SEARCH

SEARCH BY CITATION

Keywords:

  • fission yeast;
  • hygromycin B;
  • nourseothricin;
  • marker switch;
  • monomeric RFP;
  • oligonucleotide;
  • PCR

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgements
  8. References

We describe new heterologous modules for PCR-based gene targeting in the fission yeast Schizosaccharomyces pombe. Two bacterial genes, hph and nat, which display dominant drug-resistance phenotypes, are used as new selectable markers in these modules. Both genes have been used successfully in the budding yeast Saccharomyces cerevisiae, in which hph confers resistance to hygromycin B, while nat confers nourseothricin resistance (Goldstein and McCusker, 1999). Vector modules for gene disruption and C-terminal tagging with 3HA, 13Myc and GFP(S65T) are constructed using previously constructed pFA6a–MX6–derived plasmids (Bähler et al., 1998; Wach et al., 1997). In combination with the existing systems that are based upon the G418-resistance gene (kan), triple gene deletions or tags could be constructed. In addition a vector for one-step integration of a monomeric RFP (mRFP) to the C-terminus of proteins of interest is developed. Finally, oligonucleotides that allow a simple marker switch from kan to hph or nat, and vice versa, are described. The new constructs developed here should facilitate post-genomic molecular analysis of protein functions in fission yeast. Copyright © 2005 John Wiley & Sons, Ltd.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgements
  8. References

For over the past 20 years modern yeast genetics using recombinant DNA techniques, including gene deletion, targeted integration of either homologous or heterologous genes and C-terminal epitope tagging, has made a crucial contribution to advancement of yeast research. These ‘new’ methodologies have greatly facilitated our understanding not only of yeast specific phenomena but also of more general biological questions. These techniques were based upon relatively simple applications of molecular principles underlying homologous recombination between linear fragments and corresponding genomic sequences (Orr-Weaver et al., 1983; Rothstein, 1983; Szostak et al., 1983). During the last 10 years, further new methodological development has been made by use of oligonucleotide- and PCR-based targeting techniques (Baudin et al., 1993; Wach et al., 1997; see more comprehensive references cited therein). This method has allowed individual yeast researchers to accelerate the speed and quality of their work substantially, as time-consuming and sometimes difficult recombinant DNA experiments are no longer needed. Originally PCR methods were developed and used only in the budding yeast, Saccharomyces cerevisiae. This is mainly ascribable to its inherent faithful systems of homologous recombination. Subsequently, Bähler et al. (1998) reported successful application of similar PCR-based gene targeting in the fission yeast, Schizosaccharomyces pombe and provided a number of versatile vectors (Bähler et al., 1998). Since then PCR targeting has been widely used in these two yeast species and also recently started to be applied to other yeasts and fungi.

In PCR targeting methods, the other important factor besides the usage of synthetic oligonucleotides and PCR is introduction of heterologous drug-resistance genes as selection markers upon yeast transformation (Wach et al., 1994). Drug resistance genes display dominant phenotypes, thereby, unlike conventional auxotrophic markers, allowing immediate usage of virtually any strains as recipients without further crossing to introduce appropriate auxotrophic mutations. In addition, complete lack of homologous sequences corresponding to these marker genes in the yeast genome ensures the accuracy of the integrated sites. The most frequently used drug resistance marker is the G418-resistant bacterial gene kan, which encodes aminoglycoside phosphotransferase-3′ (I) (Jimenez and Davies, 1980; Wach et al., 1994). However, since successful application of PCR tagging with the kan marker, it has often become necessary to construct multiple tagged or deleted strains for more sophisticated experiments. This situation has provoked the need for additional dominant markers besides kan. Under this circumstance, again the budding yeast system has been advantageous, as three new drug resistance markers have been introduced (Goldstein and McCusker, 1999). These genes include hph (hygromycin B resistance), nat (nourseothiricin resistance) and pat (bialaphos resistance).

In addition to modern yeast genetics aforementioned, recent technological advancement in yeast cell biology is due largely, if not entirely, to the introduction of the jellyfish Aequorea victoria green fluorescence protein (GFP; Heim et al., 1995; Prasher, 1995). The biggest merit in GFP technology lies in visualization of proteins of interest in live cells. As more knowledge with regard to the subcellular localization of individual proteins is accumulated, more requirement for simultaneous observation of multiple proteins becomes inevitable. For this purpose, variants of GFP, including CFP (cyan fluorescence protein) and YFP (yellow fluorescence protein), have found widespread use. In addition, development of novel fluorescent proteins, such as RFP (red fluorescence protein) from Discosoma coral, has been made (Matz et al., 1999). Despite its potential usefulness, conventional RFP (DsRed and drFP583) (Matz et al., 1999) has some experimental disadvantages, as in order to be functional this fluorescent protein needs to be tetramerized, which is often slow and inefficient. Recently, however, a new derivative, called monomeric RFP (mRFP), has been developed and has provided technical superiority, as this modified RFP works as a fluorescence source in a monomer form (Campbell et al., 2002).

Here we describe construction of DNA modules that enable to use two drug markers in fission yeast, hph and nat, and also introduction of mRFP as a new fluorescent tag in this organism. Furthermore we introduce oligonucleotide-based systems for a convenient one-step marker switch.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgements
  8. References

Bacterial strains, media and nucleic acids preparation

Escherichia coli host strains used were XL1-Blue and DH5α. Standard recombinant DNA methodologies were used (Sambrook et al., 1989). Enzymes were used as recommended by the suppliers (New England Biolabs Ltd., Takara Shuzo Co., Invitrogen Co. and Roche Co.).

Fission yeast strains, media, chemicals and genetic methods

Sz. pombe strains used in this study are listed in Table 1. Liquid cultures or solid agar plates consisting of rich media (YE5S) or synthetic minimal media (EMM2) were used and the standard methods were followed for growth and propagation of yeast strains as described (Moreno et al., 1991). G418 disulphate, hygromycin B and nourseothiricin (clonNAT) were purchased from Sigma Co., Roche Co., and Werner BioAgents, respectively. These drugs were used in solid YE5S plates in concentrations of 100 µg/ml G418, 300 µg/ml hygromycin B and 100 µg/ml clonNAT.

Table 1. Strain list used in this study
StrainsGenotypesDerivations
  1. All the strains listed in this table contain his2-245, leu1-32 and ura4-D18.

513hleu1ura4Our stock
TP108-3Dh+leu1 his2 ura4Our stock
AFAGhase1+-GFP-kan leu1 ura4Our stock
MA003h+alp14+-GFP-kan leu1 his2 ura4Our stock
MS146halp14+-GFP-kan alp7::ura4+leu1 ura4 
 Our stock 
MS875h+alp14+-GFP-hph leu1 his2 ura4This study
MS876h+alp14+-GFP-nat leu1 his2 ura4This study
MS711hase1+-mRFP-kan leu1 ura4This study
MS772h+mis6+-CFP-kan leu1 his2 ura4Our stock
MS803h+mis6+-CFP-hph leu1 his2 ura4This study
MS814h+alp14::hph leu1 his2 ura4This study
MS832halp14::nat leu1 ura4This study
MS887halp7+-GFP-nat leu1 ura4This study
MS888halp7+-13Myc-nat leu1 ura4This study
MS889halp7+-3HA-nat leu1 ura4This study
NK04halp14::kan leu1 ura4Our stock

Construction of plasmids containing new drug markers, hph and nat, for PCR template modules

pAG25 and pAG32, which contain bacterial genes conferring resistance to nourseothiricin (nat, encoding nourseothiricin N-acetyltransferase) and hygromycin B (hph, encoding hygromycin B phosphotransferase), respectively (Goldstein and McCusker, 1999), were obtained from EUROSCARF. Each drug resistance gene was amplified from these template plasmids by PCR with oligonucleotides MD1 and MD2 (Table 2) as primers and inserted into a pCR2.1 vector, using the TOPO TA cloning kit (Invitrogen Co.). Individual subclones were designated pCR2.1-nat and pCR2.1-hph. These new plasmids were used as template DNAs for gene disruption. The 1.2 kb PmeI–BglII nat or 1.6 kb SacI–BglII hph fragments were purified from pCR2.1-nat or pCR2.1-hph, respectively, and subcloned into the same site of the plasmid cassettes pFA6a–GFP(S65T)–kanMX6, pFA6a–13Myc–kanMX6 and pFA6a–3HA–kanMX6 (Bähler et al., 1998), which were digested with the same enzymes to remove the 1.6 kb kan fragment. These new modules were designated pFA6a–GFP(S65T)–natMX6, pFA6a–13Myc–natMX6, pFA6a–3HA–natMX6, pFA6a–GFP(S65T)–hphMX6, pFA6a–13Myc–hphMX6 and pFA6a–3HA–hphMX6 (see Figure 1 for more detailed procedures).

thumbnail image

Figure 1. Construction of new heterologous drug resistance marker cassettes. Construction of new marker plasmids for gene deletion and C-terminal tagging. The new markers, hphMX6 and natMX6, were PCR-amplified from pAG32 and pAG25, respectively, with oligonucleotides MD1 and MD2 as primers (see Table 2 for nucleotide sequences for these oligonucleotides). Amplified fragments were cloned into a TOPO TA cloning vector (pCR2.1, Invitorogen Co). Generated plasmids were named as pCR2.1-hph and pCR2.1-nat. These two plasmids can be used for gene disruption. pCR2.1-nat or pCR2.1-hph was digested with BglII and PmeI or BglII and SacI, and cloned into the same sites of pFA6a–GFP–kanMX6 to replace kanMX6 with natMX6 or hphMX6, yielding pFA–6a–GFP–natMX6 or pFA–6a–GFP-hphMX6, respectively. Plasmids for C-terminal tagging with 13Myc- and 3HA were constructed in a similar manner, designated pFA6a–13Myc–natMX6 (pFA6a–13Myc–hphMX6) and pFA6a–3HA–natMX6 (pFA6a–3HA–hphMX6), respectively. Closed and open arrows indicate 20-nucleotide sequences to which the conventional 100-mer primers anneal (see Table 2) (Bähler et al., 1998)

Download figure to PowerPoint

Table 2. Oligonucleotides used in this study
PrimerSequence 5′ to 3′Comments
  • 1

    Fwd and Rev are long oligonucleotides used for PCR-based gene targeting method (Bähler et al., 1998). 20-mer sequences (italic) correspond to the black and white arrows in Figures 1 and 2.

  • 2

    This primer (Bähler et al., 1998) could be used to check correct integration of fragments containing any of the kan, hph or nat gene.

  • 3

    Underlined sequences indicate the restriction enzyme sites for PmeI (MD2), PacI (MD3) and AscI (MD4).

Fwd1(gene-specific sequence)-CGGATCCCCGGGTTAATTAAGene targeting
Rev1(gene-specific sequence)-GAATTCGAGCTCGTTTAAACGene targeting
− 42–142GCTAGGATACAGTTCTCACATCACATCCGChecking of homologous integration by PCR
MD1CGGATCCCCGGGTTAATTAAGGCGhphMX6 and natMX6 cloning
  Marker switch
MD23GAATTCGAGCTCGTTTAAACACTGGATGGCGGChphMX6 and natMX6 cloning
 GTTAGTATCGMarker switch
MD33GGGTTAATTAAGATGGCCTCCTCCGAGGACGTCmRFP cloning
MD43GTTGGCGCGCCTTAGGCGCCGGTGGAGTGGCGmRFP cloning

Another drug resistance marker, pat (bialaphos) (Goldstein and McCusker, 1999), was not used in this study, as fission yeast cells could still form colonies on standard rich or minimal medium containing bialaphos (800 µg/ml). It is possible that, like budding yeast, special minimal medium containing proline instead of ammonium ion as a nitrogen source might have to be used to see toxicity of this drug in fission yeast, but this point was not pursued further.

Construction of a plasmid template for C-terminal tagging with a monomeric RFP

Plasmid containing the mRFP gene (pRSETB–mRFP) (Campbell et al., 2002) was generously provided by Dr Takashi Morishita (Osaka University, Japan) by permission of Dr Roger Y. Tsien. In order to construct a new vector for C-terminal tagging with mRFP, the mRFP gene was amplified using MD3 and MD4 oligonucleotides (Table 2), and then the 0.7 kb PacI–AscI fragment containing mRFP was further inserted into the same sites of pFA6a–GFP(S65T)–kanMX6, in which the 0.7 kb GFP fragment was removed. This plasmid was designated pFA6a–mRFP–kanMX6. pFA6a–mRFP–natMX6 was then constructed by replacing the 1.6 kb PmeI–BglII kan fragment with the 1.2 kb PmeI–BglII nat fragment. pFA6a–mRFP–hphMX6 was constructed in a similar manner, except that the SacI and BglII sites were used to replace kan with the 1.6 kb hph fragment.

One-step marker switch

Oligonucleotides MD1 and MD2 were used as PCR primers to amplify the hph or nat fragment with pCR2.1-hph or pCR2.1-nat, respectively, as template DNA. These PCR fragments were then directly used as transformation-mediated marker switches. Fragment replacement and integration via homologous recombination occur in the regions of common 5′PTEF (400 bp) and 3′TTEF (200 bp) sequences, which are much longer than 80 bp homologous sequences for long oligonucleotide-based method. Thus, higher efficiency of integration is expected and this is indeed the case (Table 3).

Table 3. Efficiency of marker switch
MethodHostTargeted geneMarkerEfficiency (%)
  1. Marker switch displays a higher efficiency of correct homologous recombination compared to the long oligonucleotide-based gene targeting or deletion method (Bähler et al., 1998). In the top three rows, efficiency shows the number of positive colonies (correct transformants)/total drug-resistant colonies. mis6+ encodes an essential kinetochore protein (Saitoh et al., 1997). Note that, in addition to efficiency, the total number of transformants in the marker switch method is also much higher than that obtained with the long oligonucleotide-based method (bottom three rows).

Marker switch
 alp14+–GFP–kankannat73/103 (71)
 alp14+–GFP–kankanhph36/38 (95)
 mis6+–CFP–kankanhph190/200 (95)
Long oligonucleotides
C-taggingWild-typealp14+kan4/13 (31)
DeletionWild-typealp14+hph2/9 (22)
DeletionWild-typealp14+nat1/4 (25)

PCR amplification of DNA fragments for transformation

DNA fragments for transformation were amplified with PCR using LA-Taq DNA polymerase (Takara Shuzo Co.). PCR reactions were performed in 50 µl volumes that contained 5 µl X10 PCR buffer (supplied by Takara Shuzo Co.), 5 mM MgCl2, 0.8 mM each dNTP, 2 µM each primer, 30 ng template DNA and 1 µl LA-Taq DNA polymerase. For PCR, conditions described previously (Bähler et al., 1998) were followed, 35 cycles of 94 °C for 1 min, 55 °C for 1 min and 72 °C for 3 min followed by a final extension (72 °C for 10 min).

Transformation of Sz. pombe cells and selection of drug-resistant colonies

Yeast transformation was performed according to a protocol based upon lithium acetate method (Bähler et al., 1998; Ito et al., 1983; Keeney and Boeke, 1994). The following are procedures routinely used in our laboratory. 5–10 µg PCR-amplified fragments were used for one transformation. Lithium acetate-treated cells (∼2 × 107 cells) were mixed with PCR amplified DNA fragments, first spread on YE5S agar plates in the absence of drugs and incubated for 24 h at 32 °C or 27 °C. Replica-plating was then performed onto YE5S plates containing drugs for selection. Drug-resistant colonies appeared after 4–5 days of incubation. Correct integration was verified by the colony PCR method. The efficiency of transformation with the long oligonucleotide-mediated method varied depending on the oligonucleotides used; however, on average 1–10 correct integrants were usually obtained (see Table 3 for the exact number of successful integrants).

Request for plasmids

Send plasmid requests to Takashi Toda (fax: 44-20-7269-3258; e-mail: toda@cancer.org.uk). Investigators who plan to use one or more of the plasmids for commercial purposes should state this fact in their requests. For plasmids containing mRFP, a Howard Hughes Medical Institute material transfer agreement (MTA) must be signed. To obtain this document, contact Dr Roger Y. Tsien, Howard Hughes Medical Institute, Cellular and Molecular Medicine, University of California at San Diego, 9500 Gilman Drive, La Jolla, CA 92 093-0647, USA (fax: 1-619-534-5270; e-mail: rtsien@ucsd.edu) and state that you will use monomeric RFP. A copy of the material transfer agreement must be received before the plasmids can be shipped.

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgements
  8. References

Gene disruption and epitope tagging at the C-terminus of proteins with hph and nat drug-resistance markers

DNA fragments containing the new drug-resistance gene encoding hph or nat were cloned into pCR2.1 (Invitrogen Co.) by one-step PCR and these two plasmids were designated pCR2.1-hph or pCR2.1-nat, respectively (Figure 1). 5′ and 3′ regions of these genes in pCR2.1-hph and pCR2.1-nat contained upstream (PTEF) and downstream (TTEF) regulatory sequences, respectively, which were further flanked by common multicloning sites. These flanking sequences were derived from original pFA6a–MX6 plasmids (Bähler et al., 1998; Wach et al., 1994), and therefore oligonucleotides designed for gene deletion (and also C-terminal tagging, see below) by the kan marker could be used directly to amplify hph or nat fragments without designing new oligonucleotides. Successful gene deletions with the hph or nat gene could be checked by PCR using two primer oligonucleotides, in which one primer corresponds to the unique genomic region flanking amplified DNA fragments and the other primer (–42–14; see Table 2 for sequences) anneals with the common upstream region (PTEF) inside the transforming sequence (Bähler et al., 1998).

In order to construct vector modules for C-terminal tagging with new drug resistance markers, the 1.2 kb PmeI–BglII nat fragment or the 1.6 kb SacI–BglII hph fragment was subcloned into vector backbones that were purified from three pFA6a–MX6 plasmids, namely pFA6a–GFP(S65T)–kanMX6, pFA6a–13Myc–kanMX6 and pFA6a–3HA–kanMX6 (see Figure 1 and Methods for detailed construction scheme). The resulting six plasmids were designated pFA6a–GFP(S65T)–nat(or hph)MX6, pFA6a–13Myc–nat(or hph)MX6 and pFA6a–3HA–nat(or hph)MX6, respectively. Having constructed these new vector modules, we performed several gene deletions and C-terminal tagging and succeeded in creating deleted and tagged alleles with nat or hph (e.g. alp7+ and alp14+) (Garcia et al., 2001; Sato et al., 2004) (see Table 3 for the efficiency of integration).

It is of note that we have encountered difficulties in integration of the nat or hph fragment into a genomic locus using long oligonucleotide-based PCR methods when a recipient strain already contained the kan sequence somewhere else in the genome. The results obtained from such transformations are presented in Table 4. In this case, we attempted to disrupt the ura4+ gene (integrated at the alp7 locus) with the natMX6 cassette in a strain that contained the kanMX6 cassette in another genomic locus (alp14+GFP–kan) (MS146, Table 1). Upon transformation, cells were replica-plated onto YES5 plates containing either nourseothiricin (a single selection for nat) or nourseothiricin and G418 (double selection for nat and kan). Although we obtained 93 or 80 drug-resistant colonies, respectively, none of them were auxotrophic for uracil (Table 4), indicating that the natMX6 fragment was not replaced by the ura4+ gene. The same procedure was also performed using hph and hygromycin B, but again we could not obtain correct integrants (Table 4). The exact loci in which PCR fragments were inserted were not pursued in detail further, but in the case of selection by a single drug, a subset (15–30%) of transformants appeared to be integrated via homologous recombination at common 5′PTEF and 3′TTEF sequences found in kanMX6 and natMX6 (or hphMX6), as these transformants were no longer G418-resistant. The application of multiple drug-resistance markers is, therefore, in general useful and efficient, but when we construct strains containing multiple deletions or tags with different drug markers, strains containing each deletion or tag are recommended to be constructed first, and then the desired strains to be made by subsequent genetic crossing.

Table 4. Attempt to integrate the second MX6 cassette
Selection markersNumber of drug-resistant colonies (Ura colonies)
  1. We tried to disrupt the ura4+ gene with nat (or hph) drug-resistance marker cassettes by the long oligonucleotide-mediated method in a strain that had already contained the kan cassette sequence somewhere else in the genome (alp14+–GFP–kan, MS146, Table 1). Transformed cells were first selected on plates containing either nourseothiricin (or hygromycin B) alone or nourseothiricin (or hygromycin B) and G418 together. Drug-resistant colonies were then replica-plated onto minimal plates lacking uracil to verify correct replacement of the ura4+ gene with the nat (or hph) drug-resistance gene. As shown in this table, none of transformants is auxotrophic for uracil, indicating that homologous recombination between amplified drug resistance fragments and the ura4+ sequence did not occur.

Nourseothiricin93 (0)
Nourseothiricin, G41880 (0)
Hygromycin B27 (0)
Hygromycin B, G41821 (0)

C-terminal tagging with a monomeric RFP

RFP has been previously used in fission yeast study, but only conventional RFP (DsRed) was tested (Grallert et al., 2004; Yaffe et al., 2003). DsRed has some practical disadvantages attributable to its slow and incomplete maturation and obligate tetramerization (Baird et al., 2000; Grallert et al., 2004; Pereira et al., 2001). Recently we successfully introduced a monomeric RFP (mRFP) to visualize in vivo microtubule dynamics in fission yeast, in which N-terminal tagged mRFP-α2-tubulin (mRFP-Atb2) that is ectopically expressed under the thiamine-repressible nmtP3 promoter was used (Maundrell, 1990; Yamashita et al., 2005). In order to make mRFP tagging feasible for more general usage, we now constructed pFA6a–mRFP–kanMX6, pFA6a–mRFP–natMX6 and pFA6a–mRFP–hphMX6 (the scheme depicted in Figure 2A), by which tagging of mRFP to the C-terminus of proteins of interest could be performed. As this new modules contain the same flanking sequences as those found in other pFA6a–kanMX6 vectors, long oligonucleotides designed for these modules (Bähler et al., 1998) could be used directly.

thumbnail image

Figure 2. Construction of a vector for mRFP tagging and its usage for visualization of protein localization. (A) Plasmid construction for C-terminal tagging with a monomeric RFP (mRFP). The entire ORF encoding mRFP including stop codon (asterisks) was amplified by PCR, using pRSETB-mRFP as a template, and oligonucleotides MD3 and MD4 (Table 2) as primers. Amplified fragments were digested with PacI and AscI, and then cloned into the same sites of pFA6a–GFP–kanMX6 to replace GFP with mRFP. The resulting plasmid was designated pFA6a–mRFP–kanMX6. pFA6a-mRFP–natMX6 and pFA6a–mRFP–hphMX6 were then constructed from pFA6a–mRFP–kanMX6 by replacing the kan fragment with the nat and hph fragment, respectively. (B) An example of intracellular localization of mRFP-tagged Ase1 (Ase1–mRFP). An ase1+–mRFP–kan strain (MS711, Table 1) was obtained with PCR-tagging using pFA6a–mRFP–kanMX6 as a template. For comparison, Ase1 localization using an Ase1–GFP strain (AFAG, Table 1) (Yamashita et al., 2005) was also shown. The bar indicates 10 µm

Download figure to PowerPoint

In order to validate the feasibility of these new vector modules, we created several strains that produce C-terminally mRFP-tagged proteins. One such example is shown in Figure 2B. In this case, Ase1, which is a conserved microtubule-bundling factor (Yamashita et al., 2005), was tagged with mRFP. For comparison, fluorescent images from Ase1–GFP and Ase1–mRFP are shown (Figure 2B). It was evident that the localization patterns of Ase1 tagged with GFP and mRFP are very similar, indicating that mRFP is of great use in fission yeast systems.

Marker switch

In addition to new modules containing additional drug markers for gene disruption and C-terminal tagging described earlier, we designed short oligonucleotides that could be directly used as marker switches with already existing strains, which is also mentioned in budding yeast (Goldstein and McCusker, 1999). A scheme is depicted in Figure 3A, in which kan is replaced by hph or nat (MD1 and MD2; sequences of these oligonucleotides are shown in Table 2). These oligonucleotides are ‘universal’, i.e. by simply using different templates, the same oligonucleotides could be used for marker swaps between three drug-resistance genes (kan, nat and hph).

thumbnail image

Figure 3. One-step marker switch. (A) Marker switch from kan to hph or nat. Either hphMX6 or natMX6 fragment can be substituted for the kanMX6 sequence, which is integrated in the fission yeast chromosome, by homologous recombination between homologous sequences (PTEF and TTEF). An example of marker switch from alp14+–GFP–kan to alp14+–GFP–hph or alp14+–GFP–nat is shown. hphMX6 and natMX6 fragments can be easily amplified by PCR method with short primers MD1 and MD2 (Table 2). (B) Drug resistance of marker-switched transformants. A host alp14+–GFP–kanR strain (NK04, Table 1) was transformed with hphMX6 or natMX6 cassettes, as shown in (A). Transformants (MS814 and MS832) were streaked on rich YE5S plates containing G418 (left), hygromycin B (middle) or nourseothricin (right), and incubated at 30 °C for 2 days

Download figure to PowerPoint

Using MD1 and MD2 oligonucleotides, we tested the practicability of marker switch method with several strains and obtained a high efficiency of successful conversion of drug-resistance markers. One example of a marker switch from alp14+GFP–kan to alp14+–GFP–hph or alp14+GFP–nat is shown in Figure 3B. In one case, by using a G418-resistant strain (alp14+–GFP–kan) as a host strain, more than 90% (36/38) of hygromycin B-resistant transformants were sensitive to G418 (Table 3), implying a successful marker switch and, in fact, a correct replacement of drug markers was subsequently confirmed by the colony PCR method. In contrast, in the case of PCR-tagging methods using long oligonucleotides (Bähler et al., 1998), the rate of correct integration varied between 22% and 31% (Table 3). It is of note that the number of positive transformants was also much higher in the marker switch method than in the PCR- and long oligonucleotide-mediated methods. The marker switch method could be applicable not only to gene deletion or C-terminal tagging but also to kan- and nmt promoter-based N-terminal tagging (Bähler et al., 1998; Maundrell, 1990).

The merits of the marker switch method are three-fold. First, it is easy. Sometimes amplification of PCR fragments with long oligonucleotides as templates is troublesome, while PCR fragments for marker switch can be reproducibly amplified in a large quantity from short oligonucleotides. Second, it has higher efficiency, as shown here. Third, it is economical. Common short oligonucleotides are much more economical than long oligonucleotides, which are more expensive and have to be designed and ordered for each construction. Therefore, if strains that contain either a gene deletion or any tag carrying one drug marker already exist, it would be recommended to follow the marker-switch method instead of trying to create strains with other drug markers using new long oligonucleotides. A similar marker-switch approach was reported previously between auxotrophic markers (MacIver et al., 2003), but the method described here is more efficient and faster.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgements
  8. References

In combination with previously constructed modules (Bähler et al., 1998), new versatile vectors described in this work will enable fission yeast work to be easier and faster. Multiple tagged or deleted strains could be constructed and either fixed or time-lapse live imaging analysis of multiple proteins is possible.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgements
  8. References

We thank Jürg Bähler and Takashi Morishita for providing materials used in this study and Jun-ichi Nakayama for information about hph. Special thanks are extended to Roger Y. Tsien for generous permission to use the mRFP-containing plasmid. We thank Karen Crawley for critical reading of the manuscript and useful suggestions. This work was supported by Cancer Research UK.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and discussion
  6. Conclusions
  7. Acknowledgements
  8. References