Pioneering single-molecule imaging

Antibiotics used to treat bacterial infections effectively are tiny molecules. When studying them as a graduate student, I was very curious about how a tiny molecule inhibits the massive molecular machines it acts upon. In conversation, we described antibiotics as gumming up the operation of machines like a grain of sand in a gear. As I started to find my footing as a scientist, I was deeply motivated to understand the function of these cellular machines at a molecular level and to define their mechanisms precisely. 

As a graduate student in the lab of Joseph Puglisi, my focus was drawn to the ribosome, a complex assembly of RNA and protein, which was the target of nearly half of all known antibiotics. I hoped to one day apply the knowledge gained to develop new cancer treatments.

There were practical limits to structural biology methods at the time. While providing insightful static snapshots of parts of molecular machines, what I really wanted to do was watch one single ribosome as it functions. To observe this molecular machine undergoing structural changes and interacting with factors in an ordered sequence of events would be game-changing. 

As an early-career scientist, I dreamed about getting the structures of the ribosome in all the different conformations it adopts while performing its job of synthesizing proteins. It was extraordinarily motivating to cast this dream out into the future and know that’s the goal I was aiming at — I strapped myself to that goal and never let go.

Focusing on the signal, ignoring the noise

When I moved to Steven Chu’s lab in the Department of Applied Physics at Stanford University, I was fortunate to be surrounded by others with similarly lofty aspirations. My job switched from making fluorescent reagents for microscopy to learning how to build faster and more robust microscopes and establishing pipelines for understanding the behaviors of single molecules. 

The first evidence that one may be able to visualize the conformational changes of biological molecules at ambient temperatures came in 1996. This work, led by Taekjip Ha, a graduate student in the lab of Shimon Weiss at UCLA, used fluorescence resonance energy transfer (FRET), a technique developed as a molecular ruler at Stanford by Lubert Stryer. At the time of my joining, he and Xiaowei Zhuang were colleagues in the lab. They had collaborated to measure dynamic processes in a catalytic domain of an RNA molecule called the hairpin ribozyme. 

Seeing what they were able to discern using single-molecule FRET, I began to dream about what could be revealed about ribosome function and antibiotic action.

From start to finish, it took seven years to publish the first observations of ribosome function at the single-molecule scale. This work was one of the first demonstrations that single-molecule imaging could probe the dynamism of intact, complex molecular machines. While continuing to break new ground through explorations of ribosome function during my early faculty career at Weill Cornell Medicine, a second major breakthrough came through our efforts to apply single-molecule methods to integral membrane proteins, molecules roughly 10 times smaller than ribosomes.

This important collaboration with Harel Weinstein and Jonathan Javitch led to a deeper understanding of the transport mechanisms of sodium symporter family proteins, which include the molecular target of cocaine and other mood-regulating drugs. This advance convinced my faculty peers that single-molecule methodology was a potentially generalizable approach to interrogating the dynamism of both large and small molecular machines. 

This work inspired meaningful collaborations with the world’s experts on the function of other integral membrane protein families, where static structures had been determined but knowledge of a system’s dynamism was missing. The success of these collaborations solidified the power of the approach and demonstrated for the first time that medically relevant mammalian systems were amenable to quantitative biophysical measurements of function.

Consistent innovation bears fruit for single-molecule imaging

Through this research, it became increasingly clear that automated approaches for data processing were paramount. We first made progress on this front by borrowing a software tool from the ion channel research community developed in Fred Sachs’s lab called QuB. With the help of the Sachs lab, we repurposed QuB to assess our single-molecule FRET data. 

This kicked off an era of accelerated workflow and experimental throughput that has now become the standard in the field. In this space, our lab has developed and contributed an independent software pipeline called SPARTAN that is now broadly utilized by imaging centers and labs worldwide. 

In parallel, our team took aim at the fluorophores used for single-molecule research because our studies had revealed them to be inadequate for key applications, including rapid imaging. These endeavors led us to invent what are now referred to as “self-healing” fluorophores, which have revolutionized our ability to collect high-quality data. 

We also worked to improve our experimental throughput by employing state-of-the-art camera technologies. As a result of a flood that ruined our lab’s only microscope, we decided to gamble on cheaper, unproven scientific CMOS cameras. This gamble, together with our move to synchronize the outputs of multiple cameras simultaneously, led to dramatic increases in time resolution and experimental throughput, which permanently changed the landscape of what we could achieve. Using scientific CMOS cameras in place of EMCCD technologies is now standard in the field. 

Realizing the power of single-molecule imaging

Collectively, these advances contributed to collaborations with Nobel Laureate Brian Kobilka, which demonstrated that single-molecule FRET could be used to quantify ligand efficacy in G protein–coupled receptors (GPCRs). GPCRs control nearly every aspect of human physiology and are the most common drug target of clinically approved drugs. At St. Jude, and in collaboration with labs around the world, we continue to flesh out the potential uses of single-molecule imaging to inform on the mechanism of GPCR signaling. This helps us gain a deeper understanding of drug mechanisms of action and how new drugs might be developed with greater efficacy and specificity.

Moving my lab to St. Jude has fundamentally accelerated our lab’s goal of creating direct links between molecular structure and molecular dynamics. These efforts, which were made possible by the institution’s Cryo-EM Center and Single-Molecule Imaging Center and the very talented researchers who lead them, led to multiple breakthroughs in the area of protein synthesis, where we revealed the mechanisms of antibiotic action on bacterial and human ribosomes at atomic resolution for the first time. These efforts also unearthed some key differences between bacterial ribosomes and human ribosomes that contribute to the specificity of clinically used antibiotics. 

While it required more than 25 years to come to fruition, this period of change fundamentally transformed the potential of single-molecule imaging. Today, this approach is regarded as one of the only basic research tools capable of directly accessing dynamic processes in complex biological systems. 

In our efforts to push the field to new frontiers, I am reminded daily of the importance of innovating around obstacles. I am convinced that fundamental discoveries are the key to new medicines, and I feel incredibly fortunate to have the support of St. Jude Children’s Research Hospital to continue our advance towards this goal.