Quantitative proteome analysis of an antibiotic resistant Escherichia coli exposed to tetracycline reveals multiple affected metabolic and peptidoglycan processes

Quantitative proteome analysis of an antibiotic resistant Escherichia coli exposed to tetracycline reveals multiple affected metabolic and peptidoglycan processes

    Quantitative proteome analysis of an antibiotic resistant Escherichia coli exposed to tetracycline reveals multiple affected metaboli...

709KB Sizes 5 Downloads 25 Views

    Quantitative proteome analysis of an antibiotic resistant Escherichia coli exposed to tetracycline reveals multiple affected metabolic and peptidoglycan processes Daniela Jones-Dias, Ana Sofia Carvalho, Inˆes Barata Moura, Vera Manageiro, Gilberto Igrejas, Manuela Canic¸a, Rune Matthiesen PII: DOI: Reference:

S1874-3919(16)30545-0 doi:10.1016/j.jprot.2016.12.017 JPROT 2746

To appear in:

Journal of Proteomics

Received date: Revised date: Accepted date:

10 August 2016 20 December 2016 27 December 2016

Please cite this article as: Jones-Dias Daniela, Carvalho Ana Sofia, Moura Inˆes Barata, Manageiro Vera, Igrejas Gilberto, Cani¸ca Manuela, Matthiesen Rune, Quantitative proteome analysis of an antibiotic resistant Escherichia coli exposed to tetracycline reveals multiple affected metabolic and peptidoglycan processes, Journal of Proteomics (2016), doi:10.1016/j.jprot.2016.12.017

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Quantitative proteome analysis of an antibiotic resistant Escherichia coli exposed to tetracycline reveals multiple affected metabolic and peptidoglycan processes

T

Daniela Jones-Dias1,2*, Ana Sofia Carvalho3*¥, Inês Barata Moura1,2, Vera Manageiro1,2, Gilberto

¥

SC R

*These authors contributed equally to this work.

IP

Igrejas4,5, Manuela Caniça1, Rune Matthiesen3¥

NU

Current address: Computational and Experimental Biology Group, CHRONIC DISEASES

MA

RESEARCH CENTER, NOVA Medical School - Universidade Nova de Lisboa, Lisboa, Portugal

1

National Reference Laboratory of Antimicrobial Resistances, Department of Infectious Diseases, National Institute of Health Doutor Ricardo Jorge, Lisbon, Portugal; 2

TE

D

Centre for the Studies of Animal Science, Institute of Agrarian and Agri-Food Sciences and Technologies, Oporto University, Oporto, Portugal; 3

CE P

Computational and Experimental Biology Group, Department of Health Promotion and Chronic Diseases, National Institute of Health Doutor Ricardo Jorge, Lisbon, Portugal; 4

5

AC

Functional Genomics and Proteomics Unit, Department of Genetic and Biotechnology, University of Trás-os-Montes and Alto Douro, Vila Real, Portugal; UCIBIO-REQUIMTE, Faculty of Science and Technology, New University of Lisbon, Monte da Caparica, Portugal. Key words: Escherichia coli, proteome, tetracycline, antibiotic resistance, metabolism

Correspondence: Manuela Caniça National Institute of Health Doutor Ricardo Jorge Av. Padre Cruz | 1649-016 Lisboa | Portugal [email protected] Tel: (+351) 217 519 246

ACCEPTED MANUSCRIPT Abstract

Tetracyclines are among the most commonly used antibiotics administrated to farm animals

T

for disease treatment and prevention, contributing to the worldwide increase in antibiotic

IP

resistance in animal and human pathogens. Although tetracycline mechanisms of resistance are well known, the role of metabolism in bacterial reaction to antibiotic stress is still an

SC R

important assignment and could contribute to the understanding of tetracycline related stress response. In this study, spectral counts-based label free quantitative proteomics has been applied to study the response to tetracycline of the environmental-borne Escherichia coli

NU

EcAmb278 isolate soluble proteome. A total of 1,484 proteins were identified by high resolution mass spectrometry at a false discovery rate threshold of 1%, of which 108 were uniquely identified under absence of tetracycline whereas 126 were uniquely identified in

MA

presence of tetracycline. These proteins revealed interesting difference in e.g. proteins involved in peptidoglycan-based cell wall proteins and energy metabolism. Upon treatment, 12 proteins were differentially regulated showing more than 2-fold change and p<0.05 (p value

D

corrected for multiple testing). This integrated study using high resolution mass spectrometry

TE

based label-free quantitative proteomics to study tetracycline antibiotic response in the

AC

CE P

soluble proteome of resistant E. coli provides novel insight into tetracycline related stress.

ACCEPTED MANUSCRIPT 1. Introduction

Infections due to antibiotic resistant pathogens constitute a major public health concern,

T

frequently leading to high levels of morbidity, mortality and healthcare costs [1, 2].

IP

Tetracyclines are a group of broad-spectrum antibiotic agents that exhibit antibiotic activity against a wide range of microorganisms. Their cost, favorable antimicrobial properties, and

SC R

absence of major adverse side effects have led to their extensive use in human and veterinary medicine, as well as in agriculture [3]. Tetracyclines sterically block aminoacyl-tRNA binding within the bacterial ribosome, inhibiting protein synthesis. However, this association of the

NU

antibiotics with the ribosomes is reversible, hence explaining their bacteriostatic effect [4]. Resistance to tetracyclines may be transfered or arise by chromosomal mutations, and occurs through five main mechanisms: 1) production of ribosomal protection proteins (RPPs), 2)

MA

active efflux of tetracycline from the cell, 3) enzymatic inactivation of the antibiotic, 4) decreased drug permeability 5) and mutation of the antibiotic target [5]. There are currently over 40 different acquired tetracycline resistance determinants recognized: 46 tet (tetracycline

D

resistance) genes and 3 otr (oxytetracycline resistance) genes [6, 7].

TE

Considering the existing resistance to tetracycline, new generation tetracycline-based antibiotics are now starting to be developed which shows potential for the treatment of

CE P

serious multidrug resistant Gram negative infections, which highlights the importance of preserving this class of antibiotics [8, 9]. Moreover, it has been reported that tetracyclines induce the expression of multidrug resistance efflux pumps, ribosomal proteins, and iron

[10].

AC

uptake transporters, which may favor the emergence of resistance to other antibiotic classes

Additionally, there are emerging concerns on the ability of human actions to enhance the selection pressure caused by antibiotics, which may increase the mobility of antibiotic resistance genes, and their recruitment by clinically relevant pathogens [11]. Previous studies have been able to show that agriculture soils may become contaminated with clinically important antibiotics, antibiotic resistant isolates and antibiotic resistance genes through the application of manure originated from antibiotic treated livestock [12]. Moreover, specific practices, such as intensive agriculture, are known to provide major impacts on the selection of environmental-borne resistant genes, as they may supply a selective pressure, either by direct application of antibiotics on crops (such as tetracycline, flumequine or streptomycin) or by indirect exposure through manure or wastewater amendments [13]. Thus, the combination of soil contamination with the use of tetracyclines in agriculture may be creating the ideal

ACCEPTED MANUSCRIPT conditions to favor the selection and emergence of clinically important mobile and non-mobile antibiotic resistant mechanisms. The role of metabolism in the bacterial response to antibiotics has recently gained interest due

T

to the lack of novel targets for multidrug resistant microorganisms. Many studies have already

IP

focused on the evaluation of variations in protein expression caused by antibiotic exposure in tetracycline resistant microorganisms from different species [10, 14, 15]. Furthermore, these

SC R

proteomic studies have mainly targeted the membrane proteome and to date there are no quantitative tetracycline resistance studies analyzed by high resolution mass spectrometry allowing high mass accuracy in both MS and MS/MS scans. Additionally, little is known about

NU

the metabolic response of genetically resistant bacterial populations to antibiotic exposure [16].

Alterations occuring within antibiotic resistance soil bacteria, to overcome antibiotic stress

MA

conditions are reflected in the proteome [5, 17]. The use of proteomic profiling to identify and quantify molecules that might be directly or indirectly related with the response to antibiotic exposure represents an important step in determining the metabolic pathways that might be

D

associated with antimicrobial activity. By highlighting pathways involved in the acquisition of

TE

resistance, and which may themselves represent new drug targets, these approaches may be

drugs [17].

CE P

helpful not only to extend the usefulness of current antimicrobials but also to develop new

In this study, we have used a liquid chromatography-mass spectrometry-based (LC-MS/MS) proteomics approach to evaluate the global metabolic changes in the soluble protein fraction

AC

of an antibiotic resistant Escherichia coli isolate, when challenged with tetracycline. The MS analysis was performed using high mass resolution in MS and MS/MS scans and label free quantitation using three biological replicates for both control and treated cells.

ACCEPTED MANUSCRIPT 2. Materials and Methods

2.1. Characterization of bacterial isolate

T

Isolate E. coli EcAmb278 was collected in July of 2012, from a soil sample aseptically recovered

IP

from an agricultural setting near Almeirim, Portugal, used for the intensive farming of tomato

SC R

plants, as previously described [18].

The nonsusceptibility profile of the isolate, performed according to the Antibiogram Committee of the French Society of Microbiology (SFM) [19], comprised penicillins, first, second, third and fourth generation cephalosporins, monobactam, β-lactam/β-lactamase

NU

inhibitors and tetracycline. Minimum Inhibitory Concentration (MIC) was determined for tetracycline, as suggested by SFM, resulting in 256 mg/L. The isolate remained susceptible to

MA

carbapenems, quinolones, aminoglycosides, phenicols, sulphonamides and nitrofurans. Previous genomic characterization, performed as described [20], confirmed the presence of five antibiotic resistance genes encoding the penicillin TEM-135, the extended-spectrum β-

D

lactamase CTX-M-1, the dihydropteroate synthase Sul2 and the efflux pumps TetA and TetB.

TE

Furthermore, the E. coli EcAmb278 harbored four virulence factors (Gad, Iss, IroN, Cma) and displayed a 93.1% probability of acting as a human pathogen, according to Pathogen Finder

CE P

[21]. The multilocus sequence typing (MLST) [22] method defined the isolate as a ST1718. Serotype determination, performed upon the analysis of FliC-, Wzt- and Wzm-encoding genes,

AC

showed an E. coli from serotype O9:H31 [23].

2.2. Bacteria and culture conditions Frozen glycerol stocks of E. coli EcAmb278 were seeded onto MacConkey agar and trypticase soy agar plates sequentially, and grown for 18h, at 37ºC. Single colonies of the E. coli EcAmb278 were pre-inoculated in 5ml of Brain Heart infusion broth that was then used to inoculate 250ml of the same medium, in covered 500ml Erlenmeyer flasks at 37 °C, with rotary aeration at 180 rpm. For three out of the six experiments, tetracycline (Sigma-Aldrich, St. Louis, MO) was added to the culture in a final concentration of 10 mg/L, which was tested and confirmed as a sub inhibitory concentration prior to cell growth. The three remaining replicates were cultured without exposure to antibiotics. The cells were then harvested during the exponential phase (OD640nm= 0.6, approximately 2 × 109cells/ml).

ACCEPTED MANUSCRIPT 2.3. Extraction of soluble proteins and protein quantification Proteins were isolated using a protein extraction method for soluble proteins. Briefly, the bacterial cells were harvested at 10,000 g for 5 min at 4 °C. Three biological replicates of three

T

independent assays were performed with and without antibiotic. The pellet was then

IP

completely resuspended in MES [2-(N-morpholino)ethanesulfonic acid)] - NaOH 20mM pH 8.0

SC R

(Sigma-Aldrich, St. Louis, MO) and centrifuged three times using the conditions above mentioned. After centrifugation, bacterial pellets were resuspended in 10ml for each 2.5g of cells in MES-NaOH 20mM pH 8.0, DTT 1mM, with addition of bacterial protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO), disrupted by ultrasonic treatment as previously

NU

described [24], and stored at −20 °C until further analyses.

Protein concentration was

estimated using BCA protein assay kit (Pierce, Rockford, IL), according to the manufacturer’s

D

2.4. One-dimensional electrophoresis

MA

instructions.

TE

The integrity and reproducibility of the bacterial proteome was estimated by using sodium dodecyl sulfate-polyacrylamide gel electrophoresis, as previously described [24]. Furthermore,

CE P

one-dimensional SDS-PAGE together with Bradford quantitation was used to assure equal protein amount for MS sample preparation. No major protein band differences were observed in the SDS-PAGE. Briefly, approximately 15μg of protein extract was collected and resuspended

AC

in an equal volume of sample buffer containing Tris-HCl pH 6.8, glycerol, SDS (dodecyl sulphate sodium), DTT (1,4-Dithiothreitol) and bromophenol blue [4.5% (p/v) Tris-HCl (Sigma-Aldrich, St. Louis, MO), 10% glycerol (Merck KGaA, Darmstadt, Germany), 6% (p/v) SDS (Sigma-Aldrich, St. Louis, MO), 3% (p/v) DTT (BioRad, München, Germany), and 2% (p/v) bromophenol blue (Merck, Darmstadt, Germany); pH was adjusted to pH 6.8 by adding the concentrated HCl].” One-dimensional gel electrophoresis was then conducted on vertical SDS-polyacrylamide gels with final acrylamide concentrations of 12 and 5% (wt/vol) for the separating and the stacking gels, respectively, in an ENDURO VE20 Vertical Gel Electrophoresis System (Labnet International, Edison, NJ), as previously described [25]. Proteins were separated with a constant current of 125V until the dye-front reached the bottom of the gel and stained with Coomassie Blue R-250 (Sigma-Aldrich, St. Louis, MO).

ACCEPTED MANUSCRIPT 2.5. Peptide Sample Preparation Protein solution containing SDS and DTT were loaded into filtering columns and washed exhaustively with 8M urea in HEPES buffer. Proteins were then incubated for 16h with trypsin

T

sequencing grade (Promega, Madison, WI) after alkylation with iodoacetamide (GE Healthcare

IP

Europe GmbH, Lisbon, Portugal) and reduction with DTT (BioRad, München, Germany).

SC R

2.6. Mass Spectrometry

Peptides generated as described above were desalted and concentrated [26] prior to analysis by nano LC-MS/MS using a Q-Exactive (Thermo, San Jose, CA) mass spectrometer coupled to a

NU

Dionex NCP3200RS HPLC setup (Thermo, Sunnyvale, CA). A reversed phase column ReproSilPur 120 Si, 3 µm, 12 cm length (Nikkyo Technos Co., Ltd., Japan) was used to separate

MA

peptides. The analytical gradient, using a flow rate of 200 nL/min, was increasing from 5% Buffer B (0.1% formic acid in acetonitrile)/ 95% Buffer A (0.1% formic acid in H2O) to 35% Buffer B/ 65% Buffer A over 110 min followed by an increase to 90% Buffer B/ 10% Buffer A

D

during 10 min. MS survey scans were acquired from m/z 350 to m/z 1400 at 70,000 resolution

TE

(AGC: 1e6 and Maximum IT: 120 ms). An upper limit of 20 most abundant ions was subjected to MS/MS and measured at a resolution of 35,000 (AGC: 5e4 and Maximum IT: 120 ms) with

CE P

lowest mass set to m/z 100.

AC

2.7. Preprocessing of MS data

Q-Exactive data was calibrated using polycyclodi-methylsiloxane (PCMs—out gassed material from semiconductors) present in the ambient air and Bis (2-Ethylhexyl)(Phthalate) (DEHP— from plastic) [27, 28] modular VEMS [29]. VEMS further allows alternative parent ion annotations for each MS/MS spectrum which is needed if two peptide elution profiles overlap in the m/z and retention time dimension. These alternative parent ion annotations were taken into account during the database dependent search.

2.8. MS database dependent search All data were searched with VEMS [26, 30]. Mass accuracy was set to 5 ppm for peptides and 10 mDa for peptide fragments. Gaussian weight for fragment ions was set to 5 and the six

ACCEPTED MANUSCRIPT most intense fragment ions per 100 Da was used for scoring fragment ions. Four missed cleavages were specified. The data was first searched against two databases: all bacterial protein sequences in NCBI and all bacterial protein sequences in UniProtKB/TrEMBL (Release

T

2015_02). These searches confirmed that the peptide spectra assignments were mainly

IP

matching bacterial proteins from E. coli. The search was therefore repeated only against E. coli proteins in UniProtKB (Figure S1). The three databases included permutated protein

SC R

sequences, leaving Lys and Arg in place, together with common contaminants such as human keratins and proteases [31].

Fixed modification of carbamidomethyl cysteine was included in the search parameters. A list

NU

of 5 variable modifications (Figure S2) was considered for all data against the full protein database. Protein N-terminal Met-loss is not specified for VEMS searches since VEMS by default checks N-terminal Met-loss. The false discovery rate (FDR) for protein identification

MA

was set to 1% for peptide and protein identifications. No restriction was applied for minimal peptide length. Identified proteins were divided into evidence groups as defined by

TE

D

Matthiesen et al. [32].

CE P

2.9. Quantitative proteome analysis

Proteins were quantified by spectral counting [33] and mziXIC [29] followed by intensity-based absolute quantification (iBAQ) [34] estimation. For statistical comparison of regulated proteins between control and tetracycline treated spectral counting data was used. No imputation for

AC

missing values was used. This leads to less significant proteins than if imputation was performed, however, this raises the confidence on the proteins identified as statistical significant. The quantitative spectral count values were added one and log two transformed. The quantitative values were next normalized using quantile normalization and statistical calculation of p values was performed by the R package limma (Figure S3). Correction for multiple testing was done by the required false discovery rate (FDR) method [35].

2.10.

Validation of significant regulated proteins

To further validate the significant regulated proteins based on spectral counting all peptide spectra assignments using raw data were depicted together with delta mass plots and subsequently manually scrutinized (See Figure S4). Validation of spectral counts was made by

ACCEPTED MANUSCRIPT automatically extracting the ion counts using a mass accuracy of 0.005 Da and maximum allowed deviation from expected isotope distribution of 10%. The quantitation obtained by ion counts are summarized in Figure S5 and confirms all the quantitative values obtained by

Functional analysis of proteins

SC R

2.11.

IP

T

spectral counting on the significant regulated proteins.

Functional enrichment analysis was performed using DAVID Bioinformatics server [36]. All identified proteins, proteins unique for control and for tetracycline treated were submitted to

NU

DAVID using E. coli proteome as background. Gene ontology tables with the functional enrichment results for biological processes (BP), cellular component (CC) and molecular

MA

function (MF) were manually downloaded for the above mentioned three sets of proteins. The tables were sorted on based on FDR corrected p value. The tables were filtered by starting with the most significant enriched functional category and then eliminating any subsequent

D

functional category with more than 80% protein overlap with any of more significant enriched

TE

functional categories. We find that 80% is good trade off to eliminate redundant functional categories. Finally, the ten most significant functional categories were depicted using barplot

barplots.

Proteome Xchange Accession Numbers

AC

2.12.

CE P

in the statistical programming language R [37]. FDR corrected p values were used for the

The raw MS data obtained from the proteome of isolate EcAmb278 has been deposited at Proteome Xchange under the accession PDX00000. (available when manuscript is accepted and subsequent paragraph will be deleted).Project Name: E. coli EcAmb278 treated with tetracycline, Project accession: PXD003641, Project DOI: Not applicable, Reviewer account details: Username: [email protected], Password: 0s9dIx5t

ACCEPTED MANUSCRIPT 3. Results and Discussion

3.1. Overall protein identifications in the soluble fractions of environmental-borne E. coli

T

EcAmb278

IP

We have previously reported the presence of higher levels of antibiotic resistant isolates on a

SC R

collection of non-susceptible Gram negative bacteria recovered from sites of different agricultural practices (see also section 2.1 for more details on EcAmb278) [18]. E. coli EcAmb278 isolate (Gram negative) was recovered from soils of intensive agriculture. Up to now MS-based proteomics analysis of tetracycline resistance has targeted mainly pathogens

NU

membrane components. Several studies point to the necessity for further investigation of the role of cytosolic metabolic pathways in drug resistance [5, 38]. Previously use of cumbersome

MA

fractionation and sample preparation methods are less relevant with the use of newly accurate and sensitive instruments allied with powerful bioinformatics tools. The use of newly accurate and sensitive instruments allied with powerful bioinformatics tools set the ground for major

D

contribution of mass spectrometry into antibiotic drug resistance. We present here the first in

TE

depth analysis of the soluble proteome of a nonsusceptible E. coli with the highest proteome coverage using high resolution mass spectrometry. We have analyzed the soluble proteome of

CE P

E. coli EcAmb278 itself and challenged with 10 mg/L tetracycline followed by MS-based label free quantitation. The MS data was acquired with high accuracy in both MS and MSMS scans. In both conditions analyzed (in the presence and absence of tetracycline) a total number of 1,484 proteins were identified, using a 1% FDR as cut off and collapsing the proteins to

AC

encoding genes to remove bacterial strain redundancy. The UniProt annotation, quantitative and statistical result on the identified proteins is available in Table S1 both before and after collapsing the proteins into gene encoding proteins. The UniProt bioinformatics database (http://www.uniprot.org/)

and

the

ExPASy

SIB

Bioinformatics

Resource

Portal

(http://www.expasy.org/), complemented with an in-depth literature search, were used to annotate protein's functions and its association to specific biological processes at cellular level. A total of 37% (547 proteins out of 1,484) proteins identified are annotated to a cellular component (Figure 1A). The most significantly enriched component is cytosol (GO:0005829), which reflects the targeted soluble protein fraction described in the method section. However, we do also, to a less extent, identify proteins associated to periplasmic space and membranes (Figure 1A). Ribonucleoprotein complex members such as the ribosomal protein family are significantly represented as bacteria do not have a distinct nucleus that separates DNA from ribosomes, so there is no barrier to immediate translation. The ribosome is an important

ACCEPTED MANUSCRIPT target for a wide variety of antibiotics. Tetracycline blocks the binding of aminoacylated tRNA (aa‐tRNA) to the A‐site of the 30S subunit, inhibiting protein synthesis [39, 40]. GO categories translation (GO:0006412) in the GO annotation category biological process (Figure 1B), and

T

structural constituent of ribosome (GO:0003735) in molecular function category are the

IP

primary significantly enriched for identified proteins (Figure 1C). Generation of precursor

SC R

metabolites and energy (GO:0006091) was also significantly identified.

3.2. Comparison of identified proteins in controls versus tetracycline treated E. coli

NU

EcAmb278

The comparison of identified proteins in absence and presence of tetracycline is depicted in Figure 2. A total of 1,484 proteins were identified of which 108 were uniquely identified under

MA

absence of tetracycline whereas 126 were uniquely identified in presence of tetracycline. We observe enrichment of e.g. “aerobic respiration”, “phosphate transport” and “ATPase

D

activity” among proteins identified uniquely in the absence of tetracycline (Figures S6 and S7). These findings are in accordance with metabolomics studies demonstrating that antibiotic

TE

growth inhibition is associated with down regulation of bacterial cellular respiration [41]. In addition, our results also provided support for the hypothesis that changes in energy

CE P

metabolism could be an important cell survival response upon antibiotics [41, 42]. Proteins identified uniquely in the presence of tetracycline (see Figure S7) were enriched for proteins involved in catabolic processes. We speculate that tetracycline’s effect on ribosome

AC

translation cause damage requiring repair from catabolic processes. Furthermore, proteins unique for E. coli in the presence of tetracycline were enriched in gene ontology category manganese ion binding (Figure S7C). A set of peptidoglycan-based cell wall proteins were unique for control and a distinct set for tetracycline treated E. coli were observed (Figures S6A and S7A). These results suggest that these alterations can potentially occur by shifting to another set of proteins involved in peptidoglycan-based cell wall rather than just upregulating additional peptidoglycan-based cell wall proteins (see table S1, sheet “peptidoglycan-based cell wall”). For example, the tyrosineprotein kinase (etk) was only identified in the presence of tetracycline. Etk and Wzc are essential for synthesis of specific extracellular polysaccharide [43]. We have also explored the presence of the post-translational modification acetylation (Figure S2 and S8). Post-transcriptional and post-translational processes generate a remarkable diversity of mature proteins from a single gene that can influence the global metabolic and

ACCEPTED MANUSCRIPT evolutionary responses of a cell [44]. In this study, lysine acetylation appeared to be more frequent in the tetracycline treated EcAmb278 isolate although not significantly different. Although global lysine acetylation levels between the two conditions was not statistically

T

different, a target approach using immunoaffinity enrichment of lysine acetylated peptides

IP

upstream of MS analysis could evidence these observations. The physiological role of lysine

SC R

acetylation in bacteria is still unknown however several studies showed that the bacterial acetylated proteins are especially well represented within the carbohydrate and energy metabolism processes [45, 46, 47]. Lysine acetylation shields the positive charge of lysines leading to increased affinity to DNA thereby regulating transcription of DNA in eukaryotes [48].

NU

In other words the changes in lysine acetylation in E coli potentially regulate transcriptional activity. Lysine acetylation might also regulate protein stability. In eukaryotes lysine acetylation

MA

seems to regulate protein stability through cross talk with the ubiquitin system. Furthermore, nicotinamide increases acetylation of nuclear proteins [49]. Although, E coli does not possess the ubiquitin system [50] lysine acetylation might regulate or influence protein stability

D

through other yet to be discovered pathways. The role of PTMs in prokaryotes still needs more

CE P

control conditions.

TE

scientific attention. N-terminal protein acetylation was less frequently detected, compared to

3.3. Significantly differentially regulated E. coli proteins under tetracycline treatment

AC

Comparison between the proteome profiles of two E. coli EcAmb278 samples conditions revealed altered expression of several proteins under tetracycline stress. Among those, the twelve most significant proteins (FDR-corrected p≤0.05), differentially regulated by more than two-fold between the tested conditions, were further analyzed, and are shown in Table 1 and Figure 3. The identified proteins, which were either up-regulated or down-regulated in the tetracycline exposed strain, compared to the non-exposed control, are involved in the following biological functions: DNA replication/repair (Q8XBL5 and Q8FH89), transcription (P64624 and B7UQK0), virulence (P61887 and P08191) intracellular trafficking and secretion (P20966), biosynthesis of vitamins (P0AG42), other biosynthesis processes (P0A8Z1 and Q46925), other processes (P22256), and unknown function (P0AD11) (Figure 3). The low number of significantly regulated proteins could be explained by the fact that nearly 26% of the most abundant proteins had functions related to protein or amino acid biosynthesis. In fact, since regulation was considerably low for these protein categories it is

ACCEPTED MANUSCRIPT hard to define the role of regulation in these protein categories: they might not be directly relevant for survival of E. coli EcAmb278 upon tetracycline treatment, but their function might also be influenced by post-translational modifications contributing to the survival of this strain.

T

The acyl-CoA thioester hydrolase (P0A8Z1) is involved in the biosynthesis of coenzyme A,

IP

which has been implicated in the resistance to aminoglycosides, being responsible for a conformational increase in the binding site of an aminoglycoside modifying-enzyme [51]. The

SC R

fact that the acyl-CoA thioester hydrolase is here significantly up-regulated might suggest its involvement in response to tetracycline stress. This group of antibiotics displays a mode of action that, just like tetracycline, discontinues protein biosynthesis [52].

NU

Bacterial NAD+-dependent DNA ligase (Q8XBL5) plays a critical role in DNA replication, recombination, and repair in all living organisms [53]. Moreover, this protein was recently evaluated as a potential bacterial broad-spectrum drug target using adenosine analogs to

MA

inhibited the LigA activities of E. coli, Haemophilus influenzae, Mycoplasma pneumoniae, Streptococcus pneumoniae, and Staphylococcus aureus [53]. In this study, LigA was the most up-regulated protein, reaching up to 4.7 log2 fold change when compared to the control

D

group, which could suggest a bacterial survival response to the tetracycline-induced stress.

TE

Glucose-1-phosphate thymidylyl transferase (P61887), which in this study was also found to be significantly up-regulated in the treated E. coli, is part of the L-rhamnose biosynthesis pathway

CE P

[54], one of the important residues of O-antigen lipopolysaccharide. The up-regulation of such protein might increase the pathogenicity of this environmental-borne strain when exposed to this antibiotic, as L-rhamnose is a key determinant for virulence and complement cascade

AC

mediated serum killing [54, 55]. The up-regulation of a kinase (P0AG42), which is involved in the riboflavin biosynthesis pathway, is also relevant, considering that tetX encodes an unprecedented flavin-dependent monooxygenase that selectively hydroxylates this class of antibiotics, conferring resistance to all clinically relevant formulas, including tigecycline, a last resource tetracycline [56, 57]. The up-regulation of this protein could suggest a broader involvement of flavins in tetracyclineinduced stress response [56]. The adhesion type 1 fimbriae was 4.2-fold down-regulated in the tetracycline-challenged strain, which has been shown to result in the decrease of motility and adherence [58-60]. Flagellum-mediated motility and chemotaxis have been suggested to contribute to virulence by enabling uropathogenic E. coli to escape host immune responses and disperse to new sites within the urinary tract [58]. Thus, the decrease in motility in antibiotic challenged-isolates seems to decrease the global pathogenicity of the isolate.

ACCEPTED MANUSCRIPT Overall the results indicate that E. coli responses to tetracycline are related to protein translation as well as metabolic regulation. Several metabolic proteins were differentially regulated, indicating a concerted activity of the bacteria to modulate the response against

IP

T

antibiotic exposure using its own metabolism.

SC R

3.4. Identified proteins involved in virulence, antibiotic resistance and transfer of foreign DNA

Overall, considering the environmental origins of E. coli EcAmb278, it is important to highlight

NU

the proteins related with virulence, antibiotic resistance, and transfer of foreign DNA (Table S2). We were able to identify several proteins associated with antibiotic resistance in E. coli EcAmb278, such as β-lactam resistance mechanisms CTX-M β-lactamase (P28585) and TEM

MA

(P62593), confirming previous results obtained and explored, using complementary DNAbased molecular methods [18].

D

The analysis of the iBAQ values showed that CTX-M β-lactamase was the second most abundant protein and the most abundant among the different resistance mechanisms and

TE

virulence factors, showing a value of 25.7 (Table S2). As expected, CTX-M β-lactamase was not found to be significant regulated upon tetracycline treatment. Our findings corroborate recent

CE P

evidence suggesting that the preservation of this resistance mechanism imposes no or negligible fitness costs on E. coli, persisting without appropriate antibiotic selection [61]. The acquired (sul2) and chromosomal (folP) dihydropteroate synthase-encoding genes are

AC

associated with the different levels of resistance to the sulphonamides. FolP and Sul2 showed comparable abundance levels (iBAQ of 21.5 20.3, respectively), despite conferring resistance to the same class of antibiotics [62]. Although acquired versions of FolP have been frequently described, in this study the location of folP gene was not explored. We have also detected three of the proteins involved in the AcrAB-TolC and MdtEF-TolC multidrug efflux systems: AcrA (P0AE06), TolC (P02930) and MdtE (P37636). Indeed, MdtE was among the most abundant proteins according to the calculated iBAQ values in this E. coli, in both conditions tested, which suggests an increased efflux in this strain. However, high-level resistance to tetracycline is normally displayed only when the AcrAB-TolC is associated to an additional tetracycline-specific resistance mechanism or a significant decrease in permeability [63]. Besides TolC, six other porins were identified (P0AA16, P0A917, P77747, P0A915, P0A910, and Q8XE41), which are components of permeability channels that allow passive diffusion of small molecules, like antibiotics, across the outer membrane. It is known that, in E. coli strains,

ACCEPTED MANUSCRIPT the down-regulation or loss of function due to amino acid substitutions in OmpC or OmpF promote resistance to some antibiotics, including tetracyclines [64]. Overall, five other detected proteins have been associated with resistance to rifampicin

T

(P0A903), tellurite (P25397), novobiocin and mecillinam (P0A9F5), and ampicillin (P0AD68,

IP

P10121) when mutated (Table S2). In particular, the down-regulation of BamC (P0A903), a component of the β-barrel assembly machinery complex, responsible for recognition and

SC R

assembly of outer membrane proteins (Omps) [65], has been observed in E. coli isolates resistant to tetracycline and ampicillin, in proteomic studies performed in presence of those antibiotics [15].

NU

The proteins involved in virulence and DNA transfer described in Table S2, highlight the pathogenicity of this environmental-borne isolate. Indeed, we have identified proteins related with the formation of biofilms, as well as the transfer, acquisition and recombination of DNA

MA

elements (P0AF96, Q8FF56, P32885, P03835, P05846, P24218) that are associated with the dissemination of well documented resistance mechanisms, such as ISAba1 (from IS4 family) and tnsE (from class 2 integrons), among others [66, 67].

D

The production of bacteriocins in response to worsening environmental conditions is one

TE

mean of bacteria to outcompete other microorganisms, being effective against closely related Enterobacteriaceae. We detected five of these bacterial protein toxins, including colicin E1,

CE P

which has already been described as a virulence factor in uropathogenic E. coli [68, 69]. In fact, bacteriocin production is an exclusive and important pathogenic feature of E. coli strains of clinical origin, particularly associated with complicated urinary tract infections [68].

AC

In this proteome profiling of EcAmb278 upon tetracycline stress we investigated the soluble cell fraction, so that this new perspective could provide a broaden understanding of the metabolic cell responses of E. coli to a widely used antibiotic.

ACCEPTED MANUSCRIPT 4. Conclusions

Although previous calls have been made to address antibiotic resistance in an environmental

T

perspective [67], it is important to highlight that the metabolic burden on a resistant pathogen

IP

is highly dependent on the bacterial microenvironment and the metabolic adaptations required for colonizing such a habitat [16]. Considering that soils are persistently contaminated

SC R

with resistant strains and that tetracyclines are still deliberately applied in crops, understanding the metabolic processes affected by antibiotic exposure of resistant strains is crucial.

NU

The large scale quantitative proteome comparison of an environmental-borne E. coli challenged with tetracycline with the same strain without antibiotic exposure identified 8 up-

<0.05 and at least 2 fold regulation.

MA

and 4 down-regulated proteins upon tetracycline stress using a corrected p value threshold

Our observed down regulation of proteins in the functional gene ontology categories “aerobic respiration” and “phosphate transport” and “ATPase activity” upon tetracycline treatment

D

confirms previous observations based on metabolomics [41]. These findings also concur with

TE

the findings of Lin and co-workers [70] which demonstrate down regulation of proteins involved in energy metabolism upon chlortetracycline treatment of E. coli. We additionally

CE P

found that down regulation of a specific set of peptidoglycan-based cell wall proteins were replaced by up regulation of another set of peptidoglycan-based cell wall proteins upon tetracycline exposure. Furthermore, proteins unique for E. coli in the presence of tetracycline

AC

were enriched in gene ontology category manganese ion binding. The microbial physiology shows that a single protein can act in a variety of cellular processes, including antibiotic resistance, although this may not be its primary role [67]. Thus, we highlight that antibiotic stress response is not exclusively the result of a single altered protein, but rather a comprehensive and concerted metabolic process, as indicated by the proteome adjustments observed in the nonsusceptible bacteria of this study. In the end, investigating the relationships between bacterial metabolism and antibiotic susceptibility can help to uncover novel strategies for treating infections.

ACCEPTED MANUSCRIPT Conflict of interest statement The authors declare that the present study was conducted in the absence of any commercial

IP

T

or financial relationships that could be construed as a potential conflict

SC R

Acknowledgements

DJ-D has received research funding from Fundação para a Ciência e Tecnologia (FCT) through grant SFRH/BD/80001/2011. ASC and VM are supported by the FCT, financed by the European

NU

Social Funds (COMPETE-FEDER) and national funds of the Portuguese Ministry of Education and Science (POPH-QREN) fellowships SFRH/BPD/85569/2012 and SFRH/BPD/77486/2011, respectively. RM is supported FCT investigator program 2012. This work was supported by FCT

MA

(grant numbers PTDC/CVT/65713 and PEst-OE/AGR/UI0211/2011-2014). We thank Brian D. Dilland Henrik Molina from The Proteomics Resource Center at The Rockefeller University, US for MS analysis of samples. The Proteomics Resource Center at The Rockefeller University

AC

CE P

TE

spectrometer instrumentation.

D

acknowledges funding from the Leona M. and Harry B. Helmsley Charitable Trust for mass

ACCEPTED MANUSCRIPT References

Laxminarayan R, Duse A, Wattal C, Zaidi AK, Wertheim HF, Sumpradit N, Vlieghe E,

T

[1]

IP

Hara, GL, Gould IM, Goossens H, Greko C, So AD, Bigdeli M, Tomson G, Woodhouse W, Ombaka E, Peralta AQ, Qamar FN, Mir F, Kariuki S, Bhutta ZA, Coates A, Bergstrom R,

SC R

Wright GD, Brown ED, Cars O. Antibiotic resistance-the need for global solutions. Lancet Infect Dis 2013; 13:1057-1098. [2]

Giske CG, Monnet DL, Cars O, Carmeli Y, on behalf of ReAct-Action on Antibiotic

NU

Resistance. Clinical and economic impact of common multidrug-resistant gramnegative bacilli. Antimicrob Agents Chemother 2008; 52:813-21. Roberts MC. Tetracycline therapy: update. Clin Infect Dis 2003;4:462-467.

[4]

Chopra I, Roberts M. Tetracycline antibiotics: mode of action, applications, molecular

MA

[3]

biology, and epidemiology of bacterial resistance. Microbiol Mol Biol Rev 2001;

Vranakis I, Goniotakis I, Psaroulaki A, Sandalakis V, Tselentis Y, Gevaert K, Tsiotis G,

TE

[5]

D

65:232-260.

Proteome studies of bacterial antibiotic resistance mechanisms. J Proteomics

[6]

CE P

2014;97:88-99.

Roberts MC, Schwarz S. Tetracycline and phenicol resistance genes and mechanisms: importance for agriculture, the environment, and humans. J Environ Qual

[7]

AC

2016;45:576-92.

Roberts, M. C., Update on acquired tetracycline resistance genes. FEMS Microbiol Lett 2005;245:195-203.

[8]

Trudy G, Clark R, Yi X-X, Corey F, Joyce S. In vitro activities of novel tetracyclines against molecularity characterized carbapenem-resistant Gram negative pathogens. In 25th European Congress of Clinical Microbiology and Infectious Diseases: 2015;P0243.

[9]

Mawabo IK, Noumedem JA, Kuiate JR, Kuete V. Tetracycline improved the efficiency of other antimicrobials against Gram negative multidrug resistant bacteria. J Infect Public Health 2015; 8:226-233.

[10]

Yun SH, Choi CW, Kwon SO, Park GW, Cho K, Kwon KH, Kim JY, Yoo JS, Lee JC, Choi JS, Kim S, Kim SI. Quantitative proteomic analysis of cell wall and plasma membrane

ACCEPTED MANUSCRIPT fractions from multidrug resistant Acinetobacter baumannii. J Proteome Res 2011;10:459-469. [11]

Finley RL, Collignon P, Larsson DG, McEwen SA, Li XZ, Gaze WH, Reid-Smith R,

Hartmann A, Locatelli A, Amoureux L, Depret G, Jolivet C, Gueneau E, Neuwirth C,

SC R

[12]

IP

role of the environment. Clin Infect Dis 2013;57:704-710.

T

Timinouni M, Graham DW, Topp E. The scourge of antibiotic resistance: the important

Occurrence of CTX-M producing Escherichia coli in soils, cattle, and farm environment in France (Burgundy region). Front Microbiol 2012;3:83. [13]

Popowska M, Rzeczycka M, Miernik A, Krawczyk-Balska A, Walsh, F, Duffy B, Influence

NU

of soil use on prevalence of tetracycline, streptomycin, and erythromycin resistance and associated resistance genes. Antimicrob Agents Chemother 2012;56:1434-1443. Vranakis I, De Bock PJ, Papadioti A, Tselentis, Y, Gevaert K, Tsiotis G, Psaroulaki A.

MA

[14]

Quantitative proteome profiling of C. burnetii under tetracycline stress conditions.

Xu C, Lin X, Ren H, Zhang Y, Wang S, Peng X. Analysis of outer membrane proteome of

TE

[15]

D

PLoS One 2012;7:e33599.

Escherichia coli related to resistance to ampicillin and tetracycline. Proteomics

[16]

CE P

2006;6:462-473.

Bhargava P, Collins JJ. Boosting bacterial metabolism to combat antibiotic resistance. Cell Metab 2015;21:154-155. Burchmore R. Mapping pathways to drug resistance with proteomics. Expert Rev

AC

[17]

Proteomics 2014; 11: 1-3. [18]

Jones-Dias D, Manageiro V, Caniça M. Influence of agricultural practice on mobile bla genes: IncI1-bearing CTX-M, SHV, CMY and TEM in Escherichia coli from intensive farming soils. Environ Microbiol 2016; 18:260-272.

[19]

Bonnet R, Caron F, Cavallo JD, Chardon C, Chidiac P, Courvalin P, Dubreuil L, Jarlier V, Jehl F, Lambert T, Leclercq R, Lina G, Merens A, Nicolas-Chanoine MH, Plesiat P, Ploy MC, Quentin C, Soussy CJ, Varon E, Weeber P. Comité de l'Antibiogramme de la Société Française de Microbiologie. Recommandations 2013. Paris, France.

[20]

Jones-Dias D, Manageiro V, Sampaio DA, Vieira L, Caniça M. Draft Genome Sequence of a Pathogenic O86:H25 Sequence Type 57 Escherichia coli Strain Isolated from Poultry and Carrying 12 Acquired Antibiotic Resistance Genes. Genome Announc 2015;3:5.

ACCEPTED MANUSCRIPT [21]

Cosentino S, Voldby Larsen M, Møller Aarestrup F, Lund O. PathogenFinderdistinguishing friend from foe using bacterial whole genome sequence data. PLoS One 2013;8:e77302. Wirth T, Falush D, Lan R, Colles F, Mensa P, Wieler, LH, Karch H, Reeves PR, Maiden

T

[22]

[23]

SC R

perspective. Mol Microbiol 2006; 60: 1136–1151.

IP

MC, Ochman H, Achtman M. Sex and virulence in Escherichia coli: an evolutionary

Joensen KG, Tetzschner AM, Iguchi A, Aarestrup FM, Scheutz. Rapid and easy in silico serotyping of Escherichia coli using whole genome sequencing (WGS) data. J Clin

[24]

NU

Microbiol 2005; 53:2410-2426.

Mendonça N, Manageiro V, Robin F, Salgado MJ, Ferreira E, Caniça M, Bonnet R. The

MA

Lys234Arg substitution in the enzyme SHV-72 is a determinant for resistance to clavulanic acid inhibition. Antimicrob Agents Chemother 2008;52:1806-1811. [25]

Laemmli UK. Cleavage of structural proteins during the assembly of the head of

Carvalho AS, Ribeiro H, Voabil P, Penque D, Jensen ON, Molina H, Matthiesen R. Global

TE

[26]

D

bacteriophage T4. Nature 1970;227:680-685.

mass spectrometry and transcriptomics array based drug profiling provides novel

CE P

insight into glucosamine induced endoplasmic reticulum stress. Mol Cell Proteomics 2014;13:3294-3307. [27]

Schlosser A, Volkmer-Engert R. Volatile polydimethylcyclosiloxanes in the ambient

AC

laboratory air identified as source of extreme background signals in nanoelectrospray mass spectrometry. J Mass Spectrom 2003;38:523-525. [28]

Olsen JV, de Godoy LM, Li G, Macek B, Mortensen P, Pesch R, Makarov A, Lange O, Horning S, Mann M. Parts per million mass accuracy on an Orbitrap mass spectrometer via lock mass injection into a C-trap. Mol Cell Proteomics 2005; 4:2010-2021.

[29]

Matthiesen R. LC-MS spectra processing. Methods Mol Biol 2013;1007:47-63.

[30]

Matthiesen R. Algorithms for database-dependent search of MS/MS data. Methods Mol Biol 2013; 1007:119-138.

[31]

Bunkenborg J, Garcia GE, Paz MI, Andersen JS, Molina H. The minotaur proteome: avoiding cross-species identifications deriving from bovine serum in cell culture models. Proteomics 2010; 10:3040-3044.

ACCEPTED MANUSCRIPT [32]

Matthiesen R, Prieto G, Amorim A, Aloria K, Fullaondo A, Carvalho AS, Arizmendi JM. SIR: Deterministic protein inference from peptides assigned to MS data. J Proteomics 2012; 75:4176-4183. Matthiesen R, Carvalho AS. Methods and algorithms for quantitative proteomics by

Mann K, Edsinger E. The Lottia gigantea shell matrix proteome: re-analysis including

SC R

[34]

IP

mass spectrometry. Methods Mol Biol 2013;1007:183-217.

T

[33]

MaxQuant iBAQ quantitation and phosphoproteome analysis. Proteome Sci 2014;12:28. [35]

Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful

[36]

NU

approach to multiple testing. J Roy Stat Soc 1995;57:289–300. Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysis of large

[37]

MA

gene lists using DAVID bioinformatics resources. Nat Protoc 2009;4:44-57. R-Core-Team. R: A language and environment for statistical computing. R Foundation

Xie L, Liu W, Li Q, Chen S, Xu M, Huang Q, Zeng J, Zhou M, Xie J. First succinyl-

TE

[38]

D

for Statistical Computing, Vienna, Austria (http://www.R-project.org) 2014.

proteome profiling of extensively drug-resistant Mycobacterium tuberculosis revealed

[39]

CE P

involvement of succinylation in cellular physiology. J Proteome Res 2015;14:107-119. Pioletti M, Schlunzen F, Harms J, Zarivach R, Gluhmann M, Avila H, Bashan A, Bartels H, Auerbach T, Jacobi C, Hartsch T, Yonath A, Franceschi F. Crystal structures of

AC

complexes of the small ribosomal subunit with tetracycline, edeine and IF3. EMBO J 2001; 20:1829-1839. [40]

Brodersen DE, Clemons WM, Carter AP, Morgan-Warren RJ, Wimberly BT, Ramakrishnan V. The structural basis for the action of the antibiotics tetracycline, pactamycin, and hygromycin B on the 30S ribosomal subunit. Cell 2000;103:11431154.

[41]

Lobritz MA, Belenky P, Porter CB, Gutierrez A, Yang JH, Schwarz EG, Dwyer DJ, Khalil A, S, Collins JJ. Antibiotic efficacy is linked to bacterial cellular respiration. Proc Natl Acad Sci USA 2015;112:8173-80.

[42]

Liu X, Hu Y, Pai PJ, Chen D, Lam H. Label-free quantitative proteomics analysis of antibiotic response in Staphylococcus aureus to oxacillin. J Proteome Res 2014;13:1223-1233.

ACCEPTED MANUSCRIPT [43]

Shi L, Kobir A, Jers C, Mijakovic I. Bacterial protein-tyrosine kinases. Current Proteomics 2010;7: 188-194. Baer B, Millar AH. Proteomics in evolutionary ecology. J Proteomics 2015; 135:4-11.

[45]

Kim SC, Sprung R, Chen Y, Xu Y, Ball H, Pei J, Cheng T, Kho Y, Xiao H, Xiao L, Grishin

T

[44]

IP

NV, White M, Yang XJ, Zhao Y. Substrate and functional diversity of lysine acetylation

[46]

SC R

revealed by a proteomics survey. Mol Cell 2006; 23:607-618.

Weinert BT, Iesmantavicius V, Wagner SA, Schölz C, Gummesson B, Beli P, Nyström T, Choudhary C. Acetyl-phosphate is a critical determinant of lysine acetylation in E. coli.

[47]

NU

Mol Cell 2013; 51:265-272.

Ouidir T, Kentache T, Hardouin J. Protein lysine acetylation in bacteria: Current state of

[48]

MA

the art. Proteomics 2016; 16:301-309.

Beck HC, Nielsen EC, Matthiesen R, Jensen LH, Sehested M, Finn P, Grauslund M, Hansen AM, Jensen ON. Quantitative proteomic analysis of post-translational

Schölz C, Weinert BT, Wagner SA, Beli P, Miyake Y, Qi J, Jensen LJ, Streicher W,

TE

[49]

D

modifications of human histones. Mol Cell Proteomics 2006; 5:1314-1325.

McCarthy AR, Westwood NJ, Lain S, Cox J, Matthias P, Mann M, Bradner JE, Choudhary

CE P

C. Acetylation site specificities of lysine deacetylase inhibitors in human cells. Nat Biotechnol. 2015; 33:415-423. [50]

Piatkov K, Graciet E, Varshavsky A. Ubiquitin reference technique and its use in

[51]

AC

ubiquitin-lacking prokaryotes. PLoS One 2013; 8:e67952. Hu X, Norris AL, Baudry J, Serpersu EH. Coenzyme A binding to the aminoglycoside acetyltransferase (3)-IIIb increases conformational sampling of antibiotic binding site. Biochemistry 2011; 50:10559-10565. [52]

Davies JE. Aminoglycosides: ancient and modern. J Antibiot (Tokyo) 2006;59:529-532.

[53]

Mills SD, Eakin A, Buurman ET, Newman JV, Gao N, Huynh H, Johnson KD, Lahiri S, Shapiro AB, Walkup GK, Yang W, Stokes SS. Novel bacterial NAD+-dependent DNA ligase inhibitors with broad-spectrum activity and antibacterial efficacy in vivo. Antimicrob Agents Chemother 2011; 55:1088-1096.

[54]

Blankenfeldt W, Asuncion, M, Lam JS, Naismith JH. The structural basis of the catalytic mechanism and regulation of glucose-1-phosphate thymidylyltransferase (RmlA). EMBO J 2000;19:6652-6663.

ACCEPTED MANUSCRIPT [55]

Blankenfeldt W, Giraud MF, Leonard G, Rahim R, Creuzenet C, Lam JS, Naismith JH. The purification, crystallization and preliminary structural characterization of glucose-1phosphate thymidylyltransferase (RmlA), the first enzyme of the dTDP-L-rhamnose

T

synthesis pathway from Pseudomonas aeruginosa. Acta Crystallogr D Biol Crystallogr

[56]

IP

2000;56:1501-1504.

Yang W, Moore IF, Koteva KP, Bareich DC, Hughes DW, Wright GD. TetX is a flavin-

SC R

dependent monooxygenase conferring resistance to tetracycline antibiotics. J Biol Chem 2004;279:52346-52352. [57]

Volkers G, Palm GJ, Weiss MS, Wright GD, Hinrichs W. Structural basis for a new

NU

tetracycline resistance mechanism relying on the TetX monooxygenase. FEBS Lett 2011;585:1061-1066.

Snyder JA, Haugen BJ, Lockatell CV, Maroncle N, Hagan EC, Johnson DE, Welch RA,

MA

[58]

Mobley HL. Coordinate expression of fimbriae in uropathogenic Escherichia coli. Infect

[59]

D

Immun 2005;73:7588-7596.

Simms AN, Mobley HL. Multiple genes repress motility in uropathogenic Escherichia

Simms AN, Mobley HL. PapX, a P fimbrial operon-encoded inhibitor of motility in

CE P

[60]

TE

coli constitutively expressing type 1 fimbriae. J Bacteriol 2008, 190;3747-3756.

uropathogenic Escherichia coli. Infect Immun 2008;76:4833-4841. [61]

Fischer EA, Dierikx CM, van Essen-Zandbergen A, van Roermund HJ, Mevius DJ,

AC

Stegeman A, Klinkenberg D. The IncI1 plasmid carrying the blaCTX-M-1 gene persists in in vitro culture of a Escherichia coli strain from broilers. BMC Microbiol 2014;14:77. [62]

van Hoek AH, Mevius D, Guerra B, Mullany P, Roberts AP, Aarts HJ. Acquired antibiotic resistance genes: an overview. Front Microbiol 2011;2:203.

[63]

de Cristóbal RE, Vincent PA, Salomón RA. Multidrug resistance pump AcrAB-TolC is required for high-level, Tet(A)-mediated tetracycline resistance in Escherichia coli. J Antimicrob Chemother 2006;58:31-36.

[64]

Delcour AH. Outer membrane permeability and antibiotic resistance. Biochim Biophys Acta 2009;1794:808-816.

[65]

Albrecht R, Zeth K. Structural basis of outer membrane protein biogenesis in bacteria. J Biol Chem 2011;286:27792-27803.

ACCEPTED MANUSCRIPT [66]

Ramírez MS, Piñeiro S, Centrón D, Argentinian Group for Integron Study. Novel insights about class 2 integrons from experimental and genomic epidemiology. Antimicrob Agents Chemother 2010;54:699-706. Olivares J, Bernardini A, Garcia-Leon G, Corona FB, Sanchez M, Martinez JL. The

T

[67]

Petkovsek Z, Zgur-Bertok D, Starcic Erjavec M. Colicin insensitivity correlates with a

SC R

[68]

IP

intrinsic resistome of bacterial pathogens. Front Microbiol 2013;4:103.

higher prevalence of extraintestinal virulence factors among Escherichia coli isolates from skin and soft-tissue infections. J Med Microbiol 2012;61:762-765. [69]

Smajs D, Micenková L, Smarda J, Vrba M, Sevčíková A, Vališová Z, Woznicová V.

NU

Bacteriocin synthesis in uropathogenic and commensal Escherichia coli: colicin E1 is a potential virulence factor. BMC Microbiol 2010;10:288. Lin X, Kang L, Li H, Peng X. Fluctuation of multiple metabolic pathways is required for

MA

[70]

Escherichia coli in response to chlortetracycline stress. Mol Biosyst 2014;10:901-908. Willis MA, Zhuang Z, Song F, Howard A, Dunaway-Mariano D, Herzberg O. Structure of

D

[71]

TE

YciA from Haemophilus influenzae (HI0827), a hexameric broad specificity acylcoenzyme A thioesterase. Biochemistry 2008;47:2797-2805. Trotter V, Vinella D, Loiseau L, Ollagnier de Choudens S, Fontecave M, Barras F. The

CE P

[72]

CsdA cysteine desulphurase promotes Fe/S biogenesis by recruiting Suf components and participates to a new sulphur transfer pathway by recruiting CsdL (ex-YgdL), a

[73]

AC

ubiquitin-modifying-like protein. Mol Microbiol 2009;74:1527-1542. Erbel PJ, Barr K, Gao N, Gerwig GJ, Rick PD, Gardner KH. Identification and biosynthesis of cyclic enterobacterial common antigen in Escherichia coli. J Bacteriol 2003;185:1995-2004. [74]

Yang Y, Tsui HC, Man TK, Winkler ME. Identification and function of the pdxY gene, which encodes a novel pyridoxal kinase involved in the salvage pathway of pyridoxal 5'-phosphate biosynthesis in Escherichia coli K-12. J Bacteriol 1998;180:1814-1821.

[75]

Gutiérrez-Preciado A, Torres AG, Merino E, Bonomi HR, Goldbaum FA, García-Angulo VA.. Extensive identification of bacterial riboflavin transporters and their distribution across bacterial species. PLoS One 2015;10:e0126124.

[76]

Postma PW, Lengeler JW, Jacobson GR. Phosphoenolpyruvate: carbohydrate phosphotransferase systems of bacteria. Microbiol Rev 1993;57:543-594.

ACCEPTED MANUSCRIPT [77]

Garces F, Fernández FJ, Gómez AM, Pérez-Luque R, Campos E, Prohens R, Aguilar J, Baldomà L, Coll M, Badía J, Vega MC. Quaternary structural transitions in the DeoRtype repressor UlaR control transcriptional readout from the L-ascorbate utilization

Manges AR, Harel J, Masson L, Edens TJ, Portt A, Reid-Smith RJ, Zhanel GG, Kropinski

IP

[78]

T

regulon in Escherichia coli. Biochemistry 2008;47:11424-11433.

AM, Boerlin P. Multilocus sequence typing and virulence gene profiles associated with

SC R

Escherichia coli from human and animal sources. Foodborne Pathog Dis 2015;12:302310.

Guzmán GI, Utrilla J, Nurk S, Brunk E, Monk JM, Ebrahim A, Palsson BO, Feist AM.

CE P

TE

D

MA

Natl Acad Sci USA 2015;112:929-934.

NU

Model-driven discovery of underground metabolic functions in Escherichia coli. Proc

AC

[79]

ACCEPTED MANUSCRIPT

IP

T

Figure captions:

CR

Figure 1. Functional enrichment analysis of all identified soluble proteins from E. coli EcAmb278 cultured in the presence and absence of tetracycline. (A) The proteins are analyzed according to cellular component (CC), (B) biological process (BP) and (C) molecular function (MF). The probability of a protein

US

being enriched in a specific subcategory is represented by –log p, p corresponding to the p value. The values on top of each bar correspond to the number of

MA N

genes enclosed in the specific subcategory which correlates with the pattern fill.

TE D

Figure 2. Venn diagram comparing identified proteins in control and tetracycline treated E. coli.

Figure 3. Heat map depicting significant regulated proteins after p value correction based on spectral counting (FDR-corrected p value ≤ 0.05). M depicts more (green) or less (red) than 2-fold regulated. Evidence group: E= 1 (blue), 2 (green), 3 (yellow). FDR for identification: blue <=0.001, green (0.001-0.01].

AC

CE P

Spectral counts: [10-50) yellow, [50-100) green and >=100 blue.

ACCEPTED MANUSCRIPT

IP

T

Supplementary material:

CR

Figure S1. Number of proteins and genes encoding proteins identified at 1% FDR across samples. C1, C2, C3: replicas of control EcAmb278; T1, T2, T3:

US

replicas of tetracycline treated EcAmb278.

Figure S2. Relative frequency of detected protein modifications.N_acPro, N-terminal modification of proteins; M_oxidation, methionine oxidation; K_ac,

MA N

acetylation of lysine; Q_deamidation, glutamine deamidation; N_deamidation, asparagine deamidation.

TE D

Figure S3. Distribution of iBAQ values across samples. C1, C2, C3: replicas of control EcAmb278; T1, T2, T3: replicas of tetracycline treated EcAmb278.

CE P

Figures S4. Annotated raw spectra assigned to peptides and m/z versus delta m/z plots.

AC

Figures S5. Heatmaps summarizing the extracted ion counts for proteins found to be significant regulated by spectral counting.

Figures S6. Functional enrichment analysis of identified proteins unique forE. coli EcAmb278 cultured in the absence of tetracycline. The proteins are analyzed according to cellular component (CC), biological process (BP) and molecular function (MF). The probability of a protein being enriched in a specific subcategory is represented by –log p, p corresponding to the p value. The values on top of each bar correspond to the number of genes enclosed in the specific subcategory which correlates with the pattern fill.

Figures S7. Functional enrichment analysis of identified proteins unique for E. coli EcAmb278 cultured in the presence of tetracycline. The proteins are analyzed according to cellular component (CC), biological process (BP) and molecular function (MF). The probability of a protein being enriched in a specific

ACCEPTED MANUSCRIPT

subcategory is represented by –log p, p corresponding to the p value. The values on top of each bar correspond to the number of genes enclosed in the

CR

IP

T

specific subcategory which correlates with the pattern fill.

Figure S8. Comparison of protein modifications between tetracycline treated and non-treated replicas. acPro, N-terminal modification of proteins; K_ac,

US

acetylation of lysine.

MA N

Table S1. UniProt annotation, quantitative and statistical results on the identified proteins.

AC

CE P

TE D

Table S2. Specific proteins identified in E. coli EcAmb278 that are involved in the processes of resistance, virulence and transfer of foreign DNA.

ACCEPTED MANUSCRIPT

IP

T

Significance

CR

The lack of new antibiotics to fight infections caused by multidrug resistant microorganisms has motivated the use of old antibiotics, and the search for new drug targets. The evolution of antibiotic resistance is complex, but it is known that agroecosystems play an important part in the selection of antibiotic

US

resistance bacteria. Tetracyclines are still used as phytopharmaceutical agents in crops, selecting resistant bacteria and changing the ecology of farm soil. Little is known about the metabolic response of genetically resistant populations to antibiotic exposure. Indeed, to date there are no quantitative

MA N

tetracycline resistance studies performed with the latest generation of high resolution mass spectrometers allowing high mass accuracy in both MS and MS/MS scans. Here, we report the proteome profiling of a soil-borne Escherichia coli upon tetracycline stress, so that this new perspective could provide a

AC

CE P

TE D

broaden understanding of the metabolic responses of E. coli to a widely used antibiotic.

ACCEPTED MANUSCRIPT

Biological process

Protein

Function

IP

Strain

CR

Accession number

T

Table 1. Proteins significantly up- and down-regulated in E. coli EcAmb278 challenged with 10 mg/L of tetracycline.

Up-regulated proteins

Gene

Log2 ratio

FDR-corrected p value

Reference

E. coli O6:H1 ATCC700928

Other biosynthesis processes

Putative acylCoA thioester hydrolase

Catalyzes the hydrolysis of the thioester bond in palmitoyl-CoA and malonyl-CoA Catalyzes the hydrolysis of the thioester bond in palmitoyl-CoA and malonyl-CoA

yciA

2,3

1,86E-61

[69]

Q8XBL5

E. coli O157:H7

DNA repair/replication

DNA ligase

Catalyzes the formation of phosphodiester linkages between 5'phosphoryl and 3'-hydroxyl groups in double-stranded DNA; essential for DNA replication and repair of damaged DNA

ligA

4,7

0,000384

[51]

Q46925

E. coli O157:H7

Other biosynthesis processes

Cysteine sulfinate desulfinase

Functions as a selenium delivery protein in the pathway for the biosynthesis of selenophosphate; seems to participate in Fe/S biogenesis by recruiting the SufBCDSufE proteins

csdA

3

0,002996

[70]

P61887

E. coli K12

Virulence

Glucose-1phosphate thymidylyltransf erase

Catalyzes the formation of dTDP-glucose, from dTTP and glucose 1-phosphate, as well as its pyrophosphorolysis; involved in O-antigen biosynthesis

rmlA

3,4

0,00494

[71]

Q8FH89

E. coli O6:H1 ATCC700928

DNA repair/replication

Pyridoxamine kinase

Functions in a salvage pathway using pyridoxamine and phosphorylating B6 vitamers

pdxY

2,5

0,017328

[72]

P64624

E. coli K12

Transcription

Predicted

Unknown function

yheO

3,2

0,017462

-

AC

CE P

TE D

MA N

US

P0A8Z1

ACCEPTED MANUSCRIPT

IP

T

transcriptional regulator E. coli O157:H7

Biosynthesis of vitamins

Riboflavin biosynthesis protein

Involved in the riboflavin (vitamin B2) biosynthesis pathway as kinase and transferase

ribF

3,2

0,017717

[73]

P0AD11

E. coli O6:H1 ATCC700928

Unknown

Hypothetical protein

Unknown function

yecJ

2,8

0,017717

-

Down-regulated proteins

MA N

US

CR

P0AG42

E. coli K12

Intracellular trafficking and secretion

PTS system fructose-specific EIIBC component

Catalyzes the phosphorylation of incoming sugar substrates concomitantly with their translocation across the cell membrane; involved in fructose transport

fruA

-3,4

0,002996

[74]

B7UQK0

E. coli O127:H6 E2348/69

Transcription

HTH-type transcriptional regulator

L-Ascorbate is utilized under anaerobic conditions through proteins encoded by the ula regulon, which is under the control of theulaR repressor

ulaR

-2,8

0,00494

[75]

P08191

E. coli K12

Virulence

Bacterial adhesin

Involved in regulation of length and mediation of adhesion of type 1 fimbriae

fimH

-2,1

0,007356

[76]

P22256

E. coli K12

Other processes

4-aminobutyrate aminotransferas e GabT

Catalyzes the transfer of the amino group from gamma-aminobutyrate to alpha-ketoglutarate to yield succinic semialdehyde

gabT

-3,2

0,032568

[77]

AC

CE P

TE D

P20966

ACCEPTED MANUSCRIPT

Highlights

CR

IP

T

(3 to 5 bullet points, max 85 caracters)

Label free quantitative proteomics was applied to soil antibiotic resistant E. coli



Globally 1,484 proteins were identified at a false discovery rate threshold of 1%



108 proteins were exclusively identified in E. coli cells exposed to tetracycline



Interesting differences were noted in energy metabolism and cell wall proteins



12 proteins associated with distinct processes were differentially regulated

AC

CE P

TE D

MA N

US



ACCEPTED MANUSCRIPT

AC

CE P

TE D

MA N

US

CR

IP

T

Graphical abstract

ACCEPTED MANUSCRIPT

AC

CE P

TE D

MA N

US

CR

IP

T

Figure 1

ACCEPTED MANUSCRIPT

AC

CE P

TE D

MA N

US

CR

IP

T

Figure 2

ACCEPTED MANUSCRIPT

AC

CE P

TE D

MA N

US

CR

IP

T

Figure 3