Previous studies have shown that the larval epidermis of the tobacco hornworm, Manduca sexta, contains a 29 kDa nuclear protein (JP29) that binds pothoaffinity analogs of juvenile hormone (JH), but does not bind JH I with high affinity. We now find that JP29 is also associated with the insecticyanin granules, and we show that JP29 mRNA is regulated in a complex fashion by both 20-hydroxyecdysone (20E) and JH. Studies with day 2 fourth instar larval epidermis in vitro showed that a molting concentration 12 μg/ml) of 20E caused the disappearance of JP29 mRNA, irrespective of the presence or absence of JH; this effect was dependent on the concentration of 20E (ED50=200 ng/ml). The reappearance of JP29 mRNA around the time of ecdysis required the presence of JH at head capsule slippage (HCS), since little appeared in larvae allatectomized about 6 h before HCS unless JH I was applied at the time of HCS. Maintenance of JP29 mRNA in fifth instar epidermis also required the continued presence of JH in both isolated abdomens and in vitro. Culture of either day 1 or day 2 fifth instar epidermis without hormones for 24 h caused decline of JP29 mRNA, which was accelerated by 20E in a concentration-dependent manner (ED50 = 30 and 10 ng/ml 20E respectively). When day 2 epidermis was exposed to 500 ng/ml 20E for 24 h to cause pupal commitment, JP29 mRNA disappeared. Neither methoprene nor JH I (in either the presence or the absence of the esterase inhibitor O-ethyl, S-phenyl phosphamidethiolate [EPPAT]) was able to prevent this loss, although both slowed its rate. The mRNA for the larval cuticle protein LCP14 was found to be regulated similarly to that for JP29 by 20E, but differently by JH. The JP29 protein was relatively long-live, persisting after the disappearance of its mRNA for at least 19 h during the larval molt and for more than 24 h in vitro. Although trace amounts of JP29 are found for the first 12 h after pupal ecdysis, injection of 5 μg JH II into pupae during the critical period to cause the synthesis of a second pupal cuticle had no effect on the amount of JP29 present. Thus, although the presence of JP29 in larval epidermis is associated with and dependent on JH, high amounts are not associated with the “status quo” action of JH on the pupa. The role of this protein consequently remains obscure. Arch. Insect Biochem. Physiol. 34:409–428, 1997. © 1997 Wiley-Liss, Inc.