Key words
11 beta-hydroxysteroid dehydrogenase - steroid 5 alpha-reductase 1 - corticosterone - metabolic disorders - sugar
HFCS high-fructose corn syrup
GC glucocorticoid
Hsd11b1, 2 11 beta-hydroxysteroid dehydrogenase1, 2
Srd5a1 steroid 5 alpha-reductase 1
PD postnatal day
Introduction
Fructose is a monosaccharide found in fruits and vegetables and is often used in
foods and beverages as high-fructose corn syrup (HFCS). The consumption of HFCS
increased rapidly in the 1970s owing to its sweet taste and low price.
Epidemiological and experimental evidence suggests that fructose ingestion is a risk
factor for metabolic diseases, such as obesity, hypertension, dyslipidemia, and
insulin resistance [1]
[2]
[3]. Recently, HFCS consumption has increased, especially in children
[4]. Vos et al. showed that nearly
one-fourth of the adolescent population obtain more than 15% of their
calories from fructose [5]. Moreover,
rates of adolescent obesity have been on the rise in recent years, from 1%
in 1975 to 8% in 2016 [6]. Thus,
there is a concern that HFCS intake during adolescence may contribute to increasing
rates of certain metabolic disorders.
Adolescence is the period of organ development and physical growth from childhood to
adulthood, defined as 10–19 years of age in humans and as postnatal days
(PD) 21–60 in rodents [7]. The
importance of the adolescent environment is becoming clearer; a growing body of
evidence indicates that children and adolescents are more sensitive than adults to
the effects of food intake/nutrition, such as fructose consumption. Hsu et
al. showed more significant effects of fructose intake on brain function, metabolic
outcomes, and neuroinflammation in adolescence than in adulthood [8]. In addition, impaired insulin
signaling because of high-fructose intake has been observed during adolescence but
not during adulthood [8]
[9]. However, the causes of the difference
in vulnerability between adolescence and adulthood has not yet been elucidated.
Glucocorticoids (GCs; corticosterone in rodents, cortisol in humans) are hormones
important for the regulation of lipid and glucose metabolism and blood pressure
[10]
[11]. GCs are synthesized in the adrenal
gland, and their synthesis is regulated by the hypothalamic-pituitary-adrenal axis.
Synthesized GCs are regulated by 11 beta-hydroxysteroid dehydrogenase 1, 2 (Hsd11b1,
Hsd11b2) and the steroid 5 alpha-reductase 1 (Srd5a1) enzymes [12]
[13]. Hsd11b2 is mainly expressed in the kidney, while Srd5a1 is expressed
in the liver [14]
[15], and these enzymes convert active GC
to inactive form [12]
[13]. Hsd11b1 is mainly expressed in the
liver and skeletal muscle, and regenerates active GC from its inactive form [16].
The alterations in the expression of GC metabolizing enzymes disrupt GC action and
can cause disorders in the affected tissues [16]. For example, Hsd11b1 overexpression in the liver causes hepatic
dysfunction [17], the deletion of Srd5a1
in mice causes insulin resistance and hepatic steatosis [14], and kidney-specific Hsd11b2 knock-out
in mice leads to hypertension [18].
Fructose intake alters the expression of enzymes involved in GC metabolism [19]
[20]. London et al. showed that high-fructose diets increased
hepatic expression of Hsd11b1 [19].
Likewise, Vasiljević et al. reported that fructose-induced increase in
Hsd11b1 expression may lead to disruption of insulin signaling [20]. These studies suggest the induction
of abnormal GC metabolism due to fructose intake. Given that fructose intake before
adulthood is harmful, fructose-induced metabolic abnormalities in GC may
consequently be more pronounced in children and adolescents than in adulthood.
However, previous studies have not considered the effects of fructose intake on
individuals at different stages of maturation [19]
[20]. To better understand
the mechanism, fructose-induced disease pathogenesis needs to be analyzed based on
the developmental stage.
In this study, we aimed to test the effects of 20% HFCS solution (containing
11% fructose) on GC metabolism at different developmental stages of
experimental rats, including childhood and adolescence (PD21–60), young
adulthood (PD60–100), and adulthood (PD100–140).
Materials and Methods
Animals
This study was approved by the Animal Ethics Committee of Fujita Health
University (Permit No. H0862). Male Sprague-Dawley (SD) rats (SLC, Shizuoka,
Japan) were obtained from Japan SLC (Hamamatsu, Japan) and kept under standard
conditions at room temperature (23°C±3°C) under a
12:12 h light-dark cycle (the light period started at 8:00 am). In
animal studies using rodents, adolescence is indicated as the period from
weaning (PD21) to PD60. The period from PD60 to PD100 is defined as young
adulthood [21]. In this study, the
period after young adulthood was defined as adulthood. We divided the
experimental period into three parts: childhood and adolescence
(PD21–60; Period I), young adulthood (PD60–100; Period II), and
adulthood (PD100–140; Period III) [22]. The control group (C, n=7–8) received distilled
water for 40 days, whereas the HFCS-fed group (H, n=7–8)
received 20% HFCS solution for the same period. A 20% aqueous
HFCS solution was prepared using 75% HFCS (Japan Corn Starch, Tokyo,
Japan). Standard chow (MF; Oriental Yeast, Tokyo, Japan) and water were
available ad libitum to all animals throughout the experimental period.
According to data from the Third United States National Health and Nutrition
Examination Survey, the mean fructose consumption was>10% of
total caloric intake and constituted approximately 20% of total caloric
intake in the top 5% of fructose consumers [5]. As fructose is most commonly
consumed in the form of HFCS, this study modeled conditions that relatively
substituted the modern human diet. The body weight of the rats was measured
every two weeks. At the end of the experimental period, the rats were fasted for
6 h before being perfused with saline and dissected under isoflurane
anesthesia. The kidney, adrenal gland, gastrocnemius muscle, and liver were
harvested and stored at −80°C until use.
Quantification of mRNA expression
Total RNA was isolated from the kidney, adrenal gland, gastrocnemius muscle, and
liver using TRIzol Reagent (Thermo Fisher Scientific, Waltham, MA, USA)
according to the manufacturer’s instructions [23]. Quantitative polymerase chain
reaction (qPCR) was performed using the QuantStudio 7 Flex system (Thermo Fisher
Scientific, Waltham, MA, USA) and the Thunderbird SYBR qPCR Mix (Toyobo, Osaka,
Japan). Conditions were as follows: 95°C for 1 min and 40 cycles
of 95°C for 15 s, 55°C for 30 s, and
72°C for 1 min. The target gene primers were designed by Fasmac
Co., Ltd. (Kanagawa, Japan) and have been described previously [24]. The expression levels of target
genes were normalized to the mRNA levels of actin beta (Actb) as
an internal control. We calculated the fold changes in the expression between
the control and HFCS-fed groups using the 2-∆∆Ct
method [25].
Western blotting
The kidney, adrenal gland, gastrocnemius muscle, and liver were homogenized in
RIPA buffer (Wako Pure Chemicals, Osaka, Japan). In brief, the extracted protein
were boiled in EzApply Buffer (Atto, Tokyo, Japan) for 3 min. Sodium
dodecyl sulfate-treated proteins (20 μg) were separated by
electrophoresis on a 12.5% polyacrylamide gel (Atto, Tokyo, Japan) and
transferred onto membranes. (Atto, Tokyo, Japan) The membranes were incubated
overnight at 4°C with primary antibodies against Hsd11b1 (ab39364;
Abcam, Cambridge, UK), Hsd11b2 (14192–1-AP; Proteintech, Rosemont, USA),
Srd5a1 (ab110123; Abcam, Cambridge, UK), and beta-actin (ab8227; Abcam,
Cambridge, UK) [26]. Subsequently,
the membranes were incubated with horseradish peroxidase-conjugated secondary
antibody for 1 h at room temperature. Immunoreactive bands were
visualized using Immobilon Western Chemiluminescent HRP Substrate (Millipore,
Billerica, MA, USA). The intensities of specific chemiluminescence bands were
analyzed using a Lumino image analyzer (ImageQuant LAS 3000; GE Healthcare,
Amersham, UK).
Activities of 11 beta-hydroxysteroid dehydrogenase 1 and 2 (Hsd11b1and
Hsd11b2) enzymes
Liver was homogenized to 40 mg/mL concentration in 30 mM
Tris buffer (pH=7.4) containing 0.9 mM
ethylenediaminetetraacetic acid and 0.3 mM sucrose. Reaction buffers A
and B were then prepared. The activity of Hsd11b1 or Hsd11b2 was determined in
buffer A or B, respectively. Reaction buffer A was 0.01 M sodium
phosphate buffer containing 200 µM cortisone (Merck, Darmstadt,
Germany) and 2 mM NADPH. Reaction buffer B contained 0.01 M
sodium phosphate buffer containing 100 µM cortisol (VWR
International Ltd., Leuven, Belgium) and 2 mM NAD+.
Cortisone and cortisol were used as substrates for Hsd11b1 and Hsd11b2 because
these steroids show high affinity with rat Hsd11b1 and Hsd11b2 [27]
[28]. The reaction was initiated by
adding the homogenate (equivalent to 10 mg of tissue). After incubation
at 37°C for 30 min, the reaction buffer was extracted by adding
2 mL chloroform and 50 µM dexamethasone (Merck,
Darmstadt, Germany) as an extrinsic control. The chloroform phase was
transferred to a tube and evaporated at 45°C to dryness. The residue was
dissolved in 60% methanol and fractionated using a high-performance
liquid chromatography (HPLC) system (UV-2075 Intelligent UV/Vis
Detector, Jasco Inc., Tokyo, Japan) to measure cortisone, cortisol, and
dexamethasone [29]. The reverse-phase
HPLC analysis was performed on a 5C18-MS-II column (Nacalai Tesque,
Kyoto, Japan) at a column temperature of 30°C and detection wavelength
of 254 nm.
Statistical analysis
Statistical analyses were performed using JMP version 14 (SAS Institute, Cary,
NC, USA). All data are expressed as the mean±standard deviation, and
statistical analysis was performed using the Student’s t-test [30]. Statistical significance was
defined as p<0.05.
Results
The experimental rats were fed distilled water or 20% HFCS solution for 40
days during childhood and adolescence (PD21–60), young adulthood
(PD60–100), and adulthood (PD100–140) ([Fig. 1]). Assessment of the effects of
HFCS intake on body weight during each developmental period showed no significant
differences between these periods ([Fig.
1]).
Fig. 1 Experimental design. The experimental period was divided to
assess rats in three stages of their growth: childhood and adolescence
(postnatal day (PD) 21–60; Period I), young adulthood
(PD60–100; Period II), and adulthood (PD100–140; Period
III). Male rats were fed either distilled water or a 20% HFCS
solution for 40 days.
To evaluate the effect of excess HFCS intake on GC metabolism in childhood and
adolescence, we analyzed the mRNA expression of Hsd11b1 and Hsd11b2,
and Srd5a1 in the kidney, adrenal gland, muscle, and liver. Renal
Hsd11b2 levels were lowered during Period I ([Table 1]). In the liver, Hsd11b1
levels increased, and levels of Hsd11b2 and Srd5a1 decreased during
Period I. Hepatic Hsd11b1 and Hsd11b2 expression was not significantly
different in Periods II and III, and Srd5a1 expression decreased during all
periods ([Fig. 2]).
Fig. 2 Effect of HFCS intake on hepatic mRNA levels during each
period. Hsd11b1, Hsd11b2, and Srd5a1 mRNA levels were
quantified by quantitative polymerase chain reaction. The results are
expressed as the ratio of the relative intensity of the gene expression
levels to that of Actin beta. C, control group
(n=7–8); H, HFCS-fed group (n=7–8). Values
are presented as mean±standard deviation.
*p<0.05.
Table 1 List of qPCR analysis of corticosterone metabolizing
enzyme.
|
Period I
|
Period II
|
Period III
|
|
C
|
C
|
C
|
mRNA
|
Relative mRNA Level
|
Relative mRNA Level
|
Relative mRNA Level
|
Liver
|
|
Hsd11b1
|
1.00±0.16
|
1.57±0.38*
|
1.00±0.36
|
1.09±0.22
|
1.00±0.45
|
0.78±0.27
|
Hsd11b2
|
1.00±0.31
|
0.62±0.27*
|
1.00±0.23
|
0.83±0.39
|
1.00±0.26
|
0.95±0.11
|
Srd5a1
|
1.00±0.29
|
0.51±0.12*
|
1.00±0.27
|
0.69±0.34*
|
1.00±0.43
|
0.56±0.22*
|
Kidney
|
|
Hsd11b1
|
1.00±0.29
|
0.73±0.26
|
1.00±0.12
|
1.02±0.28
|
1.00±0.15
|
1.13±0.27
|
Hsd11b2
|
1.00±0.23
|
0.52±0.09*
|
1.00±0.22
|
0.98±0.20
|
1.00±0.22
|
0.88±0.28
|
Srd5a1
|
1.00±0.23
|
0.77±0.40
|
1.00±0.13
|
0.92±0.21
|
1.00±0.16
|
0.91±0.26
|
Adrenal Gland
|
|
Hsd11b1
|
1.00±0.23
|
1.75±0.86
|
1.00±0.23
|
1.00±0.25
|
1.00±0.42
|
0.99±0.23
|
Hsd11b2
|
1.00±0.31
|
0.94±0.11
|
1.00±0.24
|
1.02±0.23
|
1.00±0.26
|
1.24±0.19
|
Srd5a1
|
1.00±0.20
|
1.43±0.61
|
1.00±0.17
|
0.97±0.22
|
1.00±0.40
|
1.16±0.26
|
Muscle
|
|
Hsd11b1
|
1.00±0.46
|
0.81±0.21
|
1.00±0.30
|
1.17±0.43
|
1.00±0.22
|
1.10±0.21
|
Hsd11b2
|
1.00±0.50
|
1.06±0.42
|
1.00±0.64
|
0.61±0.38
|
1.00±0.79
|
1.15±0.25
|
Srd5a1
|
1.00±0.31
|
0.73±0.40
|
1.00±0.50
|
1.40±0.57
|
1.00±0.41
|
0.74±0.18
|
Values are presented as mean ± standard deviation.
Abbreviations: 11 beta-hydroxysteroid dehydrogenase 1, 2
(Hsd11b1, Hsd11b2); steroid 5 alpha-reductase 1 (Srd5a1). C: control group;
H: HFCS-fed group. HFCS: high-fructose corn syrup
*p<0.05.
We also quantified Hsd11b1, Hsd11b2, and Srd5a1 protein levels in these tissues. The
kidney, adrenal gland, and muscle did not show any significant changes ([Table 2]). Following hepatic mRNA
expression, Hsd11b1 protein levels increased significantly, and Hsd11b2 protein
levels decreased significantly during Period I (p=0.05), but the
levels did not differ significantly during Periods II and III ([Fig. 3]). In addition, hepatic Srd5a1
protein levels decreased during Period I, but no difference was observed during
Periods II and III ([Fig. 3]).
Fig. 3 Effect of HFCS intake on hepatic protein levels during each
period. Hsd11b1, Hsd11b2, and Srd5a1 protein levels were quantified using
western blotting. Representative samples of western blot images are shown.
The results were expressed as a ratio of the relative intensity of protein
expression to that of Actin beta. C: control group (n=4–6);
H: HFCS-fed group (n=4–6). Values are presented as
mean±standard deviation. *p<0.05.
Table 2List of western blotting analysis of corticosterone
metabolizing enzymes.
|
Period I
|
Period II
|
Period III
|
|
C
|
C
|
C
|
Protein
|
Relative Protein Level
|
Relative Protein Level
|
Relative Protein Level
|
Liver
|
|
Hsd11b1
|
1.00±0.09
|
1.20±0.12*
|
1.00±0.14
|
1.04±0.09
|
1.00±0.26
|
1.04±0.39
|
Hsd11b2
|
1.00±0.06
|
0.82±0.10*
|
1.00±0.20
|
0.88±0.13
|
1.00±0.34
|
0.93±0.25
|
Srd5a1
|
1.00±0.10
|
0.73±0.13*
|
1.00±0.18
|
0.79±0.13
|
1.00±0.09
|
0.93±0.16
|
Kidney
|
|
Hsd11b1
|
1.00±0.27
|
0.97±0.35
|
1.00±0.07
|
1.01±0.14
|
1.00±0.17
|
0.98±0.07
|
Hsd11b2
|
1.00±0.25
|
1.01±0.27
|
1.00±0.04
|
1.01±0.11
|
1.00±0.31
|
0.97±0.26
|
Srd5a1
|
1.00±0.23
|
1.11±0.31
|
1.00±0.21
|
1.09±0.20
|
1.00±0.23
|
0.85±0.15
|
Adrenal Gland
|
|
Hsd11b1
|
1.00±0.08
|
0.99±0.41
|
1.00±0.35
|
0.96±0.20
|
1.00±0.30
|
1.02±0.40
|
Hsd11b2
|
1.00±0.19
|
0.95±0.19
|
1.00±0.27
|
1.00±0.22
|
1.00±0.18
|
1.04±0.19
|
Srd5a1
|
1.00±0.21
|
1.09±0.20
|
1.00±0.03
|
0.96±0.10
|
1.00±0.22
|
0.94±0.12
|
Muscle
|
|
Hsd11b1
|
1.00±0.21
|
0.99±0.15
|
1.00±0.20
|
0.83±0.28
|
1.00±0.25
|
1.17±0.29
|
Hsd11b2
|
1.00±0.34
|
1.01±0.14
|
1.00±0.32
|
1.04±0.22
|
1.00±0.39
|
1.25±0.39
|
Srd5a1
|
1.00±0.41
|
1.14±0.29
|
1.00±0.28
|
0.82±0.41
|
1.00±0.33
|
0.90±0.26
|
Values are presented as mean ± standard deviation.
Abbreviations: 11 beta-hydroxysteroid dehydrogenase 1, 2
(Hsd11b1, Hsd11b2); steroid 5 alpha-reductase 1 (Srd5a1). C: Control group;
H: HFCS-fed group. HFCS: high-fructose corn syrup
*p<0.05.
We analyzed the activities of Hsd11b1 and Hsd11b2 enzymes that are likely to be
highly sensitive to GC metabolism in each period ([Fig. 4]). As expected, Hsd11b1 enzyme
activity increased while that of Hsd11b2 enzyme decreased during Period I, but no
difference was observed during Periods II and III. The enzymatic activity of hepatic
Hsd11b1 and Hsd11b2 were 10.4±2.5 and 11.0±0.5
pmol/min/mg tissue in the control group, respectively.
Fig. 4 Effect of HFCS intake on hepatic Hsd11b1 and Hsd11b2 enzyme
activities during each period. Hepatic Hsd11b1 and Hsd11b2 enzyme activities
according to HFCS intake are shown. Hsd11b1 activity was calculated based on
the rate of conversion of cortisone to cortisol. Hsd11b2 activity was
calculated based on the rate of conversion of cortisol to cortisone. These
metabolites were measured using high-performance liquid chromatography. C:
control group (n=5–7); H: HFCS-fed group
(n=4–7). Values are presented as mean±standard
deviation. *p<0.05.
Discussion
We analyzed the effects of HFCS intake on GC metabolism in the kidney, adrenal gland,
muscle, and liver at different developmental stages. HFCS-induced increase in
hepatic Hsd11b1 activity was observed only in the liver in Period I. Similarly, the
reduction of hepatic Hsd11b2 activity was also induced only in Period I. Thus, our
study shows that the effects of HFCS in relation to GC metabolism may vary with the
developmental period.
The significant effects of GC exposure are reported more in adolescence than in
adulthood. For example, Kinlein et al. observed chronic GC exposure-induced
reduction in bone density in adolescence but not in adulthood [31]. According to our enzyme activity
analysis, GC did not seem to accumulate in the liver during Periods II and III,
although its accumulation may increase during Period I. Adverse effects of
high-fructose intake were only observed in adolescence [8]
[9]. This phenomenon may be partially explained by the presence or absence
of abnormal hepatic GC metabolism.
In this study, we observed increased Hsd11b1 activity in the rat liver in Period I
but not in Periods II and III. Liver-specific Hsd11b1 overexpressing mice display
insulin resistance and dyslipidemia [17],
while Hsd11b1 knock-out mice are protected from metabolic disorders [32]. These reports may support our finding
that accumulated active GC in the liver exacerbates the adverse effects of fructose
in adolescence. We observed reduced hepatic Hsd11b2 activity by approximately
50% after HFCS intake during Period I. Hsd11b2 is mainly expressed in the
kidney; however, Chia et al. showed that Hsd11b2 enzyme activity in the liver
is about 70% of that in the kidney [33]. The decrease in Hsd11b2 activity in the liver may contribute to the
increase in active GC. Moreover, we observed a reduction in hepatic Srd5a1 activity
in Period I. In one study, the affinity of the Srd5a1 enzyme for GC was 1000-fold
lower than that of Hsd11b2 [34]
[35], implying that the decrease in Srd5a1
activity may not contribute to GC inactivation compared to Hsd11b2 or has only a
negligible effect.
Our results implied that adolescents are more vulnerable to environmental changes
than adults and are more likely to experience negative consequences. However,
several animal experiments have shown that adverse effects of fructose intake occur
not only in adolescence but also in adulthood. This discrepancy in findings may be
explained by differences in the concentration of fructose used. Some studies
regarding the adverse effects of fructose have been conducted using
60–70% fructose concentrations [36]
[37], which is an unlikely
concentration to be ingested as part of a modern diet. Another possible explanation
is the difference in the duration of the intake period. We administered the HFCS
solution for 40 days, unlike some studies in which adult rats were fed 10%
fructose for a longer time [38]
[39]. For example, Ibrahim et al. reported
that adult rats fed a 10% fructose solution in drinking water for 16 weeks
showed increased serum glucose and insulin levels [38]. Notably, these reports have observed
body weight gain, which differs from the results of our study (Supplemental Fig.
1). In summary, the lack of significant effects of HFCS during adulthood in
this study may be because of differences in the concentration of fructose
administration and length of the intake period.
According to our previous studies, exposure to fructose during early life stages such
as fetal and lactation induces changes in miRNA expression and DNA methylation level
in the liver [23]
[40]. These epigenetic modifications may be
maintained at least until adulthood, altering mRNA expression of genes related to
lipid metabolism. These reports suggest that the effects on GC-related gene
expression changes may be observed via epigenetic changes even after HFCS intake is
discontinued. Further study is needed to analyze this point.
The current study has a few limitations. First, we did not clarify the mechanism of
the transcriptional changes in GC metabolizing enzymes in adolescent HFCS intake.
Cooper et al. reported that interleukin-1 beta (Il1b) and tumor necrosis
factor-alpha (TNFa) decrease Hsd11b2 and increase Hsd11b1 expression and activity
[41]. Therefore, we analyzed
Il1b and TNFa mRNA levels but did not observe any increase in
their levels (data not shown). Further analysis is needed to determine the mechanism
of the change in GC metabolizing enzymes in adolescent HFCS intake. Next, this study
only analyzed GC metabolism in the liver and not steroid levels. Fructose intake
increases GC levels, but given that increased GC in the liver is associated with
abnormal lipid metabolism and insulin resistance [42], it is necessary to measure steroid
levels.
In conclusion, we showed that HFCS intake during adolescence may induce hepatic GC
accumulation. This study provides insight into the adverse effects of fructose on GC
metabolism in children in the context of increasing rates of HFCS consumption.