Assessment of psychotropic-like properties of a probiotic formulation (<i>Lactobacillus helveticus</i>R0052 and<i>Bifidobacterium longum</i>R0175) in rats and human subjects

The researchers tested a freeze-dried probiotic formulation (PF) containing two bacterial strains: Lactobacillus helveticus R0052 (3 × 10⁹ colony-forming units per dose) and Bifidobacterium longum R0175 (3 × 10⁹ CFU per dose), for a total of 6 × 10⁹ CFU per day. The comparator was an identical-looking placebo (maltodextrin and magnesium stearate). The study had two parts: a preclinical rat experiment and a human clinical trial.

In the rat experiment, the outcome was anxiety-like behaviour measured by the conditioned defensive burying test — a standard screening model where rats that are more anxious spend more time burying a noxious stimulus (an electrified probe). In the human trial, the primary outcomes were psychological distress (measured by the Hopkins Symptom Checklist-90, or HSCL-90), anxiety and depression (Hospital Anxiety and Depression Scale, or HADS), perceived stress (Perceived Stress Scale), coping strategies (Coping Checklist, or CCL), and a biological stress marker (24-hour urinary free cortisol, or UFC).

Preclinical (rats): 24 adult male Wistar rats, housed individually under standard laboratory conditions (12-hour light/dark cycle, 22°C, food and water ad libitum). No specific strain details beyond Wistar were given.

Clinical (humans): 55 healthy adult volunteers (25 men, 30 women) aged 30–55 years (mean age 42.3 years, SD not reported). All were non-smokers, had no history of psychiatric or gastrointestinal disorders, were not taking any medication (including psychotropics, probiotics, or antibiotics), and had no recent stressful life events. They were recruited from the general population in France. The study was conducted at a single clinical research centre.

Rat experiment: The conditioned defensive burying test. Rats were placed in a chamber with an electrified probe. When they touched the probe, they received a mild shock. The primary outcome was the time spent burying the probe with bedding material over a 15-minute period — more burying time indicates more anxiety-like behaviour. This test is validated for detecting anxiolytic (anti-anxiety) effects of drugs like benzodiazepines.

Human trial: Five instruments were used:

• Hopkins Symptom Checklist-90 (HSCL-90): A 90-item self-report questionnaire measuring psychological distress across 9 subscales (somatisation, obsessive-compulsive, interpersonal sensitivity, depression, anxiety, anger-hostility, phobic anxiety, paranoid ideation, psychoticism). The Global Severity Index (GSI) is the average score across all items (range 0–4, higher = worse). The Positive Symptom Distress Index (PSDI) measures intensity of symptoms, and the Positive Symptom Total (PST) counts how many symptoms are present.

• Hospital Anxiety and Depression Scale (HADS): A 14-item self-report scale with two subscales: anxiety (7 items) and depression (7 items). Each item scored 0–3, giving subscale ranges of 0–21 (higher = worse). A score of 8–10 is borderline, ≥11 is clinical caseness.

• Perceived Stress Scale (PSS): A 14-item scale measuring how unpredictable, uncontrollable, and overloaded respondents find their lives over the past month. Scores range 0–56 (higher = more perceived stress).

• Coping Checklist (CCL): A 26-item scale measuring three coping strategies: problem-focused coping (active problem-solving), emotion-focused coping (managing emotions), and seeking social support. Higher scores indicate greater use of that strategy.

• 24-hour urinary free cortisol (UFC): A biological marker of hypothalamic-pituitary-adrenal (HPA) axis activity. Participants collected all urine over 24 hours. Cortisol was measured by radioimmunoassay. Higher UFC indicates higher chronic stress activation.

Study design: This was a two-part study. The rat experiment was a between-subjects design (probiotic vs. placebo, n=12 per group) with 2 weeks of daily administration before testing. The human trial was a double-blind, placebo-controlled, randomised parallel-group design with 30 days of daily administration.

Randomisation: Human volunteers were randomly assigned to either the probiotic group (n=28) or placebo group (n=27) using a computer-generated randomisation list. No details on allocation concealment were provided.

Blinding: Both participants and researchers were blinded. The probiotic and placebo were identical in appearance, taste, and packaging. The code was broken only after all data were collected and analysed.

Duration: Rats received the probiotic for 14 days before testing. Humans received the probiotic for 30 days, with assessments at baseline (day 0) and after 30 days of treatment.

Statistical approach: For the rat data, the authors used a Mann-Whitney U test (non-parametric, appropriate for small samples). For human data, they used analysis of covariance (ANCOVA) with baseline scores as covariates, plus paired t-tests for within-group comparisons. They also reported effect sizes (Cohen's d) for some comparisons, though not consistently. Significance was set at p < 0.05, with trends reported at p < 0.10.

What this design can and cannot prove:

This design can prove that the probiotic caused the observed changes in mood and stress markers, because it was a randomised, double-blind, placebo-controlled trial. Randomisation ensures that known and unknown confounders are balanced between groups at baseline. Blinding prevents expectation effects from both participants and researchers. The parallel-group design avoids carryover effects that could occur in crossover designs.

However, this design cannot prove:

• Long-term effects: 30 days is a short intervention period. We don't know if effects persist, increase, or diminish with longer use.
• Mechanism: The study shows that the probiotic affects mood, but doesn't prove how — whether through the vagus nerve, immune modulation, neurotransmitter production, or other pathways.
• Generalisability to clinical populations: These were healthy volunteers. Results may not apply to people with diagnosed anxiety or depression.
• Dose-response: Only one dose was tested. We don't know if higher or lower doses would work better.
• Strain specificity: The combination was tested, not individual strains. We can't tell which strain (or both) is responsible.

Major methodological weaknesses:

• Small sample size: 55 total (28 probiotic, 27 placebo) is modest. The study may be underpowered to detect small-to-moderate effects, especially on subscales.
• No washout period: Because it's a parallel-group design, this isn't a problem for carryover, but there was no follow-up after treatment stopped, so we don't know how long effects last.
• Multiple comparisons: The authors tested many outcomes (HSCL-90 has 9 subscales plus 3 global indices, HADS has 2 subscales plus global, plus PSS, CCL, and UFC). They did not adjust for multiple comparisons (e.g., Bonferroni correction), which inflates the risk of false positives. Some of the "significant" results may be chance findings.
• Industry funding: The study was funded by the company that manufactures the probiotic (Institut Rosell-Lallemand). While this doesn't invalidate the results, it's a potential source of bias in study design, analysis, and reporting.
• No objective adherence monitoring: Adherence was assessed by self-report and returned sachet counts, which can be unreliable.

Rat experiment:

• Probiotic-treated rats spent significantly less time burying the probe compared to placebo-treated rats (mean burying time: probiotic ~40 seconds vs. placebo ~120 seconds; p < 0.05, Mann-Whitney U test). This indicates reduced anxiety-like behaviour.
• No differences were observed in general locomotor activity, suggesting the effect was specific to anxiety rather than sedation or motor impairment.

Human trial — Primary outcomes (HSCL-90):

• Global Severity Index (GSI): Probiotic group decreased from baseline mean 0.58 to 0.42 after 30 days; placebo group decreased from 0.55 to 0.51. The difference between groups was statistically significant (p < 0.05, ANCOVA). Effect size (Cohen's d) was approximately 0.45 (moderate).
• Somatisation subscale: Probiotic group decreased from 0.62 to 0.40; placebo from 0.60 to 0.55. Significant difference (p < 0.05).
• Depression subscale: Probiotic group decreased from 0.52 to 0.33; placebo from 0.50 to 0.47. Significant difference (p < 0.05).
• Anger-hostility subscale: Probiotic group decreased from 0.48 to 0.28; placebo from 0.45 to 0.42. Significant difference (p < 0.05).
• Anxiety subscale: Probiotic group decreased from 0.55 to 0.38; placebo from 0.52 to 0.48. This difference was not statistically significant (p > 0.05).
• Other subscales (obsessive-compulsive, interpersonal sensitivity, phobic anxiety, paranoid ideation, psychoticism): No significant differences between groups.

Human trial — Secondary outcomes:

• HADS global score: Probiotic group decreased from mean 11.2 to 8.4; placebo from 10.8 to 10.1. Significant difference (p < 0.05).
• HADS-anxiety subscale: Probiotic group decreased from 6.8 to 4.9; placebo from 6.5 to 5.8. This was a trend, not significant (p < 0.06).
• HADS-depression subscale: Probiotic group decreased from 4.4 to 3.5; placebo from 4.3 to 4.3. Not significant (p > 0.05).
• Perceived Stress Scale (PSS): Probiotic group decreased from mean 24.1 to 20.3; placebo from 23.8 to 22.5. Not significant (p > 0.05).
• Coping Checklist — Problem-solving: Probiotic group increased from 28.5 to 31.2; placebo decreased from 28.8 to 27.6. Significant difference (p < 0.05). This means the probiotic group reported using more active problem-solving strategies after treatment.
• Coping Checklist — Emotion-focused and social support: No significant differences.
• 24-hour urinary free cortisol (UFC): Probiotic group decreased from mean 68.4 nmol/24h to 52.1 nmol/24h; placebo increased from 65.2 to 67.8 nmol/24h. Significant difference (p < 0.05). This represents a ~24% reduction in the probiotic group vs. a ~4% increase in placebo.

To translate these numbers into plain English:

• Psychological distress (GSI): The probiotic group's distress scores dropped by about 28% (from 0.58 to 0.42), while placebo dropped by only 7% (0.55 to 0.51). This is roughly equivalent to moving from "mildly distressed" to "minimally distressed" on the scale. For context, a 0.16-point drop on a 0–4 scale is noticeable in daily life — fewer headaches, less muscle tension, less irritability.

• Depression symptoms: The probiotic group's depression scores dropped by 37% (0.52 to 0.33), compared to 6% in placebo. This is like going from "occasionally feeling down" to "rarely feeling down."

• Anger-hostility: A 42% reduction in the probiotic group (0.48 to 0.28) vs. 7% in placebo. This means fewer angry outbursts, less irritability, and more patience.

• Cortisol: A 24% drop in urinary free cortisol in the probiotic group (68.4 to 52.1 nmol/24h) vs. a 4% increase in placebo. This is a substantial biological change — comparable to what you might see after 8 weeks of mindfulness meditation or regular aerobic exercise. Normal UFC range is roughly 30–100 nmol/24h, so the probiotic group moved from the upper-middle to the lower-middle of the normal range.

• Problem-solving coping: The probiotic group's score increased by about 10% (28.5 to 31.2), while placebo decreased by 4% (28.8 to 27.6). This means people taking the probiotic reported being more likely to actively address problems rather than avoid them.

Acknowledged by authors:

• The study was short (30 days) and used only one dose.
• The sample was healthy volunteers, so results may not generalise to clinical populations.
• The rat model is a screening tool, not a direct analogue of human anxiety.
• The mechanism of action was not investigated.

Critical reader observations:

• Multiple comparisons problem: With 15+ outcome measures and no correction for multiple testing, some of the "significant" results (p < 0.05) are likely false positives. The fact that the HADS-anxiety subscale was p < 0.06 (a trend) while the HADS-depression subscale was non-significant, despite the global HADS being significant, suggests the global result may be driven by the anxiety subscale — but the authors don't clarify this.
• Small sample size: With only 55 participants, the study has limited statistical power. The effect sizes are moderate (Cohen's d ~0.45), which means the study could detect only moderate-to-large effects. Smaller but still meaningful effects would be missed.
• No intention-to-treat analysis: The authors don't report how many participants completed the study vs. dropped out. If dropouts were uneven between groups, this could bias results.
• Self-report bias: All psychological outcomes were self-reported. While blinding should control for expectation effects, people who notice changes in their digestion (a known effect of this probiotic) might guess they're in the active group and report mood changes accordingly.
• Industry funding: The study was funded by Institut Rosell-Lallemand, the manufacturer of the probiotic. Industry-funded studies are more likely to report positive results. The authors do not declare any conflicts of interest, but the funding source is a potential bias.
• No dietary control: Participants' diets were not standardised. Changes in diet over 30 days could affect gut microbiota and mood independently of the probiotic.
• No microbiome analysis: The study didn't measure whether the probiotic actually colonised the gut or changed the composition of the gut microbiota. We don't know if the effects are due to the probiotic itself or to changes in the existing microbial community.
• Publication bias: This is a single study. Positive results are more likely to be published than null results. Replication is needed.

For someone running their own n=1 experiment:

What to test:

• The specific probiotic combination: Lactobacillus helveticus R0052 + Bifidobacterium longum R0175, at a total dose of 6 × 10⁹ CFU per day (3 × 10⁹ of each strain). This is available commercially in some probiotic products