A strategy for generating mutations in histone genes at their endogenous location in Saccharomyces cerevisiae is presented.
We describe a PCR- and homologous recombination-based system for generating targeted mutations in histone genes in budding yeast cells. The resulting mutant alleles reside at their endogenous genomic sites and no exogenous DNA sequences are left in the genome following the procedure. Since in haploid yeast cells each of the four core histone proteins is encoded by two non-allelic genes with highly homologous open reading frames (ORFs), targeting mutagenesis specifically to one of two genes encoding a particular histone protein can be problematic. The strategy we describe here bypasses this problem by utilizing sequences outside, rather than within, the ORF of the target genes for the homologous recombination step. Another feature of this system is that the regions of DNA driving the homologous recombination steps can be made to be very extensive, thus increasing the likelihood of successful integration events. These features make this strategy particularly well-suited for histone gene mutagenesis, but can also be adapted for mutagenesis of other genes in the yeast genome.
The four core histone proteins H2A, H2B, H3, and H4 play central roles in the compaction, organization, and function of eukaryotic chromosomes. Two sets of each of these histones form the histone octamer, a molecular spool that directs the wrapping of ~147 base pairs of DNA around itself, ultimately resulting in the formation of a nucleosome1. Nucleosomes are active participants in a variety of chromosome-based processes, such as the regulation of gene transcription and the formation of euchromatin and heterochromatin across chromosomes, and as such have been the focus of intense research over the course of the past several decades. A number of mechanisms have been described by which nucleosomes can be manipulated in ways that can facilitate execution of specific processes – these mechanisms include posttranslational modification of histone residues, ATP-dependent nucleosome remodeling, and ATP-independent nucleosome reorganization and assembly/disassembly2,3.
The budding yeast Saccharomyces cerevisiae is a particularly powerful model organism for the understanding of histone function in eukaryotes. This can be largely attributed to the high degree of evolutionary conservation of the histone proteins throughout the domain eukarya and the amenability of yeast to a variety of genetic and biochemical experimental approaches4. Reverse-genetic approaches in yeast have been widely used to study the effects of specific histone mutations on various aspects of chromatin biology. For these types of experiments it is often preferable to use cells in which the mutant histones are expressed from their native genomic loci, as expression from autonomous plasmids can lead to abnormal intracellular levels of histone proteins (due to varying numbers of plasmids in cells) and concomitant alteration of chromatin environments, which can ultimately confound the interpretation of results.
Here, we describe a PCR-based technique that allows for targeted mutagenesis of histone genes at their native genomic locations that does not require a cloning step and results in the generation of the desired mutation(s) without leftover exogenous DNA sequences in the genome. This technique takes advantage of the efficient homologous recombination system in yeast and has several features in common with other similar techniques developed by other groups – most notably the Delitto Perfetto, site-specific genomic (SSG) mutagenesis, and cloning-free PCR-based allele replacement methods5,6,7. However, the technique we describe has an aspect that makes it particularly well-suited for mutagenesis of histone genes. In haploid yeast cells, each of the four core histones is encoded by two non-allelic and highly homologous genes: for example, histone H3 is encoded by the HHT1 and HHT2 genes, and the open reading frames (ORFs) of the two genes are over 90% identical in sequence. This high degree of homology can complicate experiments designed to specifically target one of the two histone-encoding genes for mutagenesis. Whereas the aforementioned methods often require the use of at least some sequences within the ORF of the target gene to drive homologous recombination, the technique we describe here makes use of sequences flanking the ORFs of the histone genes (which share much less sequence homology) for the recombination step, thus increasing the likelihood of successful targeting of mutagenesis to the desired locus. Moreover, the homologous regions that drive recombination can be very extensive, further contributing to efficient targeted homologous recombination.
NOTE: The experimental strategy for targeted in situ histone gene mutagenesis includes several steps (summarized in Figure 1). These steps include: (1) Replacement of the target histone gene with the URA3 gene, (2) Generation and purification of PCR products corresponding to two partially overlapping fragments of the target histone gene using primers harboring the desired mutation(s), (3) Fusion PCR of the two partially overlapping fragments to obtain full size PCR products for integration, (4) Co-transformation of full size PCR products and backbone plasmid, and selection for marker on plasmid, (5) Screen for 5-FOA-resistant transformants, (6) Purification of 5-FOA-resistant colonies and loss of backbone plasmid, and (7) Molecular analyses to assay for proper integration of the mutant allele.
Figure 1: Overview of the Strategy for Targeted in situ Mutagenesis of Histone Genes in Budding Yeast. In this example the targeted gene is HHT2, but any other core histone gene can also be mutagenized using this strategy. (A) Haploid yeast cells harbor two histone H3-encoding genes (HHT1 and HHT2) and two histone H4-encoding genes (HHF1 and HHF2) arranged as shown in the figure (the HHT1 and HHF1 genes are located on chromosome II and the HHT2 and HHF2 genes are located on chromosome XIV – in each case, the arrows point in the direction of transcription). In the first step of the procedure, the ORF of the HHT2 gene is replaced with the URA3 gene, giving rise to an hht2Δ::URA3 strain. (B) In part 1, a wild-type copy of the HHT2 gene from a genomic DNA sample is used as template for two PCR reactions to generate the two partially overlapping fragments of the gene. The reverse primer for the first reaction includes one or more mismatched nucleotides (indicated with a red circle) that correspond to the desired mutation(s) to be introduced into the genome. The forward primer for the second reaction has the equivalent mismatch in a reverse complementary configuration (also indicated with a red circle). The two PCR products generated in part 1 (products a and b) are then used as templates for fusion PCR using two primers that anneal to products a and b in the fashion shown in part 2. This results in the generation of full-size PCR products (product c in part 3) harboring the desired mutation(s). (C) The hht2Δ::URA3 strain is then co-transformed with the full-size PCR products and a backbone plasmid (a HIS3-marked plasmid in this example), and cells are selected for the presence of the plasmid (on media lacking histidine in this example). Transformants are then screened for 5-FOA resistance – resistant cells are candidates for having undergone a homologous recombination event leading to integration of the PCR product and excision of the URA3 gene, as shown. Subsequent loss of the backbone plasmid by mitotic cell division leads to the final desired histone mutant strain. We have found that selection of the backbone plasmid followed by screening for 5-FOA resistance results in a much higher frequency of identification of correct integration events compared to direct selection on 5-FOA plates, which mostly identifies cells that have acquired spontaneous URA3 mutations. (This figure has been modified from reference14). Please click here to view a larger version of this figure.
1. Replacement of the Target Histone Gene with the URA3 Gene
2. Generation and Purification of PCR Products Corresponding to Two Partially Overlapping Fragments of the Target Histone Gene using Primers Harboring the Desired Mutation(s)
3. Fusion PCR of the Two Partially Overlapping Fragments to Obtain Full Size PCR Products for Integration
4. Co-transformation of Full Size PCR Products and Backbone Plasmid, and Selection for Marker on Plasmid
5. Screen for 5-FOA-resistant Transformants
6. Purification of 5-FOA-resistant Colonies and Loss of Backbone Plasmid
7. Molecular Analyses to Assay for Proper Integration of the Mutant Allele
We describe the generation of an hht2 allele expressing a histone H3 mutant protein harboring a substitution at position 53 from an arginine to a glutamic acid (H3-R53E mutant) as a representative example of the targeted in situ mutagenesis strategy.
We generated a strain in which the entire ORF of HHT2 is replaced by the URA3 gene (see step 1 of the protocol). This strain, yAAD156, also harbors a his3Δ200 allele, which causes the cells to be auxotrophic for histidine. Following the procedure indicated in step 2 of the protocol, we then generated the two partially overlapping fragments of HHT2. The following primers were used for the first fragment: Forward Primer (OAD20): 5' GCGTTCATTATCGCCCAATGTG 3' and the Reverse Primer (R53Erev): 5' GTTCAGTAGATTTTTGGAATtcTCTAATTTCTCTCAAG 3'. The following primers were used for the second fragment: Forward Primer (R53Efor): 5' CTTGAGAGAAATTAGAgaATTCCAAAAATCTACTGAAC 3' and the Reverse Primer (OAD21): 5' GCGCTTGATCAGCAGTTCATCG 3'. Note that the Reverse Primer in the first reaction and the Forward Primer in the second reaction contain the desired mutated nucleotides (in lower case) – these mutated nucleotides are nestled in between two long stretches of wild-type sequences, as this allows for efficient annealing of the primer to the template DNA despite the mismatch at the mutated positions. Also note that the mutated nucleotides in these two primers are reverse complementary to each other and cause an AG to GA mutation in the sense strand, which results in a AGA to GAA codon change, which ultimately produces the R53E mutation in the encoded protein. Results from these PCR reactions are shown in Figure 2A.
Using the procedure in step 3, we then generated the full-size PCR products. The primers used were Forward Primer (OAD479): 5'TATGGCTCGGTGTCAAAACA 3' and Reverse Primer (OAD480): 5' CATGGTTTCTTGCCGGTTAT 3'. When designing these primers, it is important to make sure that the resulting fusion PCR products include at least 40 base pairs on either ends to drive the homologous recombination reaction (the longer the regions of homology, the more efficient and specific the homologous recombination events will be). In our experiment, the full-size PCR products contained a 195 base pair region and 220 base pair region homologous to the regions upstream and downstream of the HHT2 ORF, respectively. A sample of these PCR products was analyzed by agarose gel electrophoresis (Figure 2B).
The full-size PCR products were then co-transformed along with plasmid pRS413 (a centromeric, HIS3-marked plasmid13) and transformants were selected on plates lacking histidine (SC-his plates) as described in step 4 of the protocol. His+ colonies were then screened for 5-FOA resistance following the procedures described in step 5 of the protocol. Figure 3 shows an example of a transformation plate (SC-his) and 5-FOA plates following replica-plating from SC-his plates. Examples of candidate samples as well as samples unlikely to represent the desired integration events are also presented in Figure 3. In our experiment, we identified twelve candidate samples out of ~90,000 transformants screened. These candidates were then purified as described in step 6 of the protocol.
The twelve candidates were then subjected to the PCR analysis described in step 7 of the protocol to assess if they reflected authentic replacements of the URA3 gene with mutant PCR products. For our experiment, we used a Forward Primer that anneals 89 base pairs upstream from the integration site of the PCR product and a Reverse Primer that anneals 302 base pairs downstream from the integration site of the PCR product. These primers generate PCR products of 1217 base pairs in size if the HHT2 locus is occupied by HHT2 (or mutant versions of HHT2 harboring base pair substitutions), or PCR products of 2188 base pairs in size if the locus is occupied by the URA3 marker gene. The sequence of the Forward Primer used (OAD476) is: 5' GAAACTATTGGCACGCCCTA and that of the Reverse Primer used (OAD477) is: 5' CCTGCGAATCAACCGATACT 3'. Of the twelve candidates we had identified, four showed integration of a PCR product at the correct location (see Figure 4A for an example). Finally, since the AG to GA mutation generates a new EcoRI restriction site, we subjected PCR products from one of the successful integration samples to an EcoRI digestion and were able to demonstrate that the mutant sequence had indeed been integrated into the genome (see Figure 4B). In this particular case we did not sequence the PCR products to ensure that additional unwanted mutations had not been incorporated into the genome: however, we have used a procedure very similar to this to generate a large number of HHT2 mutations in a past study14, and found that the majority of the integrated mutant alleles carry no mutations other than the desired ones.
Figure 2: Generation of PCR Products Harboring Desired Mutations for Integration into the Yeast Genome. (A) PCR reaction to generate the partially overlapping fragments of HHT2 harboring the desired mutations. Top: cartoon representation of the two PCR reactions to obtain the two HHT2 fragments (refer to Figure 1B-1). Red circles represent the mismatched nucleotides on the primers used to introduce the AG to GA mutation in the sense strand of the gene. Bottom: gel electrophoresis analysis of PCR products a and b (lanes 2 and 3, respectively). DNA standards are shown in lane 1. (B) Fusion PCR reaction to generate full size PCR product for integration into yeast genome (refer to Figure 1B-2 and 3). Top: cartoon representation of the PCR reaction and expected PCR product (product c). Red circles represent desired mutation as described in (A) above. Bottom: gel electrophoresis analysis of PCR products c (lane 2). DNA standards are shown in lane 1. Please click here to view a larger version of this figure.
Figure 3: Representative Results from the Co-transformation Experiment and 5-FOA Screen. Left: representative SC-his plate plated with cells subjected to the co-transformation procedure after a 3-day incubation at 30˚C. Approximately 5,000 colonies were counted. Middle: representative 5-FOA plate following replica-plating from an SC-his co-transformation plate after a 2-day incubation at 30˚C. Samples 1 and 2 were considered candidates for having gone through a successful integration event due to their asymmetric morphologies. This is because a true replacement of the URA3 gene with the hht2 allele is likely to occur before (or very soon after) a transformed cell is plated on the SC-his plate and, as a result, the colony that will arise at that location will contain mostly, if not exclusively, 5-FOA-resistant cells – when such a colony is then subsequently replica-plated to a 5-FOA plate it will be squashed, giving rise to an asymmetric-looking patch of cells on the plate. Conversely, spontaneous mutations in the URA3 gene that can arise during colony formation are more likely to be confined within a small region of a colony, and thus more likely to give rise to papillae when replica-plated to a 5-FOA plate. Right: a different representative 5-FOA plate following replica-plating from an SC-his co-transformation plate after a 3-day incubation at 30 °C. An additional candidate (sample 3) as well as two small papillae are visible on this plate (samples 4 and 5) – such papillae, which are most clearly visible after 3 or more days of incubation, are likely to represent spontaneous URA3 mutations and not the desired integration event. Please click here to view a larger version of this figure.
Figure 4: Molecular Analyses of Candidate Integration Samples. (A) PCR assay to identify successful integration events. Top: cartoon representations of the PCR reactions used to assess successful integration of the hht2 mutant allele into the genome. Bottom: gel electrophoresis analysis of PCR reactions using genomic DNA derived from an HHT2 wild-type strain (used as a control, lane 2), strain yAAD156 harboring the hht2Δ::URA3 replacement (used as a control, lane 3), a candidate integration sample (lane 4), and a 5-FOA-resistant papillae sample (and thus unlikely to represent a correct integration event, lane 5). DNA standards are shown in lane 1. Note that the sizes of the PCR products for the candidate integration sample are consistent with a successful integration event, whereas the size of the PCR products for the papillae sample is consistent with an intact hht2Δ::URA3 locus. (B) EcoRI digestion to confirm presence of mutant allele. Top: cartoon representation of the expected digestion fragments derived from an EcoRI digestion reaction of PCR products containing either the wild-type HHT2 gene or the mutant hht2 allele. Bottom: gel electrophoresis analysis of digestion reactions of PCR products derived from a wild-type HHT2 strain (lane 2; the PCR products used here were derived from the same sample as that used in lane 2 of panel A) and a candidate integration sample (lane 3; the PCR products used here were derived from the same sample as that used in lane 4 of panel A). DNA standards are shown in lane 1. Note that the digestion patterns confirm that the AG to GA mutation has been successfully introduced into the genome of the candidate integration sample. Please click here to view a larger version of this figure.
The high level of sequence homology between the two non-allelic genes that code for each of the four core histone proteins in haploid S. cerevisiae cells can represent a challenge for investigators who wish to specifically target one of the two genes for mutagenesis. Previously described yeast mutagenesis methodologies, including the Delitto Perfetto, site-specific genomic (SSG) mutagenesis, and cloning-free PCR-based allele replacement methods5,6,7, as well as more recent yeast CRISPR-based techniques15, often rely at least in part on sequences within the ORF of a target gene to recruit the recombination or repair machineries to the desired genomic site to eventually generate the final mutant allele. On the other hand, the strategy we have presented here makes use of DNA sequences flanking the targeted ORF to drive the homologous recombination event required for the integration of the mutation, and, as a result, allows for more specific targeting of a gene over another despite the two genes having highly homologous ORFs. This feature makes our strategy particularly well-suited for histone mutagenesis. However, this strategy can also be adapted for mutagenesis of other genes in the yeast genome.
Additional features of our strategy include the facts that the lengths of the regions that drive homologous recombination of the mutant genes can be designed to be very extensive, thus increasing the efficiency of integration at the target genomic location, and that besides the desired mutation(s) no additional DNA sequences are introduced into the genome. An additional advantage of this system is that once a strain harboring replacement of a specific histone gene with URA3 has been constructed, it can be used for the generation of any desired mutant version of that particular histone gene. Thus, for example, strain yAAD156, which is available to the research community upon request, can be used to make any desired hht2 mutation without the need of constructing a de novo hht2Δ::URA3 allele. Finally, by using properly designed PCR primers, this strategy can also be used to generate histone alleles with internal deletions or with mutations at multiple codons, as long as they are relatively near each other (for example, we have generated an hht2 allele encoding an H3-K56R,L61W double mutant protein using this strategy).
The ability to specifically target one of two highly homologous histone genes for mutagenesis could be useful in a number of different experimental settings. For example, since the HHT1-HHF1 genes are expressed at different levels compared to the HHT2-HHF2 genes16, it could be of interest to determine whether a particular H3 or H4 mutation confers different phenotypes depending on which gene it is expressed from. Another example is a scenario in which an investigator wishes to generate haploid cells expressing a particular histone mutant from both corresponding histone genes – this could be achieved by first mutagenizing each gene independently in two different strains, and then obtaining double mutant haploid cells through a cross and subsequent isolation of the desired meiotic products. Yet another example relates to mutagenesis of histones H2A and H2B: given the fact that two slightly different isoforms of each protein are encoded by the corresponding non-allelic gene set, an investigator may want to assess the effects of a specific H2A or H2B mutation in the context of either isoform. When designing experiments to mutagenize the HTA1 and HTB1 loci (encoding H2A and H2B, respectively), investigators should be aware of recent findings showing that strains carrying an (hta1-htb1)Δ are viable only if they have amplified the HTA2-HTB2 locus (and nearby HHT1-HHF1 locus as well) through the generation of a small circular chromosome17, as this could complicate the interpretation of the results.
We have recently used a version of this histone mutagenesis strategy to generate histone H3 proteins expressing all possible amino acid substitutions at position 61, which is normally occupied by the amino acid leucine14. For those experiments, instead of using yAAD156 as the starting strain for the mutagenesis, we used strain yADP106, which harbors a replacement of the HHT2 ORF with both the URA3 and TRP1 nutritional markers (instead of just URA3). The presence of both marker genes facilitated the identification of candidates for integration of the mutant PCR products following the 5-FOA screen and purification step (steps 5 and 6 of the protocol), as such candidates became phenotypically Ura– and Trp–, whereas spontaneous URA3 mutation resulted in cells with a Ura– Trp+ phenotype. yADP106 is also available upon request, as is the sequence of the URA3-TRP1 cassette, which could be amplified as a unit and used for mutagenesis of other histone genes using the strategy described herein.
Inability to obtain the desired histone mutants using this strategy could be attributable to a number of possible reasons, including insufficient amount of full-size PCR products used for the integration step or an insufficient number of transformants following the co-transformation step. The former problem could be addressed by pooling together larger sets of PCR reactions or by designing different primer sets that produce a higher yield of product, whereas the latter problem could be solved by using higher amounts of backbone plasmid in the co-transformation experiment. Although, as indicated earlier, one can use this procedure to generate histone mutants harboring two or more amino acid substitutions, the targeted amino acids have to be relatively close to each other since the respective codons need to be located within the same primer molecule. Thus, this strategy cannot be easily adapted for the generation of histones with multiple mutations located distantly from each other across the length of the proteins.
The authors have nothing to disclose.
We thank Reine Protacio for helpful comments during the preparation of this manuscript. We express our gratitude to the National Science Foundation (grants nos. 1243680 and 1613754) and the Hendrix College Odyssey Program for funding support.
1 kb DNA Ladder (DNA standards) | New England BioLabs | N3232L | |
Agarose | Sigma | A5093-100G | |
Boric Acid | Sigma | B0394-500G | |
dNTP mix (10 mM each) | ThermoFisher Scientific | R0192 | |
EDTA solution (0.5M, pH 8.0) | AmericanBio | AB00502-01000 | |
Ethanol (200 Proof) | Fisher Scientific | 16-100-824 | |
Ethylenediaminetetraacetic acid disodium salt dihydrate (EDTA) | Sigma | E4884-500G | |
Lithium acetate dihydrate | Sigma | L6883-250G | |
MyCycler Thermal Cycler | BioRad | 170-9703 | |
Poly(ethylene glycol) (PEG) | Sigma | P3640-1KG | |
PrimeSTAR HS DNA Polymerase (high fidelity DNA polymerase) and 5X buffer | Fisher Scientific | 50-443-960 | |
Salmon sperm DNA solution | ThermoFisher Scientific | 15632-011 | |
Sigma 7-9 (Tris base, powder form) | Sigma | T1378-1KG | |
Sodium acetate trihydrate | Sigma | 236500-500G | |
Supra Sieve GPG Agarose (low metling temperature agarose) | AmericanBio | AB00985-00100 | |
Taq Polymerase and 10X Buffer | New England BioLabs | M0273X | |
Toothpicks | Fisher Scientific | S67859 | |
Tris-HCl (1M, pH 8.0) | AmericanBio | AB14043-01000 | |
a-D(+)-Glucose | Fisher Scientific | AC170080025 | for yeast media |
Agar | Fisher Scientific | DF0140-01-0 | for yeast media |
Peptone | Fisher Scientific | DF0118-07-2 | for YPD medium |
Yeast Extract | Fisher Scientific | DF0127-17-9 | for YPD medium |
4-aminobenzoic acid | Sigma | A9878-100G | for complete minimal dropout medium |
Adenine | Sigma | A8626-100G | for complete minimal dropout medium |
Glycine hydrochloride | Sigma | G2879-100G | for complete minimal dropout medium |
L-Alanine | Sigma | A7627-100G | for complete minimal dropout medium |
L-Arginine monohydrochloride | Sigma | A5131-100G | for complete minimal dropout medium |
L-Asparagine monohydrate | Sigma | A8381-100G | for complete minimal dropout medium |
L-Aspartic acid sodium salt monohydrate | Sigma | A6683-100G | for complete minimal dropout medium |
L-Cysteine hydrochloride monohydrate | Sigma | C7880-100G | for complete minimal dropout medium |
L-Glutamic acid hydrochloride | Sigma | G2128-100G | for complete minimal dropout medium |
L-Glutamine | Sigma | G3126-100G | for complete minimal dropout medium |
L-Histidine monohydrochloride monohydrate | Sigma | H8125-100G | for complete minimal dropout medium |
L-Isoleucine | Sigma | I2752-100G | for complete minimal dropout medium |
L-Leucine | Sigma | L8000-100G | for complete minimal dropout medium |
L-Lysine monohydrochloride | Sigma | L5626-100G | for complete minimal dropout medium |
L-Methionine | Sigma | M9625-100G | for complete minimal dropout medium |
L-Phenylalanine | Sigma | P2126-100G | for complete minimal dropout medium |
L-Proline | Sigma | P0380-100G | for complete minimal dropout medium |
L-Serine | Sigma | S4500-100G | for complete minimal dropout medium |
L-Threonine | Sigma | T8625-100G | for complete minimal dropout medium |
L-Tryptophan | Sigma | T0254-100G | for complete minimal dropout medium |
L-Tyrosine | Sigma | T3754-100G | for complete minimal dropout medium |
L-Valine | Sigma | V0500-100G | for complete minimal dropout medium |
myo-Inositol | Sigma | I5125-100G | for complete minimal dropout medium |
Uracil | Sigma | U0750-100G | for complete minimal dropout medium |
Ammonium Sulfate | Fisher Scientific | A702-500 | for complete minimal dropout medium |
Yeast Nitrogen Base | Fisher Scientific | DF0919-07-3 | for complete minimal dropout medium |
5-Fluoroorotic acid (5-FOA) | AmericanBio | AB04067-00005 | for 5-FOA medium |