ABSTRACT
Arbitrarily primed PCR fingerprinting of RNA and differential display resolved
on an acrylamide gel has been extensively used to detect differentially expressed RNAs. However, after a differentially amplified product is detected the next
steps are labor-intensive: a small portion of the fingerprinting gel that contains the
differentially amplified product is cut out, reamplified and the correct product is determined, typically by cloning and sequencing what
is often a mixture of products of similar size. Here we use a native acrylamide gel to separate DNAs in the reamplified mixture based on single-stranded conformation polymorphisms. Reamplifications are performed for the region carrying the differentially amplified product and a corresponding region from an adjacent lane where the product is less prominent or not visible. Denaturation of the reamplified DNA followed by side-by-side comparison on an SSCP gel allows the classification of
reamplified material into (i) those that can be directly cloned because the differentially
amplified product is relatively pure, (ii) those that need to be reamplified from the SSCP gel before cloning
and (iii) those that are too complex for further study. This screen should save
considerable effort now wasted on directly cloning unsuitable products from RNA fingerprinting experiments. An example is presented of cloning
a gene differentially expressed in
Trypanosoma brucei
life cycle.
RNA arbitrarily primed PCR fingerprinting and differential display (
1
,
2
) are confounded if the product is not really differentially expressed. This
particular problem is often due to variations in the fingerprints caused by slight differences in the quality or concentration of nucleic acid between samples, as originally demonstrated for
DNA (
3
). This problem is largely controlled for by comparing fingerprints generated from two or more RNA concentrations for each sample, side by side. Those products that occur in only
one concentration are eliminated from consideration. However even when intra-sample variation is controlled for, a labor-intensive, rate limiting step that remains is the isolation of differentially amplified products and their identification and confirmation as fragments from a differentially regulated
RNA.
Methods to characterize differentially amplified products generally start by cutting out the product from the fingerprinting gel and
reamplifying the product by PCR using the primers originally employed in the
fingerprinting. The main stumbling block at this stage is the fact that the
arbitrary fingerprint is always a mixture of products and the reamplification
from a tiny portion of the gel usually generates multiple products, of almost
identical size to the product of interest. Thus, in subsequent steps the
reamplified material contains a mixture of desirable and undesirable DNA
products.
Various strategies have been applied to distinguish the correct product from
other products of similar size that are co-amplified. One strategy is to clone the reamplified cDNA mixture and then
to sequence a number of independent clones, as originally shown for genomic DNA
fingerprints (
4
,
5
). More than one sequence and its complement are frequently observed, demanding
considerable work to find a statistically more abundant clone which is most
likely to be the differentially amplified cDNA of interest. Then, the most
frequent clone can be hybridized to a Southern blot of the original RAP-PCR gel to prove that the correct amplified product was cloned (
6
,
7
). The clone is generally confirmed as differentially regulated by hybridization
to a Northern blot (
8
) or by RT-PCR (
9
). Another strategy avoids cloning until differential expression is confirmed.
The amplified mixture from the original RAP-PCR gel is used directly for a Northern blot (
10
). If a discrete differentially hybridizing product is seen then this can be
reamplified from the Northern blot using the original primers. This strategy
has the virtue that the Northern blot essentially purifies the correct
differentially expressed product. However, this method is only successful if
the reamplified mixture is free of dispersed repeats that would obscure the
data. Also, if a contaminating product in the mixture hybridizes to a mRNA that
is more abundant than that targeted by the PCR product of interest, then the
Northern blot will appear to show no differential expression. Furthermore, this
strategy uses a lot of RNA, something that is not practical for many RNA
sources.
The strategies described above can all be enhanced by the use of SSCP gels as
described below.
Total RNA was prepared from
Trypanosoma brucei brucei
GUTat 3.1 at different stages of the life cycle. The procyclic stage (Pc), was
cultured
in vitro
, while the slender (Sl) and stumpy (St) bloodstream forms were grown in mice.
Reverse transcription was performed on 500, 250 and 125 ng total RNA. The
fingerprints were obtained as previously described (
1
,
6
-
8
,
11
), using either two different 10mer oligonucleotide primers of arbitrary
sequences, or a combination of a 10mer arbitrary primer and a 11mer derived
from the 5' mini-exon sequence of the trypanosomes mRNA (
12
,
13
).
The region that contained the product of interest was cut out of the gel, as was
the corresponding region in an adjacent lane where the product was apparently
not present, or present at a significantly lower level of abundance. Products
were eluted in 50-100 [mu]l TE at 65oC for 2 to 3 h. The eluted solution was diluted by 20-fold in water, and 2 [mu]l used in a 20 [mu]l volume PCR reaction mixture containing 10 mM
Tris-HCl pH 8.3, 10 mM KCl, 4 mM MgCl
2
, 0.2 mM of each dNTP, 1 [mu]Ci [[alpha]-
32
P]dCTP and 2 U Ampli
Taq
polymerase Stoffel fragment (Perkin-Elmer-Cetus, Norwalk, CT) and 0.5 [mu]M of each oligonucleotide primer (the same two primers used to
generate the RNA fingerprint). Thermocycling was performed with a GeneAmp PCR
System 9600 thermocycler (Perkin-Elmer-Cetus), using 30 cycles of 94oC for 30 s, 35oC for 30 s and 72oC for 1 min. The PCR products (4 [mu]l samples) were mixed with 18 [mu]l formamide dye solution, heated to 92oC for 3 min and 1.2 [mu]l was loaded onto MDE gel (HydroLink
®
MDEtm gel, J.T. Baker Inc., NJ), in 0.6* TBE buffer with 5% glycerol. Electrophoresis was performed
overnight at 8 W for ~16 h. The gel was dried under vacuum and placed on a Kodak BioMax X-ray film for 20 h. Reducing the number of PCR cycles to 20 or 25 may
better preserve differences between samples. However, in that case an
intensifying screen may be needed in order to visualize the products on the
SSCP gel.
Polyacrylamide gels were transferred by capillary action, overnight, onto nylon membranes (Hybond N+, Amersham, Buckinghamshire, UK) using standard conditions. Probes were labeled and hybridized to the membranes using the non-radioactive ECLtm direct nucleic acid labeling and detection system (Amersham)
according to the manufacturer's instructions. Fragments were separated from
dNTPs and primers on a low-melting-point agarose gel, and were cloned in pCR-Scripttm SK(+) plasmid vector (Stratagene, La Jolla, CA) and Epicurian Coli
®
XL1-Blue MRF' competent cells using standard conditions. Single-stranded phagemids were sequenced using the Sequenase DNA sequencing kit (US Biochemical, Amersham) and [[alpha]-
35
S]dATP.
The reverse transcription was performed on 250 ng RNA. RNA, 5 [mu]l, was mixed with the same volume RT mixture for a 10 [mu]l final reaction containing 50 mM Tris-HCl pH 8.3, 50 mM KCl, 4 mM MgCl
2,
10 mM DTT, 0.2 mM of each dNTP, 13 U MuLV-reverse transcriptase, 2 [mu]M primer (three different anchored-dT primers: (T)
12
-G, (T)
12
-A and (T)
12
-C for Fig.
2
D, lanes 1, 2 and 3, respectively). The reverse transcription was performed at
37oC for 1 h, the enzyme was inactivated by heating the samples at 94oC for 2 min and the cDNA obtained was diluted 4-fold in water. The PCR was performed using the primers 5'-AATGAAAGTTACGATAGCGG and 5'-AAAGACAACGGAGATGGCA, chosen from the
DNA sequence of the clone. Diluted cDNA, 5 [mu]l, was mixed with the same volume of PCR mixture for a 10 [mu]l final reaction containing 10 mM Tris-HCl pH 8.3, 10 mM KCl, 4 mM MgCl
2,
0.2 mM of each dNTP, 1 [mu]M of each primer and 2 U Ampli
Taq
polymerase Stoffel fragment (Perkin-Elmer-Cetus, Norwalk, CT). Thermocycling was performed with a GeneAmp PCR
System 9600 thermocycler (Perkin-Elmer-Cetus), using 35 cycles of 94oC for 30 s, 55oC for 30 s and 72oC for 1 min. Amplification products were run on a 2%
agarose gel.
In an effort to improve the ease and effectiveness of the characterization and
confirmation of PCR-reamplified products we have investigated single-stranded conformation polymorphism (SSCP) gels as a method of
purifying the cDNA product of interest away from other products of different
sequence but of similar size.
Single-stranded DNAs fold to form stable and metastable secondary structures that
affect their mobility in a native acrylamide gel. Hayashi exploited this phenomenon to distinguish between point mutations
in molecules that were otherwise identical (
14
). RAP-PCR fingerprinting product mixtures are much easier to distinguish than
point mutations because, while the contaminating products are generally about
the same length as the product of interest, they are of completely different
sequence. Thus, they can be expected to have distinct mobilities on an SSCP
gel.
The SSCP procedure was applied to products isolated from an experiment involving different stages in the development of
Trypanosoma brucei
, the unicellular eukaryote responsible for sleeping sickness in Africa. The
RNAs investigated were prepared from the procyclic form of the parasite that
divides in the midgut of the tsetse fly vector, the slender form which divides
rapidly in the blood of the mammal host and the stumpy form that has stopped
dividing and is primed for recycling into the biting fly.
In each case, the region of the RAP-PCR gel that contained the differentially amplified cDNA product of
interest were reamplified and simultaneously radiolabeled, as was the
corresponding region in an adjacent lane where the product of interest was
apparently not present, or was present at a significantly lower level of
abundance. These are referred to as the `experimental' and the `control'.
Each PCR product can be expected to produce two SSCP bands, one for each of the
two cDNA strands. In Hydrolink
®
MDE gels it is rare that one strand will migrate as two different confomers.
Occasionally these strands will have the same mobility and produce only one
band. Our results showed that it was almost impossible to obtain only one or
two bands on the SSCP gel, even if the same samples, run an agarose gel, gave
only one band. This underlines the utility of this method for purification.
There were four classes of patterns that were seen in the SSCP gels.
The PCR product exhibits one to three bands of very low intensity, in addition
to two bands of very high intensity. These two strong bands were present only
in the experimental lane, in the case of a presence versus absence of the eluted band in the RAP-PCR gel (Fig.
1
A), or present in both experimental and control lanes, in the case of a
difference of intensity of the eluted band (Fig.
1
B). In this last case, the two bands visible in the control are from a low level
of the differentially amplified product that was able to amplify because it was
purified away from competing products in the same fingerprinting lane. In both
these cases the original amplification can be used for cloning or for Southern
or Northern blots.
One or two prominent products in the experimental lane are shared with the
control and one or two others are not. An example is shown in Figure
1
C. In this case the product(s) unique to the experimental lane can be further
cut and reamplified from the SSCP gel (as shown later). The pattern then
obtained exhibits only one or two bands corresponding to the two single
strands. This purified product can now be cloned or hybridized. We have
performed this enrichment step for seven products that have been tested and
confirmed as the correct differentially amplified products. An example will be
presented later in this paper.
Both the experimental and the control lanes exhibit the same profile of a few
bands with no strong differences in intensity. To determine which band(s) is
the product of interest, a larger strip of gel, including the band
differentially amplified and one or two conserved bands, can be cut from the
original fingerprint and reamplified as well as the corresponding control. The
difference between the experimental and the control is often preserved because
each reaction now has other products to amplify in addition to the
differentially amplified product of interest. The DNA mass in the control lane
is distributed to these products rather than merely erasing differences in the
product of interest. Also, limiting the number of cycles to 20, or less, will
better preserve the stoichiometry of products between samples. Limited
experiments indicate that this may be the case (data not shown). If this
strategy is used then it is necessary to cut and reamplify the product of
interest from the SSCP gel, as in Class 2.
The experimental lane may give a complex pattern or a smear that may or may not
differ from the control. This category represents failures and should be
rejected. However, even in these cases SSCP has served well by giving the
investigator an easy assay to identify those mixtures that would be very
difficult to pursue further. Without this assay much time would be wasted on
this class of products.
Class 1 is observed most often when the background is very low (less contaminant
cut out of the RAP-PCR gel with the band). Classes 2 and 3 are most often observed when the
background is high on the RAP-PCR gel.
In total, four different pairs of primers were used for cDNA fingerprinting.
These fingerprints detected ~200 cDNA products. Of these, 22 were clearly differentially amplified
between two of the three stages of
Trypanosoma
development. These 22 products were reamplified and screened using SSCP. Of
these, 13 gave potentially usable SSCP patterns, two from class 1, six from
class 2 and five from class 3. So far nine products have been cloned, sequenced
and confirmed on Southern blots. Four of them have also been confirmed for
differential expression using RT-PCR. An example will be presented below.
Note that when the reamplified material from the fingerprinting gel is resolved
by SSCP it is possible to assess which reamplifications are likely to yield a
clear majority of one fragment when cloned. Even when the SSCP indicates a
simple pattern of one or two bands (as in Fig.
1
A and B) this does not preclude the infrequent cloning of other products from
the mixture as well. Thus, sequencing of a few clones can be performed, though,
after a purification from the SSCP gel and a second reamplification, as shown
in Figure
1
C, the chance of sampling two different sequences is very low. A major clone can
be found by sequencing no more than four clones.
As an example of the characterization procedure, Figure
2
presents data for the different steps of isolation, purification and
confirmation of the differential expression of a gene. RNA fingerprinting of
the three different stages of
T.brucei
is shown in Figure
2
A. One differentially amplified fragment, that was present in the two
bloodstream forms only, was reamplified and resolved on a SSCP gel twice in a
row. Figure
2
B illustrates reamplifications of this product which belongs to class 2. The
cDNA fragment from this second step of reamplification was cloned into the pCR-Script vector, as described in Materials and Methods. The four sequenced
clones had the same sequence. One clone was used as a probe on a Southern blot
against a polyacrylamide fingerprint gel (Fig.
2
C), confirming the right fragment was cloned. This product was apparently absent
in the procyclic but present in the two bloodstream forms. Primers were derived
from the sequence of this clone and used in RT-PCR. Differential expression was so extreme that even after 35 cycles the
difference between the two different stages was preserved (Fig.
2
D). Extra PCR products in this reaction, which act as internal positive controls
for amplification, are presumably due to arbitrarily primed PCR because the
annealing temperature used was 5-10oC below the
T
m
of the primers. A search of the GenBank database indicated strong homology to
an expression site-associated gene, ESAG 1 (
15
,
16
), a member of a gene family associated to the expression sites of
T.brucei
variable surface genes (VSGs), which are known to be expressed in the
bloodstream stages of the parasite and not generally in the procyclic stage.
This product has been deposited in the GenBank database (accession number
U49237). A detailed description of the developmental profile of some of the
novel genes discovered in these experiments is in preparation, including their
distribution in other stages of development and under various treatments.
In summary, SSCP represents a rapid method to screen for contaminating products
in reamplifications from RNA fingerprints. By comparing reamplifications from
fingerprinting lanes where the desired product is of high abundance with the
equivalent position from lanes in which the desired product is absent or of low
abundance, it is possible to identify the desired product in the mixture. If
necessary, the enriched product of interest can be purified by reamplification
from the SSCP gel. This strategy greatly aids in selecting suitable
reamplification mixtures for the labor-intensive step of purifying and identifying the differentially amplified
product of interest.
In the future it should also be possible to use the SSCP method directly on DNA
eluted from the RAP-PCR fingerprinting gel without reamplification. Elution is sufficiently
efficient and DNA can be sufficiently labeled to allow detection after SSCP.
This direct method has the advantage that it will preserve stoichiometry from
the fingerprinting gel.
We thank Dr Charles Davis and Karen Arnold (Division of Infectious Diseases,
UCSD Medical Center, CA) for providing the
T.brucei
stock, and growing the bloodstream forms of the parasite. This work was
supported by the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (no. 910020) and NIH grants AI 34829, NS 33377, CA 68822 and AI3 2644. F.M-D. was supported by a Lavoisier Fellowship (Ministère Français des Affaires Etrangères).
+
Present address: Molecular and Bioscience Department, MS P7-56, Batelle-Pacific Northwest Laboratory, Richland, WA 99352, USA
REFERENCES
Return