Gerw, 98 (1991) 295-298

295

Elsevier

GENE

03902

Novel human PDGFA gene transcripts derived by alternative mRNA splicing (Polymerase

chain reaction;

Angeles Sanchez”,

platelet-derived

growth factor A; recombinant

DNA)

Colin N. Chesterman b and Merilyn J. Sleigh a

‘ICSIRO, Division of Biomolecular Engineering, Laboratory for Molecular Biology, North Ryde, NSW 2113 (Australia) and h Department of Haematology,

Prince of Wales Hospital, Universit]l of NS W, Kensington, NS W 2033 (Australia) Tel. (612)399-4528

Received by R.W. Davies: 29 June 1990 Accepted: 2X September 1990

SUMMARY

The polymerase chain reaction (PCR) has been used to amplify sequences coding for the platelet-derived growth factor A chain (PDGFA) using mRNA populations derived from two transformed cell lines (a human osteosarcoma, U-2OS, and a human glioma, U-343) and from human umbilical vein cells. The primers used for PCR were designed to amplify both of the two transcripts previously reported for the PDGFA gene. These transcripts differ from each other by the presence or absence of sequences from a sixth exon located near the 3’ end of the gene. The PCR procedure revealed not only these expected transcripts, but additional RNAs that were shown by cloning and sequencing to lack exon 2. These species were present at variable levels in the three cell types examined. We propose that this novel splicing pattern, generating mRNAs encoding truncated and non-functional polypeptides, signals an additional, post-transcriptional mechanism for modulation of PDGFA gene expression.

INTRODUCTION

Human platelet-derived growth factor (PDGF) is the principal mitogen found in serum. Purified PDGF is a cationic 30-kDa protein. It consists of a disulfide-bonded dimer of two polypeptide chains, designated A and B, which are the products of separate but related genes (Ross et al., 1986). In almost all cases where PDGFA gene transcripts have Corrcrporukw~ to: Dr. A. SBnchez, CSIRO, Engineering,

Laboratory

for Molecular

Rydc, NSW 2113 (Australia) Abbreviations: mentary ORF, human

Tel. (612)886-4972;

of Biomolecular

P.O. Box 184, North Fax (612)888-9271.

aa. amino acid(s); bp, base pair(s); cDNA,

to RNA;

kilobaae(s)

Division

Biology,

HUVE,

human

umbilical

or 1000 bp; nt, nucleotide(s);

open reading platelet-derived

frame;

factor

cells; kb,

oligo, oligodeoxyribonucleotide;

PCR, polymerase

growth

DNA comple-

vein endothelial chain reaction;

A chain;

PDGFA,

PDGFA,

gene (DNA)

encodlng PDGFA; Tnq, Thermus aquaticus; U-2OS, human osteosarcoma cell lint; LJ-343, human clonal glioma cell line MGaC12: 6.

been examined by Northern hybridization, mRNAs of 2.8, 2.3 and 2.0 kb have been observed. At least some of this variation appears to arise from alternative splicing of the primary transcript. The U-343 cell line has been shown to produce two A-chain mRNAs, the longer one containing an additional exon of 69 nt (exon 6 as defined by Rorsman et al., 1988; Bonthron et al., 1988). Translation of this mRNA yields a polypeptide terminating within the added exon 6 sequences, and 15 aa longer than the polypeptide derived from mRNA lacking exon 6 (Rorsman et al., 1988) (see Fig. 1A). In the course of experiments using PCR to isolate PDGFA coding sequences, we identified two new RNA species for PDGFA, both lacking exon 2. One of these was present in all cell types studied (U-343, U-20s and HUVE), whereas the other was detected only in U-20s cells. We propose that these shorter RNAs reflect a posttranscriptional mechanism for regulation of PDGFA gene expression.

296 EXFERIM~NTAL

AND DISCUSSION

(a) PCR amplification of PDGFA sequences of Fig. 1 shows the strategy for PCR amplification PDGFA coding sequences and Fig. 2 shows a Southern analysis of the DNA products obtained. mRNA from all

B

A 123

12345678910

*872bp

three cell types yielded a 666-bp DNA fragment (species I in Fig. 2) with A2 and Al primers, which should amplify specifically the PDGFA mRNA containing exon 6. DNA (605 bp) generated from the exon &lacking RNA by HU and Al primers is also present in all lines (species II in Fig. 2, U-343 data not shown). Two additional major products were seen: a 569-bp fragment from A2/Al primers, generated only from U-20s RNA, and a 508-bp band from HU/Al primers, in all three cell types (species III and IV, respectively, in Fig. 2). The UN1 primer in combination with Al was expected to generate DNAs from both exon &containing and exon 6-lacking RNAs simultaneously. However, only products derived from the shorter transcript were seen in these reactions (Fig. 2), suggesting that the exon &lacking transcript is the predominant species in our preparations, or that it is preferentialiy amplified. Using a 3’ primer very similar to UNI, but amplifying a much shorter region, Matoskova et al. (1989) detected exon 6-containing and -lacking transcripts in the same PCR reaction.

ATG

Ter I 1

D

2

3

-3

5

6

7

Al

cf II El----

ATG

UN1 A2 HU

TC3r I

0

0

2

3

4

S

7

Al

Cl U

UNI HU

R

Ol.IMlDE~IXY\I

C‘l.l:O-I’lIlI:S

1’SI:ll

AS I’Rl\ll~li\

\.i,X

SLLpKll‘C

11I

s’~~~~~~~\~i\CciTA.rc;:\(;(;A(‘(‘TI(;(;(‘I’I(;(‘

i’i>\lillm

A2

~‘-ct;.~~‘!‘c!‘i\?~lC;(;TT(;(iC”i

Iii;

i’~cj(,~ATc‘~C4_l~(‘,~c_nc’~i’(tr;

I \I

iXi$ii\~l

Fig. I. Strategy

t’,