To fight agricultural pests, scientists try spray-on plant vaccines

RNA interference can be used to protect food crops and improve plants' health, no genetic engineering required

Fabiola De Marchi


University of Padua

Every year, between 20 and 40 percent of the world's food crops are lost to plant pests and pathogens. This leaves 821 million people without enough food to eat and costs the global economy around $220 billion, according to the Food and Agriculture Organization of the United Nations (FAO).

The spread of pests and diseases is exacerbated by climate change, which has an enormous impact on food supplies, alters ecosystems, and creates new niches where those pests can thrive. Driven by an altered ecosystem, insects and pests are likely to expand their range. Because of these severe threats to food security, scientists are always searching for sustainable solutions to safeguard crops.

In August 2019, a research group headed by Sven-Erik Behrens, from Martin-Luther-Universität in Halle-Wittenberg, Germany, announced the development of a rapid and reliable approach to creating a "vaccine" for plants. It can be sprayed on like a pesticide, or even injected like an animal vaccine.

It works using a system called RNA interference, or RNAi. RNAi can be thought of as a kind of immune system: the cell recognizes double-stranded RNA (a cousin of DNA that also carries genetic information) that's not its own, like that from a virus trying to take over a cell, and chops that RNA into small fragments. Then, the cell uses those fragments to identify and stop further pathogen activity. Interestingly, eukaryotic cells – like animals, plants, and fungi – also use RNAi to regulate their own genes, recognizing and suppressing their own RNA. This can turn off or even fine-tune gene expression, like a dimmer switch on a lamp.

Behrens's group recreated RNAi machinery in their laboratory using cultured tobacco plant cells and double-stranded RNA from TBSV (tomato bushy stunt virus). Then, they searched for the viral RNA which produced the strongest RNAi reaction in tobacco. Once identified, they carried out “vaccination" experiments, injecting that RNA into uninfected tobacco plants. They found that plants inoculated with the most powerful vaccine were protected at a rate of 90%.

Entomologist Wayne Buhler sweeps a soybean field for western corn rootworm adults—another way to scout a field to determine if insecticidal bait is needed.

A scientist sweeps a soybean field for corn rootworms.

Ken Hammond/USDA

So far, RNAi vaccination attempts have been based on genetically engineering plant genomes to express specific RNAs normally only seen in pests, so they grow already vaccinated. In 2017 the U.S. Environmental Protection Agency (EPA) approved a genetically engineered maize specifically engineered against corn rootworm. However, even if GMOs are recognized as safe, in some countries there are concerns about their use. Countries that have banned GMOs seek non-transgenic, sustainable alternatives for pest management strategies that can reduce pesticides use. By combining the spray-on easy-application typical of chemical pesticides with the precision offered by genetic engineering, plant “vaccines” based on RNAi represent sustainable alternative in plant protection where GMOs use is forbidden by policy regulations.

One of the major advantages of "vaccinating" plants is that RNA molecules can be externally delivered to plants, through topical application like a spray, stem injection, root drenching, or seed treatment. This simple delivery method gives RNAi plant "vaccines" another benefit: flexibility. Viruses and pathogens mutate continuously to adapt to changing environments; designing a tailored RNA "vaccine" would be quicker and easier than time consuming and laborious procedures needed for gene editing. This would be an advantage for perennial crops, such as grapevine, citrus trees, and apple trees, which often require years of experiments and are expensive to genetically modify. RNAi “vaccines” are also highly specific: only target pathogens are affected by treatments, while undesired effects on other organisms are expected to be limited.

Cassava mosaic disease (CMD) symptoms in a field in Tanzania.

Cassava mosaic disease (CMD) symptoms in a field in Tanzania.  

 H.Holmes/RTB via Flickr

Plant vaccinations would make an enormous difference for many commercially important staple crops threatened by viruses like rice, wheat, maize, sweet potato, cassava, and banana. For example, at least ten different species of viruses have been identified in cassava plants affected by cassava mosaic disease (CMD), which causes substantial yield losses and famines in east and central Africa and more recently in southeast Asia, threatening the livelihood of 800 million people worldwide.

Since eukaryotic pests like insects use RNAi to fine-tune or shut down their own genes, that can be used against them. And although RNAi can not be directly used against bacteria that infect plant (because bacteria don't use RNAi), it could be employed to control their insect vectors, which carry the bacteria with them, transporting it from an infected to a healthy plant. If an insect nibbles on a plant carrying RNA that shuts down a gene the insect needs to live, RNAi will function like a targeted, biodegradable pesticide.

This would be very useful in the case of bacterium Xylella fastidiosa, which causes a range of diseases in different plant species, like Pierce’s disease of grapevines, citrus variegated chlorosis, and olive quick decline syndrome. X. fastidiosa has recently spread in the Apulia region, in the southern part of Italy, where it is destroying centennial olive trees and threatening the local economy. However, it is estimated that up to 563 plant species belonging to 82 botanical families can host the bacterium. It is transmitted especially by insects who feed on sap, mainly sharpshooter leafhoppers (pictured at the top of the page) in America and the meadow spittle bug in Europe. To date, apart from prevention and containment measures, there is no known cure for Xylella fastidiosa. But, RNAi treatments have been able to protect crops from insect vectors of X. fastidiosa, and so may be a tool to control bacterial infections as well.

Further studies are needed to understand how RNA molecules can be produced in large, industrial quantities. However, RNAi plant "vaccines" are proving to be a highly engineered sustainable method to protect food crops. Perhaps soon, we'll see crop dusters filled to the brim with RNA.

Comment Peer Commentary

We ask other scientists from our Consortium to respond to articles with commentary from their expert perspective.

The title certainly caught my attention.  Thank you Fabiola - nicely done.

As a scientist, I am excited to see that we are advancing our  knowledge of the role of RNAi and we are using it to combat some of the  pesky plant diseases. Scale up (and cost of production) will definitely be the next hurdle to tackle.  Also, how do we expect farmers to use them? Seed treatment and spraying would definitely be most straightforward. If we are spraying - do we know whether the RNAi will interact with other plants/weed species in the field? Inoculation might  make sense for fruit trees but it might still be a very intensive  endeavor.

I only have one problem with the article. This is not meant to be a harsh critique and I hope I am not offending anyone. I understand it is simply an analogy so that the lay audience can quickly understand what a RNAi does, but there are already far too many negative images of agricultural research on the internet. If we search for GMO stock images, many of them involve a plant or a piece of fruit/vegetable with a big needle. So I worry that “vaccinating plants” will only add to  people’s bias and mislead them to associate this technology with GMO.  In my experience, people thought any kind of spraying is negative and I have a difficult time to convince people that zero-input farming practices are actually NOT good practices (whether organic or conventional). I would be very hesitant to complicate my messaging by saying that scientists can now vaccinate plants… when in reality, we  really aren’t.

But overall, it was a pleasurable read.  I look forward to hearing more about the research progress.

Fabiola De Marchi responds:

Thank you for your questions. I’ll try to answer them.
Although RNA interference was discovered more than 20 years ago, experimenting this technology is still in progress, as you said. Farmers could use this technology in a feasible way by spraying RNAi based products as they would do with a synthetic pesticide. Trunk injection or inoculation could be more laborious, in fact, it depends on how many plants you have. Targeting unintended species is possible, but here is where the research is moving towards: finding the best double stranded RNA candidates to target only those specific pests and pathogens.

Yes, probably you are right, saying that scientists “vaccinate plants” could mislead people to associate RNAi with GMO; or even recall antivax movements (hopefully not). Should we better refer to them as “sprayable vaccines”? This could resonate as something really unusual since it recalls something different from genetic editing, but also from usual vaccines; therefore people could get interested, optimistically.  Probably, RNAi based products will not be presented and sold as  “vaccines” but as plant protectors or even biopesticides. There is always a lot of bias and misleading when talking about agriculture, people have the right to be properly informed for sure. Farmers, on their part, must not get used to chemical pesticides, but be aware that more sustainable alternatives, like RNAi, exist and will soon be  available. 

Crystal Chan responds:

Thank you! I do agree that farmers should continue to improve and not become overly reliant on chemical methods. I believe in Canada the RNAi technology will fall within the “biological control agent” definition (sounds a little dry - but very straightforward). I personally am not  sure if “sprayable vaccines” would sound much different from “vaccines”  to the general public. I very much prefer the term “plant protector”, but that’s just me.

I guess at the end of the day we just have to be mindful about the terminologies we use, because there’s a difference between perception from a scientist and perception from a lay person. It is just unfortunate that new research on genome editing and related technologies tend to get lumped into the prevailing “GMO” mindset. It’s a tough battle to fight, but that’s why we are all here trying to get the right message across.

Again, thank you for the piece, keep on the hard work!

Lauren Sara McKee

Microbiology, Biochemistry, and Biotechnology

KTH Royal Institute of Technology

Nice article, Fabiola! This is a tricky technical topic - I’m mostly familiar with RNAi being used to silence genes for research purposes.  Answering questions like “what happens to the structural integrity of a tree if we silence the genes that make lignin or cellulose in the cell  wall?” So it was very interesting to see this same tech being explored for use in the field. I also worry about on-farm application, and wonder if this type of spray might be better suited to a controlled greenhouse-type environment, where the risk of interfering with the wider communities is much lower.  We don’t want to start inhibiting the growth of helpful biodiverse insect communities! 

Fabiola De Marchi responds:

Thank you! Yes, it’s tricky because RNA is both the trigger and the target of the mechanism, so it can be confusing at the beginning. I think that controlled greenhouse should be quite similar to laboratory conditions, wherein there isn’t much exchange with the outer environment. RNA interference would be for sure very useful for plants growing indoors, underground or even in hydroponic systems, which could be the future ways of making agriculture. However, in these controlled conditions, plants wouldn’t be much in need of treatments against pests and insects, hopefully. The efficiency of RNA interference relies on its ability to target only harmful organisms and not the other ones; if the RNAi based product shows off-target effects, then more research has to be done to make that “vaccine” more precise. RNAi based “vaccines” represent a sustainable alternative to non specific chemical pesticides, that’s why it would be nice to see them on the field. 

Devang Mehta


University of Alberta

Really great article Fabiola! You’ve penned the best lay description of RNAi I’ve read yet!
I personally also agree with Crystal's objections to using “vaccination” as an analogy for the use of RNAi to limit disease in plants, but mainly because plants don’t have adaptive immunity and these RNA sprays would not cause plants to “remember” a pathogen attack like a vaccine would.
I’m also kind of sceptical of the technology’s ability to kill intra-cellular pathogens like viruses and some microbes. I actually tried it out in our lab to deal with the Cassava mosaic virus that’s pictured in your story and it failed miserably, especially compared to our genetically engineered plants. It could be because we weren’t too serious about the project (it was what my PI called, a “Saturday-afternoon project”) but a number of people in the field too have reservations about the ability of the RNA sprays to get into plant cells and amplify a systemic response to pathogens that divide rapidly, like viruses. It clearly works great with insects though, and having an insecticidal spray that only kills the target pest and not beneficial insects would be fantastic for agriculture! 

Fabiola De Marchi responds:

Thank you very much Devang! If it was up to me, I would have put a lot of jargon, so, just to give the proper credit, I can only be grateful to Massive Science editors.
That’s a good point, RNAi based “vaccines” are not remembered by plants, when non transgenic. RNA interference is quite unusual as mechanism, so it’s not easy to describe this molecular kitchen. In the paper I translated for the article, RNAs were inoculated in the plants. Topical application could be trickier, requiring for example nanoparticle carriers or other more technological solutions, but this is a possibility not to leave out, in my opinion.