To elucidate the role for transforming growth factor-beta isoforms (beta(1), -beta(2), and -beta(3)) in wound repair, we used isoform-specific antibodies to detect the spatial and temporal expression of the latent and mature/active transforming growth factor-beta isoforms by immunohistochemical localization through 21 days after excisional and incisional wounding of ovine skin. Although incisional and excisional wounds showed similar patterns of transforming growth factor-beta immunoreactivity, we found a differential temporal and spatial expression of the latent and mature transforming growth factor-beta isoforms throughout wound repair. Specifically, 1 day after wounding, there was a marked increase in transforming growth factor-beta isoforms in the epithelium adjacent to the wound, epidermal appendages, and the cells and matrix of the granulation tissue. At this time, transforming growth factor-beta(3) isoform was the most abundant. Most notably, the epidermis adjacent to the wound was intensely immunoreactive for all transforming growth factor-beta isoforms 1 day after injury. However, the migrating epithelium, derived from both the hair follicles and the wound margins, was completely devoid of immunoreactive transforming growth factor-beta until reepithelialization was complete. Within the inflammatory exudate, there was a distinct band of leukocytes that was immunoreactive for transforming growth factor-beta(2) and -beta(3) 1 day after injury and 1 day later for transforming growth factor-beta(1). Although transforming growth factor-beta(1) and -beta(2), latent transforming growth factor-beta(2), transforming growth factor-beta(3), and latent transforming growth factor-beta(3) immunostaining was present in the numerous fibroblasts and other dermal cells, latent transforming growth factor-beta(1) was only associated with the extracellular matrix. In general, immunoreactivity remained high until day 7 after wounding and slowly subsided over time. However, by day 21, immunostaining had not returned to normal and the original wound was replete with immunoreactive fibroblasts and a dense, immunostained extracellular matrix. Thus, although the dynamic presence of transforming growth factor-beta isoforms exemplifies its positive role in the wound repair process, its persistence together with its known potent effects on matrix accumulation, supports its role in scar formation.