Identification of two novel nuclear import sequences on the 5-lipoxygenase protein.

The nuclear import of 5-lipoxygenase modulates its capacity to produce leukotrienes from arachidonic acid. However, the molecular determinants of its nuclear import are unknown. Recently, we used structural and functional criteria to identify a novel import sequence at Arg(518) on human 5-lipoxygenase (Jones, S. M., Luo, M., Healy, A. M., Peters-Golden, M., and Brock, T. G. (2002) J. Biol. Chem. 277, 38550-38556). However, this analysis also indicated that other import sequences must exist. Here, we identify two additional sites, at Arg(112) and Lys(158), as nuclear import sequences. Both sites were found to be common to 5-lipoxygenases from different species but not found on other lipoxygenases. Both sites also appeared to be a part of structures that were predominantly random loops. Peptide sequences at these sites were sufficient to direct nuclear import of green fluorescent protein. Mutation of basic residues in these sites impaired nuclear import and combinations of mutations at different sites were additive in effect. Mutations in all three sites were required to disable nuclear accumulation of 5-lipoxygenase in all cells. Significantly, mutation in these sites did not inhibit catalytic function. Taken together, these results indicate that nuclear import of 5-lipoxygenase may reflect the combined functional effects of three discrete import sequences. Mutation of individual sites can, by itself, impair nuclear import, which in turn could impact arachidonic acid metabolism.

Leukotrienes (LTs) 1 are lipid mediators derived from arachidonic acid. They are synthesized primarily by leukocytes and orchestrate a variety of physiological responses in both host defense and inflammatory disease states (reviewed in Ref. 1). The enzyme 5-lipoxygenase (5-LO) catalyzes the rate-limiting first two steps of LT synthesis. Therefore, the regulation of 5-LO action and how it might be modulated in disease have been a focus of interest.
The molecular components that regulate nuclear import of 5-LO remain to be fully elucidated. Commonly, a nuclear import sequence (NIS), rich in the basic amino acids Arg and Lys, mediates nuclear import of proteins, and three such basic regions (BR) have been identified (13). Truncation of 5-LO suggested the presence of an NIS in the amino-terminal region of 5-LO (13). However, limited mutagenesis of a BR at Lys 72 did not prevent nuclear import, suggesting that a non-conventional NIS may exist in the amino terminus of 5-LO. Another site, at Arg 651 , resembles a bipartite NIS. Mutational analysis of this region demonstrated that most basic residues could be replaced without affecting nuclear import (13)(14)(15), unless Arg 651 was replaced (14,15). However, replacement of Arg 651 also caused loss of catalytic activity (15), suggesting that mutagenesis caused protein misfolding, which can also impair import (13). Consistent with this, analysis of 5-LO secondary structure indicated that Arg 651 serves a critical structural role, through its association with Asp 473 (16).
Recently, we developed novel structural and functional criteria to identify functional NIS on 5-LO (16). First, we sought basic residues that were common to 5-LO from different species but not shared by other LO, since nuclear import has been observed in 5-LO from all species but not in 15-LO and 12-LO. Second, we sought BRs having a predominantly random coil/ loop secondary structure, which appears to be necessary for binding to importin-␣ proteins (17)(18)(19). Finally, mutations that altered nuclear import should not also inactivate the enzyme, since failed import may result from mutation-induced changes in protein structure (13); loss of activity, then, would be used as an indirect indication of such a false positive result. Application of these rigorous criteria to 5-LO revealed a novel site at Arg 518 , designated as BR 518 (16). This BR alone was sufficient to drive nuclear import, and replacement of basic residues impaired import without inactivating the enzyme, indicating that BR 518 is a functional NIS. Interestingly, however, mutations in BR 518 could not totally abolish nuclear import in all cells, suggesting that additional NIS(s) must exist on 5-LO.
This study applies the same structural and functional criteria to search for the unidentified NIS(s) on 5-LO. Our results support the conclusion that BR 68 , the only basic region in the ␤-barrel region of 5-LO, is unlikely to be a functional NIS. However, a novel site at Arg 112 , which links the ␤-barrel region to the catalytic domain, meets these criteria and appears to act as an NIS. However, mutation of basic residues in both BR 518 and BR 112 did not eliminate nuclear import of 5-LO. A third site at Lys 158 , which also meets structural and functional criteria, was found to be, by itself, sufficient for nuclear import. Mutation of all three sites eliminated nuclear accumulation of 5-LO without loss of function, indicating that 5-LO contains three functional NISs.

EXPERIMENTAL PROCEDURES
Sequence and Structural Analysis-Amino acid sequences were obtained from Swiss-Prot from the ExPaSy (Expert Protein Analysis System) proteomics server of the Swiss Institute of Bioinformatics. Primary accession numbers for proteins are: 5-LOs, human P09917, mouse P48999, rat P12527, hamster P51399; 15-LOs, rabbit P12630, and human P16050; human platelet-type 12-LO P18054; Clostridium perfringens ␣-toxin P15310. Alignment of protein sequences was performed using ClustalW (20). Structural analysis utilized the resolved structures of rabbit 15-LO (PDB: 1lox) and C. perfringens ␣-toxin (PDB: 1qmd), as well as published theoretical models of the 5-LO ␤-barrel region (21) and the 5-LO catalytic region (22).
Plasmids and Mutagenesis-To construct a fusion peptide joining BR 112 or BR 158 to green fluorescent protein (GFP), complementary oligonucleotides encoding the basic regions (indicated below) were annealed and ligated to the BamHI and HindIII sites of pEGFP-C1. BR 112 peptide was Leu 111 -Asp 121 (LRDGRAKLARD); BR 158 peptide was Asp 156 -Asp 166 (DAKCHKDLPRD).
Specific amino acids within the putative 5-LO NISs were substituted in the pEGFP-C1/5-LO template (14) using the QuikChange site-directed mutagenesis kit (Stratagene). Briefly, two complementary primers (125 ng each) containing the desired mutation and 20 ng of template in 1ϫ reaction buffer were denatured at 95°C for 30 s and annealed at 55°C for 30s, and DNA synthesis was carried out by Pfu polymerase at 68°C for 14 min. This cycle was repeated 12-18 times, depending on the number of bases substituted, according to the manufacturer's directions. The methylated template was removed by incubation with 10 units of DpnI at 37°C for 1 h. The mutation BR 518 was R518Q/R520Q/ K521Q/K527Q/K530Q; mutation BR 112 was R115Q/K117Q/R120Q; mutation BR 158 was K158N/H160Q/K161N. All substitutions and constructs were verified by DNA sequence analysis (DNA Sequencing Core, University of Michigan). Oligonucleotides (sequences available upon request) were synthesized and PAGE-purified by Integrated DNA Technologies Inc. (Coralville, IA).
Cell Culture, Transfection, and Imaging-NIH 3T3 cells were obtained from American Type Culture Collection (Manassas, VA) and grown under 5% CO 2 in Dulbecco's Modified Eagle's Medium (Invitrogen) supplemented with 10% calf serum, 100 units/ml penicillin, and 100 units/ml streptomycin. Cells were transfected using Polyfect (Qiagen, Inc.) transfection reagents according to the manufacturer's specifications. Transient transfectants were evaluated microscopically, live, or after fixation with 4% paraformaldehyde, 16 -20 h after transfection. Comparable results were obtained when cells were examined as early as 9 h after transfection.
Immunoblotting-As described previously (14), cells were disrupted by sonication on ice, and protein concentrations were determined by a modified Coomassie Blue dye binding assay (Pierce). Samples containing 10 g of protein were separated by SDS-polyacrylamide gel electrophoresis under reducing conditions and transferred to nitrocellulose. Membranes were probed with a rabbit polyclonal antibody raised against purified human leukocyte 5-LO (a generous gift from Dr. J. Evans, Merck Research Laboratories, Rahway, NJ) (23) or with rabbit polyclonal anti-GFP (Santa Cruz Biotechnology, Inc.; titer 1:500) followed by peroxidase-conjugated secondary antibody and enhanced chemiluminescence detection (Amersham Biosciences).
Cell Stimulation and Analysis of Leukotriene Synthesis-To stimulate 5-LO activity, cells transfected with various 5-LO constructs were washed, then incubated for 30 min at 37°C in serum-free medium containing 10 M calcium ionophore A23187 and 10 M arachidonic acid. Immunoreactive LTB 4 in conditioned media was quantitated by enzyme immunoassay (Cayman Chemical, Ann Arbor, MI) according to the supplier's instructions. For each sample, the measured value was taken as the average of duplicate determinations. Media from nontransfected, mock-transfected, or GFP-transfected cells did not contain detectable LTB 4 . LTB 4 determinations were standardized for transfection efficiency: cells were washed following stimulation, harvested by scraping, sonicated on ice, assayed by immunoblot analysis (using 10 g of protein/sample) for expression using anti-GFP, with expression quantitated by densitometry. LTB 4 synthesis, adjusted for construct expression, was evaluated for all constructs in at least two independent experiments. The detection limit for LTB 4 was 4 pg/ml; cross-reactivity for AA, 5-HETE, LTC 4 , LTD 4 and LTE 4 was Ͻ0.01%.
Alternatively, activity of constructs was evaluated by cell-free assay: 5 ϫ 10 6 COS-1 cells were transfected with 5 g of plasmid DNA using Polyfect; 40-h post-transfection, cells were harvested, washed with phosphate-buffered saline once, and sonicated for 90 s in 10-s bursts on ice in cell lysis buffer containing 50 mM Tris-HCl (pH 7.5), 0.1 mM EDTA, 1 mM dithiothreitol, and protease inhibitor mixture (Complete TM EDTA-free, Roche Molecular Biochemicals). After sonication, lysates were centrifuged (5000 rpm, 8 min, 4°C) to remove cell debris and protein expression for each construct was confirmed first by Western blot using anti-GFP. The 5-LO activity of cell lysates (200 g of total protein) was determined in 0.25 ml reaction mixtures containing 50 mM Tris-HCl (pH 7.5), 0.6 mM CaCl 2 , 0.1 mM EDTA, 0.1 mM ATP, 12 g/ml phosphatidylcholine (Avanti Polar Lipids, Alabaster, AL), 20 M AA (Cayman Chemicals, Ann Arbor, MI), including ϳ100,000 dpm [ 3 H]AA (PerkinElmer Life Sciences) and 10 M 13(S)-hydroperoxy-9-cis-11trans-octadecadienoic acid (Cayman Chemicals). After a 30-min incubation at room temperature, the reaction was stopped by adding 1 ml of ether/methanol/1 M citric acid (30:4:1, v/v/v). After vortexing thoroughly, the mixture was centrifuged at 5000 rpm for 5 min. The upper phase was removed, evaporated under nitrogen, and stored at Ϫ70°C. Lipid residues were dissolved in 250 l of 50% acetonitrile/trifluoroacetic acid (1000:1, v/v) and 50% water/trifluoroacetic acid (1000:1, v/v), analyzed by reverse-phase high performance liquid chromatography (HPLC) on a 5-m Bondapak C 18 column (30 ϫ 0.4 cm; Waters Associates, Milford, MA) using a mobile phase of acetonitrile/trifluoroacetic acid at a flow rate of 2 ml/min. 5-LO metabolites were eluted during a series of linear gradient increases of acetonitrile from initial conditions of 50:50 (v/v) to 73:27 (v/v) at 7 min, then to 85:15 (v/v) at 9 min, and finally to 100:0 (v/v) at 15 min. Radioactivity in 1-ml eluted fractions was quantitated by on-line radiodetection. There were no LTB 4 /LTB 4 isomers detected in cell lysates; 5-HPETE and 5-HETE co-eluted as a single peak, clearly separated from un-metabolized AA. The 5-LO specific activity of different mutants was calculated and compared based on the ratio of conversion of radiolabeled AA to 5-HPETE/5-HETE.
Quantitation of Subcellular Distribution-As an initial approach to quantitation, slides were fixed 16 h after transfection, and 100 positive cells were scored as to whether nuclear fluorescence was greater than, equal to or less than cytosolic fluorescence. Care was taken to avoid damaged, dead or autofluorescent cells. Results from at least three independent transfections per construct were used for statistical analysis. As a second approach, 100 individual cells per construct were scored for cytosolic and nuclear fluorescence intensity: using Adobe Photoshop 5.5, grayscale digital images were adjusted to include the full black-to-white range, and representative gray values, from 0 (white) to 100 (black), were obtained for the cytoplasm and nucleoplasm. Cytoplasmic and nuclear values for each cell were summed to give total cellular fluorescence, and the percent fluorescence values for the nuclear compartment were calculated.
Statistical Analysis-Statistical significance was evaluated by oneway analysis of variance, using p Ͻ 0.05 as indicative of statistical significance. Pairs of group means were analyzed using the Tukey-Kramer post-test.

RESULTS
Reassessment of BR 68 as a Functional NIS-As outlined above, cells expressing GFP⅐5-LO with mutated BR 518 displayed either no import or significant import, indicating the existence of at least one other functional NIS on 5-LO. Since the BR 518 NIS was identified using structural criteria noted above, these criteria were applied to other candidate sites. Previous work indicated that an NIS in the amino-terminal region of 5-LO, potentially at BR 68 , might function as an NIS (13). This site was a good candidate because its primary structure, RXXKRK, fulfills the criteria of a monopartite NIS, being a cluster of 4 of 6 basic residues. Using the structural and functional criteria, BR 68 was evaluated further. Comparison with the primary sequences of other lipoxygenases, however, indicated that this BR was not unique to 5LO (Fig. 1A). Thus, if it were a functional NIS on 5-LO, it might also be expected to direct the import of 15-LO and, perhaps, 12-LO. Regarding the secondary structure of BR 68 , no resolved structure for 5-LO is available. However, predicted structures for the ␤-barrel domain have been published using C. perfringens ␣-toxin (21) or 15-LO (24) as templates, and the structure of the 15-LO ␤-barrel domain has been published (25). The majority of the amino acids in BR 68 were found to be involved in the fifth ␤-sheet, in the predicted structure of 5-LO patterned after ␣-toxin (Fig.  1B) and in the resolved structure of 15-LO (Fig. 1C), as well as in the 5-LO structure patterned after 15-LO (Ref. 24 and not shown here). This suggests that this region serves a critical structural role in the amino-terminal ␤-barrel and is not available for binding importin. These results indicate that BR 68 is unlikely to be a functional NIS.
Evaluation of BR 112 as a Functional NIS-Alignment of LO primary sequences revealed a novel basic region, beginning at R112 on human 5-LO, which was conserved across 5-LOs and not found in 12-or 15-LOs ( Fig. 2A). Correct alignment was suggested by high levels of amino acid similarity on both sides of the region as well as alignment of the ␣-helix in the catalytic domain. This region, designated BR 112 , contained 4 basic amino acids over a stretch of 9 residues. The region was located on a random coil between the ␤-barrel and catalytic domains of 5-LO (Fig. 2B). The presence of a conserved glycine, which can serve as a "helix breaker," also indicated that this region would retain a random coil structure. Since the region was conserved across different 5-LOs, not found on other LOs and was on a coiled structural element, it met our primary and secondary structural criteria for a good candidate NIS.
To test whether BR 112 was sufficient to cause nuclear import, oligonucleotides were synthesized and inserted into the GFP vector to produce GFP with the peptide LRDGRAKLARD fused to the carboxyl terminus. As has been frequently described (e.g. (13)), GFP alone distributed evenly between nuclear and cytoplasmic compartments in transfected cells (Fig. 3), because it is small enough to diffuse freely through the nuclear pore. How-ever, the GFP⅐BR 112 fusion protein showed distinct nuclear accumulation (Fig. 3), indicating that this peptide alone is sufficient to drive nuclear import against a diffusion gradient.
To determine whether BR 112 was necessary for nuclear import, the effect of basic residue replacement in BR 112 , in the context of GFP⅐5-LO, was evaluated. Three residues were replaced by site-directed mutagenesis: R115Q/K117Q/R120Q. As described previously (13,14), the wild type (WT) GFP⅐5-LO fusion protein showed strong nuclear accumulation in most cells (Fig. 4, A and B). Cells expressing GFP⅐5-LO with mutation of BR 112 included two distinct phenotypes: some cells had little or no nuclear fluorescence, while others showed clear nuclear accumulation of the expressed protein (Fig. 4, C and  D). This result was first quantitated by scoring individual cells as having nuclear fluorescence greater than, equal to or less than the cytosolic fluorescence. Representative images and numbers for one experiment are given in Fig. 5. While the majority (65%) of cells expressing WT GFP⅐5-LO had nuclear accumulation, a significant number (31%) had a balanced distribution. A balanced distribution of 5-LO, associated with nuclear envelope breakdown during mitosis, has been described (7) and quantitated (15). Mutation at the BR 112 site reduced the number of cells with nuclear import and increased those with cytosolic fluorescence. These results indicated that mutation of BR 112 in the context of GFP⅐5-LO impaired nuclear import of the fusion protein in some 3T3 cells.
While the above results demonstrated that mutation of the BR 112 site affected the subcellular distribution of 5-LO, they do not clearly define the nature of that effect. In particular, they did not clearly indicate whether the mutation simply reduced the efficiency of nuclear import, or whether the mutation resulted in distinct subpopulations of cells. To address this question, nuclear and cytoplasmic fluorescence levels in individual cells were quantitatively analyzed as described under "Experimental Procedures." By this analysis, there were (at least) two subpopulations exhibiting nuclear import of WT GFP⅐5-LO, when expressed in 3T3 cells (Fig. 6A). A major peak, consistently found in multiple transfections, consisted of cells with 60 -70% of total fluorescence in the nucleus (peak N 1 ). A shoulder, associated with 70 -90% nuclear fluorescence, also was consistently observed (peak N 2 ). Mutation of BR 112 , as shown in Fig. 6B, reduced the number of strongly importing cells (peak N 2 ) and reduced the rate of nuclear import, as indicated by the shift of peak N 1 to the left (i.e. to 55%). More significantly, this mutation produced a new population with only 30 -40% nuclear fluorescence, designated peak C 1 . It should be noted that these cells, shown in Figs. 4 and 5, had little or no nuclear import; the relatively high value of 35% nuclear fluorescence reflects the conservative scoring of this quantitative method. In general terms, these results indicated that mutation of BR 112 reduced the capacity for the strong nuclear import that produced the N 2 peak. The resulting protein was still capable of pronounced nuclear import in some cells, as evidenced by the persistent N 1 peak. However, the resulting protein did not direct nuclear import in those cells comprising the C 1 peak.
The finding that mutation in BR 112 produced discrete import competent and non-importing populations was very similar to results previously found with mutations in BR 518 (16). This suggested the possibility that these sites might overlap in function. To address this possibility, mutations in both sites were performed. The changes in BR 518 were R518Q/R520Q/ K521Q/K527Q/K530Q. Sample subcellular distributions and frequencies, for mutation in BR 518 alone or in BR 112 plus BR 518 , are given in Fig. 7A. Mutation of the BR 518 site alone resulted in ϳ20% of the cells having impaired import, with about half of the cells still having significant nuclear accumulation of GFP⅐5-LO, as reported previously (16). The combined mutations of BR 112 plus BR 518 had an additive effect, with over half of the cells showing a failure to import GFP⅐5-LO.
Statistical analysis of results from multiple transfections showed that, for each of the two mutants, both the reduction of cells with nuclear import and the increase in cells with cytoso- lic fluorescence were statistically significant (Table I). Moreover, mutation of both basic regions produced statistically greater changes in the two distribution groups than did either mutation alone. No statistically significant change in the group showing balanced distribution was found for any mutation. The additive nature of the mutations indicated that these sites represent distinct NISs.
Significantly, all mutants were functionally active, producing LTB 4 when stimulated with the calcium ionophore A23187 in the presence of 10 M arachidonic acid (Table I). The double mutant produced marginally less LTB 4 than WT GFP⅐5-LO. Furthermore, mutated proteins of the appropriate size were expressed in similar amounts as wild type GFP⅐5-LO in transfected 3T3 cells (Fig. 7B). This result was obtained using antibodies to either GFP (Fig. 7B) or 5-LO (data not shown). Thus, the changes in nuclear import were unlikely to result from altered protein expression, conformational folding, or protein degradation.
Evaluation of BR 158 as a Functional NIS-Because mutations of both the BR 518 and BR 112 did not completely impair nuclear import of 5-LO, the protein sequence was further evaluated using the structural and functional criteria described previously. Another novel basic region was identified, beginning at Lys 158 on human 5-LO. Correct alignment was supported by amino acid similarity on both sides of the region as well as alignment of the DLP core (Fig. 8A). This region, BR 158 (KCHKDLPR), contained 3 basic residues, which were conserved across 5-LOs and not found in 12-or 15-LOs. This region was predicted to form a random coil on the catalytic domain of 5-LO (22), although a helix-like turn involving Leu-Thr on 15-LO (replaced by His-Lys on 5-LO) was evident (Fig. 8B). The presence of the conserved proline within the region, which can serve as helix breaker, also indicates that this region would retain a random coil structure. Since the region was conserved across different 5-LOs, not found on other LOs and was on a largely coiled structural element, it met our primary and secondary structural criteria for a good candidate NIS.
A vector was constructed to express the GFP⅐BR 158 fusion protein, with DAKCHKDLPRD representing BR 158 . This fusion protein showed nuclear accumulation (Fig. 8, C and D). Quantitative analysis of nuclear/cytosolic fluorescence ratios for 100 cells revealed that nuclear accumulation of the GFP⅐BR 158 fusion protein was greater than that for the GFP⅐BR 112 fusion protein (data not shown). Thus, the BR 158 peptide is sufficient to drive nuclear import.
To determine whether the BR 158 region was necessary for nuclear import, site-directed replacement of basic residues on GFP  construct was expressed in 3T3 cells, the majority of cells showed nuclear accumulation, with only 10% of the cells clearly indicating a failure to import (Fig. 9). However, when this mutation was combined with mutations at the other two NIS sites, no cells showed nuclear accumulation (Fig. 9). Statistical analysis of results from multiple transfections showed that mutation of BR 158 produced a small but statistically significant increase in cells with impaired nuclear import (Table II). As described above in Table I, the combination of mutBR 112 ϩ mutBR 518 significantly but incompletely decreased nuclear import. When all three BRs were altered, no cells showed nuclear accumulation of GFP⅐5-LO. Significantly, all mutants were functionally active, producing LTB 4 when stimulated with the calcium ionophore A23187 in the presence of 10 M arachidonic acid (Table II). The double and triple mutants produced marginally less LTB 4 than the single mutants or WT GFP⅐5-LO. Further analysis of mutants by cell-free assay confirmed that, although the multiple amino acid substitutions did indeed reduce activity relative to wild type GFP⅐5-LO, even the mutation of all three NISs did not abolish activity (Table III). These results indicated that all three basic regions, BR 518 , BR 112 , and BR 158 , were necessary for nuclear accumulation of 5-LO in all cells within a population. DISCUSSION Previously, we used novel structural and functional criteria to identify an NIS, designated BR 518 , on 5-LO, but predicted that at least one other NIS must also exist (16). In the present study, we continued to search for potential NISs on 5-LO using the same criteria. This work revealed two novel basic regions, BR 112 and BR 158 . Subsequent analysis showed that these sites were both sufficient and necessary for normal import. The body of work presented in this and the previous study confirms that three NISs exist on 5-LO and that all three are functional in determining the subcellular localization of 5-LO in 3T3 cells. Moreover, these studies indicate that the different NISs act independently from one another and that they can be activated and inactivated. Finally, these results demonstrate how muta-tions at different NISs will differ in their impact on the subcellular distribution of 5-LO.
Our mutagenesis data showed that the first basic region, BR 112 , serves a relatively strong import role, since mutation of the basic residues in this region impaired 5-LO nuclear import to at least the same extent as the mutation of BR 518 . Although mutation of BR 112 significantly impaired 5-LO nuclear import, enzyme immunoassay showed that the catalytic activity of 5-LO was not compromised. Thus, the mutation specifically impaired nuclear import without changing the general enzyme secondary structure. This further substantiates that BR 112 is a functional NIS. Mutations of both BR 112 and BR 518 had additive effects in reducing nuclear import (Table I), indicating that these sites act independently from one another.
The fact that mutations on both BR 112 and BR 518 could not totally eliminate nuclear import implied the existence of an additional import sequence. The third site, BR 158 , was found to be unique to 5-LO, structurally appropriate for binding importin, and sufficient to import GFP. Because mutation of BR 158 alone only slightly impaired nuclear import, we speculate that this region may act as a weak import sequence that has low affinity to the receptor protein importin . Alternatively, BR 158 alone may not function as an independent import sequence; it may coordinate other NISs to mediate 5-LO nuclear import. Supporting this idea, when BR 112 , BR 158 and BR 518 were all mutated at once, nuclear import of 5-LO was totally eliminated. Combined with the finding that the multiple mutant was still active, these studies strongly indicate that BR 112 , BR 158 , and BR 518 are functional NISs and that multiple NISs orchestrate 5-LO nuclear import.
Previously, it was reported that a peptide containing the first 80 amino acids of 5-LO could drive import, leading to the suggestion that this region contains an NIS (13). The role of secondary structure in determining import capacity, as stressed in this study, may help explain this result. The region BR 68 , as shown in Fig. 1   81-111 will also remove the last three ␤-sheets of the barrel. In the LOs, sheet 5 is positioned between sheets 2 and 8, across from sheets 6 and 7; loss of sheets 6 -8 might allow the fifth sheet, and BR 68 , to reform as a random coil. In this conformation, BR 68 would be able to bind importin and, misleadingly, act as an NIS. As we have found for 5-LO, an increasing number of proteins have been described as having multiple NISs. These include BRCA1 (26), Epstein-Barr virus DNase (27), herpes simplex virus products ICP22 (28), ICP27 (29), and XPG nuclease (30). The importance of having multiple NISs is unclear. In some cases, the individual NISs are weak, and the actions of multiple NISs can be additive or synergistic, as appears to be the case for 5-LO (this study) and XPG nuclease (30). Alternatively, the isoforms of importin-␣ are differentially expressed in different cell types and may bind each NIS with varied specificity (31). Thus, the NIS that is actually functional in a given cell type may depend on the isoform(s) of importin-␣ that is present.
Finally, each NIS may be regulated independently from the others. The observation of two distinct populations, one with import and one without, in cells expressing GFP⅐5-LO with either mutBR 112 (Fig. 4) or mutBR 518 (16) suggests that the remaining NISs may be subject to regulation. Protein phosphorylation in the vicinity of NISs has been repeatedly shown to play a role in regulating nuclear import (e.g. Refs. 32 and 33). There is evidence that 5-LO can be modulated by different kinases, such as protein kinases A (34) and C (35,36), protein tyrosine kinases (37), and by mitogen-activated protein kinase kinase (38,39), but these studies have not shown direct phosphorylation of 5-LO. More recently, two groups have been able to show direct phosphorylation of 5-LO by a tyrosine kinase (40) and by MAPKAP kinase 2 (41). It is not known which, if any, of these phosphorylation events regulate the three NISs of 5-LO.
In certain leukocytes, such as neutrophils and eosinophils in circulating blood, 5-LO is found exclusively in the cytoplasm (2)(3)(4)(5)12). It seems reasonable that, in cells under these conditions, no NIS is activated. As noted above, nuclear import of 5-LO can be induced by different cues, including adherence, recruitment and cytokines. Each of these cues might activate distinct kinase pathways and, in turn, activate specific NISs on 5-LO. This suggests that the purpose for having multiple NISs, combined with multiple activation pathways, would be to ensure the import of 5-LO in response to a range of conditions. Activation of a single NIS might be sufficient for significant accumulation of 5-LO in the nucleus, whereas activation of multiple NISs might drive even greater accumulation.
Elucidation of the functional NISs in 5-LO represents a first step toward our understanding of 5-LO nuclear import. Future work regarding how these nuclear targeting sequences are recognized and modulated in normal and diseased cell types may reveal the mechanisms as well as functional consequences of nuclear import of 5-LO.