Study unveils new possibility for studying biology in 3D

For two decades, scientists at ASU’s Center for Biological Physics have explored how physics shapes biological processes.

Now, they’re closer to capturing high-speed biological and soft-matter events in unprecedented detail.

Douglas Shepherd, an associate professor in the Department of Physics, is the director of the  Quantitative Imaging and Inference Lab, where they’re currently studying the hydrodynamics of bacterial motilityThe movement of a living organism.. To do this, the group uses microscopy and computational tools such as a fluorescence microscopy, which shines a light onto a molecule or subject that absorbs it and emits a longer wavelength light.

The problem? Scientists can’t use this method at a high rate of speed to accurately study certain biological processes. In order to image faster using fluorescence microscopy, scientists would have to shine more light onto the sample, which in turn causes phototoxicity, or damage caused by the light.

Shepherd used bacteria swimming near the surface of a liquid as an example. For the bacteria, water is much more viscous than it is to humans — think corn syrup. When coming to the surface, only a thin layer of fluid separates them from it, and the laws of physics cause them to circle along the surface. Until now, researchers had never been able to capture the surrounding fluid’s behavior on a 3D scale.

“We have spent a lot of time trying to understand (such events) from the molecular scale, such as gene expression up to cellular activity, and how physics interacts with biology,” said Shepherd. “How did biology come to all the schemes it has for life to exist in the context of the laws of physics?”

In a study funded by the Research Corporation for Science Advancement, the Chan Zuckerberg Initiative and the National Science Foundation, Shepherd — alongside a team of scientists from ASU and Western Washington University — has introduced a quantitative phase imaging (QPI) approach that is capable of recording the 3D refractive index at kilohertz rates in a method they refer to as Fourier synthesis optical diffraction tomography (FS-ODT).

Published in Science Advances, the paper “Fourier synthesis optical diffraction tomography for kilohertz rate volumetric imaging” suggests the possibility of using lasers to reconstruct images as holograms at a high rate of speed. The study was led by former Center for Biological Physics postdoctoral scholar Peter Brown, with additional support from Assistant Professor Navish Wadhwa and WWU’s Domenico Galati, who provided expertise and samples on swimming microorganisms.

QPI is a label-free method that measures the density of the sample using minimal light, which doesn't hurt the sample as much as other methods. Normally, it only maps density and often requires hundreds to thousands of independent pictures to create a 3D volume, making it too slow to measure the physics surrounding microswimmers such as bacteria. Speeding up the QPI to 1,000 volumes per second is one of the major contributions of the research.  

Through physics-inspired machine learning and innovations to the microscope hardware, the team has opened a new avenue of investigation into high-speed biological processes that aren’t amenable to other types of microscopy.

“The big challenge we had was how do we build on the work that people have done before us and speed everything up to quantify the physics of swimming, not just swimming itself. We designed a way to direct many laser beams into the sample at once, increasing the information but also complexity of the measurement. In parallel we had to build a new computational model to decode the result,” said Shepherd. “Put together, these two advancements let us speed up enough to where we could go at 1,000 volumes a second.”

Their approach in the study takes inspiration from the way magnetic resonance imaging (MRI) and computed tomography (CT) produce 3D scans of the body by using lasers. The new process creates holographic imaging of the density of organisms such as bacteria without disturbing them with high laser power, unlike fluorescence microscopy techniques, at the rate of speed needed to visualize both the swimmer and the physics of the fluid surrounding the swimmer.

“Instead of the CT, where you just detect the X-ray intensity, here we make a digital hologram of the sample. Usually, 3D QPI requires independently acquiring hundreds of holograms, slightly tilting the laser beam in between each one. To speed up the measurement, we came up with a way to simultaneously acquire hundreds of holograms, allowing us to reconstruct the sample in 3D at high speeds,” said Shepherd.

This new method of phase imaging unlocks the potential to dig deeper into how physics has played a role in these biological organisms’ evolution and how they interact with the world. The team’s work will contribute to better understanding not only how humans came about but also how life on Earth works.

“For example, bacteria can form biofilms. These biofilms can make you get sick, but they also maintain the health of the ecosystem. We don't fully understand the mechanical interaction between the bacteria and their environment when they go from swimming to growing on a surface,” said Shepherd. “This method opens up a new way for us to study how these things mechanically and energetically interact with their environment, which we haven't been able to do before.”

While studies are ongoing, Shepherd looks forward to the potential this new method has in the field of physics and beyond.

“The thing we're most excited about is that this was all done using physics-driven algorithms and approaches,” said Shepherd. “I think we're really proud of the fact that this can be applied to any system that physicists could use it for, including experiments that don't have anything to do with biology. Physicists, chemists and biologists can use it for things they're interested in — because it can work on any system that has density.”