When research scientist Hsue-Chia Chang imagines full-blown smart biomanufacturing, he envisions a huge factory with hundreds of reactors housing different batches of bone destined to help correct defects in the bodies of cancer patients, crash victims and injured soldiers.
Right now, though, the process for making bone or any other human tissue has more in common with the artisanal, small-batch coffee Chang might grab on the way to work than with the smart, connected factories of Industry 4.0.
Hsue-Chia Chang, Bayer professor of chemical engineering at the University of Notre Dame, has created a prototype sensor for detecting proteins in a bioreactor that improves on the enzyme-linked immunosorbent assay, or ELISA, the current gold standard for measurement. Provided by The University of Notre Dame
“Most require hand-done sample preparation, extraction, purification and so forth,” said Chang, who is Bayer professor of chemical engineering at the University of Notre Dame. “It’s too complicated, too personnel-intensive and doesn’t lend itself to an automated system.”
Until a couple of years ago he was focused on cancer and infectious diseases biomarkers for diagnostic purposes. But Chang heeded the call to nudge human tissue fabrication to a process more familiar to the 21st century than the early 1800s.
As he sees it, working on sensors for bioreactors is one doable task among the many hurdles that have to be surmounted to get to an FDA-approved implantable tissue.
BioFabUSA, a U.S. Department of Defense-funded program of the Advanced Regenerative Manufacturing Institute (ARMI), has taken on the job of working beyond the current practice of offline or at-line monitoring using destructive testing and is funding work on sensors for automated, in-line monitoring. BioFabUSA formed a community working group two years ago to understand what sensors were needed by the burgeoning cell and tissue industry.
“In order for us to really have a tightened and well-controlled process, we’re going to need to have insight into what is going on in that process,” said Mary Clare McCorry, director of technology and process development at ARMI, a Manufacturing USA innovation institute. “Sensors are the way we can look behind the curtain of the manufacturing process to be able to monitor it and implement appropriate (process) controls.”
Until now, researchers and commercial enterprises have used a recipe-driven approach, mixing a heavy portion of hope with growth-factor solutions and nutrients that are injected into a tissue reactor at predetermined times. They keep their fingers crossed that the stem cells inside will mature into the desired tissue. That’s because there’s no way of knowing whether the desired differentiation and maturation of the cells are taking place until the tissue is harvested weeks to months later. As a result, a large fraction of the cultures fails or has unacceptable batch-to-batch variations.
Also as a result, even the experts don’t know for certain all of the relevant biomarkers present in a bioreactor as stem cells grow and mature into tissue. If these would-be tissue manufacturers are ever to gain regulatory approval, they must prove that each batch is dependably and reliably the same in terms of safety and effectiveness.
“What we want to get to is this point where we can have continuous monitoring that goes into a feedback loop where we can nudge the process and have corrective actions happen throughout,” said McCorry.
Mary Clare McCorry, director of technology and process development at ARMI/BioFabUSA, describes how a capacitance-based sensor, typically used to measure the level of viable biomass inside a bioreactor, is being adapted by her organization and North Carolina State University’s Binil Starly to apply the probe to measure viability of cells embedded within a tissue matrix. Provided by ARMI/BioFabUSA
On the wish list are real-time, automated, low-cost inline sensors to tell what’s going on with cells during tissue development, including the appearance of metabolites such as glutamine and glutamate, lipids and proteins. Finding a sensor for sterility would be like finding a Holy Grail of sensors—and would be a major breakthrough for ensuring the safety of cell and tissue products.
“Right now, the gold standard for sterility is a 14-day culture test,” said McCorry. “And a lot of our products need to get to the patient within one to two days, so that doesn’t do us any good.”
So far, researchers have used well-established sensors for pH, temperature, dissolved oxygen, and the cell metabolism biomarkers glucose and lactate.
Problems arise because even though the sensors have been in use for years, biomaterial researchers are using individual detectors for a lot longer than their intended “use by” date. Sensors intended for continuous use for the duration of a culture are usually designed or validated for seven to 14 days’ duration without significant signal drift. They then need recalibrating. Some tissues being grown require more than two weeks, some as long as two months.
Another issue beyond duration of use is the form factor of available sensors for large-size, stir-tank bioreactors. Cell and tissue manufacturing has many different bioreactor form factors to which sensors need to conform.
To help fill the gap for protein detection, Chang’s lab is working on an anion-exchange membrane sensor. Projects at Georgia Institute of Technology include a thin-film elastomer that can be used for multiple points of detection for a more complete picture of what is happening in the bioreactor and a 3D printed capsule for real-time, in-situ detection.
W. Hong Yeo’s team at Georgia Tech developed a disposable, thin-film elastomer sensor with integrated wireless circuitry. A prototype attaches to a cell culture bag, fittingly designed to be low-profile while attached to the wall of any bioreactor. Used for multiple points of detection, it can provide a more complete picture of what is happening inside.
“We targeted the disposable cell bag because it can easily integrate our sensors while avoiding any contamination issues,” said Yeo, who is an associate professor of mechanical engineering and director of Georgia Tech’s Center for Human-Centric Interfaces and Engineering. “Our sensors are ultra-thin, with very small form factors. Our key focus is to miniaturize a system and then change the mechanics in a way that all the sensors and electronics are very soft so there’s no interruption of the natural fluid motion in the cell bag.”
The team is testing an electrochemical sensor designed to work for seven days that measures pH, dissolved oxygen, glucose and temperature. They already have feedback that the sensor must work for multiple weeks. “We are getting input that it would be better to monitor other values, so that’s something we are trying to integrate,” Yeo said. “There are other biomarkers of various types they want to measure—microorganisms, mycoplasma, viruses, endotoxins [all for sterility] and others.”
An advancement of the type Yeo described could be applied to in-line, real-time sterility detection.
In another lab at Georgia Tech, Billyde Brown and his team have produced a wireless sensor system with three main components: a multiplexed, microfabricated sensor chip, data acquisition and wireless transceiver electronics, and 3D-printed, biocompatible, sterilizable and waterproof packaging.
“This sensor capsule is something we’re developing to do real-time, in-situ monitoring of critical quality attributes in a tissue growth bioreactor,” said Brown, senior research faculty and education and workforce development director at the Georgia Tech Manufacturing Institute. The sensor chip has to be exposed to the solution in the bioreactor via a sealed opening. The packaging protects the electronics, but it also prevents contamination of the culture media.
The sensors he and his team are working on are for pH, glucose, lactate, interleukin (IL) 6, IL 8 and vascular endothelial growth factor (VEGF) alpha, said Brown. Like Chang, his previous work has also included cancer, but his work was for the newest therapeutics using immunotherapy.
So far, the sensor has been successful in detecting pH and glucose for 14 days with negligible loss in sensitivity over that period, Brown said.
The system is even more flexible given its ability to integrate a large number of sensors in the array. A mini-GPS could be added to provide information about where the system is working within a large bioreactor, Brown said. Also, the capsules could be fitted for varying amounts of buoyancy to help locate them at different points in the reactor, giving accurate indications of the contents’ homogeneity.
At Notre Dame, Chang and his colleagues produced a prototype sensor made with positively charged polystyrene-divinylbenzene particles supported with polyethylene binder and polyamide or polyester fibers, a material also used in water desalination and wastewater treatment.
The anion-exchange polymer membrane material is highly charged to eliminate protein fouling. Also, the membrane can be changed to a negatively charged polysulfonate cation-exchange membrane, depending on which protein a user wants to detect.
“What’s unique about these ion-selective membranes is that if you apply an electric field, only positive or negative ions can go through,” Chang said. “Nature wants to preserve electro-neutrality, so if you lose counter ions (countercharged to the charge of the membrane), you are also substantially reducing ionic strength near the surface of the membrane because of the net loss.”
If only cations go in, the anions will move away to preserve the neutrality. So, concentrations of both ions decrease near the membrane on one side.
“The result is a very thin, ion-depleted layer, at most 1 micron,” said Chang. “Almost all the voltage drop occurs across this ion-deficient region. Therefore, if you have any charged molecules on the surface, you can pick them up because the charged molecules will bring some mobile ions and because you have depleted all the ions before, so if you add just a few ions it changes the conductance significantly.”
Chang’s membrane sensor is comparable in performance to the current gold standard enzyme-linked immunosorbent assay, or ELISA. However, it’s superior because it doesn’t have to be recalibrated, requires no sample preparation, is faster (one hour vs. seven hours) and is lower in cost.
With the sensors developed by Brown, Chang, Yeo and others, cell and tissue developers will gain better insight into their processes and can move confidently toward scaled production, relegating artisanal processing to the local coffee shop.
Rohit Sharma and Prashant Tathireddy quit their university jobs to launch a startup and pursue a much bigger prize.
Their inline, multi-analyte biosensor can help revolutionize the artisanal, manual bio-manufacturing processes. Current practices include testing a bioreactor sample offline at different points in time, while the complex molecules inside are changing every second.
With more than 8,000 biopharmaceutical and more than 1,000 cell and gene therapy candidates in the pipeline, the need for better biosensors and the possibilities they can afford are immense.
“They have the potential to revolutionize the treatment of cancer or immune disorders,” said Sharma, COO of their company, Applied Biosensors LLC, Salt Lake City. “While this opportunity is great, there are many challenges facing the industry of biomanufacturers.”
Their single-use, smart probe uses a magnetometer and tiny pieces of novel polymers tuned to detect specific analytes. A single probe has five different types of polymers adhered to a permanent magnet. Each piece of polymer, measuring 1.5 × 1.5-mm, changes size volumetrically when it detects a change in the concentration of the analyte-of-interest. The change is picked up by the magnetometer.
“Any time the polymer increases in volume or decreases in volume because of the change in the analyte concentration, the magnet also moves up or down,” said Sharma.
A reader captures the change and, powered by a smart algorithm, correlates the change in volume to the change in concentration of the analyte.
“At the same time, you are collecting large quantities of data from a process that can help you develop robust predictive models to enable AI/ML that can ultimately accelerate R&D and increase the product yield of a batch,” Sharma said.
The probe overcomes two limitations of previously available sensors, said Sharma. It can be sterilized with both autoclave and gamma radiation, while available sensors are limited to autoclave sterilization. And it can be used for far longer than previous probes, which have had “use by” dates of seven or 14 days.
“This will be the world’s first in-line, multi-analyte sensor probe capable of continuous monitoring of glucose, lactate, osmolality and pH,” he said. “Internally, we have tested the sensor for 60 days and we have not seen any degradation in performance.”
Applied Biosensors’ technology is in prototype testing at pharmaceutical companies and is scheduled for commercial availability in the third quarter of 2022.
Connect With Us
