• 2018-07
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • 2021-10
  • 2021-11
  • 2021-12
  • 2022-01
  • br NLS and NES Mapping br Sequence


    NLS and NES Mapping
    Sequence Comparison of Gli NLSs—NLSs in ZF Domains Alignment of the Gli NLSs of representative metazoan animals suggests that the amino Sodium salicylate australia residues in the ZF domains are strongly conserved (Figure 4.2, Figure 4.3B). However, the length of the intervening amino acid sequence between the first and second basic clusters is shorter in sea anemone, Nematostella vectensis (11 amino acid residues), and sponge, Amphimedon queenslandica (8 amino acid residues), than in bilateria. Bipartite classical NLSs generally have 10–12 residues between the sequences that separate the two basic clusters (Dingwall and Laskey, 1991). Whether or not the sponge sequence acts as an NLS is not clear at this point. Gli/Ci NLSs are partly conserved in Glis3, but not in Glis2 and Zic3 (Fig. 4.3C). Comparison of the Gli/Ci NLSs with the Glis and Zic NLSs is intriguing because Gli, Glis, and Zic proteins are likely derived from a common metazoan ancestor (Hatayama and Aruga, 2010). These gene families are widely distributed in metazoan animals, except that placozoa lack Gli but possess Zic (; ) (Fig. 4.3A). This fact suggested that metazoan common ancestor had already acquired Gli/Glis, and Zic may have been innovated in eumetazoan ancestor. The divergence of NLS sequences may partly reflect this evolutional history. In a comparison between the eumetazoan Gli/Ci proteins, the two basic residue clusters in the NLS are strongly conserved (Fig. 4.3C), suggesting a biological significance of the bipartite Gli/Ci NLS motif. The Zic NLS sequence is not a classical NLS (dispersed NLS where NLS-forming basic residues does not form a cluster), but in its 3D conformation, the basic residues are positioned similar to those in the bipartite nucleoplasmin NLS (). The corresponding region in the Gli ZF5+C-region NLS lacks the second cluster and acts as a supplementary NLS (; ). Thus, the position of the NLS clearly differs between Gli and Zic proteins. Glis ZF sequences can be divided into two groups (data not shown): Glis1/3 and Glis2. Our sequence alignment suggests that basic amino acid residues in the Gli/Ci NLS region are partly conserved in Glis1/3, but not in Glis2 (Fig. 4.3C). reported that the Glis3 NLS was located in ZF4 and that removal of the Gli NLS-corresponding region did not affect nuclear localization. The Glis2 NLS is located in the ZF3 region (237–264) (). Therefore, it can be concluded that the NLSs in Gli/Ci, Glis2, Glis3, and Zic3 are not identical. It is intriguing that the NLSs in the closely related ZF domains are divergent. This presumably reflects the structural restriction of ZFs due to their multifunctional properties. C2H2-type ZF domain NLSs have recently been mapped in the snail (; ) and in Sp1 proteins (). In the snail, the NLS was mapped to a ZF domain that directly binds Importin β (). Further, mutational analysis of the basic residues in the snail ZF domain showed that five out of six basic residues in defined positions in the three consecutive ZFs are essential for nuclear localization (). Interestingly, the positions of six basic residues are conserved in C2H2 ZF proteins, including the Gli, Glis, and Zic proteins. The ZF C-terminal flanking basic cluster in the bipartite Gli NLS may have been uniquely acquired over the course of evolution, while the other ZF NLSs may have diversified using basic residues scattered within the ZF.
    Predicted 3D Structure of Gli NLS The crystallographic 3D structure was reported for the GLI1 ZF (Pavletich and Pabo, 1993). To visualize the structure of the NLS, we modeled the 3D structure of the human GLI1 bipartite NLS using Modeller software (version 9.9; (). In our model, the fifth ZF domain folded into a solid structure, but the flanking C-terminal region had an indeterminate structure (data not shown). This result suggests that the bipartite Gli1 NLS outside of ZF5 is flexible. Interestingly, one out of ten Gli1 NLS models fitted well with the cap-binding protein 80 NLS (CBP80; PDBID 3FEX) in terms of side-chain position of the NLS-constituting basic amino acid residues (Fig. 4.4A).