Autism spectrum disorder (ASD), also known as autism, is a heterogeneous neurodevelopmental disorder characterized by deficits in social interaction and communication, accompanied by repetitive behaviors (Schafer et al., 2019). In this review, we will use the term “autism” to refer to ASD in general for brevity. In 2020, the estimated prevalence of ASD among 8-year-old children in America was approximately one in 36 (Maenner et al., 2023). ASD is typically diagnosed in early childhood and is considered a lifelong condition (Lord et al., 2018). Despite the high prevalence and substantial impact of ASD, the development of effective therapies is hindered by our limited understanding of its underlying pathogenesis. Behavioral interventions are the primary approach for managing ASD. Although antipsychotics such as risperidone and aripiprazole are used for ASD-related irritability, there are no approved medications specifically targeting the core symptoms of autism, such as social communication deficits and repetitive behaviors (Neul and Sahin, 2015; Xu et al., 2019). The lifespan cost of supporting a person with ASD in the United States was about $2.4 million, and by 2025, the financial burden of ASD may far exceed the costs of diabetes and attention deficit hyperactivity disorder (Buescher et al., 2014; Leigh and Du, 2015). ASD places a significant burden on families, society, and the economy, suggesting that further research into the pathogenesis of ASD and finding a safe, effective, reliable, and inexpensive treatment strategy is urgent (Buescher et al., 2014). According to many epidemiological reports, ASD is caused by a complex interplay of genetic susceptibility and/or environmental factors, and potentially immunological influences (Theoharides, 2013; Colvert et al., 2015; Geschwind and State, 2015; Modabbernia et al., 2017). Current research on the etiology of ASD has mainly focused on identifying risk genes and neurological abnormalities (Hazlett et al., 2017; Satterstrom et al., 2020). Neuropathological investigations have revealed abnormal brain development during the early prenatal period in individuals with ASD (Rodier, 2002), involving disrupted functional connectivity (Courchesne and Pierce, 2005), cortical hyperexpansion (Hazlett et al., 2017), altered neuronal plasticity (Ardalan et al., 2022), and extensive atypical synaptic function (Johnson et al., 2015). However, the precise pathogenesis of ASD remains elusive.

The brain accounts for at least 20% of the body’s resting metabolic energy consumption, primarily dedicated to supporting neuronal activity (Magistretti and Allaman, 2015). The brain vasculature develops into a unique and efficient vascular system different from the peripheral vascular system, to maintain enormous energy consumption and normal brain function. Recent insights have expanded our understanding of the neurovascular interactions beyond traditional oxygen and nutrient supply. Cerebrovascular elements also emerge as critical scaffolds and paracrine signaling mediators, regulating the development, function, and maintenance of various neural cell populations (Paredes et al., 2018; Segarra et al., 2019). The precise interaction between the nervous and cerebrovascular systems is essential for normal development, functional maintenance, and aging of the central nervous system (CNS) (Peguera et al., 2021). In recent years, there has been growing interest in exploring the role of cerebrovascular abnormalities in neurodevelopmental and neurodegenerative diseases, such as schizophrenia and Alzheimer’s disease (Nation et al., 2019; Ouellette and Lacoste, 2021). The potential contribution of cerebrovascular abnormalities to ASD has only recently been gradually described (Ouellette et al., 2020). Although some studies have reported the presence of cerebrovascular abnormalities in individuals with ASD, its contribution to the pathophysiology of ASD remains incompletely understood. In this work, starting from the process of cerebrovascular formation during embryonic development, we review research findings on cerebrovascular alteration in ASD patients. By exploring the association of specific cerebrovascular characteristic defects with neurological abnormalities and behavioral disturbances in ASD, this review aims to enhance our understanding of ASD’s pathogenesis and provide a theoretical foundation for the development of novel therapeutic interventions.

Neurovascular development refers to the parallel emergence and formation of the neural and cerebrovascular during early embryonic stages (Bautch and James, 2009). While the vascular and nervous systems each establish distinct specialized networks, they also share overlapping mechanisms. This integration of signaling pathways and cellular responses is crucial for normal brain development, function, and aging (Ouellette and Lacoste, 2021). In mice, between embryonic day 8.5 (E8.5) and 10 (E10), the dorsal ectoderm region containing neuroectodermal cells undergoes specialization into the neural plate. Subsequently, the neural plate elongates, folds, and closes, forming the neural tube (Nikolopoulou et al., 2017). Incipient neural tubes have no vessels, this mild “physiological hypoxic” environment promotes the proliferation and differentiation of neural stem cells (NSCs). This process is likely mediated through the hypoxia-inducible factor-1 (HIF-1) signal pathway (Zhu et al., 2005). Studies involving conditional knockdown of HIF1-α in mouse midbrain-derived neural precursor cells (mNPCs) have demonstrated an impaired midbrain-specific proliferation of mNPCs and a significant reduction in dopaminergic differentiation in vitro. Notably, this phenotype could be rescued by 50 ng/ml VEGF treatment (Milosevic et al., 2007). Conversely, the increasing oxygen and nutrient demands of immature neurons serve as signals to promote angiogenesis (Ouellette and Lacoste, 2021). Simultaneously to the closure of neural tube, the angioblasts (endothelial cell precursors) in the adjacent presomitic mesoderm are recruited by neural tube to form perineural vascular plexus (PNVP) which constitutes the initial vascular network in CNS (James and Mukouyama, 2011). Vascular endothelial growth factor A (VEGF-A), a downstream signal of the transcription factor HIF-1 primarily secreted by the CNS-resident NPCs, acts as the initial driver to trigger angiogenesis (Ferrara et al., 2003; James et al., 2009; Peguera et al., 2021). Thereafter CNS vascularization accompanies its development, followed by massive neurovascular molecular communications. VEGF-A facilitates tip cell migration and enhances stalk cell proliferation at the tips of vascular sprouts, promoting vascular growth into non-vascular regions (Gerhardt et al., 2003). Around E10.5, vessel sprouts from the PNVP invade the neuroectoderm, extending from the pial surface to the luminal surface, resulting in the formation of the intraneural vascular plexus (INVP) (James and Mukouyama, 2011). In general, numerous angiogenic sprouts originating from the IVNP follow the radial glia cells (RGCs) fibers, branch laterally, and anastomose to form the periventricular vascular plexus (PVP) within the ventricular zone (VZ) (Tata and Ruhrberg, 2018). Subsequently, the periventricular and perineural vessels coalesce, forming a complex network of blood vessels intertwined with neural cells that undergo dynamic remodeling until postnatal stage. The instructive role of the vascular system in CNS fate has been largely overlooked in the past but is now garnering increasing attention.

At the same time as neural tube vascularization begins, RGCs transformed from NSCs, present a bipolar morphology, with a long in basal side vessels in the meningeal surface (Sun and Hevner, 2014). RGCs are a major class of neural progenitor cells as well as serve as scaffolds for pyramidal neuron migration in cortical formation (Sun and Hevner, 2014). A 2018 study identified the significant role of CNS vascular endothelial cells (ECs) in regulating neuronal migration and differentiation (Segarra et al., 2018). Reelin, a large glycoprotein secreted by Cajal-Retzius cells, plays a crucial role in neocortex development. Knockdown of reelin/Dab1 signaling specifically in the vascular system impaired the anchoring of RGCs fibers on pial vessels, induced altered pyramidal neuron location, and affect the differentiation of glial cells to neuronal cells (Segarra et al., 2018). This suggests that the vasculature may direct neuronal migration through the function of Dab1 on ECs. Moreover, reelin interacts with the VEGF/VEGF2 pathway to control EC proliferation and vascular filopodia expansion, exerting a powerful proangiogenic effect (Segarra et al., 2018). As a good example, indicated that Reelin/Dab1 signals perform bivalent functions to coordinate neurovascular communication. In addition, vascular ECs around the stem cell niche can promote NPCs proliferation by secreting Neuropilin 1 (NRP1), as one way of vascular regulation of neurogenesis (Tata et al., 2016). The neuropilin (NRP) family members NRP1 and NRP2 also work as receptors or co-receptors of VEGF and semaphorins to modulates angiogenesis (Simons et al., 2016). Knockdown of endothelial uncoupling protein 2 (UCP2) in ECs inhibited NPCs differentiation into neuronal but promoted differentiation into astrocyte, this suggests that blood vessels near NPCs influence neurogenic-to-gliogenic transition (Wang et al., 2022). Conversely, Wnt signaling is crucial for guiding cerebral angiogenesis during neural development. NPCs expressing Wnt7a/b ligand stimulate sprouting angiogenesis of the INVP and PNVP, promoting invasion of the neural tube and inducing and maintaining blood-brain barrier (BBB) properties (Stenman et al., 2008; Daneman et al., 2009). This process involves the interaction of Gpr124 (an orphan GPCR) expressed by ECs with the Norrin/Frizzled4 signaling pathway (Zhou and Nathans, 2014). Throughout the embryonic and postnatal stages, NPCs undergo a highly active process of cell division, differentiation, and migration to establish the intricate and organized CNS. Overall, balanced neurovascular interactions are essential to support the development of the CNS and cerebrovascular systems.

During vertebrate development, specialized blood vessels are formed to meet the specific requirements of different tissues. The formation of the BBB coincides with the growth of vasculature into the CNS. BBB is a specialized vascular barrier between circulating blood and nervous system, acting as a gatekeeper to maintain homeostasis (Langen et al., 2019). It consists of a specialized monolayer of brain capillary ECs. These cells exhibit unique cytological features that differ from those of peripheral ECs, including specialized tight junctions (TJs) between cells, designated transporters that highly selectively control specific substrates and low rates of transcellular vesicle trafficking (Langen et al., 2019). However, early transplantation experiments demonstrated that the BBB properties are induced by the perivascular cellular (pericytes and astrocytes) and CNS environment, rather than the intrinsic properties of CNS endothelial cells (Stewart and Wiley, 1981; Daneman et al., 2010). During embryonic development, the BBB gradually forms from the vascularization of the CNS. At an early stage of cerebrovascular development (sprouts from PNVP), they already acquire TJs under the regulation of canonical Wnt signaling (Risau, 1991) and express barrier function-related genes such as CLDN-5 (Morita et al., 1999). However, blood vessels do not exhibit barrier properties in the early stages of development due to the continued active transcellular transport of ECs. The inhibition of EC endocytosis gradually occurs as the barrier forms (Langen et al., 2019).

Previous physiological studies on cerebral blood flow (CBF) dynamics have demonstrated intensive intercellular communication between cells of the vasculature and nearby neurons and glial cells (McConnell et al., 2017). The BBB is now understood to function as part of a multicellular unit called the neurovascular unit (NVU) rather than operating independently (Figure 1) (McConnell et al., 2017). Endothelial cells are connected by tight junctions, surrounded by pericytes, and embedded in the basement membrane, the end-foot processes of astrocytes surround blood vessels, forming NVU (Abbott et al., 2006; Tam and Watts, 2010). Compared with peripheral blood vessels, cerebral vessels require dense vascularization to maintain the supply of nutrients and oxygen and precise control of CBF (Ouellette and Lacoste, 2021). In the mammalian brain, cerebral vessels respond to the needs of the nervous system by increasing the rate of CBF and oxygen transport, a mechanism known as neurovascular coupling (Iadecola, 2017). The integrity of the NVU is critical for this normal function (Stobart and Anderson, 2013). Besides, the NVU induces and regulates BBB establish (Abbott et al., 2006), and controls the selective transportation of substances in the CNS (Attwell et al., 2010; Daneman and Prat, 2015; Iadecola, 2017). Therefore, the interaction among the cellular elements of the NVU is crucial for maintaining CNS homeostasis (Segarra et al., 2019).

The neuronal activity also ensures the development of the cerebrovascular network. In the stress model of rats with social isolation, occurs BBB damage early, manifested as increased BBB permeability accompanied by increased IL-6 expression (Schiavone et al., 2017). Chronic stimulation of neonatal but not adult mice, such as repetitive sounds, whisker deflection, and chemically induced seizures, reduced endothelial proliferation and vascular sprouting, which may be attributed to inhibition of vascular growth by abnormally active interneurons and glial cells through the release of nitric oxide (Whiteus et al., 2014). Exosomes provide another mechanism for mediating neurovascular communication in the CNS. Exosomes are extracellular vesicles enriched in proteins and lipids that are produced by nearly every cell type and play a significant role in neuron-glial cell communication (Frühbeis et al., 2012; Tkach and Théry, 2016). These exosomes carry microRNAs, which can cross the BBB and regulate gene expression in various physiological and pathological processes of the CNS through the RNA-induced silencing complex (Tkach and Théry, 2016). In zebrafish larvae, embryonic neurons express a microRNA called miR-132 in exosomes, which regulates BBB integrity by controlling the expression of endothelial VE-cadherin. When miR-132 is antagonized, zebrafish larvae exhibit severe intracranial hemorrhage and impaired BBB integrity. This suggests exosome-mediated long-range communication between neurons and brain ECs (Xu et al., 2017). Overall, the neural and cerebrovascular systems develop and mature simultaneously. Molecular pathways shared by both systems converge during CNS development to regulate the complex processes of morphogenesis and functional formation in the brain.

Prior research has suggested that the etiology of ASD might involve a complex interaction between genetic and environmental factors (Geschwind and State, 2015; Modabbernia et al., 2017). However, the molecular and cellular mechanisms underlying ASD pathogenesis remain poorly understood. While most studies have focused on neuronal aspects, ASD is now thought to be associated with abnormal vasculature. The healthy development and functioning of the brain rely heavily on a well-functioning vascular system. Given the extensive crosstalk between the vasculature and nervous system during embryonic development, it is reasonable to hypothesize that structural and functional abnormalities in the cerebral vasculature could have detrimental effects on neurodevelopment. Surprisingly, few studies have explored the neurological consequences of vascular abnormalities during the embryonic period in the context of ASD. Furthermore, the brain cannot store amounts of energy and therefore depends on a constant supply of oxygen and nutrients of cerebral vasculature, to buffer its high but variable metabolic demand. The unique characteristics of the cerebral vasculature ensure a stable microenvironment for neural activity and provide adequate oxygen and nutrients. These characteristics distinguish it from the peripheral vasculature including a dense vasculature to maintain adequate perfusion, a functional BBB to maintain brain homeostasis, as well as regulating stable local CBF in response to neuronal demand (Ouellette and Lacoste, 2021). Defects in the brain vascular system can have severe and irreversible impacts on the CNS. Increasing evidence indicates that these critical features of the cerebral vasculature are altered in individuals with ASD (Figure 2). It is not clear whether cerebrovascular abnormalities are the cause of autism or the result of pathophysiologic changes. The association between genetic, environmental factors and cerebrovascular abnormalities in ASD needs to be further explored.

Indeed, these cerebrovascular characteristics mentioned earlier are not independent in ASD. Proper cerebral blood supply relies on both normal cerebrovascular network and regulation of neurovascular coupling. Disruptions in either of these factors can impair brain maturation by compromising the oxygen and blood supply necessary for proper brain function. Furthermore, systemic inflammation and immune activation have long been associated with the pathogenesis of ASD as co-morbid conditions. One potential mechanism is inflammation, derived from environmental factors (e.g., gut-brain axis dysbiosis, perinatal maternal infections, vaccines) and genetic susceptibility, as a source of cerebrovascular susceptibility (Theoharides, 2013; Zengeler and Lukens, 2021). Indeed, neuroinflammation could damage the component cells of NVC. Defective NVC cells are unable to maintain the integrity of the BBB, resulting in BBB leakage and inflammatory mediators and cells from the periphery easily entering and damaging the brain. This mechanism has been demonstrated in other psychiatric disorders, such as schizophrenia and Alzheimer’s disease (Carrier et al., 2020; Korte et al., 2020).

The vascular and nervous systems construct a highly specialized system separately, but share overlapping mechanisms during embryonic development. It is plausible that abnormalities in vascular development could adversely affect associated neurodevelopmental processes. Regrettably, these studies that have explored the neurovascular interactions of ASD are largely non-developmental, providing only a snapshot at a single point in time (Azmitia et al., 2016; Fiorentino et al., 2016). For example, studies on cerebrovascular alterations in patients with ASD have been conducted using post-mortem brains (Azmitia et al., 2016; Fiorentino et al., 2016), making it challenging to determine the timing of these defects. Although these studies have identified cerebrovascular vasculature as an important component of ASD, it is unclear whether changes in vascular structure and function are a cause or a consequence of neuropathy in ASD. More research is needed to further explore the pivotal role of cerebrovascular defects in the neuropathology of ASD. Additionally, the limited sample sizes in these studies warrant caution, considering the complex pathophysiology and heterogeneous phenotype of ASD.

However, recent studies have provided evidence of cerebrovascular alterations in individuals with ASD (Azmitia et al., 2016; Fiorentino et al., 2016; Bjørklund et al., 2018). Hypoperfusion and neuroinflammation caused by cerebrovascular alterations may emerge as a new potential pathogenesis of ASD. Thus, conceptually, treatments that improve intracerebral hypoperfusion or restore defective BBB and NVU may show beneficial effects in patients with ASD. In fact, several studies based on the repair of cerebrovascular features have made many new advances. As previously described, perfusion abnormalities in specific cortical regions are associated with key features of ASD (Ohnishi et al., 2000; Bjørklund et al., 2018). Improve hypoperfusion through vasodilatation or pro-angiogenesis may be useful. Considering that the VEGF pathway plays an important role in neurovascular communication, more and more studies have focused on this pathway as a potential therapeutic target of neurodevelopmental disorders and neurodegenerative diseases. In mice experiment, VEGF-C injections were found to dilate blood vessel diameter, restore cerebrospinal fluid flow, and ultimately improve cognitive test results (Da Mesquita et al., 2018). Besides, transplantation of VEGF-A-producing cells has been shown to promote the repair of brain damage caused by ischemia in neonatal rats (Yao et al., 2016). Other studies indicate that intrathecal transplantation of autologous bone marrow mononuclear cells (BMMNCs) in 32 patients with autism has shown significant clinical effects by promoting angiogenesis to improve perfusion and balance inflammation through immunomodulation (Sharma et al., 2013). Notably, the persistent proliferation of blood vessels in the brain of ASD patients after normal development may be related to the disorder of brain functional connectivity and the increased incidence of epilepsy. Anti-VEGF therapies may reduce abnormal neuronal activity in ASD patients by inhibiting vascular plasticity (Greenberg and Jin, 2005; Shibuya, 2009; Azmitia et al., 2016). However, given the high heterogeneity of pathophysiological conditions in individuals with ASD, further studies are warranted to evaluate the feasibility and effectiveness of pro/anti-angiogenic therapies in this population.

In addition, restoring the integrity of the BBB may also benefit patients with ASD by re-establishing vascular function and providing neuroprotection. Study has demonstrated that once-daily administration of P7C3-A20, a compound that stabilizes cellular energy levels, is beneficial to the recovery of BBB endothelial injury, neuroinflammation reduction, and cognitive recovery after traumatic brain injury (TBI) (Vázquez-Rosa et al., 2020). Alvarez et al. (2011) claimed that shh secreted by astrocytes binds to receptors in BBB ECs, has a BBB protective effect. Tumor necrosis-alpha (TNF-α) and interferon-gamma (IFN-γ) -treated astrocytes increased Shh production, to promote BBB repair and balance inflammation induced during BBB injury (Alvarez et al., 2011). This study provides new avenues for designing treatments to repair BBB. Balancing neuroinflammation in the brain is also beneficial for defective BBB and NVU. For example, pioglitazone is a non-steroidal anti-inflammatory drug (NSAID) was found can improve ASD symptoms by reducing microglial activation and subsequent inflammation in children with ASD (Emanuele et al., 2007). Immunoregulatory therapy may be an effective option as a personalized treatment for patients with a distinct inflammatory phenotype (Pape et al., 2019).

Moreover, interventions targeting ASD risk genes associated with neurovascular development, are also expected to provide potential therapeutic benefits in the management of ASD symptoms (Baranova et al., 2021). The Wnt/Shh pathway is critical for controlling neural progenitor cell differentiation, neuronal migration, synaptogenesis, NVU development, and BBB formation, the disruption of the pathway may lead to neurodevelopmental disorders including ASD (Gozal et al., 2021). Therapeutic interventions based on components of the Wnt/Shh pathway are currently being tested or used for the treatment of various neurological diseases (Baranova et al., 2021; Rahi and Mehan, 2022).

This review focuses on neurovascular communication during embryonic development and recent research advancements regarding cerebrovascular alterations in ASD, aiming to deepen our understanding of ASD pathogenesis. We also discuss the possibility of these findings as a mechanistic basis for vascular-based therapy for ASD. As we remarked above, these treatment attempts have yielded some positive results, which further confirm the presence of cerebrovascular defects in ASD individuals. Although these current treatments targeting cerebrovascular structure and function do not provide a cure for autism, it is certain that they have great potential to control disease progression and improve quality of life. The investigation of ASD has increasingly emphasized the role of the cerebrovascular component. These studies provide additional evidence for exploring the mechanisms underlying ASD and demonstrate that ASD can be considered as a neurovascular disease. The cerebral vasculature is crucial for normal brain function, and emerging evidence has linked it to various neurological disorders. It is curious that similar vascular abnormalities are found in neurological diseases with different pathogenesis and clinical features. Cerebrovascular defects make the brain more vulnerable, and it is reasonable to assume that the vascular plays a more important role than previously thought. However, much of the evidence supporting the vascular basis of neurological disease needs to be explored in additional causal studies, to determine the extent to which cerebrovascular alterations are involved in pathogenesis or whether these changes do as a result of pathophysiological changes.

YW: writing-original draft. YW, SY, and ML: writing-review and editing. All authors have read and agreed to the published version of the manuscript.