,6) in comparison to the NCDO2727 strain (Figure 1A), but my concern is about the biomass corresponding to the bacterial growth. In fact, when we look at the Figure 1B, we can see that only the NCDO2118 strain present an increase of the biomass, which is almost not the case for the NCDO2727 strain. The authors should comment on this. Does such growth defect could induce a bias in the in vivo observed data using such bacteria culture? In addition, the authors should also give similar data about the NCDO2118ΔgadB strain, just to make sure that such deletion did not induce bacterial growth defect as the NCDO2727 strain. Finally, about the in vivo experiments, I would be interested to see what would be the results on colonic sensitivity using killed bacteria (heat killed for example).

Whatever the bacterial strain tested and considering the growth profile differences, we systematically adjusted the cell amount in order to give in 1 mL by oral route the same CFU to the rats (10exp9 CFU/day/rat). The text of the manuscript has been revised accordingly to avoid any misunderstanding. Furthermore, the mutation of the gadB gene in L. lactis NCDO2118 has no impact on the cell growth. This is now added in the manuscript (supplementary data Figure S2).

In terms of basic knowledge, our input will undoubtedly open new perspectives for GRAS L. lactis strains as future therapeutic agents for the management of visceral pain and the anxious profile of IBS patients. We consider the most relevant audience for this work is large and multidisciplinary, including researchers in Nutrition, Microbiology, Digestive pathoph ysiology but also clinicians, gastroenterologists and industrial partners working on functional food (probiotics and postbiotics). General public could also be interested for human and animal well-being.

[Editors’ note: what follows is the authors’ response to the second round of review.]

Essential revisions:

1) Incorporate additional controls as outlined in detail in the individual reports are critical.

We have now incorporated additional controls (reviewer 1 point 2). Indeed, we have verified that L. lactis NCDO2118 in presence of glutamate had no impact on basal CRD sensitivity. We have now added a dedicated figure (Figure 2—figure supplement 1) on the supplementary information section. Comments have been included in the main text of the revised manuscript L164-L166. Regarding the effect of the inhibitor on sensitivity by itself, a previous in vivo study (dorsal spinal cord microdialysis) aimed at evaluating the role of GABAB receptor antagonism (using SCH-50911, the same as in our study) on GAT-1 inhibition-induced effects on evoked algesic amino acid (ASP, GLU, GlY) release in the dialysates. This study reported that SCH-50911 administered alone did not affect the evoked release of those amino acids (Smith et al. 2007). Furthermore, according to the 3R rules, our objective was to reduce at the maximum the number of animals used in our in vivo studies. For all these reasons, we did not test a possible effect per se of SCH-50911 in our model.

2) Discuss the significance/lack of significance of data.

In response to this point (reviewer 1 point 3), we have performed additional experiments on in vitro kinetics of GABA production by L. lactis NCDO2118 and L. lactis NCDO2727 in the presence of glutamate under “stomach-like” conditions (at pH=4.6 in acetate buffer or gastric juice sampled from naive rats) (now presented in Supplementary File 2). We clearly demonstrate that, notably in gastric juice sampled from naive rats and supplemented with glutamate, NCDO2118 strain produced GABA. We found that the rate of GABA production by the NCDO2727 strain was similar to the vehicle but the rate of GABA production by the NCDO2118 strain was fourfold increased. To reinforce our findings on the potential role of the gastric region, we have inverted both sets of data in the revised version of the manuscript and slightly modified the in vitro part. Comments have also been added in the Discussion section L351-L354.

3) Identify which cell types are involved in the beneficial effect on visceral hypersensitivity.

As answered to the reviewer (reviewer 2 point 3), in this study we did not identify the cell types involved in the GABA-dependent- antinociceptive properties of NCDO2118. In the gastro-intestinal tract, GABA receptors are present on the vagal afferent fibers both at the mucosal endings that respond to touch and chemical stimuli and muscular endings that respond optimally to mechanical stretch or tension (Page et al. 2002). As discussed in our manuscript, GABAB receptors are also widely distributed from the stomach to the ileum in the enteric nervous system (ENS). In rodents, the presence of GABAB receptors has also been identified on the mucosal gland cells of the stomach. Taken together, all these data suggest complex mechanisms and pathways by which GABA delivered by NCDO2118 strain exerts visceral antinociceptive effect. Based on the literature and our own observations, at this step of our knowledge, we cannot emphasize a cell type more than another and additional investigations are needed to open this black box.

4) Determine why only 2118 but not 2727 other natural L. lactis strains can produce GABA.

We have shown in the manuscript (results L153-L156 and Supplementary File 1) the very strong differences in the intracellular GAD activity in the two strains NCDO2118 and NCDO2727 explaining their major GABA production differences (and also their different antinociceptive properties).

In the manuscript, we have demonstrated by in silico analyses that these differences were not due to differences in the genetic organisation of gadCB operon and gadR gene nor to differences in GadB,C and R protein sequences (L140-L144).

To determine whether these differences in GABA production between strains might be associated to differences in the regulation of gadCB and gadR expression at the transcription level, we have now analysed the nucleotide sequences of their promoters. They were found to be almost identical and the very small number of mutations identified (2 single nucleotide polymorphisms for gadR and 1 for gadCB) seems unlikely to cause large differences in transcriptional regulation (this is described in more details in the comment 4 for the reviewer 2).

We believe that the differences between NCDO2218 and NCDO2727 strains are likely due to posttranscriptional regulations but we do not really know what are the mechanisms involved. Nothing in the literature can help us to formulate realistic hypothesis.

As suggested by the reviewer 1, the growth defect of the NCDO2727 strain could participate to the lack of efficient GABA production. To check the relevance of this mechanism, we have investigated the growth and the GABA production in another culture medium enhancing the growth performances of the NCDO2727 strain. We demonstrated that the GABA production and the growth rate are not correlated (see below in the response to comments 1 and 2 of reviewer 1).

In order to study the gadCB regulation and thanks to reporter gene constructions, we are currently investigating the gadCB expression in various environments and in various strains. Further work will be required to fully understand gad operon regulations and explain the differences observed between the strains.

In the manuscript, in the results and in the Discussion sections, we have added sentences on the promoter similarities (see the comment 4 of the reviewer 2) and also on the lack of relationship between GABA production and bacterial cell growth (see the comment 1 of the reviewer 1).

Reviewer #1

Laroute et al. investigated the potential anti-hypersensitive properties of a Generally Recognized As Safe (GRAS) lactic acid bacterium, the Lactococcus lactis food-borne bacterium. They have first demonstrated that, among the L. lactis bacteria, only some of them are able to produce active GABA, even if the encoding genes are present in the genome. In addition, those GABA-producing bacteria are able to relieve stress-induced visceral hypersensitivity. Finally, this anti-hypersensitive effect is dependent on the bacteria-produced GABA since genetic and pharmacological inhibition of the GABA leads to a loss of its ability to relieve the stress-induced visceral hypersensitivity.

In general, the presented data are really convincing, and I would need some more controls, in particular in in vivo experiments:

1. The in vitro data presented in figure 1A nicely demonstrate the difference in the GABA production between 2 strains of L. lactis (NCDO2118 and NCDO2727). I just have a concern about the figure 1B since it seems the non GABA-producing bacteria (NCD02727) present a defect in its growth in comparison to the GABA-producing L. lactis (NCDO2118). May it be a reason why the NCDO2727 did not produce GABA (different growth stage in the bacteria culture) and consequently it loses its in vivo properties on visceral hypersensitivity?

Indeed, the NCDO2727 strain grows slowly in the medium used in the experiments depicted in our manuscript (i.e. M17 supplemented with glutamate). Since then, we have tested another culture medium, YE medium with glutamate. The cell growth was significantly enhanced (growth rate of 0.7 h-1 compared to 0.24 h-1 in the condition described in the manuscript) but the GABA production remains very low (less than 0.3 mM, which is of the same order of magnitude as the level described here). This result indicates the growth limitation of the strain NCDO2727 is not the reason of its lack of GABA production (and the loss of antinociceptive effect).

In the manuscript, we have added a sentence in the Results section L147: “The GABA concentration did not increase when the biomass production increased during growth of NCDO2727 strain in YE medium with glutamate (data not shown)” and a comment in the discussion L304-L305 “The growth defect of NCDO2727 should also not be involved since higher growth of the strain in another culture medium did not restore a high GABA production.”

2. The in vivo results are beautiful and really convincing to me (Figures 2 and 3). I just have the feeling some controls are missing to reinforce the results. On figure 2, I would like to see the effect of the NCDO2118 strain + glutamate in control animals to make sure it will not reduce the basal visceral hypersensitivity? On the figure 3, what happened in PRS + Vehicle + inhibitor, to make sure that the inhibitor does not affect sensitivity by itself?

We have verified that L. lactis NCDO2118 in presence of glutamate had no impact on basal CRD sensitivity. For clarity reasons, we have added a dedicated figure (Figure 2—figure supplement 1) on the supplementary information section rather than inserting data on Figure 2. Comments have been included in the main text of the revised manuscript L164-L166. Regarding the comment of the reviewer on the effect of the inhibitor on sensitivity by itself, a previous in vivo study (dorsal spinal cord microdialysis) aimed at evaluating the role of GABAB receptor antagonism (using SCH-50911, the same as in our study) on GAT-1 inhibition-induced effects on evoked algesic amino acid (ASP, GLU, GlY) release in the dialysates. This study reported that SCH-50911 administered alone did not affect the evoked release of those amino acids (Smith et al. 2007). Furthermore, according to the 3R rules, our objective was to reduce at the maximum the number of animals used in our in vivo studies. For all these reasons, we did not test a possible effect per se of SCH-50911 in our model.

– Smith CG, Bowery NG, Whitehead KJ. GABA transporter type 1 (GAT-1) uptake inhibition reduces stimulated aspartate and glutamate release in the dorsal spinal cord in vivo via different GABAergic mechanisms. Neuropharmacology. 2007 Dec;53(8):975-81. doi: 10.1016/j.neuropharm.2007.09.008. Epub 2007 Sep 29. PMID: 17981306.

Finally, I feel that the weakness of the study is in the last figures 4 and 5, since, in my point of view, it did not present as strong data as on the figures 1-3. On the Figure 4, I would appreciate if the authors could increase the n per group since we ca not really see significant differences for the GABA concentration along the GI tract, and it should be more detailed in the discussion. On the figure 5 about microbiota analysis, the authors discussed some differences that are not "significant" as for Escherichia genus since others are not discussed (despite the significance).

Our objective was to determine in which GIT compartment GABA delivery specifically occurred for NCDO2118 strain (compared to vehicle and NCDO2727) and even though results did not reach significance, we can see in the stomach that the GABA production by L. lactis NCDO2118 is higher than that measured in vehicle or NCDO2727 strain. Interestingly, in a follow-up study (article in preparation) considering a higher GABA-producing L. lactis strain than NCDO2118, the GABA concentration in the stomach was significantly higher than for NCDO2727 strain and nearly for vehicle (p=0.07).

We have performed additional experiments presented in Supplementary File 2 on in vitro kinetics of GABA production by L. lactis NCDO2118 and L. lactis NCDO2727 in the presence of glutamate under “stomach-like” conditions (at pH=4.6 in acetate buffer or gastric juice sampled from naive rats). We clearly demonstrate that, notably in gastric juice sampled from naive rats and supplemented with glutamate, NCDO2118 strain produced GABA. In particular, we found that the rate of GABA production by the NCDO2727 strain was similar to the vehicle but the rate of GABA production by the NCDO2118 strain was fourfold increased. To reinforce our findings on the potential role of the gastric region, we have inverted both sets of data in the revised version of the manuscript and slightly modified the in vitro part. Comments have been added in the Discussion section L351-L354.

Regarding the impact on the gut microbiota, we fully agree with the reviewer that we observed a significantly increased abundance in Frisingicoccus and Papillibacter genera (p-value of 0.041 for both), complementarily to Clostridium (p-value of 0.023) and Escherichia (p-value of 0.069). According to the reviewer’s comment, we have added these two genera in the description of the results on the gut microbiota, while introducing some slight changes. In particular, the p-values have been added. However, in the discussion part, we still focus on (i) Clostridium genus since it was shown in silico as modulating GABA (i.e. GABA producer and consumer) in humans and (ii) on Escherichia genus since such enrichment in the faecal microbiota of L. lactis-treated animals could help reduce visceral pain via a lipopeptide-mediated GABA functionalization pathway. To the best of our knowledge, nothing is known on the Frisingicoccus and Papillibacter genera regarding GABA.

Reviewer #2:

[…]

Recommendations for the authors:

1. The authors state the NCDO2118 strain increased GABA in vivo, however the data in Figure 4 shows a significant decrease of GABA in the cecum and no significant differences along the rest of the GI tract in NCD02118+glutamate vs. vehicle groups (Figure 4A-C). The authors also state that the NDCO2118 strain has a higher concentration of GABA in the stomach in Figure 4C, but this is not a significant difference either. How does this lack of differences correlate with visceral pain blockade? Is it due to poor colonization of NCDO02118 after gavage or improper microenvironment?

As indicated above (see our response to comment 3 of reviewer 1), our objective was to determine in which GIT compartment GABA delivery specifically occurred for NCDO2118 strain (compared to vehicle and NCDO2727) and even though results did not reach significance, we can see in the stomach that the GABA production by L. lactis NCDO2118 is higher than that measured in vehicle or NCDO2727 strain. Interestingly, in a follow-up study (article in preparation) considering a higher GABA-producing L. lactis strain than NCDO2118, the GABA concentration in the stomach was significantly higher than for NCDO2727 strain and nearly for vehicle (p=0.07). We have performed additional experiments presented in Supplementary File 2 on in vitro kinetics of GABA production by L. lactis NCDO2118 and L. lactis NCDO2727 in the presence of glutamate under “stomach-like” conditions (at pH=4.6 in acetate buffer or gastric juice sampled from naive rats). We clearly demonstrate that, notably in gastric juice sampled from naive rats and supplemented with glutamate, the NCDO2118 strain produced a fourfold-increased GABA level while the rate of GABA production by the NCDO2727 strain was similar to the vehicle. To reinforce our findings on the potential role of the gastric region, we have inverted both sets of data in the revised version of the manuscript and slightly modified the in vitro part. Comments have been added in the Discussion section L353-L354. Concerning the detection of L. lactis in the GIT environment (from stomach to feces), a PCR-based approach would be insightful to overcome the poor suitability of the 16S rRNA gene sequencing method to low abundant bacterial populations. Interestingly, in the study of Duranti et al. in Scientific Reports in 2020 (https://doi.org/10.1038/s41598-020-70986-z) on Bifidobacterium adolescentis, the authors showed that the fecal concentration of GABA in rats treated with two high GABA-producing strains, namely B. adolescentis PRL2019 and B. adolescentis HD17T2H, was not statistically different than that found in rats treated with a non-GABA producer strain B. adolescentis ATCC15703 or rats not supplemented by B. adolescentis strains (control group). Furthermore, in the study of Pokusaeva et al. in Neurogastroenterology and Motility in 2017 (https://doi.org/10.1111/nmo.12904), demonstrating that GABA-producing Bifidobacterium dentium modulates visceral sensitivity in the intestine, the authors found that cecal GABA concentrations in the B. dentium-treated mice vs PBS-treated ones were within a similar range (10.1 ± 2.5 vs 10.3 ± 1.7 μg/g cecal content). This clearly indicates that it is not so straightforward to correlate cecal/fecal levels of GABA and in vivo efficacy.

2. The authors observed higher GABA in the stomach compared with cecum (Figure 4A). However, the colorectal distension assay was performed in the colon and rectum. How does local GABA upregulation in stomach contribute to the alleviation of visceral pain in the colon? Are circulating GABA levels changed?

We thank the reviewer for addressing these questions. Reflex among different organs along the gastrointestinal tract is well known. For example, ingestion of food into the stomach induces a response not only in the stomach but also in the small intestine and colon; distention of the rectum inhibits gastric and intestinal motility (Bampton et al. 2002; Kerlin et al. 1983; Shafik et al. 2000). Cross-talk along the gastrointestinal tract has also been observed. Gastric electrical stimulation inhibits rectal tone, an effect mediated by sympathetic pathway (Liu et al. 2005).Vagal afferent fibers innervating the upper gastro-intestinal tract play a critical role in the initiation of symptoms and reflexes controlling several functions (Page and Blackshaw, 1999). Vagal afferents are comprised of mucosal endings that respond to touch and chemical stimuli and muscular endings that respond optimally to mechanical stretch or tension (Page et al. 2002). There is a clear distribution of GABA receptors and particularly of GABAB receptors along the peripheral afferent vagal fibers. In our study, even though we have to more deeply elucidate in a follow-up study the functionalization pathway of GABA released by L. lactis into the stomach, we can hypothesize a primary mucosal activation of such receptors with central vagal pathways in the nucleus tractus solitarii (NTS) and dorsal vagal nucleus activations. In response to this central GABAergic integrative stimulation, an analgesic spinal descending outcome could in turn counteract visceral hypersensitivity induced by stress in response to colorectal distension. Besides this nervous cross-talk mechanistic pathway of GABA delivered by L. lactis into the stomach, we cannot exclude an antinociceptive effect mediated by an increase of circulating GABA. However, at this stage we have no data on this hypothesis. Thanks to the referee’s input, the measurements of circulating GABA levels are now planned in our future investigations on L. lactis studies.

– Bampton PA, Dinning PG, Kennedy ML, Lubowski DZ, and Cook IJ. The proximal colonic motor response to rectal mechanical and chemical stimulation. Am J Physiol Gastrointest Liver Physiol 282: G443–G449, 2002. doi: 10.1152/ajpgi.00194.2001. PMID: 11841994.

– Kerlin P, Zinsmeister A, and Phillips S. Motor responses to food of the ileum, proximal colon, and distal colon of healthy humans. Gastroenterology 84: 762–770, 1983. PMID: 6825988.

– Shafik A and El-Sibai O. Esophageal and gastric motile response to rectal distension with identification of a recto-esophago gastric reflex. Int J Surg Investig 1: 373–379, 2000. PMID: 11341593.

– Liu S, Wang L, Chen J. D. Z. Cross-talk along gastrointestinal tract during electrical stimulation: effectsand mechanisms of gastric/colonic stimulation on rectal tone in dogs. Am J Physiol Gastrointest Liver Physiol 288: G1195–G1198, 2005. doi: 10.1152/ajpgi.00554.2004. Epub 2005 Feb 3. PMID: 15691864.

– Page AJ and Blackshaw LA. GABA(B) receptors inhibit mechanosensitivity of primary afferent endings. J Neurosci19: 8597–8602, 1999. doi: 10.1523/JNEUROSCI.19-19-08597.1999. PMID: 10493759; PMCID: PMC6783028.

– Page AJ, Martin CM and Blackshaw LA (2002). Vagal mechanoreceptors and chemoreceptors in mouse stomach and esophagus. J Neurophysiol 87, 2095–2103. doi: 10.1152/jn.00785.2001. PMID: 11929927.

3. The authors show that GABAB receptor antagonist blocks the beneficial effect of the NCDO02118 strain on visceral hypersensitivity, but the authors do not try to identify which cell types are involved in this mechanism. Enteric, vagal, and spinal neurons that innervate the gut all have GABA receptors and identifying which subtype is involved would strengthen the study.

We agree with the reviewer that in this study we did not identify the cell types involved in the antinociceptive mechanism. As answered in the comment 2 of the reviewer, GABA receptors are present on the vagal afferent fibers both at the mucosal endings that respond to touch and chemical stimuli and muscular endings that respond optimally to mechanical stretch or tension (Page et al. 2002, see above). As discussed in our paper, GABAB receptors are also widely distributed from the stomach to the ileum in the enteric nervous system (ENS). In rodents, the presence of GABAB receptors has also been identified on the mucosal gland cells of the stomach. All these data suggest complex mechanisms and pathways by which GABA delivered by NCDO2118 strain exerts visceral antinociceptive effect. Based on the literature and our own observations, at this step of our knowledge, we cannot emphasize a cell type more than another and we agree that additional investigations are needed to open this black box.

4. The authors show an interesting phenotype that the NCDO02727 strain does not produce GABA even though it has the proper machinery. This study does not explore why this strain is not able to make GABA. Can the authors reveal some mechanistic insight into the differences between NCDO02727 vs. NCDO2118?

Why the NCDO2118 strain produces GABA and the corresponding GAD enzyme but not the NCDO2727 is a really intriguing question. In the manuscript, we have demonstrated by in silico analyses that these differences were not due to differences in the genetic organisation of gadCB operon and gadR gene nor to differences in GadB,C and R protein sequences.

In order to determine if these differences in GABA production could be associated to differences in gadCB and gadR expression regulation, we have now analysed the nucleotide sequences of their promoters. They were found identical (except 2 SNPs for gadR and 1 for gadCB, see Author response image 1 the example regarding the comparison of the gadB promoter in NCDO2118 and NCDO2727). These very few mutations seem minimal to explain major differences in transcriptional regulation of GadB.

Author response image 1
Download asset Open asset

We believe that the differences between NCDO2118 and NCDO2727 strains are likely due to post-transcriptional regulations but we do not really know the precise mechanisms involved. In the literature, very few studies on the regulation of the gadCB/R gene have been reported, even for chloride activation, for which molecular basis is still unclear (Sanders et al. 1998). In fact, nothing in the literature can helps us to formulate realistic hypothesis.In order to study the gadCB and gadR regulation and thanks to reporter gene constructions, we are currently investigating the gadCB expression in various environments and in various strains. Considerable work will be required to fully understand the gad operon regulations and identify the involved actors (protein, ncRNA, etc.). Only after understanding this, we can expect to explain the differences observed between the strains.

At this stage, we added in the manuscript the results of the promoter analyses. The following sentence was added in results L142-L144 “Only two single nucleotide polymorphisms and a deletion of one bp were identified in the gadR and gadCB promoters respectively compared to NCDO2118.” In the discussion a comment was also added L302-L304 “Because the promoters of gad genes are nearly identical in the two strains, transcriptional regulations should not explain the observed differences.”

– Sanders JW, Leenhouts K, Burghoorn J, Brands JR, Venema G, Kok J. A chloride-inducible acid resistance mechanism in Lactococcus lactis and its regulation. Mol Microbiol. 1998 Jan;27(2):299-310. doi: 10.1046/j.1365-2958.1998.00676.x. PMID: 9484886.

5. Do gut anti-inflammatory properties of NCDO02118 contribute to their antinociceptive effect in response to PRS? NCDO02118 admininstration induces the upregulation of IL-10 (DOI: 10.3389/fmicb.2021.623920), which can block pain. Is IL-10 also induced in these mice?

We thank the reviewer for this relevant question. In a previous study, we have shown that PRS increased gut paracellular permeability and endotoxemia, two responses which in cascade induced elevated mRNA expression of pro-inflammatory cytokines (IL-1β, IL-6 and TNF-α) in the PVN (Ait-Belgnaoui et al. 2012, see above). This central neuro-inflammation was prevented by a chronic daily treatment with a probiotic strain i.e. Lactobacillus farciminis. Since several probiotics have been described to be able to restore gut physical barrier damage as well as to increase anti-inflammatory cytokines like IL-10, it would be very interesting in future investigations to evaluate gut anti-inflammatory properties of NCDO2118 in this model.

6. The authors suggest that L. lactis may be a treatment option for visceral pain and anxiety associated with irritable bowel syndrome (IBS). However, the study does not test whether L. lactis has any effect on anxiety behavior. Can the authors test anxiety behaviors?

We thank the reviewer for this point. We know that the comorbidity between IBS and anxiety disorders is elevated and that anxiety and depression contribute to inefficient therapies in IBS (Vandvik et al. 2006; Popa et al. 2015). Anxiety is a complex disorder, defined as the response to prolonged, unpredictable threat, a response which encompasses physiological, affective, and cognitive changes. To our knowledge, different animal models are developed to mimic anxiety. The Open Field Maze is one of the most commonly anxiety-like models used to characterize behaviors in rodents. However, until now, we haven’t this type of model in our lab but we are exploring the possibility of a collaborative work to investigate L. lactis ability to reduce anxiety behavior in response to stress.

Besides, even if we did not use an anxiety-like model in our study, we kindly point that PRS model is a reliable model widely used for studying the pharmacological modulation of the GABAergic system in order to prevent the anxiety-like behavior induced by acute and chronic stress (Reznikov et al. 2009; Nuss, 2015; Assad et al. 2020; Farajdokht et al. 2020).

– Assad, N., Luz, W. L., Santos-Silva, M., Carvalho, T., Moraes, S., Picanço-Diniz, D., et al. (2020). Acute restraint stress evokes anxiety-like behavior mediated by telencephalic inactivation and gabaergic dysfunction in zebrafish brains. Sci. Rep. 10:5551. doi: 10.1038/s41598-020-62077-w. PMID: 32218457; PMCID: PMC7099036.

– Farajdokht, F., Vosoughi, A., Ziaee, M., Araj-Khodaei, M., Mahmoudi, J., and Sadigh-Eteghad, S. (2020). The role of hippocampal GABAAreceptors on anxiolytic effects of Echium amoenum extract in a mice model of restraint stress. Mol. Biol. Rep. 47, 6487–6496. doi: 10.1007/s11033-020-05699-7. Epub 2020 Aug 10. PMID: 32778988.

– Nuss, P. (2015). Anxiety disorders and GABA neurotransmission: a disturbance of modulation. Neuropsychiatr. Dis. Treat. 11, 165–175. doi: 10.2147/NDT.S58841. PMID: 25653526; PMCID: PMC4303399.

– Popa Stefan-Lucian, Dumitrascu Dan Lucian. Anxiety and IBS revisited: ten years later. Clujul Med. 2015;88(3):253-7. doi: 10.15386/cjmed-495. Epub 2015 Jul 1. PMID: 26609253; PMCID: PMC4632879.

– Reznikov Leah R, Reagan Lawrence P, Fadel Jim R. Effects of acute and repeated restraint stress on GABA efflux in the rat basolateral and central amygdala. Brain Res. 2009 23;1256:61-8. doi: 10.1016/j.brainres.2008.12.022. Epub 2008 Dec 24. PMID: 19124010.

– Vandvik PO, Lydersen S, Farup PG. Prevalence, comorbidity and impact of irritable bowel syndrome in Norway. Scand J Gastroenterol. 2006;41(6):650–656. doi: 10.1080/00365520500442542. PMID: 16716962.

https://doi.org/10.7554/eLife.77100.sa2

Article and author information

Author details

  1. Valérie Laroute

    Toulouse Biotechnology Institute (TBI), Université de Toulouse, CNRS, INRAE, INSA, Toulouse, France
    Contribution
    Data curation, Formal analysis, Investigation, Methodology
    Contributed equally with
    Catherine Beaufrand
    Competing interests
    No competing interests declared

  2. Catherine Beaufrand

    Toxalim (Research Centre in Food Toxicology), Université de Toulouse, INRAE, ENVT, INP-Purpan, UPS, Toulouse, France
    Contribution
    Data curation, Formal analysis, Investigation, Methodology
    Contributed equally with
    Valérie Laroute
    Competing interests
    No competing interests declared

  3. Pedro Gomes

    1. Toulouse Biotechnology Institute (TBI), Université de Toulouse, CNRS, INRAE, INSA, Toulouse, France
    2. Toxalim (Research Centre in Food Toxicology), Université de Toulouse, INRAE, ENVT, INP-Purpan, UPS, Toulouse, France
    Contribution
    Data curation, Formal analysis, Methodology
    Competing interests
    No competing interests declared

  4. Sébastien Nouaille

    Toulouse Biotechnology Institute (TBI), Université de Toulouse, CNRS, INRAE, INSA, Toulouse, France
    Contribution
    Formal analysis, Investigation, Methodology
    Competing interests
    No competing interests declared

  5. Valérie Tondereau

    Toxalim (Research Centre in Food Toxicology), Université de Toulouse, INRAE, ENVT, INP-Purpan, UPS, Toulouse, France
    Contribution
    Formal analysis, Investigation, Methodology
    Competing interests
    No competing interests declared

  6. Marie-Line Daveran-Mingot

    Toulouse Biotechnology Institute (TBI), Université de Toulouse, CNRS, INRAE, INSA, Toulouse, France
    Contribution
    Formal analysis, Investigation, Methodology
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-6884-1840

  7. Vassilia Theodorou

    Toxalim (Research Centre in Food Toxicology), Université de Toulouse, INRAE, ENVT, INP-Purpan, UPS, Toulouse, France
    Contribution
    Conceptualization, Supervision, Validation
    Competing interests
    No competing interests declared

  8. Hélène Eutamene

    Toxalim (Research Centre in Food Toxicology), Université de Toulouse, INRAE, ENVT, INP-Purpan, UPS, Toulouse, France
    Contribution
    Conceptualization, Funding acquisition, Supervision, Validation, Writing – original draft, Writing – review and editing
    Contributed equally with
    Muriel Mercier-Bonin and Muriel Cocaign-Bousquet
    For correspondence
    helene.eutamene@inrae.fr
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-2983-1938

  9. Muriel Mercier-Bonin

    Toxalim (Research Centre in Food Toxicology), Université de Toulouse, INRAE, ENVT, INP-Purpan, UPS, Toulouse, France
    Contribution
    Conceptualization, Funding acquisition, Supervision, Validation, Writing – original draft, Writing – review and editing
    Contributed equally with
    Hélène Eutamene and Muriel Cocaign-Bousquet
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-8398-2529

  10. Muriel Cocaign-Bousquet

    Toulouse Biotechnology Institute (TBI), Université de Toulouse, CNRS, INRAE, INSA, Toulouse, France
    Contribution
    Conceptualization, Funding acquisition, Supervision, Validation, Writing – original draft, Writing – review and editing
    Contributed equally with
    Hélène Eutamene and Muriel Mercier-Bonin
    For correspondence
    cocaign@insa-toulouse.fr
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-2033-9901

Funding

No external funding was received for this work.

Acknowledgements

This work received the financial support of Toulouse-Tech Transfer (Maturation project TTT P0091) and Lesaffre International (Marcq-en-Barœul, France) and was part of patent n° WO 2020/157297. The authors wish to thank the Sequencing Platform of Toulouse (GeT-Biopuces) and especially Etienne Rifa for gut microbiota analyses. The authors wish to thank Hervé Robert (Toxalim, Toulouse, France) for helpful discussion.

Ethics

Human Subjects: No Animal Subjects: Yes Ethics Statement: Allexperiments were approved by the Local Animal Care and Use Committee (APAFiS#5577-201606061639777v3, see Colorectal and Under PRS conditions sections and APAFiS#14898-2018043016031426, see Without PRS) in compliance with European directive 2010/63/UE.

Senior and Reviewing Editor

  1. Gisela Storz, National Institute of Child Health and Human Development, United States

Version history

  1. Received: January 15, 2022
  2. Accepted: June 1, 2022
  3. Version of Record published: June 21, 2022 (version 1)

Copyright

© 2022, Laroute, Beaufrand et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 776
    Page views
  • 198
    Downloads
  • 5
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Valérie Laroute
  2. Catherine Beaufrand
  3. Pedro Gomes
  4. Sébastien Nouaille
  5. Valérie Tondereau
  6. Marie-Line Daveran-Mingot
  7. Vassilia Theodorou
  8. Hélène Eutamene
  9. Muriel Mercier-Bonin
  10. Muriel Cocaign-Bousquet
(2022)
Lactococcus lactis NCDO2118 exerts visceral antinociceptive properties in rat via GABA production in the gastro-intestinal tract
eLife 11:e77100.
https://doi.org/10.7554/eLife.77100

Categories and tags

Research organism

Further reading

    1. Microbiology and Infectious Disease

    Evolution of a functionally intact but antigenically distinct DENV fusion loop

    Rita M Meganck, Deanna Zhu ... Longping V Tse
    Short Report

    A hallmark of dengue virus (DENV) pathogenesis is the potential for antibody-dependent enhancement, which is associated with deadly DENV secondary infection, complicates the identification of correlates of protection, and negatively impacts the safety and efficacy of DENV vaccines. Antibody-dependent enhancement is linked to antibodies targeting the fusion loop (FL) motif of the envelope protein, which is completely conserved in mosquito-borne flaviviruses and required for viral entry and fusion. In the current study, we utilized saturation mutagenesis and directed evolution to engineer a functional variant with a mutated FL (D2-FL), which is not neutralized by FL-targeting monoclonal antibodies. The FL mutations were combined with our previously evolved prM cleavage site to create a mature version of D2-FL (D2-FLM), which evades both prM- and FL-Abs but retains sensitivity to other type-specific and quaternary cross-reactive (CR) Abs. CR serum from heterotypic (DENV4)-infected non-human primates (NHP) showed lower neutralization titers against D2-FL and D2-FLM than isogenic wildtype DENV2 while similar neutralization titers were observed in serum from homotypic (DENV2)-infected NHP. We propose D2-FL and D2-FLM as valuable tools to delineate CR Ab subtypes in serum as well as an exciting platform for safer live-attenuated DENV vaccines suitable for naïve individuals and children.

    1. Microbiology and Infectious Disease

    Biosensor-integrated transposon mutagenesis reveals rv0158 as a coordinator of redox homeostasis in Mycobacterium tuberculosis

    Somnath Shee, Reshma T Veetil ... Amit Singh
    Research Article Updated

    Mycobacterium tuberculosis (Mtb) is evolutionarily equipped to resist exogenous reactive oxygen species (ROS) but shows vulnerability to an increase in endogenous ROS (eROS). Since eROS is an unavoidable consequence of aerobic metabolism, understanding how Mtb manages eROS levels is essential yet needs to be characterized. By combining the Mrx1-roGFP2 redox biosensor with transposon mutagenesis, we identified 368 genes (redoxosome) responsible for maintaining homeostatic levels of eROS in Mtb. Integrating redoxosome with a global network of transcriptional regulators revealed a hypothetical protein (Rv0158) as a critical node managing eROS in Mtb. Disruption of rv0158 (rv0158 KO) impaired growth, redox balance, respiration, and metabolism of Mtb on glucose but not on fatty acids. Importantly, rv0158 KO exhibited enhanced growth on propionate, and the Rv0158 protein directly binds to methylmalonyl-CoA, a key intermediate in propionate catabolism. Metabolite profiling, ChIP-Seq, and gene-expression analyses indicate that Rv0158 manages metabolic neutralization of propionate toxicity by regulating the methylcitrate cycle. Disruption of rv0158 enhanced the sensitivity of Mtb to oxidative stress, nitric oxide, and anti-TB drugs. Lastly, rv0158 KO showed poor survival in macrophages and persistence defect in mice. Our results suggest that Rv0158 is a metabolic integrator for carbon metabolism and redox balance in Mtb.

    1. Microbiology and Infectious Disease
    2. Structural Biology and Molecular Biophysics

    A pH-sensitive switch activates virulence in Salmonella

    Dasvit Shetty, Linda J Kenney
    Research Article

    The transcriptional regulator SsrB acts as a switch between virulent and biofilm lifestyles of non-typhoidal Salmonella enterica serovar Typhimurium. During infection, phosphorylated SsrB activates genes on Salmonella Pathogenicity Island-2 (SPI-2) essential for survival and replication within the macrophage. Low pH inside the vacuole is a key inducer of expression and SsrB activation. Previous studies demonstrated an increase in SsrB protein levels and DNA-binding affinity at low pH; the molecular basis was unknown (Liew et al., 2019). This study elucidates its underlying mechanism and in vivo significance. Employing single-molecule and transcriptional assays, we report that the SsrB DNA binding domain alone (SsrBc) is insufficient to induce acid pH-sensitivity. Instead, His12, a conserved residue in the receiver domain, confers pH sensitivity to SsrB allosterically. Acid-dependent DNA binding was highly cooperative, suggesting a new configuration of SsrB oligomers at SPI-2-dependent promoters. His12 also plays a role in SsrB phosphorylation; substituting His12 reduced phosphorylation at neutral pH and abolished pH-dependent differences. Failure to flip the switch in SsrB renders Salmonella avirulent and represents a potential means of controlling virulence.