Research Reports
O. Henegariu, N.A. Heerema, S.R. Dlouhy, G.H. Vance and P.H. Vogt1
Indiana University, Indianapo-lis, IN, USA and 1Heidelberg University, Heidelberg, Germany
ABSTRACT
By simultaneously amplifying more than one locus in the same reaction, multiplex PCR is becoming a rapid and convenient screening assay in both the clinical and the research laboratory. While numerous pa-pers and manuals discuss in detail condi-tions influencing the quality of PCR in gen-eral, relatively little has been published about the important experimental factors and the common difficulties frequently en-countered with multiplex PCR. We have ex-amined various conditions of the multiplex PCR, using a large number of primer pairs. Especially important for a successful multi-plex PCR assay are the relative concentra-tions of the primers at the various loci, the concentration of the PCR buffer, the cycling temperatures and the balance between the magnesium chloride and deoxynucleotide concentrations. Based on our experience, we propose a protocol for developing a mul-tiplex PCR assay and suggest ways to over-come commonly encountered problems. INTRODUCTION
Multiplex polymerase chain reac-
tion (PCR) is a variant of PCR in which
two or more loci are simultaneously
amplified in the same reaction. Since
its first description in 1988 (6), this
method has been successfully applied
in many areas of DNA testing, includ-
ing analyses of deletions (2,8), muta-
tions (14) and polymorphisms (11), or
quantitative assays (10) and reverse-
transcription PCR (7).
The role of various reagents in PCR
has been discussed (3,9,12,13), and
protocols for multiplex PCR have been
described by a number of groups. How-
ever, few studies (5,15) have presented
an extensive discussion of some of the
factors (e.g., primer concentration, cy-
cling profile) that can influence the re-
sults of multiplex analysis. In this
study, over 50 loci were amplified in
various combinations in multiplex
PCRs using a common, KCl-containing
PCR buffer. Because of specific prob-
lems associated with multiplex PCR,
including uneven or lack of amplifica-
tion of some loci and difficulties in re-
producing some results, a study of the
parameters influencing the amplifica-
tion was initiated. Based on this experi-
ence, a step-by-step multiplex PCR
protocol was designed (Figure 1), with
practical solutions to many of the prob-
lems encountered. This protocol should
be useful to those using PCR technolo-
gy in both the research and the clinical
laboratories.
MATERIALS AND METHODS
Standard Solutions and Reagents for
the PCR
Nucleotides (dNTP) (Pharmacia
Biotech [Piscataway, NJ, USA] or
Boehringer Mannheim [Indianapolis,
IN, USA]) were stored as a 100 mM
stock solution (25 mM each dA TP,
dCTP, dGTP and dTTP). The standard
10×PCR buffer was made as described
(Perkin-Elmer, Norwalk, CT, USA) and
contained: 500 mM KCl, 100 mM Tris-
HCl, pH 8.3 (at 24°C) and 15 mM
MgCl2. Taq DNA Polymerase was pur-
chased from Life Technologies
(Gaithersburg, MD, USA) or from
Perkin-Elmer. Dimethyl sulfoxide
(DMSO), bovine serum albumin (BSA)
and glycerol were purchased from Sig-
ma Chemical (St. Louis, MO, USA).
Primers were either commercially ob-
tained (Genosys [The Woodlands, TX,
USA] or Research Genetics [Hunts-
ville, AL, USA]) or synthesized locally
and were used in a final concentration
of 10–25 pmol/µL each. One set of
primer pairs (sY) was used to map dele-
tions on the human Y chromosome
(8,16). Another 10–15 primer pairs
were for the Duchenne muscular dys-
trophy (DMD) gene on human chromo-
some X (4). Other primers represent
various polymorphic loci (microsatel-
lites) on human chromosome 12 (Re-
search Genetics). Primers were com-
bined in multiplex mixtures as
described in Table 1 and Figures 2b, 3b
Multiplex PCR: Critical Parameters and Step-by-Step Protocol
BioTechniques 23:504-511 (September 1997)
and 5e. Genomic DNA was prepared
using a standard sodium dodecyl sul-fate (SDS)/proteinase K protocol
(Boehringer Mannheim).
Basic PCR Protocol
The basic PCR (25 µL vol) includ-
ed: autoclaved ultra-filtered water; PCR buffer (1×); dNTP mixture (200µM each); primer(s) (0.04–0.6 µM each); DMSO, glycerol or BSA (5% - if used); Taq DNA polymerase (1–2 U/25µL) and genomic DNA template (150 ng/25 µL). The components of the re-action can be added in any order, pro-vided that water is added first. Pipetting was done on ice, and the vials were placed from ice directly into the pre-heated metal block or water bath (94°C) of the thermal cycler. For ra-dioactive labeling, 1 µCi [32P]dCTP (Amersham, Arlington Heights, IL,
USA) was added to a 100 µL master
mixture immediately before setting up
the reaction. Results of PCR were the
same when 100- or 25- or 6.2-µL reac-
tion volumes were used. With smaller
volumes, pipetting is critical, especial-
ly for dNTP. Various thermal cyclers
were used during these studies and,
with minor cycling adjustments, all
performed well.
Gel Analysis of PCR Products
The PCR products of non-polymor-
phic loci (chromosomes X and Y) were
separated by electrophoresis on 3%
SeaKem®LE or NuSieve®(3:1)
Agarose Gels (FMC BioProducts,
Rockland, ME, USA) in 1×TAE [0.04
M Tris-acetate; 0.001 M EDTA (pH
8.0)] or 1×TBE [0.09 M Tris-borate;
0.002 M EDTA (pH 8.0)] buffer, re-
spectively, at room temperature using
voltage gradients of 7–10 V/cm. For
any given gel analysis, the same vol-
ume of PCR products was loaded in
each gel slot. Results were visualized
after staining the gels in 0.5–1 µg/mL
ethidium bromide. Sequencing gels
(6% polyacrylamide [PAA]/7 M urea)
were used for separation of the PCR
products when the loci tested were
polymorphic or a higher resolution was
required. The equivalent of about 0.2
µL radioactively labeled PCR product
was loaded in each gel lane, after mix-
ing it in loading buffer. These gels were
run in 0.6×TBE at 1800–2000 V (60
A) for about 2 h. Autoradiographs were
obtained after overnight exposure.
RESULTS AND DISCUSSION
Based on many experiments, a pro-
tocol for establishing a multiplex PCR
has been designed (Figure 1), including
a number of practical solutions to some
of the most commonly encountered
problems. For convenience and ease of
use, the words in italic characters link
the scheme with various points present-
ed in Materials and Methods and the
following subsections.
Basic Principles of the Multiplex
PCR
DNA primers (Steps 1 and 2).
Primer selection followed simple rules:
primer length of 18–24 bp or higher
and a GC content of 35%–60%, thus
having an annealing temperature of
55°-58°C or higher. Longer primers
(DMD primers, 28-30 bp) allowed the
reaction to be performed at a higher an-
nealing temperature and yielded less
unspecific products. To calculate the
melting point and test for possible
primer-primer interactions, “Primers
1.2” (a freeware that can be down-
loaded from ftp.bio.indiana.edu) was
used. To test for possible repetitive se-
quences, many of the primers used
were aligned with the sequence data-
bases at the National Center for
Biotechnology Information (NCBI) us-
ing the Basic Local Alignment Search
Tool (BLAST) family of programs.
Single locus PCR (Step 3).A PCR
Figure 1. Step-by-step protocol for the multiplex PCR.
Research Reports program to amplify all loci individually
was designed. Reaction mixture includ-
ed 1×PCR buffer, 0.4 µM each primer,
5% DMSO and 1 U Taq DNA poly-
merase/25 µL reaction volume. Results
of PCR were compared when the reac-
tions were done consecutively in the
same thermal cycler, or in parallel, in
machines of the same model and in ma-
chines of different models or manufac-
turers. Results were very reproducible
when the same machine or same ma-
chine model was used but could
markedly differ when the same exact
PCR program was used on thermal cy-
clers from different manufacturers.
However, with adjustments in only the
cycling conditions, results became re-
producible even in different types of
machines. We have observed that for
the loci tested (100–300-bp long), yield
of some products was increased by de-
creasing the extension temperature. For
individually amplified loci, the anneal-
ing time (from 30–120 s) and the exten-
sion time (from 30–150 s) did not visi-
bly influence the results, but the
specificity and yield of PCR product
were increased or decreased by
changes in annealing temperature. To
amplify the 22 Y-specific loci (Figure
2a), PCR program A gave best results
(Table 2).
Multiplex PCR: equimolar primer
mixture (Step 4).Combining the pri-
mers in various mixtures and amplify-
ing many loci simultaneously (Table 1
and Figure 2b), required alteration/opti-
mization of some of the parameters of
the reaction. When the multiplex reac-
tion is performed for the first time, it is
useful to add the primers in equimolar
amounts. The results will suggest how
the individual primer concentration and
other parameters need to be changed.
Examples of some useful changes are
illustrated and discussed below; howev-
er, these examples do not necessarily
follow the exact order as listed in the
protocol (Figure 1) since a number of parameters (e.g., extension tempera-ture) are referred to more than once.
Optimization of Multiplex PCR Cycling Conditions
Extension temperature (Step 5, A–C).Figure 2c illustrates the results obtained when four different amplifica-tion mixtures containing equal amounts
(0.4 µM each) of different Y-chromo-
some primers were subjected to multi-
plex PCR with program A and program
B (Table 2); the latter program had a
higher extension temperature (72°C)
and longer annealing and extension
times. In general, there was a visibly
higher yield of PCR products for mix-
tures Y-1, Y-3* and Y-4 with program
A. In addition, with program B, some
products are missing (in Y-1 and Y-2)
and some unspecific products appear
(in Y-1 and Y-3*). The results with
program B were considered less desir-
able overall and suggested that the
higher extension temperature in pro-
gram B decreased the amplification of Name Size Name Size Name Size Name Size (locus)(bp)(locus)(bp)(locus)(bp)(locus)(bp) Y-1Y-2Y-3Y-4
sY84 326sY143311sY86320sY14472 DYS273DYS231DYS148SRY
sY134 301sY157285sY105301sY95303 DYS224DYS240DYS201DYS280
sY117262sY81209sY82264sY127274
DS209DYS271DYS272DYS218
sY102 218sY182125Y6HP35226sY109233 DYS198KAL Y DYS274DYF43S1
sY151 183sY147100Y6PHc54166sY149132 KAL Y DYS232n.a.DYS1
sY94 150sY153139
DYS279DYS237
sY88 123sY97104
DYS276DYS281
DMD exon Size DMD exon Size Name Size No.(bp)No. (bp)(locus)(bp) X-1X-312-1 No. 45547PM535AFM263zd1 317-341
D12S332
PM535No. 3410AFM205ve5 271-291
D12S93
No. 19459No. 50271AFM205xg3 243-253
D12S310
No. 17416No. 6202AFM211wb6 228-238
D12S98
No. 51388No. 60139AFM206ze5 183-201
D12S94
No. 8360AFM299zd5 165-181
D12S349
No. 12331AFM135xe3 142-168
D12S87
No. 44268AFM122xf6 105-125
D12S85
No. 4196
n.a. = locus not assigned.
PM = promoter region
Table 1. List of Primers Used in the Multiplex Mixtures
some loci, even though we tried to compensate using a longer annealing time and slightly longer extension time.Extension time (Step 5, A, B and D).In multiplex PCR, as more loci are simultaneously amplified, the pool of enzyme and nucleotides becomes a limiting factor and more time is neces-sary for the polymerase molecules to complete synthesis of all the products.Two experiments illustrated the
influ-ence of the extension time. In one ex-periment, a Y -chromosome primer pair (Y6BaH34pr, 910bp) was added to a X-chromosome primer mixture (X-3).The results (Figure 3b) showed that in-creasing the extension time in the mul-tiplex PCR (program A vs. program D)increased the amount of longer prod-ucts. In another experiment, four Y multiplex mixtures were amplified us-ing PCR programs C and A (Figure 3a and Table 2). Visibly higher yields of PCR products were obtained for all Y mixtures when a longer extension time was used.
Annealing time and temperature (Step 5, A–D; Figure 1).Modification of the annealing time from 20 s to 2min did not alter the amplification effi-ciency (not shown), but the annealing temperature was one of the most im-portant parameters. Although many in-dividual loci could be specifically am-plified at 56°–60°C, our experience showed that lowering the annealing temperature by 4°–6°C was required for the same loci to be co-amplified in multiplex mixtures. This is demonstrat-ed in Figure 3, d–f, which depict an op-timal multiplex annealing temperature of 54°C for primers individually usable at 60°C. At 54°C, although unspecific amplification probably occurs (e.g.,Figure 3c), it is overcome by the con-current amplification of an increased number of specific loci in the multiplex reaction and thus remains invisible.Similarly, when many specific loci are simultaneously amplified, the more ef-ficiently amplified loci will negatively influence the yield of product from the less efficient loci. This is due to the fact that P
CR has a limited supply of en-zyme and nucleotides, and all products compete for the same pool of supplies. Number of PCR cycles. Primer mixture Y -3* was used to amplify two different genomic DNA samples, stop-Figure 2.(a) Single-locus PCR. Amplification of the sY loci using 1×PCR buffer and program A. On the gel, the products are arranged in increasing order of sY number (1=sY14, 2=sY81, 3=sY82, 4=sY84,5=sY86, 6=sY88, 7=sY94, 8=sY95, 9=sY97, 10=sY102, 11=sY105, 12=sY109, 13=sY117, 14=sY127,15=sY134, 16=sY143, 17=sY147, 18=sY149, 19=sY151, 20=sY153, 21=sY157 and 22=sY182). All products had the expected length, and there was no visible unspecific amplification. In all gels, lanes without a label show the size marker (1-kb ladder; Life Technologies). (b) Optimized multiplex reactions.Multiplex PCR with primer mixtures Y-1 (sY84, sY134, sY117, sY102, sY151, sY94 and sY88), Y-2(sY143, sY157, sY81, sY182 and sY147), Y-3 (sY86, sY105, sY82, Y6HP35, Y6Phc54, sY153 and sY97) and Y-4 (sY14, sY95, sY127, sY109 and sY149) in 1.6×PCR buffer (PCR program E). Mix Y-3*is mixture Y-3 without primers Y6HP35 and Y6Phc54. Arrows indicate the expected amplification prod-ucts. (c) Extension temperature. Multiplex PCR with mixtures Y-1 to Y-4 with PCR programs A and B (Table 2). All amplification products are visible in the first four lanes (extension at 65°C). In the last four lanes (extension at 72°C), bands are missing in Y-1 and Y-2, and unspecific products appear in Y-1 and Y-3*. Length marker in all figures = 1-kb ladder. In all images, electrophoresis was conducted from top to bottom. Program A
Program B Program C First Denaturing 94°C, 4 min 94°C, 4 min 94°C, 4 min Denature 94°C, 30 s 94°C, 30 s 94°C, 30 s Anneal 54°-56°C, 30 s*54°C, 1 min
54°C, 45 s
Extend 65°C, 1 min
72°C, 1 min, 20 s 65°C, 2 min 32 cycles 32 cycles 32 cycles Final Extension
65°C, 3 min 72°C, 3 min 65°C, 3 min Program D
Program E Program F First Denaturing 94°C, 4 min 94°C, 4 min none Denature 94°C, 30 s 94°C, 30 s 94°C, 30-45 s Anneal 55°C, 30 s 54°C, 45 s 56°-58°C, 45 s Extend 65°C, 4 min 65°C, 2 min 68°C, 2 min, 30 s 32 cycles 45 cycles 35 cycles Final Extension
65°C, 3 min
65°C, 5 min
none
Bold characters show most important modifications when programs are com-pared.
*Program A was used with two different annealing temperatures, according to the type of PCR amplification (see Results and Discussion).
Table 2. Cycling Conditions/PCR Programs
Research Reports
ping the reaction after increasing num-bers of cycles (Figure 4a). One of the two genomic DNAs was a better tem-plate, possibly due to the higher quality and/or amount of DNA. Both of them,however, show a gradual increase in the yield of all bands with the number of cycles. The most obvious variation in the amount of products was around 25cycles (for ethidium bromide-stained gels). Twenty-eight to thirty cycles are usually sufficient for a reaction; little is gained by increasing cycle number up to 60.
Optimization of Multiplex Reaction Components
Initially, there was some variation from test to test when the same PCR program was used (e.g., Figures 2c and 3a). Solving this reproducibility prob-lem required adjustments of PCR com-ponents.
Amount of primer (Step 5, B and C). Initially, equimolar primer concen-trations of 0.2–0.4
µM each were used in the multiplex PCR (Figure 3c), but there was uneven amplification, with some of the products barely visible even after the reaction was optimized for the cycling conditions. Overcoming this problem required changing the pro-portions of various primers in the reac-tion, with an increase in the amount of primers for the “weak” loci and a de-crease in the amount for the “strong”l
oci. The final concentration of the
primers (0.04–0.6
µM) varied consider-ably among the loci and was estab-lished empirically.
dNTP and MgCl 2concentrations (Step 5D).
dNTP .The significance of the dNTP concentration was tested in a multiplex PCR test with primer mixture Y-4.Magnesium chloride concentration was kept constant (3 mM), while the dNTP concentration was increased stepwise
from 50–1200
µM each (Figure 4b).The best results were at 200 and 400µM each dNTP, values above which the amplification was rapidly inhibited.
Lower dNTP concentration (50
µM) al-lowed PCR amplification but with visi-bly lower amounts of products. dNTP stocks are sensitive to thawing/freezing cycles. After 3–5 such cycles, multi-plex PCRs often did not work well;
products became almost completely in-visible. To avoid such problems, small aliquots (2–4 µL, 10–20 reactions) of dNTP (25 mM each) can be made and kept frozen at -20°C and centrifuged before use. This “low stability” of dNTP is not so obvious when single loci are amplified.
MgCl 2. A recommended magne-sium chloride concentration in a stan-dard PCR is 1.5 mM at dNTP con-centrations of around 200
µM each. To test the influence of magnesium chlo-ride, a multiplex PCR (mixture Y -3)was performed, keeping dNTP concen-tration at 200 µM and gradually in-creasing magnesium chloride from 1.8–10.8 mM (Figure 4c). Amplifica-tion became more specific (unspecific bands disappeared), and the products acquired comparable intensities (at 10.8 mM). In PCRs with up to 20 mM MgCl 2, products became barely visible,as if the reactions were inhibited (not shown).
reaction tooldNTP/MgCl 2balance.To work properly, Taq DNA polymerase re-quires free magnesium (besides the
Figure 3.(a) Extension time. Multiplex PCR of mixtures Y-1 to Y-4, comparing PCR programs C (2-min
extension time) and A (1-min extension time, 54°C annealing temperature). Comparison of equivalent lanes shows an improvement in yield when extension time is 2 min. Some faint unspecific bands appe
ar,possibly due to the low buffer concentration (1×). (b) Extension time. Multiplex PCR with mixture X-3(primers for DMD gene exons Nos. PM, 3, 50, 6, 60) and primer pair Y6BaH34 (910-bp product, upper arrow). Primers giving shorter amplification products are preferentially amplified with short extension times (1-min, program A). (c) Equimolar primer mixture. PCR with individual primer pairs of mixture 12–1 (separate and multiplex), using program F. Products are arranged on the gel according to their de-creasing length. Individual products have comparable intensities. When equimolar amounts of primers were mixed for the multiplex reaction (first lane), some products were not efficiently amplified but un-specific products disappeared. (d–f) Annealing temperature, buffer concentration and number of primers.Multiplex amplification of mixture Y-3* (first three lanes in each gel), primer pair sY 153 (lanes 4–6)and mixture Y-3 (lanes 7–12 in 1×or 2×PCR buffer) on three different template DNAs using three PCR programs differing in annealing temperature (48°, 54°or 59°C). Lanes 1–9 on each gel show reactions in 1×PCR buffer. Lanes 10–12 on each gel show reactions in 2×PCR buffer. Lanes 7–12 on each gel (under 1×PCR and 2×PCR) were with primer set Y-3. The very last lane in Figure 3, d and f is the marker (1-kb ladder). Small horizontal arrows indicate the expected products of mixture Y-3* (five products) including the longest specific product on the gel. Oblique arrow (3e) indicates a strong unspecific product. Solid arrowheads indicate the two extra products expected in mixture Y-3 (total of seven products) compared with Y-3*. Arrowhead outlines show positions of some missing products (e.
g., 3e, first lane). With mul-tiplex amplification at 48°C, many unspecific bands appear. In 1×PCR buffer, the sY153 product is stronger when amplified in mixture Y-3* (5 primer pairs) than in mixture Y-3 (7 primer pairs), which shows that at least for some products, an increased number of simultaneously amplified loci can influ-ence the yield at some specific loci. Raising the PCR buffer concentration from 1×to 2×allows a more even amplification of all specific products and helps to decrease the intensity of many longer unspecific products (compare lanes 7–9 vs. 10–12). The strong 470–480-bp unspecific band (oblique arrow) seen with 2×buffer was eliminated by varying the proportion of different primers in the reaction (compare with Y-3, Figure 2b). At 59°C the sY153 product can be seen only when 2×buffer is used or when the lo-cus is amplified alone.
magnesium bound by the dNTP and the
DNA) (9). This is probably why in-creases in the dNTP concentrations
(Figure 4b) can rapidly inhibit the PCR,
whereas increases in magnesium con-
centration often have positive effects (Figure 4c). By combining various amounts of dNTP and MgCl2, i
t was found that 200 µM each dNTP work well in 1.5–2 mM MgCl2, whereas 800µM dNTP require at least 6–7 mM MgCl2. The threshold for the reaction was roughly 1 mM MgCl2when 200µM dNTP was used, with reduced PCR amplification below this MgCl2con-centration.
PCR buffer (KCl) concentration.
Comparison of PCR buffers (Step 5, B–D).
KCl or PCR buffer concentration.
Raising the buffer concentration to 2×(or only the KCl concentration to 100
mM) improved the efficiency of the multiplex reaction (Figure 4d and also
Figure 3, d–f), this effect being more
important than using any of the adju-
vants tested (DMSO, glycerol or BSA). Generally, primer pairs with longer am-plification products worked better at lower salt concentrations, whereas primer pairs with short amplification products work
ed better at higher salt concentrations, where longer products become harder to denature (compare 0.4×with 2.8×in Figure 4d). For exam-ple, pair sY94 (melting point ca. 58°C) is favored over both sY88 (melting point ca. 58°C) and sY151 (melting point ca. 52°C) at 0.8×buffer but not at higher salt concentrations. The proper buffer concentration may help over-come other factors (product size, GC
Figure 4.(a) Number of cycles. Amplification with two different DNA templates using primer mixture Y-3* in 1.4×PCR buffer, with increasing numbers of cycles by units of three. (b) dNTP concentration. PCR amplification using mixture Y-4 in 2×PCR buffer (3 mM MgCl2) and increasing concentrations of dNTP (50, 100, 200, 400, 600 and 1200 µM). Most efficient amplification is seen at concentrations of 200–400 µM dNTP. Further increase in the dNTP concentration inhibits the reaction when MgCl2 concentration is kept constant. (c) MgCl2concentration. Multiplex PCR was performed with mixture Y-3 in 1.4×PCR buffer, using PCR program E and gradually raising the concentration of MgCl2. (d) PCR buffer concentration. Amplification products of mixture X-1 (DMD gene exons Nos. 45, PM, 19, 17, 51, 8, 12, 44 and 4) using increasing concentrations of PCR buffer and program E. As the stringency in the reaction mixture decreases, shorter products are amplified more efficiently, whereas the intensity of longer products gradually decreases. For this particular primer mixture, the optimal buffer concentration was 1.2×–1.6×. (e) Comparison of PCR buffers. Comparison of multiplex PCR of mixture
X-1 in the DMD buffer and the 1.6×KCl-based PCR buffer, using the same proportion of ingredients (DNA, Taq DNA polymerase, primer amount) and PCR program E. For every DNA sample tested, the amounts of products were increased when 1.6×PCR buffer was used. Only four lanes are shown, although the gel had more samples loaded, and identical results were observed. (f) Amount of template DNA. Various amounts of template DNA were amplified with primer sY153 and mixture Y-3* in 2×PCR buffer with program E. Reaction volumes were 25 µL. There were no major differences using 500 or 30 ng DNA; however, some bands became weaker as the DNA amount was further decreased to 0.5 ng/25 µL reac-tion. No major differences due to the DNA template concentration were seen when primer pair sY153 was used alone.
版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。
发表评论