Using Carbon Nanotubes To Pinpoint Optimal Harvest Time in Bioprocessing

A conversation with Sepehr Hejazi, Iowa State University

When Nigel Reuel, Ph.D., returned to academia from a stint in industry, he and one of his first students, Nathaniel Kallmyer, envisioned a nanosensor capable of measuring enzyme activity directly within a biological environment. They wanted a tool that could function in crude cell extracts, bypassing tedious purification steps that typically delay data acquisition.

That was 10 years ago. Today, Reuel’s research group at Iowa State University has moved beyond theory, publishing extensively on the use of single-walled carbon nanotubes (SWCNTs). These sensors exhibit intrinsic fluorescence in the near-infrared window II (NIR-II), a range where biological media like lysate and fermentation broth become optically transparent and have low background signals. 

Nanotubes can give process developers a real-time monitoring tool for complex environments that would otherwise stymie traditional optics and also introduce non-specific responses. Beyond confirming protein production, SWCNT sensors can indicate proper folding and activity.

They also serve a dual purpose. They aid during early-stage optimization when gathering real-time data from test batches (optimization phase) and provide a clear endpoint measurement tool (quality control phase). Simply add a sample to the SWCNT solution for a clean readout. 

The project, which began as a quest to accelerate protein prototyping, now has implications for commercial CMC and scale-up, Reuel said. The team can now screen directly from a bioreactor to optimize media and timing, pinpointing the exact moment of peak expression to trigger harvest. Process analytical technology can make that step happen automatically, the moment the signal drops.

At CHI’s PepTalk conference in San Diego, a research assistant in Reuel's lab, Sepehr Hejazi, Ph.D., presented the team’s progress in enhancing sensor quality and signal-to-noise ratios. Hejazi's work includes qualifying these SWCNTs for use in cell-free extracts to monitor enzymes for various substrates. 

We spoke with Reuel about the trajectory of the technology and later met with Hejazi after his presentation to ask follow-up questions. The transcript below is edited for clarity.

Can we start at the top with a quick overview of what you're trying to solve with single-walled carbon nanotubes?

Hejazi: Carbon nanotubes enable us to measure enzymes and other protein activity in real time while they're getting expressed. In this work I presented here today, I showed that we can measure protease activity with carbon nanotubes in real time.

This enables us to get the activity measurements at peak expression. If you only do endpoint measurements, they might pass their peak activity. With carbon nanotubes, you're able to do it in real time, which is beneficial for protein engineering campaigns.

How does the technology work? It seems broadly applicable in commercial and clinical manufacturing, and especially for timing the harvest — knowing when expression is at its peak so you can initiate harvest.

Hejazi: Carbon nanotubes are fluorescent probes, and, in cell-free systems, they are transparent in that window. So, you're just adding carbon nanotubes to your cell-free expression system. As long as you have the carbon nanotubes inside, you can get the fluorescent signal out of your cell-free reaction systems. If anything happens to those carbon nanotubes, you're able to interpret those signals to indicate substrate degradation or enzyme affinity.

Lysate in cell-free expression systems is quite cloudy. Traditional methods of monitoring don't actually shine through that lysate. Is that correct?

Hejazi: That's completely correct. We are dealing with turbid solutions. The cell-free systems we use are cell-lysate-based. We culture and lyse those E. coli and use the machinery inside to express our proteins.

You're not dealing with very pure, very transparent solutions, but the good thing about carbon nanotubes is their range of their fluorescence. Their fluorescence is around 1,000 nanometers, and most of the biological molecules are transparent in that window (roughly 1,000 nm to 1,400 nm). So, you're looking at your solution, it's cloudy; however, the fluorescence that comes out of it is clear.

About The Expert:  

Sepehr Hejazi is a Ph.D. candidate at Iowa State University where he is exploring advanced screening methods for protein characterization. He received a B.Sc. in chemical engineering at Sharif University of Technology and an M.Sc. from Iowa State. Connect with him on LinkedIn.