Abstract

Members of high (22-, 22.5-, 24-, and 34-kDa) and low (18-kDa) molecular mass forms of fibroblast growth factor-2 (FGF-2) regulate cell proliferation, differentiation, and migration. FGF-2s have been previously shown to accumulate in the nucleus and nucleolus. Although high molecular weight forms of FGF-2 contain at least one nuclear localization signal (NLS) in their N-terminal extension, the 18-kDa FGF-2 does not contain a standard NLS. To determine signals controlling the nuclear and subnuclear localization of the 18-kDa FGF-2, its full-length cDNA was fused to that of green fluorescent protein (GFP). The fusion protein was primarily localized to the nucleus of COS-7 and HeLa cells and accumulated in the nucleolus. The subcellular distribution was confirmed using wild type FGF-2 and FGF-2 tagged with a FLAG epitope. A 17-amino acid sequence containing two groups of basic amino acid residues separated by eight amino acid residues directed GFP and a GFP dimer into the nucleus. We systematically mutated the basic amino acid residues in this nonclassical NLS and determined the effect on nuclear and nucleolar accumulation of 18-kDa FGF-2. Lys(119) and Arg(129) are the key amino acid residues in both nuclear and nucleolar localization, whereas Lys(128) regulates only nucleolar localization of 18-kDa FGF-2. Together, these results demonstrate that the 18-kDa FGF-2 harbors a C-terminal nonclassical bipartite NLS, a portion of which also regulates its nucleolar localization.

Highlights

  • Members of high (22, 22.5, 24, and 34-kDa) and low (18-kDa) molecular mass forms of fibroblast growth factor-2 (FGF-2) regulate cell proliferation, differentiation, and migration

  • Lys119 and Arg129 are the key amino acid residues in both nuclear and nucleolar localization, whereas Lys128 regulates only nucleolar localization of 18-kDa FGF-2. These results demonstrate that the 18-kDa FGF-2 harbors a C-terminal nonclassical bipartite nuclear localization signal (NLS), a portion of which regulates its nucleolar localization

  • These results indicate that the 17-mer containing the NLS of 18-kDa FGF-2 can direct green fluorescent protein (GFP) and the GFP dimer into the nucleus but not as efficiently as full-length FGF-2

Read more

Summary

EXPERIMENTAL PROCEDURES

Plasmid Construction—To construct pEGFP-N1-FGF-2 encoding the fusion protein GFP-FGF-2, the full-length cDNA of human 18-kDa FGF-2 The amplified DNA was digested with BsrGI and NotI and ligated to the corresponding sites at the 3Ј-end of GFP cDNA in pEGFP-N1 (Clontech). The full-length cDNA of FGF-2 was placed at the 5Ј-end of GFP cDNA by first amplifying it using forward (5Ј-TTTAAGCTTATGGCAGCCGGGAGCATCACC-3Ј) and reverse (5Ј-TTTCTGCAGGCTCTTAGCAGACATTGGAAG-3Ј) primers containing HindIII and PstI sites, respectively. To generate pNLS-dEGFP-N1, in which the NLS of 18-kDa FGF-2 is fused to a GFP dimer, GFP cDNA was amplified with primers containing BsrGI sites. Site-directed Mutagenesis—To obtain a plasmid encoding wild type 18-kDa FGF-2, two stop codons were introduced between FGF-2 and GFP cDNA in pFGF-2-EGFP-N1 using the QuikChange® site-directed mutagenesis kit (Stratagene) according to the manufacturer’s instruction. Mutations K128G and R129G were introduced into the mutant R116G/R118G to generate R116G/R118G/K128G and R116G/R118G/R129G, respectively, using the corresponding primers described above. The secondary antibody was detected using SuperSignal chemiluminescent substrate (Pierce) according to the manufacturer’s instructions

RESULTS
Protein sequence
Findings
DISCUSSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call