Communication channel between cells and machines paves way toward bio-hybrid robots

Nov 29, 2012 by Lisa Zyga feature
Illustration of the experimental setup used to investigate the production of nitric oxide by genetically modified protein cells in response to light. Image credit: Yarkoni, et al. ©2012 IOP Publishing Ltd

(Phys.org)—While some advanced humanoid robots already look eerily lifelike, robots in the future may actually become partly alive. Currently, researchers are working on integrating living cells and other biological components with electronic components in an attempt to create bio-hybrid robots. These robots could act autonomously, imitate some animal behaviors, and have the ability to self-replicate some of their parts.

One of the major challenges in developing bio-hybrid robots consists of creating an interface to allow communication between the biological and electronic components. This task is inherently difficult because the vast majority of cellular signals are simply not compatible with electronics. Most use signals that travel too slowly to interface with , and exposure to electrical charges has negative effects on many , which can lead to cell death.

Despite these challenges, a team of researchers, Orr Yarkoni, Lynn Donlon, and Daniel Frankel from the Department of Chemical Engineering at Newcastle University in the UK, has developed a way to make communication between these two domains possible. In their study, the researchers genetically engineered protein cells from a Chinese hamster ovary to produce nitric oxide in response to visible light. When the were coupled to a platinum electrode, the resulting device could convert an incoming to a , which could then be converted into an electrical signal. This provides an important step in developing future bio-hybrid robots.

"The bio-electronic interface is notoriously difficult to manipulate," Frankel told Phys.org. "Usually the electronics are modified to fit the of interest with limited success. This is the first time that we have tried an alternative approach, to modify the biology (via genetic engineering) to fit the electronics."

One of the keys to opening a communication channel was identifying nitric oxide as a suitable chemical compound for the middle signal in the pathway. As the researchers explained, nitric oxide is one of the few compounds that meets all the requirements for this role: it can permeate cells, is highly diffusible, has a relatively short half-life, and is one of the rare cell signaling molecules that remains in its gaseous form in any cell environment. As an indication of its aptitude for signaling, nitric oxide is widely used in nature, playing a role in areas such as muscle relaxation, blood vessel widening (vasodilation), inadequate blood flow (ischaemia), penile erection, and stroke.

Several proteins naturally produce nitric oxide in response to a chemical stimulus, but the researchers wanted to produce nitric oxide in larger amounts and in response to an optical stimulus. To achieve this, the researchers explained how they were able to "hijack" the protein's electron transfer pathway by severing a particular gene coding for a binding site required by the native pathway. In addition to removing that binding site, the researchers inserted amino acid structures called light-oxygen-voltage (LOV) domains into one or two regions in the protein. Since these structures are capable of accepting an electron from an incoming photon, they make the protein produce nitric oxide in response to light.

In experiments, the researchers observed that mutant cells that were modified with LOV domains produced more nitric oxide than the wildtype cells when exposed to light. Increasing the light stimulus further increased the mutant cells' nitric oxide production, while the wildtype cells exhibited limited photosensitivity. The researchers also found that mutant cells with two LOV domains produced about twice as much nitric oxide as those with only one LOV domain.

When coupled to an electrode, the mutant cells could produce sufficient to cause a change in the electrode's current. By changing the light intensity, the researchers could control the communication signals traveling through this optical-chemical-electrical cascade. Unlike typical solid-state photodetectors, these devices have the ability to self-reproduce and the potential to combine input signals to perform computation. These advantages could be turned into useful applications in a future generation of advanced robots.

"This work is part of a bigger project to build a swimming bio-hybrid robot called 'Cyberplasm,'" Frankel said. "These cells are the 'eyes' of the robot able to respond to light and tell the robot to alter its swimming pattern in response. The complete robot is still in development, but we are working on the individual parts. Potential applications are robots that can swim into contaminated areas, detect chemicals and report back."

Co-investigators of the Cyberplasm project are Joseph Ayers at Northeastern University in Boston, Chris Voigt at MIT, and Vlad Parpura at the University of Alabama.

The next step for the researchers is to produce cells that can communicate with electronics in response to chemical stimuli, and in addition communicate with each other.

Explore further: Contamination likely explains 'food genes in blood' claim

More information: Orr Yarkoni, Lynn Donlon, and Daniel Frankel. "Creating a bio-hybrid signal transduction pathway: opening a new channel of communication between cells and machines." Bioinspir. Biomim. 7 (2012) 046017 (7pp). DOI: 10.1088/1748-3182/7/4/046017

4.6 /5 (12 votes)

Related Stories

Nitric oxide shown to cause colon cancer

Jan 20, 2009

(PhysOrg.com) -- Researchers long ago established a link between inflammation, cancer and the compound nitric oxide, which may be produced when the immune system responds to bacterial infections, including those of the colon. ...

Gas on your mind

Dec 11, 2006

Scientists at the University of Leicester are to gain a greater insight into the workings of the human mind…through the study of a snail’s brain.

Recommended for you

Identifying the source of stem cells

2 hours ago

When most animals begin life, cells immediately begin accepting assignments to become a head, tail or a vital organ. However, mammals, including humans, are special. The cells of mammalian embryos get to ...

Contamination likely explains 'food genes in blood' claim

Oct 29, 2014

Laboratory contaminants likely explain the results of a recent study claiming that complete genes can pass from foods we eat into our blood, according to a University of Michigan molecular biologist who re-examined ...

User comments : 15

Adjust slider to filter visible comments by rank

Display comments: newest first

Lurker2358
1 / 5 (2) Nov 29, 2012
Still far too big for what I'd like to see, such as the intelligent robotic immune system concept. At least they moving in a good direction here.

I would imagine such systems could be used to test all sorts of chemical and perhaps radioactive properties after accidents and other disasters.

Really need a nano-scale protonic-to-electronic converter so that you could control cell growth and behavior at the molecular level via computer.
sstritt
1 / 5 (2) Nov 29, 2012
Resistance is futile!
MrVibrating
1.3 / 5 (4) Nov 29, 2012
I get the impression that Cyberplasm (the bot mentioned above) probably isn't going to develop into Hector from the movie Saturn3 simply because, in the decades to come, digital neural networks are going overtake anything that might be achieved via neural-integration by centuries. Completion of the Human Connectome Project will lead to AIs far superior, not to mention simpler, than real 'brain-bots'...

Not to deny the benefits of such techs wrt say prosthetics of course... i just think that by the time whole-body prosthetics are viable, we'll already have the tech to copy minds to software, obviating the need for any organic element...
Lurker2358
2.3 / 5 (3) Nov 29, 2012
Not to deny the benefits of such techs wrt say prosthetics of course... i just think that by the time whole-body prosthetics are viable, we'll already have the tech to copy minds to software, obviating the need for any organic element...


While the Character "Walter Bishop" from the "Fringe" television series can probably do that with existing technology, it's unlikely you could copy a human brain to software for the foreseeable future. You need a simulator that precisely models the neural net of the brain in everyway, and you need mechanism for initializing that simulated net, and of course both the Secondary Memory (not so much of a problem) and primary memory (very big problem) to store that much information.

Humans probably have the equivalent of a couple terabytes worth of data and skills memorized, and in humans it's all interconnected in ways that existing software doesn't model very well at all.
Telekinetic
1 / 5 (2) Nov 29, 2012
"Humans probably have the equivalent of a couple terabytes worth of data and skills memorized, and in humans it's all interconnected in ways that existing software doesn't model very well at all."

Existing software, precisely. Wait till the quantum robot edges out the iPad on everyone's wish list. Then you can design your own friends.

MrVibrating
1.3 / 5 (4) Nov 29, 2012
@Lurker - simply in terms of capacity, Moore's law will surpass nature within decades, then there's the next revolution, as Tele says above... but the real groundwork is being laid in artificial neural nets and projects like the Connectome - perhaps, the function is all that's required, and the means are incidental...
Telekinetic
1 / 5 (2) Nov 29, 2012
@lite:
I see you're still one ball "lite".
socean
4 / 5 (1) Nov 30, 2012
The exciting part is crossing the wet-dry boundary so simply. Now projects like Connectome can incorporate wetware as well.

Hopefully, we will use the brain-computer interfaces to do something better than invent sexbots.

BTW, to recreate a mind you will have to recreate it physically. No model will be identical to the wetware reality. No two wetware devices can ever be identical, so no minds can be copied identically. At what point do you loose enough congruity to end up with something very different?

There is no way to predict what differences will arise from very small alterations because it is a non-linear, fractal, system. You might invent a roughly identical twin in software, but it will have its own personality, it will be a different entity.
socean
not rated yet Nov 30, 2012
What about eyes for people? Someone try this on a nerve cell.
extinct
1 / 5 (4) Nov 30, 2012
GMO's were only the beginning. keep messing with mother nature, and within a smaller-than-large number of generations you will get your ultimate psychopathic wish: human extinction
socean
5 / 5 (1) Nov 30, 2012
Why do techophobes read phys.org? Do they use the articles for scary stories around campfires?

"And then kids, the bionic boogie man cooked mother nature in a nuclear oven, frosted her with GMO, and gobbled her all up."
Osiris1
1 / 5 (2) Nov 30, 2012
Mmmmmm...self replicate. See real life remake of the movie: "The Evil Dead!"!
MrVibrating
1 / 5 (2) Nov 30, 2012
You might invent a roughly identical twin in software, but it will have its own personality, it will be a different entity.

Agreed, without a 1:1 rip of everything, they'll diverge. Still, if it has your memories and a sense of continuity of existence then "it" will still be "you" as far as it's concerned...

Proper rabbit hole innit..?
sstritt
1.8 / 5 (5) Dec 01, 2012
Roger Penrose has credibly demonstrated that consciousness is a non-computational phenomenon, and as such, cannot be achieved by any computational device.
nannasin28
not rated yet Dec 19, 2012
At what point do you loose enough congruity to end up with something very different?

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.