Pain Avoidance: Probing New Ways to Monitor Glucose

Bob Michaels and Brian Buntz
Researchers are striving to develop innovative glucose monitoring techniques that could eliminate the need for diabetics to prick their fingers

For millions of diabetes sufferers, pricking the skin to obtain a drop of blood to test glucose levels is uncomfortable at best and a potential cause of infection at worst. It’s hardly surprising, therefore, that researchers are busily investigating alternative methods for testing glucose that do not rely on traditional lancets and pinpricks. Thus, while some research teams are focusing on such needleless methods as biosensors and chips to monitor glucose levels in a range of body fluids other than blood, others are striving to make needles as small and noninvasive as possible to minimize pain.

Considering the vast and growing scope of diabetes today, pinpricks are no small matter. Worldwide, one in 10 people are afflicted with the disease, according to the World Health Organization’s World Health Statistics 2012 report. In the United States, diabetes afflicted 25.8 million in 2011, or 8.3% of the U.S. population. Behind these raw numbers stands a healthcare catastrophe in the making. Diabetes, according to the National Diabetes Fact Sheet, 2011, published by the Centers for Disease Control and Prevention, is the leading cause of kidney failure, nontraumatic lower-limb amputations, and new cases of blindness among U.S. adults. It is also a major cause of heart disease and stroke, ranking as the seventh leading cause of death. Such statistics underscore the importance of dilligent diabetes management, but it also couldn't hurt to eliminate the ouch factor from the process.

Multifluid Sensing Capability

SEM images show platinum-decorated graphene nanosheets that are key components of a new type of biosensor that can detect minute concentrations of glucose in saliva, tears, blood, and urine. (Image courtesy of Purdue University/Jeff Goecker)

Centered at Purdue University (West Lafayette, IN) and the Center for Biomolecular Science and Engineering at the U.S. Naval Research Laboratory (Washington, DC), a group of scientists involved with diabetes research have engineered a noninvasive, low-cost biosensor that could detect glucose in concentrations as low as 0.3 micromolar not only in blood, but also in urine, saliva, and tears.

“The nanostructured glucose biosensor works just like current-generation finger-prick blood glucose monitors do,” explains Jonathan Claussen, a research assistant professor at the Naval Research Lab. “Namely, glucose oxidase interacts with glucose found within a biological sample such as blood.” This interaction ultimately produces electrons that can be monitored on a digital readout. The generated electrical signal, Claussen adds, is subsequently calibrated to known glucose concentrations, enabling the researchers to correlate the digital readout to glucose concentrations even when the glucose concentration in the sample remains unknown.

The biosensor is constructed from layers of graphene nanosheets, platinum nanoparticles, and the enzyme glucose oxidase. The edges of each graphene layer have dangling, incomplete chemical bonds that attract the platinum nanoparticles. The combination of the graphene nanosheets and the platinum nanoparticles forms electrodes, which generate a signal when the glucose oxidase attaches to the platinum nanoparticles and converts glucose to hydrogen peroxide.

While this biosensor monitors glucose levels as do standard technologies, it works better, according to Claussen, because of the properties of the graphene and platinum nanoparticles. “These nanomaterials are highly electroactive on the nanoscale and have a very large surface area,” he says. “Thus, the nanomaterials are able to funnel electrons from the glucose oxidase/glucose reaction more readily than conventional materials, making our biosensor more sensitive over a wider glucose concentration range than current glucose monitoring devices.”

The biosensor’s ability to monitor a range of body fluids is attributable to its wide sensing range, comments Anurag Kumar, a Purdue University doctoral student and member of Claussen’s team. “What’s unique about our technology is that this single sensor can detect glucose levels in different body fluids simultaneously. You don’t need different sensors for different fluids.” This capability, in turn, is rooted in the sensor’s structure. “The sensor’s nanosheets—2-D materials standing vertically on a flat substrate—lend the structure added dimensionality,” Kumar says. “This dimensionality enables a very high density of platinum nanoparticles per unit area to form on top of the sensor, resulting in excellent sensing performance.”

Growing the nanostructures on a silicon chip via chemical vapor deposition and then electrodepositing platinum nanoparticles on the graphene proved to be the most challenging fabrication step, Claussen notes. After these steps were accomplished through trial and error, the team was able to use glucose oxidase enzyme immobilization techniques to transform the nanostructured chip into an active glucose sensor.

But more work remains to be done before the technology is ready for commercialization. “Since our tests to date have not involved the use of actual biological samples, we hope to test the device in such fluids as blood and tears in a clinical trial setting going forward,” Claussen states. “We predict that this type of testing and optimization will continue for another five to ten years before we have a finished product.”

Monitoring Glucose Levels in Saliva

Like its colleagues at Purdue University and the U.S. Naval Research Laboratory, a team of researchers at Brown University (Providence, RI) is also developing a biochip that can measure glucose without relying on blood. Using plasmonic interferometers, assistant professor of engineering Domenico Pacifici and his team have engineered a technology that can detect glucose at levels similar to those found in human saliva.

“The concentration of glucose in saliva is about 100 times less than that in blood,” remarks Vince Siu, a PhD candidate in biomedical engineering at Brown and a member of Professor Tayhas Palmore's group working in collaboration with Pacifici's team. “Our plasmonic interferometers are unique because they can be tuned and made more sensitive to very small concentrations of glucose.”

One of the team’s devices consists of a single slit about 100 nm wide etched between two 200-nm-wide grooves on a metal film, according to Siu. Incoming incident light hitting the surface of the chip is scattered at the grooves and interacts with the free electrons on the metal surface to create surface plasmon polaritons (SPPs), an electromagnetic wave with a wavelength smaller than a photon in free space. The SPP waves originating from the two grooves can interfere with the incident light going through the single slit, determining maxima and minima in the light intensity transmitted through the slit.

When an analyte such as glucose is present on the sensor surface, a relative phase difference occurs between the counterpropagating SPP waves, causing a measurable change in light intensity in real time. By etching thousands of plasmonic interferometers on top of a 1 × 1-cm biochip, each of which acts as its own detector, an array of thousands of plasmonic interferometers can be built, enabling the biochip to determine the fingerprints of the analytes, Siu explains. “Our plasmonic interferometers can be tuned to measure very low concentrations of analytes such as glucose in saliva by simply varying the distance between the slit and the two grooves on the chip.”

Fabricating the biochip entails several steps. The first is to deposit a thin metallic layer on top of a transparent chip using E-beam deposition. The second involves etching nanostructures on top of the metallic surface using focused ion-beam milling. And the third requires forming a polydimethylsiloxane microfluidic channel to accommodate the flow of the analyte. The chip’s optical setup consists of a broadband white-light source, a microscope, a charge-coupled-device camera, and a spectrograph.

“Detecting tiny but clinically relevant concentrations of a molecule such as glucose in saliva was challenging,” Siu comments. “We had to fine-tune the optical setup and the plasmonic interferometers to achieve this goal. To tune the device, the team characterized and modeled the optical response of hundreds of plasmonic interferometers by varying the distance between the slit and the two grooves, selecting the devices that demonstrated the best interference response for the sensing purpose.

While a positive correlation has been demonstrated between blood glucose and salivary glucose levels, a standard method of measuring salivary glucose has yet to be achieved. Nevertheless, because the plasmonic interferometers are sensitive enough to measure glucose in saliva in real time, the Brown researchers hope that they will be able to determine a more-accurate time-course correlation between blood glucose and salivary glucose levels.

Microneedles Detect Blood Glucose

A scanning electron micrograph shows a single microneedle equipped with a sensor used to detect compounds such as blood glucose. (Image courtesy of Talanta)

In contrast to the research under way at Purdue, the U.S. Naval Research Lab, and Brown, researchers at North Carolina State University (NC State; Raleigh), Sandia National Laboratories (Albuquerque, NM), and the University of California at San Diego (La Jolla, CA) are developing a diabetes-treatment technology based on microneedles. If successful, their effort could result in a painless or low-pain mechanism for sensing glucose levels and delivering insulin.

While several previous efforts have aimed at the development of a microneedle technology for monitoring glucose, these efforts involved devices in which the microneedles and the sensors were separated by channels. “Our innovation is that we have the sensing element at the tip of the microneedle,” explains Roger Narayan, a professor at NC State. “By putting the sensing component at the tip, you don’t have to worry about biofouling of channels within a microfluidic device, which could be a big problem for microneedle-based sensors that one would like to use for some period of time,” he adds.

The researchers use two different rapid prototyping methods to create their microneedles: digital micromirror device–based stereolithography and two-photon polymerization. While both methods involve performing selective polymerization of a liquid resin, the former technique produces structures with a substantially higher throughput. In addition, the researchers are evaluating new classes of materials to fabricate microneedles, Narayan says.

Optimistic about the technology’s potential clinical applications, the research team envisions that the microneedles could eventually be incorporated into a wearable device resembling a wristwatch to continuously monitor insulin and other compounds in the blood. In such applications, the device would integrate several components, including a skin interface, a sensor, and a delivery system. To that end, the researchers are actively partnering with several companies to work on commercializing the technology.

Looking further toward the future, Narayan and his colleagues imagine that the microneedles could be bundled together with sensors and pumps to create a minimally invasive wearable artificial pancreas. For example, the microneedles could be equipped with a variety of sensors for detecting glucose and other chemicals in the blood, including pH and lactate.

While microneedles for use as drug-delivery devices are starting to reach the market, work remains to be done on the devices' sensing component. “I think it will take a little bit more time to commercialize microneedle-based sensors because improvements in biocompatibility at the skin-device interface are needed if you want to have continuous monitoring at the skin’s surface,” Narayan comments. “If you are using something as just a replacement for a hypodermic needle, you don’t have to worry about the long-term biocompatibility, whereas if you want a sensor to interact with the skin for a while, improving the tissue-material interface will become really important.”