Simple CRISPR-Cas9 Genome Editing in S. cerevisiae

Abstract

CRISPR-Cas9 has emerged as a powerful method for editing the genome in a wide variety of species, since it can generate a specific DNA break when targeted by the Cas9-bound guide RNA. In yeast, Cas9-targeted DNA breaks are used to promote homologous recombination with a mutagenic template DNA, in order to rapidly generate genome edits (i.e., DNA substitutions, insertions, or deletions) encoded in the template DNA. Since repeated Cas9-induced DNA breaks select against unedited cells, Cas9 can be used to generate marker-free genome edits. Here we describe a simple protocol for constructing Cas9-expressing plasmids containing a user-designed guide RNA, as well as protocols for using these plasmids for efficient genome editing in yeast.

INTRODUCTION

Clustered regularly interspaced short palindromic repeats (CRISPR) associated enzymes have proven to be powerful tools for genome editing in a wide variety of species (Adli, 2018; Hsu, Lander, & Zhang, 2014; Sternberg & Doudna, 2015), including the budding yeast Saccharomyces cerevisiae (DiCarlo et al., 2013; Giersch & Finnigan, 2017; Stovicek, Holkenbrink, & Borodina, 2017). The most commonly used CRISPR editing system relies on the CRISPR-associated protein-9 (Cas9) endonuclease from Streptococcus pyogenes, which generates a double-stranded DNA break at DNA sites that match the Cas9-bound CRISPR RNA. Typically, this is expressed as a chimeric single guide RNA (sgRNA), containing a 20-nucleotide (nt) guide sequence and the structural portion that is important for Cas9 binding (Jinek et al., 2012). Co-expression of Cas9 and the sgRNA in yeast results in cleavage at the sgRNA target site, as long as there is a protospacer-adjacent motif (PAM) neighboring the 3’ side of the target site. S. pyogenes Cas9 requires an 5’-NGG-3’ PAM motif to cleave the target site and initiate genome editing. In mammalian cells, random mutations are frequently generated at Cas9 target sites by the error-prone non-homologous end joining (NHEJ); however, such mutagenic events are infrequent during Cas9-editing in S. cerevisiae (DiCarlo et al., 2013). In yeast, Cas9-induced breaks are frequently used to promote homologous recombination with a co-transformed template DNA, typically a long oligonucleotide or PCR product. The DNA template contains mutations to introduce the desired genome edit and destroy the Cas9 target site (typically by mutating the PAM), in order to prevent further Cas9-induced DNA breaks. Hence, Cas9 can efficiently induce genome edits in yeast without the need of a selectable marker (DiCarlo et al., 2013), since repeated Cas9-induced DNA breaks selects against the growth of unedited cells containing an intact DNA target site.

Typically, the rate-limiting step for Cas9 genome editing in yeast is designing and cloning the user-designed 20-nt guide RNA into a yeast sgRNA expression vector for subsequent transformation and genome editing. Here we describe a simple and rapid method for performing directional cloning of user-designed oligonucleotides containing the 20-nt guide into sgRNA expression vectors (Basic Protocol 1), containing either a URA3 (pML104) or LEU2 (pML107) marker gene (Figure 1). Since these vectors also express active Cas9 enzyme (Laughery et al., 2015), we further describe how the constructed guide-RNA expressing vectors can be co-transformed with a repair template (Basic Protocol 2) to induce the desired genome edit (Basic Protocol 3). Successful genome editing is typically confirmed by PCR amplification of the edited locus and sequencing (Basic Protocol 4). Following successful genome editing, the Cas9/sgRNA-expressing vector can be removed by a period of non-selective growth and replica plating (pML107) or by counter-selection on 5-fluoroorotic acid (5-FOA) containing plates (pML104; see Basic Protocol 5).

Figure 1:

Plasmid maps of the sgRNA/Cas9 expression vectors pML104 (URA3 marker) and pML107 (LEU2 marker). Both plasmids are yeast/E. coli shuttle vectors containing an ampicillin resistance (AmpR) marker, and a yeast 2 micron (2μ) origin of replication. Both vectors contain a Cas9 expression cassette and an sgRNA expression cassette, which contains unique BclI and SwaI restriction sites, to facilitate directional cloning of the user-designed guide sequence into the sgRNA expression cassette.

BASIC PROTOCOL 1

CONSTRUCTING THE GUIDE RNA EXPRESSION VECTOR

This protocol utilizes the guide RNA expression vectors pML104 or pML107 (see vector maps in Figure 1) to facilitate rapid cloning of user-designed guide RNA sequences (Laughery et al., 2015). The protocol describes the design of oligonucleotides containing the desired guide RNA sequence, the preparation of a stock of restriction enzyme-digested and gel purified vector backbone, and the ligation of hybridized DNA oligonucleotides containing the guide RNA sequence into the digested vector. The experimental strategy for guide RNA cloning is outlined in Figure 2.

Figure 2:

Experimental strategy for cloning user-designed guide sequence into sgRNA expression cassette in pML104 or pML107. The BclI site is located at the 3’ end of the pSNR52 promoter, which is used to drive sgRNA expression in yeast. The SwaI site is located in the structural portion of the sgRNA. SwaI and BclI cleavage of the pML104 or pML107 sgRNA expression cassette results in a cut vector ready for ligation of the user designed guide sequence. The user-designed guide sequence contains a 20nt DNA region encoding the guide (for targeting Cas9; shown as N’s), a 5’ structural portion of the sgRNA, and a 5’ GATC overhang on one strand to facilitate ligation into the cleaved sgRNA expression cassette. Correct ligation can be validated by DNA sequencing using the neighboring T3 primer site. This figure is adapted from (Laughery et al., 2015).

Materials

Yeast/E. coli shuttle vectors containing both a guide RNA expression cassette and S. pyogenes Cas9: i.e., pML104 (Addgene #67638), pML107 (Addgene #67639)

dam- strain of E. coli: i.e., New England Biolabs (NEB #C295I)

Restriction enzymes SwaI (NEB #R0604S) and BclI (NEB #R0160S)

10x buffer compatible with both SwaI and BclI enzymes (i.e., NEB buffer 3.1)

PCR thermal cycler

Primers specific for user-designed guide RNA (100 pmol/μL)

T4 DNA Ligase (NEB #M0202S)

T4 DNA Ligase buffer (NEB)

Competent E. coli cells (e.g., NEB #C2987I or similar)

LB plates with 100 μg/ml ampicillin

Plasmid Miniprep kit (i.e., Zymo Research D4036 or similar)

T3 primer ( GCAATTAACCCTCACTAAAGG)

LiAc-TE buffer (see recipe)

Design oligonucleotides containing guide RNA sequence

• 1
  Go to the CRISPR-Cas9 primer design website (http://wyrickbioinfo2.smb.wsu.edu/crispr.html) to design guide RNAs targeting a user-specified yeast gene. Use this or a similar guide RNA design tool to identify a guide RNA sequence with a PAM motif within 10–15 nucleotides of the intended gene edit.

  The ideal scenario is where the intended gene edit overlaps with the PAM motif of the chosen guide RNA target. In this case, a single DNA sequence change can simultaneously introduce the intended gene edit and disrupt the PAM motif. However, in most cases, at least two DNA sequence changes will be needed to introduce the gene edit and disrupt the PAM motif. Figure 3 shows an example of a guide RNA designed to target the RNR1 gene in yeast, in order to create the rnr1-D57N mutation.

  In selecting an appropriate guide RNA, try to avoid guide RNAs containing long T runs, particularly near the 3’ end of the guide RNA (i.e., near the ‘GTTTT’ segment of the sgRNA; see Figure 1). Long T runs in the sgRNA sequence can lead to premature termination of the RNA polymerase III-encoded sgRNA transcript, resulting in poor Cas9 editing.

• 2
  Order DNA oligonucleotides representing both strands of the guide RNA sequence. The oligonucleotides should have the following DNA sequences:

  • Top oligo: 5’-GATC(N)₂₀GTTTTAGAGCTAG-3’
  • Bottom oligo: 5’-CTAGCTCTAAAAC(N)₂₀-3’

  The (N)₂₀ segment in the oligonucleotides above indicates the user-designed 20-nt guide RNA segment (or its reverse complement in the bottom oligo). The 5’GATC overhang on the top strand oligonucleotide facilitates directional cloning into the pML104 or pML107 vectors. The GTTTTAGAGCTAG/CTAGCTCTAAAAC sequences contain a segment of the structural portion of the sgRNA that needs to be including when cloning into the pML104/pML107 vectors. Note, if the guide RNA was designed using our CRISPR-Cas9 primer design website (http://wyrickbioinfo2.smb.wsu.edu/crispr.html), then these sequence features will already be included in the oligonucleotide sequences designed by the website.

  An example of Top and Bottom oligonucleotide sequences for targeting the RNR1 gene are shown in Figure 3.

  The guide RNA oligonucleotides should not be 5’ phosphorylated, in order to avoid ligating a concatemer of the guide RNA sequence.

Figure 3:

Example of sgRNA and repair template oligonucleotide design to construct, by way of example, the rnr1-D57N point mutation in yeast. (A) Experimental strategy showing location of Cas9 guide target and its adjacent PAM. The guide target and PAM motif are on the opposite strand (i.e., bottom strand). The mutations needed to create the rnr1-D57N mutant and simultaneously disrupt the PAM are highlighted in blue. (B) Oligonucleotides for cloning the RNR1 guide target into the pML104 or pML107 sgRNA/Cas9 expression vectors. The 20mer guide RNA sequence is highlighted in bold. (C) Oligonucleotides representing the repair template, with the mutations to be introduced highlighted in blue.

Prepare restriction enzyme-digested sgRNA expression vector

• 3
  Transform sgRNA expression vector (e.g., pML104 or pML107) into a dam- E. coli strain (e.g., NEB #C2925I; see Current Protocols article; Seidman et al., 1997) and isolate plasmid DNA from dam- E. coli strain using a miniprep kit.

  BclI will not cut Dam methylated DNA, necessitating this step. However, continual passaging of the vector in dam- strains can be mutagenic, so it is recommended to keep frozen permanent stocks of the plasmid in both dam- and dam+ strains.

• 4
  Perform an overnight digestion (at least 12 hrs) of the plasmid with SwaI (NEB Buffer 3.1) in a thermocycler set to 25°C. A typical reaction mix is as follows:

  • 8 μL 10x Buffer 3.1 (NEB)
  • 19 μL H₂O
  • 50 μL pML104 vector (~200 ng/μL for a total of 10 μg)
  • 3 μL SwaI (10 units/μL)

  BclI does not heat inactivate as reliably as SwaI, so be sure to perform the restriction enzyme digestion in the prescribed order. Be sure to use a buffer that is compatible with both enzymes (e.g., NEB Buffer 3.1), otherwise it will be necessary to clean up the digestion reaction by ethanol precipitation or DNA cleanup kit before moving on to the next restriction enzyme digestion step.

• 5
  Heat inactivate SwaI by incubating at 65°C for 20 minutes, chill briefly (to avoid damaging the following enzyme), and add BclI to the reaction. Incubate at 50°C for 2–4 hours. For the reaction mix described in step 4 above, one would add 3 μL of BclI (10 units/μL).

  The creation of a clean, well-digested plasmid stock is critical to the success of constructing sgRNA expression vectors. If desired, a small amount of additional restriction enzyme (e.g., 1 μL) can be added into each reaction in the last 0.5–1 hour of its incubation period to ensure complete digestion.

• 6
  Purify the cut vector by agarose gel electrophoresis and recover the band for the cleaved vector from the gel using a kit (e.g., Zymoclean DNA gel recovery kit or similar). Determine the concentration of purified DNA by measuring the absorbance at 260 nm (A₂₆₀).

  When cutting out DNA bands from the gel, we advise the use of a blue light transilluminator (e.g., Dark Reader from Clare Chemical Research or similar) to allow visualization of the DNA without damaging it with UV radiation. This significantly enhances cloning efficiency in subsequent steps.

Hybridize top and bottom guide RNA oligonucleotides and clone into sgRNA expression vector

• 7
  Assemble the guide RNA oligonucleotides into a 100 μL hybridization mix consisting of the following components:

  • 3 μL of top guide RNA oligonucleotide (100 pmol/μL)
  • 3 μL of bottom guide RNA oligonucleotide (100 pmol/μL)
  • 10 μL 10x T4 Ligase buffer
  • 84 μL H₂O

  T4 Ligase buffer is used here since it is compatible with the ligation reaction in step 9, and the Mg²⁺ ions present in the ligase buffer help facilitate hybridization of the DNA oligonucleotides.

• 8
  Hybridize oligonucleotides in the mix in a thermocycler using the following cycling conditions:

  • 95°C 5 min
  • 95°C 1 min
  • Decrease 1°C per cycle for 70 cycles
  • Hold at 4°C

  The hybridized oligonucleotides can be used immediately for the ligation reaction or stored at −20°C.

• 9
  Ligate hybridized oligonucleotides from step 8 into the SwaI-BclI digested vector (see step 6) by assembling a 10μL reaction with the following components:

  • ≥10ng SwaI-BclI digested pML104 or pML107 vector (from step 6)
  • 1 μL hybridized oligonucleotides (from step 8)
  • 0.9 μL 10x T4 ligase buffer
  • 1 μL T4 ligase
  • X μL H₂O (add enough H₂O so that the final volume is 10 μL).
• 10
  Incubate ligation reaction overnight at 16°C (in thermocycler or water bath), then transform the ligation into E. coli and select for transformants by spreading on LB-ampicillin plates.

  The use of competent cells with an efficiency of 10⁸-10⁹ transformants/μg plasmid is recommended to ensure a sufficient number of transformants.

  We recommend performing a control ligation reaction and transformation by omitting the hybridized oligonucleotides in the ligation reaction (step 9), in order to assess the prevalence of unwanted backbone ligation events.

• 11
  Isolate plasmid DNA from transformants using miniprep kit. Verify correct ligation of sgRNA oligonucleotides into the pML104 (or pML107) vector by sequencing with the T3 primer.

  It is recommended that plasmid DNA from 3–6 candidate colonies are isolated and sequenced regardless of the observed frequency of background ligation events (i.e., the number of colonies on the ‘no oligonucleotide’ control ligation).

BASIC PROTOCOL 2

PREPARING DOUBLE-STRANDED OLIGONUCLEOTIDE REPAIR TEMPLATE

This protocol describes how to design and prepare the repair template for co-transformation with the Cas9/sgRNA expressing vector. Typically, long oligonucleotides (≥90-nt) are used as repair templates in yeast genome editing, which are described in this protocol, although PCR products can also be used. This protocol assumes that a double-stranded oligonucleotide will be used as the repair template, although genome editing can also be performed using single-stranded oligonucleotides (see Alternate Protocol 1), albeit with a somewhat lower editing efficiency.

Materials

Template oligonucleotides containing the desired genome edit/mutations

PCR thermal cycler

LiAc-TE buffer (see recipe)

Design oligonucleotides containing the desired genome edit

• 1
  Design and order complementary template oligonucleotides (i.e., both top and bottom DNA strands) containing the intended gene edit/mutation. The template oligonucleotides should contain mutations that disrupt the PAM motif (i.e., 5’NGG’3 sequence) associated with the guide RNA target and introduce the intended gene edit. Typically, the template oligonucleotides are at least 90 nts in length and have at least 30–40 nts of homologous flanking DNA on both sides of the intended gene edit/mutation. We also recommend having 30–40 nts of homologous flanking DNA on both sides of the Cas9-induced DNA break (typically 3–4 nts on the 5’ side of the PAM within the guide RNA target) in order to facilitate efficient homologous recombination with the targeted gene.

  The ideal scenario is when a single DNA sequence change will simultaneously introduce the intended gene edit (i.e., insertion, deletion, or substitution) and disrupt the PAM motif so that Cas9 cleavage is abrogated. Often this is not possible, so frequently two DNA sequence changes are needed, in order to both introduce the desired gene edit and disrupt the PAM motif. In this case, the PAM mutation should be a synonymous mutation so that it does not affect the encoded protein sequence. Moreover, it is important that the two DNA sequence changes (i.e., the PAM mutation and gene edit) are nearby (i.e., less than 20 nucleotides apart), as editing efficiency will likely decrease with increasing distance between the PAM mutation and gene edit (Garst et al., 2017). See Figure 3 for an example of designing the oligonucleotide repair template.

  While disrupting the PAM motif is the most common way of destroying the guide RNA target site in the edited cells, an alternative approach is to introduce multiple DNA sequence changes in the guide RNA target itself in order to inhibit Cas9 targeting and cleavage. This approach has been previously used in large-scale CRISPR screens in yeast (Roy et al., 2018).

Hybridize oligonucleotides to make a double stranded repair template

• 2Re-suspend each oligonucleotide to a final 1 nmol/μL in LiAc-TE buffer, then add 8 μL of each oligonucleotide to a PCR tube, for a total volume of 16 μL.
• 3
  Hybridize oligonucleotides in a thermocycler using the following cycling conditions:

  • 95°C 5 min
  • 95°C 1 min
  • Decrease 1°C per cycle for 70 cycles
  • Hold at 4°C

  The hybridized oligonucleotides can be used immediately for transformation (see Basic Protocol 3) or stored at −20°C.

ALTERNATE PROTOCOL 1

PREPARING A SINGLE-STRANDED OLIGONUCLEOTIDE REPAIR TEMPLATE

Using a single-stranded oligonucleotide repair template typically yields fewer transformants than a double-stranded repair template (Basic Protocol 2). However, the percentage of isolates positive for the desired mutation is nearly identical, so both methods are recommended.

1. Design a single oligonucleotide containing the desired gene edit/mutation, as described in Basic Protocol 2, step 1.

   When using a single-stranded repair template, either DNA strand can be used, although we often use the non-target DNA strand (i.e., the strand that will not hybridize to the guide RNA) for the repair template.

2. Re-suspend the oligonucleotide to a final concentration of 1 nmol/μL in LiAc-TE buffer

BASIC PROTOCOL 3

INDUCE GENOME EDITING BY CO-TRANSFORMATION OF YEAST

This protocol describes how to induce marker-free genome editing in yeast by co-transforming the sgRNA/Cas9 expression plasmid (Basic Protocol 1) and the repair template (Basic Protocol 2) into yeast. Since this essentially involves transforming yeast with a plasmid, a quick, low efficiency transformation method is utilized in this step, which is adapted from (Chen, Yang, & Kuo, 1992). Higher efficiency yeast transformation methods can also be used, but they are typically not necessary.

Materials

One Step buffer (see recipe)

Cloned sgRNA/Cas9 expression prepared in Basic Protocol 1 (~50–250 ng/μL)

Parent vector lacking guide RNA (i.e. pML104 or pML107), for control transformation reactions (~50–250 ng/μL)

Single or double-stranded DNA repair template prepared in Basic Protocol 2

Salmon Sperm DNA (10 μg/μL)

Yeast strain (e.g., BY4741 [MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0] ATCC #201388)

Yeast extract peptone dextrose (YPD) liquid medium (see Current Protocols article; Treco and Lundblad, 1993)

Plates containing rich yeast extract peptone dextrose (YPD) medium

Plates containing selective yeast medium (SC-URA or SC-LEU)

Microcentrifuge

30°C incubator and shaker/rotator

Preparing yeast culture for transformation

• 1Streak yeast strain to YPD plates and let grow at 30°C for 2–3 days until colonies are formed.
• 2
  Start overnight culture by inoculating 2–3 mL YPD media with a single colony in 10 mL test tube.

  Aim to grow the culture about 12–18 hours until it is in late log or early stationary phase.

One step transformation of yeast with sgRNA/Cas9 plasmid and repair template

• 3
  Aliquot 125 μL of overnight yeast culture to a separate tube for each transformation reaction. Pellet the cells with by centrifuging at 11,000 – 13,000 rpm for 30 seconds in a table top microcentrifuge and remove the supernatant (i.e. YPD media).

  Preparing three additional tubes for control experiments is recommended for troubleshooting. This protocol can easily be scaled up to accommodate a larger cell volume (e.g. 250 – 500 μL culture).

• 4
  Re-suspend each pellet with 25 μL One Step Buffer by pipetting/vortexing.

  Vortexing is not recommended in the next step so do your best to get the pellet thoroughly re-suspended at this time.

• 5
  Add the following additional ingredients to each tube and mix by pipetting:

  • 1.25 μL salmon sperm DNA (10 μg/μL)
  • 1 μL sgRNA/Cas9 plasmid (approximately 50–250 ng; see Basic Protocol 1)
  • 1 μL of 1:10 dilution of oligonucleotide template DNA (see Basic Protocol 2)

  To avoid shearing the carrier (salmon sperm) DNA, do not vortex the reaction at this step. Omit sgRNA/Cas9 plasmid in the first control reaction (no vector control), omit the repair template in the second control reaction (no repair template control), and substitute pML104 (or pML107) for the sgRNA/Cas9 expression plasmid in the third control reaction (empty vector control).

• 6Incubate tubes for 30 minutes in a water bath set to 45°C, then plate the entire reaction on selective medium (e.g., SC-URA for pML104 or SC-LEU for pML107).
• 7
  Incubate plates for 2–3 days at 30°C until yeast colonies appear.

  The number of colonies arising after the transformation of the empty vector control (i.e., pML104 or pML107) indicates the efficiency of the transformation experiment and should yield a relatively high number (e.g., hundreds to thousands) of colonies. No colonies should appear for the ‘no vector’ control transformation. The control reaction omitting the repair template DNA should yield very few (typically less than ten) colonies because Cas9 is selecting against unedited cell growth by repeatedly inducing double strand DNA breaks. The few colonies arising in the absence of a repair template typically have a random mutation near the Cas9 cleavage site, due to error-prone NHEJ. If a large number of colonies appear on the ‘no repair template’ control plate, this would indicate poor cutting by Cas9, likely due to a problem with guide RNA design. The experimental transformation reaction (i.e., sgRNA/Cas9 expression vector + repair template) should yield 10–100x more colonies than the no repair template control, but fewer colonies than transformation with the empty control vector (i.e., pML104 or pML107).

BASIC PROTOCOL 4

SCREENING FOR EDITED CELLS

This protocol describes how to screen yeast for the desired genome edit. Yeast colonies transformed with the sgRNA/Cas9 expression vector are initially streaked to selective media to eliminate the small fraction of unedited cells in the colony. Then genomic DNA is isolated, and the genome edit/mutation of interest is screened by PCR amplification and DNA sequencing. The quick genomic DNA isolation protocol described below is adapted from (Looke, Kristjuhan, & Kristjuhan, 2011).

Materials

Transformed yeast colonies from Basic Protocol 3

Plates containing selective yeast medium (SC-URA or SC-LEU)

PCR thermal cycler

Microcentrifuge

Heat block (optional)

30°C incubator

Yeast Lysis Buffer (see recipe)

100% Ethanol

70% Ethanol

3 M Sodium Acetate, pH 5.2

ddH₂O or TE

EconoTaq (Lucigen, #30031 or similar)

10x EconoTaq Buffer (Lucigen or similar)

Forward and reverse oligonucleotides flanking targeted gene edit (100 pmol/μL)

dNTPs

PCR cleaning kit (e.g. Zymo Research Clean &Concentrator-5 Kit #D4014, or similar)

Additional reagents and equipment needed for performing gel electrophoresis and DNA sequencing.

Re-streaking of transformants to select against unedited cells.

• 1
  Streak 3–6 colonies from the transformation plate (Basic Protocol 3) for isolation onto selective medium (e.g., SC-URA or SC-LEU), and incubate at 30°C for 2–3 days, until isolated colonies appear.

  The process of re-streaking for isolation to selective medium is important for minimizing the number of false positives, since in our experience the colonies on the transformation plate often contains a small fraction of unedited cells.

• 2Patch single colonies onto selective medium and incubate at 30°C for 2–3 days.

Genomic DNA isolation of transformants

• 3
  Prepare fresh stock of Yeast Lysis Buffer (see recipe) and aliquot 100 μL to a 0.5 mL microcentrifuge tube for each transformant to be screened.

  The Yeast Lysis Buffer should be prepared fresh for each experiment. This can be done quickly using 1 M Lithium Acetate (pH 7.5) and 10% SDS stock solutions. The use of 0.5 mL tubes allows the samples to be heated quickly in a thermal cycler; however, 1.5 mL tubes can also be used and incubated in a heat block or water bath (see step 5).

• 4
  Using a wooden applicator stick, collect cells from each patch and re-suspend in lysis buffer. Vortex 10–20 seconds to break up cell clumps.

  The volume of cells needed for this protocol is very flexible; however, more cells typically lead to greater yields of genomic DNA. Aim to collect cells from a 2–3 mm square area, being sure to leave plenty behind to use for future experiments, making frozen stocks, etc.

• 5
  Incubate cells in lysis buffer in a PCR thermal cycler set to 70°C for ≥ 5 minutes.

  This incubation can also be performed in a water bath or heat block. A five-minute incubation is sufficient, but longer incubations work as well.

• 6Allow tubes to cool to room temperature on the benchtop or by briefly incubating on ice, then add 300 μL 100% Ethanol (room temperature). Vortex 5–10 seconds to thoroughly mix.
• 7
  Centrifuge samples at high speed (about 12,000–13,000 rpm) for 5 minutes at room temperature.

  This pellets the DNA and cell debris. The pellet should be white or off-white colored and easy to see.

• 8Remove supernatant and discard. Wash pellets carefully with 0.5 mL 70% ethanol by adding the 70% ethanol slowly around the circumference of the tube and then removing it with the pipette.
• 9
  Centrifuge tubes briefly (~10 seconds) and remove any remaining 70% ethanol.

  This minimizes the drying time needed in step 10.

• 10
  Air dry for 5 minutes at room temperature, then re-suspend pellets in 100 μL ddH₂O or TE by pipetting/vortexing.

  Pipetting is recommended to initially break up pellet, after which the sample can be vortexed to help fully re-suspend the DNA. Samples may appear slightly cloudy because the precipitated cell debris will not dissolve.

• 11
  Centrifuge tubes for 30 seconds at 12,000–13,000 rpm. Transfer supernatants (~100 μL) to fresh 0.5 or 1.5 mL microcentrifuge tubes.

  The genomic DNA is in the supernatant. The unpurified supernatant can be used for PCR screening (see step 18 below); however, we recommend performing the ethanol precipitation described below to ensure a cleaner DNA sample.

Ethanol precipitation of isolated genomic DNA

• 12Add 10 μL 3 M Sodium Acetate to each sample and mix by pipetting/vortexing.
• 13
  Add 275 μL of 100% Ethanol and vortex to mix well, then incubate for ≥ 2 hours at −20°C.

  Samples may be left overnight (or longer) in a −20°C freezer. This is a convenient stopping point if you wish to do the protocol over the span of more than one day.

• 14Centrifuge 10–15 minutes at maximum speed (e.g., 13,000 rpm).
• 15
  Carefully remove supernatant and discard. Wash pellets with 0.5 mL 70% ethanol by adding the ethanol slowly around the circumference of the tube and then removing it with the pipette.

  The DNA pellet will be white and much smaller than the pellet seen in step 7. It is a good idea to keep track of the orientation of the tubes in the microcentrifuge so that you know where the pellet is likely to be located.

• 16Centrifuge tubes 20–30 seconds at maximum speed, then carefully remove remaining 70% ethanol wash.
• 17Allow pellets to air dry 5 minutes, then re-suspend in 15μL ddH₂O (or TE) by pipetting/vortexing.

Confirming genome edit by PCR amplification and DNA sequencing

• 18
  Assemble reactions to perform PCR amplification of genomic DNA using primers that flank the targeted region:

  • 15.9 μL ddH₂O
  • 2.5 μL 10x EconoTaq Buffer
  • 2.5 μL forward primer (10 pmol/μL)
  • 2.5 μL reverse primer (10 pmol/μL)
  • 0.5 μL dNTPs
  • 1 μL genomic DNA
  • 0.125 μL EconoTaq (Lucigen, #30031)

  We recommend designing flanking primers that amplify a 300–500 bp region with the gene edit centrally located. Two microliters of genomic DNA can also be used in the PCR reaction; be sure to adjust the volume of water accordingly. Reactions can be scaled up if desired.

• 19
  Perform PCR reactions under the following conditions:

  • $$\begin{array}{ll}94°\text{C}& 2\text{min}\\ 94°\text{C}& 30\text{sec}\\ 50–65°\text{C}*& 30\text{sec}\\ 72°\text{C}& 30\text{sec}**\\ 72°\text{C}& 5\text{min}\end{array}\}\text{x29 cycles}$$
  • *Will vary depending on T_m of primers. Set to 3–4°C below the lowest T_m of the primer pair
  • **Will vary depending on length of DNA to be amplified; allow 1 min/kb

  These PCR guidelines are according to the recommendations for EconoTaq DNA Polymerase (Lucigen). If another DNA polymerase is used, be sure to adjust the protocol according to its manufacture’s recommendations.

• 20
  Assess success of PCR reactions by agarose gel electrophoresis, and clean remaining reaction using a kit (e.g. Zymo Research Clean & Concentrator-5 #D4014, or similar) or by ethanol precipitation.

  Use 2–4 μL of PCR reaction for electrophoresis. The use of a kit to clean the PCR product is important for the subsequent sequencing step (step 21).

• 21Submit samples for DNA sequencing to identify isolates that are positive for the desired genome edit. Either the forward or reverse flanking primer can be used as the sequencing primer.

BASIC PROTOCOL 5

REMOVING sgRNA/CAS9 EXPRESSION VECTOR

This protocol describes how to remove from yeast the pML104 sgRNA/Cas9 expression vector, which has a URA3 selectable marker, by counter-selection using 5-fluoroorotic acid (5-FOA). The alternate protocol 2 discusses a means for removing the pML107 sgRNA/Cas9 expression vector, which has a LEU2 selectable marker, by growth on non-selective media and screening for colonies that have lost the vector.

Materials

Transformed yeast colonies from Basic Protocol 3

Plates containing 5-fluoroorotic acid (SC+FOA)

Plates containing selective yeast medium (SC-URA)

Plates containing rich yeast extract peptone dextrose (YPD) medium

30°C incubator and shaker/rotator

Removal of pML104-derived vectors from genome-edited strain

1. Patch yeast colonies/isolates that were found to be positive for the desired gene edit to YPD plate. Incubate in a 30°C for 1–2 days.

   Culturing in nonselective media is critical for allowing the isolates to lose the sgRNA/Cas9 expression vector.

2. Streak for isolation (i.e., single colonies) on SC+FOA plate. Incubate at 30°C for 2–3 days.

   5-FOA selects for URA^– yeast cells, thereby selecting for cells that have lost the pML104-derived sgRNA/Cas9 expression vector.

3. Confirm loss of pML104-derived sgRNA/Cas9 expression vector by streaking colonies from SC+FOA plate to YPD and SC-URA plates. Incubate at 30°C for 2–3 days.

   Successful removal of the pML104-derived sgRNA/Cas9 expression vector will result in colonies able to grow on YPD plates, but unable to grow on SC-URA plates.

ALTERNATE PROTOCOL 2

REMOVING pML107-DERIVED sgRNA/CAS9 EXPRESSION VECTOR

The pML107 sgRNA/Cas9 expression vector has a LEU2 marker, so to remove this vector, we grow the cells on non-selective media and screen for colonies unable to grow on SC-LEU plates.

Removal of pML107-derived sgRNA/Cas9 expression vector from genome-edited strain

1. Inoculate culture tube containing 2–3 mL YPD media with yeast colony found to be positive for the desired gene edit. Incubate in a 30°C shaker or rotator overnight.

   Culturing in nonselective media is critical for allowing the isolates to lose the expression vector.

2. Plate cultures onto YPD plates at a sufficient dilution to obtain single colonies. Grow 2–3 days at 30°C.

   50 μL of a 10⁻⁵ dilution of an overnight culture will often yield single colonies.

3. Patch 4–8 colonies first to SC-LEU plates, and then to YPD plates. Strains that have lost the pML107-derived sgRNA/Cas9 vector will no longer grow on SC-LEU media.

   Strains will likely have lost the expression vector at this point; however, if for some reason they have not, steps 1–3 can be repeated.

REAGENTS AND SOLUTIONS

Use sterile, distilled, deionized water (i.e., ddH₂O) to prepare the following buffers and in all protocol steps.

One Step Buffer

200 mM Lithium Acetate

100 mM Dithiothreitol

40% (w/v) Polyethylene Glycol 3350

Store in 1 mL aliquots at −20°C

LiAc-TE Buffer

100 mM Lithium Acetate

10 mM Tris-HCl pH 8

1 mM EDTA

Store at room temperature

Yeast Lysis Buffer

200 mM Lithium Acetate

1% SDS

Prepare fresh prior to each genomic isolation

COMMENTARY

Background Information

Budding yeast (S. cerevisiae) has long been a leading model organism, due to the diverse array of methods available for altering its genomic sequence. An important aspect of genome editing is the ability to create tailored mutations in the genome without the requirement of a selectable marker (i.e., ‘marker-free’ genome editing). This is important, because in many cases using a selectable marker to introduce a genome edit is cumbersome or may modulate the function of the edited genomic locus. This is particularly problematic for genome edits in promoter regions, for example. While other methods have been developed for marker-free genome editing in yeast, most notably the delitto perfetto method (Storici, Lewis, & Resnick, 2001; Storici & Resnick, 2006; see also Current Protocols article; Moqtaderi & Geisberg, 2018), these methods typically require multiple steps to construct the marker-free genome edit.

An important breakthrough was the discovery that CRISPR-Cas9 can be used to perform marker-free genome editing in yeast (DiCarlo et al., 2013), since repeated Cas9 cleavage of the target site selects against unedited cells, thereby obviating the need for a selectable marker. This is likely a consequence of the high efficiency of homologous recombination and the low efficiency of NHEJ in yeast, particularly for Cas9-induced blunt DNA double strand breaks, resulting in efficient homologous recombination with the repair template containing the genome edit. CRISPR can be used for a wide-variety of genome-editing applications beyond marker-free genome editing, as discussed in recent reviews (Giersch & Finnigan, 2017; Stovicek et al., 2017). Notably, certain Cas9 applications allow multiple genome edits to be introduced simultaneously in yeast (e.g., (Bao et al., 2015; Horwitz et al., 2015; Jakociunas et al., 2015; Ryan et al., 2014)).

In practical terms, CRISPR-Cas9 genome editing in yeast is often limited by the time and effort required to construct a vector expressing the user-designed guide RNA. To facilitate guide RNA cloning for yeast genome editing, we developed new yeast-E. coli shuttle vectors (i.e., pML104 and pML107) possessing unique BclI and SwaI restriction enzymes in the sgRNA expression cassette to expedite cloning of the user-designed guide sequence (Laughery et al., 2015). Importantly, these vectors also express Cas9, so that a single yeast transformation with an sgRNA/Cas9 expression vector can introduce the marker-free genome edit with high efficiency.

CRISPR-Cas9 genome editing can also be used to generate more traditional genome alterations in yeast, including gene deletions, C-terminal epitope-tagging of proteins, etc. Since traditional PCR-based methods for making these types of genome alterations in yeast are fast and robust, we do not routinely use CRISPR-Cas9 for these applications. However, CRISPR-Cas9 can be used as an alternative method for these traditional genome alterations, particularly if the genetic manipulation is technically challenging or if available selectable markers are limiting.

Critical Parameters and Troubleshooting

There are several critical parameters to consider when genome editing yeast using the pML104 or pML107 sgRNA/Cas9 expression vectors.

sgRNA design and Cas9 targeting efficiency

The design of the sgRNA is of critical importance to the success of the protocol. As it has been noted, the presence of multiple thymidine nucleotides at the 3’ end of the guide RNA can cause premature termination of the sgRNA transcript by RNA polymerase III, which would lead to poor Cas9 cleavage of the target site. Typically, this is apparent in the yeast transformation controls listed in Basic Protocol 3. If Cas9 is targeting the locus efficiently, there should be a precipitous drop in the number of transformants when the sgRNA/Cas9 expression vector is transformed into yeast without a repair template (i.e., no repair template control). A high number of transformant colonies in the no repair template control is diagnostic of poor Cas9 cleavage of the target site, and is associated with decreased genome editing efficiency. Although we have isolated edited cells under these conditions, typically more colonies must be screened to isolate edited cells. In such cases, we would recommend redesigning the guide RNA sequence.

Preparation of a well-cut vector for cloning

Obtaining a well-cut pML104 or pML107 backbone is critical for successful preparation of the sgRNA/Cas9 expression vector. The yield of the cleaved vector DNA is often very low following gel purification (e.g., as low as 20 ng/μL); however, very little is needed for a successful ligation. Incomplete digestion of the plasmid will lead to many background colonies when E. coli is transformed with the control ligation omitting the hybridized oligonucleotides. This is because incomplete digestion allows singly-cut vectors to efficiently re-ligate to themselves. Thus, it is recommended to follow the suggested guidelines for performing the restriction enzyme digests and to run gel electrophoresis long enough to maximally separate the cut and uncut vector DNA. It should be noted, however, that successful cloning of the guide RNA has been efficiently obtained even in cases where there were a high number of background transformants in the ‘no hybridized oligonucleotide’ control ligation. Therefore, we recommend sequencing 3–6 transformants, regardless of the number of background transformants.

Transformation and purification of yeast colonies prior to screening

The method used for yeast transformation is not a critical parameter for this experiment. A single step, low efficiency transformation protocol is detailed because it has been found to be sufficient for Cas9 editing in most cases. However, purification of transformant colonies by re-streaking to selective plates, prior to screening, is a critical parameter because a small fraction of unedited cells in the initial colonies is common. As a result, we highly recommend streaking transformants for isolation to selective media and patching isolates once more to selective plates to purify the strains prior to genomic DNA isolation and analysis (see Basic Protocol 4).

Understanding Results

With a well-designed guide RNA, the protocol described above should yield a very high editing efficiency in yeast. In our published study, >90% of the transformant colonies (see Basic Protocol 3) had the desired genome edit (Laughery et al., 2015), although the efficiency can vary depending on the design of the guide RNA and repair template.

As described above (see Critical Parameters and Troubleshooting), the number of yeast transformants (see Basic Protocol 3) is usually diagnostic of editing efficiency. If Cas9 is targeting the genome efficiently with the user-designed guide RNA, then there should be a precipitous drop in the number of transformants when the sgRNA/Cas9 expression vector is transformed into yeast without a repair template (i.e., no repair template control). A high number of transformant colonies in the no repair template control indicates poor Cas9 cleavage of the target site, and is usually associated with lower genome editing efficiency. While edited cells can be obtained under such conditions, the editing efficiency is usually much lower. In such cases, we suggest redesigning the guide RNA.

Time Considerations

After designing and obtaining oligonucleotides corresponding to the user-designed guide RNA and repair template, an edited yeast strain can be obtained in approximately 4 weeks. An example time line is as follows:

Cloning the sgRNA/Cas9 expression vector (Basic Protocol 1)

Day 1: Begin overnight SwaI digest of pML104 (or pML107) and hybridize sgRNA oligonucleotides.

Day 2: Heat inactivate SwaI and perform BclI digest, then electrophorese cut plasmid. Isolate linearized plasmid from agarose gel, purify, and begin overnight ligation.

Day 3: Transform ligation into E. coli and grow overnight.

Day 4: Start liquid cultures from transformant colonies and incubate overnight.

Day 5: Isolate plasmid DNA and submit for sequencing with T3 primer.

Day 7–8: Analyze sequence data to identify isolates containing sgRNA oligonucleotides. Start overnight liquid culture of the yeast strain to be edited (Basic Protocol 3), and hybridize repair template oligonucleotides (Basic Protocol 2).

Induce genome editing by co-transformation of yeast (Basic Protocol 3)

Day 9: Transform yeast with sgRNA/Cas9 expression vector and repair template.

Days 9–12: Incubate plates at 30°C until colonies appear.

Screening for edited cells (Basic Protocol 4)

Day 12: Streak individual transformant colonies for isolation onto selective medium.

Day 12–15: Incubate plates at 30°C until isolated colonies appear.

Day 14: Patch individual colonies to selective plates.

Day 14–17: Incubate plates at 30°C until significant growth appears.

Day 17: Isolate genomic DNA from patched isolates. Perform PCR of genomic DNA.

Day 18: Analyze PCR amplification by gel electrophoresis, clean reactions, and submit for DNA sequencing.

Removing sgRNA/Cas9 expression vector (Basic Protocol 5)

Days 20–21: Analyze sequencing results and patch colonies on YPD medium.

Day 22: Streak patches for isolation to SC-Uracil + 5-FOA plates (URA selection) and incubated 2–3 days at 30°C until colonies appear.

Days 24–25: Patch isolated colonies first to SC-Uracil plates and then to YPD plates and incubate 2–3 days at 30°C to confirm loss of pML104 sgRNA/Cas9 expression plasmid.