ArticlePDF Available

High-throughput genotyping of CRISPR/Cas9-mediated mutants using fluorescent PCR-capillary gel electrophoresis

Authors:

Abstract and Figures

Recent advances in the engineering of sequence-specific synthetic nucleases provide enormous opportunities for genetic manipulation of gene expression in order to study their cellular function in vivo. However, current genotyping methods to detect these programmable nuclease-induced insertion/deletion (indel) mutations in targeted human cells are not compatible for high-throughput screening of knockout clones due to inherent limitations and high cost. Here, we describe an efficient method of genotyping clonal CRISPR/Cas9-mediated mutants in a high-throughput manner involving the use of a direct lysis buffer to extract crude genomic DNA straight from cells in culture, and fluorescent PCR coupled with capillary gel electrophoresis. This technique also allows for genotyping of multiplexed gene targeting in a single clone. Overall, this time- and cost-saving technique is able to circumvent the limitations of current genotyping methods and support high-throughput screening of nuclease-induced mutants.
Content may be subject to copyright.
1
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
www.nature.com/scientificreports
High-throughput genotyping
of CRISPR/Cas9-mediated mutants
using uorescent PCR-capillary gel
electrophoresis
Muhammad Khairul Ramlee
1
, Tingdong Yan
1
, Alice M. S. Cheung
1
, Charles T. H. Chuah
2
&
Shang Li
1,3
Recent advances in the engineering of sequence-specic synthetic nucleases provide enormous
opportunities for genetic manipulation of gene expression in order to study their cellular function
in vivo. However, current genotyping methods to detect these programmable nuclease-induced
insertion/deletion (indel) mutations in targeted human cells are not compatible for high-throughput
screening of knockout clones due to inherent limitations and high cost. Here, we describe an ecient
method of genotyping clonal CRISPR/Cas9-mediated mutants in a high-throughput manner involving
the use of a direct lysis buer to extract crude genomic DNA straight from cells in culture, and
uorescent PCR coupled with capillary gel electrophoresis. This technique also allows for genotyping
of multiplexed gene targeting in a single clone. Overall, this time- and cost-saving technique is able
to circumvent the limitations of current genotyping methods and support high-throughput screening
of nuclease-induced mutants.
Genome editing is an invaluable technique in modern day genetics. Targeted modication of mammalian
genome has greatly improved in the past decade or so particularly with the introduction of program-
mable nucleases such as transcription activator-like eector nucleases (TALENs), zinc-nger nucleases
(ZFNs) and the more recently described RNA-guided clustered regularly interspaced short palindromic
repeats (CRISPR)/Cas9 system. An integral function of these programmable nucleases is the targeted
knockout of specic genes via the introduction of insertion/deletion (indel) mutations in the coding
region of these genes. ese potentially frameshi-causing mutations are mediated by the error-prone
non-homologous end joining (NHEJ) repair induced by double strand breaks (DSBs) by the nucleases.
e resultant frameshi may, in turn, lead to loss of gene expression due to premature termination of
translation and nonsense-mediated decay
1–3
. Due to its ease of use and low cost, the CRISPR/Cas9 sys-
tem has made it extremely simple to specically knock out virtually any genes in the human genome.
Despite the convenience of design oered by the CRISPR/Cas9 system, genotyping of mutated clones
remains largely a bottleneck, especially for high-throughput purposes
4,5
. ere are currently several tech-
niques available for detecting engineered nuclease-mediated mutations in cells: SURVEYOR or T7E1
assay which detects mismatches in double-stranded DNA
6,7
; DNA melting analysis which dierenti-
ates fragments based on their melting curve
8,9
; restriction fragment length polymorphism (RFLP) assay
which reports the disappearance of restriction sequences at nuclease target sites
10
; and Sanger and deep
sequencing. However, each of these techniques suers certain drawbacks that hamper their use in a
high-throughput format.
1
Cancer and Stem Cell Biology Program, Duke-NUS Graduate Medical School, Singapore 169857.
2
Department of
Haematology, Singapore General Hospital, Singapore 169608.
3
Department of Physiology, Yong Loo Lin School of
Medicine, National University of Singapore, Singapore 117597. Correspondence and requests for materials should
be addressed to S.L. (email: shang.li@duke-nus.edu.sg)
Received: 30 May 2015
Accepted: 29 September 2015
Published: 26 October 2015
OPEN
www.nature.com/scientificreports/
2
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
SURVEYOR and T7E1 assays, whilst cheap and easy to use, are unable to dierentiate between identi-
cally mutated alleles (homozygous mutants)
7
, thus mutants bearing these alleles are erroneously reported
as wildtype clones. e simplicity and cost eectiveness of the RFLP assay are oen outweighed by its
lack of appropriate restriction sites
11
. DNA melting analysis, on one hand, is a relatively simple method
to use but on the other, tends to be inconsistent and is not very informative. Lastly, sequencing methods,
which can be said to be the gold standard for genotyping mutants, is extremely informative but rather
costly especially for high-throughput purposes.
Here, we describe a technique that is able to circumvent the abovementioned limitations and is
thus amenable to high-throughput detection of nuclease-induced mutants. Using uorescent PCR
coupled with capillary gel electrophoresis, we are able to eectively and eciently screen for CRISPR/
Cas9-mediated mutant clones harbouring indels at three targeted genes (ATRX, TP53 and MIR615) in a
high-throughput fashion. is technique also supports multiplexed targeting of genes which is particu-
larly useful in synergistic and redundancy studies. Overall, the method we describe here is not merely an
alternative to the existing genotyping techniques available but a superior one on the basis that it allows
for high-throughput screening of mutants, which renders it more time- and cost-eective.
Results
Testing sgRNA targeting eciency. Current methods of genotyping engineered nuclease-mediated
mutations are not optimal for high-throughput screening of individual knockout clones due to inherent
limitations and/or high cost. We thus sought to determine whether uorescent PCR coupled with capil-
lary gel electrophoresis is able to accurately genotype mutant clones in a high-throughput and cost-eec-
tive manner and hence circumvent the abovementioned limitations (see Fig.1 for schematic of protocol).
We selected the near-diploid colorectal carcinoma cells, HCT116, as our model system, and chose
three dierent genes found on three disparate chromosomes to test our hypothesis. ATRX encodes for a
chromatin remodelling protein and is found on the X-chromosome
12,13
. TP53 gene encodes the protein
p53 which is crucial in regulating DNA repair, cell cycle progression and apoptosis
14–16
. is gene is
found on chromosome 17 in humans
17
. Lastly, the MIR615 gene encodes two microRNAs—miR615-5p
and miR615-3p—and has been implicated in prostate and colon cancer
18
. is non-coding RNA gene
is found on chromosome 12 in humans. For our gene targeting experiments, we focused mainly on the
inactivation of miR615-3p, but has also investigated the eect of the genome targeting on miR615-5p
expression.
We used the database set up by Feng Zhang and colleagues (http://tools.genome-engineering.org)
19
to search for CRISPR single guide RNA (sgRNA) targets in exon 2 and 4 of ATRX gene and the region
Figure 1. Schematic of high-throughput genotyping technique via uorescent PCR-capillary gel
electrophoresis. First, cells are transfected with plasmids expressing Cas9-GFP fusion protein and individual
sgRNA. Two days later, GFP-positive cells are sorted and plated onto 10-cm dishes. When individual
colonies of cells are visible, they are picked and arrayed on 96-well plates. When the arrayed cells reach
~80% conuence, they are lysed directly using Direct-Lyse buer and the crude lysate is used to amplify
the genomic region containing the expected indel site using uorophore-labelled primers. e labelled
amplicons are resolved via capillary gel electrophoresis and successful mutants are identied by shis in
fragment size with respect to wildtype fragment. Putative knock-out clones are expanded and validated via
Sanger sequencing, quantitative RT-PCR and/or Western blot analysis.
www.nature.com/scientificreports/
3
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
encoding miR615-3p in MIR615 gene and ultimately chose two and four targets, respectively, which
we named sgATRX-E2, sgATRX-E4 and sgMIR615-3p-T1, -T2, -T3 and -T4. We obtained two sgRNA
plasmids targeting exon 4 of TP53 gene from Voorhoeves laboratory and these were named sgTP53-E4.1
and –E4.2. In order to ascertain their eciency in generating site-specic double-strand breaks, we tran-
siently transfected HEK293 or HCT116 cells with pCas9WT-2A-GFP plasmid (Addgene) together with
one of the sgRNA expression plasmid, and performed SURVEYOR mutation detection assay on puried
genomic DNA derived from the transfected cells. Of the two ATRX sgRNA, sgATRX-E4 displayed e-
cient targeted cleavage whereas sgATRX-E2 did not show any activity within the detection limit of the
SURVEYOR mutation detection assay (Fig.2a). In contrast, all sgRNAs targeting TP53 and MIR615-3p
displayed high cutting eciency (Fig. 2b,c). Hence, virtually all the sgRNAs tested were ecacious in
mediating the cleavage of DNA at specic targets in the genome.
Testing direct lysis condition. An important factor that renders a method compatible for
high-throughput genotyping of CRISPR/Cas9-induced mutant cells is its capacity for easy and fast
extraction of genomic DNA from individual clones. us, we were interested to test whether we could
perform direct cell lysis to obtain crude genomic DNA and utilize it to amplify specic regions via PCR.
We adapted a protocol originally reported for use with plant cells
20
and tested its ecacy in the extrac-
tion of intact genomic DNA directly from human cells in culture. We concocted the Direct-Lyse buer
and performed preliminary tests to compare the optimum dilution factor for use with cultured human
cells. Lysis of cells involves physical agitation of trypsinised cell suspension via trituration and subject-
ing the lysate to thermal cycling (see Materials and Methods for details). As compared to the original
report involving plant cells which used 2× dilution factor, we discovered that 0.5× Direct-Lyse buer
(10 mM Tris pH 8.0, 2.5 mM EDTA, 0.2 M NaCl, 0.15% SDS, 0.3% Tween-20) worked optimally for the
H1 human embryonic stem cells tested (Supplementary Figure S2a). We also found that the Direct-Lyse
Figure 2. SURVEYOR mutation detection assay to test for sgRNA targeting eciency. Specic regions of
ATRX (a), TP53 (b) and MIR615 (c) genes were targeted using CRISPR/Cas9 system (top) and the eciency
of the sgRNA used were examined via SURVEYOR assay (bottom). Top: e exons/coding region of each
gene are represented by blue boxes and are numbered accordingly. CRISPR/Cas9 target sequences are given
in green (protospacer) and red (protospacer adjacent motif, PAM) and the corresponding name of the
sgRNA are shown in bold. Middle: Grey arrows represent PCR primers used to amplify targeted regions.
Red arrows indicate the expected cleavage site of each sgRNA target and the expected sizes of the cleavage
product are given next to them. Bottom: SURVEYOR mutation detection assay results for each sgRNA
tested. Red stars indicate the cleavage products of the samples indicated above each lane and the numbers
at the bottom indicate estimated indel frequency. PC: positive control (G and C control from SURVEYOR
assay kit); EV: empty vector (without sgRNA expression cassette).
www.nature.com/scientificreports/
4
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
buer is superior to sodium hydroxide lysis buer (50 mM NaOH, 0.4 mM EDTA) which has been pre-
viously used for the same purpose
21
(Supplementary Figure S2b). Moreover, lysates are stable for months
if stored at lower than 20 °C, without any noticeable loss of quality as PCR template (data not shown).
Lastly, in order to ascertain its compatibility, we performed PCR analysis of genomic regions spanning
our genes of interest and found that they were amplied eciently and consistently (Supplementary
Figure S2c). Hence with the use of our homemade lysis buer, Direct-Lyse, we were able to obviate the
need to expand clones before extraction of genomic DNA, saving both time and cost in the process.
Genotyping clones with uorescent capillary gel electrophoresis. Fluorescent capillary gel elec-
trophoresis has previously been used with CRISPR/Cas9-mediated genome disruption to measure its
cutting eciency
22
. Taking into consideration its high sensitivity, we sought to assess this technique for its
ability to accurately and eciently genotype CRISPR/Cas9-induced insertion/deletion (indel) mutations
in the targeted cells. We examined ATRX, TP53 and MIR615 in near-diploid HCT116 colorectal carcinoma
cells for this purpose. We rst cultured HCT116 cells and co-transfected them with pCas9WT-2A-GFP
plasmid and individual sgRNA expression plasmids targeting ATRX exon 4 (sgATRX-E4), TP53 exon 4
(sgTP53E4.1 and sgTP53E4.2) and MIR615 region encoding miR615-3p (sgMIR615-3p-T1 to -T4). Two
days aer the transient transfection, we performed FACS to sort for GFP-positive cells and plated them
on 10-cm dish at 500–2,500 cells per dish. When individual colonies were large enough (about 2–3 mm
in diameter or 500–1,000 cells per colony), we picked and arrayed them on 96-well plates. When these
clones reached about 80% conuence, we lysed them with 0.5× Direct-Lyse buer and used the lysate
directly to amplify genomic regions spanning the sgRNA target sites with 6-FAM-labelled PCR primers.
Lysates of untransfected parental cells were subjected to amplication with HEX-labelled oligonucleotides
in parallel as a control. e 6-FAM- and HEX-labelled amplicons were diluted accordingly and mixed in
equal parts before being resolved via capillary gel electrophoresis (CGE) on Applied Biosystems 3500xL
Genetic Analyzer. With HEX-labelled wildtype DNA fragment as reference (green channel in CGE pro-
gram), targeted clones with indel mutations are expected to show shied peaks in the blue channel cor-
responding to the 6-FAM-labelled fragments due to their altered fragment size (see Figs1 and 3–6).
Since ATRX gene is found on the X chromosome and HCT116 cell line is derived from a male
patient
23
, its mutants are expected to show a single oset peak. As expected, using sgATRX-E4, 17 out
of 26 clones (65.4%) showed a shi in fragment size ranging from deletion or insertion of one to more
than 90 base pairs (Fig. 3a; Supplementary Figure S3a; Supplementary Table 1; and data not shown).
We sequenced several representative clones and found that indels reported by the uorescent capillary
electrophoresis method tally robustly with the sequencing results (Fig. 3b; and Supplementary Figure
S3b). In addition, these clones displayed strong attenuation of mRNA expression (Fig.3c) likely due to
nonsense-mediated decay, and complete ablation of protein expression as shown by western blot analysis
(Fig.3d). Of note, clone #1 had one nucleotide insertion (Fig.3a,b) and this small change in size was
readily and accurately detected by the uorescent capillary electrophoresis technique.
Next, we applied the uorescent capillary electrophoresis technique to examine the CRISPR/
Cas9-mediated targeting eciency of a bi-allelic gene in HCT116 cells—TP53. We anticipated the tech-
nique to rigorously discern between wildtype clones, heterozygous mutants (clones containing a wild-
type and a mutant allele), compound heterozygous mutants (clones with two mutant alleles which are
non-identical) and homozygous mutants (clones comprising of two identical mutant alleles). As expected,
targeting with sgTP53-E4.1 and sgTP53-E4.2 produced 4.7% and 5.3% heterozygous mutants, 34.9%
and 34.2% compound heterozygous mutants, and 39.5% and 44.7% homozygous mutants, respectively
(Supplementary Table 1). ese mutant clones harboured indels ranging from one to more than 200 base
pairs (Fig.4a; Supplementary Figure S3c; and data not shown). We picked a few clones for each sgRNA
and sequenced their genome to ascertain the accuracy of the uorescent capillary electrophoresis results
(Fig.4b; and Supplementary Figure S3d). As with ATR X gene targeting, the uorescent capillary electro-
phoresis method is very accurate in predicting shorter indels (< 30 bp) but tend to overestimate larger
indels (> 30 bp) (Fig.4a,b; Supplementary Figures S3c,3d; and data not shown). Nevertheless, qRT-PCR
and western blot analyses corroborate the authenticity of the genotype of these clones as predicted by
uorescent capillary electrophoresis (Fig.4c,d). Clones #2 and #4, being homozygous mutants and whose
indels were expected to cause a frameshi, displayed marked decrease in mRNA expression levels and
complete ablation of protein expression. In contrast, the heterozygous mutants showed a less dramatic
decrease in mRNA levels and slight decrease in protein levels. Interestingly, clones #1 and #6, which
are homozygous mutant and compound heterozygous mutant respectively, expressed p53 comparable to
parental HCT116 cells. is is likely due to in-frame indels harboured by these clones as supported by
the discernible shi in protein size.
Lastly, we interrogated the ecacy of the uorescent capillary electrophoresis technique to geno-
type a bi-allelic non-coding RNA gene, MIR615, specically the region encoding miR615-3p. Due to its
extremely short sequence, microRNA-encoding regions are expected to be easily knocked out. In accord-
ance to this, we found that 70.8% of clones examined were mutated on at least one allele (Supplementary
Table 1) and all the clones we analysed via qRT-PCR showed at least seven-fold decrease in miR615-3p
expression level (Fig.5b; and data not shown). Due to their complementary nature and mutual require-
ment for maturation, we expected that the mutation of miR615-3p will adversely aect the expression
of miR615-5p as well. In agreement to that, we observed a similarly drastic decrease in miR615-5p
www.nature.com/scientificreports/
5
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
expression in the three mutant clones examined (Fig.5d). We also investigated the eects of miR615-3p
mutation on a known target—AKT2
24
. All three clones showed signicant increase in AKT2 mRNA
expression (Fig. 5e) and protein level (Fig. 5f) when compared to wildtype parental cells, indicating
successful targeting of the MIR615 gene.
In addition to HCT116 cells, we have successfully genotyped A2780/CP, RKO and H1 cells targeted
using the same sgRNAs described above (data not shown). On top of that, we have also used the uores-
cent capillary gel electrophoresis technique to successfully genotype HT1080 cells targeted using sgRNAs
designed against human TERT, WLS and NCOR2 genes (data not shown). Collectively, these results
showed that the uorescent capillary electrophoresis technique enables robust genotyping of CRISPR/
Cas9-targeted clones, sensitive enough to dierentiate between wildtype clones, heterozygous mutants,
compound heterozygous mutants and homozygous mutants. More importantly, this technique accurately
measures the number of nucleotide(s) inserted or deleted and thus eectively reports any shi in frame
resulting in the formation of premature stop codon.
Additional benets of uorescent capillary gel electrophoresis. As reported above, our target-
ing of exon 2 of ATRX gene with sgATRX-E2 did not yield detectable levels of cleavage in the SURVEYOR
mutation detection assay (Fig.2a). However, we were interested to nd out whether the uorescent cap-
illary gel electrophoresis technique is able to circumvent the detection limitation of SURVEYOR muta-
tion detection assay and is sensitive enough to detect mutants at this target site. We performed similar
experiments as described above with sgATRX-E2 and identied mutants at a very low rate (3 out of 32;
Supplementary Table 1; and Supplementary Figure S4a). We sequenced these clones and found that they
indeed harboured indel mutations at the expected CRISPR/Cas9 cutting site (Supplementary Figure S4b).
Hence, we conclude that uorescence PCR-capillary gel electrophoresis is an ecient way of detecting
CRISPR/Cas9-targeted mutants even when the ecacy of the sgRNA is very low. is technique, thus,
Figure 3. Genotyping of ATR X-targeted clones via uorescent PCR-capillary gel electrophoresis.
(a) HCT116 cells targeted with sgATRX-E4 were genotyped using uorescent PCR-capillary gel
electrophoresis and three representative clones are shown. Blue peaks indicate fragments obtained from PCR
amplication of the region spanning the sgRNA target site in targeted cells using 6-FAM-labeled primers.
Green peaks indicate similar fragments but from wildtype parental HCT116 cells and thus act as an internal
size control. e numbers given in each plot represent the sizes of each peak (or fragment) and those in
parentheses are the calculated dierence in size (in base pairs) with respect to individual wildtype peaks.
(b) Sanger sequencing results for the individual clones. Wildtype sequence is shown in blue and the PAM
sequence in red. Inserted nucleotides are shown in green and underlined are the anking nucleotides in
the original sequence. Deleted nucleotides are shown as dashes (−). Quantitative RT-PCR (c) and Western
blot (d) analyses were performed to corroborate knockout status of the clones shown. Error bars represent
standard deviations of values from two independent experiments (n = 2). Asterisks represent signicantly
dierent (p < 0.05) expression levels as compared to the wildtype parental clone using one-tailed t-test.
www.nature.com/scientificreports/
6
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
possibly precludes the need for testing the ecacy of sgRNA prior to its use in targeting specic regions
of the genome since the screening platform we describe here is capable of handling a large sample size.
A potential problem that one may encounter during handling of large number of clones is the
cross-contamination between clones resulting in heterogeneity of cell population, especially during the
colony picking process to array individual clones onto 96-well plates from 10-cm dishes. Due to its high
sensitivity and resolving power, we anticipated the ability of uorescent capillary gel electrophoresis to
identify mixed population of cells. Indeed, we found an sgATRX-E2-targeted clone with an unusual
peak pattern. As ATRX is a mono-allelic gene in HCT116 background, both wildtype and mutant clones
are expected to display single peaks. However, clone #S6 displayed two peaks in its uorescent capil-
lary gel electrophoresis analysis—a smaller peak corresponding to wildtype sequence and a taller peak
with deleted sequence (Supplementary Figure S5a). Sanger sequencing conrmed the heterogeneity of
the cell population (Supplementary Figure S5b, top) and deletion of nucleotides in the mutant genome
Figure 4. Genotyping of TP53-targeted clones via uorescent PCR-capillary gel electrophoresis. TP53-
targeted HCT116 cells were genotyped using uorescent PCR-capillary gel electrophoresis (a) and several
representative clones are shown. Clones #1 to #3 and clones #4 to #6 were targeted by sgTP53-E4.1 and –
E4.2, respectively. (b) Sanger sequencing was performed to validate the indel mutations harboured by each
clones. Representation of mutations is similar to that in Fig.3b; in addition, orange nucleotides indicate
substituted bases. Quantitative RT-PCR (c) and Western blot (d) analyses were performed to conrm
the genotype of the clones. Error bars represent standard deviations of values from two independent
experiments (n = 2). Asterisks represent signicantly dierent (p < 0.05) expression levels as compared to the
wildtype parental clone using one-tailed t-test.
www.nature.com/scientificreports/
7
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
(Supplementary Figure S5b, bottom). As this deletion of 17 base pairs is expected to cause a frameshi,
full-length ATRX protein is expected to be lowly expressed. However, we observed ATRX levels com-
parable to untargeted HCT116 cells in this clone (Supplementary Figure S5c). Hence, this shows that
uorescent capillary gel electrophoresis enables the identication of heterogeneous population of cells by
the presence of aberrant peak pattern.
Fluorescent capillary gel electrophoresis technique is multiplexable. An advantage of the
CRISPR/Cas9 system over other programmable nucleases (ZFNs and TALENs) is its ability to simulta-
neously target multiple genes eciently
19
. erefore, we sought to test the robustness of the uorescent
capillary gel electrophoresis technique in multiplex detection of indel mutations in two dierent genes
in the same clone. We co-transfected HCT116 cells with sgATRX-E4 and sgTP53-E4.2 expression plas-
mids and pCas9-2A-GFP plasmid and sorted for GFP-positive cells. Individual clones were arrayed on
a 96-well plate and subjected to uorescent capillary gel electrophoresis analysis for both the targeted
genes. About 18% of clones examined displayed double-gene targeting (Supplementary Table 1). We
sequenced some of these clones and found excellent agreement between the uorescent capillary gel
electrophoresis results and the sequencing results. Figure 6a,b show the results of two representative
clones. We performed RT-PCR and western blot analyses and found that the results from these exper-
iments corroborate the double knockout status of these two clones (Fig.6c,d). ese results show that
the uorescent capillary gel electrophoresis technique supports multiplex genome targeting via CRISPR/
Cas9 system, enabling fast and accurate screening of multiply targeted cells.
Figure 5. Genotyping of MIR615-3p-targeted clones via uorescent PCR-capillary gel electrophoresis.
MIR615 gene (specically the region encoding miR615-3p) was targeted using CRISPR/Cas9 system in
HCT116 cells. e genotype of individual targeted clones was determined via uorescent PCR coupled with
capillary gel electrophoresis (a) and veried using Sanger sequencing (b) and quantitative RT-PCR (c).
(d) e expression of miR615-5p was evaluated using quantitative RT-PCR. In addition, the expression of a
known target of miR615-3p, AKT2, was examined using quantitative RT-PCR (e) and Western blot analysis
(f). All symbols and representations are identical to those in Fig.3. Error bars represent standard deviations
of values from two independent experiments (n = 2). Asterisks represent signicantly dierent (p < 0.05)
expression levels as compared to the wildtype parental clone using one-tailed t-test.
www.nature.com/scientificreports/
8
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
Discussion
Inherent technical limitations and high cost of present genotyping methods render these techniques not
ideal for high-throughput detection of CRISPR/Cas9-induced indel mutants in targeted cells. Whilst
there have been various recent reports on the improvements of these current genotyping methods,
many of them tend to focus on nuclease-mediated mutations at the cell population level and are not
very informative in that they do not report the occurrence of frameshi in the sequence of individual
clones. For example, the high resolution melting analysis (HRMA) protocol described by omas and
colleagues
9
is able to distinguish distinct mutant alleles based on their intricate DNA melting proles.
However, this technique is not sensitive enough to report the presence or absence of frameshi in these
alleles and it thus necessitates a second round of screening for a knockout clone. Another example of
an advancement in the eld is the introduction of TIDE (Tracking of Indels by DEcomposition) by
Brinkman and co-workers
25
which uses a decomposition algorithm to analyse capillary sequencing traces
to determine the major induced mutations in a mixed population of cells. is method is extremely
useful for the assessment of the cutting eciency of programmable nucleases but is not compatible for
high-throughput screening due to its high cost engendered by the sequencing step.
Recently, yet another two more genotyping techniques have been described. Yu and colleagues
reported a PCR-based method which utilizes a primer overlapping the putative indel site to determine
the presence of CRISPR/Cas9-induced mutations at these sites
26
. In addition, Zhu and his team intro-
duced a polyacrylamide gel electrophoresis (PAGE)-based technique to distinguish wildtype and mutant
sequences on the basis of dierential mobility rate of homoduplex and heteroduplex DNA formed from
Figure 6. Genotyping of multiplex targeted clones via uorescent PCR-capillary gel electrophoresis.
e genome of HCT116 cells was targeted using sgATRX-E4 and sgTP53-E4.2 and clones were genotyped
via uorescent PCR-capillary gel electrophoresis (a) at both loci (exon 4 of ATRX gene and exon 4 of TP53
gene). Two double knockout clones are shown. Sanger sequencing (b), quantitative RT-PCR (c) and Western
blot (d) analyses were performed to validate the uorescent PCR-capillary gel electrophoresis results and
double knockout status of the two clones. All symbols and representations are identical to those in Figs3
and 4. Error bars represent standard deviations of values from two independent experiments (n = 2).
Asterisks represent signicantly dierent (p < 0.05) expression levels as compared to the wildtype parental
clone using one-tailed t-test.
www.nature.com/scientificreports/
9
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
the annealing of denatured DNA amplicons
27
. Whilst these techniques are advantageous especially in
circumventing certain drawbacks inherent in the other genotyping techniques mentioned above, they
too suer from a certain drawback common to most other genotyping techniques: they do not report
the presence or absence of frameshi mediated by CRISPR/Cas9 DNA cleavage, which the technique we
have described herein is capable of.
Taking all these factors into consideration, we therefore sought to assess the possibility of employing
uorescence capillary electrophoresis for the very purpose of genotyping CRISPR/Cas9-induced mutants.
is technique has previously been used to genotype populations of nuclease-mediated cells mainly to
determine the cleavage ecacy of the nuclease system
22,28
. Here, we show that this time- and cost-saving
technique is able to eciently and accurately detect indel mutations in CRISPR/Cas9-targeted sequences
in single clones and hence eectively identify knockout clones. We targeted three genes on three dier-
ent chromosomes and found that the uorescent capillary electrophoresis technique could genotype the
targeted clones highly accurately.
In addition to its amenability for high-throughput screening, the uorescent capillary electrophoresis
technique has several other advantages. Firstly, this technique reports accurately the number of nucleo-
tides inserted or deleted at the non-homologous end joining repair-mediated DSB site. is essentially
obviates the need for sequencing analysis to determine the occurrence of a frameshi which may poten-
tiate the ablation of expression of the targeted gene. is advantage stems from the high sensitivity and
resolving power of this technique and is evident in its ability to identify mutants with indel of just one
base pair. Secondly, this technique is able to support multiplex targeting of genes. As a result, users stand
to save time and cost by performing genotyping analysis in a single 96-well plate using dierent primers
for the various targeted regions. irdly, this uorescent capillary electrophoresis protocol allows for the
identication of erroneous heterogeneous cell populations which is crucial when studying the eects of
a knocked out gene where homogeneity of clones is key.
Two factors are important for this techniques amenability for high-throughput screening. First, the
use of our direct lysis buer, Direct-Lyse, allows for fast and reproducible extraction of crude genomic
DNA straight from cells in culture. Furthermore, the resultant lysates are stable for months if kept at
lower than 20 °C. We found that the Direct-Lyse buer is comparably ecacious as other commercially
available direct lysis buer and that it is more cost-eective in the long term. Secondly, as compared
to Sanger sequencing which is currently the most widely used method of genotyping targeted clones,
the use of uorescent capillary electrophoresis is markedly cheaper and therefore eectively scalable.
e cost of genotyping a targeted clone using Sanger sequencing is approximately US$4-8 per reaction,
including the DNA purication step and the actual sequencing reaction; and depending on the method
of purication and the number of samples handled at a time (throughput). In comparison, the capillary
gel electrophoresis technique described herein costs about US$3-4 per reaction. is amounts to a cost
savings of between 25% and 50%. In addition, the uorescent PCR-capillary gel electrophoresis technique
requires less than half the time to obtain the genotyping results as compared to Sanger sequencing which
typically requires third-party services.
ere are several caveats associated with the uorescent capillary gel electrophoresis technique we
have described. For one, this technique is not adequately accurate at reporting mutations with indels of
more than 30 base pairs. It tends to overestimate these mutations. However, we nd that this may not
necessarily be a problem because most of the mutations we have seen in our study involve indels shorter
than that threshold. Another limitation of this technique is its inability to detect base substitutions or
single nucleotide polymorphisms (SNPs) in the targeted genome which may result in the formation
of missense or nonsense mutations. Whilst this type of mutation may result in the decrement or even
abolition of gene expression, it has been reported that non-homologous end joining-mediated repair of
DSBs rarely results in the substitution of bases in the nucleotides found at the break sites
11
. irdly, our
protocol requires the use of a genetic analyser equipment and analysis soware which may not be readily
available. However, we found that this part of the experiment can be outsourced to companies which pro-
vides sequencing services and at a cost comparable to when done on our own. Lastly, since our targeting
strategy involves the transfection of two separate plasmids for the two components of the CRISPR/Cas9
genome-targeting system—Cas9 and sgRNA—the targeting eciencies we report are not very high. We
expect that the use of a bicistronic expression plasmid containing both CRISPR components will greatly
improve targeting eciency, coupled with the enrichment step of sorting for GFP-positive cells from the
expression of Cas9-2A-GFP cassette, as shown previously
29
.
us, we have described a high-throughput method of genotyping CRISPR/Cas9-mediated mutation
of human cells using uorescent capillary gel electrophoresis. We anticipate this technique to be appli-
cable to a wide range of targets and model systems. Moreover, with its amenability for high-throughput
screening, fast and ecient genotyping of mutants or knockout clones may no longer be a bottleneck
for gene targeting studies.
Methods
Reagents. All oligonucleotides used in our study were procured from Sigma-Aldrich Co. and Life
Technologies and are listed in Supplementary Table 2. Oligonucleotides labelled with 6-FAM or HEX were
covalently bonded with the uorophore at the 5 end. e following antibodies were used in our study:
rabbit polyclonal anti-ATRX antibody from Santa Cruz Biotechnology (sc-15408); mouse monoclonal
www.nature.com/scientificreports/
10
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
anti-p53 antibody from Santa Cruz Biotechnology (sc-126); mouse monoclonal anti-β -actin antibody
from Sigma-Aldrich Co. (A1978); rabbit monoclonal anti-Akt2 antibody from Cell Signaling Technology
(#3063); sheep anti-mouse HRP-conjugated secondary antibody from Jackson ImmunoResearch; and
goat anti-rabbit HRP-conjugated secondary antibody from Jackson ImmunoResearch. All antibodies
were diluted in 5% milk in TBST buer.
Selecting sgRNA targets. Exon 2 and 4 of human ATRX gene and miR615-3p-encoding region
of MIR615 were subjected to potential sgRNA target search using the online soware created by Feng
Zhangs group (http://tools.genome-engineering.org)
19
. e very top hits were chosen and used for the
experiments hence described. e nucleotide sequence of these sgRNA targets are given in Fig.2 and
Supplementary Table 3.
Plasmid construction. e pCas9-GFP plasmid was obtained from Addgene (#44719). Both TP53
sgRNA (sgTP53-E4.1 and -E4.2) expression plasmids were gis from Voorhoeves laboratory. ATRX sgR-
NAs (sgATRX-E2 and -E4) and MIR615 sgRNA (sgMIR615-3p-T1, -T2, -T3 and -T4) expression plasmids
were constructed by cloning U6 promoter-sgRNA-TTT [(as used by Mali et al.
30
] into pBluescript SK (-)
vector (see Supplementary Figure S1 for schematic). Briey, the sgRNA expression cassettes were divided
into twelve overlapping oligonucleotides and assembled into a contiguous DNA fragment via two rounds
of PCR. e inserts and pBSK(−) vector were then digested with SpeI and PstI (Homann-La Roche),
resolved on a 1% agarose gel, puried from excised gel using Wizard SV Gel and PCR Clean-Up System
(Promega) and ligated using T4 DNA Ligase (New England Biolabs). DH5α cells were transformed with
the ligation product and plated on Luria Bertani agar supplemented with ampicillin, IPTG and X-gal.
Individual white colonies were picked and cultured, and corresponding plasmids extracted using Wizard
Plus SV Minipreps DNA Purication System (Promega) and sequenced to ensure correct insert sequence.
Selecting lysis conditions. Comparisons of direct lysis eciency between 1× and 0.5× Direct-Lys e
buer, and between 0.5× Direct-Lyse buer and sodium hydroxide lysis buer were performed using
H1 human embryonic stem cell line and A2780/CP human ovarian carcinoma cell line, respectively.
Monolayer of cells on a 96-well plate were trypsinised using 25 μ l of 0.25% trypsin/EDTA (without phe-
nol red) at 37 °C for 10–15 minutes and 5 μ l of the cell suspension was added to 10 μ l of one of the lysis
buer: 1× Direct-Lyse buer (20 mM Tris pH 8.0, 5 mM EDTA, 0.4 M NaCl, 0.3% SDS, 0.6% Tween-20);
0.5× Direct-Lyse buer (10 mM Tris pH 8.0, 2.5 mM EDTA, 0.2 M NaCl, 0.15% SDS, 0.3% Tween-20);
or 2× sodium hydroxide lysis buer (50 mM NaOH, 0.4 mM EDTA). Direct-Lyse lysates were subjected
to a series of thermal cycling as follows: 65 °C for 30 s, 8 °C for 30 s, 65 °C for 1.5 min, 97 °C for 3 min,
8 °C for 1 min, 65 °C for 3 min, 97 °C for 1 min, 65 °C for 1 min, and 80 °C for 10 min. Sodium hydroxide
lysates were heated at 100 °C for 20–30 min aer which 10 μ l of 40 mM Tris-HCl (pH 5.0) was added to
neutralize the pH. Finally, each of these lysates was diluted with 40 μ l water before being subjected to
PCR analysis. To this end, 3 μ l of the diluted lysates together with Platinum Taq DNA polymerase (Life
Technologies) were used to amplify various genomic regions using primers listed in Supplementary Table
2 following manufacturers recommendation. Amplicons were resolved on 1% agarose gel, stained with
ethidium bromide and imaged using ChemiDoc XRS System (Bio-Rad).
Cell culture and transfection conditions. All CRISPR/Cas9 targeting experiments were performed
with HCT116 human colorectal carcinoma cell line or HEK293 human embryonic kidney cell line (both
from ATCC). Cells were maintained in normal growth media (DMEM containing 10% FBS and 1%
penicillin/streptomycin) except prior to and during DNA transfection. For targeting of individual genes,
cells were seeded at 2 × 10
5
cells per well of a 6-well plate in antibiotic-free media (DMEM with 10%
FBS) and co-transfected with pCas9-GFP and one (for single gene targeting) or two (for multiplexed
targeting) of the sgRNA expression plasmid or empty vector [pBluescript SK (−)] using Lipofectamine
2000 (Life Technologies) as per manufacturer’s recommendation. We co-transfected the cells with 0.5 μ g
pCas9-GFP plasmid and 3.5 μ g sgRNA expression plasmid using 10 μ l Lipofectamine 2000 transfec-
tion reagent for about 16 hours and changed to fresh media thereaer. For simultaneous targeting of
TP53 and ATRX genes, cells were seeded at 2 × 10
6
cells per 10-cm dish and co-transfected with 1 μ g
pCas9-GFP plasmid and 11.5 μ g each of pBSK-sgTP53-E4.2 and pBSK-sgATRX-E4 plasmids using 60 μ l
Lipofectamine 2000 transfection reagent.
Fluorescence-activated cell sorting. HCT116 cells that were used for SURVEYOR assay and uo-
rescent capillary gel electrophoresis were subjected to cell sorting. Two days aer transient transfection,
cells were sorted using FACSAria III (BD Biosciences). GFP-positive clones were collected and plated
accordingly. For purpose of performing SURVEYOR assay, 1 × 10
4
cells were plated on 6-well plate and
maintained until conuent; whereas for isolation of individual clones, cells were plated on 10-cm dish at
500 to 2,500 cells per dish until individual colonies were visible.
SURVEYOR mutation detection assay. HEK293 and HCT116 cells were used to perform
SURVEYOR mutation detection assay with ATRX and TP53 sgRNAs, respectively. e latter were sub-
jected to FACS to enrich for GFP-positive cells prior to SURVEYOR assay, whilst crude transfected cells
www.nature.com/scientificreports/
11
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
were used for the former. When cells reached conuence, they were harvested and their genomic DNA
extracted using Gentra Puregene Cell Kit (QIAGEN). 100 ng of genomic DNA was used to amplify
the regions spanning CRISPR/Cas9 cut sites using specic primers (see Supplementary Table 2) and
Platinum Taq DNA Polymerase High Fidelity (Life Technologies) as per manufacturers instruction.
Control C and G fragments (Transgenomic, Inc.) were also amplied in parallel. e PCR products were
puried using QIAquick PCR Purication Kit (QIAGEN) and diluted to 100 ng/μ l. 2.15 μ g of the puried
PCR product were added to 1× Platinum Taq DNA Polymerase High Fidelity buer (Life Technologies)
and 2 mM magnesium sulphate, and subjected to a denaturation and re-annealing program to facilitate
the formation of mismatched fragments: 95 °C for 10 minutes, ramping from 95 °C to 85 °C at 2 °C/s,
85 °C for 1 minute, ramping from 85 °C to 75 °C at 0.3 °C/s, 75 °C for 1 minute, ramping from 75 °C to
65 °C at 0.3 °C/s, 65 °C for 1 minute, ramping from 65 °C to 55 °C at 0.3 °C/s, 55 °C for 1 minute, ramping
from 55 °C to 45 °C at 0.3 °C/s, 45 °C for 1 minute, ramping from 45 °C to 35 °C at 0.3 °C/s, 35 °C for
1 minute, ramping from 35 °C to 25 °C at 0.3 °C/s, and 25 °C for 1 minute. Next, 15 μ l of the re-annealed
DNA were added to 0.5 μ l of Enhancer S and 0.5 μ l of Nuclease S (Transgenomic, Inc.) and incubated at
42 °C for an hour. e digestion products were resolved on a 10% polyacrylamide/TBE gel and stained
with ethidium bromide for 15 minutes. e gels were visualized using ChemiDoc XRS System (Bio-Rad
Laboratories, Inc.) and the fragments quantied using ImageJ. Indels were calculated using the follow-
ing formula, where a is the intensity of the undigested fragment, and b and c are the intensities of the
cleavage products.
(%)= ×
+
++
bc
abc
Indel100 1
Direct lysis of cultured cells. Around 10–12 days aer plating of GFP-positive FACS-sorted cells
onto 10-cm dish, individual colonies were picked and transferred to 96-well plates and maintained.
When cells in the 96-well plate reached about 80% conuence, they were subjected to direct lysis using
0.5× Direct-Lyse buer (10 mM Tris pH 8.0, 2.5 mM EDTA, 0.2 M NaCl, 0.15% SDS, 0.3% Tween-20).
Growth media was removed from the wells and cells were trypsinised with 0.05% trypsin-EDTA (without
phenol red) for 7 minutes. About 5 μ l of the cell suspension were added to 10 μ l of Direct-Lyse lysis buer
in a 96-well PCR plate. e cells were then subjected to a series of heating and cooling to ensure complete
lysis: 65 °C for 30s, 8 °C for 30s, 65 °C for 1.5 min, 97 °C for 3 min, 8 °C for 1 min, 65 °C for 3 min, 97 °C
for 1 min, 65 °C for 1 min, and 80 °C for 10 min. e lysates were then diluted with 40 μ l of water and
subjected to diagnostic PCR analyses.
Fluorescent PCR. Diluted lysates were used to amplify regions spanning CRISPR/Cas9 targeting
sequence in a high-throughput manner on a 96-well plate to identify clones harbouring indel mutations.
Platinum Taq DNA Polymerase (Life Technologies) was used for this purpose, together with primer
pairs with 5 modication on the forward primers (Supplementary Table 2). Lysates of parental wildtype
HCT116 cells were subjected to PCR amplication with HEX-labelled forward primers, whereas those of
the CRISPR/Cas9-targeted clones were performed with 6-FAM-labeled primers. A part of these labelled
amplicons were then resolved on a 1% agarose gel to estimate their relative amounts and the remaining
were subsequently diluted to roughly standardize their concentration, before being subjected to capillary
gel electrophoresis. Typically, the amplicons were diluted 30× with water and those which showed lower
amounts were diluted by a lower factor.
Capillary gel electrophoresis. Diluted uorescent PCR amplicons from wildtype and targeted cells
were mixed 1:1 and 1 μ l of the mixture was added to 8.7 μ l Hi-Di Formamide and 0.3 μ l GeneScan 500
LIZ dye Size Standard (Life Technologies) on a 96-well plate. As per manufacturers instruction, the
resultant mixture was heated at 95 °C for 3 minutes and subsequently cooled on ice for 3 minutes. ese
samples were resolved via capillary gel electrophoresis on a 3500xL Genetic Analyzer (Life Technologies).
e details of the instrument protocol and the size-calling protocol used are as follow—application type:
fragment; capillary length: 50 cm; polymer: POP7; dye set: G5; run voltage: 19.5 kV; pre-run voltage: 15
kV; injection voltage: 1.6 kV; run time: 1330 s; pre-run time: 180 s; injection time: 15 s; data delay: 1 s;
size standard: GS500(-250)LIZ; size-caller: SizeCaller v1.10. Results were analysed using Gene Mapper
5 soware (Life Technologies). Insertions or deletions of nucleotides at corresponding CRISPR/Cas9
cleavage sites were estimated by calculating the dierence in fragment sizes as provided by the analysis
soware.
Sequencing of targeted clones. Clones predicted to contain indel mutations due to the CRISPR/
Cas9 targeting were isolated and the regions spanning these mutations were amplied via PCR using prim-
ers listed in Supplementary Table 2 and Platinum Taq DNA Polymerase High Fidelity (Life Technologies)
following manufacturers recommendation. e resultant amplicons were sent for sequencing with
their respective 5 PCR primers. For compound heterozygous mutants, we cloned the PCR amplicons
into pCRII-Blunt-TOPO vector using the Zero Blunt TOPO PCR Cloning Kit (Life Technologies) and
www.nature.com/scientificreports/
12
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
transformed DH5α cells with the ligation product. Transformants were picked and cultured overnight
and their plasmids were extracted using Wizard Plus SV Minipreps DNA Purication System (Promega)
and sent for sequencing with their respective 5 PCR primers.
Quantitative RT-PCR. Clones were cultured on 6-well plates and harvested when they reached conu-
ence. Total RNA was extracted from the clones using NucleoSpin RNA kit (Macherey-Nagel GmbH & Co.)
and their concentration measured using Nanodrop 2000 (ermo Fisher Scientic Inc.). e RNA sam-
ples were diluted to 100 ng/μ l and 300 ng of each were used to perform quantitative RT-PCR using
KAPA SYBR FAST Bio-Rad iCycler (Kapa Biosystems) and CFX96 Touch Real-Time PCR Detection
System (Bio-Rad Laboratories, Inc.) in technical triplicates and in biological duplicates. Relative mRNA
expression of individual clones was calculated using the comparative C
T
(Δ Δ C
T
) method, normalized to
β -actin and relative to wildtype parental cells. Mean fold change, standard deviation, and p-value (using
one-tailed t-test) were calculated using Microso Excel.
Western blot analyses. Clones were cultured on 10-cm dish and harvested when they reached con-
uence. Cells were lysed with Lysis 250 Buer (50 mM Tris-HCl pH 7.4, 250 mM NaCl, 5 mM EDTA pH
8.0, 0.1% NP-40, 50 mM NaF) supplemented with Complete Protease Inhibitor (Homann-La Roche),
by a series of freeze-thaw cycles. Crude lysates were centrifuged to remove cell debris and the claried
lysates were used directly for immunoblotting (TP53 clones, and MIR615-3p clones probing for Akt2)
or for immunoprecipitation prior to blotting (ATRX clones). About 75 μ g and 8 mg of total lysates were
used for immunoblotting and immunoprecipitation, respectively. To the lysates of ATR X-targeted clones,
1 μ g of anti-ATRX antibody (sc-15408; Santa Cruz Biotechnology, Inc.) was added and incubated for
an hour before the addition of ~50 μ l GammaBind G Sepharose beads (GE Healthcare). e immuno-
precipitation mixtures were incubated overnight at 4 °C with constant rotation and the protein-bound
beads were subsequently washed three times with Lysis 250 Buer. Next, the beads were heated at 95 °C
for 10 minutes and centrifuged at top speed for a minute before the resultant supernatants were resolved
on a 6% polyacrylamide gel. Similarly for TP53- and MIR615-3p-targeted clones, samples were mixed
with SDS loading dye and heated at 95 °C for 10 minutes and centrifuged at top speed for a minute and
loaded onto a 10% polyacrylamide gel. Resolved proteins were transferred onto PVDF membranes at
400 mA for 1.5 to 2 hours. e membranes were blocked with milk (5% in TBST) for an hour at room
temperature before incubated overnight with primary antibody. ey were then washed with TBST for
at least 30 minutes and incubated at room temperature with secondary antibody for an hour. Lastly, the
membranes were washed with TBST and subjected to ECL detection. Anti-ATRX antibody (Santa Cruz)
was diluted 1:200; anti-p53 antibody (Abcam) was diluted 1:1,000; anti-Akt2 antibody (Cell Signaling)
was diluted 1:1,000; and anti-β -actin antibody was diluted 1:40,000. Sheep anti-mouse HRP-conjugated
antibody (1:20,000; Jackson ImmunoResearch) and goat anti-rabbit HRP-conjugated antibody (1:20,000;
Jackson ImmunoResearch) were used as secondary antibodies.
References
1. Perez, E. E. et al. Establishment of HIV-1 resistance in CD4+ T cells by genome editing using zinc-nger nucleases. Nat.
Biotechnol. 26, 808–816 (2008).
2. Santiago, Y. et al. Targeted gene nocout in mammalian cells by using engineered zinc-nger nucleases. Proc. Natl. Acad. Sci.
USA 105, 5809–5814 (2008).
3. Sung, Y. H. et al. nocout mice created by TALEN-mediated gene targeting. Nat. Biotechnol. 31, 23–24 (2013).
4. Shalem, O. et al. Genome-scale CISP-Cas9 nocout screening in human cells. Science 343, 84–87 (2014).
5. Zhou, Y. et al. High-throughput screening of a CISP/Cas9 library for functional genomics in human cells. Nature 509,
487–491 (2014).
6. Guschin, D. Y. et al. A rapid and general assay for monitoring endogenous gene modication. Methods Mol. Biol. 649, 247–256
(2010).
7. im, H. J., Lee, H. J., im, H., Cho, S. W. & im, J. S. Targeted genome editing in human cells with zinc nger nucleases
constructed via modular assembly. Genome es. 19, 1279–1288 (2009).
8. Dahlem, T. J. et al. Simple methods for generating and detecting locus-specic mutations induced with TALENs in the zebrash
genome. PLoS Genet. 8, e1002861 (2012).
9. omas, H. ., Percival, S. M., Yoder, B. . & Parant, J. M. High-throughput genome editing and phenotyping facilitated by high
resolution melting curve analysis. PLoS One 9, e114632 (2014).
10. Urnov, F. D. et al. Highly ecient endogenous human gene correction using designed zinc-nger nucleases. Nature 435, 646–651
(2005).
11. im, J. M., im, D., im, S. & im, J. S. Genotyping with CISP-Cas-derived NA-guided endonucleases. Nat. Commun. 5,
3157 (2014).
12. Gibbons, . J., Suthers, G. ., Wilie, A. O., Bucle, V. J. & Higgs, D. . X-lined alpha-thalassemia/mental retardation (AT-X)
syndrome: localization to Xq12-q21.31 by X inactivation and linage analysis. Am. J. Hum. Genet. 51, 1136–1149 (1992).
13. Xue, Y. et al. e ATX syndrome protein forms a chromatin-remodeling complex with Daxx and localizes in promyelocytic
leuemia nuclear bodies. Proc. Natl. Acad. Sci. USA 100, 10635–10640 (2003).
14. Livingstone, L. . et al. Altered cell cycle arrest and gene amplication potential accompany loss of wild-type p53. Cell 70,
923–935 (1992).
15. Michalovitz, D., Halevy, O. & Oren, M. Conditional inhibition of transformation and of cell proliferation by a temperature-
sensitive mutant of p53. Cell 62, 671–680 (1990).
16. Yonish-ouach, E. et al. Wild-type p53 induces apoptosis of myeloid leuaemic cells that is inhibited by interleuin-6. Nature
352, 345–347 (1991).
www.nature.com/scientificreports/
13
SCIENTIFIC RepoRts | 5:15587 | DOI: 10.1038/srep15587
17. Isobe, M., Emanuel, B. S., Givol, D., Oren, M. & Croce, C. M. Localization of gene for human p53 tumour antigen to band 17p13.
Nature 320, 84–85 (1986).
18. Hulf, T. et al. Discovery pipeline for epigenetically deregulated miNAs in cancer: integration of primary miNA transcription.
BMC Genomics 12, 54 (2011).
19. an, F. A. et al. Genome engineering using the CISP-Cas9 system. Nat. Protoc. 8, 2281–2308 (2013).
20. Yang, Y. G., im, J. Y., Soh, M. S. & im, D. S. A simple and rapid gene amplication from Arabidopsis leaves using AnyDirect
system. J. Biochem. Mol. Biol. 40, 444–447 (2007).
21. Truett, G. E. et al. Preparation of PC-quality mouse genomic DNA with hot sodium hydroxide and tris (HotSHOT). Biotechniques
29, 52, 54 (2000).
22. Cho, S. W. et al. Analysis of o-target eects of CISP/Cas-derived NA-guided endonucleases and nicases. Genome es. 24,
132–141 (2014).
23. Brattain, M. G., Fine, W. D., haled, F. M., ompson, J. & Brattain, D. E. Heterogeneity of malignant cells from a human colonic
carcinoma. Cancer es. 41, 1751–1756 (1981).
24. Bai, Y., Li, J., Li, J., Liu, Y. & Zhang, B. Mi-615 inhibited cell proliferation and cell cycle of human breast cancer cells by
suppressing of AT2 expression. Int. J. Clin. Exp. Med. 8, 3801–3808 (2015).
25. Brinman, E. ., Chen, T., Amendola, M. & van Steensel, B. Easy quantitative assessment of genome editing by sequence trace
decomposition. Nucleic Acids es. 42, e168 (2014).
26. Yu, C., Zhang, Y., Yao, S. & Wei, Y. A PC based protocol for detecting indel mutations induced by TALENs and CISP/Cas9
in zebrash. PLoS One 9, e98282 (2014).
27. Zhu, X. et al. An ecient genotyping method for genome-modied animals and human cells generated with CISP/Cas9
system. Sci. ep. 4, 6420 (2014).
28. im, H. et al. Surrogate reporters for enrichment of cells with nuclease-induced mutations. Nat. Methods 8, 941–943 (2011).
29. Gocezade, J., Siensi, G. & Duche, P. Ecient CISP/Cas9 plasmids for rapid and versatile genome editing in Drosophila.
G3 (Bethesda) 4, 2279–2282 (2014).
30. Mali, P. et al. NA-guided human genome engineering via Cas9. Science 339, 823–826 (2013).
Acknowledgements
We would like to thank Dr. Mathijs Voorhoeve for sharing some plasmids, and Dr. Koji Itahana and
Dr. Marc Fivaz for sharing some antibodies. Our gratitude also goes to Ms. Charlene Foong for helping
with the FACS experiments, and Ms. Tan Shi Min and Dr. Zhao Yi for helping with the uorescent
capillary gel electrophoresis experiments.
Author Contributions
S.L. conceived the experiments. M.K.R., T.Y. and A.M.S.C. conducted the experiments. S.L., M.K.R., T.Y.
and A.M.S.C. analysed the results. C.T.H.C provided the infrastructure for conducting the experiments.
All authors reviewed the manuscript.
Additional Information
Supplementary information accompanies this paper at http://www.nature.com/srep
Competing nancial interests: e authors declare no competing nancial interests.
How to cite this article: Ramlee, M. K. et al. High-throughput genotyping of CRISPR/Cas9-mediated
mutants using uorescent PCR-capillary gel electrophoresis. Sci. Rep. 5, 15587; doi: 10.1038/srep15587
(2015).
is work is licensed under a Creative Commons Attribution 4.0 International License. e
images or other third party material in this article are included in the article’s Creative Com-
mons license, unless indicated otherwise in the credit line; if the material is not included under the
Creative Commons license, users will need to obtain permission from the license holder to reproduce
the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
... The mutation detection on exon 7 was analyzed via the fluorescent PCR-capillary gel electrophoresis technique [33,36]. The PCR was performed using 6-FAM-labeled forward primers. ...
... After the PCR, the samples were processed as described previously [33] and the fluorescent PCR-capillary gel electrophoresis technique was performed using a Gen-eScan TM 500 LIZ Size Standard (Applied Biosystem, Thermo Fisher, Waltham, USA) and a 3500 Genetic Analyzer (Applied Biosystems, Thermo Fisher, Waltham, USA). The details of the instrumental protocol were similar to those previously described [36] When the peak obtained via capillary electrophoresis was the same size as the control peak, the samples were considered to be WT, whereas other peaks of different sizes with respect to the control peak were considered to be KO. When more than two peaks were detected in a sample, it was evaluated as mosaic. ...
Article
Full-text available
The generation of genetically modified pigs has an important impact thanks its applications in basic research, biomedicine, and meat production. Cloning was the first technique used for this production, although easier and cheaper methods were developed, such as the microinjection, electroporation, or lipofection of oocytes and zygotes. In this study, we analyzed the production of genetically modified embryos via lipofection of zona-pellucida-intact oocytes using LipofectamineTM CRISPRMAXTM Cas9 in comparison with the electroporation method. Two factors were evaluated: (i) the increment in the concentration of the lipofectamine–ribonucleoprotein complexes (LRNPC) (5% vs. 10%) and (ii) the concentration of ribonucleoprotein within the complexes (1xRNP vs. 2xRNP). We found that the increment in the concentration of the LRNPC had a detrimental effect on embryo development and a subsequent effect on the number of mutant embryos. The 5% group had a similar mutant blastocyst rate to the electroporation method (5.52% and 6.38%, respectively, p > 0.05). The increment in the concentration of the ribonucleoprotein inside the complexes had no effect on the blastocyst rate and mutation rate, with the mutant blastocyst rate being similar in both the 1xRNP and 2xRNP lipofection groups and the electroporation group (1.75%, 3.60%, and 3.57%, respectively, p > 0.05). Here, we showed that it is possible to produce knock-out embryos via lipofection of zona-pellucida-intact porcine oocytes with similar efficiencies as with electroporation, although more optimization is needed, mainly in terms of the use of more efficient vesicles for encapsulation with different compositions.
... Mutation detection was performed by a fluorescent PCR-capillary gel electrophoresis technique [14,50]. PCR was carried out using 6-FAM-labeled forward primers. ...
... The sample was incubated (3 min at 95 • C), immediately chilled on ice for 2 min, and analyzed by capillary gel electrophoresis on a 3500 Genetic Analyzer (Applied Biosystems, Thermofisher, Waltham, MA, USA). The details of the instrumental protocol were similar to that previously described [50] Samples were considered to be wild type when the peak obtained by capillary electrophoresis was the same size as the control peak. Other peaks of different sizes with respect to the control peak were considered to be mutant, and when more than two peaks were detected in a sample it was considered as mosaic. ...
Article
Full-text available
Mosaicism is the most important limitation for one-step gene editing in embryos by CRISPR/Cas9 because cuts and repairs sometimes take place after the first DNA replication of the zygote. To try to minimize the risk of mosaicism, in this study a reversible DNA replication inhibitor was used after the release of CRISPR/Cas9 in the cell. There is no previous information on the use of aphidicolin in porcine embryos, so the reversible inhibition of DNA replication and the effect on embryo development of different concentrations of this drug was first evaluated. The effect of incubation with aphidicolin was tested with CRISPR/Cas9 at different concentrations and different delivery methodologies. As a result, the reversible inhibition of DNA replication was observed, and it was concentration dependent. An optimal concentration of 0.5 �M was established and used for subsequent experiments. Following the use of this drug with CRISPR/Cas9, a halving of mosaicism was observed together with a detrimental effect on embryo evelopment. In conclusion, the use of reversible inhibition of DNA replication offers a way to reduce mosaicism. Nevertheless, due to the reduction in embryo development, it would be necessary to reach a balance for its use to be feasible.
... The fluorescent PCR-capillary gel electrophoresis technique is accurate enough to differentiate one base-pair differences between alleles, so this technique can indicate the presence or absence of a frameshift in the coding sequence of the gene (Ramlee et al. 2017). This technique involves performing a conventional PCR with one primer labelled with a fluorochrome and later the sample is run in a capillary gel electrophoresis device linked to a genetic analyser (Ramlee et al. 2015). In order to discriminate the wild type allele in comparison to the other alleles in a sample, a PCR product of the sample is mixed with another wild type amplicon labelled with a different fluorochrome (Ramlee et al. 2017). ...
... If the mutation of target gene will cause obvious phenotype, the PCR amplicons containing target site from mutants can be sequenced to identify the mutation; If the editing efficiency is high or not too many T1 transgenic plants selected, the PCR amplicons containing target site from mutants can be sequenced to identify the mutation; If the editing efficiency is low, it is better to use suitable methods for screening before sequencing large numbers of samples. There are several methods developed to screen CRISPR/ Cas9 editing mutants, such as PCR/restriction enzyme assay (Nadakuduti et al. 2019), T7 endonuclease I assay , surveyor nuclease assay (Vogiatzakis et al. 2013), polyacrylamide gel electrophoresis-based methods (Zhu et al. 2014), high-resolution melting analysis-based assays (Meistertzheim et al. 2017), fluorescent PCR capillary gel electrophoresis methods (Ramlee et al. 2015), and so on. However, these methods are not suitable for high-throughput screening due to specific sequence requirement, time consuming or high cost. ...
Article
Full-text available
D-Glycerate 3-kinase (GLYK, EC2.7.1.31) catalyzes the phosphorylation of glycerate to form 3-phosphoglycerate. However, knowledge about GLYK regulation is limited. In this study, the Arabidopsis Ler background glyk mutants were generated using CRISPR/Cas9 technology to elucidate the genetic regulation mechanisms of GLYK in future. Specifically, two sgRNAs targeting the Arabidopsis GLYK genomic sequence (At1g80380) were constructed into egg cell-specific expression CRISPR/Cas9 vector, and Ler plants were transformed by the floral dip method. A high-throughput screening and sequencing combined PCR (SSC-PCR) was developed to identify mutants generated by CRISPR/Cas9 technology using the DNA of T1 positive transgenic lines. By using this method, the edited GLYK lines were efficiently screened out from T1-positive transgenic lines. Thereafter, two independent homozygous lines, glyk-1 and glyk-3, were selected and further confirmed their genetic stability and absence of off-targets. The glyk-1 and glyk-3 showed typical photorespiration phenotype, and the chlorophyll content and Fv/Fm ratio of glyk-1 and glyk-3 gradually decreased when they were moved from 3000 ppm CO2 to 400 ppm CO2. Finally, these two mutants became lethal after nine-day exposure to 400 ppm CO2. Furthermore, six proteins have been screened out for their potential interactions with AtGLYK using yeast two-hybrid system in a previous study, while in this study, only two proteins, At1g78380 (AtGSTU19) and At4g37980 (AtCAD7), were further confirmed their interactions with AtGLYK using BiFC and LCI approaches, respectively. In summary, the developed SSC-PCR method here can be applied for the quick screening and identification of CRISPR/Cas9-edited mutants in plants; the glyk mutants and the GLYK-interacting proteins reported in this study will facilitate future understanding of the genetic regulation mechanisms of AtGLYK.
... Among them, the polymerase chain reaction (PCR)-based assay, restriction enzyme (RE) site loss assay, and high-throughput next-generation sequencing are frequently used for characterization of target mutants. Another approach is to use capillary electrophoresis-based fragment analysis for high-throughput detection of insertion/deletions in gene-edited samples [209]. Traditional Sanger sequencing methods to identify homozygous mutants are time-consuming and laborious for large numbers of samples and may be unsuitable for heterogeneous samples [210]. ...
Article
Full-text available
Advances in molecular technologies over the past few decades, such as high-throughput DNA marker genotyping, have provided more powerful plant breeding approaches, including marker-assisted selection and genomic selection. At the same time, massive investments in plant genetics and genomics, led by whole genome sequencing, have led to greater knowledge of genes and genetic pathways across plant genomes. However, there remains a gap between approaches focused on forward genetics, which start with a phenotype to map a mutant locus or QTL with the goal of cloning the causal gene, and approaches using reverse genetics, which start with large-scale sequence data and work back to the gene function. The recent establishment of efficient CRISPR-Cas-based gene editing promises to bridge this gap and provide a rapid method to functionally validate genes and alleles identified through studies of natural variation. CRISPR-Cas techniques can be used to knock out single or multiple genes, precisely modify genes through base and prime editing, and replace alleles. Moreover, technologies such as protoplast isolation, in planta transformation, and the use of developmental regulatory genes promise to enable high-throughput gene editing to accelerate crop improvement.
... Yang and colleagues [46] substituted a fluorescently labeled M13F primer with a universal primer and termed their method IDAA (Indel Detection by Amplicon Analysis). Ramlee and colleagues [47] used a traditional 2-primer PCR with a fluorescently labeled forward primer. The main advantage of using a 3-primer mixture is to cut the costs associated with the fluorescent labeling of forward primers for all amplicons and is best suited for investigators involved in the generation and analysis of several gene knockouts. ...
Article
Full-text available
The use of zebrafish in functional genomics and disease modeling has become popular due to the ease of targeted mutagenesis with genome editing nucleases, i.e., zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), and clustered regularly interspaced short palindromic repeats/Cas9 (CRISPR/Cas9). These nucleases, specifically CRISPR/Cas9, are routinely used to generate gene knockout mutants by causing a double stranded break at the desired site in the target gene and selecting for frameshift insertions or deletions (indels) caused by the errors during the repair process. Thus, a variety of methods have been developed to identify fish with indels during the process of mutant generation and phenotypic analysis. These methods range from PCR and gel-based low-throughput methods to high-throughput methods requiring specific reagents and/or equipment. Here, we provide a comprehensive review of currently used indel detection methods in zebrafish. By discussing the molecular basis for each method as well as their pros and cons, we hope that this review will serve as a comprehensive resource for zebrafish researchers, allowing them to choose the most appropriate method depending upon their budget, access to required equipment and the throughput needs of the projects.
... This fact makes genome editing and the subsequent mutation detection extremely challenging in wheat. Current methods for detecting DNA sequence mutations induced by genome editing include PCR-RE (Shan et al., 2014), PCR/RNP (Liang et al., 2018), the T7EI cleavage assay (Vouillot et al., 2015), next generation sequencing (NGS) (Liu et al., 2019b), high-resolution melting analysis (HRMA) (Dahlem et al., 2012), and fluorescent PCR-capillary gel electrophoresis (Ramlee et al., 2015). Even though each method has its shortcomings, PCR-RE and Sanger sequencing are the most direct, convenient, and widely used methods in many laboratories. ...
Article
Full-text available
The Nudum ( Nud ) gene controls the caryopsis type of cereal crops by regulating lipid biosynthetic pathways. Based on the HvNud sequence and its homologous gene sequences in wheat, a conserved sgRNA was designed to obtain the mutants from the barley variety “Vlamingh” and the wheat variety “Fielder” via Agrobacterium -mediated transformation. A total of 19 and 118 transgenic plants were obtained, and 11 and 61 mutant plants were identified in T 0 transgenic plants in barley and wheat after PCR-RE detection, and the editing efficiencies of the targeted gene were 57.9 and 51.7% in barley and wheat, respectively. The grain shape of the barley mutants was naked. Five different combinations of mutations for wheat TaNud genes were identified in the T 0 generation, and their homozygous-edited plants were obtained in the T 1 generation. Interestingly, the conjoined plants in which one plant has different genotypes were first identified. The different tillers in an individual T 0 plant showed independent transgenic or mutant events in both barley and wheat, and the different genotypes can stably inherit into T 1 generation, indicating that the T 0 transgenic plants were the conjoined type. In addition, we did not find any off-target mutations in both barley and wheat. A candidate method for detecting putative-edited wheat plants was suggested to avoid losing mutations in this investigation. This study provides not only materials for studying the function of the Nud gene in barley and wheat but also a system for detecting the mutants in wheat.
Article
The CRISPR/Cas9 system enables a versatile set of genomes editing and genetic-based disease modeling tools due to its high specificity, efficiency, and accessible design and implementation. In cancer, the CRISPR/Cas9 system has been used to characterize genes and explore different mechanisms implicated in tumorigenesis. Different experimental strategies have been proposed in recent years, showing dependency on various intrinsic factors such as cancer type, gene function, mutation type, and technical approaches such as cell line, Cas9 expression, and transfection options. However, the successful methodological approaches, genes, and other experimental factors have not been analyzed. We, therefore, initially considered more than 1,300 research articles related to CRISPR/Cas9 in cancer to finally examine more than 400 full-text research publications. We summarize findings regarding target genes, RNA guide designs, cloning, Cas9 delivery systems, cell enrichment, and experimental validations. This analysis provides valuable information and guidance for future cancer gene validation experiments.
Article
Limb girdle muscular dystrophy type R1 (LGMDR1) is an autosomal recessive myopathy described in humans resulting from a deficiency of calpain-3 protein (CAPN3). This disease lacks effective treatment and an appropriate model, so the generation of KO pigs by CRISPR-Cas9 offers a way to better understand disease ethology and to develop novel therapies. Microinjection is the main method described for gene editing by CRISPR-Cas9 in porcine embryo, but electroporation, which allows handling more embryos faster and easier, has also recently been reported. The objective of the current study was to optimize porcine oocyte electroporation to maximize embryo quality and mutation rate in order to efficiently generate LGMDR1 porcine models. We found that the efficiency of generating CAPN3 KO embryos was highest with 4 electroporation pulses and double sgRNA concentration than microinjection. Direct comparison between microinjection and electroporation demonstrated similar rates of embryo development and mutation parameters. The results of our study demonstrate that oocyte electroporation, an easier and faster method than microinjection, is comparable to standard approaches, paving the way for democratization of transgenesis in pigs.
Preprint
Full-text available
Honeybees ( Apis mellifera ) need their fine sense of taste to evaluate nectar and pollen sources. Gustatory receptors (Grs) translate taste signals into electrical responses. In vivo experiments primarily demonstrate collective responses of the whole Gr-set, but little is known about the individual impact of receptors. Here, we disentangle for the first time the contributions of three gustatory receptors (AmGr1-3) in sugar sensing of honeybees by combining CRISPR/Cas9 mediated genetic knock-out, electrophysiology and behaviour. AmGr1 responds to multiple sugars. Bees lacking this receptor have a reduced response to sucrose and glucose but not to fructose. AmGr2 acts as co-receptor of AmGr1 in a heterologous expression system, but honeybee knock-out mutants perform normally. Eliminating AmGr3 while preserving AmGr1 and AmGr2 abolishes the perception of fructose but not of sucrose. We thus dissociate the roles of AmGr1, AmGr2 and AmGr3 in honeybee taste perception.
Article
Full-text available
MicroRNAs (miRNAs) have emerged as important regulators that potentially play critical roles in various biological processes. Previous studies have shown that miR-615 regulates proliferation and apoptosis in many types of cancers. The biological function of this microRNA in breast cancer remains largely unexplored. In the present study, we found that miR-615 expression was markedly downregulated in breast cancer tissues and breast cancer cells. The enforced expression of miR-615 was able to inhibite the proliferation and anchorage-independent growth of breast cancer cells, while miR-615-in showed the opposite effect. Bioinformatics analysis further revealed AKT2, a putative tumor promoter as a potential target of miR-615. Ectopic expression of miR-615 led to downregulation of AKT2 protein, which resulted in the upregulation of p27 and p21 and the downregulation of cyclin D1. In sum, these results suggest that miR-615 represents a potential anti-onco-miR and participates in breast cancer carcinogenesis by suppressing AKT2 expression.
Article
Full-text available
With the goal to generate and characterize the phenotypes of null alleles in all genes within an organism and the recent advances in custom nucleases, genome editing limitations have moved from mutation generation to mutation detection. We previously demonstrated that High Resolution Melting (HRM) analysis is a rapid and efficient means of genotyping known zebrafish mutants. Here we establish optimized conditions for HRM based detection of novel mutant alleles. Using these conditions, we demonstrate that HRM is highly efficient at mutation detection across multiple genome editing platforms (ZFNs, TALENs, and CRISPRs); we observed nuclease generated HRM positive targeting in 1 of 6 (16%) open pool derived ZFNs, 14 of 23 (60%) TALENs, and 58 of 77 (75%) CRISPR nucleases. Successful targeting, based on HRM of G0 embryos correlates well with successful germline transmission (46 of 47 nucleases); yet, surprisingly mutations in the somatic tail DNA weakly correlate with mutations in the germline F1 progeny DNA. This suggests that analysis of G0 tail DNA is a good indicator of the efficiency of the nuclease, but not necessarily a good indicator of germline alleles that will be present in the F1s. However, we demonstrate that small amplicon HRM curve profiles of F1 progeny DNA can be used to differentiate between specific mutant alleles, facilitating rare allele identification and isolation; and that HRM is a powerful technique for screening possible off-target mutations that may be generated by the nucleases. Our data suggest that micro-homology based alternative NHEJ repair is primarily utilized in the generation of CRISPR mutant alleles and allows us to predict likelihood of generating a null allele. Lastly, we demonstrate that HRM can be used to quickly distinguish genotype-phenotype correlations within F1 embryos derived from G0 intercrosses. Together these data indicate that custom nucleases, in conjunction with the ease and speed of HRM, will facilitate future high-throughput mutation generation and analysis needed to establish mutants in all genes of an organism.
Article
Full-text available
The efficacy and the mutation spectrum of genome editing methods can vary substantially depending on the targeted sequence. A simple, quick assay to accurately characterize and quantify the induced mutations is therefore needed. Here we present TIDE, a method for this purpose that requires only a pair of PCR reactions and two standard capillary sequencing runs. The sequence traces are then analyzed by a specially developed decomposition algorithm that identifies the major induced mutations in the projected editing site and accurately determines their frequency in a cell population. This method is cost-effective and quick, and it provides much more detailed information than current enzyme-based assays. An interactive web tool for automated decomposition of the sequence traces is available. TIDE greatly facilitates the testing and rational design of genome editing strategies.
Article
Full-text available
The CRISPR-associated RNA-guided nuclease Cas9 has emerged as a powerful tool for genome engineering in a variety of organisms. In order to achieve efficient gene targeting rates in Drosophila, current approaches require either the injection of in vitro transcribed RNAs or injecting into transgenic Cas9-expressing embryos. We report a simple and versatile alternative method for CRISPR-mediated genome editing in Drosophila using bicistronic Cas9/sgRNA expression vectors. Gene targeting with this single-plasmid injection approach is as efficient as in transgenic nanos-Cas9 embryos and allows the isolation of targeted knock-out and knock-in alleles by molecular screening within two months. Our strategy is independent of genetic background and does not require prior establishment of transgenic flies.
Article
Full-text available
The rapid generation of various species and strains of laboratory animals using CRISPR/Cas9 technology has dramatically accelerated the interrogation of gene function in vivo. So far, the dominant approach for genotyping of genome-modified animals has been the T7E1 endonuclease cleavage assay. Here, we present a polyacrylamide gel electrophoresis-based (PAGE) method to genotype mice harboring different types of indel mutations. We developed 6 strains of genome-modified mice using CRISPR/Cas9 system, and utilized this approach to genotype mice from F0 to F2 generation, which included single and multiplexed genome-modified mice. We also determined the maximal detection sensitivity for detecting mosaic DNA using PAGE-based assay as 0.5%. We further applied PAGE-based genotyping approach to detect CRISPR/Cas9-mediated on- and off-target effect in human 293T and induced pluripotent stem cells (iPSCs). Thus, PAGE-based genotyping approach meets the rapidly increasing demand for genotyping of the fast-growing number of genome-modified animals and human cell lines created using CRISPR/Cas9 system or other nuclease systems such as TALEN or ZFN.
Article
Full-text available
Genome editing techniques such as the zinc-finger nucleases (ZFNs), transcription activator-like effecter nucleases (TALENs) and clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated (Cas) system Cas9 can induce efficient DNA double strand breaks (DSBs) at the target genomic sequence and result in indel mutations by the error-prone non-homologous end joining (NHEJ) DNA repair system. Several methods including sequence specific endonuclease assay, T7E1 assay and high resolution melting curve assay (HRM) etc have been developed to detect the efficiency of the induced mutations. However, these assays have some limitations in that they either require specific sequences in the target sites or are unable to generate sequencing-ready mutant DNA fragments or unable to distinguish induced mutations from natural nucleotide polymorphism. Here, we developed a simple PCR-based protocol for detecting indel mutations induced by TALEN and Cas9 in zebrafish. We designed 2 pairs of primers for each target locus, with one putative amplicon extending beyond the putative indel site and the other overlapping it. With these primers, we performed a qPCR assay to efficiently detect the frequencies of newly induced mutations, which was accompanied with a T-vector-based colony analysis to generate single-copy mutant fragment clones for subsequent DNA sequencing. Thus, our work has provided a very simple, efficient and fast assay for detecting induced mutations, which we anticipate will be widely used in the area of genome editing.
Article
Full-text available
Targeted genome editing technologies are powerful tools for studying biology and disease, and have a broad range of research applications. In contrast to the rapid development of toolkits to manipulate individual genes, large-scale screening methods based on the complete loss of gene expression are only now beginning to be developed. Here we report the development of a focused CRISPR/Cas-based (clustered regularly interspaced short palindromic repeats/CRISPR-associated) lentiviral library in human cells and a method of gene identification based on functional screening and high-throughput sequencing analysis. Using knockout library screens, we successfully identified the host genes essential for the intoxication of cells by anthrax and diphtheria toxins, which were confirmed by functional validation. The broad application of this powerful genetic screening strategy will not only facilitate the rapid identification of genes important for bacterial toxicity but will also enable the discovery of genes that participate in other biological processes.
Article
Full-text available
Restriction fragment length polymorphism (RFLP) analysis is one of the oldest, most convenient and least expensive methods of genotyping, but is limited by the availability of restriction endonuclease sites. Here we present a novel method of employing CRISPR/Cas-derived RNA-guided engineered nucleases (RGENs) in RFLP analysis. We prepare RGENs by complexing recombinant Cas9 protein derived from Streptococcus pyogenes with in vitro transcribed guide RNAs that are complementary to the DNA sequences of interest. Then, we genotype recurrent mutations found in cancer and small insertions or deletions (indels) induced in cultured cells and animals by RGENs and other engineered nucleases such as transcription activator-like effector nucleases (TALENs). Unlike T7 endonuclease I or Surveyor assays that are widely used for genotyping engineered nuclease-induced mutations, RGEN-mediated RFLP analysis can detect homozygous mutant clones that contain identical biallelic indel sequences and is not limited by sequence polymorphisms near the nuclease target sites.
Article
Full-text available
Improving Whole-Genome Screens Improved methods are needed for the knockout of individual genes in genome-scale functional screens. Wang et al. (p. 80 , published online 12 December) and Shalem et al. (p. 84 , published online 12 December) used the bacterial CRISPR/Cas9 system to power-screen protocols that avoid several of the pitfalls associated with small interfering RNA (siRNA) screens. Genome editing by these methods completely disrupts target genes, thus avoiding weak signals that can occur when transcript abundance is partially decreased by siRNA. Furthermore, gene targeting by the CRISPR system is more precise and appears to produce substantially fewer off-target effects than existing methods.
Article
Gene amplification occurs at high frequency in transformed cells (10(-3)-10(-5)), but is undetectable in normal diploid fibroblasts (less than 10(-9)). This study examines whether alterations of one or both p53 alleles were sufficient to allow gene amplification to occur. Cells retaining one wild-type p53 allele mimicked the behavior of primary diploid cells: they arrested growth in the presence of drug and failed to demonstrate amplification. Cells losing the second p53 allele failed to arrest when placed in drug and displayed the ability to amplify at a high frequency. Thus, loss of wild-type p53 may lead to amplification, possibly caused by changes in cell cycle progression. Other determinants can by-pass this p53 function, however, since tumor cells with wild-type p53 have the ability to amplify genes.