Long term effects of high fat and sucrose diets on obesity and lymphocyte proliferation in mice

Abstract

Objective

To clarify the effect of prolonged feeding of a high-fat and sucrose, and to clarify the effect of sucrose instead of other carbohydrate on obesity and immunity in C57BL/6J mice.

Methods

We investigated the development of obesity and immune cell function in four groups of mice fed high-fat, high-fat plus high-sucrose, high-sucrose, and control diet for 7 months.

Results

Mice fed high-fat and high-fat plus high-sucrose groups developed severe obesity. Body weight, adipose tissue weight, serum leptin, blood glucose, and insulin were significantly higher, while the level of serum soluble leptin receptor was significantly lower in mice fed high-fat and high-fat plus high-sucrose diets than in mice fed the control or high-sucrose diets. Splenocyte proliferation stimulated by T-cell mitogen (PHA, ConA, and anti-CD 3 antibody) and B-cell mitogen (LPS) was significantly lower in both obese, high-fat and high-fat plus high-sucrose groups than in control and high-sucrose groups. However, these parameters did not differ between high-fat and high-fat plus high-sucrose groups.

Conclusions

Long-term feeding of high-fat diet and high-fat plus high-sucrose diet similarly induced severe obesity in C57BL/6J mice. Not only T-cell, but also B-cell function may be impaired in mice made severely obese by the high-fat or high-fat plus high-sucrose diets.

Introduction

Obesity is often associated with immune dysfunction (1, 2). We already reported that several immune functions were changed in obese model mice by high-fat (lard supplemented) feeding (3, 4, 5). We showed that mitogen-stimulated or antigen-specific splenocyte proliferation, cytokine production, and antigen-specific antibody in sera were affected in high-fat diet induced obese mice.

C57BL/6 mice fed a high-fat diet are a good model of human obesity (6, 7). This obese mouse model manifests increased body weight, hyperinsulinemia, hyperglycemia, hyperleptinemia, or fatty liver by three or four months of high-fat feeding. Collins et al reported that the C57BL/6J mice fed high-fat diet show defects in -adrenergic receptor expression and function in adipocytes, while the decline is less dramatic in other strain like A/J mice (8). As genetically obese mice, such as ob/ob or db/db mice, have a mutation in leptin or leptin receptor gene, they rapidly develop obesity with increased intake of standard chow. However, obese humans rarely have these mutant genes, and they frequently show the increment of leptin. Mice with diet-induced obesity represent hyperleptinemia like obese human (7). Since leptin plays a role as an immune regulatory factor (9), mice with diet-induced obesity may be a good model for the analysis of immune function in obese humans.

It is reported that high-sucrose consumption by drinking option induces obesity in rats (10, 11, 12). Consumption of high-fat in combination with a high-sucrose diet is also reported to induce severe obesity, hyperinsulinemia, or hyperglycemia, similar to that seen in type 2 diabetes (13, 14). Furthermore, a recent paper reported that long-term intake of high-fat and high-sucrose diets have different effects on glucose intolerance in C57BL/6 mice (15). However, comparison between high-fat diet and high-fat plus high-sucrose diet in the development of obesity was not fully investigated.

As immune function is affected by adipocyte-derived hormones or metabolic changes due to obesity, it is possible that the different diets may affect immune function differently. It is reported that C57BL/6 mice fed either high-fat, high-sucrose, or combined high-fat and high-sucrose for 9 weeks develop increased hepatic NKT cell apoptosis resulting in reduced liver NKT cells (16). Interestingly, combined high-fat and high-sucrose diets caused even further reduction of hepatic NKT cells in the report (16). However, the effect of long-term obesity-inducing diets on immunity has not been fully investigated since as little as 3-4 months of feeding is sufficient to induce obesity in C57BL/6 mice (6, 7, 17). There is a need to investigate the pathophysiology of obesity induced by long-term feeding in mice, and to clear the effect of sucrose for obesity and immunity instead of other carbohydrate. Accordingly, we set up four groups, each fed a different diet (i.e., high-fat (HF), high-fat plus high-sucrose (HF+HS), high sucrose (HS), and control (Con)) for 7 months, and compared the effects of these diets on obesity and immunity.

Methods and Materials

Animals and experimental protocol

Female C57BL/6J mice were obtained from CLEA Japan (Tokyo, Japan) at 4 weeks of age. The mice were maintained at 22°C with 12-h light:dark cycle. One week after arrival, mice were randomly divided into four groups, each of which were fed one of four diets for 7 months: high-fat (HF), high-fat plus high-sucrose (HF+HS), high-sucrose (HS), and control (Con). The number of mice was 8-9 per group. The compositions of these diets are shown in Table 1. Sucrose was added as a solid to all four diets. These diets were manufactured by Oriental Yeast Co. (Tokyo, Japan). Mice were given free access to food and water. Food intake was recorded weekly for 24 hours per cage. Seven months later, all mice were sacrificed by cervical dislocation, and blood, liver, and spleen were harvested and weighed. Subcutaneous (inguinal/one side of foot pad) and visceral (preperitoneal and mesenteric) fat pads were dissected and weighed from each mouse.

Table 1.

Diet composition

Ingredients (g/100g)	High-fat diet (HF)	High-fat plus high-sucrose diet (HF+HS)	High-sucrose diet (HS)	Control diet (Con)
Milk casein	14.0	14.0	14.0	14.0
L-cystin	0.18	0.18	0.18	0.18
Corn starch	17.6	0	13.5	46.6
-corn starch	15.5	0	15.5	15.5
Sucrose	10.0	43.1	43.1	10.0
Soybean oil	4.0	4.0	4.0	4.0
Lard	28.9	28.9	0	0
Cellulose powder	5.0	5.0	5.0	5.0
AIN-93M mineral mix	3.5	3.5	3.5	3.5
AIN-93M vitamin mix	1.0	1.0	1.0	1.0
Heavy tartaric acid choline	0.25	0.25	0.25	0.25
Tertiary butyl hydroquinone	0.0062	0.0062	0.0008	0.0008
Kcal/100g	518.4	518.4	348.2	348.2
Energy basis (as % of total energy)
Fat	50.3	50.3	10.3	10.3
Carbohydrate	38.9	38.9	73.6	73.6
Protein	10.8	10.8	16.1	16.1

Blood analysis

Blood was drawn in the fed state, and serum was obtained by centrifugation. Serum samples were stored at -30°C until analysis.

Leptin and soluble leptin (s-leptin) receptor were measured by enzyme-linked immunosorbent assays (ELISA) using the Duo set system (R&D Systems Inc., Minneapolis, MN, USA) according to the manufacturer’s instructions. ELISA kits were used to measure serum insulin (Shibayagi Inc., Gunma, Japan). Serum alanine aminotransferase (ALT) were determined using test kit (WAKO, Osaka, Japan). Blood glucose level was measured using an automatic glucose analyzer (ARKRAY Factory, Inc, Shiga, Japan).

Preparation and proliferative response of splenocytes

Single cell suspensions of splenocytes in RPMI-1640 medium (NISSUI, Tokyo, Japan) were obtained by pressing the fragments of the spleen through a stainless steel mesh and filtering the resulting splenocyte suspension through a 250-µm nylon mesh. Cell suspensions were washed and diluted in RPMI-1640 medium containing 5% heat-inactivated fetal calf serum (JRH, Lenaxa, Australia), L–glutamine (2 nmol/L; GIBCO, Grand Island, NY, USA), penicillin (100 U/mL; GIBCO), and streptomycin (100 µg/mL; GIBCO). Cell numbers were adjusted to a density of 4×10⁶ cells/mL. Splenocytes were cultured in 96-well plates with phytohemagglutinin (PHA), concanavalin-A (Con A), anti-CD3 antibody (Cedarlane, Hornby, Ontario, Canada), or lipopolysaccharide (LPS) and incubated at 37°C in a 5% CO₂ atmosphere. After 72 hours of culture, the proliferative responses of splenocytes were measured by the Alamar Blue assay (18).

Statistical analysis

Data were expressed as mean ± SE. The effect of high-fat, high-sucrose, and an interaction of high-fat and high-sucrose was analyzed by two-way ANOVA. Significant difference among means in four groups was identified by Bonferroni’s t-test.

Results

Body weight, food intake and organ weight

Body weight (1 month, 4 months, and 7 months of feeding) and body weight gain were significantly increased in both HF and HF+HS groups compared with each control (HS and Con) (P<0.01). No interaction of high-fat and high-sucrose was observed by two-way ANOVA (Table 2).

Table 2.

Body weight, food intake and organ weight

	Groups				two-way ANOVA
	HF	HF+HS	HS	Con	High-fat effect	High-sucrose effect	High-fat × High-sucrose
Body weight
Initial (g)	14.6 ± 0.2	14.9 ± 0.3	15.0 ± 0.2	14.3 ±0.2
1 month (g)	20.2 ± 0.3a	21.0 ± 0.5a	18.5 ± 0.3b	18.0 ± 0.2b	P<0.0001	N.S.	N.S.
4 months (g)	30.6 ± 1.9a	34.4 ± 1.8a	21.3 ± 0.5b	21 ± 0.2b	P<0.0001	N.S.	N.S.
Final (7months) (g)	45.1 ± 1.6a	46.3 ± 1.4a	23.0 ± 0.5b	23.5 ± 0.4b	P<0.0001	N.S.	N.S.
Body weight gain (g)	30.5 ± 1.7a	31.4 ± 1.3a	7.9 ± 0.4b	9.2 ± 0.3b	P<0.0001	N.S.	N.S.
food intake (kcal/day)	14.2 ± 0.4a	13.9 ± 0.4a	9.0 ± 0.2b	8.9 ± 0.2b	P<0.0001	N.S.	N.S.
Subcutaneous fat pad weight
Inguinal fat pad weight (g)	1.54 ± 0.11a	1.79 ± 0.12a	0.27 ±0.01b	0.24 ± 0.01b	P<0.0001	N.S.	N.S.
Visceral fat pad weight
Peritoneal fat pad weight (g)	2.66 ±0.10a	2.70 ± 0.16a	0.31 ± 0.03b	0.35 ± 0.03b	P<0.0001	N.S.	N.S.
Mesenteric fat pad weight (g)	1.50 ± 0.16a	1.47 ± 0.13a	0.24 ± 0.04b	0.26 ± 0.04b	P<0.0001	N.S.	N.S.
Liver weight (g)	1.66 ± 0.08a	1.95 ± 0.11a	1.06 ± 0.05b	0.96 ± 0.03b	P<0.0001	P=0.01	N.S.
Spleen weight (mg)	106 ± 10.9ac	118 ± 12.1a	81 ± 5.7bc	65 ± 4.5b	P<0.001	N.S.	N.S.

Values are means ± SE. Means in a row without a common letter differ, P<0.01 (without spleen (HF vs. Con, and HF+HS vs. HS), P<0.05); N.S.=not significant.

High-fat, but neither high-sucrose nor the combination of high-fat and high-sucrose, affected organ weights, except for liver weight. Food intake (kcal/day) was presented as the average of measured values over 7 months, and was significantly higher in both HF and HF+HS groups than in each control (HS and Con) (P<0.01) (Table 2).

Blood biomarkers

High-fat, but neither high-sucrose nor the combination of high-fat and high-sucrose, had a significant effect on blood glucose level, insulin level, serum leptin level, soluble leptin receptor level, and ratio of leptin to soluble leptin-R by two-way ANOVA. There were no significant between group differences in ALT, and HF group had the highest ratios of leptin/soluble leptin-R compared with other three groups (P<0.01) (Table 3).

Table 3.

Blood biomarkers

	Groups				two-way ANOVA
	HF	HF+HS	HS	Con	High-fat effect	High-sucrose effect	High-fat × High-sucrose
Blood glucose (mg/dL)	206 ± 12.5a	204 ± 8.2a	124 ± 12.8b	145 ± 12.4b	P<0.0001	N.S.	N.S.
Insulin (pg/mL)	3318 ± 769a	3410 ± 637a	336 ±97b	638 ± 200b	P<0.0001	N.S.	N.S.
ALT (IU/L)	21 ± 3	27 ± 6	20 ± 8	16 ± 6	N.S.	N.S.	N.S.
Serum leptin (ng/mL)	25.7 ± 6.3a	24.3 ± 2.8a	4.2 ± 6.5b	6.3 ± 8.8b	P<0.0001	N.S.	N.S.
Serum soluble leptin receptor (ng/mL)	4.9 ± 3.3a	5.6 ± 2.8a	10.9 ± 1.2b	9.8 ± 5.5b	P<0.0001	N.S.	N.S.
Ratio of leptin/soluble leptin-R	5.4 ± 0.3a	4.4 ± 0.2b	0.4 ± 0.1c	0.6 ± 0.1c	P<0.0001	N.S.	N.S.

Values are means ± SE. Means in a row without a common letter differ, P<0.01.

Proliferative response of splenocytes

Data are expressed as relative stimulation index calculated with fluorescence intensity. High-fat, but neither high-sucrose nor the combination of high-fat and high-sucrose, had a significant effect on proliferative response of splenocytes stimulated with T-cell mitogen (PHA, ConA, and anti-CD3) and B-cell mitogen (LPS) by two-way ANOVA (Table 4).

Table 4.

Proliferative response of splenocytes

	Groups				two-way ANOVA
Stimulation	HF	HF+HS	HS	Con	High-fat effect	High-sucrose effect	High-fat × High-sucrose
PHA (SI)	1.7 ± 0.1a	1.9 ± 0.1ab	2.2 ± 0.1b	2.2 ± 0.2b	P<0.01	N.S.	N.S.
ConA (SI)	2.3 ± 0.1a	2.6 ± 0.1ab	3.2 ± 0.3b	3.1 ± 0.3ab	P<0.01	N.S.	N.S.
Anti-CD3 antibody (SI)	2.1 ± 0.2a	2.5 ± 0.1ab	3.1 ± 0.3b	3.0 ± 0.3ab	P<0.01	N.S.	N.S.
LPS (SI)	2.2 ± 0.1a	2.7 ± 0.1ab	3.3 ± 0.3b	3.2 ± 0.3b	P<0.01	N.S.	N.S.

Values are means ± SE. Means in a row without a common letter differ, P<0.05. SI (Stimulation Index) are expressed as the ratio of the data of mitogen-stimulated cultures to the data of non-stimulated cultures.

Discussion

It is reported that obesity can be induced by a high-fat diet in C57BL/6 mice. Because this high-fat diet induced obesity in mice manifests as increased adipose tissue mass and metabolic abnormalities such as hyperglycemia, hyperleptinemia, or hyperinsulinemia, it is a good model of human obesity and diabetes (6, 7, 17). High-fat plus high-sugar diets are also often used to induce obesity in animals (13, 14, 19). However, very few studies have dealt with whether high-fat and high-sucrose diets have different effects on obesity. Surwit et al reported the differential effect of feeding fat and feeding sucrose for 4 months on the development of obesity (17). In our long term feeding study, the development of obesity, blood biomarkers related to obesity, and other markers (except for liver weight) were not different between our HF and HF+HS groups. Although the group fed the high-fat plus high-sucrose diet had the highest liver weight, all groups had comparable liver function judged as ALT levels. Both HF and HF+HS groups showed markedly increased body weight, adipose tissue, hyperglycemia, hyperleptinemia, and hyperinsulinemia. These findings suggest that long term feeding of high-fat diet, even in the absence of sucrose, induced severe obesity and hyperinsulinemia and possibly insulin resistance. On the other hand, Several studies reported ingestion of sucrose solution in combination with high-fat or standard diet in normal rats induced weight gain or increased glucose intolerance, compared with ingestion of water in rats (10, 11, 12). As the period of observation in their study was the same as in our present study, the form of sucrose (solid or solution) or animal species may be associated with development of adiposis. Although we could not observe the specific effect of sucrose in this study, the comparison to other carbohydrate should be investigated to elucidate whether other carbohydrate has any different effect.

We presented new findings about serum soluble leptin receptor concentration (s-leptin-R) in diet-induced obese mice. Soluble leptin-R is considered a new parameter of leptin resistance by recent human studies. It was reported in human obesity that the level of serum leptin, one of the adipocyte-derived hormones, is markedly increased, while the level of s-leptin-R is decreased (20, 21, 22). Soluble leptin-R is a circulating leptin-binding factor that competes with the binding sites of cellular receptors and modulates steady-state leptin levels (23). It is reported that leptin and s-leptin receptor increased during infection, inflammation, or pregnancy in mice (24, 25). In our present study, in spite of high leptin level in HF and HF+HS groups, s-leptin-R level was significantly lower in these groups than in each control groups (Con and HS). Leptin affects several immune functions (9, 26). Both T-cell number and T-cell function were markedly decreased in genetically obese rat, ob/ob mice (9, 27, 28), and administration of exogenous leptin restored them in ob/ob mice (28). The level of circulating leptin and s-leptin-R may be a key factor maintaining normal immune function.

Then, we analyzed the proliferative response of splenocytes. Our present data indicated that lymphocyte proliferation in response to various T-cell mitogens (PHA, ConA, and anti-CD3 antibody) and B-cell mitogen (LPS) was decreased in both HF and HF+HS groups. The previous studies showed that obesity decreases T-cell number and proliferation but has little effect on B-cell function or antibody production (1, 2, 3, 4, 27). There are several possible mechanisms of obesity-related impairment of immunity, such us reduction of GLUT-1 or insulin receptor expression on T cells (29, 30). However, our findings suggested that severe obesity may induce impairment of not only T-cell but also B-cell proliferation. Additionally, Li Z et al. reported that C57BL mice fed high-fat or high-sucrose diet reduced liver NKT cells (16). So, further study is needed to elucidate the effect of obesity on other cell populations in addition to T cells.

Obesity-related derangement factors, such as glucocorticoids, or insulin in addition to leptin may also affect immunity. Glucocorticoids, released from the adrenal cortex in response to various stressive factors, suppress several immune functions and alter inflammation processes (31). Increased level of glucocorticoids (corticosterone) were reported in genetic obese, ob/ob mice (28). As for insulin, we previously reported that inflammatory cytokine production were altered by type-2 diabetes with obesity (32). Furthermore, insulin-receptor signaling interfaces with inflammatory pathway (33). These obesity-related factors may play role as controllers of immune function, not only in murine models, but in human.

Adipose tissue weight was significantly increased in both HF and HF+HS groups in the present study. Recently, it is described that obesity are associated with low grade inflammation. Macrophage infiltrations into adipose tissue were shown in obesity (34). It is also reported that transplantation of wild-type white adipose tissue normalizes metabolic, immune, and inflammatory alterations in ob/ob mice (35). One of the adipocyte derived hormone, leptin affect various immune function (9). As adipocyte produce other various adipocytokines, increment of adipose tissue may affect immune function and alters inflammatory reaction in obesity. Though we could not clarify the additional effect of sucrose to diet in this study, further study from the view point of immunity and inflammation should be necessary, as obesity possess both pro-inflammatory and anti-immunity aspects.

Acknowledgement: We greatly thank Kawamoto R, Mori M, and Yoneda H for their technical assistance.