In Vivo Imaging of Ca²⁺ Signaling in Astrocytes Using Two-Photon Laser Scanning Fluorescent Microscopy

Abstract

Astrocytes are the predominant nonneuronal cell type in the central nervous system. Although they are electrically nonexcitable, they have been found to play an active role in modulation of neuronal function and plasticity through Ca²⁺ excitability. Thus, Ca²⁺ signaling in astrocytes serves as a mediator of bidirectional interactions between neurons and astrocytes. Although astrocytic Ca²⁺ signaling has been extensively studied in cultured cells, the recent development of two-photon laser scanning fluorescent microscopy and astrocyte-specific dye labeling make it possible to study astrocytic Ca²⁺ signaling in live animals. Here we describe a detailed protocol for in vivo Ca²⁺ imaging of astrocytes in mice.

1. Introduction

Astrocytes are the predominant glial cell type in the central nervous system (CNS) (1, 2). They provide nutritional and structural support for neurons, and act as a K⁺ sink to maintain extracellular K⁺ homeostasis (3). In addition to these passive roles, it is also established that they can remove glutamate from the synaptic cleft through glutamate transporters, thus avoiding glutamate toxicity (4, 5). The discovery that astrocytes can mediate Ca²⁺ signaling suggests that they could play even more active roles in the CNS (6). Astrocytic Ca²⁺ signaling can be stimulated in a variety of ways including neuronal input, direct G-protein-coupled receptor (GPCR) activation, mechanical stimulations, photolysis of caged glutamate, and inositol 1,4,5-triphosphate (IP₃) (7). Spontaneous excitation of astrocytes has been observed in vivo using 2-photon microscopy (8, 9). Our own and other results show that under normal conditions, cortical astrocytes in adult mice exhibit low frequency of Ca²⁺ oscillations which are confined to the microdomains in the cellular processes (10, 11). Pharmacological or sensory stimulations can induce intercellular Ca²⁺ waves in astrocytes within a brain slice and in vivo (6, 10–18). Although few studies have been done on the properties of Ca²⁺ signaling in vivo under pathological situations, our studies and others demonstrate cell-wide Ca²⁺ signals and intercellular waves in astrocytes following status epilepticus, ischemia and in a mouse model of Alzheimer’s disease (10, 19, 20). Ca²⁺ signaling in astrocytes is now considered to be a primary form of cellular excitability that can be determined by fluorescent imaging using a Ca²⁺ indicator.

Astrocytes generally employ metabotropic receptors that are coupled with G_q/11 to activate phospholipase C (PLC) to liberate IP₃ that activates IP 3 receptor (IP 3R) to release Ca²⁺ from the internal store (1, 21, 22). Among the three types of IP₃R (IP₃R1-3), IP₃R2 is primarily expressed and is the predominant IP₃R in astrocytes in the rodent brain (23–25). IP₃R2 knock-out (IP₃R2 KO) mice do not exhibit GPCR agonist-evoked increase in astrocytic Ca²⁺, thus providing compelling evidence that IP₃R2 is a key mediator of astrocytic intracellular Ca²⁺ release (21). Astrocytes express a variety of GPCRs, e.g., glutamate, gamma-aminobutyric acid (GABA), adenosine-5′-triphosphate (ATP), serotonin, norepinephrine, and dopamine, which can all mediate astrocytic Ca²⁺ signaling through the PLC/IP₃ pathway by activation of their respective receptors, including P2Y receptors, metabotropic glutamate receptors (mGluRs), GABA_B receptors and dopamine receptors (26, 27). A number of studies in cultured astrocytes (28) as well as in acute brain slice preparations (12, 29, 30) have linked the increase in astrocytic Ca²⁺ to the release of chemical transmitters. Thus, Ca²⁺ signaling in astrocytes serves as a mediator of bidirectional interactions between neurons and astrocytes.

Although astrocytic Ca²⁺ signaling has been extensively studied in cultured cells, relatively few studies have been done in vivo. The recent development of two-photon laser scanning fluorescent microscopy and astrocyte-specific dye labeling make it possible to study the Ca²⁺ signaling and structure of astrocytes in live animals (8, 10, 11, 19, 31, 32). This chapter describes a detailed protocol for in vivo imaging of Ca²⁺ signaling in astrocytes.

2. Materials

2.1. Animals

Adult FVB/NJ or C57/6BJ mice (5–7 weeks of age) were purchased from The Jackson Laboratory. All procedures were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals, and were approved by the University Of Missouri Office Of Animal Care Quality Assurance.

2.2. Chemicals and Reagents

1. Urethane, at a working concentration of 200 mg/mL in artificial cerebral spinal fluid (ACSF). Dose: 1.75–2 mg/g body weight.

2. Artificial cerebral spinal fluid (ACSF): 120 mM NaCl, 3.1 mM KCl, 2 mM CaCl₂, 1.3 mM MgCl₂, 10 mM glucose, and 10 mM Hepes (pH 7.4).

3. Fluorescent dyes: Sulforhodamine 101 (SR101), Calcium indicator fluo-4 AM, and Dextran-Rhodamine (all from Invitrogen).

4. Low melting point agarose.

5. Pluronic F-127 (Sigma).

6. Dexamethasone sodium phosphate (APP Pharmaceuticals LLC). Working solution 4 mg/mL. Dose: 20 μL per mouse.

7. Buprenophine (Henryschein), working solution: 2 mg/mL in saline. Dose: 0.5 mg/kg body weight.

8. Vaseline or eye ointment.

2.3. Surgical Tools

1. High-speed micro drill (Fine Science Tools).

2. Trephine with side opening for clearing of material. 2.3 mm shaft diameter, 44 mm overall length (Fine Science Tools).

3. Self-regulating heating pad.

4. Dumont forceps with fine tip.

5. Straight stainless steel 12 cm surgical scissors.

6. Cyanoacrylate glue (3M Vetbond Adhesive, World Precision Instruments).

7. Glass coverslips, 0.13–0.17 mm thick.

8. Custom-made metal frame (see Fig. 1) to which brain is attached (with glue) and fixed on microscope stage.

9. Dissecting stereomicroscope.

Fig. 1.

The custom-made metal frame. The mouse brain is attached to the frame with cyanoacrylate glue. The depth of chamber for retaining ACSF solution is 1 mm. The thickness of the plate is 1.5mm.

2.4. Two-Photon Microscope

Previously, two-photon microscopes usually were custom made in the laboratory. They are now commercially available from several companies. Usually, the upright fluorescence microscope is used for in vivo imaging. Two photomultiplier tubes (PMT) are generally installed for detecting green and red fluorescence, achieved by using a 575LP nm dichroic mirror. 40×/60× long working distance water-immersion objectives are suitable for in vivo imaging. We use a 60×, NA0.9 Olympus water-immersion objective for astrocytic Ca²⁺ imaging.

3. Methods

3.1. Cranial-Window Surgery

1. To prevent or reduce brain edema, dexamethasone is usually injected 30 min prior to surgery.

2. Anesthetize mouse with an intraperitoneal (IP) injection of urethane (1.5–2.0 mg/g body weight) dissolved in ACSF.

3. After mouse reaches a surgical level of anesthesia, shave the fur over the scalp.

4. Using scissors, make an incision in the midline of the scalp and remove a flap of skin (see Note 1).

5. Using a high-speed drill over the somatosensory cortex at coordinates −0.8 mm from bregma and 2.0 mm lateral to the midline, perform a circular craniotomy 2.0 mm in diameter (see Fig. 2b) (see Note 2).

6. Attach the custom-made metal frame (see Fig. 1) to the skull with cyanoacrylate glue, then carefully remove the dura with fine forceps under a dissecting stereomicroscope (see Note 3).

7. Add a drop of ACSF solution over the cranial window (see Note 4).

Fig. 2.

Schematic diagram of in vivo 2-P imaging. (a) Diagram of the optical pathway of the Ultima dual scanning 2-P microscope. Two Ti:Sapphire laser sources: one can be tuned to 720 nm for photolysis and the other to 820 nm to excite Fluo-4 for calcium imaging. A scan control system allows performing 2-P imaging and photolysis simultaneously. (b) In vivo 2-P fluorescent imaging. Left : The head of mouse was attached to a metal plate beneath objective for 2-P imaging. Right: A craniotomy was performed on the mouse cortex, where fluorescent dyes can be loaded and imaging can be performed.

3.2. Fluorescent Dye Labeling of Astrocytes In Vivo

3.2.1. Fluo-4 Labeling of Astrocytes In Vivo

To load the Ca²⁺ indicator fluo-4 into astrocytes, 50 μg acetoxymethyl (AM) ester of fluo-4 (i.e., fluo-4 AM) is dissolved in 5 μL 20% DMSO solution of Pluronic F-1287 (e.g., 0.2 mg of Pluronic F-127 in 1 mL DMSO) to obtain a 10 μg/μL stock solution. This stock solution (2.5 μL) is mixed with 40 μL ACSF and following the craniotomy, is applied to the dura-free cortical surface for 1 h (see Note 5). The residue dye is then washed away with ACSF. This procedure leads to the selective labeling of astrocytes with fluo-4 (see Fig. 3) (8–10, 19). After dye labeling, a glass coverslip is glued over the cranial window on the metal frame, and the gap between glass and cranial window is filled with 2% agarose premelted in ACSF solution. This greatly reduces movement artifacts resulting from respiration and heartbeat. The mouse is then transferred to the stage of a two-photon microscope for in vivo imaging.

Fig. 3.

Selective labeling of astrocytes with calcium indicator fluo-4 using surface incubation method. Astrocytes were labeled with the astrocyte-specifi c dye SR101 (left) and fluo-4 (middle). Right panel shows the merged image. The image is modifi ed from Ding et al. (19), and is used with permission of Glia.

3.2.2. SR101 Labeling of Astrocytes In Vivo

Selectivity of labeling of astrocytes can be confirmed using sulforhodamine101 (SR101). In this procedure, 100 μL of 100 μM SR101 dissolved in ACSF is applied on the cortical surface in craniotomy for 1–5 min and then is washed away with ASCF. After 40–60 min, astrocytes are selectively labeled with SR101. Using two wide-field detectors, co-labeling of astrocytes with fluo-4 and SR101 can be confirmed (see Fig. 3) (see Note 6).

3.3. In Vivo Ca²⁺ Imaging in Astrocytes

After dye labeling, the mouse is transferred to the stage of microscope for imaging. The cranial window is covered by ACSF for objective immersion. Time-lapse imaging is performed to monitor Ca²⁺ oscillation in astrocytes. Astrocytes usually exhibit low frequency Ca²⁺ oscillations in anesthetized mice under normal conditions (10, 19, 33) (see Note 7). ATP can trigger robust Ca²⁺ oscillation and waves. Iontophoretic application of ATP induces a synchronous Ca²⁺ elevation in a number of astrocytes (34), while in the continuous presence of ATP, astrocytes exhibit synchronous and regenerative Ca²⁺ oscillations as waves (10, 19). For ATP stimulation, we apply 0.5 mM ATP in ACSF to the cortex. The cranial window is then filled with 2% agarose containing 0.5 mM ATP. ACSF containing the same concentration is applied on the surface of the solidified agarose before imaging. Imaging is usually performed on astrocytes 80–100 μm below the cortical surface within 15–60 min after ATP administration. Multiple time-lapse imaging is performed to monitor Ca²⁺ signals for a period of 7.5 min, with acquisition rates of one image in every 2 s. For each mouse, 15–25 astrocytes in 4–5 fields are imaged and all astrocytes imaged are used for analysis. Throughout the experiment (about 3–4 h from the beginning of surgery to the end of imaging), the mouse is maintained at 37°C using a heating pad and at a surgical level of anesthesia. If the preparation is good, one should see bright astrocytes labeled with fluo-4 and repetitive and synchronous Ca²⁺ oscillations in astrocytes can be observed (see Fig. 4).

Fig. 4.

In vivo imaging of Ca²⁺ signaling in astrocytes induced by ATP stimulation. (a) Representative images showing the fluo-4 fluorescence changes in response to 0.5 mM ATP. (b) Time courses of fluo-4 fluorescence (ΔF/Fo)invivo from somata of individual astrocytes indicated by the numbers in (a) where t = 20 s, in the presence of 0.5 mM ATP. The boxed region is corresponding to the images in (a). Notice that Ca²⁺ signals are highly synchronized among the astrocytes.

Some experiments require the relocation of individual astrocytes to determine whether the addition of agonists and antagonists affects astrocytic Ca²⁺ oscillations. Because of the three-dimensional nature of the cortex, it is difficult to identify and relocate the same astrocytes previously imaged in the other fields or at different depths. However, this relocation can be achieved using vasculature as landmarks. To label vasculature, we inject Dextran-Rhodamine (200 μL of 20 mg/mL solution), which highlights blood plasma, into the tail vein (see Fig. 5a, b). This approach can relocate the same astrocytes so that we are able to image individual astrocytes in different brain regions before and after pharmacological agents are administered.

Fig. 5.

Vasculature as landmarks to relocate the same astrocytes. (a) 3D construction with maximum projection of vasculature loaded with Dextran-Rhodamine. (b)Acolor-combined image from single frame images of vasculature (red) and astrocytes (green) labeled with fluo-4 AM showing the relative positions of vasculature and astrocytes. (c, d) shows the fluo-4 fluorescent images of the same astrocytes before (c) and after (d) they were labeled with SR101. With vasculature as landmarks, it is easy to identify the same astrocytes even after the animal is reattached to the imaging platform after drug administration.

3.4. Analysis of Astrocytic Ca²⁺ Signal

The fluorescent signals can be quantified by measuring the mean pixel intensities of the cell body of each astrocyte using MetaMorph software (Universal Imaging, CA). Ca²⁺ changes are expressed as ΔF/Fo values vs. time, where Fo is the background subtracted baseline fluorescence. To calculate the magnitude of Ca²⁺ signals without subjective selection of threshold values, we integrate the ΔF/Fo signal over the imaging period using Origin software (OriginLab Corporation, MA). The resulting value is expressed as ΔF/Fo s. Data collected from multiple cells from each individual mouse are averaged and the averaged value of these cells is used as a single value for that mouse. The summary data should be the average value from four to five mice.