Brain-Derived Neurotrophic Factor and Nerve Growth Factor Therapeutics for Brain Injury: The Current Translational Challenges in Preclinical and Clinical Research

Ischemic stroke and traumatic brain injury (TBI) are among the leading causes of death and disability worldwide with impairments ranging from mild to severe. Many therapies are aimed at improving functional and cognitive recovery by targeting neural repair but have encountered issues involving efficacy and drug delivery. As a result, therapeutic options for patients are sparse. Neurotrophic factors are one of the key mediators of neural plasticity and functional recovery. Neurotrophic factors such as brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF) serve as potential therapeutic options to increase neural repair and recovery as they promote neuroprotection and regeneration. BDNF and NGF have demonstrated the ability to improve functional recovery in preclinical and to a lesser extent clinical studies. Direct and indirect methods to increase levels of neurotrophic factors in animal models have been successful in improving postinjury outcome measures. However, the translation of these studies into clinical trials has been limited. Preclinical experiments have largely failed to result in significant impacts in clinical research. This review will focus on the administration of these neurotrophic factors in preclinical and clinical stroke and TBI and the challenges in translating these therapies from the bench to the clinic.


Introduction
Ischemic stroke, a leading cause of disability worldwide, results from limited blood flow to the brain due to the blockage or narrowing of arteries. Unfortunately, there are a lack of therapeutic options that can effectively minimize damage or aid in recovery following brain injury from ischemic stroke. The pharmacologic standard of care involves clot breakdown with the thrombolytic agent tissue plasminogen activator (tPA) and/or prevention of further ischemia using anticoagulants such as aspirin. After a stroke, tPA is often the best available option for patients but must be administered within the first 3 hours, or potentially up to 4.5 hours [1]. The short time window in which tPA can be administered, combined with potential complications such as intra-cranial hemorrhage, has greatly limited its use in some patients [2]. While clot thrombolysis and prevention can be useful in preventing further ischemic damage, by the time a stroke patient receives these treatments, brain injury has already occurred. Despite a need for new stroke treatments, there has been little success in identifying therapeutics that can be widely used to promote neural repair and improve functional recovery following brain injury. Many promising experimental treatments have failed to deliver positive results in clinical trials [3] for reasons including lack of efficacy and target validation and pharmacokinetic and pharmacodynamic issues [4].
One avenue of stroke research has involved exploring the use of neurotrophin treatments as a mechanism for neural repair and enhanced recovery. Neurotrophins are a family of growth factors that play important roles in the survival and function of neurons. There are four known members of the neurotrophin family of growth factors in mammals: brain-derived neurotrophic factor (BDNF), nerve growth factor (NGF), neurotrophin-3 (NT-3), and neurotrophin 4 (NT-4) [5]. Neurotrophins regulate development in the central and peripheral nervous systems by interacting with tropomyosin receptor kinase (Trk) receptors [6]. NGF preferentially binds with TrkA receptors, BDNF and NT-4 with TrkB receptors, and NT-3 with TrkC receptors [5]. The dimerization and autophosphorylation of Trk receptors activate major signaling pathways including PLC-gamma, MAPK/ERK, and PI3K/Akt which suppress apoptosis through their downstream mediators CREB, BCL2, and BAX and Bad, respectively [7]. BDNF and NGF increase the phosphorylation of synapsin 1 for synaptic vesicle release [8]. These four neurotrophins also bind to their low-affinity receptor, p75 NTR , which induces apoptosis, and in some cases may promote neuronal survival and neurite growth during neurodevelopment [9]. Therefore, neurotrophins are key mediators in neural plasticity postinjury to promote neuronal growth and survival [10]. Of these four neurotrophins, only two, BDNF and NGF, are well studied as potential stroke treatments.
This review will outline the current state of preclinical and clinical research surrounding the potential for BDNF and NGF to become treatment options for patients following stroke. These neurotrophins promote neuroprotection and regeneration and have been examined to determine their role in neural repair as well as their ability to improve functional recovery in preclinical and, to a lesser extent, clinical studies. Direct and indirect methods of increasing levels of these neurotrophins in animal models have demonstrated promise in improving outcome measures following brain injury. Unfortunately, the translation of these preclinical studies into clinical trials has been limited. This review will focus on both direct administration of exogenous neurotrophic factors and indirect methods of modifying endogenous neurotrophic factor levels in the central nervous system after stroke, and it will also examine the challenges involved in moving BDNF and NGF-related treatments from the bench to the clinic.
1.1. Challenges with Treatments. Currently, a need exists for the development of additional therapies that are effective for improving recovery from ischemic injury. While there have been promising advances in preclinical studies, many of these therapies have failed to translate clinically. There are several challenges that may contribute to this lack of translation. First, many treatments have poor pharmacokinetic profiles [11]. Second, the presence of the blood-brain barrier limits the ability of many systemically administered therapeutics to access the central nervous system. Finally, preclinical studies vary widely in their methods, and their outcomes have not converged on a consensus as to the effects of neurotrophin treatment or the extent of those effects. These issues may contribute to the challenges in translating preclinical studies into clinical trials. Human trials involving the administration of neurotrophins for ischemic injury are rare, and there have been no large, comprehensive clinical trials. As a result, there is not sufficient information to assess whether current preclinical models and methodologies are adequate to forecast human outcomes.
1.2. Poor Pharmacokinetics (ex., Size and Half-Life). Neurotrophins can form protein-antibody complexes which may affect their tissue distribution, metabolism, and elimination. Additionally, peptidases and proteases in the blood can degrade neurotrophins, leading to reduced bioavailability, as evidenced by poor tissue distribution and short halflives. An increased dose would be required to compensate for its poor bioavailability. This, in turn, can trigger adverse effects. Administration of neurotrophins can cause immunogenicity which can manifest in adverse effects including hypersensitivity and anaphylactic shock [12]. In order to overcome these pharmacokinetic issues, neurotrophins have been incorporated in drug delivery systems and neurotrophin mimetics with more favorable pharmacokinetics have been developed. Studies involving the implantation of BDNF polymers in the hippocampi of rats indicated that microspheres released the majority of the encapsulated BDNF within 48 hours [13]. Mimetics, which mimic the BDNF protein, were similarly created in an attempt to circumvent the pharmacokinetic issues [14]. The development of nonpeptide molecules that can activate the TrkB receptor without activating the harmful p75NTR receptor has been the subject of current research. In vitro experiments demonstrate that the molecule LM22A-4, a selective small-molecule partial agonist of TrkB, can trigger the downstream activators of the TrkB receptor [9]. Although these interventions are still in the preclinical stage, their potential to overcome pharmacokinetic challenges may lead to future use in clinical trials. An additional benefit of mimetics is that they may be better able to cross the blood-brain barrier compared to BDNF and NGF.
1.3. Poor Blood-Brain Barrier Permeability. Attempts to use BDNF and NGF as therapeutics for central nervous system (CNS) disorders typically utilize central administration routes that bypass the BBB, including intracerebroventricular (ICV) injection, intraparenchymal injection, and intranasal administration [11]. This is largely due to their severe limitation in crossing the blood-brain barrier; however, challenges related to direct central administration still remain [11]. Intracerebroventricular and intraparenchymal routes of administration are highly invasive. Although intranasal administration is noninvasive, it generally results in lower efficiency of drug delivery to brain tissues as nasal mucosa can inhibit molecule permeability which is compounded by a lack of literature on appropriate nasal delivery [15]. The lack of methods for efficient and noninvasive delivery of BDNF and NGF to the brain has therefore presented a roadblock to studying the direct administration of these neurotrophins. To circumvent this issue, there have been some studies involving indirect modification of the levels of neurotrophins, specifically BDNF.
Indirect modification is achieved by the administration of therapeutics that elicit an increase in neurotrophins in 2 Neural Plasticity the CNS, including drugs classified as NMDA receptor antagonists, cholinesterase inhibitors, statins, and sigma-1 receptor agonists. NMDA receptor antagonists, including memantine, ketamine, and dextromethorphan, have been used in humans as experimental therapeutics for stroke. NMDA receptor antagonism is a mechanism of action in several Alzheimer's disease therapeutics, including memantine. In addition to NMDA receptor antagonism, memantine was found to increase BDNF levels in macaques, as measured by upregulated mRNA and protein expression of BDNF [16]. Because of this impact on neurotrophic factor expression, memantine and other NMDA receptor antagonists are being studied as potential stroke therapeutics in animal models as well as in human clinical trials. One completed trial investigating memantine as a therapeutic for poststroke aphasia showed that memantine treatment resulted in an improvement in speech compared to placebo but did not measure BDNF levels, so it is unclear what mechanisms underlie the benefits to speech associated with memantine treatment [17]. In mice, memantine resulted in increased BDNF signaling, a reduction in reactive astrogliosis, improved vascularization, and improvements in functional recovery [18]. In addition to memantine, several other therapeutics have been shown to modify BDNF levels. Donepezil, a cholinesterase inhibitor used for the treatment of Alzheimer's disease, increased serum BDNF in Alzheimer's disease patients [19], while atorvastatin, a HMG-CoA reductase inhibitor, increased serum BDNF levels and improved functional recovery following stroke [20]. Sigma-1 agonists, which activate TrkB receptors, have been shown to increase BDNF levels in the rat hippocampus [21] and demonstrate neuroprotective effects [22,23] in a non-SOD1 motor neuron disease model, Huntington's disease model, and an SOD1 ALS model through the ERK and Akt pathways downstream of TrkB [24][25][26]. Stem cell therapy is another method currently being studied for its potential to increase BDNF levels. In vitro studies have shown that neural progenitor cells are capable of releasing neurotrophins including BDNF, NGF, and NT-3 [21]. Further preclinical studies have demonstrated the utility of stem cells to elevate BDNF in models of neurological disorders. Implantation of neural stem cells yielded elevated BDNF and increased synaptic density in a mouse model of Alzheimer's disease [22] while mouse models of ischemia reveal that administration of embryonic stem cells (ESCs) can lead to the restoration of behavioral deficits, synaptic connections, and damaged neurons through the release of neurotrophic factors such as BDNF, NGF, and GDNF [23][24][25].

Lack of Consensus on Measurable Outcomes of Recovery.
Several case studies examining the administration of neurotrophic factors, specifically NGF, were published during the 1980s and 1990s but resulted in a lack of consensus on measurable outcomes of recovery. Trials involving central administration of NGF to aid recovery after cerebral ischemia were preceded by its use in Parkinson's disease and Alzheimer's disease [26]. Clinical use of exogenous neurotrophic factors was later examined in two case studies examining the impact of NGF administered via intracerebroventricular infusion on the recovery of infants following a hypoxic/ischemic event. Results of this intervention demonstrated measurable improvements in cognitive and motor performance, including improvements in cerebral perfusion and Glasgow coma score, among other measurements [27]. However, it is difficult to draw any generalized conclusions regarding the efficacy of centrally administered neurotrophins in humans given that these results represented only several individuals within case studies.
Many of the ongoing clinical studies are aimed at using rehabilitation methods such as exercise, transcranial magnetic stimulation (TMS), or hyperbaric oxygen to increase neurotrophin levels, specifically BDNF, and NGF (Table 1). An earlier clinical trial demonstrated increased BDNF levels within the cerebrospinal fluid following intrathecal administration of recombinant BDNF in patients with amyotrophic lateral sclerosis (ALS) [28]. While patients in this clinical trial did not experience serious or painful side effects due to administration, results failed to improve outcome measures. However, a subgroup of patients with severely impaired respiratory problems did significantly improve when compared to placebo groups. This clinical trial demonstrates the possibility of using recombinant BDNF and other neurotrophins to improve severe neurological conditions as other trials involving subcutaneous and intrathecal administration of BDNF are in progress for other conditions such as ALS and spinal cord injury [29].
1.5. The Impact of BDNF Polymorphisms. The beneficial impact of treatment effectiveness is further limited by the presence of BDNF polymorphisms. One of the more studied genetic variants of BDNF is the val66met single nucleotide polymorphism (SNP), which is common in humans, particularly in Asian (40-50%) and Caucasian (25-32%) populations [30]. The val66met SNP is linked to a decrease in activity-dependent BDNF release (but not constitutive BDNF release) and reduced cortical plasticity [31]. It is possible to track and compare the recovery and outcomes of ischemic stroke patients with the normal or abnormal genetic variants of the neurotrophic factor to better understand the effects of neurotrophic factors in cognitive and motor recovery. Although it is debated, it has been suggested that val66met polymorphism may play a role in neurological and neuropsychiatric disorders. The inconsistency in results is complicated by variations in the genetic model used, age, sex, ethnicity, and other factors [32]. Furthermore, the implications of this polymorphism in stroke recovery are not well defined. Recovery of stroke patients has been tracked in several studies investigating cognitive and motor function. The results have been mixed, with some studies suggesting that this polymorphism is linked to worse outcomes compared to the normal BDNF gene and other studies reporting that worse outcomes were seen for short-term but not long-term recovery or that there was no impact on recovery [33][34][35][36].       8 Neural Plasticity  Table 2). This trial, designated NCT03686163, has recruited 106 participants to evaluate the effects of intranasal NGF for acute ischemic stroke (20 μg/day) for two weeks beginning at least 72 hours poststroke. Results are expected in late 2020. Additionally, there are several ongoing or recruiting trials seeking to evaluate the potential to enhance stroke recovery of therapeutics that can influence BDNF levels in the CNS, including memantine for enhanced stroke recovery (NCT02144584) and evaluation of memantine vs. placebo on ischemic stroke outcome (NCT02535611) as well as use of donepezil in combination with transcranial direct current stimulation and intensive speech therapy (NCT04134416). Given the abundance of preclinical research using BDNF and BDNF-enhancing therapeutics, it is notable that BDNF itself is not used as a potential stroke therapeutic in current clinical trials. This may be partially due to its severe limitation in crossing the blood-brain barrier and the resulting challenges related to drug delivery and direct central administration in humans [11] whereas several FDA-approved small molecule therapeutics that cross the blood-brain barrier have been shown to elicit increases in BDNF, including memantine, donepezil, and atorvastatin, which may present a more attractive clinical option. In addition to pharmacologic interventions, there are also interventions involving exercise or motor therapy to attempt to increase endogenous neurotrophin levels.
1.7. Traumatic Brain Injury. Although administration of neurotrophic factors to treat stroke in a clinical setting has not been a primary focus of recent literature, there is substantial interest in understanding the role that endogenous neurotrophic factors play in recovery following other forms of injury such as TBI to optimize recovery. TBI occurs following a bump, blow, or jolt to the brain that causes brain edema and results in neuronal cell death. Treatment options for TBI are similarly lacking, with typical immediate interventions including hyperosmolar therapy to relieve intracranial pressure [37,38] and invasive decompressive surgery [39]. Although there are a variety of pharmacological interventions that can be used following TBI depending on the severity and details of the brain injury, many of these are used to manage TBI sequelae including seizure, clotting, depression, and anxiety as opposed to enhancing neuropro-tection and neural repair mechanisms to address cell death. Preclinical and clinical TBI studies focused on the pathologies can be further exacerbated by severe secondary damage, which is driven by an increased inflammatory response as well as a relatively hypoxic environment [37,38,40] and invasive decompressive surgery [39]. Although there are a variety of pharmacological interventions that can be used following TBI depending on the severity and details of the brain injury, many of these are used to manage TBI sequelae including seizure, clotting, depression, and anxiety as opposed to enhancing neuroprotection and neural repair mechanisms to address cell death ( Table 2).
The direct administration of BDNF as a therapeutic option post-TBI has centered around the role of BDNF in promoting an anti-inflammatory cytokine milieu. In fact, BDNF has been implicated in downregulating the inflammatory response in other pathological states besides TBI/stroke. For example, in female rats inoculated with Streptococcus pneumoniae meningitis, intracisternal BDNF infusions were associated with a significant decrease in inflammatory cytokines, including IL-1B, TNF-a, and NF-κB, and moreover, this response was inhibited when TrkB receptor inhibitor was coadministered [41]. To specifically investigate the anti-inflammatory effects of BDNF on a TBI animal model, Yin et al. attached a collagen-binding domain (CBD) onto BDNF and delivered this combination intracerebroventricularly, as a mechanism to improve BDNF bioavailability [42]. It was revealed that BDNF-CBD was associated with a decrease in brain edema, a reduced amount of NF-κB, and an increased expression of TrkB post-TBI. Further, these effects were reversed via the administration of a TrkB receptor antagonist. To address BDNF's difficulty in crossing the BBB, Kim et al. injected BDNF-filled nanoparticles via IV three hours post-TBI injury in mice [35]. This resulted in a significantly increased level of brain BDNF that correlated with an improved neurological severity score (NSS) test in these mice.
Similar to BDNF, NGF has also been studied in the context of taming the post-TBI inflammatory response as well as its neuroprotective effects. Many TBI research studies focus on the role NGF plays in synaptic transmission, by promoting a more anti-inflammatory cytokine milieu. In an in vitro study, NGF attenuated the robust proinflammatory response    Glasgow Outcome Scale-Extended; BBB: blood-brain barrier dysfunction (vWF, BDNF, GFAP, S100β, sTau, and sNFL); RPSQ: Change in Rivermead Postconcussion Symptom Questionnaire; PROMIS: Change in Patient-Reported Outcomes Measurement Information System; NFL: change in posttraumatic epilepsy biomarker endpoints; TL: telomere length; ICT: Induced Constraint Therapy; MMP-9: matrix metallopeptidase 9; CD105-CXCR4: C-X-C chemokine receptor type 4-PS (phosphoserine).
13 Neural Plasticity to LPS-induced monocytes significantly decreasing NF-κB, IL-1b, and IL-6 mRNA levels [43]. In a similar study, Chiaretti et al. targeted cerebrospinal fluid (CSF) from children immediately following severe TBI and 48 hours after injury [44]. It was discovered that NGF concentrations in cerebrospinal fluid are a biological marker of brain damage following TBIs. The upregulation of NGF within the first 48 hours of injury, when paired with lower IL-1b expression, presents a favorable neurological outcome [44]. These papers illustrate the role NGF has in decreasing inflammation after TBI. Furthermore, these preclinical studies create a foundation for the creation and improvement of clinical trial drug therapies for TBI patients. In 2017, this same research group delivered NGF intranasally to children with severe TBI and found that there was significant cognitive improvement, cerebral perfusion, and brain glucose metabolism associated with the therapy [44].

Conclusion
Determining the role of BDNF and NGF in stroke and TBI is complex and often contradictory. While animal and in vitro studies demonstrate the potential of these two neurotrophins to improve recovery, the lack of clinical research and the lack of consensus among clinical research outcomes emphasize the gap in the translation of BDNF and NGF research from bench to clinic. From these studies, it is clear that neurotrophins, specifically BDNF and NGF, administered directly and indirectly have a growing role in increasing neurogenesis and functional recovery after stroke and TBI and that BDNF and NGF have a synergistic role in motor learning and cognitive recovery. Bearing these partial successes in mind, the lack of clinical research due to the ineffectiveness of treatment options serves as an impetus for the necessity of further investigation not only to identify but also to resolve the roadblocks of the translation of this research.

Conflicts of Interest
The authors declare that they have no conflicts of interest.