1. Project title and ADF file number.

ADF Project Final Report Format

1. Project title and ADF file number. 20140027, Investigation of Avenanthramides, a Type of New Healthy Compounds in Oat

2. Name of the Principal Investigator and contact information. Xiao Qiu, Department of Food and Bioproduct Sciences, University of Saskatchewan

3. Name of the collaborators and contact information.

4. Abstract/ Summary: This must include project objectives, results, and conclusions for use in publications and in the Ministry database. Maximum of 300 words in lay language.

Avenanthramides are a group of N-cinnamolyanthranilic acids comprising anthranilic acid and cinnamic acid connected by an amide linkage with health-promoting properties mainly found in oat (Avena sativa L.). In this research, avenanthramide A, B and C (Avn-A, B and C), the three most abundant avenanthramides (Avns) in oat, were identified and quantified from oat varieties. Subsequently, in vitro antioxidant activities of oat extracts and Avn-A, B and C were evaluated, and Avn-C had the highest in vitro antioxidant activity among the three avenanthramides. To investigate the cytoprotective activity of Avns, normal human skin fibroblasts (2DD) were treated with Avn C followed by exposure to extracellular stress and its ability to reduce cellular damage was determined. Pre-treatment of cells with Avn-C reduced hydrogen peroxide (H2O2)-induced oxidative stress significantly as demonstrated by decreased intracellular free radical levels and antioxidant gene transcripts. Avn-C pre-treatment also resulted in decreased levels of gene transcripts encoding pro-inflammatory cytokines in response to H2O2 or tumor necrosis factor α (TNF-α) stimulation. This reduction in cytokine gene transcription occurred concomitantly with reduced phosphorylated nuclear factor-κB (NF-κB) p65, indicating reduced pro-inflammatory response. To better understand the mechanisms of actions, the impact of Avn-C on cellular signaling pathways was investigated on Avn C-treated 2DD cells without exposure to stress. Avn-C was found to induce heme oxygenase-1 (HO-1) expression through increased DNA-Nrf2 binding activity. Also, it reduced basal levels of pro-inflammatory cytokines through decreased DNA-NF-κB binding activity. Moreover, anti-proliferative effect of Avn C on 2DD cells was observed via mechanisms independent of autophagy activation. Collectively, our findings suggest that Avn-C protects normal human skin fibroblasts against oxidative stress and inflammatory response through Nrf2/HO-1 activation and NF-κB inhibition.

Despite the health importance of avenanthramides, the biosynthesis of these compounds in oat is not completely understood. In the present research, we also identified and characterized three different types of genes from oat encoding 4-coumarate-CoA ligase (4CL), hydroxycinnamoly-CoA:hydroxyanthranilate N-hydroxycinnamoly transferanse (HHT) and caffeoyl-CoA O-methyltransferase (CCoAOMT)-like enzyme involved in the complete biosynthetic process of these avenanthramides. In vitro enzymatic assays using the proteins expressed in Escherichia coli as enzyme sources showed that oat 4CL could activate 4-coumaric acid, caffeic acid and ferulic acid to their CoA thioesters, respectively. Oat HHTs were only responsible for the biosynthesis of Avn-A and Avn-C using hydroxyanthranilic acid as an acyl acceptor and 4-coumaroyl-CoA and caffeoyl-CoA as an acyl donor. Avn-B was synthesized by a CCoAOMT-like enzyme through the methylation of Avn-C. Mutations of two conserved residues lysine and aspartic acid proximal to the active site to alanine completely abolished the catalytic activity of the oat CCoAOMT-like enzyme towards all substrates tested. Collectively, these results have elucidated the molecular mechanisms for the complete biosynthesis of three

Page 2 of 31

major avenanthramides in oat and paved ways for genetic improvement of the nutritional trait through marker-assisting breeding in oat, and metabolic engineering of the biosynthetic pathway in heterologous systems to produce the nutritionally important compounds.

5. Introduction: Brief project background and rationale. Human food market for oat is increasingly important as the consumers gradually recognize its health benefits (Webster and Wood, 2011). Oat is considered as a healthy grain primarily due to the presence of β-glucan, a

mixed-linkage (β13, β14) glucose polymer that can reduce the risk of heart diseases through lowering the cholesterol (Sadig-Butt, 2008). In addition, oat grain contains a higher level of nutritious storage proteins than other cereals, legumin-like globulins equivalent in nutritional quality to soybean proteins (Singh et al., 2013). Furthermore, oat comprises avenanthramides, a group of polyphenolic compounds that possess strong antioxidant, anti-inflammatory and anti-skin-irritating properties that have been linked to the prevention of cardiovascular diseases, and the protection of human skin (Adom and Liu, 2002; Liu et al., 2004; Sur et al., 2008; Meydani, 2009). Avenanthramides were first identified as phytoalexins in oat leaves infected by pathogenic fungus Puccina coronata (Mayama et al., 1981), but they were later found at a considerable level in oat grains (Collins, 1989; Wise, 2011). Avenanthramides are a group of N-cinnamolyanthranilic acids comprising anthranilic acid and cinnamic acid linked by an amide bond (Collins and Mullin, 1988). Due to the presence of various substituted groups on the two components, more than 25 different types of avenanthramides have been detected in oat grains. However, the most abundant ones are three comprising 5-hydroxyanthranilic acid conjugated with caffeic acid (as avenanthramide-C or Avn-C), p-coumaric acid (as avenanthramide-A or Avn-A) or ferulic acid (as avenanthramide-B or Avn-B) (Collins, 1989). Previously, avenanthramides have been found to possess strong antioxidant activity (Dimberg et al., 1996). In addition, several studies have also indicated that both oat phenolic-rich extract and pure avenanthramides had anti-inflammatory activity by reducing the expression of pro-inflammatory cytokines and repressing nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) function (Sur et al., 2008; Chu et al., 2013). Moreover, Avn-C was reported to inhibit the proliferation of rat vascular smooth muscle cells through modulation of cell cycle, indicating the potential health benefit of oat consumption in the prevention of coronary heart disease. Recently, the anti-proliferative effects of avenanthramides have also been examined on several cancerous cell lines and found that Avn-C was able to decrease the proliferative rates of colonic cancer cell lines, including Caco-2, HT29, LS174T, HCT116 and MDA-MB-231 breast cancer (Adom and Liu, 2002; Liu et al., 2004). However, no study has addressed the protective effect of avenanthramides against cellular oxidative stress. There is also limited evidence on the protective effect of avenanthramides against pro-inflammation in normal human cells. Therefore, exploration of multiple health benefits of avenanthramides and mechanism investigation would provide a comprehensive understanding of the functions. Despite the health importance of avenanthramides, the biosynthesis of these compounds in oat is not completely understood. Previous reports indicate that avenanthramides are synthesized through a condensation process of hydroxyanthranilic acid with hydroxycinnamoyl-CoA and related species catalyzed by hydroxycinnamoyl-CoA:hydroxyanthranilate N-hydroxycinnamoyltransferase (HHT) , an anthranilic acid

Page 3 of 31

acyl-CoA dependent acyltransferase, homologous to hydroxycinnamoyl-CoA:shikimate/quinate hydroxycinnamoyl transferase (HCT) catalyzing the coupling of p-coumaroyl-CoA with shikimate/quinate (Ishihara et al., 1997; Yang et al., 2004; D'Auria, 2006; Bontpart et al., 2015). HCT is a family of well-conserved enzymes among land plants participating in the biosynthesis of lignins and flavonoids by acylating a wide range of aroyl group-containing substrates to acceptors (Yang et al., 2004; Xu et al., 2009; Landmann et al., 2011; Tohge et al., 2013; Molina and Kosma, 2015). Therefore, acyl donors such as hydroxycinnamoyl-CoA and derivatives, precursors for the biosynthesis of avenanthramides are probably diverged from the phenylpropanoid pathway and their biosynthesis may require 4-coumarate:CoA ligase (4CL) for the activation of aroyl group-containing substrates to the corresponding thioesters for the subsequent condensation process (Gross and Zenk, 1974; Hahlbrock and Scheel, 1989; Dixon and Paiva, 1995). In addition, the exact mechanism for the biosynthesis of Avn-B has not been conclusively determined, although this avenanthramide might be synthesized by the same HHT using 4-hydroxycoumarol-CoA and feruloyl-CoA as substrates (Ishihara et al., 1997; Yang et al., 2004). The incomplete and inconclusive information on the biosynthesis of avenanthramides has rendered ineffective genetic improvement of this important nutritional trait in oat.

6. Methodology: Include approaches, experimental design, methodology, materials, sites, etc. 6.1 Exaction of phenolic compounds from oat samples Five (± 0.01) grams of ground powder of each sample was extracted three times with 50 mL of methanol:phosphoric acid solvent mixture (80:20, v:v) by magnetic stirring for 30 min. The sample mixtures were centrifuged for 10 min at 10,000 rpm and the supernatants were pooled and evaporated at 40 °C using a Buchi III Rotavapor vacuum evaporator (Brinkmann Instruments, Mississauga, ON). The residue was re-suspended in 5 mL of the extraction solvent and centrifuged at 10,000 rpm for 10 min. The final volume of the oat extract was made up to 5 mL in a volumetric flask with methanol. The oat extract samples were stored in dark at -18 °C for analysis.

6.2 HPLC analysis of Avn A, B and C in oat extracts Reversed phase HPLC was performed on an 1100 series HPLC system (Agilent Technologies Canada Incorporated, Mississauga, ON), equipped with a photodiode array (PDA) detector. Phenolic separation was achieved using a 250 x 4.6 mm Luna® 5 μm, C18 column (Phenomenex, Torrance, CA, USA) in series with a C18 guard cartridge (Phenomenex, Torrance, CA, USA). Mobile phases consisted of solvent A: H2O with 5% acetonitrile and 0.5% trifluoracetic acid by volume; Solvent B: acetonitrile. Samples were run with a linear gradient over 40 min from 5 to 40 % solvent B in solvent A with a flow rate of 1 mL/min followed by 40 % solvent B in solvent A for 5 min. The injection volume was 10 μL. All samples were syringe filtered (0.22 μm pore size; 13 mm diameter; Sigma-Aldrich, Oakville, ON, Cat #: F7148) prior to HPLC analysis. Standards run in conjunction with the samples included: Avn-A (SynInnova, Edmonton, AB, Cat #: SL707), Avn-B (SynInnova, Cat #: SL337) and Avn-C (SynInnova, Cat #: SL340). Standards were prepared at the following concentrations: 2.0, 4.0, 6.0, 8.0, 10.0, 20.0 and 40.0 (± 0.02) ppm. Detection was monitored at 340 nm with reference at 700 nm. All samples were analyzed in triplicate.

6.3 Total phenolic content analysis of oat extracts The Folin-Ciocalteau (FC) method was used to determine total phenolic content of oat extract samples. In brief, a 2.0 N stock FC solution (Sigma-Aldrich, Cat #: 47641) was diluted 1/10 with water obtained Millipore Milli-Q™ water system (Millipore Corporation, Milford, MA, USA). All oat extracts were diluted 1/20 with Milli-Q™ water. One hundred and fifty microliters of the diluted oat extract was mixed with 750 µL diluted FC reagent by vortexing. Six hundred microliters of 15% (w/v) Na2CO3 solution was then added followed by vortexing. The samples were then held static at room temperature in dark for 2 h. Following the incubation period, samples were analyzed by UV-visible spectroscopy (Genesys 10S UV-

Page 4 of 31

visible, Fisher Scientific, Edmonton, AB) at 765 nm. A control was prepared which contained all the reagents and 150 µL Milli-Q™ water in place of the diluted samples. Milli-Q™ water was used as blank. A standard curve was prepared using gallic acid at concentrations ranging from 10-50 ± 0.2 mg/L in water, which were analyzed in conjunction with samples.

Standard curves had correlation coefficients 0.9989. All samples and standards were analyzed in triplicate and the results were reported as gallic acid equivalent (GAE).

6.4 2,2'-Azino-bis (3-ethylbenzothiazoline-6-sulphonic acid) (ABTS) radical scavenging activity ABTS•+ free radical solution was prepared by mixing 7.0 mM ABTS solution (Sigma-Aldrich, Cat #: 11557) and 2.45 mM potassium persulfate (Sigma-Aldrich, Cat #: 216224) in a ratio of 1:1. This mixture was held static at room temperature for 12 h in dark for ABTS•+ formation. The resulting ABTS•+ radical cation solution was diluted approximately 1 in 50 with 70% (v/v) methanol to give an absorbance reading of 0.75 ± 0.05 at 734 nm. Each oat extract sample was diluted 1/4, 1/3, 1/2, 1/1 with methanol. The Trolox standard (Sigma-Aldrich, Cat #: 238813) was prepared in 70% (v/v) methanol at the following concentration range: 0.1 (0.4 mM) to 0.5 mg/mL (2.0 mM). Avn-A, B and C were prepared from 100 ppm to 1000 ppm in methanol. Caffeic acid (Sigma-Aldrich, Cat #: C0625), gallic acid (Fisher Scientific, Cat #: A122-500) and Trolox (Sigma-Aldrich, Cat #: 238813) were used as standards and prepared from 50 to 300 ppm in methanol. The assay was conducted by mixing 10 μL of the sample solution with 1.0 mL of ABTS•+ solution. The absorbance was read at 734 nm at time intervals of 1, 4 and 6 min. A control solution of 10 µL 70% (v/v) methanol in 1.0 mL ABTS•+ solution was prepared and analyzed at the same wavelength. Absorbance values measured at 6 min were used to calculate the percent radical inhibition.

6.5 2,2-Diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity A 500 µM DPPH solution was prepared by dissolving 9.8 ± 0.2 mg of DPPH (Sigma-Aldrich, Cat #: D9132) in 70% (v/v) methanol. Fresh DPPH solution was prepared daily for sample analysis. Oat extracts were diluted 1/4, 1/3, 1/2 with methanol. Avn-A, B and C were prepared from 100 to 1000 ppm in methanol. Caffeic acid, gallic acid and Trolox were used as references and prepared from 50 to 300 ppm in methanol. A 1.0 mL DPPH solution and 0.25 mL sample solution were mixed by vortexing. The blank consisted of methanol, and the control was prepared by mixing 1.0 mL DPPH solution with 0.25 mL of methanol. All samples were held static at room temperature in dark for 15 min, prior to absorbance measurement at 517 nm. 6.6 Cell culture Normal human skin fibroblasts 2DD cells were grown in high glucose (4.5 mg/mL) Dulbecco’s Modified Eagle Medium (DMEM) (pH 7.7; Corning, Manassas, VA, USA, Cat #: ca45000-304) supplemented with 10% (w:w) fetal bovine serum (FBS) (Gibco, Thermo Fisher Scientific, Cat #: 12483-020) and 1.0% (w:w) penicillin streptomycin (GE Healthcare Life Sciences, Logan, UT, USA, Cat #: SV30010). Cells were seeded at an initial density of 3000 cells/cm2 and incubated at 37°C in a humidified with 5% carbon dioxide (CO2). Cells were passaged every 3 to 4 days with cells never excessing 80% confluency. Cells from passage number 12 to 18 were used in experiments. Four human cancer cell lines: MCF-7 cells (human breast adenocarcinoma), HCT 116 cells (human colon carcinoma), HepG2 cells (human liver hepatocellular carcinoma) and U2OS cells (human bone osteosarcoma) were purchased from American Type Culture Collection (Manassas, VA, USA) grown in the same media and under the same culture condition as 2DD cells. Culture media was changed every 3 to 4 days.

6.7 Cell treatment Media used for Avn C treatments was identical to that used for proliferative cultures. Avn C stock solution was made by dissolving Avn C in dimethyl sulfoxide (DMSO) to a final concentration of 50 mM. DMSO was used as a vehicle control and never exceeded 0.4% (v:v) in culture media.

Page 5 of 31

Cell treatment for intracellular antioxidant activity assays: 2DD cells were seeded at a density of 3,000 cells/cm2. After 24 h, DMSO or Avn C (100 and 200 μM) were added into culture media and incubated at 37°C & 5% CO2 for 48 h. Cells were then washed two times with serum-free DMEM and hydrogen peroxide solution (H2O2) (Sigma-Aldrich, Cat #:

216763) was added to serum-free DMEM media at a final concentration of 200 M with incubation at 37°C in a humidified atmosphere with 5% CO2 for 1 h. Three biological replicates were conducted.

Cell treatment for anti-inflammatory activity assays: 2DD cells were seeded at a density of 3,000 cells/cm2. To evaluate anti-inflammatory activity of Avn C on 2DD cells, three different treatment conditions were applied: 1) Non-inducing conditions: cells were treated with DMSO or Avn C (100 and 200 μM diluted from 50mM Avn C stock solution) for 24 h; 2) H2O2-induced conditions: cells were treated with DMSO or Avn C (100 and 200 μM diluted from 50mM Avn C stock solution) for 48 h, followed by washing with serum-free DMEM. Hydrogen peroxide in serum-free media was added to achieve a final concentration of 200 μM with incubation as above for 1 h to induce cellular stress; and 3) TNFα-induced condition: cells were treated with DMSO or Avn C (100 and 200 μM diluted from 50mM Avn C stock solution) for 24 hours. TNFα was then added to media to a final concentration of 10 ng/mL and incubated for 4 h to induce the expression of pro-inflammatory markers. Three biological replicates were conducted.

Cell treatment for anti-proliferative effect assays: 2DD cells and four cancer cells were seeded at a density of 3000 cells/cm2. After 24 h, DMSO or Avn C (50, 100 and 200 μM diluted from 50mM Avn C stock solution) were added into culture media and incubated for 48 h. Three biological replicates were conducted.

6.8 Cell counts and cell viability 2DD cells and cancer cells were grown in 6-well plates. At the end of the incubation period, cells were dissociated from the plate surface using TrypLE Express (Life Technologies, Carlsbad, CA, USA, Cat #: 12604013) and centrifuged at 200 relative centrifugal force (rcf) for 5 min to pellet cells. Cells were then re-suspended in culture media. Cell counts were conducted using 10 µL of cell suspension on a 0.0025 mm2 Neubauer improved haemocytometer. To count and calculate total cell number, one drop of cell suspension was placed on the hemocytometer and then covered with a piece of cover glass. Cell numbers in five of the nine squares were counted. For cell viability test, cell suspensions were mixed 1:1 with 0.4% (v:v) trypan blue dye (VWR International, Cat #: CA97063-702). Total stained and unstained cells were counted by a haemocytometer under light microscope. Cells stained dark blue were considered not viable.

6.9 Immuno-labeling of Ki67 2DD cells were grown on sterilized 22 mm2 glass coverslips in 6-well dishes. At the end of the incubation period, cells were fixed with 3.7% formaldehyde in phosphate-buffered saline (PBS; 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4 and 2 mM KH2PO4; pH 7.4) for 10 min at room temperature, followed by incubation in ice-old methanol/acetone (1:1 mixture) for 8 min at 4°C. Cells were washed two times with PBS and then permeabilized with 0.5% (v:v) Triton X-100/PBS for 10 min at room temperature. Cells were blocked with 1% (v:v) bovine serum albumin (BSA) for 1 h at room temperature. Cells were placed on coverslips and were incubated in rabbit anti-Ki67 (1:2000 dilution; Novacastra, Newcastle, UK, Cat #: NCL-Ki67) diluted in 1.0% (v:v) BSA/PBS for 1 h in a dark humidity chamber at room temperature. Cells were then washed two times with PBS (1 mL per wash) and incubated in goat anti-rabbit A488 (1:200 dilution; Stratech/Jackson Scientific, UK, Cat #: 111545-003-JIR) diluted in 1.0% (v:v) BSA/PBS for 1 h in a dark humidity chamber. Nuclei were counterstained with VECTASHIELD® Mounting Medium with DAPI (4′,6-diamidino-2-phenylindole) (Vector Laboratories, Burlingame, CA, USA, Cat #: H1200) onto glass slides and sealed with nail varnish. A Nikon Y-IDP microscope fitted with an X-Cite fluorescence light source and a Nikon Digital Sight DS-U3 camera was used to collect images. Images were collected at 40X magnification with a constant exposure time. Images were imported into ImageJ (https://imagej.nih.gov/ij/), an image processing software. An arbitrary threshold was selected and any nuclei with a level above this value was considered positive and those below this threshold were considered negative. Percentage of Ki67 positive cells was calculated.

6.10 MitotrackerTM Orange labeling of intracellular free radicals

Page 6 of 31

MitoTrackerTM Orange CM-H2TMRos (Molecular Probe, Eugene, OR, USA, Cat #: M-7511) is a reduced, non-fluorescent dye that stains mitochondria in live cells and becomes fluorescent when oxidized. 2DD cells were grown on sterilized 22 mm2 glass coverslips in 6-well dishes. Coverslips with adhered cells were washed two times with serum-free media (2ml per wash). MitoTracker Orange was dissolved in DMSO and added to the serum-free media at a final concentration of 500 nM. After incubating for 30 min, cells on coverslips were fixed and image analysis was conducted.

6.11 RNA extraction, cDNA synthesis and genes cloning RNA extraction and cDNA synthesis and gene cloning were performed following the standard protocols. The PCR products for candidate genes were first cloned into a pGEM®-T vector (Promega, USA) and verified by sequencing. For the construction of plasmids in the expression in E. coli, verified fragments were released from pGEM®-T and subcloned into the destination vector pET-28a(+) with specific restriction enzymes. The expression plasmids were verified by restriction enzyme digestion and sequencing.

6.12 Quantitative polymerase chain reaction (qPCR)

Quantitative PCR was performed following cDNA synthesis. A series of 10 L reactions were set up using 5 μL PerfeCTa® SYBR® Green SuperMix for iQ (Quantabio, Beverly, MA, USA, Cat #: 95053-500), 2 μL cDNA template, 2 μL nuclease-free H2O and 1 μL of 3 μM of each forward and reverse primers. qPCR reactions were conducted using a Rotor-Gene® Q qPCR (Qiagen, Germantown, MD, USA). All reactions for each gene were run in triplicate with non-template controls. Melt curve analyses were conducted to confirm the presence of single products in each reaction. Results were quantified using the ΔΔCt method against two normalizing genes: SPARC and FKBP10. 6.13 Western blot All protein samples were diluted to final concentrations of 1.0 or 1.5 µg/µL. For protein extracts prepared in RIPA buffer, samples were diluted in RIPA buffer and SDS-PAGE Loading Buffer (5X: 250 mM Tris-HCl pH 6.8, 30% v/v glycerol, 0.02% w/v bromophenol blue, 20% w/v 2-mercaptoethanol and 10% w/v SDS). For protein extracts prepared in Laemmli lysis buffer, samples were diluted in 1X Laemmli lysis buffer and 1 µL of 1% bromophenol (w:v) blue was added to each sample. All diluted protein extracts were denatured by boiling at 95°C for 5 min and then cooled on ice. Samples were then centrifuged for 10 min at 15,000 rpm, and the supernatants were transferred to new labelled tubes and stored at -20°C until analyses. Denatured protein samples (15 to 30 µg cell lysate per sample) were loaded and separated on a polyacrylamide gel with a 5.0% polyacrylamide stacking top gel and a resolving bottom gel in 1X running buffer (25 mM Tris base, 192 mM glycine and 0.1% SDS) at 125 V. Proteins were then transferred to a nitrocellulose membrane (Bio-Rad Laboratories, Cat #: 1620115) in 1X transfer buffer (25 mM Tris base, 192 mM glycine and 20% methanol) using Trans-Blot® SD semi-dry transfer cell (Bio-Rad Laboratories) at 25 V. The membranes were then blocked in 5.0% skim milk powder in phosphate-buffered saline with 0.05% Tween® 20 (PBST) for 1 h at room temperature, followed by incubating overnight at 4°C in primary antibody diluted in 5.0% skim milk powder in PBST. Primary antibodies used were rabbit anti-phospho-NF-κB p65 (Ser536) antibody (1:1000 dilution; Cell Signaling Technology, Danvers, MA, USA, Cat #: 3033), mouse anti-SIRT1 antibody (1:1000 dilution; Abcam, Cat #: ab110304), mouse anti-phospho-mTOR (Ser 2448) antibody (1:500 dilution; Santa Cruz, Dallas, TX, USA, Cat #: sc-293133), rabbit anti-LC3II (1:1000 dilution; Abcam, Cat #: ab48394), mouse anti-heme oxygenase 1 antibody (1:500 dilution; Santa Cruz, Cat #: sc-136960) mouse anti p-Histone H2A.X (Ser 139) antibody (1:500 dilution; Santa Cruz, Cat #: sc-517348) and rabbit anti-β-actin (1:2000 dilution; Abcam, Cat #: ab8227). Following primary antibody incubation, membranes were washed three times in 5.0% skim milk/PBST for 10 min. Membranes were then incubated in secondary antibody diluted in 5.0% skim milk/PBST for 1 h at room temperature. Secondary antibodies used were goat anti-rabbit horse radish peroxidase (HRP) (1:2000 dilution; Abcam, Cat #: ab97069) or donkey anti-mouse HRP (1:2000 dilution, Jackson Scientific, West Grove, PA, USA, Cat #: 715-035-150). Membranes were then washed two times in 5.0% skim milk/PBST for 10 min and two times in PBST for 5 min. Protein bands were visualized with chemiluminescence using enhanced chemiluminescence (ECL) reagent (100 mM Tris-HCl pH 8.5, 0.2 mM p-coumaric acid, 1.25 mM 3-aminophtalhydrazide and 0.1% v/v H2O2). β-actin was used as a loading control.

Page 7 of 31

6.14 Chromatin immunoprecipitation-qPCR (ChIP-qPCR) 2DD cells were grown in 15 cm diameter plates and treated with DMSO (control) or 100 µM Avn C for 48 h. At the end of the treatment period, cells were fixed with 1% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA, USA, Cat #: 15714) in 15 mL serum-free culture media per plate for 10 min at room temperature. Fixation was terminated by adding glycine into culture media to a final concentration of 125 mM followed by incubation at room temperature for 10 min. Media were then removed and cells were washed two times with ice-cold PBS. Cells were scraped in ice-cold PBS and centrifuged at 200 rcf for 5 min at 4°C to pellet cells. Cells were then re-suspended in 400 µL ChIP lysis buffer (1% SDS, 10 mM EDTA, 50 mM Tris-HCl pH 8.0) containing Protease Inhibitor Cocktail 2 (Sigma-Aldrich, Cat #: P8340) and Phosphatase Inhibitor Cocktail 2 (Sigma-Aldrich, Cat #: P5726). Following 10 min of incubation on ice, cells were sonicated on ice at ~7% power output and 30% duty cycle for 1 min. Samples were then centrifuged at 12,000 rcf 4°C for 10 min and supernatants were transferred to new 1.5 mL tubes as sheared chromatin samples. Aliquots of chromatin samples were stored at -80°C until analyses. From each treatment condition, 30 µL of the chromatin sample was used as the input sample and 60 µL of the chromatin sample was used for each immunoprecipitation. A 2.5 µg aliquot of mouse anti-Nrf2 antibody or 2.5 µg of rabbit anti-phospho-NF-κB p65 (Ser536) antibody was added to 60 µL chromatin sample diluted 10 times in ChIP buffer (0.01% SDS, 1.1% Triton X100, 1.2 mM EDTA, 16.7 mM Tris-HCl pH 8.0 and 167 mM NaCl) containing Protease Inhibitor Cocktail 2 (Sigma-Aldrich, #: P8340) and Phosphatase Inhibitor Cocktail 2 (Sigma-Aldrich, Cat #: P5726). 2.5 µg donkey anti-mouse HRP was used as the non-specific antibody control. The mixture was incubated at 4°C overnight with rotation, followed by binding to 50 µL Dynabeads™ Protein A (Life Technologies, Oslo, Norway, Cat #: 10006D) at 4°C for 1 h. The samples were then washed three times with ChIP washing buffer I (0.1% SDS, 1% Triton X100, 2 mM EDTA, 20 mM Tris pH 8.0 and 150 mM NaCl) for 5 min each, three times with ChIP washing buffer II (0.1% SDS, 1% Triton X100, 2 mM EDTA, 20 mM Tris pH 8.0 and 500 mM NaCl) for 5 min each and three times with ChIP washing buffer III (1 mM EDTA and 10 mM Tris-HCl pH 8.0) for 5 min each. After washing steps, samples were eluted with 500 µL freshly made elution buffer (1% SDS and 0.1 M NaHCO3) for 1 h at room temperature. For each eluted and input sample, crosslinks were reversed by adding 200 mM NaCl, 12.5 mM EDTA and 2 µL proteinase K (Invitrogen, Cat #: 25530049), followed by incubation at 65°C for 5 h with agitation at 900 rpm in a thermal mixer. The DNA from each sample was extracted by phenol-chloroform as follows: after reverse-crosslinking, 500 µL of phenol-chloroform (1:1, pH 8.0) was added into each tube and vigorously vortexed. Samples were then centrifuged at 12,000 rcf for 10 min at 4°C. The clear upper-phase was separated from the sample and placed in a new 1.5 mL centrifuge tube. DNA was precipitated by adding 2.0 µL glycogen and 1X sample volume of isopropanol. Samples were then centrifuged at 12,000 rcf for 30 min at 4 °C, and the supernatant was removed. The DNA pellet from the input sample was re-suspended in 40 µL nuclease-free water and DNA from immunoprecipitated sample was re-suspended in 80 µL nuclease-free water. Chromatin shearing efficiency was checked monitored by running 5 µL input DNA samples on a 1.5% agarose gel and most DNA fragments were found between 200 and 1000 bp on the gel.

6.15 Protein expression and purification The pET-28a(+) recombinant constructs were introduced into E. coli BL21 (DE3) (Novagen, CA) for oat protein expression.

Expression was induced by the addition of 0.5 mM isopropyl--1-thio-galactopyranoside (IPTG) to cultures (A600 at 0.4 to 0.5), and the induced cells were incubated for overnight at 16oC. The cells were harvested by centrifugation at 5,000 g for 10 mins. The pellet was resuspended in a buffer containing 100 mM Tris/HCL, pH 8.0, 0.5 M NaCl, 20 mM imidazole and 5% glycerol. Cells were disrupted by glass beads using a Mini-Beadbeater, and lysate was subsequently centrifuged at 15,000 g for 15 min at 4 oC. The recombinant His-tagged fusion proteins were purified by Hispur Ni-NTA Resin (Thermo Fisher Scientific, USA) according to the manual. The purified proteins were further desalted and concentrated with Zeba Spin Desalting Columns (Thermo Fisher Scientific, USA). Concentration for purified protein samples was determined by the Bradford Assay. 6.16 Enzyme assays Enzymatic assays of oat 4CL were performed following the previous method with some modification (Obel and Scheller, 2000). The reaction took place in a MOPS buffer (100 mM, pH 7.5) in a total volume of 300 μL consisting of 0.4 mM substrate (p-coumaric acid, caffeic acid, or ferulic acid), 2.5 mM ATP, 2.5 mM MgCl2, 0.2 mM Coenzyme A, 1~10 μg of purified oat 4CL proteins. Enzymatic reactions were initiated by the addition of purified 4CL protein. The reaction with

Page 8 of 31

boiled (96oC, 10 min) proteins was exploited as the control. The assay was performed at 30°C and formation of CoA esters was monitored at 333 nm for p-coumaroyl-CoA, 346 nm for caffeoyl-CoA and feruloyl-CoA in a time course until the substrate conversion to CoA esters was completed. Assays were terminated by adding 10 μL of acetic acid and analyzed by HPLC. Oat HHT activity was determined by reacting 10 μL the protein extract with 10 μl of 10 mM 5-hydroxy-anthranilic acid in DMSO and 50 μL of 2 mM one of p-coumaroyl-/caffeoyl-/feruloyl-CoA at 30oC for 1 hour in 30 mM Tris-HCl (pH 7.2) in a total reaction volume of 100 μL. The reaction was initiated by adding the protein and stopped by adding 20 μL acetic acid. The reaction mixture was diluted with 0.38 mL Methanol, filtered through 0.22 μm filter and analyzed by HPLC. Mass spectrum analysis of products in the assays were performed following a method from previous study (Atila et al., 2016). The assay of oat CCoAOMT-like sequence was performed following a previous study with modification (Schoch et al., 2001). The reaction was comprised of 10 μg purified protein, 1 mM oversaturated S-Adenosyl-L-methionine

(SAM) (Sigma, USA), 0 μM to 2mM Avenanthramide C (SynInnova, CA), 0 M to 5 mM caffeoyl-CoA (company name), 0

M to 10 mM caffeic acid (Sigma, USA) 50 mM Tris- HCl, 0.2 mM MgCl2, 2 mM DTT, 10% glycerol, and 0.2 mM PMSF and incubated at 25oC for 30 min. Enzymatic Michaelis-Menten kinetic parameters Km and Vmax on avenanthramide B, feruloyl-CoA, and ferulic acid were calculated by GraphPad Prism 6.

7. Research accomplishments: (Describe progress towards meeting objectives. Please use revised objectives if Ministry-approved revisions have been made to original objectives.)

Objectives Progress

1). Survey of oat total phenolic and three major anvenathramide contents (meet)

The Folin-Ciocalteau method was used to measure the total phenolic compounds using gallic acid as a reference, and results were reported as gallic acid equivalent (GAE) in Table 1. The range of total phenolic contents in selected oat lines was found from 78.5 mg to 121.2 mg GAE/100 g oat seeds with the mean value at 92.0 mg GAE/ 100 g oat seeds. The top three total phenolic content lines were BW10, OT2021 and BW5303. HPLC-PDA was used to determine the phenolic composition of oat phenolic extracts from more than 30 oat cultivars and breeding lines (Table 2). Representative chromatograms of chemically synthetic Avn-A, Avn-B and Avn-C standards and oat cultivars were shown in Figure 1. Avn-A, Avn-B and Avn-C in oat phenolic extracts were identified based on their retention times and UV-visible spectral profiles. On average, the amount of Avn-C was the highest, followed by Avn-B and Avn-A among oat lines. Oat cultivars BW10, BW13, CDC Dancer, CDC Pro-Fi, OT2021 and CDC Ministreal were among the highest amounts of three major avenanthramides, Avn-A, Avn-B and Avn-C (from 90.58 to 27.82 mg/kg), while LSU105D563, CDC Sol-Fi, ASLAK and FL04179-L2 were among the lowest amounts of the three avenanthramides (7.56 to 8.9 mg/kg). The amount of total phenolic contents was positively correlated with the amount of the total three major avenanthramides (R2=0.68). A high level of the avenanthramides was generally reflected by a high amount of Avn-C or Avn-B or both.

2). In vitro antioxidant activity of oat total phenolic extracts and

The antioxidant activity of oat total phenolic extracts was determined by ABTS and DPPH radical scavenging assays. The results shown in Table 3 indicated the activity values obtained by the two methods were dissimilar in some lines; however the overall trend of the activities among the lines tested was highly correlated (R2=0.77). For instance, OT2021 had the greatest DPPH free radical scavenging ability followed by BW10, while BW10 showed the greatest ABTS free radical scavenging ability followed by OT2021 among the lines tested. Between the two assay methods, the ABTS gave the

Page 9 of 31

avenanthramide (exceeding)

activity more coincident with the amount of the total phenolic content and the three major avenathramides (R2=0.83 and R2= 0.84, respectively), compared with the DPPH result (R2=0.70 and R2= 0.57, respectively). The ABTS and DPPH radical scavenging activities of synthetic Avn-A, -B and -C were determined with the references of gallic acid, caffeic acid and Trolox. The amounts of DPPH and ABTS free radical reduced were linearly proportional to the amounts of the three major avenathramides over the ranges tested. The concentrations required to reduce 50% of free radicals were summarized in Table 4. A lower IC50 value indicated a higher antioxidant activity. By comparing the concentrations of avenanthramides required to reduce 50% of the DPPH free radical or ABTS free radical, the DPPH assay gave the order of activity as gallic acid>Avn-C>caffeic acid>Trolox>Avn-B>Avn-A, while the ABTS assay gave the order of activity as gallic acid>Avn-C>Avn-B>Trolox>caffeic acid>Avn-A. Although the two orders of activity were slightly different, both assays indicated Avn-C possessed the highest antioxidant activity, followed by Avn-B, and Avn-A had the least activity among the three major avenanthramides.

3). Health-promoting properties of oat avenanthramide in human cells (exceeding)

7.3.1 Avn-C protects normal human skin fibroblasts from H2O2-induced cellular damage Avn C protects 2DD fibroblasts from H2O2 induced oxidative stress: To evaluate cytoprotective effects of Avn C against H2O2-induced oxidative stress, normal human fibroblasts (2DD) were used as a cell culture model. 2DD cells were pre-treated with Avn C for 48 hours prior to 1 h exposure to H2O2 to induce oxidative stress. Free radicals inside cells were stained with a reduced, non-fluorescent dye MitoTracker

Orange that becomes fluorescent when oxidized by free radicals. Fluorescence micrographs were collected and the intensity of the Mitotracker™ dye was examined. The result indicated that Avn C penetrated into cells and was capable of scavenging intracellular free radicals. To further evaluate the protective capacity of Avn C against H2O2-induced oxidative stress, we examined transcript levels of antioxidant genes normally stimulated by oxidative stress. After pre-treatment with Avn C for 48 h followed by 1 h H2O2 exposure, total RNA was extracted, converted to cDNA and evaluated by qRT-PCR. Hydrogen peroxide treatment without cell pre-treatment with Avn C resulted in a significant increase in the transcripts from GSS, HMOX1, SOD1, GPx1 and CAT antioxidant enzyme genes. By comparison with H2O2 treatment alone, pre-treatment of 2DD cells with 200 uM Avn C significantly suppressed increased levels of all the transcripts induced by H2O2. Also, pre-treatment with 100 uM Avn C significantly suppressed increased levels of all the transcripts with the exception of CAT (Fig 2). Avn-C protects 2DD fibroblasts from H2O2-induced DNA damage: One of the major biological consequences of free radical production and exposure is increased DNA

damage. Upon DNA damage, the histone variant H2AX- becomes phosphorylated (p-

H2AX) and is deposited at the sites of damage, facilitating the recruitment of repair machinery; therefore, p-H2AX is commonly used as a sensitive marker for DNA double-strand breaks. We evaluated the ability of Avn C to protect cells from H2O2-induced DNA damage by measuring p-H2AX expression. After pre-treatment of normal human skin fibroblasts (2DD) with Avn C for 48 h followed by 1 h H2O2 exposure, a whole cell extract was obtained and subjected to western blot analysis. We observed that pre-

Page 10 of 31

treatment with Avn C (100 uM and 200 uM) significantly decreased levels of p-H2AX (as determined by γ-H2AX to β-actin ratios) induced by H2O2 (Fig. 3), further demonstrating the cytoprotective effects of Avn C on normal cells. Avn-C protects 2DD fibroblasts from H2O2-induced inflammatory responses: We measured the ability of Avn C to suppress the transcript levels of pro-inflammatory markers following H2O2 treatment. Transcripts from Il-1β, Il-6, Il-8 and TNF-α genes were evaluated by qRT-PCR. Avn C significantly reduced the amounts of transcripts from Il-6, Il-8 and TNF-α under pre-treatment conditions of 100 and 200 μM Avn C, whereas a significant reduction of Il-1β was only observed at the 200 μM pre-treatment concentration (Fig 4A). As we observed that Avn C suppressed H2O2-induced pro-inflammatory cytokine transcripts, we further evaluated if this suppression effect was mediated by inhibition of NF-κB activation in 2DD fibroblasts. Western blot analyses of phosphorylated NF-𝜅B p65 (Ser536) demonstrated that H2O2 treatment alone resulted in an increased protein level of phosphorylated p65, which was significantly reduced by 100 and 200 μM Avn C pre-treatment (Fig 4B). These observations indicate that Avn C is able to protect normal human skin fibroblasts against H2O2-induced inflammation through both the suppression of increased levels of pro-inflammatory cytokine transcription and NF-𝜅B activation. 7.3.2 Anti-inflammatory potential of Avn-C Avn-C protects 2DD fibroblasts from TNFα-induced inflammation: RT-qPCR was employed to evaluate if Avn C also could protect 2DD fibroblasts against TNFα-induced inflammation. Results demonstrated that TNF-α treatment alone strongly increased the level of transcripts from the Il-1β, Il-8 and TNF-α genes (Fig 5A). Pre-treatment with Avn C at 100 and 200 μM significantly reduced the mRNA levels of Il-1β, Il-8 and TNF-α induced by TNF-α treatment. Western blot analyses of phosphorylated NF-𝜅B p65 (Ser536) demonstrated that TNF-α treatment alone resulted in an increased protein level of phosphorylated p65, which was significantly reduced by 100 and 200 μM Avn C pre-treatment (Fig 5B). Avn-C reduces basal inflammation through NF-κB inhibition in 2DD fibroblasts: As we observed that Avn C showed a cytoprotective effect against H2O2-induced and TNFα-induced inflammation, we further investigated if Avn C could also exert its anti-inflammatory activity under normal metabolic conditions inside cells. Following Avn C treatment for 48 h on 2DD fibroblasts, RNA was extracted, converted to cDNA and evaluated by RT-qPCR. mRNA fold changes of pro-inflammatory cytokines are shown in Fig 6A. These results demonstrated a significant reduction of Il-1β, Il-8, IL-6 and TNF-α transcripts at the basal level, except for TNF-α at 100 μM Avn C treatment. Western blot analyses were performed on cell protein extracts under the same treatment

conditions. It was found that 100 and 200 M Avn C treatment significantly increased protein expression of phosphorylated NF-𝜅B p65 (Ser536) in 2DD fibroblasts (Fig 6B). To further confirm that Avn C reduced inflammatory response by inhibiting NF-κB activation, chromatin immunoprecipitation-qPCR (ChIP- qPCR) was performed to analyze changes in NF-κB binding activity to promoter regions of Il-1β and Il-8 with 48 h

of 100 M Avn C treatment. We observed that 100 μM Avn C treatment significantly decreased NF-κB binding activity to both Il-1β and Il-8 promoter regions (Fig 6C),

Page 11 of 31

indicating that NF-κB activity was down-regulated. Taken together, these findings suggest that Avn C reduces basal inflammation through reduced NF-κB activation in 2DD fibroblasts. 7.3.3 Avn-C induces HO-1 expression through Nrf2 activation in 2DD fibroblasts

Given that Avn C showed a cytoprotective effect against H2O2-induced oxidative stress through free radical scavenging which decreased H2O2-induced antioxidant gene transcription, we further investigated if Avn C could also exert protective effects by activating other cytoprotective pathways independent of free radical scavenging. 2DD cells were treated with Avn C for 48 h and RNA was extracted, converted to cDNA and evaluated by RT-qPCR to examine if there was any antioxidant enzyme that was up-regulated by Avn C treatment. Transcript quantification by qPCR demonstrated that HMOX1 was significantly up-regulated by 50, 100 and 200 μM Avn C treatment (Fig 7A), whereas other antioxidant genes either showed no significant change or significant decreases. To further confirm if HMOX1 gene expression was induced by Avn C treatment, western blot was used to analyze protein expression levels of HO-1, which is coded by the HMOX1 gene. It was found that 100 and 200 μM Avn C treatment significantly increased protein expression of HO-1 in 2DD fibroblasts (Fig 7B). Expression of HO-1 is regulated by nuclear factor-E2-related factor 2 (Nrf2). To investigate the impact of Avn C treatment on Nrf2-DNA binding activity of, 2DD cells were treated with vehicle (DMSO) or 100 μM Avn C for 48 h. Cells were then cross-linked and chromatin immunoprecipitation-qPCR (ChIP- qPCR) was performed to analyze Nrf2 binding activity to promoter regions of HMOX1 and NAD(P)H quinone dehydrogenase 1 (NQO1) genes, which have been identified as Nrf2-regulated genes. Treatment of cells with 100 μM Avn C treatment significantly increased Nrf2 binding activity to both HMOX1 and NQO1 promoter regions (Figure 7C). Taken together, these results indicate that Avn C induced HO-1 expression through increased DNA binding activity of Nrf2 in 2DD fibroblasts. This indicates that Avn C not only reduces oxidative stress through scavenging free radicals, it could also mediate signaling pathways resulting in cytoprotection. 7.3.4 Anti-proliferative effect of Avn-C To determine the anti-proliferative effect of Avn C on 2DD fibroblasts, we treated cells with either vehicle (DMSO) or 50, 100 and 200 μM Avn C for 48 h and monitored their growth. Avn C had a significant impact on proliferative rates in 2DD fibroblasts at 50, 100 and 200 μM concentrations (Fig 8A). Trypan blue staining of cells indicated that there was no significant cell death in any of the treatment conditions used (Fig 8B). To further confirm this response, we immuno-labeled Avn C treated 2DD fibroblasts for the proliferative marker Ki67, which is a nucleolar and chromatin-associated protein that is only present in actively dividing cells. 2DD cells showed a significant decreased in the number of Ki 67 positive cells at 50, 100 and 200 μM Avn C treatment (Fig 8C). These observations indicated that Avn C was modulating cell growth in primary fibroblasts without causing cell death. As we observed a significantly decrease of proliferative rate with Avn C treated normal human fibroblast cells, we investigated if this anti-proliferative effect was caused by activation of the autophagy pathway by Western blot analyses of the autophagy marker protein light chain 3 isoform II (LC3-II),

Page 12 of 31

Sirtuin 1 (SIRT1) and mTOR, upstream molecules of autophagy activation. Western blot results indicated that there was no change in the protein levels following 48 h Avn C treatment (data not shown) implying Avn C reduces the proliferative rate of normal human skin fibroblasts through an autophagy-independent pathway. To determine if Avn C can inhibit the growth of cancer cells as well, several cancer cell lines, MCF7 (breast), HTC116 (colon), HepG2 (liver) and U2OS (bone) were used. The result indicated that Avn C suppressed the growth of human breast cancer cells and bone cancer cells, but did not promote cell death, thus might not be effective at killing cancer cells (data not shown).

4). Cloning and functional analysis of oat genes in the biosynthesis of avenanthramide (exceeding)

7.4.1 Identification and functional analysis of genes encoding 4CLs in the biosynthesis of hydroxycinnamate thioesters Avenanthramides are the condensed products of anthranilic acid and hydroxycinnamic acid using hydroxycinnamoyl-CoA as an acyl donor and anthranilate as an acyl acceptor. The biosynthesis of hydroxycinnamoyl-CoAs was generally believed to be catalyzed by 4-coumarate-CoA ligase (4CL) converting hydroxycinnamic acids to corresponding thioesters (Stockigt and Zenk, 1975). To identify genes encoding this enzyme in oat, an Arabidopsis thaliana 4CL was used as a query to search a transcriptome database of oat developing seeds. Two candidate genes (As4CL1 and As4CL2) coding for putative 4CL were identified. Sequence analysis of these two genes revealed that they shared high sequence identity with each another throughout the open reading frame except for the middle region where an insertion of a few nucleotides occurred in As4CL2, which resulted in the considerable difference of amino acid sequences in the region. Functional analysis of As4CL1 by in vitro assays using the purified protein expressed in E. coli showed that the enzyme could convert three substrates, ferulic acid, p-coumaric acid and caffeic acid, to their corresponding thioesters. Under the assay condition, the preferred substrate was ferulic acid, followed by p-coumaric acid and caffeic acid. At one hour of reaction with the same concentration of substrates, over 90% of ferulic acid, 62% of p-coumaric acid and 52% of caffeic acid were converted to their corresponding CoA thioesters, respectively (Fig. 9). 7.4.2 Identification and functional analysis of new genes encoding HHT in the biosynthesis of Avn-A and Avn-C Three genes encoding HHT (AsHHT1-3) in the biosynthesis of avenanthramides were previously identified in oat (Yang et al., 2004). Encoded isozymes AsHHT1-3 shared very high amino acid identity (from 95% to 98%). By searching an oat seed transcriptome database (Gutierrez-Gonzalez et al., 2013) using these sequences as queries, three new HHT genes (AsHHT4-6) were identified, and they were highly homologous to AsHHT1-3 with more than 87% of amino acid identity. In vitro assays using the purified proteins of AsHHT1 and AsHTT4 expressed in E. coli showed that both could catalyzed the condensation of 5-hydroxyanthranilic acid and p-coumaroyl-CoA, producing a product with the retention time and mass spectra identical to standard Avn-A (Fig. 10). Mass spectrum analysis confirmed the authenticity of the product on the basis of the molecular ion at m/z 298.84 and a fragment at m/z 147.12 yielded by the break of the amide linkage. In addition, both enzymes could also condense 5-hydroxyanthranilic acid

Page 13 of 31

and caffeoyl-CoA to a product with the retention time and mass spectra identical to standard Avn-C). However, when feruloyl-CoA and 5-hydroxyanthranilic acid were supplied, no new product was produced by two enzymes, indicating AsHHTs were incapable of synthesizing Avn-B by the condensation of the two perceived substrates as we expected. 7.4.3 Identification and functional analysis of a gene encoding CCoAOMT-like enzyme in the biosynthesis of Avn-B As AsHHTs were not involved in the biosynthesis of Avn-B, we hypothesized that this avenanthramide might be synthesized by a different mechanism. A previous study showed that one of the major metabolites in mice fed with Avn-C was identified as Avn-B (Wang et al., 2015). As caffeoyl-CoA O-methyltransferases (CCoAOMT) could methylate a wide range of substrates, we assumed that this type of O-methyltransferases might be able to convert Avn-C to Avn-B by methylation. To confirm the hypothesis, we identified a CCoAOMT-like gene with the open reading frame (ORF) of 768 nucleotides encoding 256 amino acids from an oat transcriptome database. In vitro assays of this gene using the purified protein expressed in E. coli in the presence of S-adenosyl methionine showed that the enzyme indeed could convert Avn-C to a product with retention time and mass spectra identical to Avn-B (Fig 11). To examine the substrate specificity, three possible substrates, Avn-C, caffeoyl-CoA and caffeic acid, were employed in in vitro assays under the optimal condition. The result showed that the CCoAOMT-like enzyme exhibited activity towards all three substrates tested. However, the most preferred substrate was Avn-C other than caffeoyl-CoA and caffeic acid. Kinetic analysis of the enzyme activity using different concentrations of the three substrates showed that the ratio of Vmax to Km was 3.08 with Avn-C, five or ten times higher than those with caffeoyl-CoA and caffeic acid (0.70 and 0.31), respectively (Fig 12; Table 5). These results clearly indicate Avn-B is synthesized from Avn-C through the methylation process catalyzed by a CCoAOMT-like enzyme. CCoAOMTs were identified and functionally analyzed from a variety of plants species. They were highly conserved in residues involved in the binding of SAM, and metal ion. Two residues proximal to the active site, lysine and aspartic acid located in the C-terminal region, were involved in substrate binding and catalysis (Walker et al., 2016). In addition, a loop structure at the C-terminus were likely involved in the recognition of substrates (Ferrer et al., 2005). To better understand the properties of the oat CCoAOMT-like enzyme, two residues at the active sites (K174, D246) and one residue (A209) in the loop region possible for substrate recognition were mutagenized (Fig 13A). As shown in Figure 5B, mutation of either lysine at 174 or aspartic acid at 246 to alanine completely abolished the enzyme activity towards three substrates, which coincides with the previous result that these two residues are likely involved in the catalysis (Walker et al., 2016). The mutation of alanine 209 to aspartic acid, the corresponding residue in a sorghum CCoAOMT with substrate specificity to caffeoyl-CoA decreased activity to Avn-C, but increased the activity towards caffeoyl-CoA and caffeic acid. However, this change was not statistically significant (Fig 13B).

Page 14 of 31

8. Discussion: Provide discussion necessary to the full understanding of the results. Where applicable, results should

be discussed in the context of existing knowledge and relevant literature. Detail any major concerns or project setbacks.

The Folin-Ciocalteau method measures the reduction of a phosphomolybdic-phosphotungstic acid reagent (FCR) to a blue-colored complex in the presence of phenolics, which can be measured at 765 nm. Under basic conditions, dissociation of a phenolic proton leads to a phenolate anion, which is capable of reducing FCR. Thus, the Folin-Ciocalteau method has been widely used to quantify the total phenolic content. However, FCR not only react with phenolic compounds, but can also be reduced by nonphenolic compounds (e.g. vitamin C). Since the FC method actually measures the content of total reducing compounds that comprises mainly phenolics, the value can roughly represent the total phenolic content in a sample. Gallic acid is usually used as a reference in the assay, thus the total phenolic contents are often reported as gallic acid equivalent (GAE). As seen in Table 1, there is huge difference in the total phenolic contents among oat germplasm. Although there are many different types of avenanthramides in oat phenolic extracts, three of them (avenanthramides A, B & C) appear in much higher concentrations than the others (Figure 1B). Analysis of about thirty different cultivars and breeding lines shows a wide range in the amount of the three avenanthramides in oat with generally Avn-C being the highest among the oat lines, followed by Avn-B. Antioxidant activity is widely used as a parameter to characterize polyphenolics. The activity is based on the chemical nature capable of protecting a biological system against the production of reactive oxygen species. In vitro free radical scavenging tests are the most common methods used to measure the antioxidant activity. Two most widely used chemical compounds to measure the in vitro antioxidant activity of polyphenolics are chromogens ABTS• and DPPH• radicals. The DPPH (2,2-diphenyl-1-picryl-hydrazyl-hydrate) free radical method is an antioxidant assay based on an electron-transfer mechanism that produces a violet solution in a protic solvent such as alcohol. This DPPH radical is stable at room temperature and can be reduced by an antioxidant, leading to the reduced intensity of the initial colour in a sample solution. The ABTS free radical assay is based on the scavenging ability of antioxidants for stable radical cation ABTS• (2,2'-azino-bis(3- ethylbenzothiazoline-6-sulphonic acid) on electron transfer from sample antioxidants to the ABTS• and it measures the ability of sample antioxidants that react directly with the ABTS• radical to decrease the initial solution colour. Our results indicate these two methods are reliable to assess the antioxidant activity of oat total phenolic extracts and oat avenanthramides, showing that antioxidant activity is highly correlated with the total phenolic contents and the three major avenanthramides in oat. In addition, both methods indicate Avn-C had the highest antioxidant activity, followed by Avn-B and Avn-A among the three major avenanthramides. For the human cell study of oat avenanthramide, we demonstrated for the first time that Avn C protected human skin fibroblasts from H2O2-induced cellular oxidative damage as shown by significantly reduced levels of intracellular free radicals, DNA damage and pro-inflammatory response. Pre-treatment of Avn C also protected cells against signalings to promote inflammatory responses induced by TNF-α, indicating its functional role as a signaling mediator other than free radical scavenger. Furthermore, mechanistic studies revealed that Avn C suppressed basal level of inflammation through decreased NF-κB DNA binding activity and induced heme oxygenase 1 expression through increased Nrf2 DNA binding activity, indicating that Avn C exerts antioxidant and anti-inflammatory effect, either or indirectly through modulating NF-κB and HO-1/Nrf2 signaling pathways apart from free radical scavenging. In addition, Avn C slowed the growth rate of normal human skin fibroblasts through autophagy-independent mechanisms. It also suppressed the growth of human breast cancer cells and bone cancer cells but did not induce cell death.

Page 15 of 31

Avenanthramides are a group of phenolic compounds found almost exclusively in oat. Three major avenanthramides in oat are the conjugates of hydroxycinnamic acid with p-coumaroyl-CoA (Avn-A), feruloyl-CoA (Avn-B) and caffeoyl-CoA (Avn-C), respectively. The biosynthesis of these compounds was previously believed to be catalyzed by a single enzyme called hydroxycinnamoyl CoA:hydroxyanthranilate N-hydroxycinnamoyl transferase (HHT) by the condensation of hydroxyanthranilate and substituted cinnamoyl-CoA thioesters (Ishihara et al., 1997; Yang et al, 2004). However, our enzymatic assays with two oat HHT proteins demonstrated the enzymes catalyzed the N-acylation of 5-hydroxyanthranilic acid with p-coumaroyl-CoA or caffeoyl-CoA, but not with feruloyl-CoA, indicating oat HHTs are only involved in the biosynthesis of Avn-A and Avn-C, but not Avn-B. A previous in vitro assay using crude protein extracts from oat showed that Avn-B was produced in the presence of 5-hydroxyanthranilic acid and feruloyl-CoA (Ishihara et al., 1997). As oat crude protein extracts contain numerous enzymes, this result could arise from other enzymatic activities in the extract. In addition, when oat AsHHT1 was expressed in E. coli, in vitro assays using the crude protein extracts detected a low level of activity for the synthesis of Avn-B in the presence of 5-hydroxyanthranilic acid and feruloyl-CoA (Yang et al., 2004). However, our in vitro assays using purified proteins of AsHHT1 and AsHHT4 expressed in E. coli showed that both could not catalyze the condensation of 5-hydroxyanthranilic acid and feruloyl-CoA, giving rise to Avn-B. The reason why the different results were obtained by the two experiments on the same gene is currently unknown. The possible explanation is that the low level of activity might also be derived from other factors in the crude proteins from the expression host in the previous experiment. As oat HHTs are only responsible for the synthesis of Avn-A and Avn-C, the next question would be what the biosynthetic mechanism for Avn-B is. In the consideration of the structural difference of Avn-C and Avn-B, we assumed that certain CCoAOMT would be able to convert Avn-C to Avn-B by methylation as this enzyme has a wide range of substrates. In fact, a previous feeding study showed that one of the major metabolites from Avn-C in mice was Avn-B (Wang et al. 2015). Indeed, when a CCoAOMT-like enzyme from oat was expressed in E. coli, the purified protein was capable in the synthesis of Avn-B on Avn-C in the presence of S-adenosyl methionine, indicating Avn-B is synthesized by the O-methylation of Avn-C catalyzed by the CCoAOMT-like enzyme. In plants, there are two types of O-methyltransferases responsible for methylating hydroxyl groups at the 3- and 5-positions of a phenolic ring. The first type called caffeic acid O-methyltransferase (COMT) is larger (38 – 43 kD) with methylation activity mainly on caffeic acid and related species (Zubieta et al., 2002; Green et al., 2014), while the second type called caffeoyl-CoA O-methyltransferase (CCoAOMT) is smaller (23 - 27 kD) with methylation activity mainly on caffeoyl-CoA and derivatives (Hoffmann et al., 2001; Noel et al., 2003). The oat O-methyltransferase with Avn-B synthetic activity belongs to the type II O-methyltransferase (CCoAOMT) (Hoffmann et al., 2001; Noel et al., 2003). The residues involved in in the binding of SAM and metal ion, and proximal to the active site for catalysis (Walker et al., 2016), and a loop structure for the recognition of substrates were highly conserved (Ferrer et al., 2005). In our assays, mutations of two conserved residues lysine and aspartic acid proximal to the active site to alanine completely abolished the catalytic activity of the oat CCoAOMT-like enzyme towards all substrates tested, confirming involvement of these residues in the catalysis (Walker et al., 2016). However, substitution of alanine at the position 209 with the corresponding residue aspartic acid in a sorghum CCoAOMT with substrate specificity to caffeoyl-CoA in the loop did not alter activity towards Avn-C and caffeoyl-CoA significantly, indicating this amino acid in the loop might not play a vital role in defining the substrate specificity.

Page 16 of 31

Oat has been considered as a functional food with many health benefits. One of primary effective ingredients in oat grain is avenanthramides, a group of polyphenolic compounds with antioxidant, anti-inflammatory, anti-cell-proliferating and anti-skin-irritating properties. Although the health-promoting properties of avenanthramides are well known, the biosynthetic mechanism is not completely understood. In the present study, we have identified three different types of enzymes involved in the biosynthesis of the major avenanthramides in oat, 4CL in activating hydroxycinnamates to their thioesters prior to the condensation, HHTs catalyzing the condensation in the biosynthesis of Avn-A and Avn-C, and CCoAOMT-like enzyme for the methylation of Avn-C to Avn-B. Particularly, we demonstrated that oat HHTs are only responsible for the biosynthesis of Avn-A and Avn-C, but not for Avn-B, which is synthesized by a new mechanism, the methylation of Avn-C catalyzed by CCoAOMT-like enzyme. For the complete biosynthesis of three major avenanthramides in oat, 4-hydroxy-coumaric acid is initially derived from phenylalanine catalyzed by phenylalanine ammonia lyase (PAL) and coumaric acid-4-hydroxylase (C4H). Hydroxylation of 4CH by 4-hydroxycoumaric acid-3-hydroxylase (C3H) gives rise to caffeic acid. Both 4-hydroxycoumaric acid and caffeic acid can be activated into their CoA thioesters by 4-coumaric acid:CoA ligase. 4-hydroxy-coumaroyl-CoA and caffeoyl-CoA can then be condensed with 5-hydroxyanthranilic acid to Avn-A and Avn-C, respectively, by HHT. Finally, Avn-C is methylated to Avn-B by CCoAOMT-like enzyme (Fig 14). The full elucidation of the biosynthetic pathway of avenanthramides in the present study not only contribute to our understanding of the biosynthesis of this important nutraceutical compound, but would also facilitate our genetic improvement of the nutritional trait in oat by marker-assisting breeding and open opportunity to produce the active compound by metabolic engineering of the complete biosynthetic pathway in heterologous systems.

9. Conclusions and Recommendations: Highlight significant conclusions based on the previous sections, with

emphasis on the project objectives specified above. Provide recommendations for the application and adoption of the project.

1. The amount of total phenolic compounds in oat is in a range from 78.5 mg to 121.2 mg GAE/100g

seeds. The amount of three major avenanthramides was observed (7 – 90 mg/kg) with avenanthramide C being generally higher than B or A in oat seeds.

2. Avn-C has the highest antioxidant activity, followed by Avn-B and Avn-A. Oat lines OT2021 and BW10

possess high free radical scavenging abilities, which are highly correlated with the total phenolic content and the amount of the three major avenanthramides.

3. Avn-C can protect human skin fibroblasts from H2O2-induced cellular oxidative damage by significantly

reducing the levels of intracellular free radicals, DNA damage and pro-inflammatory response.

4. Avn-C can suppress basal level of inflammation through decreased NF-κB DNA binding activity and induce heme oxygenase 1 expression through increased Nrf2 DNA binding activity.

5. Avn C can slow the growth rate of normal human skin fibroblasts through autophagy-independent

mechanisms.

6. Two genes from oat encoding 4-coumarate-CoA ligase (4CL) were identified and characterized for activating 4-coumaric acid, caffeic acid and ferulic acid to their CoA thioesters.

Page 17 of 31

7. Three new genes encoding hydroxycinnamoly-CoA:hydroxyanthranilate N-hydroxycinnamoly transferanse (HHT) were identified and characterized for the biosynthesis of Avn-A and Avn-C using hydroxyanthranilic acid as an acyl acceptor and 4-coumaroyl-CoA and caffeoyl-CoA as an acyl donor.

8. One hene encoding caffeoyl-CoA O-methyltransferase (CCoAOMT)-like enzyme were identified and

characterized for the biosynthesis of Avn-B through the methylation of Avn-C.

9. The complete biosynthesis of three major avenanthramides in oat has been elucidated, which paves ways for genetic improvement of the nutritional trait through marker-assisting breeding in oat.

10. Success stories/ practical implications for producers or industry: Identify new innovations and /or

technologies developed through this project; and elaborate on how they might impact the producers /industry.

Oat is known as functional food due to the presence of several nutritional factors. One of them is avenathramides, a group of polyphenols with health-promoting properties. This project attempted to survey content of avenathramides in oat germplasm, examine antioxidant and anti-inflammatory activities in human cells and analyze the biosynthesis of the avenathramides. The information gained is useful for understanding not only the biological activity of avenathramides but also for molecular mechanisms for the biosynthesis of these important nutritional compounds in oat. Particularly, the full elucidation of the biosynthetic pathway of avenathramides for the first time would be extremely beneficial for genetic improvement of the nutritional trait through marker-assisting breeding in oat, and metabolic engineering of the compounds in heterologous systems.

11. Patents/ IP generated/ commercialized products: List any products developed from this research.

NA

12. List technology transfer activities: Include presentations to conferences, producer groups or articles

published in science journals or other magazines.

Some of the results were shared with oat producer group and oat research community (The quest for healthier oats, Oat Scoop, November, 2015; The story of oat breeding and research, Grainews, February 2, 2016). Two papers are prepared and will be submitted for peer-reviewed publications.

13. List any industry contributions or support received. This project was supported by ADF and POGA.

14. Is there a need to conduct follow up research? Detail any further research, development and/or

communication needs arising from this project.

15. Acknowledgements. Include actions taken to acknowledge support by the Ministry of Agriculture and the

Canada-Saskatchewan Growing Forward 2 bilateral agreement.

Financial support from ADF and POGA is gratefully acknowledged.

Page 18 of 31

16. Appendices: Include any additional materials supporting the previous sections, e.g. detailed data tables, maps, graphs, specifications, literature cited

Table 1. Total phenolic content of ten oat phenolic extracts from different varieties as determined by the Folin-Ciocalteau assay.

Germplasm mg GAE/100g seed

BW 10 121.2 ± 2.5

BW 513 87.7 ± 1.2

BW 5303 97.9 ± 1.1

CDC Dancer 86.8 ± 1.5

CDC Morrison 84.5 ± 1.3

CDC Sol-Fi 82.3 ± 1.4

Jumbo 89.7 ± 1.5

Marion 89.2 ± 1.4

Newburg 78.5 ± 0.7

OT 2021 102.5 ± 2.0

Average 92.0 ± 1.5

Note: all values are reported as mg gallic acid equivalent per 100 g groat.

Table 2. Oat three major avenanthramide contents. Mean values of avenanthramide contents (mg/kg) with standard deviation.

Germplasm Ave-A Ave-B Ave-C Total

1 Ogle 4.43 ± 0.21 2.95 ± 0.31 13.31 ± 0.92 20.69 2 Marion 2.22 ± 0.19 3.05 ± 0.18 4.77 ± 0.12 10.04 3 CDC Morrison 7.95 ± 0.36 11.38 ± 0.35 6.61 ± 0.06 25.94 4 BW13 22.61 ± 0.95 18.94 ± 2.35 28.65 ± 0.43 70.2 5 Ave 20.24 4.72 ± 0.48 2.98 ± 0.25 8.55 ± 0.93 16.25 6 Matilda 4.25 ± 0.11 2.84 ± 0.14 7.66 ± 0.25 14.75 7 CDC Pro-Fi 12.41 ± 0.58 9.59 ± 1.14 13.25 ± 0.26 35.25 8 SW Bentania 3.95 ± 0.18 4.29 ± 0.10 6.28 ± 1.06 14.52 9 CDC Ministreal 8.93 ± 1.07 7.62 ± 0.76 11.27 ± 0.59 27.82

10 Kanota 3.73 ± 0.30 3.56 ± 0.40 10.72 ± 0.72 18.01 11 ASLAK 1.17 ± 0.27 1.43 ± 0.28 6.11 ± 0.51 8.71 12 Newburg 2.81 ± 0.17 4.75 ± 0.07 5.54 ± 0.12 13.1 13 WINSTON 1.76 ± 0.19 2.33 ± 0.11 4.83 ± 0.70 8.92 14 HiFi 5.75 ± 0.66 5.82 ± 1.58 9.28 ± 0.43 20.85 15 Supernova 8.52 ± 0.99 5.42 ± 0.15 7.31 ± 0.25 21.25 16 WIX 9150-1 6.70 ± 0.69 3.40 ± 0.12 4.68 ± 0.28 14.78 17 LSU105D563 2.70 ± 0.92 2.00 ± 0.15 2.86 ± 0.19 7.56 18 SA080087 4.69 ± 0.36 5.06 ± 0.67 5.62 ± 0.08 15.37 19 SEABISCUIT 3.51 ± 0.25 4.83 ± 0.59 4.49 ± 0.36 12.83 20 Lutz 2.91 ± 0.59 2.75 ± 0.51 4.17 ± 0.39 9.83

Page 19 of 31

21 FL 04179-L2 2.88 ± 0.02 2.72 ± 0.04 3.30 ± 0.10 8.9 22 LA06012SBSB-S1 3.68 ± 0.35 3.02 ± 0.12 3.83 ± 1.01 10.53 23 CDC Dancer 19.04 ± 0.23 29.54 ± 0.33 19.03 ± 0.15 67.61 24 CDC Sol-Fi 1.76 ± 0.09 2.15 ± 0.11 4.45 ± 0.11 8.36 25 Jumbo 3.93 ± 0.25 4.30 ± 0.09 4.43 ± 0.18 12.66 26 IL2858-1 3.76 ± 0.37 3.79 ± 0.28 4.69 ± 0.06 12.24 27 SDO80348 2.75 ± 0.57 2.29 ± 0.32 5.58 ± 0.25 10.62 28 Mam 17-5 5.05 ± 0.29 4.64 ± 0.17 7.39 ± 0.26 17.08 29 X466 4.76 ± 0.10 3.80 ± 0.12 6.34 ± 0.39 14.9 30 Leggett 4.22 ± 0.46 4.69 ± 0.27 5.57 ± 0.66 14.48 31 BW10 21.39 ± 0.69 34.67 ± 0.94 34.52 ± 0.87 90.58 32 BW513 5.43 ± 0.16 9.29 ± 0.09 11.03 ± 0.12 25.75 33 BW5303 3.80 ± 0.14 4.41 ± 0.06 6.61 ± 0.06 14.82 34 OT2021 9.82 ± 0.18 12.59 ± 0.28 11.46 ± 0.17 33.87

Average 6.12 ± 0.39 6.66 ± 0.39 8.65 ± 0.38 21.44

Table 3. Free radical scavenging activities of oat phenolic extracts from ten oat lines.

Germplasm

DPPH assay (1/IC50)a

ABTS assay

(mmol Trolox)

Total phenolic

content (mg/100g)

Total Avn-A/B/C

(mg/kg)

BW10 3.39 ± 0.05 1.54 ± 0.04 121.2 90.58

BW513 2.66 ± 0.08 0.93 ± 0.01 87.7 25.75

BW5303 2.73 ± 0.10 1.09 ± 0.04 97.9 14.82

CDC Dancer 2.86 ± 0.15 1.24 ± 0.04 86.8 67.61

CDC Morrison 2.76 ± 0.08 1.18 ± 0.04 84.5 25.94

CDC Sol-Fi 2.66 ± 0.05 1.02 ± 0.03 82.3 8.36

Jumbo 2.16 ± 0.02 1.03 ± 0.01 89.7 12.66

Marion 2.42 ± 0.02 1.07 ± 0.05 89.2 10.04

Newburg 1.26 ± 0.08 0.92 ± 0.01 78.5 13.1

OT2021 3.59 ± 0.08 1.31 ± 0.03 102.5 33.87 a 1/IC50: inverse of IC50

Table 4. Antioxidant activities of Avn-A, B and C as compared to caffeic acid, gallic acid and Trolox as indicated by IC50 values (IC50: mM concentration required for 50% free radical inhibition)

Antioxidant DPPH IC50 ABTS IC50

Avn-A 1.26±0.03 1.20±0.01

Avn-B 0.61±0.01 0.84±0.02

Avn-C 0.41±0.01 0.82±0.02

Caffeic acid 0.45±0.03 1.19±0.02

Gallic acid 0.14±0.01 0.37±0.01

Trolox 0.54±0.02 1.18±0.01

Page 20 of 31

Table 5. Kinetic parameters of oat CCoAOMT-like enzymes with three substrates

Substrate Km (uM) Vmax (nmol/ug/min) Vmax/Km

Avenanthramide C 167.0 ± 25.2 514.4 ± 41.0 3.08

Caffeoyl-CoA 2319.0 ± 867.6 1620.0 ± 390.0 0.70

Caffeic Acid 1903.0 ± 173.5 597.6 ± 40.0 0.31

Figure 1: HPLC analysis of anvenathramides in oat seeds.

A). HPLC-PDA chromatogram of Avn A, B and C standards monitored at 340 nm.

B). HPLC-PDA chromatogram of the oat extract from the CDC Dancer variety monitored at 340 nm. Peak assignments: 1. Avn C; 2. Avn A; 3. Avn B

Page 21 of 31

Figure 2. Avn C suppresses H2O2-induced mRNA expression of antioxidant enzymes. 2DD primary fibroblasts were pre-treated with DMSO (0) or Avn C (100 μM and 200 μM) for 48 h prior to exposure to 200 μM H2O2. qRT-PCR on cDNA libraries was performed for glutathione synthetase (GSS), heme oxygenase 1 (HMOX1), superoxide dismutase 1 (SOD1), glutathione superoxidase 1 (GPx1) and catalase (CAT) gene transcripts. Graphs indicate the average fold change from three biological replicates. Error bars represent the standard deviations. *p-value <0.05, **p-value <0.01 vs H2O2 treatment alone

Figure 3. Avn C suppresses H2O2-induced DNA double-strand breaks. Normal human skin fibroblasts (2DD) were pre-treated with DMSO (0) or Avn C (100 μM and 200 μM) for 48 h followed by 1 h exposure to 200 μM H2O2. Whole cell extract was then obtained and subjected to western blot analysis. Data are presented as a ratio of γ-H2AX to β-actin. Representative blot images are shown. All graphs indicate mean values from three biological replicates. Error bars represent standard deviations. *p-value <0.05, **p-value <0.01 vs H2O2 treatment alone.

Figure 4. Avn C suppresses H2O2-induced mRNA expression of pro-inflammatory cytokines and NF-κB activation in 2DD fibroblasts. Normal human skin fibroblasts (2DD) were pre-treated with DMSO (0) or Avn C (100 μM and 200 μM) for 48 h followed by 1 h exposure to 200 μM H2O2. A) RT-qPCR on cDNA libraries was performed for interlukin-1 beta (IL-1β), interlukin-6 (IL-6), interlukin-8 (IL-8) and tumor necrosis factor (TNF) gene transcripts. Data are represented as the fold change against control. B) Western blot analysis of phospho-NF-𝜅B p65 (Ser536) (p-p65) expression. Data are

Page 22 of 31

presented as a ratio of p-p65 to β-actin. Representative blot images are shown. All graphs indicate mean values from three biological replicates. Error bars represent standard deviations. **p-value <0.01 vs H2O2 treatment alone.

Figure 5. Avn C suppresses TNF-α-induced mRNA expression of pro-inflammatory cytokines and NF-κB activation in 2DD fibroblasts. Normal human skin fibroblasts (2DD) were pre-treated with DMSO (0) or Avn C (100 μM and 200 μM) for 24 h followed by 4 hours exposure to 10 ng/mL TNF-α. A) RT-qPCR on cDNA libraries was performed for interlukin-1 beta (IL-1β), interlukin-6 (IL-6), interlukin-8 (IL-8) and tumor necrosis factor alpha (TNF) gene transcripts. Data are represented as the fold change against control. B) Western blot analysis of phospho-NF-𝜅B p65 (Ser536) (p-p65) expression. Data are presented as a ratio of p-p65 to β-actin. Representative blot images are shown. All graphs indicate mean values from three biological replicates. Error bars represent standard deviations. * p-value <0.05, ** p-value <0.01 vs TNF-α treatment alone.

Page 23 of 31

Figure 6. Avn C reduces pro-inflammatory cytokine transcriptions through decreased DNA binding activity of NF-κB in 2DD fibroblasts. A) 2DD human skin fibroblasts were treated with DMSO (control) or Avn C (100 μM and 200 μM) for 48 h. RT-qPCR on cDNA libraries was performed for interlukin-1 beta (IL-1β), interlukin-6 (IL-6), interlukin-8 (IL-8) and tumor necrosis factor (TNF) gene transcripts. Data are represented as the fold change against control. B) Western blot analysis of phospho-NF-κB p65 (p-p65) expression in 2DD cells treated with DMSO (0) or Avn C (50 μM, 100 μM and 200 μM) for 48 h. Data are presented as a ratio of p-p65 to β-actin. Representative blot images are shown. C) ChIP-qPCR analysis of phospho-NF-κB p65 binding activity to promoter regions of interlukin-1 beta (IL-1β) and interlukin-8 (IL-8) in 2DD cells treated with control (DMSO) or 100 μM Avn C for 48 h. Data are represented as the enrichment relative to % input. All graphs indicate mean values from three biological replicates. Error bars represent standard deviations. *p-value<0.05, **p-value <0.01 vs control.

Page 24 of 31

Figure 7. Avn C induces heme oxygense-1 expression through increased DNA binding activity of Nrf2 in 2DD fibroblasts. A) 2DD human skin fibroblasts were treated with DMSO (control) or Avn C (50 μM, 100 μM and 200 μM) for 48 h. qRT-PCR on cDNA libraries was performed for glutathione synthetase (GSS), heme oxygenase 1 (HMOX1), superoxide dismutase 1 (SOD1), glutathione superoxidase 1 (GPx1) and catalase (CAT) gene transcripts. Data are represented as fold changes against control. B) Western blot analysis of heme oxygenese-1 (HO-1) expression in 2DD cells treated with DMSO (control) or Avn C (100 μM and 200 μM) for 48 h. Data are presented as a ratio of HO-1 to β-actin. Representative blot images are shown. C) ChIP-qPCR analysis of Nrf2 binding activity to promoter regions of heme oxygenase 1 (HMOX1) and NAD(P)H quinone dehydrogenase 1 (NQO1) in 2DD cells treated with DMSO (control) or 100 μM Avn C for 48 h. Data are represented as the enrichment relative to % input. All graphs indicate the average values from three biological replicates. Error bars represent the standard deviations. *p-value<0.05, **p-value<0.01 vs control.

Page 25 of 31

Figure 8. Avn C reduces proliferative rate of 2DD fibroblasts without inducing cell death. 2DD fibroblasts were treated with DMSO (0) or Avn C (50 μM, 100 μM and 200 μM) for 48 h. A) Population doubling times. B) Trypan blue assays were conducted to detect percent (%) viable cells. C) Ki67 immuno-labelling of 2DD cells. Percent positive Ki67 cells was shown on the Y-axis. All graphs indicate mean values from three biological replicates. Error bars represent standard deviations. *p-value<0.05, **p-value <0.01 vs control.

Figure 9. The in vitro assays of oat 4CL gene (As4CL-1) using the protein expressed in E. coli on three substrates. HPLC analysis of the products on substrate ferulic acid (A), 4-coumaric acid (B), and caffeic acid (C). D, Activity assays of three substrates in a time course. The value in each time point represents the mean of biological triplicate. *: unknown peaks.

Page 26 of 31

Figure 10. The in vitro assays of oat HHT gene (AsHHT-4) using the protein expressed in E. coli. HPLC analysis of the products in the presence of 5-hydroxy-anthranilic acid with p-coumaroyl-CoA (A) and with caffeoyl-CoA (B). *: unknown peaks.

Figure 11. In vitro assays of CCoAOMT-like gene using the protein expressed in E. coli on three substrates. HPLC analysis of the products in the presence of S-adenosyl methionine with Avn-C (A), Caffeoyl-CoA (B) and caffeic acid (C). *: unknown peaks.

Page 27 of 31

Figure 12. Kinetic analysis of oat CCoAOMT-like enzyme. The kinetic constants were estimated from Michaelis-Menten plot using the average of three biological triplicate measurements.

Figure 13. Mutagenesis analysis of the oat CCoAOMT-like enzyme. A. Sequence alignment of oat CCoAOMT-like sequence and related sequences. Conserved residues involved in SAM binding were highlighted in red boxes. Residues involved in divalent binding were highlighted in green boxes. The loop region was between N227 to L250. Mutation site were marked by red *. B. Changes in the activity of oat CCoAOMT-like sequence on three substrates.

Page 28 of 31

Figure 14. The complete biosynthetic pathway of three major avenathramides in oat. PAL, phenylalanine ammonia lyase; C4H, coumaric acid-4-hydroxylase; C3H, 4-hydroxycoumaric acid-3-hydroxylase; 4CL, 4-coumaric acid:CoA ligase; HHT, hydroxycinnamoly-CoA:hydroxyanthranilate N-hydroxycinnamoly transferanse; CCoAOMT-like, caffeoyl-CoA O-methyltransferase-like enzyme.

17. References:

Adom, K.K. and Liu, R.H. (2002) Antioxidant activity of grains. J. Agric. Food Chem. 50, 6182-6187.

Page 29 of 31

Bontpart, T., Cheynier, V., Ageorges, A. and Terrier, N. (2005) BAHD or SCPL acyltransferase? What a dilemma for acylation in the world of plant phenolic compounds. New Phytol. 208: 695-707.

Chu, Yi-Fang, et al. (2013). In vitro antioxidant capacity and anti-inflammatory activity of seven common oats. Food chemistry 139: 426-431.

Collins, F. W. and Mullin, W. J. (1988) High-performance liquid chromatographic determination of avenanthramides, N-aroylanthranilic acid alkaloids from oats. J. Chromatogr. 445: 363-370.

Collins, F.W. (1989) Oat phenolics: avenanthramides, novel substituted N cinnamoylanthranilate alkaloids from oat groats and hulls. Journal of Agriculture and Food Chemistry 37, 60–66.

D’ Auria, J. C. (2006) Acyltransferases in plants: a good time to be BAHD. Plant Biol. 9: 331-340.

Dimberg, L.H., et al. (1996) Variation in oat groats due to variety, storage and heat treatment.I: Phenolic compounds. Journal of Cereal Science 24, 263–272.

Dimberg, L.H., et al. (2001) Stability of oat avenanthramides. Cereal Chemistry 78, 278–281.

Dixon, R.A. and Paiva, N. L. (1995) Stress-induced phenylpropanoid metabolism. Plant Cell. 7: 1085-1097.

Emmons, C.L. and Peterson, D.M. (1999) Antioxidant activity and phenolic contents of oat groats and hulls. Cereal Chemistry 76, 902–906.

Emmons, C.L. and Peterson, D.M. (2001) Antioxidant activity and phenolic content of oat as affected by cultivar and location. Crop Science 41, 1676–1681.

Ferrer, J.L., Zubieta, C., Dixon, R. A. and Noel, J. P. (2005) Crystal structures of alfalfa caffeoyl coenzyme A 3-o- methyltransferase. Plant Physiol. 137: 1009-1017.

Green, A. R., Lewis, K. M., Barr, J.T., Jones, J. P. Lu, F., Ralph, J., Vermerris, W., Sattler, S. E. and Kang, C. (2014) Determination of the structure and catalytic mechanism of sorghum bicolor caffeic acid o-methyltransferase and the structure impact of three brown midrib12 mutations. Plant physiol. 165: 1440-1456.

Gross, G.G. and Zenk, M.G. (1974) Isolation and properties of hydroxycinnamate: CoA ligase from lignifying tissue of forsythia. Eur. J. Biochem. 42: 453-459.

Gutierrez-Gonzalez, J.J., Tu, Z. J. and Garvin, D. F. (2013) Analysis and annotation of the hexaploidy oat seed transcriptome. BMC Genomics. 14: 471.

Gutierrez-Gonzalez, Juan J., Zheng J. Tu, and David F. Garvin. (2013). Analysis and annotation of the hexaploid oat seed transcriptome. BMC genomics 14:471-487.

Hahlbrock, K. and Scheel, D. (1989) Physiology and molecular biology of phenylpropanoid metabolism. Annu. Rev. Plant Physiol. Plant Mol. Biol. 40: 347-369.

Page 30 of 31

Hoffmann, L., Maury, S., Bergdoll, M., Thion, L., Erard, M. and Legrand, M. (2001) Identification of the enzymatic active site of tobacco caffeoyl-coenzyme A o-methyltransferase by site-directed mutagenesis. J Biol Chem. 276: 36831-36838.

Huang, D., Ou, B. and Prior, R.L. 2005. The Chemistry behind Antioxidant Capacity Assays. J. Agric. Food Chem., 53, 1841-1856

Ishihara, A., Matsukawa, T., Miyagawa, H., Ueno, T., Mayama, Shigeyuki. and Iwamura, H. (1997) Induction of hydroxycinnamoyl-CoA: hydroxyanthranilate N-hydroxycinnamoyl-transferase (HHT) activity in oat leaves by victorin C. Z. Naturforsch. 52c: 756-760.

Kopycki, J. G., Rauh, D., Chumanevich, A.A., Neumann, P., Vogt, T. and Stubbs, M.T. (2008) Biochemical and structural analysis of substrate promiscuity in plant. J. Mol. Biol. 378: 154-164.

Landmann, C., Hucherig, S., Fink, B., Hoffmann, T., Dittlein, D., Coiner, H.A. and Schwab, W. (2011) Substrate promiscuity of a rosmarinic acid synthase from lavender (Lavandula angustifolia L.) Planta. 234:305-320.

Liu, L., Zubik, L., Colllins, W., Marko, M. and Meydani, M. (2004) The antiatherogenic potential of oat phenolic compounds. Atherosclerosis. 175: 39-49.

Matsukawa, T., et al. (2000) Occurrence of avenanthramides and hydroxycinnamoly-CoA:hydroxyanthranilate N-hydroxycinnamoyltransferase activity in oat seeds. Zeitschrift f ¨ur Naturforschung C 55, 30–36.

Mayama, S., Tani, T. and Matsuura, Y. (1981) The production of phytoalexins by oat in response to crown rust, Puccinia coronate f. sp. avenae. Physiol Plant Pathol. 19: 217-226.

Meydani, M. (2009) Potential health benefits of avenanthramides of oats. Nutrition Reviews. 67: 731-735.

Molina, I. and Kosma, D. (2015) Role of HXXXD-motif/BAHD acyltransferases in the biosynthesis of extracellular lipids. Plant Cell Rep. 34: 587-601.

Noel, J. P., Dixon, R. A., Pichersky, E., Zubieta, C. and Ferrer, J. L. (2003) Structural, functional, and evolutionary basis for methylation of plant small molecules. Recent Advances in Phytochem. 37: 37-58.

Okazaki, Y., Isobe, T., Iwata, Y., Matsukawa, T., Matsuda, F., Miyagawa, H., Ishihara, A., Nishioka, T. and Iwamura, H. (2004) Metabolism of avenanthramide phytoalexins in oats. Plant J. 39: 560-572.

Oliver, et al. 2011. Model SNP development for complex genomes based on hexaploid oat using high-throughput 454 sequencing technology. BMC Genomics 12: 77.

Peterson, D.M. (2001) Oat antioxidants. Journal of Cereal Science 33, 115–129.

Peterson, D.M., et al. (2005) Relationships among agronomic traits and grain composition in oat genotypes grown in different environments. Crop Science 45, 1249–1255.

Sadiq-Butt, M., Tahir-Nadeem, M., Khan, M.K.I., Shabir, R., Butt, M.S. (2008) Oat: unique among the cereals. Eur J Nutr. 47: 68-79.

Page 31 of 31

Sharma, P. and Singh, R.P. 2013. Evaluation of antioxidant activity in foods with special reference to TEAC method. American Journal of Food Technology, 8, 83-101

Singh, R., De, Subrata. and Belkheir, Asma. (2013) Avena sativa (oat), a potential neutraceutical and therapeutic agent: an overview. Crit. Rev. Food Sci. Nutr. 53: 126-144.

Stockigt, J. and Zenk, M.H. (1975) Chemical syntheses and properties of hydroxycinnamoyl-coenzyme A derivatives. Z. Naturforsch. 30: 352-358.

Sur, R., Nigam, A., Grote, D., Liebel, F. and Southall, M. D. (2008) Avenanthramides, polyphenols from oats, exhibit anti-inflammatory and anti-itch activity. Arch Dermatol Res. 300: 569-574.

Tohge, T., Watanabe, M., Hoefgen, R. and Fernie, A. R. (2013) The evolution of phenylpropanoid metabolism in the green lineage. Crit. Rev. Biochem. Mol. Biol. 48: 123-152.

Walker, A. M., Sattler, S.A., Regner, M., Jones, J.P., Ralph, J., Vermerris, W., Sattler, S. E. and Kang, C. (2016) The structure and catalytic mechanism of sorghum bicolor caffeoyl-coA o-methyltransferase. Plant Physiol. 172: 78-92.

Wang, P., Chen, H., Zhu, Y., McBride, J., Fu, J. and Sang, S. (2015) Oat avenanthramide-C (2c) is biotransformed by mice and the human microbiota into bioactive metabolites. J. Nutr. 145: 239-245.

Wise, M. L. (2011) Effect of chemical systemic acquired resistance elicitors on avenanthramide biosynthesis in oat (Avena sativa). J. Agric. Food Chem. 59: 7028-7038.

Wise, M.L., et al. (2008) Association of avenanthramide concentration in oat (Avena sativa L.) grain with Crown Rust incidence and genetic resistance. Cereal Chemistry 85, 639– 641.

Xu, Z., Zhang, D., Hu, J., Zhou, X., Ye, X., Reichel, K.L., Stewart, N. R., Syrenne, R. D., Yang, X., Gao, P., Shi, W., Doeppke, C., Sykes, R.W., Burris, J.N., Bozell, J.J., Cheng, Z.M., Hayes, D. G., Labbe, N., Davis, M., Stewart Jr, C. N. and Yuan, J.S. (2009) Comparative genome analysis of lignin biosynthesis gene families across the plant kingdom. BMC Bioinformatics. 10: S3.

Yang, Q., et al. (2004) Analysis of the involvement of hydroxyanthranilate hydroxycinnamoyltransferase and caffeoyl-CoA 3-O-methyltransferase in phytoalexin biosynthesis in oat. Molecular Plant-Microbe Interactions 17, 81–89.

Yang, Q., Trinh, H. X., Imai, S., Ishihara, A., Zhang, L., Nakayashiki, H., Tosa, Y. and Mayama, S. (2004) Analysis of the involvement of hydroxyanthranilate hydroxycinnamoyltransferase and caffeoyl-coA 3-o-methyltransferase in phytoalexin biosynthesis in oat. MPMI. 17: 81-89.

Zubieta, C., Kota, P., Ferrer, J. L., Dixon, R.A. and Noel, J. P. (2002) Structural basis for the modulation of lignin monomer methylation by caffeic acid/5-hydrocyferulic acid 3/5-o-methyltransferase. Plant Cell. 14: 1265-1277.