This manuscript presents protocols for surgically inflicting controlled blunt and sharp spinal cord injuries to a regenerative axolotl (Ambystoma mexicanum).
The purpose of this study is to establish a standardized and reproducible regenerative blunt spinal cord injury model in the axolotl (Ambystoma mexicanum). Most clinical spinal cord injuries occur as high energy blunt traumas, inducing contusion injuries. However, most studies in the axolotl spinal cord have been conducted with sharp traumas. Hence, this study aims to produce a more clinically relevant regenerative model. Due to their impressive ability to regenerate almost any tissue, axolotls are widely used as models in regenerative studies and have been used extensively in spinal cord injury (SCI) studies. In this protocol, the axolotls are anesthetized by submersion in a benzocaine solution. Under the microscope, an angular incision is made bilaterally at a level just caudal to the hind limbs. From this incision, it is possible to dissect and expose the spinous processes. Using forceps and scissors, a two-level laminectomy is performed, exposing the spinal cord. A custom trauma device consisting of a falling rod in a cylinder is constructed, and this device is used to induce a contusion injury to the spinal cord. The incisions are then sutured, and the animal recovers from anesthesia. The surgical approach is successful in exposing the spinal cord. The trauma mechanism can produce contusion injuries to the spinal cord, as confirmed by histology, MRI, and neurological examination. Finally, the spinal cord regenerates from the injury. The critical step of the protocol is removing the spinous processes without inflicting damage to the spinal cord. This step requires training to ensure a safe procedure. Furthermore, wound closure is highly dependent on not inflicting unnecessary damage to the skin during incision. The protocol was performed in a randomized study of 12 animals.
The overall goal of this study was to establish a controlled and reproducible microsurgical method for inflicting blunt and sharp SCI to the axolotl (Ambystoma mexicanum), producing a regenerative spinal cord injury model.
SCI is a severe condition that, depending on the level and extent, inflicts neurological disability to the extremities along with impaired bladder and bowel control1,2,3. Most SCI are the result of high energy blunt trauma such as traffic accidents and falls4,5. Sharp injuries are very rare. Therefore, the most common macroscopic injury type is contusions.
The mammalian central nervous system (CNS) is a non-regenerative tissue, hence no restoration of neurological tissue following SCI is seen6,7,8. On the other hand, some animals have an intriguing ability to regenerate tissues, including CNS tissue. One of these animals is the axolotl. It is widely used in studies of regenerative biology and is of interest in spinal cord regeneration, because it is a vertebrate9,10,11,12.
Most SCI studies in the axolotl are performed as either amputation of the entire tail or ablation of a larger part of the spinal cord9,10,11,12. Recently, a new study was published on blunt injuries13 that mimics clinical situations better. Whereas complete appendage amputation in the axolotl results in full regeneration, some non-amputation-based regenerative phenomena are dependent on the critical size defect (CSD)14,15. This means that injuries exceeding a critical threshold are not regenerated. To develop a regenerative model with a higher clinical translational value, this study investigated whether a 2 mm blunt trauma would exceed the CSD limit.
This method is relevant for researchers working on spinal cord regeneration in small animal models, especially in the axolotl. Furthermore, it may be of more general interest, because it exhibits a way of using standard laboratory equipment to develop a blunt trauma mechanism that is suitable for use in small animals in general.
All applicable institutional and governmental regulations concerning the ethical use of animals were followed during this study. The study was conducted under the approval id: 2015-15-0201-0061 by the Danish Animal Experiment Inspectorate. Animals were Mexican axolotls (Ambystoma mexicanum, mean body mass ± STD: 12.12 g ± 1.25 g).
1. Preparation
2. Anesthesia
3. Microsurgical Laminectomy
NOTE: The laminectomy is performed under a stereomicroscope.
4. Introducing a Contusion Type Injury (Figure 2)
5. Introducing a Sharp Injury
NOTE: Perform these steps after 3.5.4.
6. Closing the Surgical Wound
7. Returning the Animal to the Anesthetic-free Solution
8. Postoperative Ultrasound
The purpose of the protocol is to produce an SCI that will paralyze the motor and sensory functions caudal to the injury. Because the axolotl is regeneration-competent it restores function within weeks, allowing researchers to study CNS regeneration during a short time span.
Anesthesia was provided for 45 min to all animals, and no episodes of preterm recovery were experienced. All animals recovered within an hour and showed no signs of damage from anesthesia in the following weeks13,16.
The laminectomy was successful in all animals. However, anatomical variation in the width of the spinal canal called for the widening of the canal using forceps and a twist in some individuals. Furthermore, residual laminae in some individuals prevented the falling rod from reaching its target, hence making it imperative that the surgeon clean the field from the residual bone and prominences.
Closing the incisions was associated with some difficulties, especially during the piloting phase of the study. Sutures in the top part of the keel would not hold and resulted in insufficient closures. The closure of one animal in the study did not hold, resulting in the keel being torn, subsequent infection, and death. This stresses the need for careful suturing along the entire incisions.
The initial mechanical injuries were obvious during the procedure. During the model development, injured and sham animals were stained with hematoxylin and eosin to validate the injury. Representative results of each group are shown in Figure 3A1,A2 and Figure 3C1,C2. Regeneration was confirmed by histological sections preparations made after nine weeks (Figure 3B1,B2 and Figure 3D1,D2), which showed a reestablished spinal cord connection in the SCI animals.
Injury and regeneration can be followed by examining neurological function. Stimulating the tail with a light touch and pinching from forceps will reveal whether tactile and nociceptive sensory functions have been lost and potentially reestablished. A neurological score was defined based on the reaction of the animal: 0 point = no response, 1 point = local tail movement, 2 points = truncal movement, 3 points = coordinated movement of limbs and/or head alongside with truncal movement, 4 points = animals with immediate coordinated fast movement. In six SCI animals versus five sham animals the loss of neurological function three weeks post injury was found, and a gradual restoration within nine weeks (Figure 4 and Supplementary Video 1).
Ultrasonographic images of the injured spinal cord can be obtained using the above protocol. Visualizing the SCI site was possible due to the obvious lack of bony spinous processes (Figure 5). Furthermore, using the B-mode the dorsal artery of the uninjured spinal cord could be visualized, yielding a marker of vessel integrity.
It is possible to test the animals immediately upon reawakening. However, some animals expressed local small amplitude, repetitive, and rhythmic tail movement upon stimulation comparable to the clonus phenomena observed in human SCI. These movements might represent clonus or a lack of central reflex suppression and could potentially cause more damage to the newly injured spinal cord. Therefore, testing the animals is not recommend before one-week post injury.
From simple qualitative observation of the animals, it will be evident that the tail is paralyzed, and swimming is significantly inhibited, making the animals completely dependent on moving their limbs. These observations will also validate the success of the protocol.
High-field MRI scans (9.4 T) were performed immediately after injury to visualize the injury in vivo (Figure 6). However, the scans were generally low in signal-to-noise ratio compared to those of non-operated animals, likely due to bleeding and hemosiderin. Hence, it was concluded that MRI was a suboptimal method to validate the injury and success of the protocol.
Figure 1: Schematic drawing of the microsurgical laminectomy. Please click here to view a larger version of this figure.
Figure 2: Schematic drawing of the contusion trauma mechanism. (A) The entire setup, showing the falling rod above the animal. (B) The disassembled mechanism, showing how the rod is disconnected from the electromagnet. (C) The falling rod is connected to the electromagnet. The falling height adjustment cylinder is installed, and the electromagnet and rod loaded into the cylinder. Height adjustment of the entire system is controlled by an adjusting wheel. (D) Turning off the electromagnet will cause the rod to fall without the operator touching the system. Figure was originally published by Thygesen et al.13. Please click here to view a larger version of this figure.
Figure 3: Histological sections hematoxylin and eosin stained immediately and nine weeks post injury. (A1) SCI animal immediately after injury. (B1) SCI animal at nine weeks. (C1) Sham surgery animal immediately after injury. (D1) Sham animal at nine weeks. Red square = marks the injury of the SCI animals, and the laminectomy of the sham animal. Figure 2A, Figure 2B, Figure 2C are magnifications of these areas at 5x. Blue arrow = uninjured spinal cord. This figure was originally published by Thygesen et al.13. Please click here to view a larger version of this figure.
Figure 4: Graph of response to tactile stimuli. The response of the SCI groups is lower after three weeks, compared to the sham group. WPI = weeks post injury, Black line = SCI, Grey color = sham. Sham n = 5, SCI n = 6. Figure was originally published by Thygesen et al.13. Please click here to view a larger version of this figure.
Figure 5: Ultrasonographic image showing the spinal cord in a sagittal section. Yellow lines mark the spinal cord, yellow circle the injury site, and white arrows mark the vertebrae. Please click here to view a larger version of this figure.
Figure 6: MRI scans at different time points post injury or sham surgery. CSF surrounding the spinal cord is lacking, especially at three WPI for the SCI animal, indicating swelling of the spinal cord. Darkening of the spinal cord indicates edema as well. Notice how these changes disappear as regeneration progresses. Yellow arrow = the area of laminectomy. Figure was originally published by Thygesen et al.13. Please click here to view a larger version of this figure.
Supplemental Video 1: Video showing the neurological function after tactile stimuli and later a nociceptive stimulus. First a healthy control animal, and then an animal suffering from SCI. Please click here to download this video.
Because risk of injury to the spinal cord is significant, the critical steps of the protocol are removing the spinous processes and widening of the bony access to the spinal canal if needed. As mentioned in the protocol, removing the most cranial process first is highly recommended. This will mean that the more caudal processes protect the spinal cord from being hit by the scissors. It is recommended to ensure enough surgical access, meaning to not make too small a primary incision. Also, when grasping anything with forceps, the direction of the pull applied must always be considered. Applying a gentle pull away from the spinal cord will protect it in the event of the grasp failing and a slip of the instrument.
The surgical procedure in the axolotl is not different from other animals. However, certain important differences do exist, primarily attributable to the tissue composition and size of the animal. The axolotl keel skin is very fragile, and paradoxically does not heal well upon small damages inflicted during incision. Caution should be taken, especially upon the primary incisions, because damage will substantially complicate the suturing. The bones of very young axolotls are very soft. This means that often basic anatomical forceps may suffice in bone removal. This presents another element of caution, because pinching the spinous processes could inflict substantial damage. The subcutaneous and muscle fascia layers are not available for suturing, due to their fragile tissue compositions. It is imperative to ensure a calm postoperative week. The animals may not rest sufficiently after the operation. Hence, they may inflict secondary damage to their spinal cord postoperatively. Their small anatomy does not allow for neither internal nor spline fixation.
Weight and falling height of the falling rod system is crucial to inflicting a contusion injury. During extensive piloting for an earlier study, the rod weight and falling height needed was found to be 25 g and 3 cm13. This was enough to induce paralysis in 12 g axolotls without cutting or disintegrating the spinal cord. Added weight or falling height might be needed in bigger animals. Furthermore, the diameter of the falling rod might need to be bigger in the case of bigger animals and shorter for smaller animals.
The model has some limitations. Because axolotls are not used for learned behavior studies, one cannot test complex neurological functions. The injury was introduced caudal to the limbs, sparing the hind limbs and bowel and bladder from being paralyzed. The reason for this was ethical, to reduce the impact on the animal to a minimum. However, it does limit the opportunity to study the effects on limb movements, which may be easier to describe and categorize. A large part of the SCI-associated morbidity stems from the loss of control of bowel and bladder. This model does not allow for future research in these fields. Inflicting damage rostral to the hind limbs would be possible, but it was not attempted.
Studying SCI in a regenerative model such as the axolotl allows for a different approach in SCI research. Because the animal model can regenerate, elimination studies will be able to reveal critical factors of regeneration. Conventional studies on SCI are performed in non-regenerative models, meaning that one will need to intervene on all critical factors to induce a regenerative response.
This model and protocol are in concordance with Krogh’s principle stating that: “For such a large number of problems there will be some animal of choice or a few such animals on which it can be most conveniently studied”17. Mammalian regeneration is inhibited by multiple factors. Inhibiting these in a mammalian model usually does not induce any effects. However, increasing levels of inhibitors in the axolotl should eliminate regeneration, and thereby reveal whether that inhibitor is critical or not10.
The authors have nothing to disclose.
Michael Pedersen, Aarhus University for his expertise and time on developing MRI protocols and setting up the entire project. Peter Agger, Aarhus University for his expertise and time on developing the MRI protocols. Steffen Ringgard, Aarhus University for his expertise and time on developing the MRI protocols. The development of the SCI model in the axolotl was kindly supported by The A.P. Møller Maersk Foundation, The Riisfort Foundation, The Linex Foundation, and The ELRO Foundation.
25 g custom falling rod | custom home made | ||
30 mm PVC pipe | custom home made | ||
Acetone | Sigma-Aldrich | 67-64-1 | Propanone |
Axolotl (Ambystoma mexicanum) | Exoterra GmbH | N/A | 12-22 cm and 10 g – 80 g, All strains (wildtype, melanoid, white, albino, transgenic white with GFP) |
Benzocain | Sigma-Aldrich | 94-09-7 | ethyl 4-aminobenzoate |
electromaget | custom home made | ||
Excel 2010 | Microsoft | N/A | Excel 2010 or newer |
ImageJ | National Institutes of Health | ImageJ 1.5e or newer. Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, https://imagej.nih.gov/ij/, 1997-2016. | |
kimwipes | |||
microsurgical instruments | N/A | N/A | Forceps and scissors |
MS550s | Fujifilm, Visualsonics | MS550s | 40 MHz center frequency, transducer |
MS700 | Fujifilm, Visualsonics | MS700 | 50 MHz center frequency, transducer |
Petri dish | any maker | ||
Soft cloth | N/A | N/A | Any piece of soft cloth measuring appromixately 70 x 55 cm^2 e.g. a dish towel |
Stereo microscope | |||
Vevo 2100 | Fujifilm, Visualsonics | Vevo 2100 | High frequency ultrasound system |