We studied the effect of arcuate nucleus (ARC) lesions induced pharmacologically by the perinatal treatment of monosodium l-glutamate (MSG) on the cardiovascular, metabolic, and behavioral responses to fasting. Saline and MSG-treated male Sprague-Dawley rats were instrumented with telemetry devices for measurement of mean arterial pressure (MAP) and heart rate (HR) and housed in room calorimeters at an ambient temperature (T(a)) of 23 degrees C for assessment of oxygen consumption (VO(2)). At baseline, controls and MSG-treated rats had similar MAP (control=95+/-4; MSG=91+/-2 mmHg), HR (control=323+/-4; MSG=324+/-2 bpm), and VO(2) (control=8.7+/-0.3; MSG=8.6+/-0.2 ml/min). There were no differences in fasting-induced reductions in body weight or in food intake upon refeeding. MSG-treatment significantly attenuated fasting-induced reductions in HR and VO(2). This effect was specific to reduced caloric availability, as MSG-treated rats exhibited intact capacity to both increase and decrease HR and VO(2) in response to cold (T(a)=15 degrees C) and to thermoneutrality (T(a)=30 degrees C). Additional studies were performed in saline- and MSG-treated rats chronically treated with beta(1)-adrenergic receptor blockade (atenolol) prior to and during fasting. In controls, the cardiovascular responses to fasting during beta(1)-blockade were blunted and generally mimicked the effects of MSG-treatment, while beta(1)-blockade had no additional effect on MSG-treated rats. The results are consistent with the hypothesis that ARC neuronal signaling is requisite for intact homeostatic responses to fasting and may participate in fasting-induced withdrawal of cardiac sympathetic activity.