THE MILK EXOSOME - BACTERIAL VESICLE - HOST HEALTH TRIAD

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: ACTIVE
• Funding Source: AFRI COMPETITIVE GRANT
• Reporting Frequency: Annual
• Accession No.: 1029779
• Grant No.: 2023-67017-39056
• Cumulative Award Amt.: $638,000.00
• Proposal No.: 2022-09417
• Project Start Date: Mar 1, 2023
• Project End Date: Feb 28, 2027
• Grant Year: 2023
• Program Code: [A1343]- Food and Human Health

Project Director
Zempleni, J.

Recipient Organization
UNIVERSITY OF NEBRASKA
(N/A)
LINCOLN,NE 68583

Performing Department
(N/A)

Non Technical Summary
Human and cow's milk contain large amounts of natural nanoparticles (milk exosomes, MEs), whereas plant-based milk and infant formula contain no MEs. Prior research has shown that a ME-deficient diet impairs spatial learning and memory (SLM) and increases the severity of epilepsy-like seizures. This projects studies whether the positive effects of MEs in the host are mediated by ME-dependent changes in nanoparticles secreted by the gut microbiome. [Gram-negative bacteria secrete outer membrane vesicles (OMVs), whereas Gram-positive bacteria secrete cytoplasmic membrane vesicles (CMVs).] This proposal tests the hypothesis that dietary ME depletion alters the quantity (number) and quality (content of proteins and RNA) of OMVs and CMVs, thereby impairing SLM in the host. The researchers will study 1) effects of MEs on OMV and CMV quantity and quality in cultures of gut bacteria, 2) absorption and tissue distribution of OMVs, CMVs and their protein and RNA cargo in the host, 3) effects of ME-dependent change in OMVs and VMVs on gene expression in the host, and 4) whether ME-dependent changes in OMV and CMV are responsible for the effects of MEs on SLM n the host observed in prior research.

Animal Health Component

30%

Research Effort Categories

Basic

70%

Applied

30%

Developmental

0%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
724	3450	1010	100%

Knowledge Area
724 - Healthy Lifestyle;

Subject Of Investigation
3450 - Milk;

Field Of Science
1010 - Nutrition and metabolism;

Keywords

bacterial extracellular vesicles

brain

exosomes

microbiome

milk

Goals / Objectives
Objective 1. Assess the effects of MEs on OMV and CMV biogenesis and cargo content in human gut bacteria. Objective 1 tests the hypothesis that MEs alter the secretion of OMVs and CMVs and their cargos content in fecal (mixed) cultures ex vivo and in bacterial monocultures.Objective 2. Assess the bioavailability and tissue distribution of OMVs and CMVs in mice. Objective 2 tests the hypothesis that bacterial EVs, secreted by human gut bacteria, and their RNA and protein cargos are bioavailable and accumulate in peripheral tissues.Objective 3. Determine whether OMVs and CMVs play a causal role in mediating effects of ME on SLM. Objective 3 tests the hypothesis that depletion of bacterial EVs abolishes effects of MEs on spatial learning and memory in mice.

Project Methods
Objective 1Bacteria. We will test both bacterial monocultures and infant feces to achieve controlled conditions (monocultures) and capture the full spectrum of gut bacteria and interactions among species (feces). In monocultures, we will test E. coli strain B, E. coli Nissle 1917, B. infantis, and Lactobacillus gasseri.Milk exosome (ME)-defined cultures. MEs will be purified from human milk using ultracentrifugation and authenticated by nanoparticle tracking analyzer (NTA), transmission electron microscopy (TEM), and immunoblots. Bacteria will be cultured in an anaerobic chamber in ME-free (MEF) and ME-supplemented (MES) media. The ME concentration in MES media is nutritionally relevant. Bacteria will be cultured for up to 20 hours, and fresh MEs will be added every two to four hours, mimicking feeding frequency in infants. Outer membrane vesicles (OMVs) and cytoplasmic membrane vesicles (CMVs) will be harvested from exponentially growing Gram-negative and Gram-positive bacteria, respectively.OMV and CMV analysis. OMVs and CMVs will be harvested and authenticated as described for MEs. In addition to rates of bacterial OMV and CMV secretion, we will assess the content of RNA by sequencing analysis, and proteins by mass spectrometry analysis.Experimental endpoints. The primary experimental endpoints will be the rate of EV secretion, and the RNA and protein profiles in OMVs and CMVs secreted by bacteria grown in ME-defined cultures.Statistical analysis. Homogeneity of variances among groups will be assessed using Bartlett's test. If variances are not equal, data will be log transformed. Conformity to normal distribution will be determined using the Kolmogorov-Smirnov test. For comparisons of paired samples, we will use the paired t-test or Wilcoxon test, depending on the normality of data distribution. One-way ANOVA will be used for comparison of more than two treatments. Dunnett's posthoc test or Fisher's Least Significant Difference test will be used for posthoc testing, depending on data distribution. Repeated measures analysis will be used in time course experiments. Differences will be considered significant if P<0.05.Power calculation. We conducted sample size calculations to estimate the power of the proposed studies to detect differences among groups. Four bacterial cultures will be sufficient to detect a 3% change in OMV secretion. A 3% change is biologically important for the host.Objective 2Labeling of bacterial extracellular vesicles (EVs) and their RNA and protein cargos. We will isolate and authenticate OMVs and CMVs as described in Objective 1. We are well versed in ME and cargo labeling strategies. In preliminary studies we adopted these technologies for use in OMVs and CMVs. Our previous studies suggest that distinct ME cargos have distinct tissue distribution profiles. Because of that observation we will separately assess the distribution of OMVs, CMVs, protein cargo and RNA cargo as follows.EVs: OMVs and CMVs will be labeled with carbonyl-reactive HiLyte 750.Proteins: The tissue distribution of proteins will be assessed by transforming bacteria to express Cre recombinase in OMVs and CMVs. Cre-dependent recombination events in Cre reporter mice lead to a change from expressing tdTomato to expressing EGFP in tissues and will be used as readout for protein accumulation in tissues.RNA: For assessing the tissue distribution of RNA we will load OMVs and CMVs with synthetic, IRDye-labeled RNA as previously described. RNAs will be chosen from the ten most abundant RNAs in sequencing analysis.Bioavailability and distribution. Studies will be conducted in male and female C57BL/6J mice, ages two and four weeks.EVs: HiLyte-labeled OMVs and CMVs will be administered by oral gavage using a dose-response time-course design. In a previous study and preliminary studies, OMVs and CMVs accumulated in the intestinal mucosa, heart, liver kidney, spleen, and brain. We will focus on these tissues but also monitor other tissues. Feces will be collected for estimating the percent absorption Time courses will also be established following intravenous administration to establish a reference for 100% bioavailability.Proteins: The analysis of protein cargos in bacterial EVs depends on Cre-recombination events in reporter mice colonized with transgenic bacteria transformed to secrete Cre-positive EVs. Reporter mice will be treated with a mixture of antibiotics known to deplete the microbiome via drinking water for six days. Antibiotic-treated mice will be colonized with Cre-positive bacteria in drinking water following cessation of antibiotics for the duration of the study.RNA: The accumulation of IRDye-labeled RNA, delivered by OMVs and CMVs from bacterial monocultures and infant feces cultures will the assessed using a dose-response, time course protocol.Experimental endpoints. The primary experimental endpoints will be the accumulation of OMVs and CMVs, and their protein and RNA cargos in tissues and their fecal excretion.Statistical analysis. As described in Objective 1.Power calculation. Six mice will be sufficient to detect a 50% change in OMV accumulation in the liver, which is well within the 60% decrease in the levels of plasma microRNA levels in mice fed ME-defined diets.Objective 3Mice. We will determine whether OMVs and CMVs play a causal role in mediating effects of ME on SLM in male and female C57BL/6J mice ages two five weeks.Treatment groups. We will use a 2x2 factorial design with ME-defined and treatment with antibiotics (depletion of OMVs and CMVs) as independent variables. Treatment will be initiated at age two weeks and SLM will be assessed periodically until age five weeks. Until weaning, mice will be fed human milk (rich in MEs) or infant formula (free of MEs) with or without an antibiotics cocktail. At weaning, the mice will be transitioned to ME-defined diets.Assessment of spatial learning and memory (SLM). SLM will be assessed in the Barnes maze.Host gene expression. We will assess effects of antibiotics on ME-dependent gene expression in the intestinal mucosa and the hippocampus by RNA-sequencing analysis and KEGG pathway analysis by RNA-sequencing analysis.Experimental endpoints. The primary experimental endpoints will be the time needed to locate the escape hole in the Barnes maze, and gene expression patterns in the intestinal mucosa and hippocampus.Statistical analysis. As described in Objective 1.Power calculation. Seven mice will be sufficient to detect a 30% change in SLM. An effect that size is well within the 9-fold difference of SLM in mice fed ME-defined diets.

Progress 03/01/24 to 02/28/25

Outputs
Target Audience:The target audience includes faculty, postdocs, and students in nutrition sciences; pediatricians and neonatologists; stakeholders in dairy research, production and industry; policy makers; journalists; and the lay public. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The project has provided professional development opportunities for the program director and postdoctoral associate. Members of the Zempleni laboratory have acquired knowledge in protocols of the gut microbiome and signaling by bacterial vesicles. How have the results been disseminated to communities of interest?We have disseminated our results to our audience through and invited seminars, presentations at local and national conferences, and blogs [see Products (previous year) and Other Products (this year)]. What do you plan to do during the next reporting period to accomplish the goals?Objective 1 Complete the RNA mapping and annotation of RNA-sequencing data from cultures of Bifidobacterium infantis and E. coli. Complete wet lab omics studies (RNA-seq, proteomics/mass spec, and lipidomics/mass) from ME-defined monocultures of E. coli Nissle 1917, E. coli Migula 1895, and L. gasseri. Complete wet lab omics studies (RNA-seq, proteomics/mass spec, and lipidomics/mass) from fecal cultures from adults. Submit samples of fecal cultures from infants for wet lab omics analysis (RNA-seq, proteomics/mass spec, and lipidomics/mass). Objective 2 Conduct bioavailability studies of Cre-loaded CMVs and OMVs from E. coliNissle 1917,E. coliMigula 1895, andL.. gasseri in tdTomato/EGFP reporter mice. Covalent labeling: Complete. Objective 3 Initiate studies of spatial learning and memory in Objective 3. Other Ensure that the manuscripts that are currently under review will be accepted for publication. Submit one or two original research manuscripts to reputable peer-review journals.

Impacts
What was accomplished under these goals? Central hypothesis. We hypothesize that gut bacteria transmit food signals (MEs) to the host through altering the quantity and quality of OMVs and CMVs, thereby eliciting cognition phenotypes in the host. Long-term goal. The long-term goal of this project is to understand the crosstalk between animal and bacterial kingdoms through MEs, OMVs and CMVs and its importance to human health: MEs (human/animal) → OMVs, CMVs (bacteria) → health (human/animal). Objective 1. Assess the effects of MEs on OMV and CMV biogenesis and cargo content in human gut bacteria. Objective 1 tests the hypothesis that MEs alter the secretion of OMVs and CMVs and their cargos content in fecal (mixed) cultures ex vivo and in bacterial monocultures. Objective 2. Assess the bioavailability and tissue distribution of OMVs and CMVs in mice. Objective 2 tests the hypothesis that bacterial EVs, secreted by human gut bacteria, and their RNA and protein cargos are bioavailable and accumulate in peripheral tissues. Objective 3. Determine whether OMVs and CMVs play a causal role in mediating effects of ME on SLM. Objective 3 tests the hypothesis that depletion of bacterial EVs abolishes effects of MEs on spatial learning and memory in mice. Objective 1 In the previous reporting period, we used reporter plasmids to demonstrate that Bifidobacterium infantis and Escherichia coli (lab strain) internalize milk exosomes (MEs). Consistent with our central hypothesis MEs promoted bacterial growth and in the culture media. MEs increased the secretion of cytoplasmic membrane vesicles (CMVs) by Gram-positive B. infantis and decreased the secretion of outer membrane vesicles (OMVs) by Gram-negative E. coli. In this reporting period we have completed the proteomics and lipid analyses of CMVs and OMVs from B. infantis and E. coli (lab strain). RNA-sequencing analysis is complete. We identified 2,000 sequencing reads in CMVs, and 4,000 sequencing reads in OMVs. The annotation is ongoing. We have initiated the ME-defined monocultures of the other bacteria that we named in our grant proposal, E. coliNissle 1917,E. coliMigula 1895, andLactobacillus. gasseri. Effects of ME supply on the secretion on CMVs and OMVs, and the vesicle cargo (RNA, proteins, lipids, and small metabolites) are in progress. We have completed cultures of adult feces in media defined by the content of cow's milk exosomes, and infant feces defined by the content of human milk exosomes. The samples from adults have been submitted for RNA-sequencing analysis, metagenomics analysis, and metabolomics analysis. Objective 2 We used two strategies for assessing bioavailability and distribution of CMs and OMVs in mice: Cre recombinase and reporter mice: In the previous reporting period we engineered Bifidobacterium infantis and E. coli (lab strain) to secrete CMVs and OMVs, respectively, loaded with Cre recombinase. Bioavailability and distribution were assessed in tdTomato/EGFP reporter mice. The bioavailability of CMVs was greater than that of OMVs. CMVs accumulated primarily in liver, kidneys, and lung; Cre-driven recombination events were also detected in the heart and intestinal mucosa. OMVs are localized almost exclusively to the intestinal mucosa. In this reporting period, we engineered the three other species, E. coliNissle 1917,E. coliMigula 1895, andL.. gasseri to express and secrete CMVs and OMVs loaded with Cre recombinase. Bioavailability studies in reporter mice are in progress. Covalent labeling: CMVs were purified from cultures of B. infantis and Lactobacillus gasseri using ultracentrifugation. OMVs were isolated from cultures of E. coli Nissle 1917, E. coli (Migula 1895) Castellani and Chalmers. CMVs and OMVs were covalently labeled with carbonyl-reactive HiLyte-750 and administered by oral gavage [1.5x10 (exp)10 particles per g body weight], and tissues were harvested before and at timed intervals (3, 6, 9, and 12 hours) after gavage. Both CMVs and OMVs were detected primarily in the lung, followed by accumulation (in rank order) in heart, kidneys, spleen, and liver. No CMVs and OMVs were detectable in the brain. We consider this arm of the project to be complete. Objective 3 We have not yet initiated studies of spatial learning and memory in Objective 3. Other We have submitted two manuscripts to peer-reviewed journals. The manuscripts are under review. We disseminated research discoveries at four national and six state-wide conferences.

Publications

Progress 03/01/23 to 02/29/24

Outputs
Target Audience:The target audience includes faculty, postdocs, and students in nutrition sciences; pediatricians and neonatologists; stakeholders in dairy research, production and industry; policy makers; journalists; and the lay public. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The project has provided professional development opportunities for the program director and postdoctoral associate. Members of the Zempleni laboratory have acquired knowledge in protocols of the gut microbiome and signaling by bacterial vesicles. How have the results been disseminated to communities of interest?We have disseminated our results to our audience through and invited seminars, presentations at local and national conferences, and blogs (see Products). What do you plan to do during the next reporting period to accomplish the goals?Objective 1 Complete the analysis of data generated in cultures of Bifidobacterium infantis and E. coli. Initiate ME-defined monocultures of E. coli Nissle 1917, E. coli Migula 1895, and L. gasseri and analyze CMV and OMV secretion, respectively. Submit samples for RNA-seq, proteomics, and metabolomics analysis. Objective 2 Assess the bioavailability and distribution CMVs and OMVs secreted by of E. coli Nissle 1917, E. coli Migula 1895, and Lactobacillus gasseri using both HiLyte labeling and Cre reporter mice. Conduct dose-response studies of HiLyte-labeled CMVs and OMVs secreted B. infantis and L. gasseri, respectively. Objective 3 Studies of spatial learning and memory in Objective 3 will not be initiated until Year 3. Other Submit one or two original research manuscripts to reputable peer-review journals.

Impacts
What was accomplished under these goals? We loaded bovine milk exosomes (MEs) with reporter plasmids and demonstrated that Bifidobacterium infantis and Escherichia coli (lab strain) internalize MEs. We cultured Bifidobacterium infantis and E. coli in media defined by their content of human milk exosomes (MEs). We chose levels of MEs that are nutritionally relevant: ME-supplemented (MES) media represent breastfed infants; ME-free (MEF) media represents formula-fed media. (Human milk is rich in MEs, whereas formula is free of MEs). Bifidobacterium infantis and E. coli grew 19% and 2% faster in MES compared to MEF cultures, respectively. The secretion of cytoplasmic membrane vesicles (CMVs) by B. infantis was 8% higher in MES compared to MEF cultures. The secretion of outer membrane vesicles (OMVs) by E. coli was 7% lower in MES compared to MEF cultures. (Secretion of CMVs and OMVs was normalized by bacterial mass.) We completed RNA-seq and proteomics analysis in bacteria (cells) and we completed RNA-seq, proteomics, and metabolomics analysis in CMVs and OMVs in the above cultures. The annotation of RNA is ongoing. We identified 1199 unique proteins in B. infantis, and we identified 217 and 141 unique proteins and metabolites in CMVs secreted by B. infantis, We identified 1,448 unique proteins in E. coli, and we identified 110 and 141 unique proteins and metabolites in OMVs secreted by E. coli. The data are undergoing pathway and statistical analysis. Objective 2 We used two strategies for assessing bioavailability and distributution of CMs and OMVs in mice: Cre recombinase and reporter mice: We engineered Bifidobacterium infantis and E. coli (lab strain) to secrete CMVs and OMVs, respectively, loaded with Cre recombinase. Bioavailability and distribution were assessed in tdTomato/EGFP reporter mice. The bioavailability of CMVs was greater than that of OMVs. CMVs accumulated primarily in liver, kidneys, and lung; Cre-driven recombination events were also detected in the heart and intestinal mucosa. OMVs localized almost exclusively to the intestinal mucosa. Covalent labeling: CMVs were purified from cultures of B. infantis and Lactobacillus gasseri using ultracentrifugation. OMVs were isolated from cultures of E. coli Nissle 1917, E. coli (Migula 1895) Castellani and Chalmers. CMVs and OMVs were covalently labeled with carbonyl-reactive HiLyte-750 and administered by oral gavage [1.5x10 (exp)10 particles per g body weight], and tissues were harvested before and at timed intervals (3, 6, 9, and 12 hours) after gavage. Both CMVs and OMVs were detected primarily in the lung, followed by accumulation (in rank order) in heart, kidneys, spleen, and liver. No CMVs and OMVs were detectable in the brain. Objective 3 We have not yet initiated studies of spatial learning and memory in Objective 3.

Publications

• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Mumtaz P. T and Zempleni J. Exploring Cross-Kingdom Communication with Microbial Messengers: Bifidobacterium infantis EVs' Diverse Cargo, Bioavailability in Mice, and Interaction with Human Intestinal Cells (Caco 2). University of Nebraska, NPOD 9th Annual Research Symposium, September 26, 2023
• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Zempleni J. Milk exosomes and their relevance in human nutrition and the delivery of therapeutics. University of Wisconsin-Madison, October 12, 2023 [invited talk]
• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Zempleni J. Biopharming: engineering EVs in milk for delivering therapeutics. Annual Conference of the American Association of Extracllular Vesicles, Boston, MA, October 29, 2023 [invited talk]
• Type: Other Status: Published Year Published: 2023 Citation: Exosome RNA (Exosome RNA Research & Industry News) Milk nanoparticles biological signaling may have important benefits for cognitive development. https://exosome-rna.com/milk-nanoparticles-biological-signaling-may-have-important-benefits-for-cognitive-development/
• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Zempleni J. Gene delivery by milk exosomes restores Syngap expression in mouse brains. Annual meeting of the Syngap Research Fund. Orlando, FL, November 30, 2023 [invited talk]
• Type: Other Status: Published Year Published: 2023 Citation: USDA funding supports Zempleni research on breastmilk consumption and brain development (by Geitner Simmons). https://cehs.unl.edu/nhs/news/usda-funding-supports-zempleni-research-breastmilk-consumption-and-brain-development/