Isobaric Labeling Quantitative Metaproteomics for ... - ACS Publications

demonstrated herein show the significant improvement in quantitative ...... Janda, J. M.; Abbott, S. L., 16S rRNA gene sequencing for bacterial identi...
0 downloads 0 Views 931KB Size
Subscriber access provided by University of Rhode Island | University Libraries

Article

Isobaric Labeling Quantitative Metaproteomics for the Study of Gut Microbiome Response to Arsenic Chih-Wei Liu, Liang Chi, Pengcheng Tu, Jingchuan Xue, Hongyu Ru, and Kun Lu J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.8b00666 • Publication Date (Web): 14 Dec 2018 Downloaded from http://pubs.acs.org on December 16, 2018

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 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 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.

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 43 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

Isobaric Labeling Quantitative Metaproteomics for the Study of Gut Microbiome Response to Arsenic Chih-Wei Liu,1,# Liang Chi,1,# Pengcheng Tu1, Jingchuan Xue1, Hongyu Ru2, Kun Lu1,* 1.Department of Environmental Sciences and Engineering, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599, United States 2. Department of Population Health and Pathobiology, North Carolina State University, Raleigh, North Carolina 27607

#

Chih-Wei Liu and Liang Chi have contributed equally to this work

Corresponding Author * Kun Lu, Phone: 919 966 7337; e-mail: [email protected].

ACS Paragon Plus Environment

1

Journal of Proteome Research 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 43

ABSTRACT

Quantitative metaproteomics is a relatively new research field by applying proteomics technique to study microbial proteins of microbiome, and holds the great potential to truly quantify the functional proteins actually expressed by microbes in the biological environment such as gastrointestinal tract. The significant association between arsenic exposure and gut microbiome perturbations has been reported; however, metaproteomics has not yet been applied to study arsenic induced proteome changes of microbiome. Most importantly, to our knowledge, isobaric-labeling based large-scale metaproteomics has not been reported using the advanced database search approaches such as MetaPro-IQ and matched metagenome database search strategies to provide high quantification accuracy and less missing quantification values. In the present study, a new experimental workflow coupled with isobaric labeling and MetaPro-IQ was demonstrated for metaproteomics study of arsenic induced gut microbiome perturbations. The advantages of this workflow were also discussed. For all 18 fecal samples analyzed, 7,611 protein groups were quantified without any missing values. The consistent results of expression profiles were observed between 16S rRNA gene sequencing and metaproteomics. This isobaric labeling based workflow demonstrated the significant improvement

of quantitative

metaproteomics for gut microbiome study.

KEYWORDS. Metaproteomics, gut microbiome, arsenic, isobaric labeling, tandem mass tags, label free, MetaPro-IQ

ACS Paragon Plus Environment

2

Page 3 of 43 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

INTRODUCTION Chronical exposure with arsenic especially through drinking water is a seriously and globally public health issue, however the molecular mechanisms remain unclear, and several hundreds of millions of people are suffering the diseases induced by arsenic toxicity including multiple types of cancers (1, 2). Recently, the association between arsenic exposure through drinking water and gut microbiome perturbations such as the changes of profiles and compositions of microbiome and secreted metabolites has been observed by metagenomics and metabolomics strategies (3-6). Since human microbiome project launched in 2008 (7), microbiome has been attracted a lot of attention and efforts to address the complicated interactions between host and microbe (8, 9), especially using metagenomic techniques such as 16S rRNA gene sequencing (10) and nextgeneration sequencing (11). In contrast with gene level study using metagenomics, metaproteomics should truly reflect the functional protein products actually expressed by microbes. However, metaproteomics research has significantly lagged behind the tremendous progress of metagenomics on microbiome study, mainly due to the lack of a suitable database for database search in which only the sequences of all potentially expressed proteins in the analyzed samples are included (12). Several remarkable advances have recently been made for metaproteomics, for example, a two-step database search approach (13) and matched metagenome database (14) generated from shotgun metagenomics to greatly increase the sensitivity of peptide/protein identification rates. In that two-step database search, a smaller refined database is generated in the first search to include all the possible microbial sequences within the analyzed samples. Next, this refined database is further used for second search in target-decoy database search by including falsediscovery rate cut off. Of note, additional cost is needed to obtain this matched metagenome

ACS Paragon Plus Environment

3

Journal of Proteome Research 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 43

database which limits the wide applications in metaproteomics, and it is hard to financially support the cost for sequencing the microbial genes from all analyzed samples in the general laboratory. Recently, large-scale metagenomics has been applied by Xiao et al. to analyze the fecal samples collected from 184 mice and their results revealed the significantly different microbiome gene profiles from different mouse strains, animal providers, housing facilities, diet supply and gender (15). In addition, they also found only 4% of similarity of microbial genes between mouse and human gut microbiome. Those observations highlight the importance of “sample-specific” database used for metaproteomics database search. Zhang et al. recently proposed a universal database search approach for metaproteomics study, MetaPro-IQ, in which three-step searches are sequentially conducted for generation of “samplespecific” database that is further used for protein identification and quantification (12). In MetaPro-IQ, “sample-specific” database is generated from the original gut microbial gene catalog using X! Tandem algorithm (15, 16), and then protein identification and quantification are performed against this “sample-specific” database by advanced MaxLFQ algorithm in MaxQuant (17). Using this approach, the largest number of protein groups identified from gut metaproteome has been reported to date (12, 18). This new approach is very useful for identification of gut microbial proteins; however, protein quantification is still limited to labelfree quantification (LFQ) which is generally considered as low quantification accuracy compared to labeling-based quantification (19, 20). Moreover, the missing quantification value issue resulted from data-dependent MS/MS acquisition impedes the statistical analysis to reveal the proteins with truly differential expressions (21-23). Isobaric labeling-based quantitative proteomics, such as tandem mass tag (TMT), has been used for large-scale relative quantification of proteins with multiple advantages including high

ACS Paragon Plus Environment

4

Page 5 of 43 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

accuracy, less missing quantification values, and multiplexing capability to increase analytical throughput (24-27). This work aims to develop an isobaric-labeling based workflow for the study of quantitative metaproteomics and to demonstrate, for the first time, the feasibility of this workflow for mice gut microbiome response to arsenic. In this new workflow, label free samples are used for construction of “sample-specific” database, and, on the other hand, TMT-10plex labeled samples are utilized for accurate quantification of microbial proteins. The results demonstrated herein show the significant improvement in quantitative metaproteomics and its great potential for microbiome study.

ACS Paragon Plus Environment

5

Journal of Proteome Research 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 43

EXPERIMENTAL SECTION All reagents and chemicals used in this study were purchased from Sigma Aldrich (St. Louis, MO), unless otherwise stated. The kits for BCA protein assay and TMT-10plex labeling and sodium arsenite were obtained from ThermoFisher Scientific (Rockford, IL). Sequencing-grade trypsin was purchased from Promega (Madison WI). Specific-pathogen-free grade C57BL/6J female mice (approximate 7 weeks old) were purchased from Jackson Laboratories (Bar Harbor, ME). All solvents used were at least HPLC-grade. Fecal samples collected from control and arsenic-treated mice. Mice were maintained in the animal facility of University of North Carolina at Chapel Hill under standard environmental conditions (22 °C, 40-70% humidity, and a 12:12 light:dark cycle), and purified rodent diet was provided. All mice were treated humanely as described in the approved animal protocol. Mice were kept in the facility for one week after received before starting experiment. The background variations of gut microbiome between mice were normalized by fecal transplant from three mouse donors. The mice were randomly switched between cages weekly for 4 weeks to further minimize individual differences of gut microbiota. Next, the mice were randomly divided into the control and arsenic-treated groups (6 cages for each group and 3 mice in each cage). In this study, the mice were administered with sodium arsenite in the drinking water, freshly prepared twice a week, at a concentration of 250 ppb on the basis of arsenic for an experimental period of 3 months (13 weeks). The fecal samples were collected before and after exposure, and stored in -80 °C until microbial protein extraction. All animal experiments were approved by the University of North Carolina on Animal Care. 16S rRNA gene sequencing and data analysis.

ACS Paragon Plus Environment

6

Page 7 of 43 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

DNA was isolated from fecal samples using a PowerSoil® DNA Isolation Kit according to manufacturer’s instructions (MO BIO Laboratories, Carlsbad, CA). DNA was amplified using universal primers (iTRU-A 515 F and iTRU-1 806 R) to target the V4 region of bacterial 16S rRNA (3). The individual samples were barcoded and pooled to construct the sequencing library, and sequenced by Illumina MiSeq to generate pair-ended 250 × 250 reads. The pair-ended reads were analyzed using quantitative insights into microbial ecology (28). Operational taxonomic units (OTUs) were annotated and classified, and UCLUST was used to determine OTUs under a threshold of 97% sequence similarity. Next, a representative set of sequences was selected from each OTU for taxonomic identification using the ribosomal database project (RDP) classifier. The latest Greengenes OTU reference sequences with 97% sequence similarity were used as the training sequences for RDP (29). Principal coordinate analysis (PCoA) was performed to evaluate intrinsic clusters within the observations with the β-diversity as the metrics. Metaproteomic sample preparation. The microbial protein extraction from mouse fecal samples was prepared by differential centrifugation method described previously with minor modifications (18). Fecal samples were resuspended in cold phosphate buffer (PBS) through vortexing, and then fecal slurry was centrifuged (300 g, 4 °C for 5 min) to collect supernatant. The pellets were subjected to additional three times of resuspension and all supernatants were pooled. The pooled supernatant was undergone to another three centrifugation events (300 g, 4 °C for 5 min) to remove remaining debris or large particles. The resulting supernatant was collected for high-speed centrifugation (14,000 g, 4 °C for 20 min) to pellet the microbial cells followed by additional three wash steps with PBS.

ACS Paragon Plus Environment

7

Journal of Proteome Research 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 8 of 43

The pelleted microbial cells were resuspended with 4% sodium dodecyl sulfate (SDS) and 6 M urea in 50 mM Tris-HCl buffer (pH 8.0) by vortexing before mechanical disruption of bacterial cells by TissueLyser (50 Hz for 5 min, QIAGEN) and sonication in water bath for 3 min. The cell debris was further removed by centrifugation (16,000 g, 4 °C for 10 min). Microbial proteins were precipitated using acidified acetone/ethanol (5-fold volume, 50:50 (v/v), with 0.1% acetic acid) at -20 °C for overnight. The microbial proteins were spun down by centrifugation (16,000 g, 4 °C for 20 min) followed by washing with cold acetone for three times. The washed microbial protein pellets were resuspended in 6 M urea in 50 mM ammonium bicarbonate (ABC, pH 8.0) before protein quantification by BCA assay. The proteins were sequentially reduced and alkylated by dithiothreitol (10 mM) and iodoacetamide (40 mM) for 1 hr at 37 °C, and then the alkylation was quenched by adding dithiothreitol for 15 min at 37 °C. The protein solution was further diluted with 50 mM ABC to reduce urea concentration (