Superfusion with low glucose solution increases catecholamine release from type I cells of the rat carotid body (Pardal & Lopez-Barneo, 2002) and insulin infusion increases baseline ventilation in a carotid body-dependent manner (Bin-Jaliah & Kumar, 2003). Whether this latter effect was due to hypoglycaemia per se or hypermetabolism was not determined. In this study, we hypothesized that insulin infusion should, like hypoxia, increase CO₂ chemosensitivity in vivo and we have tested the direct effect of low glucose on an in vitro carotid body preparation.

Ventilatory CO₂ chemosensitivity was measured from integrated airflow using a modified rebreathing technique (Read, 1967) in hyperoxia (P_a,J{special} > 300 mmHg) in adult Wistar rats (300-350 g), anaesthetized with urethane (650 mg kg^-1, I.V.). Chemoreceptor afferent discharge was recorded in vitro from carotid bodies (Pepper et al. 1995) isolated from adult Wistar rats (120-150 g) anaesthetized with halothane (2-3 % in O₂). All animals were humanely killed at the end of the experiment. Data are expressed as means ± S.E.M. and significance (P < 0.05) was tested with ANOVA and, as appropriate, the post hoc Bonferroni/Dunn test.

Insulin infusion (0.4 U min^-1 kg^-1) lowered blood glucose from 6.7 ± 0.1 to 3.3 ± 0.1 mmol l^-1 (P < 0.0001) and in sham-operated, control animals (n = 6) increased CO₂ chemosensitivity from 12.58 ± 1.10 to 19.19 ± 1.51 ml min^-1 kg^-1 mmHg^-1 (P < 0.01). In contrast, CO₂ chemosensitivity in animals with bilateral carotid sinus nerve section (n = 6) remained unchanged (12.12 ± 0.92 to 13.78 ± 0.95 ml min^-1 kg^-1 mmHg^-1; P > 0.25). Chemoafferent recordings from few-fibre preparations of the carotid sinus nerve (n = 5) showed that, whilst all fibres responded to falls in PJ{special} or elevations in P_CJ{special}, lowering superfusate glucose from 10 mM to 2 mM had no effect upon baseline discharge (0.45 ± 0.26 to 0.36 ± 0.20 Hz, respectively; P > 0.30) or upon CO₂ chemosensitivity measured as Δ discharge between 40 and 80 mmHg P_CO2 at a PJ{special} of ca 400 mmHg (P > 0.15).

These data demonstrate that a lowered glucose concentration does not increase baseline carotid body chemoafferent discharge or CO₂ chemosensitivity during hyperoxia in vitro and suggest that the in vivo effects of insulin infusion upon baseline ventilation and CO₂ chemosensitivity are, therefore, more likely due to its effects upon metabolism rather than upon glucose concentration.

We acknowledge the financial support of King Khalid University, Saudi Arabia and the British Heart Foundation.