“It’s interesting and novel food for thought,” said Jerrold Vitek at the University of Minnesota, Minneapolis, an expert in DBS who was not involved in the research. “It still remains to be determined what [the findings] mean in the big picture.”

Deep-brain stimulation has been calming the tremors of Parkinson’s patients for more than a decade, and has recently garnered interest as a treatment for other disorders as well (see ARF related news story and ARF series). Some scientists are now pursuing this approach for dementia, with clinical trials starting up in Nice, France; Rouen, France; and Cologne, Germany. Andres Lozano at Toronto Western Hospital, a coauthor on the Frankland paper, is recruiting for a Phase 2 trial. He became intrigued by the potential of DBS to enhance memory when he stimulated a patient’s brain near the columns of the fornix and unexpectedly triggered old memories (see ARF related news story). Part of the brain’s memory circuitry, the fornix is a bundle of fibers that carries signals from the hippocampus to other regions. Based on this finding, Lozano and colleagues ran a Phase 1 trial on six people with early AD, reporting that DBS of the fornix appeared safe, improved brain glucose metabolism, and showed some suggestions of cognitive benefit (see ARF related news story on Laxton et al., 2010).

First author Scellig Stone wanted to investigate the mechanisms behind the apparent memory enhancement. Previous studies in the field had shown that electrical activation of hippocampal inputs can increase neurogenesis in the dentate gyrus region (see, e.g., Bruel-Jungerman et al., 2006; Chun et al., 2006). For this reason, Stone chose to excite the entorhinal cortex (EC), the last stop on the way into the hippocampus. He delivered high-frequency stimulation for 60 minutes to one side of the EC and used the unstimulated side as an internal control. As expected, the procedure activated neurons in the dentate gyrus on the stimulated side, as seen by increased expression of the activity-dependent gene c-Fos one hour later.

To look for an effect on cell division, the authors used the proliferation marker BrdU. On the stimulated side of the brain, cell proliferation nearly doubled three to five days after DBS, then dropped back to normal. This pulse of extra cells survived and differentiated into neurons in the same proportions as newborn cells on the unstimulated side. After six weeks, the neurons had developed mature shapes, with dendritic arbors and numerous spines, and appeared to be integrated into brain circuitry. The new cells became activated (i.e., increased c-Fos expression) after mice learned a task, suggesting the neurons were functional. One hour of DBS also modestly improved the survival of dentate gyrus neurons that were between one and three weeks old at the time of stimulation. Neurons of this age have been shown to be especially sensitive to survival signals (see Zhao et al., 2008).

The authors then looked for behavioral effects. They bilaterally stimulated the entorhinal cortex of wild-type mice, and trained the animals in the Morris water maze one and a half weeks later, an age when the newborn neurons were still immature. Stimulated animals learned no better than controls. In contrast, when mice learned the water maze six and a half weeks after DBS—after the newborn neurons had matured and integrated into the brain circuitry—they appeared to have better spatial memory and use more effective search strategies than unstimulated animals. This suggested to the authors that the six-week-old neurons contributed to learning. To test for a causal link between neurogenesis and learning, the researchers inhibited neurogenesis at the time of stimulation by using the DNA-alkylating agent TMZ. Mice who got both TMZ and DBS learned no better than unstimulated controls at six and a half weeks.

The results do not imply that neurogenesis is the only mechanism, or even the most important mechanism, by which DBS might enhance memory, Stone told ARF. “DBS likely causes many different things to happen in the brain,” he noted. Nonetheless, knowing even one of the mechanisms is valuable, because it gives researchers something they can measure and optimize in order to improve treatments, Stone said. “If we have a way of titrating our treatment to produce the best neurogenesis response, that might be associated with the best clinical response, and that gives us a biologic target to reach for.”

In ongoing work, Stone, now at Toronto Western Hospital, is looking at the effects of DBS in AD model mice to see if the procedure improves learning when extensive pathology is present. He is also looking at what happens to mice who receive continuous, long-term DBS, which more closely models the treatment people get. Since one hour of DBS increased both the birth of new cells and the survival of young neurons, continuous stimulation might have cumulative beneficial effects, Stone speculated. In addition, Lozano’s group is imaging the brains of people who received DBS in the fornix to see if their hippocampi grew larger, which would suggest increased neurogenesis.

Many studies have shown that neurogenesis underlies the beneficial effects of exercise in mice (see, e.g., ARF related news story; ARF news story; ARF news story; and ARF news story). Some studies have reported that antidepressants boost neurogenesis, although others question this finding (see, e.g., Marlatt et al., 2010 and Hanson et al., 2011). Henriette van Praag at the National Institute on Aging, Bethesda, Maryland, who studies exercise and neurogenesis, told ARF that exercise produces a greater boost in new neurons than DBS did in this study, up to tripling new cell production. Van Praag praised the innovative approach of the Toronto group, noting that many previous neurogenesis studies have focused exclusively on what happens in the dentate gyrus, whereas this one broadens the perspective to look at neurogenesis in the context of other brain structures. The findings are promising, van Praag said, but she suggested more work should be done to nail down the causal connection between DBS-activated neurogenesis and learning. Orly Lazarov at the University of Illinois at Chicago recommended ablating neurogenesis, which can be done by radiation or genetic methods, as the next step in making this connection.

Another question is how this might translate to people. “I thought the paper was exciting, but we have to be careful about the implications for treating disease,” Vitek told ARF. Not only would the procedure need to work in AD brains, but the enhancement it brings has to be meaningful to people, he said. “Let’s say we can improve some memory in certain tasks. What does that mean for the patient’s quality of life in general?” The benefits would have to be weighed against the potential risks of the surgery, Vitek pointed out. DBS requires electrodes to be implanted in the brain, an invasive procedure that carries risk of infections or bleeding. A complication with AD patients is that many of them cannot give informed consent (see ARF series on this topic). There are also practical questions, such as, Which stimulation parameters and electrode locations provide optimal cognitive enhancement? Vitek recommended that scientists answer these questions in animal studies before moving into clinical trials.—Madolyn Bowman Rogers.

Reference:
Stone SS, Teixeira CM, Devito LM, Zaslavsky K, Josselyn SA, Lozano AM, Frankland PW. Stimulation of entorhinal cortex promotes adult neurogenesis and facilitates spatial memory. J Neurosci. 2011 Sep 21;31(38):13469-84. Abstract