Efficient CRISPR Editing With A Hypercompact Cas12
Efficient CRISPR Editing With A Hypercompact Cas12
Efficient CRISPR Editing With A Hypercompact Cas12
https://doi.org/10.1038/s41587-021-01009-z
Gene therapy would benefit from a miniature CRISPR system that fits into the small adeno-associated virus (AAV) genome
and has high cleavage activity and specificity in eukaryotic cells. One of the most compact CRISPR-associated nucleases yet
discovered is the archaeal Un1Cas12f1. However, Un1Cas12f1 and its variants have very low activity in eukaryotic cells. In the
present study, we redesigned the natural guide RNA of Un1Cas12f1 at five sites: the 5′ terminus of the trans-activating CRISPR
RNA (tracrRNA), the tracrRNA–crRNA complementary region, a penta(uridinylate) sequence, the 3′ terminus of the crRNA
and a disordered stem 2 region in the tracrRNA. These optimizations synergistically increased the average indel frequency by
867-fold. The optimized Un1Cas12f1 system enabled efficient, specific genome editing in human cells when delivered by plas-
mid vectors, PCR amplicons and AAV. As Un1Cas12f1 cleaves outside the protospacer, it can be used to create large deletions
efficiently. The engineered Un1Cas12f1 system showed efficiency comparable to that of SpCas9 and specificity similar to that
of AsCas12a.
T
he clustered regularly interspaced short palindromic repeats base editors8,9, prime editors10 or regulators of epigenetic features5.
(CRISPR) system, which functions as an adaptive immune Moreover, a miniature CRISPR system would allow the use of mul-
system in bacteria, archaea and huge bacteriophages, has tiple guide (g)RNAs and/or additional regulatory genes in a single
been developed into versatile genome-editing tools1–4. Conventional AAV particle.
genome-editing tools induce double-stranded DNA (dsDNA) Classified as type-V CRISPR nucleases, miniature Cas proteins,
breaks (DSBs), which frequently result in indel mutations through including Cas12f (also known as Cas14) and Cas12j (CasΦ), were
the nonhomologous end-joining (NHEJ)-mediated repair pro- identified from archaea25 and huge bacteriophages2. Miniature
cess5,6. Precise genetic modifications have also been achieved by CRISPR–Cas effectors consist of ~400–700 amino acid residues
homology-driven repair7, base editing systems8,9 and prime edit- and include a single RuvC nuclease domain25. So far, it has not been
ing technology10. These diverse genome-editing tools have facili- demonstrated that miniature Cas proteins can be used as robust
tated cell engineering, generation of model animals, development genome-editing tools in eukaryotic cells. Cas12f1 was originally
of new plant varieties11 and genetic screening12. In particular, these reported to show only single-stranded DNA (ssDNA) cleavage
methods hold promise for gene therapy in the treatment of cancer13, activity25. A subsequent study revealed indel activity of Cas12f in
genetic disorders14 and infectious diseases15,16. eukaryotic cells, but the efficiency was only marginal and only two
Gene therapy can be performed in vivo or ex vivo, depending on targets were tested26. Cas12j was functionally validated in plant cells,
the accessibility of target cells. Ex vivo therapy is performed on indi- but the indel efficiency was also <1%2. Furthermore, in neither case
vidual types of patient-derived cells, which include immune cells17, was the feasibility of AAV delivery demonstrated.
adult stem cells18 or, in principle, germ cells19. However, most mono- In the present study, we intensively remodeled the gRNA for
genic diseases require systemic delivery of genetic materials through Un1Cas12f1 (hereafter referred to as Cas12f) through five rounds
an in vivo strategy, which necessitates an efficient vehicle for their of gRNA engineering. The individual changes to the gRNA struc-
delivery. AAV is a US Food and Drug Administration-approved ture were synergistic, thereby transforming the CRISPR–Cas12f
vehicle as a result of its safety, persistence and compatibility with system into a highly efficient and specific genome-editing tool.
mass production20. In this regard, an AAV-loadable CRISPR tool Moreover, the gRNA remodeling substantially reduced the gRNA
would hold promise for the treatment of genetic disorders in vivo21. size. When delivered by an AAV vector, this compact CRISPR
However, AAV has a limited payload size of <4.7 kb which hampers system enabled multiplexed and efficient genome editing. In the
clinical applications of most CRISPR tools22. Smaller Cas proteins, present study, we propose that our engineered Cas12f1 system is
including SaCas9 (ref. 23) and CjCas9 (ref. 24), have been identified useful for gene editing with AAV delivery and that Un1Cas12f1
and validated with respect to their use as genome-editing tools might form the basis for developing AVV-deliverable base and
delivered through an AAV. These can be delivered with an AAV, but prime editors, epigenetic regulators and constructs for site-specific
even smaller CRISPR systems are needed to enable AAV delivery of gene regulation.
Genome Editing Research Center, Korea Research Institute of Bioscience & BioTechnology, Daejeon, Republic of Korea. 2KRIBB School of Bioscience, Korea
1
University of Science and Technology, Daejeon, Republic of Korea. 3GenKOre, Daejeon, Republic of Korea. 4Department of Biomedical and Pharmaceutical
Sciences, Graduate School, Kyung Hee University, Seoul, Republic of Korea. 5Department of Pharmaceutical Science, College of Pharmacy, Kyung Hee
University, Seoul, Republic of Korea. 6These authors contributed equally: Do Yon Kim, Jeong Mi Lee. ✉e-mail: omsys1@kribb.re.kr
a –160 MS3
Engineering flow
MS5 –130 Stem 1
MS1
Stem 2
–105 MS2
tracrRNA
–60
MS3
–1 –24 MS4
MS2
Spacer
crRNA MS5
MS4 –30 –20 MS1 –10 –1
b (position) c d
tracrRNA nucleotide P = 0.019
4 30 P < 0.001
–24 –23 –22 –21 –20
U U U U U P = 0.023
1%
Percentage indel
3
Percentage indel
P < 0.001
U 0.01 20
n.s.
2 P = 0.001
P < 0.001
Substitution
e f 100 h
Target 1 Target 2 Target 3 P = 0.002
MS3 Change of length (nt)
Exp 1 Exp 2 Exp 3 Exp 1 Exp 2 Exp 3 Exp 1 Exp 2 Exp 3 80
Percentage indel
MS2 +9 (crRNA)
WT 16.4 17.5 18.4 5.4 4.9 5.8 13.3 16.4 13.1 60 MS3 –20 (tracrRNA)
∆12 nt 13.2 18.7 21.5 8.9 7.0 6.7 15.3 22.1 13.5
40 MS4 –28 (tracrRNA), –27 (crRNA)
∆13 nt 4.5 6.5 5.5 5.4 7.6 8.8 13.4 16.4 15.0
MS5 –27 (tracrRNA)
20
∆14 nt 21.6 23.6 20.4 9.1 7.6 8.0 21.2 22.4 21.4
∆15 nt 10.7 10.6 10.7 1.9 2.8 2.8 13.2 11.9 16.4 0 5,000 gRNA
Cas12f1
Cas12f_ge4.1
– + – + –
LbCas12a
SaCas9
Cas12f_ge3.0
CjCas9
CasΦ-2
AaCas12b
Cas12f_ge4.0
SpCas9
– – + + –
∆23 nt 6.8 7.8 6.9 10.8 18.2 11.8 0.1 1.3 1.0 – – + – +
– – – + +
– + – – –
∆24 nt 13.0 8.4 12.9 2.8 3.5 1.6 9.7 8.2 4.9 – – + – –
– – – + –
– – – – +
– – – – –
0 60% 0 30% 0 40%
Fig. 1 | Engineering Cas12f gRNA. a, Structure of the canonical Cas12f1 gRNA consisting of tracrRNA and crRNA. Five MSs for gRNA engineering are
indicated. The gRNA engineering steps were performed sequentially, from MS1 to MS2 and MS3, and finally MS4; MS5 modifications are discussed
elsewhere. b, Increased Cas12f-mediated indel frequencies caused by substitutions of uridine in the tracrRNA penta(uridinylate) site (MS1). c, Combined
effects of sequence modifications in both the tracrRNA and the crRNA at the penta(uridinylate) site on indel frequencies (n = 3). d, Synergistic modulation
of indel frequencies by modifications in MS1 and MS2 (addition of poly(uridinylate) 3′ overhang on the crRNA) (n = 3). e, Optimal length of truncation
of the 5′ terminus of the tracrRNA (MS3). Values were obtained from independent triplicate experiments. f, Changes in indel frequencies induced by the
truncation of the crRNA–tracrRNA complementary region (MS4). At each position, the crRNA and tracrRNA were truncated and connected with a GAAA
tetraloop (n = 3). g, Increased indel frequencies induced by various combinations of gRNA modifications (n = 3). h, Changes in the length of the crRNA
and tracrRNA caused by each step of gRNA engineering and comparison of the lengths of sequences encoding the components of several representative
CRISPR–Cas systems. NS, not significant. c,d,f, Two-group and multiple comparisons were performed by the two-sided Student’s t-test and one-way
ANOVA test, respectively. All error bars represent s.d.
our extensive gRNA engineering efforts yielded a potent, extremely suggested previously32, the U-rich 3′ overhang appeared to sta-
compact CRISPR–Cas12f1 system. bilize the sgRNA transcript in cells. The MS3 engineering was
Finally, we sought to explain how each gRNA modification not associated with changes in gRNA expression, but further
(MS1–MS5) contributes to increased indel frequencies using tar- increased the dsDNA cleavage activity of Cas12f1 when stacked
geted RNA-sequencing (RNA-seq) analysis (Supplementary Fig. 4a). to the MS1/2 modifications (Supplementary Fig. 4d). The MS4
As expected, the MS1 engineering led to a drastic increase in the and MS5 engineering further increased the cleavage activity. To
expression of the full-length sgRNA (Supplementary Fig. 4b,c). validate the effects of the MS3 modification on indel frequencies,
Besides increasing the affinity of the Cas–gRNA interaction as MS1/2- and MS1/2/3-modified gRNAs were compared with respect
Percentage indel
ge_3.0 60
AsCas12a DR Spacer (N23)
40
Spacer (N20) ge_4.0
Canonical Cas12f1 DR
20
Cas12f_ge3.0 DR(MS1/3) Spacer (N20) MS2(T4AT6) ge_4.1
0
Cas12f_ge4.0 DR(MS3/4) Spacer (N20) MS2(T4AT6) SpCas9
a
0
Sp 4.1
0
C s9
al
12
4.
3.
AsCas12a
ic
Spacer (N20)
As Ca
Cas12f_ge4.1 DR(MS3/4/5) MS2(T4AT6)
ge
ge
ge
as
on
0%
an
C
d e 100
Range of Distribution of percentage indel (n = 88) n = 67 n = 15 n=6
percentage 80 ge3.0
indel ge_3.0 ge_4.0 ge_4.1 Cas9 Cas12a
Percentage indel
ge4.0
≤1% 35 24 7 3 8 60 ge4.1
1–10% 25 23 12 11 25
40
10–20% 4 10 28 32 24
20–30% 8 6 10 19 20
20
30–50% 6 14 19 17 9
≥ 50% 10 11 12 6 2 0
Validation targets (n = 88)
f g gRNA1
h i 15
Target 1 Target 2 20 20
gRNA2 MS1/2
5′ TTTR-N20---spacing---N20-YAAA 3′ gRNA1+2 MS1/2/3
10 15
MS1/2
Percentage indel
MS2/3/4/5 10
Fold-change
Fold-change
gRNA1 gRNA1+gRNA2 gRNA2 0 10
5
MS1/2/3
10 5
T1 T2 T1 T2 T1 T2
20 0 0
(Low indel) (High indel) (Low indel) a b c d e f g h i j 0 20 40 60 80
a b c d e f g h i j
Target ID Target ID Length of spacing (bp)
Fig. 2 | Large-scale validation of the engineered CRISPR–Cas12f system. a, Common sequences targetable by the SpCas9, AsCas12a and Cas12f systems
and the gRNA formulations for each system. TTTR indicates TTTA or TTTG. b, A heatmap for indel frequencies per target obtained by SpCas9, AsCas12a
or Cas12f. Measurement of indel frequencies in HEK293T cells transfected with SpCas9, AsCas12a, canonical Cas12f or engineered Cas12f vector
constructs. Cells (1.75 × 105) were transfected with 2 μg of plasmid vector using a Fugene lipofection kit and grown for 96 h. c, A box-and-whisker plot for
SpCas9-, AsCas12a- and Cas12f-induced indel frequencies merged with a dot plot. Whole data points (n = 88) were plotted with mean values as indicated
by the horizontal cyan-colored line. Box plots represent the median with interquartile ranges (25–75%); whiskers extend to 1.5× the interquartile distance
from the box. P values were derived by a Mann–Whitney U-test. NS, not significant. Error bars represent the s.d. d, Distribution of the number of targets
per indel frequency subdivision. Values indicate the number of targets with indel efficiency belonging to the indicated ranges. The indel efficiencies and
target information were provided in a source file and Supplementary Table 2. e, Distribution of the targets that show the highest efficiencies by
the _ge3.0, _ge4.0 or _ge4.1 version. f, Schematic representation of the paired gRNA strategy for increasing Cas12f-mediated indel frequencies. DNA
cleavages are centered in the spacing region, which is 10–30 bp in length. The size of the triangle indicates the frequency of a DNA-strand cleavage.
g, Indel frequencies induced by Cas12f with engineered gRNA at ten targets when using either or both gRNAs. The upper and lower panels indicate indel
frequencies for gRNAs with MS1/2 and MS1/2/3 engineering, respectively. h, Comparison of fold-changes in indel frequencies caused by MS1/2- and
MS1/2/3-engineered gRNAs. The fold-changes were calculated from the indel frequencies induced by paired gRNAs compared with that of a gRNA that
induces a higher indel frequency at a target located between the two paired gRNAs. i, Fold-changes in indel efficiencies by paired gRNAs according to the
length of spacing.
to indel-generating efficiency in vivo. Out of 19 targets tested, 17 targets showing frequencies of <0.1%. However, use of our engi-
(90%) showed increased indel frequencies, by at least twofold, with neered gRNAs led to significant increases in indel frequencies at
the average fold increase being 3.12 (Supplementary Fig. 4e). The most target sites (Fig. 2b). The average efficiency of Cas12f_ge4.1
structures of the Cas12f_ge3.0, Cas12f_ge4.0 and Cas12f_ge4.1 was comparable to that of SpCas9 (P > 0.05) and was even higher
gRNAs are presented in Supplementary Fig. 5. than that of AsCas12a (Fig. 2c). The average increase in efficiency
induced by Cas12f_ge4.1 sgRNA was 867-fold. Notably, Cas12f_
Large-scale validation of Cas12f. We next investigated whether the ge4.1 had more targets with high indel frequencies (≥50%) than
increased genome-editing efficiency of the engineered gRNAs can SpCas9 and AsCas12a (Fig. 2d). In addition, Cas12f_ge4.1 showed
be validated at a wider range of targets. We searched in silico for higher efficiencies for 76.1% (67 of 88) of targets, compared with the
endogenous targets containing the sequence 5′-TTTR-N20-NGG-3′, Cas12f_ge3.0 and Cas12f_4.0 versions, whereas the Cas12f_ge4.0
which are targetable with SpCas9, AsCas12a and Cas12f1 (Fig. 2a). and Cas12f_ge3.0 versions were most effective for 17.0% and 6.8%
We randomly selected 88 such endogenous loci (for target infor- of targets, respectively (Fig. 2e).
mation, please refer to Supplementary Table 2) and measured the We then sought to refine the Cas12f system further, because
SpCas9-, AsCas12a- and Cas12f-mediated indel frequencies in there still remained targets resistant to genome editing by Cas12f
HEK293T cells. Cas12f with canonical gRNAs generated indel fre- (in fact, the situation is also true for Cas9 and Cas12a, but Cas12f1
quencies of <1.0% over all tested targets, with 91% (80 of 88) of showed more targets with indel frequencies <1% than SpCas9).
a b c
Cas12f_ge4.1 SpCas9 LbCas12a
Cas12f 30 60 Cas12f1
Cas12f
Cas12f_ge4.1
5’ PAM 3’
25 50 SpCas9
Protospacer
LbCas12a
Length of deletion
Percentage indel
20 40
15 30
1st DSB/NHEJ
10 20
5 10
Protospacer
Multiple 0
events 0
1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 1 2 3 4 5 6 7
2nd DSB/NHEJ Days after transfection Days after transfection
Ladder
Cas12f LbCas12a
Control DS1 DS2 DS1 DS2 50
Percentage indel
nth DSB/NHEJ 40
1.0 kb
0.5 kb 30
Protospacer 0.3 kb 20
f g P = 0.002
h i
3d 9d 2.5 P = 0.002
SaCas9 4
Ladder
1.5
Relative expression
3
1.0
3 kb 0.5 2
2 kb 20
0
1
1 kb
PO Ta BE P
Ta BE d-T 1
Ta + U *
d* 2x GI
x I
-M I
VPLV
KR 64
AB
N t1
D 1
LSC1
Ez 1
p3 2
00
+2 UG
M UG
d- C1 ad
H MT
h
C
D
F
D Te
A
PO G
rA e
0
0
l
tro
3d 9d
rA
on
OCT4 promoter
C
Fig. 3 | Highly efficient correction of pathogenic mutations through AAV-delivered Cas12f. a, A schematic illustration showing multiple chances for
dsDNA cleavage and NHEJ cycles for the Cas12f system. Cleavage sites are marked with triangles and different colors indicate changes in the DNA
sequences. The protospacer regions are colored sky-blue and green. b, Increased frequency of long deletion mutations over time in HEK293T cells
transfected with CRISPR–Cas12f_ge4.1. c, Time-course of Cas12f-, SpCas9- and LbCas12a-induced indel frequencies in HEK293T cells transfected with
plasmid vectors (n = 3). d. Comparison of Cas12f- and LbCas12a-mediated rates of exon 51 deletion from the human dystrophin gene in AC16 cells. The
lower bands indicate the PCR amplicons of the exon 51-deleted locus. The intensity of the lower bands is indicative of the deletion efficiency. Deletion
strategy 1 (DS1) and DS2 target identical loci for Cas12f and Cas12a. The data represent three experiments. e, Screening of targets for the deletion
of the c.2991+1655A>G mutation from the CEP290 gene. f, Comparison of Cas12f- and SaCas9 (EDIT101)-mediated frequencies of deletion of the
c.2991+1655A>G mutation. HEK293T cells (2 × 105) were seeded into 12-well plates, transduced with AAV2 harboring the Cas12f or SaCas9 system at
5.0 × 109 vector genomes (vg) ml−1, and harvested 3 and 9 d post-transduction. NT, nontransduction. The data represent three experiments. g, Quantitative
analysis of deletion frequencies using RT-qPCR. Percentage deletion indicates the percentage ratio of PCR amplicons containing the deletion versus intact
amplicons (n = 3). P values were derived using a two-sided Welch’s t-test. h, Possible applications of Cas12f using an AAV delivery system. For AAV
delivery of vector constructs harboring the Cas12f sequence with two nuclear localization signals, a BGH poly(A) signal, and an XTEN linker sequence
under the control of an EF-1α core promoter and a ge4.1 sequence under the control of a U6 promoter, a protein encoded by a gene ≤2.1 kb in size could
be fused to Cas12f. i, Application of dCas12f-VP64 to CRISPRa. The fusion protein guided by gRNAs (ge4.1) targeting promoter regions of OCT4 led to
transcriptional activations in HEK293T cells (n = 3). P values were derived using a two-sided Student’s t-test. All error bars represent the s.d.
We hypothesized that the low efficiency of Cas12f1 at certain sites indel frequencies with the paired gRNAs. The fold increase varied
may originate from different cleavage efficiency between target and among targets, but all tested targets showed indel frequencies of
nontarget strands, because the compact size of Cas12f might cause >1%. Moreover, final indel frequencies were further improved by
less efficient nontarget strand cleavage. To test this hypothesis, we using MS1/MS2/MS3- versus MS1/MS2-modified gRNAs (Fig. 2g),
selected targets that carry a 5′-TTTR-N20-spacing-N20-YAAA-3′ mainly because indel-generating efficiencies of each gRNA were
sequence, where ‘spacing’ is a 10- to 80-bp-long dsDNA segment increased by MS3 engineering. However, the fold increase was more
(Fig. 2f). These sequences are targetable by a pair of gRNAs ori- pronounced for MS1/MS2 engineering, compared with the MS1/
ented in opposite directions; two dsDNA cleavage events occur in MS2/MS3 and MS2/MS3/MS4/MS5 versions (Fig. 2h). This result
the spacing region. Although each gRNA alone mediated relatively would be explained by our hypothesis that Cas12f1 displays unequal
low indel frequencies, targets in ten loci showed sharply increased cleavage kinetics for the target and nontarget strands, and that the
Matched
1/2
2/3
3/4
4/5
5/6
6/7
7/8
8/9
9/10
10/11
11/12
12/13
13/14
14/15
15/16
16/17
17/18
18/19
19/20
OF03 0 0 0 0
Mismatched position Mismatched position OF04 0 0 0 0
c 1 LbCas12a
e
NS geCas12_4.0 RPL32P3 CLIC4 P2RX5-TAX1BP3
(off-target:on-target)
0.1 geCas12_4.1
Indel ratio
NS
0.01 NS
34 23 34 74 13 38
0.001 22 5 14
0.0001
OF1
OF2
OF3
OF4
OF1
OF2
OF3
OF4
PRKCH EMX1
d Gap f
On-target
Percentage indel
0.001 0.01 0.1 1 10
PAM
1 10
Off-target AsCas12a
Cas12f_ge4.1
(chr13) Control
(chr22)
Fig. 4 | Unbiased and targeted analysis of Cas12f specificity as assessed by Digenome-seq analysis. a, Tolerance of Cas12f_4.1 to mismatched gRNA. The
engineered gRNAs with a singly mismatched base and pairs of mismatched bases were used for the investigation of indel frequencies in HEK293T cells
(n = 3). b, Indel frequencies at off-targets identified by OFFinder for AsCas12a, Cas12_ge4.0 and Cas12_ge4.1. An intergene corresponds to target 3 targeted
throughout the main text. c, Indel frequencies at previously validated off-targets for AsCas12a, Cas12_ge4.0 and Cas12_ge4.1. The ratio of indel frequency at
off-target to that at on-target was considered as an index for specificity (n = 3). Statistical analysis was performed by a two-sided Student’s t-test. NS, not
significant. d, IGV files at on-target and off-target loci after in vitro digestion of genomic DNA. The gap indicates a region where sequences were missing
in both forward and reverse reads. e, The number of potential off-target loci identified by Digenome-seq analysis for AsCas12a and Cas12f. f, Validation of
off-target sites identified by Digenome-seq analysis using Cas12f and AsCas12a as endonucleases. Indel frequencies were measured at both on-target and
potential off-target loci after transfection with either Cas12f- or AsCas12a-encoding vector. Control refers to Cas12f-untreated cells (n = 3). All error bars
represent the s.d.
degree of difference is reduced by MS3 engineering. A longer spac- mutations induced by Cas12f. Most mutation patterns included
ing region of ≥50 bp did not yield this pair gRNA-assisted increase relatively long deletions that affected the protospacer sequence
in indel frequencies (Fig. 2i). (Supplementary Fig. 6a,b). In contrast, indel mutations outside the
protospacer were relatively rare. We interpreted these long dele-
Favorable kinetic property for Cas12f-induced DNA cleavage. In tions to be the products of multiple cutting-and-joining processes.
addition to the compactness of Cas12f1, this system has an addi- In fact, this assumption was confirmed through a time-course
tional advantage for gene therapy: it induces dsDNA cleavages investigation of indel patterns. In the early phase of transfection,
outside the protospacer sequence33,36. This property implies that, deletions of <5 bp were dominant (Fig. 3b; the radius of a bubble
even after the initial round of NHEJ-mediated indel mutations, the indicates the mutation frequency). However, the frequency of long
protospacer sequence is likely to remain unchanged. Then, further deletions increased over time until 4 d later. In contrast, the pattern
rounds of the dsDNA cleavage–NHEJ process can continue (Fig. 3a). of indel mutations was almost consistent over time for SpCas9 and
This property is even more desirable for a large DNA-deletion strat- LbCas12a. Moreover, Cas12f caused a more persistent increase in
egy involving a pair of gRNAs. We analyzed the profile of indel indel frequencies, compared with SpCas9 and LbCas12a (Fig. 3c).
A handful of genetic disorders can potentially be treated by dele- levels of tolerance were observed for single-base mismatches, par-
tion of pathogenic introns or exons using paired gRNAs and Cas ticularly at positions 1–3, 5 and 17–20 (Fig. 4a). To compare the
proteins, including Duchenne muscular dystrophy37, Leber congen- results with that of Cas12a32,41, Cas12f showed lower tolerance in
ital amaurosis 10 (LCA10)38 and Usher’s syndrome type 2A39. We the protospacer-adjacent motif (PAM)-proximal regions and simi-
explored the potential utility of the Cas12f system for those appli- lar or slightly higher tolerance in the PAM-distal regions (positions
cations. As a case study, we selected a pair of sites in the vicinity 17–20). However, Cas12f exhibited less tolerance for mismatches in
of exon 51 of the human dystrophin gene that are common targets the middle region (positions 6–16). Moreover, Cas12f showed neg-
for LbCas12a and Cas12f. Screening experiments identified target ligible levels of tolerance for two-base mismatches, except for posi-
sequences that show similar indel frequencies for LbCas12a and tions 19/20, again similar to Cas12a.
Cas12f. Despite the similar indel efficiencies of individual gRNAs, Next, we employed targeted approaches to assess specificity. Using
Cas12f resulted in a higher level of deletions, compared with Cas-OFFinder42, we selected potential off-target sites that contained
LbCas12a (Fig. 3d). These results indicate that Cas12f might be par- three base mismatches, but no bulges, with a set of on-target sites in
ticularly useful for AAV delivery in gene therapy applications that P2RX5-TAX1BP3, CLIC4, NLRC4 and an intergene region, for which
require deletions. Cas12f showed higher on-target efficiencies than Cas12a (Fig. 4b
and Supplementary Table 3). Deep-sequencing analysis revealed
AAV delivery of the engineered Cas12f system. Next, we inves- that Cas12f was more specific than AsCas12a: whereas AsCas12a
tigated the genome-editing performance of a recombinant AAV2 resulted in residual levels of indels (<0.1%) at two off-target sites
(rAAV2)–Cas12f vector. We constructed an rAAV vector carry- and an indel frequency of 0.36% at one other site among a total of
ing sequences encoding either Cas12f_ge4.1 or a control vector 26 potential off-target sites, Cas12f_ge4.0 and _ge4.1 resulted in
(scrambled sgRNAs). Cas12f1 and sgRNA expression were driven an indel frequency of 0.04% at each one of the potential off-target
under the control of the chicken β-actin and the human U6 pro- sites. We also compared genome-editing specificity for targets in
moters, respectively (Supplementary Fig. 7a). The total length RPL32P3, PRKCH and EMX1, for which Cas12a was previously
of these sequences (4.40 kb) fell within the permissive size for an observed to induce off-target effects41,43 (Supplementary Table 3).
AAV payload, even in the presence of two sgRNA sequences and On the whole, Cas12f and AsCas12a induced similar off-target
an enhanced green fluorescent protein (eGFP)-encoding reporter effects, except for the off-target sites that had a single mismatch in
sequence. The rAAV2 particles were produced in HEK293T cells the PAM-distal region (OF1–3 for RPL32P3; Fig. 4c).
after transfection with an rAAV vector, pAAVED2/2 and a helper We next employed the Digenome-sequencing (Digenome-seq)
plasmid. The sgRNAs respectively targeted an intergenic locus (tar- analysis to further examine the specificity of Cas12f44. Three tar-
get 1) and the KRT1 gene (target 2). gets (RPL32P3, CLIC4 and P2RX5-TAX1BP3) were selected to
AAV delivery to HEK293T cells led to an increase of the frequen- compare the specificity of AsCas12a and Cas12f_ge4.1. Analysis
cies of indel mutations over time (Supplementary Fig. 7b) and with of the Integrative Genomics Viewer (IGV) files from the Cas12f_
increasing numbers of rAAV2 particles (Supplementary Fig. 7c). ge4.1 experiments shows a presence of gaps between forward- and
The infection was monitored by green fluorescence, which was per- reverse-strand reads at both on-target and off-target sites (Fig. 4d),
sistent for 2 weeks post-transduction (Supplementary Fig. 7d). which is assumed to arise from either the ssDNA cleavage activ-
Next, we explored the targeting of therapeutically useful loci ity by the cleavage-activated Cas12f25 or a generation of 3′ over-
for the deletion of a pathogenic cryptic exon in the CEP290 gene hang. The Digenome-seq analysis revealed that gRNAs targeting
for the treatment of LCA10 (ref. 38). We tested on both sides of the RPL32P3, CLIC4 and P2RX5-TAX1BP3 showed off-target activity
c.2991+1655A>G mutation site and identified a pair of highly for Cas12f at 57, 51 and 19 loci, respectively, which were similar to
potent sgRNAs (Fig. 3e). We then constructed an rAAV vector car- or smaller in number than 57, 87 and 27, respectively, for AsCas12a
rying the Cas12f1_ge4.1 system, and the deletion-inducing effi- (Fig. 4e and Supplementary Table 4). Intrinsically, Cas12f would be
ciency of Cas12f was compared with that of SaCas9 (EDIT101, a expected to show fewer off-target sites than Cas12a because of the
gene therapeutic agent under clinical trial) in HEK293T cells. We more restricted preference of PAMs36. We then validated the nine
observed higher levels of deletions on agarose gels for the Cas12f potential off-target sites for RPL32P3 by measuring Cas12f- and
system (Fig. 3f). Quantitative analysis using droplet digital PCR AsCas12a-mediated indel frequencies. The indel frequencies at
showed a 46% higher deletion rate of Cas12f, compared with the on-target site were similar for Cas12f and AsCas12a. Similarly,
EDIT101 (Fig. 3g). These results indicate that Cas12f might provide the indel frequencies at off-target sites were not significantly dif-
a versatile and valid genome-editing platform for gene therapy. ferent between the two CRISPR systems, although Cas12f showed
When using an elongation factor (EF)-1α core promoter, a slightly higher off-target activity at the sites with a mismatch in the
bovine growth hormone (BGH) poly(A) signal sequence, a U6 pro- PAM-distal region, in line with Fig. 4a. In addition, a certain level of
moter and an XTEN linker between sequences encoding Cas12f1 indel frequencies was observed for noncanonical TTTR PAM, such
and a potential fusion partner are used, we have an upper limit as GTTG and ATTG, for both enzymes. However, the overall indel
of approximately 2.1 kb for a fusion partner gene for AAV deliv- frequencies at the investigated sites were similar between Cas12f
ery. Considering the sizes of genes encoding validated regulators, and AsCas12a, indicating that Cas12f shows high genome-editing
we propose that the Cas12f system could provide a scaffold for specificity comparable to Cas12a (Fig. 4f). The Cas12f system
various applications including CRISPR interference (CRISPRi)40, not only recognized fewer off-target sites, but also resulted in
CRISPR activation (CRISPRa), base editing8,9, prime editing10 lower off-target/on-target indel frequency ratios. Despite the lower
and site-specific epigenetic regulations5,6 (Fig. 3h). The possibility off-target activity, Cas12f showed long deletions (up to ~10 kb), as is
of such applications was explored in a CRISPRa strategy, where observed for SpCas9 and AsCas12a (Supplementary Fig. 8), which
dCas12 (D510A) fused to VP64 activated transcriptional expression requires further scrutiny45.
of OCT4 gene in a gRNA-dependent manner (Fig. 3i).
Discussion
Genome-editing specificities of Cas12f. Considering the persistent Cas12f1 has an extra-long gRNA for its compact protein size,
activity of Cas12f in cells (Fig. 3c,f,g), it is particularly important to which might be related to Cas12f1’s ssDNA cleavage activity. Our
examine the specificity of this system. First, we assessed the activity engineered sgRNA_ge4.1, although still a little bit longer, was
of Cas12f when gRNA_ge4.1 contained single- or adjacent two-base structurally similar to the crRNA used by Cas12a or Cas12j. This
mismatches with the protospacer complementary sequence. Certain architecture was obtained mainly by trimming the 5′-tracrRNA,
1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue, royalties,
rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal content cannot be
used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any other, institutional
repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or content on
this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature may revoke this
licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied with
respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law, including
merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed from
third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com