We present a surgical protocol detailing how to perform a cut or crush axotomy on the facial nerve in the mouse. The facial nerve axotomy can be employed to study the physiological response to nerve injury and test therapeutic techniques.
The goal of this surgical protocol is to expose the facial nerve, which innervates the facial musculature, at its exit from the stylomastoid foramen and either cut or crush it to induce peripheral nerve injury. Advantages of this surgery are its simplicity, high reproducibility, and the lack of effect on vital functions or mobility from the subsequent facial paralysis, thus resulting in a relatively mild surgical outcome compared to other nerve injury models. A major advantage of using a cranial nerve injury model is that the motoneurons reside in a relatively homogenous population in the facial motor nucleus in the pons, simplifying the study of the motoneuron cell bodies. Because of the symmetrical nature of facial nerve innervation and the lack of crosstalk between the facial motor nuclei, the operation can be performed unilaterally with the unaxotomized side serving as a paired internal control. A variety of analyses can be performed postoperatively to assess the physiologic response, details of which are beyond the scope of this article. For example, recovery of muscle function can serve as a behavioral marker for reinnervation, or the motoneurons can be quantified to measure cell survival. Additionally, the motoneurons can be accurately captured using laser microdissection for molecular analysis. Because the facial nerve axotomy is minimally invasive and well tolerated, it can be utilized on a wide variety of genetically modified mice. Also, this surgery model can be used to analyze the effectiveness of peripheral nerve injury treatments. Facial nerve injury provides a means for investigating not only motoneurons, but also the responses of the central and peripheral glial microenvironment, immune system, and target musculature. The facial nerve injury model is a widely accepted peripheral nerve injury model that serves as a powerful tool for studying nerve injury and regeneration.
Many peripheral nerve injury models exist, but one that stands out for the study of motoneurons is the facial nerve axotomy model. The facial nerve, also known as cranial nerve VII, originates in the pons and innervates the muscles of facial expression1,2. In this surgical protocol, the facial nerve is exposed at its exit from the stylomastoid foramen and either cut or crushed. The severity of nerve injury can be classified following the Sunderland3 classifications, which differentiates the injury based on the intactness of the axons, endoneurium, perineurium, and epineurium, which are connective tissue layers that sequentially wrap around the axon bundles. In the crush injury (axonotmesis), the axons are severed, but the perineurium and epineurium are preserved. Complete functional recovery from facial nerve crush occurs in about 11 days because the intact nerve sheath serves as a conduit within which the axons regrow4,5. On the other hand, in the cut injury (neurotmesis), the axons and all 3 connective tissue layers are severed, and the entire distal nerve must regrow to restore musculature innervation. Surgical reconnection of the epineurium is often performed in human patients with nerve transection injuries, however the recovery outcomes are seldom optimal. Further study is required to understand why the nerve fails to regrow to its target and what therapies can be employed to improve and accelerate the regenerative process.
There are many advantages to studying nerve injury using the facial nerve axotomy model. First, the facial nerve axotomy procedure is quick, easy, and highly reproducible; and the resultant paralysis of the facial muscles does not impact vital functions and is well tolerated by the animal. Because this is a cranial nerve injury model, studying the motoneuron cell bodies is simplified because the motoneurons reside in a relatively homogenous population in the facial motor nucleus in the pons. The population does differ based on the subnuclear pattern within the facial motor nucleus, as there are seven subnuclei each specific to innervating a specific group of muscles, so subnuclear differences in response to axotomy may impact results2,6,7.
A major benefit of the facial nerve injury model is that the unaxotomized side can serve as a paired internal control because the nerve innervation is highly symmetrical and there is no crosstalk between the facial motor nuclei8. Another advantage of using this surgical method is the lack of direct trauma to the CNS or disruption of the blood brain barrier9. Complications such as excessive bleeding and infection are rare with this procedure.
A variety of analyses can be performed to assess the physiologic response to nerve injury. The recovery of the eye blink reflex and whisker activity can be used as a behavioral measure of functional recovery10,11. Video recording of vibrissae activity is currently the most powerful method for detecting recovery of facial nerve innervation12,13. After euthanasia, histological analysis of the brainstem can be performed on the motoneuron cell bodies within the facial motor nucleus. The facial motor nucleus is subdivided into seven subnuclei, each specific to certain facial muscles, allowing for differential examination of responses to injury2,6. Facial motoneurons can be counted to quantify cell survival, or immunohistochemistry can be used to identify biomarkers and specific cell populations14. The facial motor nucleus can be accurately microdissected using laser capture for molecular analysis of the cellular response to nerve injury15,16. Impacts of the facial nerve axotomy can be analyzed in the motor cortex17,18. Also, the nerve can be dissected to study Wallerian degeneration19 or axon regeneration20, and the muscles can be removed to study neuromuscular junctions21. The facial nerve axotomy can also be used to study the accompanying central and peripheral glial cells22, target musculature21, and the immune system response23. Although much has been accomplished in studying the facial nerve axotomy model24, further study of peripheral nerve injury is required because nerve damage is a significant problem for patients and current treatments fail to produce optimal results. This model is a powerful tool for examining the physiologic response to nerve injury and analyzing the effectiveness of nerve regeneration therapies.
All procedures executed are approved by the Indiana University School of Medicine Institutional Animal Care and Use Committee and follow National Institute of Health guidelines.
1. Surgical Technique
2. Anesthesia and Preparation
3. Surgical Approach
4. Closing and Recovery
After the facial nerve axotomy is performed, motoneuron loss occurs as a result of the injury. Motoneuron survival after injury depends on many variables, such as gender, animal age at time of surgery, and the timepoint at which the motoneuron counts are done, and the Moran and Graeber review24 and Jinno and Yamada review22 both summarize motoneuron survival data. Typically, about 86% of motoneurons survive at 28 days post-axotomy14,15,26. Kinetics of motoneuron loss are described in Serpe et al. 2000. Figure 1 illustrates the variation in motoneuron survival in multiple genetically modified mice. No significant differences are observed in facial motoneuron counts of the control side, indicating that the genetic alterations do not impact baseline counts. Compared to the motoneuron survival in wild type mice (84% ± 2.0; Figure 1A,D), significant cell loss is observed in a mouse model of amyotrophic lateral sclerosis (SOD1G93A; 68% ± 1; Figure 1B,E) as well as the immunodeficient recombination-activating gene-2 knockout mouse (RAG-2-/-; 57% ± 2.5; Figure 1C,F)27.
Figure 2 demonstrates the laser capture microdissection technique applied to the facial motor nucleus. The entire facial motor nucleus can be captured (Figure 2A-C), or subnuclei can be collected separately (Figure 2D-F). For greater precision, motoneurons can be captured individually, and the remaining neuropil can be collected for analysis (Figure 2G-I). Figure 3 depicts qPCR results of the RNA material extracted from the subnuclear samples comparing the ventromedial and ventrolateral subnuclei. The four genes tested, βII tubulin, growth associated protein-43 (Gap-43), hemopoietic- and neurologic-expressed sequence-1 (Hn1), and brain-derived neurotrophic factor (BDNF) are all associated with the nerve regeneration response and there are interesting differences between the two subnuclei and their gene expression profiles after axotomy16.
Figure 1. Representative coronal facial motor nucleus sections stained with thionin and quantified 28 days after facial nerve transection. Facial motor nuclei are shown from (A,D) WT, (B,E) SOD1G93A, and (C,F) RAG-2-/- mice (control side, axotomized side). Scale bars = 120 μm. This figure has been modified from 27. Please click here to view a larger version of this figure.
Figure 2. Laser microdissection of the facial motor nucleus. (A) Thionin-stained section of axotomized facial nucleus, (B) with partial laser microdissection of the axotomized facial nucleus, and (C) collection of laser microdissected tissue. A template (D) of the subnuclei was superimposed on the computer screen to identify the ventromedial and ventrolateral facial subnuclei for laser microdissection (E,F). Facial motoneurons were laser microdissected based on their morphology with a visible nucleus and nucleolus (* indicates motoneuron, G,H), while FMN cell body fragments, indicated by the arrows (G), were laser microdissected separately and disposed of to eliminate FMN mRNA in the neuropil samples. After all FMN and cell body fragments were collected, the remaining facial nucleus tissue was laser microdissected as the neuropil sample (I). Scale bars = 100 μm. This figure has been modified from 16. Please click here to view a larger version of this figure.
Figure 3. Pro-regeneration and pro-survival mRNA expression in the ventromedial (VM) and ventrolateral (VL) facial motor subnuclear regions following facial nerve axotomy. Average percent of mRNA expression ± SEM in the transected VM and VL facial subnuclei relative to the unoperated control subnuclei (A-D). Time-course of mRNA expression includes no injury (0), 3, 7, 14, and 28 dpo for βII tubulin (A), GAP-43 (B), Hn1 (C), and BDNF (D). # represents significant differences of VL compared to VM, at p < 0.05. This figure has been modified from 16.
The critical step for this protocol is positioning the mouse properly before surgery is begun. If the mouse is not lying flat on its side, the ear is not taped at the correct angle, or the incision is made in an incorrect location, then finding the facial nerve becomes much more difficult. When this technique is mastered, surgeries will take only minutes per mouse.
Either sutures, glue, or wound clips can be used to close the wound. Wound clips are preferred because of the small size of incision and rapidity with which they can be applied. We have determined that if wound clips are used and the mice are housed together, female mice fare better than male mice because males housed together often remove their cagemate’s wound clips, leading to infection of the surgical site.
Modifications can be made to the injury, such as changing the location along the nerve to induce injury. For example, the injury could be administered along one of the branches of the facial nerve28, or the skull could be opened to deliver an intracranial injury29. In addition, a variety of treatments can be implemented, such as applying drugs to the nerve stump14, injecting medicines systemically30, or even performing cell transfers31. Motoneuron diseases such as amyotrophic lateral sclerosis can also be studied using the facial nerve axotomy model32. Ongoing applications of this technique involve studying nerve reconnection strategies after transection and the application of pharmaceutical agents to accelerate nerve regeneration after injury.
One of the most useful aspects of this surgery is that it is easy to confirm successful axotomy by monitoring the mouse’s vibrissae orientation and movements as well as the eye blink reflex.
The paralysis does not affect the animal’s vital functions and overall is well tolerated in the mouse.
A major advantage of the facial nerve injury model is that the isolated study of motoneurons is possible. However, because this is a cranial nerve injury, results found using this method may not translate to spinal, autonomic, or sensory nerve injury models. To study these other nerve injury types, a sciatic33 or tibial34 nerve injury model is recommended. The drawback of these models is that there is crosstalk from the contralateral neurons in the spinal cord that can confound results. There is no crosstalk between the facial motor nuclei, so the contralateral facial motor nucleus can serve as a paired internal control. The facial nerve axotomy model presents many advantages for studying nerve injury and regeneration.
The authors have nothing to disclose.
This work is funded by NIH RO1 NS 40433 (K.J.J.).
Name of Material/ Equipment | Company | Catalog Number | Comments/Description |
Stereo Microscope | Leica | M60 | |
Labeling tape | Fisher Scientific | 15-952 | |
Vannas-Tübingen Spring Scissors – Straight/Sharp/8.5cm/5mm cutting edge | Fine Science Tools | 15003-08 | Sterilize before use |
Dumont #5/45 Forceps – Standard tips/Angled 45°/Dumoxel/11cm | Fine Science Tools | 11251-35 | Sterilize before use |
Michel Suture Clips – 7.5mm x 1.75 mm | Fine Science Tools | 12040-01 | Described as "wound clip" in protocol, sterilize before use |
Hagenbarth Cross Action Wound Clip Applier 5" | George Tiemann & Co | 160-910 | Used to apply wound clip, sterilize before use |
Michel Suture Clip Applicator & Remover – For 7.5 mm Clips | Fine Science Tools | 12029-12 | Used to remove wound clip |
0.9% Sodium Chloride Injection, USP | Hospira | 0409-4888-10 | |
Betadine, 16 oz, with dispenser | Fisher Scientific | 19-027132 | |
70% Ethanol | |||
Glass Bead Sterilizer |