Roadmap on nanomedicine for the central nervous system

In recent years, a great deal of effort has been undertaken with regards to treatment of pathologies at the level of the central nervous system (CNS). Here, the presence of the blood-brain barrier acts as an obstacle to the delivery of potentially effective drugs and makes accessibility to, and treatment of, the CNS one of the most significant challenges in medicine. In this Roadmap article, we present the status of the timeliest developments in the field, and identify the outstanding challenges and opportunities that exist. The format of the Roadmap, whereby experts in each discipline share their viewpoint and present their vision, reflects the dynamic and multidisciplinary nature of this research area, and is intended to generate dialogue and collaboration across traditional subject areas. It is stressed here that this article is not intended to act as a comprehensive review article, but rather an up-to-date and forward-looking summary of research methodologies pertaining to the treatment of pathologies at the level of the CNS.

1. Introduction 3 2. Microphysiological systems for preclinical testing of drug-loaded nanoparticle transport across the human blood-brain barrier 5 3. Advanced nanomaterials for ischemic stroke treatment 10 4. Nanotechnology-based neuronal interfaces 13 5. Brain cancer nanomedicine: State of the art and challenges 16 6. Nanoparticles-based developments for Parkinson's disease treatment 20 7. Nanomedicine advances in Alzheimer's disease treatment 23 8. Nanoparticles-mediated deep brain stimulation 26 9. Application of nanotechnologies in the treatment of neuroinflammation in neurodegenerative disorders 30 10.Nanoparticles-mediated immune therapy for the central nervous system 33 Data availability statement 35 References 35

Introduction
Gianni Ciofani Smart Bio-Interfaces, Istituto Italiano di Tecnologia, Pontedera, Italy (DBS), trying to overcome limitations of the current state of the art and indicating the challenges such technologies are expected to face in the next years. Section 9 reports on the importance of the evaluation and treatment of inflammatory conditions in the CNS, often associated to important pathological conditions. The possibility to modulate the function of neuro-inflammatory-related cells and preserving neuronal health is shown. Section 10, eventually, provides an overview of immune modulation at CNS level, highlighting its pivotal role in the treatment of CNS cancer and neurodegenerative diseases.  [22]. Schematic created with Biorender.com.
questionable their predictive value for human response [19]. In this context, there exists an unmet need for innovative in vitro models of the BBB that closely mimic in vivo brain endothelium in preclinical investigations, to serve as reliable tools to elucidate the role of the BBB in brain pathogenesis or for preclinical drug screening, with the aim of reducing the number of experimental animals, increase the efficiency of pharmacological research, and perform patient-specific studies to develop personalized treatments [13].

Advances in science and technology to meet challenges
Nanomaterial-based delivery systems have the capability to improve drug treatment specificity and shortand long-term efficacy. NP physical characteristics, such as size, shape, stiffness, surface and bulk composition can be tailored for optimal drug delivery. In particular, small NPs (<100 nm size) exhibit more favorable delivery across the BBB than larger ones [15]. Furthermore, although rod-shaped NPs have superior ability to cross vessel walls than spherical ones, they show decreased selectivity for brain tissue [23]. Low stiffness NPs have shown a superior ability to cross the vessel wall, including the BBB, than stiffer ones. Additionally, NP interactions with the physiological fluids should be minimized to avoid the corona effect and NP aggregation phenomena as well as to allow NP escape from the immune response. While pegylated NPs have been commonly used, anti-PEG (polyethylene glycol) antibodies have been recently identified [24], suggesting the need for new improved strategies avoiding rapid clearance of NPs for long-term efficacy. In addition, targeting ligands should be used in tandem with antifouling molecules to allow NP crossing of the BBB to reach the target brain cells [25]. To enhance delivery, the surface of NPs has been functionalized with antibodies, carbohydrates and other ligands to facilitate their transport across the BBB via transcellular pathways. As an example, low density lipoprotein receptor related protein-1 (LRP-1) has been identified as a selective receptor on both the BBB endothelial cells and glioma cells: NP surface functionalization with Angiopep-2 (a peptide binding LRP-1) have been shown to facilitate receptor-mediated transcytosis (RMT) of NPs across the BBB and targeted cargo delivery to glioma cells in the brain [26]. Furthermore, Figure 2. 3D microphysiological BBB model. (a) Schematic explanation of BBB model and protocol, from 2D culture of induced pluripotent stem cells-derived endothelial cells (iPSC-ECs), brain pericytes (PCs) and brain astrocytes (ACs) to generate a 3D microphysiological BBB model by self-assembled vasculature within a microfluidic device. The PDMS microfluidic platform was fabricated using soft lithography techniques and designed with inlet ports for injecting cell-gel suspensions, and large medium reservoirs and fluidic channels for culture medium. (b) Schematic of dynamic culture of the microphysiological BBB model over time in a section of 3D microfluidic system. Experimental steps and seeding configuration of vasculogenesis process of microphysiological BBB model including iPSC-ECs + PCs + ACs as self-assembled microvascular network that undergoes maturation within seven days of culture. Three-dimensional ECs layer covering top, bottom and side surfaces of the fluidic channels. (c) Confocal image of self-assembled microvasculature of the microphysiological BBB model including iPSC-ECs (CD31, green), PCs (F-actin, red) and ACs (GFAP, magenta), and nuclei (DAPI, blue). (d) Confocal images of xy and xz (cross-section) planes of the 3D microphysiological BBB model with iPSC-ECs + PCs + ACs, including EC layers in the side channels. Scale bars 200 µm. (e) Schematic and methods of 3D microvascular permeability measurements. Confocal images of transport of NPs microvasculature (in red) are displayed. (c), (d) Reproduced with permission from [29]. (e) Reproduced with permission from [30]. All schematics were created with BioRender.com. Preclinical models for testing nanoparticles transporting drugs across the blood-brain barrier using a microphysiological system. Schematic model of transport of nanoparticle across the blood-brain barrier in vivo and in vitro using microphysiological systems. Schematics created with Biorender.com. advancements in administration approaches and exploration of alternative delivery routes could improve the efficiency of drug delivery through the BBB. In this context, non-invasive methods, able to deliver drugs bypassing the BBB have been widely studied. On the other hand, invasive modalities for drug release to the brain have been also proposed, including convection enhanced delivery for intracranial injection of therapeutics, by generating a pressure gradient at the tip of an infusion catheter to deliver payloads directly into the interstitial spaces of the CNS [27,28]. NP design may be optimized by exploiting in vitro preclinical BBB models as testing platforms, as recently demonstrated in previous studies [29][30][31]. Particularly, BBB spheroids, organ-on-chip and micro-phsiological systems MPSs (figures 1(e) and (f)) have the potential to more closely recapitulate the microenvironmental characteristics and primary functions of human BBB. With respect to 2D models, MPS provide improved representation of complex dynamic cell interactions, and can reproduce blood flow and the whole 3D tissue structure with its barrier functions. Such characteristics are combined with high-throughput screening ability. Personalized nanomedicine design could also be possible through patient-specific MPS models making use of patient-derived human induced pluripotent stem cells [19] (figures 2(a)-(d)).

Concluding remarks
Human BBB models are of great interest to the scientific community and pharma industries for testing drug transport across microvessels in a 3D microenvironment with close similarity to the in vivo human brain microvasculature [29]. Advanced microphysiological models using microfluidic technology have demonstrated the potential to accelerate in vitro pre-clinical validation and screening of novel drugs and their nanovectors for effective therapeutic treatments [30,32]. Such systems are expected to facilitate a more comprehensive comparison among different drug candidates for an accurate preclinical assessment of their ability to cross the human BBB (figures 2(e) and 3). Hence, BBB models support the paradigm change in preclinical investigation from animal to alternative testing, according to the '3Rs Principle' (Replacement, Reduction, and Refinement). Important advancements in this field are represented by self-assembled MPSs of the human BBB [29][30][31]. The availability of simple and cost-effective protocols for the design of human BBB models and for their use in drugs and drug-loaded NPs testing is expected to have a high socio-economic impact, reducing the time and cost required for translation of basic science discoveries into clinical settings [22].

Status
Stroke is an acute brain disease caused by the lack of blood flow to brain cells either due to a hemorrhage or occlusion of cerebral blood vessels often leading to the dysfunction of brain cells and neuronal death. Approximately 87% of stroke cases are ischemic and the remaining 13% have been reported to be related to hemorrhage [33]. Cerebral ischemia is accompanied by a series of pathophysiological changes including oxidative stress, localized inflammation, neuronal damage and loss of integrity of BBB. All these events have significant role in brain damage [34]. Current management strategy for ischemic stroke includes reperfusion to restore blood flow in the brain by administration of tissue plasminogen activator or by performing thrombectomy (figure 4). But reperfusion is accompanied by generation of reactive oxygen species (ROS) which in turn can initiate inflammatory responses and often leads to tissue damage [35]. Administration of neuroprotectants has shown relatively low efficacy in alleviating reperfusion-induced injury in part because they are not able to tackle the biological complexity after an ischemic insult and due to their low efficacy to penetrate the brain [36,37]. Hence there is need to search for alternate treatment strategies to maximize clinical efficacy and transport through the BBB. Synthetic or biological NPs (e.g. EVs) offer potential to deliver therapeutic molecules in ischemic stroke such as neuroprotectants drugs across BBB, enhance their circulation half-life and promote their accumulation at ischemic sites [38][39][40][41]. This opinion article will focus on the advanced treatment strategies based on nanomaterials for the management of ischemic stroke.

Current and future challenges
In many cases, patients who survive a stroke event have limited functional recovery due to a limited remodeling and restorative process in the lesion area. Neuroprotective strategies targeting the cascade of cellular and molecular events that lead to ischemic damage, and strategies to promote post-ischemic regeneration, have been pursued in the last years, although without clinical translation [37,42]. Neuroprotective strategies targeting the cascade of molecular events that lead to ischemic damage as well as strategies to promote post-ischemic regeneration, have shown potential in preclinical models [43]. These include pharmacological interventions (e.g. free-radical trapping agents; magnesium; drug compounds like NA-1 which is a peptide fused with Tat protein-1) that target a single molecular target [43]. Nonetheless, so far, no therapy has demonstrated clear efficacy in clinical trials. However, interventions employing a cocktail of factors which can simultaneously target several molecular targets have not been tested in clinical trials. EVs are lipidic vesicles, with a diameter between 50 and 200 nm, that transport a cocktail of active molecules of the parental cell (DNA, mRNA, miRNAs, enzymes and growth factors) and act as mediators of cell-to-cell communication [44]. EVs present attractive features for therapeutic purposes because they regulate angiogenesis [45], neurogenesis [46] and synaptogenesis [46]. Currently, there is a running clinical trial that aims to evaluate mesenchymal stroma cell-derived EVs in the context of acute ischemic stroke (NCT03384433). Although functional benefits of EVs from different sources have been observed in animal models of ischemic stroke, their local neuroprotective/regenerative potential may be hampered by extremely low levels of accumulation in the brain after systemic delivery (typically below 1% of the initial dose) [47]. Thus, a main challenge is to develop strategies for the increased accumulation of EVs in the brain.
Better understanding of the physicochemical interactions of nanocarriers with BBB is needed to accomplish their clinical translation. Recent reports suggest that nanocarriers can cross BBB more effectively in the venules where blood flow is low and perivascular space is available [48]. Retention and absorption of the nanocarriers from venules can be achieved by active targeting of nanocarriers using ligands specific to receptors of brain endothelial cells in venules. But our understanding of proteins expressed by brain endothelial cells of venules, arteries and capillaries is limited. In addition, permeability of BBB is not uniform throughout and it has been reported that intravenously administered transferrin conjugated gold NPs could preferentially accumulate at neurogenic niches after activation with near infrared (NIR) radiation [49]. This depicts potential of exploring the feasibility of efficient BBB targeting across all parts of the brain such as cerebellum, thalamus, cortex etc. Stroke alters the BBB permeability and better understanding of BBB characteristics after stroke is needed to design more effective formulations for stroke therapy [50].

Advances in science and technology to meet challenges
Several NP-based delivery platforms have been investigated for the delivery of neuroprotectants, anti-inflammatory agents and imaging probes for ischemic stroke treatment (figure 5). Among these, liposomes were one of the first generation of nanoformulations employed. Induction of stealth property to the liposomes by PEGylation and use of active targeting ligands like transferrin depicted efficient drug delivery across BBB [51]. Stimuli sensitive polymer and metal based NPs were explored to enhance drug disposition to ischemic areas utilizing biological stimuli such as high ROS levels in ischemic regions [38][39][40][41]. Research sheds light in to the significance of using synthetic multivalent epitopes with tunable avidity to promote RMT across BBB [52]. Moderate avidity of the epitopes towards low density LRP1 [53] or transferrin receptor [49] facilitated the transcytosis transport across the BBB. EVs are also emerging as therapeutic delivery vehicles for brain diseases [54]. EVs have been used successfully for the delivery of siRNA into the brain after intravenous injection [55]. A recent study showed that tumor-derived EVs showed enhanced transcytosis across BBB through modulation of the endothelial recycling endocytic pathway. Specifically, EVs reduced expression of rab7 which in turn promoted the efficiency of BBB transport [56]. Despite the exact mechanism by which EVs breach BBB is still unclear these findings can help in designing more efficient EV based formulations for BBB transport. Another significant advance in the BBB targeted drug delivery is the selective targeting of NPs to BBB by labeling the brain endothelial cell surfaces with specific ligands. NPs with specific affinity to these ligands can preferentially bind the brain endothelial cell surface thereby minimizing off target effects [57]. Designing an optimal in vitro BBB model mimicking the properties and complexity of BBB is also critical in screening efficient formulations. Researchers have developed promising microfluidics based in vitro BBB model which can be a useful tool to investigate the permeability of drug molecules [31]. Yet, the possibility to perform these screenings in high-throughput while taking in account the heterogeneity of the BBB in terms of composition and flow dynamics remains elusive.

Concluding remarks
Successful clinical translation of various NP based formulations in ischemic stroke needs thorough understanding of pathogenesis of ischemic stroke and mechanism of interaction of these NPs with ischemic tissues. Route and frequency of administration of NP based therapeutics and optimization of their therapeutic windows are also important determinants of successful clinical outcome. Apart from commonly employed intravenous injection, alternative routes of administration such as nose to brain delivery can also be explored for drug delivery to ischemic brain areas. Facilitation of the transport of drug molecules across BBB by the NPs also needs to be addressed in more detail with the help of advances in design and development of nanoformulations and in vivo imaging systems such as intravital microscopy. Development of new drug molecules and novel NP based delivery systems will improve the clinical outcome in the treatment and long-term management of ischemic stroke.

Status
Implantable neural interfaces (INIs) are devices capable to stimulate or record nervous system activity to restore a lost or compromised neurophysiological function. INIs function requires the intimate contact of the device with the nervous tissue to modulate neuronal activity for the treatment and diagnosis of cognitive or sensory-motor disorders [58]. A wide plethora of INIs examples has been presented nowadays, ranging from DBS electrodes for mental disorders management (including PD), cochlear implants, retinal prostheses, and electrocorticogram electrodes to record brain activity. Furthermore, research on microfabrication technique enabled progress in electrode miniaturization, reaching extremely high spatial resolution. In this regard, Utah array, a silicon-based INI capable of dense sampling of multiple brain regions simultaneously, flexible polyimide-based microelectrodes with improved cells/implant interface and micromagnetic stimulation technology guarantee better integration with surrounding tissue [59,60].
However, there is a common bottleneck that impedes long-term usage of this technology: geometry and physiochemical properties of INIs materials differ completely from those of the tissue with which they must interface. This occurrence causes INI electrical performances drop due to fibrotic tissue encapsulation. Furthermore, tissue damage due to chronic inflammation response and micromotion at the tissue/electrode interface are other severe consequences of the consistent mechanical mismatch between INIs and the brain tissue. Moreover, most of the present INIs typologies requires percutaneously cabled electronics to connect the electrode with the power source. This poses further biocompatibility and encumbrance issues and a second invasive surgery to remove the implant is needed to avoid the risk to worsen nervous tissue damage.
There has been a tremendous effort in the last decade to overcome these limitations to envision long-term usage of INIs. Research on material science is allowing remarkable improvement to enhance INIs biocompatibility by developing new polymeric conductive materials, nature-derived compounds, and hybrid materials to reduce chronic inflammation response and to further miniaturize the implant. Efforts have been also made to avoid cable communication, by developing new wireless-controlled devices. However, despite these progresses INIs technology is still far from a successful long-term use and more efforts are needed to enhance its clinical relevance.

Current and future challenges
In the recent years nanotechnology is giving a significant contribution in overcoming the lack of long-term efficiency of INIs (figure 6). Nanoscale materials and nanofabrication techniques could in this sense operate a sort of revolution that would allow to build the ideal INI: a device capable to modulate neuronal activity with single-neuron resolution for long-term usage that does not affect the integrity of the nervous tissue and capable of wireless control. As an example, nanomaterials as carbon nanotubes (CNT) and graphene have been reported to successfully improve electrode signal-to-noise ratio and to enhance charge injection capacity, when used as conductive materials in electrode fabrications [61]. By employing such nanostructures, enhancement of mass transport phenomena with radial diffusion occurs, which results in improved sensitivity of the electrode due to. Moreover, nanostructures such as gold and platinum NPs and zinc oxide nanowires exploited as electrode nano-coatings have been reported to ameliorate recording sensitivity by reducing electrical impedance due to their high surface-to-volume ratio [62,63]. In addition to electrical performances, nanotechnology advances cold help to improve INIs flexibility, and electrode/tissue interface properties as well. Conductive polymers (CPs) are another class of nanostructured materials that combine biocompatibility, similar mechanical properties respect to native tissue and metal-like conductivity. These materials can be employed as substitute to bare metals as conductive elements of INIs as demonstrated by studies that reported the fabrication of highly flexible, porous and injectable hydrogel using poly (3, 4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) as CPs [64][65][66].
These reported examples illustrated some of the most promising research areas that demonstrate the impact of nanotechnology in overcoming the bottleneck of current INIs technology. However, in our opinion nanotechnology could represent the most promising future challenge in NIs technology, giving the opportunity to even change its actual paradigm: by employing nanostructured entities as the INI itself instead of probes, electrodes, or flexible structures. In this regard, scaling down materials dimensions allows the exploitation of new physical properties of the matter, because different energies such as electrical, optical, thermal, and mechanical all converge at the nanoscale, enabling energy conversion and transduction from an energy type to another.

Advances in science and technology to meet challenges
Switching from electrodes to nanostructured materials as the INI core represents the most promising challenge in this technology. In the past years there has been a strong interest to investigate the use of nanostructured materials as neuromodulating structures.
Nanostructured materials could work as nanotransducers, by converting an external energy source to another type of energy to stimulate or sense neuronal activity at the cellular level [66]. Importantly, this phenomenon occurs in wireless configuration, avoiding the needs of cables and bulky connectors. Several class of nanotransducers for wireless neuromodulation have been presented nowadays (figure 7). Optoelectrical nanotransducers such as quantum dots and conjugated polymer NPs have been demonstrated to trigger neuromodulating effect upon light conversion into electrical potential. A recent example showed that subretinal injection of poly [3-hexylthiophene] (P3HT) NPs allowed light-evoked stimulation of retinal neurons and rescue visual functions in a rat model of retinitis pigmentosa [68].
Temperature has been reported to modulate nervous activity by stimulating or inhibiting it. Although its mechanism is not fully understood, changes in membrane capacity and the activation of temperature-gated ion channels are thought to play a crucial role in triggering temperature-mediated neuromodulating effect. Gold NPs have been already demonstrated to efficiently convert NIR light into local heat to stimulate or inhibit action potential propagation [67]. Other promising examples of wireless neuromodulation are piezoelectric NPs. These nanostructures can convert external mechanical force, such as ultrasound, into local electric fields to modulate voltage-dependent ion channels, in order to trigger action potential. In this regard, barium titanate NPs have showed neuromodulating potential, as local currents generated by ultrasound can induce Ca 2+ influx into cell cytoplasm [69].
All these nanomaterials show remarkable potential to change the current paradigm of INIs technology. For this reason, future research is strongly needed to develop strategies to efficiently target these nanostructures across the brain, overcoming physiological barriers such as the BBB. Furthermore, more studies are also needed to understand nanomaterials interactions with immune system, as well as the mechanisms of cellular interaction and clearance of these nanostructures. This is of fundamental importance to envision long-term usage and future clinical translation of nanotransducers-based neuromodulation.

Concluding remarks
Scaling down materials dimensions allows to explore extraordinary physical properties only exhibited at the nanoscale. The employment of nanostructured and hybrid materials (such as carbon-based materials and CPs) has already been shown to enhance INIs electrical performances and to reduce the mechanical mismatch at the electrode/tissue interface, which is primarily responsible for their lack of efficiency in chronic experiments.
In our opinion, the most promising contribution driven by nanotechnology to extend the clinical relevance of INIs will be given using nanostructured materials as wireless nanotransducers for neuromodulation. Future studies to better understand their cellular interactions and brain targeting will be of primarily importance to understand the correct dose to trigger the desired effect and the most appropriate route of administration to reach the target neurons. Furthermore, development of portable energy sources and closed-loop control devices for energy transduction will allow the use of this technology even outside of healthcare facilities.

Status
Brain cancer represents one of the most difficult conditions to treat. In particular, glioblastoma multiforme (GBM), which represent 47.7% of brain tumors, is one of the deadliest tumors, and is characterized by an extremely dismal prognosis [70]. The current gold standard treatment for GBM includes surgery, chemotherapy and radiotherapy; in particular, the Stupp protocol, based on the administration of the drug temozolomide in combination with radiotherapy, is the most common therapeutic regimen [71]. Another therapeutic approach consist of the application of a carmustine-loaded biodegradable wafer (Gliadel ® ) directly in the resection cavity during surgery [72]. Nevertheless, despite the efforts, GBM prognosis is still very poor, with an average five-year survival rate of 5.6% [70]. This poor outcome is related to the complex nature of brain cancer and the therapeutic challenges related to it, that conventional therapies fail to address in an efficient way. Therefore, there is an urgent need of new approaches to improve efficacy, while reducing the severe side effects related to the aspecific distribution of conventional drugs within the body after systemic administration. In this sense, a lot of effort has been put in designing drugs that are highly bioavailable and that specifically target cancer cells. Owing to their peculiar physicochemical properties (high surface-to-volume ratio, sizes in the nanoscale), tunable morphology and composition, easy preparation protocols and possible surface functionalization, biocompatible NPs emerged as a promising tool to treat cancer, creating a new branch of nanotechnology, called nanomedicine. NPs offer several advantages with respect to conventional chemotherapy: (a) they can encapsulate hydrophobic molecules that are difficult to administer in biological fluids, improving their solubility and biocompatibility; (b) they can release drugs in a controlled manner and, in some cases, the release can be triggered by external stimuli; (c) they can be easily conjugated to ligands that are specific for cancer cells, imparting targeting abilities to the NPs and favoring their accumulation in tumor tissues [70]. Besides being a delivery agent for drugs or other active compounds, NPs can also have an active role in cancer medicine, being themselves therapeutic or contrast agents (superparamagnetic iron oxide NPs, gold NPs or quantum dots, for instance). Currently, there are a plethora of NPs that can be employed to treat brain cancer, and, depending on their composition, they can be classified as organic (e.g. liposomes, polymeric NPs or nanostructured lipid carriers) or inorganic NPs.

Current and future challenges
Brain cancer presents several therapeutic challenges that make its treatment very problematic. First of all, the BBB that normally protects the brain from harmful substances, represents one of the main obstacles for the delivery of therapeutic compounds to the brain, since most of the conventional drugs are unable to efficiently cross it [73]. On the other hand, NPs can be functionalized with ligands that interact with receptors overexpressed on endothelial cells and that can trigger active transport mechanisms, such as RMT [73]. For instance, peptides derived from the specific amino acid sequence of the apolipoprotein E (ApoE) bind to the low-density lipoprotein receptor of capillary endothelial cells; thus, NPs functionalized with ApoE have a higher BBB crossing efficiency [74,75].
Another goal in nanomedicine is to favor the accumulation of therapeutics within diseased tissues, in order to reduce side effects on the healthy ones. NPs are known to preferentially accumulate in tumors via passive targeting exploiting the so-called enhanced permeation and retention effect. This phenomenon depends on the size of the NPs, on the abnormal vascular architecture around tumors, that favors extravasation, and on the lack of a proper lymphatic drainage [76]. However, passive targeting is difficult to control, with consequent poor drug diffusion and aspecific accumulation in liver and spleen and it can induce multidrug resistance [77]. Active targeting, instead, relies on the interaction between a ligand (e.g. antibodies, peptides, small molecules or aptamers) attached to the NPs surface and a receptor overexpressed by the target cells. For instance, NPs functionalized with the peptide angiopep-2 can selectively target GBM cells and have improved BBB crossing abilities [78]. More recently, a new strategy exploiting the ability of cancer cells to recognize each other has been developed. In fact, tumor cells can form multicellular aggregates thanks to the interaction between specific proteins in hemophilic adhesion domains of plasma membranes, or between tumor-specific binding proteins [79]. By mimicking this natural tendency of cancer cells to homotypic recognition, coating NPs with cancer cell membranes extracts could significantly increase their uptake in the tumor cells. This targeting strategy, referred as 'homotypic targeting' , accounts for the intricate interactions that requires simultaneous binding of different ligands to efficiently target cancer cells [80].
Another future challenge in brain cancer nanomedicine will be to design NPs able to exert a therapeutic action when remotely activated by an external stimulus (e.g. magnetic fields, light irradiation and ultrasound). This will guarantee a less invasive and 'on demand' treatment, activated only when the NPs are effectively located in the tumor area, avoiding potential harmful effects on healthy cells.

Advances in science and technology to meet challenges
In recent years, nanomedicine has made huge progresses in overcoming the main issues related to the treatment of brain cancer. As mentioned before, the improvement of the BBB crossing abilities of NPs is one of the biggest challenges to meet. Several NPs functionalized with ligands interacting with receptors on endothelial cells surface (transferrin, lactoferrin, insulin and low-density lipoprotein receptors) that activate RMT have been developed to enhance their crossing efficiency [71]. Another proposed solution is to exploit immune cells (e.g. neutrophils, monocytes, and macrophages) to activate cell-mediated transcytosis; for instance, drug-loaded liposomes can be absorbed by immune cells circulating in the blood and transported through the BBB towards the inflammation site in the brain exploiting immune cells properties called diapedesis and chemotaxis [81].
In order to improve the accumulation of nanomaterials in tumors, researchers have proposed several strategies. Among them, functionalization with peptides or aptamers interacting with receptors overexpressed on tumor cells has demonstrated to be an effective targeting strategy. Nevertheless, when dealing with genetic heterogeneous tumors such as GBM, approaches relying on one or two single interactions are often inefficient. In these cases, homotypic targeting has shown very promising results in the accumulation of therapeutics in the tumor site. For instance, boron nitride nanotubes loaded with doxorubicin and coated with GBM cell membrane extracts were shown to selectively target GBM cells, while the uptake and, as a consequence, the cytotoxicity on other healthy cells used in the study were not observed (figure 8) [82].
New NPs exerting an anticancer action only when activated by an external stimulus have been also developed. Nanostructured lipid carriers loaded with superparamagnetic iron oxide NPs were able to induce apoptosis in GBM cells through hyperthermia-induced lysosomal membrane permeabilization after a chronic stimulation with a proper alternated magnetic field [76]. An innovative approach to treat brain cancer is represented by the ultrasound stimulation of organic piezoelectric NPs. Treatment with these NPs loaded with a drug, followed by chronic ultrasound stimulation, led to the activation of cell apoptosis and anti-proliferation pathways, induction of cell necrosis, inhibition of cancer migration, and reduction of cell invasiveness in drug-resistant GBM cells (figure 9) [75].  , encapsulating the non-genotoxic drug, nutlin-3a, and functionalized with a ApoE-derived peptide, that enhances the BBB crossing abilities of the nanoparticles. Upon ultrasound stimulation, the nanoparticles were able to reduce cell migration, thanks to a reduce f-actin/g-actin ratio, and to foster apoptotic and necrotic events (Reproduced from [75] Copyright 2021, with permission from Elsevier).

Concluding remarks
Nanomedicine has shown promising results for the treatment of brain cancer by offering several approaches to cross the BBB, improving the systemic delivery of drugs and favoring their accumulation in tumor tissues. Nevertheless, in the future, research will have to focus on making the treatment specific for the patient, moving towards precision medicine. Since brain tumors are extremely heterogeneous between different patients, it would be desirable to be able to develop nanotherapeutics that can be adapted to the specific needs of the patients. This will increase treatment efficacy, while reducing side effects. Nanotherapeutics should be tested directly on patient-derived cancer cells in order to choose the best therapeutic approach and/or should contain features of the patient's cancer cells (e.g. coating with cell membrane extracts) to adapt the targeting abilities to the specific membrane proteins expression of the tumor.

Status
Degeneration and loss of dopaminergic neurons in the substantia nigra pars compacta, and subsequent reduction of DA levels in striatum, are associated with motor symptoms that characterize PD. In fact, since PD is a multifactorial disease where both genetic and non-genetic factors are involved, the most prominent mechanisms related to the development of this disease include the accumulation of misfolded proteins aggregates (i.e. α-synuclein, ubiquitin, PTEN-induced kinase-1, parkin, and other proteins), failure of protein clearance pathways, mitochondrial damage, oxidative stress, excitotoxicity, neuroinflammation, and genetic mutations [83]. Several treatments are available, but none of them is notably effective to reduce neuronal loss and restoring DA levels. It has been introduced some promising alternative strategies, such as stem cell transplantation and gene therapy. However, most of them are still under investigation and safe/efficacy need to be adequately addressed before clinical trials. During the last years, extensive studies to understand the molecular signaling pathways involved in PD, the involvement of molecular chaperones, autophagy-lysosomal pathways, and proteasome systems, have been addressed. In addition, emerging therapies such as pharmacological manipulations, or surgical procedures, are proposed as alternative treatments.
The incorporation of conventional drugs into NPs, the so-called nanocarriers, has represented a step forward in conventional medicine for the treatment of different diseases and specifically of PD [84]. Of the many advantages, NPs stabilize hydrophobic drugs and facilitate crossing the BBB of different nowadays active systems such as enzymes, proteins and even DA, seeding novel drug therapies beyond the gold-standard L-Dopa therapy (figure 10). Nanocarriers also offer major pharmacokinetics advantages such as controlled drug release over time while avoiding early metabolism and phagocytosis, facilitate targeting to specific cells (improving efficacy, safety, sensitivity and personalization) and allow for the delivery to the brain of combined drugs, antioxidant agents, neurotrophic, and neuroprotective factors, as well as antiapoptotic factors, or even gene therapy. All these advantages allow us to predict a great advance in this field for the years to come though further studies are still needed, especially to overcome the BBB. In this context, the intranasal administration has been proposed over the last years as a novel route to bypass the BBB and reduce systemic side effects with respect to oral or systemic administration [85]. Several nanoformulations have been already developed showing an enhancement of nose-to-brain drug transport for neurodegenerative diseases [86], including coordination polymer NPs for intranasal DA replacement in PDs [87]. Based on the intranasal physiognomy, there are two possible passages from nose to brain (i.e. the olfactory nerves that end up at the olfactory bulb and the maxillary branch of the trigeminal nerve). The goal of different studies is to track the translocation of NPs and the payloads along these nose-to-brain pathways.

Current and future challenges
As previously stated, nanotechnology is a promising approach to facilitate PD patients management and design more selective and effective therapies. Moreover, it could be very important to understand the pathophysiology of the disease, achieve earlier stages of diagnosis, and offer better treatment options. Though, major challenges are still to be faced before its implementation becomes true: • Nanocarrier improvement. Future developments involve biodegradable therapeutic nanocarriers with optimized drug encapsulation yields and controlled release in response to external stimuli. • Drug research. Further developments in using novel DA drugs (e.g. use of IPX066, XP21279, and Opicapone) or non-DA drugs (e.g. α2-adrenergic antagonists, serotonergic, or adenosine A2a antagonists), or even the use of micro-RNA or Si-RNA approach to inhibit mRNA of misfolded protein aggregates, which may offer beneficial effects in late-stage developments of motor symptoms in PD, are needed; The exhaustive study of the PD mechanisms and use of novel gene editing techniques (e.g. CRISP-Cas9) for correction of mutated genes involved in PD, will improve the election of the adequate drug and therapy design for an effective PD treatment. • Intranasal administration. Further research is required to better understand the drug passage mechanism through the intranasal route to specific areas of the brain depending on the NP formulation and its membrane permeability, mucocilliary clearance, or enzymatic degradation [88]. • Targeting. The specific targeting of dopaminergic neurons would be strongly desired to reduce secondary effects while ensuring a proper biodistribution (drug bioavailability) [89]. • Side-effects. Although nanocarriers overall reduce the toxicity of the free drugs, NPs can induce the triggering of inflammation, oxidative stress, and gene overexpression [90].

Advances in science and technology to meet challenges
Until now, several nanocarrier systems have been successfully applied such as chitosan NPs, target functionalized liposomes, (bio)hydrogels, CNTs, graphene-based nanosystems, polymeric nanocapsules or magnetic NPs [91]. For most of them, size, shape and charge can be systematically tuned, though much work is still needed to develop responsiveness in front of pH changes, enzymatic action or redox stimuli. Exploration of novel and disruptive nanocarriers would also benefit the field and offer rational designs for drug transport and effective and selective drug release in the site of action, minimizing doses and side effects as well as increasing the colloidal and chemical stability of the NPs. The development of novel and non-invasive administration routes, such as nose-to-brain delivery, opens up a host of possibilities to reach the brain in an effective manner. In this sense, the use of a suitable mucoadhesive coating on the NPs (e.g. mucoadhesive polymers, gelatins, hydrogels) would favored the mucoadhesion enhancing the retention time in nasal cavity and reducing mucociliary clearance ( figure 11). Apart from pharmacological treatments, the monitoring of DA levels in vivo and in real time is necessary to understand its physiological roles. In this sense, different sensors and biosensors have been developed, as those based in measuring neurotransmitters with fast-scan cyclic voltammetry (FSCV) using multielectrode Figure 11. (a) Different nanocarriers developed for other administration routes can be modified for nose-to-brain administration. For the validation of the resulting nanoparticles, it is essential to investigate pathogenic pathways at the whole-organism level in PD animal models (b) for further optimization of the nanoformulation before its clinical use (c).
arrays [92]. These studies help to better understand the complex brain heterogeneity, the dynamic neurochemical environment, and how disease states or drugs affect separate brain areas concurrently.

Concluding remarks
The limited efficacy of drugs to treat neurodegenerative diseases has become a major challenge over the years due to different physiological factors such as enzymatic degradation, systemic clearance, peripheral side effects and reduced bioavailability. The encapsulation of drugs within NPs has shown a clear improvement in this respect. The use of nanoformulated systems allow using new insoluble or chemically unstable drugs (or a combination of them) as well as design an adequate biodistribution and targeting. So far, many examples have been described in the literature, some of them even reaching clinical phases, leading us to expect very promising results soon even though the development of these new nanopharmaceuticals is at a very early stages. A very relevant aspect of this research is the drastic limitations imposed by the BBB to improve current therapies. In this scenario, the investigation of alternative administration modes to bypass or avoid the BBB crossing, such as intranasal administration, is an important aspect to improve PD treatment. In fact, a large number of active principles have already being successfully delivered through this non-invasive route to the CNS. If clinical studies support such preclinical data, intranasal drug delivery may revolutionize treatments for brain disorders. Finally, we would not like to end this section without mentioning precision medicine therapies as one of the greatest expectations for the future in this area.

Giulia Sierri and Francesca Re School of Medicine and Surgery, University of Milano-Bicocca, Milano, Italy
Status AD is an age-related, irreversible form of dementia characterized by the progressive degeneration of cognitive performance [93]. In developed countries, about 50 million people are affected by dementia and the incidence of AD will increase because of the ongoing increase of the aged population [94]. This makes AD a growing concern that harms the global health care system. The cardinal features of Alzheimer pathology are the cerebral accumulation of β-amyloid (Aβ) peptide in toxic oligomers and amyloid plaques, neurofibrillary tangles of aggregated tau protein, synaptic dysfunction, neuronal death, and cognitive impairment. However, AD is a multifactorial disease with several additional pathogenic mechanisms, including inflammation, oxidative damage, iron dysregulation, mitochondrial dysfunction, and altered cholesterol metabolism [93,95]. Almost all current AD treatments merely delay the onset of symptoms, without modifying the course of the disease.
The majority of disease-modifying treatments (DMTs), i.e. those proven to alter the underlying disease pathology or disease course, are nearly every designed to specifically target Aβ. Antibodies tackling Aβ are in advanced clinical trials, and Aducanumab, which clears Aβ, was approved by the US Food and Drug Administration (FDA) in 2021, not without controversy. Indeed, the scarce beneficial effects on cognitive decline reported by treated patients drove the European Medicines Agency (EMA) to the refusal of the marketing authorization for Aducanumab.
Overall, the development of DMTs for AD is complicated by the presence of the BBB, which prevents certain drugs and large-molecule therapeutics from entering the brain.
Therefore, strategies boosting the passage of drugs able to modulate multiple disease-associated pathways across the BBB represent a key opportunity to treat AD.
Nanotechnology has emerged as an exciting and promising strategy to treat neurological disease, with the potential to fundamentally change the way to approach brain-targeted therapeutics. In this field, NPs can be tuned by controlling their physiochemical properties, such as size, shape, surface charge, and hydrophobicity to cross the BBB, to bind Aβ, to enhance their anti-amyloid capacity and/or to deliver DMTs to the brain. Accordingly, one phase 2 trial based on the intranasal administration of APH-1105, an alpha (α)-secretase modulator formulated in NPs, is being estimated to start in June 2023. Its safety, tolerability, and efficacy are being evaluated for the treatment of subjects with mild-moderate AD [96].
However, further advances in NPs engineering could be a breakthrough in AD treatment, taking into account the timing of interventions and the population to be recruited [97].

Current and future challenges
There are several issues in seeking effective medicines for AD. One of the biggest obstacles is the presence of the BBB, a formidable challenge to the delivery drugs into the brain. Several compounds have not shown efficacy in clinical trials, because they generally fail to cross the BBB, and the use of NPs as drug delivery system offers an alternative approach for promising and innovative therapeutic solutions for AD.
The NPs, suitably functionalized and with a long blood-residency time, can cross the BBB and release active molecules at target sites in the brain, minimizing side effects. It has been demonstrated that the concentration of drug that reaches the brain is higher if it is formulated in NPs, rather than when administered alone [98]. For example, it was shown that the concentration of rivastigmine that reaches the brain is about four times higher when embedded in NPs than when it is free [99]. Also the administration route of NPs may affect the bioavailability of the drug delivered into the body. For example, nasal administration route is the most practical and non-invasive method to administer drugs. However, the nasal cavity has enzymes that can could affect the bioavailability of the drug and NPs [100].
An additional challenge for AD therapy is the multifactorial nature of the disease ( figure 12). Indeed, the different pathogenic mechanisms involved in AD, other than the most known disease hallmarks, make the situation more complicated, impacting on the strategy adopted.
Another issue could be the potential 'ancillary effect' of NPs, which is affected by the inability of the various clearance systems to remove them from the brain, and by their physicochemical properties that could interfere with physiological pathways, as seen for inorganic or metallic NPs. Another limitation could be the expensive process that leads to the production of NPs, because specific materials, instruments, and optimal conditions are needed in order to obtain specific multifunctional NPs (figure 13). Alzheimer's disease is a multifactorial disorder. Multitarget drugs have been increasingly sought after over the last decades, however strategies to enhance their entry in the brain thus enhancing the therapeutic efficacy are sought.

Advances in science and technology to meet challenges
The application of nanotechnology in the treatment of AD includes NPs targeting Alzheimer's Aβ or NPs active on other pathways involved in AD pathogenesis or progression.
Among the most promising NPs designed to tackle Aβ, we can cite: (a) Poly(lactic-co-glycolic acid) (PLGA)-NPs embedding the peptide (iAβ5) able to inhibit the Aβ fibrillogenesis, and surrounded with anti-transferrin receptor monoclonal antibody (OX26) and anti-Aβ (DE2B4) useful to deliver the drug across the BBB [101]. (b) Liposomes functionalized with phosphatidic acid and surface decorated with a modified peptide derived from the receptor binding domain of apolipoprotein E (mApoE), that have the dual ability to affect Aβ 42 aggregation/disaggregation processes and to cross the BBB [102]. Among the NPs designed to target AD-related pathways, we can cite: 1. NPs with antioxidant activity useful to counteract ROS-mediated cerebrovascular dysfunctions, which are considered worsening factors in the progression of AD [103]. 2. Discoidal NPs functionalized with ApoE A-I suitable as a potential supportive treatment to compensate the depletion of cerebral HDLs occurring in AD [95]. 3. Selenium NPs encapsulated PLGA nanospheres with curcumin to decrease Aβ-dependent inflammation [104].
The application of nanotechnologies includes also NPs for AD diagnosis. Polymeric NPs and SPIONs developed for the imaging of amyloid aggregated, have demonstrated great potential in this field [105]. A new strategy is theranostics that combines diagnosis with therapeutic approaches in order to have a unique device that recognizes and identifies the biomarkers of the disease, for example Aβ plaques or the tau tangles, and at the same time reaches them and releases specific drugs.
In general, the attention has focused on lipid NPs, because as being made-up of lipids, they are biocompatible and tolerated by our body. They show several advantages over other systems, including an Figure 13. Multifunctional nanoparticles. Nanoparticles with multiple properties (moieties) can be designed to deliver drugs active on AD in the brain. easy large-scale production, biodegradability and biocompatibility materials, low toxic potential, the ability to control or modify drug release, and the ability to incorporate hydrophilic and lipophilic drugs.

Concluding remarks
Achieving sufficient delivery across the BBB is a clinical need in the development of drugs to treat cerebral disorders, because many biopharmaceuticals are largely excluded from the brain after peripheral administration. When it comes to treating AD, the per se difficult drug discovery enterprise becomes a titanic challenge, because it is a multifactorial disorder.
Different big pharma are retreating from drug development for AD because of the negative results of more than 200 investigational programs failed in the last decade. Nevertheless, the recent (2021) FDA approval of Aducanumab in mild-moderate AD demonstrate that Aβ-centered strategy is still alive, revitalizing the AD drug development and opening new scenarios for nanotechnologies. In fact, pharmaceutical and biotechnology companies are joining forces with academic institutions and public-private consortia active in this field to develop nanotechnologies against AD.

Nanoparticles-mediated deep brain stimulation
Prachi Kumari 1 , Kristen L Kozielski 1 and Hannah Wunderlich 1 1 School of Computation, Information and Technology, Technical University of Munich, Munich, Germany Status DBS is an effective and standard treatment for a range of neurological disorders, most commonly for PD and essential tremor. Clinical DBS implants require invasive surgery, and can be associated with various risks due to the large implanted devices. Recently, in order to mitigate these risks, interest in smaller, less invasive methods for stimulation have led to a focus on the use of NPs. Various materials have been explored for these NPs, which can act as transducers of external stimuli such as magnetic fields, ultrasound, or light. In addition, considerable research has been conducted in less invasive NP delivery routes, such as through the BBB and via cell-or tissue-specific NP surface modifications. As NPs used specifically for DBS is a rather new and less explored field, this section will address nanomaterials used for neuronal stimulation, that may be applied to DBS in the future.

Current and future challenges
In clinical DBS, an electrode in the deep brain is wired under the skin to a battery-powered device that provides electrical signals for neural stimulation. Despite the long history and success of clinical DBS, these devices are susceptible to hardware damage and infection risk, and thus often require corrective surgery [106]. The battery, which is implanted under the skin, must also be replaced at regular intervals. As a result, fewer patients undergo this treatment than who would be medically suitable for it [107,108]. To overcome the drawbacks of a larger implanted electrode, new strategies have been presented in preclinical research in recent years in the form of NP-mediated neuromodulation.
A common strategy in the earliest NP neuromodulation technologies is to utilize genetic engineering to introduce heat-, mechano-, or light-responsive ion channels into neurons. Magnetothermal stimulation was among the earliest of these, first showing stimulation of deep brain tissue in mice using 22 nm magnetic NPs and heat-sensitive ion channel TRPV1. To enhance colloidal stability and improve biocompatibility NPs were coated with poly(acrylic acid) and PEG [109] Using this approach, magnetothermal DBS has also shown alleviation of PD symptoms in mice [110]. Magnetothermal stimulation can also be used for neuronal suppression. Using superparamagnetic MnFe 2 O 4− CoFe 2 O 3 core-shell NPs (12.9 ± 1.4 nm) to provide heating, Munshi et al inserted a temperature sensitive chloride channel, which when heated, causes Cl − influx and hyperpolarization with 2 s latency ( figure 14(c)). To improve suspension properties NPs were coated with poly-isobutylene-maleic anhydride which could be further functionalized with required macromolecules [111].
Mechanosensitive ion channel introduction also enables remote magnetic stimulation, but via the application of mechanical force. Ferromagnetic NP 'm-Torquer' with an overall dimension of 500 nm was used to remotely apply torque to a Piezo-1 mechanosensitive cation channel in vivo, as verified by c-fos staining. Piezo-1 was modified with a Myc-tag, and m-Torquer was PEGylated and conjugated with an anti-Myc antibody to enable force transduction (figures 14(a) and (b)) [112] Another mechanosensitive ion channel, TRPV4, which is naturally expressed in dorsal root ganglion (DRG) cells, was used for transduction of neural stimulation via torque from ferrimagnetic iron oxide nanodiscs with sizes varying from 98-226 nm (MNDs). These nanodiscs were modified with oleic acid which was functionalized with amphiphilic poly(maleic anhydride-alt-1-octadecene) for better colloidal stability. MNDs use the transition from vortex to in-plane magnetization, such that they have a near-zero net magnetic moment in the absence of a magnetic field, and therefore less risk of aggregation. Despite the advantages of using endogenous ion channels and MNDs, one drawback of this approach highlighted by the authors was the difficulty of delivery of MNDs due to their substantial size [113].
Another strategy in neural modulation is NPs that use light to stimulate the brain. The photoelectric approach uses light as the input source of energy, and is converted into electrical signals by nanomaterials [68,114]. Conversely, in optogenetics, light is the stimulus signal (i.e. transmitted from the NPs to neurons), which requires genetic transduction to express light-sensitive ion channels in cells [115]. The spatial selectivity of light enables even single neuron resolution. The disadvantage, however, is that the penetration depth of the light often does not reach the areas of the brain that need to be targeted for DBS.
Overall, wireless signal transduction via NPs allows flexibility in input energy sources to produce neural stimulation. However, NPs have their own challenges, which need to be overcome. Additionally, studies in primates, and long-term studies have yet to be conducted, which will be necessary for successful to translation of NP DBS therapy to humans. Reproduced with permission [111,112,114,115,118].

Advances in science and technology to meet challenges
The earliest technologies in NP-based neural stimulation opened up a field in neuroengineering which had previously been reserved only for centimeter-scale, battery-operated, and surgically-implanted devices. While the translational potential of some early NP technologies is limited, we will now present the progress that has been gained in recent years.
Upconversion nanoparticles can be used to overcome the limited penetration depth of light with visible or ultraviolet wavelengths. This was achieved by Yadav et al by doping 39 ± 1.5 nm silica NPs with lanthanides (Yb 3+ , Tm 3+ ) that absorb light in the near-infrared region. Longer input wavelengths can thus reduce power loss with increasing tissue penetration depth ( figure 14(d)). However, this technology still required transduction with the light-sensitive ion channel channelrhodopsin-2 (ChR2) [115]. Also, while longer, infrared wavelengths may extend penetration depths, this is still insufficient to reach deep brain tissue from outside the human skull.
An approach also based on light as an input stimulus, which does not rely on genetic modification, was developed by using semiconducting 280.5 nm silicon nanowires (SiNWs). Here, SiNWs were stimulated with a focused light beam after spontaneous uptake into cultured oligodendrocytes. Photoelectric stimulation of the oligodendrocytes led to downstream electric stimulation of cocultured DRGs (figure 14(e)) [114]. In this case, it was shown that the intentional uptake of the NPs into the cells did not affect viability, did not interfere with cell mitosis, and did not affect the functionality of the NPs [114]. Whether this is also the case with other NPs where uptake is unintended remains to be elucidated, as well as if uptake is advantageous in electrophysiological stimulation.
Another example of photoelectric stimulation, in which light is converted into an electrical stimulus, is the use of conjugated polymer NPs (P3HT) in the retina. Using a rat model of degenerative retinal disease, it was shown that spared retinal neurons could be stimulated by light using these NPs which have a diameter of 300 nm. The effect lasted up to eight months after a single injection, restoring the response to visual stimuli as well as cortical and subcortical activity [68]. Like all light stimulation-based technologies, the disadvantage of using SiNWs and P3HT NPs for DBS is the loss of power with increasing penetration depth into the deep brain, as light needs to be delivered in close proximity of the desired target for stimulation. However, the longevity of P3HT NPs still provides some insight with regard to long-term functionality of nanostimulators. Not only was functionality proven eight months after the first injection, but also the NP distribution did not quantitatively change, and thus did not get cleared over time (figures 15(a) and (b)) [68]. The previously described magnetothermal approach by Chen et al was also able to confirm their functionality one month after NP injection [109].
Ultrasound-induced electrical stimulation transduced by piezoelectric materials is a growing technology in neuroengineering, but new and much less well-studied at the nanoscale. A dual targeting system, consisting of 300 nm barium titanate nanoparticles (BTNPs) coated with a copolymer of a phosphatidylethanolamine and PEG coupled with an antibody against the transferrin receptor, enabled crossing of a BBB model, and selective cell uptake in glioblastoma cells (figures 15(c) and (d)) [116]. As the BTNPs have demonstrated neuronal stimulation previously [117], they could be used in the future as intravenously injectable neural stimulators. This surface modification approach also provides insight into how other nanostimulators can be transported across the BBB to provide a less invasive method of NP delivery. In addition, this demonstrates a possibility for specific stimulation of individual cell or tissue types that does not rely on prior genetic modification of the cells being stimulated [109, 111-113, 115, 116].
Similarly, magnetoelectric (ME) neurostimulator materials are a growing field, and also under-studied at the nanoscale. Like BTNPs, ME materials use piezoelectric transduction, but in combination with magnetostrictive materials, in order to achieve magnetic-to-electric signal transduction. ME materials have recently demonstrated by stimulation of the subthalamic region through NPs of size 224-27 nm in mice, which promoted behavioral change (figures 14(f) and (g)) [118].
An in vivo study with live mice as a new approach to solve depressive symptoms employed not a continuous magnetic field, but instead exposure to pulses at a low (10 Hz) frequency. This was conducted with superparamagnetic iron oxide NPs, injected into the left pre-limbic cortex. These NPs were coated with polyglucose sorbitol carboxymethylether and had dimensions of 9 and 30 nm. The combined mechanisms/effects increased the susceptibility of the cell membrane under pulsed stimulation. Using this mechanism, symptoms of depression in mice improved, with the therapeutic effect demonstrated using various biomarkers [119].
Overall, the new but growing interest in less invasive neuromodulation has provided a good starting point for the field of nanoscale neural stimulators. Nonetheless, some technologies are only beginning to be understood, and further research is needed to address unanswered questions. Among these is the selective stimulation of single neurons or tissue types without genetic modification, which has so far achieved only early findings. Research is also needed to determine what, if any, effects there are on other areas of the brain away from the stimulation site, and what the physiological side effects are. In addition, research is being conducted on less invasive methods of NP delivery beyond injection, such as through the BBB.
The long-term effects of nanoscale neurostimulators in the brain are not yet well studied. To date, there have been few studies on how the distribution of NPs changes, or whether the nanomaterials degrade or lose function over time. In addition, it is often unclear what, if any, options are available to remove NPs if necessary. Further research is also needed on whether NPs are taken up by cells over time, and what the consequences of this are.

Concluding remarks
The need for less invasive methods of neural stimulation has driven the relatively new field of NPs for DBS. Injectable NPs have to potential to significantly reduce the adverse effects of invasive devices that typically require wires and batteries. At the same time, their use brings a number of challenges, ranging from synthesis to long-term viability. Presumably, there will be more than one technique for translational applications in the future, adapted to the needs of patients and the goals of therapy.

Status
Neuroinflammation is critically involved in the CNS-related diseases including neurodegenerative disorders like AD, PD and HD, multiple sclerosis (MS) or amyotrophic lateral sclerosis [120]. The process endorses the involvement of the innate immune responses by glial resident cells (i.e. microglia and astrocytes) and the production of pro-inflammatory cytokines including IL1β, IL-6, IL18 and TNF-α, and chemokines such as C-C motif chemokine ligand 1 (CCL1), CCL5 and C-X-C motif chemokine ligand 1 (CXCL1), but also small-molecule messengers, including nitric oxide, and ROS [120]. Neuroinflammation may also involve during disease path the transfer of antigen-experienced peripheral immune cells into the CNS ensuring adaptive immune responses ( figure 16). The specificity of the CNS is the existence of protective walls (i.e. BBB, and the blood-cerebrospinal fluid barrier) that prevent neurons from external insults. However, these barriers likely impede the access of therapeutic agents to the brain. Advances in drug delivery design have focused on nanotechnology-based approaches with the aim not only to improve drug targeting to the brain, but also to increase the bioavailability and pharmacokinetics relative to free drugs. To date, several nanomaterials have been employed including organic (e.g. liposomes, dendrimers, polymeric NPs, micelles, nanogels, EVs, and red blood cell membranes,), inorganic (e.g. metal nanostructures, magnetic NPs, and quantum dots), and carbon-based (e.g. graphene and CNTs) materials [121].
Noticeably, NPs size, type and hydrophobicity will determine their biological fate, toxicity, distribution and targeting ability [121]. Several NPs have been used as carriers for drugs (i.e. curcumin, okadaic acid, quercetin, anthocyanin, and levodopa), increasing their bioavailability in the brain and thereby modulating neuroinflammatory responses and the release of pro-inflammatory cytokines. Moreover, inorganic NPs themselves have shown therapeutic benefit eliciting an anti-inflammatory phenotype of microglia. Besides, recent studies have highlighted the use of chimeric small EVs, conjugated to NPs, biomolecules and drugs, both as biomarkers and for therapeutic purposes in diseases. Among various therapeutic strategies, nanotechnologies targeting neuroinflammation represents a promising approach to improve neurodegenerative diseases conditions. Although, precautions have to be made in data interpretation obtained with NPs in experimental cellular and animal study models before their application in clinical studies.

Current and future challenges
Nanomaterials have been largely employed to alleviate neuroinflammation in CNS diseases ( figure 16). Lipid-based NPs used as carrier of curcumin attenuates neuroinflammatory and reactive gliosis in astroglia cells and organotypic brain slices. Interestingly, curcumin-and sesamol-loaded solid lipid NPs provide anti-inflammatory, neuroprotective and behavioral outcomes in symptomatic AD-like mice [122,123]. Polymer-based NPs have also shown some beneficial effects towards neuroinflammation in cellular study models and in a PD-like mice model that occurs through a downregulation the NF-κB signaling pathway and the inhibition of lipid peroxidation [124]. Particularly, a novel synthetic NP, poly (lactide-co-glycolide)-block-PEG provided efficient brain targeting and beneficial learning and memory abilities in an AD mice model when conjugated with B6 peptide and loaded with curcumin [125]. Synthetic NPs have also been applied in epilepsy and cerebral ischemic reperfusion injury (IR) injury mouse models showing a reduction of pro-inflammatory markers and gliosis [126,127]. In addition, some studies have reported the potential of NP dendrimers-mediated drug delivery to alleviate neuroinflammation in vitro and in cerebral palsy and in AD in vivo study models [128,129]. However, the benefit-cost of dendrimers is high compared to linear polymers and their toxicity has been raised in some studies. Several evidences reported that inorganic NPs alone or modified drugs provides anti-inflammatory effects in vitro and in vivo models of neurodegenerative diseases. Ex vivo experiments have shown that L-DOPA conjugated inorganic NPs increases their transport across brain endothelial monolayers and are readily taken up by brain macrophages without inflammatory effects. At most interest, antocyanine carried PEG coated inorganic NPs reduced the levels of amyloid beta (Aβ) and BACE-1 in Aβ 1-42 -injected mice brains and treated microglia cells in vitro. Moreover, they stimulated p-GSK-3β Ser9 /p-CDK5 signaling, reduced the microgliosis and astroglyosis and the level of pro-inflammatory markers including p-NF-kB, iNOS, TNFα, and IL-1β in vivo [130]. These observations were further supported by another study demonstrating that inorganic NPs may increase the level of anti-inflammatory IL-4 in a drug-induced AD mice model [131]. Besides neurodegenerative diseases, the combination of organic NPs and n-acetylcysteine also decreased pro-inflammatory cytokines production in sepsis-induced brain dysfunction in rats [132]. Recently developed NPs based on graphene quantum dots are considered a promising therapeutic approach to alleviate neuroinflammation as demonstrated in experimental autoimmune encephalomyelitis via the activation of MAPK/Akt signaling, and in PD and AD study models through the disaggregation of α-synuclein and Aβ 1-42 peptides respectively [121].

Advances in science and technology to meet challenges
Besides being involved in several neurodegenerative diseases, neuroinflammation is common to several psychiatric disorders including major depression, anxiety, post-traumatic stress, and bipolar disorders. Thus, advances in drugs targeting brain inflammation remain a scientific challenge that will benefit to a large set of CNS diseases. NPs are increasingly recognized as a valuable therapeutic approach to alleviate deleterious chronic neuroinflammation in animal CNS disease study models. However, many NPs have been reported to exhibit neurotoxicity and pro-inflammatory responses. Therefore, it is imperative to develop better nanotechnologies, and to have a better mechanistic understanding of nanomaterials. Conjugating NPs with specific antigens/peptides may also be envisioned to target a specific set of CNS inflammatory cells (i.e. microglia or astrocytes). Similarly, loading NPs with existing drugs/small molecules is another avenue to pursuing and gaining better pharmacokinetic properties (i.e. a long half-life, better target specificity, high lipophilicity) and BBB penetrability. Recent studies reported the potential use of EVs for treating brain cancer. In fact, EVs have the advantage being biocompatible and may be considered immunologically inactive and able to cross the BBB. However, it is still somewhat premature to predict the therapeutic applications by modified-EVs as drug delivery systems for CNS diseases. Recent studies have highlighted that peripheral myeloid cells may play a fundamental role in CNS diseases development. Data obtained in animal models have demonstrated the beneficial effects in targeting peripheral myeloid cells in AD, PD and MS. This includes blockade of their migration to the CNS, modulation of their biological functions, immunological activity and cytokine production. Thus, NPs represent an exciting new tool for regulating myeloid cell functions, likely impacting pro-inflammatory response in the CNS.

Concluding remarks
Neurodegenerative disorders still pose several therapeutic challenges. The development of nanotechnologies has improved the targeting of therapeutic agents to brain forcing the brain barriers, enhancing the bioavailability and pharmacokinetics of drugs. Moreover, nanotechnologies, able to recognize specific brain receptors or transporters, favor the selective release of drugs to the target site, reducing side effects and systemic exposure. NPs guided delivery of therapeutic agents has demonstrated benefits alleviate chronic neuroinflammation in preclinical study models. Looking forward, the regulatory process for nanotechnology should consider both the risk and the reward in a balanced manner, to enable translation of nanomedicine from bench to bedside in the CNS disorders application. The short-term objective will be to prove the efficiency of nanotechnologies in pre-clinical experimental models with pathophysiological similarities closer to human inflammatory disease.

Status
The difficulty to safely and precisely release drugs in the diseased CNS limits the outcome of the current treatments. CNS immunotherapy aims at modulating the activity of the immune system to recognize and treat diseased tissues and neuropathological conditions. In this scenario, smart nanomaterials can be designed to target specific cell types, cross biological barriers, and modulate the pathologic microenvironment. The passive immune nanomedicines expose or release traditional immunotherapy agents (e.g. antigens and cytokines) while the active ones (e.g. iron oxide NPs and aAPC nanoconjugates) are designed to directly boost the host immunoresponse [133]. Nano-enabled CNS immunotherapy primarily finds application in counteracting brain cancer but also in treating neurodegenerative conditions, such as AD, and spinal cord injuries [134].
In brain cancer, immune nanomedicines are formulated to teach the innate immune system to target and attack cancer cells. The approach involves the activation of the cytotoxic T cells by the dendritic cells (DC). Scarcely immunogenic brain tumors like GBM show multiple protection systems to resist the immune system attacks (e.g. the immune checkpoint-ICP-ligands). Different therapeutic nanomedicines are currently focused to overcome this challenge [135].
Nano-enabled immunotherapy has been also recently applied to trigger an antibody response against the amyloid-beta (Aβ; figure 17) [136]. In the brain of AD patients, abnormal deposition of this peptide and its accumulation in the amyloid plaques are observed. Researchers proposed innovative nanomaterials able to recruit microglia and promote Aβ phagocytosis in APPswe/PS1dE9 mice, therefore improving neuronal functionality, decreasing inflammation, and rescuing memory. This pioneering approach of 'Aβ cleaning' , despite is still at an early stage of development, represents an innovative strategy for AD therapy. In the future, alternative immune nanomedicines can be developed to reduce the neural toxicity associated with the accumulation of the tau protein in AD and α-synuclein in PD.

Current and future challenges
In neuro-oncology, the treatment of GBM with immunotherapy remains a huge challenge due to its elevated grade of heterogeneity and scarce immunogenicity.
GBM is defined as an immunologically 'cold' tumor microenvironment (TME) characterized by a reduced T cell activity and infiltration capability. Likely other 'cold' tumors, the outcomes of the immunotherapy against GBM with antibodies targeting ICPs are poor in most cases [137]. This represents a relevant therapeutic challenge. Nanomaterials can be specifically designed to address this obstacle, finely modulating the TME and promoting the expansion/engagement of T cells [138]. Furthermore, the multifunctionality of smart nanomedicines allows to efficiently cross the blood-brain tumor barrier (BBTB), promote the release of immunomodulators or chemotherapy drugs, remotely stimulate the tumor tissue, and absorb tumor antigens upon tumor damage for priming immune response.
Another fundamental challenge is associated with the elevated interindividual heterogeneity in immunoresponse. The high variability in therapeutic outcome derives not only from the specific molecular characteristics of the tumor (e.g. the levels of the tumor mutational burden and the pattern of the neoantigen expression) but also from extrinsic interindividual differences. In this regard, an important role is mediated by the microbiota diversity, hormonal-related sex-specific peculiarities, and the aging of the immune system [139]. The use of NPs for immunotherapy increases the level of complexity of the therapy approach, amplifying the variability of the treatment responses. As an example, female patients show a faster plasma clearance rate when treated with doxorubicin-loaded liposomes compared to males. Sex-related differences in NP-cell interaction may be attributed to hormonal levels. In this regard, recent studies are showing as hormonal concentrations affect the NP internalization in cancer cells [140].
Finally, the adverse effects of the adopted nanomedicines on the immune cells should be carefully evaluated. The active or passive modulation of the immune system can indeed change the cytotoxicity response to a specific nanomaterial [139]. Also, the nano-enabled immunotherapy can induce the release of inflammatory cytokines and, as a consequence, induce side effects to distal organs [141].

Advances in science and technology to meet challenges
Advances in science revealing in detail the interactions between nanoagents and the immune cells will be required. Future studies will have to take into account and highlight the differences in immunotherapeutic response due to sex, age, hormone levels, and specific molecular characteristics of the disease (e.g. brain cancer and AD). The experimental design will have to adequately consider the abovementioned aspects [139]. Furthermore, the neurotoxicity of immune nanomedicines will have to be thoroughly evaluated, preferring in vitro the adoption of highly biomimetic systems, such as the 3D brain-on-chip models incorporating immune cells. Basic research evaluating the effects of nanomaterial morphology, size, and composition on immune cell behavior still requires elucidations.
Improvements in the targeting specificity of the nanomaterials toward selected types of immune cells, cytokine receptor-expressing populations, and cancer cells are highly demanded to improve the therapy outcomes and reduce side effects. In addition to the multi-ligand functionalization strategies, a new trend is the nanomaterial surface modification with patient-derived cell membranes [141]. In this regard, NPs camouflaged with cancer cell membranes display a dual function to selectively target cancer cells owing to homotypic membrane-membrane interactions [142] but also to expose tumor antigens and elicit the systemic anticancer immune response. First experimental proofs are revealing the positive role of this coating strategy in promoting the BBB crossing. Plasma membranes from different cell types, such as those from leukocytes, red blood cells, platelets, DCs, and natural killer cells, have been exploited as a coating strategy for imparting specific functionalities to the NPs. Improved performances and multifunctionality can be provided by exploiting hybrid membranes (i.e. obtained by mixing membranes from different cell types) and genetically engineered cell membranes (figure 18) [143]. The long-term safety, biodistribution, and excretion of these innovative nanomedicines should be carefully examined since the literature data in this regard are very limited.
Finally, to complement the function of immune nanomedicines in 'cold' immunosuppressive TME, the development of new strategies promoting tumor recognition and infiltration by immune cells will be required.

Concluding remarks
Immunotherapy, despite being primarily applied for cancer treatment, shows high potential also for AD, PD, and spinal cord injury. Nanomaterials represent a potential solution for most of the challenges of brain immunotherapy since they can be specifically designed to cross BBB/BBTB, target the diseased cells/ microenvironment, and trigger a specific immune cell response. On the other hand, the use of nanomedicines increases the complexity of the therapeutic approach, amplifying the interindividual variability of the therapeutic response and limiting the clinical applicability in the short-term period. More sophisticated preclinical investigations on nanomedicine safety and future advancement in BBB-crossing and targeting technologies will pave the way for the clinical exploitation of different immune nanomedicines. Although clinical studies on nano-enabled immunotherapy have not been applied to the treatment of CNS diseases yet, the approach is being tested in clinical trials for the treatment of other non-CNS tumors by using different nanoplatforms. Examples of such therapeutic nanoplatforms are the PLGA NPs loaded with