Progress 09/01/17 to 08/31/21
Outputs
Target Audience:The primary target audience for this project are scientists working in the area of horizontal gene flow and the development of probiotics. The work has importance for regulators of genetically modified organisms, livestock producers, environmental conservationists, and biotechnology companies. Regulatory agencies, including the USDA, FDA, EPA, APHIS and also the NIH will be interested in this work. We have introduced our work via a poster at the annual USDA-BRAG meeting, where we met administrators and officials from these various institutions. We have also presented the work at scientific conferences and departmental seminars. We have published some of the work in peer-reviewed journals. Changes/Problems:We were unable to obtain several of the compounds needed to determine whether we could reduce the rates of HGT (Objective #3). Rather, we switched to determining how selection (by antibiotics or by a prebiotic) served as a method for enhancing HGT. To accomplish this, we designed and constructed plasmids carrying either antibiotic resistance genes and a gene for degrading and utilizing 6'SL (6'-Sialyllactose). We showed that these organisms can be selected for, and we have performed experiments in mouse models treated with antibiotics or 6'SL to determine the spread of these plasmids. Using OIL-PCR, a method developed as part of Objective #2, we can determine which organisms have acquired the plasmid. What opportunities for training and professional development has the project provided?The students involved in this study have participated in work-in-progress meetings on campus and regular meetings with their committee members. Our ability to go to conferences has been curtailed by the pandemic, although two students have presented in online conferences and are registered to present at a Keystone microbiome meeting in Banff in early 2022. Several students have taken online workshops to improve their ability to analyze metagenomic data and they have participated in the Cornell BEST program, which exposes students to a variety of career options after obtaining their PhD. How have the results been disseminated to communities of interest?The community of interest is largely academic researchers. The PI has presented at several departmental seminars at: University Of Pennsylvania, Penn State College, Memorial Sloan Kettering Cancer Center, Queen's University, Columbia University, the University of Louisville, the University of Washington and Duke University (scheduled for January 2022). The PI has also presented at the ASM Microbe meeting, the McGill Bicentennial Event on Antimicrobial Resistance and the Microbiome, and the Intelligent Systems for Molecular Biology meeting. The students have presented at the International Conference on Microbiome Engineering and are planning on presenting at the Keystone Meeting on Microbiomes and Antimicrobial Resistance in Banff in early 2022. With support from the NSF, we also held two microbiome hackathons in 2018 and 2019 (the 2020 was canceled due to pandemic related shutdowns.) Here, students were exposed to microbial genetics in a fun, low-stress environment. What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
Objective 1: Model the gene flow between livestock, wild animals, humans and the environment by analyzing overlap between their microbiome compositions and their mobile gene pools. (Years 1-2, 90% complete) We performed metagenomic sequencing on dairy cows, pigs and goats (approximately 10 animals per farm) from the region. To model gene flow, we developed a large-scale model of over 12,000 isolated genomes (from more than 10,000 species). We hypothesized that functional traits could be used to determine 'who transfers genes with whom' within a network of interspecies horizontal gene exchange. We find significant numbers of high-probability edges within this network that do not have identified gene exchange. This suggests that there may be more gene flow between pathogens and gut microbiota (from humans and animals) than we observe. This work was published in Science Advances. We are gearing up for follow-up experiments using Salmonella enterica in chickens and mouse models to follow up on these findings. We have also pulled out transfers between probiotic species that are used in livestock settings to examine their potential to exchange genes with other species. Objective 2: Determine the "spillover potential" of genes to transfer between livestock and wild microbiomes through targeted experiments using a set of genetically engineered probiotics. (Years 1-4, 75%) We have developed several assays that enable monitoring the spread of mobile genetic elements. Specifically, we have developed reporter lines of bacteria that allow us to distinguish between donor (i.e. a probiotic strain) and recipient (i.e. unintended transcongugant) strains in the gut. This is accomplished by labeling the donor strains with RFP (red fluorescent protein) on their chromosomes; whereas the plasmid is labeled with GFP (green fluorescent protein). The donor strain also contains a repressor to reduce GFP expression in the donor strain, thereby making donor and recipient easy to distinguish. We have also developed a plasmid with a barcoding system so that we can determine the number of potential transfers and the specific bacterial species involved in the transfer. This will highlight promiscuous donors. Our preliminary tests show that we can confidently distinguish transfer events within a cultured bacterial sample. We have performed plasmid tracing experiments in a gut microbiome community to determine our confidence in tracing the route of plasmid-based transmission. Next, we will apply this method to the gut microbiomes of a library of samples that we have collected, in collaboration with the Wild Animal Health Center at Cornell University and local farms, of wild animals, mainly birds, and domesticated livestock species. Objective 3: Test whether compounds that inhibit gene flow in laboratory experiments, can effectively reduce gene flow between genetically engineered probiotics and members of the microbiome in animal models. (Years 2-4, 60%) Due to inabilities to obtain the reagents outlined in the proposal that were previously indicated as reducing HGT rates, we have instead examined whether the use of either a prebiotic or antibiotics can induce or reduce the spread of mobile genetic elements. We have a plasmid that is labeled with an antibiotic and also has the ability to degrade 6'SL, an oligosaccharide normally found in milk, that has been shown in humans to be beneficial for immune phenotypes. Antibiotics are purported to induce reactive oxygen species that may increase HGT. We have performed studies in mouse models to determine how administration of these substrates, in the context of exposure to an organism carrying this plasmid, contribute to the spread and/or maintenance of the donor strain and/or plasmid. The next step of these experiments is sequencing the gut microbiome samples using OIL-PCR and 16S rRNA taxonomic profiling.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Zhou H, Beltr�n JF, Brito IL. Functions predict horizontal gene transfer and the emergence of antibiotic resistance. (2021) Science Advances. doi: 10.1126/sciadv.abj5056.
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Diebold P, New F, Hovan M, Satlin M, Brito IL. Detecting mobilized antibiotic resistant genes in natural bacterial communities using one-step isolation and lysis PCR. (2021) eLife. doi: 10.7554/eLife.66834.
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Brito IL. Examining horizontal gene transfer in microbial communities. (2021) Nature Reviews Microbiology. doi: 10.1038/ s41579-021-00534-7.
- Type:
Journal Articles
Status:
Other
Year Published:
2022
Citation:
Diebold P, Zakharevich I, Brito IL. Tracking horizontal gene transfer using compact CRISPR-based barcoding. In preparation.
|
Progress 09/01/19 to 08/31/20
Outputs
Target Audience:The primary target audience for this project are scientists working in the area of horizontal gene flow and the development of probiotics. The work has importance for regulators of genetically modified organisms, livestock producers, environmental conservationists, and biotechnology companies. Regulatory agencies, including the USDA, FDA, EPA, APHIS and also the NIH will be interested in this work. The only outreach we have performed thus far is introducing our work via a poster at the annual USDA-BRAG meeting, where we met administrators and officials from these various institutions. Changes/Problems:We had difficulties in obtaining some of the compounds we wanted to test in Aim 3. Instead of focusing on compounds that could selectively reduce HGT in the livestock population,we focused on the effects of having a positive selective pressure (a metabolic gene or antibiotics) in the livestock population, that would not be present in a wild population. We have designed plasmids and donor probiotics to test these rates. Donor E. coli strains with an RP4 (broad-host range) plasmid have been engineered with broad-host range promoters to express either different resistance genes or nanH2, which allows for degradation and utilization of 6'SL which itself is considered prebiotic. What opportunities for training and professional development has the project provided? This year was both a damper and a boon for training and professional development. Once the labs were shut down, there were more (free) opportunities online for students to gain professional development advice and develop their skills (16S rRNA microbiome analysis workshops, metagenomic sequencing workshops). I involved several students in reviewing papers during this time. I met with students regularly, once per week, as usual. How have the results been disseminated to communities of interest?We are in the midst of preparing several papers on our methods. We have published one Review paper and one research paper, the latter on HiC used to identify bacterial carriage of mobile genes, this year. What do you plan to do during the next reporting period to accomplish the goals?Objective #1: In addition to continuing our modeling and surveying of the mobile gene pools in animal microbiomes, we will perform Hi-C on these samples to determine whether they are harbored by the same organisms within a population of animals and between farms or livestock species. We have targeted collection of wild animals through Cornell and will continue surveying farms in the area. Objective #2: We will also start on experiments in the lab where we introduce organisms into both mice and livestock species to determine how wide-spread the MGEs can transfer within the gut microbiomes of animals. We will administer our reporter bacterial lines to determine whether other species' microbiomes are able to uptake the organism or whether the MGE is dispersed within the microbiome. Objective #3: We will perform our experiments in animal models and in animal gut microbiomes (ex vivo) to determine the spread of plasmids in the context of selection with antibiotic or HMO. Objective #1: In addition to continuing our modeling and surveying of the mobile gene pools in animal microbiomes, we will perform Hi-C on these samples to determine whether they are harbored by the same organisms within a population of animals and between farms or livestock species. We have targeted collection of wild animals through Cornell and will continue surveying farms in the area. Objective #2: We will also start on experiments in the lab where we introduce organisms into both mice and livestock species to determine how wide-spread the MGEs can transfer within the gut microbiomes of animals. We will administer our reporter bacterial lines to determine whether other species' microbiomes are able to uptake the organism or whether the MGE is dispersed within the microbiome. Objective #3: We will perform our experiments in animal models and in animal gut microbiomes (ex vivo) to determine the spread of plasmids in the context of selection with antibiotic or HMO.
Impacts
What was accomplished under these goals?
Objective 1: Model the gene flow between livestock, wild animals, humans and the environment by analyzing overlap between their microbiome compositions and their mobile gene pools. (Years 1-2, 90% complete) In year 3, we have near-completed the modeling portion of our project. We assembled a large database of ~12,000 genomes, for which we have developed a model for gene transfer that achieves high predictability. After quality filtering, we had roughly 9,000 genomes, for which we had near-full length 16S sequences and which had low contamination and high completion, as determined by CheckM. We assigned each gene to a specific KEGG function. We determined co-occurrence networks using SPARCC to determine correlation coefficients, in consideration of relative abundances, and the Earth Microbiome Project which has thousands of samples from animal, plant, and human-associated organisms. We also examined horizontal gene transfer between organisms using a heuristic outlined in Smillie et al, 2009 (500bp or greater of 99% or greater identical DNA). Using this network, we applied several different machine learning approaches that utilize the genetic composition of functions within microbial genomes to predict horizontal gene transfer between organisms. Using a random forest-based method or a graphical convolutional neural network, we achieve extremely high accuracy in our predictions (AUROC=0.97). Our ability to predict gene flow in species associated with animals only is not quite as good as with humans. We have performed enrichment analyses to examine the genes involved in gene transfer and these tend to be associated with the process of horizontal gene transfer itself or environmental determinants (shared ecologies enhance gene transfer). We are currently examining the specific environments that promote sharing within agricultural systems. We have also finished our collection of livestock (10 cows, 10 pigs, 10 chickens and 10 sheep) and wildlife samples (157 wild bird and mammals). Objective 2: Determine the "spillover potential" of genes to transfer between livestock and wild microbiomes through targeted experiments using a set of genetically engineered probiotics. (Years 1-4, 70%) Our work was slightly delayed due to COVID-related shutdowns of our research labs. Nevertheless, we have been developing assays to apply to Aim 2. Specifically, we are developing reporter lines of bacteria that allow us to distinguish between donor (i.e. a probiotic strain) and recipient (i.e. unintended transcongugant) strains in the gut. We have created a barcoded plasmid library. The barcode that we can use with OIL-PCR (One-step Oil and Lysis PCR), a method we recently developed. This method essentially encapsulates organisms within individual oil-in-water emulsions. Within the droplet, single cells are lysed and the plasmid gene of interest is amplified and bound to a taxonomic marker (typically the 16S rRNA gene). The final amplicon provides information on which taxa harbor the plasmid of interest. We have benchmarked this method on chicken and human gut microbiome samples and the method is remarkably sensitive and accurate. We have also been designing and constructing plasmids that carry a barcode so that they can be traced using OIL-PCR. We have created libraries using two broad-host range plasmids: pKJK5 and RP4 that carry barcodes that can be amplified using universal primers and fused to the 16S rRNA marker. Objective 3: Test whether compounds that inhibit gene flow in laboratory experiments, can effectively reduce gene flow between genetically engineered probiotics and members of the microbiome in animal models. (Years 2-4, 40%) We have had difficulties in obtaining the compounds that we wanted to use for these experiments. We decided to shift gears instead of trying to halt HGT, to examine whether HGT was exacerbated or reduced in the context of a prebiotic, 6'SL or antibitoics. 6'SL is a human milk oligosaccharide that exists in other animals as well. The adult animal gut microbiome should not have organisms capable of degrading this compound. In our experiments, it shows that it is has higher growth in 6'SL than WT E. coli Nissle, the probiotic strain used in our experiments, and that it is able to outcompete WT E coli in the presence of HMO.
Publications
|
Progress 09/01/18 to 08/31/19
Outputs
Target Audience:The primary target audience for this project are scientists working in the area of horizontal gene flow and the development of probiotics. The work has importance for regulators of genetically modified organisms, livestock producers, environmental conservationists, and biotechnology companies. Regulatory agencies, including the USDA, FDA, EPA, APHIS and also the NIH will be interested in this work. In addition to presenting a poster at the BRAG meeting, in 2019, students presented at the Synthetic Biology: Engineering, Evolution & Design (SEED) and the Biomedical Engineering Society Meetings. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?The training opportunities regarding scientific communication have consisted of participating in lab journal clubs, presenting at work-in-progress meetings for each of the graduate fields in which students participate and at the Cornell Microbiome supergroup. Students are attending progress report meetings with committee members. Wetlab skills development has mainly been through peer training in lab, consultation with members of labs on campus, and meetings with committee members for students. Drylab skills development have included our 2-day microbial genomics hackathon and workshops led by the Cornell Bio-High Performing Computing Cluster director and assistants. How have the results been disseminated to communities of interest?
Nothing Reported
What do you plan to do during the next reporting period to accomplish the goals?Objective #1: We aim to move forward with our modeling procedures. Once we settle on a model, we will examine putative transfers between members of the gut microbiomes of livestock and animal species. We will continue to collect samples and perform additional sequencing on the wild animal microbiomes. Objective #2: We will continue to validate OIL-PCR and perform HGT experiments using animal stool. We will continue to design and test our donor strains and develop statistics to determine spread of mobile genetic elements within microbial communities. Objective #3: We will obtain the remaining compounds outlined to halt HGT in order to perform the outlined experiments.
Impacts
What was accomplished under these goals?
Objective 1: Model the gene flow between livestock, wild animals, humans and the environment by analyzing overlap between their microbiome compositions and their mobile gene pools. (Years 1-2, 50% complete) In year 2, we have made significant progress on modeling of gene flow related to our project. We have settled on a modeling approach that focuses on the functions of specific genes rather than phenotypes. The idea is that the gene functions will largely integrate phenotypic differences. Using machine learning models, we can identify the key components involved in the predictions of edges depicting horizontal gene transfer (HGT+) and HGT- edges. We have started to identify genome sets and filter out putative contamination arising from vector contamination, human contamination, and metagenomic assembled genomes. We have also collected gut microbiome samples from a set of cows, pigs, goats and chickens. We have launched our collaboration with the Swanson Center for Wildlife Health and they have started collecting samples from wild birds and mammals that they see in their clinic. We have performed metagenomic sequencing on the pig, goat and cow samples and identified mobile genetic elements to determine whether we see the same elements in different animal samples. Objective 2: Determine the "spillover potential" of genes to transfer between livestock and wild microbiomes through targeted experiments using a set of genetically engineered probiotics. (Years 1-4, 30%) We have settled on construction of our plasmids used to detect transfer of plasmids and have been constructing them and performing preliminary tests. We have placed GFP in the plasmid, RFP in the donor strain, as well as a repressor for GFP in the donor, such that donor strains fluoresce red and recipients fluoresce green. We have also placed a barcode in the donor strain that can be detected in a single PCR read. We have created a control strain that is incapable of HGT. It has the TraG, a gene required for conjugation and the OriT (origin of transfer) deleted. We have ported all of this into an RP4 plasmid, which is a natural broad-range plasmid. We have sequence-verified these plasmids and strains. We have also been testing our fusion PCR based protocol to determine whether it can be used to detect transfer with high fidelity. We have revamped the EPIC-PCR protocol so that it is shorter (1 day versus 5 days), more robust (plasmids are contained within droplets avoiding false associations), and less toxic. We have validated this with a mock community and it is ready for use with animal samples. Objective 3: Test whether compounds that inhibit gene flow in laboratory experiments, can effectively reduce gene flow between genetically engineered probiotics and members of the microbiome in animal models. (Years 2-4, 0%) We are in the process of obtaining the chemicals listed in the protocol. We have performed some experiments with NAC (N-acetyl cysteine) an antioxidant, although we are concerned with the amount of NAC that is reduced in the starting material.
Publications
|
Progress 09/01/17 to 08/31/18
Outputs
Target Audience:The primary target audience for this project are scientists working in the area of horizontal gene flow and the development of probiotics. The work has importance for regulators of genetically modified organisms, livestock producers, environmental conservationists, and biotechnology companies. Regulatory agencies, including the USDA, FDA, EPA, APHIS and also the NIH will be interested in this work. The only outreach we have performed thus far is introducing our work via a poster at the annual USDA-BRAG meeting, where we met administrators and officials from these various institutions. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?One graduate student, one postdoc and one undergraduate have so far been associated with the project. A senior research scientist in the lab has overseen sample collection and metagenomic library preparation. I have weekly meetings with these students. How have the results been disseminated to communities of interest?As we are only at the start of the project, we have not participated in communication with outside parties except for the annual BRAG meeting. The undergraduate student is an undergraduate fellow of the Cornell Institute for Host-Microbe Interactions and Disease (CIHMID) and has presented a poster at two of their on-campus events. The audience for these events consists of Cornell professors and students. What do you plan to do during the next reporting period to accomplish the goals?In addition to continuing our modeling and surveying of the mobile gene pools in animal microbiomes, we will perform Hi-C on these samples to determine whether they are harbored by the same organisms within a population of animals and between farms or livestock species. We have targeted collection of wild animals through Cornell and will continue surveying farms in the area. We will also start on experiments in the lab where we introduce organisms into both mice and livestock species to determine how wide-spread the MGEs can transfer within the gut microbiomes of animals. We will administer our reporter bacterial lines to determine whether other species' microbiomes are able to uptake the organism or whether the MGE is dispersed within the microbiome.
Impacts
What was accomplished under these goals?
Objective 1: Model the gene flow between livestock, wild animals, humans and the environment by analyzing overlap between their microbiome compositions and their mobile gene pools. (Years 1-2, 33% complete) In year 1, we have focused primarily on Aim 1. We have obtained fecal samples from several farms in the region, including from dairy cows, pigs and goats (approximately 10 animals per farm). We have performed DNA metagenomic shotgun sequencing on those samples, assembled contigs and subjected those contigs to a series of pipelines to identify mobile genetic elements. We compared these mobile genetic elements across animals within a farm and across farms, to determine which mobile genetic elements were the most promiscuous. As for the modeling aspect of Aim 1, we have assembled a large database of genomes for which we have strain-level phenotypic data and microbiome studies (profiling the 16S rRNA gene) in which cells with a specific phenotype were tested. Phenotypic traits include: mucosal versus luminal localization, oxygen tolerance, association with inflammation, presence of the same mobile genetic elements specifically focusing on insertion sequences, motility, gram staining, association with plant/animal based diet, and whether the bacteria forms spores. These data have gone into a large model to understand gene flow within a given animal. Our model also takes into account phylogeny which is a known predictor of gene flow. We are also incorporating the data gleaned from the metagenomic assemblies (i.e. presence of overlapping mobile genetic elements) to test predictions about gene flow amongst a specific population. Objective 2: Determine the "spillover potential" of genes to transfer between livestock and wild microbiomes through targeted experiments using a set of genetically engineered probiotics. (Years 1-4, 10%) We have been developing experimental assays to apply to Aim 2. Specifically, we are developing reporter lines of bacteria that allow us to distinguish between donor (i.e. a probiotic strain) and recipient (i.e. unintended transcongugant) strains in the gut. We are using several methods for this--one is based on fluorescent and antibiotic resistance reporters, the other is purely genetic. For the latter, we have developed a method to be able to get a readout on which organisms have acquired the mobile genetic element of interest. Objective 3: Test whether compounds that inhibit gene flow in laboratory experiments, can effectively reduce gene flow between genetically engineered probiotics and members of the microbiome in animal models. (Years 2-4, 0%)We are in the planning stages of these experiments.
Publications
|
|