Recipient Organization
DELAWARE STATE UNIVERSITY
1200 NORTH DUPONT HIGHWAY
DOVER,DE 19901
Performing Department
(N/A)
Non Technical Summary
Snacks are enjoyed by both children and adults. However, the growing prevalence of metabolic syndrome has been attributed to physical inactivity, unhealthy eating habits, and obesogenic environments such as the availability of calorie-dense nutrient-poor snacks. With increasing consumer appeal for ready-to-eat (RTE) products including snacks with nutritional, sensorial, and healthful properties, consumers are drawn to plant-based food (PBF) products. One strategy is to leverage the increasing interest in plant-based products, that also meet consumer demand for food and nutrition security, sustainability, and healthier options. Alternatives like lentil and quinoa have received relatively little attention as ingredients for food application due in part to the many limitations in their nutritional profile, functionality, and sensory properties. It is still unclear how they impact gut microbiota and health. Our overarching goal is to improve the acceptability and economic value of lentil and quinoa by designing alternative processing conditions that develop value-added ingredients and products with enhanced nutritional, techno-functional bioactive, and sensory properties. This project is a collaboration between PIs at Delaware State University, and Alabama A&M University with support from researchers at NCAT, UMD, USDA-ARS-ERRC, and InfraReady Products Ltd, Canada. Our project proposes to: (1) determine effect of processing on key ingredients properties; (2) develop and characterize formulated puffed snacks; and (3) assess impact of metabolites on gut microbiota and ameliorating effect on obesity. We will demonstrate the quality and value of these snack products through in vitro, consumer studies, animal models to validate their nutritional and health impact, consumer acceptability compared to traditional counterparts. We will incorporate immersive machine learning training simultaneously with students' experiential research, that will improve students' employability, and equip PIs with resources to translate research to commercial innovation.
Animal Health Component
40%
Research Effort Categories
Basic
20%
Applied
40%
Developmental
40%
Goals / Objectives
There is increasing consumer appeal for ready-to-eat (RTE) products including snacks with nutritional, sensorial, and healthful properties. Consumers are also increasingly drawn to plant-based food (PBF) products due to multiple factors such as food and nutrition security, health benefits, rising access and interest, sustainability, ethics, and the impacts of modern food production and supply. With the growing prevalence of metabolic syndrome, largely due to unhealthy diet, physical inactivity, tobacco use, and obesogenic environments, efforts aimed at improving lifestyle and diet have garnered attention. Although there are inconsistencies in the direct link between snacking and weight gain, recent studies have reported increased content of sugar and unhealthy fat compared to a few decades ago. The obesogenic environment includes a dizzy array of choices of often cheap, easily accessible, and calorie-dense nutrient-poor snacks. Our goal is to develop snacks with abundant and unexplored nutrient?rich plant-based ingredients such as legumes (e.g., pulses - lentil) and cereals including pseudo-cereals (e.g., quinoa). This project seeks to demonstrate the quality and value of snack formulated using plant-based ingredients through in vitro, consumer studies, animal models to validate their nutritional and health impact, consumer acceptability compared to traditional counterparts. The project will also incorporate immersive machine learning training simultaneously with students' experiential research, improve students' employability, and equip PIs with resources to translate research to commercial innovation. Our project proposes to: (1) determine effect of processing on key ingredients properties; (2) develop and characterize formulated puffed snacks; and (3) assess the impact of metabolites on gut microbiota and ameliorating effect on obesity. These will be accomplished with various research projects designed for MS and Ph.D. students. Undergraduate students will collaborate on all aspects of the research outlined.
Project Methods
Processing of lentil and quinoa ingredients and characterization: Lentil and quinoa will be milled using a knife mill, micronized by InfraReady Products Ltd., Canada and protein isolate will be produced by isoelectric precipitation, and freeze-drying. The flours and protein powder will be collectively called ingredients. Proximate composition will be determined according to the AOAC methods. In vitro starch digestibility, insoluble and soluble dietary fiber contents will be determined using assay kits from Megazyme. Protein quality will be established from digestibility and amino acid composition and in vitro protein digestibility corrected amino acid score will be calculated. Antinutritional factors (trypsin inhibitors, phytic acid, tannins, and saponins) and techno-functional properties (solubility, water and fat holding capacity, gelation, pasting and emulsification) will be determined by standard protocols and assays kits. Content of selected minerals, molecular weight, color, particle size distribution, structural, thermal properties, TPC and phenolic profile of the alcoholic extracts and in vitro antioxidant capacity will be determined.Formulation of extruded puffs: Fourteen different combinations of extruded puffs will be developed from the blend of the lentil and quinoa ingredients. Specific control of processing variables in extrusion-cooking will be optimized to meet product acceptability. Extrusion-cooking will be conducted using theTS-45 twin screw extrusion-cooker after conditioned to specific dough moisture content, mixed, and proofed before undergoing the three stages of extrusion. The diameter of the extruder screw and the barrel inside diameter, and the screw compression ratio will be optimized. The product will be shaped with selected circular openings die and screw speeds. Post-extrusion, the products will be vacuum oven dried, cooled, and packaged in plastic bags at rt. Specific mechanical energy consumption and pressure values of the extruder die will be obtained from the transducer. Radial expansion, puffing percentage and bulk density will be determined. Morphology, textural, thermal, color, proximate, mineral, ANF content, digestibility and bioaccessibility of the extrudates will be determined. Sensory evaluation: Ethical approval for the involvement of human subjects will be obtained from the DSU's IRB. The sensory analysis of the extrudates will be performed by consumers in three different steps on the same day. In the first phase (screening), the duo-trio test (ASTM - E2610-082011) will be employed with 150 panelists using Compusense software version 5.0 for sensory evaluation, data collection, and statistical analysis. The objective will be to select the panelists with better acuity to distinguish the differences between the samples, proving that they have more accurate sensory perceptions and that they are thus more suitable to participate in the second phase (focus group). The 15 panelists who had the highest number of correct answers will be selected for the second phase. The focus group will be invited to discuss the samples in four one-hour sessions. Open-ended questions will be carefully asked to not induce panelists' responses to select product attributes that please or dislike consumers and how the product met their expectations. The panelists will develop a list of attributes regarding the appearance, smell, taste, texture, and global impression of puffs and the PIs and panelists will exclude attributes that have the same meaning, which could confuse consumers of the third phase and the ones that were least mentioned. The list will be applied in the third phase (consumer test), using the Check-All-That-Apply methodology, 120 panelists, with acceptance of each sample and purchase intention on a nine- and five-points hedonic scale, respectively.Predicting consumer acceptability using ML models: ML algorithms will be used for predictive modelingto analyze historical consumer perference data to predict future trends and increase their chances of market success, without the continue use of sensory scores.Relevant features will be extracted from sensory data (appearance, smell, taste, texture, and global impression attributes) and cleaned, handling any missing values to ensure consistency in formatting. Data will be split the into training and validation sets, ensuring that each set represents a diverse range of panelists.Various ML learning algorithms (e.g., SVM, SVR, LightGBM, KNN, and XGBoost regressors) will be trained using k-fold validation to assess performance in accurately predicting panelists' acuity. For each iteration of training and validation, performance metrics such as accuracy, precision, recall, or mean squared error will be computed based on the model's predictions. Feature importance provided by ML models will be used to identify which sensory attributes contribute most to distinguishing products with higher acuity.?Effect on adiposity, glucose tolerance and insulin resistance: Diets will be formulated using the nutrient profile of the puffed snack. The intervention study in C57BL/6J male mice (n=10) will test the impact of the snack on protection against diet induced obesity, insulin resistance and inflammation. Mice will be randomized into 3 groups: Low fat diet (LF with 10% fat); High fat diet (HF with 45% fat) and High fat diet (45% fat supplemented with 4% puffed snack (HFPS). The feeding will last 16 weeks. Weekly food intake and body weight, insulin resistance (at baseline, week 8 and week 15) and glucose tolerance by the oral glucose tolerance test (week 15) will be determined. At euthanasia, the epididymal, perirenal, renal, and mesenteric fat pads will be weighed to determine fat accumulation. Effect on small intestinal and colon microbiome community: Bacterial DNA will be isolated from the small intestine (duodenum, jejunum) and colon contents. Microbiota membership (16S rRNA sequencing) and bioinformatics (Ion ReporterTM Software) will be evaluated. Specific bacterial taxa of interest identified by 16S sequencing will be validated by qPCR and short-chain fatty acids by GC. Effect on fat absorption, oxidation, lipolysis, and lipogenesis markers: Small intestine tissue, epididymal fat pad, skeletal muscle, and liver tissues collected and frozen at euthanasia will be used to determine differential expression of genes (by qPCR) and proteins (by western blotting). Targets will be genes/proteins involved in lipid uptake, transport, lipolysis, fat oxidation, TAG formation/lipogenesis and fatty acid oxidation. TAGs will be extracted and quantified in the fecal and cecal contents as a measure of intestinal fat absorption and in liver and muscle as a measure of fat accumulation. Effect on intestinal inflammation and gut barrier integrity: In week 15, fresh fecal samples will be used to quantify Lipocaline 2 (Lcn-2) using the Duoset Lcn-2 ELISA kit. Lcn-2 expressed mainly by neutrophils is a real-time marker of gut inflammation. Fresh feces collected in RNAlater will be used as source of gut-exfoliated cells to determine expression of inflammatory cytokines using an inflammation pathway focused panel RT² Profiler PCR Array (Qiagen, MD). Significant changes observed will be followed up by specific qPCR to confirm mRNA expression. We will determine protein and gene expression of IL-10, ICOS, and TGF-β as health-promoting cytokines enhanced by the puffed snack. Expression and enzyme activity of intestinal alkaline phosphatase (IAP) in fecal contents will be determined by ELISA and colorimetric phosphatase kit, respectively. IAP released by enterocytes constitutes the functional gut barrier; it inactivates inflammatory lipopolysaccharide (LPS) and enhances the expression of tight junction proteins in the epithelial layer. Expression of Tight junction proteins and Adheren junction proteins in intestinal tissue will be determined by qPCR and western blotting.