Think of the mind like a library, Dheeraj Roy says, and a piece of information like a book.
Once the book is stowed away on a shelf, it is encoded and becomes a memory. In a healthy brain-library, the mind makes an index. But what happens if that reference gets muddled, or misplaced? “If you don’t know the index, there’s no way me and you could find one out of a million books,” Roy explains.
Roy, a PhD candidate at Massachusetts Institute of Technology (MIT), is describing a quandary that has baffled cognitive psychologists for decades: if there is no recall, is a memory gone? In the absence of the means to test the hypothesis in the lab, the assumption was affirmative. Ask a patient with early Alzheimer’s disease what they ate for breakfast, and their failure to remember toast versus porridge is taken as evidence that the memory never took root.
This is why Roy’s PhD thesis paper, published in Nature in March, caused such intrigue. By studying mice in the early stages of Alzheimer’s, Roy found that not only could the animals form new memories, once those memories had been “forgotten” they could be retrieved – given a little help. This has consequences for the 40 million people with Alzheimer’s worldwide. It offers a strategy for improving memory that could go beyond the modest benefit of available drugs. It also suggests that deep-brain stimulation, a neurosurgical therapy most commonly used to treat Parkinson’s disease, is of limited benefit to those with Alzheimer’s.
“People ask, how did you think of this? And it’s the opposite: we didn’t think of it,” Roy says. “If I’m very honest, my guess was that these Alzheimer’s animals would not remember. We thought like everyone else, and then when we couldn’t make sense of our results and we had to switch our thinking.”
Roy, who is 29, migrated to neuroscience late. “Coming from an Indian household, all you hear is ‘doctor or engineer’, and it’s a running joke but it’s actually true,” he says. He went down the bio-engineering route, only becoming interested in memory after a difficult trip back to India in 2011. Hugging his grandmother, Roy realised that she no longer knew who he was. His parents had hidden the gravitas of her condition. “If someone tells you on the phone that someone is ‘forgetting’, you don’t really understand the impact until you actually meet them and have to re-introduce yourself and tell them you are the younger grandchild, you’re the one that likes sweets.” Roy joined the Tonegawa neuroscience research lab at MIT in 2012.
His timing was fortuitous. That year, colleagues at the Tonegawa lab published landmark research in Nature in which, for the first time, scientists activated the cells in the hippocampi of mice that are crucial for the storage of memory. Until then, researchers had only been able to discern which regions of the brain serve particular functions by de-activating cells. They used a new, more precise tool called optogenetics, which can “fire” neurons deep in the brain using pulses of light. “I said, wait, can we use this for Alzheimer’s too?” Roy says.
If you place a healthy mouse in a box and administer a small electric shock to its feet, the next time it encounters the box it will freeze in fear. A seven-month-old mouse with early Alzheimer’s, however, will not remember the box the second time around. Rather than freezing, it will scamper about, exploring its “new” environment carefree.
Read more: Matt Eagles has had Parkinson's since he was 7. A brain implant gave him back his life
Roy spent a year trying to devise a means to manipulate the memory cells in Alzheimer’s mice. He was trying to develop a harmless virus that would tag memory cells of brain, called engram cells, with a gene – the “genetics” part of “optogenetics”. The gene, a protein called channelrhodopsin, forms pores in the membranes of neurons. When light shines on the pores they open, flooding the cells with positively charged ions that “fire” the neurons. This was no easy feat: “A year of troubleshooting,” says Roy. “If you want to know what that looks like, I was hitting my head on the bench when everything kept failing and going to the bar at night.”
Eventually, he concocted a virus that worked. But he soon found that the circuitry of the brains of older mice with more developed Alzheimer’s were too riddled with the sticky amyloid-protein plaques that form in the brains of those with disease. Seven-month-old mice, in contrast, did not yet have plaques but did have memory loss: put into the box, they did not remember the shock. By drilling a tiny hole in the mouse’s skull and inserting a fibre-optic cable, Roy was able to stimulate the engram cells with light pulses. This, he discovered, strengthens the synaptic connections between neurons, helping memories to re-form. Once they were fired-up, the mice would freeze. Not only did they remember the box, they could hold onto the memory for up to six days if given a three-hour light treatment.
“I thought that the results were both remarkable and exciting,” says Raymond Kelleher, assistant professor of neurology at Massachusetts General Hospital. “It’s really stimulated a lot of interest in a relatively neglected area, this phase of memory called retrieval or recall.” Alzheimer’s patients and their families often ask Kelleher if there is anything they can do to improve memory. Roy’s study, says Kelleher, provides a scientific foundation for the “cueing” technique, where a patient is fed partial cues to a seemingly forgotten memory.
The light-pulsing tool is, of course, far too invasive to be used on humans. Yet Roy’s findings may have a more immediate implication for those with Alzheimer’s. Deep-brain stimulation (DBS), where a large part of the brain is zapped using electrodes implanted under the skull, is being developed as a treatment for Alzheimer’s. Patients can recover some cognitive functions through DBS, but often only temporarily. When Roy applied the optogenetic method to every neuron in a mouse hippocampus, the animal showed no improved memory recall – it only worked when he targeted precise clusters of memory cells.
“I think what deep-brain is doing is broadly strengthening many inputs to the hippocampus,” Roy says. He puts it another way: on a cloudy day it can be, hard to find the Sun. Charging the entire hippocampus will make the sun brighter but it will also create more clouds. Roy’s next challenge is to figure out how to target and fire memory neurons without using an invasive fibre-optic cable. Or, borrowing from his metaphor, he is in the business of clearing out the clouds to sharpen the Sun.
This article was originally published by WIRED UK
