Progress 05/15/19 to 05/14/23
Outputs
Target Audience:The target audiences for this project were the international cereal science and gut microbiome academic communities, member companies of the Whistler Center for Carbohydrate Research, and other companies that use cereals in production (mostly milling and ingredient companies). Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?The project provided training opportunities for 3 graduate students in this final year, as well as partial FTE for a postdoctoral researcher (though funded by internal funds instead of the grant). It also provided training opportunities for 3 undergraduate researchers, who performed independent research projects and communicated their results via posters. It also permitted the graduate students and postdoctoral researcher to present at the Midwest Microbiome Symposium. The graduate students and postdoctoral researchers also had significant opportunities to learn how to mentor more junior scientists, troubleshoot experiments, and communicate their results. All students received personal mentorship from the PI, but also learned to work in a research team in which individuals interacted around their individual portions of the project at least once a week. Finally, the work provided opportunity for students and postdocs to develop protocols for better control of in vitro batch fermentations, which will permit us to bring more rigor and reproducibility to the field as these recommendations are incorporated by other groups. How have the results been disseminated to communities of interest?Presentations by the PI delivered during this project period containing work under this project: "The devil's in the details: different microbial and metabolic fates arise in fermentation of divergent fiber fine structures." Nutrition and Dietetics Seminar Series, Florida International University, April 7, 2023 "Polysaccharide fine structure controls the ecology and function of human gut microbial communities." Biology Seminar Series, Baylor University, February 24, 2023 "Manipulating fiber polysaccharide structure to alter targeting to gut microbiota." International Society of Microbiota 9th World Congress on Targeting Microbiota, Paris, FRANCE, October 21, 2022. "Tuning gut microbiome structure and function using dietary fiber fine structures - at precision and population scales." Council for Responsible Nutrition Science in Session, October 11, 2022 "Using polysaccharides to modulate gut microbiomes for improved health." PI4D Symposium, Purdue University, West Lafayette, IN, September 23, 2022. In addition, the work supported 3 graduate student posters, one postdoc poster, and 3 undergraduate research posters. These posters were presented at the Whistler Center for Carbohydrate Chemistry Technical Conference, the Midwest Microbiome Symposium, and the Purdue Undergraduate Research Symposium. Three manuscripts (with a possible fourth) are being readied for submission. What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
The project resulted in a wealth of data linking processing methods of brans to differential gut microbiome compositional and functional outcomes, resulting in an overarching change of knowledge that processing methods are likely to be influential on how cereals interact with the human gut microbiome. The data produced through this project is now being converted into four manuscripts for publication. Specifically, our project discovered the following: 1. Particle size is influential in recruitment of degrading microbiota and relates to organismal motility. Particle size ranges generated from a single parent bran recruit distinct microbial communities, both of particle-attached and suspended microbes, and those exhibit different successional trajectories. Metagenome sequencing of particle-associated and suspected microbiota revealed that genes involved with flagellar biosynthesis are correlated with organisms that colonize particles over those suspended in solution. These data suggest that microbes are adapted to seek and attach to specific food particles in the relatively less-viscous, less-spatially-constrained cecum and ascending colon, and that these outcomes suggest that specificity of microbes for these particles may be high. Continuing work aims to identify the signals that drive microbial motility towards bran particles. 2. Milling methods alter arabinoxylan composition in resulting particles as relates to particle size. Alkaline-extractable arabinoxylans associated with larger particles are increasingly complex in branch structure compared to those derived from small particles, although milling method used alters the relationship between arabinoxylan complexity and particle size (especially in intermediate size ranges). These data suggest that the chemical structure of particles varies, which provides distinct opportunities for different microbiota to colonize and metabolize. We hypothesis this difference is largely driven by tissue composition and specific modes of damage imparted by different milling methods. These data suggest that divergent processes of milling cereals are likely to 1) provide distinct opportunities for extraction of arabinoxylans of distinct structures, which may serve as a base for new industrial approaches and 2) promote distinct communities of bacteria in the gut microbiomes of humans and productions animals consuming cereal brans. 3. Milling methods alter the relationship of particle size to microbial composition and metabolic outcomes in fermentation. Although there are some general features of particle size that transcend milling method - chiefly that small particles generate increased propionate where larger particles elicit more butyrate production - the relationship of particle size to metabolic and microbial outcomes of fermentations are altered by the milling method used. Some milling methods (e.g. roller) show much larger particle size effects than others (e.g. stone). These outcomes become even clearer when other metabolites besides short-chain fatty acids are considered. In addition, though there are generalities in relationships of particle size to microbial community composition, e.g. larger populations of Bacteroides and Prevotella on smaller particles and larger populations of Lachnospiraceae members on larger ones, there are milling method-dependent influences on how particle size impacts microbiome composition. This likely suggests that different milling methods afford different niches to microbes. The key impact is that it is likely that milling methods and particle sizes can be optimized to increase healthfulness to humans or animals. 4. Particles - not initial microbial community composition - governs the microbiome and metabolic outcomes of fermentation. Although it is clear that gut microbiota adapt to particle size, especially over time, bran-selected communities are flexible to adapt to many different particle structures. Within the space of a single culture, communities selected on one type of wheat bran particle structure adapt to an alternate particle structure, both with respect to microbiome composition and metabolic outputs. The key impact of this discovery is that it appears that the particle's structure - not the initial microbiome composition - governs the fermentative fate of these particles. This suggests that particles engineered or optimized for microbiome function are likely to operate in the desired way across the population, regardless of variation in initial microbiome compositions across individuals. If true, this increases the probability that efforts to optimize bran processing methods are impactful on health of both humans and production animals. 5. Particle size and milling method cause differential extractability and biotransformation of wheat bran phenolics in ways that likely influence their bioavailability and bioactivity in vivo. One unexpected observation of the metabolite analyses of the project's experiments was that fermentations of particles produced different phenolic metabolites in linear relationship to size (either direct or inverse). These phenolic metabolites were further altered by the milling method used to generate the particles. We demonstrated that milling methods were related to the ability of these metabolites to be extracted from brans using solvents, and also to the liberation of unmetabolized phenolics by microbiota in fermentations as well as the anaerobic biotransformation of these metabolites. Given that these phenolics and their metabolites are increasingly implicated in human health, especially with respect to neurobiology and immunity, it seems likely that wheat bran processing influences their health via phenolics in addition to fibers. This observation caused us to redesign how animal experiments (which could not be completed in the grant's time window due to COVID but are now continuing) to detect likely phenolic-driven impacts. In sum, the project's activities convincingly establish wheat bran milling and size as important parameters mediating their interaction with the gut microbiome, and likely their influence on health. These data provide strong justification for future feeding trials that examine whether these differences are influential in humans.
Publications
|
Progress 05/15/21 to 05/14/22
Outputs
Target Audience:In this project period, our primary audience was the national and international cereal science and gut microbiome scientific communities. Our secondary audience were ingredient and food processing companies that employ cereal bran fibers and/or cereal-derived fibers in their products. This second audience was primarily reached through the activities of Purdue University's Whistler Center for Carbohydrate Research. Changes/Problems:There were no major problems or changes to scope during this project period - the work proceeds as described. What opportunities for training and professional development has the project provided?This project supported two graduate students and, partially, one postdoc in my laboratory. In addition, the project benefited (and, in turn, provided training for) four undergraduate students. Both graduate students and postdoc were able to present their work at the Whistler Center for Carbohydrate Research Technical Conference and the Purdue Microbiome Symposium. How have the results been disseminated to communities of interest?The research we performed during this project period was disseminated through the oral presentations below, in whole or in part, given by SRL: April, 2022: "The devil is in the details: how dietary fiber fine structures divergently interact with gut microbiota." Department of Integrative Physiology, University of Colorado, Boulder. (Virtual) March, 2022: "Fiber structure impacts microbiome compositions and metabolic function." PepsiCo, virtual. February, 2022: "Fine cereal fiber structure influences on gut microbiome responses: the effects of genotype and processing." Centro de Desarrollo de Productos Bóticos, Instituto Politécnico Nacional, San Ysidro, MEXICO (Virtual) January, 2022: "Polysaccharide structure and niche structure: how fine structure mediates selection and maintenance of diversity among gut microbial communities." Department of Genetics, Evolution, and Environment, University College London, London, UNITED KINGDOM January, 2022: "Dietary fiber structure influences on microbial diversity, community composition, and metabolic function," Mars Wrigley, virtual. Data generated under this poster was presented by graduate students Miguel Alvarez and Adam Quinn at both the Whistler Center for Carbohydrate Research Technical Conference and the Purdue Microbiome Symposium. What do you plan to do during the next reporting period to accomplish the goals?Having learned the lessons described above, in the final project period we will perform two experiments to determine to what extent varying bran milling methods influence microbiome function and organismal health: 1. Objective 3 animal feeding trials in which various cereal brans (sorghum, wheat, maize) milled to two size ranges (coarse or fine) using two methods (roller or stone) are fed to mice in a high-fat diet context, with the goal of looking for differences in how mouse morphology and microbiota differentially respond over time. 2. In vitro human experiments using standardized cereal bran sizes as substrates across cereals (sorghum, maize, wheat) to identify cereal-specific microbiome effects.
Impacts
What was accomplished under these goals?
In this project period, we completed in vitro sequential cultivation experiments of wheat brans milled to different size fractions and using different techinques (roller, hammer, stone). In addition, we completed the microbial community and metabolic characterization of these fermentations. Together, these fermentations reveal that 1) there are signficiant size-dependencies in microbial community selection on wheat brans and 2) that these size dependencies are not equivalent across milling methods. Specifically, we identified a species from Lachnospiraceae, Kineothrix spp., which is highly adapted to large wheat bran particles, though this effect was magnified in stone-milled brans compared to other methods. In contrast, Prevotella spp. were strongly preferential to small particles, across milling methods. Interestingly, roller-milled particles maintained substantial Prevotella populations at multiple size fractions, where particles generated by stone or hammer milling were significantly less permissive to Prevotella colonization. Scanning electron microscopy analysis revealed that the microbiota degraded the wheat bran particles largely from the aleurone side, suggesting that tissue composition of various size fractions may be driving colonization (rather than damage to plant tissues, per se). Evidence from arabinoxylan extracted from different sized particles suggests that, indeed, carbohydrate linkages vary across particle sizes in tissue-specific ways. Together, these data suggest that 1) milling methods differentially create preferential attachment points for microbial biofilms with respect to particle size and 2) that the differing hemicellulose structures of distinct layers interact differentially with gut microbiota, which then drives differences in community structure. If true, this suggests that milling methods (and size distributions) are potentially influential on human (and animal, as many brans are discarded as waste and are major components of animal feed) health. Further, they suggest that responses to wheat brans across individuals might be fairly deterministic (in that organisms who consume the wheat brans are likely to be similar across humans). The impact of this aspect of the project is that it reveals the non-equivalence of bran fibers in their interaction with the gut microbiome, and may necessitate substantially increased structural resolution on nutritional labeling and in fiber-microbiome interaction studies. In support of this point, further work on this using sorghum arabinoxylans as model systems reveals that these differences in hemicellulose (arabinoxylan) structure are likely determinative for microbiome structure and function. These sorghum arabinoxylans (extracted from red vs. white sorghum) vary subtly in fine structure, but select for significantly different microbial communities and metabolic functions. Together, these data suggest that the responses of gut microbiome to a bran fiber substrate likely are predictable given the consitutient arabinoxylan structure. If true, this provides a new opportunity to both process food with an eye towards increased health and also a new opportunity to inform consumers' dietary habits to select foods most strongly beneficial to them. Inspired by this result, we have investigated genotypic variation in wheat within a whole food context to determine whether differences in arabinoxylan structure observed there may be influencing our results. Briefly, our initial research here suggests that wheat class has a differentiable effect, but that various lines within classes 1) exhibit differences in arabinoxylan structure and 2) interact differently with gut microbiota. The impact of this development is that it opens a new line of inquiry into optimization of cereal genetics for improvement of human health.
Publications
|
Progress 05/15/20 to 05/14/21
Outputs
Target Audience:During this project period, our target audience were the academic and industrial communities; with respect to the latter, specifically, grain- and food carbohydrate-producing members of the Whistler Center for Carbohydrate Research. As is probably expected, COVID-19 significantly reduced our ability to present our research; however, work performed under this project was presented virtually to the academic community at the General Meeting of the American Chemical Society, the 4th Microbiome Microbiome Movement - Human Nutrition Summit, the Global Prebiotic Association's Future of Microbiome Virtual Conference, and the Gut Microbiome Webinar Series of the Institute for the Advancement of Food and Nutrition Sciences. Results from this project were also presented to industrial stakeholders virtually to Nestlé, Kellogg's, Cargill, Mars Wrigley, and PepsiCo through the Whistler Center for Carbohydrate Research. Changes/Problems:As is common, our research on this project was meaningfully delayed by COVID-19. Although we anticipate that this will require us to seek a no-cost extension for our project to complete the aims we proposed, we do not anticipate any significant alterations to the scope of the project. What opportunities for training and professional development has the project provided?This project supported two graduate assistantships and mentoring in their research programs. COVID-19 unfortunately hampered our ability for students to present their work and engage in internal professional development activities. Students were able to present virual posters at Whistler Center meetings to industrial stakeholders, however. Additionally, students participated in laboratory meetings, online journal clubs, seminars available virtually, and in the broader Purdue Applied Microbiome Sciences community, which included virtual meetings among 50+ faculty, postdoctoral fellows, graduate and undergraduate students. How have the results been disseminated to communities of interest?Results produced from this project were disseminated in three academic virual meetings and at five industrial stakeholders. Two publications are currently in preparation for submission. What do you plan to do during the next reporting period to accomplish the goals?Our future plans in the next reporting period are very much in line with the original timeline. We will complete our in vitro fermentaiton experiments, metagenomics, and metabolomics work, and we will also perform our in vivo mouse experiments to determine whether the differences in microbiome responses we observe in vitro translate to in vivo and, if so, to what extent these differences influence mouse physiology.
Impacts
What was accomplished under these goals?
Although COVID-19 significantly hampered our research operations during this project period, we were still able to make significant advances towards our project objectives. Impact Statement: During this project period, we uncovered two key pieces of knowledge that are relevant to optimizing cereal processing for improved health. First, we were able to determine that the differential interaction of the gut microbiome with different sizes of wheat bran particles depend upon the milling method used to reduce particle size. That is, the relationship between microbiome outputs - with respect to bioactive short-chain fatty acids - depends upon which milling method is used. This is important in that it suggests that different methods of processing grains - both with respect to the size to which those grains are milled and the mills employed - may exert differential gut microbiome-mediated health impacts. Second, we learned that within-class variability of wheat cultivars influences microbiome responses, also suggesting that genetics may be employed not only for increased yield but also for improved health characteristics of wheat. Finally, we learned that subtle changes in the polysaccharide structure of bran-derived arabinoxylan fibers significantly influences which microbes are most competitive for their consumption. These data suggest that even the seemingly-minor differences that occur between lines or between cultivation years could be influential in determining how they will interact with gut microbiota. Further, our data suggest that there is a subset of microbiota, which is largely present across many different individuals, that is most efficient in degrading specific fiber types. This may suggest that specific fiber structures may be used across wide swaths of the population to improve gut microbiome function. Progress on Objectives: 1. Quantify the impacts of whole cereal bran type, particle size, and milling methods on the diversity, composition and metabolic function of gut microbiota in vitro. In this project period, primarily using wheat brans processed using different methods (stone, roller, cyclone, hammer, and disc mills), we determined that the outcomes of microbial fermentation depended upon not only size, as we had previously observed, but also which method was used. Most interestingly, the total amount of short-chain fatty acids produced and their ratios varied markedly for a given particle size across methods. We further determined that the reason for this difference is likely to occur due to differences in aleurone content and modifications to aleurone structure, as this tissue displays the most significant microbial attachment and modification in in vitro fermentations. The differences in milling method manifested as alterations in the structural profile of extracted arabinoxylans, as we hypothesized. Furthermore, we have determined that the different particle sizes host substantially different microbial communities, suggesting resource availability to microbes is substantially affected by milling method. Further work is underway to determine to what extent these phenotypes vary across individuals, and further whether these differences in in vitro responses manifest in different in vivo behavior of bran particles. We further determined that flours from different wheat cultivars perform differently in interaction with gut microbes, and that, in some cases, these interactions may be as large among different lines as they are between classes. We are presently following this work up to confirm whether this effect repeats with different initial microbiota. 2. Quantify the impact of soluble bran arabinoxylan chemical structure on diversity, composition and metabolic function of gut microbiota in vitro. We have determined that there are significant arabinoxylan structural differences among brans milled in different ways, and that the arabinoxylan extraction is particle size dependent. In wheats, fermentations are presently underway to determine whether these different arabinoxylans perform distinctly in interactions with the gut microbiome, and to determine to what extent these arabinoxylan structural modifications influence the outcome of microbial competition. In sorghum, we have determined that arabinoxylan from red and white lines behaves quite differently in interaction with the same gut microbiota, despite apparently minor structural differences. These differences manifest more intensely over multiple passages, suggesting the possibility that long-term consumption of specific cereal types may, over time, select for distinct populations of microbiota. We reconstructed member genomes from the fermenting consortia from metagenomes and determined their functional capacity. Specifically, red sorghum arabinoxylan maintained a more diverse population, at the genomic level, than did white sorghum arabinoxylan. Further, these fermentations differed substantially in metabolic outputs, especially with respect to metabolites from bifidobacteria. Finally, we determined that enzymatic debranching of white sorghum arabinoxylan retargeted that polymer to different microbiota, with different microbiome outputs. These data suggest both that small differences in arabinoxylan structure result in very different microbiome interactions, and further that structures may be modified for targeted impact on the gut microbiome. Work continues to determine what metabolic mechanisms undergird these differences.
Publications
|
Progress 05/15/19 to 05/14/20
Outputs
Target Audience:Our target audiences are 1) the scientific community working on cereal grain-microbiome interactions and 2) industrial partners working in cereal products and/or interactions with the microbiome. Although some plans to present this reporting period were disrupted by the COVID-19 pandemic, portions of this work were presented to scientific audiences at the Purdue Microbiome Symposium, at invited seminars at Pacific Northwest National Laboratory, the University of Michigan Center for Gastrointestinal Research, and Case Western Reserve University as well as to industrial stakeholders virtually at the Whistler Center Technical Conference. Changes/Problems:Although there were no major changes to the approach, recruiting for the project was difficult due to timing (compared with timing of graduate student recruitment). However, we are now well-staffed for this project and moving forward at full pace. The second major challenge has been COVID-19, which has substantially slowed our progress late in this reporting period by completely barring access to the laboratory (which was not restored during this project period). What opportunities for training and professional development has the project provided?This project has provided research assistantships for two graduate students and undergraduate research projects for two additional undergraduates. How have the results been disseminated to communities of interest?The project is in its early days and dissemination of results has been significantly impacted by COVID-19. However, this work has been presented in part to in one virtual conference presentation (American Chemical Society Fall Meeting) and four invited talks (Pacific Northwest National Laboratory, University of Michigan, Case Western Reserve University, and University of Nebraska). What do you plan to do during the next reporting period to accomplish the goals?In the next reporting period, we aim to perform the following experiments: Objective I: 1. Identify how stone and roller milling impacts arabinoxylan structure in wheat. 2. Determine how bran tissue composition is different in milled particles. 3. Perform test fermentations of differently sized particles (and different milling methods) in wheat. Objective II: 1. Determine how extracted arabinoxylans from differently sized particles (and different milling methods) are structurally different 2. Determine how extracted arabinoxylans from differently milled particles are fermented by microbiota Objective III: 1. Perform animal experiment with differently sized particles to determine whether there is any impact on host physiology or metabolism.
Impacts
What was accomplished under these goals?
IMPACT: As the gut microbiome is increasingly known to be involved in modulating human health, most principally in influencing the risk and progression of chronic diseases like obesity and diabetes, and diet is increasingly known to be the major driver of the gut microbiome, linking the diet and its components to gut microbiome structure and function is critical for improving health and wellness. Furthermore, most food processing approaches have not been developed and employed with the microbiota in mind; consequently, the impact of processing choices on the healthfulness of products is poorly understood. This is especially true where modern processing technologies far outstrip previous approaches in efficiency (for example, modern roller mills replacing older stone mills). These technologies afford increased efficiency in generating very small particles over older approaches, leading to decreases in the average particle size of grain products consumed as nations westernize.As whole grains (mostly, their bran components) are a major contributor to the dietary fiber intake of Americans, processing-based differences in how microbiota respond to cereal bran fibers may be exerting a significant - but hidden - role in shaping the American gut microbiome and, in turn, health. Thus, the purpose of this project is to determine whether the size ranges and processing methods chosen for cereal brans influences the responses of gut microbiota and, in turn, how those differences impact host health. Despite significant disruption to this project imposed by the COVID-19 pandemic, we have made substantial progress on Objectives I and II. I. Using multiple milling types (cyclone mill, disc mill, and hammer mill, initially), we have determined that arabinoxylan fiber structure and content in wheat bran particles within size ranges (e.g. 180-250 um, 250-300 um, etc.) varies depending upon which processing method was employed. In general, smaller particle sizes revealed fewer branches and simpler branching structure than larger particle sizes, with less 2,3,4-linked (doubly branched) xylose backbones and 2,3,5-linked arabinose branches (more structurally complex branch structures). Conversely, we observed increased 4-linked xylose (undecorated backbone) and 2-, 3-, and 3,5-linked arabinose (single or double sugar substituents on the backbone). Although this trend was observed for all three, this process was dependent upon milling method - cyclone milling exerted the largest impact, followed by hammer and disc milling. Work is currently underway to replicate this effect with roller and stone milling. Also, as one key aspect of milling methods might be differential separation of tissues (rather than modification of polysaccharides), we are also in the process of determining whether tissue composition of milled brans are different as a function of particle size and milling method. In addition, we have completed a trial long-term fermentation of cyclone-milled wheat bran particles in three size fractions. This experiment revealed that the microbiota selected are particle size dependent, as are their outputs. Furthermore, it revealed that very similar microbial species arise in fermentation of wheat brans from three very different donors, suggesting that these organisms are specifically adept at fermenting these substrates. These data validate that the method is working and also that investigating particle size effects on the microbiome is likely to reveal important linkages between bran processing and microbiome responses. II. We have determined that 1) subtle, naturally occurring variation in arabinoxylan structure substantially alters microbial communities and 2) that in vitro debranching of arabinoxylans leads to a very different succession of organisms, compared with native arabinoxylans. The majority of these trial experiments have been conducted in sorghum arabinoxylan, as it is relatively evenly substituted and with a relatively simple branch structure. However, we have determined that red and white sorghum arabinoxylan select for different communities across multiple initial sets of human microbiota and, further, pretreatment with enzymes that debranch the arabinoxylan alter which microbes are successful. These data suggest that the differences in processing observed in Obj. I are indeed likely to modulate microbiome responses via differences in arabinoxylan structure. III. This objective has not yet begun.
Publications
|
|