nutrients

Article
High Fat-High Fructose Diet-Induced Changes in the
Gut Microbiota Associated with Dyslipidemia in
Syrian Hamsters
Rachael G. Horne 1,† , Yijing Yu 2,† , Rianna Zhang 2 , Nyan Abdalqadir 2,3 , Laura Rossi 4 ,
Michael Surette 4 , Philip M. Sherman 1,3,5 and Khosrow Adeli 2,3,6,7, *
1 Cell Biology, Research Institute, Hospital for Sick Children, Toronto, ON M5G 0A4, Canada;
[email protected] (R.G.H.); [email protected] (P.M.S.)
2 Molecular Medicine Programs, Research Institute, Hospital for Sick Children, Toronto,
ON M5G 0A4, Canada; [email protected] (Y.Y.); [email protected] (R.Z.);
[email protected] (N.A.)
3 Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
4 Department of Biochemistry and Biomedical Sciences, McMaster University Medical Centre, Hamilton,
ON L8N 3Z5, Canada; [email protected] (L.R.); [email protected] (M.S.)
5 Department of Paediatrics, University of Toronto, Toronto, ON M5S 1A8, Canada
6 Department of Biochemistry, University of Toronto, Toronto, ON M5S 1A8, Canada
7 Department of Physiology, University of Toronto, Toronto, ON M5S 1A8, Canada
* Correspondence: [email protected]
† Both authors contributed equally to this work.




Received: 26 October 2020; Accepted: 17 November 2020; Published: 20 November 2020


Abstract: Aim: The objective of this study was to characterize the early effects of high
fructose diets (with and without high fat) on both the composition of the gut microbiota and lipid
metabolism in Syrian hamsters, a reproducible preclinical model of diet-induced dyslipidemia.
Methods: Eight-week-old male hamsters were fed diets consisting of high-fat/high-fructose,
low-fat/high-fructose or a standard chow diet for 14 days. Stool was collected at baseline (day 0), day 7
and day 14. Fasting levels of plasma triglycerides and cholesterol were monitored on day 0, day 7 and
day 14, and nonfasting levels were also assayed on day 15. Then, 16S rRNA sequencing of stool samples
was used to determine gut microbial composition, and predictive metagenomics was performed to
evaluate dietary-induced shifts in deduced microbial functions. Results: Both high-fructose diets
resulted in divergent gut microbiota composition. A high-fat/high-fructose diet induced the largest
shift in overall gut microbial composition, with dramatic shifts in the Firmicute/Bacteroidetes ratio,
and changes in beta diversity after just seven days of dietary intervention. Significant associations
between genus level taxa and dietary intervention were identified, including an association with
Ruminococceace NK4A214 group in high-fat/high-fructose fed animals and an association with
Butryimonas with the low-fat/high-fructose diet. High-fat/high-fructose feeding induced dyslipidemia
with increases in plasma triglycerides and cholesterol, and hepatomegaly. Dietary-induced changes
in several genus level taxa significantly correlated with lipid levels over the two-week period.
Differences in microbial metabolic pathways between high-fat/high-fructose and low-fat/high-fructose
diet fed hamsters were identified, and several of these pathways also correlated with lipid profiles
in hamsters. Conclusions: The high-fat/high-fructose diet caused shifts in the host gut microbiota.
These dietary-induced alterations in gut microbial composition were linked to changes in the
production of secondary metabolites, which contributed to the development of metabolic syndrome
in the host.

Keywords: microbial metabolites; lipids; metabolism; metabolic dysfunction; 16S rRNA

Nutrients 2020, 12, 3557; doi:10.3390/nu12113557 www.mdpi.com/journal/nutrients

Nutrients 2020, 12, 3557 2 of 20

1. Introduction
Human gut microbiota, comprising of bacteria, viruses, phages, fungi and protists, play a profound
role in mammalian health, contributing not only to host metabolism [1], but also to mucosal and
systemic immune responses [2] and the metabolism of xenobiotics [3]. Numerous studies have found
that the impact of the gut microbiota on host health relates to overall microbial composition and to
bacterial diversity [4,5].
Over the last two decades, studies have focused on understanding the links between changes
in the gut microbiota and the development of metabolic syndromes in humans, such as diabetes,
obesity, and nonalcoholic fatty liver disease [1,6,7]. Both murine models and human studies reveal
strong associations between alterations in the relative proportions of dominate gut phyla (that is,
Bacteroidetes and Firmicutes) and obesity [8,9]. Changes in the gut microbiota are associated with host
metabolic conditions, including changes in lipid profiles (such as low levels of high-density lipoprotein
and increases in cholesterol and triglycerides) [10]. Recent work has determined one of the metabolic
links between the gut microbiota and host cholesterol levels, with a microbial enzyme identified as
a key factor in metabolizing dietary cholesterol, and subsequently limiting cholesterol absorption [11].
These findings indicate the critical role the gut microbiota play in maintaining host lipid homeostasis.
Animal studies have found that varying levels of dietary fat and carbohydrates can have profound effects
on the gut microbiota and host metabolism [12–14]. Specifically, a high-fat, high-sugar and low-fiber diet,
commonly referred to as the “Western diet”, results in a shift in the gut microbiota composition,
leading to increased lipopolysaccharide in circulation, causing pro-inflammatory responses that
precede the development of insulin resistance and obesity [15]. The use of dietary interventions
has also been explored as a means of correcting metabolic dysfunction in humans with decreased
levels of triglycerides and low-density lipoproteins while improving microbial dysbiosis in nearly half
of patients [16].
While previous studies reported links between diet manipulations, changes in the composition of
gut microbiota, and subsequent metabolic disease, the underlying mechanisms are only now beginning
to be delineated. Specifically, dietary manipulations have been found to impact levels of specific
bacterial species within the gut microbiota, which are known to produce a range of bioactive metabolites,
such as short-chain fatty acids, amino acids and vitamins [17].
Fructose, a common component of the western diet, has been shown to have profound effects
on the liver, with high fructose diets shown to contribute to the development of nonalcoholic fatty
liver disease [18]. High fructose diets have also been shown to cause shifts in the gut microbiota
composition, reducing the abundance of beneficial Bifidobacterium and Lactobacillus [19]. In addition,
high levels of fructose play a direct role in liver lipogenesis by increasing microbial short-chain fatty acid
acetate [20], whereas diets high in fat reduce beneficial short-chain fatty acid butyrate [21]. In addition
to the complex relationship between diet and microbial production of bioactive short-chain fatty acids,
a bidirectional interaction between amino acids and gut microbiota has also been observed [22]. The gut
microbiota aids in maintaining homeostasis of host amino acids by synthesizing several essential
amino acids for the host, while microbial produced amino acids can also provide nutrients for the
growth of amino acid-dependent commensal intestinal bacteria [23]. Diet composition influences
microbial production of amino acids, with high-fat diets resulting in increases in microbial production
of the branched-chain amino acids, i.e., isoleucine, valine and leucine [24], which are known to play
an essential role in host lipid metabolism [25].
While much of the foundational work regarding how diet and the gut microbiota affect host
metabolism has been performed in rodents, few studies have yet to be undertaken on Syrian hamsters.
Importantly, Syrian hamsters have a reproducible response to dietary manipulation [26–28] and lipid
metabolism that closely resembles that of humans [29]. Specifically, Syrian hamsters, like humans,
but unlike mice and rats, have cholesterol ester transport protein activity [30], and respond to diets
high in fat and cholesterol with the major plasma cholesterol being LDL-cholesterol (LDL-C) [26–28,31].
In addition to previous work on rodents, many previous studies have focused on the long-term
Nutrients 2020, 12, 3557 3 of 20

effect of dietary manipulation, often evaluating the impact of months-long exposure to high fat and
high carbohydrate diets [13,14,32,33]. In this study, we provided either a high-fat/high-fructose or
low-fat/high-fructose diet to Syrian hamsters for a period of two weeks, in order to assess potential links
between diet, gut microbiota, and host lipid metabolism in the early stages of metabolic derangement.

2. Materials and Methods

2.1. Animal Handling and Experimental Design

Syrian male hamsters with ~100 g body weight were purchased from Envigo (Mississauga, ON,
Canada), acclimatized for one week after arrival and fed standard rodent chow (PicoLab rodent diet;
catalog # 5058, St. Louis, MO, USA) prior to experimental manipulation. Hamsters were individually
housed in bedded cages with access to environment enrichment objects. Cages were situated in
a temperature-controlled room with a 12 h light-dark cycle. Animals were monitored closely for
changes in body weight and baseline triglyceride levels prior to randomization. Animals were allowed
free access to both food and water for the duration of the study protocol. The food from all diet groups
was weighed every seven days to monitor consumption, and fresh chow was provided. Body weight
was measured at baseline on day 0, and then again after 2 weeks of dietary intervention on experimental
day 14. Food intake of individual hamsters was monitored weekly over the course of the experiment.
This study was conducted in accordance with the Declaration of the Hospital for Sick Children, Toronto,
ON, Canada, and the protocol approved by the Animal Care and Use Committee (project identification
code #46136).

2.2. Experimental Diets

Hamsters were randomly divided into three groups (n = 9 per group) fed with varying diets
for two weeks: (1) low-fat/high-fructose (HF) diet, (2) high-fat/high-fructose with 0.05% cholesterol
(FFLC) diet, or (3) standard proprietary hamster diet (Chow) (PicoLab catolog #007689) (Table 1).
Customized diets consisting of high sugar, high fat diet with 0.05% cholesterol (FFLC) (catalog # 180507)
and 22% casein modified hamster diet with 60% fructose (HF) (catalog #161506) were purchased from
Dyets Inc. (Bethlehem, PA, USA). A detailed comparison of experimental diet components can be
found in Table 1A detailed comparison of experimental diet components can be found in Table 1 and
Supplementary Table S1.

Table 1. Dietary protein, fat, carbohydrate, and total energy of each diet employed in the current study.

kcal/kg g/kg
Chow FFLC HF Chow FFLC HF
Protein 872 613 796 218 171 222
Fat 819 2700 540 91 300 60
Carbohydrate 2072 1572 2280 518 414 600
Gross Energy 4600 4971 3655

2.3. Fecal, Blood and Tissue Collection

On days 0, 7 and 14 after the start of dietary intervention, hamsters were fasted for 5 h before
blood was drawn into lithium heparin-coated tubes via saphenous veins. Blood samples were
centrifuged at 6000 rpm for 10 min at 4 ◦ C, plasma was collected, and a mixture of protease inhibitors
(SIGMAFAST™, Sigma-Aldrich, catalog # S8830-20TAB, St. Louis, MO, USA) added. All chemical
assays were performed using plasma samples on the day of collection, with remaining plasma stored
at −20 ◦ C. Fecal samples were collected on concurrent days and stored at −80 ◦ C. At day 15 post diet
manipulation, hamsters were euthanized via exsanguination under isoflurane anesthetic by cardiac
puncture. Liver and adipose tissues were then harvested, weighed and flash frozen.
Nutrients 2020, 12, 3557 4 of 20

2.4. Fasting Triglycerides and Cholesterol

Concentrations of plasma triglycerides (TG) and cholesterol (CH) were measured using commercial
assay kits (ESBE-Randox, catalog # TR213, # CH200, respectively, Markham, ON, Canada). Briefly,
2 µL of each plasma sample plus 200 µL of reaction buffer were loaded in 96-well plates. After a 10
min incubation at room temperature, the colorimetric reactions were measured at a wavelength of 500
nm using a plate reader (Molecular Devices VersaMax 190, San Jose, CA, USA).

2.5. 16S rRNA Gene Sequencing

DNA was extracted from stool pellets, as previously described [34]. Sequences of the 16S rRNA
gene variable 3–4 (V3–V4) regions were amplified using modifications previously described [34,35],
and sequenced using the Illumina MiSeq platform. Primer and adaptor sequences were trimmed
from the resulting sequences using Cutadapt [36]. DADA2 was used to filter and trim paired reads.
Reads were discarded if the quality score was 2 or less, and if reads aligned to Phix human genome.
DADA2 error correction was performed for each paired reads, de-noised reads merged, and any
identified chimeric sequences then removed. Taxonomy was assigned to the resulting Amplicon
Sequence Variants (ASVs) using RDP classifier trained with Silva v123 16S rRNA database [37].
Alpha and beta diversity analyses were performed on amplicon sequence variant (ASV) table using
phyloseq and vegan packages in R v3.5.6. The predicted metagenomic function of the gut microbiota
composition was performed using Phylogenetic Investigation of Communities by Reconstruction of
Unobserved States (Picrust2) [38].

2.6. Statistical Analyses

Statistical analysis of 16S rRNA data was performed in R v3.5.6. Alpha diversity was assessed
using linear regression, with diet and time as the fixed factors, animal ID as a random factor, and total
reads per sample as an offset. Pairwise comparisons were assessed using the emmeans package in R,
and multiple comparisons were corrected with Tukey’s post hoc adjustment. Negative binominal
mixed model regression analysis was performed on the top 50 most abundant ASVs assigned to
the genus level using NBZIMM r package (https://github.com/nyiuab/NBZIMM) [39]. Models were
designed as follows: read counts of the selected ASV were used as the predicted variable, with dietary
group and day used as the fixed factors, animal ID as a random factor, and the log of the total reads
per sample used as an offset in the individual models. Models were evaluated using maximum
likelihood testing and analysis of variance, significant interaction between dietary groups across time
were evaluated by employing Tukey’s post hoc comparisons. Predicted microbial functional pathway
abundance was normalized using total sum scaling, and the Kruskal Wallis test with False discovery
rate (FDR) used for multiple comparison correction to assess changes in pathway abundance across
days within treatment groups. Two-sided permutation Student’s t-test with FDR was used to assess
differences in pathway abundance between groups at day 7 and day 14. Spearman correlation between
microbial abundance and metabolic biomarkers was performed using a center log ratio transformed
ASV table filtered to retain the top 25 ASVs with genus level taxonomy, and then corrected for multiple
comparisons using FDR. The same approach was used for correlation of predicted pathway abundance;
however, analysis was restricted to the top 200 abundant pathways. All other statistical analyses were
performed using GraphPad prism software (GraphPad, San Diego, CA, USA), using one-way and
two-way ANOVA where appropriate with Tukey’s post hoc testing. Data normality was assessed by
Shapiro-Wilk test and results are presented as means, ±SEM.

3. Results

3.1. High-Fructose Diets Do Not Have a Significant Impact on the Weight Gain of Hamsters
Hamsters were fed with either low-fat/high-fructose, high-fat/high-fructose, or control diet (chow)
for two weeks, and the body weight of each hamster was measured before and after the two-week
3. Results

3.1. High-Fructose Diets Do Not Have a Significant Impact on the Weight Gain of Hamsters
Nutrients 2020, 12, 3557 5 of 20
Hamsters were fed with either low-fat/high-fructose, high-fat/high-fructose, or control diet
(chow) for two weeks, and the body weight of each hamster was measured before and after the two-
week dietary
dietary intervention.
intervention. No significant
No significant difference
difference in body
in body weight wasweight was found
found between between
animals animals
assigned to
the three dietary treatment groups on day 0 (data not shown). After 14 days of dietary
assigned to the three dietary treatment groups on day 0 (data not shown). After 14 days of dietary intervention,
no significant
intervention, nodifference in body
significant weightin
difference gain wasweight
body observed among
gain wasthe three dietary
observed among groups
the (Figure 1A).
three dietary
In addition, we evaluated total gross energy intake over the two-week experimental
groups (Figure 1A). In addition, we evaluated total gross energy intake over the two-week period and found
no significant
experimental difference
period in energy
and found intake between
no significant experiment
difference diets intake
in energy when compared to control diet;
between experiment diets
however, animals fed with high-fat/high-fructose diet exhibited an increase in energy intake
when compared to control diet; however, animals fed with high-fat/high-fructose diet exhibited an compared
to low-fat/high-fructose
increase in energy intake fed hamsters
compared to(Figure 1B).
low-fat/high-fructose fed hamsters (Figure 1B).

Figure
Figure1. 1.
High-fat/high-fructose
High-fat/high-fructosediet dietresults
results in
in increased adiposityand
increased adiposity andhepatomegaly
hepatomegalywithout
without weight
weight
gain after
gain 1414
after days. (A)(A)
days. BodyBodyweight
weightgain across
gain 2 weeks.
across (B) (B)
2 weeks. Total gross
Total energy
gross intake
energy over
intake thethe
over two-
week study study
two-week periodperiod
(kcal);(kcal);
# denotes no significant
# denotes difference
no significant when
difference compared
when comparedto control chow.
to control chow.(C)
(C) Epididymal
Epididymal white white
adiposeadipose
tissuetissue
(eWAT)(eWAT)
weight.weight. (D) Percent
(D) Percent liver weight
liver weight per body
per total total body
weight
measured at study day 14. Values are expressed as means, ±SEM.
measured at study day 14. Values are expressed as means, ±SEM. Significance denoted asdenoted
weight Significance * p < 0.05,
** as p < 0.05,
p <* 0.01, < 0.01,
*** p**<p0.001 vs*** p < 0.001
control Chowvs control chow using
using one-way one-way
ANOVA, ANOVA,
with with
Tukey’s Tukey’scomparison
multiple multiple
comparison test. FFLC denotes high-fat/high-fructose diet, HF indicates low-fat/high-fructose
test. FFLC denotes high-fat/high-fructose diet, HF indicates low-fat/high-fructose diet intervention, diet
intervention, ns indicated no significance.
ns indicated no significance.
3.2. High-Fat/High-Fructose Diet Causes Hepatomegaly and Increased Adipose Tissue
3.2. High-Fat/High-Fructose Diet Causes Hepatomegaly and Increased Adipose Tissue
To determine the effects of dietary fat and fructose on the accumulation of body fat and the
To determine
development the effects of
of hepatomegaly, wedietary
measured fatepididymal
and fructose onadipose
white the accumulation
tissue weightof body and
(eWAT) fat found
and the
development of hepatomegaly, we measured epididymal white adipose tissue
that hamsters fed a high-fat/high-fructose diet developed significantly higher amounts of eWATweight (eWAT) and
found that hamsters
compared to animalsfedfedathe control diet (p < 0.05),diet
high-fat/high-fructose developed
whereas significantly higher
the low-fat/high-fructose amounts
diet feeding of
eWAT
groupcompared to animals
did not affect fed the
eWAT (Figure control
1C). diet (p < 0.05), whereas
The high-fat/high-fructose thesignificantly
diet also low-fat/high-fructose
increased thediet
feeding
ratio ofgroup did not
liver weight affectbody
to total eWAT (Figure
weight 1C).
(Figure The
1D). By high-fat/high-fructose diet alsoalteration
contrast, there was no significant significantly
in
liver weight
increased the in the low-fat/high-fructose
ratio of liver weight to totaldiet fed animals
body weightcompared
(Figureto1D).
thoseByoncontrast,
the chow there
controlwas
dietno
(p > 0.05).alteration in liver weight in the low-fat/high-fructose diet fed animals compared to those
significant
on the chow control diet (p > 0.05).
 Nutrients
Nutrients 12,12,
2020,
2020, x 3557
FOR PEER REVIEW 6 of620
of 20

3.3. High-Fat/High-Fructose Diet Induces Dyslipidemia

3.3. High-Fat/High-Fructose Diet Induces Dyslipidemia
To evaluate the impact of diet on lipid metabolism, we measured fasting levels of triglycerides
To evaluate the impact of diet on lipid metabolism, we measured fasting levels of triglycerides
and cholesterol in plasma collected at baseline (day 0), day 7, and day 14 post-dietary treatments. The
and cholesterol in plasma collected at baseline (day 0), day 7, and day 14 post-dietary treatments.
high-fat/high-fructose diet resulted
The high-fat/high-fructose in a significant
diet resulted increase
in a significant of fasting
increase triglycerides
of fasting by day by
triglycerides 7 (pday
< 0.05),
7
which remained
(p < 0.05), whichelevated
remained atelevated
day 14 (p at <day
0.001)
14 (p(Figure
< 0.001)2A). We 2A).
(Figure alsoWemeasured the nonfasting
also measured level of
the nonfasting
triglycerides on day 15;
level of triglycerides onelevated triglycerides
day 15; elevated were found
triglycerides only inonly
were found theinhigh-fat/high-fructose
the high-fat/high-fructosegroup
(Figure
group2B). Similar
(Figure 2B). effects
Similarwere
effectsobserved for fasting
were observed levels of
for fasting cholesterol:
levels hamsters
of cholesterol: fed with
hamsters fed a high-
with
fat/high-fructose diet exhibited
a high-fat/high-fructose increased
diet exhibited levels levels
increased of cholesterol at day
of cholesterol 7 and
at day 14,14,
7 and compared
comparedtotothe
control diet fed
the control dietgroup
fed group = 0.01
(p =(p0.01 andand = 0.04,
p =p 0.04, respectively)
respectively)(Figure
(Figure 2C). By Bycontrast,
contrast,there
there were
were nono
significant
significantdifferences
differencesnoted
notedin inthe
the low-fat/high-fructose
low-fat/high-fructose fed fedanimals
animalsatateither
eithertime
timepoint
point > 0.05),
(p (p > 0.05),
when
when compared
comparedtotocontrol
controlchow
chow fedfed study group.
group. Additionally,
Additionally,the thelevel
levelofofnonfasting
nonfasting cholesterol
cholesterol
measured
measured onon day
day 1515 post
post dietmanipulation
diet manipulationwas waselevated
elevatedonlyonlyininhigh-fat/high-fructose
high-fat/high-fructose fedfed group of
of hamsters
hamsters (p < 0.01),
(p < 0.01), compared
compared to control
to control chowchow fed group
fed group (Figure
(Figure 2D). Thus,
2D). Thus, we found
we found that athathigh-
a high-fat/high-fructose diet induces dyslipidemia in hamsters, whereas a high-fructose/low-fat
fat/high-fructose diet induces dyslipidemia in hamsters, whereas a high-fructose/low-fat diet does diet
does not.
not.

Figure
Figure2.2.High-fat/high-fructose diet induces
High-fat/high-fructose diet inducesdyslipidemia.
dyslipidemia. (A)(A)
5 h5-h fasting
fasting plasmaplasma
TG onTGday (n =06),
on 0day (n =
6),day = 6),
day7 7(n(n = 6), and
and dayday 14=(n
14 (n 9);=(B)
9);nonfasting
(B) nonfasting
plasma plasma TG;
TG; (C) (C) Fasting
Fasting plasma
plasma total CHtotal CH day
on study on study
on
dayday (n = 6),
on0 day 0 (nday
= 6), = 6),
7 (nday 7 and
(n = 14 = 9);
6),(nand 14(D)
(n Nonfasting plasma total
= 9); (D) Nonfasting CH ontotal
plasma day 15 CH = 9).
(n on dayValues
15 (nare
= 9).
expressed
Values as means,as
are expressed ±SEM.
means, Significance * p < 0.05,
denoted asdenoted
±SEM. Significance as ***p p< <0.05,
0.01,***** < 0.001
p <p0.01, *** pvs< Chow,
0.001 vs
usingusing
Chow, Dunnett’s multiple
Dunnett’s comparison
multiple test. TG test.
comparison denotes
TG triglycerides, CH cholesterol,
denotes triglycerides, FFLC indicates
CH cholesterol, FFLC
a high-fat/high-fructose diet, and HF a, low-fat/high-fructose diet, ns denotes
indicates a high-fat/high-fructose diet, and HF a, low-fat/high-fructose diet, ns denotes no no significant difference.
significant
3.4.difference.
High-Fat/High-Fructose and Low-Fat/High-Fructose Diets Alter the Intestinal Microbiota
To understand the underlying
3.4. High-Fat/High-Fructose mechanisms by which
and Low-Fat/High-Fructose the diets
Diets Alter contribute
the Intestinal to the regulation of
Microbiota
lipid metabolism, we investigated the effects of dietary manipulations on the composition of the
gutTo understand
microbiota. As the underlying
measured by betamechanisms by was
diversity, there which the diets difference
a significant contributeintooverall
the regulation
microbial of
lipid metabolism,
composition we investigated
between the effects
dietary groups. of dietarymicrobial
This divergent manipulations on thewas
composition composition of theatgut
highly variable
microbiota. As measured by beta diversity, there was a significant difference in overall
each time point, as there was a significant interaction between dietary group and treatment day microbial
when
composition betweenrepeated
using two distinct dietary measure
groups. This
tools divergent microbial
for the analysis composition
of beta was highly
diversity, Bray < 0.05; at
variable
Curtis (p
each time point, as there was a significant interaction between dietary group and treatment day when
using two distinct repeated measure tools for the analysis of beta diversity, Bray Curtis (p < 0.05;
Figure 3A) and Binary Jaccard (p < 0.05; Figure 3B). Pairwise comparisons of dietary treatment groups
with control chow diet revealed that after seven days of dietary intervention, hamsters fed the high-
 No significant differences in alpha diversity between the dietary groups were found, as
measured by Shannon index between day 0 to day 14 (p > 0.05) (Figure 3D). A trend towards
decreased alpha diversity was observed on day 7 for high-fat/high-fructose diet fed hamsters,
compared to control diet (p = 0.06). However, there were no significant differences when comparing
within
Nutrientsdietary
2020, 12, treatment
3557 groups across sampling days (p > 0.05). 7 of 20
In addition to overall microbial composition, gut microbiota was evaluated at the phylum level.
We observed that the Firmicutes/Bacteroidetes ratio differed significantly between dietary groups
Figure
and 3A)days
across and ofBinary
intervention: (p < 0.05; Figure 3B).fed
Jaccard high-fat/high-fructose Pairwise
animalscomparisons of dietaryincrease
exhibited a significant treatment in
groups with control chow diet revealed that after seven days of dietary intervention,
Firmicutes between day 0 and day 7 (58% versus 87% respectively, p = 0.0042; Figure 3E). Between hamsters fed the
high-fat/high-fructose
day diet exhibited the greatestratio
7 and day 14, the Firmicutes/Bacteroidetes shift was
in the composition
decreased of gut microbiota (p <diet
in high-fat/high-fructose 0.0001).
fed
On day 14 post dietary manipulation, pairwise comparisons indicated that
hamsters, although the ratio remained elevated compared to standard chow and low-fat/high-the high-fat/high-fructose
diet groupdiet
fructose maintained
fed groupsthe microbiota
(Figure 3E). compositional
By contrast, change,
therewhereas
was nothesignificant
difference change
in dissimilarity
in the
distance between low-fat/high-fructose
Firmicutes/Bacteroidetes ratio for either and chow
the fed animals
control was reduced fructose
diet or low-fat/high (Figure 3C).
dietNo significant
groups from
differences
day 7 to daywere
14 (pfound within
> 0.05). the control
Evaluating grouprelevant
clinically across the three
phyla time points using
Proteobacteria, both Binary
we found Jaccard
no significant
and Unweighted
changes in any ofUnifrac metrics
the dietary (p > 0.05).
treatment groups (p > 0.05).

Figure 3.3. Feeding

Feedingof either a high-fat/high-fructose
of either or low-fat/high-fructose
a high-fat/high-fructose diet fordiet
or low-fat/high-fructose 14 days leads
for 14 to
days
shifts
leads in
to intestinal
shifts in microbial
intestinal composition and changesand
microbial composition in dominant
changes in bacterial
dominant Phyla. (A) Principle
bacterial Phyla.
coordinate
(A) analysis
Principle by Bray-Curtis
coordinate analysis similarities distances;
by Bray-Curtis (B) Principle
similarities coordinate
distances; analysiscoordinate
(B) Principle by Binary
Jaccard
analysis index,
by Binarysignificance assessed
Jaccard index, using repeated
significance measure
assessed using (PERMANOVA);
repeated (C) Pairwise
measure (PERMANOVA);
dissimilarity
(C) distance comparison
Pairwise dissimilarity between between
distance comparison dietary groups on dayon0,day
dietary groups day 7, and
0, day 7, andday
day1414 using
Unweighted Unifrac
Unifrac distances,
distances, one-way
one-way ANOVA
ANOVA followed
followed by by Tukey’s multiple comparison
comparison testing;
testing;
(D) Alpha
Alpha diversity
diversitymeasured
measuredbybyShannon
Shannon index, two-way
index, two-way ANOVA,
ANOVA, with totaltotal
with readread
depth usedused
depth as a
as a model
model offset,offset,
Tukey’sTukey’s multiple
multiple comparison
comparison testing; testing; and (E) Firmicutes/
and (E) Firmicutes/ BacteroidetesBacteroidetes ratio,
ratio, two-way
two-way ANOVA,
ANOVA, Tukey’scomparison
Tukey’s multiple multiple comparison
testing; ** ptesting; ** p <<0.001,
< 0.01, ***p 0.01, ns p < 0.001,
***denotes ns denotes
no significance,
no significance,
FFLC denotes a FFLC denotes a high-fat/high-fructose
high-fat/high-fructose diet, HF a low-fat/high-fructose
diet, HF a low-fat/high-fructose diet and ASV diet and
denotes
ASV denotes
amplicon amplicon
sequence sequence variant.
variant.

No significant differences in alpha diversity between the dietary groups were found, as measured
by Shannon index between day 0 to day 14 (p > 0.05) (Figure 3D). A trend towards decreased alpha
diversity was observed on day 7 for high-fat/high-fructose diet fed hamsters, compared to control diet
(p = 0.06). However, there were no significant differences when comparing within dietary treatment
groups across sampling days (p > 0.05).
In addition to overall microbial composition, gut microbiota was evaluated at the phylum level.
We observed that the Firmicutes/Bacteroidetes ratio differed significantly between dietary groups
and across days of intervention: high-fat/high-fructose fed animals exhibited a significant increase in
Firmicutes between day 0 and day 7 (58% versus 87% respectively, p = 0.0042; Figure 3E). Between day 7
and day 14, the Firmicutes/Bacteroidetes ratio was decreased in high-fat/high-fructose diet fed hamsters,
although the ratio remained elevated compared to standard chow and low-fat/high-fructose diet
Nutrients 2020, 12, x FOR PEER REVIEW 8 of 20

3.5. High-Fat/High-Fructose and Low-Fat/High-Fructose Diets Induce Changes in Bacterial Taxa of Gut
Nutrients 2020, 12, 3557 8 of 20
Microbiota
Negative binomial regression modeling identified amplicon sequence variants (ASVs) at the
fed
genus level(Figure
groups 3E).with
associated By contrast, there was
each individual no significant
dietary change
group and in the Firmicutes/Bacteroidetes
the temporal relationship across the
ratio for either the control diet or low-fat/high fructose diet groups from
two-week intervention period. As shown in Figure 4A, three ASVs were positivelyday 7 to day 14 (p > 0.05).
associated with
Evaluating clinically relevant phyla Proteobacteria, we found no significant changes in any
the high-fat/high-fructose diet: Alistipes, Lachnospiraspeae FCS020, and Coprococcus. Low-fat/high of the
dietary treatment groups (p > 0.05).
fructose fed hamsters exhibited a positive association with Coprococcus, a unique positive association
with Butyricimonas, and a negative association with Ruminoclostridum 5 and UBA 1819.
3.5. High-Fat/High-Fructose and Low-Fat/High-Fructose Diets Induce Changes in Bacterial Taxa of
A total of 25 taxa were identified with a significant interaction between dietary group
Gut Microbiota
assignment and time measured in days (Figure 4B). Of these 25 taxa, ten were differentially affected
Negative
between binomial regression
high-fat/high-fructose andmodeling identified amplicon
low-fat/high-fructose sequence (Figure
diet fed hamsters variants4C).
(ASVs) at thea
Notably,
genus level associated with each individual dietary group and the temporal relationship across
significant increase in Ruminococceace NK4A214 group, Intestinimonas, Roseburia and Ruminiclostridum the
two-week
9 was observed in high-fat/high-fructose diet fed hamsters, whereas low-fat/high-fructose diet the
intervention period. As shown in Figure 4A, three ASVs were positively associated with fed
high-fat/high-fructose diet: Alistipes, Lachnospiraspeae FCS020, and Coprococcus. Low-fat/high fructose
animals exhibited increases in Bacteroides, Butyricimonas, Parabacteroides and Prevotella. Interestingly,
fed hamsters
Prevotella exhibited
responded a positiveinassociation
divergently with Coprococcus,
the two dietary a unique
groups: decreasing positive
in the association with
high-fat/high-fructose
Butyricimonas, and a negative association with Ruminoclostridum 5 and UBA 1819.
diet fed animals while increasing in those receiving the low-fat/high-fructose diet.

Figure 4.4.Dietary
Dietaryinterventions
interventions overover 14 days
14 days result result in differential
in differential association
association oftaxa
of bacterial bacterial taxa
abundance
abundance
at the genusatlevel.
the genus
(A) level. (A) Negative
Negative binomialbinomial
regressionregression coefficients
coefficients for significant
for significant genusASVs
genus level level
ASVs associated
associated with dietary
with dietary interventions
interventions and changes
and changes acrossacross
time. time.
Dots Dots represent
represent coefficient
coefficient and
and bars
bars represent
represent standard
standard error error
of theofestimate;
the estimate; (B) Heat-map
(B) Heat-map of theofrelative
the relative abundance
abundance of bacterial
of bacterial taxa
taxa with
significant interactions
with significant between
interactions dietary
between groupgroup
dietary and time; (C) Relative
and time; abundance
(C) Relative of bacterial
abundance taxa with
of bacterial taxa
divergent responses
with divergent to either
responses study study
to either diet across two weeks
diet across of intervention.
two weeks High-fat/high-fructose
of intervention. diet is
High-fat/high-fructose
indicated by FFLC,
diet is indicated byHF denotes
FFLC, low-fat/high-fructose
HF denotes (HF) diet(HF)
low-fat/high-fructose intervention and ASV and
diet intervention denotes
ASVamplicon
denotes
sequence
amplicon variant.
sequence variant.
Nutrients 2020, 12, 3557 9 of 20

A total of 25 taxa were identified with a significant interaction between dietary group assignment
and time measured in days (Figure 4B). Of these 25 taxa, ten were differentially affected between
high-fat/high-fructose and low-fat/high-fructose diet fed hamsters (Figure 4C). Notably, a significant
increase in Ruminococceace NK4A214 group, Intestinimonas, Roseburia and Ruminiclostridum 9 was
observed in high-fat/high-fructose diet fed hamsters, whereas low-fat/high-fructose diet fed animals
exhibited increases in Bacteroides, Butyricimonas, Parabacteroides and Prevotella. Interestingly, Prevotella
responded divergently in the two dietary groups: decreasing in the high-fat/high-fructose diet fed
animals while increasing in those receiving the low-fat/high-fructose diet.

3.6. Dietary-Induced Shifts in Gut Microbiota Composition Correlate with Triglyceride and Cholesterol Levels
We observed that high-fat/high-fructose diet induced both dyslipidemia and alterations of gut
microbiota in hamsters, and further analyzed whether there was a correlation between these two
parameters. As shown in Figure 5, three ASVs that significantly correlated with fasting cholesterol
levels were unique to the high-fat/high-fructose diet fed hamsters: Ruminiclostridium 9 (ρ = 0.45)
Tyzzerella (ρ = 0.48) and Ruminiclostridium 6 (ρ = −0.53). No significant correlations between genus level
taxa and fasting levels of cholesterol were found in the low-fat/high-fructose diet fed group of animals.

Figure 5. Correlation heatmap representing Spearman correlation coefficients between changes in center
log ratio normalized abundance of bacterial taxa (at the genus level) and the lipid profile biomarkers
fasting CH and fasting TG within dietary treatment groups. Significance denoted as * padj < 0.05,
** padj < 0.01, *** padj < 0.001. CH denotes cholesterol, TG denotes triglyceride, and FFLC denotes
a high-fat/high-fructose diet, and HF a low-fat/high-fructose diet.

When analyzing fasting levels of triglyceride in plasma, we observed that the high-fat/high-fructose
diet fed study group had a total of eight significant correlations, with four positive correlations,
i.e., Tyzzerella (ρ = 0.73), Ruminococceace NK4A214 group (ρ = 0.73), GCA.900066575 (ρ = 0.58)
and Ruminiclostridium 9 (ρ = 0.44), and four negative correlations, i.e., Muribaculum (ρ = −0.71),
Nutrients 2020, 12, 3557 10 of 20

Prevotellaceae_UCG.003 (ρ = −0.64), Bifidobacterium (ρ = −0.55) and Ruminiclostridium 6 (ρ = −0.59).

Paradoxically, within the low-fat/high-fructose group, opposite correlations were observed in the
Ruminiclostridium genus: Ruminiclostridium 6 exhibited a negative association with plasma levels of
cholesterol and triglycerides, whereas Ruminiclostridium 9 presented a positive correlation with these
two lipid metabolism markers.
Two taxa were identified as having similar correlations with triglycerides in both dietary
intervention groups, with hamsters fed a low-fat/high-fructose diet exhibiting almost identical
correlation with Bifidobactrium (ρ = −0.56) and Tyzzerella (ρ = 0.51) compared to the high-fat/high-fructose
diet fed group. These findings indicate that the alterations in these taxa may relate to triglyceride
levels regardless of specific dietary manipulations.
The low-fat/high-fructose diet group exhibited four unique correlations with fasting levels
of triglyceride, including three positive correlations: Butyrimonas (ρ = 0.67), Bacteroides (ρ = 0.61),
Alistipes (ρ = 0.65), and one negative correlation with Ruminococcus 1 (ρ = −0.67).

3.7. Dietary Intervention Impacts Functional Changes in the Gut Microbiota

In addition to the microbial composition, an investigation was conducted to determine whether
low-fat/high-fructose or high-fat/high-fructose diets had any impact on the function of gut microbiota.
In the low-fat/high-fructose diet fed hamsters, 26 pathways were notably changed, which included
increases in fucose and rhamnose degradation, and biosynthesis of fatty acids (e.g., oleate and mycolate)
(Figure 6A). Interestingly, a shift toward increased energy production was observed, with increases
in the electron transport chain including 1,4-dihydroxy-6-napthoate biosynthesis, and menaquinol-8
biosynthesis II. An increase in the biosynthesis of the important cofactor biotin was also observed
in low-fat/high-fructose fed animals. Previous studies have reported that biotin affects lipid and
carbohydrate metabolism [40] and reduces both hyperglycemia and hypertriglyceridemia [41].
In the high-fat/high-fructose diet fed group, a total of 34 pathways were significantly altered
from day 0 to day 14 (Figure 6B). Notably, there was a shift towards catabolism of nucleotides,
as evidenced by a reduction in nucleotide biosynthesis accompanied by an increase in nicotinamide
adenine dinucleotide (NAD), thiamine and vitamin B12 salvage pathways. Fatty acid elongation,
lipid biosynthesis, and gluconeogenesis were also decreased. Both high-fructose diets resulted in
significant increases in urea cycle pathways from day 0 to day 14, likely resulting from an increase in
available energy capacity.
The comparison of gut microbiota metabolic pathways affected by the low-fat and high fat with
high-fructose diets on day 7 and day 14 revealed an increase in amino acid biosynthesis pathways,
including lysine, arginine, glutamate, glutamine and the branched-chain amino acid isoleucine
in the microbiome of high-fat/high-fructose diet fed animals. This predictive metagenomic
analysis also suggested that the low-fat/high-fructose diet group had an increase in fatty acid
elongation and gluconeogenesis, accompanied by an increase in vitamin B cofactor biosynthesis,
including tetrahydrofolate and thiamine (Supplemental Figure S1).
from day 0 to day 14 (Figure 6B). Notably, there was a shift towards catabolism of nucleotides, as
evidenced by a reduction in nucleotide biosynthesis accompanied by an increase in nicotinamide
adenine dinucleotide (NAD), thiamine and vitamin B12 salvage pathways. Fatty acid elongation,
lipid biosynthesis, and gluconeogenesis were also decreased. Both high-fructose diets resulted in
significant increases
Nutrients 2020, 12, 3557 in urea cycle pathways from day 0 to day 14, likely resulting from an increase
11 ofin
20
available energy capacity.

Figure 6. Predicted metagenomic changes within dietary intervention groups across sampling days.
(A) Metagenomic pathways significantly changed across treatment days for the low-fat/high-fructose
fed group of hamsters; and (B) Metagenomic pathways significantly changes across treatment
days in the high-fat/high-fructose fed group of animals. Kruskal Wallis test, with multiple
comparison corrections with FDR. Cytosine monophosphate, CMP; Tricarboxylic acid cycle, TCA;
nicotinamide adenine dinucleotide, NAD; Adenosine diphosphate, ADP.

3.8. High-Fat/High-Fructose Diet Leads to Functional Changes in the Gut Microbiome which Contribute to
Host Dyslipidemia
For hamsters fed a high-fat/high-fructose diet, a total of 22 pathways significantly correlated
with metabolic markers: 19 with fasting levels of triglycerides (15 negatively correlated and four
positively correlated) and three with fasting levels of cholesterol (Tables 2 and 3). Interestingly,
all four pathways that positively correlated with fasting levels of triglyceride in hamsters receiving
a high-fat diet were related to nucleotide degradation. Notably, a negative correlation was observed
between fasting levels of triglycerides and pyruvate fermentation for propionate in hamsters on
a high-fat/high-fructose diet. Propionate has previously been shown to regulate triglyceride levels
through expression of the peroxisome proliferator-activated α receptor [42]. Moreover, there was
a significant negative correlation between aspartate and asparagine biosynthesis pathways and levels
of both triglycerides and cholesterol in hamsters fed a high-fat/high-fructose diet.
Nutrients 2020, 12, 3557 12 of 20

Table 2. Significant correlations between bacterial metabolic pathway abundance and

fasting triglycerides.

Predicted Metabolic Pathways ρ Diet AdjP

Superpathway of l-alanine biosynthesis −0.59 FFLC 0.011
l-histidine degradation I −0.59 FFLC 0.012
Anhydromuropeptides recycling −0.56 FFLC 0.019
Superpathway of l-aspartate and l-asparagine biosynthesis −0.54 FFLC 0.024
Pyruvate fermentation to propanoate I −0.54 FFLC 0.025
UDP N-acetylmuramoyl pentapeptide biosynthesis II
−0.52 FFLC 0.032
lysine containing
Peptidoglycan biosynthesis III mycobacteria −0.52 FFLC 0.034
Peptidoglycan biosynthesis I meso
−0.51 FFLC 0.037
diaminopimelate containing
CMP 3 deoxy d-manno-octulosonate biosynthesis I −0.50 FFLC 0.042
Superpathway of geranyl-geranyl diphosphate biosynthesis II
−0.50 FFLC 0.044
via MEP.
Starch degradation V −0.50 FFLC 0.045
Aminoimidazole ribonucleotide biosynthesis I −0.50 FFLC 0.046
UDP N-acetylmuramoyl pentapeptide biosynthesis I meso
−0.49 FFLC 0.049
diaminopimelate containing
Thiamin salvage II −0.49 FFLC 0.049
Glycolysis II from fructose 6-phosphate −0.45 FFLC 0.049
Purine nucleotides degradation II aerobic 0.54 FFLC 0.027
Guanosine nucleotides degradation III 0.54 FFLC 0.025
Adenosine nucleotides degradation II 0.56 FFLC 0.019
Purine nucleobases degradation I anaerobic 0.56 FFLC 0.018
Anhydromuropeptides recycling −0.69 HF 0.001
TCA cycle VI obligate autotrophs −0.67 HF 0.003
TCA cycle I prokaryotic −0.66 HF 0.003
Reductive TCA cycle I −0.64 HF 0.005
tRNA processing −0.63 HF 0.005
Superpathway of fucose and rhamnose degradation 0.62 HF 0.007
Arginine ornithine and proline interconversion 0.59 HF 0.012
l-rhamnose degradation I 0.57 HF 0.017
TCA cycle V 2-oxoglutarate ferredoxin oxidoreductase −0.54 HF 0.024
Superpathway of l-alanine biosynthesis −0.54 HF 0.026
Fucose degradation 0.52 HF 0.035
l-lysine fermentation to acetate and butanoate 0.51 HF 0.040
ρ Spearman’s rho correlation.

Table 3. Significant correlations between bacterial metabolic pathway abundance and fasting cholesterol.

Predicted Metabolic Pathways ρ Diet AdjP

Glycolysis II from fructose 6-phosphate −0.44 FFLC 0.049
Superpathway of L-aspartate and L-asparagine biosynthesis −0.47 FFLC 0.034
Teichoic acid poly glycerol biosynthesis −0.49 FFLC 0.050
ρ Spearman’s rho correlation.

4. Discussion
This study uncovers the early impacts of dietary fat and carbohydrates on the gut microbiota
and metabolic profiles of golden Syrian hamsters. Feeding either high-fat/high-fructose or
low-fat/high-fructose diets resulted in shifts in overall composition of the gut microbiota,
with high-fat/high-fructose diet resulting in more profound changes in both the gut microbiota
and lipid metabolic profiles compared to a low-fat/high-fructose diet. A few previous studies have
also employed this model to determine how diet can impact the composition of gut microbiota [43–47].
However, most of these reports assessed the addition of a single dietary component over a longer
Nutrients 2020, 12, 3557 13 of 20

dietary intervention time to determine the resulting effects on host lipid metabolism [43–47],
thereby overlooking earlier metabolic changes.
The high-fat/high-fructose diet, which more closely resembles the so-called “Western” diet [48],
elevated plasma levels of triglycerides and cholesterol within just seven days. By contrast,
a low-fat/high-fructose diet did not result in a significant increase in host circulating lipid levels.
Previous work has described the lipogenic nature of fructose [49,50], and the synergistic effects of
fructose and high fat [14,51]. In the current study, we found that low-fat/high-fructose diet did not
induce dyslipidemia after two weeks, which is consistent with studies by Lozano et al, reporting that
the addition of fructose to a normal diet did not produce any metabolic changes in rats, whereas the
combination of high-fructose with high-fat resulted in metabolic changes greater than those observed
with a high-fat diet alone [14]. Notably, we did not see any significant difference in gained body weight
after the two-week intervention between the three diets. The lack of weight gain observed in both
fructose diets, also observed in several other studies [19,52,53], may be a result of the thermogenic effect
of fructose, previously observed in both humans [54,55] and rodents [56]. We also observed significant
increases in adiposity and liver weight only in high-fat/high-fructose fed hamsters. These findings
are in contrast with our previous work demonstrating an impact of fructose feedings on host lipid
profiles [50,57]. However, previous studies did not evaluate the role of the gut microbiota in contributing
to the host response to dietary interventions.
In this study, we found that the combined high-fat and high-fructose diet resulted in a dramatic
shift in the composition of gut microbiota. Specifically, seven days of dietary intervention resulted in
a significant shift in microbial composition, which was still evident at day 14, albeit to a lesser degree.
This may suggest a transient change in microbiota was induced by high-fat/high- fructose diet and
that the microbiota profile was under recovery after 14-days of feeding. Such a finding has been
observed in previous reports with various diets and oral antibiotics [58–60]. Interestingly, there were
no marked changes in overall diversity of commensal bacterial species, indicating that changes in the
composition of the gut microbiota resulted primarily from shifts in the abundance of bacterial species
that were previously presented at lower levels. The Firmicute/Bacteroidetes ratio, a metric commonly
associated with both obesity and metabolic dysfunction [61], was found to be altered only in the
high-fat/high-fructose diet. We also observed unique changes at the genus level that were significantly
associated with dietary interventions. For instance, high-fat/high-fructose diet increased the abundance
of Ruminococceace NK4A214 group, Intestinimonas, Roseburia and Ruminiclostridium 9, each of which
has previously been linked to obesity [62–64]. In addition, we found an increase in the abundance
of Parabacteroides, a genus with anti-inflammatory properties [65], in the low-fat/high-fructose diet
fed group. Recent work notes that Parabacteroides distasonis alleviates obesity in mice fed with a high-fat
diet via regulating the succinate and secondary bile acid metabolism [66]. The increased abundance of
Parabacteroides observed only in the low-fat/high-fructose diet group of hamsters may explain the lack
of demonstrable host lipid responses to dietary intervention.
Previous studies using germ-free mice showed that host lipid profiles can be transferred via
conventionalization of the intestinal microflora [67]. The phenomenon also has been observed in
humans following fecal microbial transplantation [68]. We observed a comparable relationship by
evaluating correlations between changes in microbial abundance at the genus level and fluctuations
in host lipid profiles. Several notable correlations were identified, including a positive correlation
between plasma levels of triglycerides and cholesterol with Tyzzerella, a Gram-negative member of
the Lachnospireacae family, in the gut. Recent work indicates that the abundance of Tyzerella is a risk
factor related to the development of cardiovascular diseases [69]. Correlations were also found in the
present study between Ruminiclostridium and host lipid levels, which is of interest given previous work
linking Ruminiclostridium with obesity [64] and gestational diabetes [70]. Notably, Ruminiclostridium 6
was previously found to have a strong positive correlation with ghrelin levels, indicating a potential
role in mediating the hunger response and resulting changes in energy intake [71].
Nutrients 2020, 12, 3557 14 of 20

Correlations between microbial abundance and metabolic markers reveal a relationship between
gut microbial taxa and host metabolism. However, this approach alone provides little information
regarding underlying mechanisms. By evaluating functional shifts related to changes in the
gut microbiota between various dietary interventions, we uncovered additional details about
potential mechanistic contributions to host dyslipidemia via the production of secondary metabolites,
although validation studies employing shotgun metagenomic sequencing and metabolomics
are warranted.
In the current study, the metabolic pathway of biotin biosynthesis was increased across treatment
days in hamsters fed a high-fructose/low-fat diet. A previous study demonstrated that the treatment
of biotin in mice protects against metabolic syndrome induced by a diet high in fructose [72].
The increase in biotin by gut microbes following the high-fructose feedings observed here may
serve as a compensatory response, accounting for the lack of increase in plasma lipids in hamsters
fed with a low-fat/high-fructose diet. Notably, we observed an increase in urea cycle for both
high-fructose/high-fat and high-fructose/low-fat diets. Previous findings have shown an effect of
fructose feeding on liver uric acid production [73], which is known to contribute to hepatic injury and
systemic inflammation [74]. Our findings indicate that the gut microbiome may also be a contributing
factor for increased circulating levels of uric acid following fructose feedings.
Low-fat/high-fructose diet fed hamsters also exhibited an increase in levels of the
butyrate-producing bacterium, Butryimonas [75]. Butyrate levels are linked to improved metabolite
outcomes both in humans and in animal studies, by acting as a regulator of adipocytes [76,77].
Butyrate supplementation has also been found to improve the dyslipidemia caused by a high-fat diet
in mice [78]. While not measured directly in the current study, an increase in levels of butyric acid in
the low-fat/high-fructose diet fed group is one potential mechanism that could account for the lack of
host dyslipidemia, as well as the lack of increase in epididymal adipose tissue observed herein.
By contrast, high-fat/high-fructose fed hamsters exhibited a decrease in fermentation pathways
that control production of the short-chain fatty acid, propionate. In humans, propionic acid stimulates
glucagon-like peptide-1 and pancreatic polypeptide [79], and long-term delivery of propionate
into the colon reduces weight gain and intra-abdominal fat accretion in overweight adults [79].
Changes in this pathway in high-fat/high-fructose diet fed hamsters correlated with fasting levels
of triglycerides. This relationship is likely a reflection of the effects of propionate on host lipid
metabolism. Propionate is known to directly affect lipid metabolism by decreasing the enzymatic
activities of 3-hydroxy-methlyglutaryl-CoA synthase and reductase [80], critical enzymes in cholesterol
synthesis [81] and ketogenesis [82]. Additionally, propionate regulates triglyceride production through
increasing expression of PPAR-α receptor, which subsequently leads to an increase in β-oxidation and
a reduction in triglycerides [42]. Other short-chain fatty acids also affect PPAR-γ receptor expression [83]
and β-oxidation in colonocytes. This relationship between microbial production of short-chain fatty
acids and host energy metabolism is critical in maintaining a healthy homeostatic relationship between
the host and constituents of the gut microbiota [84].
In addition to changes in the biosynthesis of short-chain fatty acids, there were negative correlations
between amino acid (aspartate and asparagine) biosynthesis and fasting metabolic parameters in
the high-fat/high-fructose diet group. Previous studies in humans indicate that plasma levels of
asparagine are associated with a lower risk of type-2 diabetes [85]. Asparagine levels are inversely
correlated with levels of plasma lipids. Moreover, feeding of exogenous L-aspartate has been shown
to limit the progression of fatty liver in cholesterol-fed rabbits [86]. The high-fat/high-fructose diet
also induced a significant increase in production of branched-chain amino acid isoleucine, which is
of potential importance since high levels of circulating isoleucine are associated with inactivity [87],
type 2 diabetes [88], and metabolic syndrome [89].
Taken together, these findings suggest that a two-week period of dietary intervention with
a high-fat/high-fructose diet not only induces changes in gut microbial composition, but also affects
microbial metabolic functions, which have the potential to produce metabolites contributing to host
Nutrients 2020, 12, 3557 15 of 20

dyslipidemia and development of the metabolic syndrome. An improved understanding of the

complex relationships between diet, the gut microbiota, and host metabolism provides clinicians and
researchers with the knowledge needed to improve outcomes in humans affected by metabolic diseases.
Herein, we identify several key taxa and microbial metabolic pathways that are influenced by fructose
and fat dietary supplementation. To further evaluate the specific relationship between dietary fats
and fructose, future studies should evaluate the impact of high fat, low fructose diets using the same
animal model. Here, we employed 16S rRNA sequencing and predictive metagenomics; while these
methods yield robust preliminary findings, further validation is required using shotgun metagenomic
and metabolomic evaluations.

Supplementary Materials: The following are available online at http://www.mdpi.com/2072-6643/12/11/3557/s1,

Supplementary Figure S1. Significant differences in predicted microbial metabolic pathway abundances between
dietary interventions with combined measurements taken at day 7 and day 14. Two-sided permutation t-test,
with multiple comparison correction with FDR Value expressed as absolute relative change; colors denote
significant increases for each dietary intervention study group. FFLC denotes a high-fat, high-fructose diet,
HF a low-fat/high-fructose diet, and FDR multiple correction for false discovery rate. Supplementary Table S1.
Dietary components of customized experimental diets: high-fat/high fructose (FFLC) and high-fructose/low
fat (HF).
Author Contributions: Y.Y. and K.A. designed the research; Y.Y., N.A., R.Z., conducted the experiments; L.R.
and M.S. performed the 16S rRNA gene sequencing. Data analysis and interpretations were done by R.G.H.;
R.G.H., Y.Y. and R.Z. wrote the manuscript. P.M.S. reviewed and edited the manuscript. All authors have read
and agreed to the published version of the manuscript.
Funding: This research was funded by operating grants from the Canadian Institutes of Health Research
(CIHR Foundation Grant 201509FDN-353989-FDN to KA and CIHR MOP-89894 and IOP-92890 to PMS).
Acknowledgments: PMS is the recipient of a Canada Research Chair in Gastrointestinal Disease.
Conflicts of Interest: P.M.S. serves on Advisory Boards for Nestle-Gerber, Cargill, and Antibe Therapeutics
and receives support for the research stipend of a post-doctoral fellow from Lallemand Human Nutrition
(Montreal, Quebec, Canada) through a Mitacs Accelerates Internship Award. PMS has received honoraria from
Nestle Nutrition and Abbott Nutrition for speaking at continuing medical education activities. K.A. has no
reported conflicts of interest.

References
1. Turnbaugh, P.J.; Hamady, M.; Yatsunenko, T.; Cantarel, B.L.; Duncan, A.; Ley, R.E.; Sogin, M.L.;
Jones, W.J.; Roe, B.A.; Affourtit, J.P.; et al. A core gut microbiome in obese and lean twins. Nature
2009, 457, 480–484. [CrossRef]
2. Hooper, L.V.; MacPherson, A.J. Immune adaptations that maintain homeostasis with the intestinal microbiota.
Nat. Rev. Immunol. 2010, 10, 159–169. [CrossRef]
3. Koppel, N.; Maini Rekdal, V.; Balskus, E.P. Chemical transformation of xenobiotics by the human
gut microbiota. Science 2017, 356, eaag2770. [CrossRef] [PubMed]
4. Durack, J.; Lynch, S.V. The gut microbiome: Relationships with disease and opportunities for therapy.
J. Exp. Med. 2019, 216, 20–40. [CrossRef] [PubMed]
5. McBurney, M.I.; Davis, C.; Fraser, C.M.; Schneeman, B.O.; Huttenhower, C.; Verbeke, K.; Walter, J.;
Latulippe, M.E. Establishing What Constitutes a Healthy Human Gut Microbiome: State of the Science,
Regulatory Considerations, and Future Directions. J. Nutr. 2019, 149, 1882–1895. [CrossRef] [PubMed]
6. Qin, J.; Li, Y.; Cai, Z.; Li, S.; Zhu, J.; Zhang, F.; Liang, S.; Zhang, W.; Guan, Y.; Shen, D.; et al. A metagenome-wide
association study of gut microbiota in type 2 diabetes. Nature 2012, 490, 55–60. [CrossRef] [PubMed]
7. Henao-Mejia, J.; Elinav, E.; Jin, C.; Hao, L.; Mehal, W.Z.; Strowig, T.; Thaiss, C.A.; Kau, A.L.; Eisenbarth, S.C.;
Jurczak, M.J.; et al. Inflammasome-mediated dysbiosis regulates progression of NAFLD and obesity. Nature
2012, 482, 179–185. [CrossRef] [PubMed]
8. Ley, R.E.; Backhed, F.; Turnbaugh, P.; Lozupone, C.A.; Knight, R.D.; Gordon, J.I. Obesity alters gut
microbial ecology. Proc. Natl. Acad. Sci. USA 2005, 102, 11070–11075. [CrossRef]
9. Ley, R.E.; Turnbaugh, P.J.; Klein, S.; Gordon, J.I. Microbial ecology: Human gut microbes associated
with obesity. Nature 2006, 444, 1022–1023. [CrossRef]
10. Lee, L.; Sanders, R.A. Metabolic syndrome. Pediatr. Rev. 2012, 33, 459–466, quiz 467–458. [CrossRef]
Nutrients 2020, 12, 3557 16 of 20

11. Kenny, D.J.; Plichta, D.R.; Shungin, D.; Koppel, N.; Hall, A.B.; Fu, B.; Vasan, R.S.; Shaw, S.Y.; Vlamakis, H.;
Balskus, E.P.; et al. Cholesterol Metabolism by Uncultured Human Gut Bacteria Influences Host
Cholesterol Level. Cell Host Microbe 2020, 28, 245–257.e6. [CrossRef] [PubMed]
12. Cani, P.D.; Bibiloni, R.; Knauf, C.; Waget, A.; Neyrinck, A.M.; Delzenne, N.M.; Burcelin, R. Changes in
gut microbiota control metabolic endotoxemia-induced inflammation in high-fat diet-induced obesity and
diabetes in mice. Diabetes 2008, 57, 1470–1481. [CrossRef] [PubMed]
13. Ijaz, M.U.; Ahmed, M.I.; Zou, X.; Hussain, M.; Zhang, M.; Zhao, F.; Xu, X.; Zhou, G.; Li, C. Beef, casein, and
soy proteins differentially affect lipid metabolism, triglycerides accumulation and gut microbiota of high-fat
diet-fed C57BL/6J mice. Front. Microbiol. 2018, 9, 2200. [CrossRef] [PubMed]
14. Lozano, I.; Van der Werf, R.; Bietiger, W.; Seyfritz, E.; Peronet, C.; Pinget, M.; Jeandidier, N.; Maillard, E.;
Marchioni, E.; Sigrist, S.; et al. High-fructose and high-fat diet-induced disorders in rats: Impact on
diabetes risk, hepatic and vascular complications. Nutr. Metab. 2016, 13, 15. [CrossRef]
15. Guerville, M.; Leroy, A.; Sinquin, A.; Laugerette, F.; Michalski, M.C.; Boudry, G. Western-diet consumption
induces alteration of barrier function mechanisms in the ileum that correlates with metabolic endotoxemia
in rats. Am. J. Physiol. Endocrinol. Metab. 2017, 313, E107–E120. [CrossRef] [PubMed]
16. Guevara-Cruz, M.; Flores-Lopez, A.G.; Aguilar-Lopez, M.; Sanchez-Tapia, M.; Medina-Vera, I.; Diaz, D.;
Tovar, A.R.; Torres, N. Improvement of Lipoprotein Profile and Metabolic Endotoxemia by a Lifestyle
Intervention That Modifies the Gut Microbiota in Subjects With Metabolic Syndrome. J. Am. Heart Assoc.
2019, 8, e012401. [CrossRef]
17. Roager, H.M.; Dragsted, L.O. Diet-derived microbial metabolites in health and disease. Nutr. Bull. 2019,
44, 216–227. [CrossRef]
18. Jegatheesan, P.; De Bandt, J.P. Fructose and NAFLD: The Multifaceted Aspects of Fructose Metabolism.
Nutrients 2017, 9, 230. [CrossRef]
19. Jegatheesan, P.; Beutheu, S.; Ventura, G.; Sarfati, G.; Nubret, E.; Kapel, N.; Waligora-Dupriet, A.J.; Bergheim, I.;
Cynober, L.; De-Bandt, J.P. Effect of specific amino acids on hepatic lipid metabolism in fructose-induced
non-alcoholic fatty liver disease. Clin. Nutr. 2016, 35, 175–182. [CrossRef]
20. Oh, J.H.; Alexander, L.M.; Pan, M.; Schueler, K.L.; Keller, M.P.; Attie, A.D.; Walter, J.; van Pijkeren, J.P. Dietary
Fructose and Microbiota-Derived Short-Chain Fatty Acids Promote Bacteriophage Production in the Gut
Symbiont Lactobacillus reuteri. Cell Host Microbe 2019, 25, 273–284.e6. [CrossRef]
21. Jakobsdottir, G.; Xu, J.; Molin, G.; Ahrne, S.; Nyman, M. High-fat diet reduces the formation of butyrate, but
increases succinate, inflammation, liver fat and cholesterol in rats, while dietary fibre counteracts these effects.
PLoS ONE 2013, 8, e80476. [CrossRef] [PubMed]
22. Hoyles, L.; Fernandez-Real, J.M.; Federici, M.; Serino, M.; Abbott, J.; Charpentier, J.; Heymes, C.; Luque, J.L.;
Anthony, E.; Barton, R.H.; et al. Molecular phenomics and metagenomics of hepatic steatosis in non-diabetic
obese women. Nat. Med. 2018, 24, 1070–1080. [CrossRef]
23. Ma, N.; Tian, Y.; Wu, Y.; Ma, X. Contributions of the Interaction Between Dietary Protein and Gut Microbiota
to Intestinal Health. Curr. Protein. Pept. Sci. 2017, 18, 795–808. [CrossRef] [PubMed]
24. Do, T.T.; Hindlet, P.; Waligora-Dupriet, A.J.; Kapel, N.; Neveux, N.; Mignon, V.; Delomenie, C.; Farinotti, R.;
Feve, B.; Buyse, M. Disturbed intestinal nitrogen homeostasis in a mouse model of high-fat diet-induced
obesity and glucose intolerance. Am. J. Physiol. Endocrinol. Metab. 2014, 306, E668–E680. [CrossRef] [PubMed]
25. Xiao, F.; Du, Y.; Lv, Z.; Chen, S.; Zhu, J.; Sheng, H.; Guo, F. Effects of essential amino acids on lipid metabolism
in mice and humans. J. Mol. Endocrinol. 2016, 57, 223–231. [CrossRef]
26. Hsieh, J.; Trajcevski, K.E.; Farr, S.L.; Baker, C.L.; Lake, E.J.; Taher, J.; Iqbal, J.; Hussain, M.M.; Adeli, K.
Glucagon-Like Peptide 2 (GLP-2) Stimulates Postprandial Chylomicron Production and Postabsorptive
Release of Intestinal Triglyceride Storage Pools via Induction of Nitric Oxide Signaling in Male Hamsters
and Mice. Endocrinology 2015, 156, 3538–3547. [CrossRef]
27. Dalboge, L.S.; Pedersen, P.J.; Hansen, G.; Fabricius, K.; Hansen, H.B.; Jelsing, J.; Vrang, N. A Hamster Model of
Diet-Induced Obesity for Preclinical Evaluation of Anti-Obesity, Anti-Diabetic and Lipid Modulating Agents.
PLoS ONE 2015, 10, e0135634. [CrossRef]
28. Wang, P.R.; Guo, Q.; Ippolito, M.; Wu, M.; Milot, D.; Ventre, J.; Doebber, T.; Wright, S.D.; Chao, Y.S. High fat
fed hamster, a unique animal model for treatment of diabetic dyslipidemia with peroxisome proliferator
activated receptor alpha selective agonists. Eur. J. Pharm. 2001, 427, 285–293. [CrossRef]
Nutrients 2020, 12, 3557 17 of 20

29. Bravo, E.; Cantafora, A.; Calcabrini, A.; Ortu, G. Why prefer the golden Syrian hamster (Mesocricetus auratus)
to the Wistar rat in experimental studies on plasma lipoprotein metabolism? Comp. Biochem. Physiol. Part B
Comp. Biochem. 1994, 107, 347–355. [CrossRef]
30. Gaynor, B.J.; Sand, T.; Clark, R.W.; Aiello, R.J.; Bamberger, M.J.; Moberly, J.B. Inhibition of cholesteryl
ester transfer protein activity in hamsters alters HDL lipid composition. Atherosclerosis 1994,
110, 101–109. [CrossRef]
31. Nistor, A.; Bulla, A.; Filip, D.A.; Radu, A. The hyperlipidemic hamster as a model of experimental atherosclerosis.
Atherosclerosis 1987, 68, 159–173. [CrossRef]
32. Shan, K.; Qu, H.; Zhou, K.; Wang, L.; Zhu, C.; Chen, H.; Gu, Z.; Cui, J.; Fu, G.; Li, J.; et al. Distinct Gut
Microbiota Induced by Different Fat-to-Sugar-Ratio High-Energy Diets Share Similar Pro-obesity Genetic
and Metabolite Profiles in Prediabetic Mice. mSystems 2019, 4. [CrossRef] [PubMed]
33. Do, M.H.; Lee, E.; Oh, M.J.; Kim, Y.; Park, H.Y. High-Glucose or -Fructose Diet Cause Changes of the
Gut Microbiota and Metabolic Disorders in Mice without Body Weight Change. Nutrients 2018, 10, 761.
[CrossRef] [PubMed]
34. Whelan, F.J.; Verschoor, C.P.; Stearns, J.C.; Rossi, L.; Luinstra, K.; Loeb, M.; Smieja, M.; Johnstone, J.;
Surette, M.G.; Bowdish, D.M. The loss of topography in the microbial communities of the upper respiratory
tract in the elderly. Ann. Am. Thorac. Soc. 2014, 11, 513–521. [CrossRef]
35. Horne, R.; St Pierre, J.; Odeh, S.; Surette, M.; Foster, J.A. Microbe and host interaction in
gastrointestinal homeostasis. Psychopharmacology 2019, 236, 1623–1640. [CrossRef] [PubMed]
36. Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet J. 2011,
17, 10. [CrossRef]
37. Quast, C.; Pruesse, E.; Yilmaz, P.; Gerken, J.; Schweer, T.; Yarza, P.; Peplies, J.; Glockner, F.O. The SILVA
ribosomal RNA gene database project: Improved data processing and web-based tools. Nucleic Acids Res.
2013, 41, D590–D596. [CrossRef]
38. Douglas, G.M.; Maffei, V.J.; Zaneveld, J.R.; Yurgel, S.N.; Brown, J.R.; Taylor, C.M.; Huttenhower, C.;
Langille, M.G.I. PICRUSt2 for prediction of metagenome functions. Nat. Biotechnol. 2020, 38, 685–688. [CrossRef]
39. Zhang, X.; Pei, Y.F.; Zhang, L.; Guo, B.; Pendegraft, A.H.; Zhuang, W.; Yi, N. Negative Binomial Mixed
Models for Analyzing Longitudinal Microbiome Data. Front. Microbiol. 2018, 9, 1683. [CrossRef]
40. Fernandez-Mejia, C.; Lazo-de-la-Vega-Monroy, M.-L. Biological Effects of Pharmacological Concentrations
of Biotin. J. Evid. Based Integr. Med. 2011, 16, 40–48. [CrossRef]
41. Revilla-Monsalve, C.; Zendejas-Ruiz, I.; Islas-Andrade, S.; Baez-Saldana, A.; Palomino-Garibay, M.A.;
Hernandez-Quiroz, P.M.; Fernandez-Mejia, C. Biotin supplementation reduces plasma triacylglycerol and
VLDL in type 2 diabetic patients and in nondiabetic subjects with hypertriglyceridemia. Biomed. Pharm.
2006, 60, 182–185. [CrossRef] [PubMed]
42. Higashimura, Y.; Naito, Y.; Takagi, T.; Uchiyama, K.; Mizushima, K.; Yoshikawa, T. Propionate Promotes
Fatty Acid Oxidation through the Up-Regulation of Peroxisome Proliferator-Activated Receptor alpha in
Intestinal Epithelial Cells. J. Nutr. Sci. Vitam. 2015, 61, 511–515. [CrossRef] [PubMed]
43. Carr, T.P.; Weller, C.L.; Schlegel, V.L.; Cuppett, S.L.; Guderian, D.M., Jr.; Johnson, K.R. Grain sorghum lipid
extract reduces cholesterol absorption and plasma non-HDL cholesterol concentration in hamsters. J. Nutr.
2005, 135, 2236–2240. [CrossRef] [PubMed]
44. Martinez, I.; Wallace, G.; Zhang, C.; Legge, R.; Benson, A.K.; Carr, T.P.; Moriyama, E.N.; Walter, J. Diet-induced
metabolic improvements in a hamster model of hypercholesterolemia are strongly linked to alterations of
the gut microbiota. Appl. Env. Microbiol. 2009, 75, 4175–4184. [CrossRef] [PubMed]
45. Kim, H.; Kim, D.H.; Seo, K.H.; Chon, J.W.; Nah, S.Y.; Bartley, G.E.; Arvik, T.; Lipson, R.; Yokoyama, W.
Modulation of the intestinal microbiota is associated with lower plasma cholesterol and weight gain in
hamsters fed chardonnay grape seed flour. J. Agric. Food Chem. 2015, 63, 1460–1467. [CrossRef]
46. Pongking, T.; Haonon, O.; Dangtakot, R.; Onsurathum, S.; Jusakul, A.; Intuyod, K.; Sangka, A.;
Anutrakulchai, S.; Cha’on, U.; Pinlaor, S.; et al. A combination of monosodium glutamate and high-fat
and high-fructose diets increases the risk of kidney injury, gut dysbiosis and host-microbial co-metabolism.
PLoS ONE 2020, 15, e0231237. [CrossRef]
47. Wang, Y.; Tong, Q.; Shou, J.W.; Zhao, Z.X.; Li, X.Y.; Zhang, X.F.; Ma, S.R.; He, C.Y.; Lin, Y.; Wen, B.Y.; et al.
Gut Microbiota-Mediated Personalized Treatment of Hyperlipidemia Using Berberine. Theranostics 2017,
7, 2443–2451. [CrossRef]
Nutrients 2020, 12, 3557 18 of 20

48. Zinocker, M.K.; Lindseth, I.A. The Western Diet-Microbiome-Host Interaction and Its Role in
Metabolic Disease. Nutrients 2018, 10, 365. [CrossRef]
49. Le, K.A.; Ith, M.; Kreis, R.; Faeh, D.; Bortolotti, M.; Tran, C.; Boesch, C.; Tappy, L. Fructose overconsumption
causes dyslipidemia and ectopic lipid deposition in healthy subjects with and without a family history of
type 2 diabetes. Am. J. Clin. Nutr. 2009, 89, 1760–1765. [CrossRef]
50. Stanhope, K.L.; Schwarz, J.M.; Keim, N.L.; Griffen, S.C.; Bremer, A.A.; Graham, J.L.; Hatcher, B.; Cox, C.L.;
Dyachenko, A.; Zhang, W.; et al. Consuming fructose-sweetened, not glucose-sweetened, beverages increases
visceral adiposity and lipids and decreases insulin sensitivity in overweight/obese humans. J. Clin. Investig.
2009, 119, 1322–1334. [CrossRef]
51. Softic, S.; Gupta, M.K.; Wang, G.X.; Fujisaka, S.; O’Neill, B.T.; Rao, T.N.; Willoughby, J.; Harbison, C.;
Fitzgerald, K.; Ilkayeva, O.; et al. Divergent effects of glucose and fructose on hepatic lipogenesis and insulin
signaling. J. Clin. Investig. 2018, 128, 1199. [CrossRef] [PubMed]
52. Tillman, E.J.; Morgan, D.A.; Rahmouni, K.; Swoap, S.J. Three months of high-fructose feeding fails to induce
excessive weight gain or leptin resistance in mice. PLoS ONE 2014, 9, e107206. [CrossRef] [PubMed]
53. Messier, C.; Whately, K.; Liang, J.; Du, L.; Puissant, D. The effects of a high-fat, high-fructose, and combination
diet on learning, weight, and glucose regulation in C57BL/6 mice. Behav. Brain Res. 2007, 178, 139–145.
[CrossRef] [PubMed]
54. Simonson, D.C.; Tappy, L.; Jequier, E.; Felber, J.P.; DeFronzo, R.A. Normalization of carbohydrate-induced
thermogenesis by fructose in insulin-resistant states. Am. J. Physiol. 1988, 254, E201–E207.
[CrossRef] [PubMed]
55. Mizobe, T.; Nakajima, Y.; Ueno, H.; Sessler, D.I. Fructose administration increases intraoperative core
temperature by augmenting both metabolic rate and the vasoconstriction threshold. Anesthesiology 2006,
104, 1124–1130. [CrossRef] [PubMed]
56. DeBosch, B.J.; Chen, Z.; Finck, B.N.; Chi, M.; Moley, K.H. Glucose transporter-8 (GLUT8) mediates
glucose intolerance and dyslipidemia in high-fructose diet-fed male mice. Mol. Endocrinol. 2013,
27, 1887–1896. [CrossRef]
57. Taghibiglou, C.; Carpentier, A.; Van Iderstine, S.C.; Chen, B.; Rudy, D.; Aiton, A.; Lewis, G.F.; Adeli, K.
Mechanisms of hepatic very low density lipoprotein overproduction in insulin resistance. Evidence
for enhanced lipoprotein assembly, reduced intracellular ApoB degradation, and increased microsomal
triglyceride transfer protein in a fructose-fed hamster model. J. Biol Chem. 2000, 275, 8416–8425. [CrossRef]
58. Leeming, E.R.; Johnson, A.J.; Spector, T.D.; Le Roy, C.I. Effect of Diet on the Gut Microbiota: Rethinking
Intervention Duration. Nutrients 2019, 11, 2862. [CrossRef]
59. Voreades, N.; Kozil, A.; Weir, T.L. Diet and the development of the human intestinal microbiome.
Front. Microbiol. 2014, 5, 494. [CrossRef]
60. Wei, S.; Mortensen, M.S.; Stokholm, J.; Brejnrod, A.D.; Thorsen, J.; Rasmussen, M.A.; Trivedi, U.; Bisgaard, H.;
Sorensen, S.J. Short- and long-term impacts of azithromycin treatment on the gut microbiota in children:
A double-blind, randomized, placebo-controlled trial. EBioMedicine 2018, 38, 265–272. [CrossRef]
61. Magne, F.; Gotteland, M.; Gauthier, L.; Zazueta, A.; Pesoa, S.; Navarrete, P.; Balamurugan, R. The
Firmicutes/Bacteroidetes Ratio: A Relevant Marker of Gut Dysbiosis in Obese Patients? Nutrients 2020,
12, 1474. [CrossRef] [PubMed]
62. Murugesan, S.; Ulloa-Martinez, M.; Martinez-Rojano, H.; Galvan-Rodriguez, F.M.; Miranda-Brito, C.;
Romano, M.C.; Pina-Escobedo, A.; Pizano-Zarate, M.L.; Hoyo-Vadillo, C.; Garcia-Mena, J. Study of the
diversity and short-chain fatty acids production by the bacterial community in overweight and obese
Mexican children. Eur. J. Clin. Microbiol. Infect. Dis. 2015, 34, 1337–1346. [CrossRef] [PubMed]
63. Lin, H.; An, Y.; Tang, H.; Wang, Y. Alterations of Bile Acids and Gut Microbiota in Obesity Induced by High
Fat Diet in Rat Model. J. Agric. Food Chem. 2019, 67, 3624–3632. [CrossRef] [PubMed]
64. Zhao, L.; Chen, Y.; Xia, F.; Abudukerimu, B.; Zhang, W.; Guo, Y.; Wang, N.; Lu, Y. A Glucagon-Like Peptide-1
Receptor Agonist Lowers Weight by Modulating the Structure of Gut Microbiota. Front. Endocrinol. 2018,
9, 233. [CrossRef] [PubMed]
65. Hiippala, K.; Kainulainen, V.; Suutarinen, M.; Heini, T.; Bowers, J.R.; Jasso-Selles, D.; Lemmer, D.; Valentine, M.;
Barnes, R.; Engelthaler, D.M.; et al. Isolation of Anti-Inflammatory and Epithelium Reinforcing Bacteroides
and Parabacteroides Spp. from A Healthy Fecal Donor. Nutrients 2020, 12, 935. [CrossRef] [PubMed]
Nutrients 2020, 12, 3557 19 of 20

66. Wang, K.; Liao, M.; Zhou, N.; Bao, L.; Ma, K.; Zheng, Z.; Wang, Y.; Liu, C.; Wang, W.; Wang, J.; et al.
Parabacteroides distasonis Alleviates Obesity and Metabolic Dysfunctions via Production of Succinate and
Secondary Bile Acids. Cell Rep. 2019, 26, 222–235.e5. [CrossRef]
67. Ridaura, V.K.; Faith, J.J.; Rey, F.E.; Cheng, J.; Duncan, A.E.; Kau, A.L.; Griffin, N.W.; Lombard, V.; Henrissat, B.;
Bain, J.R.; et al. Gut microbiota from twins discordant for obesity modulate metabolism in mice. Science 2013,
341, 1241214. [CrossRef]
68. Alang, N.; Kelly, C.R. Weight gain after fecal microbiota transplantation. Open Forum Infect. Dis 2015,
2, ofv004. [CrossRef]
69. Kelly, T.N.; Bazzano, L.A.; Ajami, N.J.; He, H.; Zhao, J.; Petrosino, J.F.; Correa, A.; He, J. Gut Microbiome
Associates with Lifetime Cardiovascular Disease Risk Profile among Bogalusa Heart Study Participants.
Circ. Res. 2016, 119, 956–964. [CrossRef]
70. Kuang, Y.S.; Lu, J.H.; Li, S.H.; Li, J.H.; Yuan, M.Y.; He, J.R.; Chen, N.N.; Xiao, W.Q.; Shen, S.Y.; Qiu, L.; et al.
Connections between the human gut microbiome and gestational diabetes mellitus. Gigascience 2017,
6, 1–12. [CrossRef]
71. Rossell, J.; Brindefalk, B.; Baena-Fustegueras, J.A.; Peinado-Onsurbe, J.; Udekwu, K.I. Diet change affects
intestinal microbiota restoration and improves vertical sleeve gastrectomy outcome in diet-induced obese rats.
Eur. J. Nutr. 2020, 59, 3555–3564. [CrossRef] [PubMed]
72. Aguilera-Mendez, A.; Hernandez-Equihua, M.G.; Rueda-Rocha, A.C.; Guajardo-Lopez, C.; Nieto-Aguilar, R.;
Serrato-Ochoa, D.; Ruiz Herrera, L.F.; Guzman-Nateras, J.A. Protective effect of supplementation with biotin
against high-fructose-induced metabolic syndrome in rats. Nutr. Res. 2018, 57, 86–96. [CrossRef] [PubMed]
73. Nakagawa, T.; Hu, H.; Zharikov, S.; Tuttle, K.R.; Short, R.A.; Glushakova, O.; Ouyang, X.; Feig, D.I.;
Block, E.R.; Herrera-Acosta, J.; et al. A causal role for uric acid in fructose-induced metabolic syndrome.
Am. J. Physiol. Ren. Physiol. 2006, 290, F625–F631. [CrossRef] [PubMed]
74. Lanaspa, M.A.; Sanchez-Lozada, L.G.; Choi, Y.J.; Cicerchi, C.; Kanbay, M.; Roncal-Jimenez, C.A.; Ishimoto, T.;
Li, N.; Marek, G.; Duranay, M.; et al. Uric acid induces hepatic steatosis by generation of mitochondrial
oxidative stress: Potential role in fructose-dependent and -independent fatty liver. J. Biol. Chem. 2012,
287, 40732–40744. [CrossRef] [PubMed]
75. Cherbuy, C.; Bellet, D.; Robert, V.; Mayeur, C.; Schwiertz, A.; Langella, P. Modulation of the Caecal Gut
Microbiota of Mice by Dietary Supplement Containing Resistant Starch: Impact Is Donor-Dependent.
Front. Microbiol. 2019, 10, 1234. [CrossRef] [PubMed]
76. Pelgrim, C.E.; Franx, B.A.A.; Snabel, J.; Kleemann, R.; Arnoldussen, I.A.C.; Kiliaan, A.J. Butyrate Reduces
HFD-Induced Adipocyte Hypertrophy and Metabolic Risk Factors in Obese LDLr-/-.Leiden Mice. Nutrients
2017, 9, 714. [CrossRef]
77. Yan, H.; Ajuwon, K.M. Mechanism of Butyrate Stimulation of Triglyceride Storage and Adipokine Expression
during Adipogenic Differentiation of Porcine Stromovascular Cells. PLoS ONE 2015, 10, e0145940. [CrossRef]
78. Yu, C.; Liu, S.; Chen, L.; Shen, J.; Niu, Y.; Wang, T.; Zhang, W.; Fu, L. Effect of exercise and
butyrate supplementation on microbiota composition and lipid metabolism. J. Endocrinol. 2019,
243, 125–135. [CrossRef]
79. Chambers, E.S.; Viardot, A.; Psichas, A.; Morrison, D.J.; Murphy, K.G.; Zac-Varghese, S.E.; MacDougall, K.;
Preston, T.; Tedford, C.; Finlayson, G.S.; et al. Effects of targeted delivery of propionate to the human
colon on appetite regulation, body weight maintenance and adiposity in overweight adults. Gut 2015,
64, 1744–1754. [CrossRef]
80. Preiss, B. Regulation of HMG-CoA Reductase in Extrahepatic Tissues. Regul. Hmg-Coa Reductase 1985, 183–200.
81. Greenspan, M.D.; Yudkovitz, J.B.; Lo, C.Y.; Chen, J.S.; Alberts, A.W.; Hunt, V.M.; Chang, M.N.; Yang, S.S.;
Thompson, K.L.; Chiang, Y.C.; et al. Inhibition of hydroxymethylglutaryl-coenzyme A synthase by L-659,699.
Proc. Natl. Acad. Sci. USA 1987, 84, 7488–7492. [CrossRef] [PubMed]
82. Hegardt, F.G. Mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase: A control enzyme in ketogenesis.
Biochem. J. 1999, 338, 569–582. [CrossRef] [PubMed]
83. Alex, S.; Lange, K.; Amolo, T.; Grinstead, J.S.; Haakonsson, A.K.; Szalowska, E.; Koppen, A.; Mudde, K.;
Haenen, D.; Al-Lahham, S.; et al. Short-chain fatty acids stimulate angiopoietin-like 4 synthesis in human
colon adenocarcinoma cells by activating peroxisome proliferator-activated receptor gamma. Mol. Cell Biol.
2013, 33, 1303–1316. [CrossRef] [PubMed]
84. Byndloss, M.X. Microbial management. Science 2020, 369, 153. [CrossRef]
Nutrients 2020, 12, 3557 20 of 20

85. Ottosson, F.; Smith, E.; Melander, O.; Fernandez, C. Altered Asparagine and Glutamate Homeostasis Precede
Coronary Artery Disease and Type 2 Diabetes. J. Clin. Endocrinol. Metab. 2018, 103, 3060–3069. [CrossRef]
86. Yanni, A.E.; Agrogiannis, G.; Nomikos, T.; Fragopoulou, E.; Pantopoulou, A.; Antonopoulou, S.; Perrea, D.
Oral supplementation with L-aspartate and L-glutamate inhibits atherogenesis and fatty liver disease in
cholesterol-fed rabbit. Amino Acids 2010, 38, 1323–1331. [CrossRef]
87. Kujala, U.M.; Makinen, V.P.; Heinonen, I.; Soininen, P.; Kangas, A.J.; Leskinen, T.H.; Rahkila, P.; Wurtz, P.;
Kovanen, V.; Cheng, S.; et al. Long-term leisure-time physical activity and serum metabolome. Circulation
2013, 127, 340–348. [CrossRef]
88. Wang, T.J.; Larson, M.G.; Vasan, R.S.; Cheng, S.; Rhee, E.P.; McCabe, E.; Lewis, G.D.; Fox, C.S.; Jacques, P.F.;
Fernandez, C.; et al. Metabolite profiles and the risk of developing diabetes. Nat. Med. 2011,
17, 448–453. [CrossRef]
89. Kujala, U.M.; Peltonen, M.; Laine, M.K.; Kaprio, J.; Heinonen, O.J.; Sundvall, J.; Eriksson, J.G.; Jula, A.;
Sarna, S.; Kainulainen, H. Branched-Chain Amino Acid Levels Are Related with Surrogates of Disturbed
Lipid Metabolism among Older Men. Front. Med. 2016, 3, 57. [CrossRef]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional
affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).