Amplification-Refractory Mutation System UNIT 9.

8
(ARMS) Analysis of Point Mutations
The amplification-refractory mutation system (ARMS), also known as allele-specific
polymerase chain reaction (ASPCR) or PCR amplification of specific alleles, is a simple,
rapid, and reliable method for detecting any mutation involving single base changes or
small deletions. ARMS is based on the use of sequence-specific PCR primers that allow
amplification of test DNA only when the target allele is contained within the sample and
will not amplify the nontarget allele. Following an ARMS reaction the presence or absence
of a PCR product is diagnostic for the presence or absence of the target allele.
The protocols detailed here outline methods that can be used to analyze human genomic
DNA for one or more mutations. The basic protocol describes the development and
application of an ARMS test for a single mutation; the alternate protocol extends this to
multiplex ARMS for the analysis of two or more mutations. The support protocol
describes a rapid DNA extraction method from blood or mouthwash samples that yields
DNA compatible with the type of tests described.
Optimized ARMS tests are extremely straightforward and reliable to use—the key to
success lies in the selection of appropriate primer sequences and reaction conditions. This
unit describes not only the use of ARMS tests but also the process of test development.
NOTE: High-quality water (e.g., tissue culture grade) should be used in all solutions.
Sigma double-processed tissue culture water has been shown to work well.
CAUTION: Radioactive, biological, and chemical substances require special handling;
see APPENDIX 2 for guidelines.

STRATEGIC PLANNING
The ARMS technique is based upon the observation that oligonucleotides that are
complementary to a given DNA sequence except for a mismatched 3′ terminus will not
function as PCR primers under appropriate conditions. For a robust and reliable ARMS
test it is necessary to determine a combination of primer sequence and reaction conditions
that will generate a detectable ARMS product from the target allele while minimizing
false priming at the nontarget allele.
The solution suggested here is empirical and is based upon the development of a large
number of ARMS tests. The approach is to standardize as many variables as possible (e.g.,
thermal cycling parameters, enzyme concentration, etc.) and to optimize the reaction by
altering just two parameters, namely ARMS primer sequence (discussed in the following
paragraphs) and ARMS primer concentration (which is optimized based on results of a
pilot experiment, as described in the basic protocol).
The following guidelines are used to generate ARMS primers that will detect point
mutations when used in combination with the reaction conditions outlined in the basic
protocol. An example is given in Figure 9.8.1.
1. The ARMS primers should be oligonucleotides of around 30 or more bases. Primers
less than 28 bases long should be avoided. Longer primers (up to 60-mers) may be
used.
2. For the mutant-specific primer (M), the 3′ terminal base of the ARMS primer should
be complementary to the mutation; for the normal-specific primer (N), the 3′ terminal
Clinical
base should be complementary to the corresponding normal sequence. Molecular
Diagnostics

Contributed by Stephen Little 9.8.1

Current Protocols in Human Genetics (1995) 9.8.1-9.8.12
Copyright © 2000 by John Wiley & Sons, Inc. Supplement 7
 Normal sequence Mutant sequence
C
C C
Normal C A G A T A G . . .5 ′ A G A T A G . . .5 ′
primer 5 ′...GAAC G CT CTA T C GCGAT...3 ′ 5 ′. . . G A A C A CT C T A T C G C G A T . . .3 ′
482 482

T
C C
Mutant A G A T A G . . .5 ′ T A G A T A G . . .5 ′
primer 5 ′. . . G A A C G CT C T A T C G C G A T . . .3 ′ 5 ′. . . G A A C A CT C T A T C G C G A T . . .3 ′
482 482

Figure 9.8.1 ARMS primer sequences for a single ARMS test. Sequences of the ARMS primers
and target DNA sequences around the R117H mutation of the CFTR gene (G→A at position 482;
Dean et al., 1990). The base that is altered is indicated in the normal and mutant DNA sequences
by a box. The presence of an arrow indicates that primer/target combinations can be extended by
Taq DNA polymerase; an “X” indicates extension does not occur. Bases in the ARMS primers that
are not complementary to the target are shown displaced from the target sequence. A single
mismatch (in this case a C/C) at the penultimate base is not sufficient to prevent extension whereas
a primer with two adjacent mismatches, at the terminal and the penultimate base, is not extended.

Table 9.8.1 ARMS Primer Additional Mismatch Selectiona

Coding strand nucleotide corresponding to

Terminal penultimate nucleotide in the primer
mismatch
A G C T
AA A G A G
AG C T A G
AC G A C T
TT C T A G
TG G A T C or T
TC C T A G
CC C T A G
GG A G A G
aThe table indicates which base to include at the penultimate position of the
ARMS primer. The left-hand column lists the 3′ terminal mismatches speci-
fied (i.e., the mutation in the coding strand with the normal base in the
anti-coding strand) and the top row indicates the target base in the coding
strand corresponding to the penultimate base in the primer.

3. Additional deliberate mismatches should normally be introduced at the penultimate

base of the ARMS primer to increase the specificity of the ARMS reaction. Because
different mismatches have been found to have different destabilizing effects, it is
necessary to consider both terminal and penultimate mismatches together. If the
mutation-induced terminal mismatch is strong, a weak additional mismatch should
be selected, and vice versa, as indicated in Table 9.8.1.
4. The remainder of the ARMS primer should be complementary to the target sequence.
5. The design of the common primer is straightforward. The primer should be 30 bases
long; selected to have ∼50% G + C content, no 3′ complementarity with the ARMS
primer or the internal control primers, and no repeated or unusual sequences (e.g.,
ARMS runs of a single base or palindromes); and, for use with the control PCR reactions
Analysis of Point
Muataions suggested in this unit, should give a 150- to 250-bp PCR product.

9.8.2
Supplement 7 Current Protocols in Human Genetics
ANALYSIS OF SINGLE MUTATIONS BY ARMS TESTS BASIC
PROTOCOL
A typical single ARMS test is comprised of two complementary reactions each conducted
using the same substrate DNA. The first reaction contains an ARMS primer specific for
the normal DNA sequence and cannot amplify mutant DNA at a given locus. Similarly,
the second reaction contains a mutant-specific primer and cannot amplify normal DNA.
Both reactions contain the same common PCR primer and also a pair of internal control
primers that coamplify a different region of the genome (see Fig. 9.8.1).
The protocol describes a two-stage approach to designing ARMS tests. Primer sequences
selected on the basis of the guidelines provided (see Strategic Planning) are used in ARMS
reactions with known DNA samples. Based on the outcome of these pilot reactions, the
tests are optimized by modifying the ARMS primer sequence and concentration.
Materials
For recipes, see Reagents and solutions in this unit (or cross-referenced unit); for common stock
solutions, see APPENDIX 2; for suppliers, see SUPPLIERS APPENDIX.
50 µM normal ARMS primer
50 µM mutant ARMS primer
50 µM common primer
50 µM each control primer A and B (or other appropriate pair; see recipe)
1 mM 4dNTP mix (APPENDIX 2)
10× PCR amplification buffer containing 12 mM MgCl2 (APPENDIX 2)
Control DNA samples of known genotype (10 to 50 ng/µl in H2O)
Light mineral oil
5 U/µl Taq DNA polymerase (Perkin-Elmer Cetus AmpliTaq or equivalent)
Nusieve 3:1 agarose (FMC Bioproducts)
1× TBE buffer (APPENDIX 2) containing 0.5 µg/ml ethidium bromide
Loading buffer (see recipe)
DNA molecular size markers
Test DNA samples (10 to 50 ng/µl in H2O; support protocols)
Perkin-Elmer Cetus thermal cycler (480 or TC) and suitable reaction tubes
Additional reagents and equipment for agarose gel electrophoresis (UNIT 2.7)

Perform pilot ARMS reactions

1. Follow the ARMS primer guidelines outlined above (see Strategic Planning) to design
normal, mutant, and common ARMS primers suitable for the mutation of interest.
Select an appropriate control primer pair (see recipe). Obtain all primers and prepare
50 µM solutions of each.
In general, use control primers A and B. If it is necessary to amplify a >300-bp ARMS
product, use primer pair C and D. If the ARMS product has a high G+C content (>65%)
and is >300 bp in length, use primer pair E and F.
2. Prepare normal (N) and mutant (M) ARMS reaction premixes according to the
guidelines presented in Table 9.8.2. Dispense 40-µl aliquots of the premix into
reaction tubes suitable for use in the thermal cycler.
The amounts of reagents shown in the table are sufficient for ten ARMS reactions; they can
be scaled up for larger numbers of reactions.
The use of different-colored reaction tubes for the normal and mutant reaction mixes
eliminates the need for labeling the premix aliquots. Use a fresh pipet tip for each successive
transfer step.
ARMS reaction mixes are stable at this stage and can be stored several days at room Clinical
temperature or several months at −20°C. Molecular
Diagnostics

9.8.3
Current Protocols in Human Genetics Supplement 3
 Table 9.8.2 ARMS Reaction Premixes

Normal ARMS Mutant ARMS

Reagent
reaction (N) reaction (M)
50 µM common primer 10 µl 10 µl
50 µM normal ARMS primer 10 µl —
50 µM mutant ARMS primer — 10 µl
50 µM control primer A 10 µl 10 µl
50 µM control primer B 10 µl 10 µl
1 mM 4dNTP mix 50 µl 50 µl
10× PCR amplification buffer 50 µl 50 µl
H2O 260 µl 260 µl

3. Add 5 µl of 10 to 50 ng/ml control DNA to each tube of an N and M premix pair at

room temperature. If possible set up separate reactions containing DNA from normal,
heterozygote, and homozygote individuals. Include a negative control test in which
no DNA is added to the N and M tubes.
4. Add one drop mineral oil to each reaction tube and cap firmly. Microcentrifuge 10
sec at high speed.
5. Dilute Taq DNA polymerase by adding 12.5 µl of 10× PCR amplification buffer and
5 µl Taq DNA polymerase (1 U/5 µl final) to 107.5 µl sterile water. Mix by pipetting.
This amount is sufficient for ten tests.
Because the specificity of an ARMS reaction involves the nonextension of a 3′ mismatched
nucleotide, it is essential that a DNA polymerase without 3′ to 5′ proofreading activity is
used. Therefore, Taq DNA polymerase is suitable for ARMS reactions, but other DNA
polymerases such as Vent and Pfu cannot be used (see CPMB UNIT 7.4A for more extensive
discussion of the properties of different DNA polymerases).
6. Place tubes in thermal cycler block. Run 94°C hold program. After 5 min at 94°C,
remove one tube from the block and add 5 µl Taq DNA polymerase dilution to the
lower (aqueous) phase through the mineral oil layer. Return tube to 94°C block and
repeat process until enzyme has been added to all tubes.
This 94°C “hot start” for the PCR amplification helps eliminate nonspecific annealing of
the primer to the template. Minimize the time each tube is out of the block at the enzyme
addition stage to ensure that the temperature of the reaction mix does not drop significantly.
Care should be taken to avoid depositing the enzyme into the oil layer.
7. Stop the 94°C hold program and immediately run the following amplification
program:
35 cycles: 1 min 94°C (denaturation)
1 min 60°C (annealing)
1 min 72°C (extension)
1 cycle: 10 min 72°C (extension).
If a rapid thermal cycler such as the Perkin-Elmer Cetus 9600 is used, follow the
manufacturer’s recommendations to alter the cycling times to suit the characteristics of
this type of machine.
At the end of the amplification program, it is advisable to leave the reaction tubes until the
block temperature has reached 50°C to prevent the formation of artifact products.
ARMS The amplified samples are stable up to 1 week at room temperature.
Analysis of Point
Muataions

9.8.4
Supplement 3 Current Protocols in Human Genetics
Visualize the reaction products
8. Prepare a 3% agarose gel using 4.5 g Nusieve 3:1 agarose and 150 ml of 1× TBE
buffer with 0.5 µg/ml ethidium bromide.
A 150-ml gel is suitable for the analysis of at least ten tests. Larger gels can be used if
required. This gel results in good separation of the ARMS products. Other agaroses that
give good separation in the product size-range may be substituted.
CAUTION: TBE buffer and gel containing ethidium bromide should not be allowed to
come into contact with skin.
9. Label a 0.5-ml microcentrifuge tube for each reaction. Add 10 µl loading buffer to
each tube.
10. Transfer 25 µl ARMS reaction from beneath the oil layer into the corresponding
labeled tube. Mix by pipetting.
To avoid contaminating the laboratory with ARMS-product aerosols, it is preferable to
uncap the reaction tubes in a fume hood or a separate laboratory.
11. Load 20 µl of each reaction per lane and include a DNA molecular size marker in
one lane of the gel.
12. Electrophorese 1 to 2 hr at 120 V and photograph under UV transillumination.

Analyze the pilot reactions

13. Analyze each lane for presence of the expected-size control fragment and ARMS
fragment (control fragments should be 360 bp, 813 bp, or 825 bp for primer pairs A
and B, C and D, or E and F respectively). For both the normal and the mutant ARMS
reactions, there are three potential outcomes:
Expected outcome. ARMS product when target allele is present in sample; no visible
product when target allele is absent.
Nonspecific outcome. ARMS product when target allele is present in sample; ARMS
product still visible when target allele absent.
Insensitive outcome. ARMS product absent or faint even when target is present; no
visible product when target allele is absent.
a. If the expected result is obtained for both the normal and mutant ARMS primers,
then the optimization is complete and the ARMS test is ready for use—omit steps
14 and 15 and proceed to step 16.
b. If one or both of the ARMS primers is nonspecific, proceed to step 14a.
c. If one or both of the ARMS primers is insensitive, proceed to step 14b.
It is only possible to test for nonspecificity of an ARMS primer if a sample that is
homozygous for the nontarget allele is available. This is usually straightforward for the
mutant ARMS primer, as the appropriate sample is wild-type. For rare mutant alleles,
testing the specificity of the normal primer can be difficult due to the rarity of
homozygous mutant individuals.

Optimize the ARMS reactions

For nonspecific ARMS primer(s):
14a. Increase the reaction specificity by repeating the procedure starting at step 2 with
two sets of reactions. Reduce the amount of ARMS primer in the two reaction mixes
by two- and five-fold respectively. Clinical
Molecular
The concentration of the common primer is not altered. Diagnostics

9.8.5
Current Protocols in Human Genetics Supplement 3
 Table 9.8.3 ARMS Primer Mismatch Strength Tablea

Destabilization strength Mismatch pairing(s)

Maximum GA, CT, TT
Strong CC
Medium AA, GG
Weak CA, GT
None AT, GC
aMismatches are grouped according to destabilization strength. At
least two mismatch strengths are available for each base.

15a. If nonspecific bands are still observed, increase the strength of the additional
mismatch at the penultimate base. Repeat the procedure starting at step 1.
See Table 9.8.3 for guidelines about how to increase mismatch strength.

For insensitive ARMS primer(s):

14b. Increase the sensitivity by repeating the procedure starting at step 2 with two sets
of reactions. Increase the amount of ARMS primer in the two reaction mixes by two-
and four-fold respectively.
An alternative option is to reduce the concentration of the control primers by two- and
four-fold. The concentration of the common primer is not altered.
15b. If bands for the expected ARMS products are still weak or absent, decrease the
strength of the additional mismatch at the penultimate base. Repeat the procedure
starting at step 1. Decrease the additional mismatch strength by following the
guidelines shown in Table 9.8.3.
Changing the additional mismatch in an ARMS primer often causes a large alteration in
the specificity and sensitivity of an ARMS reaction. Altering the primer concentration has
a smaller effect and can be used to fine-tune the reaction.
Following optimization of the ARMS test using DNA samples of known genotype, it is
recommended that a large batch of ARMS reactions be made and divided into aliquots. The
individual ARMS reactions are stable at least 6 months at −20°C.

Perform optimized ARMS tests

16. Using the information obtained from the pilot experiments to optimize reaction
conditions, analyze test DNA samples by following steps 2 through 13 but replacing
the control DNA with the test samples to be analyzed.
It is recommended that both negative controls without DNA and positive control samples
of known genotype be included.

ALTERNATE ANALYSIS OF MULTIPLE MUTATIONS BY MULTIPLEX ARMS

PROTOCOL It is often necessary to analyze a particular gene for the presence or absence of several
different point mutations. One option is to carry out a series of single ARMS tests; the
alternative approach—described in this protocol—is to develop a multiplex ARMS test
that will allow simultaneous analysis of several mutations.
The approach is the same as that used for the single ARMS tests in that most of the
variables are standardized and only primer sequences and concentrations are altered to
achieve the desired result. The principle difference is that there are several primer
combinations to be optimized simultaneously, which increases the complexity of the
procedure. In view of this increased complexity it is not possible to outline a standard
ARMS
Analysis of Point protocol for multiplex ARMS test development. The basic steps in the protocol are the
Muataions same as for the single test development with the following differences:
9.8.6
Supplement 3 Current Protocols in Human Genetics
 1. In general, use less destabilizing mismatches in an ARMS test that is part of a
multiplex than if it had been a single ARMS test. Use Table 9.8.1 (mismatch selection)
followed by Table 9.8.3 (mismatch strength) to select less destabilizing mismatches.
2. The different ARMS products in the reaction are distinguished on the basis of size.
It is best to plan the ARMS reactions so there is a ≥50-bp difference in the ARMS
products’ sizes. In some cases it will be necessary to use a control PCR amplification
that gives a larger product; two alternatives are listed among the oligonucleotide
primers (see Reagents and Solutions).
3. Combine several ARMS reactions into two multiplex reactions. Split the normal and
mutant reactions so both multiplex reactions contain normal ARMS primers for some
reactions and mutant ARMS primers for others. An example of a multiplex reaction
containing two normal and two mutant primers is shown in Figure 9.8.2.
4. In testing for an autosomal recessive inherited disease gene, the PCR control reaction
can be omitted if at least four mutations are analyzed simultaneously and both
reaction tubes contain at least two normal specific ARMS reactions.
5. If the multiplex test is for several mutations in close proximity in the same exon, it
is possible to use a single common primer in combination with a number of ARMS
primers.
6. The multiplex protocol is essentially the same as for the single ARMS tests, except
that it may be necessary to repeat the full procedure (steps 1 to 16) multiple times
because there are several amplification reactions to optimize simultaneously.

1 2 3 4 blank
N M N M N M N M N M

control
ARMS product

1 2 3 4 5 6 7 8
N M N M N M N M N M N M N M N M
621+1
G551D
G542X
F508

Figure 9.8.2 Single and multiplex ARMS tests. The upper panel shows the results from four
individuals tested for the R117H mutation of the CFTR gene. Each ARMS test consists of two ARMS
reactions specific for the normal (N) or mutant (M) sequences. The 3′ sequences of the ARMS
primers are given in Figure 9.8.1. All sample lanes contain the 360-bp PCR control product amplified
using control primer pair A and B. Samples 1 and 2 are from an individual heterozygous for the
R117H mutation, whereas samples 3 and 4 are from normal individuals. The lower panel shows
the results of seven individuals analyzed using a multiplex ARMS test for four common mutations
of the CFTR gene (∆F508, G542X, G551D and 621+1G→T). There are two lanes for each sample.
The first lane contains products of the normal ARMS primers for 621+1G→T and ∆F508 and the
mutant ARMS primers for G551D and G542X. The second lane contains the corresponding
products for the 621+1G→T and 508 mutant primers and the G551D and G542X normal primers.
The location of the products of each primer set is indicated at the side of the figure. The genotypes
Clinical
of the seven samples are: 1, normal; 2, ∆F508 heterozygote; 3, G551D heterozygote; 4, G542X Molecular
heterozygote; 5, 621+1G→T heterozygote; 6, G542X, ∆F508 heterozygote; 7, G551D, ∆F508 Diagnostics
heterozygote.
9.8.7
Current Protocols in Human Genetics Supplement 3
 SUPPORT RAPID DNA EXTRACTION FROM MOUTHWASH
PROTOCOL AND BLOOD SAMPLES
Because of their ease of use, speed, and reliability, ARMS tests are often used to screen
large numbers of DNA samples. The following protocol describes a rapid DNA extraction
method for blood and mouthwash samples that provides DNA compatible with the ARMS
reaction conditions described in this unit.
Materials
For common stock solutions, see APPENDIX 2.
170 mM ammonium chloride (prepare fresh)
Blood sample, fresh or frozen
10 mM NaCl/10 mM EDTA
4% (w/v) sucrose in H2O
50 mM sodium hydroxide
1 M Tris⋅Cl, pH 7.5 (APPENDIX 2)
1.5-ml screw-cap microcentrifuge tube
Rotator
25-ml plastic sample tube, sterile
Low-speed centrifuge
Boiling water bath

To extract DNA from blood:

1a. Add 800 µl freshly prepared 170 mM ammonium chloride to 200 µl blood in a 1.5-ml
screw-cap microcentrifuge tube. Mix 20 min on a rotator.
The method is suitable for fresh or frozen blood with or without the addition of anticoagu-
lating agents such as EDTA.
2a. Microcentrifuge 2 min at high speed to obtain a white cell pellet. Discard supernatant.
3a. Wash cell pellet with 300 µl of 10 mM NaCl/10 mM EDTA. Microcentrifuge 15 sec
at high speed to pellet cells. Repeat wash three times to remove all visible hemoglo-
bin.

To extract DNA from mouthwash samples:

1b. Vigorously agitate 10 ml of 4% sucrose in the mouth for 20 sec to produce a
suspension of buccal epithelial cells. Collect suspension in a sterile 25-ml plastic
sample tube.
To avoid any potential contamination problems, it is recommended that food and drink be
avoided in the 30 min prior to sampling.
2b. Centrifuge 10 min at 1200 × g, room temperature, to collect cells. Discard super-
natant.
3b. Resuspend cells in 500 µl of 10 mM NaCl/10 mM EDTA and transfer to screw-cap
microcentrifuge tubes. Microcentrifuge 15 sec at high speed and discard supernatant.
4. Resuspend pellet in 500 µl of 50 mM sodium hydroxide and vortex 10 sec to
resuspend white cell pellet. Incubate 20 min in boiling water bath.
5. Add 100 µl of 1 M Tris⋅Cl to neutralize and vortex 5 sec.
6. Microcentrifuge 15 sec at high speed to remove cell debris. Store supernatant at
−20°C until use.
ARMS
Analysis of Point 5 ìl of DNA prepared in this way contains ∼100 ng of DNA and is sufficient for a single
Muataions ARMS reaction (2 × 5 ìl per ARMS test). DNA can be stored at least 1 year at −20°C.
9.8.8
Supplement 3 Current Protocols in Human Genetics
REAGENTS AND SOLUTIONS
Loading buffer
70% (v/v) 1× TBE buffer (APPENDIX 2)
30% (v/v) glycerol
0.1% (w/v) bromphenol blue
Oligonucleotide primers
Control primer A
5′-CCCACCTTCCCCTCTCTCCAGGCAAATGGG-3′
Control primer B
5′-GGGCCTCAGTCCCAACATGGCTAAGAGGTG-3′
These primers amplify a fragment of the ATT gene and generate a 360-bp PCR product.
Long control primer C
5′-CCAAGCCCAACCTTAAGAAGAAAATTGGAG-3′
Long control primer D
5′-CCAAACCCACGGTACGCATGGGAACACTGC-3′
These primers amplify a fragment of the APC gene and generate a 813-bp PCR product.
High G+C control primer E
5′-CTCACCTGCGTCAGGAGAGCACACACTTGC-3′
High G+C control F
5′-CATCGAGACCTCGGCCAAGACCCGGCAGG-3′
These primers amplify a fragment of the RAS gene and generate a 825-bp PCR product.

COMMENTARY
Background Information tially consists of the determination of reaction
Over the past few years there has been an conditions that will allow this discrimination
exponential increase in the number of human between target and nontarget alleles. The pro-
genes cloned and sequenced coupled with an tocols outlined in this unit describe one ap-
increase in the number of identified mutations proach to the development of ARMS tests based
and polymorphisms that are of clinical rele- on the alteration of the ARMS primer sequence
vance. As a consequence there is now an in- and concentration. Alternative strategies are
creasing demand for analysis of human genetic also possible (Sommer et al., 1992).
variation in the fields of inherited disease test- The ARMS technique involves the exten-
ing, cancer diagnostics, tissue typing, and dis- sion or nonextension of an ARMS primer hy-
ease predisposition analysis. bridized to target DNA. The yield of PCR
For clinical applications, the analysis tech- product from a particular ARMS primer and
nique must be simple and reliable and should target DNA combination is governed by two
allow analysis of several mutations simultane- variables—the rate of hybridization of the
ously and be suitable for all point mutations. ARMS primer to target DNA and the rate at
The amplification-refractory mutation system which the bases at the 3′ end of the ARMS
(ARMS) technique (Newton et al., 1989; see primer form a suitable substrate for Taq DNA
also Okayama et al., 1989; Wu et al., 1989; polymerase (Wu et al., 1989). The approach in
Sarkar et al., 1990) has a number of features this unit is to use long (>28-mer) ARMS prim-
that make it particularly suited for routine ap- ers to minimize any differences in the stability
plications. of annealed primers between the target and
Reliability and test design. An ARMS test nontarget allele. It is then the rate of suitable
is specific when, using the detection system of substrate formation that governs the overall rate
choice (e.g., agarose gel electrophoresis), the of the reaction. The advantage of this approach
yield of product from the target allele can be is that the rate of hybridization is sensitive to
detected and the yield of product from the small changes in ionic strength and tempera-
nontarget allele is not detectable. The process ture, whereas the rate of extension appears to Clinical
Molecular
of ARMS test design and development essen- be less affected by such changes. As a conse- Diagnostics

9.8.9
Current Protocols in Human Genetics Supplement 3
 quence, the temperature window in which an reliable, it is still possible that on rare occasions
ARMS reaction will give the correct result is there may be problems of specificity or sensi-
larger than the corresponding window for tech- tivity with a particular reaction. The following
niques based solely on hybridization, such as are additional suggestions for these circum-
allele-specific oligonucleotide (ASO) hybridi- stances:
zation (UNIT 9.4). This makes ARMS a reliable (1) Increase specificity by increasing the
technique that, once optimized, will perform in annealing temperature. This is not normally
a reproducible fashion. recommended as the resulting test will tend to
Applicability. There are many published ex- be temperature-sensitive with an increased risk
amples of single and multiplex ARMS tests that of producing a false-positive result.
include all possible combinations of transitions (2) Alter the primer sequences by switching
and transversions (Sommer et al., 1992). It the direction of the ARMS reaction. There is no
seems likely that the ARMS technique is suit- particular advantage in designing the ARMS
able for all point mutations. ARMS is particu- primer to complement the sense rather than the
larly relevant where there is a need to analyze antisense strand of a given gene. If a reaction
a large number of samples for one or more is proving problematical to optimize, switching
mutations. the direction of the ARMS primers allows a
An additional application of ARMS is based second attempt at test development.
on the ability of the system to selectively am- There are a number of other parameters,
plify rare mutant sequences in a background of some common to all PCR amplifications and
wild-type DNA. This makes ARMS testing some specific to ARMS reactions, that are im-
suitable for cancer screening, where only a portant in test design and application.
small fraction of the DNA being tested may ARMS primers. The ARMS primers must
carry a mutation (Sidransky et al., 1992). be of good quality. There will always be a small
Multiplex test development. Development fraction of failure sequences in any oligonu-
of multiplex ARMS tests is more complex than cleotide synthesis reaction, but because oli-
for single tests, but once optimized the use of gonucleotides are synthesized in a 3′→5′ direc-
the tests is straightforward and their perform- tion, failures will not produce truncated primers
ance is reliable (Ferrie et al., 1992; Fortina et that can allow mispriming. It is unnecessary to
al., 1992). A particular advantage of the ARMS purify the ARMS primers.
approach is that there is no need for the muta- PCR contamination. As with all PCR–
tions to be located in the same exon, or even the based techniques, precautions must be taken to
same gene, because the method does not require prevent contamination with the products of
initial amplification of a region of DNA for earlier reactions. This is particularly the case
subsequent analysis. when an ARMS test is used regularly as part of
a screening program. Because the tests are
Critical Parameters and stable at least 6 months at −20°C, it is recom-
Troubleshooting mended that a large batch of test aliquots be
Criteria for success. The key to developing prepared under clean conditions and stored for
a successful ARMS reaction is to ensure that future use. It is also good practice to use sepa-
the ARMS primer allows efficient amplifica- rate laboratories for PCR setup and analysis,
tion from the target allele while minimizing preferably under positive and negative air pres-
amplification from the nontarget allele. This sure respectively. See APPENDIX 2A for additional
unit outlines a reliable iterative approach to this guidelines for preventing PCR contamination.
problem. At present, the rules governing ARMS DNA samples. The least controllable vari-
primer specificity are not completely under- able in the ARMS test process is the test DNA
stood and the available ARMS primer design sample. The support protocol outlines a rapid
guidelines are based on empirical observations. DNA extraction process that has been shown to
These guidelines generally work well, but it work well with the reaction conditions described
should be noted that the DNA sequences around in this unit. One potential concern over the use
the priming sites are also important. The com- of mouthwash samples is inhibition of the PCR
bination of terminal and penultimate mis- by foodstuffs carried over in the sampling pro-
matches that gives an optimized ARMS test cedure. In a limited study of likely contaminants
may not be that initially selected using Table (snack foods, fruit, chocolate, coffee, etc.), no
9.8.1. inhibition was observed. Nevertheless, as a pre-
ARMS
Analysis of Point Although the test optimization protocol is caution it is recommended that no food be con-
Muataions

9.8.10
Supplement 3 Current Protocols in Human Genetics
sumed for 30 min prior to sampling. 100-fold. If this fails, it may be helpful to boil
DNA prepared by other methods is also the sample 5 min in 50 mM NaOH, then neu-
suitable for ARMS reactions. Any problems tralize by adding 1⁄10 vol of 1 M Tris⋅Cl, pH 7.5.
encountered are usually due to the presence of Other parameters. Other potential prob-
PCR inhibitors in the sample and can generally lems with ARMS are addressed in the trou-
be solved by diluting the DNA sample 10- to bleshooting guide in Table 9.8.4.

Table 9.8.4 Troubleshooting Guide

Problem Possible cause Solution

ARMS and control bands faint Inactive enzyme Use fresh batch of Taq DNA
and/or absent in all reactions polymerase
Component omitted from Repeat ARMS from
reaction reaction preparation

ARMS bands faint and/or absent Failure to denature high- Increase denaturation
in all reactions; control bands nor- G+C ARMS product temperature to 96°C
mal
Add 10% (v/v) glycerol to
mix
Insensitive ARMS test Increase primer
concentration, decrease
mismatch strength

ARMS and control bands faint Sample-borne PCR Dilute sample 1:100
and/or absent in some reaction inhibition
pairs
Insufficient DNA in affected Add 10× more sample
samples

ARMS bands absent in some reac- Control reaction too Reduce concentration of
tions; control bands normal efficient control primers
Faint additional bands same size Contamination Check negative controls
as ARMS products
Excess enzyme Check enzyme dilution
Nonspecific ARMS test Reduce primer
concentration, increase
mismatch strength

Faint additional bands, different Excess enzyme Check enzyme dilution

size from ARMS products
Cooling artifacts Allow reaction tubes to cool
to room temperature in PCR
block
Spurious priming Identify specific primer(s)
elsewhere in genome causing problem and
redesign
Streakiness in some lanes Excess DNA Add less DNA
PCR products ∼50-60 bp Primer dimers Ensure hot start is used
Check 3′ ends of primers for
complementarity
First lanes on gel give extra Insufficient mixing of Increase mixing
bands; later lanes give faint bands enzyme dilution
Clinical
Molecular
Diagnostics

9.8.11
Current Protocols in Human Genetics Supplement 7
 Anticipated Results conductance regulator gene by multiplex allele
DNA samples that contain the target allele specific polymerase chain reaction. Hum. Genet.
90:375-378.
will allow amplification of an ARMS product.
The use of two separate ARMS reactions (nor- Mullis, K.B. 1991. The polymerase chain reaction
in an anemic mode: How to avoid cold oligode-
mal and mutant) will allow discrimination of
oxyribonuclear fusion. PCR Meth. Appl. 1:1-4.
homozygotes and heterozygotes for a given
Newton, C.R., Graham, A., Heptinstall, L.E., Pow-
allele. For some applications (e.g., genetic
ell, S.J., Summers, C., Kalsheker, N., Smith, J.,
screening) such discrimination may not be re- and Markham, A.F. 1989. Analysis of any point
quired and the normal ARMS reaction may be mutation in DNA: The amplification refractory
omitted. Examples of typical results from a mutation system (ARMS). Nucl. Acid Res.
single ARMS test for the R117H mutation of 17:2503-2516.
the CFTR gene and a multiplex ARMS test for Okayama, H., Curiel, D.T., Brantly, M.L., Holmes,
four additional mutations of the CFTR gene are M.D., and Crystal, R.G. 1989. Rapid nonra-
dioactive detection of mutations in the human
given in Figure 9.8.2.
genome by allele specific amplification. J. Lab.
Clin. Med. 114:105-113.
Time Considerations
Sarkar, G., Cassady, J., Bottema, C.D.K., and Som-
Following optimization, the complete proc- mer, S.S. 1990. Characterization of polymerase
ess from sample preparation to analysis can chain reaction amplification of specific alleles.
easily be performed in a single day. The ther- Anal. Biochem. 186:64-68.
mal-cycling parameters have not been opti- Sidransky, D., Tokino, T., Hamilton, S.R., Kinzler,
mized for a rapid thermal cycler such as the K.W., Levin, B., Frost, P., and Vogelstein, B.
Perkin-Elmer Cetus 9600; use of such an in- 1992. Identification of ras oncogene mutations
in the stool of patients with curable colorectal
strument could allow further reduction in the
tumors. Science 256:102-105.
time needed to complete the protocol.
Sommer, S.S., Groszbach, A.R., and Bottema,
Any commercial inquiries regarding ARMS
C.D.K. 1992. PCR amplification of specific al-
should be addressed to Cellmark Diagnostics. leles (PASA) is a general technique for rapidly
ARMS is the subject of European Patent No. detecting known single base changes. BioTech-
0332435 (ZENECA). niques 12:82-86.
Wu, D.Y., Ugozzoli, L., Pal, B.K., and Wallace, R.B.
Literature Cited 1989. Allele specific enzymatic amplification of
Dean, M., White, M.B., Amos, J., Gerrard, B., Ste- β-globin genomic DNA for the diagnosis of
wart, C., Khaw, K.-T., and Leppert, M. 1990. sickle cell anemia. Proc. Natl. Acad. Sci. U.S.A.
Multiple mutations in highly conserved residues 86:2757-2760
are found in mildly affected cystic fibrosis pa-
tients. Cell 61: 863-70. Key Reference
Ferrie, R.M., Schwarz, M.J., Robertson, N.H., Ferrie et al., 1992. See above.
Vaudin, S., Super, M., Malone, G., and Little, S. Provides background information on ARMS and
1992. Development, multiplexing and applica- gives clinical data that support the reliability claims
tion of ARMS tests for common mutations in the for the method.
CFTR gene. Am. J. Hum. Genet. 51:251-262.
Fortina, P., Conant, R., Monokian, G., Dotti, G.,
Parrella, T., Hitchcock, W., Kant, J., Scanlin, T.,
Rappaport, E., Schwartz, E., and Surrey, S. 1992.
Contributed by Stephen Little
Nonradioactive detection of the most common Zeneca Diagnostics
mutations in the cystic fibrosis transmembrane Cheshire, United Kingdom

ARMS
Analysis of Point
Muataions

9.8.12
Supplement 7 Current Protocols in Human Genetics