Convo with Emily Bailes: Breeding crop flowers to be more attractive to bees

emily bailes GRAPHIC

By MATT KELLY

Here’s a recent conversation I had with Emily Bailes, a postdoctoral research assistant in the School of Biological Sciences at Royal Holloway, University of London. She and her colleagues, Jonathan Pattrick and Beverly Glover, recently published a paper showing how the flowers of crops we eat might be bred to be more attractive to bees. The transcript below has been lightly edited for length and readability.

Emily Bailes: Hi Matt.

Matt Kelly: Hi Emily. Thanks for taking the time today.

Emily: Not at all.

Matt: Can you give me the short story of how you got into studying bees?

Emily: Good question. Well, the brief answer is that I became very interested in interactions between plants and other organisms during my undergraduate degree, but I also wanted to research something useful. Pollinators are the obvious choice because of how important they are for our food production. Plus, there’s something about bees — they seem to have their own personalities and are also cute and fluffy!

Matt: Except those little metallic green sweat bees…

Emily: Well quite! But I work mainly with bumblebees.

Matt: What particular interactions were you originally looking at?

Emily: I did a project on herbivores, looking if plants can use toxic metals such as cadmium to defend against caterpillars. I also have an interest in associations of plants with bacteria and fungi which can help them obtain extra nutrients. But the flowers and pollinators has definitely won out.

Matt: Can you give me a quick summary of your findings in this recent paper? The elevator pitch version.

Emily: The flowers of plants have generally not been thought about during breeding programmes. As they’re so important for improving the yield of many crops (including the faba bean) by attracting pollinators, we thought there could be a gap to fill there: breeding for flowers that would be more attractive to pollinators. Bees, in our case.

This study is looking at whether this would be possible in the faba bean, focusing on the pollen and nectar that bees visit the flowers for. We found there was a huge amount of variation in the different genetic lines of the beans we looked at, and so there is definitely potential to breed for flowers with more optimal levels of pollen and nectar to attract more bees and improve crop yield.

IMG_6267-2
Bee in V. faba flowers. / Photo: courtesy of Emily Bailes

Matt: And why start with beans? Vicia faba are self pollinating, right? Why not squash or brassicas, plants that need cross pollination?

Emily: Well, they can produce some seed without bees but previous work has shown that bee visitation can increase yield by 50 percent. Plus, bee visitation reduces yield variability, which means that the farmers can be more sure of what they’ll get.

The project really began from conversations between my PhD supervisor Beverley Glover and some bean growers and breeders who had noticed that yields had become more variable over the past few years, which could be linked to declines in the bee populations. In Cambridge we also have a large collection of different types of beans at the National Institute of Botany, so it was also a combination of timings and resources. Plus, brassicas are pollinated by a much wider range of pollinators, so it’s a bit simpler in beans. Particularly given that bees are what my lab was experienced working in.

Matt: In terms of breeding crop flowers to have more sugar in their nectar to make them more attractive to bees… Is this a “zero sum” game in plants? If you increase resources in one part of the plant, does that take away from development in other parts of the plant that might contribute to productivity or vigor?

Emily: Potentially there is a bit of a trade off, but I don’t think it’s that simple. In an agricultural environment, crops are usually provided with an excess of the nutrients they need. Carbon, which forms the basis of nectar production, is a pretty cheap resource for plants to use because they can get it from the air. I think the quantity we’re talking probably wouldn’t be detrimental to the plant yield.

Matt: Is this the sort of thing you might look at in field trials?

Emily: Yes. I’m currently working on a different project entirely, but Beverley has a new PhD and they are going to do some field trials to see if they improve pollinator visitation. The ultimate aim is to breed the traits into commercial varieties and see if they improve yield. At the moment, the different nectar productions we have are not necessarily in plants bred for high yields, so it’s hard to tell.

Matt: Also, in the paper you mentioned the possibility of increased visit time at a single flower and decreased overall visits to many flowers, which could impact necessary cross pollination. Big concern? Or just something to keep in mind? Would it vary by crop type?

Emily: Yes, that’s true for nectar volume. But I think it’s something to keep in mind for now. It depends on the trade off with greater numbers of bee visits versus each doing less work. If you’re trying to help in the long term by benefiting the local populations, it shouldn’t matter. And I can’t think of a reason why that logic would vary between crop types.

Matt: Your assessments of sugar-content preference and opening force only involved B. terrestris, correct?

Emily: Yes for sugar concentration. For force, it’s a bit more general.

We quantified how much effort the flower took to open, and a previous study has measured the strength of different species: bumblebees, honeybees and some solitary bees. The energetic calculation was for any bee. Bumblebees are all pretty strong it seems!

Matt: So, in terms of sugar concentration at least, would we need to breed crop flowers differently to be attractive to different species?

Emily: I don’t think it should be too different between bees. They all drink the nectar in the same way. It will be different for non-bees because of how efficiently they can take up the nectar. And there might be other reasons that bees need a different concentration that I can’t think of off the top of my head. For example, in really hot climates they may need the extra water content.

Matt: So, stepping back a bit from your findings in this particular paper… If we’re going to breed crops to be attractive to a variety of bee species and other pollinators, nectar might make a difference. But so might scent or coloration or petal features. There’s a whole chain of things we might need to consider?

Emily: Yes, exactly. There was a paper that came out recently that suggested short tube lengths in sunflowers improve bee visitation — although they didn’t look if that correlated with sugar reward, etc.

We see a similar thing when we use artificial flowers with different tube lengths (work I’m currently writing up). The shape of flowers will affect accessibility by scent, colour, etc. It will all help advertise the flowers and make them more visible or easy to find by pollinators.

Matt: So we think that preference for certain characteristics may vary by pollinator or bee species?

Emily: Yes, I think there are lot more differences in the colours, scents and shapes of flowers that different pollinators prefer. A coarse example is that bees cannot see red easily (they have a UV receptor instead), whereas birds can see red well like ourselves. So birds tend to prefer red flowers. But different pollinators will also possess different scent receptors, be different shapes and sizes, and so be able to manipulate flowers with different abilities.

IMG_6225-2
V. faba plants / Photo: courtesy of Emily Bailes

Matt: At the end of the paper, you talk briefly about making a comparison with landraces — presumably to see which features are most attractive to bees? Is there any sort of presumption that we could actually do better than landraces at attracting pollinators?

Emily: Hmmm… I think landraces have still been selected for to some extent. In our study, one landrace had really good nectar production but one of the others produced basically nothing. So, yes, I think in some cases we can do better than the landraces. But it really just depends on the line. I think that the environments the landraces were selected in will not necessary be the same as we see now in terms of number of pollinators.

Some of the commercial variety derivatives were actually not that bad for one trait or another.

Matt: One thing I found really interesting was the systemic view you and the findings suggest for growing crops. Example 1: Breeding is about more than selecting for productivity or fruit size; it can be about benefiting the pollinators in the larger farming system. Example 2: The benefit of breeding more attractive crop flowers isn’t something that exists in isolation; rather, it fits into a bigger systemic approach of planting wild flower strips, increasing future yields, etc.

Emily: I agree. Thinking not just from a short-term perspective and a single magic bullet but how we can use a combination of sustainable and long-term strategies. In the end, whilst humans need food, we should do that in a way that is as supportive to nature as possible. Especially given that the world wouldn’t work without pollinators.

Matt: Ideally, how do you see the findings from this research playing out in the real world? How would plant breeding and farming change? Or not?

Emily: I think that in the long term, and I imagine this will take many years, this approach would be taken in all of our pollinator-dependent and non-cereal crops to try and improve forage and crop production (if field trials show that it has a direct effect). I think it’s a sustainable way to support food production hand-in-hand with pollinator communities, and supports a move away from making agriculture more and more intensive. Even though that may have its place in some situations.

Matt: Last question… You mentioned before that you’re now working on a “different project entirely”. Related to bees?

Emily: Yes, still bees but from the other side of things: I’m looking at the transmission of diseases between bumblebees when they share flowers. An infected bee visits a flower, leaves a virus or parasites behind, and then the next bee that visits picks them up. We’re hoping to see if there are certain types of flowers that are more or less likely to do this. But the work is still in it’s early stages at the moment.

Matt: Wow. That’s interesting.

Thanks so much for your time today. Really enjoyed the chat.

Emily: No problem! Lovely to talk to you.