We have studied the adenovirus-induced inhibition of host cell protein synthesis and the effect of infection on the overall metabolism of host cell mRNA during the late phase of adenovirus infection by following the fate of a number of cellular mRNAs complementary to specific cloned DNA segments. At a time in infection when the rate of total cellular protein synthesis is drastically (>90%) reduced, transcription of specific cellular genes is undiminished. However, the transport of newly synthesized cellular mRNA to the cytoplasm is greatly decreased. This decreased appearance of new mRNA in the cytoplasm cannot account for the observed cessation of cell specific protein synthesis, however, since the concentration of several preexisting cellular mRNAs, including the mRNA for actin, remains unchanged throughout the course of infection. The preexisting mRNA is intact, capped, and functional as judged by its ability to direct protein synthesis in vitro in a cap-dependent fashion. The interruption in host translation appears to operate at the level of initiation directly, since we find that fewer ribosomes are associated with a given cellular mRNA after infection than before infection. Furthermore, the in vivo inhibition of cellular protein synthesis does not appear to be the result of competition with viral mRNA, since conditions which prevent the efficient initiation of translation of viral mRNA (infection with a viral mutant) do not result in the recovery of cell translation. Thus, it appears that a late adenovirus gene product directly mediates a shutoff of host protein synthesis.The interaction of a virus with its host cell often leads to an alteration in the synthesis and metabolism of macromolecules in the cell. The study of these alterations offers the opportunity to gain information concerning the normal course of events within the cell. For example, many different cytolytic virus infections result in decreased host protein synthesis at the same time that viral protein synthesis increases (22). The molecular basis for this viral "takeover" is best understood for poliovirus (9,19,39,42) and reovirus (33) infections, where differentiation between capped and uncapped mRNAs allows discrimination between viral and host mRNAs. However, the long-standing observation that host protein synthesis is greatly depressed late in adenovirus infection (2, 4, 5) cannot be easily explained; adenovirus mRNAs are all capped (12,23,35,45) and polyadenylated [poly(A)+] (28) and therefore would appear to be indistinguishable from host mRNAs on a gross level (6,32). Virtually all of the newly labeled mRNA that reaches the polyribosomes is virus specific (5,20,27), suggesting that transcription of cellular genes may be inhibited. However, neither the rate of synthesis of specific cellular mRNAs nor the fate of preexisting cellular mRNA has been examined with cloned cell DNA sequences.Beltz and Flint (5) approached the question of host cell mRNA manufacture and host protein synthesis by using cDNA hybridization to total cellular mRNA...