Taken together, the data presented herein indicate that targeting of pUL
34 to the INM and aggregation in the absence of Us3 kinase activity are not dependent on expression of lamin A/C or B1, despite the observation that these proteins were shown to interact with pUL
34 in infected cell lysates. On the other hand, the absence of lamin A/C altered the normal smooth distribution of pUL
34 as revealed by indirect immunofluorescence, suggesting that it is directly or indirectly involved in pUL
34 distribution within the nuclear rim. In contrast, lamin B1 was entirely dispensable for pUL
34 targeting to the nuclear rim, as viewed by indirect immunofluorescence, and to the INM, as viewed by immunogold electron microscopy. Thus, lamin A or lamin A-interacting components of the nuclear lamina are more relevant to pUL
34 targeting than lamin B1 or lamin B1-interacting components. The effects of lamin A on pUL34 targeting may reflect effects of lamin A on lamina mechanics and stiffness, functions which are largely independent of lamin B1 (16
). Alternatively, data obtained in this study and in previous work in which lamin A, pUL
31, and pUL
34 were shown to interact in pull-down assays from rabbit reticulocyte lysates suggest that the lamin A-pUL
34 interaction is direct and may play a role in pUL34 targeting (26
). The data are also consistent with the observation that emerin, a lamin A-interacting protein, can interact with pUL
34 and is displaced in HSV-infected cells in a UL
34-dependent manner (17
). Whether the emerin-pUL
34 interaction is mediated through interactions with lamin A/C will require further studies. A final possibility is that lamin B1 is involved in INM targeting of pUL
34, but lamin A confers redundant functions that compensate for the absence of lamin B1.
An untested but widely accepted hypothesis is that the nuclear lamina poses a barrier that herpesviruses must breach to allow nucleocapsids access to the INM for envelopment. Along these lines, previous observations indicated that the lamina is disrupted in HSV-infected cells in a UL
34-dependent manner (2
). Several studies have identified mechanisms by which pUL
34 of HSV may accomplish nuclear lamina disruption. Specifically, protein kinase Cδ and -α have been shown to be recruited to the nuclear membrane in a pUL
34-dependent manner to augment lamin B phosphorylation, whereas US
3 has been shown to phosphorylate lamin A directly in vitro and to be required for its optimal phosphorylation in infected cells (23
). Moreover, pUL
34 may have their own lamina-depolymerizing activities inasmuch as overexpression of these proteins in the absence of other proteins are sufficient to disrupt the lamina, and locally high concentrations of pUL
34 exaggerate adjacent lamina perforations (2
). The observation in this study that lamin A knockout cells are more permissive to replication of a US
3 kinase-dead virus at high multiplicities of infection further argues that at least lamin A poses a barrier to replication (and presumably nucleocapsid envelopment) and that US
3 helps to overcome this barrier. Thus, disruption of the lamina might indeed reflect an important role for these proteins in promoting virion budding at the INM.
Despite their relative permissiveness to virus replication at high multiplicities of infection, the lamin A knockout cells exhibited altered NM morphology when infected with the US
3 kinase-dead virus. Specifically, exaggerated extensions of the INM and increased space between the INM and ONM were noted. Because these effects were precluded by US
3 kinase activity, it follows that substrates of US
3 kinase other than lamin A are responsible for these phenotypes. Previous studies reported increased spacing between the INM and ONM upon overexpression of UL
34-dependent alterations of the nuclear lamina, and budding from the INM upon coexpression of pUL
31 and pUL
34 of pseudorabies virus in the absence of capsids (14
). Thus, the effects on nuclear membrane morphology induced by the combined absence of US
3 kinase activity and lamin A might be consequential to increased concentration of pUL
31 in certain regions of the INM. This is consistent with the observation by immunogold electron microscopy that pUL
34 was specifically detected in the unusual nuclear membrane extensions observed in this study.
Also interesting to us was the absence of virions trapped within the perinuclear space of the LmnA knockout MEFs infected with the US3 kinase-dead virus, despite the fact that this feature was commonplace in similarly infected normal MEFs. Thus, lamin A is somehow required for the observed retention of virions in discrete regions within the perinuclear space. We speculate that this reflects the general permissivity of primary envelopment in the lamin A knockout cells. Thus, capsids may bud through the INM at multiple sites in the absence of lamin A, whereas the position of envelopment sites are more restricted by the lamin A-imposed barrier in normal MEFs. The absence of US3 kinase activity would still delay fusion of the virion envelope and ONM, but the wide dispersal of virions in the perinuclear space of lamin A knockout cells would preclude virion aggregation. Alternatively, we cannot rule out the possibility that the nuclear envelope is generally more fragile in the lamin A knockout cells, thus precluding stable association of virions within the perinuclear space of cells as they are pelleted prior to fixation for electron microscopy.
We also noted that pUL34-containing regions mostly localized external to the nuclear membrane in lamin A knockout cells infected with the US3 kinase mutant but internal to the nuclear membrane in MEFs or lamin B knockout cells infected with this virus. We speculate that lamin A, by enhancing lamina stiffness, restrains locally high concentrations of the pUL31/pUL34 complex from inducing cytoplasmic protrusions of the nuclear membrane, although we cannot exclude more indirect mechanisms.
In striking contrast to the results with lamin A, LmnB1
knockout MEFs were less permissive to viral replication at high multiplicities of infection, and this was observed in both the presence and absence of US
3 kinase activity. The role of lamin B1 in HSV infection is unclear, but it is required for a number of important nuclear functions including transcription, DNA replication, cell signaling, and optimal reassembly of the nucleus after mitosis (36
). Thus, disturbance of any of these or related functions might render the cells less permissive by impairing the machinery that the virus ultimately needs to commandeer for optimal infection. This may be one reason why alteration of the nuclear lamina during viral infection is mostly limited to very discrete regions. Such a strategy might limit detrimental effects on the cell while still allowing nucleocapsids to bud through the INM.