Regional differences in actomyosin contraction shape the primary vesicles in the embryonic chicken brain

• Published in: Physical biology 2012-12, Vol.9 (6), p.066007-18
• Main Authors: Filas, Benjamen A, Oltean, Alina, Majidi, Shabnam, Bayly, Philip V, Beebe, David C, Taber, Larry A
• Format: Article
• Language: English
• Subjects: Actomyosin - analysis; Actomyosin - metabolism; Actomyosin - ultrastructure; Animals; biomechanics; Brain - cytology; Brain - embryology; Brain - metabolism; Brain - ultrastructure; brain development; Cell Shape; Chick Embryo - cytology; Chick Embryo - embryology; Chick Embryo - metabolism; Chick Embryo - ultrastructure; chicken embryo; computational modeling; cytoskeletal contraction; Models, Biological; morphogenesis
• ISSN: 1478-3975; 1478-3967; 1478-3975
• DOI: 10.1088/1478-3975/9/6/066007

In the early embryo, the brain initially forms as a relatively straight, cylindrical epithelial tube composed of neural stem cells. The brain tube then divides into three primary vesicles (forebrain, midbrain, hindbrain), as well as a series of bulges (rhombomeres) in the hindbrain. The boundaries between these subdivisions have been well studied as regions of differential gene expression, but the morphogenetic mechanisms that generate these constrictions are not well understood. Here, we show that regional variations in actomyosin-based contractility play a major role in vesicle formation in the embryonic chicken brain. In particular, boundaries did not form in brains exposed to the nonmuscle myosin II inhibitor blebbistatin, whereas increasing contractile force using calyculin or ATP deepened boundaries considerably. Tissue staining showed that contraction likely occurs at the inner part of the wall, as F-actin and phosphorylated myosin are concentrated at the apical side. However, relatively little actin and myosin was found in rhombomere boundaries. To determine the specific physical mechanisms that drive vesicle formation, we developed a finite-element model for the brain tube. Regional apical contraction was simulated in the model, with contractile anisotropy and strength estimated from contractile protein distributions and measurements of cell shapes. The model shows that a combination of circumferential contraction in the boundary regions and relatively isotropic contraction between boundaries can generate realistic morphologies for the primary vesicles. In contrast, rhombomere formation likely involves longitudinal contraction between boundaries. Further simulations suggest that these different mechanisms are dictated by regional differences in initial morphology and the need to withstand cerebrospinal fluid pressure. This study provides a new understanding of early brain morphogenesis.