Until recently, fish that consume algae and detritus — grazers — were thought to keep reefs healthy, and fish that eat coral — corallivores — were thought to weaken reef structures. The researchers found high levels of coral pathogens in grazer feces and high levels of beneficial bacteria in corallivore feces, which they say could act like a “coral probiotic.”

“Corallivorous fish are generally regarded as harmful because they bite the corals,” said Carsten Grupstra, a former graduate student at Rice and lead author of the open-access study in Frontiers in Marine Science. “But it turns out that this doesn’t tell the whole story.”

Grupstra received his doctorate in 2022, and is now a postdoctoral researcher at Boston University and Woods Hole Oceanographic Institution. In 2021, he, his doctoral adviser, Adrienne Correa, and others from her Rice research group discovered that feces of coral predators are packed with living symbiotic algae that corals depend upon.

The new study builds on that with data and evidence compiled from two years of field research and laboratory experiments at Rice and theMoorea Coral Reef Long-term Ecological Research station on the South Pacific island of Moorea, French Polynesia. The research showed corallivore feces contained many bacteria that are found in healthy corals under normal conditions. Grazer feces were both found to contain pathogenic bacteria and shown to damage or kill living coral fragments in controlled laboratory experiments.

To understand why corals might benefit from their predators’ feces, Grupstra said it is important to consider that coral-eating fish do not devour their prey. Ever wary of being eaten themselves, they spend their days repeating a simple two-step process: grab a mouthful, swim to a new location. Because they relieve themselves as they go, they naturally disperse poop — and any beneficial organisms it contains — over a wide area.

Grupstra said the new study suggests corallivore feces could be an important source of beneficial microbes for corals.

“It’s analogous to fecal microbiota transplantation therapy in humans,” Grupstra said.

The study examined bacteria from both corallivore and grazer feces, and compared their effects on live corals. The researchers began by collecting fresh feces from corallivores and grazers during research dives. They both assayed bacteria in the samples and conducted controlled experiments to see how each type of feces affected coral.

They placed pieces of coral in jars with microbe-free seawater. To some jars, they added fresh feces from either corallivores or grazers. To determine whether the physical characteristics of the feces alone might harm the coral, they sterilized some fecal samples and added them to other jars. To the final group of jars, an experimental control, nothing was added. At the end of the experiment, coral fragments from all jars were examined and categorized as apparently healthy, containing lesions or dead.

The experiments showed that some feces could kill or smother corals. In most cases, the effect was localized, producing lesions on the coral fragment. In others, the entire fragment died. Feces from grazers caused lesions or death in all jars. Corallivore feces produced fewer and smaller lesions and rarely led to fragment death, and sterilized feces produced comparable damage.

“The bacterial assays from our field samples helped explain the results from the laboratory experiments,” said Correa, an assistant professor of biosciences. “We found coral pathogens were more abundant in grazer feces, and beneficial microbes were more abundant in corallivore feces.”

Grupstra said researchers need to further test how fish feces affect corals in the ocean. For example, damaging or beneficial effects could be limited if fecal pellets disintegrate or are eaten or removed by another organism. Better understanding the drivers of fecal effects could allow reef managers to implement treatments that promote beneficial effects or minimize negative impacts.

“Together, these findings result in a more nuanced understanding of the roles of fish on coral reefs and may help us better understand the interactions that are happening on reefs around the world,” he said. “Both corallivores and grazers have important ecological roles and understanding those roles can help us better manage and conserve these important ecosystems.”

The research was supported by the National Science Foundation (2145472 and 1635798), Rice University startup funds and fellowships awarded by the Wagoner Foreign Study Scholarship Program and the Kirk W. Dotson Endowed Graduate Fellowship in Ecology and Evolutionary Biology.

Human cells shield their genetic material inside the cell nucleus, protected by the nuclear membrane. As the control center of the cell, the nucleus must be able to exchange important messenger molecules, metabolites or proteins with the rest of the cell. About 2000 pores are therefore built into the nuclear membrane, each consisting of about 1000 proteins.

For decades, researchers have been fascinated by the three-dimensional structure and function of these nuclear pores, which act as guardians of the genome: substances that are required for controlling the cell are allowed to pass, while pathogens or other DNA-damaging substances are blocked from entry. The nuclear pores can therefore be thought of as molecular bouncers, each checking many thousands of visitors per second. Only those who have an entrance ticket are allowed to pass.

How do the nuclear pores manage this enormous task? About 300 proteins attached to the pore scaffold protrude deep into the central opening like tentacles. Until now, researchers did not know how these tentacles are arranged and how they repel intruders. This is because these proteins are intrinsically disordered and lack a defined three-dimensional structure. They are flexible and continuously moving — like spaghetti in boiling water.

Combination of microscopy and computer simulations

As these intrinsically disordered proteins (IDPs) are constantly changing their structure, it is difficult for scientists to decipher their three-dimensional architecture and their function. Most experimental techniques that researchers use to image proteins only work with a defined 3D structure. So far, the central region of the nuclear pore has been represented as a hole because it was not possible to determine the organization of the IDPs in the opening.

The team led by Gerhard Hummer, Director at the Max Planck Institute of Biophysics, and Edward Lemke, Professor for Synthetic Biophysics at Johannes Gutenberg University Mainz, and Adjunct Director at the Institute of Molecular Biology Mainz has now used a novel combination of synthetic biology, multidimensional fluorescence microscopy and computer-based simulations to study nuclear pore IDPs in living cells.

“We used modern precision tools to mark several points of the spaghetti-like proteins with fluorescent dyes that we excite by light and visualize in the microscope,” Lemke explains. “Based on the glow patterns and duration, we were able to deduce how the proteins must be arranged.” Hummer adds, “We then used molecular dynamics simulations to calculate how the IDPs are spatially organized in the pore, how they interact with each other and how they move. For the first time, we could visualize the gate to the control center of human cells.”

Dynamic protein network as transport barrier

The tentacles in the transport pore take on a completely different behavior compared to what we knew before, because they interact with each other and with the cargo. They move permanently like the aforementioned spaghetti in boiling water. So, in the center of the pore there is no hole, but a shield of wiggly, spaghetti-like molecules. Viruses or bacteria are too big to get through this sieve. However, other large cellular molecules needed in the nucleus can pass as they carry very specific signals. Such molecules have an entry ticket, whereas pathogens usually do not. “By disentangling the pore filling, we enter a new phase in nuclear transport research,” adds Martin Beck, collaborator and colleague at the Max Planck Institute of Biophysics.

“Understanding how the pores transport or block cargo will help us identify errors. After all, some viruses manage to enter the cell nucleus despite the barrier,” Hummer sums up. “With our combination of methods, we can now study IDPs in more detail to find why they are indispensable for certain cellular functions, despite being error-prone. In fact, IDPs are found in almost all species, although they carry the risk of forming aggregates during the aging process which can lead to neurodegenerative diseases such as Alzheimer’s,” Lemke says. By learning how IDPs function, researchers aim to develop new drugs or vaccines that prevent viral infections and help healthy aging.

‘Hexbug Nano v2’ microbots use vibrations to propel themselves forward. By connecting several of these toys with an elastic silicon rubber chain, the resulting structure is ‘elastoactive’. This means that it will return to its original shape after being deformed, while the self-propelling, active constituents that it is comprised of constantly try to push the structure in a certain direction.

Depending on the size of the chain links and whether the chains were fixed at one or both ends, the elastoactive chains showed a range of movement types, including self-oscillatory, self-synchronising and self-snapping.

“By experimenting with these elastoactive chains, we discovered that there is an interplay between activity and elasticity: when activity dominates, the chains self-oscillate and synchronize,” says Corentin Coulais, head of the Machine Materials Laboratory at the University of Amsterdam.

He continues: “Mechanical self-oscillation and synchronization are a key feature of biological machines, features that are useful for making new types of autonomous robots. These active chains really allow us to single out the nature of these nonlinear phenomena.”

Self-oscillation, self-synchronisation and self-snapping

When a structure self-oscillates, this means that it bends back and forth by itself. In the chains, the microbots might start by bending the chain to the left. However, because the chain is pinned at one end, the elastic links resist this movement, reorienting the bots in such a way that they start pushing and bending the chain to the right. This movement will again be resisted by the elastic chain, until the bots start moving to the left again.

Synchronisation happens when two elastoactive chains are connected at one end by a stiff enough rod. By wriggling around, the two connected chains automatically start oscillating with the same frequency, like sea grasses moved by the same waves.

Finally, taking a single elastoactive chain and pinning both of its ends, it shows ‘self-snapping’ behaviour. When you bend a playing card with your fingers, you can make it ‘snap’ to bending the other way by pushing hard enough from the side. The elastoactive chains do this by themselves, repeatedly snapping from bending to the left to bending to the right.

Instructive play

“We started this research by just playing around with the microbot toys. But more generally, the idea was to explore materials out of equilibrium. In soft matter, active fluids have been studied extensively in the last 25 years, but their solid counterparts were investigated much less,” says Coulais.

Next on the menu is exploring elastoactive behaviour at smaller scales, for instance in so-called colloidal systems, consisting of small particles suspended in a fluid. Even though these are still model systems, they are closer to the biological system due to similar length scales and the presence of the fluid. At any scale, it would also be interesting to use smart design to embed multiple self-oscillations within a single structure to obtain more complex movement patterns. With a better understanding of self-oscillations, the hope is that it becomes possible to create new types of autonomous robots.