Fullerenols Prevent Neuron Death and Reduce Oxidative Stress in Drosophila Huntington’s Disease Model

Huntington’s disease (HD) is one of the human neurodegenerative diseases for which there is no effective treatment. Therefore, there is a strong demand for a novel neuroprotective agent that can alleviate its course. Fullerene derivatives are considered to be such agents; however, they need to be comprehensively investigated in model organisms. In this work, neuroprotective activity of C60(OH)30 and C120O(OH)44 fullerenols was analyzed for the first time in a Drosophila transgenic model of HD. Lifespan, behavior, oxidative stress level and age-related neurodegeneration were assessed in flies with the pathogenic Huntingtin protein expression in nerve cells. Feed supplementation with hydroxylated C60 fullerene and C120O dimer oxide molecules was shown to diminish the oxidative stress level and neurodegenerative processes in the flies’ brains. Thus, fullerenes displayed neuroprotective activity in this model.


Introduction
Huntington's disease (HD) is an autosomal dominant neurodegenerative disorder caused by the mutation in the huntingtin (Htt) gene which generates a mutant protein (HTT) with an expanded polyglutamine (polyQ). Oligomers accumulation or mutant HTT aggregates leads to the death of striatal and cerebral cortex neuron populations. PolyQ expansion is believed to cause abnormal β-sheets formation in protein structure with increased stability due to hydrogen bonding between the main and side chain amides [1,2]. During mutant protein folding, mechanisms targeting misfolded proteins for their eventual degradation by molecular chaperones, as well as ubiquitination and autophagy, cease to function [3]. This negatively influences transcription, energy metabolism and mitochondria, increases reactive oxygen species (ROS) and inflammation processes, and eventually leads to cell damage [4][5][6][7]. Moreover, the mutant HTT is thought to affect some cytoskeleton components, resulting in axonal transport disruption and cell death [5].
Oxidative stress (OS) is one important HD pathogenesis environment. Its negative role has been described both in studies of HD patients and in experimental models [6, 8,9]. Therefore, the study of the role of OS and the search for new effective antioxidants is key to developing a complex treatment of HD and other neurodegenerative diseases (ND) [10].
Fullerenes and their derivatives are powerful antioxidants due to their electron deficiency property accounted for by interaction with free radicals; therefore, they are called "Radical sponges" [11]. A number of studies have experimentally confirmed their antioxidant activity [12,13]. Moreover, fullerenes have been shown to prevent amyloid proteins aggregation [14][15][16]. The major or dominant factors of HD pathogenesis have not been identified so far. Perhaps they are all complementary and interdependent [4,5]. The unique properties of fullerenes and their derivatives open the possibility of not only correcting some aspects of pathology development, but also studying the role of individual components in this process on model organisms. At the same time, fullerenes hydroxylated derivatives show greater compatibility with biological systems than do native fullerenes; in addition, they have low toxicity [17] and leave the body through the urinary tract [18][19][20].
The aim of this work was to analyze the C 60 (OH) 30 and C 120 O(OH) 44 fullerenols in terms of its neuroprotective activity on a Drosophila HD model. Drosophila melanogaster is a good model to study diseases with relatively simple genetic backgrounds such as HD [21,22]. It can be used to study the effects of an increase in the CAG repeats number, as well as the influence of genetic modifiers and environmental factors on the disease pathogenesis [23][24][25]. Using Drosophila as a model organism makes it possible to study and modulate tissue-specific genes expression, test existing and novel hypotheses in pathology development and take into account a large number of animals [26][27][28]. The latter is very important, as it allows for drugs primary screening, which significantly speeds up, and reduces the cost of, the discovery and development of new drugs.

Fullerenols
Fullerenols C 60 (OH) 30 and C 120 O(OH) 44 were obtained at the Petersburg Nuclear Physics Institute, named by the B.P. Konstantinov of National Research Center as the "Kurchatov Institute" [17,29]. The structure of fullerenols is qualitatively identified by Fourier infrared spectroscopy in the range 4000-400 cm -1 using a spectrometer IRTracer-100 (Shimadzu) and MALDI-TOF/TOF mass spectrometry (Axima Resonance Mass Spectrometer-Shimadzu, Kyoto, Japan). Dynamic light scattering and zeta potential experiments were performed on a Zetasizer nano (Malvern Panalytical, Malvern, UK) size analyzer. The measurements were carried out three times. The data were collected at 25 • C by monitoring the scattered light intensity at a 90 • detection angle. Initial concentrations of fullerenols were 20 mg/mL for C 60 (OH) 30 and 1.05 mg/mL for C 120 O(OH) 44 . The optimal experimental concentrations were identified based on toxicity and antioxidant activity (AA) data in vitro.

In Vitro Toxicity Analysis of Fullerenols
For this study, we used ECV cells (human umbilical vein endothelial cells), A549 cells (human lung adenocarcinoma cell) and Hela cells (human cervical carcinoma cell) (National Research Center "Kurchatov Institute"-PNPI collection). Cells were cultured in DMEM with L-Glutamine (Capricorn Scientific, Ebsdorfergrund, Germany), antibiotics (penicillin and streptomycin, Biolot, Saint-Petersburg, Russia) and 10% bovine serum (Biolot, Saint-Petersburg, Russia). Cells were incubated at 37 • C in 5% CO 2 . Toxicity was determined using a standard MTT test (13). The culture medium was replaced with one containing C 60 (OH) 30 (0.2 and 0.02 mg/mL) or C 120 O(OH) 44 (0.01 and 0.001 mg/mL) a day after cell seeding. After 24 h, an MTT test was performed. The absorbance was determined at a wavelength of 540 nm using a Multiskan FC microplate photometer (Thermo Scientific, Waltham, MA, USA).

In Vitro Antioxidant Activity Assay
Antioxidant activity (AA) of fullerenols in vitro was measured by their ability to inhibit adrenaline autoxidation, according to [30,31] with modifications. The reaction mixture contained 100 µL 0.1% adrenaline hydrochloride solution and 2 mL 0.2M sodium carbonate buffer (pH 10.7). The mixture was mixed quickly and placed in a spectrophotometer. The absorbance was determined at 347 nm every 30 s for 15 min. Then, 2 mL 0.2 M sodium carbonate buffer, 50 µL of the investigated fullerenol solution and 100 µL 0.1% adrenaline hydrochloride solution were mixed. The absorbance was determined as described above. A similar mixture, but not containing adrenaline, was placed in the comparison cuvette. AA was expressed as the percentage of inhibition of adrenaline autoxidation and was calculated by the formula: where Abs 1 is the absorbance of adrenaline, and Abs 2 is the absorbance of adrenaline with fullerenol. A value greater than 10% indicates significant AA. The optical absorption spectra of adrenaline autoxidation products were recorded on an SF-2000 spectrophotometer (LLC OKB Spectr, Saint-Petersburg, Russia). The rootmean-square error of the arithmetic mean optical density for three experiments did not exceed 10%.

In Vivo ROS Measurement
Oxidative stress was measured using the H2DCF-DA probe. A total of 20 flies' heads were used for the lysate preparation. The heads were homogenized in a mixture containing 100 µL of 10 mM Tris and 3 µL of protease inhibitor Cocktail tablets (Roche, Penzberg, Germany) (pH = 7.4). The homogenate was centrifuged for 10 min at 10,000 rpm. A total of 5 µL of lysate and 60 µL of 5 µM H2DCF-DA (Invintrogen, Waltham, MA, USA) were dropped into a well of a 96-well plate and were incubated for 60 min at 37 • C. The signal was measured on a flatbed analyser with λwt = 490 nm, λem = 530 nm, and h = 4.7 mm. The signal was normalized to protein concentration measured by the Bradford method.

Drosophila Lifespan Experiment and Climbing Assay
Flies were placed in test tubes with agar (30 flies, male and female, per tube) at 25 • C. The flies were fed with yeast inactivated at 65 • C with the addition of 0.2 mg/mL of fullerenols C 60 (OH) 30 , or 0.01 mg/mL C 120 O(OH) 44 . First, yeast was thoroughly mixed with nanoparticles using vortex, and then 80 µL of feed was dripped onto the agar surface, which is not a nutrient medium for Drosophila. Every two days, living flies were transferred to fresh food. Simultaneously, the number of dead specimens was counted. Males and females were counted separately. The observation was carried out until the last dead fly was detected, and the maximum lifespan was determined. There were at least 300 flies in each experiment. Each experiment was run using triplicates.

Analysis of Neurodegeneration
Brain preparations for confocal microscopy were prepared according to the method [33]. A series of images were obtained using a Leica TCS SP5 microscope (Leica, Wetzlar, Germany) with a built-in 35-mW argon laser at a wavelength of 488 nm. The shooting parameters were an optical thickness of 2 µm and image resolution of 1024 × 1024 and 40× (oil) objective. The cells number of interneurons (IN), belonging to the cholinergic neurons (CN), was counted on 3D projections in Leica Application Suite X and ImageJ. The total neurodegeneration level was assessed in paraffin brain sections. Fly heads were fixed in 3.7% paraformaldehyde (Sigma-Aldrich, Saint Louis, MO, USA) for 24 h, embedded in paraffin, and 6 µm brain slices were stained with hematoxylin/eosin (Bio Optica, Milano, Italy). The paraffin slices were examined using a Leica DM 2500 microscope. To estimate the extend of degeneration, we calculated the ratio between areas free of cells and the total area of the brain.

Immunoblotting
Western blot analysis was carried out according to the method, which is described in detail in the article with the modifications [33]. A total of 100 heads were taken for each experiment. Blots probed using the primary Anti-Huntingtin Protein antibody, a.a. 181-810, clone 1HU-4C8 (Millipore Cat # MAB2166) or mouse monoclonal anti-b-tubulin (Santa Cruz Biotechnology Cat # sc-365791, RRID: AB_10841919) antibodies were diluted in BlockPRO buffer. Samples with antibodies were incubated at 4 • C overnight.
Bands were quantified using ImageJ software (NIH, Bethesda, MD, USA). The density of the background staining was subtracted from all signals, and the densities of the 348 kDa Huntingtin bands were normalized to the densities of the tubulin bands (55 kDa).

Statistical Analysis
Statistical analysis was performed using KyPlot 6.0 software. The Dunnetts-test was used for a comparison of two samples (toxicity, in vivo OS analysis, climbing assay and total neurodegeneration) and the Tukey-Kramer was used for multiple comparisons (to analyze the IN number). Differences at p < 0.05 were considered statistically significant.

Toxicity Analysis and Antioxidant Activity of Fullerenols In Vitro
The fullerenes synthesis and characterization used in the work have previously been described in detail [17]. As shown in Figure S1 (Supplementary Materials), the four intramolecular vibrational characteristic modes T 1u of pristine C60 (A) at 526 cm −1 , 576 cm −1 , 1184 cm −1 and 1428 cm −1 disappear from the FTIR spectra of dimer oxide C 120 O (D) and fullerenols (B, C). The dimer oxide C 120 O contains characteristic bands at~800 cm −1 , 1028 cm −1 , 1080 cm −1 , 1261 cm −1 [34]. The fullerenols spectra (B, C) has broad hydroxyl absorption around 3390 cm −1 , a C-O stretching absorption at 1080 cm −1 and C = C absorption at 1618 cm −1 . The peak centered at 1370 cm −1 can be assigned to bending vibrations of hydroxyl groups. IR spectra of fullerenols also showed the appearance of a C = O stretching peak at 1720 cm −1 .
The mass spectrum of C 60 (OH) 30 ( Figure S2A, Supplementary Materials) showed peaks with a high intensity at m/z = 720, which demonstrates that the fullerene cage has not been damaged during fullerenol synthesis. Detection of a molecular ion at m/z 1230 was indicative for the composition of fullerenol as containing 30 hydroxyl groups per C 60 cage. Fullerenols under the conditions of MALDI-TOF analysis lose their hydroxyl groups, and the use of various matrices (dihydroxybenzoic acid, dithranol) did not change the results. A similar situation is observed in the mass spectrum of C 120 O(OH) 44 fullerenol ( Figure S2B, Supplementary Materials). In addition to the detachment of hydroxyl groups, defragmentation of the initial oxide molecule of the C 120 O dimer also occurs, with successive detachment of two carbon atoms from the cages, as evidenced by the appearance of lines in the spectra that differ by 24 Da.
Previously, we found that fullerenols C 60 (OH) 30 and C 120 O(OH) 44 in the concentration range of 0.01-1.0 mg/mL (C 60 (OH) 30 ) and 0.005-0.1 mg/mL (C 120 O(OH) 44 ) in an aqueous solution form polydisperse aggregates without a clear dependence on the concentration of particle sizes and zeta potential [17]. Figure S3 shows the zeta potential and size distribution for fullerenols C 60 (OH) 30 and C 120 O(OH) 44 at concentrations of 2 mg/mL and 0.1 mg/mL, respectively.
As can be seen from Figure S3 (Supplementary Materials), there are aggregates of two sizes in the fullerenol solution-144 ± 44 nm and 814 ± 143 nm for C 60 (OH) 30 and 14 ± 4 nm and 516 ± 45 nm in C 120 O(OH) 44 solution. Zeta potential was negative with the mean value of −15.2 ± 6.3 mV for C 60 (OH) 30 and −28.8 ± 6.3 mV for C 120 O(OH) 44 .
According to modern concepts, all newly synthesized carbon nanoparticles derivatives should be tested for toxicity. Fullerenols synthesized for this study were analyzed using the MTT test. C 120 O(OH) n was additionally purified from traces of unreacted hydrogen peroxide during the hydroxylation reaction, since in previous work C 120 O(OH) n showed greater toxicity than C 60 (OH) 30 [17]. Also, the OH groups number in C 120 O(OH) n was determined to be 44 ± 2 using the thermal estimation method of the substituents number [35]. It has been shown that C 60 (OH) 30 at the studied concentrations is safe for ECV, A549 and Hela cells. C 120 O(OH) 44 was non-toxic to ECV and A549 cells and Hela cell viability was reduced by only 10-15% ( Figure 1A). Changes in cell morphology were not observed when exposed to nanoparticles.
It is known that the attachment to the fullerene cage of even one addend can lead to a significant change in its reactionary ability, including in relation to free radicals. Figure  S4 (Supplementary Materials) presents the absorption spectra of autoxidation products of pure adrenaline and adrenaline in the presence of C 60 (OH) 30 in a 0.2 mg/mL concentration and C 120 O(OH) 44 in a 0.01 mg/mL concentration 5 min after the reaction started. As can be seen from Figure S4 (Supplementary Materials), the C 60 (OH) 30 and C 120 O(OH) 44 addition does not lead to the appearance of a new absorption peak, while the absorbance value of the intermediate product of adrenaline autoxidation is reduced at 347 nm. A similar situation is observed with the addition of C 120 O(OH) 44 to adrenaline. It is assumed [36] that the fullerenols do not directly interact with adrenaline and its oxidation products, and the inhibiting of the adrenaline autoxidation reaction is achieved by the competing reaction of the fullerenols' interaction with the superoxide anion radicals (O 2−˙) that occur during intramolecular restructuring of the adrenaline molecule, as Figure S5 (Supplementary Materials) shows. Superoxide radicals can join sp 2 -hybridized carbon atoms with the formation of fullerenol anion-radical. The water present in the system then gives a proton to a superoxide radical with the formation of the OH − and HOO˙groups on the carbon cage. Then these two stages are repeated and two HOO˙groups on the carbon cage are isomerized with the formation of oxygen and hydrogen peroxide. The schemes of these processes are shown in Figure S6 (Supplementary Materials). Figure 1B,C shows the dependence of the absorbance of the intermediate product of adrenaline oxidation on the reaction time of adrenaline autooxidation in the absence and in the presence of fullerenols. The results presented in Table 1 reveal that the AA of the fullerenols depends on the concentration being non-linear. The AA change during the adrenaline autoxidation is associated with the transition of the intermediate products of the reaction with an absorption maximum at 347 nm to adrenochrome with an absorption maximum at 480 nm. Therefore, it is more reliable to evaluate AA at a time of 10 to 15 min from the start of the adrenaline autoxidation reaction, when the rate of intermediate products formation is equal to the rate of their transition to adrenochrome. It is known that the attachment to the fullerene cage of even one addend can lead to a significant change in its reactionary ability, including in relation to free radicals. Figure  S4 (Supplementary materials) presents the absorption spectra of autoxidation products of pure adrenaline and adrenaline in the presence of C60(OH)30 in a 0.2 mg/mL concentration

Lifespan and Behavior Analysis of Drosophila in a Huntington's Disease Model
Characteristic features of HD are reduced lifespan and movement disorders. To model HD, we activated HTT neuronal-specific expression with normal (HTT.16Q) and expanded (HTT.128Q) polyQ repeats in flies using GAL4/UAS binary system. The lifespan of HTT.128Q-expressing flies was reduced by more than half compared to those expressing HTT.16Q (Figure 2A). Sex was taken into account when counting the dead flies. The lifespans of males and females with HTT.16Q was equal, while in the HTT.128Q flies, the lifespan of males was shorter than that of females. Locomotor activity was monitored by a climbing assay (negative geotaxis analysis). The ability to climb decreased with age in the HTT.128Q flies at a significant rate compared to the HTT.16Q flies ( Figure 2B), and was minimal on the 25-day-old flies. We did not observe a significant difference between the males' and females' locomotor activity, and thus combined them in the analysis in a 1:1 ratio.  We tested the fullerenols effects on lifespan and locomotion activity on this model, since the HTT.128Q neuronal-specific expression in flies led to a violation of the main vital signs. The lifespan and locomotion activity of males and females with HTT.128Q did not decrease when C60(OH)30 at a dose of 2 mg/mL or 0.2 mg/mL and C120O(OH)44 at a dose of 0.1 mg/mL or 0.01 mg/mL were added to the feed (Figures 3A-C). Thus, fullerenols were proven to be safe for flies.

Analyses of Oxidative Stress
Next, we assessed the level of ROS in the HTT.128Q fly brain using 2',7'-dichlorodihydrofluorescein diacetate. As shown in Figure 3D, we found that ROS levels were higher in the brains of the 15-day-old and 25-day-old HTT.128Q flies We tested the fullerenols effects on lifespan and locomotion activity on this model, since the HTT.128Q neuronal-specific expression in flies led to a violation of the main vital signs. The lifespan and locomotion activity of males and females with HTT.128Q did not decrease when C 60 (OH) 30 at a dose of 2 mg/mL or 0.2 mg/mL and C 120 O(OH) 44 at a dose of 0.1 mg/mL or 0.01 mg/mL were added to the feed ( Figure 3A-C). Thus, fullerenols were proven to be safe for flies.

Analysis of Neurodegeneration
It is well known that oxidative stress is a hallmark of neurodegeneration, although the relationship between cell death and the accumulation of reactive oxygen species remains unclear [37]. We performed histological analyses of brain sections stained with hematoxylin and eosin on days 15 and 25-30 of the fly's adult life ( Figure S7, Supplementary materials). We observed an increase in vacuolization and overall neurodegeneration in the brain of HTT.128Q flies with age ( Figure 4A). These parameters were corrected by adding C60(OH)30 and C120O(OH)44 fullerenols to food. The vacuoles area on the brain sections was less in the 15-day-old HTT.128Q flies, fed with fullerenols, unlike the control group. However, on the 25th day, a statistically significant slowdown in neurodegeneration was observed only when fullerenol C120O(OH)44 was added to the flies' food ( Figure 4A).
It has previously been shown that during the development of HD, mainly the GA-BAergic and cholinergic systems are depleted [5]. At the same time, cholinergic neurons are the most numerous and play an important role in Drosophila [38]. Therefore, we expressed HTT.128Q in acetylcholinergic neurons and determined their number on days 5, 15 and 25 by confocal microscopy. Intermediate cholinergic neurons (INs) were selected for analysis, which are well distinguishable in the fly brain ( Figure 4C). It was shown that the INs number in HTT.128Q flies decreases with age ( Figure 4B). Adding fullerenols to food for 15 days prevented neuron death and increased numbers in the brain. However,

Analyses of Oxidative Stress
Next, we assessed the level of ROS in the HTT.128Q fly brain using 2',7'-dichlorodihydrofluorescein diacetate. As shown in Figure 3D, we found that ROS levels were higher in the brains of the 15-day-old and 25-day-old HTT.128Q flies compared to the control flies. Fullerenols are active scavengers of ROS, unlike some other antioxidants imitating the activity of antioxidant defense enzymes. Therefore, we analyzed the effect of fullerenols on the level of ROS in the brain of HTT.128Q flies. The content of Drosophila on the medium, including fullerenols (C 60 (OH) 30 at a dose of 0.2 mg/mL or C 120 O(OH) 44 at a dose of 0.01 mg/mL), led to a significant decrease in the ROS level in 15-day-old flies.

Analysis of Neurodegeneration
It is well known that oxidative stress is a hallmark of neurodegeneration, although the relationship between cell death and the accumulation of reactive oxygen species remains unclear [37]. We performed histological analyses of brain sections stained with hematoxylin and eosin on days 15 and 25-30 of the fly's adult life ( Figure S7, Supplementary Materials). We observed an increase in vacuolization and overall neurodegeneration in the brain of HTT.128Q flies with age ( Figure 4A). These parameters were corrected by adding C 60 (OH) 30 and C 120 O(OH) 44 fullerenols to food. The vacuoles area on the brain sections was less in the 15-day-old HTT.128Q flies, fed with fullerenols, unlike the control group. However, on the 25th day, a statistically significant slowdown in neurodegeneration was observed only when fullerenol C 120 O(OH) 44 was added to the flies' food ( Figure 4A). at 25 days old, the effectiveness of fullerenols dropped ( Figure 4B,C) and there was no statistically significant difference between the control group and the HTT.128Q flies.  It has previously been shown that during the development of HD, mainly the GABAergic and cholinergic systems are depleted [5]. At the same time, cholinergic neurons are the most numerous and play an important role in Drosophila [38]. Therefore, we expressed HTT.128Q in acetylcholinergic neurons and determined their number on days 5, 15 and 25 by confocal microscopy. Intermediate cholinergic neurons (INs) were selected for analysis, which are well distinguishable in the fly brain ( Figure 4C). It was shown that the INs number in HTT.128Q flies decreases with age ( Figure 4B). Adding fullerenols to food for 15 days prevented neuron death and increased numbers in the brain. However, at 25 days old, the effectiveness of fullerenols dropped ( Figure 4B,C) and there was no statistically significant difference between the control group and the HTT.128Q flies.

Analyses of HTT Level
Human HTT was detected in the fly brain using the HTT-specific antibody MAB2166. We detected a band corresponding to the full-length protein (∼350 kDa). We did not detect insoluble HTT, which is consistent with data obtained by other researchers [39]. The soluble HTT level in the fly brain did not differ between those who received normal food and food with fullerenols (C 60 (OH) 30 at a dose of 0.2 mg/mL or C 120 O(OH) 44 at a dose of 0.01 mg/mL) for 30 days ( Figure 5A,B).

Analyses of HTT Level
Human HTT was detected in the fly brain using the HTT-specific antibody MAB2166. We detected a band corresponding to the full-length protein (∼350 kDa). We did not detect insoluble HTT, which is consistent with data obtained by other researchers [39]. The soluble HTT level in the fly brain did not differ between those who received normal food and food with fullerenols (C60(OH)30 at a dose of 0.2 mg/mL or C120O(OH)44 at a dose of 0.01 mg/mL) for 30 days (Figure 5A,B).

Discussion
The various HD models differing in the length of polyglutamine repeats have been created in Drosophila to study pathogenesis and search for potential therapeutic agents. The flies expressing elongated polyQ repeats imitate HD in terms of reduced lifespan, reduced motor activity, photoreceptor degeneration, etc. [40]. It has been shown that the pathological effects of HTT are not necessarily associated with its aggregation. Thus, Akbergenova Y et al. [28] revealed that pathogenic HTT can be present in a significant amount in Drosophila synapses in a non-aggregated form, disrupting synaptic growth. Its accumulation in neuromuscular junctions is stochastic. At the same time, increased mortality of the third instar larvae has been observed. Romero E et al. also used a model

Discussion
The various HD models differing in the length of polyglutamine repeats have been created in Drosophila to study pathogenesis and search for potential therapeutic agents. The flies expressing elongated polyQ repeats imitate HD in terms of reduced lifespan, reduced motor activity, photoreceptor degeneration, etc. [40]. It has been shown that the pathological effects of HTT are not necessarily associated with its aggregation. Thus, Akbergenova Y et al. [28] revealed that pathogenic HTT can be present in a significant amount in Drosophila synapses in a non-aggregated form, disrupting synaptic growth. Its accumulation in neuromuscular junctions is stochastic. At the same time, increased mortality of the third instar larvae has been observed. Romero E et al. also used a model with no accumulation of mutant HTT aggregates [39]. They have demonstrated that the expression of HTT.16Q, as well as HTT.128Q, resulted in a full-length~350 kDa human protein accumulation in the flies' heads. The protein has been located in the cytoplasm and did not form aggregates in axons. At the same time, HTT.128Q expression was associated with the progressive degeneration of photoreceptor cells and degenerative phenotypes in the peripheral nervous system, as well as with those correlated with them: early death, behavioral disorders and impaired electrophysiological parameters. According to the authors, neurodegeneration in this case is based on Ca 2+ -dependent increased neurotransmission in the early stages of pathogenesis, caused by an increased CAG triplets number in the protein.
Using the model of Romero et al. [39], we obtained similar results: a decrease in life expectancy, impaired behavior and neurodegeneration against the background of the formation of only soluble HTT forms. We also observed ROS level increase in the brains of HTT.128Q flies with age. According to modern concepts, OS is a significant factor leading to the emergence of many NDs and one of the central nonspecific pathogenesis components [41]. There is evidence that mitochondrial disfunctions and energy metabolism defects in the CNS are important in triggering HD [8]. The mutant HTT is known to change the PGC-1α factor functional activity after binding. The PGC-1α factor regulates the genes expression of the main antioxidant enzymes SOD1, SOD2 and Gpx-1 [42]. Impairment of antioxidant defense system leads to ROS excessive formation, causing lipid peroxidation, cell structures, proteins, DNA, RNA damage and mitochondrial dysfunction [41]. Altered biomolecules themselves become OS sources and a vicious circle arises [43]. Severe and prolonged OS can trigger apoptosis and necrosis, and, accordingly, lead to selective damage to neurons [8].
The neuroprotective and antioxidant activity of carbon nanoparticles, in particular fullerenes and their derivatives, has been shown in studies in vitro [44][45][46], as well as in vivo models of Alzheimer's, Parkinson's, etc. [47]. We have previously shown the neuroprotective effect of fullerenols on the models of Parkinson's disease and Alzheimer's disease in Drosophila. In the first case, C 60 (OH) 30 and C 70 (OH) 30 reduced OS and the level of the alpha synuclein insoluble form in the flies' brains, and prevented dopaminergic neurons death [33]. In the second case, C 60 (OH) 30 prevented the aggregation of amyloid peptide beta, and reduced the ROS number and the overall neurodegeneration level in the brain [48]. The perspectives of fullerene applications for HD treatment have been broadly discussed in the literature, but we are aware of only one experimental work devoted to this problem [49].
In this work, we have shown that fullerenols C 60 (OH) 30 at a dose of 0.2 mg/mL or C 120 O(OH) 44 at a dose of 0.01 mg/mL led to a significant decrease in the level of ROS in the brain of 15-and 25-day-old HTT.128Q flies. However, the decrease in the level of ROS maximally correlated with the slowing down of neurodegeneration only on the 15th day, and the effectiveness of fullerenols in preventing neuronal death decreased on the 25th day. Our results suggest that OS plays a considerable, but not dominant, role in neuronal death. The vital function for neurons is mediated by HTT: it promotes the movement of various vesicles along axons, acting as a connecting element (framework) between motor proteins, "cargo" and microtubules. Mutations in Htt disrupt the organelles transport, including mitochondria, into axons. At the same time, abnormal forms of the protein have been shown to interact with the Drp-1 protein regulating mitochondria division. As a result, the accumulation of fragmented organelles takes place, followed by apoptosis activation [50]. Note that we did not observe a large difference in the effects of C 60 (OH) 30 and C 120 O(OH) 44 , which may be due to the comparable sizes of their particles and zeta potential [17].
We were unable to affect the lifespan and locomotor activity of flies with the help of fullerenols. But it should be noted that the integral indicator of vital activity-lifespan of D. melanogaster-is highly conservative and is not always amenable to correction by drugs. The test for locomotor activity also did not show a correlation with a decrease in neuron death. We assume that in our model, as in cases of ND in humans, the ROS level reaches critical values with age, and the antioxidants used are no longer able to prevent neuronal death or locomotor impairments or increase life expectancy. According to Forman and Zhang [51], antioxidant enzymes play an even larger role in antioxidant protection than ROS scavengers. That may explain why exogenous antioxidants demonstrate limited effectiveness.
Thus, using the Drosophila HD model, we were able to demonstrate for the first time that C 60 (OH) 30 and C 120 O(OH) 44 fullerenols have antioxidant and neuroprotective activity. Although the involvement of oxidative stress in the pathogenesis of HD is now well established [52], it is still not clear whether oxidative stress is a cause, a consequence or simply a by-product of the neurodegenerative process that occurs in the brains of HD patients. In addition, Huntington's disease is currently considered a multifactorial disease, which implies the creation of a complex therapy that affects various cellular processes, part of which could be water-soluble fullerenols. However, further investigations are required to elucidate the detailed mechanism of fullerenols activity.  Figure S4. The optical absorption spectra of adrenaline autoxidation products at 5 min of the reaction; Figure S5. Scheme of the formation of adrenochrome and superoxide anions in the process of autoxidation of adrenaline along the quinoid pathway; Figure S6. The possible O 2 •− -scavenging mechanism for fullerenols C 60 (OH) 30 and fullerenol C 120 O(OH) 44; Figure S7. Paraffin brain section of the flies with HTT.128Q expression in neurons at different ages. Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available in the article and Supplementary Materials.

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