We have demonstrated previously that all-trans retinoic acid (tRA) stimulates collagen synthesis in ultraviolet B (UVB)-exposed but not in nonirradiated hairless mouse skin, suggesting that UVB exposure is required to prime certain tRA-induced events. The current study investigated further whether this tRA stimulatory effect on collagen synthesis could be regulated quantitatively by the accumulative dose of UVB. Mice were irradiated with the total amount of 0.25 (low), 0.77 (medium) or 1.43 J/cm2 (high) UVB radiation over 2, 5 and 10 weeks, respectively. Post-irradiation, 50 microliters of 0.1% tRA was applied topically 5 times a week. An initial delay was observed before tRA-induced collagen stimulation was detected in all 3 treatment groups. The first statistically significant increase in collagen synthesis was observed at the 6th week of tRA treatment in the high-UVB group. In the groups receiving medium and low UVB doses, onset occurred at the 8th week. Furthermore, at the end of tRA treatment, the degree of collagen stimulation correlated positively with the total UVB dose applied (increasing 183%, 224% and 250% over the vehicle control in the low-, medium- and high-UVB groups). By contrast, no collagen stimulation was detectable in the non-irradiated animals treated with tRA for 10 weeks, as previously reported. Although the mechanism by which UVB exposure primes skin to respond to tRA for collagen synthesis still remains unknown, the data indicate that the accumulative UVB dose positively affects the length of the initial delay and the magnitude of the stimulatory effect of tRA on collagen synthesis in photodamaged skin.