Simple and reliable in situ CRISPR-Cas9 nuclease visualization tool is ensuring efficient editing in Streptomyces species

Abstract

CRISPR-Cas9 technology has emerged as a promising tool for genetic engineering of Streptomyces strains. However, in practice, numerous technical hurdles have yet to be overcome when developing robust editing procedures. Here, we developed an extension of the CRISPR-Cas toolbox, a simple and reliable cas9 monitoring tool with transcriptional fusion of cas9 nuclease to a beta glucuronidase (gusA) visual reporter gene. The Cas9-SD-GusA tool enables in situ identification of cells expressing Cas9 nuclease following the introduction of the plasmid carrying the CRISPR-Cas9 machinery. Remarkably, when the Cas9-SD-GusA system was applied under optimal conditions, 100% of the colonies displaying GusA activity carried the target genotype. In contrast, it was shown that the cas9 sequence had undergone major recombination events in the colonies that did not exhibit GusA activity, giving rise to “escaper colonies” carrying unedited genotype. Our approach allows a simple detection of “escaper” phenotype and serves as an efficient CRISPR-Cas9 optimisation tool.

Full text

Journal of Microbiological Methods 200 (2022) 106545
Available online 1 August 2022
0167-7012/© 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Simple and reliable in situ CRISPR-Cas9 nuclease visualization tool is
ensuring efcient editing in Streptomyces species
Alen Pˇ
seniˇ
cnik , Roman Reberˇ
sek , Lucija Slemc , Tim Godec , Luka Kranjc , Hrvoje Petkovi´
c
*
a
University of Ljubljana Biotechnical Faculty, Chair of Biotechnology, Microbiology and Food Safety, Jamnikarjeva ulica 101, Ljubljana, Ljubljana, SI 1000, Slovenia
ARTICLE INFO
Keywords:
CRISPR-Cas9
gusA visual screening
“CRISPR escaper colonies”
Streptomyces
ABSTRACT
CRISPR-Cas9 technology has emerged as a promising tool for genetic engineering of Streptomyces strains.
However, in practice, numerous technical hurdles have yet to be overcome when developing robust editing
procedures. Here, we developed an extension of the CRISPR-Cas toolbox, a simple and reliable cas9 monitoring
tool with transcriptional fusion of cas9 nuclease to a beta glucuronidase (gusA) visual reporter gene. The Cas9-
SD-GusA tool enables in situ identication of cells expressing Cas9 nuclease following the introduction of the
plasmid carrying the CRISPR-Cas9 machinery. Remarkably, when the Cas9-SD-GusA system was applied under
optimal conditions, 100% of the colonies displaying GusA activity carried the target genotype. In contrast, it was
shown that the cas9 sequence had undergone major recombination events in the colonies that did not exhibit
GusA activity, giving rise to “escaper colonies” carrying unedited genotype. Our approach allows a simple
detection of “escaper” phenotype and serves as an efcient CRISPR-Cas9 optimisation tool.
1. Introduction
Streptomyces species and taxonomically closely related bacteria have
been one of the most abundant sources of biologically active drugs for
over 70 years (Myronovskyi and Luzhetskyy, 2016; Horbal et al., 2018;
Tong et al., 2020). To further exploit this potential, it is of key impor-
tance to develop advanced gene tools, since Streptomyces species are
known to be difcult to engineer (Paradkar et al., 2003; Mitousis et al.,
2020). Despite their morphological and physiological diversity, CRISPR-
Cas technology has emerged as a promising tool for the genetic engi-
neering of Streptomyces strains (Lee et al., 2019). The rst CRISPR tools
for Streptomyces were reported in 2015 by four independent research
groups (Cobb et al., 2015; Huang et al., 2015; Tong et al., 2015; Zeng
et al., 2015; Zhao et al., 2020). Since then, the use of CRISPR-Cas tools in
Streptomyces species has grown signicantly (Alberti and Corre, 2019).
However, in practice, a number of technical hurdles must still be over-
come when attempting to develop efcient and robust CRISPR-Cas
procedures for any bacterial species (Ye et al., 2020). The CRISPR sys-
tem is preferentially delivered to Streptomyces species via high copy-
number plasmids (Wang et al., 2020; Wlodek et al., 2017), and
considering that Cas9 nuclease can represent a serious burden to the
host, the reproducibility of CRISPR-Cas9 editing methods is often very
low. For this reason, when developing a CRISPR-Cas tool for any
Streptomyces species, it is important to evaluate the Cas9 expression level
from the selected plasmid/promoter, thus ensuring efcient editing
without excessive toxicity to the host. Classically, a set of trans-
formations with control plasmid constructs is performed when the
CRISPR plasmid is introduced into Streptomyces. The toxicity of active
Cas9 nuclease to the host can be evaluated indirectly based on the
substantial difference that exists between the number of viable trans-
formants that are obtained when transforming the culture with the
control plasmid (usually empty plasmid vector or vector lacking gRNAs)
and those with an active Cas9-gRNA plasmid vector (Huang et al., 2015;
Ye et al., 2020; Zhang et al., 2020; Jiang et al., 2017). However, the
indirect monitoring of Cas9 activity based on a reduced number of viable
bacterial colonies is a work-intensive and time-consuming approach.
Recently, the use of CRISPR-Cas technology has also been reported in
Streptomyces rimosus (Jia et al., 2017), which has been used for oxytet-
racycline (OTC) production for over 70 years (Petkovi´
c et al., 2006). We
aimed to introduce this technology in our laboratory, however, we
observed low efcacy of CRISPR-Cas9 in S. rimosus. We also observed
frequent recombination events when pIJ101-based high copy-number
Abbreviations: SD, Shine-Dalgarno sequence; X-Gluc, 5-bromo-4-chloro-3-indolyl-beta-D-glucuronic acid; OTC, oxytetracycline; StAp, secretory tripeptidyl
aminopeptidase gene; CRISPR, clustered regularly interspaced short palindromic repeats.
* Corresponding author.
E-mail address: Hrvoje.Petkovic@bf.uni-lj.si (H. Petkovi´
c).
Contents lists available at ScienceDirect
Journal of Microbiological Methods
journal homepage: www.elsevier.com/locate/jmicmeth
https://doi.org/10.1016/j.mimet.2022.106545
Received 19 July 2022; Received in revised form 27 July 2022; Accepted 27 July 2022

Journal of Microbiological Methods 200 (2022) 106545
2
plasmid was used as delivery vector for CRISPR-Cas9 in S. rimosus.
Therefore, to better understand the underlying cause of the low repro-
ducibility of the CRISPR-Cas9-based editing method in S. rimosus and to
improve its robustness, we assessed the expression of the cas9 gene by
applying the GusA reporter system, a visually quantiable marker
encoding beta-glucuronidase, which is commonly used in Streptomyces
species (Lee et al., 2019; Wang et al., 2020; Jefferson, 1989; Myr-
onovskyi et al., 2011; Siegl et al., 2013). We constructed a plasmid
composed of the cas9 gene that is transcriptionally coupled to a pro-
motorless gusA gene, thereby ensuring the correlation of the colour
signal with cas9 transcription. As a model system and to evaluate the
efcacy and robustness of the developed method, we carried out the
deletion of the stAp gene, encoding a secretory tripeptidyl aminopepti-
dase (StAp), which was identied in our laboratory as one of the most
abundant extracellular proteins in S. rimosus. Remarkably, when the
Cas9-SD-GusA system was applied, 100% of the colonies displaying
GusA activity showed the correct genotype, thus demonstrating the high
reproducibility of the developed method. Therefore, we have proven
that this approach can serve as an efcient CRISPR-Cas9 optimisation
tool.
2. Materials and methods
2.1. Bacterial strains, plasmids and cultivation methods
The bacterial strains and plasmids used in the present study are listed
in Table S1. E. coli DH10β cells were used for cloning. E. coli ET12567/
pUB307 strain was used for conjugal transfer of plasmids to S. rimosus
strain. We used Streptomyces rimosus strain ATTC 10970 (NRRL 2234)
from the American Type Culture Collection (R7 in most publications)
(Pethick et al., 2013) with a deletion of the otc gene cluster, designated
as ATCCΔotc (Pikl et al., 2021). S. rimosus ATCCΔotc colonies appeared
completely white on MS agar plates, which simplies the observation of
the GusA phenotype. E. coli strains were grown in 2TY (yeast extract-
tryptone) medium and cultivated at 28 ◦C. S. rimosus was cultivated at
28 ◦C on MS agar medium or in liquid TSB (Kieser et al., 2000) for total
genomic DNA isolation. E. coli cultures were supplemented with 100
μ
g/
mL apramycin (apr) and 50
μ
g/mL kanamycin (kan) when required. 30
μ
g/mL of thiostrepton (tio) was used for the selection of S. rimosus. After
conjugation, E. coli was eliminated using 35
μ
g/mL nalidixic acid (nal).
5-bromo-4-chloro-3-indolyl-beta-D-glucuronic acid (X-gluc), used for
visualization of GusA activity, was supplied by X-gluc Direct (Spain). A
stock solution (0,5 M) of X-gluc was prepared by diluting the appro-
priate amount of X-gluc in DMSO. When adding X-gluc directly into MS
agar plates, we used a nal concentration of 4 mM. Alternatively, to
enhance the intensity of GusA visualization, MS agar plates were over-
laid with 4 mL of 30 mM X-gluc solution.
2.2. Cas9-SD-GusA screening system construction
Primers (Table S2) were supplied by Integrated DNA Technologies
(IDT, USA) and contained 22–45 bp overlapping regions to enable
homology-based cloning procedures. We used the repliQa HiFi Tough-
Mix DNA polymerase mix (Quantabio, USA) for all PCR amplications.
Genomic DNA was isolated using the peqGOLD Bacterial DNA Isolation
Kit (VWR, USA). Plasmid isolation was performed using an EZNA
Plasmid DNA Mini Kit (Omega Bio-tek, USA). Plasmid constructs and
graphics used in this study were designed using Geneious R10 and
R11.1.5 software (https://www.geneious.com).
To allow efcient selection of the plasmid in S. rimosus, we rst
introduced a thiostrepton resistance cassette (tsr) into the pREP_P1_cas9
plasmid (Fig. 1A), which was kindly provided by Novartis/Lek (Mrak
et al., 2018). The tsr cassette was amplied from the plasmid pAB04
(Carrillo Rinc´
on et al., 2018) and inserted, using a T4 DNA ligase
(Thermo Fischer Scintic, USA), into a pREP_P1_cas9 plasmid digested
with BspOI and ClaI (FastDigest, Thermo Fischer Scintic, USA)
(Fig. 1A), thus generating the plasmid construct pREP_P1_cas9_tsr
(Fig. 1B). pREP_P1_cas9_tsr was linearised with SpeI (FastDigest, Thermo
Fischer Scintic, USA) (Fig. 1B) to allow for further insertions. The re-
gion containing the gusA sequence was amplied from plasmid pSET-
GUS (Myronovskyi et al., 2011). The Shine-Dalgarno sequence (SD),
identical to the SD sequence at the cas9 transcription start site, was
introduced upstream of the gusA start codon using the SD-GusA-Fw
primer. Due to the small size of the stAp gene deletion (1422 bp), we
designed a single sgRNA that targeted the 5′side of stAp coding sequence
using Geneious R10 software (Fig. 3C). The gRNA cassette was syn-
thesised by ATG Biosynthetics GmbH (Germany). Then, a 20 bp-gRNA
region targeting stAp was introduced into the NcoI-digested (FastDi-
gest, Thermo Fischer Scintic, USA) gRNA cassette as ssDNA with the
NEBuilder reaction (New England Biolabs, USA), similarly to what has
been described by Tong et al. (Tong et al., 2019), and PCR amplied
(Fig. 1D). This approach allows for fast replacement of the targeted
sequence. Upstream (UP, 1121 bp) and downstream (DOWN, 1094 bp)
homology regions that begin in close proximity to the stAp gene (Fig. 3C)
Fig. 1. A. pREP_P1_cas9 plasmid (Mrak et al., 2018) with marked ClaI and BspOI restriction sites. B. Plasmid construct pREP_P1_cas9_tsr with an inserted thiostrepton
resistance cassette (tsr) and marked SpeI restriction site. C. Final plasmid construct pREP_cas9_SD_gusA_stAp with gusA, gRNA part and homology regions stAp_UP
(UP) - stAP_DOWN (DOWN) inserted into a SpeI restriction site. D. Homology based cloning strategy used in pREP_cas9_SD_gusA_stAp plasmid assembly; gRNA,
stAp_UP and stAp DOWN parts were rst joined with OE-PCR. Then, the OE-PCR amplicon was joined with a gusA PCR fragment and SpeI digested pREP_P1_cas9_tsr
plasmid to construct pREP_cas9_SD_gusA_stAp.
A. Pˇ
seniˇ
cnik et al.

Journal of Microbiological Methods 200 (2022) 106545
3
were amplied from the genomic DNA of S. rimosus ATCC 10970Δotc.
Because assembling many fragments is always a challenge, we sought to
minimise the number of fragments that need to be assembled in a single
reaction. This was achieved by combining the PCR method of overlap
extension (OE-PCR) (Nelson and Fitch, 2011), followed by a nal as-
sembly using a homology-based cloning technique, SLiCE cloning
(Zhang et al., 2012). The stAp_UP homology, stAp_DOWN homology,
and stAp_gRNA were rst joined into an OE-PCR fragment (2625 bp)
(Fig. 1D). The OE PCR reaction consisted of equimolar amounts of each
part, initially run for 10 cycles without primers. After the rst step, the
external primers (stAp_gRNA_Fw and stAp DOWN_Rw) were added to
the reaction and run for another 30 cycles (Fig. 1D). The stAp OE-PCR
fragment was then joined to the gusA PCR fragment and SpeI-linear-
ized pREP_P1_cas9_tsr in a SLiCE reaction to obtain the nal plasmid
construct pREP_cas9_SD_gusA_stAp (Fig. 1C).
Fig. 2. A. Design of the Cas9-SD-GusA system for the co-transcriptional evaluation of Cas9 expression with GusA. Following transcription from the P1 promoter, Cas9
and GusA enzymes are simultaneously translated from a single mRNA. After translation and subsequent protein folding Cas9-sgRNA complex binds to the target
sequence and creates a double-stranded break inside the target gene (stAp gene). The simultaneously translated GusA enzyme catalyses the conversion of the X-gluc
chromogenic substrate (separately added to medium), yielding blue-coloured colonies. The double-stranded break is then repaired via homology directed repair
(HDR), facilitated by the HDR template, also supplied by the plasmid. B. Relevant part of the pREP_cas9_SD_gusA_stAp plasmid; the cas9-SD-gusA transcription unit is
followed by a sgRNA unit and a HDR template – stAp_UP and stAp DOWN homology regions. C. cas9-SD-gusA junction on mRNA; the gusA gene is out of the reading
frame in relation to the upstream gene cas9. These two genes are also separated by two stop codons, a HindIII restriction site and a Shine-Dalgarno sequence (SD) –
identical to SD, constituting the P1 promoter. HDR template: homology directed repair template; PAM: protospacer adjacent motif; sgRNA: single guide RNA.
A. Pˇ
seniˇ
cnik et al.

Journal of Microbiological Methods 200 (2022) 106545
4
2.3. Microbiological methods
2.3.1. Intergeneric conjugation and screening of exconjugants
Conjugation was performed to introduce CRISPR-Cas9 carrier plas-
mids into the recipient S. rimosus ATCC Δotc. pREP_cas9_SD_gusA_stAp
was rst electroporated into E. coli ET12567/pUB307 (Flett et al., 1997).
The plasmids were then transferred to S. rimosus using the classical
E. coli - Streptomyces conjugation method (Kieser et al., 2000; Flett et al.,
1997). The conjugation mixture was spread onto MS agar plates sup-
plemented with 10 mM MgCl
2
and 10 mM CaCl
2
. After 8 h of incubation
at 28 ◦C, plates were overlaid with 2 mL aqueous solution of tio +nal at
a nal concentration of 30
μ
g/mL. After 6 days of incubation at 28 ◦C,
emerging exconjugants (Fig. S1) were patched onto fresh MS +tio +nal
plates.
2.3.2. Identication of colonies with successful deletion of stAp gene
Visual detection of GusA activity was carried out by adding X-gluc
directly to MS plates to a nal concentration of 4 mM or by overlaying
cultures after a 6-day incubation period with 4 mL of a 30 mM X-gluc
solution (Fig. 3A). Eight exconjugants that displayed blue colouration
were selected for verication of stAp gene deletion. We selected 10
exconjugants that did not display blue colouration after growth on MS +
tio +X-gluc plates for further investigation. Genomic DNA was isolated
from both groups of exconjugants, and PCR was performed with primers
ΔstAp_Fw and ΔstAp_Rw to conrm stAp deletion (Fig. 3). The plasmid
construct pREP_cas9_SD_gusA_stAp was not cleared from S. rimosus
cultures before PCR was performed. Therefore, both primers for the
conformation of stAp deletion were designed to anneal outside the ho-
mology regions (Fig. 3C).
2.3.3. S. rimosus – E. coli plasmid rescue experiment
We analysed the integrity of the plasmid constructs obtained directly
from the exconjugants. Genomic DNA was isolated from S. rimosus col-
onies with conrmed stAp deletion. These colonies showed GusA activity
and were termed “blue colonies”. Genomic DNA was also isolated from
colonies in which stAp deletion could not be conrmed by PCR. These
colonies showed no GusA activity and where thus termed “white col-
onies” (Fig. 3A). Electrocompetent E. coli DH10β were transformed with
1
μ
l of each total DNA isolate from “blue” and “white” colonies and
spread onto 2TY +apr agar plates. Single E. coli colonies that appeared
after 24 h of incubation at 28 ◦C were inoculated into 2TY +apr liquid
medium and incubated for 16 h before plasmid isolation.
3. Results
3.1. Construction and use of the pREP_cas9_SD_gusA_stAp plasmid for co-
transcriptional screening
Our CRISPR system (Fig. 2A) is based on the plasmid pRE-
P_P1_WT_cas9 (Mrak et al., 2018), containing a pTJU412 (pIJ101-
derived) origin of replication (Sun et al., 2009), a codon-optimised
version of the Streptococcus pyogenes Cas9 coding sequence, and the
theophylline inducible-ErmEp1-based promoter P1 (Myronovskyi et al.,
2016). The pREP_P1_cas9 plasmid contained TraJ and oriT for transfer
functions and an apramycin resistance marker. sgRNA expression is
under the control of a synthetic P21 promoter (Myronovskyi et al., 2016)
and the native S. pyogenes trans-activating CRISPR RNA terminator
(Mrak et al., 2018). For direct chromogenic monitoring of cas9 tran-
scription, we constructed the genetic element Cas9-SD-GusA (Fig. 2C),
where the cas9 gene is followed by two stop codons, an inserted HindIII
restriction site enabling further modications, and a 13 bp part of the P1
promoter containing the Shine-Dalgarno (SD) sequence. To prevent the
translation of a fused Cas9 - GusA dual protein, the start codon of gusA
(ATG) is transcribed in a reading frame different from that of the cas9
gene. The transcription of Cas9-SD-GusA creates a single mRNA that is
translated into two proteins via two ribosome-binding sites (Fig. 2C).
According to this strategy, translation should occur simultaneously, but
independently, for Cas9 and GusA. Therefore, when the GusA substrate
X-gluc is added directly to MS plates, colonies expressing Cas9 nuclease
should also produce a blue colour (Fig. 3A). In our model system, stAp
gene was targeted (S. rimosus ATCC 10970 genome: 6064838
bp–6066259 bp; Slemc et al., 2022), which encodes one of the extra-
cellular peptidases responsible for degrading peptides involved in cell
wall rearrangements in Streptomyces (Butler et al., 1995).
3.2. In situ monitoring of Cas9 expression and conrmation of the stAp
deletion
After completion of the E. coli ET/pUB307 - S. rimosus ATCC
10970Δotc conjugation procedure (Fig. S1), exconjugants which
Fig. 3. A. S. rimosus ATCC 10970Δotc/pRep_P1_cas9_SD_gusA_stAp exconjugant patches on MS agar plates overlaid with a X-gluc solution. Most of the patches
displayed blue colour (b), except two that were not coloured – “white” phenotype (w). B. Gel electrophoresis of PCR-amplied stAp region from genomic DNA of
“blue” (lanes 1–8) and “white” (lanes 9–18) S. rimosus colonies transformed with pREP_cas9_ SD _gusA_stAp plasmid. All tested “blue” colonies of S. rimosus
transformed with pREP_cas9_ SD _gusA_stAp plasmid displayed the ΔstAp genotype (Samples 1–8) with expected band size of 2303 bp. In contrast, and with one
exception, all “white” colonies displayed a WT genotype with a band size of 3717 bp. C. Targeted genomic region containing the stAp gene; the PCR reaction was
performed by the set of primers cPCR_stAp_FW and cPCR_stAp_RW.
A. Pˇ
seniˇ
cnik et al.

Journal of Microbiological Methods 200 (2022) 106545
5
appeared on MS plates were patched onto fresh MS +tio +nal agar
plates. When overlaid with X-gluc, numerous exconjugants developed
blue colouring and could be clearly distinguished from those which
remained uncoloured (Fig. 3A). Colonies with both blue and white
phenotype were selected for further evaluation and therefore patched to
selective SM medium, supplemented with X-gluc. All patched S. rimosus
ATCCΔotc/pRep_P1_cas9_SD_gusA_stAp exconjugants (Fig. 3A) were
analysed for targeted deletion of the stAp gene via PCR with DNA iso-
lated from individual patches. As shown in Fig. 3B, all the S. rimosus/
pRep_P1_cas9_SD_gusA_stAp colonies displaying a blue-coloured
phenotype also showed the expected PCR amplicon (2303 bp), thus
conrming the deletion of the stAp gene. In contrast, only a single un-
colored S. rimosus/pRep_P1_cas9_SD_gusA_stAp colony appeared to have
the desired ΔstAp genotype, whereas the other nine colonies displayed a
WT genotype (3717 bp).
3.3. Identication of plasmid recombination events by the Cas9-SD-GusA
system
As spontaneous resistance to thiostrepton never occurs in S. rimosus,
“white colonies” were subjected to further analysis. Using a plasmid
rescue approach (Methods; 2.3.3), plasmid DNA was isolated and sub-
sequently digested with the PstI restriction enzyme, which cuts the
plasmid into DNA fragments of distinguishable sizes (Fig. 4B). As shown
in the restriction analysis (Fig. 4A), the DNA fragment of the plasmid
DNA encoding most of Cas9, GusA, gRNA, and the entire DNA sequence
Fig. 4. A. Restriction analysis of plasmid DNA isolated via a plasmid rescue approach from S. rimosus exconjugants transformed with pREP_cas9_SD_gusA_stAp. Lanes
1–6: PstI restriction of pDNA isolated from 6 independent “blue” colonies exhibiting GusA activity and with conrmed deletion of the stAp gene. Lanes 7–17: PstI
digest of pDNA isolated from 10 “white” colonies not exhibiting GusA activity and without stAp deletion – WT genotype. B. Predicted restriction pattern after PstI
digestion of plasmid pREP_cas9_SD_gusA_stAp (C) and rearranged plasmid pREP_cas9_SD_gusA_stAp containing a large deletion (D). C. Plasmid map of pRE-
P_cas9_SD_gusA_stAp, showing marked region of spontaneous deletion containing Cas9, GusA and gRNA coding sequences, together with UP and DOWN homology
regions that occurred in tested S. rimosus “white” colonies containing rearranged plasmid (D). D. Plasmid map of rearranged pREP_cas9_SD_gusA_stAp with deleted
CRISPR-specic regions. PstI restriction sites are marked on both plasmids. GR =GeneRuler DNA Ladder Mix (Thermo Fischer Scientic) (k =kbp).
A. Pˇ
seniˇ
cnik et al.

Journal of Microbiological Methods 200 (2022) 106545
6
containing the upper and lower sides of the homology region was
deleted in the “white” colonies. Therefore, an 8,2 kbp fragment within
the plasmid DNA was deleted, resulting in a much smaller plasmid (7856
bp) that only contained essential functions, that is, to ensure replication,
conjugal transfer, and resistance to thiostrepton. The exact DNA
sequence of the plasmids with large deletions was obtained from Sanger
sequencing data (Fig. S3). Surprisingly, all analysed plasmids obtained
from “white” colonies carried an identical deletion. This could indicate
that the recombination event occurred at an early stage of the conju-
gation procedure in the E. coli donor. We evaluated the state of the
pREP_cas9_SD_gusA_stAp plasmid isolated directly from the stored E. coli
ET12567/pUB307 stocks which were used for conjugal transfer of the
plasmid from E. coli into S. rimosus. According to restriction analysis
with the PstI multicutter enzyme, all E. coli ET12567/pUB307 clones
carried the intact plasmid pREP_cas9_SD_gusA_stAp. Therefore, we can
conclude that the plasmid pREP_cas9_SD_gusA_stAp was stably main-
tained in the E. coli ET12567/pUB307 (Fig. S2).
4. Discussion
The difculties associated with the use of CRISPR-Cas tools in
Streptomyces species often stem from their potential toxicity to the host
cell (Zhao et al., 2020; Ye et al., 2020; Wang et al., 2020; Alberti et al.,
2018; Vento et al., 2019). Editing efciency can vary greatly when
CRISPR-Cas tools are applied in Streptomyces strains (Cobb et al., 2015;
Huang et al., 2015; Zhang et al., 2020). Viable colonies containing either
an unintended edit or the wild-type sequence are often observed after
the transformation of CRISPR-Cas-bearing plasmids into the host cul-
ture. These cells, referred to as “escaper colonies”, often complicate the
identication of target clones that contain the correct edit (Vento et al.,
2019). In various instances, their characterisation suggests that deacti-
vation of the cas nuclease gene or the spacer region is a common escape
mechanism, as seen by the deactivation of cas9 gene in Corynebacterium
glutamicum (Fischer et al., 2012) and type I cas genes in the archaea
Haloferax volcanii, where 77% of escaper colonies had deletions or
mutations in the cas gene cluster (Liu et al., 2017). Thus, Cas9 toxicity
not only results in a large proportion of non-viable cells but also exerts
selection pressure for recombination events that preferentially abolish
Cas9 activity (Liu et al., 2017). Although overexpression of CRISPR
components often leads to severe cytotoxicity, a very weak level of
expression can result in inefcient genome editing (Ye et al., 2020). To
establish the optimal level of expression for the CRISPR system, the most
important factors are the choice of the plasmid replicon and promoter
sequences. We focused on a pIJ101 replicon-based vector, which,
together with pSG5, is one of the two most commonly used CRISPR-
associated replicons in Streptomyces species (Cobb et al., 2015; Huang
et al., 2015; Tong et al., 2015; Zeng et al., 2015; Zhao et al., 2020). In
addition to the drawback posed by the pSG5 replicon, which is reported
to be problematic when editing highly repetitive genomic DNA regions
(Wang et al., 2020; Wlodek et al., 2017), the advantage of the pIJ101-
based replicon is the rapid loss of the plasmid during subcultivation
procedures carried out in the absence of antibiotic pressure following
CRISPR-mediated editing.
However, we observed that the pIJ101-based plasmid vectors dis-
played a relatively high degree of instability in S. rimosus (Carrillo
Rinc´
on et al., 2018). These observations led us to believe that the se-
lection pressure of an active Cas9-gRNA complex expressed from a
pIJ101-based high-copy plasmid could also lead to recombination
events, thus reducing the efcacy of CRISPR tools. To reduce the puta-
tive selection pressure caused by high-copy vectors, we used the pRE-
P_P1_Cas9 plasmid (Mrak et al., 2018), where the cas9 gene is placed
under the control of a weak P1 promoter with an additional riboswitch
(Mrak et al., 2018). The use of strong promoters for gRNA expression is
seemingly non-toxic to cells, with stronger promoters showing increased
editing efciency (Zhang et al., 2020). For this reason, we focused only
on monitoring Cas9 nuclease expression. To gain visual insight into the
performance of the combination of the P1 promoter, the pIJ101-based
replicon pTJU412, and the Cas9 nuclease plasmid in S. rimosus, we
transcriptionally fused gusA (Myronovskyi et al., 2011) to cas9. In this
way, we not only monitored cas9 gene expression, but also ensured the
selection of transformants containing the cas9 gene. To avoid low levels
of the reporter protein, GusA, which could reduce the functionality of
our system, the riboswitch region was not included upstream of the
GusA coding sequence. Therefore, we aimed to achieve “full strength”
translation of the gusA gene. In contrast, potential toxicity from the
translation of the cas9 gene is controlled by the riboswitch, thus
reducing the translation efciency of the cas9 gene.
Importantly, we observed that deletion of the stAp gene occurred at a
high frequency when no theophylline was added to induce the P1-
riboswitch, therefore at low Cas9 expression levels. The addition of
theophylline only resulted in reduced growth of S. rimosus cultures. A
negative effect of theophylline addition in inducing Cas9 translation was
recently reported in S. coelicolor M145 and S.lividans TK24 (Ye et al.,
2020). Therefore, we conclude that in an S. rimosus background, the
theophylline riboswitch is sufciently leaky even without the addition of
its inducer, ensuring sufcient Cas9 expression when combined with the
P1 promoter and a high-copy pIJ101 replicon. These results also suggest
that the high toxicity of the Cas9 protein is the main factor affecting the
efcacy and reproducibility of CRISPR-Cas9-based tools in S. rimosus.
Altogether, we observed that the expression of the gusA reporter gene
coincided entirely with the occurrence of the target CRISPR-Cas9 editing
event in independent transformants, thereby conrming the efcacy of
our visual exconjugant chromogenic screening approach. Interestingly,
we also observed a relatively large proportion of S. rimosus exconjugants
that did not display GusA activity (“white” colonies) and, without
exception, contained a rearranged CRISPR-Cas9 plasmid. Independent
plasmid isolates from the so-called “escaper colonies” in an S. rimosus
background underwent extensive recombination events, in which most
CRISPR-specic parts of the carrier plasmid were deleted.
5. Conclusion
The Cas9-SD-GusA system, where the GusA reporter system is tran-
scriptionally fused to a Cas9 nuclease that we have developed, is a very
efcient system contained within a single plasmid. We demonstrated
that in S. rimosus, GusA activity directly correlates with the efcacy of
CRISPR-Cas9-mediated editing. Our results demonstrate that this
method can greatly simplify the optimisation of CRISPR-Cas tools and
procedures in Streptomyces species, including the optimisation of culti-
vation conditions, selection of suitable vectors, and other gene tools
such as promoters of appropriate strength. Importantly, the method we
have developed allows for simple and rapid visual selection of single
Streptomyces colonies that express Cas9, while simultaneously enabling
the rapid identication of undesired “escaper colonies”. With a certain
re-adjustments, such as a selection of suitable vector, resistance gene
and G +C DNA content readjustment, Cas9-SD-GusA system that we
developed can be applied to a diverse range of microbial hosts for the
purpose of CRISPR-Cas tools efcacy optimization.
Funding
This study was supported by the Slovenian Research Agency
(P4–0116). A.P. is supported by a young researcher grant from the
Slovenian Research Agency (No. 53621).
CRediT authorship contribution statement
Alen Pˇ
seniˇ
cnik: Conceptualization, Methodology, Investigation,
Writing – original draft, Writing – review & editing, Visualization.
Roman Reberˇ
sek: Methodology, Investigation, Writing – original draft,
Writing – review & editing. Lucija Slemc: Methodology. Tim Godec:
Visualization. Luka Kranjc: Methodology, Supervision, Writing –
A. Pˇ
seniˇ
cnik et al.

Journal of Microbiological Methods 200 (2022) 106545
7
review & editing. Hrvoje Petkovi´
c: Conceptualization, Resources,
Writing – original draft, Writing – review & editing, Supervision, Project
administration, Funding acquisition.
Declaration of Competing Interest
The authors have no conicts of interest to disclose.
Data availability
Data will be made available on request.
Acknowledgements
We thank Novartis/Lek Mengeˇ
s for kindly providing us with plasmid
pREP_P1_cas9 (Mrak et al., 2018) and Andriy Luzhetskyy from the
Helmholtz Institute for Pharmaceutical Research Saarland, who pro-
vided us with plasmid pSETGUS (Myronovskyi et al., 2011)
.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.mimet.2022.106545.
References
Alberti, F., Corre, C., 2019. Editing streptomycete genomes in the CRISPR/Cas9 age. Nat.
Prod. Rep. 36 (9), 1237–1248. https://doi.org/10.1039/c8np00081f.
Alberti, F., Leng, D.J., Wilkening, I., et al., 2018. Triggering the expression of a silent
gene cluster from genetically intractable bacteria results in scleric acid discovery.
Chem. Sci. 10 (2), 453–463. https://doi.org/10.1039/c8sc03814g.
Butler, M.J., Binnie, C., DiZonno, M.A., et al., 1995. Cloning and characterisation of a
gene encoding a secreted tripeptidyl aminopeptidase from Streptomyces lividans 66.
Appl. Environ. Microbiol. 61 (8), 3145–3150. https://doi.org/10.1128/
aem.61.8.3145-3150.
Carrillo Rinc´
on, A.F., Magdevska, V., Kranjc, L., et al., 2018. Production of extracellular
heterologous proteins in Streptomyces rimosus, producer of the antibiotic
oxytetracycline. Appl. Microbiol. Biotechnol. 102 (6), 2607–2620. https://doi.org/
10.1007/s00253-018-8793-z.
Cobb, R.E., Wang, Y., Zhao, H., 2015. High-efciency multiplex genome editing of
streptomyces species using an engineered CRISPR/Cas system. ACS Synth. Biol. 4
(6), 723–728. https://doi.org/10.1021/sb500351f.
Fischer, S., Maier, L.K., Stoll, B., et al., 2012. An archaeal immune system can detect
multiple protospacer adjacent motifs (PAMs) to target invader DNA. J. Biol. Chem.
287 (40), 33351–33363. https://doi.org/10.1074/jbc.M112.377002.
Flett, F., Mersinias, V., Smith, C.P., 1997. High efciency intergeneric conjugal transfer
of plasmid DNA from Escherichia coli to methyl DNA-restricting streptomycetes.
FEMS Microbiol. Lett. 155 (2), 223–229. https://doi.org/10.1111/j.1574-6968.
Horbal, L., Siegl, T., Luzhetskyy, A., 2018. A set of synthetic versatile genetic control
elements for the efcient expression of genes in Actinobacteria. Sci. Rep. 8 (1), 491.
https://doi.org/10.1038/s41598-017-18846-1.
Huang, H., Zheng, G., Jiang, W., et al., 2015. One-step high-efciency CRISPR/Cas9-
mediated genome editing in Streptomyces. Acta Biochim. Biophys. Sin. Shanghai 47
(4), 231–243. https://doi.org/10.1093/abbs/gmv007.
Jefferson, R.A., 1989. The GUS reporter gene system. Nature 342 (6251), 837–838.
https://doi.org/10.1038/342837a0.
Jia, H., Zhang, L., Wang, T., et al., 2017. Development of a CRISPR/Cas9-mediated gene-
editing tool in Streptomyces rimosus. Microbiology (Reading) 163 (8), 1148–1155.
https://doi.org/10.1099/mic.0.000501.
Jiang, Y., Qian, F., Yang, J., et al., 2017. CRISPR-Cpf1 assisted genome editing of
Corynebacterium glutamicum. Nat. Commun. 8, 15179. https://doi.org/10.1038/
ncomms15179.
Kieser, T., Bibb, M.J., Buttner, M.J., et al., 2000. Practical Streptomyces Genetics. Int
Microbiol. John Innes Foundation, Norwich.
Lee, N., Hwang, S., Lee, Y., et al., 2019. Synthetic biology tools for novel decondary
metabolite discovery in Streptomyces. J. Microbiol. Biotechnol. 29 (5), 667–686.
https://doi.org/10.4014/jmb.1904.04015.
Liu, J., Wang, Y., Lu, Y., et al., 2017. Development of a CRISPR/Cas9 genome editing
toolbox for Corynebacterium glutamicum. Microb. Cell Factories 16 (1), 205. https://
doi.org/10.1186/s12934-017-0815-5.
Mitousis, L., Thoma, Y., Musiol-Kroll, E.M., 2020. An update on molecular tools for
genetic engineering of actinomycetes—the source of important antibiotics and other
valuable compounds. Antibiotics 9 (8), 494. https://doi.org/10.3390/
antibiotics9080494.
Mrak, P., Krastel, P., Pivk Lukanˇ
ciˇ
c, P., et al., 2018. Discovery of the actinoplanic acid
pathway in Streptomyces rapamycinicus reveals a genetically conserved synergism
with rapamycin. J. Biol. Chem. 293 (52), 19982–19995. https://doi.org/10.1074/
jbc.RA118.005314.
Myronovskyi, M., Luzhetskyy, A., 2016. Native and engineered promoters in natural
product discovery. Nat. Prod. Rep. 33, 1006–1019. https://doi.org/10.1039/
c6np00002a.
Myronovskyi, M., Welle, E., Fedorenko, V., et al., 2011. Beta-glucuronidase as a sensitive
and versatile reporter in actinomycetes. Appl. Environ. Microbiol. 77 (15),
5370–5383. https://doi.org/10.1128/AEM.00434-11.
Myronovskyi, M., Br¨
otz, E., Rosenkr¨
anzer, B., et al., 2016. Generation of new compounds
through unbalanced transcription of landomycin A cluster. Appl. Microbiol.
Biotechnol. 100 (21), 9175–9186. https://doi.org/10.1007/s00253-016-7721-3.
Nelson, M.D., Fitch, D.H.A., 2011. Overlap extension PCR: an efcient method for
transgene construction. Methods Mol. Biol. 772, 459–470. https://doi.org/10.1007/
978-1-61779-228-1_27.
Paradkar, A., Trefzer, A., Chakraburtty, R., et al., 2003. Streptomyces genetics: a
genomic perspective. Crit. Rev. Biotechnol. 23 (1), 1–27. https://doi.org/10.1080/
713609296.
Pethick, F.E., Macfadyen, A.C., Tang, Z., et al., 2013. Draft genome sequence of the
oxytetracycline-producing bacterium Streptomyces rimosus ATCC 10970. Genome
Announc 1 (2), e0006313. https://doi.org/10.1128/genomeA.00063-13.
Petkovi´
c, H., Cullum, J., Hranueli, D., et al., 2006. Genetics of Streptomyces rimosus, the
oxytetracycline producer. Microbiol. Mol. Biol. Rev. 70 (3), 704–728. https://doi.
org/10.1128/MMBR.00004-06.
Pikl, ˇ
S., Carrillo Rinc´
on, A.F., Slemc, L., et al., 2021. Multiple copies of the
oxytetracycline gene cluster in selected Streptomyces rimosus strains can provide
signicantly increased titers. Microb. Cell Factories 20 (1), 47. https://doi.org/
10.1186/s12934-021-01522-5.
Siegl, T., Tokovenko, B., Myronovskyi, M., et al., 2013. Design, construction and
characterisation of a synthetic promoter library for ne-tuned gene expression in
actinomycetes. Metab. Eng. 19, 98–106. https://doi.org/10.1016/j.
ymben.2013.07.006.
Slemc, L., Jakˇ
se, J., Filisetti, A., et al., 2022. Reference-grade genome and large linear
plasmid of Streptomyces rimosus: pushing the limits of nanopore sequencing.
Microbiol. Spectr. 10 (2), e0243421 https://doi.org/10.1128/spectrum.02434-21.
Sun, Y., He, X., Liang, J., et al., 2009. Analysis of functions in plasmid pHZ1358
inuencing its genetic and structural stability in Streptomyces lividans 1326. Appl.
Microbiol. Biotechnol. 82 (2), 303–310. https://doi.org/10.1007/s00253-008-1793-
7.
Tong, Y., Charusanti, P., Zhang, L., et al., 2015. CRISPR-Cas9 based engineering of
actinomycetal genomes. ACS Synth. Biol. 4 (9), 1020–1029. https://doi.org/
10.1021/acssynbio.5b00038.
Tong, Y., Whitford, C.M., Robertsen, H.L., et al., 2019. Highly efcient DSB-free base
editing for streptomycetes with CRISPR-BEST. Proc. Natl. Acad. Sci. U. S. A. 116
(41), 20366–20375. https://doi.org/10.1073/pnas.1913493116.
Tong, Y., Whitford, C.M., Blin, K., et al., 2020. CRISPR-Cas9, CRISPRi and CRISPR-BEST-
mediated genetic manipulation in streptomycetes. Nat. Protoc. 15 (8), 2470–2502.
https://doi.org/10.1038/s41596-020-0339-z.
Vento, J.M., Crook, N., Beisel, C.L., 2019. Barriers to genome editing with CRISPR in
bacteria. J. Ind. Microbiol. Biotechnol. 46 (9–10), 1327–1341. https://doi.org/
10.1007/s10295-019-02195-1.
Wang, Q., Xie, F., Tong, Y., et al., 2020. Dual-function chromogenic screening-based
CRISPR/Cas9 genome editing system for actinomycetes. Appl. Microbiol. Biotechnol.
104, 225–239. https://doi.org/10.1007/s00253-019-10223-4.
Wlodek, A., Kendrew, S.G., Coates, N.J., et al., 2017. Diversity oriented biosynthesis via
accelerated evolution of modular gene clusters. Nat. Commun. 8 (1), 1206. https://
doi.org/10.1038/s41467-017-01344-3.
Ye, S., Enghiad, B., Zhao, H., et al., 2020. Fine-tuning the regulation of Cas9 expression
levels for efcient CRISPR-Cas9 mediated recombination in Streptomyces. J. Ind.
Microbiol. Biotechnol. 47 (4–5), 413–423. https://doi.org/10.1007/s10295-020-
02277-5.
Zeng, H., Wen, S., Xu, W., et al., 2015. Highly efcient editing of the actinorhodin
polyketide chain length factor gene in Streptomyces coelicolor M145 using CRISPR/
Cas9-CodA(sm) combined system. Appl. Microbiol. Biotechnol. 99 (24),
10575–10585. https://doi.org/10.1007/s00253-015-6931-4.
Zhang, Y., Werling, U., Edelmann, W., 2012. SLiCE: a novel bacterial cell extract-based
DNA cloning method. Nucleic Acids Res. 40 (8), e55 https://doi.org/10.1093/nar/
gkr1288.
Zhang, J., Zhang, D., Zhu, J., et al., 2020. Efcient multiplex genome editing in
streptomyces via engineered CRISPR-Cas12a systems. Front. Bioeng. Biotechnol. 8,
726. https://doi.org/10.3389/fbioe.2020.00726.
Zhao, Y., Li, G., Chen, Y., et al., 2020. Challenges and advances in genome editing
technologies in streptomyces. Biomolecules 10 (5), 734. https://doi.org/10.3390/
biom10050734.
A. Pˇ
seniˇ
cnik et al.