Divergent Effects of Unilateral and Subtotal Nephrectomy on Insulin Sensitivity in Rats

Abstract

Objectives. Insulin resistance is associated with chronic renal failure, which may amplify its cardiovascular pathologic manifestations. We previously showed the presence of insulin resistance in mild renal insufficiency due to chronic glomerulonephritis. These observations may be explained by a decrease in insulin sensitivity due to renal dysfunction. The aim of the present study was to evaluate the effects of unilateral and subtotal nephrectomy on insulin sensitivity. Methods. Unilateral heminephrectomy and five-sixths nephrectomy (5/6Nx) were performed in male Sprague–Dawley rats, measuring steady-state plasma glucose (SSPG) during the insulin suppression test. Results. The glomerular filtration rate (GFR) decreased significantly from sham-operated rats, to unilateral heminephrectomy group, to 5/6Nx rats. SSPG was unexpectedly lower in the unilateral heminephrectomy group than in controls, suggesting that unilateral heminephrectomy increased the sensitivity to insulin despite a mild decrease in renal function. However, when the analysis was limited to the 5/6Nx group, SSPG was inversely correlated with GFR (r = − 0.65, p < 0.05). When renal failure caused by 5/6Nx had reached end stage, the rats became insulin resistant, despite a profound reduction in renal mass. Conclusion. It is suggested that unilateral heminephrectomy reduced GFR and increased the sensitivity to insulin. When rats became uremic, insulin sensitivity decreased, even, in 5/6Nx.

• insulin resistance
• steady-state plasma glucose
• nephrectomy
• end-stage renal disease
• gluconeogenesis

Introduction

Insulin resistance is associated with chronic renal failure (CRF).[1-5] Although insulin resistance accompanied by end-stage renal disease (ESRD) can be mitigated by hemodialysis,[4] it is an independent predictor of fatal cardiovascular events in nondiabetic patients treated with long-term hemodialysis.[6] Metabolic changes taking place during the evolution of CRF may play an important role in the progression of nephrosclerosis and survival of patients with ESRD.[7] Insulin resistance and hyperinsulinemia are not limited to patients with ESRD and may be present very early in the course of renal disease, when the glomerular filtration rate (GFR) is still within the normal range.[8] We previously reported a correlation between GFR and insulin sensitivity in nonobese, nondiabetic patients in the early phase of chronic glomerulonephritis (CGN) by the euglycemic hyperinsulinemic clamp technique.[9] In the general population, the sum of the fasting and 2 h post-loading insulin concentrations during a 75 g oral glucose tolerance test was correlated with the reciprocal of serum creatinine.[10] However, Stenvinkel et al. reported insulin resistance in patients with nephrotic syndromes and in renal transplant recipients with slightly impaired renal function, but not in patients with nonnephrotic IgA nephropathy with similar renal function.[11] Therefore, it remains unclear whether mild renal insufficiency interferes with insulin sensitivity early in the course of renal disease, regardless of its etiology.

In ESRD, hepatic glucose production is not increased and is usually suppressed in response to insulin. The site of insulin resistance is in peripheral tissues, i.e., skeletal muscle and brown fat. The association between insulin resistance and ESRD probably reflects mechanisms such as coexisting metabolic acidosis and accumulation of glucose utilization inhibitors.[12&13] Insulin resistance associated with uremia may be primarily attributable to altered postreceptor events caused by circulating factors.[14] The role of the kidney as a glucose-producing organ on insulin resistance remains to be determined, especially early in the course of chronic renal disease, although the kidney is known as the second most important glucose-producing organ. Reaven et al. reported that rats made uremic by bilateral nephrectomy developed a significant decrease in serum glucose level, whereas rats made similarly uremic by bilateral ureterotomy demonstrated a significant elevation of both serum glucose and insulin levels.[15] From observations made during the anhepatic phase of liver transplantation, the kidney is the main source of extrahepatic glucose production, at a rate close to that measured during the postabsorptive phase in healthy subjects.[16] In an earlier study, unilateral nephrectomy had no effect on the glycemic response to an oral glucose tolerance test in male Wistar Kyoto rats. In contrast to unilateral heminephrectomized rats, rats rendered uremic by the removal of 90% of their renal tissue had an abnormal sensitivity to insulin demonstrable by the euglycemic hyperinsulinemic clamp technique.[17] The focus of this study is from what degree of renal dysfunction caused by renal mass reduction insulin resistance emerges.

Materials and Methods

Animal Preparations

The experiments were performed in male Sprague–Dawley (SD) rats obtained from Charles-River Laboratories and bred at the animal facility of Keio University School of Medicine. The rats were fed standard chow (Clea Japan Inc., Tokyo, Japan) containing 0.39% NaCl and 24.8% protein and were given free access to water. They were divided into a sham-operated group, a unilateral nephrectomy group, and a five-sixths nephrectomy (5/6Nx) group.

Nephrectomy

At 13 weeks of age, the animals were anesthetized with ether and underwent a sham operation (n =11), unilateral heminephrectomy (n =13), or 5/6Nx (n =11). The 5/6Nx was achieved by first removing the upper and lower poles of the right kidney, then performing a left nephrectomy 1 week later.

Measurements of Glomerular Filtration Rate

The GFR was measured by the creatinine clearance method. At 15 weeks of age, each rat was anesthetized with an intraperitoneal injection of sodium pentobarbital. The body temperature was maintained at 37°C, and the trachea was cannulated to facilitate respiration. The right common carotid artery and right internal jugular vein were each exposed and cannulated with a PE50 catheter to collect blood samples and infuse solutions. A PE90 catheter was inserted into the dome of the bladder via a midline suprapubic incision and exteriorized through the abdominal muscle layers and skin for the collection of urine. To measure the creatinine clearance, isotonic saline was infused intravenously at a rate of 0.03 mL/min for a 45 min equilibration phase before collection of a 30 min urine sample to determine the baseline excretory parameters. A 0.4 ml arterial blood sample was obtained at the end of each experiment for the measurement of serum creatinine. Clearance values were corrected for body weight.

Insulin Suppression Test

After measurement of the GFR, each rat received a continuous infusion of epinephrine, 0.08 µg/kg/min, propranolol, 1.7 µg/kg/min, glucose, 8.0 mg/kg/min, and 2.660 mU/kg/min of pork insulin. This technique to measure the action of insulin in vivo was described previously.[18] These amounts of epinephrine and propranolol suppress the endogenous release of insulin, allowing an assessment of insulin-stimulated glucose utilization under conditions in which circulating exogenous insulin concentrations in the sham-operated and unilateral heminephrectomized rats are comparable. The infusions were continued for 3 h, and blood samples were collected at 15 min intervals during the last 60 min of the study to measure plasma glucose and insulin concentrations. Steady-state plasma glucose (SSPG) and insulin (SSPI) concentrations were the average values of the five measurements made during the last 60 min of infusion in each experiment. This experimental protocol was approved by the institutional animal care and usage committee, and the survival rate of 5/6Nx was 85% after the skill of the experimental procedure became sophisticated.

Glucose and Insulin Measurements

Plasma glucose concentrations were measured by the glucose dehydrogenase method with a Hitachi 7150 automated glucose analyzer (Hitachi, Tokyo, Japan).[19] In preliminary studies, insulin concentrations were measured by radioimmunoassay using kits for human and rat insulin. The immunoreactive insulin levels were similar between the two methods, and the data using the kit for rat insulin were analyzed in this study.

Data Analysis

Values are expressed as the means ± SEM. Statistical comparisons were made by analysis of variance followed by Scheffé's F-test to examine between-group differences. Pearson's correlation was used to assess relationships between variables. A p value < 0.05 was considered statistically significant.

Results

The mean body weight, serum creatinine, and creatinine clearance in the three experimental groups are summarized in Table 1. Body weight was significantly lower in the 5/6Nx rats than in the sham-operated rats (p < 0.01). The plasma creatinine concentration was significantly higher in the 5/6Nx rats than in the sham-operated or the heminephrectomized rats (p < 0.01). Similarly, the creatinine clearance decreased significantly from the sham-operated, to the unilateral heminephrectomy group, to the 5/6Nx group (p < 0.001; Table 1). There were no significant differences among the groups in plasma glucose or insulin concentrations before the insulin suppression test.

Table 1 Mean body weight, serum creatinine, and creatinine clearance in the three experimental groups

Group	Sham-operated (n = 11)	Unilateral heminephrectomy (n =13)	5/6Nx (n =11)
Body weight (g)	328 ± 12	296 ± 12	267 ± 11
Serum creatinine (mg/dL)	0.29 ± 0.02	0.45 ± 0.04	0.85 ± 0.13,
Ccr (mL/min/100 g)	1.31 ± 0.08	0.75 ± 0.06	0.36 ± 0.05,

Values are expressed as mean ± S.E.M.

Ccr, creatinine clearance.

*p < 0.01 versus sham-operated rats.

^†p < 0.01 versus unilateral heminephrectomy.

SSPG was unexpectedly and significantly lower in the unilateral heminephrectomy than in the sham-operated group (p < 0.02), while SSPI was similar (16.6 ± 1.2 ng/mL versus 17.8 ± 0.9 ng/mL). As in the presence of similar SSPI values, SSPG is a marker of insulin resistance, these results indicate that unilateral heminephrectomy significantly increased insulin sensitivity.

SSPG in the sham-operated rats (199 ± 11 mg/dL) was significantly different than in the 5/6Nx rats (133 ± 15 mg/dL, p < 0.01, Figure 1). However, the insulin suppression test resulted in a markedly higher SSPI concentration in the 5/6Nx rats (21.3 ± 0.8 ng/mL) than in the sham-operated rats (16.6 ± 4.1 ng/mL, p < 0.01) and unilateral heminephrectomized rats (17.7 ± 3.1 ng/mL, p < 0.05). The SSPG concentration as a marker of insulin resistance in the 5/6Nx rats versus the other groups could not be directly compared because of the markedly higher SSPI in the 5/6Nx group than in both other groups.

Figure 1 Steady-state plasma glucose (SSPG) (A) and insulin (SSPI) (B) concentrations in each group during the infusion of epinephrine, propranolol, glucose, and insulin. (Bars represent mean ± SEM. ns = not significant. *p < 0.05; **p < 0.01.)

No significant correlation between GFR and SSPG was observed when the unilateral heminephrectomy or sham-operated groups were analyzed separately. However, when the analysis was limited to the 5/6Nx group, SSPG was inversely correlated with GFR (r = − 0.65, p < 0.05, Figure 2). Moreover, no significant correlation was observed between GFR and SSPI when the sham-operated or unilateral heminephrectomy rats were analyzed separately. SSPI was inversely and strongly correlated with GFR in the 5/6Nx group (r = − 0.64, p < 0.05). Therefore, some 5/6Nx rats that had reached end-stage renal failure became insulin resistant despite a profound reduction in renal mass.

Figure 2 Relationship between steady-state plasma glucose concentration (SSPG) and creatinine clearance. [Open circles, closed triangles, and open squares indicate sham-operated rats, unilateral heminephrectomy, and 5/6Nx, respectively. When the relationship was limited to the 5/6Nx rats, an inverse correlation was observed between SSPG and creatinine clearance (r = − 0.65, p < 0.05, dashed line).]

Discussion

Unilateral Heminephrectomy Increases the Sensitivity to Insulin

This study demonstrated unambiguously that mild renal dysfunction caused by unilateral heminephrectomy was associated with a higher sensitivity to insulin than in control rats. However, severe renal dysfunction by 5/6Nx was associated with insulin resistance. The pathogenic mechanisms between a decrease in renal mass and insulin sensitivity remain unclear. Little is known regarding renal glucose utilization and its role in the regulation of plasma glucose. Studies using a sensitive isotopic dilution technique have confirmed that the kidneys of intact rats contribute approximately 25% of gluconeogenesis in the fed state.[20] In addition, while renal gluconeogenesis is cooperatively regulated by hepatic glucose production, several important organ-specific differences have been observed. For example, phosphoenolpyruvate carboxykinase is an enzyme unique to gluconeogenesis in both liver and kidney. It is expressed in a highly organ-specific manner and is suppressed by insulin in the liver but not the kidney.[21] This suggests that renal gluconeogenesis is not sufficiently suppressed during the steady-state phase of the insulin suppression test, and it is likely that a decrease in renal gluconeogenesis caused by unilateral heminephrectomy increases the sensitivity to insulin. Because body weight in unilateral heminephrectomy was slightly decreased (although not significant) compared with the sham-operated rats, it cannot be ruled out that fasting and catabolism might play a role on enhancing insulin sensitivity in unilateral heminephrectomy.

Insulin resistance invariably accompanies CRF,[1-5] and the main site of resistance to insulin-mediated glucose uptake in uremia is skeletal muscle. We previously showed a correlation between insulin sensitivity and GFR in nonobese, nondiabetic patients by the euglycemic hyperinsulinemic clamp technique in the early phase of CGN.[9] Although it is assumed that the output of hepatic glucose is completely suppressed during euglycemic hyperinsulinemia,[22] we hypothesize that the peripheral glucose utilization stimulated by insulin is decreased even in the presence of mild renal insufficiency. We further hypothesized when launching this study that unilateral heminephrectomy and 5/6Nx would both be associated with a decrease in insulin sensitivity, as is observed in patients with CGN and mild renal insufficiency and in patients with ESRD. Unexpectedly, however, unilateral heminephrectomy increased the sensitivity to insulin. There was a clear difference between the unilateral heminephrectomy rats and patients with early CGN in the effects on insulin sensitivity, because renal gluconeogenesis in unilateral heminephrectomy should have reduced by half. In an earlier study, unilateral heminephrectomy had no effect on the glycemic response to an oral glucose tolerance test in male Wistar Kyoto rats. However, the diagnostic value of the oral glucose tolerance test in peripheral insulin sensitivity is limited, because an impaired pancreatic β-cell response cannot be distinguished from peripheral resistance to the action of insulin.[23] In contrast to unilateral heminephrectomized rats, rats rendered uremic by the removal of 90% of their renal tissue had an abnormal sensitivity to insulin demonstrable by the euglycemic hyperinsulinemic clamp technique.[17] In that model, the ability of insulin to enhance both glycolysis and glucose oxidation by the soleus muscle was significantly reduced. Metabolic clearance rates of insulin and glucagon also influence the glucose metabolism. An alternate explanation is that its anatomic remnant kidney plays an important role in the lower insulin sensitivity of patients with CGN. Tubulo-interstitial injury and hyperfunction of remnant tubules may influence renal glucose metabolism.[24] Other pathological mechanisms, such as activation of the renin–angiotensin system, may also contribute to the association between insulin resistance and the very early phase of CGN.

If we attempt to compare the SSPG values of the unilateral heminephrectomized and 5/6Nx rats at a similar GFR level, we can select five unilateral heminephrectomized rats with the lowest GFR and five 5/6Nx rats with the highest GFR to match their GFR levels to approximately 0.5 mL/min/100 g. The SSPI values are similar between them (20 ng/mL, not significant). However, the SSPG values in unilateral heminephrectomized rats with the lowest GFR are apparently and significantly higher than those of 5/6Nx rats with the highest GFR [(177 ± 30 versus 111 ± 19 mg/dL) mg/dL, p < 0.01] (Figure 2). Therefore, the larger the renal mass, the more insulin resistant, even at the substantially decreased GFR level.

Correlation Between GFR and Metabolic Parameters in 5/6Nx

In the 5/6Nx group, SSPG was inversely correlated with GFR. When GFR had fallen to ≤ 0.3 mL/min/100 g, corresponding to approximately 20% of normal, the effect of renal failure on insulin resistance from the accumulation of glucose utilization inhibitors became more prominent than the insulin-sensitizing effect of decreased gluconeogenesis by renal mass reduction. In some of the 5/6Nx rats that revealed ESRD, prominent insulin resistance was detected, probably because the effects of renal mass reduction on glucose metabolism became negligible compared with the effects of renal failure caused by extensive nephrectomy.

Although insulin resistance regularly accompanies CRF, occasional episodes of spontaneous hypoglycemia may also occur. Spontaneous, uremic hypoglycemia has been attributed to a diminished renal gluconeogenesis, impaired renal insulin degradation and clearance, poor nutrition, deficiency of precursors of gluconeogenesis, and, in a few cases, deficiency of counter-regulatory hormones such as catecholamines and glucagon.[25] Growth impairment in the postsurgical catabolic state has also been described in 5/6Nx rats.[26] In the present study, the body weight of the 5/6Nx was significantly lower than that of the sham-operated rats (Table 1), and poor nutrition may have contributed to the low SSPG concentrations. There is a criticism that stress and changes in body weight modulate insulin sensitivity on 5/6Nx. However, in 5/6Nx rats, SSPG and SSPI were inversely and more strongly correlated with GFR.

Insulin resistance and compensatory hyperinsulinemia cause glomerular hypertension and hyperfiltration, which may predispose to progressive glomerulosclerosis and renal insufficiency. Nosratola et al. reported that CRF causes a marked downregulation of tissue lipoprotein lipase expression, which may contribute to dyslipidemia and altered energy metabolism in uremia caused by 5/6Nx.[27] However, in this study, insulin resistance was observed only in the 5/6Nx rats with severe CRF, because the sensitizing effects on glucose metabolism of less gluconeogenesis by renal mass reduction and poor nutrition were greater, until ESRD exerted its own adverse effects on glucose metabolism.

Methodological Considerations

Our study suffered from methodological limitations with respect to the measurements of GFR and insulin sensitivity. Neither creatinine clearance nor the insulin suppression test are gold standards to examine renal function or insulin sensitivity. While we would have favored more precise methods, we could not apply them because of the technical complexity of simultaneous measurements of GFR and insulin sensitivity. Though the euglycemic hyperinsulinemic clamp technique would, in theory, have been ideal, it is technically difficult to achieve a stable glucose infusion rate in small animals because of the relatively larger amount of blood extraction and transfusion than in humans. Water and electrolytes balance may, therefore, become unstable during the euglycemic hyperinsulinemic clamp technique, markedly influencing GFR. The relationship between GFR measured by inulin clearance and amount of renal mass reduction and time course of changes in GFR after renal mass reduction has been reported by Fleck et al.[28] Our results are completely concordant with their data. Since the period from nephrectomy to the insulin suppression test was relatively short, stress and changes in body weight might modulate insulin sensitivity. However, we could not regulate renal function to a certain level, because compensatory glomerular hyperfiltration will occur after 2 weeks in the remnant kidney. This study first demonstrated the complex relationship between GFR and insulin sensitivity in nephrectomized rats by obtaining simultaneous measurements of them.

In conclusion, mild renal insufficiency caused by unilateral heminephrectomy increased the sensitivity to insulin. Insulin resistance emerges only when renal function is severely impaired by subtotal nephrectomy. Therefore, paradoxically in patients with mild renal dysfunction, the residual renal mass may play a substantial role in glucose metabolism.