This document is The Aquaponics Association’s response to a recent publication on E. coli in Aquaponic and Hydroponic systems.

PDF Version: Food Safety and E. Coli in Aquaponic and Hydroponic Systems

April 27, 2020

By Tawnya Sawyer; Nick Savidov, Ph.D.; George Pate; and Marc Laberge 

Overview of the Study

On April 6, 2020, Purdue Agriculture News published a story about a study related to the contamination risk of Shiga toxin-producing E. coli (STEC) in Aquaponic and Hydroponic production. The full study was published in MDPI Journal Horticulturae in January 2020.

Researchers conducted the study from December 2017 through February 2018. The Study consisted of side-by-side aquaponic and hydroponic systems in a controlled environment lab growing lettuce, basil, and tomatoes with tilapia. The purpose of the study was to identify the food safety risks associated with soilless systems. The study indicates that both the aquaponic and hydroponic systems contained Shiga toxin-producing E. coli (STEC) at the time of sampling. It did not find the presence of Listeria spp., or Salmonella spp. 

The authors contend that the aquaponic system and specifically the fish feces were likely the sources of E. coli. However, we believe that there is no evidence to prove that this was the actual source of contamination since the authors admit traceback was not performed, and there were several other possible introductions.

The pathogen was present in the water and on the root system of the plants. The researchers did not detect it in the edible portion of the plants. However, if the water is positive for a contaminant, and it accidentally splashes onto the edible portion of the crop throughout its life, or during harvest, this could still result in a food safety concern.

History of E. coli in Soil-less growing systems 

Until now, researchers have only discovered environmental E. coli in soilless growing systems. It is essential to note that there are hundreds of types of non-fecal coliform bacteria in the air, water, soil, as well as the fecal coliform bacteria represented mostly by E.coli in the waste of all mammals, humans, and some birds. A vast majority of these coliforms are perfectly harmless.

The E. coli found in this Study — Shiga toxin-producing O157:H7 — historically has been associated with warm-blooded mammals, more specifically bovine fed corn in feedlots (Lim JY et al. 2007), as well as swine and turkeys. Further research must be performed to prove that cold-blooded, non-mammal aquatic species such as tilapia can harbor this strain of pathogenic E. coli. A wide group of studies, university professors, and industry professionals currently refute the possibility that tilapia can harbor this strain. The lack of evidence detailing the ability of aquatic animals to harbor E. coli makes the fish contaminated with this specific strain of bacteria very rare and suspect.

Many foodborne illnesses from fresh produce such as romaine lettuces, green onions, herbs, and sprouts, are traced back to the soil; the irrigation water used in these crops (Solomon et al. 2002); the seed stock; or poor sanitation in handling facilities.

There are a wide variety of community and commercial aquaponic and hydroponic growing facilities that routinely perform pathogen testing and have not identified this pathogen present. If it was present, traceback procedures would be followed to identify and remove the source, as well as any necessary food safety precautions and recalls performed.

Our Position

The Aquaponic Association and its members agree that food safety and proper handling practices are critical to commercializing our industry and the safety of our customers. One thing that the study points out is that a contaminant can occur in a soilless system, which creates a potential food safety concern. We agree on this; however, we have numerous concerns with the procedures and statements made in the publication.

We have reached out to the professional investigator on this study Hye-Ji Kim to get answers to essential questions that the study publication does not adequately address. There are significant gaps and questions with the study.

 Concerns About the Study Findings and Publication 

Lack of Traceability

The study group is unsure how the pathogen was introduced into the two systems. They admit that no traceback was performed to identify the source of contamination. They speculate both in the study and in their email response that this pathogen was:

1) Accidentally introduced

2) That it is from the fish feces in the aquaponics system that splashed into the hydroponic system through the open top of the fish tank during feeding,

3) that it was from contaminated fish stock (which were provided by the Purdue Animal Sciences Research and Education Center)

4) That it was human contamination from visitors or operator handling issues.

A traceback was not conducted as it was not within the scope of the study (Kim personal communications). We disagree; the discovery of O157:H7 strain in the university greenhouse with the suspicion of fish being contaminated should have resulted in immediate action in order to track down the source of contamination and prevent infection of the university students and staff. Outside of a University setting, traceback would have been mandatory in a commercial facility. It is questionable that the University did not perform these procedures because it was “out of the scope of the study”.

Questioning Fish Feces as the Source of Contamination

Blaming fish feces as the contaminating source seems incredibly misleading when so many other options exist, and no traceback proved that as the source. The contents of the fish intestines were tested for the presence of E. coli, and none was found (Kim personal communications). It seems that if the fish does not have STEC E. coli inside its gut, then it is more likely the fish feces being positive would be related to the contaminated water that the feces was floating in.

In wild fish species, levels of E. coli appear to follow trends similar to ambient water and sediment concentrations; as concentrations in their environments rise, so do concentrations within the fish (Guillen et al., 2010).

Furthermore, it seems very suspect that a two-month-old system in a controlled environment lab could have been so quickly contaminated. It is well-known that E.coli cannot survive in a biologically-active environment, such as an anaerobic digester or aquaponic system (T.Gao et al., 2011). E. coli are outcompeted by other microorganisms, which adapted to survive in the environment outside animal guts much better than E. coli. Thus, E. coli O157:H7, which is specially adapted to live in cattle guts, will inevitably be replaced by other microorganisms.

As for the hydroponic system showing positive results, this also seems suspect if the nutrients were synthetic, as there would be very little chance for the E. coli to survive without a biological host or continuous contamination source being present. An accidental exposure in the hydroponic system would have become diluted over time, or the pathogen died off to the point that they would have been undetectable. The fact is the organic matter in hydroponics is virtually absent and, therefore, provides a poor environment for E. coli growth and propagation (Dankwa, 2019). Therefore. one would need a continuous source, not an accidental one (like splashing), in order to maintain the E. coli population in hydroponics.

Since both systems were contaminated, we suggest that there is a more likely common pathogen source that the researchers did not correctly identify and remove. The source of contamination could be from source water, filtering system, repurposed equipment, airborne in the greenhouse or HVAC system, human vector, lab equipment, the seed stock, nutrients, or other inputs.

The Purdue Animal Research and Education Center, where the researchers sourced the fish, is an operation that also has swine, cattle, and poultry production. Research suggests that pathogenic E. coli can travel 180 m through airborne exposure (Berry et al., 2015). Airborne exposure poses a more significant risk to controlled environments as pathogens can persist in the HVAC system (Riggio et al., 2019). STEC has the potential to live in dust particles for up to 42 weeks, which can act as a possible vector of contamination if there is a continuous source. Therefore, even a slight possibility of the pathogenic Shiga-producing O157:H7 strain of E. coli transfer from the Animal Research and Education Center resulting in the uncontrolled cross-contamination of other research labs and facilities certified below Biosafety level 2 not designed to work with the pathogenic bacteria would raise a serious concern about the existing safety practices (Boston University).

Lack of 3rd Party or Peer University Test Verification

It has also been recognized that there is a high frequency of false-positive signals in a real-time PCR-based “Plus/Minus” assay (Nowrouzian FL, et al., 2009). Hence the possibility that the PCR verification method may have resulted in inaccurate results. The pathogen was not verified by a 3rd party lab to be actual STEC E.coli O157:H7. Only positive or negative results were obtained for this study.

We recommend several other universities and third-party labs to run samples and validate the results. However, no samples have been provided, which may be impossible to obtain based on the study being conducted in early 2018. Without this verification, there are questions about the possibility of false-positives due to the presence of environmental E.coli, fecal coliforms, or a wide variety of other bacteria commonly found in nutrient-rich environments (Konstantinidis et al., 2011).

Impact of Sterilization

The study conclusion suggests that sterilization efforts are critical. “Our results indicated that contamination with bacterial pathogens could likely be reduced in aquaponic and hydroponic systems if the entire systems were thoroughly sanitized before each use and pathogen-free fish were used for the operation.” This statement is inaccurate and could be detrimental to proper food safety practices. As the microflora of the system develops, it creates an environment that can suppress phytopathogens (Bartelme et al., 2018) and other zoonotic pathogens as a result of antibiotic compounds released by beneficial bacteria (Compant et al., 2005). In Recirculating Aquaculture Systems (RAS), some microbial communities take over 15 years to develop (Bartelme et al., 2017), resulting in greater stability over time.

Many papers support this hypothesis with regards to probiotics in wastewater treatment, aquaculture, and hydroponics. Microbial community analysis also depicts a greater microbial diversity in aquaponics over decoupled or aquaculture systems (Eck et al., 2019), indicating a more significant potential for suppression of pathogens in coupled aquaponic systems over RAS or decoupled aquaponic system. No pathogens were discovered in a mature coupled aquaponics system during 18 years of continuous research in Canada since 2002 (Savidov, personal communications).

These findings support the argument that more biologically mature systems are less likely to develop pathogens and that periodic sanitation should not be done outside of initial start-up unless a zoonotic pathogen (Henderson 2008), is detected. If a pathogen is found, producers should follow proper sanitation and recall procedures.

Conclusion

Overall, this and other research into food safety are ongoing, and new information becomes available continuously to help shape the best practices for proper greenhouse management. As the Aquaponic Association, we hope to provide the most accurate and reliable resources for this purpose. At the same time, we hope to reduce the possibility of studies like this creating unnecessary fear, or unsubstantiated claims that could harm the growth of the aquaponic (and hydroponic) industry. When a document like this is published, it will be quoted by the media, and referenced in other studies as if it is an absolute. Other research must be performed to validate or negate this study’s outcomes.

Our findings conclude that while there is a low chance of the persistence of a pathogen in properly designed aquaponic and hydroponic systems, there is still a potential concern. No agricultural system is immune to this. Compared to soil production, soil-less crops grown in a controlled environment are far less likely to become infected pathogens from mammals, birds, and other creatures which are difficult to prevent in field crop production. Human contamination or poor handling practices are of significant concern (Pattillo et al., 2015). The best way to avoid risk is to adhere to food safety guidelines set forth by the USDA, GlobalGAPs, the Aquaponic Association, and other accredited organizations.

contact: info@aquaponicsassociation.org

References

Bartelme, R.P., McLellan, S.L., Newton, R.J., 2017. Freshwater Recirculating Aquaculture System Operations Drive Biofilter Bacterial Community Shifts around a Stable Nitrifying Consortium of Ammonia-Oxidizing Archaea and Comammox Nitrospira. Front. Microbiol. 8. https://doi.org/10.3389/fmicb.2017.00101

Bartelme, R.P., Oyserman, B.O., Blom, J.E., Sepulveda-Villet, O.J., Newton, R.J., 2018. Stripping Away the Soil: Plant Growth Promoting Microbiology Opportunities in Aquaponics. Front. Microbiol. 9, 8. https://doi.org/10.3389/fmicb.2018.00008

Berry, E.D., Wells, J.E., Bono, J.L., Woodbury, B.L., Kalchayanand, N., Norman, K.N., Suslow, T.V., López-Velasco, G., Millner, P.D., 2015. Effect of Proximity to a Cattle Feedlot on Escherichia coli O157:H7 Contamination of Leafy Greens and Evaluation of the Potential for Airborne Transmission. Appl. Environ. Microbiol. 81, 1101–1110. https://doi.org/10.1128/AEM.02998-14

Compant, S., Duffy, B., Nowak, J., Clément, C., Barka, E.A., 2005. Use of Plant Growth-Promoting Bacteria for Biocontrol of Plant Diseases: Principles, Mechanisms of Action, and Future Prospects. Appl. Environ. Microbiol. 71, 4951–4959. https://doi.org/10.1128/AEM.71.9.4951-4959.2005

Dankwa, A.S., 2019. Safety  Assessment of Hydroponic Closed System 127. https://digitalcommons.library.umaine.edu/cgi/viewcontent.cgi?article=4052&context=etd

Eck, M., Sare, A., Massart, S., Schmautz, Z., Junge, R., Smits, T., Jijakli, M., 2019. Exploring Bacterial Communities in Aquaponic Systems. Water 11, 260. https://doi.org/10.3390/w11020260

Guillen, Wrast, Environmental Institute of Houston, 2010, Fishes as Sources of E. coli Bacteria in Warm Water Streams, https://www.uhcl.edu/environmental-institute/research/publications/documents/10-015guillenetalfishreport.pdf

Henderson, H., 2008. Direct and indirect zoonotic transmission of Shiga toxin-producing Escherichia coli. J. Am. Vet. Med. Assoc. 232, 848–859. https://doi.org/10.2460/javma.232.6.848

Konstantinidis, Chengwei Luo, 2011. Georgia Tech Institute, Environmental E. coli: New way to classify E. coli bacteria and test for fecal contamination, https://www.sciencedaily.com/releases/2011/04/110411152527.htm

Lim JY et al., Escherichia coli O157:H7 colonization at the rectoanal junction of long-duration culture-positive cattle. Appl Environ Microbiol. 2007;73:1380–1382 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1828644/

Boston University Agent Sheet E.coli EHEC or STEC) (https://www.bu.edu/researchsupport/safety/rohp/agent-information-sheets/e-coli-0157h7-agent-information-sheet/).

Nowrouzian FL1, Adlerberth I, Wold AE., 2009 High frequency of false-positive signals in a real-time PCR-based “Plus/Minus” assay. https://www.ncbi.nlm.nih.gov/pubmed/19161539

Riggio, G., Jones, S., Gibson, K., 2019. Risk of Human Pathogen Internalization in Leafy Vegetables During Lab-Scale Hydroponic Cultivation. Horticulturae 5, 25. https://doi.org/10.3390/horticulturae5010025

Solomon et al., Effect of Irrigation Method on Transmission to and Persistence

of Escherichia coli O157:H7 on Lettuce Journal of Food Protection, Vol. 65, No. 4, 2002, Pages 673–676 https://www.ncbi.nlm.nih.gov/pubmed/11952218

1. Gao*, T. Haine,  A. Chen,  Y. Tong, and X. Li, 2011, 7 logs of toxic strain of E. coli  were removed by mesophilic AD process while ~ 5 logs increase of the strain were seen in water control with the same condition for 7 days

Pattillo*, Shaw, Currey, Xie, Rosentrater, 2015, Aquaponics Food Safety and Human Health, https://southcenters.osu.edu/sites/southc/files/site-library/site-documents/abc/aquaponics_workshop/AquaponicsFoodSafetyandHumanHealthAllenPatillo.pdf