Plasma short-chain fatty acid concentrations in social anxiety disorder and changes after cognitive behavioral therapy.

Plasma short-chain fatty acid concentrations in social anxiety disorder and changes after cognitive behavioral therapy - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Transl Psychiatry . 2026 Jun 3;16:295. doi: 10.1038/s41398-026-04134-y Plasma short-chain fatty acid concentrations in social anxiety disorder and changes after cognitive behavioral therapy Wenjie Cai 1, 2 , Miranda Stiernborg 1, 2 , Alexander Wolthon 3, 4 , Rikard Landberg 5 , Amirhossein Manzouri 3, 6, 7, 8 , Tomas Furmark 9 , Catharina Lavebratt 1, 2 , Kristoffer N T Månsson 6, 7, 8, ✉ Author information Article notes Copyright and License information 1 Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden 2 Karolinska University Hospital Solna, Center for Molecular Medicine, Stockholm, Sweden 3 Department of Psychology, Stockholm University, Stockholm, Sweden 4 Department of Comparative Medicine, Karolinska Institutet, Stockholm, Sweden 5 Department of Life Sciences, Division of Food and Nutrition Science, Chalmers University of Technology, Gothenburg, Sweden 6 Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden 7 Centre for Psychiatry Research, Department of Clinical Neuroscience, Karolinska Institutet, & Stockholm Health Care Services, Region Stockholm, Stockholm, Sweden 8 Department of Clinical Psychology and Psychotherapy, Babeș-Bolyai University, Cluj-Napoca, Romania 9 Department of Psychology, Uppsala University, Uppsala, Sweden ✉ Corresponding author. Received 2025 Nov 23; Revised 2026 Apr 16; Accepted 2026 May 19; Collection date 2026. © The Author(s) 2026 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . PMC Copyright notice PMCID: PMC13233848  PMID: 42236485 See the article " Pre-analytical considerations in the interpretation of plasma SCFA profiles ", 468. Abstract Emerging evidence indicates that the gut microbiota are linked to variation in social behavior and anxiety, with short-chain fatty acids (SCFAs) proposed as key microbial metabolites that may mediate microbiota-brain communication. Cognitive behavioral therapy (CBT) is an effective treatment for social anxiety disorder (SAD), but it remains unclear whether gut microbiota changes following treatment. In this longitudinal study, 46 individuals with SAD underwent nine weeks of internet-delivered CBT. Plasma concentrations of nine SCFAs were measured using liquid chromatography-mass spectrometry: twice at pre-treatment, once at post-treatment and once at a 36-month follow-up in patients, with > 80% data completeness. Healthy controls ( n = 42) underwent two SCFA assessments 11 weeks apart. Generalized additive mixed models were used to assess longitudinal changes with CBT, linear mixed models for patient-control differences, and intraclass correlation coefficients for test-retest reliability. CBT led to significant reductions in social anxiety symptoms, and effects were sustained at 36 months follow-up. Plasma concentrations of butyric, isobutyric, propionic, and valeric acids increased after CBT, with trajectories predominantly explained by time rather than individual variability. Before CBT, isobutyric acid was statistically significantly lower in SAD patients, relative to healthy controls. These findings provide initial evidence that sustained therapeutic benefits of CBT may be linked to alterations in circulating microbial metabolites, as indexed by lower isobutyric acid in SAD patients at pre-treatment, and increases in isobutyric, butyric, propionic and valeric acids after therapy. Lifestyle factors were not assessed but may have contributed to the observed long-term metabolic changes. Subject terms: Psychology, Psychiatric disorders Introduction Social anxiety disorder (SAD) is a debilitating psychiatric disorder characterized by an excessive fear of social situations and negative evaluation by others, with a lifetime prevalence exceeding 12% [ 1 ]. SAD severely impairs individuals’ social functioning, manifesting as decreased work productivity, increased school dropout, elevated unemployment rates, and impaired social and romantic relationships [ 2 , 3 ]. The disorder is also associated with substance abuse, severe depression, and an increased risk of suicide [ 3 , 4 ]. Twin studies have shown that genetic factors play an important role [ 5 ] as well as interactions with environmental factors, including parenting styles and adverse life events [ 6 ]. Animal studies have consistently shown that an altered microbiota-gut-brain axis state is associated with social behavior. Recent reports suggest that patients with SAD exhibit an altered microbial composition compared with healthy controls [ 7 ] and that germ-free mice receiving faecal transplants from SAD patients exhibit high sensitivity to social fear [ 8 ]. Thus, an altered microbiota-gut-brain axis state may contribute to the pathophysiology of SAD. Short-chain fatty acids (SCFAs) are produced by the bacterial fermentation of dietary fibre in the gut and are considered essential mediators of microbiota-gut-brain communication. SCFAs play vital roles not only in maintaining gastrointestinal health but also in systemic metabolism [ 9 ], immune regulation [ 10 , 11 ], and critical neurological processes, such as synaptic plasticity, neural differentiation, and neurotransmitter regulation [ 12 , 13 ]. Emerging evidence links altered SCFA profiles with various mental disorders, including ADHD [ 14 ], autism [ 15 ] and depression [ 16 ]. A recent experimental study shows that acute stress can rapidly reduce SCFAs availability and disrupt gut-brain barrier function [ 17 ], highlighting SCFAs as dynamic, stress-sensitive microbial metabolites with direct relevance for psychopathology. Understanding how these SCFA patterns influence treatment outcomes is therefore a key research topic. Cognitive behavioral therapy (CBT) has emerged as a partly effective treatment for SAD, with internet-delivered guided self-help programs being an accessible and efficacious alternative [ 18 , 19 ]. These programs involve structured modules with homework assignments and therapist guidance, facilitating treatment adherence and consistency across participants [ 20 , 21 ]. The therapeutic gains are well maintained over the long term [ 22 ]. Despite the effectiveness of CBT, many patients’ outcomes remain unsatisfactory. One approach to better understand how effective treatment works is to investigate the biological mechanisms underlying the therapeutic response. To this end, we have previously demonstrated that CBT is associated with structural and functional neural plasticity, including reduced amygdala volume [ 23 ] and that pre-treatment neural variability reliably predicts CBT outcome [ 24 , 25 ]. Further, we have previously reported that improvements in cellular protective markers (i.e., glutathione peroxidase and telomerase activity) increased after CBT [ 26 ], and that circulating cell-free mitochondrial DNA in peripheral blood was reduced in SAD patients relative to matched controls [ 27 ]. To our knowledge, there are currently no data on plasma concentrations of SCFA in patients with SAD, and no longitudinal studies investigating changes in plasma SCFA profiles after effective treatment. Our study aimed to (1) examine the short- and long-term changes in SCFAs concentrations and treatment outcomes following internet-delivered CBT in SAD patients, and (2) to compare pre-treatment SCFA concentrations in SAD patients relative to matched healthy controls. Materials and methods Participant and study design The recruitment procedures have been described in detail, and some clinical outcomes have been reported elsewhere [ 24 , 26 , 28 ]. In brief, participants aged over 18 were recruited through media advertising as previously described. The Mini-International Neuropsychiatric Interview (M.I.N.I; v7) and the Structured Clinical Interview for DSM-IV-Axis I Disorders (SCID-I) were used for the diagnosis of SAD [ 29 , 30 ]. Somatic and neurological conditions were screened via self-report at baseline, and structural magnetic resonance imaging scans were reviewed by a neuroradiologist to exclude clinically relevant abnormalities. Individuals with neurological diseases, bipolar disorder, psychotic disorders, severe depression, alcohol or substance abuse disorders, or antisocial personalities were excluded. Ultimately, 46 patients diagnosed with SAD were enrolled in the study (Table 1 ). None of the participants were undergoing psychotherapy concurrently during the study’s active treatment period, and four participants (9%) were on psychoactive drug treatment and agreed to maintain a stable medication dosage for at least three months before and during CBT. In addition to the SAD patients, a matched healthy control (HC) group was also recruited (Table 1 ), comprising individuals with no life-time history or current diagnosis of psychiatric disorders, as assessed by M.I.N.I. (modified to assess life-time history). There were no significant differences between SAD patients and HC participants regarding sex, age, BMI, or smoking status (all P’s > 0.20). Screening and enrollment of SAD participants and healthy controls were conducted on separate occasions. For clarity, Supplementary Fig. S1 - 2 presents the timing of all assessments in relative weeks. The Insomnia Severity Index (ISI) [ 31 ] is a self-administered questionnaire used to assess insomnia symptoms. Table 1. Demographic characteristics. SAD patients ( n = 46) Healthy controls ( n = 42) Sex, n (%) Females 29 (63.0) 24.0 (57.1) Male 17 (37.0) 18.0 (42.9) Age, mean (SD) 30.7 (8.3) 31.9 (9.5) BMI, mean (SD) 24.8 (4.4) 23.8 (3.2) SAD, duration in years, mean (SD) 17.0 (9.9) n/a Smoking regularly the past 3 months, n (%) 4 (8.7) 8 (19.0) Marital status, n (%) Married/cohabiting with children 16 (34.8) 15 (35.7) Married/cohabiting without children 10 (21.7) 12 (28.6) Non-cohabiting partner 4 (8.7) 3 (7.1) Single with children 4 (8.7) 2 (4.8) Single without children 9 (19.6) 9 (21.4) Other 3 (6.5) 1 (2.4) Education, n (%) Completed primary school 3 (6.5) 0 (0.0) Completed secondary school 7 (15.2) 5 (11.9) Completed vocational education 2 (4.3) 2 (4.8) Ongoing university education 16 (34.8) 13 (31.0) Completed university education 18 (39.1) 22 (52.4) Concurrent medications, n (%) No medication 18 (39.1) n/a SSRIs 4 (8.7) n/a Hormonal contraceptives 16 (34.8) 11 (26.2) Antihistamines 1 (2.2) n/a Hormone medications 1 (2.2) n/a Thyroid hormone substitution 2 (4.4) n/a Concurrent comorbidity, n (%) No concurrent psychiatric comorbidity 12 (26.1) n/a Major depressive disorder 3 (6.5) n/a Panic disorder 3 (6.5) n/a Agoraphobia 5 (10.9) n/a Generalized anxiety disorder 3 (6.5) n/a Binge eating disorder 1 (2.2) n/a Obsessive compulsive disorder 2 (4.3) n/a Open in a new tab As shown in Fig. 1 , all participants completed two baseline assessments, and SAD patients underwent four additional follow-ups after CBT (post-treatment, and at ~13, ~18, and ~36 months). Exact numbers of biological and self-report assessments across time-points are shown in Fig. 1 . Fig. 1. Study design and timeline of assessments for patients with social anxiety disorder and matched healthy controls. Open in a new tab All participants underwent two baseline assessments, including clinical self-reports and blood draws. Patients completed four additional assessments after receiving cognitive behavioural therapy: (1) follow-up assessment at post-treatment (~6 months after recruitment, including blood draw); (2) assessment at ~13 months (self-reports only); (3) assessment at ~18 months (self-reports only), and (4) final assessment at ~36 months including both self-reports and blood draw. Created with BioRender.com. Internet-delivered cognitive behavioral therapy for social anxiety disorder Internet-delivered CBT is a self-guided psychological treatment intervention based on principles of traditional CBT [ 19 ]. Patients received one module per week over nine weeks, with each module including text materials and homework assignments. A clinical psychologist interacted weekly with participants through messages to provide guidance and feedback, see also [ 32 ] for details. Clinical assessments The Liebowitz Social Anxiety Scale Self-Report version (LSAS-SR) was used as the primary outcome measure. LSAS-SR is a widely used and validated psychological instrument specifically designed for assessing the severity of SAD [ 33 – 35 ]. In the Supplementary Table S1 - 3 , data on the Montgomery-Åsberg Depression Rating Scale – Self-rated version (MADRS-S) scores are reported. Analysis of plasma short-chain fatty acids Participants were instructed to begin fasting at 10:00 p.m. the evening before blood collection, and all blood samples were collected by an experienced research nurse the following morning between 7:00 and 11:30 a.m. Plasma was isolated from lithium-heparin tubes immediately after sampling and stored at −80 °C. Without prior freeze-thawing, the plasma was thawed, and plasma concentrations of nine SCFAs (i.e., formic, acetic, propionic, butyric, isobutyric, succinic, valeric, isovaleric, and caproic acids) were analyzed using liquid chromatography-mass spectrometry (LC-MS) [ 36 ] at the Department of Life Sciences, Chalmers University of Technology, Gothenburg, October 2021. Per analysis batch, 40 samples from SAD and controls were analyzed in singlets, together with 3 quality control samples (QC) in triplicates. All samples from a given individual were processed in the same batch, ensuring even distribution of timepoints within batches. In total, 6 batches were run. The QC samples had SCFAs concentrations within the range of the human samples and were used to calculate the within-batch coefficient of variation (CV) of 7.4%, 3.4%, 9.5%, 8.7%, 8.6%, 11.4%, 17.3%, 7.4%, and 8.2% for formic, acetic, propionic, butyric, isobutyric, succinic, valeric, isovaleric, and caproic acid, respectively. Between-batch variation was controlled by normalization of the individual data across batches, where each value was divided by the normalization factor (mean of the QC values of the individual batch)/(mean of the QC values from all batches). SCFA concentrations have not been reported in previous publications on this cohort. Statistical analysis All analyses were conducted in R (v4.5.0; R Core Team, 2022) via RStudio (v.2025.5.0.496, Posit, Boston, MA, USA), and all statistical tests were two-tailed with an alpha level set at 0.05. The primary analysis was a longitudinal within-patient analysis across repeated assessments. For reference, 46 patients provide 91% power to detect a moderate pre–post effect (Cohen’s d = 0.5) at α = 0.05. To examine short- and long-term changes in clinical outcomes and SCFA concentrations, generalized additive mixed models (GAMMs) were used. Non-linear trajectories over time were represented by penalized regression splines, and individual variability was accounted for via random effects. For clinical data, which included repeated measurements at seven time points, both random intercepts and random slopes for time were included. For SCFA outcomes, which were sampled at four time points per patient, random intercepts were used exclusively, as models with random slopes frequently resulted in singular fits or unstable variance estimates. Models were estimated using the mgcv package [ 37 ], v.1.9.1), and summaries included effective degrees of freedom (edf), F-statistics, P values, and deviance explained. Marginal and conditional R² values were estimated according to the Nakagawa method [ 38 ], using the performance package [ 39 ], v.0.13.0) based on refitted models using the gamm4 package ( https://CRAN.R-project.org/package=gamm4 , v.0.2.7). Paired Cohen’s d effect sizes were computed to quantify the magnitude of change in LSAS-SR scores. Specifically, effect sizes were calculated for paired comparisons between the second baseline and both the first and the last post-treatment follow-ups. The statistical analysis of SCFA concentrations was conducted using normalized values. SCFA values were missing for four HCs at the first baseline, and for one HC at the second baseline. Three participants, with samples identified as extreme outliers exceeding 8 times the interquartile range (IQR), were removed from all analyses (see Supplementary Fig. S3 for details). Group comparisons of LSAS-SR scores and SCFA concentrations between the SAD patients and HC participants were examined between the two baseline time points, using linear mixed-effects models (LMM) implemented in the lme4 package ([ 40 ], v.1.1.37), estimating fixed effects of group, time, and their interaction. Random intercepts were included for each participant to account for within-subject dependencies. The models used sum-coded contrasts to reflect the structure of a 2 × 2 repeated-measures ANOVA, with main effects representing marginal means across levels of the other factor. Corresponding Cohen’s d effect sizes were computed for group and time comparisons. P values for SCFA outcomes in the longitudinal (GAMM) and baseline group comparison (LMM) analyses were adjusted for multiple comparisons using the Benjamini-Hochberg method to control the false discovery rate. Adjusted P values ( P FDR ) below 0.05 were considered statistically significant. To assess the test-retest reliability between the two baseline assessments in both SAD patients and HC participants, single-measurement, absolute-agreement, two-way mixed-effects intraclass correlation coefficients (ICC) were computed by using the psych package ([ 41 ], v2.5.6), and interpreted based on the guidelines by McGraw & Wong [ 42 ]. Exploratory analyses included within-group paired comparisons of ISI across time (pre- to post-treatment; and post-treatment to 36-month follow-up) using paired-samples t-tests, with effect sizes quantified using Cohen’s d . Associations between SCFA concentrations and LSAS-SR were examined using Pearson correlations at each baseline time point. Further, we also investigated whether pre-treatment SCFA concentrations were predictive of subsequent changes in LSAS-SR (pre- to post-treatment; and post-treatment to the 36-month follow-up). In addition, Pearson correlation tests were used to examine associations between changes in SCFA concentrations and changes in ISI scores over the corresponding intervals (baseline to post-treatment; post-treatment to the 36-month follow-up). Assumptions for Pearson-based inference were evaluated using residual diagnostics from corresponding linear models (outcome ~ predictor), including Shapiro-Wilk tests of model residuals and inspection of QQ-plots. Results Social anxiety severity changes after cognitive behavioral therapy LSAS-SR scores decreased non-linearly over time following CBT, with the largest reduction during the treatment period, followed by smaller continued reductions over the subsequent months (Fig. 2 ). The fixed effect of time was significant and explained most of the variance ( P < 0.001, R 2 = 0.61, relative to the random effect P < 0.001, R 2 = 0.23). The paired Cohen’s d representing the change from the second baseline to the post-treatment was 1.22 (95% CI: 0.81–1.64), and 1.51 (95% CI: 0.97–2.05) for the change from the second baseline to the 36-month follow-up. Fig. 2. Trajectories of social anxiety symptoms over time estimated using generalized additive mixed models. Open in a new tab Total scores on the Liebowitz Social Anxiety Scale Self-Report version (LSAS-SR), for individuals with social anxiety disorder (SAD), and healthy controls are presented. Four SAD patients with concurrent psychotropic medication are marked in orange color. Each dot represents an individual measurement, and lines connect repeated measures for each participant. The grey shaded area marks the 9-week internet-delivered cognitive behavioral therapy (CBT) treatment period. Error bars represent 95% CI. BL1 1st baseline assessment, BL2 2nd baseline assessment, FU1 1st follow-up assessment, FU2 2nd follow-up assessment, FU3 3rd follow-up assessment, FU4 4th follow-up assessment. Short-chain fatty acid changes during and after cognitive behavioral therapy As presented in Fig. 3 and Table S1 , butyric, isobutyric, propionic, and valeric acid increased non-linearly over time, with limited change between baseline and post-treatment, and more pronounced increases between post-treatment and the 36-month follow-up. Most of the variance was explained by time ( P FDR < 0.001, R² = 0.60–0.89), rather than individual variability ( P FDR = 0.001–0.47, R 2 = 0.001–0.22). Isovaleric acid changed non-linearly over time as well, but with a different trajectory, peaking at post-treatment and declining below baseline levels at the 36-month follow-up (with individual variability accounting for most variance, P FDR < 0.001, R 2 = 0.43, instead of time ( P FDR = 0.003, R² < 0.001). Fig. 3. Short-chain fatty acids change in patients with social anxiety disorder (SAD) across the weeks. Open in a new tab The concentration of each SCFA (μM) in SAD patients over four time points: two baseline assessments (1st baseline assessment at 2 weeks and 2nd baseline assessment at 3 months after recruitment), first follow-up assessments at post-treatment (~6 months) and the final assessment at 39 months after recruitment. Orange lines represent individual SAD patients with concurrent psychotropic medication ( n = 4). Outliers are excluded, and solid black error bars represent 95% CI. FDR corrected P ( estimated using generalized additive mixed models), based on the effect of time are displayed. Caproic, formic, and succinic increased more linearly over time. Of these, formic acid showed the greatest change and was largely explained by time ( P FDR < 0.001, R 2 = 0.54), while the relatively modest changes in caproic and succinic acid were only to a smaller extent explained by time ( P FDR 0.001–0.016, R 2 = 0.029–0.038). Acetic acid did not change over time ( P FDR = 0.96, R² < 0.001), and individual variability was statistically significant for this measure ( P FDR < 0.001). For model statistics and acid trajectories over time, see Supplementary Table S1 and Fig. 3 , respectively. Differences between patients and healthy controls LSAS-SR scores were significantly higher in SAD (M ± SD = 74.63 ± 19.81) patients than in HC (M ± SD = 10.49 ± 10.47) participants at baseline ( P < 0.001, d = 2.01; see Fig. 2 ). At both baselines, isobutyric concentration in SAD patients (M ± SD = 0.12 ± 0.05) was lower relative to HCs (M ± SD = 0.16 ± 0.06; P FDR = 0.006, Cohen’s d = 0.34). No other SCFA showed FDR corrected statistically significant differences between SAD patients and HCs ( P FDR ≥ 0.08; Fig. 4 , Supplementary Table S2 ). Fig. 4. Short-chain acid concentration differences in plasma between patients with social anxiety disorder (SAD) and healthy controls (HCs) at baseline. Open in a new tab The concentration (μM) of each SCFA in SAD patients (blue) and HCs (red) at the first baseline. The black dot represents the mean, and colored solid error bars represent 95% CI. Test-retest reliability of clinical assessments and short-chain fatty acid concentrations Test-retest reliability between the two baselines (spanning 2.5 months) was good to excellent for the LSAS-SR for both SAD (ICC = 0.89) and HCs (ICC = 0.83). SCFA outcomes demonstrated higher variability, with ICCs ranging from poor to moderate reliability (ICC = 0.17–0.71) (Supplementary Table S3 , Figure S4 - 14 ). Associations between acid concentration changes and anxiety reductions No statistically significant association between changes in SCFA concentrations and changes in LSAS-SR scores was found niether from the baseline to post-treatment, nor from the first to the 36-month follow-up (Supplementary Table S4 , Figure S15 - 16 ). Exploratory analyses Insomnia symptoms significantly decreased from pre- to (M ± SD = 9.33 ± 5.34) post-treatment (M ± SD = 6.67 ± 5.08; P < 0.001, Cohen’s d = 0.54), and the effect remained at the 36-month follow-up (M ± SD = 6.29 ± 5.18) ( P = 0.82, Cohen’s d = 0.037; Supplementary Table S5 , Figure S17 ). No statistically significant associations were found between changes in SCFA concentrations and changes in ISI scores, either from pre- to post-treatment ( r = 0.005–0.318, P FDR ≥ 0.34) or from post-treatment to the 36-month follow-up ( r = 0.06–0.24, P FDR ≥ 0.90). Further, no statistically significant associations were found between pre-treatment SCFA concentrations and pre-treatment LSAS-SR social anxiety symptoms, and all correlation coefficients were negligible to weak ( r = 0.01–0.29, P FDR ≥ 0.31). No statistically significant associations were found between pre-treatment SCFA concentrations and LSAS-SR treatment response, either at post-treatment ( r = 0.002–0.14, P FDR ≥ 0.88) or at the 36-month follow-up ( r = 0.06–0.23, P FDR ≥ 0.56). Discussion This study examined longitudinal changes in plasma SCFA concentrations following CBT in individuals with SAD, as well as multiple baseline differences relative to healthy controls. Over the 36-month study period, we observed increases in butyric, isobutyric, propionic and valeric acid concentrations, all of which followed non-linear trajectories primarily explained by time rather than individual variability, indicating a delayed but potentially therapy-related shift in microbial metabolite profiles. At pre-treatment (i.e., at both baseline assessments), SAD patients had significantly lower isobutyric acid concentrations compared to healthy controls. Butyric acid, in particular, is known for its beneficial effects on the intestinal barrier and its potent anti-inflammatory properties [ 43 – 46 ]. It may also influence brain function by reducing neuroinflammation, regulating neurotransmitter concentrations, and strengthening the integrity of the blood-brain barrier [ 47 – 50 ]. Supporting this, butyrate supplementation has been shown to improve behavioral and physiological abnormalities in individuals with autism spectrum disorder [ 51 , 52 ]. Moreover, a therapy combining sodium butyrate and fluoxetine has demonstrated greater antidepressant-like efficacy compared to using fluoxetine alone [ 53 ]. Further, an increase in plasma butyric acid concentrations was observed in schizophrenia patients after initial treatment with risperidone [ 54 ]. Here, we found that SAD patients treated with a non-pharmacological psychosocial intervention also showed increases in butyric acid at long-term follow-up. We also found that propionic acid increased after CBT for patients with SAD. Accumulating evidence suggests that propionic acid has significant anti-inflammatory properties [ 55 – 58 ]. Propionic acid also protects the blood-brain barrier by activating FFAR3 receptors on human brain endothelial cells, reducing inflammation and oxidative stress [ 11 ]. In a rodent model, oral administration of propionic acid alleviated depressive-like behaviors and ameliorated chronic stress-induced brain abnormalities by modulating intestinal flora composition and histone-3-related epigenetic mechanisms, suggesting its antidepressant potential [ 59 ]. Valeric acid has also been associated with a range of beneficial properties, including anticancer, antidiabetic, antihypertensive, anti-inflammatory, and immunomodulatory effects. It has been reported to influence molecular pathways involved in diseases such as Alzheimer’s disease, Parkinson’s disease, and epilepsy [ 60 – 62 ]. Here, we add that valeric acid may also be involved in patients’ response to a psychosocial intervention for anxiety disorders. Changes in caproic, formic, succinic, and acetic acids followed more linear trajectories in our study. Formic acid increased greatly, while caproic and succinic acids showed moderate increases, and acetic acid remained unchanged. Isovaleric acid increased until post-treatment, but was the only acid to later decline below baseline values. Notably, SCFAs with more linear trajectories were generally those that changed less over time, and were more influenced by individual-level factors, with minimal contribution from the fixed-effect of time (and indirectly the intervention). Our pre-treatment analyses revealed initial differences in gut microbiota-derived metabolites between SAD patients and healthy controls. Isobutyric acid concentrations were significantly lower in SAD patients compared to healthy controls ( P FDR = 0.006). Similar trends were observed for propionic and succinic acid ( P ≤ 0.027), though these did not remain significant after multiple comparison corrections ( P FDR = 0.080). Notably, these three acids were among the SCFAs that increased after CBT. Lower concentrations of these acids in SAD patients would support existing hypotheses about gut microbiota dysbiosis in anxiety disorders [ 7 , 63 ]. Jiang et al. [ 63 ] reported that patients with SAD exhibited reduced gut microbial diversity and a marked depletion of SCFA-producing genera such as Faecalibacterium and Eubacterium rectale . Consistent with our findings, reduced concentrations of propionic and succinic acids were also observed in patients with depression [ 16 ] and ADHD [ 14 ]. Notably, succinic acid is an intermediate metabolite in the fermentation towards propionic acid and derives also from the host’s mitochondrial activity [ 11 , 55 – 59 ]. Plasma levels of isobutyric acid were reported to be lower in patients with anorexia nervosa than in healthy controls and exhibited a positive correlation with body mass index (BMI) [ 64 ]. In the present study, this group difference remained stable across two baseline assessments conducted 10 to 11 weeks apart, and isobutyric acid levels subsequently normalized and increased following CBT. As previously reported from this cohort, CBT yielded a significant reduction in LSAS-SR scores over time, and the most pronounced improvement was observed immediately after treatment, followed by continued modest improvement during the long-term follow-up [ 28 ]. These findings are consistent with previous studies, confirming sustained and long-term effectiveness of internet-delivered CBT in the treatment of SAD [ 22 – 24 , 26 ]. Although SCFA concentrations changed significantly over time, particularly in butyric, formic, isobutyric, propionic and valeric acids, these changes were not significantly associated with changes in social anxiety symptom severity. This may be because of the temporal discrepancy between psychological and metabolic changes. This lag effect could obscure potential associations in a cross-timepoint correlation analysis. It could also be due to the heterogeneity of SAD and measurement of noise. The clinical features vary between individuals, and SAD is also associated with considerable variability in personality traits [ 56 ]. Likewise, SCFAs are highly sensitive to dietary intake and individual metabolic differences. Fibre-rich diets, such as the mediterranean diet, are known to enhance SCFA production [ 65 – 67 ], which can, in turn, affect the gut microbiota composition and subsequently alter bacterial metabolite profiles. It is thus possible that symptom improvements in SAD patients may gradually lead to changes in both lifestyle and dietary habits. While mixed models were applied to account for individual variability, the poor to moderate test-retest reliability observed for several SCFAs suggests measurement variability, which may have diluted real effects and reduced statistical power. Although the reliability of individual SCFA measures was modest, we observed a consistent and replicable difference in isobutyric acid concentration between patients and controls at two independent time points. This convergence of evidence supports the robustness of the group-level difference despite notable within-subject variability. Importantly, this pattern is consistent with a scenario in which within-individual measurements are noisy or state-dependent, yet systematic biological or pathological processes produce robust shifts at the group level. To our knowledge, this is the first study to examine longitudinal changes in plasma SCFA concentrations before and after psychosocial treatment for any disorders. An important strength is that we investigated not only the immediate alterations in SCFA concentrations but also monitored these metabolite changes longitudinally for two years. We quantified SCFAs in plasma rather than faeces, providing a longitudinal readout of circulating, systemically available metabolites directly relevant to gut–brain signalling. This contributes to a better understanding of the microbiota-gut-brain axis in psychiatric conditions. Limitations Long-term follow-ups with the healthy control group are lacking, as the study was statistically designed for within-patient group comparisons, but the study did utilize two baseline assessments to control the effects of time and regression to the mean in both patients and controls. Intrinsic to the study design, the samples from the last follow-up (4th follow-up) were stored at −80 °C for two years shorter than the other samples. However, we find this unlikely to explain the increased propionic, butyric, isobutyric and valeric acid concentrations in the fourth follow-up samples, relative to the first follow-up samples as (i) the other SCFAs did not increase from the first to the fourth follow-ups, (ii) SCFAs are reported stable for a few years at −80 °C [ 68 ], (iii) all samples had identical handling protocol when collected, and (iv) all samples were kept in sealed cryotubes and were freeze-thawed only once. The limited 10–11 week test–retest reliability of SCFA concentrations with ICCs ranging from poor to moderate (ICCs ranging from 0.17 to 0.71) suggests that SCFA concentrations may be influenced by transient or unmeasured factors such as recent dietary intake or metabolic fluctuations. The absence of faecal samples precluded microbiome metagenomic profiling, and we were therefore unable to directly link plasma SCFAs alterations to specific microbial taxa or functional pathways. Prior studies have reported weak, non-significant, or even inverse correlations between faecal and circulating SCFAs levels [ 69 , 70 ], highlighting the complex absorption and utilisation processes. Some of the gut-microbiome-derived SCFAs are consumed locally after absorption by the colonic epithelium, particularly butyrate. As the primary energy substrate for colonocytes, intracellular butyrate is rapidly oxidised via mitochondrial β-oxidation [ 71 – 73 ]. Following intestinal absorption, SCFAs are first transported via the portal vein to the liver, where a proportion undergoes hepatic metabolism before the remainder enters the systemic circulation. Consequently, faecal SCFAs primarily reflect microbial production within the gut lumen, whereas plasma SCFAs represent systemically available metabolites after epithelial utilisation and first-pass hepatic metabolism. Therefore, observed plasma SCFA changes are interpreted potentially as downstream metabolic readouts of host exposure and gut-brain signalling rather than direct proxies for microbial abundance or faecal production. This study did not include the assessments of dietary, physical activit, or other lifestyle-related factors, which may influence gut microbiota composition and SCFA production and may have changed gradually following anxiety symptom improvement. The observed SCFA alterations may reflect indirect effects of CBT mediated through lifestyle modifications rather than direct psychotherapeutic mechanisms. Participants fasted overnight before blood sampling, likely minimising the influence of dietary SCFAs absorbed in the small intestine. Future studies with dietary and lifestyle control and multi-omics integration are highly warranted to disentangle these interactions and assess the potential of microbiota-targeted interventions in SAD. Conclusions This longitudinal study demonstrated that plasma concentrations increased in propionic, butyric, isobutyric and valeric acids over 2 years after SAD patients received effective CBT, and that plasma concentration of isobutyric acid was lower in SAD patients relative to matched healthy controls before treatment onset. Our findings suggest that the effects of a psychosocial intervention may involve a microbiota-associated biological pathway. Supplementary information Supplemental Material (7.1MB, docx) Acknowledgements First and foremost, we sincerely thank all the participants for providing behavioral data and blood samples in this clinical trial. This study was performed at the Centre for Functional Brain Imaging (UFBI), Umeå University, and the University Hospital of Umeå, and we are grateful for all support from Lars Nyberg, and Carl-Johan Boraxbekk. The authors would also like to express their sincere gratitude to the staff of Kliniskt Forskningscentrum, Biobanken Norr, and Klinisk Kemi at Umeå University Hospital for providing excellent research conditions for collecting and storing blood samples. Gerhard Andersson, Martin Kraepelien, Lise Bergman Nordgren, Erik Hedman-Lagerlöf, Samir El Alaoui, Jens Högström, Nils Hentati Isacsson and Örn Kolbeinsson provided state-of-the-art clinical expertise. Author contributions Concept and design: Lavebratt, Furmark and Månsson. Acquisition, data management, analysis and interpretation of data: Wolthon, Stiernborg, Manzouri, Cai, Lavebratt and Månsson. Drafting of the manuscript: Cai. Preparing the final version of the manuscript: Cai, Wolthon, Furmark, Lavebratt and Månsson. Critical revision of the manuscript and acceptance to submit for publication: All authors. Supervision: Lavebratt and Månsson Funding source This work was supported by StratNeuro (to KM), Karolinska Institutet (to KM), and the Swedish Research Council (Grant Nos. 2018-06729 and 2023-01362 [to KM], 2016-02228 and 2020-02426 [to TF], and 2022-01188 [to CL]). This work was further supported by the Swedish Brain Foundation (Grant No. FO-2016-0106 [to KM and TF] and FO2025-0276-HK-280 [to CL]). WC was supported by China Scholarship Council. Further, KM is supported by the European Union - Nextgeneration EU and Romanian Government, under the National Recovery and Resilience Plan for Romania, contract no. 760043/23.05.2023, code PNRR-C9-I8-CF 278/30.11.2022, through the Romanian Ministry of Research, Innovation and Digitalization, within Component 9, Investment I8. Open access funding provided by Karolinska Institute. Data availability The code used to perform the statistical analyses is publicly available at https://github.com/neuronssonlab/Cai_etal_2026_Transl_Psychiatry . Data supporting the findings of this study are available from the corresponding author upon reasonable request. Competing interests The authors declare no competing interests. Ethics declarations The study was registered at ClinicalTrials.gov ( NCT02592564 ), and ethical approval was obtained from the regional committee at Umeå University, Umeå, Sweden. All participants gave written informed consent prior to participation. All methods were performed in accordance with the relevant guidelines and regulations. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41398-026-04134-y . References 1. Stein DJ, Lim CCW, Roest AM, de Jonge P, Aguilar-Gaxiola S, Al-Hamzawi A, et al. The cross-national epidemiology of social anxiety disorder: data from the world mental health survey initiative. BMC Med. 2017;15:143. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Aderka IM, Hofmann SG, Nickerson A, Hermesh H, Gilboa-Schechtman E, Marom S. Functional impairment in social anxiety disorder. J Anxiety Disord. 2012;26:393–400. [ DOI ] [ PubMed ] [ Google Scholar ] 3. Leichsenring F, Leweke F. Social anxiety disorder. N Engl J Med. 2017;376:2255–64. [ DOI ] [ PubMed ] [ Google Scholar ] 4. Ruscio AM, Brown TA, Chiu WT, Sareen J, Stein MB, Kessler RC. Social fears and social phobia in the USA: results from the national comorbidity survey replication. Psychol Med. 2008;38:15–28. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Scaini S, Belotti R, Ogliari A. Genetic and environmental contributions to social anxiety across different ages: a meta-analytic approach to twin data. J Anxiety Disord. 2014;28:650–6. [ DOI ] [ PubMed ] [ Google Scholar ] 6. Spence SH, Rapee RM. The etiology of social anxiety disorder: an evidence-based model. Behav Res Ther. 2016;86:50–67. [ DOI ] [ PubMed ] [ Google Scholar ] 7. Butler MI, Bastiaanssen TFS, Long-Smith C, Morkl S, Berding K, Ritz NL, et al. The gut microbiome in social anxiety disorder: evidence of altered composition and function. Transl Psychiatry. 2023;13:95. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Ritz NL, Brocka M, Butler MI, Cowan CSM, Barrera-Bugueño C, Turkington CJR, et al. Social anxiety disorder-associated gut microbiota increases social fear. Proc Natl Acad Sci USA. 2024;121:e2308706120. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Morrison DJ, Preston T. Formation of short chain fatty acids by the gut microbiota and their impact on human metabolism. Gut Microbes. 2016;7:189–200. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Duscha A, Gisevius B, Hirschberg S, Yissachar N, Stangl GI, Dawin E, et al. Propionic acid shapes the multiple sclerosis disease course by an immunomodulatory mechanism. Cell. 2020;180:1067–80. e16. [ DOI ] [ PubMed ] [ Google Scholar ] 11. Hoyles L, Snelling T, Umlai UK, Nicholson JK, Carding SR, Glen RC, et al. Microbiome-host systems interactions: protective effects of propionate upon the blood-brain barrier. Microbiome. 2018;6:55. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Dalile B, Van Oudenhove L, Vervliet B, Verbeke K. The role of short-chain fatty acids in microbiota-gut-brain communication. Nat Rev Gastroenterol Hepatol. 2019;16:461–78. [ DOI ] [ PubMed ] [ Google Scholar ] 13. Yang LL, Millischer V, Rodin S, MacFabe DF, Villaescusa JC, Lavebratt C. Enteric short-chain fatty acids promote proliferation of human neural progenitor cells. J Neurochem. 2020;154:635–46. [ DOI ] [ PubMed ] [ Google Scholar ] 14. Yang LL, Stiernborg M, Skott E, Gillberg T, Landberg R, Giacobini M, et al. Lower plasma concentrations of short-chain fatty acids (SCFAs) in patients with ADHD. J Psychiatr Res. 2022;156:36–43. [ DOI ] [ PubMed ] [ Google Scholar ] 15. Wang L, Christophersen CT, Sorich MJ, Gerber JP, Angley MT, Conlon MA. Elevated fecal short chain fatty acid and ammonia concentrations in children with autism spectrum disorder. Dig Dis Sci. 2012;57:2096–102. [ DOI ] [ PubMed ] [ Google Scholar ] 16. Skonieczna-Żydecka K, Grochans E, Maciejewska D, Szkup M, Schneider-Matyka D, Jurczak A, et al. Faecal short chain fatty acids profile is changed in polish depressive women. Nutrients. 2018;10:1939. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Rosell-Cardona C, Leigh SJ, Knox E, Tirelli E, Lyte JM, Goodson MS, et al. Acute stress-induced alterations in short-chain fatty acids: implications for the intestinal and blood brain barriers. Brain Behav Immun Health. 2025;46:100992. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Tulbure BT, Szentagotai A, David O, Ștefan S, Månsson KN, David D, et al. Internet-delivered cognitive-behavioral therapy for social anxiety disorder in Romania: a randomized controlled trial. PLoS One. 2015;10:e0123997. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Andersson G, Carlbring P, Holmström A, Sparthan E, Furmark T, Nilsson-Ihrfelt E, et al. Internet-based self-help with therapist feedback and in vivo group exposure for social phobia: a randomized controlled trial. J Consult Clin Psychol. 2006;74:677–86. [ DOI ] [ PubMed ] [ Google Scholar ] 20. Pelissolo A, Abou Kassm S, Delhay L. Therapeutic strategies for social anxiety disorder: where are we now?. Expert Rev Neurother. 2019;19:1179–89. [ DOI ] [ PubMed ] [ Google Scholar ] 21. Kindred R, Bates GW, McBride NL. Long-term outcomes of cognitive behavioural therapy for social anxiety disorder: a meta-analysis of randomised controlled trials. J Anxiety Disord. 2022;92:102640. [ DOI ] [ PubMed ] [ Google Scholar ] 22. Hedman E, Furmark T, Carlbring P, Ljótsson B, Rück C, Lindefors N, et al. A 5-year follow-up of internet-based cognitive behavior therapy for social anxiety disorder. J Med Internet Res. 2011;13:e39. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Månsson KN, Salami A, Frick A, Carlbring P, Andersson G, Furmark T, et al. Neuroplasticity in response to cognitive behavior therapy for social anxiety disorder. Transl Psychiatry. 2016;6:e727. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Månsson KNT, Waschke L, Manzouri A, Furmark T, Fischer H, Garrett DD. Moment-to-Moment brain signal variability reliably predicts psychiatric treatment outcome. Biol Psychiatry. 2022;91:658–66. [ DOI ] [ PubMed ] [ Google Scholar ] 25. Tsikonofilos K, Kumar A, Ampatzis K, Garrett DD, Månsson KNT. The promise of investigating neural variability in psychiatric disorders. Biol Psychiatry. 2025;98:195–207. [ DOI ] [ PubMed ] [ Google Scholar ] 26. Månsson KNT, Lindqvist D, Yang LL, Svanborg C, Isung J, Nilsonne G, et al. Improvement in indices of cellular protection after psychological treatment for social anxiety disorder. Transl Psychiatry. 2019;9:340. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Lindqvist D, Furmark T, Lavebratt C, Ohlsson L, Månsson KNT. Plasma circulating cell-free mitochondrial DNA in social anxiety disorder. Psychoneuroendocrinology. 2023;148:106001. [ DOI ] [ PubMed ] [ Google Scholar ] 28. Garke M, Hentati Isacsson N, Kolbeinsson Ö, Hesser H, Månsson KNT. Improvements in emotion regulation during cognitive behavior therapy predict subsequent social anxiety reductions. Cogn Behav Ther. 2025;54:78–95. [ DOI ] [ PubMed ] [ Google Scholar ] 29. Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, et al. The mini-international neuropsychiatric interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry. 1998;59:22–33. [ PubMed ] [ Google Scholar ] 30. First MB Structured clinical interview for DSM-IV-TR axis I disorders, research version, patient edition (SCID-I/P). New York: Biometrics Research, New York State Psychiatric Institute; 2002. 31. Bastien CH, Vallières A, Morin CM. Validation of the insomnia severity index as an outcome measure for insomnia research. Sleep Med. 2001;2:297–307. [ DOI ] [ PubMed ] [ Google Scholar ] 32. Andersson G, Carlbring P, Ljótsson B, Hedman E. Guided internet-based CBT for common mental disorders. J Contemp Psychother. 2013;43:223–33. [ Google Scholar ] 33. Romm KL, Rossberg JI, Berg AO, Hansen CF, Andreassen OA, Melle I. Assessment of social anxiety in first episode psychosis using the Liebowitz Social Anxiety scale as a self-report measure. Eur Psychiatry. 2011;26:115–21. [ DOI ] [ PubMed ] [ Google Scholar ] 34. Rytwinski NK, Fresco DM, Heimberg RG, Coles ME, Liebowitz MR, Cissell S, et al. Screening for social anxiety disorder with the self-report version of the Liebowitz Social Anxiety Scale. Depress Anxiety. 2009;26:34–8. [ DOI ] [ PubMed ] [ Google Scholar ] 35. Fresco DM, Coles ME, Heimberg RG, Liebowitz MR, Hami S, Stein MB, et al. The Liebowitz Social Anxiety Scale: a comparison of the psychometric properties of self-report and clinician-administered formats. Psychol Med. 2001;31:1025–35. [ DOI ] [ PubMed ] [ Google Scholar ] 36. Han J, Lin K, Sequeira C, Borchers CH. An isotope-labeled chemical derivatization method for the quantitation of short-chain fatty acids in human feces by liquid chromatography-tandem mass spectrometry. Anal Chim Acta. 2015;854:86–94. [ DOI ] [ PubMed ] [ Google Scholar ] 37. Wood S. Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J R Stat Soc Ser B Stat Methodol. 2011;73:3–36. [ Google Scholar ] 38. Nakagawa S, Johnson PCD, Schielzeth H. The coefficient of determination R(2) and intra-class correlation coefficient from generalized linear mixed-effects models revisited and expanded. J R Soc Interface. 2017;14:20170213. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Lüdecke D, Ben-Shachar MS, Patil I, Waggoner P, Makowski D. performance: an R package for assessment, comparison and testing of statistical models. J Open Source Softw. 2021;6:3139. [ Google Scholar ] 40. Bates D, Mächler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. J Stat Softw. 2015;67:1–48. [ Google Scholar ] 41. Revelle W. psych: procedures for psychological, psychometric, and personality research. R Package Version 1.0–95. Evanston, IL: Northwestern Universit. 2013. 42. McGraw KO, Wong SP. Forming inferences about some intraclass correlation coefficients. Psychol Methods. 1996;1:30–46. [ Google Scholar ] 43. Chen G, Ran X, Li B, Li Y, He D, Huang B, et al. Sodium butyrate inhibits inflammation and maintains epithelium barrier integrity in a TNBS-induced inflammatory bowel disease mice model. EBioMedicine. 2018;30:317–25. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 44. Peng L, Li ZR, Green RS, Holzman IR, Lin J. Butyrate enhances the intestinal barrier by facilitating tight junction assembly via activation of AMP-activated protein kinase in Caco-2 cell monolayers. J Nutr. 2009;139:1619–25. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 45. Kovarik JJ, Tillinger W, Hofer J, Hölzl MA, Heinzl H, Saemann MD, et al. Impaired anti-inflammatory efficacy of n-butyrate in patients with IBD. Eur J Clin Invest. 2011;41:291–8. [ DOI ] [ PubMed ] [ Google Scholar ] 46. Meijer K, de Vos P, Priebe MG. Butyrate and other short-chain fatty acids as modulators of immunity: what relevance for health?. Curr Opin Clin Nutr Metab Care. 2010;13:715–21. [ DOI ] [ PubMed ] [ Google Scholar ] 47. Stilling RM, van de Wouw M, Clarke G, Stanton C, Dinan TG, Cryan JF. The neuropharmacology of butyrate: the bread and butter of the microbiota-gut-brain axis?. Neurochem Int. 2016;99:110–32. [ DOI ] [ PubMed ] [ Google Scholar ] 48. Wei H, Yu C, Zhang C, Ren Y, Guo L, Wang T, et al. Butyrate ameliorates chronic alcoholic central nervous damage by suppressing microglia-mediated neuroinflammation and modulating the microbiome-gut-brain axis. Biomed Pharmacother. 2023;160:114308. [ DOI ] [ PubMed ] [ Google Scholar ] 49. Li H, Sun J, Du J, Wang F, Fang R, Yu C, et al. Clostridium butyricum exerts a neuroprotective effect in a mouse model of traumatic brain injury via the gut-brain axis. Neurogastroenterol Motil. 2018;30:e13260. [ DOI ] [ PubMed ] [ Google Scholar ] 50. Fock E, Parnova R. Mechanisms of blood-brain barrier protection by microbiota-derived short-chain fatty acids. Cells. 2023;12:657. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 51. Rose S, Bennuri SC, Davis JE, Wynne R, Slattery JC, Tippett M, et al. Butyrate enhances mitochondrial function during oxidative stress in cell lines from boys with autism. Transl Psychiatry. 2018;8:42. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 52. Kratsman N, Getselter D, Elliott E. Sodium butyrate attenuates social behavior deficits and modifies the transcription of inhibitory/excitatory genes in the frontal cortex of an autism model. Neuropharmacology. 2016;102:136–45. [ DOI ] [ PubMed ] [ Google Scholar ] 53. Schroeder FA, Lin CL, Crusio WE, Akbarian S. Antidepressant-like effects of the histone deacetylase inhibitor, sodium butyrate, in the mouse. Biol Psychiatry. 2007;62:55–64. [ DOI ] [ PubMed ] [ Google Scholar ] 54. Li X, Fan X, Yuan X, Pang L, Hu S, Wang Y, et al. The role of butyric acid in treatment response in drug-naïve first episode schizophrenia. Front Psychiatry. 2021;12:724664. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 55. Ali I, Yang M, Wang Y, Yang C, Shafiq M, Wang G, et al. Sodium propionate protect the blood-milk barrier integrity, relieve lipopolysaccharide-induced inflammatory injury and cells apoptosis. Life Sci. 2021;270:119138. [ DOI ] [ PubMed ] [ Google Scholar ] 56. Tobin D, Vige R, Calder PC. Review: the nutritional management of multiple sclerosis with propionate. Front Immunol. 2021;12:676016. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 57. Tedelind S, Westberg F, Kjerrulf M, Vidal A. Anti-inflammatory properties of the short-chain fatty acids acetate and propionate: a study with relevance to inflammatory bowel disease. World J Gastroenterol. 2007;13:2826–32. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 58. Nastasi C, Candela M, Bonefeld CM, Geisler C, Hansen M, Krejsgaard T, et al. The effect of short-chain fatty acids on human monocyte-derived dendritic cells. Sci Rep. 2015;5:16148. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 59. Behrens LMP, Gasparotto J, Rampelotto PH, Escalona MAR, da Silva LDS, Carazza-Kessler FG, et al. Sodium propionate oral supplementation ameliorates depressive-like behavior through gut microbiome and histone 3 epigenetic regulation. J Nutr Biochem. 2024;130:109660. [ DOI ] [ PubMed ] [ Google Scholar ] 60. Kumari B, Kumari U, Singh DK, Husain GM, Patel DK, Shakya A, et al. Molecular targets of valeric acid: a bioactive natural product for endocrine, metabolic, and immunological disorders. Endocr Metab Immune Disord Drug Targets. 2024;24:1506–17. [ DOI ] [ PubMed ] [ Google Scholar ] 61. Jayaraj RL, Beiram R, Azimullah S, Mf NM, Ojha SK, Adem A, et al. Valeric acid protects dopaminergic neurons by suppressing oxidative stress, neuroinflammation and modulating autophagy pathways. Int J Mol Sci. 2020;21:7670. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 62. Liu M, Zhang Y, Liu J, Xiang C, Lu Q, Lu H, et al. Revisiting the role of valeric acid in manipulating ulcerative colitis. Inflamm Bowel Dis. 2024;30:617–28. [ DOI ] [ PubMed ] [ Google Scholar ] 63. Jiang HY, Zhang X, Yu ZH, Zhang Z, Deng M, Zhao JH, et al. Altered gut microbiota profile in patients with generalized anxiety disorder. J Psychiatr Res. 2018;104:130–6. [ DOI ] [ PubMed ] [ Google Scholar ] 64. Xu J, Landberg R, Lavebratt C, Bulik CM, Landén M, Nilsson IAK. Plasma concentrations of short-chain fatty acids in active and recovered anorexia nervosa. Nutrients. 2022;14:5247. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 65. Haskey N, Estaki M, Ye J, Shim RK, Singh S, Dieleman LA, et al. A mediterranean diet pattern improves intestinal inflammation concomitant with reshaping of the bacteriome in ulcerative colitis: a randomised controlled trial. J Crohns Colitis. 2023;17:1569–78. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 66. Joseph J, Depp C, Shih PB, Cadenhead KS, Schmid-Schönbein G. Modified mediterranean diet for enrichment of short chain fatty acids: potential adjunctive therapeutic to target immune and metabolic dysfunction in schizophrenia?. Front Neurosci. 2017;11:155. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 67. Seethaler B, Nguyen NK, Basrai M, Kiechle M, Walter J, Delzenne NM, et al. Short-chain fatty acids are key mediators of the favorable effects of the mediterranean diet on intestinal barrier integrity: data from the randomized controlled LIBRE trial. Am J Clin Nutr. 2022;116:928–42. [ DOI ] [ PubMed ] [ Google Scholar ] 68. Wagner-Golbs A, Neuber S, Kamlage B, Christiansen N, Bethan B, Rennefahrt U, et al. Effects of long-term storage at −80 °C on the human plasma metabolome. Metabolites. 2019;9:99. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 69. Nogal A, Asnicar F, Vijay A, Kouraki A, Visconti A, Louca P, et al. Genetic and gut microbiome determinants of SCFA circulating and fecal levels, postprandial responses and links to chronic and acute inflammation. Gut Microbes. 2023;15:2240050. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 70. Chen SJ, Chen CC, Liao HY, Lin YT, Wu YW, Liou JM, et al. Association of fecal and plasma levels of short-chain fatty acids with gut microbiota and clinical severity in patients with Parkinson disease. Neurology. 2022;98:e848–e58. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 71. Roediger WE. Role of anaerobic bacteria in the metabolic welfare of the colonic mucosa in man. Gut. 1980;21:793–8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 72. Koh A, De Vadder F, Kovatcheva-Datchary P, Bäckhed F. From dietary fiber to host physiology: short-chain fatty acids as key bacterial metabolites. Cell. 2016;165:1332–45. [ DOI ] [ PubMed ] [ Google Scholar ] 73. Clausen MR, Mortensen PB. Kinetic studies on the metabolism of short-chain fatty acids and glucose by isolated rat colonocytes. Gastroenterology. 1994;106:423–32. [ DOI ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplemental Material (7.1MB, docx) Data Availability Statement The code used to perform the statistical analyses is publicly available at https://github.com/neuronssonlab/Cai_etal_2026_Transl_Psychiatry . Data supporting the findings of this study are available from the corresponding author upon reasonable request. Articles from Translational Psychiatry are provided here courtesy of Nature Publishing Group ACTIONS View on publisher site PDF (5.3 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Tachikawa Project for Prevention of Posttraumatic Stress Disorder With Polyunsaturated Fatty Acid (TPOP): TPOP-02 Study#1172272
• Tachikawa Project for Prevention of Posttraumatic Stress Disorder With Polyunsaturated Fatty Acid (TPOP): TPOP-01 Study#833886
• Plasma metabolomics and red blood cell fatty acid profiles in adolescent mental health#123219
• An Open Label Trial of Omega 3(o3mega+Joy)in the Treatment of Social Phobia#885757
• Dietary exposures and common mental illness in the Netherlands Study of Depression and Anxiety (NESDA): a cohort-level GLAD project analysis#422486