2900 Nucleic Acids Research, Vol. 20, No. 11
1992
Oxford University Press
A random-POR method (rPCR) to construct whole cDNA library from low amounts of RNA Patrick Froussard Departement des R6trovirus, Institut Pasteur, 28 rue du Docteur Roux, 75724 Paris Cedex 15, France Submitted February 25, 1992 In contrast with random-primed RACE (1) and RNA-PCR (2) which can be done to amplify respectively specific 5'- or 3'-cDNA ends, random-PCR (rPCR) can be used to amplify the whole cDNA population, from 5' to 3' ends, derived from small amounts of RNA transcripts. This quick rPCR method allows construction of whole cDNA libraries from as little as one eukaryotic cell, 100 parasites or bacteria, or 105 particles of a 9 kb retrovirus, so lowering the threshold. Libraries constructed in this manner can then be screened with specific DNA probes or antibodies. Moreover, cDNA amplified by rPCR are powerful tools for the screening of substracted libraries. The rPCR was tested here on several MS2 phage RNA dilutions, either respectively 5 x 10-4, 5 x l0-5, 5x 10-6 and 5 x 10-7 Fg MS2 RNA (Serva), and on a control assay with no RNA. Using a 26 nucleotides primer containing a random hexamer at its 3' end (Universal primer-dN6; 5'-GCCGGAGCTCTGCAGAATTCNNNNNN-3'), the first-strand cDNA was prepared as follows. MS2 RNA was suspended in 6 pl of distilled water, heated to 650C for 5 mi, rapidly cooled on ice and reverse transcribed after addition of 0.5 Il (20 units) RNAsin, 1.25 ;d lOxreverse transcription buffer (500 mM Tris-HCI pH 8.3 at 43°C, 800 mM NaCl, 80 mM MgCl2, 50 mM DT), 1.25 Al each dNTP (10 mM), 1.5 jd Universal primer-dN6 (0.1 pg4l), 2 tl (16 units) AMV reverse transcriptase. Incubation was at 430C for 1 h. The reaction was then boiled for 2 min and rapidly cooled on ice. For second-strand cDNA synthesis, the following components were then added; 24.25 1l distilled water, 10 ILI 5 xKlenow buffer (kit multiprime/Phrmacia), 1.25 p1 dCTP (100 mM), 2 ,ll Klenow fragment (8 units). After 30 min incubation at 37°C, the sample was purified on a Chroma Spin400 column (Clontech) to eliminate the excess of Universal primer-dN6. Amplification of the randomly synthetised double-strad cDNA population was then performed on a 1 I1 aliquot in presence of the Universal primer essentially as described by Cetus (3). Briefly, amplification took place in 50 ILI reaction mixtures containing double-strand cDNA in 10 mM Tris-HCI pH 8.3, 50 mM KCI, 1.5 mM MgCl2, 0.01% gelatin, 500 F&M each dNTP, 1 isM of Universal primer and 1.5 unit Taq polymerase (Cetus). The samples were subjected to 40 cycles of amplification, 94°C-1 min, 55°C-I min, 72°C-3 min. Final amplification products analysed on agarose gel (Figure 1) showed DNA populations essentially distributed from 0.4 to 3 Kb with a maximum number of copies around 0.8 Kb; rPCR on total eukaryotic RNA gives routinely DNA fragments around 0.5 Kb. Care has to be taken with the Universal primer-dN6 concentration as it govems both the size of the synthesized cDNA
and the efficiency of the reactions. Using 1 pl from a 50 ul assay, as little as 10-6 sg RNA was efficiendy amplified with the rPCR method (lane d). However, less sting material can generate amplification of one or a few cDNA species that are not represnative of the initial RNA populato (lane f); this was supported here by the use of a pure MS2 RNA which is only
3.6 kb long. The rPCR is actually used in our laboraty in attempt to identify putative retrovirus sequences poorly represented in human culture cell supernatant. This technique should also be applicable for amplification of genomic DNA by replacing the reverse transcriptase with a Klenow fragment step.
ACKNOWLEDGEMENTS This work was supported by grants from the nstitut Pasteur and the Centre National de la Recherche Scientifique (CNRS). We thank Doctor Simon Wain-Hobson and Professo Luc Montagnier for their encouragement in this work as well as for laboratory facilities.
REFERENCES 1. Harvey,J. and Darlison,M.G. (1991) Nucleic Acids Res 14, 4002. 2. Fritz,J.D., Greaser,M.L. and Wolff,J.A. (1991) uclkic Acds Res. 13, 3747. 3. Saild,R.K., Gelfand,D.H., Stroffel,S., Scharf,S.L, Higuchi,R., Horn,G.T., Mullis,K.B. and Erlich,H.A. (1988) Science 239, 487-491.
a b c d e f
Kb 12.2 5.0
-
3.0
_
1.0
-
0.5
-
0.3
-
Flgure 1. Agarose gel electrophoretic analysis of rPCR prohdct encoding MS2 virus seqences. For Southern analysis, 10 #1 of the PCR products were sed in 1.0% agarose gel, transfered to nitod1lulowe aid hybridized eleCtro with3 P-labeled MS2 cDNA probes primed with a rndon hexamer. Lanes c to f correspond respectively to 10-5, 10-6, 10-7 and i-3 g of MS2 RNA. Lane a is an amplification of water and lane b is a rPCk product done with no RNA. The relative positions of size markers are shown in kb.
1992
Oxford University Press
A random-POR method (rPCR) to construct whole cDNA library from low amounts of RNA Patrick Froussard Departement des R6trovirus, Institut Pasteur, 28 rue du Docteur Roux, 75724 Paris Cedex 15, France Submitted February 25, 1992 In contrast with random-primed RACE (1) and RNA-PCR (2) which can be done to amplify respectively specific 5'- or 3'-cDNA ends, random-PCR (rPCR) can be used to amplify the whole cDNA population, from 5' to 3' ends, derived from small amounts of RNA transcripts. This quick rPCR method allows construction of whole cDNA libraries from as little as one eukaryotic cell, 100 parasites or bacteria, or 105 particles of a 9 kb retrovirus, so lowering the threshold. Libraries constructed in this manner can then be screened with specific DNA probes or antibodies. Moreover, cDNA amplified by rPCR are powerful tools for the screening of substracted libraries. The rPCR was tested here on several MS2 phage RNA dilutions, either respectively 5 x 10-4, 5 x l0-5, 5x 10-6 and 5 x 10-7 Fg MS2 RNA (Serva), and on a control assay with no RNA. Using a 26 nucleotides primer containing a random hexamer at its 3' end (Universal primer-dN6; 5'-GCCGGAGCTCTGCAGAATTCNNNNNN-3'), the first-strand cDNA was prepared as follows. MS2 RNA was suspended in 6 pl of distilled water, heated to 650C for 5 mi, rapidly cooled on ice and reverse transcribed after addition of 0.5 Il (20 units) RNAsin, 1.25 ;d lOxreverse transcription buffer (500 mM Tris-HCI pH 8.3 at 43°C, 800 mM NaCl, 80 mM MgCl2, 50 mM DT), 1.25 Al each dNTP (10 mM), 1.5 jd Universal primer-dN6 (0.1 pg4l), 2 tl (16 units) AMV reverse transcriptase. Incubation was at 430C for 1 h. The reaction was then boiled for 2 min and rapidly cooled on ice. For second-strand cDNA synthesis, the following components were then added; 24.25 1l distilled water, 10 ILI 5 xKlenow buffer (kit multiprime/Phrmacia), 1.25 p1 dCTP (100 mM), 2 ,ll Klenow fragment (8 units). After 30 min incubation at 37°C, the sample was purified on a Chroma Spin400 column (Clontech) to eliminate the excess of Universal primer-dN6. Amplification of the randomly synthetised double-strad cDNA population was then performed on a 1 I1 aliquot in presence of the Universal primer essentially as described by Cetus (3). Briefly, amplification took place in 50 ILI reaction mixtures containing double-strand cDNA in 10 mM Tris-HCI pH 8.3, 50 mM KCI, 1.5 mM MgCl2, 0.01% gelatin, 500 F&M each dNTP, 1 isM of Universal primer and 1.5 unit Taq polymerase (Cetus). The samples were subjected to 40 cycles of amplification, 94°C-1 min, 55°C-I min, 72°C-3 min. Final amplification products analysed on agarose gel (Figure 1) showed DNA populations essentially distributed from 0.4 to 3 Kb with a maximum number of copies around 0.8 Kb; rPCR on total eukaryotic RNA gives routinely DNA fragments around 0.5 Kb. Care has to be taken with the Universal primer-dN6 concentration as it govems both the size of the synthesized cDNA
and the efficiency of the reactions. Using 1 pl from a 50 ul assay, as little as 10-6 sg RNA was efficiendy amplified with the rPCR method (lane d). However, less sting material can generate amplification of one or a few cDNA species that are not represnative of the initial RNA populato (lane f); this was supported here by the use of a pure MS2 RNA which is only
3.6 kb long. The rPCR is actually used in our laboraty in attempt to identify putative retrovirus sequences poorly represented in human culture cell supernatant. This technique should also be applicable for amplification of genomic DNA by replacing the reverse transcriptase with a Klenow fragment step.
ACKNOWLEDGEMENTS This work was supported by grants from the nstitut Pasteur and the Centre National de la Recherche Scientifique (CNRS). We thank Doctor Simon Wain-Hobson and Professo Luc Montagnier for their encouragement in this work as well as for laboratory facilities.
REFERENCES 1. Harvey,J. and Darlison,M.G. (1991) Nucleic Acids Res 14, 4002. 2. Fritz,J.D., Greaser,M.L. and Wolff,J.A. (1991) uclkic Acds Res. 13, 3747. 3. Saild,R.K., Gelfand,D.H., Stroffel,S., Scharf,S.L, Higuchi,R., Horn,G.T., Mullis,K.B. and Erlich,H.A. (1988) Science 239, 487-491.
a b c d e f
Kb 12.2 5.0
-
3.0
_
1.0
-
0.5
-
0.3
-
Flgure 1. Agarose gel electrophoretic analysis of rPCR prohdct encoding MS2 virus seqences. For Southern analysis, 10 #1 of the PCR products were sed in 1.0% agarose gel, transfered to nitod1lulowe aid hybridized eleCtro with3 P-labeled MS2 cDNA probes primed with a rndon hexamer. Lanes c to f correspond respectively to 10-5, 10-6, 10-7 and i-3 g of MS2 RNA. Lane a is an amplification of water and lane b is a rPCk product done with no RNA. The relative positions of size markers are shown in kb.
