This protocol validates a reliable, easy-to-perform and reproducible rodent model of brain diffuse axonal injury (DAI) that induces widespread white matter damage without skull fractures or contusions.
Traumatic brain injury (TBI) is a major cause of death and disability. Diffuse axonal injury (DAI) is the predominant mechanism of injury in a large percentage of TBI patients requiring hospitalization. DAI involves widespread axonal damage from shaking, rotation or blast injury, leading to rapid axonal stretch injury and secondary axonal changes that are associated with a long-lasting impact on functional recovery. Historically, experimental models of DAI without focal injury have been difficult to design. Here we validate a simple, reproducible and reliable rodent model of DAI that causes widespread white matter damage without skull fractures or contusions.
Traumatic brain injury (TBI) is a major cause of death and disability in the United States. TBIs contribute to about 30% of all injury-related deaths1,2. The leading causes of TBI differ among age groups and include falls, high-speed collisions during sports, intentional self-harm, motor vehicle crashes and assaults1,2,3.
Brain diffuse axonal injury (DAI) is a specific type of TBI induced by rotational acceleration, shaking or blast injury of the brain resulting from unrestricted head movement in the instant after injury4,5,6,7,8. DAI involves widespread axonal damage leading to long-lasting neurological impairment that is associated with poor outcome, burdensome health-care costs, and a 33-64% mortality rate1,2,4,5,9,10,11. Despite significant recent research into the pathogenesis of DAI, there has not been a consensus on best treatment options11,12,13,14.
Over the last decades, numerous experimental models have attempted to accurately replicate different aspects of DAI11,12,15,16. However, these models have limitations given the unique presentation of DAI compared to other focal injuries. These prior models not only cause axonal injury in white matter regions but also result in focal cerebral injuries. Clinically, DAI is accompanied by micro hemorrhages, which may constitute a major cause of damage to white matter.
Only two animal models have been shown to replicate the key clinical features of DAI. Gennarelli and colleagues produced the first lateral head rotation device in 1982, using nonimpact head rotational acceleration to induce coma with DAI in a nonhuman primate model15. This primate model employed controlled single rotation for acceleration and deceleration to displace the head through 60° within 10-20 ms. This technique was able to emulate impaired consciousness and widespread axonal damage that resembled the effects of severe TBI observed in human brains. However, primate models are very expensive4,11,16. Based in part on the previous model, a pig model of rotational acceleration brain injury was designed in 1994 (Ross et al.) with similar results14.
These two animal models, though they produced different presentations of typical pathology, have added greatly to the concepts of DAI pathogenesis. Rapid head rotation is generally accepted as the best method for inducing DAI, and rodents provide a less expensive model for the rapid head rotation studies11,16. Here, we validate a simple, reproducible and reliable rodent model of DAI that causes widespread white matter damage without skull fractures or contusions. This current model will enable better understanding of the pathophysiology of DAI and development of more effective treatments.
The experiments were performed following the recommendations of the Declarations of Helsinki and Tokyo and to the Guidelines for the Use of Experimental Animals of the European Community. The experiments were approved by the Animal Care Committee of Ben-Gurion University of the Negev.
1. Preparing rats for the experimental procedure
NOTE: Select adult male Sprague-Dawley rats weighing 300-350 g.
2. Induction of diffuse axonal injury
NOTE: The device consists of the following components: 1) transparent plastic cylinder, 2) iron weight (1308 g), 3) rotation mechanism consisting of a cylindrical tube, two bearings upon which the axis rotate and a head fixation (for ear pins); 4) horizontal platform on which are fixed two bearings.
3. Measurement of rotational Kinematics/Biomechanical parameters.
4. Evaluation of Neurological Severity Score after 48 hours
NOTE: Neurological deficits were assessed and graded using a Neurological Severity Score, as previously described17,18,19. Alterations in motor function and behavior are assessed by a point-system such that a maximum score of 24 represents severe neurological dysfunction. A score of 0 indicates intact neurological status. The following behavioral functions are assessed.
5. Brain collection for histological examination after 48 hours
6. Immunochemical staining and examination
Table 1 illustrates the protocol timeline. The mortality rate in this model of DAI was 0%. A Mann-Whitney test indicated that neurological deficit was significantly greater for the 15 DAI rats compared to the 15 sham rats at 48 hours following intervention (Mdn = 1 vs. 0), U = 22.5, p < 0.001, r = 0.78 (see Table 2). The data are measured in counts and are presented as median and 25–75 percentile range.
Representative photomicrographs of thalamic sections of brain tissue are shown in Figure 1. Photomicrographs revealed axonal and neuronal βAPP immunoreactivities following isolated DAI in rats 48 hours post injury compared to the control group (67.46 ± 30 vs. 0 ± 0), U = 0, p < 1.1E-06, r = 0.92. The data are measured as counts and presented as mean ± SD.
Groups | Time | Procedures |
DAI (15 rats) | 0 h | Induction Diffuse Axonal Injury |
Sham (15 rats) | 48 h | Neurological Severity Score Assessment, |
DAI (15 rats) | Immunochemical staining of BAPP. |
Table 1: Demonstration of the protocol timeline. The various groups of rats at different times are shown: DAI = Diffuse axonal brain injury at the beginning of the experiment; At 48 hours, a Neurological Severity Score was determined and immunochemical staining of βAPP was performed in both groups.
NSS values of the various groups at 48 hours | ||
Animal Group | N | NSS 48 hours after DAI |
Sham | 15 | 0 (0-0) |
DAI | 15 | 1 (1-1)* |
Table 2: Neurological severity score. Neurological deficit 48 hours following DAI for 2 study groups. A Mann-Whitney test indicated that neurological deficit was significantly greater for the 15 DAI rats compared to the 15 sham rats at 48 hours following intervention (Mdn = 1 vs. 0), U = 22.5, p < 0.001, r = 0.78. The data are measured in counts and are presented as median and 25–75 percentile range.
Figure 1: Immunochemical examination. Representative photomicrographs of thalamic sections of brain tissue revealed axonal and neuronal immunoreactivities following isolated DAI in rats (B) 48 hours post injury compared to the control group (A). βAPP immunoreactivity was detected in the region of interest in all 15 DAI rats, and not at all in any of the sham-operated rats. Mann-Whitney test indicated that number of βAPP -positive axons was significantly greater for 15 DAI rats than for sham-injured animals at 48h following DAI (67.46 ± 30 vs. 0 ± 0), U = 0, p < 1.1E-06, r = 0.92. Images are at the original magnification * 200. The data are measured as counts and presented as mean ± SD. Please click here to view a larger version of this figure.
This protocol describes a rodent model of DAI. In DAI, rotational acceleration on the brain causes a shear effect that triggers axonal and biochemical changes that lead to loss of axonal function in a progressive process. Secondary axonal changes are produced by a rapid axonal stretch injury and are variable in their extent and severity4,5,10. Within hours to days after the primary injury, biochemical changes will lead to the loss of axonal function4,5,10. Following the injury, the permeability of the axon membrane changes, allowing a massive calcium influx. The intake of calcium causes the mitochondria to swell and break, releasing caspases and triggering caspase mediated progressive cell death4,5,10,11,20. Secondary axonal injury can present in the form of the axonal bulbs at the ruptured end or in the form of varicosities along the length of the axon4,21,22. The loss of nerve impulse transition is expressed by the aggregation of the β-amyloid precursor protein (βAPP), a single transmembrane protein present in most cells and tissues4,23,24,25,26. Immunohistochemical analysis of βAPP accumulation is currently the gold standard clinical and experimental technique for assessment of DAI4,9,10,20,27. Studies have reported βAPP immunoreactivity starting approximately 2 hours after injury, but there is evidence that ongoing changes continue for one or more years post-injury23,28,29. The most vulnerable areas are the brainstem, parasagittal white matter of the cerebral cortex, and corpus callosum11.
Common in vivo animal models of DAI are the lateral fluid percussion model30,31, the impact acceleration injury32,33 and the controlled cortical impact model34,35,36. These models provide some useful results but with significant limitations.
Fluid percussion models in animal models induce brain injury by injecting varying volumes of saline into the closed cranial cavity at the midline, especially in cat and rabbit models, or laterally in rodent models30,31. Injury severity can be varied from mild to severe by adjusting the fluid pressure. Although this model is reliable and reproducible, it is not an ideal model of human DAI, because percussion injury produces contusion and/or subarachnoid hemorrhage and the type of primary impact is distinct from real life injuries37,38. Furthermore, the effects of brain geometry and intracranial structure on direction, displacement and velocity make it very difficult to perform a precise biomechanical analysis of the injury39.
The impact acceleration injury model32,33 uses segmented brass weights free-falling from a specified height through a Plexiglas guide tube onto a metallic helmet fixed by dental acrylic to the skull vertex of the rat. This model is inexpensive, easy to perform, and can produce graded DAI, but there is also a possibility of contusions and skull fractures, compromising the reproducibility of the model. In addition, the induced injury involves a disproportionately smaller volume of the brain than in humans39.
The model of controlled cortical impact employs a pneumatic or electromagnetic impact device to drive a rigid impactor onto the exposed, whole dura through a unilateral craniotomy, which leads to deformation of the underlying cortex16,17. Air pressure is responsible for the impact velocity, and the depth of cortical deformation is regulated by vertical adjustment of the crossbar where the cylinder is attached. Like the fluid percussion model, it causes mainly focal injury.
Regarding these disadvantages, a new modified rodent model has been developed with opening of the dura mater over the contralateral hemisphere to produce more widespread axonal injury40. However, most previous models require craniotomy, and results of axonal pathogenesis may be affected by contusion and hemorrhage that usually appear in previous models. Moreover, the mechanism of injury in these models is different from the human DAI caused by the acceleration–deceleration movements of the brain.
There are several steps in the protocol that are critical and merit careful consideration. One should consider that head of the rat should be tightly fixed to the ear pins, or the rat may fall from device. When falling, other forces may play a role that will affect the accuracy of any calculations. Also, the iron weight must be the specific weight and dropped at the specific height noted in this protocol. These measurements have been determined empirically and are mandatory conditions for the reproduction of the model. The installation of the plastic cylinder should be at an angle of 90° relative to the rotational mechanism, namely the bolt. This is because it is the hit to the bolt that drives the rotational mechanism. Otherwise, the friction of the iron weight relative to the plastic cylinder is introduced, which will lead to a decrease in the force applied to the rat’s head.
There are some limitations to this model. The development of DAI in humans is mainly secondary to an impact from another object. In this case, either the person moves towards the object, the object moves towards the person, or they both move toward each other. In such a collision, a patient develops a combined head injury, where diffuse axonal damage is only part of the TBI. Here, the applied rotational acceleration is the main mechanism that leads to the development of DAI without others elements of head injury.
The model proposed here appears to alleviate the complications of skull fractures and contusions that caused widespread white matter damage without limited additional injury. Similar to other recent rodent models, this model is effective and provides a low (0%) mortality rate. It is a reproducible and affordable technique that could serve as a valuable resource for better understanding the pathophysiology of DAI to develop more effective treatments.
The authors have nothing to disclose.
The authors gratefully acknowledge Dr. Nathan Kleeorin (Department of Mechanical Engineering, Ben-Gurion University of the Negev) for his assistance with the biomechanical measurements. Also, we thank Professor Olena Severynovska, Maryna Kuscheriava, Maksym Kryvonosov, Daryna Yakumenko and Evgenia Goncharyk of the Department of Physiology, Faculty of Biology, Ecology, and Medicine, Oles Honchar Dnipro University, Dnipro, Ukraine for her support and helpful contributions to our discussions.
0.01 M sodium citrate | SIGMA – ALDRICH | ||
2.5% normal horse serum | SIGMA – ALDRICH | H0146 | Liquid |
4 % buffered formaldehyde solution | |||
Anti-Amyloid Precursor Protein, C – terminal antibodyproduced in rabbit | SIGMA – ALDRICH | Lot 056M4867V | |
biotinylated secondary antibody | Vector | BA-1000-1.5 | 10 mM sodium phosphate, pH 7.8, 0.15 M NaCl, 0.08% sodium azide, 3 mg/ml bovine serum albumin |
bone-cutting forceps | |||
DAB Peroxidase (HRP) Substrate Kit (with Nickel), 3,3’-diaminobenzidine | vector laboratory | ||
embedding cassettes | |||
ethanol 99.9 % | ROMICAL | Flammable Liquid | |
guillotine | |||
Hematoxylin | SIGMA – ALDRICH | H3136-25G | |
Hydrogen peroxide solution | Millipore | 88597-100ML-F | |
Isofluran, USP 100% | Piramamal Critical Care, Inc | ||
Olympus BX 40 microscope | Olympus | ||
paraffine | paraplast plus leica biosystem | Tissue embedding medium | |
phosphate-buffered saline (PBS) | SIGMA – ALDRICH | P5368-10PAK | Contents of one pouch, when dissolved in one liter of distilled or deionized water, will yield 0.01 M phosphate buffered saline (NaCl 0.138 M; KCl – 0.0027 M); pH 7.4, at 25 °C. |
Streptavidin HRP | ABCAM | ab64269 | Streptavidin-HRP for use with biotinylated secondary antibodies during IHC / immunohistochemistry. |
xylene |