©
1997 Oxford University Press
1080-1081
Footnote
PCR-mediated direct gene disruption in
Schizosaccharomyces pombe
PCR-mediated direct gene disruption in Schizosaccharomyces pombe
Rupinder
Kaur
,
Susham S.
Ingavale
and
Anand K.
Bachhawat*
Institute of Microbial Technology, Sector 39-A,
Chandigarh
160 036,
India
Received October 21, 1996;
Revised and Accepted January 9, 1997
ABSTRACT
We have examined the feasibility and efficiency of PCR-mediated direct gene disruptions in the fission yeast
Schizosaccharomyces pombe
. In the present study, the
S.pombe
ura4
+
gene was amplified by PCR with oligonucleotides that had short flanking regions
(
~
40 bp) to the target gene. Using this purified PCR product we were able to
disrupt genes in an
S.pombe
strain bearing a
ura4
deletion, with an efficiency ranging between 1 and 3% among selected
transformants. The results indicated that despite
S.pombe
's preference for non-homologous or illegitimate recombination, even very short stretches of
homologous regions could be used to target genes at a defined frequency in this
organism. The successful disruption of four independent genes (
sts1
+
, gcs1
+
, gsh2
+
and
hmt1
+
) by this method further demonstrates that, despite the relatively low
efficiency, the method is very feasible, and it's simplicity, especially when
coupled to phenotype-based screening, should greatly facilitate disruption of genes in
S.pombe
.
The ability to carry out targeted gene disruptions by homologous recombination
is one of the major means for functional analysis of different genes in yeast.
The need for rapid and efficient strategies for carrying out such gene
disruption have grown with the progress, and now the completion of the
Saccharomyces cerevisiae
genome sequencing project. Different disruption strategies and methods have
recently been described for use in
S.cerevisiae
. The simplest method for gene disruption in this yeast uses a PCR-mediated gene disruption strategy where the selection marker is amplified
by PCR with oligonucleotides that have ~35-50 bp flanking region on either side. This short flanking region on
either side of the selection marker gene has been shown to provide sufficient
length of homology to permit a high frequency of gene disruptants at the
targeted locus upon transformation in
S.cerevisiae
(
1
-
4
).
The fission yeast
Schizosaccharomyces pombe
is also being intensely studied as a model eukaryote, and indeed, in many
features it appears to have a closer relationship to higher eukaryotes than
S.cerevisiae
. A major drawback to the functional analysis of genes in
S.pombe
, however, is the relative difficulty of gene disruption through homologous
recombination in this organism. This deficiency is being even more acutely felt
with the rapid release of sequences from the Project for the sequencing of the
S.pombe
genome at the Sanger Sequencing Centre, UK. Although gene disruptions by
homologous recombination in this yeast can be carried out, the requirement for
longer regions of homology, and the apparent preference for illegitimate or non-homologous recombination in this yeast have been a major deterrent to the
functional analysis of genes by use of the rapid and simple methods available
for
S.cerevisiae
(
5
,
6
)
.
Gene disruption in
S.pombe
therefore has invariably required multiple cloning steps. We have sought to
determine whether the simple PCR disruption methods in use in
S.cerevisiae
were at all feasible in
S.pombe
and, if so, with what kind of efficiency. In this report we describe that,
contrary to expectations, the disruption of genes in
S.pombe
with flanking regions of ~40 bp of the target gene are indeed possible. Furthermore, the frequency of
correct disruptions with such short flanking regions range between 1 and 3%
among selected transformants.
We initially carried out our investigations with the
S.pombe
sts1+
gene (
7
). This gene, which is the
S.pombe
homologue of the
S.cerevisiae
ERG4
gene, encodes the last enzyme in ergosterol biosynthesis (
8
). Disruption of
sts1
+
has been shown to lead to an increased drug-sensitivity and calcium ion-sensitivity (
7
). Oligonucleotides for carrying out the PCR-mediated disruption of this gene were designed that could amplify the
S.pombe ura4
+
gene from a
ura4
+
containing plasmid with ~40 bp of flanking region on either side that had homology to
sts1
+
. These oligonucleotides were STSDEL1, 5'-ATGGCTTATGAAAAGTTTGGTTTTATGCTTATCTTTTGG
CTATATGTATGCAT
-3' and STSDEL2, 5'-AATCAGCAGTATAATGAATTTTACGAGCGTACCTATACCAAC
ACCAATGTTTATAACC
-3'. The underlined regions correspond to
ura4
+
sequences, and the remaining sequences are the flanking residues; 39 bp
flanking residues in the case of STSDEL1 and 42 bp in the case of STSDEL2. The
template DNA was plasmid pBSK-
ura4
+
, a pBSK plasmid containing the
S.pombe ura4
+
gene cloned into the
Hin
dIII site of pBSK. The PCR reaction was carried out using a mixture of
Taq
and vent polymerases. After amplification by PCR, the PCR product was purified
by silica particles (
9
) and ~0.5-1.0 [mu]g of DNA was used to transform an
S.pombe ura4-
[Delta]
18
strain that carried a complete deletion in the
ura4
+
gene (
10
). (A
ura4
[Delta]
-18
strain that carried a complete disruption in the
ura4
+
locus has been used in all these experiments so as to minimize recombinations
at the
ura4
+
locus.)
Transformation was carried out by the lithium acetate method exactly as
described earlier (
11
), and transformants were selected on minimal medium containing adenine but
lacking uracil. Transformants obtained were patched onto YES (Yeast Extract +
Supplements) medium and then replica plated onto minimal medium plus adenine
but lacking uracil (to record the stable Ura
+
colonies), as well as on YES medium containing either cycloheximide, crystal
violet or calcium chloride. Approximately 45% of the colonies were stable Ura
+
colonies. Of these, colonies that showed sensitivity to all three media were
picked up as putative
sts1
disruptants. These were confirmed by PCR (data not shown). The combined data
from several independent experiments is shown in Table
1
. An overall frequency of ~2% was obtained indicating that
S.pombe
genes could be disrupted by this PCR-mediated strategy. To examine if this frequency extended to other unlinked
genes, we carried out disruptions of
gcs1
+
([gamma]-glutamyl cysteine synthetase) located on chromosome I (cosmid
c22F3),
gsh2
+
(glutathione synthase) located on chromosome I (cosmid c3F10) and
hmt1
+
(vacuolar heavy metal transporter) located on chromosome III, using similarly
designed oligonucleotides that also had short flanking residues to the target
gene that have been described elsewhere (
11
). The oligonucleotides designed for the
gcs1
+
deletion had 40 and 42 bp flanking residues, the oligonucleotides for the
gsh2
+
deletion had 42 and 41 bp flanking residues, while those for the
hmt1
+
deletion had 42 and 40 bp flanking residues. We were able to obtain successful
disruptions of all these genes using the PCR products obtained with these
oligonucleotides, and the frequencies at which they were obtained are shown in
Table
1
. The disruptants were initially screened by their phenotypes.
gcs1
and
gsh2
disruptants are both glutathione auxotrophs and cadmium-sensitive (
11
) while
hmt1
disruptants are cadmium-sensitive (
12
). The putative disruptants were picked up and disruptions at the target locus
were confirmed by PCR (data not shown). The frequencies obtained in the case of
these genes also lay between 1 and 3%, and reveal that, despite the propensity
of
S.pombe
to carry out illegitimate recombination (
13
,
14
), PCR-mediated gene disruption with short flanking regions is indeed feasible in
S.pombe.
To examine whether in the targeted disruptants we were also observing unlinked
integrations in addition to the target site, we examined the targeted
disruptions for the possibility of multiple integrations by Southern blotting
using the
ura4
+
gene as a probe. Among the disruptants tested (two
sts1
[Delta], three
gcs1
[Delta], one
hmt1
[Delta] and one
gsh2
[Delta]) we were unable to observe a second unlinked integration in any of the
seven cases (data not shown). Our results indicate that although the frequency
of gene disruptions seen in
S.cerevisiae
with similar flanking regions and with equivalent strains carrying deletions in
the selection markers is much higher (>50%) (
3
), the 1-3% frequency that we have observed in
S.pombe
still makes this method a very workable one, as we have demonstrated through
successful disruption of four independent genes. This observation should
encourage
S.pombe
researchers to carry out gene disruptions of their target genes of interest
using the PCR-mediated strategy. The method could be coupled with either phenotype-based screening (wherever the phenotype is either known or
suspected) or alternatively through PCR screening by using pools of putative
disruptants. While phenotype-screening makes the method almost as convenient as it is for
S.cerevisiae
, we believe that even screening by PCR of pooled samples would be generally
preferable to multiple cloning steps that are currently being used for gene
disruption in
S.pombe.
Table 1
Frequency of gene disruption with short flanking regions in
S.pombe
|
Target gene
|
No. of stable Ura
+
|
Colonies with disruption
|
% Frequency
|
|
|
colonies
|
at target locus
|
|
sts1
+
|
301
|
5
|
2
|
|
gcs1
+
|
90
|
3
|
3
|
|
gsh2
+
|
94
|
1
|
1
|
|
hmt1
+
|
61
|
1
|
2
|
ACKNOWLEDGEMENTS
We would like to thank Dr Jagmohan Singh for generous gifts of
Taq
polymerase and for the pBSK-
ura4
+
plasmid. R.K. and S.I. are Senior Research Fellows of the Council of Scientific
and Industrial Research, India. This work was supported in part by a Grant-in-aid project (No. BT/R&D/15/40/93) from the Department of Biotechnology, Government of
India.
REFERENCES
1 Baudin, A., Ozier-Kalogeropoulos, O., Denouel, A., Lacroute, F. and Cullin, C. (1993) Nucleic Acids Res., 21, 3329-3330. MEDLINE Abstract
2 Wach, A., Brachat, A., Pohlmann, R. and Philippsen, P. (1994) Yeast, 10, 1793-1808. MEDLINE Abstract
3 Manivasakam, P., Weber, S. C., McElver, J. and Schiestl, R. E. (1995) Nucleic Acids. Res. 23, 2799-2800. MEDLINE Abstract
4 Lorenz, M. C., Muir, R. S., Lim, E., McElver, J., Weber, S. C. and Heitman, J. (1995) Gene 158, 113-117.
5 Tateyabashi, K., Kato, J. and Ikeda, K. (1994) Mol. Gen. Genet., 244, 111-119.
6 Keeney, J. B. and Boeke, J. (1994) Genetics, 136, 849-856.
7 Shimanuki, M., Goebl, M., Yanagida, M. and Toda, T. (1992) Mol. Biol. Cell, 3, 263-273. MEDLINE Abstract
8 Lai, M. H., Bard, M., Pierson, C. A., Alexander, J. F., Goebl, M., Carter, G. M. and Kirsch, D. R. (1994) Gene, 140, 41-49.
9 Kaur, R., Kumar, R. and Bachhawat, A. K. (1995) Nucleic Acids Res., 23, 4932-4933. MEDLINE Abstract
10 Grimm, C., Kohli, J., Murray, J. and Maundrell, K. (1988) Mol. Gen. Genet., 215, 81-86. MEDLINE Abstract
11 Chaudhuri, B., Ingavale, S. and Bachhawat, A. K. (1997) Genetics, 145, 75-83. MEDLINE Abstract
12 Ortiz, D. E., Kreppel, D., Speiser, G., Scheel, G., McDonald, G. and Ow, D. W. (1992) EMBO J., 11, 3491-3499.
13 Wright, A. P. H., Maundrell, K. and Shall, S. (1986) Curr. Genet., 10, 503-508.
14 Grallert, B., Nurse, P. and Patterson, T. E. (1993) Mol. Gen. Genet., 238, 26-32. MEDLINE Abstract
Return
*To whom correspondence should be addressed. Tel: +91 172 690004; Fax: +91 172
690632; Email: anand@imtech.ernet.in
