Amphiphilic Nanocarrier Systems for Curcumin Delivery in Neurodegenerative Disorders

Neurodegenerative diseases have become a major challenge for public health because of their incurable status. Soft nanotechnology provides potential for slowing down the progression of neurodegenerative disorders by using innovative formulations of neuroprotective antioxidants like curcumin, resveratrol, vitamin E, rosmarinic acid, 7,8-dihydroxyflavone, coenzyme Q10, and fish oil. Curcumin is a natural, liposoluble compound, which is of considerable interest for nanomedicine development in combination therapies. The neuroprotective effects of combination treatments can involve restorative mechanisms against oxidative stress, mitochondrial dysfunction, inflammation, and protein aggregation. Despite the anti-amyloid and anti-tau potential of curcumin and its neurogenesis-stimulating properties, the utilization of this antioxidant as a drug in neuroregenerative therapies has huge limitations due to its poor water solubility, physico-chemical instability, and low oral bioavailability. We highlight the developments of soft lipid- and polymer-based delivery carriers of curcumin, which help improve the drug solubility and stability. We specifically focus on amphiphilic liquid crystalline nanocarriers (cubosome, hexosome, spongosome, and liposome particles) for the encapsulation of curcumin with the purpose of halting the progressive neuronal loss in Alzheimer’s, Parkinson’s, and Huntington’s diseases and amyotrophic lateral sclerosis (ALS).

In recent years, several studies have shown that curcumin is a safe natural compound which may prevent the deleterious effects of risk factors causing brain damage as well as slowing down the progressive neuronal loss via different pathways . However, clinical studies performed with AD patients with various degrees of progression have reported poor results on the AD symptoms following curcumin treatment [51][52][53][54]. This did not allow firm conclusions about the therapeutic or neuroprotective potential of curcumin to be drawn. The obstacles for curcumin utilization as a drug originate from its limited water solubility, poor physicochemical stability, high-grade metabolism, and low plasma concentrations [36,[53][54][55]. The development of nanoparticulate delivery systems for curcumin has attracted scientific interest in order to improve its bioavailability and stability as a drug compound [56][57][58][59][60][61][62][63][64][65]. Curcumin administration to neurodegenerative disease models by nanoparticles has been realized using liposomes, solid lipid nanoparticles, and polymeric particles. Delivery by other carriers such as amphiphilic proteins, e.g., casein, is also possible, but has not been examined as a means of transporting curcumin across the BBB towards neuro-regeneration.
In this review, we briefly summarize the in vitro and the in vivo evaluations of curcumin, which are linked to multiple risk factors and the multi-target mechanisms of neurodegenerative diseases, and discuss the reported clinical investigations of varying efficacy in humans. Then, we highlight the variety of amphiphilic curcumin-loaded nanocarriers including liposomes, liquid crystalline nanoparticles (cubosomes, hexosomes, and spongosomes), solid lipid nanoparticles, micelles, and polymeric nanoparticles as potential nanomedicine formulations in regeneration therapies of the major neurological disorders.

Risk Factors for Neurodegenerative Disorders
Alzheimer's disease (AD) is the most common cause of dementia. It currently affects about 10% of the world's population over 60-65 years of age, and about 50% over 85 years of age. More than 30 million people may expect to be affected by AD during the next 20 years due to the increasing lifespan of the world population [1,2]. Major pathological features of AD include the accumulation of extracellular amyloid plaques and fibrils, intracellular neurofibrillary tangles, and disruption of the cholinergic transmission, including reduced acetylcholine levels in the basal forebrain ( Table 1). The most common symptom is the short-term memory loss, i.e., difficulty in remembering recent events [2][3][4][5]. Other symptoms include disorientation, mood, language, and behavioral issues, and loss of motivation, depending on the progression of the disease. The treatments of AD have employed acetylcholinesterase inhibitors (tacrine, rivastigmine, galantamine, and donepezil) to overcome the decrease of the ACh levels as a result of the death of cholinergic neurons. The NMDA receptor antagonist (memantine) acts by inhibiting the overstimulation by glutamate, which can cause cell death. Atypical antipsychotics have modest efficacy in reducing the aggression and psychosis of AD patients. These medications provide little benefit, and provoke various adverse effects [6,7].
The second most common disorder, Parkinson's disease (PD), affects more than 1% of the population over 60 years of age and 5% over 85. PD is characterized by progressive impairments in locomotive ability such as tremor, rigidity, and bradykinesia. These symptoms are attributed to the loss of dopaminergic neurons in the substantia nigra and the formation of Lewy bodies in the brain [8,9]. Treatments are symptomatic and aim at boosting the depleted levels of dopamine (Table 1). The most used drug is levodopa. Dopamine agonists are used when the treatment by levodopa becomes less efficient. The inhibitors of MAO-B and COMT (safinamide, selegiline, rasagiline, and tolcapone) are used to inhibit the activity of the enzymes which degrade dopamine. These medications become less effective as the neurons are continuously lost during disease progression. At the same time, they produce complications marked by the involuntary movements of the patients [8,9].
Huntington disease (HD) is a rare disease which affects about 1/10,000 people (usually between 30 to 50 years of age) in the United States and 1/18,000 people in Europe. It is a poly-glutamine (PolyQ) autosomal genetic disorder characterized by impairments of cognitive, psychiatric, and motor functions [10]. The hallmark of the HD pathology is the abnormal accumulation of misfolded mutated huntingtin protein (mHTT) as intracellular aggregates. They cause selective neuronal loss, primarily in the cortex and in the medium spiny neurons of striatum. Symptoms develop from a general lack of coordination to apparent uncoordinated, jerky body movements [11]. The physical abilities of the patients gradually worsen until coordinated movement becomes difficult. There is no effective cure available to HD ( Table 1). The only approved medication, tetrabenazine, and other tested drugs (neuroleptics and antipsychotics) help to reduce chorea and psychiatric symptoms.
Amyotrophic lateral sclerosis (ALS) is a severe debilitating disease caused by motors neurons degeneration in the brain and the spinal cord. It is generally characterized by progressive paralysis starting at the limbs and ultimately leading to death caused by respiratory failure within 3 to 5 years after the onset of the symptoms. There is no cure for ALS (Table 1). The approved medication, riluzole, may extend life by just a few months [12,13].
The pathological characteristics, genetic factors, clinical symptoms, and actual medications of these diseases are summarized in Table 1. It should be emphasized that the existing therapeutic approaches do not exert disease-modifying effects on the neurodegeneration. The associated economic and societal challenges lead to a growing public health burden. Although the etiology and the pathological mechanism of the neurodegenerative diseases are not completely understood, it has been established that the progressive loss of neurons results from the combination of multiple factors ( Figure 1). First, genetic factors are involved in the appearance of misfolded amyloid-Aβ protein and other misfolded mutant forms like hyperphosphorylated Tau (p-Tau) and Huntingtin proteins [3][4][5]10,14]. All these mutated proteins aggregate and form deposits. The resulting aggregates can be toxic, and can affect the intracellular organelles such as mitochondria [14,21,25,27]. The disruption of the mitochondrial membrane causes neuronal cell death [25,27]. Second, neurotrophic factors deficiency has been reported in the severe neurodegenerative disorders [11,13,[29][30][31][32][33]. Neurotrophins regulate the neuronal survival, differentiation, growth, and regeneration, as well as the synaptic plasticity. Studies have shown that the levels of brain derived neurotrophic factor (BDNF) and its tropomyosin kinase B (TrkB) receptor are decreased in the hippocampus and the cerebral cortex at the beginning of the Alzheimer's disease [11]. In addition, the administration of BDNF mimetics into transgenic mouse models of AD has enhanced learning and memory capacities [31].
Fourth, neuro-inflammation is a crucial factor that favors neurodegenerative disease development. Several inflammatory markers (such as chemokines, cytokines, or proteins in the acute phase) are upregulated and cause inflammation. In fact, elevated levels of the inflammatory markers have been found during the progression of the neurodegenerative diseases [27,28]. Second, neurotrophic factors deficiency has been reported in the severe neurodegenerative disorders [11,13,[29][30][31][32][33]. Neurotrophins regulate the neuronal survival, differentiation, growth, and regeneration, as well as the synaptic plasticity. Studies have shown that the levels of brain derived neurotrophic factor (BDNF) and its tropomyosin kinase B (TrkB) receptor are decreased in the hippocampus and the cerebral cortex at the beginning of the Alzheimer's disease [11]. In addition, the administration of BDNF mimetics into transgenic mouse models of AD has enhanced learning and memory capacities [31].
Fourth, neuro-inflammation is a crucial factor that favors neurodegenerative disease development. Several inflammatory markers (such as chemokines, cytokines, or proteins in the acute phase) are upregulated and cause inflammation. In fact, elevated levels of the inflammatory markers have been found during the progression of the neurodegenerative diseases [27,28].

Curcumin Potential for Neuroprotection against Neurodegenerative Diseases
Curcumin is a hydrophobic polyphenolic substance ( Figure 2) produced in the root of the plant Curcuma Longa L. This antioxidant compound is extensively marketed worldwide as a nutraceutical in various preparations, because it has a very safe nutraceutical profile with low side effects. Curcumin has been reported to be well tolerated at doses up to 8 g per day over short periods in humans [32]. Research on the pharmacological activities of curcumin has attracted strong attention in relation to its multiple actions of therapeutic interest, e.g., the anti-inflammatory, antioxidant, antiviral, antibacterial, antifungal, and antitumor activities. These activities appear to be dose-dependent [33].

Curcumin Potential for Neuroprotection against Neurodegenerative Diseases
Curcumin is a hydrophobic polyphenolic substance ( Figure 2) produced in the root of the plant Curcuma Longa L. This antioxidant compound is extensively marketed worldwide as a nutraceutical in various preparations, because it has a very safe nutraceutical profile with low side effects. Curcumin has been reported to be well tolerated at doses up to 8 g per day over short periods in humans [32]. Research on the pharmacological activities of curcumin has attracted strong attention in relation to its multiple actions of therapeutic interest, e.g., the anti-inflammatory, antioxidant, antiviral, antibacterial, antifungal, and antitumor activities. These activities appear to be dosedependent [33].

In Vitro and In Vivo Studies of Curcumin Properties in Neurodegenerative Disease Models
The neuroprotective potential of curcumin ( Figure 2) and its antioxidant, anti-inflammatory, and amyloid Aβ binding properties have been highlighted in in vitro and in vivo investigations of different neurodegenerative disease models [33][34][35][36][37][38][39][40][41][42]. Curcumin has been found to increase the levels of glutathione (GSH) and malondialdehyde (MDA), as well as the antioxidant enzyme [superoxide dismutase (SOD), glutathione peroxidase (GPx), glutathione reductase (GR), and catalase (CAT)] activities in the rat brain, thus preventing the stress-induced oxidative damage of brain [37,39]. The anti-inflammatory properties of curcumin have been characterized by the inhibition of the inflammatory chemokines, by increased levels of the anti-inflammatory cytokines, and by enhanced expression levels of induced nitric oxide synthase (iNOS) and the transcription factor NF-Kb [39]. Curcumin has been shown to prevent the fibrillation of α-synuclein at the earliest stage of the aggregation process, as well as the fibrillation of the globular protein hen egg-white lysozyme (HEWL) [40]. Both proteins are known to form amyloid-like fibrils. These results have suggested that curcumin might be a potential therapeutic agent for preventing protein aggregation in Alzheimer's, and Parkinson's diseases [40]. Recent in vitro and in vivo investigations of curcumin's activities in neurodegenerative disease models [41][42][43][44][45][46][47][48][49][50] are summarized in Table 2. Table 2. Recently reported curcumin (CU) activities in in vitro and in vivo models of neurodegenerative diseases [41][42][43][44][45][46][47][48][49][50].

In Vitro and In Vivo Studies of Curcumin Properties in Neurodegenerative Disease Models
The neuroprotective potential of curcumin ( Figure 2) and its antioxidant, anti-inflammatory, and amyloid Aβ binding properties have been highlighted in in vitro and in vivo investigations of different neurodegenerative disease models [33][34][35][36][37][38][39][40][41][42]. Curcumin has been found to increase the levels of glutathione (GSH) and malondialdehyde (MDA), as well as the antioxidant enzyme [superoxide dismutase (SOD), glutathione peroxidase (GPx), glutathione reductase (GR), and catalase (CAT)] activities in the rat brain, thus preventing the stress-induced oxidative damage of brain [37,39]. The anti-inflammatory properties of curcumin have been characterized by the inhibition of the inflammatory chemokines, by increased levels of the anti-inflammatory cytokines, and by enhanced expression levels of induced nitric oxide synthase (iNOS) and the transcription factor NF-Kb [39]. Curcumin has been shown to prevent the fibrillation of α-synuclein at the earliest stage of the aggregation process, as well as the fibrillation of the globular protein hen egg-white lysozyme (HEWL) [40]. Both proteins are known to form amyloid-like fibrils. These results have suggested that curcumin might be a potential therapeutic agent for preventing protein aggregation in Alzheimer's, and Parkinson's diseases [40]. Recent in vitro and in vivo investigations of curcumin's activities in neurodegenerative disease models [41][42][43][44][45][46][47][48][49][50] are summarized in Table 2.  [47].

Diazepam-induced cognitive impairment
In vivo: diazepam-treated rats, oral supplementation Downregulation of the extracellular signal-regulated kinase (ERK 1/2)/nuclear transcription factor-(NF-)κB/pNF-κB pathway in the hippocampus and the iNOS 12 expression in the hippocampus and frontal cortex Improvement of the cognitive performance, Decrease of blood and brain oxidative stress levels [48].

Clinical Trials and Curcumin Limits
A serious obstacle to the pharmaceutical application of curcumin has arisen from its limited water-solubility and low bioavailability. In addition, this compound is chemically instable, which may cause a loss of biological activities. The failure of free curcumin in clinical trials is likely due to its limited bioavailability. For instance, curcumin has been delivered in doses between 1 and 4 g/per day as capsules or as powder mixed with food in trials for treatment of Alzheimer's disease patients. The performed 6-month treatment study found no differences in the Aβ-amyloid levels between the treatment groups or in the Mini Mental State Examination (MMSE) scores [51]. Similarly, oral curcumin in a 24-week, randomized, double blind, and placebo-controlled study for AD treatment has shown no detectable differences in the measured biomarkers from the different treatment groups [52]. A clinical study with three single cases of patients receiving curcumin (100 mg/day) reported that only one patient increased his MMSE score from 12/30 to 17/30 after 12 weeks of treatment (improved calculation, concentration, transcription of the figure, and spontaneous writing). Two of the patients were on donepezil treatment before starting the curcumin trial [53]. Based on all performed trials with AD patients, it was, however, difficult to conclude if curcumin has positive effects on the AD symptoms [54].
In fact, the major fraction (35-89%) of orally-administered curcumin can be lost due to its low bioavailability. The intestinal mucosa and mucus form a physical barrier to curcumin adsorption. The drug cannot reach the circulation in a bioactive form as it undergoes reduction and metabolism/conjugation in the liver. Reductases enzymatically reduce curcumin to dihydrocurcumin, tetrahydrocurcumin, and hexahydrocurcumin. Furthermore, curcumin may be conjugated to sulfates and glucuronides [55][56][57]. Thus, most of the circulating curcumin is in a conjugated form.
The necessity of the development of a delivery system for the protection of curcumin from rapid metabolism and for the improvement of its bioavailability has become evident [58]. A randomized, double-blind, placebo-controlled clinical trial examined the acute administration (effects 1 h and 3 h after a single dose application), chronic (4 weeks) administration, and acute-on-chronic (1 h or 3 h after a single dose followed by a chronic treatment) effects of solid-lipid-nanoparticle (SLNP) loaded by curcumin. The results of the SLNP formulation of curcumin (400 mg in capsules Longvida ® ) on cognitive function, mood, and blood biomarkers were obtained with 60 healthy adults (aged 60-85). SLNP-loaded curcumin significantly improved the performance in sustained attention and working memory tasks one hour after its administration (as compared to placebo). Working memory and mood (general fatigue and change in the calmness state, contentedness, and fatigue induced by psychological stress) were essentially improved following chronic treatment. A significant acute-on-chronic treatment effect on alertness and contentedness was also observed [59].

Nanocarrier-Mediated Curcumin Delivery
Nanotechnology for nanomedicine development employs functional materials with appropriate nanoscale organization that can interact with biological systems and induce desired physiological responses while minimizing undesirable side effects [60]. Nanotechnology-based delivery systems can influence drug capacity to cross the biological barriers (e.g., the BBB) and reach the targeted brain regions [58][59][60][61]. Therefore, nanocarriers are promising for the development of personalized medicines for the treatment of neurological disorders [62][63][64][65][66][67].
Lipid-based nanoparticles, including solid lipid nanoparticles (SLNPs), nanostructured lipid carriers (NLC), liposomes and liquid crystalline nanocarriers (LCN), as well as polymer-based nanoparticles (Figure 3), have been developed to overcome the poor solubility, stability, and bioavailability of curcumin, and to promote its utilization as a drug in disease treatments . Lipid-based nanoparticles have the advantage of being the least toxic carriers for in vivo applications. The lipids used to prepare biocompatible and biodegradable nanoparticles are usually naturally-occurring molecules with low acute and chronic toxicity. In the case of polymeric nanoparticles, the in vivo degradation of the polymer matrices might cause toxic effects [94]. The biocompatibility and the physico-chemical diversity of lipids and their capacity to enhance the oral bioavailability of drugs have made this kind of nanocarriers very attractive systems for drug delivery. As a matter of fact, lipid-based formulations may positively influence drug absorption in several ways, e.g., by influencing the solubilization properties, preventing the drug precipitation upon intestinal dilution, increasing the membrane permeability, inhibiting the efflux transporters, reducing the CYP enzymes, or enhancing the lymphatic transport [94,122].
Among the lipid-based nanoparticles, SLNPs have been intensively developed because they combine the advantages of different carrier systems like liposomes and polymeric particles. Similarly to liposomes, SLNPs are composed of physiologically-biocompatible excipients (lipids and fatty acids). In the same way to polymeric NPs, their solid matrix core can efficiently protect the loaded active pharmaceutical ingredient against chemical degradation under the harsh conditions of biological milieux. Therefore, SLNPs provide controlled release profiles of the encapsulated drugs [95].
In addition to the above advantages, liposomes can encapsulate and transport both lipophilic and hydrophilic drugs. They have a high degree of similarity to cell membranes in terms of lipid composition and organization, which facilitates the bioavailability of the pharmaceutical compounds [102]. Liquid crystalline nanocarriers (LCN) such as cubosomes and hexosomes ( Figure 3) involve multiple compartments for encapsulation of either lipophilic or hydrophilic drugs. They display structural advantages which enable high encapsulation efficacy for molecules of various sizes and hydrophilicity [113,116]. LCNs are formed by self-assembly of lyotropic lipids such as unsaturated monoglycerides, phospholipids, glycolipids, and other amphiphilic molecules. For example, monoolein, which is a nontoxic, biodegradable, and biocompatible lipid, is classified as a GRAS (Generally Recognized As Safe), and ω-3 polyunsaturated fatty acids (n-3 PUFA) have been shown to be highly beneficial in various disease models of neurodegeneration [123,124].
In the following section, we summarize recently reported works on curcumin delivery to in vitro and in vivo models of neurodegenerative diseases. Lipid-based nanoparticles have the advantage of being the least toxic carriers for in vivo applications. The lipids used to prepare biocompatible and biodegradable nanoparticles are usually naturally-occurring molecules with low acute and chronic toxicity. In the case of polymeric nanoparticles, the in vivo degradation of the polymer matrices might cause toxic effects [94]. The biocompatibility and the physico-chemical diversity of lipids and their capacity to enhance the oral bioavailability of drugs have made this kind of nanocarriers very attractive systems for drug delivery. As a matter of fact, lipid-based formulations may positively influence drug absorption in several ways, e.g., by influencing the solubilization properties, preventing the drug precipitation upon intestinal dilution, increasing the membrane permeability, inhibiting the efflux transporters, reducing the CYP enzymes, or enhancing the lymphatic transport [94,122].
Among the lipid-based nanoparticles, SLNPs have been intensively developed because they combine the advantages of different carrier systems like liposomes and polymeric particles. Similarly to liposomes, SLNPs are composed of physiologically-biocompatible excipients (lipids and fatty acids). In the same way to polymeric NPs, their solid matrix core can efficiently protect the loaded active pharmaceutical ingredient against chemical degradation under the harsh conditions of biological milieux. Therefore, SLNPs provide controlled release profiles of the encapsulated drugs [95].
In addition to the above advantages, liposomes can encapsulate and transport both lipophilic and hydrophilic drugs. They have a high degree of similarity to cell membranes in terms of lipid composition and organization, which facilitates the bioavailability of the pharmaceutical compounds [102]. Liquid crystalline nanocarriers (LCN) such as cubosomes and hexosomes ( Figure 3) involve multiple compartments for encapsulation of either lipophilic or hydrophilic drugs. They display structural advantages which enable high encapsulation efficacy for molecules of various sizes and hydrophilicity [113,116]. LCNs are formed by self-assembly of lyotropic lipids such as unsaturated monoglycerides, phospholipids, glycolipids, and other amphiphilic molecules. For example, monoolein, which is a nontoxic, biodegradable, and biocompatible lipid, is classified as a GRAS (Generally Recognized As Safe), and ω-3 polyunsaturated fatty acids (n-3 PUFA) have been shown to be highly beneficial in various disease models of neurodegeneration [123,124].
In the following section, we summarize recently reported works on curcumin delivery to in vitro and in vivo models of neurodegenerative diseases.

Curcumin Delivery by Polymeric Nanoparticles
Polymeric nanoparticles of a biocompatible and biodegradable nature are of essential interest as drug delivery nanocarriers. The release of the encapsulated drug can be modulated by altering the polymer composition and amphiphilicity. Poly(lactic-co-glycolic acid) (PLGA) is one of the most commonly-used biodegradable synthetic polymers. It is a FDA-(US) and EMA-approved platform for the delivery of drugs to humans. PLGA-derived nanoparticles have been successfully used for the encapsulation of different hydrophobic compounds (such as curcumin) by nanoprecipitation or by single emulsion techniques [71]. Hydrophilic molecules can be encapsulated by means of double emulsions or by two-step nanoprecipitation methods. On the other site, polymeric micelles have been studied towards site-specific drug delivery [68][69][70].
Micelles are formed by amphiphilic macromolecules, which self-assemble into nano-sized (10-100 nm in diameter) core/shell structures in excess aqueous media (Figure 3, top). The core-shell organization facilitates the incorporation of curcumin inside the hydrophobic core, while the water solubility of the nanocarriers is ensured by their hydrophilic corona [68]. Amphiphilic co-polymers self-assemble into micelles in aqueous solutions due to the hydrophobic interactions among their water-insoluble segments. Curcumin-loaded polymeric micelles have received attention due to various features (Table 3), like (i) the enhanced solubility of the drug; (ii) the sustained CU release profile; and (iii) the small size of the PEG-decorated carriers (<200 nm), which stabilizes them in biological fluids [64][65][66]. Micelles formed by the PLGA-PEG-PLGA synthetic copolymer have shown potential in modifying the pharmacokinetics and tissue distribution of curcumin. Evaluation of pharmacokinetics and biodistribution has demonstrated a prolonged half-life of the CU-micelles and a more efficient drug delivery to brain areas as compared to the carrier-free CU administration [71].  Higher curcumin concentration in plasma, 6 times higher area under the curve and mean residence time in brain than ordinary CU, Improved memory function [83].

AD Nanoemulsion
In vitro: SK-N-SH cell line, Sheep nasal mucosa Safe for intranasal delivery for brain targeting, Higher flux and permeation across sheep nasal mucosa [84].
Mitochon drial dysfunction in brain aging

Micelles
In vitro: PC12 cells In vivo: NMRI mice; Ex vivo: isolated mouse brain mitochondria Improved bioavailability of native curcumin around 10-to 40-fold in plasma and brain of mice, Prevention of mitochondrial swelling in isolated mouse brain mitochondria, Protection of PC12 cells from nitrosative stress as compared to free CU [85].

Solid Lipid Nanoparticles (SLNPs) and Nanostructured Lipid Carriers (NLCs)
Solid lipid nanoparticles (SLNPs) are submicron colloidal lipid carriers (from 50 nm to 1000 nm in diameter) which maintain a solid, spherical shape at room temperature. They possess a solid lipid core matrix stabilized by emulsifiers that can solubilize lipophilic molecules. The CU-SLNPs are usually small, ranging from 100 to 300 nm in diameter. The total drug content can reach up to 92% when the SLNPs are manufactured using the micro-emulsification technique [87]. In an experimental rat model of cerebral ischemic reperfusion injury, animals fed with CU-loaded SLNPs have had a 90% improvement in their cognitive function along with a 52% inhibition of the acetyl cholinesterase activity [88]. The investigated formula has been shown to increase the levels of superoxide dismutase (SOD), catalase (CAT), glutathione (GSH), and the activities of mitochondrial enzymes, while decreasing the lipid peroxidation and the peroxynitrite levels. Furthermore, this formulation showed a 16.4 to 30-fold improvement in the bioavailability of CU in the brain upon oral and intravenous (IV) administrations, respectively [88]. The product Longvida ® (Verdure Sciences Inc.) is a SLNP-formulation of curcumin which can yield from 0.1 to 0.2 µM plasma levels of CU with associated 1-2 µM brain levels of free CU in animals [89][90][91]. This formula was later optimized as "lipidated Cur", which can yield more than 5 µM CU in mouse brain [93]. Other formulations of CU-loaded SLNPs, tested in Alzheimer's disease models, are outlined in Table 4. Table 4.
Curcumin-loaded lipid nanoparticles studied in in vitro and in vivo models of neurodegenerative diseases.  [88].
AD Lipoprotein (LDL)-mimic nanostructured lipid carrier (NLC) modified with lactoferrin (Lf) and loaded with CU In vivo: Rat, oral Cellular uptake mediated by the Lf receptor, Permeability through the BBB and preferentially accumulation in the brain, Efficacy in controlling the damage associated with AD [96].

AD
Liposomes functionalized with TAT-peptide In vitro Permeability across the BBB enhanced [98]. Increased entrapment efficiency of NGF and CU, of NGF release and cell viability, Decreased release of CU, Permeability of NGF and CU across the blood-brain barrier [102].

AD Liposomes
In vivo: Mice, stereotaxic injection in the right hippocampus and neocortex Decrease in Aβ secretion and toxicity [97]. Nanostructured lipid carriers (NLC) are referred to as the "second generation" of SLNPs. NLCs are composed of mixtures of sterically different amphiphilic molecules. Often, mixtures of liquid-phase lipids and solid-phase lipids yield matrices with imperfections, which may incorporate increased quantities of drug molecules as compared to the SLNPs. Despite of the presence of the liquid-phase lipid, the NLC matrix appears to be in a solid state at room and body temperatures. The solid state is controlled by the fraction of the included liquid-phase lipid [94]. Sadegh-Malvajerd et al. have reported an enhanced entrapment efficiency of curcumin in NLCs (94% ± 0.74) as compared to SLNPs (82% ± 0.49). The pharmacokinetic studies, performed after intravenous (IV) administration of 4 mg/kg dose of curcumin in rats, have indicated that the amount of curcumin available in the brain was significantly higher for curcumin-loaded NLCs (AUC 0-t = 505.76 ng/g h) as compared to free curcumin (AUC 0-t = 0.00 ng/g h) and curcumin-loaded SLNs (AUC 0-t = 116.31 ng/g h) (P < 0.005) [95]. The outcomes of other recent investigations of CU-loaded NLCs in models of neurodegenerative diseases are summarized in Table 4.

Liposomes
Liposomes are lipid bilayer-based, self-assembled, closed colloidal structures, typically 25 nm to 5 µm in diameter [97]. They usually have a spherical shape comprising an aqueous core surrounded by a hydrophobic lipid membrane ( Figure 3). The lipid bilayer can be loaded with hydrophobic or amphiphilic molecules, whereas the hydrophilic molecules can be encapsulated in the aqueous reservoir of the liposomes. Often, liposomes are composed of phospholipids (e.g., phosphatidylcholines) or mixtures of phospholipids with co-lipids. Various liposome architectures can form depending on the preparation methods; for instance, multilamellar vesicles (MLV, involving a stack of several lipid bilayers), small unilamellar vesicles (SUV, constituted by a single lipid bilayer), large unilamellar vesicles (LUV), tubular vesicles, and cochleate vesicles.
Curcumin encapsulated in liposomes has been proven to be a safe formulation, which enhances the CU solubility and its cellular uptake [97][98][99][100][101]. Liposomes deliver CU into the cells via membrane fusion or endocytosis process. Liposomal formulations with a PEG surface coating provide a longer circulation time for the encapsulated drug. Biomolecular ligands can be anchored to the liposome surface in order to enhance the receptor targeting capacity, and hence, the permeability across the brain-blood barrier (BBB) [102,103]. The outcomes of the investigated CU-loaded liposomes in models of neurodegenerative diseases are summarized in Table 4.

Liquid Crystalline Nanoparticles (LCNPs) with Internal Structure
Liquid crystalline nanoparticles (LCNPs) are self-assembled architectures of lyotropic lipids, co-lipids (surfactants or oils), and water. They are typically formed upon dispersion and fragmentation of bulk lyotropic liquid crystalline phases (e.g., bicontinuous cubic, sponge, or inverted hexagonal phases) [104][105][106]. The amphiphilic molecules spontaneously organize into compartments with hydrophobic and hydrophilic domains (Figure 3), which can encapsulate lipophilic or hydrosoluble guest compounds. The structures formed by this self-assembly process are thermodynamically stable. The initial liquid crystalline phases are usually viscous and have a short-range order in comparison to solids, but long-range order in comparison to liquids. A typical example of a lyotropic liquid crystalline phase is the inverted bicontinuous cubic phase formed upon mixing of unsaturated monoglyceride lipids with water [105]. Cubosomes are produced upon dispersion of the bicontinuous cubic liquid crystalline phases in excess aqueous medium. Their periodic structures comprise folded bicontinuous lipid bilayer membranes and periodic networks of aqueous channels (Figure 3). The latter enable high encapsulation capacity for hydrophilic guest macromolecules [106][107][108][109][110][111]. Lipid nanocarriers of liquid crystalline internal structures have received considerable attention as delivery vehicles through the BBB [111,112].
Curcumin has been successfully entrapped into monolein-based LCNPs with almost 100% encapsulation efficiency [113]. LCNPs dispersion was very stable in terms of nanocarrier sizes and surface charge upon storage. LCNPs were efficiently taken up by cultures cells following the sustained release of curcumin. In addition, they provided inhibition of the cell proliferation and apoptosis induction in an anticancer study [113]. A recent investigation of an inverse hexagonal (H II ) liquid crystalline phase encapsulating curcumin has demonstrated that the release of curcumin was a concentration-diffusion controlled process in the early stages, whereas multiple diffusion mechanisms coexisted in the later stages of drug release. Radical scavenging experiments have shown that curcumin-loaded LCNPs exert the desired antioxidant activity [114]. Thus, curcumin-loaded LCNPs may be promising for neurodegenerative disease treatments using sustained-release nanoformulations for combination therapies [115,116]. Further results obtained with lipid-based LCNPs in models of neurodegenerative disease are presented in Table 4.

Conclusions
The naturally occurring compound curcumin is increasingly studied in neurodegenerative disease models due to its neurogenesis-stimulating properties and its anti-amyloid and anti-tau potential. Nanotechnology-based delivery systems of curcumin have been developed with the purpose of improving its solubility, stability, and bioavailability in potential treatment strategies of neurodegenerative disorders. We summarized recent advances in research on safe liquid crystalline lipid-based nanocarriers (cubosome, spongosome, hexosome, and liposome particles) and solid lipid nanoparticles, as well as on selected biodegradable polymer-based nanocarriers. The emphasis is given on the observed biological outcomes of the curcumin nanoformulations in in vitro and in vivo models of the multifactor neurodegenerative diseases (AD, PD, HD and ALS). Despite the difficulty of overcoming biological barriers, promising results on the enhancement of the permeability of the BBB and receptor-mediated delivery across the BBB have been reported with liposome and cubosome nanocarriers. Further investigations will be required in order to understand the involved mechanisms of action of curcumin nanoformulations in the proposed neurodegenerative disease models, and to optimize the delivery systems and strategies towards translation into clinics. Acknowledgments: This work is supported by the "IDI 2017" project funded by the IDEX Paris-Saclay, ANR-11-IDEX-0003-02. M.R. thanks S. Lesieur for helpful advices during the doctoral study.

Conflicts of Interest:
The authors declare no conflict of interest.