Abstract
Purpose :
The development of new innovative therapies implies to monitor activity in the whole visual brain system with high spatial and temporal resolutions. To increase the resolution obtained with fMRI, we investigated if the new functional imaging technique based on ultrasounds (US) can accurately map activity in the Visual Cortex (VC), Superior Colliculus (SC) and Lateral Geniculate Nucleus (LGN) in Long-Evans rats. As in fMRI, ultrasound imaging is based on Cerebral Blood Volume (CBV), a functional indicator correlated with the neuronal activity thanks to the neurovascular coupling.
Methods :
Long Evans rats aged from 7 to 9 weeks were anesthetized to generate a craniotomy. Animals were then placed in front of a screen to deliver visual stimuli while the ultrasound probe was approached 500μm above the brain. An echographic gel is inserted between the probe and the brain to ensure proper transmission of ultrasounds. Our visual stimulations consist in 5 repetitions of 30s stimulations followed by 30s of darkness. Different parameters were assessed such as flickering frequency, stimulus contrast. 3D maps of the visual brain system were reconstructed by shifting the US probe position (pitch of 500μm) and repeating the acquisition.
Results :
The cerebral blood volume increased in a correlated fashion with the visual stimulation reaching up to 47% increase in SC (SEM: 1.8%), 46% in LGN (SEM: 1.6%) and 18% in VC (SEM: 2.4%) with correlation coefficients r=0.7 in SC, LGN and r=0.6 in VC. These values are obtained with our optimal visual stimulation parameters: 100%contrast and a 3Hz flickering frequency. Using these parameters, we were able to reconstruct the 3D map of the whole brain visual system with a high temporal resolution (sampling frequency=1Hz), and spatial resolution (100μmX100μmX400μm). Finally, we were also able to generate a retinotopic map of the different structures.
Conclusions :
We demonstrated the value of ultrasound imaging to investigate the rodent visual system, even in deep structures like LGN and SC. Spatial and temporal resolutions are high enough to discriminate and accurately monitor these structures - and substructures - involved in the brain visual information processing according to the parameters of the visual stimulation.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.