Downloaded from http://cshprotocols.cshlp.

org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by

Cold Spring Harbor Laboratory Press

Protocol

Amplification of cDNA Generated by Reverse Transcription

of mRNA: Two-Step Reverse Transcription-Polymerase Chain
Reaction (RT-PCR)
Michael R. Green and Joseph Sambrook

Reverse transcription-polymerase chain reaction (RT-PCR) is a powerful method to detect and syn-
thesize cDNA copies of low-copy-number mRNAs. Two enzymes are used: reverse transcriptase to
produce single-stranded cDNA copies, which are then used as templates in an amplification reaction
catalyzed by a thermostable DNA polymerase. For this reason, the method is known as “two-step RT-
PCR.” This protocol describes the traditional method of RT-PCR in which the two synthetic reactions
are performed separately and sequentially.

MATERIALS

It is essential that you consult the appropriate Material Safety Data Sheets and your institution’s Environmental
Health and Safety Office for proper handling of equipment and hazardous materials used in this protocol.

Reagents
Amplification buffer (10×) containing MgCl2 (as supplied with the thermostable DNA polymerase)
dNTP mix (all four dNTPs, 20 mM each; pH 8.0)
Ethidium bromide (optional; see Step 12)
Exogenous reference RNA (optional; see Step 6.iv)
Gel, agarose or polyacrylamide (see Step 12)
MgCl2 (50 mM)
Oligo(dT)12–18 (100 µg/mL, in TE; pH 8.0) (see Step 3 and Discussion section entitled “Primers for
First-Strand Synthesis”)
Oligonucleotide, gene-specific (20 pmol/µL, in H2O) (see Step 3 and Discussion section entitled
“Primers for First-Strand Synthesis”)
The gene-specific oligonucleotide should be complementary to a known sequence in the target mRNA.

Oligonucleotide primers for cDNA amplification, sense and antisense

The primer used to generate cDNA can also be used as the antisense primer in the amplification stage of standard
RT-PCR. However, the specificity of amplification can be improved by using an antisense primer that binds to an
upstream sequence in the target transcript. Both sense and antisense primers are gene-specific synthetic oligo-
nucleotides that should be 20–30 nt in length and contain approximately equal numbers of the four bases, with a
balanced distribution of G and C residues, and a low propensity to form stable secondary structures. Wherever
possible, use specific oligonucleotides that bind to sequences located in different exons of the target RNA as sense

From the Molecular Cloning collection, edited by Michael R. Green and Joseph Sambrook.
© 2019 Cold Spring Harbor Laboratory Press
Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190

405
 Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by
Cold Spring Harbor Laboratory Press

M.R. Green and J. Sambrook

and antisense primers for amplification of the cDNA product. In this way, amplification products derived from
cDNA and contaminating genomic DNA can be easily distinguished.
Bases encoding restriction sites can be added to the 5′ termini of the sense and antisense primers used in PCR. This
addition greatly facilitates cloning of the amplified product.
Oligonucleotide primers synthesized on an automated DNA synthesizer can generally be used in standard
RT-PCR without further purification. However, amplification of low-abundance mRNAs is often more efficient
if the oligonucleotide primers are purified by chromatography on commercially available resins or by denaturing
polyacrylamide gel electrophoresis.
Use the following formula to calculate the molecular mass of the oligonucleotides:

Mr = (C × 289) + (A × 313) + (T × 304) + (G × 329),

where C is the number of C residues in the oligonucleotide, A is the number of A residues, T is the number of
T residues, and G is the number of G residues. The molecular mass of a 20-mer will be 6000 Da; 100 pmol of the
oligonucleotide will be equivalent to 0.6 µg.

Placental RNase inhibitor (20 units/µL)

Poly(A)+ RNA (10 µg/mL, in H2O) (optional; see Step 1)
Random hexanucleotides (1 mg/mL, in TE; pH 8.0) (see Step 3 and Discussion section entitled
“Primers for First-Strand Synthesis”)
Reverse transcriptase (RNA-dependent DNA polymerase)
See the Discussion section entitled “Reverse Transcriptases Used in Two-Step RT-PCR” below.

SYBR Gold (optional; see Step 12)

Thermostable DNA-dependent DNA polymerase
Taq DNA polymerase is the standard and appropriate enzyme for the amplification stage of most forms of
RT-PCR. However, where elongation from 3′ -mismatched primers is suspected, a thermostable DNA polymerase
with 3′ → 5′ proofreading activity might be preferable (see Tables 1 and 2).

Total RNA (100 µg/mL, in H2O) (see Step 1 and Box 1)

To reduce the chance of contamination with exogenous DNAs, prepare and use a special set of reagents and
solutions for PCR only. Buffers and other solutions should be treated with diethylypyrocarbonate (DEPC) or a
commercial product to destroy RNase (e.g., RNaseZap [Ambion] or an equivalent product).

Equipment
Barrier tips (for automatic micropipettes)
Gel electrophoresis equipment (for polyacrylamide or agarose gels)
Ice bucket
Microcentrifuge tubes, 0.5 mL, thin-walled (optional; see Step 1)
Microtiter plates (optional; see Step 1)
Positive-displacement pipette
Thermal cycler
A number of programmable thermal cyclers marketed by different commercial companies (e.g., Mastercycler
[Eppendorf], PTC-100 [MJ Research]) are licensed by PerkinElmer for use in PCR. The choice among these
instruments depends on the investigator’s inclination, the available budget, and the range of uses to which the
machine will be put. Before purchasing a thermal cycler, we recommend soliciting as many opinions as possible
to discover the advantages and disadvantages of different machines.

Water baths or heat blocks preset to 37˚C, 75˚C, and 95˚C

Bake all glassware for 6 h at 150˚C and autoclave all plasticware.

METHOD

Before beginning work, clean all devices with RNaseZap (or any RNase commercial decontaminant). Wear gloves
throughout experiments to prevent contamination with human RNases. Change gloves after touching common sur-

406 Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190

 TABLE 1. Thermostable DNA polymerases: properties
Polymerase 3′ → 5′ 5′ → 3′ 3′ -End Recommended target length Error rate
Vendor blend Exonuclease Exonuclease modification (genomic) Uracil sensitivitya (×106)

PfuUltra II Agilent Nob Strong No Blunt ≤19 kb Yes 0.4

Herculase II Agilent Nob Strong No Blunt ≤23 kb Yes 1.3
Phusion NEB, No Strong No Blunt ≤7.5 kb Yes 1.3
Finnzymes
PicoMaxx Agilent Taq + Pfu b Weak Yes dA > blunt ≤10 kb Weak 4.0
Easy-A Agilent Taq + Strong Yes dA ≤5 kb Weak 1.3
exonucleaseb
Platinum Taq High Thermo Fisher Taq + Deep Vent Weak Yes dA > blunt ≤15 kb Weak 5.8
Fidelity
Expand High Roche Taq + Tgo Weak Yes dA > blunt ≤5–9 kb Weak 3.3
Fidelity
Expand Long Roche Taq + Tgo Weak Yes dA > blunt ≤20–27 kb Weak 7.4
PfuTurbo Agilent Nob Strong No Blunt ≤19 kb Yes 1.3
PfuTurbo Cx Agilent No Strong No Blunt ≤10 kb No (uracil-resistant 1.3
mutant)
Vent NEB No Strong No Blunt ≤2 kb Yes 2.8
Platinum Pfx Thermo Fisher No Strong No Blunt ≤12 kb Yes 3.5
Taq Multiple No No Yes dA ≤1–4 kb No 8.6
a
PCR inhibited by dU primers or dUTP-containing nucleotide mix (dATP, dCTP, dGTP, dUTP).
b
Enzyme contains dUTPase to eliminate uracil sensitivity caused by deamination of dCTP → dUTP (Hogrefe et al. 2002).

Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190

Cold Spring Harbor Laboratory Press

TABLE 2. Thermostable DNA polymerases: Applications

Routine High- Fast-cycling Long GC-rich TA dUTP/Ung
Type Examples PCR fidelity PCR PCR PCR template cDNA template cloning method

Fusion PfuUltra II,a Herculase II, Phusiona ++ +++ +++ ++/+++ +++ ++ nr nr
(Proofreader)
Blend PicoMaxx,a EasyA,a Platinum Taq High ++ + (+++, + ++/+++ ++ +++ ++/+++ +
Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by

Fidelity,a Expand High Fidelity, Expand Long EasyA)

Proofreader PfuTurbo,a PfuTurbo Cx,a Vent, Platinum Pfxa + +++ nr +/++ ++ + nr nr (+++,
PfuTurbo Cx)
Non- Taq,a Tth +++ nr nr nr + ++ (RNA template: +++ +++
proofreader Tth with Mn2+)
Relative performance is shown on a scale from + (low/poor) to +++ (highest/best). nr, not recommended.
a
Two-Step RT-PCR

407
Hot start versions available.
 Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by
Cold Spring Harbor Laboratory Press

M.R. Green and J. Sambrook

BOX 1. PREPARATION OF RNA FOR RT-PCR

Total RNA extracted from cells with chaotropic agents is generally the template of choice for RT-PCR for
mRNAs that are expressed at moderate-to-high abundance (Liedtke et al. 1994). Poly(A)+ is preferred as a
template for all forms of RT-PCR when the target mRNA is expressed at low abundance. However, when only
small numbers of cultured cells are available, RNA suitable for use as a template in RT-PCR can be prepared
as follows:

1. Pellet 10–1000 cells by centrifugation for 5 sec at 4˚C in a microcentrifuge.

2. Remove the supernatant by aspiration. Add 10–20 µL of an ice-cold lysis solution consisting of 0.5%
Nonidet P-40 (NP-40), 10 mM Tris–Cl (pH 8.0), 10 mM NaCl, and 3 mM MgCl2 . Vortex the tube very
gently to disperse the cells throughout the lysis solution.
Note that it might be necessary to perform preliminary experiments to determine the optimum concen-
tration of NP-40 in the lysis buffer. The concentration of 0.5% recommended in this protocol works well
for lymphocytes. Higher concentrations could be required for other types of mammalian cells.
3. Incubate the tube for 5 min on ice, then centrifuge at maximum speed for 2 min at 4˚C in a
microcentrifuge.
Before use as a template in PCR, remove contaminating cellular DNA by treatment with RNase-free
DNase (see below).
RT-PCR can succeed only if the preparation of RNA is of the highest quality—intact and free of ribonu-
clease and of contamination by genomic DNA. Unfortunately, almost all preparations of eukaryotic mRNA
are contaminated with small amounts of fragmented genomic DNA, which is best removed by treatment with
RNase-free DNase I (Rio et al. 2010), as described below.

1. Dissolve the RNA in RNase-free H2O at a concentration of 1 µg/µL.

2. Combine:
RNA solution 25 µL
DNase I buffer (10 × ; supplied by manufacturer) 10 µL
RNasin 4000 units
DNase I 1000 units
H2O to 100 µL
3. Incubate the mixture for 30 min at 37˚C.
4. Add 100 µL of 2× DNase I inactivation buffer (supplied by manufacturer).
5. Add 20 µL of 3 M sodium acetate (pH 5.2).
6. Extract the solution with phenol and precipitate the RNA with 2.5 volumes of ethanol.
7. Dissolve the semidried pellet of RNA in 20 µL of DEPC-treated, sterile H2O.
8. Measure the concentration of RNA.

faces (doorknobs, handles of refrigerators, etc.). Use a set of automatic pipettors that are used solely for working with
RNA. Use barrier tips.

1. Transfer 1 pg to 100 ng of poly(A)+ mRNA or 10 pg to 1 µg of total RNA to a fresh micro-

centrifuge tube or microtiter plate well. Adjust the volume to 10 µL with H2O.
2. Denature the RNA by heating for 5 min at 75˚C, followed by rapid chilling on ice.
3. To the denatured RNA add:
Amplification buffer (10×) 2 µL
dNTP mix (20 mM each, pH 8.0) 1 µL
Oligonucleotide primers for first-strand synthesis 1 µL
Placental RNase inhibitor (20 units/µL) 1 µL
MgCl2 (50 mM) 1 µL
Reverse transcriptase (100–200 units/µL) 1 µL
H2O to 20 µL

408 Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190

 Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by
Cold Spring Harbor Laboratory Press

Two-Step RT-PCR

Depending on the experiment, oligo(dT)12–18, random hexanucleotides, or gene-specific antisense oli-

gonucleotides can be used as primers for synthesis of first-strand cDNA (see Discussion). Before use in
experiments, determine the optimum ratio of primer:template empirically for each preparation of RNA.
As a starting point, add varying amounts of primers to 20-μL reactions: Synthetic oligonucleotide
complementary to the target RNA (5–20 pmol), Oligo(dT)12–18 (0.1–0.5 µg), or random hexanucleotides
(1–5 µg).
4. If using random hexamers or oligo(dT) to prime synthesis of first-strand cDNA in Step 3,
assemble the reaction mixture at room temperature and incubate for 10 min at 25˚C before
proceeding.
Preincubation at lower temperature allows reverse transcriptase to begin primer extension and prevents
premature melting of the primer–RNA complex.
5. Prepare control reactions.
Process the controls through all subsequent steps of the protocol in parallel with the experimental
samples.

i. In the first negative control, include all of the components of the first-strand reaction, except
the RNA template.
ii. In the second negative control, include all of the components of the first-strand reaction,
except the reverse transcriptase.
iii. In the third negative control, include all of the components of the first-strand reaction,
except the primers.
These controls provide reassurance that the cDNA product is not the result of contamination or self-
priming by RNA. See Discussion.
iv. If possible, prepare positive controls containing varying amounts of an exogenous reference
RNA equipped with a set of primer annealing sites identical to those used in the authentic
target RNA (Wang et al. 1989).
With some foresight, it is often possible to generate an appropriate reference by cloning synthetic copies
of the sense and antisense primer-binding sites on either side of an exogenous DNA sequence. The
recombinant clone can then be transcribed in vitro into an RNA template that can serve as a positive
control in RT-PCR.

6. Incubate the reaction for 60 min at 37˚C.

cDNA synthesis can be measured by determining the proportion of radioactivity incorporated in reactions
that have been supplemented with 10–20 μCi of [32P]dCTP (sp. act. 3000 Ci/mmol). The size of the first-
strand cDNA molecules can be estimated by electrophoresis through alkaline agarose gels.
See Troubleshooting.

7. Inactivate the reverse transcriptase and denature the template–cDNA complexes by heating the
reaction for 5 min to 95˚C.
The reverse transcriptase should be inactivated to ensure efficient synthesis during the amplification phase of
RT-PCR (Sellner et al. 1992; Chumakov 1994). Sometimes, inactivation of reverse transcriptase by heat is
insufficient. It is then necessary to purify the first-strand cDNA by extraction with phenol:chloroform and
precipitation with ethanol before setting up the PCRs. Presumably, contaminants in the reverse transcriptase
enzyme lead to a decreased efficiency of the amplification reactions.
8. Adjust the reaction mixture such that it contains 20 pmol each of the sense and antisense
amplification primers.
The 20 pmol of antisense oligonucleotide primer in the amplification reaction includes any gene-specific
oligonucleotide used to prime the synthesis of the first-strand cDNA. The presence of excess oligonucleotide
primers can lead to amplification of undesirable (i.e., nontarget) sequences, whereas a dearth of primers
reduces the efficiency of amplification. It might be necessary to remove excess oligonucleotide and random
hexamer primers from the cDNA preparation and then to optimize the concentrations of the sense and
antisense primers in the amplification reaction.
9. To each reaction mixture, add sufficient 1× Amplification buffer to bring the volume of the
reaction mixture to 99 µL (i.e., 77 µL) and 1 µL (1–2 units) of thermostable DNA polymerase.
10. Place the tubes or the microtiter plate in the thermal cycler fitted with a heated lid.

Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190 409

 Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by
Cold Spring Harbor Laboratory Press

M.R. Green and J. Sambrook

11. Amplify the nucleic acids using the listed denaturation, annealing, and polymerization times and
temperatures:

Cycle number Denaturation Annealing Polymerization

1–35 45 sec at 94˚C 45 sec at 55˚C 1 min 15 sec at 72˚C
Last cycle 1 min at 94˚C 45 sec at 55˚C 1 min 15 sec at 72˚C
Polymerization should be performed for 1 min for every 1000 bp of length of the target DNA.
Most thermal cyclers have an end routine in which the amplified samples are incubated at 4˚C until they are
removed from the machine. Samples can be left overnight at this temperature, but should be stored there-
after at −20˚C.
12. Analyze a sample (5–10 µL) from the test reaction mixture and the four control reactions by
electrophoresis through an agarose or polyacrylamide gel. Be sure to include DNA markers of
an appropriate size. Stain the gel with ethidium bromide or SYBR Gold to visualize the
DNA.
A successful amplification reaction should yield a readily visible DNA fragment of the expected size.
The identity of the band can be confirmed by DNA sequencing. If no product is visible after 30 cycles
of amplification, add fresh Taq polymerase and continue the amplification reaction for a further 15–20
cycles.
See Troubleshooting.

13. Extract the reaction mixtures with 150 µL of chloroform.

The aqueous phase, which contains the amplified DNA, will form a micelle near the meniscus. The micelle
can then be transferred to a fresh tube with an automatic micropipette.
14. Clone the amplified products into an appropriately prepared vector.

TROUBLESHOOTING

Problem (Step 6): There is poor yield of first-strand cDNA. Degraded RNA is the cause.
Solution: Check the integrity and concentration of the mRNA preparation by denaturing agarose gel
electrophoresis. If the RNA is degraded, prepare a new batch.

Problem (Step 6): There is poor yield of first-strand cDNA. Inhibitors of reverse transcriptase, such as
sodium dodecyl sulfate or guanidinium salts, might be present in the RNA preparation.
Solution: Re-precipitate the RNA with ethanol.

Problem (Step 6): There is poor yield of first-strand cDNA. Spurious priming, which can occur when
the stoichiometry of the reverse transcription reaction is suboptimal, is present in the PCR.
Solution: Try one or more of the following.

• Check the size of the amplified products by agarose gel electrophoresis. Spurious priming generally
leads to the synthesis of shorter cDNAs, which migrate as a smear through agarose gels. If this is the
case, optimize the reverse transcriptase reaction by systematically varying the ratio of template to
primer, the concentration of Mg2+, and the annealing temperature.
• Try changing the primers (e.g., from oligo(dT) to random primers).
• Try raising the temperature of the reverse transcriptase reaction to 42˚C or higher.

Problem (Step 12): The amplification reaction is inefficient.

410 Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190

 Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by
Cold Spring Harbor Laboratory Press

Two-Step RT-PCR

BOX 2. RIBONUCLEASE H
Ribonuclease H (RNase H) catalyzes the endonucleolytic degradation of the RNA moiety of DNA–RNA
hybrids, generating oligoribonucleotides of varying chain lengths with 3′ -hydroxyl and 5′ -phosphate termini.
It was first recognized and isolated from calf thymus (Stein and Hausen 1969; Hausen and Stein 1970), but it
is now known to be present in a wide variety of mammalian tissues, yeasts, prokaryotes, and virus particles.
Many types of cells contain more than one RNase H.
In many retroviruses, RNase H is associated with the multifunctional enzyme reverse transcriptase and
performs important functions at several stages during the transcription of the viral genome into DNA. In
eubacteria, it is believed to be required for the removal of RNA primers from Okazaki fragments, for
processing of transcripts into primers used by DNA polymerase I to initiate DNA synthesis, and to remove
R-loops that provide sites for opportunistic initiation of unregulated DNA synthesis at the chromosomal
origin of replication in Escherichia coli. RNase H is presumed to perform similar functions in eukaryotic cells.
RNase H has been reported to increase markedly the inhibition of gene expression by antisense oligo-
deoxynucleotides. Hybrids between these oligonucleotides and specific sequences in mRNAs are sensitive
to degradation by the enzyme. RNase H is required for initiation of replication at the origin (ori) of colicin E1
(colE1)–type plasmids in vitro. The enzyme also seems to suppress initiation of DNA synthesis at sites other
than ori.
X-ray crystallographic analysis shows that E. coli RNase H consists of two domains, one of which
contains an Mg2+-binding site enmeshed in β-strands—a fold previously recognized in DNase I. For
further information and references, see Crouch (1990), Wintersberger (1990), Hostomsky et al. (1993),
Jung and Lee (1995), Kanaya and Ikehara (1995), Rice et al. (1996), and Crooke (1998).

Solution: It is likely there was too little first-strand cDNA. To remedy the problem, try one or more of
the following.
• Increase the amount of cDNA in the amplification reaction.
• Optimize the PCR by varying the concentration of first-strand cDNA and primers, using a new
batch of dNTPs. If results are unsatisfactory, consider redesigning the primers.
• Treat the products of the first-strand reaction with 2 units of RNase H (see Box 2) for 20 min at
37˚C to remove the template RNA from the single-stranded DNA product.

Problem (Step 12): The size of the amplified product is larger or smaller than expected.
Solution: It is likely that there is contamination with genomic DNA. Perform two-step RT-PCR in the
presence and absence of reverse transcriptase. If necessary, treat the RNA preparation with DNase
I. If necessary, redesign the primers so that they anneal to genomic DNA sequences in different
exons; this greatly reduces the possibility of efficient amplification of contaminating genomic DNA.

DISCUSSION

Primers for First-Strand Synthesis

Depending on the purpose of the experiment, the primer for first-strand cDNA synthesis can be
specifically designed to hybridize to a particular target gene or, when oligo(dT) or random primers are
used, can bind generally to all mRNAs.
• Oligo(dT), which binds to the endogenous poly(A) tails of mammalian mRNAs, can be used as a
universal primer for conventional first-strand cDNA synthesis. Subsequent amplification by PCR
generally uses one or more internal, gene-specific primers and oligo(dT) to generate copies of the
3′ -terminal sequences of a particular mRNA (3′ -RACE; see Protocol: Rapid Amplification of
Sequences from the 3′ Ends of mRNAs: 3′ -RACE [Green and Sambrook 2019]).
• cDNA synthesis can be primed by a synthetic antisense oligonucleotide that hybridizes specifically
to a chosen region in a particular target RNA or family of mRNAs. Amplification of the desired

Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190 411

 Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by
Cold Spring Harbor Laboratory Press

M.R. Green and J. Sambrook

portion of cDNA can be achieved in PCRs primed, for example, by sense and antisense oligonu-
cleotide primers corresponding to specific sequences in particular cDNAs. For maximum specif-
icity, the antisense primer should be located upstream of the oligonucleotide used to prime
cDNA synthesis.
Wherever possible, oligonucleotides that bind to sequences located in different exons of the
target RNA should be used as sense and antisense primers for amplification of the cDNA product.
In this way, amplification products derived from cDNA and contaminating genomic DNA can be
easily distinguished. However, transcripts of intronless genes cannot be differentiated unambig-
uously from contaminating genomic sequences. In these circumstances, treating the RNA prep-
aration with RNase-free DNase can be helpful (Grillo and Margolis 1990).
• Random hexanucleotides, which are capable of priming cDNA synthesis at many points along RNA
templates, generate fragmentary copies of the entire population of RNA molecules. They are useful
when the target RNA is extremely long or contains so much secondary structure that cDNA
synthesis cannot be efficiently primed by oligo(dT) or synthetic oligonucleotide(s) (Lee and
Caskey 1990). Because all mRNAs in the population can serve as templates, gene-specific ampli-
fication is achieved by using sense and antisense primers targeted to sequences in the cDNA.
Different methods of priming yield cDNA with different efficiencies (Lekanne Deprez et al. 2002;
Ståhlberg et al. 2004a,b). In most cases, investigators aim to generate a first strand of cDNA that is as
long as possible and contains a high proportion of molecules complementary to the target RNA.
Synthetic gene-specific oligonucleotides that bind to the 3′ -untranslated regions of the target mRNA
are therefore the primers of choice. Oligo(dT) is the next best option, and random hexamers, which
have no specificity and generate both long and short molecules of cDNA, are used when other
methods of priming disappoint.

Controls
Positive and negative controls should always be included when setting up RT-PCRs. Negative controls
are generated by omitting essential components of the RT stage of the RT-PCR. Positive controls,
however, are much trickier because they require standards that can be transcribed into cDNA and
amplified in parallel with the authentic target sequence. Ideally, these standards should consist of
known amounts of a synthetic RNA transcribed in vitro from a cloned, mutated segment of the target
DNA. The perfect standard would differ in sequence from the target RNA by only one or two bases
required to create one or more novel restriction site(s) in the amplified cDNA products. Both the
target RNA and the control should be able to be amplified with the same pair of primers (Becker-
André and Hahlbrock 1989; Wang et al. 1989; Gilliland et al. 1990; Siebert and Larrick 1992). After
amplification, the two PCR products are distinguished by digestion with restriction enzyme(s) and
agarose gel electrophoresis (for review, see Becker-André 1993).
Generating a perfect standard requires detailed planning and considerable labor. Most investiga-
tors therefore settle for internal standards that fall short of perfection: for example, a cloned segment
of cDNA that can be added in varying quantities to the RNA template, or, less desirably, a completely
unrelated, endogenous RNA species that can be amplified only with separate primers. Such imperfect
controls might be malodorous to purists, but they are certainly preferable to no controls at all. At least
they provide a means to measure the overall efficiency of reverse transcription and amplification and
provide some indication of the quality of the RNA preparation.

Reverse Transcriptases Used in Two-Step RT-PCR

The discovery of DNA-dependent RNA polymerases in 1970 solved the long-standing puzzle of how
the RNA genomes of retroviruses could be copied into DNA in infected and/or transformed cells
(Baltimore 1970; Temin and Mizutani 1970). The name “reverse transcriptase” was coined as a joke by

412 Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190

 Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by
Cold Spring Harbor Laboratory Press

Two-Step RT-PCR

John Tooze, who first used it in an anonymous “News and Views” article in Nature. Much to the
annoyance of purists and the delight of Tooze, the name stuck and quickly became iconic.
Several types of RNA-dependent DNA polymerases are now used in vitro to catalyze the synthesis
of DNA complementary to an RNA template:
• Mesophilic enzymes encoded by avian myeloblastosis virus (AMV) and the Moloney strain of murine
leukemia virus (Mo-MLV) both require an RNA or DNA template with an RNA or DNA primer
bearing a 3′ -hydroxyl group. Lacking an editing 3′ → 5′ exonuclease activity, both enzymes are
prone to error. The Km values for their dNTP substrates are very high (in the millimolar
range). To ensure complete transcription of template RNAs, it is essential to include a high
concentration of dNTPs in the reaction.
The enzyme encoded by AMV has a powerful RNase H activity that can digest the RNA moiety
of RNA–DNA hybrids and can cleave the template near the 3′ terminus of the growing DNA strand
if reverse transcriptase pauses during synthesis (Kotewicz et al. 1988). Thus, the high level of
RNase H activity associated with the avian reverse transcriptase tends to suppress the yield of
cDNA and restrict its length. (For further details, see Box 2.)
Mesophilic reverse transcriptases used to catalyze first-strand cDNA synthesis are easily blocked
by secondary structure in the RNA template. In addition, the RNase H activity of the enzymes is
thought somehow to interfere with the synthesis of full-length cDNA. Mo-MLV reverse transcrip-
tase is far better suited for RT-PCR than the avian enzyme because its RNase H activity is com-
paratively weak (Gerard et al. 1997). However, the Mo-MLV enzyme reaches maximum activity at
a lower temperature (37˚C) than the avian enzyme (42˚C), which could be a slight disadvantage if
the RNA template has a high degree of secondary structure. Variants of Mo-MLV reverse tran-
scriptase that lack RNase H activity are sold commercially (e.g., SuperScript, StrataScript). The
modified reverse transcriptases transcribe a greater proportion of the template molecules and
synthesize longer cDNA molecules than the wild-type enzyme (Gerard et al. 1988, 1997; Kotewicz
et al. 1988; Telesnitsky and Goff 1993). In addition, they are capable of cDNA synthesis at high
temperatures, which is an advantage when the template RNA is rucked into secondary structures.
Modified forms of reverse transcriptase with increased thermal stability (up to 60˚C) are com-
mercially available (e.g., SuperScript III, ThermoScript, AffinityScript) (Arezi and Hogrefe 2009).
These enzymes generate longer cDNAs than mesophilic reverse transcriptases and are now used
extensively in both one-step (see below) and two-step RT-PCR.
• Thermostable rTth DNA polymerase is encoded by the thermophilic eubacterium Thermus ther-
mophilus and shows reverse transcriptase activity in the presence of Mn2+ (Myers and Gelfand
1991). The chief advantage of using rTth polymerase in RT-PCR is that both phases of the reaction
(reverse transcription and amplification) are performed in the same reaction tube (i.e., one-step
RT-PCR) (Myers et al. 1994). However, Tth DNA polymerase cannot be used with oligo(dT) or
random hexamers as primers because the hybrids are unstable at temperatures at which reverse
transcription is performed (between 60˚C and 70˚C). Several manufacturers sell improved ver-
sions of thermostable DNA polymerases that are capable of efficiently copying RNA into cDNA.
These enzymes are used chiefly in one-step RT-PCR (see below).
For further information about the efficiencies of different reverse transcriptases in RT-PCR, see
Ståhlberg et al. (2004b).

One-Step RT-PCR
In one-step RT-PCR, all of the reagents and enzymes for the reverse-transcription and amplification
steps are assembled in the same tube. The two reactions are then performed sequentially without
further addition of reagents. In contrast, in the more traditional two-step method, cDNA synthesized
in a separate reaction is used as template in a conventional PCR. The first-strand DNA generated in the
reverse-transcription reaction can therefore be used in a number of different amplification reactions,

Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190 413

 Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by
Cold Spring Harbor Laboratory Press

M.R. Green and J. Sambrook

which can be an advantage. However, two-step PCR also has some deficiencies. For example, the
mesophilic reverse transcriptases conventionally used to catalyze first-strand cDNA synthesis are
easily blocked by secondary structure in the RNA template. In addition, the reverse transcription
and amplifications must be optimized separately. These problems can be avoided in one-step RT-PCR
because the coupled transcription and amplification reactions can both be catalyzed by thermostable
enzymes, which are highly active at elevated temperature. Consequently, priming of first-strand cDNA
synthesis is more specific, and the coupled transcription reaction is quicker and simpler to set up and
to optimize. Nevertheless, because the amplification reaction is performed with gene-specific primers,
separate one-step RT-PCRs must be set up for each pair of primers.
The reverse transcriptases and thermostable DNA polymerases used in one-step RT-PCR can be
purchased separately from many manufacturers, and the reaction mixtures can then be assembled
from scratch and optimized for use with particular template RNAs. However, one-step RT-PCR is one
of the few procedures for which it makes more sense to buy a kit. In most laboratories, the method will
be used only a few times for a specific project. Purchasing a 50-reaction kit will certainly be the less
expensive option in time, and probably in dollars.
Kits for one-step RT-PCR are available from several manufacturers. The enzymes supplied in these
kits are different, and there are no published reports of head-to-head comparison of their efficiency
and ease of use with a range of templates. As usual, the best way forward in this situation is to ask the
company representatives for the names of people who have bought kits and then to find out
their opinions.
The Ung/dUT system can be used to control contamination of PCRs with DNA, a particularly
important issue in diagnostic and forensic laboratories (see Box 3). However, the Ung/dUT system
should not be used with one-step RT-PCR unless rTth DNA polymerase is used both for reverse
transcription and DNA amplification (e.g., in the TaqMan Real-Time EZ RT-PCR Kit).

BOX 3. THE dut AND ung GENES OF E. coli

dut
The deoxyuridine triphosphatase (dut) gene of E. coli encodes dUTPase, a zinc-containing tetrameric pyro-
phosphatase (Mr = 64,000) (Shlomai and Kornberg 1978; Lundberg et al. 1983). In dut mutants, the cell’s
ability to convert dUTP to dUMP is impaired. This decrease results in a 25- to 30-fold increase in the
intracellular pool of dUTP. The elevated ratio of dUTP:dTTP leads to an increase in the frequency of
misincorporation of dUTP at a random subset of the sites normally occupied by thymine.
The dut − mutants of E. coli used in the Kunkel method all contain a trace of dUTPase activity (Konrad and
Lehman 1975; Hochhauser and Weiss 1978), perhaps because a complete lack of the enzyme is lethal (el-
Hajj et al. 1988). Viable dut − mutants show increased rates of recombination and spontaneous mutation,
which result from the temporary strand breaks and gaps in DNA caused by excision of the uracil residues by
uracil-DNA glycosylase, encoded by the ung gene.

ung
Uracil residues incorporated into DNA in E. coli are normally removed by the small monomeric enzyme
uracil-DNA glycosylase (deduced Mr = 25,664), which cleaves the N-glycosydic bond between the base
and the deoxyribose phosphate backbone, generating an apyrimidinic site (Lindahl 1974; for review, see
Lindahl 1982; Tomilin and Aprelikova 1989). The phosphodiester backbone at these sites is cleaved by
endonucleases (exonuclease III or endonuclease IV) (for review, see Lloyd and Linn 1993) and then repaired
by DNA polymerase I and ligase. Cells deficient in uracil-DNA glycosylase cannot cleave the N-glycosydic
bond and display increased rates of spontaneous mutation that are biased toward G:C → A:T transitions
(Duncan and Miller 1980; Duncan and Weiss 1982).
The ung gene of E. coli has been cloned (Duncan and Chambers 1984), as have highly conserved ung
genes from many other organisms including yeast, humans, and herpesviruses (see Olsen et al. 1991; Lloyd
and Linn 1993). These genes probably evolved to remove mutagenic deaminated cytosine residues from
DNA.

414 Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190

 Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by
Cold Spring Harbor Laboratory Press

Two-Step RT-PCR

Avoiding Trouble
• In preliminary experiments, perform the reverse-transcription step at a range of different tem-
peratures. Choose the minimum temperature that generates an acceptable yield of the desired
product. Higher temperatures should be avoided to suppress metal ion-catalyzed hydrolysis of
RNA (Brown 1974).
• Once an appropriate temperature has been selected, set up another series of pilot experiments, this
time varying the concentration of Mg2+.
• If synthesis of nonspecific PCR products is a problem, add a cosolvent such as betaine, dimethyl
sulfoxide, or a commercial product such as Perfect Match polymerase enhancer (Agilent).
• To minimize the chances of amplifying segments of DNA that might contaminate the RNA
preparation, consider using the Ung/dUT system (Longo et al. 1990; Park and Lee 2003) (see
Box 3). However, bear in mind that some DNA polymerases that display 3′ -exonuclease activity
(e.g., Pfu, Deep Vent) are inhibited by dUTP (Barnes 1994; Cheng et al. 1994; Lasken et al. 1996).
• In general, make sure that all the appropriate negative controls (no RNA, no reverse transcriptase, no
thermostable DNA polymerase) and positive controls (positive RNA control, if available) were
included.

ACKNOWLEDGMENTS

Tables 1 and 2 were provided by Holly Hogrefe and Michael Borns, Agilent Technologies.

REFERENCES

Arezi B, Hogrefe H. 2009. Novel mutations in Moloney murine leukemia Gerard GF, Schmidt BJ, Kotewicz ML, Campbell JH. 1988. cDNA synthesis
virus reverse transcriptase increase thermostability through tighter by Moloney murine leukemia virus RNase H-minus reverse transcrip-
binding to template-primer. Nucleic Acids Res 37: 473–481. tase possessing full DNA polymerase activity. Focus 14: 91–93.
Baltimore D. 1970. RNA-dependent DNA polymerase in virions of RNA Gerard GF, Fox DK, Nathan M, D’Alessio JM. 1997. Reverse transcriptase.
tumour viruses. Nature 226: 1209–1211. The use of cloned Moloney murine leukemia virus reverse transcriptase
Barnes WM. 1994. PCR amplification of up to 35-kb DNA with high fidelity to synthesize DNA from RNA. Mol Biotechnol 8: 61–77.
and high yield from λ bacteriophage templates. Proc Natl Acad Sci 91: Gilliland G, Perrin S, Blanchard K, Bunn HF. 1990. Analysis of cytokine
2216–2220. mRNA and DNA: Detection and quantitation by competitive polymer-
Becker-André M. 1993. Absolute levels of mRNA by polymerase chain ase chain reaction. Proc Natl Acad Sci 87: 2725–2729.
reaction–aided transcript titration assay. Methods Enzymol 218: Green MR, Sambrook J. 2019. Rapid amplification of sequences from the 3′
420–445. ends of mRNAs: 3′ -RACE. Cold Spring Harb Protoc doi: 10.1101/pdb
Becker-André M, Hahlbrock K. 1989. Absolute mRNA quantification using .prot095216.
the polymerase chain reaction (PCR). A novel approach by a PCR aided Grillo M, Margolis FL. 1990. Use of reverse transcriptase polymerase chain
transcript titration assay (PATTY). Nucleic Acids Res 17: 9437–9446. reaction to monitor expression of intronless genes. BioTechniques 9:
Brown DM. 1974. Chemical reactions of polynucleotides and nucleic acids. 262–268.
In Basic principles in nucleic acid chemistry (ed. Ts’o POP), pp. 43–44. Hausen P, Stein H. 1970. Ribonuclease H. An enzyme degrading the RNA
Academic, New York. moiety of DNA-RNA hybrids. Eur J Biochem 14: 278–283.
Cheng S, Fockler C, Barnes WM, Higuchi R. 1994. Effective amplification of Hochhauser SJ, Weiss B. 1978. Escherichia coli mutants deficient in deoxy-
long targets from cloned inserts and human genomic DNA. Proc Natl uridine triphosphatase. J Bacteriol 134: 157–166.
Acad Sci 91: 5695–5699. Hogrefe HH, Hansen CJ, Scott BR, Nielson KB. 2002. Archaeal dUTPase
Chumakov KM. 1994. Reverse transcriptase can inhibit PCR and stimulate enhances PCR amplifications with archaeal DNA polymerases by pre-
primer–dimer formation. PCR Methods Appl 4: 62–64. venting dUTP incorporation. Proc Natl Acad Sci 99: 596–601.
Crooke ST. 1998. Molecular mechanisms of antisense drugs: RNase H. Hostomsky Z, Hostomska Z, Matthews DA. 1993. Ribonucleases H. In
Antisense Nucleic Acid Drug Dev 8: 133–134. Nucleases, 2nd ed. (ed. Linn SM, et al.), pp. 341–376. Cold Spring
Crouch RJ. 1990. Ribonuclease H: From discovery to 3D structure. New Biol Harbor Laboratory Press, Cold Spring Harbor, NY.
2: 771–777. Jung YH, Lee Y. 1995. RNases in ColE1 DNA metabolism. Mol Biol Rep 22:
Duncan BK, Chambers JA. 1984. The cloning and overproduction of Esche- 195–200.
richia coli uracil-DNA glycosylase. Gene 28: 211–219. Kanaya S, Ikehara M. 1995. Functions and structures of ribonuclease H
Duncan BK, Miller JH. 1980. Mutagenic deamination of cytosine residues in enzymes. Subcell Biochem 24: 377–422.
DNA. Nature 287: 560–561. Konrad EB, Lehman IR. 1975. Novel mutants of Escherichia coli that accu-
Duncan BK, Weiss B. 1982. Specific mutator effects of ung (uracil-DNA mulate very small DNA replicative intermediates. Proc Natl Acad Sci 72:
glycosylase) mutations in Escherichia coli. J Bacteriol 151: 750–755. 2150–2154.
el-Hajj HH, Zhang H, Weiss B. 1988. Lethality of a dut (deoxyuridine Kotewicz ML, Sampson CM, D’Alessio JM, Gerard GF. 1988. Isolation of a
triphosphatase) mutation in Escherichia coli. J Bacteriol 170: cloned Moloney murine leukemia virus reverse transcriptase lacking
1069–1075. ribonuclease H activity. Nucleic Acids Res 16: 265–277.

Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190 415

 Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by
Cold Spring Harbor Laboratory Press

M.R. Green and J. Sambrook

Lasken RS, Schuster DM, Rashtchian A. 1996. Archaebacterial DNA poly- Park K, Lee J. 2003. Dealing with carryover contamination in PCR: An
merases tightly bind uracil-containing DNA. J Biol Chem 271: enzymatic strategy. In PCR primer: A laboratory manual (ed. Dieffen-
17692–17696. bach CW, Dveksler GS), pp. 15–20. Cold Spring Harbor Laboratory
Lee CC, Caskey CT. 1990. cDNA cloning using degenerate primers. In PCR Press, Cold Spring Harbor, NY.
protocols: A guide to methods and applications (ed. Innis MA, et al.), pp. Rice P, Craigie R, Davies DR. 1996. Retroviral integrases and their cousins.
46–53. Academic, San Diego. Curr Opin Struct Biol 6: 76–83.
Lekanne Deprez RH, Fijnvandraat AC, Ruijter JM, Moorman AFM. 2002. Rio DC, Ares M Jr, Hannon GJ, Nilsen TW. 2010. Removal of DNA from
Sensitivity and accuracy of quantitative real-time polymerase chain RNA. Cold Spring Harb Protoc doi: 101101/pdb.prot5443.
reaction using SYBR green I depends on cDNA synthesis conditions. Sellner LN, Coelen RJ, Mackenzie JS. 1992. Reverse transcriptase inhibits
Anal Biochem 307: 63–69. Taq polymerase activity. Nucleic Acids Res 20: 1487–1490.
Liedtke W, Battistini L, Brosnan CF, Raine CS. 1994. A comparison of Shlomai J, Kornberg A. 1978. Deoxyuridine triphosphatase of Escherichia
methods for RNA extraction from lymphocytes for RT-PCR. PCR coli. Purification, properties, and use as a reagent to reduce uracil in-
Methods Appl 4: 185–187. corporation into DNA. J Biol Chem 253: 3305–3312.
Lindahl T. 1974. An N-glycosidase from Escherichia coli that releases free Siebert PD, Larrick JW. 1992. Competitive PCR. Nature 359: 557–558.
uracil from DNA containing deaminated cytosine residues. Proc Natl Ståhlberg A, Håkansson J, Xian X, Semb H, Kubista M. 2004a. Properties of
Acad Sci 71: 3649–3653. the reverse transcription reaction in mRNA quantification. Clin Chem
Lindahl T. 1982. DNA repair enzymes. Annu Rev Biochem 51: 61–87. 50: 509–515.
Lloyd RS, Linn S. 1993. Nucleases involved in DNA repair. In Nucleases, 2nd Ståhlberg A, Kubista M, Pfaffl M. 2004b. Comparison of reverse transcrip-
ed. (ed. Linn SM, et al.), pp. 263–316. Cold Spring Harbor Laboratory tases in gene expression analysis. Clin Chem 50: 1678–1680.
Press, Cold Spring Harbor, NY. Stein H, Hausen P. 1969. Enzyme from calf thymus degrading the RNA
Longo MC, Berninger MS, Hartley JL. 1990. Use of uracil DNA glycosylase to moiety of DNA–RNA hybrids: Effect on DNA-dependent RNA poly-
control carry-over contamination in polymerase chain reactions. Gene merase. Science 166: 393–395.
93: 125–128. Telesnitsky A, Goff SP. 1993. RNase H domain mutations affect the
Lundberg LG, Thoresson HO, Karlström OH, Nyman PO. 1983. Nucleotide interaction between Moloney murine leukemia virus reverse tran-
sequence of the structural gene for dUTPase of Escherichia coli K-12. scriptase and its primer-template. Proc Natl Acad Sci 90:
EMBO J 2: 967–971. 1276–1280.
Myers TW, Gelfand DH. 1991. Reverse transcription and DNA amplification Temin HM, Mizutani S. 1970. RNA-dependent DNA polymerase in virions
by a Thermus thermophilus DNA polymerase. Biochemistry 30: 7661–7666. of Rous sarcoma virus. Nature 226: 1211–1213.
Myers TW, Sigua CL, Gelfand DH. 1994. High temperature reverse tran- Tomilin NV, Aprelikova ON. 1989. Uracil-DNA glycosylases and DNA
scriptase and PCR with a Thermus thermophilus DNA polymerase. uracil repair. Int Rev Cytol 114: 125–179.
Nucleic Acids Symp Ser 30: 87. Wang AM, Doyle MV, Mark DF. 1989. Quantitation of mRNA by the
Olsen LC, Aasland R, Krokan HE, Helland DE. 1991. Human uracil-DNA polymerase chain reaction. Proc Natl Acad Sci 86: 9717–9721.
glycosylase complements E. coli ung mutants. Nucleic Acids Res 19: Wintersberger U. 1990. Ribonucleases H of retroviral and cellular origin.
4473–4478. Pharmacol Ther 48: 259–280.

416 Cite this protocol as Cold Spring Harb Protoc; doi:10.1101/pdb.prot095190

 Downloaded from http://cshprotocols.cshlp.org/ at NYU MED CTR LIBRARY on July 17, 2019 - Published by
Cold Spring Harbor Laboratory Press

Amplification of cDNA Generated by Reverse Transcription of mRNA: Two-Step

Reverse Transcription-Polymerase Chain Reaction (RT-PCR)
Michael R. Green and Joseph Sambrook

Cold Spring Harb Protoc; doi: 10.1101/pdb.prot095190

Email Alerting Receive free email alerts when new articles cite this article - click here.
Service

Subject Browse articles on similar topics from Cold Spring Harbor Protocols.
Categories
Amplification of DNA by PCR (79 articles)
cDNA (60 articles)
Molecular Biology, general (1192 articles)
mRNA (122 articles)
Polymerase Chain Reaction (PCR) (119 articles)
Polymerase Chain Reaction (PCR), general (173 articles)
RNA (287 articles)
RNA, general (242 articles)
RT-PCR (48 articles)

To subscribe to Cold Spring Harbor Protocols go to:

http://cshprotocols.cshlp.org/subscriptions

© 2019 Cold Spring Harbor Laboratory Press