investigates Parkinson’s disease with an unusual tool: light. Han is a pioneer in the young field of , in which scientists reengineer nerve cells, or neurons, to respond to light, using molecules called opsins. Like ice cream, opsins come in many flavors—there’s rhodopsin in the human eye and halorhodopsin in bacteria, for instance—but they all share one key characteristic: they change shape when exposed to light.
By finding ways to implant opsins into neurons, Han, a College of Engineering assistant professor of biomedical engineering, has given researchers a simple tool to turn neurons on and off, and thereby study their function. The technique is now widely used to study brain activity, and it is leading to a better understanding of diseases and treatments.
In April 2014, Han traveled to Washington, D.C., where President Obama awarded her a , the US government’s highest honor for science and engineering professionals in the early stages of their independent research careers.
spoke recently with Han.
: Who came up with the idea of using light to turn neurons on and off?
Using light to control cells is not so new. In our retina there are all these rhodopsins that naturally are sensitive to light, but we can’t easily engineer the whole system into neurons. So the really novel part was sensitizing neurons to light so they’re easy to use.
So how did I get involved in this whole thing? I started my postdoc at Stanford in 2005, and that’s the same time that Ed Boyden, now an MIT associate professor, along with Karl Deisseroth, used this molecule called channelrhodopsin. They put it in neurons, and they were able to drive neural activities with the light. And the beauty of channelrhodopsin is that it’s a very small protein and it’s very easy to use.
Then Ed and I were thinking, since there’s a technology to excite neurons, can we also silence them? That led to the discovery of halorhodopsin, which allowed us just to do that. But it doesn’t do it really well. Who knows why? These are from bacteria, and you’re putting them in mammalian cells. That’s the complexity of biology.
So we said, let’s find a better one. We screened a whole bunch of proteins similar to halorhodopsins, and we found some other things that were similar, like proton pumps. We did not think they would work, but we thought, you know what? Let’s throw a couple in and see what happens. And we did that, and found that these proton pumps are way more effective in silencing neurons. And more importantly, from what we have tested, it’s safe for the neurons. It’s a powerful engineering tool—that we can excite or silence neurons now.
Are the tools getting closer to being used in patients?
Right now, my group is interested in how, in a disease like Parkinson’s, deep brain stimulation works. There are all these hypotheses about deep brain stimulation and its therapeutic effects. So the idea is that if you use light, then we can understand the mechanism and simultaneously see how the neurons respond and how they are contributing to Parkinson’s disease. These neurons in the Parkinsonian brain tend to oscillate or synchronize at a frequency of 20 hertz or so.
All the neurons in the brain, or just the Parkinsonian ones?
The Parkinsonian ones in a particular part of the brain.
All the neurons affected with Parkinson’s in a certain part of the brain are talking to each other at 20 hertz?
Not all of them. But somehow, more are talking to each other than normal.
But that’s what people find. If you think about Parkinson’s in particular, it’s a dopamine neuron loss. We are trying to figure out how this dopamine loss leads to the generation of these pathological oscillations in the brain. And then what’s the relationship of these oscillations to the symptoms?
Do other neurological diseases have different pathological oscillations?
That’s a great question. Can we establish some sort of oscillations as biomarkers of specific mental disorders? I think this is definitely a very interesting area. There’s certain evidence that a frequency around 40 hertz is associated with schizophrenia, but a coherent understanding would really help.
Why does Parkinson’s interest you?
I think for Parkinson’s, we are at a stage that things are converging. There’s a very good animal model for Parkinson’s, and the symptoms can be easily quantified, more easily than major depression or other types of mental disorders, like schizophrenia.
Well, we collaborate still. It’s hard not to collaborate, right? But you know, we have two small kids, so as soon as we start a conversation someone spills milk, and that conversation goes nowhere.
Do you tell your kids about your work? Are they interested?
Certainly there are scientific terms we use that our babysitter would not really understand. This morning my son was asking me how the Earth was generated. I told him the Earth was here when he was born, and I told him I was here, and so I started to explain it to him a little bit. It’s hard not to.
Where do you think diagnosing and treating neurological diseases will be 20 or 30 years from now? How will your work fit it?
A lot of parts can be replaced, but when it comes to the brain, we are not there yet. In Parkinson’s and Alzheimer’s, we know the neurons are dying. Is there some replacement we can do? If we have a biomarker, we can probably start to develop more human therapies. So that’s what I’m hoping: we’ll treat these disorders before we’re old enough to get them ourselves. So we need to hurry up.
See video (courtesy of WBUR).