Experimental diffuse brain injury results in regional alteration of gross vascular morphology independent of neuropathology

Abstract

Primary objective

A dynamic relationship exists between diffuse traumatic brain injury and changes to the neurovascular unit. The purpose of this study was to evaluate vascular changes during the first week following diffuse TBI. We hypothesized that pathology is associated with modification of the vasculature.

Methods

Male Sprague-Dawley rats underwent either midline fluid percussion injury or sham-injury. Brain tissue was collected 1d, 2d, or 7d post-injury or sham-injury (n=3/time point). Tissue was collected and stained by de Olmos amino-cupric silver technique to visualize neuropathology, or animals were perfused with AltaBlue casting resin before high-resolution vascular imaging. The average volume, surface area, radius, branching, and tortuosity of the vessels were evaluated across three regions of interest.

Results

In M2, average vessel volume (p<0.01) and surface area (p<0.05) were significantly larger at 1d relative to 2d, 7d and sham. In S1BF and VPM, no significant differences in the average vessel volume or surface area at any of the post-injury time points were observed. No significant changes in average radius, branching, or tortuosity were observed.

Conclusions

Preliminary findings suggest gross morphological changes within the vascular network likely represent an acute response to mechanical forces of injury, rather than delayed or chronic pathological processes.

1. Introduction

Traumatic brain injury (TBI) is a major cause of death and disability throughout the world ^1,2. Despite improvements in critical care, morbidity and mortality remain high. TBI is a heterogeneous disorder with varying symptoms based on mode and severity of injury. It can lead to short and long-term cognitive and behavioural deficits and increases the risk for neurodegenerative disease ^3–6. Injury can be divided into two phases: the primary mechanical damage is irreversible and only amenable to preventative measures to minimize the extent or incidence of injury including the wearing of seat belts and helmets and reducing vehicle speed; the second phase is a multifactorial process initiated at the time of impact and evolving over the subsequent hours to days. Secondary injury leads to physiological, cellular and molecular responses aimed at restoring homeostasis, which if not controlled, will lead to further injury.

The neurovascular unit of the brain is comprised of neurons, astrocytes, oligodendrocytes, microglia, endothelial cells and pericytes ⁷. Disruption to this neurovascular unit can lead to local and global events which affect neurological function and overall outcome ⁷. Processes involved in the secondary phase of injury are under ongoing investigation as targets for novel therapeutic strategies to limit delayed brain damage. These strategies often aim to reduce cell death, inflammation and behavioural morbidities ^8–12. However, limited literature is available on injury-induced alterations to the vasculature. Pathological changes to the vasculature, such as redirection of blood flow, blood-brain barrier (BBB) permeability, angiogenesis, and oedema may propagate secondary injury cascades and worsen outcome ¹³.

Midline fluid percussion injury (FPI) can produce a concussion-like diffuse brain injury in rats, characterized by brief neurological and systemic physiological alterations without remarkable structural damage ¹⁴. Furthermore, it has been demonstrated that this model is associated with decreases in cerebral blood flow ¹⁵. With decreased blood flow, endothelial cells of the BBB become distressed, leading to disruption of the BBB and the influx of solutes and proteins from the periphery, defined as vasogenic oedema ¹⁶. The period of vasogenic oedema is followed by cytotoxic oedema, where cells such as astrocytes take up excess fluid to restore homeostasis ¹⁶. With the transient opening of the BBB following diffuse brain injury ¹⁷, both vasogenic and cytotoxic oedema can occur ^18,19. Oedema has been associated with poor outcome following TBI, predominantly by raising intracranial pressure, impairing cerebral perfusion and oxygenation ¹⁶. These data suggest that vascular remodelling may be a meaningful component of delayed brain injury processes. To this end, a study of lateral FPI, where cavitation and contusion occur, reported a diffuse and heterogeneous change in the microvasculature ²⁰. Decreased capillary density and diameter as well as reduced perfusion of microvessels were reported. Regardless of injury severity, there was a reduction in the density of microvasculature at 1 day post-injury. However, by 2 weeks post-injury density was comparable to sham rats, indicative of recovery. The authors speculated that these changes in vasculature may exacerbate pathology and ultimately outcome.

On the other hand, midline FPI results in rapid secondary axotomy within the cortex and thalamus without cell death ^21–24. These physical forces which can tear axons may similarly tear blood vessels. Our previous studies using midline FPI have demonstrated temporal and regional pathology as well as circuit dysfunction in discrete regions such as the primary somatosensory barrel field cortex (S1BF) and ventral posteromedial thalamus (VPM) ^25–30. The pathways between these two regions control whisker somatosensation ^31,32. Damage to the S1BF and VPN manifests as an increased behavioural sensory sensitivity to manual whisker stimulation ^25,27. In fact, whisker stimulation in the chronic period following midline FPI produces functional hyperactivation of the S1BF and VPM ³³. Midline FPI-induced pathology in the thalamocortical pathway is associated with increased numbers of activated microglia, indicative of local inflammatory cascades, likely contributing to secondary injury processes ^30,34. Despite these overt changes to cells of the neurovascular unit, no research has investigated the morphological alterations to the vasculature in these regions following midline FPI. Here, we explore whether midline FPI modifies the vasculature in brain regions which are known to harbour pathology associated with behavioural morbidity. We hypothesize that pathology is associated with gross modification to the vasculature.

2. Methods

2.1 Surgical preparation and diffuse brain injury

Adult male Sprague-Dawley rats (350–375 g) were subjected to midline fluid percussion injury (mFPI) consistent with methods described previously ^{23,25–27,35,36}. Briefly, rats were anesthetized with 5% isoflurane in 100% O₂ prior to the surgery and maintained at 2% isoflurane via nose cone. Rats were placed in a stereotaxic frame and a midline scalp incision was made to expose the skull. A 4.8-mm circular craniectomy was performed (centered on the sagittal suture midway between bregma and lambda) without disrupting the underlying dura or superior sagittal sinus. An injury hub was fabricated from the female portion of a Luer-Loc needle hub, which was cut, bevelled, and scored to fit within the craniotomy. A skull screw was secured in a 1-mm hand-drilled hole into the right frontal bone. The injury hub was affixed over the craniotomy using cyanoacrylate gel and methyl-methacrylate (Hygenic Corp., Akron, OH) was applied around the injury hub and screw. The incision was sutured at the anterior and posterior edges and topical Lidocaine ointment was applied. Animals were returned to a warmed holding cage and monitored until ambulatory.

For injury induction, animals were re-anesthetized with 5% isoflurane 60–90 min after surgery. The dura was inspected through the injury-hub assembly, which was then filled with physiological saline and attached to the male end of the fluid percussion device (Custom Design and Fabrication, Virginia Commonwealth University, Richmond, VA). As the rat’s reflexive responses returned, a moderate injury (1.9–2.0 atm) was administered by releasing the pendulum onto the fluid-filled cylinder. Animals were monitored for the presence of a forearm fencing response as well as the return of the righting reflex as indicators of injury severity ³⁵. Sham animals were connected to the FPI device, but the pendulum was not released. The injury-hub assembly was removed en bloc, integrity of the dura was observed, and bleeding was controlled prior to the incision being stapled. Brain-injured animals had righting reflex recovery times of 6 to 10 minutes, indicating a moderate injury severity; sham-injured animals recovered within 15 seconds. Surgical recovery was monitored post-operatively for three days and no overt differences (e.g. weight, movement, grooming) were observed between animals.

2.2 Brain preparation for Aminocupric silver technique

Argyrophilic reaction product, which has come to indicate neuropathology, was examined using the de Olmos aminocupric silver histochemical technique as previously described ^26,37,38. Sham (n = 3) or brain-injured rats (n = 3 per time point; 1d, 2d, and 7d) were given an overdose of sodium pentobarbital (200 mg/kg i.p.) and transcardially perfused with 0.9% sodium chloride, followed by a fixative solution containing 4% paraformaldehyde and 4% sucrose in 0.1 M phosphate buffered saline (PBS). Following decapitation, the heads were stored in a fixative solution containing 15% sucrose for 24 hr, after which the brains were removed, placed in fresh fixative, and shipped for histological processing to Neuroscience Associates Inc. (Knoxville, TN). The rat brains were embedded into a single gelatin block (Multiblock Technology™; Neuroscience Associates Inc.), cut at 40 μm, stained with the de Olmos aminocupric silver technique according to proprietary protocols (Neuroscience Associates Inc.), counterstained with Neutral Red, and cover-slipped. Qualitative analysis of silver staining was conducted on photomicrographs which were acquired on a Zeiss Imager A2 microscope with AxioCam MRc5 digital camera. All three regions of interest (ROIs) were analysed at 20x magnification.

2.3 Brain preparation for vascular casting and blood brain barrier disruption

On 1d, 2d, or 7d post-injury or sham-injury, a separate cohort of animals (n = 3/group) were transcardially perfused with heparinized (100–200 U/ml) 0.9% saline followed by 10% neutral buffered formalin (NBF). Then the vascular imaging reagent AltaBlu™ (Numira Biosciences, Salt Lake City, UT), warmed to 50°C was perfused as per the manufacturer instructions. The perfused animal head was immersed in cold 10% NBF and stored in the refrigerator overnight to allow the AltaBlu to cure. The brains were then removed, placed in 10% NBF, and sent to Numira Biosciences (Salt Lake City, UT) for MicroCT vasculature imaging. Atlas registration was performed on each brain based on MicroCT data. Atlas data ³⁹ were loaded for the whole brain, as well as for the regions of interest (ROI: motor cortex (M2) region directly underneath the injury site; primary somatosensory barrel field (S1BF), and ventral posteromedial nucleus (VPM) associated with our prior work demonstrating both pathology and functional changes ^23,26,33; Figure 1C). Using the data from the MicroCT, each section of the brain was masked and the boundaries automatically detected. Sampled points from the boundaries of the MicroCT data and atlas template were used as control points for landmark-based registration. A transformation of the MicroCT image, using a second order polynomial transformation, was applied to the atlas ROI. A 3D ROI was produced based on the interpolation between the ROI slices as seen in Figure 1A and B. High resolution vasculature imaging was used to extract the average volume, surface area, radius, branching, and tortuosity (curvature or twisting) of the vessels in the three selected ROIs.

Figure 1.

Neurovascular imaging and regions of interest. Atlas registration was performed on each brain included in the vascular casting study. Registration was accomplished using MicroCT and an atlas template to identify anatomical landmarks. A 3D region of interest (ROI) was produced based on the interpolation between the ROI on individual sections/slices. (A, B) Axial and sagittal images, respectively, from a 3D model showing each ROI, based on atlas registration of the M2 (light blue), S1BF (red), and VPM (green). Analyses and imaging was performed by Numira Biosciences (Salt Lake City, UT). (C) Figure from The Rat Brain in Stereotaxic Coordinates, Paxinos and Watson (2007) showing a coronal view of ROIs using the same color scheme as the 3D images. Secondary motor cortex (M2); primary somatosensory barrel field (S1BF); ventral posteromedial nucleus of the thalamus (VPM); region of interest (ROI).

2.4 Statistical analysis

Statistical analyses on quantitative vascular data were performed using GraphPad Prism 6 (GraphPad Software, Inc., La Jolla, CA). One-way analysis of variance (ANOVA) was used to compare averages for sham (7d) and injured (1d, 2d, and 7d) animals on each measure of vascular morphology: volume, surface area, radius, branching, and tortuosity. Tukey’s HSD post-hoc analysis was performed where appropriate. Significance was set at p < 0.05.

3. Results

3.1 Neuropathology evidenced by argyrophilic staining

The amino-cupric de Olmos silver staining technique was used to identify neuropathology in brain regions of interest. Limited silver accumulation was noted in the sham animals, as seen in figure 2. The M2 region, located directly under the fluid pulse, did not show substantial argyrophilic reaction product at any of the three post-injury time points in brain-injured animals, as seen in figure 2. However, silver accumulation was present in the S1BF at all post-injury time points, with the greatest staining intensity observed at 7d post-injury. Consistent silver accumulation was seen in the VPM region at 1d and continued through 7d.

Figure 2.

De Olmos silver stain shows region-specific neuropathology. Deposits of aminocupric staining are indicative of neuropathology and show a regional and temporal pattern following diffuse TBI. As expected, no staining was observed in the sham animals. The M2 did not show evidence of neuropathology at 1d, 2d or 7d after fluid percussion injury (FPI). The S1BF and the thalamus, including the VPM, had silver stain visible at 1d which spread more diffusely over time post-injury and the intensity of staining increased over 2d and 7d. Dark and intense staining in S1BF and VPM suggests neuropathology and circuit disruption.

3.2 Acute change only in M2 vascular morphology after diffuse TBI

Uninjured and brain-injured animals were perfused with AltaBlu™ at 3 different time points post-injury to investigate vascular changes after diffuse brain injury. Vasculature was visualized and reconstructed via microCT. A one-way ANOVA was used to evaluate differences between sham and brain-injured groups for each measure of vascular morphology (average volume, surface area, radius, branching, and tortuosity) for each ROI, as shown in figures 3 and 4. Among the three ROI, only the M2 had significant injury-induced alterations to vascular morphology, and only in two of the five outcome measures: volume and surface area.

Figure 3.

Changes in neurovascular volume and surface area following diffuse TBI. Analysis of the vasculature in the secondary motor cortex (M2) region showed significant differences in vessel volume and surface area. Tukey’s HSD post hoc analysis revealed that both vessel volume and surface area were significantly increased in M2 at 1d after fluid percussion injury (FPI) relative to 2d, 7d and sham. There were no significant changes observed in vessel volume or surface area between 2d and 7d post-injury relative to sham or each other in M2. In the primary somatosensory barrel field cortex (S1BF) there were no significant changes in vessel volume or surface area (p > 0.05). Analysis of the ventral posteromedial thalamus (VPM) also failed to show any significant differences in vessel volume (p = 0.0573) or surface area (p = 0.3806). However, there was a similar, but non-significant, pattern of vascular change in vessel volume and surface area in the S1BF at 1d, 2d, and 7d post-injury. Data are presented as the mean ± SEM. *p < 0.05, **p < 0.01.

Average vessel volume in M2 was significantly increased at 1d compared to sham, 2d, and 7d groups (F(3, 8) = 8.551, p = 0.0071; Figure 3). There were no significant differences among sham, 2d and 7d time points. Changes in vessel volume were not observed in the S1BF (F(3, 8) = 2.775, p = 0.1104). There was an increase in vessel volume in the VPM, however, this was not statistically significant (F(3, 8) = 2.955, p = 0.0573). Average vessel surface area was significantly increased in M2 at 1d post-injury compared to sham, 2d, and 7d (F(3, 8) = 7.317, p = 0.0111) groups. No other comparisons were significant. No significant changes in surface area were found in the S1BF (F(3, 8) = 2.092, p = 0.1796) or VPM (F(3, 8) = 1.168, p = 0.3806; Figure 3).

For the average radius, average branching and average tortuosity, no significant differences were detected, as shown in table 1. In the M2 region, there was no significant difference between groups in the average radius (F(3,8) = 2.506, p = 0.1329), branching (F(3,8) = 0.9660, p = 0.4547), or tortuosity (F(3,8) = 0.3423, p = 0.7957). In the S1BF, there was no significant difference between groups in the average radius (F(3,8) = 1.368, p = 0.3205), branching (F(3,8) = 0.9212, p = 0.1586), or tortuosity (F(3,8) = 0.8992, p = 0.4827). Similarly in the VPM, there was no significant difference between groups in the average radius (F(3,8) = 0.7480, p = 0.5533), branching (F(3,8) = 1.569, p = 0.2711), or tortuosity (F(3,8) = 2.284, p = 0.1558).

Table 1.

Changes in vessel radius, branch number and tortuosity were not evident following injury. Neither diffuse brain injury (fluid percussion injury, FPI) nor sham-injury produced significant changes in the average vessel radius, branching or tortuosity at any of the post-injury time points in M2, S1BF or VPM (p > 0.05). Data are presented as the mean ± SEM.

Average radius, branching and tortuosity of vasculature in the M2, S1BF and VPM within the first week post-injury.

	Sham			1d FPI			2d FPI			7d FPI
	M2	S1BF	VPM	M2	S1BF	VPM	M2	S1BF	VPM	M2	S1BF	VPM
Average radius (μm)	27.34±0.8	28.41±2.2	29.05±2.4	27.34±0.8	29.66±0.7	28.07±1.7	35.23±4.6	33.67±3.5	29.53±1.5	27.09±1.7	28.26±1.0	26.23±0.8
Average branching (vessel branches/mm)	1.48±0.1	1.49±0.1	1.20±0.2	1.53±0.04	1.50±0.11	1.41±0.1	1.40±0.02	1.57±0.1	1.77±0.3	1.37±0.1	1.45±0.15	1.329±0.06
Average tortuosity	1.32±0.03	1.29±0.03	1.303±0.06246	1.31±0.02	1.29±0.03	1.33±0.02	1.30±0.02	1.34±0.02	1.44±0.04	1.30±0.03	1.29±0.01	1.34±0.004

4. Discussion

Clinically, disturbances to the vasculature are routinely noted following TBI. PET studies have allowed investigation into vascular dilation, cerebral blood flow as well as cerebral blood volume ⁴⁰. Changes to the vasculature are associated with cerebral oedema and poor outcome ¹³. Limited research has focused on vascular changes following diffuse brain injury. In this study, we examined morphological characteristics of the vasculature in three brain regions following diffuse TBI in the rat. The S1BF and VPM were selected based on demonstrated neuropathology corresponding to impairments in functional performance ^25–30, whereas the M2 was selected due to its location directly underneath the injury site. The primary aim was to determine if TBI-induced neuropathology, as evidenced by silver accumulation, was accompanied by alterations in vascular morphology. Contrary to our hypothesis, neuropathology was not associated with morphological changes in vasculature.

The primary injury associated with TBI initiates complex secondary injury cascades, which further exacerbate the initial injury. Secondary injury cascades include BBB permeability, inflammatory cascades, release of reactive oxygen species, neuropathology as well as oedema ^19,41–44. With increased permeability of vasculature and the breakdown of the BBB, vasogenic oedema occurs. As the secondary injury propagates, cellular metabolism is impaired leading to the disruption of sodium and potassium pumps, leading to cellular retention of sodium and water, or cytotoxic oedema ^45,46. These factors culminate in neuropathology. Silver-impregnation allows visualization of neuropathology ³⁸. Consistent with previous findings, the argyrophilic reaction product of silver staining has been quantified in the S1BF, with peak accumulation occurring at 7 days post-injury ²⁶, a region with astrocyte and microglia activation ^28,30,34. Moreover, silver accumulation has been documented to mirror FluoroJade staining, another marker for neuropathology, in both rats and mice following controlled cortical impact ⁴⁷. Therefore, in this study silver staining was used to indicate injury-induced neuropathology. Silver accumulation was observed over 7 days post-injury, particularly evident in the S1BF and to some extent the VPM. Compared to the S1BF and VPM, little evidence for neuropathology was observed in the M2 region directly beneath the injury site. The multifocal, diffuse nature of midline FPI is evidenced in the regional distribution of neuropathology, however the presence distal to the injury site remains a unique feature of this experimental model.

The vascular system is more heterogeneous in brain tissue than in other organs, with vessels possessing endothelial cell tight junctions that prevent or considerably hinder the influx of water-soluble molecules ⁴⁸. In rats, cerebral arteries and veins almost never run in pairs, although they connect through a complex network of capillaries ⁴⁸. The rat cerebrovascular system lacks direct communication between arteries and veins (A-V shunts), but retains arterial and venous anastomoses, which create considerable redundancy and precludes complete localized ischemia of the brain ⁴⁸. In the wake of diffuse brain injury, complete occlusion of vessels is unlikely and has yet to be documented, indicating that experimental TBI is discrete from experimental ischemia, despite shared cellular mechanisms. Unfortunately, the technique used in this study cannot distinguish arteries from veins and may not have been able to label capillaries exhaustively. However, it was able to show gross changes in the M2 vasculature following diffuse brain injury. These data therefore indicate further study is warranted to more thoroughly investigate vascular changes in veins, arteries and capillaries.

Reduced cerebral blood flow has been established as a component of contusion and pericontusional brain injury in the clinical population ⁴⁰. In the clinical population, the decreased blood flow in the contusion area occurred in the absence of changes to cerebral blood volume. However, in the pericontusional area reduced blood flow was coupled with reduced blood volume ⁴⁰. Taken together these data suggest changes to the vasculature, such as engorgement, can occur following brain injury. In the current study high resolution imaging and volumetric analysis was performed on the vascular casts in the M2, S1BF, and VPM regions at each post-injury time point. We expected to find evidence of injury-induced alterations in vascular morphology in regions with positive silver staining. For these reasons, the present approach was undertaken to observe gross changes in angiogenesis or vascular pruning, in addition to regional changes in vascular morphology. However, the only changes in vasculature were found at the earliest time point in the M2 region. Quantitative analysis of the vasculature at 1d post-FPI showed transient and significant increases in vessel volume and surface area that returned to sham levels by 2d and remained unchanged at 7d. One possible reason for the findings could be related to the proximity of M2 to the injury site. Others have demonstrated reduced vascular density in the ipsilateral cortex 1d post-lateral FPI ²⁰, which returned to uninjured levels by 2 weeks post-injury, suggesting injury-related vascular remodelling. Additionally, arterial vessels viewed through a cranial window adjacent to the injury site have been reported to dilate after midline FPI in the cat ^49,50, where vasodilation persisted through the 8-hour observation period. Increased vessel dilation could be directly related to the increase in volume and surface area observed in the present study. Of note, none of the vascular casts indicated overt haemorrhage or pooling of blood, even at the 1d post-injury time point, suggesting that any compromise to the BBB would be microscopic.

Delayed neuropathology may be secondary to vascular changes, such as cerebral blood flow and vessel occlusion. Reduced cerebral blood flow is a feature of experimental diffuse brain injury, with reports suggesting that following FPI there is a 40–50% reduction in cerebral blood flow which lasts approximately 4 hours ⁵¹. For that study, 4 hours post-injury was the latest time point examined; however, in areas of decreased blood flow, neuropathology and vascular alterations are likely to follow. Regional cerebral blood flow after lateral FPI recovered to near normal rates by 2 hours post-injury, everywhere but the injury site ⁵¹. This prolonged oligemia at the injury was associated with development of cystic necrosis at the injury site 4 weeks after injury ⁵¹. In the present study, no changes in vascular morphology were observed concomitant with neuropathology, since the time points chosen for analysis may be too late for the acute changes (2 hours) and too early for the chronic alterations (4 weeks).

Immediately following midline FPI in the cat, the pial arterioles dilate among other morphological and functional vascular abnormalities ⁵². The authors reported the vascular changes to be mediated by the generation of free radicals. Free radicals are known to propagate secondary injury cascades leading to neuropathology. In the current study, changes in vessel function were not assessed, however the same processes which contributed to silver accumulation (neuropathology) may equally impact vessel function. Moreover, vascular dysfunction in terms of altered vascular reactivity to vasodilators has also been reported ^53,54. These studies show altered vascular function for weeks post-injury, suggesting future studies need to examine vasoreactivity and functional changes in both the acute (<12 hours) and chronic phases post-injury. The relatively acute resolution of changes in vascular structure outlined in the current communication do not undermine the potential for long term functional vascular changes. Indeed, such vascular impairment is hypothesized to play a role in the brain’s increased vulnerability to secondary insults after TBI as the impaired vessels cannot adequately respond to additional challenges. Moreover, recent clinical data indicate that cerebral blood flow may serve as a marker for enduring neurological impairments after concussion ⁵⁵.

The current study assessed gross changes in vasculature via resin casting. This method may only detect gross changes with microscopic alterations being obscured by the 3D reconstruction procedures. Others have used microcorrosion casts in different cortical regions of human tissue to examine perivascular changes resulting from TBI ⁴⁶. Their immunohistochemical analysis of endothelial markers revealed loss of smooth normal endothelium. For the present study, regional analysis focused on areas of known neuropathology, given our hypothesis that underlying cellular processes would most likely contribute to gross vascular change; yet, no changes were evident. Further studies could pursue vascular changes at the microscopic level and incorporate functional assessments.

5. Conclusion

The acute increase in average volume and surface area among vessels in the M2 region likely resulted from mechanical injury associated with proximity to the primary insult. However, the morphological vascular response was transient, having recovered by 2d post-injury. Comparison of the vessel volume and surface area within S1BF and VPM across all time points failed to show any TBI-induced effects. These preliminary findings suggest that morphological changes within the vascular network likely represent an acute response to the injury, rather than delayed or chronic pathological processes. Further studies investigating injury severity and type would help decipher these injury processes. More detailed analysis of vascular networks may reveal subtle neovascularization and verify reduced vasoreactivity.