TYPE Original Research PUBLISHED 22 August 2023 DOI 10.3389/fvets.2023.1213287 Age-associated changes in OPEN ACCESS intestinal health biomarkers in dogs EDITED BY Shengfa F. Liao, Mississippi State University, United States REVIEWED BY Zhongyue Yang, Anna Fernández-Pinteño1*, Rachel Pilla2 , Xavier Manteca3 , University of California, Davis, United States Emily Coffey, Jan Suchodolski2 , Celina Torre1 and Anna Salas-Mani1 University of Minnesota Twin Cities, 1 United States Department of Research and Development, Affinity Petcare, L’Hospitalet de Llobregat, Spain, 2 Gastrointestinal Laboratory, Department of Small Animal Clinical Sciences, Texas A&M University, *CORRESPONDENCE College Station, TX, United States, 3 School of Veterinary Science, Universitat Autònoma de Barcelona, Anna Fernández-Pinteño Bellaterra, Spain anfernandez@affinity-petcare.com RECEIVED 27 April 2023 ACCEPTED 13 July 2023 The gut microbiome is critical for maintaining host health. In healthy humans, the PUBLISHED 22 August 2023 aging process is one of the main factors modulating the changes in the intestinal CITATION Fernández-Pinteño A, Pilla R, Manteca X, microbiota. However, little is known about the relationship between gut health, Suchodolski J, Torre C and Salas-Mani A (2023) microbiota, and the aging process in dogs. The present study aims to explore Age-associated changes in intestinal health the differences in the intestinal microbiota and intestinal health based on fecal biomarkers in dogs. Front. Vet. Sci. 10:1213287. doi: 10.3389/fvets.2023.1213287 biomarkers in a population of dogs of different ages. The study involved 106 dogs of different breeds aged between 0.2 and 15 years categorized as senior (>7 years; COPYRIGHT © 2023 Fernández-Pinteño, Pilla, Manteca, n = 40), adult (2–7 years; n = 50), and junior (<2 years; n = 16). Fecal samples Suchodolski, Torre and Salas-Mani. This is an were collected during the same period at the same facilities. The analysis included open-access article distributed under the terms the following gut health indicators: 16S rRNA gene sequencing to investigate the of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction differences in the fecal microbiota; qPCR to determine the dysbiosis index; fecal in other forums is permitted, provided the short-chain fatty acid concentrations; fecal calprotectin; and immunoglobulin A. original author(s) and the copyright owner(s) Beta diversity analysis revealed a significant difference with a small effect size (p are credited and that the original publication in this journal is cited, in accordance with = 0.003; R = 0.087) among age categories based on the unweighted UniFrac accepted academic practice. No use, metric, but no significance was observed based on the weighted UniFrac metric or distribution or reproduction is permitted which Bray–Curtis distances. There were no significant differences in the alpha diversity does not comply with these terms. measures or the fecal dysbiosis index among age categories. Senior dogs had significantly higher relative abundance proportions in phyla Bacteroidota and Pseudomonadota and the genus Faecalibacterium, but not on qPCR analysis. At the family level, Ruminococcaceae, Uncl. Clostridiales.1, Veillonellaceae, Prevotellaceae, Succinivibrionaceae, and Bacteroidaceae abundances were higher in the senior category than in the adult and/or junior categories. Relative proportions, but not concentrations of fecal acetate, were higher in the senior category, while butyrate, isovaleric acid, and valeric acid were lower. The valeric acid concentration was significantly lower in the senior category than in the adult category. Calprotectin and immunoglobulin A levels did not differ significantly across groups. In conclusion, this study observed multiple minor changes in the fecal microbiota composition and the relative amount of short-chain fatty acids in dogs among different age groups, but studies in larger populations representative of all ages are warranted to refine the present results. KEYWORDS fecal microbiota, canine, aging, intestinal health, nutrition Frontiers in Veterinary Science 01 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 1. Introduction TABLE 1 Descriptive information regarding the age distribution of dogs sorted by age category. The gut microbiome is defined as the entire collection of Junior Adult Senior microorganisms living in the gastrointestinal tract, dominated by n = 16 n = 50 n = 40 bacteria and complemented by commensal populations of fungi, Age (in years) 0.92 [0.23; 1.14] 5.00 [4.58; 7.11] 8.25 [8.15; 9.06] viruses, archaea, and protists (1). It has the largest microbiota population in the body when compared to other colonized organs. The intestinal microbiota is estimated to be composed of 1013 - 1014 cells (2), and its genome is 150 times larger than the certain pathologies. Garrigues et al. (13) recently reviewed the human genome. development of the gut microbiota during the early stages of The gut microbiome plays an important role in the canine life, supporting observed changes in bacterial communities physiological and pathological state of its host. It participates from day 2 of age to up to 52 weeks. On day 2 after birth, gut in multiple metabolic functions, protects against pathogens, and microbiota richness increases, and from day 2 to 21, Bacillota plays a crucial role in the immune response. These processes predominance is substituted by a codominance of Bacillota, are critical to maintaining the health status of the host (3). The Bacteroidota, and Fusobacteriota (14). Puppies in the first few composition of the microbiota is crucial to maintaining balanced weeks of life have immature microbiota, characterized by an gut functionality. An imbalanced or disrupted microbiota is related increased dysbiosis index (DI), Clostridium difficile abundance, to several intestinal and extraintestinal diseases (4). and decreased Clostridium hiranonis compared to adult dogs (15). There is an increasing number of studies evaluating the Approximately after 4–6 months of age, the microbiota resembles different elements modulating the intestinal microbiota (5–8). The that observed in adult dogs and remains largely stable in adulthood main factors can be classified as external and internal, including, (13). However, the relationship between the intestinal microbiota, for example, diet and pharmacological treatments and age and gut health conditions, and the aging process in older dogs is not genetics, respectively. In addition, pathological disorders (e.g., well established. Studies in this field are needed first to understand inflammation and type 2 diabetes) may induce deviations in the gut this relationship and then to explore effective strategies to improve microbiota, causing an imbalance in the gut microbiota, defined as the quality of the aging process in dogs from this perspective. dysbiosis. Among all these factors, age is one of the most important The present study aimed to investigate the differences in variables that must be considered when studying the evolution and the intestinal microbiota and intestinal health based on fecal changes of the gut microbiota. biomarkers in a population of dogs of different ages, with a special Aging and its relationship with gut health and gut microbiota focus on the senior category. are currently being explored in mammals, and they are already well-established in humans. Badal et al. (9) summarized the current knowledge of the human gut microbiota, considering composition, 2. Materials and methods function, and metabolic products from the microbiota in relation to aging and lifespan. Changes in the human microbiota related to age All samples collected from dogs were in accordance with the have been described and are apparently associated with the host’s 2010/63/UE directive, and any additional procedures to the dogs’ health scenario. For example, bifidobacteria belonging to members daily routine were undertaken. Before the study, the protocol was of the genus Bifidobacterium, which are considered a beneficial shared with the Affinity Ethical Committee to ensure good practices bacterial group, decrease during the shift from middle age to old and the animals’ welfare. All dogs were housed at the Affinity age. Contrarily, Clostridium perfringens, lactobacilli, enterococci, Nutrition Center (Barcelona, Spain). and Enterobacteriaceae increase during the aging process (10). In contrast to the human microbiota, bifidobacteria may not play an important role in dog gut health, according to the results reported 2.1. Animals by Masuoka et al. (10), in which bifidobacteria were only found in half of the youngest dogs and none of the adult dogs sampled, The study involved 106 multi-breed dogs (n = 53 male dogs; although differences in methodologies need to be considered when n = 53 female dogs) aged between 0.2 and 15.0 years (mean age interpreting these results. = 6.3 years). The dog breeds that participated in the study were The existence of a core microbiota in healthy dogs has as follows: Beagle, Bichon Maltese, Jack Russell Terrier, Boxer, already been described (11). Moreover, Ziese and Suchodolski German Shorthaired Pointer, Poodle, Chihuahua, Cocker Spaniel, (12) associated shifts in the canine fecal microbiota with Dalmatian, German Shepherd, Spanish Water Dog, Miniature Pinscher, Andalusian Podenco, Pointer, Pomeranian, Miniature, Schnauzer, Shih Tzu, Brittany Spaniel, and Yorkshire Terrier Abbreviations: DI, Dysbiosis index; J, Junior; A, Adult; S, Senior; SCFA, Short [detailed information is in the Supplementary Table 1]. Animals Chain Fatty Acid; cCP, Calprotectin; IgA, Immunoglobulin A; PCoA, Principal were grouped into junior (J), adult (A), and senior (S) categories, Coordinate Analysis; DM, Dry Matter; GC-MS, Gas Chromatography-Mass considering the respective age ranges: up to 2 years, from 2 to 7 Spectrometry; RIA, Radio Radioimmunoassay; ANOSIM, Analysis Of Similarity; years, and over 7 years (Table 1). R, Rotation; S, Stable; Emmeans, Estimated Means; FDR, False Discovery Rate; Animals lived in kennels in groups of between two and seven ANCOM, Analysis of Composition of Microbiomes; IBD, Inflammatory Bowel individuals, depending on the size of the dogs. Dogs living in pairs Disease. shared an area of 15 m2 , whereas those living in larger groups Frontiers in Veterinary Science 02 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 shared an area of 57–60 m2 . All the animals had free access to the from sample collection to separation into aliquots, were sterile and outdoor and indoor parts of their kennels, and they spent between 2 meant for single use. and 4 h per day in a bigger outdoor park where they could socialize Feces were divided into two different aliquots according to the with a larger group of dogs. analysis to be conducted. For the microbiota analysis, between 0.5 The water supply was ad libitum. Non-specific diets were and 1 g of feces were prepared in a 1.5-ml sterile, RNAse-free tube. manufactured and offered because of the study. Dogs were fed For short-chain fatty acid (SCFA) analysis, calprotectin (cCP) and commercially available, dry-complete, and balanced diets adequate immunoglobulin A (IgA) were filled in one sterile stool tube with to their needs. Furthermore, 54% of dogs were fed a constant diet, 5 g of feces. Both tubes were stored at −80◦ C before being sent whereas the diet was changed for 46% of dogs before and during and analyzed at the Small Animal Clinical Sciences Department at the collection period. All diets met all requirements for protein, Texas A&M University (College Station, TX). fat, vitamins, and minerals recommended by FEDIAF (16). Most of the dogs in the study (89%) were fed diets formulated with the following composition range: 7–8% moisture, 22–30% protein, 15– 2.3. Fecal biomarkers 21% fat, and 2–5% fiber. The rest of the dogs (11%) were out of range because of the protein, fat, and/or fiber composition [detailed 2.3.1. Microbiota analysis information is in the Supplementary Table 2]. Mineral and vitamin Illumina sequencing of the bacterial 16S rRNA levels were similar in all the diets that were present throughout genes was performed using primers 515F (5′ - the study. GTGYCAGCMGCCGCGGTAA) (18) to 806RB (5′ - The inclusion criteria for enrollment were generally healthy GGACTACNVGGGTWTCTAAT) (19) at the MR DNA laboratory animals and comprised sterilized male and female dogs of all pure (Shallowater, TX, USA). breeds and ages. Based on the literature, the exclusion criteria Sequences were processed and analyzed using the Quantitative included the presence of clinical disease or any pharmacological Insights Into Microbial Ecology 2 (QIIME 2) (20) v 2021.8 pipeline. treatment potentially interfering with the gut microbiota and/or The raw sequences were uploaded to the NCBI Sequence Read intestinal health. Specifically, no antibiotics, NSAIDs, PPIs, Archive under the BioProject identification PRJNA901473. In prebiotics, probiotics, or deworming treatments were administered brief, the sequences were demultiplexed, and the ASV table was during or 4 weeks before the sampling. None of the animals created using DADA2 (21). Before downstream analysis, sequences participating in this study were fed diets containing probiotics. assigned as chloroplast, mitochondria, and low abundance ASVs, Dogs with watery, soft, or unformed stools in three consecutive containing <0.01% of the total reads in the dataset, were removed. samples were excluded from the study. No clinical signs were All samples were rarefied to even sequencing depth, based on the compatible with gastrointestinal disorders (such as vomiting, lowest read depth of samples, to 21,025 sequences per sample. regurgitation, and hyporexia) were observed in any of the animals Alpha diversity was measured with the Chao1 (richness) and participating in the study. Shannon diversity metrics within QIIME2. Beta diversity was evaluated with the unweighted and weighted phylogeny-based UniFrac (22) distance metric (measures that consider phylogenetic 2.2. Fecal collection information) and the Bray-Curtis distance metric and visualized using Principal Coordinate Analysis (PCoA) plots, generated Samples were collected during 3 months (March, April, and within QIIME2. May) without interfering with the dogs’ daily routines. Feces were The dysbiosis index (logDNA) is based on a validated algorithm collected through direct observation while the animals were kept in that considers a panel of eight bacterial groups identified by a their kennels or outdoor recreational areas. The majority of samples Quantitative PCR assay for total bacteria, Faecalibacterium spp., were obtained during the first month, but due to the volume of Turicibacter spp., Escherichia coli, Streptococcus spp., Blautia spp., samples and the physiologically atypical routines of some animals, Fusobacterium spp., and Clostridium hiranonis (23). A DI of <0 the collection extended up to 3 months. In all cases, individualizing was defined as normal, a DI between 0 and 2 was defined as mild the animals and/or using cages to collect the samples were avoided. to moderate microbiota shift, and a DI of >2 was considered One stool sample was collected per dog, and it was first scored significant dysbiosis. according to an adapted five-point scale of fecal consistency (0 = watery stool; 25 = soft unformed stool; 50 = soft formed and moist stool retaining shape when being collected; 75 = hard formed stool 2.3.2. Intestinal health indicators remaining shaped but soft; and 100 = hard dry stool). The scoring Concentrations of SCFAs (acetate, propionate, butyrate, system was adapted from the five-point scale previously described isobutyric acid, isovaleric acid, and valeric acid) in feces were by Strickling et al. (17). measured using a stable isotope dilution gas chromatography-mass If fecal consistency was equal to or lower than 50, the sample spectrometry (GC-MS) assay with some modifications (24). To was not collected, and the researchers waited for a higher fecal consider the difference in water content between fecal samples, score. The entire stool or enough quantity to fill the sterile fecal the final concentrations of fecal SCFAs were adjusted by fecal dry collection tube with 250 ml was taken. External contaminants matter (DM) and expressed as µmol/g of fecal DM (25). (stones, grass, sand, etc.) were discarded. All samples were The fecal IgA (mg/g) was quantified by using the commercial processed within 4 h after the deposition. All the materials used, ELISA kit for canine IgA determination (Bethyl Laboratories, Frontiers in Veterinary Science 03 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 TABLE 2 Results of the quantitative real-time PCR (abundance expressed as LogDNA) and values of the calculated dysbiosis index (ratio). Junior Adult Senior emmeans ± SE emmeans ± SE emmeans ± SE p-value Dysbiosis index −4.51 ± 0.642 −3.98 ± 0.370 −3.49 ± 0.518 0.468 Universal LogDNA 10.9 ± 0.067 11.0 ± 0.038 11.0 ± 0.054 0.723 Faecalibacterium 5.43 ± 0.285 5.44 ± 0.164 5.76 ± 0.230 0.476 Turicibacter 7.23 ± 0.237 7.74 ± 0.137 7.53 ± 0.192 0.125 Streptococcus 4.57 ± 0.406 5.10 ± 0.235 5.31 ± 0.328 0.362 E.coli 4.89 ± 0.441 5.12 ± 0.254 5.26 ± 0.356 0.814 Blautia 10.7 ± 0.088 10.5 ± 0.051 10.6 ± 0.071 0.124 Fusobacterium 8.91 ± 0.189 8.81 ± 0.109 8.91 ± 0.152 0.774 Clostridium hiranonis 6.80 ± 0.91 6.86 ± 0.523 6.81 ± 0.073 0.729 b a a,b Bifidobacterium 4.09 ± 0.530 6.04 ± 0.306 5.35 ± 0.428 0.004 Bacteroides 5.72 ± 0.195 5.84 ± 0.113 6.00 ± 0.158 0.525 Lactobacillus 4.88 ± 0.454 5.58 ± 0.262 6.11 ± 0.367 0.118 Different superscript letters indicate significant differences between groups (p-value < 0.05). Montgomery, TX, USA). The quantification of the fecal cCP (ng/g) where animals were allocated. Finally, as for the breed, the was also analyzed by using the ELISA kit previously validated in dogs were categorized into two groups: beagles and non-beagle dogs (26, 27). dogs. The different breeds were not statistically analyzed due to the high number of different breeds but due to the low number of dogs within each breed; however, the dog size was 2.4. Statistical analysis introduced in the statistical analysis to reduce the heterogeneity of dogs participating in the study. The response variables For the statistical analysis of microbiota data, an analysis considered in the general statistical analysis were as follows: fecal of similarity (ANOSIM) test within the PRIMER 7 software consistency; SCFA concentration (acetate, propionate, butyrate, package (PRIMER-E Ltd., Luton, UK) was performed to analyze isobutyric acid, isovaleric acid, valeric acid, and total SCFA); significant differences in microbial communities and the size SCFA relative amounts (acetate, propionate, butyrate, isobutyric effect (R-values between 0 and 1; a higher R-value indicates acid, and isovaleric acid); calprotectin; immunoglobulin A; alpha a larger size effect) among age groups (J, A, and S). All diversity (Chao 1 and Shannon diversity); and finally, from qPCR, datasets were tested for normality by performing the Shapiro– total bacteria, Faecalibacterium, Turicibacter, Streptococcus, E. coli, Wilk test (JMP Pro 11, SAS Software Inc.). The Kruskal– Blautia, Fusobacterium, Clostridium hiranonis, Bifidobacterium, Wallis test was performed (Prism v. 9, GraphPad Software Inc.), Bacteroides, Lactobacillus, and dysbiosis index. followed by a post-hoc Dunn’s multiple comparison tests, to For the general statistical analysis, a first summary statistic determine the age group differences in bacterial taxa (including was performed, in which quantitative variables were analyzed phylum, class, order, family, genus, and species). All p-values using the mean and standard deviation. Qualitative variables were adjusted for multiple comparisons using Benjamini and were tested using relative and absolute frequencies. The Hochberg’s False Discovery Rate (28) at each taxonomic level, existence of differences among age groups was tested by and an adjusted p-value of <0.05 (q-value) was considered performing the appropriate tests (ANOVA and Kruskal– statistically significant. Wallis test), considering the equality among groups’ null For the general statistical analysis, the model included hypothesis. The compliance of the application criteria was demographic information as an explanatory variable and fecal assessed by performing the Shapiro–Wilk normality test. The indicators as response variables. The primary explanatory variable relationship between quantitative variables was analyzed using was age, which was grouped into the three different categories Spearman’s correlation. described. The potential explanatory variables included the To analyze the relationship between the age categories and fecal following: body weight, sex, “feeding routine,” “housing,” and markers, the appropriate linear model was considered, including “breed.” Two categories were established for feeding routines: potential explanatory variables such as body weight, sex, feeding rotation (R), if animals were changing the type of diet in routines, and housing breed. Estimated means (emmeans) for age periods shorter than 2 weeks, and stable (S), if animals were categories were calculated using the adjusted model. Pairwise post- eating the same diet for at least 4 weeks. No animals had a hoc comparisons were also performed, and the model validation change of diet in the 2–4 week range. Each animal was only was analyzed by performing a graphical residual analysis. For the assigned to one category during the whole study. Housing was general statistical analyses, differences were considered significant the variable defined to identify the three different buildings with a p-value of <0.05. Frontiers in Veterinary Science 04 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 3. Results 3.1. Quantitative real-time PCR analysis and dysbiosis index The results from the qPCR analysis are summarized in Table 2. Bifidobacterium abundance was different between J and A, being higher in A. However, no difference was found when considering S vs. J and A. The calculated dysbiosis index was not different among the age categories (Figure 1). In J, 0 of 16 (0%) dogs had a DI of >0, while in A and S, 8 of 50 (16%) and 5 of 40 (13%) dogs had a DI of >0. FIGURE 2 3.2. 16s rRNA sequencing 3D beta diversity patterns of canine fecal microbiota comparing Adult (red), Junior (blue) and Senior (green) categories based on unweighted UniFrac distance. A total of 3,856,065 quality bacterial 16S rRNA sequences were obtained from the 106 fecal samples analyzed. The range count per sample was between 21,030 and 60,827 (median: 33,688 and mean: 36,378). The beta diversity was analyzed by followed by Actinomycetota, Fusobacteriota, Bacteroidota, the weighted and unweighted UniFrac distances and the Bray– and Pseudomonadota. The same distribution, considering Curtis distance. The unweighted UniFrac analysis of similarities, relative abundance, was described in the three age categories. which only considers the presence or absence of individual taxa, Significant differences were found for the phylum Bacteroidota revealed significant differences in the microbial communities and Pseudomonadota, in which S had a higher abundance among age categories (PCoA plot shown in Figure 2). However, than A and J (Table 4). However, the corrected q-value did not no significance was observed based on weighted UniFrac metric reach significance. or Bray–Curtis distances (Table 3). Alpha diversity indices (Chao1 and Shannon) were not significantly different among the age groups (Figures 3, 4). 3.2.2. Family taxonomic level Falling into the main phylum (Bacillota), Ruminococcaceae, Uncl. Clostridiales.1, Veillonellaceae, and Lachnospiraceae families 3.2.1. Phylum taxonomic level were significantly different among the age categories (Figures 5– When the bacterial relative abundance was studied at 8). Specifically, the abundance of Ruminococcaceae and Uncl. the tax level, the major phylum identified was Bacillota, Clostridiales.1 was higher in the S category than in the A category, and Veillonellaceae showed a higher abundance in the S category than in the J category. Lachnospiraceae were significantly higher in the J category than in the A and S categories and did not differ significantly. Within the Bacteroidota phylum, the relative abundance of Bacteroidaceae and Prevotellaceae was different among the age categories; Bacteroidaceae was higher in the S category than in the A and J categories, and Prevotellaceae was higher in the S category than in the A. For Pseudomonadota, the relative abundance of Succinivibrionaceae was higher in the S category than in the A category. Finally, the Bifidobacteriaceae family (Actinomycetota) was significantly higher in the A category than in the J category; however, the adjusted q-value was not significant (Table 5). 3.2.3. Genus taxonomic level The two predominant genera identified in the dogs’ population belonged to the phylum Bacillota, Uncl. Clostridiaceae.1, and Allobaculum. Within Bacillota, Faecalibacterium, and Uncl. Clostridiales.3 showed the highest values in S, being significantly FIGURE 1 different when compared with A (p = 0.006 and 0.008, A representative plot of the dysbiosis index (DI) split by age category. respectively). The relative abundance observed in the genus Faecalibacterium was 18.6 times higher in the S category than Frontiers in Veterinary Science 05 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 TABLE 3 Results of beta diversity analysis including unweighted UniFrac, weighted UniFrac, and Bray–Curtis distances. Senior vs. Adult vs. Senior vs. Adult Senior vs. Junior Junior vs. Adult Junior R p-value R p-value R p-value R p-value Unweighted UniFrac 0.087 0.003 0.071 0.003 0.150 0.018 0.070 0.117 Weighted UniFrac −0.023 0.759 0.047 0.024 −0.145 0.993 −0.127 0.970 Bray–Curtis 0.001 0.459 0.037 0.028 −0.077 0.863 −0.045 0.715 The table shows the sample statistic (R) and the significance level of the sample statistic (p-value < 0.05), considering the three age categories and the pairwise test comparison. FIGURE 3 FIGURE 4 Chao 1 alpha diversity index plot by category. Shannon alpha diversity index plot by age category. in the A category. However, the difference in Faecalibacterium category (p = 0.016); however, the adjusted q-value was was not confirmed by a targeted qPCR. Within this genus, the not significant. The relative abundance of Fusobacterium relative abundance of F. prausnitzii was 18.4 times higher in the (phylum Fusobacteriota) was higher in the S category than S category (median 1.3% of sequences) than in the A category in the A category when considering only the significance (median 0.07%; p = 0.006 and q = 0.030). Detailed information of the p-value (p = 0.045), but this was not confirmed by at the species level is in the Supplementary Table 3. In relation to targeted qPCR. Detailed information at the genus level is in the Phascolarctobacterium abundance, S was higher than the A and J Supplementary Table 4. categories (p = 0.011). Although the S category also showed higher values for Ruminococcus when compared with A (p = 0.035), the adjusted q-value did not reach significance. Contrarily, the abundance of Eubacterium, Uncl. Erysipelotrichaceae, and Uncl. 3.3. Fecal SCFAs, IgA, and cCP Lachnospiraceae was not significantly different when the S category vs. the J and A categories were considered. However, J showed Valeric acid was significantly lower in the S category than in significantly higher values for the three genera when compared the A category. However, the concentrations for total SCFA were with the A category (p = 0.005, p = 0.001, and p = 0.002). numerically higher (but not significant) in the S category and lower Specifically, J showed 8.6 and 7.2 times more relative abundance in the A and J categories (Table 6). Acetate, butyrate, isovaleric in Eubacterium and Uncl. Erysipelotrichaceae genera, respectively, acid, and valeric acid reached significant results when they were compared with A. Considering the Uncl. Peptostreptococcaceae analyzed as relative percentages (Table 6). In the post-hoc analysis, genus, the relative abundance was higher in J than in S and A the S category showed higher values than A for acetate (p < 0.001) categories (p = 0.009). and lower values for butyrate, isovaleric acid, and valeric acid (p The relative abundance of Bacteroides (phylum Bacteroidota) = 0.001, p = 0.045, and p = 0.026, respectively). For isovaleric was 24.3 and 10.4 times higher in the S category than in acid, the S category had lower values than J. Finally, no significant the J and A categories, respectively (p < 0.001). Prevotella differences among age categories were found for the fecal IgA and (Prevotellaceae family) was higher in S than in the A cCP concentrations (Table 7). Frontiers in Veterinary Science 06 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 TABLE 4 Fecal microbiota composition (% of relative abundance from rarefied data) at the phylum level split by age category. Phylum Junior Adult Senior Junior vs. Adult vs. Senior Median Range Median Range Median Range p-value q-value a,b a b Bacteroidota 0.06 0–3.67 0.53 0–11.01 1.74 0.21–12.02 0.036 0.087 Bacillota 89.60 77.20–97.52 85.28 71.62–95.08 82.02 64.30–88.92 0.052 0.087 Actinomycetota 6.87 1.47–14.05 5.35 1.66–14.13 8.20 2.60–15.70 0.569 0.569 Fusobacteriota 1.20 0.15–10.52 5.08 0.13–20.34 3.60 0.89–27.9 0.145 0.181 a a,b b Pseudomonadota 0.01 0–0.60 0.27 0–1.73 0.28 0.04–2.14 0.019 0.087 Significantly different results are indicated in bold (p-value < 0.05). Superscript letters indicate differences between age categories. FIGURE 5 FIGURE 6 A box plot representation of the relative abundance from rarefied data of the family Ruminococcaceae family by age category. A box plot representation of the relative abundance from rarefied data of the Uncl. Clostridiales.1 family by age category. 3.4. Correlations among qPCR, IgA, cCP, and SCFA 4. Discussion Considering significant correlations (rho) equal to or Our study supports previously published results on the >0.3, Faecalibacterium abundance correlated positively with relationship between the aging process and gut health and gut propionate and valeric acid concentrations, while Fusobacterium microbiota in dogs. Notably, we included a wider age range correlated positively with isobutyric concentration. Clostridium (between 0.2 and 15 years old) than previous studies on puppies hiranonis correlated positively with isovaleric acid percentage, and adult dogs. In addition, we studied our dog population from and Bifidobacterium correlated positively with valeric acid different perspectives based on the analysis of fecal microbiota concentration. This genus correlated positively with the acetate together with other intestinal health biomarkers to broadly relative percentage, while the correlation with the butyrate understand the overall changes. Our main focus for the discussion relative percentage was negative. Bacteroides genus correlated of the results is the senior life stage, the period when the quality of positively with valeric acid concentration and its relative life is threatened by the aging process itself. Exploring gut health percentage. IgA concentrations correlated negatively with and microbiota based on fecal indicators may help find specific diet isobutyric, isovaleric, and valeric acid concentrations. This interventions to improve the quality of life of aging dogs. indicator also correlated negatively with isobutyric and valeric To compare the selected indicators along the different life absolute percentages. Correlation indexes are detailed in the stages, this study included dogs with specific diet needs because of Supplementary Table 5. their age, which may be considered a confounder to understanding Frontiers in Veterinary Science 07 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 FIGURE 7 FIGURE 8 A box plot representation of the relative abundance from rarefied A box plot representation of the relative abundance from rarefied data of the Veillonellaceae family by age category. data of the Lachnospiraceae family by age category. the results among the age categories. However, the diet, together of methodological aspects in 16S rRNA gene sequencing on the with body weight, sex, feeding routine, housing, and breed, results is important to consider, so different studies are difficult to were considered potential explanatory variables for the general compare. Previous studies have reported significant differences in statistical analysis. the abundance of bacterial groups based on variations in labs and methodologies (38–40). The Bifidobacteriaceae family is an important bacterial group 4.1. Microbiota results for humans. Considering the results for Bifidobacterium in the qPCR and the sequencing analysis, this genus was present in Bacillota, Actinomycetota, Fusobacteriota, Bacteroidota, and all three age categories (senior dogs, adult dogs, and puppies), Pseudomonadota were the five main phyla found in the fecal being more representative in adults and less representative in samples analyzed in our studied population of dogs using 16S puppies. Although intestinal microbiota composition has been rRNA gene sequencing. Previous studies described that most studied in different animal species, including dogs, the transition bacterial sequences identified in the gut belong to these five of the intestinal microbiota with age has not been thoroughly different phyla (29–31). In this regard, previous literature may have investigated. Masuoka et al. (10) studied a different age population, used the old phylum terminology for Bacillota, Actinomycetota, from pre-weanling puppies to senior dogs, to evaluate age- Fusobacteriota, Bacteroidota, and Pseudomonadota, which dependent differences and changes in the intestinal microbiota by were previously called Firmicutes, Actinobacteria, Fusobacteria, employing a culture-based method. Bifidobacteria were detected Bacterioidetes, and Proteobacteria, respectively (32). Although the in puppies but were not present when analyzing dogs older names have recently changed, these phyla have been previously than 3 years old (adult dogs and senior dogs). Based on the identified as the main contributors to gut microbial composition in literature on human beings, the presence of this genus in the dogs. More specifically, Bacillota, Bacteroidota, and Fusobacteriota three categories should be understood as positive. Bifidobacteria were described as the three predominant phyla in the healthy is the most prevalent bacteria in infants and adults, and it canine fecal microbiome in previous studies (33, 34). However, is presumed to promote health benefits in the host (41). in our study, within the three main bacterial groups, the phylum However, the extrapolation of the results among animal species Actinomycetota was found instead of Bacteroidota. Interestingly, should be interpreted carefully. Bifidobacterium abundance in the distribution of relative abundance was the same across the three the qPCR correlated positively with valeric acid concentration categories. The low abundance in Bacteroidota phyla could be and acetate relative percentage. Positive correlations between because of obesity and gastrointestinal disease (35–37). Although Bifidobacterium and acetate have been previously described (42, Actinomycetota is present in the small intestine of healthy animals, 43), although further studies investigating links with valeric acid the relative abundance found in feces is lower, and its increase may may be warranted. be associated with pathological conditions (29). Considering the Senior dogs had a higher relative abundance of the health status of the population studied, obesity and gastrointestinal families Ruminococcaceae, Veillonellaceae, Bacteroidaceae, pathologies do not explain the low numbers obtained, and the and Lachnospiraceae (phylum Bacillota). Previous studies have differences may be explained by the analysis itself. The influence found decreases at this level in dogs with inflammatory bowel Frontiers in Veterinary Science 08 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 TABLE 5 Fecal microbiota composition (% of relative abundance from rarefied data) at the family level split by age group. Family Junior Adult Senior Adult vs. Junior vs. Senior Median Range Median Range Median Range p-value q-value b a a,b Bifidobacteriaceae 0.06 0–0.70 1.43 0–13.44 0.18 0–4.78 0.024 0.066 Coriobacteriaceae 6.73 1.17–14.05 3.84 0.68–13.01 7.10 0.88–15.69 0.204 0.264 [Paraprevotellaceae] 0.03 0–1.08 0.03 0–1.24 0.15 0–1.05 0.115 0.211 S24-7 0 0–0.5 0.1 0–10.8 0 0–7.8 0.201 0.264 a a b Bacteroidaceae 0.03 0–1.33 0.07 0–0.44 0.73 0.21–3.09 <0.001 0.009 Prevotellaceae 0a,b 0–0.74 0a 0–0.19 0.03b 0–6.93 0.016 0.049 Erysipelotrichaceae 10.01 2.18–20.99 22.31 2.25–83.44 5.58 3.71–52.05 0.199 0.264 Clostridiaceae 34.47 8.34–46.61 24.94 3.07–65.36 39.12 5.27–57.80 0.628 0.641 Uncl. Clostridiales.1 0.17a,b 0–0.39 0.05a 0–0.28 0.19b 0.05–0.72 0.008 0.049 Peptostreptococcaceae 1.17 0–11.38 0.70 0–8.22 0.23 0.03–1.46 0.094 0.189 b a a,b Lachnospiraceae 15.13 6.16–24.89 7.44 1.17–19.26 13.81 3.66–23.45 0.011 0.049 Turicibacteraceae 7.86 0.14–25.78 1.48 0.11–13.28 2.56 0.33–8.39 0.167 0.263 Streptococcaceae 0.23 0–16.38 1.02 0–22.00 0.19 0–14.31 0.641 0.641 Lactobacillaceae 0.69 0–67.59 0.09 0–20.13 1.01 0–47.93 0.242 0.296 a,b a b Ruminococcaceae 0.31 0–3.45 0.25 0.01–2.19 1.99 0–11.46 0.014 0.049 Peptococcaceae 0.32 0–0.91 0.23 0–1.82 0.17 0–0.89 0.570 0.627 Uncl. Clostridiales.2 0.32 0–0.51 0.29 0–0.71 0.22 0–0.48 0.494 0.572 Veillonellaceae 0a 0–0.18 0.01a,b 0–0.2 0.15b 0–1.25 0.009 0.049 Fusobacteriaceae 1.20 0.15–10.52 5.08 0.13–20.34 3.60 0.89–27.90 0.145 0.246 Succinivibrionaceae 0a,b 0–0.51 0a 0–0.11 0.03b 0–0.18 0.013 0.049 Alcaligenaceae 0 0–0.20 0.14 0–1.60 0.10 0–2.10 0.051 0.124 Enterobacteriaceae 0 0–0.04 0.02 0–1.59 0.03 0–1.01 0.094 0.189 Significantly different results are indicated in bold (p-value < 0.05). Superscript letters indicate differences between age categories. disease (IBD) (25, 44), but there is no clear relationship with Although Faecalibacterium genera were not significantly aging. Senior dogs also had higher values in the relative percentage different in the qPCR analysis, Faecalibacterium prausnitzii was of acetate, one of the main products of bacterial fermentation. found to be higher in older dogs, which may be considered a minor These two findings are in the same direction as previously positive change for this aged population. This discrepancy may be published data; there is a well-known positive correlation caused by the fact that F. prausnitzii falls into the Faecalibacterium between the bacterial groups classified in Bacillota phyla and genera together with other species. Another possible reason the production of SCFA (25). However, our results supported could be that Facecalibacterium is frequently undetected in dogs the fact that relative percentages of butyrate and valeric acid during sequencing due to its low abundance, but it could be were significantly lower in the population of senior dogs. The detected using qPCR. In addition, Faecalibacterium qPCR was relative percentages are logically influenced by the high numbers significantly positively correlated with propionate and valeric acid in the acetate concentration. Focusing on the valeric result, the concentrations, which is positive for gut health. The absence of relative percentage was supported by the concentration result, this bacterium has been associated with gastrointestinal disorders which was also lower in the senior group. These findings may in dogs (36) and humans (47). Focusing on the species level, be interpreted as negative considering gut health since SCFA F. prausnitzii has been clearly associated with increased fiber production provides an appropriate pH environment to maintain utilization and different fiber types in the diets of dogs (8, 48). healthy microbiota, inhibits the growth of pH-sensitive pathogenic However, the diet’s fiber content was not considered a cause of the bacteria, and preserves gut integrity (45). Moreover, valeric higher abundance described in the S category for F. prausintzii. acid has been positively associated with the gut-brain axis and In our study, only 5% of the dogs were out of the defined fiber cognitive function in mice (46). Therefore, senior dogs may not composition range (2–5%), and they were similarly distributed in benefit from the mechanisms surrounding the production of the three different age categories (J = 2/106; A = 2/106; and S SCFA that promote intestinal health and positively contribute to = 1/106). From this information, considering the importance of cognitive function. F. prausnitzii within the Faecalibacterium genera, we could state Frontiers in Veterinary Science 09 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 TABLE 6 Fecal SCFA values expressed as concentration (µmol/g DM) and relative percentages (%) reported by age category. Fecal SFCA Junior Adult Senior emmeans ± SE emmeans ± SE emmeans ± SE p-value Concentration (µmol/g DM) Acetate 169 ± 26.1 210 ± 15.0 249 ± 21.1 0.061 Propionate 10.2 ± 1.449 12.4 ± 0.837 14.0 ± 1.171 0.149 Butyrate 52.8 ± 8.25 71.4 ± 4.76 61.2 ± 6.66 0.079 Isobutyric acid 6.91 ± 1.109 7.92 ± 0.640 7.27 ± 0.896 0.634 Isovaleric acid 9.72 ± 1.404 8.16 ± 0.811 6.69 ± 1.134 0.249 Valeric acid 4.55 ± 2.06a,b 8.59 ± 1.19a 4.70 ± 1.66b 0.049 SCFA total 254 ± 34.7 319 ± 20.0 343 ± 28.0 0.130 Relative percentages (%) Acetate 69.4 ± 1.78a,b 66.0 ± 1.02a 72.9 ± 1.43b <0.001 Propionate 4.17 ± 0.342 4.18 ± 0.197 4.23 ± 0.276 0.986 a,b a b Butyrate 18.8 ± 1.552 22.0 ± 0.896 17.3 ± 1.253 0.003 Isobutyric acid 2.63 ± 0.304 2.62 ± 0.175 2.22 ± 0.246 0.358 a a,b b Isovaleric acid 3.43 ± 0.359 2.69 ± 0.207 2.01 ± 0.290 0.011 a,b a b Valeric acid 1.56 ± 0.507 2.44 ± 0.293 1.38 ± 0.410 0.043 Significantly different results are indicated in bold (p-value < 0.05). Superscript letters indicate differences between age categories. TABLE 7 Fecal cCP and IgA values by age category. performed. However, the size effect as estimated by the R-value was extremely small, and other distance metrics were not significant. Junior Adult Senior The changes found in our population, and especially the categories emmeans emmeans emmeans p-value defined, may not be enough to change the weighted UniFrac ± SE ± SE ± SE and Bray–Curtis beta diversity. Another possible explanation for cCP (ng/g) 4.67 ± 0.564 3.90 ± 0.326 3.79 ± 0.456 0.420 the observed differences in the unweighted UniFrac but not in IgA (mg/g) 11.5 ± 3.14 11.1 ± 1.81 12.2 ± 2.54 0.935 the weighted UniFrac and Bray–Curtis beta diversity could be Different superscript letters indicate significant differences between groups (p-value < 0.05). attributed to the fact that these two metrics consider relative abundances instead of only the presence/absence of different taxa. Furthermore, alpha diversity analysis measured by the Chao1 and that the difference found in senior dogs is a positive change in the Shannon indexes was not significantly different among the age microbiota population related to aging. groups. Our findings are consistent with previous studies. You and At the genus level, the two predominant groups identified Kim (31) studied different individual traits in healthy dogs and also belong to the phylum Bacillota (Uncl. Clostridiaceae.1 and reported no differences at the age level when studying microbial Allobaculum). These findings support previously published data diversity patterns. Interestingly, none of the dogs belonging to the on dogs, in which major taxa fell within the Bacillota group and junior category had a moderate or significant change in the index the bacterial class Clostridia was considered the most abundant (DI > 0). Blake et al. (15) showed that the DI increased in younger taxon (29). In our study, the Fusobacterium genus (within the puppies, but many dogs at 9 weeks and most dogs after 6 months Fusobacteriota phylum) was present in all the age populations of age had normal DI. In our study, the junior age ranged from and was specifically higher in dogs over 7 years old. This finding 2 to 19 months; therefore, our data are consistent with previous can be considered positive for the microbiome of senior dogs studies, as most dogs over 6 months are expected to have a DI of <0. since this genus has previously been related to a healthy intestinal From those results, we could state that, in our study, small changes microbiome in dogs (13). in the microbiota may have occurred due to aging. To conclude with the microbiota data, the performance of the Kruskal–Wallis test for assessing differential abundance between groups could be a limitation. This statistical test considers the overall abundance 4.2. Beta diversity, alpha diversity, and between groups but not the frequency of organism identification dysbiosis index in samples within a group. Further statistical tests for assessing differential abundance between groups (e.g., ANCOM and DESeq) For beta diversity, unweighted UniFrac was significantly could complement the information analyzed by the Kruskal– different among age categories when ANOSIM analysis was Wallis test. Frontiers in Veterinary Science 10 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 4.3. Fecal SCFAs, IgA, and cCP Data availability statement Fecal cCP and IgA concentrations in feces have been studied The datasets presented in this study can be found in as non-invasive biomarkers of gut health in adult dogs. Fecal cCP online repositories. The names of the repository/repositories and is an indicator of intestinal inflammation in dogs (49, 50), whereas accession number(s) can be found below: https://www.ncbi.nlm. IgA is associated more with the gut immune response (51). In the nih.gov/, PRJNA901473. present study, the fecal concentration of cCP and IgA was not related to the aging process since no differences were observed among the three age categories. Based on the literature, aged mice Ethics statement and humans did not show a clear relationship between age and the concentration of mucosal IgA (52). Zaine et al. (53) reported no The animal study was reviewed and approved by Affinity significant differences in IgA fecal values in senior dogs compared Petcare Ethical Committee. with adults and puppies. However, 5-month-old puppies had lower IgA concentrations than adult dogs. Fecal cCP concentrations were recently correlated with age in healthy humans in Korea (54), but to Author contributions our knowledge, the results on cCP concentrations in healthy dogs of different ages are not consistent. The lack of clear reference values AF-P, AS-M, and CT designed and performed the study. AF- for fecal cCP and IgA considering the different age stages is a clear P and RP analyzed the results and prepared the tables and graphs limitation to thoroughly understanding our results in relation to for inclusion in the manuscript. AF-P wrote the manuscript. age and the SCFA correlation found in our study. However, in our Overall, all authors contributed to the paper, discussed its results, study, age did not change the intestinal inflammation status or gut reviewed the drafts, and approved the final version submitted immune response based on the two indicators analyzed. for publication. Funding 4.4. Main features This study received funding from Affinity Petcare. The Considering the overall results of the present study, minor funder was not involved in the study design, collection, analysis, changes in the intestinal health biomarkers analyzed were interpretation of data, the writing of this article, or the decision to found among the junior, adult, and senior categories. The submit it for publication. microbiota community and SCFA production showed age- dependent differences. However, the IgA and cCP were not significantly different across the three categories. Conflict of interest Our age category definition may be a limitation to thoroughly understanding the continuous changes of the microbiome in AF-P, AS-M, and CT were employed by the pet food company relation to the aging process. Significantly, no clear consensus exists Affinity Petcare, which funded the study. JS and RP were employed about this classification in animal research or veterinary medicine by the Gastrointestinal Laboratory at Texas A&M University, which (55). The definition of age categories may reduce the sensitivity provides microbiome assessment on a fee-for-service basis. of understanding whether different ages equally affect gut health The remaining author declares that the research was conducted and microbiota. Thus, a larger number of animals representative of in the absence of any commercial or financial relationships that all age categories may help define more categories to increase the could be construed as a potential conflict of interest. sensitivity of these first results. In addition, the extrapolation of this trial in a field setting study may help better explore effective strategies, such as dietary Publisher’s note interventions, in the dog population. All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, 5. Conclusion the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by In conclusion, in this study, the canine gut microbiota and its manufacturer, is not guaranteed or endorsed by the certain SCFAs showed minor variations among age groups in dogs. publisher. Other intestinal health biomarkers (IgA and cCP) were similar among age groups. Considering all the parameters, we can state that, once the microbiota becomes stable in healthy dogs, minor Supplementary material changes occur during the aging process. The modulation of these minor changes, considering the The Supplementary Material for this article can be found intestinal microbiota and short-chain fatty acid production, could online at: https://www.frontiersin.org/articles/10.3389/fvets.2023. improve the overall gut health during aging in dogs. 1213287/full#supplementary-material Frontiers in Veterinary Science 11 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 References 1. Matijašić M, Meštrović T, Paljetak HC, Perić M, Barešić A, Verbanac D. 24. Moreau NM, Goupry SM, Antignac JP, Monteau FJ, Bizec BJL, Champ Gut microbiota beyond bacteria—mycobiome, virome, archaeome, and eukaryotic MM, et al. Simultaneous measurement of plasma concentrations and 13C- parasites in IBD. Int J Mol Sci. (2020) 21:2668. doi: 10.3390/ijms21082668 enrichment of short-chain fatty acids, lactic acid and ketone bodies by gas chromatography coupled to mass spectrometry. J Chromatogr B. (2003) 784:395– 2. Gill SR, Pop M, DeBoy RT, Eckburg PB, Turnbaugh PJ, Samuel BS, et al. 403. doi: 10.1016/S1570-0232(02)00827-9 Metagenomic analysis of the human distal gut microbiome. Science. (2006) 312:1355– 9. doi: 10.1126/science.1124234 25. Minamoto Y, Minamoto T, Isaiah A, Sattasathuchana P, Buono A, Rangachari 3. Jandhyala SM, Talukdar R, Subramanyam C, Vuyyuru H, Sasikala M, Reddy VR, et al. Fecal short-chain fatty acid concentrations and dysbiosis in dogs with DN. Role of the normal gut microbiota. World J Gastroentero. (2015) 21:8787– chronic enteropathy. J Vet Intern Med. (2019) 33:1608–18. doi: 10.1111/jvim. 803. doi: 10.3748/wjg.v21.i29.8787 15520 4. de Vos WM, Tilg H, Hul MV, Cani PD. Gut microbiome and health: mechanistic 26. Heilmann RM, Suchodolski JS, Steiner JM. Development and analytic validation insights. Gut. (2022) 71:1020–32. doi: 10.1136/gutjnl-2021-326789 of a radioimmunoassay for the quantification of canine calprotectin in serum and feces from dogs. Am J Vet Res. (2008) 69:845–53. doi: 10.2460/ajvr.69.7.845 5. Bajinka O, Darboe A, Tan Y, Abdelhalim KA, Cham LB. Gut microbiota and the human gut physiological changes. Ann Microbiol. (2020) 27. Heilmann RM, Ruaux CG, Grützner N, Cranford SM, Bridges CS, Steiner JM. 70:65. doi: 10.1186/s13213-020-01608-2 Biological variation of serum canine calprotectin concentrations as measured by ELISA in healthy dogs. Vet J. (2019) 247:61–4. doi: 10.1016/j.tvjl.2019.03.002 6. Wen L, Duffy A. Factors Influencing the gut microbiota, inflammation, and type 2 diabetes. J Nutrition. (2017) 147:1468S–75S. doi: 10.3945/jn.116.240754 28. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B. (1995) 57:289– 7. Hasan N, Yang H. Factors affecting the composition of the gut microbiota, and its 300. doi: 10.1111/j.2517-6161.1995.tb02031.x modulation. PeerJ. (2019) 7:e7502. doi: 10.7717/peerj.7502 29. Pilla R, Suchodolski JS. The role of the canine gut microbiome and 8. Pilla R, Suchodolski JS. The gut microbiome of dogs and cats, and the metabolome in health and gastrointestinal disease. Front Vet Sci. (2020) influence of diet. Vet Clin North Am Small Animal Pract. (2021) 51:605– 6:498. doi: 10.3389/fvets.2019.00498 21. doi: 10.1016/j.cvsm.2021.01.002 30. Deschamps C, Humbert D, Zentek J, Denis S, Priymenko N, Apper E, et al. From 9. Badal VD, Vaccariello ED, Murray ER, Yu KE, Knight R, Jeste DV, et al. Chihuahua to Saint-Bernard: how did digestion and microbiota evolve with dog sizes. The gut microbiome, aging, and longevity: a systematic review. Nutrients. (2020) Int J Biol Sci. (2022) 18:5086–102. doi: 10.7150/ijbs.72770 12:3759. doi: 10.3390/nu12123759 31. You I, Kim MJ. Comparison of gut microbiota of 96 healthy dogs by 10. Masuoka H, Shimada K, Kiyosue-Yasuda T, Kiyosue M, Oishi Y, Kimura S, et al. individual traits: breed, age, and body condition score. Anim Open Access J. (2021) Transition of the intestinal microbiota of dogs with age. Biosci Microbiota Food Heal. 11:2432. doi: 10.3390/ani11082432 (2016) 36:27–31. doi: 10.12938/bmfh.BMFH-2016-021 32. Oren A, Garrity GM. Valid publication of the names of forty-two phyla of 11. Suchodolski JS. Analysis of the gut microbiome in dogs and cats. Vet Clin Path. prokaryotes. Int J Syst Evol Microbiol. (2021) 71:005056. doi: 10.1099/ijsem.0.005056 (2022) 50:6–17. doi: 10.1111/vcp.13031 33. Hand D, Wallis C, Colyer A, Penn CW. Pyrosequencing the canine 12. Ziese A-L, Suchodolski JS. Impact of changes in gastrointestinal microbiota in faecal microbiota: breadth and depth of biodiversity. PLoS ONE. (2013) canine and feline digestive diseases. Vet Clin North Am Small Animal Pract. (2021) 8:e53115. doi: 10.1371/journal.pone.0053115 51:155–69. doi: 10.1016/j.cvsm.2020.09.004 34. Middelbos IS, Boler BMV, Qu A, White BA, Swanson KS, Fahey GC. 13. Garrigues Q, Apper E, Chastant S, Mila H. Gut microbiota development in Phylogenetic characterization of fecal microbial communities of dogs fed diets with the growing dog: a dynamic process influenced by maternal, environmental and host or without supplemental dietary fiber using 454 pyrosequencing. PLoS ONE. (2010) factors. Front Vet Sci. (2022) 9:964649. doi: 10.3389/fvets.2022.964649 5:e9768. doi: 10.1371/journal.pone.0009768 14. Guard BC, Mila H, Steiner JM, Mariani C, Suchodolski JS, 35. Salas-Mani A, Jeusette I, Castillo I, Manuelian CL, Lionnet C, Iraculis N, et al. Chastant-Maillard S. Characterization of the fecal microbiome during Fecal microbiota composition changes after a BW loss diet in Beagle dogs. J Anim Sci. neonatal and early pediatric development in puppies. PLoS ONE. (2017) (2018) 96:3102–11. doi: 10.1093/jas/sky193 12:e0175718. doi: 10.1371/journal.pone.0175718 36. Minamoto Y, Otoni CC, Steelman SM, Büyükleblebici O, Steiner JM, Jergens 15. Blake AB, Cigarroa A, Klein HL, Khattab MR, Keating T, Coevering PVD, AE, et al. Alteration of the fecal microbiota and serum metabolite profiles in et al. Developmental stages in microbiota, bile acids, and clostridial species in healthy dogs with idiopathic inflammatory bowel disease. Gut Microbes. (2015) 6:33– puppies. J Vet Intern Med. (2020) 34:2345–56. doi: 10.1111/jvim.15928 47. doi: 10.1080/19490976.2014.997612 16. FEDIAF. The European Pet Food Industry. Nutritional Guidelines for Complete 37. Söder J, Wernersson S, Höglund K, Hagman R, Lindåse S, Dicksved and Complementary Pet Food for Cats and Dogs. (2021). Available online at: https:// J. Composition and short-term stability of gut microbiota in lean and europeanpetfood.org/wp-content/uploads/2022/03/Updated-Nutritional-Guidelines. spontaneously overweight healthy Labrador retriever dogs. Acta Vet Scand. (2022) pdf 64:8. doi: 10.1186/s13028-022-00628-z 17. Strickling JA, Harmon DL, Dawson KA, Gross KL. Evaluation of oligosaccharide 38. Handl S, Dowd SE, Garcia-Mazcorro JF, Steiner JM, Suchodolski JS. Massive addition to dog diets: influences on nutrient digestion and microbial populations. Anim parallel 16S rRNA gene pyrosequencing reveals highly diverse fecal bacterial and Feed Sci Technol. (2000) 86:205–19. doi: 10.1016/S0377-8401(00)00175-9 fungal communities in healthy dogs and cats. Fems Microbiol Ecol. (2011) 76:301– 18. Parada AE, Needham DM, Fuhrman JA. Every base matters: 10. doi: 10.1111/j.1574-6941.2011.01058.x assessing small subunit rRNA primers for marine microbiomes with mock communities, time series and global field samples. Environ Microbiol. (2015) 39. Garcia-Mazcorro JF, Lanerie DJ, Dowd SE, Paddock CG, Grützner N, Steiner 18:1403–14. doi: 10.1111/1462-2920.13023 JM, et al. Effect of a multi-species synbiotic formulation on fecal bacterial microbiota of healthy cats and dogs as evaluated by pyrosequencing. Fems Microbiol Ecol. (2011) 19. Apprill A, McNally S, Parsons R, Weber L. Minor revision to V4 78:542–54. doi: 10.1111/j.1574-6941.2011.01185.x region SSU rRNA 806R gene primer greatly increases detection of SAR11 bacterioplankton. Aquat Microb Ecol. (2015) 75:129–37. doi: 10.3354/ame 40. Guard BC, Barr JW, Reddivari L, Klemashevich C, Jayaraman A, Steiner JM, et al. 01753 Characterization of microbial dysbiosis and metabolomic changes in dogs with acute diarrhea. PLoS ONE. (2015) 10:e0127259. doi: 10.1371/journal.pone.0127259 20. Bolyen E, Rideout JR, Dillon MR, Bokulich NA, Abnet CC, Al-Ghalith GA, et al. Reproducible, interactive, scalable and extensible microbiome data science using 41. Mitsuoka T. History and evolution of probiotics. Jpn J Lact Acid Bact. (2011) QIIME 2. Nat Biotechnol. (2019) 37:852–7. doi: 10.1038/s41587-019-0209-9 22:26–37. doi: 10.4109/jslab.22.26 21. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. 42. Oliphant K, Allen-Vercoe E. Macronutrient metabolism by the human gut DADA2: high resolution sample inference from Illumina amplicon data. Nat Methods. microbiome: major fermentation by-products and their impact on host health. (2016) 13:581–3. doi: 10.1038/nmeth.3869 Microbiome. (2019) 7:91. doi: 10.1186/s40168-019-0704-8 22. Lozupone C, Knight R. UniFrac: a new phylogenetic method 43. Koh A, De Vadder F, Kovatcheva-Datchary P, Bäckhed F. From dietary fiber for comparing microbial communities. Appl Environ Microb. (2005) to host physiology: short-chain fatty acids as key bacterial metabolites. Cell. (2016) 71:8228–35. doi: 10.1128/AEM.71.12.8228-8235.2005 165:1332–45. doi: 10.1016/j.cell.2016.05.041 23. AlShawaqfeh MK, Wajid B, Minamoto Y, Markel M, Lidbury JA, Steiner 44. Suchodolski JS, Dowd SE, Wilke V, Steiner JM, Jergens AE. 16S rRNA gene JM, et al. A dysbiosis index to assess microbial changes in fecal samples of dogs pyrosequencing reveals bacterial dysbiosis in the duodenum of dogs with idiopathic with chronic inflammatory enteropathy. FEMS Microbiol Ecol. (2017) 93:fix136. inflammatory bowel disease. PLoS ONE. (2012) 7:e39333. doi: 10.1371/journal.pone.0 doi: 10.1093/femsec/fix136 039333 Frontiers in Veterinary Science 12 frontiersin.org Fernández-Pinteño et al. 10.3389/fvets.2023.1213287 45. Mackei M, Talabér R, Müller L, Sterczer Á, Fébel H, Neogrády Z, 50. Otoni CC, Heilmann RM, García-Sancho M, Sainz A, Ackermann MR, et al. Altered intestinal production of volatile fatty acids in dogs triggered by Suchodolski JS, et al. Serologic and fecal markers to predict response to induction lactulose and psyllium treatment. Vet Sci. (2022) 9:206. doi: 10.3390/vetsci90 therapy in dogs with idiopathic inflammatory bowel disease. J Vet Intern Med. (2018) 50206 32:999–1008. doi: 10.1111/jvim.15123 46. Lai Z, Shan W, Li J, Min J, Zeng X, Zuo Z. Appropriate exercise level 51. Grellet A, Mila H, Heilmann RM, Feugier A, Gruetzner N, Suchodolski JS, et al. attenuates gut dysbiosis and valeric acid increase to improve neuroplasticity Effect of age, gestation and lactation on faecal IgA and calprotectin concentrations in and cognitive function after surgery in mice. Mol Psychiatr. (2021) 26:7167– dogs∗ . J Nutritional Sci. (2014) 3:e41. doi: 10.1017/jns.2014.44 87. doi: 10.1038/s41380-021-01291-y 52. Fujihashi K, McGhee JR. Mucosal immunity and tolerance in the elderly. Mech 47. Varela E, Manichanh C, Gallart M, Torrejón A, Borruel N, Casellas F, Ageing Dev. (2004) 125:889–98. doi: 10.1016/j.mad.2004.05.009 et al. Colonisation by faecalibacterium prausnitzii and maintenance of clinical remission in patients with ulcerative colitis. Aliment Pharm Therap. (2013) 38:151– 53. Zaine L, Ferreira C, de OS Gomes M, Monti M, Tortola L, Vasconcellos RS, 61. doi: 10.1111/apt.12365 et al. Faecal IgA concentration is influenced by age in dogs. Brit J Nutr. (2011) 106:S183–6. doi: 10.1017/S0007114511000559 48. Mori A, Goto A, Kibe R, Oda H, Kataoka Y, Sako T. Comparison of the effects of four commercially available prescription diet regimens on the fecal microbiome in 54. Park SY. Age-related fecal calprotectin concentrations in healthy adults. Korean healthy dogs. J Vet Med Sci. (2019) 81:1783–90. doi: 10.1292/jvms.19-0055 J Clin Laboratory Sci. (2020) 52:181–7. doi: 10.15324/kjcls.2020.52.3.181 49. Grellet A, Heilmann RM, Lecoindre P, Feugier A, Day MJ, Peeters D, et al. Fecal 55. Harvey ND. How old is my dog? identification of rational age groupings calprotectin concentrations in adult dogs with chronic diarrhea. Am J Vet Res. (2013) in pet dogs based upon normative age-linked processes. Front Vet Sci. (2021) 74:706–11. doi: 10.2460/ajvr.74.5.706 8:643085. doi: 10.3389/fvets.2021.643085 Frontiers in Veterinary Science 13 frontiersin.org