N-terminal proteomics of Mycobacterium marinum using bottom-up label-free quantitative analysis in data-dependent acquisition mode on a timsTOF Pro mass spectrometer

ABSTRACT N-terminal acetylation in Mycobacterium tuberculosis is correlated with pathogenic activity. We used genomics and bottom-up proteomics to identify protein Emp1 as the sole acetyltransferase responsible for acetylation of EsxA, a known virulence factor. Using custom data analysis, we screened the proteome to identify 22 additional putative substrates of Emp1.

parameters were set to default for proteomic studies with the following differences: quadrupole low-mass set to 20 m/z, pre-pulse storage to 5 μs.
Peptide identification and label-free quantitation were performed using PEAKS:Online Xpro (build 1.4.2020-10-21_171258,Bioinformatics Solutions Inc.).Results were filtered to a false discovery rate of <1%.The database used was an M. marinum M strain FASTA file from Mycobrowser (v.4).Carbamidomethylation of cysteines was set as a fixed modification.Variable modifications were protein N-terminal acetylation, deamidation of asparagine and glutamine, oxidation of methionine, and pyroglutamate formation on glutamine and glutamatic acid.Quantitation was normalized to total ion current.All other search parameters were default.
Custom scripting was used for analysis, written in R (v.4.3.0)(13) using Rstudio (build 463) (14).The code can be found on Zenodo (https://doi.org/10.5281/zenodo.10635943).The bio-replicate with the highest number of peptide identifications were analyzed.Briefly, technical replicates within bio-replicates were median normalized to the most intense injection.Filtering was performed to remove peptides with three or more missing values between the six injections of wild-type (WT) and Δemp1/comp.The Δemp1 strain was not considered in this filtering step because it was hypothesized to contain missing acetylated peptides.N-terminally acetylated peptides were identified, and technical replicates were averaged.Peptidoforms from the same protein were collapsed using weighted averages by peptide area.k-Means clustering (k = 2) was performed to cluster each protein, with the inputs to the clustering being the pairwise ratios of N-terminally acetylated peptide area between all strains.Analysis of variance and Tukey's statistical tests were performed within the two protein clusters (EsxA-like and other) between the three strains (WT, Δemp1, and Δemp1/comp) using the mediannormalized, log 10 -transformed N-terminal-acetylated peptide areas.Missing values from the proteins of interest were imputed using a normal distribution around the limit of quantification, which was determined using the average and standard deviation of the minimum normalized peptide area for each injection.
The final result of these analyses was a list of proteins that showed N-terminal acetylation patterns similar to EsxA across the WT, Δemp1, and Δemp1/comp strains as shown in Fig. 1.The cluster of proteins that included EsxA shows significantly lower acetylation in the Δemp1 strain compared to WT and Δemp1/comp, whereas the other cluster shows no significant differences.Overall, 41 proteins showed measurable N-terminal acetylation levels in the WT and Δemp1/comp strains, 22 of which clustered with EsxA, indicating that they are likely substrates of Emp1.These proteins are listed in Table 1.

FIG 1
FIG 1 Proteins with measured N-terminally acetylated peptides across three samples and between the two clusters.****P < 0.0001 by Tukey's honestly significant difference test.ns, not significant.
Research reported in this publication was supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under award number R01AI106872 to P.A.C. S.D.W. was supported through the Chemistry-Biochemistry-Biology Interface program at the University of Notre Dame from the National Institute of General Medicine of the National Institutes of Health under award number T32GM075762 and is a Berry Family Foundation Fellow through the Berthiaume Institute for Precision Health.The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

TABLE 1
(12)s and accessions of proteins in cluster containing EsxA a N-terminal acetylation intensity is reduced in the ∆emp1 sample compared to WT and ∆emp1/comp.This table is excised from a larger supplemental information from Collars et al.(12). a