Three human elastase-like genes coordinately expressed in ... - PNAS

Proc. Nati. Acad. Sci. USA Vol. 89, pp. 8215-8219, September 1992

Genetics

Three human elastase-like genes coordinately expressed in the myelomonocyte lineage are organized as a single genetic locus on l9pter MICHAEL ZIMMER*, ROBERT L. MEDCALFt, THOMAS M. FINKS, CHANTAL MATTMANN§, PETER LICHTERt, DIETER E. JENNE§¶

AND

*Abteilung for Molekulare Neuroendokrinologie, Max-Planck-Institut fur Experimentelle Medizin, D-3400 Gottingen, Federal Republic of Germany; tHematology Division, Department of Medicine, University Hospital Center, CH-1011 Lausanne, Switzerland; tDeutsches Krebsforschungszentrum,

D-6900 Heidelberg, Federal Republic of Germany; and §Institut de Biochimie, UniversitM de Lausanne, Chemin des Boveresses, CH-1066 Epalinges s/Lausanne, Switzerland

Communicated by Hans Neurath, June 1, 1992

ine proteases are thus regarded as pathogenic factors in degenerative and inflammatory diseases with abnormal tissue injury-e.g., in the formation of pulmonary emphysema, the adult respiratory distress syndrome, and the reperfusion damage of small blood vessels after ischemia (6) as well as rheumatoid arthritis and glomerulonephritis (7). Patients with congenital a1-antitrypsin deficiency develop early pulmonary emphysema due to poorly controlled NE activity (8). The proteolytically inactive serine protease homologue AZU has, besides its bactericidal activity, monocyte chemotactic activity and is involved in the recruitment of immune defense cells to the site of inflammation (9). Granule-associated serine proteases of neutrophils are also the targets of autoantibody formation. Cathepsin G-, NE-, and PR3-binding antibodies have been detected in the sera of patients suffering from chronic inflammatory diseases (10, 11). In Wegener granulomatosis, autoantibodies against PR3 are regularly found as an obligate feature of this autoimmune vasculitis (10, 11). Furthermore, PR3 appears to regulate growth and terminal differentiation of the myelomonocyte lineage (12, 13). Elastase-like enzymes of neutrophils are also involved in shedding of tumor necrosis factor receptor (14) and release of soluble transforming growth factor a from membrane-bound precursors (15). The cDNAs for NE (16, 17), cathepsin G (18), PR3 (12, 13, 19), and AZU (20, 21) have been cloned and sequenced and the genomic structure of the NE (22-24) and cathepsin G (25) genes has been determined. According to the distribution and phase type of their introns, both genes fall into the sixth class of serine proteases, which represents a separate subfamily (26) that diverged early in evolution. Here we report the cloning of a 50-kilobase (kb) genomic segment that contains the functional genes for AZU, PR3, and NE. II The three genes are expressed at high levels in the neutrophil and monocyte lineage and are coordinately downregulated by phorbol 12-myristate 13-acetate (PMA) in the promonocytic leukemia cell line U937. We also show that this cluster of elastase-related genes is located in the terminal region of the short arm of chromosome 19. MATERIALS AND METHODS Isolation of cDNA Probes for AZU and PR3. Two redundant oligonucleotides, JT91 and JT92, were designed using the published sequences of two peptide fragments from AZU (27). JT91 (5'-ccgaattcCC-CAY-CAR-TTY-CCN-TTY-HT-

The human neutrophil and monocyte-derived ABSTRACT serine protease homologues neutrophil elastase (NE), proteinase 3 (PR3), and azurocidin (AZU) are involved in a variety of immnune defense reactions. NE and PR3 assist in the destruction of phagocytosed microorganisms, cleave the important connective-tissue protein elastin, and generate chemotactic activities by forming a 1-proteinase inhibitor complexes and elastin peptides. AZU is cytotoxic to certain microorganisms and chemotactic for monocytes. AU three proteins are produced and packaged into azurophil granules in large quantities during neutrophil differentiation. We have isolated several cosmid clones each of which contains the functional genes for AZU, PR3, and NE in this order. The PR3 gene is separated by 8 kilobases from the 3' end of the AZU gene and by 3 kilobases from the 5' end of the NE gene. We report a physical map of the gene cluster, its location on chromosome 19pter, and the exon-intron organization of the AZU and PR3 genes. Our fluorescence in situ hybridization studies disprove the previous chromosomal assignment of the human NE gene to 11q14. The five exons of AZU and PR3 are organized like those of NE and other granule-associated serine proteases of hematopoietic cells. NE, PR3, and AZU are coordinately downregulated in the premonocytic cell line U937 during induced terminal differentiation. The cluster-like physical organization of these genes and concerted regulation during hematopoietic differentiation suggests that they are located in a developmentally activated chromatin domain promoting high-level, cell-specific expression in the monocyte-myelocyte lineage.

Neutrophilic granulocytes and macrophages eliminate bacterial and fungal pathogens by two mechanisms: generation of reactive oxygen intermediates through the respiratory burst, and secretion of antibiotic cytotoxic polypeptides into phagocytic vacuoles (1, 2). The oxygen-independent killing depends on proteins packaged into cytosolic dense granules at a distinct stage of myelocyte differentiation. Among the various antibiotic proteins of neutrophilic granules are the highly abundant serine protease homologues cathepsin G, neutrophil elastase (NE), proteinase 3 (PR3) (synonyms: myeloblastin; AGP7, azurophil granule protein 7; Wegener autoantigen) and azurocidin (AZU) (synonym: CAP37, cationic protein 37), that contribute to the destruction of ingested microorganisms (1, 2). In addition, NE and PR3 have important extracellular effects at sites of neutrophil accumulation. NE and PR3 cleave connective-tissue proteins, in particular collagen type IV and elastin (3-5), which confers elastic properties to lungs, arteries, skin, and ligaments. Neutrophil-derived ser-

Abbreviations: NE, neutrophil elastase; PR3, proteinase 3; AZU, azurocidin; PAI-2, plasminogen activator inhibitor 2; PMA, phorbol 12-myristate 13-acetate. $To whom reprint requests should be addressed. I'The sequences reported in this paper have been deposited in the GenBank data base (accession nos. M96326, M96628, M96837, M96838, and M96839).

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

8215

8216

Genetics: Zimmer et al.

Proc. Natl. Acad. Sci. USA 89 (1992)

3') is backtranslated from the amino-terminal sequence ProHis-Gln-Phe-Pro-Phe-Leu (Met), JT92 (5'-ccgaattcGT-SCCNCC-RTC-NCC-RTT-RCA-3') is complementary to the nucleotide sequence encoding the internal peptide Cys-AsnGly-Asp-Gly-Gly-Thr. With JT91 and JT92 as primers, reverse-transcribed poly(A)+ mRNA from U937 cells was amplified with Taq DNA polymerase (Perkin-Elmer). Similarly, the first strand cDNA of PR3 was amplified using the oligonucleotides JT88 and JT90. JT88 (5'-ccgaattcTG-AGCGGT-GCT-GCC-CGA-GCT-3') is identical to the first 20 bases of the PR3 cDNA sequence (12, 13), whereas JT90 (5'-ccgaattC-GAG-GGT-TTG-GAG-CCA-GGC-TC-3') represents its complementary strand from position 680 to 700 (12, 13). EcoRI linkers (lowercase letters) were attached at the 5' ends of the oligonucleotides to facilitate subcloning. Characterization of Genomic Clones. Cosmid clones were isolated from a human genomic library constructed in the cosmid vector pcos2 (28). Nylon filters (NEN) were hybridized to PCR-generated cDNA probes as described (29). The restriction sites for BamHI, Bgl II, EcoRI, Sal I, Sac II, Not I, BssHII, and Mlu I were mapped after linearization of cosmid DNA with A terminase (30). DNA restriction fiagments containing the AZU and PR3 genes were identified by using the PCR-generated probes and additional oligonucleotides for AZU (JT101, 5'-ccgaattCGC-CCC-AAC-AACATG-TGC-3') and PR3 (JT114, 5'-GCT-CAG-CAG-CAAGGC-CAG-CA-3'). The genomic sequence of AZU was assembled from M13 shotgun clones (26) generated for the adjacent 2-kb and 3-kb BamHI fragments (Fig. la) and by double-strand sequencing of selected pUC13 subclones with synthetic oligonucleotides. The 7.6-kb BamHI fragment containing PR3 exons 1-4 and part of exon 5 was characterized by partial sequence and restriction-site analysis. Gene Regulation in PMA-Treated U937 Cells. U937 cells were maintained in RPMI 1640 medium containing 5% fetal bovine serum (GIBCO/BRL) and 2 mM L-glutamine. PMA (25 ng/ml) was added to the cells for various periods up to 48 hr. For Northern blots, poly(A)+ mRNA was prepared as described (31). Nylon filters were hybridized successively to cDNA probes for AZU, PR3, plasminogen-activator inhibitor 2 (PAI-2), and 1-actin (32). Fluorescence in Situ Hybridization. In situ hybridization to metaphase spreads of human chromosomes (33, 34) and differential labeling of the probes PR3-1 (this paper), ZNF61 (34), and pG-A16 (35) by nick-translation were carried out as described. Hybridization signals and the 4',6-diamidino-2phenylindole staining pattern of chromosomes were recorded

by applying conventional epifluorescence microscopy. Digital images were generated by a cooled CCD camera (Photometrics, Tuscon, AZ) and processed electronically. RESULTS AZU, PR3, and NE Genes Are Clustered Within 50 kb. To obtain probes for the AZU and PR3 genes, we have amplified cDNA fragments from reverse-transcribed U937 poly(A)+ mRNA by using AZU- and PR3-specific primers. Five independent cosmid clones, AZU-3, AZU-11, AZU-13, and AZU-14, were obtained by screening a human genomic cosmid library with the AZU probe and two clones, PR3-1 and PR3-3, were identified by subsequent screening with the PR3 probe. The restriction sites for BamHI, EcoRI, Bgl II, and Sal I were mapped in all cosmid clones. Unexpectedly the two sets overlapped to a large extent. The composite restriction map of the 50-kb region covered by these clones is shown in Fig. 1. Southern blot analysis of restriction fragments generated by BamHI, EcoRI, HindIII, and Pst I digestion of cosmid DNA confirmed that both the AZU gene and the PR3 gene were located within the same genomic region. When the restriction map ofthis region was compared with the map of the NE gene (23, 24), the latter gene was found within 3 kb on the 3' side ofthe PR3 gene. The positions of the restriction sites for EcoRJ, BamHI, Bgl II, Sac II, BssHII, and Mlu I (Fig. 1) coincided exactly with the sequence of the NE gene (23, 24). The position of the indicated Not I site (5'-GCGGCCGC-3) falls into a region of the NE sequence (positions 4114-4121) that differs by one base (position 4117, underlined) from the Not I recognition sequence. The absence of this Not I site in the published NE sequence may be explained by a base polymorphism. The three elastase-related genes are arranged in the same transcriptional orientation. AZU and PR3 are separated by about 8 kb and PR3 and NE by 3 kb of genomic DNA. AZU, PR3, and NE Genes Share the Same Exon-Intron Organization. Fig. lb shows the location of the five exons for each of the three genes. The location of the PR3 exons was determined by mapping the restriction sites for Eag I, Nco I, and Dra III within the partially sequenced 7.6-kb BamHI fragment (Fig. lb). These sites occur in different exons-Nco I sites in exons 1 and 4; Eag I sites in the 5' promoter region, in exon 2, and in intron 3; and Dra III sites in exons 3 and 4. The size of intron 4 was confirmed by PCR using primers that hybridize to the antisense strand of exon 4 (JT149, 5'-TGGAC-CCC-ACC-ATG-GCT-CA-3') and to the sense strand of exon 5 (JT85, 5'-ccgaattCAGAAGTTGGCTGCCCG-

Mu I Neu I Mlu I BssH II Sal I Sal I

a BssH II BssH II

Mlu I

Not I BssH II

BssH

Not I

Sal I

I~~fl2~~ ~IL ~~RJ i II BR G RGIH lBll- R~ -IBl II _I _VI_B||

4 ||BI G

B

B

B B

G B Tl _

|

B

1 -

PR3-i1

bBam HI Sac II

AZI-3, AZI- I 1 AZI-13, AZI-14

PR3-3 BamHI BssHII

SacII

MluI BamrHI

AZU 1

2

Nco I Bam HI 1

BssH II Dram

Il; Eagl

Ba HI

PR3 Nlot I

Ec,co RII IB~ I1I

1kb

Nco I Dra4fi

Mflu I

ISac - * H1-

BamnHI BssH II

Not I Sac III

Eco RI

NE 4 Bam HI

FIG. 1. Physical map and organization of the AZU-PR3-NE gene cluster. (a) Composite restriction map of cosmid clones isolated with AZU and PR3 probes. Positions of the three genes are indicated by black boxes. Relative positions of the cosmid clones are depicted below by lines. The cosmid clone PR3-3 (arrow) extends further in 3' direction. Restriction enzymes: B, BamHI; G, Bgi II; R, EcoRI. (b) Exon-intron structure of the AZU, PR3, and NE genes. Exons are depicted by black boxes and intervening sequences by connecting lines. Exon distribution of the NE gene is taken from previous reports (22-24). Horizontal arrows mark the position and orientation of Alu sequences.

Genetics: Zimmer et al.

Proc. Natl. Acad. Sci. USA 89 (1992)

GGTT-3') respectively. Exon-intron organization of the AZU and PR3 genes is very similar to that of the NE gene (22-24). Intron 1 occurs in phase 1 and falls at the codon for Gly-(-7), which is located in the prepropeptide. Intron 2 splits the codon for residue 62 (chymotrypsin residue numbering) in phase 2. Introns 3 and 4 occur in phase 0 between the codons for Gln-107 and Leu-108 (the actual residue numbers in mature AZU are Gln-94 and Leu-95) and the codons for residue 192 and Gly-193, respectively. Thus all three serine protease homologues in this gene cluster are members of the sixth class of serine protease genes according to a previously extended classification scheme (26). Fig. 2a shows parts of the genomic AZU sequence. The sequence includes a TATAA box, located 41 bases upstream of the translation start codon, and a putative CAAT box 75 bp further upstream, with typical Spl binding elements at positions 28 and 179. Similar sequence elements are also found in the promoter region of the PR3 gene (Fig. 2b). The putative TATA box is located at position 643, a CAAT box in reverse direction (CATTG) at position 610, a Spl element at positions 498 and an AP1 binding sequence, 5'-TGACTCA3', in the first intron at position 943 (36). Most interestingly, a binding sequence of the immunoglobulin heavy chain enhancer, 5'-CAGGTGGC-3' (37), is perfectly conserved in each of the six-fold repeats (positions 388-705) of the NE gene (23). Two copies differing by a single base pair are also present in the promoter region of the PR3 gene. One copy, 5'-CAGGTGTC-3', is at position 105, the second copy, 5'-CATGTTGC-3', is at position 443. E box-like sequences conforming to the consensus CANNTG (38) are present at positions 105, 230, and 418. Ets-like binding site motifs,

8217

AGGAGGA(T/A)G (36, 38), are found at positions 208 and 584. A sequence motif recognized by c-Myb (36), 5'CAACGG-3', is located at position 622, immediately upstream of the TATA box. These sequence elements may be binding sites that are recognized by known or structurally related DNA-binding factors. Highly repetitive sequence elements are present in the 5-kb genomic region containing the AZU gene. Full-length copies of Alu sequences are found in the third intron between positions 2417 and 2703, in the fourth intron between positions 3525 and 3810 in reverse orientation, and in the 3' flanking region between positions 4661 and 4951. In addition, partial copies of Alu sequences are inserted into intron 3 between 2246 and 2369 and between 2895 and 3014 in reverse orientation. A sequence that is homologous to the IR3 repetitive region of the Epstein-Barr virus genome (39) occurs in the large second intron of the AZU gene between positions 800 and 1800. AZU and PR3 Genes Are Downregulated During U937 Differentiation. U937 cells can be differentiated to macrophage-like cells by treatment with PMA. During this process, the transcription of several genes is downregulated-e.g., c-myc, cathepsin G, and PR3 (12, 40). PR3 has been identified as a regulator of growth and differentiation (12, 13). We compared mRNA levels for AZU, PR3, PAI-2, and 1-actin during PMA stimulation. AZU and PR3 mRNA levels remained constant for the first 4 hr (Fig. 3) but then declined steadily to undetectable levels after 24 hr of exposure to PMA. In contrast, PAI-2 mRNA, undetectable in unstimulated U937 cells, increased strongly, peaking after 8 hr.

a

1 101 201

GGATCCACTG GTTCCTGACA CCCTCACCTG CCCCTGGGGG TGTGGCCATC TTCTAGAGAG CTGGCCCAGG GAGCAGTTGG CGGTGGAGGC CITGGGA&LMCCCGTGTT CCCACTGAGT CAAGGCCTGT G!ATMGGGC AGCCGCCGCC TTAGCCACAG ACCTGCCCCG CCATGACCCG -26 M T R 301 TCGAGGGCCG =GAGTGCCT CTCTGTGCCG GTGGTCCCCC ATCTGTGCTA GGGCCCGGCT S R A (G) -7 601 CTTGGGGATC TCAGAGCTGT CTCCCCCCGA CCCMGCTCC AGCCCCCTTT TGGACATCGT -6 S S P L L D I V 701 GCCTCCATTC AGAATCAAGG CAGGCACTTC TGCGGGGGTG CCCTGATCCA TGCCCGCTTC A S I 0 N Q G R H F C G G A L I H A R F 2000 AOGAACCCCGG GGTTAGCACC GTGGTGCTGG GTGCCTATGA CCTGAGGCGG CGGGAGAGGC 47 N P G V S T V V L G A Y D L R R R E R Q 2101 TGGCTACGAC CCCCAGCAGA ACCTGAACGA CCTGATGCTG CTTCAG=GA GAGGATGGTG G Y D P Q Q N L N D L M L L Q 94 3101 CTCCAGTCCC CAGGGCCACC CTCCCCTGAC TCCATTTCCT TCCCCAXTG GACCGTGAGG 95 L D R E A 3201 GCAGAACGCC ACGGTGGAAG CCGGCACCAG ATGCCAGGTG GCCGGCTGGG GGAGCCAGCG T V E A G T R Q N A C Q V A G W G S Q R 3301 GTGACTGTGA CCCCCGAGGA CCAGTGTCGC CCCAACAACG TGTGCACCGG TGTGCTCACC 0 C R P N N V C T G V L T V T V T P E D 4101 GGGTGGCGTG GGAGCCAGGC CCTGGGACGC CCTGACACAG CTGCTGCCTG CCC2aGGGGA 173 G D 4201 GTGGCCTCCT TTTCCCTGGG GCCCTGTGGC CGAGGCCCTG ACTTCTTCAC CCGAGTGGCG V A S F S L G P C G R G P D F F T R V A 4301 GACCGGGGCC AGCCTAGGGG GGCCTGTGAC CTCCCATGGA GCCCAGCCCC CGCCCTCCAC A * P G P 225

GGAAACTGAG GATCAGTGCA GAATGTAGGG GGAGCCCAGG GGGGCTGTCC CTGGGCCTGG GCGGGGACGC CACCAACTGC GCTGACAGTC CTGGCCCTGC TGGCTGGTCT GCTGGCATCC L T V L A L L A G L L A S GCCAGGCAG AACTCAGACT TAAAGCACAGAGAAGGC...

TGGCGGCCGG AAGGCGAGGC G G R K A R P GTGATGACCG CGGCCAGCTG V M T A A S C ATCCCGCCA GACGTTTTCC S R Q T F S CCACCTGTGA TCCCAGCACC

CCCGCCAGTT CCCGTTCCTG R Q F P F L

CTTCCAAAGC CACIjGAG... 0 46 F 0 S ATCAGCAGCA TGAGCGAGAA I S S M S E N TCGGGAGGCC GACGTTA...

CCAACCTCAC CAGCAGCGTG ACGATACTGC CACTGCCTCT N L T S S V T I L P L P L CAGTGGGGGG CGTCTCTCCC GTTTTCCCAG GTTTGTCAAC S G G

R L S R

F P R

F V N

CGCCGCGGTG GCATCTGCAA T=IGAGTGCT CCCTGTG... G I C N 172

R R G

CGGGGGCACC CCCCTCGTCT GCGAGGGCCT GGCCCACGGC P L V C E G L G G T A H G CTCTTCCGAG ACTGGATCGA TGGTGTTCTC AACAACCCGG L F R D G V L W I D N N P G ACCTCCGGCG CTCCGCACCC ACCTCCCACG GCCCCGC...

b 583 683

AAGGAGGAAG TGGGGACCCA GCCTGGCA1L0GCAACTC CATGGCTCAC CGGCCCCCCA GCCCTGCCCT GGCGTCCGTG R P P S P A L A S V TGGACTCCCC CCCTGC2GGT GCTGCCCGAG CTGCGGAGAT -6 A A R A A E I GGGGAACCCG GGCAGCCACT TCTGCGGAGG CACCTTGATC G N P C G G T L I G S H F GACCACCCCA CCCCCGCAA CCCCAGCGCC TGGTGAACGT 50 P Q R L V N V TCAGGTGTTT CTGAACAACT ACGACGCGGA GAACAAACTG L N N Y Q V F N K L D A E CCGCCTGCCT TCTGCCCCS CTGAGCAGCC CAGCCAACCT 97 L S S P A N L CCAGTGCCTG GCCATGGGCT GGGGCCGCGT GGGTGCCCAC Q C L A M GN G R V G A H CCACATAACA TTTGCACTTT CGTCCCTCGC CGCAAGGCCG P H N I C T F V P R R K A G

-27 M A H ...

...

...

...

AACGGCCTCT GGCATTGGGC TIAT GAGGA CTGCTGGCCT TGCTGCTGAG CGGCGAGTGA L L A L L L S (G) -7 CGTGGGCGGG CACGAGGCGC AGCCACACTC V G G H E A 0 P H S

GCTTGACCGT GGGTGCACCC TGGACCCCAC GCCACGTGCC CATCCATCCA GCTTCCA...

CCGGCCCTAC ATGGCCTCCC TGCAGATGCG R P Y M A S L Q M R ACATMIGAGC GGCCGCC... V L T A A H C L R D (I) 49 GCCCACAACG TGCGGACGCA GGAGCCCACC CAGCAGCACT TCTCGGTGGC A H N V R T Q E P T S V A Q Q H F TCCTCATCCA GICGGGCGGG ... L I 0 96 GTCGCCACAG TCCAGCTGCC ACAGCAGGAC CAGCCAGTGC CCCACGGCAC 0 L P 0 0 D V A T V Q P V P H G T CCCAGGTCCT GCAGGAGCTC AATGTCACCG TGGTCACCTT CTTCTGCCGG Q V L Q E L N V T V V T F F C R CGIAAGTAAC CGTGCCCCCA C...

CACCCCAGCT TCGTGCTGAC GGCCGCGCAC TGCCTGCGGG

H P S F GGTGCTCGGA V L G AACGACGTTC N D V L CAGTGCCTCC S A S GACCCCCCAG D P P A GCATCTGCTT I C F TGGCCGTCCC TGTCCTCC2a GGAGACTCAG GTGGCCCCCT GATCTGTGAT 174 G D S G G P L I C D CCTTTTCCCT GACTTCTTCA CGCGGGTAGC CCTCTACGTG GACTGGATCC L F P D F F T R V A L Y V D W I R

173

GGCATCATCC AAGGAATAGA CTCCTTCGTG ATCTGGGGAT GTGCCACCCG G I I Q S F V I W G C G I D A T R GTTCTACGCT GCGCCGTGTG GAGGCCAAGG GCCGCCCCTG AACCGCCCCT S T L R R V E A K G R P * 229

FIG. 2. Nucleotide sequences of the AZU und PR3 genes. (a) Only the 5' flanking region, the five exons, and the exon-intron boundaries of the AZU gene are shown together with the translation of exons in single-letter code under the first nucleotide of each codon. The putative CAAT box, TATA box, and 5' and 3' splice junctions (GT and AG) are underlined. Numbering starts with the first nucleotide sequenced. (b) The 5' sequence of the PR3 gene and the sequences for all exons including exon-intron boundaries are shown. Bases are numbered only from the first base of the BamHI site to the last base of the EcoRI site downstream of exon 1. The putative CAAT box (opposite orientation), TATA box, and the splice junctions of exons 1-5 are underlined. The most likely start site of transcription is located at position 669 in agreement with consensus criteria and cDNA sequencing data (19). Amino acid residues interrupted by introns are in parentheses. Amino acid residues of the AZU prepropeptide are numbered from -26 to -1, and those of PR3 from -27 to -1. The residues of mature AZU are numbered 1-225 and agree with the cDNA-derived AZU sequence of ref. 20; the sequence of mature PR3 (numbered 1-229) differs only by Arg-228, which is missing in ref. 19.

8218

~ ~ ~.

Genetics: Zimmer et al. 0

30' *

2

4

8

Proc. Natl. Acad. Sci. USA 89 (1992) 24h

....

AZLU PR3

........ ..' .........W..-

PAI-2

_-actin FIG. 3. Downregulation of AZU and PR3 mRNA in U937 cells. U937 cells were treated with 25 ng of PMA per ml of culture medium for the indicated times (30 min to 24 hr). Partially purified poly(A)+ RNA (4 ,ug per lane) was separated by electrophoresis, blotted onto GeneScreenPlus membranes (NEN), and probed with 32P-labeled cDNA probes for AZU, PR3, PAI-2, and ,-actin as indicated at right. The 28S and 18S ribosomal markers are indicated at left. Note the downregulation of AZU and PR3 transcripts, the upregulation of PAI-2 transcripts, and the unchanged P-actin mRNA levels.

3-Actin mRNA remained constant. mRNA decay curves for AZU and PR3 showed virtually identical time courses. AZU-PR3-NE Gene Cluster Maps to l9pter. Although the human NE gene has been mapped to 11q14 by radioactive in situ hybridization (22), we have reexamined the chromosomal position of the AZU-PR3-NE gene cluster by fluorescence in situ hybridization. Specific signals were found on both sister chromatids in the telomeric region ofthe short arm of chromosome 19 (Fig. 4A). The short arm of chromosome 19 was identified by simultaneously counterstaining with 4',6-diamidino-2-phenylindole. During the evaluation of 70 metaphase spreads from four independent hybridization experiments, we obtained signals from 85% of the target sequences and no additional fluorescent signals on any other chromosome. To document the presence of the elastase-like gene cluster on l9pter directly, the biotinylated PR3-1 probe, the l9qter-specific biotinylated ZNF61 probe (34), and the digoxigenin-labeled pG-A16 probe, which specifically binds to the centromeric region of chromosomes 5 and 19 (35), were simultaneously hybridized to elongated chromosomes. Strong pG-A16-specific rhodamine signals (pairs of partially overlapping red spots) were found in the centromeric region of chromosomes 5 (data not shown) and 19 (Fig. 4B and C); whereas pairs of yellow fluorescein signals were elicited in both telomeric regions of chromosome 19. Thus, our findings clearly reveal the location of the AZU-PR3-NE gene cluster on chromosome l9pter. DISCUSSION In this study, the functional genes for AZU, PR3, and NE, all encoding neutrophil-derived elastase homologues, have been shown to be physically linked within a 50-kb region on chromosome l9pter. The AZU, PR3, and NE genes form a second cluster of genes encoding hematopoietic serine proteases, the first one being the cluster of the cathepsin G, granzyme H (CGL-2), and granzyme B (CGL-1) genes on 14q11.2 (41, 42). AZU is the oldest branch of the neutrophil serine protease cluster on 19pter and shares only 44% identical amino acid residues with NE and PR3, whereas cathepsin G, the oldest branch of the 14q11.2 gene cluster, shows 55% identity with granzymes B and H (42). In view of the greater degree of sequence divergence it is surprising that all three elastase homologues have maintained a myelomonocyte-specific pattern of gene expression through evolution,

FIG. 4. AZU-PR3-NE gene cluster maps to 19pter. (A) Hybridization of cosmid PR3-1 to human metaphase chromosomes yields highly specific fluorescent signals (arrows). (B and C) Simultaneous visualization of the biotinylated PR3-1 probe that covers the AZU, PR3, and NE genes; the digoxigenin-labeled pG-A16 probe (35), which is specific for the centromeres of chromosome 5 and 19; and the biotinylated zinc finger probe ZNF61, which has been mapped to 19qter (34). Five nanograms ofthe probe pG-A16, 60 ng of the probes PR3-1 and ZNF61, and 30 1Lg of Cot-1 DNA (GIBCO-BRL) were added to 30 Al of hybridization solution. The pG-A16 probe was detected via rhodamine-conjugated anti-digoxigenin antibodies (red spots); PR3-1 (arrows) and ZNF61 (arrowheads) probes were detected via fluorescein-conjugated avidin (yellow). Two chromosome 19 homologues (B and C) from different metaphase spreads are depicted. A-C represent overlaid digitized images.

whereas the cathepsin G homologues have acquired expression specificity for different hematopoietic lineages (42). To firmly establish the genetic locus of AZU, PR3, and NE on chromosome 19, we have performed cohybridization experiments with two chromosome 19-specific reference probes (Fig. 4). The genomic probe for the NE gene, which includes long intron sequences, is unlikely to detect homologous fully processed (intronless) pseudogenes of NE at other chromosomal loci. In contrast, the 0.65-kb cDNA fragment, which was previously used for radioactive in situ localization of the NE gene, has a clear potential to hybridize to NE-related pseudogene sequences. Nevertheless, only a single binding site on 11q14 was reported for the NE cDNA probe. Strong evidence against multiple NE and PR3 loci was provided by Southern blot experiments, which showed that NE (16, 22) as well as PR3 (21, 43) are encoded by single-copy genes in the human haploid genome. Thus we conclude that the previous chromosomal assignment of the NE gene is incorrect. Organization of three elastase homologues in a single cluster is reflected by a common tissue-specific and stagerelated pattern of gene expression during hematopoietic differentiation. Human promonocyte-like U937 cells express NE (17, 44) and PR3 (12) as well as AZU (this study). Steady-state levels of PR3 and AZU mRNA were downreg-

Genetics: Zimmer et al. ulated by PMA, becoming undetectable after 24 hr. The kinetics observed agrees with the previously reported decline of elastinolytic activity after 12 hr of phorbol ester treatment. These findings, together with previous studies (12, 17, 44, 45), suggest that high-level, lineage-specific gene expression of all elastase homologues is developmentally restricted to the promyelomonocyte stage of phagocyte maturation, when biogenesis and assembly of microbicidal storage granules occurs. Common or conserved cis-acting regulatory DNA elements in the NE locus may synchronize the temporal and tissue-specific pattern of gene expression. That the three genes evolved from a common ancestor in the same region of the genome and that their protein products are sequestered in the same cellular organelle during hematopoietic differentiation support the hypothesis that the three proteins have multiple, partially overlapping biological functions that pertain to defense against invading pathogens (46-48). The pathophysiological significance of NE and PR3 is highlighted by the fact that a genetic deficiency of alantitrypsin, a specific potent inhibitor of both PR3 and NE, is associated with increased tissue damage and premature development of pulmonary emphysema (8). PR3 and NE are produced at high levels during normal differentiation of the myelocyte lineage (48). The daily turnover of NE amounts to =-250 mg per kilogram of body weight. We hypothesize that genetic predisposition to pulmonary emphysema and degenerative connective-tissue diseases may be determined not only by the locus for a1-antitrypsin but also by the elastase locus on l9pter itself. Certain alleles of PR3 and NE may display amino acid substitutions in those surface regions of the enzymes which are important (49) for the tight irreversible binding of specific endogenous inhibitors. These dominant enzyme variants would then partially escape the control of inhibitors and would have a greater potential to damage connective tissues during normal aging or inflammatory reactions. Another interesting feature of the elastase locus is the involvement of NE and PR3 antigens in autoimmune reactions (10, 11). Two circumstances may contribute to the antigenic potential of these granule proteins: their high catabolic turnover in antigen-presenting phagocytic cells, in particular during microbial infections, and the strongly cationic nature of their polypeptide chains. All three elastase homologues show clusters of highly charged amino acid residues, which are a characteristic feature in the majority of identified autoantigens (50). The knowledge of the genetic locus and genomic organization for these neutrophil-derived autoantigens will help clarify whether polymorphisms in the coding region of elastase homologues predispose to increased turnover of extracellular matrix components and to autoantibody formation against neutrophils. Note. Using a PCR-generated cDNA probe for human NE, we confirmed that NE transcripts in U937 cells were also downregulated during PMA treatment. We thankJ. Tschopp and J. Spiess for their help and interest during this study and Christina Czerny for technical assistance. The work was supported by grants from the Swiss National Science Foundation (no. 31-30809.91 to D.E.J. and no. 31-26399.89 to R.L.M.) and the Roche Research Foundation (to D.E.J.) and by the Verein zur Forderung der Krebsforschung in Deutschland (to P.L.). 1. Spitznagel, J. K. (1990) J. Clin. Invest. 86, 1381-1386. 2. Lehrer, R. I. & Ganz, T. (1990) Blood 76, 2169-2181. 3. Kao, R. C., Wehner, N. G., Skubitz, K. M., Gray, B. H. & Hoidal, J. R. (1988) J. Clin. Invest. 82, 1963-1973. 4. Lfidemnann, J., Utecht, B. & Gross, W. L. (1990) J. Exp. Med. 171, 357-362. 5. Rao, N. V., Wehner, N. G., Marshall, B. C., Gray, W. R., Gray, B. H. & Hoidal, J. R. (1991) J. Biol. Chem. 266, 9540-9548. 6. Smedly, L. A., Tonnesen, M. G., Sandhaus, R. A., Haslett, C., Guthrie, L. A., Johnston, R. B., Jr., Henson, P. M. & Worthen, G. S. (1986) J. Clin. Invest. 77, 1233-1243.

Proc. Natl. Acad. Sci. USA 89 (1992)

8219

7. Weiss, S. J. (1989) N. Engl. J. Med. 320, 365-376. 8. Crystal, R. G. (1989) Trends Genet. 5, 411-417. 9. Pereira, H. A., Shafer, W. M., Pohl, J., Martin, L. E. & Spitznagel, J. K. (1990) J. Clin. Invest. 85, 1468-1476. 10. Kallenberg, C. G. M., Tervaert, J. W. C., Van der Woude, F. J., Goldschmeding, R., von dem Borne, A. E. G. K. & Weening, J. J. (1991) Immunol. Today 12, 61-64. 11. Jennette, J. C., Charles, L. A. & Falk, R. J. (1991) Int. Rev. Exp. Pathol. 32, 193-221. 12. Bories, D., Raynal, M. C., Solomon, D. H., Darzynkiewicz, Z. & Cayre, Y. E. (1989) Cell 59, 959-968. 13. Jenne, D. E., Tschopp, J., Lddemann, J., Utecht, B. & Gross, W. L. (1990) Nature (London) 346, 520. 14. Porteu, F., Brockhaus, M., Wallach, D., Engelmann, H. & Nathan, C. F. (1991) J. Biol. Chem. 266, 18846-18853. 15. Mueller, S. G., Paterson, A. J. & Kudlow, J. E. (1990) Mol. Cell. Biol. 10, 4596-4602. 16. Farley, D., Salvesen, G. & Travis, J. (1988) Biol. Chem. Hoppe-Seyler 369, 3-7. 17. Takahashi, H., Nukiwa, T., Basset, P. & Crystal, R. G. (1988) J. Biol. Chem. 263, 2543-2547. 18. Salvesen, G., Farley, D., Shuman, J., Przybyla, A., Reilly, C. & Travis, J. (1987) Biochemistry 26, 2289-2293. 19. Campanelli, D., Melchior, M., Fu, Y., Nakata, M., Shuman, H., Nathan, C. & Gabay, J. E. (1990) J. Exp. Med. 172, 1709-1715. 20. Morgan, J. G., Sukiennicki, T., Pereira, H. A., Spitznagel, J. K., Guerra, M. E. & Larrick, J. W. (1991) J. Immunol. 147, 3210-3214. 21. Almeida, R. P., Melchior, M., Campanelli, D., Nathan, C. & Gabay, J. E. (1991) Biochem. Biophys. Res. Commun. 177, 688-695. 22. Takahashi, H., Nukiwa, T., Yoshimura, K., Quick, C. D., States, D. J., Whang-Peng, J., Knutsen, T. & Crystal, R. G. (1988) J. Biol. Chem. 263, 14739-14747. 23. Farley, D., Travis, J. & Salvesen, G. (1989) Biol. Chem. Hoppe-Seyler 370, 737-744. 24. Nakamura, H., Okano, K., Aoki, Y., Shimizu, H. & Naruto, M. (1987) Nucleic Acids Res. 15, 9601-9602. 25. Hohn, P. A., Popescu, N. C., Hanson, R. D., Salvesen, G. & Ley, T. J. (1989) J. Biol. Chem. 264, 13412-13419. 26. Jenne, D. E., Lichter, P., Zimmer, M., Garcia-Sanz, J. A. & Tschopp, J. (1991) J. Immunol. 147, 1045-1052. 27. Pohl, J., Pereira, H. A., Martin, N. M. & Spitznagel, J. K. (1990) FEBS Lett. 272, 200-204. 28. Poustka, A., Rackwitz, H. R., Frischauf, A. M., Hohn, B. & Lehrach, H. (1984) Proc. Natl. Acad. Sci. USA 81, 4129-4133. 29. Church, G. M. & Gilbert, W. (1984) Proc. Natl. Acad. Sci. USA 81, 1991-1995. 30. Rackwitz, H. R., Zehetner, G., Murialdo, H., Delius, H., Chai, J. H., Poustka, A., Frischauf, A. & Lehrach, H. (1985) Gene 40, 259-266. 31. Thomas, P. S. (1980) Proc. Natl. Acad. Sci. USA 77, 5201-5205. 32. Schleuning, W. D., Medcalf, R. L., Hession, C., Rothenbuhler, R., Shaw, A. & Kruithof, E. K. (1987) Mol. Cell. Biol. 7, 4564-4567. 33. Lichter, P., Tang, C. J., Call, K., Hermanson, G., Evans, G. A. & Housman, D. (1990) Science 247, 64-69. 34. Lichter, P., Bray, P., Ried, T., Dawid, I. B. & Ward, D. C. (1992) Genomics 13, in press. 35. Hulsebos, T., Schonk, D., van Dalen, I., Coerwinkel Driessen, M., Schepens, J., Ropers, H. H. & Wieringa, B. (1988) Cytogenet. Cell Genet. 47, 144-148. 36. Faisst, S. & Meyer, S. (1992) Nucleic Acids Res. 20, 3-26. 37. Ephrussi, A., Church, G. M., Tonegawa, S. & Gilbert, W. (1985) Science 227, 134-140. 38. Leiden, J. M. (1992) Immunol. Today 13, 22-30. 39. Heller, M., Flemington, E., Kieff, E. & Deininger, P. (1985) Mol. Cell. Biol. 5, 457-465. 40. Hanson, R. D., Connolly, N. L., Burnett, D., Campbell, E. J., Senior, R. M. & Ley, T. (1990) J. Biol. Chem. 265, 1524-1530. 41. Hanson, R. D., Hohn, P. A., Popescu, N. C. & Ley, T. J. (1990) Proc. Natl. Acad. Sci. USA 87, 960-963. 42. Haddad, P., Jenne, D., Tschopp, J., Clement, M. V., Mathieu-Mahul, D. & Sasportes, M. (1990) Int. Immunol. 3, 57-66. 43. Labbaye, C., Musette, P. & Cayre, Y. E. (1991) Proc. Natl. Acad. Sci. USA 88, 9253-9256. 44. Welgus, H. G., Connolly, N. L. & Senior, R. M. (1986) J. Clin. Invest. 77, 1675-1681. 45. Fouret, P., du Bois, R. M., Bernaudin, J. F., Takahashi, H., Ferrans, V. J. & Crystal, R. G. (1989) J. Exp. Med. 169, 833-845. 46. Wilde, C. G., Snable, J. L., Griffith, J. E. & Scott, R. W. (1990) J. Biol. Chem. 265, 2038-2041. 47. Campanelli, D., Detmers, P. A., Nathan, C. F. & Gabay, J. E. (1990) J. Clin. Invest. 85, 904-915. 48. Gabay, J. E., Scott, R. W., Campanelli, D., Griffith, J., Wilde, C., Marra, M. N., Seeger, M. & Nathan, C. F. (1989) Proc. Natl. Acad. Sci. USA 86, 5610-5614. 49. Bode, W., Meyer, E. J. & Powers, J. C. (1989) Biochemistry 28, 19511963. 50. Brendel, V., Dohlman, J., Blaisdell, B. E. & Karlin, S. (1991)Proc. Natl. Acad. Sci. USA 88, 1536-1540.