Dietary supplementation with strawberry induces marked changes in the composition and functional potential of the gut microbiome in diabetic mice

Abstract

Gut microbiota contribute to the biological activities of berry anthocyanins by transforming them into bioactive metabolites and anthocyanins support the growth of specific bacteria indicating a two-way relationship between anthocyanins and microbiota. In the present study, we tested the hypothesis that strawberry supplementation alters gut microbial ecology in diabetic db/db mice. Control (db/+) and diabetic (db/db) mice (7-wk) consumed standard diet or diet supplemented with 2.35% freeze-dried strawberry (db/db+SB) for 10 weeks. Colon contents were used to isolate bacterial DNA. V4 variable region of 16S rRNA gene was amplified. Data analyses were performed using standardized pipelines (QIIME 1.9 and R packages). Differences in predictive metagenomics function was identified by PICRUSt. Principle coordinate analyses confirmed that the microbial composition was significantly influenced by both host genotype and strawberry consumption. Further, α-diversity indices and β-diversity were different at the phylum and genus levels, and genus and OTU levels, respectively (P<0.05). At the phylum level, strawberry supplementation decreased the abundance of Verrucomicrobia in db/db+SB vs. db/db mice (P<0.05). At the genus level, db/db mice exhibited a decrease in the abundance of Bifidobacterium and strawberry supplementation increased Bifidobacterium in db/db+SB vs. db/db mice (P<0.05). PICRUSt revealed significant differences in 45 predicted metabolic functions among the three groups. Our study provides evidence for marked changes in the composition and functional potential of the gut microbiome with strawberry supplementation in diabetic mice. Importantly, strawberry supplementation increased the abundance of beneficial bacteria Bifidobacterium which play a pivotal role in the metabolism of anthocyanins.

1. Introduction

Anthocyanins are one of the major flavonoid compounds present in many fruits and vegetables including blueberries and strawberries [1]. Anthocyanins are glycosides comprised of an aglycone component (anthocyanidins such as cyanidin, delphinidin, malvidin, peonidin, pelargonidin, and petunidin) and a sugar moiety (glucose, xylose, galactose, and arabinose) [1]. Epidemiological and clinical studies support the beneficial health effects of berry anthocyanins such as reducing the risk of cardiovascular disease [2–5]. The health benefits of berry anthocyanins are ascribed to their antihypertensive, antioxidant, and anti-atherosclerotic effects [3, 4]. Anthocyanins are extensively metabolized by digestive enzymes and intestinal microbiota in humans suggesting a significant portion of the biological activity of anthocyanins may be linked to their metabolites [6, 7]. Data from our laboratory and others suggest that the circulating metabolites mediate the beneficial vascular and endothelial effects of anthocyanins [8–10].

Strawberry is an excellent source of dietary anthocyanins and the most commonly found anthocyanins in strawberries are the glycosidic derivatives of pelargonidin and cyanidin [5]. Epidemiological study demonstrated that habitual consumption of 2-3 servings (~160 - 240 g) of strawberries per week reduce the risk of myocardial infarction in humans [2]. Further, strawberry intake attenuated postprandial inflammation in overweight adults and reduced inflammatory molecules in humans with cardiovascular risk factors [11–13]. Recently we showed that dietary supplementation of strawberry at a nutritional dosage (equivalent to 2 human servings) reduced vascular inflammation and ameliorated vascular dysfunction in diabetic mice [10]. Further, our study suggested that circulating metabolites mediate the vascular effects of strawberry anthocyanins [10]. Gut microbiota contributes significantly to the biological activities of berry anthocyanins by transforming them into more readily absorbable bioactive metabolites [6, 7]. Interestingly, anthocyanins promote intestinal colonization and support the growth of specific groups of bacteria indicating a two-way relationship between anthocyanins and microbiota [7, 14]. Indeed, anthocyanins may act as prebiotics and play a role in reshaping the gut microbiome which enhance the host microbial interaction to provide beneficial health effects in humans [7].

Healthy gut microbiota play a major role in converting anthocyanins into bioactive metabolites. Therefore, it is one of the important factors required for the potential biological activities of anthocyanins. Anthocyanins in strawberry extracts were shown to enhance the beneficial effects of diets with fructooligosaccharides (a constituent of dietary fiber which has prebiotic properties) in the rat cecal environment [15]. However, to our knowledge the effect of strawberry on gut microbiota in general and diabetes in particular are unknown. In the present study, we tested the hypothesis that dietary supplementation of strawberry induces changes in the composition and functional potential of the gut microbiome in diabetic db/db mice.

2. Materials and methods

2.1. Experimental animals

db/db mice homozygous for the diabetes spontaneous mutation (Lepr^db) with C57BL/6J background manifest morbid obesity, pancreatic cell atrophy and chronic hyperglycemia. These leptin receptor deficient db/db mice is a widely-used type 2 diabetic animal model which exhibits dysbiosis of gut microbiota [16–18]. Six-week-old male diabetic db/db mice and control db/+ mice (Stock no. 000642) were obtained from the Jackson Laboratories (Bar Harbor, ME). The mice were held under humane conditions in the animal facility at the University of Utah and acclimated for a week before experiments were performed. Mice were housed 4 per cage and maintained under artificial light in a 12-hour light/dark cycle, 23 ± 1° C, and 45 ± 5% humidity. The Institutional Animal Care and Use Committee at the University of Utah approved the animal experiment protocols which conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institute of Health.

2.2. Experimental groups

The freeze-dried strawberry powder was provided by FutureCeuticals (Momence, IL). The customized pelleted diets (control diet and strawberry supplemented diet) were prepared as reported in our recent study and supplied by Dyets Inc. (Bethlehem, PA) [10]. The strawberry supplemented diet was adjusted to compensate for the fiber and additional sugars provided by the freeze-dried strawberry powder. The composition of the control diet and strawberry supplemented diet was reported in our recent study [10]. The amount of freeze-dried strawberry powder used in this study was calculated based on the Food and Drug Administration recommendation for the extrapolation of doses from humans to animals by normalization to body surface area [19]. The nutritional dose of freeze-dried strawberry powder was based on average human consumption. The dose used in these studies [2.35% Freeze dried strawberry powder in the diet pellets (w/w)] is equivalent to two human servings of fresh strawberries (~160 g strawberries) [10]. After 1 week of acclimation, diabetic mice (7 weeks old) were divided into two groups and received control diet (n=8) or 2.35% freeze-dried strawberry supplemented diet for 10 weeks (n=8). db/+ mice were fed control diet for 10 weeks served as controls (n=8). At the end of the 10-week treatment, mice were euthanized and colon contents were collected and snap-frozen in liquid nitrogen, and stored at −80 °C.

2.3. Microbial community profiling using 16S rRNA amplicon sequencing

Genomic DNA was extracted from colon contents using the DNeasy PowerSoil Kit (Qiagen, MD). Fifty nanogram of genomic DNA was utilized for amplification of the V4 variable region of the 16S rRNA gene using 515F/806R primers. Forward and reverse primers were dual-indexed as described by Kozich et al [20] to accommodate multiplexing of up to 384 samples per run. Paired-end sequencing (2× 250 bp) of pooled amplicons was carried out on an Illumina MiSeq [21] with ~30% PhiX DNA.

2.4. Bioinformatics Analysis

Processing and quality filtering of reads were performed using scripts in QIIME (v1.9.1) [22] and other in-house scripts. Paired reads were stitched with PEAR, an over-lapping paired-end reads merging algorithm which evaluates all possible paired-end read overlaps minimizing false positive hits [23]. Reads were further filtered based on Phred quality scores and for chimeric reads using USEARCH61 [24]. Filtered reads were de-multiplexed within QIIME and samples with less than 5000 reads were excluded from further analysis. UCLUST was used to cluster sequences into operational taxonomical units (OTUs based on >97% identity) [24]. OTU picking was performed using open-reference method which encompasses clustering of reads against a reference sequence collection and also performs de novo OTU picking on the reads which fail to align to any known reference sequence in the database [25]. To eliminate erroneous mislabeling, the resulting OTU tables were checked for mislabeling sequences [26]. Representative sequences were further aligned using PyNAST with the Greengenes core-set alignment template [27]. Construction of the phylogenetic tree was performed using the default (FASTTREE) method in QIIME [28].

All samples were clustered based on their between-sample distances using UPGMA, and subsequent jack-knifing was performed by resampling methods. Comparisons of intergroup and intragroup diversity were performed using ANOVA including correction for multiple comparisons. OTU reads were summed at genus levels and then assessed for group differences with negative binomial regression using the DESeq2 package. PICRUSt was used to identify differences in predictive metagenome function [29]. OTUs were normalized by the predicted 16S copy number, and functions were predicted with the use of GreenGenes 13_5 databases for KEGG Orthologs.

2.5. Statistical Analysis

Microbiota OTU reads were imported into R version 3.2.1 and all statistical analyses were performed using the vegan and phyloseq packages unless specifically noted. OTU richness was measured by Chao1 and evenness was measured by several diversity indices (Shannon, Simpson, Inverse Simpson, and Fisher). Group differences in α-diversity (richness and diversity) were assessed by ANOVA. Between-specimen diversity (β-diversity) was assessed by calculating a matrix of dissimilarities using the Bray-Curtis method and then visualized using non-metric multidimensional scaling (NMDS). Group differences in β-diversity were assessed using permutational multivariate analysis of variance (PERMANOVA) with 500 permutations. Group differences among genus level OTUs were assessed by pairwise comparisons on read counts using Negative Binomial Wald Tests from the DESeq2 package. OTU relative abundance is given as median % relative abundance when described in text. All statistical tests used on 16S-rRNA gene sequencing data were considered significant at p ≤ 0.05. All tests were corrected for multiple comparisons using the false discovery rate (FDR) correction by Benjamini and Hochberg. Associations among selected variables were assessed with Spearman’s correlations. An in-house developed R-based shiny app (DAME) was developed to facilitate procedures and statistical analysis described above [30]. All statistical analyses were performed and figures were made using R. Correlations between bacterial abundance with predicted metagenomic function was performed using the corrplot package in R and utilized bacteria at family or genus levels as described in the specific comparison. Statistical significance was determined at p ≤ 0.05.

3. Results

3.1. α-Diversity and β-Diversity

In the present study, a total of 480321 reads were assessed. A total of 320 of OTUs were assigned to taxonomic classifications of 6 phyla, 17 families and 30 genera. Principal coordinate analysis (PCoA) of unweighted UniFrac distances performed on the OTU abundance matrix showed that the β-diversity of gut microbial communities was significantly different between the groups and microbial composition was significantly influenced by both genotype (db/db) and strawberry consumption (Fig. 1A).

Fig. 1.

(A) Principle Component Analysis (PCA) plot. Non-metric multidimensional scaling analysis of the OTU abundance matrix of β-diversity of gut microbial communities at phylum (B), family (C), genus (D) and OTU (E) levels. Values are mean ± SEM, n=8. db/+, Standard diet fed control mice; db/db, standard diet fed diabetic mice; db/db+SB, strawberry fed diabetic mice.

Measurements of α-diversity are indicative of phylogenetic species richness and evenness within a sample. Indices of α-diversity such as ACE, Chao1, Fisher, InvSimpson, Observed, Shannon, and Simpson were measured. At the phylum level, α-diversity indices ACE, Chao1, Fisher, and Observed were significantly different at phylum level among groups (Table 1). At the genus level, α-diversity indices such as Fisher and Observed were significantly different among groups (Table 2). β-diversity represents compositional differences between samples (Fig. 1B - 1E). β-diversity, a measure of global microbial composition, was significantly different at the genus and OTU levels (Fig. 1D and 1E). β-diversity was influenced by genotype and diet at OTU levels (Fig. 1E).

Table 1.

α-Diversity indices at phylum level

Index	db/+	db/db	db/db+SB	P value
ACE	5.167 ± 0.144	6.017 ± 0.217	6.187 ± 0.162	0.006
Chao1	5.125 ± 0.125	5.75 ± 0.164	5.875 ± 0.125	0.002
Fisher	0.486 ± 0.016	0.542 ± 0.018	0.565 ± 0.013	0.006
InvSimpson	2.129 ± 0.103	2.364 ± 0.187	2.025 ± 0.041	0.173
Observed	5.125 ± 0.125	5.75 ± 0.164	5.875 ± 0.125	0.002
Shannon	0.807 ± 0.044	0.943 ± 0.084	0.778 ± 0.043	0.142
Simpson	0.524 ± 0.019	0.558 ± 0.036	0.505 ± 0.01	0.305

Data are expressed as means ±SE (n=8). db/+, Standard diet fed control mice; db/db, standard diet fed diabetic mice; db/db+SB, strawberry fed diabetic mice.

Table 2.

α-Diversity indices at genus level

Index	db/+	db/db	db/db+SB	P value
ACE	23.63 ± 0.501	22.069 ± 0.858	25.4 ± 3.57	0.554
Chao1	23.062 ± 0.538	21.562 ± 0.741	22.562 ± 1.551	0.590
Fisher	2.539 ± 0.069	2.244 ± 0.062	2.247 ± 0.066	0.006
InvSimpson	3.108 ± 0.322	3.421 ± 0.449	2.491 ± 0.062	0.139
Observed	22.625 ± 0.596	20.625 ± 0.532	20.25 ± 0.491	0.012
Shannon	1.463 ± 0.071	1.48 ± 0.109	1.257 ± 0.025	0.095
Simpson	0.657 ± 0.03	0.668 ± 0.046	0.597 ± 0.01	0.266

Data are expressed as means ±SE (n=8). db/+, Standard diet fed control mice; db/db, standard diet fed diabetic mice; db/db+SB, strawberry fed diabetic mice.

3.2. Relative abundance of microbiota at the phyla level

The distribution of bacterial taxa and the relative abundance of bacteria at the phyla level are shown in Fig. 2A-G. Bacterial sequences were distributed among 6 bacterial phyla including Actinobacteria, Bacteroidetes, Firmicutes, Proteobacteria, Tenericutes and Verrucomicrobia. The taxonomic abundance indicated a significant (P<0.05) decrease in Actinobacteria in db/db mice compared with db/+ mice (Fig. 2B). Strawberry supplementation did not increase the abundance of Actinobacteria in db/db+SB mice (Fig. 2B). The most abundant phyla were Bacteroidetes and Firmicutes (>90%). In the present study, the abundance of Bacteroidetes and Firmicutes were similar among three groups (Fig. 2C and 2D). The taxonomic abundance indicated a significant increase (p<0.05) in Proteobacteria in db/db vs db/+ mice but strawberry supplementation did not alter the abundance of Proteobacteria in db/db+SB mice vs db/+ mice (Fig. 2E). The abundance of Tenericutes was similar among the groups (Fig. 2F). There was a non-significant increase in Verrucomicrobia in db/db compared with db/+ mice but this was significantly reduced (p<0.05) in db/db+SB mice compared with db/db mice (Fig. 2G).

Fig. 2.

(A) The relative abundance of bacterial population at phylum level. The abundance of Actinobacteria (B), Bacteroidetes (C), Firmicutes (D), Proteobacteria (E), Tenericutes (F) and Verrucomicrobia (G) in db/+, db/db and db/db+SB mice treated for 10 wk. *, P<0.05, values are mean ± SEM, n=8. NS, non-significant; db/+, Standard diet fed control mice; db/db, standard diet fed diabetic mice; db/db+SB, strawberry fed diabetic mice.

3.3. Relative abundance of microbiota at the genus level

The abundance of Bifidobacterium, which belong to the phylum Actinobacteria, is significantly decreased in db/db mice as compared to db/+ mice (P<0.05) (Fig. 3A). However, strawberry supplementation significantly increased the abundance of Bifidobacterium in db/db+SB mice (P<0.05) (Fig. 3A). The abundance of Bacteroides was unchanged between db/+ and db/db mice (Fig. 3B). Strawberry supplementation significantly increased Bacteroides in db/db+SB mice (P<0.05) (Fig. 3B). Dehalobacterium and Dorea, which belong to Firmicutes phyla, were significantly increased in db/db mice compared with db/+ mice (P<0.05) (Fig. 3C and 3D). Lactobacillus, SMB53, and Turicibacter which belong the phylum Firmicutes were significantly decreased in db/db mice compared with db/+ mice (P<0.05) (Fig. 3E-G). Strawberry supplementation did not change the abundance of Dehalobacterium, Dorea, Lactobacillus, SMB53, and Turicibacter in db/db+SB mice compared with db/db mice. Akkermansia is significantly reduced in db/db+SB mice compared with db/db mice (P<0.05) (Fig. 3H). In addition, the unassigned genera belong to Enterobacteriaceae and Ruminococcaceae were increased whereas Peptostreptococcaceae was decreased in db/db mice compared with db/+ mice (P<0.05) (data not shown). Strawberry supplementation did not alter these genera in db/db+SB mice compared with db/db mice.

Fig. 3.

The abundance of Bifidobacterium (A), Bacteroides (B), Dehalobacterium (C), Dorea (D), Lactobacillus (E), SMB53 (F), Turicibacter (G) and Akkermansia (H) in db/+, db/db and db/db+SB mice treated for 10 wk. *, P<0.05, values are mean ± SEM, n=8. NS, non-significant; db/+, Standard diet fed control mice; db/db, standard diet fed diabetic mice; db/db+SB, strawberry fed diabetic mice.

3.4. Predicted metabolic pathways

Predicted functional metagenomic profiles based on KEGG pathways were generated using PICRUSt. Comparison between groups revealed significant differences in 45 predicted metabolic functions (p<0.05) (Fig. 4). Most of these features revealed similar abundances between control db/+ compared to db/db+SB. The correlations between bacterial abundance with predicted metagenomic function indicate that G protein-coupled receptors, insulin signaling pathway, fatty acid elongation in mitochondria, glycerophospholipid metabolism, lipid biosynthesis protein, fatty acid biosynthesis and steroid hormone biosynthesis were significantly different among the three groups. Specifically, lipid biosynthesis proteins, insulin signaling pathway, and the phosphatidyl inositol signaling pathway were modified in db/db+SB compared with db/db mice (p<0.05).

Fig. 4.

Heat map for the effect of strawberry supplementation on functional potential of gut microbiome in diabetic mice. Metabolic pathways from KEGG module predictions using 16S rRNA data with PICRUSt and sequenced shotgun metagenome (n=8). NS, non-significant; db/+, Standard diet fed control mice; db/db, standard diet fed diabetic mice; db/db+SB, strawberry fed diabetic mice.

4. Discussion

The gut microbiome plays a pivotal role in the metabolism of anthocyanins and is one of the important mechanisms of the health-promoting properties of anthocyanins. We investigated the effect of dietary supplementation of strawberry on gut microbiota in diabetic mice. In our study, strawberry supplementation at a nutritional dosage induced marked changes in the composition and functional potential of the gut microbiome in diabetic db/db mice.

Evidence from epidemiological and clinical studies indicates beneficial cardiovascular effects following the consumption of strawberries [2, 5, 12, 13]. Our group has also recently reported that dietary supplementation of strawberry attenuates vascular inflammation and improves vascular dysfunction in diabetic mice [10]. It has been speculated that such cardiovascular benefits may be mediated by circulating metabolites of the bioactive compounds of strawberries. These are produced by human digestive enzymes and intestinal microbiota, and highlight the potential importance of microbiota [31–33]. A human study indicated that 21 polyphenolic metabolites appear in the plasma following the consumption of strawberries [31]. We recently assessed the role of circulating metabolites of strawberry in mediating the vascular effects of strawberry. We used serum that was obtained from strawberry fed mice (serum containing circulating metabolites of strawberry) and control diet fed mice (control serum) for this study. In our study, serum from strawberry fed mice reduced high glucose and palmitate-induced endothelial inflammation in mouse aortic endothelial cells indicating the possible role of circulating metabolites in mediating the vascular effects of strawberry [10].

Intestinal microbiota play a major role in the conversion of anthocyanins to metabolites. Therefore, they modulate the biological activities of dietary anthocyanins. The commensal bacteria such as Bifidobacterium and Lactobacillus possess β-glucosidase activity and have the ability to metabolize anthocyanins into phenolic metabolites [34, 35]. Indeed, a recent randomized clinical study showed that high levels of Bifidobacterium are associated with increased urinary concentrations of anthocyanin metabolites [35]. On other hand, anthocyanins act as prebiotics and support the growth of Bifidobaterium and Lactobacillus, indicating a two-way relationship between anthocyanins and gut microbiota [36]. Hence, a healthy microbiome is essential to benefit from the effects of anthocyanins as intestinal microbiota play a key role in the conversion of parent anthocyanins in to metabolites.

In our study, db/db mice exhibited marked changes in the microbial abundance at phylum and genus levels. This is consistent with previous studies that showed the compositional changes in gut microbiota at phylum and genus levels in both Type 1 and type 2 diabetes [37–39]. Actinobacteria, Proteobacteria and Verrucomicrobia were significantly altered among the three groups at the phylum level. There was a significant decrease in the abundance of Actinobacteria in db/db mice compared with db/+ mice. Actinobacteria represent only a small percentage but still are pivotal in the maintenance of gut homeostasis [40]. The classes of Actinobacteria, importantly Bifidobacterium, are widely used as probiotic indicating their beneficial effects in many pathological conditions [40]. Actinobacteria was shown to decrease in type 1 diabetic children compared with healthy children [37]. Though strawberry supplementation increased the Actinobacteria in db/db+SB mice in our study, the difference did not reach significance. Consistent with a previous study, db/db mice exhibited a significant increase in Proteobacteria compared with control mice [38]. A bloom of Proteobacteria in the gut is an indication of an unstable gut microbial community and/or gut dysbiosis [41]. The levels of Proteobacteria was shown to increase in type 2 diabetic mice and diabetic patients [38, 39]. There was a non-significant increase in Verrucomicrobia in db/db compared with db/+ mice but was significantly reduced in db/db+SB mice compared with db/db mice.

At the genus level, many of the bacterial genera were altered in db/db vs db/+ and db/db vs db/db+SB mice. The abundance of Bifidobacterium and Lactobacillus were significantly decreased in db/db mice compared to db/+ mice. Bifidobacterium and Lactobacillus are considered beneficial bacteria and are associated with positive effects for the host in the large intestine. Evidence shows that the etiology and development of type 2 diabetes is closely associated with changes in the gut microbiota including a decrease in the abundance of Bifidobacterium and Lactobacillus [42, 43]. These microbes modulate lipid and glucose metabolism, improve insulin resistance, reduce low-grade inflammation, improve the gut barrier function and stimulate the host immune system [7, 44, 45]. Indeed, the Bifidobacterium abundance was shown to be lower in overweight, obese or type 2 diabetic patients than in lean subjects [46, 47]. Further, a significant decrease in the number of Lactobacillus and Bifidobacterium was reported in children with diabetes [37]. In the present study, strawberry supplementation significantly increased the abundance of Bifidobacterium in db/db+SB mice compared to db/db mice but did not change the abundance of Lactobacillus. A recent study showed that incubating malvidin-3-glucoside, one of the major berry anthocyanins, with fecal slurry enhance the growth of Bifidobacterium and Lactobacillus and exhibited a synergic effect to support the growth of beneficial bacteria when mixed with other anthocyanins [36]. Further, blueberry consumption for 6 weeks was shown to increase Bifidobacterium in human volunteers [48]. These studies indicate that the anthocyanins can act as a prebiotic to support the growth of beneficial gut bacteria such as Bifidobacterium. An increase in the Bifidobacterium can enhance the bioactivity of strawberry as it increases bioavailability of metabolites of anthocyanins [35].

Akkermansia was non-significantly increased in db/db mice compared to db/+ mice but was significantly reduced in db/db+SB mice compared with db/db mice. Studies suggest that Akkermansia muciniphila possess anti-inflammatory properties though the underlying mechanisms are unknown [49]. Akkermansia muciniphila, was shown to protect against atherosclerosis and prevent the development of high-fat diet-induced obesity by improving the gut barrier and metabolic inflammation in animal models [50]. Interestingly in our study strawberry supplementation reduce Akkermansia in db/db+SB mice. Given that we have previously shown that strawberries significantly improve vascular function and indices of vascular inflammation, the biological relevance of Akkermansia in the db/db model remains unclear.

PICRUSt revealed significant differences in 45 predicted metabolic functions among the three groups. Specifically, lipid biosynthesis proteins, the insulin signaling pathway, and the phosphatidyl inositol signaling pathway were modified in db/db+SB vs db/db. Evidence suggests that oral supplementation of prebiotics (fermented oligosaccharides) and/or probiotics may improve metabolic disorders such as obesity and type 2 diabetes [51, 52]. A recent study showed that strawberry extracts modulate the effects of fructooligosaccharaides on microbiota in the gastrointestinal tract [53]. To our knowledge, the present study is the first to show the complex interaction between strawberry and intestinal microbiota in diabetes.

Our findings indicate that dietary supplementation of strawberry at a nutritional dosage induced marked changes in the composition and functional potential of gut microbiota in db/db mice. Importantly, strawberry supplementation increased the abundance of Bifidobacterium, which plays a pivotal role in the metabolism of anthocyanins and the formation of metabolites. Hence, the reported beneficial health effects of strawberry could be due to an increased abundance of Bifidobacterium, which may enhance the metabolism of strawberry and the bioactive metabolites formed by Bifidobacterium metabolism. However, mechanistic studies are still needed to provide evidence for the prebiotic effects of strawberry anthocyanins and to understand how this could modulate the biological activity of metabolites. Our results promote further exploration into the analysis of strawberry metabolites and correlating the functional aspects of strawberries with metabolites and microbiome. In conclusion, our study provides a strong proof of concept for further considering strawberry as an adjunct therapy to improve intestinal microbiota and thereby to prevent or reverse the complications associated with diabetes.

Highlights.

• Gut microbiota convert berry anthocyanins into bioactive metabolites

• Anthocyanins support the growth of specific bacteria

• Strawberry increased beneficial bacteria Bifidobacterium in diabetic mice

• Strawberry induces changes in the composition of gut microbiome in diabetic mice

• Strawberry induces marked changes in the functional potential of gut microbiome