Release of viruses and viral DNA from nucleus to cytoplasm of HeLa cells at late stages of productive adenovirus infection as revealed by electron microscope in situ hybridization

Considerable progress has been made over the past 10 years towards a full understanding of the functional significance of the structural changes resulting from the production of adenoviruses in permissive cells. Similarly, the host-virus interactions which are involved in viral replication and gene expression as well as in RNA nuclear export have been investigated. Post-embedding nonisotopic in situ hybridization has been proven to be a powerful tool for the study of nucleic acids in infected cells provided that controlled elimination of artifacts by appropriate treatments was undertaken. Adenovirus infected cells present two biological characteristics which could lead to false positive or negative results. First, they contain large amounts of single-stranded portions of viral DNA which are revealed with viral RNA molecules. Second, DNA-binding proteins are present which hide some nucleic acid sequences. By using a DNA probe and appropriate variations in the experimental protocol, it is possible to reveal specifically different kinds of targets, simultaneously single-stranded DNA (ssDNA) and double-stranded DNA (dsDNA), or only ssDNA, or only dsDNA, or only RNA. By using two probes labeled with different haptens, biotine and digoxigenin, it is possible to detect concomitantly two nucleic acid targets and, therefore, to study their relationships. In order to gain insight into the changes in the nucleus before cell lysis and to improve our knowledge on the series of steps leading to the release of adenoviruses from the nucleus, examination of cells at 41 h post-infection and identification of structures containing adenoviral nucleic acids were undertaken. In addition to the ultrastructural changes and precise distribution of cellular DNA and viral nucleic acid molecules already described in cells up to 24 h post-infection (for a review, see Puvion E, Puvion-Dutilleul F (1996) Exp Cell Res 229, 217–225), new results were obtained. Routine observation revealed the presence of: i) viruses in the cytoplasm, some being located next to nuclear pores; ii) abnormally large portions of the nuclear envelope devoid of underlying condensed chromatin; iii) proliferation of either the inner nuclear membrane only or both membranes of the nuclear envelope; and iv) electron-opaque grains in the nuclear compartment involved in viral genome transcription, and also in the clusters of interchromatin granules known to contain mature viral messenger RNA (Bridge E et al (1996) J Cell Biol 135, 303–314). In situ hybridization revealed the presence of: i) dsDNA in the cytoplasmic viruses indicating that they were mature viruses; ii) free viral dsDNA and ssDNA molecules in the cytoplasm whereas host DNA remained confined at the nuclear border; and iii) viral RNA in the newly-described electron-opaque grains we call, therefore, viral-RNA containing grains. Immunodetection of bromodeoxyuridine (BrdU) incorporated into DNA in pulse and pulse-chase experiments allowed us to ascertain that cells at 41 h post-infection were truly living cells and that at least part of the newly-synthesized viral DNA migrated from the nucleus to the cytoplasm. Taken together, the data suggest that modifications of the nuclear matrix, cytoskeleton, and nucleo-cytoplasmic interactions might occur near the termination of adenovirus infection inducing the progressive release of viruses, viral dsDNA and ssDNA molecules in the cytoplasm. In addition, the observation of a new structural support for the intranuclear viral RNA in the clusters of interchromatin granules emphasizes the role of these cellular structures in the intranuclear trafficking of messenger RNA leading to the regulation of its nuclear export.