Miniature microscopes for large-scale imaging of neuronal activity in freely behaving rodents

Highlights

• We discuss the use of the miniature fluorescence microscopes in neuroscience.

• The recent development of the integrated microscope allows large-scale, longitudinal, and cell-type specific Ca²⁺ imaging in freely behaving mice.

• We propose that the integrated microscope enables new types of neuroscience studies.

Recording neuronal activity in behaving subjects has been instrumental in studying how information is represented and processed by the brain. Recent advances in optical imaging and bioengineering have converged to enable time-lapse, cell-type specific recordings of neuronal activities from large neuronal populations in deep-brain structures of freely behaving rodents. We will highlight these advancements, with an emphasis on miniaturized integrated microscopy for large-scale imaging in freely behaving mice. This technology potentially enables studies that were difficult to perform using previous generation imaging and current electrophysiological techniques. These studies include longitudinal and population-level analyses of neuronal representations associated with different types of naturalistic behaviors and cognitive or emotional processes.

Introduction

A major approach in investigating how neurons process information and bring about behavior has been to record neuronal activity in intact brains of awake and behaving animals, and correlate the recorded neuronal activities to specific behaviors, sensory inputs, and cognitive processes. Extracellular electrophysiological recordings have been the technique of choice for such studies, yielding over the last several decades numerous seminal discoveries, such as spatial representation by place cells and grid cells [1, 2]. With improved recording techniques, neuroscientists can now record from larger neuronal populations rather than from a few individual neurons. Dense recordings using multi-tetrode arrays or silicon probes enable the recordings of tens to hundreds of neurons simultaneously in freely behaving rodents [3, 4, 5, 6]. Such recordings allow investigators to adopt a population-level perspective to the data analysis [7, 8] and address questions that could not be addressed from the recordings of small numbers of individual cells. For example, in the hippocampus, dense electrophysiological recordings led to the discoveries of cell assemblies whose activity is synchronized over timescales of tens of milliseconds [9], and that sequential reactivations of multiple individual place cells follow previously traversed paths [10], or predict future paths to remembered goals [11]. Recordings from a large enough population were essential in making these discoveries.

However, many important dynamic aspects cannot be obtained by dense recordings as electrophysiological recordings are inherently technically limited. For instance, the spacing between electrodes in dense recordings limits the mapping of fine spatial/anatomical organization of neurons within a circuit and their relationship to the recorded coding dynamics. Electrophysiology only allows for subset differentiation based on waveform shapes, making it difficult to determine the differential contribution of different neuronal cell types within the sampled population to the observed dynamics. Chronic electrophysiological recordings are difficult to do on the same populations of individual cells for more than ∼2 days, making assessment of coding stability and plasticity over longer timescales difficult [12]. In addition, the recording of activities from cellular structures such as dendrites, spines and axons in behaving rodents cannot be performed using electrophysiology. In recent years, fluorescence microscopy techniques emerged as complementary neural recording activity approaches, facilitating the investigation of dynamic aspects that are difficult to investigate using electrophysiology. These approaches are based on different fluorescence imaging modalities, namely one-photon or two-photon microscopy, with the experiments typically done either under head fixation or freely behaving conditions.