Animal studies have shown that electroacupuncture (EA) at Zusanli (ST-36) and Zhongwan (CV-12) acupoints reduces plasma glucose concentrations in rats with type II diabetes. However, whether EA reduces plasma glucose levels in type I diabetes is still unknown. In this study, we explore the various non-insulin-dependent pathways involved in EA-induced lowering of plasma glucose. Streptozotocin (STZ) (60 mg kg−1, i.v.) was administered via the femoral vein to induce insulin-dependent diabetes in non-adrenalectomized and in adrenalectomomized rats. EA (15 Hz) was applied for 30 min to bilateral ST-36 acupoints after administration of Atropine (0.1 mg kg−1 i.p.), Eserine (0.01 mg kg−1 i.p.), or Hemicholinium-3 (5 μg kg−1 i.p.) in non-adrenalectomized rats. Rats administered acetylcholine (0.01 mg kg−1 i.v.) did not undergo EA. Adrenalectomized rats underwent EA at bilateral ST-36 acupoints without further treatment. Blood samples were drawn from all rats before and after EA to measure changes in plasma glucose levels. Expression of insulin signaling proteins (IRS1, AKT2) in atropine-exposed rats before and after EA was measured by western blot. Atropine and hemicholinium-3 completely blocked the plasma glucose lowering effects of EA, whereas eserine led to a significant hypoglycemic response. In addition, plasma glucose levels after administration of acetylcholine were significantly lower than the fasting glucose levels. In STZ-adrenalectomized rats, EA did not induce a hypoglycemic response. EA stimulated the expression of IRS1 and AKT2 and atropine treatment blocked the EA-induced expression of those insulin signaling proteins. Taken together, EA at the ST-36 acupoint reduces plasma glucose concentrations by stimulating the cholinergic nerves.