Complex bacterial engineering made easy with novel CRISPR-based tools—all at once or one-gene-at-a-time

CRISPR-Cas technologies are a powerful tool for microbial genome engineering, yet in non-traditional bacteria their use is still limited in efficiency and throughput. We have repurposed a CRISPR base editor for complex engineering, both for fundamental and applied aspects of bacterial research.
Complex bacterial engineering made easy with novel CRISPR-based tools—all at once or one-gene-at-a-time

The ever-expanding CRISPR-Cas toolbox is arguably the most widely used gene and genome-editing tool across biological systems (both prokaryotic and eukaryotic), and these approaches have found broad applications in both basic research and biotechnological applications, including biomedicine and sustainable chemical production1,2. Among other purposes, intense ongoing research aims at engineering CRISPR-Cas9 proteins with enhanced efficiency and accuracy for precise genome editing or to establish specific base modifications. These exciting developments have provided the scientific community with tools and protocols for microbial engineering, yet some limitations still stand on the way towards wide application of the cognate technologies. For instance, metabolic engineering applications based on non-traditional bacterial hosts, e.g., Pseudomonas species3, has not been developed to the extent currently seen in other microbes—as the available toolset had been mostly tailored for model organisms. In an attempt to mitigate these practical constraints, we have developed a widely-applicable genome engineering toolset for Gram-negative bacteria (with a focus on their use to manipulate Pseudomonas species)4. The challenge included the modification of a CRISPR base editor such that it can enable single-nucleotide resolution manipulations (i.e., C·G→T·A) with >90% efficiency in a broad-host-range format.


Cas6-mediated processing of the guide RNAs (gRNAs, which lead and support the intended modifications in the genome) was also incorporated in our pipeline by creating a streamlined protocol for plasmid assembly and deployment (Fig. 1). The resulting series of plasmids (termed pMBEC, and endowed with four different antibiotic resistances to facilitate their utilization in different bacteria) supported multiplex base editing in Escherichia coli, Pputida and Paeruginosa with >85% efficiency. Furthermore, to facilitate cloning procedures and to simplify the selection of positive clones containing the intended plasmid constructs, a fluorescent protein marker gene (monomeric super-folder GFP, msfGFP) was added to the pMBEC backbones in our study—which facilitated rapid counter-selection of template (i.e., empty) vectors. Thus, the base editing-plasmids contain a constitutively-expressed msfGFP module flanked by BsaI recognition sites. This marker module is replaced by either one spacer sequence or multiple gRNAs through Golden Gate assembly and, upon transformation of recipient bacteria (E. coli, used as a cloning host for these purposes), empty plasmids (i.e. those that would lead to fluorescent colonies) can be easily discarded.


How is the multiplexing of genome targets achieved? In a first step, the desired gRNAs are generated by PCR. In this procedure, the Cas9 handle is amplified along with a Cas6 recognition sequence from a template vector (i.e., with a conserved sequence in the oligonucleotide), while the specific spacer sequence (i.e., the sequence that defines the gene target itself) and Golden Gate-flanking motifs are introduced in the oligonucleotide sequences. By following this design, the resulting gRNAs will harbor unique BsaI sites that can be used to compose multiplex arrays by Golden Gate cloning5. Moreover, a script is made available along with our article such that the user can obtain the oligonucleotide sequences to be ordered for assembly of the pMBEC plasmid from the desired spacer sequence. SacB-mediated curing of these pMBEC plasmids facilitates multiple editing cycles upon confirmation and analysis of the resulting bacterial phenotypes.

Fig. 1. The upgraded cytidine base editor (CBE) described in our work. A nicking Cas9 (nCas9) is guided by a single guide RNA (sgRNA) to the cognate protospacer sequence. Cytidine deaminase, fused to nCas9, converts cytosine to uracil within an editing window; uridine is subsequently transformed into thymidine during DNA replication. This strategy can be multiplexed as needed either for the construction or deconstruction of bacterial phenotypes. Further details and abbreviations are detailed in the article by Volke et al.4.

With this base-editing toolset at hand, we aimed at constructing and deconstructing complex phenotypes in the soil bacterium P. putida. As indicated above, this microorganism is increasingly being used as a microbial host for the production of added-value molecules owing to its very rich and versatile metabolism6. Single-step engineering of an aromatic-compound production phenotype and multi-step deconstruction of the intricate redox metabolism illustrated the versatility of multiplex base editing afforded by our toolbox. In the first example, we bestowed production of protocatechuic acid (PCA) onto P. putida KT2440 by boosting precursor availability and suppressing product degradation. De novo biosynthesis of PCA from sugar feedstock would normally suppose a labor-intense microbial engineering program, involving gene deletion, overexpression and pathway fine-tuning to enable production of the aromatic compound. We based our design on native activities of P. putida KT2440, as this strain is endowed with all the genes encoding the enzymes needed for PCA biosynthesis through the core shikimate route. By a combined knock-in and knock-out strategy that involved the manipulation of 9 separate targets, we obtained an engineered P. putida strain that produced up to 15 mM PCA from glucose as the main substrate in 72-h shaken-flask fermentations. Prompted by these results, whereby a complex production phenotype can be engineered in a single-step, multiplex manipulation of the host, we decided to decompose the redox metabolism of P. putida towards creating a NADPH-deficient strain. In this case, we focused on a modular, iterative gene-editing strategy, such that key metabolic nodes were sequentially blocked or isolated to limit NADPH formation. In this way, three different editing modules were designed to target NADPH production (i) in the upper metabolism, (ii) in the lower metabolism, and (iii) through pyridine nucleotide transhydrogenation. These operations led to a series of modified strains where NADPH biosynthesis is impaired, including a NADPH ‘auxotrophic’ P. putida that could not grow on pyruvate as the only substrate—yet the growth deficient phenotype could be rescued by an external, orthogonal NADPH-producing activity (i.e., an engineered formate dehydrogenase) or by awakening silent, endogenous NADP+-dependent dehydrogenases.


Taken together, the results afforded by this approach, as presented in our article, illustrate how the CRISPR base-editing tool overcomes the typical limitations of previous technologies adopted for gene and genome engineering of Pseudomonas, and empowers engineering programs in Gram-negative bacteria that were out of reach thus far. We envision that this novel CRISPR system and related protocols for microbial engineering will not only benefit the broad field of basic and applied microbiology, but will also provide a deeper understanding towards the mechanisms of CRISPR-mediated gene editing in bacteria.


  1. Schultenkämper, K., Brito, L.F. & Wendisch, V.F. Impact of CRISPR interference on strain development in biotechnology. Biotechnol. Appl. Biochem. 67, 7-21 (2020).
  2. Zhao, D. et al. CRISPR-based metabolic pathway engineering. Metab. Eng. 63, 148-159 (2021).
  3. Volke, D.C., Calero, P. & Nikel, P.I. Pseudomonas putida. Trends Microbiol. 28, 512-513 (2020).
  4. Volke, D.C., Martino, R.A., Kozaeva, E., Smania, A.M. & Nikel, P.I. Modular (de)construction of complex bacterial phenotypes by CRISPR/nCas9-assisted, multiplex cytidine base-editing. Nat. Commun. 13, 3026 (2022).
  5. Pryor, J.M. et al. Enabling one-pot Golden Gate assemblies of unprecedented complexity using data-optimized assembly design. PLoS One 15, e0238592 (2020).
  6. Weimer, A., Kohlstedt, M., Volke, D.C., Nikel, P.I. & Wittmann, C. Industrial biotechnology of Pseudomonas putida: Advances and prospects. Appl. Microbiol. Biotechnol. 104, 7745–7766 (2020).