In a study into possible mechanisms of desensitisation to arginine vasopressin (AVP), it was discovered serendipitously that [Ca²⁺]_i increases in response to caffeine were enhanced following exposure to AVP. We now report investigations designed to characterize this effect and the signalling pathway responsible for it. Eyes were removed from freshly killed pigs at the local abattoir. Arterioles were mechanically dispersed from the retinas using a fire-polished Pasteur pipette and microvascular smooth muscle (MVSM) cells were loaded with 10mM Fura-2AM for 2 hours. Changes in [Ca²⁺]_i were recorded from fragments of intact arteriole using ratiometric microfluorimetry. Caffeine (10 mM) was used to release stored Ca²⁺ and the resulting increase in the fluorescence ratio used as a measure of the increase in [Ca²⁺]_i. Caffeine responses were compared in the same vessels before and after treatment with AVP or other agents under test, and the statistical significance of any changes assessed ANOVA with the Tukey Kramer post-hoc test unless otherwise indicated. Caffeine responses were increased by 52±20% (mean±SEM) 10 min after exposure to AVP (10nM) (P<0.005, paired t test, n=17). When caffeine was applied a third time, some 90 min after washout of AVP, the response was still 55±23% greater than the control caffeine response before AVP in the same vessels (P<0.05, n=6). There were no significant changes in caffeine responses when the drug was applied repeatedly in the absence of AVP (n=8). Previous studies in vascular smooth muscle have shown that cAMP can promote store filling (Porter et al. 1998). Experiments were carried out to determine whether this was the possible signalling pathway responsible in this case. Ten minutes exposure to forskolin, which elevates cAMP levels, increased caffeine responses by 30±10%, (P<0.05, n=12). When tissues were pre-incubated for 1-2 h with Rp-8-HA-cAMPS (10mM), a cell permeant inhibitor of protein kinase A, control caffeine responses remained brisk, but there was no enhancement of these responses after treatment with either AVP (n=19) or forskolin (n=12). We conclude that AVP, a physiological agonist in retinal microvessels, also stimulates enhancement of store loading via a cAMP, PKA-dependent mechanism in these vessels.