RESPONSE
A response to this article has been published. To view the response, click here.

Comment on Phosphorylation adjacent to the nuclear localization signal of human dUTPase abolishes nuclear import: structural and mechanistic insights by Róna et al. (2013)

The authors comment on the article by Róna et al. [(2013), Acta Cryst. D69, 2495–2505].

Phosphorylation can regulate nuclear targeting of classical nuclear localization signal (NLS)-bearing cargoes by modulating their affinity for the cellular transporters of the importin (IMP) superfamily, which is critical to many important biological processes (Jans et al., 2000; Poon & Jans, 2005). In particular, phosphorylation within or immediately upstream of the NLS can prevent IMP recognition and nuclear import (Jans et al., 1991, 1995), whereas distinct phosphoryl­ation generally up to 30 amino acids upstream of the NLS can promote IMP recognition/nuclear import (Hübner et al., 1997; Rihs & Peters, 1989). This pattern, originally described for the prototypal NLS identified from Simian Virus 40 Large T-antigen (T-ag) (Kalderon et al., 1984; Rihs & Peters, 1989) was subsequently proven to apply to a plethora of proteins of viral and cellular origins that mediate important biological effects, potentially representing a target for therapeutic intervention (Alvisi et al., 2005, 2011, 2008, 2013; Poon & Jans, 2005).

However, the mechanistic details of how cargo-phosphorylation really influences IMP binding remained elusive until now. In the case of phosphorylation enhancing nuclear transport events, the crystal structure of a complex of a truncated form of IMPα and a phosphorylated peptide from the prototypical NLS of SV40 T-ag was solved, but revealed no direct interaction between the phosphoryl­ated cargo and IMPα (Fontes et al., 2003), suggesting that the effect of phosphorylation in influencing IMP binding in this context might only be observed in the full-length protein (Alvisi et al., 2008, 2013); clearly crystal structures of the full IMPα/β heterodimer complexed with a whole protein, rather than an NLS peptide, is ultimately required to enable understanding of how phosphorylation can indeed promote IMP recognition and nuclear import.

The study from Róna et al. (Róna et al., 2013) in the August 2013 issue of Acta Crystallographica Section D provides new structural insight into the nature of the phosphorylation-regulated recognition of cargoes by IMPs. In particular, the structure of phosphomimetic and phosphonull-NLS peptides derived from human dUTPase (DUT) complexed with a truncated form of IMPα is correlated with IMP-NLS binding ability. Critically, the phosphomimetic S11E mutation altered the conformation of R15 DUT-NLS, leading to a change in the binding arrangement of the peptide in the IMPα binding site, and to a loss of interaction between the P12 and R15 NLS residues and the IMPα surface. Róna et al. (Róna et al., 2013) conclude that this conformational change explains the ca tenfold lower binding affinity of the phosphomimetic peptide to the IMPα subunit, as observed in their in vitro experiments.

These findings are of great interest in the nuclear transport field because they provide a structural explanation of how phosphoryl­ation might negatively affect the NLS-IMPα interaction. However, a more careful consideration of the biological relevance of the results is justified since the present study seems to ignore the fact that other proteins exist in the cell that can bind NLSs/NLS-like sequences. In short, Róna et al. (2013) do not consider the possibility that phosphorylation of DUT S11 may regulate DUT's interaction with other cellular factors.

In a seminal paper, Hodel et al. (2001) clearly showed that only mutations diminishing the binding affinity of IMPα to its NLS bearing cargo by two orders of magnitude significantly impact on nuclear targeting (Hodel et al., 2001); mutations impairing recognition by up to tenfold in fact, have little effect on nuclear accumulation (Hodel et al., 2001; Harreman et al., 2004). Thus, nuclear exclusion of the S11E-DUT, rather than being explicable solely in terms of reduced binding to IMPα, may in fact indicate that other factors are playing an important role in the the observed cytoplasmic localization in cells.

Cytoplasmic retention factor(s) that contribute to negatively regulate NLS activity upon phosphorylation was postulated over 20 years ago (Jans et al., 1991), and more recently shown to be a key factor limiting nuclear accumulation of both SV40 T-ag (through the T124 phosphorylation site) and Human Cytomegalovirus processivity factor UL44 (through T427) (Fulcher et al., 2010), both of which closely resemble S11 of DUT in terms of position relative to the NLS core of the respective proteins (see Fig. 1). The cytoplasmic retention factor involved in this case is BRCA-1 binding protein 2 (BRAP2) (Fulcher et al., 2010).

Figure 1 DUT, T-ag and UL44 NLSs all possess phosphorylation sites that negatively regulate nuclear import adjacent to the basic NLS. Phosphorylation sites are underlined, basic residues within NLS core are in bold, the single-letter amino-acid code is used. Numbers indicate amino-acid position within the indicated proteins.

Róna et al. (Róna et al., 2013) should consider the possible contribution of negative regulators of nuclear targeting within eukaryotic cells, which compete with IMPs to finely tune the nuclear levels of certain cargoes. Given the modest (ca tenfold) reduction in IMPα binding of the entirely cytosolic S11R-DUT as compared to entirely nuclear S11A-DUT, and the fact that in light of previous work (Hodel et al., 2001; Harreman et al., 2004) this is unlikely to impact so strongly on nuclear targeting, it seems feasible that S11 phosphorylation may confer interaction with a factor such as BRAP2, and that this may be the mechanism responsible for the strong cytoplasmic localization.

Given the importance of structural information to our understanding of key physiopathological processes such as gene expression, cell growth and transformation, and virus–host interactions, the work of Róna et al. (Róna et al., 2013) is of great importance. We now wait with expectation for new, highly informative crystal structures of IMPs bound to full cargoes, and arguably of even more interest will be crystal structures of negative regulators of nuclear import such as BRAP2 to phosphorylated NLS-containing cargoes.