Species-Level Discrimination of Psychrotrophic Pathogenic and

Subscriber access provided by HACETTEPE UNIVERSITESI KUTUPHANESI

Article

Species Level Discrimination of Psychrotrophic Pathogenic and Spoilage Gram-Negative Raw Milk Isolates using a Combined MALDI-TOF MS Proteomics-Bioinformatics Based Approach Nuwan R. Vithanage, Thomas Richard Yeager, Jeevana Bhongir, Snehal R. Jadhav, Chaminda Senaka Ranadheera, Enzo A. Palombo, and Nivedita Datta J. Proteome Res., Just Accepted Manuscript • Publication Date (Web): 18 Apr 2017 Downloaded from http://pubs.acs.org on April 19, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Species Level Discrimination of Psychrotrophic Pathogenic and Spoilage Gram-Negative Raw Milk Isolates using a Combined MALDI-TOF MS Proteomics-Bioinformatics Based Approach

†, Ø

Nuwan R. Vithanage

†

§

†, Ø

, Jeevana Bhongir , Snehal R. Jadhav , Chaminda S. Ranadheera

, Enzo A.

Palombo§, Thomas R. Yeager *, ‡, €, Ø, Nivedita Datta †, €, Ø †

‡

College of Health and Biomedicine, Victoria University, Werribee, Victoria 3030, Australia; College

of Engineering and Science, Victoria University, Australia; §Faculty of Science, Engineering and €

Technology, Swinburne University of Technology, Hawthorn, Victoria 3122, Australia; Institute for Sustainability and Innovation, Victoria University, Werribee, Victoria 3030, Australia; ØAdvanced Food Systems, Victoria University, Werribee, Victoria 3030, Australia.

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 34

ABSTRACT Identification of psychrotrophic pathogenic and spoilage Gram-negative bacteria using rapid and reliable techniques is important in commercial milk processing, as these bacteria can produce heatresistant proteases and act as post-processing contaminants in pasteurised milk. Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) is a proven technology for identification of bacteria in food, however may require optimisation for identification of pathogenic and spoilage bacteria in milk and dairy products. The current study evaluated the effects of various culture conditions and sample preparation methods on assigning of raw milk isolates to the species level by MALDI-TOF MS. The results indicated that culture media, incubation conditions (temperature and time), and sample preparation significantly affected the identification rates of bacteria to the species level. Nevertheless, development of spectral libraries of isolates grown on different media using a web tool for hierarchical clustering of peptide mass spectra (SPECLUST) followed by a ribosomal protein based bioinformatics approach significantly enhanced the assigning of bacteria, with at least one unique candidate biomarker peak identified for each species. Phyloproteomic relationships based on spectral profiles were compared to phylogenetic analysis using 16S rRNA gene sequences and demonstrated similar clustering patterns with significant discriminatory power. Thus, with appropriate optimisation, MALDI-TOF MS is a valuable tool for species level discrimination of pathogenic and milk spoilage bacteria.

KEY WORDS: Gram-negative bacteria, MALDI-TOF MS, Culture condition, 16S rRNA gene sequencing, Phyloproteomic, Phylogenetic, Ribosomal proteins.

INTRODUCTION The presence of psychrotolerant microorganisms and their heat-stable enzymes in raw milk can cause bitterness, increased viscosity and gelation in milk and dairy products, resulting in spoilage and significant economic losses in commercial milk processing. 1, 2 Likewise, some Enterobacteriaceae isolates are considered important histamine producers in food that can cause histamine poisoning.3 In addition, the presence of pathogenic microorganisms in dairy products can result in food safety concerns. In general, the indigenous microbiota of products of dairy origin comprises a variety of Gram negative genera, such as Pseudomonas, Acinetobacter, Hafnia, Aeromonas, Serratia and


Page 3 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Klebsiella.4 Thus, identification of these bacteria in milk and dairy products and tracking their sources on-farm and in processing facilities using a rapid and reliable tool are of utmost importance.5, 6 Molecular methods based on housekeeping genes such as 16S rRNA, gyrB and rpoD genes are often used to discriminate psychrotrophic bacteria in milk. 4, 7 Nevertheless, sequencing and phylogenetic methods are costly and time-consuming often requiring skilled labour.5 These attributes make them less suitable for use in a routine high throughput dairy microbiology laboratory. Matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) is a rapid, reliable and cost-effective proteomics based approach that can be used as an alternative to phenotypic and molecular methods.8 To date, the technique has been successfully applied in various aspects of microbiology including

medical, food safety, food spoilage, environmental, counter-

terrorism and taxonomic studies 8, and is in agreement with the gold standard 16S rRNA gene sequencing as an identification tool. 9 MALDI-TOF MS sample preparation involves mixing bacterial proteins with a UV-absorbing organic compound called matrix such as α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxy benzoic acid (DHB), and 3,5-dimethoxy-4-hydroxycinnamic acid (sinapinic acid) that are also contain highly purified water, organic solvent (i.e. acetonitrile: ACN or ethanol) and a counter ion source (i.e. trifluoroacetic acid: TFA).

10

The co-crystallized proteins with matrix is then exposed to short laser pulses, which

convert bacterial proteins (or peptides) into singly charged ions [M+H] + by addition or loss of protons, without significant loss of sample integrity.10, 11 The protonated ions are then vaporized into the gas phase (desorption) and accelerated at a fixed potential in an electric field through high voltage, allowing them to move through a field-free flight tube, where these ions are separated from each other on the basis of their mass-to-charge ratio (m/z ).

10-12

Since the protein ions are singly charged, the 10, 11

spectral profiles generated for these bacteria depend on the molecular weight.

The charged

analyte are then detected and measured using a time of flight mass analyser, which determines the time required for it to travel the length of the flight tube.10, 11 Some TOF analysers incorporate an ion mirror at the rear end of the flight tube, which serves to reflect back ions through the flight tube to a detector, increasing the length of flight tube as well as correcting the small differences in energy among protonated ions.

13

A characteristic spectrum referred to as peptide mass fingerprint (PMF) is

generated for analytes in the sample, based on the TOF information.

10

The peaks representing



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 34

average or monoisotopic masses of bacterial proteins are then compared with reference profiles in a database for identification.

8, 14

Bacteria contain unique protein profiles in the m/z range of 1000 to 20000 Da, comprising various intracellular proteins. These are highly abundant ribosomal, housekeeping proteins and structural proteins in the bacterial cell, which represent about 60–70% of the dry weight of a microbial cell that have been used for discrimination of microorganisms and their tentative assignment at the genus, species or sub-group levels.9,

10, 12

Previous studies have indicated that changes in the culture

conditions, such as media, incubation time and incubation temperature, can significantly affect the spectral profiles of bacteria by MALDI-TOF MS.

12, 14-16

However, two recent studies used ribosomal

proteins coded by S10-spc alpha operon for the tentative assignment of Pseudomonas and Neisseria to the strain levels. 17, 18 Unlike structural proteins, ribosomal proteins are coded by the S10-spc alpha operon, which are constitutively expressed and comprise a highly conserved amino acid sequence similar to the 16S rRNA gene sequence, which is often used for taxonomic classification of bacteria.

17,18

Thus, the aims of the present study were, first, to investigate the effects of various

analytical (sample preparation methods), pre-analytical (culture conditions) and post-analytical (databases) factors on tentative assignment of raw milk isolates to the species level using MALDITOF MS and, second, to develop a highly reliable method for assigning of these bacteria at species level using a combined proteomic and bioinformatics-based approach to eliminate the effect of the culture conditions. Additionally, 16S rRNA gene sequencing was used to validate those assignments derived from MALDI-TOF MS analysis. MATERIALS AND METHODS Bacterial strains Representative proteolytic and pathogenic psychrotolerant Gram-negative bacteria belonging to several bacterial families, including (i) Pseudomonas (n=68) (ii) non-fermentative Gram-negative bacilli (NFGNB: n=54) and (iii) Enterobacteriaceae (n=60) that were previously isolated from raw milk were used for this analysis.4, 5 Standard strains from the American Type Culture Collection (ATCC) were also included in the analysis (Table 1). Culture conditions


Page 5 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In order to examine the impact of culture media on the generated MALDI-TOF MS spectra, the raw milk isolates were cultured on different solid and broth media, including: enriched media, i.e. sheep blood agar (SBA: tryptone soya agar supplemented with 5% sheep blood); skim milk agar (SMA); non-selective media, i.e., brain heart infusion agar (BHIA), nutrient agar (NA) and nutrient broth (NB); selective media, i.e. Pseudomonas selective agar (CFC: containing cetrimide, fucidin and cephaloridine as selective agents); differential media (MacConkey agar: MAC); and, minimal media (M9) (Table S-1). All media were purchased from the Media Preparation Unit, University of Melbourne, Australia. SBA was used as the reference media for MALDI-TOF MS analysis. 16S rRNA gene sequencing and phylogenetic analysis 5

The isolates were identified using 16S rRNA gene sequencing with universal primers. The sequence homogeneity was calculated using the maximum composite likely method (MEGA 6.0)19, 20, based on alignments from CLUSTAL W. A phylogenetic tree was constructed using the UPGMA method.21 The bootstrap values obtained were from 1000 iterations (< 50.0%). 22 MALDI-TOF MS analysis The MALDI TOF MS was conducted using two different methods: (i) direct smear method; and (ii) extraction method.

14

Direct smear method A small fraction of a bacterial colony was picked using a sterile toothpick and spotted onto a disposable Flexi Mass-DS (Shimadzu-Biotech, Japan) target plate, followed by overlaying of 1µl of matrix solution [10mg mL-1 α-cyano4-hydroxy-cinnamicacid (CHCA: BioMérieux, Marcy l'Etoile, France)] and allowing to air dry prior to analysis.16 Extraction method Five isolated colonies of each bacteria were used to prepare the lysates using ethanol, formic acid and acetonitrile (ACN) according to the method described by Anderson et al. (2012). Subsequently, the samples were each overlaid with 1µl of CHCA matrix solution and air-dried again.14 Assignment of unknown bacteria by MALDI-TOF MS Target plates were then loaded into the Axima Performance instrument and the mass spectra was generated according to the method described by Jadhav et al. (2015). Peak lists corresponding to the relative molecular mass to charge ratio (m/z) were generated in the range of 2000–20,000 Da. Each spectrum was obtained by accumulating 100 laser shots. Escherichia coli (ATCC 8739) grown on



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 34

SBA was used as an external calibrant strain for MALDI-TOF MS analysis. The resulting spectra were overlaid using the Launchpad software. Outliers were deleted from the data set, and then a single spectrum was selected from each overlapping set of spectra for analysis using Microsoft Excel 2010 (version 14.0; Microsoft Corporation, North Ryde, NSW, Australia). The raw spectra of the unknown bacteria were exported to Spectral Archive And Microbial Identification System (SARAMISTM: version 4.10, BioMérieux, Marcy l'Etoile, France) database to compare with reference spectra of the type strains using a standard pattern-matching algorithm, resulting in tentative assignment of unknown isolates to their accurate taxonomic level with a confidence score ranging from 75%-99.9%. The assignment of bacteria to the genus and species level was considered as reliable when confidence scores were ≥75.0%.

16

Samples were analysed

representing two biological (two separate time points) and two technical (two sub-samples each day) replicates (n = 4). Phyloproteomic analysis and Heat Map The consensus spectrum for each isolate grown in different media (n=4) using bacterial protein extract was constructed using the ‘peaks in common’ option in the freely available web-based application, SPECLUST (http://bioinfo.thep.lu.se/speclust.html). spectral

profiles

available

in

the

23

All spectra were compared with

SPECLUST

and

FoodBIMS

(http://bioinformatica.isa.cnr.it/Bact_Dbase.htm) databases.3, 9, 24 A hierarchical cluster analysis (HCA) was defined as a group of consensus spectra having spectral distances of ≤ 0.25. For each cluster, one or more isolates (depending on the cluster size) were selected and used in further investigation. The average linkage method was used to merge the two clusters with the smallest average of pairwise distances and the width in peak match score was set to 10 Da.

23

Heat maps were generated in Microsoft Excel 2010 (version 14.0; Microsoft Corporation,

North Ryde, NSW, Australia). Identification of peaks derived from MALDI-TOF MS based on MASCOT database searches Protonated ion masses [M+H]+ were used for searches in the NCBInr database taxonomically restricted to bacteria (eubacteria) with MASCOT sequence query according to the method described before. 25 with minor modification. The parameters were usually set as follows: enzyme: no-cleave, missed cleavages: 0, protein mass: unrestricted, product ion matches: b- and y-type ions, average peptide ion mass tolerance: ±0.3 Da, average production mass tolerance: ±1.5 Da. If partial sequence


Page 7 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


information obtained from bacterial peptides resulted in no database search hit, additional MASCOT search queries were conducted by allowing up to 1 and 2 missed cleavage sites and by stepwise increasing peptide and product ion mass tolerances to ± 1.5 Da and ± 2.0 Da, respectively. The identity of the peaks are given with a Protein score which is calculated using 10*Log (P), where P is the probability that the observed match is a random event, and Protein scores greater than 68 are significant (P

Species-Level Discrimination of Psychrotrophic Pathogenic and

Recommend Documents