Lactic acid bacteria (LAB) are a group of Gram-positive, non-sporulating, aerotolerant bacteria characterized by their ability to produce lactic acid as the principal final product, including seven genera: Lactococcus, Lactobacillus, Enterococcus, Pediococcus, Streptococcus, Leuconostoc and Oenococcus (Stefanovic et al., 2017). LABs have great potential for application in biological preservation and naturally dominate the microflora of many foods because most LABs are generally considered safe (Ghanbari et al., 2013).

Aquatic products are popular for their delicious taste and high nutritional value as they are rich in protein, fat, vitamins, and minerals. However, the high contents of various nutrients and moisture in aquatic products will lead to rapid microbial growth, metabolism, and biochemical reactions in postmortem aquatic products (Olatunde and Benjakul 2018). These fast-growing bacteria are called specific spoilage organisms (SSOs), which will become dominant as storage time increases and ultimately lead to the spoilage of aquatic products, such as Pseudomonas spp., Aeromonas spp., and Shewanella spp. (Zhang et al., 2019).

LABs can produce a number of bacteriocins, which are proteins or polypeptides that can be used to inhibit the growth of SSOs in aquatic products (Bali et al., 2016). The growth of LABs can also inhibit many SSOs by competing for nutrients or producing one or more metabolites with antimicrobial activity (Ghanbari et al., 2013). The most commonly used LABs in aquatic products preservation are Lactobacillus spp., followed by Lactococcus spp. (Dong et al., 2022), which are compatible with their environments, such as modified atmosphere packaging, low temperatures, and pH, and so on. The improvement of genome editing tools is particularly important for the development of new strains of LABs with excellent biopreservation properties for aquatic products (Figure 1).

Several gene-editing tools have been developed to achieve stable integration of sequences in Lactobacillus spp. and Lactococcus spp., for example, the use of insertion sequence (IS) elements (Walker and Klaenhammer 1994), Cre-lox-based systems (Lambert et al., 2007) and methods using selectable markers. However, the employment of IS elements is restricted by the presence and distribution of the IS sequences in the genome. The Cre-lox-based system will leave remnant sequences (scars) at the targeted site. And the selection markers should be eliminated in the following step for subsequent modifications and many methods using selectable markers can only be used in strains that have a specific gene mutated (Dong et al., 2022).

The genome engineering tools based on the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein (Cas) systems have been widely used in microorganisms. The CRISPR/Cas systems are classified into two classes (class I and class II) and six types (type I ∼ VI), and types I, III, and IV belonging to class I and types II, V, and VI belonging to class II (Mohanraju et al., 2016). Recently, many methods based on recombineering and CRISPR/Cas system had been successfully established in Lactobacillus spp. and Lactococcus spp. Here, we will summarize the development and application of CRISPR/Cas-based genome editing tools in Lactobacillus spp. and Lactococcus spp. according to different types of genome editing and transcriptional regulation scenarios. And we will discuss the different shortcomings of current CRISPR/Cas-based genome editing technologies and suggest possible directions for the future development of LABs.

Type I CRISPR/Cas systems with Cas3 protein as hallmark can be divided into seven subtypes: I-A to I-F and I-U. Hidalgo-Cantabrana et al identified widespread existence of CRISPR-Cas systems in Lactobacillus crispatus, including type I-B, complete type I-E, and type II-A (Hidalgo-Cantabrana et al., 2019). They characterized the native type I-E CRISPR-Cas system and developed an efficient chromosomal targeting and genome editing tool based on this system with a 5′-AAA-3′ protospacer adjacent motif (PAM). In the genetic target of the p-gtf gene, which encodes the exopolysaccharide priming-glycosyl transferase, the 643-bp gene deletion efficiency is 100%, the stop codon insertion efficiency is 36%, and the single nucleotide substitution efficiency is 19%. However, in the prophage DNA packaging Nu1, the 308-bp deletion efficiency is only 20%, and the insertion efficiency of a 730-bp green fluorescent protein gene in the downstream of enolase is 23%.

Types II CRISPR/Cas system relies on a single effector nuclease Cas9. Cas9, which contains HNH and RuvC nuclease domains, was used as a scissor to cleave target DNA to generate double-strand breaks (Jiang and Doudna 2017). CRISPR/Cas9 is a fast-growing and powerful tool for genome engineering in various organisms, and it is the most widely used in LAB (Hidalgo-Cantabrana et al., 2019). The natural CRISPR/Cas9 system consists primarily of two components: Cas9 effector nucleases and guide RNA (gRNA), which is composed of crRNA and tracrRNA (Jinek et al., 2012; Panda and Ray 2022). The 20–24 NT bases at the 5′-end of sgRNA were used to recognize the target sequences by base pairing with target DNA with the presence of an appropriate PAM sequence (the commonly used SpyCas9 from Streptococcus pyogenes is 5′-NGG-3′) at the 3′-end. The CRISPR/Cas9 genome editing process is as follows: the gRNA binds to Cas9 to recruit Cas9 to the target DNA sites and activates Cas9 nuclease activity to cleave the double-strand DNA sequence to generate DSBs (Figure 2) (Anders et al., 2014). With several recombination techniques, including plasmid-assisted recombination, ssDNA and dsDNA-assisted recombination, the DSBs can be repaired by homology-directed repair (HDR) for precise gene editing (Figure 3). Besides, the DSBs can be repaired by non-homologous end joining (NHEJ) to produce inaccurate editing (Hsu et al., 2014). The DSBs in Lactobacillus are often repaired by HDR rather than NHEJ (Song et al., 2020).

The type III CRISPR/Cas systems can be classified into four subtypes A-D. The Type III-A system is one of the most widely distributed CRISPR/Cas systems across prokaryotic phyla and cleaves DNA and RNA molecules.

The III-A CRISPR/Cas module from L. lactis, which include a Cas6 protein, a CRISPR locus for crRNA production, and Csm effector complex proteins, was heterologously expressed in E. coli (Ichikawa et al., 2017). The expressed module specifically eliminated an invasive plasmid recognized by the crRNA. When appropriate crRNA sequences were added to the module, the module could be programmed to recognize plasmids with novel target sequences. This system lays the foundation for developing it as a gene editing tool in E. coli or other novel organisms.

Replacing Cas9 in the CRISPR/Cas9 system with dCas9 enabled CRISPR interference-mediated silencing of genes (Figure 4A). Berlec et al used CRISPRi to mediate the upp gene silencing in L. lactis, which decreased upp mRNA transcription and prevented the toxicity of 5-fuorouracil (Berlec et al., 2018). Xiong et al constructed a two-plasmid CRISPRi system in L. lactis (Xiong et al., 2020). The dCas9 was expressed under inducible promoter P_nisin and in one plasmid and the sgRNA for single or multiple target genes was expressed under a strong constitutive promoter P₄₄ and expressed in the other plasmid. This system mediated silencing of single or multiple genes significantly reduce gene expression by up to 99%.

Myrbraten et al (2019) developed a similar two-plasmid CRISPRi system for knockdown of gene expression in L. plantarum, in which the dCas9 and sgRNAs are expressed on separate plasmids. The CRISPRi system was used to preliminarily understand the functions of many key cell cycle genes in L. plantarum.

Although the CRISPR/Cas systems have been developed for robust genetic manipulations, base editors can simultaneously perform genome editing at multiple endogenous loci without inducing DSBs (Figure 4B). Tian et al (2022) successfully established the Cytosine-to-Thymine base editor (CBE) and Adenine-to-Guanine base editor (ABE) in L. lactis by combining dCas9 and deaminase. CBE can be used to simultaneously inactivate multiple genes using a single plasmid. Continuous gene editing by CBE and ABE could be achieved by using temperature-sensitive plasmids, which can be cured quickly. It was found that Cas9 variants SpG and SpRY can expand the editing scope of gene inactivation in L. lactis, and the preference of these variants for the PAM in L. lactis was evaluated.

Natural preservatives like bacteriocins can effectively replace synthetic preservatives because of their excellent antimicrobial and antioxidant properties. However, the low yield and the high cost of production restricted their application as biopreservatives in aquatic products. The introduction of acid-tolerant genes or over-expression of lactic acid synthesis pathway genes can increase the production of bacteriocins by LABs (Zhang et al., 2016). The bacteriocin production levels could also be increased by increasing the carbon conversion rates in the central pathway under oxidizing conditions through the expression of relevant genes (Papagianni and Avramidis 2012). LABs as biological preservatives in aquatic products have been explored in recent years as they can inhibit the growth of spoilage bacteria to prolong the shelf -life of aquatic products. However, some LAB metabolites may influence the sensory characteristics of aquatic products. Genome engineering can be used to improve the production of natural preservatives such as bacteriocins and the performance of LABs.

Lactobacillus or Lactococcus strains are widely used in aquatic products biopreservation, and it is important to develop them with excellent performance. However, conventional gene-editing methods of Lactobacillus or Lactococcus strains are inefficient. For example, the employment of insertion sequence (IS) elements is restricted by the presence and distribution of IS sequences in the genome that have low genetic stability; the employment of Cre-lox-based systems will leave remnant scars at the targeted site; and other systems using selectable markers (such as antibiotic resistance genetic markers that result in low biosafety) can only be used in strains that have a specific gene mutated. Compared with conventional gene-editing systems, the CRISPR/Cas9-based system was easy to operate, trace-free, and had high genetic stability and biosafety.

Recently, the technique of producing superior Lactobacillus or Lactococcus strains through CRISPR/Cas-based genome editing has attracted a large number of researchers’ interest. Currently, the most commonly used Cas protein in Lactobacillus or Lactococcus is Cas9. CRISPR/Cas9 mediated genome editing in Lactobacillus or Lactococcus strains can be mainly divided into three types according to their working principles. One type is CRISPR/Cas9-mediated double-stranded breaks, endogenous or exogenous recombinases-mediated ssDNA or dsDNA recombination to facilitate gene editing, and CRISPR/Cas9 to eliminate unedited cells. The other type is CRISPR/nCas9-mediated single-stranded breaks, where endogenous or exogenous recombinases are used to repair the breaks. Another type is CRISPR/dCas9-mediated gene targeting with double-strand break-free, cytidine and adenosine deaminases to make target base editing. Furthermore, dCas9 itself can be used to suppress gene transcription.

The application of CRISPR/Cas-based genome editing to improve Lactobacillus or Lactococcus for aquatic product biopreservation may be a promising method. Genome editing of Lactobacillus or Lactococcus strains is now approaching the goal of increasingly efficient editing and lower off-target rates, but it is important to see the limitations of CRISPR/Cas technology, such as the cytotoxicity of Cas9 to some Lactobacillus or Lactococcus strains. Many CRISPR/Cas systems perform well in model strains but not in non-model strains. It is difficult to find a universal gene editing tool for all Lactobacillus or Lactococcus strains. Therefore, the development of diverse gene-editing tools with complementary editing strategies is essential.

A series of CRISPR/Cas gene editing systems were developed to allow for wide application. Anzalone et al. (2019) developed a Prime Editor for precision gene editing. They used CRISPR/Cas9 and reverse transcriptase to effectively create all forms of base alterations without DNA templates and DSBs, as well as precise multiple base insertion and deletion. Solving the problem of inserting large gene fragments with recent tools is difficult. Strecker et al. built a CRISPR/Cas-based transposase system that consists of Cas12k (a Type V-K CRISPR effector with a naturally inactivated RuvC-like nuclease domain) and a Tn7-like transposase (Strecker et al., 2019). This system could integrate ∼10 kb of DNA fragments into the E. coli genome without positive selection and generating DSBs. Furthermore, Sternberg et al. reported a type I-F CRISPR/Cas-based Tn7-like transposon system to insert DNA fragments into the E. coli genome that is more efficient than homologous recombination-mediated gene insertion (Klompe et al., 2019). And there are many other Cas proteins widely used in other microorganisms, such as Cas12a, which can also be used in Lactobacillus or Lactococcus, all of which can mediate DSB.

Except for the exogenous CRISPR/Cas system, many of the Lactobacillus or Lactococcus strains possessed one or more endogenous CRISPR/Cas systems, which could be exploited for genome editing after developing an appropriate gRNA. These CRISPR systems offer novel prospects for gene editing in strains of Lactobacillus or Lactococcus, and they are anticipated to be utilized in future studies.