The capsule affects different stages of biofilm formation of K. pneumoniae. It controls the initial adhesion through a series of behaviors, such as improving the regular initial spatial distribution and preventing bacterial interactions as much as possible (Balestrino et al., 2008). Davis et al. pointed out that the capsule was necessary for constructing an appropriate initial covering of mature biofilm structure (Davis and Brown, 2020). The expression of capsular polysaccharide also ensures that K. pneumoniae forms a typical three-dimensional mature biofilm structure (Balestrino et al., 2008).

The capsule is a toxic bacterial component. So far, researchers have identified 134 capsule synthesis loci (K loci) (Wyres et al., 2016). In a study using signature-tagged mutagenesis (STM) to screen a K. pneumoniae mutant library with unique characteristic markers to identify genes related to biofilm formation, the authors found that mutations in capsule gene cluster sites could lead to defects in biofilm formation (Boddicker et al., 2006). Mutants that insert transposons in the capsular wza and wzc loci show defective biofilm formation (Wu et al., 2011). Balestrino et al. (2008) discovered that the biofilm forming ability of K. pneumoniae ORF4 (wza homologous, transport of capsular polysaccharides) and ORF14 (glycosyl transferase, capsule biosynthesis) mutants on polyvinyl-chloride (PVC) was significantly reduced.

Further, treC and sugE have been shown to affect biofilm formation of K. pneumoniae by regulating the production of capsular polysaccharide (CPS). TreC encodes trehalose-6-phosphate hydrolase, and its deletion affects bacterial utilization of trehalose. TreC mutants show reduced mucus viscosity and produce less CPS, thereby reducing biofilm formation and preventing formation of advanced biofilm structures in K. pneumoniae. Adding glucose to the culture medium of K. pneumoniae treC mutant strains can restore CPS production and biofilm formation. However, a sugE (encoding an intima protein) mutant that increases biofilm formation in K. pneumoniae shows higher mucus viscosity and produces more CPS. It was suggested that the absence of sugE in K. pneumoniae would lead to changes in bacterial membrane structure and activate the downstream cascade, thus increasing CPS production during biofilm formation (Wu et al., 2011).

Biofilm formation in isolates containing magA (K1), rmpA+, rmpA2+, the virulence factors related to capsule production, is more obvious than in isolates with negative virulence factors. However, multivariate regression analysis showed that wcaG was the only independent risk factor for biofilm (Zheng et al., 2018). The wcaG positive genotype was involved in K1 and K54 capsular types, and was less associated with K16 and K58 capsular types (Turton et al., 2010). In addition, wcaG encodes the protein participating in the biosynthesis of fucose, and the deletion mutation of wcaG affects most capsule polysaccharide genes (Ho et al., 2011), thereby suggesting that wcaG may affect biofilm formation by changing the composition of capsule polysaccharides (Zheng et al., 2018).

One study also showed that the capsule could inhibit biofilm formation in K. pneumoniae. A recent study showed that strains lacking the wbaP gene, which was related to capsule production, formed stronger biofilms. While the strains with the super capsule containing wzc mutation could not form biofilms (Ernst et al., 2020). When carbohydrates were added to the medium, the biosynthesis of CPS increased, but the biofilm formation in K. pneumoniae decreased (Chen et al., 2020). A previous study showed that the expression of CPS in K. pneumoniae physically interrupted the function of type 1 fimbriae, hindered the biofilm formation mediated by fimbriae, and reduced the adhesion of bacteria to the surface (Schembri et al., 2005). In addition, CPS could inhibit bacterial surface interactions on non-biological substrates (dos Santos Goncalves et al., 2014). It was found that the capsule was costly in nutrient rich media, but it provided obvious adaptive advantages under conditions of malnutrition. Further, among strains forming more biofilms, the capsule often played a positive role in biofilm formation. The authors suggested that this was not because of the presence or absence of capsules, but was instead caused by the amount of capsule expressed by a given strain, which then affected biofilm formation. Moreover, the function of the capsule was not conservative in different isolates, but relied on other elements of the genome or serotype (Buffet et al., 1946).

The mechanism by which the capsule influences biofilm formation and the conditions under which it has positive or negative regulation on biofilm formation are still unclear. However, there seems to be a relationship with the O antigen, because the polysaccharide capsule is retained on the outer surface of the bacteria by interacting with the repeat sugar molecule of the lipopolysaccharide (LPS) molecule, namely “O antigen.” The waaL gene encodes a ligase involved in the connection of the LPS repeat O antigen to the LPS core, and inactivation of this gene is understood to lead to a significant reduction in capsule retention and an increase in biofilm formation (Singh et al., 2022).

The K. pneumoniae genome encodes for several types of fimbriae. Fimbriae are hair-like protein appendages extending from the cell surface (Wilksch et al., 2011). Fimbriae promote K. pneumoniae adhesion to non-biological surfaces, resulting in catheter related infections (Schroll et al., 2010). Type 1 and type 3 fimbriae, the most studied fimbriae, are encoded by fim and mrk gene clusters, respectively. In addition, ecp and kpa to kpg gene clusters found in recent years also encode fimbriae (Wu et al., 2010; Alcántar-Curiel et al., 2013).

In K. pneumoniae, biofilm formation is mainly mediated by type 3 fimbriae, and the Mrk protein is encoded by the operon containing the mrkABCDF gene (Allen et al., 1991). Type 3 fimbriae are mainly made up of the main fimbriae subunit, MrkA, which polymerizes to form a spiral fimbriae axis (Murphy and Clegg, 2012). In addition, ΔmrkA mutants are unable to attach to the abiotic surface to form biofilms (Di Martino et al., 2003; Jagnow and Clegg, 2003). Further, MrkA protein expression is significantly upregulated during biofilm thickening (Vuotto et al., 2017). MrkB and MrkC have sequence characteristics representing the periplasmic chaperone and usher translocatase, respectively. MrkD, present on the top of the fimbriae surface, also has adhesion characteristics of appendages, and determines specificity of fimbriae binding (Murphy and Clegg, 2012). The mrkA and mrkD genes play a key role in the biofilm formation of K. pneumoniae (Fang et al., 2021). The mrkA gene contributes to rapid biofilm formation while mrkD contributes to form dense K. pneumoniae biofilms (Ashwath et al., 2022). A gene cluster, mrkHIJ, adjacent to the type 3 fimbriae operon is related to the regulation of type 3 fimbriae expression. MrkH is a new transcription activator of the mrk gene cluster, which regulate mrkHI expression and contains the PilZ domain. MrkH binds to the region upstream of the mrkA promoter and activates the expression of the mrkABCDF operon. Therefore, mrkH is often referred to as a “biofilm switch” as it can initiate expression of genes involved in producing type 3 fimbriae (Wilksch et al., 2011; Tan et al., 2015). The biofilm formation capacity of K. pneumoniae carrying the mrkH box was clearly higher than strains without it (Fang et al., 2021). Wu et al. (2012) found that mrkHI transcription could be activated by MrkI. MrkI is a LuxR-like regulatory factor. The mrkI mutant reduces the mannose-resistant Klebsiella-like (MR/K) hemagglutinins (HA) activity and the number of type 3 fimbriae on the cell surface, leading to a significant reduction in biofilm formation, which can be rescued when providing wild type mrkI copies (Johnson et al., 2011; Wilksch et al., 2011). The expression of mrkHI is also actively regulated by Fur, which usually acts as a transcriptional activator to directly activate the transcription of mrkHI. The deletion of Fur reduces mrkH, mrkI, and mrkA transcription, thereby reducing type 3 fimbriae expression and biofilm formation (Wu et al., 2012). In addition, at least two components of pulmonary surfactant, phosphatidylcholine and cholesterol, promote the transcription of type 3 fimbriae genes and biofilm formation of K. pneumoniae (Willsey et al., 2018).

The type 3 fimbriae-dependent adhesion is probably the initial stage of K. pneumoniae colonization and biofilm formation on non-biological surfaces (Duguid, 1959). Type 3 fimbriae mediate the binding to the surface of damaged epithelium as they can bind to the extracellular matrix of urinary and respiratory tissues (Tamayo et al., 2007).

Type 3 fimbriae not only participate during the initial stages of K. pneumoniae biofilm formation, but also mediate the c-di-GMP dependent bacterial growth mode transformation from planktonic to biofilm. c-di-GMP is an important second messenger in bacteria (Tamayo et al., 2007). The activity of diguanylate cyclase (DGC) and phosphodiesterase (PDE) can regulate the intracellular concentration of c-di-GMP in bacteria (Simm et al., 2004; Hengge, 2009). The mrkHIJ gene cluster is associated with the regulation and sensing of c-di-GMP (Lin et al., 2016). When activated by c-di-GMP, MrkH recruits RNA polymerase to the mrkHI promoter to auto-activate mrkH expression. Increased MrkH production subsequently drives the expression of mrkABCDF, leading to type 3 fimbriae biosynthesis and biofilm formation (Tan et al., 2015). MrkJ encodes a hypothetical phosphodiesterase (PDE) which contains an EAL domain (the sequences encoding diguanylate cyclase and phosphodiesterase A share a lengthy consensus motif, comprising two adjacent domains termed GGDEF and EAL) mediating the hydrolysis of c-di-GMP (Tal et al., 1998; Johnson and Clegg, 2010). Because of intracellular accumulation of c-di-GMP, the absence of mrkJ leads to an increase in the production of type 3 fimbriae and biofilm formation (Johnson and Clegg, 2010; Wilksch et al., 2011). YjcC possesses PDE activity in the recombinant protein of its EAL domain. After receiving oxidative stress signal input, YjcC actively regulates oxidative stress responses by changing the level of c-di-GMP and has a negative impact on type 3 fimbriae expression and biofilm formation (Huang et al., 2013). YfiN harbors DGC domain plays a positive role in the expression of type 3 fimbriae (Wilksch et al., 2011). OmpR/EnvZ is a two-component system that senses osmotic signals and controls downstream gene expression in many species of Enterobacteriaceae. In response to osmotic stresses, the phosphorylated form of OmpR of K. pneumoniae regulates the expression of type 3 fimbriae to influence biofilm formation via modulating the level of intracellular c-di-GMP and MrkHIJ (Lin et al., 2018).

Type 1 fimbriae, also known as mannose sensitive fimbriae, can bind soluble mannose as a competitive inhibitor, as the name suggests. Type 1 fimbriae are encoded by the fim gene cluster, which is composed of eight genes (fimAICDFGHK) (Gomes et al., 2020). Regulation of fim gene expression is controlled by reversible DNA elements (fimS). Type 1 fimbriae are composed of a major fimbriae subunit, FimA, and a minor apical adhesion protein, FimH (Alcántar-Curiel et al., 2013). FimH is a component that promotes adhesion to the host surface and contains mannose (Sahly et al., 2008). The increased expression of the fimH gene plays an important role in the binding of bacteria to surfaces, leading to strong biofilm formation (Ashwath et al., 2022). The regulatory gene fimK constitutes the fim operon, and FimK has an EAL domain with PDE characteristics, which can regulate intracellular levels of c-di-GMP (Johnson and Clegg, 2010). Mutant strains that cannot produce FimK have a higher fimbriate shape than wild type K. pneumoniae and can be planted in the urinary tract of mice (Johnson and Clegg, 2010). FimK can reduce type 1 fimbriae, and inhibit biofilm formation and intracellular bacterial communities (Rosen et al., 2008). Type 1 fimbriae are key causes of UTIs (Struve et al., 2008) and they have high affinity of mannose residues on bladder cell surfaces (Rozen and Skaletsky, 2000), which can promote the adhesion and invasion of epithelial bladder cells, thus forming biofilm-like intracellular bacterial communities (Rosen et al., 2008). The regulation mechanisms of the biosynthesis of type 3 and type 1 fimbriae are shown in Figure 1.

Type 1 and type 3 fimbriae can contribute to biofilm formation and compensate for each other (Stahlhut et al., 2012; Murphy et al., 2013; Ashwath et al., 2022). One study using the catheter bladder model, found that type 1 and type 3 fimbriae enhance biofilm formation on catheters (Stahlhut et al., 2012). Moreover, it was reported that gene clusters of type 3 and type 1 fimbriae have a cross regulatory effect, and the up-regulation of type 1 fimbriae can make up for the loss of type 3 fimbriae expression (Schroll et al., 2010). Type 3 fimbriae may have a more significant effect on biofilm formation than type 1 fimbriae (Fang et al., 2021). Bacterial strains that cannot produce type 1 fimbriae are as proficient as bacterial strains that can produce such fimbriae in biofilm formation (Clegg and Murphy, 2016).

Alcantar-Curiel et al. found that except for type 3 and type 1 fimbriae, the Escherichia coli common pilus (ECP) fimbriae gene cluster in the K. pneumoniae genome has an operon that is homologous to the E. coli ECP fimbriae. The ECP fimbriae gene cluster contains the ecpRABCDE gene and importantly, 90% of K. pneumoniae strains can produce ECP fimbriae. Ultrastructural and immunoassay analysis of K. pneumoniae showed that ECP can bind bacteria to each other, thus forming specific micro-colonies on cultured epithelial cells and stable biofilms on inert surfaces. ECP likely also plays an important role in cell adhesion, biofilm formation and several niche colonization, especially for isolates lacking MrkD adhesin or the entire type 3 fimbriae (Alcántar-Curiel et al., 2013).

Wu et al. (2010) found seven new fimbriae gene clusters in K. pneumoniae, namely kpa, kpb, kpc, kpd, kpe, kpf, and kpg. The loss of kpgC resulted in an obvious decrease in biofilm formation, adhesion to animal cells, and intestinal colonization in mice. Further, ΔkpaC and ΔkpeC mutants were also found to weaken biofilm formation and adhesion to Arabidopsis cells, respectively. The deletion of the kpjC usher-coding gene was shown to significantly reduce biofilm formation, while the loss of the kpaC usher gene was shown to only affect early and late stages of biofilm formation (Khater et al., 2015). The kpf gene cluster encodes type 1-like fimbriae, while the kpfR gene encoding the transcription inhibitor of the kpf gene cluster negatively regulates the expression of fimbriae. K. pneumoniae lacking the kpfR gene showed a hyperfimbriated phenotype and enhanced adhesion to epithelial host cells and biofilm formation (Gomes et al., 2020).

During biofilm formation, QS mediates inter-specific or intra-specific interactions through which bacterial cells communicate with each other (Bassler et al., 1993; Miller and Bassler, 2001). The QS system regulates the synthesis of fimbriae, exopolysaccharides, adhesins, and other substances through signaling molecules, thus affecting biofilm formation in bacteria (Yang et al., 2013; Gu et al., 2021). Depending on bacterial cell density, bacteria will produce and detect specific signaling molecules called auto inducers (AIs) to coordinate their gene expression (Bassler et al., 1993; Miller and Bassler, 2001). There are two main types of intercellular QS regulatory systems, namely type I and type II.

Type I QS is mainly used for intraspecific communication, which is usually related to the LuxI/LuxR system. LuxI synthetase produces N-acyl homoserine lactone (AHL) as an AI, and the LuxR transcription factor is their homologous receptor. However, K. pneumoniae does not produce AHL (Balestrino et al., 2005), rather encodes SdiA, which is an orphan LuxR receptor that reacts with exogenous AHL molecules produced by other bacteria (Pacheco et al., 2021). SdiA plays a repressive role in the expression of type 1 fimbriae in K. pneumoniae. Cells lacking SdiA regulator presents a hyperfimbriated phenotype that render the ΔsdiA mutant strain with a greater ability to form biofilm and agglutinate yeast cells (Pacheco et al., 2021).

Type II QS has an interspecific communication function, enabling bacteria to react not only to AI-2 produced by other species, but also to their own AI-2 (Chen et al., 2002; Zhang et al., 2020). De Araujo et al. observed that K. pneumoniae lacking AI-2 output (ΔtqsA) or input (ΔlsrCD) systems showed an increased surface coverage after growth in dynamic micro-fermentation but decreased biofilm thickness. In addition, production of AI-2 relies on the presence of luxS but the biofilm structure of ΔluxS mutants is different. In these mutants the surface coverage rate is lower, and fewer large colonies are formed. Mutations related to luxS and AI-2 transport systems both induce increased expression of wbbM and wzm in connection with LPS synthesis, which indicates that QS affects biofilm formation through LPS in K. pneumoniae (De Araujo et al., 2010). The regulation mechanisms of the QS in biofilm formation of K. pneumoniae is shown in Figure 2.

Nutritional conditions are also an important factor for biofilm formation. Excess nutrition may promote the planktonic growth model, while malnutrition environments are more favorable for the biofilm growth model (Stanley and Lazazzera, 2004).

Previous studies have found that high concentration of sugars (such as glucose) prohibit biofilm formation in K. pneumoniae and E. coli (Jackson et al., 2002; Sutrina et al., 2015). Glucose-rich medium inhibits the production of cyclic AMP (cAMP), a well-known second messenger that has important effects on gene regulation (Rickenberg, 1974). Furthermore, cAMP forms a homodimer (CRP-cAMP) with its signal transduction target, cAMP receptor protein (CRP), and then combines with the CRP binding site in the DNA promoter region to regulate mRNA transcription. External glucose inhibits the function of CRP-cAMP in K. pneumoniae. CRP indirectly regulates the expression of type 3 fimbriae through the c-di-GMP signal pathway (Lin et al., 2016). In addition, CRP mediates catabolite repression. The absence of CRP increases the concentration of c-di-GMP and reduces the activity of PDE in cells. The expression of mrkHI depends on c-di-GMP which in turn increases the expression of MrkH and MrkI, leading to the high expression of type 3 fimbriae. It was reported that inserting an open-reading frame containing CRP-activation domain into K. pneumoniae resulted in biofilm deficiency (Boddicker et al., 2006). However, other studies found that crp mutant K. pneumoniae strains could not express MrkA, the major subunit of the fimbrial shaft, which indicated that CRP was required for fimbriae production and biofilm formation (Ou et al., 2017; Panjaitan et al., 2019). These studies indicate the important regulating role of CRP in biofilm formation of K. pneumoniae.

Cellobiose also affects biofilm formation in K. pneumoniae. It has previously been shown that the celB deletion mutation, which leads to cellobiose deficiency, clearly decreased biofilm formation in K. pneumoniae. Moreover, celB encodes the cellobiose-specific subunit IIC of enzyme II (EIIC) of a carbohydrate phosphotransferase system (PTS, a sugar transport system in bacteria) (Wu et al., 2012). Horng et al. (2018) showed that a non-characteristic enzyme II complex homolog of PTS in K. pneumoniae actively regulated biofilm formation by enhancing eDNA and capsular polysaccharide production.

Different carbon sources can also affect the biofilm formation of K. pneumoniae. The isolates formed more robust biofilms when grown with fucose as the sole carbon source than with glucose or glycerol. It was related to the positive modulates of fucose to hypermucoviscosity of K. pneumoniae (Hudson et al., 2022).

The presence of bile salts can stimulate biofilm formation in K. pneumoniae, which is related to the production of poly-β-1,6-N-acetyl-d-glucosamine (PNAG) (Chen et al., 2014). PNAG is a common bacterial surface polysaccharide and a significant component of the biofilm EPS (Chen et al., 2020). PAGA, which is encoded by pgaABCD (Cywes-Bentley et al., 2013), mediates the intercellular binding of bacterial species and surface adhesion. The biofilm formation in pgaA mutants was shown to be significantly decreased (Wu et al., 2011). The loss of pgaC in K. pneumoniae reduces PNGA production, and significantly affects the enhancement of 1% bile salt mixture on K. pneumoniae biofilm (Chen et al., 2014).

Iron is indispensable in K. pneumoniae growth and virulence factor expression (Chhibber et al., 2013; Chen et al., 2020). A study showed that a certain concentration of iron (0.16 mM FeCl₂) could promote biofilm formation in K. pneumoniae by inhibiting succinic acid. This may be due to a reduction in protein and polysaccharide expression in the biofilm EPS since succinic acid participates in pyruvate metabolism and amino acid synthesis (Liu et al., 2022). Chen et al. observed that biofilm formation was strongest when K. pneumoniae was cultured in LB broth supplemented with 50 μM iron. When the strain was cultured with an iron chelator, biofilm formation decreased (Chen et al., 2020). Chhibber et al. studied the biofilm formation of K. pneumoniae in the presence of Co [II] (iron antagonist ions) and depolymerase producing phage (degrading extracellular polysaccharides on biofilm structure). A significant reduction was observed in the growth of younger biofilms (1–3 days old) when 500 μM CoSO₄ and 10 μM FeCl₃ supplemented medium was used. Moreover, a complete eradication of the younger biofilms was observed when both elements were present (Chhibber et al., 2013).

The use of some drugs will instead promote biofilm formation. In the presence of sub-MICs of cefotaxime, the biomass increased and was positively related to the antibiotic concentration (Hennequin et al., 2012). When CRKP strain was under antibiotic pressure, the expression of the Psp and Pho family genes [PspB-PspC complex is a pressure receptor that plays a role in molecular switch during the process of biofilm pressure response (Flores-Kim and Darwin, 2016)] was induced, thus further mediating the downstream stress responses, and compensating for the adsorption, colonization, and biofilm formation (Bowen et al., 2021). Cadavid et al. (2018) found that in K. pneumoniae, hydrochlorothiazide and acetaminophen could promote biofilm formation.

Antibiotic-resistant genes in special plasmids can regulate the biofilm formation of K. pneumoniae (Maeyama et al., 2004). Multidrug-resistant (MDR) K. pneumoniae often forms stronger biofilms than non-MDR strains (Shadkam et al., 2021). The plasmid encoding cephalosporin enzyme was shown to obtain a transcription factor, namely AmpR, which was involved in upregulating capsule synthesis and antiserum killing, regulating the expression of type 3 fimbriae and biofilm formation (Hennequin et al., 2012). In addition, compared to the control strains, strong biofilm formation was found in NDM-1 producing K. pneumoniae. Moreover, the resistance genes blaNDM-1 of K. pneumoniae were observed to be maximally up-regulated in 24 h-biofilms (Al-Bayati and Samarasinghe, 2022).

Bacterial efflux pumps are made up of transmembrane proteins, which export a variety of harmful substances including different types of antibiotics from the intracellular environment to the external environment. This process is one of the causes of MDR (Du et al., 2018). The role of efflux pumps in biofilm formation is still controversial. Tang et al. found that the efflux pump inhibitor, CCCP, had a dose-dependent effect on biofilm formation (Knight et al., 2018). In another study, researchers found that the up-regulation of the AcrAB multidrug efflux system was observed only in XDR strains with biofilm growth that could be considered an essential factor in the biofilm-forming ability in K. pneumoniae (Vuotto et al., 2017). In turn, biofilms have been shown to up-regulate K. pneumoniae efflux pump genes acrA, emrB, oqxA, and qacEΔ1 (Tang et al., 2020). However, some studies have suggested that there is no correlation between the expression of efflux pump genes (acrA, kexD, kdeA, kpnEF and ketM) and biofilm formation (Türkel et al., 2018).

The physical environment of bacteria will affect biofilm formation. Some physical and chemical properties of the surface on which bacteria grow may interfere with biofilm formation by damaging the initial bacterial attachment to the surface (Bos et al., 1999; Li and Logan, 2004). Biofilm production of K. pneumoniae decreased at 37°C compared to at 30°C, but the difference was not significant (Hostacká et al., 2010). In another experiment with 17 CRKP isolates, biofilm formation was greater at 37° C than at 25° C (Gual-de-Torrella et al., 2022). An increase in the pH of culture medium led to an increase in biofilm formation, and in K. pneumoniae this increased by 151–319% at pH 8.5 and by 113–177% at pH 7.5, respectively compared to at pH 5.5 (Hostacká et al., 2010).

Klebsiella pneumoniae growing under simulated microgravity (SMG) conditions formed a thicker biofilm than those growing under normal gravity conditions. Moreover, under SMG conditions, the cellulose production and expression of type 3 fimbriae of K. pneumoniae were enhanced. Therefore, K. pneumoniae isolated from orbital spacecrafts poses a potential threat to the health of astronauts (Wang et al., 2016).

Recently, the CRISPR-Cas9 technique has been implemented to eliminate certain bacteria by use of bacteriophages or bacterial conjugation. This technique allows targeted editing of genomes by inducing double-stranded-DNA breaks (DSBs). However, a novel type of biofilm (“R-biofilm”) was found in clinical isolates of K. pneumoniae after DSBs. R-biofilms are mainly made up of extracellular proteins and/or DNA, which may be released by dead bacteria. In addition to bacterial SOS reaction (severe DNA damage in cells will result in SOS reaction), new signaling pathways also participate in the formation of R-biofilms. Furthermore, R-biofilms form a fixed ring or disk shape with better ductility, which can protect living bacterial cells in the body from harmful conditions such as exposure to ethanol, hydrogen peroxide, and ultraviolet radiation. The discovery of R-biofilms indicated the limited effect of the current popular Cas9-mediated sterilization tools, because the resulting DSBs may facilitate the formation of these new protective biofilms (Liu et al., 2020).