CRISPRi screen for enhancing heterologous α-amylase yield in Bacillus subtilis

Abstract

Yield improvements in cell factories can potentially be obtained by fine-tuning the regulatory mechanisms for gene candidates. In pursuit of such candidates, we performed RNA-sequencing of two α-amylase producing Bacillus strains and predict hundreds of putative novel non-coding transcribed regions. Surprisingly, we found among hundreds of non-coding and structured RNA candidates that non-coding genomic regions are proportionally undergoing the highest changes in expression during fermentation. Since these classes of RNA are also understudied, we targeted the corresponding genomic regions with CRIPSRi knockdown to test for any potential impact on the yield. From differentially expression analysis, we selected 53 non-coding candidates. Although CRISPRi knockdowns target both the sense and the antisense strand, the CRISPRi experiment cannot link causes for yield changes to the sense or antisense disruption. Nevertheless, we observed on several instances with strong changes in enzyme yield. The knockdown targeting the genomic region for a putative antisense RNA of the 3' UTR of the skfA-skfH operon led to a 21% increase in yield. In contrast, the knockdown targeting the genomic regions of putative antisense RNAs of the cytochrome c oxidase subunit 1 (ctaD), the sigma factor sigH, and the uncharacterized gene yhfT decreased yields by 31 to 43%.

Keywords: CRISPRi; Transcriptomics; fermentation; screening; α-amylase.

Fig. 1.

RNA-seq of fermentation samples of Bacillus subtilis strains. (A) Fermentation growth curve of a codon optimized strain prsA + je1zyn (orange) and control strain prsA + je1 (gray) (n = 2, error bars depict standard deviation). Time-points for sampling for RNA-seq are indicated with vertical dashed lines. (B) Yield of the α-amylase enzyme as in (A). (C) Principal component analysis of RNA-seq samples based on DESeq2 r-log normalized expression of the 500 most variable expressed genes, excluding rRNA, tmRNA, and SRP.

Fig. 2.

Novel ncRNA prediction. (A) Predicted transcribed regions are regions for which the RNA-seq coverage (black curve) is above a height cut-off (red line). The prediction method tolerated if the coverage for a few base pairs is below the cut-off; this tolerance was larger if these bases were within a reference annotation (see S2 for details). We determined these parameters by benchmarking the transcribed region predictions compared to known reference annotations, particularly their differences in the 5′ and 3′ positions. We measured the predicted transcribed region ends assigned upstream (black arrows) or downstream (gray dashed arrow) of the reference annotation. These values were used in the benchmark (section S2Supplementary file 1). (B) Novel ncRNAs were identified from parts of transcribed regions (gray) that intersect with strand-specific gaps in the reference annotations (orange). The novel ncRNA predictions are presumably non-coding and depending on the position of these fragments relative to the annotation, they could be classified as novel ncRNAs (green), potentially with antisense overlap (red), or as a UTR (purple). + and—denote strand identity. (C) Number of predicted ncRNAs classification according to length.

Fig. 3.

Heatmap and pathway enrichment. (A) Heatmap of the log2 expression profile as implied by the kmeans cluster centroids sorted by a hierarchical complete linkage clustering. The columns contain the expression values at each of the four conditions (days and strain). (B) Enrichment of KEGG pathway by kmeans clusters. The plot shows a pathway and the enriching cluster (y-axis) over the ratio of enrichment (x-axis) with point sizes indicating the number of genes annotated in a pathway and color the p-value of enrichment.

Fig. 4.

Maximum logFCs. For annotations with differentially expression, the cumulative density of the maximal observed logFC at each statistically significant pairwise comparison (see Differential Gene Expression and Pathway Analysis) is shown (x-axis). Coding genes are shown separately from coding genes that are located in an operon located antisense to a novel predicted asRNA.

Fig. 5.

CRISPR-dCas9 functionality. (A) Quantitative RT-PCR of tmRNA in strains expressing sgRNA:: gfp or sgRNA:: ssrA normalized to sgRNA:: gfp (n = 3, error bars depict standard error of the mean). RNA was extracted from exponentially growing flask cultures (see methods sections RNA Purification and Quantitative RT-PCR). (B) JE1 relative activity in deep-well plate fermentation strains expressing sgRNA:: gfp or sgRNA:: ssrA normalized to sgRNA:: gfp (n = 3, error bars depict standard deviation). (C) GFP fluorescence in strains expressing GFP, dCas9, empty plasmid (pE194), or sgRNA:: gfp.

Fig. 6.

Non-coding RNA interference for impact on yield. (A) JE1 relative activity in deep-well plate fermentation of strains expressing sgRNA against ncRNA candidates. Each candidate was tested with two sgRNAs. Samples were normalized to JE1 activity of a strain expressing sgRNA:: GFP (n = 3, error bars depict standard deviation). Based on the observation of a consistent relative JE1 activity combined with the overall CRISPRi impact (see results), the median activity (solid black line) of all guides (including second panel B) indicates a baseline for retrieving the changes in yield. The dashed lines indicate the upper and lower interquartile range. (B) as in (A) but for 31 asRNA candidates. Data for folB sgRNA1 and yhaX sgRNA2 is not shown since these strains did not grow in liquid cultures. The average differences in yield relative to the median for the four targets ctaD, skfH, yhfT, and sigH are highlighted. (C) The skfA-H operon showing amplicons in qRT-PCR (green boxes), sgRNAs (red boxes), and putative asRNA (yellow box). (D) Quantitative RT-PCR of skfH coding region in strains expressing sgRNA:: gfp, sgRNA:: skfH sgRNA1 or sgRNA:: skfH sgRNA2 normalized to sgRNA:: gfp (n = 3, error bars depict standard error of the mean). RNA was extracted from exponentially growing flask cultures (see methods sections RNA Purification and Quantitative RT-PCR). (D) As in (C) but qRT-PCR of skfE coding region.