Exploring the Enzymatic and Antibacterial Activities of Novel Mycobacteriophage Lysin B Enzymes.

Mycobacteriophages possess different sets of lytic enzymes for disruption of the complex cell envelope of the mycobacteria host cells and release of the viral progeny. Lysin B (LysB) enzymes are mycolylarabinogalactan esterases that cleave the ester bond between the arabinogalactan and mycolic acids in the mycolylarabinogalactan-peptidoglycan (mAGP) complex in the cell envelope of mycobacteria. In the present study, four LysB enzymes were produced recombinantly and characterized with respect to their enzymatic and antibacterial activities. Examination of the kinetic parameters for the hydrolysis of para-nitrophenyl ester substrates, shows LysB-His6 enzymes to be active against a range of substrates (C4–C16), with a catalytic preference towards p-nitrophenyl laurate (C12). With p-nitrophenyl butyrate as substrate, LysB-His6 enzymes showed highest activity at 37 °C. LysB-His6 enzymes also hydrolyzed different Tween substrates with highest activity against Tween 20 and 80. Metal ions like Ca2+ and Mn2+ enhanced the enzymatic activity of LysB-His6 enzymes, while transition metal ions like Zn2+ and Cu2+ inhibited the enzymatic activity. The mycolylarabinogalactan esterase activity of LysB-His6 enzymes against mAGP complex was confirmed by LC-MS. LysB-His6 enzymes showed marginal antibacterial activity when tested alone against Mycobacterium smegmatis, however a synergetic activity was noticed when combined with outer membrane permealizers. These results confirm that LysB enzymes are lipolytic enzymes with potential application as antimycobacterials.


Introduction
Mycobacterial infections remain among the deadliest and disabling diseases globally, the most common being tuberculosis (TB), a respiratory contagious disease caused by a direct contact with the acid-fast bacterium, Mycobacterium tuberculosis (Mtb). The increasing emergence of antibiotic resistance makes the disease a public health crisis [1]. In 2018, an estimated 10 million people developed TB, with half million rifampicin-resistant new cases (of which 78% had multi-drug resistant TB). Moreover, 1.2 million deaths were reported due to infection with Mtb in HIV-negative patients and 251,000 deaths produced in Escherichia coli. This paper presents the production and characterization of these LysB enzymes with respect to their enzymatic and antibacterial activities. The esterase activity using p-nitrophenyl esters, and lipase activity with Tween substrates were determined. The antibacterial activity of the enzymes was measured alone as well as in combination with anti-TB drugs (Rifampicin, Isoniazid, Pyrazinamide, and Ciprofloxacin), and with antimicrobial cationic polypeptides (Colistin and Protamine sulfate) against Mycobacterium smegmatis, a surrogate strain to the pathogenic Mtb.

Cloning and Expression of LysB-His 6 Genes
Cloning and transformation of the DNA sequences encoding LysB-His 6 proteins (D29, Omega, Saal, Obama12, Enkosi, Echild, DS6A and Pumpkin) into E. coli BL21(DE3) resulted in successful expression of only 4 LysB proteins (D29, Omega, Obama12, and Saal) in soluble, active form in the recombinant bacteria grown in LB or auto-induction media. SDS-PAGE analysis of the enzymes revealed molecular weights consistent with predicted (https://web.expasy.org/protparam/) molecular masses of 29.3 kDa for LysB-D29, 31.5 kDa for LysB-Omega, 37.4 kDa for LysB-Saal, and 36.7 kDa for LysB-Obama12 with His 6 tags (Supplementary Materials Figure S1). Expression of the other 4 LysB proteins (Enkosi, Echild, DS6A and Pumpkin) was not successful in spite of several attempts including different culture media, growth temperature post-induction, IPTG concentration, and even E. coli strains.

Activity of the LysB-His 6 Enzymes Against p-Nitrophenyl Esters
Assay of the esterase activities of the recombinant enzymes against p-nitrophenyl esters, with different carbon chain lengths (C4-C18) showed that the enzymes exhibited varying trends in kinetic parameters against the different substrates (Table 1). The highest activities (V max ) for all the enzymes were obtained with the shortest ester pNPB (C4); the highest V max recorded was 122.3 U/mg for LysB-D29 followed by the LysB-Omega (111.8 U/mg), while LysB-Saal and LysB-Obama12 exhibited significantly lower V max values. There was a significant drop in V max values on increasing the acyl chain length beyond C4, however increase in reaction rate to varying degrees was observed with pNPL (C12) for LysB-Omega, LysB-Saal, and LysB-Obama12, followed again by a decrease in V max . The trend in K m values with the different substrates varied among the four enzymes, LysB-D29 exhibiting the lowest value with pNPL, LysB-Omega, and LysB-Obama12 with pNPP and LysB-Saal with pNPM. Interestingly, the highest catalytic efficiency (k cat /K m ) for LysB-D29, LysB-Omega and LysB-Saal was observed with the C12 substrate, pNPL; the highest value being for LysB-Saal (4.31 µM -1 . min -1 ). On the other hand, the highest k cat /K m for LysB-Obama12 was 2.72 µM -1 . min -1 with the C4 substrate pNPB. For all the LysB-His 6 enzymes no activity was detected with pNPS (C18).

Activity of the LysB-His 6 Enzymes Against Tween Substrates
LysB-D29 was the most active enzyme against Tween substrates with prominent activities of 42.1 and 41.4 kU/mg against Tween 80 and 20, respectively. On the other hand, LysB-Saal exhibited highest activity against Tween 40 and 20 with activity values of 10.8 and 9.6 kU/mg, respectively, while LysB-Obama12 showed highest activity against Tween 80 (4.5 kU/mg) and lowest against Tween 60 (0.31 kU/mg). LysB-Omega exhibited relatively low activity against Tween substrates; the highest activity was 1.14 kU/mg against Tween 40 and 1.03 kU/mg against Tween 80 ( Figure 1).

Activity of the LysB-His6 Enzymes Against Tween Substrates
LysB-D29 was the most active enzyme against Tween substrates with prominent activities of 42.1 and 41.4 kU/mg against Tween 80 and 20, respectively. On the other hand, LysB-Saal exhibited highest activity against Tween 40 and 20 with activity values of 10.8 and 9.6 kU/mg, respectively, while LysB-Obama12 showed highest activity against Tween 80 (4.5 kU/mg) and lowest against Tween 60 (0.31 kU/mg). LysB-Omega exhibited relatively low activity against Tween substrates; the highest activity was 1.14 kU/mg against Tween 40 and 1.03 kU/mg against Tween 80 ( Figure 1).

Figure 1.
Specific activities of LysB-His6 enzymes against different Tween substrates. Fifty microliters of LysB-His6 solution (containing 0.175 µg of the enzyme) were mixed with 250 µL of the assay buffer (0.33% Tween, 50 mM Tris-HCl, 33 mM CaCl2, pH 8) and incubated at 30 °C for 1 h. The absorbance at 400 nm was recorded at 1 min intervals. One unit of lipase activity is defined as the amount of LysB-His6 enzyme that increases the optical density OD400nm of 0.01/min under the assay conditions.

Data Correlation of Esterase and Lipase Activities
The correlation coefficients (−1 < r <1) for the esterase activity data sets against p-nitrophenyl esters and Tweens were calculated to determine if there is positive correlation (large values of one set are associated with large values of the other set) or negative correlation (large values of one set are associated with small values of the other set) or if there is no correlation at all between data sets. Table 2 shows the correlation coefficients displayed in the correlation matrix. Results for Tween 20, 60 and 80 hydrolysis exhibited strong correlation only with pNPB and pNPP, while Tween 40 showed strong correlation with all the tested pNP substrates. Specific activities of LysB-His 6 enzymes against different Tween substrates. Fifty microliters of LysB-His 6 solution (containing 0.175 µg of the enzyme) were mixed with 250 µL of the assay buffer (0.33% Tween, 50 mM Tris-HCl, 33 mM CaCl 2 , pH 8) and incubated at 30 • C for 1 h. The absorbance at 400 nm was recorded at 1 min intervals. One unit of lipase activity is defined as the amount of LysB-His 6 enzyme that increases the optical density OD 400nm of 0.01/min under the assay conditions.

Data Correlation of Esterase and Lipase Activities
The correlation coefficients (−1 < r <1) for the esterase activity data sets against p-nitrophenyl esters and Tweens were calculated to determine if there is positive correlation (large values of one set are associated with large values of the other set) or negative correlation (large values of one set are associated with small values of the other set) or if there is no correlation at all between data sets. Table 2 shows the correlation coefficients displayed in the correlation matrix. Results for Tween 20, 60 and 80 hydrolysis exhibited strong correlation only with pNPB and pNPP, while Tween 40 showed strong correlation with all the tested pNP substrates.

LysB-His 6 Activity Characteristics
Determination of the esterase activity profiles of the LysB-His 6 enzymes against p-nitrophenyl substrates in a pH range of 5-9 showed LysB-Omega, LysB-Saal and LysB-Obama12 to be optimally active at pH 8, and LysB-D29 at pH 7.4 ( Figure 2A). Further increase in pH resulted in a sharp decrease in enzyme activities. The optimum pH for lipase activity against Tween substrates was 8 except for LysB-Saal that was optimally active at pH 7.4 ( Figure 2B).

LysB-His6 Activity Characteristics
Determination of the esterase activity profiles of the LysB-His6 enzymes against p-nitrophenyl substrates in a pH range of 5-9 showed LysB-Omega, LysB-Saal and LysB-Obama12 to be optimally active at pH 8, and LysB-D29 at pH 7.4 ( Figure 2A). Further increase in pH resulted in a sharp decrease in enzyme activities. The optimum pH for lipase activity against Tween substrates was 8 except for LysB-Saal that was optimally active at pH 7.4 ( Figure 2B). The optimum temperature for the esterase activity against pNPB was 37 °C for all the enzymes; LysB-Obama12 and LysB-D29 retained over 80% activity at temperatures up to 60 °C ( Figure 3A). On the other hand, the optimum temperature for lipase activity against Tween 80 was 30 °C, and 50% of the activity was lost at 60 °C ( Figure 3B). The optimum temperature for the esterase activity against pNPB was 37 • C for all the enzymes; LysB-Obama12 and LysB-D29 retained over 80% activity at temperatures up to 60 • C ( Figure 3A). On the other hand, the optimum temperature for lipase activity against Tween 80 was 30 • C, and 50% of the activity was lost at 60 • C ( Figure 3B). Effect of different additives including metal ions, EDTA and PMSF on the enzyme activities was also investigated. Mn 2+ and Ca 2+ ions stimulated the esterase activity of LysB-D29, -Omega and -Saal enzymes against pNPB but had an inhibitory effect on LysB-Obama12. Zn 2+ and Cu 2+ ions, on the other hand, decreased the activity of all the enzymes, and drastic inhibition of the activity was seen with phenylmethane sulfonyl fluoride (PMSF) ( Table 3). The metal ions (except for Mn 2+ ) showed adverse effects on the lipase activity against Tween 80 of LysB-D29, Zn 2+ , and Cu 2+ ions being the most inhibitory. In contrast, the lipase activity against Tween 80 of LysB-Omega, was boosted to 300%, 225%, and 212% of the original activity with Zn 2+ , Mn 2+ and K + , respectively. Regarding LysB-Saal, Zn 2+ totally abolished its activity, however EDTA enhanced the activity to 256% of original activity. The activity of LysB-Obama12 was stimulated by all the metal ions (except Zn 2+ and Cu 2+ ) as well as EDTA. The lipase activity of all LysB-His6 enzymes was inhibited by PMSF (Table 3). Effect of different additives including metal ions, EDTA and PMSF on the enzyme activities was also investigated. Mn 2+ and Ca 2+ ions stimulated the esterase activity of LysB-D29, -Omega and -Saal enzymes against pNPB but had an inhibitory effect on LysB-Obama12. Zn 2+ and Cu 2+ ions, on the other hand, decreased the activity of all the enzymes, and drastic inhibition of the activity was seen with phenylmethane sulfonyl fluoride (PMSF) ( Table 3). The metal ions (except for Mn 2+ ) showed adverse effects on the lipase activity against Tween 80 of LysB-D29, Zn 2+ , and Cu 2+ ions being the most inhibitory. In contrast, the lipase activity against Tween 80 of LysB-Omega, was boosted to 300%, 225%, and 212% of the original activity with Zn 2+ , Mn 2+ and K + , respectively. Regarding LysB-Saal, Zn 2+ totally abolished its activity, however EDTA enhanced the activity to 256% of original activity. The activity of LysB-Obama12 was stimulated by all the metal ions (except Zn 2+ and Cu 2+ ) as well as EDTA. The lipase activity of all LysB-His 6 enzymes was inhibited by PMSF (Table 3).

Hydrolysis of mAGP
The hydrolytic activity of LysB-His 6 enzymes against mAGP was evaluated using LC-MS in negative-ion mode. LysB attacks the ester bond between the arabinosyl units and mycolic acid, resulting in free mycolic acid. Figure 4 shows HPLC chromatogram for the different samples. mAGP treated with LysB-D29 reveals a peak corresponding to the liberated mycolic acid, while no peak was observed in the sample treated with Rhizopus oryzae lipase and the negative control. The observed m/z value of 1278.19 confirmed the presence of mycolic acid in LysB-D29 treated mAGP samples ( Figure 5). The hydrolytic activity of LysB-His6 enzymes against mAGP was evaluated using LC-MS in negative-ion mode. LysB attacks the ester bond between the arabinosyl units and mycolic acid, resulting in free mycolic acid. Figure 4 shows HPLC chromatogram for the different samples. mAGP treated with LysB-D29 reveals a peak corresponding to the liberated mycolic acid, while no peak was observed in the sample treated with Rhizopus oryzae lipase and the negative control. The observed m/z value of 1278.19 confirmed the presence of mycolic acid in LysB-D29 treated mAGP samples ( Figure 5).

Antibacterial Activity
LysB-His6 enzymes did not show any antibacterial activity against M. smegmatis as a function of MIC and MBC, when tested in different media. However, a marginal Log10 reduction of 1.1, 1.32, 1.44, and 1.36 for LysB-D29, -Omega, -Saal, and -Obama12, respectively could be detected with 100 µg/mL of LysB-His6 in Cation Adjusted Mueller-Hinton Broth (CAMHB) ( Table 4). Moreover, combining LysB-His6 enzymes with different anti-TB drugs did not exert synergistic effect. On the other hand, the inhibitory effect of LysB-His6 enzymes was enhanced by addition of half MIC values of Colistin and Protamine sulfate, respectively; the highest reduction being obtained with Obama12 (Table 4 and Supplementary material Table S1). Incubation of M. smegmatis cells with Colistin and Protamine sulfate at half MIC values did not result in a decrease in the Log10 values.

Antibacterial Activity
LysB-His 6 enzymes did not show any antibacterial activity against M. smegmatis as a function of MIC and MBC, when tested in different media. However, a marginal Log 10 reduction of 1.1, 1.32, 1.44, and 1.36 for LysB-D29, -Omega, -Saal, and -Obama12, respectively could be detected with 100 µg/mL of LysB-His 6 in Cation Adjusted Mueller-Hinton Broth (CAMHB) ( Table 4). Moreover, combining LysB-His 6 enzymes with different anti-TB drugs did not exert synergistic effect. On the other hand, the inhibitory effect of LysB-His 6 enzymes was enhanced by addition of half MIC values of Colistin and Protamine sulfate, respectively; the highest reduction being obtained with Obama12 (Table 4 and Supplementary Materials Table S1). Incubation of M. smegmatis cells with Colistin and Protamine sulfate at half MIC values did not result in a decrease in the Log 10 values.

Thermostability of LysB-His 6 Enzymes
Thermostability of the enzymes measured as the melting temperature (T m ) using Nano DSF, showed LysB-Omega to be the most stable enzyme with a T m value of 57.7 • C followed by LysB-D29 (T m of 54.7 • C). On the other hand, LysB-Obama12 and -Saal showed lower T m values of 47.9 • C and 45.7 • C, respectively (Table 5).

Discussion
Despite the availability of ≈1885 mycobacteriophage complete genome sequences available on the Actinobacteriophage database (https://phagesdb.org/) (accessed April 2020), the knowledge of their lytic enzymes is fairly limited. Concerning LysB enzymes, crystal structure of only LysB-D29 is available so far and some studies on the antimycobacterial properties of a few other LysB candidates have been reported [26][27][28]32,33]. The present study was started by selection of 8 LysB candidates which exhibited 30-76% sequence analogy with LysB-D29. Only 4 (including LysB-D29) of the 8 LysB enzymes could be expressed and purified to homogeneity. Payne and coworkers have earlier observed high variations in expression and solubility levels of LysB proteins and were able to obtain good expression levels of only LysB-D29 among trials with expression of several LysB enzymes [28].
LysB enzymes hydrolyze the ester bonds between the mycolic acid and arabinogalactan as well as the disaccharide trehalose. The three LysB enzymes characterized so far have exhibited structural relatedness and activity patterns similar to the lipolytic enzymes belonging to the α/β hydrolase family including esterases, lipases, and cutinases [32]. While esterases catalyze the hydrolysis of glycerol esters with short acyl chains and lipases act on long chain triacylglycerol substrates [34], cutinases act on wide range of substrates from soluble p-nitrophenyl esters to insoluble long-chain triglyceride esters besides its natural substrate cutin [35,36]. LysB enzymes are considered to be a link between lipases and cutinases as they have the ability to hydrolyze soluble esters and emulsified triglycerides [27]. As reported earlier, amino acid sequence analysis revealed the cutinase motif to be conserved among the majority of LysB enzymes [30,31]. Our recent study has further reported several common structural features of Lys B enzymes with the α/β hydrolase members [30].
The lipolytic activities i.e., esterase and lipase activities of LysB-His 6 enzymes were evaluated against p-nitrophenyl esters and Tween substrates, respectively, both with varying carbon chain lengths. In general, the activity levels of the different enzymes varied significantly with respect to the different substrates. Measurement of the esterase activity against p-nitrophenyl esters (C4-C16) showed 3 of the enzymes to exhibit highest catalytic efficiency towards C12 (pNPL) ( Table 1), which is in accordance with the earlier report on LysB-Ms6 having the highest catalytic efficiency of 1.14 µM -1 . min -1 against pNPL [27]. The decrease in the esterase activity of LysB-D29 and other LysB enzymes with increasing chain lengths of the p-nitrophenyl ester substrates is also in agreement with the previous report on LysB-D29 showing the highest activity against the C4 substrate pNPB (0.72 U/mg) [28]. It was further reported that the K m values of LysB-Ms6 and -Bxz2 decreased with the increase in carbon chain length of the pNP esters [33]. LysB-Bxz2 showed 10-fold higher esterase activity than Ms6 but with lower catalytic efficiency [33]. This might be attributed to the different conformations of the active sites of LysB proteins (tunnel, shallow bowl, deep, superficial funnels and deep buried cave) influencing the binding of different substrates as suggested in our previous study [30]. We noted that LysB enzymes with longer hydrophobic acyl binding site, e.g., LysB-D29 (tunnel) and LysB-Omega (shallow bowl) have higher V max values against long fatty substrates than the LysB enzymes with shorter ones, e.g., LysB-Saal (deep funnel) and LysB-Obama12 (inverted tunnel).
Tunnel-shaped active sites of LysB-D29 and -Obama12 confer long, open hydrophobic cervix that accept long chain fatty substrates (Supplementary Materials Figure S3a). However, the inverted positioning of pNP ligands upon docking to LysB-Obama12 active site (Supplementary Materials Figure  S3b) directs the long hydrophobic ligand tail towards the narrow hydrophilic mouth of the tunnel where no sufficient hydrophobic acyl binding site is provided. This might explain the highest activity of LysB-D29 on pNPP (C16) among the tested LysB-His 6 enzymes in contrast to LysB-Obama12 with the lowest activity against the same substrate (Table 1). Furthermore, the shallow bowl conformation, was shown earlier to accept long substrates up to pNPS with C18 acyl chain length [37]; it provides a long hydrophobic superficial groove for fitting the long hydrophobic tail of pNP ligands, which might account for LysB-Omega achieving the second highest specific activity on pNPP (Supplementary Materials Figure S3c). On the other hand, LysB-Saal has a deep funnel conformation, which was reported earlier to suit bulky ligands with acyl chain length up to C14, with a narrow opposing hydrophobic wall wrapping the fatty pNP tail (Supplementary Materials Figure S3d). The narrow and short acyl binding site of LysB-Saal might explain its diminished activity against fatty substrates with long acyl side chain and its highest affinity towards pNPM (Table 1) [30]. In contrast, shorter substrates (up to C8) are better accepted by shorter and shallower funnel conformation e.g., LysB-Bxz2 [30,31,37].
Tweens 20, 40, 60, and 80 are esters of lauric (C12), palmitic (C16), stearic (C18), and oleic (C18) acids. Tween 20 and 80 turned out to be the best substrates for most of the enzymes, probably due to the relatively lower melting point of the corresponding fatty acids, especially for oleic acid. However, LysB-Saal showed higher activity against Tween 40 (10.8 kU/mg) (Figure 1). The higher enzymatic activity recorded against Tweens compared with p-nitrophenyl substrates may be attributed to the high emulsifying properties of the detergents forming stable oil-in-water emulsions, allowing for greater interactions between the substrate and the enzyme [38,39]. It is worthy to note that the mycobacterial cutinases as well as phospholipase A isolated from M. smegmatis show the same patterns of activity correlation between Tweens and pNP substrates [40][41][42][43].
LysB-His 6 enzymes were shown to be active over a wide range of temperature (with optimum of 37 • C against pNPB and 30 • C against Tween 80) and in the pH range of 7.4-8, which seems to be similar to the activity pattern of different lipolytic enzymes reported earlier [44]. The melting temperatures of LysB-His 6 enzymes ranged between 45-60 • C which correlates well with both lipase and esterase activities of the enzymes; increasing the reaction temperature above 60 • C, resulted in drop of the enzymatic activity of more than 60% for some of the enzymes. As reported earlier for LysB-Ms6 [27], the metal ions Mn 2+ , Ca 2+ , and Mg 2+ generally showed a stimulating effect on the esterase and lipase activities of most LysB-His 6 enzymes, while the transition metal ions, Zn 2+ and Cu 2+ had an inhibitory effect (Table 3). Activation by Ca 2+ and inhibition by Zn 2+ and Cu 2+ have also been reported for esterase/lipase from Geobacillus thermoleovorans [45]. The effect of Ca 2+ is related to the structural stabilization and activation of lipases [46,47], while Zn 2+ and Cu 2+ ions lead to activity loss due to the metal ion catalyzed oxidation of the sensitive amino acids such as histidine that is an important part of the catalytic triad in the α/β hydrolases [48,49]. EDTA had a slight inhibitory effect on the esterase activity that may suggest chelation of some stimulatory metal ions present in the samples [50]. On the other hand, the reason underlying the significant increase in the lipase activity of most of the enzymes with EDTA is not entirely clear but may be related either to increased flexibility of the enzyme or complexation of potential inhibitory compounds in the reaction. The activity inhibition by PMSF is attributed to its interaction with the conserved nucleophilic Ser residue in the catalytic triad, which is typical of esterases, lipases as well as proteases, hence confirming that LysB enzymes belong to the group of serine hydrolases [27,51].
The hydrolytic activity of LysB-His 6 enzymes against the isolated mAGP substrate is also in accordance with the previously reported mycolylarabinogalactan esterase activity of other LysB homologous enzymes Ms6 [26], Bxz2 [33], D29 [28], and Bxb1 [52]. This confirms that the LysB enzymes are capable of hydrolyzing the ester bonds not only from inside the cells when viral progeny is released but also when applied from outside.
LysB-His 6 enzymes in this study did not exhibit significant antibacterial activity either alone or in combination with anti-TB drugs. The activity was determined without Tween 80, which has been suggested to be required for removal of M. smegmatis cell clumps and aggregates due to surface hydrophobicity and also for promoting the antibacterial activity of LysB enzymes [33,[57][58][59][60]. There is also a possibility that the antibacterial activity observed in the presence of Tween 80 could be due to the action of the oleic acid released by hydrolysis of the surfactant by LysB [26,33]. The differences in the antibacterial activity among the LysB-His 6 enzymes tested in this study are very subtle when compared with the high diversity of their catalytic profiles (in terms of esterase and lipase activities), which is clearly related to the diversity of active site conformations among LysB enzymes.
While LysB-His 6 enzymes did not exert notable inhibitory effect against M. smegmatis cells, half MIC values of the polycationic peptides colistin and protamine sulfate enhanced the Log 10 reduction of M. smegmatis treated with 100 µg/mL LysB-His 6 enzymes (Table 5). Colistin and Protamine sulfate are cyclic cationic polypeptides with a hydrophobic tail, which destabilize the bacterial outer membrane by quenching the divalent cations (Ca 2+ and Mg 2+ ) from the lipopolysaccharides and phospholipids creating pores and subsequently cell death [61]. The antibacterial activity of cationic polypeptides extends beyond Gram-negative bacteria to different species of Mycobacteria including M. avium [62], M. aurum, M. vaccae [63], M. xenopi [64], M. smegmatis [65], and Mtb [60,65]. Due to the permeabilizing effect of Colistin on Mycobacteria cell membrane, it has been used in combination with anti-TB drugs to facilitate their uptake [66,67].
Even with the combination of LysB-His 6 and polycationic peptides, we could not detect MIC/MBC values. For achieving significant antimycobacterial activity, a combination of antimicrobial peptides (outer membrane permeabilizer), LysB (mAGP hydrolase) and LysA (peptidoglycan hydrolase) should be considered. Earlier reports have demonstrated that fusions of antimicrobial peptide with either LysA or LysB did not exert antimycobacterial activity, however mixing the two fusion proteins together resulted in enhanced antimycobacterial activity even for intracellular mycobacteria in lung macrophages cell line [68].

Cloning and Expression of LysB-His 6 Genes
Cloning and expression of LysB-His 6 gBlocks were performed as reported earlier [30]. This involved cloning in EcoRI and NdeI restriction sites of pET22b(+) expression vector, ligation with T4 DNA ligase, and transformation of the ligation mixtures into E. coli BL21(DE3) expression host. The bacteria were grown overnight at 37 • C on LB agar plates supplemented with 100 µg/mL ampicillin. The plasmids (pET22b(+)-LysB-His 6 ) extracted from the transformant colonies were sequenced (GATC Biotech AB, Solna, Sweden), and those with correct sequences were used for protein expression in E. coli BL21(DE3), which was grown overnight at 37 • C, 200 rpm in LB medium supplemented with the antibiotic as above. The recombinant cells were suspended in 50% glycerol, distributed in xx µL aliquots and stored at −80 • C.
Small scale production of the recombinant LysB-His 6 enzymes was done in LB medium supplemented with 100 µg/mL ampicillin. The respective glycerol stocks were inoculated into 10 mL of the medium in 50 mL sterile falcon tubes and grown overnight at 37 • C and 200 rpm, and 5 mL of the culture were used to inoculate 50 mL medium in 250 mL Erlenmeyer flask and grown under the same conditions. When the optical density (OD600 nm ) reached 0.5−0.6, the cells were induced for protein expression using IPTG at a final concentration of 1mM and reducing the cultivation temperature to 30 • C. After 4 h, the cells were harvested by centrifugation at 3900× g and 4 • C for 15 min (Sigma 3−16PK, Sigma Laborzentrifugen GmbH, Osterode am Harz, Germany).
Protein production at a larger scale was performed in auto-induction medium (1% tryptone, 0.5%, yeast extract, 25 mM Na 2 HPO 4 , 25 mM KH 2 PO 4 , 25 mM (NH 4 ) 2 SO 4 , 2 mM MgSO 4 , 0.05% glucose and 0.2% α-lactose) supplemented with 100 µg/mL ampicillin. One liter of the auto-induction medium in 5 L Erlenmeyer flask was inoculated with 15 mL of recombinant cell culture grown in LB medium as above, incubated at 37 • C and 180 rpm for 4 h, and then at 30 • C for 24 h prior to harvesting the cells by centrifugation at 6000× g and 4 • C for 20 min (Sorvall Lynx 4000 centrifuge, Thermo Scientific, Waltham, MA, USA).

Purification of LysB-His 6 Enzymes
The cell pellet obtained above from 1 L auto-induction medium was suspended in 50 mM Tris-HCl buffer, pH 8 supplemented with 50 mM NaCl and a cocktail of protease inhibitors (Calbiochem) and then sonicated on ice (5 × 60 s, cycle 0.5) using UP400S sonicator (Dr. Hielscher GmbH, Stahnsdorf, Teltow, Germany). After removal of the cell debris by centrifugation (18,500× g, 30 min, 4 • C, Sorvall RC5C, Sorvall Instruments, Dupont, Wilmington, DE, USA), the clarified lysate was subjected to immobilized metal ion affinity chromatography (IMAC) for purification of the LysB-His 6 enzymes using 5 mL HisTrap FF TM nickel column (GE Healthcare Bio-Sciences AB, Uppsala, Sweden). Briefly, 15 mL of the clarified lysate was applied on the column pre-equilibrated with the binding buffer

Activity Against p-Nitrophenyl Esters
The esterase activity of the purified LysB-His 6 enzymes was measured against p-nitrophenyl esters with variable carbon chain length as substrates. Stock solutions (250 mM) of pNP-esters were prepared in dichloromethane and stored at −20 • C. For the assay, 180 µL of pNP-esters (final concentration of 1 mM dissolved in 20 mM Tris-HCl, 100 mM NaCl, 0.1% Triton X-100, pH 8) was mixed with 20 µL LysB-His 6 enzyme solution (containing 0.175 µg enzyme). The reaction mixture was incubated at 37 • C and the release of p-nitrophenol was followed at 410 nm with 1 min intervals for 30 min (Multiskan TM GO Microplate Spectrophotometer, Thermo Scientific, Vantaa, Finland). Each assay was performed in three independent replicates, using the corresponding buffer as negative control unless otherwise stated.
The activity was calculated and quantified with the following equation using calibration curve for p-nitrophenol.
where ε is the extinction coefficient of p-nitrophenol which was determined under the assay conditions and at different pH values. One unit (U) of enzyme activity corresponds to the amount of enzyme liberating 1 µmol of p-nitrophenol per min under the assay conditions. Optimum conditions for the enzyme activity were determined using 1 mM pNPB as substrate. To determine the optimum temperature, the assay was performed in a temperature range of 25-60 • C, while optimum pH was determined in a pH range of 5-9 using 20 mM sodium phosphate buffer (pH 5-7) and 20 mM Tris-HCl (pH 7.4-9), respectively.
The substrate specificity of the LysB-His 6 enzymes was measured using pNP esters with variable carbon chain length ranging from C4 to C18. The kinetic parameters were determined by running the enzymatic activity assay using the substrates in a concentration range of 10 µM-4 mM and calculated according to the Michaelis-Menten model.
The effect of metal ions (5 mM Ca 2+ , Mn 2+ , Cu 2+ , Mg 2+ , Zn 2+ , K + , Na + ), 5 mM EDTA and 10 mM phenylmethane sulfonyl fluoride (PMSF), respectively, was determined by incubation with LysB-His 6 enzymes at room temperature for 2 h, and then determining the residual esterase activity using pNPB as substrate under the standard assay conditions at pH8 and 37 • C. The relative activity (%) was calculated in relation to the activity of the enzymes incubated under the same conditions but without any metal ions. %Relative activity = (Enzymatic activity with metal ion) × 100 Enzymatic activity without metal ion

Activity Against Tween Substrates
The lipase activity of LysB-His 6 enzymes was determined against Tween 20, 40, 60, and 80 as substrates [32], which are esters of lauric, palmitic, stearic, and oleic acids, respectively. The assay is based on the cleavage of Tween into alcohol and the corresponding fatty acids, which in the presence of calcium ions form insoluble calcium salts of fatty acids that can be measured turbidimetrically at 400 nm. In a microtiter plate, 250 µL of the assay buffer (composed of 0.33% Tween dissolved in 50 mM Tris-HCl, 33 mM CaCl 2 , pH 8) was mixed with 50 µL of LysB-His 6 enzymes (final amount of 0.175 µg in the reaction mixture) and incubated at 30 • C for 1 h. The absorbance at 400 nm was recorded at 1 min intervals (Multiskan TM GO Microplate Spectrophotometer). One unit of lipase activity is defined as the amount of LysB-His 6 enzyme that increases the optical density OD400 nm of 0.01/min under the assay conditions. Optimum temperature, pH, as well as the effect of metal ions and enzyme inhibitors on the lipase activity were also determined as described above using Tween 80 as substrate.
A correlation matrix between the esterase activities against different pNP substrates and the lipase activity against different Tweens was performed using IBM SPSS Statistics 25 (IBM, Armonk, NY, USA).

Determination of Protein Stability
Stability of LysB-His 6 enzymes was determined using Nano-Differential Scanning Fluorimetry (DSF) on a Prometheus NT.48 instrument (NanoTemper Technologies, GmbH, Germany). Ten microliter samples of 2 µM LysB-His 6 enzymes in 50 mM Tris-HCl buffer (pH 8) were loaded into UV capillaries (NanoTemper Technologies), and subjected to increasing temperature from 20 to 95 • C with a temperature gradient of 1 • C/min. Protein unfolding was detected by tracking changes in the intrinsic tryptophan/tyrosine fluorescence (350 nm/330 nm fluorescence ratio) as a function of temperature increase. Data was analyzed using the ThermControl software V 2.0.4 (NanoTemper Technologies) and melting temperature (T m ) was calculated.
4.6. Hydrolysis of M. smegmatis mAGP 4.6.1. Preparation of Mycolylarabinogalactan-Peptidoglycan (mAGP) Substrate mAGP was extracted from M. smegmatis according to the method described earlier [33] with modification. Briefly, M. smegmatis cells were grown overnight in 1L of LB medium supplemented with 0.05% Tween 80 at 37 • C and 200 rpm. The cells were harvested by centrifugation at 3900× g and 4 • C for 20 min and washed three times with phosphate buffered saline (PBS) pH 7.4. The extractable lipids of the cell pellet were extracted with a 2:1 mixture of chloroform/methanol (15 mL per g cell pellet) at 55 • C overnight, followed by 3 successive washing steps with MilliQ quality water. Subsequently, the pellet was resuspended in 30 mL of 2% SDS and stirred gently overnight at room temperature and then collected by centrifugation (3900× g, 4 • C, 20 min), resuspended again in 30 mL of 2% SDS and refluxed for 1 h. SDS was removed through five successive washings (with MilliQ quality water)/centrifugation steps. The pellet was then washed with 30 mL of 80% acetone followed by 30 mL of diethyl ether, 30 mL of 80% acetone, and twice with water at the end. The resulting mAGP was lyophilized and stored at 4 • C for further experiments.

Treatment of mAGP with LysB-His 6 Enzymes
The purified mAGP (1 mg) was suspended in 1 mL PBS pH 7.4 containing 0.2% Triton X-100 through sonication (2 × 60 s, cycle 0.5), followed by addition of LysB-His 6 (100 µg), and incubation for 24 h at room temperature with shaking at 300 rpm. The liberated mycolic acid was extracted with 3 mL of diethyl ether and washed with water. The organic phase (upper layer) was separated from the aqueous phase by centrifugation (3900× g, 4 • C, 5 min), collected and air dried. After drying, the samples were dissolved in 1 mL of organic solvent (80% n-propanol, 20% hexane) and analyzed by liquid chromatography-mass spectrometry (LC-MS) using an Accela 600 high-pressure liquid chromatograph (HPLC) coupled to an LTQ-Orbitrap hybrid mass spectrometer (Thermo Scientific). The LC was equipped with a C18 reversed-phase column (Phenomenex Kinetex XB-C18 50 × 2.1 mm, 1.7 µm particle size) and elution was performed with a mobile phase comprising 40% solvent A (100% methanol) and 60% solvent B (80% n-propanol and 20% hexane) at a flow rate of 0.2 mL/min over 10 min. The MS was operated in electrospray ionization mode with detection of negatively charged ions in the m/z range 500-1500 Da at a resolution of ∼100,000. Negative control (buffer replacing the LysB-His 6 enzymes) and a control using Rhizopus oryzae lipase were included in the assay. The microtiter plates were incubated at 37 • C for 24 h and examined for growth. For viable counting plating assay, 100 µL aliquot from each well was diluted in PBS containing 0.05% Tween 80 and plated on 7H10 agar supplemented with OADC (Oleic acid-Albumin-Dextrose Complex), 0.5% glycerol and 0.05% Tween 80. The plates were incubated at 37 • C for 48 h followed by colony counting. Since 7H9 medium contains mono-and divalent cations (Na + , K + , Mg 2+ , Zn 2+ , and Ca 2+ ), we noticed formation of a precipitate upon addition of LysB-His 6 enzymes which might be due to formation of oleic acid salts upon hydrolysis of Tween 80 by LysB-His 6 enzymes. Also, adding LysB-His 6 enzymes to Tween 80 free 7H9 medium resulted in a faint turbidity, so we decided to exclude 7H9 medium when testing the antibacterial activity of LysB-His 6 enzymes. Instead, Mueller-Hinton broth (MHB), Cation Adjusted Mueller-Hinton Broth (CAMHB), and LB culture media, respectively, were tested.

In Vitro Testing of the Combined Effect of LysB-His 6 Enzymes with Anti-TB Drugs
The following antibiotics were tested: Rifampicin (RIF), Isoniazid (INH), Ciprofloxacin (CIP), and Pyrazinamide (PYZ). Firstly, the minimum inhibitory concentrations (MIC) and the minimum bactericidal concentrations (MBC) of those antibiotics were determined. For MIC, antibiotics stock solutions were serially diluted (2-fold) in MHB and mixed with M. smegmatis (1 × 10 6 CFU/mL) in a total volume of 200 µL, the reaction mixture was incubated at 37 • C for 24 h and examined for bacterial growth. MIC was defined as the lowest antibiotic concentration of the antibiotic that inhibited the growth of M. smegmatis (no visible growth at the end of the experiment).
MBC was evaluated by plating 100 µL suspension from the wells with no visible growth on 7H10 agar medium and incubating at 37 • C for 48 h and checking for growth. MBC was defined as the lowest antibiotic concentration that showed no growth. Positive controls (100 µL of bacterial culture + 100 µL MHB) and negative controls (MHB mixed with antibiotics) were included. The experiment was performed in triplicates independently. To test the combined effect of LysB-His 6 enzymes with anti-TB drugs, modified checkerboard technique was applied [69]. Briefly, in a microtiter plate, antibiotics (half the MIC values) were diluted two-fold vertically (columns) in MHB, while LysB-His 6 enzymes were diluted two-fold horizontally (rows). M. smegmatis cells were added to the wells to a final volume of 200 µL to give (1 × 10 6 CFU/mL). The plates were incubated at 37 • C for 24 h and examined for growth. The combination effect is synergistic when the effect of the combined drugs is significantly greater than the effects of each drug individually, antagonistic when the combination effect is lower than the effect of each drug individually or additive when the effect is equal to the sum of the separate effect of each drug [70] The MIC and MBC values for Colistin (Col) and Protamine sulfate (Prot) were determined as described earlier in CAMHB medium, and modified checkerboard technique was performed to determine the combinatorial effect of LysB-His 6 enzymes with Col and Prot as described above. M. smegmatis cells incubated with half MIC values of Colistin and Protamine sulfate were included as controls.

Conclusions
Our study clearly demonstrates that LysB enzymes are lipolytic enzymes (combining features of esterases and lipases) that hydrolyse a range of fatty substrates besides the mAGP complex, their natural substrate in the mycobacterial cell envelope. When applied externally to M. smegmatis cells, LysB-His 6 enzymes showed marginal antimycobacterial activity. However, combining outer membrane permealizers with LysB-His 6 enzymes enhanced the antimycobacterial activity. The potential application of LysB enzymes to disrupt and kill the mycobacterial cells is currently being investigated in combination with both LysA and antimicrobial peptides.