The Beetle and the Virus: Scientific Discovery During a Trip to the Loo

Graph

Ceratoma trifurcata (photo by Marlin E. Rice) on a soybean leaf: the primary vector of Bean pod mottle virus throughout the midwestern U.S.

Graph

Leaves of Desmodium illinoense with (top) and without (bottom) Ceratoma trifurcata injury.

My deep appreciation for observations of nature was inspired by Marlin Perkins' Wild Kingdom television documentaries of the natural world. The thrill of studied preparation and careful observation were characteristics of the natural sciences that strongly appealed to me in my introverted youth. I would later understand scientific observation as a fundamental first step in the scientific method that can offer a path to discovery. The endorphin rush of discovery—to be the first person to shed light on a new attribute of nature—is a scientific addiction. With that in mind, I certainly did not anticipate such serendipity while on a search for the loo at a rest area on Interstate 35 in Iowa.

In June of 2006, I was one year from graduating with a dual doctoral degree in entomology and plant pathology from Iowa State University. Due to a heavier co-major course load, virtual reams of spreadsheets, and a crashed computer hard drive, my degree program was taking longer than planned and inspiration had become alien. I had been in this headspace before, and I often found comfort and motivation escaping to the natural world. However, on this day, personal physiology compelled me to stop at a rest area, and a small designated prairie area nearby grabbed my attention. There, I found a short trail through a mixed stand of old hardwood trees and brush that guided me toward a soybean field. On this short hike, I happened to look down, and what I saw led me down a most fruitful rabbit hole of discovery.

My doctoral research was focused on a major challenge to midwestern U.S. soybean production caused by an endemic pest—the bean leaf beetle, Ceratoma trifurcata (Forster) (Coleoptera: Chrysomelidae). A series of mild winters had led to the unprecedented overwinter survival success of this beetle and a concomitant outbreak of its population in soybean fields. This outbreak, in turn, led to a regional plant disease epidemic resulting from the beetle-transmitted virus Bean pod mottle virus (Secoviridae: Comovirinae), hereafter "the virus."

When it comes to plant viruses, arthropod vectors are the leading cause of pathogen dispersal throughout the landscape ([5]). During my research, questions remained about the contribution of alternate hosts to the initial inoculum of the virus into soybean fields ([6]). However, there was consensus that bean leaf beetles were the primary vector for the virus (at least in the Midwest); thus, part of my research was wrapped up in further defining the overlapping host range of the beetle and the virus. Given this beetle's overwintering habitat preference, and the knowledge gap regarding the annual source of the initial inoculum of the virus, I had developed a kind of ecological "search image" toward this problem.

A year prior to my 2006 rest stop, I had gone on a collection spree throughout Iowa, searching for possible leguminous host plants for the beetle and the virus. I had taken copious field notes and challenged adult bean leaf beetles to no-choice host tests in plastic arenas containing the leaves of several species of perennial Fabaceae from native, naturalized, and cultivated landscapes. Additionally, all plant samples were returned to the lab and tested for the virus. Rarely did I return with plant samples collected from nature showing any hallmarks of adult beetle feeding or evidence of beetle populations among established stands of a probable host plant (except for soybean; [2]). So here I was, a year after conducting extensive collections and bioassays and nearing submission of my findings, when I looked down to see a small clump of Illinois tick trefoil, Desmodium illinoense A. Gray (a native plant), covered in adult bean leaf beetles with its leaves pitted with the hallmarks of leaf beetle injury! I let out a loud "Whoa!" and immediately realized I might need to accommodate this new observation into my doctoral research.

Graph

Odontata horni (photo by Marlin E. Rice) on a soybean leaf. This leaf-mining beetle beetle may be an important secondary vector of Bean pod mottle virus throughout the midwestern U.S.

What caused my rest-stop exclamation were the numerous beetles on the plants and the extent of their feeding throughout the Lilliputian thicket of tick trefoil. Importantly, not only were bean leaf beetle adults and leaf injury conspicuous; so too were the feeding activities and adults of a soybean leafminer, Odontata horni J. Smith (Coleoptera: Chrysomelidae), another vector for the virus. Lastly, this species of tick trefoil is a perennial species that is not fire tolerant ([7]), which suggests that it inhabits protected habitats that might be more permanent and well suited for bean leaf beetle overwintering. This evidence suggested that should this plant test positive for the virus, there might be an interesting (and perhaps novel) ecological interaction at play between this wild legume, the beetles that were infesting it, and the seedling soybean field just a few feet further at the trail's end.

Prior to this observation, I had spent several years poring over the minutiae of plant–virus–insect interactions. My studies had impressed me with the unmatched adaptation rates of RNA viruses and their quasispecies arms race within and between landscapes ([4]). My mind was saturated with this information as I peered from the woodlot and into the soybean field beyond, and a hunch began to take form. In short, I wondered if a systemically infected, perennial tick trefoil plant might select for a virus population with mild symptomology as mutualistic populations of the virus may evolve within the plant over time—given enough springtime reinoculation events as the virus moves from dormant rootstock to fresh meristem. Because (so I thought) of the relative genetic similarity between tick trefoil and soybean, maybe a virus isolate from this new host might have mild symptomology in soybean too. As I broke through the clearing and glanced across the soybean seedbed before me, I wondered if this crop could be protected from severe virus symptoms—similarly to a IPM tactic in plant virology referred to as "cross protection" ([1])—because of that stand of tick trefoil and the conservation area at this rest stop.

When I returned to the lab, I immediately began sample preparations and confirmed by three quantitative methods that the tick trefoil sample (collected from a single plant) was indeed infected with the virus. As I had never had so much as a molecular biology class, the weeks that followed were deeply reliant on willing and patient expert instruction from lab mentors and a great deal of trial and error at the bench. This virus isolate turned out to be a little trickier than usual to sequence, which was partly due to the peculiarities of this virus' genome. Each of the independently encapsulated RNA strands that make up the Bean pod mottle virus genome (creatively referred to as RNA1 and RNA2) can be further subdivided into two subgroups (sgI and sgII) whose type members are designated based on host symptomology and genomic characterization. While I did eventually elucidate the full genome sequence (the first full sequence of this virus ever taken from a wild host), it turned out to be a unique strain that was diploid for RNA1 subgroups. That is, the virus contained full-length sequences of both sgI and sgII types for RNA1 within the single isolate! Further analysis revealed that there were also recombinant genotypes of the two RNA1 subgroups within the isolate, providing further evidence of a naturally occurring recombinant replicating in this wild host. Genes contained on RNA1 are responsible for symptom expression in its host, and previous RNA1 recombinants in soybean expressed severe symptoms. So, did this virus isolate (designated as IA-Di1) conform to my hunch, or was tick trefoil a harbinger of soybean doom?

Meanwhile, in the greenhouse, I had seeded and inoculated several replicate plants of virus-free species of Desmodium and soybean with my new isolate to observe host responses. To my excitement, many plants were asymptomatic when inoculated with IA-Di1, and if it were not for immunoassay detection, they would appear healthy. This observation contradicted previous observations of recombinant strains of this virus, which expressed severe symptoms in soybean. However, this was supportive evidence for my hunch that Illinois tick trefoil might be host to a mutualistic strain of this virus.

A responsible student who was this close to graduation would reasonably end this inquiry and be satisfied with submitting these findings for publication. However, my little hike from the rest area had given me a metaphorical face slap from which I had not yet recovered. So, later that growing season, I headed back out to the soybean field adjacent to the natural area to sample for the virus and compare those isolates for evidence of belonging to this new IA-Di1 strain.

Overall, 10 Bean pod mottle virus isolates were extracted and either fully or partially sequenced from 10 different soybean plants. I fully sequenced both RNA1 and RNA2 of the soybean isolate that was closest to the rest stop (designated as IA-P10) and compared it to IA-Di1. The comparison was interesting, but inconclusive. While the subgroup II types of IA-Di1 appeared to be highly related to the isolates from the soybean field, the full recombinant IA-Di1 was not recovered from soybean. Therefore, evidence of a direct linkage between the natural area, the soybean field, and the virus was not found ([3]). However, there was a very close relatedness between the RNA2 of IA-Di1 and IA-P10. Furthermore, because RNA2 is responsible for this virus' coat protein (which might be involved in the beetle's transmission of the virus), there was still some room for further inquiry. But, alas, at this point, I had run out of time and funding for additional inquiry.

While my sampling scheme was very limited, I also did not have clear evidence that this natural area was a reservoir for virus strains that might expose the adjacent soybean field to any clear risk. However, because of some of the unique properties of IA-Di1, I continued collaboration with my lab mentors to develop this isolate into a unique viral vector for gene expression and RNA silencing ([8]). The resulting research publication tops my "cited by" chart on Google Scholar with 206 citations to date, having been used by numerous scientists to improve our fundamental understanding of plant disease. I clearly got more out of my "bio break" to the loo than I anticipated, and the follow-up experience continues to inform my inquiries into the natural world when in the field and at the bench.

Graph

Desmodium illinoense (top row) and Glycine max var. Clark (bottom row) not inoculated (left column) or inoculated (right column) with Bean pod mottle virus isolate I-Di1. Plants were all grown in controlled greenhouse conditions.

Jeffrey D. Bradshaw is a professor of entomology in the Department of Entomology and Director of the Doctor of Plant Health Program at the University of Nebraska–Lincoln. His research program focuses on agroecology on the High Plains. His teaching and student mentorship is focused on growing plant health practitioners globally. He is President of the North Central Branch of ESA.

Acknowledgement

The author appreciates the careful review and feedback offered by Marlin Rice on the early versions of this manuscript.