Renal mass reduction increases the response to exogenous insulin independent of acid-base status or plasma insulin levels in rats

Abstract

Impairments in insulin sensitivity can occur in patients with chronic kidney disease (CKD). Correction of metabolic acidosis has been associated with improved insulin sensitivity in CKD, suggesting that metabolic acidosis may directly promote insulin resistance. Despite this, the effect of acid or alkali loading on insulin sensitivity in a rodent model of CKD (remnant kidney) has not been directly investigated. Such studies could better define the relationship between blood pH and insulin sensitivity. We hypothesized that in remnant kidney rats, acid or alkali loading would promote loss of pH homeostasis and consequently decrease insulin sensitivity. To test this hypothesis, we determined the impact of alkali (2 wk) or acid (5-7 days) loading on plasma electrolytes, acid-base balance, and insulin sensitivity in either sham control rats, 2/3 nephrectomized rats, or 5/6 nephrectomized rats. Rats with 5/6 nephrectomy had the greatest response to insulin followed by rats with 2/3 nephrectomy and sham control rats. We found that treatment with 0.1 M sodium bicarbonate solution in drinking water had no effect on insulin sensitivity. Acid loading with 0.1 M ammonium chloride resulted in significant reductions in pH and plasma bicarbonate. However, acidosis did not significantly impair insulin sensitivity. Similar effects were observed in Zucker obese rats with 5/6 nephrectomy. The effect of renal mass reduction on insulin sensitivity could not be explained by reduced insulin clearance or increased plasma insulin levels. We found that renal mass reduction alone increases sensitivity to exogenous insulin in rats and that this is not acutely reversed by the development of acidosis.NEW & NOTEWORTHY Impairments in insulin sensitivity can occur in patients with chronic kidney disease, and previous work has suggested that metabolic acidosis may be the underlying cause. Our study investigated the effect of acid or alkali loading on insulin sensitivity in a rodent model of chronic kidney disease. We found that renal mass reduction increases the blood glucose response to insulin and that this is not acutely reversed by the development of acidosis.

Keywords: Zucker obese rat; blood glucose; chronic kidney disease; insulin resistance; metabolic acidosis.

Graphical abstract

Figure 1.

Relationship between creatinine clearance (CrCl) and the change in pH (ΔpH) and ${\text{HCO}}_3^{-}$ (Δ${\text{HCO}}_3^{-}$) following alkali and acid loading in remnant kidney rats. A: relationship between CrCl and ΔpH following 2 wk of treatment with 0.1 M NaHCO₃ in remnant kidney rats. y-Axis, ΔpH following treatment with 0.1 M NaHCO₃; x-axis, CrCl (mL/min). Individual values are shown. n = 11 animals. B: relationship between CrCl and Δ${\text{HCO}}_3^{-}$ (mmol/L) following 2 wk of treatment with 0.1 M NaHCO₃ in remnant kidney rats. y-Axis, Δ${\text{HCO}}_3^{-}$ following treatment with 0.1 M NaHCO₃; x-axis, CrCl (mL/min). Individual values are shown. n = 11 animals. C: relationship between CrCl and ΔpH following 5–7 days of treatment with 0.1 M NH₄Cl in remnant kidney rats. y-Axis, ΔpH following treatment with 0.1 M NH₄Cl; x-axis, CrCl (mL/min). Individual values are shown. n = 19 animals. D: relationship between CrCl and Δ${\text{HCO}}_3^{-}$ (mmol/L) following 5–7 days of treatment with 0.1 M NH₄Cl in remnant kidney rats. y-Axis, Δ${\text{HCO}}_3^{-}$ following treatment with 0.1 M NH₄Cl; x-axis, CrCl (mL/min). Individual values are shown. n = 19 animals. P values are the results of Deming (model II) linear regression. r indicates Pearson’s correlation. *P < 0.05 was considered significant.

Figure 2.

Insulin sensitivity and release following treatment with either NaHCO₃ or NaCl. A: blood glucose response to insulin (0.75 U/kg) following 2 wk of either 0.1 M NaHCO₃ or 0.1 M NaCl. y-Axis, blood glucose (mg/dL); x-axis, time after insulin injection (min). Data were merged from the appropriate treatment phase and are means ± SE; n = 11 animals/group. P values are the results of two-way ANOVA. *P < 0.05 was considered significant. B: inverse area under the curve (AUC) analysis for the blood glucose response to insulin between NaHCO₃ or NaCl treatment. y-Axis, AUC; x-axis, treatment groups. Data were merged from the appropriate treatment phase and are means ± SE; n = 11 animals/group. P values are the results of an unpaired Student’s t test. *P < 0.05 was considered significant. C: plasma insulin levels during the insulin tolerance test following NaHCO₃ or NaCl treatment. y-Axis, plasma insulin (ng/mL); x-axis, time after insulin injection (min). Data were merged from the appropriate treatment phase and are means ± SE; n = 11 animals per group. P values are the results of two-way ANOVA. *P < 0.05 was considered significant. D: blood glucose response to glucose (1 g/kg) following 2 wk of either 0.1 M NaHCO₃ or 0.1 M NaCl. y-Axis, blood glucose (mg/dL); x-axis, time after glucose gavage (min). Data were merged from the appropriate treatment phase and are means ± SE; n = 11 animals/group. P values are the results of two-way ANOVA. *P < 0.05 was considered significant. E: AUC analysis for the blood glucose response to insulin between NaHCO₃ or NaCl treatment. y-Axis, AUC; x-axis, treatment groups. Data were merged from the appropriate treatment phase and are means ± SE; n = 11 animals/group. P values are the results of an unpaired Student’s t test. *P < 0.05 was considered significant. F: plasma insulin levels during the glucose tolerance test following NaHCO₃ or NaCl treatment. y-Axis, plasma insulin (ng/mL); x-axis, time after glucose gavage (min). Data were merged from the appropriate treatment phase and are means ± SE; n = 11 animals/group. P values are the results of two-way ANOVA. *P < 0.05 was considered significant.

Figure 3.

Insulin sensitivity and release following treatment with NH₄Cl. A: the blood glucose response to insulin (0.75 U/kg) at baseline and following 5−7 days of 0.1 M NH₄Cl. y-Axis, blood glucose (mg/dL); x-axis, time after insulin injection (min). Data are means ± SE; n = 15 animals. P values are the results of two-way ANOVA. *P < 0.05 was considered significant. B: inverse area under the curve (AUC) analysis for the blood glucose response to insulin. y-Axis, AUC; x-axis, testing time points (baseline and following NH₄Cl treatment). Data are means ± SE; n = 15 animals. P values are the results of an unpaired Student’s t test. *P < 0.05 was considered significant. C: plasma insulin levels during insulin tolerance tests performed at baseline and following 5–7 days of 0.1 M NH₄Cl treatment. Data are means ± SE; n = 15 animals. P values are the results of two-way ANOVA. *P < 0.05 was considered significant. D: the blood glucose response to glucose (1 g/kg) at baseline and following 5–7 days of 0.1 M NH₄Cl. y-Axis, blood glucose (mg/dL); x-axis, time after glucose gavage (min). Data are means ± SE; n = 5 animals. P values are the results of two-way ANOVA. *P < 0.05 was considered significant. E: AUC analysis for the blood glucose response to glucose. y-Axis, AUC; x-axis, testing time points (baseline and following NH₄Cl treatment). Data are means ± SE; n = 5 animals. P values are the results of an unpaired Student’s t test. *P < 0.05 was considered significant. F: plasma insulin levels during glucose tolerance tests performed at baseline and following 5−7 days of 0.1 M NH₄Cl treatment. Data are means ± SE; n = 5 animals. P values are the results of two-way ANOVA. *P < 0.05 was considered significant.

Figure 4.

Impact of renal mass reduction on insulin sensitivity prior to and following treatment with NH₄Cl. A: the blood glucose response to insulin (0.75 U/kg) at baseline in sham rats and rats with 2/3 and 5/6 Nx nephrectomy (Nx). y-Axis, blood glucose (mg/dL); x-axis, time after insulin injection (min). Data are means ± SE; n = 6 for sham, n = 7 for 2/3 Nx, and n = 8 for 5/6 Nx. 2/3 Nx rats in this study had a percent renal mass reduction of 60% < x < 70%; 5/6 Nx rats in this study had a percent renal mass reduction of 70% ≤ x < 85%. P values are the results of two-way ANOVA. *P < 0.05 was considered significant. B: inverse area under the curve (AUC) analysis for the blood glucose response to insulin in Sham, 2/3 Nx, and 5/6 Nx rats at baseline. y-Axis, AUC; x-axis, level of Nx (sham, 2/3 Nx, and 5/6 Nx). Data are means ± SE; n = 6 for sham, n = 7 for 2/3 Nx, and n = 8 for 5/6 Nx. P values are the results of one-way ANOVA. *P < 0.05 was considered significant. C: plasma insulin levels during a baseline insulin tolerance test in sham, 2/3 Nx, and 5/6 Nx rats. y-Axis, plasma insulin (ng/mL); x-axis, time after insulin injection (min). Data are means ± SE; n = 6 for sham, n = 7 for 2/3 Nx, and n = 8 for 5/6 Nx. P values are the results of two-way ANOVA. *P < 0.05 was considered significant. D: the blood glucose response to insulin (0.75 U/kg) prior to (baseline) and following 7 days of 0.1 M NH₄Cl in 2/3 Nx rats. y-Axis, blood glucose (mg/dL); x-axis, time after insulin injection (min). Data are means ± SE; n = 7 animals. P values are the results of two-way ANOVA. *P < 0.05 was considered significant. E: inverse AUC analysis for the blood glucose response to insulin in 2/3 Nx rats at baseline and following NH₄Cl treatment. y-Axis, AUC; x-axis, testing time point (baseline and NH₄Cl). Data are means ± SE; n = 7 animals. P values are the results of a paired Student’s t test. *P < 0.05 was considered significant. F: the blood glucose response to insulin (0.75 U/kg) prior to (baseline) and following 5−7 days of 0.1 M NH₄Cl in 5/6 Nx rats. y-Axis, blood glucose (mg/dL); x-axis, time after insulin injection (min). Data are means ± SE; n = 8 animals. P values are the results of two-way ANOVA. *P < 0.05 was considered significant. G: inverse AUC analysis for the blood glucose response to insulin in 5/6 Nx rats at baseline and following NH₄Cl treatment. y-Axis, AUC; x-axis, testing time point (baseline and NH₄Cl). Data are means ± SE; n = 8 animals. P values are the results of a paired Student’s t test. *P < 0.05 was considered significant.

Figure 5.

Impact of renal mass reduction on insulin sensitivity before and after treatment with NH₄Cl in Zucker obese rats. A: the blood glucose response to insulin (0.75 U/kg) at baseline in sham and 5/6 nephrectomized (Nx) rats as well as 5/6 Nx rats following 4 days of 0.1 M NH₄Cl. y-Axis, blood glucose (mg/dL); x-axis, time after insulin injection (min). Data are means ± SE; n = 3 for sham and n = 4 for 5/6 Nx. P values are the results of two-way ANOVA. *P < 0.05 was considered significant. B: inverse area under the curve (AUC) analysis for the blood glucose response to insulin in sham and 5/6 Nx rats at baseline and following NH₄Cl treatment. y-Axis, AUC; x-axis, experimental groups (sham, 5/6 Nx baseline, and 5/6 Nx NH₄Cl). Data are means ± SE; n = 3 for sham and n = 4 for 5/6 Nx. P values are the results of one-way ANOVA. *P < 0.05 was considered significant. C: plasma insulin levels during the insulin tolerance test at baseline in sham and 5/6 Nx rats and following NH₄Cl in 5/6 Nx rats. y-Axis, plasma insulin (ng/mL); x-axis, time after insulin injection (min). Data are means ± SE; n = 3 for sham and n = 4 for 5/6 Nx. P values are the results of two-way ANOVA. *P < 0.05 was considered significant.

Figure 6.

Relationship between insulin sensitivity and acid-base status in remnant kidney rats. A: relationship between pH and inverse area under the curve (AUC) throughout the study. y-Axis, inverse AUC analysis for the blood glucose response to insulin; x-axis, pH. Individual values from both Sprague Dawley and Zucker obese rats at all insulin tolerance test time points and blood gas measurements are shown. B: relationship between plasma ${\text{HCO}}_3^{-}$ and inverse AUC throughout the study. y-Axis, inverse AUC analysis for the blood glucose response to insulin; x-axis, plasma ${\text{HCO}}_3^{-}$ (mmol/L). Individual values from both Sprague Dawley and Zucker obese rats at all insulin tolerance test time points and blood gas measurements are shown. P values are the results of Deming (model II) linear regression. r indicates Pearson’s correlation. *P < 0.05 was considered significant.