Targeting virus transmission in a vital crop for African food security

In East and Central Africa beans are a vital crop that naturally enriches the soil with nitrogen, providing natural fertilizer for other important crops such as maize and cassava. In the regional diet beans are important because they are rich in protein and critically important micronutrients. Since beans are mostly grown and traded by women, this crop provides important direct economic benefits to families & children. Unfortunately, a number of viruses attack bean plants, causing severe crop losses from disease.

Although some bean varieties have resistance to one of these viruses, bean common mosaic virus, a closely related virus that occurs widely in sub-Saharan Africa (bean common necrotic mosaic virus) causes plants of these 'resistant' lines to die. Thus, many farmers prefer to plant susceptible bean lines despite the risk of crop loss and it is important to develop new strategies to defend this vital crop.

The viruses are transmitted on the needle-like mouthparts of aphids. Importantly, the virus particles are lost by the aphid as soon as it feeds on a healthy plant; a weak link in the infection chain that we can exploit. Aphids are attracted to volatile chemicals released by plants. Virus infection can change these plants 'odours' so that aphids are encouraged to land on infected plants, acquire virus particles and then transport them to healthy plants. However, some bean varieties emit odours that are more attractive to aphids even when the plants are healthy. This means that if we plant mixtures of bean plants that include 'attractive' varieties we could decoy aphids away from the rest of the crop and slow disease transmission.

Even better, if we can decoy virus-bearing aphids to land on plants that are resistant to the virus, then as soon as the aphid feeds it will lose the virus particles. This would effectively 'sanitize' the aphids and stop them spreading the virus, as well as dumping the virus into plants in which it cannot grow. An additional benefit of using a mixture containing a limited number of resistant plants is that viruses will be under less pressure to evolve new strains that can overcome resistance and, in cases where virus infection kills resistant plants, only a small proportion of the crop will be lost. Since aphids and related insects transmit the majority of plant viruses the method could be applied to a wide range of crops.

Experiments under controlled conditions using varieties of a plant called Arabidopsis and an aphid transmitted virus called cucumber mosaic virus showed that aphid and virus spread can be altered using combinations of plants with different levels of attractiveness to aphids. A pilot field experiment in Uganda using mixtures of bean varieties with different degrees of attractiveness to aphids showed that aphids can be detained on more attractive varieties and inhibited from moving to other plants until close to harvest time, showing that our approach can work under farm conditions.

Our multinational team will:

A. Screen bean varieties that are popular in bean-growing regions in Kenya & Rwanda with respect to their attractiveness to aphids. We will identify the chemical odours emitted by plants that are most attractive to aphids. It is important to work with local varieties since farmers prefer certain bean types for hardiness under local conditions, cooking properties and local marketability. We will focus particularly on identifying aphid-attractive varieties carrying one or more virus resistance genes.

B. Mathematically model the consequences of different arrangements and mixtures of attractive/repellent/susceptible/resistant plant lines on aphid and virus spread in fields. This will not only help us design our field and lab experiments but also to analyze the data we obtain.

C. Test our models and lab experiments in Kenya & Rwanda to devise the most effective & practical field designs that minimize crop losses due to viral diseases.

We will devise methods to make better use of existing limited, genetically-determined virus resistance in common bean by creating scenarios in which aphids are lured to settle on plants with resistance to the bean-infecting viruses most prevalent in East & Central Africa i.e. BCMV and BCMNV. For non-persistently transmitted viruses (including these viruses) viral particles detach from aphid mouthparts as soon as aphids feed. 'Pulling' aphids towards resistant plants will limit virus spread since aphids subsequently migrating away from these decoys will no longer carry infection. We will devise field designs using mixtures consisting predominantly of farmer-preferred varieties (which typically are BCMV/BCMNV susceptible) protected by inclusion of aphid-attractive, resistant lines. We reason that this is a sustainable approach to protection against pathogens that have a propensity to evolve resistance-breakage and/or when deployment of 100% resistant plants is inadvisable due to problems with resistance (e.g. black-root disease).
The programme comprises three work packages:
A. Identify Kenyan and Rwandan farmer-preferred common bean varieties and virus-resistant CIAT lines that can be used in field designs. We will use a combination of aphid choice and feeding assays at Cambridge and analysis of headspace volatiles by GC-MS and electroantennagram analysis at Rothamsted.
B. Mathematically model the spread of non-persistently transmitted viruses in mixtures. This process will be iterative, using field data (from Kenya and Rwanda) and experiments with beans under controlled conditions to improve and refine models.
C. Validate models under controlled conditions and in field experiments in Kenya and Rwanda. This will be done with project partners with capabilities to monitor aphid distribution, virus infection and yield over four growing seasons in three locations in Kenya (BecA-ILRI campus and KALRO centres at Kiboko & Mugaga) & Rwanda at RAB, Kigali.

The project builds on work by the collaborators on the effects of virus infection on interactions between common bean & aphids (esp. Aphis fabae, Myzus persicae). Outputs of this work include findings that infection of common bean remodels host gene expression, inhibits prolonged aphid feeding, and alters emission of aphid-perceivable plant volatiles. Collectively, these effects enhance aphid-mediated transmission of non-persistently transmitted viruses (incl. the most important: bean common and bean common mosaic necrosis virus) to new hosts. Modelling indicates that decoying virus-bearing aphids to attractive plants will inhibit virus dissemination- born out by experiments under controlled conditions (using mixed Arabidopsis accessions, cucumber mosaic virus & M. persicae) & in the field by tracking aphid movement through bean mixtures. We hypothesize that including aphid-attractive, resistant plants in mixtures will decrease the proportion of aphids carrying viruses and thus decrease the basic reproduction number R0 to >1, inhibiting disease spread.

Translating this research & its outputs (incl. field designs, seed mix combinations, epidemiological models, semiochemicals) is vital for addressing problems posed by non-persistently transmitted viruses in many crops & in particular where food security is further threatened by spread of insecticide resistance within aphid populations or where access to chemical inputs is limited, where novel aphid-vectored diseases are emerging or where genetic resistance may be limited in its effectiveness. However, the approach's impact would be particularly important for common bean cultivation by resource-limited smallholder farmers in East and Central Africa, in which mixed cropping is the norm; indeed, bean is a vital intercrop required for high yields (via N fixation) and biotic and abiotic stress resistance in other crops (incl. maize, cassava, banana). Bean is an important source of protein & trace elements in the East and Central African diet and is of critical economic as well as nutritive value to women & children in the region.

Impact will be achieved through three interrelated activities.

1. Impact through Capacity Building. Dr. Appolinaire Djikeng our sub-contract partner & BecA-ILRI via their unique collaborator network incl. alumni, Governments, Growers, National Research Organizations & Industry in East and Central Africa & via the CG system including CIAT's linkages with the Pan Africa Bean Research Alliance provides a clear pipeline for technology transfer & knowledge sharing to National Agricultural Research services (NARS), growers, breeders, seed companies and extension personnel to promote income generation and improved nutrition in the region. Monitoring, evaluation, impact assessment and delivery activity through national research partners & extension services is managed Helen Altshul, BecA's development partnerships specialist. At Rothamsted Professors Toby Bruce & John Pickett have unparalleled experience in translation of basic research to improvement of smallholder agriculture in Africa, most notably the push-pull system, through their close collaborations with scientists in Africa.

2. Impact through Publication and Dissemination of Scientific Results including novel epidemiological models and elucidation of mechanisms underpinning virus-plant-aphid interactions via peer-reviewed papers & conference presentations.

3. UK Public engagement will be through existing outreach activities at Cambridge & Rothamsted. For outputs with UK commercial potential pathways exists via Cambridge's Innovation & Enterprise Project Officer, Dr Mariana Fazenda via the CambPlants Industry Club a hub hosted linking institutes, SMEs & UK industry & at Rothamsted via the Knowledge Exchange & Commercialisation Office.