A small molecule exerts selective antiviral activity by targeting the human cytomegalovirus nuclear egress complex.

Human cytomegalovirus (HCMV) is an important pathogen for which new antiviral drugs are needed. HCMV, like other herpesviruses, encodes a nuclear egress complex (NEC) composed of two subunits, UL50 and UL53, whose interaction is crucial for viral replication. To explore whether small molecules can exert selective antiviral activity by inhibiting NEC subunit interactions, we established a homogeneous time-resolved fluorescence (HTRF) assay of these interactions and used it to screen >200,000 compound-containing wells. Two compounds, designated GK1 and GK2, which selectively inhibited this interaction in the HTRF assay with GK1 also active in a co-immunoprecipitation assay, exhibited more potent anti-HCMV activity than cytotoxicity or activity against another herpesvirus. At doses that substantially reduced HCMV plaque formation, GK1 and GK2 had little or no effect on the expression of viral proteins and reduced the co-localization of UL53 with UL50 at the nuclear rim in a subset of cells. GK1 and GK2 contain an acrylamide moiety predicted to covalently interact with cysteines, and an analog without this potential lacked activity. Mass spectrometric analysis showed binding of GK2 to multiple cysteines on UL50 and UL53. Nevertheless, substitution of cysteine 214 of UL53 with serine (C214S) ablated detectable inhibitory activity of GK1 and GK2 in vitro, and the C214S substitution engineered into HCMV conferred resistance to GK1, the more potent of the two inhibitors. Thus, GK1 exerts selective antiviral activity by targeting the NEC. Docking studies suggest that the acrylamide tethers one end of GK1 or GK2 to C214 within a pocket of UL53, permitting the other end of the molecule to sterically hinder UL50 to prevent NEC formation. Our results prove the concept that targeting the NEC with small molecules can selectively block HCMV replication. Such compounds could serve as a foundation for development of anti-HCMV drugs and as chemical tools for studying HCMV.

Human Cytomegalovirus Nuclear Egress Complex Subunit, UL53, Associates with Capsids and Myosin Va, but Is Not Important for Capsid Localization towards the Nuclear Periphery.

After herpesviruses encapsidate their genomes in replication compartments (RCs) within the nuclear interior, capsids migrate to the inner nuclear membrane (INM) for nuclear egress. For human cytomegalovirus (HCMV), capsid migration depends at least in part on nuclear myosin Va. It has been reported for certain herpesviruses that the nucleoplasmic subunit of the viral nuclear egress complex (NEC) is important for this migration. To address whether this is true for HCMV, we used mass spectrometry and multiple other methods to investigate associations among the HCMV NEC nucleoplasmic subunit, UL53, myosin Va, major capsid protein, and/or capsids. We also generated complementing cells to derive and test HCMV mutants null for UL53 or the INM NEC subunit, UL50, for their importance for these associations and, using electron microscopy, for intranuclear distribution of capsids. We found modest associations among the proteins tested, which were enhanced in the absence of UL50. However, we found no role for UL53 in the interactions of myosin Va with capsids or the percentage of capsids outside RC-like inclusions in the nucleus. Thus, UL53 associates somewhat with myosin Va and capsids, but, contrary to reports regarding its homologs in other herpesviruses, is not important for migration of capsids towards the INM.

Linking indirect effects of cytomegalovirus in transplantation to modulation of monocyte innate immune function.

Cytomegalovirus (CMV) is an important cause of morbidity and mortality in the immunocompromised host. In transplant recipients, a variety of clinically important "indirect effects" are attributed to immune modulation by CMV, including increased mortality from fungal disease, allograft dysfunction and rejection in solid organ transplantation, and graft-versus-host-disease in stem cell transplantation. Monocytes, key cellular targets of CMV, are permissive to primary, latent and reactivated CMV infection. Here, pairing unbiased bulk and single cell transcriptomics with functional analyses we demonstrate that human monocytes infected with CMV do not effectively phagocytose fungal pathogens, a functional deficit which occurs with decreased expression of fungal recognition receptors. Simultaneously, CMV-infected monocytes upregulate antiviral, pro-inflammatory chemokine, and inflammasome responses associated with allograft rejection and graft-versus-host disease. Our study demonstrates that CMV modulates both immunosuppressive and immunostimulatory monocyte phenotypes, explaining in part, its paradoxical "indirect effects" in transplantation. These data could provide innate immune targets for the stratification and treatment of CMV disease.

A Role for Myosin Va in Human Cytomegalovirus Nuclear Egress.

Herpesviruses replicate and package their genomes into capsids in replication compartments within the nuclear interior. Capsids then move to the inner nuclear membrane for envelopment and release into the cytoplasm in a process called nuclear egress. We previously found that nuclear F-actin is induced upon infection with the betaherpesvirus human cytomegalovirus (HCMV) and is important for nuclear egress and capsid localization away from replication compartment-like inclusions toward the nuclear rim. Despite these and related findings, it has not been shown that any specific motor protein is involved in herpesvirus nuclear egress. In this study, we have investigated whether the host motor protein, myosin Va, could be fulfilling this role. Using immunofluorescence microscopy and coimmunoprecipitation, we observed associations between a nuclear population of myosin Va and the viral major capsid protein, with both concentrating at the periphery of replication compartments. Immunoelectron microscopy showed that nearly 40% of assembled nuclear capsids associate with myosin Va. We also found that myosin Va and major capsid protein colocalize with nuclear F-actin. Importantly, antagonism of myosin Va with RNA interference or a dominant negative mutant revealed that myosin Va is important for the efficient production of infectious virus, capsid accumulation in the cytoplasm, and capsid localization away from replication compartment-like inclusions toward the nuclear rim. Our results lead us to suggest a working model whereby human cytomegalovirus capsids associate with myosin Va for movement from replication compartments to the nuclear periphery during nuclear egress.IMPORTANCE Little is known regarding how newly assembled and packaged herpesvirus capsids move from the nuclear interior to the periphery during nuclear egress. While it has been proposed that an actomyosin-based mechanism facilitates intranuclear movement of alphaherpesvirus capsids, a functional role for any specific myosin in nuclear egress has not been reported. Furthermore, the notion that an actomyosin-based mechanism facilitates intranuclear capsid movement is controversial. Here we show that human cytomegalovirus capsids associate with nuclear myosin Va and F-actin and that antagonism of myosin Va impairs capsid localization toward the nuclear rim and nuclear egress. Together with our previous results showing that nuclear F-actin is induced upon HCMV infection and is also important for these processes, our results lend support to the hypothesis that nascent human cytomegalovirus capsids migrate to the nuclear periphery via actomyosin-based movement. These results shed light on a poorly understood viral process and the cellular machinery involved.