In a paper published in Oecologia in 2000, Richard Karban, Ian Baldwin, Kimberly Baxter, Grit Laue and Gary Felton experimentally demonstrated that plants that had not been attacked by herbivores could sense and respond, by upping their defenses, to cues released by neighboring plants that had been attacked. Seventeen years after the paper was published, I asked Richard Karban about his motivation to do this study, memories of experimental work, and what we have learnt since about induces resistance in plants.
Citation: Karban, R., Baldwin, I. T., Baxter, K. J., Laue, G., & Felton, G. W. (2000). Communication between plants: induced resistance in wild tobacco plants following clipping of neighboring sagebrush. Oecologia, 125(1), 66-71.
Date of interview: Questions sent by email on 9th December 2017; responses received by email on 30th April 2019.
Hari Sridhar: I would like to start by asking you about the origins of this paper? What brought this group of authors together and what did each bring to this study?
Richard Karban: I had been intrigued for a while by the idea that plants could communicate. Around 1981, David Rhoades told me that his work suggested this possibility. He observed that trees that were near neighbors that had been chewed by caterpillars became poor hosts for subsequent caterpillars. He was unable to repeat these results, and they were not properly replicated or controlled. Ian Baldwin and Jack Schultz conducted lab experiments and found a similar phenomenon. However, the whole line of inquiry was shut down when an influential paper by John Lawton convinced most ecologists that the notion that plants communicated was rubbish. Then, in 1990, a paper by Ted Farmer and Bud Ryan was sent to me to review. They presented rigorous evidence for communication between cut sagebrush and potted tomato plants, albeit in sealed jars in the lab. I found this paper quite convincing although it wasn’t clear to me that similar things occurred in nature, involving plants that co-occurred (unlike sagebrush and agricultural tomato).
I had been working in Arizona on induced resistance in wild cotton prior to 1991. After my daughter was born, fieldwork in Arizona seemed too difficult and dangerous with a toddler. I decided to look for a project at a field station that was on the east side of the Sierras where the managers, Dan and Leslie Dawson, had a daughter the same age as mine. They told me that I was likely to find many woolly bear caterpillars at the station and this possibility interested me. I never did find the caterpillars but wild tobacco was reasonably abundant at the field site and I had been impressed with the work that Ian Baldwin was doing on induced resistance in that species. I started several preliminary studies involving induced resistance with wild tobacco, none of which ever panned out. However, in attempting to access the tobacco, I kept stepping on sagebrush, causing it to emit its strong characteristic odor. This reminded me of the paper I had reviewed, since both tobacco and tomato are in the same plant family. Perhaps the sagebrush and wild tobacco was a natural analog of the system that Farmer and Ryan had described in the lab. I set up a preliminary experiment to test this idea during the summer of 1995. This first test involved experimentally clipping sagebrush plants and leaving others unclipped. I then observed levels of damage to neighboring tobacco throughout the season. Since I am a field ecologist with little aptitude for chemistry, the only measurements I made were on naturally occurring damage by insects to the neighboring tobacco plants. Tobacco with clipped sagebrush neighbors looked like they had less herbivore damage than tobacco near unclipped sagebrush, although the sample sizes were too small to draw any definitive conclusions. Towards the end of the summer, my two first year graduate students, Anurag Agrawal and Jennifer Thaler, came up to the field station for a visit. We talked about the results and about what experiments could be done if this phenomenon was real. There are many hot springs in the vicinity of the field station and they were interested in going for a soak in the evening. I am not a huge fan of hot springs so I passed on that opportunity – a fortuitous decision as this was the start of their lifelong partnership. During this time, Ian Baldwin drove his camper van over from Utah to visit for a day. The four of us discussed the advantages of collecting volatiles from clipped sagebrush foliage. The following season, Ian sent traps and a protocol for collecting samples of air from above the clipped sagebrush plants which I mailed to him in Germany for analysis.
Since this paper was rejected by many journals over several years, it gave us the unplanned and unwelcomed opportunity to repeat the experiment numerous times and to test alternative hypotheses for our results and possible mechanisms. Kim Baxter was an undergrad I hired to help with the experiments over two summers. Grit Laue was a technician at the Max Planck Institute who helped Ian analyze the volatiles collected from clipped and unclipped sagebrush. I had known Gary Felton when he was a graduate student with Sean Duffey at UC Davis. His thesis examined induced chemical responses of tomato foliage to herbivory by various insects. Gary was well equipped to analyze enzyme activity in the tobacco foliage which he suspected was a reliable marker of induced resistance. Unfortunately, these enzymes degrade quickly if they are not kept cold and Gary was an assistant professor at the University of Arkansas at the time. I was able to get dry ice from a fishing outfitter an hour from the field station and drive the samples 5 hours to Davis to mail to Gary overnight.
HS: Stepping back a bit, could you tell us how you got interested in the topic of herbivory and induced resistance in plants? I notice that your early work, including your PhD, was on the population ecology of cicadas. At what stage did your interest shift to plant-insect interactions?
RK: My PhD attempted to understand the factors that controlled populations of 17-year cicadas. I worked in the field with two other ecologists interested in cicadas, JoAnn White and Monte Lloyd. Monte was on my thesis committee and was incredibly helpful and inspiring to me as a naïve graduate student. (I didn’t realize at the time how unusual it was for Monte to fly from Chicago to Philadelphia to attend my committee meetings and exams). JoAnn and Monte noticed that trees responded to cicada eggs that were placed in twigs by producing gums and callous tissue capable of suffocating and killing the eggs. Since cicada nymphs spend their 17-year development feeding on root xylem fluid and move only short distances during their development underground, a tree that kills the eggs can reduce the number of cicada nymphs that will feed on it over the next 17 years. Although JoAnn and Monte did not call this induced resistance, the phenomenon that they described was shocking and exciting to me. It was the first time that I had considered that plants may not be the passive victims of their herbivores, but may actually fight back. When I got to Davis I was jazzed to investigate this idea.
Hari Sridhar: A part of this study was done in six sites in Utah and Arizona. How were these sites chosen?
RK: The parts of this study that were done in Utah and Arizona were the chemical analyses of air collected from clipped and unclipped sagebrush plants. These were the sites that Ian Baldwin was using for his studies. I don’t know how they were selected although they were sites that had recently burned and had large populations of wild tobacco.
HS: Do you continue to work at your field site in the Sierra Nevada Aquatic Research Laboratory today? When was the last time you visited this place? In what ways has this site changed since the time you worked there? Do you continue to work on wild tobacco and sagebrush?
RK: I have continued to work at the field site because I love spending time there. This past summer, I finished a study examining the effects of sticky surfaces on wild tobacco plants (in press in Ecology). Small insects get stuck on the plant and a community of predatory bugs is specialized for walking on sticky plants. These bugs are more numerous on plants that have been damaged and provide some protection against herbivores for the sticky plants. This sticky plant syndrome may be as common and important as extrafloral nectar. I don’t think the site has changed much since the 1990s.
I continue to work on this system even though it has some severe limitations. My focus has shifted from communication between sagebrush and tobacco to communication between sagebrush individuals and among branches within an individual. The chemical nature of induced responses is poorly known for sagebrush and it is not an easy plant to clone and manipulate. Nonetheless, it has been a fantastic system in many ways and exhibits considerable natural variation in cues emitted by damaged individuals.
HS: When you think back to the time of these field experiments what are your most striking memories?
RK: I dragged my kids (Claire and Jesse) to the field for most of the summers of their childhoods. We generally rented a condo in town that was large enough for their friends and their parents. This kept them entertained, gave them a big dose of the outdoors, and gave me a chance to really get to know them. During the rest of the year, our lives were much more separate – I went to work, they went to school, and we only saw each other briefly in the mornings and evenings. During the summers our lives overlapped to a much greater extent and I got to know who they were. This was one of the best parenting decisions I made.
HS: Would you remember how the figures for this paper were drawn and who drew them?
RK: Ian Baldwin prepared Figure 1. I prepared the other figures using SigmaPlot. This was one of my first times using this software and I complained about it a lot. Anurag Agrawal had to hold my hand repeatedly as I attempted to shrug off my Luddite tendencies.
HS: You acknowledge a number of people at the end of your paper. Could you tell us a little more about how you knew them and their contribution to this paper?
RK: This question is the one that has kept me from returning this questionnaire to you sooner. I do not remember what each of these people did but I know that this paper was a long time in the works and benefited greatly from the help that I got from many generous colleagues.
HS: How long did the writing of this paper take? When and where did you do most of the writing?
RK: I wrote this paper in 1997 and then continued rewriting it and repeating the basic experiment for five years as the paper was repeatedly rejected by at least five journals. I don’t remember how long this paper took to write initially but I generally write papers quickly. I revised the paper many times as I attempted unsuccessfully to publish it.
HS: Did this paper have a relatively smooth ride through peer-review? Was Oecologia the first place this was submitted to?
RK: This paper was rejected by five different journals; one of them rejected two versions of it separated by two years. It took three years to eventually find a home for it. The ecological community was not willing to believe the results that we found. At the same time, Teja Tscharntke and coworkers were attempting to publish a study reporting that alder trees were exhibiting a similar phenomenon. I was a reviewer for their manuscript and although I gave it a favorable review, the journal rejected it. It eventually was published in Oecologia as well.
HS: What kind of attention did this paper receive when it was published?
RK: The paper did not receive a lot of attention when it was finally published. It didn’t fit the ideas that many ecologists held at the time. I get a lot more correspondence about it now than I did when it was published from both the scientific community and the popular press. Following its publication, Marcel Dicke and Jan Bruin organized a special issue of Biochemical Systematics and Ecology dealing with communication between wounded and unwounded plants. That special issue helped bring attention and acceptance of the idea.
HS: What kind of impact did this paper have on your career and the future course of your research?
RK: This was the first paper that I published on volatile cues that plants recognize and respond to, adjusting their defenses to the risks they perceive in their environments. I have continued working on this question and this system during the 25 years since I inadvertently damaged sagebrush attempting to get to my wild tobacco plants. Although I am a dabbler by nature, induced resistance and the cues that trigger it have dominated my professional career.
HS: Today, 17 years after it was published, what is your take on your main conclusion: “undamaged plants respond to cues released by neighbors to induce higher levels of resistance against herbivores in nature”? Is this a conclusion that has found a lot of support subsequent to your study, in other systems as well?
RK: I think that this conclusion has stood the test of time in this system and in others. Andre Kessler and Ian Baldwin were somewhat skeptical of the result and repeated the basic experiment using a different sagebrush subspecies and wild tobacco in Utah. Their findings confirmed our results and added priming as a mechanism that we were not aware of. More recently, I conducted a meta-analysis with Louie Yang and Kyle Edwards of published and unpublished studies that examined volatile communication between plants of other species. This meta-analysis revealed that communication resulting in induced resistance is a widespread phenomenon, although it was not found in all systems. I believe that plants are under strong selection to evaluate their risk of being eaten. The most reliable cue of future herbivory is probably direct damage to the plant and this is the cue that has been described most commonly in studies of induced resistance. In addition to direct damage, many plants perceive volatile cues that are emitted by neighboring tissues of the same individual or neighboring individuals; to the extent that these cues provide reliable information about future risk, plants will be selected to respond to them. However, volatiles are not the only cues that can be perceived and offer reliable information.
HS: If you were to redo this study today, would you do anything differently?
RK: I would trust my data more and not second guess myself and my results as much. During my work with induced resistance and plant communication, I have run into many colleagues, reviewers, and others who have told me that the results cannot be true, are unimportant, uncommon, or poorly conducted and interpreted. Some of these critics have improved the science. Almost all of them have made me question whether I made some big mistake and whether I am about to embarrass myself. If I could redo my career, I would proceed more confidently. I would look harder at my data and let it lead me to the next questions. I would worry less about how my work was being perceived by the community.
HS: You say “We hypothesize that 3R, 7S MeJA is the wound signal, a hypothesis that needs further testing”. Was this confirmed by subsequent research?
RK: There is no question that MeJA can act as a signal to induce resistance in tobacco and other plants. This was shown by the seminal lab experiments conducted by Farmer and Ryan that motivated this study. Since that study, many workers have used MeJA to stimulate plants to elevate resistance under field conditions. What is not clear is whether MeJA acts as a volatile signal that plants perceive in nature. Cathy Preston and Ian Baldwin measured concentrations of MeJA that were emitted by damaged sagebrush and concluded that they would not be sufficient to induce resistance in tobacco plants under natural conditions. Therefore, the nature of the active signal in this and other systems remains unresolved.
HS: You say “It is not clear at this time that either of the plant species tested benefit from this communication”. Did subsequent research reveal whether the two plant species benefit from this communication?
RK: I conducted subsequent experiments with John Maron to examine the fitness consequences of communication between sagebrush and tobacco. Tobacco plants with clipped sagebrush neighbors produced more flowers and seed-bearing capsules than tobacco plants near unclipped sagebrush neighbors. This does not mean that growing near a sagebrush neighbor necessarily benefits a tobacco individual. Annual tobacco plants growing close to perennial sagebrush neighbors experienced less overall success than those without sagebrush neighbors, although they suffered less herbivory. Competition with a larger sagebrush neighbor is probably not beneficial for a tobacco plant although communication may mitigate some of the negative competitive effects by reducing herbivory. From the point of view of the sagebrush plant, there may be no benefit of alerting neighboring tobacco plants. However, the same volatiles that neighboring tobacco perceive may allow a sagebrush individual to coordinate its own defenses and may alert sagebrush kin to the relative risks of herbivory.
HS: You say “Other plant-derived volatile molecules have recently been identified as possible cues in other interactions”. Did your study also serve as an experimental template for further investigations of these molecules?
RK: Many other volatile molecules and other modes of communication have been identified for plants. As a field ecologist, I have not had much of a role in advancing our understanding of these mechanisms. However, I believe that our study may have convinced other biologists with more mechanistic approaches and tool kits that these phenomena were real and worth investigating.
HS: Have you ever read this paper after it was published? If yes, in what context?
RK: Since I have continued working in this system I have reread this paper many times.
HS: Would you count this paper as a favorite, among all the papers you have written?
RK: This is certainly one of my favorite papers. Because the notion of plant communication was widely dismissed it took more persistence than most other projects I have been involved with. The work itself was also thoroughly enjoyable. Fieldwork is the part of my job that I most enjoy. I conducted this study in a beautiful place doing something that was challenging and fun. My family helped out and I got to work with some great colleagues. Experiences like this one are why I get up in the morning.
HS: What would you say to a student who is about to read this paper today? What should he or she take away from this paper written 30 years ago? Would you add any caveats?
RK: I think there are several take home messages from this study that have more to do with how science is done than with the science itself.
First, where did this study come from? The impetus for this study came from two places – reading a paper by Farmer and Ryan that was somewhat outside of my immediate area of expertise and spending time in the field. It has become impossible to keep up with the ecological literature and many students don’t try. My advice is to read broadly and go to seminars that are not necessarily exactly what you are studying. You never know what will be relevant and help you make a connection. I find it much easier to come up with interesting ideas and projects working with the organisms than sitting at my desk in Davis being a ‘research administrator’. Messing around with your study organisms, particularly in the field, may help you come up with a better project.
Second, be opportunistic! I didn’t set out to study communication at my field site. When I had the idea that it might be occurring, I was willing to carve out enough time to conduct a preliminary experiment. Most of the ideas that I have don’t turn into anything. However, I’m willing to try a lot of projects that don’t work and I’m willing to drop what I’m doing when I see one that does seem promising.
Third, while I had some funding to work on related questions, I had no money to work specifically on this project. Indeed, it would have been impossible for me to convince a granting panel to give me money to do this study. Don’t let funding constrain the questions that you ask. Many ecological questions can be addressed with very little money. What you work on is too important and too personal to allow other scientists to impose their priorities on your decisions. I doubt you would let them direct your choice in relationships or other key choices.
Fourth, there are still many widespread and important ecological phenomena that we don’t fully appreciate. I thoroughly reject the idea that all the low hanging fruit has already been picked. Ecology is still a very immature discipline and ecologists tend to study the same phenomena that they believe to be important over and over again. If you are willing to get outside and keep your eyes and your mind open, there are many interactions that have not been well studied.