Cachd1 interacts with Wnt receptors and regulates neuronal asymmetry in the zebrafish brain

Abstract

Neurons on the left and right sides of the nervous system often show asymmetric properties, but how such differences arise is poorly understood. Genetic screening in zebrafish revealed that loss of function of the transmembrane protein Cachd1 resulted in right-sided habenula neurons adopting left-sided identity. Cachd1 is expressed in neuronal progenitors, functions downstream of asymmetric environmental signals, and influences timing of the normally asymmetric patterns of neurogenesis. Biochemical and structural analyses demonstrated that Cachd1 can bind simultaneously to Lrp6 and Frizzled family Wnt co-receptors. Consistent with this, lrp6 mutant zebrafish lose asymmetry in the habenulae, and epistasis experiments support a role for Cachd1 in modulating Wnt pathway activity in the brain. These studies identify Cachd1 as a conserved Wnt receptor-interacting protein that regulates lateralized neuronal identity in the zebrafish brain.

Fig. 1. cachd1 mutants show bilaterally symmetrical, double-left habenulae.

(A) Dorsal views of whole-mount 5 dpf wild-type sibling and rorschach (rch/u761) mutant larvae showing expression of an asymmetric habenular marker (kctd12.1, indicated with asterisk; box indicates approximate epithalamic region) and markers for liver (selenop2, indicated with arrow), pancreas (prss1, indicated with arrowhead), and ventral retina (aldh1a3). (B) Schematic of Cachd1 protein: two dCache domains (cyan and dark blue), a VWA domain (purple stripes), a FZD interaction domain (FZI; gray stripes), a transmembrane domain (white), and an unstructured cytoplasmic tail. Residues affected in sa17010 and u761 alleles are marked in red at approximate positions in primary sequence. (C) Fluorescence images of transfected HEK293T cells expressing constructs encoding EGFP-tagged wild-type (top; cyan) or rch/u761 mutant Cachd1 (bottom; cyan) and KDEL-tRFP (red) to mark the endoplasmic reticulum. Scale bar, 10 μm. (D) Dorsal views of brains of dissected 4 dpf transgenic siblings from a complementation cross of sa17010 and u761 alleles, stained with antibody to Cachd1 (cyan). The Et(gata2a:EGFP)pku588 (pku588Et) transgene is expressed in dHb_L neurons (magenta). (E) Dorsal view of 2 dpf habenulae after double fluorescent in situ hybridization with cachd1 antisense riboprobe (cyan) and the dHb_L marker kctd12.1 (magenta). (F) Dorsal views of 2 dpf habenulae after immunohistochemistry with antibody to Cachd1 (cyan) co-stained with antibody to HuC/D to mark differentiating neurons (red). The dotted lines in (E) and (F) indicate the approximate position of the posterior commissure; open arrowheads indicate the dorsal habenulae. Shown are maximum projections of [(D) and (E)] confocal z-stacks or (F) single confocal slice. Scale bars, 50 μm.

Fig. 2. Loss of function of cachd1 disrupts habenular efferent connectivity, is epistatic to removal of the parapineal signal, and causes precocious neurogenesis.

(A and C) Dorsal views and (B and D) sagittal projections (dorsal left) of the IPN showing DiI (magenta) and DiD (green) labelling of left- and right-sided habenula neuron axon terminals predominantly innervating the dIPN and vIPN respectively, and raphe (r), in 5 dpf wild-type [(A) and (B), n = 3] or cachd1^u761 mutant [(C) and (D), n = 8] larvae. (E) Dorsal views of 4 dpf wild-type or cachd1^u761 mutant epithalami in which the parapineal was ablated before leftward migration (pineal complex marked by zf104Tg, u711Tg alleles with antibody to GFP; white) after double FISH with kctd12.1 (magenta; n = 26 of 29 wild-type siblings, 11 of 12 cachd1^u761 mutants) or kctd12.2 (green; n = 19 of 23 wild-type siblings, 5 of 5 cachd1^u761 mutants). (F) Dorsal views of 4 dpf larvae from a cross of carriers of cachd1^u761 and sox1a^ups8 alleles after FISH with kctd12.1 [magenta; pineal complex as (C), white]. n = 4 wild-types, 3 cachd1^u761 mutants, 4 sox1a^ups8 mutants, 3 sox1a^ups8, cachd1^u761 double mutants. (G) Dorsal views of Et(gata2a:EGFP)pku588 wild-type or cachd1^u761 mutant habenulae incubated at 48 hours post fertilization (hpf) with a pulse of BrdU to label newly born neurons, then processed for immunohistochemistry at 5 dpf with antibody to GFP (magenta) and antibody to BrdU (green). DAPI (4′,6-diamidino-2-phenylindole) counterstain marks nuclei (gray). (H to J) Segmentation of confocal stacks from Et(gata2a:EGFP)pku588 wild-type or cachd1^u761 mutant larvae incubated at (H) 24, (I) 32 and (J) 48 hpf with a pulse of BrdU, then processed at 5 dpf as in (G). Double-positive cells are represented in magenta; BrdU-positive only cells are indicated in green. Times of pulse indicated at top right. (K) Quantification of the proportion of BrdU-positive neurons that also expressed Et(gata2a:EGFP)pku588 (magenta) in 5 dpf wild-type or cachd1^u761 larvae incubated with a pulse of BrdU at 24 hpf (all timepoints presented in fig. S11). Error bars represent 95% confidence intervals. Total number of cells and larvae for each genotype indicated in axis label in parentheses. Q′ test of equality of proportions [24 hpf, degrees of freedom (DF) = 3, χ² = 40.94, P = 6.7 × 10^-9], post hoc pairwise comparisons using a modified Marascuilo procedure with Benjamini-Hochberg correction for multiple testing, **** P ≤ 0.005. Scale bars, 50 µm in (A) to (H).

Fig. 3. CACHD1 physically interacts with Wnt receptors LRP6 and FZD family members.

(A) (Left) Representative scatter plot of flow cytometry testing binding of FLAG-tagged CACHD1 prey protein to human FZD7-EGFP transiently transfected HEK293E cells detected by means of phycoerythrin (PE)-conjugated secondary antibody. (Right) Without (blue) or with (red) preincubation with antibody to Frizzled OMP-18R5; secondary only negative control (gray). n = 3; one-tailed paired t test (DF = 2, t = 9.53, *** P = 0.0054). (B) Dot plot of human (blue diamonds) or zebrafish (blue circles) Cachd1 or negative control CD200R (gray) prey protein binding (standardized as ΔM_PE (supplementary materials, materials and methods) to cells transiently transfected with EGFP fusion protein constructs indicated (transfections verified by means of antibody labeling are indicated in bold). Each dot indicates a single experiment; horizontal bars indicate the mean, and error bars indicate 95% confidence intervals. One way Welch test of means (Cachd1 prey versus CD200R prey, not assuming equal variances; F = 132.32, DF_num = 30.00, DF_denom = 34.67, P = 5.09 × 10^-28), post hoc pairwise t tests with non-pooled standard deviations, Benjamini-Hochberg correction for multiple testing. Only statistically significant differences between Cachd1 and CD200R prey for individual transfections are presented here for clarity; ** 0.05 ≥ P > 0.01, *** 0.01 ≥ P > 0.005, **** P ≤ 0.005. (C) SPR-based determination of K_D (left) for mouse CACHD1_ECD analyte binding to immobilized mouse FZD5_CRD (cyan) or human LRP6_P3E3P4E4 (3-4, red), and normalized response curves (right) for different CACHD1_ECD:FZD_CRD interactions. RU, response units. (D) TOPFlash responses of HEK293 cells to WNT3A treatment after transfection with a control plasmid (white) or plasmids containing full length rodent Cachd1 (cyan), Cachd1 extracellular domain only (blue; ΔTM-C), or Cachd1 transmembrane and intracellular domains only (gray; ΔN). Mean responses are shown (n = 3 experiments; black dots indicate mean of quadruple technical replicates in each experiment); error bars idicate 95% confidence intervals. One way Welch test of means (not assuming equal variances; F = 13.202, DF_num = 3.00, DF_denom = 4.19, P = 0.014), post hoc pairwise t tests with non-pooled standard deviations, Benjamini-Hochberg correction for multiple testing. ns, P > 0.1, ** 0.05 ≥ P > 0.01.

Fig. 4. CACHD1 forms a ternary complex with FZD5 and LRP6.

(A) Cartoon representation of mouse CACHD1_ECD, [rainbow-colored from N terminus (blue) to C terminus (red)] in complex with mouse FZD5_CRD (cartoon and surface in teal) and human LRP6_P3E3P4E4 (cartoon and surface in salmon pink). The position of the four cache (C-1, -2, -3 and -4), VWA and FZD interaction (FZI) domains of CACHD1_ECD are indicated (PDB ID 8S7C). (B) Superimposed structures of the FZD8:WNT3 complex (PDB ID 6AHY) with the FZD5_CRD:CACHD1_ECD complex. WNT3 is shown as a violet cartoon tube with palmitoleic acid (PAM) as spheres. (C) Superimposed structures of the LRP6:DKK1-C complex (PDB ID 5FWW) with the CACHD1_ECD:LRP6_P3E3P4E4 complex. DKK1-C is shown as a magenta cartoon tube.

Fig. 5. cachd1 interacts genetically with Wnt pathway components.

(A) Dorsal views of wholemount 4 dpf wild-type sibling (n = 12) and lrp6^u351 mutant (n = 9) heads stained for expression of kctd12.1. (B and C) Graphs showing the percentage of (B) 4 dpf wild-type siblings and lrp6^u349/+ larvae or (C) wild-type siblings and tcf7l2^zf55/+ larvae with (gray; sym) or without (white; asym) a symmetric bilateral left phenotype in uninjected larvae and larvae injected with a suboptimal (1 ng) or standard dose (2 ng) of cachd1 morpholino (MO1). Error bars indicate 95% confidence intervals of the proportion. Q′ test of equality of proportions [(B) DF = 2, χ² = 18.71, P = 8.66 × 10^-5; (C) DF = 2, χ² = 7.93, P = 0.019) and post hoc modified Marascuilo procedure with Benjamini-Hochberg correction for multiple testing. (D) Dorsal views of the habenulae of wholemount 4 dpf larvae from an incross of cachd1^u761 and axin1^tm213 mutants, showing expression of kctd12.1. n = 5 wild-types, 6 cachd1^u761 mutants, 3 axin1^tm213 mutants and 3 cachd1^u761, axin1^tm213 double mutants. (E to G) Quantitative RT-PCR data showing relative expression (-ΔΔC_t values) of CACHD1 and known Wnt-responsive genes (CCND1, AXIN2) in (E) HEK293 (HEK) cells untreated, incubated with WNT3A alone (WNT) or WNT3A and RSPONDIN1-conditioned media (W + R), or stable APC mutant cells (APC4); (F) wild-type (Control) and Apc mutant (APC5) mouse organoids; and (G) colorectal cancer-derived cell lines with mutations in Wnt pathway genes (APC mutants, DLD1, SW480; CTNNB1 mutants, HCT116, Ls174T). Data is presented as mean -ΔΔC_t values compared to expression of ACTB (human) or Hrpt1 (mouse) reference genes and untreated controls (HEK293 cells or wild-type organoid). Individual points indicate biological replicates (each an average of three technical replicates), n = 3; error bars indicate 95% confidence intervals. One way Welch test of means [not assuming equal variances; (A) F = 58.83, DF_num = 11.00, DF_denom = 9.41, P = 3.03 × 10^-7; (B) F = 225.66, DF_num = 5.00, DF_denom = 5.16, P = 5.12 × 10^-6; (C) F = 236.49, DF_num = 14.00, DF_denom = 11.33, P = 7.67 × 10^-12], post hoc pairwise t tests with non-pooled standard deviations and Benjamini-Hochberg correction for multiple testing; only statistically significant differences with control samples (HEK or Control) are presented here for clarity. * 0.1 ≥ P > 0.05, ** 0.05 ≥ P > 0.01, *** 0.01 ≥ P > 0.005, **** P ≤ 0.005.