Article

Efficacy of Lactobacillus paracasei HA-196 and Bifidobacterium longum R0175 in Alleviating Symptoms of Irritable Bowel Syndrome (IBS):

1. A Randomized, Placebo-Controlled Study

Erin D. Lewis 1, Joseph M. Antony 1, David C. Crowley 1, Amanda Piano 2, Renu Bhardwaj 2, Thomas A. Tompkins 2 and Malkanthi Evans 1,*

1. 1 KGK Science Inc., London, ON N6A 5R8, Canada; [email protected] (E.D.L.); [email protected] (J.M.A.); [email protected] (D.C.C.)
2. 2 Lallemand Health Solutions, Montreal, QC H4P 2R2, Canada; [email protected] (A.P.); [email protected] (R.B.); [email protected] (T.A.T.)

* Correspondence: [email protected]

Received: 5 February 2020; Accepted: 16 April 2020; Published: 21 April 2020

Abstract: Specific probiotic strains can alleviate the gastrointestinal (GI) symptoms and psychiatric comorbidities of irritable bowel syndrome (IBS). In this randomized, double-blind, placebo-controlled study, the efficacy of Lactobacillus paracasei HA-196 (L. paracasei) and Bifidobacterium longum R0175 (B. longum) in reducing the GI and psychological symptoms of IBS was evaluated in 251 adults with either constipation (IBS-C), diarrhea (IBS-D), or mixed-pattern (IBS-M). Following a 2-week run-in period, participants were randomized to one of three interventions: L. paracasei (n = 84), B. longum (n = 83) or placebo (n = 81). IBS symptoms, stool frequency and consistency and quality of life were assessed by questionnaires. The differences from baseline in the severity of IBS symptoms at 4 and 8 weeks were similar between groups. Participants in this study were classified, after randomization, into subtypes according to Rome III. Within the L. paracasei group, complete spontaneous and spontaneous bowel movement frequency increased in participants with IBS-C (n = 10) after 8 weeks of supplementation (both p < 0.05) and decreased in participants with IBS-D (n = 10, p = 0.013). Both L. paracasei and B. longum supplementation improved the quality of life in emotional well-being and social functioning compared with baseline (all p < 0.05). In conclusion, L. paracasei and B. longum may reduce GI symptom severity and improve the psychological well-being of individuals with certain IBS subtypes.

Keywords: probiotics; Bifidobacterium longum; Lactobacillus paracasei; Irritable Bowel Syndrome; IBS

Irritable bowel syndrome (IBS) is one of the most common functional gastrointestinal disorders (GI) worldwide, affecting approximately 11% of the global population [1]. IBS is characterized by the co-occurrence of abdominal pain or discomfort, changes in bowel habits, and defecation [2]. Based on bowel habits and stool consistency, IBS patients are grouped into three subtypes, namely the constipation-predominant IBS (IBS-C), diarrhea-predominant IBS (IBS-D), and the mixed bowel habits or cyclic patterns IBS (IBS-M) characterized by the occurrence of both constipation and diarrhea episodes [2]. The physical symptoms of IBS are variable in severity, ranging from unpleasant to debilitating, and as a result, patients’ psychological well-being and quality of life can be adversely affected in more severe cases. This variability in the severity of IBS symptoms, coupled with our incomplete understanding of its pathophysiology, often results in extended delays for patients before

Nutrients 2020, 12, 1159; doi:10.3390/nu12041159 www.mdpi.com/journal/nutrients

receiving a diagnosis. This also leads to inconsistencies in the efficacy of symptom management regimes. Currently, the first step in IBS symptom management involves lifestyle and dietary modifications, such

1. as the adoption of low fermentable oligo-, di-, and monosaccharides and polyols (FODMAPs) and gluten-free diets [3]. In cases where dietary changes are ineffective, adjunctive pharmacotherapies such as antidiarrheal agents, antibiotics, laxatives, antispasmodics, or antidepressants are introduced. Pharmacological interventions often target specific but not all symptoms of IBS and may result in serious adverse events [4,5]. Based on accumulating research evidence, gut microbial dysbiosis is thought to play a role in the pathophysiology of IBS and symptom severity. Consequently, promising leads in IBS management now include the potential use of specific probiotic bacteria to improve both the GI symptoms and psychological comorbidities in IBS [6].
2. at KGK Science Inc., London, ON, Canada between September 2014 and February 2018. The study

included a 2-week run-in period, during which regular bowel habits were reported, and an 8-week intervention period (day 1 to day 57), consisting of a total of four clinic visits (FigureNutrients2020,12, x FOR PEER REVIEW 3 of 221).

Figure 1. Schematic representation of the study design. This 10-week study consisted of 2 periods: a 2-week run-in and an 8-week supplementation period with one of two probiotics, Lactobacillus paracasei (L. paracasei) HA-196 or Bifidobacterium longum (B. longum) R0175 or a placebo (days 1–56).

Clinical and qualitative assessments were conducted at screening, day 0 (baseline), 4 weeks and 12 weeks (end-of-study). At the screening visit, participants were provided with a 3-day food record and a 2-week daily diary to complete before the baseline visit. The daily diary assessed the number of spontaneous bowel movements (SBM; defined as a stool not induced by rescue medication), complete spontaneous bowel movements (CSBM; defined as an SBM associated with a sensation of complete evacuation), and stool consistency based on the Bristol stool scale [18]. Participants were also instructed to collect a fecal sample (5–10 g) using the provided stool collection kits 3 days before taking the investigational product (IP). The fecal sample was to be frozen immediately upon collection and brought to the clinic before or at the baseline visit.

12 weeks (end-of-study). At the screening visit, participants were provided with a 3-day food record and a 2-week daily diary to complete before the baseline visit. The daily diary assessed the number of spontaneous bowel movements (SBM; defined as a stool not induced by rescue medication), complete spontaneous bowel movements (CSBM; defined as an SBM associated with a sensation of complete evacuation), and stool consistency based on the Bristol stool scale [18]. Participants were also instructed to collect a fecal sample (5–10 g) using the provided stool collection kits 3 days before taking the investigational product (IP). The fecal sample was to be frozen immediately upon collection and brought to the clinic before or at the baseline visit.

At all study visits, the Irritable Bowel Syndrome Symptom Severity Scale (IBS-SSS), Short Form (36) Health Survey v2TM (SF-36), Irritable Bowel Syndrome Quality of life (IBS-QOL), and Hospital Anxiety Depression Scale (HADS) questionnaires were administered and assessed. Additionally, 3day food records and diaries were reviewed for daily IBS-SSS, concomitant therapies, adverse events, and study product use. Vital signs and anthropometric measurements were undertaken at each visit while hematology and clinical chemistry parameters for all safety endpoints were assessed at the screening and end-of-study visits.

At all study visits, the Irritable Bowel Syndrome Symptom Severity Scale (IBS-SSS), Short Form (36) Health Survey v2TM (SF-36), Irritable Bowel Syndrome Quality of life (IBS-QOL), and Hospital Anxiety Depression Scale (HADS) questionnaires were administered and assessed. Additionally, 3-day food records and diaries were reviewed for daily IBS-SSS, concomitant therapies, adverse events, and study product use. Vital signs and anthropometric measurements were undertaken at each visit while hematology and clinical chemistry parameters for all safety endpoints were assessed at the screening and end-of-study visits.

At week 8, participants provided a fecal sample (5–10 g) collected within the 3 days preceding their last visit and frozen immediately after collection and then brought to the clinic.

2.2. Participants

Individuals diagnosed with IBS were recruited from Southwestern Ontario, Canada using the KGK Science Inc.’s internal participant database along with local electronic and physical advertisement devoid of gender or racial bias. Participants met the following inclusion criteria: aged 18 years or older, IBS diagnosed according to the Rome III criteria [19], and willingness to discontinue probiotic consumption for the duration of the study. After the broadening of eligibility criteria to all

At week 8, participants provided a fecal sample (5–10 g) collected within the 3 days preceding their last visit and frozen immediately after collection and then brought to the clinic.

1. 2.2. Participants

randomization using the IBS Rome III Questionnaire completed at screening.

Individuals diagnosed with IBS were recruited from Southwestern Ontario, Canada using the KGK Science Inc.’s internal participant database along with local electronic and physical advertisement devoid of gender or racial bias. Participants met the following inclusion criteria: aged 18 years or older, IBS diagnosed according to the Rome III criteria [19], and willingness to discontinue probiotic consumption for the duration of the study. After the broadening of eligibility criteria to all subtypes, the distribution of IBS subtypes in each intervention group was determined after randomization using the IBS Rome III Questionnaire completed at screening.

the past month, history of gastrointestinal surgery, gastrointestinal disease (except hemorrhoids and

Participants were excluded if they used medications to manage IBS symptoms or narcotics in the past month, history of gastrointestinal surgery, gastrointestinal disease (except hemorrhoids and

uncomplicated diverticula) or family history of colorectal cancer, inflammatory bowel disease, or celiac sprue.

1. 2.3. Interventions
2. 2.4. Randomization and Blinding
3. 2.5. Assessments of IBS Symptom Severity, General Health and Psychological Well-Being
4. 2.6. Fecal Microbe Composition Analysis

Stool samples collected at baseline and week 8 were used to confirm participant compliance and microbe composition through quantitative real-time polymerase chain reaction (qPCR). Total DNA from 250–350 mg of homogenized fecal samples was extracted using the QIAamp Fast DNA Stool Mini kit (QIAGEN, ON, Canada) with the following modifications: (1) two washes with 0.05 M phosphate buffer prior to the InhibitEx step and (2) a 0.1 mm zirconia/silica bead beating step

(3 × 4 m/s for 1 min) prior to the centrifugation of samples to pellet particles. Ratios of A260/A280 were assessed to determine sample purity. Absolute quantification of the two probiotic strains and Bifidobacterium genus and relative quantification of Akkermansia mucinphilia and Faecalibaterium prausnitzii was conducted. Template DNA generated for the standard curve formation involved spiking 109 of the appropriate lyophilized powder into a fecal mix and proceeding with the DNA extraction. DNA was diluted 10-fold to generate a standard curve ranging from 109–104 bacteria for L. paracasei and B. longum Primers for B. longum R0175 forward: 5’-GTCGCCACA TTTCATCGCAA-3’, reverse: 5’-GAGAGCTTCGATTGGCGAAC-3’; L. paracasei HA-196 forward: 5’-ACCGAAGTCTATCACCCGGA-3’, reverse: 5’-TCGCCAAATTTGCTGTCGTG-3’; Bifidobacterium genus forward: 5’-TGGAAGGTCTCGATGGAGGT-3’, reverse: 5’-CTGGACAAGCCGTTCCTGAT-3’). DNA was diluted 1/5 prior to qPCR. Each qPCR reaction contained 300 nM of the appropriate forward and reverse primer, 1X SYBR Select Master Mix (Thermo Fisher Scientific, Waltham, MA, USA) and 1 µL of diluted DNA. The epMotion 5075 liquid handling robot (Eppendorf, Hamburg, Germany) was used to dilute DNA from the fecal samples 5-fold, and add 9 mL of mastermix and 1 µL of diluted DNA to the reaction plate. The cycling conditions for the B. longum R0175, L. paracasei HA-196, Bifidobacterium species, and 16S rRNA universal bacterial primers included a 2-min hold at 50 ◦C, a 2-min hold at 95 ◦C and 40 cycles of 95 ◦C for 15 s, 60 ◦C for 30 s and 72 ◦C for 30 s. The relative quantifications of A. muciniphila and F. prausnitzii was performed via species-specific primers and DNA was normalized by bacterial primers targeting the 16S rRNA gene. The primers and cycling conditions used for A. muciniphila and F. prausnitzii quantification were obtained from literature [22–24]. The CFX384™ Touch Real-Time polymerase chain reaction (PCR) Detection System was used to perform quantitative PCR (qPCR) analyses and results were viewed using the CFX Maestro Software 1.1 (Bio-Rad, Hercules, CA).

1. 2.7. Compliance
2. 2.8. Statistical Analyses

The proposed sample size for this study was 285 enrolled participants, with 95 participants randomized into each of the three study arms in a double-blinded manner. The sample size calculation was based on a standard deviation (SD) of 60, a significance level of 5% (two-sided α), 83% power (β = 0.17), 20% attrition rate and a 110-point detectable difference in IBS-SSS scores from baseline to week 8, compared between probiotics and placebo. The assumptions were based on two previous trials, by Dapoigny et al. (2012) [25] and Williams et al. (2009) [26], which examined the effect of probiotic interventions with Lactobacillus and Bifidobacterium strains on the symptom severity scores in adult IBS patients.

All statistical analyses were completed using SAS® software Version 9.3 (SAS Institute, Cary, NC, USA) for Microsoft Windows, where probabilities ≤0.05 were considered statistically significant.

1. A blinded interim analysis was conducted on the primary outcome (IBS-SSS) after 50% recruitment. The blind was maintained for all study personnel including the statistician. The per protocol (PP) population consisted of all participants who consumed at least 80% of either product dose, did not have any major protocol violations, and completed all study visits and procedures connected with measurement of the primary variable. The safety population consisted of all participants who received either product and on whom any post-randomization safety information was available. An efficacy analysis based on the PP population and IBS subtypes was performed. Variables were tested for

normality and log-normality, and variables showing a log-normal distribution were analyzed in the logarithmic domain.

Appropriate non-parametric tests were used to analyze non-normal variables. Continuous, normally distributed variables were analyzed by linear mixed-effects models with visit (baseline, week 4, and week 8), intervention (L. paracasei, B. longum, and placebo groups), and their interaction as fixed effects, and subject as a random effect to capture the repeated observations on each subject. Additionally, within-group changes from Day 0 were obtained from the linear mixed-effects models.

1. 3.1. Study Participant Dispositions

Out of the 494 participants screened, 285 were eligible and enrolled in the study. Ninety-five participants were randomized into each of the three groups to receive either L. paracasei, B. longum, or placebo (Figure 2). Following randomization, six participants withdrew consent and 1 participant was lost to follow up in the L. paracasei group. In the B. longum group, two participants withdrew consent, two were withdrawn by the Qualified Investigator (QI) and two were lost to follow up. In the placebo group, eight participants discontinued the study (three withdrew consent, two were withdrawn by the QI and three were lost to follow up). Thirteen participants who completed the study were removed from per protocol (PP) analysis. Four participants were removed from the L. paracasei group due to consumption of magnesium throughout the study (n = 1), use of antibiotics (n = 1), or insufficient compliance (n = 2). Three participants were excluded from the B. longum group for consumption of NSAIDs (n = 1), antibiotics (n = 1), and laxatives (n = 1). In the placebo group, six participants were removed from PP analysis due to the consumption of NSAIDs and PPIs (n = 1) or antibiotics (n = 4), while another was enrolled 18 days outside of the screening window (n = 1). The demographic information of the PP population is presented in Table 1, along with the IBS subtype distribution within each treatment conditions.

Nutrients 2020

Figure 2. Participant disposition throughout the study. At the end of the study, 251 participants were included in the per protocol population. ITT, intent-to-treat; PP, per protocol.

Table 1. Demographic characteristics for all participants in the per protocol (PP) population and their IBS-subtypes.

Age (years) Mean+/−SD 42.42 ± 12.30 (84) 42.31 ± 16.88 (86) 41.84 ± 16.14 (81)

Median (Min–Max) 43.00 (21.00–72.00) 41.00 (19.00–87.00) 40.00 (18.00–75.00)

1. p-value * 0.9573 0.9706
2. p-value * 1.0000 0.5836

Hispanic or Latino 7 (8.3%) 5 (5.8%) 1 (1.2%) Not Hispanic or Latino 77 (91.7%) 81 (94.2%) 80 (98.8%)

p-value * 0.0641 0.2115 Race [n (%)] Black or African American

2 (2.3%) 2 (2.5%) Central American 1 (1.2%) 1 (1.2%)

East Asian 2 (2.5%) Eastern European White 10 (11.9%) 9 (10.5%) 8 (9.9%)

Middle Eastern 8 (9.5%) 2 (2.3%) 3 (3.7%) North American Indian 1 (1.2%) 1 (1.2%)

South American 5 (6.0%) 3 (3.5%) 1 (1.2%) South Asian 2 (2.3%) 1 (1.2%) South East Asian 1 (1.2%) 1 (1.2%)

Western European White 59 (70.2%) 66 (76.7%) 62 (76.5%)

p-value * 0.1463 0.8926 Type of IBS: ROME III [n (%)]

IBS-C 10 (11.9%) 5 (5.8%) 13 (16.0%) IBS-D 10 (11.9%) 15 (17.4%) 13 (16.0%) IBS-M 64 (76.2%) 66 (76.7%) 55 (67.9%)

p-value * 0.5434 0.1041

IBS, irritable bowel syndrome; IBS-C, constipation-predominant IBS, IBS-D, diarrhea-predominant IBS; IBS-M, mixed bowel habits IBS; n, number; SD, standard deviation; Min, minimum; Max, maximum; * Continuous variables: p-values for comparison of each probiotic strain to placebo generated by ANOVA with Group as a fixed effect and subject as a random effect. Dunnett’s method was used to adjust for multiple comparisons; * Categorical variables: p-values for comparison of each probiotic strain to placebo generated by Fisher’s Exact (2-tail) test.

1. 3.2. Improvement in the IBS Symptom Severity Score and Rescue Medication Use

There were no significant between group changes; however, there were significant reductions in IBS-SSS at week 4 and 8 week from baseline in all three groups. The IBS-SSS scores were reduced significantly at week 4 vs. baseline in participants supplemented with L. paracasei (−20%), B. longum (−20%), and placebo (−17%) (all p < 0.001; Table 2 and Figure 3). Similarly, IBS-SSS scores were significantly reduced at week 8 vs. baseline in the L. paracasei (−30%), B. longum (−22%), and placebo (−31%) groups (all p < 0.001; Table 2 and Figure 3). Rescue medication (bisacodyl 5 mg tablets) was permitted for the relief of severe constipation, defined as 72 h since last BM or when symptoms became intolerable. Participants from both probiotic supplemented groups reported consuming less rescue medication compared to the placebo group; this difference was found to be statistically significant only for the L. paracasei group (p < 0.05; Figure 4, Supplementary Table S2).

Table 2. Irritable Bowel Syndrome Symptom Severity Scale (IBS-SSS) score at baseline, week 4, and week 8 and change in IBS-SSS from baseline to week 4, and week 8 for participants in the PP

Nutrients 2020, 12, 1159 8 of 19

Table 2. Irritable Bowel Syndrome Symptom Severity Scale (IBS-SSS) score at baseline, week 4, and week 8 and change in IBS-SSS from baseline to week 4, and week 8 for participants in the PP population.

IBS-SSS Study Day Statistic

L. paracasei HA196

B. longum R0175

Placebo

IBS-SSS Study Day Statistic

238.96 ± 64.55 (81) p-value * 0.7011 0.9960

246.94 ± 75.53 238.13 ± 76.59

Baseline (Day 0) Mean ± SD (n)

L. paracasei HA-196

(83)

B. longum R0175 Placebo

Baseline (Day 0) Mean ± SD (n) 246.94 ± 75.53 (84) 238.13 ± 76.59 (83) 238.96 ± 64.55 (81) p-value * 0.7011 0.9960

199.39 ± 72.13 (77) Mean ± SD (n)

198.22 ± 69.07 (82)

189.57 ± 82.55 (83)

Week 4 Mean ± SD (n)

164.56 ± 73.25 (81) Change from Baseline to Week 4

172.23 ± 80.90 (84)

185.71 ± 77.74 (84)

Week 4 Mean ± SD (n) 198.22 ± 69.07 (82) 189.57 ± 82.55 (83) 199.39 ± 72.13 (77) Week 8 Mean ± SD (n) 172.23 ± 80.90 (84) 185.71 ± 77.74 (84) 164.56 ± 73.25 (81)

−42.32 ± 66.93 (77) Within Group pValue +

−50.20 ± 64.42 (82)

−50.78 ± 72.12 (82)

Mean ± SD (n)

Change from Baseline to Week 4

Mean ± SD (n) −50.20 ± 64.42 (82) −50.78 ± 72.12 (82) −42.32 ± 66.93 (77) Within Group p-Value +

p < 0.001 (r) p < 0.001 (r) p < 0.001 (r) Between Group pvalue **

p < 0.001 (r) p < 0.001 (r) p < 0.001 (r) Between Group p-value **

0.4246 (r) 0.1390 (r) Change from Baseline to Week 8

0.4246 (r) 0.1390 (r) Change from Baseline to Week 8

−74.71 ± 71.52 (84)

−53.25 ± 65.65

Mean ± SD ( )

(83) (81) Within Group pValue +

Mean ± SD (n) −74.71 ± 71.52 (84) −53.25 ± 65.65 (83) −74.41 ± 62.83 (81) Within Group p-Value +

p < 0.001 (r) p < 0.001 (r) p < 0.001 (r) Between Group p-

p < 0.001 (r) p < 0.001 (r) p < 0.001 (r) Between Group p-value **

0.9632 (r) 0.0763 (r)

, number; SD, standard deviation; Min, minimum; Max, maximum; * For Baseline (Day 0),

0.9632 (r) 0.0763 (r)

for comparison of each probiotic strain to placebo generated by ANOVA with Group as a fixed effect and subject as a random effect. Dunnett’s method was used to adjust for multiple comparisons; ** Values generated from a Repeated Measures ANCOVA with baseline as a covariate and Group, Study Day and Group by Study Day interaction as fixed effects with subject as a random effect. p-values for comparison to placebo were adjusted for multiple comparisons using Dunnett’s method; + Within group p-values generated from the Repeated Measures ANCOVA specified above; (r) indicates values were ranked prior to generating ANOVA or ANCOVA.

n, number; SD, standard deviation; Min, minimum; Max, maximum; * For Baseline (Day 0), p-values for comparison of each probiotic strain to placebo generated by ANOVA with Group as a fixed effect and subject as a random effect. Dunnett’s method was used to adjust for multiple comparisons; ** Values generated from a Repeated Measures ANCOVA with baseline as a covariate and Group, Study Day and Group by Study Day interaction as fixed effects with subject as a random effect. p-values for comparison to placebo were adjusted for multiple comparisons using Dunnett’s method; + Within group p-values generated from the Repeated Measures ANCOVA specified above; (r) indicates values were ranked prior to generating ANOVA or ANCOVA.

1. Figure 3. Effect of Lactobacillus paracasei HA-196 and Bifidobacterium longum R0175 on the IBS symptom severity score. (A) Mean Total IBS-SSS at baseline, week 4 and week 8; and (B) Mean change in IBS-SSS from baseline to week 4 and week 8 for participants in the PP population. L. paracasei group n = 84, B. longum group n = 86, placebo group n = 81; data represented as mean ± SD; *** p < 0.001; IBS-SSS, irritable bowel syndrome symptom severity scale. L. paracasei, Lactobacillus paracasei HA-196; B. longum, Bifidobacterium longum R0175.

1. Figure 4. Effect of Lactobacillus paracasei HA-196 and Bifidobacterium longum R0175 on the use of rescue medication (bisacodyl 5 mg tablets) in participants with IBS compared to placebo. L. paracasei group n = 84, B. longum group n = 86, placebo group n = 81; data represented as mean ± SD; * p < 0.05. L. paracasei, Lactobacillus paracasei HA-196; B. longum, Bifidobacterium longum R0175.

Figure 4. Effect of Lactobacillus paracasei HA-196 and Bifidobacterium longum R0175 on the use of rescue medication (bisacodyl 5 mg tablets) in participants with IBS compared to placebo. L. paracasei group n = 84, B. longum group n = 86, placebo group n = 81; data represented as mean ± SD; * < 0.05. L. paracasei, Lactobacillus paracasei HA-196; B. longum, Bifidobacterium longum R0175.

3.3. Increase in Bowel Movement Frequency in Participants with IBS-C

1. 3.3. Increase in Bowel Movement Frequency in Participants with IBS-C

and stool consistency among the three groups (Table 3). However, L. paracasei consumption led to increased number of within-group SBM and CSBM. At week 4, participants with IBS-C reported significant 33% and 24% improvements for the number of weekly SBM from baseline when consuming either L. paracasei or placebo, respectively (both p = 0.007). Although, participants receiving B. longum reported a 19% increase in SBM, it was not statistically significant until week 8, when a 44% improvement was reported (p = 0.03). At week 8, participants with IBS-C consuming L. paracasei continued to report a significant 33% improvement in the number of weekly SBM from baseline ( = 0.03). Conversely, a non-significant 11% decrease in the number of weekly SBM from baseline was reported by the placebo group (Figure 5A,B).

In participants with IBS-C, there were no significant between-group differences in SBM, CSBM, and stool consistency among the three groups (Table 3). However, L. paracasei and B. longum consumption led to increased number of within-group SBM and CSBM. At week 4, participants with IBS-C reported significant 33% and 24% improvements for the number of weekly SBM from baseline when consuming either L. paracasei or placebo, respectively (both p = 0.007). Although, participants receiving B. longum reporteda 19% increase in SBM, it was not statistically significant until week 8, when a 44% improvement was reported (p = 0.03). At week 8, participants with IBS-C consuming L. paracasei continued to report a significant 33% improvement in the number of weekly SBM from baseline (p = 0.03). Conversely, a non-significant 11% decrease in the number of weekly SBM from baseline was reported by the placebo group (Figure 5A,B).

Figure 5. Effect of Lactobacillus paracasei HA-196 and Bifidobacterium longum R0175 on frequency of bowel movements in participants with IBS-C. (A) Weekly SBM at baseline, week 4 and week 8, (B) Change in total SBM from baseline to week 4 and week, (C) Total CSBM at baseline, week 4 and week 8, and (D) Change in weekly CSBM from baseline to week 4 and week 8. L. paracasei group n = 9, B. longum group n = 5, placebo group n = 13; data represented as mean ± SD; * p < 0.05, ** p < 0.01; SBM; spontaneous bowel movement; CSBM, complete spontaneous bowel movement; L. paracasei, Lactobacillus paracasei HA-196; B. longum, Bifidobacterium longum R0175.

IBS-C participants consuming

IBS-C participants consuming L. paracasei reported a significant 49% increase in CSBM at week 4, and a significant 43% increase at week 8 (p = 0.004 and p = 0.02, respectively). Whereas, the placebo group had reported a significant 39% increase (p = 0.04) at week 4 in the number of CSBM, but by the end of the study period a non-significant 2.2% decrease was reported from baseline (Figure 5C,D).

Table 3. Stool consistency at baseline, week 4, and week 8 and change in stool consistency from baseline to week 4, and week for participants with IBS-C in the PP population.

L. paracasei HA-196

B. longum R0175 Placebo Baseline (average of Weeks −2 and −1)

Study Day Statistic

Mean ± SD (n) 2.84 ± 1.43 (10) 2.55 ± 1.57 (5) 2.69 ± 0.78 (12) p-value * 0.9467 0.9656

Week 4 Mean ± SD (n) 3.14 ± 1.05 (10) 3.21 ± 1.15 (5) 2.91 ± 0.82 (13) Week 8 Mean ± SD (n) 3.21 ± 1.32 (10) 2.44 ± 0.74 (5) 2.49 ± 0.95 (12)

Change from Baseline to Week 4

Mean ± SD (n) 0.44 ± 1.09 (9) 0.49 ± 1.74 (5) −0.05 ± 0.80 (12) Within Group p-Value +

p = 0.221 (r) p = 0.210 (r) p = 0.295 (r) Between Group p-value **

0.2585 (r) 0.9940 (r) Change from Baseline to Week 8

Mean ± SD (n) 0.62 ± 1.12 (9) −0.28 ± 1.54 (5) −0.41 ± 1.00 (11) Within Group p-Value +

p = 0.186 (r) p = 0.789 (r) p = 0.858 (r) Between Group p-value **

0.0721 (r) 0.9805 (r)

n, number; SD, standard deviation; Min, minimum; Max, maximum; * For Baseline (Day 0), p-values for comparison of each probiotic strain to placebo generated by ANOVA with Group as a fixed effect and subject as a random effect. Dunnett’s method was used to adjust for multiple comparisons; ** Values generated from a Repeated Measures ANCOVA with baseline as a covariate and Group, Study Day and Group by Study Day interaction as fixed effects with subject as a random effect. p-values for comparison to placebo were adjusted for multiple comparisons using Dunnett’s method; + Within group p-values generated from the Repeated Measures ANCOVA specified above; (r) indicates values were ranked prior to generating ANOVA or ANCOVA.

1. 3.4. Decrease in the Bowel Movement Frequency in Participants with IBS-D

In participants with IBS-D, L. paracasei and B. longum consumption led to lower within group SBM and CSBM frequencies. The L. paracasei group reported a significant 21% decrease in SBM from baseline to week 8 that was decreased compared to placebo at week 8 (p < 0.05) (Figure 6A,B). Additionally, IBS-D participants supplemented with B. longum reported significant decreases in their SBM stool frequency compared to placebo at week 4 (Figure 6B). With respect to CSBM, there were no significant within-group differences in participants with IBS-D. The greatest reduction, however, in CSBM was reported by participants receiving L. paracasei (24%), followed by the B. longum group (8%); whereas participants in the placebo group reported a 13.5% increase from baseline to week 8 (Figure 6C,D).

Lactobacillus paracasei HA-196 and Bifidobacterium longum R0175 on frequency of bowel movements in participants with IBS-D. (A) Weekly SBM at baseline, week 4 and week 8, (B) Change in total SBM from baseline to week 4 and week 8, (C) Total CSBM at baseline, week 4 and week 8, and ( ) Change in weekly CSBM from baseline to week 4 and week 8. group n 10, B. longum group n = 15, placebo group n = 12; data represented as mean ± SD; * p < 0.05, ** p < 0.01; SBM; spontaneous bowel movement; CSBM, complete spontaneous bowel movement; L. paracasei, Lactobacillus paracasei HA-196; B. longum, Bifidobacterium longum

Furthermore, IBS-D participants in the L. paracasei

a significant 15% within group decrease in stool consistency after 8 weeks (Table 4). There were no significant between-group differences for CSBM or stool consistency among the three groups at the end of 4 or 8 weeks of the supplementation period.

Furthermore, IBS-D participants in the L. paracasei and placebo groups reported significant decreases in stool consistency at 4 weeks; however, the L. paracasei group was the only group to report a significant 15% within group decrease in stool consistency after 8 weeks (Table 4). There were no significant between-group differences for CSBM or stool consistency among the three groups at the end of 4 or 8 weeks of the supplementation period.

Table 4. Stool consistency at baseline, week 4, and week 8 and change in stool consistency from baseline to week 4, and week 8 for participants with IBS-D in the PP population.

Study Day Statistic L. paracasei HA-196 B. longum R0175 Placebo Baseline (average of Weeks −2 and −1)

Mean ± SD (n) 4.34 ± 0.65 (10) 4.46 ± 0.89 (15) 4.79 ± 0.77 (13) p-value * 0.3106 0.4416

Week 4 Mean ± SD (n) 3.73 ± 0.75 (10) 4.13 ± 0.64 (15) 4.26 ± 0.57 (12) Week 8 Mean ± SD (n) 3.68 ± 0.95 (10) 4.22 ± 0.97 (15) 4.66 ± 0.76 (12)

Change from Baseline to Week 4

Mean ± SD (n) −0.39 ± 0.80 (10) −0.36 ± 1.01 (15) −0.60 ± 0.74 (12) Within Group p-Value + p = 0.023 p = 0.132 p = 0.016

Between Group p-value ** 0.8523 0.9868 Change from Baseline to Week 8

Mean ± SD (n) −0.45 ± 1.14 (10) −0.26 ± 1.12 (15) −0.20 ± 0.78 (12) Within Group p-Value + p = 0.013 p = 0.281 p = 0.450

Between Group p-value ** 0.2896 0.5955

n, number; SD, standard deviation; Min, minimum; Max, maximum; * For Baseline (Day 0), p-values for comparison of each probiotic strain to placebo generated by ANOVA with Group as a fixed effect and subject as a random effect. Dunnett’s method was used to adjust for multiple comparisons; ** Values generated from a Repeated Measures ANCOVA with baseline as a covariate and Group, Study Day and Group by Study Day interaction as fixed effects with subject as a random effect. p-values for comparison to placebo were adjusted for multiple comparisons using Dunnett’s method; + Within group p-values generated from the Repeated Measures ANCOVA specified above; (r) indicates values were ranked prior to generating ANOVA or ANCOVA.

1. 3.5. No Significant Changes in the Bowel Movement Frequency in Participants with IBS-M
2. 3.6. Improvements in Mental Health Measures in Participants Given Probiotics

There were no significant within group differences in HADS score, but participants receiving

1. B. longum reported the greatest reduction (4%, p = 0.09) in HADS score after 8 weeks (Figure 8A). As assessed by the SF-36, participants supplemented with B. longum reported a significant increase of 12% (p < 0.001) and those in the L. paracasei group reported a non-significant 3.7% increase in energy from baseline to week 8 (Figure 8B). Evaluation of the emotional well-being parameter by SF-36 showed, significant 3% and 4% improvements in the L. paracasei and B. longum groups, respectively, from baseline to week 8 (all p < 0.05, Figure 8C). Participants receiving L. paracasei and B. longum reported significant 8% and 6% within-group increases in social functioning, respectively (all p < 0.05, Figure 8D). These within-group significant improvements in SF-36 parameters were not reported by participants given placebo. There were also no significant between-group differences among the three groups in the SF-36 parameters assessed.

Figure 8. Effect of Lactobacillus paracasei HA-196 and Bifidobacterium longum R0175 on psychological symptoms in participants with IBS from baseline to week 4 and week 8 (A) Mean change from baseline in HADS score, (B) Mean change from baseline in SF-36 energy levels, (C) Mean change from baseline in SF-36 emotional well-being, and ( ) Mean change from baseline in SF-36 social functioning.

paracasei group n = 84, B. longum group n = 86, placebo group n = 81; data represented as mean ± SD; * p < 0.05, ** p < 0.01, *** p < 0.001; HADS, Hospital Anxiety and Depression Scale; SF, Short Form; L. paracasei, Lactobacillus paracasei HA-196; B. longum Bifidobacterium longum R0175.

1. 3.7. L. paracasei and B. longum are Safe to Consume
2. 3.8. High Compliance and Increased Number of Participants with Bifidobacterium Species In addition to assessing compliance based on counting capsules at each study visit, qPCR

1. 3.7. L. paracasei and B. longum are Safe to Consume
2. 3.8. High Compliance and Increased Number of Participants with Bifidobacterium Species

Compliance was found to be of 85% and 83% for participants receiving B. longum and L. paracasei respectively. Probiotic strains were detected in their respective groups and in nine and eleven participants of the placebo group at baseline and the end of study, respectively (Figure 9A–C). We observed no significant changes in fecal Bifidobacterium species levels at baseline and at the end of study (median ~9 Log bacteria/g of wet feces). Notably, following B. longum administration, the number of participants without detection of Bifidobacterium species was only 4 participants at the end of the supplementation period (Figure 9D). There were no significant within group changes in relative quantification of A. muciniphila and F. prausnitzii in any of the study groups (Figure 9E). There were no differences in L. paracasei, B. longum or Bifidobacterium species levels in participants with IBSC, -D or -M after supplementation (Supplementary Figure S1).

In addition to assessing compliance based on counting capsules at each study visit, qPCR analysis wasusedtoquantifyprobioticstrainsinfecalsamplesandestablishparticipantcompliance. Compliance was found to be of 85% and 83% for participants receiving B. longum and L. paracasei, respectively. Probiotic strains were detected in their respective groups and in nine and eleven participants of the placebo group at baseline and the end of study, respectively (Figure 9A–C). We observed no significant changes in fecal Bifidobacterium species levels at baseline and at the end of study (median ~9 Log bacteria/g of wet feces). Notably, following B. longum administration, the number of participants without detection of Bifidobacterium species was only 4 participants at the end of the supplementation period (Figure 9D). There were no significant within group changes in relative quantification of A. muciniphila and F. prausnitzii in any of the study groups (Figure 9E). There were no differences in L. paracasei, B. longum or Bifidobacterium species levels in participants with IBS-C, -D or -M after supplementation (Supplementary Figure S1).

Figure 9. Effect of Lactobacillus paracasei HA-196 and Bifidobacterium longum R0175 on fecal microbial composition in participants with IBS. qPCR analysis of absolute quantification of B. longum (red dots) and L. paracasei (blue dots) in participants receiving (A) L. paracasei (B) B. longum, or (C) placebo; (D) Absolute quantification of Bifidobacterium species at baseline and 8 weeks with limit of quantification for each species represented by corresponding coloured dotted line; (E) relative quantification of Akkermansia muciniphila and Faecalibacterium prausnitzii in the three groups. A. muciniphila, Akkermansia muciniphila; F. prausnitzii, Faecalibacterium prausnitzii; B. longum, Bifidobacterium longum; L. paracasei, Lactobacillus paracasei.

In this randomized, double-blind, placebo-controlled study, both L. paracasei HA-196 and B. longum R0175 were found to be safe, and significantly improved participants’ experience of IBS-related symptoms after 8 weeks as reflected by reduced scores on the IBS-SSS. A similar decrease in symptom severity was also observed in the placebo group. The significant reduction in symptom severity was marked by a 30% decrease in L. paracasei and a 22% decrease in the B. longum groups compared to a 31% decrease in the placebo group. IBS-SSS is a widely used self-administered questionnaire that measures severity and frequency of abdominal pain, abdominal distension/tightness, bowel habit satisfaction, and quality of life. The total IBS-SSS score ranges from 0 to 500, with higher scores indicating more severe symptoms. A decrease in the IBS-SSS score indicates a reduction in symptom severity. When participants were grouped into IBS subtypes (IBS-C, IBS-D, IBS-M), a reduction in IBS-SSS was also observed in all three groups. However, L. paracasei supplemented IBS-C and IBS-D participants reported improvements in their bowel habits and stool consistency that were not observed in the placebo group.

In this randomized, double-blind, placebo-controlled study, both L. paracasei HA-196 and B. longum R0175 were found to be safe, and significantly improved participants’ experience of IBSrelated symptoms after 8 weeks as reflected by reduced scores on the IBS-SSS. A similar decrease in symptom severity was also observed in the placebo group. The significant reduction in symptom severity was marked by a 30% decrease in L. paracasei and a 22% decrease in the B. longum groups compared to a 31% decrease in the placebo group. IBS-SSS is a widely used self-administered questionnaire that measures severity and frequency of abdominal pain, abdominal distension/tightness, bowel habit satisfaction, and quality of life. The total IBS-SSS score ranges from

1. 0 to 500, with higher scores indicating more severe symptoms. A decrease in the IBS-SSS score indicates a reduction in symptom severity. When participants were grouped into IBS subtypes (IBSC, IBS-D, IBS-M), a reduction in IBS-SSS was also observed in all three groups. However, L. paracasei supplemented IBS-C and IBS-D participants reported improvements in their bowel habits and stool consistency that were not observed in the placebo group.

In addition to IBS-SSS score, BM frequency was examined based on IBS subtype. At the end of the 8-week study, IBS-C participants in the L. paracasei group had significant improvements in both SBM and CSBM. Although not statistically significant, IBS-C participants in the placebo group reported fewer numbers of SBM and CSBM at the end of the study. For IBS-D participants, B. longum significantly reduced SBM compared to placebo from baseline to week 4, but this significant betweengroup difference was not sustained to the end of study. L. paracasei significantly improved the weekly frequency of SBM at week 8 in IBS-D participants compared to the placebo. Additionally, L. paracasei had a greater effect on reducing the number of CSBM in IBS-D participants than B. longum. As with IBS-C participants, the placebo did not alter bowel movement frequency with IBS-D. Although, not significantly different, L. paracasei supplemented IBS-M participants were the only group to report increases in SBM and CSBM after 8 weeks. IBS-M participants in the B. longum group did not report

In addition to IBS-SSS score, BM frequency was examined based on IBS subtype. At the end of the 8-week study, IBS-C participants in the L. paracasei group had significant improvements in both SBM and CSBM. Although not statistically significant, IBS-C participants in the placebo group reported fewer numbers of SBM and CSBM at the end of the study. For IBS-D participants, B. longum significantly reduced SBM compared to placebo from baseline to week 4, but this significant between-group difference was not sustained to the end of study. L. paracasei significantly improved the weekly frequency of SBM at week 8 in IBS-D participants compared to the placebo. Additionally, L. paracasei had a greater effect on reducing the number of CSBM in IBS-D participants than B. longum. As with IBS-C participants, the placebo did not alter bowel movement frequency with IBS-D. Although, not significantly different, L. paracasei supplemented IBS-M participants were the only group to report

increases in SBM and CSBM after 8 weeks. IBS-M participants in the B. longum group did not report increases in SBM in participants, and those in the placebo had decreases in CSBM by the end of the study.

All participants with IBS who received L. paracasei reported improvements in stool consistency after 8 weeks, unlike IBS-C and IBS-M participants in the B. longum group as well as all participants given placebo. This shift in stool consistency in the L. paracasei groups was marked by a significant within-group reduction in stool consistency reported by IBS-D participants after 8 weeks. The monitoring of stool consistency and BM frequencies provide evidence of efficacy of the two probiotic strains, with L. paracasei being more efficacious in improving symptoms of IBS, particularly IBS-C and IBS-D.

AnimportantaspectofthisstudywasthatofallIBSsubtypes, thehighestpopulationofparticipants enrolled was those with IBS-M. IBS as a condition can be very transient in presentation of its type and intensity [27]. Participants in this study often changed their subtype between screening and baseline run-in. Furthermore, the recording of IBS intensity is highly subjective making clinical studies in this area challenging. Despite the challenges associated with studies in this area, this study suggests that both L. paracasei HA-196 and B. longum R0175 play a role in mitigating the effects of IBS in certain subtypes. These data would suggest that probiotic efficacy may be dependent on the IBS subtype. This warrants further investigation and studies focused on specific IBS symptoms within the IBS symptomatology continuum rather than considering all IBS patients as a homogenous population.

Psychological comorbidities are commonly observed in IBS patients, who are significantly more likely to develop mental health conditions such as depressive disorders, anxiety, and sleep disorders [28]. Considering this, probiotics consumption is viewed as a means to alleviate the psychological comorbidities of IBS [29]. However, similar to the efficacy of probiotics and the physical symptoms of IBS, there is heterogeneity among studies linking probiotics and modulation of behaviour in humans. A narrative review of seven systematic reviews of RCTs in individuals with depressive symptoms noted that five reviews reported a beneficial effect of probiotics on depressive symptoms. However, the authors stated that heterogeneity in the type of populations studied warranted further studies in clinically diagnosed depression before more definitive conclusions could be made [30]. More recently, a study on IBS participants with mild to moderate anxiety and/or depression reported that B. longum NCC3001 improved depression scores based on the HADS scale. A reduction of two or more points was reported in 64% of the participants in the probiotic group compared to 32% in the placebo group (p = 0.04) [31], although anxiety or IBS symptoms scores were not different compared to placebo. In the current study, we did not observe any significant improvements in the HADS score nor in general health assessment (based on the SF-36) among IBS participants compared to placebo. However, participants in the B. longum group reported having increased energy levels after 8 weeks, which can directly impact their willingness and ability to perform everyday tasks. Furthermore, participants in both B. longum and L. paracasei groups reported significant within-group improvements in social functioning after 8 weeks. In contrast, none of these improvements pertaining to mental health were found in participants receiving the placebo.

Ensuring compliance with the probiotic for the duration of the study is an essential component of examining efficacy. The ability to detect the probiotic strains in participant’s fecal samples can confirm the participant compliance calculated from the returned IP. Fecal samples from baseline and

1. 8 weeks were analyzed by qPCR and revealed a high degree of compliance within the probiotic supplemented groups. In the placebo group, probiotic strain detection was not expected. However, a handful of participants in the placebo group did have detection, which may potentially be due to the participants harbouring similar sequences from the bacteria in their gut, which may occur naturally due to bacterial diversity.

The abundance of selected relevant bacteria, including A. muciniphila, F. prausnitzii and Bifidobacterium species was also investigated in fecal samples by qPCR. Compared to baseline, participants in all three treatment groups had no significant changes in the two species and one

genus quantified, though more participants had detection of Bifidobacterium species in their fecal matter following 8-weeks of B. longum supplementation. The latter may have contributed to the BM improvement observed in the B. longum supplemented groups, as Bifidobacterium species have been shown to improve bowel habits in all IBS subtypes [32]. An increase in A. muciniphila has been associated with a decrease in abdominal pain in IBS patients, and F. prausnitzii has been shown to reduce visceral hypersensitivity in a non-inflammatory animal model of IBS [33,34]. Thus, it is possible that the unchanged relative abundance of A. muciniphila and F. prausnitzii following probiotic supplementation may have contributed to the absence of significant differences in the IBS-SSS score between the groups. It also suggests that the potential mechanism by which L. paracasei, exerted its impact on bowel habits and stool consistency may be independent of the specific bacterial species investigated.

A placebo effect of 33% was observed in this study, consistent with reports from previous IBS clinical trials. A review of RCTs evaluating complementary health products showed an overall placebo response of 43%, with higher placebo responses associated with longer treatment duration and greater number of clinic visits [35]. A meta-analysis reported that the placebo response ranges from 16% to 71% in participants with IBS, with a population-weighted average of 40% in studies using natural health products and pharmaceutical interventions [36]. The subjective nature of the assessments and the intrinsic nature of IBS patients who display high levels of suggestibility and a strong placebo effect may have contributed to the placebo effect seen in this study. Additionally, this placebo effect may partially explain why some outcomes assessed were significant at either week 4 or 8. The placebo effect in studies within this area of research also exemplifies the importance of a run-in period, which was included in this study. However, examining bowel habits for a longer period prior to intervening may be an important aspect of understanding probiotic efficacy while mitigating the placebo effect.

Other limitations of this study include the relatively small sample size of the population, more specifically when considering the IBS subtypes, which may have introduced variability in the responses and reduced effect sizes. Moreover, the severity of IBS symptoms using IBS-SSS was only captured at baseline, week 4, and week 8 and not at any of the intermediate weeks. A characteristic feature of IBS is that some weeks were worse than the others, and due to the lack of data from intermediate weeks, temporary improvements experienced by participants associated with probiotic supplementation may have been missed. Furthermore, the duration of IBS in participants prior to enrollment into the study was not accounted for, which may have influenced the outcomes measured. Additionally, participants in this study were diagnosed and classified by IBS subtype according to Rome III, which has now been updated to Rome IV [37]. A recent study has shown differences in the diagnosis and classification of participants with IBS based on Rome III or Rome IV [38].Of note, at the time of classification based on the Rome III questionnaire, no participants in this study were classified as the IBS-unclassified (IBS-U) subtype.

Low grade inflammation has been suggested to play a role in the pathogenesis of IBS. Abnormal mast cell activation in the gut has been associated with the development of IBS symptoms [39]. Although beyond the scope of this study, it would be important to evaluate the anti-inflammatory effect of L. paracasei or B. longum in IBS patients. Overall, the results of the current study suggest a clinically relevant role for L. paracasei or B. longum as probiotics to reduce symptoms in IBS. Furthermore, they may add to the armamentarium of IBS management tools without the safety concerns that may accompany pharmaceuticals. The effects of each probiotic strain may be dependent on the IBS subtype, and future studies should target IBS-D and IBS-M for probiotic therapy.

Supplementary Materials: The following are available online at http://www.mdpi.com/2072-6643/12/4/1159/s1, Table S1: CONSORT 2010 checklist of information to include when reporting a randomised trial. Table S2: Number of rescue medication pills taken by participants with IBS-C, -D or -M during the 8-week intervention period. Figure S1: L. paracasei, B. longum or Bifidobacterium species levels in participants with IBS-C, -D or -M after supplementation.

Author Contributions: Conceptualization, T.A.T. and M.E.; methodology, T.A.T., M.E., A.P., D.C.C.; validation, M.E., J.M.A.; formal analysis, M.E., J.M.A.; investigation, D.C.C.; data curation, D.C.C.; manuscript—writing original draft, E.D.L., M.E.; Manuscript—review and editing, A.P., R.B. E.D.L., J.M.A., M.E.; visualization, E.D.L.,

R.B.; supervision, M.E., D.C.C.; funding acquisition, T.A.T., R.B., A.P. All authors have read and agreed to the published version of the manuscript.

Funding: This research was funded by Lallemand Health Solutions (LHS).

Acknowledgments: We would like to thank Mitchell Hendry, Jeremie Auger, and Justin Knapp, Sharif Shajib and Annie Tremblay for their valued discussions in the preparation of this manuscript. We also thank Jeremie Auger for data analysis.

Conflicts of Interest: LHS owns Bifidobacterium longum R0175 and Lactobacillus paracasei HA-196. TAT, RB and AP are employees of Lallemand Health Solutions. LHS was involved in the design of the study, revising the manuscript and decision to publish the results but was not involved in data collection and analyses, or writing the manuscript. The corresponding author oversaw collection and analysis of data and had final responsibility for the decision to submit for publication.