Posted on 05/30/2006 6:04:16 AM PDT by S0122017
Dawn of the zombies 27 May 2006 From New Scientist Print Edition. Subscribe and get 4 free issues. Robin Orwant
IT STARTS out like any silicon chip: intricate patterns are drawn with light and etched with acid. But this is no microprocessor destined for a computer. Instead there are minuscule chambers filled with human cells: liver cells, lung cells, fat cells, all connected by tiny channels. A nutrient fluid is pumped through the channels, flowing from one chamber to another just as blood flows from organ to organ in the body.
That's the whole point. The chip, created by Michael Shuler of Cornell University in New York, could be the first step towards a revolution in medicine, one that several teams around the world are working towards. "Our vision is building the human body on a chip," says Linda Griffith at the Massachusetts Institute of Technology. "A living, 3D, interconnected set of tissues."
“Our vision is building a human on a chip. A living, 3D interconnected set of tissues”The idea is to create not feeling entities but brainless mini-bodies that will be ideal for studying diseases and testing drugs. They could be infected with a virus and used to test treatments, for instance, or made to develop a degenerative disease, or given mutated genes to see what happens. They could also act as sentinels, warning us about nasty pollutants or an attack by bioweapons. In a few years, labs around the world might be filled with these "zombie" chips.
Shuler started out big rather than small. He was trying to improve computer models of drug metabolism, but even the best models are extremely crude approximations of what goes on in the body after someone pops a pill. "It's very difficult to anticipate all the possible consequences," Shuler says.
To check the models' predictions, he decided to build physical replicas of the body, connecting flasks of cells with tubing. It made for a cluttered lab bench. "They were very awkward to work with," he says.
Nor were these systems very realistic. The cells were bathed in relatively large quantities of fluid, for instance, diluting the toxic effects of any drug breakdown products. Instead, Shuler and his colleague Greg Baxter decided to go micro. Taking their lead from microfluidics and "lab-on-a-chip" technology, they reconstructed the entire bench-top set-up on a silicon chip. Over the years, they have built several different prototypes with chambers representing various organs, but always including the liver, the most important organ when it comes to testing drug safety.
When you pop a pill, the first place the drug goes after it is absorbed into the bloodsteam is the liver, which promptly tries to get rid of it. Liver enzymes convert drugs into forms that are easily excreted, and these "metabolites" can behave quite differently from the original drug. A drug might be toxic while its metabolites are harmless, or vice versa.
To prove that his chip is useful for drug testing, Shuler exposed one version to naphthalene, the chemical found in mothballs. Naphthalene itself is not toxic, but some of the metabolites produced by the liver are deadly to lung cells. Fat cells absorb some of these metabolites, however, reducing the amount that reaches the lungs. All this happens on the chip; cells in the lung chamber start dying as the dose of naphthalene increases. On the basis of such results, Shuler and Baxter have founded a company called Hurel to commercialise the device.
More recently, Shuler has built chips for testing multi-drug combinations for cancer. "We can explore a much larger combination of drugs than would be possible in animal studies," he says. The chips built for one such study, for example, have chambers containing "ordinary" uterine cancer cells and uterine cancer cells resistant to a chemotherapy drug. Other chambers contain bone marrow, as this tissue is most sensitive to cancer drugs, as well as liver and fat cells. Hundreds of drug combinations can be tested on these chips to see if any kill the normal and resistant cancer cells without destroying the marrow or liver, and Shuler's team has already got some promising results.
The big question is whether the results obtained with the chips apply to full-size people. Some researchers think Shuler is jumping the gun with his prototypes. That's because fresh human liver cells, taken from donor organs, die within days of being removed. They are also hard to come by. So in his chips Shuler is using cheap liver cell lines that grow indefinitely in a dish. The trouble is that these cells thrive only because of mutations that mean they no longer behave exactly like liver cells in the body.
Other groups are focusing on developing more realistic "organs-on-a-chip" before trying to combine them in a human-on-a-chip. At MIT, for instance, bioengineer Salman Khetani and his adviser Sangeeta Bhatia have come up with a way to trick fresh human liver cells into believing they are still in the body, so they function normally for weeks or months. The key is mimicking the interactions that occur between primary liver cells and the support cells, such as fibroblasts, that make up a third of the liver.
Restoring communications Although the support cells are isolated along with primary liver cells, the intricate structure of the liver is lost when cells are placed in a dish, destroying normal cell-to-cell communications. What Khetani has discovered is that surrounding tiny circular "islands" of human liver cells with a ring of fibroblasts seems to restore some of the signalling. By simply placing cells in this pattern, he has been able to keep fresh liver cells alive and well for more than six weeks. Tests with known toxins suggest these islands respond more like real livers than conventional cell cultures.
The plan is to put thousands of these islands on a plastic plate, allowing thousands of drugs to be screened at once. Unlike any existing cell-based tests, the system should reveal drugs or metabolites that kill liver cells slowly, over weeks, as well as those that kill within days. "That is crucial," says Khetani.
Khetani's islands may be more realistic than simply chucking liver cells into a chamber, but they are still quite different to a real liver. There, cells grow within intricate structures constantly buffeted by flowing blood, and this mechanical stimulation has important effects. Another group at MIT, led by Linda Griffith, is trying to recreate these conditions.
Griffith starts by etching a 3D pattern of microchannels in a silicon chip, coats the surfaces of the channels with a cell-binding protein and then adds rat liver cells. A nutrient fluid is pumped round the channels at a rate designed to ensure the cells get enough oxygen but are not too stressed by the mechanical forces of the flow. Unlike Khetani, Griffith does not try to arrange cells in a specific pattern: she has found that given the right environment, the primary and support cells start to arrange themselves into structures resembling those in the liver.
They also behave much like cells in a real liver. Griffith compared the levels of activity of genes encoding key liver enzymes in cells grown for a week in her system with levels in liver cells in the body and in various culture systems. She found that the activity levels in her system were similar to those in the body, whereas they fell in the other systems tested (Current Drug Metabolism, vol 6, p 569).
Griffith is now building chips with multiple "livers" for drug testing and is now working with human rather than rat cells. She is also getting defence funding to turn her liver-on-a-chip into a sensor for chemical or biological weapons. The unique advantage of a tissue-based sensor is that it should detect even unknown viruses or bacteria, she points out.
Griffith started her tissue engineering work with the idea of building replacement organs. Now her vision has changed: she hopes devices such as her liver-on-a-chip will help reduce the need for replacement organs. For instance, she hopes her system can be used to study hepatitis C, an increasingly common virus that slowly destroys the liver and is very hard to treat. The virus refuses to infect cells in a dish, so it is difficult to study using conventional methods. Another potential use is studying how viruses designed to deliver therapeutic genes will behave in the liver.
As promising as all this work is, the liver is only one organ. Other groups are trying to recreate different parts of the body. Shuler is working with various collaborators to create a gastrointestinal tract, another key organ involved in drug metabolism, as well as trying to build an artificial blood-brain barrier to reveal which molecules can pass through it and reach the brain. He says the work is progressing, though he will not reveal details.
At the University of Michigan in Ann Arbor, the various parts of the body Shuichi Takayama is trying to recreate include the circulatory system, the bone marrow, which nurtures various kinds of stem cells, and the lung. In Shuler's chip, the lung chamber consisted merely of lung cells immersed in liquid, Takayama points out. "But lung cells don't really behave like lung cells in just liquid."
Instead, he is engineering chips in which lung cells interact with liquid on one side and air on the other. One of the advantages of this system compared with animal testing is that the transparent chips allow the researchers to watch what happens to the cells in real time. "We have really solid, good results that might explain clinical situations," Takayama says, though details have yet to be published.
One of the reasons researchers like Shuler started creating microfluidic devices was to do experiments faster and more cheaply. Now, says Takayama, there is increasing recognition that they are also ideal for recreating conditions inside the body, where most cells grow in microfluidic environments.
Big pharma jumps aboard The next step is to combine improved versions of organs to create a better human-on-a-chip. Such devices could soon to be used to screen potential drugs for toxicity by pharmaceutical firm Johnson & Johnson. Subrahmanyam Vangala, a regional head of drug development for the US east coast, was so impressed when he heard about the human-on-a-chip concept about a year ago that he joined forces with Hurel. His team is improving that company's human-on-a-chip prototypes by using cells taken directly from human livers instead of cell lines and by incorporating 3D structures similar to Griffith's in the liver chamber.
Vangala hopes new approaches like the human-on-a-chip will give drug discovery a much-needed speed boost. In 2004 the number of new drugs reaching the market worldwide hit a 20-year low. There are many reasons for this crisis: a major one is that despite all the amazing advances in biology in recent years, researchers still lack reliable methods for predicting whether a compound that looks promising in the lab will be safe. This means enormous resources can be wasted on drugs doomed to fail.
Better ways of weeding out the duds could make a huge difference. Ideally, potential drugs would have to pass a whole series of toxicity screens long before animal testing begins, says biochemical engineer Jonathan Dordick of Rensselaer Polytechnic Institute in Troy, New York. Thousands of chemicals might be quickly tested for short-term toxicity using devices such as his MetaChip (see left). Drugs that pass this test could then be screened for longer-term liver toxicity in a system such as Khetani's islands. Those that still show no signs of toxicity could then be tested on a 3D liver chip like Griffith's and finally in a human-on-a-chip to reveal more subtle forms of toxicity affecting other organs. Only after passing all these stringent tests would a drug go to animal tests.
Humans-on-a-chip given various diseases, such as Shuler's uterine cancer chip, will also give researchers a more realistic way to test the effectiveness of therapies at an early stage. Together, all these technologies could make drug development faster, cheaper and easier. "A lot of people are still sceptical about these things," Vangala says. "But we need this kind of technology right now and we believe it will work. Eventually, we have to recreate a human or an animal on a chip."
It should be good news in terms of animal welfare, too. "We're not saying we can replace animal studies," cautions Shuler. "We just hope to make them more efficient and reduce our use of animals."
“We are not saying we can replace animal studies, but we hope to reduce animal use”One potential problem with humans-on-a-chip or organs-on-a-chip is that the cells come from specific individuals, yet different people can respond to drugs in drastically different ways depending on levels of the various liver enzymes. Some people's livers break down a drug so fast that a normal dose is not enough, while in others enzyme activity is so low that they need only a fraction of the usual amount.
The answer could be to use a mixture of cells in chips that reflects the genetic variation in the general population. But specific humans-on-a-chip could also help doctors tailor medicines to different groups, and ultimately perhaps even to individuals. Who knows, one day everyone might have their very own "mini-me" on a chip.
Robin Orwant is a freelance science writer based in Boston From issue 2553 of New Scientist magazine, 27 May 2006, page 40
The MetaChip While the human-on-a-chip concept could revolutionise drug testing, such devices will be too expensive to use at the very earliest stages. So Jonathan Dordick of Rensselaer Polytechnic Institute in Troy, New York, and Douglas Clark at the University of California, Berkeley, are mimicking what happens to drugs in the liver without using liver cells.
Dordick and Clark take samples of the main liver enzymes, called P450s, and fix microscopic dots of them to a glass slide. To each dot they add tiny amounts of a drug, which is converted into various metabolites by the enzymes much as it would be in the liver. Finally, cells of various kinds are placed on top of the dots to mimic what would happen when blood carries metabolites to different tissues. Regions of cell death reveal which metabolites are toxic to specific cell types. In their proof-of-concept study, the team showed the chip can detect the liver toxicity of acetaminophen (paracetamol or Tylenol).
Dordick can get up to 10,000 enzyme spots on a single slide, so he can test thousands of potential drugs on any given cell type in a few hours. "We can do this at speeds that allow drug companies to test at the earliest stages of drug development," he says.
The MetaChip would also make it easy to test how particular groups of people might respond to a drug, by adjusting enzyme ratios to reflect those typical of various populations. Dordick and Clark have founded a company called Solidus Biosciences to produce a commercial version of the device.
The MetaChip, however, cannot reveal long-term toxicity. And a few isolated enzymes can never capture the complexity of the body's response. "The way they're going about it is interesting," says Linda Griffith of MIT. "But there are an incredible number of ways a drug can be toxic."
human cell on chip ping
Terminator Seeds.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.