Here, we present a protocol based on c-FOS protein immunohistological detection, a classical technique used for the identification of neuronal populations involved in specific physiological responses in vivo and ex vivo.
Many studies seek to identify and map the brain regions involved in specific physiological regulations. The proto-oncogene c-fos, an immediate early gene, is expressed in neurons in response to various stimuli. The protein product can be readily detected with immunohistochemical techniques leading to the use of c-FOS detection to map groups of neurons that display changes in their activity. In this article, we focused on the identification of brainstem neuronal populations involved in the ventilatory adaptation to hypoxia or hypercapnia. Two approaches were described to identify involved neuronal populations in vivo in animals and ex vivo in deafferented brainstem preparations. In vivo, animals were exposed to hypercapnic or hypoxic gas mixtures. Ex vivo, deafferented preparations were superfused with hypoxic or hypercapnic artificial cerebrospinal fluid. In both cases, either control in vivo animals or ex vivo preparations were maintained under normoxic and normocapnic conditions. The comparison of these two approaches allows the determination of the origin of the neuronal activation i.e., peripheral and/or central. In vivo and ex vivo, brainstems were collected, fixed, and sliced into sections. Once sections were prepared, immunohistochemical detection of the c-FOS protein was made in order to identify the brainstem groups of cells activated by hypoxic or hypercapnic stimulations. Labeled cells were counted in brainstem respiratory structures. In comparison to the control condition, hypoxia or hypercapnia increased the number of c-FOS labeled cells in several specific brainstem sites that are thus constitutive of the neuronal pathways involved in the adaptation of the central respiratory drive.
The c-fos gene was identified for the first time at the beginning of 19801,2 and its product was characterized in 1984 as a nuclear protein having gene-activator properties3,4. It participates in long-term mechanisms associated with neuron stimulation. Indeed, changes in neuronal activity lead to second messenger signaling cascades that induce the expression of the immediate early gene c-fos, which induces the production of the transcription factor c-FOS. The latter initiates the expression of late genes and thus participates in adaptive responses of the nervous system to many different types of stimuli4. Thus, since the end of 19805,6, c-FOS protein detection has been frequently used to study the effects of exogenous factors on gene transcription in general4 and on the activity of the central nervous system (CNS) for mapping out neuronal pathways involved in different physiological conditions.
Basal c-fos expression has been studied in various species including mice, rat, cat, monkey, and human4. Thereby, the kinetics of its expression is relatively well known. The transcription activation is rapid (5 to 20 min)7,8, and the mRNA accumulation reaches a maximum between 30 and 45 min after the onset of stimulation9 and declines with a short half-life of 12 min. The c-FOS protein synthesis follows mRNA accumulation and could be detected by immunohistochemistry at 20 to 90 min post stimulation6.
Analysis of c-fos expression is classically used in in vivo studies to identify the central respiratory network involved in the ventilatory responses to hypoxia or hypercapnia10-14. More recently, this tool was also used in ex vivo brainstem preparations to explore central respiratory network adaptations to hypoxia or hypercapnia15-18. Indeed, these preparations generate a rhythmic activity classically assimilated to the central respiratory drive19. Thus, this type of preparation has the advantage of being completely deafferented, and therefore, results regarding c-fos expression only reflect the consequences of a central stimulation without any intervention of peripheral structures.
The c-FOS detection could be made by immunohistochemical or immunohistofluorescence approaches. Indirect immunodetection necessitates the use of a primary antibody against c-FOS and a secondary antibody directed against the species in which the primary antibody was produced. For the immunohistochemical method, the secondary antibody is conjugated with an enzyme (peroxidase, for example) that acts on a substrate (H2O2 for the peroxidase). The product of the enzymatic reaction is developed by a chromogen (3.3-diaminobenzidine tetrahydrochloride), which stains it and can be observed under light microscopy. The reaction could be reinforced using nickel ammonium sulphate. These methods allow the detection of actives neurons during different physiological challenges and therefore the identification and/or the mapping of peripheral and central pathways involved in the consecutive physiological responses.
Note: c-FOS detection is a standardized procedure involving several steps (Figure 1). All experiments were performed on rats or mice. Experimental protocols were approved by the Ethics Committee in Animal Experiment Charles Darwin (Ce5/2011/05), done in accordance with the European Communities Council Directive of September 22, 2010 (2010/63/EU) for animal care, and conducted in accordance with French laws for animal care.
1. Preparation of Solutions
2. c-fos Induction and Tissue Processing In Vivo
Note: Experiments were performed in rats (Sprague Dawley) or mice (C57BL/6).
3. c-fos induction and Tissue Processing Ex Vivo
Note: Experiments were performed in newborn rats or mice only.
4. Brainstem Sectioning
Note: From this step to the end, the protocol is strictly the same for in vivo and ex vivo inductions.
5. Immunohistological Procedures (Table 1)
Note: Throughout the procedure, the sections remain in the well inserts.
6. Data and Statistical Analysis
The c-FOS detection is a useful tool that allows identifying groups of activated cells under specific conditions such as hypoxia and hypercapnia in vivo (Figure 2A) or in situations that mimic these conditions ex vivo (Figure 2B). In vivo, newborn, young, or adult rodents were placed in an airtight box in which the gaseous environment is continually renewed by a gas mixture with a composition precisely defined for 30 to 180 min13,25,26 (Figure 2A). As the stimulation acts on the whole body, the related change in c-FOS could reflect both central and peripheral activated pathways. Ex vivo, deafferented preparations containing the medulla oblongata and spinal cord were placed in a chamber continuously superfused with an aCSF with different levels of O2 or pH in order to model hypoxic or hypercapnic conditions (Figure 2B) for 30 min16-18,27. As the stimulation acts only in these deafferented preparations, it is possible to conclude that only central mechanisms were involved in the activation of cells in the c-FOS identified structures.
The CNS was fixed with PFA, either by a perfusion via the systemic circulation in vivo, or by an immersion ex vivo (Figure 3). Due to this difference, the fixative diffusion is longer ex vivo than in vivo. Thus, the c-FOS signal might be affected, particularly in deep structures. However, because the c-FOS analysis is always performed in a comparative approach between control and stimulated in vivo animals or ex vivo preparations, this difference would not interfere with the results from a strict comparison of control and stimulated data. After that, the CNS was cryoprotected, sectioned, and engaged in the immunohistochemical procedure (Figure 3, Figure 4). In this way, some respiratory brainstem areas have been identified in vivo as modifying their activity under hypoxic or hypercapnic challenges, including the retrotrapezoid nucleus (RTN), which is the main central site of CO2 chemoreception10,13,28,29, the ventrolateral medulla (VLM), which contains the inspiratory rhythm generator 13, the commissural and median parts of the nucleus tractus solitarius (c/mNTS), which are the projection areas of the carotid bodies input13,25,26,30 (Figure 5), and the medullary raphe nuclei12,13.
Ex vivo, in deafferented conditions, neurons of the RTN and the VLM have been described to change their c-FOS immunoreactivity under conditions of reduced O216-18. The c-FOS detection may also be combined with other immunohistochemical detections to characterize the phenotype of activated cells14,31
Figure 1. Study Design for the Use of c-FOS Detection to Map the Neuronal Structures Involved in the Respiratory Adaptations to Hypoxia or Hypercapnia. Chronological steps of c-FOS detection technique in the CNS. CNS: Central Nervous System; IHC: Immunohistochemistry. Please click here to view a larger version of this figure.
Figure 2. C-FOS induction. (A) In vivo induction protocol. An animal was placed in a hermetic chamber and exposed to either air (21% O2; 79% N2), hypercapnic (21% O2; 4% CO2; 75% N2), or hypoxic (10% O2; 90% N2) gas mixture for 2 hr. (B) Ex vivo induction protocol. CNS was dissected out from newborn rodent aged 0 to 4 days. Medulla oblongata and spinal cord were isolated under magnification to obtain the preparations. They were placed in a chamber with the ventral surface facing upward, superfused with aCSF (pH 7.4) at 26 ± 1 °C. Modeling hypoxic stimulation in vivo was performed by reducing the fraction of O2 in the aCSF by bubbling an anoxic gas mixture containing 95% N2 and 5% CO2. Modeling hypercapnic stimulation in vivo was performed using an acid aCSF that differed from normal aCSF in terms of NaHCO3 concentration (decrease). The acid aCSF is saturated with O2 and adjusted to pH 7.23 by bubbling with 95% O2 and 5% CO2. Each stimulation condition was applied for 30 min. aCSF: artificial cerebrospinal fluid. Please click here to view a larger version of this figure.
Figure 3. Central Nervous System Sampling. In vivo, the CNS was obtained by dissection after fixative procedure using 4% paraformaldehyde perfused via the vascular system. Ex vivo, the CNS was directly immersed in 4% paraformaldehyde after dissection. Subsequently, the CNS was cryoprotected by immersion in a cryoprotective solution and sliced into 40 µm thick coronal sections using a cryostat. The rostro-caudal sections of medulla were placed in a 12-well plate containing PBS before immunohistochemical procedures. CNS: Central Nervous System; PBS: Phosphate-Buffered Saline. Please click here to view a larger version of this figure.
Figure 4. Indirect Immunohistochemical Procedures. The sections were first incubated with a rabbit polyclonal primary antibody against the c-FOS protein. Afterwards, they were treated with a biotinylated goat anti-rabbit secondary antibody against the primary antibody. Following this incubation, sections were treated with an Avidin-Biotin-peroxidase complex, which binds tightly to the secondary antibody to amplify the signal. Indeed, the complex contains several peroxidase molecules leading to a high staining intensity. Finally, sections were stained using a solution of 3.3-diaminobenzidine tetrahydrochloride and nickel ammonium sulfate, which produces a dark gray precipitate in the presence of the peroxidase enzyme. Please click here to view a larger version of this figure.
Figure 5. c-FOS-positive neurons on sections of the medulla oblongata in control mice and mice submitted to hypoxia in vivo. (A) Drawing of a section of the medulla oblongata adapted from Paxinos and Franklin24 (Bregma -7, 48 mm). The dotted area delimitates the commissural and median parts of the nucleus tractus solitarius c/mNTS. The box illustrates the structures illustrated by photomicrographs (B1 and B2). (B) Photomicrographs performed (x100) at the c/mNTS level in normoxia (B1. control, black frame) or under hypoxia (B2. blue frame). The black arrows show c-FOS-positive cells. (C) Histogram showing the mean number of c-FOS-positive cells per section at c/mNTS level in normoxia (black bar) or under hypoxia (blue bar). Values are expressed as mean ± SD. * indicates a significant difference between normoxia and hypoxia values – Student's unpaired t test; ***p< 0.001. CC: central canal of spinal cord; c/mNTS: commissural and median parts of the nucleus tractus solitarius. Please click here to view a larger version of this figure.
Table 1. Immunohistological Procedure. Chronological steps of the c-FOS procedure. Ab-I, primary antibody; Ab-II, secondary antibody; BSA, Bovine Serum Albumin; DAB, 3.3-diaminobenzidine tetrahydrochloride; Ni, nickel ammonium sulphate; PBS 0.1 M, 0.1 M phosphate-buffered saline (pH 7.4); PBST, 0.1 M Phosphate-buffered saline supplemented with 0.3% Triton X-100; Serum, serum from species used for secondary antibody production. Please click here to view a larger version of this figure.
C-fos is an immediate early gene, and the detection of its product, the c-FOS protein, is classically used to identify neuronal populations involved in specific respiratory responses in vivo11,13,25,28 and ex vivo16-18,27,32,33.
Critical Steps Within the Protocol
Be careful during the perfusion step. The 4% PFA solution must be well prepared and the fixation and post-fixation steps must be long enough to obtain optimal slicing and staining. Furthermore, the revelation is the most important step of the procedure; the intensity of the staining should be controlled to avoid background noise.
Advantages of the Technique and Significance with Respect to Existing Methods
Although some neurons do not seem to express the c-fos gene5, this method has been proved to be useful to determine neural pathways activated, for example in respiratory rhythm adaptation during chemical challenges. Compared to other immediate early genes, c-fos has a low expression in the absence of stimulation, allowing easier quantification of neuronal activity under a test situation6.
The c-FOS detection is a cellular technique that indicates global changes in activity in order to analyze neuronal populations involved in response to several stimuli. Compared with neuronal activity analysis using an electrophysiological approach, which most of the time concerns only a small number of cells in a defined brain area, c-FOS detection permits one to appreciate changes of activity in a large number of cells and therefore to define active areas in the whole CNS. C-FOS detection is easily compatible with unrestrained and awake animals, which is not the case during neuronal activity recording by electrophysiology. This is an advantage because it avoids stress and interference of anesthetics with the obtained results.
Ex vivo, we used a 30 min period of stimulation. It was previously shown that this stimulation is sufficient to induce changes in c-fos expression at the neuronal level34-36. Five min of stimulation are sufficient to induce changes in c-fos expression, which are observed after a latent period of 15 to 20 min37. The related increases in c-fos expression are related to changes in activity taking place during the first 10 min of stimulation.
Limitations of the Technique
Following stimulation, a delay is required before the accumulation of c-FOS in the nucleus. This delay of about 30 min corresponds to mRNA and protein synthesis. This item is opposed to the immediacy of the results that may be obtained by recording the activity of neurons with electrophysiology. This can cause a loss of information about rapid changes in neuronal activity after a specific and transient stimulus.
As c-FOS protein may have a half-life of 90 to 100 min4, its expression could be affected by the surgical procedures or stress associated with the manipulation of the animal, its constraint, or changes in its environment. Thus, it is necessary to minimize manipulations that could induce changes in neuronal activity not related to the studied stimulus and to perform tissue sampling immediately after the end of the physiological stimulus.
Modifications and Troubleshooting of the Technique
In this protocol, the revelation step used a solution that contained 3.3-diaminobenzidine tetrahydrochloride, nickel ammonium sulphate, and hydrogen peroxide, but the commercial kits for peroxydase could also be used for this step.
The authors have nothing to disclose.
The University Paris 13 supported this work. ASPT was supported by a University Paris 13 fellowship and the “Association Française pour le Syndrome d’Ondine”. FJ was supported by a Laboratory of Excellence GR-Ex fellowship. The GR-Ex (ref ANR-11-LABX-0051) is funded by the program “Investissement d’avenir” of the French National Research agency (ref ANR-11-IDEX-0005-02).
Cell culture plate 12-Well | Costa | 35/3 | |
15 mm Netwell inserts with mesh polyester membrane | Corning | 3477 | The 15mm diameter well inserts have 74µm polyester mesh bottoms attached to polystyrene inserts |
primary antibody (rabbit polyclonal antibody against the c-Fos protein) | Santa Cruz Biotechnology | sc-52 | |
Vectastain Elite ABC KIT | Vector laboratories | PK-6101 | |
(Rabbit IgG-secondary antibody) | |||
NaH2PO4*2H2O | Sigma | 71505 | |
Na2HPO4 | Sigma | S7907 | |
Paraformaldehyde | Sigma | P6148 | |
NaOH 0.1N | Sigma | 43617 | |
Polyvinyl-Pyrrolidone | Sigma | PVP-360 | |
Sucrose | Sigma | S7903 | |
NaCl | Sigma | S7653 | |
Ethylene-glycol | Sigma | 33068 | |
Triton X100 | Sigma | T8787 | |
Trisma HCl | Sigma | T5941 | |
Trisma Base | Sigma | T1503 | |
3.3-diaminobenzidine tetrahydrochloride | Rockland | DAB50 | |
Nickel ammonium sulphate | Alfa Aesar | 12519 | |
H2O2 | Sigma | H1009 | |
Xylene | Sigma | 33817 | |
Entellan Neo | Merck Millipore | 107961 | |
Slide | Thermo-scientific | 1014356190 | Superfrost ultraplus |
Cover glass | Thermo-scientific | Q10143263NR1 | 24 x 60mm |
BSA | Sigma | A2153 |