Diabetes could affect nearly 370 million people worldwide by 2030, according to the World Health Organization. More worrying still, diabetes is now the second most chronic disease in children. For reasons that are still unclear, the rate of type-1 diabetes (T1D, also known as autoimmune diabetes) in children is increasing by about 3% every year, with a projected increase of a staggering 70% between 2004 and 2020.

Although T1D was once thought of as being exclusively a paediatric disease, around one-fourth of individuals now contract it as adults. The rate of type-2 diabetes (T2D, also called metabolic or diet-induced diabetes), normally seen in overweight adults, has also alarmingly escalated in children since the early 1990s, in part because of the global obesity epidemic. Until quite recently, it was fairly simple to distinguish between T1D and T2D populations but this is becoming more and more difficult, as the two are beginning to overlap. The main problem is that existing diagnostic tests are slow and expensive, and it would be better to detect diabetes as early as possible to ensure the best possible treatment.

Higher concentration of autoantibodies

T1D is different from T2D in that patients with the disorder have a much higher concentration of autoantibodies against one or more so-called pancreatic islet antigens (such as insulin, glutamic acid decarboxylase and/ or tyrosine phosphatase). Detecting these autoantibodies, and especially those against insulin (which are the first to appear), is thus a good way to detect T1D. Again, standard tests are not very efficient here and even the most widely used technique, radioimmunoassay (RIA) with targeted antigens, is far from ideal because it is rather slow and, of course, relies on toxic radioisotopes.

In an attempt to try and overcome these problems, researchers at Stanford have now developed a simple point-of-care autoantibody test that is more reliable, simpler and faster than RIA and similar tests. It works thanks to an islet antigen microarray on a plasmonic gold (pGOLD) chip and can be used to diagnose T1D by detecting autoantibodies against insulin, GAD65 and IA-2, and potentially new biomarkers of the disease. It is able to detect different types of autoantibodies in just 2 µL of whole human blood (from a finger prick sample, for example) and results can be obtained in the same day.

Enhancing the fluorescence emission

The team, led by Hongjie Dai, made its pGOLD chip by uniformly coating glass slides with gold nanoparticles that have a surface plamson reasonance in the near-infrared part of the electromagnetic spectrum. The pGOLD chip is capable of enhancing the fluorescence emission of near-infrared tags of biological molecules by around 100 times. Together with Brian Feldman's group, the researchers robotically printed the islet antigens in triplicate spots onto the plasmonic gold slide to create a chip containing a microarray of antigens.

"We tested our device by applying 2 µL of human serum or blood (diluted by 10 or 100 times) to it," explains Dai. "If the sample contains autoantibodies that match one or more of the islet antigens on the chip, those antibodies bind to the specific antigens, which are then tagged by a secondary antibody with a near-infrared dye to make the islet spots brightly fluoresce."

Antibody detected at much lower concentrations

The samples came from Feldman's patients that had new-onset diabetes. They were tested against non-diabetic controls at Stanford University Medical Center. The antigen spots fluoresce 100 times more brightly thanks to the plasmonic gold substrate, which allows the antibody to be detected at much lower concentrations (down to just 1 femtomolar) than if ordinary gold were employed in the microarray platform.

Plasmonics is a relatively new branch of photonics and is beginning to prove itself in the field of medical research with the discovery that specifically engineered metallic nanoparticles can be made to resonate in the infrared. The nanoparticles interact strongly with light via localized surface plasmons (collective oscillations of electrons on a metal's surface) and so act as efficient optical nanoantennas that capture more light. Such plasmonic resonances can be used as probes that can be tuned to and away from various vibrational modes in biological molecules, which themselves vibrate when excited with infrared light.

"We believe that our technology will be able to address the current clinical need for improved diabetes diagnostics," Dai tells our sister site nanotechweb.org. "The pGOLD platform is also being commercialized by a new start-up company Nirmidas Biotech Inc., based in San Francisco, aimed at better detecting proteins for a wide range of research and diagnostic applications. It might even be able to detect biomarkers for other diseases such as heart disease with ultrahigh sensitivity."

Related stories

• CNT transistors detect cancer biomarkers
• Shining a light on diabetes
• Nanosensor detects cancer in whole blood
• Eyes give easy access to glucose levels