Gene, 77 (1989) 95-105

Elsevier

95

GEN 02904

Structure of the 5’ flanking region of the gene encoding human parathyroid-hormone-related protein (PTHrP)

(Alternate splicing; cDNA; gene regulation; genomic sequence; exon; intron; 5’ untranslated)

L.J. Suvaa, K.A. Mather ‘, M.T. Gillespie b,G.C. Webbd,K.W.Nga,G.A. Winslowe, W.I. Wood”,T.J.Martin” and P. J. Hudson ’

a St. Vincent’s Institute of Medical Research and University of Melbourne, Department of Medicine, Fitzroy (Australia) 3065, Tel. (03)4182375: b Department of Veterinary Clinical Sciences, University of Melbourne, Werribee (Australia) 3030, Tel. (03)7413500; ’ C.S.I.R.O. Division of Biotechnology, Parkville (Australia) 3052, Tel. (03)3424312; d Human Genetics Group, John Curtin School of Medical Research, Australian National University, Canberra (Australia) 2601, Tel. (062)494710 and e Genentech. Inc., South San Francisco, ‘CA 94080 (U.S.A.) Tel. (41.5)2661000

Received by P.A. Manning: 28 May 1988

Revised: 27 September 1988

Accepted: 28 November 1988

SUMMARY

We have characterized a human genomic clone that contains the 5’ coding and 5’ flanking sequences of the

human parathyroid hormone-related protein gene (PTHrP). The 5’ end of the gene contains three exons

separated by two small introns of 60 and 165 bp, respectively. The coding region of the PTHrP gene exhibits

significant structural homology to the human parathyroid hormone gene (PTH), including the position of at

least two introns. However, there is no significant nucleotide sequence homology to the PTH gene within the

intragenic region nor in the flanking genomic sequences. The PTHrP gene has been localized, by chromosomal

in situ hybridization to bands pll or ~12, on human chromosome 12. Analysis of the 5’-noncoding DNA

reveals a complex, putative regulatory region, with multiple potential transcription start points. Nucleotide

sequence analysis shows the position of one consensus TATA sequence, at -514 bp, from the start of

translation whereas the other regulatory domain is located at least 1 kb further 5’ to this consensus TATA

sequence. Evidence from the structure of a number of cDNA clones, as well as Sl nuclease and primer

extension studies supports the hypothesis that the PTHrP gene contains at least two mRNA transcription start

points that define two putative regulatory domains. The result of expression from these different promoters

combined with an alternative splicing event would be to produce multiple forms of PTHrP mRNA that differ

in the 5’-untranslated region. This analysis of the human PTHrP gene is the first report of a PTHrP gene for

any species.

Correspondence to: Dr. T.J. Martin, St. Vincent’s Hospital,

Department of Medicine, University of Melbourne, 41 Victoria

Parade, Fitzroy, 3065 Victoria (Australia) Tel. (03)4182375;

Fax. (61-3)4173340.

nt, nucleotide(s); oligo, oligodeoxyribonucleotide; PolIk, Klenow

(large) fragment ofE. coli DNA polymerase I; PTH, parathyroid

hormone; PTH, gene (or DNA) coding for PTH; PTHrP, PTH-

related protein; PTHrP, gene (or DNA) coding for PTHrP; SDS,

Abbreviations: aa, amino acid(s); bp, base pair(s); HHM, hu- sodium dodecyl sulfate; S SC, 0.15 M NaCl/O.OlS M Na, citrate

moral hypercalcaemia of malignancy; kb, kilobase or 1000 bp; pH 7.6; SV40, simian virus 40; UTR, untranslated region.

0378-1119/X9/$03.50 0 1989 Elsevier Science Publishers B.V. (Biomedical Division)

96

INTRODUCTION

Cancers associated with the HHM produce a

factor that is immunologically distinct from PTH,

but resembles PTH in its biological activity (Martin

and Atkins, 1979; Stewart et al., 1980). We have

recently purified and sequenced this PTHrP from a

human lung cancer cell line (BEN) (Moseley et al.,

1987), determined the mRNA sequence encoding the

141 aa mature protein from cloned cDNA (Suva

et al., 1987) and synthesized N-terminal peptides

with potent PTH-like biological activity (Kemp

et al., 1987). These important functional data sup-

port the idea that the HHM syndrome is mediated

by a secreted tumor factor with biological activity

similar to PTH. Subsequen~y, two other groups

have published cDNA cloning data describing a

number of distinct PTHrP cDNA species (Mangin

et al., 1988; Thiede et al., 1988). It was, therefore, of

considerable interest to explore the regulation of

expression of the PTHrP gene and for this reason we

undertook to isolate genomic DNA encoding

PTHrP and ultimately those regions imposing tran-

scriptional control.

Using the cDNA structure (Suva et al., 1987) and

considering the obvious biochemical similarities

between PTHrP and PTH (Powell et al., 1973;

Stewart et al., 1983; Moseiey et al., 1987) we specu-

lated that the gene encoding PTHrP may well be

similar to the gene encoding human PTH (Vasicek

et al., 1983). We therefore designed synthetic oligos

that would hybridize preferentially to the presump-

tive separate exons, assuming that the intron po-

sitions are similar between the two genes.

These strategies enabled isolation of a human

genomic clone containing part of the PTHrF gene

sequence in addition to a considerable 5’ flanking

sequence, including the regions that should contain

promoter activity.

MATERIALS AND METHODS

(a) Materials

Restriction endonucleases, T4 polynucleotide

kinase and PolIk were purchased from Pharmacia.

B&I, nick-tr~slation kits and T4 RNA ligase were

purchased from Bethesda Research Laboratories,

the sequenase kit was obtained from United States

Biochemical Corporation, and the [ a-32P]dATP,

[E-~~S ]dATP and [ Y-~~P]ATP from Amersham

International.

(b) Analysis of cDNA clones

From the BEN cell cDNA library described else-

where (Suva et al., 1987) we isolated three cDNA

clones, pBRF50, pBRF52 and pBRF61. The nucle-

otide sequence of the 5’ ends was obtained after

subcloning the EcoRI insert from ,%gtlO into pUC19

as described (Suva et al., 1987).

(c) Screening of human genomic library

A human genomic library containing Suu3AI frag-

ments cloned into /ZEMBL3 was screened by the

plaque hybridization method (Maniatis et al., 1982)

with a nick-translated 1103-bp XhoI fragment of a

PTHrP cDNA clone, pBRF50, corresponding to the

entire coding region of PTHrP (Suva et al., 1987).

Approximately 5 x lo5 plaques of the human

genomic library were screened with a-32P-labelled

cDNA pBRF50 (Suva et al., 1987); hybridization

was performed in 50% form~de, containing

6 x SSC, 0.1% SDS and 1% non-fat dry milk at

42°C. Washing was performed in 2 x SSC contain-

ing 0.1 y0 SDS at 42°C. Positive plaques were

rescreened using oligo probes labelled with

[ Y-~~P]ATP by T4 polynucleotide kinase. Hybridi-

zation conditions were the same as described for the

cDNA probes above, except that washing was per-

formed in 6 x SSC containing 0.1% SDS at 42°C.

(d) Cbromosomal localization

A probe consisting of a 900-bp fragment of PTHrP cDNA clone pBRF50 (Suva et al., 1987) was nick-

translated using three tritiated dNTPs to a specific

activity of 8.4 x IO’ cpm/pg. The probe, at a concen-

tration of 200 ng/ml, with 1200-3600 times this con-

centration of carrier DNA was hybridized in situ to

chromosome slides from three individuals banded

with 5-bromodeoxyuridine by the method of Buckle

and Craig (1986). 75 % of the slides were acetylated

by the method of Pardue (1985). Acetylation did not

affect the slides in this case. The slides were

stringently rinsed with 50% formamide (Donlon,

1986) dipped in Ilford L4 emulsion, and exposed for

17 days.

(e) Northern-blot analysis

Poly(A) * RNA and poly(A)- RNA was prepared

from BEN cells as described (Suva et al., 1987),

electrophoresed through 1 y0 agarose-formaldehyde

gels and transferred to nitrocellulose (Maniatis et al.,

1982). Blots were hybridized in a 20% formamide

solution (Suva et al., 1987) at 42°C and washed at

42°C in 0.2 x SSC containing 0.1% SDS.

RESULTS AND DISCUSSION

(a) Analysis of cDNA clones

We have previously isolated three cDNA clones,

pBRF50, pBRF52 and pBRF61, from a human lung

cancer cell line (BEN) cDNA library and reported

that they contained identical coding and 3’-UTRs

(Suva et al., 1987). Further analysis revealed, how-

ever, that one of these clones, pBRF52, was clearly

divergent at the 5’ end (Fig. 1). Presumably,

pBRF52 was the result of alternate splicing in the

5 ’ -UTR or due to a cDNA cloning artefact. To

address this question, we analyzed the genomic

sequence of the PTHrP gene in the region upstream

from the translation start codon.

(h) Isolation of genomic clones containing the

5’4mtranslated region of PTHrP

From a genomic library, two positive plaques were

isolated and rescreened with kinase-labelled oligos

complementary to the 5’-UTR of pBRF61 (Fig. 1;

pBRF61 underlined: 5’-GGGCTGGGTTGCTT-

CCGGAAAGTTG-3’) and the 3’ coding region

corresponding to residues Ser-14 to His-25 of

PTHrP (Suva et al., 1987; 5’-CAGGTAGGTT-

CTAAATGCCGCTGCTAAGAAGGAAGT-3’).

Both plaques hybridized with the 5’-UTR pBRF61

oligo, but not the 3’-coding region oligo. Subsequent

restriction endonuclease and Southern-blot analyses

demonstrated that these clones contained an identi-

cal 15-kb genomic fragment, that did not include the

pBRF52 -265

-243

-212

-181

-150

pBRF52 -119 pBRF61

pBRF52 -88

pBRF61

pBRF52 -57

pBRF61

pBRF50

pBRF52 -26 pBRF61

pBRF50

91

RRRGRRGCTGRCTTCRGRGGGG

GARACTTTCTTCTTTTRGGRGGCGGTTRGCC

CTGTTCCRCGRRCCCRGGAGRRCTGCTGGCC

RGRTTRATTRGRCATTGCTRTGGGRGRCGTG

TRRRCRCACTRCTTRTCRTTGflTGCATATAT

RRRRCCRRTTTTRTTTTCGCTRTTRTTTCRG TGGCCCTCCTGRTCCGRCGRCRCRCGC

AGGTTTTGCTCTTTCTGGCTGTGTGGTTTGG

RCTTGRRRCTTGTTCTCRGGGTGTGCTGGRR

RGRRRGCRCRGTTGGRGTRGCCGGTTGCTRR TCRRCTTTCCGGRRGCRRCCRGCCCRCCRGR

RCTTTCCGGRRGCRRCCRGCCCRCCRGR

RTRR'GTCCCGRGCGCGRGCGGRGRCGRTGCR GGRdGTCCCGRGCGCGRGCGGRGRCGkR GGRdGTCCCGRGCGCGRGCGGRGRCGECR

-1-

Fig. 1. Alternative 5’-untranslated region (5’-UTR) of human

PTHrP cDNA. Boxed region indicates the homology between

pBRF50, pBRF52 and pBRF61. The sequence of pBRF52

diverges from the other sequences at the position indicated by

the arrow. The oligos corresponding to pBRF6 1 and pBRF52 are

underlined and correspond to oligos 1 and 2 in Fig. 2, respec-

tively. The nt numbering begins at nt -1 (counted from the start

AUG codons; underlined).

3’ -coding region. Within this genomic region, a

2.1-kb BarnHI-Sal1 fragment, located at the 3’ end

of the genomic DNA insert (Fig. 2), was found to

hybridize with both the cDNA probe, pBRF50 and

the 5’-UTR-specific oligo of pBRF61 (Fig. 1). Thus,

the genomic fragment contained only the 5’ end of

the gene.

Another oligo (5’-TAAACACACTACTTATC,

ATTGATGC-3’; Fig. l), based on the differing

5’-UTR sequence of pBRF52, was also synthesized

and hybridized with the genomic clone, demon-

strating that both 5’-UTR sequences are present in

the PTHrP gene region. Therefore, the PTHrP cDNA

clones are probably products of alternative splicing

as their 5’-UTR sequence is contained within the

PTHrP gene.

(c) Chromosomal localization

By in situ hybridization we have assigned the

PTHrP gene to chromosome 12 (Fig. 3). Grains

scored over all chromosomes, mainly from several

females, gave an obvious peak over the pale sub-band

12~11.2 on the short arm of chromosome 12

(Fig. 3a).

Ikb n

Fig. 2. Structure of the 5’ region of the PEfrP gene. PTHrP-coding region begins with the ATG in exon 2 (see Fig. 4). Restriction map showing relevant sites is indicated at the top; restriction sites are indicated by: B, BarnHI; BL, BaZI; ET EcoRI; H, ZCzdIII; Sal, S&I; S, Smni. The S&I site comes from IEMBL3. The positions of the exons in the 5’ terminai region are illustrated below, along with the potential alternate splicing mechanism, that results in the variable 5’-IJTR of PZ’HrP cDNA. ‘Ihe sequencing oligos numbered 1 through 2 (see Fig. 1) and 3-8 are shown in the relative positions around the intron-exon boundaries. Arrows indicate the direction of sequence extension. The positions of the introns are indicated by A, B and C (singie line).

Scoring over good quality prophasic chromosome

12 from ah three subjects also showed a peak over sub-band 12~11.2 (Fig. 3b). Of 150 grams over a target region 12~13.3 to 12q13.3, 61% were over sub-bands 12~11.1 to 12~12.2 and this segment we interpret as the location of the PTHrP gene.

The conservation of the position of the intron-exon boundaries of PTHVP with PTH, together with their similar biological activity supports our earher sug- gestion of an evolution~y relationship between PTH and PTHrP. The sequence around the in~on-exon boundaries in the PTHrP gene, believed to be involved in splicing are consistent with those reported by Mount (1982).

Mangin et al. (1988) localized the gene which they called parathyroid hormone-like peptide to sub- bands 12~11.2 to 12~12.11, a slightly narrower range than that reported here. The peak of grains obtained by Mangin et al. (1988), over the pale sub-band 12~11.2 is identical to the peak presented here (Fig. 3b). We both have essentially obtained the

same chromosomal localization, suggesting that sub- baud 12~11.2 may be considered to be the point location of the PTNrP gene. The remarkably similar result obtained by both groups is an example of the accuracy of the in situ method of gene localization.

Human chromosomes 11 and 12 are similar in size, centromere index and banding pattern. The PT’ and PTHrP genes join a number of related pairs of genes, one being on chromosome 11 (Mannens et al., 1987) and the other on chromosome 12. Chro- mosomes 11 and 12 are thought to have a common origin, perhaps in tetraploidy (Comings, 1972). The chromosomal relationship of PTHrP and PTH genes suggests that they have a common evolutionary origin through duplication of the two chromosomes.

(d) Genomic sequence analysis

The nudeotide sequence of the human genomic clone as illustrated in Fig. 4 was determined by dideoxy sequencing (Sanger et al., 1980) of frag-

99

26r I b

p 12 q Fig. 3. Chromosomal localization of the PTHrP gene. (a) Score

of grains over all chromosomes showing a peak of 30 grains over

sub-band 12~11.2 with a background of no more than five grains

over the other chromosomes. The location on chromosome 11 of

the gene coding for PTH, band 11~15, is shown. The low peak

of grains over the long arm of chromosome 17 is seen on most

of our in situ hybridization slides (Board et al., 1988) and is

regarded as an artefact. (b) Score of grains over prophasic chro-

mosome 12. Of the grains 61 y0 is over a large region consisting

of the short arm and the proximal quarter of the long arm over

sub-bands 12~11.2 and 12~11.1 and the peak of grams is over

sub-band 12~11.2.

ments subcloned into Ml3 bacteriophage and

pGEM vectors (Tabor and Richardson, 1987).

The genomic sequence was obtained using a

battery of oligo primers, the relative positions of

which are shown in Fig. 2. For exon 2, encoding the

signal peptide domain, the primers were based on the

sequence of PTHrP cDNA in the region surrounding

the intron-exon boundaries as predicted from the

PTH gene (Vasicek et al., 1983). For exons la and

lb, the sequencing primers were based on the

alternative 5’-UTRs which had been found in

PTHrP cDNA clones (Fig. 1).

(e) Characterization of the promoter region

The position of the exons in the 5’ end of the

PTHrP gene was determined by comparison with the

cDNA sequences of clones pBRF52 and pBRF61

and the intron-exon junctions of the homologous

PTH gene (Vasicek et al., 1983). The 5’ terminus of

the PTHrP gene was found to contain three exons,

separated by two introns (Fig. 2). The 5’ region of

the PTHrPgene comprises, in order, exon la (at least

900 bp), intron A (60 bp), exon lb (283 bp), intron B

(165 bp), exon 2 (123 bp) and intron C (approx.

1 kb) before the Sal1 site in the cloning linker termi-

nates the genomic insert. Intron B interrupts the

cDNA sequence 22 bp upstream from the initiating

ATG and corresponds precisely to the point where

the cDNA clones pBRF52 and pBRF61 diverge

(Fig. 1). It should be noted that the exon sequences

in the PTHrP gene agree exactly with those for the

cDNA derived from BEN cells (Suva et al., 1987).

The promoter region of the PTHrP gene appears

to contain two transcription start points that could

be utilized for the efficient and correct transcription

by RNA polymerase II (Breatnach and Chambon,

1981). We believe that one ‘TATA’ consensus

sequence (promoter B, Fig. 4) is located in in&-on A,

-22 bp upstream from the putative exon lb cap site

(Fig. 4), as expected. This putative promoter region

(B), would generate mRNA transcripts equivalent to

cDNA clone pBRF52 (Fig. 1). By inference, from

our characterization of divergent cDNA clones,

another transcription start point (promoter A,

Fig. 5) is presumably located some distance 5’ of the

exon la cap site. The exact position of this putative

100

-1610

-1525

-1440 C

atatgaaatctggaaCTGTTTTTGTTCTTLTAAGCAAAAGATCTCCCTCTCTCTAGCCGATGCTCCCCACTCAGTTC-ATCCCGGG

-1355

-1270

*** AATGGGCCAGGGAGGAAGGTTCTCATGCATCGCCCCGAGCTGCCAGGCGACCTTCGGGCTCCTTAAATTCACAGGCCAACAGCCC -

GCGTCCTCTCCGCGCAGGCTCCCGGTTGCCCGCGGTCCCCGGCCCAGCTCCTTGGCCTCCTCCTCGTCGGTCCGCCCCTGGTGGT

-1185 CTTGGCGCCCGCTCGTCCAGCTCGGCGCGCCGGGGACCGCCGGCTGCCCGGGGCAGTCCGCACGCCCTCGGGGATCTCGGCTCCC

-1100

-1015

- 930

- 845

G C GG C C GGATCCGCCGCGCCGGCAGGAGCCGGCCGGGCCTGGAGGGAGCAAGCGGATCGCCACGCCCCCGGCACGGATGGGCGACAGGGCC -

w C GGGCTCCGGGGTGGGGCTCGGCAGAGCTCCTGACAGCTCCGGGGTCGGCAGCG'GGGAGGGGGGAGCTCCGCCGCTCGCCGCTCAT

CG C TCCCGGCTCGGGGCTCCCCTCCACTCGCTCGGGCGGCGCGGGGCCCGTTGCGCCGCCCGTCGCGCCCCCGCCCCCCGCGCGCCCG

. . ...*

T C TCGCT CCG CCCGCCAGCCCGCCCGTGCCCGCTTCGCCCCGCGCGCGTTCCTAGGGCGCCACCTCTTTGCGACTAGCTCACTTCTAGCAGGTTT

- 760 cc CCCG

GCCTCGGAGCGTG~ACATTCCTCCGCTCGGTTTTCAACTCGCCTCCAACCTGCGGCCCGGCCAGCATGTCTCCGCCCGTGAAG -

- 675 +Lr*

CGGGCTCGCCTCCCTGGCTCCGCCGCCACTAACGACCCGCCCTCGCC~~~CCTGGCCCTCCTGATCGACGACACACGCACTTGA . ..*..

- 590

- 505 gtgggtttcgctacaagtggctctggaacgaaagGGCCTGGTTCGCAAAGAAGCTGACTTCAGAGGGGGAAACTT~CTTCTTTTA

- 420 GGAGGCGGTTAGCCCTGTTCCACGAACCCAGGAGAACTGCTGGCCAGATTAATTAGACATTGCTATGGGAGACGTGTAAACACAC

- 335 TACTTATCATTGATGCATATATAAAACCATTTTATTTTCGCTATTATTrCAGAGGAAGCGCCTCTGATTTGTTTCTTTTTTCCCT

- 250 TTTTGCTCTTTCTGGCTGTGTGGTTTGGAGAAAGCACAGTTGGAGTAGCCGGTTGCTAAATAAgtaagtgctgagaggctccaga

- 165

- 80

+ 6

f- 91

+ 176

+ 261

+ 356

gaaattttttttcttttcaacttgggagatgcccttgatgttgaagaggcttttt~agagcgggctaaaaagggggagcggagta

-lM Q gtgcggggagatggagagtcctgactgacacctcgggtcccattcccttctgttgcagGTCCCGAGCGCGAGCGGAGACGATGCA

R K L V QQWSVAVFLLSYAVPSCGKSVEGL GCGGAGACTGGTTCAGCAGTGGAGCGTCGCGGTGTTCCTGCTGAGCTACGCGGTGCCCTCCTGCGGGCGCTCGGTGGAGGGTCTC

s K K L AGCCGCCGCCTgtaagtcccccatcctccccagggcgccgg~ttggggaggccagggggaggggctgccaagctgggatgctgcc

gaggcgttgcagcggtcaccgatcgtccttgcccgggttagggagagggaccatcccgcatacctgccgggcctgagccgttctc

aaacctggcaggagaactggttgatcttcaaccggagacaggcaagagagagactttatgtgtgtttccataa~agggagctttc

acagaatctcttctagggaaagatccttgcctcta

Fig. 4. Nucleotide sequence ofthe 5’ region ofthe PTHrP gene. The nt are numbered (-1610 to -1) from the start ATG codon (A = + 1).

The positions of the exons are represented in upper-case letters, the introns in lower-case letters and the potential TATA sequences

are boxed; putative promoter A at -1558 bp and promoter B at -5 14 bp. The positions of the upstream ATGs are underlined and the

in-frame stop codons are indicated by stars. The Sp-1 binding site and its inverted repeat are indicated by dots. The bases given as

superscripts represent the differences between IHHM-8 (Mangin et al., 1988) and the genomic sequence.

101

(i)

(ii)

(iil)

(iv)

Fig. 5. Alternative transcripts and splicing of the PTHrP gene. The relative positions of the two promoter regions A and B, as well as

the exons and introns in the gene are as indicated. The position of the ATG start codon and the N-terminal alanine of the mature protein

are indicated. The four species of PTHrP mRNA predicted from this gene structure are numbered (i) through (iv) (Suva et al., 1987;

Thiede et al., 1988; Mangin et al., 1988). Signal refers to the leader sequence of PTHrP that is cleaved off to release the mature protein.

transcription start point has not yet been accurately

mapped.

We predict that pBRF52 is derived by initiation

from promoter B, rather than by initiation of tran-

scription upstream with some alternate splicing

mechanism. Consistent with this hypothesis are

recent data from Thiede et al. (1988), who analyzed

cDNA clones from a human renal carcinoma cell

line (786-O) and found by sequence analysis and

primer extension only a single type of 5’-untranslated

sequence for PTHrP mRNA, that would be initiated

15-17 bp downstream from promoter B. This has

now been confirmed by our own Sl nuclease

mapping and primer extension studies that have lo-

calized the 5’ end of the mRNA (data not shown).

The nucleotide sequence of the 5’-UTR of 786-O

mRNA is identical to our cDNA clone, pBRF52

(Fig. 1).

Evidence for initiation at the proposed promoter

A region (Fig. 5) arises from the BEN cell cDNA

clones (Suva et al., 1987) since this is the only way

to generate cDNA clones pBRF50 and pBRF61.

Further support for the existence of this 5’ tran-

scription start point comes from the work of Mangin

et al. (1988) who identified a cDNA clone

(LHHM-8) from a renal carcinoma cell line

(SKRC-1) which possessed a 5’-untranslated

sequence, identical to pBRF61, but was extended a

further 826 bp upstream. We believe that pBRF61 is

equivalent to AHHM-8 since their 5’-untranslated

sequences are identical, with the exception of the

nucleotides indicated in Fig. 4. Presumably, se-

quences defining promoter A are located 5’ to the

AHHM-8 cDNA sequence as depicted in Fig. 4; it is

possible that the CAAAAA sequence (Corden et al.,

1980) 26 nt 5’ to the AHHM-8 cDNA could function

as promoter A. These data support the conclusion

that the production of PTHrP mRNA species with

different 5’-UTRs occurs by initiation from at least

two separate promoter regions.

Upstream from promoter regions A and B

(Fig. 5) we would expect to find the enhancer ele-

ments or control elements responsible for the regu-

lation of the rate of transcription. These enhancer

elements can act on promoters at great distances in

an orientation-independent manner and can also act

downstream from the transcription unit (Dynan and

Tjian, 1985). In the region of putative promoter A

there are two ‘GCAAT’ sequences located at

-1449 bp to -1445 bp and -1537 bp to -1533 bp.

This sequence has recently been shown to have the

same enhancer activity as the consensus ‘CCAAT

enhancer sequence in a number of systems

(Hatamochi et al., 1988).

102

The position of promoter B as proposed in Fig. 5

is supported and has been partially characterized by

the primer extension analysis of Thiede et al. (1988).

There are no apparent consensus enhancer se-

quences within exon la which immediately precedes

promoter B. However, exon la does have a high

G + C content of around 70% and contains one

‘GGGCGG’ sequence, which is the core sequence

that binds the DNA-binding protein Sp-1 in vitro

(Dynan and Tjian, 1985) and is essential for the

function of the SV40 promoter (Dynan and Tjian,

1983). The ‘CCGCCC’ inverted repeat of the Sp-1

binding sequence is located at -638 bp, near TATA

lb. A 9-nt repeat ‘AGCGGAGAC’ of unknown

function is located symmetrically at either side of the

initiating AUG. It is possible that this sequence plays

some role in aiding the identification of the correct

initiating AUG.

(f) Alternate mRNA species

To investigate whether the alternate cDNA

species isolated were representative of functional

BEN cell mRNAs, we hybridized BEN cell

poly(A) + RNA with probes derived from the dif-

fering 5’-UTR of pBRF52/pBRF61 and with probes

from the differing 3’-UTR of pBRF52/pBRF61 and

clone lOB5, PTHrP cDNA clone with a differing

3’-UTR (Thiede et al., 1988; Fig. 6).

Hybridization with the 5’-UTR-specific probes

(Fig. 6, lanes .l and 2) revealed four major species of

PTHrP mRNA (lane 1: pBRF52 5’-UTR-specific

oligo, approx. 1600 and 1300 bp, and lane 2:

pBRF61 5’-UTR-specific oligo, approx. 1550 and

1250 bp) and a number of other higher molecular-

size species, that are presumably pre-spliced inter-

mediates. Presumably, the 300-bp differences in the

size of these pairs of mRNA transcripts are attributa-

ble to alternate 3’-untranslated sequences.

To examine this possibility, probes based on the

common 3’-UTR of pBRF52/pBRF61 (lane 3) and

the differing 3’-UTR of lOB5 (Thiede et al., 1988;

lane 4) were hybridized with BEN cell mRNA. The

3’-UTR-specific probes basedon pBRF52/pBRF61,

hybridized with the 1250-1300-bp band (Fig. 6,

lane 3) and the lOB5 3’-UTR-specific probe hybrid-

ized with the 1550-1600-bp band (Fig. 6, lane 4).

Thus, BEN cells produce four major PTHrP

mRNA species which are alternately spliced at both

Fig. 6. Northern gel analysis of BEN cell mRNA. Poly(A) + and

poly(A)- BEN cell RNAs were prepared as described previously

(Suva et al., 1987). A total of 5 pg of poly(A))RNA and 2.5 pg

poly(A) + RNA per lane was run on a 7 y0 formaldehyde denatur-

ing - 1 y0 agarose gel at 100 V for 3 h, and transferred to nitro-

cellulose (Maniatis et al., 1982). Size markers were 28s and 18s

ribosomal RNA. Four specific end-labelled 50-mer oligos were

hybridized in 20% formamide at 42°C 16 h (Suva et al., 1987).

The filters were washed twice in 0.2 x SSC, 0.1% SDS at 42°C.

Small arrows indicate specific hybridization signals detected

with the oligos. Lanes 1: obrf. 15.1(5’-ATTTAGCAACCGGCT-

ACTCCAACTGTGCTT’TC’TCCAAACCACACAGCCAG-3’)

5’ end of pBRF52, 1600-bp and 1300-bp bands; 2, obrf.15.2

(5’-CCTCTGGTGGGCTGGTTGCTTCCGGAAAGTTGA-

TTCCAGCACACCCTGAG-3’) 5’ end of pBRF61, 1550-bp

and 1250-bp bands; 3, obrf.15.3 (5’-ATCCTGCAATATGTC-

CTTGGAAGGTCTCTGCTGAAAATTTCAATGCCTCC-

3’) 3’ end ofpBRF52/pBRF61, detects the 1250-bp and 1300-bp

transcripts which are not resolved; 4, obrf.15.4 (5’-CCCAGC-

TGAGAGCACCCCGCTGAGGCTACGGGCCAGAGA-

AGCCTGTTACC-3’) alternate PTHrP cDNA, lOB5 3’-UTR

(Thiede et al., 1988) detects the 1550-m and 1600-nt transcripts

which are not resolved.

the 5’ and 3’ ends. Clearly, the two separate PTHrP

promoters produce multiple mRNA species.

(g) Conclusions and discussion

Although at present there are no data on the regu-

lation of PTHrP gene expression, the evidence

103

presented here for two promoter regions raises many

possibilities. The two promoters could confer tissue

specificity on the expression of PTHrP in a way

analogous to the expression of the mouse cc-amylase

gene (Hagenbuchle et al., 1981; Young et al., 1981).

In this situation, alternative mRNAs specific for

particular tiss,ues differ only in their 5’-UTRs, leav-

ing the protein-coding sequences unaffected. The

alternative splicing of a single gene serves as a poten-

tial mechanism for developmental and tissue-specific

gene expressi’on (Leff et al., 1986). Support for this

idea comes from the data of Shen et al. (1988) who

demonstrated1 alternative 5’-UTRs in the human

insulin-like growth factor gene.

Our cDNA cloning data suggest that the human

lung cancer cell line, BEN, produces at least two

types of mRNA encoding PTHrP, with different

5’-untranslated sequences as depicted schematically

in Fig. 5. In renal carcinoma cell lines (Mangin et al.,

1988; Thiede et al., 1988) several species of mRNA

are produced that can be accounted for simply by

divergent 3’-UTRs, according to the cDNA struc-

tures reported. In one case (Mangin et al., 1988) the

5’-UTR was, identical to pBRF61, in the other

(Thiede et al., 1988) it was identical to pBRF52. This

suggests that either of the promoters is able to pro-

duce alternate mRNA species. Mangin et al. (1988)

present a cDNA structure corresponding to pro-

moter A transcription that results in an unusually

long 5’-UTR. There are some minor differences

(Fig. 4) compared to our genomic sequence that

could result from allelic variations or sequencing

errors. The cDNA sequence derived by Thiede et al.

(1988) corresponds to promoter B transcription.

In the human PTHrP gene (Fig. 4) the 5’ UTR

from promoter A is unusually long and suggests that

a novel mechanism for the control of PTHrP gene

expression may exist. This 5’-UTR contains only

minor sequence variation to the cDNA clone of

Mangin et al., (1988). Several AUGs are located

within this long 5’-untranslated sequence, each of

which is followed by in-frame stop codons. None of

these upstream AUGs conform to the consensus

sequence proposed by Kozak (1983) for utilization

of the eukaryotic mRNA translation. The role of long

5’-untranslated sequences containing unused up-

stream AUGs or secondary structure loops in

mRNA translation is unclear. In other genes the

presence of these upstream AUGs is relatively rare,

but they have recently been suggested to be involved

in the control of translation in both yeast GCN#

(Mueller et al., 1986) and mouse bcl-2 genes (Negrini

et al., 1987). Presumably, to be consistent with the

scanning model of translation (Kozak, 1981), ribo-

somes must first bind at the cap site and move down-

stream reading through several AUGs without

initiating translation at any of these sites. However,

this model provides an opportunity for translational

control since the ribosomes could be retarded when

encountering potential initiation sites. In the PTHrP gene, the existence of alternative promoters together

with a mechanism for translational control presents

a complex picture of gene regulation for this

important hormone in calcium metabolism. In sup-

port of this concept, the Northern-blot analysis of

Mangin et al. (1988) shows multiple-sized tran-

scripts of PTHrP mRNA, which is similar to our

earlier published data (Suva et al., 1987).

Neither of our genomic clones contain the 3’ end

of the PTHrP gene. However, the data of Thiede

et al. (1988) showed clearly that 3’ alternate mRNA

splicing also occurs and suggests that the transcrip-

tional regulation of the PTHrP gene is extremely

complex and may involve an additional intron to that

of the PTH gene. This intron is believed to interrupt

the PTHrP mRNA at Arg-139, 9 bp upstream from

the stop codon of full-length PTHrP mRNA (Suva

et al., 1987). Further support for the 3’ alternate

splicing of PTHrP mRNA comes from Mangin et al

(1988) who have found another cDNA clone that

encodes a longer peptide with a different C terminus.

However, neither of these workers consider the pos-

sibility of the existence of two distinct promoter

regions, resulting in the expression of multiple forms

of PTHrP mRNA with distinct 5’-UTRs (Fig. 5).

The putative alternate splicing mechanisms of the

PTHrP gene are illustrated in Fig. 5. This figure

shows the possible mechanism for the production of

all PTHrP mRNA species so far described (Fig. 5(i),

(ii), (iii), (iv); see Fig. 6). The alternative promoters

and splicing mechanisms may explain, at least in

part, the multiple mRNA species of PTHrP observed

in Northern gel analyses (Suva et al., 1987; Mangin

et al., 1988; Thiede et al., 1988). Northern analysis

(Fig. 6) using specific 5’-UTR and 3’-UTR oligos

confirms that two promoters are functioning in BEN

cells. These data are suggestive of another level of

complexity, in that the cDNA clone of Mangin et al.

104

(1988) may represent a pre-spliced mRNA inter-

mediate, due to the shorter mRNA transcripts pro-

duced from the upstream promoter (Fig. 6, lane 2).

Examples of genes producing mRNAs with

heterogeneous 3’-non-coding regions as well as mul-

tiple promoter sites are very rare, although the mouse

dihydrofolate reductase gene is one such example

(Setzer et al., 1982). The function of alternative 3’-

noncoding regions in the regulation of gene expres-

sion remains unclear, although Miyata et al. (1980)

and Yaffe et al. (1985) would suggest that 3’-non-

coding sequences perhaps play some role in develop-

mental or tissue-specific gene expression.

Recent work in our laboratory has demonstrated

that there is PTH-like activity in extracts of ovine

placenta and that PTHrP promotes placental cal-

cium transport in the intact ovine foetus (Rodda

et al., 1988). As yet, it is unknown whether one or

more types of PTHrP mRNA are produced in

placental tissue. Further studies are required to

determine the sites of PTHrP expression in both

normal and abnormal tissues.

In conclusion, this study represents the first iso-

lation of the PTHrP gene for any species. One regula-

tory domain has been identified in the human PTHrP gene, and the existence of further domains is strongly

suggested by evidence presented in this paper and

reviewed. The actual role of these promoter regions

and the possible tissue specificity will be clarified

only by the functional analysis of gene activity. These

studies are currently in progress.

ACKNOWLEDGEMENTS

We gratefully acknowledge Dr. Ulrike Novak for

providing us with the human genomic DNA library

and thank Dr. David Findlay for valuable dis-

cussion. This work is supported by grants from the

National Health and Medical Research Council of

Australia, the Commonwealth Department of

Veterans’ Affairs and the CSIRO/Melbourne Uni-

versity Collaborative Project Fund. M.T.G. is the

recipient of a J.M. Higgins Research Fellowship.

REFERENCES

Board, P.G., Webb, G.C., McKee, J. and Ichinose, A.: Localisa-

tion of the coagulation factor XIIIA subunit gene (F13A) to

human chromosome bands 6~24-25. Cytogenet. Ceil Genet.

(1988) in press.

Breatnach, R. and Chambon, P.: Organization and expression of

eukaryotic split genes coding for proteins. Annu. Rev.

Biochem. 50 (1981) 349-383.

Buckle, V.J. and Craig, I.W.: In situ hybridization. In Davies,

K.E. (Ed.), Human Genetic Diseases - A Practical

Approach, IRL Press, Oxford, 1986, pp. 85-100.

Comings, D.E.: Evidence for ancient tetraploidy and conserva-

tion of linkage groups in mammalian chromosome. Nature

238 (1972) 455-457.

Corden, J., Wasylyk, B., Buchwalder, A., Sassone-Corsi, P.,

Kedinger, C. and Chambon, P.: Promoter sequences of

eukaryotic protein-coding genes. Science 209 (1980)

1406-1414.

Donlon, T.A.: Practical approaches to in-situ hybridization.

Karyogram 12 (1986) 3-10.

Dynan, W.S. and Tjian, R.: The promoter-specific transcription

factor SP-1 binds to upstream sequences in the SV40 early

promoter. Cell 35 (1983) 79-87.

Dynan, W.S. and Tjian, R.: Control of eukaryoiic messenger

RNA synthesis by sequence-specific DNA-binding proteins.

Nature 316 (1985) 774-778.

Hagenbuchle, O., Tosi, M., Schibler, U., Borey, R., Wellaver,

P.K. and Young, R.A.: Mouse liver and salivary gland

a-amylase mRNA differ only in their 5’untranslated

sequences. Nature 289 (1981) 643-646.

Hatamochi, A., Golumbek, P.T., Van Schaftingen, E. and de

Crombrugghe, B.A.: CCAAT DNA binding factor consisting

of two different components that are both required for DNA

binding. J. Biol. Chem. 263 (1988) 5940-5947.

Kemp, B.E., Moseley, J.M., Ebeling, P.R., Wettenhall, R.E.H.,

Stapleton, D., Diefenbach-Jagger, H., Ure, F., Michelangeli,

V.P., Simmons, H.A., Raisz, L.G. and Martin, T.J.: Para-

thyroid hormone-related protein of malignancy: active

synthetic fragments. Science 238 (1987) 1568-1570.

Kozak, M.: Possible role of flanking nucleotides in recognition

of the AUG initiator codon by eukaryotic ribosomes. Nucleic

Acids Res. 9 (1981) 5233-5252.

Kozak, M.: Comparison of initiation of protein synthesis in

prokaryotes, eukaryotes and organelles. Microbial. Rev. 47

(1983) l-45.

Leff, S.E., Rosenfeld, M.G. and Evans, R.M.: Complex tran-

scriptional units: diversity in gene expression by alternative

RNA processing. Annu. Rev. Biochem. 55 (1986) 1091-l 117.

Mangin, M., Webb, A.C., Dreyer, B.E., Posillico, J.T., Ikeda, K.,

Weir, E.C., Stewart, A.F., Bander, N.H., Milstone, L.,

Barton, D.E., Francke, U. and Broadus, A.E.: Identification

of a cDNA encoding a parathyroid hormone-like peptide

105

from a human tumor associated with humoral hyper-

calcaemia of malignancy. Proc. Natl. Acad. Sci. USA 85

(1988) 597-601.

Maniatis, T., Fritsch, E.F. and Sambrook, J.: Molecular Cloning.

A Laboratory Manual. Cold Spring Harbor Laboratory, Cold

Spring Harbor, NY, 1982.

Mannens, M., Slater, R.M., Heyting, C., Guerts van Kessel, A.,

Goedde-Saiz, B., Fran& R.R., Van Ommen, G.J.B. and

Pearson, P.L.: Regional localization of DNA probes on the

short arm of chromosome 11 using aniridia-Wilms’ tumor-

associated deletions. Hum. Genet. 75 (1987) 180-187.

Martin, T.J. and Atkins, D.: Biochemical regulators of bone

resorption and their si~ificance in cancer. Essays Med.

Biochem. 4 (1979) 49-82.

Miyata, T., Yasunoga, T. and Nishida, T.: Nucleotide sequence

divergence and functional constraint in mRNA evolution.

Proc. Natl. Acad. Sci. USA 77 (1980) 7328-7332.

Moseley, J.M., Kubota, M., Diefenbach-Jagger, H.D., Wetten-

hall, R.E.H., Kemp, B.E., Suva, L.J., Rodda, C.P.,

Ebeling, P.R., Hudson, P.J., Zajac, J.D. and Martin, T.J.:

Parathyroid hormone-related protein purified from a human

lung cancer cell line. Proc. Natl. Acad. Sci. USA 84 (1987)

5048-5052.

Mount, S.M.: A catalogue of splice junction sequences. Nucleic

Acids Res. 10 (1982) 459-472.

Mueller, P.P. and Hinnebusch, A.G.: Multiple upstream AUG

codons mediate translational control of GCN4. Cell 45 ( 1986)

201-207. Negrini, M., Silini, E., Zozak, C., Tsujimoto, Y. and Croce C.M.:

Molecular analysis of mbcl-2: structure and expression of the

murine gene homologous to the human gene involved in

follicular lymphoma. Cell 49 (1987) 455-463.

Pardue, M.L.: In situ hybridization. In Hawes, B.D. and Higgins,

S.J. (Eds.), Nucleic Acid Hybridization - A Practical

Approach, IRL Press, Oxford, 1985, pp. 179-202.

Powell, D., Singer, F.R., Murray, T.M., Minkin, C. and Potts Jr.,

J.T.: Non-parathyroid humoral hypercalcaemia in patients

with neoplastic disease. N. Engl. J. Med. 289 (1973) 176-181.

similar protein implicated in humoral hypercalcaemia of

malignancy. J. Endocrinol. 117 (1988) 261-271.

Sanger, F., Coulson, P.R., Barrel& B.G., Smith, A.J.H. and Rot,

B.A.: Cloning in single stranded bacteriophage as an aid to

rapid DNA sequencing. J. Mol. Biol. 143 (1980) 161-178.

Setzer, D.R., McGrogan, M. and Schimke, R.T.: Nucleotide

Rodda, C.P., Kubota, M., Heath, J.A., Ebeling, P.R., Moseley,

J.M., Care, A.D., Caple, I.W. and Martin, T.J.: Evidence for

a novel parathyroid hormone-related protein in fetal lamb

parathyroid glands and sheep placenta: comparisons with a sequences surrounding multiple polyadenylation sites in the

mouse dihydrofolate reductase gene. J. Biol. Chem. 257

(1982) 5143-5147.

Shen, S.J., Daimon, M., Wang, C.Y., Jansen, M. and Ban, J.:

Isolation of an insulin-like growth factor II cDNA and a

unique 5’ untranslated region from human placenta. Proc.

Natl. Acad. Sci. USA 85 (1988) 1947-1951.

Stewart, A.F., Horst, R., DefIos, L.J., Cadman, E.C., Lang, R.

and Broadus, A.E.: Biochemical evaluation of patients with

cancer-associated hypercaleaemia. Evidence for humoral

and non-humoral groups. N. Engl. J. Med. 303 (1980)

1377-1383.

Stewart, A.F., Insogna, K.L., Goltzman, D. and Broadus, A.E.:

Identification of adenylate cyclase-stimulating activity and

cytochemical glucose-6-phosphate dehydrogenase-stimulat-

ing activity in extracts of tumors from patients with humoral

hypercalcaemia ofmalignancy. Proc. Natl. Acad. Sci. USA 80

(1983) 1454-1458.

Suva, L.J., Winslow, G.A., Wettenhall, R.E.H., Hammonds,

R.G., Moseley, J.M., Diefenbach-Jagger, H.D., Rodda, C.P.,

Kemp, B.E., Rodriguez, H., Chen, E.Y., Hudson, P.J.,Martin,

T.J. and Wood, WI.: A parathyroid hormone-related protein

implicated in malignant hyperc~caemia: cloning and expres-

sion. Science 237 (1987) 893-896.

Tabor, S. and Richardson, CC.: DNA sequence analysis with a

modified bacteriophage T7 DNA polymerase. Proc. Natl.

Acad. Sci. USA 84 (1987) 4267-4271.

Thiede, M.A., Strewler, G.J., Nissenson, R.A., Rosenblatt, M.

and Rodan, G.A.; Human renal carcinoma expresses two

messages encoding a parathyroid hormone-like peptide: evi-

dence for the alternate splicing of a single gene. Proc. Natl.

Acad. Sci. USA 85 (1988) 4605-4609.

Vasicek, T.J., McDevitt, B.E., Freeman, M.W., Fennick, B.J.,

Hendy, F.N., Potts Jr., J.T., Rich, A. and Kronenberg, H.M.:

Nucleotide sequence of the human parathyroid hormone

gene. Proc. Nati. Acad. Sci. USA 80 (1983) 2127-2131.

Yaffe, D., Nudel, U., Mayer, Y. and Neuman, S.: Highly con-

served sequences in the 3’ untranslated regions of mRNAs

coding for homologous proteins in distantly related species.

Nucleic Acids Res. 13 (1985) 3723-3737.

Young, R.A., Hagenbuchle, 0. and Schibler, V.: A single mouse

cc-amylase gene species produces two different tissue-specific

mRNAs. Cell 23 (1981) 451-458.