Acetylcholinesterase plays a non-neuronal, non-esterase role in organogenesis

Abstract

Acetylcholinesterase (AChE) is crucial for degrading acetylcholine at cholinergic synapses. In vitro studies suggest that, in addition to its role in nervous system signaling, AChE can also modulate non-neuronal cell properties, although it remains controversial whether AChE functions in this capacity in vivo Here, we show that AChE plays an essential non-classical role in vertebrate gut morphogenesis. Exposure of Xenopus embryos to AChE-inhibiting chemicals results in severe defects in intestinal development. Tissue-targeted loss-of-function assays (via microinjection of antisense morpholino or CRISPR-Cas9) confirm that AChE is specifically required in the gut endoderm tissue, a non-neuronal cell population, where it mediates adhesion to fibronectin and regulates cell rearrangement events that drive gut lengthening and digestive epithelial morphogenesis. Notably, the classical esterase activity of AChE is dispensable for this activity. As AChE is deeply conserved, widely expressed outside of the nervous system, and the target of many environmental chemicals, these results have wide-reaching implications for development and toxicology.

Keywords: Acetylcholinesterase; Fibronectin; Gut; Intestine; Morphogenesis; Xenopus laevis.

Fig. 1.

AChE plays a non-neuronal, non-esterase role in intestine organogenesis. Normal intestinal elongation and rotation are observed in DMSO-exposed control tadpoles (A,A′). Exposure to malathion (MTH; B,B′), chlorpyrifos-methyl (CPF; C,C′) or Huperzine A (HupA; D,D′) increases the percentage (E) of tadpoles with short/malrotated intestines. AChE activity assays (F) confirm that the applied compounds inhibit AChE in vivo. RT-PCR (G) indicates that isolated intestinal endoderm (indicated by the expression of ifabp, but not foxf1) expresses ache. −RT, control lacking reverse transcriptase. At NF 41, AChE (red) colocalizes with E-cadherin (green) at endoderm cell membranes (H-H″, arrows). By NF 46, AChE is apically enriched (I-I″, arrowheads), with reduced lateral membrane expression (arrows). Intestinal development is normal in control MO-injected embryos (J,J′,N), whereas microinjection of AChE MO results in short/malrotated intestines (K,K′,N). Intestinal malformations are rescued by co-injection of RNA encoding wt AChE (L,L′,N) or mutAChE that lacks catalytic activity (M,M′,N). AChE activity assays (O) confirm that AChE MO knocks down AChE, that wt AChE mRNA increases activity, and that mutAChE mRNA has no effect on AChE activity, relative to controls (uninjected, control MO, GFP mRNA). Higher magnification views of the boxed regions in A-D,H-M are shown in A′-D′,H′-M′, respectively. The number of tadpoles with the phenotype shown among the total number of tadpoles in that experimental group is indicated (A-D,J-M). Bar charts show mean±s.e.m. Significant differences between the percentage of tadpoles with abnormal gut phenotypes or between AChE activities from n=3-16 independent experiments (16-30 embryos per condition per experiment) are indicated by lowercase letters (P<0.05). Scale bars: 1000 μm in A-D′,J-M′; 100 μm in H,I; 25 μm in H′-I″.

Fig. 2.

AChE is required for endoderm cell rearrangement and polarization. In NF 46 intestine sections of embryos injected with control MO (A), AChE MO (B), AChE MO plus wt AChE mRNA (C), or AChE MO plus mutAChE mRNA (D), β-catenin (red) outlines membranes of injected (GFP labeled, green) and uninjected cells. Serial sections from embryos in A-D were immunostained for integrin (green) and aPKC (red) (E-H); the boxed regions are shown at higher magnification in I-L. A single-layer columnar epithelium of polarized cells forms in control MO-injected intestines (I). AChE MO-injected cells (J) are rounder (asterisks), unpolarized [absence of aPKC (red), arrowhead] and fail to form a single layer. Defects are rescued by co-injection with wt AChE mRNA (K) or mutAChE mRNA (L). Scale bars: 100 μm in A-H; 25 μm in I-L.

Fig. 3.

AChE is required for microtubule organization and endoderm differentiation. In NF 46 intestine sections of embryos injected with control MO (A), AChE MO (B), AChE MO plus wt AChE mRNA (C), or AChE MO plus mutAChE mRNA (D), β-catenin (red) outlines cell membranes of injected (GFP labeled, green) and uninjected cells. Serial sections from embryos in A-D were immunostained to visualize microtubules (α-tubulin, green) (E-H); boxed regions are shown at higher magnification in I-L. Microtubules are apically enriched and oriented along the apical-basal axis of columnar epithelial cells (I). Microtubule organization is disrupted when AChE is knocked down (J), but is rescued by co-injection of the AChE MO with wt AChE mRNA (K) or mutAChE mRNA (L). Serial sections from embryos in A-H were immunostained for E-cadherin (green) and IFABP (red) (M-P); boxed regions are shown at higher magnification in Q-T. IFABP is expressed in control MO-injected intestine (Q). AChE knockdown (R) prevents differentiation of the endoderm, as indicated by the absence of IFABP in AChE MO-injected cells. Differentiation is restored by co-injection of the AChE MO with wt AChE mRNA (S) or mutAChE mRNA (T). Nuclei, blue (TO-PRO-3). Scale bars: 100 μm in A-H,M-P; 25 μm in I-L,Q-T.

Fig. 4.

AChE is not required for cell-cell adhesion, but is necessary for cell-substrate adhesion to fibronectin. Dissociated intestinal endoderm cells from control MO-injected (A) or AChE MO-injected (B) embryos reaggregated 30 min (30′) after introduction of Ca²⁺ ions into the medium (0′; see supplementary Materials and Methods). Brightfield/fluorescent images show that both injected and uninjected cells from each injection group reaggregated. Assays were performed using at least three different embryos per condition. Transverse sections through wild-type guts (NF 41) were immunostained for laminin (LM; C,C′) or fibronectin (FN; D,D′). LM (red) is found at basement membranes (arrow) surrounding the gut tube. FN (green) is found at the basement membrane (arrow), but is also enriched at endoderm cell basal poles (arrowheads). Endoderm cells from control MO-injected or AChE MO-injected intestines were plated on LM (E) or FN (F). There is no difference in cell adhesion on LM (E), but cells from AChE MO-injected embryos are less adherent than control cells on FN (F). Mean±s.e.m. for the percentage of adherent cells for six to eight independent embryos. *P<0.05. Scale bars: 1000 μm in A,B; 100 μm in C,D; 25 μm in C′,D′.