Supplementation of Lactobacillus plantarum Improves Markers of

990,256 reads were retained based on the following criteria: (1) no errors in .... P = 0.13, effect size = 0.9; Isoleucine: P = 0.13, effect size = 1...
0 downloads 0 Views 1MB Size
Subscriber access provided by - Access paid by the | UCSB Libraries

Article

Supplementation of Lactobacillus plantarum improves markers of metabolic dysfunction induced by a high fat diet Alice Martinic, Javad Barouei, Zach Bendiks, Darya Mishchuk, Dustin D Heeney, Roy Martin, Maria L. Marco, and Carolyn M Slupsky J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.8b00282 • Publication Date (Web): 22 Jun 2018 Downloaded from http://pubs.acs.org on June 23, 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 56 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

Supplementation of Lactobacillus plantarum improves markers of metabolic dysfunction induced by a high fat diet

Alice Martinic1, Javad Barouei2, Zach Bendiks2, Darya Mishchuk1,2, Dustin D Heeney2, Roy Martin3, Maria L Marco2, Carolyn M Slupsky1,2*

1

Department of Nutrition, University of California, Davis. 2

3

Department of Food Science and Technology

Western Human Nutrition Research Center, USDA, Davis, CA

*Corresponding Author [email protected] 530-219-5757

1 ACS Paragon Plus Environment

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 56

ABSTRACT

Obesity is a prevalent chronic condition in many developed and developing nations that raises the risk for developing heart disease, stroke, and diabetes. Previous studies have shown that consuming particular probiotic strains of Lactobacillus is associated with improvement in the obese and diabetic phenotype; however, the mechanisms of these beneficial effects are not well understood. In this study, C57BL/6J male mice were fed a lard-based high fat diet for 15 weeks with Lactobacillus plantarum supplementation NCIMB8826 (Lp) between weeks 10 and 15 (n=10 per group). Systemic metabolic effects of supplementation were analyzed by NMR metabolomics, protein expression assays, gene transcript quantification, and 16S rRNA marker gene sequencing. Body and organ weights were not significantly different with Lp supplementation and no microbiota community structure changes were observed in the cecum; however, L. plantarum numbers were increased in the treatment group according to culturebased and 16S rRNA gene quantification. Significant differences in metabolite and protein concentrations (serum, liver, and colon), gene expression (ileum and adipose), and cytokines (colon) were observed between groups with increases in the gene expression of tight junction proteins in the ileum and cecum, and improvement of some markers of glucose homeostasis in blood and tissue with Lp supplementation. These results indicate Lp supplementation impacts systemic metabolism and immune signaling before phenotypic changes and without large-scale changes to the microbiome. This study supports the notion that Lp is a beneficial probiotic, even in the context of a high fat diet.

Keywords: Lactobacillus plantarum, metabolism, microbiome, high-fat diet

2 ACS Paragon Plus Environment

Page 3 of 56 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 Poor nutrient intake has been linked with numerous conditions including obesity, diabetes, high blood pressure, heart disease, inflammation and liver injury, and neurological conditions1, 2. Over the past 60 years, substantial changes in agricultural practices and food processing have resulted in more shelf-stable and readily available foods in the typical North American diet that, with the added sugar and fats, now account for >50% of the energy with a concomitant decrease in essential nutrients. These dietary changes coincide with the increase in the prevalence of overweight adults in the United States, which has more than doubled since the 1960’s3 with an astounding almost four-fold increase in overweight children4. It is now understood that the gut microbiota plays a substantial role in our health, and associations have been made between the gut microbiota and the above mentioned conditions (reviewed by Postler et al.5 and Tse et al.6). In particular, the dysbiotic gut microbiome of humans on high fat and high sugar diets is associated with increased subclinical systemic inflammation and impaired barrier function. However, the mechanisms leading to the increased inflammation and impaired barrier function are not well-understood7. Recent meta-analyses have described the benefits of probiotic Lactobacillus supplementation on weight loss and glycemic control in human studies8, 9. In rodents, probiotic Lactobacillus have been linked with reduced risk factors for type 2 diabetes mellitus (T2DM) and metabolic syndrome (MetS), such as decreasing serum insulin and systemic low-grade inflammation10-12. Probiotics support intestinal homeostasis likely through mechanisms involving modification to the intestinal metabolome and changes in the innate and adaptive immune response of the host that ultimately strengthens the intestinal barrier13, 14. However, the effects arising from the use of probiotics are variable such that probiotic supplementation can result in decreased body weight 3 ACS Paragon Plus Environment

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 56

or BMI whereas other human studies did not find any effect (reviewed by Kobyliak et al.15 and Seganfredo et al.16). Lactobacilli are an essential part of the microbiota in the mouth, gastrointestinal (GI), and genital tracts of humans and animals, and have been associated with food production and preservation for thousands of years17. The genus Lactobacillus includes over 200 species that are recognized for their fermentative metabolism18. Of these species, 12 are considered Generally Recognized as Safe (GRAS) for use as probiotics by the FDA. L. plantarum is one of the most well-studied Lactobacillus species due to its versatility which allows for its distribution in a wide variety of environments such as fermented foods and the human body. Its metabolic capacity is characterized by lactic acid production during rapid growth on mono- and disaccharides and adaptation to conditions in the GI tract by the modular expression of distinct extracellular proteins19-22. There is a growing body of evidence detailing its health benefits in both animals and humans17. L. plantarum WCFS1 was shown to induce cytokine production by immune cells in vitro23,

24

and increase the expression of immune response pathways25 and epithelial tight

junction proteins26, 27 in the human duodenum. L. plantarum NCIMB8826 (Lp) was shown to survive transit through the human GI tract21, 28, 29, but animal studies suggest that the host diet is one of the primary factors in how effective Lp supplementation will be30, 31. The present study investigates the systemic metabolic effects of Lp supplementation in the context of a high fat diet. Addition of Lp in this context with systemic metabolic analysis has not been previously investigated. Characterization of the gene expression of gut barrier proteins, lipid metabolism, insulin signaling, and inflammation in response to the supplementation help to elucidate the mechanisms by which Lp exerts beneficial effects.

4 ACS Paragon Plus Environment

Page 5 of 56 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

MATERIALS AND METHODS Animal study design. The mouse

Table 1. High-fat diet used in this study

study was conducted under approval of the UC Davis Animal Care and Use

Ingredients (g/kg)

HF* (TD.130588)

Committee (protocol #17500). A total

Casein

195.0

L-Cystine

3.0

AMIOCA™ Corn Starch†

280

Maltodextrin

77.4

Sucrose

80.0

Cellulose

79.0

Lard

195.0

Soybean Oil

30.0

Mineral Mix, AIN-93G-MX (9404)

43.0

Vitamin Mix, AIN-93-VX (94047)

15

Choline Bitartrate

2.5

Food color

Blue 0.1

of 20 C57BL/6J male mice at six weeks old were obtained from the Jackson

Laboratory

(The

Jackson

Laboratory, Sacramento, CA, USA). Mice were singly housed in cages lined with a 4:1 ratio of TEK-Fresh Laboratory Animal Bedding (Harlan, Madison, (Shepherd Watertown,

WI)

and

paper

specialty TN).

Animals

chips papers, were

maintained on a 12 h light/dark cycle

Nutrient Information

% by weight

% of kcal

Protein

17.3

15.5

Carbohydrate

43.0

38.6

Fat

22.7

45.9

kcal/g

4.5

* HF, high-fat diet †

(lights on at 06:00 am) and given free

AmiocaTM (Ingredion Incorporated, Bridgewater, NJ) consists of primarily amylopectin (100% glycemic starch).

access to food and water for the duration of the study. The mice were acclimated to the housing for two weeks while maintained on a standard Harlan/ENVIGO mouse chow (Harlan Laboratories, Inc., Madison, WI). The mice were then fed a high-fat (HF) diet (Harlan, Madison, WI) (Table 1) for nine weeks. After nine weeks on the HF diet, the mice were divided into two groups (10 mice per group) and continued on the HF diet for another six weeks (Table 1). During the last six weeks of the study, on alternate days, one group of HF-fed mice received regular 5 ACS Paragon Plus Environment

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 56

supplements of 1.2 x 109 cells of L. plantarum NCIMB8821 (group referred to as HF + Lp) (see below for details). Mouse weight and food intake were monitored every other day for the duration of the 15-week study. Mice were not fasted overnight before sacrifice but were sacrificed in random order to ensure time of sacrifice did not skew results. At sacrifice, mice were anesthetized under 2% isoflurane and blood was collected by cardiac puncture following diaphragotomy. Termination was then ensured by cervical dislocation. Epididymal and retroperitoneal adipose tissues (eAT and rAt), liver, spleen tissues, and ileal, cecal and colonic tissues and contents were collected, weighed, snap-frozen in liquid nitrogen, and stored at −80 °C until further analysis. L. plantarum feeding. Mice were fed a naturally Rifampicin-resistant variant of Lp NCIMB8826 (the parental strain of strain WCFS132) as previously described30. Briefly, a fresh overnight culture of Lp NCIMB8826-R grown at 37 °C in de Man Rogosa and Sharpe (MRS) (BD Biosciences, San Jose, CA, USA) supplemented with 50 µg/mL Rifampicin (Fisher Scientific, Fair Lawn, NJ, USA) was washed once with sterile PBS buffer. Bacterial cells were then concentrated in sterile PBS to achieve a bacterial concentration of 6×1010 cells/mL. Mice in the Lp feeding group (HF + Lp) received 20 µL of the Lp suspension via mouth feeding using gavage needles (Gage 20) every other day. Mice in the control HF subset received 20 µL phosphate buffered saline (PBS) vehicle with no Lp added. Fresh fecal pellets were collected 24 and 48h after each bacterial administration. The pellets were weighed and diluted with PBS in a ratio of 1:30 (w:v), and homogenized using four to six 2 mm-glass beads in a Fast Prep 24 instrument (MP Biomedicals, Solon, OH, USA) at a speed setting 4 for 10 seconds. Fecal suspensions were then serially diluted, plated onto MRS-rifampicin (50 µg/mL) agar and incubated at 37 °C for 48 h prior to colony enumerations. 6 ACS Paragon Plus Environment

Page 7 of 56 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

RNA extraction and complementary DNA synthesis. Total RNA was isolated from liver, ileal, cecal and colonic tissues using the RNeasy Mini kit according to manufacturer’s instructions (Qiagen Inc., Valencia, CA, USA). The RNeasy Lipid Tissue Mini kit (Qiagen) was used to extract RNA from eAT. Approximately 20-30 mg (100 mg for fat pads) of the tissue was cut on dry ice and immediately placed into a pre-chilled 2 mL microfuge tube containing 1.4 mm Matrix D ceramic beads (MP Biomedicals, Solon, OH, USA) and 600 µL RLT lysis buffer (RNeasy mini kit) or 1 mL QIAzol lysis reagent (Qiagen) for fat pads. The tissues were then disrupted and homogenized in a Fast Prep 24 instrument (MP Biomedicals, Solon, OH, USA) at speed setting 6 for 40 seconds (or 2×30 s for fat pads with 1 min cooling rest on ice between the two runs). RNA extraction/purification protocols were then followed as instructed. Purified RNA was treated with DNase using TURBO DNA-free™ Kit (Ambion, Austin, TX, USA) following the manufacturer's instructions. The quantity and quality of RNA was determined using the NanoDrop 2000c (Thermo Fisher Scientific, Wilmington, DE, USA) and the Agilent 2100 Bioanalyzer (Agilent Technologies, Waldbronn, Germany), respectively. RNA integrity numbers (RIN) were between 7.1 − 9.9. The RNA was immediately stored at –80 °C until use. Firststrand complementary DNA (cDNA) was generated using the M-MLV reverse transcriptase and random decamers (Ambion, Grand Island, NY, USA) according to the manufacturer’s instructions. cDNA was then stored at –20 °C. Real-time quantitative PCR. Real-time quantitative PCR (qPCR) was performed using a 7500 Fast Real-time PCR system (Applied Biosystems, Carlsbad, CA, USA). Each reaction mix (20 µL) consisted of approximately 2.5 ng cDNA template, 10µL Fast SYBR Green Master Mix (Applied Biosystems), 200 nmol of each forward and reverse primer33 and nuclease free water (Ambion). All reactions were performed in duplicate. Amplification was initiated at 95 °C for 20 7 ACS Paragon Plus Environment

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 56

s, followed by 40 cycles of 95 °C for 3 s (denaturation), and 60 °C for 30 s (annealing extension). Primer specificity was assessed by adding a melting curve step at the end of amplification. Data were analysed using the 2-∆∆Ct method34 using HF group (mean) as the reference condition. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) or ribosomal protein L19 (RPL19) were used for intestinal or non-intestinal tissues respectively for transcript normalization. Protein measurements in blood, colon, and adipose tissue. Blood was incubated at room temperature for 30 min, centrifuged at 2,350 ×g for 10 min, and the supernatant (serum) was collected and frozen at –80 °C until further analysis. Metabolic markers (amylin active, Cpeptide 2, ghrelin, glucose-dependent insulinotropic polypeptide (GIP), glucagon, IL-6, insulin, leptin, monocyte chemoattractant protein-1 (MCP-1), pancreatic polypeptide (PP), peptide YY (PYY), resistin, and TNF-α) were measured in the sera using a Millipore’s MILLIPLEX® MAP Mouse Metabolic panel kit (EMD Millipore Corp. Billerica, MA, USA) according to the manufacturer’s instructions. Amylin active, ghrelin, Il-6, MCP-1, glucagon, PYY and TNF-α were all below the limit of detection and are not reported. Serum adiponectin was also measured using an EMD Millipore’s MILLIPLEX MAP Mouse Adiponectin Magnetic Bead kit (EMD Millipore). Approximately 200 mg of eAT or 50 mg of proximal colon tissue from each mouse was cut on dry ice, weighed, and placed into a pre-chilled 2 mL microfuge tube containing 1.4 mm Matrix D ceramic beads (MP Biomedicals). Tissues were suspended in five volumes of cold PBS (137 mM sodium chloride, 2.7 mM potassium chloride, 1.44 mM sodium phosphate dibasic and 1.8 mM potassium phosphate monobasic) containing a protease inhibitor cocktail (Sigma, St. Louis, MO, USA) and 1% Triton X-100 (Fisher Scientific, Fair Lawn, NJ, USA). The tissues were then 8 ACS Paragon Plus Environment

Page 9 of 56 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

homogenized twice in a Fast Prep 24 instrument (MP Biomedicals) for 30 s at speed setting 6.5 with 1 min cooling on ice between the two runs. The extracts were centrifuged at 14,000 ×g for 10 min at 4 ºC, and the supernatants were collected and stored at −80 °C until analysis. Adipose tissue concentrations of adiponectin, IL-6, leptin, MCP-1, PAI-1, resistin and TNFα were quantified using a Mouse Adipocyte Magnetic Bead kit (EMD Millipore). The colon tissue concentrations of IL-1β, IL-2, IL-4, IL-5, IL-6, IL-10, IL-12p70, IL-13, IL-15, IL-17A, IL-25/IL17E, IL-17F, IL-21, IL-22, IL-23, IL-27, IL-28B, IL-31, IL-33, CD40L, GM-CSF, IFNγ, MIP3α, TNFα, and TNFβ were measured using a Mouse Th17 Magnetic Bead kit (EMD Millipore). IL-1β, IL-2, IL-4, IL-5, IL-6, IL-10, IL-12p70, IL-13, IL-15, IL-17F, IL-21, IL-22, IL-27, IL28B, IL-31, CD40L, GM-CSF, IFNγ, MIP-3α, and TNFα, were all below the limit of detection, and are thus not reported. All assays were run using a Bio-plex Magpix multiplex reader (BioRad), and data were analysed with the Bio-Plex Manager software Ver. 6.1. When concentrations were below the limit of detection, a value of 0 was assigned for the concentration. The colon tissue from mouse HF4 was lost during extraction, resulting in n=9 for the HF group. Metabolome. Urine and serum samples were prepared as previously described35. In brief, samples were defrosted from –80 °C and filtered through a 3,000 MW cutoff filter (Amicon, Millipore Sigma). The filtrate volume was adjusted to 207 µL with ultrapure water for small sample volumes. 23 µL of 5 mM 3-(trimethylsilyl)-1-propanesulfonic acid-d6 (DSS-d6) internal standard in 100% D2O was added and pH was adjusted to 6.85 ± 0.1. 180 µL aliquots were transferred to 3 mm Bruker NMR tubes and stored at 4 °C until acquisition. Contents from the cecum and colon were freeze dried and weighed. Liver samples were cut on dry ice and weighed to approximately 200 milligrams. Liver tissue was homogenized with 1.4 mm ceramic spheres for one minute (MP Biomedicals FastPrep®-24 and Lysing Matrix D). For 9 ACS Paragon Plus Environment

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 10 of 56

cecal content, colon content, and liver samples, polar metabolites were separated with a 2:1 methanol:chloroform extraction and the methanol/water layer was dried. Polar metabolites were reconstituted into 10 mM phosphate buffer and centrifuged at 10,000 x g for 10 minutes to pellet debris. 207 µL of supernatant and 23 µL of a DSS-d6 internal standard were combined, pH was adjusted to 6.85 ± 0.1, and samples were prepared for NMR as described above. Proton (1H) NMR spectra were collected on a Bruker NMR spectrometer operating at a proton NMR frequency of 600 MHz that was equipped with a SampleJet autosampler using the noesypr pulse sequence as previously described. The acquisition temperature was 298K for all sample types35. Metabolites were quantified using NMRSuite v8.1. Profiler (Chenomx, Inc.) as described36. All compounds have been verified, and have shown to be reproducible and accurate37, 38. Urine samples were collected in the morning and evening for two consecutive days and pooled. An approximately 40:60 ratio of morning to evening collection was combined to account for diurnal variation and concentrations were normalized to creatinine. Concentrations are expressed as nanomole per gram dry weight of cecal or colon material, nanomole per gram of wet weight of liver tissue, and micromole per liter of urine or serum. Eight serum samples (four from each of the groups (HF or HF + Lp)), and seven urine samples (four from the HF and three from the HF + Lp group) were omitted from analysis due to insufficient sample volume. Two colon samples were omitted in the HF + Lp group: one due to no visible contents, and one due to sample contamination. Final samples numbers for analysis by sample type were: cecum (HF, n=10; HF + Lp, n=10), colon (HF, n=10; HF + Lp, n=8), serum (HF, n=6; HF + Lp, n=6), liver (HF, n=10; HF + Lp, n=10), urine (HF, n=6; HF + Lp, n=7). Total metabolites quantified were: (cecum, 28; colon, 28; serum, 44; liver, 45, urine, 46). 10 ACS Paragon Plus Environment

Page 11 of 56 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

Bacterial diversity analysis. DNA was extracted using the QIAamp fast DNA stool mini kit (Qiagen, Hilden, Germany) with modifications. Cecal contents or feces were suspended in 100 µL lysis buffer consisting of 200 mM NaCl, 100 mM Tris-HCl, 20 mM EDTA and 20 mg/mL lysozyme, and incubated at 37 ºC for 30 min. Prior to purification, the suspension was mechanically disrupted in InhibitEX buffer (Qiagen) and 300 mg of 0.1 mm zirconium beads (BioSpec Products, Inc. Bartlesville, OK, USA) in a FastPrep-24 instrument (MP Biomedicals LLC. Santa Ana, CA, USA) run twice at 6.5 m/s for 1 min. DNA was stored at -20 ºC. The V4 region of the 16S rRNA gene was amplified using barcoded F515 primers and the R806 primer39,

40

. Each 50 µL reaction contained 5 ng DNA template, 0.8 U Ex Taq DNA

polymerase (Takara Bio Inc. Kyoto, Japan), 1X Ex Taq buffer, 2.5 mM MgCl2, 200 µM dNTPs and 0.1 µM of each primer. Amplification was initiated at 94 ºC for 3 min, followed by 25 cycles of 94 ºC for 45 s, 50 ºC for 60 s and, 72 ºC for 40 s, prior to a final extension at 72 ºC for 10 min. A negative control was included in each PCR run to confirm the absence of contamination. The concentration of each PCR product was measured by Quant-iT PicoGreen dsDNA (Invitrogen, Waltham, MA, USA) according to the manufacturer’s recommendations. Approximately 50 ng of each amplicon was pooled and purified using Wizard® SV Gel and PCR Clean-Up System (Promega, Fitchburg, WI, USA). The purified amplicon mixture was used for DNA library preparation and paired-end Illumina Mi-Seq (PE250) sequencing (Illumina Inc., San Diego, CA) at the UC Davis Genome Center (http://dnatech.genomecenter.ucdavis.edu/). DNA sequence analysis was performed using the pipeline Quantitative Insights Into Microbial Ecology (QIIME) version 1.9.041. The forward and reverse Illumina reads were assembled using the join_paired_ends.py script with a minimum overlap of 100 bp allowing 1 % differences within the overlap region. Only joined-end reads were used for further analysis. After extracting 11 ACS Paragon Plus Environment

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 12 of 56

barcode sequences and trimming the primers, reads were demultiplexed using the split_libraries_fastq.py script. 990,256 reads were retained based on the following criteria: 1) no errors in barcode, 2) no ambiguous bases in sequence, 3) possess at least the minimal acceptable Phred quality score of 30. USEARCH 6142 was used to identify and remove 15,752 chimeric sequences. The remaining high quality reads with a median sequence length of 294 bp were clustered into Operational Taxonomic Units (OTUs) based on 97% sequence similarity against the Greengenes reference database version 13_843 using a uclust-based method. A representative sequence from each OTU was used to build a phylogenetic tree for downstream UniFrac distance measurements, and these sequences were aligned by the RDP-based consensus taxonomy assigner for taxonomic assignment. OTUs comprising