The presented study shows that cellular ORFV infection leads to structural dispersion of the Golgi/TGN compartments and enrichment of COP-I vesicular structures. These processes are accompanied by an increase in the steady state expression of β-COP (Figure b), defective carbohydrate trimming of MHC I within the Golgi (Figure b), reduction of surface expressed MHC I molecules and a prolonged half-life of pre-existing MHC I on the plasma membrane (Figure ). Upcoming studies have to prove whether the described interferences of ORFV with the MHC I expression also occur in natural host cells.
Our findings demonstrate that in ORFV-infected cells the intra-Golgi- and endosome/TGN-transport of MHC I was severely disturbed. ORFV seems to utilize early gene expression to block MHC I export within the late secretory route and thereby reduces MHC I surface expression. As shown by our experiments ORFV alters the perinuclear localization as well as the overall structure of the Golgi and TGN in infected Vero cells. Similar effects on the Golgi have also been described for a variety of different viruses. Early gene expression of Varicella zoster virus leads to MHC I down-regulation by impairing its transport to the cell surface [23
]. A late event in the reproductive cycle of Herpes simplex virus type 1 causes fragmentation and dispersal of the Golgi in infected Vero cells, which coincides with virion assembly [24
]. The infection with human rhinovirus 1A (HRV-1A) induces Golgi-fragmentation into vesicles that appear to be used as a substrate for viral RNA replication [25
]. Another positive-strand RNA virus, the poliovirus, induces dramatic disruption of the Golgi with consequences for the secretory complex [26
]. Furthermore, it is known that vaccinia virus becomes enwrapped by cisternae derived from the intermediate compartment between ER and Golgi stacks as well as the TGN [28
]. Recently Tan et al. also observed fragmentation of the Golgi during ORFV infection, and reported the Golgi localization of an ORFV envelope protein during late stage of infection [29
]. The authors suggested that it is concealed between two Golgi membranes, which are forming wrapped mature virions. In the present study, the destruction of the Golgi structure is clearly not linked to virus envelope formation since the observed structural modifications are also visible in the presence of AraC, which prevents the expression of late ORFV genes essentially required for the virus envelope.
ORFV-infected cells are characterized by a reduced amount of newly synthesized MHC I on the plasma membrane as well as a prolonged half-life of the remaining pre-existing surface MHC I molecules (Figure ). Down-regulation of MHC I is clearly AraC-insensitive and thus apparently linked to the expression of early ORFV genes whereas it cannot be excluded that the observed MHC I half-life effect might be also controlled by late ORFV gene expression. It is tempting to speculate that the respective viral gene products target compartments within the late secretory route. Since structural and functional integrity of the TGN are essentially required for endosomal/TGN-trafficking, the observed disruption of the TGN in infected cells (Figure ) might be suspected to interfere with endocytosis as well as endosomal recycling of MHC I. A similar phenotype has been described for the HPV16 protein E5 [30
], which mediates disruption of the exo- and endocytic trafficking, including transport of the MHC I [30
], which causes reduced MHC I surface presentation and extends the half-life of the remaining MHC I molecules on the plasma membrane (M. R. Knittler, manuscript in preparation).
The ORFV infection leads to an accumulation of MHC I in COP-I vesicles (Figure a). COP-I is the cytoplasmic membrane-coat complex (coatomer) of seven distinct proteins and is required for both anterograde and retrograde transport in the secretory pathway [31
]. The observation that ORFV infection increases the cellular expression levels of β-COP (Figure b) and the amount of COP-I vesicular structures suggests inhibition of uncoating of COP-I vesicles by ORFV. The identification of responsible ORFV protein(s), as found in Coxsackievirus [33
], requires further detailed studies. In contrast to vaccinia virus, which hijacks the COP-I coatomer for viral particle formation [34
], no correlation between accumulation of COP-I vesicles and viral biogenesis was observed, since the ORFV-mediated effect was also detectable in the presence of AraC.
The Endo H-experiments suggest that destruction of Golgi and TGN structures as well as intracellular accumulation of MHC I in COP-I vesicles is accompanied by impaired post-ER maturation of the N-linked carbohydrates of MHC I. In contrast to non-infected cells, a substantial amount of the MHC I molecules exhibits partial Endo H-resistance in ORFV-infected cells indicating that these molecules are not correctly processed by carbohydrate-trimming within Golgi. This reminds of the defective maturation of MHC I in the presence Concanamycin B, a specific inhibitor of the vacuolar type H(+)-ATPase [35
], suggesting that ORFV infection not only affects the intracellular location and structure of Golgi and TGN, but also the functional pH conditions within these two compartments.
In addition to MHC I, ORFV infection also interferes with the surface expression of the transferrin receptor (TfR, CD71) (data not shown), which suggests that the ORFV-induced reduction of MHC I-antigen presentation is mediated by subversion of the host cell export machinery and not via specific targeting of MHC I molecules. Thus, one could assume that the ORFV-mediated modulation of vesicular transport has a more pleiotropic effect that also includes the reduction of antigen presentation and thereby provides an immune subversion strategy in advantage of the viral pathogen.
ORFV does clearly not interfere with the expression of MHC I molecules (Figure a) but uses an evasion strategy that accumulates newly synthesized MHC I molecules within the late secretory pathway (COP-I vesicles) possibly to down-modulate MHC I presenting viral antigens (for evasion of a cytotoxic T cell -mediated response), while simultaneously increasing the half-lives of pre-existing self peptide MHC I complexes at the plasma membrane (for evasion of an NK cell-mediated response). This suggests that ORFV like other large DNA viruses (e.g. Herpesviruses) uses different evasion strategies to interfere with antigen presentation at different levels of MHC I processing.