Moderate Recurrent Hypoglycemia Markedly Impairs Set-Shifting Ability in a Rodent Model: Cognitive and Neurochemical Effects

Abstract

Recurrent hypoglycemia (RH) is the major complication of intensive insulin treatment for diabetes mellitus. Of particular concern is the perceived potential for long-term impact of RH on cognition. Because diabetic patients have been reported to have deficits in mental flexibility and judgment, both generally considered to be mediated predominantly by the prefrontal cortex, the purpose of the present study was to determine whether RH would affect prefrontal cortex function. Medial prefrontal cortex (mPFC)-mediated set-shifting ability was tested in male Sprague-Dawley rats using a maze-based, food-reward Set-Shift task analogous to the Wisconsin card-sorting task. The performance measure was the number of trials to criterion on both day 1 (initial rule-learning) and day 2 (set-shifting in response to a changed contingency). In vivo microdialysis was used to measure mPFC extracellular glucose, lactate, pyruvate, glutamate, and dopamine. Post-mortem measures within the mPFC included glucose transporter 3 (GluT3) and c-Fos. RH animals had enhanced performance on day 1, consistent with previous work that showed RH to improve subsequent hippocampal function when euglycemic. The key finding of the present work is that RH led to impaired set-shifting performance on day 2, suggesting impairment in e.g. mental flexibility. Consistent with this finding, RH animals show decreased mPFC glycolysis on day 2 compared to controls. Our data show that RH can lead to subsequent impaired judgment, accompanied by reduced prefrontal cortex function. The findings suggest a potential underlying mechanism for the impaired judgment seen in diabetic patients.

Keywords: Dopamine; glucose; medial prefrontal cortex; recurrent hypoglycemia; set-shifting ability.

Fig. 1

Behavioral performance on the Set-Shift task on day 1 and day 2 of testing in control animals (CTL) and in animals induced with recurrent bouts of hypoglycemia (RH) on each of preceding 3 days of testing. A previous history of recurrent hypoglycemia aided acquisition learning on day 1, however impaired mental flexibility, tested on day 2 in RH group. Data are means + SEM. Asterisks indicate significantly different behavioral performance between the treatment groups (p<0.05).

Fig. 2

Medial prefrontal cortex (mPFC) extracellular fluid (ECF) glucose levels, measured before, during, and after Set-Shift testing of control animals (CTL) and animals with prior history of recurrent hypoglycemia (RH) on day 1 (A) and day 2 (B) of the behavior task. ECF samples were collected in 20 min bins. Data are expressed as percentage of baseline + SEM. Asterisks indicate significantly different group findings (p<0.05).

Fig. 3

Medial prefrontal cortex (mPFC) extracellular fluid (ECF) pyruvate levels, measured before, during, and after Set-Shift testing of control animals (CTL) and animals with prior history of recurrent hypoglycemia (RH) on day 1 (A) and day 2 (B) of the behavior task. ECF samples were collected in 20 min bins. Data are expressed as percentage of baseline + SEM. Asterisks indicate significantly different group findings (p<0.05).

Fig. 4

Fig. (4A). The mean (+ SEM) number of stained nuclei for c-Fos protein within the medial prefrontal cortex (mPFC) at the end of Set-Shift testing in control animals (CTL) and in animals induced with recurrent bouts of hypoglycemia on each of preceding 3 days of testing (RH). No significant differences were observed in the number of c-fos stained cells between the treatment groups. Fig. (4B). A representative medial prefrontal cortex (mPFC) c-Fos immunostaining from CTL and RH animals. Bar = 100μm.

Fig. 4

Fig. (4A). The mean (+ SEM) number of stained nuclei for c-Fos protein within the medial prefrontal cortex (mPFC) at the end of Set-Shift testing in control animals (CTL) and in animals induced with recurrent bouts of hypoglycemia on each of preceding 3 days of testing (RH). No significant differences were observed in the number of c-fos stained cells between the treatment groups. Fig. (4B). A representative medial prefrontal cortex (mPFC) c-Fos immunostaining from CTL and RH animals. Bar = 100μm.

Fig. 5

Fig. (5A). The mean (+ SEM) staining intensity of glucose transporter 3 (GluT3) immunoreactivity (quantified by thresholded pixels) within the medial prefrontal cortex (mPFC) at the end of Set-Shift testing in control animals (CTL) and in animals induced with recurrent bouts of hypoglycemia on each of preceding 3 days of testing (RH). No significant differences were observed in the mean staining intensity of GluT3 between the treatment groups. Fig. (5B). A representative medial prefrontal cortex (mPFC) GluT3 immunostaining from CTL and RH animals. Bar = 10μm.

Fig. 5

Fig. (5A). The mean (+ SEM) staining intensity of glucose transporter 3 (GluT3) immunoreactivity (quantified by thresholded pixels) within the medial prefrontal cortex (mPFC) at the end of Set-Shift testing in control animals (CTL) and in animals induced with recurrent bouts of hypoglycemia on each of preceding 3 days of testing (RH). No significant differences were observed in the mean staining intensity of GluT3 between the treatment groups. Fig. (5B). A representative medial prefrontal cortex (mPFC) GluT3 immunostaining from CTL and RH animals. Bar = 10μm.