Millions of people worldwide depend on insulin for managing diabetes. However, the insulin molecule can cause complications when administered in large quantities. New research shows that injectable nanoclusters offer a potential solution for longer-lasting doses without complications. The development of these nanoclusters could have broader implications for treating people with various diseases requiring higher volumes of medication.
An estimated 150–200 million people worldwide rely on insulin to manage their diabetes. But the properties of the insulin molecule itself cause problems at higher doses. Injecting too much insulin at once can cause the proteins to tangle into a gloopy mess in the bloodstream.
New research, published in February in the International Journal of Biological Macromolecules, explores a way to sidestep the problem. Scientists from the Technical University of Denmark (DTU) are developing insulin nanoclusters that could one day enable people to take larger, longer-acting doses of insulin without gumming up their blood vessels.
Why reinvent the insulin injection?
People with type 2 diabetes can become even less responsive to insulin over time and “need to administer more and more insulin per day,” says co-author Andrew Urquhart, a chemist and Associate Professor at DTU who studies drug delivery. This means larger injections of insulin, but when the concentration in the bloodstream gets excessive, insulin proteins coalesce into a viscous goo.
Today, people who require high doses of insulin have to spread out their medication over several smaller injections – but researchers such as the DTU team hope to develop techniques to deploy higher doses of insulin that last longer, avoiding multiple jabs.
Previous research has made strides toward this goal by chemically packing the insulin so it cannot tangle in the bloodstream, Urquhart explains. A drug already on the market called NPH insulin clusters into microparticles after injection, enabling a higher dose to be delivered safely. However, NPH insulin struggles with shelf stability – meaning that the insulin mixture is prone to forming those microparticles prematurely while it is sitting in a jar, rendering it unusable. So Mia Danielsen, a pharmacist and lead author, set out to take alternative insulin delivery to another level – from micro to nano – and circumvent the storage issues.
The smaller, the better
Danielsen and her team are not the first to attempt to pack insulin at the nanoscale. But previous nanoparticles developed to deliver insulin have been woefully inefficient. Most of them are devoted to the matrix, essentially a vehicle made of a less chemically reactive substance, carrying a puny payload of insulin in the centre. “Mia’s work has focused on basically eliminating the matrix,” Urquhart explains. In an all-insulin nanocluster, “the drug is kind of carrying itself”.
Danielsen and the team initially tried to exploit insulin’s chemical structure to encourage it to clump on its own.
Molecules of insulin contain regions that are hydrophobic: strongly repelled by water. In the same way that oil in your pasta water forms beads, molecules of insulin cluster together with the hydrophobic regions at the centre to reduce the surface area exposed to the water.
But after trying various hydrophobicity-based tactics, Danielsen found that the clusters formed by insulin alone varied between 130 and 300 nanometres in diameter – too inconsistent for use as a drug and larger than hoped for. In nanomedicine, the target size is less than 200 nanometres, Urquhart says. Scientists are not certain why, but cells seem most likely to interact with particles beneath that threshold. (Most pathogens have also adapted to this and fall in the same size range.)
To get under 200 nanometres, Danielsen turned to chemical agents called crosslinkers to bring the insulin molecules closer together. Crosslinkers are almost like chemical carabiners that snap onto parts of two molecules and can help to stabilise a structure like a nanocluster. But adding another component to the nanocluster carries its own risks – it, too, has to be processed by the body. Danielsen chose a crosslinker with a disulfide bond, a common structure in cell biology that is easily broken down by specialised proteins.
The addition of the crosslinking agent successfully whittled the nanoclusters down to the target size. But would the mixture work as a drug?
Does a nanocluster still work as insulin?
Packing proteins very close together, especially with the help of a crosslinker to force them tighter, runs the risk of ruining their structure. It is like shoving too many action figures into a box – you may be able to squeeze more in by twisting their arms into strange configurations, but Superman’s famous right hook might be stuck behind his back forever.
To determine whether the insulin proteins in their nanoclusters were still in good enough shape to do the job, Danielsen and the team checked their structure using a device that tracks how microscopic particles absorb light based on their shape. They identified several structural changes caused by the crosslinking that could undermine the insulin’s effectiveness – and even make it toxic to cells.
Danielsen initially administered the nanoclusters to cells in petri dishes and found no evidence of toxicity. This established that the nanoclusters would not do harm, but would they help?
The next step was to determine whether the nanoclusters could actually achieve the task of insulin – improving glucose metabolism. Danielsen and her team found that rats that received nanocluster injections responded unusually to the insulin. Their glucose metabolism increased right after the injection, like a standard injection of insulin, but also spiked several hours later – evidence of a burst release followed by extended release of the insulin, according to the authors.
The researchers think that this delayed activity could be the effect of the disulfide bonds breaking down, freeing up more proteins. But ultimately, Urquhart explains, we do not know how nanomedicines interact with cells. How does the cell detect a nanocluster? Does the whole ball attach to receptors on the surface of a cell – maybe even multiple cells at a time – or do individual molecules calve off the cluster and make their way to receptors?
The future of the nanocluster
The researchers were pleased to find that the nanoclusters remained shelf-stable for at least 3 months when stored in water, but the insulin nanoclusters are still a long way from being used by people with diabetes, the team says.
Further experiments with the rats showed that it took dramatically more insulin in nanocluster form to achieve similar results to free insulin proteins – 30–35 times the dose. Combined with the effort and expense to form the nanoclusters in the first place, that jump in dosage would make these nanoclusters too expensive.
But learning how to make a matrix-free protein nanocluster could be its own reward – the lessons learned from insulin nanoclusters could be used to treat people with other diseases. “If you look through therapeutics and proteins, we have a problem with needing higher doses not just for people with diabetes. We need bigger volumes for many diseases, and maximum solubility can be a problem,” Danielsen concludes.
