Recurrent glucose deprivation leads to the preferential use of lactate by neurons in the ventromedial hypothalamus

Abstract

Increased GABAergic output in the ventromedial hypothalamus (VMH) contributes to counterregulatory failure in recurrently hypoglycemic (RH) rats, and lactate, an alternate fuel source in the brain, contributes to this phenomenon. The current study assessed whether recurring bouts of glucose deprivation enhanced neuronal lactate uptake and, if so, whether this influenced γ-aminobutyric acid (GABA) output and the counterregulatory responses. Glucose deprivation was induced using 5-thioglucose (5TG). Control rats received an infusion of artificial extracellular fluid. These groups were compared with RH animals. Subsequently, the rats underwent a hypoglycemic clamp with microdialysis. To test whether 5TG affected neuronal lactate utilization, a subgroup of 5TG-treated rats was microinjected with a lactate transporter inhibitor [cyano-4-hydroxycinnamate (4CIN)] just before the start of the clamp. Both RH and 5TG raised VMH GABA levels, and this was associated with impaired counterregulatory responses. 4CIN reduced VMH GABA levels and restored the hormone responses in the 5TG group. We then evaluated [¹⁴C]lactate uptake in hypothalamic neuronal cultures. Recurring exposure to low glucose increased monocarboxylate transporter-2 mRNA expression and augmented lactate uptake. Taken together, our data suggest that glucose deprivation, per se, enhances lactate utilization in hypothalamic neurons, and this may contribute to suppression of the counterregulatory responses to hypoglycemia.

INTRODUCTION

Hypoglycemia is a clinical problem that leads to greater incidences of morbidity in patients with type 1 and increasingly in those with type 2 diabetes that can sometimes be fatal. Hypoglycemia in diabetes typically stems from a combination of excess insulin and a compromised ability to generate an appropriate hormone response to counter a fall in blood glucose levels, conditions that lead to the development of hypoglycemia-associated autonomic failure (HAAF) (15). The mechanisms underlying the development of HAAF are not well understood.

Several mechanisms have been proposed to explain this phenomenon, and one of these, the alternate fuel hypothesis, postulates that hypoglycemia causes the brain to adapt to using other types of fuels such as lactate to meet its needs (17). Lactate is a critically sensed fuel substrate in the brain that can influence the activity of glucose-responsive neurons to glucose and inhibit glucose production (5, 13, 25, 26, 30). We previously reported the inhibitory neurotransmitter γ-aminobutyric acid (GABA) acts within the ventromedial hypothalamus (VMH) to suppress glucose counterregulatory responses and that a reduction of this inhibitory signal is important for full activation of the counterregulatory responses to hypoglycemia in normal hypoglycemia-naïve animals (9, 12). In rats with prior antecedent hypoglycemia, GABA output in the VMH is increased, and this contributes to counterregulatory failure, although the precise mechanism(s) leading up to the increase in GABA release have yet to be fully elucidated (16). It has been suggested that glial cells can support neuronal function under conditions of limited glucose availability by shuttling lactate to neurons through monocarboxylate transporters (MCTs) (31, 32, 34). Earlier work by Borg and colleagues showed that, when lactate was given locally in the VMH of normal hypoglycemia-naïve rats, it suppressed the counterregulatory responses to hypoglycemia (1). We later reported that lactate may achieve this goal by stimulating GABA release in the VMH (11). Recent studies reported that recurrent hypoglycemia can increase the utilization of monocarboxylic acids in the brain (23, 24, 35). However, the question of whether repeated exposure to a glucoprivic environment causes adaptations in hypothalamic neurons that enhance lactate uptake and/or utilization is not known. The current study investigates whether recurrent exposure to low glucose itself causes adaptations that increase the uptake of lactate by hypothalamic GABAergic neurons that, in turn, may enhance GABA release (11).

RESEARCH DESIGN AND METHODS

Animals

Adult male Sprague-Dawley rats (Harlan Laboratories, Indianapolis, IN) weighing 275–300 g were individually housed in the Yale Animal Resources Center in temperature (22 ± 2°C)- and humidity-controlled rooms. The animals were fed rat chow (Harlan Teklad, Indianapolis, IN) and water ad libitum and were acclimated to a 12:12-h light-dark cycle (lights on between 0700 and 1900) for a period of 1 wk before experimental manipulation. Principles of laboratory animal care were followed, and experimental protocols were approved by the Yale University Institutional Animal Care and Use Committee.

Surgery

Vascular surgery.

After a 1-wk acclimation period, the animals underwent aseptic surgery to have vascular catheters implanted in the left carotid artery and right jugular vein as described previously (Fig. 1 and Ref. 12).

Fig. 1.

Schematic diagram depicting the time course of surgeries, treatments, and clamp procedures. aECF, artificial extracellular fluid; 5TG, 5-thioglucose; 4CIN, α-cyano-4-hydroxycinnamate.

Stereotaxic surgery.

Subsequently, the animals were placed in a stereotaxic frame (David Kopf Instruments, Tujunga, CA), and a bilateral stainless steel guide cannula for microinjection (from bregma: 2.6 mm posterior, ± 0.6 mm lateral, and 7.6 mm ventral at an angle of 0°; Plastics One, Roanoke, VA) and two microdialysis guide cannulas (from bregma: 2.6 mm posterior, 3.8 mm lateral, and 8.8 mm ventral at an angle of 16°; Eicom, San Diego, CA) were inserted down to the level of the VMH and secured in place with screws and dental acrylic as described previously (10). The animals were then left in their home cages for 1 wk to recover before being randomly assigned to one of the following treatment groups.

Recurrent Hypoglycemia

Hypoglycemia was induced using a single intraperitoneal injection of insulin (5–10 U/kg Humulin R; Eli Lilly). Insulin doses were tapered down from 10 U/kg on day 1 to 8 U/kg on day 2 and to 5 U/kg on day 3 to account for the development of counterregulatory failure and to avoid seizures. Blood glucose concentrations were monitored from a tail nick every 30 min over 3 h using an AlphaTRAK 2 glucometer (Abbott Laboratories). Target glucose levels were maintained between 30 and 50 mg/dl for at least 2 h during this time. Following each episode of hypoglycemia, plasma glucose levels were recovered to euglycemia, and the animals were then returned to their home cages. This procedure was repeated for three consecutive days. Control animals received an intraperitoneal saline injection under similar sampling conditions.

Recurrent Glucose Deprivation in the VMH

Glucose deprivation was locally induced in the VMH for 3 h each day for three consecutive days using the glucose antimetabolite 5-thioglucose (5TG, 10 mM; Tocris). 5TG is thought to inhibit the first step of glycolysis, thus creating a low glucose or low energy state within the cells, a stimulus that is similar to hypoglycemia at the cellular level. Microinjection needles designed to extend 1 mm beyond the tip of the microinjection guide cannula were inserted, and 5TG was continuously infused into the VMH at a rate of 0.1 µl/min for 3 h. This infusion rate has been reported to localize the injection to small hypothalamic centers of interest (20). Blood glucose was monitored from a tail nick every 30 min. At the end of the 3-h period, the animals were returned to their home cages with free access to food and water. Control animals were perfused with artificial extracellular fluid (aECF) vehicle under similar conditions.

Hypoglycemic Clamp

The day following the final episode of recurrent hypoglycemia or 5TG treatment, overnight-fasted animals had bilateral 1-mm microdialysis/microinjection probes (Eicom) inserted through the microdialysis guide cannulas in the VMH of the animals, and their vascular catheters were connected to the infusion pumps. aECF was perfused through the microdialysis probes at a rate of 1.5 µl/min for the duration of the study. After a 2.5-h recovery period, baseline blood samples were collected, and baseline microdialysate samples were collected over 45 min. Following collection of the baseline microdialysate samples, the animals were injected over the course of 1 min with 0.1 µl of either aECF or the monocarboxylic acid transporter inhibitor α-cyano-4-hydroxycinnamate (4CIN, 15 nmol/side; Sigma-Aldrich, St. Louis, MO) dissolved in the aECF vehicle. Although 4CIN is a nonspecific inhibitor of monocarboxylic acid transporters, this lower concentration of 4CIN is more specific for MCT2 rather than MCT1 (19). Four groups of animals were used as follows: 1) control animals microinjected with aECF (n = 7), 2) recurrently hypoglycemic animals microinjected with aECF (RH, n = 10), 3) recurrent 5TG microinjected with aECF (5TG, n = 6), and 4) recurrent 5TG microinjected with 4CIN (5TG + 4CIN, n = 6). A small injection volume was used to limit spread of the compound outside of the VMH as described previously (3). Following microinjection, a constant insulin (50 mU·kg⁻¹·min⁻¹) and variable 20% glucose infusion were used to lower and maintain plasma glucose levels at 45 ± 5 mg/dl for 90 min. Blood samples were collected at 30, 60, and 90 min during the clamping period. Once the plasma was aspirated and frozen, the red blood cells were resuspended in an equivalent volume of artificial plasma and reinfused back in the animal to prevent volume depletion and anemia (28).

Probe placement.

At the end of the experiment, the animals were euthanized with an overdose of pentobarbital sodium (Sleepaway; Fort Dodge Animal Health, Fort Dodge, IA). The brains were rapidly removed, frozen, and stored at −80°C until they were analyzed. The brains were sectioned on a cryostat to 1) verify probe position and 2) collect frozen micropunches through the VMH for PCR analysis. Accuracy of probe position was histologically verified, and only those animals with properly positioned probes were included in the analysis [Supplemental Fig. S1 (Supplemental data for this article can be found at https://doi.org/10.6084/m9.figshare.c.4396754.v1.)].

Hormone and Microdialysate Analysis

Plasma catecholamine concentrations were analyzed by HPLC using electrochemical detection, whereas plasma hormone concentrations were determined using commercially available radioimmunoassay kits. VMH GABA concentrations from microdialysate samples were determined using LC/MS/MS after butylation, as described previously (19). Microdialysate lactate concentrations were analyzed using a photometric assay on the CMA600 small clinical chemistry analyzer (CMA Microdialysis, Holliston, MA).

Immunohistochemistry for Glutamic Acid Decarboxylase 65

Male Sprague-Dawley rats (n = 3) were deeply anesthetized with pentobarbital sodium and perfused with 0.01 M PBS followed by 10% neutral buffered formalin. The brains were then removed and immersed in 30% sucrose. Once sunk, the brains were cryosectioned in 30-µm sections and stored in an antifreeze solution at −20°C. The sections were then washed two times in 0.01 M PBS for 16 min before being incubated in PBS containing 0.1% Triton X-100 and the primary antibody [anti-glutamic acid decarboxylase 65 (GAD65, no. ab26113, 1:200 dilution; Abcam) or mouse IgG (no. ab81032, 1:200 dilution; Abcam)] overnight at room temperature. The next day, the sections were washed in PBS for 16 min before being incubated in PBS containing the Alexa Fluor 488 donkey anti-mouse IgG secondary antibody (no. A-21202, 1:1,000 dilution; Thermo). The sections were then mounted and imaged using an Olympus F10-SV confocal microscope and Fluoview software (Supplemental Fig. S2).

Glucose Deprivation of Hypothalamic Neurons

To determine whether glucose deprivation had direct effects on neuronal lactate uptake, we cultured a murine hypothalamic neuronal cell line (N43/5; Cellutions Biosystems) in DMEM (GIBCO) containing 10% FBS and 1% penicillin-streptomycin. This particular cell line was chosen because it not only expresses GAD65 but it was also shown to be glucose responsive (8). Once the cells reached 90% confluence, they were seeded at a density of 1 × 10⁶ cells in six-well plates in triplicate. The following treatment conditions were used: 1) the control group was incubated in media containing 2.5 mM glucose, and 2) the recurrent glucose deprivation group was incubated in media containing 0.2 mM glucose for 3 h each day for 3 consecutive days before being transferred back to medium containing 2.5 mM glucose.

[¹⁴C]lactate uptake.

Following the last day of treatment, the neurons were transferred to fresh medium containing 2.5 or 0.2 mM glucose to assess lactate uptake under baseline or glucoprivic conditions, respectively. To evaluate uptake, the neurons were washed two times with HEPES buffer and preincubated at 37°C for 10 min with 1 ml HEPES buffer. The HEPES buffer was then removed, and 0.5 ml of fresh HEPES buffer containing l-[¹⁴C]lactic acid (0.2 µCi/ml; PerkinElmer) and 2 mM cold lactic acid was added to each well. Later (30 s), uptake was stopped with the addition of ice-cold HEPES buffer. The cells were then washed two times with ice-cold HEPES buffer and solubilized in 0.5 ml of 1% SDS/0.2 N NaOH. The radioactivity was measured in a scintillation counter. Uptake values were calculated based on a standard curve. Protein concentrations were determined using a Bradford protein assay, and uptake values were standardized to the protein content.

MCT2 mRNA expression.

A second set of cells was subjected to the same treatment regimen as the control and recurrent glucose-deprived cells described above and were harvested and frozen on day 4 for quantification of MCT2 mRNA expression.

Quantitative RT-PCR for MCT2 mRNA

Total RNA was extracted from frozen VMH micropunches collected from the animal studies or pelleted neuronal cells from the cell culture studies using the Qiagen RNA extraction kit. MCT2 mRNA levels were evaluated using the Taqman assay (assay ID: Rn00568872_m1). Briefly, reverse transcription was performed by adding 10 µl of the Taqman master mix to 125 ng of RNA and running through the thermal cycler using the following settings: 25°C for 10 min, 37°C for 120 min, 85°C for 5 min, and 4°C hold. PCR reactions were carried out by adding 18 µl of the reaction mix to 2 µl of the cDNA and running the following reactions: 50°C for 2 min, 95°C for 2 min, 95°C for 1 s, and 60°C for 20 s for 40 cycles. Normalization was carried out using β-actin (assay ID: Rn01410374_m1) as the endogenous control. The ${{\displaystyle 2}}^{{{\displaystyle -\Delta \Delta \text{C}}}_{\text{T}}}$ method was used to calculate relative differences between experimental and control groups as a fold change in gene expression normalized to β-actin as the endogenous control (27).

Statistical Analysis

Treatment effects were analyzed using one- or two-way analysis of variance (ANOVA) for independent or repeated measures as appropriate, followed by Tukey’s multiple-comparisons post hoc analysis using the GraphPad Prism suite of analytical software for personal computers. P < 0.05 was set as the criterion for statistical significance.

RESULTS

Both Recurrent Hypoglycemia and Recurrent 5TG Increased Lactate and GABA Concentrations in the VMH

During the three treatment days before the clamp experiment, local delivery of 5TG into the VMH raised the average plasma glucose concentrations by ~20% from baseline levels (P < 0.05 vs. basal; Table 1). In contrast, plasma glucose levels were lowered to between 30 and 50 mg/dl during the hypoglycemic episodes in RH animals (P < 0.01 vs. basal; Table 1).

Table 1.

Average blood glucose concentrations from tail nick samples during the 3 days of treatment assessed at baseline and following initiation of treatment with either aECF, insulin-induced Hypo, or 5TG

	Control(n = 7)		RH(n = 10)		5TG(n = 6)		5TG + 4CIN(n = 6)
	Basal	aECF	Basal	Hypo	Basal	5TG	Basal	5TG
Day 1	118 ± 3	127 ± 2	104 ± 3	38 ± 5**	121 ± 5	141 ± 7*	120 ± 8	142 ± 5*
Day 2	120 ± 2	131 ± 4	105 ± 8	33 ± 2**	123 ± 3	139 ± 5*	121 ± 4	138 ± 4*
Day 3	119 ± 4	126 ± 2	99 ± 12	33 ± 4**	122 ± 3	142 ± 8*	121 ± 2	143 ± 8*

Data are presented as means ± SE; n, no. of animals. Units are mg/dl. RH, recurrently hypoglycemic; CIN4, cyano-4-hydroxycinnamate; aECF, artificial extracellular fluid; Hypo, hypoglycemia; 5TG, 5-thioglucose.

^*P < 0.05 versus basal.

^**P < 0.01 versus basal.

Baseline plasma insulin (Table 2) and glucose (Fig. 2A) concentrations on the day of the clamp were similar between the treatment groups. During the hypoglycemic clamp, plasma insulin and glucose concentrations were maintained at similar levels between the four treatment groups. In contrast, glucose infusion rates during the last 30 min were significantly higher in the RH- and 5TG-treated animals compared with controls (P < 0.05 vs. control; Fig. 2B). Blocking lactate uptake in 5TG-treated animals during the clamp reduced exogenous glucose infusion requirements to normal (P < 0.01 vs. 5TG; Fig. 2B).

Table 2.

Plasma insulin concentrations in control, RH, recurrent 5TG-treated, and recurrent 5TG-treated animals given 4CIN under the basal condition and during hypoglycemia

	Control (n = 7)	RH (n = 10)	5TG (n = 6)	5TG + 4CIN (n = 6)
Basal	10 ± 2	15 ± 2	8 ± 1	10 ± 2
Hypoglycemia	3,998 ± 208	4,345 ± 608	4,417 ± 529	4,465 ± 395

Data are presented as means ± SE; n, no. of animals. Units are µU/ml. RH, recurrently hypoglycemic; 5TG, 5-thioglucose; CIN4, cyano-4-hydroxycinnamate; basal, baseline conditions; hypoglycemia, hypoglycemic clamping period. No significant differences (1-way ANOVA, P = not significant) were observed between groups at baseline or during the hypoglycemic clamp.

Fig. 2.

A: plasma glucose concentrations during the hyperinsulinemic-hypoglycemic glucose clamp were matched between the four treatment groups. B: average glucose infusion rates during the last 30 min of the hypoglycemic clamp. RH, recurrently hypoglycemic; 5TG, 5-thioglucose; 4CIN, α-cyano-4-hydroxycinnamate. Data are presented as means ± SE. *P < 0.05 and **P < 0.01.

Three days of recurrent glucose deprivation with 5TG in the VMH led to significant increases in basal extracellular lactate (P < 0.05 vs. control; Fig. 3A) and GABA concentrations (P < 0.01 vs. control; Fig. 3B), similar to what was observed in the recurrent hypoglycemia group. During the hypoglycemic clamp, VMH lactate concentrations did not change significantly from their baseline levels over the 90-min clamping period. In contrast, VMH GABA levels declined significantly in controls (P < 0.05 vs. baseline) but failed to do so in RH- and 5TG-treated animals. Following administration of the monocarboxylic acid transporter inhibitor 4CIN in the VMH of 5TG-treated animals, extracellular lactate concentrations remained elevated (P < 0.05 vs. control; Fig. 3A), but the levels did not change significantly during the hypoglycemic clamping period, and this may be because of the inefficient recovery of the microdialysis procedure and/or the relatively short duration of our clamp procedure (38). Inhibiting neuronal lactate transport did, however, reduce VMH GABA levels significantly (P < 0.01 vs. baseline; Fig. 3B) during the hypoglycemic clamp despite the elevated extracellular lactate concentrations.

Fig. 3.

Extracellular lactate (A) and γ-aminobutyric acid (GABA, B) concentrations in the ventromedial hypothalamus (VMH) under baseline (open bars) conditions and during the hypoglycemic clamping period (closed bars). Recurrent hypoglycemic (RH) and 5-thioglucose (5TG) treatment increased basal lactate and GABA levels. In response to hypoglycemia, VMH GABA levels decreased in control but not in RH- or 5TG-treated rats. Blocking lactate uptake with cyano-4-hydroxycinnamate (4CIN) restored the decrease in VMH GABA levels during hypoglycemia. Peak plasma glucagon (C) and epinephrine (D) concentrations during the hypoglycemic clamp. RH and 5TG suppressed the counterregulatory hormone responses, and these were restored with 4CIN. Data are presented as means ± SE. *P < 0.05 and **P < 0.01.

The elevated GABA levels in the VMH of RH- and 5TG-treated groups were associated with significantly (P < 0.01 vs. control) reduced peak plasma glucagon (Fig. 3C) and epinephrine (Fig. 3D) responses during the hypoglycemic clamp. Blocking neuronal lactate transport in the VMH of 5TG-treated animals restored the hormonal defects to normal, suggesting it was the uptake and possible utilization of lactate by VMH neurons that stimulated GABA release which, in turn, suppressed glucagon and epinephrine secretion.

Recurrent hypoglycemia and recurrent glucose deprivation both led to significant (P < 0.05 vs. control) increases in MCT2 mRNA expression in the VMH (Fig. 4), suggesting recurring exposure to a low glucose environment caused adaptations that enhanced the neuronal capacity to take up lactate.

Fig. 4.

Compared with hypoglycemia-naïve controls (n = 7), monocarboxylate transporter (MCT)-2 mRNA expression in the ventromedial hypothalamus (VMH) is significantly increased under conditions of both recurrent hypoglycemia (RH; n = 10) and recurrent glucose deprivation [5-thioglucose (5TG); n = 6]. Data are presented as means ± SE. *P < 0.05.

Immunohistochemistry for GAD65 in the VMH

Immunohistochemical staining for GAD65, the rate-limiting enzyme involved in the synthesis of GABA, indicates the presence of GABAergic terminals throughout the VMH (Supplemental Fig. S2).

Neuronal Cell Culture Studies

To evaluate whether recurrent glucose deprivation influenced lactate uptake in hypothalamic neurons, we repeatedly incubated the N43/5 hypothalamic neuronal cells in media containing 0.2 mM glucose for 3 h/day before transferring them back to medium containing 2.5 mM glucose. This procedure was repeated for three consecutive days. On the day following the third treatment, we evaluated the uptake of [¹⁴C]lactate under normal (2.5 mM) glucose conditions and in response to a low (0.2 mM) glucose challenge. Repeated exposure of hypothalamic neurons to low glucose not only increased basal lactate uptake significantly (P < 0.05 vs. control baseline; Fig. 5A) but it also significantly increased lactate uptake in response to a low glucose challenge. The increases in lactate uptake were accompanied by significant (P < 0.05 vs. control) increases in MCT2 mRNA expression in the recurrent low glucose group (Fig. 5B, bottom).

Fig. 5.

N43/5 neuronal cell culture study. A: uptake of [¹⁴C]lactate in N43/5 cells cultured in medium containing 2.5 mM glucose (control) or following repeated incubations in medium containing 0.2 mM glucose (recurrent glucose deprivation) under baseline conditions (open bars) and following incubation in low-glucose medium for 90 min (closed bars). B: monocarboxylate transporter (MCT)-2 mRNA expression in N43/5 cells incubated in medium containing 2.5 mM glucose (control) or following repeated incubations in medium containing 0.2 mM glucose (recurrent glucose deprivation). Data are presented as means ± SE. * P < 0.05 versus control baseline. **P < 0.01 versus baseline.

DISCUSSION

Repeated exposure to hypoglycemia has been shown to elicit multiple adaptive changes within the brain and periphery that act to preserve normal function, including changes in glucose transport and also changes in the use of alternate fuel substrates (2–4, 11, 23, 24). Although beneficial in some respects, these adaptations may also hinder the ability of the brain to recognize a fall in blood glucose concentrations and trigger an appropriate counterregulatory hormone response to hypoglycemia. Our previous research revealed that one such fuel substrate, lactate, can dramatically enhance GABAergic neurotransmission and action in the VMH of normal rats when it is given directly, and elevated VMH lactate levels may be one factor contributing to counterregulatory failure in RH animals (11). Support for the latter stems from the observation that VMH lactate and GABA levels are not only elevated in RH animals, but inhibiting VMH lactate utilization using 4CIN reduced VMH GABA levels dramatically and helped to restore the counterregulatory responses in RH animals (11). Whereas this framework provides circumstantial evidence that repeated bouts of hypoglycemia can lead to changes within the central nervous system that enhance lactate uptake in neurons, these early studies did not identify the mechanisms by which this occurred. In the current study, we show through the use of in vivo microdialysis and in vitro hypothalamic neuronal cultures that repeated exposure of the VMH to a low glucose environment may increase its reliance on other fuel substrates, such as lactate, to supply and/or maintain neuronal fuel metabolism. In particular, we sought to determine whether recurrent glucose deprivation per se causes changes in lactate utilization in VMH neurons by 1) locally administering the glucoprivic agent 5TG directly in the VMH of hypoglycemia-naïve rats and 2) observing the changes in lactate uptake and monocarboxylic acid transporter expression in hypothalamic neurons that were repeatedly incubated in low glucose medium. Our data suggest that, when the supply of glucose is limited, MCT2 expression in VMH neurons is increased, leading to greater uptake of lactate.

In a similar manner to the recurrent insulin-induced hypoglycemia animals, three bouts of glucose deprivation in the VMH led to significant increases in VMH GABA levels, and this was associated with reductions in the counterregulatory hormone responses to hypoglycemia. Because 5TG has been reported to enhance lactate release in astrocytes (29), it is likely that a 5TG-induced rise in extracellular lactate is in part responsible for stimulating GABA release from GABAergic neurons in the VMH, since the pharmacological blockade of lactate uptake in VMH neurons was sufficient to restore both the fall in VMH GABA levels and the counterregulatory hormone responses. These data are consistent with those of Song and Routh who reported that incubation of glucose-excited neurons in medium containing lactate increased their firing rate in response to a drop in glucose from 2.5 to 0.1 mM (33). Whether lactate is acting as an alternate fuel substrate that is itself metabolized by neurons and/or whether it acts as a signal to sustain glucose metabolism in neurons (23) cannot be fully discerned in the current study; however, the increase in VMH MCT2 expression in both RH and recurring glucose deprivation, along with the observed increase in lactate uptake in cultured hypothalamic neurons following recurring exposure to low glucose, is supportive of the former. Additionally, work from Waagepetersen and colleagues indicates that lactate can be metabolized and incorporated into neurotransmitter GABA in neurons (36, 37), suggesting lactate can be metabolized and potentially serve as a substrate for GABA synthesis.

Infusion of 5TG in the VMH likely influenced a number of different cell types, including neurons and astrocytes alike. Impairing glucose metabolism in astrocytes can raise extracellular lactate concentrations through the breakdown of intracellular glycogen stores (7, 14, 18, 21, 22). On the other hand, acutely reducing glucose metabolism in glucose-responsive neurons may trigger an appropriate counterregulatory response initially, but, over the long term, it is possible that repeated impairment of the initial step in glucose metabolism with 5TG administration leads to an overall switch to the use of other fuel substrates to meet the metabolic needs of the neuron. It has been postulated that 5TG may remain in and possibly accumulate within neurons for prolonged periods, preventing glucose from being used effectively, and, hence, lactate may become a viable alternative. Although we did not perform continuous monitoring of central nervous system glucose or lactate levels during each of the recurrent bouts of glucose deprivation, we observed that blocking lactate uptake in neurons following three bouts of glucose deprivation was sufficient to restore the counterregulatory responses to normal during the clamp, suggesting lactate was a key factor in suppressing the hormonal responses to hypoglycemia in the 5TG animals. Preventing the uptake of lactate with 4CIN may help to enhance the fuel-deficient signal in hypothalamic glucose-responsive neurons and thereby restore the counterregulatory response.

To better understand the mechanism by which this metabolic switch occurs, we evaluated the direct effects of glucose deprivation on neuronal lactate uptake in a mouse hypothalamic neuronal cell line. Repeated exposure to low-glucose medium led to significant increases in MCT2 mRNA expression in hypothalamic neurons, and this, in turn, was associated with a significant increase in lactate uptake compared with neurons that were cultured in medium containing 2.5 mM glucose. These in vitro data further support the findings in the above animal studies, indicating that it was the lack of glucose itself and not cellular toxicity stemming from 5TG administration that enhanced lactate utilization. Our data are consistent with those of other groups who showed that 5TG administration in glucose-responsive areas of the hindbrain can also elicit very similar increases in MCT2 expression (6, 30, 35).

Taken together, our study suggests that, during recurring episodes of hypoglycemia, it is the repeated presentation of a low glucose stimulus, rather than repeated exposure to elevated insulin levels, that causes adaptations in the brain which raises extracellular lactate levels and increases the capacity of hypothalamic neurons to take up lactate by upregulating MCT2 expression (Fig. 6). In turn, the increase in lactate uptake by glucose-responsive GABAergic neurons in the VMH enhances GABA release and inhibition of its downstream targets (11). These findings suggest that a lack of glucose itself may be one component that fuels a metabolic switch to the preferential use of lactate in RH animals.

Fig. 6.

Recurring glucose deprivation associated with recurrent hypoglycemia enhances lactate production and release from astrocytes. Additionally, recurring low glucose also increases monocarboxylate transporter (MCT)-2 expression in GABAergic neurons, which increases its capacity to take up lactate. Enhanced lactate uptake, in turn, stimulates γ-aminobutyric acid (GABA) release from GABAergic neurons.

GRANTS

This work was supported by research grants from the Juvenile Diabetes Research Foundation (2-2008-480), the American Diabetes Association (7-11-BS-21), and the National Institute of Diabetes and Digestive and Kidney Diseases (R01-DK-099315).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

M.S., A.A., P.W., W.Z., and O.C. conceived and designed research; M.S., A.A., P.W., W.Z., and O.C. performed experiments; M.S., A.A., and P.W. analyzed data; M.S., A.A., P.W., and O.C. interpreted results of experiments; M.S. prepared figures; M.S. drafted manuscript; O.C. edited and revised manuscript; O.C. approved final version of manuscript.

ENDNOTE

At the request of the authors, readers are herein alerted to the fact that additional materials related to this manuscript may be found at https://doi.org/10.6084/m9.figshare.c.4396775.v1. These materials are not a part of this manuscript, and have not undergone peer review by the American Physiological Society (APS). APS and the journal editors take no responsibility for these materials, for the website address, or for any links to or from it.