Graded effects of unregulated smooth muscle myosin on intestinal architecture, intestinal motility and vascular function in zebrafish

Abstract

Smooth muscle contraction is controlled by the regulated activity of the myosin heavy chain ATPase (Myh11). Myh11 mutations have diverse effects in the cardiovascular, digestive and genitourinary systems in humans and animal models. We previously reported a recessive missense mutation, meltdown (mlt), which converts a highly conserved tryptophan to arginine (W512R) in the rigid relay loop of zebrafish Myh11. The mlt mutation disrupts myosin regulation and non-autonomously induces invasive expansion of the intestinal epithelium. Here, we report two newly identified missense mutations in the switch-1 (S237Y) and coil-coiled (L1287M) domains of Myh11 that fail to complement mlt Cell invasion was not detected in either homozygous mutant but could be induced by oxidative stress and activation of oncogenic signaling pathways. The smooth muscle defect imparted by the mlt and S237Y mutations also delayed intestinal transit, and altered vascular function, as measured by blood flow in the dorsal aorta. The cell-invasion phenotype induced by the three myh11 mutants correlated with the degree of myosin deregulation. These findings suggest that the vertebrate intestinal epithelium is tuned to the physical state of the surrounding stroma, which, in turn, governs its response to physiologic and pathologic stimuli. Genetic variants that alter the regulation of smooth muscle myosin might be risk factors for diseases affecting the intestine, vasculature, and other tissues that contain smooth muscle or contractile cells that express smooth muscle proteins, particularly in the setting of redox stress.

Keywords: Intestine; Myosin; Smooth muscle; Zebrafish.

Fig. 1.

Schematic overview of the mlt dominant modifier screen. (A) Mutagenized wild-type adult fish were mated to mlt heterozygotes (myh11^mlt/+). F1 mlt heterozygotes are heterozygous for a large number of ENU-induced mutations, here indicated by genes ‘A’ and ‘B’. (B) Random matings of F1 mlt heterozygotes carrying rare modifier mutations. Fully penetrant dominant suppressor mutations reduce the frequency of larvae with the typical mlt phenotype from 25% to 12.5% of total larvae. Fully penetrant dominant enhancer mutations induce a mlt phenotype in 25% of total larvae. Red larvae represent mlt homozygotes and modified heterozygotes.

Fig. 2.

Modifier mutations induce cell invasion in mlt heterozygotes. (A-D) Lateral views of live 6-dpf larvae. Arrowhead points to the mid-intestine. (A) A wild-type (WT) larva. (B) A homozygous mlt larva in which there is invasive expansion of the mid- and posterior intestine. (C) A mlt/S237Y compound heterozygote. Arrowhead points to a region of altered intestinal morphology. Inset shows a more severely affected mlt/S237Y compound heterozygote. (D) mlt/L1287M compound heterozygote. Mild and moderate (inset) phenotypes are shown as in C. (E,F) Histological section through the posterior intestine of 6-dpf mlt/S237Y and mlt/L1287M compound heterozygotes following their immunostaining with anti-Keratin (red) and anti-Laminin-1 antibodies (green). Dashed line indicates predicted position of basement membrane disrupted by the invasive epithelial cells; e, intestinal epithelium; sm, smooth muscle. (G,H) Histological section through the posterior intestine of 3-dpf Tg(miR194-Lifeact-GFP) wild-type and sibling mlt/S237Y larvae following immunostaining with anti-GFP (green) and anti-Laminin-1 (red) antibodies, showing GFP-labeled invadopodia (arrowheads) of epithelial cells extending through gaps in the basement membrane. Inset shows a higher-magnification view. (I-K) Confocal projections through the mid- and posterior intestine of 4-dpf Tg(miR194-Lifeact-GFP) wild-type and sibling mlt/S237Y (myh11^mlt/S237Y) larvae following anti-GFP immunostaining (green). Arrowheads point to invadopodia protruding from the basal epithelial cell plasma membrane in the mlt/S237Y mutants. Blue, DAPI-stained nuclei.

Fig. 3.

mlt modifier mutations encode newly identified myh11 alleles. (A,B) Sequencing of intestinal cDNA from modified mlt heterozygous larvae identifies distinct myh11 mutations. Arrows indicate the location of the cytosine-to-adenine transversion mutations that change serine 237 to tyrosine (S237Y) and leucine 1287 to methionine (L1287M). Both amino acid substitutions are outlined in red boxes. Comparable amino acid sequences of human and chicken Myh11 proteins are indicated. (C) Cartoon depicting conserved domains within the MYH11 protein and the corresponding locations of the zebrafish myh11 mutations.

Fig. 4.

mlt modifier mutations alter myosin regulation. (A) Steady-state (normalized) ATPase activity of wild-type (WT) HMM, S237Y HMM and full-length L1287M. Data from phosphorylated and unphosphorylated assays are shown. There is low level activity of the S237Y and L1287M myosins in the absence of phosphorylation. Phosphorylation increases the activity of wild-type greater than it does S237Y myosins, but does not alter the activity of L1287M myosin. (B) Table showing values for maximum ATPase and ADP release for phosphorylated and unphosphorylated WT, S237Y and L1287M myosins. RLC, regulatory light chain.

Fig. 5.

mlt modifier mutants are sensitized to oxidative stress. (A-D) Lateral views of 82-hpf wild-type (WT), homozygous mlt, heterozygous mlt and L1287 homozygous larvae. Wild-type (A), mlt heterozygote (C) and L1287M homozygote (D) larvae treated with menadione (1.5 μM; mena). Morphological changes of the mid- and posterior intestine that are characteristic of epithelial invasion (arrowheads) are seen in the mlt larva (B) and the menadione-treated mlt heterozygote and L1287M homozygote (C,D). In all three mutant larvae, the ventral intestine appears to extend into the yolk. Intestinal morphology is normal in the wild-type larva (A). (E-H) Lateral views of 6-dpf KRAS-axin larvae that either have homozygous wild-type myh11 (E,F), S237Y (G) or L1287M (H) alleles. The intestine is expanded in KRAS-axin larvae as a result of epithelial hypertrophy (brackets, E,F). Further expansion as a result of epithelial invasion is seen in menadione-treated KRAS-axin S237Y and L1287M homozygotes (brackets, G,H), as previously reported for menadione-treated KRAS-axin mlt heterozygotes (Seiler et al., 2012). (I) Quantification of intestinal width in KRAS-axin compound-mutant larvae treated with menadione compared with untreated siblings. ****P<0.001; n.s., not significant. Unpaired Student's t-test performed; mean±s.e.m. (J,K) Histologic analyses of immunostained larvae showing epithelial cell invasion in the intestine of a menadione-treated homozygous S237Y KRAS-axin larva (K) vs a wild-type KRAS-axin larva (J). Green, anti-Laminin-1; red, anti-Keratin; blue, DAPI-stained nuclei. Arrowhead (K) points to invasive cells. Scale bars: 100 µm.

Fig. 6.

Zebrafish myh11 mutations alter intestinal transit. (A,B) Intestinal transit as measured by the expulsion of fluorescent microspheres in wild-type and sibling homozygous mlt larvae injected with a myh11 morpholino (MO) (A), and wild-type and sibling S237Y homozygous larvae (B). Anterior: beads remain in anterior intestinal bulb; posterior: beads remain in mid- and posterior intestine; expelled: no beads remaining in the intestine. ****P<0.001. Chi-squared test performed with 2 degrees of freedom. (C) Time lapse of a wild-type (WT) larva beginning when the fluorescent beads are present in the anterior intestine and followed until bead expulsion. A, M, P, location of anterior intestine, mid-intestine and posterior intestine, respectively.

Fig. 7.

Zebrafish myh11 mutations alter vascular function. (A-F) Recordings of blood flow (dorsal aorta), heart rate and aortic diameter in live 4-dpf larvae injected with control and smooth muscle actin (SMA; acta2) morpholinos (MO) (A-C). (D-F) Identical recordings from uninfected 6-dpf wild-type larvae (WT) and sibling S237Y homozygous mutants, and heterozygous mlt larvae along with myh11-MO-injected homozygous mlt larvae. ****P<0.0001; ***P<0.001; **P<0.01; n.s., not significant. SMA MO (A-C) and mutant (D-F) experiments performed independently. Unpaired Student's t-test performed. Mean±s.e.m.

Fig. 8.

MYH11 variants detected via exome sequencing. (A) Bar graph depicting location and frequency of 682 MYH11 variants retrieved from the ExAc server; the location of the mlt (W512R), S237Y and L1287M mutations is indicated. Blue arrowheads point to the location of the Ala1241Thr and Val1296Ala variants, which are likely to be benign substitutions, and pink arrowheads point to the C7 and C9 variants in the SM2 allele. (B) Table listing the predicted and observed number of variants in different domains and regions of human MYH11. Ratio of observed to predicted listed in parentheses.