Pathophysiological changes in the cardiac autonomic nervous system, especially in its sympathetic branch, contribute to the onset and maintenance of ventricular arrhythmias. In the present protocol, we show how to characterize murine stellate ganglia to improve the understanding of the underlying molecular and cellular processes.
The autonomic nervous system is a substantial driver of cardiac electrophysiology. Especially the role of its sympathetic branch is an ongoing matter of investigation in the pathophysiology of ventricular arrhythmias (VA). Neurons in the stellate ganglia (SG) – bilateral star-shaped structures of the sympathetic chain – are an important component of the sympathetic infrastructure. The SG are a recognized target for treatment via cardiac sympathetic denervation in patients with therapy-refractory VA. While neuronal remodeling and glial activation in the SG have been described in patients with VA, the underlying cellular and molecular processes that potentially precede the onset of arrhythmia are only insufficiently understood and should be elucidated to improve autonomic modulation. Mouse models allow us to study sympathetic neuronal remodeling, but identification of the murine SG is challenging for the inexperienced investigator. Thus, in-depth cellular and molecular biological studies of the murine SG are lacking for many common cardiac diseases. Here, we describe a basic repertoire for dissecting and studying the SG in adult mice for analyses at RNA level (RNA isolation for gene expression analyses, in situ hybridization), protein level (immunofluorescent whole mount staining), and cellular level (basic morphology, cell size measurement). We present potential solutions to overcome challenges in the preparation technique, and how to improve staining via quenching of autofluorescence. This allows for the visualization of neurons as well as glial cells via established markers in order to determine cell composition and remodeling processes. The methods presented here allow characterizing the SG to gain further information on autonomic dysfunction in mice prone to VA and can be complemented by additional techniques investigating neuronal and glial components of the autonomic nervous system in the heart.
The cardiac autonomic nervous system is a tightly regulated equilibrium of sympathetic, parasympathetic, and sensory components that allows the heart to adapt to environmental changes with the appropriate physiological response1,2. Disturbances in this equilibrium, for example, an increase of sympathetic activity, have been established as a key driver for the onset as well as maintenance of ventricular arrhythmias (VA)3,4. Therefore, autonomic modulation, achieved via pharmacological reduction of sympathetic activity with beta-blockers, has been a cornerstone in the treatment of patients with VA for decades5,6. But despite pharmacological and catheter-based interventions, a relevant number of patients still suffers from recurrent VA7.
Sympathetic input to the heart is mostly mediated via neuronal cell bodies in the stellate ganglia (SG), bilateral star-shaped structures of the sympathetic chain, which relay information via numerous intrathoracic nerves from the brainstem to the heart8,9,10. Nerve sprouting from the SG after injury is associated with VA and sudden cardiac death11,12, emphasizing the SG as a target for autonomic modulation13,14. A reduction of sympathetic input to the heart can be achieved temporarily via percutaneous injection of local anesthetics or permanently by partial removal of the SG via video-assisted thoracoscopy15,16. Cardiac sympathetic denervation presents an option for patients with therapy-refractory VA with promising results14,16,17. We have learned from explanted SG of these patients that neuronal and neurochemical remodeling, neuro-inflammation and glial activation are hallmarks of sympathetic remodeling that might contribute or aggravate autonomic dysfunction18,19. Still, the underlying cellular and molecular processes in these neurons remain obscure to date, for example, the role of neuronal transdifferentiation into a cholinergic phenotype20,21. Experimental studies present novel approaches to treat VA, for example, the reduction of sympathetic nerve activity via optogenetics22, but in-depth characterization of the SG is still lacking in many cardiac pathologies that go in hand with VA. Mouse models mimicking these pathologies allow to study neuronal remodeling that potentially precedes the onset of arrhythmias12,23. These can be completed by further morphological and functional analyses for autonomic characterization of the heart and the nervous system. In the present protocol, we provide a basic repertoire of methods allowing to dissect and characterize the murine SG to improve the understanding of VA.
All procedures involving animals were approved by the Animal Care and Use Committee of the State of Hamburg (ORG870, 959) and the North Rhine-Westphalian State Agency for Nature, Environment and Consumer Protection (LANUV, 07/11) and conform to the National Institutes of Health's Guide for the Care and Use of Laboratory Animals (2011). Studies were performed using male and female (aged 10-24 weeks) C57BL/6 mice (stock number 000664, Jackson Laboratories) and mice homozygous (db/db) or heterozygous (db/het; control) for the diabetes spontaneous mutation (Leprdb; BKS.Cg-Dock7m+/+ Leprdb /J, stock number 000642, Jackson Laboratories). The authors have used the protocols at hand without variations for mice aged up to 60 weeks.
1. Location and dissection of murine stellate ganglia
NOTE: Even though descriptions and drawings are mostly available in bigger species, some publications have previously described the location of the SG in rats24 and mice25 using anatomical methods and fluorescent reporter lines, respectively.
2. Whole mount immunohistochemistry protocol
NOTE: This protocol is adapted from cardiac whole mount stainings4,29. Perform incubation steps for every single SG in one well of a 96-well plate and use 100 μL (for antibody-containing solutions) to 200 μL (for all other solutions) of the solution to ensure complete coverage. Regularly check the coverage and correct immersion of the SG with binoculars. Remove liquids manually with a 200 μL pipette with an additional 10 μL tip on top of the 200 μL tip. This will prevent aspiration of the SG in the pipette tip. Use freshly prepared solutions and sterile liquids to prevent bacterial growth.
3. Whole mount in situ hybridization
NOTE: Whole-mount in situ-hybridization of the SG is adapted from the organ of corti32 and the commercial RNA fluorescence in situ protocol (see Table of Materials). Obtain probes for the genes of interest and buffers and solutions from the supplier. All incubation steps are performed at RT, if not mentioned otherwise. Use sterile PBS. If interested in staining several SG in one well, use at least 150 μL of buffers and solutions.
4. Imaging and analyses of murine stellate ganglia
5. Molecular analyses of murine stellate ganglia
NOTE: Include controls depending on your experimental design. This could be SG with different genotypes and disease background and/or other autonomic ganglia, such as the sympathetic superior cervical ganglion (located in the neck area, see detailed description in Ziegler et al.35) or parasympathetic ganglia (such as intracardiac ganglia, see Jungen et al.4).
Figure 1 visualizes how to identify and dissect the SG. Figure 1A shows a schematic drawing of the location, while Figure 1B presents the view into the thorax after removal of the heart-lung-package. The left and right longus colli muscles medial from the SG and the rib cage are important landmarks for orientation. Dissection is performed along the dotted lines between muscles and the first rib. The SG and the sympathetic chain become visible as white structures (Figure 1C). Figure 1D shows a magnification of the region between the left longus colli muscle and the first rib, where the left SG is located. Morphology of the SG differs between individuals. It often consists of a fusion of the inferior cervical and the first to the third thoracic ganglia24. Some variety that the experimenter can expect in murine SG is depicted in Figure 1E,F, where left and right SG of five male C57Bl6 wild type mice are photographed.
Evaluation of gross anatomic overview, as well as cellular and subcellular analyses of cells in the SG innervating the heart can be performed by whole mount techniques on protein and RNA level. An overview of a SG is presented in Figure 2. Myocardial sympathetic fibers originate from cell bodies in the SG. These are visualized by staining with an antibody against tyrosine hydroxylase (TH). TH-expressing neuronal somata are surrounded by nerve fibers staining positive for choline acetyltransferase (ChAT). These are most likely presynaptic fibers37,38. An exemplary magnification from TH and ChAT co-labeling is presented in Figure 2A. Glial cells surrounding neuronal cell bodies can be visualized by staining for S100B. This is depicted in Figure 2B in combination with the neural marker PGP9.5. Figure 2C-F shows exemplary analyses to study the SG on a subcellular level, using whole mount in situ hybridization and immunofluorescent co-staining. The protein TH (Figure 2C, red) and mRNA molecules of Tubb3 (Figure 2D, white) are expressed in large neuronal cell bodies, while mRNA of S100b (Figure 2E, green) is also detectable in surrounding glia cells. In the merge (Figure 2F), it is visible that some neurons are negative for TH but express Tubb3, while S100b mRNAs can also be detected in surrounding cells, as depicted in the magnification in Figure 2G.
Figure 3 presents potential quantitative analyses and pitfalls for studying the murine SG. Images from TH-stained SG (Figure 3A) can be used for cell size measurements as was performed exemplary for a mouse model of diabetes. Neuronal somata from control (db/het) SG were 388.8 ± 123.8 μm2 vs. in diabetic SG (db/db) 407.33 ± 139.6 μm2 (Figure 3B, n = 2 SG, 100 cells per SG per genotype, P = 0.348, data were compared using Mann-Whitney test). Figure 3C shows the expression of genes from different cell types of the SG (n = 6-7). Pooling of both SG from one animal allows gene expression measurements of approximately 24 assays (12 genes in duplicates). We typically normalize samples for Cdkn1b (detected at Ct values of 25.4 ± 0.97) as well as the neuronal marker Neun/Rbfox3 (32.5 ± 0.7) if it is necessary to account for other cell types and neuronal purity of the dissection. Genes that we found useful for characterizing molecular processes in the SG include the sympathetic gene Th (22.4 ± 1.6), Chat, which could indicate cholinergic transdifferentiation (expressed at Ct values of 30.8 ± 1.3) and Gap43, a marker for neuronal sprouting (detectable at Ct values of 22.4 ± 1.4). Genes expressed in non-neuronal cell type include S100b (for glial cells, 27.3 ± 1.2), Ki-67 (for proliferating cells, 33.0 ± 1.6) and Cd45 (for immune cells, 30.2 ± 1.1).
The SG is surrounded by a capsule of connective tissue26, visualized via hematoxylin and eosin staining in Figure 3D,E. Occasionally, we observed inconsistencies in antibody-based staining as demonstrated in Figure 3F, most likely due to incomplete removal of the capsule. While ChAT and TH staining are only detectable in some parts of the SG, nuclei counterstained with DAPI are detectable throughout. The dotted line in the merged image separates successful staining (right of the line) from unsuccessful staining (left of the line).
Data are presented as mean ± standard deviation. Statistical significance was defined as a P value of <0.05; statistical analysis was performed using commercial software.
Figure 1: Location, dissection, and morphology of the murine stellate ganglia. (A) Schematic drawing of the location of the stellate ganglia (SG). (B) View into the thorax after removal of the heart-lung package. It is important to note that the SG are not immediately visible most of the time. The longus colli muscles are located lateral from the spine. The SG are located lateral from the muscles at the junction with the first rib. Carefully dissect lateral to the muscles (area marked by dotted line) to uncover the ganglia. After dissection, ganglia (left and right, LSG and RSG, respectively) and the sympathetic chain can be made out as white, long structures. (C) An exemplary dissection showing the ganglia and anatomical landmarks. (D) Magnification of the LSG. (E) LSG and (F) RSG from wild type, male C57Bl6 mice (16 weeks) were dissected and photographed to show the variations in morphology and size. Scale bar represents 1,000 μm. Please click here to view a larger version of this figure.
Figure 2: Visualization of different cell types in murine stellate ganglia via whole mount immunohistochemistry and in situ hybridization. (A) Gross overview of a murine stellate ganglion (SG) stained for the sympathetic marker tyrosine hydroxylase (TH) and choline acetyltransferase (ChAT). The magnification shows TH-positive cell bodies and the presence of ChAT-positive, most likely presynaptic, nerve fibers surrounding neuronal somata. (B) Glial cells ensheathing neuronal cell bodies can be visualized by staining for S100B, here in combination with the neuronal marker PGP9.5. (C,D,E,F) Microscopic images from one SG stained whole mount via a combination of immunohistochemistry for TH (red) and in situ hybridization for Tubb3 (D, white) and S100b (E, green). Nuclei are counterstained with DAPI (blue). (F) The merge shows that not all neuronal (Tubb3-positive) cells are TH positive. S100b mRNAs can be detected within neuronal somata, but also surrounding cells, as marked by an arrow in the magnification in (G). Please click here to view a larger version of this figure.
Figure 3: Potential quantitative analyses and pitfalls. (A) Images from Tyrosine-hydroxylase (TH)-stained SG can be used for cell size measurements using ImageJ. (B) This was performed in a mouse model of diabetes (100 cells per SG, n = 2 SG per genotype, data was compared using Mann-Whitney test). (C) Exemplary genes expressed in the SG that might be useful for characterizing molecular processes. Cdkn1b as well as the neuronal marker Neun/Rbfox3 (to account for the presence of other cell types) can be used for normalization. Tyrosine hydroxylase (Th) serves as a sympathetic marker, choline acetyltransferase (Chat) for cholinergic transdifferentiation, Gap43 for neuronal sprouting. Genes expressed in non-neuronal cell types include S100b (glial cells), Ki-67 (proliferating cells) and Cd45 (immune cells). (D) Hematoxylin and eosin staining of a formalin-fixed SG visualizes connective tissue and cells on top of the SG. (E) Magnification from the boxed area. (F) On some occasions, we observed failure in antibody-based staining, most likely due to incomplete removal of the capsule. While ChAT (red) and TH (green) staining are only detectable in some parts of the SG, nuclei counterstained with DAPI (dark blue) are detectable throughout. The dotted line in the merged image separates successful staining (right of the line) from unsuccessful staining (left of the line). Please click here to view a larger version of this figure.
The understanding of cellular and molecular processes in neurons and glial cells of the sympathetic nervous system that precede the onset of VA is of high interest, as sudden cardiac arrest remains the most common cause of death worldwide5. Therefore, in the current manuscript, we provide a basic repertoire of methods to identify the murine SG – a murine element within this network – and perform subsequent analyses on RNA, protein, and cellular level.
One challenge of the murine SG is its size and the limited number of cells39. Due to this, different animal models, such as rats40, dogs22, and pigs41 are used for studies on the SG. Still, there is a variety of well-established disease models for cardiac phenotypes available in mice and some of these have already been characterized in different aspects of cardiac innervation and arrhythmia, such as models for diabetes23, myocardial infarction42, or myocarditis43. Therefore, further studies of the murine SG are warranted to further characterize autonomic dysfunction in the light of VA. These can be completed by other approaches to study innervation of the murine heart such as functional experiments4,23,44 including in-vivo stimulation of the SG23.
Due to its small size and its location within the thoracic cavity24, manipulation of the murine SG in vivo is challenging, although it has been performed successfully23. For this reason, some studies therefore focus on the superior cervical ganglia, which are located more accessibly in the neck, upstream of the SG in the sympathetic chain behind the carotid bifurcation into internal and external carotid arteries24,35. Cardiac denervation via removal of the superior cervical ganglia has been shown to attenuate myocardial inflammation, hypertrophy, and cardiac dysfunction after myocardial infarction35. However, it is important to note that the superior cervical ganglia innervate different regions of the heart, most prominently the anterior side45. In addition, a recent in-depth literature review came to the conclusion that the role of cervical ganglia on sympathetic innervation of the heart remains unclear in humans46. This highlights the importance of characterizing the SG for studies of cardiac sympathetic innervation.
It is important to note that the heart is not the only target of the SG. Among others, lungs47 and sweat glands in the forepaw48 are also innervated from fibers originating in the SG, the latter are an exception to sympathetic physiology as they express choline acetyltransferase37. Temporary blockade of the SG is studied with regard to inflammatory processes in acute lung injury49 or for treatment of hot flushes and sleep dysfunction50; therefore, the protocols at hand might offer a repertoire for mechanistic questions in these fields. When focusing on cardiac disease models, it should be kept in mind for interpretation of results that cardiac neurons cannot be differentiated by morphology or electrophysiological properties from non-cardiac neurons51. This can be achieved by retrograde tracing, thereby the location of neurons projecting to the heart was shown to be located in the cranio-medial parts of the SG52.
Additionally, it is important to note that besides different types of neurons, sympathetic ganglia are made up of ensheathing glia, so-called satellite glial cells or satellite cells marked by expression of the glial marker S100B53. While little is known about the role of these cells in cardiovascular pathologies, glial activation and expression of the glial fibrillary acidic protein (GFAP) has been described in SG from patients with arrhythmias18.
Some pitfalls should be kept in mind with the presented methods: we observed inconsistencies in antibody-based staining at some occasions and hypothesized that incomplete removal of the connective tissue capsule ensheathing the SG might be at fault, as they have been described to vary in permeability among different types of ganglia26. Mechanical removal of the capsule using fine forceps has been described in the superior cervical ganglion of rats up to postnatal day 1028 and desheathing is mentioned in literature for adult rat SG54,55 and mice56. Removal of the SG capsule might vary between age28 and – due to size differences – species. In our experience, fresh dissection, removal of as much connective tissue as possible using fine forceps and thorough permeabilization as described in the protocol at hand, are important factors for successful staining. Regarding quantitative real-time PCR, quick work and efficient lysis are essential. Pooling both SGs from one animal reliably allowed for the analysis of up to 12 different genes (when performing duplicates).
Even though function and gross anatomy of SG has been studied for decades now and every single cardiomyocyte is innervated by sympathetic fibers46, many open questions remain. For example, it remains unclear, why sympathetic neurons of the SG transdifferentiate transiently to a cholinergic phenotype in ischemic21 and non-ischemic heart failure, in animal models as well as in patients20. Recently, our group described a role of S100B-positive glial cells, which are also present in the murine SG, on nerve sprouting in the cardiac nervous system29. Whether these cells are relevant for sympathetic nerve sprouting after injury associated with VA11,12, needs to be elucidated in future studies. Importantly, innovative approaches, such as optogenetics52 and transcriptome analyses36 can complement established methods such as neuronal tracing in order to deepen the understanding of the sympathetic nervous system and its role on cardiac electrophysiology.
In conclusion, this repertoire allows the inexperienced investigator to perform a basic characterization of the SG in murine models of cardiac pathologies. We hope that this will stimulate the usage, combination, and creation of novel methods. This might help to increase the understanding of the underlying cellular and molecular processes in sympathetic neurons that might be responsible for the onset and maintenance of VA.
The authors have nothing to disclose.
The authors would like to thank Hartwig Wieboldt for his excellent technical assistance, and the UKE Microscopy Imaging Facility (Umif) of the University Medical Center Hamburg-Eppendorf for providing microscopes and support. This research was funded by the DZHK (German Centre for Cardiovascular Research) [FKZ 81Z4710141].
96-well plate | TPP | 92097 | RNAscope |
Adhesion Slides SuperFrost plus 25 x 75 x 1 mm | R. Langenbrinck | 03-0060 | Microscopy |
Albumin bovine Fraction V receptor grade lyophil. | Serva | 11924.03 | Whole mount staining |
bisBenzimide H33342 trihydrochloride (Hoechst) | Sigma-Aldrich, St. Louis, MO, USA | B2261 | Whole mount staining |
Chicken anti neurofilament | EMD Millipore | AB5539 | Whole mount staining |
Dimethyl sulfoxide (DMSO) | Merck, KGA, Darmstadt, Germany | D8418 | Whole mount staining |
Donkey anti chicken IgY Alexa 647 | Merck, KGA, Darmstadt, Germany | AP194SA6 | Whole mount staining |
Donkey anti goat IgG Alexa 568 | Thermo Fisher Scientific | A11057 | Whole mount staining |
Donkey anti rabbit IgG Alexa 488 | Thermo Fisher Scientific | A21206 | Whole mount staining |
Drying block 37-100 mm | Whatman (Sigma Aldrich) | WHA10310992 | Whole mount staining |
Eosin Y | Sigma Aldrich | E4009 | Whole mount staining |
Ethanol 99 % denatured with MEK, IPA and Bitrex (min. 99,8 %) | Th.Geyer | 2212.5000 | Whole mount staining |
Eukitt mounting medium | AppliChem | 253681.0008 | Whole mount staining |
Fluoromount-G | Southern Biotech | 0100-01 | Whole mount staining |
Fluoromount-G + DAPI | Southern Biotech | 0100-20 | Whole mount staining |
Goat anti choline acetyltransferase | EMD Millipore | AP144P | Whole mount staining |
H2O2 30% (w/w) | Merck, KGA, Darmstadt, Germany | H1009 | Whole mount staining |
Heparin Sodium 25.000 UI / 5ml | Rotexmedica | PZN: 3862340 | Preparation SG |
High-capacity cDNA reverse transctiption kit | Life technologies | 4368813 | RNA isolation |
Isoflurane (Forene) | Abbott Laboratories | 2594.00.00 | Preparation SG |
Mayer's hemalum solution | Merck | 1.09249.0500 | Whole mount staining |
Methanol | Sigma-Aldrich | 34860 | Whole mount staining |
Microscope cover glasses 20×20 mm or smaller | Marienfeld | 0101040 | Whole mount staining |
miRNeasy Mini Kit | Qiagen | 217004 | RNA isolation |
NanoDrop 2000c | Thermo Fisher Scientific | ND-2000C | RNA isolation |
Opal 570 Reagent Pack | Akoya Bioscience | FP1488001KT | RNAscope |
Paraformaldehyde, 16% w/v aq. soln., methanol free | Alfa Aesar | 43368 | Whole mount staining |
Pasteur pipettes, LDPE, unsterile, 3 ml, 154 mm | Th.Geyer | 7691202 | Whole mount staining |
Phosphate-buffered saline tablets | Gibco | 18912-014 | Whole mount staining |
Pinzette Dumont SS Forceps | FineScienceTools | 11203-25 | Preparation SG |
QIAzol Lysis Reagent | Qiagen | 79306 | RNA isolation |
Rabbit anti tyrosine hydroxylase | EMD Millipore | AB152 | Whole mount staining |
RNAlater | Merck | R0901-100ML | RNA isolation (optional) |
RNAscope Multiplex Fluorescent Reagent Kit v2 | biotechne (ACD) | 323100 | RNAscope |
RNAscope Probe-Mm-S100b-C2 | biotechne (ACD) | 431738-C2 | RNAscope |
RNAscope Probe-Mm-Tubb3 | biotechne (ACD) | 423391 | RNAscope |
Stainless steel beads 7 mm | Qiagen | 69990 | RNA isolation |
Sudan black B | Roth | 0292.2 | Whole mount staining |
TaqMan Gene Expression Assay Cdkn1b (Mm00438168_m1) | Thermo Fisher Scientific | 4331182 | Gene expression analysis |
TaqMan Gene Expression Assay Choline acetyltransferase (Mm01221880_m1) | Thermo Fisher Scientific | 4331182 | Gene expression analysis |
TaqMan Gene Expression Assay MKi67 (Mm01278617_m1) | Thermo Fisher Scientific | 4331182 | Gene expression analysis |
TaqMan Gene Expression Assay PTPCR (Mm01293577_m1) | Thermo Fisher Scientific | 4331182 | Gene expression analysis |
TaqMan Gene Expression Assay S100b (Mm00485897_m1) | Thermo Fisher Scientific | 4331182 | Gene expression analysis |
TaqMan Gene Expression Assay Tyrosin Hydroxylase (Mm00447557_m1) | Thermo Fisher Scientific | 4331182 | Gene expression analysis |
TaqMan mastermix | Applied biosystems | 4370074 | Gene Expression analysis |
Tissue Lyser II | Qiagen | 85300 | RNA isolation |
Triton X-100 10% solution | Sigma-Aldrich | 93443-100ml | Whole mount staining |
Tween-20 | Sigma-Aldrich | P9416-100ML | RNAscope |
Wacom bamboo pen | Wacom | CTL-460/K | Cell size measurements |
Whatman prepleated qualitative filter paper, Grade 595 1/2 | Sigma-Aldrich | WHA10311647 | Whole mount staining |
Wheat Germ Agglutinin, Alexa Fluor 633 Conjugate | Thermo Fisher Scientific | W21404 | RNAscope |