Magnitude of Exercise-Induced β-Endorphin Response Is Associated with Subsequent Development of Altered Hypoglycemia Counterregulation

Abstract

Context:

β-Endorphin release in response to recurrent hypoglycemia is implicated in the pathogenesis of hypoglycemia-associated autonomic failure.

Objective:

We hypothesized that exercise-induced β-endorphin release will also result in the deterioration of subsequent hypoglycemia counterregulation and that the counterregulatory response will negatively correlate with the degree of antecedent β-endorphin elevation.

Design, Setting, Participants, and Interventions:

Sixteen healthy subjects (six females, aged 26 ± 4.3 yr, body mass index 26.1 ± 5.6 kg/m²) were studied with three experimental paradigms on 2 consecutive days. Day 1 consisted of one of the following: 1) two 90-min hyperinsulinemic hypoglycemic clamps (3.3 mmol/liter); 2) two 90-min hyperinsulinemic euglycemic clamps while subjects exercised at 60% maximal oxygen uptake; or 3) two 90-min hyperinsulinemic euglycemic clamps (control). Day 2 followed with hyperinsulinemic (396 ± 7 pmol/liter) stepped hypoglycemic clamps (5.0, 4.4, 3.9, and 3.3 mmol/liter plasma glucose steps).

Main Outcome Measures:

Day 2 hypoglycemia counterregulatory hormonal response and glucose turnover ([3-³H]-glucose) as indicators of recovery from hypoglycemia.

Results:

There was a significant inverse correlation between plasma β-endorphin levels during exercise and catecholamine release during subsequent hypoglycemia. Subjects with an exercise-induced rise in β-endorphin levels to above 25 pg/ml (n = 7) exhibited markedly reduced levels of plasma epinephrine and norepinephrine compared with control (2495 ± 306 vs. 4810 ± 617 pmol/liter and 1.9 ± 0.3 vs. 2.9 ± 0.4 nmol/liter, respectively, P < 0.01 for both). The rate of endogenous glucose production recovery in this group was also much lower than in controls (42 vs. 89%, P < 0.01).

Conclusions:

The physiological increase in β-endorphin levels during exercise is associated with the attenuation of counterregulation during subsequent hypoglycemia.

The benefits of glycemic control in patients with type 1 diabetes (T1DM), manifested as reductions in many diabetes associated complications, are well established (1). However, achievement of near-normal glycemia carries a significant risk of hypoglycemia. In the Diabetes Control and Complications Trial, episodes of biochemical and symptomatic hypoglycemia occurred more frequently during intensive therapy, and the risk of severe hypoglycemia increased 3-fold (2). Despite medical advances, insulin therapy in diabetes remains imperfect and a source of iatrogenic hypoglycemia. Recurrent episodes of hypoglycemia in T1DM patients result in the compromise of counterregulatory hypoglycemic responses during subsequent bouts of hypoglycemia, ultimately culminating in impaired recovery from hypoglycemia (3). This phenomenon, termed hypoglycemia-associated autonomic failure (HAAF), is manifested biochemically by blunted glucagon release and sympathoadrenal activation and clinically as hypoglycemia unawareness (3).

The regulation of glucose homeostasis during exercise has parallels to the counterregulatory response elicited by hypoglycemia (4). A major barrier to physical activity in T1DM is hypoglycemia, which frequently occurs during or after exercise (5, 6). Experimental evidence suggests that antecedent exercise in normal subjects and in those with T1DM attenuates autonomic counterregulatory responses during subsequent hypoglycemia, thereby contributing to the development of exercise-associated autonomic failure (EAAF) (7, 8). This implies that shared mechanisms may be responsible for hypoglycemia-related and exercise-related autonomic failure.

The precise mechanism of HAAF has not been elucidated yet; however, a number of factors have been proposed to contribute to its pathogenesis, including cortisol release, changes in brain glycogen concentrations, and other neuroendocrine signals that mediate the counterregulatory response (9–11). Various stressors, including exercise and hypoglycemia, have been shown to induce the release of endogenous opioids, including β-endorphin, which in turn modulate the autonomic/sympatoadrenomedullary response (12, 13). Using a blockade of endogenous opioids during exercise and hypoglycemia, we and others have demonstrated an augmentation of sympatoadrenal activation, suggesting that endogenous opioids may play a role in the development of HAAF (12, 14, 15). Based on the above findings, we hypothesized that endogenous opioid release during exercise would also result in attenuation of counterregulation during subsequent hypoglycemia, mimicking the compromised counterregulatory response observed with recurrent hypoglycemia, and that the counterregulatory response will negatively correlate with the degree of antecedent β-endorphin elevation.

Materials and Methods

We studied 16 healthy volunteers [10 men, six women, aged 26 ± 4.3 yr, body mass index (BMI) 26.1 ± 5.6 kg/m², glycosylated hemoglobin (HbA1c) 4.8 ± 0.5%]. Each subject participated in three sets of studies, in random order, separated from each other by at least 5 wk. All studies were performed after an overnight fast. At least 2 wk before the initial study, all subjects were admitted to the Clinical Research Center to determine their maximal oxygen uptake (VO_2max). Incremental exercise was performed on a stationary cycle ergometer and expired gases were collected and analyzed using computerized open-circuit indirect calorimetry (VMax-29; SensorMedics, Yorba Linda, CA), as previously described (16, 17). VO_2max averaged 41 ± 1.1 ml/kg · min.

Each set of studies consisted of 2 consecutive days. Day 1 in each set of studies consisted of one of the following: 1) two 90-min hyperinsulinemic hypoglycemic clamps, with plasma glucose maintained at 3.3 mmol/liter (H); 2) two 90-min hyperinsulinemic euglycemic clamps while the subject exercised at 60% of his or her VO_2max on a stationary ergonomic bicycle (E); or 3) two 90-min hyperinsulinemic euglycemic clamps, while the subject rested [control (C)]. See Fig. 1 for the graphical representation of the d 1 experimental conditions. Day 2 was identical in all studies and included a hyperinsulinemic stepped hypoglycemic clamp, with quantification of counterregulatory hormonal responses and glucose kinetics.

Fig. 1.

Experimental protocol on d 1.

Informed written consent was obtained in accordance with policy of the Institutional Review Board of the Albert Einstein College of Medicine. Subjects were admitted to the Clinical Research Center for each experiment.

Day 1

At 0800 h on the study day, all subjects had two indwelling cannulae inserted, one in the antecubital vein for infusions and the second placed in a distal vein of the contralateral forearm for blood sampling. To obtain arterialized venous blood samples, this hand was maintained at 65 C in a thermoregulated sleeve. At t = −30 min, a constant insulin infusion (Humulin Regular; Eli Lilly, Indianapolis, IN) was initiated at a rate of 1.0 mU/kg · min, and a variable infusion of 20% dextrose was administered to maintain the plasma glucose concentration at euglycemia until t = 0. Blood samples were collected at 5-min intervals for measurements of plasma glucose. At t = 0, subjects were assigned to 90 min of hypoglycemia, exercise, or euglycemia, at which point the 20% dextrose infusion was adjusted to maintain the desired plasma glucose concentration, depending on the study performed. At the completion of 90 min, the insulin infusion was discontinued and the subjects rested for 90 min. During this time a small snack containing 15 g of carbohydrate was administered. At t = 180, the experimental conditions were resumed, with subjects assigned to the same conditions as during the first 90 min. In all the groups, blood samples were obtained for the determinations of serum β-endorphin.

Day 2

The study conducted on d 2 was identical in all protocols. At 0800 h, the subjects had two indwelling cannulae inserted. At t = −120 min, a primed continuous infusion of HPLC-purified [3-³H] glucose was initiated with a bolus of 21.6 μCi, followed by a continuous infusion of 0.15 μCi/min for the entire period of the study. The specific activity of infused dextrose was kept equivalent to plasma specific activity by addition of [3-³H] glucose to the infusate, as previously described by Finegood et al. (18). At t = 0 min, a primed continuous infusion of insulin was initiated at a rate of 1.0 mU/kg · min for the first 10 min and thereafter was continued at 0.5 mU/kg · min throughout the study. At t = 10 min, a variable infusion of 20% dextrose was also begun to maintain the plasma glucose concentration at 5 mmol/liter for 50 min. At t = +50 min and every 50 min thereafter, the plasma glucose concentration was decreased by 0.6-mmol/liter decrements for 50 min each by reducing the dextrose infusion rate accordingly. Plasma glucose was clamped at the desired range according to plasma glucose measured at 5-min intervals with targets of 5.0, 4.4, 3.9, and 3.3 mmol/liter. Blood samples were obtained for the determinations of plasma insulin, C-peptide, glucagon, epinephrine, norepinephrine, and cortisol as well as for glucose turnover.

Plasma glucose was measured with a Beckman glucose analyzer (Beckman Coulter, Fullerton, CA), using the glucose oxidase method. Plasma [3-³H] glucose radioactivity was measured in duplicate on the supernatants of barium hydroxide-zinc sulfate precipitates of plasma samples, after evaporation to dryness to eliminate tritiated water (19).

The methods for measurement of plasma insulin, C-peptide, glucagon, cortisol, and their intra- and interassay variations have been previously reported (20). Plasma β-endorphin was measured using ELISA (MD Bioproducts, St. Paul, MN). Plasma epinephrine and norepinephrine levels were determined using radioenzymatic assays described by Evans et al. (21).

The data are presented as mean ± sem. Steele's equation was used for calculation of glucose turnover as described elsewhere (22). Values for endogenous glucose production (EGP) and glucose uptake, obtained at 10-min intervals, were averaged over the final 30 min of each glucose step for each individual subject. Statistical analyses were performed using repeated-measures ANOVA for multiple comparisons. Linear regression was calculated using ordinary least squares. A value of P < 0.05 was considered significant.

Results

Day 1

All subjects achieved the glycemic goals set by our protocol (Fig. 1).

Plasma β-endorphin levels at baseline (time 0) were 5.1 ± 0.9, 3.3 ± 0.7, and 4.2 ± 0.8 pg/ml in the H, E, and C studies, respectively. At the end of the studies (time 270 min), plasma β-endorphin concentrations increased significantly in the H and E studies (19.3 ± 2.1 and 23.5 ± 5.0 pg/ml, respectively, P < 0.01 compared with time 0) but remained unchanged in the C studies (5.6 ± 0.5 pg/ml, P = NS compared with time 0).

Day 2

Plasma glucose concentrations during the stepped hyperinsulinemic hypoglycemic clamps on d 2 are shown in Fig. 2A. Plasma glucose concentrations at t = 0 averaged 5.2 ± 0.1 mmol/liter in the H studies, 5.1 ± 0.1 mmol/liter in the E studies, and 5.3 ± 0.2 mmol/liter in the C studies (P = NS). No significant difference was noted between all studies during the 5.0, 4.4, and 3.9 mmol/liter target glucose steps. However, during the 3.3-mmol/liter target glucose step, plasma glucose could not be decreased below 3.6 ± 0.1 mmol/liter in the C studies despite the fact that the dextrose infusion was discontinued. In the H and E studies, plasma glucose averaged 3.3 ± 0.1 and 3.4 ± 0.2 mmol/liter, respectively, at the 3.3-mmol/liter target glucose step.

Fig. 2.

A, Plasma glucose concentrations (d 2) at each glucose step in the antecedent exercise studies (white circles), hypoglycemia (black squares), and control studies (white triangles). B, Glucose infusion rates over time. C, Plasma insulin concentrations. D, Plasma C-peptide concentrations.

Glucose infusion rates are depicted in Fig. 2B. During the first (5 mmol/liter) and second (4.4 mmol/liter) target glucose steps, average glucose infusion rates were comparable in all studies (9.4 ± 1.1 μmol/kg · min in the H, 10 ± 1.1 μmol/kg · min in the E, and 10 ± 1.1 μmol/kg · min in the C studies, P = NS). However, during the 3.9-mmol/liter glucose step, the mean rate of glucose infusion was 9.4 ± 1.1 μmol/kg · min and 8.9 ± 2.2 μmol/kg · min in the H and E studies, respectively, and 6.7 ± 0.6 μmol/kg · min in the C studies (P = 0.01 vs. the H and E studies). During the 3.3-mmol/liter glucose step, the average glucose infusion rate was also significantly lower in the C studies (0.6 ± 0.0 vs. 6.7 ± 0.6 μmol/kg · min and 5.6 ± 3.9 μmol/kg · min in the H and E studies (P < 0.01).

Basal plasma insulin concentrations were nearly identical in all studies, averaging 45.8 ± 2.8 pmol/liter in the H, 49 ± 4.2 pmol/liter in the E, and 57.6 ± 6.3 pmol/liter in the C studies (P =NS). Similarly, there was no significant difference in plasma insulin concentration during all clamps averaging 395.2 ± 34.7 pmol/liter in the H, 383.4 ± 36.1 pmol/liter in the E and 402.8 ± 34 pmol/liter in the C studies (P = NS, Fig. 2C). Plasma C-peptide concentrations were comparable in all sets of studies at baseline (0.46 ± 0.1, 0.52 ± 0.1, and 0.43 ± 0.1 nmol/liter in the H, E, and C studies, respectively, P = NS). C-peptide values remained similar in all clamps, suppressing to an average of 0.1 ± 0.01 nmol/liter at the hypoglycemic nadir, at t = 200 min (Fig. 2D).

During the 5- and 4.4-mmol/liter glucose steps, plasma epinephrine concentrations remained at basal values and were similar in all studies (219 ± 24.6, 192 ± 34.9, and 180 ± 29.5 pmol/liter, in the H, E, and C studies, respectively, P = NS). Further reduction in plasma glucose to 3.3 mmol/liter was associated with an expected increment in plasma epinephrine in the C studies to maximum value of 4810 ± 617 vs. 2796 ± 240 pmol/liter in the H studies (P < 0.01) and 3773 ± 683 pmol/liter in the E studies (P = NS), (Fig. 3A). Thus, antecedent hypoglycemia was associated with blunting of the epinephrine response to subsequent hypoglycemia. Antecedent exercise was also associated with a decreased epinephrine response compared with the C studies, although the difference was only statistically significant at the 160-and 180-min time points. However, when these results were analyzed using the area under the curve, a significant difference in epinephrine was found between the C and all other groups (12558 ± 1127 vs. 6454 ± 521 and 8417 ± 1049 pmol/liter for C vs. H and E, respectively, P < 0.05).

Fig. 3.

Day 2 plasma epinephrine (*, P < 0.001 vs. hypoglycemia, A) and norepinephrine (*, P < 0.001 vs. hypoglycemia, B) concentrations over time and plasma glucagon (*, P < 0.01 vs. hypoglycemia, C) and cortisol (D) concentrations over time in the antecedent exercise studies (white circles), hypoglycemia (black squares), and control studies (white triangles).

Plasma norepinephrine concentrations were equivalent during the 5- and 4.4-mmol/liter glucose steps in all studies (Fig. 3B). However, during the 3.9- and 3.3-mmol/liter glucose steps, plasma norepinephrine increased significantly in the C studies to a maximum value of 2.9 ± 0.4 nmol/liter and to a lesser degree in the H studies (1.8 ± 0.2 nmol/liter, P < 0.01 vs. C), reflecting altered counterregulatory response. In the E studies, plasma norepinephrine increase (2.6 ± 0.5 nmol/liter) in response to hypoglycemia was lower than in the C study; however, the difference was not significant. No significant difference in norepinephrine response was found between C and E studies using the area under the curve analysis either.

Plasma glucagon was also equivalent in all studies during the 5- and 4.4-mmol/liter glucose steps (Fig. 3C), and the increase in glucagon release was blunted in the H studies (47 ± 8 ng/liter) compared with the C and E studies (96 ± 6 and 104 ± 8 ng/liter, respectively, P < 0.01). Plasma cortisol concentrations were similar at baseline and increased proportionally and equally in all studies (Fig. 3D).

Mean baseline EGP was similar in all studies (11.1 ± 0.6, 10.5 ± 1.1, and 11.7 ± 1.1 μmol/kg · min, in the H, E, and C studies, respectively, P = NS). During the 5- and 4.4-mmol/liter glucose steps, EGP was equally suppressed by approximately 71% in all studies. During the next two clamp steps (3.9 and 3.3 mmol/liter), EGP rose by 6 and 31%, respectively, in the H studies, showing only modest EGP recovery. In contrast, EGP rose to 55 and 63%, respectively, in the E studies (P < 0.01 compared with the H studies), suggesting adequate recovery and to a similar extent in the C studies (50 and 65%, respectively, P < 0.01 vs. the H studies) (Fig. 4).

Fig. 4.

EGP (d 2) averaged for the final 30 min of each glucose step (*, P < 0.01 vs. hypoglycemia).

To evaluate the correlation between the rise in plasma β-endorphin levels and next-day hypoglycemia counterregulation, we analyzed the association between plasma β-endorphin and the hormonal counterregulation using statistical stepwise regression. There was a significant inverse relationship between exercise-induced rise in plasma β-endorphin concentrations and subsequent epinephrine/norepinephrine responses to hypoglycemia (Fig. 5). Furthermore, other factors that were analyzed in this model, including the level of VO_2max, gender, BMI, HbA1c, plasma glucagon, and plasma cortisol had no significant effect on the hormonal response to hypoglycemia. Thus, we reexamined in a post hoc analysis the counterregulatory response to hypoglycemia in subjects that exhibited an exercise-induced rise in plasma β-endorphin to above 25 pg/ml (EN). In this subgroup (n = 7), the β-endorphin level averaged 43.7 ± 4.5 pg/ml. As shown in Fig. 6, A and B, during the 3.9- and 3.3-mmol/liter glucose steps, plasma epinephrine and norepinephrine were significantly blunted in the EN studies, similar to the H studies (with maximum epinephrine and norepinephrine values of 2495 ± 306 pmol/liter and 1.9 ± 0.3 nmol/liter vs. 2796 ± 240 pmol/liter and 1.8 ± 0.2 nmol/liter, respectively, P = NS between groups). Furthermore, EGP failed to increase appropriately with counterregulation in the EN studies, analogously to the H studies (42 vs. 43%, P = NS). As expected, these values differed significantly from the C group, which demonstrated prompt counterregulatory responses characterized by a significant rise in epinephrine and norepinephrine and a robust increase in EGP (4810 ± 617 pmol/liter, 2.9 ± 0.4 nmol/liter, and 89%, respectively). Thus, the exercise studies that were associated with significant increases in plasma β-endorphin levels exhibited altered hypoglycemia counterregulation on d 2, independent of any other factors.

Fig. 5.

Correlation between plasma β-endorphin concentrations (d 1) and d 2 plasma epinephrine (A) and plasma norepinephrine (B) in the exercise studies. Values represent the difference between the point of maximal rise and baseline for all determinations.

Fig. 6.

Day 2 plasma epinephrine (*, P < 0.001 vs. hypoglycemia, A) and norepinephrine (*, P < 0.001 vs. hypoglycemia, B) concentrations over time in the antecedent exercise studies with significantly elevated plasma β-endorphins concentrations (>25 pg/ml, white circles), hypoglycemia (black squares), and control studies (white triangles).

Discussion

Our results demonstrate that in nondiabetic humans, significant β-endorphin release during antecedent exercise is associated with deterioration of the counterregulatory response to subsequent hypoglycemia on the following day, similar to the effect of antecedent hypoglycemia. Furthermore, the effect of β-endorphin was found to be independent of exercise intensity, hormone levels, BMI, gender, or HbA1c measures. These findings suggest that endogenous opioids may contribute to the development of exercise-associated autonomic failure observed in patients with diabetes. The blunting of the counterregulatory response to subsequent hypoglycemia was characterized by a less significant epinephrine and norepinephrine rise, compared with the C group. The inadequate hormonal response to hypoglycemia after exercise culminated in the impaired recovery of glucose levels, as evidenced by lower EGP and higher levels of dextrose infusion required by the exercise group to maintain plasma glucose at the desired levels.

A number of previously published studies have examined the effects of exercise on the subsequent autonomic response to hypoglycemia, with some, but not all, demonstrating that antecedent exercise may lead to the subsequent development of impaired hypoglycemia counterregulation in healthy subjects (23–25). Blunting of the catecholamine response to hypoglycemia after antecedent exercise was shown by Galassetti et al. (23). Others, however, observed deterioration of the epinephrine response only, despite exercise performed at higher intensity, as measured by the percentage VO_2max, (24), or no change at all (25). As in our experiments, Rattarasarn et al. (25) and McGregor et al. (24) did not observe decrements in the glucagon levels in response to subsequent hypoglycemia after exercise, whereas Galassetti et al. (23) reported a decrease in the glucagon response. The fact that we did not find a decrement in glucagon release in response to exercise induced β-endorphin rise suggests that a different mechanism may be involved in the attenuation of the glucagon response as opposed to the catecholamine response. The glucose infusion rates required to maintain the plasma glucose steps during the hypoglycemia on d 2 were higher and the EGP was lower in the studies after exercise (23, 24), similar to our findings. These differences in results between our findings and others' findings (23–25) may be related to variations in protocols, including differences in exercise regimens and insulin infusion rates.

The above studies all varied in exercise duration and intensity, measured as percentage VO_2max, which was required to induce subsequent impaired hypoglycemia counterregulation. Furthermore, none of the studies measured plasma β-endorphin concentrations. This inconsistency in findings suggests that the development of exercise-associated subsequent impaired hypoglycemia counterregulation may not be solely related to the intensity, measured as percent VO_2max, or duration of exercise, but to the exercise-induced release of endogenous β-endorphin. Notably, the variability of β-endorphin responses to exercise in our study was large; however, after adjustment for multiple factors, including VO_2max, gender, BMI, HbA1c, and other hormonal parameters, β-endorphin level emerged as the only factor predictive of subsequent attenuation of the counterregulatory response to hypoglycemia. In addition, when we compared the counterregulatory response to hypoglycemia in the group with exercise-induced β-endorphin release of less than 25 pg/ml with the group whose β-endorphin level reached above 25 pg/ml, we found a significant deterioration of the counterregulatory response to hypoglycemia, similar in degree to antecedent hypoglycemia, only in the latter. This implies that there may be a β-endorphin threshold that needs to be attained before significant hypoglycemia counterregulatory failure ensues. Interestingly, we did not find a similar negative correlation between hypoglycemia-induced rise in β-endorphin and subsequent hormonal counterregulatory responses. This suggests that additional factors may be responsible for the development of HAAF or that autonomic activation and substrate metabolism may differ during antecedent hypoglycemia. Another potential explanation for not observing a negative correlation between hypoglycemia-induced rise in β-endorphin and subsequent hormonal counterregulatory responses may be that the hypoglycemia-induced rise in β-endorphin was more uniform among the subjects (sem 2.1 pg/ml) compared with the exercise-induced rise in β-endorphin (sem 5 mg/dl); thus, with a greater variability in β-endorphin levels, an inverse correlation was more likely to be detected.

Various factors affect the release of endogenous opioids during exercise. These include the subject's fitness level and intensity of exercise performed. Physical training and increased exercise intensity have been shown to elevate β-endorphin levels during exercise; thus, exercise at the same percentage of VO_2max stimulates greater β-endorphin release in trained compared with untrained subjects (26, 27). Interindividual variations also exist, such that certain highly trained individuals exercising to exhaustion may not exhibit β-endorphin elevation (28). This evidence suggests that the measure of β-endorphin elevation during exercise may be a better predictor for the subsequent development of exercise-associated HAAF than relying on the intensity of exercise or fitness of the subject, as confirmed by our results; however, further study is required.

β-Endorphins are endogenous opioids, which are synthesized and released by the proopiomelanocortin neurons of the pituitary gland (10) in response to a variety of stressors, including hypoglycemia and exercise (12, 13). The precise stimulus for β-endorphin release during exercise remains uncertain; however, proposed mechanisms include anaerobic metabolism and exercise intensity, as previously described (12). β-Endorphin has been implicated in analgesic response and hormonal regulation of glucose metabolism during exercise (12). The latter is supported by studies of opioid antagonist-induced alterations of hormonal response during exercise, such as enhancement of the sympatoadrenal and glucagon responses (15, 29, 30). In this study we showed that exercise-induced release of β-endorphin is associated with the impaired sympathoadrenal counterregulation in response to subsequent hypoglycemia, similar to the effect induced by antecedent hypoglycemia. The attenuated hormonal response to repeated hypoglycemia has also been shown to improve with the administration of a nonselective opioid antagonist, naloxone (14, 31, 32). This suggests that the mechanism for HAAF and EAAF is likely the same and is mediated, at least in part, by β-endorphin. In light of this evidence, β-endorphin emerges as a potential, hitherto unrecognized, pathogenic factor in the development of exercise-associated HAAF.

The effects of β-endorphin in HAAF and EAAF are presumably mediated via the δ-, κ-, and μ-opioid receptors, which have been localized to various areas in the hypothalamus responsible for glucose sensing, including the ventromedial hypothalamus and the arcuate nucleus (33–36), thus suggesting that β-endorphin acts centrally to induce HAAF and EAAF. Repeated bouts of hypoglycemia result in suppression of genes in the hypothalamic neurons that inhibit glycolysis and stimulate fatty acid oxidation, including Pdk4, Angpt14, Cpt1a, and Gpd1 genes, which leads to the inability to shift from glycolysis toward alternate fuel use at the time of hypoglycemia (37) and may contribute to impaired recovery from hypoglycemia. However, this gene suppression is reversible with naloxone (37), thus implying that endogenous opioids are the mediators of these responses. Whether this gene regulation results directly from the action of endogenous opioids on the hypothalamic neurons or indirectly via opioid effect on sympathoadrenal system remains to be elucidated.

At present, there is no precise method to estimate β-endorphin concentrations in the hypothalamus compared with periphery, which is a limitation to the study. However, given that naloxone is known to cross the blood-brain barrier and has been shown to alter hypothalamic gene expression in response to hypoglycemia (37), β-endorphin is likely to be affecting hypoglycemic counterregulation centrally. However, β-endorphin is not the only endogenous opioid released in response to exercise, with β-lipotropin being another (38); thus, the role of the other endogenous opioids on hypoglycemia counterregulation should be explored. Nonetheless, the administration of exogenous β-endorphin directly into the rat brain inhibited some of the hypothalamic responses to hypoglycemia (39), thus implicating β-endorphin, at least in part, in central nervous system hypoglycemia counterregulation.

Opioid antagonists have been shown to enhance not only the autonomic/sympatoadrenomedullary response during exercise but also during subsequent episodes of hypoglycemia (14) in healthy people, making it an attractive target for study in the prevention of exercise-associated HAAF in T1DM. Although naloxone infusion during hypoglycemia attenuated HAAF in a group with T1DM (32), the exercise-induced release of β-endorphin in T1DM patients may be blunted (40); therefore, further inquiry is required into the effect of opioid antagonism during exercise in this population.

In summary, exercise-induced release of β-endorphin is associated with subsequent development of impaired hypoglycemia counterregulation. In light of the hitherto accumulated evidence and this newly acquired knowledge, opioid inhibition could now be considered for evaluation as a potential therapy for EAAF in T1DM.