The Man‐PTS subunit ⅡC is responsible for the sensitivity of Listeria monocytogenes to durancin GL

Abstract Target cell recognition is an important issue in the realization of bacteriocin's activity. In this report, we provide genetic and biochemical evidence of durancin GL, a new bacteriocin produced by Enterococcus durans 41D, and use ⅡC subunit in the mannose phosphotransferase system (Man‐PTS) of Listeria monocytogenes as target/receptor. First, the L. monocytogenes mutants with Man‐PTS IIC or IID deletion were constructed with the vector pHoss1. Then, the utilization of glucose and mannose and the sensitivity to durancin GL of the mutant strains were investigated. Afterward, the interactions between durancin GL and the subunits of IIC or IID in Man‐PTS of L. monocytogenes were characterized by yeast two‐hybrid system. The results showed that the L. monocytogenes mutants with either IIC or IID deletion were not only resistant to durancin GL, but also their absorption and utilization of glucose and mannose were not disturbed by the presence of durancin GL. Finally, in situ detection of the interaction between durancin GL and Man‐PTS subunits of IIC or IID by yeast two‐hybrid system revealed that there was a strong interaction between durancin GL and Man‐PTS subunit IIC. However, the interaction between durancin GL and Man‐PTS subunit IID was not present or weak. Based on the experimental evidence above, the Man‐PTS subunit IIC is responsible for the sensitivity of L. monocytogenes to bacteriocin durancin GL.

mild gastroenteritis to severe blood and/or central nervous system infections, as well as abortion in pregnant women (Vazquez-Boland et al., 2001). The lethal rate of listeriosis is more than 25% (Pamer, 2004). In addition, L. monocytogenes was recently reported to have the ability to form biofilm, which brought a bigger problem to food processing (Giaouris et al., 2015;Mathur et al., 2018). Therefore, it is of vital practical significance to inhibit the growth of Listeria and control its contamination in foods to ensure the quality and safety of food.
Recent studies have shown that bacteriocin can be degraded in human body, nontoxic, and target bacteria are not easy to produce resistance (Franz et al., 2018;Mathur et al., 2017). Besides nisin, bacteriocins produced by lactic acid bacteria (LAB) are attracting considerable interest for use as alternative food preservatives (Favaro, Penna, & Todorov, 2015), especially the class IIa bacteriocins that is well recognized for their high antilisterial activity. However, the target cell recognition mechanism of class IIa bacteriocins is poorly understood.
Previous research suggested that mannose phosphotransferase system (Man-PTS) functions in the regulation of various bacterial physiological processes (Postma, Lengeler, & Jacobson, 1993;Reizer et al., 1988;Saier, 1989). The Man-PTS is consisted by four components named IIA, IIB, IIC, and IID. It was reported that there was a putative Man-PTS IIAB component in a leucocin A-resistant strain of L. monocytogenes (Ramnath, Beukes, Tamura, & Hastings, 2000). Also, interruption of either the proximal (mptA) or distal (mptD) gene in Enterococcus faecalis resulted in resistant to mesentericin Y105 (Karine et al., 2001). Thus, the subclass IIa bacteriocin is likely to use a Man-PTS component as specific target (Hechard & Sahl, 2002). Furthermore, sensitivity to class IIa bacteriocins of lactic acid bacteria was recently associated with Man-PTS permease in L. monocytogenes, and it shows that expression of mptC alone is sufficient to confer sensitivity to class IIa bacteriocins in Lactococcus lactis (Ramnath, Arous, Gravesen, Hastings, & Hechard, 2004). The antibacterial mechanism of lactococcin A, a class Ⅱa bacteriocin, on target bacteria was described in detail by Diep, Skaugen, Salehian, Holo, and Nes, (2007). Kjos, Nes, and Diep, (2009) pointed that the level of bacteriocin susceptibility for a bacterial is primarily determined by differences in its Man-PTS proteins, although the expression levels of the corresponding genes also play an important role.
Durancin GL, a new bacteriocin produced by Enterococcus durans 41D, was found to have high antilisterial activity (Du, Somkuti, Renye, & Huo, 2012). By alanine-scanning mutational analysis with site-directed mutagenesis, it showed that durancin GL residues were important for antimicrobial activity and specificity, such as three mutations lost their antimicrobial activity, 10 mutations demonstrated a decreased effect on the activity, and seven mutations exhibited relatively high activity; besides, four mutants demonstrated a narrower antimicrobial spectrum than wild-type durancin GL, and another four mutants displayed a broader target cell spectrum and increased potency relative to wild-type durancin GL (Ju et al., 2015). Study on phenotypic and genotypic alterations of durancin GL-resistant enterococcus durans strains showed that durancin GL can cause damage to bacterial cells of wild bacteria and the increased unsaturated fatty acid and decreased mannose phosphotransferase system gene expression in resistant strains could contribute to durancin GL resistance (Du, Liu, Liu, Ju, & Yuan, 2016). These findings broaden our understanding antimicrobial activity of durancin GL; however, the related mechanism is not clear yet, and this study focused on clarifying the targeting inhibition of L. monocytogenes by durancin GL.

| Bacterial strain and culture condition
Bacterial strains, plasmids, and primers used in this study are listed in

| Effect of durancin GL on Listeria monocytogenes growth curve
Durancin GL was prepared according to the method of Ju et al., (2015) Bacteriocin titer was determined according to the literature (Van Reenen, Dicks, & Chikindas, 1998), and the unit of bacteriocin titer is AU/ml. L. monocytogenes Scott A was inoculated into BHI liquid medium with 1% (v/v) inoculation volume. A final concentration of durancin GL at 100 AU/ml or 200 AU/ml was added to the BHI culture when L. monocytogenes was in logarithmic growth period. The growth of L. monocytogenes was spectrophotometrically measured.
According to the literature (Andrews, 2001), the live bacteria in 100 μl culture above were enumerated on BHI agar plate after serial dilution.

| Effect of durancin GL on carbohydrate utilization by Listeria monocytogenes
According to Premaratne, Lin, & Johnson, (1991), the basic medium with the lowest requirement for the growth of L. monocytogenes was prepared, and the only carbohydrates used in the basic medium above were glucose, fructose, mannose, cellobiose, trehalose, and maltose, respectively. All the compounded carbohydrate medium was sterilized at 121°C for 15 min. After the medium was cooled to room temperature, a certain concentration of durancin GL was added in with a final concentration of 0 AU/ml (as control), 50 AU/ml, 100 AU/ml, and 200 AU/ml, respectively. Then, the growth of L. monocytogenes was followed by OD 600 value.

| Gene knockout box construction
According to the manufacturer's protocol of bacterial genome extraction kit (Sangon Biotech), the genomic DNA in L. monocytogenes was extracted by SZ-10 column silica gel membrane. Man-PTS IIC knockout box was constructed with primer pairs Pr01 (containing Sal I restriction site)/Pr02 and Pr03/Pr04 (containing Bgl II restriction site), and Man-PTS IID knockout box was constructed with primer pairs Pr05 (containing Nco I restriction site)/Pr06 and Pr07/Pr08

| Screening of Listeria monocytogenes gene mutants
Referring to the optimized method by Park & Stewart, (1990 The cells were plated on BHI agar plates containing penicillin (the final concentration was 50 μg/ml) to screen for the transformants.
Single colonies on the plates were selected for PCR verification and then sent to Sangon Biotech Co., Ltd for sequence verification. The Man-PTS IIC and IID gene mutant strains were designated L. monocytogenes GKC and L. monocytogenes GKD, respectively.

| Characterization of Listeria monocytogenes gene mutants
The sensitivity to durancin GL of L. monocytogenes mutants was tested on BHI agar plates with reference to the literature (Seegal & Holden, 1945). The concentration of durancin GL in the test was 50 AU/ml, 100 AU/ml, and 200 AU/ml, respectively. The effect of durancin GL (200 AU/ml) on the growth of L. monocytogenes mutants and the effect on carbohydrate (glucose and mannose) utilization of L. monocytogenes mutants were tested as the same method stated above.

| Construction of prey and bait vectors
The gene of durancin GL (GenBank: HQ696461.1), Man-PTS IIC, and IID (GenBank: AF397145.1) was designated as GL, IIC, and IID, respectively. According to the target gene, primers with two Sfi I restriction sites were designed to amplify these target genes. The gen-

| Self-activation test, cytotoxicity test, and intermolecular interaction test
The recombinant plasmids prepared above were transformed into yeast reporter strain NMY51 chemically competent cells ( Table 2).
The yeast cells were plated on the YPAD (Land Bridge Co. Ltd) agar plate and cultured at 30°C for 4 days. Three colonies of 28 transformed strains (Table 2) were randomly selected from each YPAD plate and diluted with 1 ml sterile water, and then inoculated on SD-TL agar plate and SD-TLHA agar plate. SD-TLHA agar plate contained a certain concentration of 3AT (the final concentration was 0 mM, 5 mM, 10 mM, and 30 mM, respectively).

| Determination of β-galactosidase in transformants
β-galactosidase in transformants was determined by HTX β-galactosidase kit (Dualsystems Biotech). The brief steps were as follows: Firstly, the transformant strains were inoculated with 1% (v/v) inoculum in SD-TL liquid medium and cultured at 30°C with agitation at 250 rpm for 24 hr. Secondly, the supernatant was discarded after centrifugation at 2,500 × g for 5 min. Thirdly, in 96-well plates, 100 μl pyrolysate mixture (mixed by 995 μl one-step lysis and assay reagent and 5 μl dye stock solution) was added into each reaction and incubated for 90 min, and then, OD 615 and at last OD 546 for each well were determined by microplate reader (SpectraMax M2e, Molecular Devices). Each sample was analyzed in triplicate.

| Effect of durancin GL on Listeria monocytogenes growth
Previous studies have indicated that durancin GL has a targeted inhibitory effect on L. monocytogenes (Du et al., 2012;Ju et al., 2015).
This study further confirmed the effect of durancin GL on L. monocytogenes growth (Figure 1). During the logarithmic growth period

| Preparation and characterization of Listeria monocytogenes mutants
Genomic DNA of L. monocytogenes was extracted, and the result of agarose gel electrophoresis showed it was mainly concentrated in  (Figure 3a). The results were consistent with previous experiments (Wu et al., 2017). After sequencing the knockout vectors,

| Intermolecular interaction between durancin GL and Man-PTS subunit IIC or IID
To identify the interaction between durancin GL and Man-PTS subunit IIC or IID, we first constructed the yeast two-hybrid splitubiquitin system to detect this interaction. For this purpose, the  Table 2 were constructed, and a pair of plasmids was introduced into yeast reporter strain NMY51, which has two nutritional compensatory reporter genes (HIS3, ADE2) and a color indicator reporter gene (LacZ). Thus, the interaction between durancin GL and Man-PTS subunit IIC or IID can be determined on the growth of yeast cells on the SD-TL and SD-TLHA plates, as well as

LacZ-dependent color change after incubation in the presence of
LacZ substrates (Figure 4).
In order to ensure the accuracy and correctness of intermolecular interaction results, cytotoxicity test and self-activation test of yeast reporter strain NMY51 with recombinant plasmids were verified (Figure 4a, 4, and 4). Cytotoxicity test showed that all the six bait vectors constructed (pBT3-STE-GL, pBT3-SUC-GL, pBT3-STE-IIC, pBT3-SUC-IIC, pBT3-STE-IID, and pBT3-SUC-IID, in Table 2) can be expressed smoothly in yeast cells (Figure 4c), and LacZ reporter gene was activated with β-galactosidase activity was much higher than the negative control group (Figure 4c and e). However, there is little self-activation for target gene (gene GL, IIC, and IID) located in bait vector (Figure 4b, Figure 4b). As we all know, a certain concentration of 3-AT could eliminate some background growth on the selection plates (Vidalain, Boxem, Ge, Li, & Vidal, 2004), and thus, it was necessary to add a certain concentration of 3-aminotriazole (3-AT, from 0 mM to 30 mM) for inhibiting self-activation (Figure 4a and 4). The results showed that 5 mM 3-AT was sufficient to inhibit the self-activation above.
As the positive control, yeast reporter strain NMY51 with vector pNubG-Fe65 and pTSU2-APP grew well on SD-TL and SD-TLHA plates and showed strong β-galactosidase activity (OD 615 / OD 546 = 1.103) because of the interaction of expressed cytosolic protein Fe65 (amyloid beta A4 precursor protein-binding family B member 1) and APP (amyloid A4 precursor protein), but yeast reporter strain NMY51 with vector pPR3-N and pTSU2-APP did not The plates contain 200 AU/ml durancin GL. (D) Sensitivity of L. monocytogenes and its gene mutants to durancin GL. "-" means no antimicrobial zone, "+" means that it has an antimicrobial zone, and the bacteriocin titer is 50 AU/ml; "++" means that it has an antimicrobial zone, and the bacteriocin titer is 100 AU/ml; and "+++" means that it has an antimicrobial zone, and the bacteriocin titer is 200 AU/ml. grow on SD-TLHA plates and showed little β-galactosidase activity (OD 615 /OD 546 = 0.890), which was used as a negative control. The results showed that yeast reporter strain NMY51 with vector pPR3-N-GL and pBT3-SUC-IIC or pPR3-N-IIC and pBT3-STE-GL or pPR3-N-IIC and pBT3-SUC-GL grew well on the SD-TL and SD-TLHA plates and showed a higher β-galactosidase activity than negative control, which OD 615 /OD 546 was 1.059, 1.153, and 1.129, respectively; however, yeast reporter strain NMY51 with vector pPR3-N-GL and pBT3-STE-IID or pPR3-N-GL and pBT3-SUC-IID did not grow on the SD-TLHA plates, and yeast reporter strain NMY51 with vector pPR3-N-IID and pBT3-STE-GL or pPR3-N-IID and pBT3-SUC-GL grew well same as the self-activation of pPR3-N-IID (Figure 4b).
Based on the above experimental results, it has indicated that there was a strong interaction between durancin GL and Man-PTS subunit IIC. However, the interaction between durancin GL and Man-PTS subunit IID was not present or weak.  (Deegan, Cotter, Hill, & Ross, 2006), Apb 118 (Corr et al., 2007), and sakacin P (Tessema, Moretro, Kohler, Axelsson, & Naterstad, 2009). Besides, the effect of mesentericin Y105, a class IIa bacteriocin, on four sugar utilization by sensitive bacteria showed that the sensitivity of E. faecalis was highly increased in the presence of glucose or mannose, compared to cellobiose or fructose (Hechard, Pelletier, Cenatiempo, & Frere, 2001), and the sensitivity of L. monocytogenes was affected in a medium supplemented with mannose or glucose but not with cellobiose or fructose (Karine, Yves, Pascale, & Yann, 2001). Many reports about Man-PTS components as bacteriocin receptors have been proved experimentally. It was reported that Lactococcin A affected on the carbohydrate utilization by L. monocytogenes, and the correlation between ptn gene and Lactococcin A sensitivity was verified by gene knockout and gene replacement experiments (Diep et al., 2007 (Opsata et al., 2010). Stoll and Goebel, (2010) examine the major PEP-dependent phosphotransferase systems of L. monocytogenes by a systematic deletion analysis and identified the major PTSs involved in glucose, mannose, and cellobiose transport. A growing number of reports suggest that Man-PTS components act as a receptor targeting for bacteriocin (Colombo et al., 2018). Besides, several researches revealed that the Man-PTS subunit IID also was target/receptor for the sensitivity of target bacteria on bacteriocin (Stoll & Goebel, 2010;Tymoszewska, Diep, & Aleksandrzak-Piekarczyk, 2018;Zhou et al., 2016). However, credible evidence was not observed in this study because of interaction between durancin GL and Man-PTS subunit IID was not present or weak.

| D ISCUSS I ON
Although several plasmids have been used for L. monocytogenes generating mutants by allelic exchange, construction of L. monocytogenes mutants has been inefficient due to lack of effective selection markers for first and second recombination events (Abdelhamed, Lawrence, & Karsi, 2015). A new suicide plasmid pHoss1 provides answers to the above questions. Gene knockout Despite its great popularity, the greatest disadvantage of the classical yeast two-hybrid system is the obligatory nuclear localization of the proteins and, hence, their site of interaction. Five years after the initial description of the yeast two-hybrid system, an alternative with the potential to overcome these limitations was described. The split-ubiquitin system never managed to gain as much popularity as the yeast two-hybrid system despite providing the same ease of application, yet it allows virtually all protein types to be tested without the need to truncate or mislocalize these proteins and without introducing additional artifacts to those associated with the yeast two-hybrid system (Xing, Wallmeroth, Berendzen, & Grefen, 2016). The interaction between bacteriocin durancin GL and Man-PTS IIC or IID was detected by yeast two-hybrid split-ubiquitin system. Although both IIC and IID are components of Man-PTS, they interact with bacteriocins durancin GL in the opposite way. There are many possible reasons for this. Firstly, there is no interaction between bacteriocin durancin GL and Man-PTS IID. Secondly, because of the structural differences between Man-PTS IIC and Man-PTS IID, the former has 7 transmembrane regions, and the latter has four transmembrane regions. Thirdly, it may also be related to the location of membrane subunit in yeast cells. All of the above may need further confirmation in future research.
Although this study has revealed the mechanism of bacteriocins durancin GL inhibiting L. monocytogenes, there are still a lot of unclear aspects in its physical and chemical properties, bacteriostasis process and related mechanism and so on. As a new bacteriocin, clarifying the relevant mechanism of durancin GL inhibiting L. monocytogenes is not only provides a new choice for the development of efficient natural food preservatives, but also provides a scientific basis for further clarifying the role of bacteriocins.

| CON CLUS ION
In this study, results indicated that durancin GL, a new bacteriocin, had has obvious antibacterial activity against L. monocytogenes. The presence of durancin GL affected the absorption and utilization of glucose and mannose by L. monocytogenes; however, the L. monocytogenes Man-PTS gene mutants exposed to bacteriocin durancin GL still grow and utilize glucose and mannose in the medium.
Furthermore, an obvious intermolecular interaction between durancin GL and Man-PTS subunit IIC is confirmed in this report. This provides a basis for sensitivity of L. monocytogenes to durancin GL.

ACK N OWLED G M ENTS
This project was financially supported by the National Natural Science Foundation of China (No 31271930) and the Innovation of Graduate Culturing of Jiangsu Province (KYLX_1167).

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

E TH I C A L A PPROVA L
The study did not involve any human or animal testing.

I N FO R M E D CO N S E NT
Written informed consent was obtained from all study participants.