THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 263, No. 5, Issue of February 15. pp. 2543-2547, 1988 Printed in U.S.A.

Myelomonocytic Cell Lineage Expression of the Neutrophil Elastase Gene*

(Received for publication, October 21, 1987)

Hideki Takahashi, Toshihiro Nukiwa, Paul Basset, and Ronald G . Crystal$ From the Pulmonary Branch, National Heart, Lung and Blood Institute, National Institutes of Health, Bethesda, Maryland 20892

Human neutrophil elastase (NE) functions as a pow- erful serine protease capable of attacking a broad range of proteins. To examine the cellular site(s) of NE gene expression, a 0.65-kilobase cDNA (pPB15) com- plementary to the coding region of the NE gene was cloned from the cell line U937 using an oligonucleotide based on the known NE protein sequence. The sequence of pPB15 demonstrated that it coded for the 173 C- terminal residues of the 218 amino acids that comprise the mature NE protein, plus an additional 3‘ 60 base pairs prior to the in-frame stop codon, suggesting the NE mRNA contains sequences for a 20-residue C-ter- minal “pro” peptide that is not found in the mature protein. Northern analysis using 32P-labeled pPB15 as a probe revealed that neutrophils do not contain de- tectable NE mRNA transcripts despite the fact that this cell carries large amounts of this protein. Further- more, resting and activated blood monocytes also con- tained no detectable NE mRNA transcripts, although these cells also carry detectable NE. In contrast, bone marrow precursor cells contained NE transcripts, sug- gesting the NE gene is expressed in blood precursor cells. In this regard, evaluation of HL-60 cells, a human cell line with myelomonocytic lineage features, dem- onstrated NE transcripts in resting cells and increased NE mRNA levels when the cells were induced toward the myelocytic lineage with dimethyl sulfoxide. How- ever, when the HL-60 cells were induced toward the monocytic lineage with phorbol 12-myristate 13-ace- tate, NE transcripts were lost even though transcripts for interleukin-lj3 were plentiful. Together, these ob- servations are consistent with the concept that the NE gene is not expressed in the blood cells that carry the protein, but in bone marrow precursors that express NE transcripts about the time of commitment to the myelocytic series.

Human neutrophil elastase (NE)’ is a 218-amino acid single chain glycoprotein that functions as a potent proteolytic

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in thispaper has been submitted to the GenBankTM/EMBL Data Bank with accession number(s) 503545.

6D03, NIH, Bethesda, MD 20892. $To whom reprint requests should be addressed Bldg. 10, Rm.

The abbreviations used are: NE, neutrophil elastase; DMEM, Dulbecco’s modified Eagle’s medium; FBS, fetal bovine serum; PMA, phorbol 12-myristate 13-acetate; HEPES, 4-(2-hydroxyethyl)-l-pi- perazineethanesulfonic acid; IL, interleukin; kb, kilobase; LPS, lipo- polysaccharide.

enzyme capable of destroying a broad range of substrates in the physiologic conditions of the extracellular milieu (1-6). There are at least three forms of NE; current evidence sug- gests all have the same polypeptide backbone but differ in the two asparaginyl N-linked carbohydrate side chains (2-4). The protein contains four intramolecular disulfide links and is very rich in arginyl residues, located mostly on the outside of the molecule, giving it a very high isoelectric point (2-4). NE is classified as a “serine protease”; its reactive site is comprised of the “catalytic triad” H i ~ ~ ’ - A s p ~ ~ - S e r ’ ~ ~ , in which a “charge- relay system” allows the His4’ and Asprn to transiently bind a proton from the Ser173 which becomes a powerful nucleophile that attacks the peptide bond in the target protein (2-4). As its name suggests, NE is found in larger abundance in the blood neutrophil where it is stored at 0.5-3 pg/cell in cyto- plasmic azurophilic granules and is rapidly released when the neutrophil is activated (1, 2).

NE is classified as an “elastase” (EC 3.4.21.37) because it is one of a small group of proteases capable of attacking insoluble elastin, a highly cross-linked rubber-like macromol- ecule that modulates the structural integrity and mechanical properties of connective tissue matrixes (1, 2, 7). However, despite the specificity of its name, NE can also destroy most other matrix components (including collagen types I-IV, fi- bronectin, and proteoglycans), as well as coagulation factors, immunoglobulins, complement components, and Escherichia coli cell walls (2). As such, it has been hypothesized that NE plays a role in normal tissue turnover, the clearing of extra- cellular macromolecular debris during wound healing, the regulation of coagulation and immune responses, and anti- bacterial defenses (1,2). In the context of its powerful destruc- tive capabilities, NE is considered to be the major proteolytic enzyme causing the progressive destruction of the alveolar walls that results in the emphysema associated with a,-anti- trypsin deficiency and cigarette smoking (7).

Although NE is carried and released by neutrophils, it is not clear that neutrophils can synthesize this enzyme. Fur- thermore, despite the specificity of the name “neutrophil” elastase, there is evidence that bone marrow-derived cells other than those of the myelocytic lineage are capable of expressing the NE gene. For example, NE has been recovered from the cells of the mononuclear phagocyte lineage, including blood monocytes and alveolar macrophages as well as U937, a cell line with many characteristics of the mononuclear phagocyte lineage (8-10). From this evidence, and with the knowledge that neutrophils and mononuclear phagocytes are derived from a common bone marrow precursor (Io-lz), it is likely that the NE gene is expressed in the precursor marrow cells committed to the myelomonocytic lineage of cells and that the cells of this lineage that are released into the blood have lost, or are in the process of losing, this capability.

TO help to define what cells are capable of expressing the

2543

2544 Neutrophil Elastase Gene Expression

NE gene, we have isolated a human NE cDNA from cells known to produce NE and, using this cDNA as a probe, evaluated a variety of cells in which functional NE has been observed for the presence of NE mRNA transcripts. Interest- ingly, the data demonstrate that neutrophils, monocytes, and alveolar macrophages do not appear capable of expressing the NE gene, but suggests the NE gene is expressed by myelo- monocytic lineage precursor cells, particularly those differ- entiated toward the myelocytic lineage.

MATERIALS AND METHODS

Isolation of a Human Neutrophil Elastase cDNA Probe-A cDNA complementary to NE was isolated from a X g t l O cDNA library prepared from the human cell line U937 (Clonetech Laboratory), a cell known to produce NE (8), using standard techniques. The NE cDNA was identified using a 5’-[32P]ATP-labeled 36-mer unique oligonucleotide probe (GGCAATGGCGTGCAGTGCCTGGCCAT- GGGCTGGGGC) designed to correspond to amino acids 117 to 128 of human NE (3). Sequencing of the cDNA insert was carried out by the dideoxynucleotide chain termination method with synthetic bi- directional oligonucleotide primers.

Cells and Cell Culture-Neutrophils, monocytes, and T-lympho- cytes were purified from blood of normal individuals using conven- tional methods (13). The neutrophil populations were >95% pure, the monocyte populations were >90% pure, and the T-cells were >95% pure. Bone marrow cells were obtained from normal volunteers. After dilution with an equal volume of Dulbecco’s modified Eagle’s medium (DMEM), the population of marrow precursor cells were enriched using Ficoll-Hypaque density centrifugation (14); the gra- dient interface had enrichment of marrow precursors (28 f 4% of total cells), while the pellet contained relatively few marrow precur- sors (4 f 2% of total cells). Alveolar macrophages (>go% pure) were obtained by bronchoalveolar lavage of normal nonsmoking volun- teers.

Each of the purified cell populations was activated in culture using standard stimuli appropriate for each cell. Neutrophils were activated (2 X lo6 cells/ml) in RPMI 1640 containing 10% heat-inactivated fetal bovine serum (FBS) with phorbol myristate acetate (PMA, 10 ng/ml) or N-formyl-methionyl-leucyl-phenylalanine M) plus cytochalasin B (5 pg/ml) for 12 h at 37 ‘C. Monocytes were activated (lo6 cells/ml) in DMEM/10% FBS for 24 h at 37 “C with lipopoly- saccharide (LPS) (10 yg/ml E. coli serotype 0127 b8), PMA (50 ng/ ml), calcium ionophore (A-23187, 1 y ~ ) and 8-bromoadenosine 3’:5’- monophosphate (1 FM). T-cells were activated (lo6 cells/ml) in RPMI 1640/10% FBS for 24 h at 37 “C with phytohemagglutinin (5 pg/ml) plus PMA (2 ng/ml). Alveolar macrophages were activated (lo6 cells/ ml) in DMEM/10% FBS for 24 h at 37 “C with LPS (10 yg/ml).

The U937 cell line (American Type Culture Collection (ATCC); CRL1593) was maintained in RPMI 1640/10% FBS, 50 units/ml penicillin, and 50 pg/ml streptomycin. The HL-60 human myelomon- ocytic cell line (ATCC CCL240) was maintained in DMEM/lO% FBS, (50 units/ml penicillin and 50 pg/ml streptomycin). The HL- 60 cell line can differentiate to either the granulocyte or monocyte lineages, depending on the nature of the inducer (10). To evaluate the NE mRNA expression relevant to such differentiation, HL-60 cells were induced to differentiate toward the granulocyte lineage by using dimethyl sulfoxide (2 X lo5 cells/ml DMEM/10% FBS alone, or with 1.25% (v/v) dimethyl sulfoxide for 1 to 5 days) or toward the monocyte lineage using PMA (lo6 cells/ml in serum-free DMEM, 10 mM HEPES, pH 7.4, alone or with 100 ng/ml PMA for 1 to 3 days).

As additional controls, diploid human fetal lung fibroblasts (HFL1, ATCC CCL153), the T-cell line MOLT3 (ATCC CRL1552), and the B-cell line Raji (ATCC CCL86) were cultured using standard meth- ods. Finally, samples of human liver were also used as a source of RNA.

Northern and Dot Blot Analysis-The presence of NE mRNA transcripts was evaluated by Northern analysis (12) using 32P-labeled NE cDNA clone pPB15. As a control, y-actin mRNA transcripts were identified using a y-actin cDNA (plasmid pHFyA-1; P. Gunning and L. Kedes, Stanford University). As further controls, RNA sam- ples from the LPS-activated monocytes and alveolar macrophages were evaluated with a 0.8-kb tumor necrosis factor cDNA probe (pTNFtrp, Genetch) and T-cell RNA samples were evaluated using an interleukin-2 (IL-2) cDNA probe (pTG26, Transgene) and an IL- 2 receptor cDNA probe (pIL2R2, T. Waldman, National Cancer Institute). All probes were labeled by nick translation.

Quantification of neutrophil elastase mRNA in the cellular RNA of HL-60 cells was carried out by dot blot analysis. As a control, interleukin-10 (IL-113) mRNA transcripts were evaluated in parallel using a 32P-labeled IL-10 cDNA probe (M. Tocci, Merck Sharp and Dohme). As arbitrary standards for NE and IL-1s mRNA levels, 2.5 yg of total HL-60 RNA was defined as containing 1 unit of NE mRNA, and 2.5 yg of total U937 RNA as containing 1 unit of IL-1 mRNA.

RESULTS

Neutrophil Elastase cDNA-Characterization of pPB15 re- vealed a 0.65-kb insert that contained a sequence complemen- tary to residues Val46 through the C-terminal Gin'" of the known NE sequence plus an additional 132-base pairs 3’ sequence (Fig. 1). For the region complementary to residues Val46 through Gln2I8 of the mature protein, the protein se- quence derived from the cDNA was identical to the protein sequence of human NE determined by Sinha et al. (3). In this regard, it is reasonable to conclude that this region of cDNA pPB15 codes for the 173 C-terminal residues of the total 218 residues of the mature NE protein.

Interestingly, the cDNA sequence 3’ to the CAA coding for the C-terminal Gln218 of the mature protein also included 60 additional bases prior to reaching the stop codon TGA. This 3’ sequence codes for a 20-amino acid peptide; presumably this represents a “pro-peptide’’ precursor sequence of NE that is translated but processed away from the mature NE poly- peptide before (or as) it is stored in the azurophilic granules of the neutrophil. The sequence of the cDNA 57 base pairs beyond the stop codon revealed the sequence AATAAA that is typical for a polyadenylation signal.

Expression of the Neutrophil Elastase Gene-Using the NE cDNA as a probe, Northern analysis of RNA extracted from U937 cells revealed the presence of a single NE mRNA transcript of 1.3 kb (Fig. 2, lane 1 ). Similarly, NE mRNA transcripts were observed in unfractionated bone marrow (lane 2 ) . When the bone marrow cells were partially fraction- ated, the fraction enriched for marrow precursors had sever- alfold greater amounts of the 1.3-kb NE mRNA transcripts than the fraction enriched for mature polymorphonuclear leukocytes (lanes 3 and 4 ) . In contrast, mature neutrophils purified from blood did not contain detectable NE transcripts (lane 5 ) , although they did contain y-actin transcripts (lane 17). This was true even when %fold more RNA was electro- phoresed on the gels, and the resulting autoradiograms were exposed for 14 days instead of the usual 3 days (lanes 6 and 18). Furthermore, when the neutrophils were stimulated with PMA or with N-formyl-Met-Leu-Phe plus cytochalasin B, no NE mRNA transcripts were observed (not shown). From these observations, it is reasonable to conclude that blood neutro- phils do not have the ability to synthesize this protease. In contrast, cells in the bone marrow contain NE mRNA and thus are likely the cellular site a t which this protein is pro- duced.

Like blood neutrophils, blood monocytes, whether resting or activated with LPS, did not contain NE transcripts (lanes 7 and 8) although y-actin transcripts were observed (lanes 19 and 20). Furthermore, when monocytes were stimulated with a protein kinase C activator (PMA), a calcium ionophore (A- 23187), or an agent that mimicked increased intracellular CAMP levels (8-bromoadenosine 3’:5’-monophosphate), the results were similar, i.e. no NE mRNA transcripts were de- tectable, although y-actin transcripts were present (not shown). Likewise, neither resting alveolar macrophages nor activated alveolar macrophages contained NE transcripts (lanes 9 and 10) although y-actin mRNA transcripts were evident (lanes 21 and 22). Furthermore, in the activated

Neutrophil Elastase Gene Expression

1 5 10 15 20 25

Protein Ile Val Gly Gly Arg Arg Ala Arg Pro His Ala Trp Pro Phe Met Val Ser Leu Gln Leu Arg Gly Gly His Phe

2545

26 30 35 40 His‘ Glv’ Cys Gly Ala Thr Leu Ile Ala Pro Asn Phe Val Met Ser Ala Ala His Cys Val Ala Asn Val Asn Val Arg Ala

45 r - - - - ASP’ :“-J

I 50

Deduced protetn Val Asn Val Arq Ala cDNA GTA AAC GTC CGC GCG

51 75

Val Arg Val Val Leu Gly Aia His Asn Leu Ser Arg Arq Glu Pro Thr Arg Gln Val Phe Ala Val Gln Arg Ile

GTG CGG GTG GTC CTG GGA GCC CAT AAC CTC TCG CGG CGG GAG CCC ACC CGG GAG GTG TTC GCC GTG CAG CGC ATC Val Arg Val Val Leu Gly Ala His Asn Leu Ser Arg Arg Glu Pro Thr Arg Gln Val Phe Ala Val Gln Arg Ile

Glu’ 55 60 65

Thrb Pro’ Thr’ 70 Leu’

76 80 85 90 95 100

Phe Glu Asp Gly Tyr Asp Pro Val Asn Leu Leu Asn Asp Ile Val Ile Leu Gln Leu Asn Gly Ser Ala Thr Ile

TTC GAA C GC TAC GAC CCC GTA AAC TTG CTC AAC GAC ATC GTG ATT CTC CAG CTC AAC GGG TCG GCC ACC ATC Phe Glu Asn Gly Tyr Asp Pro Val Asn Leu Leu Asn Asp lie Val Ile Leu Gln Leu Asn Gly Ser Ala Thr Ile

101 105 110 115 120 125

Asn Ala Asn Val Gln Val Ala Gln Leu PIO Ala Gln Gly Arg Arg Leu Gly Asn Gly Val Gln Cys Leu Ala Met

AACGCCAACGTGCAGGTGGCCCAGCTGCCG GCTCAGGGACGCCGCCTGGGCAACGGGGTGCAGTGCCTGGCCATG Asn Ala Asn Val Gln Val Ala Gln Leu Pro Ala Gln Gly Arg Arg Leu Gly Asn Gly Val Gln Cys Leu Ala Met

El. *

126 130 135 140 145 150

Gly Trp Gly Leu Leu Gly Arg Asn Arg Gly Ile Ala Ser Val Leu Gln Glu Leu Asn Val Thr Val Val Thr Ser

GGCTGGGGCCTT CTGGGCAGGAAC CGTGGGATC GCCAGCGTC CTG CAGGAG CTC AAC GTGACGGTGGTGACG TCC Gly Trp Gly Leu Leu Gly Arg Asn Arg Gly Ile Ala Ser Val Leu Gln Glu Leu Asn Val Thr Val Val Thr Ser

*

151 155 160 165 170 175

Leu Cys Arg Arg Ser Asn Val Cys Thr Leu Val Arg Gly Arq Gln Ala Gly Val Cys Phe Gly Asp Sei Gly Ser

CTC TGC CGTCGCAGCAAC GTCTGCACT CTC GTGAGGGGCCGGCAGGCCGGCGTC TGT TTCGGGGACTCAGGCAGC Leu Cys Arg Arg Ser Asn Val Cys Thr Leu Val Arg Gly Arg Gln Ala Gly Val Cys Phe Gly Asp Ser Gly Ser

176 180 185 190 195 200

Pro Leu Val Cys Asn Gly Leu Ile His Gly Ile Ala Ser Phe Val Arg Gly Gly Cys Ala Ser Gly Leu Tyr Pro

CCC TTG GTC TGC AAC GGG CTA ATC CAC GGA ATT GCC TCC TTC GTC CGG GGA GGC TGC GCC TCA GGG CTC TAC CCC Pro Leu Val Cys Asn Gly Leu Ile HIS Gly Ile Ala Ser Phe Val Arg Gly Gly Cys Ala Ser Gly Leu Tyr Pro

201 205 210 215 218 220 225

Asp Ala Phe Ala Pro Val Ala Gln Phe Val Asn Trp Ile Asp Ser Ile Ile Gln

GAT GCC TTT GCC CCG GTG GCA CAG TTT GTA AAC TGG ATC GAG TCT ATC ATC CAA CGC TCC GAG GAC AAC CCC TGT Asp Ala Phe Ala Pro Val Ala Gln Phe Val Asn Trp Ile Asp Ser Ile Ile Gln Arg Ser Glu Asp Asn Pro Cys

226 230 235 230

CCC CAC C C C C G G G A C C C G G A C C C G G C C A G C A G G A C C CAC TGAGAAGGGCTG CCCGGGTCA CCTCAG CTG CCC ACA Pro HIS Pro Arg Asp Pro Asp Pro Ala Ser Arq Thr His

s lop

CCC ACA CTC TCC AGC ATC TGG CAC AAT A M CAT TCT

FIG. 1. Nucleotide sequence of the human neutrophil elastase cDNA clone (pPB16) compared to the known protein sequence of NE. The complete amino acid sequence of NE described by Sinha et al. (3) is shown on the first line; a few residues of that sequence differ from previously reported partial protein sequences; such differences are indicated above (a, partial sequence of Heck et al. (6); b, partial sequence of Travis et al. (5)). Numbers above the lines indicate amino acid position starting with the N-terminal Ile’ of the mature protein. The putative N-linked glycosylation sites a t Ame5 and Asn“‘ are indicated by asterisks. The nucleotide sequence of the cDNA clone and the amino acid sequence deduced from the cDNA sequence are shown on the second and third lines, respectively. The 0.65-kb cDNA clone insert encompasses the protein coding region from V a P through the termination codon (TGA “stop”) and a 3”untranslated region. A putative polyadenylation signal sequence (ATT AAA) is underlined. The amino acid sequence from Val46 to Gln”’ deduced from the cDNA sequence is identical to the protein sequence presented in Fig. 1 of the report by Sinha et al. (3) except for one amino acid residue (Asn” in the deduced sequence) indicated by the box; the AspT8 in Sinha et al. (3) is a misprint and is actually Asn” (J. Travis, personal communication) as predicted by the cDNA sequence. The dashed box (residues 47 and 48) are amino acids found by Sinha et al. (3) and predicted by the cDNA sequence, but are missing from the partial amino sequence of Travis et al. (5).

2546 Neutrophil Elastase Gene Expression

A. Nwtrophil elrstrcm cDNA probr IY I-

lOm1.d N""., .,"..,., UIhll.d

W) I,., I-, h,* *"im#n,,' *t",m,",m *O"rn,, rnwml' n.lrwh.I.I '.".P*.,.' l-ralls 1-1111 L"l"**ld

2s-

189- 1.3kb' 0 a -

1 2 3 4 5 6 7 8 9 1 0 1 1 1 2

B. y-octin cDNA probe

-.Pi.

I.

2.0 kb+ q 20s-

10s- q 0 b r . P -

13 14 15 16 17 18 19 20 21 22 23 24

FIG. 2. Presence of neutrophil elastase mRNA transcripts in various cells. Shown are Northern analyses of total cellular RNA (10 pg/lane and autoradiograms were exposed for 3 days unless otherwise indicated) from various sources using the "P-labeled NE cDNA probe. Lane 1, U937; lane 2, total human bone marrow; lane 3, cells from the interface fraction of Ficoll-Hypaque density centrifu- gation of bone marrow; lane 4, bone marrow cells in the pellet of the Ficoll-Hypaque gradient; lane 5 , blood neutrophils; lane 6, identical to lane 5 , but with 20 pg of RNA and the autoradiograms were exposed for 14 days; lane 7, resting blood monocytes; lune 8, LPS-activated monocytes; lane 9, resting alveolar macrophages; lane 10, LPS-acti- vated alveolar macrophages; lane 11, resting blood T-cells; lane 12, phytohemagglutinin + PMA-activated T-cells. As a control to dem- onstrate that all cells evaluated contained mRNA transcripts, lanes 13-24 are identical to lanes 1-12, respectively, but were probed with a "P-labeled y-actin cDNA. The size of the NE and y-actin mRNA transcripts and position of 18 S and 28 S were determined using conventional markers.

A. Neutrophil elastase cDNA probe B. Interleukin-lp cDNA probe

3 days cullure 3 days culture

Fresh Alone +PYI +omso Fresh Alone +PMA +OMSO

28s - 285 - 185-

0 -1.3 kb el.8 kb

1 2 3 4 5 6 7 8

FIG. 3. Expression of neutrophil elastase mRNA transcripts in the HL-60 cell line. A , Northern analysis of total cellular RNA (10 pgllane) from HL-60 cells under various conditions evaluated with the "'P-labeled NE cDNA probe. Lane 1, fresh HL-60 cells; lane 2, after culture in DMEM/10% FBS for 3 days; lane 3, as lane 2 but cultured in the presence of PMA; lane 4, as lane 2 but cultured in the presence of dimethyl sulfoxide. B, similar to lanes 1-4 but evaluated with a '"P-labeled interleukin-l@ cDNA probe. The position of the NE and 11-1p mRNA transcripts and the 18 S and 28 S RNA were determined using conventional markers. DMSO, dimethyl sulfoxide.

monocyte and activated alveolar macrophage RNA samples, mRNA transcripts typical of activated mononuclear phago- cytes were observed, including 1.8-kb interleukin-18 tran- scripts, and 1.9-kb tumor necrosis factor transcripts (not shown). Like the other blood cells, T-cells and activated T - cells did not contain NE mRNA transcripts (lunes 11 and 12), although both contained y-actin transcripts (lunes 23 and 24) and activated T-cells contained the 0.85-kb IL-2 transcript

A. Neutrophil elastase cDNA probe

Alone Fresh 3 davs

+ PMA

+ PMA t - - I I

8. Interleukin-le cDNA probe -

Alone +PMA +DMSO Fresh 3days 5days 1

1 c 0

2 - -

+DMSO " ? , e " " _ _ .

0 1 2 3 4 5

Duration of culture (days)

FIG. 4. Quantification of neutrophil elastase mRNA tran- scripts in resting and differentiated HL-60 cells. A , dot blot analysis of NE mRNA transcripts in HL-60 cells cultured in DMEM/ 10% FBS for 0 to 5 days ("alone"), in the presence of dimethyl sulfoxide (DMSO), or in the presence of PMA. The cells begin to lose viability when cultured in PMA for more than 3 days thus limiting the +PMA analysis to 3 days. Inset, example of the dot blot analysis; the initial dot contained 2.5 pg of RNA and the dilutions in consecutive dots were 1:2. B, dot blot analysis of RNA from cells in A, but evaluated for the presence of IL-16, mRNA transcripts. For both panels, the data is presented as arbitrary mRNA units/lOfi cells a t each time point. Error estimated represents mean rt standard error of the mean of three different experiments.

and the 1.5- and 3.5-kb IL-2 receptor transcripts (not shown). Likewise, human liver, diploid human fetal lung fibroblasts, the T-cell line MOLT3, or the B-cell line Raji did not contain NE transcripts, but did contain y-actin transcripts (not shown).

Neutrophil Elastase mRNA Expression in Myelomonocytic Precursors-When evaluated fresh, HL-60 cells contained the 1.3-kb NE transcripts (Fig. 3, lune 1 ). However, when stimu- lated with PMA to develop monocytic lineage characteristics, the NE transcripts disappeared (lane 3), even though mRNA transcripts for IL-18 that are typical for activated monocytic cells were easily detected (lune 7). In contrast, when stimu- lated with dimethyl sulfoxide to develop myelocytic lineage characteristics, the HL-60 cell line clearly contained NE transcripts (lune 4) but no IL-18 transcripts (lane 8).

Quantitative evaluation of the HL-60 cells using cytoblot analysis revealed major differences in the content of NE and IL-18 mRNA transcripts under resting, PMA-stimulated, and dimethyl sulfoxide-stimulated conditions (Fig. 4). When stim- ulated with dimethyl sulfoxide, the HL-60 cells increased their content of NE transcripts but did not contain IL-18 tran- scripts while PMA had the opposite effect. Together, these observations are consistent with the concept that the NE gene can be expressed by bone marrow precursors, with emphasis toward the myelocytic, and not monocytic cell lineages.

DISCUSSION

Neutrophil elastase is a potent protease capable of attacking most proteins to which it is exposed (1-7). In this regard, NE

Neutrophil Elastase Gene Expression 2547

is considered both a valuable and dangerous enzyme: valuable in its role in normal tissue turnover, inflammatory reactions, and antibacterial defenses, but dangerous in its ability to destroy fragile tissues (2, 7 ) . Although large quantities are stored in the neutrophils (1, 2, 7, 11, 12) and smaller amounts in monocytes (8), the present study demonstrates that these cells do not contain detectable NE mRNA transcripts when resting or activated, making it unlikely that either cell is capable of synthesizing this enzyme. In contrast, bone marrow cells, particularly those cells enriched for precursor cells, clearly contained NE mRNA transcripts. Furthermore, HL- 60 cells, a leukemic cell line with immature myeloid cell characteristics, contained NE transcripts and increased the amounts of these transcripts when induced toward the mye- locytic lineage. Together, these observations suggest that the NE gene is normally expressed in the bone marrow by mye- lomonocytic lineage precursors, particularly those that have differentiated toward the myelocytic lineage, but at the time these cells leave the bone marrow, they have lost the potential to synthesize this enzyme. Together with the evidence that NE is present in promyelocytes but not myeloblasts ( E ) , it is likely that the NE gene is expressed transiently during mye- loid cell differentiation, beginning with promyelocyte to mye- locyte differentiation and ending as metamyelocytes differ- entiate into mature neutrophils.

Despite the fact that neutrophils are capable of synthesizing RNA and protein (15), have detectable mRNA transcripts for secretory proteins such as fibronectin (16), and can be induced to synthesize heat shock protein (17), the observation that neutrophils contain no detectable NE mRNA transcripts is consistent with the general view that neutrophils function essentially as cellular “mules,” terminally differentiatied cells that carry stored proteins that they cannot produce. From the present evidence it is not clear where in myeloid lineage differentiation the NE gene is shut down, but it may occur at a point different from other prominent components of the azurophilic granules of neutrophils. In this context, while the number of NE mRNA transcripts increases when HL-60 cells are stimulated to differentiate toward the myelocytic series by dimethyl sulfoxide, there are reduced levels of myeloper- oxidase mRNA under similar conditions (18).

The observations that neither blood monocytes nor alveolar macrophages appear able to express the NE gene and that monomyeloid lineage precursors apparently lose the capacity to express this gene as they differentiate toward the monocytic lineage argues that, while mononuclear phagocytes may con- tain and release small amounts of NE, these cells are likely not capable of synthesizing this enzyme. In this regard, since mononuclear phagocytes carry far less NE per cell than do neutrophils (8) and are very long-lived cells (13), these obser- vations suggest that the contribution of mononuclear phago- cytes to the total NE burden of tissue over time must be very small. In contrast, while neutrophils also cannot synthesize NE de nouo, the amount of NE per neutrophil is large and neutrophils are very short-lived cells, i.e. as a cellular mule

for NE, the mononuclear phagocyte is clearly outclassed by the neutrophil.

Knowledge of the potential for cell expression of NE is relevant to the concepts of the role of NE in the pathogenesis of emphysema and in defending against bacterial infection. First, emphysema is conceptualized to result from the chronic burden of neutrophils (and hence NE) that overwhelms the anti-NE screen of the lower respiratory tract, allowing the NE to destroy the alveolar walls ( 2 , 7 ) . If, however, monocytes and their progeny alveolar macrophages had the capacity to produce NE (8, 9), strategies to protect the lung from NE would have to be expanded to include the concept of dealing with a burden of NE released by mononuclear phagocytes as well as that released by neutrophils. However, since mono- nuclear phagocytes apparently cannot express the NE gene, it is unlikely that mononuclear phagocytes contribute signif- icantly to the burden of NE in the lower respiratory tract. Second, the fact that there is little NE in mononuclear phag- ocytes and that mononuclear phagocytes cannot manufacture NE makes it unlikely that NE contributes significantly to the antibacterial defensive weapons of these cells.

Acknowledgments-We would like to thank K. Konishi, S. Tsuji, C. Saltini, Y. Martinet, D. States, and A. Rosolen for their assistance in this study.

REFERENCES 1. Janoff, A., and Scherer, J. (1968) J. Enp. Med. 128, 1137-1155 2. Bieth, J. G. (1986) in Regulation of Matrix Accumulation (Me-

cham, R., ed) pp. 217-320, Academic Press, New York 3. Sinha, S., Watorek, W., Karr, S., Giles, J., Bode, W., and Travis,

J. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 2228-2232 4. Bode, W., Wei, A. Z., Huber, R., Meyer, E., Travis, J., and

Neumann, S. (1986) EMBO J. 5, 2453-2458 5. Travis, J., Giles, P. J., Porcelli, L., Reilly, C.F., Baugh, R., and

Powers, J. (1980) CZBA Found. Symp. 75, 51-68 6. Heck, L. W., Darby, W. L., Hunter, F. A., Bhown, A,, Miller, E.

J., and Bennett, J. C. (1985) Anal. Biochem. 149, 153-162 7. Janoff, A. (1985) Annu. Reu. Med. 36,207-216 8. Senior, R. M., Campbell, E. J., Landis, J . A,, Cox, F. R., Kuhn,

C., and Koren, H. S. (1982) J. Clin. Znuest. 69, 384-393 9. Rodrigeuz, R. J., White, R. R., Senior, R. M., and Levine, E. A.

(1978) Science 198, 313-314 10. Harris, P., and Ralph, P. (1985) J. Leukocyte Biol. 37,407-422 1. Bainton, D. F., Ull;ot, J . L., and Farquha;, M. G. (1971) J. Enp.

Med. 134,907-934 2. Kramps, J. A., Van Der Valk, P., Van Der Sandt, M. M., Linde-

man, J., and Meijer, C. J . L. M. (1984) J. Histochem. Cytochem.

3. Mornex, J. F., Chytil-Weir, A., Martinet, Y., Courtney, M., LeCocq, J. P., and Crystal, R. G. (1986) J. Clin. Inuest. 77, 1952-1961

32,389-394

4. Bovum. A. (1976) S c a d . J. Immunol. 5. 9-15

1

1

1

1 - . , 15. Granelli-Piperno, A., Vassalli, J. D., and Reich, E. (1979) J. Enp.

16. Fleur, M. L., Beaulieu, A. D., Kreis, C., and Poubelle, P. (1987)

17. Eid, N. S., Kravath, R. E., and Lanks, K. W. (1987) J. Exp. Med.

18. Yamada, M., and Kurahashi, K. (1984) J. Biol. Chem. 259,3021-

Med. 149,284-289

J. Biol. Chem. 262, 2111-2115

165, 1448-1452

3025