Probe into memory

Around the world, 50 million people are currently suffering from dementia, a syndrome the symptoms of which include memory loss, decreased thinking speed, disorientation and other problems associated with impaired brain functioning. As the global population ages, the number of people living with the affliction is expected to triple by 2050.

In the US, Alzheimer’s, the most common form of dementia, now kills more people than prostate cancer and breast cancer combined. In the UK, the condition has overtaken heart disease as the country’s number one killer – costing the lives of more than 70,000 people a year. But until recently, advocacy organisations say research into dementia has been largely neglected, with expenditure on the syndrome overshadowed by spending on other conditions, such as heart disease and cancer.

“Compared with the number of people developing dementia, one every three seconds, the amount of money devoted to research is tiny,” said Paola Barbarino, the CEO of Alzheimer’s Disease International in the 2018 ‘World Alzheimer Report’. “For a start, there isn’t enough original research. The global ratio of publications on neurodegenerative disorders versus cancer is an astonishing 1:12. At the same time, not enough people are getting into research on dementia.”

According to the report, more than 100 drugs have been tested but just four authorised for use on dementia patients since 1998. And while some have helped manage symptoms of the devastating condition, none have provided the silver bullet.

“For these diseases, there is simply no cure,” says Dong Song, a research professor at the University of Southern California (USC)’s Viterbi School of Engineering. “There is no drug for the patient to take.”

The lack of an effective solution for treating different types of dementia, such as Alzheimer’s, has not dissuaded Dong from searching. For the past 20 years, the research professor has been attempting to build what is known in scientific circles as a hippocampal memory prosthesis, a brain-machine interface device capable of restoring or enhancing memory functions.

The hippocampus is a brain structure located in the medial temporal lobe, which plays a central role in the formation of new, longterm memories. When it is damaged by diseases like Alzheimer’s, people find themselves unable to form new memories. The idea of the hippocampal prosthesis is to bypass the point of injury and restore memory to patients.

In March, Dong was part of a team of scientists from USC and Wake Forest Baptist Medical Center that published a paper in the Journal of Neural Engineering demonstrating the “successful implementation” of a prosthetic system capable of boosting memory.

The participants of the study were epilepsy patients who had electrodes already implanted in their hippocampi to locate the origin of their seizures. They were asked to perform a simple computerised memory task in which an image was shown, followed by a blank screen, then a request to identify the initial image out of a selection of four or five options.

The researchers then analysed the recordings and synthesised an “MIMO-based code” for correct memory performance. This code was then played back to the participants while they performed the memory task again. The result: a staggering 37% improvement in recall.

“This is the first time scientists have been able to identify a patient’s own brain cell code or pattern for memory and, in essence, ‘write in’ that code in order to make existing memory work better, an important first step in potentially restoring memory loss,” explained the study’s lead author, Robert Hampson, professor of physiology, pharmacology and neurology at Wake Forest Baptist Medical Center, in a press release.

“We showed that we could tap into a patient’s own memory content, reinforce it and feed it back to the patient,” Hampson added. “Even when a person’s memory is impaired, it is possible to identify the neural firing patterns that indicate correct memory formation and separate them from the patterns that are incorrect. We can then feed in the correct patterns to assist the patient’s brain in accurately forming new memories, not as a replacement for innate memory function, but as a boost to it.

Studying specific neurons

In order to understand how memories are formed and how mathematical models can be developed to mimic the encoding and decoding processes of the hippocampus, Dong says researchers also require long-term, stable recordings from individual neurons in the brain structure.

This means “designing a device that enables us to bidirectionally interact with the brain, and building a microelectrode array to record and manipulate a large population of neurones”, he says.

Most of the current recording probes in use are made of thin, metal microwires with exposed tips. Arrays designed with multiple microwires have been produced to “anatomically match” hippocampal architecture, with longer and shorter wires targeting different parts of the brain structure. For higher-density recordings, microfabricated silicon-based neural probes have also been used. The shank of each probe can cater for hundreds of recording sites, providing information from multiple neurons along the length of the neural track.

But despite the relative success of silicon probes, Dong says their material composition is not ideal for reliable, long-term recording.

“There is a mismatch with the stiffness of the brain tissue,” he says. “This causes a lot of immune response, which means the quality of the recording will degrade in the long term.”

In January, Dong and a team from USC, including Ellis Meng and Theodore William Berger, published another paper in the Journal of Neural Engineering suggesting the use of parylene C, a flexible, biocompatible polymer, could solve this design challenge.

“The main advantage of using parylene is flexibility,” says Dong. “It is soft and can conform to the tissue, costing less immune response.”

“The information that we can get out is equivalent, but the damage is much less,” added Professor Ellis Meng, in a USC news release. “Polymers are gentler on the brain, and because of that, these devices get recordings of neuronal communication over long periods of time.”

To test the hypothesis, the team designed and fabricated a parylene C neural probe array with eight electrodes on eight individual probe shanks. The quality of the electrical recordings, taken from four rat hippocampi, was then tested against microwire electrode arrays. To avoid the long, thin probes from buckling upon insertion, the team also added a disposable brace made from polyethylene glycol.

“What we found is that using a parylene-based material is comparable to silicon or metal wire in terms of the recording quality but offers better performance in chronic recording because of its softness,” says Dong. “This means that we can record neural activity for a longer period of time compared with silicon-based arrays.”

According to Dong, parylene-based electrodes could have further medical applications and be used in other regions of the brain.

“There are many medical applications for this type of array,” Dong continues. “In addition to cortical areas and subcortical areas, by which I mean the hippocampus, we can record from different areas of the cortex. We are also working on using a parylene-based material to build cuff electrodes for recording peripheral nerves. These electrodes look like cuffs because they wrap around the peripheral nerve and record the neural activity from those nerves.”

If and when it comes to actually commercialising parylene-based microelectrode arrays, Dong says various options will hopefully be available.

“All universities have certain mechanisms for connecting researchers to possible investors,” he says. “One possibility is that the faculty itself can take the innovation, start its own company and get its own investment. Another option, if a big company is interested in the invention, is to license the technology and build medical products based on the invention.”

How long it will take for the hippocampal memory prosthesis to make a mark in front-line healthcare remains to be seen. Designing such a device and putting it into the hands of medical practitioners remains incredibly complex, as Dong knows better than anybody.

“It’s a challenging task,” he says. “You have to record from the brain region and, based on the brain signal, design a mathematical model to predict what the output signal should be. You then need that output signal to drive stimulation back to the brain, and by doing that, try to bypass the damaged brain region to restore, or even enhance, brain functions.”

According to Rob Malenka, a psychiatrist and neurologist at Stanford University, who was not part of the USC and Wake Forest Baptist Medical Center study, the road is long.

“This kind of approach is certainly worth pursuing with vigour, but I think it will still be decades before this kind of approach will ever be used routinely in large numbers of patient populations,” he told Wired.

Cautiously optimistic

But the results of the pilot study by USC and Wake Forest Baptist Medical Center, and progress on the parylenebased electrode, suggest the team is making strong progress. Even Hampson was shocked by what the team found in the memory test.

“We weren’t surprised to see improvement, because we’d had success in our preliminary animal studies. We were surprised by the amount of improvement,” he said in an interview with Wired. “We could tell as we were running the patients that they were performing better. But we didn’t appreciate how much better until we went back and analysed the results.”

For his part, Dong remains confident that the technology he has spent around two decades working on will have a massive impact on patients around the world in the not-toodistant future.

He concludes, “This whole idea – of using this kind of fine spatialtemporal resolution recording and stimulation in the brain in order to cure disease – I believe is the future for neuromodulation.”