Tailored viruses that make brain cells manufacture fluorophores could make two-photon microscopy techniques viable for studying primate brains. Tested in marmosets, the technique could, for the first time, allow scientists to study neural activity related to the complicated cognitive and social behaviors of the order of mammals that includes humans. Representative images of GCaMP6f fluorescent signals. Neurons indicated by yellow arrowheads were assumed to be the same between post-injection day 10 and post-injection day 118. Image depth was 200 μm from the cortical surface. Courtesy of Riken. "While many people use two-photon microscopy with mice, they cannot study certain complex behaviors shared by primates, including us," said Tetsuo Yamamori, a team leader at the Riken Brain Science Institute. "With our new method, we will be able to investigate how primate brains work at a level not previously possible." Two-photon microscopy is a technique used to image living cells by exciting fluorophores within them with IR lasers. Because neurons use calcium when they transmit signals, neuronal activity can be visualized by genetically attaching fluorophores to molecules that bind to calcium. These genetically encoded calcium indicators (GECIs) are introduced via adeno-associated viruses (AAVs). GECIs have been used successfully in small animals, such as mice, but have had limited success in primates because weak expression levels produced fluorescent signals that were too faint to be seen. A marmoset. "Simply adopting the same technique used in mice did not work at all because the fluorescent signal was too weak," said Riken research scientist Osamu Sadakane. "But, by improving the viral technique, we amplified the signal and succeeded in observing a large group of neurons in the marmoset neocortex." The researchers used two AAVs rather than one, as is typically done in other research animals. GECI expression was also tied to doxycycline, which could be added to the marmosets' drinking water. This allowed the team to express the fluorophore GCaMP6f in very high quantities, which dramatically boosted the fluorescence level. "Not only was fluorescence easily to see," Sadakane said, "our reversible gene expression system enabled us to image the same neurons over several months." Normally, extended expression of AAV transgenes can damage neurons, making long-term imaging studies difficult. But by removing doxycycline from the monkeys' drinking water, the team was able to pause fluorophore expression for more than a month and prevent neuron damage. They were then able to visualize the same neurons again after reintroducing doxycycline into the monkeys' drinking water. The method also allowed visualization at the subcellular level, with activity being detected in axons and individual dendrites and in cortical layers up to 400 μm below the surface of the brain. This allowed individuals neurons to be imaged in 3D, which is critical for investigating the organization of cortical circuits. "The next step is to improve the technique so it can be used with awake, behaving marmosets," said Yamamori. "Then we will be able to relate neural activity to behavior and see changes in individual neurons that occur through experience." The work was carried out as part of Japan's Brain Mapping by Innovative Neurotechnologies for Disease Studies (Brain/MINDS) initiative and published in Cell Reports (doi: http://dx.doi.org/10.1016/j.celrep.2015.10.050 [open access]).