Association of the sulfur microbial diet with metabolic profile, gut-related biomarkers, and adiposity indicators in adults with overweight and obesity Authors: Amr Ali Mohamed Abdelgawwad El-Sehrawy, Ismail Kandil, Gehad Mohsen ELmazny, Elsayed A Eid, Sherif M Yusuf, Mansuor A Alanazi, Safa Labib, Taghreed Saud M Almansouri, Rufaida Abdalla, Ahmed M Salah Published: 2026 ABSTRACT AbstractThe sulfur microbial diet (SMD), characterized by high intake of processed meats, alcohol, and artificially sweetened beverages, has been suggested to be associated with gut microbial dysbiosis by influencing sulfur-reducing bacterial activity. These microbial changes may be associated with altered production of hydrogen sulfide (H₂S) and lipopolysaccharide (LPS), potentially contributing to systemic inflammation, metabolic disorders, alterations in gut-related biomarkers such as nerve growth factor (NGF). However, the metabolic and neuroimmune consequences of this dietary pattern in overweight and obese individuals remain unclear. In the current study we aimed to examine the association between adherence to the SMD and metabolic risk factors, gut–brain axis biomarkers (LPS and NGF), and adiposity indicators in a population of overweight and obese adults. In this cross-sectional study, 633 overweight and obese adults (BMI ≥ 25 kg/m²) aged 20–60 years were recruited from outpatient nutrition clinics. Dietary intake was assessed using a previously validated food frequency questionnaire, and SMD scores were calculated based on the intake of eight food groups included in the original SMD scoring methodology. Participants were categorized into quartiles of SMD adherence. Anthropometric measures and fasting blood samples were collected to assess metabolic markers, LPS, and NGF. Participants with higher SMD scores had significantly higher total cholesterol levels (adjusted P = 0.028), while adjusted comparisons showed a significant difference in circulating NGF levels between SMD quartiles (adjusted P = 0.047). LPS concentrations differed significantly across SMD quartiles after multivariable adjustment (adjusted P = 0.037), with a significant linear trend across quartiles (P for trend = 0.011). No significant differences were observed for fasting glucose, insulin, triglycerides, or blood pressure. Higher adherence to the sulfur microbial diet was associated with differences in total cholesterol, NGF, and LPS concentrations among adults with overweight and obesity. These findings suggest that greater adherence to this dietary pattern may be accompanied by alterations in metabolic and gut-related biomarkers. Given the cross-sectional design of the study, causal relationships cannot be inferred, and prospective studies are needed to confirm these associations.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-63258-9.Keywords: Sulfur microbial diet, SMD, Gut microbiota, Obesity, Lipopolysaccharide, Nerve growth factor, AdipositySubject terms: Biomarkers, Diseases, Endocrinology, Neurology, Neuroscience Introduction Obesity remains a growing global health crisis, contributing to the increasing prevalence of chronic conditions such as type 2 diabetes, cardiovascular diseases, and several types of cancer. The development of obesity is influenced by a complex interplay of genetic, behavioral, and environmental factors, among which dietary habits play a central role. In recent years, emerging evidence has highlighted the importance of gut microbiota as a key mediator between dietary intake and metabolic health. Disruptions in microbial composition and function—often referred to as dysbiosis—have been associated with increased adiposity, insulin resistance, and systemic inflammation, all of which are central features of metabolic disorders1,2. The Sulfur Microbial Diet (SMD) refers to a dietary pattern characterized by a higher intake of foods that potentially enhance sulfur-reducing microbial activity in the gut. This diet typically emphasizes consumption of processed meats, alcoholic beverages (particularly spirits), and artificially sweetened drinks—foods positively associated with elevated levels of sulfur-metabolizing bacteria. In contrast, it is marked by lower consumption of items such as legumes, non-starchy vegetables, fruit juices, sweets, and beer, which tend to have inverse associations with sulfur-metabolizing microbial abundance3. These dietary choices contribute to increased intestinal sulfur availability, fostering the growth of specific gut bacteria like Bilophila wadsworthia, Desulfovibrio spp., and Fusobacterium spp., which utilize sulfur compounds to generate hydrogen sulfide (H₂S)4. Excess H₂S may disrupt mucosal barrier function, enhance intestinal permeability, and promote systemic inflammation through microbial translocation and metabolic endotoxemia5,6. Moreover, growing epidemiologic evidence suggests that long-term adherence to sulfur-promoting diets may increase the risk of cardiometabolic abnormalities, chronic inflammation, and gastrointestinal disorders7,8. The gut–brain axis is a dynamic, bidirectional communication network between the gastrointestinal tract and central nervous system, regulated through neural, endocrine, and immune pathways. Disruptions in the gut microbiota—particularly those triggered by unbalanced diets—can alter this communication and contribute to both metabolic and neuroinflammatory disorders9. One key mechanism involves the translocation of lipopolysaccharide (LPS), a potent endotoxin derived from gram-negative bacteria, into systemic circulation. This process, often referred to as metabolic endotoxemia, results from increased intestinal permeability and activates pro-inflammatory signaling cascades, primarily through Toll-like receptor 4 (TLR4)10,11. Elevated LPS levels have been shown to promote insulin resistance, visceral adiposity, and low-grade systemic inflammation in individuals with obesity12. Emerging evidence also suggests that microbial endotoxins such as LPS may influence the expression of neurotrophic factors, including nerve growth factor (NGF). NGF is a key regulator of neuronal survival, neuroplasticity, and neuroimmune responses. Altered NGF levels have been associated with chronic inflammatory states, obesity, and metabolic dysfunction13. Furthermore, experimental models indicate that LPS exposure may suppress NGF signaling and contribute to neuroinflammation, possibly affecting mood, appetite regulation, and energy homeostasis14. Given that the SMD fosters the growth of LPS-producing and sulfur-reducing bacteria, it may represent a novel dietary factor influencing gut-derived inflammation and NGF-related neurobiological pathways in obesity15. Despite growing interest in the role of diet–microbiota interactions in obesity and metabolic diseases, limited research has explored the effects of the SMD on host metabolic profile, gut-derived inflammatory markers, and neurotrophic signaling pathways in humans. While prior studies have linked sulfur-promoting diets to colorectal cancer and systemic inflammation3,6,8,11,12, little is known about their relationship with NGF and LPS levels—key mediators of the gut–brain axis and chronic inflammation. Moreover, no study to date has simultaneously examined how the SMD relates to adiposity indicators, including central and general obesity, in obese individuals. To address this gap, the present study aimed to investigate the association between adherence to the SMD and metabolic parameters, gut–brain axis biomarkers (NGF and LPS), and adiposity indicators in a population of overweight and obese adults. By integrating dietary, biochemical, and neuroimmune data, this research offers a novel perspective on how diet-driven microbial pathways may contribute to obesity-related metabolic and neurological dysfunction. The findings may support the development of microbiome-informed dietary strategies for metabolic disease prevention and management. Methods Participants’ recruitment This cross-sectional study was conducted among adults with overweight or obesity who were recruited from six obesity and weight-management outpatient clinics affiliated with Mansoura University, Mansoura, Egypt, between 2024 and 2025. Eligible participants were identified through recruitment announcements displayed in the participating clinics and university facilities. Interested individuals were screened by the research team according to the predefined inclusion and exclusion criteria. Participants were recruited using a convenience sampling approach. Before enrollment, all participants were informed about the study objectives, procedures, and potential risks, and written informed consent was obtained from each participant. The study protocol was approved by the Institutional Review Board of Mansoura University (Approval No. R.26.02.3600) before participant recruitment and data collection. All study procedures were conducted in accordance with the Declaration of Helsinki. Eligible participants were adults aged 20–60 years with a body mass index (BMI) ≥ 25 kg/m², based on World Health Organization (WHO) classification criteria16. Individuals were excluded if they had a history of chronic diseases (e.g., type 1 diabetes, cardiovascular disease, cancer, inflammatory bowel disease), recent acute illness, gastrointestinal surgery, current use of antibiotics or probiotics within the last three months, or if they were pregnant, lactating, or following a special diet. The required sample size was calculated using G*Power software (Heinrich Heine University Düsseldorf, Germany) with the following parameters: effect size (f²) of 0.05 (small to moderate), α-error probability of 0.05, power (1 − β) of 0.9017. The initial sample size required was estimated at 570 participants. Considering an anticipated dropout and missing data rate of approximately 10%, the final required sample size was calculated as 633. The flowchart of participant recruitment is represented in Fig. 1. The study flowchart. Dietary and physical activity assessment Dietary intake was assessed using a semi-quantitative food frequency questionnaire (FFQ) adapted from a previously developed FFQ18, which has been widely applied in Arab populations. This instrument was selected based on previous applications and validation studies in Arab populations, including the Mutaba’ah Study18. The adapted FFQ was reviewed by nutrition experts to ensure the relevance and clarity of food items for the study population. The final FFQ included 146 food items covering major food groups, including dairy products, fats, vegetables, fruits, meat, fish, eggs, mixed dishes, bread and cereals, beverages, sweets and baked products, nuts, and seeds. Participants reported the frequency of consumption and standard portion sizes for each food item, which were converted into average daily intake estimates (g/day) using household measures and national food composition tables. The average daily intake of foods and nutrients was calculated based on reported frequency and serving sizes. The FFQ was administered in Arabic by trained researchers through self-completion or interviewer-assisted interviews when needed. Dietary data were analyzed using Nutritionist IV software to estimate daily energy, macronutrient, and food group intakes. Alcoholic beverages were included as dietary items in the FFQ because they represent one of the components of the SMD score. Given the very low frequency of alcohol consumption in this population, alcohol intake was not evaluated as an independent dietary exposure, and former drinkers and non-drinkers were not analyzed as separate categories. To calculate the SMD score, food items were categorized based on previously established associations with sulfur-metabolizing bacterial abundance, as described by Nguyen et al.3. In their model, food group intakes were linked to the log-transformed relative abundance of 43 sulfur-related microbial taxa using β-coefficients derived from multivariable regression analyses. Accordingly, three food groups showed positive β-values—processed meats (β = 0.64), low-calorie drinks (β = 0.38), and liquor (β = 0.31)—indicating direct associations with sulfur-metabolizing bacteria. Conversely, five groups displayed negative β-values—beer (β = −0.54), fruit juice (β = −0.21), legumes (β = −0.64), other vegetables (β = −0.30), and sweets/desserts (β = −0.23)—reflecting inverse relationships. The SMD score in this study was calculated using eight food groups, where the intake of each group (g/day) was multiplied by its corresponding β-coefficient. The total SMD score was obtained by summing these weighted values, and participants were subsequently categorized into quartiles based on their overall SMD score. SMD components alongside derived β-coefficients and their corresponding standard errors are illustrated in Fig. 2. Physical activity (PA) was assessed using the International Physical Activity Questionnaire (IPAQ), and metabolic equivalent of task (MET) minutes per week were calculated according to the IPAQ scoring protocol. Total physical activity was calculated by summing the MET-minutes/week from walking, moderate-intensity, and vigorous-intensity activities. Participants were categorized into low, moderate, and high physical activity levels according to the established IPAQ classification criteria; low physical activity was defined as < 600 MET-min/week, moderate physical activity as 600–2999 MET-min/week, and high physical activity as ≥ 3000 MET-min/week19. Sulfur Microbial Diet (SMD) components with β-coefficients attributed for each food item. Anthropometric assessments Body weight was measured using a calibrated digital scale (Seca, Hamburg, Germany) with participants in light clothing and without shoes, and recorded to the nearest 0.1 kg. Height was measured to the nearest 0.1 cm using a stadiometer in a standing position with no shoes. Body mass index (BMI) was calculated by dividing weight (kg) by height squared (m²), and classified according to the World Health Organization criteria. Waist circumference (WC) was measured using a non-stretchable tape at the midpoint between the lowest rib and the iliac crest, and recorded to the nearest 0.1 cm. WHR was calculated accordingly. All anthropometric measurements were performed by trained personnel under standardized conditions. Biochemical analysis After an overnight fast (10–12 h), venous blood samples were collected from all participants. Serum was separated by centrifugation at 3000 rpm for 10 min and stored at − 80 °C until analysis. Biochemical markers, including total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), triglycerides (TG), and fasting blood glucose (FBG) were measured using enzymatic colorimetric methods with commercial kits (Pars Azmoon Co., Tehran, Iran) on an automated analyzer (BT 1500, Biotecnica Instruments, Italy). Low-density lipoprotein cholesterol (LDL-C) was calculated using the Friedewald formula for participants with TG < 400 mg/dL. Serum levels of LPS and NGF were assessed as indicators of microbial endotoxemia and neurotrophic activity, respectively. Concentrations of LPS were measured using a quantitative enzyme-linked immunosorbent assay (ELISA) kit (CUSABIO Biotech Co., Wuhan, China; Catalog No. CSB-E09945h), with a sensitivity lower than 1.56 pg/mL and intra-assay and inter-assay coefficients of variation (CVs) below 10%. NGF levels were determined using a high-sensitivity human NGF ELISA kit (Elabscience, USA; Catalog No. E-EL-H6035), with a detection range of 7.81–500 pg/mL. All assays were performed according to the manufacturer’s instructions, and measurements were conducted in duplicate to ensure accuracy. Laboratory personnel involved in biochemical analyses were blinded to participants’ dietary exposure categories and SMD quartile assignments. Statistical analysis Data were analyzed using IBM SPSS Statistics for Windows, version 26.0 (IBM Corp., Armonk, NY, USA). The distribution of continuous variables was evaluated using the Kolmogorov–Smirnov test to determine normality. Results for continuous variables are presented as mean ± standard deviation (SD), while categorical variables are summarized as frequencies and percentages. To compare general characteristics across quartiles of the SMD score, one-way analysis of variance (ANOVA) was employed for continuous variables with approximately normal distributions, whereas the Kruskal–Wallis test was used for variables with non-normal distributions (i.e., triglycerides and insulin). The chi-square test was used for categorical variables. Post hoc analyses were performed using the Bonferroni correction test. General linear models (GLM) were applied to obtain adjusted P-values for the comparison of serum concentrations of metabolic and biochemical biomarkers after adjustment for potential confounders (e.g., age, sex, total energy intake, physical activity, BMI, education, and marital status, as appropriate). Likewise, in the comparison of dietary food group intakes across SMD quartiles, residual confounding by total energy intake was controlled using GLM20,21. For comparative analyses, the SMD score was categorized into quartiles, and all group comparisons were conducted according to these quartile classifications. The total SMD score ranged from − 23.60 to 98.45. The quartile boundaries were as follows: Q1 (− 23.60 to 13.85), Q2 (> 13.85 to 27.31), Q3 (> 27.31 to 43.12), and Q4 (> 43.12 to 98.45). In addition, P for trend across SMD quartiles was assessed using multivariable linear regression models. Before fitting the regression models, model assumptions were evaluated. Multicollinearity was assessed using variance inflation factor (VIF) and tolerance statistics. The normality of standardized residuals was examined by visual inspection of the normal probability (P–P) plot. Independence of residuals was assessed using the Durbin–Watson statistic, with a value of 1.86, indicating no evidence of residual autocorrelation. Collinearity diagnostics are presented in Supplementary Tables 1, and the residual normality assessment is provided in Supplementary Figure S1. All statistical tests were two-tailed, and a P value < 0.05 was considered statistically significant. The graphical abstract in the current manuscript was developed by the authors, with visual support provided by the ChatGPT platform (OpenAI) based on the data generated in this study. Results A total of 633 adults with overweight or obesity were categorized into quartiles based on their adherence to the SMD. Table 1 presents the general demographic and anthropometric characteristics of the participants across SMD score quartiles. Significant differences were observed across SMD quartiles for age (P = 0.039), sex distribution (P = 0.026), and waist circumference (P = 0.039). After adjustment for age, sex, physical activity, education, marital status, and total energy intake, no significant differences remained in the anthropometric variables across SMD quartiles (all adjusted P > 0.05). Smoking status, educational level, marital status, physical activity, weight, height, BMI, and waist-to-hip ratio did not differ significantly among the quartiles. Table 1General demographic and anthropometric characteristics of study participants across quartiles of SMD.VariableSMD quartilesP-value*P-value **1st quartile (n = 158)2nd quartile (n = 158)3rd quartile (n = 159)4th quartile (n = 158)Age (y)40.73 ± 10.0440.32 ± 9.7640.71 ± 9.6538.09 ± 8.23 0.039† –Male sex, n (%)70 (44.3)67 (42.4)76 (47.8)88 (55.7) 0.026 –Physically active, n (%)14 (8.9)14 (8.9)11 (6.9)17 (10.8)0.613–Smoking [smokers, n (%)]20 (12.7)11 (7.0)20 (12.6)18 (11.4)0.743Education [≥ 12 years, n (%)]605954750.843-Marital status [married, n (%)]26 (34.7)34 (43.0)26 (36.1)35 (36.1)0.386-Weight (kg)91.17 ± 15.2991.13 ± 14.0992.64 ± 12.9494.66 ± 14.140.0890.844Height (cm)165.85 ± 12.15165.03 ± 9.70164.71 ± 16.78167.01 ± 9.660.3560.775BMI (kg/m²)32.52 ± 6.8333.85 ± 4.8833.66 ± 4.4733.94 ± 4.340.0580.420WC (cm)104.79 ± 9.92106.37 ± 10.24107.40 ± 9.06107.65 ± 9.73 0.039 ‡ 0.582 WHR 0.92 ± 0.080.91 ± 0.070.92 ± 0.090.93 ± 0.090.0880.701SMD, sulfur microbial diet; SD, standard deviation; BMI, body mass index; WC, waist circumference; WHR, waist-to-hip ratio. *P values were obtained using one-way ANOVA for continuous variables and the chi-square test for categorical variables. **Adjusted P values were obtained using the general linear model after adjustment for age, sex, physical activity, education, marital status and total energy intake. † Significant difference between forth quartiles and other quartiles according to the Bonferroni post hoc test.‡ Significant difference between first quartile and other quartiles according to the Bonferroni post hoc test. P < 0.05 was considered statistically significant. Bolded values are statistically significant. General demographic and anthropometric characteristics of study participants across quartiles of SMD. SMD, sulfur microbial diet; SD, standard deviation; BMI, body mass index; WC, waist circumference; WHR, waist-to-hip ratio. *P values were obtained using one-way ANOVA for continuous variables and the chi-square test for categorical variables. **Adjusted P values were obtained using the general linear model after adjustment for age, sex, physical activity, education, marital status and total energy intake. † Significant difference between forth quartiles and other quartiles according to the Bonferroni post hoc test.‡ Significant difference between first quartile and other quartiles according to the Bonferroni post hoc test. P < 0.05 was considered statistically significant. Bolded values are statistically significant. Table 2 presents the biochemical characteristics of the participants across quartiles of the sulfur microbial diet (SMD) score. Significant differences across SMD quartiles were observed for TC (P = 0.022) and NGF (P = 0.015). After adjustment for age, sex, physical activity, education, marital status, and total energy intake, significant differences remained for TC (adjusted P = 0.028), NGF (adjusted P = 0.047), and LPS (adjusted P = 0.037). A significant linear trend across SMD quartiles was observed for TC (P for trend = 0.043) and LPS (P for trend = 0.011). No significant differences or linear trends were observed for SBP, DBP, glucose, TG, HDL-C, LDL-C, or insulin. Table 3 presents the dietary energy, macronutrient intake, and selected food group consumption across quartiles of the SMD score. Total energy intake differed significantly across SMD quartiles (P = 0.014). After adjustment for total energy intake, no significant differences were observed in the consumption of grains, nuts, vegetables, fruits, dairy products, or red meat across SMD quartiles (all adjusted P > 0.05). In the trend analysis, a significant linear trend was observed only for fruit intake (P for trend = 0.047), whereas no significant trends were identified for carbohydrate, fat, protein, grains, nuts, vegetables, dairy products, or red meat. Table 2The comparison of biochemical biomarkers of study participants across quartiles of SMD.VariableSMD quartilesP-value*P-value **P- for trend1st quartile (n = 158)2nd quartile (n = 158)3rd quartile (n = 159)4th quartile (n = 158) SBP (mmHg) 122.41 ± 16.02119.66 ± 16.67121.57 ± 14.30118.41 ± 17.350.1950.4650.466 DBP (mmHg) 80.50 ± 11.2179.27 ± 12.2281.51 ± 11.3478.42 ± 12.570.2010.3200.355 Glucose (mg/dL) 95.50 ± 22.3996.87 ± 21.4896.45 ± 27.2096.64 ± 24.870.9600.9710.782 TC (mg/dL) 193.40 ± 43.06190.01 ± 38.82189.02 ± 40.83202.48 ± 48.04 0.022† 0.028 0.043 TG (mg/dL) 146.75 ± 93.32132.63 ± 85.56142.25 ± 83.44128.41 ± 74.210.1900.0820.104 HDL-C (mg/dL) 42.98 ± 10.3542.76 ± 10.7044.25 ± 12.7544.96 ± 10.970.2510.3260.716 LDL-C (mg/dL) 125.41 ± 38.61122.68 ± 36.32119.56 ± 36.34123.02 ± 37.920.5360.4290.328 Insulin (mIU/L) 17.54 ± 27.5816.10 ± 14.3116.15 ± 12.4515.90 ± 16.990.8840.9340.261 NGF (pg/mL) 13.83 ± 1.2113.97 ± 1.1813.82 ± 1.1613.53 ± 1.30 0.015‡ 0.047 0.178 LPS (pg/mL) 23.20 ± 2.2823.15 ± 2.1923.53 ± 2.0723.73 ± 2.100.071 0.037 0.011 SMD, sulfur microbial diet; SD, standard deviation; SBP, systolic blood pressure; DBP, diastolic blood pressure; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; NGF, nerve growth factor; LPS, lipopolysaccharide. *P values were obtained using one-way ANOVA for all continuous variables except triglycerides and insulin, which were analyzed using the Kruskal–Wallis test due to their non-normal distribution. **Adjusted P values were obtained using the general linear model after adjustment for age, sex, physical activity, education, marital status, and total energy intake. *P for trend was obtained using multivariable linear regression across increasing SMD quartiles. †Significant difference between the third and fourth quartiles according to the Bonferroni post hoc test. ‡Significant difference between the second and fourth quartiles according to the Bonferroni post hoc test. P < 0.05 was considered statistically significant. Bolded values are statistically significant. The comparison of biochemical biomarkers of study participants across quartiles of SMD. SMD, sulfur microbial diet; SD, standard deviation; SBP, systolic blood pressure; DBP, diastolic blood pressure; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; NGF, nerve growth factor; LPS, lipopolysaccharide. *P values were obtained using one-way ANOVA for all continuous variables except triglycerides and insulin, which were analyzed using the Kruskal–Wallis test due to their non-normal distribution. **Adjusted P values were obtained using the general linear model after adjustment for age, sex, physical activity, education, marital status, and total energy intake. *P for trend was obtained using multivariable linear regression across increasing SMD quartiles. †Significant difference between the third and fourth quartiles according to the Bonferroni post hoc test. ‡Significant difference between the second and fourth quartiles according to the Bonferroni post hoc test. P < 0.05 was considered statistically significant. Bolded values are statistically significant. Table 3The comparison of dietary energy, macronutrients and some main food groups’ consumption among study participants across quartiles of SMD.VariableSMD quartilesP-value*P-value **P- for trend1st quartile (n = 158)2nd quartile (n = 158)3rd quartile (n = 159)4th quartile (n = 158) Energy (kcal/day) 2938.13 ± 670.112787.22 ± 854.762816.22 ± 760.283134.04 ± 565.63 0.014† –0.850 CHO (% energy) 59.85 ± 18.0756.75 ± 10.1157.50 ± 8.5158.22 ± 6.610.264–0.214 Fat (% energy) 30.66 ± 9.7129.83 ± 7.4630.71 ± 7.1330.08 ± 7.610.870–0.331 Protein (% energy) 16.49 ± 1.5612.99 ± 2.5312.98 ± 2.5112.03 ± 2.790.060–0.274 Grains (g/day) 600.74 ± 232.11516.25 ± 243.07507.17 ± 247.44541.07 ± 244.670.0530.0790.162 Nuts (g/day) 14.28 ± 18.1013.95 ± 15.9919.64 ± 12.8227.10 ± 23.890.3590.5030.146 Vegetables (g/day) 385.72 ± 67.05352.16 ± 87.22330.53 ± 74.34369.87 ± 55.280.2590.3860.297 Fruits (g/day) 647.98 ± 72.20529.02 ± 46.06536.07 ± 58.09565.61 ± 51.940.1510.1070.047 Dairy (g/day) 334.20 ± 50.17333.60 ± 46.53342.42 ± 31.10314.04 ± 34.410.7990.0820.252 Red meat (g/day) 22.12 ± 5.2425.18 ± 8.3722.70 ± 9.0123.44 ± 7.870.7010.4850.971SMD, sulfur microbial diet; SD, standard deviation; CHO, carbohydrate. *P values were obtained using one-way ANOVA. **Adjusted P values were obtained using the general linear model after adjustment for total energy intake. †Significant difference between the first and fourth quartiles according to the Bonferroni post hoc test. P < 0.05 was considered statistically significant. Bolded values are statistically significant. The comparison of dietary energy, macronutrients and some main food groups’ consumption among study participants across quartiles of SMD. SMD, sulfur microbial diet; SD, standard deviation; CHO, carbohydrate. *P values were obtained using one-way ANOVA. **Adjusted P values were obtained using the general linear model after adjustment for total energy intake. †Significant difference between the first and fourth quartiles according to the Bonferroni post hoc test. P < 0.05 was considered statistically significant. Bolded values are statistically significant. Table 4 summarizes the intake of the individual food groups comprising the SMD score across its quartiles. Among the negatively associated components, fruit juice and legume intakes differed significantly across quartiles in both the crude (P = 0.008 and P = 0.016, respectively) and energy-adjusted analyses (adjusted P = 0.006 and P = 0.005, respectively). Significant linear trends were also observed for fruit juice (P for trend = 0.001), legumes (P for trend < 0.001), and sweets and desserts (P for trend = 0.021). Among the positively weighted components, significant linear trends were identified for processed meat (P for trend = 0.044), liquor (P for trend = 0.024), and low-calorie drinks (P for trend < 0.001), whereas no significant differences were observed across quartiles in crude or adjusted analyses. No significant differences or trends were found for beer or other vegetables. Table 4The comparison of SMD diet components intake among study participants across quartiles of SMD.VariableSMD quartilesP-value*P-value **P- for trend1st quartile (n = 158)2nd quartile (n = 158)3rd quartile (n = 159)4th quartile (n = 158)Negative association componentsBeer (g/day)0.31 ± 0.060.38 ± 0.090.51 ± 0.120.50 ± 0.160.7800.1200.11Fruit juice (g/day)38.60 ± 18.3224.08 ± 12.3925.90 ± 6.5916.52 ± 9.200.008†0.0060.001Legumes (g/day)65.43 ± 34.3247.42 ± 36.1548.98 ± 22.2551.06 ± 34.620.016‡0.005< 0.001Other vegetables (g/day)20.23 ± 10.1522.52 ± 15.1518.65 ± 11.8320.46 ± 9.820.4720.6190.999Sweets and desserts (g/day)14.19 ± 6.6714.05 ± 8.6011.66 ± 5.2311.91 ± 5.120.2550.4690.021Positive association componentsProcessed meat (g/day)4.01 ± 7.055.33 ± 9.436.41 ± 13.157.31 ± 16.440.5960.070 0.044 Liquor (g/day)0.85 ± 0.180.64 ± 0.130.52 ± 0.100.40 ± 0.070.0790.098 0.024 Low-calorie drinks (g/day)30.49 ± 5.3346.84 ± 8.8953.48 ± 9.8862.68 ± 11.730.1810.220 < 0.001 SMD, sulfur microbial diet; SD, standard deviation. *P values were obtained using one-way ANOVA. **Adjusted P values were obtained using the general linear model after adjustment for total energy intake. †Significant difference between the first and fourth quartiles with others and with each other according to the Bonferroni post hoc test. ‡ Significant differences between the first quartile with other quartiles according to the Bonferroni post hoc test. P < 0.05 was considered statistically significant. Bolded values are statistically significant. The comparison of SMD diet components intake among study participants across quartiles of SMD. SMD, sulfur microbial diet; SD, standard deviation. *P values were obtained using one-way ANOVA. **Adjusted P values were obtained using the general linear model after adjustment for total energy intake. †Significant difference between the first and fourth quartiles with others and with each other according to the Bonferroni post hoc test. ‡ Significant differences between the first quartile with other quartiles according to the Bonferroni post hoc test. P < 0.05 was considered statistically significant. Bolded values are statistically significant. Table 5 presents the results of the multivariable linear regression analyses. After adjustment for age, sex, physical activity, education, marital status, smoking status, and total energy intake, total cholesterol (β = 0.011, 95% CI: 0.000, 0.021; P = 0.043) and LPS (β = 0.129, 95% CI: 0.024, 0.233; P = 0.016) remained significantly associated with the SMD score. No significant associations were observed for weight, height, BMI, waist circumference, waist-to-hip ratio, SBP, DBP, glucose, triglycerides, HDL-C, LDL-C, insulin, or NGF (all P > 0.05). Table 5Multivariable linear regression analyses examining the association between SMD, anthropometric and biochemical variables.VariableUnstandardized coefficientStandardized β95% CIP-value *βSEWeight (kg)0.0060.0120.075-0.018, 0.0300.604Height (cm)-0.0180.015-0.148-0.047, 0.0120.246BMI (kg/m²)-0.0040.018-0.023-0.039, 0.0310.419WC (cm)0.0120.0150.106-0.018, 0.0430.433WHR-2.6993.434-0.072-9.488, 4.0900.819SBP (mmHg)-0.0040.005-0.056-0.015, 0.0070.466DBP (mmHg)0.0070.0080.070-0.008, 0.0220.355Glucose (mg/dL)-0.0010.002-0.014-0.005, 0.0040.782TC (mg/dL)0.0110.0050.4280.000, 0.021 0.043 TG (mg/dL)-0.0020.001-0.132-0.004, 0.0000.104HDL-C (mg/dL)0.0030.0070.026-0.012, 0.0170.716LDL-C (mg/dL)-0.0100.006-0.354-0.021, 0.0010.072Insulin (mIU/L)-0.0030.003-0.057-0.009, 0.0020.261NGF (pg/mL)0.1300.0960.124-0.059, 0.3190.178LPS (pg/mL)0.1290.0530.2330.024, 0.233 0.016 SE, standard error; CI, confidence interval; SBP, systolic blood pressure; DBP, diastolic blood pressure; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; NGF, nerve growth factor; LPS, lipopolysaccharide. β, unstandardized regression coefficient. ** β coefficients, standard errors, 95% confidence intervals, and P values were obtained from multivariable linear regression models adjusted for age, sex, physical activity, education, marital status, smoking status, and total energy intake. P < 0.05 was considered statistically significant. Bolded values are statistically significant. Multivariable linear regression analyses examining the association between SMD, anthropometric and biochemical variables. SE, standard error; CI, confidence interval; SBP, systolic blood pressure; DBP, diastolic blood pressure; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; NGF, nerve growth factor; LPS, lipopolysaccharide. β, unstandardized regression coefficient. ** β coefficients, standard errors, 95% confidence intervals, and P values were obtained from multivariable linear regression models adjusted for age, sex, physical activity, education, marital status, smoking status, and total energy intake. P < 0.05 was considered statistically significant. Bolded values are statistically significant. Discussion This study investigated the association between the SMD and metabolic, inflammatory, and neurotrophic markers in adults with overweight and obesity. Higher adherence to the SMD was associated with higher total cholesterol concentrations and altered circulating levels of LPS after adjustment for potential confounding factors. Serum NGF concentrations differed across SMD quartiles; however, the observed differences should be interpreted cautiously (Fig. 3). These findings suggest that greater adherence to the SMD is associated with variations in selected circulating metabolic and inflammatory biomarkers. Although previous studies have proposed potential roles for gut microbial pathways in the relationship between sulfur-containing dietary patterns and host metabolism, the present study did not assess gut microbiota composition, microbial metabolites, or intestinal permeability; therefore, these mechanisms remain hypothetical and require confirmation in future studies. Conceptual model illustrating the associations between higher adherence to the sulfur microbial diet (SMD) and key metabolic and neuroimmune parameters. In nutritional epidemiology, dietary exposures are often associated with relatively modest differences in biochemical and neuroimmune biomarkers because diet–health relationships are influenced by multiple interacting factors and long-term cumulative processes22–24. Unlike pharmacological interventions involving single agents at therapeutic doses, dietary patterns generally exert more subtle effects on systemic metabolism and inflammatory pathways over time25,26. Previous cohort studies and dietary intervention trials have reported small variations in metabolic and inflammatory biomarkers in response to dietary patterns or nutrient intake27. However, the clinical and biological relevance of small biomarker differences should be interpreted within the context of the specific biomarker, assay characteristics, and study design. In the present study, the observed differences in circulating biomarkers were modest, and their interpretation should be considered exploratory rather than indicative of established clinical alterations. Further longitudinal studies incorporating repeated measurements, microbiome-related assessments, and detailed biomarker validation are needed to determine whether these changes represent persistent biological adaptations associated with SMD adherence25,28–31. In the present study, higher SMD scores were associated with greater waist circumference in the unadjusted analysis; however, this association was attenuated after adjustment for potential confounders. This finding differs from the results reported by Liu et al.7 who observed significant associations between higher SMD scores and increased odds of general and abdominal obesity in a large cohort study. Several factors may explain these differences, including variations in study population characteristics, sample size, dietary assessment methods, distribution of SMD scores, and the extent of adjustment for lifestyle and metabolic confounders. In addition, differences in background dietary patterns and environmental factors may influence the relationship between sulfur-containing dietary patterns and obesity-related outcomes. Although previous experimental studies have suggested that sulfur-related microbial pathways, including alterations in sulfur metabolism and hydrogen sulfide production, may contribute to metabolic regulation4–6, these mechanisms were not directly evaluated in the present study and should therefore be considered as potential explanations requiring further investigation. We observed a significant association between higher SMD adherence and increased total cholesterol concentrations. Similar findings have been reported in previous studies linking sulfur microbial diet patterns with adverse metabolic profiles3,7. One possible explanation is that some dietary components contributing to higher SMD scores, such as processed meats, may also provide sources of saturated fat and other nutrients that can influence lipid metabolism. In addition, previous experimental studies have suggested that gut microbial pathways, including alterations in bile acid metabolism and inflammatory signaling, may contribute to the regulation of host lipid homeostasis4,11. Caesar et al. demonstrated that gut microbial profiles can influence lipid metabolism through inflammatory pathways, including Toll-like receptor (TLR) signaling11. These proposed pathways may provide a biological context for the observed association between SMD and cholesterol levels. Serum NGF concentrations differed across SMD quartiles, with lower levels observed in the highest SMD quartile compared with the second quartile. NGF is a neurotrophin involved in neuronal survival, energy homeostasis, and appetite regulation13. However, the magnitude of the observed differences was small, and their biological relevance should be interpreted cautiously. Previous studies have suggested that inflammatory processes, including LPS-related signaling pathways, may influence neurotrophin regulation through mechanisms involving Toll-like receptor pathways and neuroinflammatory responses11,14. In the present study, LPS concentrations showed a significant increasing pattern across SMD quartiles after adjustment, although the absolute differences were small. Previous reviews have highlighted potential interactions between gut microbial-derived inflammatory signals and neuroinflammatory pathways, including effects on NGF signaling14. These mechanisms may provide a potential biological context for the observed associations, although further studies directly assessing microbiota-related pathways are needed. No significant differences were observed in fasting glucose, insulin, blood pressure, or triglyceride concentrations across SMD quartiles. The absence of associations with these metabolic markers may reflect the complexity of dietary influences on metabolic regulation and the potential contribution of factors beyond dietary patterns alone. Previous studies have suggested that dietary patterns may have heterogeneous effects across different metabolic pathways, with variations depending on population characteristics, duration of exposure, and individual metabolic susceptibility7,15. In addition, genetic, environmental, and lifestyle factors may modify the relationship between dietary exposures and metabolic outcomes2,6. Our findings extend previous reports by Nguyen et al. and Wang et al., who demonstrated associations between SMD scores, colorectal cancer risk, and inflammatory outcomes3,8. To our knowledge, this is among the first studies to investigate the relationship between SMD and NGF concentrations, while simultaneously evaluating metabolic and inflammatory biomarkers in an overweight/ obese population. The observed associations suggest that higher SMD adherence may be related to differences in selected metabolic and neuroimmune-related biomarkers; however, these findings should be interpreted within the context of the cross-sectional design. Further studies incorporating direct assessments of gut microbial composition, microbial metabolites, and inflammatory pathways are needed to clarify the biological mechanisms underlying these associations. A major strength of this study is its integrative approach, combining dietary pattern assessment with metabolic, inflammatory, and neurotrophic biomarkers in a relatively large and well-characterized sample of obese adults. The use of a validated food frequency questionnaire and standardized biochemical assays strengthens the reliability of dietary and biomarker measurements. In addition, the simultaneous assessment of LPS and NGF provides an exploratory perspective on systemic inflammatory and neurotrophic responses potentially related to dietary patterns. We acknowledge that direct gut microbiome profiling, microbial metabolite assessment, and measurements of intestinal barrier function would provide more comprehensive evidence regarding the biological pathways linking dietary sulfur exposure with host responses. Nevertheless, circulating biomarkers such as LPS and neurotrophic factors have been widely investigated in human studies as peripheral indicators associated with metabolic and inflammatory states. Previous clinical studies have reported relationships between circulating LPS levels, obesity, dietary patterns, and cardiometabolic characteristics32–35, while experimental studies have contributed to the understanding of metabolic endotoxemia and its potential role in metabolic dysfunction10. Similarly, circulating neurotrophic factors, including NGF and BDNF, have been used as peripheral markers of neuroimmune and neurotrophic status in clinical research35. Therefore, our findings should be interpreted as associations with systemic biomarker profiles rather than direct evidence of microbiota-mediated mechanisms. Future studies incorporating microbiome and metabolomic approaches are warranted to further elucidate the pathways underlying these observations. The SMD score used in the present study was calculated based on the original β-weighted algorithm proposed by Nguyen et al.3, in which food-group intakes were weighted according to regression coefficients derived from the association between dietary patterns and sulfur-metabolizing microbial taxa. This approach has subsequently been applied in independent populations, where FFQ-derived food-group intakes were combined with the published β coefficients to generate SMD scores without additional standardization or energy adjustment of dietary variables7,36. Although some later studies have developed modified versions of the SMD using population-specific ranking or scoring approaches8,37, the present study retained the original β-weighted methodology to maintain comparability with the initial microbiome-derived dietary pattern. Nevertheless, because dietary habits, food availability, and intake distributions may differ substantially across populations, we used study-specific quartile categorization to assess relative adherence to the sulfur microbial dietary pattern within our study population. Some limitations of the current study should also be acknowledged. First, the cross-sectional design precludes causal inference, and reverse causality cannot be excluded. Second, dietary intake was self-reported and may be subject to recall bias and measurement error. In addition, the intake of alcoholic components of the SMD score may have been affected by very low consumption levels and potential underreporting, which should be considered when interpreting these components38. Third, although the analyses were adjusted for several important potential confounders, including age, sex, total energy intake, physical activity, education, marital status, and smoking status, residual confounding by unmeasured factors such as genetic background, psychological stress, or other lifestyle characteristics cannot be completely ruled out. Fourth, direct assessment of gut microbiota composition, sulfur-metabolizing bacteria, microbial metabolites, intestinal permeability, and inflammatory mediators was not performed; therefore, the proposed biological mechanisms should be interpreted cautiously. Finally, although serum LPS and NGF provided additional information regarding systemic inflammatory and neurotrophic status, they do not fully capture the complexity of gut–host interactions. Future longitudinal studies incorporating microbiome and metabolomic analyses are needed to clarify the underlying mechanisms and temporal relationships. Conclusion This study provides insights into the relationship between the sulfur microbial diet and metabolic and inflammatory biomarkers in adults with overweight and obesity. Higher SMD adherence was associated with higher total cholesterol concentrations and alterations in circulating biomarkers, including differences in NGF levels and an increasing pattern of LPS concentrations across SMD quartiles. However, the magnitude of these differences was modest, and the findings should be interpreted cautiously within the context of the cross-sectional design. While previous experimental studies have proposed potential links between sulfur-related dietary patterns, microbial pathways, and host metabolic responses, the underlying mechanisms require further investigation. Longitudinal studies incorporating direct assessments of gut microbiota composition and microbial metabolites are warranted to clarify the long-term implications of SMD adherence for metabolic and neuroimmune health. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements I would like to express my sincere gratitude to the faculty of Medicine, Mansoura University for their continuous support, guidance, and mentorship throughout this work. Their commitment to academic excellence and research advancement has been a driving force in the completion of this study. Author contributions In the current work AAMAE was the main researcher, he was involved in data collection, hypothesis generation, conceptualization and data analysis. He was the main supervisor, managed the study’s recruitment and managed data collection. IK, GME, AMS, SMMY and EAE were involved in sample recruitment and data collection, patient management and data entry. TSMA, MAA and SL were involved in hypothesis generation, data validation, conceptualization and visualization. They were also involved in data analysis. IK was also involved in data processing, questionnaires item validation and entry. RA, was the co-supervisor that helped AAMAE with management of the study and data processing, validation, analysis and review of the article. All of the authors were involved in writing the manuscript and final review before submission. Funding Research division of Mansoura University for their financial support. Data availability The datasets of the current study are available from the corresponding author on reasonable request. Declarations Competing interests The authors declare no competing interests. Ethics approval and consent to participate Written informed consent was obtained from all of the participants. The study protocol was approved by the Institutional Review Board of Mansoura University (R.26.02.3600). Also, all methods were performed in accordance with Declaration of Helsinki. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Amr Ali Mohamed Abdelgawwad El-Sehrawy, Email: amralimohamedabdelgawwadelsehr@gmail.com. Rufaida Abdalla, Email: rafiaabdalla@outlook.com. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The datasets of the current study are available from the corresponding author on reasonable request.