Detection of ras point mutations by polymerase chain reaction using mutation-specific, inosine-containing oligonucleotide primers

Vol. 160, No. 2, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

April 28, 1989

Pages

441-447

DETECTIONOF RAS POINT MUTATIONS BY POLYMERASE CHAIN REATION USINGMUTATION-SPECIFIC, INOSINE-CONTAINING OLIGONUCLEOTIDE PRIMERS

Thomas Ehlen and Louis Dubeau’

Departments of Gynecologic Oncology and Pathology Kenneth Norris Jr. Comprehensive Cancer Center USCSchool of Medicine LosAngeles, California 90033 Received

March

2, 1989

SUMMARY: The use of DNA primers with 3’-ends complementary to specific genetic point mutations allowed for the rapid detection of such mutations in genomic DNA by polymerase chain reaction. The sensitivity of this approach was such that mutations could be detected in DNA samples mixed with a 107-fold excessof normal non-mutated DNA. To increase the practicality of this approach for the detection of point mutations affecting all 3 of the known ras oncogenes we synthesized mutation-specific primers complementary to all 3 enes by substituting inosine residues at positions corresponding to ambigous %aseson the genes. 0 1989 Academic Press, Inc.

The occurrence of single base substitutions within specific human genes is related to a variety of different pathologic conditions. Examples include hereditary diseases such as hemoglobinopathies and activation of specific proto-oncogenes such as those belonging to the ras super family (1). Point mutational inactivation of a recessive cancer gene has also been recently reported (2) and such lesions may play an important role in tumor development. The detection of single base substitutions within genomic DNA has been greatly facilitated recently by the use of PCR technology (3-41, which allows enzymatic amplification of specific DNA fragments containing potential point mutations. The advantage of this approach is that it allows molecular studies on minute quantities of crude DNA preparations.

Point mutations have been detected in the PCR products

using a variety of different approaches which include hybridization to mutationspecific oligomer probes (5), hybridization to RNA probes followed by cleavage of single base mismatches with RNase A (6). oligonudeotide ligation assay (7), hybridization to DNA probes followed by chemical cleavage of mismatched pairs (8) or

‘To whom correspondence should be addressed. Abbreviations:

PCR: polymerase chain reaction. OO!X-291X/89

441

M

$1.50

CopyrQht 6 1989 by Academic Press, Inc. rights of reproduction in any form reserved.

Vol. 160, No. 2, 1989

BIOCHEMICAL

AND BIOPHYBICAL RESEARCH COMMUNICATIONS

followed by restriction endonuclease digestion (3), denaturing gradient gel electrophoresis (9), and direct sequencing of the amplified fragments (2). The above-mentioned techniques vary in their sensitivities and are often time-consuming multistep procedures. We now report on a simple and highly sensitive approach for the detection of DNA point mutations which takes advantage of the fact that the Taql DNA polymerase enzyme, which is used in PCR studies, lacks a 3’exonuclease activity (10) and is therefore unable to repair single base mismatches at the 3’-end of DNA primers. Thus, if synthetic oligonucleotides complementary to a given genetic sequence containing a specific point mutation are used in PCR studies there should be no detectable enzymatic amplification of homologous non-mutated genes if the base complementary to the mutation is located at the 3’-end of the primers. Such primers would therefore be mutation-specific because they would only allow enzymatic amplification of genetic sequences containing the specific point mutation. MATERIALS AND METHODS Synthetic oliqonucleotides. The following oligomer is complementary to a sequence extendin? from codon 6 to the second position of codon 12 on the non-coding strand o the normal human H-ras gene: 5’-CTGGTGGTGGTGGGCGCCGG-3’. Our mutation-specific oligomer used fortheetection of the point mutation present in EJ DNA was identical except for the presence of a T instead of a G at its 3’-end. Other primers used in Fig. 2 were similar except for the presence of either A or C residues at the 3’-ends. The following primer is complementary to a sequence extending from codons 30 to 37 on the coding strand of H-s and was used in pair with any one of the above primers for PCRstudies: 5’-CTCTATAGTGGGGTCGTATTCGTC-3’. The sequence of the inosine containing primers is shown in Fig. 3. All oligomers were synthesized using a model 380A DNA sythesizer (Applied Biosystems, Foster City, CA) at the Microchemical Core Facility of the Kenneth Norris Jr. Comprehensive Cancer Center. Enzymatic amplification by PCR. The composition of the 100 pl PCR mix was essential1 as suggested b y Saiki (4) except for the concentration of nucleotides (obtaine J from Fisher-Biotech) which was 2 pM unless indicated otherwise. Each reaction mix contained 1 pg of genomic DNA. Taql polymerase was obtained from Perkin-Elmer-Cetus (Emeryville, CA). PCR was done for 30 cycles using an automatic thermal DNA cycler (Perkin-Elmer-Cetus). Machine settings for each cycle were: one min at 94X, 2 min at 55”C, and 1 min at 72°C. Analysis of PCRproducts. After completion of PCR, 20 pl aliquots of the reaction mixes were electrophoresed on gels containing 3% NuSieve and 1% Seakem agaroses (FMC, Rockland, ME). The procedures for transfer to Zeta probe (Bio-Rad, Richmond, CA) membranes, hybridization to radiolabelled DNA probes, and autoradiography have been described (11). The H-E and N-ras probes were obtained from the American Type Culture Collection (Cat #41,000 ani41.031, respectively). The K-m (12) probe was obtained from Dr. R.A. Weinberg from the Whitehead Institute, MA. Source of DNA samples. Normal DNA was extracted from normal human blood specimens using the procedure of Bell et al. (13). EJ cells were obtained from Dr. E. Stanbrid e, University of California at IrGne, CA. DNA was extracted from serum-free cultures P14) of those cells using the procedure of Marmur (15).

RESULTS Our strategy to detect single base substitutions in genomic DNA is illustrated in Fig. 1. The figure represents the non-coding strand of the first exon of the H-E gene in EJ bladder carcinoma and normal DNA. The base sequence at codon 12 of each DNA, 442

Vol. 160,

BIOCHEMICAL

No. 2, 1989

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

INTROY

txoll 1 CR” Cl..I.bbb

El:

cadon 12

NORMIL:

KC

Fi ure 1 Strategy for the detection of specific point mutations by PCR using mutation&primers. The drawings represent part of the first exon of the H-ras gene in either EJ bladder carcinoma cells (top) or normal cells (bottom). The mgenome contains a point mutation on codon 12 of this gene. Also represented is a synthetic oligonucleotide which shows full complementarity to the EJ DNA segment, allowing chain elongation when used as a primer for the Taql DNA polymerase. The same oligonucleotide is not a suitable primer on the normal DNA segment because of a base mismatch at its 3’-end, which cannot be repaired byTaq1 polymerase.

which contains a point mutation in the EJgenome (16), isshown. Also represented is a synthetic oligomer that shows perfect complementarity to El DNA but not to normal DNA because its 3’-end is complementary to the mutated base in the EJ gene. The oligomer

template,

is a suitable

primer

for

the Taql

DNA

polymerase

if EJ DNA

is used

as

but not with a normal DNA template because the resulting 3’-mismatch

cannot be repaired by the enzyme. We have tested the validity of the above approach to detect the specific point mutation present in the H-s gene of EJ cells (16) using the above mutation-specific oligomer as a primer for PCR studies (Fig. 2). The results showed abundant PCR products when EJ DNA was used as template (Fig. 2a). A smaller quantity of products could also be detected with the normal DNA when a nucleotide concentration of 200 PM was used for the reaction (Fig. 2a). This concentration isthe one recommended for PCR studies

(4).

However,

we reasoned

that

a

this background

could

be diminished

b 200 BP

--20

El N El N EJ2 N

G:C A:C c:C

118-

118-

72-

72-

Fi ure 2 Inhibition of PCR by primers containing single 3’-base mismatches. (a) PCR was per ormed using a DNA primer complementary to the mutated H-ras gene present in the El genome but showing a single base mismatch with normaTbNA (Fig. 1). Nucleotide concentrations of 200 PM, 20 PM, or 2 PM were used in the reaction mixes with either EJ or normal (N) DNA templates. The products were electrophoresed on agarose, transferred to Zetaprobe membranes, and analyzed by hybridization to a radiolabelled H-ras probe followed by autoradiography. (b) Oligomers identical to the mutation-specificprimer used in (a) except for their 3’-ends, which contained G, A, or C residues respectively, were used as primers for PCR studies of normal DNA. The first primer, which resulted in G:C base pairing with normal DNA, allowed enzymatic amplification to proceed whereas the other 2, which produced A:C or C:C mismatches, inhibited the reaction.

b

443

by

Vol. 160, No. 2, 1989

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

using lower nucleotide concentrations closer to the Km value for the Taql polymerase. The intensity of the signal obtained with a normal DNA template was indeed reduced when a nucleotide with concentrations

concentration of 20 pM was used and no signal could be detected of 2 pM (Fig. 2a).

To determine if the Taql enzyme would mismatches we synthesized 3 different oligomers primer used in the above experiment C residues H-*gene, primers, mutated

also be sensitive to other base identical to the mutation-specific

except for their 3’-ends, which

contained

C, A, or

respectively. Only the first one, which was complementary to the normal was a suitable primer for PCR studies on normal DNA (Fig. 2b). The other 2 which had either A:C or C:C mismatches H-E DNA, yielded no detectable

at their 3’-ends if hybridized to nonPCR products using a nucleotide

concentration of 2 pM for the reaction (Fig. 2b). Thus, the presence mismatch at the 3’end of DNA primers inhibits chain elongation polymerase

with

high efficiency

provided

that the nucleotide

of a single base by Taql DNA

concentration

used for

the reaction is 2 PM or less. One can thus detect known specific base substitutions in genomic DNA by performing PCR analyses using DNA primers with 3’-ends complementary to such mutations only if those mutations are present There are 6 possible

point

because the primers in the DNA template. mutations

involving

allow codon

enzymatic

amplification

12 of ras genes that can

result in amino acid substitutions in the ras protein products, 6 involving codon I 3, and 7 involving codon 61. Point mutations at these 3 codons are the only ones known to have resulted in ras oncogene activation in spontaneously arising tumors (see I). It would therefore be possible to detect all possible activating point mutations involving any of these codons on a given ras gene with a set of 19 different mutation-specific primers using the above approach. However, such point mutations can be found in any of the 3 known members of the ras gene family and the total number of primers required to examine all 3 genes therefore has to be multiplied by 3. To minimize the number of synthetic oligomers required and increase the practicality of our approach we have attempted to use primers that can bind with nearly equal efficiency to all 3 + genes simultaneously. This was achieved by introducing inosine residues on the primers at each position corresponding to sites showing lack of homology on the 3 ras genes (Fig. 3). lnosine residues have no destabilizing effect on DNA duplexes regardless of the opposing base (17). When mutation-specific, inosine-containing primers were used for PCR studies of m genes in EJ and normal DNAs we were able to detect enzymatic amplification of EJ but not of normal sequences (Fig. 4), indicating that the inosinecontaining primers were suitable for PCR analyses and could distinguish between mutated and non-mutated genes. The amplified sequences hybridized to DNA probes for H-E, but not to K-m or N-s probes (Fig. 4). confirming the specificity of the PCR products. Thus, the use of inosine-containing primers should allow examination of specific point mutations on all 3 ras genes simultaneously. One of the advantages of our approach is that it shows positive selectivity. Because it results in the amplification of DNA sequences harboring specific point 444

Vol. 160, No. 2, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

TTI

GIT

AAACAI

CIT

CT,

ATI

CT,

GGI

TG,

TAT

Fiqure 3. Nucleotide sequences of the entire first exons of the H-ras (18), K-ras (19), and N-4 (20) genes together with the sequence of the 2 oligonucleotide pritii&s used in Fig. 4 and 5. Codons 12 and 13, which are common sites for activating point mutations, are underlined. The primerscontain inosine residues at each position corresponding to ambiguous bases on the 3 genes. One of the primers has a base substitution at its 3’-end which corresponds to the H-w mutation found in EJ cells.

mutations mutated

only, DNA.

and the number small fraction

the This

is not

of neoplastic of the total

and demonstrate quantities

technique is important

its ability

of non-mutated

DNAs and attempted samples.

The results

to allow

detection

affected tumor

specimens

cells present

within

a given

cell population. to detect

To illustrate genetic

the specific

of Fig. 5 show of the mutated

DNA among

of H-*

EJ and

large normal

in the resulting

was sensitive

containing

a

of our approach

we mixed

technique

in samples

non-

heterogeneous

mass may represent

of mutated

12 mutation

that our proposed DNA even

tumor

sequences,

codon

of endogenous

are often

the sensitivity

very small amounts

homologous

to detect

by the presence

because

enough

up to a 107-fold

Thus, the sensitivity of this technique is non-mutated sequences. such that specific point mutations can be detected in tumor specimens even if only one

excess of homologous

out of lo7 cells are affected

H-RAS BP EJ N

by such mutations.

K-RAS El N

BP l:o

KRAS El N

118-

118-

72-

72-

0

1102

1:105 1:1d

0

5

4

Fi .+. ure 4

PCR analysis of EJ and normal (N) DNA with inosine-containing mutation. in Fig. 3 were used and the PCR products were

specs IC primers The primers shown analyzed as in Fig. 2 usin radiolabelled by alkali treatment and tR e membrane

probes. Fiqure 5. Sensitivity

probes for H-ras. was rehybridizafirst

of the mutation-specific

primer

The

probe was then removed to K-Eand then to N-E

assay in detecting

specific point

mutations. EJ DNA was mixed with normal DNA using the indicated ratios and the DNAs were analyzed by PCR using the inosine-containing mutation-specific primers shown in Fig. 3. The left lane (1:O) contained only EJ DNA and the right lane (O:l) contained only normal DNA. Total amount of DNA in each reaction mix was 1 pg.

445

0:1

CTC

Vol. 160, No. 2, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

DISCUSSION

The results of our experiments demonstrate the possibility of using mutationspecific primers for the detection of specific base substitutions in genomic DNA samples analyzed by PCR. Our data also show that the specificity of the primers can be markedly increased by using lower nucleotide concentrations for the reaction. This approach has several advantages over currently available techniques. Its simplicity allows for a more rapid screening of specific point mutations than other presently available methods. In fact, there is no need for hybridization studies because PCR products can be detected simply by ethidium bromide-staining after gel electrophoresis of the samples. It is more sensitive than other known techniques because of its positive selectivity effect. The results of Fig. 5 showed that the technique allowed detection of a specific point mutation in the presence of up to a 107-fold excess of normal DNA. Finally, our proposed use of inosine-containing primers can facilitate the study of mutations affecting

ras proto-oncogenes

because it allows simultaneous screening of point

mutations on all 3 of the known -ras genes. This technique should prove useful for the rapid detection of known recurring point mutations affecting ras proto-oncogenes as well as any other genetic loci of interest. Its high sensitivity can facilitate the early detection of such mutations in small quantities of human tissue specimens. It should also be valuable in the detection of minimal

residual disease during follow-up

of cancer patients.

It may therefore

contribute to the cost-effective use of molecular knowledge in clinical medicine.

ACKNOWLEDGMENTS

We thank Dr. Myron Goodman from USCfor his helpful advice during the course of this work. Thomas Ehlen is a recipient of the McEachern Award of the Canadian Cancer Society. This work was supported by a Wright Foundation Award to Louis Dubeau.

REFERENCES 1.

2. 3. 4. 5. 6. ;:

Bos, J.L. (1988) Mutation Res. 195, 225-271. Horowitz, J.M., Yandell, D.W., Park, S.-H., Canning, S., Whyte, P., Buchkowich, K., Harlow, E., Weinberg, R.A., and Dryja, T.P. (1989) Science 243, 937-940. Saiki, R.K., Scharf, S.! Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A., and Arnheim, N. (1985) Scrence 230, 1350-1354. Saiki, R.K., Gelfand, D.H., Stoffel, S., Scharf, S.J., Higuchi, R., Horn, G.T., Mullis, K.B., and Erlich, H.A. (1988) Science 239, 487-491. Verlaan-de-Vries, M., Bogaard, M.E., van den Elst, H., van Boom, J.H., van der Eb, A.J., and Bos, J.L. (1986) Gene 50, 313-320. Almo uera C. Shibata D., Forrester, K., Martin, J., Arnheim, N., and Perucho, M. (1988y Cell 53,‘549-554: Landegren, U., Kaiser, R., Sanders, L., and Hood, L. (1988) Science 241, 1077-1080. Cotton, R.G.H., Rodrigues, N.R., and Campbell, D. (1988) Proc. Natl. Acad. Sci. USA 85, 43974401. 446

Vol. 160, No. 2, 1989

9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Meyers, R.M., Lumelsky, N., Lerman, L.S., and Maniatis, T. (1985) Nature 313,495498. Chien,A., Edgar, D.B., andirela, J.M. (1976) J. Bacterial. 127, 1550-1557. Dubeau, L., Weinberg, K., Jones, P.A., and Nichols, P.W. (1988) Am. J. Pathol. 130, 558-594. McCoy, M.S., Bargmann, C.T., and Weinberg, R.A. (1984) Mol. Cell. Biol. 4, 15771582. Bell, G.I., Karam, J.H., and Rutter, W.J. (1981) Proc. Natl. Acad. Sci. USA 78, 57595763. Dubeau, L., and Jones, P.A. (1987) Cancer Res. 47, 2107-2112. Marmur, J. (1961) J. Molec. Biol. 3, 208-218. Tabin, C.J., Bradley, S.M., Bargmann, Cl., Weinberg, R.A., Papageorge, A.G., Scolnick, E.M., Dhar, A., Lowy, D.R., and Chang, E.H. (1982) Nature 300, 143-149. Ohtsuka, E., Matsuki, S., Ikehara, M., Takahashi, Y., and Matsubara, K. (1985) J. Biol. Chem. 260, 2605-2608. Reddy, E.P., Reynolds, R.K., Santos, E., and Barbacid, M. (1982) Nature 300, 149152. Capon, D.J., Seeburg, P.H., McGrath, J.P., Hayflick, J.S., Edman, U., Levinson, A.D., and Goeddel, D.V. (1983) Nature 304,507-513. Taparowski, E., Shimizu, K., Goldfarb, M., and Wigler, M. (1983) Cell 34,581-596.

447