The gut microbiota consists of a community of microorganisms including bacteria, viruses, protozoa, and fungi in the gastrointestinal tract, and is defined as the totality of these microbes and their genomic components [1
]. A substantial body of work indicates that the human gut microbiome is implicated in mental health via bidirectional interactions within the brain–gut–microbiome (BGM) axis, which involves the immune, neural, endocrine, and metabolic pathways between the gut and the brain [2
]. The perturbation of this axis leads to altered stress response and behaviors, and has been proposed to be involved in several psychiatric disorders, including depression [5
], anxiety [6
], and insomnia [7
The composition and activity of the gut microbiota can be modulated by dietary intake [8
]. Diet modification rapidly alters the gut microbial community [9
], and a complex and varied diet is associated with a more diversified gut microbiome [10
]. From a clinical perspective, diet-induced improvements in the diversity and function of the gut microbiome have the potential to improve mental health by affecting the BGM axis. Thus, the oral intake of probiotics, a preparation of live beneficial microorganisms [11
], may have therapeutic effects on psychiatric symptoms by promoting a healthy and balanced gut microbiome.
Probiotics contain a variety of microbes and may exert varying health effects depending on their microbial composition. In recent years, accumulating literature suggests that probiotics composed of specific strains of genera Lactobacillus
may have the potential to prevent and treat various psychiatric conditions such as depression and anxiety [12
]. However, to the best of our knowledge, only a few randomized controlled clinical trials have investigated the effects of probiotics containing Lactobacillus
strains on mental health in healthy individuals, and these studies have produced conflicting results [13
]. Furthermore, these studies had small sample sizes and did not conduct gut microbiome analyses to confirm the restoration of homeostasis of the gut microbiome, one of the most reliable markers of probiotic effects [16
The probiotic NVP-1704 is a mixture of Lactobacillus reuteri
NK33 and Bifidobacterium adolescentis
NK98 isolated from the healthy human gut microbiota [17
], and several preclinical trials have showed that NVP-1704 can alleviate anxiety and depression in mice by modulating gut immune responses and gut microbiota composition [17
]. Translational research to confirm these psychotropic effects of NVP-1704 may be the first step to providing a novel therapeutic approach to alleviate psychiatric symptoms.
The present study aimed to examine the efficacy and safety of NVP-1704 administration for the management of stress-related symptoms, such as depression, anxiety, and insomnia, in healthy adults. To overcome the methodological shortcomings of previous studies, we conducted a randomized, double-blind, placebo-controlled parallel study with a relatively large sample size, and included a gut microbiome analysis to confirm the effects of our intervention.
2. Materials and Methods
2.1. Study Population
The present study was conducted at the sleep clinic of Seoul National University Bundang Hospital. Healthy adults aged 19 to 65 years with psychological stress and subclinical symptoms of depression or anxiety were recruited by advertising in the hospital and the local community, from October 2018 to August 2019. Enrolled participants underwent screening tests at the first visit (within two weeks prior to initiation of the intervention), and we included those who met both of the following criteria: (1) Stress Response Inventory (SRI) score ≥ 50 and ≤ 100; and (2) either Beck Depression Inventory-II (BDI-II) score ≥ 20 and ≤ 45 or Beck Anxiety Inventory (BAI) score ≥ 16 and ≤ 45. The Korean version of SRI [19
], BDI-II [20
], and BAI [21
] were developed and validated in the Korean population. The exclusion criteria were as follows: (1) psychiatric disorders including mood disorders, anxiety disorders, and psychotic disorders; (2) withdrawal syndrome from alcohol or smoking; (3) history of treatment for stress, depression, and anxiety within four weeks prior to study; (4) any severe stressful life events within four weeks prior to study; (5) use of oral steroids, sleeping pills, anorectic agents, beta-blockers, antibiotics, drugs related to colon diseases, any type of pre- or probiotics including yogurt and other functional foods for stress or insomnia within two weeks; and (6) medical illness such as uncontrolled hypertension, uncontrolled diabetes mellitus, thyroid diseases, and impaired renal or hepatic function. All participants agreed to participate in the study and signed informed consent before the initiation of the study. This study was approved in August 2018 by the Institutional Review Board of Seoul National University Bundang Hospital (B-1807/483–005) and was registered retrospectively in the Clinical Research Information Service, Korea (registration number KCT0004801).
In the present study, the probiotic NVP-1704 was chosen as the active agent. NVP-1704 is composed of Lactobacillus reuteri NK33 and Bifidobacterium adolescentis NK98 freeze-dried with maltodextrin. Each 500 mg capsule of NVP-1704 contained 2.5 × 109 colony-forming units of microorganisms (2.0 × 109 CFU for Lactobacillus reuteri NK33 and 0.5 × 109 CFU for Bifidobacterium adolescentis NK98). The placebo capsule, made of maltodextrin, was created to look identical to the NVP-1704 capsule. All study products were prepared at a Good Manufacturing Practice-certified manufacturing facility and stored at 2–8 °C. The participants were instructed to keep the products refrigerated.
At the first visit, demographic, medical, and anthropometric data were gathered, and electrocardiogram, blood cell count, biochemistry profile, and urinalysis were conducted for each participant. At the second visit, two weeks after the first visit, eligible participants were randomly assigned in a 1:1 ratio to either the experimental group receiving the probiotic NVP-1704, or the control group receiving a placebo. The random assignment was performed via the block randomization method, using SAS version 9.4, and both participants and investigators were blinded to the treatment conditions. Participants received either the probiotic NVP-1704 or a placebo depending on their randomized group assignment, and intervention began immediately. They were instructed to take two capsules with water once a day, daily, for eight weeks. The third and final visits took place at four and eight weeks after initiation of intervention, respectively. Treatment compliance was assessed twice over the study period by counting the remaining capsules at the last two visits. Figure 1
describes the timeline of the study protocol.
All participants were asked to complete self-report questionnaires regarding symptoms of stress, depression, and anxiety at the first, third, and final visits. The SRI was adopted to assess emotional, somatic, cognitive, and behavioral stress responses. The SRI consists of 39 items with total scores ranging from 0 to 156, with higher scores indicating more severe stress symptoms [19
]. Symptoms of depression and anxiety were evaluated using the BDI-II [22
] and BAI [23
], respectively. As an additional analysis, overall depression and anxiety symptoms were evaluated by the sum of the BDI-II and BAI scores, because depression and anxiety tend to co-occur and often overlap in phenomenology and pathophysiology [24
Subjective sleep quality and symptoms of insomnia were assessed using the Pittsburgh Sleep Quality Index (PSQI) and Insomnia Severity Index (ISI), respectively. The Korean version of PSQI [25
] and ISI [26
] were developed and validated in the Korean population. The PSQI consists of seven subscales that assess various aspects of sleep quality; we performed sub-analyses for each subscale score of the PSQI to identify the specific effects of our intervention. We also compared the rate of clinically meaningful improvements in sleep quality and insomnia symptoms between the two groups for additional analysis. Clinically meaningful improvements were defined as a score reduction of more than or equal to the minimal clinically important difference (MCID), which refers to a threshold value for a clinically meaningful and worthwhile change [27
]. Based on several previous studies, we determined the MCID for the PSQI [28
] and ISI scales [30
] as 3 and 6 points, respectively.
2.4. Blood Biomarkers
The pre- and post-intervention blood tests for biomarkers were conducted at a designated time in the morning (from 9 AM to 11 AM) of the second and final visits, respectively, as initial and follow-up assessments for each participant. To assess the inflammatory response and hypothalamic–pituitary–adrenal (HPA) axis activity, we measured the serum levels of interleukin 6 (IL-6), tumor necrosis factor alpha (TNF-alpha), adrenocorticotropic hormone (ACTH), and cortisol. Simultaneously, the blood level of brain-derived neurotrophic factor (BDNF) was evaluated as a neuroplasticity index. Blood samples were drawn from the antecubital vein, processed per protocol, and transported to GCCL Co., Ltd. (Gyeonggi-Do, Yongin-si, Korea), a global clinical testing laboratory. The ratios of the concentrations of IL-6 to log (BDNF) and TNF-alpha to log (BDNF) were also determined.
2.5. Gut Microbiome Analysis
The participants were asked to submit fecal samples at the final visit. Bacterial genomic DNA was extracted from fresh feces using a QIAamp DNA Stool Mini Kit (Qiagen, Hilden, Germany), according to a protocol by Kim et al. [31
]. Genomic DNA was amplified using barcoded primers targeting the bacterial 16S rRNA V4 region gene. Each amplicon was sequenced using an Illumina iSeq 100 (Illumina, San Diego, CA, USA). Functional genes were predicted using the phylogenetic investigation of communities by the reconstruction of unobserved states (PICRUSt) [31
]. Linear discriminant analysis (LDA) and cladograms were captured using the LDA effect size on the Galaxy platform [33
]. Pyrosequencing reads were deposited in the Short Read Archive of the National Center for Biotechnology Information under accession number PRJNA 678145.
2.6. Safety Assessment
To evaluate the safety of probiotic administration, we monitored for the occurrence of any adverse event by participants’ self-reports at four and eight weeks after the start of the intervention. Follow-up blood cell count, biochemistry profile, urinalysis, and an electrocardiogram were also performed at the final assessment to screen for any medical problems that may have occurred during the intervention. All adverse reactions were coded to preferred terms, as specified in the Medical Dictionary for Regulatory Activities (MedDRA) version 21.0.
2.7. Nutritional Data
Nutrition plays a central role in the modulation of gut microbiota composition [8
]. Therefore, we examined participants’ habitual dietary patterns during the intervention period. Based on 24 h dietary recalls and food diaries reported at the second, third, and final visits, the daily intakes of calories, carbohydrates, proteins, fat, water, dietary fiber, vitamins, and mineral nutrients were calculated. Nutritional data were analyzed using the Computer Aided Nutritional Analysis Program (CAN-Pro version 5.0, http://canpro5.kns.or.kr/
, accessed on 24 December 2019) developed by the Korean Nu-trition Society.
2.8. Statistical Analysis
Sample size was determined based on the randomized controlled trial of Ali Talaei and colleagues, who adopted the BDI-II and BAI scores as primary outcome measures [32
]. The minimal required sample size was calculated as 104 subjects (52 for each group) using the two-sample, continuous outcome, two-tailed, independent t
-test approach, with a significance level of 5% and power of 80%. Considering a drop-out rate of 30%, we planned to enroll more than 150 subjects (75 per group). We performed per-protocol (PP) analysis, excluding the data of individuals who dropped out as well as those with incomplete measurements of outcomes.
Paired t-tests were used to evaluate intra-group differences in pre- and post-intervention assessments, including questionnaires and blood tests. Independent t-tests were used to evaluate inter-group differences in the changes in parameters after treatment. Effect size was calculated using Cohen’s d to estimate the magnitude of clinical improvement in each group. As for additional analysis, analysis of covariance (ANCOVA) adjusted for age, sex, baseline BDI-II score, and baseline IL-6 level was performed to compare the improvement in each subscale score of PSQI between the two groups. All statistical analyses were carried out using SAS version 9.4 (SAS Institute, Cary, NC, USA), and a two-tailed p-value of less than 0.05 was considered statistically significant.
The present study revealed that NVP-1704 is a safe and well-tolerated probiotic with beneficial effects on depression and sleep in healthy adults. Our study also revealed a significant reduction in serum pro-inflammatory cytokine IL-6 levels after NVP-1704 treatment. In line with these findings, our microbiome analysis demonstrated that the individuals treated with NVP-1704 had a gut microbiota composition with reduced ratios of Enterobacteriaceae to Bifidobacteriaceae and Enterobacteriaceae to Lactobacillaceae, which could be associated with better mental health. Our evaluation of safety revealed that no serious adverse reactions occurred during the study period, and the risk of adverse events with NVP-1704 treatment was low (<3%), yielding no significant difference compared to the placebo.
Our findings are similar to the results reported in a previous mouse study, which concluded that treatment with NVP-1704 led to the alleviation of depression/anxiety, changes in serum IL-6 levels, and an altered gut microbiome [17
]. Therefore, the beneficial effects of NVP-1704 on depression/anxiety and the potential physiological processes underlying the psychotropic effects of NVP-1704 treatment have been implicated in both preclinical and clinical studies. We also found a significant improvement in both subjective sleep quality and insomnia symptoms after treatment with NVP-1704. To the best of our knowledge, our work is the first study to report a significant decrease in insomnia symptoms, measured using both the PSQI and ISI scales, which is attributable to probiotic treatment. A recent review of randomized controlled trials regarding the effects of probiotics on sleep reported that only a few studies revealed meaningful improvements in the PSQI score, whereas no positive changes were observed on other subjective sleep scales [34
]. Although we found these clinical improvements, a non-negligible placebo effect was also observed in some of the mood and sleep parameters. The placebo effect has been frequently observed in clinical trials, especially when the evaluating parameters are subjective [35
]. This study was conducted on a subclinical sample; therefore, the placebo effect could have been more pronounced.
The present study found a significant reduction in the serum pro-inflammatory cytokine IL-6 levels. The human gut microbiome is closely linked to the production of pro-inflammatory cytokines, including IL-6 [36
]. Dysbiosis of the gut microbiome, such as a decrease in commensal Bifidobacterium
strains and an increase in pathogenic gut microbes, can stimulate the secretion of pro-inflammatory cytokines via increased permeability of the intestinal epithelium [38
]. In contrast, the administration of Bifidobacterium
] and Lactobacillus
] can downregulate pro-inflammatory cytokine secretion. Consistent with these results from previous studies, our findings contribute to the evidence supporting the anti-inflammatory effects of the probiotic NVP-1704.
We hypothesized that NVP-1704 treatment may increase serum BDNF levels; previous randomized clinical trials have shown that the administration of specific probiotic strains of Bifidobacterium
increased serum BDNF levels [41
]. In contrast to our expectation, however, we could not observe a significant difference in the change in serum BDNF levels between the two study groups. Instead, a significant decrease in the ratio of IL-6 to log (BDNF) was found after NVP-1704 treatment. A previous study suggested that an increased BDNF to IL-6 ratio can act as a surrogate maker of recovery and neuroplasticity [44
]. However, because no significant difference was observed in serum BDNF levels, the changes in the ratio of IL-6 to log (BDNF) might be principally due to the impact of IL-6, not that of BDNF.
As we expected, NVP-1704 treatment induced an increase in Bifidobacterium
strains and a decrease in the Proteobacteria
population at the phylum level. Previous studies have shown that dysbiosis of the gut microbiome might be associated with psychiatric disorders, suggesting that the gut microbiome could be a useful therapeutic and preventive target. Jiang et al. [45
] reported that the gut microbiota composition of patients with depressive disorders showed higher levels of Proteobacteria
compared to healthy controls. An altered gut microbiota profile was also reported in patients with generalized anxiety disorder [46
], and gut microbiome diversity appears to be related to sleep physiology [7
]. In line with previous studies, our clinical data support the notion that the oral administration of probiotic NVP-1704 can restore a balanced gut microbiota composition, thereby mitigating psychiatric symptoms.
Although the exact mechanism needs to be further elucidated, the modulation of neuroinflammatory pathways due to beneficial modification of the gut microbiome may be the key components underlying the beneficial effects of NVP-1704. Neuroinflammation induced by various medical conditions plays an important role in the pathophysiology of depression [47
], anxiety [48
], and insomnia [49
]. Therefore, the suppression of IL-6 might be responsible for the psychotropic effects of NVP-1704 treatment.
A plethora of research has indicated that psychological stress, often represented by HPA hyperactivity, is closely related to the development of various mental health disorders, including depression [50
], anxiety [51
], and insomnia [52
]. Hence, we hypothesized that the beneficial effects of NVP-1704 on mood and sleep may be mediated by HPA axis modulation. However, we did not detect any significant differences in either the serum ACTH or cortisol concentrations between the two groups after intervention. Although many animal studies have demonstrated that probiotics containing specific strains of Bifidobacterium
modulate the HPA axis [53
], clinical studies examining the effects of probiotics on HPA axis activity have revealed inconsistent findings. One study showed a correlation [14
], whereas others did not [13
]. This discrepancy may, in part, stem from methodological heterogeneity, such as the difference in the timing of blood tests (morning versus other times of the day) and differences intrinsic to the population being studied (healthy individuals versus patients with psychiatric disorders). Further methodologically robust studies are required to examine the effects of probiotics on HPA axis activity.
Our study had some limitations. First, we performed PP analysis, which may result in an overestimation of treatment efficacy and underestimation of adverse reactions, compared to intention-to-treat analysis. However, our data reported very few adverse events and good adherence (>90%), which may reduce the possibility of selection bias. Secondly, we did not analyze the gut microbiota composition prior to treatment. However, no significant differences were observed in the dietary patterns of the two study groups, suggesting that post-treatment analysis may predominantly reflect the effects of the intervention. Thirdly, we did not apply appropriate statistical corrections such as the Benjamini–Hochberg correction. We performed multiple hypothesis testing; therefore, not doing this correction may have caused false-positive findings. Additionally, we registered our study protocol to the clinical trials registry retrospectively, because Clinical Research Information Service, a Korean clinical trials registry platform, allows the retrospective registration of clinical trials. However, we should have registered the study protocol in advance, to meet international standards. In addition, a relatively large dropout rate (over 20%) may have affected the results and the interpretation of our findings. Finally, we cannot generalize our findings to the elderly aged over 65 years or to the clinical population.