Combinatorial CRISPR Interference Library for Enhancing 2,3-BDO Production and Elucidating Key Genes in Cyanobacteria

Abstract

Cyanobacteria can convert CO₂ to chemicals such as 2,3-butanediol (2,3-BDO), rendering them promising for renewable production and carbon neutralization, but their applications are limited by low titers. To enhance cyanobacterial 2,3-BDO production, we developed a combinatorial CRISPR interference (CRISPRi) library strategy. We integrated the 2,3-BDO pathway genes and a CRISPRi library into the cyanobacterium PCC7942 using the orthogonal CRISPR system to overexpress pathway genes and attenuate genes that inhibit 2,3-BDO formation. The combinatorial CRISPRi library strategy allowed us to inhibit fbp, pdh, ppc, and sps (which catalyzes the synthesis of fructose-6-phosphate, acetyl-coenzyme A, oxaloacetate, and sucrose, respectively) at different levels, thereby allowing for rapid screening of a strain that enhances 2,3-BDO production by almost 2-fold to 1583.8 mg/L. Coupled with a statistical model, we elucidated that differentially inhibiting all the four genes enhances 2,3-BDO synthesis to varying degrees. fbp and pdh suppression exerted more profound effects on 2,3-BDO production than ppc and sps suppression, and these four genes can be repressed simultaneously without mutual interference. The CRISPRi library approach paves a new avenue to combinatorial metabolic engineering of cyanobacteria.

Keywords: CRISPR; CRISPRi library; PCC7942; combinatorial metabolic engineering; cyanobacteria.

FIGURE 1

Validation of orthogonal CRISPR systems for gene editing in PCC7942. (A) Plasmids encoding SpCas9, SaCas9, and St1Cas9 as well as plasmids encoding their corresponding sgRNA. Each sgRNA comprised identical spacer targeting NSI site but different scaffold backbone. sgRNA was driven by J23119 promoter. (B) Survival rate relative to the control. Cas9-expressing plasmid was transformed alone at different doses (1,000–4,000 ng). Nontransformed cells (0 ng) served as the control. Survival rate was assessed by (CFU/CFUcontrol) × 100%. (C) Death rate induced by different cognate Cas9/sgRNA pairs. Cas9-expressing plasmid was co-transformed with the corresponding sgRNA-expressing plasmid at different doses, and the death rate was assessed using nontransformed cells (0 ng) as the control. The death rate was calculated by (1-CFU/CFUcontrol) × 100%. (D) Orthogonality test for three Cas9 variants. The Cas9-and sgRNA-expressing plasmids were co-transformed into PCC7942 and streaked to plates. The plates were observed at day 7–9. The data represent the mean ± SD of three independent culture experiments and were analyzed by one-way ANOVA. p < 0.05 was considered statistically significant.

FIGURE 2

Generation of 2,3-BDO-producing PCC7942 using SaCas9. (A) Pathway leading to 2,3-BDO production. CBB, Calvin–Benson–Bassham cycle; 3 PG, 3-phosphoglycerate; PEP, phosphoenolpyruvate; PYR, pyruvate. (B) Plasmids encoding 2,3-BDO pathway genes driven by repressor-less LacO promoter (pNSII-23BDO) and sgRNA plasmids corresponding to SpCas9 (psgRNA-NSII-Sp) and SaCas9 (psgRNA-NSII-Sa). sgRNA spacer was designed to target the NSII site. NRII L and NSII R are the homology arms for the NSII site. Km^R, kanamycin resistance gene. P1 to P4 are primers used for PCR check. (C) CFU after co-transformation of 2000 ng pNSII-23BDO, 500 ng of pSpCas9/pSgRNA-NSII-Sp (Sp group), or pSaCas9/pSgRNA-NSII-Sa (Sa group) and Km selection. The control was transformed with only 2000 ng pNSII-23BDO. (D) PCR analysis of five colonies from the Sa group using two primer pairs (P1/P2 and P3/P4) targeting the left and right junctions at the integration site. The expected PCR amplicon size is 2.7 kb. (E) Illustration of the 7942-BDO strain, which was picked after Km selection and re-streaking of colony 4 as shown in (D). (F) 2,3-BDO titer after shake flask culture of 7942-BDO strain for 11 days. The quantitative data represent the mean ± SD of at least three independent culture experiments and were analyzed using Student’s t-test.

FIGURE 3

Development of the CRISPRi library for enhanced 2,3-BDO production. (A) Pathways leading to the synthesis of desired 2,3-BDO and undesired metabolites. AcCoA, acetyl-coenzyme A; OAA, oxaloacetate; F6P, fructose-6-phosphate; FBP, fructose-1,6-bisphosphate; fbp, fructose bisphosphatase; pdh, pyruvate dehydrogenase; ppc, phosphoenolpyruvate carboxylase; and sps, sucrose phosphate synthase. (B) Plasmid encoding SpdCas9 and the engineered 7942-BDOdCas strain. 7942-BDOdCas strain was engineered using SaCas9 to integrate SpdCas9 into NSI and 2,3-BDO pathway genes into the NSII site. (C) Plasmids encoding the sgRNA and unique barcode. The sgRNA spacer was designed to target the protospacer downstream the transcription start site (TSS) of fbp, pdh, ppc, or sps (sg1 or sg2) with the highest targeting scores. We also designed a plasmid encoding the scrambled sgRNA comprising a nontargeting spacer (sg0). (D) Illustration of the combinatorial sgRNA library with 81 combinations. Each sgRNA was labeled with a barcode (A: fbp; B: pdh; C: ppc; and D: sps; Supplementary Table S1) and was expressed from the same J23119 promoter. Each gene has three possible suppression levels (0: none; 1, weak; and 2: strong); thus, there are 3⁴ (=81) possible combinations. The combinatorial sgRNA library contained the homology arms for NSIII sites (NSIII L and NSIII R) The sequences were confirmed by PCR analyses of the unique barcode using primers P1/P2. (E) 7942-BDOdCas-library which was constructed by integrating the sgRNA library into the NSIII site of 7942-BDOdCas using SaCas9. (F) 2,3-BDO titer from 7942-BDOdCas-library. After antibiotic selection, 210 colonies were picked from 7942-BDOdCas-library and cultured in shaker flasks for 11 days for extracellular 2,3-BDO analyses using GC-BID. The quantitative data represent the mean ± SD of at least three independent culture experiments and were statistically analyzed by one-way ANOVA.

FIGURE 4

Effects of combinatorial gene suppression on 2,3-BDO production. (A,B) Correlation between the extracellular 2,3-BDO titer and the 81 sgRNA combinations (also see Supplementary Table S2). All the 210 clones were sequenced and designated based on the sgRNA combination. (B) Illustration of the targeting site (sg0, sg1, or sg2) at each gene (fbp, pdh, ppc, and sps) and the designated sgRNA combination. For instance, BDO2222 indicates that the sgRNA targeted four genes at sg2. BDO2202 indicates that the sgRNA targeted fbp, pdh, and sps at sg2 and ppc with sg0. (C–F) Effects of suppression of fbp (C), pdh (D), ppc (E), or sps (F) on the 2,3-BDO titer. The gene expression levels were measured by qRT-PCR, and suppression levels were calculated for each clone. The data represent the mean ± SD of at least three independent experiments and were statistically analyzed by one-way ANOVA. p < 0.05 was considered statistically significant.

FIGURE 5

Effects of gene suppression on intracellular metabolite levels. (A) Effects of fbp suppression on F6P titer. (B) Effects of pdh suppression on the AcCoA titer. (C) Effects of ppc suppression on the OAA titer. (D) Effects of sps suppression on sucrose. The data represent the mean ± SD of at least three independent experiments and were statistically analyzed by one-way ANOVA. p < 0.05 was considered statistically significant. The intracellular F6P, AcCoA, OAA, and sucrose for all 210 clones were analyzed by HPLC-MS. N.D., not detectable.