Water availability is not the single biggest constraint on yield in African agriculture. Surprising, right? Nutrient limitation is the chief antagonist across large swathes of the continent (Mueller et al., 2012).

Biologically available nitrogen in the form of nitrate (NO3), in particular, is a major issue for the intensification of agriculture. Under natural conditions, nitrate is recycled through the decomposition of plants and animals, by microbial ammonification and nitrification. These are slow processes, counterbalanced by denitrification. In order to enhance crop yield, we have had to tip the balance of this cycle. The green revolution transformed agricultural practice and productivity across much of the world. High yielding cereals, mechanisation, irrigation, pesticides, herbicides and nitrogenous fertilisers combined to powerful effect, safeguarding millions from a long-predicted Malthusian catastrophe. In fact, data from Mueller et al. highlight the inequity of nitrogen consumption globally. The developed world consumes substantially more than the developing world.

Cost is the major issue. The Haber-Bosch process turns air into fertiliser, using brute force chemistry to fracture atmospheric dinitrogen into component parts. Liberated nitrogen combines with hydrogen to create ammonia (NH3), which is quickly converted into nitrate in the soil. This is an energetically expensive process, and represents a significant barrier to cheap nitrogen fertiliser across the developing world. Fertilisers are bulky products, and transportation to smallhold farmers living in rural areas of Africa with poor road infrastructure, makes distribution a challenge, adding additional cost. The same fertiliser costs three to four times as much in Africa as in the UK.  Most smallhold farmers can’t afford that, using small amounts of household manure instead – crop yields are correspondingly low.

Global reliance on nitrogen fertilisers does however have an impact on the wider environment. A lot of the fertiliser is lost from the soil as it leeches into water courses, lakes, and seas, where it triggers devastating algal blooms. These blooms deplete oxygen levels, spew toxins, and destroy natural ecosystems. An outstanding issue that we will come back to…

Nitrogen is found in every cell of every organism, and whilst the atmosphere we breathe is 78% nitrogen, we can’t actually use it; neither can plants. It is bound together in a very stable triple-bonded dinitrogen (N2) form that requires a large amount of energy to crack. Legume plants (clover, peas, beans and what have you) possess the truly remarkable ability to form extremely close and mutually beneficial interactions with soil bacteria that can grab biologically inert atmospheric nitrogen, and turn it into a usable form; a process known as nitrogen fixation. So-called nitrogen-fixing bacteria (or rhizobia) are biological freaks, armed with a specialised enzyme *nitrogenase* that lowers the energy barrier to splitting atmospheric dinitrogen.

Whereas small hold farmers typically apply in the region of 3 to 5 kg of nitrogen fertiliser per hectare, an average legume can fix the equivalent of over 100 kg of nitrogen per hectare!

The nitrogen-fixing symbiosis in plants involves a series of complicated interactions. Once both parties, plant and microbe, have acknowledged each other, root hairs curl around the rhizobium, like a warm embrace between old friends. The rhizobium then starts to set up camp inside the root hair itself, extending microscopic infection threads into the plant tissue. Plant cells divide in the area around this infection event, producing characteristic nodules that create a suitable environment for rhizobial replication, and ultimately, nitrogen fixation. Rhizobia are then released into the plant cells from this infection thread – a phenomenon truly unique to nitrogen fixing symbiosis.

The rhizobia become in planta tenants, paying a generous rate in nitrogenous rent.

Root nodule TEM
An electron microscope cross-section of a nitrogen-fixing nodule in soybean. The bacterial cells (darkest staining) are encapsulated within so-called “symbiosomes” – plant derived capsules that facilitate the symbiosis.

A ‘greener’ revolution for Africa?

When I first heard from a colleague about Giles Oldroyd’s plan to engineer nitrogen-fixation into cereal crops, my eyebrows quickly ran out of forehead. Engineering entire biochemical pathways into a new context, and getting them to work together in a coordinated fashion seemed a very big ask to me. Curiosity got the better of me, and several months later I had invited Giles to speak about his ENSA project (Engineering Nitrogen Symbiosis for Africa) at a departmental seminar.

As Giles’ himself explained during his presentation, nitrogen-fixing is incredibly complex. However, we already have a working prototype in the case of legume plants. He outlined explicitly how the process of engineering self-fertilising cereals could be broken down into four logical and definable steps:

1) Engineer rhizobial recognition into cereals – the crop must discriminate between harmful and beneficial microbes – opening the door exclusively to the latter, and responding appropriately.

2) Engineer nodule formation in cereals – once the rhizobia has gained access to the plant root, the plant must create an appropriate space for rhizobial colonisation – the nodules that you will see in the banner image.

3) Engineer the bacterial infection process – the plant must not resist attempts by the rhizobia to invade and replicate within plant cells.

4) Engineer the appropriate environment for Nitrogen fixation – plant cells must create microscopic vesicular pockets, known as symbiosomes, within which the rhizobia will fix nitrogen, and interact with the plant.

I was rocked. A system that seemed so intractably difficult was begining to make sense. Giles’ would be the first to admit that his program of research is aspirational, but by the end of the presentation, I could see a logical path ahead. If this worked, it would transform agriculture. Farmers in Africa could grow these crops without the need for additional nitrogen input, saving money and boosting yield. This would also avoid the nitrogenous pollution that comes with heavy fertiliser use. That is an exciting prospect – made possible by radical thinking, some serious financial beef (props to the Gates Foundation), and of course, the creative fractal of genetic engineering!


Several years later, I caught up with Giles’ at a leadership training course organised by the Gates Foundation. I asked him how the project was going. He took a long deliberate sip of his tea, before slowly breaking into a wry smile.  They had managed to address point one of the agenda above. It was now possible to rationally engineer a cereal crop to understand a new chemical language, and to interact with the rhizobia. This initial interaction plugged into another symbioitc pathway (that is present in cereals) for mycorrhizal fungi, triggering a characteristic spiking pattern in nuclear calcium. The first major hurdle, dealt with, just as he explained it could be done. He told me that they had managed this just before the grant was due for renewal. We laughed.

This past few days I’ve been at the Gates Foundation grand challenges meeting in Washington DC- a melting pot of big ideas, and correspondingly big budgets. Although Giles couldn’t be here (his lab is moving from the John Innes Centre to The Sainsbury Lab, and there was simply too much going on), there are many here working on his project. It has been fascinating to hear about their approach to the next big hurdle – engineering nodule formation. I’m a big fan of what Giles’ and the ENSA team are trying to achieve. It’s going to be fascinating to see how this pans out.

For more information on the project, check out Giles’ presentation below, along with the references and web links:


Bozoski et al. 2017. Receptor-mediated chitin perception in legume roots is functionally separable from Nod factor perception. PNAS 114(38): E8118-E8127. https://dx.doi.org/10.1073/pnas.1706795114  (this paper addresses the molecular basis of plant perception of pathogens relative to beneficial symbiotic bacteria)

Mueller et al. 2012. Closing yield gaps through nutrient and water management. Nature 490: 254-257. https://dx.doi.org/10.1038/nature11420 (this paper is pay-walled, but you can learn something of the content from Giles’ video presentation above)

Roy et al. 2017. MtLAX2, a functional homologue of the Arabidopsis auxin influx transporter AUX1, is required for nodule organogenesis. Plant Physiology 174(1): 326-338. https://dx.doi.org/10.1104/pp.16.01473  (this paper addresses the molecular basis of nodule formation, paving the way for translation into cereal crops)

ENSA project website

Alliance for Science – fixing the nitrogen problem