Yasushi Kawaguchi

Viral replication exhibits substantial heterogeneity among individual cells. Although temporally ordered gene-expression cascades are thought to contribute to such heterogeneity in DNA viruses, whether and how their tempo is regulated, and how cell-to-cell variability shapes population-level infection outcomes, remain unknown. Here, we developed a high-throughput live-cell imaging platform to track HSV-1 gene-expression dynamics in individual cells and, in combination with single-cell RNA sequencing, identified calmodulin (CaM) as a host regulator of HSV-1 gene-expression tempo. CaM accelerates progression from immediate-early to early and late phases at the single-cell level by advancing their onset. By increasing the fraction of virus-producing cells without altering burst size, CaM reshapes the distribution of infection outcomes, thereby promoting population-level viral replication. Pharmacological inhibition of CaM protects mice from lethal infection, exceeding acyclovir. These findings establish CaM-mediated temporal control as a key determinant of population-level viral output and a regulator of infection heterogeneity.

Although numerous phosphorylation sites on herpes simplex virus 1 (HSV-1) proteins have been identified through phosphoproteomic analyses, their virological significance remains largely unclear. Here, we developed an evolution-guided prioritization strategy that integrates phosphoproteomic data with residue-level conservation across the genus Simplexvirus to identify phosphorylation sites as candidates for functional relevance. Applying this strategy to a previously reported phosphoproteomic data set, we identified 46 phosphorylation sites with a conservation level of 88% or higher. Among them, we focused on two phosphorylation sites, UL6 Tyr-234 and UL7 Tyr-89, which are conserved in 100% and 88% of Simplexvirus species, respectively. Phosphomimetic mutations at either site significantly reduced progeny virus yields in cultured cells, suggesting that phosphorylation at these sites can function as an inhibitory switch regulating HSV-1 replication. In agreement with this, in-depth analyses of phosphorylation at UL7 Tyr-89 during HSV-1 infection revealed that the phosphomimetic mutation led to phenotypes similar to those of the UL7 null mutation in virion morphogenesis, HSV-1 replication in the central nervous system (CNS) and eyes of mice, as well as CNS pathogenicity and ocular pathogenic manifestations. Notably, the non-phosphorylatable mutation at this site had little effect on HSV-1 infection in cultured cells, whereas it significantly reduced HSV-1 replication and pathogenicity in the CNS, but not in the eyes of mice. These results suggest that our strategy effectively prioritizes phosphorylation sites for functional investigation and that phosphorylation at UL7 Tyr-89 serves as a context-dependent inhibitory switch, fine-tuning UL7 activity in a tissue-specific manner, particularly in the CNS. IMPORTANCE: Intricate phosphorylation-dependent regulatory mechanisms enable viruses to diversify the functions of their proteins, profoundly shaping viral replication and pathogenicity. Although phosphoproteomic analyses have produced an expanding catalog of phosphorylation sites on viral proteins, a considerable proportion of these modifications are likely non-functional. This creates a pressing need for a prioritization strategy to predict functionally relevant phosphorylation sites. To address this, we developed an evolution-guided prioritization strategy that integrates phosphoproteomic data with genus-level conservation. Using this strategy, we prioritized two tyrosine phosphorylation sites for functional analysis. Notably, phosphorylation at one of these sites, UL7 Tyr-89, functions as a tissue-specific regulatory mechanism that fine-tunes UL7 activity and modulates herpes simplex virus 1 replication and pathogenicity in the central nervous system. This strategy provides a practical framework for prioritizing candidate regulatory sites for functional investigation in viral replication and pathogenicity.

Although previous studies reported that the herpes simplex virus 2 (HSV-2) UL13 protein kinase mediates the phosphorylation of elongation factor 1δ (EF-1δ) in infected cells, we found here that individual expression of UL13 was insufficient to induce phosphorylation of EF-1δ in mammalian cells. This led us to hypothesize that HSV-2 UL13 requires viral cofactors for full kinase activity and prompted us to identify such cofactors. Our results were as follows. (i) Co-expression of UL13 with UL55 or Us10 significantly enhanced phosphorylation of EF-1δ compared to UL13 alone. (ii) UL13 was co-precipitated with UL55 or Us10 upon co-expression, and its kinase activity was significantly increased in their presence, as demonstrated by in vitro kinase assays. (iii) In HSV-2-infected cells, UL13 was co-precipitated with Us10 and UL55. (iv) The UL55-null mutation significantly reduced phosphorylation of EF-1δ in HSV-2-infected cells, whereas the Us10-null mutation had little effect; however, the double-null mutation further decreased the phosphorylation compared to the UL55-null mutation alone. (v) The UL55-null mutation, but not the Us10-null mutation, significantly reduced HSV-2 replication and cell-cell spread in U2OS cells to levels comparable to those observed with the UL13 kinase-dead mutation. These results suggest that UL55 acts as a principal activator of UL13 in HSV-2-infected cells, whereas Us10 serves as an auxiliary activator. Moreover, the role of UL13 kinase activity in HSV-2 replication and cell-cell spread in U2OS cells appears to be largely dependent on UL55. IMPORTANCE: Herpesviruses encode conserved protein kinases (CHPKs) that often target cellular cyclin-dependent kinase (CDK) phosphorylation sites. CHPKs from beta- and gammaherpesviruses can exhibit these CDK-like functions even when individually expressed in mammalian cells. In contrast, CHPKs from alphaherpesviruses display these CDK-like functions in infected cells, but not upon individual expression, suggesting that they require additional viral factors to exhibit full kinase activity. In this study, we focused on HSV-2 UL13, an alphaherpesvirus CHPK, and identified HSV-2 UL55 and Us10 as viral activators of UL13. In HSV-2-infected cells, UL55 functions as a principal activator of UL13, while Us10 serves as an auxiliary activator. Importantly, the contribution of UL13 kinase activity to HSV-2 replication and cell-cell spread appears to be largely dependent on the presence of UL55. Our findings uncover a previously unrecognized mechanism of CHPK regulation in alphaherpesviruses and provide new insights into the evolutionary diversification of viral kinase control.

Herpesviruses replicate their genomes and package them into capsids within the host cell nucleus. These capsids must then translocate from the nucleus to the cytoplasm through a process designated nuclear egress. The virus-encoded nuclear egress complex (NEC), consisting of a nuclear matrix protein and a nuclear membrane protein, plays a crucial role in this process. Although the role of NEC for nucleo-cytoplasmic transport of capsids is conserved in Herpesviridae , some of the binding partners of the NEC components are specific for individual viruses. The NECs of alpha- and gammaherpesviruses recruit the Endosomal Sorting Complex Required for Transport III (ESCRT-III) to the inner nuclear membrane for efficient nuclear egress of capsids. In contrast, the role of ESCRT-III for nuclear egress of betaherpesviruses, including human cytomegalovirus (HCMV) and human herpesvirus 6A (HHV-6A), has not been elucidated. Here, we show that ESCRT-III is recruited to the nuclear rim in cells expressing the NEC of HCMV or of HHV-6A. Inhibition of ESCRT-III impaired HHV-6A replication and nuclear egress of the capsids. Mechanistically, ESCRT-III adaptor ALIX interacts with HHV-6A NEC membrane protein U34 and thus contributes to HHV-6A replication. From these observations, we conclude that, like at least some alpha- and gammaherpesviruses, HHV-6A NEC recruits ESCRT-III through ALIX to promote viral capsid nuclear egress. IMPORTANCE: ESCRT-III performs reverse-topology scission involved in many diverse cellular processes, including cytokinesis, endosome maturation, autophagy, membrane repair, and viral budding. Nucleo-cytoplasmic transport of herpesvirus capsids requires scission at the inner nuclear membrane. In alpha- and gammaherpesviruses, this process requires ESCRT-III, but it is not known whether this is also the case for betaherpesviruses. Here, we show that ESCRT-III is also important for nuclear egress of capsids of the betaherpesvirus human herpesvirus 6A. These results imply that ESCRT-III-mediated inner nuclear membrane scission is a conserved feature in the virion maturation process of Herpesviridae . Our findings thus suggest that ESCRT-III is a potential therapeutic target also for betaherpesvirus infections.

A serine-threonine protein kinase (PK), Us3, encoded by herpes simplex virus 1 (HSV-1), shares substrate specificity with host cellular PKs, protein kinase A (PKA), and AKT. Many Us3 substrates have been identified, and it is thought that during HSV-1 infection, Us3 fine-tunes the phosphorylation levels of individual substrates within their repertoire. However, the significance of this regulatory fine-tuning by Us3 during HSV-1 infection is poorly understood. Here, we found alanine at position 326 (Ala-326) in Us3 was required for the proper fine-tuning of Us3-mediated phosphorylation across the target repertoire in HSV-1-infected cells. Using recombinant viruses in which Us3 Ala-326 was replaced with valine (A326V) or isoleucine (A326I), we showed these mutations selectively altered the phosphorylation of only a subset of at least 14 Us3 target proteins tested in HSV-1-infected cells, with each mutation generally affecting different targets. Of note, (i) both mutations significantly reduced plaque sizes without affecting viral replication in cell cultures; (ii) the Us3 A326I mutation impaired viral replication in the brains of mice and improved survival following intracranial infection, whereas the Us3 A326V mutation had little effect; and (iii) the Us3 A326V mutation reduced ocular pathogenic manifestations and viral replication in the trigeminal ganglia and brains of mice, thereby improving survival following ocular infection. Taken together, these results suggest that the proper fine-tuning of Us3-mediated phosphorylation across its target repertoire is required for efficient cell-to-cell spread of HSV-1 in vitro , and its replication and pathogenicity in vivo . IMPORTANCE: The activation loop (A-loop) is a conformationally flexible loop that critically regulates cellular protein kinases (PKs), but its role in viral PKs during infection remains unclear. We demonstrated alanine at position 326 (Ala-326) in the A-loop of herpes simplex virus 1 (HSV-1) PK Us3 was important for the proper fine-tuning of Us3-mediated phosphorylation across the target repertoire in HSV-1-infected cells. This fine-tuning was necessary for efficient HSV-1 cell-cell spread in cell cultures, and replication and pathogenicity in mice. Taken together, fine-tuning phosphorylation levels of individual Us3 targets within its repertoire is important for HSV-1 infection in vitro and in vivo . Different amino acid substitutions at Us3 Ala-326 selectively affected the phosphorylation of most distinct Us3 targets, leading to varied phenotypic outcomes in viral replication and pathogenicity in mice. These results provide important clues to elucidate the mechanisms by which Us3 regulates HSV-1 infection in vivo .

Although viral protein expression and progeny virus production were independently shown to be highly heterogeneous in individual cells, their direct relationship, analyzed by considering their heterogeneities, has not been investigated to date. To elucidate the direct relationship between viral protein expression and progeny virus production, we constructed a reporter herpes simplex virus 1 (HSV-1) by tagging Venus to the late protein Us11. We then separated the HSV-1-infected cell population into multiple subpopulations according to the fluorescence intensity of Venus—which reflected the expression of L proteins, largely constituting virion structural proteins—and titrated virus yields and performed electron microscopic analysis in each subpopulation. Our results revealed that infectious progeny virus production, as well as nucleocapsid maturation, was triggered only when L protein expression exceeded a specific threshold. This suggested the existence of a rate-limiting step in progeny virus production, with nucleocapsid maturation potentially being one such step. IMPORTANCE: Earlier single-cell studies of virus-infected cells have revealed high heterogeneity in the state of viral gene expression and progeny virus yield. Notably, these two aspects have been shown independently, and therefore, the direct relationship between progeny virus production and viral gene expression has been unclear. This study, for the first time, demonstrated the direct and quantitative relationship between viral protein expression and progeny virus production by taking into account their heterogeneities and revealed a threshold for the levels of herpes simplex virus 1 protein expression for progeny virus production, thereby suggesting the existence of a rate-limiting step in progeny virus production.

Herpes simplex virus 1 (HSV-1) is the most common cause of viral encephalitis, which can be lethal or result in severe neurological defects despite antiviral therapy. The apolipoprotein B messenger-RNA editing enzyme, catalytic polypeptide-like (APOBEC) group of proteins can act as viral restriction factors. How HSV-1 evades this intrinsic immune mechanism is unclear. Here, using human carcinoma HEp-2 cells, we find that phosphorylation and therefore activation of HSV-1 uracil-DNA glycosylase counteracts mouse APOBEC1 DNA-editing activity on HSV-1 genomes. This protects viral genomes, promotes viral replication and encephalitis in the central nervous system of mice. Presence of Apobec1 improved encephalitis outcomes in mice challenged with HSV-1 carrying a mutation in the phosphorylation site of uracil-DNA glycosylase. Treatment with an UNG inhibitor, adeno-associated virus vector expressing UGI, protected wild-type HSV-1-infected mice from lethal encephalitis. These findings identify uracil-DNA glycosylase as a viral factor enabling evasion from intrinsic antiviral immunity mediated by APOBEC1 in the central nervous system and suggests a potential therapeutic approach to treat HSV-1 encephalitis.

Herpesviruses encode conserved protein kinases (CHPKs) that target cellular cyclin-dependent kinase (CDK) phosphorylation sites; thus, they are termed viral CDK-like kinases. Tyrosine 15 in the GxGxxG motifs of CDK1 and CDK2, whose phosphorylation down-regulates their catalytic activities, is conserved in the corresponding motifs of CHPKs. We found that CHPK UL13, the corresponding Tyr-162 in herpes simplex virus 2 (HSV-2), was phosphorylated in HSV-2-infected cells. Mutational analyses of HSV-2 UL13 Tyr-162 suggested that phosphorylation of UL13 Tyr-162 reduced the phosphorylation of all UL13 substrates tested in HSV-2-infected cells. These findings suggested that HSV-2 UL13 mimicked the regulatory mechanism of CDKs and that this CHPK has regulatory and functional mimicry with CDKs. Furthermore, phosphorylation of HSV-2 UL13 Tyr-162 was suggested to be required for the downregulation of viral replication and pathogenicity, specifically in the brains of mice, and for efficient viral recurrence in guinea pigs. These findings highlight the dual impact of the regulatory mimicry of CDKs by CHPK on the fine-tuned regulation of lytic and latent HSV-2 infections in vivo.

Facilitates chromatin transcription (FACT) interacts with nucleosomes to promote gene transcription by regulating the dissociation and reassembly of nucleosomes downstream and upstream of RNA polymerase II (Pol II). A previous study reported that herpes simplex virus 1 (HSV-1) regulatory protein ICP22 interacted with FACT and was required for its recruitment to the viral DNA genome in HSV-1-infected cells. However, the biological importance of interactions between ICP22 and FACT in relation to HSV-1 infection is unclear. Here, we mapped the minimal domain of ICP22 required for its efficient interaction with FACT to a cluster of five basic amino acids in ICP22. A recombinant virus harboring alanine substitutions in this identified cluster led to the decreased accumulation of viral mRNAs from UL54, UL38, and UL44 genes, reduced Pol II occupancy of these genes in MRC-5 cells, and impaired HSV-1 virulence in mice following ocular or intracranial infection. Furthermore, the treatment of mice infected with wild-type HSV-1 with CBL0137, a FACT inhibitor currently being investigated in clinical trials, significantly improved the survival rate of mice. These results suggested that the interaction between ICP22 and FACT was required for efficient HSV-1 gene expression and pathogenicity. Therefore, FACT might be a potential therapeutic target for HSV-1 infection. IMPORTANCE: ICP22 is a well-known regulatory factor of HSV-1 gene expression, but its mechanism(s) are poorly understood. Although the interaction of FACT with ICP22 was reported previously, its significance in HSV-1 infection is unknown. Given that FACT is involved in gene transcription, it is of interest to investigate this interaction as it relates to HSV-1 gene expression. To determine a direct link between the interaction and HSV-1 infection, we mapped a minimal domain of ICP22 required for its efficient interaction with FACT and generated a recombinant virus carrying mutations in the identified domain. Using the recombinant virus, we obtained evidence suggesting that the interaction between ICP22 and FACT promoted Pol II transcription from HSV-1 genes and viral virulence in mice. In addition, CBL0137, an inhibitor of FACT, effectively protected mice from lethal HSV-1 infection, suggesting FACT might be a potential target for the development of novel anti-HSV drugs.

Although the herpes simplex virus type 1 (HSV-1) genome was thought to contain approximately 80 different protein coding sequences (CDSs), recent multi-omics analyses reported HSV-1 encodes more than 200 potential CDSs. However, few of the newly identified CDSs were confirmed to be expressed at the peptide or protein level in HSV-1-infected cells. Furthermore, the impact of the proteins they encode on HSV-1 infection is largely unknown. This study focused on a newly identified CDS, UL31.6. Re-analyzation of our previous chemical proteomics data verified that UL31.6 was expressed at the peptide level in HSV-1-infected cells. Antisera raised against a viral protein encoded by UL31.6 (pUL31.6) reacted with a protein with an approximate molecular mass of 37 kDa in lysates of Vero cells infected with each of three HSV-1 strains. pUL31.6 was efficiently dissociated from virions in high-salt solution. A UL31.6-null mutation had a minimal effect on HSV-1 gene expression, replication, cell-to-cell spread, and morphogenesis in Vero cells; in contrast, it significantly reduced HSV-1 cell-to-cell spread in three neural cells but not in four non-neural cells including Vero cells. The UL31.6-null mutation also significantly reduced the mortality and viral replication in the brains of mice after intracranial infection, but had minimal effects on pathogenic manifestations in and around the eyes, and viral replication detected in the tear films of mice after ocular infection. These results indicated that pUL31.6 was a tegument protein and specifically acted as a neurovirulence factor by potentially promoting viral transmission between neuronal cells in the central nervous system. IMPORTANCE: Recent multi-omics analyses reported the herpes simplex virus type 1 (HSV-1) genome encodes an additional number of potential coding sequences (CDSs). However, the expressions of these CDSs at the peptide or protein levels and the biological effects of these CDSs on HSV-1 infection remain largely unknown. This study annotated a cryptic orphan CDS, termed UL31.6, an HSV-1 gene that encodes a tegument protein with an approximate molecular mass of 37 kDa, which specifically acts as a neurovirulence factor. Our study indicates that HSV-1 proteins important for viral pathogenesis remain to be identified and a comprehensive understanding of the pathogenesis of HSV-1 will require not only the identification of cryptic orphan CDSs using emerging technologies but also step-by-step and in-depth analyses of each of the cryptic orphan CDSs.