Herpes simplex virus 1 (HSV-1) is a neurotropic virus that establishes a life-long latent infection in the trigeminal ganglia (TG) of the infected human host. From time to time the virus asymptomatically or symptomatically reactivates from the latency stage producing epithelial lesions, most of the time on the face but also in the eye, inducing severe pathologies such as keratitis. HSV-1 infection is also associated with pathologies of the central nervous system such as encephalitis.
During latency the virus is in a transcriptionally restricted state. Of the about 80 genes transcribed during lytic infection, only a family of long non-coding RNAs is produced abundantly during latency. These latency associated transcripts (LATs) arise from the transcription of an 8.3 kb primary RNA that is processed in two major LATs of 1.5 kb and 2 kb and several microRNAs with cellular and viral targets. The precise role of LATs is a matter of debate; however, a point of convergence among the many studies of LATs is that their initial production would favor the survival of the infected neurons and the coordination of the infectious process towards the latency transcriptional program and reactivation. The lytic cycle is the alternative transcriptional program and is characterized by a temporarily regulated transcriptional program, which starts with the expression of immediate early (IE), then early (E), and finally late (L) genes.
In the virus particle, HSV-1 genome is a 150-kb double stranded naked linear DNA. Upon entry into the host nucleus, the viral genome does not integrate, remaining as an extrachromosomal plasmid-like molecule. Chromatinization of the viral genome during latency plays a major regulatory role, and post-translational modifications of histones associated to key viral promoters determines the fate of the latency/reactivation process. Latent viral genomes are present in multiple copies within the nucleus of infected neurons in mouse models and human, and little is known about the molecular determinants that enable one neuron rather than another to sustain reactivation.
This paper describes a fluorescent in situ hybridization (FISH) approach combined with immunofluorescence to investigate the interaction between viral genomes and nuclear proteins within TG neurons of latently infected mice and during the whole process of latency establishment (from 4 to 28 days post infection, dpi). Viral genomes were found in neurons and satellite cells at 4 and 6 dpi, but only in neurons at > 6 dpi. In satellite cells, viral genomes showed only replication compartment (RC) patterns, whereas in neurons both RC and “multiple-acute” patterns were detected. From 4 to 14 dpi both patterns progressively disappeared, and transformed from14dpi onwards to the latency-associated “single” and “multiple-latency” patterns. Immuno-FISH analyses of human TG showed a close spatial distribution between latent HSV-1 genomes and PML protein in neurons, which suggests that, similar to the situation in the mouse model, HSV-1 latency in human is probably tightly linked to the activity of PML-NBs.
We have been trying for decades to understand the biology of latency in HSV. The strength of this approach is to point out that this cannot be achieved by looking at the virus alone, the biology of the host cell also has to be considered. This study describes the nuclear architecture and nuclear distribution of viral genomes as major determinants of HSV-1 latency. It confirms the close interrelation between PML-NBs and HSV-1 genomes in the establishment of latency through the formation of vDCP-NBs. Finally, it confirms that the spatial organization of HSV-1 genomes and PML is conserved in latently infected neurons in human TG, which indicates PML-NBs to be major HSV-1 genome interactants during latency and probably reactivation.
Latency Entry of Herpes Simplex Virus 1 Is Determined by the Interaction of Its Genome with the Nuclear Environment. (2016) PLoS Pathog 12(9): e1005834. doi: 10.1371/journal.ppat.1005834
Herpes simplex virus 1 (HSV-1) establishes latency in trigeminal ganglia (TG) sensory neurons of infected individuals. The commitment of infected neurons toward the viral lytic or latent transcriptional program is likely to depend on both viral and cellular factors, and to differ among individual neurons. In this study, we used a mouse model of HSV-1 infection to investigate the relationship between viral genomes and the nuclear environment in terms of the establishment of latency. During acute infection, viral genomes show two major patterns: replication compartments or multiple spots distributed in the nucleoplasm (namely “multiple-acute”). Viral genomes in the “multiple-acute” pattern are systematically associated with the promyelocytic leukemia (PML) protein in structures designated viral DNA-containing PML nuclear bodies (vDCP-NBs). To investigate the viral and cellular features that favor the acquisition of the latency-associated viral genome patterns, we infected mouse primary TG neurons from wild type (wt) mice or knock-out mice for type 1 interferon (IFN) receptor with wt or a mutant HSV-1, which is unable to replicate due to the synthesis of a non-functional ICP4, the major virus transactivator. We found that the inability of the virus to initiate the lytic program combined to its inability to synthesize a functional ICP0, are the two viral features leading to the formation of vDCP-NBs. The formation of the “multiple-latency” pattern is favored by the type 1 IFN signaling pathway in the context of neurons infected by a virus able to replicate through the expression of a functional ICP4 but unable to express functional VP16 and ICP0. Analyses of TGs harvested from HSV-1 latently infected humans showed that viral genomes and PML occupy similar nuclear areas in infected neurons, eventually forming vDCP-NB-like structures. Overall our study designates PML protein and PML-NBs to be major cellular components involved in the control of HSV-1 latency, probably during the entire life of an individual.