Vol.
169,
No.
June
29.
1990
3, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS Pages
1021-l
027
CLONING AND CHARACTERIZATION OF THE PROXIMAL PROMOTER REGION OF THE RAT INSULIN-LIKE GROWTH FACTOR I (IGF-I) RECEPTOR GENE Haim Werner, Bethel Stannard, Mark A. Bach, Derek LeRoith and Charles T. Roberts, Jr. Diabetes Branch, Building 10, Room 83-243, National Institutes of Health, Bethesda, Maryland 20892 Received
May
3,
1990
We have isolated genomic clones that contain the promoter region of the rat IGF-I receptor gene. A unique transcriptional start site was suggested by the results of primer extension and RNase protection assays, which also defined a 940-base S-untranslated region. Despite the single start site, the proximal 415 base pairs of S-flanking region were devoid of TATA or CCAAT elements. The region surrounding the start site was, however, similar to a recently described “initiator” sequence that can direct specific transcription initiation in the absence of a TATA element. The S-flanking region was GC-rich and contained several possible SPl sites, but also included potential ETF and AP-2 binding sites. The rat IGF-I receptor gene promoter region appears to have some sequences similar to both “housekeeping” and highly regulated promoters and may be an example of an intermediary class of regulatory region. O1990 Academic Press. Inc.
The biological effects of insulin-like in most cases by its activation transmembrane
protein
with
growth factor I (IGF-I)
of the IGF-I
an extracellular
cytoplasmic tyrosine kinase domain (1).
receptor, ligand-binding
The IGF-I
are mediated
a heterotetrameric domain and a
receptor is a member of a
family of tyrosine kinase receptors (2), and is most closely related to the insulin receptor. The IGF-I receptor can also mediate the effects of insulin under certain circumstances (3). Additionally, the IGF-I receptor can bind IGF-II with high affinity (4) and may be the major signal-transducing receptor for this peptide,
since the IGF-II/mannose-6-phosphate
convincingly IGF-I
receptor (5) has yet to be
demonstrated to mediate a biological effect of IGF-II.
Thus, the
receptor is pivotal in the biological actions of several hormones encoded
by the insulin gene family in that it can bind and mediate the effects of insulin, IGF-I and IGF-II. The rat IGF-I
receptor gene is developmentally regulated in a tissueInterestingly, in several tissues, the pattern of IGF-I specific manner (6). receptor gene expression is the opposite of that seen for the IGF-I ligand gene (7). In order to study the transcriptional control mechanisms that regulate IGF-I
1021
0006-291X/90 $1.50 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
169,
No.
3, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
receptor gene expression, we have cloned the putative promoter region of the rat IGF-I receptor gene. We have found that the IGF-I receptor gene promoter has a number of interesting features that suggest it is a novel intermediate between promoters for highly-regulated genes and those for “housekeeping” genes. MATERIALS
AND METHODS
Mapping of the transcription start site by primer extension: For primer extension studies, a 22-base oligonucleotide (Y-GCTGGTAAACAAGAACCTCAGC-3’) complementary to the 5’ end of the rat IGF-13seceptor cDNA clone 1 (6) was employed. This primer was end-labeled with 7- P-ATP using polynucleotide kinase (Boehringer Mannheim), and 1 x lo6 dpm were annealed to 15 pg of poly(A)+ RNA from SV40-transformed rat granulosa cells (8), neonatal brain and adult liver as well as to yeast tRNA. The granulosa cells and neonatal brain contain high levels of IGF-I receptor mRNA while there is very little IGF-I receptor mRNA in adult liver (6,8). Following incubation at 42’C for 18 hrs, samples were ethanol-precipitated and resuspended in 40 ~1 of a solution containing 50 mM Tris-HCl, pH 8.0, 100 mM KCl, 10 mM MgC12 and 0.5 mM each of dNTPs. AMV reverse transcriptase (Boehringer Mannheim, 20 U) was added and the reaction was incubated for a further 60 min at 42’C, after which the extended products were extracted with phenol/chloroform, ethanolprecipitated, and resolved on an 8% polyacrylamide/8M urea denaturing gel. Isolation of rat IGF-I receptor genomic clones: A rat genomic library (EyR1 ~3 rtial digest) was obtained from Clontech (Palo Alto, CA) and screened usrng a P-labeled 52-base oligonucleotide (5’-CCTTTTTTTCCGCTCAGCGAGTTAATGCTGGTAAACAAGAACCTCAGCCTCA-3’). This probe includes in its sequence the 22-base oligonucleotide used for primer extension (shown underlined). To prevent cross-hybridization with related sequences, filters were washed in the presence of tetramethylammonium chloride as described (9). After rescreening and plaque purification, two clones (XIGF-I-R.3 and XIGF-I-R.6) were selected for further characterization. Restriction analyses showed that these two clones contained overlapping, but not identical inserts. Since the inserts in both clones were extremely large (~16 kb), a 647-bp Alul fragment and an 849bp Smal fragment (both of which hybridized to the oligonucleotide used for the initial screening of the genomic library) were subcloned into pGem vectors (Promega) and sequenced as described (10). RNase protection assays: Antisense RNAs were generated from the subcloned genomic Alul and EcoRl-Smal cDNA fragments by linearization of the recombinant plasmids followed by in vitro transcription with SP6 RNA polymerase (for the Alul pro ) or T7 RNA polymerase (for the EcoRl-Smal probe) in the presence of 95 P-UTP according to the supplier’s directions (Promega). RNase protection assays were performed as previously described (11) and protected probe fragments were analyzed on 8% polyacrylamide/8M urea gels. RESULTS AND DISCUSSION The transcriptional start site of the rat IGF-I receptor gene was determined by primer extension (Figure 1). We found a single major initiation site which was only 9 bases upstream of the 5’ end of a rat IGF-I receptor cDNA clone that contained a 931-base 5’-untranslated region that we had previously cloned (6). The exact position of the start site and the surrounding sequences were determined from sequencing reactions that used the same primer with an appropriate genomic fragment as template. The position of this start site was 1022
Vol.
169,
No.
BIOCHEMICAL
3, 1990
AND
BIOPHYSICAL
RESEARCH
COMMUNiCATlONS
bp
67 -
58-
G T
4 5’ end of cDNA
G
u
Mapping of the 5’ end of the rat IGF-I receptor mRNA by primer extension. The sequence ladder shown on the four right lanes was obtained by using the primer extension oligonucleotide as a sequencing primer and the Smal genomic fragment as a template. The sequence shown at the right is similar to the “initiator” sequence described in ref. 12. An as isk and an arrow indicate the transcription initiation site. Markers were %‘-labelled HpaII-digested pBR325 DNA.
determined
independently
probes generated
from genomic
415 bases upstream from genomic the
transcription
protection
clones that were complementary region
failed
to
detect
RNA
any
riboprobes
up to derived
to the first 520 bases of additional
downstream
start sites (Fig. 2).
of S-untranslated
region
of the rat IGF-I
This sequence is a combination which
antisense
sequences extending
The use of antisense
The sequence of 415 bases of S-flanking
Smal (-329
assays using
clones that contained
of this start site.
or cDNA
5’-untranslated
in RNase
to 520) subfragments
overlapped
sequence as well as the 940 bases receptor
gene is shown in Figure
of the sequences of the Alul of clone XIGF-I-R.3
and had identical
sequences 1023
(-415
and cDNA
in the region
3.
to 232) and
clone 1, all of
corresponding
to
Vol.
169,
No.
BIOCHEMICAL
3, 1990
Probes
AND
BIOPHYSICAL
EcoRl-Smal
Alu 1 genomic
:
RESEARCH
COMMUNICATIONS
cDNA
+
510 bases
230 bases -b
G
L
G
*
L
EcoRl-Smal
Mapping of the 5’ end of the IGF-I receptor mRNA by RNase protection analyses. Twenty fig of total RNA from SV40-tran ormed rat granulosa cells (G) or adult rat liver (L) were hybridized to 3%P-labelled antisense RNA probes derived from genomic Ah11 and EcoRl-Smal cDNA fragments subcloned into pGem vectors. The Alul riboprobe was complementary to bases -415 to +232, 230 bases of which would be expected to be protected based upon the primer extension results, and the EcoRl-Smal riboprobe was complementary to bases t10 to t520, all of which (510 bases) were expected to be otected (Fig. 3). The numbers adjacent to each panel represent the position of 9 P-labelled molecular weight markers (HaeIII-digested 0X174 DNA for the gel on the left and the 555base full-length RNA probe for the gel on the right) and the arrows depict the position and predicted size of the protected probe bands based upon the primer extension results of Fig. 1. The schematic below shows the position of the two riboprobes with respect to the S- flanking and untranslated regions of the IGF-I receptor gene. Fia.
bases 10 to 232 of the S-untranslated Alul
and Smal subfragments
overlapping
region.
of clone XIGF-I-R.6
regions, but had a number and cDNA
most likely
for these relatively
explanation
IGF-I
receptor Despite
extension
major
and RNase protection
two polymorphic to clone XGF-I-R.3.
transcription analyses,
start site determined the proximal
sequence were devoid of any obvious TATA-like
normally
required
to a recently Adenovirus initiation
for transcription
the initiation described middle-late
initiation
site (underlined “initiator”
in the absence of a TATA
been found untranslated
sequences, which
at a specific
in Figure
site.
3) is, however,
in a putative T cell oncogene (13). regions contain a number of potential 1024
A similar
are
The sequence very similar
transferase
this sequence can direct specific element (12).
by primer
415 base pairs of 5’-
sequence found in the terminal
promoters;
is that the
alleles of the rat
from which cDNA clone 1 was derived
flanking surrounding
3. We feel that the
minor sequence differences
clones represent
from the allele corresponding the single
to each other in
from the sequences derived
clone 1 shown in Figure
gene, and that the mRNA
was transcribed
were identical
of differences
from clone AIGF-I-R.3
inserts in the two genomic
The sequences of the analogous
and
transcription
sequence has also
The Y-flanking and Sbinding sites for the Spl
Vol.
169,
No.
3, 1990
-415
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
aagcttccgggccgcact~~ccgcgcccgca
-360
~ccccttggccaccgagt)cccgcgcc~ggggcattgttttttggagtcctgcg
-300
ggtggggagggtgcggacagggccggccggcgcagtgcgggt~gggcgga~
-240
gcgcgcgtggctcagtgtgcgcgcgggggccggcgcgggt~cc
-180
tcctgggcccggctccccacqcccgcgctccgtccgcacgtccctgcgatcccgaactcc
-120
ggctcttggcgactgccgagtcgggcccggccctcggcgcgccgggactcggctgtagcc
-60
cgtLJwc
gcttggagtgtgcgcgcgggcacgtgtgcgcggccccgagagcgcgcgcgtaga~ 1
GCGGCGGCGCGGTGAGGCTGAGGTTCTTGTTTACCAGCATTAACTCGCT
b
61
GAGCGGAAAAAAAGGAGGAGGGGGaACCGAGGAGGAGGAGCGAGCGCACCGGGCG~CTCGAG
121
AGAGGCGAGCGACAGGGAAGCCGCCAGCGAGCCTATCCCGCGCGCCCGGCGCTCGCAGAC
181
CCTCGGCCCCGCTCCTTGGACCCTCCACGCCTCTCCCGTGCCTGC~GAGCTCCACGGCA
241
CGCGGCGTGCTCGGCTTTGACCTTCAGCGAGCCGGAGCCCCCGCGCACGGAGTCGGCGGC
301
I;GGCGGGGI;C~GACG~~~GGGACGG~TGAG~TG~G~GG~~GCCGCTTTGTGTGT
361
CCTGGATTTGGATTTGGGAAGGAGCTCGCGGCGGCGGCGCTGAGGGAGGAGGCGGCGGCG
421
AGCGGAGCCCAGAGGAGGAGGAGGAGCCGGAGGAGGGGGACCGCTCATTCATTTCCACT
481
CAGCATTTCTGCCCCTCGCCGGCCTCGCCCGCGCCCGGGACTTCGGACCATCTCGCCAAC
541
TGCGTCGGGCTC~~~CGTAGGCTCGGGTGGTCGTCCCCTCCGGATCGGGGGCGTT
601
TG~~~CGGCATTTGGGCTTTGCTCCTCTTTCTGTACATTTTCCTCCCCTCTTC
661
TGCATCTCTGGGTTTGCAAATGGAGGCCGACGACACCGACA
721
GGCTTCCCTACTCCGCCGCCCCGTGCGCGCTGCTGCCGGCGCTGAGGGGCCGCCCGCGCA
781
CCTCCTGTCCAGCGTTTCCGAGGATCTTCGCTCTTGTTTTTGGACGAGAGTGAGGATGAG
841
TTGAAGACTTTTTTCTTTTTTTTTTCTTTTTTTTTCTTTTCTTTCTTTCTTTCTTTTTTT
901
TTTTTTTTGAGAAAAGGGAATTTCGTCCCAAATAAFlAGG~
CCCGCCCC
CGCGCCCC
Nucleotide sequence of the Y-flanking and 5’-untranslated regions of u the IGF-I receptor gene. Nucleotide 1 corresponds to the A of the transcription initiation site. Y-flanking sequences are shown by lower case letters, whereas The “initiator” upper case letters are used for the 5’-untranslated region. sequence encompassing the transcription start site is underlined. Potential SPl binding sites are boxed. The putative ETF-binding element is overlined by a thick bar (position -369 to -357), and the potential AP-2 binding site is underlined by a thick line (-165 to -158). The ATG codon which initiates translation of the preproreceptor is at the end of the sequence shown in the figure (underlined and in boldface).
transcription interact
factor (14); since it has already been demonstrated
with
receptor
the initiator
promoter
may
sequence represent
that SPl sites can
in artificial
constructs
(12),
a
occuring
version
naturally
the IGF-I of
this
arrangement. The 5’-flanking transcription
region
factors, a potential
also contains ETF binding
binding
sites for two other
site (S-GCCCTGCCGCCGC-3’)
position
-369 to -357, and a potential
position
-166 to - 159. The presence of these sites is not entirely
the ETF transcription
AP-2
potential binding
site (5’-CCCCACGC-3’)
factor has been shown to specifically 1025
at
suprising,
initiate
at since
transcription
Vol.
169,
No.
BIOCHEMICAL
3, 1990
of genes which lack a TATA
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
element (15). The AP-2 element mediates both
cyclic AMP and phorbol ester control of transcription
(16) and the presence of a
potential AP-2 site may explain previous studies which suggested that IGF-I receptor gene expression may be regulated by both cyclic AMP (17) and phorbol esters (18). The putative ETF binding site overlaps that SPl site which exhibits the most homology to the SPl decamer consensus sequence. Overlapping cisacting sites have been described for several promoters (19-21), and in the case of the growth hormone gene promoter it was demonstrated that SPl could displace binding of the GHF-1 factor to an overlapping binding site (22). We are currently determining those regions of the IGF-I receptor gene promoter that are required for promoter activity and regulation by gel-shift, DNA footprinting and transient expression assays. From this preliminary receptor gene differs both are GC-rich
analysis, it is clear that the promoter of the IGF-I
from the promoter of the insulin receptor gene, although
and lack TATA or CCAAT elements. Both the human (23-27)
and mouse (28) insulin receptor gene promoters contain multiple transcription start sites and, apart from a potential SPl site, do not contain other known cisacting regulatory elements. The insulin receptor gene promoter, then, appears to conform to the pattern generally seen in “housekeeping” genes, which are not highly regulated. The IGF-I receptor gene, on the other hand, is highly regulated during development in a tissue-specific manner (6) as well as in other experimental models (8,18,29) and this regulation may involve the potential regulatory elements we have described in this report. We have used probes containing the rat transcription
start site to isolate
human genomic clones containing the putative promoter of the human IGF-I receptor gene. Preliminary analyses indicate that the initiator sequence and the flanking sequences are highly conserved.
REFERENCES 1. Ullrich, A., Gray, A., Tam, A.W., Yang-Feng, T., Tsubokawa, M., Collins, C., Henzel, W., LeBon, T., Kathuria, S., Chen, E., Jacobs,S., Francke, U., Ramachandran,J. and Fujita-Yamaguchi, Y. (1986) EMBO J. 5, 2503-2512. 2. Yarden, Y. and Ullrich, A. (1988)Ann. Rev. B&hem. 57, 443-478. 3. Flier, J. S., Usher, P. and Moses,A.C. (1986) Proc. Natl. Acad. Sci. (USA) 83, 664-668. 4. Steele-Perkins,G., Turner, J., Edman, J.C., Hari, J., Pierce, S.B., Stover, C., Rutter, W.J. and Roth, R.A. (1988)J. Biol. Chem. 263, 11486-l1492. 5. Morgan, D-O., Edmaa, J.C., Standring, D-N., Fried, V.A., Smith, M. C., Roth, R.A. and Rutter, W.J. (1987)Nature 329, 301-307. 6. Werner, H., Woloschak,M., Adamo, M., Shen-Grr, Z., Roberts, C.T., Jr. and LeRoith, D. (1989) Proc. Natl. Acad. Sci. USA 86, 7451-7455. 7. Adamo, M., Lowe, W.L., Jr., LeRoith, D. and Roberts, CT., Jr., (1989) Endocrinology 124, 2737-2744. 1026
Vol.
169, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
8. Zilberstein, M., Chou, J.Y., Lowe, W.L., Jr., Shen-Orr, Z., Roberts, C.T., Jr., LeRoith, D. and Catt, K.J. (1989) Mol. Endocrinol. 3, 1488-1492. 9. Wood, W.I., Gitschier, J., Lasky, L.A. and Lawn, R.M. (1985)Proc. Natl. Acad. Sci. (USA) 82, 1585-1588. 10. Hattori, M. and Sakaki, Y. (1986) Anal. Biochem. 152, 232-238. 11. Lowe, W.L., Jr., Roberts, C.T., Jr., Lasky, S.R. and LeRoith, D. (1987) Proc. Natl. Acad. Sci. (USA) 84, 8946-8950. 12. Smale,S.T. and Baltimore, D. (1989) Cell 57, 103-l 13. 13. Boehm, T., Greenberg, J.M., Buluwela, L., Lavenir, I., Forster, A. and Rabbitts, T.H. (1990) EMBO J. 9, 857-868. 14. Kadonaga,J.T. and Tjian, R. (1986) Proc. Natl. Acad. Sci.(USA) 83, 5889-5893. 15. Kageyama, R., Merlino, G.T. and Pastan,I. (1989) J. Biol. Chem. 264, 15508-15514. 16. Roesler, W.J., Vandenbark, G.R. and Hanson,R.W. (1989) J. Biol. Chem. 263, 9063-9066. 17. Adashi, E.Y., Resnick, C.E., Svoboda,M.E. and Van Wyk, J.J. (1986) J. Biol. Chem.261, 3923-3926. 18. Ota, A., Shen-Orr, Z., Roberts, C.T., Jr. and LeRoith, D. (1989) Mol. Brain Res.6, 69-76. 19. Barberis, A., Superti-Furga, G. and Busslinger,M. (1987) Cell 50, 347-359. 20. Lichtsteiner, S., Wuarin, J. and Schibler, U. (1987) Cell 51, 963-973. 21. Mercurio, F. and Karin, M. (1989) EMBO J. 8, 1455-1460. 22. Lemaigre, F.P., Lafontaine, D.A., Courtois, S.J., Durviaux, S.M. and Rousseau,G.G. (1990) Mol. Cell. Biol. 10, 1811-1814. 23. Araki, E., Shimada,R., Uzawa, H., Mori, M. and Ebina, Y. (1983) J. Biol. Chem. 262, 16186-16191. 24. Mamula, P.W., Wong, K.-Y., Maddux, B.A., McDonald, A.R. and Goldfine, I.D. (1988) Diabetes37, 1241-1246. 25. Seino, S., Seino, M., Nishi, S. and Bell, G.1. (1989) Proc. Natl. Acad. Sci. (USA) 86, 114-I 18. 26. Tewari, D.S., Cook, D.M. and Taub, R. (1989) J. Biol. Chem. 764, 16238-16245. 27. McKeon, C., Moncada, V., Pham, T., Salvatore,P., Kadowaki, T., Accili, D. and Taylor, S.I. (1990) Mol. Endocrinol. 4, 647-656. 28. Sibley, E., Kastelic, T., Kelley, T.J. and Lane, M.D. (1989) Proc. Natl. Acad. Sci. (USA) 86, 9732-9736. 29. Lowe, W.L., Jr., Adamo, M., Werner, H., Roberts, C.T., Jr., and LeRoith, D. (1989) J. Clin. Inves. 84, 619-626.
1027