Evidence of Early Vasogenic Edema Following Minor Head Injury That Can Be Reduced With a Vasopressin V1a Receptor Antagonist

Background: Does minor head injury without signs of structural brain damage cause short-term changes in vasogenic edema as measured by an increased apparent diffusion coecient (ADC) using diffusion weighted imaging? If so, could the increase in vasogenic edema be treated with a vasopressin V1a receptor antagonist? We hypothesized that SRX251, a highly selective V1a antagonist, would reduce vasogenic edema in response to a single minor head injury. Methods: Lightly anesthetized male rats were subjected to a sham procedure or a single hit to the forehead using a closed skull, momentum exchange model. Animals recovered in ve min and were injected with saline vehicle (n=8) or SRX251 (n=8) at 15 min post head injury and again 7-8 hrs later. At 2 hrs, 6 hrs, and 24 hrs post injury, rats were anesthetized and scanned for increases in ADC, a neurological measure of vasogenic edema. Sham rats (n=6) were exposed to anesthesia and scannedat all time points but were not hit or treated. Images were registered to and analyzed using a 3D MRI rat atlas providing site-specic data on 150 different brain areas. These brain areas were divided into 11 major brain regions. Results: Untreated rats with head injury showed a signicant increase in global brain vasogenic edema as compared to sham and SRX251 treated rats. Edema peaked at 6 hrs in injured, untreated rats in three brain regions where changes in ADC were observed, but returned to sham levels by 24 hrs. There were regional variations in the time course of vasogenic edema and drug ecacy. Edema was signicantly reduced in cerebellum and thalamus with SRX251 treatment while the basal ganglia did not show a signicant response to treatment. Conclusion: A single minor impact to the forehead causes regional increases in vasogenic edema that peak at 6 hrs but return to baseline within a day in a subset of brain regions. Treatment with a selective V1a receptor antagonist can reduce much of the edema.

regional variations in the time course of vasogenic edema and drug e cacy. Edema was signi cantly reduced in cerebellum and thalamus with SRX251 treatment while the basal ganglia did not show a signi cant response to treatment.
Conclusion: A single minor impact to the forehead causes regional increases in vasogenic edema that peak at 6 hrs but return to baseline within a day in a subset of brain regions. Treatment with a selective V1a receptor antagonist can reduce much of the edema. Background Traumatic brain injury (TBI) affects more than 50 million people each year worldwide across all demographic and socioeconomic groups [1]. In the United States and Europe, TBI accounts for over 4 million emergency hospital visits annually [1][2][3]. Major traumatic brain injuriesare a contributing factor to a third of injury-related deaths in the United States and are one of the leading causes of death and disability in persons under 35 and over 65 [2]. In the United States alone, the annual economic burden of TBI is conservatively estimated at $86 billion [1]. While these statistics point to the signi cance of TBI as a health issue, one form of head injury, the effects of a single, minorimpact that doesnot require hospitalization or necessitate immediate medical attention, are not well understood [4]. The health community lists guidelines for diagnosing these"mild" head injuries that include self-reports of transient confusion, disorientation, impaired consciousness, or dysfunction in memory around the time of the injury and, importantly, no apparent structural damage as determined with imaging [4][5][6]. Despite these recognized symptoms, interventions other than rest with self-monitoring are rarely undertaken.
There areneurological complications following any closed head injury, including mild TBI [7,8]. Edema is among the most signi cant complications because it is a critical component in the pathophysiology of brain injury and occurs in stages. Vasogenic edema is caused by disruption of the blood-brain barrier (BBB) and, in moderate and severe TBI, is followed by cytogenic edema from cellular swelling and lastly, by a combination of both [9]. The timing and severity of each is dependent upon the level of brain injury.The extent of edema occurring after TBI can be determined by measures of apparent diffusion coe cient (ADC) using diffusion weighted imaging (DWI) [8]. ADC is a quantitative measure of water mobility. Vasogenic edema, characterized by higher ADC values, results from a change in permeability across the BBB, with an in ux of water and solute into the interstitial space contributing to the extracellular volume. This increase in brain water contributes to parenchymal swelling and increase in intracranial pressure. Cellular swelling due to loss of homeostatic regulation of osmolarity across the plasma membrane results in lower ADC values.
There are currently no approved pharmaceutical treatments that can prevent, minimize, or reverse brain edema following any level of brain damage [10,11].Several labs have studied the role of the neuropeptide arginine vasopressin (AVP) in intracranial water regulation [12][13][14][15][16]. AVP is a naturally occurring neuropeptide hormone involved in brain water permeability and vasoconstriction [17,18]. The AVP V1a (V1a) receptor subtype is highly expressed in cortical and subcortical brain areas across all mammals [19,20] and AVP signaling through this receptor is strongly tied to the pathology of cerebral edema in TBI.
For instance, V1a receptor density increases following TBI and infusion of V1a antagonists block increased intracranial pressure after head injury or brain lesion in animal models [21][22][23]. Knockout mice lacking V1a receptors show reduced edema, secondary injuries, and behavioral impairments in the days following brain damage [24]. We recently showed that systemic treatment with a highly selective V1a receptor antagonist for 5 days starting within 24 hrs of moderate/severe brain TBI signi cantly reduced edema and prevented cognitive de cits [25].
The present study was designed to address two questions: 1) Does minor head injury without signs of structural brain damage cause short-term changes in vasogenic edema as measured by an increase in ADC and,if so, 2) could these changes be mitigated by treatment with an AVP receptor V1a antagonist.

Animals
Male Sprague Dawley rats (n=22) weighing between 300 and 325gm were obtained from Charles River Laboratories (Wilmington,Massachusetts, USA). Rats were housed in Optirat® GenII (Animal Care Systems, Inc., Centennial, CO) cages with Bed-o'Cobs® corn-cob bedding (Anderson lab bedding, Delphi, IN) mixed with Envirodri™ shredded paper (Shepards Specialty Papers, Richland, MI). The vivarium was maintained atan ambient temperature of (22-24°C),humidity (35-45%) on a 12:12 light:dark cycle (lights on at 07:00 a.m.).Rats were allowed access to food and fresh water ad libitum. Rats were imaged during the light phase of the circadian cycle. All rats were acquired and cared for in accordance with the NIH Guide for the Care and Use of Laboratory Animals. All methods and procedures described below were preapproved by the Northeastern University Institutional Animal Care and Use Committee.

Treatment
Lightly anesthetized rats were subjected to a sham procedure or a single mild hit at 7m/sec to the forehead, midline and rostral to Bregma, using a closed skull, momentum exchange model [26]. Originally developed by Viano and colleagues to reliably in ict mild, medium, or severe rat head injury [27], this same model was further re ned by Mychasiuket al., and used to test the behavioral effects of TBI controlling for the axis of injury, rotational force, and head acceleration in different directions [28]. Animals were awake and moving in 5 min and were injected with saline vehicle (n=8) or SRX251 (n=8) at 15 min post head injury and again 7-8 hrs later. At 2 hrs, 6 hrs, and 24 hrs post head injury, rats were anesthetized and scanned for structural brain damage and increases in ADC, a neurological measure of vasogenic edema. Sham rats (n=6) were exposed to anesthesia and scanned at all time points, but were not hit or treated. Scanning sessions lasted about 70 min, after which animals were returned to their respective home cage.There was no evidence of structural brain damage in the 16 rats exposed to a single mild head injury during the 24 hr study period(Supplementary Data Fig S1) Neuroimaging Imaging sessions were conducted using a Bruker Biospec 7.0 T/20-cm USR horizontal magnet (Bruker, Billerica, MA, USA) and a 20-G/cmmagnetic eld gradient insert (ID=12 cm) capable of a 120-μs risetime. Radio frequency signals were sent and received with a quadraturevolume coil built into the animal restrainer (Animal Imaging Research,Holden, Massachusetts). The design of the restraining system included apadded head support obviating the need for ear bars, helping to reduce discomfort while minimizing motion artifact. All ratswere imaged under 1-2% iso urane whilekeeping a respiratory rate of 40-50 breathes/min. At the beginning of eachimaging session, a high-resolution anatomical data set was collected for assessment of structural damage using the RARE pulse sequence with followingparameters, 35  Diffusion Weighted Imaging -Quantitative Anisotropy DWI was acquired with a spin-echo echo-planar-imaging (EPI) pulse sequence having the following parameters: TR/TE = 500/20 msec, eight EPI segments, and 10 non-collinear gradient directions with a single b-value shell at 1000 s/mm 2 and one image with a B-value of 0 s/mm 2 (referred to as B 0). Geometrical parameters were: 48 coronal slices, each 0.313 mm thick (brain volume) and with in-plane resolution of 0.313×0.313 mm 2 (matrix size 96×96; FOV 30 mm 3 ). The imaging protocol was repeated two times for signal averaging. Each DWI acquisition took 35 min and the entire MRI protocol including the anatomy lasted about 90 min. Image analysis included DWI analysis of the DWI-3D-EPI images to produce the maps of ADC. DWI analysis was completed with MATLAB andMedINRIA (1.9.0; http://wwwsop.inria.fr/asclepios/software/MedINRIA/index.php) software. Because sporadic excessive breathing during DWI acquisition can lead to signi cant image motion artifacts that are apparent only in the slices sampled when motion occurred, each image (for each slice and each gradient direction) was screened prior to DWI analysis. If found, acquisition points with motion artifacts were eliminated from analyses.
For statistical comparisons among rats, each brain volume was registered to a 3D MRI Rat Brain Atlas (© 2012 Ekam Solutions LLC, Boston, MA) allowing voxel-and region-based statistics. All image transformations and statistical analyses were carried out using the in-house MIVA software (http://ccni.wpi.edu/). For each rat, the B 0 image was co-registered with the B 0 template (using a 6parameter rigid-body transformation). The co-registration parameters were then applied on the DWI ADC maps. Normalization was performed on the maps because they provided the most detailed visualization of brain structures and allowed for more accurate normalization. The normalization parameters were then smoothed with a 0.3-mm Gaussian kernel. To ensure that ADC values werenot affected signi cantly by the pre-processing steps, the 'nearest neighbor' option was used following registration and normalization.Statistical differences in measures of DWI between experimental groups were determined using two-way ANOVA followed by Bonferroni post hoc tests (alpha set at 5%).
The 3D MRI rat atlas has 173 segmented, annotated brain areas. For this study, 150areas were chosen for analysis because they could be organized into well-de ned neuroanatomical regions. Areas excluded were white matter tracts because they traverse several brain regions. Circumventricular organs, e.g. anterior and posterior pituitary, pineal gland, area postrema, median eminence and small adjacent areas like the arcuate and retrochiasmatic nuclei, were also excluded because of their larger, more fenestrated blood vessels.Also excluded were areas with no clear regional organization e.g., prerubral eld. The remaining 150 brain areas were divided into 11 brain regions: cerebellum (20), cortex (19), thalamus (20), basal ganglia (10), hypothalamus (14), hippocampus (9), prefrontal cortex (9), olfactory bulb/cortex (8), amygdala (8), midbrain/pons (12), brainstem (21). The areas and their average ADC values for each experimental condition and for 2, 6 and 24 hrs, is provided in Supplementary Data Excel S1. The organization was based on conventional neuroanatomy and an effort to keep individual brain areas localized and contiguous within a region. The olfactory bulb/cortex is the exception as the piriform cortex extends some distance caudally along the ventral lateral cortex away from the bulbs. To determine which areas of the brain were most sensitive to this single mild head impact directed to the forebrain, ANOVA's with post hoc analyses were run on the 11 brain regions. Only the cerebellum, thalamus, and basal ganglia showed a signi cant increase in ADC between experimental groups. Fig 2   shows scatter plots for ADC measures in the cerebellum. The two-way ANOVA results and list of twenty brain areas that comprise the cerebellum taken from the rat 3D MRI atlas are shown. At 2 and 6 hrs post impact, injured vehicle-treated rats present with signi cantly greater edema than sham (p<0.01), a difference that was no longer found at 24 hrs. At 6 hrs, ADC values peak in injured vehicle animals and signi cantly exceed the two hr levels (p< 0.02). SRX251 treatment signi cantly reduced edema at the 2 and 24 hr time points compared to vehicle (p< 0.001), but the effect was not signi cant at the 6 hr time point. There are no signi cant differences at any time point between SRX251 and sham rats.  list of ten brain areas that comprise the basal ganglia taken from the rat 3D MRI atlas. The change in ADC measures at 2 hr is not signi cantly different among conditions. By 6 hrs,injured vehicle-treated rats show signi cantly greater edema than sham but not SRX251. At 24 hrs, ADC values show no signi cant differences between conditions. At no time point is SRX251 signi cantly different from sham.

Results
The 3D reconstruction shown in Fig 5 summarizes the brain regions with signi cant increases in edema over the 24 hr period following impact to the forebrain (location depicted in black). The thalamus (light red) and cerebellum (dark red) were responsive to SRX251 treated and we found signi cantly reduced edema in these areas. While the basal ganglia only showed an increase in edema at 6 hrs, as noted above, a decrease in ADC was present in response to SRX2521 but did not differ signi cantly when compared to vehicle (area in blue).

Discussion
This was an initial study testing the effects of a single, minor head impact on the development of early edema in a closed head, momentum exchange model of brain injury and the ability of a V1a receptor antagonist to reduce or block observed changes. The ADC from diffusion weighted imaging was used as a proxy for vasogenic edema. In this model of mild head injury, edema was signi cantly increased in whole brain and a subset of the 11 regions that were tested, including basal ganglia, cerebellum, and thalamus, without evidence of structural damage. The observed edema was transient because at 24 hours post-injury, vehicle-treated rats with mild head injury did not differ from sham controls in ADC. V1a antagonist treatment signi cantly reduced edema compared to vehicle treatment in whole brain, thalamus, and cerebellum.
Edema is a critical component in the pathophysiology of TBI and can occur within the rst few hrs postinjury [29][30][31]. The type ofedema occurring after TBI can be determined using DWI with ADC, a quantitative measure of water mobility [29,[32][33][34].Vasogenic edema, characterized by higher ADC values, contributes to parenchymal swelling and an increase in intracranial pressure. In the present study, only higher ADC values were found in the affected regions, indicating that a single, mild head injury can induce vasogenic edema.
An important question is why the cerebellum and thalamus, regions caudal to the site of impact on the forehead, but not the prefrontal cortex immediately beneath the site of impact, show the greatest increases in ADC.The transmission of mechanical force down white matter tracts and along brain areas abutting the ventricular system may offer an explanation [35]. The cerebellum is recognized as beingparticularly vulnerable to mild TBI [36][37][38][39] andneuroradiological evidence of cerebellar dysfunction has beenadvanced as a diagnostic biomarker of TBI [39]. Because this is the only studywe knowof that used a closed-head, mild injury protocol, we cannot readily compare our ndings to other rodent models that producestructural brain damage.
The increase in ADC, which re ects vasogenic edema, suggests a change in the permeability of the BBB with a single mild head impact. BBB permeability increases within 1-4 hr hrs after brain damage in rodent models of TBI based on extravasation of blood biomarkers, e.g. Evan blue, plasma immunoglobulins, horse radish peroxidase, and albumin [40][41][42][43]. Similar results, albeit in moderate TBI models, have been reported in rats usingdynamic contrast enhanced (DCE) MRI to measure BBB permeability [44,45] and in humans, DCE MRI shows increased BBB permeability around damaged tissue within 24 hrs of injury that persists for days [46].Athletes playing American football and thought to have mild,sub-concussive head injuries show evidence of increased BBB permeability with DCE MRI [47]. Using DCE-MRI, O'Keeffe and colleagues reported BBB disruption in response to repetitive mild concussions in rugby players after a season and mixed martial arts ghters within days after a competitive ght [48]. There were no signi cant differences in self-reported concussions between these players and control subjects (track and eld athletes).Players with multiple subconcussive collisions in American football show elevated blood levels of astrocytic protein S100B thought to be due to BBB disruption [49]. Interestingly, professional boxers show an increase in ADC values in large white matter tracts and a general increase in brain diffusivity attributed to the cumulative effect of multiple non-severe head impacts [50][51][52].To our knowledge, these are the rst data reporting evidence of global and region-speci c increases in brain edema to a single, mild head injury in the absence of structural brain damage.
These data reinforce concerns expressed by others about the long-term effect of mild head injury [48]. For example, mild head injury begins with an impact that by itself may be recorded but not pursued as needing medical attention. Yet here we show there are global changes in vasogenic edema presumably caused by disruption in the BBB. Newsomeet al. [53]reported modest neurocognitiveabnormalities and increased functional connectivity in adolescent athletes with sports related concussions compared to age-matched, orthopedically injured controls. Despite these signs of altered cortical function, postconcussive symptoms were judged to be resolvedand athletes could return to competition. Indeed, more adolescent players are suffering from undetected neurological problems than previously thought [54]. A single season of high school American football producesDWI changes in diffusivity in the absence of clinical concussion [55]. Multiple mild head injuries are a risk factor for cognitive and emotional impairment and neurodegenerative disorders in athletes [56].
Evidence for the involvement AVP in the development of edema associated with TB is substantial. Following a moderate impact to the brain, the tissue surrounding the damage parenchyma increases the expression of V1a receptors on neurons, astrocytes and endothelium of microvessels, an effect that occurs within hrs and persists for days [57,58]. AVP appears to have a role in post-traumatic neuroin ammation as the proin ammatory chemokines in endothelial cells and astrocytes of damaged tissue are reduced in AVP-de cient Brattleboro rats compared to wild-type controls [59,60].With brain damage,the water content from vasogenic edema goes up dramatically in the rst 5 to 24 hrs [23,30,31].
The increase in brain edema can be reduced by the continuous intravenous infusion of SR49059, a V1a receptor antagonist,beginning immediately after head injury [22,23,30]. The intracerebroventricular injection of V1a receptor antagonists e.g., SR-49059, V1880 [61] or the peptidic deamino-Pen(1), O-Me-Tyr(2), Arg (8)]-vasopressin [31,62], are also effective but only when given within the rst few hrs of injury [61].Moderate head injury is associated with elevated levels of the water channel AQP4, GFAP, and V1a receptor, and disruption in sodium/potassium balance, all of which can be corrected by continuous exposure to SR49059 [22,23,30]. The edema in the these moderate TBI models is reduced in V1a null mice [24].In a recent study, we reported that AVN576, an orally active V1a receptor antagonist,givenipbeginning 24 hrs post injury for 5 days, could effectively reduce edema following TBI with brain damage [25].

Limitations and Data Interpretation
This was aninitial study with a limited number of animals to explore the possibility that we could measure signi cant changes in ADC following a single mild head impact. As such, there were several limitations: 1) The sensitivity of DWI did not allow us to examine 150 individual brain areas; instead, we were only able to observe change in ADC values in the whole brain and in three brain regions, cerebellum, thalamus, and basal ganglia;2) Would there have been sex differences if female rats also had been tested? We are aware of the many studies reporting sex differences in response to TBI, e.g. [63][64][65][66][67][68]. Recent reviews report there are male/female differences in morbidity and mortality following TBI [69,70]. However, the data collected over decades of preclinical and clinical research are inconsistent, certainly as it relates to the role of estrogen and progesterone [71]. Further, these initial studies were not designed to address the issues surrounding hormones and head injury;3) While there was no evidence of structural brain damage, post mortem histology looking at site-speci c or regional signs of in ammation or alterations in BBB permeability would have helpedto interpret the ndings. 4) Animals were exposed to light iso urane anesthesia four times over a 24 hr period. This necessary condition over this abbreviated time span for imaging precluded taking any meaningful measures of motor or cognitive behavior.

Conclusion
We were interested in two primary questions 1) could we detect global changes in brain edema using ADC, and 2) could these changes be mitigated by drug treatment with a vasopressin 1a antagonist?This study differs from previous animal TBI studies as it relies on imaging, without animal behavior, to evaluate the consequences of a mild head impact.The Glasgow Coma Scale of 13-15 de nes a head injury as "mild" based on measures of motor behavior, verbal responses and eye opening, with loss of consciousness and short hospitalization [72]. With CT and MRI, physical damage to the brain can be con rmed, adjusting the classi cation to "mild with complication" [73]. Organized sports in adolescence added another dimension to the characterization of "mild." The Centers for Disease Control and Prevention, World Health Organization and the Veteran's Administration have included self-reports of transient confusion, disorientation, impaired consciousness or dysfunction in memory around the time of the injury -importantly there should be no structural damage as determined with imaging [4][5][6]. This study in rats used a single head impact that caused no structural damage to the brain but clearly showed global and regional increases in vasogenic edema that appear within 2 hrs, increase at 6 hrs, but return to baseline within a day. Systemic treatment with a selective V1a receptor antagonist that crosses the BBBreduced much of the vasogenic edema. These ndings highlight the potential risk to a single mild head injury and the role AVP in acute vasogenic edema. 3D reconstruction of affected brain areas Shown is a color-coded reconstruction of brain areas showing signi cant increases in edema following a single mild hit to the forehead (black ball and circle) Figure 2 Edema in the basal ganglia Shown are scatter plots of average ADC values taken from the basal ganglia (10 areas shown below) for sham (black), vehicle (blue) and SRX251 (red) treatments 2, 6 and 24 hrs post head injury. ns -not signi cant, * p<0.05 Edema in thalamus Shown are scatter plots of average ADC values taken from thalamus (20 areas shown below) for sham (black), vehicle (blue) and SRX251 (red) treatments 2, 6 and 24 hrs post head injury. ns -not signi cant, *** p<0.001, ** p<0.01 Figure 4 Edema in cerebellum Shown are scatter plots of average ADC values taken from the cerebellum (20 areas shown below) for sham (black), vehicle (blue) and SRX251 (red) treatments 2, 6 and 24 hrs post head injury. ns -not signi cant, *** p<0.001, ** p<0.01 Figure 5 Edema in whole brain Shown are scatter plots of average ADC values taken from whole brain (150 areas) for sham (black), vehicle (blue) and SRX251 (red) treatments 2, 6 and 24 hrs post head injury. The analysis from a two-way ANOVA is shown below. ns -not signi cant, *** p<0.001, * p<0.05

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.