Combined In Vivo Anatomical and Functional Tracing of Ventral Tegmental Area Glutamate Terminals in the Hippocampus

Summary

The current protocol demonstrates a simple method for tracing of ventral tegmental area (VTA) glutamate projections to the hippocampus. Photostimulation of VTA glutamate neurons was combined with CA1 recording to demonstrate how VTA glutamate terminals modulate CA1 putative pyramidal firing rate in vivo.

Abstract

Optogenetic modulation of neuron sub-populations in the brain has allowed researchers to dissect neural circuits in vivo and ex vivo. This provides a premise for determining the role of neuron types within a neural circuit, and their significance in information encoding relative to learning. Likewise, the method can be used to test the physiological significance of two or more connected brain regions in awake and anesthetized animals. The current study demonstrates how VTA glutamate neurons modulate the firing rate of putative pyramidal neurons in the CA1 (hippocampus) of anesthetized mice. This protocol employs adeno-associated virus (AAV)-dependent labeling of VTA glutamate neurons for the tracing of VTA presynaptic glutamate terminals in the layers of the hippocampus. Expression of light-controlled opsin (channelrhodopsin; hChR2) and fluorescence protein (eYFP) harbored by the AAV vector permitted anterograde tracing of VTA glutamate terminals, and photostimulation of VTA glutamate neuron cell bodies (in the VTA). High-impedance acute silicon electrodes were positioned in the CA1 to detect multi-unit and single-unit responses to VTA photostimulation in vivo. The results of this study demonstrate the layer-dependent distribution of presynaptic VTA glutamate terminals in the hippocampus (CA1, CA3, and DG). Also, the photostimulation of VTA glutamate neurons increased the firing and burst rate of putative CA1 pyramidal units in vivo.

Introduction

In the past decade, an array of genetic tools was developed to increase the specificity of neuron-type modulation, and the mapping of complex neural networks¹. Notably, neurotropic viruses with an inherent ability to infect and replicate in neuronal cells have been deployed to express or ablate specific proteins in neuron sub-types. When harboring fluorescence proteins or genetically encoded synaptic activity indicators, transfected AAV vectors label and delineate neural networks across brain regions^2,3. The choice of a promoter in the AAV construct directs the expression of the vector in neuron types with some level of specificity (promoter-dependent expression). However, through Cre-lox recombination, AAV constructs are deployed with greater specificity for neuron labeling^4,5,6,7. Of note, photoactivated microbial opsins and fluorescence proteins packaged in AAV vectors can be expressed in various neuron subtypes⁸, and are ideal for imaging, neuron-type circuit tracing, and photomodulation^9,10.

AAVs constructs stereotactically injected into a brain region (or nucleus) drives the expression of the reporter protein in the soma, dendrite, and axons terminals. Neural expression of AAV harboring a reporter gene (eYFP) facilitates the labeling of neuron cell bodies and anatomical tracing of projections to and from other brain regions^11,12,13,14. AAV-eYFP constructs carrying light-controlled opsin (e.g., hChR2), can be deployed as a tool for imaging^6,15 and stimulation-based physiological tracing of neural projections to target brain areas in vivo¹⁶. Depending on the AAV serotype, the direction of neuron labeling may be anterograde or retrograde^11,12. Previous studies have established that AAV5 travels anterogradely in neurons¹². Thus, photostimulation of cell bodies expressing hChR2 produces presynaptic effects elsewhere in the brain (target)¹⁷.

Here, AAV (serotype 5) with a CaMKIIα promoter was used to express eYFP (reporter) and hChR2 (opsin) in VTA glutamate neurons and axonal projections. Results from this study demonstrate the layer-dependent distribution of VTA-glutamate presynaptic terminals in the CA1, CA3, and DG hippocampal regions. Also, photostimulation of VTA glutamate neurons increased CA1 multi-unit and single-unit firing rates in vivo when compared with baseline values. This protocol utilizes affordable tools and commercially available software that can increase the quality of data obtained from neural circuit tracing experiments.

Protocol

All experimental and animal handling procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of the Louisiana State University School of Veterinary Medicine.

1. Experimental animal

1. Use 5–6 weeks old mice.
2. House 3–5 animals per cage under standard conditions of 12 h alternating light and dark cycle. Food and water should be provided ad libitum.

2. Craniotomy and animal preparation

NOTE: This section describes pre- and peri-operative procedures for mouse craniotomy. Use standard stereotactic apparatus and appropriate coordinates (Anteroposterior: AP, Mediolateral: ML, and Dorsoventral: DV). Refer to a mouse brain atlas to determine the coordinates for the brain region of interest.

1. Anesthetize mice by intraperitoneal ketamine (100 mg/kg)/xylazine (10 mg/kg) cocktail injection. Perform a toe pinch test to ensure the absence of sensation before the commencement of surgery.
   NOTE: Alternatively, Isoflurane anesthesia with an appropriate nose chamber can also be used for this step.
2. Gently fix the head of the animal on the stereotaxic apparatus.
   NOTE: It is important to handle animals with caution during this process. Also, check the breathing rate and other vitals before proceeding with surgery. Allow the animal to rest for ~7 min to reduce stress.
   1. Place a heating pad on the stereotaxic frame such that the body of the animal is lying on it. This will help maintain the body temperature and keep the animal warm though out the procedure. Retain the heating pad for postoperative care and recovery.
3. Use a clipper to remove hair over the scalp and clean the skin with an iodine solution.
   1. Apply topical lidocaine to block sensation on the scalp.
   2. Use a scalpel to make a midline scalp incision extending from the frontal to the occipital region.
   3. Clean the incision area with iodine solution, then expose the calvaria.
      NOTE: A small amount of 3% hydrogen peroxide solution can be applied to remove periosteum from the calvaria. This will increase the visibility of landmarks (bregma and lambda) and sutures.
   4. Using a mouse brain stereotactic atlas, determine AP and ML coordinates relative to the bregma.
   5. For VTA injection, position an ultrafine blunt-point needle syringe (e.g., Neuros syringe) at coordinates AP: -3.08 mm/ML: 0.5 mm relative to the bregma.
   6. Use a drilling tool to bore a 1 mm hole (craniotomy) in the skull at the marked AP/ML coordinate.
      NOTE: This step requires a significant level of caution. Minimal pressure should be applied while drilling to prevent the drill bit from crushing the brain tissue. In the current study, drill bit was 0.8 mm and drill speed was set to 15,000 rpm.
      CAUTION: If hydrogen peroxide was used in cleaning the incision or calvaria, ensure total removal of solution or allow complete dryness before drilling.

3. AAV cocktail injection

NOTE: This section describes the process for AAV injection into the VTA of adult C57BL/6 mice (23–27 g). The described method can be used for AAV injection into any brain region using standard stereotactic apparatus and appropriate coordinates. To demonstrate this protocol, eYFP and hChR2 were expressed in VTA glutamate neurons using AAV5-mediated transfection under a CaMKIIα promoter (Figure 1). Cre-lox recombination can also be used for this step.

1. Mount a syringe with ultrafine blunt-point needle (32 G; 5 µL capacity with 100 nL accuracy) on an injector. Fix the syringe holder on a micromanipulator.
   NOTE: For this procedure, a manual syringe holder, with 1.6 nL calibration, was used. An automated injector can also be used.
2. Fill the syringe with double distilled water to clean and test the flow of fluid.
3. Thaw aliquots of AAV (serotype 5) cocktail on ice. AAVs are best stored at -80 °C to maintain viral titer for good expression.
4. Fill the mounted syringe with 1,000 nL of AAV solution (10 mM in phosphate-buffered saline (PBS), pH 7.4). Dispense 10 nL of the AAV solution to confirm the flow of the liquid before lowering the needle into an injection site.
5. Use a micromanipulator to lower the needle stereotactically into the desired depth (DV coordinate). After lowering the needle to the desired depth, allow the syringe to remain in place for 10 min before injection.
   NOTE: For this procedure, the needle was lowered into the VTA at a depth (DV) of 4.5 mm from the pial surface of the brain. For other brain areas, use the appropriate dorsoventral coordinate (refer to the mouse brain atlas).
6. Inject 600–800 nL of AAV into the VTA. Injection volume may be adjusted depending on the size of the target site.
   1. Deliver the AAV solution at the rate of 60 nL/min (3 min interval).
   2. To express eYFP and hChR2 in VTA glutamate neurons and projections, inject AAV-CaMKIIα-ChR2-eYFP.
      NOTE: Cre-lox recombination method can also be used depending on the goal of the experiment.
   3. Allow the needle to remain in place for 15 min after AAV injection. This will reduce diffusion and backflow.
7. Withdraw the syringe needle, and then suture the incision to close the wound.
8. Administer antibiotics and analgesia as a part of post-operative care. Place the mouse on a warm padded platforms and monitor the animal until awake.
   NOTE: After 3 weeks of injection, AAV expression can be observed by fluorescence detection of reporter protein (eYFP) in mouse brain sections. Also, photostimulation experiments can be performed after 3 weeks (Figure 2).

4. Set up for in vivo neural recordings with optogenetics

NOTE: This section describes the steps for acute neural recording with the optogenetic tracing of a brain circuit (VTA glutamate neuron projection to the CA1). If necessary, check the system (amplifier and connections) for electrical noise and grounding issues before commencing this step. Performing this step in a Faraday cage can help eliminate electrical noise in the recording.

1. At 3 weeks after AAV injection, anesthetize mice by intraperitoneal urethane injection (0.2 mg/kg).
   CAUTION: Urethane is carcinogenic and must be handled cautiously while wearing protective gear. Also, administering urethane anesthesia at this concentration is non-survival.
2. Affix the head of the animal on a stereotactic frame as described previously (step 2.2, Figure 3A). Perform a craniotomy to expose the dura (Figure 3A). Use a drilling tool (bit size: 0.8 mm, speed: 15,000 rpm) to remove part of the parietal bone.
   1. The craniotomy should be about 3 mm x 4mm wide. Apply drops of artificial cerebrospinal fluid (aCSF) over this area to prevent dryness.
3. Under a dissection (digital) microscope, use a bent needle tip (27 G) to excise the exposed dura. Be careful not to pull apart the delicate pial covering and cortical tissues in this area.
4. Bore a hole in the occipital bone (bit size: 0.8 mm, speed: 10,000 rpm) to hold the ground screw (pan head Phillips screw; M0.6 x 2.0 mm).
5. Connect a stainless-steel ground wire.
   NOTE: Other types of wires can also be used (Figure 3b).
6. For combined photostimulation (VTA) and neural recording (CA1), use a multi-rail stereotaxic apparatus fitted with ultra-fine (10 µm or 1 µm) micromanipulators. Mount the optical fiber and recording neural probe on 10 μm-range and 1 μm-range micromanipulator, respectively.
   1. If necessary, use a head stage-probe adapter (Figure 3c).
      NOTE: In the current protocol, a 32-channel head stage was fitted with an adapter to hold a 4-channel recording electrode (Figure 3d). The stainless-steel ground wire was soldered to the ground connection port of the adapter.
7. At co-ordinates AP: -3.08 mm ML: 0.5 mm, lower the 400 μm diameter optic fiber into the VTA.
   1. Before lowering the optic fiber, connect the ferrule to a fiber optic cable (with stainless steel or ceramic ferrule) linked to a fiber-coupled LED source.
      NOTE: Make sure light intensity and focus are set as desired. The choice of the LED light source should correspond to the opsins being targeted. Here, a 470 nm (blue light) LED driver was used for hChR2 photoactivation (Figure 4a).
   2. To synchronize the light pulse with the neural recording, connect the LED driver and recording controller digital IN port to a transistor-transistor logic (TTL) pulser (Figure 4b). This can be achieved using a BNC splitter.
      NOTE: The use of a TTL pulser provides the flexibility of digitally setting the desired pulse train frequency and duration. In the current protocol, 10 ms light pulses were triggered at 50 Hz¹⁸. For TTL control of the LED driver, switch the driver to “trigger”. In this mode, the light intensity can be regulated by turning the “knob”. Stimulation frequency can be adjusted to accommodate experimental variables and could range from 1 to 100 Hz. For neural circuit tracing, a stimulation frequency of >20 Hz is recommended.
   3. Adjust the knob to determine the effective intensity that can generate a response without producing photoelectric artifacts. If necessary, reposition the optic fiber to eliminate artifacts¹⁹.
      NOTE: The LED driver used for the current protocol produces light power of ~21.8 mW.

5. Neural recording

1. Use an acute neural probe with a 15–50 µm thick shank, measuring at least 5 mm in length.
   NOTE: Recording from deep brain structures will require neural probes with a longer shank. In the current study, a probe with 20 mm shank length was used.
2. Connect the pre-amplifier head stage to the recording controller via a serial peripheral interface (SPI) cable. Check LEDs color lighting on the recording controller ports. Green and yellow LEDs indicate proper voltage on the connected amplifier board. Red LED indicates a working software-head stage control (Figure 5b).
3. Using a micro-manipulator, position the electrode contact sites in the pyramidal cell layer of the CA1 (AP: -1.94mm, ML: +1.0mm, DV: +1.1 to 1.2mm).
   NOTE: The optic fiber and electrode can be positioned in other desired brain regions for photostimulation and recording.

6. Amplifier and filter settings

1. Connect the recording controller amplifier system to a computer through a USB 3.0 port. Connect the head stage to the amplifier using the SPI cable. Launch the controller software. If not properly connected, the system will display a “device not found” message.
2. Click Run to view activity on the channels. Each waveform is plotted as the voltage (y-axis) versus time (x-axis). Depending on the region of interest, adjust the voltage and time scales to adjust the waveform display.
3. If only a few channels are active, disable unused channels to reduce file size on the storage disk.
4. Set the sampling rate and cut-off frequencies by editing bandwidth parameters on the acquisition software. For single-unit recording, set the upper and lower cutoff frequencies to 300 Hz and 5,000 Hz respectively. Change the amplifier sampling rate to 30 kS/s.
   NOTE: Choosing a high sampling rate will produce a larger file size.
5. Before recording, check the integrity of the electrode channels by measuring impedance at 1,000 Hz. This can be done by launching the function in the user interface (controller software). A working channel should have an impedance value that ranges from 0.1 to 5 MΩ.
6. If an audio output (speaker) is connected to ANALOG OUT port, adjust gain and silencer for optimum spike sound.
7. To display the TTL pulse train marker in the spike recording, enable the display for the DIGITAL IN marker. To record the pulse train time stamp, enable the DIGITAL IN channel display before the main experiment.
   NOTE: There is usually more than one “DIGITAL IN” channel (1 or 2). Ensure that the BNC cable from the TTL pulser is connected to the “DIGITAL IN” port that is selected for display in the recording controller software.
8. For real-time viewing of the spike waveform, open the spike scope window, and set threshold using the mouse click.
9. Inspect the noise level by monitoring the Root Mean Square (RMS in µV). For clean spike recording, it is preferred that the neural spike threshold is at least 5x RMS.
10. Once the setup is complete, select the file output format.
    NOTE: Here, CA1 neural spike was recorded in .rhd format. Other file formats (.rhs and .dat) can also be selected to match the file formats permissible by the analysis software.

7. In vivo optogenetic recording and settings

1. Open the pulse control software (TTL) which displays various adjustable pulse parameters. Select the appropriate COM port on the software. Set pulse parameters by adjusting the Train and Group settings.
   NOTE: The pulse condition should be determined by considering experimental variables and design.
2. On the amplifier controller software, click Run to detect neural spikes in the hippocampus or desired brain area. If necessary, adjust the electrode depth to detect viable and active neurons.
3. Once extracellular voltage activity is detected, observe baseline spikes for 15 min and monitor the vitals of the animal. Also, test the light pulse to detect responsive neurons within the CA1 or desired anatomical target region.
   NOTE: Check for photoelectric artifacts and eliminate by adjusting light intensity or optic fiber position.
4. Record the baseline activity for ~10 min before triggering the light pulse at the desired frequency (Figure 5a,b). This will allow for comparison of firing or burst rates with and without illumination.
   NOTE: In the current study, baseline activity was recorded for 10 min without illumination, followed by illumination at 50 Hz (Movie 1).

8. Analysis

1. Convert the .rhd file into Nex5 output format.
2. Process filename.Nex5 file in an offline sorter software to detect single unit raster trains and waveforms (Figure 5a–b).
   1. Select the desired channel from the drop-down window.
      NOTE: The continuously recorded spike data can be viewed on a separate window of the OFSS. The time scale can also be adjusted. If a tetrode is used, the OFSS can be set to sort the 4 channels as a tetrode.
   2. Set the voltage level for threshold crossing waveform extraction.
      CAUTION: Use a minimum of 5x RMS for proper spike detection.
   3. Detect waveforms, then perform spike sorting using valley-seeking (automatic) or K-means (semi-automatic) method. Where necessary, merge units that occupy similar regions of the 3D-Principal Component Analysis (PCA) space.
   4. Export sorted .Nex5 files for further analysis (Figure 5c–d).
      NOTE: Analysis outcomes for the interspike interval, firing rate, and burst rate can be exported into other analysis platforms.

9. Fluorescence detection of AAV expression

1. After the recording, euthanize the animal in an isoflurane chamber.
2. Perform transcardial perfusion with 10 mM PBS (~10 mL), followed by 4% phosphate-buffered paraformaldehyde (PB-PFA, ~10 mL).
3. Remove the whole brain intact and fix it in 4% PB-PFA for 48 h.
4. Transfer the fixed brain into 4% PB-PFA containing 30% sucrose for cryopreservation at 4 °C. After 72 h, section the brain in a cryostat and mount slices on a gelatin-coated slide.
   NOTE: In the current protocol, sections containing the closed part of the hippocampus (rostral) were selected to demonstrate AAV-labeled terminals in the DG, CA3, and CA1.

Results

Anterograde tracing

AAV expression was verified by immunofluorescence imaging of reporter protein (eYFP) in the VTA of C57BL/6 mice 21 days post-injection (Figure 2). Successful anterograde labeling of presynaptic VTA glutamate projections in the hippocampus was also verified by eYFP detection in the layers of the DG, CA3, and CA1 (Figure 6a–d; Movie 2 and 3).

Discussion

In the past decade, the design of AAV constructs has advanced significantly. As such, more neuron-specific promoters have been incorporated into an array of AAV serotypes for improved transfection specificity¹⁴. By combining genes for fluorescence proteins, transporters, receptors, and ion channels, libraries of AAV now exists for imaging, neuromodulation, and synaptic activity detection. In commercially available AAV-constructs, a combination of a genetically encoded fluorophore and ion channels ...

Materials

Name	Company	Catalog Number	Comments
3% Hydrogen peroxide	Fisher chemical	H312
AAV-CaMKIIα-ChR2-eGYP	Addgene	Plasmid #26969
BNC cable	Amazon
BNC Splitter	Amazon
Ceramic Split Mating Sleeve for Ø1.25mm Ferrules.	Thorlabs	ADAL1-5
Drill	Dremel	LR 39098
Gelatin coated slides	Fisher scientific	OBSLD01CS
Hamilton's syringe (Neuros)	WPI Inc.	06H
Head stage adapter	Neuronexus	Adpt-Q4-OM32
High impedance silicon probe	Neuronexus	Q1x1-tet-5mm-121-CQ4
INTAN 512ch Recording Controller	INTAN	RHD2000
Iodine solution	Dynarex	1425
Isoflurane	Piramal	NDC 66794-017-25
Ketamine	Spectrum	K1068
LED Driver	Thorlabs	LEDD1B
LED light source (470 nm)-blue light	Thorlabs	M470F3
Micromanipulator	Narishige	M0-203
Optic fiber	Thorlabs	CFMLC14L05
Pan head philips screw (M0.6 X 2mm)	Amazon	M0.6 X 2mm
Pre-amplifier headstage (32 Channel)	INTAN	C3314
Stereotaxic frame	Kopf	1530
TTL pulser	Prizmatix	4031
Urethane	Sigma	U2500
Xylazine	Alfa Aesar	J61430
Software	Company	Version
Graphpad Prism
Intan Recording Controller
Neuroexplorer
Plexon Offline Spike Sorter
ACSF Composition:
oxygenated ACSF with 95% Oxygen/5%CO2 constantly being bubbled through the ACSF (ACSF; in mM 125 NaCl, 25 NaHCO3, 3 KCl, 1.25 NaH2PO4, 1 MgCl2, 2 CaCl2 and 25 Glucose).