I mmmunotherapy in a Rodent Model of Traumatic Brain Injury Student: Amanda Weissman firstname.lastname@example.org College of Agricultural and Life Sciences CALS Honors Program Faculty Mentor: Dr. Kevin Wang
! I. Abstract Traumatic brain injuries (TBIs) affect millions of people yearly and can result in chronic post concussive syndrome, neuropsychological and cognitive deficits, or chronic traumatic encephalopathy. Clinical trials have found that the gliosis induced autoimm une response, which occurs through glial fibrillary acidic protein (GFAP) activation, generated after TBI results in a second brain injury. The purpose of this study is to prevent the autoimmune response by injecting mice with a GFAP antigen. The e xperimental group was injected with GFAP antigen, and then blood was collected. Next, TBIs were induced by performing a craniotomy then using the controlled cortical impact model. Post TBI, the Morris water maze (MWM) evaluated memory function, while the e levated plus maze (EPM) assessed anxiety levels. The mice then underwent perfusion, and their brains were dissected post mortem. Results were gathered to determine how autoimmunity was triggered, how to prevent autoimmunity from being triggered, an d on the behavioral tests. By injecting mice with GFAP antigens, it was discovered that autoimmunity by endogenous GFAP could be prevented to some extent. The MWM concluded immunized mice have better memory, and the EPM demonstrated non immunized mice are more anxious. Overall, developing an immunization for TBI would be beneficial for millions of at risk TBI candidates.
! # II. Introduction a. Traumatic Brain Injury Importance Sports related mild traumatic brain injuries (TBI) have increasingly attracted public health awareness (Harmon, 2013) The Center for Disease Control says that every year there are 1.6 to 3.8 million reported concussions occurring due to recreationa l and sports activities. Concussions can result in neuropsychological and cognitive deficits, chronic traumatic encephalopathy, or chronic post concussive syndrome. However, post concussive syndrome is usually hard to diagnose using the usual labora tory, clinical, or neuroimaging (Solomon, 2014) Post concussive syndrome includes dizziness, drowsiness, headaches, loss of memory and concentration, and anxiety. Currently, chronic traumatic encephalopathy can only be ultimately diagnosed after death. Even though most patients recover from concussions, there is increasing evidence showing that multiple concussions may be more severe and harder to recover from (Solomon, 2014) It is even shown that multiple mild traumatic brain injuries are linked t o developing chronic traumatic encephalopathy, which is a progressive degenerative disease. b. Mechanisms behind Traumatic Brain Injuries Increasing pathological changes associated with traumatic brain injury have also been reported ( Lucke Wold 2014) This includes long term brain volume loss and histological changes, involving Tau proteins, the hallmark of chronic traumatic encephalopathy, in the brain found in people with multiple brain traumas. Glial fibrillary acidic protein (GFAP) that is proteoly tically modified (GFAP BDP).
! $ During a brain injury, the protein Glial fibrillary acidic protein (GFAP) is also expressed in the central nervous system and has breakdown products (GFAP BDP) that are released following traumatic brain injury ( Kalisch 2006) Furthermore, significant cell death occurs within the first few days after the TBI incident, causing a release of brain proteins and the breakdown products into biofluids. Examples of breakdown products include cerebrospinal fluid (CSF) and circulating blood that is assisted by the brain blood barrier (BBB) ( Kalisch, 2006) It is already known that autoantibodies have pathological roles in neuroimmunlogical diseases. Autoimmunity in humans, however, has not been fully explored. Since experiments of the pathology of brain injury cannot be performed on humans, our research laboratory has developed an animal model to incite human repetitive concussive head injury. Through a pilot unbiased investigation involving the serum from post TBI patients, it has been discovered that a significant number of subacute sera from TBI patients contains an autoantibody response to a major astrocyte protein GFAP that is proteolytical ly modified, or GFAP BDP ( Zhang, 2014). The background for the hypothesis of our experiment is that the autoantibodies generated after the injury will cause brain damage ( Zhang, 2014). Protease modified GFAP could breakdown self tolerance and be a primary autoantigen to trigger the autoimmune response following TBI. Then, GFAP BDP induced autoimmune response will cause an increased and sustained glial cell and brain targeted neuroinflammatory response in conjunction with TBI. It will either cause GFAP activation, which is gliosis from astrocytes, or GFAP antibodies, which are an autoimmune reaction ( Zhang, 2014).
! % c. Research O bjective and Hypothesis: Zhang et. al. (2014) found that the gliosis induced autoimmune response generated after traumatic brain injury will result in a second injury to the brain Gliosis occurs due to GFAP activation. Our laboratory's hypothesis is that after inject ing mice with a GFAP antigen, the mice will develop antibodies, which will prevent an autoimmune response post TBI. After performing the controlled cortical impact (CCI) model, which induces the TBI, our lab conduct tests to see if the immunotherapy was successful. Included in these tests are behavioral tests and post mortem tests on the mice brains. III. Materials and Methods To examine this hypothesis, a study is conducted that involves testing clinical samples as well as conducting rodent autoimmune in vivo studies. The University of Flori da IACUC approved the animal testing protocols (Wang 2014; Wang, 2016) Our test subjects are mice. Initially, our laboratory injects the experimental group of mice with the GFAP antigen. Following the injection blood samples are collecte d pre injury Then our lab performs either closed or open head surgeries on mice, in which the position of the surgeon is alternated between two individuals, including myself. After the surgeries, behavioral tests are conducted to te st the functioning of mic e post TBI. Finally, the mice are killed and their brains are examined using various tests. a. Drawing Blood Our laboratory routinely draws blood from mice in their facial vein. First, the rodent is anesthetized briefly to ensure that it can be picked it up with ease (Golde, 2005). Next,
! & the rodent is placed in a prone position on a blue "diaper" pad that covers a heating pad, and the rodent must be picked up. Once the rodent has been grasped, use a gauge needle to pierce the freckle near the mouth, below the eye, and in front of the ear. It is important to break through the muscle to reach the facial vein, and it can be done by piercing the skin at an a ngle toward the ear going away from the mouth (Golde, 2005). Once the area is pierced, blood continually flows out and is collected into small, labeled test tubes. Lastly, use a swab stick to ensure the bleeding clots. b. Open Controlled Cortical Impact Model The purpose of the controlled cortical impact model is to create a head impact that will cause a traumatic brain injury. The mouse will be prepared in a different area than the surgical table to prevent contamination ( Yang 2014 ) This preparatory area involves the weighing, hair removal, and disinfection of the animal. The area is also covered with sterile drapes, in which the sterile instruments can be placed during surgery, preventing contamination. For the actual animal surgery, the animal is pl aced into the stereotaxic device. Then the surgeon puts on the sterile gloves. The surgical area is also disinfected, as well as the surgical equipment. To begin the open head controlled cortical impact surgery, the adult rodent will be initially an esthetized with 4% isoflurane for 4 to 5 minutes ( Yang 2014 ) Then, anesthesia is utilized for the entire duration of the procedure at 2 3% isoflurane Next, the hair over the skull will be shaved and the animal will be mounted in a stereotaxic fr ame in a prone position and secured by blunt ear and incisor bars ( Yang 2014 ) The rodent will be maintained under anesthesia during all procedures, and a heating pad covered with a blue "diaper" pad will be placed under the body. This
! maintains the core body temperature of the rodent at 37 degrees Celsius. Throughout the procedure, a rectal thermistor monitors the rodent's temperature to ensure the body temperature does not fluctuate more than one degree Celsius in either direction from 37 degrees. Before any procedures, it is important to make sure the animal is fully sedated using tail and/or toe pinches ( Yang 2014 ) Then sterile eye lubricant is placed in the rodent's eyes. Before beginning the first surgical incision, the head is sterilized wit h germicidal povidone iodine scrub, followed by a 70% isopropyl alcohol rinse. This combination for sterilization is used three times. It is important to note that swab sticks are dipped into the solutions then placed on the rodent head in circular motions starting from the center and circling outward. To begin the surgical procedure, a midline cranial incision of the soft tissues will be made ( Yang 2014 ) Then a unilateral craniotomy will be performed midway between the lambda and bregma, adjacen t to the central suture The dura matter will be kept intact over the cortex. For the impact to be successful, the clip contact sensor must be connected to the mouse's tail for the stereotaxic device to connect to the control box ( Yang 2014 ) Brain trauma will be produced using a Head Impactor with a 4 mm diameter impactor tip that goes a speed of 5m/s. The compression depth is at 1.8mm, and there is a 200ms dwell time. The impactor machine cannot be sterilized due to the sensitive nature of the controlling electronics. However, the impactor tip will be initially steam sterilized together with the surgical instruments and re sterilized in a dry sterilizer between each animal.
! ( The craniotomy and CCI may cause occasional bleeding and will b e controlled with sterile gauze ( Yang 2014 ) Once the bleeding is controlled, 4 0 Nylon sutures are used to close the skin incision. Typically, the entire surgical procedure lasts no longer than 15 minutes. After closing the skin incision, the animals wil l be placed on disposable blue "diaper" pads with a heating pad kept at a low temperature underneath. Animals typically recover from anesthesia and CCI surgery in from 5 to 30 minutes ( Yang 2014 ) After the recover, the rodents will be placed back in a new cage and returned to the animal facility in a few hours. If mice exhibit signs of pain, 1 2mg/kg of Meloxicam is injected every 12 to 24 hours. Finally, the mice are given additional injuries on days 4, 7, and 10 after the initial injury. c. Morris Water Maze Behavioral Test The Morris water maze behavioral test measures memory function. The actual water maze consists of a 4ft diameter pool filled with water and clouded by non toxic paint (Yang, 2015). Next, the pool is divided into four quadr ants. Each quadrant contains a platform position equidistant from the center to the wall. The pool is filled to 1cm below the surface of the water during cue training. Cue training helps assess the motor ability and visual acuity of mice as they are trying to escape the platform. This training is independent of their spatial learning ability, and mice perform tests where the platform and start positions vary with each trial. To assess memory function, the platforms are hidden (Yang, 2015). There is o ne test where the platform remains in the southwest quadrant, and another test where the platform is constantly moved throughout the trial. The mouse is placed in the same pool of water as the pool for cue training, so it has the same visual cues (for exam ple, colored
! ) shapes). The goal of the test is for the mouse to find the hidden platform, which is usually about 1cm below the surface of the water. In between mice, the water is stirred to prevent the following mouse from picking up the previous mouse's sc ent. One person releases the animal into the pool, while the other person records the time it takes for the mouse to find the platform using a computer based video tracking system (Yang, 2015). The video tracking system records the time it takes for the mouse to reach the platform, as well as the time spent in each quadrant of the pool. The start locations are always randomized for each trial the mouse participates in. Depending on which platform test is being performed, the platform is eithe r randomized or left in the southwest quadrant of the pool (Yang, 2015). Each mouse performs four run per block of trials. There are three blocks of trials per day, and the test is performed for four days. When the platform is in the southwest quadrant, on the fourth run, the platform is removed. The goal of this run is to test the memory of the mouse and see if it remains in the southwest quadrant longer than the other quadrants. If a mouse cannot find the visual platform for three out of five trials, it i s excluded from future studies. Data from the Morris water maze behavioral test provides insight on affects the traumatic brain injury has on the behavior of each mouse. d. Elevated Plus Maze Behavioral Test The elevated plus maze consists of two closed arms and two open arms (Yang, 2015) Mice that avoid the open arms of the plus maze whether they have decreased entries or spend less time in the open arms, are thought to have heightened anxiety levels. Mice were placed individually in the center of a 33cm long and 5 cm wide maze, with 25 cm high walls on closed arms, and allowed free access for 5 min utes The mice spent
! *+ time either in the closed arms or the "safe area", in the open arms, or in the middle, intermediate zone. Each session was videotaped with a compu ter based video tracking system for later analysis by an observer blind to the experimental treatment. The apparatus is wi ped with 70% ethanol and air dried between mice. The system records the moving distance and time spent in the open arms of the maze Then the data is analyzed with Student's t test. e. Perfusion CSF, blood, and tissue collection are done while performing mice perfusion. The animals die after comple tion of the procedures of CSF and blood collection. After the mice die, the tissues are collected. CSF will be collected from the mice after drug injections. The mice will be anesthetized at specific time points ( Yang 2014 ) The rodents are placed in a stereotaxic instrument with a longitudinal axis that is available for their heads to move freely. Once placed in the device, it is important that the neck is prominent by flexing the head to expose the external occipital protuberance. Isoflurane is d elivered throughout the procedure using a nose cone. Additionally, the mouse's body is on top of a blue "diaper" pad that covers a heating pad, so the rodent's core body temperature does not fluctuate. Before beginning any procedures, it is necessary to pe rform toe or nail pinches to check that the mouse is fully sedated. Once the mice do not respond to the pinches, the rodent head is shaved. When they do not respond to the pinches, the rodent's stomach hair is shaved and cleaned with iodine. A 25 ga uge needle will be attached to polyethylene tubing, and lowering the needle into the cisterna magna will collect CSF ( Yang 2014 ) Then, 5 10 l of CSF will be
! ** frozen on dry ice. Using a cardiac puncture technique, enough blood is collected to terminate the rodent while under anesthesia. Then, immediately after CSF collection, the mouse is flipped over with the nose cone still emitting isoflu rane. Next, the chest cavity is fixed open with locking forceps to expose the heart using scissors to make an incision Using a 10 cc syringe w ith 25G needle, about 0.5 to 1 ml of blood will be carefully collected from the heart ventricle. At the end of the procedure, the collected blood is either places in a tube for serum or a tube for plasma. f. Brain Dissection Bra in dissection occurs once the mice are dead to examine brain tissue. After cardiac perfusion, the rodent's head will be severed from its body Of course, the mouse is still under anesthesia to avoid any pain. Next, the brain is removed from the mouse's sku ll carefully to not pierce any parts of the brain. Then, the hippocampus, stratum, c ortex, and cerebellum are dissected rapidly to preserve these parts of the brain ( Yang 2014 ) Each brain part is placed in its own test tube, and then all test tubes are rinsed in a cold phosphate buffered saline and snap frozen in liquid nitrogen. IV. Results a. How Autoimmunity is Triggered To confirm Zhang and his colleagues' (2014 ) hypothesis that the autoantibodies generated after the injury will cause brain damage mice were tested to see how autoimmunity was triggered. Using the open head controlled cortical impact model, head impacts were delivered, leading to TBI Then, the mice were perfused and their brains were dissected. Using an immunohistochemistry test pe rformed by the lab manager, GFAP patterns in injured and activated glia cells were found (Figure 1).
! The results compared the controlled mouse cortex and the experimental mouse cortex one day after experimental TBI, showing injured glia cells, and three days after experimental TBI, showing activated glia cells. The GFAP was stained red, while the counterstain was DAPI ( 4',6 diamidino 2 phenylindole ) for nuclear DNA (blue) Additionally, western blotting of the mouse cortexes was conducted by the lab manager after brain dissections to reveal increased GFAP patter n s as days increased up to fourteen days after TBI before slightly decreasing (Figure 2). However, the western blot test also revealed that the GFAP protein developed cleaved fragments or BDP, after one to three days GFAP BDP then continued to increase until fourteen days and slightly decreased by 28 days. This data suggests the overexpression of GFAP proteolysis and gliosis and that GFAP BDP accumulates over time post TBI. Our lab assumed that the GFAP antibodies generated after an impact to the head caused TBI and want to confirm this assumption in future studies b. How to Prevent Autoimmunity from being Triggered To prevent autoimmunity from being triggered, experimenta l mice were injected with the GFAP antigen pre TBI. After a month of allowing the body to react to the TBI, perfusion was performed. The mice brains were dissected and the hippocampus was sliced and stained (Figure 4). The lab manager then performed a western blot to compare immunized mice with control mice. The ipsilateral hippocampus (IH) is the injured side of the hippocampus, while the contralateral hippocampus (CH) is the non injured side of the hippocampus. Thus, the data suggests that the experimental group had less GFAP in the hippocampus, which means the injection prevented autoimmunity to an extent.
! *# The data from the staining and western blot showed that mice injected with the antigen before TBI could generate antibodies to f ight off the antigens after TBI (Figure 3) Therefore, there was less cell death. The data was found using spectrin, which is a cell death marker. spectrin revealed that the 120KD bands, which show apoptosis, reduced over 7 days in the immunized mice. Also, spectrin revealed that the 150KD/145KD bands, which show cell necrosis, also reduced slightly over a 7 day period. This data suggests that immunized mice have reduced cell death overtime post TBI. c. Elevated Plus Maze Behavioral Test Since there was no difference between the control and experimental group for the distance moved and velocity moved, then it can be concluded that the injection does not affect motor function (Figure 5A and 5B). In other words, there was no motor fun ction deficit in the immunized mice. To test for anxiety post TBI, the elevated plus maze was utilized. After a TBI, anxiety increases. If the mouse spent more time in the open arms of the maze or travel more frequently to the open arms, it means those mic e are less anxious. The data showed that mice who received the injection pre injury were less anxious and traveled to the open arms more often (Figure 5 C and 5D ) Meanwhile, mice without the injection pre injury stayed in the closed arms more frequently to feel safer. d. Morris Water Maze Behavioral Test The goal of the Morris water m aze (MWM) is to test for memory function post TBI. Since there was no difference between the control and experimental group for the velocity moved, then it can be conclude d that the injection does not affect motor function (Figure 6A) After conducting many runs, the data showed that the injection did not
! *$ significantly improved spatial learning in the mice post TBI (Figure 6B) However, it was shown that immunized mice have a better memory (Figure 6C) Immunized mice spent more time in the quadrant that the platform should ha ve been in once it was removed. V. Discussion For our lab to conduct our experiment, we verified and agreed with Zhang et al. (2014) and found that the autoantibodies did cause brain damage following TBI. By conducting CCI on either immunized or non immunized mice our lab demonstrated that it is possible to prevent the triggering of autoimmunity by reducing endogenous GFAP Prevention is accomplished by injecting the GFAP antigen pre injury so the body can build antibodies and fight against future antigens, similar to the goal of a vaccine. The subcutaneous injections of GFAP antigens on experimental mice decrease d GFAP gliosis and pro teolysis post TBI. This conclusion was found by confirming the hypothesi s through collecting blood post injection and running it through the manifold immunoblotting system to examine if
! *% antibodies formed 14 days after the immunization (Diagram 1). Then, CCI was performed to induce TBI and after 28 days of recovery and behavioral tests, perfusion and brain dissection occurred. The antibodies formed in the immunized mice could bind to the GFAP and GFAP BDP antigens released p ost TBI which decrease d brain damage post TBI. Additionally, the behavioral tests suggested that immunized mice had better memories and were less anxious than non immunized mice. Overall, our lab concluded that GFAP imm unization could be beneficia l for those who are at risk of TBIs. Further studies must be done to assure the GFAP immunization causes no detrimental negative side effects and to ensure that no other proteins are involv ed in causing brain damage post TBI. In this study, we only tested for one month post injury. Therefore, additional studies should be conducted over longer periods of time, closer to 6 months, to verify the immunotherapy benefits on the autoimmunity in a mouse model. Once these studies are concluded, clinical tria ls must be introduced to willing participants at risk for TBI. Nevertheless, work on this immunization is important because TBI affects millions of people and can lead to irreparable damage.
! *& VI. Literature Cited Golde, W. T., Gollobin, P., & Rodriguez, L. L. (2005). A rapid, simple, and humane method for submandibular bleeding of mice using a lancet. Lab animal 34 (9), 39. Harmon, K. G., Drezner, J. A., Gammons, M., Guskiewicz, K. M., Halstead, M., Herring, S. A., ... & Roberts, W. O (2013). American Medical Society for Sports Medicine position statement: concussion in sport. British journal of sports medicine 47 (1), 15 26. Kalisch, R., Schubert, M., Jacob, W., KeÂ§ler, M. S., Hemauer, R., Wigger, A., ... & Auer, D. P. (2006). Anxi ety and hippocampus volume in the rat. Neuropsychopharmacology 31 (5), 925 932. Lucke Wold, B. P., Turner, R. C., Logsdon, A. F., Bailes, J. E., Huber, J. D., & Rosen, C. L. (2014). Linking traumatic brain injury to chronic traumatic encephalopathy: ide ntification of potential mechanisms leading to neurofibrillary tangle development. Journal of neurotrauma 31 (13), 1129 1138. Solomon, G. S., & Zuckerman, S. L. (2015). Chronic traumatic encephalopathy in professional sports: retrospective and prospective views. Brain injury 29 (2), 164 170. Wang, K., et al ( 2014). IACUC Study 201507692 Wang, K., et al ( 2016). IACUC Study 201308116 Yang, Z., Lin, F., Robertson, C. S., & Wang, K. K. (2014). Dual vulne rability of TDP 43 to calpain and caspase 3 proteolysis after neurotoxic conditions and traumatic brain injury. Journal of Cerebral Blood Flow & Metabolism 34 (9), 1444 1452. Yang, Z., Wang, P., Morgan, D., Bruijnzeel, A. W., Lin, D., Pan, J., ... & Febo, M. (2015). Temporal MRI characterization, neurobiochemical and neurobehavioral changes in a mouse repetitive concussive head injury model. Scientific reports 5 11178. Zhang, Z., Zoltewicz, J. S., Mondello, S., Newsom, K. J., Yang, Z., Yang, B., ... & Heaton, S. (2014). Human traumatic brain injury induces autoantibody response against glial fibrillary acidic protein and its breakdown products. PloS one 9 (3), e92698.
! *' VII. Figures a. How Autoimmunity is Triggered i. Figure 1 ii. Figure 2
! *( b. How to Prevent Autoimmunity from being Triggered i. Figure 3 ii. Figure 4
! *) c. Elevated Plus Maze Behavioral Test i. Figure 5 d. Morris Water Maze Behavioral Test i. Figure 6
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`ZM % ;\]1&10)#a34'(5234'2)/3 % %