|Home | About | Journals | Submit | Contact Us | Français|
We have designed the most efficient strategy to knock out genes in fission yeast Schizosaccharomyces pombe on a large scale. Our technique is based on knockout constructs that contain regions homologous to the target gene cloned into vectors carrying dominant drug-resistance markers. Most of the steps are carried out in a 96-well format, allowing simultaneous deletion of 96 genes in one batch. Based on our knockout technique, we designed a strategy for cloning knockout constructs for all predicted fission yeast genes, which is available in a form of a searchable database http://mendel.imp.ac.at/Pombe_deletion. We validated this technique in a screen where we identified novel genes required for chromosome segregation during meiosis. Here, we present our protocol with detailed instructions. Using this protocol, one person can knock out 96 S. pombe genes in 8 days.
Large-scale knockout screens in fission yeast have been hindered so far by lack of an efficient knockout technique. A technique using long oligonucleotides (80 bp homology to the target sequence)1 was used in a pilot gene deletion project. As many as 20 out of 85 selected genes could not be deleted in a diploid strain2. Using the same approach, Martin-Castellanos et al.3 deleted 160 out of 184 selected meiotically upregulated genes. Recently, Korean BIONEER Corporation embarked on knocking out all fission yeast genes (http://pombe.bioneer.co.kr) using a long-oligonucleotide strategy. Although the long-oligonucleotide strategy1 has been widely used to knock out and modify genes in fission yeast, the low efficiency of this strategy suggests that it may not be optimal for large-scale screens.
We tested and compared different knockout strategies and decided that the most efficient strategy was to use constructs containing left and right homology regions 150–700 bp long cloned into a vector (Fig. 1). The vector was carrying dominant drug resistance markers conferring resistance to either nourseothricin or hygromycin B4-6. We chose the nourseothricin and hygromycin B resistance genes because these have so far not been widely used in fission yeast, which will facilitate the generation of deletions in most of the existing fission yeast strains. Most of the steps are carried out in a 96-well format, allowing simultaneous deletion of 96 genes in one batch.
We validated this technique in our initial screen where we knocked out 180 uncharacterized genes and identified proteins required for meiotic chromosome segregation, most notably the protector of centromeric cohesion Sgo1 (ref. 4). In this screen, we were able to knock out 180 genes out of 192 selected. The remaining genes, which resisted deletion, may be essential genes or possibly genes that are refractory to targeting by homologous recombination owing to a repressed state during vegetative growth7. Besides its superior knockout efficiency, our strategy has the added advantage that a library of knockout plasmids is created.
We further designed a strategy for cloning knockout constructs for all predicted fission yeast genes that is available in a form of a searchable database (http://mendel.imp.ac.at/Pombe_deletion). Apart from the cloning strategy, this webpage includes gel previews indicating mobility of the PCR-amplified homology regions, ligation products and checking PCR products as well as basic information about the protein, DNA sequence of the target gene and link to S. pombe genome database (GeneDB). Web-based tool that automatically suggests primer sequences for deletion, tagging and regulatable expression of S. pombe genes using long oligonucleotides is now also available8.
Apart from the fission yeast, this knockout technique should also be applicable to other yeast species (including pathogenic yeasts) where homologous recombination can be used to knock out genes.
For checking of bacterial inserts, two primers are required: upch-uni located approximately 90 bp upstream of the MCS and dwch-uni located approximately 60 bp downstream of the MCS. For checking of yeast transformants, the same primers are used, but there are two alternative primers, upch-uni2 located approximately 150 bp upstream of the MCS and dwch-uni2 located approximately 94 bp downstream of the MCS. upch-uni: GTCGTTAGAACGCGGCTACA; dwch-uni: TCTGGGCCTCCATGTCGCTGG; upch-uni2: GGCTGGCTTAACTATGCGGC; dwch-uni2: GCTGCGCACGTCA AGACTGTC.
Prepare a midi-prep (Qiagen) of the cloning vector (pCloneNat1 or pCloneHyg1).
Prepare 0.1 M lithium acetate in 1× Tris-EDTA (TE), pH 7.5.
Prepare 0.1 M lithium acetate and 40% (wt/vol) PEG 3350 in 1 × TE, pH 7.5
Make transformation-competent Escherichia coli DH5α using CaCl2 method. Alternatively, use commercially available competent E. coli cells.
Prepare standard 2× TY media for E. coli. For selection, add ampicillin (100 μg ml−1). For yeast cultivation, prepare standard YES medium supplemented with 0.15 g l−1 adenine and 0.1 g l−1 uracil, l-histidine, l-lysine and l-leucine. For selection, add clonat at 100 μgml−1 and hygromycin B at 200 μg ml−1.
Dissolve oligonucleotide primers in TE buffer to 100 μM concentration. To knock out 96 genes (one batch), you will need six 96-well plates with primers (upout, upin, dwout, dwin, upch, dwch).
Two 20-mer oligonucleotides, four 30-mer oligos, standard synthesis, desalted; 460 μl PCR reaction; 8.5 μlrestriction enzymes; 4 μl T4 DNA ligase; four DNA purification columns; competent bacteria and yeast cells plus reagents for transforming and growth media.
▲ CRITICAL STEP Do not rush with preparing the vector. Vector that is not digested to completion and not prepared properly may result in high numbers of negative E. coli transformants containing empty vector without homology regions.
■ PAUSE POINT Digested PCR products can be stored for months at −20 °C.
▲ CRITICAL STEP Exposure to UV light can damage DNA. Work quickly and minimize the UV exposure.
■ PAUSE POINT Digested PCR products can be stored for months at −20 °C.
▲ CRITICAL STEP Checking of the PCR product by Enz1 digestion is necessary. We observed and confirmed by sequencing that some of the ligated homology regions contained altered Enz1 recognition sequencethatcould notbecleaved by theEnz1.
■ PAUSE POINT E. coli transformants can be stored for years at −80 °C. Plasmid DNA can be stored for years at −80 °C.
■ PAUSE POINT The transformants can be stored at −80 °C for years.
Steps 1–10 (day one)
Steps 11–18 (day two)
Steps 19–24 (day three)
Steps 25–30 (day four)
Steps 31–32 (day seven)
Steps 33–41 (day eight)
Our suggestion is to start on Tuesday morning, transform yeast on Thursday and plate yeast transformants on Friday. Saturday and Sunday are free. On Monday, you should see colonies of yeast transformants on selective plates. Identify positive clones on Tuesday.
Troubleshooting advice can be found in Table 1.
Using this protocol, one person can knock out 96 S. pombe genes in 8 days. It is also possible that one person can do two batches (192 knockouts) simultaneously; however, if more than 192 knockouts are required, additional help is needed. We successfully used this protocol to knock out functionally uncharacterized genes whose expression is upregulated during meiosis9. In this screen, we were able to knock out 180 genes out of 192 selected.
Most of the steps are carried out in a 96-well format. However, single tubes are used for extraction of DNA from agarose gel and for plasmid purification. Recently, 96-well mini-prep kits became available (e.g., from Qiagen). In the future, gel extraction kits compatible with the 96-well system may also become available. Using these kits will eliminate single tubes and may speed up the whole procedure. This protocol can also be scaled down to knock out a single gene. We routinely use this protocol in our lab to knock out single genes. A diploid strain can be used to knock out essential genes. Knockout constructs can be verified by sequencing, which makes the whole procedure safe.
Knocking out your favorite gene may affect neighboring genes if they are in close proximity to the target gene. If this is the case, new primers should be designed to move the regions of homology. We have not considered this criterion when designing knockout constructs for all S. pombe genes (http://mendel.imp.ac.at/Pombe_deletion). To exclude the effect of neighboring genes on the knockout phenotype, we suggest to test complementation of the knockout phenotype by introducing the wild-type allele of the target gene.
This work was supported by Boehringer Ingelheim and partly by Austrian Industrial Research Promotion Fund (FFF) and Austrian Science Fund (FWF) grants. We thank Georg Dietzl and Barry Dickson (IMP, Vienna) for help with setting up the cloning protocol and Stephen Kearsey (University of Oxford, UK) for helpful suggestions and reagents.
COMPETING INTERESTS STATEMENT The authors declare that they have no competing financial interests.
Rights and permissions information is available online at http://npg.nature.com/reprintsandpermissions