Cardiac myocytes isolated from adult rats and cultured for up to 5 days in a defined serum- and 3,5,3'-triiodothyronine-(T3) free medium were processed for in situ hybridization using [35S]cRNA probes specific for alpha- or beta-myosin heavy chain (MHC) mRNAs. A computer-assisted image analysis system was used to quantitate the hybridization signals within individual myocytes (100 cells/experimental point). The method was validated by comparison with dot-blot quantitation. The mean alpha-MHC mRNA density per cell decreased by 50% (P < 0.01) after 2 days in culture and remained stable thereafter, whereas the relative amount of beta-MHC mRNA did not increase until day 5. Addition of 10(-12) M T3 to the culture medium for 2 or 3 days was sufficient to maintain alpha-MHC mRNA levels similar to the day 0 values, whereas 10(-9) M T3 was necessary to completely inhibit beta-MHC mRNA expression. The independent analysis of myocytes exhibiting different morphological phenotypes with time in culture demonstrated that rounded myocytes contain relatively more alpha-MHC mRNA and were as sensitive to T3 as their rod-shaped counterparts. Their beta-MHC RNA content was similar to that found in rod-shaped cells and was still depressed by T3. In conclusion, we show that 1) physiological doses of T3 are sufficient to maintain in vitro a MHC phenotype close to that observed in vivo in adult, 2) the dose responsiveness of adult myocytes to T3 differs from that reported in neonatal myocytes, and 3) the alpha-MHC mRNA content and the T3 sensitivity of spheroidal myocytes imply that there is no alteration in their state of maturation.