Engineered glycomaterial implants orchestrate large-scale functional repair of brain tissue chronically after severe traumatic brain injury

 WHOM very specifically do I have to beat over the head to get research going on this for stroke in humans?

Engineered glycomaterial implants orchestrate large-scale functional repair of brain tissue chronically after severe traumatic brain injury

Charles-Francois V. Latchoumane 1,2, 

Martha I. Betancur 3 , 

Gregory A. Simchick 4,5, 

Min Kyoung Sun1,6, 

Rameen Forghani 1 , 

Christopher E. Lenear 1,2, 

Aws Ahmed 1,2, 

Ramya Mohankumar 1 , 

Nivedha Balaji 1 , 

Hannah D. Mason 1 , 

Stephanie A. Archer-Hartmann 7 , 

Parastoo Azadi 7 , 

Philip V. Holmes 6,8, 

Qun Zhao1,4,5, 

Ravi V. Bellamkonda 3 , 

Lohitash Karumbaiah 1,2,6* Severe traumatic brain injury (sTBI) survivors experience permanent functional disabilities due to significant volume loss and the brain’s poor capacity to regenerate. Chondroitin sulfate glycosaminoglycans (CS-GAGs) are key regulators of growth factor signaling and neural stem cell homeostasis in the brain. However, the efficacy of engineered CS (eCS) matrices in mediating structural and functional recovery chronically after sTBI has not been investigated. We report that neurotrophic factor functionalized acellular eCS matrices implanted into the rat M1 region acutely after sTBI significantly enhanced cellular repair and gross motor function recovery when compared to controls 20 weeks after sTBI. Animals subjected to M2 region injuries followed by eCS matrix implantations demonstrated the significant recovery of “reach-to-grasp” function. This was attributed to enhanced volumetric vascularization, activity-regulated cytoskeleton (Arc) protein expression, and perilesional sensorimotor connectivity. These findings indicate that eCS matrices implanted acutely after sTBI can support complex cellular, vascular, and neuronal circuit repair chronically after sTBI. INTRODUCTION Severe traumatic brain injuries (sTBIs) caused by blunt force or penetrating trauma to the brain lead to extensive tissue loss and lifelong disabilities (1). Although sTBIs account for only 10% of the approximately 1.7 million TBI cases reported in the United States annually, they are responsible for over 90% of all TBI-associated costs (1). There are no effective treatments to prevent cognitive impairments and tissue loss encountered after sTBI. Neural stem cells (NSCs) have long held a privileged position in neural repair strategies for their ability to mediate neuroprotective “bystander” signaling (2). However, inadequate control over NSC differentiation and the risk of immune rejection of xenografted NSCs pose serious limitations for their clinical application (3, 4). Although autologous cell therapies could help mitigate some of these concerns (5), issues related to poor survivability of transplanted cells (6), and the risk of tumorigenesis in the case of human embryonic and pluripotent stem cells (7, 8), pose significant barriers to clinical success. Chondroitin sulfate proteoglycans (CSPGs) are major constituents of the brain extracellular matrix (ECM) and stem cell niche (9–15). CSPG-linked sulfated CS glycosaminoglycan (CS-GAG) side chains regulate growth factor signaling (16–19), neuronal development (20), and neuroplasticity (21, 22). Fibroblast growth  factor 2 (FGF2) and brain-derived neurotrophic factor (BDNF) bind with high affinity to sulfated CS chains via specific sulfation motifs that act as molecular recognition sites (23). FGF2 is expressed by NSCs and radial glial cells found in neurogenic niches of the brain (24), where it is known to promote neuroprotection (25), NSC migration and proliferation (26, 27), and neurogenesis (24, 26, 28). BDNF is an important mediator of neuroplasticity at all stages of brain development (29) and is known to promote neuroprotection (30) and functional repair of the injured brain (29, 31, 32). FGF2 and BDNF also promote angiogenesis after injury (16, 33–35) and are critical for the maintenance of oxygen perfusion and tissue viability after injury. The development of acellular three-dimensional (3D) constructs that can potentiate cellular and functional brain repair has received much attention (18, 19, 23, 36, 37). A range of natural (38–41) and synthetic (42, 43) polymer composites, as well as natural ECMderived biomaterials (44, 45) have been used for mediating brain tissue repair. When compared to these approaches, sulfated CS-GAGs have native functionality and sulfation motif–dependent high affinity to neurotrophic factors (18, 19, 23, 37), making them ideal materials for mediating signal transduction of neurotrophic factors and complex endogenous tissue repair chronically after sTBI. In recent studies, acutely implanted 3D CS-GAG scaffolds with and without encapsulated NSCs enhanced the efficacy of endogenous NSCs and mediated neuroprotection 4 weeks after sTBI (46). However, the chronic functional implications of acellular, neurotrophic factor functionalized engineered CS (eCS) matrix implants after sTBI have so far not been assessed. eCS matrix implants that are designed to match the composition and biophysical properties of native brain ECM present an attractive approach for the treatment of sTBIinduced tissue and functional loss. In this study, we conducted a systematic assessment of the longterm outcomes of implanting FGF2 and BDNF-laden sulfated eCS matrices in rats subjected to sTBI. We performed magnetic resonance imaging (MRI) phase gradient analyses of regional cerebral blood flow (rCBF) and tissue vascularization in conjunction with longitudinal behavioral performance and functional recovery, followed by detailed terminal immunohistochemical assessments of tissue-specific biomarkers. We also quantified activity-related cytoskeletal associated protein (Arc) expression and volumetric vasculature tracing in cleared brain tissue to localize newly formed functional clusters of neurons intra- and perilesionally in sTBI rats implanted with eCS scaffolds when compared to controls. RESULTS Acutely implanted eCS matrices promote chronic neuroprotection and recovery of gross motor function To assess the chronic neuroprotective effects of neurotrophic factor– laden eCS matrix implants (fig. S1), we performed controlled cortical impact (CCI)–induced sTBI in the caudal forelimb area (CFA) in rats and implanted eCS matrices 48 hours after CCI in the lesion epicenter. We performed longitudinal assessments of motor function and MRI-based quantification of tissue volume loss and conducted terminal histological quantification of injury volume (Fig. 1A). While general locomotion was found to be similar across all groups (fig. S2A), we observed that the performance of eCS matrix– implanted rats was found to be comparable to Sham controls, whereas sTBI rats demonstrated significant motor deficits (Fig. 1B and fig. S2B). From results of longitudinal T2-weighted MRI at 4- and 20-week time points and lesion volume analyses (Fig. 1, C and D), as well as terminal Nissl staining of tissue and quantification at 20 weeks after sTBI (Fig. 1, E and F), we were able to detect the significantly greater tissue atrophy in the sTBI-only animals when compared to eCS matrix–implanted rats and Sham controls (Fig. 1, D and F). These results demonstrated that eCS matrix–implanted rats exhibited tissue and motor function recovery that was comparable to Sham control group animals chronically after sTBI. eCS matrices mediate enhanced proliferation of endogenous NSCs and neurotrophic factor expression CS matrices have been demonstrated to promote FGF2-mediated neural progenitor homeostasis and neuroprotection 4 weeks after sTBI (17, 46). Considering these previous observations, we quantified the effects of neurotrophic factor–laden eCS matrices on the chronic maintenance of proliferating NSCs in the lesion site using Ki67 and Sox-1 biomarkers (Fig. 2, A and C) and CS-mediated FGF2 retention (Fig. 2, B, D, and E). Endogenous NSC proliferation in brain tissue from eCS matrix–implanted rats was found to be comparable to Sham controls (Fig. 2C and fig. S3, A, C, and D) but was significantly decreased in sTBI-only rats. Although a significant decrease in Ki67+ cells in tissue from eCS-treated animals was observed when compared to Sham and TBI controls (fig. S3, A, B, and E), the significantly increased percentage of Sox-1+ Ki67+ cells in eCS matrix–treated animals when compared to TBI controls (Fig.  2C) suggests that NSCs do not contribute to the observed overall decrease in Ki67+ cells. These findings were further corroborated by FGF2 expression levels (Fig. 2D), which followed a similar trend to Sox-1+ staining. CS56 antibody staining of implanted eCS matrices and endogenously produced CS-GAGs in the lesion site revealed no significant differences in CS56 staining between TBI and eCS groups but indicated a significantly greater CS56+ staining in both groups when compared to Sham control (Fig. 2E). Qualitative differences in locational presence of CS56+ staining were observed throughout the lesion sites in eCS and sTBI-only animals. When compared to sTBI-only animals, which predominantly demonstrated CS56+ staining only in the lesion penumbra and indicated a complete lack of cellular presence in the lesion volume, eCS matrix–treated animals indicated a uniform presence of CS56+ staining throughout the lesion volume. The CS56+ staining of eCS matrix–implanted brain tissue indicates the potential residual presence of eCS matrix along with what appears to be endogenously produced CS-GAGs/CSPGs 20 weeks after sTBI. These results suggest that eCS matrix implants might enhance NSC proliferation by facilitating the sustained expression and presence of FGF2  in the lesion site 20 weeks after sTBI. We used laser capture microdissection (LCM) to isolate perilesional tissue labeled by rat immunoglobulin G (IgG) staining (Fig. 2F), along with any tissue present in the lesion cavity to specifically identify regional changes in expression of BDNF, FGF2, CXCL12, and CXCR4 transcripts (Fig. 2G). We identified the significant and tissue-specific increase in expression levels of transcripts encoding neuroplasticity (BDNF) and neural progenitor proliferation (FGF2) factors, along with NSC homing (CXCL12 and CXCR4) transcripts in brain tissue explanted from eCS matrix–treated animals when compared to Sham and TBI controls (Fig. 2G; >2-fold, P < 0.05). Fold differences in target gene expression in eCS matrix– implanted animals were compared to that of TBI rats after normalizing to Sham group expression levels and endogenous controls (GADPH and HPRT1). eCS matrices promote chronic neurogenesis and neuroplasticity Because cell proliferation, neuronal differentiation, and synaptic plasticity are indicators of normal brain tissue homeostasis (47, 48), we considered the occurrence of these processes within the lesion site as measures of functional neuronal network activity, tissue maintenance, and recovery. We quantified the presence of newly formed neurons using markers for immature/migrating neuroblasts [doublecortin-positive (DCX+ )] and dividing [5-bromo-2′-deoxyuridine– positive (BrdU+ )] neuronal cells (Fig. 3, A and B). We found a significantly increased number of dividing cells in the eCS-implanted rats when compared to Sham and TBI groups (fig. S4, A to D). A similar trend was found for DCX+ cells (fig. S4E) and DCX+ /BrdU+ / DAPI+ (4′,6-diamidino-2-phenylindole–positive) colabeled cells in eCS-treated animals (Fig. 3B), demonstrating a significant increase in these markers when compared to Sham and TBI groups. These results indicate that eCS matrix implants might promote the differentiation of proliferating cells into new neurons at the lesion site up to 20 weeks after sTBI. Because synaptic vesicle presence is indicative of functional neurons and synaptic plasticity (49), we evaluated the potential change in synaptic plasticity at the 20-week time point using antibodies against the synaptic vesicle marker synaptophysin I (Syn; Fig. 3C). We observed a significant increase in Syn1+ signal in both DAPI+ (P = 0.014; fig. S5, A to D) and NeuN+ cells (Fig. 3D) in eCS matrix– implanted animals when compared to Sham and TBI-only rats. Despite a net reduction in Olig2+ cells in eCS-implanted animals (fig. S6, C and D), we observed a significantly higher percentage of Olig2+ cells colocalized with NeuN+ cells in eCS rats when compared to those in the Sham and TBI groups (fig. S6, A to C and E). In contrast to these findings, the colocalization of NeuN+ and Olig2+ relative to DAPI+ was found to be significantly reduced in TBI rats (fig. S6, B and F) when compared to Sham and eCS matrix– implanted animals. Together, these results suggest that despite a significant reduction in the number of neurons following sTBI, eCS matrix implant promoted cell proliferation, neuronal differentiation, synaptic plasticity, and potential myelination of newly formed neurons present intra- and perilesionally. eCS matrix implants attenuate the chronic presence of neuroinflammatory cells Attenuated influx of neuroinflammatory cells chronically after sTBI could prevent prolonged tissue damage and atrophy and is a marker of a favorable tissue response after TBI (50). To mark the chronic presence of activated macrophages and reactive astrocytes that are characteristic of a neuroinflammatory cellular response to brain injury, we quantified CD68+ -activated macrophages and glial fibrillary acidic protein–labeled (GFAP+ ) reactive astrocytes (fig. S7, A to C). We observed that eCS matrix–implanted rats had similar levels of CD68+ cells (fig. S7D) as Sham controls, while TBI rats showed a significant decrease in CD68+ cells compared to Sham and eCS matrix–implanted rats. While Sham and eCS rats showed similar GFAP+ expression levels, TBI rats in comparison demonstrated a significant increase in GFAP+ area (fig. S7E). Although the TBI animals demonstrated a significant reduction in activated macrophage response, we speculate that this is likely due to the significant tissue loss and absence of intralesional tissue in TBI animals when compared to Sham controls and eCS-treated animals. Overall, these results suggest the prevalence of an attenuated neuroinflammatory cellular response in eCS animals 20 weeks after sTBI. eCS matrices enhance local vascularization and global blood flow Because inadequate vascularization is often responsible for the failure of implanted biomaterials (51), we investigated the extent of tissue neovascularization in eCS matrix–implanted animals when compared to controls, using collagen IV (Col-IV+ ) and rat endothelial cell antigen (Reca1+ ) markers (Fig. 4, A and B). We also used MRI phase gradient imaging of CBF as a measure of vascular function (Fig. 4, C to F). We found a significant increase in Reca1+ (fig. S8D) and Col-IV+ (fig. S8E) expression percentage area in the eCS matrix–treated animals compared to both Sham and TBI controls. Notably, Reca1 and Col-IV colocalization was detected to be ~60% in Sham and eCS matrix–treated groups, which was significantly enhanced when compared to the TBI group (Fig. 4B). Because a significant reduction of CBF chronically is linked to cognitive dysfunction and poor prognosis in humans with sTBI (52–56) and is also associated with the lack of neuronal activity and loss of neuronal volume in humans and rats (57, 58), we assessed CBF using MRI-based normalized phase gradient (nPG) mapping of identified blood vessels in the lesion ( Fig. 4C). We observed a significantly enhanced overall CBF in eCS matrix–implanted rats (Fig. 4D and fig. S9; estimation statistics) when compared to Sham animals, while TBI control animals showed a statistically insignificant decrease in measured CBF. Using ipsi- and contralesional region of interest (ROI) measurements (Fig. 4E), we detected a specific increase in cortical CBF in eCS matrix–implanted rats with a significantly higher CBF detected in the contra- versus ipsilesional side (Fig. 4F). Together, these results indicate that eCS matrix– implanted rats demonstrate chronically enhanced neovascularization perilesionally, along with significantly enhanced CBF both contraand ipsilesionally when compared to TBI-only controls. eCS matrices promote chronic forelimb-specific functional recovery and activation of activity-regulated cytoskeleton-associated protein (Arc) in RFA We used a skilled reach task (SRT) to investigate the circuit-specific implications of the neuroprotective, neurogenic, and angiogenic attributes of eCS matrix implants following sTBI. The SRT was used as a forelimb-specific assessment of motor recovery followed immediately by terminal volumetric imaging of activity-regulated cytoskeleton-associated protein (Arc+ ). This made it possible to assess task-specific responses following a lesion of the M2–reach-to-grasp region in the rostral forelimb area (RFA). We found that eCS matrix–implanted rats that trained for 2 weeks on the SRT (Fig. 5A) demonstrated reach-to-grasp performance recovery from week 2 that was comparable to Sham animals and that persisted up to week 8 (Fig. 5B and fig. S10). TBI control rats showed significant forelimb functional deficits throughout the 8 weeks of testing, with a transient improvement that lasted about 3 weeks (weeks 4 to 6) before significantly worsening at week 8. The quantification of ipsi- and contralesional Arc+ neurons (Arc+ / NeuN+ coexpression; Fig. 5C) immediately after reach-to-grasp activity showed strong activation within anterior and posterior RFA regions and low contralateral signal in Sham animals (Fig. 5D). TBI control rats, in comparison, showed little to no ipsilesional RFA activity with notable contralesional activation when compared to eCS matrix–implanted rats, which demonstrated Arc+ signal within the RFA lesion in addition to contralesional activation. Perilesional laminar recordings of multiunit activity reveal the preservation of sensorimotor responses 10 weeks after sTBI in eCS matrix–implanted rats We stimulated the contralesional paw using low-intensity electrical pulses to further investigate whether eCS matrix implants directly facilitated recovery of sensorimotor responses (fig. S11A). We evaluated the spontaneous and evoked (fig. S11, B and C) electrophysiological responses from the intact M1 and M1/S1 regions using a 32-channel laminar electrode (fig. S11D). We observed that following sTBI, a rapid response to electrical stimulation of the left paw reduced in occurrence rate and increased in time jittering in all cortical layers and in the two recording positions CFA and CFA/S1 (fig. S11E). Notably, the late response recorded in the CFA region in response to paw stimulation showed an increased delay in TBI animals (fig. S11E, left; mean jitter: 36.5 ms), which was found to be faster in eCS matrix–implanted rats (mean jitter: 15.1 ms). This effect was not detected in CFA/S1 region (fig. S11E, right). The TBI control rats also showed a sustained post-stimulation activity registered during nonstimulation recording periods when compared to the prestimulation epochs (fig. S12), which was not observed in Sham and eCS matrix–implanted rats. We also found that paw stimulation induced differential activation of the CFA/S1 and CFA motor area, with TBI control animals demonstrating a marked reduction and eCS matrix-implanted animals showing a marked increase in activation respectively. These results indicate that the response and activation of the perilesional circuitry associated with sensorimotor function of forelimb were reduced in TBI rats and partially recovered in eCS matrix–implanted rats. RFA-specific revascularization explains neuronal presence and behavioral performance We performed brain tissue clearing to specifically investigate whether vascular architecture could reveal a stronger correspondence between vasculature features (Fig. 6), neuronal presence, and behavioral performance. Following Imaris-based vasculature tracing (Fig. 6, A and B), we observed that the anterior RFA region in eCS rats showed an increased vessel segment density (Fig. 6C) and cumulative vessel length (Fig. 6D) when compared to TBI control rats, whereas both TBI and eCS matrix–implanted animals demonstrated a marked reduction in these features compared to Sham (movies S1 to S4). We also observed that the posterior RFA showed a decreased density of vasculature in both TBI and eCS matrix–implanted animals. Notably, only three of five rats showed tissue presence in the RFAa ROI in TBI controls, whereas all animals in Sham and eCS matrix– implanted groups showed cellular and tissue presence. Tissue clearing and volumetric imaging revealed a larger propensity of tissue preservation in eCS matrix–implanted rats (movies S3 and S4) compared to TBI controls (fig. S13 and movie S2), which is consistent with lesion volume analysis obtained in Fig. 1. Using strong anion exchange (SAX) high-performance liquid chromatography (HPLC) analysis of CS sulfation profiles in brain tissue, we demonstrate that brain tissue isolated intralesionally from eCS matrix– implanted animals closely mimicked the CS sulfation profiles of Sham and perilesional tissue in TBI animals. These results also indicate a change in CS sulfation profile in brain tissue obtained from eCS matrix–implanted animals 4 weeks after TBI when compared to eCS matrix alone, suggesting integration and remodeling of the implanted eCS matrix (fig. S14 and table S1). We found that the mean vessel diameters of vasculature in TBI controls and eCS matrix–implanted animals were not significantly different from those in Sham rats (Fig.  6E). However, consistent with the increased CBF observed in eCS matrix–implanted animals (Fig.  4, –20 weeks), the distribution of both vessel length density and diameter in RFAa was enhanced in eCS matrix–implanted rats when compared to Sham and TBI controls, as shown in Q-Q plots (fig. S15, A and B). We found a significant correlation between NeuN+ cells and vessel density (fig. S16A), and NeuN+ cells and forelimb performance (fig. S16B). The correlation between vessel density and SRT performance also returns a strong Pearson’s R value (fig. S16C), although not statistically significant. Together, these results indicate that eCS matrix implants promoted vascularization intralesionally (RFAa), which is strongly correlated with neuronal presence and forelimb performance in rats. DISCUSSION Parenchymal volume loss of brain tissue is highly correlated with the level of TBI severity in humans, with patients sustaining severe lesions experiencing significantly greater volume loss when compared to those who sustained mild TBIs (59). As a result, functional losses are inevitable and neuroplastic changes are limited, leaving patients with chronic and debilitating impairments (1). Studies on the intracortical implantation of brain-mimetic CS and hyaluronic acid GAG scaffolds after sTBI and stroke have demonstrated the ability to mediate complex structural and functional repair of brain tissue (38, 46, 60). The acute implantation of neurotrophic factor functionalized acellular eCS scaffolds that are compositionally similar to brain ECM could present a rational approach to mitigating the significant volume and functional losses encountered chronically after sTBI. Using a materials design strategy that exploits the native functionality of CS, we demonstrate that neurotrophic factor–laden eCS matrix implants were neuroprotective and neurogenic and significantly enhanced the peri- and intralesional presence of newly formed blood vessels chronically after sTBI when compared to controls. The role of a global reduction of CBF and hypoperfusion in chronic dysfunction and poor prognosis after sTBI is well established (53–57). Our observations correlating enhanced vascular density to neuronal presence and reach-to-grasp function recovery suggest that eCS matrix implants orchestrate complex vascular repair that directly contributes to neuronal function and task-specific recovery chronically after sTBI. These effects are likely mediated by the enhanced presence and signaling of FGF2 and BDNF, which is potentiated by CS binding of these factors as demonstrated previously (17, 60, 61). We nonetheless interpret the implications of observed rCBF increase cautiously in the absence of comparative real-time cerebral oxygenation studies. The compensatory reorganization of the motor cortex has been observed in perilesional circuitry and the contralateral hemisphere after TBI (62–65). In our studies, the volumetric quantification of Arc+ neuronal populations following left-limb usage confirmed enhanced activity in both the perilesional and contralateral regions in TBI-only rats compared to Sham controls. eCS matrix–implanted rats showed a pattern of functional activity similar to TBI-only rats, with the added presence of intralesionally active neurons within the implant. Because BDNF expression is known to be regulated via autocrine signaling mechanisms (66) and could be contributing to enhanced neuroplasticity observed after brain injury (62, 63, 67), we speculate that enhanced transduction of neurotrophic factor signaling could be contributing to the observed enhancement in motor remapping (Fig. 3) and potentiation of FGF2 and BDNF expression in eCS matrix–implanted animals (60, 61, 68, 69) when compared to TBI-only controls (Fig. 2). Our results also showed a close correspondence between Arc-dependent activation patterns and both electrophysiological response and recovery of forelimb-specific sensorimotor functions. This evidence suggests that intracortical eCS implants mediated the reorganization of neuronal circuitry, leading to chronic recovery of reach-to-grasp function after sTBI. Subacute studies conducted 4 weeks after TBI have demonstrated the neuroprotective attributes of unmodified CS matrix implants (46). However, in light of evidence suggesting the poor prognostic implications of low BDNF levels acutely after sTBI (70), and function blocking studies demonstrating the direct role of BDNF in facilitating motor function recovery (71), strategies to enhance local signal transduction of neurotrophic factors as demonstrated in this study could offer significant therapeutic benefits. Despite the lack of eCS matrix presence 4 weeks after TBI as determined by SAX-HPLC analysis of brain tissue, our results imply that the complex cellular and vascular repair and functional recovery observed chronically in eCS matrix–implanted animals may have been influenced by acutely implanted eCS matrices. Future investigations of the temporal effects of indwelling CS and eCS matrix implants on neurotrophic factor signaling will help better inform the design of more effective acellular eCS matrix implants for sTBI. In summary, our results demonstrate that rationally designed, brain-mimetic, acellular eCS implants have native structure-function attributes required to mediate chronic tissue level repair and functional recovery after sTBI. This study opens up new avenues for the design and application of synthetic sulfated GAG constructs for brain tissue repair and informs future investigations of neuronal connectivity and electrophysiological responsiveness between the eCS matrix–implanted tissue and perilesional cortical columns.

 

More at link.