Next Article in Journal
Association of Sugar-Sweetened Beverage Consumption and Moderate-to-Vigorous Physical Activity with Childhood and Adolescent Overweight/Obesity: Findings from a Surveillance Project in Jiangsu Province of China
Next Article in Special Issue
Artificial Intelligence Technology for Food Nutrition
Previous Article in Journal
Teaching Kitchens and Culinary Gardens as Integral Components of Healthcare Facilities Providing Whole Person Care: A Commentary
Previous Article in Special Issue
Uncovering Predictive Factors and Interventions for Restoring Microecological Diversity after Antibiotic Disturbance
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Nutrients
Volume 15
Issue 19
10.3390/nu15194163
nutrients-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Weight Loss Promotion in Individuals with Obesity through Gut Microbiota Alterations with a Multiphase Modified Ketogenic Diet

by
Hongchao Wang
1,2,†,
Xinchen Lv
1,2,†,
Sijia Zhao
1,2,
Weiwei Yuan
1,2,
Qunyan Zhou
3,
Faizan Ahmed Sadiq
4,
Jianxin Zhao
1,2,
Wenwei Lu
1,2,* and
Wenjun Wu
5,*
1
State Key Laboratory of Food Science and Technology, Jiangnan University, Wuxi 214122, China
2
School of Food Science and Technology, Jiangnan University, Wuxi 214122, China
3
Department of Nutriology, The Affiliated Wuxi People’s Hospital of Nanjing Medical University, No. 299, Qingyang Road, Wuxi 214023, China
4
Flanders Research Institute for Agriculture, Fisheries and Food (ILVO), Technology & Food Sciences Unit, 9090 Melle, Belgium
5
Department of Endocrinology, The Affiliated Wuxi People’s Hospital of Nanjing Medical University, No. 299, Qingyang Road, Wuxi 214023, China
*
Authors to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Nutrients 2023, 15(19), 4163; https://doi.org/10.3390/nu15194163
Submission received: 15 August 2023 / Revised: 13 September 2023 / Accepted: 19 September 2023 / Published: 27 September 2023
(This article belongs to the Special Issue Combining Artificial Intelligence for Nutrition Applications: Recent Advancements)
Browse Figures
Versions Notes

Abstract

:
The occurrence of obesity and related metabolic disorders is rising, necessitating effective long-term weight management strategies. With growing interest in the potential role of gut microbes due to their association with responses to different weight loss diets, understanding the mechanisms underlying the interactions between diet, gut microbiota, and weight loss remains a challenge. This study aimed to investigate the potential impact of a multiphase dietary protocol, incorporating an improved ketogenic diet (MDP-i-KD), on weight loss and the gut microbiota. Using metagenomic sequencing, we comprehensively analyzed the taxonomic and functional composition of the gut microbiota in 13 participants before and after a 12-week MDP-i-KD intervention. The results revealed a significant reduction in BMI (9.2% weight loss) among obese participants following the MDP-i-KD intervention. Machine learning analysis identified seven key microbial species highly correlated with MDP-i-KD, with Parabacteroides distasonis exhibiting the highest response. Additionally, the co-occurrence network of the gut microbiota in post-weight-loss participants demonstrated a healthier state. Notably, metabolic pathways related to nucleotide biosynthesis, aromatic amino acid synthesis, and starch degradation were enriched in pre-intervention participants and positively correlated with BMI. Furthermore, species associated with obesity, such as Blautia obeum and Ruminococcus torques, played pivotal roles in regulating these metabolic activities. In conclusion, the MDP-i-KD intervention may assist in weight management by modulating the composition and metabolic functions of the gut microbiota. Parabacteroides distasonis, Blautia obeum, and Ruminococcus torques could be key targets for gut microbiota-based obesity interventions.

1. Introduction

Obesity is considered a global health issue that not only affects the quality of life through depression, illness, and disabilities but also reduces life expectancy because it is associated with higher all-cause mortality [1,2]. The last few decades have witnessed an alarming surge in obesity and its associated metabolic disorders. Globally, approximately two billion adults are considered overweight, of which almost half are obese [3,4]. Current approaches to obesity primarily involve lifestyle interventions [5], with dietary interventions being widely regarded as one of the most promising avenues for weight management [6]. While such interventions often yield short-term success, weight regain is almost inevitable [7]. Novel weight management strategies are still in demand.
Current research has demonstrated substantial disparities in gut microbiota composition between obese individuals and those with normal weight and that the gut microbiome plays a crucial role in host energy metabolism [8,9,10]. The human gut microbiota affects the body’s nutrient acquisition, energy expenditure, and several metabolic pathways, owing to a wide spectrum of metabolites, including bile acids, amino acids, short-chain fatty acids, indole derivatives, tryptophan, and trimethylamine N-oxide [11,12,13,14]. Furthermore, alterations in the gut microbiota can modulate the efficacy of dietary interventions, underscoring the potential benefits of fine-tuning the gut microbiota [15,16,17]. However, the extent to which different interventions impact the gut microbiota varies [7], and research investigating the intricate interplay between dietary interventions, obesity, and gut microbiota at the species level and microbial functional composition is still limited. Identifying key responsive species and critical microbial functions in promoting dietary intervention for obesity management remains a valuable avenue for further exploration. In this context, it is worth noting that metagenomic shotgun sequencing, in conjunction with bioinformatics tools, offers an enhanced means to characterize the microbiota [18]. This approach enables a more precise prediction of the biological features of microorganisms and their potential influence on host physiology [19].
In dietary interventions for obesity, low-carbohydrate or very low-carbohydrate ketogenic diets (KDs) have gained popularity in weight management [20]. These diets induce metabolic adaptations, utilizing fatty acids and ketogenic amino acids instead of carbohydrates as the primary energy source, producing ketone bodies as a molecular byproduct and resulting in ketosis [21]. Despite numerous studies demonstrating the efficacy of ketogenic diets in promoting weight loss in humans over the years [22,23,24], their underlying mechanisms are not yet fully understood [25]. Research has indicated that the ketogenic diet also modulates the subjects’ immune responses through the mediation of gut microbiota, as evidenced in a ketogenic diet intervention study involving 17 overweight or grade I obese nondiabetic men, where a significant reduction in the relative abundance of Bifidobacterium was observed following the ketogenic diet intervention [26]. Both in vitro and in vivo experiments showed that ketogenic diet-induced β-hydroxybutyrate selectively inhibited the growth of Bifidobacterium, consequently reducing intestinal proinflammatory Th17 cell levels, demonstrating a causal role of gut microbiota in mediating the host’s immune response to diet [26]. These studies provide valuable insights into the interplay among obesity, ketogenic diets, and gut microbiota, but further research is needed to expand our understanding of how different types of KD affect the gut microbiota.
Therefore, we conducted pre- and post-intervention fecal gut microbiota species-level and functional compositional analyses of the MDP-i-KD, a safer variant of the traditional ketogenic diet. We aimed to identify specific microbial species and metabolic functions closely associated with the weight reduction induced by ketogenic diet interventions. Given the intricate interplay among species within the human gut microbiota, hypothesis testing methods may not be entirely suitable for human gut microbiota research, as they assume interdependence among microbes [27]. Therefore, we employed machine learning approaches, allowing for multivariate and nonlinear analyses, to more accurately identify species crucial for obesity treatment in the MDP-i-KD intervention. Furthermore, we correlated obesity traits with specific gut metagenomic species and microbial functions to elucidate the interactions among obesity, diet, and the human gut microbiota. This study offers insights to support future microbiome-based weight management interventions, thus enabling personalized treatments to better achieve long-term weight management goals for individuals with obesity.

2. Materials and Methods

2.1. Subjects

The study protocol was approved by the hospital ethics committee (KYLKS 201806) and registered with ClinicalTrials.gov (ChiCTR180001523). Thirteen patients with obesity were recruited from The Affiliated Wuxi People’s Hospital of Nanjing Medical University. Inclusion criteria comprised: (1) age range of 18–65 years, (2) BMI ≥ 28 kg/m2, and (3) stable body weight over the preceding three months. All participants had signed an informed consent form. Detailed information regarding the study design, exclusion criteria, and dietary specifics has been previously reported [28].
All participants began with a 4-week hypocaloric balanced diet (HBD) as an introductory period, followed by a 12-week MDP-i-KD intervention consisting of two cycles of the same “2 + 2 + 2” dietary regimens (Supplementary Material Figure S1): two weeks of KD, followed by two weeks of a carbohydrate-based transition diet (TD), and finally two weeks of an HBD. The nutrient limits observed at each stage are listed in Supplementary Material Table S1. In addition, the registered dietitian developed a range of food choices for each diet phase for the subjects’ reference (Supplementary Material Tables S1 and S2). During the KD phase, morning urine ketone results for all participants showed that they had entered a state of ketosis.
Each subject also performed aerobic and resistance exercises. This component was described in detail in our previous study [28]. During the experiment, the dietitian monitored the patients’ dietary information, weight data, and the presence of adverse effects in the form of pictures and text reports.

2.2. Anthropometric Assessment and Blood Chemistry

Before and after the MDP-i-KD intervention, anthropometric assessment and blood biochemical indicator detection were carried out for obese subjects. The detailed detection indicators and details are clarified in the research of Yuan et al. [28].

2.3. Sample Collection

All subjects were instructed to collect fecal samples using sterile collection tubes and to collect stools once before the start of the KD intervention and once after the end of the MDP-i-KD, respectively. Collected stools were mixed with 15% glycerol and frozen at −80 °C and were used for metagenomic analysis.

2.4. Metagenomic Data Processing and Quality Control

Metagenome sequencing was performed on the Illumina NovaSeq 6000 platform (Illumina Inc., San Diego, CA, USA) of Shanghai Meiji Biomedical Technology Co., Ltd. (Shanghai, China) The 26 samples submitted for examination were stool samples of 13 subjects who received the MDP-i-KD at 0 and 12 weeks of intervention. The average sequencing amount of the 26 samples was 49.2 ± 4.4 (mean ± standard deviation) million reads. The preprocessing of the original sequence includes the following procedures: Trimmomatic (version 0.39) was employed to filter low-quality sequences [29]. Sequences with an average base quality score below 30 were trimmed, and sequences longer than 60 bp after filtering were retained as a quality-controlled output. Subsequently, filtered sequences were aligned to the human reference genome (Homo sapiens genome assembly GRCh38, hg38) using BWA (version 0.7.17), Samtools (version 1.9), and BEDTools (version 2.30.0), effectively removing host-origin genes from the samples [30,31,32]. After the above quality control, an average of 37.1 ± 3.8 (mean ± standard deviation) million reads were retained per sample.

2.5. Analysis of Gut Microbiota Species and Functions

The high-quality sequences post-quality control were subjected to taxonomic and functional annotation using MetaPhlAn3 and HUMAnN3, respectively [33]. Notably, HUMAnN3 utilized merged paired-end data for annotation and subsequently normalized the obtained counts to relative abundance values. The stratified information for individual bacterial functional contributions was derived from the normalized abundance data of metabolic pathways. Visualization of bacterial contributions to individual metabolic pathways was achieved using the humann_barplot script. Species abundance information tables were used to calculate alpha diversity metrics. Beta diversity was calculated using the Bray–Curtis metric. Linear discriminant analysis effect size (LEfSE) was used to identify bacterial taxa driving differences before and after the MDP-i-KD [34]. For the construction of the bacterial co-occurrence network, species with a relative abundance of less than 0.01% were excluded. Spearman correlation analysis was performed on bacterial correlations before and after the MDP-i-KD. Only strong correlations (r > |0.6| and p < 0.01) were considered and visualized by Gephi 0.9.2.

2.6. Machine Learning Analysis

Metagenomic species’ ability to classify two states before and after the MDP-i-KD intervention was assessed using machine learning methods. The area under the ROC curve (AUC) served as the model evaluation metric. Feature selection was performed on the original dataset using seven methods, including univariate ANOVA (f_classif, FC), mutual information estimation, and model-based approaches like Random Forest (RF), Gradient Boosting (GB), XGBoost (XGB), LightGBM (LGB), and L1 Regularization (L1) [35]. Model-based feature selection retained features with importance scores greater than 0.001.
The obtained feature subsets from each selection method were applied to individual classifiers. To mitigate random errors, we conducted five-fold cross-validation ten times and utilized the area under the curve (AUC) score as the evaluation metric to select the feature subset and corresponding model with the best classification performance. Finally, the SHapley Additive exPlanations (SHAP) package (0.42.1) was employed to calculate the SHAP value of each feature within the subset [36], revealing their importance and impact on classification.

2.7. Association Analysis of Gut Microbiota and Physiological Indicators

Correlation analysis between species was conducted and we identified physiological indicators of obese subjects before and after intervention. The relative abundance of bacteria species with a mean relative abundance greater than 0.1% in 20% of the samples before and after the multi-stage ketogenic diet intervention was compared with the weight, BMI, waist circumference, hip circumference, waist-to-hip ratio, body fat, body fat percentage, and internal organs of obese subjects. Spearman correlation analysis was performed on the fat area. The correlation coefficient was −0.35 < r < 0.35, p < 0.05.

2.8. Differential Analysis of Metabolic Pathways of Gut Microbiota

To identify the key gut microbiota metabolic pathways that exhibited significant changes before and after the MDP-i-KD intervention, the Wilcoxon rank-sum test was performed using the Wilcox.test function of the R language to analyze the metabolism of the gut microbiota with significant differences before and after the intervention pathway, and we calculated the Log2 Foldchange of the metabolic pathway before and after the intervention.

2.9. Correlation Analysis between Metabolic Pathways of Gut Microbiota and Physiological Indicators

In order to explore the metabolic pathways that were significantly changed before and after the MDP-i-KD, an association analysis between the metabolic pathways and the physiological indicators of obese subjects was carried out. Spearman correlation analysis was performed between the metabolic pathways existing in 20% of the samples before and after the intervention and the body weight, BMI, waist circumference, hip circumference, and waist-to-hip ratio of obese subjects, and the correlation coefficient was −0.4 < R < 0.4, p < 0.05 correlation.

2.10. Statistical Analysis

A statistical analysis of clinical physiological indicators was applied to measure the effect of weight loss in the MDP-i-KD and was analyzed using the paired Wilcoxon rank-sum test. Values are expressed as mean ± SD or n (%). Spearman correlations were calculated using the corr.test function (a function of the psych package in R). The Wilcoxon rank-sum test was performed using the Wilcox.test function of the R statistical package to analyze the metabolic pathways with significant differences between the gut microbiota before and after the intervention. When p < 0.05, the results were statistically significant.

3. Results

3.1. MDP-i-KD Changes Biochemical Measurements and Anthropometric Characteristics of Obese Subjects

Before the intervention with the ketogenic diet, the average BMI of 13 subjects (62% male and 38% female) was 31.0 ± 2.6 kg/m2, and the average body weight was 86.6 ± 14.7 kg. After 12 weeks of treatment, there was a significant weight loss of 8.2 ± 2.5 kg, and the average body weight of the subjects decreased by 9.4% when compared to that before the intervention. Notably, BMI (p = 0.0002), waist (p = 0.0017), hip (p = 0.0016), systolic blood pressure (p = 0.0118), blood pressure (p = 0.0291), AST (p = 0.0357), ALT (p = 0.0134), Triglyceride (p = 0.0.0081) and HbA1c (p = 0.0058) all showed significant decreases after the MDP-i-KD. In addition, we observed that the total cholesterol concentration did not significantly change. Before and after the intervention, the LDL cholesterol concentration did not change significantly, but the HDL cholesterol concentration increased (Table 1). The MDP-i-KD intervention did not significantly affect BUN, creatinine, albumin, UA and ALP levels (Table 1).

3.2. Effect of the MDP-i-KD on Gut Microbiota

Based on the results of observed OTUs (Figure 1A), Pielou’s evenness (Figure 1B), the Shannon index (Figure 1C), and the Simpson index (Figure 1D), an assessment was made regarding the impact of the intervention on the species richness and diversity of the gut microbiota. The results indicated that none of the four diversity indices exhibited significant changes. Specifically, observed OTUs demonstrated a slight increase after the intervention, whereas the Pielou, Shannon, and Simpson indices all displayed a decreasing trend post-intervention. The increase in observed OTUs indicates a certain expansion in the variety of species within the gut microbiota after the intervention. However, the declines in the Pielou, Shannon, and Simpson indices signify reduced evenness in species distribution and decreased overall diversity in the gut microbiota. These findings suggest a potential dominance shift among certain microbial species post-intervention, resulting in heightened unevenness and decreased overall gut microbiota diversity. In addition, the beta diversity calculated by the Bray–Curtis distance did not have significant changes before and after the intervention (Figure 1E).
LEfSe analysis was used to identify six species that were significantly enriched post-intervention (D12 group): Bacteroides nordii, Streptococcus infantarius, Lactococcus petauri, Parabacteroides distasonis, Bacteroides ovatus, and Klebsiella variicola. A significant reduction in Actinomyces odontolyticus was observed after the intervention (Figure 1F). The above results suggest that the MDP-i-KD intervention was able to alter the gut microbiota of the participants.

3.3. Machine Learning Identifies Key Microbial Changes before and after MDP-i-KD Intervention

Machine learning analysis was conducted on metagenomic species-level data to investigate the main microbial features affected by the MDP-i-KD intervention. Seven feature selection methods were used to filter potentially redundant features for each model to reduce data dimensionality and improve classification performance. The GB and L1 methods effectively reduced features without compromising model classification performance. Specifically, the GB model identified a small subset of seven key features, including Parabacteroides distasonis, Weissella cibaria, Eisenbergiella tayi, Bacteroides vulgatus, Oscillibacter sp_57_20, Bacteroides thetaiotaomicron, and the Enterobacter cloacae complex. This subset achieved the best classification performance, with an AUC of 0.838 in GB (Figure 2A).
Based on the GB classification model, we ranked the relative importance of the SHAP values for the seven feature species. P. distasonis was identified as the most crucial species in distinguishing the pre- and post-intervention states of the MDP-i-KD. The Wilcox test confirmed a significant increase in its relative abundance after the intervention (Supplementary Material Figure S2A). Other feature species selected through the GB model did not exhibit significant changes in relative abundance before and after the intervention, indicating a low perturbation of effect values, determined by changes in their relative abundance (Supplementary Material Figure S2B–G).
Among the above characteristics, the relative abundance of P. distasonis, the most critical microbial feature, increased significantly after the intervention. Previous studies have identified a decrease in the abundance of P. distasonis in obese and metabolic syndrome populations [37,38]. Additionally, other studies of dietary interventions for treating obesity have shown an upward trend in the relative abundance of P. distasonis after the intervention. For example, in clinical trials with Mediterranean or low-fat, high-complex-carbohydrate diets for obesity treatment, the relative abundance of P. distasonis increased significantly one year after consumption [39].
Furthermore, our findings suggest an upward trend of B. vulgatus and B. thetaiotaomicron after the intervention. Recent studies have suggested a cross-feeding between B. vulgatus and Akkermansia muciniphila, positively influencing obesity reduction [40]. Additionally, other studies found a significant decrease in the abundance of B. thetaiotaomicron in individuals with obesity, and gavage with B. thetaiotaomicron protected mice from obesity [41]. These results underscore the potential importance of B. vulgatus and B. thetaiotaomicron in regulating obesity during the MDP-i-KD.

3.4. Species-Level Co-Occurrence Network Changes after the Intervention

To evaluate the response of microbial interactions to the MDP-i-KD, we constructed co-occurrence networks filtered using the following criteria: strong correlation (|r| > 0.6) and p-value < 0.01. Prior to the intervention, a total of 722 positive and 38 significant negative correlations were observed between the 299 species of participants (Table 2). Post-intervention, on the other hand, 885 positive and 16 significant negative correlations were observed between 318 species (Table 2).
We calculated the topological features of each node in the networks (Table 2). When comparisons were made between the two networks, it was found that the average clustering coefficients in the species-level co-occurrence networks increased from 0.905 to 0.920 due to the MDP-i-KD, which indicates a transition of species interactions to a more complex microbial network after the intervention. Using the modularity algorithm in the Gephi software, we clustered the closely linked nodes, and the nodes in different categories were marked differently and shown in different colors. Three major subclusters with a node count greater than ten were found in the D0 network, whereas six major subclusters with a node count greater than ten were found in the D12 network (Figure 3). While the number of modularizations decreased from 141 pre-intervention to 138 post-intervention, there was an increase in the number of subclusters containing more than 10 nodes and an overall expansion in the number of nodes within these subclusters post-intervention. These results suggest that the interactions at the species level of the gut microbiota within each subcluster are more complex and concentrated after the MDP-i-KD intervention.
In Figure 3, dashed boundaries highlight alterations in interactions among obesity-related species pre-and post-intervention. In module A, interactions involving Blautia obeum were lost after the intervention. Module B illustrates a negative correlation between Prevotella copri and Ruminococcus gnavus—the dominant species in the corresponding module post-intervention. In module C, Blautia producta, which had close associations with other species at D0, displayed disrupted interactions by D12. Conversely, two feature species, B. thetaiotaomicron and E. tayi, identified through feature selection in modules D and F, re-established their interrelationships with other species post-intervention. B. thetaiotaomicron exhibited a negative correlation with other anamorphic bacilli, including Bacteroides ovatus, Bacteroides uniformis, and Baeteroides nordii. Meanwhile, E. tayi showed a positive correlation with the weight loss-related species A. muciniphila. These findings suggest that following the MDP-i-KD intervention, subjects’ gut microbiota symbiotic network shifted toward a healthier and more intricate state.

3.5. MDP-i-KD Alters the Correlation between Intestinal Microbiota Abundance and Physiological Indicators

Spearman correlation analysis was performed between body weight, BMI, waist circumference, hip circumference, waist-to-hip ratio, body fat, body fat percentage, visceral fat area of obese subjects before and after the MDP-i-KD, and gut bacteria obtained by metagenomic sequencing. The results showed that Bacteroides intestinalis, Paraprevotella xylaniphila, Alistipes putredinis, Parabacteroides goldsteinii, Clostridium disporicum, Eubacterium ramulus, Blautia obeum, Ruminococcus torques, Coprococcus eutactus, Dorea longicatena, Fusicatenibacter saccharivorans, R. bromii, and Oxalobacter formigenes were significantly and positively associated with BMI. P. xylaniphila, A. putredinis, C. disporicum, E. ramulus, B. obeum, R. torques, C. eutactus, D. longicatena, F. saccharivorans, and O. formigenes were also significantly positively correlated with body weight, waist circumference, hip circumference, and other indicators (Figure 4). The results showed that changes in gut microbiota were closely related to the phenotypic symptoms observed in patients with obesity.

3.6. The MDP-i-KD Changes the Gut Microbiota Profile of the Obese Subjects

The results of the differential analysis of metabolic pathways of gut microbiota showed that before and after the multi-stage ketogenic diet intervention, the following metabolic pathways were detected: secondary metabolite degradation, amino acid biosynthesis, carbohydrate degradation, nucleotide synthesis, and aromatic compound biosynthesis of the gut microbiota in obese subjects. The results showed that ten metabolic pathways were significantly different before and after the intervention (Figure 5). The relative abundances of selenium-based amino acid biosynthesis (PWY-6936), lysine biosynthesis II (PWY-2941), the superpathway of 5-aminoimidazole ribonucleotide biosynthesis (PWY-6277), 5-aminoimidazole ribonucleotide biosynthesis II (PWY-6122), 5-aminoimidazole ribonucleotide biosynthesis I (PWY-6121), starch degradation V (PWY-6737), chorismate biosynthesis I (ARO-PWY), chorismate from 3-dehydroquinine (PWY-6163), and superchannels for aromatic amino acid biosynthesis (COMPLETE-ARO-PWY) were significantly reduced in the intestinal microbiota of post-intervention obese subjects. The relative abundance of the mannitol cycle in the gut microbiota was significantly enhanced following the intervention. These results suggest that multi-stage KD intervention alters the metabolic pathways of the gut microbiota in individuals with obesity.

3.7. The MDP-i-KD Alters Metabolic Pathways and Correlates with Physiological Indicators Studied

Body weight, BMI, hip circumference, waist circumference, and waist-to-hip ratio of obese subjects before and after the MDP-i-KD intervention were analyzed using Spearman correlation analysis with metabolic pathways of the gut microbiota obtained by metagenomic sequencing. The results showed that the metabolic pathway of the gut microbiome could be analyzed using the Spearman test. The changes in the subjects were closely related to their obese phenotypic symptoms (Figure 6). Including amino acid biosynthesis, lipid biosynthesis, cofactor biosynthesis, nucleotide biosynthesis, carbohydrate degradation, polyamine synthesis, aromatic compound degradation, aromatic compound biosynthesis, carbohydrate synthesis, cellular structure biosynthesis, and fermentation, a total of 93 metabolic pathways in secondary metabolite biosynthesis, energy metabolism, and other pathways were significantly correlated with obesity epigenetic indicators. Among them, 5-aminoimidazole ribonucleotide biosynthesis I (PWY-6121), 5-aminoimidazole ribonucleotide biosynthesis II (PWY-6122), 5-aminoimidazole ribonucleotide biosynthesis super pathway (PWY-6277), synthesis of chorismate from 3-dehydroquinine (PWY-6163), starch degradation V (PWY-6737), superchannel for aromatic amino acid biosynthesis (COMPLETE-ARO-PWY), and chorismate biosynthesis Synthetic I (ARO-PWY) were not only significantly positively correlated with BMI, but their relative abundance was significantly reduced after the intervention. Therefore, these seven metabolic pathways may play a vital role in reducing obesity after the MDP-i-KD intervention.

3.8. Species Contribution of the Metabolic Pathways of the Gut Microbiota

3.8.1. Analysis of Species Contribution to Key Metabolic Pathways

An analysis of the metabolic pathways significantly positively correlated with BMI, with significant differences after the MDP-i-KD, including 5-aminoimidazole ribonucleotide biosynthesis I (PWY-6121), 5-aminoimidazole ribonucleotide biosynthesis I (PWY-6121), 5-aminoimidazole ribonucleotide biosynthesis II (PWY-6122), Super pathway for 5-aminoimidazole ribonucleotide biosynthesis (PWY-6277), synthesis of chorismate from 3-dehydroquinine (PWY-6163), starch degradation V (PWY-6737), superchannel for aromatic amino acid biosynthesis (COMPLETE-ARO-PWY), and chorismate biosynthesis I (ARO-PWY), was performed. Bacteria that have been reported to play an important role in these metabolic pathways were screened by calculating the contribution of species to the metabolic pathways, and the results are shown in Figure 7. R. torques makes a high contribution to the PWY-6121, PWY-6122, PWY-6277, ARO-PWY, PWY-6163, COMPLETE-ARO-PWY, and PWY-6737 metabolic pathways. After the MDP-i-KD and weight loss, the relative contribution of R. torques to key metabolic pathways was reduced, and the relative abundances of the seven key metabolic pathways were significantly reduced. Therefore, R. torques was significantly positively correlated with BMI, possibly by altering the key metabolic pathways in individuals with obesity. Eubacterium hallii contributed to a higher degree to the PWY-6163 metabolic pathway. After the intervention, the relative abundance of E. hallii, its relative contribution to PWY-6163, and the relative abundance of PWY-6163 significantly decreased. B. obeum has a high contribution to the ARO-PWY, PWY-6163, and COMPLETE-ARO-PWY metabolic pathways. The relative abundance of Blautia is highly correlated with obesity and thus diminishes with the reduction in obesity following the MDP-i-KD. B. obeum was significantly positively correlated with multiple obesity phenotypic indicators. After the MDP-i-KD and weight loss, the relative contributions of B. obeum to ARO-PWY, PWY-6163, and COMPLETE-ARO-PWY metabolic pathways decreased. We speculate that the MDP-i-KD can improve obesity symptoms by altering the relative abundance and metabolic function of B. obeum and E. hallii.

3.8.2. B. obeum and R. torques Are Involved in the Regulation of Various Metabolic Activities in the Gut of Subjects with Obesity

A comparative analysis of the metabolic pathways with a relatively high contribution of B. obeum before and after the intervention revealed that B. obeum had a high relative contribution in ARO-PWY, PWY-6163, COMPLETE-ARO-PWY, and 21 other metabolic pathways (Supplementary Material Table S3). Among them, B. obeum can degrade formaldehyde assimilation II (hip cycle) (PWY-1861), guanosine nucleotide degradation superchannel (PWY-6595), and guanosine nucleotide degradation II (PWY-6606), which had the highest relative contribution. The analysis of the contributing pathways of R. torques found that in addition to the 7 key metabolic pathways, the relative contribution of bacterial species was higher, and the other 56 metabolic pathways had higher relative contributions (Supplementary Material Table S3). The above results show that B. obeum and R. torques or their metabolites can participate in a variety of metabolic activities in the guts of obese and normal-weight individuals.

4. Discussion

A person’s diet is considered a major determinant of the diversity, ecology, and functionality of their gut microbiota, with implications for health. Adaptation to different dietary patterns dramatically affects the onset and continuation of obesity. Although a ketogenic diet has been introduced as an effective means of weight loss, the mechanisms underlying its ameliorative effects have not yet been fully explored. Previous studies have demonstrated the potential of the ketogenic diet (KD) to modify the diversity and species composition of the gut microbiota in individuals with various health conditions, including obesity [26], type 2 diabetes [42], cancer [43], epilepsy [44,45,46], and cognitive impairment [47,48], and Alzheimer’s disease [49], as well as healthy individuals [50]. However, the majority of these investigations have predominantly centered on genus-level discussions of microbial composition, often overlooking the examination of the correlation between individual microbial species and their functional alterations in response to obesity and dietary patterns. Additionally, different types of ketogenic diets appear to exert varying effects on the gut microbiota species among subjects, highlighting the need for tailored analyses in specific contexts. In this trial, the effect of an MDP-i-KD intervention on the intestinal flora of subjects was analyzed using shotgun metagenomic sequencing. We observed significant changes in the gut microbiota composition, bacterial interaction networks, and metabolic function as a result of the multiphase dietary regimen; therefore, certain species related to weight loss were identified. Our study provides in-depth information on the effect of dietary interventions on the intestinal microbiota of individuals with obesity. We investigated not only the fecal microbiota of subjects before and after the intervention but also assessed whether any gut microbiota changes were associated with body weight loss. Therefore, our findings help clarify the interplay between obesity, diet, and the gut microbiota in human subjects.
Recent meta-analyses have shown that ketogenic diets can serve as a helpful tool for treating obesity by reducing body weight, waist circumference, and body mass index (BMI) [23]. Nonetheless, researchers recommend that a KD should be followed with caution, as its treatment of obesity may be accompanied by adverse conditions, such as dehydration, hypoproteinemia, poor adherence, and hair loss [51]. In contrast, a multi-stage dietary therapy program that includes the gradual reintroduction of carbohydrates is an efficient and safe strategy to combat obesity [52]. Therefore, the MDP-i-KD was developed in order to counteract the nutritional deficiency and low compliance of people with obesity, which may be due to KD-only interventions. It was essential to determine which species-specific alterations were caused by the MDP-i-KD intervention that led to reduced body weight. We performed shotgun metagenomic sequencing of treated feces before and after the intervention. Our results demonstrate a reduction in the alpha diversity of the gut microbiota in the intervention group, which may be attributed to the fact that carbohydrates serve as the fundamental substrate for microbial energy generation through decomposition [21]. A lower carbohydrate content in the KD could lead to an overall decrease in microbial diversity [53]. Machine learning methods identified seven crucial species associated with MDP-i-KD intervention: P. distasonis, W. cibaria, E. tayi, B. vulgatus, O. sp_57_20, B. thetaiotaomicron, and the E. cloacae complex. Among them, P. distasonis played a dominant role in distinguishing between pre-intervention and post-intervention states. Lefse analyses further confirmed a significant enrichment of P. distasonis post-intervention. A previous study in Mexico found that obese children and those suffering from metabolic syndrome had a reduced abundance of P. distasonis in their gut when compared to healthy children [37]. Similar trends were observed in a three-week intermittent fasting intervention for obesity. P. distasonis and B. thetaiotaomicron were significantly enriched after the intervention and negatively correlated with obesity-related parameters [54]. Therefore, the above findings also support the hypothesis that the decreased abundance of P. distasonis observed in our study may be related to metabolic changes related to weight loss after the intervention. Wang et al. demonstrated that live P. distasonis treatment attenuated weight gain in ob/ob and HFD-fed mice. This effect might be attributed to P. distasonis’s ability to convert primary bile acids into secondary bile acids (LCA and UDCA) and produce succinic acid [55]. Therefore, the above findings also support the hypothesis that the decreased abundance of P. distasonis observed in our study may be related to metabolic changes linked to weight loss after the intervention. Additionally, although B. vulgatus and B. thetaiotaomicron showed only an upward trend after the intervention without significant differences, they were still identified as important traits by machine learning methods, considering the complex interactions between microorganisms. B. vulgatus has been shown to have a cross-feeding relationship with the weight loss-associated species A. muciniphila [40,56,57]. Moreover, B. vulgatus also exhibits a cross-feeding relationship with B. ovatus, which we found to be significantly enriched post-intervention in the Lefse analysis [58]. As for B. ovatus, the trend of a childhood obesity study in Malaysia is consistent with our results, with a higher relative abundance of B. ovatus in the gut of normal-weight children than in obese children [59]. Moreover, B. thetaiotaomicron exhibited consistent results, with reduced abundance in individuals with obesity [41] and increased abundance in dietary interventions for weight loss [54]. We found that the MDP-i-KD intervention can effectively reduce A. odontolyticus, which causes meningitis and cervical abscess [60], indicating that the MDP-i-KD can reduce the relative abundance of harmful bacteria in the gut. These findings assure that strategies targeting the gut microbiota to treat obesity are effective. Although these differential species’ physiological function or biological behavior is not yet clear, these signatures are still worthy of attention as fine-grained gut microbiome features.
The gut microbiota is one of the most complex microbial ecosystems known in which a plethora of microbial interactions occur. Currently, most of the knowledge on the role of diet in ameliorating several metabolic disorders is limited to studying changes in the abundance of different microbial groups rather than focusing on microbial interspecific interactions [61]. Co-occurrence networks were constructed in order to search for interactions between species using metagenomic sequencing. We found significant changes in the intestinal flora co-occurrence network before and after the intervention. The co-occurrence network at the species level appeared to become more concise after the intervention, showing that the number of subclusters decreased with weight loss, which is consistent with our previous analysis of 16S rRNA. However, it is worth noting that the number of nodes in the remaining four main modules increased after the MDP-i-KD intervention, suggesting that the species-level interactions of the microbiota within each module are more complex and concentrated. Considering this, in this study, the interaction network at the microbial-species level was analyzed by metagenomic analysis, as when compared with the genus level obtained by 16S rRNA analysis, the degree of interaction between bacterial species must be more complex. In addition, pathogenic bacteria, such as pro-inflammatory and even cancer-promoting Streptococcus infantis, disappeared in the co-occurrence network after the intervention, suggesting they were less abundant and interacted with other species. In the D12 network, a reciprocal inhibition relationship was observed between P. copri and R. gnavus. Recent studies have indicated a significant enrichment of R. gnavus in insulin-resistant and obese subjects, with a notable association with low cognitive traits [62]. On the other hand, P. copri has been implicated in triggering host inflammatory responses and exhibits a significant correlation with adipose accumulation in pigs [63]. Several population cohort studies have indicated that B. obeum is associated with obesity [41]. The detailed parameters of the D12 network map show that the interaction of B. obeum with other species is lower than that on D0, which seems to further explain that the influence of B. obeum in the intestine is also reduced with weight loss. Although the interaction between microbiota species remains unclear, further studies are still needed. Overall, the above phenomena suggest that the gut flora interactions of obese participants are reorganized into a healthier state after weight loss. Our results indicate that microbial interactions in KD–obesity interactions should be considered in addition to changes in microbial abundance levels. Through co-occurrence network analysis, we have identified the impact of KD interventions on the interactions between several species that are closely associated with obesity. These species have the potential to be targeted for regulating gut microecology in obesity.
Changes in the gut microbiota are strongly associated with phenotypic symptoms in people with obesity. Previously, using AGP big data analysis, dietary interventions affecting genus-level gut microbiota, and other results, Blautia was consistently screened as a genus highly correlated with obesity of the gut microbiota in people with obesity. Correlation analysis between obesity phenotype indicators and intestinal bacteria showed that Blautia obeum was positively correlated with BMI, waist circumference, body weight, body fat, hip circumference, waist-to-hip ratio, and visceral fat area. B. hydrogenotorophica was significantly positively correlated with the waist-to-hip ratio. Furthermore, our findings are consistent with those of published studies. Liu et al. conducted a metagenomic association analysis and serum metabolomic analysis of obese and non-obese Chinese populations and identified a significant positive correlation between B. obeum and BMI [41]. Kasai et al. studied the species composition of the gut microbiota of obese and non-obese Japanese people and found that the relative abundance of B. hydrogenotorophica and B. obeum was higher in the intestines of individuals with obesity [64]. Therefore, B. obeum is a key bacterial species in the gut of individuals with obesity and may serve as a target for probiotics or dietary interventions for obesity. In addition, the results of the metagenomic analysis before and after intervention showed that R. bromii, R. torques, D. longicatena, and E. ramulus were significantly positively correlated with BMI. D. longicatena was significantly increased in the gut of the obese Chinese population in previous studies, which is consistent with our findings. R. torques was also confirmed to be enriched in the gut of individuals with obesity. Rosés et al. studied the effect of the Mediterranean diet on the gut microbiota of normal-weight, overweight, and obese individuals, and the results showed that the Mediterranean diet was associated with an increase in the production of butyrate and abundance of R. bromii [65].
We analyzed the differences seen in the metabolic pathways of the intestinal microbiota before and after the intervention and combined the correlation analysis between obesity-related physiological indicators and metabolic pathways to explain the mechanisms underlying weight loss. We observed that ten metabolic pathways were significantly altered in the intestinal microbiota of obese participants, including amino acid biosynthesis, nucleotide synthesis, secondary metabolite degradation, aromatic compound biosynthesis, and carbohydrate degradation. Among differential metabolic pathways, the relative abundance of the mannitol cycle was significantly increased after the intervention, suggesting that the ketogenic diet can promote its synthesis and metabolism. Mannitol, a known diuretic that rapidly excretes water from tissues, has not been shown to correlate with any obesity-related physiological markers. Therefore, the reasons for weight loss remain to be further explored. The graph (heatmap) shows that all nucleotide biosynthesis pathways were significantly positively correlated with BMI. Following KD intervention, a substantial decrease was observed in the relative abundance of 5-aminoimidazole ribonucleotide biosynthesis pathways I (PWY-6121) and II (PWY-6122) and the super pathway (PWY-6277). Moreover, 5-Aminoimidazole ribonucleotide is a critical intermediary in the biosynthesis of purine nucleotides, and the attenuation of its biosynthesis potentially translates into diminished purine production, thereby mitigating the accumulation of uric acid. Notably, serum uric acid levels have been positively correlated with BMI across various studies [66,67,68]. Research has observed that umami induces obesity and metabolic syndrome through the purine nucleotide degradation pathway [69]. Intriguingly, specific purine metabolites such as inosine, hypoxanthine, and uric acid have been implicated in heightened caloric intake and weight gain, even in the absence of monosodium glutamate intervention [69]. In addition, 3-dehydroquinine synthesis chorismate (PWY-6163), starch degradation V (PWY-6737), aromatic amino acid biosynthesis superchannel (COMPLETE-ARO-PWY), and chorismate biosynthesis I (ARO-PWY) were consistent with the trend in the nucleotide biosynthesis pathway, which was not only positively correlated with BMI but also significantly decreased in relative abundance after the intervention. These findings are consistent with the well-established microbial functional changes associated with obesity [13]. In a metagenomic study conducted on an obese population, it was observed that the gut microbiota in obese individuals may demonstrate elevated biosynthesis of aromatic amino acids and branched-chain amino acids, along with an enhanced capacity for carbohydrate utilization in comparison to lean controls [41]. The serum metabolomic profiles of pregnant women who had undergone malabsorptive bariatric surgery indicated reduced circulating levels of branched-chain and aromatic amino acids [70]. Additionally, a three-month supplementation with live or pasteurized A. muciniphila led to weight loss and the down-regulation of serum metabolomic pathways associated with tyrosine, phenylalanine, and tryptophan metabolism in obese subjects [71]. These findings suggest that the MDP-i-KD intervention may effectively diminish the relative prevalence of purine synthesis pathways, the aromatic amino acid synthesis pathway, and the starch degradation pathway within the gut microbiota, thereby curtailing the generation of purine compounds, aromatic amino acids, and their metabolites. This reduction is presumed to exert a positive influence on ameliorating obesity and metabolic syndrome. Nonetheless, comprehensive investigations are warranted to elucidate the precise mechanisms and biological ramifications of this association. It is worth noting that the contribution of these seven metabolic pathways of R. torques is higher, and after the multi-stage ketogenic diet intervention and weight loss, the relative contribution of R. torques to the seven key metabolic pathways was reduced. Therefore, R. torques was significantly positively correlated with BMI, possibly by altering the key metabolic pathways in individuals with obesity. B. obeum has a high contribution to the ARO-PWY, PWY-6163, and COMPLETE-ARO-PWY metabolic pathways, and after multi-stage ketogenic diet intervention and weight loss, the relative contribution of B. obeum to these three metabolic pathways was reduced. Blautia is a genus that is significantly associated with obesity and is significantly positively correlated with multiple obesity phenotypic indicators. Therefore, we speculated that MDP-i-KD can improve obesity symptoms by altering the relative abundance and metabolic functions of B. obeum and R. torques. As key bacteria, B. obeum and R. torques contributed to the metabolic pathway analysis results before and after the intervention, showing that they or their metabolites could participate in various metabolic activities in the intestines of obese and normal-weight individuals. Therefore, B. obeum and R. torques may be key targets for obesity intervention based on gut bacteria.

5. Conclusions

An increasing body of evidence suggests the relevance of the gut microbiota in the pathophysiology of diet-related metabolic disorders such as obesity. Given the intricate metabolic interactions between the host and its microbial community, the causative relationship between the gut microbiota and obesity remains elusive. Our study underscores the usefulness of the analyses of the taxa and functional groups of microbes to improve our understanding of the interactions among diet, gut microbiota, and weight loss. Parabacteroides distasonis exhibited the most robust response to the multiphase modified ketogenic diet (MDP-i-KD) intervention, suggesting a potential predominant role in facilitating weight loss, yet the precise mechanisms warrant further investigation. Network analysis revealed a shift in the interactions among gut microbiota toward a healthier state post-weight loss, characterized by positive associations among weight-loss-related species and inhibitory relationships among obesity-associated species after the intervention. Microbial metabolic functions indicated that the gut microbiota may promote weight loss by reducing purine compounds, aromatic amino acid generation, and carbohydrate utilization. Notably, species B. obeum and R. torques, positively correlated with BMI, contributed significantly to these pathways, with reduced relative contributions post-intervention, highlighting their potential as key targets for future obesity interventions. In summary, this work extends our previous findings [28], providing insights into future microbiome-based weight management interventions, facilitating personalized obesity treatment, and better achieving long-term weight management goals. Nevertheless, further research is required to delve into the biological mechanisms of these specific microbial signatures and their impact on obesity.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/nu15194163/s1, Figure S1: study design; Figure S2: Alterations in relative abundance of seven microbial features Identified by GB feature selection method pre- and post-intervention; Table S1: nutrient component of the KD, TD, and HBD; Table S2: food selection range of the KD, TD, and HBD; Table S3: gut microbiota metabolic pathways with relatively high contribution of (a) Blautia obeum; (b) Ruminococcus torques.

Author Contributions

Conceptualization: H.W., X.L., W.L. and W.W.; Data curation: H.W., X.L., S.Z., W.Y. and W.W.; Formal Analysis: X.L., S.Z., H.W., W.Y. and Q.Z.; Investigation: H.W. and Q.Z.; Methodology: H.W., W.L., W.W. and Q.Z.; Project administration: J.Z.; Resources: W.L. and W.W.; Supervision: J.Z.; Validation: X.L. and S.Z.; Visualization: X.L. and S.Z.; Writing—original draft: H.W., X.L., W.L. and W.W.; Writing—review and editing: F.A.S. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Key Research and Development Program of China (2022YFF1100403).

Institutional Review Board Statement

This study was approved by the Affiliated Wuxi People’s Hospital of Nanjing Medical University ethics committee (KYLKS 201806).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Raw sequencing data have been submitted to the National Center for Biotechnology Information (NCBI) under study accession number PRJNA915838. The data presented in this study are available on GitHub: https://github.com/hcwang-jn/gut-MDP-i-KD.git (accessed on 10 August 2023).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Di Angelantonio, E.; Bhupathiraju, S.N.; Wormser, D.; Gao, P.; Kaptoge, S.; Berrington de Gonzalez, A.; Cairns, B.J.; Huxley, R.; Jackson, C.L.; Joshy, G.; et al. Body-mass index and all-cause mortality: Individual-participant-data meta-analysis of 239 prospective studies in four continents. Lancet 2016, 388, 776–786. [Google Scholar] [CrossRef] [PubMed]
  2. GBD 2015 Obesity Collaborators. Health Effects of Overweight and Obesity in 195 Countries over 25 Years. N. Engl. J. Med. 2017, 377, 13–27. [Google Scholar] [CrossRef] [PubMed]
  3. Hoffman, D.J.; Powell, T.L.; Barrett, E.S.; Hardy, D.B. Developmental origins of metabolic diseases. Physiol. Rev. 2021, 101, 739–795. [Google Scholar] [CrossRef]
  4. Ridaura, V.K.; Faith, J.J.; Rey, F.E.; Cheng, J.Y.; Duncan, A.E.; Kau, A.L.; Griffin, N.W.; Lombard, V.; Henrissat, B.; Bain, J.R.; et al. Gut Microbiota from Twins Discordant for Obesity Modulate Metabolism in Mice. Science 2013, 341, 1241214. [Google Scholar] [CrossRef] [PubMed]
  5. Perdomo, C.M.; Cohen, R.V.; Sumithran, P.; Clément, K.; Frühbeck, G. Contemporary medical, device, and surgical therapies for obesity in adults. Lancet 2023, 401, 1116–1130. [Google Scholar] [CrossRef] [PubMed]
  6. Bray, G.A.; Ryan, D.H. Evidence-based weight loss interventions: Individualized treatment options to maximize patient outcomes. Diabetes Obes. Metab. 2021, 23, 50–62. [Google Scholar] [CrossRef]
  7. Carmody, R.N.; Bisanz, J.E. Roles of the gut microbiome in weight management. Nat. Rev. Microbiol. 2023, 21, 535–550. [Google Scholar] [CrossRef]
  8. Wang, Y.; Wang, M.; Chen, J.; Li, Y.; Kuang, Z.; Dende, C.; Raj, P.; Quinn, G.; Hu, Z.; Srinivasan, T.; et al. The gut microbiota reprograms intestinal lipid metabolism through long noncoding RNA Snhg9. Science 2023, 381, 851–857. [Google Scholar] [CrossRef]
  9. Anhê, F.F.; Zlitni, S.; Zhang, S.Y.; Choi, B.S.; Chen, C.Y.; Foley, K.P.; Barra, N.G.; Surette, M.G.; Biertho, L.; Richard, D.; et al. Human gut microbiota after bariatric surgery alters intestinal morphology and glucose absorption in mice independently of obesity. Gut 2023, 72, 460–471. [Google Scholar] [CrossRef]
  10. Qi, Q.; Li, J.; Yu, B.; Moon, J.Y.; Chai, J.C.; Merino, J.; Hu, J.; Ruiz-Canela, M.; Rebholz, C.; Wang, Z.; et al. Host and gut microbial tryptophan metabolism and type 2 diabetes: An integrative analysis of host genetics, diet, gut microbiome and circulating metabolites in cohort studies. Gut 2022, 71, 1095–1105. [Google Scholar] [CrossRef]
  11. Agus, A.; Clément, K.; Sokol, H. Gut microbiota-derived metabolites as central regulators in metabolic disorders. Gut 2021, 70, 1174. [Google Scholar] [CrossRef] [PubMed]
  12. Martin, A.M.; Sun, E.W.; Rogers, G.B.; Keating, D.J. The Influence of the Gut Microbiome on Host Metabolism Through the Regulation of Gut Hormone Release. Front. Physiol. 2019, 10, 428. [Google Scholar] [CrossRef] [PubMed]
  13. Fan, Y.; Pedersen, O. Gut microbiota in human metabolic health and disease. Nat. Rev. Microbiol. 2021, 19, 55–71. [Google Scholar] [CrossRef] [PubMed]
  14. Farhat, E.K.; Sher, E.K.; Džidić-Krivić, A.; Banjari, I.; Sher, F. Functional biotransformation of phytoestrogens by gut microbiota with impact on cancer treatment. J. Nutr. Biochem. 2023, 118, 109368. [Google Scholar] [CrossRef] [PubMed]
  15. von Schwartzenberg, R.J.; Bisanz, J.E.; Lyalina, S.; Spanogiannopoulos, P.; Ang, Q.Y.; Cai, J.; Dickmann, S.; Friedrich, M.; Liu, S.-Y.; Collins, S.L.; et al. Caloric restriction disrupts the microbiota and colonization resistance. Nature 2021, 595, 272–277. [Google Scholar] [CrossRef] [PubMed]
  16. Meslier, V.; Laiola, M.; Roager, H.M.; De Filippis, F.; Roume, H.; Quinquis, B.; Giacco, R.; Mennella, I.; Ferracane, R.; Pons, N.; et al. Mediterranean diet intervention in overweight and obese subjects lowers plasma cholesterol and causes changes in the gut microbiome and metabolome independently of energy intake. Gut 2020, 69, 1258–1268. [Google Scholar] [CrossRef]
  17. Zhang, S.; Wu, P.; Tian, Y.; Liu, B.; Huang, L.; Liu, Z.; Lin, N.; Xu, N.; Ruan, Y.; Zhang, Z.; et al. Gut Microbiota Serves a Predictable Outcome of Short-Term Low-Carbohydrate Diet (LCD) Intervention for Patients with Obesity. Microbiol. Spectr. 2021, 9, e0022321. [Google Scholar] [CrossRef]
  18. Zhu, F.; Ju, Y.; Wang, W.; Wang, Q.; Guo, R.; Ma, Q.; Sun, Q.; Fan, Y.; Xie, Y.; Yang, Z.; et al. Metagenome-wide association of gut microbiome features for schizophrenia. Nat. Commun. 2020, 11, 1612. [Google Scholar] [CrossRef]
  19. Gibbons, S.M. Microbial community ecology: Function over phylogeny. Nat. Ecol. Evol. 2017, 1, 0032. [Google Scholar] [CrossRef]
  20. Yannakoulia, M.; Poulimeneas, D.; Mamalaki, E.; Anastasiou, C.A. Dietary modifications for weight loss and weight loss maintenance. Metabolism 2019, 92, 153–162. [Google Scholar] [CrossRef]
  21. Zhu, H.; Bi, D.; Zhang, Y.; Kong, C.; Du, J.; Wu, X.; Wei, Q.; Qin, H. Ketogenic diet for human diseases: The underlying mechanisms and potential for clinical implementations. Signal Transduct. Target. Ther. 2022, 7, 11. [Google Scholar] [CrossRef] [PubMed]
  22. Amini, M.R.; Aminianfar, A.; Naghshi, S.; Larijani, B.; Esmaillzadeh, A. The effect of ketogenic diet on body composition and anthropometric measures: A systematic review and meta-analysis of randomized controlled trials. Crit. Rev. Food Sci. Nutr. 2022, 62, 3644–3657. [Google Scholar] [CrossRef]
  23. Castellana, M.; Conte, E.; Cignarelli, A.; Perrini, S.; Giustina, A.; Giovanella, L.; Giorgino, F.; Trimboli, P. Efficacy and safety of very low calorie ketogenic diet (VLCKD) in patients with overweight and obesity: A systematic review and meta-analysis. Rev. Endocr. Metab. Disord. 2020, 21, 5–16. [Google Scholar] [CrossRef]
  24. Cunha, G.M.; Guzman, G.; Correa De Mello, L.L.; Trein, B.; Spina, L.; Bussade, I.; Marques Prata, J.; Sajoux, I.; Countinho, W. Efficacy of a 2-Month Very Low-Calorie Ketogenic Diet (VLCKD) Compared to a Standard Low-Calorie Diet in Reducing Visceral and Liver Fat Accumulation in Patients with Obesity. Front. Endocrinol. 2020, 11, 607. [Google Scholar] [CrossRef] [PubMed]
  25. Van Hul, M.; Cani, P.D. The gut microbiota in obesity and weight management: Microbes as friends or foe? Nat. Rev. Endocrinol. 2023, 19, 258–271. [Google Scholar] [CrossRef] [PubMed]
  26. Ang, Q.Y.; Alexander, M.; Newman, J.C.; Tian, Y.; Cai, J.; Upadhyay, V.; Turnbaugh, J.A.; Verdin, E.; Hall, K.D.; Leibel, R.L.; et al. Ketogenic Diets Alter the Gut Microbiome Resulting in Decreased Intestinal Th17 Cells. Cell 2020, 181, 1263–1275.e1216. [Google Scholar] [CrossRef] [PubMed]
  27. Salim, F.; Mizutani, S.; Zolfo, M.; Yamada, T. Recent advances of machine learning applications in human gut microbiota study: From observational analysis toward causal inference and clinical intervention. Curr. Opin. Biotechnol. 2023, 79, 102884. [Google Scholar] [CrossRef]
  28. Yuan, W.W.; Lu, W.W.; Wang, H.C.; Wu, W.J.; Zhou, Q.Y.; Chen, Y.T.; Lee, Y.K.; Zhao, J.X.; Zhang, H.; Chen, W. A multiphase dietetic protocol incorporating an improved ketogenic diet enhances weight loss and alters the gut microbiome of obese people. Int. J. Food Sci. Nutr. 2022, 73, 238–250. [Google Scholar] [CrossRef]
  29. Bolger, A.M.; Lohse, M.; Usadel, B. Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics 2014, 30, 2114–2120. [Google Scholar] [CrossRef]
  30. Li, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv 2013, arXiv:1303.3997. [Google Scholar]
  31. Li, H.; Handsaker, B.; Wysoker, A.; Fennell, T.; Ruan, J.; Homer, N.; Marth, G.; Abecasis, G. Genome Project Data Processing S: The Sequence Alignment/Map format and SAMtools. Bioinformatics 2009, 25, 2078–2079. [Google Scholar] [CrossRef] [PubMed]
  32. Quinlan, A.R.; Hall, I.M. BEDTools: A flexible suite of utilities for comparing genomic features. Bioinformatics 2010, 26, 841–842. [Google Scholar] [CrossRef] [PubMed]
  33. Beghini, F.; McIver, L.J.; Blanco-Míguez, A.; Dubois, L.; Asnicar, F.; Maharjan, S.; Mailyan, A.; Manghi, P.; Scholz, M.; Thomas, A.M.; et al. Integrating taxonomic, functional, and strain-level profiling of diverse microbial communities with bioBakery 3. eLife 2021, 10, e65088. [Google Scholar] [CrossRef] [PubMed]
  34. Segata, N.; Waldron, L.; Ballarini, A.; Narasimhan, V.; Jousson, O.; Huttenhower, C. Metagenomic microbial community profiling using unique clade-specific marker genes. Nat. Methods 2012, 9, 811–814. [Google Scholar] [CrossRef] [PubMed]
  35. Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; Weiss, R.; Dubourg, V.; et al. Scikit-learn: Machine Learning in Python. J. Mach. Learn. Res. 2011, 12, 2825–2830. [Google Scholar]
  36. Lundberg, S.M.; Erion, G.; Chen, H.; DeGrave, A.; Prutkin, J.M.; Nair, B.; Katz, R.; Himmelfarb, J.; Bansal, N.; Lee, S.-I. From local explanations to global understanding with explainable AI for trees. Nat. Mach. Intell. 2020, 2, 56–67. [Google Scholar] [CrossRef]
  37. Gallardo-Becerra, L.; Cornejo-Granados, F.; García-López, R.; Valdez-Lara, A.; Bikel, S.; Canizales-Quinteros, S.; López-Contreras, B.E.; Mendoza-Vargas, A.; Nielsen, H.; Ochoa-Leyva, A. Metatranscriptomic analysis to define the Secrebiome, and 16S rRNA profiling of the gut microbiome in obesity and metabolic syndrome of Mexican children. Microb. Cell Factories 2020, 19, 61. [Google Scholar] [CrossRef]
  38. Yan, P.; Sun, Y.; Luo, J.; Liu, X.; Wu, J.; Miao, Y. Integrating the serum proteomic and fecal metaproteomic to analyze the impacts of overweight/obesity on IBD: A pilot investigation. Clin. Proteom. 2023, 20, 6. [Google Scholar] [CrossRef]
  39. Haro, C.; Montes-Borrego, M.; Rangel-Zúñiga, O.A.; Alcalá-Díaz, J.F.; Gómez-Delgado, F.; Pérez-Martínez, P.; Delgado-Lista, J.; Quintana-Navarro, G.M.; Tinahones, F.J.; Landa, B.B.; et al. Two Healthy Diets Modulate Gut Microbial Community Improving Insulin Sensitivity in a Human Obese Population. J. Clin. Endocrinol. Metab. 2016, 101, 233–242. [Google Scholar] [CrossRef]
  40. You, H.J.; Si, J.; Kim, J.; Yoon, S.; Cha, K.H.; Yoon, H.S.; Lee, G.; Yu, J.; Choi, J.S.; Jung, M.; et al. Bacteroides vulgatus SNUG 40005 Restores Akkermansia Depletion by Metabolite Modulation. Gastroenterology 2023, 164, 103–116. [Google Scholar] [CrossRef]
  41. Liu, R.; Hong, J.; Xu, X.; Feng, Q.; Zhang, D.; Gu, Y.; Shi, J.; Zhao, S.; Liu, W.; Wang, X.; et al. Gut microbiome and serum metabolome alterations in obesity and after weight-loss intervention. Nat. Med. 2017, 23, 859–868. [Google Scholar] [CrossRef] [PubMed]
  42. Choi, Y.J.; Jeon, S.M.; Shin, S. Impact of a Ketogenic Diet on Metabolic Parameters in Patients with Obesity or Overweight and with or without Type 2 Diabetes: A Meta-Analysis of Randomized Controlled Trials. Nutrients 2020, 12, 2005. [Google Scholar] [CrossRef] [PubMed]
  43. Muscogiuri, G.; Barrea, L.; Campolo, F.; Sbardella, E.; Sciammarella, C.; Tarsitano, M.G.; Bottiglieri, F.; Colao, A.; Faggiano, A. Ketogenic diet: A tool for the management of neuroendocrine neoplasms? Crit. Rev. Food Sci. Nutr. 2022, 62, 1035–1045. [Google Scholar] [CrossRef] [PubMed]
  44. Zhang, Y.J.; Zhou, S.Z.; Zhou, Y.F.; Yu, L.F.; Zhang, L.M.; Wang, Y. Altered gut microbiome composition in children with refractory epilepsy after ketogenic diet. Epilepsy Res. 2018, 145, 163–168. [Google Scholar] [CrossRef] [PubMed]
  45. Fan, Y.Y.; Wang, H.; Liu, X.Y.; Zhang, J.M.; Liu, G. Crosstalk between the Ketogenic Diet and Epilepsy: From the Perspective of Gut Microbiota. Mediat. Inflamm. 2019, 2019, 8373060. [Google Scholar] [CrossRef]
  46. Lindefeldt, M.; Eng, A.; Darban, H.; Bjerkner, A.; Zetterstrom, C.K.; Allander, T.; Andersson, B.; Borenstein, E.; Dahlin, M.; Prast-Nielsen, S. The ketogenic diet influences taxonomic and functional composition of the gut microbiota in children with severe epilepsy. NPJ Biofilms Microbiomes 2019, 5, 5. [Google Scholar] [CrossRef]
  47. Ma, D.; Wang, A.C.; Parikh, I.; Green, S.J.; Hoffman, J.D.; Chlipala, G.; Murphy, M.P.; Sokola, B.S.; Bauer, B.; Hartz, A.M.S.; et al. Ketogenic diet enhances neurovascular function with altered gut microbiome in young healthy mice. Sci. Rep. 2018, 8, 6670. [Google Scholar] [CrossRef]
  48. Nagpal, R.; Neth, B.J.; Wang, S.; Mishra, S.P.; Craft, S.; Yadav, H. Gut mycobiome and its interaction with diet, gut bacteria and alzheimer’s disease markers in subjects with mild cognitive impairment: A pilot study. EBioMedicine 2020, 59, 102950. [Google Scholar] [CrossRef]
  49. Dilmore, A.H.; Martino, C.; Neth, B.J.; West, K.A.; Zemlin, J.; Rahman, G.; Panitchpakdi, M.; Meehan, M.J.; Weldon, K.C.; Blach, C.; et al. Effects of a ketogenic and low-fat diet on the human metabolome, microbiome, and foodome in adults at risk for Alzheimer’s disease. Alzheimer’s Dementa 2023. [Google Scholar] [CrossRef]
  50. Nakamura, K.; Hagihara, K.; Nagai, N.; Egashira, R.; Takeuchi, M.; Nakano, M.; Saito, H.; Moriguchi, M.; Tonari, S.; Watanabe, S.; et al. Ketogenic Effects of Multiple Doses of a Medium Chain Triglycerides Enriched Ketogenic Formula in Healthy Men under the Ketogenic Diet: A Randomized, Double-Blinded, Placebo-Controlled Study. Nutrients 2022, 14, 1199. [Google Scholar] [CrossRef]
  51. Muscogiuri, G.; Barrea, L.; Laudisio, D.; Pugliese, G.; Salzano, C.; Savastano, S.; Colao, A. The management of very low-calorie ketogenic diet in obesity outpatient clinic: A practical guide. J. Transl. Med. 2019, 17, 356. [Google Scholar] [CrossRef] [PubMed]
  52. Basciani, S.; Costantini, D.; Contini, S.; Persichetti, A.; Watanabe, M.; Mariani, S.; Lubrano, C.; Spera, G.; Lenzi, A.; Gnessi, L. Safety and efficacy of a multiphase dietetic protocol with meal replacements including a step with very low calorie diet. Endocrine 2015, 48, 863–870. [Google Scholar] [CrossRef] [PubMed]
  53. Swidsinski, A.; Dörffel, Y.; Loening-Baucke, V.; Gille, C.; Göktas, Ö.; Reißhauer, A.; Neuhaus, J.; Weylandt, K.H.; Guschin, A.; Bock, M. Reduced Mass and Diversity of the Colonic Microbiome in Patients with Multiple Sclerosis and Their Improvement with Ketogenic Diet. Front. Microbiol. 2017, 8, 1141. [Google Scholar] [CrossRef]
  54. Hu, X.; Xia, K.; Dai, M.; Han, X.; Yuan, P.; Liu, J.; Liu, S.; Jia, F.; Chen, J.; Jiang, F.; et al. Intermittent fasting modulates the intestinal microbiota and improves obesity and host energy metabolism. NPJ Biofilms Microbiomes 2023, 9, 19. [Google Scholar] [CrossRef] [PubMed]
  55. Wang, K.; Liao, M.; Zhou, N.; Bao, L.; Ma, K.; Zheng, Z.; Wang, Y.; Liu, C.; Wang, W.; Wang, J.; et al. Parabacteroides distasonis Alleviates Obesity and Metabolic Dysfunctions via Production of Succinate and Secondary Bile Acids. Cell Rep. 2019, 26, 222–235.e225. [Google Scholar] [CrossRef] [PubMed]
  56. Depommier, C.; Everard, A.; Druart, C.; Plovier, H.; Van Hul, M.; Vieira-Silva, S.; Falony, G.; Raes, J.; Maiter, D.; Delzenne, N.M.; et al. Supplementation with Akkermansia muciniphila in overweight and obese human volunteers: A proof-of-concept exploratory study. Nat. Med. 2019, 25, 1096–1103. [Google Scholar] [CrossRef]
  57. Wang, C.; Xiao, Y.; Yu, L.; Tian, F.; Zhao, J.; Zhang, H.; Chen, W.; Zhai, Q. Protective effects of different Bacteroides vulgatus strains against lipopolysaccharide-induced acute intestinal injury, and their underlying functional genes. J. Adv. Res. 2022, 36, 27–37. [Google Scholar] [CrossRef]
  58. Rakoff-Nahoum, S.; Foster, K.R.; Comstock, L.E. The evolution of cooperation within the gut microbiota. Nature 2016, 533, 255–259. [Google Scholar] [CrossRef]
  59. Joseph, N.; Clayton, J.B.; Hoops, S.L.; Linhardt, C.A.; Hashim, A.M.; Yusof, B.N.M.; Kumar, S.; Nordin, S.A. Alteration of the Gut Microbiome in Normal and Overweight School Children from Selangor with Lactobacillus Fermented Milk Administration. Evol. Bioinform. 2020, 16, 1176934320965943. [Google Scholar] [CrossRef]
  60. Jain, H.; Singh, G.; Eranki, A. Actinomyces odontolyticus causing meningitis and cervical abscess. In Baylor University Medical Center Proceedings; Taylor & Francis: Abingdon, UK, 2021. [Google Scholar] [CrossRef]
  61. Chen, L.M.; Collij, V.; Jaeger, M.; van den Munckhof, I.C.L.; Vila, A.V.; Kurilshikov, A.; Gacesa, R.; Sinha, T.; Oosting, M.; Joosten, L.A.B.; et al. Gut microbial co-abundance networks show specificity in inflammatory bowel disease and obesity. Nat. Commun. 2020, 11, 4018. [Google Scholar] [CrossRef]
  62. Arnoriaga-Rodríguez, M.; Mayneris-Perxachs, J.; Burokas, A.; Contreras-Rodríguez, O.; Blasco, G.; Coll, C.; Biarnés, C.; Miranda-Olivos, R.; Latorre, J.; Moreno-Navarrete, J.M.; et al. Obesity Impairs Short-Term and Working Memory through Gut Microbial Metabolism of Aromatic Amino Acids. Cell Metab. 2020, 32, 548–560.e547. [Google Scholar] [CrossRef] [PubMed]
  63. Chen, C.; Fang, S.; Wei, H.; He, M.; Fu, H.; Xiong, X.; Zhou, Y.; Wu, J.; Gao, J.; Yang, H.; et al. Prevotella copri increases fat accumulation in pigs fed with formula diets. Microbiome 2021, 9, 175. [Google Scholar] [CrossRef] [PubMed]
  64. Kasai, C.; Sugimoto, K.; Moritani, I.; Tanaka, J.; Oya, Y.; Inoue, H.; Tameda, M.; Shiraki, K.; Ito, M.; Takei, Y.; et al. Comparison of the gut microbiota composition between obese and non-obese individuals in a Japanese population, as analyzed by terminal restriction fragment length polymorphism and next-generation sequencing. BMC Gastroenterol. 2015, 15, 100. [Google Scholar] [CrossRef] [PubMed]
  65. Roses, C.; Cuevas-Sierra, A.; Quintana, S.; Riezu-Boj, J.I.; Martinez, J.A.; Milagro, F.I.; Barcelo, A. Gut Microbiota Bacterial Species Associated with Mediterranean Diet-Related Food Groups in a Northern Spanish Population. Nutrients 2021, 13, 636. [Google Scholar] [CrossRef]
  66. You, A.; Li, Y.; Shen, C.; Fan, H.; He, J.; Liu, Z.; Xue, Q.; Zhang, Y.; Zheng, L. Associations of non-traditional cardiovascular risk factors and body mass index with metabolic syndrome in the Chinese elderly population. Diabetol. Metab. Syndr. 2023, 15, 129. [Google Scholar] [CrossRef]
  67. Li, S.; Feng, L.; Sun, X.; Ding, J.; Zhou, W. Association between serum uric acid and measures of adiposity in Chinese adults: A cross-sectional study. BMJ Open 2023, 13, e072317. [Google Scholar] [CrossRef]
  68. Su, S.Y.; Lin, T.H.; Liu, Y.H.; Wu, P.Y.; Huang, J.C.; Su, H.M.; Chen, S.C. Sex Difference in the Associations among Obesity-Related Indices with Hyperuricemia in a Large Taiwanese Population Study. Nutrients 2023, 15, 3419. [Google Scholar] [CrossRef]
  69. Andres-Hernando, A.; Cicerchi, C.; Kuwabara, M.; Orlicky, D.J.; Sanchez-Lozada, L.G.; Nakagawa, T.; Johnson, R.J.; Lanaspa, M.A. Umami-induced obesity and metabolic syndrome is mediated by nucleotide degradation and uric acid generation. Nat. Metab. 2021, 3, 1189–1201. [Google Scholar] [CrossRef]
  70. Kiana Ashley, W.; Chidimma, K.; Tanya, M.; Julie Anne Kathryn, M.; Jeremy, K.N.; Jia, V.L.; Mark, R.J.; Elaine, H.; Makrina, D.S. Longitudinal metabolic and gut bacterial profiling of pregnant women with previous bariatric surgery. Gut 2020, 69, 1452. [Google Scholar] [CrossRef]
  71. Depommier, C.; Everard, A.; Druart, C.; Maiter, D.; Thissen, J.P.; Loumaye, A.; Hermans, M.P.; Delzenne, N.M.; de Vos, W.M.; Cani, P.D. Serum metabolite profiling yields insights into health promoting effect of A. muciniphila in human volunteers with a metabolic syndrome. Gut Microbes 2021, 13, 1994270. [Google Scholar] [CrossRef]
Nutrients 15 04163 g001
Figure 1. Effect of the MDP-i-KD on gut microbiota. Alpha diversity was assessed using several metrics: (A) observed OTUs, (B) Pielou’s evenness, (C) Shannon index, (D) Simpson index (NS.: p > 0.05). D0: before the intervention, D12: after the intervention; (E) Beta diversity. (F) LEfSe analysis identified the microbes with significant differences in abundance before and after the intervention (LDA score > 2, p < 0.05).
Figure 1. Effect of the MDP-i-KD on gut microbiota. Alpha diversity was assessed using several metrics: (A) observed OTUs, (B) Pielou’s evenness, (C) Shannon index, (D) Simpson index (NS.: p > 0.05). D0: before the intervention, D12: after the intervention; (E) Beta diversity. (F) LEfSe analysis identified the microbes with significant differences in abundance before and after the intervention (LDA score > 2, p < 0.05).
Nutrients 15 04163 g001
Nutrients 15 04163 g002
Figure 2. Machine learning identifies key microbial changes before and after MDP-i-KD intervention. (A) Impact of various feature selection methods on the performance of different models (evaluation metric is AUC). (B) Feature interpretation of the contribution of the seven features in the GB model to pre- and post-intervention state classification using SHAP analysis.
Figure 2. Machine learning identifies key microbial changes before and after MDP-i-KD intervention. (A) Impact of various feature selection methods on the performance of different models (evaluation metric is AUC). (B) Feature interpretation of the contribution of the seven features in the GB model to pre- and post-intervention state classification using SHAP analysis.
Nutrients 15 04163 g002
Nutrients 15 04163 g003
Figure 3. Species-level co-occurrence network changes after diet intervention. Each node was labeled by the corresponding species. A connection represents a strong (|r| > 0.6) and significant (p-value < 0.01) correlation. Nodes sharing the same color represent identical modules, while node size corresponds to their betweenness centrality scores. Dashed boundaries highlight alterations in interactions among obesity-related species pre-and post-intervention, (A) Blautia obeum module; (B) Prevotella copri module; (C) Blautia producta module; (D) B. thetaiotaomicron module; (E) E. tayi module.
Figure 3. Species-level co-occurrence network changes after diet intervention. Each node was labeled by the corresponding species. A connection represents a strong (|r| > 0.6) and significant (p-value < 0.01) correlation. Nodes sharing the same color represent identical modules, while node size corresponds to their betweenness centrality scores. Dashed boundaries highlight alterations in interactions among obesity-related species pre-and post-intervention, (A) Blautia obeum module; (B) Prevotella copri module; (C) Blautia producta module; (D) B. thetaiotaomicron module; (E) E. tayi module.
Nutrients 15 04163 g003
Nutrients 15 04163 g004
Figure 4. Correlation analysis of gut microbiota and physiological indicators. (A) Species significantly and positively linked to obesity-related physiological indicators; (B) Species significantly and negatively linked to obesity-related physiological indicators (*** p < 0.001; ** p < 0.01; * p < 0.05).
Figure 4. Correlation analysis of gut microbiota and physiological indicators. (A) Species significantly and positively linked to obesity-related physiological indicators; (B) Species significantly and negatively linked to obesity-related physiological indicators (*** p < 0.001; ** p < 0.01; * p < 0.05).
Nutrients 15 04163 g004
Nutrients 15 04163 g005
Figure 5. The MDP-i-KD changes the gut microbiota functional characteristics of subjects with obesity (* p < 0.05).
Figure 5. The MDP-i-KD changes the gut microbiota functional characteristics of subjects with obesity (* p < 0.05).
Nutrients 15 04163 g005
Nutrients 15 04163 g006
Figure 6. Correlation between the metabolic pathways of the gut microbiota and physiological indicators studied (*** p < 0.001; ** p < 0.01; * p < 0.05).
Figure 6. Correlation between the metabolic pathways of the gut microbiota and physiological indicators studied (*** p < 0.001; ** p < 0.01; * p < 0.05).
Nutrients 15 04163 g006
Nutrients 15 04163 g007
Figure 7. Contribution of species to key metabolic pathways.
Figure 7. Contribution of species to key metabolic pathways.
Nutrients 15 04163 g007
Table 1. Anthropometric assessment and biochemical measurements of participants before and after the intervention.
Table 1. Anthropometric assessment and biochemical measurements of participants before and after the intervention.
CharacteristicBeforeAfterp-ValueChange
Body composition
 Bodyweight, kg86.6 ± 14.778.4 ± 13.20.0017 −8.2 ± 2.5
 BMI, kg/m231.0 ± 2.628.1 ± 2.3 *0.0002 −2.9 ± 0.8
 Waist, cm99.3 ± 10.091.9 ± 8.90.0017 −7.3 ± 3
 Hip, cm105.4 ± 5.0100.2 ± 5.2 *0.0016 −5.2 ± 1.7
Blood pressure, mm Hg
 Systolic 132.3 ± 14.1123.5 ± 14.80.0118 −8.8 ± 10.6
 diastolic 78.4 ± 11.572.4 ± 11.20.0291 −6.0 ± 10
Liver function, U/L
 AST26.5 ± 10.321.2 ± 5.80.0357 −5.2 ± 7.3
 ALP78.4 ± 26.484.0 ± 37.00.4214 5.6 ± 18.8
 ALT41.9 ± 30.727.5 ± 22.70.0134 −14.4 ± 19.5
Renal function
 Albumin, g/L48.3 ± 1.346.6 ± 2.10.0645 −1.7 ± 2.5
 BUN, mmol/L4.7 ± 1.55.3 ± 0.80.0803 0.6 ± 1.5
 Creatinine, μmol/L71.5 ± 13.672.0 ± 13.20.7869 0.5 ± 5.0
 UA, μmol/L452.35 ± 29.73438.05 ± 22.190.4973−14.3 ± 37.10
Lipids, mmol/L
 Total cholesterol5.2 ± 0.95.2 ± 0.90.6848 0.0 ± 0.8
 LDL cholesterol3.4 ± 0.83.4 ± 0.90.5292 −0.1 ± 0.7
 HDL cholesterol1.0 ± 0.21.1 ± 0.20.1259 0.1 ± 0.2
 Triglyceride2.3 ± 2.11.5 ± 0.70.0081 −0.9 ± 1.6
HbA1c [%]5.4 ± 0.55.0 ± 0.4 *0.0058 −0.4 ± 0.5
Abbreviations: BMI, body mass index; AST, aspartate transaminase; ALP, alkaline phosphatase; ALT, alanine aminotransferase; BUN, blood urea nitrogen; UA, uric acid; LDL, low-density lipoprotein; HDL, high-density lipoprotein; HbA1c, hemoglobin Alc. A Wilcoxon signed-rank test was used: * p < 0.05.
Table 2. The topological features of the network.
Table 2. The topological features of the network.
VariableD0D12
Number of edges 760.000901.000
Number of positive edges 722.000885.000
Number of negative edges 38.00016.000
Number of vertices 299.000318.000
Average degree5.0845.667
Average path length 2.8672.395
Diameter8.00010.000
Average clustering coefficient0.9050.920
Centralization degree 0.0360.028
Modularity0.9410.888
Number of modularity 141.000138.000
D0: before the intervention; D12: after the intervention.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Wang, H.; Lv, X.; Zhao, S.; Yuan, W.; Zhou, Q.; Sadiq, F.A.; Zhao, J.; Lu, W.; Wu, W. Weight Loss Promotion in Individuals with Obesity through Gut Microbiota Alterations with a Multiphase Modified Ketogenic Diet. Nutrients 2023, 15, 4163. https://doi.org/10.3390/nu15194163

AMA Style

Wang H, Lv X, Zhao S, Yuan W, Zhou Q, Sadiq FA, Zhao J, Lu W, Wu W. Weight Loss Promotion in Individuals with Obesity through Gut Microbiota Alterations with a Multiphase Modified Ketogenic Diet. Nutrients. 2023; 15(19):4163. https://doi.org/10.3390/nu15194163

Chicago/Turabian Style

Wang, Hongchao, Xinchen Lv, Sijia Zhao, Weiwei Yuan, Qunyan Zhou, Faizan Ahmed Sadiq, Jianxin Zhao, Wenwei Lu, and Wenjun Wu. 2023. "Weight Loss Promotion in Individuals with Obesity through Gut Microbiota Alterations with a Multiphase Modified Ketogenic Diet" Nutrients 15, no. 19: 4163. https://doi.org/10.3390/nu15194163

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop