Grantee Research Project Results
Final Report: A comprehensive methodology to track genetically recoded organisms and assess their impacts on freshwater microbial communities
EPA Grant Number: R840203Title: A comprehensive methodology to track genetically recoded organisms and assess their impacts on freshwater microbial communities
Investigators: Konstantinidis, Konstantinos (Kostas) T. , Hatt, Janet K , Vereen, Ethell
Institution: Georgia Institute of Technology - Main Campus (School of Civil & Environmental Engineering) , Morehouse College
EPA Project Officer: Callan, Richard
Project Period: July 1, 2021 through June 30, 2024
Project Amount: $759,980
RFA: Assessment Tools for Biotechnology Products (2020) RFA Text | Recipients Lists
Research Category: Safer Chemicals , Chemical Safety for Sustainability
Objective:
The research objectives of the project, as outlined in the original proposal, are:
- Perform dialysis bag incubations with a synthetic E. coli expressing a recoded enzyme able to degrade benzalkonium chloride disinfectants (BAC) vs. E. coli expressing the wildtype proteins under BAC-supplemented conditions in order to assess the effects of the former on the natural microbial community relative to the latter.
- Develop biomarkers representing different sources of pollution and synbio organisms by examining large metagenomic/genomic libraries from around the world.
- Make the underlying gene and genome biomarkers searchable online as part of a 'MST webserver' to help users identify the source(s) of pollution in their samples and presence of synbio organisms by simply uploading their metagenomic dataset(s) to the webserver.
- Using the best performing biomarkers identified under #2 above, to quantify the relative contributions of different the sources of pollution of impaired creeks in the Atlanta metro area. The overarching hypothesis is that the modified E. coli will not pass BAC-degrading functional genes to indigenous microbes due to inability of these genes to function in non-modified cells.
Summary/Accomplishments (Outputs/Outcomes):
We have performed the proposed laboratory incubations, including an E. coli carrying recoded amino acids, and the corresponding manuscripts are already published or are being written (for the recoded E. coli incubation) at the time of this writing. The only difference compared to the original plan was that the recoded (SynBio) E. coli does not carry recoded BAC-degradation genes, but instead three other genes in the genome were recoded because it was challenging to produce the BAC-recoded strain and the recoded strain we used is equally appropriate for our purposes (a gift of Prof. Aditya Kunjapur of U. of Delaware). Our results show that the E. coli with the recoded amino acids dies out quickly, within about 3 days since the start of the incubation. We employed shotgun metagenomic sequencing to see if the presence of recoded E. coli has caused any significant effects on the indigenous microbial communities present in the freshwater inoculum of the mesocosms (from a nearby river) relative to the control (no recoded E. coli added), as well as if there have been any horizontal gene transfer events from members of the community to the SynBio organism to help it escape the dependence on the modified amino acid. The short answer to these questions is no; that is, we have not seen any significant effects on the indigenous microbes other than the (dead) E. coli cells serving as a carbon source, or any escape events. Thus, we can conclude that, for the organism and conditions tested here, the use of SynBio organism does not have a significant public health risk effect.
Further, developing biomarkers representing different sources of pollution and SynBio organisms, which can be used in microbial source tracking (MST) studies and pollution monitoring, are complicated by a lack of knowledge regarding the genetic content and distribution of fecally shed microbial populations from different sources. To address this gap, we performed a systematic literature review and curated a large collection of genomes (n=26,018) representing fecally shed prokaryotic species across broad and narrow source categories commonly implicated in FST studies of recreational waters (i.e., cats, dogs, cows, seagulls, chickens, pigs, birds, ruminants, human feces, and wastewater). We find that across these sources, the total number of prokaryotic genomes recovered from materials meeting our initial inclusion criteria varied substantially across fecal sources: from none in seagulls to 9,085 in pigs. We examined genome sequences recovered from these metagenomic and isolation-based studies extensively via comparative genomic approaches to characterize trends across source categories and produce a finalized genome database for each source category which is available online (n=12,730). On average 81% of these genomes, representing species-level populations, occur only within a single source. We expect this resource to be useful to MST-related objectives, One Health research, and sanitation efforts globally (Lindner et al., ES&T Letter 2024). Therefore, we have fully accomplished our Objective #2 above.
Further, using these previously developed genome libraries (described above), we developed a bioinformatic pipeline, called SourceApp, for inferring source attribution (which sources are present) and apportionment (what is the relative contribution of each source detected) by mapping the metagenomic data to these genome databases. The tool is freely available online and can be used by external users with their own metagenomic datasets. The users only need to upload their metagenomes to the website or download SourceApp and run it locally. To further validate the tool and benchmark or contextualized it for practice, we performed shotgun sequencing experiments (n=35) of mesocosms constructed from the water of a well-studied recreational and drinking water reservoir spiked with various fecal (n=6 animal sources, 3 wastewater sources, and 1 septage source) and mock microbiome spike-ins (n=1) introduced at predetermined concentrations based on cell counts to simulate fecal pollution events of known origin and contribution. Metagenome-based source attribution by SourceApp varied substantially based on parameter selection, and our parameter tuning led to sensitivity and specificity near 90% overall. The greatest number of false negative calls were for cow feces, largely due to under sampling of cow fecal microbiomes. False positive calls were most frequently associated with pig feces due to uncontrolled cross-reactivity with populations shared between sources. These results identify the areas that need improvement (e.g., more genomes) in future iterations of the tool. Source apportions of wastewater, septage, and cat sources had the least error using top parameters.
Conclusions:
The framework provided by SourceApp can assist researchers with analyzing and interpreting shotgun sequencing in efforts to develop standard operating procedures, in field work, and for interlaboratory collaborations on the frontiers of metagenomic MST. Notably, SourceApp represents a significant advancement over culture-dependent MST methods in cases such as mixed pollution sources (which are likely quite common) where culture-independent methods are of limited use with respect to source partitioning, and alleviates some of the computational barriers that may otherwise prohibit the broad adoption of the culture-independent MST methods. The aim of more granular MST approaches - such as the workflow and associated webserver described herein - is to better support decisions made by officials to protect public health and assist with developing remediation plans to restore water environments. Importantly, SynBio organisms are likely to vary in their genetic make-up and be industry-specific e.g., to represent genomes with recoded amino acids or simply, transgenic organisms with foreign genes cloned into their genome. By representing a general-purpose bioinformatic approach (and webserver), SourceApp allows the tracking of either whole genomes or genes of interest, providing an approach to track a great variety of SynBio genomes. Therefore, we have fully accomplished our Objective #3 and at least partially Objective #4 above. To fully accomplish Objective #4, we need to run SourceApp on field metagenomes, which is ongoing at the time of this writing with metagenomes originating from the Chattahoochee River (GA, USA) as well as (unfunded) collaborators at US EPA. We are working closely with our Morehouse College collaborators (Dr. Vereen and his team), and we are characterizing metagenomic samples from several impaired creeks in the Atlanta metro area that the Vereen team has been sampling. We will use these samples to fully test our Objective #4 above; that is, to use our MST webserver and underlying methodology to quantify the sources of pollution affecting these creeks. This work, as well as the work conducted by our EPA collaborators (Drs. Ben Davis and David Linz), is expected to be completed in 2025, but this represents a rather minor objective compared to what has been accomplished thus far. Therefore, we believe that, overall, we have accomplished the major objectives and deliverables of our project, and we have been highly successful and productive with it.
Journal Articles on this Report : 4 Displayed | Download in RIS Format
| Other project views: | All 5 publications | 5 publications in selected types | All 5 journal articles |
|---|
| Type | Citation | ||
|---|---|---|---|
|
|
Cha G, Zhu K, Fischer J, Flores C, Brown J, Pinto A, Hatt J, Konstaninidis K, Graham K. Metagenomic evaluation of the performance of passive Moore swabs for sewage monitoring relative to composite sampling over time resolved deployments. Water Research 2024;253(121269). |
R840203 (2023) R840203 (Final) |
Exit |
|
|
Lindner B, Gerhardt K, Feistel J, Rodriguez-R L, Hatt J, Konstantinidis K. A user's guide to the bioinformatic analysis of shotgun metagenomic sequence data for bacterial pathogen detection. International Journal of Food Microbiology 2024;410(110488). |
R840203 (2023) R840203 (Final) |
Exit |
|
|
Lindner B, Choudhry R, Pinamang P, Bingham L, D'Amico I, Hatt J, Konstantinidis K, Graham K. Advancing source tracking: systematic review and source-specific genome database curation of fecally shed prokaryotes. Environmental Science & Technology Letters 2024;11(9):931-939. |
R840203 (2023) R840203 (Final) |
Exit |
|
|
Lindner BG, Suttner B, Zhu KJ, Conrad RE, Rodriguez-R LM, Hatt JK, Brown J, Konstantinidis KT. Toward shotgun metagenomic approaches for microbial source tracking sewage spills based on laboratory mesocosms. Water Research 2022;210:117993. |
R840203 (2022) R840203 (Final) |
Exit |
Supplemental Keywords:
Microbial source tracking, FST, MST, limit of detection, SourceAppRelevant Websites:
Environmental Microbial Genomics Laboratory - GA Tech Exit
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.