Description

Retinoic acid (RA), the active metabolite of vitamin A, broadly regulates gene expression. The RA signaling pathway plays an essential role in embryonic development, including the development of the body axis, eye, brain, and heart (Ghyselinck & Duester, 2019). One of the key enzymes in the biosynthesis of RA is aldehyde dehydrogenase 1 family member A3 (ALDH1A3). ALDH1A3 converts retinaldehyde to retinoic acid and it is expressed early in forebrain development (McCaffery & Drager, 1994). Several mutations in ALDH1A3 have been implicated in patients with autosomal recessive microphthalmia and other neurological disorders (Fares-Taie et al., 2013; Roos et al., 2014). However, current animal models of ALDH1A3 have large truncations of the protein. Direct evidence of the effects of missense variants on ALDH1A3 protein activity has not yet been obtained.

In this study, we aimed to determine the functional consequence of ALDH1A3(C174Y) missense variants implicated in patients (Roos et al., 2014). Patients with the Cys174Tyr missense variant presented with autosomal recessive microphthalmia, autistic symptoms, and intellectual disabilities. The human ALDH1A3 C174Y missense residue is located at the amino-terminal nicotinamide adenine dinucleotide (NAD)-binding domain (Moretti et al., 2016), which is evolutionarily conserved between human and commonly used model organisms, such as C. elegans, Drosophila, zebrafish, and mouse. The ALDH1A3 orthologs in these model organisms have similar amino acids at >75% of positions of the human ALDH1A3 protein. We believe the conservation of protein sequences among various species indicates the essential role of this domain in protein functions. Here, we use C. elegans as a prime candidate for studying the causality of missense variants in ALDH1A3. We generated the targeted ALDH1A3 C174Y missense variant in its C. elegans orthologous gene alh-1 (Figure 1A).

The short lifespan and genetically modifiable nature of C. elegans allow rapid screening of functional impactful missense mutation implicated in human diseases. We previously established a working pipeline to introduce autism-associated missense mutations into the genome of C. elegans using CRISPR/Cas9 and homology-directed repair (Wong et al., 2019). Here, we used this pipeline to generate a missense mutant, alh-1 (C172Y), that matches the human ALDH1A3(C174Y) variant. We then characterized this C. elegans missense mutant in a fecundity assay to quantify the robustness of germline development. The C. elegans germline development is a highly sensitive process, which reflects subtle changes in sensory perception, feeding behavior, and metabolism (Hubbard et al., 2013). As a result, we used the C. elegans fecundity assay as a screening tool to identify functional changing missense mutations, as defects in germline development will result in a decreased brood size.

We evaluated the impact of C172Y missense mutation on development using a fecundity assay, which measures the number of viable progeny per hermaphrodite. The alh-1(C172Y) missense mutant nematode, alh-1(sy899), showed only 54% of wild-type fecundity (Figure 1B. Wild-type: 234 ± 7; alh-1(C172Y): 127 ± 8. p < 0.0001). We observed that two of the eighteen mutant animals had very low brood size and a lot of unhatched eggs (17 unhatched eggs to a brood size of 34, and 4 unhatched eggs to a brood size of 38). The other animals had essentially no unhatched eggs or dead larvae. This result indicated a malfunction in the aldehyde dehydrogenase protein, resulting in partial sterile and germline developmental defects. Our result is consistent with the observation that the C. elegans alh-1(tm5823) putative null mutant strain produces some dead animals and a highly expressed sterile phenotype (personal communication with Dr. Shohei Mitani, NBPJ). The alh-1(tm5823) mutant harbors a 618 bp deletion, located outside of our C172Y missense site and removes part of the NAD-binding and catalytic domains. Our conserved C172Y missense residue is also located in the NAD-binding domain. In addition, the developmental phenotype is displayed in other animal models of the aldehyde dehydrogenase family. For example, Aldh null larvae and adults are less viable in the presence of ethanol in Drosophila (Fry & Saweikis, 2006). Moreover, mouse Aldh3^-/- homozygotes were impaired in early forebrain development, possibly through failure to induce genes needed to establish regionalization (Molotkova et al., 2007). The Aldh1a3 null mouse showed perinatal lethality that could be rescued by maternal RA treatment (Dupé et al., 2003).

The developmental defects caused by mutations in ALDH1A3 are likely to act through alteration in the RA signaling pathway. RA binds to its nuclear RA receptor (RAR), which forms a heterodimer with retinoid X receptors (RXRs) and RA response elements (RARE). The RAR/RXR/RARE complex regulates gene transcription at specific locations during various developmental stages. In addition, RA is known to promote the differentiation of neurons prenatally; RA concentration, indicated indirectly by ALDH1A3 expression, influences the maturation of selected parts of the cerebral cortex postnatally (Wagner et al., 2006). This could be a possible explanation of a connection between ALDH1A3 mutations and the neurological symptoms displayed in patients with intellectual disabilities and autism spectrum disorder. Our result illustrated that ALH-1(C172Y) residue is necessary for wild-type ALH-1 protein function in C. elegans. Since the ALH-1(C172Y) residue is conserved between C. elegans and humans, it is likely that this residue also has a functionally significant role in human ALDH1A3 protein. Further study in other model organisms is needed to validate this and potentially elucidate the mechanistic basis of the functional consequences of the ALDH1A3(C174Y) variation.