Heat shock-induced extracellular vesicles derived from neural stem cells confer marked neuroprotection against oxidative stress and amyloid-β caused neurotoxicity

Abstract

Alzheimer’s disease (AD) is the most prevalent neurodegenerative disease and a leading cause of dementia. Although the amyloid-β (Aβ) peptide is deemed a crucial driver of AD, there are no effective therapeutics available to treat Aβ-caused neurotoxicity. Extracellular vesicles (EVs) are membrane-bound small particles mediating intercellular traffic of nucleic acids, lipids, proteins, and metabolites. Exosomes are a subtype of EVs with a size range of 30 – 150 nanometers in diameter. Stem cell-derived EVs are a potential therapeutic for AD, while EVs isolated from normal stem cell cultures generally have a low yield. Here, we studied the EVs secreted by the rat neural stem cells in the presence of heat shock (HS) stimulus. Nanoparticle tracking analysis confirmed HS-derived EVs exhibit significantly higher concentration and larger diameter in comparison to the non-heat shock (NHS)-derived EVs. Mass spectrometric studies of EV proteins reveal that HS-derived EVs contained fewer diverse proteins than NHS-derived exosomes. GO enrichment analysis of the proteins suggested that the top two biological functions of the proteins in HS-derived EVs are involved in the negative regulation of apoptotic process and positive modulation of DNA repair. Importantly, the therapeutic efficacy of the NHS- and HS-derived EVs were tested in a cell culture model of AD: HS-derived EVs exhibited greater neuroprotection against not only oxidative stress but also amyloid-β (Aβ) induced neurotoxicity compared to NHS-derived EVs. Moreover, HS-derived EVs were also able to dramatically attenuate Aβ-induced apoptosis and oxidative stress. These data indicate that in response to HS, neural stem cells increase EV production and alter EV morphology and cargo to confer better neuroprotection against oxidative stress and Aβ caused neurotoxicity, suggesting that HS-induced EVs from neural stem cells can be a therapeutic reagent for AD and possibly other neurological disorders.

Introduction

Alzheimer’s disease (AD) is a prevalent neurodegenerative disorder and the most common cause of dementia in the elderly. The disease is associated with two defining neuropathological characteristics: the accumulation of extracellular senile plaques and intracellular neurofibrillary tangles. By the year 2060, 13.8 million Americans over the age of 65 are projected to be diagnosed with the disease [1]. Despite this, current therapeutics for this disorder are not able to effectively attenuate or prevent the progression of the disease [2].

Membrane-bound vesicles secreted into the extracellular space are referred to as extracellular vesicles (EVs). This generic term represents a heterogeneous population of vesicles with different functions, biogenesis, and properties that can be further divided into 3 major subtypes: exosomes, microvesicles (MVs), and apoptotic bodies. MVs are 50 – 1,000 nanometers in diameter EVs released from the plasma membrane by clathrin-mediated shedding. Apoptotic bodies are 1,000 – 5,000 nanometers in diameter EVs generated by blebbing and fragmentation of apoptotic cells. Exosomes are 30 – 150 nanometers in diameter EVs generated from the inward budding of multivesicular bodies and released upon fusion with the plasma membrane [3]. The small-sized EVs or exosomes derived different stem cells exhibit favorable characteristics as a therapeutic reagent: low immunogenicity and toxicity, ability to cross the blood-brain barrier (BBB), and biodegradability and contain neuroprotective molecules [4,5]. Previous studies have demonstrated that exosomes are a diagnostic tool and potential therapeutic reagent in cancer immunotherapy, ischemic stroke, diabetes and cardiovascular diseases [6].

Even though exosomes exhibit advantageous characteristics, the low yield from cultured cells, mostly stem cells, has limited application as a therapeutic reagent. For instance, the yield of exosomes from culture media is usually less than 1 μg of exosomal protein from 1 mL of media [7]. To overcome this hurdle, the present study aims to determine whether the highly proliferating rat neural stem cell, HC2S2 [8], can increase the yield of EVs and exosomes in response to heat shock (HS) and whether HS-induced exosomes confer neuroprotection against oxidative stress and Aβ caused neurotoxicity in neuronal cell cultures.

Materials and methods

Cell culture

The HC2S2 rat neural stem cell line that was described previously [9,10] was utilized for the isolation of EVs and exosomes. Cell culture petri-dishes (10 cm) were coated with Matrigel for 2 hours at room temperature prior to use. Cells were cultured in DMEM/F12 supplemented with N2 medium and 20 ng/ml basic fibroblast growth factor. Cells were grown at 37°C in 5% CO₂ to 80 – 85% confluency before passage.

Heat-shock of HC2S2 cells

Once cells reached 80% confluency, media was collected and utilized for non-heat shock (NHS) exosome isolation. Then, fresh media was added, and cells were incubated at 42°C in 5% CO₂ for three hours to stimulate HS conditions [11]. Subsequently, cells were returned to 37°C in 5% CO₂ for 21 hours, and media was then collected for isolation of EVs and exosomes from heat shock-derived cells.

Exosome isolation

Exosomes were isolated and purified using a sucrose gradient ultracentrifugation method adapted from our previously described protocols [12,13]. To collect the crude exosomes, or EVs, the cell culture media was collected by centrifugation at 300 x g for 20 minutes to remove cellular debris. The supernatant was centrifuged at 1,000 x g for 15 minutes to remove apoptotic bodies. The supernatant was then pooled and ultracentrifuged at 100,000 x g for 90 minutes in the Beckman SW41Ti Rotor using Beckman Coulter ultracentrifuge tubes. For fractionation, the supernatant was discarded, and the resulting pellets (crude exosomes, or EVs) were resuspended in 0.32 M sucrose and added to the top of a sucrose gradient with the following concentrations of sucrose in 5 mM HEPES dissolved in dH₂O with a pH of 7.4: 0.72 M, 1.16 M, 1.58 M and 2.0 M. The sucrose gradient was then ultracentrifuged at 100,000 x g for 16 hours. Next, ten 1 mL fractions were collected using a fractionation machine and each individual fraction was added to 1X PBS and further centrifuged at 100,000 x g for 60 minutes. The supernatant was discarded, and the pellets were resuspended in 1X PBS and stored at −80°C until use. All centrifugation steps were carried out at 4°C.

Exosome characterization

Nanoparticle Tracking Analysis (NTA) was conducted to assess the concentration and size of exosomes collected from HS- and NHS-derived cells. The NanoSight NS300 system utilized 100 μg of exosomal protein diluted in 1X PBS for comparison between the size and concentration of exosomes isolated from the two conditions (Malvern Instruments). Transmission Electron Microscopy (TEM) was conducted to further validate the morphology of the collected samples adapted from a previously described protocol [14]. Briefly, NHS and HS-derived exosomes were resuspended in water containing 0.2% paraformaldehyde (PFA). Samples (10 microliters each) were prepared by drop-casting on a graphine-based grid and then 1% phosphotungstic acid was applied as a contrast agent. Samples were imaged using the FE-TECNAI-G2 transmission electron microscope (FEI) with an exposure of 1.0 sec and magnification of SA 59,000.

Western blot analysis was conducted on the 10 isolated fractions to determine the presence of known exosomal markers. Proteins from each fraction were separated onto 12% Bis-Tris gels and transferred to polyvinylidene difluoride membranes. Antibodies against HSP-90 and Flotillin-1 were used to characterize the isolated fractions.

Mass spectrometric analysis of exosomal proteins

The exosomal proteins were separated by SDS-PAGE, and the gel was stained with Coomassie G250. Each gel line corresponding to each replicate/group was cut into small slices before being subjected to destaining and in-gel digestion as previously described [15,16,13]. The tryptic peptides were analyzed in an Easy nLC 1200 using a trapping and desalting online trap column (300 μm x 20 mm Acclaim PepMap C18 100Å (Thermo Scientific), and were separated by an Acclaim PepMap RSLC 2 μm, 75 μm x 15 cm, nano Viper (Thermo Scientific) coupled to nanoESI QExactive Plus Orbitrap HR/MA mass spectrometer as previously reported [17,15]. The ions were analyzed in positive MS ion mode (m/z 300 – 1800) with 70,000 resolution (m/z 200) after accumulation with target ions to 1 × 10⁶ value based on predictive AGC. The MS/MS ions selection was set at m/z 65-1800 to 1 × 10⁵ counts. The spectrum was deconvoluted and analyzed using the Mascot Distiller v2.6 (www.matrixscience.com) and Discovery and Proteome Discoverer v2.1 (Thermo Scientific). A mascot generic format list (MGF format) was generated to identify +1 or multiple charged precursor ions from the MS data file.

Mascot server v2.7.0.1 (www.matrix-science.com, UK) in MS/MS ion search mode (local licenses) was applied to conduct peptide matches (peptide masses and sequence tags) and protein searches against SwissProt 2021_04 (565,928 sequences; 204,173,280 residues) using taxonomy filter for Rattus norvegicus. The protein redundancy that appeared at the database under different GI and accession numbers was limited to Human. All the proteins identified in the current study were found in these domains.

Bioinformatic analysis of NHS and HS exosomal proteins

The DAVID functional annotation tool (https://david.ncifcrf.gov/summary.jsp) was utilized to perform gene ontology (GO) and pathway analysis of significant interactors. The UNIPROT IDs for each protein were converted to the official gene ID with Rattus norvegicus selected as the species. The default settings for thresholds of 2 genes per term and an EASE threshold of 0.1 were utilized for the analysis. The p-value was converted to the −log(p). The Venny 2.1 software (https://bioinfogp.cnb.csic.es/tools/venny/) was utilized to determine differences in GO terms and pathways between NHS and HS exosomal proteins.

Treatment of neuronal cultures with NHS- or HS-derived exosomes or EVs in the presence of oxidative stress or Aβ peptide

Mouse neuronal cells obtained from the Coriell Institute for Medical Research were cultured in 12-well plates in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and antibiotics as previous described [18]. The cells were co-treated with 250 μM H₂O₂ in the presence of 2 μg/ml of crude NHS- or HS-derived crude exosomes, or EVs, for 8 h. Alternatively, neuronal cultures were treated with 250 nM of Aβ_1-42 oligomers as previously described [19] in the presence of 4 μg/ml of purified NHS or HS-exosomes (pooled fraction 3-8) or for 24 h. Following the treatments, the cells were collected for assessment of cell viability by measuring ATP levels using an ATP assay kit as previously described [20]. Aβ treated cells were also assessed for apoptosis using an Annexin V conjugates kit (ThermoFisher) by following the manufacturer’s instructions. Annexin V positively stained cells were captured with a fluorescence microscope equipped with the ZEN 2.5 software (Carl Zeiss) and the total cells were determined by Hoechst 33342 staining. To examine the level of reactive oxygen species (ROS), a ROS indicator, H2DCFDA (ThermoFisher), was utilized as advised by the manufacturer’s guide. The dye-loaded cells were homogenized in a buffer (50 mM Tris-HCl, pH 7.5, 250 mM sucrose, 5 mM MgCl₂, 0.5 mM EDTA, 2 mM ATP and 1mM dithiothreitol) by passing 10 times through a 27-gauge needle attached to a 1-ml syringe. The cell debris was removed by centrifugation and the resultant supernatant was subjected measuring fluorescence intensity with a fluorescence microplate reader (Biotek).

Statistical analysis

All numerical data are presented as mean ± SD or SEM. All experiments were repeated for at least three times (n ≥ 3). For comparison between NHS- and HS-derived EVs in terms of concentration and diameter, a student’s t test was utilized. All other analyses were carried out using a one-way ANOVA followed by a Tukey’s post hoc test. Statistical significance was accepted when p < 0.05. For statistical analysis and graphical displays, GraphPad Prism Statistical Software version 8.2.1 was utilized.

Results

HS-derived EVs/exosomes exhibited increased concentration and diameter compared to NHS-derived EVs/exosomes

EVs and exosomes were isolated from the HC2S2 neural stem cells incubated at 37°C (NHS) or 42°C (HS). Exosomes were then subjected to characterization via WB, TEM and NTA to determine protein levels of known exosomal markers, morphology, size and concentration of the particles, respectively. WB analysis revealed the isolation of exosomes based on two known exosomal markers from NHS- (Fig. 1A) and HS-derived cells (Fig. 1B). Using TEM, we observed that the exosomes isolated from both NHS- (Fig. 1C) and HS-derived cells (Fig. 1D) exhibited characteristics consistent with membrane-bound vesicles. NTA determined that NHS-derived exosomes exhibited an average diameter of 111.3 nanometers with an average concentration of 2.63 x 10¹⁰ particles/mL (Fig. 1E). Further, NTA determined that HS-derived exosomes displayed an average diameter of 129.2 nanometers and an average concentration of 3.46 x 10¹¹ parti cles/mL (Fig. 1F). HS-derived exosomes exhibited a statistically significant increase in diameter in comparison to the NHS-derived exosomes (p < 0.05; data analyzed by a student’s t test) (Fig. 1G). HS-derived exosomes also exhibited a statistically significant increase in concentration in comparison to NHS-derived exosomes (p < 0.0001; data analyzed by a student’s t test). The concentration of HS-derived exosomes are approximately 13-times higher than NHS-derived exosomes (Fig. 1H). Consequently, the yield of exosomes from the HS-induced cell cultures was 269.36 μg of total exosomal protein from 1 ml medium, which is dramatically higher than the yield from NHS treated cell culture (27.3 μg/ ml). Thus, we have isolated and characterized exosomes from two treatment groups from neural stem cells: cells exposed to HS conditions and cells exposed to normal conditions. Hence, the neural stem cells utilized in this study produced a significantly higher concentration and an increase in diameter of exosomes in response to HS stress compared to the NHS condition.

Fig. 1. HS-derived EVs/exosomes exhibit increased diameter and concentration compared to NHS-derived exosomes.

Western blot analysis of known exosomal markers in 10 isolated fractions: NHS (A) and HS (B). Transmission electron microscopy highlighting the morphology of NHS- (C) and HS-derived exosomes (D). Nanoparticle tracking analysis to determine the concentration and average size of the particles in solution: NHS- (E) and HS-derived EVs/exosomes (F). Size (G) and concentration (H) comparison between NHS- and HS-derived EVs/exosomes analyzed from the NTA. Data were shown as mean ± SEM; n = 5, * p < 0.05; *** p < 0.0001 (Student’s t test).

HS-derived EVs/exosomes contained fewer diverse proteins than NHS-derived EVs/exosomes but were enriched in anti-apoptotic and DNA repair proteins

Mass spectrometric analysis of exosomal proteins revealed that HS-derived exosomes contained 131 different proteins, which is much fewer than 301 proteins identified exclusively in NHS exosomes (Fig. 2A). Gene Ontology (GO) analysis of NHS and HS derived exosomal proteins identified differential molecular functions. Intriguingly, the top two biological functions for HS-derived exosomal proteins were involved in negative regulation of apoptotic processes and modulation of DNA repair (Fig. 2C, Table 1 & 2). On the other hand, the NHS-derived exosomes contained proteins responsible for protein transport and cell-cell adhesion (Fig. 2B). These data suggest that HS-derived exosomes contain different proteins that may have differential functions when compared to NHS derived exosomes.

Fig. 2. HS-derived EVs/exosomes are enriched in anti-apoptotic and DNA repair proteins.

Mass spectrometric analysis of HS- and NHS-derived EV/exosome proteins (A). GO enrichment analysis of biological functions for proteins unique to NHS- (B) and HS-derived exosomes (C).

Table 1.

HS-exosome proteins involving negative regulation of apoptotic process

UNIPROT ID	Name of protein
BAG6	BCL2-associated athanogene 6(Bag6)
GNAQ	G protein subunit alpha q(Gnaq)
SMCA4	SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily a, member 4(Smarca4)
ALDR	Aldo-keto reductase family 1 member B(Akr1b1)
AATF	Apoptosis antagonizing transcription factor(Aatf)
CSF1R	Colony stimulating factor 1 receptor(Csf1r)
NCPR	Cytochrome p450 oxidoreductase(Por)
DDB1	Damage-specific DNA binding protein 1(Ddb1)
HS71B	Heat shock 70kD protein 1A(Hspa1a)
HXK1	Hexokinase 1(Hk1)
ILK	Integrin-linked kinase(Ilk)
STAT3	Signal transducer and activator of transcription 3(Stat3)

Table 2.

HS-exosome proteins involving DNA repair

UNIPROT ID	Name of protein
PDS5B	PDS5 cohesin associated factor B(Pds5b)
RAD50	RAD50 double strand break repair protein(Rad50)
DDB1	Damage-specific DNA binding protein 1(Ddb1)
HS71A	Heat shock 70kD protein 1A(Hspa1a)
SMC1A	Structural maintenance of chromosomes 1A(Smc1a)
SMC3	Structural maintenance of chromosomes 3(Smc3)
SSRP1	Structure specific recognition protein 1(Ssrp1)

HS-derived EVs/exosomes exhibited a marked neuroprotection against oxidative stress and Aβ-induced neurotoxicity

To determine whether HS derived EVs and exosomes play a differential role in protection of neuronal cells under conditions of stress, such as oxidative stress, we co-treated a neuronal culture with 250 μM of H₂O₂ and 2 μg/ml of NHS- or HS-derived crude exosomes, or EVs. Previous data has indicated that H₂O₂ induces neuronal apoptosis [21]. As shown in Fig. 3A, even crude HS-derived EVs could significantly suppress H₂O₂ induced cell death. It is worth noting that NHS-derived EVs were also able to suppress H₂O₂ induced cell death with a lower efficacy. Additionally, we also tested the effect of the purified HS- and NHS-derived exosomes on Aβ-induced neurotoxicity due to studies highlighting the role of Aβ in inducing neuronal apoptosis [22]. In the absence of NHS- or HS-derived exosomes, Aβ caused a marked decrease in neuronal viability (Fig. 3B). Both NHS- and HS-derived exosomes were also able to reverse Aβ induced neurotoxicity, although HS-derived exosomes exhibited the potential to confer greater neuroprotection against Aβ-induced neurotoxicity compared to NHS-derived exosomes (Fig. 3B).

Fig. 3. Treatment of neuronal cells with HS-induced EVs/exosomes confers greater neuroprotection against oxidative stress and Aβ-induced neurotoxicity compared to NHS-derived EVs/exosomes.

A. Neuronal cells were treated with 250 μM H₂O₂ in presence of 2 μg/ml of NHS or HS-crude exosomes. After 8 h following the treatments, cells were harvested for viability analysis. Data are presented as Mean ± SD; n = 3-6. B. Neuronal cell cultures were treated with indicated conditions for 24 h before cell viability was assessed. Data are presented as Mean ± SEM; n = 3. *p<0.05; ***p < 0.001; **** p < 0.0001 (one-way ANOVA and Tukey’s post hoc test).

As mentioned above, the HS-derived EVs contained an increased amount of anti-apoptosis proteins. We next examined the effect of the HS-derived EVs in Aβ-induced apoptosis. Following 24 h of incubation, Aβ caused apparent apoptotic cell death, while NHS- or HS-derived EVs reversed the effect (Figs. 4A, 4B), despite no significant difference between NHS- and HS-derived EVs (Fig. 4B). Importantly, Both NHS- and HS-derived EVs also significantly reduced Aβ caused oxidative stress (Fig. 4C), suggesting that HS-derived EVs could be a therapeutic agent for AD and other neurodegenerative diseases.

Fig. 4. Treatment of neuronal cells with NHS- or HS-induced EVs/exosomes reverses Aβ-induced apoptosis and oxidative stress.

Neuronal cell cultures were treated with indicated conditions for 24 h before apoptosis (Figs. 4A, 4B) or ROS (Fig. 4C) was assessed. Data are presented as Mean ± SEM; n = 3. ***p < 0.001; **** p < 0.0001 (one-way ANOVA and Tukey’s post hoc test).

Discussion

Owing to their favorable characteristics of being able to cross the BBB and reflect the content of their donor cells, exosomes derived from neural stem cells are a promising therapeutic for AD. However, previous studies have highlighted the low yield of exosomes following isolation. Here, we demonstrated that exosomes isolated from cells under HS conditions exhibit a significant increase in diameter and concentration in comparison to exosomes isolated from cells under physiological conditions. Previous studies have shown that exosomes isolated from HS conditions increased the quantity of exosomes in comparison to NHS conditions in B-lymphoblastoid cells [11]. Of note, this study utilized the bicinchoninic acid assay to determine the change in exosomal proteins following HS. Our study utilized NTA, a well-accepted approach for exosome characterization, to determine the specific concentration and size of exosomes in solution; however, it remains unclear how HS alters the size and quantity of exosomes, despite previous studies suggesting that heat shock proteins may play a role in modulation of exosome biogenesis [23,24].

Although HS causes the neural stem cells to secret dramatically more exosomes, the protein diversity of HS-derived exosomes was much lower than NHS-derived exosomes. However, our bioinformatics data indicated that these fewer diverse HS exosomal proteins were enriched in anti-apoptotic and DNA repair proteins, suggesting that these samples may provide therapeutic efficacy. Importantly, our cell culture results validated the role of the HS-derived EVs and exosomes in providing marked neuroprotection against H₂O₂-induced oxidative stress and Aβ-caused neurotoxicity, both of which have been reported to trigger neuronal cell death via apoptotic manner [21,22]. One potential mechanism that could explain the ability of NHS- and HS-derived EVs to protect against oxidative stress and Aβ-caused neurotoxicity could be that these vesicles contain neuroprotective molecules, such as numerous chaperon proteins, or even organelles, such as mitochondria [25]. Following addition of either NHS- or HS-derived EVs alone into cell cultures, we observed a significant increase of ATP levels at 8 h after the incubation. This increase may be caused mainly by the mitochondria harbored in the EVs. Given the fact that apoptosis plays a key role in the progression of AD and several other neurologic disorders as revealed by studies on animal models and cell lines [26], HS-derived EVs/exosomes from neural stem cell and possibly other stem cell cultures could be an invaluable therapeutic reagent to treat these diseases. However, here we did not include any further mechanistic and animal-related studies. Moreover, in our experiments, the EVs/exosomes were given directly to the cells at the same time and same place as H₂O₂ or Aβ. In this condition, it was possible that the EVs/exosomes could directly interact with H₂O₂ or Aβ, leading to reduced toxicity, rather than the EV cargos translocated into the cells to confer protection. In the future, it is necessary to test whether treating the cells at a different time from the H₂O₂ or Aβ still confers neuroprotection. These may represent the limitations of the work.

In conclusion, this study shows promising results in utilizing HS-derived EVs/exosomes as a therapeutic agent to treat oxidative stress- and Aβ-induced neurotoxicity, due to their ability to significantly enhance the production of exosomes, enriched anti-apoptotic components, and ability to cause marked neuroprotection against oxidative stress and Aβ caused neurotoxicity and apoptosis. These results provide useful information for the consideration of HS-derived EVs/exosomes from other stem cells, which would allow investigators to overcome the low yield of EVs/exosomes seen in previous studies. HS-derived EVs/exosomes from other stem cells could be a therapeutic agent to treat AD and possibly other neurological diseases. Future studies are necessary to elucidate the mechanism underlying the increase in concentration, size and composition of HS-derived EVs/exosomes.