A thermostable type I-B CRISPR-Cas system for orthogonal and multiplexed genetic engineering

Abstract

Thermophilic cell factories have remarkably broad potential for industrial applications, but are limited by a lack of genetic manipulation tools and recalcitrance to transformation. Here, we identify a thermophilic type I-B CRISPR-Cas system from Parageobacillus thermoglucosidasius and find it displays highly efficient transcriptional repression or DNA cleavage activity that can be switched by adjusting crRNA length to less than or greater than 26 bp, respectively, without ablating Cas3 nuclease. We then develop an orthogonal tool for genome editing and transcriptional repression using this type I-B system in both thermophile and mesophile hosts. Empowered by this tool, we design a strategy to screen the genome-scale targets involved in transformation efficiency and established dynamically controlled supercompetent P. thermoglucosidasius cells with high efficiency ( ~ 10⁸ CFU/μg DNA) by temporal multiplexed repression. We also demonstrate the construction of thermophilic riboflavin cell factory with hitherto highest titers in high temperature fermentation by genome-scale identification and combinatorial manipulation of multiple targets. This work enables diverse high-efficiency genetic manipulation in P. thermoglucosidasius and facilitates the engineering of thermophilic cell factories.

Fig. 1. Identification of the thermostable type I-B CRISPR-Cas from P. thermoglucosidasius.

a Genetic architecture of the Type I-B and III-B CRISPR-Cas operon. The diamond: repeat sequence. The square: spacer sequence. b Protospacers predicted by web CRISPRTarget tool. The yellow sequences: spacers from genome. The black sequences: protospacer sequence. The potential PAM is indicated by a blue background. c Sequence alignment of seven repeats from Type I-B and Type III-B CRISPR-Cas system. Non-conservative sites are highlighted in red. d Plasmid interference activity of the type I-B CRISPR-Cas system in strain NCIMB 11955. The transformation efficiency was indicated by the rate of recovered clones. A low-transformation efficiency indicates a high plasmid interference activity. e Plasmid interference activities of the type I-B CRISPR-Cas system at different temperatures in strain NCIMB 11955. f Plasmid interference activities of the type I-B CRISPR-Cas system at 37 °C in E. coli. Data represented three biological repeats and displayed as mean ± SD. Source data are provided as a Source Date file.

Fig. 2. Transcription repression by using shortened crRNA.

a Principle for evaluating transcription repression using sfGFP expression. b Design of truncated noncanonical spacers used for generating crRNA. c Transformation efficiency of these plasmids expressing the full-length and truncated spacers. Transformation efficiency is determined by the colony recovery rate relative to control plasmid without any spacer targeting the genome. d Fluorescence intensity repressed by these crRNAs. The exact p values for T-11, T-15, T-20, and T-25 were 0.00072, 2.98e−5, 1.06e−7, and 8.76e−8, respectively. e Repression of sfgfp gene characterized at transcription level using RT-qPCR. The exact p values for T-11, T-15, T-20, and T-25 were 0.00018, 1.88e−5, 5.66e−6, and 5.62e−6, respectively. f Transformation efficiency of the plasmids expressing truncated spacers from 26 to 29 nt. g Fluorescence intensity repressed by 26-nt crRNA. P value = 7.76e−8. h Effect of different length crRNA on the persistence of the cognate target plasmid. The test plasmid was purified from equal volumes of cultures of strain T-27 and T-26 grown for 4 h with or without crRNA inducers and resolved by agarose gel electrophoresis. Data are the mean of three biological repeats and are expressed as mean ± SD. Statistical significance is calculated based on two-tailed Student’s t test (***P < 0.001). Source data are provided as a Source Date file.

Fig. 3. Characterization of this type I-B CRISPR-Cas system.

a Logo showing the putative PAM motif based on three identified protospacer. The sequences were aligned and analyzed with WebLogo (http://weblogo.berkeley.edu/logo.cgi). b Effect of mutants at positions −5 and −4 of PAM on interference activity. Interference activity is determined by the colony recovery rate of targeting plasmid transformation. c Effect of mutants of the 3-nt PAM on interference activity. d Schematic illustration of the developed sfGFP report system for evaluating the specific PAM. The detail of P_xylA* was shown in Supplementary Fig. S6A, B. e Comparison the repression of the 3-nt and 5-nt PAM sequence on sfGFP report system by measuring fluorescence intensity. f Evaluating the specificity of transcription repression on a genome-wide scale based on deep sequencing (RNA-seq). g Evaluation of prime adaptation by truncated crRNA. CRISPR expansion at the leader end of CRISPR1, CRISPR2 or CRISPR3 were assayed by PCR. DNA sample from no crRNA expression (control) or truncated crRNA expression (T-26, T-25, T-20, or T-15) strain was used as PCR templates. Lane M, dsDNA size marker. h Evaluation of prime adaptation by truncated crRNA. Data are the mean of three biological repeats and are expressed as mean ± SD. Source data are provided as a Source Date file.

Fig. 4. Orthogonal tool development in thermophilic P. thermoglucosidasius and mesophilic E. coli.

a Schematic of the editing plasmid. This plasmid equipped Esp3I cleavage sites for the insertion of crRNA. ColE1 ori: replication origin used in E. coli, amp^R: ampicillin resistance gene, R: repeat, lacZ used for blue-white selection of the spacer insertion, repB ori: replication origin used in Geobacillus or Parageobacillus genus, kan^R: kanamycin resistance gene. b Transformation efficiency of targeting amylase gene for allelic recombination based on deletion. P_RplsWT and P_xylA indicate the constitutive and inducible promoter driving crRNA, respectively. c The workflow of deletion in P. thermoglucosidasius. d Confirmation of amylase gene deletion by PCR. e Schematic of the transferrable plasmid pCasIB harboring the type I-B Cas effectors. The plasmid pCasIB and editing/repressing plasmid together were used to editing or repressing target gene in E. coli. f Deleting pyrF gene in E. coli using the transferrable tool. g Growth rate influenced by the transcription repression of pyrF gene. Data are the mean of three biological repeats and are expressed as mean ± SD. Source data are provided as a Source Date file.

Fig. 5. Workflow for systematic improvement of transformation efficiency.

a Workflow for improving transformation efficiency. b Transformation efficiency of strain GtCom1–GtCom6. Control: WT strain. GtCom1 (ΔBCV53_09985), GtCom2 (ΔBCV53_05690), GtCom3 (ΔBCV53_12685), GtCom4 (ΔBCV53_09960), GtCom5 (ΔBCV53_8635), GtCom6 (ΔBCV53_09985, ΔBCV53_05690, ΔBCV53_12685, and ΔBCV53_09960), n = 5 biologically independent samples. c Schematic for screening the new targets contributing to transcription efficiency. d Abundance changes of tgRNAs assayed by deep sequencing. e Testing the transformation efficiency of strain GtCom6 by re-transform selected positive tgRNAs, n = 5 biologically independent samples. The exact P values for BCV53_10935, BCV53_12375, BCV53_09555, BCV53_01915, BCV53_09535, and BCV53_08150 were 0.25, 8.69e−8, 6.05e−11, 2.62e−6, 0.0076, and 5.09e−8, respectively. f Schematic of xylose-induced CRISPR array containing multiple tgRNAs. Strain status switch between competent cell and production host by inducer. The symbol ‘+’ indicates the medium with inducer, and ‘–’ indicates without. g Transformation efficiency of strain GtCom7 with or without inducer, n = 5 biologically independent samples. h Growth rate of strain GtCom7 with or without inducer. Data are the mean of three biological repeats and are expressed as mean ± SD. Statistical significance is calculated based on two-tailed Student’s t test (**P < 0.01; ***P < 0.001). Source data are provided as a Source Date file.

Fig. 6. Workflow for systematic improvement of riboflavin production.

a Workflow for improving riboflavin production. b Metabolic pathways for the biosynthesis of riboflavin. Abbreviations in Supplementary Note 1. c Schematic for colorimetric screening the new targets contributing to riboflavin production. d Testing the riboflavin production of strain Gt-08 by re-transform selected positive tgRNAs. The exact p values for BCV53_04550, BCV53_12070, BCV53_16240, BCV53_12180, and BCV53_00110 were 2.83e−5, 6.23e−6, 1.62e−5, 1.39e−5, and 3.86e−5, respectively. e Schematic of CRISPR array containing multiple tgRNAs expressed in the plasmid. f Fed-batch fermentation profile of the final riboflavin strain Gt-09. g Microscopic examination for identifying no bacterial contaminants. Data are the mean of three biological repeats and are expressed as mean ± SD. Statistical significance was calculated based on two-tailed Student’s t test (***P < 0.001). Source data are provided as a Source Date file.

Fig. 7

A versatile genetic engineering tool contributing to building thermophilic factories.