Caveolae are associated with molecules crucial for calcium handling. This review considers the roles of caveolae in calcium handling for smooth muscle and interstitial cells of Cajal (ICC). Structural studies showed that the plasma membrane calcium pump (PMCA), a sodium-calcium exchanger (NCX1), and a myogenic nNOS appear to be colocalized with caveolin I, the main constituent of these caveolae. Voltage dependent calcium channels (VDCC) are associated but not co-localized with caveolin 1, as are proteins of the peripheral sarcoplasmic reticulum (SR) such as calreticulin. Only the nNOS is absent from caveolin 1 knockout animals. Functional studies in calcium free media sugest that a source of calcium in tonic smooth muscles exists, partly sequestered from extracellular EGTA. This source supported sustained contractions to carbachol using VDCC and dependent on activity of the SERCA pump. This source is postulated to be caveolae, near peripheral SR. New evidence, presented here, suggests that a similar source exists in phasic smooth muscle of the intestine and its ICC. These results suggest that caveolae and peripheral SR are a functional unit recycling calcium through VDCC and controlling its local concentration. Calcium handling molecules associated with caveolae in smooth muscle and ICC were identified and their possible functions also reviewed.

We studied pacing and neurotransmission in longitudinal (LM) and circular muscle (CM) in intestine of W/W++ and W/W(V) mice. Electrical field-stimulation (EFS) of nerves in LM segments was more inhibitory in W/W(V) mice than in W/W++ mice. No inhibitory input to CM segments of W/W(V) mice was found. The EFS, after nerve block, entrained segments of both W/W++ and mutant mice with 10 ms pulses, and entrained those of mutant mice more readily at 1 and 3 ms pulses. Pacing with external electrodes did not depend on interstitial cells of Cajal in the myenteric plexus (ICC-MP). 2-Aminoethoxydiphenyl borate (2-APB), putative antagonist at IP3 receptors, store-operated channels and the Sacro-endoplasmic reticulum Ca2+ ATPase pump, reduced frequency and amplitudes of pacing of LM segments from W/W(V) mice as it did in BALB/c mice. Thus, its actions may not require ICC-MP. SKF 96365, a putative inhibitor of store-operated channels, reduced frequencies and amplitudes of intestinal segments in W/W++ mice at 10 or 30 micromol L-1. This resulted from blocking L-Ca2+-channels. Thus, no evidence was found that store-operated channels play a role in pacing. In LM segments of W/W(V), SKF 96365 had no effects on frequency of contractions. We conclude, results from models of severely reduced systems may not be applicable to intact ICC networks.