Broad-Spectrum In Vitro Activity of Nα-Aroyl-N-Aryl-Phenylalanine Amides against Non-Tuberculous Mycobacteria and Comparative Analysis of RNA Polymerases

This study investigates the in vitro activity of Nα-aroyl-N-aryl-phenylalanine amides (AAPs), previously identified as antimycobacterial RNA polymerase (RNAP) inhibitors, against a panel of 25 non-tuberculous mycobacteria (NTM). The compounds, including the hit compound MMV688845, were selected based on their structural diversity and previously described activity against mycobacteria. Bacterial strains, including the M. abscessus complex, M. avium complex, and other clinically relevant NTM, were cultured and subjected to growth inhibition assays. The results demonstrate significant activity against the most common NTM pathogens from the M. abscessus and M. avium complexes. Variations in activity were observed against other NTM species, with for instance M. ulcerans displaying high susceptibility and M. xenopi and M. simiae resistance to AAPs. Comparative analysis of RNAP β and β′ subunits across mycobacterial species revealed strain-specific polymorphisms, providing insights into differential compound susceptibility. While conservation of target structures was observed, differences in compound activity suggested influences beyond drug–target interactions. This study highlights the potential of AAPs as effective antimycobacterial agents and emphasizes the complex interplay between compound structure, bacterial genetics, and in vitro activity.


Introduction
Within the field of infectious diseases, the exploration of mycobacterial infections has always revolved around the formidable presence of Mycobacterium tuberculosis (Mtb), rightfully claiming its place as a global health concern [1].However, amidst this predominant focus, a group of less discussed yet clinically relevant entities is emerging-non-tuberculous mycobacteria (NTM) [2][3][4][5].NTM, comprised of over 190 species, are ubiquitous in the environment, inhabiting soil, water sources, and various organic substrates.These mycobacterial species, distinct from their tuberculosis-causing congener, have garnered increasing attention in recent years due to their diverse clinical manifestations and growing significance in healthcare settings worldwide [6].
While historically deemed harmless environmental dwellers, their potential to cause diseases in immunocompromised populations and individuals with chronic respiratory con-

Selection of Nα-Aroyl-N-Aryl-Phenylalanine Amide Compounds
The compounds selected for assessment against the strains described in this study were synthesized and characterized as described previously [25,32].The selection of the test set considered their activity against previously tested mycobacteria, as well as diverse structural features that could affect the activity against different NTM species.Compounds 1 to 7 were synthesized during a detailed SAR study that varied the ring systems of the chemical scaffold in the search for better activity against NTM.The morpholine moiety was substituted by thiomorpholine sulfoxides and sulfones, which proved advantageous for the activity and solubility of the compound class, in addition to the exchange of the thiophene carboxylic acid amide to 2-fluorobenzoic acid amides.Compound 6 was included because it showed high activity while harbouring the hydroxyl group [25].We also wanted to determine the influence of 5-fluoro substitution in para-position to the morpholine moiety (compounds 4 to 6).Compounds 8 to 10 showed high anti-NTM activity and have been synthesized and tested in a previous study with the aim to improve the stability of the compound class by additional sterical hindrance of the amide bonds [32].The molecular structures of the tested compounds are given in Scheme 1 while their previously published activities against NTM can be found in Table 1.For comparative purposes, the initial hit compound of phenylalanine amides, MMV688845 (MMV), was included.Additionally, clarithromycin (CLR) was used as a positive control due to its clinical relevance.mycobacterial resistance.

Selection of Nα-Aroyl-N-Aryl-Phenylalanine Amide Compounds
The compounds selected for assessment against the strains described in this study were synthesized and characterized as described previously [25,32].The selection of the test set considered their activity against previously tested mycobacteria, as well as diverse structural features that could affect the activity against different NTM species Compounds 1 to 7 were synthesized during a detailed SAR study that varied the ring systems of the chemical scaffold in the search for better activity against NTM.The morpholine moiety was substituted by thiomorpholine sulfoxides and sulfones, which proved advantageous for the activity and solubility of the compound class, in addition to the exchange of the thiophene carboxylic acid amide to 2-fluorobenzoic acid amides Compound 6 was included because it showed high activity while harbouring the hydroxy group [25].We also wanted to determine the influence of 5-fluoro substitution in paraposition to the morpholine moiety (compounds 4 to 6).Compounds 8 to 10 showed high anti-NTM activity and have been synthesized and tested in a previous study with the aim to improve the stability of the compound class by additional sterical hindrance of the amide bonds [32].The molecular structures of the tested compounds are given in Scheme 1 while their previously published activities against NTM can be found in Table 1.For comparative purposes, the initial hit compound of phenylalanine amides, MMV688845 (MMV), was included.Additionally, clarithromycin (CLR) was used as a positive control due to its clinical relevance.

Inhibition of M. abscessus Complex
All the AAPs that were tested displayed activity against the selected Mycobacterium abscessus complex strains.An overview of the calculated MIC 50 values is given in Table 2.As microbial populations are often diverse and different strains may respond differently to antimicrobial agents under assay conditions, we decided to utilize the MIC 50 as a comparative measure of activity, because in some cases, MIC 50 values may reflect a more representative and comparable average than MIC 90 values, as the latter can be influenced by outliers and/or growth and plate effects [33].The respective MIC 90 values for each compound and strain calculated from the same data sets are displayed in the Supplementary Materials.Testing CLR and the AAP hit compound MMV resulted in MIC 50 values well comparable to those reported in the literature [22,33] (see Figure 1 for exemplary dose-response curves).The obtained MIC 50 values of the AAPs are generally in the low micromolar concentration range between 0.1 µM and 2.1 µM.While the MIC 50 values against subsp.abscessus were comparable to those against subsp.massiliense, there was a pattern of slightly lower activities against subsp.bolletii, which was also reported for other antibiotics [34].Structures containing a sulfone moiety generally exhibit lower MIC 50 values, with compound 8 displaying the highest activity against all subspecies (subsp.abscessus: 0.2 µM, subsp.massiliense: 0.1 µM, subsp.bolletii: 0.3 µM) translating to a 7-19-fold enhancement in activity against the various subspecies when compared to the hit compound MMV.Compound 7 and 8 that exhibited promising in vitro activity (MIC 90 of 0.78 µM) against M. abscessus subsp.abscessus ATCC 19977 were further investigated against a range of clinical isolates of the M. abscessus complex, expanding the evaluation to a broader range of genotypes within this species (Table 3).The results demonstrated a comparable potency to the type strains, confirming their potential as effective antimicrobial agents for the treatment of M. abscessus infections.
Antibiotics 2024, 13, 404 5 against subsp.abscessus were comparable to those against subsp.massiliense, there w pattern of slightly lower activities against subsp.bolletii, which was also reported for o antibiotics [34].Structures containing a sulfone moiety generally exhibit lower M values, with compound 8 displaying the highest activity against all subspecies (su abscessus: 0.2 µM, subsp.massiliense: 0.1 µM, subsp.bolletii: 0.3 µM) translating to a fold enhancement in activity against the various subspecies when compared to th compound MMV.Compound 7 and 8 that exhibited promising in vitro activity (MIC 0.78 µM) against M. abscessus subsp.abscessus ATCC 19977 were further investig against a range of clinical isolates of the M. abscessus complex, expanding the evalua to a broader range of genotypes within this species (Table 3).The results demonstrat comparable potency to the type strains, confirming their potential as effe antimicrobial agents for the treatment of M. abscessus infections.

Inhibition of M. avium Complex
The examined AAPs demonstrate substantial activity against the selected M. avium complex strains, with the determined MIC 50 values being in line with those against the M. abscessus complex (Table 4).A promising observation is that these derivatives show potent activity against the M. avium complex, with a 5-10-fold improvement in in vitro activity over the hit compound MMV688845.The high clinical relevance of M. avium complex infections, for which rifampicin's therapeutic benefits are controversial, emphasizes the potential of AAPs as novel RNAP inhibitors [29,35].

Inhibition of Other NTM
The panel of AAPs was tested against a variety of NTM that occasionally occur as human pathogens.The respective MIC 50 values are displayed in Table 5.The selection showed activities in a comparable range to those seen against the M. abscessus complex and M. avium complex.Notable differences were seen for M. xenopi and M. simiae, against which activities were reduced.While clarithromycin showed an MIC 50 of 0.03 µM against M. xenopi, M. simiae was less susceptible (MIC 50 10.7 µM) to clarithromycin, which is consistent with what has been reported in the literature [36][37][38].M. ulcerans showed a high susceptibility to AAPs with MIC 50 values as low as 10 nM.A further difference is that certain mycobacterial species react differently to 6, the only compound in which phenylalanine is replaced by tyrosine.The presence of an additional hydroxyl group reduces the activity in M. kansasii, M. malmoense, M. marinum and M. szulgai, with activities always lower than those of the hit compound MMV.For M. ulcerans and M. xenopi, 6 shows the highest activities among all tested compounds.This suggests that a higher polarity of the compounds may be advantageous for in vitro activity against these strains.

Comparative Analysis of RNAP β and β' Subunits
In this study, we employed comparative alignment analysis to investigate the genomic diversity among the selected mycobacterial strains and elucidate whether alterations of the protein primary structure within the binding pocket of AAPs could explain the observed differences in AAP susceptibility.By integrating the primary target sequences of the β and β ′ RNAP subunits from the mycobacterial strains in protein-protein alignment (performed within positions 450-600 and 800-880 for the β and β ′ subunits, respectively, the complete alignment in this area can be found in the Supplementary Materials), we identified strainspecific amino acid variations in comparison to the reference sequence of a published protein structure of Mtb RNAP that was co-crystallized with the AAP [29] analog D-AAP1 (PDB: 5UHE).An overview of the contacts of D-AAP1 and its RNAP binding site based on PDB: 5UHE is given in Figure 2. AAPs are highly active against the Mtb-type strain ATCC 25618 H37Rv [22,24,25].The results of the alignment analysis are displayed in Figure 3.The table was constrained to show amino acid variations located within a 7 Å distance of target-bound D-AAP1 [29] to the surrounding amino acids to limit the analysis to the area around the binding site.Mtb RNAP positions that did not show variations for any strain were excluded from the depiction as well as strains that did not show any polymorphisms in these areas.
The overall sequence identity of the Mtb β subunit to the NTM β subunits is high (89% to 95%), with the trend that the fast-growing mycobacteria (M.chelonae, M. fortuitum and M. abscessus) show lower identity values (89-91%), while the skin pathogens M. ulcerans and M. marinum have 95% identity each and even 99% identity in the region that contains the AAP binding site (positions 450-600).M. chelonae, M. fortuitum, M. xenopi and M. abscessus all display an alanine-to-glycine variation at position 565 of the β subunit in close proximity to the AAP binding site.However, this variation does not appear to affect their susceptibilities possibly due to the minor differences in volume and polarity between alanine and glycine.In M. fortuitum, a leucine to methionine exchange (Grantham's distance 15, a measure of the similarity of amino acids in protein structures that combines the composition, volume and polarity for comparison, in which small values indicate high similarity and high values indicate low similarity [39]) occurs at position 560, which does not seem to result in lower activities in this case.This particular position reportedly resulted in a resistant M. abscessus Bamboo strain after a leucine to proline exchange [28] (Grantham´s distance 98).The leucine at position 560 ensures a transition to a random coil formation (R562-V568) that is in direct contact with AAP structures and is therefore crucial for the right orientation of the binding pocket.As proline disrupts secondary structures, this variation could cause a conformational change in the binding site that cannot be compensated.Additionally, M. fortuitum and M. xenopi display a proline to serine exchange at position 477, which is an essential lipophilic binding contact to AAP's anilide aromatic system and the phenylalanine aromatic system.Its importance for the interaction was demonstrated by the formation of a resistant mutant after a proline to leucine variation [28].Leucine exhibits similar lipophilic properties as the proline side chain, but its higher spatial demand causes clashes with the anilide aromatic system and the random coil formation (R562-V568), altering the arrangement of the binding pocket and leading to AAP resistance.The proline-serine exchange of M. fortuitum and M. xenopi leads to a binding pocket that is less hydrophobic but has a similar volume to the native proline conformation, resulting in AAP activity.However, the binding of AAPs could be restricted due to the differences in polarity, which might be a part of the explanation for the slightly lower activities in M. fortuitum and the loss of activity against M. xenopi.
genomic diversity among the selected mycobacterial strains and elucidate whether alterations of the protein primary structure within the binding pocket of AAPs could explain the observed differences in AAP susceptibility.By integrating the primary target sequences of the β and β′ RNAP subunits from the mycobacterial strains in proteinprotein alignment (performed within positions 450-600 and 800-880 for the β and β′ subunits, respectively, the complete alignment in this area can be found in the Supplementary Materials), we identified strain-specific amino acid variations in comparison to the reference sequence of a published protein structure of Mtb RNAP that was co-crystallized with the AAP [29] analog D-AAP1 (PDB: 5UHE).An overview of the contacts of D-AAP1 and its RNAP binding site based on PDB: 5UHE is given in Figure 2. AAPs are highly active against the Mtb-type strain ATCC 25618 H37Rv [22,24,25].The results of the alignment analysis are displayed in Figure 3.The table was constrained to show amino acid variations located within a 7 Å distance of target-bound D-AAP1 [29] to the surrounding amino acids to limit the analysis to the area around the binding site.Mtb RNAP positions that did not show variations for any strain were excluded from the depiction as well as strains that did not show any polymorphisms in these areas.Comparing the sequence identities of the Mtb β ′ subunit with their NTM counterparts, we observed high degrees of homology (90-97% sequence identity), whereas the lowest value of 90% was found for M. abscessus and the highest values were again found for M. ulcerans and M. marinum, which showed 97% each.The β' subunit of Mtb shows two clusters of amino acids that build up the binding surface to AAPs.Only two variations were observed within a 7 Å distance to the AAP binding site.A prominent variation that all displayed NTM strains exhibit is the valine to isoleucine exchange at position 836.This position is close to the aryl carboxylic acid amide structure of the AAPs and contributes to the lipophilic surface that interacts with the aromatic system.We do not expect that this amino acid exchange affects activities, as the additional methylene group in isoleucine is not largely affecting the properties of the binding pocket in the matter of polarity.The additional expansion does not appear to influence binding and activity.The only other variation present was found in M. abscessus ATCC 19977, where phenylalanine 831 is changed to tyrosine (Grantham's distance 22).However, this alteration does not seem to affect the activity against M. abscessus ATCC 19977.The side chain of the phenylalanine is oriented away from the AAP binding site into an unrelated, open cleft resulting in no direct interaction.Phenylalanine and tyrosine share similar properties, making it probable that the same is true for tyrosine.Still, the exchange could lead to differences in the geometry of the binding pocket due to its proximity to the binding site.The overall sequence identity of the Mtb β subunit to the NTM β subunits is high (89% to 95%), with the trend that the fast-growing mycobacteria (M.chelonae, M. fortuitum and M. abscessus) show lower identity values (89-91%), while the skin pathogens M. ulcerans and M. marinum have 95% identity each and even 99% identity in the region that contains the AAP binding site (positions 450-600).M. chelonae, M. fortuitum, M. xenopi and M. abscessus all display an alanine-to-glycine variation at position 565 of the β subunit in close proximity to the AAP binding site.However, this variation does not appear to affect their susceptibilities possibly due to the minor differences in volume and polarity between alanine and glycine.In M. fortuitum, a leucine to methionine exchange (Grantham's distance 15, a measure of the similarity of amino acids in protein structures that combines the composition, volume and polarity for comparison, in which small values indicate high similarity and high values indicate low similarity [39]) occurs at position 560, which does not seem to result in lower activities in this case.This particular position reportedly resulted in a resistant M. abscessus Bamboo strain after a leucine to proline exchange [28] (Grantham´s distance 98).The leucine at position 560 ensures a transition to a random coil formation (R562-V568) that is in direct contact with AAP structures and is therefore crucial for the right orientation of the binding pocket.As proline disrupts secondary structures, this variation could cause a conformational change in the binding site that cannot be compensated.Additionally, M. fortuitum and M. xenopi display a proline to serine exchange at position 477, which is an essential lipophilic binding contact to AAP's anilide aromatic system and the phenylalanine aromatic system.Its importance for the interaction was demonstrated by the formation of a resistant mutant after a proline to leucine variation [28].Leucine exhibits similar lipophilic properties as the proline side chain, but its higher spatial demand causes clashes with the anilide aromatic system and the random coil formation (R562-V568), altering the arrangement of the binding pocket and leading to AAP resistance.The proline-serine exchange of M. fortuitum and M. xenopi leads to a binding pocket that is less hydrophobic but has a similar volume to the native proline conformation, resulting in AAP activity.However, the binding of AAPs could be The aromatic system of the phenylalanine part of AAPs extends into a lipophilic cleft of the β subunit.One of the constituents of the surface of this cleft is proline at position 477, which was previously discussed for M. fortuitum and M. xenopi.This amino acid is in close proximity to the para position of the phenylalanine group of AAPs.In M. xenopi, the exchange from proline to serine could induce a geometric shift that provides an additional hydrogen bond between the serine backbone and the tyrosine hydroxyl group.This could explain the higher potency of 6.However, for the strains that show reduced susceptibility to 6, no variations in the lipophilic cleft were found, making it challenging to explain the difference in activity.

Bacterial Cultures and Strains
For general bacteria culturing and inhibition experiments, Middlebrook 7H9 broth (BD Difco) was supplemented with 0.5% albumin, 0.2% glucose, 0.085% sodium chloride, 0.0003% catalase, 0.2% glycerol, and 0.05% Tween 80 ® .Most bacterial strains analyzed in this study were purchased as type strains either from the American Type Culture Collection (ATCC) or the Culture Collection University of Goteborg (CCUG) as indicated.M. abscessus subsp.abscessus Bamboo was isolated from the sputum of a patient with amyotrophic lateral sclerosis and bronchiectasis and was provided by Wei Chang Huang, Taichung Veterans General Hospital, Taichung, Taiwan.Clinical isolates covering the M. abscessus complex (M9, M199, M337, M404, M422, M232, M506, and M111) were provided by Jeanette W. P. Teo (Department of Laboratory Medicine, National University Hospital, Singapore).Detailed information on the origin of these isolates is given in reference [40].M. avium subsp.hominissuis strain 109 (MAC109) was isolated from the blood of a patient with AIDS and was provided by Petros C. Karakousis (Johns Hopkins University) [41].M. avium subsp.hominissuis strain 11 originates from the bone marrow of an AIDS patient suffering from a disseminated infection caused by M. avium [22,42].The isolate was provided by Jung-Yien Chien and Po-Ren Hsueh, National Taiwan University Hospital, Taipei.

Growth Inhibition Assay
Growth inhibition assays were performed in 96-well plate format.The wells were filled with 100 µL of supplemented 7H9 medium before dispending 10 mM compound stock solutions in DMSO into the wells using a Tecan D300e digital dispenser.For each compound, a 10-point 3-fold dilution series or a 10-point 2-fold dilution series was prepared that typically started at a concentration of 100 µM.The DMSO concentrations were normalized to 2%.The cultures of the respective bacterial strains were grown to the mid-log phase, which was indicated by a measured OD 600 between 0.4 and 0.6.A sufficient aliquot of the culture was taken from the culture and diluted to an OD 600 of 0.1 with fresh 7H9 medium (1 × 10 7 CFU/mL).Next, 100 µL of the resulting bacterial suspension was used to inoculate the prefilled wells, which resulted in a total volume per well of 200 µL with an OD 600 of 0.05 (5 × 10 6 CFU/mL, 1% DMSO).Each plate included 8 untreated wells containing 1% DMSO and 8 sterile wells for blank corrections.The plates were sealed with Parafilm ® (Bemis Company, Nennah, WI, USA), wrapped in damp paper towels, and placed in tightclosing plastic boxes, before incubation at 37 • C and shaking at 110 rpm.Fast-growing NTM (M.abscessus and subsp., M. fortuitum and M. chelonae) were incubated for 3 days as a standard procedure, while the slow-growing strains (all the other strains) were incubated for 5 days.Due to its particularly slow growth rate, M. ulcerans was incubated for 10 days.

Determination of MIC Values
To determine the minimal inhibitory concentration at 50% growth inhibition relative to an untreated control (MIC 50 ), OD 600 values of each well were measured with a Tecan Infinite M200 plate reader on day 0 and day 3, day 5 or day 10.Before measuring the OD on the final day of analysis, the sedimented bacterial cells were resuspended with either manual pipetting or with the use of an Eppendorf epMotion 5070 pipetting robot.On day 0 and the final day of analysis, the average OD of the sterile wells was subtracted from the remaining wells for blank correction.To generate the bacterial growth values for every well, the blank-corrected day 0 values were subtracted from the blank-corrected day 3/5/10 values.For each compound, the growth values of the two corresponding untreated wells gave the average drug-free growth, which is equal to 100% growth/0% inhibition.To calculate the % growth of each drug-containing well, their growth values were related to their respective drug-free growth values.GraphPad Prism 10.0 was used for graphical analysis, curve fitting and calculations.Dose-response curves were plotted with % inhibition (=100% growth) versus compound concentration.For the calculation of MIC values, the obtained data points were fitted utilizing a standard variable slope Hill function (bottom asymptote value constrained to equal 0).The resulting function was used to calculate MIC 50 .The values were calculated from two technical replicates and averaged for each compound.

Protein-Protein Primary Structure Alignment
To compare the primary structures of the RNAP β and β ′ subunits of different NTM, we utilized the protein-protein BLAST algorithm provided by NIH (https://blast.ncbi.nlm.nih.gov/Blast.cgi(accessed on 4 April 2024).The primary structures of the type strains were obtained from the Pathosystems Resource Integration Center (PATRIC) database, which is provided by the Bacterial and Viral Bioinformatics Resource Center (BV-BRC, University of Chicago, https://www.bv-brc.org/(accessed on 4 April 2024)).Additionally, we retrieved the genome of MAC109 from Matern et al. [41].The comparative genomics analysis excluded M. abscessus subsp.massiliense, M. abscessus subsp.bolletii, M. abscessus clinical isolates and M. avium subsp.hominissuis strain 11.

Visualization of Protein Models
For surface and interaction analysis as well as visualization of the Mtb RNAP 3D structure (PDB: 5UHE [29]), we utilized UCSF ChimeraX (Resource for Biocomputing, Visualization, and Informatics at the University of California, San Francisco, CA, USA) [43], as

Scheme 1 .
Scheme 1.Chemical structures and identifiers of the AAPs evaluated against the test set of mycobacteria.Scheme 1.Chemical structures and identifiers of the AAPs evaluated against the test set of mycobacteria.

Figure 1 .
Figure 1.Exemplary dose-response curves of Clarithromycin and 7 against Mycobacterium absc subsp.abscessus ATCC 19977.Green data points are average values of two technical replicates error bars display the respective standard deviation.The basis of MIC50 determination was a c fit performed with GraphPad Prism 10.0 utilizing a standard variable slope Hill function (bo asymptote constrained to equal 0).

Figure 1 .
Figure 1.Exemplary dose-response curves of Clarithromycin and 7 against Mycobacterium abscessus subsp.abscessus ATCC 19977.Green data points are average values of two technical replicates.Red error bars display the respective standard deviation.The basis of MIC 50 determination was a curve fit performed with GraphPad Prism 10.0 utilizing a standard variable slope Hill function (bottom asymptote constrained to equal 0).

Figure 2 .
Figure 2. D-AAP1 target interactions based on PDB: 5UHE.(A) Two-dimensional summary of interactions.Purple arrows: hydrogen bonds; green lines: pi stacking interaction.(B) Threedimensional depiction of the binding pose.Yellow dashed lines: hydrogen bonds, blue dashed line: pi stacking interaction; turquoise dashed line: aromatic hydrogen bonds.Visualization generated

Figure 3 .
Figure 3. Variations in β and β′ RNAP subunits of mycobacterial strains compared to Mtb (sequence numbering refers to Mtb β and β′ subunits derived from PDB: 5UHE).Violet indicators show positions with direct drug-target contacts.Yellow indicators show positions that resulted in resistant M. abscessus Bamboo mutants (P477L; L560P).Dots represent amino acid identity.Green indicates above-average activity of the respective compound, while red indicates below-average activity.a: Published MIC50 value for M. tuberculosis ATCC 25618 [22].

Figure 3 .
Figure 3. Variations in β and β ′ RNAP subunits of mycobacterial strains compared to Mtb (sequence numbering refers to Mtb β and β ′ subunits derived from PDB: 5UHE).Violet indicators show positions with direct drug-target contacts.Yellow indicators show positions that resulted in resistant M. abscessus Bamboo mutants (P477L; L560P).Dots represent amino acid identity.Green indicates above-average activity of the respective compound, while red indicates below-average activity.a: Published MIC 50 value for M. tuberculosis ATCC 25618 [22].

Table 1 .
Overview of previously published MIC 90 values for the compound test set utilized in this study.Shown values were calculated from two technical replicates.For detailed information regarding the determination of these values, see the respective reference.

Table 2 .
MIC 50 values of a selection of AAPs against type strains of the Mycobacterium abscessus complex.Cell shading from green to red indicates high to low activity.The displayed values are average

Table 3 .
MIC50 values of a selection of AAPs against a panel of Mycobacterium abscessus com clinical isolates.Cell shading from green to red indicates high to low activity.The displayed v are average values of two technical replicates.For detailed information on the origin of cl isolates, see Section 3.M.

Table 3 .
MIC 50 values of a selection of AAPs against a panel of Mycobacterium abscessus complex clinical isolates.Cell shading from green to red indicates high to low activity.The displayed values are average values of two technical replicates.For detailed information on the origin of clinical isolates, see Section 3.

Table 4 .
MIC 50 values of a selection of AAPs against type strains of the Mycobacterium avium complex.Cell shading from green to red indicates high to low activity.The displayed values are average values of two technical replicates.

Table 5 .
MIC 50 values of a selection of AAPs against type strains of the different NTM.Cell shading from green to red indicates high to low activity.The displayed values are average values of two technical replicates.