key: cord-0425847-30c3norx
authors: van Schie, Loes; Borgers, Katlyn; Michielsen, Gitte; Plets, Evelyn; Vuylsteke, Marnik; Tiels, Petra; Festjens, Nele; Callewaert, Nico
title: Exploration of synergistic action of cell wall-degrading enzymes against Mycobacterium tuberculosis
date: 2021-04-06
journal: bioRxiv
DOI: 10.1101/2021.04.05.438542
sha: 0b975bda26d55e0cf66e4f3df2e232946ced07e4
doc_id: 425847
cord_uid: 30c3norx

Background The major global health threat tuberculosis is caused by Mycobacterium tuberculosis (Mtb). Mtb has a complex cell envelope – a partially covalently linked composite of polysaccharides, peptidoglycan and lipids, including a mycolic acid layer – which conveys pathogenicity but also protects against antibiotics. Given previous successes in treating gram-positive and -negative infections with cell wall degrading enzymes, we investigated such approach for Mtb. Objectives (i) Development of an Mtb microtiter growth inhibition assay that allows undisturbed cell envelope formation, to overcome the invalidation of results by typical clumped Mtb-growth in surfactant-free assays. (ii) Exploring anti-Mtb potency of cell wall layer-degrading enzymes. (iii) Investigation of the concerted action of several such enzymes. Methods We inserted a bacterial luciferase-operon in an auxotrophic Mtb strain to develop a microtiter assay that allows proper evaluation of cell wall degrading anti-Mtb enzymes. We assessed growth-inhibition by enzymes (recombinant mycobacteriophage mycolic acid esterase (LysB), fungal α-amylase and human and chicken egg white lysozymes) and combinations thereof, in presence or absence of biopharmaceutically acceptable surfactant. Results Our biosafety level-2 assay identified both LysB and lysozymes as potent Mtb-inhibitors, but only in presence of surfactant. Moreover, most potent disruption of the mycolic acid hydrophobic barrier was obtained by the highly synergistic combination of LysB, α-amylase and polysorbate 80. Conclusions Synergistically acting cell wall degrading enzymes are potently inhibiting Mtb – which sets the scene for the design of specifically tailored antimycobacterial (fusion) enzymes. Airway delivery of protein therapeutics has already been established and should be studied in animal models for active TB.

Background 27 The major global health threat tuberculosis is caused by Mycobacterium tuberculosis (Mtb). Mtb has a complex 28 cell envelope -a partially covalently linked composite of polysaccharides, peptidoglycan and lipids, including a 29 mycolic acid layer -which conveys pathogenicity but also protects against antibiotics. Given previous successes 30 in treating gram-positive and -negative infections with cell wall degrading enzymes, we investigated such 31 approach for Mtb. 32 Objectives 33 (i) Development of an Mtb microtiter growth inhibition assay that allows undisturbed cell envelope formation, to 34 overcome the invalidation of results by typical clumped Mtb-growth in surfactant-free assays. (ii) Exploring anti- 35 Mtb potency of cell wall layer-degrading enzymes. (iii) Investigation of the concerted action of several such 36 enzymes. 37 Methods 38 We inserted a bacterial luciferase-operon in an auxotrophic Mtb strain to develop a microtiter assay that allows 39 proper evaluation of cell wall degrading anti-Mtb enzymes. We assessed growth-inhibition by enzymes 40 (recombinant mycobacteriophage mycolic acid esterase (LysB), fungal α-amylase and human and chicken egg 41 white lysozymes) and combinations thereof, in presence or absence of biopharmaceutically acceptable surfactant. 42 

Our biosafety level-2 assay identified both LysB and lysozymes as potent Mtb-inhibitors, but only in presence of 44 surfactant. Moreover, most potent disruption of the mycolic acid hydrophobic barrier was obtained by the highly 45 synergistic combination of LysB, α-amylase and polysorbate 80. Bacteriophage PG-degrading enzymes (endolysins) are increasingly investigated as antibacterial agents, with 65 several products for topical use against Gram-positive infections already on the market, 12,13 and progress being 66 made for Gram-negative pathogens. 14-16 Such products are rapidly gaining attention due to the looming antibiotics 67 resistance crisis, together with improved know-how in biopharmaceutical protein production and formulation for 68 nebulization or dry powder inhalation. In the context of respiratory diseases such as TB, it is encouraging that several recombinant biopharmaceutical protein treatments (e.g. dornase alpha to reduce viscosity of airway 70 mucus) have been successfully developed for inhalation. 17 However, due to their distinctive cell wall, 71 bacteriophage-mediated lysis of mycobacteria is more complicated than that of other Gram-positives. In addition 72 to endolysin, mycobacteriophages employ a mycomembrane-targeting mycolylarabinogalactan esterase (LysB) to 73 lyse mycobacteria. 18 Obviously this could affect the validity of results obtained with cell wall degrading enzymes, and precludes 92 exploration of surface-active components in the enzyme cocktail. In this study, we solve both problems by developing a bioluminescent derivative of a biosafety level 2 triple auxotrophic Mtb. 33 The bacterial lux operon 94 used does not require cellular uptake of any cofactor, 34 which allows for quantifying metabolic activity in static, 95 clumped cultures. 96 We set out to systematically assess the potency of enzymes that degrade the various layers of the Mtb cell wall, 97 using cultivation conditions that keep the Mtb cell wall/biofilm intact. Apart from PG and the mycomembrane, we 98 targeted the outer capsule of Mtb. Considering that α-glucan is a main constituent of this capsule, 10 we speculated 99 that Aspergillus oryzae α-amylase, an α-glucan hydrolyzing enzyme already used in human medicine to treat 100 pancreatic insufficiency, could have a capsule-destabilizing effect and potentially synergise with the known 101 capsule destabilization imparted by polysorbate 80. Furthermore, we hypothesized that the anti-TB potency of 102 cell wall-degrading enzyme treatments would be enhanced by using the enzymes in cocktails, which could 103 potentially 'peel' the cell envelope layer-by-layer and synergistically weaken it (Figure 1 ). Amylase-induced 104 hydrolysis of the capsular layer may, for instance, increase permeability for enzymes such as LysB that degrade 105 the mycomembrane and/or for enzymes such as lysozyme that degrade the PG layer underneath. 106 

Materials 108 Chicken egg white lysozyme was purchased at Sigma-Aldrich (L6876, ≥90% protein), as was human lysozyme 109 recombinant from rice (L1667, ≥90% pure) and Aspergillus oryzae α-amylase (A8220, ≥800 fungal amylase units/g). The release of p-nitrophenol after hydrolysis of the pNPB substrate was measured as increase in absorbance at 400 nm for 30 minutes, and the ∆A400nm/minute was calculated from the linear region of the blank-corrected values 138 after a lag-phase. The specific enzyme activity was calculated using the micromolar extinction coefficient of 139 p-nitrophenol at 400 nm at 37°C of 0.0148. One LysB enzyme unit releases 1 nmol of p-nitrophenol per minute at 140 pH 8 at 37 °C using pNPB as substrate. capacitance, 800 Ω resistance). After selection on kanamycin, the obtained Mtb mc 2 7902_Lux strain was stored in 158 1 ml aliquots at OD600 1 in 20% glycerol at -80 °C. For each drug testing assay, two Mtb mc 2 7902_Lux aliquots were 159 thawed, combined and cultured at 37 °C, shaking, in a 60 ml square bottle in 20 ml of Middlebrook 7H9 broth (BD Diagnostics) supplemented with 0.05%V/V polysorbate 80, 10% oleic acid-albumin-dextrose-catalase supplement 161 (OADC; BD Diagnostics), 0.5%V/V glycerol, 1 mM L-arginine, 50 µg/ml L-leucine and 24 µg/ml L-pantothenate. At 162 days 5 and 8 after thawing, bacteria were subcultured by 1/100 and 1/20 dilution, respectively, for use in setting 163 up the antimicrobial assays on day 9. Middlebrook 7H10 agar (BD Diagnostics) supplemented with 10% OADC, 164 0.5%V/V glycerol, 1 mM L-arginine, 50 µg/ml L-leucine and 24 µg/ml L-pantothenate was used for growth of Mtb 165 on solid culture; colony counts after incubation at 37 °C for up to 12 weeks were given as colony forming units per 166 ml plated (cfu/ml). 167 A bioluminescence microtiter assay for mycobacterial growth inhibition 168 To remove bacterial clumps before setting up the microtiter assay, mycobacteria were passed three times through 169 a 27G needle. A Mtb mc 2 7902_Lux inoculum at an OD600 of 0.008 (2•10 6 cfu/ml) was prepared in assay medium 170 (Middlebrook 7H9 broth supplemented with 10% OADC, 0.5%V/V glycerol, 1 mM L-arginine, 50 µg/ml L-leucine and 171 24 µg/ml L-pantothenate) without any polysorbate 80 unless specifically stated otherwise. (Figure 2a-b) . Minor low molecular weight bands were observed in purified Lys-Ms6, probably indicating low level 202 protein degradation. Enzymatic activity on a pNPB substrate seemed highly dependent on the bacteriophage from 203 which the enzyme was derived, with LysB-D29 displaying three-fold higher specific activity than LysB-Ms6 (Table   204 1, not significant, p=0.2, Mann-Whitney test). Enzymatic activity of commercially available lysozyme was validated 205 using a Micrococcus PG degradation assay (Table 1, 

The effect of sublethal injury by heating, freezing, drying and gamma-radiation on the duration of the lag 394 phase of Salmonella typhimurium

Combenefit: an interactive platform for the analysis and visualization of drug combinations

SciPy 1.0: fundamental algorithms for scientific computing in Python

Section 14.7 Spearman's rank correlation coefficient

Shortened multidrug-resistant tuberculosis regimens overcome low-level fluoroquinolone resistance. Eur Respir 402

Safety and efficacy of inhaled nebulised interferon beta-1a (SNG001) for treatment of SARS-CoV-404 2 infection: a randomised, double-blind, placebo-controlled, phase 2 trial

Development of a drug delivery system for efficient alveolar delivery of a neutralizing monoclonal 407 antibody to treat pulmonary intoxication to ricin

Measurements of Deposition, Lung Surface Area and Lung Fluid for Simulation of

Inactive Ingredient Search for Approved Drug Products

Mucus hypersecretion: a common symptom, a common mechanism?

Sputum Volume Predicts Sputum Mycobacterial Load during the First 2 Weeks of 415 Antituberculosis Treatment

Antimicrobial activity of Mycobacteriophage D29 Lysin B during Mycobacterium ulcerans infection

Mechanisms of tolerance and allergic sensitization in the airways and the lungs

Figure 3. The effect of various antimycobacterial compounds on bioluminescence of Mtb mc 2 7902_Lux. (a) Scheme of the generalized logistic function as 434 applied to a bioluminescence growth assay and its main parameters. (b) Effect of antimycobacterial agents in absence (purple circles) and presence (cyan 435 triangles) of 0.05% polysorbate 80 (PS80). Luminescence detected after 4 days of incubation at 37 °C. If data allowed

If inhibition was observed but no sigmoid curve could be validly fitted, the MIC was 439 defined as lower than or equal to the lowest value for which all datapoints were lower than 200 RLU (corresponding to half the inoculum). Data shown were 440 derived from three independent experiments, with each data point an average of duplicate plates within a repetition of the experiment. Mean and standard 441 deviation are shown. (c) For several conditions from the high and low ends of inhibition curves of LysB-Ms6 (circles) and human lysozyme (triangles)

We thank the W.R. Jacobs lab for kindly providing the Mtb mc 2 7902 strain and Dr. B. Robertson, Dr. S. Wiles and 313 Dr P. Sander for expression plasmids used in this study.

Tables 456 Table 1 . In vitro activity-testing of potential antimycobacterial enzymes. Lipolytic activity of LysB enzymes was determined in vitro on a pNP-butyrate 457 substrate. One LysB enzyme unit (U) will release 1 nmol of p-nitrophenol per minute at pH 8 at 37 °C. Peptidoglycan (PG)-degrading activity of human and 458 hen egg white lysozyme was determined on a fluorescein-labeled Micrococcus lysodeiticus cell wall substrate. One lysozyme enzyme unit (U) will produce a 459 0.001 units per minute change in the absorbance at 450 nm of at pH 6. 24