Cell Therapies for Autism Spectrum Disorder Based on New Pathophysiology: A Review

Abstract

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder characterized by difficulties in social communication, repetitive behaviors, and restricted interests, with onset early in life. The prevalence of ASD has increased worldwide in the last two decades. However, there is currently no effective therapy for ASD. Therefore, it is important to develop new strategies for ASD treatment. Evidence for the relationship between ASD and neuroinflammation, ASD and microglia, and ASD and glucose metabolism has increased rapidly in recent decades. We reviewed 10 clinical studies on cell therapies for individuals with ASD. Almost all studies showed good outcomes and no remarkable adverse events. Over the past decades, the neurophysiological characteristics of ASD have been shown to be impaired communication, cognition, perception, motor skills, executive function, theory of mind, and control of emotions. Recent studies have focused on the roles of immune pathology, such as neuroinflammation, microglia, cytokines, and oxidative stress, in ASD. We also focused on glucose metabolism in patients with ASD. The significance of gap junction–mediated cell–cell interactions between the cerebral endothelium and transplanted cells was observed in both bone marrow mononuclear cells and mesenchymal stromal cells transplantation. Owing to the insufficient number of samples, cell therapies, such as umbilical cord blood cells, bone marrow mononuclear cells, and mesenchymal stromal cells, will be a major challenge for ASD. As a result of these findings, a new paradigm for cell therapy for autism may emerge.

Introduction

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder characterized by difficulties in social communication, repetitive behaviors, and restricted interests, with an onset early in life. To meet the diagnostic criteria for ASD according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), a child must have persistent deficits in each of the three areas of social communication and interaction, and at least two of the four types of restricted, repetitive behaviors¹ (Table 1). The prevalence of ASD worldwide has been increasing over the last two decades with a universal screening of 91%; ASD prevalence has been documented at 2.2%^2,3.

Table 1.

Diagnostic or Evaluation Tools for ASD.

Behavior	DSM-5	Persistent deficits in social communication and social interaction across multiple contexts, as manifested by the following, currently or by history	Deficits in social-emotional reciprocity, ranging, for example, from abnormal social approach and failure of normal back-and-forth conversation; to reduced sharing of interests, emotions, or affect; to failure to initiate or respond to social interactions.
			Deficits in nonverbal communicative behaviors used for social interaction, ranging, for example, from poorly integrated verbal and nonverbal communication; to abnormalities in eye contact and body language or deficits in understanding and use of gestures; to a total lack of facial expressions and nonverbal communication.
			Deficits in developing, maintaining, and understanding relationships, ranging, for example, from difficulties adjusting behavior to suit various social contexts; to difficulties in sharing imaginative play or in making friends; to absence of interest in peers.
Cognitive and behavior	ADOS (Autism Diagnostic Observation Schedule)	Semi-structure standardized assessment	Social interaction
			Play
			Imaginative use of materials
Behavior	ADI-R (Autism Diagnostic Interview–Revised)	Interview from parents	Reciprocal social interactions
			Communication
			Language
			Restricted and repetitive Stereotyped interests
			Restricted and repetitive stereotyped behaviors
Behavior	CARS (Childhood Autism Rating Scale–Second Edition)		Observations on different areas of behavior
Cognitive	WISC-4 (Intelligent Quotient)	Investigation to use instruments on one to one	VCI (Verbal Comprehension Index)
			PRI (Perceptual Reasoning Index)
			WMI (Working Memory Index)
			PSI (Processing Speed Index)
Cognitive	Eye-tracking test	Investigation to individual’s gaze and eye movements	Socially relevant information and face processing
			Smaller fraction of time looking at the nose and eyes than controls
			Theory of mind
Social interactions	Vineland-II (Vineland Adaptive Behavior Scales–Second Edition)	Two survey forms, the survey interview form and the parent/caregiver rating form	Communication
			Daily living skills
			Socialization
			Motor skills
Social responsiveness	SRS-2 (Social Responsiveness Scale)	The presence and severity of social impairment	Total score for severity of social deficits
			Five treatment subscale scores
			Two DSM-5-compatible subscale scores
Social communication	ScQ (Social Communication Questionnaire)	A quick, easy, and inexpensive way to routinely screen for ASD	Total score with cutoff points
			Patient Questionnaire with 40 yes-or-no items. Current and Lifetime Forms

ASD: autism spectrum disorder; DSM: Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition; WISC-4: Wechsler Intelligence Scale for Children–Fifth Edition.

We have previously published about review articles on cell therapies for individuals with ASD. In this article, we focus more precisely on mechanisms of cell therapy for ASD⁴. Fig. 1 shows the current pathophysiology and treatment of ASD. The main cognitive and behavioral characteristics of ASD are impairment of communication, cognition, perception, motor skills, executive function, theory of mind, and control of emotions^5–9. Genetic factors, environmental factors, and hormone/neurotransmitter abnormalities likely predispose to ASD. Various behavioral therapies based on neurophysiological viewpoints have been considered the foremost strategies for the management of individuals with ASD¹⁰.

Figure 1.

Pathophysiology and treatment of ASD. The main cognitive and behavioral characteristics of ASD are impairment of communication, cognition, perception, motor skills, executive function, theory of mind, and emotion control,^3–9 and various behavioral therapies based on psychological viewpoints have been considered the foremost strategies for management of individuals with ASD¹⁰. Genetic factors, environmental factors, and hormones/neurotransmitters abnormality are likely predisposing to ASD. Inherited and de novo genetic risk factors such as genome-wide changes splicing and regional gene expression patterns in long noncoding RNAs (lncRNAs)^11–13, environmental factors such as air pollution and environmental toxins^14,15, and hormones/neurotransmitters abnormality¹⁶ are likely predisposing factors to ASD. Many medications targeting hormones and neurotransmitters are being proposed as potential candidates for new therapies. Yet, there are currently no apparent effective therapies for ASD¹⁷. Recent studies focused on the roles of immune pathology such as neuroinflammation, impairment of microglia, and glucose metabolism in ASD. New treatment strategy with stem cell therapy includes UCBs, UC-MSCs, BM-MSCs, and BM-MNCs for individuals with ASD. ASD: autism spectrum disorder; BM-MNC: bone marrow mononuclear cell; MSC: mesenchymal stromal cell; UC-MSC: umbilical cord–derived MSC; UCB: umbilical cord blood cell.

Many medications that target hormones and neurotransmitters, such as oxytocin, dopamine, serotonin, glutamate, GABA, and noradrenaline, have been proposed as potential candidates for new therapies^18–27. However, there are currently no effective therapies for ASD¹⁷. Recent studies have focused on the roles of immune pathology, such as neuroinflammation, microglia, cytokine or oxidative stress, and glucose metabolism, in ASD. A new treatment strategy for individuals with ASD includes “stem cell therapy” (Fig. 1). This review focuses on the recently proposed pathophysiology and various cell therapies for ASD with regard to neuroinflammation, microglia, and glucose metabolism.

Pathophysiology

“In the Viewpoint of Neuroinflammation, Microglia, and Glucose Metabolism”

Abnormal activation of microglia by activation of the TLR4 signaling pathway following maternal lipopolysaccharide exposure, which in turn is involved in excessive synaptic pruning to decrease synaptic plasticity in the offspring may be one of the reasons for the autism-like behavior in offspring mice²⁸. However, early postnatal allergic airway inflammation has been reported to induce dystrophic microglia that exhibit defective synaptic pruning upon short-term and long-term allergen exposure, resulting in excitatory postsynaptic surplus and ASD-like behavior²⁹. Post-mortem studies have further implicated neuroinflammation, microglial dysfunction, and oxidative stress in the pathogenesis of ASD³⁰. Voineagu et al. reported the overexpression of immune-related gene networks³¹. Braunschweig et al. reported the presence of maternal antibodies in fetal brain tissue³². Vargas et al. described atypical levels of pro-inflammatory cytokines (IL-6 and TNF-α) in the cerebrospinal fluid (CSF) of patients³³, and further reports suggested that excessive microglial activation leads to aberrant neural connections^34,35. These reports suggest that abnormal microglial activation, responsible for synaptic pruning, partially contributes to ASD pathology. Impairment of microglia affects neural development, possibly contributing to ASD³⁶. Therefore, modulation of microglial phenotype, immunomodulation, or repair of microglial function may be a novel therapeutic strategy for the treatment of neurological disorders accompanied by inflammation. Umbilical cord blood contains hematopoietic stem cells, endothelial progenitor cells, and mesenchymal stromal cells (MSCs) that alter brain connectivity and modulate inflammation^37,38. The infusion of autologous umbilical cord blood cells (UCBs) has been shown to be safe in individuals with cerebral palsy (CP), ASD, and other acquired brain injuries. Dawson et al. reported safe and feasible autologous UCBs infusion in young children with ASD, and several promising outcomes have been published^39–42. Recently, UC-MSCs (umbilical cord–derived MSC) and bone marrow mononuclear cells (BM-MNCs) from individuals with ASD had favorable outcomes^43–49 (Table 2). Fetal stem cells for individuals with ASD have shown favorable outcomes, albeit instigating an ethical dilemma regarding the use of fetal stem cells.⁴⁸

Table 2.

Recent Preclinical Review of Cell Therapy for Autism Spectrum Disorder.

	Animal	Source	Number of cells	Results	Route of administration	Dose	Pathology
Autism Research Segal-Gavish et al.50	BTBR Mice	MSC		A reduction of stereotypical behavior—a decrease in cognitive rigidity	Intracerebroventricular
Mol Autism Perets et al.51	Mouse of autism Shank3B; 8–10 weeks, male	Exosomes derived from MSC	107 particles/microl	Improve the social behavior deficit; increase vocalization; reduce repetitive behavior	Intranasal	4 dose (8days)	Increase of GABARB1 in the prefrontal cortex
ACS Appl Bio Mater Liang et al.52	VPA-treated mice	Exosomes from hUC-MSCs		Restore the social ability; correct the repeated stereotyped behaviors	Intranasal
Brain Res Perets et al.53	BTBR Mice; 6–8 weeks	MSC induced to secrete higher amounts of neutrophic factors	Improve communication skills	Reduce stereotypic behavior; improve cognitive flexibility; improve social behavior	1 dose		Brain Res.
Mol Autism Perets et al.54	BTBR Mice; 6–7 weeks, male	Exosomes derived from MSC vs MSC	3.81 × 108 particles/microl	Increase male to female interaction; reduce repetitive behaviors; improve maternal behaviors to pup retrieval	Intranasal/intravenous	1 dose
Ann N Y Acad Sci Zeng et al.55	Maternal diabetes mouse 6 weeks, male	Hematopoietic stem cell (HSC)	2 × 106 (total)	Ameliorates gastrointestinal symptoms ameliorates autism-like behavior	Intravenous	1 dose
Neuropsychopharmacology Donegan et al.56	The Poly I: C SD rat (gestational day 12) 40–45 days	J27 mouse embryonic stem cell line	4 × 104 (right and left hemisphere)	Alleviate deficits in social interaction; alleviate deficits in cognitive flexibility	The medial prefrontal cortex (mPFC)	1 dose	Decrease pyramidal cell firing in the mPFC
Behav Brain Res Ha et al.57	VPA-induced mouse; p2 or p3	Human adipose-derived stem cells (hASCs)		Ameliorates repetitive behavior; ameliorates social deficit and anxiety	Intraventriculary	1 dose
Stem Cell Res Ther Zhanget al.58	BTBR mouse; 2 months	Human amniotric epithelial cell (hAECs)	5 × 104 cells /microl	Ameliorates social deficits	Bilateral intraventriculary (1micro × 2)	1 dose	Decline neurogenesis and neuroprogenitor cell in the hippocampus by expanding the stem cell pool. Rescue the decrease levels of BDNF and TrkB in the hippocampus

MSC: mesenchymal stromal cell; VPA: Valproic acid; BTBR mouse: BTBR T⁺Itpr3^tf/J mouse; SD rat: Sprague-Dawley rat.

Clinical Presentation

Clinical manifestations such as speech delay, echolalia, lack of eye contact, and sleep disturbances are present in the toddler period. Difficulties in group activity, hyperactivity, motor disturbances, and learning delays are usually witnessed in elementary schools. Epilepsy, tic disorder, attention-deficit hyperactivity disorder, and self-injurious behaviors may be associated with ASD⁵⁹. Clinical manifestations such as anxiety, obsessive-convulsive behavior, and depression are usually present during junior high school, especially when children encounter difficult situations at home or school, wherein some last throughout their lifetime⁵⁹. The level of adaptation to social interactions and clinical manifestations are key factors affecting the quality of life of the child and family.

Assessment and Diagnosis

The criteria for diagnosing and evaluating ASD are shown in Table 1. The Autism Diagnostic Observation Schedule–2 (ADOS-2) and Autism Diagnostic Interview–Revised (ADI-R) are considered the “gold standard” for diagnosis⁶⁰. The ADOS consists of a semi-structured, standardized assessment of social interactions as well as the play and imaginative use of materials. The ADI-R is often used in combination with the ADOS and interview tests to assess reciprocal social interactions, communication, language, and restricted and repetitive stereotyped interests and behaviors. The ADOS-2 and ADI-R are diagnostic tools for cognitive and behavioral viewpoints. The Childhood Autism Rating Scale–Second Edition (CARS-2) is another diagnostic tool for ASD that aims to observe different areas of behavior. The Wechsler Intelligence Scale for Children–Fifth Edition (WISC-5) is another useful tool for assessing intelligence quotient (IQ) from the viewpoint of cognitive function. The eye-tracking test is a diagnostic tool for ASD to investigate individuals’ gaze and eye movements, clarifying socially relevant information, and relevant information of the theory of mind. These diagnostic tools are based on cognitive and behavioral viewpoints.

There are no definitive biomarkers, electroencephalography (EEG), magnetic resonance imaging (MRI), or other investigations to diagnose ASD. The evaluation tools of the level of adaptation to social community, such as the Vineland Adaptive Behavior Scales–Second Edition (Vineland-II), Social Responsiveness Scale (SRS-2), and the Social Communication Questionnaire (ScQ), are useful for obtaining information on coping with individuals with ASD (Table 1).

Methods

Definition of cell therapy in this article includes cell therapy using human stem cell or human cell–derived stem cell such as UC-MNC, UC-MSC, BM-MNC, BM-MSC, and fetal stem cell. We conducted a PubMed (MEDLINE) search for studies on neuroinflammation, microglia, or glucose metabolism that included the terms “ASD,” “every 3 years,” “in the last two decades,” in English. We conducted a PubMed (MEDLINE) search for clinical studies using cell therapy for ASD in English. Furthermore, we conducted a PubMed (MEDLINE) search for preclinical studies using cell therapy that included the terms “ASD” or “Autism,” in English.

Results

Fig. 2 shows that the relationship between ASD and neuroinflammation, ASD and microglia, and ASD and glucose metabolism has increased rapidly in the last two decades. The number of studies that include ASD and microglia totals over 150, that include ASD and neuroinflammation increase over 200, that include ASD and microglia totals over 150, and that include ASD and glucose metabolism amounts to over 50. In light of these results, cell therapies which focus on neuroinflammation, microglia, and glucose metabolism may serve as a new strategy for the treatment of ASD.

Figure 2.

Article numbers related to autism by PubMed Medline between 1998 and 2021. The relationship between ASD and neuroinflammation, ASD and microglia, and ASD and glucose metabolism has increased rapidly in the last two decades. ASD: autism spectrum disorder.

Source: “PubMed” Medline.

In a review of recent preclinical studies of various cell therapies for ASD (Table 2), MSC (n = 2), exosomes derived from MSC, hematopoietic stem cells (n = 1), embryonic stem cells (n = 1), human adipose-derived stem cells (n = 1), and human amniotic epithelial cell (n = 1) were studied. Based on these results, MSCs may be a promising candidate for cell therapy in the treatment of ASD.

Conversely, umbilical cord blood and bone marrow have become widely used as sources of cell therapies owing to their popularity as treatments for malignant leukemia over the last several decades. Despite the recent focus on umbilical cord stem cells for the treatment of various diseases, fetal stem cells are not widely used due to ethical issues. In a review of the clinical trials on ASD in the past two decades (Table 3), MSC (n = 2), UCB (n = 3), UC-MSC (n = 2), BM-MSC (n = 1), fetal stem cells (n = 1) and UCB + UC-MSC (n = 1) were used. Administration took place one (n = 5), two (n = 2), three (n = 2), or four times (n = 1). Variability was noted in the number of cells administered (1 × 10⁶ to 5 × 10⁷ cells/kg). UCB, fetal stem cells, and UC-MSC, BM-MNC, and BM-MSC were performed using intravenous (IV) and intrathecal (IT) routes, respectively. IV injection is easier compared with IT, wherein UC-MSC and UCB are dispersed to the central nervous system without being trapped in the lung and blood–brain barrier. UC-MSCs therapy for ASD is performed considering the neurotrophic effects of UC-MSCs, in addition to immunomodulation against chronic inflammation. A few severe adverse events were observed after transplantation (Table 2). Autologous BM-MNC transplantation is required to avoid rejection, but the BM-MNC procedure is invasive and carries its own risks, and administration of BM-MNC via IT injection in people with ASD may also pose risks due to IT access. Moreover, most reports suggest that UC-MSCs have therapeutic potential with relative safety, and allogeneic MSCs can be ordered at the most suitable critical timing. Almost all studies showed good outcomes and no notable adverse events. Cell therapies, such as UCBs, BM-MNCs, and MSCs, for ASD will be challenging because the number of samples is insufficient.

Table 3.

Recent Clinical Review of Cell Therapy for Autism Spectrum Disorder.

	Source	Number (age; range)	Route of administration	Number of cells	Results	Severe adverse events	Study design
Lv et al.43	UCB / UC-MSC	37 (3–12)	IV/IT+ IV	UCB: 2 × 106/kg UC-MSC: 1 × 106/kg	Improvement in the CARS, Clinical Global Impression (CGI) scale, and Aberrant Behavior Checklist (ABC)	No	Non-randomized, open-label, single-center, phase 1/2 trial invested safety and efficacy
Bradstreet et al.49	Fetal stem cells	45 (3–15) allo	IV	30 × 106/mL	Improvement in cognitive ability, behaviors, sociability Improvements in immune functions	No	Open-label, prospective pilot study invested safety and efficacy
Li et al.39	UCB	14 (3–12)	IV/IT	2–6 × 106 cells	Improvement in the CARS and increase of NGF levels in the CSF	No	Non-randomized, open-label, single-center trial invested safety and efficacy
Dawson et al.41	UCB	25 (2.3–6.0)	IV	1–5 × 107cells	Significant improvements in children’s behavior	No	Open-label, single-center, phase 1 trial invested safety and feasibility
Riordan et al.44	UC-MSC	20 (6–16) allo	IV	3.6 × 107cells x 4	The CARS and ATEC scores of eight subjects decreased	Mild or moderate and short in duration	Non-randomized, open-label, single-center, phase 1/2 trial invested safety and efficacy
Sun et al.45	UC-MSC	12 (4–9) allo	IV	2 × 106 /kg × 3	6/12 participants demonstrated improvement in at least two ASD-specific measures	5 participants developed new class I anti-human leukocyte antigen (HLA) antibodies	Open-label, single-center, phase 1 trial invested safety and feasibility
Sharma et al.47	BM-MNC	254 (Under 5—over 15) auto	IT	1 time immediately after isolation	Improvement of eye contact, attention and concentration, hyperactivity, sitting tolerance, social interaction, stereotypical behavior, aggressiveness, communication, speech, command following and self-stimulatory behavior	No major procedure-related adverse events. 5 patients, with history of seizure and abnormal EEG, had an episode of seizure which was managed using medications.	Non-randomized, open-label, single-center trial invested safety and efficacy
Dawson et al.40	UCB	56 (2–7) auto 63 (2–7) allo	IV	<2.5 × 107 cells/kg × 1	Analysis of the entire sample showed no evidence that CB was associated with improvements in the primary outcome. There was also no overall evidence of differential effects by type of CB infused. In a subanalysis of children without ID, allogeneic, but not autologous, CB was associated with improvement in a larger percentage of children on the clinician-rated Clinical Global Impression–Improvement scale, but the OR for improvement was not significant. Children without ID treated with CB showed significant improvements in communication skills and exploratory measures, including attention to toys and sustained attention (eye-tracking), and increased alpha and beta electroencephalographic power.	There were 6 SAEs reported in 6 unique participants, including 3 in the placebo arm (viral gastroenteritis, dehydration, and aggression), 1 in the autologous CB cohort (concussion), and 2 in the allogeneic CB cohort (pediatric autoimmune neuropsychiatric disorders associated with streptococcal infection [PANDAS] and dehydration). None of these events was related to the study product.	A phase2 randomized, placebo-controlled, double-blind clinical trial invested safety and efficacy
Sharifzadeh et al.46	BM-MSC	32 intervention 14 control 18 (5–15) (Mean;9.5)	IT	First 0.5–1 × 108 Second 0.3–0.5 × 108	The improvements in CARS total score, CARS-II autism index, and CGI global improvement showed no significant differences between the groups over 12 months. However, the main effect for time group interaction was significant regarding the CGI-severity of illness, showing a significantly more pronounced improvement in the intervention group (P = .002).	Injection-related side effects, such as hospital complications, short-term and long-term complications of 12 months were not observed in any of the patients.	Randomized, open-label, single-center trial invested safety and efficacy
Nguyen et al.48	BM-MNC	30 (3–7) auto	IT	First: 4.2 × 107/kg Second: 4.1 × 107/kg	Significantly Improvement in the CARS, and the median Vineland Adaptive Behavior Scales.	No severe adverse events	Non-randomized, open-label, single-center trial invested safety and efficacy

ASD: autism spectrum disorder; BM-MNC: bone marrow mononuclear cell; CARS: Childhood Autism Rating Scale; CB: cord blood; CSF: cerebrospinal fluid; EEG: electroencephalography; ID: intellectual disability; IT: intrathecal; IV: intravenous; MSC: mesenchymal stromal cell; UC-MSC: umbilical cord–derived MSC; UCB: umbilical cord blood cell; ATEC: Autism Treatment Evaluation Checklist. SAE: Serious adverse event.

Discussion

Cell Therapies: Treatment Based on New Pathophysiological Principles

UCBs for ASDs

UCBs prevent impairment of microglia, chronic inflammation, and oxidative stress, as well as enhance neurological regeneration. Human CD34 positive cells have been shown to secrete various growth factors such as brain-derived neurotrophic factor (BDNF), glial cell line–derived neurotrophic factor (GDNF), vascular endothelial growth factor (VEGF), and numerous angiogenic factors, including hepatocyte growth factor (HGF) and insulin-like growth factor-1^61,62. Considering that human CD34 cells have an effect on BDNF production and that this neurotrophic factor is widely described as altered in ASD, this mechanism could be a potential therapeutic effect of UCBs^63–65. Immunological and microglial changes are related to behavioral outcomes in ASD⁶⁶. For example, researchers have suggested that strategies to restore the physiological gut microbiome, such as probiotic supplementation and fecal microbiota transplantation, may improve behavioral symptoms in ASD⁶⁷. Considering the angiogenic and vascular reparative capabilities of endothelial progenitor cells, there is an altered expression of genes associated with blood–brain barrier integrity coupled with increased neuroinflammation and possibly impaired gut barrier integrity in the ASD brain⁶⁸. In 2004, we reported that endothelial progenitor cells have angiogenic and vascular reparative capabilities that make them ideal for neurovascular repair⁶⁹. Endothelial progenitor cells enhance subsequent neuronal regeneration with a rich vascular environment, along with the generation of other nurturing neuronal mediators from CD34 positive cells, such as VEGF, epidermal growth factor 2, and insulin-like growth factor 1⁷⁰. UCBs can suppress chronic inflammation in addition to paracrine function, angiogenesis, and restoration of immunological balance^71,72. Min et al.⁷³ reported that allogeneic UCBs therapy for CP improved glucose metabolism, as observed in a Positron Emission Tomography (PET) study.

BM-MNCs for ASDs

BM-MNCs are not a single population⁷⁴, and each population is disputed in terms of its significance⁷⁵. Regarding glucose metabolism in ASD, Serge et al. reported that glucose metabolic rates decreased in the parietal lobe, frontal premotor, eye-field areas, and amygdala⁷⁶. Anil Kumar et al.⁷⁷ reported that 4 of 10 patients with autism had abnormal PET scan findings, whereas none of the patients in the control group had abnormal PET scans. Their findings support the hypothesis of hypometabolism of glucose in patients with ASD. Sharma et al.⁷⁸ performed [18F] 2-fluoro-2-deoxy-D-glucose PET scans on 45 patients with ASD to study age-related developmental changes in brain metabolism. The results showed that, in contrast to the control data, the median standardized uptake values in patients with ASD decreased linearly with increasing age. Compared with controls, children with autism aged < 5 years showed greater metabolism, and older children showed lower metabolism. In the ASD group, comparison of absolute standardized uptake values within different regions of the brain revealed relatively lower metabolism in the amygdala, hippocampus, para-hippocampal gyrus, caudate nucleus, cerebellum, mesial temporal lobe, thalamus, superior and middle temporal poles, and higher metabolic uptake in the calcarine fissure and Heschl’s gyrus. These results explain the baseline and developmental changes in brain metabolism among the different age groups in patients with ASD. Zhao et al.⁷⁹ reported the characteristic changes in glucose metabolism and ASD-like behaviors in the first and second generations of 12- and 18-month-old male mice, respectively. Whole genome bisulfite sequencing of sperm from advanced paternal age mice identified differentially methylated regions within the whole genome and differentially methylated regions within promoter regions, suggesting that specific genes and relevant pathways might be associated with ASD and aberrant glucose metabolism in the offspring of advanced paternal age males. These results suggest that epigenetic reprogramming induced by aging in the male sperm may lead to high risks of aberrant glucose metabolism and the development of ASD behaviors in intergenerational and transgenerational offspring. UCBs and peripheral blood mononuclear cell infusion therapy for patients with CP improved brain glucose metabolism, as observed in PET studies^73,80. Autologous BM-MNC therapies for patients with ASD also showed improvements in glucose metabolism and various cognitive and behavioral symptoms such as eye contact, attention and concentration, hyperactivity, sitting tolerance, social interaction, stereotypical behavior, aggressiveness, communication, speech, command following, and self-stimulatory behavior, as well as improvement of glucose metabolism and motor function in patients with CP^47,81. In 2020, we reported that angiogenesis is activated by BM-MNCs via gap junction–mediated cell–cell interaction soon after cell transplantation and that cell–cell interaction via gap junctions is the prominent pathway for the activation of angiogenesis in endothelial cells and improvement of glucose uptake⁸². Transplanted BM-MNCs transferred small molecules to endothelial cells via gap junctions, followed by activation of hypoxia inducible factor 1α (Hif-1α) and suppression of autophagy in endothelial cells⁸² (Fig. 3A). Furthermore, we reported that neurogenesis is activated in the hippocampus of aged mice following BM-MNC transplantation⁸³. The mechanism by which BM-MNCs improve glucose metabolism and neurogenesis may have therapeutic potential in individuals with ASD.

Figure 3.

Schema of therapeutic mechanism of stem cell transplantation. (A) Therapeutic mechanism of hematopoietic stem cell transplantation on angiogenesis. Hematopoietic stem cells rely on anaerobic metabolism and are rich in energy sources, including glucose (i). In contrast, injured endothelial cells with ischemia are poor in energy sources (ii). Energy sources transfer from transplanted hematopoietic stem cells to injured endothelial cells occurs by gap junction formation (iii) that activates Hif1α at injured endothelial cells (iv). (B) Therapeutic mechanism of mesenchymal stem cell transplantation on immune suppression. Mesenchymal stem cells rely on aerobic metabolism and are not rich in energy sources, including glucose (i). In contrast, excessively activated endothelial cells possess much energy sources (ii). Energy sources transfer from excessively activated cells to transplanted mesenchymal stem cells occurs by gap junction formation (iii) that inactivates excessively activated cells (iv).

MSCs for ASDs

Microglia in patients and animals with ASD symptoms can be in the apoptotic phase, with high turnover rates of microglia found in some pathological conditions⁸⁴. MSCs secrete heterogeneous lipid bilayer vesicles called extracellular vesicles (EVs), which act as mediators of inter-cell communication. These exosomes and EVs secreted from MSCs improve neuronal function in neurologically injured models⁸⁵. Perets et al. reported the long-term beneficial effects of neurotrophic factor-secreting MSC transplantation in a mouse model⁵¹. Transplantation of MSC resulted in a reduction of stereotypical behaviors, a decrease in cognitive rigidity, and an improvement in social behavior. Tissue analysis revealed elevated BDNF protein levels in the hippocampus accompanied by increased hippocampal neurogenesis in MSC-transplanted mice compared with sham-treated mice. This result might indicate a possible mechanism underlying the behavioral improvement⁵⁰. Exosomes derived from MSC would have a direct beneficial effect on the behavioral autistic-like phenotype of the genetically modified Shank3B knock-out mouse model of ASD. They indicated that intranasal treatment with exosomes derived from MSC improves the core ASD-like deficits in this mouse model of ASD and, therefore, has the potential to treat ASD patients carrying the Shank3 mutation⁵³. Our group also demonstrated the amelioration of neuronal injury followed by functional improvement in MSC-administered mouse models, which resulted from the secretion of trophic factors such as BDNF and HGF rather than neuronal differentiation and external cell replacement by MSCs⁸⁶. Human CD34 positive cells have been shown to secrete various growth factors, such as BDNF, GDNF, and VEGF, and numerous angiogenic factors, including HGF and insulin-like growth factor-1.

We also demonstrated that secretomes from MSCs can alter the phenotype of activated microglia. UC-MSCs immunomodulate microglia and change the phenotype of LPS-activated microglia, restoring actin dynamics and phagocytosis by increasing active Rho GTPase activity, in which microglia change their amoeboid to a more ramified pattern⁸⁷. This suppression and immunomodulation of activated microglia by MSCs may have therapeutic potential for individuals with ASD, in which excessive microglial activation leads to aberrant neural connections. In 2021, we reported that enhanced inflammation in the injured brain is suppressed by MSC through gap junction mediated cell–cell interactions between the cerebral endothelium and intravenously transplanted MSC soon after cell transplantation⁸⁸. Fig. 3B shows that MSCs are dependent on aerobic metabolism and do not contain a high amount of energy sources, including glucose (1). Conversely, excessively activated endothelial cells and white blood cells possess too much energy (2). Excessively activated cells transfer energy to transplanted MSCs through the formation of gap junctions (3), which inactivates excessively activated cells (4). In contrast, human UC-MSC therapies for individuals with CP showed improvement in motor function and increased glucose metabolism by PET-CT scan⁸⁹. The mechanism by which MSCs improve glucose metabolism and suppress enhanced inflammation may lead to therapeutic potential in individuals with ASD.

Conclusion

Nine out of 10 clinical studies of cell therapies for individuals with ASD showed good results with no noteworthy adverse events. The significance of gap junction–mediated cell–cell interactions between the cerebral endothelium and transplanted cells was observed in both BM-MNCs and MSC transplantation. These new findings provide a novel paradigm for cell therapy for ASD.