Development of fast multi-photon microscope for kHz-imaging of in vivo neuronal network activity

双光子显微镜（TPM）可以对大脑进行高分辨率的三维成像。TPM已经广泛应用于阿尔茨海默氏症和痴呆症等退行性疾病的研究。然而，为了研究整个神经网络的神经活动和疾病相关的修饰，必须以高帧率获得图像。突触信息传输发生在毫秒级和亚毫秒级的时间尺度上，因此需要千赫的帧率。加州大学洛杉矶分校（UCLA）的Bahram Jalali教授小组已经用单光子显微镜（FIRE显微镜）实现了这一目标。这项DFG研究计划的目的是开发一个快速TPM系统，从而在高分辨率下实现更高的组织成像深度。通过将快速激光与时间编码成像技术相结合，可以在256x256像素处达到100 MHz的像素率和1.5 kHz的帧率。为了实现这一目标，信号将使用先进的调制技术与高灵敏度光电倍增管（PMT）的双光子信号检测相匹配。pmt具有亚秒级的时间分辨率，因此也适用于并行荧光寿命成像（FLIM）。这也将导致所提出的系统的薄膜成像速度的急剧增加。总的来说，与传统的TPM系统相比，设想的多模态显微镜将产生超过10倍的速度增强，因此适合于在体内神经网络成像中应用轴突潜在尖峰事件固有的时间分辨率。

Two-Photon microscopy (TPM) enables high-resolution, 3-D imaging of the brain. TPM is already widely used for the study of degenerative diseases like Alzheimer´s and dementia. However, in order to study neural activity and disease-related modifications of whole neural networks, the images must be obtained at high frame rates. The synaptic information transport occurs at the millisecond and sub-millisecond timescale, so kHz frame-rates are desired. This was already achieved with one-photon microscopy (FIRE microscope) in the group of Prof. Bahram Jalali at the University of California Los Angeles (UCLA). The aim of this DFG research proposal is the development of a fast TPM system, thus enabling much higher imaging depth in tissue at high resolution. By employing fast lasers in combination with a time-encoded imaging technique, pixel rates of 100 MHz and frame rates of 1.5 kHz at 256x256 pixels shall be reached. In order to accomplish this, the signals will be generated using advanced modulation techniques paired with the detection of the two-photon signals by highly sensitive photomultiplier tubes (PMT). PMTs have time-resolutions of sub-ns and are thus also suited for fluorescence lifetime imaging (FLIM) in parallel. This would also lead to a dramatic increase of FLIM imaging speeds by the proposed system. Over all, the envisaged multi-modal microscope would yield a speed enhancement of more than a factor of 10 as compared to conventional TPM systems and is thus suited for application in imaging of neural networks in vivo at time-resolutions intrinsic to axon potential spiking events.