Category: Blog

To mark the first anniversary of Moa’s in house synthetic chemistry lab, Moa chemists Benjamin Chappell, James Grosvenor and Charles Kingston reflect on how the lab is already making a difference to Moa’s vital work: getting better herbicides into farmers’ hands.

Early days

“Shuji (our CTO) told me on a long car ride two years ago that he was thinking about launching our own synthesis lab and asked if I’d be interested in building it,” says Ben, Head of Synthetic Chemistry at Moa. “At the time most companies were doing the opposite and attempting to outsource everything, so the idea was exciting.”

An external chemistry review and tough questions on project progress were part of the catalyst. “Some of the chemistry on a few early projects hadn’t progressed much beyond library material we’d bought in,” Ben explains. “We thought an internal synthesis lab could help us push chemistry forward on our projects.”

But setting up an in house lab wasn’t risk-free. “Indian and Chinese contract research organisations (CROs) employ very good scientists, and their services are competitively priced. Even if our in-house chemists were as or more productive, I worried about how would we make it work on a ‘molecules per pound’ basis.”

For Ben, the answer lay in creativity and alignment. “Chemistry is about creativity. Your own chemists can see all of the biological data and understand the project context. You often cannot share this information with synthesis CROs. This enables in-house chemists to contribute to projects in ways that synthesis CROs can’t. They have skin in the game, they are really motivated to find creative solutions that can make a meaningful impact to the speed and quality of what Moa is delivering. A synthesis CRO will adeptly make what you tell them to. An internal chemist will tell you what you should make and more importantly, what you shouldn’t.”

Making an impact

Moa chemist Charles says the lab has been a game-changer, in terms of getting new analogue compounds to trials in the field faster. “In one sense, we aren’t making anything that a CRO couldn’t produce for us. But some tasks take so long at a CRO that they were effectively off the table before. For example, here we’ve done stability studies on project compounds to understand degradation in the field, glasshouse or storage, and we’ve helped with formulation by making salts to address solubility. You could in theory contract that out, but the turnaround would be far too long. Here, we can often do it the same day or within the week, which can make all the difference to sending the best possible molecule out to test in the field.”

Having synthetic chemists embedded in projects also means a wider range of problems get solved. “CROs don’t get to see the full context of our projects for obvious IP reasons.” Ben says. “But our own synthetic chemists can sit in glasshouse tours, talk regularly with project teams and spot issues before they become an issue. A CRO several thousand miles away would never even know it was a problem.”

“The team has been able to suggest and create several new molecules themselves, after observing performance in glasshouses and reviewing test results which we would not be able to share with a CRO,” he adds.

Getting the right working environment

What’s it like working in the new lab? “It’s laid back in the sense that people feel comfortable to do their best work.” James says. “But we are also very passionate about what we do. There’s a lot of ‘come and look at this’ when something goes well, and that enthusiasm is contagious”.

Ben says the culture in the lab is intentional. “Some synthesis labs I’ve been in are very process metric driven – number of reactions, number of targets completed. Chemistry is partly about production, but I think it’s mainly about ideas. The atmosphere here gives space for those ideas.”

James adds “I came here from a large pharma focused CRO where I was one of 300 chemists. I learned a lot and had good colleagues, but the world you see is quite narrow: you make compounds, hand them over, and most people you interact with are chemists. At Moa, you can see how your work fits into the whole company – biology, glasshouse, field, operations. You meet people from different backgrounds and feel the impact of your work directly. It feels like you’re genuinely making a difference, not just filling a pipeline.”

Looking to the future

As Head of Synthetic Chemistry, Ben is clear about how to measure success. “It will be having a commercialised molecule in the field that came from this lab,” Ben says. “We’re not selling compounds; we’re trying to change things for farmers. Success will be when 10 years from now, we’re eating food grown with the help of a molecule that started here.”

Shorter term, there are clear signs they are on the right track. “We went from nothing to a fully functioning lab in about six months, with productivity exceeding our expectations and significant project impact,” Charles notes. “Given the age of the lab and resources we have, that’s pretty remarkable.”

James is keen to emphasise the role the wider team has played. “This isn’t just about the three of us. So many others across the company have been essential in getting the lab off the ground. We’re just the ones lucky enough to pick up chemicals from the cupboard and do the reactions.”

At Moa, our aim is to develop new herbicides for farmers that are safe, effective, affordable and – crucially- that function via a novel mode of action ( i.e. work within a plant in a completely different way to any other known herbicide).

But why are new herbicides needed – and why are novel modes of action so important?

Put simply, weeds have always been tough to control, because they continually develop new survival strategies to evade the control methods used by farmers.

Weeds have learned to dodge mechanical controls such as weed seed crushing mills attached to harvesters, by dropping their seeds early before harvest or by growing closer to the ground to pass under the mill.  They are also evolving to avoid cultural controls like shifting drilling dates or reverting to tillage – tactics which also have environmental and financial consequences.  Even the herbicides used by farmers for decades to control their toughest weeds are less effective than they used to be, with more weed species becoming resistant to their chemistries, and many regulators worried about the impact of increasing use-rates in attempts to overcome resistance.

Doing nothing, however, is not an option, if we want to ensure our future food security.  Weeds can reduce yields by as much as 70%, and Rothamsted Research has estimated that resistant blackgrass – one of the most problematic weeds in Northern Europe – could cost the UK economy alone £1 billion a year if it is not tackled.[1]

The good news is that a new generation of herbicides, with novel modes of action, has the potential to overcome existing resistance mechanisms and provide weed control at lower application rates to minimise environmental impacts. They will form a crucial part of integrated weed management strategies that combine chemical treatments with cultural, mechanical and biological approaches to achieve sustainable weed control in the future [2][3].

The problems that need to be solved

Weeds can develop resistance via two primary mechanisms: Target Site Resistance (TSR) and Non-Target Site Resistance (NTSR).

Herbicides function by binding to and inhibiting important proteins within plant cells. Target Site Resistance occurs when a naturally occurring DNA mutation in an individual plant leads to increased expression of one of these target proteins or or a change in the protein’s structure [4]. Structural changes can prevent herbicide molecules binding to their target as effectively, allowing the plant to survive a chemical treatment. If the plant is allowed to set seed, a new resistant population of weeds carrying the same mutation is born!

Non-Target Site Resistance is the result of genetic changes that reduce the amount of herbicide that reaches the target protein, rather than altering the target protein itself. Plants can achieve this in several ways such as enhanced metabolism (i.e. faster breakdown) of the herbicide molecule, reduced uptake and transport of the molecule or storage of the molecule away from its target protein (sequestration) [4].

The most difficult weeds to manage are those that have developed resistance via both these mechanisms. Blackgrass (Alopecurus myosuroides) populations in the UK are a good example of this. Mutations in the acetyl-CoA carboxylase (ACCase) and acetolactate synthase (ALS) proteins have resulted in TSR to the most common post-emergence graminicides (grass herbicides). Meanwhile, increased activity of metabolic enzymes like cytochrome P450s (CYPs) and glutathione S-transferases (GSTs) has provided NTSR to many herbicides, including microtubule assembly inhibitors like pendamethlin [5].  Other “superweeds” with significant levels of both TSR and NTSR include species in North America, such as Amaranthus palmeri (pigweed), and Lolium rigidum (annual ryegrass) in Australia.

Overcoming resistance in weeds

Resistance is the reason that finding novel modes of action is so important.  Weeds like annual ryegrass have, over the years, developed TSR and NTSR resistance to as many as 12 known modes of action, rendering many herbicides ineffective. However, because novel modes of action herbicides work in different ways to any existing products, current resistance mechanisms can be bypassed. This means farmers can eliminate out-of-control weed populations, reclaiming their fields for crop use and increasing their yields.

At Moa, we find mode of action herbicides in our GALAXY platform, but this is just the first step. The new herbicides then undergo a range of tests at our labs in Oxford and our glasshouses in Yorkshire. Here, and in collaboration with international research institutions, we test our molecules against some of the most problematic resistant weed species in the world.

Our resistance trials aim to find out the dose of Moa novel mode of action herbicide needed to control herbicide-sensitive and herbicide-resistant populations of the same species. We then compare this to the performance of existing commercial herbicides.  Excitingly, many of our novel compounds can overcome weed resistance mechanisms and achieve full control of both populations at an equivalent rate, such as the examples with waterhemp and blackgrass below.

We also use modern sequencing technologies to determine the specific mechanisms each resistant weed population uses. This helps us to confirm that our Moa novel herbicides can overcome a range of mechanisms and provide broad and robust resistance breaking.

The exciting results that we’ve seen against resistant weeds so far give us real hope that our compounds will soon be an important weapon in the fight against herbicide resistant weeds and a key part of sustainable weed-free food production.

Illustrations:

Above: Waterhemp (Amaranthus tuberculatus) with Target Site Resistance is strongly resistant to a commercial herbicide in comparison to a sensitive population (left), but a Moa compound is equally effective against both the sensitive and resistant populations.

Above: Blackgrass (Alopecurus myosuroides) collected from fields in the UK is strongly resistant to a commercial herbicide in comparison to a sensitive population from Germany (left), but a Moa compound is equally effective against both the sensitive and resistant populations.

1. https://www.rothamsted.ac.uk/news/herbicide-resistant-weed-could-cost-uk-ps1-billion-year

2. Riemens, M. et al. (2022). An Integrated Weed Management framework: A pan-European perspective. European Journal of Agronomy., 133, 126443.

3. Parven, A. et al. (2025). Herbicides in modern sustainable agriculture: environmental fate, ecological implications, and human health concerns. International Journal of Environmental Science and Technology., 22, 1181-1202

4. Gaines, T. et al. (2020). Mechanisms of evolved herbicide resistance. Journal of Biological Chemistry., 295(30):10307–10330

5. Cook, S. et al. (2023). Herbicide resistance in Alopecurus myosuroides: The value of routine testing of seed samples submitted by farmers since 1985. Weed Research., 63(6), 339–347

Weeds are a major challenge in agriculture: competing for critical resources and killing crops. While herbicides have been pivotal in weed control in the last fifty years, their overuse has led to herbicide resistance in 273^[1] species globally, with annual losses to herbicide resistant blackgrass in the UK estimated to be $500million^[2] , and annual weed losses of $3.3billion in Australia and $11billion in the USA^[3].

As the world’s population grows and demand for food increases, so too does the need for new herbicide modes of action that can overcome these resistant “superweeds” and help farmers protect their crops.

At Moa, new mode of action herbicide candidates are discovered using high-throughput screening of chemical libraries in GALAXY followed by extensive profiling in glasshouse. Profiling of chemistries in the glasshouse is not only labour-intensive but can also be subjective, relying on the expertise and the judgment of the person analysing the growth and damage to plants.

Through the Knowledge Transfer Partnership (KTP) program sponsored by UKRI and in collaboration with the Department of Computer Science in Loughborough University, we have been harnessing deep learning to automate chemical library phytotoxicity testing process. Specifically, I  have developed an advanced AI-driven solution to analyse glasshouse imagery, improving the accuracy, efficiency and scalability of herbicide activity assessment.

To achieve this, our collaboration started by collecting image data from our glasshouse testing, capturing images of control and treated plant samples across multiple weed species. Each image was then manually annotated to specify the plant species and identify visible leaf matter, creating an adequate dataset for deep learning model training.

We then used these annotations to build object localisation models, enabling automated identification of plant species and precise detection of plant matter in both control and treated samples. By analysing these paired images contrastively, our approach quantifies relative growth differences, allowing a statistical estimation of each compound’s efficacy with improved speed and stability. This process was iteratively refined to optimise model performance by expanding the dataset, including more species, and analysing relationships between model predictions and manual human evaluations.

This new AI tool scored an impressive 95% accuracy when evaluated against real glasshouse data, demonstrating robustness across different levels of herbicide activity and species.

Our AI tool is now being integrated into Moa’s glasshouse pipelines, to give scientists in a standardised, scalable, reproducible process, reducing manual labour and improving the overall quality of assessment.

What next?

AI cannot find and develop new, safe and effective herbicides all on its own, out of nowhere.  It needs data: the higher quality the better.  Moa has a huge bank of data, from the 750,000+ compounds our platforms have already screened. So  now we are starting to use this new tool to help clean up and interrogate our data: re-analysing archived experiments, identifying potential misclassifications and recovering previously overlooked compounds.  This will help us – and AI – to make better decisions.

This project was made possible by a Knowledge Transfer Partnership between Moa Technology and the Department of Computer Science at Loughborough University. The author would like to thank Professor Baihua Li, Dr Haibin Cai, and Professor Qingqing Meng at Loughborough and Dr Laura Davies and Dr Nasir Rajabi at Moa Technology for their collaboration on the project. 

The results of the collaboration will be presented at VISAPP 2025 in Porto, Portugal and have been submitted for publication in Machine Learning with Applications journal.

^[1] Heap, I. (2024). The International Herbicide-Resistant Weed Database. Retrieved from https://www.weedscience.org.

^[2] Varah, A. et al. (2020). The costs of human-induced evolution in agriculture. Nat. Sustain., 3(1), 63–71.

^[3] Chauhan, B. S. (2020). Grand challenges in weed management. Frontiers in Agronomy, 1, 3.

Synthetic chemistry and Lego: building cost effective herbicides from molecular building blocks

Lego is arguably the greatest toy ever invented.

With simple, small building blocks that reliably fit together, we can build very creative and complex structures.

Lego blocks build almost anything. There are only two constraints; imagination and the availability of blocks. For example, if we want to build a six foot dinosaur from Lego, we will need a lot of blocks.

Designing useful herbicides is the same. Agricultural herbicides are high volume products. Today, farmers around the world use almost two million tonnes of herbicides a year¹ to protect their harvests from crop-killing weeds. Identifying a new herbicidal molecule that is safe to use and effective at breaking weed resistance is not enough; the molecule must be both scalable and affordable to ensure food security globally. This means thinking about the chemical building blocks we will use to make our new mode of action herbicides. How much do they cost? Are they available on sufficient scale? These are questions we think about early on.

There are different types of Moa chemist designing these new herbicides. We have discovery chemists, who analyse data generated across the company and use this to design new molecules to drive our advanced novel herbicidal mode of action projects forward. We have computational chemists, who apply a range of computational tools to both design new molecules and predict how they may behave. Finally, we have synthetic chemists, who drag these ideas for molecules into reality, providing the first tangible, physical sample of the molecule in the world. Synthetic chemists spend a lot of time thinking about how chemical building blocks can be combined to make these new molecules, both in our own lab and how we will ultimately translate this to a much larger, commercial scale.

By understanding the price and availability of different chemical building blocks, applying some reasonable assumptions and our knowledge of how chemical building blocks can be combined to form new molecules, synthetic chemists can estimate the cost of these new herbicides in $/kg. Combining these final molecule price estimates ($/kg) with weed efficacy data (g/ha), we can understand whether our molecules would be economically attractive to farmers ($/ha) or not. At Moa, we use this information to refine the design of our herbicidal molecules.

As a recent example, we examined a lead molecule within one of our lead novel mode of action projects. We identified parts of the molecule that added three steps to our putative manufacturing route and significant anticipated cost, but we questioned whether it afforded us any efficacy benefit. By designing a simplified molecule, we were able to retain efficacy, but significantly reduce the projected cost of the herbicide to farmers.

In synthetic chemistry, making a molecule requires thinking several steps ahead; what sequence of reactions will result in the production of the target molecule in a useful quantity? Identifying a useful herbicide requires similar foresight. The herbicidal molecule must be safe for humans and the environment, effective at controlling weeds and importantly, produced economically at sufficient scale. These requirements can be considered at an early stage.

If this interests you, please get in touch: we’re adding to our synthetic chemistry team!

Sources

1 FAOSTAT   Global herbicide consumption 2022