Optimizing genome editing efficiency in Streptomyces fradiae via a CRISPR/Cas9n-mediated editing system

ABSTRACT

Streptomyces fradiae is an important bioresource to produce various antibacterial natural products, however, the time-consuming and labor-intensive genome editing toolkits hindered the construction and application of engineered strains, and this study aimed to establish an efficient CRISPR/Cas9n genome editing system in S. fradiae. Initially, the CRISPR/Cas9-mediated editing tool was employed to replace those awkward genome editing tools that relied on homologous recombination, while the off-target Cas9 exhibited high toxicity to S. fradiae Sf01. Therefore, the nickase mutation D10A, high-fidelity mutations including N497A, R661A, Q695A, and Q926A, and thiostrepton-induced promotor P_tipA were incorporated into the Cas9 expression cassette, which reduced its toxicity. The deletion of single gene neoI and long fragment sequence (13.3 kb) were achieved with efficiencies of 77.8% and 44%, respectively. Additionally, the established tool was applied to facilitate the rapid deletion of nagB, replacement of P_frr with P_ermE*, and integration of exogenous vgbS, with respective efficiencies of 77.8%, 100%, and 67.8%, and all of the above modification strategies benefited neomycin synthesis in S. fradiae. Taken together, this research established an efficient CRISPR/Cas9n-mediated genome editing toolkit in S. fradiae, paving the way for developing high-performance neomycin-producing strains and facilitating the genetic modification of Streptomyces.

IMPORTANCE

This study describes the development and application of a genome editing system mediated by CRISPR/Cas9n in Streptomyces fradiae for the first time, which overcomes the challenges associated with genome editing caused by high GC content (74.5%) coupling with complex genome structure, and reduces the negative impact of “off-target effect.” Our work not only provides a facile editing tool for constructing S. fradiae strains of high-yield neomycin but also offers the technical guidance for the design of a CRISPR/Cas9n mediated genome editing tool in those creatures with high GC content genomes.

INTRODUCTION

Streptomyces fradiae, a member of the Streptomyces family, has been primarily utilized for the production of various antibiotics (1, 2), alkaloids (3), auxin (4), etc. However, the lack of a suitable genome editing tool has been a barrier to enhancing biochemical production in S. fradiae through metabolic engineering. The genome of S. fradiae demonstrates a spontaneous mutation rate exceeding 0.1%, and approximately 0.05% of spores exhibit chromosome deletions (5), without harmful effects on growth (6). Furthermore, S. fradiae possesses a GC content of 74.5% (NCBI ID: ASM870442v1), resulting in the formation of an intricate three-dimensional genetic structure and presenting significant obstacles for genome modification (7).

Recently, the construction of industrial Streptomyces strains has largely relied on editing vectors via homologous recombination, and a number of gene editing strategies have subsequently been derived, including PCR targeting (8–10), I-SceI editing (11, 12), and Cre-loxP editing systems (13). Nevertheless, the common limitation of the aforementioned methods is that they relied on the occasional crossovers in the division phase, which is subject to a great deal of contingency and uncertainty. In particular, given the long lifespan of Streptomyces, the process of subculturing and verifying each generation is time-consuming, and the potential for reverse mutation would result in an incalculable extension of processing time. Furthermore, PCR targeting systems do not allow for seamless editing, the I-SceI editing system has species differences in editing efficiency, and the Cre-loxP editing system needs repetitive preparation. Therefore, an alternative editing tool for S. fradiae genome modification was eagerly awaited.

CRISPR/Cas9, derived from the defensive system in Streptococcus pyogenes (14), has been extensively studied and used as a generalist gene editing tool. In comparison to the aforementioned genome editing methods via homologous recombination, the CRISPR/Cas9-mediated genome editing tool has the distinct advantage of achieving traceless editing and resistant marker deletion simultaneously, negating the necessity for an additional step (15), which enhances the efficiency of the experimental process and has also led to the development of CRISPR/Cas9 editing technology in Streptomyces. For instance, Cobb et al. developed the pCRISPomyces system for genome editing of Streptomyces lividans, which was shown to be capable of deleting large chromosomal fragments (up to 30 kb) with editing efficiencies from 70% to 100% (16). Tong et al. constructed the CRISPR-Cas9 and CRISPR-base editing systems in Streptomyces coelicolor, achieving double strands break (DSB) dependent genome editing and DSB-free genome editing, respectively (17, 18). Je et al. developed the CaExTun platform enabled rapid screening of a suitable promoter to Cas9, and was also employed to modify four model Streptomyces strains (19). The successful application of CRISPR technology in model Streptomyces prompted us to develop the same approach in S. fradiae, the non-model Streptomyces.

However, CRISPR/Cas9 has a tolerance for the mismatch between sgRNA and target DNA, which may result in non-specific cutting (20). This phenomenon is defined as an “off-target effect,” and is also considered a toxicity of Cas9. Meanwhile, high GC content in the genome might reduce the targeting specificity of CRISPR/Cas9 and increase the risk of “off-target effect.” Unfortunately, according to the NCBI genome database, S. fradiae (NCBI ID: ASM870442v1) is one of the Streptomyces with the highest GC content (74.5%), which is vulnerable to “off-target effect.” In light of the above considerations, Cas9n (Cas9 nickase) has been identified as an ideal substitute for Cas9, and the D10A mutated Cas9 only leads to single-strand breaks (21), which reduces the negative impact from the “off-target effect” (22).

Neomycin is an important antibiotic produced by S. fradiae, which is widely used in the fields of veterinary drugs, feed additives, etc. However, the low genetic transformation efficiency hinders metabolic engineering breeding for high-yield production of neomycin. In this study, we aimed to develop a CRISPR/Cas9n-mediated genome editing system in S. fradiae, including an editing vector and optimized conjugation conditions. It was expected not only to provide an efficient tool for targeted genome editing in S. fradiae but also to offer technical guidance for the design of CRISPR/Cas9-related editing tools in those high GC content creatures.

RESULTS

Optimization of conjugation conditions for the transfer from Escherichia coli to S. fradiae

S. fradiae Sf01 is an industrial strain used to produce neomycin, which was isolated and purified for many years, resulting in a more complex genome structure and background. The traditional genome editing plasmid pKC1139, which was based on homologous recombination, presented a significant challenge in terms of transfer into Sf01 by conjugation. The conjugation efficiency was observed to be less than 10⁻⁹. On account of this, the initial investigation tended to rectify the unsuccessful outcome by optimizing the conjugation conditions.

An appropriate solid medium was a crucial factor in the formation of strong strains, and the introduction of Mg²⁺ might enhance the conjugation efficiency (23). We first compared the capability of five solid media including SFM, IWL4, 2CMY, GS-1, and ISP2. In SFM solid medium, Sf01 exhibited optimal growth, with about 2 × 10⁹ spores (Fig. S1A). Subsequently, SFM was further optimized through the addition of a gradient-increasing concentration of MgCl_₂ (0 to 60 mM), with the objective of identifying an appropriate MgCl_₂ supplementation that could enhance the conjugation efficiency. pKC1139 was used as donor plasmids for the conjugation experiment, and about 1 × 10⁸ S. fradiae spores were used as recipients. When SFM contained 50 mM MgCl₂, the conjugation efficiency of pKC1139 was the highest, which was 2.3 × 10⁻⁷ (23 conjugants) (Fig. S1B). SFM media containing 50 mM MgCl₂ was called SFMM media.

On the basis of SFMM media, we also explored the influence of three important parameters on the conjugation efficiency, including the heat shock temperature (Fig. S1C), the antibiotic addition time (Fig. S1D), and the ratio of donor to recipient of conjugation (Fig. S1E). Optimal conditions were determined as 50°C of heat shock, 18 h of antibiotic addition time after conjugation, and 20:1 of donor to recipient, improving the conjugation efficiency to 1.3 × 10⁻⁶ (about 132 conjugants), which ensured the viability of subsequent experiments.

Design of CRISPR/Cas9 vector for genome editing in S. fradiae

NeoI (NCBI ID: CP974_27380) is a repressor for structural genes (neoE-D) transcription, and deletion of neoI might enhance the transcription levels of target genes for neomycin synthesis in S. fradiae. In the beginning, the neoI deletion plasmid was constructed on the basis of pKC1139, named pKC-ΔneoI (Fig. S2A). Although the optimized conjugation conditions permitted pKC-ΔneoI to transfer in Sf01, we did not get any successfully edited S. fradiae strain but reverse mutation one after double crossovers.

CRISPR/Cas9 represents the latest generation of gene editing tools, with a wide range of applications in Streptomyces (18, 21), and its efficacy in DNA transformation has been well documented. Here, the Cas9 editing vector was designed on the basis of pKC1139 by introducing the Cas9 and single-guided RNA (sgRNA) expressing cassette (Fig. 1). The Cas9 expression cassette included the Cas9 gene sequence with optimized codons according to the preference of S. fradiae, the constitutive promoter P_ermE*, and FD terminator. sgRNA expression cassette included another constitutive promotor P_kasO* (removed sequence of ribosome binding site) and λ-T0 terminator. The recombinant vector was named pECas9 (Fig. S2B). After designing the 20 nt for targeting neoI by CRISPy-web online, the sgRNA was linked with upstream and downstream homologous arms of neoI to construct the recombinant plasmid pECas9-ΔneoI. However, no conjugant was obtained after repeated conjugation. By comparing the number of conjugants, using pKC1139, pECas9 without sgRNA, and pECas9-ΔneoI as donor plasmids (Fig. 2B), the lack of sgRNA did not affect the conjugation efficiency, while the addition of sgRNA would severely impair the conjugation. This result was consistent with previous research (24, 25), which suggested that the “off-target effect” was caused by mismatches between the sgRNA-Cas9 complex and DNA strands, and should be blamed for the low conjugation efficiency.

Fig 1.

Expression cassette maps of Cas9 and sgRNA.

Fig 2.

Construction and ability assessment of Cas9 mediated vectors. (A) pTHF-Cas9 plasmid map. (B) Conjugation efficiency of vectors in S. fradiae Sf01. (C) Conjugation efficiency of vectors in S. coelicolor M145. (D) The effect of D10A mutation introduction on conjugation efficiency (statistical analyses by t-test; * P < 0.05, ** P < 0.01, *** P < 0.001; N.S., no significant difference).

Optimization of Cas9 expression cassette to reduce the off-target rate

To achieve the conjugation, the following optimizations were made to the vector pECas9. Previous report has implied that the imperfection of TraJ (conjugation promoting factor) might influence conjugation efficiency (26). Consequently, the incomplete traJ in the vector pECas9 was modified by the full-length one. To reduce the frequency of the “off-target effect,” four high-fidelity mutations (N497A, R661A, Q695A, and Q926A) were introduced in the Cas9 DNA sequence. The four mutated amino acids, N497, R661, Q695, and Q926, were substituted by alanine at the location where they linked with the DNA strand through hydrogen bonds/salt bridges. It has been demonstrated that if a partial mismatch occurred between sgRNA and DNA, the cleavage activity of Cas9 would be inhibited. Cas9 carrying these four mutations had undetectable off-target events and its capacity for cutting DNA remained comparable to that of wild-type Cas9 (27). The recombinant plasmid was named pEHF-Cas9. Moreover, the constitutive promoter P_ermE* was replaced by thiostrepton-inducible promoter P_tipA for Cas9 expression, correspondingly, thiostrepton resistance gene thiR was integrated in the vector, locating on the upstream of Cas9 expressing cassette and owning the opposite direction. The optimized vector was named pTHF-Cas9, and the optimization parts were summarized in Fig. 2A.

To verify the editing ability of pTHF-Cas9 in different Streptomyces, we designed pTHFCas9-ΔneoI for S. fradiae and pTHFCas9-Δsco5087 for S. coelicolor M145, respectively. A large number of transformants were shown in M145/pTHFCas9-Δsco5087, while only one transformant was presented in Sf01/pTHFCas9-ΔneoI. The previous Cas9-mediated editing tools were also tested in S. coelicolor M145 for the deletion of sco5087. The introduction of four high-fidelity mutations did reduce the toxicity of Cas9 (Fig. 2C). However, this was also insufficient for Cas9-mediated gene editing tools to be applied in S. fradiae.

Design of Cas9n-mediated editing vector and evaluating its editing efficiency

To reduce the negative impact of the “off-target effect,” D10A mutation was introduced in pEHF-Cas9 and pTHF-Cas9, and two reconstructed vectors were named pEHF-Cas9n and pTHF-Cas9n, respectively (Fig. S3 and S4).

On the evaluation of these two Cas9n-mediated editing vectors, gene neoI deletion vectors pEHFC9n-ΔneoI and pTHFC9n-ΔneoI were, respectively, constructed, and then applied to establish the recombined strains Sf01ΔneoI (Fig. 3A). The editing efficiencies of pEHFC9n-ΔneoI and pTHFC9n-ΔneoI were 88% and 77%, respectively (Fig. 3B and C).

Fig 3.

Evaluation of pEHF-Cas9n and pTHF-Cas9n for the construction of Sf01ΔneoI. (A) Editing processing of Sf01ΔneoI using pHF-Cas9n. (B) PCR validation of Sf01ΔneoI through pEHF-Cas9n. Line M, DL 5000 DNA marker (5,000, 3,000, 2,000, 1,500, 1,000, 750, 500, 250, and 100 bp); line W, PCR product of Sf01 using ΔneoI-YF/ΔneoI-YR, 948 bp; line 1–9, PCR product of Sf01ΔneoI using ΔneoI-YF/ΔneoI-YR, 597 bp for expected size. (C) Effect of thiostrepton concentration on the editing capability of pTHF-Cas9n.

Another comparison was made between the abilities of these two plasmids to knock out the large fragment, and the staurosporine biosynthetic gene cluster was designed as the target for gene deletion (Fig. S5). pTC9n-Δsta and pEC9n-Δsta were designed to knock out the staN-staC region (13.3 kb), while only pTC9n-Δsta achieved the deletion strain, with 44% editing efficiency. Consequently, pTHF-Cas9n was accepted as the editing vector for gene editing in S. fradiae.

Metabolic engineering of S. fradiae for high-level production of neomycin based on pTHF-Cas9n

Gene deletion

Glucosamine-6-phosphate deaminase NagB catalyzes the conversion of GlcN-6P to Fru-6P (28). The deletion of nagB might enhance UDP-GlcNAc concentration, thus improving neomycin production. pTC9n-ΔnagB was designed to construct gene nagB deletion strain, and the assembling process was shown in Fig. 4A. The result showed that nagB was knocked out successfully with the editing efficiency of 77.8% (Fig. 4B), and the recombinant strain was named as Sf01ΔnagB. The neomycin potency value of Sf01ΔnagB was 13,158.75 U/mL, which was 17.58% higher than that of Sf01 (11,191.29 U/mL) (Fig. 4C).

Fig 4.

Deletion of gene nagB and its effect on neomycin production. (A) Assembly procession of pTC9n-ΔnagB. (B) PCR validation of Sf01ΔnagB. Line W, PCR product of Sf01 using ΔnagB-YF/ΔnagB-YR, 994 bp; line M, DL 5000 DNA marker (5,000, 3,000, 2,000, 1,500, 1,000, 750, 500, 250, and 100 bp); line 1–9, PCR product of Sf01ΔnagB using ΔnagB-YF/ΔnagB-YR, 447 bp for expected size. (C) Neomycin titer assessment (statistical analyses by t-test; * P < 0.05).

Gene-segment replacement

Overexpression of frr (NCBI ID: CP974_22500, encoding for ribosome recycling factor) was reported to be beneficial for ribosome recycling (29). Therefore, we tended to improve the expression of FRR by replacing its natural promoter with a strong promotor P_ermE*, and attained the vector pTC9n-ΔP_frr::P_ermE* (Fig. 5A). The recombinant strain Sf01ΔP_frr::P_ermE* was established with an efficiency of 100% (Fig. 5B), and transcription level of frr was 3.5-fold higher than the control strain (Fig. 5C). Additionally, neomycin potency value was 12,579.21 U/mL, which was 12.4% higher than that of Sf01 (Fig. 5D).

Fig 5.

Replacement of frr promotor and its effect. (A) Assembly procession of pTC9n-ΔP_frr::P_ermE*. (B) PCR validation of Sf01ΔP_frr::P_ermE*. Line W, PCR product of Sf01 using P_frr-YF/P_frr-YR, 286 bp; line M, DL 5000 DNA marker (5,000, 3,000, 2,000, 1,500, 1,000, 750, 500, 250, and 100 bp); line 1–7, PCR product of Sf01ΔP_frr::P_ermE* using P_frr-YF/P_frr-YR, 619 bp for expected size. (C) Transcription levels of gene frr. (D) Neomycin titer assessment (statistical analyses by t-test; * P < 0.05, ** P < 0.01, *** P < 0.001).

Integration of external gene cassette located on ΔnagB

The lack of oxygen in deep liquid fermentation limits the cell growth and secondary metabolism of S. fradiae, which brings a negative impact on neomycin synthesis. VHb is a hemoglobin from Vitreoscilla, and overexpression of its encoding gene vgbS could improve the absorption rate of oxygen in the anoxic situation (30). The vgbS expression cassette, including gene vgbS, promotor P_ermE*, and signal peptide SP_phoD, was inserted between upstream and downstream homologous arms of pTC9n-ΔnagB, resulting in the vgbS integration vector pTC9n-ΔnagB::P_ermE*-SP_phoD-vgbS (Fig. 6A). Then, gene vgbS expression cassette was introduced into Sf01 genome to replaced nagB, achieved the recombinant strain SF2 with the editing efficiency of 67.8% (Fig. 6B).

Fig 6.

Integration of vgbS expression cassette and its effect on neomycin production. (A) Assembly procession of pTC9n-ΔnagB::P_ermE*-SP_phoD-vgbS. (B) PCR validation of SF2. Line W, PCR product of Sf01 using ΔnagB-YF/ΔnagB-YR, 994 bp; line M, DL 5000 DNA marker (5,000, 3,000, 2,000, 1,500, 1,000, 750, 500, 250, and 100 bp); line 1–7, PCR product of SF2 using ΔnagB-YF/ΔnagB-YR, 1,442 bp for expected size. (C) Neomycin titer assessment (statistical analyses by t-test; * P < 0.05; N.S., no significant difference).

By comparing fermenting results between using the liquid media of 25 mL and 50 mL, discrepancies were presented in the group of 50 mL. While the group of 25 mL showed little difference between Sf01ΔnagB and SF2 (Fig. 6C). Therefore, the integration of the vgbS expression cassette was demonstrated to enhance neomycin production under the oxygen-limited condition, and have no adverse effects under the oxygen-rich condition.

DISCUSSION

CRISPR/Cas9n editing tools have been extensively utilized for the genome modification of Streptomyces, however, there has been no documented evidence of the development or application of genome editing tools in S. fradiae using CRISPR/Cas9n. The majority of studies developing CRISPR editing tools in Streptomyces have employed model organisms as the experimental objects, with S. coelicolor being a particularly favored choice (17). The conjugation of S. coelicolor and S. fradiae with CRISPR/Cas9 editing vectors exhibits distinct states, demonstrating that S. fradiae exhibited high sensitivity to the toxicity of Cas9, indicating that the method of constructing recombinant strain of S. coelicolor using CRISPR/Cas9-mediated editing tool was incompatible in S. fradiae. S. fradiae (NCBI ID: ASM870442v1) has a GC content of 74.5% and is one of the Streptomyces with the highest GC content. In contrast, S. coelicolor M145 (NCBI ID: ASM893130v1) just has a GC content of 72%. It was postulated that GC content might be a determining factor in the compatibility of Cas9. A GC content of 74.5% led to an elevated off-target rate of Cas9 and a concomitant increase in the occurrence of double-strand breaks, thereby constraining the study and development of CRISPR/Cas9-based genome editing tool in S. fradiae. Nowadays, relative researchers have made great efforts to reduce the negative impact of “off-target effects,” including modifying the structure of Cas9 (27), regulating Cas9 expression (31), and improving the specificity of sgRNA (32). During the construction of the pTHF-Cas9 editing vector, four high-fidelity mutations were introduced in the initial vector pECas9. We simulated the interactions between the wild-type SpCas9 and the target DNA–sgRNA duplex, based on PDB accession 4OO8 and 4UN3 (27). By calculating the difference in required energy before and after the introduction of mutations, we found that the introduction of four mutations increased the free energy, thereby reducing the stability of the target DNA–sgRNA duplex (Table S1). It was speculated that the introduction of the four mutations ensured fidelity by reducing the binding stability between sgRNA and target DNA. By comparing the number of conjugants between M145/pECas9-Δsco5087 and M145/pEHFCas9-Δsco5087, the introduction of four high-fidelity mutations did reduce the toxicity of Cas9 (Fig. 2C). Nevertheless, the diminished toxicity variant proved inadequate for the implementation of Cas9 mediated editing tool in S. fradiae until the D10A mutation was introduced. The D10A mutation replaced double-strand breaks with single-strand breaks, so pTHF-Cas9n showed a further decrease in toxicity (Fig. 2D). In addition, the editing efficiency of pEHF-Cas9n was higher than that of pTHF-Cas9n on the construction of Sf01ΔneoI, while only pTHF-Cas9n achieved large fragment knock out in Sf01. This was interpreted that the efficacy of genome editing via Cas9n-mediated tools was limited by the DNA repair capacity of the host strain. In Sf01, this limited capacity allowed continuous cutting of constitutive Cas9n when editing small fragments, simultaneously ensuring the proportion of positive colonies in transformants. However, this capacity was insufficient to repair the continuous cutting of Cas9n when editing large fragments, which resulted in failed editing. In contrast, genome cleavage was only possible in the presence of an inducer, guaranteeing a timely DNA repair. Although the proportion of positive colonies in the deletion of large fragments was less than half, pTHF-Cas9n shows better versatility than pEHF-Cas9n in genome editing of Sf01.

This enabled the development and application of a Cas9n-mediated genome editing tool in S. fradiae but also addressed a gap in genome editing in the Streptomyces family by providing a related editing tool.

CRISPR-mediated genome editing is a process that relies on the collaboration of three key components: the sgRNA, Cas9 endonuclease, and homologous arms. Editing vectors containing homologous arms between 0.5 kb and 2 kb always demonstrate proficiency in genome editing in Gram-positive bacteria (33, 34). The plasmid pTHF-Cas9n contained 1 kb homologous arms, which served as the template for DNA repair process, and its editing efficiency of P_frr replacement (132 bp) was 100%, the nagB and neoI deletion (450 bp of nagB and 351 bp of neoI) were both 78%, the vgbS expression cassette integration (800 bp) was 67%, the sta gene cluster deletion (13.3 kb) was just 44%. The results above revealed an inverse relationship between the editing efficiency of pTHF-Cas9n and the size of edited DNA fragments, while the editing efficiency of large DNA fragments in S. fradiae can be improved by increasing the size of homologous arms. However, enlarging homologous arms increased the size of pTHF-Cas9n (exceeding 16 kb), potentially raising the bar for conjugation and genome editing, that cannot be ignored.

Streptomyces family has been responsible for approximately 70% of antibiotic production globally, representing a significant bioresource (5). To improve the biosynthetic capacity of the target antibiotic, the modification of Streptomyces included the reinforcement of precursor supply pathways (28, 35, 36), optimization of expression regulatory elements (37, 38), and improvement of energy utilization (39, 40). Neomycin is an aminoglycoside antibiotic produced by S. fradiae, and the above strategies were also applied to improve neomycin production in Sf01. The deletion of nagB in this study prevented the conversion of GlcN-6P to Fru-6P, thereby increasing the concentration of UDP-GlcNAc, which led to an improvement in neomycin yield, although it was previously reported that high concentrations of GlcN-6P caused weak growth (41). Furthermore, the effect of nagB deletion and frr overexpression on neomycin synthesis in S. fradiae was comparable, resulting in a 20% increase in neomycin production relative to the original strain. Furthermore, the integration of the vgbS cassette resulted in an enhanced synthesis level of neomycin. These strains served as reliable neomycin-producing strains for subsequent gene stacking and modification of engineering strains, laying the foundation for subsequent increases in industrial production of neomycin.

Taken together, the development of pTHF-Cas9n for genome editing in S. fradiae offers a successful case of applying CRISPR/Cas9n in the Streptomyces family. This study also demonstrates the feasibility of CRISPR/Cas9n-mediated genome editing tools in organisms with high GC content. Moreover, our study has also facilitated the construction of robust and stable engineering strains for the enhancement of antibiotic production.

MATERIALS AND METHODS

Bacterial strains and culture conditions

The strains and plasmids used in this study are listed in Table 1. S. fradiae Sf01 served as the initial strain for the construction of recombinant strains. Escherichia coli XL10-Gold was used to construct recombinant plasmids, and ET12567/pUZ8002 was used as a donor for conjugation.

TABLE 1.

Strains and plasmids used in this study

Strains or plasmids	Description	Source
Strains
S. fradiae Sf01	Wild type	Stored in lab
S. fradiae Sf01ΔneoI	Sf01ΔneoI	This study
S. fradiae Sf01ΔnagB	Sf01ΔnagB	This study
S. fradiae Sf01ΔPfrr::PermE*	Sf01ΔPfrr::PermE*	This study
S. fradiae SF2	Sf01ΔnagB::PermE*-SPphoD-vgbS	This study
S. coelicolor M145	Wild type	(42)
E. coli XL10-Gold	Plasmid construction, weak Tetr	Stratagene, US
E. coli ET12567/pUZ8002	Conjugation intermediate, demethylated strain containing plasmid pUZ8002, Chlr, Kanr	(43)
Plasmids
pSET152	Integrated vector with attP from φC31 located on attB; Apmr	(44)
pKC1139 (truncated traJ)	Streptomyces genome editing vector, Apmr	(45)
pCas9	Genome editing vector, Kanr	(22)
pKC1139 (full-length traJ)	Streptomyces genome editing vector, Apmr	This study
pKC-ΔneoI (truncated traJ)	pKC1139 + HindIII + 1 kb neoI-UHA + XbaI + 1 kb neoI-DHA + EcoRI, Apmr	This study
pKC-ΔneoI (full-length traJ)	pKC1139 + HindIII + 1 kb neoI-UHA + XbaI + 1 kb neoI-DHA + EcoRI, Apmr	This study
pECas9 (truncated traJ)	pKC1139 + PermE* + Cas9 + Tfd + HindIII + PkasO* + sgRNA (no target sequence) + EcoRI, Apmr	This study
pECas9 (full-length traJ)	pKC1139 + PermE* + Cas9 + Tfd + HindIII + PkasO* + sgRNA (no target sequence) + EcoRI, Apmr	This study
pECas9-ΔneoI (truncated traJ)	pECas9 + HindIII + PkasO* + sgRNA (target neoI) + 1 kb neoI-UHA + XbaI + 1 kb neoI-DHA + EcoRI, Apmr	This study
pECas9-ΔneoI (full-length traJ)	pECas9 + HindIII + PkasO* + sgRNA (target neoI) + 1 kb neoI-UHA + XbaI + 1 kb neoI-DHA + EcoRI, Apmr	This study
pTHF-Cas9 (truncated traJ)	pKC1139 + PtipA + HF-Cas9 + Tfd + HindIII + PkasO* + sgRNA (no target sequence) + EcoRI, Thir, Apmr	This study
pTHF-Cas9 (full-length traJ)	pKC1139 + PtipA + HF-Cas9 + Tfd + HindIII + PkasO* + sgRNA (no target sequence) + EcoRI, Thir, Apmr	This study
pTHFCas9-ΔneoI (truncated traJ)	pTHF-Cas9 + HindIII + PkasO* + sgRNA (target neoI) + 1 kb neoI-UHA + XbaI + 1 kb neoI-DHA + EcoRI, Thir, Apmr	This study
pTHFCas9-ΔneoI (full-length traJ)	pTHF-Cas9 + HindIII + PkasO* + sgRNA (target neoI) + 1 kb neoI-UHA + XbaI + 1 kb neoI-DHA + EcoRI, Thir, Apmr	This study
pKC-Δsco5087 (truncated traJ)	pKC1139 + HindIII + 1 kb sco5087-UHA + XbaI + 1 kb sco5087-DHA + EcoRI, Apmr	This study
pKC-Δsco5087 (full-length traJ)	pKC1139 + HindIII + 1 kb sco5087-UHA + XbaI + 1 kb sco5087-DHA + EcoRI, Apmr	This study
pECas9-Δsco5087 (truncated traJ)	pECas9 + HindIII + PkasO* + sgRNA (target sco5087) + 1 kb sco5087-UHA + XbaI + 1 kb sco5087-DHA + EcoRI, Apmr	This study
pECas9-Δsco5087 (full-length traJ)	pECas9 + HindIII + PkasO* + sgRNA (target sco5087) + 1 kb sco5087-UHA + XbaI + 1 kb sco5087-DHA + EcoRI, Apmr	This study
pEHFCas9-Δsco5087 (truncated traJ)	pEHF-Cas9 + HindIII + PkasO* + sgRNA (target sco5087) + 1 kb sco5087-UHA + XbaI + 1 kb sco5087-DHA + EcoRI, Apmr	This study
pEHFCas9-Δsco5087 (full-length traJ)	pEHF-Cas9 + HindIII + PkasO* + sgRNA (target sco5087) + 1 kb sco5087-UHA + XbaI + 1 kb sco5087-DHA + EcoRI, Apmr	This study
pTHFCas9-Δsco5087 (truncated traJ)	pTHF-Cas9 + HindIII + PkasO* + sgRNA (target sco5087) + 1 kb sco5087-UHA + XbaI + 1 kb sco5087-DHA + EcoRI, Thir, Apmr	This study
pTHFCas9-Δsco5087 (full-length traJ)	pTHF-Cas9 + HindIII + PkasO* + sgRNA (target sco5087) + 1 kb sco5087-UHA + XbaI + 1 kb sco5087-DHA + EcoRI, Thir, Apmr	This study
pEHF-Cas9n	pKC1139 + PermE* + HF-Cas9n + Tfd + HindIII + PkasO* + sgRNA (no target sequence) + EcoRI, Apmr	This study
pTHF-Cas9n	pKC1139 + PtipA + HF-Cas9n + Tfd + HindIII + PkasO* + sgRNA (no target sequence) + EcoRI, Thir, Apmr	This study
pEHFC9n-ΔneoI	pEHF-Cas9n + HindIII + PkasO* + sgRNA (target neoI) + 1 kb neoI-UHA + XbaI + 1 kb neoI-DHA + EcoRI, Apmr	This study
pTHFC9n-ΔneoI	pTHF-Cas9n + HindIII + PkasO* + sgRNA (target neoI) + 1 kb neoI-UHA + XbaI + 1 kb neoI-DHA + EcoRI, Thir, Apmr	This study
pEC9n-Δsta	pEHF-Cas9n + HindIII + PkasO* + sgRNA (target sta) + 1 kb sta-UHA + XbaI + 1 kb sta-DHA + EcoRI, Thir, Apmr	This study
pTC9n-Δsta	pTHF-Cas9n + HindIII + PkasO* + sgRNA (target sta) + 1 kb sta-UHA + XbaI + 1 kb sta-DHA + EcoRI, Thir, Apmr	This study
pTC9n-ΔnagB	pTHF-Cas9n + HindIII + PkasO* + sgRNA (target nagB) + 1 kb nagB-UHA + XbaI + 1 kb nagB-DHA + EcoRI, Thir, Apmr	This study
pTC9n-ΔPfrr::PermE*	pTHF-Cas9n + HindIII + PkasO* + sgRNA (target Pfrr) + 1 kb Pfrr-UHA + (λ-T0 + PermE*) + 1 kb Pfrr-DHA + EcoRI, Thir, Apmr	This study
pTC9n-ΔnagB::PermE*-SPphoD-vgbS	pTC9nΔnagB + ΔXbaI::(PermE* + SPphoD + vgbS) Thir, Apmr	This study

LB soil medium (1% tryptone, 0.5% yeast extract, 1% NaCl, and 2% agar, pH 7.2) was used for the cultivation of E. coli. SFMM soil medium (2% soybean cake powder hydrolysate, 2% mannitol, 50 mM MgCl₂, 2% agar, pH 7.5) was used for the cultivation of S. fradiae and S. coelicolor. Antibiotics (50 mg/L apramycin, 25 mg/L kanamycin, 25 mg/L chloramphenicol, 25 mg/L naphthoquinone acid, or 25 mg/L thiostreptone) were added into the media as required. E. coli was cultivated at 37°C, while S. fradiae and S. coelicolor were cultivated at 30°C. For neomycin production, the seed was cultivated in 25 mL TSBY medium (3% trypticase soy broth, 0.5% yeast extract, 1% NaCl, pH 7.2) for 18 h, and then transferred into neomycin fermentation medium (8% corn starch, 3% peanut cake powder, 2% glucose, 1.2% peptone, 0.6% yeast powder, 0.6% (NH₄)₂SO₄, 0.45% NaCl, pH 7.2–7.5) for 7 days. All of the fermentation experiments were performed in triplicate.

The formulation of other culture media for testing the effectiveness of S. fradiae cultivation is listed as follows. IWL4: 1% soluble starch, 0.1% K₂HPO₄, 0.1% NaCl, 0.2% (NH₄)₂SO₄, 0.2% CaCO₃, 0.4% tryptone, 0.1% yeast extract, 0.1% MgSO₄·7 H₂O, 0.8% MgCl₂, 2% agar, pH 7.2; 2CMY: 1% soluble starch, 0.2% tryptone, 0.1% NaCl, 0.2% K₂HPO₄, 0.2% CaCO₃, 2% agar, pH 7.2; GS-1: 2% soluble starch, 0.1% KNO₃, 0.05% K₂HPO₄, 0.05% MgSO₄·7 H₂O, 0.001% FeSO₄·7H₂O, 0.05% NaCl, 2% agar, pH 7.4; ISP2: 1% malt extract, 0.4% yeast extract, 0.4% glucose, 2% agar, pH 7.2.

Construction of editing plasmids

All primers used in this study are listed in Table 2. To construct pKC-ΔneoI, pKC1139 was double digestion with Hind III and EcoR I, obtaining a linearized vector. Homologous arms were obtained by PCR amplification with pKCΔneoI-UF/pKCΔneoI-UF and pKCΔneoI-DF/pKCΔneoI-DF using the Sf01 genome as the template. 2× MultiF Seamless Assembly Mix (ABclonal Technology) was used to assemble all of the linearized vector and DNA segments in this study.

TABLE 2.

Primers used in this study

Primers	Sequence (5’–3’)
pKCΔneoI-UF	TAAAACGACGGCCAGTGCCAAGCTTGCCAGCGCACCGACCTTGA
pKCΔneoI-UR	GTTCTGCACCCTGAGCCGCTGGTCGAACCTCC
pKCΔneoI-DF	GACCAGCGGCTCAGGGTGCAGAACCTCGCATC
pKCΔneoI-DR	ACAGCTATGACATGATTACGAATTCCCTTCGACGGCCATGAGGTA
ΔneoI-YF	CCGTGCCAGACGCCCTTGT
ΔneoI-YR	CGGCTTCGTTATGGACAACCC
H3-F	TAAAACGACGGCCAGTGCCAAGCTT
T0-agR	TCCAGTAATGACCTCAGAACTCCATC
ΔneoI-sgR	GTCGTCTGCGCTCGTCCGGAGGCCACGACTTTACAACACC
ΔneoI-sgF	TCCGGACGAGCGCAGACGACGTTTTAGAGCTAGAAATAGCAAG
PE-F1	CGTCGTGACTGGGAAAACCCTGCGGTCGATCTTGACGGCT
fd-R	GTTATGTTGATCGGCACTTTGAGCCTCAGCGATCGAATATA
traJ-GF	CGCTATAATGACCCCGAAGCAG
traJ-GR	GTTCGGTGATGCCACGATCC
traJ-F	CTGCTTCGGGGTCATTATAGCG
traJ-R	GGATCGTGGCATCACCGAAC
PtipA-F	CGTCGTGACTGGGAAAACCCTTTATCGGTTGGCCGCGAGATT
PtipA-R	ATGGAGTACTTCTTGTCCATATGTCCGCTCCCTTCTCTGA
Cas9-F	ATGGACAAGAAGTACTCCATCG
N497A-R	CTTGTCGAAGGCAGTCATCCGCTCAATGAAGG
N497A-F	CGGATGACTGCCTTCGACAAGAACCTGCCTAAC
R661A-R	TGCGGGAGAGGGCGCCCCACCCGGTGTACCGC
R661A-F	GGTGGGGCGCCCTCTCCCGCAAGCTGATAAACG
Q695A-R	TGGATGAGGGCCATGAAGTTACGGTTCGCGAA
Q695A-F	TAACTTCATGGCCCTCATCCACGACGACTCCCT
Q926A-R	GCTTGGTGATCGCTCGCGTTTCTACCAGCTGGC
Q926A-F	GAAACGCGAGCGATCACCAAGCACGTCGCGC
D10A-R	CCGACCGAGTTGGTGCCGATCGCGAGGCCGATGGAGTAC
D10A-F	ACTCCATCGGCCTCGCGATCGGCACCAACTCGGTCGGGTG
Δsta-sgR	AATCCCCGCGAGTGCGGAGCGGCCACGACTTTACAACACC
Δsta-sgF	GCTCCGCACTCGCGGGGATTGTTTTAGAGCTAGAAATAGCAAG
Δsta-UF	GAGAGAGAGAGAGGAGAGAGAGCCCGCACCTGCCGATGCTC
Δsta-UR	CCTCCGTCTAGAGACACGCCACCGCCCTGTCGTC
Δsta-DF	GGCGTGTCTCTAGACGGAGGAGGTCGCCAAGCTCT
Δsta-DR	ACAGCTATGACATGATTACGAATTCACAGCTTCACCGAGGGCATG
Δsta-YF	CCGCACGCCACCTGTTCATC
Δsta-YR	GAGGAGCTGTTCGCCCATGT
ΔnagB-sgR	GAGGGGGAAAGGGAGTCACAGGCCACGACTTTACAACACC
ΔnagB-sgF	TGTGACTCCCTTTCCCCCTCGTTTTAGAGCTAGAAATAGCAAG
ΔnagB-UF	GTTCTGAGGTCATTACTGGAGAGCTGCCACTCCGAGACC
ΔnagB-UR	GCGATCCTCTAGAGTCTCCGCCCCAGACCCCCCTTCG
ΔnagB-DF	CGGAGACTCTAGAGGATCGCACCCGCATCAAGACCCT
ΔnagB-DR	ACAGCTATGACATGATTACGAATTCCACCCCCGTGGCCAACATC
ΔnagB-YF	CGCGCAGCCGCCTGACGGA
ΔnagB-YR	AAGTACCCGGCGAGCTTCA
Pfrr-sgR	GGACAAACCCCAAGACACGCGGCCACGACTTTACAACACC
Pfrr-sgF	GCGTGTCTTGGGGTTTGTCCGTTTTAGAGCTAGAAATAGCAAG
Pfrr-UF	GTTCTGAGGTCATTACTGGACATCAAGGTCGGCATCTGAGCC
Pfrr-UR	GCCTTATTGTTGCGGATCCTTCTTGTCCTGCACGGTGTC
PE-F	AGGATCCGCAACAATAAGGC
PE-R	TATATATTCCTCCTTTCTAATATACCTG
Pfrr-DF	TTAGAAAGGAGGAATATATAGTGGTGATCGAAGAGACCCTCC
Pfrr-DR	ACAGCTATGACATGATTACGAATTCGGCCTTCTGGACGAAGAGC
Pfrr-YF	TCACCCTGTGCCGCGACAAC
Pfrr-YR	TCGGCTTCGAGGAGGGTCTCT
PE-BF	GGGGGGTCTGGGGCGGAGACAGGATCCGCAACAATAAGGC
vgbS-F	GTATATTAGAAAGGAGGAATATATAATGCTGGACCAGCAGACCAT
vgbS-R	TCTTGATGCGGGTGCGATCCTCACTCGACGGCCTGGGCGT
SPPhoD-F	TTAGAAAGGAGGAATATATAGTGACCAGTCGAAACCTCGTTCC
SPPhoD-R	GCCCGGCAGCTCCGGGCCCGGCAGGGGGGCGGAGGTGGTCGC
RT-sigF-F	TACCAGTACGTCCGGAACAC
RT-sigF-R	GCTTGATCTCGCCGACCA
RT-frr-F	TGTTCAACAAGATCGTGGCC
RT-frr-R	CCTTGGCGACCTTGATGTAC

Cas9 and Cas9n-mediated editing vectors were designed on the basis of pKC1139. Using the construction of pECas9-ΔneoI as an example. Linearized vector was obtained by pKCΔneoI digestion with Hind III. Cas9 expression cassette was obtained by amplification with PE-F1/fd-R using pCas9 as the template, and its expression direction was from the Hind III digestion site to traJ. As for the sgRNA expression cassette, a specific 20 nt of sgRNA was designed by CRISPy-web (https://crispy.secondarymetabolites.org/#/input). Two fragments of sgRNA (neoI) were obtained by amplification with H3-F/ΔneoI-sgR, ΔneoI-sgF/T0-agR, and pCas9, then using SOE-PCR with H3-F/T0-agR to link a whole sgRNA expression cassette, and its expression direction was from Hind III digestion site to homologous arms of pKC-ΔneoI.

Elements optimization was done on the basis of pECas9. Completed traJ was obtained by amplification with traJ-F/traJ-R using pSET152 as the template, while the linearized vector used pECas9 as the template and traJ-GF/traJ-GR as primers for amplification. Next, four mutants were introduced using Cas9-F/N497A-R, N497A-F/R661A-R, R661A-F/Q695A-R, Q695A-F/Q926A-R, Q926A-F/fd-R as primers, using pECas9 as template, and SOE-PCR was used to link five fragments with Cas9-F/fd-R, reconnecting Cas9 with HFCas9. The artificial synthesized fragment including promotor P_tipA and thiR gene was used PtipA-F/PtipA-R for amplification, followed by a replacement with P_ermE*.

The introduction process of the D10A mutation was similar to the introduction flow of the four mutants above. Here, the construction of the Cas9n-mediated editing plasmid showed the construction of pTC9n-ΔnagB as an example. sgRNA (nagB) expression cassette obtained from SOE-PCR of two fragments which used H3-F/ΔnagB-sgR, ΔnagB-sgF/T0-agR and pECas9 for PCR amplification and H3-F/T0-agR was used as primers for SOE-PCR. Homologous arms were obtained by PCR amplification with ΔnagB-UF/ΔnagB-UF and ΔnagB-DF/ΔnagB-DF using the Sf01 genome as the template. Xba I digestion site was induced in pTC9n-ΔnagB and located between the upward homologous arm and the downward homologous arms of nagB. Furthermore, a gene segment integration to replace nagB would use linearized pTC9n-ΔnagB digested by Xba I to construct an editing plasmid.

Construction of S. fradiae strains

Sf01 spores as receptors were washed twice with TES buffer (0.05 M TES, pH 8.0), and then heated at 50°C for 10 min. Treated spores were suspended at TES buffer and preincubated at 30°C for 3 h. ET12567 as a donor was inoculated in LB medium at 37°C for 20 h. Then, ET12567 was collected and washed twice with LB medium. Next, a mixed donor with receptor (20:1), was culturing in SFMM medium at 30°C for 5–7 days. Nalidixic acid and apramycin were added within 18–20 h after conjugation.

Plasmid elimination

Colonies were cultivated at 45°C, followed by an antibiotic sensitivity check. Apramycin-sensitive colonies were verified by colony PCR and Sanger sequencing.

Editing efficiency calculation

The editing efficiency (%) = (the number of successfully edited colonies/total number of colonies containing the edited vector) × 100%. Specifically, after conjugation, all of the conjugants were grown in a medium containing apramycin, which ensured that the editing vector was transferred to the target. Then, these colonies with positive resistance to apramycin were induced to express Cas9n in another medium carrying apramycin and thiostreptone. Finally, colonies will be validated by comparing the consistency of PCR products with the expected size, followed by calculating the number of edited colonies.

Transcription level analysis

Mycelium in the logarithmic phase was lysed by lysozyme, followed by RNA extraction via RNA isolation kit (FORE GENE), and extracting steps referred to in the user manual. Total RNA was treated with UnionScript First-strand cDNA mix kit (ABclonal Technology), and quantitative real-time PCR reactions were performed on the Applied Biosystems Step-One Plus system with Maxima SYBR Green/ROX qPCR Master Mix (ABclonal Technology), and gene sigF was used as the internal control to normalize samples.

Neomycin quantification

Neomycin titer was determined by Agilent HPLC-ELSD, using the Gzflm Titan F5 chromatographic column (2.1 mm × 150 mm, 3 µm), with 0.5%–0.7% trifluoroacetic acid solution as mobile phase, mobile phase flow rate of 0.15 mL/min, column temperature of 30°C, injection volume of 5 µL. The drift tube temperature was 75°C, and the carrier gas flow rate was 1.8 L/min.

DATA AVAILABILITY

The data sets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/aem.01953-24.

Fig. S1. aem.01953-24-s0001.tiff.

Optimization of conjugant parameters for S. fradiae Sf01.

Fig. S2. aem.01953-24-s0002.tiff.

Construction of pKC-ΔneoI, and pECas9 plasmid map.

Fig. S3. aem.01953-24-s0003.tiff.

pEHF-Cas9n plasmid map.

Fig. S4. aem.01953-24-s0004.tiff.

pTHF-Cas9n plasmid map.

Fig. S5. aem.01953-24-s0005.tiff.

Construction of Sf01Δsta.

Supplemental material. aem.01953-24-s0006.doc.

Table S1; legends for Fig. S1 to S5.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.