Back in 2012, DARPA wanted to create a version of the human body that could be put onto a series of chips: a collection of 10 separate organs connected with blood vessel-type structures and regulated automatically -- and it was prepared to pay $37m to get it built.
The funding for the project went to the Wyss Institute, a Harvard unit that works on biologically inspired engineering projects, and the organisation expects to have a working system by next summer.
Human organs are hugely complex things -- made of different cell types, specialised environments, conduits, and interactions. Yet the Wyss has been able to make working versions of them on a chip the size of a USB drive.
The organs-on-a-chip consist of a block of silicon rubber with one main channel running through it, which is divided into an upper and lower chamber by a porous membrane. The membrane acts like the natural boundary that the body creates between two tissues in an organ. In the case of the lung, that boundary is formed between the cells of the alveoli and blood vessel cells, for example: a boundary where oxygen and carbon dioxide cross back and forth between air and the blood.
In the chip, there are two layers of human tissue (the upper layer of lung, the lower layer of capillary) that, rather than being separated simply by human cell membranes as they would be in the human body, are divided instead by a super-thin bit of rubber, and controlled by a computer unit rather than a human breathing.
In the Wyss Institute's lung-on-a-chip, air flows over the upper layer and blood flows through the bottom, exactly like in a human lung. Air is suctioned through two side channels, separated from the main channel by only the thinnest layer of polymer, at a rate and flow that mimics human breathing, causing the walls of the main chamber to bow in and out just as if they were part of a human lung.
The chips were inspired by microchip manufacturing techniques, Professor Donald Ingber, head of the Wyss Institute, told ZDNet. "We used photolithographic etching like the computer chip manufacturing people do to make these fine features. You pour liquid polymer that's a clear silicon rubber onto an etched surface. Once it polymerises, you peel it off and you have all these fine features. That method was being to be used to make microfluidic systems -- hollow channels that have branches much like microvascular networks. I started exploring combining that with cells."
So far, the Wyss Institute has made several different types of organs-on-chips: the kidney, lung, heart, gut, bone marrow, and more. The aim is to eventually link them together to recreate the physiology of the human body, with all the chip organs working together in a single system, as per DARPA's request.
It's already got four organs on a chip linked and kept functional for two weeks. By this summer it wants to do the same with seven separate organs for three weeks, and one year later, have the DARPA vision up and running.
Why did DARPA want to embark on this project in the first place? The hope it that this system can help people make drugs quicker. When a new disease arrives it can take years to invent a cure, especially as there is often a long period of testing with new drugs. But if you can test using organs on a chip, that testing time could be cut dramatically -- meaning you could release drugs more quickly, and maybe save lives.
"When we had all these pandemics, it became very clear that no matter how fast you develop a potential countermeasure or therapy, it would take years to get that to humans through the convential drug development process," Ingber said. "They felt that one of the real bottlenecks in the process is animal testing. They thought that developing a replacement for animal testing would be an enabler of a faster response to biological threats. DARPA also thinks about deference in terms of economic competitiveness, and they thought it would advance the whole biotech and pharmaceutical industry."
Before too long, the systems should be commercialised, with the automated testing unit able to automatically analyse the liquids flowing out of the chips, and see what products are there -- useful for measuring things like inflammation. It could also have visualisation capabilities, so researchers can take the chip and get an insight into the physiological processes at work.
Already, the Wyss Institute has working chips where scientists can use different imaging modalities -- microscopic or fluorescence imaging, microfluorimetry -- or mass spectrometry to analyse what's going on in a healthy or diseased organ. In some chips, it's possible to measure the barrier function of the organ by tracing which charged particles move across it by putting one electrode above the membrane and one below. In the heart on a chip, multiple electrode arrays are used to measure the electrical activity in the cells, or stimulate them with an electrical charge.
The hope is that one day, organs-on-chips can be used to replace the animal testing typically used for exploring the safety of drugs and other chemicals before they reach the market. As well as being expensive and time-consuming, animal tests are not always reliable, and come with ethical concerns that the organs-on-a-chip don't.
One of the first organs that the Wyss Institute started working on was the lung, due to its relative straightforwardness. "I thought we could get a beautiful tissue-tissue interface, because it's a really simple tissue -- one cell layer of epithelium and one cell layer of endothelium and they share the same extracellular matrix and basement membrane, and we really had a shot at mimicking it," Ingber said.
Some of the earliest experiments with the lung-on-a-chip involved tracking how the lung absorbed nanoparticulates -- the tiny microscopic materials found in smog.
"Not only could we show absorption of [the nanoparticulates], we could show that the breathing motions were critical for absorption," Ingber said -- a research first.
After talking to a pharmaceutical company, the institute gave the miniaturised lung with a drug used in chemotherapy, IL-2, that was known to have the side effect of causing pulmonary oedema -- fluid on the lung, typically seen in pneumonia. The lung-on-a-chip responded like a lung-in-a-person -- and demonstrarted that the oedema was exacerbated by the lung's breathing motion.
"We showed that we had a disease model and a drug toxicity model. We saw [IL-2 caused pulmonary oedema to develop on the lung-on-a-chip] with the same dose and over the same time course that you see it develops in humans, we saw it develop in chip. And we discovered, again, that the breathing motions were important for this toxicity -- which was never known before," Ingber said.
In future, the organs-on-chips look likely to become more specific -- for example, being tailored to pharmaceutical, cosmetic, or chemical companies specific needs. But that's not all -- one day, they could be used to mimic individual humans, ushering in an era of personalised medicine. Chips could be produced using a person's own cells, each of which contain a genome that's unique to them -- allowing them to get a window into, say, how their body would react to two different treatment options, without having to start the treatment themselves. The organ-on-a-chip will absorb the drug, respond, heal or otherwise, while the human patient doesn't have to raise -- or risk -- so much as a little finger.
Read more about medical and health technology
- AI, MD: How artificial intelligence is changing the way illness is diagnosed and treated
- Two arms good, four arms better: How robots are redefining the future of surgery
- Retinal implants: How in-the-eye wireless tech is giving blind people back their sight
- The ghost in the machine: Vicarious and the search for AI that can rival the human brain