RENAL AND METABOLIC EFFECTS OF CAFFEINE IN OBESE (fa/fa^cp), DIABETIC, HYPERTENSIVE ZSF1 RATS^*

Abstract

In Western society, the triad of hypertension, metabolic syndrome and obesity (which caries a high risk for renal disease) is increasing, as is the intake of caffeine. However, no information is available regarding the metabolic or renal consequences of caffeine consumption in this complex disease entity. The purpose of this study was to investigate the effects of chronic caffeine consumption on renal function and metabolic status in obese ZSF1 rats, an animal model of obesity, hypertension and the metabolic syndrome. Fifteen, 18-week-old male, obese ZSF1 rats were randomized to drink tap water (Cont, n = 8) or 0. 1% solution of caffeine (Caff, n = 7) for 8 weeks. Metabolic and renal function measurements were performed at baseline and after 4 and 8 weeks of treatment. Caffeine treatment significantly (p < 0.05) reduced body weight, food, and fluid consumption and improved insulin sensitivity (fasting insulin 129.6 + 8.1 vs 97.5 ± 3.6 μIU/mL; fed insulin 146.3 ± 8.5 vs 110.6 ± 3.4 μIU/mL; fasting glucose 138.7 ± 13.4 vs 145 ± 8.0 mg/dL; fed glucose 373 ± l9.4 vs 283.3 ± 19.6 mg/dL, Cont vs Caff, respectively). After 8 weeks of caffeine treatment, animals were less glycosuric as compared with control group. Area under glucose curves (AUC-glucose) in oral glucose tolerance test did not differ between the two groups (AUC- glucose: 592.5 ± 42.7 vs 589.5 ± 20.5 mg/dL × h, Cont vs Caff), whereas caffeine treatment significantly decreased AUC of insulin (AUC-insulin: 257.77 ± 12.9 vs 198.0 ± 5.9 μIU/mL × h, Cont vs. Caff, p<0.05). No differences were found with regard to plasma triglycerides and glycerol levels; however, caffeine significantly increased cholesterol levels after 4 and 8 weeks (2F-Anova, p<0.001). Moreover, caffeine significantly decreased creatinine clearance after 4 and 8 weeks (CrCl, Cont:3.5 ± 0.4, Caff: 2.2 ± 0.2 L/kg/day, p<0.05) and increased protein/CrCl ratio (Cont:323 ± 30, Caff: 527 ± 33 mg/L/day). Caffeine treatment for 8 weeks tended to increase plasma norepinephrine levels (p<0.06), but the two groups did not differ with regard to plasma renin activity, blood pressure, renal blood flow or and renal vascular resistance. The study indicates that caffeine improves insulin sensitivity but increases plasma cholesterol levels and impairs renal function in obesity with the metabolic syndrome and hypertension. Our results imply that the health consequences of chronic caffeine consumption may depend heavily on underlying pathophysiology process.

• Caffeine
• Metabolic syndrome
• Hypertension
• Chronic renal failure
• Rats

INTRODUCTION

Because caffeine is the most widely used pharmacologically active agent known, the health benefits and risk of caffeine consumption have received considerable attention (for review see 1). Importantly, the use of caffeine is not unequivocally associated with either benefits or risk in healthy humans. Little information, however, is available concerning the interaction of caffeine with complex disease states. In this regard, as populations in modern societies age and the use of caffeine rises, the impact (either positive or negative) of caffeine consumption in complex disease states associated with aging becomes increasingly important.

In aging populations, the prevalence of hypertension is high [[2]], and our recent preliminary studies in a lean animal model indicate that chronic caffeine consumption accelerates the development of nephropathy [[3]] in the setting of high blood pressure.

In modern societies, hypertension is frequently associated with the metabolic syndrome and obesity rather than normal body weight and this triad carries high risk for renal disease [4-6]. However, no information is available regarding the metabolic and renal effects of chronic caffeine consumption in the setting of obesity, hypertension and the metabolic syndrome.

Given the increasing prevalence of obesity and cardiovascular disease in aging modern populations, we decided to examine carefully the renal and metabolic consequences of chronic caffeine consumption in a genetically-obese animal model of hypertension and the metabolic syndrome. Our results demonstrate that although caffeine consumption improves insulin sensitivity, this positive effect is offset by the ability of caffeine to increase chronically plasma cholesterol levels and to induce nephropathy as characterized by markedly enhanced urinary protein excretion and a reduction in glomerular filtration rate. Our results imply that the health consequences of chronic caffeine consumption are importantly determined by underlying pathophysiology and that studies in normal humans do not exclude possible adverse effects of caffeine in complex disease states associated with aging in modern societies.

MATERIAL AND METHODS

1. Animals

Total of fifteen, 18 weeks old, male, obese (627 ± 10g) ZSF1 rats (Genetic Model Inc. Indianapolis, IN) were used in this study. These animals, in addition to being obese, are hypertensive, have the metabolic syndrome and express renal dysfunction [7-8]. Rats were housed in the University of Pittsburgh Medical Center animal care facility (temperature, 22°C; light cycle, 12 hours; relative humidity 55%). Animals were fed Pro Lab RMH 3000 rodent diet (PMI Nutrition Inc., St Louis, MO) and were given water ad libitum. Institutional guidelines for animal welfare were followed.

2. Metabolic Cages Study

Before 4 and 8 weeks into the treatment, rats were placed in metabolic cages for 48 hours. The first 24-hour period was considered an acclimatization period, whereas food and water intake and urine volume were measured during the second 24-hour period. Urine samples were analyzed for creatinine, protein, glucose, sodium and potassium concentrations. A volume of blood (1.5 mL) was drawn from the tail vein, and used for determination of plasma levels of creatinine, lipids, sodium, potassium and renin activity (PRA). Plasma and urine creatinine were measured with a creatinine analyzer (Creatinine Analyzer 2, Beckman Instruments, Inc, Fullerton, CA, USA) and a flame photometer (model IL943 flame photometer, Instrumentation Laboratory, Lexington, MA, USA) was used to measure sodium and potassium. Total protein concentration was measured in the urine samples by a spectrophotometric assay using bicinchoninic acid reagent (Pierce; Rockford, IL) and a modification of the method described by Lowry [[9]]. 24-Hour urinary sodium and potassium excretion, creatinine clearance and 24-hour urinary protein excretion were calculated. Plasma samples were analyzed in duplicates for glucose, triglycerides and cholesterol levels (Sigma Diagnostics, St Louis, MO). Insulin levels were measured in duplicate by a double antibody radioimmunoassay specific for rat insulin (Incstar Corp., Stillwater, MN) and PRA was measured by Angiotensin I [¹²⁵I] radioimmunoassay (NEN, Boston, MA)

3. Glucose Tolerance Test

After completing metabolic cage measurements (week 8), blood samples were taken for measurement of prandial glucose and insulin levels, before animals were fasted for 16 hours and glucose tolerance test was performed. Blood samples for measurement of fasted glucose and insulin levels were taken and animals were given 2 gr/kg/4mL glucose by oral gavage. Blood withdraws were repeated after 30, 60, and 120 minutes.

4. Acute Experiments

At the end of the 8-week treatment period, each rat was anesthetized with pentobarbital (45 mg/kg i.p.), and a short section of PE-240 polyethylene catheter was placed in the trachea to facilitate breathing. The left carotid artery was exposed and cannulated with PE-50 tubing for measurement of systolic (SBP), diastolic (DBP) and mean arterial blood pressures (MABP) and heart rate via a digital blood pressure analyzer (Micro-Med, Inc., Louisville, KY). Another PE-50 catheter was placed in the jugular vein for administration of supplemental anesthetic. A midline abdominal incision was made, and a PE-10 catheter was placed in the left ureter to facilitate collection of urine. A flow probe (Model IRB; Transonic Systems, Inc., Ithaca NY) was placed on the left renal artery for measurement of renal blood flow.

A one-hour stabilization period was permitted before data and samples were collected. Following the stabilization period, blood pressure and renal blood flow(RBF) were recorded for 30 minutes and 30-minute averages for SBP, DBP, MABP, heart rate, RBF and renal vascular resistance (RVR) were calculated. One milliliter of blood was taken, and analyzed for plasma norepinephrine levels by HPLC with electochemical detection, as described previously [[10]]. Animals were euthanized by anesthetic overdose. The right kidney was immediately removed and processed for histopathological analysis.

5. Renal Histopathology

The right kidney tissue sample stored in 10% formalin buffer was sectioned and was processed into paraffin blocks for light microscopy. Three histological sections (3 microns) were cut and stained with hematoxylin-eosin, periodic acid-Schiff (PAS) and methenamine silver-trichrome(MST). Kidney slices were examined by light microscopy and scored in a blinded fashion by one of the investigators (S.B.). A total of at least 150 glomeruli from each rat were studied and the percentage of glomeruli showing segmental (FSGS) and global (FGGS) glomerulosclerosis was counted. Other histopathologic features assessed semi-quantitatively included tubular atrophy (0–3⁺), interstitial inflammation (0–3⁺), interstitial fibrosis (0–3⁺), tubular dilation (0–4⁺), arterial medial hypertrophy (0–3⁺), and arteriolar sclerosis (0–3⁺).

6. Data Analysis

All data are presented as mean ± S.E.M. Statistical analyses were performed using the Number Cruncher Statistical System (Kaysville, Utah). Group comparisons for data from metabolic studies (repeated measurements) were performed by one- (1F) or two- way (2F) hierarchical analysis of variance (ANOVA) as appropriate, followed by Fisher's LSD test for post-hoc comparison. Comparison of data from acute experiments (single point data) was performed by student's t-test. The probability value of p<0.05 was considered statistically significant.

RESULTS

Adult obese ZSF1 rats are characterized by hyperphagia, polydipsia, polyuria, glycosuria (Table 1, baseline), insulin resistance (Figure 1), hypercholesterolemia, hypertrigyceridemia and, proteinuria in nephrotic range (Figures 1-2, and 4, week 0). The experimental groups did not differ with regard to metabolic status and renal excretory function parameters measured at baseline (Table 1 and Figures 1-4, baseline and week 0, respectively). From 18 to 26 week of age there was significant increase in the body weight in control group (p<0.001, time effects, 1F-Anova) and caffeine treatment significantly attenuated the age-related increase in body weight (p<0.001, 2F-Anova). The caffeine-induced reduction in body weight observed at weeks 4 and 8 was accompanied by a reduction in food and fluid consumption. Caffeine tended to decrease creatinine and glucose excretion at 4 weeks, and after 8 weeks of treatment caffeine treated animals excreted significantly less creatinine and glucose (p<0.05, Table 1).

Figure 1. Fasted and fed plasma levels of glucose and insulin and areas under the glucose and insulin curves in glucose tolerance test after 8 weeks of caffeine consumption in adult, obese ZSF1 rats.

Figure 2. Plasma lipids levels in adult, obese ZSF1 rats treated with caffeine for 8 weeks.

Figure 3. Plasma creatinine and creatinine clearance in adult, obese ZSF1 rats treated with caffeine for 8 weeks.

Figure 4. Urinary protein excretion in adult, obese ZSF1 rats treated with caffeine for 8 weeks.

Table 1. Effect of Caffeine on Metabolic and Renal Excretory Function Parameters in Adult ZDFxSHHF Rats

Parameter	Treatment	Baseline	4 Weeks	8 Weeks
Body weight	Control	625±15.3	658±12.2	670±14.8
(g)	Caffeine	629±5.3	604±4.8^*^a	614±4.2^*^a
Food intake	Control	67.5±2.9	61.0±2.0	56.1±2.3
(g/kg/day)	Caffeine	65.1±1.9	43.1±1.3^*^a	47.4±1.4^*^a
Fluid intake	Control	138.8±10.9	127.8±14.1	137.0±19.8
(mL/kg/day)	Caffeine	138.7±9.6	84.5±4.7 ^*^a	79.5±5.0 ^*^a
Urine volume	Control	105.4±9.7	82.8±12.6	109.6±15.1
(mL/kg/day)	Caffeine	108.5±7.1	63.8±6.1^*^a	72.9±3.3^*^a
Sodium excretion	Control	4.42±0.18	3.38±0.10	5.53±0.44
(mEq/kg/day)	Caffeine	4.44±0.24	3.71±0.12	5.61±0.29
Potassium excretion	Control	15.5±0.7	13.9±0.4	13.6±0.6
(mEq/kg/day)	Caffeine	15.3±0.8	12.4±0.3	12.3±0.5
Creatinine excretion	Control	36.5±1.2	28.6±0.3	42.3±3.6
(mg/kg/day)	Caffeine	34.8±0.6	26.9±0.5	31.8±0.6^a
Urinary protein	Control	910±25	910±27	1050±28
excretion (mg/kg/day)	Caffeine	890±17	1010±24	1150±25^*
Glucose excretion	Control	4.92±0.65	4.76±0.82	5.95±0.82
(g/kg/day)	Caffeine	5.32±0.50	4.56±0.69	4.05±0.27^a

^*−p<0.05 vs baseline

^a−p<0.05 vs control

Significantly lower fasting insulin levels and prandial glucose and insulin levels were found in caffeine treated animals (Figure 1). The groups did not differ with regard to glucose load response and had similar areas under the glucose concentration curves (AUC glucose 940 ± 50.2 and 932 ± 36.6 mg/dL × h, Control and Caffeine groups, Figure 1, top panel, right axes). However, significantly lower plasma concentrations of insulin in response to the glucose load were observed in the caffeine group (AUC insulin 400.9 ± 20.7 vs 307.4 ± 8.9 μIU/mL × h, Control vs. Caffeine, Figure 1, bottom panel, right axes). Caffeine had no effect on triglycerides and glycerol levels, yet significantly increased total cholesterol levels at 4 and 8 weeks of treatment (Figure 2).

From 18 to 26 weeks of age there was an age related increase in plasma creatinine and decrease in creatinine clearance (2F-Anova, time effects p<0.001). Caffeine consumption significantly increased plasma creatinine and reduced creatinine clearance (Figure 3). Caffeine treatment for 8 weeks increased urinary protein excretion expressed either as percent changes from baseline (Figure 4 top panel) or corrected for creatinine clearance (Figure 4 bottom panel).

Caffeine treatment for 8 weeks tended to increase plasma renin activity (Figure 5 left axis) and plasma norepinephrine levels (Figure 5 right axis); however, the and two groups did not differ with regard to blood pressure, heart rate, renal blood flow or renal vascular resistance (Table 2).

Figure 5. Plasma renin activity (PRA) and norepinephrine levels in adult, obese, ZSF1 rats treated with caffeine for 8 weeks.

Table 2. Blood Pressure and Renal Hemodynamics in Adult ZDFxSHHF Rats Treated with Caffeine for 8 Weeks

Parameter	Control (n = 8)	Caffeine (n = 7)
Heart rate (b.p.m.)	334.6±5.7	356.1±10.4
Systolic blood pressure (mmHg)	181.5±5.0	187.8±5.2
Diastolic blood pressure (mmHg)	140.7±3.6	143.5±2.3
Mean blood pressure (mm/Hg)	160.7±4.1	166.0±3.4
Pulse pressure (mmHg)	40.8±3.2	44.4±3.9
Renal blood flow (mL/min/g kid)	6.37±0.65	7.48±0.56
Renal plasma flow (mL/min/g kid)	3.44±0.34	4.10±0.26
Renal vascular resistance (mL/min/g kid)	5.88±1.0	4.45±0.36

Renal histology revealed focal segmental glomeruloselerosis (FSGS), tubular dilatation and hyaline casts (Table 3). However, no differences were found between caffeine-treated animals and control group in this regard.

Table 3. Renal Histopathology in Adult ZDFxSHHF Rats Treated with Caffeine for 8 Weeks

	Control (n = 8)	Caffeine (n = 7)
Total kidney weight (g)	4.00±0.25	4.48±0.22
Kidney/B.W. ratio (mg/g)	6.04±0.29	7.41±0.34^*
Mesangial. proliferation (0–3^+)	1.38±0.21	1.36± 0.19
FSGS+FGGS (%)	9.57±2.16	7.98±1.49
Tubular atrophy (0–3^+)	1.44±0.15	1.64±0.18
Tubular dilation (0–4^+)	2.16±0.22	2.46±0.27
Cast (0–3^+)	1.44±0.15	1.64±0.17
Interst. inflammation (0–3^+)	0.44±0.22	0.21±0.15
Interstitial fibrosis (0–3^+)	0	0
Medial hypertrophy (0–3^+)	1.06±0.11	1.00±0.22
Arteriolar sclerosis (0-3^+)	0.62±0.12	0.57±0.17

^*p < 0.05

DISCUSSION

The main finding of this study is the observation that caffeine consumption for 8 weeks was associated with an accelerated decline in renal function in obese ZSF1 rats, an animal model that in addition to the metabolic syndrome and hypertension is obese. The augmented reduction in creatinine clearance in caffeine treated rats was accompanied by increases in urinary protein excretion; however, caffeine did not potentiate renal histopatholocical changes. The absence of histopathological changes may be due either to the fact that adult, obese ZSF1 rats at baseline have significant renal histopathology [7-8], or that the treatment duration for caffeine must be longer to produce significant renal structural changes. Also, the observed adverse renal effects of caffeine occurred in the face of improved insulin sensitivity (reduced fasting glucose and insulin levels, reduced glycosuria, reduced body weight and food consumption) which would be expected to ameliorate renal hitopathological changes.

The data from this study are in congruence with our previous findings. We have shown that caffeine consumption for 20 weeks in lean SHHF/Mcc-facp rats (one of the parents of obese ZSF1 rats) accelerates the decline in renal function and augments urinary protein excretion [[3]]. Furthermore, in a model of experimental nephropathy (puromycine aminonucloisede(PAN)-induced nephropathy), caffeine consumption for 24 weeks augmented proteinuria, reduced glomerular filtration and induced severe glomerular and tubulointerstitial damage [[11]].

Previously we have shown [[10]], [[12]] that caffeine via a dual mechanism (i.e. by blocking intrarenal adenosine type 1 receptors and by increasing central and renal sympathetic activity) increases renin release and that short-term caffeine consumption increases renal renin secretion in one of the parents of ZSF1 rats, i.e., SHHF/Mcc-fa^cp rats [[13]]. It may be expected that chronic activation of the renin-angiotensin system would have adverse effects on renal function in ZSF1 rats. Accordingly, we measured both PRA and plasma norepinephrine levels. Although in the present study caffeine did not increase PRA (measured in systemic circulation), an increase in the activity of the intrarenal RAS cannot be ruled out.

In this study, caffeine tended to increase plasma norepinephrine levels and this affect almost reached statistical significance (p < 0.06). This findings is consistent with our previous demonstration that caffeine consumption for 20 weeks increases renal vein norepinephrine levels in SHHF/Mccfa^cp rats [[3]] and that acute administration of caffeine augments renal sympathetic activity [[10]]. Recent study suggests that activation of A_2A receptors in the nucleus tractus solitarius inhibits renal sympathetic nerve activity [[14]] and it may be expected that blockade of these receptors by caffeine would increase renal sympathetic nerve activity. Data from the literature together with our data [[10]], [[12]], [15-16] indicates that caffeine increases total as well as renal sympathetic activity. A large body of evidence [17-18] indicates that there is increased sympathetic activity in experimental and human chronic renal failure. Although the role of renal sympathetic activity in progression of renal disease in humans has not yet been elucidated, it has been shown that renal denervation attenuates development of hypertension, renal damage and renal production of free radicals in experimental nephropathy [19-21]. Furthermore, recently it has been demonstrated that inhibition of sympathetic activity by imidazoline receptor agonists moxonidine (given in subantihypertensive doses) attenuates progression of renal failure in 5/6 nephropathy in rats [[22]]. Whether augmented sympathetic activity by caffeine contributes to the caffeine-induced renal dysfunction need to be examined further.

Caffeine consumption for 8 weeks had mixed effects on the metabolic status in adult obese ZSF1 rats. Specifically, we observed improved insulin sensitivity and increased plasma cholesterol levels in caffeine treated animals. There is compelling evidence for significant tissue specific effects of adenosine on insulin sensitivity. Adenosine increases the sensitivity of the rate of glucose utilization and lipolysis to insulin in adipose tissue [[23]] but decreases the sensitivity of glucose utilization to insulin in skeletal muscle [[24]]. Since muscle is considered to be the most important tissue for disposal of glucose in response to insulin and because the inhibitory effects of adenosine on insulin stimulated glucose utilization in skeletal muscles are mediated by A₁ receptors it may be expected that blockade of A₁ receptors would improve insulin sensitivity. Therefore it is not surprising that caffeine, a non-selective adenosine receptor antagonist that blocks both A₁ and A₂ adenosine receptors, improved insulin sensitivity in obese ZSF1 rats. This is also in accordance with previous studies where adenosine A₁ receptor antagonists have been shown to increase sensitivity to insulin in skeletal muscle from normal rats [[25]], genetically obese Zucker rats and from rats fed a high sucrose diet [26-27]. Moreover, A₁ receptor antagonists improve glucose tolerance in obese Zucker rats in vivo [28-29].

Improved insulin sensitivity observed in animals that were consuming caffeine may be in part due to increased sympathetic activity and elevated plasma catecholamines levels by caffeine. In contrast to acute activation of sympathetic nervous system where decreased sensitivity of glucose utilization to insulin is observed [[30]] in conditions that are characterized by chronic elevation of circulating catecholamines (i.e., during cold exposure or training) there is increased sensitivity of glucose metabolism to insulin [31-33].

Another important finding of this study is that caffeine consumption increased cholesterol levels. Although the consumption of caffeinated beverages has been associated with elevated serum cholesterol in humans, this relationship is inconsistent and controversial [34-35]. In rats there is little, data regarding the effects of caffeine consumption on lipids and existing data are contradictory. Some early studies suggest very modest (within physiological range) increases in cholesterol levels during the first seven days, but not after 25 days of caffeine consumption [[35]]. When rats were given a high cholesterol diet there was a marked increase in plasma cholesterol levels in animals that were consuming caffeine for 21 days. However, in rats on a high-cholesterol diet, caffeine consumption for 4 months had no effect on cholesterol concentrations in plasma and aorta and total body cholesterol content, as well as on atherosclerotic changes in the aorta and heart [[35]]. In a more recent study in rats [[36]], consumption of boiled coffee failed to increase serum cholesterol levels. Therefore caffeine's effects on plasma cholesterol levels observed in the present study suggests that the effects of caffeine on plasma cholesterol may be more severe in animals with the metabolic syndrome. There is a accumulating evidence that elevated lipids level from any primary cause exacerbate renal damage [37-40] and antilipemic drugs ameliorate renal function and structure in different models of experimental nephropathy [[41]], [[43]].

In summary, chronic caffeine consumption induces real and significant adverse effects on lipid metabolism and renal function in the setting of obesity, hypertension and the metabolic syndrome. These data provide a strong rational for examining the effects of caffeine on lipid metabolism and renal function/structure in aging individuals with obesity, hypertension and the metabolic syndrome.