Abstract

Herpes simplex virus 2 (HSV-2) establishes latent infection in dorsal root ganglion (DRG) neurons after productive (lytic) infection in peripheral tissues. A neuron-specific microRNA, miR-138, favors HSV-1 latency by repressing viral ICP0 and host Oct-1 and Foxc1 genes, yet the role of miR-138 in HSV-2 infection was unknown. The ICP0 mRNAs of HSV-1, HSV-2, and chimpanzee herpesvirus each have one to two canonical miR-138 binding sites. The sites are 100% conserved in 308 HSV-1 and 300 HSV-2 published sequences of clinical isolates. In cotransfection assays, miR-138 repressed HSV-2 ICP0 expression through the seed region and surrounding interactions that are different from HSV-1. An HSV-2 mutant with disrupted miR-138 binding sites on ICP0 showed increased ICP0 expression in Neuro-2a cells. Photoactivatable ribonucleoside-enhanced cross-linking and immunoprecipitation confirmed miR-138 binding to HSV-2 ICP0 and identified UL19 and UL20 as additional targets whose expression was repressed by miR-138 during cotransfection. In Neuro-2a cells, transfected miR-138 and its antagomir decreased and increased HSV-2 replication, respectively, and a knockout experiment showed that miR-138’s host targets OCT-1 and FOXC1 were important for HSV-2 replication. In primary mouse DRG neurons, both ICP0 and FOXC1 positively regulated HSV-2 replication, but both overexpressed and endogenous miR-138 suppressed HSV-2 replication primarily by repressing ICP0 expression. Thus, miR-138 can suppress HSV-2 neuronal replication through multiple viral and host pathways. These results reveal functional similarities and mechanistic differences in how miR-138 regulates HSV-1 and HSV-2 infection and indicate an evolutionary advantage of using miR-138 to repress lytic infection in neurons.IMPORTANCE HSV-1 and HSV-2 are closely related viruses with major differences. Both viruses establish latency in ne