Comparative mutagenesis of nuclear localization signals reveals the importance of neutral and acidic amino acids

Abstract

Nuclear proteins contain information within their primary structures which causes them to accumulate selectively in the nucleus [1,2] by associating with the cytosolic receptor importin [3]. The α subunit of importin binds the nuclear localization signal (NLS), and the β subunit docks at the nuclear pore complex. The NLS of the simian virus 40 large T-antigen (SV40 T-ag) is a single cluster of basic amino acids (PKKKRKV132; single-letter code, the basic amino acids are shown in bold; [4,5]), whereas the NLS of nucleoplasmin is bipartite. The nucleoplasmin NLS requires two essential clusters of basic amino acids, separated by a mutation-tolerant spacer (KRPAATKKAGQAKKKK171; [6,7]). A SwissProt database search shows that more than 50% of nuclear proteins contain a match to this consensus, and many NLSs have since been found to conform to this type of motif in yeast, plants and animals [8–10]. A different NLS (PAAKRVKLD) has been reported in the oncoprotein c-Myc, but it has received little attention because, unlike other known NLSs, only three of nine residues are basic [11], and one residue is even acidic. Here, we report that constructs containing an inactive basic cluster downstream of the bipartite signal of nucleoplasmin can be directed to the nucleus by flanking them with specific neutral and acidic residues taken from the signal reported for c-Myc. Nuclear targeting by the single cluster KKKK is dependent on it being preceded by PAA and is stimulated if it is followed by the dipeptide LD. The relative positions of these elements are crucial to the function of these NLSs. All regions of the unconventional signal of c-Myc are functionally important. Contrary to conventional views, neutral and even acidic amino acids can play crucial roles in NLSs.