Materials - Bioengineering

In recent years, researchers have been studying the principles supporting such perfectly adapted biological structures, and materials developers are now trying to put that research to practical use. Inspired by nature's capabilities, these experts are using cells, biomolecules and biological concepts to create new materials. "Nature has optimized its matter over millions of years—we're trying to profit from that," says Rainer Nies, who is working on potential applications in the field of bioengineering at Siemens Corporate Technology (CT) in Erlangen, Germany.

Researchers would like to duplicate organic materials' precise structuring, which can be measured in nanometers (one billionth of a meter). Similarly precise synthetic materials would make it possible to further miniaturize electronic and optical components and enhance their properties. For instance, Prof. Peter Greil and his team at the University of Erlangen are using biomaterials as templates for industrial materials. In one process, Greil's team decomposes a piece of wood in a nitrogen atmosphere at about 1,800 °C, leaving behind a skeleton of pure carbon. Liquid or gaseous silicon is then pumped into the chamber, bonding with the carbon to form silicon carbide, an extremely hard compound (see image below). The key point is that the wood's cellular structure is preserved in a kind of "petrified" image; it's almost impossible to produce a comparably porous ceramic material using conventional methods. Such biomorphic ceramics could someday be used as catalyst carriers, filters, high-temperature insulation or construction materials.

Bacterial Cages for Precious Metals. Wolfgang Pompe and his team at the Technical University of Dresden are taking a different approach. They are using bacterial proteins to generate densely packed nanoclusters of precious metals for use in catalysts and sensors. Many types of bacteria, such as Bacillus sphaericus, have numerous uniformly sized pores in their protein coverings, allowing materials to freely move in and out of the cell. "It's like a molecular strainer," explains Michael Mertig, a member of Pompe's team.

The researchers isolate protein molecules and then exploit their capacity for self-organization. If chemical conditions are right, the proteins will reorganize themselves into two-dimensional layers with perfect pore structures, even in an artificial environment. "These surfaces can have a much larger area than that of a single bacterium," says Mertig. They can also be mounted on solid substrates such as the semiconductors and metals used in microelectronics. In effect, they act as nano-scaled "egg cartons," whose cavities can be used selectively to deposit metals that are effective catalysts, such as platinum and palladium.

The metal complexes in the cavities can not outgrow their biomolecular cages—the bacterial pores. A regular pattern of particles is thus created in which the particles have a diameter of just two nanometers. This pattern simultaneously emerges at millions of locations, a key requirement for future mass production of nanostructures. The precious-metal particles are also situated at intervals of just a few nanometers, meaning that their specific surface area is vast. The larger a catalyst's surface, the more reactive it becomes.

Siemens plans to exploit this catalytic potential to develop devices such as highly sensitive gas sensors. Here, the protein membrane, metal particles and all, will be mounted on a pyrosensor, where the mini-catalysts can then accelerate a chemical reaction such as the oxidation of carbon monoxide. Since these clusters are more than one order of magnitude smaller than those conventionally used, chemical reactions can be initiated even at relatively low temperatures. The pyrosensor measures the reaction heat that is generated and transforms it into an electrical signal that indicates the concentration of the toxic gas.

This project is still in its infancy. The key components—the pyrosensor and the protein layers on technical carriers—have been developed, but they still need to be combined. One thing that won't be a problem is the lifetime of the biological structure involved. There are indications that the proteins remain stable for over a year. In any case, they are not indispensible for the sensor's proper functioning. "They are only a means to an end in the production process," says project manager Dr. Reinhard Gabl of Siemens CT. He estimates that a finished product will be ready in about three years.

At first glance, the growth of catalysts in proteins doesn't seem to have much to do with natural processes. However, this procedure is based on the same principle of biomineralization that applies to bone formation. In both cases, the biological template—the bacterial protein or collagen framework—guides the germ formation and the growth of a solid inorganic mass. The difference is that in the case of the bone the inorganic material is hydroxyapatite, while the surface layers contain metal particles.

A Liver Grows in a Reactor. If artificial materials can be created by means of biological processes, why not create new materials identical to natural ones? Man-made biomaterials are, for instance, in great demand in prosthetic devices. But this application requires living cells preferably taken from the patients themselves. Tissue engineering in bioreactors can be used to transform the cells into customized replacement parts (e.g. bones, cartilage, liver tissue). These receive all the nutrients they need to grow, and, if necessary, a framework to attach themselves to. The cells then grow into the desired tissue in accordance with their respective genetic programs.

However, such growth can take place only if the physical and chemical conditions in the reactor resemble those in the human body. This in turn requires the presence of numerous sensors and sophisticated controls. If you want to grow a bone implant, for example, the cells must be put under pressure, just as they are under natural conditions in an organism. Only then will they be stimulated to grow in the desired direction. Pompe's group is trying to generate this pressure with the help of tiny piezo actuators that vibrate at high frequencies. These actuators are very similar to the structural components that Siemens produces for direct fuel injection in diesel engines (see article Invisible Revolutions). The rapidly increasing demand for tissue produced by biotechnological means is in any case a strong incentive to overcome any difficulties that still remain, according to Pompe.

Nerve Cells on Microchips. Even more sophisticated approaches are being pursued in an attempt to unite living cells and technical components—for example by combining nerve cells and semiconductor electronics. The long-term goal is a hybrid neurochip that could be used to build neuroprostheses such as those that would enable blind people to see again. Another possibility would be a neurocomputer that combines the capacities of biological and electronic intelligence.

Prof. Peter Fromherz at the Max Planck Institute of Biochemistry in Martinsried, Germany have already succeeded in making two or three neurons grow on a silicon chip according to a preset pattern. The chip is now being used to stimulate a nerve cell. The cell conducts an electrical impulse via biological contact points called synapses to another neuron, whose activity, in turn, leads to a change in the voltage at the transistor lying under it. In other words, the nerve cell and the chip communicate with each other, and they do so in a manner that does no harm to either of them. When connected to a pond snail's nerve cells, a silicon chip of this kind will function for weeks or even months.

But that's only the beginning. As soon as you have a network of roughly 100 neurons, in which each individual neuron can be monitored or selectively stimulated, it will be possible to experimentally test the basic concepts of brain research for the first time ever. A number of theories today attempt to explain how living neural networks function, and some computers also operate according to this model. But only a neurochip will enable researchers to observe the behavior of an actual nerve network cell by cell.

Non-biological applications are also conceivable. For example, the human brain easily performs many tasks that are difficult or impossible for a computer. But a future "mini-brain" on a chip might be able to connect items stored in a memory bank by means of associations. However, the process of developing such a chip may be long and difficult. Fromherz doubts whether researchers will be able to connect more than ten neurons in the next five years.

Fromherz's team is therefore pursuing a parallel strategy that promises faster results. They are using naturally grown neural networks consisting of sections of rat brains connected with microchips. One problem is that the rat neurons cannot be triggered individually, only in groups. The scientists are focusing on the hippocampus, the area of the brain that plays a key role in learning. Infineon has built a semiconductor chip containing 10,000 transistors to meet the special needs of these experiments. The researchers can use the chip to investigate the activities of nerve cells at previously impossible resolutions. Fromherz has high expectations for their research. "I'd like to use the brain sections as a learning network controlled by a microchip," he says. Such basic research, he believes, will make it possible to find out how nerve tissue communicates with microchips.

The results of such research could, for example, help speed up the development of an artificial human retina. Although the creation of an "electronic eye" may still lie far in the future, bioengineering is already making tremendous strides. Whether it's nanocatalysts in bacterial proteins, artificial bones or artificial organs—bioengineers are creating materials with previously undreamed-of properties. These materials are well on their way to creating a new symbiosis of nature and technology.

Carola Hanisch