Science1 publisher3 min readPublished
Leiden biologists narrow more than 8,000 marine microbe extracts to five compounds for scale-up
Leiden biologists ran more than 160,000 biological tests on over 8,000 marine microbe extracts and advanced five compounds to scale-up. Their account puts the bottleneck at sorting, and so far the evidence is suppression of disease-causing organisms in biological tests.
The Scientist · Science desk

What happened
- The extracts came from more than 1,200 microbial strains, which the team looked for in sponges, algae, fish and salt-tolerant plants.
- To pick out the promising molecules, the Leiden team built an analytical tool for identifying candidates across thousands of samples.
- Gilles van Wezel's group tested signaling molecules called elicitors to switch on biosynthetic gene clusters that stay inactive under lab conditions.
- Jos Raaijmakers and Victor Carrion searched salt-tolerant plants for microbes that could help crops withstand increasingly salty soils.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- constraint Until the tool's performance is published, other labs cannot judge whether adopting it would reduce the roughly 20 tests per extract that this screen needed.
- capability If elicitors reliably switch on silent clusters, the more than 1,200 strains already in hand become a larger chemical library without another round of sampling at sea.
- decision The five candidates now compete on fermentation yield as well as activity. A potent compound from a strain that grows poorly could lose its place to a weaker one that scales.
Divided out, the team's figures come to about 20 biological tests per extract [1] and six or seven extracts per strain [2]. Five compounds from more than 8,000 extracts means each candidate stands for more than 1,600 extracts [3]. An extract is a mixture that can hold many compounds. The ratio shows how far the pool was narrowed and cannot be read as a per-molecule hit rate.
Mariana Avalos Garcia, the project manager, compared the task to looking for a needle in a haystack [1][5]. Her reason for searching marine microbes at all is their chemistry. "Marine ecosystems are particularly promising," she said [3]. "The organisms there have developed unique ways of surviving intense competition, disease pressure and environmental stress. In doing so, they often produce new compounds to protect themselves, and these can be very different from those found on land." [4]
The thing this account doesn't tell you is how well the sorting worked. It does not describe how the tool ranks samples, report a hit rate, or name the target pathogens and effective doses. The team is still confirming the five compounds' chemical structures and activity [12], so it is not yet settled whether any of them is new to science. I think the team's numbers support putting the bottleneck at sorting, with one condition. Narrowing more than 160,000 tests to five candidates [9][11] only counts as a success if those five survive structure confirmation.
The next stage is about supply: whether the substances can be produced sustainably at larger scale [11]. In fermentation, microorganisms are grown under controlled conditions so that they make the desired compound [13]. The group is improving that process alongside the structure work [12]. Growing strains in the lab is meant to avoid repeatedly collecting sponges and other organisms from the sea [14]. "Biodiscovery only makes sense if it doesn't harm or undermine the biodiversity it depends on," Avalos Garcia said [15].
The elicitor work points at a limit of the screen itself. Some microbes carry gene clusters for interesting compounds that stay switched off in culture [16]. That means the 8,000-plus extracts [8] capture what the strains made in the lab, which is only part of what their genomes encode. "Microorganisms often contain a library full of books, but most of those books are locked," Avalos Garcia said. "Finding an elicitor is like finding the key that lets you open those unread books." [18]
The applications are conditional. According to the Leiden account, new medicines, crop protection products and treatments for fish diseases could have their origins on the seabed [22]. Within MARBLES, bioinformaticians led by Marnix Medema worked on how the microbes suppress disease-causing organisms [23][19]. The project also trained dozens of Ph.D. candidates and postdoctoral researchers [21].
What to watch
- Publication of the analytical tool's method and hit rates, so its sorting can be compared with other screening approaches.
- Confirmed chemical structures for the five compounds, showing whether any is a molecule not previously described.
- Fermentation yields at larger scale for the producing strains, the step that decides which of the five can be supplied.