Effects of different diets used to induce obesity/metabolic syndrome on bladder function in rats

Abstract

Preclinical and human studies on the relationship between obesity/metabolic syndrome (MetS) and lower urinary tract dysfunction (LUTD) are inconsistent. We compared the temporal effects of feeding four different diets used to induce obesity/MetS, including 60% fructose, 2% cholesterol +10% lard, 30% fructose + 20% lard, or 32.5% lard diet, up to 42 wk, on metabolic parameters and bladder function in male Sprague-Dawley rats. Rats fed a 30% fructose + 20% lard or 32.5% lard diet consumed less food (grams), but only the 32.5% lard diet group took in more calories. Feeding rats a 60% fructose or 30% fructose + 20% lard diet led to glucose intolerance and increased blood pressure. Higher body weight and increased cholesterol levels were observed in the rats maintained on a 2% cholesterol +10% lard diet, whereas exposure to a 32.5% lard diet affected most of the above parameters. Voiding behavior measurement showed that voiding frequency and the total voided volume were lower in the experimental diet groups except for the 30% fructose + 20% lard group. The mean voided volume was lower in the 30% fructose + 20% lard and 32.5% lard groups compared with the control group. Cystometric analysis revealed a decreased bladder capacity, mean voided volume, intermicturition interval, and compliance in the 32.5% lard diet group. In conclusion, experimental diets including 60% fructose, 30% fructose + 20% lard, or 2% cholesterol + 10% lard diet differently affected physiological and metabolic parameters and bladder function to a limited extent, while exposure to a 32.5% lard diet had a greater impact.

INTRODUCTION

According to the Global Burden of Disease Study, the prevalence of obesity has been increasing over the past 30 yr worldwide (1). The Centers for Disease Control data showed that the United States obesity prevalence increased from 30.5% to 42.4% from 1999–2000 through 2017–2018, while severe obesity increased from 4.7% to 9.2%. A common complication of obesity is metabolic syndrome (MetS), a cluster of metabolic disorders that include abdominal obesity, high blood pressure, insulin resistance, and dyslipidemia. In an analysis of 12,047 adults, Shi et al. (2) reported the prevalence of MetS in the obese, overweight, and normal-weight populations was 61.6%, 33.2%, and 8.6%, respectively. A long-term epidemiologic survey estimated that MetS prevalence in the United States increased from 32.5% in 2011–2012 to 36.9% in 2015–2016 (3).

Many clinical studies have suggested an association between obesity/MetS and lower urinary tract dysfunction (LUTD) (4). In a survey of 6,000 men and women ages 18 to 79 yr (5), Vaughan et al. found obesity was associated with urinary frequency in men, stress urinary continence in women, and nocturia in men and women. A larger waist circumference, as a manifestation of abdominal obesity, is associated with an increased risk of lower urinary tract symptoms (LUTS) (6, 7). Penson et al. (8) assessed the risk factors of LUTS in 7,318 men ages 40–79 yr old. They found the risk for moderate to severe LUTS increased by 38% in patients with a body mass index (BMI) of at least 35 kg/m². The prevalence of overactive bladder (OAB) increased as waist circumference or BMI increased, although the relationship varies by gender (9). On the other hand, bariatric surgery can significantly improve LUTS in obese patients (10, 11). In the Boston Area Community Health survey, Kupelian et al. (12) reported a significant association between MetS and the American Urologic Association symptom index. In the Third National Health and Nutrition Examination Survey, men with three or more components of the MetS had a greater risk of LUTS than control subjects (13). A population-based European study also suggested a strong positive association between MetS and the severity of LUTS in patients with benign prostatic enlargement (14).

However, other clinical investigations showed that obesity/MetS was not directly or even inversely associated with LUTS (15). A relationship between age and OAB was observed in a study conducted in Japan, but MetS did not show a clear association with OAB (16). A South Korean study including 33,841 men aged ≥30 yr showed a negative association between MetS and LUTS (17). Another study, including 5,355 men (≥40 yr) who underwent health check-ups, found decreased high-density lipoprotein (HDL) cholesterol was the only variable related to moderate to severe LUTS but not MetS or other metabolic components (18). Park et al. (19) did not observe statistically significant differences in voiding symptoms between the MetS and control groups. A systematic review and meta-analysis revealed the presence of MetS was not significantly associated with the risk of having moderate to severe LUTS (20). Gacci et al. (21) did not find a significant difference in total International Prostate Symptom Score (IPSS) or voiding or storage subscores between men with and without MetS in their meta-analysis. Moreover, Ohgaki et al. (22) found MetS was significantly inversely correlated with storage symptoms in middle-aged men (50–64 yr). Similarly, total IPSS and severity of weak urinary stream were lower in men with MetS compared with control subjects. This effect was more pronounced in men with enlarged prostates (23). The reasons for the above conflicting results are unclear and may include the heterogeneity of the study population/ethnicity, the different concomitant diseases/conditions in the patients, and weaknesses in experimental design. For example, most positive association studies were performed in the Western country, and a negative association was reported in studies conducted in Asia. Vaughan et al. (5) mentioned that the validity of self-report information had not been established. Kupelian et al. (12) measured glucose and lipid levels in nonfasting blood samples; the results were likely to be inaccurate. Ohgaki et al. (16) did not exclude OAB symptoms caused by bladder cancer, bladder calculus, cystitis, prostate cancer, prostatitis, urethritis, and urinary retention. In addition, most studies (14, 17) did not include data on lifestyle variables, such as smoking, coffee and alcohol intake, and physical activity, which might affect urinary symptoms and confound the results; the effects of medications used for the treatment of MetS components were also not analyzed.

Compared with clinical investigations, more controlled studies can be performed in animal models. Diet-induced rodent obesity/MetS models are commonly used for studying the complications of obesity/MetS. High-fructose (24, 25), high-cholesterol (26, 27), and/or high-fat diet (HFD) (28, 29) were used to induce obesity/MetS. Lee et al. (24, 30) showed that 66% of Wistar rats fed a 60% fructose diet for 3 mo developed insulin resistance, hyperinsulinemia, hypertriglyceridemia, hypertension, and detrusor overactivity. Rahman et al. (26) showed that Sprague-Dawley rats fed a diet consisting of 2% cholesterol and 10% lard for 6 mo presented hyperlipidemia, bladder overactivity, and erectile dysfunction. HFD-fed male C57BL6/J mice exhibited symptoms of OAB (31, 32). However, different results were also reported in animal studies. Aizawa et al. (33) found that bladder function was not affected in male C57BL/6J mice fed HFD for 20 wk. Gonzalez et al. (34, 35) showed obese-prone female rats fed a HFD for 15 wk present detrusor underactivity. Ossabaw pigs fed a HFD for 10 mo developed MetS and detrusor underactivity (36). The inconsistent results might be due to the difference in animals used (species, strain, age, and gender), diet composition, and duration. In addition, most studies compared the effects of commercial purified HFD with a standard grain-based chow, which includes different ingredient compositions and nutrient contents changing from batch to batch (37, 38). This raises concerns as the presented results cannot be attributed to obesity alone.

This study aims to clarify the above discrepancies by comparing the temporal effects of obesity/MetS, induced by feeding the custom-made purified ingredient diets with almost identical nutrients differing only in relative amounts of fructose, cholesterol, and/or fat up to 42 wk, on bladder function in male Sprague-Dawley rats.

MATERIALS AND METHODS

Experimental Design, Animals, and Diets

Male Sprague-Dawley rats (6 wk old, Envigo, Inc., Indianapolis, IN) were used for the experiments. All experimental procedures were approved by the Institutional Animal Care and Use Committee of Case Western Reserve University (No. 20120180), Cleveland, OH. Rats were kept in an animal facility under controlled temperature (22°C), humidity (50% ± 5%), and lighting (12:12-h artificial light-dark cycle), and had free access to water and food.

After a 1-wk acclimatization, rats were randomly allocated to feed one of the following five Teklad custom-made purified ingredient diets (Envigo, Inc. Madison, WI) in pellet form (n = 18): 1) control diet (control, TD130186); 2) high-fructose diet [60% (wt/wt) fructose, TD130187]; 3) high-cholesterol moderate high-fat diet [2% (wt/wt) cholesterol 10% (wt/wt) lard diet, TD130188]; 4) high-fructose, high-fat diet [30% (wt/wt) fructose 20% (wt/wt) lard diet, TD130189]; and 5) very high-fat diet [32.5% (wt/wt) lard diet, TD130190]. Table 1 shows the key ingredients of different diets. Besides fructose, cholesterol, and lard component, the amounts of other nutrients in different diets were matched as much as possible. Five groups of rats were fed different diets for 42 wk.

Table 1.

Compositions of diets in different groups

Diet	Control Diet	60% Fructose	2% Cholesterol + 10% Lard	30% Fructose + 20% Lard	32.5% Lard
	TD.130186	TD.130187	TD.130188	TD.130189	TD.130190
Casein	207	207	207	207	240
dl-Methionine	3	3	3	3	3
Corn starch	456	0	360	0	180
Fructose	0	600	0	300	0
Cholesterol	0	0	20	0	0
Lard	30	30	100	200	325
Maltodextrin	200	0	130	150	150
Soybean oil	20	20	20	20	20
Cellulose	23.96	79.81	99.81	53.81	9.81
Mineral mix	50	50	50	55	60
Zinc carbonate	0.04	0.04	0.04	0.04	0.04
Vitamin mix	10	10	10	11	12
Total weight, g	1,000	1,000	1,000	1,000	1,000
Total calorie, kcal	3,600	3,600	3,600	4,500	5,200
Protein, %kcal from	20.2	20.2	20.1	16.2	16.3
Carbohydrate, %kcal from	66.9	66.8	49.7	39.6	23.8
Fat, %kcal from	12.9	12.9	30.2	44.2	59.9

TD, Teklad custom diet number.

Twenty-four-hour food and water intake were measured every 3 wk in the first 24 wk on diets and then at 42 wk. Body weight, body length (nose to anus), abdominal circumference (the largest zone of the abdomen), fasting blood glucose (16-h fast), and 24-h voiding behavior were measured every 3 wk until 36 wk, and the last measurement was at the end point of the study (42 wk on diets). Body mass index (BMI) was calculated by the formula BMI = body weight (g)/the square of body length (cm²). Systolic blood pressure and heart rate were measured every 3 or 6 wk by the tail-cuff method with photoelectric pulse detection (IITC Life Science, Inc., Woodland Hills, CA). Three to five measurements were taken, and the mean was calculated. After 42 wk on different diets, the above parameters were measured, and an intraperitoneal glucose tolerance test was conducted. One day later, a suprapubic bladder tube was surgically implanted, and conscious cystometry was performed 3 days later to evaluate the bladder function. Then, blood was collected in EDTA tubes by cardiac puncture under isoflurane anesthesia and centrifuged at 3,000 rpm for 10 min at 4°C. The supernatant (plasma) was stored in a –80°C freezer for biochemical parameters analysis.

Daily Food, Calorie, and Water Intake

Three rats were housed in one cage. A known amount of food (in the form of pallets) and water were placed in the corresponding slots of the cages in the morning. Twenty-four hours later, the remaining water and food (including any fallen bits) were weighed on an electronic scale. The amount was subtracted from the initial amount to account for the consumed food and water. The mean intake per rat was calculated by dividing the total intake (g) of each cage by the number of rats in the cage. Daily calorie intake was calculated by multiplying the daily food intake (g) by the total energy of the diet (kcal/g).

Intraperitoneal Glucose Tolerance Test

Rats were fasted for 16 h. The blood sample was collected by puncturing the tail end vein using a 25-gauge syringe needle. Blood glucose levels were determined using a One Touch glucometer (OneTouch SureStep, Johnson & Johnson, Inc., New Brunswick, NJ). A 20% glucose solution (2 g/kg of animal weight) was injected intraperitoneally, and glucose levels were measured immediately before injection and 30, 60, and 120 min after injection. The glucose area under the curve (AUC) of intraperitoneal glucose tolerance test (IPGTT) was calculated using the trapezoidal method in Graphpad Prism 6 software.

Twenty-Four-Hour Voiding Behavior

Twenty-four-hour urinary behavior was measured as described previously (39, 40). Briefly, rats were placed in individual metabolic cages (Nalgene Metabolic Cage, Nalge Company, Rochester, NY). Regular water and experimental diets were supplied ad libitum. With the nonwetting funnel and cone, urine and feces were collected separately. Clean plastic beakers used for collecting urine were placed on the force-displacement transducers (FT03, Grass Instruments, Warwick, RI) underneath cages. Data signals were amplified (ETH-401 Transducer Interface Amplifier, iWorx, Dover, NH) and recorded at a sampling speed of 10 events/s using the PolyView data acquisition system (Astro-Med, Inc., West Warwick, RI). Urine output was measured in real time. Data analysis included the number of voids (voiding frequency), total urine volume (sum of all voided volumes), and mean volume per void in the daytime (6 AM to 6 PM), nighttime (6 PM to 6 AM), or 24 h (6 AM to 6 AM).

Suprapubic Bladder Tube Implantation and Conscious Cystometry

A suprapubic tube (SPT) implantation was performed as described previously (39–41). Simply, under 1–2% isoflurane anesthesia, the bladder was exposed, and a circular purse-string suture of 7-0 vicryl absorbable sutures was placed around the top of the bladder. A small incision was made in the bladder apex, and the catheter (PE-50 tubing with a flared tip) was implanted. The suture was tightened. The catheter was tunneled subcutaneously and externalized at the back of the neck. The distal end of the tubing was sealed. The incision was closed. Carprofen (5 mg/kg sc) was administered before surgery and 24 h and 48 h after surgery for analgesia.

Conscious cystometry was performed 72 h later. Rats were placed in individual metabolic cages. A pressure transducer (BP-100, CB Sciences, Dover, NH) and a flow pump (Kent Scientific Corporation, Torrington, CT) were connected to the implanted bladder catheter. The bladder was filled with room temperature saline solution at 5 mL/h while bladder pressure was recorded. Urine was collected in a beaker on a force transducer (FT03, Grass Instruments, Warwick, RI) placed beneath each cage. Data signals were amplified (ETH-401 Transducer Interface Amplifier, iWorx, Dover, NH) and recorded at a sampling speed of 10 events/s using the PolyView data acquisition system (Astro-Med, Inc., West Warwick, RI) for future analysis. After an initial 30-min stabilization period, the data from at least 4 representative micturition cycles were collected. Means of the cystometry parameters, including peak micturition pressure, intercontraction interval, and voided volume, were calculated. Functional bladder capacity (infusion rate multiplied by intercontraction interval) and bladder compliance (functional capacity divided by the difference between basal and threshold pressures) were calculated.

Biochemical Parameters

Plasma insulin levels were measured by a sandwich enzyme-linked immunosorbent assay (ELISA) (rat/mouse insulin ELISA kit, catalog no. EZRMI-13K, EMD Millipore Co., St. Charles, MO). Plasma levels of triglyceride (triglyceride quantification colorimetric/fluorometric kit, catalog no. K622-100, BioVision, Milpitas, CA) and cholesterol (total cholesterol and cholesteryl ester colorimetric/fluorometric assay kit, catalog no. K603-100, BioVision, Milpitas, CA) were measured using the colorimetric assay kits. Plasma TNF-α levels were determined using an ultra-sensitive ELISA kit (rat tumor necrosis factor-α ELISA kit, product no. RAB0479, Sigma-Aldrich, Inc., St. Louis, MO).

Statistical Analysis

Some parameters in some time points were not measured due to time or equipment conflicts. In addition, the tubing was pulled or bit in three rats after SPT implantation: one in the 2% cholesterol + 10% lard group, one in the 30% fructose + 20% lard group, and one in the 32.5% lard group. Statistical analysis was performed using GraphPad Prism 6 (GraphPad Software, La Jolla, CA). All data were expressed as means ± SD. For data obtained over time, two-way ANOVA was used to compare the main effects of diet, time, and their interaction on food (calorie) and water consumption, body weight, abdomen circumference, BMI, blood pressure, heart rate, fasting blood glucose, and 24-four voiding behavior parameters. Post hoc comparisons to the control group were performed with Tukey’s test when initial two-way ANOVA indicated statistical differences among groups. For biochemical measurements, AUC, and cystometry parameters at the end point (after 42 wk of different diets), one-way ANOVA, followed by Tukey’s post hoc test, was used for analysis. P < 0.05 was considered statistically significant.

RESULTS

Daily Food, Calorie, and Water Intake

The average daily consumption of food (calories) and water for an adult rat on the control diet was around 20 g (70 kcal) and 35 mL, respectively (Fig. 1). Two-way ANOVA analysis of the daily food, calorie and water consumption revealed significant main effects of types of diets [F_(4,195) = 21.68, P < 0.0001; F_(4,195) = 3.350, P = 0.0111; F_(4,202) = 3.238, P = 0.0133, respectively] and weeks on diets [F_(8,195) = 5.235, P < 0.0001; F_(8,195) = 5.233, P < 0.0001; F_(8,202) = 6.752, P < 0.0001, respectively] and nonsignificant interactions [F_(32,195) = 0.8738, P = 0.6649; F_(32,195) = 0.8251, P = 0.7354; F_(32,202) = 0.9937, P = 0.4833, respectively]. Post hoc tests indicated rats fed a 30% fructose + 20% lard or a 32.5% lard diet consumed less food in grams in general (Fig. 1A), but the 32.5% lard diet group took in more calories compared with the control group (Fig. 1B). Fluid intake was higher in rats on a 30% fructose + 20% lard diet compared with other groups (Fig. 1C).

Figure 1.

The average daily food consumption (A), calorie intake (B), and water intake (C) over time in male Sprague-Dawley rats fed different diets at the age of 7 wk for another 42 wk. Calculated from 4 to 6 cages/group. Fru, fructose; Cho, cholesterol. Data are expressed as mean ± SD. Two-way ANOVA was performed to compare the main effects of diet, time, and their interaction, followed by Tukey’s test as a post hoc method to compare experimental diet groups to the control diet group. *P < 0.05, #P < 0.01, compared with the control group.

Body Weight, Abdomen Circumference, and BMI

Two-way ANOVA test showed significant main effects of weeks on diets and types of diets, and their interaction on the body weight [F_(13,1130) = 985.2, P < 0.0001; F_(4,1130) = 71.63, P < 0.0001; F_(52,1130) = 3.247, P < 0.0001, respectively], abdomen circumference [F_(13,1115) = 1569, P < 0.0001; F_(4,1115) = 90.83, P < 0.0001; F_(52,1115) = 5.548, P < 0.0001, respectively], and BMI [F_(13,1115) = 238.9, P < 0.0001; F_(4,1115) = 70.16, P < 0.0001, F_(52,1115) = 3.356, P < 0.0001, respectively] (Fig. 2). Post hoc test indicated rats fed a 32.5% lard or 2% cholesterol + 10% lard diet had a higher body weight, abdominal circumference, and BMI, but those fed a 60% fructose or 30% fructose + 20% lard diet had a lower body weight and BMI than those fed the control diet in the first 24 wk (Fig. 2, A, C, and D). The body weight (Fig. 2B), abdomen circumference, and BMI at the experimental end point are presented in Table 2.

Figure 2.

A–C: body weight (A), abdomen circumference (C), and BMI (D) over time in male Sprague-Dawley rats fed different diets at the age of 7 wk for another 42 wk; n = 12–18. B: Comparison of the body weight of rats fed different experimental diets for 42 wk. Fru, fructose; Cho, cholesterol. Data are expressed as mean ± SD; n = 18. Two-way ANOVA was performed to compare the main effects of diet, time, and their interaction, followed by Tukey’s test as a post hoc method to compare experimental diet groups to the control diet group in A, C, and D. One-way ANOVA followed by Tukey’s post hoc test was used for analysis of data in B. *P < 0.05, #P < 0.01, compared with the control group in A, C, and D. *P < 0.01, compared with the other 4 groups in B. #P < 0.05, compared with the 60% fructose group in B.

Table 2.

General characteristics and biochemical parameters at the end of the study (42 wk after feeding different diets)

	Control	60% Fructose	2% Cholesterol 10% Lard	30% Fructose 20% Lard	32.5% Lard
Body wt, g	556.2 ± 47.52	539.6 ± 36.08	588.2 ± 46.65#	548.9 ± 47.68	634.2 ± 56.68*#Δ†
Abdominal circumference, cm	24.47 ± 0.96	24.59 ± 0.98	25.15 ± 0.97	24.78 ± 1.12	27.64 ± 1.55*#Δ†
BMI, g/cm2	0.74 ± 0.05	0.72 ± 0.03	0.77 ± 0.05#	0.73 ± 0.04	0.82 ± 0.06*#Δ†
Blood pressure, mmHg	132.4 ± 5.09	156.8 ± 9.69*	144.2 ± 19.45	171.0 ± 7.13*Δ$	146.6 ± 14.57†
Heart rate, beats/min	433.0 ± 47.30	461.0 ± 54.07	436.7 ± 36.08	433.5 ± 34.68	457.7 ± 45.54
Blood glucose, mg/dL	109.7 ± 22.24	113.3 ± 19.00	121.3 ± 21.00	129.4 ± 32.24	129.7 ± 30.98
Insulin, ng/mL	1.23 ± 0.37	1.21 ± 0.40	1.13 ± 0.49	1.09 ± 0.60	1.84 ± 0.72*#Δ†
Cho, mg/dL	112.8 ± 7.02	118.4 ± 11.85	129.0 ± 5.63*	116.2 ± 15.25	123.8 ± 12.63
TG, mM	0.31 ± 0.15	0.32 ± 0.26	0.31 ± 0.36	0.34 ± 0.22	0.81 ± 0.25*#Δ†
TNF-α, pg/mL	19.0 ± 18.60	35.5 ± 34.50	32.5 ± 36.21	26.4 ± 20.27	531.2 ± 564.7*#Δ†

Values are expressed as means ± SD; n = 8–18 rats. BMI, body mass index; wt, weight; Cho, cholesterol; TG, triglyceride; TNF-α, tumor necrosis factor-α. One-way ANOVA, followed by Tukey’s post hoc test, was used for the analysis. *Significantly different from the corresponding value in the control group. #Significantly different from the corresponding value in the 60% fructose group. ΔSignificantly different from the corresponding value in the 2% cholesterol 10% lard group. †Significantly different from the corresponding value in the 30% fructose 20% lard group. $Significantly different from the corresponding value in the 32.5% lard group.

Blood Pressure and Heart Rate

The systolic blood pressure gradually increased in the first 12 wk and was then stable in rats on the control diet (Fig. 3A). Two-way ANOVA revealed significant main effects of weeks on diets and types of diets with interaction on blood pressure [F_(13,1045) = 59.21, P < 0.0001; F_(4,1045) = 36.13, P < 0.0001; F_(52,11045) = 2.539, P < 0.0001, respectively] but without interaction on heart rate [F_(13,1045) = 10.4, P < 0.0001; F_(4,1045) = 6.95, P < 0.0001; F_(52,11045) = 1.121, P = 0.2619, respectively] (Fig. 3). In general, the systolic pressure of rats exposed to a 60% high-fructose, 2% cholesterol + 10% lard, 30% fructose + 20% lard, or 32.5% lard diet was higher than rats fed the control diet, whereas the heart rate was higher only in rats fed a 32.5% lard diet. The final systolic blood pressure and heart rate are shown in Table 2. Rats fed a 30% fructose + 20% lard diet had the highest systolic blood pressure after 42 wk of feeding, followed by those fed a 60% fructose diet.

Figure 3.

Blood pressure (A) and heart rate (B) over time in male Sprague-Dawley rats fed different diets at the age of 7 wk for another 42 wk. Fru, fructose; Cho, cholesterol. Data are expressed as mean ± SD. Two-way ANOVA was performed to compare the main effects of diet, time, and their interaction, followed by Tukey’s test as a post hoc method to compare experimental diet groups to the control diet group. #P < 0.01, compared with the control group.

Fasting Blood Glucose, Insulin, and IPGTT

Two-way ANOVA results indicated significant main effects of weeks on diets [F_(13,1111) = 62.36, P < 0.0001] and types of diets [F_(4,1111) = 16.59, P < 0.0001] and a significant interaction [F_(52,1111) = 1.467, P = 0.0185] on fasting blood glucose level (Fig. 4A). In general, rats fed a 60% high-fructose, 2% cholesterol + 10% lard, 32.5% lard, and a 30% fructose + 20% lard diet had a higher fasting blood glucose than the control diet group. Plasma insulin levels were increased only in rats fed a 32.5% lard diet for 42 wk (Table 2). The results of the IPGTT are presented in Fig. 4B. The AUC was significantly greater in the 60% high-fructose, 30% fructose + 20% lard, and 32.5% lard diet groups but not in the 2% cholesterol + 10% lard diet group, compared with the control group (Fig. 4C), indicating an impaired glucose tolerance.

Figure 4.

A: blood glucose over time in male Sprague-Dawley rats fed different diets at the age of 7 wk for another 42 wk; n = 12–18. B: intraperitoneal glucose tolerance test (IPGTT) at the end of the experiment after overnight fasting. C: comparison of the area under the curve (AUC) of IPGTT; n = 16–18. Fru, fructose; Cho, cholesterol. Data are expressed as mean ± SD. Two-way ANOVA was performed to compare the main effects of diet, time, and their interaction, followed by Tukey’s test as a post hoc method to compare experimental diet groups to the control diet group in A. One-way ANOVA followed by Tukey’s post hoc test was used for analysis of data in C. #P < 0.01, compared with the control group.

Plasma Cholesterol, Triglyceride, and TNF-α

Plasma cholesterol levels were significantly increased only in rats fed a 2% cholesterol + 10% lard diet. Triglyceride and TNF-α levels were significantly increased in rats fed a 32.5% lard diet for 42 wk compared with those kept on other diets (Table 2). There were no statistically significant differences among other groups.

Twenty-Four-Hour Voiding Behavior

Two-way ANOVA showed significant main effects of weeks on diets [F_(11,910) = 10.03, P < 0.0001] and types of diets [F_(4,910) = 22.73, P < 0.0001] but no significant interaction [F_(44,910) = 0.9141, P = 0.6327] on 24-h voiding frequency (Fig. 5, A–C). Pairwise comparison of main effects of types of diets on 24-h voiding frequency indicated a significant decrease in the 60% high-fructose (P < 0.05), 2% cholesterol + 10% lard (P < 0.0001), and 32.5% lard diet (P < 0.0001) groups compared with the control diet group, particularly during nighttime from 3 to 24 wk. We also detected significant main effects of weeks on diets [F_(11,910) = 15.23, P < 0.0001] and types of diets [F_(4,910) = 4.755, P = 0.0008] but no significant interaction [F_(44,910) = 1.029, P = 0.4221] on 24-h mean voided volume (Fig. 5, D–F). Although the changes were not consistent over time, the mean voided volume on average decreased in rats fed a 30% fructose + 20% lard or a 32.5% lard diet compared with those fed the control diet during daytime (P = 0.0041, P = 0.0372, respectively), nighttime (P = 0.0001, P = 0.0147, respectively), or 24 h (P < 0.0001, P = 0.0377, respectively). The mean voided volume in rats fed a 2% cholesterol + 10% lard diet decreased only during nighttime (P = 0.0013). Both weeks on diets and types of diets [two-way ANOVA: interaction F_(44,910) = 1.125, P = 0.2695; effect of week F_(11,910) = 3.148, P = 0.0003; effect of diet F_(4,910) = 32.47, P < 0.0001] affected total voided volume, which was lower on average in all four experimental diet groups during 24 h (P < 0.0001) and nighttime (P < 0.0001) than the control diet group (Fig. 5, G–I).

Figure 5.

Voiding behavior measurements of voiding frequency (A, B, and C), voided volume (D, E, and F), and total voided volume (G, H, and I) during 24 h (A, D, and G), daytime (B, E, and H), or nighttime (C, F, and I) over time in male Sprague-Dawley rats fed different diets at the age of 7 wk for another 42 wk; n = 12–18. Fru, fructose; Cho, cholesterol. Data are expressed as mean ± SD. Two-way ANOVA was performed to compare the main effects of diet, time, and their interaction, followed by Tukey’s test as a post hoc method to compare experimental diet groups to the control diet group. *P < 0.05, #P < 0.01, compared with the control group.

Cystometry Measurement

All rats showed regular and periodic emptying of the bladder. Peak micturition pressure was comparable among the five diet groups (one-way ANOVA, P = 0.0807) (Fig. 6A). There were significant decreases in bladder capacity (Fig. 6B) and mean voided volume (Fig. 6C) in rats fed a 32.5% lard diet compared with those fed other diets, whereas no statistical differences were found among the other four diet groups. The inter-micturition interval (Fig. 6D) in rats fed a 32.5% lard diet decreased compared with other groups but did not reach a statistical difference compared with the control diet group. The compliance was significantly lower in rats fed a 32.5% lard diet than those fed a 60% fructose diet or a 2% cholesterol + 10% lard diet (Fig. 6E).

Figure 6.

Cystometric parameters including peak micturition pressure (A), bladder capacity (B), voided volume (C), intercontraction interval (D), and bladder compliance (E) in male Sprague-Dawley rats fed different diets at the age of 7 wk for another 42 wk; n = 11–12. Fru, fructose; Cho, cholesterol. Data are expressed as mean ± SD. One-way ANOVA followed by Tukey’s post hoc test was used to compare all 5 groups. *P < 0.05, compared with the other 4 groups. #P < 0.05, compared with the 60% fructose, 30% fructose + 20% lard, and 2% cholesterol + 10% lard diet group. @P < 0.05, compared with the 60% fructose and 2% cholesterol + 10% lard diet group.

DISCUSSION

Both preclinical and human studies have produced contradictory results regarding the relationship between obesity/MetS and LUTD manifestations (15). The reasons for these discrepancies are not entirely clear. In this study, we examined the temporal effects of different diets used to induce obesity/MetS on bladder function. We found feeding rats four experimental diets affects physiological and metabolic parameters, and bladder function to a limited extent, with the largest effect produced by exposure to a 32.5% lard diet.

MetS can be diagnosed in clinical practice if three or more of these conditions are met: large waist, high blood pressure, elevated fasting blood glucose, high triglyceride level, or low high-density lipoprotein (HDL) cholesterol (42). In the current study, rats fed a very high-fat diet (32.5% lard) presented the most signs of MetS. Rats in the 2% cholesterol + 10% lard diet group had an increased body weight and plasma cholesterol, whereas feeding 60% fructose or 30% fructose + 20% lard diet resulted in higher blood pressure and impaired glucose tolerance compared with those fed the control diet. The different effects of these diets on the development of obesity/MetS may be due to the differences in food (calorie) intake and diet composition.

Food intake (grams) can be affected when rats are maintained on a HFD (43). Rats fed a 30% fructose + 20% lard or 32.5% lard diet consumed less food in grams, although the daily energy intake increased in the 32.5% lard diet group. Appetite is partially controlled by energy homeostatic processes through the measure of calories consumed (44). Rats may adjust their food intake and energy expenditure according to the energy density of the diet. The increase in energy ingestion enhances the satiety signals, decreasing food consumption to prevent uncontrolled weight gain (45). Another factor is the palatability of the diet. Rats fed a high-fat, high-sugar, and high-salt cafeteria diet (containing cookies, liver pate, bacon, water, and milk) consumed more food and grew faster than rats fed a classical HFD (46). Palatable foods may activate brain reward circuits, driving consumption of food beyond homeostatic needs (47).

Chronic consumption of a Western diet, characterized by a high daily intake of refined carbohydrates (glucose, fructose, and sucrose) and saturated fats, was proven to be one of the reasons causing obesity/MetS (48, 49). Therefore, diet-induced models are commonly used for studying the complications of obesity/MetS. High carbohydrate (fructose or sucrose) (24, 25, 50), high fat (animal or vegetable based) (28, 29), or their combination (51), or combined with high cholesterol (26, 27) were widely used. Fructose can be delivered by replacing starch (isocaloric) (24, 25, 52) or by adding it to drinking water (hypercaloric) (53, 54). Some previous studies showed that high-fructose-diet feeding could lead to impairment of lipids metabolism and insulin resistance with limited or no change in body weight gain (24, 25, 52). The changes were more significant in the hypercaloric high-fructose diet (55). In comparison, HFD has been shown to cause increases in body weight, visceral adiposity, and plasma triglyceride, glucose, and insulin levels (56, 57). A high-fructose, high-fat diet can induce most of the metabolic disorders that occur in MetS, including obesity, dyslipidemia, glucose intolerance, hyperinsulinemia, and inflammation (51, 58, 59).

In this study, we used custom-made purified ingredient diets. We tried to match the nutrients in different diets as much as possible, except for fructose, cholesterol, and lard. As expected, rats fed a very HFD (32.5% lard) consumed more calories, developed obesity, hyperinsulinemia, hypertension, and hyperlipidemia and had higher levels of plasma TNF-α. Surprisingly, the other three experimental diets had no or moderate impact on metabolic parameters. These results are comparable with some previous studies (60–63) but not others (26, 64). For example, a high-fructose diet has been shown to cause an increase in fasting blood glucose in Sprague-Dawley (65) and Wistar rats (66). However, some studies have found no high-fructose diet-induced change in blood glucose levels in either Sprague-Dawley or Wistar rats (60). The discrepancy may be due to the differences in the species and strains used, the diet composition, the start age, and the duration of the diet. Marques et al. found Wistar rats fed a HFD exhibited more pronounced weight gain, hyperlipidemia, and hyperinsulinemia than Sprague-Dawley rats (67). C57BL/6J mice showed profound increases in their body fat content, while A/J mice and C57BL/KsJ mice were relatively resistant to a HFD (68). Animals initiating diet at early stages of life, for example, immediately after being weaned, can enhance diet-induced pathological effects in the rats (69), and metabolic disorders occur more rapidly than induction performed in adult rats. Duration of diet exposure also strongly determines the metabolic outcome (70). Sex differences in the development of diet-induced obesity, insulin resistance, and glucose intolerance were also observed in rodents, with males being more affected (70, 71). Improper use of control diets, such as grain-based standard chow as a control diet to compare with a purified diet, may lead to different conclusions (37, 38).

We evaluated voiding behavior in rats fed different diets. The results of 24-h urination behavior measurement showed that average voiding frequency was lower in 60% high-fructose, 2% cholesterol + 10% lard, and 32.5% lard diet groups, and total voided volume was lower in all four experimental diets compared with the control diet group, particularly during nighttime. These changes may be due to less water intake during the test period (33). Water intake decreased in mice fed a HFD during 24-h urination behavior measurement (33). Statistical analysis of the main effect of diets on mean voided volume showed that it was lower in rats fed a 30% fructose + 20% lard, or a 32.5% lard diet. However, this variation is not consistent across the investigation period. No significant changes in mean voided volume were found in rats fed the other three types of diets. All the obesity/MetS-associated symptoms, including high blood pressure, hyperlipidemia, and high blood glucose, could potentially affect voiding behavior. Spontaneously hypertensive rats voided less volume but had equal urinary frequencies compared with age-matched controls (72). FVB^db/db mice, a type 2 diabetic mouse model, exhibited increased mean volume per void but decreased voiding frequency compared with the control mice at 12 and 24 wk (73). Leptin-deficient obese male B6.V-Lepob/J mice presented a significantly increased voiding frequency and decreased average urine volume per void (74). In the current study, feeding a 60% high-fructose, 2% cholesterol + 10% lard, or 30% fructose + 20% lard diet produced a mild or moderate impact on metabolic parameters, which may be one of the reasons that mean voided volume was not significantly affected. On the other hand, Aizawa et al. (33) found although the body weight and blood glucose level in mice fed a HFD were remarkably higher than those fed the control diet for 25 wk, the mean voided volume was not significantly different between the two groups. The voiding frequency and total voided volume in the HFD group were lower than those in the control diet group due to decreased water intake (33). This study concluded that HFD-induced obesity did not affect LUT function substantially. Similarly, Kim et al. (75) did not see evidence of bladder dysfunction in C57BL/6J mice on a HFD for 16 mo, although the mice exhibited more weight gain.

Conscious cystometry evaluation showed no difference in peak micturition pressure among the five groups. Rats fed a 32.5% lard diet for 42 wk had a decreased bladder capacity, mean voided volume, and intermicturition interval, indicating an overactive bladder. However, the decreased mean voided volume in cystometry is not consistent with the results from 24-h urination behavior measurement assessed after 42 wk of HFD feeding. We do not know the reasons for this inconsistency. Previous studies showed that HFD feeding could induce low-grade systemic inflammation (76), which was supported by the current data showing increased plasma TNF-α levels in rats fed a 32.5% lard diet. We performed cystometry measurement 72 h after bladder catheter implantation. The HFD-induced chronic systemic inflammation could exacerbate the injury induced by the bladder catheter implantation surgery and delay wound healing, which may contribute to the overactive bladder characteristics in the cystometry measurement.

Our experiments had some limitations. First, we used Sprague-Dawley rats instead of Wistar rats, which probably are more susceptible to diet-induced obesity/MetS (67), and this may be one of the reasons for the mild/moderate metabolic response we observed in experimental diet groups. Second, our experiments were restricted to male rats since some studies showed females were less susceptible to dietary obesity and related metabolic complications (77, 78). In addition, the estrous cycle (4–5 days) in female rats is a confounding variable affecting food intake (79, 80) and micturition pattern (81, 82). Therefore, the data generated from female animals need to be interpreted carefully. Third, we monitored the body weight, abdomen circumference, blood pressure, heart rate, fasting blood glucose, and 24-h voiding behavior every 3–6 wk. These procedures inevitably stressed the rats, potentially affecting their food intake and body weight gain. Finally, we did not measure visceral white adipose tissue, which may increase even though the body weight was not significantly increased (83).

In conclusion, feeding a 60% fructose, 30% fructose + 20% lard, or 2% cholesterol 10% lard diet in male Sprague-Dawley rats for 42 wk had no or moderate effects on physiological and metabolic parameters. In comparison, the 32.5% lard diet had a larger impact on the above parameters. Rats fed a 30% fructose + 20% lard or 32.5% lard diet had a decreased voiding frequency, mean voided volume, and total voided volume, whereas those fed a 32.5% lard diet showed the signs of overactive bladder in cystometry measurement. Further studies are needed to determine the relationship between obesity/MetS and LUTS and the potential mechanisms.

Perspectives and Significance

The relationship between MetS and benign prostatic hyperplasia/LUTS is complicated. There are mainly three versions of the definition for MetS (42). The National Cholesterol Education Program Adult Treatment Panel III definition (3 or more of the following 5 criteria are present: waist circumference >102 cm, triglyceride level ≥150 mg/dL, HDL level <40 mg/dL, fasting glucose level ≥110 mg/dL, and blood pressure ≥130/85 mmHg) was mostly used. The conclusion drawn from the studies including patients with different components of MetS may be different since different combinations affect LUT to varying degrees. In addition, the impacts of lifestyle-related factors and the drugs the patients take for MetS components on LUTS need to be considered and analyzed. Therefore, a future study including a large database with extensive information is required. To better understand the underlying mechanisms, the impacts of MetS or its component(s) on the lower urinary tract in animal models still need to be studied.

GRANTS

F.D. and G.L. were supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant R01-DK110567.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

G.L. conceived and designed research; L.W. and M.W. performed experiments; L.W., G.L., M.W., S.M., and P.F. analyzed data; L.W. P.F., D.C., B.W., S.G., A.H., F.D., and G.L. interpreted results of experiments; L.W. and G.L. prepared figures; L.W. and G.L. drafted manuscript; L.W., M.W., S.M., P.F., D.C., B.W., S.G., A.H., F.D., and G.L. edited and revised manuscript; L.W., M.W., S.M., P.F., D.C., B.W., S.G., A.H., F.D., and G.L. approved final version of manuscript.