Activation of Central Lactate Metabolism Lowers Glucose Production in Uncontrolled Diabetes and Diet-Induced Insulin Resistance

We studied 8-week-old male Sprague-Dawley rats (Charles River Laboratories, Montreal, Canada). Rats were housed in individual cages and subjected to a standard light-dark cycle. We implanted a single catheter into the third cerebral ventricle, for intracerebroventricular (ICV) infusions, by stereotaxic surgery. After ∼1 week of recovery, additional catheters were placed in the internal jugular vein and carotid artery for infusion and sampling, respectively, during the eventual in vivo infusion procedure. Recovery from surgery was monitored by measuring daily food intake and body weight gain in days preceding the infusion procedure. All study protocols were reviewed and approved by the institutional animal care and use committee of the Toronto General Hospital Research Institute, University Health Network.

Uncontrolled diabetes was induced with a single injection of streptozotocin (STZ) (60 mg/kg i.v.; Sigma, St. Louis, MO), a potent diabetogenic agent that is cytolytic against pancreatic β-cells, dissolved in sterile saline, at the end of the vascular surgery. The glycemic status of diabetic rats was monitored daily using a hand-held glucometer (Accu-Chek Compact Plus; Roche Diagnostics, Laval, Canada); only rats that were sufficiently hyperglycemic (blood glucose >15 mmol/l) were included in the diabetes studies. Fifty-four to sixty hours post-STZ injection, the in vivo ICV infusion experiments were carried out, and lasted a total of 3 h, in rats that were limited to 20 g of standard chow the previous night. We infused ICV vehicle (0.9% wt/vol NaCl) or ICV 5 mmol/l l-lactate throughout the experiments. A primed-continuous infusion of [3-³H]glucose (40 μCi bolus, 0.4 μCi/min thereafter; Perkin Elmer) was initiated at 90 min and maintained throughout the study to assess glucose kinetics. Plasma samples were collected in the final 0.5 h (150–180 min) for determination of [³H]glucose specific activity as well as hormone levels.

These in vivo infusion experiments lasted a total of 270 min and were carried out in normal rats that were limited to 20 g of standard chow the previous night. A primed-continuous infusion of [3-³H]glucose (40 μCi bolus, 0.4 μCi/min thereafter; Perkin Elmer) was initiated at the start of the protocol along with somatostatin (1 μg · kg⁻¹ · min⁻¹; Sigma). An infusion of ICV saline or 5 mmol/l lactate was initiated at 90 min and maintained until the end of the experiments (270 min). Plasma glucose level was monitored every 10 min throughout the protocol. Plasma samples were collected throughout the protocol for the determination of [³H]glucose specific activity as well as hormone levels. This experimental approach decreased plasma insulin levels by 50% in both ICV saline- or lactate-infused rats. In ICV saline-infused rats, plasma glucose level was elevated by 180–200 min, and this elevation was sustained until the end of the experiments. Thus, this is an acute model of hypoinsulinemia/hyperglycemia that does not have long-term metabolic disturbances, as seen in uncontrolled diabetes.

Animals were fed a high-fat (lard oil enriched) diet (catalog no. 9389; Purina Mills) that was generated by supplementing the standard chow with 10% lard for 3 days. The caloric content of the high-fat diet was 45% carbohydrate, 22% protein, and 33% fat and 44, 24, and 6.19%, respectively, for the standard chow. The total calories provided by digestible nutrients was 5.14 kcal/g for the high-fat diet versus 3.3 kcal/g for the standard chow. The in vivo infusion experiments lasted a total of 4.5 h. A primed-continuous infusion of [3-³H]glucose (40 μCi bolus, 0.4 μCi/min thereafter; Perkin Elmer) was initiated at 0 min. We infused ICV vehicle (0.9% wt/vol NaCl) or ICV 5 mmol/l l-lactate from time 90 min and onward until 270 min. A basal pancreatic insulin clamp was performed in the final 2 h (150–270 min) of the study: a continuous coinfusion of insulin (0.8 mU · kg⁻¹ · min⁻¹) and somatostatin (3 μg · kg⁻¹ · min⁻¹) was administered, and a variable infusion of 25% glucose solution was administered as needed to clamp and maintain the plasma glucose concentration at levels similar to those of the basal state. Plasma samples were collected in the final 0.5 h (240–270 min) for determination of [³H]glucose specific activity as well as hormone levels. Plasma insulin (Fig. 3C) and glucagon (Fig. 3D) levels were maintained at ∼0.9 ng/ml and ∼37 pg/ml, respectively. A separate set of hyperinsulinemic (insulin dosage: 3 mU · kg⁻¹ · min⁻¹)-euglycemic clamp experiments was performed in this high-fat–fed model to evaluate whether the high-fat diet induces insulin resistance.

Plasma glucose concentrations were measured by the glucose oxidase method (Glucose Analyzer GM9; Analox Instruments, Lunenberg, MA). Plasma insulin, glucagons, and leptin concentrations were measured by RIA (Linco Research, St. Charles, MO).

One-way ANOVA was performed to compare differences between treatments (ICV saline versus ICV lactate) in specific individual groups. Two-way ANOVA was performed to compare differences between the effects of ICV lactate in STZ, experimentally induced hypoinsulinemia, and high-fat–fed rodents using treatment (ICV saline vs. ICV lactate) and experimental (STZ vs. experimentally induced hypoinsulinemia vs. high-fat fed) groups as independent variables. Statistical calculations were performed using SAS software (SAS, Cary, NC). Analysis revealed no interaction between the treatment and the experimental groups, indicating that ICV lactate similarly lowered glucose production in all three models. Data are presented as means ± SEM. The final 30 min of the experiments were averaged for the “ICV” time period and used for the calculation of plasma glucose, insulin, and glucagon levels and glucose production rates. The start of the experiment (t = 0 min) was used for the “pre-ICV” time period and for the calculation of starting plasma glucose levels.

To examine whether the activation of central nervous system (CNS) lactate sensing mechanisms lower glucose production and glucose levels in diabetes (Fig. 1A), we first established an early-onset uncontrolled diabetic rat model with an intravenous STZ injection (11). Within 30 h of the STZ injection (60 mg/kg), plasma glucose levels were elevated to at least 15 mmol/l. The metabolic effects of CNS lactate sensing were tested in the STZ rats whose plasma glucose levels were elevated to ∼20 mmol/l for ∼24–30 h (or 54–60 h post-STZ injections) (Fig. 1C). We first examined the peripheral glucose metabolism in these STZ rats using tracer-dilution methodology (Fig. 1B) (10). Hepatic glucose production in ICV saline–administered STZ rats was elevated to ∼22 from 11.2 ± 0.7 mg · kg⁻¹ · min⁻¹ (as seen in normal rats) (Fig. 1D). This elevation of glucose production increased plasma glucose levels to ∼20 mmol/l (Fig. 1C) in the presence of very low plasma insulin (Fig. 1E) but normal glucagon levels (Fig. 1F).

To test the metabolic impact of CNS lactate sensing in STZ rats, lactate was administered ICV at a concentration (10) that rapidly lowers glucose production and plasma glucose levels in normal rodents (Fig. 1B). Lactate ICV administration lowered glucose production by ∼6 mg · kg⁻¹ · min⁻¹ (Fig. 1D), which led to a drop in plasma glucose levels of 4.3 ± 0.8 mmol/l or a suppression of 25 ± 4% from pre-ICV lactate (Fig. 1C) in STZ rats. Importantly, this occurred in the presence of very low insulin (Fig. 1E) but normal glucagon levels (Fig. 1F). Given the fact that the activation of hypothalamic insulin signaling via direct insulin injection lowers glucose levels in uncontrolled diabetes (11), activation of hypothalamic lactate metabolism could serve as a complementary approach that targets the brain to lower glucose levels in diabetes. An activation of central lactate metabolism lowers glucose levels in uncontrolled diabetes in the absence of insulin injections.

Because uncontrolled diabetes is associated with metabolic disturbances and hypoinsulinemia, we alternatively and selectively examined whether central lactate metabolism regulates glucose homeostasis in the presence of hypoinsulinemia. We first established an experimentally induced hypoinsulinemic model in normal rodents with a method similar to that used previously in dogs (12) (Fig. 1A and Fig. 2A). Somatostatin administered at a dosage of 1 μg · kg⁻¹ · min⁻¹ in normal rats decreased plasma insulin levels by 50% (Fig. 2D), but plasma glucagon levels were not affected (Fig. 2E, as seen in STZ rats). This experimental approach resulted in an elevation of glucose production (15.8 ± 1.4 mg · kg⁻¹ · min⁻¹) and plasma glucose levels (10.8 ± 0.3 mmol/l) by 180–200 min. Glucose production was elevated to 15.6 ± 1.1 mg · kg⁻¹ · min⁻¹ (Fig. 2C) and plasma glucose levels to 10.9 ± 0.5 mmol/l (Fig. 2B) in normal rodents with ICV saline administration by the end of the infusion studies (240–270 min). In contrast, in rats that received ICV lactate administration starting at 90 min, the rise in glucose production (10.2 + 2.1 mg · kg⁻¹ · min⁻¹) (Fig. 2C) and plasma glucose levels (7.3 + 0.6 mmol/l) was prevented (Fig. 2B) throughout the infusion protocol. Together with the findings on STZ rats, these data indicate that insulin signaling is not required for central lactate metabolism to lower glucose production and glucose levels and suggest that the activation of hypothalamic lactate metabolism could bypass insulin resistance and lower glucose production in diet-induced insulin-resistant models.

To directly test this hypothesis, the metabolic effects of lactate ICV were tested in an early-onset high-fat diet–induced insulin-resistant model (13,14) (Fig. 1A). Here, we confirm that rats fed a lard oil–enriched diet rapidly increased their food intake (Fig. 3A), developed hepatic insulin resistance (% glucose production suppression from basal obtained by hyperinsulinemic-euglycemic clamps: 75 ± 5% [regular chow, n = 4] vs. 23 ± 3% [high-fat diet, n = 6], P < 0.001), and had elevated plasma insulin (0.8 ± 0.2 ng/ml [regular chow] vs. 1.9 ± 0.3 ng/ml [high-fat diet], P < 0.01) and leptin (1.1 ± 0.3 nmol/l [regular chow] vs. 2.1 ± 0.2 nmol/l [high-fat diet], P < 0.01) levels. In this high fat–fed model, ICV lactate administration lowered glucose production during the pancreatic basal insulin clamps (Fig. 3B). Importantly, this metabolic effect of lactate ICV was comparable with that in regular chow–fed rats (Fig. 3B). These data indicate that central lactate metabolism bypassed insulin resistance to lower glucose production in a diet-induced insulin-resistant model.

Here we report that a selective central activation of lactate metabolism similarly lowered glucose production in the early-onset models of uncontrolled diabetes, hypoinsulinemia, and diet-induced insulin resistance. It still remains unknown whether hypothalamic lactate sensing regulates glucose homeostasis in chronic models of diabetes and obesity. Furthermore, metabolic effects of lactate sensing in other regions of the brain (i.e., hindbrain) warrant future investigations, since inhibition of neuronal lactate transport (the reverse of direct lactate administration, as done in the current study) in the hindbrain increases plasma glucose levels (15). Finally, although selective activation of central lactate metabolism lowered glucose production and plasma glucose levels in uncontrolled diabetes in the absence of insulin injections, glucose production and glucose levels were not fully normalized. Taken together, we postulate that therapeutic strategy designed to activate hypothalamic lactate metabolism could bypass insulin resistance and serve as one of the complementary approaches to insulin injections to lower glucose levels in diabetes and obesity.

This work was supported by a research grant to T.K.T.L. from the J.P. Bickell Foundation. M.C. was supported by a Banting and Best Diabetes Centre Charles Hollenberg summer studentship. T.K.T.L. holds the John Kitson McIvor Endowed Chair in Diabetes Research at the University of Health Network and University of Toronto.