2.1 Red homologous recombination technologies
Knockouts in certain Gram-negative bacteria (e.g., Escherichia coli (E. coli), Salmonella, and Klebsiella) are often performed using a two-step homologous recombination approach. The development of this method has already been explored over a long period, and the conventional method mainly uses the RecA and RecBCD proteins encoded by the strains themselves as mediators[15]. However, this system has obvious shortcomings (e.g., serious problems with the operation steps, very low recombination rates, and the need for a long homologous arm of the target gene). As a result, it is difficult obtaining the desired recombinant, the knockout efficiency is greatly limited, and the development of this technique is affected.
In 1998, Zhang et al.[16] have suggested that the RedE/RecT system presented in E. coli exhibits the recombination function. In 2000, it was also found that the Exo and Beta proteins of λ- phage exhibit the same function[17], and the E. coli can be specifically modified using a small homologous arm. It was later termed ET Recombination, which can significantly improve the shortcomings of the conventional method, reduce homologous arms required in recombination, and increase the recombination rate. Thus, it has been widely used in the genetic modification of E. coli.
Recombination of double-stranded DNA (dsDNA) requires three bacteriophage λ-Red proteins (i.e., Gam, Exo, and Beta)[18]. Exons encode exonucleases that degrade dsDNA from the 5’ end, while β proteins bind the single-stranded regions derived from exons and facilitate recombination by facilitating pairing with cognate genomic targets[19]. It prevents the degradation of linear dsDNA by Gam E. coli RecBCD protein and SbcCD nuclease[20].
Murers et al.[21] initially constructed a plasmid expressing phage recombinase in 1999. Since then, numerous researchers have modified the plasmid on that basis. The most extensively used one is pKD46 constructed by Datsenko et al.[22], which can express the complete Red homologous recombination system in the auxotrophic plasmid pKD46, make an innovation in the two-step homologous recombination method, and successfully perform gene knockout in E. coli K-12. In its established Red recombinant technology, the primer requires only 36 nucleotides for the homologous arm, and pKD46 serves as the homologous recombinant plasmid with benzyl-resistant low copy temperature-sensitive type. The exo, bet, and gam genes can be integrated under the control of the phage arabinose promoter. The templates for targeting DNA comprise pKD3 and pKD4 plasmids with chloramphenicol and kanamycin resistance genes, respectively, with FRT sites (flip-flop binding sites)[23]. Depending on the removal of the gene of interest, the resistance gene can be removed with a helper plasmid expressing FLP recombinase that recognizes direct repeat FRT (FLP Recognition Target) sites flanking the resistance gene. Red and FLP helper plasmids contain temperature-sensitive replicons that can be eliminated through incubation at 37°C (Fig. 1).
Feng et al.[24] constructed a gene deletion strain of the E. coli efflux protein YddA, AcrB using the Red recombinant system and examined the drug susceptibility of the deletion strain. Their result has suggested that YddA mainly effluxes norfloxacin, and it is an ATP-dependent efflux protein. Ogawa et al.[25] investigate the mechanism of multidrug and toxic compound extrusion (MATE) resistance of Klebsiella using the above-mentioned recombinant plasmid, and their results indicated that the MIC values of antibacterial drugs (e.g., kanamycin) do not change after knocking down the ketM gene of the exocytosis pump, thus confirming that the MATE exocytosis pump is not the direct cause of drug resistance in Klebsiella.
2.2 CRISPR/Cas9 technologies
CRISPR technology was originally discovered in 1987[26], whereas it was not until 2012 that Professor Jennifer Doudna and Professor Emmanuel Charpentier confirmed in vitro experiments that the CRISPR-cas9 system can "localize" DNA breaks[27]. CRISPR knockout screens can be employed in functional genomics studies to detect genomic loci of cellular drug resistance[28,29,30], elucidate how cells induce host immune responses, and determine how certain viruses cause cell death[31]. The above-mentioned technique has been extensively applied in prokaryotes and eukaryotes.
CRISPR/Cas systems currently fall into two categories. Types I, III, and IV pertain to the first category of CRISPR/Cas systems, and types II. V and VI belong to the second category of CRISPR/Cas systems[32]. Type 1 systems share some common features. Precursor crRNAs are processed using specialized Cas endonucleases. and after the maturation, the respective crRNA is assembled into a large multi-Cas protein complex that is capable of secreting nucleic acids complementary to the crRNA and split. In Class 2 systems, a single multinomial large protein serves as the effector. However, in the CRISPR Type II (CRISPR/Cas9) system found in Streptococcus, CRISPR-RNA (tracrRNA) is linked to a small crRNA region complementary and lays a basis for the formation of partial dsRNA. This dsRNA can be attached to Cas9 and target the prototypical spacer sequence, which is subsequently degraded by the nucleic acid endonuclease Cas9[33] (Fig. 2).
Normally, artificial double-strand breaks in E. coli must be repaired by RecA-mediated homologous recombination using homologous sequences as editing templates; however, natural homologous recombination pathways are generally considered to be difficult repairing double-strand breaks caused by CRISPR-Cas9 cutting of chromosomes. The introduction of phage-derived λ-Red recombinase into the CRISPR system improves the probability of obtaining mutant strains, and precise gene modification can be obtained when editing templates are supplemented by the target plasmids [34,35,36]. The processing efficiency is further increased by introducing an exogenous DNA repair system (e.g., when editing templates and λ-Red systems were introduced simultaneously)[37].
Liu et al.[38] targeted and disrupted plasmids encoding kanamycin resistance genes in Escherichia coli using the CRISPR-Cas9 system, thus restoring kanamycin susceptibility in over 99% of bacteria for 32 h, demonstrating that the CRISPR-Cas9 system is effective against resistant plasmids. Wu et al.[39] used the CRISPR/Cas9 genome editing function for modification and removal. They introduced a repair template for homologous recombination, such that the gene editing efficiency was significantly increased. Qiu et al.[40] used the CRISPR-Cas9 system to create a gyrA gene mutation, which altered the susceptibility of the strain to quinolones, thus demonstrating a causal relationship between the E. coli gyrA mutation and its resistance to quinolone antimicrobials.
2.3 Suicide Plasmid Vector System
The suicide plasmid vector system refers to a chromosome modification technique developed in the 1980s, which has a wide host range and is transferable[41]. Conventional suicide plasmids contain the R6K replication initiator that replicates only in recipient bacteria with the pir gene, which encodes a protein required for R6K initiation function and is eliminated once the pir gene is missing[42,43]. Accordingly, it cannot be replicated in general bacteria and should be incorporated into the bacterial chromosome and co-replicated with the bacterial chromosome, which is the characteristic condition and special feature of suicide plasmids.
Suicide plasmids have two options when transferred into the host cell since they do not contain replicon structures that can replicate in the host cell. The first option is automatically removed without replication, and the other is to be incorporated into the chromosome and replicates as it replicates. Using the above function, the homologous arm is cloned into the suicide plasmid vector through enzymatic cleavage and DNA ligation. Moreover, after splicing transfer, the homologous fragment on the suicide plasmid will integrate with the homologous fragment on the bacterial genome. After the introduction of suicide plasmid into bacteria, the bacteria can rely on their recombination function (i.e., RecA recombination system), and genes can be precisely deleted[15].
Gene deletion using suicide plasmids is based on a two-step homologous recombination process involving a counter-selection system that substitutes the homology arms of the suicide plasmid with the bacterial target gene[44]. Thus, the first exchange integration refers to the replacement of the target gene with the homologous arm in the suicide plasmid, which integrates the suicide plasmid into the genome in its entirety, and the second exchange integration is the removal of the plasmid backbone from the bacterial genome to obtain a trace-free gene knockout. After successful replacement, the suicide plasmid is removed by rejection for its inability to replicate in an environment lacking relevant replication conditions, thus causing a trace-free knockout deletion strain[45] (Fig. 3).
Oh et al.[46] developed a method for the knockout of drug resistance genes in Acinetobacter baumannii (A. baumannii) using suicide plasmids by amplifying fragments with the upstream and downstream fragments of the target gene and antibiotic resistance genes by overlapping PCR. Subsequently, the recombinant suicide plasmids are ligated into A. baumannii for homologous recombination via flat ends to obtain mutants. TSAI Y K et al.[47] deleted and recovered copies of the ompK35 and/or ompK36 genes directly from the chromosome of Klebsiella using suicide plasmid allele exchange method. Compared with the parental strain, only the strain lacking OmpK36 is resistant to cefazolin, cephalothin, and cefoxitin. Deletion of OmpK35 further increased MICs, suggesting that strains with double deletions are significantly resistant to the above-described drugs.