Nanodrugs for the Treatment of Ischemic Stroke: A Systematic Review

Ischemic stroke, a significant neurovascular disorder, currently lacks effective restorative medication. However, recently developed nanomedicines bring renewed promise for alleviating ischemia’s effects and facilitating the healing of neurological and physical functions. The aim of this systematic review was to evaluate the efficacy of nanotherapies in animal models of stroke and their potential impact on future stroke therapies. We also assessed the scientific quality of current research focused on nanoparticle-based treatments for ischemic stroke in animal models. We summarized the effectiveness of nanotherapies in these models, considering multiple factors such as their anti-inflammatory, antioxidant, and angiogenetic properties, as well as their safety and biodistribution. We conclude that the application of nanomedicines may reduce infarct size and improve neurological function post-stroke without causing significant organ toxicity.


Introduction
Stroke, which accounts for approximately 5.5 million deaths annually, stands as the second leading cause of death worldwide. Beyond the significant mortality rate, stroke also inflicts substantial morbidity, with nearly half of survivors experiencing long-term disabilities. As a result, stroke carries considerable economic and social implications, making it a matter of utmost importance in public health. According to a study conducted by Donkor ES et al. in 2018, it is projected that by 2030, there will be a staggering 23 million stroke cases, resulting in 7.8 million associated deaths [1]. Recent findings from an American Heart Association meeting reveal that ischemic stroke constitutes nearly 87% of all stroke cases [2].
Ischemic stroke, a severe neurological condition, has limited therapeutic options available. Thrombolytic therapy is the sole FDA-approved clinical treatment for ischemic stroke. However, the presence of a physiological barrier complicates the development of effective therapeutic and diagnostic strategies for ischemic stroke [3]. After an ischemic injury, the blood-brain barrier represents a significant obstacle to the penetration of both small-and large-scale therapeutic agents into the brain [4]. 2

of 18
After an ischemic stroke, local microglia undergo polarization towards the M1 phenotype due to exposure to proinflammatory cytokines such as IFN-γ and TNF-α. Subsequently, these cells show significant expression of inducible NO (iNOS) to synthesize NO and generate pro-inflammatory cytokines such as TNF, interleukin (IL)-1, and IL-12. The outcome of this M1 polarization event depends on various factors, including the production of iNOS, reactive oxygen species (ROS), and the activation of the inflammasome complex known as NOD-like receptor family pyrin domain-containing 3 (NLRP3), all of which contribute to secondary brain damage [5,6].
The production of anti-inflammatory cytokines (IL-10, IL-13, TGF-β, VEGF, EGF, and Arg1) by M2 microglia has a beneficial impact in cases of ischemia or hypoxia following an ischemic stroke. This production helps to deactivate pro-inflammatory cell types, restore balance within the body, and enhance the outcome of strokes [5,6]. Consequently, modulating the shift from proinflammatory M1 microglia to anti-inflammatory M2 microglia shows promise as a therapeutic strategy for stroke treatment [7].
Nanotechnology holds promising prospects for the treatment of various central nervous system diseases in the field of medical research [8]. Nanoparticles (NPs) are solid colloidal particles, with sizes ranging from 1 to 1000 nm, and they find diverse applications in medicine. Utilizing their small size and mobility, NPs can effectively reach different tissues and cells, enabling both extracellular and intracellular administration. Several materials and techniques can be employed to fabricate nanoparticles, adjusting various parameters to achieve the desired NPs. Key considerations for drug delivery include size, payload encapsulation efficacy, zeta potential, and payload release characteristics [4]. Since the brain's smallest capillaries have a diameter of only 5-6 µm, administering NPs via microcirculation emerges as a practical approach to facilitate drug delivery to the brain. NPs provide a means to deliver drugs, proteins, vaccines, biological macromolecules, and gene therapy, regardless of their hydrophilic or hydrophobic nature [9]. Nanomaterials offer the potential for molecular and supramolecular interactions with biological systems, which are advantageous. Utilizing these interactions makes it possible to induce appropriate physiological responses in cells while minimizing negative effects on the overall system [10].
In this systematic review, our objective was to evaluate the scientific quality of current research focusing on nanoparticle-based treatments for ischemic stroke. Additionally, we aimed to assess the potential effectiveness of future therapies. This study also discusses the application of nanoparticles for stroke treatment in animal models. We conducted an analysis of their efficacy, taking into account a range of parameters including their anti-inflammatory, antioxidant, and angiogenetic properties, as well as safety and biodistribution.

Search Strategy
An extensive search was conducted in the PubMed and Web of Science databases using the keywords "nano-drug", "treatment", and "stroke" (Supplementary Table S1). To refine the results, we filtered for English language publications, studies on animal species, original work with free access, and articles published within the past 5 years (2018-2022). Duplicate studies were eliminated. The remaining studies were initially assessed based on their abstracts to remove irrelevant records. Subsequently, all the remaining articles were thoroughly read, with only the most relevant ones being included in the study.

Inclusion and Exclusion Criteria
Only original articles written in English were considered for inclusion. The following inclusion criteria were used: (1) studies that used in vivo models of ischemic brain lesions and (2) studies that used nanoparticles, loaded or not with some type of drug, for stroke treatment. Research involving cell cultures, nano-treatment for animals without stroke lesions, studies using drug therapy on a larger scale than the nanoscale, and studies utilizing nanotechnology for paraclinical exploration were excluded (Supplementary Table S2).
Articles that were indexed in more than one database (duplicates), incomplete articles, abstracts, reviews, and book chapters, as well as studies involving ischemic injury in locations other than the brain were excluded

Data Extraction and Quality Assessment
Each study provided various details. These included author information, publication year, and the journal in which the study was published. The study's 2021 impact factor was also listed. Information regarding the animal species used and the number of animals involved was included. Details about the targeted nanoparticles, the potential drug treatment, and the stroke procedure model were provided as well. Other aspects noted were the administration route, treatment time points, and methodological and reporting features. These were assessed through a quality questionnaire designed specifically for pre-clinical studies [11].

Results
Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram [12], the search strategy identified 230 potentially related titles. Three duplicate studies were removed, leaving 227 records, of which only 38 studies (16.74%) met the inclusion criteria. Of the remaining 189 PubMed records, 19 studies (10.05%) met the eligibility criteria. Some 55 studies (29.1%) were excluded because they were not conducted on an animal model of ischemic stroke, 20 studies (10.58%) were not pre-clinical trials, 15 (7.94%) were not based on nanomedicine, and 29 (15.34%) did not involve drug therapy. Out of the 89 Web of Science articles, 19 (21.34%) met the inclusion criteria. Some 35 (39.32%) were excluded because they were not conducted on an animal model of ischemic stroke, 16 (17.97%) were not pre-clinical trials, 6 (6.74%) were not based on nanomedicine, 12 (13.48%) did not involve drug therapy, and 1 (1.12%) was not available for full free access ( Figure 1). lesions, studies using drug therapy on a larger scale than the nanoscale, and studies utilizing nanotechnology for paraclinical exploration were excluded (Supplementary Table  S2).
Articles that were indexed in more than one database (duplicates), incomplete articles, abstracts, reviews, and book chapters, as well as studies involving ischemic injury in locations other than the brain were excluded

Data Extraction and Quality Assessment
Each study provided various details. These included author information, publication year, and the journal in which the study was published. The study's 2021 impact factor was also listed. Information regarding the animal species used and the number of animals involved was included. Details about the targeted nanoparticles, the potential drug treatment, and the stroke procedure model were provided as well. Other aspects noted were the administration route, treatment time points, and methodological and reporting features. These were assessed through a quality questionnaire designed specifically for preclinical studies [11].

Results
Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram [12], the search strategy identified 230 potentially related titles. Three duplicate studies were removed, leaving 227 records, of which only 38 studies (16.74%) met the inclusion criteria. Of the remaining 189 PubMed records, 19 studies (10.05%) met the eligibility criteria. Some 55 studies (29.1%) were excluded because they were not conducted on an animal model of ischemic stroke, 20 studies (10.58%) were not pre-clinical trials, 15 (7.94%) were not based on nanomedicine, and 29 (15.34%) did not involve drug therapy. Out of the 89 Web of Science articles, 19 (21.34%) met the inclusion criteria. Some 35 (39.32%) were excluded because they were not conducted on an animal model of ischemic stroke, 16 (17.97%) were not pre-clinical trials, 6 (6.74%) were not based on nanomedicine, 12 (13.48%) did not involve drug therapy, and 1 (1.12%) was not available for full free access ( Figure 1).

Characteristics of the Study
The eligible articles were published between 2018 and 2022. These studies were conducted in various countries, distributed as follows: China (35.84%), the USA (31.57%), England (13.15%), Australia (7.89%), Japan (2.63%), Switzerland (5.26%), and the Republic of Korea (2.63%). The sample size in these studies ranged from 12 to 287 rodents. However, in five studies, the number of animals was not clearly defined. All studies involved animals with ischemic stroke and used the following ischemia models: the MCAO procedure (86.84%), the phototrombotic model (2.63%), the BCCAO procedure (2.63%), the RCCAO procedure (2.63%), and permanent MCAO (2.63%); one study used the first two models of ischemic stroke (2.63%). For the research, studies on rodents with stroke were conducted on mice (55.26%) and rats (42.10%), and one study included both species (2.63%). Six studies were indexed in journals with a 2021 Impact Factor (IF) below 5 (15.79%), twenty-one were indexed in journals with a 2021 IF between 5 and 10 (55.26%), six were in journals with a 2021 IF between 10 and 15 (15.78%), and five were indexed in journals with a 2021 IF above 15 (13.15%). The characteristics of the studies can be found in Table 1.

Methodological Assessment
The SYRCLE criteria are presented in Supplementary Table S3 and Figure 2. Utilizing SYRCLE's risk of bias tool, which comprises ten items across six categories of bias, we evaluated the risk of bias in the included studies. These categories included selection, performance, detection, attrition, reporting, and other biases [51]. The term "Yes" indicates a low risk of bias, while "No" signifies a high risk of bias. A "?" symbolizes an unknown risk of bias. Overall, none of the studies included satisfied all the SYRCLE criteria, as the risk from other potential sources of bias remained unclear.  Table S3). Table 2. Nanoparticles associated with stroke treatment in animal models.   Table S3).

Nr First Author Year Species Nanoparticle Combined with Route Timepoint Results
On average, the included studies reported six out of the ten characteristics. The range of scores varied, with the lowest score being 2 out of 10 items (5.12%), and the highestscoring studies reporting 9 out of 10 items (15.38%). All the studies fulfilled two of the three selection bias criteria, as they described the methods used to generate the allocation sequence in enough detail to evaluate whether it should produce comparable groups, and they used similar groups of animals at baseline. In 64.10% of the included research, the animals were randomly housed during the experiment, and 43.58% of the outcome results were analyzed in a blind manner.
The animal model of ischemia was treated with nanoparticles in multiple doses in 18 records (46.15%) and in a single dose in the remaining 21 records (53.84%). In 32 records (82.05%), nanoparticles were administered after ischemic or reperfusion surgery, while they were used as a pretreatment in 5 studies (12.82%).

Discussion
Ischemic stroke is a severe cerebrovascular disease for which no existing restorative drugs are available. Recently developed nanodrugs offer fresh hope to mitigate the effects of ischemia and stimulate the recovery of neurological and physical functions. Ischemia and reperfusion can trigger a series of events, including oxidative stress and inflammatory responses. These are primarily caused by neutrophils infiltrating the post-stroke brain and by the activation of microglia due to neurological damage. Hence, nanodrugs can limit neutrophil infiltration into the ischemic brain and shift the ratio of M1/M2 phenotypes towards the neuroprotective M2 phenotype. Treatments with nanodrugs also reduce infarct volume and neurological dysfunction, with no notable organ toxicity.

The Biodistribution of Nanoparticles
The potential for harmful side effects on primary organs significantly hinders the future scope of nanomedicine in biological applications. Several studies have been conducted to investigate the distribution and potential toxicity of nanoparticles in the body. The methodologies employed include histological examination, in vivo imaging, and the measurement of transaminase activity levels, as reported in various reviewed articles.
Research conducted by Lizhen He and his colleagues [37], Lin Guo and her team [30], Daozhou Liu and his group [17], and Chao Li and his associates [11] has involved detailed histological examination of major organs in rats that had suffered a stroke, using H&E staining. The key organ sections examined included those from the heart, liver, spleen, lung, and kidney. No visible inflammation or other pathological changes were observed following exposure of these organs to nanoparticles. This suggests that there was no organ toxicity as a result of nanomedicine administration. The safety of nanomedicine was further corroborated by histological analysis correlated with the levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST), which were found to be within normal limits [11,17].

The Antioxidant Activity of Nanoparticles in Stroke Models
Ischemia and reperfusion can trigger a cascade of events, including oxidative stress and an inflammatory response, which may eventually lead to irreversible neuron damage. A major cause of oxidative stress is the excessive production of reactive oxygen species (ROS), which activate apoptotic and necrotic pathways, leading to cell necrosis or death [11].
Ma@(MnO 2 +FTY), a macrophage-disguised, FTY-loaded MnO 2 nanoparticle, was first investigated by Chao Li et al. [11]. Its effectiveness was initially assessed in vitro by measuring the residual H 2 O 2 after incubation with various concentrations of MnO 2 for specific time periods. The results showed that MnO 2 could effectively scavenge H 2 O 2 and that the scavenging process was dependent on both the concentration of MnO 2 and the duration of incubation. When Ma@(MnO 2 + FTY) was administered in vivo, it was found to alleviate brain injury and modulate the pro-inflammatory microenvironment [11]. An elevated concentration of reactive oxygen species (ROS) triggers the opening of the mitochondrial permeability transition pore (mPTP) in neurons, subsequently diminishing the potential of the mitochondrial membrane. This initiates a damaging cycle between the mPTP opening and ROS generation. The outbreak of ROS brought about by cerebral ischemia/reperfusion, along with the excessive opening of mPTP in neuronal cells, can be curbed through CsA-loaded HFn nanoparticles (CsA@HFn). These nanoparticles have been shown to mitigate neuronal damage by decreasing the production of ROS [17].
Malondialdehyde (MDA) and superoxide dismutase (SOD) serve as critical benchmarks in the assessment of oxidative stress related to cerebral ischemia/reperfusion damage. An imbalance in oxidative stress during ischemic brain injury leads to a decline in SOD levels and a surge in MDA levels. In the research conducted by Xinyi Yuan and team, the stroke-affected animals that were administered DBZ or DBZ-NLC exhibited a reduction in MDA levels [29]. Following the intrathecal administration of POM nanoclusters, the study led by Shiyong Li and his team revealed that SOD and MDA levels were similar to those found in control groups [34].
Nanoparticles carrying drugs have been found to enhance the formation of new blood vessels in models of stroke. This is important because, after a stroke, the core area of the brain that is affected by the lack of blood flow experiences damage to its vasculature. The process of angiogenesis, which leads to the development of new blood vessels, is a crucial factor in repairing and remodeling these damaged areas. This process largely depends on the successful reinstatement of blood supply. Consequently, recent research has shifted its focus to enhancing vascularization after a stroke [22].
Prior work by Cheng Wang and colleagues [22] discovered that FE is packed with various growth factors, including BDNF, GDNF, TGF-beta, and bFGF. In their most recent study, they devised a biomimetic nanocarrier that can accurately deliver FE as a treatment for stroke. The peptide sequence Arg-Gly-Asp (RGD) is particularly attractive to the αvβ3 integrin, which is found on the surface of blood vessels undergoing angiogenesis. Meanwhile, PLTs have numerous receptors that can adhere specifically to damaged and inflamed vasculature. The interaction between FE particles coated with RGD-PLT facilitates the sustained release of FE at the stroke site, delivering the neurotrophic factors BDNF, GDNF, and bFGF to the brain. This interaction ultimately results in an increase in blood flow and recovery in neurobehavioral functions [22]. Huixin Zhang and colleagues utilized exosomes to carry miR-210, a substance known to enhance angiogenesis via the VEGF signaling pathway [28]. Taking advantage of the affinity between RGD and αvβ3 integrin-a component found on angiogenic blood vessels-the team linked the exosomes with RGD to transport miR-210 to the site of a lesion. The study confirmed that mice treated with RGD-exo:miR-210 showed an increase in microvessel density within the lesion site, as evidenced by CD34+ per field, at both 7 and 14 days post-treatment, compared to the control groups [28].
In a separate study, Haoan Wu and his team discovered that treatment with glyburideloaded ASPTT NPs significantly encouraged angiogenesis in the ischemic brain [32]. The researchers found an impressive increase in the number of new blood vessels in the brains of mice receiving this treatment, compared to controls, as confirmed by CD31 immunohistochemistry (IHC) in the ischemic stroke brains.

The Role of Nanomedicines in Reducing Inflammation in Stroke Models
In the process of injury, inflammation plays a critical role, amplifying the extent of neuronal damage following ischemic injury to the central nervous system. Studies investigating the levels of pro-inflammatory cytokines TNF-α, IL-6, and IL-1β have shown dose-dependent reductions, illustrating the anti-inflammatory effects of nanoparticles [11,13,16,25,31,37,39,41].
Research assessing myeloperoxidase (MPO) activity in the brain-an indicator of neutrophil tissue infiltration-by Zhang and colleagues [15] and Dong and his team [31] has demonstrated that neutrophil infiltration into the ischemic brain can be limited by Doxorubicin and Resolvin D2-loaded nanovehicles.
Research by Koteswara Rao Nalamolu et al. [13] and Tsogzolmaa Ganbold et al. [26] found that Toll-like receptors (TLRs) are instrumental in controlling post-ischemic inflammatory responses. They developed nanoparticles carrying short interference RNAs to target TLRs 2 and 4. The outcome was a substantial reduction in TLR2 and TLR4 gene expression, which reversed cytokine expression patterns and limited the release of numerous proinflammatory cytokines, such as iNOS. This also encouraged the production of several anti-inflammatory cytokines and neurotrophils, leading to an improvement in neurological function. The intricate nature of the signaling systems involved, along with their accompanying inflammatory responses, poses considerable challenges for the development of anti-stroke drugs. Signaling networks play a crucial role in neuronal growth and metabolism in mammalian CNSs, and any disruption to these pathways during or post-stroke can lead to neuronal injury and death [36]. Research by Tian-Tian et al. [10] suggested that the anti-inflammatory action induced by EVReN strongly relies on the MAPK pathway being inhibited by the miRNAs incorporated into EVReN. The findings by Hamed Amani et al. [36] showed that the Jak2 pathway was substantially lowered in the presence of OX26-PEG-Se NPs.
Microglia constitute approximately 10% of all cells in the brain [8]. In the event of an ischemic stroke, these microglia become stimulated and may differentiate into one of two phenotypes: the M1 phenotype, which promotes inflammation, or the M2 phenotype, which mitigates inflammation. M2 microglia assist in recovery following a stroke. Conversely, M1 microglia release a range of pro-inflammatory factors, including IL-1, IL-6, IFN, and TNF, which contribute to additional brain damage. These factors significantly contribute to the worsening of cerebral injury.
In their studies, researchers used Iba-1 immunostaining, a method for tracking the activation of microglia, to observe the response of microglia after being treated with various substances delivered by nanoparticles [10,17,19]. Specifically, in the research conducted by Tian and colleagues [10], microglia exhibited extensive branching of processes after being exposed to extracellular vesicles derived from neural progenitor cells. This was in stark contrast to the microglia in the control group, which maintained thin ramified processes and small cellular bodies.
In an effort to ascertain the ratio of M1 and M2 microglia types within the area of cerebral infarction, research conducted by Daozhou Liu and his colleagues [17], as well as Po-Wah So and his team [19], employed different labeling methods. The M1 type was marked by CD16/32, while CD206 was used to indicate the M2 type. When compared with the control group, a decline in the quantity of cells co-expressing CD16/32 was noted following treatment involving heavy chain ferritin nanoparticles loaded with cyclosporine A and liposomal nanoparticles encapsulated with acetate. Conversely, there was an uptick in the count of cells co-expressing CD206.
Hyo Jung Shin and colleagues analyzed the potential of Perampanel for neuroprotection [16]. The researchers utilized PLGA NPs to enhance the delivery of Perampanel to microglia. Their results indicated that when Perampanel was successfully delivered to microglia, it managed to decrease the emission of pro-inflammatory cytokines while promoting the presence of M2 phenotype microglia, thereby contributing to neuroprotection.
Functional recovery of the nervous system was evaluated by conducting a variety of tests that comprised the components of mNSS, such as motor, sensory, reflex, and balance assessments. These testing methodologies were extensively applied in numerous studies, which utilized the filament stroke model, a method adapted by Longa and his team [49], to gauge the severity of post-stroke damage and the pace of recovery [13,22,34,35,47]. Notably, animal groups that received nanodrug treatments displayed significantly reduced neurological dysfunction, as demonstrated by their lower mNSS scores compared to the control group. This points towards faster recuperation [18,21,25].

Conclusions
Ischemic stroke, a significant neurovascular disorder, currently lacks effective restorative medication. However, recently developed nanomedicines bring renewed promise in alleviating ischemia effects and facilitating the healing of neurological and physical functions. The process of ischemia and reperfusion initiates a chain reaction of events, primarily involving oxidative stress and inflammatory responses due to neutrophils infiltrating the brain after a stroke and the activation of microglia resulting from neurological injury. Recently, it has been recognized that the production of anti-inflammatory cytokines by M2 microglia benefits ischemia or hypoxia after an ischemic stroke by deactivating pro-inflammatory cells, restoring balance, and improving stroke outcomes. Therefore, modulating the shift from pro-inflammatory M1 microglia to anti-inflammatory M2 microglia holds promise as a therapeutic approach for stroke treatment. Nanodrugs may have potential benefits in the treatment of cerebral ischemia by restricting neutrophil penetration into the ischemic brain and encouraging a shift in the M1/M2 phenotype ratio in favor of the neuroprotective M2 phenotype. The application of nanodrugs also may reduce infarct size and improve neurological recovery without causing significant organ toxicity.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.