Transcriptional regulation of human stromelysin.

THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1989 by The American Society for Biochemistry and Molecular Biology, Inc

Val. 264, No. 14,Issue of May 15, pp. 8339-8344,1989 Printed in U S A .

Transcriptional Regulation of Human Stromelysin” (Received for publication, September 26, 1988)

Susan Quinones, Juan Sauss, Yoshihide Otani, Edward D. Harris, Jr.8, and Markku Kurkinen From the DeDartrnent of Medicine. Uniuersitv ” of. Medicine and Dentistry of New Jersey-Robert Wood Johnson Medical School, Piscataway, NewJersey 08854

’

We have determined that human stromelysin mRNA can be induced by interleukin-18 (IL-18) and that the induced mRNA levels can be suppressed by retinoic acid and dexamethasone (Saus, J., Quinones, S., Otani, Y., Nagase, H., Harris, E. D., Jr., and Kurkinen, M. (1988) J. Biol. Chem. 263,6742-6745). Here we show, by nuclear run-on and transient gene expression analyses, that IL-18 induction is a promoter function and that dexamethasone suppresses 1L-18-induced gene activity. For transient gene expression assays, 1.3 kilobase pairs of the stromelysin promoter region (-1303 to -11 relative to the transcription start site) and shorter fragments thereof were cloned into a human growth hormone reporter vector. In transfected human fibroblast cultures all the constructs, with the exception of the one containing the shortest promoter fragment (-53 to - l l ) , responded to lL-1B induction. Interestingly,the ability of IL-18 to induce human growth hormone expression decreased as the length of the promoter fragment was reduced. Dexamethasone treatment suppressed the induced human growth hormone levels by approximately 50%irrespective of the promoter length. These results suggest that the 1.3kilobase pairs stromelysin promoter fragment contains DNA elements required for IL- 18 induction and dexamethasone suppression.

specificity of all, can degrade fibronectin, laminin, collagen IV, and cartilage proteoglycans (11). These three enzymes exhibitcomplementary,butlittle or nocommoncatalytic function. However, comparison of cDNA-derived amino acid sequencesfor these proteinases indicate that they share a high degree of homology in primary and secondary structure (12-14). Precisely how the productionof these metalloproteinasesis regulated is important in understanding the physiologic and pathologic processes involvingextracellular matrix. In human and rabblt fibroblast cultures, collagenase and stromelysin production is coordinately regulated in response t o several stimuli. Enzyme synthesis isinduced by a number of factors that include interleukin-1P and the tumor promoter TPA’ (12-0-tetradecanoylphorbol-13-acetate).Induced levels can be subsequently suppressed by retinoic acid or dexamethasone (14-17). Transin (18), the rathomologue of stromelysin, is induced in fibroblasts transformedwith polyoma virus, Rous sarcoma virus, or oncogene H-ras and in fibroblastsexposed to epidermal growth factor. The induction of transin transcription by epidermal growth factor canbe blocked or suppressed by prior or subsequent exposure of the cells to transforming growth factor p (19). Recently, we haveisolatedcDNAclones that cover the complete coding region for human stromelysin and have used them to establish that IL-lp, retinoic acid, and dexamethasone modulate the accumulationof stromelysin mRNA (14). Maintenance of the equilibriumbetween deposition and We havesince isolated genomic clones for human stromelysin. degradation of extracellularmatrixisessentialtonormal In this report we demonstrate by transient gene expression tissuedevelopmentandrepair of wound or inflammatory assay that1.3 kb of the stromelysin promoterregion contains damage. In thepathological states of rheumatoid arthritis and DNA elementsresponsible for IL-lP induction and dexamethtumor invasion, degradation is disproportionately increased asone suppression. Induction by IL-lP is decreasedfrom and the equilibrium destroyed (1-3). Matrix degradation re- approximately 14.4-fold with the 1.3-kb promoter fragment quires the existence of specific metalloproteinases. To date to approximately 7.6- and 6.5-fold with the 0.74- and 0.47-kb threemetalloproteinases, collagenase,gelatinase, and strofragments, respectively, thereby suggesting that the stromemelysin, have been isolated that exhibit matrix degrading lysin promoter contains multiple IL-lp-responsive elements. activity (4). Collagenase cleaves collagen types I, 11, and 111 Withallthreepromoterconstructs,dexamethasonesupat a single locus and collagen X a t two (4, 5). Gelatinase has pressed the inductive capability of IL-lp by approximately activityagainstdenatured collagen (gelatin)and collagen 50%. types IV and V (6-10). Stromelysin, with thewidest substrate MATERIALSANDMETHODS

* This work was supported by Grants AR 33714 and GM 34090

Chemicals and Enzymes- Klenow enzyme, restriction endonuclefrom the National Institutes of Health. The costs of publication of ases, and calf thymus DNA were from New England Biolabs (Beverly, this article were defrayed in part by the payment of page charges. MA) and Boehringer Mannheim (Indianapolis, IN). Bluescribe and This article must therefore be hereby marked “eduertisernent” in Bluescript vectors andTSand T, polymerases were from Stratagene accordance with 18 U.S.C. Section 1734 solely to indicate this fact. (La Jolla, Ca). MI3 vector and linkerswere obtained from Pharmacia T h e nucleotide sequence($ reportedin this paper has been submitted LKB Biotechnology, Inc. (Piscataway, NJ), and the human growth to theGenBankTM/EMBt DataBank with accession numberrs) hormone vectors came from Nichols Institute (San Juan Capistrano, J04 732. $ Profesor Titular en Comision de Servicio of the University of Valencia, Spain. Present address: Dept. of Biochemistry, Faculty of ’ The abbreviations used are: TPA, 12-O-tetradecanoylphorbol-13Pharmacy, University of Valencia, Valencia 46010, Spain. acetate; IL-lp, interleukin-lp;kb, kilobase pair; bp. base pair; hGH, § Present address: Dept. of Medicine, Stanford University School human growth hormone; SDS, sodium dodecyl sulfate; Hepes, N-2of Medicine, Stanford, CA 94305. hydroxyethylpiperazine-N’-2-ethanesulfonic acid.

8339

Regulation of Human Stromelysin Gene

8340

C.4). [a-”PIUTP (800 Ci/mmol),[35S]dATP (1200 Ci/mmol), and [y3’P]ATP (6000 Ci/mmol) were from Du Pont-New England Nuclear (Wilmington, DE). Human recombinant IL-10 was obtained from either Cistron (Pine Brook, NJ) or Boehringer Mannheim. Isolation of Stromelysin Genomic Clones-A human stromelysin cDNA (14) was cloned into Bluescript KS (+)-vector and used to make RNA probes to screen a human “X fix” genomic library (Stratagene). The plasmid DNA was restricted at the3’-end of the coding region so that only 3ZP-labeledtranscripts originating at the 5’-end were synthesized. The filters were hybridized in 5 X SSPE (1X SSPE, 150 mM NaCl, 10 mM NaH2P04, 1 mM EDTA), 0.1% SDS at 68 “C overnight and then washed in 0.1 X SSC (1X SSC, 150 mM NaC1,15 mM Na-citrate), 0.1% SDS at 75 “C. Of the approximately 5 X lo5 plaquesscreened, 10 signals were recovered and 5 of those were purified. One clone (HAF10) that was characterized by restriction mapping and partial nucleotide sequencing is 14 kb in length and contains approximately 10 kb of sequence upstream of the transcription startsite. Nucleotide Sequence Analysis-Restriction enzyme fragments of HAFlO were cloned into M13 vectors (20) and sequenced according to the dideoxynucleotide chain termination method (21) using [35S] dATP (22), modified T7DNA polymerase (23), and the “sequenase” kit from United States Biochemicals Co. (Cleveland, OH). Sequence specific oligonucleotides (17-mers) used as primers were made on an automated oligonucleotide synthesizer (Biosearch, Inc., New Brunswick, NJ). Each strandwas sequenced at least twice. Transcription Start Site-A 40-mer oligonucleotide (24 ng) complementary to stromelysin mRNA sequence (nucleotides 66-105of the protein coding region) was end-labeled by [Y-~’P]ATP and TI polynucleotide kinase and hybridized with total RNA (30 pg) from induced fibroblasts in 50 pl of 80% formamide, 0.4 M NaCl, 10 mM Hepes, pH 7.5, at 30 “C overnight. After ethanol precipitation, the reverse transcriptase reaction was carried out as described (24, 25). For accurate determination of the transcription start site, the primer-

CONTROL

IL-1 STROMELYSIN

r*

fl-ACTIN FIG. 1. Transcriptionalactivation of stromelysinafter stimulation withIL-la.Nuclei were isolated from cultured human skinfibroblasts after induction with IL-1P (5units/ml) for 3 h. Initiated transcripts were labeled as described under “Materials and Methods” and hybridized to stromelysin and &actin cDNAs (5 pg of plasmid DNA/slot).

FIG. 2. Partial restriction map of HAFlO a n dp r o m o t e fr r a g m e n t s used in the transient gene expression constructs. At the top of the figure is a schematicof the X genomic clone HAF10. The positions of the SstI ( S ) and XbaI ( X ) sites are drawn to scale. Directly below is an enlargement of the SstI partial XbaI 2.1-kb fragment containing 1.3 kb of promoter sequence extending through the first exon (box) and into the second (half-box). The restriction sites used to obtain the transient gene expression constructs are noted. In the constructs the promoterelements used to drive the hGH gene were derived from the SstI-AuaI (1293 nucleotides; -1303 to -ll), Asp718-AuaI (744; -754 to -ll), X ~ U I - A U (468; U I -478 t o - l l ) , and Sau3A-AuaI (43; -53 to -11) fragments.

extended productwas run on asequencing gel along with the sequencing reactions of the 1179-bp XbaI-XbaI fragment containingthe first exon using the same 40-mer as primer. Plasmid Construction-A 2.1-kb fragment of HAF10, containing 1.3 kb of promoter sequence and extending into the second exon of stromelysin, was isolated by SstI and partialXbaI digestion and cloned into Bluescribe. This plasmid was digested with AuaI which cuts at -11 respective to the transcription start site. The ends were filled in with Klenow enzyme and BamHI linkers were attached to replace the AuaI site. Subsequently, using HindIII linkers, the SstI site of this plasmid was converted to a HindIII site. The resultant plasmid was designated as B8.218. To obtain the 1293-bp promoter construct (-1303 to -ll), the HindIII-BamHI fragment of B8.218 was cloned into the human growth hormone reporter vector p$GH (26). To make the 744-bp construct, the 1293-bp construct was digested with HindIII and Asp718 to remove 560 nucleotides on the 5’-end of the promoter, followed by blunt-ending and religation. The 468-bp construct was obtained by cloning the corresponding XbaIBamHI fragment of B8.218 into p$GH. A minimal construct containing only 43 nucleotides of promoter sequence (-53 to -11) including the TATA box was obtained by cloning the corresponding Sau3ABamHI fragment of B8.218 into BamHI-digested p$GH. Plasmid constructs containing the correct orientation of this promoter fragment were identified by sequencing. Transfections and Gene Expression Assays-Human fibroblasts prepared from explant of infant foreskin by standard techniques were cultured and maintained as described (14). For the transfection experiments, cells between 5 and 15 passages were seeded at a density of 7 X lo5 cells/lOO-mm dish and maintained in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum and antibiotics. The next day the media were changed and 4 h later the cells were transfected with 40 pg of total DNA (20 pg of reporter and 20 pg of carrier calf thymus) incalcium phosphate (27) for 16 h. At 24 h posttransfection, the cells were treated with IL-1P (5 units/ml) for an additional 48 h and the secreted hGH levels were then measured using asolid-phase radioimmunoassay kit (Nichols Institute).In other experiments, the cells were induced for 3 h with IL-1P, washed 3 times with Hanks’ balanced salt solution, and the media replaced M dexamethasone.Forty-eighthours along with the addition of later themedia were assayed for hGH. To provide an internalcontrol for DNA uptake, some experiments included co-transfection with the RSVcat vector (28). Chloramphenicol acetyltransferase assays were performed on the cell extracts as described (28). We detected no measurable plate-to-plate or treatment-derived variation in chloramphenicol acetyltransferase activity. Nuclear Run-on Assay-Nuclei were isolated by a modification of the methods as described (29, 30). For each experiment, 2.5 X lo7 nuclei were incubated in a reaction mixture containing 30 mM TrisHCl, pH 8.0, 5 mM MgClz, 150 mM KCl, 2.5 mM dithiothreitol, 100 units/ml RNasin, 1 mM ATP, GTP, and CTP, and 200 pCi [a-”P] UTP for 30 min at 30 “C. After DNase and proteinase K treatment the labeled transcripts were hybridized at a concentration of 7 X lo6

1 kb

S

S I

I

I

X

Sstl

Asp718 Xbal U Sau3A

hQH

-1160

Regulation Stromelysin of Human

Gene

8341

cpm/ml to cDNA bound onto nitrocellulose a filter in 5 X SSPE, RESULTS AND DISCUSSION 0.1% SDS at 68 "C for 48 h. The filters were washed at 68 "C in 0.1 1 ~ - 1 promotes p the accumulation of stromelysin m~~~ in x SSC, 0.1% SDS. cultured human fibroblasts and, conversely, retinoic acid or Promoter Sequence Comparisons-Sequences were compared using a dot matrix program written by T. Burglin, Massachusetts General dexamethasone suppress the induced levels (14). In Order to Hospital, and subsequently aligned using the multiple sequence editordetermine whether the IL-lp modulation of stromelysin gene expression reflects changes in promoter activity, we perwritten by W. Gilbert, Whitaker College.

Sst I

-

GAGCTCTGGGATCMGTGATTCTCCTGCCTCAACCTCTCAAAGTGCTAGGATTACAGACATGGGTCACGGCACCTGGCCTAAAGACATTTTAAMCTAGTATTCTATGGTTCTCCATTCC

- 1300

- 1280

- 1260

- 1240

- 1220

- 1200

TTTGATGGGGGGAAAAMCCATGTCTTGTCCTGATTG~TACAGG~TATTTGGCCACATTGATATGAGGACMGGA~CAGAGTGGTGGCAGTGATGTGMTTCCAGGMGAM

-1180

- 1120

-1140-1100

- 1080

CAGTGGAGCCCMCAGATAAAATATTGAGAGGTTAAAAGTGTAGMGCTGTGGCAGCTCTMTTCCTTCTATATTMGGTTTATCCTTTGTAMTTTTCTCTGGAGTTATMCTTATAM

- 1060

- 1040

- 1000

- 1020

- 960

- 980

ACATATTTAAAGTTTAGAGGAAAGCCTGATAGATAATATGTTTGCTTTTCCMCTGTTMGATGTGATTGGCAAMTGGGCCATGCTGTTCCT~CTAMTACAAMGTTTGCTATATT

- 940

-900

-920

-840 -880

-860

Asp718 TTAAMCAGCTTCGACGACTTCCCTGGAAGATGTTTCCCAAGTCCCAGGMGAGTCATTMTCATTGTTCMT~TTGTACCTGCCTCTGCTACTCTATTTTCCACACTGMCTGT

-820

- 800

- 760

- 780

- 740

- 720

AGTCAMTGGTTTATCTGTCTCTCTCATTCTAATGTMGCTACTTGAGGGTATGGACCATGCCTCATTTGATTCTGTATTCCTAGTACTTAGCATGGTATCTGATATATAGTAGGTMCC

- 700

-680

-640

-660

- 620

-600

XbaI

AAMATTACTTAGAGMCTAAATAATATATTAGTTATGAGTGAMGGCTCATAGCAMTGTCAMCAGGTATTAAMGTC~TCCAAMATCATGCAGACATTTCTAGATATTA

- 580

-560

-540

.

Ava

-520

- 480

-500

I -

Sau3A

-

7

TGGAMTGGTCCTGCTGCCATTTGGATGAAAGCAAGG~~~CTGCGGGT~TCCAMCAMCACTGTCACTCTTTAAMGCTGCGCTCCCGAGGTTGGACCTACMGGAGGCAG

-100

- 80

-40

-60

- 20

1

Met

-

GCMGACAGCMGGCATAGAGACAACATAGAGCTAAGTAMGCCAGTGCAMTG

20

FIG. 3. Nucleotide sequence of the first 1.3 kb of the human stromelysin promoter. The presented sequence extends 5' from the first ATG-codon (Met) through the transcription start site (arrow) to an SstI site (underlined) 1.3 kb further upstream. All the restriction sites utilized for making the transient gene expression constructs are marked. The TATAbox is located between positions -30 and -24. The boxed sequence centering around position -65 shares strong homology (87.5%) with the TPA responsive element (36). Also underlined are three sequences that share stronghomology with the consensussequence for the glucocorticoid-responsive element (33).

Regulation of Human Stromelysin Gene

8342

formed nuclear run-on assays. Nuclei from IL-I/?-treated human fibroblast cultures containedincreased amounts of stromelysin transcripts when compared to nuclei from control cells (Fig. 1).By densitometric scanning, the difference in intensity of these bands is about %fold (2.9 & 1.1).In contrast, IL-10 treatment did not change the amount of radioactivity in P-actin transcripts, suggesting that IL-lP does not have a generalized effect on transcription. Based on these results, the modulated expression of stromelysin can be at least partially attributed to changes in promoter activity. In order to further study the promoter regulation of human stromelysin, we isolated and characterized a 14-kb genomic clone, HAFlO (Fig. 2). A 2.1-kb region, containing 1.3 kb of promoter sequence and extending into the secondexon of stromelysin, was characterized by nucleotide sequencing. In Fig. 3 the nucleotide sequence of the 1.3-kb promoter and the 5”untranslated region is presented. The transcription start site was determined by primer extension of a 40-mer oligonucleotide complementary toa segment of the first exon using total RNA from induced fibroblasts as template (see “Materials and Methods”). Onemajor product was detected aswell asfaintproductsone nucleotide above and one and two nucleotides below (Fig. 4). When similar analysiswas carried out with total RNA from control cells, which make only small amounts of stromelysin (14), the primer-extended product was barely visible (data not shown). For the transient gene expression assays, restriction fragments of stromelysin promoter were cloned into the hGH reporter vector (26). After transfection of human fibroblast cultures, one set of cultures was treated with interleukin-1P. After 48 h (72 h post-transfection)theamount of hGH secreted into the culture media was determined byradioimmunoassay(TableI).Withthe 1293-bp promoterconstruct, IL-l/?induced the accumulationof 14.4-fold more hGH

PE

/ \ T

A*

I

3’

FIG.4. Mapping of the transcription start site of human stromelysin. Single lane at left ( P E ) is the product of the primer extension of a 40-mer oligonucleotide using induced total RNA as template (see “Materials and Methods”). The four right lanes are the sequence of a stromelysin genomic fragment using the same oligonucleotide as primer (orderof lanes from left to right isG, A, T, and C). Thesequence 5’-GACCTACAAGG-3’ is the complementof that given by the sequence reactions. The arrow and asterisks denote the positions in thesequencing gel and in thecorresponding complementary sequence, respectively, of the major primer-extension product.

TABLE I IL-10 and dexamethasone modulation of stromelysin promoter activity as evidenced by transient gene expression assay A, 24 h post-transfection, IL-1p was added to the media. Secreted hGH was assayed 72 h post-transfection (48 h after addition of ILI@).IL-lp-induction denotes the ratio of hGH in media of IL-10induced cells to that of transfected cells incubated for same time without IL-1p. Thecontrolplasmid was pXGH5. B, 24 hposttransfection, the cultures were stimulated for 3 h with IL-16, the media changed, and dexamethasone addedalong with fresh media (as in Ref. 14). Secreted hGH was assayed 75 h post-transfection (48 h after addition of dexamethasone). IL-lp induction was calculated as in A . The control plasmid was pTKGH. Each value is the average of a t least four experiments with the exception of the dexamethasone studies using the 744, 468-, and 43-bp constructs which are the average of two. A B Stromelysin promoter

IL-1p induction (48 h)

bp

Control 1293 3.7 744 468 43

1.2 14.4 7.6 6.5 1.0

IL-10 induction Dexamethasone suppression (3 h) %

1.2 6.5 3.4 1.0

0 57 62 50 0

ascomparedtountreated cultures. As thelength of the promoter fragment attached to the hGH gene was reduced, the IL-lPinducibility of the promoter diminished. Using the 744- and 468-bp constructs, IL-1/? induction dropped off to 7.6- and 6.5-fold, respectively. Finally, with the43-bp minimal promoter construct (-53 to -11) containing only the TATA box, there was no measurable response to IL-l/? treatment. Endogenous hGH gene activity was not detectably affected by IL-l/? treatment as determined in mock-transfections of carrier DNA alone norwas the controlplasmid driven by the metallothionein promoter (pXGH5)induced by IL-1P. Based on these results, we suggest that there are at least two DNA elements in the human stromelysin promoter that are responsive to the IL-1/? induction, one within the 744-bp fragment and another further upstream in the 1293-bp fragment. We havealso studied the suppression of IL-lfl-induced stromelysin expression by dexamethasone. After transfection, cell cultures were treated with IL-l/? for 3 h followed by the addition of dexamethasone. After a further incubation of 48 h, hGH was quantitated in the culture medium (Table I). With all the promoter constructs tested, dexamethasone suppressed the IL-1/? inducibilityby about 50%. The hGHcoding region contains a sequence which is positively regulated by glucocorticoids in some cell types (26). However, in our experiments dexamethasone treatment had significant no effect on hGH levels in mock-transfected cultures nor in cultures transfected with control plasmid pTKGH (thymidinekinase promoter attached to hGH gene) (data not shown). These data indicate that the 1.3-kb stromelysin promoter contains DNA element(s) involved in suppression by dexamethasone. Dexamethasone is known to both stimulate and, in some cases, suppress gene expression throughthe binding of a receptor-hormone complex (31, 32) to the same consensus DNA sequence GGTACANNNTGTYCT (33). When we looked for such a consensus sequence in the human stromelysin promoter we could findthree sequences (-1300 to -1286, -871 to -857, and -150 to -136) which had 2 mismatches on the 5’-end sequence and 1 or 2 mismatches on the 3’-end sequence (see Fig. 3). Comparison of the stromelysin promoter sequences from human, rabbit (34), and rat(35) reveal a very high degree of sequence similarity between the promoters (Fig. 5). For ex-

Regulation of Human Stromelysin Gene

a343

CATGAGGTAGGAGTGAGTTGGTGAGTGGGAGAGCATCCTCATAGMCCAGGGGAGGGAGGACTGGATAGGGGGTTTGCAAAGGGMAACCACAAMGCGGATMCATTTGAAATGTAAAA

I

I

I

I

I

I

I

AGCTTGATATCCMTTCCTGCAGCCACTTATGGAGMGATATTTGGAGTCTATCMCAAAATATTMGCMCCAAGMCACACGATTTCTCCTTTCTACAGACCCTCGAT

I

I

I

I

I

I

I

AGTCAMTGGTTTATCTGTCTCTCTCATTCTMTGTMGCTACTTGAGGGTATGGACCATGCCTCATTTGATTCTGTATTCCTAGTACTTAGCATGGTATCTGATATATAGTAGGTMCC

.....I ......... I ......... I ......... I......... I .........I .........I ......... I .........I ......... 1 .........I

- 740

- 730

-

-

-710720

-690700

- 670

-680

-650

-660

.........I .... -630

-640

AAMAAMAAMTMCCMTAAAAMTATCATAAAATMGAAAGACAAATCTGACAG~GATATMGATAAAACCAGGCCATTCTACTTCAGTAMTCATATCMTTATATGAGCCTT

II

I

I

I

I

I II

II

I

CTGTTCTTGTCCATGAGACGTTCCTCCTCMGGACTMCAAATCCCACAGAAATCACTMTAAATATATCAGATCACAGACTTAGAAAACGMCTCTCCTTCCAMTTTTCTTTAGATAC

II

I

I

I

I

II

I

I II

MAMTTACTTAGAGMCTAMTMTATATTAGTTATGAGTGAAAGGCTCATAGCAMTGTCAMCAGGTMATTMMGTCAAAAAATCCAAAAATCATGCAGACATTTCT--AGATAT

.....I ......... I.........I .........I .........I .........I ......... I......... I.........I .......:.I - 620

- 600

-610

- 590

- 580

.........I .........I ....

-

-570

-510 560 -520-530-540-550

TATAGAAAAMCTATTATAGACCCATCTCCTTTTMTATATGGGGTACAGMTGTGGGTCGTAGTGACAGACAGCGTGGGCTGTGGCTACAGAGGCACACCTTGTCCTTACCTCATCTCA

II

I

I I

I

I

I

I

I I

I

II

I

I I II I I IIII

CATTTCCTGTATTTTAGAGTATMTMTG-------CAGGCTCTAGCTATTCACGTGCAGTTT------------ TTCATTTGCTTAGATTTTGGTGCATTTMTTCTTAAACATGTTCT

II

I

I I

I

I

I I

I

I

I

II

Ill

TATCTTTTATGTTTT-GAGTATMTTGTATATAGTATAGACTATAGCTATGTATGTACACTTTCCACTTACATCTTTTATTTGCTT---------------------

.....I

TTATA-ATGTCTT

......... I .........I .........I ......... I .........I .........I .........I ......... I .........I .........I .........I ....

- 500

- 490

- 480

- 470

- 460 -410 -420 -430 -440 -450

- 390

- 400

CTGTATTCAGCTTTGACTTCTG~GTTCTTTGTACMTTTGGACTTTTTACCMGTTAGGCCACTACT---ATCCMGTCATAAACATTACAGCTTCT---GMGGATAGTTACATTTT

II I IIIIIII I IIIIII IIIIIIIIIIII II I I I l l II

II

I II

I IIIII I

II

IIII

Ill I IIII

TCCTMMTAMGCCACTCTTMMGTTCT--GCACMTTCTGACTTTTTACCAGGTCAACC---TTCTTCTCCCAMCACATGMCATMGTTTTTCTAGTAAATTCTAGTCAMTTTT

II I IIIIIII I IIIIII IIIIIIIIIIII II I I I l l IIIII I I I

II

I IIIII I

I IIIII II

I l l I IIII

TCTTMMTMMCTGCTTTTAGMGTTCT--GCACMTTCTGA-TTTTTACCMGTCMCCTACTTCTTCTCTCMMGGACAAACATAMTTGT-CTAGTGMTTCCAGTCMTTTTT

.....I

......... I ......... I.........I......... I .........I .........I .........I ......... I .........I.........I ......... I ....

- 380

- 370

-360

-350

- 340

-310 -320 -330

-300

- 290

- 280

- 270

CCAM-GTAGM-----AMMTGCCCCAGTTTTCTCTTTTGCTMGGCAGGMGCATTTCCTGGAGATTMTCACCAT-TCGCTTTGCAAMTTMGMGGTTTGAA~CT~GTAM

Ill I I I II

IIIII II IIIIIIIIIII I I I II IIIIIIIII IIIIIIIIIIIIIIIIII I I IIII

j9+I

IIIIIIIIIIIIII

IIII I I IIII

[

CCAGAGGTMAAMMMMAAAGCCCCAGTTTTCTCTTTTGCCMCACAGGMGCATTTCCTGGAGATTMTCACTGTGTTGCCTTACAAMTTGGGAAGGTTGAGAGAMTTACTAM

I l l I I I II

I l l II IIlllIIIIII I I I II IlllIIIII IIIIIIIIIIIIIIIIII Ill II II IIIIIII IIIIIII

I I I I I I IIII

C C A G M G A A A " - - - - - - AMTGCTCCAGTTTTCTCCTCTACCMGACAGGMGCACTTCCTGGAGATTAATCACTGTGTTGCCTTGCMMTTGGGAAGGTTGAGAGAMTTAGTAGTAM

.....I

......... I.........I.........I.........I......... I.........I.........I .........I .........I ......... I ......... I.... - 250

- 260

-240

- 230

- 220

-210

- 200

GMGATTATATCACTCTTC--TGATTTTT--MTTTTTGGAMTGGTCC-----CATTTGGATGGAAGCMTT

II I I l l Ill

I II IIIIIII

I l l IIIIIIIIIIIIII

I I IIIIII I IIII IIIIIIII

GTAGATTGTATCATCCTACTTTGAGTTTAAAAATCTTTGGMATGGTCCTGGTGTCGTATGGATGAAGGCA---

I II II IIIII IIIIIIII Ill

GAGTCA

GAGTC

IllIIIIIIIIIIIIIIII I I I I IIIIII I IIIIIIIIIIII

GTAGGTTGTATCATCCTACTTTCMTTTGG-MTGTTTGGAMTGGTCCTGCTGCCATTTGGATGAAAGCMG

GAGTCA

- 180

-160 -170

TTGCGGGTG-ACTCTGCAMTACTGCCACT

I l l I II I

Ill1 I I

II II Ill

CTGCAGATGTACTMACAAACATTCCTACC

I l l I II I

IIII I I

ATM

ATM

I I II Ill

CTGCGGGTGATCCAAACAMCACTGTCACT

TTM

......... I .........I......... I ......... I......... I .........I ......... I .........I.........I ......... I .........I.... - 140 - 130 -110-120 - 100 - 90 -80 - 70 -60 - 50 -40 - 30

.....I

3

TTGGGCTCAGAMGGTGGACCTC

I l l IIIIII

I

I II I

TTGAGCTCTGGAGAGTAGATGTA

Ill II IIII

I

I II I

CTGCGCTCCCGAGGTTGGACCTA

.....I

.........I.........I

- 20

-10

+1

FIG. 5. Nucleotide sequence comparison of the stromelysin promoters of rat (top line), rabbit (middle line), and human (bottom line).The lines between the sequences of each species denote nucleotide residues shared by all three promoters. The mark +I is the transcription start site for human stromelysin. The TPA responsive element and the TATAboxes are boxed.

RegulationStromelysin of Human

a344

ample, the first upstream 370 nucleotides of human and rabbit stromelysin promoter are 82% identical in sequence. Furthermore, between all three promoters there is a 69% sequence homology in the same region. The high degree of nucleotide sequence similarity suggests that this region is important for stromelysin regulation. In all species examined TPA and ILI@induce stromelysin production, and retinoic acid and dexamethasone suppress it (14-17, 35). TPA exerts its effects at the level of transcription (15, 35, 36). The DNA element responsive to TPA has been identified in the promoter of human collagenase and the homologous sequence has been found in the human metallothionein andrat stromelysin promoters (36, 37). The TPA-responsive element of the rat stromelysin promoter falls within the region of strong homology discussed above (see Fig. 5) andshares all but one nucleotide with its rabbit and human counterpart. It is, therefore, very likely that the same regions of the rabbit and human promoters are involved in the TPA response. Previously, a 700-bp promoter fragment of rabbit stromelysinwas shown to be responsive to induction by IL-1P, epidermal growth factor, TPA, or CAMP and to subsequent suppression by dexamethasone (34). By nuclear run-on and transient gene expression assay we show that human stromelysin is induced by IL-l@ at the level of transcription and that the 1.3-kb promoter fragment contains IL-l@-and dexamethasone-responsive DNA elements. In fact, the human promoter appears to contain at least two DNA elements responsive to IL-1P, one of which is located a minimum of 754 by upstream of the transcription start site. It would be interesting to determine whether rabbit and human stromelysin share the same distribution of IL-l@-responsive elements. Acknowledgments-We thank Larry Kedes for the human @-actin cDNA, Thomas Burglin for the promoter sequence comparisons, and Francine Mittleman for help in the preparation of this manuscript. REFERENCES 1. Krane, S. M. (1982) J. Znuest. Dermatol. 79, 835-865 2. Liotta, L., Rao, C. N., and Barsky, S. H. (1983) Lab Znuest. 49, 636-647 3. Reddi, A. H. (1984) in Extracellular Matrix Biochemistry (Piez, K. A., and Reddi, A. H.,eds) pp. 375-412, Elsevier/NorthHolland, New York 4. Harris, E. D., Jr., Welgus, H. G., and Krane, S. M. (1984) Collagen Relat. Res. 4, 493-512 5. Schmid, T. M., Mayne, R., Jeffrey, J. J., andLinsenmayer, T. F. (1986) J. Biol. Chern. 261,4184-4189 6. Seltzer, J . L., Adams, S. A., Grant, G. A., and Eisen, A. Z. (1981) J . Biol. Chem. 2 5 6 , 4662-4668 7. Murphy, G., Reynolds, J. J., Bretz, U., and Baggiolini, M. (1982) Biochem. J . 2 0 3 , 209-221

Gene

M. S., Hasty, K. A., Seyer, J. M., Kang, A. H., and Mainardi, C. L. (1985) J. Biol. Chem. 2 6 0 , 2493-2500 9. Murphy, G., McAlpine, C. G., Poll, C. T., and Reynolds, J. J. (1985) Biochirn. Biophys. Acta 831,49-58 10. Vartio, T. (1985) Eur. J. Biochen. 1 5 2 , 323-329 11. Okada, Y., Nagase, H., and Harris, E. D., Jr. (1986)J. Biol. Chern. 2 6 1 , 14245-14255 12. Whitham, S. E., Murphy, G., Angel, P., Rahmsdorf, H-J., Smith, B. J., Lyons, M., Harris, T. J. R., Reynolds, J. J., Herrlich, P., and Docherty, A. J. P. (1986) Biochern. J. 240,913-916 13. Collier, I. E., Wilhelm, S. M., Eisen, A. Z., Marmer, B. L., Grant, G. A., Seltzer, J. L., Kronberger, A., He, C., Bauer, E. A., and Goldberg, G. I. (1988) J. Biol. Chem. 263, 6579-6587 14. Saus, J., Quinones, S., Otani, Y., Nagase, H., Harris, E. D., Jr., and Kurkinen, M. (1988) J . Biol. Chern. 263, 6742-6745 15. Fini, M. E., Karmilowicz, M. J., Ruby, P. L., Beeman, A. M., Borges, K. A., and Brinkerhoff, C. E.(1987) Arthritis Rheum. 30,1254-1264 16. Frisch, S. M., Clark, E. J., and Werb, Z. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 2600-2604 17. Wilhelm, S. M., Collier, I. E., Kronberger, A., Eisen, A.Z., Marmer, B. L., Grant, G. A., Bauer, E. A., and Goldberg, G. A. (1987) Proc. Natl. Acad. Sci. U. S. A. 84,6725-6729 18. Matrisian, L. M., Glaichenhaus, N., Gesnel, M-C., and Breathnach, R. (1985) EMBO J . 4,1435-1440 19. Machida, C. M., Muldoon, L. L., Rodland, K. D., and Magun, B. E. (1988) Mol. Cell. Biol. 8, 2479-2483 20. Messing, J . (1983) Methods Enzyrnol. 101, 20-78 21. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. U. S. A. 7 4 , 5463-5467 22. Biggin, M. D., Gibson, T. J., and Hong, G. R. (1983) Proc. Natl. Acad. Sci. U. S. A. 80,3963-3965 23. Tabor, S., and Richardson, C.C. (1977) Proc. Natl. Acad. Sci. U. S. A. 84,4767-4771 24. McKnight, S. L., and Kingsbury, R. (1982) Science 217, 316324 25. Jones, K. A., Yamamoto, K.R., and Tjian, R. (1985) Cell 42, 559-572 26. Selden, R. F., Howie, K. B., Rowe, M. E., Goodman, H. M., and Moore, D. D. (1986) Mol. Cell. Biol. 6, 3173-3179 27. Graham, F. L., and van der Eh, A. J. (1973) Virology 52, 456467 28. Gorman, C. M., Moffat, L. F., and Howard, B. H. (1982) Mol. Cell. Biol. 2, 1044-1051 29. Greenberg, M. E., and Ziff, E. B. (1984) Nature 311,433-438 30. Nepveu, A,, and Marcu, K. B. (1986) EMBO J. 5,2859-2865 31. Yamamoto, K. R. (1985) Annu. Reu. Genet. 10, 209-252 32. Akerblom, I. E.,Slater, E. P., Beato, M., Baxter, J. D., and Mellon, P. L. (1988) Science 2 4 1 , 350-353 33. Beato, M., Arnemann, J., Chalepakis, G., Slater, E., and Willman, T. (1987) J. Steroid Biochern. 27, 9-14 34. Frisch, S. M., and Ruley, H. E. (1987) J . Biol. Chem. 262,1630016304 35. Matrisian, L. M., Leroy, P., Ruhlmann, C., Gesnel, M-C., and Breathnach, R. (1986) Mol. Cell. Biol. 6, 1679-1686 36. Angel, P., Imagawa, M., Chiu, R., Stein, B., Imbra, R. J., Rahmsdorf, H. J., Jonat, C., Herrlich, P., and Karin, M. (1987) Cell 49,729-739 37. Lee, W., Mitchell, P., and Tjian, R. (1987) Cell 49,741-752 8. Hibbs,