Here, we present a procedure for performing large-scale Ca2+ imaging with cellular-resolution across multiple cortical layers in freely moving mice. Hundreds of active cells can be observed simultaneously using a miniature, head-mounted microscope coupled with an implanted prism probe.
In vivo circuit and cellular level functional imaging is a critical tool for understanding the brain in action. High resolution imaging of mouse cortical neurons with two-photon microscopy has provided unique insights into cortical structure, function and plasticity. However, these studies are limited to head fixed animals, greatly reducing the behavioral complexity available for study. In this paper, we describe a procedure for performing chronic fluorescence microscopy with cellular-resolution across multiple cortical layers in freely behaving mice. We used an integrated miniaturized fluorescence microscope paired with an implanted prism probe to simultaneously visualize and record the calcium dynamics of hundreds of neurons across multiple layers of the somatosensory cortex as the mouse engaged in a novel object exploration task, over several days. This technique can be adapted to other brain regions in different animal species for other behavioral paradigms.
The cortex is an essential player in many complex mental and behavioral functions, from attention, sensory perception and top-down cognitive control1,2,3 to motivational, reward, and addiction pathways4,5. Understanding the computational processes that underlie its function is an important goal to drive a better clinical understanding of many mental and behavioral disorders.
Many current theories of psychiatric disease center around the idea that cortical neural circuit dysfunction or discoordination may underlie cognitive and behavioral abnormalities that are the hallmarks of conditions such as schizophrenia6, autism7 or obsessive-compulsive disorder8. Thus, obtaining population level neural activity data from cortical circuits within the proper context of simultaneous behavioral information is of great importance, and ideally can be targeted to specific cell types for finer neural circuit dissection.
Miniaturized microscopes in conjunction with implantable gradient refractive index (GRIN) microlenses enable optical access to neuronal ensembles under freely moving conditions from a diversity of possible brain regions9,10,11,12,13, including the cortex14,15,16. Using a mobile microscopy system coupled with genetically encoded calcium indicators allows for consistent imaging of the same cellular population encompassing hundreds of neurons over days to weeks in many brain regions9, and can be genetically targeted to specific cell types using viral vector or transgenic techniques.
As the cortex is known to support different functions and connect to different brain regions depending on the location of the cells within the cortical lamina17,18,19, we are interested in obtaining simultaneous multi-layer neural activity in awake behaving subjects. Here we demonstrate how to image hundreds of fluorescently labeled neurons in freely behaving mice over days, using the miniaturized fluorescence microscope20 paired with an implanted prism probe, which offers a multi-layer view of the cortex (Figure 1).
The prism probe used here is composed of two separate GRIN lenses: a prism and a cylindrical relay lens (Figure 1). The light from the microscope excites the fluorescently labeled cells located along the imaging face of the prism probe, after being reflected off the hypotenuse of the prism portion of the probe. The emitted light from the cells also reflects off the hypotenuse of the prism, is collected through the objective of the microscope and reaches the sensor in the microscope. The prism probe used in this procedure is adapted for easy use with standard stereotaxic equipment.
The miniaturized fluorescence microscope20 detects action potential-evoked Ca2+ transients in neuronal populations with single cell resolution, after those cells have been specifically labeled with Ca2+-sensitive genetically encoded fluorescent indicators. In this protocol, we inject Ca2+ indicator encoded in a viral vector (AAV1.CaMKII.GCaMP6f.WPRE.SV40), implant a prism probe, install the microscope, then obtain multiple days of somatosensory (S1 hind limb) neural activity data from an animal exposed to novel object surfaces during free exploration (Figure 2).
Procedures involving animal subjects have been approved by the Institutional Animal Care and Use Committee (IACUC) at LifeSource Biomedical Services, NASA Ames Research Center, California.
1. Pre-operative Preparation
2. Virus Injection Surgery
3. Prism Probe Implant Surgery
4. Baseplate Attachment for Miniature Microscope Installation
5. Imaging Multiple Cortical Layers in a Freely Moving Mouse
6. Evaluating Large-Scale Ca2+ Imaging Data
The protocol outlined here describes an effective and efficient way to perform longitudinal multi-layer Ca2+ imaging from hundreds of cortical neurons in freely behaving mice using prism probes (Figure 1). Previous approaches towards multi-layer cortical imaging have primarily been restricted to head fixed animals 22,23,24,25,26,27. In order to acquire this level of data in a freely behaving context, a miniaturized microscope platform was used for behavioral flexibility; a genetically encoded calcium indicator (GCaMP6f) was used to target a specific cell population (CAMKII+ cells in the cortex); and a prism probe was chosen to provide a chronic, multi-layer field of view.
We demonstrated the workflow for preparing the animal for imaging. A viral vector encoding an appropriate calcium indicator was injected into the cortex (Figure 2, Step 1), before chronically implanting a prism probe to enable optical access to the labeled cells (Figure 2, Step 2). A baseplate that serves as a secure, temporary dock for positioning of the microscope during imaging sessions was then installed over the head of the animal (Figure 2, Step 3), enabling the visualization of cortical activity across multiple cell layers in an awake behaving experimental setup (Figure 2, Step 4).
To ensure that the desired cellular population was being targeted, a post- mortem coronal brain section from a representative mouse is shown in Figure 3 with the prism probe tract and field of view marked relative to the GCaMP6f labeled neurons in Layers 2/3 and 5 of the somatosensory cortex.
During awake behaving imaging with the system, the activity of the somatosensory cortical neurons was recorded when the mouse was exposed to three different environments- Open Field (Day 1), Object Familiarization (Day 2-4) and Novel Object (Day 5) (Figure 4). On Day 1 the mouse was placed in a behavioral arena devoid of any objects. On Day 2-4 the mouse was placed in the arena with the same two texturally different objects (a gel pad and a wood block). On Day 5, one of the objects was replaced with a novel object. The animal was imaged across 5 days for 20 min each day.
Following cell extraction using the Ca2+ image data analysis software, spatial filters corresponding to cell locations were overlaid on the mean fluorescent intensity projection of the microscope recording data (Figure 5). A white dashed line separates layers 2/3 and 5 cells. Corresponding Ca2+ traces from 5 cells from each of the layers show the firing pattern of the cells in two different behavioral contexts- Object Familiarization and Novel Object Exposure. Layer 2/3 cells were more active compared to layer 5 cells on the day when the mouse was exposed to a novel object. This is also evident from the raster plots that show the thresholded firing activity of all imaged cells on days 1, 4 and 5.
Figure 1: In Vivo Ca2+ Imaging Across Multiple Cortical Layers in Freely Moving Mice. (A) Prism probe specifications and depiction of imaging plane. The reflective coating on the inside of the hypotenuse allows for imaging 90° from the insertion plane of the prism probe. The lens cuff integrates with the lens holder, which streamlines the implantation procedure and allows for potential viewing of ambient tissue fluorescence during implantation (B) (i). Illustration of placement of prism probe craniotomy and knife incision relative to the virus injection site, and (ii) illustration of location of prism probe flat side relative to the knife incision and virus injection site. (C) Illustration of in-vivo Ca2+ imaging setup showing the light path for a small area within the full field of view through prism probe implanted in mouse cortex. (D) Example field of view during prism probe installation. Miniature microscope is attached to the lens holder, which holds the prism probe allowing for checking the virus expression during prism probe installation. (E) Integration of microscope with prism probe for multi-layer cortical imaging of GCaMP6f labeled S1 cells. F Example field of view during baseplate installation. Clear blood vessel pattern is visible at the time of baseplate install with some cells in raw image. More cells are clearly visible when DF/F is turned on in the acquisition software window. Please click here to view a larger version of this figure.
Figure 2: Schematic Showing the Timeline of Workflow Events for Prism Probe Implantation and Microscope Installation. Number of weeks is represented on the X-axis and the workflow steps of the procedures along the Y axis. (A) Graphic illustrating viral injection (AAV1.CaMKII.GCaMP6f.WPRE.SV40) along the same dorso-ventral axis, to label multiple layers of the mouse somatosensory cortex. (B) 2 weeks post virus injections, a prism probe is implanted at an axis that is parallel to the virus injection sites. (C) Approximately one week after the prism probe implantation, the animal is checked for expression with the microscope and a baseplate is mounted on the head if a population of cells is visible. (D) The animal is then ready for chronic imaging during relevant behavioral tasks (Mouse clip art modified after permission from- UW-Madison Biochemistry MediaLab). Please click here to view a larger version of this figure.
Figure 3: Postmortem Histological Validation of Prism Probe Location and GCaMP Expression. (A) Coronal section from a representative mouse brain showing the prism probe tract and with its imaging side facing the GCaMP6f expressing cells (AAV1.CaMKII.GCaMP6f expressed in neurons in layers 2/3 and 5). (B) Same coronal brain section following staining for DAPI. Scale bar = 250 µm (C) Zoomed in view of GCaMP6f expressing cells in somatosensory cortex. Scale bar = 100 µm. Please click here to view a larger version of this figure.
Figure 4: Mouse Activity During Habituation, Familiarization, and Novel Object Exposure Testing was Video-tracked Using Video Software. (A) On Day 1, the animal was placed in a behavioral arena devoid of any objects (Open Field). (B) On days 2-4, the same two texturally different objects (gel pad and wood block) were placed in the arena (Object Familiarization). (C) On Day 5, one of the objects was replaced with a novel object (wood block with sand paper) (Novel Object Exposure). Please click here to view a larger version of this figure.
Figure 5: Calcium Dynamics from Superficial and Deep Layers of Somatosensory Cortex of a Representative Mouse Imaged with the Microscope. (A) Merged image of neuronal spatial filters (green blobs) and mean fluorescence intensity projection of the microscope recording through prism probe field of view. Border between supragranular and infragranular layers indicated by a white dashed line. Scale bar = 100 µm. (B) Calcium traces from five example superficial and deep layer cells (filled blue and red cells in panel A), indicating units of standard deviation of fluorescence following principal and independent component analysis. Horizontal scale bar 50 s and vertical scale bar 10 S.D. (C) Raster plot of cells from superficial (layers 2/3) and deep layers (layer 5) shown over Open field, Object Familiarization and Novel object exploration. Scale bar = 100 s. Please click here to view a larger version of this figure.
Understanding neural circuit activity during awake behavior is a vital level of neuroscientific investigation needed to effectively dissect brain function in health and disease. The cortex is a particularly important region to study in the context of awake behavior, as it plays an important role in many vital sensory, cognitive and executive functions28,29.
The cortical column is thought to be the basic functional unit in the cortex and population-level activity of cortical cells is known to differ based on their physical location within the column. For example, excitatory neurons in layers 2/3 in the somatosensory cortex project primarily to other neocortical regions and modulate other cortical networks30, while cells in deeper layers project primarily to subcortical regions like the thalamus31. Recording the activity of hundreds of pre-specified cortical cells simultaneously and reliably over time across different lamina in freely behaving subjects would greatly advance our understanding of cortical information flow, allowing for finer functional dissection of cortical columns informed by real-time behavioral information and task-relevant time-scales.
Collecting this level of neural circuit data is made possible with the use of an efficient and streamlined miniaturized microscopy platform to conduct large-scale Ca2+ imaging in freely behaving subjects (or head-fixed subjects as desired). Used with genetically encoded calcium indicators to enable cell-type specific targeting, and imaging a multi-layer field of view provided by a chronically implanted prism probe, this protocol explored one case among many possible applications: observing laminar differences in somatosensory cortical processing when mice physically engaged with a novel object (Figure 5). This is the first procedural illustration of this kind of cell-type specific, in vivo approach to study multiple cortical layers in awake, freely behaving animals, and broadens the spectrum of experimental methods available to understand laminar structures in the active brain.
The periscopic field of view enabled by the prism probe in this technique can feasibly be applied to other brain structures when preservation of the tissue directly dorsal to a region of interest is desired; for example, CA3 imaging could be achieved without disruption of hippocampal function.
Prism probe based approach for imaging Ca2+ activity requires the physical insertion and permanent implantation of a microprism into the cortex, which equates to the creation of a cortical lesion where the lens probe is inserted. This may result in disruptions to the local neural circuitry, including the severing of apical dendrites and processes. This procedure will also cause an initial activation of glial cells in the region, though this is expected to be localized to the tissue about 150 µm from the prism face, and to subside after the brain has healed22. It is very important to consider if this technique will affect the normal circuit anatomy and/or behavior of the animals when planning experiments. Behavioral control groups should always be conducted to ensure that there are no significant alterations in baseline behaviors that could produce confounding experimental results.
Using this miniaturized, mobile Ca2+ imaging technique with neuropharmacological manipulation, various cognitive, social, motor or intrinsic behavioral paradigms, and combining it with other physiological metrics can deepen and enrich studies focused on understanding the functional roles of neural circuits in behavior and signal processing32. Suppression or activation of certain pathways modulated by drugs can influence associated behaviors, which can be easily studied using this technology33. Branching out into different cell types by modifying the targeting of the calcium indicator is another powerful and useful application, and enables many creative combinations of experimental tools to address diverse neural circuit questions.
The authors have nothing to disclose.
The authors would like to thank V. Jayaraman, D.S. Kim, L.L. Looger and K. Svoboda from the Genetically- Encoded Neuronal Indicator and Effector (GENIE) Project at Janelia Research Campus of the Howard Hughes Medical Institute for their generous donation of AAV1-GCaMP6f to University of Pennsylvania Vector Core. They would also like to thank the A. Olson and Neuroscience Microscopy Core at Stanford University supported by NIH NS069375 grant for their confocal microscopy services.
Neurostar Motorized Ultra Precise Small Animal Stereotaxic Instrument |
Kopf | Model 963SD | Surgery |
Stereoscope | Labomed | Prima DNT | Surgery and Imaging |
Mini Rectal Thermistor Probe (.062"/1.6mm diameter) – 1/4" Jack | FHC | 40-90-5D-02 | Surgery |
Heating Pad 5 X 12.5cm | FHC | 40-90-2-07 | Surgery |
DC Temperature Controller | FHC | 40-90-8D | Surgery |
Microsyringe Pump | World Precision Instruments | UMP3 model; serial 155788 F110 | Surgery |
NanoFil 10μL Syringe | World Precision Instruments | NANOFIL | Surgery |
35 G Beveled Tip Nanofil NDL 2PK | World Precision Instruments | NF35BV-2 | Surgery |
Omnidrill35, 115-230V | World Precision Instruments | 503598 | Surgery |
Burrs for Micro Drill | Fine Science Tools | 19007-05 | Surgery |
nVista | Inscopix | 100-001048 | Imaging |
Name | Company | Catalog Number | コメント |
AAV1.CaMKII.GCaMP6f.WPRE.SV40 | Penn Vector Core | AV-1-PV3435 | Surgery |
Ketoprofen | Victor Medical | 5487 | Surgery |
Carprofen | Victor Medical | 1699008 | Surgery |
Isoflurane | Victor Medical | 1001054 | Surgery |
Gelfoam (Patterson Veterinary Supply Inc Gelfoam Sponge 12cmx7mm) | Pfizer (Fisher Scientific) | NC9841478 | Surgery |
Dumont #5/45 forceps | Fine Science Tools | 11251-35 | Surgery |
Dumont #5 forceps | Fine Science Tools | 11251-30 | Surgery |
Dissecting knives | Fine Science Tools | 10055-12 | Surgery |
ProView Implant Kit | Inscopix | 100-000756 | Surgery and Imaging |
ProView Prism Probe 1.0mm-Dia. ~4.3mm Length | Inscopix | 100-000592 | Surgery and Imaging |
Kwik-Sil adhesive pack of 2 | World Precision Instruments | KWIK-SIL | Surgery |
Kwik-Cast Sealant | World Precision Instruments | KWIK-CAST | Surgery and Imaging |
Miniature Optical Mounting Post | Newport | M-TSP-3 | Imaging |
Microscope Baseplate | Inscopix | BPL-2 | Imaging |
Microscope Baseplate Cover | Inscopix | BPC-2 | Imaging |