PCR/RFLP Assay for

Copy Number of Mutant

and Wild-Type Alleles
BioTechniques 21:1055-1060 (December 1996)

ABSTRACT
A PCR method for quantitating the copy
number of mutant vs. wild-type alleles in
DNA from cell lines is described. The assay
can be used to detect a point mutation in
any gene that creates or destroys a restriction site. The alleles of interest are amplified by nested PCR and labeled in a second
round of PCR. The product is digested with
a restriction enzyme specific for that site,
resolved on a non-denaturing gel and quantified by phosphor imaging techniques. Cell
types with known numbers of wild-type and
mutant alleles of c-Harvey-ras are used to
validate the assay. The method is then applied to a cell line, homogygous for the mutation, to determine the gene copy number.
The applicability of the method to other
genes is shown using the matrix metalloproteinase gene, matrilysin. A cell line transfected with a plasmid carrying a mutant,
auto-activated form of the gene is compared
to its parent cell line. Advantages of this
technique compared with Southern analysis
are ease of screening a large number of
clones or foci and accuracy of quantitation.

INTRODUCTION
There are several research situations
where gene dosage is important. Cell
lines, derived from tumors may carry
an altered gene dose of mutant genes.
These cell lines may change this gene
dose during passage in tissue culture.
The copy number of transfected genes
integrated into the genome may need to
be determined. In the past, Southern
analyses were used to quantify the ratio
of mutant-to-wild-type c-Ha-ras alleles
(16). Quantifying Southern analyses,
however, can be a problem due to transfer and hybridization differences of
DNA fragments of various lengths,
background in the lane, incomplete cutting of DNA and the uneven loading of
lanes. Cytogenetics have been used as
an addition to Southern analyses to determine the copy number (4). An alternative method of determining gene
Vol. 21, No. 6 (1996)

dose, the probing of polymerase chain

reaction (PCR)-amplified alleles with
mutation-specific oligonucleotides (3),
requires two hybridization steps and the
adjustment of exposure times for
probe-specific activity and hybridization efficiency.
The PCR method described here is
time-efficient and accurate for establishing the copy number of mutant and
wild-type alleles. Previously, singlestage PCR amplification of alleles combined with mutation-specific restriction
fragment formation has been the basis
for detection of rare mutations in a
background of wild-type alleles (10).
Nested PCR, previously used to detect
rare mutant alleles (12) or rare viral
copies (18), is used here to enrich the
gene of interest in relationship to the
genome in the first stage, allowing the
second stage to be used for the incorporation of radioactive nucleotides. Optimization for Mg++ ion, nucleotide levels or annealing temperature was not
necessary for each new gene. Phosphor
imaging techniques provide wide linear
range with ease of quantitation and
without use of X-ray film.
MATERIALS AND METHODS
DNA Sources
CD-1 strain mouse epidermal tissue
(7) was the source of wild-type c-Haras alleles. PDV is a tumorigenic
mouse keratinocyte cell line, shown by
Southern analyses of XbaI digests to
have a 2:1 ratio of wild-type c-Ha-ras
alleles per mutant 61st codon allele,
(CAACTA) (1). A total of three cHa-ras gene copies was established by
cytogenetics (4). This cell line serves as
a control in the validation of the assay
for gene copies. CarB is a tumorigenic
spindle cell carcinoma keratinocyte
line, homozygous for the same c-Haras 61st codon AT mutation (3).
DU145 is a human prostate cell line
(15), with no known mutations in the
matrix metalloprotease matrilysin gene.
MR-15 clone 24 is a DU145 line, transfected with a cDNA plasmid carrying a
mutant 89th codon allele GTGGGG
in the second exon (13). An additional
Sau96I restriction site is formed by this
mutation.
BioTechniques 1055

Short Technical Reports

Primers
c-Ha-ras primers are based on the
published second exon of mouse cDNA
(2): Stage 1 Upstream 5 CTGTGAATTCTCTGGTCTGAGGAG 3 (nucleotide [nt] 244267); Stage 1 Downstream, 5 TAGGTGGCTCACCTGTACTG 3 (nt 491510); Stage 2 Upstream, 5 CTAAGCCTGTTGTTTTGCAGGAC 3 (nt 302322); and Stage 2
Downstream, 5 GGAACTTGGTGTTGTTGATGGC 3 (nt 455475). Matrilysin primers are based on the published second exon human cDNA (6):
Stage 1 Upstream, 5 GCCAACAGTTTAGAAGCCAAAC 3 (nt 201222);
Stage 1 Downstream, 5 CTGTAGGTGACCACTTTGGAAG 3 (nt 361
382); Stage 2 Upstream, 5 GAATGTTAAACTCCCGCGTCATA 3 (nt
261284); and Stage 2 Downstream, 5
TGACCACTTTGGAAGTCCATTTTG
3 (nt 352375).

Table 1. First-Stage PCR

1 g of digested genomic DNA

0.3 M each stage 1 primers
1 UlTma Buffer (Perkin-Elmer)
4 M each dNTP (Perkin-Elmer)
2.4 U UlTma DNA Polymerase (Perkin-Elmer)
2 mM Mg++ carried over by the DNA in the No. 2 Restriction Buffer

Protocol
The nested PCR protocol described
below produces a nonradioactive band
in the first stage of 267 bp for c-Ha-ras
and 182 bp for the matrilysin gene. The
second stage produces a double-stranded radioactive band of 176 bp for c-Haras and 115 bp for the matrilysin gene.
The diagnostic XbaI digest produces
bands of 91 and 85 bp for the mutant

c-Ha-ras. Sau96I produces diagnostic

bands of 61 and 54 bp for the mutant
matrilysin gene.
DNA Preparation
DNA was purified by proteinase K,
RNase and phenol/chloroform/isoamyl
alcohol extractions (5). High-molecular-weight DNA was spooled out, quantitated by spectrophotometry and digested with 5 U/g PstI and 5 U/g HindIII

in No. 2 Restriction Buffer (New England Biolabs, Beverly, MA, USA).

First-Stage PCR
Table 1 lists the components of the
first-stage PCR. A 50-L amplification
reaction was started with the addition
of UlTma DNA Polymerase (PerkinElmer, Norwalk, CT, USA) in 10 L of
1 UlTma Buffer into a DNA mixture
preheated to 85C on a Cetus DNA
Thermal Cycler (Perkin-Elmer). After
an initial denaturation for 2 min at
97C, 20 cycles of PCR were carried
out for 1 min at 97C, 1 min at 60C
and 1 min at 72C. The product was diluted 1:100 in H2O.
Second-Stage PCR
Table 2 lists the components of the
second-stage PCR. The 50-L reaction
was started with the addition of
Stoffel Fragment, AmpliTaq DNA
Polymerase (Perkin-Elmer) in 10 L of
1 Stoffel Buffer to DNA mixture preheated to 85C. After a denaturation
step of 2 min at 97C, 12 cycles of PCR
were carried out with 1 min at 97C
and 6 min at 67C. A final extension of
20 min at 67C was used.

Digestion for Restriction FragmentLength Polymorphisms (RFLP)

The second-stage product (10 L)
was pipetted into the appropriate restriction buffer with or without 10 U of
the diagnostic enzyme in a total volume
of 30 L and was incubated for 316 h
under oil.
Electrophoresis on a NonDenaturing TBE Vinyl Gel
Eight microliters of glycerol gelloading Buffer III (11) were added to
the digestion. The sample was heated to
65C for 5 min, and 10 L were loaded
onto a 0.4-mm-thick 1 TBE 6%8%
D-600 Vinyl Gel (American Technologies, Malvern, PA, USA) using a sequencing size vertical apparatus. Identical results were obtained using 10%
12% polyacrylamide gel electrophoresis (PAGE)/1 TBE. Electrophoresis
was carried out for 4 h at 500 V, and the
gel was dried for 30 min at 80C.
Phosphor Imaging
The gel was exposed to a PhosphorImager plate (Molecular Dynamics,
Sunnyvale, CA, USA) for 214 h.

Figure 1. Phosphor image of nested PCR products for c-Ha-ras (A) and matrilysin (B) with and
without diagnostic digestions. DNAs, prepared from tissue (CD-1) and various c-Ha-ras mutant cell
lines (PVD and CarB), were amplified by nested PCR individually (Panel A, lanes 16) and 1:1 DNA
mixtures (Panel A, lanes 712). The product, labeled by incorporation with [32P]dCTP in the second
stage, was run with (+) and without (-) digestion with the diagnostic enzyme XbaI to detect 61st codon
AT mutations in c-Ha-ras. Electrophoresis on a non-denaturing 1 TBE 6% D-600 vinyl gel (Panel
A) shows the proportion of undigestible wild-type c-Ha-ras alleles at 176 bp to digested mutant fragments at the 85/91-bp doublet. Panel B, cell line MR-15 Cl24 (lanes 34), transfected with mutant matrilysin, and its parent prostate carcinoma cell line DU145 (lanes 12) were assayed by the same protocol. Diagnostic Sau96I fragments that result from an 89th codon CTCCCC mutation in the integrated
cDNA for matrilysin were separated on a 10% 1 TBE polyacrylamide gel with the undigestible wildtype matrilysin alleles at 115 bp and digested mutant fragments at the 61/54-bp doublet. The 1:1 DNA
mixture is shown in Panel B, lanes 56.
Vol. 21, No. 6 (1996)

Short Technical Reports

Scanning was done on the ImageQuant program (Molecular Dynamics) with a rectangle created slightly
larger than the undigested band and one
rectangle around the mutant doublet.
Background was defined as 0 during
volume-integration. Since any radiation
incorporated into the undigested band
is displayed completely in the doublet
when digested, volume-integration of
uncut vs. doublet sections of the gel
represent directly the ratios of wildtype to mutant alleles carrying the diagnostic restriction site.

Table 2. Second-Stage PCR

10 L of first-stage dilution
0.6 M each stage 2 primers
1 Stoffel Buffer (Perkin-Elmer)
2.5 mM MgCl2
8 M each dNTP (Perkin-Elmer)
1 Ci [32P]dCTP at 3000 Ci/mM (Du Pont, Wilmington, DE USA)
5 U Stoffel Fragment, AmpliTaq DNA Polymerase (Perkin-Elmer)

Table 3. Gene Copy by Nested PCR/RFLP

RESULTS AND DISCUSSION

Percent
Expected
(wt/total)

Validation of the Assay

The protocol was first used to assay
the two cell types for which copy number of wild-type and mutant c-Ha-ras
has been established: CD-1 epidermis
with two copies of wild-type allele and
PDV cells with two copies of wild-type
and one copy of mutant allele. Figure 1
is a phosphor-imaged gel of the PCR
products from various sources, undigested (odd lanes) and digested with
the diagnostic enzyme (even lanes). In
Panel A, none of the CD-1 c-Ha-ras
PCR product at 176 bp (lane 1) was digested with XbaI (lane 2). This was expected since CD-1 has only wild-type
alleles. Part of the PCR product from
the PDV DNA could be digested with
XbaI (lanes 3 and 4). Only the digested
lane is needed to quantitate the ratio of
the two alleles. When radioactive volumes from the phosphor imager for the
undigested band at 176 bp were compared to the mutant fraction at doublet
91/85 and then expressed as a ratio of
wild-type/total ras alleles, the observed
gene dose of 64% closely matched the
expected dose (67%) from published
data (Table 3). Therefore, the alleles
present in the original genomic DNA
sample had been amplified and detected proportionally throughout the various steps of the assay.
To determine absolute gene copies
in a cell line, two DNAs combined
must demonstrate the correct proportion of alleles. In a CD-1/PDV mixture,
PDV would be contributing 2 wild-type
alleles and 1 mutant allele per genome,
and the CD-1 would contribute 2 wild1058 BioTechniques

Percent
Gene Copy No.
Known
Observeda
(wt/total)
wt mutant

Gene Copy No.

from Data
wt
mutant

c-Ha-ras
CD-1

100%

99.7 0.1%

PDV

67%

64.0 1.2%

CarB

0%

1.6 1.3%

CD-1/PDV Mix

80%

81.5 1.8%

CD-1/CarB Mix

unknown

55.0 5.0%

PVD/CarB Mix

unknown

36.8 1.8%

DU145

100%

98.5 0.2%

MR-15 Cl 24

unknown

75.3 2.5%

DU145/
MR-15 Cl 24 Mix

unknown

85.2 0.2%

Matrilysin

aobserved

values are means () 95% confidence level

type alleles per genome (Table 3).

When equal numbers of genomes from
each source are added in a mixture
(equal A260 units), a ratio of 4 wild-type
alleles to 1 mutant allele (80% wildtype/total alleles) would be expected.
Figure 1A shows the data for the double-source c-Ha-ras DNA PCRs in
lanes 712, and the quantitation of the
observed bands is given in Table 3. The
PCR product from CD-1/PDV (lanes 7
and 8) produces 81% wild-type. These
results establish the accuracy of the assay for a mixture of genomic DNAs.

Use of Assay for Cell Lines of

Unknown Gene Copy Number
Figure 1A (lanes 5 and 6) and Table
3 show the results of CarB tested as a
single-source DNA and confirm the
CarB to be homozygous for mutant alleles (4). The mixed sources of DNA
can then be used to determine the absolute number of copies of c-Ha-ras.
Because CD-1 has two copies of c-Haras and the ratio of 50% wild-type alleles were seen in the CD-1/CarB mixture (Figure 1A, lane 10 and Table 3),
Vol. 21, No. 6 (1996)

CarB can be deduced to have two

copies of mutated c-Ha-ras. This is
confirmed with the other possible mixture, CarB with PDV.
The assay was used to establish percent of wild-type to total alleles for the
matrilysin gene using a parental
DU145 cell line (15) and a DU145 cell
line stably transfected with a cDNA for
matrilysin that contained a mutant 89th
codon. The triploid parental cell line,
DU145 PCR-amplified matrilysin gene
product at 115 bp could not be digested
with Sau96I (Figure 1B, lane 2 and
Table 3), which confirms the absence of
the specific point mutation in this gene.
However, 25% of the alleles from transfected cell line MR-15 (13) clone 24
can be digested (lane 4). Upon mixing
the two cell lines (lanes 56), the observed percent wild-type alleles of 85%
is as expected for 6 wild-type alleles in
a total of 7 gene copies. Therefore,
MR-15 clone 24 has 1 mutant copy per
3 wild-type alleles of the matrilysin
gene.

the gene greatly reduces the number of

fragments seen on the gel. Southern
blots that would show both wild-type
and mutant alleles from the matrilysin
cell line transfected with plasmid
cDNA would be difficult to interpret
because of the number of the Sau96I
sites in the gene and unknown intron
sequences.
In cases where a mutation of interest

does not produce a diagnostic restriction fragment, rare mutations have been
detected by the introduction of a restriction site into a PCR band by the incorporation of a mismatched primer
(8,9,14,17). Using such a primer in the
second stage could extend this quantitative assay for use with point mutations
that do not produce a natural diagnostic
restriction site.

ADVANTAGES AND DISADVANTAGES

When Southern blots are used for
quantitation of copy number, comparisons are made between lanes of various cell lines with known copy number.
Although steps have been taken to ensure consistent treatment, each lane has
undergone many steps independently.
With the assay described here, all the
ratios are determined within a single
lane. The two alleles are amplified during PCR by the same primer set, with
the radioactivity incorporated equally
into mutant and wild-type sequences
according to the specific activity in that
tube. The two types of products (enzyme digestible and resistant) are transferred to a digestion reaction as a proportional mixture and loaded as the
same proportional mixture (digested
and undigested) onto the gel. Complete
digestion of short mutant PCR products
is easier to achieve than the digestion of
high-molecular-weight genomic DNA
on which the accuracy of Southern
analyses depends. Enzymes that recognize only four bases can be chosen as a
diagnostic enzyme because the digestion of only a short amplified section of
Vol. 21, No. 6 (1996)

BioTechniques 1059

Short Technical Reports

Linear range of cycling, with low
product-to-product re-annealing, is important for the accuracy of this assay.
Because restriction enzymes only recognize sites within complementary
strands, re-annealed products that form
wild-type to mutant heteroduplexes in
plateau stage create an overabundance
of the undigested allele.
When determining copy number,
mixtures need to be made of equal
amounts of genomic DNA. However, if
only a gene ratio is wanted as in screening clones, purification can be simplified and quantitation omitted.
PCR combined with RFLP has previously been used to detect mutations.
By adding a nested primer step and
choosing direct incorporation as a labeling method, this protocol is easier
and more quantitative than the methods
in use to detect allelic gene dose. The
numbers derived by the phosphor imaging system are directly proportional to
the allelic copies in the cell lines. When
this assay is used on a mixture of two
cell lines, including one for which the
copy number is known, the total number of copies can be determined without need for another step such as cytogenetics.
REFERENCES
1.Bremner, R. and A. Balmain. 1990. Genetic
changes in skin tumor progression: correlation
between presence of a mutant ras gene and
loss of heterozygosity on mouse chromosome
7. Cell 61:407-417.
2.Brown, K., B. Bailleul, M.K. Ramsden, F.
Fee, R. Krumlauf and A. Balmain. 1988.
Isolation and characterization of the 5 flanking region of the mouse c-Harvey ras gene.
Mol. Carcinog. 1:161-170.
3.Buchmann, A., B. Ruggeri, A.J.P. KleinSzanto and A. Balmain. 1991. Progression of
squamous carcinoma cells to spindle carcinomas of mouse skin is associated with an imbalance of H-ras alleles on chromosome 7.
Cancer Res. 51:4097-4101.
4.Fusenig, N.E., R.T. Dzarlieva-Petrusevska
and D. Breitkreutz. 1985. Phenotypic and
cytogenetic characteristics of different stages
during spontaneous transformation of mouse
keratinocytes in vitro. Carcinog. Compr. Surv.
9:293-326.
5.Futscher, B.W., R.O. Pieper, W.S. Dalton
and L.C. Erickson. 1992. Gene-specific
DNA interstrand cross-links produced by nitrogen mustard in the human tumor cell line
Colo320HSR. Cell Growth Differ. 3:217-223.
6.Gaire, M., Z. Magbanua, S. McDonnell, L.
McNeil, D.H. Lovett and L.M. Matrisian.
1994. Structure and expression of the human
1060 BioTechniques

gene for the matrix metalloproteinase matrilysin. J. Biol. Chem. 269:2032-2040.

7.Holladay, K., H. Fujiki and G.T. Bowden.
1992. Okadaic acid induces the expression of
both early and secondary response genes in
mouse keratinocytes. Mol. Carcinog. 5:16-24.
8.Kumar, R. and M. Barbacid. 1988. Oncogene detection at the single cell level. Oncogene 3:647-651.
9.Kumar, R., S. Sukumar and M. Barbacid.
1990. Activation of ras oncogenes preceding
the onset of neoplasia. Science 248:11011104.
10.Lu, S.J. and M.C. Archer. 1992. Ha-ras oncogene activation in mammary glands of Nmethyl-N-nitrosourea-treated rats genetically
resistant to mammary adenocarcinogenesis.
Proc. Natl. Acad. Sci. USA 89:1001-1005.
11.Maniatis, T., E.F. Fritsch and J. Sambrook.
1982. Molecular Cloning: A Laboratory Manual, p. 160. Cold Spring Harbor Laboratory,
Cold Spring Harbor, NY.
12.Nakazawa, H., D. English, P.L. Randell, K.
Nakazawa, N. Martel, B.K. Armstrong and
H. Yamasaki. 1994. UV and skin cancer: specific p53 gene mutation in normal skin as a
biologically relevant exposure measurement.
Proc. Natl. Acad. Sci. USA 91:360-364.
13.Park, A.J., L.M. Matrisian, A.F. Kells, R.
Person, Z. Yuan and M. Navre. 1991. Mutational analysis of the transin (rat stromelysin)
autoinhibitor region demonstrates a role for
residues surrounding the cycteine switch. J.
Biol. Chem. 266:1584-1590.
14.Shinder, G.A., S. Manam and W.W. Nichols. 1993. A sensitive restriction fragment
length polymorphism method to detect CAA
to AAA mutations at codon 61 of C-Ha-ras.
Mol. Carcinog. 7:263-267.
15.Stone, K.R., D.D. Mickey, H. Wunderli,
G.H. Mickey and D.F. Paulson. 1991. Isolation of a human prostate carcinoma cell line
(DU-145). Int. J. Cancer 21:274-281.
16.Strickland, J.E., D.A. Greenhalgh, A. Koceva-Chyla, H. Hennings, C. Restrepo, M.
Balaschak and S.H. Yuspa. 1988. Development of murine epidermal cell lines which
contain an activated rasHa oncogene and form
papillomas in skin grafts on athymic nude
mouse hosts. Cancer Res. 48:165-169.
17.Sukumar, S., B. Armstrong, J.P. Bruyntjes,
I. Leav and M.C. Bosland. 1991. Frequent
activation of the Ki-ras oncogene at codon
12 in N-methyl-n-nitrosourea-induced rat
prostate adenocarcinomas and neurogenic sarcomas. Mol. Carcinog. 4:362-368.
18.Zimmermann, K., K. Pischinger and J.W.
Mannhalter. 1994. Nested primer PCR detection limits of HIV-1 in the background of increasing numbers of lysed cells. BioTechniques 17:18-20.

Campbell, Tucson, AZ 85724, USA. Internet: bowden@azcc.arizona.edu

Received 22 November 1995; accepted
29 May 1996

Joanne S. Finch and G. Tim

Bowden
University of Arizona
Tucson, AZ, USA

We thank members of our laboratory for

their helpful suggestions and critical readings of this manuscript. This work was supported by National Cancer Institute Grant
CA40584 to G.T.B. Address correspondence
to Joanne S. Finch, Dept. of Radiation Oncology, University of Arizona, 1515 N.
Vol. 21, No. 6 (1996)