 .

13: 763–768 (1997)

Applications of the Long and Accurate Polymerase

Chain Reaction Method in Yeast Molecular Biology:
Direct Sequencing of the Amplified DNA and Its
Introduction into Yeast
YOKO TAKITA1, MANABU TAKAHARA1, SATORU NOGAMI1, YASUHIRO ANRAKU1 AND
YOSHIKAZU OHYA1*
1
Department of Biological Sciences, Graduate School of Science, University of Tokyo, Hongo, Bunkyo-ku,
Tokyo 113, Japan

Received 22 September 1996; accepted 2 January 1997

A DNA fragment longer than 10 kb can be amplified by the long and accurate polymerase chain reaction (LA-PCR)
method. We demonstrate here applications of this technique in molecular biological studies of Saccharomyces
cerevisiae. We have shown that DNA fragments amplified by LA-PCR can be directly used as a template in the
chain-termination sequencing protocol, making it possible to quickly identify the DNA insert of yeast genomic
library clones. We have also shown that the amplified yeast DNA can easily be introduced into yeast by
co-transformation with linearized vector DNA. Overlapping DNA between the amplified yeast fragment and the
vector must be more than 20 bp long in order to obtain 90% or more correct recombinant plasmids. These results
suggest that simple amplification of yeast clones by LA-PCR can replace the previous procedures of yeast clone
recovery, consisting of transformation of Escherichia coli, propagation of plasmids in E. coli and preparation of
plasmid DNA. ? 1997 by John Wiley & Sons, Ltd.

Yeast 13: 763–768, 1997.

No. of Figures: 1. No. of Tables: 3. No. of References: 15.

  — Saccharomyces cerevisiae; long and accurate (LA)-PCR; homologous recombination; co-
transformation

INTRODUCTION 1976). A number of hybrid plasmids have been

In order to characterize DNA clones that comp-
lement or suppress yeast mutations in a gene of
interest, the first step of a research program is to
isolate preparative amounts of the pure DNA
clone. From the beginning of yeast molecular
biology, this has been achieved by recovering yeast
clones into Escherichia coli and preparing their
plasmids (Ratzkin and Carbon, 1977; Struhl et al.,
*Correspondence to: Yoshikazu Ohya.
Contract grant sponsor: Ministry of Education, Science, Sports
and Culture of Japan.
designed and constructed thus far for the purpose
of shuttling DNA fragments between yeast and
E. coli. In the late 1980s development of the
polymerase chain reaction (PCR) made it possible
to amplify yeast clones with yeast DNA as a
template. Although the original PCR method
(Saiki et al., 1988) was limited to amplification
of less than 1–2 kb, a new technique, called long
and accurate (LA)-PCR was developed more
recently, with which DNA fragments up to 35 kb
long can be amplified with high fidelity (Barnes,
1994; Cheng et al., 1994). Despite its potential

CCC 0749–503X/97/080763–06 $17.50

? 1997 by John Wiley & Sons Ltd
764
usefulness, however, few papers have yet reported
application of the LA-PCR technology in yeast
(Ohya, 1996).
Here, we demonstrate direct sequencing of DNA
fragments longer than 10 kb generated by LA-
PCR. Efficient sequencing of DNA fragments over
10 kb long prompted us to examine whether this
protocol is amenable to quick identification of
yeast genomic library clones. After obtaining the
PCR product and its sequence, it is necessary to
reintroduce the DNA fragment into yeast. Taking
this into account, we also examined whether the
amplified yeast DNA was introduced efficiently
into yeast by co-transformation with linearized
vector DNA.
MATERIALS AND METHODS
Yeast strains and plasmids used in this study
YTY228 (MATa ade2 his3 lys2 leu2 trp1 ura3
ade3::cmd1–228:HIS3 cmd1::TRP1 [pST26]) was
made by transformation of YOC228 (Ohya and
Botstein, 1994) with pST26. YOC729 (MATá ade2
his3 lys2 leu2 trp1 ura3 ade3::rho1–3:HIS3
rho1::LYS2) and YOC755 (MATá ade2 his3 lys2
leu2 trp1 ura3 ade3::rho1–5:LEU2 rho1::HIS3)
(Qadota et al., 1996) were used in co-
transformation experiments.
pYO702 and pYO922, containing a 2·9 kb
BamHI-ClaI fragment of RHO1, were derived
from pRS314 and pRS316, respectively. pYEC101
(Qadota et al., 1996), containing a 10 kb genomic
RHO1 fragment, and pYO922 were used as
templates for LA-PCR.
Preparation of yeast DNA for LA-PCR
Yeast cells were grown to saturation in 2 ml of
synthetic minimal medium, harvested by centrifu-
gation to obtain 2–4#107 cells, suspended in 30 ìl
of STES buffer (0·5 -NaCl, 0·2 -Tris–HCl
(pH 7·6), 0·01 -EDTA, 1% SDS) and broken by
vortexing for 2 min with 0·4 mm glass beads. After
proteins were removed by phenol/chloroform
extraction, the yeast DNA was further purified
with DNA prep (Asahi-Glass, Tokyo, Japan).
Finally, the yeast DNA was dissolved in 20 ìl TE
(10 m-Tris–HCl (pH 7·6), 1 m-EDTA) contain-
ing 100 ìg/ml RNase A and used as a template to
amplify yeast DNA by LA-PCR.
Amplification of yeast DNA for DNA sequencing
Long DNA fragments were amplified to give a
high yield with a LA-PCR kit (Version 2; Takara,
? 1997 by John Wiley & Sons, Ltd
.   .
Kyoto, Japan). The final reaction mixture (50 ìl)
contained 0·2 ìg of each of the primers and 1 ìl of
the purified yeast DNA solution. The reaction
was taken through 40 cycles in a DNA Engine (MJ
Research, Watertown, MA, U.S.A.): each cycle
consisted of 94)C for 20 s (denaturation) and 68)C
for 4 min (annealing and extension). After 0·7%
agarose gel electrophoresis, the LA-PCR product
was purified with DNA prep, and used for the
sequencing reaction. The Taq DNA polymerase
dye terminator sequencing protocol and auto-
mated DNA sequencing apparatus (model 373A;
Applied Biosystems Inc., Foster City, CA, U.S.A.)
were used to determine the DNA sequence.
Introduction of amplified DNA into yeast
DNA fragments were amplified by LA-PCR,
and purified with DNA prep as described above.
In order to amplify RHO1 fragments containing
various lengths of overlap (20 bp or less) with the
vector DNA (pRS314), RN20 and RC20, RN15
and RC15, RN10 and RC10, RN5 and RC5, or
RN0 and RC0 (Table 1) were used as primers. To
amplify the fragments with overlapping regions of
40 bp or more, RN100 and RC100, RN80 and
RC80, RN60 and RC60 or RN40 and RC40
(Table 1) were used as primers, the fragment
amplified by RN20 and RC20 was used as a
template. pRS314 was digested by BamHI and SalI
or KpnI and ClaI to yield a linear plasmid. After
the DNA concentration was determined by its
absorbance at 260 nm, the amplified DNA and the
linearized vector were mixed and used to trans-
form yeast strains (YOC729 and YOC755) with
the standard Li-method (Schiestl and Gietz, 1989).
RESULTS AND DISCUSSION
Since many yeast libraries have been constructed
by insertion of about 10 kb of random yeast
genomic sequences into the BamHI site of YCp-
type (Ohya et al., 1986) or YEp-type vectors
(Carlson and Botstein, 1982; Yoshihisa and
Anraku, 1989), the inserted DNA can be easily
amplified with LA-PCR by using a pair of primers
upstream and downstream of the BamHI site
(Figure 1). At first, we examined the primers suited
for efficient amplification and direct sequencing of
the DNA inserts for commonly used yeast genomic
libraries (Carlson and Botstein, 1982; Ohya et al.,
1986; Yoshihisa and Anraku, 1989).
We analysed a pST26 (Ohya, 1996) plasmid
containing the yeast calmodulin gene which was
 . 13: 763–768 (1997)
-  765
Table 1. Primers used in this study.

Primera Sequence

Primers used for LA-PCR amplification

Tc-N ATC GAC TAC GCG ATC ATG GCG ACC ACA CCC GTC CT
Tc-R CGA TAT AGG CGC CAG CAA CCG CAC CTG TGG CGC CG

Sequencing primers
N1 GCG ACC ACA CCC GTC CTG T
N2 CGC GAT CAT GGC GAC CAC ACC CG
N3 CGA TCA TGG CGA CCA CAC CC
N4 CTA CGC GAT CAT GGC GAC CAC ACC
N5 CGA CTA CGC GAT CAT GGC GA
C1 CCA CGA TGC GTC CGG CGT AG
C2 GGT GAT GCC GGC CAC GAT GCG TCC
C3 CAG CAA CCG CAC CTG TGG CGC
C4 GCC AGC AAC CGC ACC TGT GGC
C5 CGA TAT AGG CGC CAG CAA CCG

for introduction of RHO1 into yeastb

Primers
RN500 GTGAACCATCACCCTAATCAAGTTTTTTGG
RC500 CGTCGATTTTTGTGATGCTCGTCAGGGGGG
RN0 CGACGACATATTCGAGGTTG
RC0 CAAGTAGCTATAGTTAGAGC
RN5 GGATCCGACGACATATTCGAGGTTG
RC5 ATCGACAAGTAGCTATAGTTAGAGC
RN10 CTAGTGGATCCGACGACATATTCGAGGTTG
RC10 ACGGTATCGACAAGTAGCTATAGTTAGAGC
RN15 TAGAACTAGTGGATCCGACGACATATTCGAGGTTG
RC15 GGTCGACGGTATCGACAAGTAGCTATAGTTAGAGC
RN20 CGCTCTAGAACTAGTGGATCCGACGACATATTCGAGGTTG
RC20 CTCGAGGTCGACGGTATCGACAAGTAGCTATAGTTAGAGC
RN40 AGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGGATC
RC40 AGCTGGGTACCGGGCCCCCCCTCGAGGTCGACGGTATCGA
RN60 CTCACTATAGGGCGAATTGGAGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGGATC
RC60 CCCTCACTAAAGGGAACTAAAGCTGGGTACCGGGCCCCCCCTCGAGGTCGACGGTATCGA
RN80 GCCAGTGAATTGTAATACGACTCACTATAGGGCGAATTGGAGCTCCACCGCGGTGGCG-
GCCGCTCTAGAACTAGTGGATC
RC80 TACGCCAAGCTCGGAATTAACCCTCACTAAAGGGAACTAAAGCTGGGTACCGGGCCC-
CCCCTCGAGGTCGACGGTATCGA
RN100 CACGACGTTGTAAAACGACGGCCAGTGAATTGTAATACGACTCACTATAGGGCGAATTG-
GAGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGGATC
RC100 CAGCTATGACCATGATTACGTACGCCAAGCTCGGAATTAACCCTCACTAAAGGGAACT-
AAAGCTGGGTACCGGGCCCCCCCTCGAGGTCGACGGTATCGA

a
Suitable primers for sequencing analysis are shown in Figure 1.
b
The overlapping regions are underlined.

screened from a YEp-13-based genomic library
(Yoshihisa and Anraku, 1989) based on its comp-
lementation of a temperature-sensitive calmodulin
mutation (Ohya and Botstein, 1994). Yeast DNA
was isolated from the strain (YTY228) harboring
pST26 by using a smash-and-grab method
(Strathern and Higgins, 1991) as described in
Materials and Methods. We found that primers
efficient for amplification of the DNA insert are
the Tc-N and Tc-R oligonucleotides described in
Table 1, which are complementary to segments
upstream and downstream of the BamHI site in the
tetracycline gene on YEp13 (Figure 1). After 0·7%
agarose gel electrophoresis, the LA-PCR product

? 1997 by John Wiley & Sons, Ltd  . 13: 763–768 (1997)
766 .   .

Figure 1. Schematic representation of the YEp13 library clone marked with primers
for LA-PCR amplification (Tc-N and Tc-R) and primers for sequencing analysis (N1,
N2, N3, N4, N5, C1, C2, C3, C4 and C5). Arrows indicate the 5* ] 3* orientation of
the primers. Bold arrows represent suitable primers for sequencing analysis. Only
DNA regions covered by Tc-N and Tc-R are to scale.

was purified with DNA prep, used in the sequenc-
ing reaction, and applied to the automated DNA
sequencing apparatus.
Among ten primers that we prepared for the
DNA sequencing analysis (Table 1), the most
accurate base detection (>98%) was possible with
N1 for the analysis of the left end of the insert.
Satisfactory base calling was provided equally with
C1, C2 and C4 for the right end in the automated
DNA sequencing system. The better sequencing
primers contain a slightly higher GC content than
the other primers. Sequencing about 400 bases at
both ends of the pST26 insert revealed that the
DNA insert maps from 1207 bases downstream of
the end of YBR107 to 2545 bases downstream of
the end of YBR112 on chromosome II (Feldmann
et al., 1994), which contains the calmodulin gene.
Besides identification of the calmodulin clone, this
protocol was utilized successfully in the easy iden-
tification of several other genomic library clones
(unpublished results).
Isolation of S. cerevisiae genes by complement-
ation is one of the most powerful tools in yeast
molecular biology. After finding genomic DNA
library clones, the next step is determination of the
complementing genes. With the direct sequence
technique using a LA-PCR product as a template,
one can quickly obtain information about the
yeast genomic clone. It takes only several hours to
prepare the DNA template for sequencing, starting
from the grown yeast culture. This simple pro-
cedure is substituted for the conventional 3-day
procedure that includes transformation of the
yeast plasmid DNA into E. coli, propagation of
E. coli cells harboring the plasmid, and purifi-
cation of the plasmid. Now that the complete
sequence of the yeast genome is known, infor-
mation from both ends of the DNA insert is
enough to provide the entire sequence of the insert.
In addition to determining the sequence of library
inserts, this technique should be generally appli-
cable to analysis of any DNA at long range that
the conventional PCR technique fails to cover.
When DNA clones are recovered from yeast,
they must then be reintroduced into yeast. Taking
this situation into account, we next examined
whether the amplified DNA fragment could be
introduced easily into yeast. LA-PCR is also
applicable here, because its high fidelity is sufficient
for functional complementation experiments. The
average error rate was determined to be 1·05 per
105 bp for 16 cycles of LA-PCR (Barnes, 1994).

? 1997 by John Wiley & Sons, Ltd  . 13: 763–768 (1997)
-  767
Table 2. Frequency of complementing transformants Table 3. Frequency of complementing transformants
after co-transformation of the amplified RHO1 with after co-transformation of the amplified RHO1 with
linearized pRS314. linearized pRS314.

Amplified Primers Length of

Plasmid (ìg) DNA (ìg) Ts + (%) overlapping Ts +
1st 2nd region (bp) (%)
Linearized pRS314
1 10 93 RN20+RC20 RN100+RC100 100 92
1 1 83 RN20+RC20 RN80+RC80 80 92
10 1 75 RN20+RC20 RN60+RC60 60 93
1 0 0 RN20+RC20 RN40+RC40 40 92
Circular pRS314 RN20+RC20 — 20 88
1 0 0 RN15+RC15 — 15 24
Circular pYO702 RN10+RC10 — 10 4
1 0 96 RN5+RC5 — 5 0
RN0+RC0 — 0 0
After transformation of YOC755 (rho1–5), the temperature-
sensitivity of more than 200 transformants was examined. RHO1 fragments containing various lengths of overlap with
pRS314 were constructed by LA-PCR. After co-transformation
of YOC729 (rho1–3) with the amplified fragment (10 ìg) and
linearized pRS314 (1 ìg), the temperature-sensitivity of more
We tested co-transformation of the LA-PCR- than 100 transformants was examined.
amplified DNA with a linearized vector (Ma et al.,
1987). A 2·9 kb RHO1 fragment was amplified by
LA-PCR with the RN500 and RC500 primers
(Table 1), and pYO922 as a template. A centro- the recombinant plasmid (Table 3). When 15 bp or
mere plasmid, pRS314, was digested with BamHI less overlapped, less than 80% of the transformants
and SalI to yield a linear DNA fragment contain- harbored the wild-type RHO1 gene (Table 3).
ing a 0·5 kb overlapping region with the amplified Similar results were obtained with amplified 1·6,
DNA at both ends. The insert and vector DNAs 1·4, 0·9 and 0·8 kb fragments of FKS1, a gene
were mixed in different ratios and used for trans- encoding a subunit of 1,3,-â-glucan synthase
formation of a yeast temperature-sensitive rho1 (unpublished results).
strain, YOC755. Without the amplified RHO1 We recovered 15 complementing and 5 non-
fragment, all transformants showed a temperature- complementing plasmids in the yeast transform-
sensitive phenotype (Table 2), indicating low ants that were obtained with the 20 bp overlapped
reversion frequency of YOC755. As the ratio of RHO1 fragment, and analysed their DNA
the amplified fragment increased, however, the sequences. We found that all complementing
number of complementing colonies increased. Co- plasmids contained the inserted RHO1 fragment
transformation of 10 ìg of the RHO1 fragment identical to the one made with a classical ligation
with 1 ìg of the linear vector resulted in more than reaction. The non-complementing plasmids had no
90% of the transformants containing recombinant inserted sequence, instead being identical to the
plasmids (Table 2). Given that 4% of the control one resulting from ligation of the gapped plasmid
RHO1 plasmids (pYO702) failed to complement after addition of one nucleotide. These results
the rho1 mutation (Table 2), the co-transformation suggested that an overlap of more than 20 bp
method seems to be an effective method to is necessary for the homologous recombination
introduce LA-PCR-amplified DNA fragments. between the amplified fragment and the linear
We next determined the minimum length of the vector.
overlapping region required for the efficient intro- We have proposed here that a DNA fragment
duction of the amplified DNA. A series of frag- amplified by LA-PCR can be used as a good
ments were constructed that contained a 2·2 kb source of yeast DNA. We have shown that a
RHO1 fragment with a zero to 100 bp overlapping fragment longer than 10 kb amplified by LA-PCR
region at both ends. With 20 bp or more of can be used directly as a template for DNA
overlap, the amplified fragments were introduced sequencing. We have also shown that a DNA
efficiently; 90% or more transformants harbored fragment amplified by LA-PCR can be introduced

? 1997 by John Wiley & Sons, Ltd  . 13: 763–768 (1997)
768
effectively into yeast by co-transformation. These
experimental procedures will be useful not only for
yeast genes which are difficult to propagate in
E. coli, but also for yeast genes which are difficult
to manipulate with classical molecular biological
techniques such as restriction enzyme digestion
and ligation. Moreover, with this protocol, one
can reduce the time required for the DNA recovery
and move on to the functional analysis of the gene.
We assume, however, that this protocol is not
suitable in all circumstances. When two different
pBR322-derived plasmids are present in the same
yeast strain, both insert DNAs would be amplified,
resulting in a mixed population. Also, if the yeast
strain contains plasmids or a DNA region homolo-
gous to the end of the linearized vector, the linear
vector can circularize frequently without incorpor-
ating the amplified DNA fragment. Nevertheless,
we believe that correct understanding of the
homologous recombination process will allow
application of this protocol to other yeast molecu-
lar biological techniques, including random and
site-directed mutagenesis, functional analysis of
genes, and protein engineering.
ACKNOWLEDGEMENTS
We would like to thank K. Richards for critical
reading of the manuscript. This study was sup-
ported in part by a grant-in-aid for Scientific
Research from the Ministry of Education, Science,
Sports and Culture of Japan (Y. O.).
REFERENCES
Barnes, W. M. (1994). PCR amplification of up to 35-kb
DNA with high fidelity and high yield from ë bac-
teriophage templates. Proc. Natl. Acad. Sci. USA 91,
2216–2220.
Carlson, M. and Botstein, D. (1982). Two differentially
regulated mRNAs with different 5* ends encode
secreted with intracellular forms of yeast invertase.
Cell 28, 145–154.
? 1997 by John Wiley & Sons, Ltd
.   .
Cheng, S., Fockler, C., Barnes, W. M. and Higuchi, R.
(1994). Effective amplification of long targets from
cloned inserts and human genomic DNA. Proc. Natl.
Acad. Sci. USA 91, 5695–5699.
Feldmann, H., Aigle, M., Aljinovic, G., et al. (1994).
Complete DNA sequence of yeast chromosome II.
EMBO J. 13, 5795–5809.
Ma, H., Kunes, S., Schatz, P. J. and Botstein, D. (1987).
Plasmid construction by homologous recombination
in yeast. Gene 58, 201–216.
Ohya, Y. (1996). LA-PCR-based quick method for the
identification of genes responsible for the comp-
lementation of Saccharomyces cerevisiae mutations.
BioTechniques 20, 778–779.
Ohya, Y. and Botstein, D. (1994). Diverse essential
functions revealed by complementing yeast calmodu-
lin mutants. Science 263, 963–966.
Ohya, Y., Miyamoto, S., Ohsumi, Y. and Anraku, Y.
(1986). Calcium-sensitive cls4 mutant of Saccharomy-
ces cerevisiae with a defect in bud formation. J.
Bacteriol. 165, 28–33.
Qadota, H., Python, C., Inoue, S. B., et al. (1996).
Identification of yeast RHO1 GTPase as a regulatory
subunit of the 1,3-â-glucan synthase. Science 272,
279–281.
Ratzkin, B. and Carbon, J. (1977). Functional expres-
sion of cloned yeast DNA in Escherichia coli. Proc.
Natl. Acad. Sci. USA 74, 487–491.
Saiki, R. K., Gelfand, D. H., Stoffel, S., et al. (1988).
Primer-directed enzymatic amplification of DNA
with a thermostable DNA polymerase. Science 239,
487–491.
Schiestl, R. H. and Gietz, R. D. (1989). High efficiency
transformation of intact yeast cells using single
stranded nucleic acids as a carrier. Curr. Genetics 16,
339–346.
Strathern, J. N. and Higgins, D. R. (1991). Recovery of
plasmids from yeast into Escherichia coli: shuttle
vectors. Methods Enzymol. 194, 319–329.
Struhl, K., Cameron, J. R. and Davis, R. W. (1976).
Functional genetic expression of eukaryotic DNA
in Escherichia coli. Proc. Natl. Acad. Sci. USA 73,
1471–1475.
Yoshihisa, T. and Anraku, Y. (1989). Nucleotide se-
quence of AMS1, the structure gene of vacuolar
á-mannosidase of Saccharomyces cerevisiae. Biochem.
Biophys. Res. Commun. 163, 908–915.
 . 13: 763–768 (1997)