Elsevier

Virus Research

Volume 209, 2 November 2015, Pages 76-86
Virus Research

Herpesvirus nuclear egress: Pseudorabies Virus can simultaneously induce nuclear envelope breakdown and exit the nucleus via the envelopment–deenvelopment-pathway

https://doi.org/10.1016/j.virusres.2015.02.001Get rights and content

Highlights

Abstract

Herpesvirus replication takes place in the nucleus and in the cytosol. After entering the cell, nucleocapsids are transported to nuclear pores where viral DNA is released into the nucleus. After gene expression and DNA replication new nucleocapsids are assembled which have to exit the nucleus for virion formation in the cytosol. Since nuclear pores are not wide enough to allow passage of the nucleocapsid, nuclear egress occurs by vesicle-mediated transport through the nuclear envelope. To this end, nucleocapsids bud at the inner nuclear membrane (INM) recruiting a primary envelope which then fuses with the outer nuclear membrane (ONM). In the absence of this regulated nuclear egress, mutants of the alphaherpesvirus pseudorabies virus have been described that escape from the nucleus after virus-induced nuclear envelope breakdown. Here we review these exit pathways and demonstrate that both can occur simultaneously under appropriate conditions.

Introduction

Herpesviruses are an extensive family of large DNA-viruses. Within the order Herpesvirales, the family Herpesviridae can be further divided into the three subfamilies Alpha-, Beta- and Gammaherpesvirinae (Davison, 2010, Davison et al., 2009), which differ in their host specificity, replication efficiency and target cells for establishment of latency (Pellett and Roizman, 2013). However, all herpes virions have a similar structure. The double-stranded DNA-genome, which is enclosed in an icosahedral capsid, is surrounded by aproteinaceous tegument, and an envelope, in which viral glycoproteins are embedded (McGeoch et al., 2006).

Herpesviruses encode far more genes than many other viruses. Approximately 40 of them are conserved within the Herpesviridae and are designated as “core” genes. These genes encode proteins which are important for general aspects of lytic replication, like DNA-replication, -processing and -encapsidation, capsid assembly and nuclear egress, or for structural proteins of capsid, tegument or envelope (McGeoch et al., 2006).

The complex structure of Herpesviruses is also reflected by their replication cycle. The nucleocapsid enters the host cell after fusion of the virion envelope with the plasma membrane, or with the endosomal membrane if endocytosis occurs. For this process the conserved core fusion machinery is essential. It consists of glycoprotein (g)B and the heterodimeric gH/gL complex. Other glycoproteins, which mediate attachment of the virion to cellular surface receptor proteins, enhance the fusion process (Eisenberg et al., 2012). Subsequently the virion is transported along microtubules to the nuclear pore (Sodeik et al., 1997, Zaichick et al., 2013), where the viral genome is released and enters the nucleus through the nuclear pore. Following circularization of the linear viral DNA (Strang and Stow, 2005), the viral genome is transcribed and replicated. Protein translation occurs in the cytoplasm and, after transport of capsid proteins into the nucleus, the capsid is assembled around a protein scaffold which is autoproteolytically cleaved and extruded when DNA is packaged to form nucleocapsids (Homa and Brown, 1997).

These nucleocapsids exit the nucleus through the nuclear membranes. Final maturation of the virion then takes place in the cytosol. Acquisition of proteins of the inner tegument during or immediately following nuclear egress is followed by addition of an outer tegument layer in the trans-Golgi region and final envelopment by budding into trans-Golgi vesicles. Interactions between tegument and capsid proteins, among tegument proteins, as well as between tegument proteins and cytoplasmic domains of glycoproteins drive virion formation. Subsequently, the enveloped intravesicular virions are transported to the plasma membrane and released into the extracellular space (Johnson and Baines, 2011, Mettenleiter, 2004, Mettenleiter, 2006, Mettenleiter et al., 2009).

Exit from the nucleus is an essential and highly regulated step in herpesviral replication. In contrast to the plasma membrane, the nuclear envelope is a formidable barrier consisting of two membranes with high complexity. The two membrane layers surround the periplasmic space and are linked at nuclear pores, which allow transport between nucleus and cytoplasm (Beck et al., 2004, Beck et al., 2007). Although nuclear pores have an outer diameter of ∼125 nm, the open transport channel only averages to a diameter of ∼50 nm (Frenkiel-Krispin et al., 2010, Pante and Kann, 2002). Interestingly, although both the inner (INM) and outer nuclear membrane (ONM) are contiguous, they contain different sets of proteins (Hetzer, 2010). Below the INM lies the nuclear lamina, a meshwork of intermediate filaments composed of type A and B lamins. The lamina has been shown to be important for nuclear stability, genome organization, transcription, DNA repair and signal transduction (Dittmer and Misteli, 2011).

It has long been disputed how herpesvirus nucleocapsids leave the nucleus, considering their diameter of 125 nm. Initially three different mechanisms were proposed, the single-envelopment-pathway, egress through dilated pores and the envelopment–deenvelopment-pathway (recently reviewed in Mettenleiter et al. 2013).

For the single-envelopment-pathway, it was assumed that the capsid buds into the INM obtaining a lipid envelope from the perinuclear space which is maintained during transport of the virion through the endoplasmic reticulum and the secretory pathway. However, even though some experimental data was interpreted in favor of this pathway (Johnson and Spear, 1982, Torrisi et al., 1992), it was shown that the composition (Klupp et al., 2000, Kopp et al., 2002, Reynolds et al., 2002, Skepper et al., 2001, van Genderen et al., 1994) and the appearance (Granzow et al., 2001, Mettenleiter et al., 2009) of primary and mature virions differ considerably. Moreover, fusion events between primary enveloped virions and the ONM, as well as naked cytoplasmic capsids, were observed for different herpesviruses, contradicting the single-envelopment-pathway (Granzow et al., 1997, Granzow et al., 2001, Stackpole and Mizell, 1968). Therefore, it is highly unlikely that herpesviruses use the single-envelopment-pathway to exit the nucleus. Alternatively it was proposed that herpesviruses dilate nuclear pores to exit the nucleus (Wild et al., 2005, Wild et al., 2009, Leuzinger et al., 2005). However, no solid evidence for herpesvirus transfer through nuclear pores has been provided. In contrast, it has been shown that reorganization of the nuclear pore network is not required for herpesviral nuclear egress (Nagel et al., 2008). The same holds true for the proposed combination of both postulated pathways (Leuzinger et al., 2005).

Meanwhile, it is accepted that nuclear egress of herpesviruses proceeds via a two-step mechanism designated as “envelopment–deenvelopment-pathway” (Fig. 1A; Skepper et al., 2001). Here, the nucleocapsid buds at the INM into the perinuclear space followed by fission of the INM-derived vesicle to form a primary, transient envelope. This primary envelope then fuses with the ONM, thereby releasing the nucleocapsid into the cytosol (Johnson and Baines, 2011, Mettenleiter et al., 2009, Mettenleiter et al., 2013). Mechanistically, this process equals a vesicular (=primary envelope mediated) transport of cargo (=the nucleocapsid) through the nuclear envelope, a process so far unknown in cell biology.

The conserved heterodimeric nuclear egress complex (NEC) is an essential component of this pathway (Mettenleiter, 2004). It is anchored in the nuclear envelope by its transmembrane protein component which has been designated as pUL34 in herpes simplex virus 1 (HSV-1) (Reynolds et al., 2001, Roller et al., 2000) and pseudorabies virus (PrV) (Klupp et al., 2000), pUL50 in human (HCMV) (Milbradt et al., 2007) and M50 in murine cytomegalovirus (MCMV) (Muranyi et al., 2002) and BFRF1 in Epstein–Barr Virus (Farina et al., 2005, Lake and Hutt-Fletcher, 2004). The transmembrane protein interacts with a second protein designated as pUL31 in HSV-1 and PrV (Fuchs et al., 2002, Reynolds et al., 2001), pUL53 or M53 in HCMV and MCMV (Dal Monte et al., 2002, Muranyi et al., 2002) and BFLF2 in EBV (Gonnella et al., 2005, Lake and Hutt-Fletcher, 2004). When expressed separately, the type II tail-anchored pUL34 localizes intrinsically to the nuclear membrane, whereas pUL31 is distributed diffusely throughout the nucleus and is recruited to the nuclear membrane only by interaction with pUL34 (Mettenleiter, 2004). Until now little is known about the structure of the NEC, since no crystal structure is available. However, pUL34 homologs consist of a non-conserved carboxy-terminal part and a highly conserved amino-terminal part, which can be subdivided into two (Milbradt et al., 2012) or three (Haugo et al., 2011) conserved regions (CR), depending on the selection of sequences used for the comparison and the alignment algorithm.

The pUL31 interaction domains have been located within the conserved parts of the pUL34 homologs, although the exact location may differ between the different homologs (Bubeck et al., 2004, Fuchs et al., 2002, Liang and Baines, 2005, Milbradt et al., 2012, Passvogel et al., 2013). For HSV-1 pUL34 the pUL31 interaction domain was localized to amino acids 137 to 181 (Liang and Baines, 2005), whereas in PrV pUL34 it is located between aa 5 and 161 (Passvogel et al., 2013) and the pUL53 interaction domain of HCMV pUL50 was mapped between aa 10 to 169 (Milbradt et al., 2012). However, residues that are not part of the identified interaction domains may also influence NEC formation (Roller et al., 2010).

To further examine the pUL31–pUL34 interaction several mutational studies were performed to identify amino acids important for interaction with pUL31 homologs and NEC function. For the MCMV M50 amino acids Glu56 and Tyr57, which are strictly conserved among herpesviruses (EY motif), have been shown by random transposon mutagenesis to be important for complex formation (Bubeck et al., 2004). Using site-directed mutagenesis, the same two amino acids were also shown to be required for interaction of HCMV pUL50 with pUL53 (Milbradt et al., 2012) and PrV pUL34 with pUL31 (Passvogel et al., 2013). Remarkably, a charge cluster mutation at that position in HSV-1 pUL34 resulted in a mutant protein which did not localize correctly to the nuclear membrane and was unable to complement replication of a UL34-null virus, despite continuing interaction with pUL31 (Bjerke et al., 2003).

Recently, site directed-mutagenesis of PrV pUL34 has been used to identify other amino acids important for interaction with pUL31. In transfection experiments NEC complex formation was impaired after mutation of Asn75 and Gly77. However during infection, pUL34 carrying a mutation in Gly77 was still able to form a complex with pUL31 and largely complemented replication of a PUL34 deficient virus mutant, indicating that other viral components might be involved in NEC formation. Both, Asn75 and Gly77 are part of a conserved NTG motif, which, together with the conserved EY motif, may form a structure necessary for pUL31 binding (Passvogel et al., 2014). Interestingly, mutation of Asn103 or a dileucine motif (L166/167) separating the conserved amino-terminus and the variable carboxy-terminus, impaired viral replication distinctly, while not affecting NEC formation (Passvogel et al., 2014).

Remarkably, most of the mutants analyzed in this study, including those that did not affect viral replication, exhibited reduced plaque sizes, indicating that pUL34 is also involved in direct viral cell-to-cell spread (Passvogel et al., 2014). Another study on HSV pUL34 showed that also the combined mutation of amino acids Arg158 and Arg161 impaired nuclear envelope localization and interaction with pUL31. This mutant exhibited a defect in cell-to-cell spread as well (Bjerke et al., 2003, Roller et al., 2011), pointing to a function of the NEC beyond nuclear egress.

The variable carboxy-termini of HSV-1 and PrV pUL34 containing the transmembrane domain were substituted by various viral or cellular heterologous sequences without compromising nuclear membrane localization, pUL31 interaction or NEC function, as long as a hydrophobic domain sufficient to anchor the protein in the membrane is available. This could be provided by other viral tail-anchored membrane proteins like HSV-1 pUS9 or pUL56, homologous proteins of other Herpesviruses like HCMV pUL50 or EBV BFRF1, cellular INM proteins like Emerin, Lap2β or lamin B receptor or cellular tail-anchored membrane proteins like Bcl-2 (B-cell lymphoma 2) or Vamp (vesicle associated membrane protein) (Ott et al., 2011, Schuster et al., 2012). Even shortening the transmembrane domain to 15 aa did not interfere with membrane insertion (Ott et al., 2011). In general, these results indicate that the carboxy-terminal part of the protein may only be required for anchoring the NEC in the nuclear membrane.

At the border between the conserved amino-terminal and the variable carboxy-terminal parts of PrV pUL34 RQR is located between amino acids 173 and 175 (Passvogel et al., 2013). An RXR motif had been identified before in the cytoplasmic domains of HSV-1 and HCMV gB, as well as in the cellular lamin B receptor. Fusion of this motif to the membrane protein CD8 caused CD8-relocation to the INM (Meyer et al., 2002, Meyer and Radsak, 2000). Based on these results, it was initially assumed, that the RQR motif could be involved in INM targeting of PrV pUL34. However, mutational studies indicated that it serves as a Golgi retrieval signal, relocating Golgi compartment-localized molecules back to the ER, from where they can be transported to the nuclear membrane (Passvogel et al., 2013). For MCMV M50 a potentially essential sequence has been identified between aa 178 and 207. Even though mutants comprising a deletion of this region were able to bind M53, they could not rescue viral replication. Further analysis showed that this region contains a polyproline stretch that is conserved among the betaherpesvirus homologs (Bubeck et al., 2004).

By comparison of sequences of 36 members of the pUL31 family, an amino-terminal region with low sequence conservation, which is variable in length, and a conserved C-terminal region have been identified. The conserved region can be further subdivided into four conserved regions (CR) (Lotzerich et al., 2006).

In many pUL31 homologs, the amino-terminal part contains a classical mono- or bipartite nuclear localization signal (NLS) (Lotzerich et al., 2006, Schmeiser et al., 2013, Zhu et al., 1999). In PrV pUL31 the NLS was predicted between aa 5 and 20 (Schmeiser et al., 2013) and in HCMV pUL53 it is attributed to aa 18 to 27 (Schmeiser et al., 2013). Presumably, pUL31 is imported into the nucleus via the classical importin mediated pathway (Pemberton and Paschal, 2005, Schmeiser et al., 2013). Small molecules like pUL31 may also be able to travel through the nuclear pores by passive diffusion (Terry and Wente, 2009).

The pUL34- interaction domain has been located within CR1 in HSV-1 pUL31, PrV pUL31, HCMV pUL53, MCMV M53 and EBV BFLF2 by protein complementation assays (Lotzerich et al., 2006, Schnee et al., 2006). It is assumed that an amphipathic helix, which has been found in this region in HCMV pUL53, is responsible for complex formation through hydrophobic and charge–charge-interactions (Sam et al., 2009). Efforts to identify individual amino acids in HCMV pUL53 responsible for binding of pUL50 failed since no single amino acid exchange abolished NEC formation. However, viral replication was reduced after mutation of K128, Y129 and L130 to alanine, but only combined mutation of two or three of these critical amino acids resulted in inhibition of NEC formation and suppression of viral replication (Lotzerich et al., 2006).

In addition, sequences in CR2 and/or CR3 are possibly involved in NEC interaction (Pogoda et al., 2012, Roller et al., 2010). CR3 of HSV-1 pUL31 was shown to be involved in membrane curvature of the INM (Roller et al., 2010) and mutagenesis of MCMV M53 CR4 resulted in dominant-negative mutants with defects in capsid release (Popa et al., 2010). Yeast-two hybrid studies in PrV indicated that the amino-terminus of pUL31 (up to aa 41) containing the predicted NLS is dispensable for pUL34 interaction (Fuchs et al., 2002).

In contrast to pUL31, pUL34 does not cross the nuclear membrane by active nuclear import, but may instead diffuse passively to the INM along peripheral channels of the NPC. There it is retained by NEC formation (Schmeiser et al., 2013). Although a protein complementation assay indicated that PrV pUL31 function can be complemented by homologs of the same herpesvirus subfamily, i.e. HSV pUL31 (Schnee et al., 2006), this effect could not be validated in infection experiments (Klupp et al., personal communication).

In summary, the structure and function of the two components of the NEC are not clear yet and full understanding of the results of the mutational analyses will have to await elucidation of the NEC crystal structure.

For nuclear egress, nucleocapsids bud at the INM into the perinuclear cleft. Below the INM lies the nuclear lamina, which stabilizes and shields the nuclear envelope, and prevents direct access of the nucleocapsids to the INM. Therefore, the NEC recruits different kinases to the INM, where they phosphorylate the lamins, resulting in local dissolution of the nuclear lamina to allow interaction of the nucleocapsid with the INM. For HSV-1 and MCMV protein kinase C (PKC) was shown to be involved in nuclear lamina dissolution (Muranyi et al., 2002, Park and Baines, 2006), whereas for HCMV participation of PKC could not be detected (Sharma and Coen, 2014, Sharma et al., 2014). Instead, the conserved pUL13 homolog pUL97 causes nuclear lamina disruption in HCMV infected cells (Hamirally et al., 2009, Krosky et al., 2003, Sharma et al., 2014). Besides pUL13, which is important for lamin phosphorylation in HSV-2 infected cells (Cano-Monreal et al., 2009), the viral kinase pUS3, which is present only in alphaherpesviruses, has been shown to be involved in nuclear lamina disruption (Mou et al., 2007). pUS3 also phosphorylates pUL31 to regulate primary envelopment and fusion (Mou et al., 2009). In addition, HSV-1 tegument protein pUL47 and immediate early protein pUS1 interact with the NEC and/or pUS3 and may also contribute to regulation of efficient primary envelopment (Liu et al., 2014, Maruzuru et al., 2014). A recent proteomic analysis of the HCMV NEC demonstrated that during the course of replication it may associate with different proteins. Especially at late time points of infection, it interacted with several cellular proteins, like Emerin, which possibly serves as a docking molecule to guide the NEC to specific sites of the nuclear envelope. Interestingly, knockdown of Emerin impaired HCMV replication (Milbradt et al., 2014). Other herpesviruses, like KSHV and HSV, have also been shown to phosphorylate Emerin during infection, which may be necessary to induce nuclear lamina disruption (Farina et al., 2013, Leach et al., 2007, Morris et al., 2007).

Simultaneous expression of both NEC components is sufficient to induce formation and fission of intraluminal vesicles in the perinuclear space which resemble primary enveloped virions (Desai et al., 2012, Klupp et al., 2007, Lee et al., 2012), indicating that the NEC mediates curvature of the nuclear membrane. For EBV, formation of vesicle-like structures was already observed after expression of only the pUL34-homolog BFRF1, which was inhibited in the presence of dominant-negative forms of the ESCRT (endosomal sorting complex required for transport)-proteins Vps4 and Chmp4b. This may indicate that components of the ESCRT, which are important for intracytoplasmic vesicular transport processes, are involved in vesicle formation (Lee et al., 2012). However, these data could not be reproduced in other herpesviruses. Only recently it was shown that the NEC binds to unilamellar model membranes and induces budding and scission (Bigalke et al., 2014). Therefore, it is now clear that the NEC represents the minimal membrane deformation and scission machinery that can function without any other viral or cellular factors. However, these results do not exclude the possibility that herpesviruses recruit other factors to regulate the budding process in vivo (Bigalke et al., 2014).

During nuclear egress, mature nucleocapsids are preferred over immature capsid forms (Klupp et al., 2011). This selection is regulated by pUL25 (Klupp et al., 2006), which is part of another conserved heterodimeric complex between the capsid-associated proteins pUL17 and pUL25, called CVSC (capsid vertex specific component) (Cockrell et al., 2011, Toropova et al., 2011, Trus et al., 2007). For some time, the CVSC was called CCSC (C capsid specific component), as it was at first only detected on mature (or C-) capsids (Trus et al., 2007). Only the detection of small amounts of the complex on A- and B-capsids resulted in the change of name (Cockrell et al., 2011). HSV-1 pUL31 directly interacts with the carboxy-terminal 20 amino acids of pUL25 (Yang and Baines, 2011, Yang et al., 2014). However, in the absence of pUL25, pUL31 still interacted with the capsid, presumably via pUL17 (Yang et al., 2014), which is consistent with data obtained for PrV (Leelawong et al., 2011). Nevertheless, in the absence of pUL25 nuclear egress was inhibited, even though nucleocapsids were closely associated with the INM (Klupp et al., 2006, Kuhn et al., 2008), indicating a functional role for pUL25 in nuclear egress. Additional, interactions between the NEC and the capsid could be involved in mediating nucleocapsid recruitment to the INM. For instance, interactions between the NEC and pUL33, a conserved protein which is part of the encapsidation machinery (Fossum et al., 2009, Vizoso Pinto et al., 2011), or directly with the core capsid (Leelawong et al., 2011, Ye et al., 2000) have been observed by yeast-two-hybrid analyses.

The composition of primary herpes virions within the periplasmic space has not been unraveled yet. However, it was shown that, in addition to the nucleocapsid, also the NEC and the protein kinase pUS3 are present (Granzow et al., 2004, Reynolds et al., 2002). Absence of pUS3 results in accumulations of primary enveloped particles within the perinuclear space, indicating that pUS3 is involved in fusion of the perinuclear virion with the ONM (Klupp et al., 2001, Reynolds et al., 2002, Schumacher et al., 2005) and disruption of the NEC during de-envelopment (Mou et al., 2009). In contrast, it is still disputed whether glycoproteins gB, gD or gM are part of perinuclear virions. While they have been detected in primary HSV virions (reviewed in Johnson and Baines (2011)), they were not detected in PrV (Klupp et al., 2008). Similar results were obtained for tegument proteins pUL49, pUL41, pUL48 and pUL11 (reviewed in Mettenleiter et al. (2013)), as well as pUL47 (Kopp et al., 2002, Liu et al., 2014). Also, there is evidence for (Bucks et al., 2007, Leelawong et al., 2012, Luxton et al., 2006, Morrison et al., 1998) and against (Fuchs et al., 2004, Granzow et al., 2004, Möhl et al., 2009, Trus et al., 2007) presence of the large tegument protein pUL36 in primary enveloped virions. However, none of these proteins has been shown to be essential for nuclear egress. Primary enveloped particles and mature virions differ significantly in their ultrastructure, already indicating major differences in virion composition. Whereas primary enveloped particles lack the surface spikes observed on mature virions, they contain a distinct electron-dense ring closely apposed to the primary envelope (Granzow et al., 1997, Granzow et al., 2004, Mettenleiter et al., 2009). To determine the composition of primary enveloped virions in more detail, it will be necessary to purify these particles from the perinuclear cleft for detailed proteomic analysis. Protocols for isolation of primary enveloped virions have been published (Padula et al., 2009, Remillard-Labrosse and Lippe, 2011), but require further validation.

The nucleocapsid is released into the cytosol through fusion of the primary envelope with the ONM. This process resembles fusion of the virion envelope with the plasma membrane during entry. The glycoproteins gB, gH, and gL are essential for fusion during entry (reviewed in Eisenberg et al. (2012)), but deletion of either one does not block nuclear egress indicating that both fusion processes are mechanistically different. HSV-1 nuclear egress has been shown to be slightly impaired after simultaneous deletion of gB and gH (Farnsworth et al., 2007), whereas combined deletion of different glycoproteins did not influence PrV nuclear egress (Klupp et al., 2008). Hence, the entry fusion machinery may modulate nuclear egress, but is not essential for this process. Instead, other components of the perinuclear virion could be involved. pUL34 is the only membrane protein proven to be present in perinuclear virions. However, the luminal domain of pUL34 is very short and has been shown to be exchangeable (Ott et al., 2011, Schuster et al., 2012), arguing against a role for pUL34 in deenvelopment. Instead it is more likely that cellular proteins are involved in deenvelopment of perinuclear virions (Maric et al., 2011).

Recently, TorsinA, an AAA+(ATPases associated with diverse cellular activities)-ATPase important for nuclear envelope maintenance (Goodchild et al., 2005, Grundmann et al., 2007, Naismith et al., 2004) has been shown to influence nuclear egress of herpesviruses (Maric et al., 2011). Overexpression of TorsinA reduced HSV-1 replication, and resulted in formation of virus-like-vesicles in the cytoplasm and the perinuclear space of the host cell. They resembled primary enveloped virions ultrastructurally and in their composition, e.g. they contained pUL34, an intrinsic component of perinuclear virions (Granzow et al., 2004). The presence of TorsinA in many of the cytoplasmic vesicles supports a functional role for TorsinA in nuclear egress (Maric et al., 2011). Possibly, overexpression of TorsinA impairs fusion of perinuclear virions with the ONM and, therefore, results in accumulation of capsids in the perinuclear space. The virions then translocate within the perinuclear space to the ER, where they are trapped. Alternatively, nuclear egress may proceed normally but cytoplasmic capsids become enveloped in a membrane containing TorsinA and pUL34 (Maric et al., 2011).

For a long time it has been thought that the herpesvirus-mediated envelopment–deenvelopment-pathway is the only vesicular transport mechanism across the nuclear envelope and, thus, unique in cellular biology (Mettenleiter, 2004, Mettenleiter, 2006, Mettenleiter et al., 2009, Mettenleiter et al., 2013, Johnson and Baines, 2011, Roller, 2008). However, recently it was shown for Drosophila that large ribonucleoprotein complexes (RNPs) are transported out of the nucleus in a similar process (Speese et al., 2012). Previously, it had been assumed that large RNPs, which exceed the nuclear pore diameter, also exit through the nuclear pore complexes, if necessary after structural remodeling (Grunwald et al., 2011). It is unclear why large ribonucleoprotein complexes are transported through the nuclear membrane, as transport through nuclear pores is very efficient and compromises cellular processes to a lesser extend. The choice of the transport mechanism could simply be a matter of size of the cargo, but it could also be advantageous to co-transport functionally related mRNAs or to transport mRNAs in a translationally repressed state. Alternately, transport through the nuclear membrane could also serve as an additional, independently regulated transport mechanism, to allow export of RNP complexes when general transport is inhibited (Montpetit and Weis, 2012).

Nuclear exit of RNP complexes and nuclear egress of herpesviruses appear very similar. Both processes depend on the phosphorylation of cellular lamins by kinases, like PKC, resulting in lamina disruption (Muranyi et al., 2002, Speese et al., 2012). In addition, herpesviral nuclear egress, as well as vesicular transport of RNP complexes across the nuclear envelope, might involve the AAA+-ATPase Torsin (Jokhi et al., 2013, Maric et al., 2011). Knockdown of the Torsin gene in Drosophila resulted in the accumulation of large RNP complexes within the perinuclear space. These RNP complexes were abnormally shaped and remained attached to the INM. In addition, fewer large RNP complexes could be observed in Torsin-deficient cells, indicating that in Drosophila Torsin is involved in regulating RNP complex scission of the INM (Jokhi et al., 2013).

Nuclear egress is a highly regulated process, allowing Herpesviruses to exit the nucleus without impairing nuclear envelope integrity. Although the NEC is essential for this pathway, deletion of either pUL34 (Klupp et al., 2000) or pUL31 (Fuchs et al., 2002) drastically reduced but did not abolish PrV replication. To analyze whether PrV nucleocapsids can use an alternative exit pathway, the residual infectivity of UL34- (Klupp et al., 2011), and UL31-deleted PrV (Grimm et al., 2012), was used for reversion analysis. After multiple passages on rabbit kidney (RK13) cells, the mutant viruses PrV-ΔUL34Pass and PrV-ΔUL31Pass replicated to titers comparable to wild type virus. Ultrastructural analyses showed that both passaged viruses did not exit the nucleus via the envelopment–deenvelopment-pathway, but induced nuclear envelope breakdown (NEBD) in ca. 50% of infected nuclei (Fig. 1B). Nucleocapsids leaked through the fragmented nuclear envelope in the cytosol, where final maturation proceeded (Grimm et al., 2012, Klupp et al., 2011). As a consequence the selection of mature capsids, which occurs during nuclear egress, was impaired, resulting in a larger amount of immature capsids in the cytosol as well as in the extracellular space. In general, after infection with different herpesviruses the nuclear membrane remains intact even at late times of infection or during pseudomitosis (Hertel et al., 2007, Hertel and Mocarski, 2004). Only recently it was shown that HSV infection can also cause NEBD in cells deficient in TorsinA. Interestingly, nuclear envelope fragmentation caused a reduction in virus replication. gB and gH, as well as NEC component pUL34, enhanced NEBD, whereas pUS3 had an inhibitory effect on NE breakdown. However, in contrast to NEBD induced by the PrV mutants, HSV NEBD did not rescue replication of an UL34 deficient virus (Maric et al., 2014).

Congruent mutations were detected in both passaged PrV virus mutants in seven genes, encoding for tegument and envelope proteins, as well as the maturational protease (Grimm et al., 2012). Parallel, inhibitor studies were performed to determine, which cellular pathway might be manipulated. Roscovitine, an inhibitor of cdk1, cdk2 and cdk5, and, in higher concentrations, ERK1/2, as well as U0126, a MEK1/2 inhibitor, specifically impaired replication of both passaged viruses (Grimm et al., 2012) indicating involvement of these pathways in NEBD.

The VZV pUL46 homolog ORF12 induces ERK1/2 phosphorylation (Liu et al., 2012). Since PrV pUL46 was found to be mutated in and inhibitors of the ERK signaling cascade selectively impaired replication of both passaged viruses, the involvement of the ERK signaling cascade in NEBD induction was analyzed in more detail. However, although wild type and mutated pUL46 were able to induce ERK phosphorylation and activate ERK target genes, the mutated pUL46 was not responsible for NEBD. In contrast to expectations, the mutations present in pUL46 of both passaged viruses restricted the extent of NEBD (Schulz et al., 2014).

Since pUL46 did not target the ERK signaling cascade to induce NEBD, other factors possibly involved in NEBD induction were analyzed. Both passaged virus mutants exhibit an enhanced fusion activity, resulting in formation of large syncytia. To analyze whether the fragmentation of the nuclear envelope is the result of deregulated fusion processes, core components of the Herpesvirus fusion machinery were deleted from the genomes of both passaged viruses. However, neither deletion of gB, nor of gH had an influence on nuclear envelope fragmentation (Schulz et al., 2013). pUL34 is part of the nuclear egress complex and therefore essential for the nuclear exit of Herpesviruses via the envelopment–deenvelopment pathway. However, it is not clear if NEBD is induced only in the absence of the NEC. To determine whether and how pUL34 influences fragmentation of the nuclear envelope, pUL34 was reintroduced into its original locus in PrV-ΔUL34Pass. The resulting PrV-ΔUL34Pass/UL34wt was analyzed for mechanisms of nuclear exit.

Section snippets

Cells and viruses

Rabbit kidney (RK13), African green monkey kidney (Vero) and porcine kidney (PSEK) cells were used. Isolation of PrV-ΔUL34Pass has been described (Klupp et al., 2011). For substitution of gfp, present in the UL34 gene locus of PrV-ΔUL34Pass, by wild type UL34, the 1.3 kbp XhoI fragment containing UL34 was ligated into plasmid pcDNA and cotransfected with PrV-ΔUL34Pass genomic DNA, followed by screening for non-fluorescent plaques. Correct deletion of gfp sequences and insertion of UL34 was

Isolation of PrV-ΔUL34Pass/UL34wt

PrV-ΔUL34Pass was isolated after multiple passaging of a UL34 deficient PrV in RK13 cells (Klupp et al., 2011). Although PrV-ΔUL34Pass replicates comparable to wild type PrV, it does not exit the nucleus via the NEC but induces NEBD. To analyze whether the absence of UL34 is necessary for NEBD induction and to determine the nuclear exit strategy of PrV-ΔUL34Pass in the presence of a functional nuclear egress complex, wild type UL34 was reinserted in the PrV-ΔUL34Pass genome. To this end,

Discussion

In this study we analyzed whether absence of pUL34 and inability to use the NEC for nuclear egress is necessary for NEBD induction by PrV. We found that fragmentation of the nuclear envelope is still induced by PrV-ΔUL34Pass/UL34wt, which expresses wild-type pUL34 from its original gene locus. However, this mutant virus was not only able to induce NEBD, but also used the envelopment–deenvelopment-pathway. Apparently, the vesicular transport through the nuclear membrane was beneficial early in

Acknowledgments

We thank C. Meinke, M. Sell and P. Meyer for expert technical assistance, M. Jörn for photographic help and M. Ziller for help with statistical analysis. This work was supported by the Deutsche Forschungsgemeinschaft (DFG Me 854/12-1).

References (127)

  • M. Maric et al.

    Nuclear envelope breakdown induced by herpes simplex virus type 1 involves the activity of viral fusion proteins

    Virology

    (2014)
  • D.J. McGeoch et al.

    Topics in herpesvirus genomics and evolution

    Virus Res.

    (2006)
  • T.C. Mettenleiter

    Glycoprotein gIII deletion mutants of pseudorabies virus are impaired in virus entry

    Virology

    (1989)
  • T.C. Mettenleiter

    Budding events in herpesvirus morphogenesis

    Virus Res.

    (2004)
  • T.C. Mettenleiter

    Intriguing interplay between viral proteins during herpesvirus assembly or: the herpesvirus assembly puzzle

    Vet. Microbiol.

    (2006)
  • T.C. Mettenleiter et al.

    Herpesvirus assembly: an update

    Virus Res.

    (2009)
  • G. Meyer et al.

    A three-residue signal confers localization of a reporter protein in the inner nuclear membrane

    Biochem. Biophys. Res. Commun.

    (2002)
  • G.A. Meyer et al.

    Identification of a novel signal sequence that targets transmembrane proteins to the nuclear envelope inner membrane

    J. Biol. Chem.

    (2000)
  • J. Milbradt et al.

    Specific residues of a conserved domain in the N terminus of the human cytomegalovirus pUL50 protein determine its intranuclear interaction with pUL53

    J. Biol. Chem.

    (2012)
  • J. Milbradt et al.

    Proteomic analysis of the multimeric nuclear egress complex of human cytomegalovirus

    Mol. Cell. Proteomics

    (2014)
  • G. Remillard-Labrosse et al.

    In vitro nuclear egress of herpes simplex virus type 1 capsids

    Methods

    (2011)
  • M. Beck et al.

    Nuclear pore complex structure and dynamics revealed by cryoelectron tomography

    Science

    (2004)
  • M. Beck et al.

    Snapshots of nuclear pore complexes in action captured by cryo-electron tomography

    Nature

    (2007)
  • J.M. Bigalke et al.

    Membrane deformation and scission by the HSV-1 nuclear egress complex

    Nat. Commun.

    (2014)
  • S.L. Bjerke et al.

    Effects of charged cluster mutations on the function of herpes simplex virus type 1 UL34 protein

    J. Virol.

    (2003)
  • A.R. Brack et al.

    Role of the cytoplasmic tail of pseudorabies virus glycoprotein E in virion formation

    J. Virol.

    (2000)
  • A. Bubeck et al.

    Comprehensive mutational analysis of a herpesvirus gene in the viral genome context reveals a region essential for virus replication

    J. Virol.

    (2004)
  • S.K. Cockrell et al.

    Residues of the UL25 protein of herpes simplex virus that are required for its stable interaction with capsids

    J. Virol.

    (2011)
  • P. Dal Monte et al.

    Analysis of intracellular and intraviral localization of the human cytomegalovirus UL53 protein

    J. Gen. Virol.

    (2002)
  • A.J. Davison et al.

    The order Herpesvirales

    Arch. Virol.

    (2009)
  • P.J. Desai et al.

    Reconstitution of the Kaposi's sarcoma-associated herpesvirus nuclear egress complex and formation of nuclear membrane vesicles by coexpression of ORF67 and ORF69 gene products

    J. Virol.

    (2012)
  • T.A. Dittmer et al.

    The lamin protein family

    Genome Biol.

    (2011)
  • R.J. Eisenberg et al.

    Herpes virus fusion and entry: a story with many characters

    Viruses

    (2012)
  • A. Farina et al.

    BFRF1 of Epstein–Barr virus is essential for efficient primary viral envelopment and egress

    J. Virol.

    (2005)
  • A. Farnsworth et al.

    Herpes simplex virus glycoproteins gB and gH function in fusion between the virion envelope and the outer nuclear membrane

    Proc. Natl. Acad. Sci. U. S. A.

    (2007)
  • E. Fossum et al.

    Evolutionarily conserved herpesviral protein interaction networks

    PLoS Pathog.

    (2009)
  • W. Fuchs et al.

    Essential function of the pseudorabies virus UL36 gene product is independent of its interaction with the UL37 protein

    J. Virol.

    (2004)
  • W. Fuchs et al.

    The interacting UL31 and UL34 gene products of pseudorabies virus are involved in egress from the host-cell nucleus and represent components of primary enveloped but not mature virions

    J. Virol.

    (2002)
  • R. Gonnella et al.

    Characterization and intracellular localization of the Epstein–Barr virus protein BFLF2: interactions with BFRF1 and with the nuclear lamina

    J. Virol.

    (2005)
  • H. Granzow et al.

    Egress of alphaherpesviruses: comparative ultrastructural study

    J. Virol.

    (2001)
  • H. Granzow et al.

    The pseudorabies virus US3 protein is a component of primary and of mature virions

    J. Virol.

    (2004)
  • H. Granzow et al.

    Ultrastructural analysis of the replication cycle of pseudorabies virus in cell culture: a reassessment

    J. Virol.

    (1997)
  • K.S. Grimm et al.

    Analysis of viral and cellular factors influencing herpesvirus-induced nuclear envelope breakdown

    J. Virol.

    (2012)
  • D. Grunwald et al.

    Nuclear export dynamics of RNA–protein complexes

    Nature

    (2011)
  • S. Hamirally et al.

    Viral mimicry of Cdc2/cyclin-dependent kinase 1 mediates disruption of nuclear lamina during human cytomegalovirus nuclear egress

    PLoS Pathog.

    (2009)
  • A.C. Haugo et al.

    Herpes simplex virus 1 pUL34 plays a critical role in cell-to-cell spread of virus in addition to its role in virus replication

    J. Virol.

    (2011)
  • L. Hertel et al.

    Viral and cell cycle-regulated kinases in cytomegalovirus-induced pseudomitosis and replication

    PLoS Pathog.

    (2007)
  • L. Hertel et al.

    Global analysis of host cell gene expression late during cytomegalovirus infection reveals extensive dysregulation of cell cycle gene expression and induction of Pseudomitosis independent of US28 function

    J. Virol.

    (2004)
  • M.W. Hetzer

    The nuclear envelope

    Cold Spring Harb. Perspect. Biol.

    (2010)
  • F.L. Homa et al.

    Capsid assembly and DNA packaging in herpes simplex virus

    Rev. Med. Virol.

    (1997)
  • Cited by (0)

    View full text