Industrial Physics Forum 2013: Frontiers of biophysics

Five researchers who split their time between biology and physics conferences detailed the “Frontiers of Biophysics” at the final session of this year’s Industrial Physics Forum.

Single-molecule imaging

Stanford’s W. E. Moerner brought the audience up to date on recent progress in single-molecule imaging, a field in which he has worked for more than a quarter of a century. Super-resolution imaging, he said, is providing ever-clearer pictures of molecules in motion.

Traditional imaging can’t resolve objects smaller than the wavelength of light used for illumination. Therefore, nanometer-scale molecules in bacteria tend to blur together under the microscope.

But Moerner can spot individual molecules by pining fluorescent tags to them. Photons spat out by the blinking tags show up as a smear spread over several pixels of a microscope’s detector. The symmetric shape of that smear allows its center, and thus the precise position of the molecule, to be calculated. With this technique, aggregates of huntingtin protein inside a neuron can now be localized in two dimensions.

Tags that generate a double-helix-shaped smear allow a particle’s motion to be tracked in three dimensions . Slices of the helix look different depending on the particle’s vertical position, revealing the 3D position of a piece of RNA with a precision of tens of nanometers.

Other single-molecule imaging techniques can detect the orientation of a molecule, or its interactions with other atoms and molecules. In one study, a dye attached to an enzyme lit up when a copper atom transferred an electron to the enzyme.

“It’s pretty clear that there’s more information available now at the higher resolutions,” said Moerner.

Molecular simulation

Sara Nichols of the University of California, San Diego, described how biophysics can speed up the drug discovery process, which has grown lengthier and more expensive in recent years. Her computational simulations flag parts of proteins that could be appropriate targets for drugs.

“It takes on average a billion dollars to develop a drug,” said Nichols. “This presents a need for innovation.”

Molecular simulations have become remarkably complex since the first simulation of a protein 36 years ago. One dynamic model shown by Nichols included all of the RNA, proteins, water molecules, and ions that make up the satellite tobacco virus, about a million atoms in total.

Another revealed how a particular molecule latches onto a protein and snakes its way through the protein to its final binding site. The discovery through simulation of an unexpected binding site for HIV-1 integrase led to new drugs that inhibit the protein.

Nichols wants to better understand how proteins bend, fold, and wriggle. She has found that some, such as HIV co-receptor CXCR4, have hinges. Blocking the hinge with a drug can prevent conformational changes. The most potent HIV reverse transcripase inhibitor developed to date works in this way.

Bacteria versus cancer

Another scientist on the West Coast, Adam Arkin of the University of California, Berkeley, explores the promise of synthetic biology for combating disease. Instead of developing new drugs, he is engineering an army of bacteria specialized for attacking cancer.

“These are fantastic little microrobots that we can program to do a job,” said Arkin.

By inserting new genes into bacteria, Arkin can augment them with custom components. One gene, for instance, ensures that the bacteria can’t survive without a substance ingested by a patient—keeping the living robots from going haywire in the body and multiplying out of control. Others genes encode protein that sense the low-oxygen conditions present in a cancerous mass, or proteins that trigger the bacteria to attack only when they reach a certain density.

Biological circuits inside the bacteria govern all of these activities. DNA is transcribed to RNA, which produces proteins. The interaction of those proteins switches on and off different processes, much like a logic gate in a computer chip.

Figuring out how to build those circuits is easy on paper, says Arkin. But the genetic programs rarely function cleanly in actual cells. Genetic differences between bacteria can lead to a lot of variability.

Nevertheless, after detailed studies of how different stretches of DNA interact with the new genes he inserts, Arkin can now predict the amount of protein produced by a gene in a living cell to within a factor of two—giving him more confidence that his bacterial soldiers will actually do what they’re programmed to do when in the field.

“We can pretty much program any function with high reliability,” said Arkin.

Single-molecule mass

Michael Roukes at Caltech is building a new kind of instrument for measuring the mass of single molecules, made of tiny metal beams. It is potentially faster and more sensitive to a range of masses than today’s mass spectrometers.

Clamped at either end, each beam vibrates at a known frequency. When a free-floating molecule sticks to a beam, it decreases that frequency by a value that depends on the mass of the molecule and its position on the beam. By measuring two different modes of vibration, Roukes can calculate both parameters.

Proteins can be sprayed on a grid of thousands of beams, arrayed like tic-tac-toe boxes. Slowing beam vibrations show up as changes in electrical current.

Sensitivities of a single atomic mass unit should, in theory, be possible. But right now the instrument specializes in large masses out of the range of mass spectrometers, allowing it to, for instance, count the number of antibodies in conglomerations of those molecules. Those higher-mass ranges could be useful for identifying individual bacteria, such as those that causes sepsis in the bloodstream. Spraying bacterial proteins onto beams could provide an instant fingerprint and an alternative to today’s state-of-the-art tests for sepsis, which can take a day or more.

“We think it’s feasible to do a billion proteins in 100 seconds,” said Roukes.

Heavy metal spectroscopy

Diversifying California’s hold on the session, Scott Tanner of the University of Toronto said he also wants to fingerprint individual cells. To that end, he has put a new twist on the mass spectrometer, adapting it into something much like a flow cytometer.

Traditional flow cytometry detects components of cells by first tagging those components with fluorescent dyes, using different colors for different molecules. As cells flow by a detector, the strength of each color indicates the relative quantities of each component.

But with few colors available, cellular biologist can only study a few things about a cell at once. Tanner wants to help them ask bigger questions. Instead of fluorescent molecules, he sticks antibodies loaded with heavy metals to different cell components. That gives biologists 37 different isotopes in the lanthanide family to choose from.

After labeling different parts of a cell with different metals, Tanner then tosses the cell into plasma heated to 7500 K. Molecules break apart into a 2-mm cloud of atoms. The cloud passes through a modified mass spectrometer that detects the amount of each now-ionized heavy metal isotope present—measuring, for instance, the relative quantity of DNA versus surface proteins.

In 2011 a group at Stanford used Tanner’s device to study 34 parameters in human hematopoietic cells. Generating an enormous amount of data in the process, they constructed a family tree that revealed the relationships between different populations of cells and tested the effects of various drugs.

Devin Powell is a freelance science writer based in Washington, DC. His stories have appeared in Science News, Wired, US News & World Report, and other outlets.