Abstract
Purpose :
In vivo calcium imaging is an efficient and minimally invasive method to study retinal physiology, for example in animal models of blindness and following vision restoration therapy. However, the imaging light used to excite the indicator also activates the photoreceptors. This problem can be addressed by using longer wavelength light, ultimately in the infrared two-photon excitation regime. Here we evaluate the use of a red-shifted indicator (JRGECO1a) using single photon excitation. We also compared JRGECO1a responses to the established green indicator GCaMP6s.
Methods :
Intravitreal injections were made in 3 week old C57BL/6 mice with either AAV2.CAG.JRGECO1a or AAV9.Syn.GCaMP6s. Imaging was performed using an adaptive optics scanning light ophthalmoscope. A 561 nm laser was used to excite JRGECO1a and 488 nm for GCaMP6s. Light power for both sources was ~100 µW. A uniform square-wave stimulus was delivered using a 365 nm LED, which stimulates mouse short wavelength cones. Semi-automated image segmentation was performed to identify single cells. Stimulus driven responses for each cell were computed as the amplitude and phase of the stimulus frequency component (f1). Each cell was categorised as either ON or OFF based on its response to the increment or decrement of the stimulus. Responses to the excitation light sources were also quantified using a ΔF/F metric.
Results :
Bright expression of JRGECO1a and GCaMP6s was present by 5 weeks of age. Robust responses were recorded from JRGECO1a in 12 week old mice (n = 2) and GCaMP6s in 21 week old mice (n = 2). Stimulus response phase of ON and OFF cells for JREGCO1a and GCaMP6s was similar: for JREGCO1a, ON cells 71° ± 6° (mean ± SD, n = 62), and OFF cells 206° ± 20° (n = 139); for GCaMP6s, ON cells 56° ± 38° (n = 207), and OFF cells 179° ± 20° (n = 266). Stimulus response amplitude (f1/f0) for JRGECO1a 0.09 ± 0.03 (n = 201) was smaller than GCaMP6s 0.15 ± 0.08 (n = 473). We quantified the responses to the imaging light. Fewer cells showed significant responses (ΔF > 5.SD[F]) to the 561 nm imaging light, 36% [77/212] (JRGECO1a), compared to the 488 nm light, 74% [364/493] (GCaMP6s).
Conclusions :
JRGECO1a can be used for in vivo functional imaging of retinal neurons with similar fidelity as GCaMP6s while reducing responses to the imaging light. Future studies will use two-photon excitation of jRGECO1a to further reduce responses to the imaging light.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.