Research on probiotic supplementation in the prevention and treatment of depressive and anxiety disorders and stress coping skills has been conducted for many years. This research project was designed to investigate the effects of both, probiotic supplementation and diet quality on a wide range of psychological, microbiological, biochemical, and clinical outcomes. The aim of the study was to assess the effects of targeted supplementation with Lactobacillus helveticus and Bifidobacterium longum on gut microbiota composition and quality, mental health, perceived stress intensity, psychosomatic response, and coping as well as body composition and bone density in young women who differed in diet quality. The study enrolled 120 healthy women in early adulthood (20-30 years of age) who were equally and randomly assigned to the intervention (Lh+Bl) or placebo (PLA) group. The intervention consisted of the administration of combined freeze-dried strains of Bifidobacterium longum Rosell®-175 and Lactobacillus helveticus Rosell®-52 (Lh+Bl) at dose of one capsule per day for 12 weeks. The total bacterial content per capsule was 3x109 CFU. At baseline, dietary habits were assessed using the KomPAN FFQ questionnaire and faecal samples were collected for gut microbiota analysis using Illumina next-generation sequencing technology. Anthropometric measurements were taken, and body composition and bone density were assessed using DEXA and displacement plethysmography using Bod Pod. Perceived stress and mental health were assessed using the psychological scales: PSS 10, Mini Cope, GHQ 28, and SOC-13. Salivary cortisol, DHEA and cortisol:DHEA ratio were measured as physiological markers of stress. After 12 weeks of supplementation, measurements were repeated and compared with baseline values.

Introduction The number of people receiving psychiatric services has been increasing over the past few years. In Poland, the number of people with mental disorders in outpatient care has increased by 150,000 in recent years, reaching a total of more than 1.2 million patients. The number of people requiring psychological and psychiatric care is expected to grow in the coming decades. The psychopharmacotherapy used does not always produce the expected treatment results. As a result, more and more attention is being paid to non-pharmacological interactions. In recent years, there has been a growing body of research on the relationship between nutrition and its impact on the mental state of people with mental disorders. Food and nutrition science, psychodietetics and nutripsychiatry, thriving scientific fields, are gaining prominence. Research findings indicate that nutrition and stress-coping resources affect an individual's health. The role of the probiotic bacteria Lactobacillus Helveticus and Bifidobacterium longum as part of the human microbiome has also been studied in this sphere, and results from a growing number of papers confirm their wide range of applications and positive effects on various domains of health. Previous studies have shown that supplementation with Bifidobacterium longum or a composite of them with other bacterial strains positively affected, among other things, the bone health of postmenopausal women, alleviated gastrointestinal discomfort, reduced perceived stress, and improved depression, anxiety, and mood indicators in adults of various ages. On the other hand, supplementation with a composite of several bacterial strains, including Lactobacillus Helveticus and Bifidobacterium longum, affected reducing fat mass in obese and overweight individuals. Probiotic supplementation appears to have great potential for improving health and quality of life, including the prevention and treatment of depressive and anxiety disorders and the ability to cope with stress. Nevertheless, there is still a lack of research in this area, especially concerning Lactobacillus Helveticus alone and the combination of Lactobacillus Helveticus and Bifidobacterium longum strains.

With this in mind, a research project was planned to investigate the effects of both probiotic supplements and dietary quality on a wide range of psychological, microbiological, biochemical, and clinical outcomes. The study aimed to evaluate the effects of targeted supplementation with a new composition of probiotic strains Lactobacillus Helveticus and Bifidobacterium longum on gut microbiota composition and quality, stress-related neuroendocrine markers, mental health, perceived stress intensity, psychosomatic response and coping strategies as well as body composition and bone density in young women who differed in diet quality. We hypothesised that supplementation with probiotic bacteria at a daily dose of 3 x 109 CFU, irrespective of diet quality, would result in changes in the gut microbiota towards higher proportions of beneficial bacteria, reduce stress-related neuroendocrine markers and the intensity of perceived stress and its psychosomatic symptoms, and improve stress-coping strategies, mental health potential and have a beneficial effect on body composition components and bone density in healthy young women, compared with placebo.

Study design and population This randomized, placebo-controlled, 2-arm, parallel, double-blind intervention trial was conducted in a few waves from September 2023 to March 2024 2023 at the Department of Human Nutrition and Dietetics (Poznań Univrsity of Life Sciences). Each wave started during the fall months and was completed during the winter months. The sample size was calculated using GPower software (version 3.1.9.7, Universität Düsseldorf, Germany). Briefly, to detect a significant change in perceived stress level with a two-sided t-test, a significance level of 5%, and a statistical power of 80%, a sample size of 110 was determined, considering a 30% dropout rate. Initially, 120 women were enrolled in this study. These were only women in early adulthood. Eligibility criteria for the study included female sex, age 20-30 years, BMI < 30 kg/m2. Exclusion criteria were: diagnosed chronic and/or serious health disorders (e.g. diabetes, asthma, hypertension, renal diseases), pregnancy, taking any probiotics for 8 weeks prior to enrollment to the study protocol, recent and current antibiotic treatment, ore use of pharmaceutical antidepressant treatment. All participants were healthy volunteers who were fully informed about the study's objectives and provided informed consent for participation before entering the study. The study protocol was reviewed and approved by the Bioethics Committee of the Karol Marcinkowski University of Medical Sciences in Poznan, Poland (permission no. 244/23, issued on March 9, 2023). The study complies with the CONSORT Statement on randomized trials. All procedures followed the guidelines of the 2013 Declaration of Helsinki and Good Clinical Practice.

The enrolled volunteers were evenly, randomly, and blindly assigned to either the intervention (Lh+Bl) group, receiving probiotic bacteria, or the placebo (PLA) group for 12 weeks of supplementation. Randomization was conducted by the supplement manufacturer with computer-generated codes using Research Randomizer (www.randomizer.org). The entire study protocol included familiarization and two laboratory visits in at the Department of Human Nutrition and Dietetics at baseline and at the end of the study (T1, T2; before/after Lh+Bl supplementation [Lh+Bl PRE and Lh+Bl POST] and PLA supplementation [PLAPRE and PLAPOST]). The study's main protocol included an assessment of dietary habits at the T1 study visit. Measurements of weight, height, body composition, bone density, stress response and stress-coping strategies, mental health assessment as well as saliva and stool sampling were conducted at each of the T1 and T2 study visits. The primary outcome measures are microbiota composition, perceived stress intensity and stress-related neuroendocrine markers (salivary cortisol, salivary DHEA and salivary cortisol:DHEA ratio, faecal cortisol); the secondary outcome measures are stress-coping strategies, psychosomatic symptoms, sense of coherence, body composition and bone density). All testing was performed in the morning at the same time for the participant. Participants declared that they had not introduced any changes in their lifestyles, nutrition, or supplementation during the study protocol and were prepared for each visit in the same manner. In addition, telephone contact was maintained in case of non-compliance with the protocol or gastrointestinal symptoms or adverse events.

There were 20 dropouts from the study protocol. The main reasons for dropping out were antibiotic therapy during the protocol (n=10) and discontinuation of the study without a reason before the second visit (n=10). There were 11 dropouts in the Lh+Bl and 9 in the PLA groups. Finally, 100 women (23.5 ± 2.7 years; 61.6 ± 9.4 kg body mass; 1.67 ± 0.06 m height) completed the entire study protocol and were included in the analyses.

Supplementation In the experimental procedure, each participant was supplemented with a chronic dose of 1 capsule per day of Lh+Bl or PLA in a randomized, parallel sequence. Participants were provided with blister packs of Lh+Bl capsules containing 3 x 109 CFU of combined freeze-dried probiotic strains of the probiotic bacteria Bifidobacterium longum Rosell®-175 and Lactobacillus Helveticus Rosell®-52 (trade name: Sanprobi®Stress) or PLA capsules containing an isoenergetic product prepared for the trial. The PLA capsules contained only potato starch and magnesium stearate as fillers. The capsule shell consisted of hydroxypropylmethylcellulose, and the weight of a single capsule was 420 mg. The probiotics were in identical capsules as PLA and could not be distinguished by appearance and taste. Lh+Bl and PLA capsules were obtained from Sanprobi sp. z o.o. sp. k., (Szczecin, Poland). The combination and the concentration of probiotics used in this study were based on the manufacturer's recommendations. The preparations were administrated to each participant in envelopes marked with a unique code. According to the recommended blinding procedure, the formulations were prepared in advance by the manufacturer's laboratory, which was not directly involved in the study. As for double-blinding, neither the researchers nor the participants knew whether Lh+Bl or PLA was administered. Details of randomization were anonymized and disclosed after the protocol's cessation.

Diet and lifestyle

Dietary data were collected using the validated KomPAN® food frequency questionnaire posted on the Google Drive file-sharing platform. The questionnaire included foods commonly consumed by the Polish population, grouped into 25 food categories. Participants were asked to select how often they consumed each type of food in the past 12 months. Available frequency responses were converted in daily frequencies and consisted of never (0 times/day), 1-3 times a month (0.06 times/day), once a week (0.14 times/day), a few times a week (0.5 times/day), once a day (1.0 times/day), or a few times a day (2.0 times/day). The diet quality was described by two diet quality scores: the pro-healthy diet index (pHDI) and the unhealthy diet index (nHDI). Both diet quality scores were calculated as the sum of daily frequencies (in times/day) of food item consumption. The pHDI included 10 items representing potentially pro-healthy foods (whole grain bread, whole grain cereals, milk, fermented milk drinks, cottage cheese, white meat, fish, legumes, fruits, and vegetables) with a total score ranging from 0-20 times/day. The nHDI included 14 food items representing potentially unhealthy foods (white bread, refined cereals, fast food, fried foods, butter, lard, cheese, cold cuts, red meat, sweets, canned meats, sweetened beverages, energy drinks, alcohol) with a total score range 0-28 times/day. Both diet quality scores were recalculated to standardize the total score range to 0-100 points for each. The following formulas were used:

pHDI (in points) = (100/20) × sum of the frequency of consumption of 10 food items (times/day) nHDI (in points) = (100/28) × sum of the frequency of consumption of 14 food items (times/day) Estimated lifestyle behaviors included frequency of daily meals, snacking, salt and sugar use, physical activity, smoking, etc.

Body mass and body composition assessment Anthropometric measurements were taken at the beginning of each study visit to ensure equal conditions for the test procedures. Before body composition analysis, weight and height were measured to the nearest 0.1 cm and 0.1 kg, for height and weight, respectively, using a calibrated scale with a stadiometer (WB3007311, Tanita Corporation, Japan), and body mass index (BMI) was calculated. Body composition analysis (FM and FFM) was performed by both dual-energy X-ray absorptiometry (GE Lunar Prodigy device - General Electrics Healthcare Medical Systems, Europe, Belgium) and displacement plethysmography using Bod Pod (Life Measurement Inc., Concord, CA, USA). All measurements were performed in the morning (7:30-10:00 am) and while fasting (water intake was recommended), with the ambient temperature maintained at 20-22 °C. All participants avoided strenuous exercise for at least 24 hours prior to each visit.

Bone mineral density measurements All study participants underwent dual-energy X-ray absorptiometry (DXA) by the same technician using a GE Lunar Prodigy densitometer (General Electrics Healthcare Medical Systems, Europe, Belgium). The effective dose (μSv) for this densitometer is 0.05. Parameters obtained from the measurement included bone mineral content (BMC, g) and bone mineral density (BMD, g/cm2). BMD Z-scores were generated using the sex-, race-, and age-specific normative DXA database software Enco. The International Society for Clinical Densitometry (ISCD) standards recommend the use of Z-scores to assess BMD in premenopausal women. Bone mineral density (BMD) was compared to healthy young women at each measurement (lumbar spine, total hip, femoral neck, radius) and a T-score was calculated. According to the World Health Organization definition, osteoporosis is indicated by a T-score less than -2.5 SD, and osteopenia by a T-score between -1 and -2.5 SD. Bone parameters were measured according to the adopted densitometry methodology and the International Society for Clinical Densitometry recommendations. A standard quality control program was used, which included checking the DXA equipment to verify system stability and a visual review of all images by an experienced examiner. The scanner was calibrated daily as the DXA GE Lunar Prodigy manufacturer recommended. A standard calibration block was used. The study was performed by a team with the necessary certifications to perform peripheral and axial densitometry and experience in densitometry.

Saliva collection and sample analysis Stress-related neuroendocrine markers, namely salivary cortisol and DHEA, will be measured, as well as the cortisol:DHEA ratio. Prior to saliva collection in the laboratory, participants were instructed to refrain from consuming caffeine and/or nicotine for at least 4 hours and alcohol for at least 24 hours. Unstimulated saliva samples were collected from fasting participants in disposable Eppendorf plastic tubes using the spitting method for 10 minutes while the subjects rested. Participants were first asked to rinse their mouths with water for approximately 30 seconds and 10 minutes prior to providing a saliva sample. Saliva was collected between 9:00 and 11:00 am to minimize circadian differences. Samples of the material (2 tubes per person), immediately after collection, were placed in a freezer at -80°C and stored until shipment to the laboratory (SANPROBI Research and Development Center in Szczecin). Prior to analysis, samples will be warmed and centrifuged at 1000 x g for 20 minutes to remove particulate material. Cortisol and DHEA in human saliva will be measured with commercially available kits using an immunoenzymatic method according to the manufacturer's instructions. All samples will be processed in duplicate during the same assay. Statistical analyses will be performed using the mean of the two replicate measurements. Raw values will be log-transformed, and observations greater than three SDs from the mean for each sampling time will be excluded.

Faecal sample collection and microbiota and cortisol analysis Faecal sampling was carried out as follows: feces were collected at home on the day of arrival at the laboratory or the day before, cooled immediately after defecation in the test subject's refrigerator and transferred to the laboratory. Samples were stored at -80 ◦C until analysis. Analysis of the gut microbiota will be performed in SANPROBI Research and Development Center laboratory in Szczecin using Illumina next-generation sequencing technology. DNA extraction from the faecal sample will be performed using commercially available isolation kits, followed by an assessment of the quality (integrity) of the isolate. The next step is DNA amplification by PCR using specific primers complementary to conserved gene sequences on either side of the target region. The V2 and V3 hypervariable fragments of the region, consisting of approximately 400-500 base pairs of the prokaryotic 16S rRN gene, are duplicated. The amplicons from each sample are barcoded with a short "barcode" sequence unique to each sample. Finally, high-throughput "multiplexed" sequencing of all samples is performed in a single sequencing run.

Faecal samples for cortisol analysis will be warmed and diluted with water, and the subsequent procedure is the same as for saliva samples prepared for cortisol analysis, as described above.

Personality dimensions The Ten Item Personality Inventory (TIPI) scale is a validated 10-item self-administered questionnaire designed to evaluate the psychosocial factors. The TIPI was designed by Gosling et al. as a brief instrument to measure the Big Five personality dimensions with two items for each trait (a standard item and a reverse-scored item in each trait). These personality dimensions comprise extraversion (being social, enthusiastic), agreeableness (trustworthiness, being empathetic), conscientiousness (self-discipline, reliability, self-efficacy), emotional stability (the opposite dimension to neuroticism; being balanced, calm, capability of remaining stable), and openness (curiosity and creativity, being open-minded). The responses in the personality scale were measured on a seven-point Likert scale and scored as 1= strongly disagree, 2 = disagree moderately, 3 = disagree a little, 4 = neither agree nor disagree, 5 = agree a little, 6 = agree moderately, 7 = agree strongly. Responses to reverse-scored items were recoded to be consistent with standard items. Then, the average of the standard item and the recoded reverse-scored item were calculated to make up each dimension's scale score, which ranged from 1 to 7. Higher scale scores reflect a higher level of each trait.

Each TIPI scale score was categorized by the conceptual approach, which was dividing the scale according to whether the scores were equal to being "agree" or higher (on average) to create two categories: lower TIPI (<5 as disagree) and higher TIPI categories (5-7 as agree). Any dimension scale score that produced agree and higher scores indicated a higher level of that trait and were considered as a higher TIPI category.This questionnaire's internal reliability with a Cronbach's alpha value was 0.59. The TIPI questionnaire was administered only once during the initial T1 visit.

Perceived stress and mental health assessment We collected 5 self-report (sub)scales related to broad domains of perceived stress and physical/mental health: Perceived Stress Scale (PSS-10), Brief Coping Orientations To Problems Experienced Inventory (Brief-COPE), Sense of Coherence 13-item scale (SOC-13), General Health Questionnaire GHQ-28, and Ten Item Personality Inventory (TIPI). All tests (in a standardized Polish version) were based on self-assessment questionnaires posted on the Google Drive file-sharing platform.

The PSS-10 form was used to assess the intensity of stress related to one's own life experiences in the past month and was a modified version of the Perceived Stress Scale developed by Cohen. Respondents reported on a 5-point Likert scale (0 = never, 1 = almost never, 2 = sometimes, 3 = quite often, 4 = very often) the extent to which situations in their lives during the past month were unpredictable, uncontrollable, and overwhelming. The scores for the four positively identified items (items 4, 5, 7, 8) are reversed. The maximum score is 40, with higher scores indicating higher levels of perceived stress. According to the instructions, scores in the range of 0-13 indicate a low level of stress, scores in the range of 14-26 indicate a moderate level of stress, and scores in the range of 27-40 indicate a high level of perceived stress. In this study, the scale had a Cronbach's alpha (α) of 0.85. Scores on the scale were compared between the initial and final assessments.

The Brief COPE is a shortened 28-item version of the COPE Inventory questionnaire by Carver et al. and was used to assess situational coping strategies (ways people cope with a specific stressful event) and dispositional coping strategies (usual ways people cope with stress in everyday life). In its original version, this inventory revealed adequate levels of internal consistency, in addition to showing proof of the validity of its factor structure. We used Mini-COPE, the shortened Polish version of COPE Inventory questionnaire.The scale measures 14 theoretically or empirically grounded coping strategies (coping subscales, each with two items): self-distraction, active coping, denial, substance use, emotional support, instrumental support, behavioral disengagement, venting, positive reframing, planning, humor, acceptance, religion, and self-blame. For each item, respondents indicate whether they use the coping response on a four-point Likert scale (0 = I usually do not do this almost at all; 1 = I usually do this rarely; 2 = I usually do this often; 3 = I usually do this almost always).

These subscales were futher categorized into second-order factor model, which consisted of "Adaptive coping strategies" (use of emotional support, positive reframing, acceptance, religion, humor, active coping, planning, and use of instrumental support) and "Maladaptive coping strategies" (venting, denial, substance use, behavioural disengagement, self-distraction, and self-blame). Cronbach's alpha for the original subscales ranged from 0.66 to 0.70. The overall score of each coping strategy is the sum of the scores of two items.

The Sense of Coherence Scale SOC-13 is a 13-item scale adapted from a longer 29-item scale developed by Antonovsky. This scale is a self-report questionnaire and consists of three subscales: meaningfulness (four items), comprehensibility (five items), and manageability (four items). Each item contains responses that change between questions and are rated from 1 to 7 on a Likert scale. Five variables (items 1, 2, 3, 7, and 10) are negatively worded and are therefore reverse-coded when items are summed to obtain an item total. SOC was calculated as the mean for each individual who had answered at least 10 of the 13 items (range, 1.0-7.0) A higher score represents a greater sense of coherence. Studies of the psychometric properties of the SOC-13 scale have shown high levels of validity and reliability, and internal consistency is very good, with Cronbach's alphas ranging from 0.70 to 0.92 in the 127 studies using the SOC-13. In the current study, reliability estimates for this scale range from α = 0.77-0.79.

The GHQ-28 General Health Questionnaire was developed by Goldberg in 1978. It is a measure of emotional distress over the past few weeks that contains 28 items and could be used as a community screening tool and for the detection of non-specific psychiatric disorders among individuals in primary care setting. Participants were asked about symptoms and/or discomfort they had recently experienced. Each item is rated on a 4-point Likert scale to indicate symptom severity from 0 to 3 (0 = not at all, 1 = not more than usual, 2 = a little more than usual, 3 = much more than usual). Scores for the seven positively identified items (items 4, 5, 7, 8) are reversed (1=more than usual, 2=as usual, 3=less than usual, 4=much less than usual). Items are summed across four subscales: somatic symptoms (items 1-7), anxiety/insomnia (items 8-14), social dysfunction (items 15-21), and major depression (items 22-28). A high GHQ-28 score indicates a probable psychiatric case. Normative data for the general population evaluated indicates a cutoff point ≥ 5. Reliability was calculated using Cronbach alpha coefficient and was 0.89-0.90.

Socioeconomic status and stressful events An additional questionnaire was formulated by the authors, including socioeconomic status (SES) (age, place of residence, family status, level of education) and questions on any stressful events (SEs) experienced during the previous month.

Statistical analysis Continuous variables are presented as mean and standard deviation and variables from scaling as medians and interquartile ranges (25% to 75%). Categorical variables are presented as frequencies. Differences among categorical variables are analyzed by Fisher's exact test. All continuous variables are tested for normal distribution using the Shapiro-Wilk test, as well as kurtosis, skewness, and a graphical evaluation of the distribution of each variable data. Data transformation procedures (e.g., Box-Cox transformation) are considered if they result in a normal distribution of all transformed variables. Comparisons between baseline and post-supplementation time points (T1 and T2 visits) in the Lh+Bl or PLA groups are determined using the T-test for dependent variables and effect size expressed as Cohen's d. Comparisons between groups in pre-post changes within variables are determined using the T-test for independent variables. If normality is violated, data are analyzed using the Wilcoxon signed-rank test with effect size expressed as the rank correlation coefficient (rc). Mann-Whitney U test is used to compare the variables to test for differences between the placebo and prebiotic groups. An alpha of <0.05 is considered statistically significant. Data are analyzed using STATISTICA 13.3 software (StatSoft Inc., USA).

Inclusion Criteria:

• female sex,
• age 20-30 years,
• BMI < 30 kg/m2

Exclusion Criteria:

• diagnosed chronic and/or serious health disorders (e.g. diabetes, asthma, hypertension, renal diseases),
• pregnancy,
• taking any probiotics for 8 weeks prior to enrollment to the study protocol, -
• recent and current antibiotic treatment, ore use of pharmaceutical antidepressant treatment

Only biological women are eligibilitible.

Neither the researchers nor the participants knew whether intervention or placebo was administered. The preparations were administrated to each participant in envelopes marked with a unique code. The supplements were in the same capsules as the plecebo and could not be distinguished by appearance and taste. Assignment of study material (A - placebo or B - intervention) to individuals was randomised by computer-generated codes, using Research Randomizer (www.randomizer.org), obtained from the supplement manufacturer. The study was decoded after results collection was completed.