Index Introduction Materials and Methods Cell culture and transfection Plasmid construction Western immunoblot analysis Subcellular localization assays Results and Discussion Subcellular localization of full-length dMLF and its deletion mutants Identification of NLS motifs Acknowledgements References

Introduction

The myeloid leukemia factor (MLF) family is a group of proteins newly identified in human, mouse and fly (Kuefer et al., 1996; Ohno et al., 2000; Williams et al., 1999; Yoneda-Kato et al., 1996). The human MLF1, was first found in the form of a fusion protein with nucleophosmin (NPM) generated by the t(3;5)(q25.1;q34) chromosomal translocation, which is associated with myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) (Yoneda-Kato et al., 1996). It has been reported that MLF1 is associated with several proteins such as 14-3-3ζ, Madm, MLF1IP/KLIP1, Manp and CSN3, and that hMLF1 regulates the cell cycle via the CSN3/COP1 pathway (Hanissian et al., 2004; Lim et al., 2002; Winteringham et al., 2006; Yoneda-Kato et al., 2005), although the biochemical activity of MLF1 has not been well characterized.

The cDNA for Drosophila MLF (dMLF) was originally cloned by two-hybrid screening using DNA replication-related element-binding factor (DREF) as the bait (Ohno et al., 2000). DREF is a transcription factor in Drosophila that regulates proliferation-related genes such as PCNA, Cyclin A and DNA polymerase α (Hirose et al., 1993; Hirose et al., 1996; Ohno et al., 1996; Takahashi et al., 1996). The dMLF protein has also been reported to interact physically and genetically with dSu(fu), a negative regulator of the Gli/Ci transcription factor (Fouix et al., 2003). Furthermore, in the Drosophila model of polyglutamine disorders, overexpression of dMLF suppresses the toxicity of abnormally long polyglutamine tracts expressed in the eye and central nervous system (Kazemi-Esfarjani and Benzer, 2002). These observations suggest that dMLF plays multiple roles in vivo. However, dMLF contains no recognizable domains or motifs except for a putative 14-3-3 protein binding site (Ohno et al., 2000).

MLF1 has been described as mostly cytoplasmic, but it is also present within nuclear spots in mammalian cells (Williams et al., 1999; Yoneda-Kato et al., 1996). In contrast, in Drosophila dMLF localizes mainly in the nucleus and seems to be associated with chromosomes in polytenic tissues (Fouix et al., 2003). It has been reported that four isoforms of the dMLF protein are present that are probably the result of alternative splicing and that two forms of dMLF (dMLFA and dMLFB) differ in their subcellular localizations (Martin-Lannerée et al., 2006). In addition, MLF1 has been reported to translocate between the nucleus and cytoplasm (Winteringham et al., 2006; Yoneda-Kato and Kato, 2008).

Nuclear import and export of proteins occurs by passive diffusion, active transport or by binding in a complex with an actively transported protein (Hicks and Raikhel, 1995; Lange et al., 2007). Directed nuclear entry of a protein is determined by the presence of nuclear localization signals (NLSs) that recognize and associate with the nuclear import receptors. Although there is no strict consensus sequence for all NLSs, in general they contain a high content of the basic amino acids arginine and lysine and may contain residues, such as proline. The best-characterized nuclear targeting signal is the classical nuclear localization signal, which can be monopartite or bipartite. The monopartite NLS contains a single cluster of four to six basic amino acids, typified by the SV40 large T antigen NLS, while a bipartite NLS contains two stretches of basic amino acids separated by a stretch of 10–12 spacer residues (Nigg, 1997). These classical targeting sequences can be, and often are, identified by computer searches. However, in many cases the function of these sequences is not experimentally verified.

In order to identify the region required for dMLF nuclear localization, we investigated the nuclear/cytoplasmic distribution of a variety of fusion proteins constructed through deletion and site-directed mutagenesis. We demonstrated that the dMLF protein contains NLS motifs within its C-terminal region.

Materials and Methods

Cell culture and transfection

Kc cells derived from Drosophila melanogaster embryos were grown at 25°C in M3 medium (Sigma, St Louis, MO) supplemented with 2% fetal calf serum in the presence of 5% CO₂. Cells were transfected with pAct5C-Gal4 as the effector plasmid and pUAST constructs using CELLFECTIN reagent (Invitrogen, Carlsbad, CA) and harvested 22 h after transfection.

Plasmid construction

To construct the plasmids pUAS-EGFP and a series of plasmids expressing EGFP-dMLFA (EGFPdMLF_1–309, EGFPdMLF_53–309, EGFPdMLF_96–309, EGFPdMLF_203–309, EGFPdMLF_270–309, EGFPdMLF_1–202, EGFPdMLF_1–252, EGFPdMLF_1–269, EGFPdMLF_230–269), the cDNA fragments were amplified by polymerase chain reaction (PCR) and inserted into the pUAST vector (Brand and Perrimon, 1993). For the construction of the plasmids pUAS-EGFPdMLFNLS1, pUAS-EGFPdMLFNLS2 and pUAS-EGFPdMLF_307–309, oligonucleotides were synthesized. EGFPdMLFNLS1 oligonucleotide, 5'-GATCCCGAGGCGCCAACAACGTGCCGTAAAACATTTTCACTAG-3' and 5'-TCGACTAGTGAAAATGTTTTACGGCACGTTGTTGGCGCCTCGG-3'. EGFPdMLFNLS2 oligonucleotide, 5'-GATCTACAAGGCGGTCAAGCGCGGCAAGAAGAAGTAG-3' and 5'-TCGACTACTTCTTCTTGCCGCGCTTGACCGCCTTGTA-3'. EGFPdMLF_307–309 oligonucleotide, 5'-GATCAAGAAGAAGTAG-3' and 5'-TCGACTACTTCTTCTT-3'. For NLS mutants, site-directed mutagenesis was performed using the QuickChange method (Stratagene, La Jolla, CA). The oligonucleotides used were: pUAS-EGFPdMLFNLS1mut (5'-CCCTATGCCGCCAATCCGGCGGCCCAACAAGCTGCCGTAAAACAT-3' and 5'-ATGTTTTACGGCAGCTTGTTGGGCCGCCGGATTGGCGGCATAGGG-3'), pUAS-EGFPdMLFNLS2mut (5'-GCGGCCAGCTCCTACGCGGCGGTCGCGGCCGGCAAGAAGAAGTAG-3' and 5'-CTACTTCTTCTTGCCGGCCGCGACCGCCGCGTAGGAGCTGGCCGC-3'). The plasmid pUAS-dMLF was as described previously (Ohno et al., 2000).

Western immunoblot analysis

Whole cell extracts were prepared in sample buffer (125 mM Tris-HCl [pH 7.6], 4% SDS, 10% glycerol, 0.01% bromophenol blue, 10% β-mercaptoethanol) and boiled for 5 min. The eluted proteins were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to polyvinylidene difluoride (PVDF) membranes (Bio-Rad, Hercules, CA). The blotted membranes were incubated with mouse monoclonal anti-GFP antibody (Sigma, St Louis, MO) at 1:4,000 dilution. After washing, the membranes were incubated with horseradish peroxidase (HRP)-conjugated anti-mouse IgG (GE Healthcare, Buckinghamshire, England) at 1:10,000 dilution. Proteins were detected with the ECL blotting system (GE Healthcare, Buckinghamshire, England) according to the manufacturer’s instructions. The same amount of protein loading was confirmed by probing with anti-α-tubulin IgG (Sigma) at 1:2,000.

Subcellular localization assays

Kc cells grown on 35 mm dishes were fixed in 4% paraformaldehyde for 20 min at 25°C followed by washing with phosphate buffered saline (PBS) three times. The cells on coverslips were treated with 2 mg/ml RNaseA for 1 h at 25°C, washed extensively, and stained with 100 μg/ml propidium iodide (PI) for 10 min. The coverslips were mounted in Fluoroguard Antifade Reagent (Bio-Rad, Hercules, CA) and the cells were viewed by a LSM510 confocal laser-scanning microscope (Zeiss, Jena, Germany). For these assays transfections were performed on a minimum of two independent replicates.

Results and Discussion

Subcellular localization of full-length dMLF and its deletion mutants

Previously we found that dMLFA is largely found in the nucleus in Kc cells and embryos (Sugano et al., 2008). To refine the region of dMLFA required for the nuclear localization, we introduced expression plasmids encoding a series of dMLFA deletion mutants fused with enhanced green fluorescent protein (EGFP) into Kc cells (Fig. 1A). First we confirmed the localization of EGFP-dMLFA fusion protein (Fig. 1C). In transiently transfected Kc cells, full-length dMLFA protein (amino acids 1–309) localized exclusively in the nucleus, in agreement with previous reports on dMLFA localization (Martin-Lannerée et al., 2006). Next, we examined subcellular localization of six deletion mutants of dMLFA. These fusion proteins were expressed in cells at similar levels and degradation products of the fusion proteins were rarely observed (Fig. 1B). EGFP-dMLFA mutants lacking the N-terminal region (dMLF53–309, dMLF96–309, dMLF203–309, dMLF270–309) were detected in the nucleus, whereas mutants lacking the C-terminal region (dMLF1–202, dMLF1–269) were detected diffusively in both the cytoplasm and the nucleus (Fig. 1C). These results suggest that the amino acid region 270–309 is responsible for the nuclear localization of dMLFA. It has been reported that dMLFA shows nuclear localization while dMLFB localizes both in the nucleus and cytoplasm (Martin-Lannerée et al., 2006). The dMLFB isoform differs in the C-terminal region, and therefore the present data are consistent with the idea that the C-terminal amino acids 270–309 region plays a critical role in nuclear localization of the protein. Since the other variants (dMLFC and dMLFD (dMLF1 and dMLF2)) contain the amino acid sequence corresponding to the 270–309 region of dMLFA it is likely that these isoforms may also localize in the nuclei.

View Details

Fig. 1. Subcellular localization of EGFP-dMLFA full-length and deletion mutant fusion proteins. (A) Schematic representation of dMLFA wild-type and deletion mutants. (B) Each of the deletion mutants was tagged with EGFP and transiently expressed in Kc cells. Cell lysates were immunoblotted with the antibody to GFP or α-tubulin. (C) Confocal laser scanning microscopy of Kc cells transfected with EGFP-tagged proteins (green). Nuclei were stained with propidium iodide (PI, red). Scale bars, 10 μm.

The dMLF protein contains a putative 14-3-3 protein binding motif (Fig. 2A) (Ohno et al., 2000). The 14-3-3 family of proteins is expressed in a broad range of organisms and is highly conserved throughout evolution. The 14-3-3 proteins control the activity of their partner molecules by binding to the partner, and this binding often leads to an altered subcellular localization of the partner (Muslin and Xing, 2000). We therefore examined the effect of the 14-3-3 domain on the localization of dMLF protein (Fig. 2). EGFP-dMLF1–269 containing the 14-3-3 binding motif was detected in both the nucleus and cytoplasm, and the mutant without the motif (dMLF1–252) was also broadly dispersed throughout the cytoplasm (Fig. 2B). Furthermore, a dMLFA mutant containing the 14-3-3 binding motif (dMLF230–269) was distributed in both the nuclear and cytoplasm. These results indicate that the putative 14-3-3 protein binding motif is unlikely to be involved in the subcellular localization of dMLF, at least in Kc cells.

View Details

Fig. 2. Examination of the role of the putative 14-3-3 protein binding motif in dMLFA nuclear localization. (A) Schematic representation of dMLFA wild-type and deletion mutants. (B) Expression levels of EGFP-dMLF mutant proteins are shown by Western blotting with the antibody to GFP. The same amount of protein loading was confirmed by probing with anti-α-tubulin IgG. (C) Confocal laser scanning microscopy of Kc cells transfected with EGFP-tagged proteins (green). Nuclei were stained with PI (red). Scale bars, 10 μm.

Identification of NLS motifs

Although we scanned the amino acid sequence of dMLFA for potential NLS motifs (PSORTII program; http://psort.hgc.jp/), we found no recognizable motifs within dMLFA. However, the N-terminal deletion mutant dMLF270–309 that exclusively localized in nuclei (Fig. 1C) contained two short stretches of positively-charged basic amino acid residues (Fig. 3A). In many cases, NLS motifs are rich in basic amino acids. We therefore examined the role of these basic amino acid stretches in nuclear localization of dMLFA. This region contains two possible NLS motifs located at amino acids 280–291 (hereafter NLS-1) and amino acids 300–309 (hereafter NLS-2) (Fig. 3A). To examine whether or not the two NLS motif candidates might play a role in nuclear localization, we constructed site-directed mutants of the putative NLS domains in an EGFP-dMLF expression construct (Fig. 3A). Subsequently, NLS-mutant plasmids were transfected into Kc cells and the subcellular localization of the expressed proteins was assessed. These fusion proteins were expressed in cells at similar levels and degradation products of the fusion proteins were rarely observed (Fig. 3C). Mutation of the basic amino acid residues to alanine residues in putative NLS-1 strongly abrogated the nuclear localization activity of EGFP fusion protein (Fig. 3B). Furthermore, mutations in NLS-2 also resulted in cytoplasmic localization of the fusion protein (Fig. 3B). These data indicate that the mutation of either NLS site is sufficient for loss of nuclear localization.

View Details

Fig. 3. Mutational analyses of putative NLS sites in dMLFA. (A) Candidate NLS sites within the amino acids 270–309 region and the residues mutated by site-directed mutagenesis. (B) Confocal laser scanning microscopy of Kc cells transfected with EGFP-tagged proteins (green). Nuclei were stained with PI (red). Scale bars, 10 μm. (C) Expression levels of EGFP-dMLF mutant proteins as shown by Western blotting with the antibody to GFP. The same amount of protein loading was confirmed by probing with anti-α-tubulin IgG.

The putative NLS-2 region within dMLFA protein contains a cluster of basic amino acids in the C-terminal end of the protein (³⁰⁷KKK³⁰⁹) (Fig. 4A). It has been reported that ETS transcription factor family protein Elf3 contains a similar cluster within the ETS domain, and that this cluster is important for nuclear localization of the protein in mouse (Do et al., 2006). In order to investigate this activity for nuclear localization, we made an EGFP protein containing the KKK sequence EGFP-dMLF307–309 and determined its subcellular localization (Fig. 4A). This protein displayed diffuse localization in both nucleus and cytoplasm (Fig. 4C), suggesting that this amino acid region is not critical for the nuclear localization of dMLFA.

View Details

Fig. 4. Neither of the two putative NLS sites alone is sufficient for dMLFA nuclear localization. (A) Candidate NLS sites within the amino acids 270–309 region. (B) Expression levels of EGFP-dMLF mutant proteins as shown by Western blotting with the antibody to GFP. The same amount of protein loading was confirmed by probing with anti-α-tubulin IgG. (C) Confocal laser scanning microscopy of Kc cells transfected with EGFP-tagged proteins (green). Nuclei were stained with PI (red). Scale bars, 10 μm.

In order to examine whether NLS-1 and NLS-2 alone are sufficient for directing nuclear import, we created fusion proteins containing the NLS motifs (Fig. 4A). The sequences coding for NLS-1 and NLS-2 were inserted in frame into the EGFP expression vector. These fusion proteins were expressed in cells at similar levels, and degradation products of the fusion proteins were rarely observed (Fig. 4B). In transiently transfected Kc cells, both EGFP-NLS-1 and EGFP-NLS-2 were detected diffusely in both the cytoplasm and the nucleus (Fig. 4C). These data suggest that both NLS-1 and NLS-2 are necessary for the nuclear localization of dMLFA protein. These NLS motifs may function cooperatively for efficient nuclear localization. They could be also regarded as a bipartite NLS. Classical bipartite NLSs are exemplified by the nucleoplasmin NLS, which is composed of two peptide regions containing basic residues that are separated by a spacer of ten residues (KRPAATKKAGQAKKKK) (Hicks and Raikhel, 1995; Nigg, 1997). The NLS motifs within the dMLFA protein might be a single bipartite NLS that contains a long spacer region.

It has been reported recently that mammalian MLF1 protein contains both NLS and nuclear export signal (NES), and that MLF1 translocates between the nucleus and cytoplasm (Winteringham et al., 2006; Yoneda-Kato and Kato, 2008). In Drosophila, dMLF interacts with Su(fu) and consequently translocates Su(fu) from cytoplasm to nucleus (Martin-Lannerée et al., 2006). Therefore the shuttling activity of dMLF between cytoplasm and nucleus may play an important role in the biological function of dMLF. To clarify the NLS motifs in dMLF is a first step toward understanding the biological activity of dMLF. In this study, we demonstrated that dMLF contains functional NLS motifs. In addition, dMLFA protein seems to contain a putative NES motif in the N-terminal region (⁷LMGDFDDDLGL¹⁷), and the dMLFA, dMLFB and dMLFD isoforms also have the corresponding NES sequence. Although these NLS and NES sequences are not conserved between MLF1 and dMLF (hMLF1 NES sequence, ⁸⁹LERNFGQLSV⁹⁸; hMLF1 NLS sequences, ¹⁶⁸KKSKNKK¹⁷⁴ and ²³²KRREK²³⁶) (Yoneda-Kato and Kato, 2008), it is likely that dMLF also shuttles between cytoplasm and nucleus depending on other protein factors or physiological conditions of cells, as is the case with MLF1. Alternatively, each dMLF isoform might function individually and differentially in different cell types. Our study is the first to demonstrate the presence of NLS motifs within dMLFA isoform, and suggests that the NLS motifs might function cooperatively for efficient nuclear localization.

Acknowledgements

This study was supported by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan. We thank Dr. S. Cotterill of St Georges Hospital Medical School, London, for critical reading of the manuscript.