Neurodevelopmental disorders (NDDs) and congenital heart diseases (CHDs) represent two major global public health challenges, affecting over 15% of children and nearly 1% of newborns worldwide, respectively (Bouma and Mulder, 2017; Liu Y. et al., 2019; Romero-Ayuso, 2021). NDDs are characterized by deficits in cognitive function and adaptive behavior (Micai et al., 2020; Hanly et al., 2021), spanning a broad spectrum of neurological conditions of varying severity including notably intellectual disability, developmental delay, autism spectrum disorders, communication and learning disorders, as well as attention deficit/hyperactivity disorder (Ismail and Shapiro, 2019). CHDs, commonly referred to as heart defects, are defined as anatomical and/or functional abnormalities of the heart or its associated great vessels present at birth. The spectrum of CHDs varies widely, encompassing mild defects with minimal clinical impact to severe and life-threatening conditions. This diversity complicates the classification of CHDs, making it both complex and challenging. NDDs represent the most prevalent morbidity among CHD patients, exerting a substantial impact on academic achievement, the transition to independence, and overall quality of life (Peyvandi et al., 2023). Herein, CHD children often demonstrate lower scores for cognition, language, attention, executive functions, as compared to their peers (Derridj et al., 2021). Despite their synergistic and cumulative nature, described as “the cumulative burden of injury,” well-known postnatal risk factors such as postnatal hypoxia/hypoperfusion, perioperative stroke, etc., account for less than 30% of adverse neurocognitive outcomes in NDD (Marelli et al., 2016; Patt et al., 2023). On the contrary, it seems that neurological defects within CHDs are primarily shaped by prenatal factors. This notion is substantiated by well-established evidence suggesting that CHDs can impact fetal brain development, leading to reduced fetal brain volume and disruptions in utero metabolic brain trajectories (Sadhwani et al., 2022; Andescavage et al., 2023; Peyvandi and Rollins, 2023). The heart and brain are indeed intricately interconnected during embryogenesis (Polat et al., 2011; Martinez-Biarge et al., 2013). For instance, certain CHDs can diminish blood oxygenation due to ineffective blood pumping by the heart, leading to insufficient oxygen and nutrients reaching the developing brain. This prenatal hypoxia can alter brain structures, potentially causing long-term neurological deficits (Wang et al., 2021). Lately, a genetic cause has also been proposed to explain the prevalence of NDDs in individuals with CHDs. This hypothesis is notably supported by a large study based on exome sequencing of CHDs patients and their parents, revealing a significantly increased burden of de novo variants in genes involved in both brain and heart development (Homsy et al., 2015).

Recent studies have brought to light that individuals with NDDs carrying pathogenic variants in genes related to the ubiquitin-proteasome system (UPS) may not only exhibit neurological symptoms but also present with CHDs. The UPS, a critical and conserved pathway across eukaryotes, consists of approximately 1,200 genes, enabling the covalent modification of damaged and/or unneeded proteins with ubiquitin for their subsequent degradation by 26S proteasomes (Çetin et al., 2021; Goetzke et al., 2021). Prime example of UPS dysfunction that may result in a dual phenotype of NDDs and CHDs include the recently described neurodevelopmental proteasomopathies caused by genomic alterations in proteasome genes such as PSMD12, PSMC3 and PSMC1 (Table 1).

From a structural perspective, the 26S proteasome consists of a 20S core particle (CP) encased with a 19S regulatory particle (RP). The 19S RP comprises 19 distinct subunits, which can be subdivided into two sub-modules: the base and the lid (Bard et al., 2018; Sahu and Glickman, 2021), both susceptible to loss-of-function mutations (Figures 1A, B). The base segment comprises six diverse AAA + ATPase subunits (designated as PSMC1-6) alongside four non-ATPase subunits, namely, PSMD1, PSMD2, PSMD4, and ADRM1, respectively (Bard et al., 2018; Sahu and Glickman, 2021). The lid component consists of nine structural subunits, namely, PSMD3, PSMD6, PSMD7, PSMD8, PSMD11, PSMD12, PSMD13, PSMD14, and SEM1 (Bard et al., 2018; Sahu and Glickman, 2021). As illustrated in Figure 1A, the 20S CP is a cylindrical structure formed by stacking 28 subunits into four hetero heptameric rings: two external α-rings and two internal β-rings, encompassing the PSMA1-7 and PSMB1-7 subunits, respectively (Tanaka, 2009; Collins and Goldberg, 2017). The 19S RP detects ubiquitin-modified protein substrates via the subunits PSMD4 and ADRM1 which act as ubiquitin receptors (Deveraux et al., 1995; Husnjak et al., 2008). This recognition is followed by their subsequent de-ubiquitination by PSMD14 and their unfolding by PSMC1-6 (Verma et al., 2002; Bar-Nun and Glickman, 2012). After translocation into the 20S CP, linearized substrates are degraded into short peptides via the PSMB5, PSMB6, and PSMB7 catalytic subunits (Schmidt and Finley, 2014). Beyond maintaining protein homeostasis in the cell, the UPS plays a significant role in regulating numerous pathways by selectively targeting kinases, transcription factors, cyclins, enzymes, and/or other key cellular components for degradation (Çetin et al., 2021; Goetzke et al., 2021; Papendorf et al., 2022).

Neurodevelopmental proteasomopathies were first documented in 2017, with PSMD12 loss-of-function mutations identified in patients presenting typical signs of syndromic intellectual disability, including speech delay and abnormal facial features (Küry et al., 2017; Khalil et al., 2018). Remarkably, in addition to neurodevelopmental features, a substantial portion of patients exhibited visceral anomalies, among which cardiac issues, particularly patent ductus arteriosus, emerged as the predominant feature (Cuinat et al., 2023). Five years later, a similar pattern of neurodevelopmental delay coupled with heart defects was observed in individuals carrying PSMC3 loss-of-function variants (Ebstein et al., 2023). Both neurological and cardiac features were similar to those seen in patients with PSMD12 variants, suggesting that both diseases follow the same molecular pathogenesis. This observation was not surprising, considering that both PSMD12 and PSMC3 encode subunits that are constituents of the same protein complex, namely, the 19S RP within 26S proteasomes (Çetin et al., 2021). The notion of a causal link between variants of the 19S RP and NDDs with CHDs gained further support when one patient, afflicted with a neurological syndrome arising from mutations in the PSMC1 subunit of the 19S RP, was also found to have congenital heart malformations (Aharoni et al., 2022). As of today, all proteasome variants causing neurodevelopmental proteasomopathies are consistently found in genes encoding subunits of the 19S RP, specifically PSMD12, PSMC1, PSMC3, and PSMC5 (Figure 1B) (Küry et al., 2017; Khalil et al., 2018; Aharoni et al., 2022; Ebstein et al., 2023; Küry et al., 2024).

The frequent association of CHDs with facial dysmorphism in neurodevelopmental proteasomopathies suggests a significant role of the 19S RP – and by extension, 26S proteasomes – in heart and craniofacial development. Several explanations might account for this connection, including the possibility that proteasome genes autonomously contribute to heart and brain development. Besides, given the expression of proteasomes in the placenta (Wang et al., 2013), it is also tempting to speculate that this dual phenotype may arise from defects in placental development. However, it is unlikely that impaired neurodevelopment is solely due to hypoxia caused by CHDs, as some patients with NDDs do not develop CHDs (Cuinat et al., 2023). One particularly compelling hypothesis suggests that the 26S proteasome dysfunction, which defines these disorders, may arise during the early stages of embryonic development, impacting specific cell populations that contribute to the formation of both organs.

Both the cardiac OFT and the craniofacial skeleton partially find their early origins in neural crest cells (NCCs), a multipotent migratory progenitor cell population derived from the ectodermal layer (McQuillen et al., 2010). The ectoderm is one of the three primary germ layers formed early in embryonic development and gives rise to various tissues and structures in the body, comprising the nervous system, the skin and the NCCs (Trounson, 2002). The process of embryonic development is highly orchestrated and involves a series of complex events, including migration, cell differentiation, tissue formation, and morphogenesis (Gerri et al., 2020). During early embryogenesis, the fertilized egg undergoes multiple cell divisions, leading to the formation of the blastocyst. The blastocyst subsequently differentiates into the three germ layers, one of which is the ectoderm. The dorsal ectoderm gives rise to the neural plate, which then folds and transforms into the neural tube. This process is known as neurulation (Copp et al., 2003). The neural tube will eventually differentiate into the brain and the spinal cord. As shown in Figure 2, as the neural tube forms, a group of multipotent cells referred to as NCCs emerge at its borders (Douarin et al., 2004).

Specifically, NCCs are assigned their identity through the fine-tune activity of several signaling pathways, among which bone morphogenetic protein (BMP), wingless-related (Wnt) factors, fibroblast growth factor (FGF), as well as neurogenic locus notch homolog protein (Notch)/Delta signaling interactions originating from neighboring regions including the neural plate, non-neural ectoderm, and mesoderm (Rogers et al., 2012; Alkobtawi et al., 2021). The early neural crest specifiers include the transcription factors FoxD3, Pax3/7, Ets1, Sox8/9/10, Twist1 and Snail1/2 (Schussler et al., 2021). The primary roles of the neural crest specifiers include three critical functions: 1) they establish NCC fate, 2) they initiate the epithelial-mesenchymal transition (EMT) process, characterized by the switch of cadherins, the secretion of extracellular matrix proteins such as laminin, fibronectin, collagen, vitronectin, thrombospondin, the activation of matrix metalloproteinases, the reorganization of the cytoskeleton, and the presence of distinct surface receptors including integrins and homing receptors, and 3) they play a role in preserving multipotency, by cooperation with pluripotency genes such as Nanog and Oct4 (Kirby and Hutson, 2010; Simões-Costa and Bronner, 2015; Scerbo and Monsoro-Burq, 2020; Zalc et al., 2021).

As the neural tube closes, NCCs begin migrating dorsolaterally under the surface ectoderm or ventrally along the neural tube (Theveneau and Mayor, 2012). As illustrated in Figure 2, NCCs can be categorized into four distinct subpopulations–cranial (sometimes referred to as cephalic), vagal, trunk and sacral–based on their axial origin along the anterior-posterior axis, spanning from head to tail (Milet and Monsoro-Burq, 2012; Simões-Costa and Bronner, 2015; Gandhi and Bronner, 2018). Cranial NCCs primarily reside anterior to the otic placode, extending across the forebrain, midbrain, and the anterior segments (or rhombomeres) of the hindbrain (Figure 2). The vagal NCCs, situated posterior to the cranial NCCs, are found at the levels of somites 1-7 and include cardiac NCCs (at the level of somites 1–3). Posteriorly, the vagal cell population is succeeded by trunk NCCs (somites 8–24) and sacral NCCs (somites 25–33).

More than thirty cell types are derived from NCCs in vertebrates. However, according to their axial level of origin, NCCs may display different potentials. From all levels, NCCs give rise to neurons and glia of the autonomic and sensory peripheral nervous system and to pigment cells progenitors. In addition, cranial NCCs provide mesenchymal precursors of most of head structures including bones, cartilage, connective tissue and tendons (Cordero et al., 2011). Moreover, crucial to our hypothesis, cranial NCCs also form the pericytes of forebrain blood vessels, which regulate the properties of the blood-brain-barrier (BBB) (Etchevers et al., 2001; Korn et al., 2002; Stebbins et al., 2019). Importantly, in addition to maintaining BBB integrity, capillary pericytes are believed to interact with various cell types besides endothelial cells, including astrocytes, microglia, and neurons, thereby participating in processes such as the regulation of neurogenesis (Zhou et al., 2022). In this regard, recent research has shown that pericyte dysfunction may impair neuronal differentiation by disrupting microglial function (Hattori et al., 2022). Furthermore, the notion that blood vessels play a role in brain development is supported by the observation that proper reelin signaling in endothelial cells is required for the development of cortical neurons (Segarra et al., 2018).

In contrast, cardiac NCCs actively contribute to the formation of the aorticopulmonary septum, heart valves and OFT (Figure 2) (Srinivasan and Toh, 2019). It has become clear in the last two decades, that a complex gene regulatory network controls the emergence of the neural crest fate, comprising both regulations common to all NCCs and transcriptional modules specific to NCCs subsets. This notion implies that during the early stages of development, NCCs destined to the heart or the head share signaling pathways, transcription factors, and other cell biology mechanisms that guide their specification, migration, and differentiation. This also suggests that any dysfunction affecting NCCs from all axial levels will have repercussions for both the heart and craniofacial development, as well as indirectly for brain formation via the activity of BBB. Understanding the shared developmental origins and pathways can provide valuable insights into the complex interactions between the heart and the head during embryogenesis and how disruptions in these processes can give rise to diverse clinical manifestations in affected individuals.

As discussed above, the co-occurrence of neurodevelopmental proteasomopathies with CHDs strongly suggests that the repercussions of proteasome dysfunction are particularly prominent within NCCs. A plausible rationale for the heightened susceptibility of NCCs to defective proteasomes might stem from the absence of fully adapted compensatory mechanisms. It is indeed plausible that the mechanisms could either be insufficient or overly robust, leading to diverse detrimental impacts on NCC physiology.

It is widely recognized that the physiology of NCCs depends on the proper functioning of the UPS. For instance, SoxE, an essential transcription factor guiding NCC differentiation into cartilage, glia, and melanocytes (Haldin and LaBonne, 2010), has been found to undergo modification by the ubiquitin-like protein SUMO (Taylor and Labonne, 2005; Lee et al., 2012). While the exact role of SUMO modification in this process remains incompletely understood, it is plausible that it could interfere with the ubiquitination process and thereby alter protein turnover, as previously described for the transcription factor MYC (Sun et al., 2021). Similarly, epithelial-to-mesenchymal transition (EMT) factors are recognized as proteasome substrates targeted for degradation by cullin-RING E3 ligases (Vernon and LaBonne, 2006; Lee et al., 2012).

Nonetheless, to the best of our knowledge, virtually nothing is known about the repercussions of proteasome dysfunction on the intricate series of events of NCC induction, specification, migration and/or cell fate determination. It can only be conjectured that the consequences of proteasome defects observed in other cellular or animal models might have implications for NCCs–and a fortiori in heart and head development–as well, as discussed below.

As highlighted repeatedly, our current understanding of proteasome biology and/or the regulatory mechanisms governing protein homeostasis in NCCs is ill-defined. Addressing this knowledge gap is, however, an essential prerequisite for unveiling the molecular pathogenesis underlying neurodevelopmental proteasomopathies associated with CHDs, as well as for contemplating therapeutic interventions. The reason for the lack of information on this subject is likely attributable to the constrained accessibility of NCCs from animal embryos resulting in a very limited amount of biologic material that is not suitable for conducting biochemical experiments.

However, this limitation can now be readily overcome by using human induced pluripotent stem cells (iPSCs). Indeed, in recent years, several studies have documented the successful generation of SOX10+ NCCs from iPSCs, often achieved through concurrent BMP inhibition and activation of Wnt signaling pathways (Menendez et al., 2011; 2013; Hackland et al., 2017). An ideal strategy to pinpoint the link between NDDs and CHDs would involve the further differentiation of iPSCs-derived NCCs into distinct cranial and cardiac NCC subtypes, each of which serving as precursor cells for craniofacial structures, pericytes and the OFT of the heart, (Stebbins et al., 2019). Of note, cranial identity can be established by adding BMP-4 during the process of NCC differentiation, which results in increased expression levels of cranial-specific DLX genes (Mimura et al., 2016). Unfortunately, as far as our current understanding goes, there is no existing protocol for generating cardiac NCCs from iPSCs. However, it has been demonstrated that supplementing iPSCs-derived NCCs with fetal bovine serum (FBS) and TGF-β results in the development of smooth muscle cells (SMCs) (Srinivasan and Toh, 2019) which participate in the heart OFT formation (Liu X. et al., 2019). In any case, deciphering the impact of proteasome loss-of-function variants on NCC biology would consist of 1) reprograming biological samples from patients with neurodevelopmental proteasomopathies into iPSCs before subsequent differentiation into various NCC subtypes or 2) introducing recurring loss-of-function proteasome mutations into control iPSCs by gene editing. The approach would then consist of determining the ability of these cells to differentiate into NCC and derivatives and uncover the impact of each of the compensatory pathways on this process.

The presented work highlights indeed a series of converging factors that point towards protein homeostasis disruptions within NCCs as a possible underlying cause for the concurrent emergence of NDDs and CHDs. Supporting this hypothesis, NCCs exhibit a particular vulnerability to pathogenic variants in ribosome genes commonly associated with ribosomopathies (Yelick and Trainor, 2015). This susceptibility suggests a high protein synthesis demand in NCCs, making them significant producers of potentially harmful DRiPs that require efficient clearance by the proteasome. Therefore, partial overlap in the clinical phenotype of neurodevelopmental proteasomopathies with certain ribosomopathies such as Treacher Collins syndrome (Vincent et al., 2016), Diamond-Blackfan anemia (Pallanti et al., 2008) or Roberts syndrome (Bermejo-Sánchez et al., 2011) is unsurprising. This overlap is particularly evident in craniofacial anomalies, cognitive impairment and, the frequency of cardiac malformations. While NCCs were discovered 150 years ago (Achilleos and Trainor, 2012), the acknowledgment of the vital significance of protein homeostasis regulation by the UPS is relatively recent, as underscored by the awarding of Nobel Prize in Chemistry to Aaron Ciechanover, Avram Hershko and Irwin Rose in 2004 (Behuliak et al., 2005). Although the exploration of this research field in NCCs is still in its early stages, its significance is expected to endure, given the pivotal role these cells play in the development of both the head and the heart. In this context, future research will have to assess NCCs for their equipment that preserve protein homeostasis, with the objective of understanding how disruptions of this equilibrium lead to the development of heart malformations and neurodevelopmental delay.