Portal de Periódicos da CAPES

cDNA Cloning, Expression, Mutagenesis, Intracellular Localization, and Gene Chromosomal Assignment of Mouse 5-Lipoxygenase

1995; Elsevier BV; Volume: 270; Issue: 30 Linguagem: Inglês

10.1074/jbc.270.30.17993

1083-351X

Xinsheng Chen, Todd A. Naumann, Usha Kurre, Nancy A. Jenkins, Neal G. Copeland, Colin Funk,

Estrogen and related hormone effects

5-Lipoxygenase of mouse macrophages and bone marrow-derived mast cells (BMMC) was investigated. Indirect immunocytofluorescence combined with confocal microscopy provided evidence for distinct intracellular expression patterns and trafficking of 5-lipoxygenase upon cellular activation. In resting BMMC, 5-lipoxygenase was found within the nucleus co-localizing with the nuclear stain Yo-Pro-1. When BMMC were IgE/antigen-activated the 5-lipoxygenase immunofluorescence pattern was changed from nuclear to perinuclear. The absence of divalent cations in the incubation medium, or calcium ionophore A23187 challenge, altered the predominantly nuclear expression pattern to new sites both cytosolic and intranuclear. The cDNA for murine macrophage 5-lipoxygenase was cloned by the polymerase chain reaction and would predict a 674 amino acid protein. Using control cells obtained from 5-lipoxygenase-deficient mice it was determined that a single isoform accounts for both soluble and membrane-bound and nuclear and cytosolic-localized enzyme in macrophages and BMMC. A mutation at amino acid 672 (Val ⟶ Met) introduced serendipitously during the cloning process was found to completely abolish 5-lipoxygenase enzyme activity when the enzyme was expressed in human embryonic kidney 293 cells. This subtle change is proposed to affect the ability of the COOH-terminal isoleucine to coordinate the essential non-heme iron atom. In macrophages and BMMC obtained from 5-lipoxygenase-deficient mice, compensatory changes in expression of genes involved in the biosynthesis of leukotriene B4 were investigated. 5-Lipoxygenase-activating protein expression was reduced by 50%, while leukotriene A4 hydrolase expression was unaltered. The 5-lipoxygenase gene was mapped to the central region of mouse chromosome 6 in a region that shares homology with human chromosome 10 by interspecific backcross analysis. These studies provide a global picture of the murine 5-lipoxygenase system and raise questions about the role of 5-lipoxygenase and leukotrienes within the nucleus. 5-Lipoxygenase of mouse macrophages and bone marrow-derived mast cells (BMMC) was investigated. Indirect immunocytofluorescence combined with confocal microscopy provided evidence for distinct intracellular expression patterns and trafficking of 5-lipoxygenase upon cellular activation. In resting BMMC, 5-lipoxygenase was found within the nucleus co-localizing with the nuclear stain Yo-Pro-1. When BMMC were IgE/antigen-activated the 5-lipoxygenase immunofluorescence pattern was changed from nuclear to perinuclear. The absence of divalent cations in the incubation medium, or calcium ionophore A23187 challenge, altered the predominantly nuclear expression pattern to new sites both cytosolic and intranuclear. The cDNA for murine macrophage 5-lipoxygenase was cloned by the polymerase chain reaction and would predict a 674 amino acid protein. Using control cells obtained from 5-lipoxygenase-deficient mice it was determined that a single isoform accounts for both soluble and membrane-bound and nuclear and cytosolic-localized enzyme in macrophages and BMMC. A mutation at amino acid 672 (Val ⟶ Met) introduced serendipitously during the cloning process was found to completely abolish 5-lipoxygenase enzyme activity when the enzyme was expressed in human embryonic kidney 293 cells. This subtle change is proposed to affect the ability of the COOH-terminal isoleucine to coordinate the essential non-heme iron atom. In macrophages and BMMC obtained from 5-lipoxygenase-deficient mice, compensatory changes in expression of genes involved in the biosynthesis of leukotriene B4 were investigated. 5-Lipoxygenase-activating protein expression was reduced by 50%, while leukotriene A4 hydrolase expression was unaltered. The 5-lipoxygenase gene was mapped to the central region of mouse chromosome 6 in a region that shares homology with human chromosome 10 by interspecific backcross analysis. These studies provide a global picture of the murine 5-lipoxygenase system and raise questions about the role of 5-lipoxygenase and leukotrienes within the nucleus. The enzyme 5-lipoxygenase (arachidonate:oxygen 5-oxidoreductase, EC 1.13.11.34) catalyzes the formation of 5-hydroperoxy-6,8,11,14-eicosatetraenoic acid (5-HPETE)1 1The abbreviations used are: 5-H(P)ETE5-hydro(pero)xy-eicosatetraenoic acidLTleukotrieneFLAP5-lipoxygenase-activating proteinBMMCbone marrow-derived mast cellPCRpolymerase chain reactionHEKhuman embryonic kidneyRACErapid amplification of cDNA endsDNP-BSAdinitrophenyl bovine serum albumin13-H(P)ODE13-hydro(pero)xy-octadecadienoic acidRP-HPLCreversed phase-high performance liquid chromatographykbkilobase(s). and its subsequent conversion to leukotriene (LT)A4 (5,6-oxido-7,9,11,14-eicosatetraenoic acid). LTA4 is a pivotal intermediate in the biosynthesis of inflammatory and anaphylactic mediators which include leukotriene B4 (5S,12R)-dihydroxy-6,14-cis-8,10-trans-eicosatetraenoic acid and the peptidyl leukotrienes (LTC4, LTD4, and LTE4; see Refs. 1, 2 for reviews). In human neutrophils, 5-lipoxygenase undergoes a Ca2+-dependent translocation from the cytosol to a membrane site which appears to be the nuclear envelope(3Rouzer C.A. Kargman S. J. Biol. Chem. 1988; 263: 10980-10988Google Scholar, 4Woods J.W. Evans J.F. Ethier D. Scott S. Vickers P.J. Hearn L. Heilbein J. Charleson S. Singer I.I. J. Exp. Med. 1993; 178: 1935-1946Google Scholar). 5-Lipoxygenase activating protein (FLAP), an 18-kDa membrane protein found in the nuclear envelope, acts apparently as an arachidonate-binding protein to facilitate the concerted formation of LTA4(5Dixon R.A.F. Diehl R.E. Opas E. Rands E. Vickers P.J. Evans J.F. Gillard J.W. Miller D.K. Nature. 1990; 343: 282-284Google Scholar, 6Abramovitz M. Wong E. Cox M.E. Richardson C.D. Li C. Vickers P.J. Eur. J. Biochem. 1993; 215: 105-111Google Scholar). 5-hydro(pero)xy-eicosatetraenoic acid leukotriene 5-lipoxygenase-activating protein bone marrow-derived mast cell polymerase chain reaction human embryonic kidney rapid amplification of cDNA ends dinitrophenyl bovine serum albumin 13-hydro(pero)xy-octadecadienoic acid reversed phase-high performance liquid chromatography kilobase(s). In alveolar macrophages there is evidence for the existence of two 5-lipoxygenase “pools,” cytosolic and membrane-bound forms(7Coffey M. Peters-Golden M. Fantone J.C. Sporn P.H.S. J. Biol. Chem. 1992; 267: 570-576Google Scholar). Recent data by the same investigators has established nuclear soluble and nuclear bound 5-lipoxygenase expression patterns in rat basophilic leukemia cells(8Brock T.G. Paine R. Peters-Golden M. J. Biol. Chem. 1994; 269: 22059-22066Google Scholar). Lepley and Fitzpatrick (9Lepley R.A. Fitzpatrick F.A. J. Biol. Chem. 1994; 269: 24163-24168Google Scholar) have obtained in vitro data that 5-lipoxygenase can bind to cytoskeletal proteins and the signaling protein Grb2 via an SH3-binding domain interaction. Thus, in addition to the carrier-mediated export of leukotrienes(10Lam B.K. Owen Jr., W.F. Austen K.F. Soberman R.J. J. Biol. Chem. 1989; 264: 12885-12889Google Scholar, 11Lam B.K. Gagnon L. Austen K.F. Soberman R.J. J. Biol. Chem. 1990; 265: 13438-13441Google Scholar), the concept is emerging that 5-lipoxygenase may have novel intracellular functions, possibly independent of leukotriene biosynthesis, and its intracellular location may be dictated by specific protein-protein interactions. cDNAs encoding the human (12Matsumoto T. Funk C.D. Rdmark O. Hg J.-O. Jrnvall H. Samuelsson B. Proc. Natl. Acad. Sci. U. S. A. 1988; 85 (and correction p. 3406): 26-30Google Scholar, 13Dixon R.A.F. Jones R.E. Diehl R.E. Bennett C.D. Kargman S. Rouzer C.A. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 416-420Google Scholar) and rat (14Balcarek J.M. Theisen T.W. Cook M.N. Varrichio A. Hwang S.-M. Strohsacker M.W. Crooke S.T. J. Biol. Chem. 1988; 263: 13937-13941Google Scholar) 5-lipoxygenases have been isolated, and the human 5-lipoxygenase genomic structure (15Funk C.D. Hoshiko S. Matsumoto T. Rdmark O. Samuelsson B. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 2587-2591Google Scholar) has been elucidated. Recently, in our efforts to better understand the biology of 5-lipoxygenase and leukotrienes we created 5-lipoxygenase-deficient mice by gene targeting(16Funk C.D. Kurre U. Griffis G. Ann. N. Y. Acad. Sci. 1994; 714: 253-258Google Scholar, 17Chen X.-S. Sheller J.R. Johnson E. Funk C.D. Nature. 1994; 372: 179-181Google Scholar). We realized the importance of ascertaining more clearly the existence, or not, of 5-lipoxygenase isoforms and their relationship to distinct intracellular pools of 5-lipoxygenase in resting and activated cells. Our studies have focused primarily on bone marrow-derived mast cells (BMMC) and macrophages. Here we show evidence for different 5-lipoxygenase intracellular locations depending on the state of cell activation in BMMC. Moreover, studies with 5-lipoxygenase-deficient mice show that the different 5-lipoxygenase pools in alveolar macrophages derive from the same gene product. Additionally, we demonstrate the importance of the amino acid residue two positions upstream of the carboxyl terminus for 5-lipoxygenase activity, the chromosomal location of the murine 5-lipoxygenase gene, and studies with macrophages and BMMC of 5-lipoxygenase-deficient mice designed to examine compensatory expression of other proteins key to the formation of leukotrienes (FLAP and LTA4 hydrolase). C57BL/6 × 129 mixed genetic background mice were maintained in the animal barrier facility of Vanderbilt University on a 12 h light/12 h dark cycle with water and food provided ad libitum. The generation of 5-lipoxygenase-deficient mice has been described(17Chen X.-S. Sheller J.R. Johnson E. Funk C.D. Nature. 1994; 372: 179-181Google Scholar). BMMC were prepared from cells flushed from femurs and tibiae of wild-type and 5-lipoxygenase-deficient mice and were cultured in the presence of 50% WEHI-3b conditioned medium, 50% RPMI 1640 (complete medium) for 3-6 weeks(18Razin E. Mencia-Huerta J.-M. Stevens R.L. Lewis R.A. Liu F.-T. Corey E.J. Austen K.F. J. Exp. Med. 1983; 157: 189-201Google Scholar). Cell purity estimated by cell morphology and staining with toluidine blue was 90-95%. Peritoneal macrophages (19Cohn Z.A. Benson B. J. Exp. Med. 1965; 121: 153-170Google Scholar) were obtained by lavage from the peritoneal cavity with 3 ml of Dulbecco's modified Eagle's medium containing 10% fetal calf serum and 5 units/ml heparin. Cells were plated in a humidified 95% air, 5% CO2 atmosphere at 37°C in tissue culture dishes. After adherence for 1 h, the cells were washed three times and used for experiments. Cell purity was estimated to be >97% based on cell morphology and staining with nonspecific esterase. Pulmonary alveolar macrophages were isolated by a published procedure(20Rouzer C.A. Scott W.A. Hamill A.L. Cohn Z.A. J. Exp. Med. 1982; 155: 720-733Google Scholar). Cultured BMMC obtained from wild-type and 5-lipoxygenase-deficient mice were washed three times with modified Tyrode's buffer (contains 0.32 mM Ca2+; (18Razin E. Mencia-Huerta J.-M. Stevens R.L. Lewis R.A. Liu F.-T. Corey E.J. Austen K.F. J. Exp. Med. 1983; 157: 189-201Google Scholar)) at 4°C. One group of cells was sensitized with monoclonal IgE directed against DNP-human serum albumin (Sigma; 100 μg/ml) for 1 h at 37°C, followed by three washes, and subsequent incubation with DNP-BSA (50 ng/ml) for 30 min. Other groups of cells were incubated with calcium ionophore A23187 (0.5 μM), EDTA (2 mM) or no additions for 30 min in modified Tyrode's. The cells were quickly washed two times with complete medium, with or without EDTA, at 4°C. Cells were placed on glass microscope slides using a Shandon cytocentrifuge (550 revolutions/min for 5 min). Cells were fixed with 4% paraformaldehyde in phosphate-buffered saline for 15 min and permeabilized with 0.2% Triton X-100 for 10 min. Cells were incubated with 3% bovine serum albumin for 30 min followed by 5% donkey serum for 30 min to block nonspecific binding. The cells were incubated with a rabbit polyclonal anti-5-lipoxygenase antiserum (1:2500 dilution; see below) for 5 h at room temperature or overnight at 4°C. The slides were washed three times with phosphate-buffered saline and incubated with Cy3-labeled donkey anti-rabbit antibody (Jackson ImmunoResearch Laboratories; 1:4000 dilution). The slides were washed three times with phosphate-buffered saline and incubated with the nuclear stain Yo-Pro-1 (Molecular Probes; 1:5000 dilution) for 15 min. The slides were air-dried and mounted with Aqua-Poly/Mount (Polysciences Inc.). Slides were examined under oil immersion with a Zeiss LSM410 laser scanning confocal microscope using ×40 or ×63 objectives. Excitation/emission settings were 543/650 nm (HeNe laser) for Cy3 and 488/520 nm for Yo-Pro-1 (ArKr laser). Raw data images were processed further using Adobe Photoshop (MacIntosh) and Showcase (Silicon Graphics Indigo2) programs. Cells were sonicated for two bursts of 15 s on ice and centrifuged at 10,000 × g at 4°C. Soluble and pellet fractions were obtained. Protein was quantitated by Bradford assay (Bio-Rad reagent) and prepared for SDS-polyacrylamide gel electrophoresis. Immunoblot analysis was carried out as described in the Fig. 5 legend using previously characterized polyclonal anti-human 5-lipoxygenase, FLAP and leukotriene A4 hydrolase antisera, and purified recombinant human 5-lipoxygenase as standard (generous gifts of Dr. J. Evans, Merck Frosst). Detection was by enhanced chemiluminescence combined with autoradiography using reagents from Amersham Corp. Relative band densities were estimated by densitometric analysis of x-ray films using Image 1.38 software (Wayne Rasburn, NIH, Bethesda, MD). BMMC (107 cells) in 1 ml of modified Tyrode's buffer were incubated in the presence and absence of 0.5 μM A23187, or IgE/antigen for 20 or 30 min at 37°C as described above. No exogenous arachidonic acid was added. Incubations were terminated with 4 volumes of ethanol, extracted with ODS-silica columns, and products were separated by RP-HPLC as described(17Chen X.-S. Sheller J.R. Johnson E. Funk C.D. Nature. 1994; 372: 179-181Google Scholar). To isolate the complete coding region for the murine 5-lipoxygenase cDNA degenerate oligonucleotide primers based on the known human and rat sequences (12Matsumoto T. Funk C.D. Rdmark O. Hg J.-O. Jrnvall H. Samuelsson B. Proc. Natl. Acad. Sci. U. S. A. 1988; 85 (and correction p. 3406): 26-30Google Scholar, 13Dixon R.A.F. Jones R.E. Diehl R.E. Bennett C.D. Kargman S. Rouzer C.A. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 416-420Google Scholar, 14Balcarek J.M. Theisen T.W. Cook M.N. Varrichio A. Hwang S.-M. Strohsacker M.W. Crooke S.T. J. Biol. Chem. 1988; 263: 13937-13941Google Scholar) were designed (Primer 1, 5′-GCCATGCCNTCCTACACNGTCAC-3′ and Primer 2, 5′-TTAGATGGCYACACTGTTYGGAAT-3′; underlined bases indicate start and stop codons, respectively). Two additional primers were prepared based on on the sequence we had obtained from a genomic clone containing exons 4-6 of the murine gene (Primer 3, 5′-ATGGATGGAGTGGAACCCCGG-3′ and Primer 4, 5′-CTGTACTTCCTGTTCTAAACT-3′). RNA was prepared by the method of Chomczynski and Sacchi(21Chomczynski P. Sacchi N. Anal. Biochem. 1987; 162: 156-159Google Scholar). Total RNA (1 μg) obtained from resident peritoneal macrophages of C57BL/6 × 129 F1 mice was used as the starting material for reverse transcriptase-PCR (RT-PCR) carried out by standard procedures(22Funk C.D. FitzGerald G.A. J. Biol. Chem. 1991; 266: 12508-12513Google Scholar). Amplification conditions for primer 2/primer 3 set using one-fifth of the cDNA mixture were: 94°C, 45 s; 46°C, 45 s; 72°C, 1 min 30 s for 35 cycles. A 1.6 kb band was purified by agarose gel electrophoresis and glass powder extraction (Qiagen), and an aliquot was reamplified for an additional 25 cycles. A 0.8-kb product was amplified using primer 1/primer 4 set using the same conditions mentioned above except without subsequent reamplification. Both PCR products were cloned into the pCRII vector (Invitrogen), and the inserts were entirely sequenced by the dideoxy chain termination method. An expression construct (m5LO/Met672) was prepared in the pcDNA1 vector (Invitrogen) by ligation of the two PCR-generated 5-lipoxygenase cDNA fragments. First, the 5′ end of the 5-lipoxygenase cDNA was inserted into the vector as an EcoRI (polylinker derived)/EcoRV 0.5-kb fragment. After verification of the preceding construct, the 3′ end was inserted as an EcoRV/NsiI (polylinker site) 1.5-kb fragment. DNA purified by ion-exchange resin (Qiagen) was transfected into human embryonic kidney 293 cells as described previously by the calcium phosphate method(23Chen X.-S. Brash A.R. Funk C.D. Eur. J. Biochem. 1993; 214: 845-852Google Scholar, 24Gorman C.M. Gies D.R. McGray G. DNA, Protein Engin. Techniques. 1990; 2: 3-10Google Scholar). 48 h later enzyme activity was assayed(25Funk C.D. Gunne H. Steiner H. Izumi T. Samuelsson B. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 2592-2596Google Scholar). A second construct was also prepared for expression studies. The codon for Met672 in the original 3′ end PCR product was changed to Val672 in a PCR reaction using primers 5 and 6 (Fig. 1; 5′-AAGTCTAGATTTAGATGGCCACACTGTTTGG-3′ and 5′-TTCAAGCTGCTGGTA-3′; altered base is underlined and restriction site for cloning is italics). The change was verified by DNA sequencing, and a 0.4-kb PstI/XbaI fragment was replaced into m5LO/Met672 to produce m5LO/Met672Val. Authenticity of the construct was checked by restriction site mapping and DNA sequencing. Subsequently, a 3′ RACE protocol using an oligo(dT) adapter-primer and primer 7 (5′-ATCAGCGTGATCGCCGAG-3′) with newly synthesized cDNA was employed to verify the codon at position 672. Interspecific backcross mapping was performed as described (26Copeland N.G. Jenkins N.A. Trends Genet. 1991; 7: 113-118Google Scholar) by using progeny generated from mating (C57BL/6J × Mus spretus)F1 females and C57BL/6J males. A total of 205 N2 mice were used to map the Alox5 locus (see “Results” for details). DNA isolation, restriction enzyme digestion, agarose gel electrophoresis, Southern blot transfer, and hybridization were performed as essentially described(27Jenkins N.A. Copeland N.G. Taylor B.A. Lee B.K. J. Virol. 1982; 43: 26-36Google Scholar). The probe was a 403-base pair cDNA (bases 432-834) derived by PCR with primer 3/primer 4 set, labeled with [α-32P]dCTP using a nick translation labeling kit (Boehringer Mannheim). Washing was done to a final stringency of 1.0 × SSCP, 0.1% SDS at 65°C. A fragment of 3.8 kb was detected in HindIII-digested C57BL/6J (B) DNA, and a fragment of 10.0 kb was detected in HindIII-digested M. spretus (S) DNA. The presence or absence of the 10.0-kb HindIII M. spretus-specific fragment was followed in backcross mice. A description of the probes and restriction fragment length polymorphisms for the loci linked to the Alox5 locus including microphthalmia (mi), ras-related fibrosarcoma oncogene (Raf1), and ret proto-oncogene (Ret) has been reported previously(28Hodgkinson C.A. Moore K.J. Nakayama A. Steingrimsson E. Copeland N.G. Jenkins N.A. Arnheiter H. Cell. 1993; 74: 395-404Google Scholar, 29Hogan A. Heyner S. Charron M.J. Copeland N.G. Gilbert D.J. Jenkins N.A. Thorens B. Schultz G.A. Development. 1991; 113: 363-372Google Scholar). Recombination distances were calculated as described (30Green E.L. Genetics and Probability in Animal Breeding Experiments. Oxford University Press, New York1981: 77-113Google Scholar) using the computer program SPRETUS MADNESS. Gene order was determined by minimizing the number of recombination events required to explain the allele distribution patterns. Previously, we had isolated a mouse genomic 5-lipoxygenase clone that coded for a small region of 5-lipoxygenase(16Funk C.D. Kurre U. Griffis G. Ann. N. Y. Acad. Sci. 1994; 714: 253-258Google Scholar). To examine potential 5-lipoxygenase isoforms, we sought to clone the complete murine 5-lipoxygenase cDNA by RT-PCR from macrophage RNA. Two overlapping fragments were obtained (Fig. 1). The 2.0-kb cDNA encodes a protein of 674 amino acids (including the initiator Met residue) with a molecular weight of 78,000. Various PCR primer combinations revealed no evidence for splice variants by agarose gel electrophoresis size selection of amplified products and subsequent sequence analysis. Moreover, hybridization of mouse genomic DNA with various 5-lipoxygenase cDNA restriction fragments under reduced stringency conditions did not reveal cross-hybridizing bands or complex band patterns (data not shown). These results are consistent with our previous data that there is only a single 5-lipoxygenase gene(16Funk C.D. Kurre U. Griffis G. Ann. N. Y. Acad. Sci. 1994; 714: 253-258Google Scholar). An expression construct (m5LO/Met672) was prepared by splicing together the two PCR fragments at an unique EcoRV site (Fig. 1). Introduction of this expression vector into human embryonic kidney 293 cells led to expression of immunoreactive 5-lipoxygenase protein of the correct size, but devoid of enzyme activity (Fig. 2). Alignment of the mouse sequence with human and rat sequences and examination of the primer 2 sequence indicated a PCR-based error in the codon for the amino acid two residues from the COOH terminus (amino acid 672), a residue that is conserved in most mammalian lipoxygenases. Methionine would be present at this position instead of valine. Mutagenesis to introduce valine and reconstruction of the expression construct (m5LO/Met672Val) was carried out. In contrast to m5LO/Met672 expression in HEK 293 cells, m5LO/Met672Val exhibited high 5-lipoxygenase enzyme activity (measured as 5-H(P)ETE) with comparable immunoreactive protein levels (Fig. 2, A and B). No 5-lipoxygenase protein and enzyme activity were detected in mock-transfected cells. A 3′ RACE procedure starting with newly synthesized mouse macrophage cDNA and an alternate upstream primer (primer 7) verified the presence of a valine codon at position 672. The intracellular expression pattern of 5-lipoxygenase in paraformaldehyde-fixed, cytospun BMMC preparations was studied by indirect immunocytofluorescence labeling and confocal fluorescence imaging microscopy. In resting, unstimulated BMMC from wild-type mice 5-lipoxygenase was localized primarily within the nucleus (Fig. 3C). This was evident by the extensive co-localization with the nuclear stain Yo-Pro-1 (Fig. 3C) and Z-plane sectioning at 1-μm intervals followed by image reconstruction (not shown). No specific immunofluorescence signal was detected in BMMC from 5-lipoxygenase-deficient mice or when the primary antiserum was substituted with non-immune serum (Fig. 3, A and B). If the divalent cation chelator EDTA was added to the modified Tyrode's buffer for the 30-min incubation period, there was a marked difference in the pattern of 5-lipoxygenase expression. Besides significant enzyme within the nucleus there was now clear evidence for 5-lipoxygenase throughout the cytoplasm (Fig. 3D). Washing the cells with, or without, divalent cations prior to the cell fixation step which took approximately 5-10 min did not appreciably alter this pattern (not shown). When BMMC were activated with IgE/antigen the 5-lipoxygenase “translocated” to a perinuclear location along the nuclear envelope but what appeared to be now mainly on the cytoplasmic side (Fig. 3E). In contrast, if the cells were stimulated with Ca2+ ionophore A23187 the immunofluorescence pattern was different. Signal was detected as a punctate/reticular pattern, possibly associated with structural proteins, both nuclear and perinuclear localized (Fig. 3F). Leukotriene synthesis was associated with IgE/antigen- and A23187-challenged, but not resting and EDTA-treated, BMMC obtained from wild-type mice (Fig. 4). All stimuli using BMMC obtained from 5-lipoxygenase-deficient mice failed to produce detectable leukotriene products (Fig. 4).Figure 4:RP-HPLC detection of lipoxygenase products synthesized by IgE plus antigen-stimulated and calcium ionophore A23187 challenged BMMC cells. +/+, cells obtained from wild-type mouse; −/−, cells obtained from 5-lipoxygenase-deficient mouse. No detectable leukotrienes were synthesized by −/− cells (bottom panel), resting or EDTA-treated cells (not shown). The products eluting at 11.5 (peak I) and 24 (peak II) min co-elute with authentic LTB4 and LTC4 standards, respectively.View Large Image Figure ViewerDownload (PPT) Recent data on the localization of 5-lipoxygenase in alveolar macrophages indicated the presence of enzyme in both membrane and soluble fractions, including nuclear localization in rat basophilic leukemia cells(7Coffey M. Peters-Golden M. Fantone J.C. Sporn P.H.S. J. Biol. Chem. 1992; 267: 570-576Google Scholar, 8Brock T.G. Paine R. Peters-Golden M. J. Biol. Chem. 1994; 269: 22059-22066Google Scholar, 31Peters-Golden M. McNish R.W. Biochem. Biophys. Res. Commun. 1993; 196: 147-153Google Scholar). This raised the possibility of the existence of distinct 5-lipoxygenase isoforms. Protein blot analysis indicated a single immunoreactive 5-lipoxygenase band in alveolar macrophages from normal wild-type mice but no band in 5-lipoxygenase-deficient mice generated by gene targeting (Fig. 5A). Moreover, a product co-eluting with leukotriene C4 was synthesized by A23187-stimulated alveolar macrophages from wild-type mice but not 5-lipoxygenase-deficient mice (not shown). We also examined the expression of genes involved in the production and regulation of leukotriene formation (FLAP and LTA hydrolase) at the protein level to see if there was a compensatory increase, or decrease, in expression in the absence of 5-lipoxygenase. Analysis of BMMC and macrophages from three separate mice using various amounts of protein indicated that leukotriene A4 hydrolase protein levels were not changed; however, FLAP levels dropped approximately 50% in 5-lipoxygenase-deficient mice (Fig. 5B) as revealed by densitometric analysis. The 5-lipoxygenase chromosomal location (designated Alox5) was determined by interspecific backcross analysis using progeny derived from matings of ((C57BL/6J × M. spretus)F1× C57BL/6J) mice. This interspecific backcross mapping panel has been typed for over 1700 loci that are well distributed among all the autosomes as well as the X chromosome(26Copeland N.G. Jenkins N.A. Trends Genet. 1991; 7: 113-118Google Scholar). C57BL/6J and M. spretus DNAs were digested with several enzymes and analyzed by Southern blot hybridization for informative restriction fragment length polymorphisms using a mouse cDNA Alox5 probe. The 10.0-kb HindIII M. spretus fragment length polymorphisms (see “Experimental Procedures”) was used to follow the segregation of the Alox5 locus in backcross mice. The mapping results indicated that Alox5 is located in the central region of mouse chromosome 6 linked to mi, Raf1, and Ret. Although 134 mice were analyzed for every marker and are shown in the segregation analysis (Fig. 6), up to 175 mice were typed for some pairs of markers. Each locus was analyzed in pairwise combinations for recombination frequencies using the additional data. The ratios of the total number of mice exhibiting recombinant chromosomes to the total number of mice analyzed for each pair of loci, and the most likely gene orders are: centromere - mi − 15/175 - Raf1 − 0/138 - Alox5 − 1/151 - Ret. The recombination frequencies (expressed as genetic distances in centiMorgans (cM) ± the standard error) are - mi − 8.6 ± 2.1 - (Raf1, Alox5) − 0.7 ± 0.7 - Ret. No recombinants were detected between Raf1 and Alox5 in 138 animals typed in common suggesting that the two loci are within 2.1 cM of each other (upper 95% confidence limit). 5-Lipoxygenase exists as a single isoform in mice and traffics to various intracellular sites in activated BMMC. Using indirect immunocytofluorescence labeling combined with confocal fluorescence microscopy, 5-lipoxygenase was found almost exclusively within the nucleus of resting BMMC. A similar expression pattern was seen with the transformed RBL-1 rat basophilic leukemia-derived cell line. These basophilic cells bear some resemblance to mucosal mast cells since they secrete mast cell protease II(8Brock T.G. Paine R. Peters-Golden M. J. Biol. Chem. 1994; 269: 22059-22066Google Scholar, 32Zheng Y.L. Chan B.M. Rector E.S. Berczi I. Froese A. Exp. Cell Res. 1991; 194: 301-309Google Scholar). Interestingly, when resting BMMC were incubated in the absence of divalent cations (2 mM EDTA) for 30 min there was diffusion or leakage of 5-lipoxygenase from the nucleus throughout the cytoplasm. Enzyme also remained within the nucleus. The EDTA treatment probably caused a depletion of intracellular divalent cations, in addition to extracellular depletion, by disruption of ion pumps and transporter proteins. Although not proven, these results are suggestive of a primary or secondary divalent cation requirement to maintain nuclear 5-lipoxygenase localization. An intracellular Ca2+ change with ionophore stimulation resulted in a significant rearrangement of nuclear 5-lipoxygenase to a punctate/reticular pattern around the nuclear envelope. Given the recent data that 5-lipoxygenase can bind cytoskeletal proteins by SH3 domain interactions (9Lepley R.A. Fitzpatrick F.A. J. Biol. Chem. 1994; 269: 24163-24168Google Scholar) and previous data that 5-lipoxygenase undergoes a Ca2+-dependent translocation to membrane sites that requires extracellular Ca2+(3Rouzer C.A. Kargman S. J. Biol. Chem. 1988; 263: 10980-10988Google Scholar, 33Wong A. Cook M.N. Foley J.J. Sarau H.M. Marshall P. Hwang S.M. Biochemistry. 1991; 30: 9346-9354Google Scholar) it is possible that 5-lipoxygenase is associating with nuclear filament proteins (lamins) or other cytoskeletal proteins that attach to the nuclear envelope through protein-protein interactions. More precise localization data should be achieved with high resolution electron microscopy. In fact, recent findings in human alveolar macrophages using this technique indicated 5-lipoxygenase association with the euchromatin in resting cells. A23187 stimulation resulted in translocation to the nuclear envelope(34Woods J.W. Coffey M.J. Brock T.G. Singer I.I. Peters-Golden M. J. Clin. Invest. 1995; 95: 2035-2046Google Scholar). Activation of BMMC by IgE/antigen, a challenge known to elicit transient elevation of intracellular Ca2+ in these and rat basophilic leukemia cells(35Wong A. Cook M.N. Hwang S.M. Sarau H.M. Foley J.J. Crooke S.T. Biochemistry. 1992; 31: 4046-4053Google Scholar), also caused translocation of 5-lipoxygenase. The enzyme shifted predominantly to a juxtanuclear position with some localized distribution in the cytoplasm. Malaviya et al.(36Malaviya R. Malaviya R. Jakschik B.A. J. Biol. Chem. 1993; 268: 4939-4944Google Scholar) noticed a reversible translocation of 5-lipoxygenase in mast cells upon IgE/antigen stimulation from a supernatant to pellet fraction by Western blot analysis. The detection of membrane bound 5-lipoxygenase was dependent upon quick-freezing of the cells(36Malaviya R. Malaviya R. Jakschik B.A. J. Biol. Chem. 1993; 268: 4939-4944Google Scholar). How their results correlate with ours is uncertain due to the different means of analysis. Although many questions remain to be answered including: (i) what sequence-specific signals (e.g. nuclear localization signal) control trafficking of 5-lipoxygenase; (ii) what protein-protein interactions govern localization and movement; (iii) how Ca2+ ion or other divalent cation fluxes regulate 5-lipoxygenase compartmentalization; and (iv) how in vitro data obtained on fixed, immobilized cells correlate with in vivo cellular activation, it is becoming clear that the simple 5-lipoxygenase-initiated generation of leukotrienes and their subsequent extracellular transport will have to be modified with novel roles of 5-lipoxygenase within the nucleus. The murine 5-lipoxygenase cDNA was cloned by PCR based on homology with the human (12Matsumoto T. Funk C.D. Rdmark O. Hg J.-O. Jrnvall H. Samuelsson B. Proc. Natl. Acad. Sci. U. S. A. 1988; 85 (and correction p. 3406): 26-30Google Scholar, 13Dixon R.A.F. Jones R.E. Diehl R.E. Bennett C.D. Kargman S. Rouzer C.A. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 416-420Google Scholar) and rat (14Balcarek J.M. Theisen T.W. Cook M.N. Varrichio A. Hwang S.-M. Strohsacker M.W. Crooke S.T. J. Biol. Chem. 1988; 263: 13937-13941Google Scholar) sequences. All are the same size, taking into account a putative error noted at the deduced COOH terminus of the rat sequence(14Balcarek J.M. Theisen T.W. Cook M.N. Varrichio A. Hwang S.-M. Strohsacker M.W. Crooke S.T. J. Biol. Chem. 1988; 263: 13937-13941Google Scholar, 37Minor W. Steczko J. Bolin J.T. Otwinowski Z. Axelrod B. Biochemistry. 1993; 32: 6320-6323Google Scholar). The mouse sequence is 96% identical to the rat sequence and 93% identical to the human 5-lipoxygenase. It shares the conserved histidine and COOH-terminal isoleucine residues found in all lipoxygenases. Based on the crystal structure of the soybean 15-lipoxygenase, these residues act as ligands for the non-heme iron atom(37Minor W. Steczko J. Bolin J.T. Otwinowski Z. Axelrod B. Biochemistry. 1993; 32: 6320-6323Google Scholar, 38Boyington J.C. Gaffney B.J. Amzel L.M. Science. 1993; 260: 1482-1486Google Scholar). Serendipitously, a PCR-generated cloning error revealed the stringent requirements of the amino acid 2 residues upstream of the COOH-terminal isoleucine during expression experiments in HEK 293 cells. A conservative valine to methionine substitution abolished 5-lipoxygenase activity at position 672. This alteration would result in a small increased side chain volume in the vicinity of isoleucine 674, perhaps perturbing the ability of this residue to coordinate the iron atom. Previously, we carried out deletion of the COOH-terminal isoleucine and mutagenesis to 8 different residues using mouse 12-lipoxygenases and verified the essential importance of this residue for enzyme activity(39Chen X.-S. Kurre U. Jenkins N.A. Copeland N.G. Funk C.D. J. Biol. Chem. 1994; 269: 13979-13987Google Scholar). Only valine could be substituted for isoleucine with minimal loss of activity. Moreover, Zhang et al.(40Zhang Y. Rdmark O. Samuelsson B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 485-489Google Scholar) had found that deletion of 6 amino acids from the COOH terminus of human 5-lipoxygenase abolished enzyme activity. Taken together these results indicate the importance of the integrity of the 3 COOH-terminal amino acids of lipoxygenases which probably relates to the ability of the polypeptide chain to fold back and interact with the essential iron atom. 5-Lipoxygenase exists as a single form in the mouse unlike 12-lipoxygenase which has two distinct isoforms encoded by separate, linked genes(39Chen X.-S. Kurre U. Jenkins N.A. Copeland N.G. Funk C.D. J. Biol. Chem. 1994; 269: 13979-13987Google Scholar). First, we were unable to clone any cDNA variants indicative of splice variants. Second, disruption of the 5-lipoxygenase gene removed all 5-lipoxygenase protein and enzyme activity in alveolar macrophages (known to contain both soluble and membrane-bound species) and in IgE/antigen-activated BMMC (where nuclear and perinuclear expression patterns were seen). The polyclonal antibody used in these studies cross-reacts with human, rat(7Coffey M. Peters-Golden M. Fantone J.C. Sporn P.H.S. J. Biol. Chem. 1992; 267: 570-576Google Scholar, 8Brock T.G. Paine R. Peters-Golden M. J. Biol. Chem. 1994; 269: 22059-22066Google Scholar), and mouse 5-lipoxygenases and would predictably detect alternative isoforms, if generated by different genes since 5-lipoxygenases display very high homology across species. Finally, our past data using Southern blot analysis with genomic DNA has indicated a single copy gene with no related cross-hybridizing bands at moderate stringency conditions in mice and humans(15Funk C.D. Hoshiko S. Matsumoto T. Rdmark O. Samuelsson B. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 2587-2591Google Scholar, 16Funk C.D. Kurre U. Griffis G. Ann. N. Y. Acad. Sci. 1994; 714: 253-258Google Scholar). A single report describing human 5-lipoxygenase alternative transcripts (41Boado R.J. Pardridge W.M. Vinters H.V. Black K.L. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 9044-9048Google Scholar) could mean there is heterogeneity in the 3′- or 5′-untranslated regions, or possibly, that these transcripts were not entirely processed after transcription. Disruption of the 5-lipoxygenase gene in mice did not alter expression of leukotriene A4 hydrolase, an enzyme downstream in the pathway of leukotriene B4 synthesis, in macrophages and BMMC. However, FLAP expression was reduced about 50%. FLAP may act as an arachidonic acid transfer protein(6Abramovitz M. Wong E. Cox M.E. Richardson C.D. Li C. Vickers P.J. Eur. J. Biochem. 1993; 215: 105-111Google Scholar). The reason, or mechanism, for the reduced FLAP expression is unknown. The human FLAP and 5-lipoxygenase genes reside on different chromosomes and have different promoter elements(15Funk C.D. Hoshiko S. Matsumoto T. Rdmark O. Samuelsson B. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 2587-2591Google Scholar, 42Kennedy B.P. Diehl R.E. Boie Y. Adam M. Dixon R.A.F. J. Biol. Chem. 1991; 266: 8511-8516Google Scholar, 43Funk C.D. Funk L.B. FitzGerald G.A. Samuelsson B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 3962-3966Google Scholar). Perhaps, intracellular leukotrienes can act by a feedback mechanism to positively regulate FLAP gene expression, or inhibit degradation, and this pathway is abrogated in 5-lipoxygenase-deficient mice. The 5-lipoxygenase gene maps to mouse chromosome 6 by interspecific backcross analysis. We have compared the interspecific map of chromosome 6 with a composite mouse linkage map that reports the map location of many uncloned mouse mutations (compiled by M. T. Davisson, T. H. Roderick, A. L. Hillyeard, and D. P. Doolittle and provided from GBASE, a computerized data base maintained at The Jackson Laboratory, Bar Harbor, ME). Alox5 mapped in a region of the composite map that lacks mouse mutations with a phenotype that might be expected for an alteration in this locus (data not shown). Consistent with this finding was the lack of an observable mutated phenotype in 5-lipoxygenase-deficient mice under normal, non-stressed physiological conditions(17Chen X.-S. Sheller J.R. Johnson E. Funk C.D. Nature. 1994; 372: 179-181Google Scholar, 44Funk C.D. Chen X.-S. Kurre U. Griffis G. Adv. Prostaglandin, Thromboxane, Leukotriene Res. 1995; 23: 145-150Google Scholar). However, these mice exhibited blunted inflammatory responses in certain models of inflammation. The central region of mouse chromosome 6 shares regions of homology with human chromosomes 3 and 10 (summarized in Fig. 6). The human homolog of Alox5 has previously been assigned to human chromosome 10(43Funk C.D. Funk L.B. FitzGerald G.A. Samuelsson B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 3962-3966Google Scholar). The placement of the mouse gene in this region of mouse chromosome 6 confirms and extends this region of homology between mouse and human chromosomes. In conclusion, the murine 5-lipoxygenase has been characterized at several levels. A single 5-lipoxygenase form is distributed within the nucleus in mast cells and apparently traffics to different sites upon cellular activation. The availability of 5-lipoxygenase-deficient mice should prove useful in the elucidation of putative nuclear functions. We thank Ginger Griffis and Mary Barnstead for excellent technical assistance and Drs. Bill Serafin, Lee Limbird, and Alan Brash for helpful discussions. We are grateful to Dr. Tom Jetton and the Vanderbilt Imaging Resource Center for assistance with the immunocytofluorescence experiments and laser scanning confocal microscopy. Dr. Jilly Evans (Merck Frosst) is kindly acknowledged for supplying antisera.