This study demonstrates that p10.8 targets the nucleus of transfected cells and that this targeting does not require other viral factors, thereby suggesting that the viral protein itself has the ability to translocate across the nuclear pore complex (NPC) and accumulate in the nucleus. Because p10.8 has a mass of ~10.8 kDa, it was possible that it might enter the nucleus by passive diffusion; however, we showed that p10.8 is actively imported into the nucleus, because (1) GFPp10.8 and fragments GFPp10.8(1–40) and GFPp10.8(1–65) were able to transport the GFP protein into the nucleus, but this property was not displayed by fragments p10.8 (65–95) and p10.8(40–95); (2) GFP(1–40) could translocate GST to the nucleus, and molecular mass of GFPp10.8(1–40)GST exceeded the limit for free diffusion through NPCs. Our data thus suggest that p10.8 is imported into the nucleus through the NPC via a signal-mediated nuclear localization.
The active import of proteins into nuclei requires NLSs [11–15]. The best-characterized transport sequences of classic NLSs comprise one or two short stretches of basic lysine- or arginine-rich residues. Recently, a variety of nonconforming NLSs that are not particularly rich in lysine or arginine have been identified in various viral and cellular proteins [16–19]. However, in addition to linear NLSs, discontinuous epitopes or a tertiary structure to contribute to the nuclear import of proteins have been described [20–22]. Our study indicate that the nuclear targeting signal, p10.8(1–40), does not contain any stretches of basic residues that typify the classic NLS sequences, even if it is highly conserved among MDRV. Thus, p10.8(1–40) shows characteristics that distinguish it from classic NLSs. Moreover, nuclear target motif experiments demonstrated that three deletion constructs, GFP(1–10), GFP(10–30), and GFP(30–40) are indispensable, suggesting that full sequences or a complex structure of p10.8(1–40) appears to be essential for nuclear import activity. Our finding might be in keeping with the hypothesis, derived from the import of STAT1 and pUL84, that not only non-canonical sequences but complex folding may generate NLSs [21, 23]. Report from well-described human ribonucleoprotein A1 import signal, M9 domain, indicate that aromatic amino acids may generate NLSs . Our sequence analysis of p10.8(1–40) revealed that it is rich in aromatic amino acids (Figure 2C). Determination of mutation of these aromatic amino acids may help to identify whether they are responsible for the active nuclear import of the MDRV p10.8 protein. Recently, another non-conventional nuclear transport mechanisms besides the nuclear pore itself have been well described . This might suggest future experiments to identify the exact transport mechanism used by p10.8.
Similar to nuclear import, the export of a protein from the nucleus depends on the presence of a specific signal, nuclear export signal (NES), with a leucine-rich motif [12, 13]. Interestingly, inspection of the p10.8 sequence revealed the presence of one leucine-rich motif (33LSVLAELSDLFDLAI47) that matches the consensus for leucine-rich export sequences (Figure 2C). It will therefore be of interest to investigate whether this motif constitutes a functional nuclear export sequence and, consequently, whether p10.8 is a nucleocytoplasmic shuttling protein.
The finding that MDRV, which replicates exclusively in the cytoplasm of the infected cell, expresses a nuclear protein was not surprising, since the small nonstructural protein p17 of avian reovirus and the 14-kDa nonstructural protein σ1s of mammalian reovirus have also been shown to accumulate in the nucleus of transfected and infected cells [6, 7]. However, a comparative analysis of the deduced amino acid sequences of p10.8, p17, and σ1s revealed no significant similarities in their primary sequences and showed no conserved functional motifs. Yet, the fact that both p10.8 of MDRV and σ1s of mammalian reovirus can localize to the nucleus and cause apoptosis of infected or transfected cells [7, 8], suggests that σ1s and p10.8 may be functionally related. Cytoplasmic reovirus infection profoundly affects the host cell nucleus and its functions. In MDRV-infected cells, the p10.8 protein localizes to the nucleus indicate that p10.8 protein may accomplish an important function inside the nucleus during the phase of the viral replication and ultimately influence disease pathogenesis in the infected host.