[Mb-hair] Gluing Rat Neurons...

[Jim Burns]

Whoops!  Here's the column...

 
SCIENCE
Edward Willett
 
Brains on a chip
 
Our current primary means of communicating with computers, typing, is
inelegant, to say the least. Some people never learn how to do it. Even
those who are good at must use a keyboard actually designed to slow
typists down. 
 
Wouldn't it be great if we could just think at a computer? Or better
yet, if a computer could do what we needed done without us having to
think about it at all?
 
A lot of scientists think so, which is why there's a lot of research
going on into the possibilities of directly interfacing brain cells and
computers.
 
In just the last few days, European researchers reported they have
created an interface between mammalian neurons and silicon chips.
 
Working with German microchip company Infineon, the research team, a
partnership known as NACHIP involving the Max Planck Institute in
Munich, the University of Zurich and the University of Padua, placed
16,384 transistors and hundreds of capacitors on a one-square-millimetre
chip.
 
Using special proteins found in the brain, they "glued" rat neurons to
the chip. This both kept the neurons in place (neurons tend to wander)
and helped transmit signals from the neurons to the transistors.
 
The electrical activity in neurons is the result of the movement of
charged particles, specifically sodium ions, in and out of special
pores. The neurons in the European experiment were genetically modified
to have more pores than usual. As well, the neural glue helped ensure
that extra sodium channels collected around the transistor and capacitor
connectors. This made it easier for the transistors to detect the
electrical activity in the cells and for the capacitors to alter the
movement of sodium ions within the channels and thus cause the neurons
to react.
 
With the transistors detecting cell activity and the capacitors
stimulating it, a two-way communication circuit was established between
the animate and the inanimate.
 
Previous work focused on snail neurons, which are larger and easier to
work with. That work taught the researchers how to deal with some of the
daunting challenges of interfacing neurons and silicon, most of which
relate to the fact that they process information very differently. In a
computer, for example, different functions usually take place in
different places within the circuitry. In the brain, many functions are
often carried out by the same piece of tissue. And if you cut away some
neurons from the brain, other neurons will move in and take over their
functions. Try that with a computer chip!
 
The successful creation of an interface between mammalian neurons and a
computer chip is a major step forward in creating an interface that is
actually useful. Years from now, it could lead to the creation of
organic computers that use living neurons as their CPU—or silicon
computers that use living neurons for data storage. It could make
possible neural prostheses to combat various neurological disorders,
giving sight to the blind or making possible prosthetic limbs that can
be moved as naturally as real ones.
 
In the shorter term, it may provide us with more insight into how
neurons form networks in our brains, and enable the development of
special chips that would enable pharmaceutical companies to more easily
and quickly test the efficacy of new drugs on neurons.
 
Of course (and don't I wish I got a dollar every time I wrote or said
this) more research is needed. The methods used for stimulating the
neurons have to be refined; current methods too often damage them.
 
The researchers say it should be possible to the chip to signal a neuron
to alter its membrane to allow it to take up the DNA for a new
gene—or, alternatively, to accept a compound that switches off an
existing gene.  Turning genes on and off in specific neurons will thus
be the next focus of the research team's work.
 
The day may be coming when the lines between the brain and the computer,
between the animate and the inanimate, blur to the point where it will
be hard for those who are truly "plugged in" to distinguish between
them.
 
I'm rather more intrigued than alarmed by the prospect, myself...as long
as the computer I plug myself into isn't using Windows. I'd hate for the
last words I see to be "This brain has performed an illegal operation
and must be shut down."
    

Edward Willett is a freelance writer in Regina, Saskatchewan, Canada.
E-mail comments or questions to ewillett at sasktel.net. Ed's website is
www.edwardwillett.com, and his blog is at edwardwillett.blogspot.com.
Ed's latest novel is the exciting science-fiction adventure Lost in
Translation (Five Star, ISBN 1594143056 ); his latest non-fiction
book is Genetics Demystified (McGraw-Hill, ISBN 0071459308).