ARMIES of proteins are helping the builders of nanomachines examine their
handiwork. The proteins behave like miniature robots that wander randomly all
over the machines’ nanoscale surfaces, reporting back on the shape of the
device.
At the moment nanoengineers use high-resolution atomic force microscopes to
inspect their machines. But these only work if you can touch the surface of the
object with a probe. If the object’s surfaces are too hard to get at, it’s not
an option.
So Viola Vogel and Henry Hess at the University of Washington in Seattle
tried a different approach. They decided to try creating a fleet of randomly
moving robots, using two types of human proteins: kinesin motor proteins and
microtubules.
Advertisement
Microtubules are hollow rod-like proteins that provide pathways for other
proteins to move along. Kinesins normally latch on to the microtubules, and move
along them with inchworm-like “feet”, carrying essential molecules such as
neurotransmitters from one nerve cell to another.
But Vogel and Hess have turned this process on its head. Using a binder
called casein, they anchored kinesin proteins onto a polyurethane surface with
their wiggly “feet” sticking upwards
(see Graphic). Then they dropped
microtubules tagged with fluorescent markers onto the wriggling kinesin carpet.
Fuelled by the chemical ATP—their energy source in the cell— the
kinesin molecules pushed the glowing microtubules randomly across the surface.
As they did so, an optical microscope took 500 time-lapse snapshots of the
glowing proteins—one every 5 seconds. The researchers then simply merged
the snapshots on a computer to create an image of where the protein tubes had
wandered.
The surface used to test their technique sported raised columns about 1
micrometre high and 10 micrometres across. Because the stiff microtubules are
about 1.5 micrometres long, they couldn’t climb onto the raised area, and the
columns appeared dark in the pictures. At the moment, the system has a
resolution of about 300 nanometres, but Hess reckons further image processing
could refine it to about 50 nanometres—twice the diameter of a
microtubule.
For an uneven surface, Hess says you could focus the microscope at different
focal planes to gather information about the “landscape” at each one. Combining
these snapshots from different heights would provide a complete topographical
image of the surface. The research will be published in a forthcoming edition of
the journal Nano Letters.
This is just one example of how nanoscale robots could be used for imaging.
The Washington team suggests that molecular motors could be tagged with markers
such as pH-sensitive dyes, to check on the distribution of a chemical
in a nanoscale biosensor, for example. Such a device might be used to sense the
early signs of a bioweapons attack. Or a mixture of different types of probes
could measure various properties simultaneously.
The technique would also be ideal for studying contaminated or toxic
surfaces. “You’d just throw these nanoscale robots in, and leave them there,”
says Hess.
Hess thinks the technique could one day be extended to gather diagnostic
information about surfaces inside the human body, perhaps using radioactive
markers rather than fluorescent tags.
Markus Porto, a nanotechnologist at the Max Planck Institute for Molecular
Cell Biology and Genetics in Dresden is intrigued by the idea of using
protein-powered particles to image surfaces. “But the main strength of using
autonomous scouts to explore a surface lies in the fascinating possibility of
exploring [our bodies’] internal surfaces,” he says.