1644-1645 Nucleic Acids Research, 1995, Vol. 23, No. 9
Spilinkerettes improved vectorettes for greater
efficiency in PCR walking
Rebecca S. Devon*, David J. Porteous and Anthony J. Brookes
MRC Human Genetics Unit, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, UK
Received October 3, 1994; Revised and Accepted November 25, 1994
Vectorettes enable PCR amplification of DNA sequences which
lie between a single known primer and a nearby restriction site.
They have been applied in the isolation of end fragments from
YAC recombinants (1) and also in PCR walking (2). Vectorettes
consist ofa double-stranded linker sequence with a central region
ofmismatch and a cohesive end suitable for ligation to restriction
enzyme-digested DNA (Fig. 1). The vectorette primer employed
in the PCR is exactly the same sequence as the mismatched
portion of the upper strand, and therefore cannot anneal to and
initiate priming from the vectorette until its complementary
sequence has been synthesised by polymerase extension from the
specific target DNA primer.
The specificity of the vectorette PCR reaction is, however, not
absolute. Illegitimate products result from non-specific annealing
ofeitherprimerandfrom areaction we term 'end-repair' priming.
End-repair priming involves the free cohesive ends of unligated
vectorettes and inserts which are based on restriction sites that
produce 5' overhangs. These ends are filled in during the first
cycle of the PCR reaction. After the subsequent denaturing step,
these ends are able to anneal to each other with sufficient stability
to initiate priming (3). Extension across a vectorette sequence at
either end of the insert molecule will result in the production of
a sequence complementary to the vectorette primer, allowing
exponential PCR amplification without involvement of the
specific target DNA primer. Vectorette dimers will also be
produced by end-repair priming, but will not be amplified in the
PCR because the two halves will form a stable hairpin structure
which will be refractory to amplification.
The splinkerette we describe here is designed to decrease
end-repair priming. Rather than a central DNA mismatch, the
splinkerette incorporates a hairpin structure on the bottom strand
(Fig. 1). The primer is ofthe same sequence as the top strand and
therefore, as with the vectorette, is unable to act as a primer until
the complement of this strand has been synthesised. In a PCR
reaction the free 3' end ofthe bottom strand will flipback on itself
to form a hairpin and begin elongation further along the bottom
strand. The resulting double-stranded structure is stable and is
functionally removed from further reaction. It is therefore not
able to cause end-repair priming. Furthermore, in the splinkerette
system, only the top strand is available to act as a non-specific
primer whereas in the vectorette system both the top strand
(which after end-repair is four bases longer than that of the
splinkerette) andthebottom strandcould cause mispriming in this
way. Splinkerettes are not kinased, and as such there is no
Vectorette
5 -J
31 _
primer _
I -~i 3' cohesive
_ 5' end
i mismatched region---
Splinkerette
priner >
5; 3' cohesive
_ end
Figure 1. Schematic diagram of the vectorette and splinkerette.
covalent bond between the splinkerette bottom strand and the
DNA to be amplified. This avoids elongation from the hairpin
along the whole length of the insert molecule.
Here we demonstrate the practical advantages of splinkerettes
over vectorettes by comparing the efficiency ofboth in producing
unique products from the complex resource of total human
genomic DNA. To enable direct comparison, splinkerettes and
vectorettes were designed to be compatible with a single primer.
For small target PCR products (. 500 bp) it was found that the
efficiency ofboth the splinkerette and vectorette was adequate to
produce the expected products after one or two rounds of PCR
(data not shown). However, Figure 2 shows the advantages of
splinkerettes over vectorettes in the amplification of larger
fragments where formation ofthe target product may be hindered
by increased competition from end-repair primed artefacts. A 2.3
kb fragment from the WTl gene (4) was successfully amplified
only with the splinkerette. We ascribe the reproducible failure of
vectorette PCR to amplify this fragment, and the smear of
non-specific products instead resulting, to end-repair priming.
The usefulness of splinkerette PCR has been further demonstrated
by the routine isolation of long end fragments (up to 3.6
kb) from YAC recombinants (5).
*
To whom correspondence should be addressed
Q-D, 1995 Oxford University Press
Nucleic Acids Research, 1995, Vol. 23, No. 9 1645
M 1 2 3 4 REFERENCES
I Riley, J., Butler, R., Ogilvie, D., Finniear, R., Jenner, D., Powell, S.,
Anand, R., Smith, J.C. and Markham, A.F. (1990) Nucleic Acids Res. 18,
2887-2890.
2 Arnold, C. and Hodgson, I.J. (1991) PCR Methods Appl. 1, 39-42.
3 Tadokoro, K., Oki, N., Fujii, H., Oshima, A., Inoue, T. and Yamada, M.
(1992) Jpn. J. Cancer Res. 83, 1198-1203.
4 Sommer, R. and Tautz, D. (1989) Nucleic Acids Res. 17, 6749.
5 Evans, K.L., Devon, R.S., Brookes, A.J., Porteous, D.J. et al., (1995)
Genomics, in press.
6 Chou, Q., Russell, M., Birch, D.E., Raymond, J. and Bloch, W. (1992)
Nucleic Acids Res. 20, 1717-1723.
7 Don, R.H., Cox, PT., Wainwright, B.J., Baker, K. and Mattick, J.S. (1991)
Nucleic Acids Res. 19, 4008.
Figure 2. Comparison of splinkerettes and vectorettes in PCR walking. The
target PCR product was a 2.3 kb fragment derived from the WTI gene (3).
Vectorettes were obtained from ICI. Splinkerettes were made by duplexing the
top strand (5'-CGA ATC GTA ACC GTT CGT ACG AGA ATT CGT ACG
AGA ATC GCT GTC CTC TCC AAC GAG CCA AGG-3') and the bottom
strand (5'-GATCCC TTG GCTCGTTTTTTrTTG CAA AAA-3')by mixing
the oligonucleotides (150 gg/ml each) at 90°C in 10 mM Tris-HCI (pH 7.4),
5 mM MgCl2 and leaving to cool to room temperature. Total human genomic
DNA was digested with BamHI and ligated with a 15 x molar excess of
splinkerettes or vectorettes in 20 ,l for 41/2 h at room temperature. PCR
combined Hot Start (6) and Touchdown (7) protocols and conditions were as
follows: denaturation, 95°C for 30 s in the first cycle and 15 s thereafter;
annealing, for 1 min at 71 'C initially, decreasing by 2°C to 61 'C per cycle and
61 'C thereafter; extension, 72°C for 2 min (cycles 1-10), then 4 min (cycles
11-20) and finally 6 min (cycles 21-30). In the primary PCR reaction 1 1l of
ligation product was amplified in 50 p1 using 400 ng WT1 primer (5'-CCA
AGG GCC GTG AGG ATA GCG GAA G-3') and 250 ng splinkerette/vectorette
primer (5'-CGA ATC GTA ACC GTT CGT ACG AGA A-3'). Secondary
PCR was performed using 1 jl primary PCR product, 400 ng internal WT1
primer (5'-GCA CGC AGG CAC TGG CCC CCG ACA T-3') and 250 ng
internal splinkerette/vectorette primer (5'-TCG TAC GAG AAT CGC TGT
CCT CTC C-3'). Tracks 1 and 2 show 10 1l vectorette and splinkerette
secondary PCR product respectively resolved on a 1% agarose gel. The marker
is the 1kb ladder (Gibco-BRL). No product was visible after primary PCR for
either the vectorette or the splinkerette. Tracks 3 and 4 show results ofSouthem
blot analysis of tracks 1 and 2, respectively, with a 1.1 kb probe derived from
widtin the target PCR product.