Gut Microbiome and Metabolome Modulation by High-Hydrostatic-Pressure-Processed Tomato Juice

High hydrostatic pressure (HHP) is a non-thermal pasteurization technology for the enhancement of food products’ safety and quality. The components of tomato juice can be affected by HHP processing. Little is known about the effects of HHP-processed tomato juice on the gut microbiome and metabolism. Here, we performed high-throughput sequencing and metabolomics profiling to determine the critical differences in gut microbiota structure and metabolic profiles in mice administered with HHP-processed tomato juice. Tomato juice administration significantly increased the gut bacterial alpha diversity and the relative abundance of Bacteroides. The mice administered with HHP-processed tomato juice were characterized by the enrichment of Bacteroidetes, Alistieps, and Faecalibaculum compared with those administered with HTST-processed tomato juice. Moreover, HHP-processed tomato juice promoted SCFA levels, which were positively correlated with the enriched Alistieps. Our results show that HHP-processed tomato juice may drive healthy gut microbes and metabolites.


Introduction
The demand to make food healthier, safer, tastier, and more shelf stable promotes the development of innovative technology.As one of the most important non-thermal processing technologies, high hydrostatic pressure (HHP) can inactivate foodborne microorganisms and enzymes to extend food products' shelf-life.Compared to conventional thermal processing, HHP processing can maintain the nutritional and sensory properties of fresh products due to the lower treatment temperature [1].Therefore, HHP processing is commonly used to manufacture and preserve fruit-and vegetable-based products which maintain diverse bioactive compounds, flavors, and colors [2].A study showed that HHP processing (600 MPa, 5 min) preserved greater levels of color, total phenolics, total flavonoid, and antioxidant properties of grapefruit juice than thermal treatment at 85 • C for 45 s [3].Another study reported that the contents of volatile compounds (aldehydes and alcohols) were higher in mulberry juice subjected to HHP processing at 500 MPa for 10 min compared to that processed by thermal treatment [4].In addition, a randomized cross-over clinical trial study revealed that the glycemic index was improved after consumption of HHP-processed mango puree [5].HHP processing promotes the preservation of functional components and may potentially enhance the nutritional value of food products for human health.There is an urgent need to evaluate the health-promoting impacts of HHP-processed food products on human health.processed tomato juice (HTST group), HHP-processed tomato juice (HHP group), and sterile saline (Blank group) at a dose of 10 mL/kg by oral gavage twice daily after 1 week of acclimatization.After 4 weeks, mice were killed after 12 h of food deprivation.Serum samples were obtained from blood samples with centrifugation (3000 rpm, 4 • C, 15 min).Cecum content samples were collected in sterilized tubes and then stored at −80 • C.
Animal experiments were conducted under the approval number AW09211202-4-1 at the Laboratory Animal Centre of China Agricultural University (Beijing, China) based on national legislation and local guidelines.

Blood Biochemical Assay
Low-density lipoprotein, high-density lipoprotein, total cholesterol, total triglycerides, alanine aminotransferase, and aspartate aminotransferase in serum were determined by a biochemical analyzer (AU480, Japan Olympus Corporation, Tokyo, Japan).

Metabolomics Analysis
The metabolomics study was carried out by Majorbio Bio-Pharm Technology (Shanghai, China).The fecal sample (50 mg) was mixed with 400 µL extraction solution (methanol:water = 4:1 (v/v), containing internal standard), treated with a high-throughput tissue crusher at 50 Hz for 6 min (−10 • C) and then at 40 kHz for 30 min, placed at −20 • C for 30 min, and centrifugated at 13,000× g for 15 min (4 • C).The supernatant was collected for analysis.

Statistical Analysis
The Kruskal-Wallis test was performed to compare medians between nonnormally distributed groups.The results from the experiments were calculated by a one-way ANOVA analysis.We used GraphPad Prism7.0 for graphing.For multivariate analysis, Principal Component Analysis (PCA) and Orthogonal Least-Partial-Squares Discriminant Analysis (OPLS-DA) were carried out by the R software package ropls (version 1.6.2).One-way PERMANOVA analysis was calculated by Past 4.02 (http://folk.uio.no/ohammer/past(accessed on 10 November 2020)).Variable importance in the projection (VIP) ≥ 1 and p < 0.05 were set as screening criteria for differential metabolite screening.Kyoto Encyclopedia of Genes and Genomes (KEGG, http://www.genome.jp/kegg/(accessed on 25 September 2020)) was applied to analyze metabolic enrichment and pathways.Interventionary studies involving animals or humans, and other studies that require ethical approval, must list the authority that provided approval and the corresponding ethical approval code.

Effect of Tomato Juice on Biochemical Parameters
There was no significant difference between mice administered with fresh, HHP-, and HTST-treated tomato juice in body weight, food consumption, and water intake (Figure S1).Consistently, there was no significant difference in the serum parameters, including lowdensity lipoprotein, high-density lipoprotein, total cholesterol, total triglycerides, alanine aminotransferase, and aspartate aminotransferase, between the treatment groups (Table S2).

Effect of Tomato Juice on Gut Microbiota Composition
To investigate whether administration with fresh, HHP-, and HTST-treated tomato juice altered the composition of the gut microbiota in mice, 16S rRNA gene sequencing was used.A total of 4,171,550 raw reads were picked up from 48 samples, averaging 86,907 ± 9121 reads per sample.A total of 1943 operational taxonomic units (OTUs) were determined (97% similarity) after quality checks, with an average of 431 ± 77 OTUs per sample.
The bacterial alpha diversity refers to the bacterial species (OTUs) richness and evenness within a given region or ecosystem.The Chao index, ACE index, and observed species are measured to reflect species richness, while the Shannon index is used to reflect community evenness.As shown in Figure 1, the Shannon index was significantly higher in the three tomato juice administration groups than in the control group, suggesting that tomato juice administration increased the microbial community evenness in the mice.Interestingly, the observed species, ACE index, and Chao1 index were enhanced in mice administered with HHP-and HTST-treated tomato juice compared to the mice administered with fresh tomato juice, while there was no significant difference in the bacterial α diversity between the HHP-and HTST-treated groups (Figure 1).
PCoA based on OTUs showed that the groups administered with fresh, HHP-, and HTST-treated tomato juice displayed a similar clustering of the gut bacterial community relative to the control group.Notably, the PERMANOVA test showed a significant difference in the gut bacterial structure of these tomato juice treatment groups (p < 0.01) (Figure 1E).UPGMA clustering tree analysis revealed a significant difference in the bacterial β diversity between the HHP-and HTST-processed tomato juice groups (Figure 1F).Thus, these data suggest that administration of tomato juice, especially the tomato juice processed by HTST and HHP, significantly changed the structure of the intestinal microbial community in the mice.
PCA analysis showed that there was an apparent clustering of metabolic profiles in mice administered with the fresh, HHP-treated, and HTST-treated tomato juice in comparison with the control mice, suggesting that tomato juice administration altered the fecal metabolome in the mice (Figure 3A,B).Although no significant difference in metabolic profiles was found between the HHP-and HTST-treated tomato juice groups, PLS-DA analysis revealed dramatic metabolic differences between each tomato juice treatment group and the control group (Figure 3C-H).
The levels of SCFAs in feces were measured to evaluate the effect of HHP-and HTST-processed tomato juice on intestinal homeostasis.Our results demonstrated that the concentration of acetic, propionic, butyric acids, and total SCFAs was higher in the HHPtreated tomato juice group than in the control and other tomato juice groups (Figure 6A).Spearman rank correlation analyses further revealed the relationship between alterations in the abundance of different bacterial genera and the content of SCFAs (Figure 6B).It is noteworthy that the contents of acetic acid, butyric acid, and propionic acid were significantly and positively correlated with the abundance of Alistieps, which was bacterially enriched in mice administered with HHP-processed tomato juice.In contrast, the contents of these SCFAs were negatively associated with the abundance of Staphylococcus, which was enriched in mice administered with HTST-processed tomato juice (Figure 6C).Thus, the increased abundance of Alistieps and reduced abundance of Staphylococcus may have contributed to the higher SCFA levels in the HHP-treated tomato juice administration group compared to the HTST-processed tomato juice administration group.The levels of SCFAs in feces were measured to evaluate the effect of HHP-and HTSTprocessed tomato juice on intestinal homeostasis.Our results demonstrated that the concentration of acetic, propionic, butyric acids, and total SCFAs was higher in the HHPtreated tomato juice group than in the control and other tomato juice groups (Figure 6A).Spearman rank correlation analyses further revealed the relationship between alterations in the abundance of different bacterial genera and the content of SCFAs (Figure 6B).It is noteworthy that the contents of acetic acid, butyric acid, and propionic acid were significantly and positively correlated with the abundance of Alistieps, which was bacterially enriched in mice administered with HHP-processed tomato juice.In contrast, the contents of these SCFAs were negatively associated with the abundance of Staphylococcus, which was enriched in mice administered with HTST-processed tomato juice (Figure 6C).Thus, the increased abundance of Alistieps and reduced abundance of Staphylococcus may have

Discussion
The beneficial impact of HHP processing on fruit juice has long been described, such as retention of better sensory and nutritional compounds than conventional thermal processing [2].Studies have shown that the intestinal microbiome plays a significant role in host health.However, no study has compared the effects of HHP-treated and thermaltreated tomato juice on the gut microbiome and metabolome.This study demonstrated that the administration of tomato juice affects the structure of gut microbiota and metabolic profiles in mice.HHP-processed tomato juice promoted the enrichment of Alistieps and SCFAs, which can contribute to the physiological activities of host health.
According to the World Health Organization, a daily intake of 400-600 g of fruits and vegetables is recommended to decrease the risk of noncommunicable diseases and pre-

Discussion
The beneficial impact of HHP processing on fruit juice has long been described, such as retention of better sensory and nutritional compounds than conventional thermal processing [2].Studies have shown that the intestinal microbiome plays a significant role in host health.However, no study has compared the effects of HHP-treated and thermaltreated tomato juice on the gut microbiome and metabolome.This study demonstrated that the administration of tomato juice affects the structure of gut microbiota and metabolic profiles in mice.HHP-processed tomato juice promoted the enrichment of Alistieps and SCFAs, which can contribute to the physiological activities of host health.
According to the World Health Organization, a daily intake of 400-600 g of fruits and vegetables is recommended to decrease the risk of noncommunicable diseases and prevent malnutrition.Moderate fruit juice consumption (75-224 mL/d) is consistent with dietary health guidelines for the U.S. and several European countries [18].In this study, 150 mL/d of tomato juice intake was selected as the human dose (60 kg) to convert into the mice equivalent dose (20 mL/kg/d) based on body surface area.A previous study also observed that replacing drinking water with tomato juice did not affect weight, food and water intake, or ALT and AST activities in the liver of SD rats [16].Our results also showed that 4 weeks of tomato juice administration (20 mL/kg/d) had no effects on the body weight and serum parameters of mice.Thus, the human equivalent dose of tomato juice at a dose of 150 mL per day for adults could be recommended as no-weight-gain energy intake.
Diet is a major factor influencing host gut microbiota [7].A previous study found that two weeks of tomato powder administration significantly increased the number of OTUs and the Shannon index [15].Moreover, a decreased F/B value was observed in colitis mice with tomato powder supplementation [19].It was reported that obese animals and humans had a higher abundance of Firmicutes and a lower abundance of Bacteroidetes, and the F/B value normalized with concomitant weight loss following a calorie-restricted diet [20].In addition, Krajmalnik-Brown et al. [21] proposed that Firmicutes are more efficient at extracting energy from food than Bacteroidetes, making them more efficient at absorption.Tomato juice changed the structure of gut microbiota in mice by increasing the bacterial alpha diversity and decreasing the F/B value observed in our study.Our results suggest that tomato juice may be developed as a promising dietary supplement for the prevention of obesity through the reduction of the F/B value.
The relative abundance of Staphylococcus decreased, and the abundance of Faecalibaculum, Alistipes, Alloprevotella, Bacteroides, Dubosiella, unidentified-Ruminococcaceae, Blautia, Butyricicoccus, and Akkermansia increased in tomato juice administration groups compared to in the control group.In addition, Ruminococcaceae were the bacteria commonly enriched in the three tomato juice groups.Staphylococcus aureus is the primary pathogen that causes human clinical infection, while Staphylococcus epidermidis occasionally causes diseases.Faecalibaculum rodentium could slow down tumor growth by producing shortchain fatty acids [22].Alloprevotella, Bacteroides, Ruminococcaceae, Blautia, Butyricicoccus, and Akkermansia are producers of SCFAs [23][24][25][26][27][28].Alistipes was correlated with prolonged skin graft survival in mice [29], and high-fat diet-induced intestinal microbiota dysbiosis was reversed by the increased abundance of Alloprevotella [30].Bacteroides distasonis, Bacteroides uniformis, and Bacteroides ovatus are involved in the hydrolysis of glycosidic bonds of flavonols (kaempferol, quercetin, etc.) [23].Blautia is involved in the deglycosylation of polyphenols and the catabolism of lignans; its metabolites (such as butyric acid) contribute to the relief of inflammatory and metabolic diseases [25].The increase in the relative abundance of Dubosella helps to prevent the development of salt-sensitive hypertension [31], and the abundance of Ruminococcaceae was significantly negatively correlated with the occurrence of alcoholic liver cirrhosis, hepatic encephalopathy, and non-alcoholic fatty liver disease [26].Butyricicoccus supplementation alleviated colitis in rats, and the supernatant of butyric acid bacteria culture enhanced intestinal epithelial barrier function [27].Akkermansia muciniphila is an intestinal symbiotic bacterium that colonizes the mucosal layer and has a high value in improving host metabolic function and immune response, as well as altering cancer treatment [28].Previous studies found that tomato pomace decreased the abundance of Escherichia coli in vitro [32], and lycopene increased the bacterial alpha diversity and reduced the abundance of Clostridium in mice on a high-fat diet [33].Moreover, tomato juice increased mice fecal concentrations of food-derived phenolics, which are essential in modifying the gut microbiota [16].These results indicated that lycopene, dietary fiber, and polyphenols might contribute to the impact of tomato juice on gut bacterial community structure.
Diet regulates the structure of the gut microbial community.In turn, the gut microorganisms metabolize dietary components not utilized by the host, thereby influencing the co-metabolism of the host and its gut microbiota [8].Metabolites, including N-acetyl-Lglutamate, (2S,3R)-2-aminooctadecane-1,3-diol, N-acetyl-L-glutamic acid, dihydroceramide, L-glutamate, L-arginine, and D-ornithine, were the common differential metabolites of the three tomato juice intervention groups.Notably, these metabolites were involved in sphingolipid metabolism, arginine biosynthesis, D-arginine, and D-ornithine metabolism.It has been reported that sphingolipids can be obtained from the diet by de novo synthesis in mammalian tissues.Moreover, the metabolism of gut bacteria Bacteroidetes is also considered an essential source of sphingolipids.Bacteroides and Prevotella, which possess serine palmitoyltransferase, are the Bacteroidetes that can synthesize sphingolipids from free sphingosine [34].A significantly increased ceramide level was found in the liver of insulin-resistant mice supplemented with Bacteroides thetaiotaomicron [34].Moreover, Bacteroides-derived sphingolipid levels were lower in IBD patients than in the average population.Bacteroides-derived sphingolipids, such as ceramide phosphatidol and deoxysphingolipids, were negatively correlated with inflammation and were influential in the maintenance of intestinal homeostasis and symbiotics [35].Another study demonstrated that glycerophospholipids were broken down by phospholipases to generate arachidonic acid, while sphingomyelinases could regulate arachidonic acid to produce ceramide and phosphorylcholine [36].The relative abundance of Bacteroides was positively correlated with PI(16:0/18:1(11Z)), dihydroceramid, and SM(d18:1/12:0), and negatively correlated with glycerol-3-phosphate, LysoPC(22:6(4Z,7Z,10Z,13Z,16Z,19Z), phytosphingosine, and LysoPC(20:4(8Z,11Z,14Z,17Z), in our study.Thus, the significant increase in Bacteroides may have contributed to the enrichment of sphingolipid metabolism in mice after tomato juice intervention.Previous studies indicated that Clostridia is the most common species involved in the fermentation of amino acids [37], and Corynebacterium glutamicum is a glutamate producer [38].The reasons for the enriched arginine biosynthesis and D-arginine after tomato juice administration might be related to the rise of Unidentified_Clostridiales, and the Dornithine metabolism change might be related to the enriched Corynebacterium glutamicum.
A primary finding of our study was that HHP-and HTST-processed tomato juice resulted in different gut microbiota structures and metabolic profiles of mice.HHP-processed tomato juice administration enhanced the relative abundance of Bacteroides and Alistipes and promoted SCFA production compared with HTST-processed tomato juice administration.SCFAs including acetic, lactic, propionic, n-butyric, i-butyric, n-valeric, and i-valeric acids are the essential fermentation products of gut microbiota that have a potential beneficial effect on human health [39].Gomez et al. [40] showed that supplementation with high-dose lycopene (12 mg/100 g) tomato juice had no significant effect on the level of SCFAs in the intestinal system of rats, but low-dose (2.7 mg/100 g) lycopene increased butyrate levels.Our previous study observed that the concentration of total lycopene in HHP-processed tomato juice (7.98 mg/100 g) was higher than in HTST-processed juices (5.45 mg/100 g) [17].Thus, the increased production of SCFAs in the HHP-processed tomato juice group might be related to the elevated concentration of lycopene content in tomato juice following HHP processing.Additionally, it has been reported that insoluble dietary fiber extracted from carrot pomace and mango pulp was modified by HHP [41].Microbial fermentation can be affected by dietary fibers' chemical composition and physicochemical properties [42].All types of pectins could increase the level of acetic acid, but butyric acid was only enriched by pectin L13 [43].Hence, the dietary fiber in tomato juice modified by HHP treatment was another possible reason for increased SCFA production.Notably, the levels of acetic acid, butyric acid, and propionic acid were significantly positively correlated with the relative abundance of Alistieps, which have been reported to produce acetic acids and iso-valeric acids [44].It has also been reported that Alistipes has a beneficial role in cancer immunotherapy [29].Staphylococcus is a potentially pathogenic bacterium, and studies found that Staphylococcus increased in the gut of inflammatory bowel disease and chronic rheumatic disease patients compared with ordinary people [45].SCFAs can decrease Staphylococcus aureus internalization in the mammary glands [46].Thus, the higher abundance of Alistipes might have promoted the accumulation of SCFAs, which, in turn, led to the lower abundance of Staphylococcus in mice administered with HHP-processed tomato juice.However, these hypotheses still need to be proven by further experiments, and the specific mechanisms that regulate the structure of the gut microbiota in mice treated with HHP-treated tomato juice still need to be further explored.

Conclusions
Our study demonstrated that tomato juice administration increased the gut microbiota α diversity and relative abundance of Bacteroides and regulated sphingolipid metabolism and arginine biosynthesis in mice.Moreover, we found that HHP-processed tomato juice increased the abundance of Alistipes and promoted the production of SCFAs compared with HTST-processed tomato juice.Our results provide a new insight into the advantage of HHP over HTST processing.This will benefit the application of non-thermal processing technology for the improvement of the gut microenvironment.

Supplementary Materials:
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/nu16050710/s1, Figure S1: (A) The body weight, (B) food consumption, and (C) water intake of mice administered with tomato juice.; Figure S2: Linear discriminant analysis effect size analysis to identify taxonomic differences in the gut microbiota of the mice administered with different treated tomato juice.Histograms of LDA scores in (A) Fresh and Blank, (B) HTST and Blank, and (C) HHP and Blank (LDA score (log10) ≥ 4.0) groups.Taxa enriched in the Blank group are displayed by a red bar (negative LDA score), and taxa enriched in the Fresh, HTST, and HHP tomato juice groups are described by a green bar (positive LDA score); Figure S3: Classification of mice fecal metabolites based on (A) KEGG library and (B) HMDB library.The horizontal axis represents the KEGG compounds classification, and the vertical axis represents the number of compounds in that category.The selected HMDB classifications (each pie chart) and the percentage of metabolites (each area in a pie chart) are displayed in descending order by the number of metabolites.Table S1: Chromatographic conditions; Table S2: Serum biochemical parameters in mice administered with tomato juice; Table S3: Differential metabolites discriminated between mice administered fresh, HPP-, and HTST-processed tomato juice.Institutional Review Board Statement: The animal study protocol was approved by the Institutional Review Board of the Laboratory Animal Centre of China Agricultural University (approval code: no.AW09211202-4-1, approval date: 1 August 2019).

Figure 4 .
Figure 4. Differential metabolite analysis based on the top-30 expression profile and functional enrichment analysis of KEGG topology analyses; bubble graphs show comparison between control group and the (A,B) fresh, (C,D) HTST-processed, (E,F) HHP-processed tomato juice groups (* p<0.05, ** p <0.01, *** p <0.001).The Y axis is the impact value of the relative importance of metabolites in the pathway, and the X axis is the enrichment of significant metabolites involved in the pathway

Figure 4 .
Figure 4. Differential metabolite analysis based on the top-30 expression profile and functional enrichment analysis of KEGG topology analyses; bubble graphs show comparison between control group and the (A,B) fresh, (C,D) HTST-processed, (E,F) HHP-processed tomato juice groups (* p < 0.05, ** p < 0.01, *** p < 0.001).The Y axis is the impact value of the relative importance of metabolites in the pathway, and the X axis is the enrichment of significant metabolites involved in the pathway −log10 (p-value).The bubble represents a KEGG pathway, and its size indicates the impact value.The text with the pathway name represents the significant corrected p-value (<0.05).

Figure 5 .
Figure 5.The Spearman correlation heatmap analysis between feature metabolites and gut bacteria.A significant correlation is indicated by an asterisk symbol (* p ≤ 0.05, ** p ≤ 0.01).

Figure 5 .
Figure 5.The Spearman correlation heatmap analysis between feature metabolites and gut bacteria.A significant correlation is indicated by an asterisk symbol (* p ≤ 0.05, ** p ≤ 0.01).

Figure 6 .
Figure 6.HHP-processed tomato juice increases SCFA concentrations and relative abundance of specific bacteria.(A) Acetic acid, propionic acid, isobutyric acid, butyric acid, isovaleric acid, valeric acid, and total short-chain fatty acid content; (B) correlation diagram of SCFAs and gut bacteria; (C) linear discriminant analysis effect size analysis to identify taxonomic differences in the gut microbiota of the mice administered with HHP-and HTST-treated tomato juice.Different letters indicate statistical differences between groups (* p ≤ 0.05, ** p ≤ 0.01).

Figure 6 .
Figure 6.HHP-processed tomato juice increases SCFA concentrations and relative abundance of specific bacteria.(A) Acetic acid, propionic acid, isobutyric acid, butyric acid, isovaleric acid, valeric acid, and total short-chain fatty acid content; (B) correlation diagram of SCFAs and gut bacteria; (C) linear discriminant analysis effect size analysis to identify taxonomic differences in the gut microbiota of the mice administered with HHP-and HTST-treated tomato juice.Different letters indicate statistical differences between groups (* p ≤ 0.05, ** p ≤ 0.01).