Scanning, data analysis and protein identification by MALDI-TOF-MS

2. Chapter: Material and methods

2.6. Scanning, data analysis and protein identification by MALDI-TOF-MS

Differential 2D gels were scanned on a Typhoon Trio™ Scanner (GE Healthcare) at a resolution of 100 dots / cm using filters with specific excitation and emission wavelengths for Cy3 (filter BP 30; 532 nm / 580 nm), and Cy5 (filter BP 30; 633 nm/670 nm). Protein spot abundance was analyzed by DeCyder^™ version 6.0 (Differential Analysis Software, GE Healthcare) using the differential in-gel analysis module (DIA). Quantification was applied for filter-confirmed spots with slope > 1.4, area < 420, and volume < 130,000. For membrane fractions, gels from 2D DIGE experiments were stained with colloidal Coomassie G-250 and prominent protein spots were excised for analysis. For cytoplasmic fractions, in parallel to 2D DIGE experiments, preparative 2D GE of respective mucosa-derived cytoplasm were performed using 0.5 mg - 1 mg of protein load. Isolated protein spots were trypsinized (Shevchenko et al., 1996), and dried samples were resolubilized in 3 µl 50% ACN, 0.1%

TFA, 1 μl of the peptide solution was mixed with 1 μl of 2 mg/ml CHCA, 50% ACN, 0.1%

TFA and spotted on the target plate. Samples were analyzed on a VoyagerDE™ Pro as described earlier (Buettner et al., 2009) or an AB Sciex ToF/ToF 5800™ mass spectrometer (both AB Sciex., Foster City, CA). For MALDI-TOF/TOF analysis, internal calibration on autolytic porcine trypsin peptides was applied for precursor MS spectra and external calibration with Glu-Fib fragments was used for MS/MS spectra. Mass spectrometrical data were searched against the SwissProt Database with carbamidomethylation of cysteins, oxidation of methionine and N-terminal acetylation as variable modification. A precursor mass deviation of 120 ppm and 0.5 Da for MS/MS fragments was used.

Chapter 2 – Material and methods

32 2.7. Tube-gel trypsin digestion

For tube-gel trypsin digestion (Lu & Zhu, 2005) 10 µl containing 100 µg of protein were mixed with 30 µl acrylamide-bisacrylamide solution (37.5:1; 30% [w/v]) and 30 µl distilled water containing 1% TEMED and 1% ammonium peroxodisulfate. Subsequently, proteins were fixed with 40% methanol / 10% acetic acid, and the gel was cut in small pieces. Proteins were trypsinized using Trypsin Gold-Mass Spec Grade (Promega) supplementing with 0.025% of trypsin enhancer ProteasMax^TM Surfactant (Promega) in relation of 2 µg trypsin/100 µg protein. Extraction of peptides from the gel matrix was performed in 20%, 50%, and 80% acetonitrile-water with 0.5% formic acid for 30 min in an ultrasonic bath.

2.8. SDS-PAGE-based trypsin digestion

Four of the MDC and CDC preparations (two from each isolate) were additionally analysed by gel-based liquid chromatography-tandem mass spectrometry (GeLC-MS/MS, Fig. 1-5) and tube-gel trypsin digestion followed by LC-MS/MS (Weigoldt et al., 2011). For GeLC-MS/MS, equal amounts of protein extracts (30 µg) of MDC and CDC were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and stained with Coomassie brilliant blue. From each lane seven gel slices were excised and in-gel trypsin digestion were performed as described for tube-gel trypsin digestion (see above).

Fig. 1-5. SDS-PAGE and GeLC-MS/MS of mucosa-and culture-derived cytoplasm.

10.8% (A) and 15% (B) polyacrylamide gel concentration was used for protein separation according to size.

Lanes were loaded with 30 µg protein of MDC from cow 2 (1) and cow 3 (2) as well as CDC from cow 2 (3) and

Chapter 2 – Material and methods

2.9. Nanoflow-liquid-chromatography-coupled tandem mass spectrometry (nUPLC-ESI Q-TOF-MS/MS)

A 100 µm x 100 mm nUPLC column (ACQUITY UPLC^® 1.7 µm BEH130 C18; Waters, Milford, MA, USA) was used for peptide separation. Each sample was run over 45 min with a flow rate of 300 nl min^-1. Mobile phase A consisted of 0.1% formic acid in 1% acetonitrile.

Mobile phase B included 0.1% formic acid in 99% acetonitrile. The gradient was 99% mobile phase A for 0.33 min and then ramped linearly to 65% of mobile phase A over 30 min. Over the next 1 min, it was ramped to 15% phase A and held for 1 min before equilibrating the column with 99% phase A for 13 min. Before each other run, calibration of the Quadrupole Time-Of-Flight tandem mass spectrometer (Q-TOF Ultima, Waters) was performed using [Glu1]fibrinopeptide B as standard. The capillary voltage was 1800 V and was tuned for signal intensity. The two most intense ions with charge states between 2 and 4 and mass ranges between 450 and 1600 were selected in each survey scan if they met the switching criteria. Different collision energies were used to fragment each peptide ion on the basis of its mass-to-charge (m/z) values. For tube-gel trypsin digestion, each sample was run up to eight times. For each new measurement, previously detected peptides in a mass range within ±100 mDa were excluded from analysis in order to identify fragmentation on peptides represented in weaker signals. By this approach each sample was measured until no further increase in the number of identified peptides was observed. The search algorithm was set to allow for carbamidomethylation on cysteine residues, oxidation on methionine residues, and a maximum of one missed cleavage. A minimum of one validated peptide containing four or more consecutive y-ions was set for protein matches.

2.10. High throughput protein analysis using nanoflow-liquid-chromatography-coupled tandem mass spectrometry (LC-orbitrap-MS/MS)

The LC-orbitrap-MS has been applied in corporation with Prof. Dr. rer. nat. Andreas Pich, MS lab in the Institute for Toxicology, Medical School Hannover.

Peptide extracts were combined, dried and redissolved in 10 μl 2% ACN, 0.1% TFA. LC-MS/MS analysis was performed on an LTQ Orbitrap Velos mass spectrometer (Thermo Fisher Scientific) as described recently (Boer et al., 2011). Briefly, an appropriate sample amount was loaded onto a nanoflow ultra-high pressure liquid chromatography system (RSLC, Dionex) equipped with a trapping column (5 mm C18 particle, 75 mm ID, C18

material, 2 cm length, PepMap, Dionex) and separating column (2 mm C18 particle, 75 mm ID, 50 cm length, PepMap, Dionex). After trapping, peptides were eluted with a linear gradient of buffer B (80% ACN, 0.1% formic acid) in buffer A (0.1% formic acid) from 4%

to 25% in 60 min, 25%-50% in 25 min and 50%-90% in 5 min, after which the column was flushed for 10 min isocratically with 90% B and reconditioned to 4% B in 20 min. Flow rate was 250 nl/min with a column temperature of 40 °C. Peptides were ionized in the nanoESI source with 1.2 kV. Overview scans were acquired at a resolution of 60 k in a mass range of m/z 300-1600 in the orbitrap. The top 10 most intensive ions of charge two or three and a minimum intensity of 2000 were selected for CID fragmentation with a normalized collision energy of 38.0, an activation time of 10ms and an activation Q of 0.250 in the linear ion trap mass analyzer of the LTQ Orbitrap Velos. Fragmentation mass spectra were also recorded in the LTQ. The m/z values in a 10 ppm mass window of the selected ions were subsequently excluded from the fragmentation for 70 s. Data analysis was facilitated by proteome discoverer software 1.2 (Thermo Fisher Scientific, Langenselbold, Germany) and the Mascot search algorithm. Mascot was set up to search a customized database generated using the UniProt database (release 2012_03). It includes M. avium subspecies paratuberculosis K10 (NCBI Reference Sequence: NC_002944.2; 4350 genes, 4323 protein entries in UniProt), and a total of 6760 reviewed bovine protein entries (searched for Bos taurus). A false discovery rate of 0.01 and a Peptide-Score of 30 were used. Proteins were stated identified if at least two unique peptides were detected.

2.11. Criteria for protein identification and differential expression

Proteins from mucosa- and culture-derived MAP were considered as identified if they were detected at least twice by GeLC-MS or tube-gel trypsin digestion followed by MS in independent biological repeats. Alternatively, proteins were also considered as identified if isolated from 2D gels and giving a significant database match of the peptide mass fingerprint.

The proteins identified in MDCs were considered to be present also in CDCs after a single detection by GeLC-MS or tube-gel trypsin digestion followed by MS.

Proteins were considered as differentially expressed only if an at least 1.5-fold increase in expression could be observed in MDC preparations from each of the two cows in comparison to the corresponding CDCs.

Chapter 2 – Material and methods

36 2.12. Data processing and bioinformatics

LC-MS/MS raw data were processed using ProteinLynx™ Global Server (Version 2.1, Waters) by searching against the Mycobacterium species SWISS-PROT database downloaded on 25.09.2008 [99,308 entries]. The identification of bovine proteins was performed by searching against the NCBI nonredundant database as downloaded on August 16th 2006. All proteins identified in MDM or CDM as being specific for MAP were copied into separate Microsoft Excel data sheets. The mycobacterial origin of all proteins detected in MDM by a one peptide-hit only was confirmed by searching against the entire NCBI data base.

Prediction for membrane association or cytoplasmatic localisation of these proteins was performed with PSORT (http://psort.ims.u-tokyo.ac.jp/form.html). Verified datasets were organized according to their distribution in the Cluster of Orthologous Groups (COGs).

Pathway reconstruction was performed using the cellular overview tool from SRI’s pathway tools software (http://ecocyc.org/background.shtml) for proteins with a Reference Common Name (RCN). In order to obtain information possibly missed using the cellular overview tool the KEGG database was searched using MAP annotation numbers and M. tuberculosis orthologues identified by protein homology blast using the Multi-Genome Homology Comparison (Comparative Tools) of Comprehensive Microbial Resource (CMR) available at http://cmr.jcvi.org and the TB database (http://genome.tbdb.org) and NCBI.

Reference List

Boer, U., Lohrenz, A., Klingenberg, M., Pich, A., Haverich, A. & Wilhelmi, M. (2011). The effect of detergent-based decellularization procedures on cellular proteins and immunogenicity in equine carotid artery grafts. Biomaterials 32, 9730-9737.

Buettner, F. F., Bendalla, I. M., Bosse, J. T., Meens, J., Nash, J. H., Hartig, E., Langford, P. R. & Gerlach, G. F. (2009). Analysis of the Actinobacillus pleuropneumoniae HlyX (FNR) regulon and identification of iron-regulated protein B as an essential virulence factor. Proteomics 9, 2383-2398.

Choy, E., Whittington, R. J., Marsh, I., Marshall, J. & Campbell, M. T. (1998). A method for purification and characterisation of Mycobacterium avium subsp. paratuberculosis from the intestinal mucosa of sheep with Johne's disease. Vet Microbiol 64, 51-60.

Lu, X. N. & Zhu, H. N. (2005). Tube-gel digestion - A novel proteomic approach for high throughput analysis of membrane proteins. Molecular & Cellular Proteomics 4, 1948-1958.

Möbius, P., Fritsch, I., Luyven, G., Hotzel, H. & Kohler, H. (2009). Unique genotypes of Mycobacterium avium subsp. paratuberculosis strains of Type III. Vet Microbiol.

Shevchenko, A., Wilm, M., Vorm, O. & Mann, M. (1996). Mass spectrometric sequencing of proteins from silver stained polyacrylamide gels. Analytical Chemistry 68, 850-858.

Weigoldt, M., Meens, J., Doll, K., Fritsch, I., Möbius, P., Goethe, R. & Gerlach, G. F. (2011). Differential proteome analysis of Mycobacterium avium subsp. paratuberculosis grown in vitro and isolated from cases of clinical Johne's disease. Microbiology 157, 557-565.

3. Chapter:

Differential proteome analysis of Mycobacterium avium subsp.

paratuberculosis grown in vitro and isolated from cases of clinical Johne’s disease

Mathias Weigoldt¹, Jochen Meens¹, Klaus Doll², Isabel Fritsch³, Petra Möbius³, Ralph Goethe¹†, Gerald-F. Gerlach¹†‡

1 Institute for Microbiology, Department of Infectious Diseases, University of Veterinary Medicine Hannover, Hannover, Germany

2 Clinic for Ruminants and Swine (Internal Medicine and Surgery), Justus-Liebig-University, Giessen, Germany

3 Institute of Molecular Pathogenesis, Friedrich-Loeffler-Institut (Federal Research Institute for Animal Health), Jena, Germany

† These authors contributed equally to this work.

‡ Corresponding author. Current address: IVD GmbH, Heisterbergallee 12, 30453 Hannover, Germany.

Correspondence Gerald-F. Gerlach gfgerlach@gmx.de

Running title: Johne’s disease - bacterial proteins in the host Contents category: Microbial Pathogenicity

Keywords: M. avium subsp. paratuberculosis; differential protein expression; protein expression in the host; 2D DIGE; nUPLC-ESI Q-TOF-MS/MS

(Manuscript has been published in Microbiology 2011, 157, 557-565) (Online available: doi: 10.1099/mic.0.044859-0)

ABSTRACT

Bovine Johne’s disease (paratuberculosis), caused by M. avium subspecies paratuberculosis, poses a significant economic problem to the beef and dairy industry worldwide. Despite its relevance, however, pathogenesis of Johne’s disease is still only partially resolved. Since mycobacterial membrane proteins expressed during infection are likely to play an important role in pathogenesis, membrane-enriched fractions, namely mucosa-derived membranes (MDM) and culture-derived membranes (CDM), of M. avium subsp. paratuberculosis from three cows with clinical paratuberculosis were investigated. An initial analysis by two-dimensional difference gel electrophoresis (2D DIGE) and MALDI-TOF-MS analysis revealed four differentially expressed proteins with only one predicted membrane protein.

Due to this limited outcome membrane preparations were subjected to a tube-gel trypsin digestion and investigated by nanoflow-liquid-chromatography-coupled tandem mass spectrometry. Based on this approach a total of 212 proteins were detected in MDM including 32 proteins of bovine origin; 275 proteins were detected in CDM; 59% of MDM and CDM proteins were predicted to be membrane-associated. A total of 130 of the proteins were detected in both MDM and CDM and 48 predicted membrane proteins were detected in MDM from at least two cows. Four of these proteins were not detected in CDM, implying differential expression in the host. All membrane-associated proteins, especially the four identified as being differentially expressed, might be relevant targets for further analyses into the pathogenesis of bovine paratuberculosis.

4. Chapter:

Proteome profiling of Mycobacterium avium subsp. paratuberculosis isolates obtained from cows with clinical Johne’s disease reveals

metabolic adaptation in the natural host

Mathias Weigoldt¹, Jochen Meens¹, Klaus Doll², Gerald-F. Gerlach^1§*, Ralph Goethe^1§#

1 Institute for Microbiology, Department of Infectious Diseases, University of Veterinary Medicine Hannover, Hannover, Germany

2 Clinic for Ruminants and Swine (Internal Medicine and Surgery), Justus-Liebig-University, Giessen, Germany

§ Both senior authors contributed equally

* Current address: IVD GmbH, Heisterbergallee 12, 30453 Hannover, Germany

# Corresponding author: Institute for Microbiology, Department of Infectious Diseases, University of Veterinary Medicine Hannover, Hannover, Germany

Running title: Johne’s disease - bacterial proteins in the host

Contents category: Microbial Pathogenicity

Keywords: protein expression; metabolism, 2D DIGE; mass spectrometry; energy production and conversion; antimicrobial response

(Manuscript in preparation)

ABSTRACT

The knowledge about the adaptation of pathogenic mycobacteria such as M. tuberculosis or M. avium subspecies paratuberculosis to the environment in their natural hosts is limited. M.

avium subsp. paratuberculosis causes Johne’s disease a chronic and incurable granulomatous enteritis of ruminants and is discussed as a putative etiological agent of Crohn’s disease.

Despite the diversity of disease, M. avium subsp. paratuberculosis shares many virulence mechanisms with M. tuberculosis, which indicates common pathogenicity mechanisms of mycobacteria.

Since M. avium subsp. paratuberculosis is obtainable in large numbers from the intestinal tissue of diseased cows in the present study we used a comprehensive LC-MS/MS and 2D DIGE approach to determine the metabolome of M. avium subsp. paratuberculosis in its natural host. We found that, at large, the central metabolism of M. avium subsp.

paratuberculosis in the host differed only slightly from that in culture. The central metabolism of M. avium subsp. paratuberculosis in the host seems to be driven by β-oxidation of lipids, most probably host cells cholesterol. β-β-oxidation is accompanied by an alternative TCA pathway in which a type 2-oxoglutarate ferredoxin oxidoreductase might contribute to the conversion of α-ketoglutarate to succinyl-CoA. Carbon efflux from the TCA cycle seems to be compensated by the generation of α-ketoglutarate from glutamate via enhanced degradation of proline. Adaptation to antimicrobial host reactions is indicated by enhanced expression of SodA and KatG as well as other chaperon like proteins. Within the host, metabolism of M. avium subsp. paratuberculosis seems to be accelerated as indicted by an enhanced activity of the PPP. Enhanced ATP generation through respiratory phosphorylation gives evidence for an increased demand for energy. In conclusion, this is the first comprehensive proteomic view of the metabolism of a pathogenic mycobacterium in its natural host. Our results overall provided a highly specific picture about the metabolic adaptation of M. avium subsp. paratuberculosis in its natural host.

Chapter 4

46 INTRODUCTION

M. avium subsp. paratuberculosis is the causative agent of Johne’s disease (paratuberculosis), a chronic and incurable granulomatous enteritis of ruminants (Kreeger, 1991). The disease occurs worldwide (Manning & Collins, 2001) and has considerable economic impact on the dairy industry (Harris & Barletta, 2001). M. avium subsp. paratuberculosis is the only mycobacterial species responsible for an intestinal disease. This is of interest since M. avium subsp. paratuberculosis has long been suggested to be associated with Crohn’s disease (CD) in humans (Greenstein, 2003). CD is a chronic inflammatory bowel disease with an unknown aetiology. The incidence in the human population is increasing in Western Europe and North America (Loftus, Jr. et al., 1998; Loftus, Jr. & Sandborn, 2002). Interestingly, recent studies revealed that a high percentage of CD patients were infected with M. avium subsp.

paratuberculosis (Feller et al., 2007; Grant, 2005; Greenstein, 2003). In most cases, M. avium subsp. paratuberculosis was detected in intestinal tissues samples. Nevertheless, there is no direct experimental evidence for M. avium subsp. paratuberculosis as the primary etiological agent for CD thus far.

In cattle, M. avium subsp. paratuberculosis is transmitted primarily via the faecal-oral route to neonatal calves. During the subsequent preclinical phase (2-5 years) bacteria persist and multiply in subepithelial macrophages causing a chronic transmural inflammatory reaction (Clarke, 1997; Harris & Barletta, 2001). Pathological alterations are preferentially found in the distal jejunum, ileum and the ileoceacal junction (Sigurethardottir et al., 2004). Here, M.

avium subsp. paratuberculosis is found intracellularly in sub-epithelial macrophages. In contrast to other mycobacterial diseases granuloma formation at the site of infection is diffuse resulting in a granulomatous enteritis with mucosal thickening, and even at late stages of disease no caseous necrosis or ulceration occurs. Lesions in other areas are found less common, confirming the intestinal region as the major site of disease (Buergelt et al., 1978;

Clarke, 1997).

The molecular mechanisms of M. avium subsp. paratuberculosis pathogenicity are still incompletely understood. Experiments with macrophage cell lines or primary macrophage cells confirmed that M. avium subsp. paratuberculosis shares many virulence mechanisms with M. tuberculosis, particularly the ability to survive in the hostile environment of macrophages (Coussens, 2001; Stabel, 2006). This indicates common general mechanisms in the pathobiology of mycobacterial infections.

Nevertheless, in the host M. avium subsp. paratuberculosis is multiplying in macrophages in the small intestine which is a completely different ecological niche compared to M.

tuberculosis which affects the lung of its human host where it is found within macrophages in distinct granulomas in a dormant or latent state remaining viable for many years without the patient showing clinical signs of tuberculosis (Wayne, 1994).

In the recent years it became clear that the carbon metabolism is a major determinant of the pathogenicity of M. tuberculosis. Thus mRNA expression analysis of M. tuberculosis obtained from macrophages in vitro and from the lungs of mice and humans demonstrated that M. tuberculosis seems to adapt its intermediary metabolism in vivo by utilising host-derived lipids during the course of infection rather than glucose and glycerol, the primary carbon sources metabolised in vitro (Bloch & Segal, 1956; Schnappinger et al., 2003; Talaat et al., 2004; Timm et al., 2003). M. tuberculosis strains defective in enzymes of the pyruvate dehydrogenase complex, the glyoxylate shunt or the gluconeogenic enzyme phosphoenolpyruvate carboxykinase were attenuated during the chronic phase of infection in a mouse model of pulmonary TB (Marrero et al., 2010; McKinney et al., 2000; Munoz-Elias

& McKinney, 2005; Shi & Ehrt, 2006). Nevertheless, gene expression analyses indicate that M. tuberculosis also expresses numerous genes to evade the host immune responses and to adapt to its intracellular life style in the macrophage. Among the intracellularly induced genes several have functions in oxidative stress resistance, cell wall synthesis, iron uptake, or inhibition of apoptosis (Mukhopadhyay et al., 2011).

The knowledge about M. avium subsp. paratuberculosis metabolism in its natural host is limited. To identify M. avium subsp. paratuberculosis factors expressed in the host RNA and protein based approaches have been conducted. However, the outcome was limited regarding the characterisation of the metabolic situation of M. avium subsp. paratuberculosis in the host (Egan et al., 2008; Hughes et al., 2007; Janagama et al., 2010b; Wu et al., 2007b).

As stated above most if not all pathomechanism of M. avium subsp. paratuberculosis and other pathogenic mycobacteria have been deduced to a major extend from the murine and other small animal models or from in vitro systems using murine or human cell lines as well as primary cell (especially macophages) from the appropriate host. Since M. avium subsp.

paratuberculosis is obtainable in large numbers from the intestinal tissue of diseased cows the present study, using a comprehensive 2D DIGE and LC-MS/MS approach, could determine for the first time the metabolome of a pathogenic mycobacterium in its natural host. It was found that, at large, the central metabolism of M. avium subsp. paratuberculosis in the host

Chapter 4

differed only slightly from that in culture. Adaptation to the host environment included specific aspects regarding responses to antimicrobial host reactions, nutrient availability, and increased need for energy.

METHODS

Mucosa- and culture-derived M. avium subsp. paratuberculosis. The two strains (MW080610-2 and MW080821-2) were obtained from cows of different age (3.6 and 7.9 years) with clinical signs of Johne’s disease and have been described previously (Weigoldt et al., 2011).

Preparation of cytoplasmic fractions from M. avium subsp. paratuberculosis. Cell disruption and separation of mycobacterial cytoplasm from membrane fractions by high speed centrifugation were performed as described (Weigoldt et al., 2011). Protein concentration of

Im Dokument Antigen expression and metabolism of Mycobacterium avium subsp. paratuberculosis in vivo (Seite 31-0)