The large tegument protein pUL36 is essential for formation of the capsid vertex-specific component at the capsid-tegument interface of herpes simplex virus 1.
Bottom Line: In addition, the presence of full-length pUL36 results in weak density that extends the CVSC toward the penton, suggesting either that this extra density is formed directly by pUL36 or that pUL36 stabilizes other components of the vertex-tegument interface.Herpesviruses have complex particles that are formed as a result of a carefully controlled sequence of assembly steps.We show that the largest viral protein, pUL36, which occupies the layer of tegument closest to the capsid, is essential for formation of structurally normal connections to the capsid.
Affiliation: MRC-University of Glasgow Centre for Virus Research, Glasgow, United Kingdom.Show MeSH
Related in: MedlinePlus
Mentions: The apparent requirement for pUL36 was unexpected, since it was not believed to contribute to the CVSC density (38, 39). As this analysis was carried out on cytoplasmic capsids, one possibility is that the CVSC becomes destabilized during passage of capsids across the nuclear envelope and reforms in the cytoplasm only if pUL36 is present. To determine whether this was the case, nuclear C-capsids were prepared and reconstructed to final resolutions of 24 Å for the WT and ARΔUL36 and 22 Å for FRΔUL37. In each case the structure recapitulated that seen for cytoplasmic C-caspids, with normal-size CVSCs in the WT and FRΔUL37 but a smaller CVSC in ARΔUL36 (Fig. 1C). Thus, it appears that the effects seen in this study are not related to the subcellular origin of the capsid but are intrinsic to the mutants analyzed. These results imply that the presence or absence of pUL36 has a direct effect on the nature of the CVSC density and, in addition, that this effect is realized in both the nucleus and the cytoplasm. However, the role of the inner tegument proteins in maturation of nuclear capsids remains controversial, and the possibility of cross-contamination with capsids from the cytoplasm cannot be ruled out. Such contamination would not be unexpected given the relative crudity of the method used for nuclear isolation. Indeed, cosedimentation of WT and FRΔUL37 cytoplasmic capsids with isolated nuclei had been reported to account for the poor yields seen in previous analyses (20). To investigate this, we employed a gene UL34 deletion mutant of HSV-1 (vRR1072) (42). Together with its binding partner pUL31, pUL34 is essential for primary envelopment of capsids. Capsid assembly and DNA packaging occur normally in the absence of pUL34, but the capsids are retained in the nucleus and have no exposure to cytoplasmic modifications (42). Both pUL31 and pUL34 become transiently associated with perinuclear virions during nuclear egress but are not components of the mature virion (56, 57). To confirm the block on nuclear egress, we examined the distribution of capsids in thin sections of infected cells. Cells were infected at 37°C with 5 PFU/cell of vRR1072 for 20 h and then fixed and prepared for electron microscopy as described previously (20). As expected, capsids were present in both the nucleus and cytoplasm in UL34-expressing Vero (143/1099E) cells (see Fig. S2A in the supplemental material) but were seen only in the nucleus in control Vero (see Fig. S2B) and BHK (see Fig. S2C) cells. Therefore, C-capsids were isolated from the nuclei of vRR1072-infected BHK cells and reconstructed to a final resolution of 22 Å. As with all the other types of C-capsid described here, vRR1072 capsids had CVSC density around the vertices. In this case, the density was intermediate in size between the archetypal WT and FRΔUL37 CVSCs and the small CVSC of ARΔUL36. This is clearly illustrated by the maps shown in Fig. 4. The appearance of the CVSC after superimposition of the vRR1072 map on those of the B-capsid (Fig. 4A) and the ARΔUL36 capsid (Fig. 4B) was identical, reflecting the small size of the ARΔUL36 CVSC density, which is completely contained within that of vRR1072. Similarly, the maps of vRR1072 compared with those of the WT (Fig. 4C) and FRΔUL37 (Fig. 4D) are essentially the same. However, in these cases the WT and FRΔUL37 CVSCs are considerably more extensive than that present in vRR1072, with the additional densities representing the proximal part of the CVSC, which extends toward the penton.
Affiliation: MRC-University of Glasgow Centre for Virus Research, Glasgow, United Kingdom.