Fighting pandemics in the digital age

Embriette Hyde

It’s almost impossible to remember what life was like just six short months ago before COVID-19 became the constant nightly news feature and hashtag. For far too many, that life will forever remain a memory: tens of thousands of lives and livelihoods have been lost to the virus and to the repercussions of global economic strain.

As most of us anxiously await a return of office work, going to school, concerts, sporting events, and other components of “normal life,” thousands of scientists in laboratories around the world battle tirelessly as front-line “soldiers” in the war on the disease by building better, faster tests and developing treatments and vaccines. Yet the battle is not just a scientific one, it’s also a supply chain problem. Countries struggle to get the supplies needed for testing. Just as toilet paper, yeast, and PPE were hard to come by in our homelives, now limited or unpredictable supplies of reagents and plastics hinder laboratories’ ability to run the millions of diagnostic tests needed.

Amidst the early chaos — when most of us were still in denial about the virus — two visionaries at Imperial College London foresaw the supply chain issues, and started developing a new kind of COVID-19 test. In late January, Dr. Paul Freemont, Professor of Synthetic Biology and Co-Director of the London Biofoundry, and Michael Crone, a PhD student in Dr. Freemont’s lab, began quietly working to develop a rapid, robust testing infrastructure that can accurately detect COVID-19 even under variable and uncertain supply chain conditions. Critically, his test relies on the already validated and trusted qPCR assay — but it’s fully automated, it’s scalable to thousands of tests per day, and it’s reagent agnostic, thus circumventing the supply shortages that have contributed to bottlenecks in testing capacity globally.

In just 9 weeks, Crone’s test workflow was developed from scratch, validated, and rolled out by the U.K. National Health Service to test 1,000 patient samples per day. This rapid roll out would have been impossible without the unique process, automation, and data driven approach of the London Biofoundry backed by the advanced capabilities of Riffyn Nexus, SAS JMP and other “lab of the future” systems. At the Biofoundry, Freemont and Crone develop diagnostics for other serious infectious diseases, such as human immunodeficiency virus (HIV). It is also the place where researchers and entrepreneurs engineer biology every day to tackle other serious problems such as food security, renewable biomaterials, and plastic pollution in the oceans. Now the Foundry can add COVID-19 to that list.

A storm brewing

In November 2019 Freemont was in Shanghai, China, a place he found himself often over the years to grow the global synthetic biology community. He already helped build the Imperial College Centre for Synthetic Biology and Innovation and the National UK Innovation and Knowledge Centre for Synthetic Biology (SynbiCITE) in London to help realize the power of engineering biology to solve some of the world’s biggest problems.

Together, these organizations have turned the U.K. into one of the global leaders in synthetic biology innovation. The addition of the London Biofoundry just a few years ago has enabled rapid commercialization of novel synthetic biology-based technologies and products.

As Freemont was nurturing relationships with colleagues in China — another global hotspot for synthetic biology innovation — disaster was brewing in the shadows. Soon Freemont would pivot the London Biofoundry to fight one of the biggest pandemics in modern history.

A pandemic begins

The New Year festivities had barely quieted down when the world began to realize the sobering reality of a rapidly spreading new flu-like disease originating in Wuhan, China. Freemont had been in close contacts with his colleagues on the ground in China since early January, and therefore was one of the first to understand just how quickly the disease was spreading and stressing testing resources, while the rest of the world was still happily naive to just how serious the problem was. Very early on it was apparent that the testing infrastructure there couldn’t keep up with demand, and Freemont wondered if there was something he could do. His graduate student, Michael Crone, was already developing automated, high-throughput disease diagnostics infrastructure; surely he could pivot to focus on the novel coronavirus.

Then everything on the Chinese end of the email chain went quiet as Freemont’s colleagues directed all of their focus toward fighting what would soon become known as COVID-19. It was the deafening quiet before the storm, because shortly after that, the freight train hit the rest of the world.

Freemont and Crone were no longer trying to help their Chinese friends. Now they were helping their own neighbors, friends, and colleagues in the same building.

Pivoting to fight the spread of COVID-19

Michael Crone — originally hailing from South Africa — has dedicated the last two years of his life to developing novel diagnostics for HIV and for urinary tract infections, which are a serious complication for individuals suffering from dementia. At his fingertips are the instruments, software, and high-throughput capabilities of the London Biofoundry — as well as access to patient samples through partnerships with hospitals and clinics.

IMG 9148 adj

IC PhD student Michael Crone. Image credit: Michael Crone

Last year, Crone had finished developing a critical piece of his diagnostic toolkit: virus-like particles (VLPs). The VLPs Crone was working with for his other projects are called “armored RNA” because they are comprised of short RNA sequences corresponding to a diagnostic target contained within a non-infectious MS2 bacteriophage capsid. Such armored RNA VLPs are especially useful in cases like the novel coronavirus: they enable testing and validation of diagnostics tests without putting anyone at risk to develop disease. It was these VLPs that were on Freemont’s mind when he decided he had to do something about the emerging pandemic.

It wasn’t long before the situation in the U.K. closely mirrored that in China: there was a serious shortage of reagents and other supplies needed to run the qPCR-based diagnostic test developed for SARS-CoV-2. Unlike typical diagnostics needs, the rapid spread of the emerging virus necessitated thousands of tests each day, and supplies were rapidly stressed. To make things worse, most testing centers used instrumentation that could only work with one type of kit.

Freemont remembers one of his colleagues in mid-February struggling to get reagents. “That’s when I realized that the problem wasn’t just testing capacity, but actually reagent supply chain issues. I realized it was a massively big problem. We had NHS testing capacity but all the tests were being done on platforms dominated by one or two suppliers and you could only use their reagents on their platforms.”

This was a problem, added Crone, because allocation of reagents by the government was severely limited. “Sometimes they would provide enough for only 34,000 tests per week for the entire country,” he said. “If you can use a different system that’s not dependent on that, it’s quite attractive,” explains Crone.

For Freemont and Crone, it was clear what was needed: a robust, validated assay agnostic to reagents and liquid handling instruments. This would allow test centers to mix and match instruments and reagents as available and relieve bottlenecks due to supply shortages.

A diagnostic framework built on automation and digital data management

Using different combinations of instruments, robots, reagents, and experiment design and analysis software, Crone and his colleagues at the Biofoundry quickly developed several open-access high throughput workflows for testing hundreds to thousands of samples in a 24 hour period. His VLPs — engineered to contain RNA sequences complementary to the SARS-CoV-2 N protein primer pair sets specified by the CDC 2019-Novel Coronavirus (2019-nCoV) Real-Time RT-PCR Diagnostic Panel — performed as expected. The biggest hurdles were differences in capacity on the different liquid handling robots used. To resolve this, Crone utilized Riffyn Nexus and JMP software to set up and track well randomizations and plate transfers to speed up the process of testing and validating each robot and reagent combination.

According to Crone, the combination of the design and data management software with the high-throughput capability afforded by automated liquid handling were critical for rolling out his assay. In a typical diagnostic scenario, he says, only tens or maybe a hundred daily tests are required, so clinical labs aren’t prepared for the stress on the system posed by SARS-CoV-2. But for Crone, who works with automation daily at the Biofoundry, it was an easy lift — and perfectly suited for the thousands of daily tests needed for the virus.

But validation, of course, requires two steps. It’s not enough to show proof of concept in the laboratory with VLPs. The next thing Crone needed to do was test his workflows on real patient samples. Typically, this is the step where things are slowed due to regulations around working with patient samples and ensuring test accuracy. But a collaboration with the U.K. National Health Service (NHS) enabled rapid validation for Crone’s workflow.

It helped that the workflow developed at the Biofoundry was based on “a qPCR workflow that people trust because it's ‘bomb proof’ for things like this,” says Crone. “If you can just implement something with qPCR, you’re already halfway, because then they don’t have to take the time to validate a whole new assay.” The NHS was attracted to the promise of a reagent agnostic platform, and, because it is already accredited as a clinical organization, any diagnostic assay they develop is already approved for use on patient samples.

“We took our infrastructure into one of the [NHS] labs and within three days we realized that we were going to be able to validate it,” says Freemont. “We did a pilot study on 100 patient samples and we were spot on … in fact, we were detecting some things that their diagnostic labs were not detecting. We realized we had a highly sensitive assay. So we got it revalidated in another diagnostic lab — the same platform for two validations.”

The NHS immediately added Crone’s workflow to their assay suite, rapidly scaling testing capacity one weekend when over 600 samples were run in just under 12 hours — by only two people. Now, the NHS is in the process of buying more instrumentation so they can increase capacity to 3000 samples per day on robots running 24/7 using Crone’s assay workflow. To further raise testing capacity, the Biofoundry team is designing new workflows using Riffyn Nexus’ workflow design and sample tracking. This will facilitate the implementation of pooling approaches that will have manifold impact on testing capacity.


Crone’s assay pipeline up and running. Image source: Imperial College London

In tandem, Crone also developed and validated CRISPR and LAMP-based diagnostic assays for SARS-CoV-2. These assays aren’t as quick or sensitive as the standard qPCR test and therefore it doesn’t make sense to use them for day-to-day testing — but they are useful in certain circumstances, such as in point-of-care diagnostics and for high-throughput community testing.

Impossible without a Biofoundry

Crone’s work was recently published in Nature Communications, and he is actively exploring ways to collaborate with colleagues in his home country to roll out a similar diagnostic infrastructure there. Unlike in the U.K., however, he is likely to face stricter regulatory hurdles, so it is likely to be some time before his assay reaches his fellow countrymen.

When he talks about what he did, and how quickly he did it, Crone is modest. He laughs and talks about the Biofoundry — how the resources and atmosphere there made it possible for him to do what he already knew how to do.

“What we did was really quite simple,” he says. “We already had the instrumentation, an approved and working qPCR assay, and the VLPs — we just had to validate everything together and make sure we could create a robust supply chain.”

According to Freemont, the Biofoundry itself was the critical piece for the project’s success, particularly given the rapid timeline.

“What I learned from this is that not only do Biofoundries have the liquid handling infrastructure, but they also have a mindset to develop robust workflows, reproducible measurements, reliable protocols, quality control,” he explains. “It’s not like a typical molecular biology lab because there you’re not thinking on the same scale — the idea of a workflow being critical, and how do you stress test workflows, how do you make sure that each step is working correctly as it should, and how do you automate and scale it? These are things that we think about all the time in a Biofoundry, because that's what we do.”

As for Crone, he plans to continue working on his original diagnostics projects, perhaps expanding his repertoire to include other diseases. Now that he knows how to support thousands of tests per day, typical diagnostics needs are no sweat.

And as we slowly return to the new normal, the Biofoundry’s researchers and entrepreneurs will one by one return to continue engineering biology to make our world a better place to live. And that’s good news for the rest of us. We no longer have to watch to see whether they can do it — they already have.