THE OF BIOLOGICAL Vol. 266, No. 30, 25, pp. 20544 …Vol. 266, No. 30, Issue of October 25, pp. 20544-20549,1991 Printed in U.S.A. The a581 Fibronectin Receptor CHARACTERIZATION OF

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 266, No. 30, Issue of October 25, pp. 20544-20549,1991

Printed in U.S.A.

The a581 Fibronectin Receptor CHARACTERIZATION OF THE a5 GENE PROMOTER*

(Received for publication, November 14, 1990)

Thomas M. BirkenmeierS, Jay J. McQuillanS, Elizabeth D. BoedekerS, W. Scott Argravesjq, Erkki RuoslahtiO, and Douglas C. Dean*(( From the $Department of Internul Medicine and the )I Department of Cell Biology and Physiology, Washington University School of Medicine, St. Louis, Missouri 631 10 and the §Cancer Research Center, La Jolla Cancer Research Foundation, La Jolh, California 92037

A genomic clone containing the gene for the a5 sub- unit of the human a5@1 integrin complex was isolated by screening a human genomic library with the previ- ously described a6 cDNA (Argraves, W. S., Suzuki, S., Arai, H., Thompson, K., Pierschbacher, M. D., and Ruoslahti, E. (1987) J. CeZZBioZ 105,1183-1190). The a6 gene 5“flanking region lacks both TATA and CCAAT boxes, and it is located in a CpG island. This region was an active promoter in transfection assays using the HT-1080 cell line (fibrosarcoma), which ex- presses a6, but was inactive in the Raji cell line (B cell), which does not express a5. These results indicate that the a5 gene 5”flanking region acts as a promoter that exhibits the expected cell-type specificity. Dele- tion of a6 promoter sequences from position -667 to -178 caused transcription stimulation, suggesting that a silencer is located between these sites. Successive 6’ deletions from position -178 decreased promoter ac- tivity until activity was essentially eliminated on dele- tion to position -27. Isolation of a functional promoter for the a6 gene is a first step in understanding how expression of this gene is controlled at the molecular level.

The integrin family of adhesion receptors is comprised of cell surface proteins that mediate adhesion to the extracellular matrix as well as cell-cell interactions (1-4). Integrins are heterodimeric, containing both an (Y and a D subunit. The family can be roughly divided into subgroups based on their @ subunits; many receptors within subgroups share a common ,8 subunit whereas unique a subunits dictate ligand specificity.

The D l subgroup (also known as Yery Late Antigens or VLAs) contains at least six different receptors for extracel- lular matrix proteins (4). a5Dl is one of the receptors within this subgroup for fibronectin (Fn),’ a large extracellular ma-

* This work was supported by National Institutes of Health Grants HL02428 (to T. M. B.), HL43418 (to D. C. D.), CA42507 (to E. R.), Cancer Center Support Grant CA30199 (to E. R.), an American Lung Association research grant (to T. M. B.), and a FUR Nabisco Research Scholar award (to T. M. B.). 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.

li Present address: American Red Cross, Biochemistry Dept., 15601 Crabbs Branch Way, Rockville, MD 20855.

11 To whom correspondence should be addressed Box 8052, Wash- ington University School of Medicine, 660 S. Euclid, St. Louis, MO 63110.

The abbreviations used are: Fn, fibronectin; CAT, chloramphen- icol acetyltransferase; Hepes, 4-(2-hydroxyethyl)-l-piperazineeth- anesulfonic acid; MES, 4-morpholineethanesulfonic acid; bp, base pair(s); PCR, polymerase chain reaction.

trix glycoprotein that promotes cell attachment, growth, mi- gration, and differentiation (1). The cy581 integrin binds to the sequence containing the Arg-Gly-Asp motif located in the C-terminal half of Fn. Antibodies to 05B1 or to the a5@1 binding domain of Fn inhibit receptor-ligand binding as do peptides containing the Arg-Gly-Asp sequence, which act as competitive inhibitors (5-7). Both antibodies and Arg-Gly- Asp peptides have been used to disrupt Fn-a5@1 interactions in biological systems. These studies indicate that interaction of cells with Fn is important in such processes as gastrulation, neural crest cell migration, organogenesis, wound healing, extracellular matrix assembly, and tumor cell metastasis (6- 16).

Recent evidence suggests that binding of a581 to Fn may play a role in regulating cell proliferation and phenotype. Osteosarcoma cells that overexpress 05B1 on their surface morphologically resemble osteocytes (17). A subpopulation of K562 cells (an erythroleukemia cell line), which overexpress a5@1, can assemble filamentous actin, are fibroblastoid in appearance, and are nontumorigenic in nude mice (18). Sim- ilarly, overexpression of a5@1 on the surface of Chinese ham- ster ovary cells suppressed their transformed phenotype; they lost their ability to grow in soft agar and to form tumors in nude mice (19). In contrast to these studies, a marked decrease in a 5 expression is observed in cells transformed with the oncogenes src and ras (20).

There is evidence that binding of Fn to a581 may signal the nucleus to alter gene expression. Cells incubated with Fn fragments containing the a501 binding site increased their expression of the extracellular matrix degrading enzymes collagenase and stromelysin (21). The cytoplasmic domains of integrins interact with the cytoskeletal proteins talin, vin- culin and a-actinin (22-24); however, it is unknown if these molecules can transmit a signal from a581 on the cell surface to the nucleus.

Control of a5 expression may play a role in regulating the different biological processes that are dependent on a5Dl-Fn interactions. Erythrocytes, lymphocytes, and keratinocytes lose their ability to adhere to Fn during the processes of maturation and differentiation (25-30). It has been proposed that a loss of a5 expression is one mechanism that cells use to detach from and migrate out of the Fn-rich stroma in which they mature. In addition, regulation of a5pl expression is important in T-lymphocyte activation and differentiation (31, 32). There is a 3-4-fold increase in a5Dl expression on the surface of CD4+ memory T-cells as compared with CD4+ naive T-cells and this increase is associated with increased binding of CD4’ memory T-cells to Fn. Virtually nothing is known about the molecular mechanisms controlling a5Pl expression in these developmental processes. It is known that a5 expres-

20544

a5 Integrin Promoter 20545

sion is stimulated by transforming growth factor-pl (33), interleukin-1 (34), and phorbol ester (35, 36). Transforming growth factor-Dl acts to stimulate transcription of the a5 gene (33); however, the mechanisms of action of the other modu- lators have not been investigated.

We report the results of our examination of the structure of the 5’ end and 5’-flanking region of the a5 gene. The 5’- flanking region lies within a CpG island, lacks TATA and CCAAT boxes, and contains consensus binding sites for nu- clear proteins such as the ets family of proto-oncogenes, Spl, AP-2, and AP-1. This region is active as a promoter in transfection assays, indicating that it will be a useful tool in the study of molecular events controlling expression of the a5 gene.

MATERIAL AND METHODS

Cell Culture and Transfection Assays-HeLa and HT-1080 cell lines were grown in Dulbecco’s modified Eagle’s media supplemented with 10% fetal bovine serum. The Raji and Jurkat cell lines were grown in Rosewell Park Memorial Institute (RPMI) media with 10% fetal bovine serum. The HT-1080 cell line was transfected using the Capol technique (37). Raji and Jurkat cell lines were transfected by electroporation using a BTX Transfector (BTX Inc., San Diego, CA) as follows. Approximately 1 X 10’ cells were suspended in 0.1 ml of RPMI media containing Hepes-buffered saline (10 mM Hepes, 70 mM NaCI, 2.5 mM KC1,0.35 mM Na,HPO,, and 3.0 mM dextrose), salmon sperm DNA at 1.25 mg/ml, and 30 pg of plasmid DNA and subjected to electroporation at 200 V and 950 microfahrads. Chloramphenicol acetyltransferase (CAT) assays were done 36 h after transfection (37). pRSVLuc (0.5 mg), which contains the Rous sarcoma virus long terminal repeat directing transcription of the firefly luciferase gene, was cotransfected as an internal control. For luciferase assays, 10 pl of the cell lysate from the CAT assays were added to 190 pl of luciferase reaction buffer (10 mM Mg(OAc)Z, 50 mM Tris-MES, pH 7.8, and 2 mM ATP) and luciferase activity was analyzed using a Monolight 2010 luminometer (Analytical Luminescence Laboratory, San Diego, CA) as described (38). Percent conversion of chloram- phenicol to acetylated chloramphenicol was determined by cutting the unacetylated and acetylated [‘4C]chloramphenic~l products out of the thin layer chromotography plate and counting the radioactivity in a scintillation counter. These counts were divided by the relative light units obtained from the luciferase assay to normalize for trans- fection efficiency.

Stable transfectants expressing pa5-657CAT were created by co- transfecting the HT-1080 cell line with pa5-657CAT and pSVzneo (20:1, w/w). Thirty-six h after transfection cells were changed to media containing 350 pg/ml ‘2418, and the resulting colonies were pooled 1 week later. The cells were allowed to grow for 2 additional weeks in G418 before isolation of RNA.

DNA Characterization-An a 5 primer extension cDNA library from HeLa cell mRNA was constructed as described (39) using an oligonucleotide complementary to the a5 cDNA sequence from posi- tions +lo3 to +152 as a primer (see Fig. 3 for sequence). The resulting cDNA was cloned into the EcoRI site of X-ZAP (Stratagene Inc., La Jolla, CA). The library was screened with a second ”P-labeled oligo- nucleotide complementary to positions +4 to +23 in the a5 gene. Inserts from positive clones were subcloned into the plasmid SK- Bluescript (Stratagene), and both strands were sequenced with oli- gonucleotide primers using the dideoxynucleotide method (40). Chem- ical sequencing was done as described (41).

A human spleen genomic library 10165 (42) was created by partial digestion of genomic DNA with MboI followed by ligation into the the EamHI site of the phage vector Charon 28. The original library contained more than 1 X lo6 members prior to amplification, with an average insert size of 18 kilobase pairs. This library was screened with a partial a 5 cDNA clone extending from positions +266 to +577 in the published a 5 cDNA (43). Two positive clones were obtained, only one of which contained the 5’-flanking region. Southern blot analysis of positive clones was performed as described (44). Restric- tion enzyme fragments of the genomic clone were subcloned into the plasmid SK-Bluescript, and both strands were sequenced using the dideoxynucleotide method (40). A PstI subclone (Pst-lA), which contains part of the first intron, all of the first exon, and approxi- mately 2 kilobase pairs of 5‘-flanking region, was used in the PCR described below.

A

9.4 - 6.6-

4.4-

El

2.3. 2.0.

0.56,

8 1- 17kb

Xba I (-1.4kb)

PSI I (+400)

Hind 111 (+2.4kb)

Hind 111 Eco RI

( -657 ) +1 (+1.4kb) Barn HI/Mbo I., .Barn HllMbo I

i. left arm *-. I I ..... ,-’ A nghl arm ...” ,

Xba I ’ Sac I f.l 4kb) (.goo)

Apa I PSI I

(+as) (+400)

FIG. 1. Genomic Southern blot analysis of the a5 gene. A, 10 pg of genomic DNA were digested with the indicated restriction enzymes, subjected to agarose gel electrophoresis, and transferred to nylon membrane. The membrane was hybridized to a PstI (+400)/ HindIII (-657) fragment from the a5 genomic clone (see panel E ) . The size standard is HindIII-digested X DNA; fragment sizes are given in kilobase pairs. E , restriction enzyme map of the 5’ end of the a5 gene. The arrow marked +1 is the 3’-most transcriptional start site (Fig. 2). This map was determined by DNA sequence analysis and by Southern blots of restriction enzyme digests of the genomic clone hybridized to a synthetic oligonucleotide that corre- sponds to sequences between positions +4 and +23 of the a 5 gene (see Fig. 3 for sequence). Note that the genomic Southern in panel A shows hybridization to fragments of the size predicted from the restriction map in panel B (Le. the PstI/HindIII, PstIIXbaI, and PstI/ Sac1 fragments).

Plasmid Constructs-To create the CAT expression vectors, the CAT gene was cut from the vector p65CAT (37) with PstI and EamHI and the ends of the fragment were blunt-ended with T4 polymerase. The insert was then cloned into the SmaI site (SK-Sma-CAT) or the XhoI site (which had been blunt-ended with T4 polymerase) (SK- Xho-CAT) of SK-Bluescript (Stratagene). The orientation of the insert was determined by restriction enzyme mapping; in SK-Sma- CAT the direction of transcription of the CAT gene was toward the T3 promoter, and in SK-Xho-CAT it was toward the T7 promoter. The CAT sequence in these plasmids extended 23 bp 5’ of the first ATG codon in the CAT gene. These promoterless constructs showed no CAT activity in transfection assays. 5”deletions in the a5 gene 5’-flanking region were created using PCR (45). Plasmids pa5-

pa5-41CAT, and pa5-27CAT were made using a common 3’ oligo- nucleotide (corresponding to the reverse complement of the sequence between positions +4 and +23 with a Sal1 or PstI site on its 5’ end) and distinct 5’ oligonucleotides (with a HindIII or PstI site on their 5’ end) in PCR reactions with the template plasmid Pst-IA (digested with PstI). The oligonucleotides used as 5’ PCR primers are as follows (restriction enzyme sites are underlined): pa5-657CAT, 5’-d(TG-

657CAT, pa5-250CAT, pa5-178CAT, pa5-132CAT, pa5-92CAT,

AAGAAGCTTGGAGGAAGCAGC)-3’; pa5-250CAT, 5”d(GACT- A A G m C T A A C C C A G T C C A G A C A ) - 3 ’ ; pa5-178 CAT, 5”d-

20546 a5 Integrin Promoter

B

147

160 - - go - .* ! *

76 - ti++’ 67 - . -

Nar I Sma I

”“ 5’ ”” 5’

112- e, *I (75 mer)--

probe

5’3’ 6 8 - 1 5’3’ G C C G C G C G C G

‘ C G * A T

G C 36 - G C A T 28 -

3’5’

59- i

A T -45” I

G C ‘ C G C G 3’5’

- x

o w c ZQACGT

5’3

G C A T

C G T A

‘ A T T A

C G T A

G C C G 35’

Exon I > 75mer 10bp/ ,

21bp, 44bp 5 , - - - - - - - * 5’

20bp I 35bp 5 , 65mer lObp/

FIG. 2. Identification of a5 gene transcriptional start sites. A, S1 nuclease analysis using a 5’ end-labeled SmaI fragment (position +190) hybridized to 50 pg of total HeLa cell RNA. The standards (std) are MspI-digested pBR322 and the fragment sizes are given in bp. B, S1 nuclease analysis using a 5’ end-labeled NarI fragment (position +82) hybridized to 50 pg of total HeLa cell RNA. The standards are the same as in panel A. C, S1 nuclease analysis using a synthetic oligonucleotide. The oligonucleotide 65mer was labeled on its 5’ end and hybridized to 15 pg of poly(A+) RNA from HeLa cells. Next to the S1 reaction are chemical sequencing reactions of the 65-mer. The standards (std) are MspI-digested SK Bluescript, and the fragment sizes are given in bp. D, S1 nuclease analysis using a second synthetic oligonucleotide. The oligonucleotide labeled 75mer was used in an S1 nuclease analysis as in panel C. The sequence of this region of the a5 gene using a primer with the same 5’ end as the 75-mer is shown. Size standards are the same as in panel C. E , primer extension analysis. A 5’ end-labeled oligonucleotide was hybridized to 15 pg of poly(A+) RNA from HeLa cells. A dideoxy sequencing reaction of the genomic subclone using the same primer is shown. The product correlates to the S1 nuclease site a t position -45 as marked by the arrow, and the sequence of the region is shown with the base at -45 marked by an asterisk. The standards (std) are the same as in panel D. +I and -45 indicate the major transcriptional start sites. Primers and probes for the assays are shown schematically below each figure.

(TTAACTGCAGCCCACGCCCCTTAGGGGTGG)-3’,pa5-132 CAT,5”d(TATACTGCAGAGGGACCCAGGAATGCCCCC)-3’; pa5-92CAT 5’-d(TATACTGCAGGGAGGGCTCAGCCGGGAGT- TT)-3’; pa5-41CAT, 5”d(TACACTGCAGCCTCTGGGAGGT’M” AGGAAG)-3’; pa5-27CAT, 5’-d(TACACTGCAGAGGAAGCGGCT- CCGGGT)-3’. To make pa5-178CAT, pa5-132CAT, pa5-92CAT, pa5-41CAT, and pa5-27CAT, the PCR products were digested with PstI and SalI and ligated into the corresponding sites in SK-Xho- CAT. To make pa5-657CAT and pa5-250CAT, the PCR products were digested with HindIII and SalI and ligated into the correspond- ing sites in SK-Sma-CAT. To create pa5+1CAT, two complementary oligonucleotides corresponding to a5 gene sequences between posi- tions +1 and +23 were synthesized such that, after annealing, they contained a HindIII site on the 5’ end and a PstI on the 3’ end. The annealed oligonucleotides were cloned into the HindIII and PstI sites of SK-Sma-CAT. To create pa5-657CAT-R, the PstIIHindIII insert from pa5-657CAT was ligated into the corresponding sites of SK- Xho-CAT. The a5 gene insert in each plasmid was sequenced to assure that no mutations occurred during PCR. Construction of pTA- CAT, which contains the SV40 early gene TATA box directing transcription of the CAT gene, was described previously (37).

RNA Isolation and Characterization-Total RNA was isolated by the guanidinium isothiocyanate-CsC1 method (46), and poly(A+) RNA was selected by oligo(dT) affinity chromotography (47). S1 nuclease assays, primer extension reactions, Southern blot hybridization, and Northern blot hybridization were performed as described (42).

RESULTS AND DISCUSSION

Isolation of an a5 cDNA-A cDNA encoding the a5 subunit of the a5B1 integrin complex was described previously (43 ) . A synthetic oligonucleotide complementary to the 5’ end of

the a 5 mRNA (position +4 to +23; see Fig. 3 ) was used in primer extension assays to determine the distance to the 5’ end of the gene. Surprisingly, multiple extension products were observed (data not shown). A primer extension cDNA library was made using this oligonucleotide; however, all of the clones selected from this library encoded a cDNA for the 28 S rRNA gene. Analysis of the 5’ end of the a5 cDNA revealed similarity to the 5’ end of the 28 S rRNA gene. This region of the a5 cDNA hybridized to 28 S rRNA on Northern blots and to the 28 S rRNA gene on genomic Southern blots (data not shown).

To circumvent this problem, an oligonucleotide comple- mentary to the a 5 cDNA sequence between positions +lo3 and +152, which is not similar to the 28 S gene, was used to make a second primer extension cDNA library. Most of the clones analyzed from this library contained a5 cDNA in their 3’ ends but diverged into either the 28 S cDNA or another unknown cDNA sequence before reaching the 5’ end of the a 5 cDNA (data not shown). Interestingly, the region of the a 5 cDNA where the clones diverge hybridized to several different mRNA species, distinct in size from a5 mRNA (data not shown). Therefore, most of the cDNAs that we obtained appeared to be fusions between two different RNAs; their 3’ ends contained a5 sequence, whereas their 5’ ends contained another cDNA. However, one clone isolated from this library extended through the a5 cDNA and continued 28 bp 5’ of the known sequence. This cDNA was derived entirely from the

a5 Integrin Promoter 20547 Hind 111

- 7 I AATACTGAAGAAGCTTGGAGG-CAGCGGACTTGAGGAGCCCGTGTAGGTCC

-6?0 els I

TCTAGCAAGCTGACTGCGAACGCTGTCTCTGTACCTCAACTGTTGTGTCCC-TC

CA~CTZGCCAGAACCZAGGCACCCG~CGGCCCCGG~GGC~GGG~GAATCCCAGT

T~GCC~GACGCCAGAGGAG~CTCCTGTCT~TTTAGACCC~AAAGTCT~C~CCCTAA

AGCACT~AGGGAGACCCCACACA;ATATACAGG~TTCCTCCGCCCACCAGAGG~GA

-5pO

-490

TTCCTTTCCTCATTAGGAAATTCTCCGCTCCCTTTTCCGACTCGTTTTCCGAGCGT

-3pO

T~A~GTTGTA~ATCTGG-GAAGGGGA~GAAGGAGGAAGGAAGCA~AGAG

G A G G G ~ ~ A A G A A C C C A A G T ~ T A A ~ ~ ~ A G T ~ ~ A G A C ~ C C G G C T T C C ~

GCTGG~~CT~G~GAAAGGGGGTTGGAGGG~TGCGCCCCC~CCCCACGCC~CTTAGG

G G T G G G G G A C G C G ~ G C T C A G A G T ~ T C C A G G G A C ~ C A G G A A T G C ~ C ~ ~ C C

SPl ets

-2po

SPl

*loo G+G CCG CTG C T ~ TTG CTG CTC ~ T G CCG CCG C ~ A ccc AGO G T ~ GGG

V P L L L L L V P P P P R V G I

GGC TTC AAC TTA GAC GCG GAG GCC CCA GCA GTA CTC TCG GGG CCC G F N L D A E A P A V L S G P

* Z O O ChG GGC TCC TTC TTC GGA TTC ~ C A GTG GAG T ~ T TAC CGG cct GGA

P G S F F G F S V E F Y R P G I

ACA GAC GGG TGAGTGAGGAGGGCTGGAGACGGGATGGGGGTGGGGGAGCCGCC

t .300 TGGAGACTGGGGGCGCGGCTGGGGTCTGGAGGGGACCAGGGCTTAAGTTT

FIG. 3. DNA sequence of the 5’-flanking region and first exon of the human a5 gene. Horizontal arrows indicate the two major sites of transcriptional initiation. The vertical arrow indicates the 3’ end of exon 1. Consensus binding sites for several different elements are shown. The sequence of the a5 cDNA obtained from the primer extension library is contained entirely within the genomic sequence between positions -29 and +151.

a5 gene, because it was identical in sequence to a genomic clone containing this region of the gene (see Fig. 3 below).

Isolation of an a5 Genomic Clone-A genomic clone con- taining the a5 gene was isolated as described under “Material and Methods.” Genomic Southern blot analysis (Fig. 1A) showed that this clone had the same restriction enzyme map as genomic DNA (compare the size of the XbaIIPstI, Sad/ PstI, and HindIIIIPstI restriction enzyme fragments from the a5 gene in the genomic Southern (Fig. lA) to those predicted from the sequence and the restriction map of the genomic clone (Fig. lB)), indicating that there was no recombination of DNA sequences within this region of the clone.

Identification of the 5’ End of the a5 Gene-A combination of primer extension and S1 nuclease analysis were used to map the 5’ end of the a5 gene. S1 nuclease analysis revealed two major transcription initiation sites (Fig. 2, A-D). The 3’- most site was designated position +1, placing the other major site at position -45. Primer extension analysis using a primer complementary to the sequence between the two major S1 sites confirmed that the 5’-most site is a point of transcrip- tional initiation (Fig. 2E). The 5’-most site is located within an Inr consensus, a sequence that may be important in defin- ing the site of transcriptional initiation (48). Interestingly, the Inr consensus sequence is also found at the site of tran- scriptional initiation of the platelet integrin a-chain gene, gpIIb (49,50), which like the a5 gene, does not have a TATA or CCAAT box.

28s-&

18s-

. ’.

, 657 . L b .: .23.109

; ; L/ 1 bosd ‘CAT < %conversion 0 I 013 52 . . . . . . . . . . . -.

. . . . . . . - . FIG. 4. The a5 gene 5”flanking region acts as a promoter

and shows tissue specificity in transfection assays. A, expres- sion of a5 mRNA in HT-1080 and Raji cell lines. Five pg of poly(A+) RNA from these cell lines were subjected to Northern blot analysis. The blot was hybridized to an a5 cDNA (XP7) (43). B, diagram of pa5-657CAT. C, expression of pa5 -657CAT in HT-1080 cells. pa5- 657CAT was transfected in parallel with pSVzCAT, which contains the strong promoter for the SV40 early gene and pTA-CAT, which contains a very weak promoter consisting of only the SV40 early gene TATA box. D, comparison of the activity of pa5-657CAT, pSVzCAT, and pTA-CAT in Raji cells. E, identification of transcriptional initi- ation sites in pa5-657CAT. Twenty pg of poly(A+) RNA from HT- 1080 cells stably expressing pa5-657CAT was used in a primer extension assay. A 32P-labeled primer complementary to the sequence beginning 42 bp 3’ of the ATG of the CAT gene, 5’- d(TTGGGATATATCAACGGTGGTATATCC)-3‘, was used in these extension assays. Arrows indicate the location of extension products initiating from the two major a5 transcriptional start sites located at positions +1 and -45 (shown schematically at the bottom of the figure). CAT assays were performed at least three times with no significant variation in the results. Standards (std) are MspI-digested SK-Bluescript.

Primer extension analysis using primers 3’ to position +1 were inconclusive; multiple bands resulted due to the similar- ity between this region of the a5 gene and other genes dis- cussed above (data not shown). The first exon is either 242 or 287 bp in length, depending on which major transcription initiation site is used, whereas the 5”untranslated region is either 23 or 68 bp.

The 5”Flanking Region of the a5 Gene Lacks TATA and CCAAT Boxes and Is Contained in a CpG Island-The 5’- flanking region of the a5 genomic clone lacks TATA and CCAAT boxes, and its sequence (the sequence shown in Fig. 3) contains 70% C+G. In this region, the ratio of CpG to GpC is 0.8, indicating the absence of CpG suppression. Addi- tionally, this region contains 11 HpaII restriction enzyme sites (CCGG). Both of these properties indicate that the 5’ end and 5”flanking region of the a5 gene are found in a CpG island, also referred to as a HpaII tiny fragment island (51).

Analysis of the Fn gene promoter, which contains a TATA and a CCAAT box and is regulated by a variety of hormones and growth factors (37, 52,53), indicates that it also is found in a CpG island (data not shown). Methylation of CpG islands affects chromatin structure, which in turn may dictate regu- lation and tissue specificity of gene expression (54). Therefore, selective methylation of CpG-rich regions in the Fn and a5

20548 a5 Integrin Promoter A. n

pa5-92CAT Da5-41CAT

FIG. 5. Effect of 5’ deletions on the activity of the a5 gene promoter. A , diagram of the construction of a5-CAT fusion genes. See “Materials and Methods” for details of how these plasmids were constructed. Arrows below the CAT gene indicate direction of tran- scription. B, CAT constructs containing various amounts of the a5 gene 5”flanking region were transfected in the HT-1080 cell line. pa5-657CATR is identical to pa5-657CAT, but it contains the a5 promoter fragment in reverse orientation. pTA-CAT is described in the legend to Fig. 4. These results are the average of three separate experiments in which the transfections were performed in duplicate.

gene promoters may provide a mechanism for coordinate regulation of this receptor-ligand pair.

The a5 gene flanking region and first exon contain consen- sus binding sites for several transcription factors, such as AP- 1, the ets family of proto-oncogenes (consensus binding sites for ets are also found in the 5”flanking region of the a4 and gpIIb integrin genes; see Refs. 49, 50, and 55), Spl, and AP-2 (Fig. 3). An AP-1 site mediates transcriptional stimulation by phorbol esters (56), serum (57), and transforming growth factor-8 (58). Since expression of a5 is increased by each of these inducers (33, 36), an AP-1 site may play a role in regulating its expression.

The 5‘-Flanking Region of the a5 Gene Acts as a Promoter and Shows Cell-type Specificity in Transfection Assays”pa5- 657CAT, which contains a5 gene 5”flanking sequence be- tween positions -657 and +23 fused to the CAT gene, was constructed to determine if the the a5 gene 5‘-flanking region would act as a promoter and show the proper cell-type speci- ficity in transfection assays (a diagram of the plasmid is shown in Fig. 4A). The activity of pa5-657CAT was examined in two different cell types: the fibrosarcoma cell line HT-1080, which expresses a5, and the B cell line Ftaji, which does not (Fig. 4B) . The a 5 gene 5”flanking region acted as a promoter in HT-1080; it was only slightly less active than the strong viral promoter from the SV40 early gene and was much more active than a very weak promoter containing only the SV40

early gene TATA box (Fig. 4C). In contrast, the a5 gene 5‘- flanking region was essentially inactive as a promoter in Raji cells (Fig. 4 0 ) . Therefore, the 5”flanking region of the a5 gene functions as a promoter and shows the expected cell- type activity.

TO insure that transcription was initiating properly in the a5-CAT fusion gene, mRNA from HT-1080 cells stably ex- pressing pa5-657CAT was used in a primer extension assay to map transcriptional start sites in the fusion gene, Tran- scription of the fusion gene initiates at the same sites as the endogenous a5 gene (Fig. 4E) , indicating that CAT activity is a reflection of accurate transcriptional initiation from the a5 promoter.

Effect of 5’ Deletions on a5 Gene Promoter Actiuity-Spe- cific deletions in the 5”flanking region of the the a5 gene were made using PCR as described under “Materials and Methods.” The effect of these deletions on promoter activity was examined in transfection assays in the HT-1080 cell line. Deletion of a5 gene 5”flanking sequences from positions -657 to -178 increased promoter activity (Fig. 5), suggesting that a silencer is located between these sites. Subsequent deletions 5‘ of position -178 progressively decreased pro- moter activity, until activity was essentially eliminated on deletion to position -27. Similar results were obtained using Jurkat cells, a T cell line that expresses a581 (data not shown). These results suggest that several elements that are required for efficient transcription are located between posi- tions -178 and -27. Further characterization of this promoter should lead to a better understanding of how the tissue distribution of a5 is controlled and how its level of expression is regulated by different hormones and growth factors.

Recently we have isolated and characterized the promoter of the a4 gene, a second 81-integrin-Fn receptor (55). a481 and a581 bind to different sites on Fn, suggesting that these integrins are functionally distinct. Expression of a4 is re- stricted to lymphoid and myeloid cells; in contrast, a5 is expressed on most adherent cell types. Unlike the a5 pro- moter, the a4 promoter has both a CAAT and a TATA box and is not located within a CpG island. Aside from the presence of ets binding sites (discussed above), the two pro- moters exhibit little similarity. Therefore, differences in tissue specificity between these two Fn receptors are likely the result of alternate sets of promoter elements. Analysis of the struc- ture of both promoters may ultimately provide insight into how different cells control their adhesion to Fn.

REFERENCES 1. Ruoslahti, E. (1988) Annu. Rev. Biochem. 5 7 , 375-413 2. Hynes, R. 0. (1987) Cell 48,549-554 3. Ruoslahti, E., and Pierschbacher, M. D. (1987) Science 238,491-

4. Hemler, M. E. (1990) Annu. Reu. Zmmunol. 8, 365-400 5. Pierschbacher, M. D., and Ruoslahti, E. (1984) Nature 309,30-

33 6. McDonald, J. A., Quade, B. J., Broekelmann, T. J., Lachance,

R., Forsman, K., Hasegawa, E., and Akiyama, S. (1987) J. Biol. Chem. 262,2957-2967

7. Akiyama, S. K., Yamada, S. S., Chen, W. T., and Yamada, K. M. (1989) J. Cell Bwl. 109,863-875

8. Bronner-Fraser, M. (1985) J. Cell Bwl. 101,610-617 9. Boucaut, J. C., Darribere, T., Poole, T, Aoyama, H., Yamada, K.

M., and, Thiery, J. P. (1984) J. Cell Biol. 99, 1822-1830 10. Boucaut, J. C., Darribere, T., Boulekbache, H., and Thiery, J. P.

(1984) Nature 3 0 7 , 364-367 11. Bronner-Fraser, M. (1986) Dew. Biol. 117 , 528-536 12. Duband, J. L., Rocher, S., Chen, W. T. Yamada, K. M., and

13. Humphries, M. J., Olden, K., and Yamada, K. M. (1986) Science

14. Linask, K. K., and Lash, J. W. (1988) Dew. Biol. 129,324-329

497

Thiery, J. P. (1986) J. Cell Biol. 102, 160-178

233,467-470

15. 16. 17.

18. 19. 20. 21.

22.

23.

24.

25.

26. 27.

28. 29.

30.

31.

32.

33.

34. 35.

36.

37.

a5 Integr~

Linask, K. K., and Lash, J. W. (1988) Deu. Biol. 129, 315-323 Grinnell, F. (1984) J. Cell Biochem. 26, 107-116 Dedhar, S., Argraves, W. S. , Suzuki, S., Ruoslahti, E., and

Pierschbacher, M. D. (1987) J. Cell Biol. 105, 1175-1182 Symington, B. (1990) Cell Regul. 1,637-648 Giancotti, F. G., and Ruoslahti, E. (1990) Cell 60,849-859 Plantefaber, L. C., and Hynes, R. 0. (1989) Cell 56, 281-290 Werb, Z., Tremble, P., Behrendtsen, O., Crowley, E., and Damsky,

C. (1989) J. Cell Biol. 109,877-889 Horwitz, A. F., Duggan, K., Buck, C. A., Beckerle, M. C., and

Burridge, K. (1986) Nature 320, 531-533 Tamkun, J. W., DeSimone, D. W., Fonda, D. Patel, R. S., Buck,

C., Horwitz, A. F., and Hynes, R. 0. (1986) Cell 46, 271-282 Otey, C. A., Pavalko, F. M., and Burridge, K. (1990) J. Cell Bwl.

Giancotti, F. G., Comoglio, P. M., and Tarone, G. (1986) J. Cell

Patel, V. P., and Lodish, H. F. (1986) J. Cell Biol. 102,449-456 Vuillet-Gaugler, M. H., Breton-Gorius, J., Vainchenker, W., Gui-

chard, J., Leroy, C., Tchernia, G., and Coulombel, L. (1990)

11 1,721-729

Biol. 103, 429-437

Blood 76,865-873 Adams, J. C., and Watt, F. M. (1990) Cell 63,425-435 Cardarelli, P. M., and Pierschbacher, M. D. (1986) Proc. Natl.

Cardarelli, P. M., Crispe, I. N., and Pierschbacher, M. D. (1988)

Shimizu, Y., Van Seventer, G. A., Horgan, K. J., and Shaw, S.

Shimizu, Y., Van Seventer, G. A., Horgan, K. J., and Shaw, S.

Roberts, C. J., Birkenmeier, T. M., McQuillan, J. J., Akiyama, S. K., Yamada, S. S., Chen, W. T., Yamada, K. M., and McDonald, J. A. (1988) J. Biol. Chem. 263,4586-4592

Dedhar, S. (1989) J. Cell Physiol. 138, 291-299 Symington, B. E., Symington, F. W., and Rohrschneider, L. R.

Suzuki, S., Argraves, W. S., Arai, H., Languino, L. R., Piersch- bacher, M. D., and Ruoslahti, E. (1987) J. Biol. Chem. 262, 14080-14085

Dean, D. C., Newby, R. F., and Bourgeois, S. (1988) J. Cell Biol.

Acad. Sci. U. S. A. 83, 2647-2651

J. Cell Biol. 106, 2183-2190

(1990) Nature 346, 250-253

(1990) J. Immuml. 145,59-67

(1989) J. Biol. Chem. 264,13258-13266

106,2159-2170

in Prc 38.

39. 40.

41.

42.

43.

44. 45.

46.

47.

48. 49.

50.

51. 52.

53.

54. 55.

56.

57.

58.

Imoter 20549

Nguyen, V. T., Morange, M., and Bensaude, 0. (1988) Anal.

Gubler, U., and Hoffman, B. J. (1983) Gene (Amst.) 25,283-287 Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl.

Maxam. A. M.. and Gilbert. W. (1977) Proc. Natl. Acad. Sci.

Biochem. 171,404-408

Acad. Sci. U. S. A. 74, 5463-5467

U. S. A. 74, 560-564 . .

Dean. D. C.. Bowlus. C. L., and Bourgeois. S. (1987) Proc. Natl. A c k . Sci.'U. S. A. '84, 1876-1880

Argraves, W. S., Suzuki, S., Arai, H., Thompson, K., Piersch- bacher, M. D., and Ruoslahti, E. (1987) J. Cell Biol. 105,1183- 1190

- .

Southern, E. M. (1975) J. Mol. Biol. 98,503-517 Saiki, R. K., Gelfand, D. H., Stoffel, S. J., Higuchi, R., Horn, G.

T., Mullins, K. B., and, Erlich, H. A. (1988) Science 488,487- 491

Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J. and, Rutter, W. J. (1979) Biochemistry 18,5294-5299

Aviv. H.. and Leder. P. (1972) Proc. Natl. Acad. Sci. U. S. A. 69. I . .

14o8-i412 Smale. S. T.. and Baltimore. D. (1989) Cell 57, 103-113 Prandini, M: H., Denarier, E. , Frachet, P., Uzan, G., and Mar-

guerie, G. (1988) Biochem. Biophys. Res. Commun. 156, 595- 601

Heidenreich, R., Eisman, R., Surrey, S., Delgrosso, K., Bennett, J. S., Schwartz, E., and Poncz, M. (1990) Biochemistry 29, 1232-1244

Bird, A. P. (1986) Nature 321,209-213 Blatti, S. P., Foster, D. N., Ranganathan, G., Moses, H. L., and

Getz, M. J. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 1119- 1123

Dean, D. C., Blakeley, M. S., Newby, R. F., Ghazal, P., Hennig- hausen, L., and Bourgeois, S. (1989) Mol. Cell. Biol. 9, 1498- 1506

Tazi, J., and Bird, A. (1990) Cell 60,909-920 Rosen, G. D., Birkenmeier, T. M., and Dean, D. C. (1991) Proc.

Natl. Acad. Sci. U. S. A. 80,4094-4098 Angel, P., Imagawa, M., Chiu, R., Imbra, R. J., Rahmsdorf, H. J.,

Jonat, C., Herrlich, P., and Karin, M. (1987) Cell 49, 729-739 Imler, J. L., Schatz, C., Wasylyk, C., Chatton, B., and Wasylyk,

B. (1988) Nature 332, 275-278 Kim, S. J., Angel, P., Lafyatis, R., Hattori, K., Sporn, M. B.,

Karin, M., and Roberts, A. B. (1990) Mol. Cell. Bwl. 10, 1492- 1497