Effects of Obesity and Feeding Avocado Extract on Gut Microbiota and Fecal Metabolomic Profile in Overweight/Obese Cats

Abstract

Background/Objectives: Obesity is a growing problem in the feline population worldwide. An extract of unripe avocado (AvX) has been shown to attenuate gains in body weight and body fat in mice fed a high-fat diet. The aim of this study was to evaluate the effects of overweight/obesity and AvX on gut microbiota (GM) and fecal metabolomics in cats with natural overweight/obesity. Methods: Ten naturally overweight/obese and ten lean purpose-bred domestic shorthair cats were included in this study. In a prospective, randomized, double-blind study, one group of overweight/obese cats received AvX, while the second group received maltodextrin for 16 weeks. Fecal samples were collected after spontaneous defecation at specific time points and submitted for analysis. Fecal samples of overweight/obese cats collected before administration of AvX/maltodextrin were also compared to fecal samples of lean cats. Results: There was a significant difference in the clustering of GM over time in the AvX group and between lean and overweight/obese cats. The abundance of Firmicutes in the group of cats receiving AvX decreased compared to baseline. AvX induced a trend toward an increased abundance of Dialister sp. and a trend of decreased abundances of SMB53, Roseburia sp., Blautia producta, Acidaminococcus sp., Akkermansia sp., Adlercreutzia sp., and Collinsella aerofaciens. The metabolites that significantly differed after AvX administration included tryptophan, indole-3-acetate, nicotinamine, and glycyl-proline. At the species level, abundances of Prevotella sp., Turicibacter sp., Clostridium sp., Veillonella sp., Dialister sp., Catenibacterium sp., Eubacterium biforme, Desulfovibrio sp., and Campylobacter sp. were significantly higher in lean cats. Abundances of Coriobacterium sp. and Ruminococcus gnavus were significantly higher in overweight/obese cats. Additionally, LEfSe analysis identified Dialister as the genus associated with AvX administration and Dialister, Prevotella, Ruminococcus, Campylobacter, Catenibacterium, Clostridium, Helicobacter, Eubacterium, Pseudoramibacter, Veillonella, S247, Turicibacter, and Phascolarctobacterium as bacteria associated with the lean state. Genus Coriobacterium and Enterococcus were associated with overweight/obesity. A correlation between the concentration of metabolites significantly different between the AvX/placebo groups and the abundances of detected bacterial taxa at the genus level was assessed and described. Conclusions: There are significant differences in the GM between lean and overweight/obese cats. AvX consumption appears to affect the composition of GM and fecal metabolite concentrations in naturally overweight/obese cats.

1. Introduction

Obesity is a growing problem in the human and pet population worldwide. Approximately 40.3% of people in the USA [1] and 27–60% of domestic cats are overweight/obese, and the prevalence is increasing [2,3,4,5,6,7,8,9,10]. Obesity is the most common nutritional disorder of pet cats and predisposes them, similar to people, to the development of numerous medical conditions [2,3,9,10,11]. Associated diseases that lead to the most significant morbidity and mortality rates are diabetes mellitus, hepatic lipidosis, and osteoarthritis [2,3,10].

The cat is considered a valuable animal model for studying the pathophysiology and complications of obesity and diabetes mellitus in people [12,13,14]. While a variety of interventions are available for people (e.g., nutritional, pharmaceutical, nutraceutical, surgical), the approach for cats is currently limited to calorie restriction and exercise, which is often unsuccessful [2,10]. It is therefore necessary to continue improving our understanding of the underlying pathogenesis to develop new treatment strategies.

Data from studies in animal models and humans suggest that obesity and related diseases are associated with shifts in the gut microbiota (GM) [15,16,17], which result in changes to the production of metabolites. These organisms produce a diversity of metabolites from both exogenous dietary substrates and endogenous host compounds [18], including short-chain fatty acids (SCFAs), indole derivatives, polyamines, and secondary bile acids [19].

SCFAs are some metabolites produced by fermentation reactions, as are amino acids and vitamins [20]. Acetate, propionate, and butyrate are the most abundant SCFAs in the intestinal tract [21]. Once produced by the bacteria, SCFAs are absorbed into the bloodstream and bind to G-protein-coupled receptors, which participate in cellular signaling mechanisms, including those involved in lipid, glucose, and cholesterol metabolism [22], and gut inflammation [19,23]. SCFAs produced by GM activate receptors on the surface of neutrophils, macrophages, and dendritic cells, which promotes IL-18, IL-22, IgA, and glucagon-like peptide-1 (GLP 1) production.

Metabolomic studies have linked gut dysbiosis to changes in host metabolism, with changes in tryptophan metabolism as one of the key factors. Tryptophan is an essential amino acid, and growing evidence from both human and animal studies suggests a complex interaction between obesity, inflammation, intestinal mucosal health, and tryptophan metabolism [24,25]. Data from cross-sectional studies indicate that host tryptophan catabolism is altered in individuals with obesity [26,27], with higher plasma kynurenine levels observed in obese and type 2 diabetes mellitus (T2DM) individuals compared to healthy controls [28]. Additionally, these patients show a reduced production of indole-3-acetate in the feces, indicating a shift in tryptophan metabolism [28]. This shift is associated with increased intestinal indoleamine 2,3-dioxygenase-1 (IDO1) activity, which drives tryptophan breakdown toward kynurenine rather than microbial indole derivatives and IL-22 production [29]. These findings suggest that proinflammatory stimuli in obesity enhance intestinal IDO1 activity and suppress the microbial indole pathway, similar to the findings on tryptophan metabolism in inflammatory bowel disease (IBD) [27,30].

In addition, GM creates a homeostatic imbalance within the host toward adiposity, inflammatory response, oxidative stress, and metabolic dysfunction [31,32]. Although the smaller relative colon volume in cats correlates with the generation of less energy by GM compared to people, the altered GM of obese cats can generate significantly more energy from food than the GM of lean cats [33,34].

To our knowledge, limited research has evaluated the role of GM in feline obesity. When compared to lean cats, the GM of overweight/obese shelter cats was significantly different [35]—obese cats had enriched taxa (mainly Firmicutes) [36]—and significant differences in GM were observed with weight loss and between neutered lean and neutered obese cats [37,38].

An extract of unripe avocados enriched with D-mannoheptulose (AvX) has been proposed as a calorie restriction mimetic because of its ability to reversibly inhibit hexokinase II, the first step in intracellular glycolysis [39,40]. It is hypothesized that by inhibiting glycolysis, cellular metabolism is redirected toward more efficient metabolism of fatty acids, increased mitochondrial numbers, and oxygen metabolism, resulting in a loss of body fat. Results from diet-induced obesity studies in mice indicate that the AvX attenuates gains in body weight and body fat in mice fed a high-fat diet without a reduction in food intake [41].

The aim of this study was to evaluate the effects of AvX on the GM and fecal metabolomic profile in cats with naturally occurring overweight/obesity and to compare the GM of overweight/obese and lean cats. Evaluating the results of this study might provide insight into the role that GM and metabolomics play in the ability of AvX to attenuate increases in body weight and fat.

2. Materials and Methods

2.1. Animals

Ten naturally overweight/obese purpose-bred domestic shorthair cats were included in the study evaluating the effects of AvX on the GM and fecal metabolomics in overweight/obese cats (Study 1). These cats were also included in the study evaluating differences in the GM between overweight/obese and lean cats (Study 2). Additionally, ten naturally lean purpose-bred domestic shorthair cats were included in Study 2. All cats were kept indoors at an Association for Assessment and Accreditation of Laboratory Animal Care International-accredited facility. The cats were group-housed with 4–6 cats in each room for the duration of the study. Short-duration individual housing was required during fecal sample collection. All cats had been vaccinated against common infectious diseases [42] and had no known history or clinical signs of illness. The study protocols were approved by an institutional animal care and use committee. Physical examination, a CBC, serum biochemical analysis, and total T4 concentration determination were performed on each cat. All cats were deemed healthy with results within reference intervals for these assessments. A nine-point system (1 = extremely thin, 5 = optimal, and 9 = obese) was used for body condition scoring [43].

2.2. Avocado Dietary Supplement, Placebo

The D-mannoheptulose (MH)-enriched avocado extract powder produced by Procter&Gamble and analyzed by HPLC is certified to contain 25.5% MH of dry weight powder. It was stored frozen at −20 °C until being thawed for use. The MH content of avocado extract has been found to be generally stable for over 2 years after thawing. The avocado extract was combined with a flavored (chicken) Polox oral gel (poloxamer 30%) by a commercial compounding pharmacy (Wedgewood Pharmacy and Diamondback Drugs, Scottsdale, AZ, USA) to create the oral dietary supplement. This is a standard compounding technique for oral medications, and the Polox oral gel has no known biological activity. The resulting oral gel contained 100 mg/mL of avocado extract. This allowed the daily dose to be administered in a small (1–2 mL) volume of supplement. The initial dose of MH had been 5 mg/kg/day (weeks 1–8) and then increased to 10 mg/kg/day during the second half of Study 1 (weeks 9–16). The placebo consisted of maltodextrin combined with Polox oral gel to a final concentration of 100 mg/mL, which had been administered at a dose of 5 mg/kg/day (weeks 1–8) and then 10 mg/kg/day (weeks 9–16) because the MH dose was increased.

2.3. Study Design

For Study 1, the overweight/obese cats were randomized based on age, sex, body weight, and body condition score (BCS). In a prospective, randomized, double-blind study, the AvX group (n = 5) received the AvX PO once daily, while the placebo group (n = 5) received maltodextrin PO once daily for 16 weeks. The cats in both groups were fed an unrestricted amount of LabDiet^® 5003 Laboratory Feline Diet. Daily scoring of fecal consistency using a previously published 7-point pictorial fecal scoring system (Nestlé Purina PetCare, St Louis, MO, USA) was performed throughout the entire study. Fecal samples were collected after spontaneous defecation (no longer than 2 h) at 0, 4, 8, 12, and 16 weeks. The fresh feces were weighed, aliquoted into plastic bags, and immediately frozen at −80 °C until shipped as a batch to the Gastrointestinal Laboratory at Texas A&M University for further analysis. For Study 2, fecal samples of ten lean cats were collected after spontaneous defecation. Handling, processing, and evaluation of the GM of the fecal samples was the same as in Study 1. The GM of fecal samples from lean cats was compared to that of overweight/obese cats collected before administration of AvX/placebo.

2.4. Analysis of Fecal Microbiota (DNA Extraction and Sequencing of 16S rRNA Genes)

DNA was extracted from fecal samples using a MoBio Power soil DNA isolation kit (MoBio Laboratories), following the manufacturer’s instructions. Illumina sequencing of the V4 region of the bacterial 16S rRNA genes was performed using primers 515F (5′-GTGCCAGCMGCCGCGGTAA-3′) to 806R (5′-GGACTACVSGGGTATCTAAT-3′) at the MR DNA laboratory (www.mrdnalab.com, Shallowater, TX, USA, accessed on 9 August 2025), as previously described [44,45,46,47,48].

Raw sequences were analyzed using QIIME v1.9 (Quantitative Insights into Microbial Ecology) for the AvX vs. placebo group comparison and QIIME v2.0 for the comparison between lean and overweight/obese cats, as previously described [49,50]. In both pipelines, sequences were demultiplexed, barcodes and primers were removed, and sequences of low quality (e.g., short reads <150 bp, ambiguous bases, or long homopolymers) were excluded. Chimeric sequences were identified and removed using USEARCH. Operational taxonomic units (OTUs) were defined at 97% similarity using the Greengenes database (v13.8) and an open-reference approach (QIIME v1.9) or via DADA2 denoising (QIIME v2.0). Sequences identified as chloroplasts, mitochondria, unassigned, cyanobacteria, or low-abundance OTUs (<0.01% of total reads) were excluded from downstream analysis. Rarefaction was performed to an even sequencing depth of 69,190 sequences per sample for the AvX vs. placebo group (Study 1) and 51,414 sequences per sample for the lean vs. overweight/obese group (Study 2). Alpha diversity was assessed using Chao1 (richness), observed ASVs (amplicon sequence variants; species richness), and Shannon diversity (evenness). Beta diversity was evaluated using phylogeny-based weighted UniFrac distance metrics and visualized through Principal Coordinate Analysis (PCoA) [51]. Differences in microbial community structures were tested using Analysis of Similarity (ANOSIM) within PRIMER 6 software (PRIMER-E Ltd., Luton, UK) in both Study 1 and Study 2. Linear discriminant analysis effect size (LEfSe) was performed at the genus level to identify potential microbial biomarkers [52,53].

2.5. Untargeted Fecal Metabolomics

The fecal metabolome of samples in Study 1 was assessed using an untargeted approach at the West Coast Metabolomics Center (University of California, Davis, CA, USA) via gas chromatography time-of-flight mass spectrometry, as previously described for canine fecal samples [46,47]. Peak height data were obtained and uploaded to MetaboAnalyst 4.0 (Xia Lab, McGill University, Montreal, QC, Canada). Before statistical analysis, the data were log transformed and Pareto scaled [46,48].

2.6. Statistical Analysis

All data sets were tested for normality using the Shapiro–Wilk normality test. Unless otherwise stated, data are reported as means and standard deviations. Because of a limited sample size (5 cats in AvX and placebo groups), the Mann–Whitney U test was used to assess the difference in age and body weight in Study 1. The Mann–Whitney U test was also used to assess the difference in BCS between groups in both studies, the difference in fecal scores between groups in Study 1, and the difference in the Firmicutes/Bacteroidetes ratio between groups in Study 2. A two-sample t-test was used to assess the difference in age and body weight between lean and overweight/obese groups in Study 2. Fischer’s exact test was used to assess the difference in sex distribution in both studies. The Wilcoxon matched pairs signed rank test was used to assess the difference in the Firmicutes/Bacteroidetes ratio in Study 1.

For assessment of the GM, Friedman tests were performed for data over time and adjusted for multiple comparisons using Benjamini and Hochberg’s False Discovery Rate [54] at each taxonomic level and diversity parameters (both studies). Post hoc Dunn’s multiple comparison test was used to determine the bacterial taxa that were different between the time points.

A mixed-effect ANOVA on ranks was used for an analysis of fecal metabolomics (the variables measured over time) and adjusted for multiple comparisons using Benjamini and Hochberg’s False Discovery Rate [54]. When a fixed effect was detected, Tukey–Kramer post hoc comparisons were performed with the least square means for the effect.

Spearman’s rank correlation coefficient was calculated between metabolites (those showing significant difference between the AvX and placebo groups at the end of the 16-week period, Study 1) and the abundances of detected bacterial taxa at the genus level, also assessed at the end of the 16-week period in the AvX group of cats.

Three statistical software programs (GraphPad Prism 7.0, GraphPad Software Inc; JMP Pro 17 and SAS 9.4, release TS1M8 SAS Institute Inc., Cary, NC, USA) were used as appropriate. p or q values < 0.05 were considered statistically significant.

3. Results

3.1. Demographic Data

Study 1 (

Table 1. Demographic data for naturally overweight/obese domestic shorthair cats of Study 1 (n = 10).

	AvX Group	Placebo Group
Number	5	5
Sex
neutered males	2	3
spayed females	3	2
Age (mean ± SD)	3.7 ± 1.7 years	4.7 ± 2.6 years
	(Range: 3 to 7 years)	(Range: 2.5 to 9.5 years)
Body weight (mean ± SD)	5.7 ± 1.4 kg	5.5 ± 1 kg
	(Range: 4.8 to 8.2 kg)	(Range: 4.6 to 7.5 kg)
BCS (median)	7/9	8/9
	(Distribution: three cats with BCS 7, one cat with BCS 8, one cat with BCS 9)	(Distribution: one cat with BCS 6, one cat with BCS 7, one cat with BCS 8, two cats with BCS 9)
Fecal score (median)	Week 0 2/7 (range, 2/7–3/7), week 8 2/7 (range, 2/7–4/7), week 16 2/7 (range, 2/7–3/7)	Week 0 2/7 (range, 2/7–2/7), week 8 2/7 (range, 2/7–2/7), week 16 3/7 (range, 2/7–3/7)

AvX, extract of unripe avocado; BCS, body condition score.

): There was no significant difference in age (p = 0.5), body weight (p = 0.9), body condition score (BCS) (p = 0.9), or sex distribution (p = 1) between the AvX and placebo groups. There was no significant difference in fecal score at week 0 (p = 1), 8 (p = 1), and 16 (p = 0.52) between the two groups. No significant change in body weight (p = 0.07) or BCS (0.08) was noted in the AvX group compared to the placebo group at the end of the 16-week period.

Study 2 (

Table 2. Demographic data for naturally overweight/obese and lean domestic shorthair cats of Study 2 (n = 20).

	Overweight/Obese	Lean
Number	10	10
Sex
neutered males	5	5
spayed females	5	5
Age (mean ± SD)	4.2 ± 1.9 years	1.3 ± 0.08 years
	(Range: 2.5 to 9.5 years)	(Range: 1.3 to 1.5 years)
Body weight (mean ± SD)	5.7 ± 1.1 kg	4.7 ± 0.9 kg
	(Range: 4.6 to 8.2 kg)	(Range: 3.2 to 5.72 kg)
BCS (median)	8/9	5/9
	(Distribution: one cat with BCS 6, four cats with BCS 7, two cats with BCS 8, three cats with BCS 9)	(Distribution: four cats with BCS 4/9 and six cats with BCS 5/9)

BCS, body condition score.

): The overweight/obese cats were significantly older (p = 0.0004), heavier (p = 0.04), and had a higher BCS than lean cats (p < 0.0001). There was no significant difference in the sex distribution (p = 1) between the two groups.

3.2. Gut Microbiota Composition

3.2.1. Univariate Statistics

Study 1: AvX administration appeared to impact abundances of bacteria of the GM (p < 0.05) of naturally overweight/obese cats; however, after performing a Benjamini–Hochberg adjustment, significant differences between groups were no longer identified for any taxa (q > 0.05).

At the species level, abundances of Dialister sp. (p = 0.04, q = 0.6) and Rickettsiella sp. (p = 0.02, q = 0.6) were trending up compared to baseline in the AvX group at the end of the 16-week period, while a trending decrease in abundances of SMB53 (p = 0.02, q = 0.6), Roseburia sp. (p = 0.02, q = 0.6), Blautia producta (p = 0.04, q = 0.6), Helicobacter sp. (p = 0.001, q = 0.07), and Vibrio sp. (p = 0.04, q = 0.6) was observed. Additionally, a trending decrease in Acidaminococcus sp. (p = 0.01, q = 0.4), Akkermansia sp. (p = 0.01, q = 0.4), Adlercreutzia sp. (p = 0.02, q = 0.4), and Collinsella aerofaciens (p = 0.02, q = 0.4) was noted in the AvX group compared to the placebo group at week 16. The results are also summarized in

Table 3. Impact of an extract of unripe avocado administration on the gut microbiota of overweight/obese domestic shorthair cats in Study 1 (n[AvX] =5; n[placebo] = 5) at the species level.

	Phylotypes with Altered Abundances	p/q Value
At the species level, abundances of the following bacteria were trending up compared to baseline in the AvX group at the end of the 16-week period in overweight/obese cats (n = 5) At the species level, abundances of the following bacteria were trending down compared to baseline in the AvX group at the end of the 16-week period in overweight/obese cats (n = 5) At the species level, abundances of the following bacteria were trending down in the AvX group compared to the placebo group at week 16.	Dialister sp. Rickettsiella sp. SMB53 Roseburia sp. Blautia producta Helicobacter sp. Vibrio sp. Acidaminococcus sp. Akkermansia sp. Adlercreutzia sp. Collinsella aerofaciens	(p = 0.04, q = 0.6) (p = 0.02, q = 0.6) (p = 0.02, q = 0.6) (p = 0.02, q = 0.6) (p = 0.04, q = 0.6) (p = 0.001, q = 0.07) (p = 0.04, q = 0.6) (p = 0.01, q = 0.4) (p = 0.01, q = 0.4) (p = 0.02, q = 0.4) (p = 0.02, q = 0.4)

AvX, extract of unripe avocado.

, with the bacterial frequencies provided in Supplementary File S1.

Study 2: The lean state had a significant impact on the GM. At the species level, abundances of Prevotella sp. (p = 0.001, q = 0.01), Turicibacter sp. (p = 0.003, q = 0.02), Clostridium sp. (p = 0.003, q = 0.02), Veillonella sp. (p = 0.01, q = 0.04), Dialister sp. (p = 0.001, q = 0.02), Catenibacterium sp. (p = 0.002, q = 0.02), Eubacterium biforme (p = 0.01, q = 0.04), Desulfovibrio sp. (p = 0.001, q = 0.01), and Campylobacter sp. (p = 0.001, q = 0.01) were significantly higher in lean than in obese cats. At the species level, obesity was associated with significantly increased abundances of Coriobacterium sp. (p = 0.004, q = 0.03) and Ruminococcus gnavus (p = 0.001, q = 0.01). Results are also summarized in

Table 4. Impact of naturally lean and overweight/obese state on the gut microbiota of domestic shorthair cats in Study 2 (n = 20) at the species level.

	Phylotypes with Altered Abundances	p/q Value
At the species level, abundances of the following bacteria were significantly higher in lean cats (n = 10) At the species level, abundances of the following bacteria were significantly higher in overweight/obese cats (n = 10)	Prevotella sp. Turicibacter sp. Clostridium sp. Veillonella sp. Dialister sp. Catenibacterium sp. Eubacterium biforme Desulfovibrio sp. Campylobacter sp. Coriobacterium sp. Ruminococcus gnavus	(p = 0.001, q = 0.01) (p = 0.003, q = 0.02) (p = 0.003, q = 0.02) (p = 0.01, q = 0.04) (p = 0.001, q = 0.02) (p = 0.002, q = 0.02) (p = 0.01, q = 0.04) (p = 0.001, q = 0.01) (p = 0.001, q = 0.01) (p = 0.004, q = 0.03) (p = 0.001, q = 0.01)

, with the bacterial frequencies provided in Supplementary File S2.

3.2.2. Linear Discriminant Analysis Effect Size (LEfSe)

LEfSe analysis identified the biomarkers in the GM (at the genus level).

Study 1: A comparison of the AvX to the placebo group revealed two operational taxonomic units that were associated with AvX administration at week 16—Dialister and Rickettsiella (LDA score greater than 3).

Study 2: Bacteria associated with a lean state included genus Dialister, Prevotella, Ruminococcus, Campylobacter, Catenibacterium, Clostridium, Helicobacter, Eubacterium, Pseudoramibacter, Veillonella, S247, Turicibacter, and Phascolarctobacterium (LDA score greater than or equal to 3). Bacteria associated with overweight/obesity included genus Coriobacterium and Enterococcus (LDA score of more than 3). Results are also summarized in

Table 5. Linear discriminant analysis effect size (LEfSe) identified the following biomarkers (at the genus level) in the gut microbiota of domestic shorthair cats of Study 1 (n = 10) and Study 2 (n = 20).

	Enriched Phylotypes	LDA Score
Study 1, phylotypes differentially enriched in AvX compared to the placebo group (week 16) Study 2, naturally lean cats Study 2, naturally overweight/obese cats	Dialister Rickettsiella Dialister Prevotella Ruminococcus Campylobacter Catenibacterium Clostridium Helicobacter Eubacterium Pseudoramibacter Veillonella S247 Turicibacter Phascolarctobacterium Coriobacterium Enterococcus	(LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score > 3) (LDA score = 3) (LDA score > 3) (LDA score > 3)

AvX, extract of unripe avocado; LDA, linear discriminant analysis.

and bar plots are provided in Supplementary File S3.

3.2.3. Firmicutes/Bacteroidetes Ratio

Study 1: The abundance of Firmicutes decreased in the AvX group at week 16 compared to baseline (at phylum level; p = 0.02). The Firmicutes/Bacteroidetes ratio decreased from week 0 to week 16 in the AvX group as well; however, the reduction did not reach statistical significance (p = 0.06) (Figure 1a).

Study 2: There was no significant difference in the Firmicutes/Bacteroidetes ratio between overweight/obese and lean cats (p = 0.31) (Figure 1b).

3.2.4. Diversity Within Samples (Alpha Diversity)

The results are summarized in

Table 6. Alpha diversity, diversity within fecal samples, of domestic shorthair cats in Study 1 (n = 10) and Study 2 (n = 20).

	Alpha-Diversity Parameter	p Value
Study 1, AvX vs. placebo group at baseline Study 1, AvX vs. placebo group on week 16 Study 1, placebo group over 16-week period Study 1, AvX group over 16-week period Study 2, lean vs. overweight/obese cats	Chao1 Observed ASVs Shannon index Chao1 Observed ASVs Shannon index Chao1 Observed ASVs Shannon index Chao1 Observed ASVs Shannon index Chao1 Observed ASVs Shannon index	p = 1 p = 0.4 p = 0.5 p = 0.8 p = 0.4 p = 0.8 p = 0.8 p = 0.6 p = 0.7 p = 0.8 p = 0.6 p = 0.7 p = 0.4 p = 0.4 p = 0.8

ASVs, amplicon sequence variants; AvX, extract of unripe avocado.

, with comprehensive diversity metrics available in Supplementary File S4.

Study 1: No significant differences in alpha diversity were observed between the placebo and AvX groups at baseline or at week 16. Furthermore, alpha diversity remained stable over the 16-week period in both groups.

Study 2: Alpha diversity parameters did not differ significantly between lean and overweight/obese cats.

These findings suggest that neither the AvX administration nor body condition status had a measurable impact on overall microbial richness or evenness within the GM in this study.

3.2.5. Diversity Between Samples (Beta Diversity)

Study 1: Beta diversity assessed by a weighted UniFrac distance matrix test indicated a significant difference in the clustering of microbiota over time in the AvX group (week 0–week 16, ANOSIM, R = 0.36, p = 0.024) and no significant change in the placebo group (week 0–week 16, ANOSIM, R = −0.068, p = 0.635) (Figure 2a,b).

Study 2: Beta diversity assessed by a weighted UniFrac distance matrix test indicated a significant difference between the GM of lean and overweight/obese cats (ANOSIM, R = 0.377, p = 0.012) (Figure 3).

3.3. Analysis of the Fecal Metabolome

Study 1: A total of 204 named metabolites were detected in fecal samples (Supplementary File S5). After performing a Benjamini–Hochberg adjustment, we found that four metabolites were significantly different between the AvX and placebo groups at the end of the 16-week period. AvX administration was associated with a reduction in the concentration of tryptophan (p = 0.0006, q = 0.02; Figure 4a) and nicotianamine (p = 0.002, q = 0.04) and an increase in the concentration of indole-3-acetate (p = 0.0004, q = 0.02; Figure 4b) and glycyl-proline (p = 0.0003, q = 0.02). The results, including metabolite concentrations, are summarized in

Table 7. Concentration (peak height) of fecal metabolites was significantly different between the AvX (n = 5) and placebo (n = 5) groups of overweight/obese domestic shorthair cats in Study 1 at the end of the 16-week period.

Metabolite	Concentration in AvX Group (Median and Range))	Concentration in Placebo Group (Median and Range)	p/q Value
Tryptophan Nicotianamine Indole-3-acetate Glycyl-proline	7449 (2476 to 20,805) 279 (158 to 379) 108,255 (78,679 to 146,151) 8826 (1251 to 16,732)	26,731 (24,431 to 62,562) 1251 (609 to 2157) 44,404 (7841 to 59,547) 1310 (418 to 15,538)	p = 0.0006, q = 0.02 p = 0.002, q = 0.04 p = 0.0004, q = 0.02 p = 0.0003, q = 0.02

AvX, extract of unripe avocado.

.

3.4. Correlation Between the Concentration of Metabolites Significantly Different Between the AvX and Placebo Groups and the Abundances of Detected Bacterial Taxa in the Group of Overweight/Obese Cats Receiving AvX

Study 1: The results of Spearman’s rank correlation for metabolites significantly different between the AvX and placebo groups, and associated bacterial taxa are summarized in

Table 8. Correlation between concentration of metabolites and abundances of detected bacterial taxa at the genus level in the group of overweight/obese domestic shorthair cats (n = 5) receiving AvX (Study 1) at the end of the 16-week period.

Metabolite	Bacterial Taxa	r Value	p Value
Bacterial taxa significantly positively correlated with tryptophan Bacterial taxa significantly negatively correlated with tryptophan Bacterial taxa significantly positively correlated with indole-3-acetate Bacterial taxa significantly negatively correlated with indole-3-acetate Bacterial taxa significantly positively correlated with glycyl-proline Bacterial taxa significantly negatively correlated with glycyl-proline Bacterial taxa significantly positively correlated with nicotianamine Bacterial taxa significantly negatively correlated with nicotianamine	Bifidobacterium Eubacterium Blautia Roseburia Veillonella Desulfovibrio Veillonella Eubacterium Lactobacillus Roseburia Mitsuokella Bulleidia Rickettsiella Bacteroides Odoribacter Veillonella Desulfovibrio Bifidobacterium Eubacterium Blautia	0.9 1 0.9 0.9 −0.9 −0.9 0.9 −0.9 0.9 0.9 0.9 0.9 0.9 −0.9 −0.9 1 0.9 −0.9 −1 −0.9	p = 0.04 p = 0.005 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.04 p = 0.005 p = 0.04

.

At the end of the 16-week period in the AvX group, a positive correlation was observed between tryptophan concentration and the abundances of the genera Bifidobacterium (r = 0.9, p = 0.04), Eubacterium (r = 1, p = 0.005), Roseburia (r = 0.9, p = 0.04), and Blautia (r = 0.9, p = 0.04). In contrast, negative correlations were found between tryptophan and the genera Veillonella (r = −0.9, p = 0.04) and Desulfovibrio (r = −0.9, p = 0.04).

A positive correlation was identified between indole-3-acetate levels and Veillonella (r = 0.9, p = 0.04), while a negative correlation was observed between indole-3-acetate and Eubacterium (r = −0.9, p = 0.04) at the genus level.

A positive correlation was detected between the glycyl-proline concentration and Lactobacillus (r = 0.9, p = 0.04), Roseburia (r = 0.9, p = 0.04), Mitsuokella (r = 0.9, p = 0.04), Bulleidia (r = 0.9, p = 0.04), and Rickettsiella (r = 0.9, p = 0.04). Genera Bacteroides (r = −0.9, p = 0.04) and Odoribacter (r = −0.9, p = 0.04) were negatively correlated with the level of glycyl-proline.

Additionally, a positive correlation was identified between nicotianamine and genera Veillonella (r = 1, p = 0.04) and Desulfovibrio (r = 0.9, p = 0.04). Negative correlation between Bifidobacterium (r = −0.9, p = 0.04), Eubacterium (r = −1, p = 0.005) Blautia (r = −0.9, p = 0.04), and nicotianamine was noticed.

4. Discussion

The present study evaluated the GM and fecal metabolomic profile following a randomized controlled trial where naturally overweight/obese cats consumed an MH-enriched AvX or placebo for 16 weeks. The GM of overweight/obese cats was also compared to the GM of lean cats. AvX consumption had an impact on GM composition and resulted in differences in microbial metabolite concentrations. In addition, significant differences in the GM of overweight/obese and lean cats have been identified and described. Current studies are continuously adding to our knowledge of the importance of GM in aging and the caloric restriction/mimetic paradigm, which includes the MH-enriched AvX evaluated here.

Studies demonstrating the ability of fecal microbiota transplantation to partially revert obesity-associated host metabolic phenotypes suggest that GM alteration is not merely a consequence but rather a contributing factor of obesity-related metabolic imbalance [55]. The GM can promote metabolic pathways that affect adipose tissue metabolism [56,57].

However, there are inconsistencies in findings among human studies associating GM with obesity, examining specific alterations of the microbial community structure, such as diversity and the Firmicutes/Bacteroidetes ratio [58,59]. Microbial dissimilarities (beta diversity) were observed among the body mass index (BMI) categories of people in many studies, confirming that the microbial population abundances in the gut of obese and non-obese groups are distinct from each other [60]. Several studies revealed different compositions of the GM between obese and lean individuals characterized by a high Firmicutes/Bacteroidetes ratio [36,61]. Based on the results obtained from obese animals and humans, it has been proposed that Firmicutes are more effective in extracting energy from food than Bacteroidetes, thus promoting a more efficient absorption of calories and subsequent weight gain [56,62,63,64,65,66,67]. Accordingly, the high Firmicutes/Bacteroidetes ratio has been frequently considered as a possible hallmark for obesity [68,69,70]. However, other studies failed to find such changes in the Firmicutes/Bacteroidetes ratio in obesity [71,72,73], and contradictory results, including a similar microbiota composition in obese and lean subjects [72], or even an opposite change in the Firmicutes/Bacteroidetes ratio in obesity [74], have been reported.

These conflicting data may be attributable to many factors, including diet [75], different obesity grades [61,76], and the presence of obesity-associated comorbidities [76,77]. Furthermore, it has been recognized that the interlaboratory reproducibility of microbiome sequencing analysis is not optimal. As a result, effort has been made to benchmark its analytical performance in terms of reproducibility and comparability, particularly between different research centers and studies [78,79].

Fecal microbial communities of overweight/obese and lean cats in this study showed a significant difference (beta diversity), which is in accordance with the results of human studies [60]. Additionally, a comparison of similarities among fecal microbial communities revealed significant differences in microbial community structures between day 0 and the last day of our study in the AvX group of cats, suggesting that the composition of their gut microbial population was modified by the consumption of AvX. The abundance of Firmicutes in the group of cats receiving AvX decreased compared to baseline, and the Firmicutes/Bacteroidetes ratio decreased in the AvX group as well, although the reduction did not reach statistical significance. Thus, it is conceivable that the effect of AvX on the microbial community structure at least partially reverted obesity-associated GM composition of overweight/obese cats in this study.

Genus Dialister, a Gram-negative anaerobic bacterium from the phylum Firmicutes, has been identified as a biomarker in the GM of cats receiving AvX in this study. Genus Dialister harbors genes coding for butyrate production [80]. Butyrate concentration was positively associated with the abundance of Dialister spp. in the colon of pigs [81]. Butyrate is considered to have anti-inflammatory potential that might alleviate metabolic complications [82]. The GM of obese children had lower proportions of Dialister compared to the normal weight group in the study of alteration of the GM associated with childhood obesity [83]. Obese adults had lower counts of Dialister compared to lean people in another study [84]. A higher abundance of Dialister has also been associated with lower BMI in people [85]. Diabetic cats exhibited lower abundances of Dialister spp. than lean cats [86]. The bacterium was completely absent in children genetically predisposed to type 1 diabetes mellitus (T1DM) [87]. In addition, genus Dialister was one of the bacteria with abundances significantly higher in lean compared to overweight/obese cats of this study.

Genus SMB53, Gram-positive anaerobic bacteria from the phylum Firmicutes, was found in the small intestine of farmed pigs with high body fatness, whose microbiota was enriched in inflammation-related genes speculated to trigger increased adiposity [88], and was also reported to be associated with obesity in adult people [89] and enriched in obese groups of children [83,90]. An increased relative abundance of SMB53 in the GM of high-fat diet-fed mice rather than mice fed a normal-fat diet was consistent with its positive correlation with host adiposity [91]. In addition, SMB53 was one of the bacterial species detected in high abundances in the GM of obese diabetic mice [92]. Moreover, findings in a mouse model suggested that the genus SMB53 may be an important factor for the abnormal metabolism of individuals with type 2 diabetes mellitus (T2DM) [92]. Thus, the trend of a reduction in the abundance of SMB53 detected in the AvX group of this study at the end of the 16-week period might be viewed as a positive effect of AvX on the composition of the gut microbial population.

An abundance of Roseburia sp., Gram-positive anaerobic bacteria from the phylum Firmicutes, SCFA producers, and an important reservoir of enzyme bile salt hydrolase in the gut, was trending down in the GM of the AvX group of overweight/obese cats in this study at the end of the 16-week period compared to baseline. This bacterium is more commonly detected in adult people with obesity compared to normal weight individuals [84,93], and the GM of overweight/obese children also comprised higher abundances of genus Roseburia compared to the lean group [94]. Percent body fat was positively correlated with an abundance of Roseburia spp. in adult people [95]. In addition, several studies have found a higher abundance of genus Roseburia in the GM of obese dogs when compared to lean dogs [96,97].

Blautia producta, a Gram-positive anaerobic bacterium from the phylum Firmicutes, plays an important role in the metabolism of glucose, which it converts to acetate, lactate, ethanol, and succinate in the gut. In obese children, microbial assemblages were notably enriched in Blautia compared to lean individuals [98]. In addition, Blautia producta was suggested to be related to the obesity model in high-fat-diet-fed mice [99]. Blautia was significantly elevated in both obese and T2DM rats. While obesity and T2DM show distinct differences, results suggest that in both conditions, Blautia species were increased, indicating a possible mechanistic link [100]. This is likely related to disrupted carbohydrate and glucose metabolism with a reduction in butyrate production and an increase in other SCFAs [100]. In line with these results, significantly increased levels of acetic acid and propionic acid in rats fed a high-fat diet have been reported [101]. Acetate is used as a substrate for cholesterol and fatty acid synthesis, thus promoting hypercholesterolemia, hypertriglyceridemia, and the development of liver steatosis, as observed in high-fat-diet-fed rats [100]. Propionate has been shown to promote the inhibition of lipolysis and adipocyte differentiation, leading to increased adiposity [102]. A trend of reduction in the abundance of Blautia producta was detected in the GM of the AvX group of overweight/obese cats in this study at the end of the 16-week period compared to baseline, suggesting a positive effect of AvX consumption on the composition of the gut microbial population.

In a recent study, Acidaminococcus, a Gram-negative anaerobic bacterium from the phylum Firmicutes, was identified as an important factor in distinguishing lean versus obese neutered cats. A comparison of lean and obese neutered cats on the same diet showed that in neutered cats, obesity led to increased abundances of Acidaminococcus [37]. After energy restriction, there was a significant decrease in Acidaminococcus in the same cats [37]. The abundance of Acidaminococcus spp. showed a highly significant positive correlation with obesity in people [84,103]. In addition, Acidaminococcus spp. has been identified as being associated with T2DM in people [104]. Interestingly, a trend toward a reduced abundance of Acidaminococcus sp. in the GM of overweight/obese cats receiving AvX was detected at week 16 of this study compared to placebo.

Additionally, a trend of reduction in the abundance of Akkermansia sp. in the GM of overweight/obese cats of the AvX group was also detected. Akkermansia sp., a Gram-negative anaerobic mucin-degrading bacterium from the phylum Verrucomicrobiota, was reported to be associated with obesity in people [93]. Despite the evidence suggesting the anti-inflammatory effects of A. muciniphila in obesity, diabetes, and colitis, others have failed to demonstrate such health benefits. A. muciniphila-related genes were more abundant in T2DM adults compared with healthy controls. It has also been suggested that A. muciniphila could facilitate intestinal inflammation through mucin degradation [105]. Further studies are needed to decipher the precise role of A. muciniphila in inflammatory-related diseases [106]. Akkermansia is not commonly detected in the fecal microbiota of cats and dogs. An important consideration is whether the presence of mucus degraders in the mucus correlates with its abundance in luminal contents and feces, i.e., whether the shed bacteria are the same or different from those that are fixed into the mucosal layer. This might explain the apparent low concentration of Akkermansia in the feces of cats and dogs because these bacteria may be closely attached to the mucosa, which reduces their loss in feces [107].

The increase in Adlercreutzia, a Gram-positive anaerobe from the phylum Actinobacteria, was significantly correlated with the group of obese people compared to the overweight and normal groups [103]. A trend of reduced abundance of Adlercreutzia sp. in the GM of overweight/obese cats receiving AvX was detected at week 16 of this study compared to placebo, as well as a trend toward decreased abundance of Collinsella aerofaciens, a Gram-positive anaerobic bacterium from the phylum Actinobacteria. In obese children, microbial assemblages were notably enriched in Collinsella aerofaciens [98], and taxon Collinsella is considered a pathobiont. It can affect metabolism through the alteration of intestinal cholesterol absorption, decreasing glycogenesis in the liver, and increasing triglyceride synthesis. Currently, the molecular mechanisms by which Collinsella affects the host metabolism are unknown; however, its abundance has been associated with T2DM, rheumatoid arthritis, and abnormal cholesterol metabolism. In addition, gnotobiotic approaches have shown that the administration of Collinsella reduces the expression of tight junction proteins in enterocytes and stimulates gut leakage, which are features associated with metabolic endotoxemia. This indicates that high Collinsella abundance may contribute to the progression of insulin resistance in people [108].

Genus Prevotella, Gram-negative anaerobic bacteria from the phylum Bacteroidetes, includes branched-chain amino acid producers and mucin-degrading and dietary fiber-fermenting bacteria. An abundance of Prevotella was significantly higher in the GM of the normal-weight children in the study assessing childhood obesity [83] and lower in the microbiota profile of a group of obese adults compared to lean people [84]. A negative correlation was found between BMI and an abundance of Prevotella spp. in people [109]. An abundance of Prevotella sp. was significantly lower in overweight/obese compared to lean cats in this study. A significant negative correlation between the genus Prevotella and plasmatic levels of leptin and a positive correlation between Prevotella spp. and plasmatic ghrelin have been noticed in people and kittens [110,111]. The Prevotella genus has large genetic variability between different species and strains [112]. This may be an explanation for conflicting results regarding whether the presence of Prevotella in the gut should simply be viewed as a reflection of dietary composition or if its presence is beneficial or harmful to the host.

It might be of interest that diabetic cats were shown to have a decreased proportion of the Prevotella genus in the GM, since a negative correlation was detected between the Prevotellaceae family and serum fructosamine levels [86]. Prevotella was also found in smaller proportions in T1DM cases compared to healthy people [106]. A study on germ-free mice showed improved glucose tolerance and homeostasis when they received daily gavage with live Prevotella copri isolated from human feces [113,114].

An abundance of Turicibacter, a Gram-positive anaerobic bacterium from the phylum Firmicutes (SCFA producers), was significantly lower in overweight/obese compared to lean cats in this study. A lower abundance of Turicibacter spp. was found in the GM of obese compared to lean rodents, and lower abundance was associated with more significant systemic inflammation in obese rodents in the same study [115]. When the microbiota of diet-induced obese mice was compared to that of lean mice, significantly less Turicibacter was observed in obese mice [116,117]. In addition, the relative proportions of Turicibacter spp. in obese dogs were lower compared to a lean group [96]. In a study evaluating time-dependent alterations of the GM during the progression of diabetes in rats, random blood glucose was negatively correlated with an abundance of Turicibacter spp. [118]. Cats with inflammatory bowel disease (IBD) or small-cell lymphoma all displayed lower abundances of Turicibacteraceae [119]. Data from several studies showing a depletion of Turicibacter in animal models of IBD, proposing the hypothesis that Turicibacter is an anti-inflammatory taxon [120,121]. Administration of liraglutide (GLP-1 receptor agonist) was associated with a higher abundance of Turicibacter spp. in the GM of rats. The results suggest that liraglutide could modulate the composition of GM, leading to a leaner-related profile that is consistent with its weight loss effect [104,122].

Genus Clostridium includes Gram-positive anaerobic luminal bacteria from the phylum Firmicutes. Genus Clostridium has the enzymatic capacity to modify bile acids [123]. Clostridia are also the most prevalent species involved in the hydrolysis of amino acids. Similarly to our results, reduced proportions of Clostridium spp. were observed in obese neutered cats compared to lean neutered cats [37]. Previous studies have found Clostridium to positively correlate with carbohydrate oxidation and negatively correlate with fat oxidation [124]. In the GM of restricted eaters with access to exercise, a significant increase in the abundance of Clostridium was found in a rat study [125]. The abundance of Clostridium was negatively correlated with serum leptin levels in rats [125].

Genus Veillonella, Gram-negative anaerobes from the phylum Firmicutes, SCFA producers, can produce lactate, which could induce a decrease in the colonic pH and promote the production of butyrate [126]. In addition, an inverse correlation between Veillonella spp. and cholesterol levels was observed in people with severe obesity [127]. Interestingly, the abundance of Veillonella sp. was significantly lower in overweight/obese compared to lean cats in this study.

The GM of overweight children contained a lower abundance of Catenibacterium spp., Gram-positive anaerobic bacteria from the phylum Firmicutes (SCFA producers), compared to the normal children [128], similar to the overweight/obese cats in this study. Inverse relationship between obesity and an abundance of Eubacterium biforme has been found in dogs [129], which is in accordance with results of this study. Obese dogs had a lower abundance of E. biforme compared to overweight and lean dogs [129]. The abundance of Desulfovibrio was significantly lower in overweight/obese cats compared to lean cats in this study, which is similar to the significantly decreased abundance of Desulfovibrio spp detected in the GM of obese compared to normal-weight children [64]. Genus Desulfovibrio includes Gram-negative, anaerobic bacteria from the phylum Deltaproteobacteria, producing hydrogen sulfide gas as a terminal by-product of their metabolic activity. Genus Desulfovibrio was found in significantly lower abundances in obese compared to normal-weight people [84,130]. The abundance of Desulfovibrio decreased in diet-induced obese mice [118] and was enriched after liraglutide administration [122].

A higher abundance of Phascolarctobacterium was associated with a greater likelihood of achieving weight loss in a study comparing individuals who were and were not able to achieve 5% weight loss [131]. In people, a higher abundance of Phascolarctobacterium, Gram-negative anaerobes from the phylum Firmicutes (SCFA producers), was detected in the insulin-sensitive group compared to the insulin-resistant group [95]. In addition, a negative correlation between Phascolarctobacterium and percent body fat was found in adults [95]. Phascolarctobacterium has been identified as a biomarker in the GM of lean cats in this study.

People with obesity had lower counts of genus Ruminococcus, Gram-positive anaerobes from the phylum Firmicutes and SCFA producers [83,84]. Genus Ruminococcus has been identified as a biomarker in the GM of lean cats in this study. A high-fat diet led to a decrease in the abundance of Ruminococcus spp. in mice [132]. Since a number of Ruminococcus species are known to be associated with metabolic diseases, the identification of Ruminococcus to the species level might be critical for a further understanding of the relationship between genus Ruminococcus and obesity/diabetes and its effect on the development of metabolic diseases [118].

Genus Eubacterium, Gram-positive anaerobic bacteria from the phylum Firmicutes, belongs to the luminal bacteria and SCFA producers with the enzymatic capacity to modify bile acids. A high-fat diet led to a decrease in the abundance of Eubacterium spp. in mice [133]. Similarly to the result of this study where Eubacterium has been identified as a biomarker in the GM of lean cats, children and adults with obesity had lower counts of Eubacterium in the GM profile compared to lean groups [83,84].

A higher abundance of genus Coriobacterium, Gram-positive anaerobes from the phylum Actinobacteria, was detected in obese human patients with T2DM compared to healthy individuals [30], which is in accordance with our results, indicating that overweight/obesity was associated with increased abundances of Coriobacterium sp. Also similar to our results, a higher abundance of Ruminococcus gnavus, a mucin-degrading organism, was observed in obese compared to lean individuals in a rat study [134]. This species was also shown to be enhanced in people with diverticulitis [135], which is commonly associated with obesity [136]. Enterococcus is a Gram-positive mucin-degrading facultative anaerobe from the phylum Firmicutes [137]. Obese children and adults had higher counts of Enterococcus compared to lean children and adults [84,98]. The counts of Enterococcus in obese dogs were higher than in lean dogs [96]. In our study, genus Enterococcus has been identified as a biomarker in the GM of overweight/obese cats.

Obesity is associated with a significantly decreased production of microbial indole catabolites of tryptophan (MICT), notably indole-3-acetate (IAA), indole-3-lactic acid, and indole-3-propionic acid [25,138]. Fecal levels of MICT, including IAA, are diminished in obese or diabetic individuals [139]. These alterations rely on changes in microbiota composition and function [25,138]. The production of IAA increases the abundance of CD103+/CD11c+ cells in the immune system [140], which are critical for maintaining intestinal immune homeostasis and inducing tolerogenic immune responses [141]. IAA binds to the aryl hydrocarbon receptor on dendritic cells and drives the production of IL-22. IAA promotes the production and accumulation of IL-35 + B regulatory cells in intestinal tissues, the key immune regulators of many diseases [142,143]. IL-35 expression is dysregulated in inflammatory diseases such as IBD, obesity, T1DM, and autoimmune hepatitis [144].

A study comparing the fecal amino acid profiles of treatment-naïve pediatric IBD patients with matched healthy controls found that the level of fecal tryptophan was increased in the IBD group. An elevated concentration of tryptophan in the fecal samples of patients with IBD suggests a role for decreased absorption/increased loss by inflamed intestines and for alterations in microbial tryptophan metabolism as potential causes for these differences [145]. Similarly, an increased fecal concentration of tryptophan has been observed in cats with chronic enteropathy [24].

In this study, AvX administration induced a reduction in tryptophan concentration and an increase in IAA in fecal samples. The mechanism is unknown; however, it is conceivable that the effect of AvX on the microbial community structure, abundances of specific bacterial species of the GM, and their metabolic activity could help to modulate microbial tryptophan metabolism and thus increase the production of IAA and decrease the fecal concentration of tryptophan. The effect on intestinal mucosal health and inflammation, contributing to improved absorption and reduced tryptophan loss through the GI mucosa, can also be considered.

Since genera Blautia and Roseburia have been reported to be more prevalent in individuals with obesity [84,94,98,99], the observed decrease in the abundance of these genera, along with decreased tryptophan levels in overweight/obese cats receiving AvX at the end of this study, may indicate a potentially important and beneficial association. Additionally, the concurrent negative correlation between the genus Veillonella and tryptophan levels, coupled with a positive correlation with indole-3-acetate, warrants further attention, as certain Veillonella species are known to metabolize tryptophan [146].

A negative correlation was also observed between the abundance of Desulfovibrio and tryptophan levels, which is noteworthy given that Desulfovibrio abundance was significantly lower in overweight/obese cats compared to lean cats in this study.

The genus Eubacterium comprises a phylogenetically diverse group of bacteria, with species exhibiting a broad spectrum of metabolic capabilities. These bacteria contribute to important intestinal functions such as fermentation, vitamin synthesis, and immune modulation. However, these roles can vary considerably between species and even within individuals over time [147]. Thus, the observed positive correlation with tryptophan and negative correlation with indole-3-acetate levels warrant further investigation.

No adverse effects were observed in association with the administration of AvX. Significant weight loss was not achieved; thus, an adjustment of the administered dose and the length of administration of AvX could be given consideration.

Limitations of our study include a small sample size. It is possible that small differences in the Firmicutes/Bacteroidetes ratio, individual bacterial taxa in Study 1, and in concentrations of some fecal metabolites, including MICT, between groups might have been missed. There was a significant difference in the age of lean and overweight/obese cats in Study 2. Longitudinal studies in cats concluded that GM is predominantly determined by age when diet and environment are controlled. They identified that at 42 weeks of age, one of the most abundant genera was Prevotella [148,149]. The age range of our group of lean cats was 67–78 weeks. It is possible that the difference in beta diversity between lean and overweight/obese cats and the higher abundance of Prevotella spp. in this group was not associated solely with their lean condition, but it might have been partially attributed to the age of these cats. However, genus Enterococcus is predominant in pre-weanling and young cats, and the number of enterococci decreases in elder individuals [148,149]. If age was the major determinant affecting the GM, enterococci should be higher in our lean cats than obese cats. Thus, it appears that leanness/obesity had a stronger impact on GM than age in this study since diet and environment were the same for all cats.

5. Conclusions

Evidence from animal and human studies suggests that obesity and related diseases are linked to shifts in the GM and altered metabolite production. The GM is a dynamic ecosystem that, while relatively stable in core composition, adapts to internal and external factors. Taxonomic changes can influence substrate availability, microbial interactions, and metabolic pathways, ultimately affecting host health. Our results indicate that AvX consumption can affect the GM composition and fecal microbial metabolite profiles in naturally overweight/obese cats, potentially offering benefits for managing overweight/obesity and associated comorbidities. There are significant differences in the GM between naturally lean and overweight/obese cats, likely related to better energy-harvesting abilities of the host that might be the target of future therapeutic interventions. However, additional larger-scale studies are needed to evaluate the effects of AvX on the GM, fecal metabolomic profile, and microbial tryptophan metabolism in cats with overweight/obesity and associated comorbidities.