Background
Alzheimer's disease (AD) is a major form of senile dementia, a chronic progressive disease characterized by memory and cognitive impairment leading to death within 10 years of diagnosis [
1,
2]. It is estimated that the global prevalence of AD will increase from the current 50 million to 150 million by 2050 with the increase in the size of the elderly population [
3]. According to the amyloid cascade hypothesis first advocated by Hardy et al. [
4] in the 1990s, amyloid beta (Aβ) is known to be the main cause of AD. However, AD has been reported as a multifactorial disease involving highly phosphorylated tau protein neurofibrillary tangles and the
APOE gene [
5‐
7]. The AD brain, which has complex neuropathological characteristics, is permanently affected by various inflammatory substances, and the released cytokines create a neurotoxic environment that causes overall brain atrophy, leading to chronic diseases requiring long-term treatment [
8,
9]. However, most studies on AD drugs have so far aimed at single pathological targets such as Aβ. In 2021, aducanumab (marketed as “Aduhelm”) was approved as the first disease-modifying agent for treating AD by the Food and Drug Administration; however, it failed to significantly affect cognition in patients with severe form of the disease [
10,
11]. When used at a higher dose, a modest impact on the cognitive decline of patients was observed in the early stage of AD or in those with mild cognitive impairment; however, the drug did not reverse prior memory loss associated with AD [
10,
12].
Due to the multifactorial pathogenesis of AD, therapeutic strategies targeting multiple targets, rather than single-target therapies targeting molecules such as Aβ or tau, are emerging [
13,
14]. In this context, stem cell therapy targeting various pathological mechanisms in AD has emerged as a new approach [
15,
16]. The most commonly used stem cell types in recent AD treatment research are brain-derived neural stem cells (NSCs) [
17‐
20] and mesenchymal stem cells (MSCs) [
21‐
23]. Among them, MSCs are commonly used multipotent stem cells with self-renewing and immunomodulatory properties with limited differentiation capacity, mainly observed in mesodermal lineage cells [
20,
24]. The anti-amyloid efficacy of MSCs has been proven to repair damaged astrocytes when used as a cell therapy [
20,
25,
26]. Despite the reported advantages, various challenges limit the clinical use of MSCs, including the loss of potency and proliferative ability owing to their dedifferentiation and limited lifespan [
21,
27]. In addition, studies using existing MSCs have demonstrated an improved anti-amyloid environment, but the limitation of insufficient potency persists [
28,
29].
A disease-specific intranasal treatment strategy using stem cells exposed to the AD brain environment has been reported [
27,
30]. These studies inspired Alzheimer-specific treatment strategies, which, compared to the existing research directions, rely on the inherent homeostasis of existing stem cells. To select a more AD-specific candidate group, we selected cortical neural stem cells (CNSCs) located in the cerebrum, which perform memory-related and cognitive functions, as more competitive candidates for treatment than MSCs. CNSCs are located in the cortex that holds memory-related and cognitive functions of the brain, and studies suggest that they are more effective than other types of stem cells in treating AD [
31].
The aim of this study was to present a potential candidate group of cells as an AD-specific cell-free stem cell-based therapy, by evaluating the paracrine effect of induced pluripotent stem cell (iPSC)-derived cortical neural stem cell secretome (CNSC-SE) that is delivered via intranasal administration [
32,
33].
Methods
Mice
The Korea Excellence Animal Laboratory Facility of Korea Food and Drug Administration was accredited by the Institutional Animal Care and Use Committee (IACUC) and the Department of Laboratory Animal (DOLA) in the Catholic University of Korea, Songeui Campus in 2017, and acquired AAALAC (Association for Assessment and Accreditation of Laboratory Animal Care) complete international accreditation in 2018. All procedures of animal research were conducted in accordance with the Laboratory Animals Welfare Act, the Guide for the Care and Use of Laboratory Animals, and the Guidelines and Policies for Rodent Experiment provided by the IACUC School of Medicine at the Catholic University of Korea. The animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of School of Medicine, The Catholic University of Korea (Approval number: CUMS-2020–0217-08).
The B6.CG-Tg (APPSwFILon on PSEN1*M146L*L286V) 6799Vas/J transgenic (5×FAD) mice which have high amyloid deposition from 2 months of age [
34] and C57BL/6 (wild-type, WT) mice were purchased from Jackson Laboratory (Bar Harbor, MN). All mice were maintained under specific pathogen-free conditions. A previous study by Santamaria et al. reported that a single intranasal dose of MSC secretome resulted in sustained memory recovery in AD mice for 7 days [
27]. Based on this, we administered iPSC-derived CNSC-SE (5 μg/g) in 12-week-old male mice once a week for 4 weeks (4 times in total) by inserting a 26-G intravenous catheter (Poly Medicure POLYPEN IV Catheter, 26 G/19 mm) along the nasal cavity.
The following four groups were set in this study: (i) WT mice as controls (WT group); (ii) 5×FAD AD mouse model (AD group); (iii) CNSC-SE-treated 5×FAD mice (CNSC-SE-5×FAD group); and (iv) MSC-treated 5×FAD mice (MSC-SE-5×FAD group).
Barnes maze
The Barnes maze was used to assess spatial learning and memory [
35]. The maze consisted of an elevated circular platform (92 cm in diameter), containing 18 equally spaced holes around the circumference, each 5 cm in diameter. Visual cues were placed around the walls (periphery) during the test. For training, a mouse was placed in the center of the maze under a transparent beaker for 1 min, and it was then guided to the target hole, which contained sweet-flavored chips, for 1 min. The mouse was trained for 2 consecutive days followed by testing on day 3. On the test day, the mouse was placed in the center of the maze and observed for 3 min to record the following parameters: (i) distances from the target hole; (ii) the resting time in the target hole; (iii) the first latency to reach the target hole; and (iv) the error rate, i.e., the number of times the mouse entered a zone other than the target zone. After all trials, the maze was completely cleaned with 70% alcohol solution. All experimental tests were recorded by a video camera and analyzed with SMART 3.0 (Panlab, Harvard Apparatus, Barcelona, Spain).
Cell culture
All hiPSC lines used in this study were generated through our previous research using cord blood cells isolated from a healthy individual [
36]. Human iPSCs were seeded on cell culture dishes coated with vitronectin (Gibco, Waltham, MA; #A14700), and cultured in Essential 8 media (Gibco #A1517001) in a 10% CO
2 environment at 37 °C.
To induce differentiation of iPSCs into cortical neurons, we referred to the protocol of Shi et al. [
37]. For neural induction, iPSCs were seeded on vitronectin-coated 60-mm dishes at
\(4.6\times {10}^{6}\) cells/dish with Essential 8 media and 10 µM rho-associated kinase inhibitor for 24 h. After 24 h of incubation, the culture medium was changed to a neuronal differentiation medium (NDM), which is a 1:1 mixture of N-2 (consisting of Dulbecco’s Modified Eagle Medium, Nutrient Mixture F12 GlutaMAX, 1 × N-2, 5 μg/ml insulin, 1 mM
L-glutamine, 100 μm nonessential amino acids, 100 μM 2-mercaptoethanol, and 10% penicillin/streptomycin) and B-27 (consisting of Neural basal, 1 × B-27, 200 mM
L-glutamine and 10% penicillin/streptomycin, with 10 µM SB431542 [Tocris Bioscience, Bristol, UK; #1614]) with 1 µM Dorsomorphin (Tocris Bioscience, #BML-275), and incubated for 10 days. Neuronal progenitor cells were collected by Gentle Cell Dissociation Reagent (STEMCELL Technologies, Vancouver, BC, Canada; #100-0485) for NSC expansion. The cells were seeded on a laminin-coated (Biolamina, Stockholm, Sweden; #LN521-05) 60-mm dish (9 × 10
6 cells per dish) and cultured with NDM for 6 days, followed by maintenance in NDM supplemented with 20 ng/ml FGF2 for an additional 2 days. After 2 days, the NSCs were gently dissociated with accutase (Innovative Cell Technologies, San Diego, CA; #AT-104) for CNSC expansion and seeded on laminin-coated (Biolamina, #LN111) dishes under the same condition as for NSCs for 6 days. CNSCs, called cortical neurons, were cultured for 5 days and were passaged once more with accutase under the same conditions and maintained for up to 90 days.
Bone marrow-derived human MSCs were purchased from the Catholic Institute of Cell Therapy, South Korea. Human MSCs were maintained in a 10% CO2 environment at 37 °C and were cultured in DMEM (Gibco) with 20% FBS (Gibco) and 10% penicillin (Gibco).
Secretome preparation
After 10 days of neuroepithelial differentiation from iPSCs, 8 days of NSC differentiation, and subsequent CNSC differentiation by the previously described cell culture method, the culture medium was changed every 48 h, and the culture supernatants of the last 6 days were used as the material for CNSC-SE. For MSC-SE, the same conditions as that for CNSC-SE were used; cells were seeded and culture supernatant was collected after 3 days of culture. The collected culture supernatants were filtered through a 0.2-μm filter to remove cell debris. The filtered supernatants were stored at − 80 °C for 3 days, and the frozen samples were lyophilized for 7 days. The solid samples were stored at − 20 °C.
Immunocytochemical analysis
iPSCs and differentiated cells (iPSC-derived CNSCs) used for in-vitro experiments were fixed with 4% paraformaldehyde (PFA), permeabilized with 0.1% Triton X-100, and blocked with 2% bovine serum albumin (BSA) for 40 min, before being incubated overnight with primary antibodies. The primary antibodies used are given in Table
1. Secondary antibodies were then treated for 1 h. DAPI was used for nuclear staining. Confocal microscopy was performed on LSM900 w/Airyscan (Carl Zeiss, Oberkochen, Baden-Württemberg, Germany). Images represent confocal Z-stack taken with identical laser and detection settings.
Table 1
List of antibodies used for immunocytochemistry
BDNF | Abcam | ab108319 |
GDNF | Santa cruz | Sc-13147 |
VEGF | Santa cruz | Sc-7269 |
MAP-2 | Santa cruz | Sc-74421 |
NEUN | Abcam | Ab104224 |
vGLUT | Synaptic system | 135,303 |
Tuj1 | Gentex | GTX631836 |
FOXG1 | Abcam | Ab29359 |
Tbr1 | Abcam | Ab31940 |
SATB2 | Abcam | Ab51502 |
CUX1 | Santa Cruz | Sc-514008 |
GFAP | Abcam | Ab7260 |
anti-Iba1 | Fujifilm | #019–19741 |
6E10 | BioLegend | 803,001 |
Multielectrode array recording
The iPSC-derived cortical neurons were treated with 130 μg/ml CNSC-SE for 20 days. After treatment, \(1.3\times {10}^{7}\) cells were seeded on a poly-L-ornithine/laminin-coated CytoView 24-well plate (Axion Biosystems, Atlanta, GA; #M384-tMEA-24W), and iPSCs and non-treated (0 μg/ml group) iPSC-derived cortical neurons were used as controls.
The AxIS software was used to measure the spontaneous potential activity of cortical neurons. Cells were seeded at a density of 1.3 × 107 per well in a CytoView 24-well plate (Axion Biosystems, #M384-tMEA-24W) and cultured in 5% CO2 at 37 °C. The spontaneous activity was recorded for 15 min at the same time every day, and temperatures were maintained during all experiments.
Real-time polymerase chain reaction (PCR)
RNA was extracted with Trizol reagent (Invitrogen, Waltham, MA). The RevertAid First strand cDNA Synthesis kit (Invitrogen, #K1622) was used for cDNA synthesis. Real-time PCR was performed with a PowerSYBR Green PCR Master Mix (Applied Biosystems, Waltham, MA; #436759) using the Real-Time PCR System (Applied Biosystems). The primer sequences are shown in Table
2.
BDNF | F | CGGAAGGACCTATGTTTGCT | 106 |
R | TATTTCAGAACGCGCAACTG |
GDNF | F | AGCTGCCAACCCAGAGAAT T | 87 |
R | AAATGTATTGCAGTTAAGACACAACC |
VEGF | F | CATTGGAGCCTTGCCTTG | 87 |
R | ATGATTCTGCCCTCCTCCTT |
NEUN | F | GCGGCTAACGTCTCCAACAT | 188 |
R | ATCGTCCCATTCAGCTTCTCCC |
vGlut | F | CCATGACTAAGCACAAGACTC | 81 |
R | AGATGACACCTCCATAGTGC |
MAP2 | F | GGAGACAGAGATGAGAATTCC | 82 |
R | GAATTGGCTCTGACCTGGT |
Nestin | F | ACCAAGAGACATTCAGACTCC | 303 |
R | CCTCATCCTCATTTTCCACTCC |
PAX6 | F | GTGTCCAACGGATGTGGAG | 254 |
R | CTAGCCAGGTTGCGAAAGAAC |
Tbr1 | F | GGGCTCACTGGATGCGCCAAG | 157 |
R | TCCGTGCCGTCCTCGTTCACT |
TUJ1 | F | GGCCTTTGGACATCTCTTCA | 241 |
R | ATACTCCTCACGCACCTTGC |
FOXG1 | F | AGGAGGGCGAGAAGAAGAAC | 213 |
R | TCACGAAGCACTTGTTGAGG |
OCT4 | F | ACCCCTGGTGCCGTGAA | 190 |
R | GGCTGAATACCTTCCCAAATA |
GAPDH | F | ACCCACTCCTCCACCTTTGA | 101 |
R | CTGTTGCTGTAGCCAAATTCGT |
mIGFBP-2 | F | CAGACGCTACGCTG-CTATCC | 142 |
R | CTCCCTCAGAGTGGTCGTCA |
mIGF-2 | F | TGGCCCTCCTGGAGACGTACTGTGC | 116 |
R | TTGGAAGAACTTGCCCACGGGGTATC |
mIGF-1r | F | CTACCTCCCTCTCTGGGAATG | 185 |
R | GCCCAACCTGCTGTTATTTCT |
mGAPDH | F | GCCAAACGGGTCATCATCTC | 377 |
R | GACACATTGGGGGTAGGAAC |
Western blot
The whole brain of each mouse was homogenized with a protein extraction reagent (Thermo Fisher Scientific, Waltham, MA) supplemented with 1 protease inhibitor cocktail tablet (Roche, Basel, Switzerland) and 1 mM phenylmethylsulfonyl fluoride. The homogenate was sonicated and centrifuged at 4 °C for 20 min at 16,000 rpm. Proteins were quantified using the Bradford dye (Bio-Rad, Hercules, CA) with BSA as a standard. Next, 30 μg of total proteins was separated by sodium dodecyl sulphate–polyacrylamide gel electrophoresis and transferred to a nitrocellulose membrane. The membrane was blocked with 3% BSA for 1 h at room temperature (RT) and incubated at 4 °C overnight with primary antibodies (Table
3). After washing with TBST, the membrane was incubated with the secondary antibody for 1 h at RT. Protein bands were visualized using enhanced chemiluminescence. Band intensities were normalized to β-actin.
Table 3
List of primary antibodies used for Western blotting
anti-APP | Millipore | A8717 | 1:1000 |
anti-pTau | Cell signaling | D4H7E | 1:500 |
BACE | Cell signaling | D10E5 | 1:1000 |
anti-Iba1 | Fujifilm | 019-19741 | 1:500 |
anti-TNFα | Abcam | Ab9739 | 1:500 |
anti-IL-1β | Abcam | Ab9722 | 1:500 |
GAPDH | Abcam | Ab9485 | 1:1000 |
Immunohistochemistry for Aβ plaque
Mice were intracardially perfused with saline, and their brains were collected and fixed in 4% PFA at 4 °C, followed by dehydration in 15% sucrose and then 30% sucrose. The brains were then embedded in OCT Compound (Sakura Finetek, Torrance, CA) and sectioned (9 μm) with a microtome (Thermo Fisher Scientific) onto gelatin-coated slides (Masterflex, Gelsenkirchen, Germany; #HV-75955-51). These brain section slides were stored at − 80 °C.
For immunohistochemistry, the slides were dried overnight at RT and fixed with cold acetone for 10 min. Endogenous peroxidase activity was blocked by treating all slides with 0.3% H2O2. The slides were incubated for 1 h with 10% normal goat serum and thereafter overnight at 4 °C with anti-Aβ (1:500, Abcam, Cambridge, UK, #ab201060). After incubation with biotinylated secondary antibodies (1:200) for 40 min, the slides were incubated with the ABC HRP reagent (Vector Laboratories, Newark, CA; #PK-7100) for 10 min and DAB (Vector Laboratories) for 1 min. The slides were counterstained with Mayer’s hematoxylin.
Immunofluorescence analysis of mouse brain tissue
The brain sections were dried overnight at RT and fixed with cold acetone for 10 min. Endogenous peroxidase activity was blocked by treating the slides with 0.3% H2O2. Then the sections were blocked with Mouse on Mouse (M.O.M.) IgG blocking reagent (M.O.M. Fluorescein Kit, Vector Laboratories, #FMK-2201) diluted in 1% BSA for 5 min, followed by incubation overnight at 4 °C with 6E10 (1:1000, BioLegend, #803001). The next day, the slides were incubated with M.O.M. Biotinylated anti-mouse IgG and with the secondary antibody attached to fluorescein Avidin DCS.
For GFAP and IBa-1 immunofluorescence staining, the sections were incubated with anti-GFAP (1:1000, Abcam, #ab7260) or anti-IBa-1 (1:1000, Fujifilm, #019-19741) in 10% normal goat serum at 4 °C overnight. Then the sections were incubated with secondary antibodies such as goat anti-rabbit IgG Alexa 594 (1:200, Thermo Fisher Scienctific) for 1 h. Confocal microscopy was performed on LSM900 w/Airyscan, Carl Zeiss. Confocal Z-stack images were taken with identical laser and detection settings.
Cresyl violet staining
The brain sections were dried overnight at RT, fixed with cold acetone for 10 min, rehydrated in 70% and 95% alcohol for 2 min each, and stained with cresyl violet solution (Abcam, ab246816) for 1 min. Then they were washed in 70% and 95% alcohol for 2 min each, dehydrated in 100% ethanol for 30 s, cleared with xylene twice and mounted.
Metabolomics analysis for iPSC secretomes and CNSC-SE was performed using capillary electrophoresis mass spectrometry (Human Metabolome Technologies, Inc, Yamagata, Japan). Samples were prepared to allow for metabolite extraction. To perform metabolomics analysis, we used three samples of iPSC secretomes and CNSC-SE each, and Essential 8 media and neural basal media were used as a control.
Human cytokine array
Cytokines in CNSC-SE were assayed using a human XL cytokine array kit (ARY022B, R&D Systems, Minneapolis, MN) according to the manufacturer's protocol. The membranes of the kit were blocked at RT for 1 h, and then incubated with culture medium overnight at 4 °C. After 3 days of culture, the medium was debris removed with a 0.22 um syringe filter and used for experiments. The detection antibody cocktail diluted in the assay buffer was added and incubated at 25 °C for 1 h, and 1 × Streptavidin-HRP was added and incubated at RT for 30 min. All experiments were performed on a shaker. Images were acquired using a bio-image analysis system (Amersham Imager 600), and quantitative evaluation was performed using the Image J software.
Growth factor array
The human growth factor antibody array (ab134002; Abcam) was performed according to the manufacturer’s protocol. The lyophilized CNSC-SE and MSC-SE samples used in the experiments were diluted to 0.1 g/ml in blocking buffer. Membranes blocked at RT for 1 h were incubated with samples overnight at 4 °C. The biotin-conjugated anti-cytokines cocktail, diluted 1000 × , was added and incubated for 2 h. After washing, HRP-Conjugated Streptavidin was added and incubated for 1 h at RT. The membrane was then exposed using a bio-image analysis system (Amersham Imager 600).
Statistical analysis
All experiments were repeated at least three times, and results are expressed as the mean and standard error of the mean (shown as error bars). Statistical analysis was performed using GraphPad Prism 9.0. Comparison of multiple groups was performed by one-way analysis of variance (ANOVA) followed by post-hoc Dunnett’s multiple comparison test. A t-test was used to analyze nonparametric quantitative datasets, and one-tailed P values were calculated. Differences between groups were examined for statistical significance using Welch’s t-test, which was also used for analyzing metabolomics. Kruskal–Wallis and Mann–Whitney analyses were performed for intergroup comparison. Statistical significance was set at P < 0.05.
Discussion
AD is a progressive neurodegenerative disease characterized by memory loss and cognitive impairment, caused by synaptic disorders and excessive accumulation of incorrectly folded proteins [
4,
34]. To date, almost all advanced clinical trials on specific AD-related pathways have failed due to the loss of large numbers of neurons in the brains of most AD patients [
4,
13]. Stem cell-based treatments have emerged as new treatment strategies for various neurodegenerative diseases due to their self-renewability, versatility, and the ability to differentiate into major cell phenotypes in the CNS [
15]. Recent preclinical studies of stem cell therapy for AD have proven promising; however, many hurdles exist for stem cell therapy to become a clinically feasible treatment for human AD and related diseases [
3,
16].
Therefore, based on the report that the anti-amyloid efficacy of MSCs in AD is due to the side secretion effect [
16], the secretome was separated from the iPSC-derived CNSC and used as a new AD-specific treatment candidate for this study. In addition, an intranasal administration method was used to efficiently deliver it to the brain in a non-invasive manner in consideration of elderly patients with AD.
Our main findings are as follows: (1) iPSC-derived cortical neurons facilitated neuroelectromagnetic signaling by increasing cortical neural network development and neuron maturation in vitro. (2) CNSC-SE treatment improved the memory of 5×FAD mice. (3) The number of Aβ plaques was decreased in the CNSC-SE-treated mice. This is thought to be potentially caused by a decrease in β-secretase, an enzyme that produces Aβ. (4) IGFBP-2 is the possible candidate responsible for the efficacy of iPSC-derived CNSC-SE in AD. (5) We also established a protocol for efficient delivery using intranasal administration via the olfactory nerve pathway without damaging the brain.
In this study, we successfully differentiated iPSCs into cortical neurons based on the previously reported protocol [
37] and isolated CNSC-SE from the cell culture medium. In the process of cortical nerve differentiation of iPSCs, it was confirmed that nerve cells exposed to CNSC-SE showed dose-dependent increase of neural marker expression compared to those that were not exposed (Fig.
1f). Furthermore, electrical network activity, which is a major feature of neurodevelopment, was present in cortical neurons treated with CNSC-SE from day 55, confirming that CNSC-SE had a potential impact on electrical network development through increasing spike area and density (Fig.
2e).
One of the ultimate goals of AD treatment is memory recovery. Here, CNSC-SE was efficiently delivered to the brain of 5×FAD mice through a non-invasive route, and Barnes maze test was conducted to evaluate neuronal development, cognitive impairment, and spatial learning recovery. The 5×FAD mice treated with CNSC-SE showed memory improvement compared with those without CNSC-SE treatment (Fig.
3). In addition, histological analysis of brain samples showed that CNSC-SE treatment caused plaque reduction. The expression of neuroinflammatory factors and BACE (a β-secretase that promotes Aβ cleavage) was also decreased (Fig.
4). Compared to MSC-SE, CNSC-SE showed better efficiency in reducing APP and BACE. Taken together, the isolated CNSC-SE had similar effects to that of the MSC-SE, with several additional beneficial effects.
We also demonstrated that CNSC-SE reduced the inflammatory neuroimmunocyte phenotype by significantly reducing TNFα and IL-1β expression (Fig.
4b). In addition, the expression of inflammatory cytokines in the CNSC-SE group was significantly lower than that in the MSC-SE group. The AD brain is hyperinflammatory due to abnormally deposited plaques, and most neuroimmune cells are activated; however, CNSC-SE treatment resulted in local activation of neuroinflammatory cells only around the plaques. The anti-inflammatory effects of CNSC-SE in the AD brain affected microglial activation Excessive neuroinflammatory reactions in AD promote synaptic loss and cognitive deficiency, which are correlated with active microglial cells confirmed by the morphological changes of Iba-1-positive cells (Fig.
6). This was confirmed by the Western blots of Iba-1 (Fig.
4a, b). However, CNSC-SE treatment reduced both GFAP- and Iba-1-positive cells in the whole brain and in the cortex (Figs.
5c,
6c). While many studies on microglial activation mostly relied on Iba-1 staining to characterize its morphology, quantity, and distribution, it was previously reported that Iba-1 expression (protein and mRNA) might not reflect microglial activation and this still remains a debatable issue [
57‐
59]. Increased Iba-1 in activated microglia has been reported [
59,
60], but several studies found that the activation of microglia in brain tissue is not always accompanied by increased expression of Iba-1 [
57,
59,
61]. It is suggested that Iba-1 can only identify microglia and its expression may not relate to microglial activation in the brain tissue [
62,
63]. Although Iba-1 has been reported as a microglia/macrophage specific marker [
64,
65], most studies have relied on immunostaining to determine the extent of microglial activation by checking its morphology and distribution [
57‐
59]. It is still debatable whether increased protein expression of Iba-1 can be used as a sensitive indicator of microglial activation [
66]. Therefore, further confirmation with additional markers such as CD11b or ICAM-1 might be necessary and the same might be applicable in the case of astrocytes as well.
Our human cytokine and growth factor arrays showed that IGFBP-2 was present at a higher level in the CNSC-SE than in the iPSC-SE. IGFBP-2 is the most abundant type of IGFBPs in the cerebrospinal fluid, the developing brain and the hippocampus and cortex [
43]. Also, IGFBP-2 is a pleiotropic polypeptide that functions as a autocrine and/or paracrine growth factor [
43]. Mice with depressive-like behavior caused by chronic immobilization stress showed decreased IGFBP-2 expression in the central amygdala [
42]. Also, prenatal stress resulted in reduced IGFBP-2 expression in the hippocampus and frontal cortex in adult male rats [
67]. Previous studies also suggest that IGFBP-2 may enhance regenerative sprouting and contribute to neuronal repair in a rat model of sensory spinal axonal injury [
68].
IGF-2 is involved in memory enhancement [
41]. While both IGF-1 and IGF-2 can be regulated by IGFBP-2, only IGF-2 expression correlates with the expression of IGFBP-2 throughout the CNS development, suggesting that IGFBP-2 may regulate the function of IGF-2 in the CNS [
69]. When relatively higher IGFBP-2 concentrations were administered with IGF-2, increases in the percentage of neurite-bearing cells and the average neurite length were confirmed [
42]. The ability of IGFBP-2 to improve neuronal survival is related to its role in apoptosis inhibition [
43]. The expression of Bcl-2 and IGF/IGFBPs was localized at the same site in the hippocampus, suggesting that the IGF/IGFBPs system and the pro-survival proteins protect the cells from apoptosis and play a critical role during brain development [
70]. IGFBP-2 is also significantly increased around the injury site [
71].
On the other hand, a possible link between increased IGFBP-2 and mitochondrial dysfunction in AD has been suggested [
72]. Blood protein analysis showed increased IGFBP-2 levels in serum before the onset of clinical AD features [
73]. Elevated plasma IGFBP-2 levels are suggested to be linked with lower hippocampal volumes [
74,
75]. Interestingly, smaller hippocampal volumes are associated with higher IGFBP-2 levels only in the amyloid-negative individuals [
74]. IGFBP-2 may differentially modulate normal physiological and pathological functions [
76]. IGFBP-2 can be found in extracellular matrix, plasma, and nucleus [
77]. Circulating plasma is suggested as a novel biomarker for various brain diseases [
75]. Therefore, the detection of plasma IGFBP-2 in AD mice with or without CNSC-SE treatment might be useful for further analysis of this study. Also, the multiple functional domains of IGFBP-2 are thought to contribute to the spatial regulation of IGFBP-2 tumor biology, inducing different regulatory mechanisms operating in the extracellular, intracellular, and nuclear environments [
77]. Therefore, further investigations of IGFBP-2 high expression in AD and high presence in the CNSC-SE might be interesting.
Metabolomics represents the final results of interactions between genes, RNAs, and proteins, therefore it has several advantages over other analysis [
46,
78]. Comparison of normalized semi-quantitative metabolite profile of CNSC-SE versus neural basal medium using principal component analysis showed a clear distinction, which indicates that CNSC-SE and neural basal medium are biochemically distinct (Fig.
8b). In addition, 47 (36.2%) of the 130 metabolites were present only in CNSC-SE, including products with neuroprotective effects, such as GSH [
49,
50] and creatine [
51‐
53]. As such, it can be confirmed that the CNSC-SE contains several components that exert neuroprotective functions or amplify the effect of cognitive function recovery. Furthermore, we report that CNSC-SE contains various components with neuroprotective, anti-amyloid, memory-restoring effects, and excessive neuroinflammatory response-reducing effects.
In conclusion, we found that the iPSC-derived CNSC-SE enhanced neurorestorative activity of cells than MSC-SE with conventional anti-amyloid effects. The efficacy of CNSC-SE was confirmed both in vitro and in vivo, and the memory restoration, which is an important point in the treatment of AD, was significantly enhanced by CNSC-SE. Although these results clearly demonstrate the paracrine action of CNSC-SE, further studies are needed to conclude the process by which CNSC-SE exerts its therapeutic effects on AD. Further investigation of the components of CNSC-SE identified through metabolomics analysis is also warranted.