In Vivo Calcium Imaging of Dorsal Root Ganglia Neurons' Response to Somatic and Visceral Stimuli

Acupuncture, an integral part of Traditional Chinese medicine, has gained global recognition primarily for its effectiveness in pain management, including the alleviation of chronic visceral pain¹. Over the past decades, our knowledge of the central nervous mechanisms underlying acupuncture analgesia has undergone considerable growth^1,2. However, little attention has been paid to exploring the functional roles of dorsal root ganglia (DRG) neurons in inducing the analgesic effect of acupuncture in visceral nociception. Visceral nociception and acupuncture analgesic studies are potentially carried out on primary sensory neurons using electrophysiological techniques or other neural recording methods^3,4. Such research aids in comprehending the relationship between somatic and visceral input from specific target tissues or target organs, offering valuable insights into conditions related to acupuncture, visceral pain, autonomic nervous system regulation, and related medical conditions.

Being the first-order neurons in the somatosensory system, neurons in DRG are referred to as primary sensory neurons which have important roles in transducing information about the external environment as well as the internal state into electrical signals and transmitting signals to the central nervous system (CNS). Numerous studies have suggested that visceral nociception was dominantly relayed by sensory neurons whose cell bodies are in the DRG^5,6. Although numerous researches have elucidated the cellular and molecular mechanism of DRG neurons in acupuncture-induced analgesic effect on visceral pain^7,8, very little literature exists on its functional characteristics due to technical difficulties⁹. Several methods for recording neural activity in the DRG, such as peripheral fiber recording, single-cell electrophysiology recording, and in vivo calcium imaging, can be used to record the patterns and properties of the action potentials passed along axons¹⁰. Loosely patched glass electrode recording of the DRG has been one of the most widely used techniques to investigate the correlation between neuronal activities and different stimuli in vivo¹¹. However, traditional methods such as electrophysiological recording cannot efficiently examine sufficient cell numbers and distinct specific cellular subtypes to identify visceral-responsive neurons in vivo.

In addition to encoding peripheral sensation, DRG neurons play a significant role in the transmission of acupuncture signals to the central nervous system. Traditional electrophysiological recording has already been widely applied to explore the regulation of acupuncture on abnormal activities of DRG neurons induced by pathological pain¹¹. Appropriate segments of DRG need to be observed in relation to sensory innervation. Lumbar (L) 6 DRG was generally observed to investigate colon modulation⁴.

Recent advances in the development of optical and genetic methods make it possible to investigate the activity of large populations of genetically labeled neurons simultaneously¹². However, there is still a lack of detailed calcium imaging methods for monitoring neuronal activity in DRG under visceral and somatic stimulation. Hence, this protocol explains the procedures for in vivo observation of responsiveness of L6 DRG neurons to intracolonic and acupuncture stimulation. The method described here can also be used to detect characteristics of somatic and visceral sensory neurons.

The broad application and promotion of calcium imaging deliver a very effective and practical tool for acupuncture research. Considering the advantages of calcium imaging mentioned above, this method ought to have been widespread and applied in acupuncture research. However, the utilization of calcium imaging in acupuncture research is still relatively uncommon. The key reason for this limitation may be the difficulty of operational and recording procedures. The primary purpose of this article is to give an overview of some critical points in the conduct of calcium imaging recordings of L6 DRG neurons in mice. Most importantly, we hope to promote the advancement and development of acupuncture research by using this cutting-edge tool in vivo.

This animal protocol was approved by the Animal Care and Use Ethics Committees of the Institute of Acupuncture and Moxibustion, China Academy of Chinese Medical Sciences, and complied with the National Institutes of Health Guide for the Care and Use of Experimental Animals to ensure minimal animal use and discomfort. Pirt-cre mice were kindly donated by Dr. Xinzhong Dong from Johns Hopkins University (Baltimore, MD). Rosa26-loxP-STOP-loxP-GCaMP6s mice were obtained from a commercial source (see Table of Materials). The following procedure is optimized for either gender of mice weighing 20-35 g. The recommended age for mice undergoing the operation was between 8 and 14 weeks. Either genetically encoded calcium indicators (GECIs), such as Pirt-GCaMP6s mice, or normal C57 mice intrathecally injected with viral GECIs, were adopted to show the main populations of DRG neurons. Genetically engineered mice intrathecally injected with viral GECIs or crossed with Rosa26-GCaMP mice would show the specific population of DRG neurons. Laboratory gowns, gloves, and masks were worn throughout the protocol.

1. Pre-operative setup

1. Verify that the anesthesia ventilation device (see Table of Materials) has been properly installed.
2. Ensure that the imaging system (see Table of Materials) has been pre-opened and is functioning correctly.
   ​NOTE: The operation and imaging procedures were conducted sequentially, while the analysis process can be completed offline later.

2. Anesthetization

1. Intraperitoneally inject the mouse with a tribromoethanol solution (1.25%, 250 mg/kg, see Table of Materials). Assess the depth of anesthesia via toe pinch before proceeding.
2. Shave the hair in the lower back area and the ventral cervical part.
3. Place the mouse in a supine position on the heating pad (see Table of Materials).
4. Apply ointment to the animal's eyes to prevent them from drying out.
   ​NOTE: The depth of anesthesia was maintained as long as the mouse showed no response to a hindpaw pinch during all surgical and recording procedures. If the animal exhibited any signs of pain, an additional 0.2 mL of tribromoethanol solution was administered intraperitoneally, or 0.5% isoflurane was given intratracheally after the ventilator was applied to prevent muscle twitching and ensure the depth of anesthesia.

3. Tracheotomy

1. Make a midline incision of approximately 2 cm in the skin of the throat, using a surgical blade, from the sternum to the chin.
2. Extend the incision and gently push aside the sub-maxillary glands, ensuring full exposure of the digastric muscles located beneath the glands, running longitudinally along the middle of the trachea.
3. Once the seam of the paratracheal musculature in the middle of the muscles is seen, expose the trachea by gently separating the musculature and retracting it open.
4. Create a transverse incision on the trachea, large enough to accommodate the trachea cannula (outer diameter 1.3 mm, length 13 mm, with a Y-adapter, see Table of Materials) directed toward the lung (Figure 1A).
5. Cover the trachea with the paratracheal muscles and sub-maxillary glands, then suture the surrounding tissues to the skin layer by layer using 5-0 surgical sutures.

4. Exposure of lumbar vertebrae

1. Place the animal in a prone position on the heating pad.
2. Create a midline incision of approximately 7 cm in the lower back, extending from lumbar-5 to sacral-1 vertebra.
3. Carefully expose the spinous processes of the three vertebrae by moving aside the longissimus lumborum muscles.
4. Expose the successive three articular processes and mamillary processes, with the recording segment in the middle of the vertebrae, by moving aside the inner parts of the bilateral longissimus lumborum, dorsal cervical spine muscles, and dorsal cervical semispine muscles (Figure 1B).
   ​NOTE: The articular processes and mamillary processes of the three vertebrae were exposed to allow the spinal clamp to secure the animal without displacement during recording.

5. Exposure of dorsal root ganglion

1. Securely immobilize the animal with a toothed forceps (see Table of Materials) on the bilateral articular processes of the neighboring vertebrae. Then, use rongeur forceps to remove either the left or right articular and mamillary processes of the lumbar-6 vertebrae.
2. Carefully expose the left or right lumbar-6 DRG after clearing away the connective tissues above it, keeping the intact epineurium on the left or right side of the lumbar-6 vertebrae.
3. Ensure the active conditions of the exposed DRG by covering the exposed areas with small cotton balls soaked in saline.
   ​NOTE: Change the cotton balls until there is no subtle leakage of blood near the exposed DRG and the surrounding tissue, as this could affect the clear imaging of the DRG neurons. Replace the saline-wetted small cotton balls until there is no red or yellow leakage of tissue fluids.

6. Immobilization and ventilation

1. Place the animal with the exposed lumbar-6 DRG onto a heating pad on the stage of a customized spinal clamp^13,14 (Figure 1C).
2. Secure the animal by using two clamps on the spinal columns of the neighboring vertebrae to minimize movement during testing (Figure 1D).
3. Continue to change the saline-wetted cotton balls until the surgical area is sufficiently clean (Figure 1E).
4. Monitor ventilation and pulse oximetry throughout the experiment.
   ​NOTE: Immobilize the spinal columns using the customized spinal clamp and then adjust the pad's angle so that the microscope's objective is perpendicular to the DRG. Make necessary adjustments to align the objective and the DRG using angle adjusters.

7. Hardware and software setup for imaging

1. Insert the customized spinal clamp into the custom-designed microscope stage (Figure 1F,G).
2. Position a 10x/0.32 long-working distance air objective lens of a confocal microscope (see Table of Materials) over the exposed DRG for imaging (Figure 1G).
3. Capture time-lapse z-stacks of the intact L-6 DRG with a 25 µm/step and either 512 x 512 or 1024 x 1024 pixel resolution. The duration of one stack scan of the entire DRG varies from 5-15 s based on factors such as the perpendicularity of the DRG's position to the objective lens, the depth and speed of each scanning step, and the imaging resolution.
4. Conduct XYZT scanning of the DRG with 4-8 stacks, which include 1-3 baseline states, 2-3 acupuncture or visceral colorectal distension (CRD) stimuli, and 1-2 post-stimuli states.

8. Acupuncture/visceral stimuli and imaging

1. Apply brushing or pinching stimuli to the lower back, hindlimb, or hindpaw areas of the animal to assess the most sensitive receptive field of the imaged DRG neurons (Figure 1H).
2. Administer acupuncture stimuli manually or with a commercially available acupoint and nerve stimulator (Figure 1F) (see Table of Materials).
3. Induce visceral stimuli by using colorectal distension (CRD) with a self-made air pressure gauge (Figure 1H).
   ​NOTE: Test the most sensitive receptive field of the imaged DRG neurons by brushing or pinching the receptive areas of the lumbar-6 DRG. If these stimuli cause fluorescence changes during XYZT stack recording, select acupoints in these areas, such as the lower back acupoint BL25 (Dachangshu)¹⁵. Choose acupoints (as shown in Figure 2) that have the skin and muscle within the receptive field of lumbar-6 innervation.

9. Data analysis and processing

1. Assess evoked calcium responses using a confocal microscope by measuring the increase in green fluorescence of the neuronal GCaMP upon binding to intracellular calcium following somatic or visceral stimuli.
2. Manually trace visible cells and determine cell size and relative fluorescence intensity using the imaging software (Fiji, see Table of Materials). Define small, medium, and large neurons as having soma diameters of <20 µm, 20 to 30 µm, and >30 µm, respectively, based on previous studies^14,16.
3. Express fluorescence intensity as a ratio of the increase in maximum evoked fluorescence to the basal level (ΔF/F0) or a ratio of the maximum evoked fluorescence to the basal level (Ft/F0), where F0 represents the maximum fluorescence intensity measured during the baseline period.
4. Measure the percentage of activated neurons and the intensity of evoked calcium transients to assess changes in neuronal responsiveness following somatic or CRD stimuli.
5. Define activation of a cell as an increase in fluorescence intensity (ΔF) ≥30% of the baseline (F0).
   NOTE: Before tracing visible cells, adjust the brightness of the images to the maximum level to aid in identifying cell borders. Do not trace cells if there is any uncertainty. Review the recorded movie forward and backward to ensure no cells are overlooked.

Following the above protocol, the lumbar-6 DRG of a transgenic Pirt-GCaMP6s mouse was exposed, and visceral CRD or somatic acupuncture stimuli were applied to the colorectum or receptive field. This experiment aimed to observe the number and types of neurons elicited by different visceral CRD and somatic stimuli.

As shown in Figure 2A, most of the neurons in the lumbar-6 DRG do not exhibit GFP fluorescence under baseline conditions. This baseline fluorescence may be influenced by two factors: the expression level of GCaMP and the possible damage done to the DRG during the surgery. CRD stimuli resulted in a rapid, transient increase in GCaMP fluorescence, and the numbers and intensities of GFP increased, as can be seen in Figure 2B. Similar changes occurred in Figure 2C when BL25 (Dachangshu) electroacupuncture (EA) was applied¹⁵. BL25 is located 3 mm beside the median dorsal line of the lower 4^th lumbar spine, which is commonly used when there are lower intestinal disorders¹⁷. The selected and numbered cells are circled in Figure 2D using the imaging software.

As described in the above protocol, changes in fluorescence intensity above a threshold level of ≥30% F0 were considered as positive responses. Pseudo colors were added to show the merged image of neurons that responded to both CRD (red) and EA (green) in Figure 2E. There was one merged neuron that responded to both CRD and EA, as pointed out by the arrow in Figure 2E. The heat map and line chart in Figure 2F and Figure 2H represent the responses of all the neurons circled in the imaging software to CRD. Figure 2G displays a histogram chart of the different diameters of the responsive neurons to CRD. The heat map and line chart in Figure 2I and Figure 2K represent the responses of all the neurons circled in the imaging software to EA at BL25. Figure 2J displays a histogram chart of the different diameters of the responsive neurons to EA. In addition to analyzing the number and fluorescent intensities of the responsive neurons, it is possible to analyze different responses of neurons in the DRG of various sizes when different stimuli are applied, as shown in Figure 2G,J.

Once a single neuron is identified, it is also possible to analyze the responses of the same or adjacent cells to different stimuli. This allows for the examination of temporal and spatial interactions between different visceral and somatic stimuli at the DRG level.

Figure 1: Surgical procedure for lumbar-6 DRG. (A) The tracheotomy of the experimental animal. (B) Superior view displaying three exposed vertebrae, with the left side being rostral and the right side caudal. (C) The customized spinal clamp. Note the adjustable pad, allowing for precise alignment of the exposed DRG and the imaging objective, maintained at a perpendicular angle. The spinal clamps are controlled by universal ball units, facilitating fine-tuning of the recorded DRG and the objective. (D) Application of the spinal clamp and exposure of the DRG before imaging. (E) Enlarged images provide a closer look at the exposed left lumbar-6 DRG. (F) Depiction of the anesthesia ventilator and the custom stage used in this procedure. (G) The imaging process of the DRG under the objective lens. (H) A schematic representation of somatic and visceral stimulation during recording. Please click here to view a larger version of this figure.

Figure 2: Representative images and analysis of in vivo calcium imaging of lumbar-6 DRG neurons. (A-C) Representative images depict lumbar-6 DRG neurons during baseline, following CRD, and after BL25 EA stimuli. (D) Images illustrate labeled and numbered cells within a single lumbar-6 DRG after cell tracing using the imaging software. (E) Merged images display lumbar-6 DRG neuron ensembles responding to both CRD (in red) and EA (in green) stimuli. Scale bars = 100 µm. (F), (H), (I), (K) Heatmap and line chart display fluorescence intensity changes in the traced cells in response to CRD and EA stimuli. (G), (J) Depiction of the number of responsive cells of various sizes to CRD and EA stimuli. The error bars denote mean ± SEM. Please click here to view a larger version of this figure.

It is believed that acupuncture analgesia is modulated by integrative processes in the DRG, involving an interplay between afferent impulses from pain regions and impulses from acupoints. Here, we describe an elaborate procedure for L6 DRG imaging. The advantages of imaging are manifold, including remarkable spatial resolution, the possibility for high-efficiency imaging of large areas of neurons simultaneously, and the ability to monitor specific cellular subtypes and subcellular domains using gene-targeting probes¹⁸. In the spinal cord, multiple laminectomy techniques have been developed for in vivo imaging of spinal neuron activity^19,20. Although the DRG is in the vicinity of the spinal cord, in vivo imaging of DRG cells remains more challenging due to its location surrounded by connective tissue and muscles. A previous study has shown an approach to survey neural responses of L5 DRG under somatic stimuli²¹. Since L5 DRG neurons receive mostly somatic afferents instead of visceral afferents, questions remain about surgical procedures for surveying the activity of DRG neurons receiving somatic and visceral input.

The method described here allows for stable calcium imaging in genetically defined neurons of the DRG while applying acupuncture stimulation and visceral stimulation in intact mice. As is mentioned in most in vivo techniques, the achievable imaging stability strongly depends on the spinal stabilization device and proper anesthesia²². In this study, a custom-built spinal adaptor equipped with angle adjusters was used to conveniently angle the spine to obtain clear images of the DRG. A pair of Adson forceps was modified to serve as spinal vertebrae clampers attached to articulating arms, allowing enough space to lower an air lens over the exposed DRG. Alternative forceps can be applied as needed for different spinal cord segments to offer better stability. It is advisable to suspend the spinal cord and its imaging field slightly from the clamps, which also reduces respiratory displacement. Another important strategy to minimize fluctuations during imaging is the use of tracheal cannula anesthesia and muscle relaxants combined with mechanical ventilation.

This study can also be adapted for two-photon imaging, which may provide better cellular resolution in deeper tissues. Recently, Chen et al. developed an imaging technique that allows researchers to examine the activity of neurons in the DRG of awake mice over extended periods²³. However, it is not yet readily available to most laboratories due to the difficulty in surgical preparation. Furthermore, although the method described here is not suitable for imaging DRG over weeks in awake mice, it has advantages in drug application and dorsal root stimulation since the animal's skin can be sewn to a ring to hold a pool for perfusion of ACSF.

There are also some limitations to this protocol. Restricting the recording to lumbar dorsal root segments limits the applicability of this technique to various ganglion segments that innervate different organs throughout the body, such as for the study of thoracic organs by recording thoracic DRGs. Calcium imaging videos were analyzed using the imaging software. There are challenges in calcium imaging data analysis, such as cell detection and Ca²⁺ signal extraction. Custom-written scripts in MATLAB seem to be an effective tool, but they require the researcher to have a programming background.