Convergent Extension Drives Elongation in Drosophila
Elongation in Drosophila Occurs Primarily by Junctional Remodeling
In Drosophila, cells in the blastoderm acquire both anteroposterior (AP) and dor- soventral (DV) patterning prior to germband elongation. The elongating field of cells occupies almost the entire egg, along both the AP and DV axes, and the cells are relatively uniform in size and shape. During elongation, rows of cells along the DV axis simultaneously intercalate, collapsing the cell field along the DV axis and lengthening it along the AP axis (Figure 3.8A; Irvine and Wieschaus, 1994). This intercalation is the primary mode of elongation, although oriented cell division occurs in the posterior of the embryo during late elongation (da Silva and Vincent, 2007). Recently, basal-lateral protrusions have been discovered to be required as an additional and, to some degree, independent mechanism for germband elongation (Sun et al., 2017). Prior to junctional remodeling at the apical surface of the cells, basolateral protrusions arise in DV cells. These actin-rich protrusions require Rac (a Rho GTPase) for DV cells to actively migrate toward and sometimes past one another as they intercalate. Blocking apical junctional remodeling does not halt basolateral protrusions and when protrusions are blocked alone, germband extension is compromised, suggesting that the tw'o mechanisms operate in parallel but somewhat independently to extend the germband (Sun et al., 2017).
Intra- and Intercellular Effectors of Cell Movements Are Polarized in Drosophila
In Drosophila, the contractile cell intercalation driving elongation apically depends on the asymmetrical distribution of proteins associated with adherens junctions: myosin II and F-actin concentrate between anterior and posterior faces of adjacent cells, while DE-cadherin, armadillo/(3-catenin, and Bazooka/PAR-3 concentrate in the DV interfaces. This polarization within cells is required for differential junctional remodeling between neighboring cells: membrane in the DV axis is reduced, while membrane contacts along the AP axis increase in a fashion that depends directly on myosin II (Bertet et al., 2004; Zallen and Wieschaus, 2004). Differential junctional remodeling results in active cell-cell intercalation along the DV axis, which can be coordinated between quartets of cells or larger multicellular rosettes (Bertet et al., 2004; Blankenship et al., 2006).
FIGURE 3.8 Comparison of cell interaction in Tribolium versus Drosophila shows less ordered cell movements. A-D. Traced cells in the extending Drosophila germband move almost in register together along the DV axis and spread in a correspondingly equidistant way along the AP axis. E-H. Traced cells in the extending Tribolium germband, while clearly intercalating, show unequal movement in both the DV and AP axes. (Panels are modified from Irvine and Wieschaus, 1994, and Benton, 2018.)
Pair-Rule Genes Drive Periodic Expression of the Toll Receptors Required for Convergent Extension
Germband elongation relies on inputs from the segmentation genes that pattern the AP axis, specifically, the pair-rule genes. In Drosophila, pair-rule genes form a series of stripes alternating along the AP body axis; these genes provide positional information to intercalating cells. Pair-rule gene mutants (even-skipped [eve], runt, odd paired [odd]) lose polarized distribution of intracellular effector molecules that drive cell intercalation (Zallen and Wieschaus, 2004; Blankenship et al., 2006; Simoes et ah, 2010) and fail to elongate normally (Irvine and Wieschaus, 1994; Blankenship et ah, 2006). Conversely, mutants that ectopically express the pair-rule genes eve and runt cause local reorientation of the polarized effector molecules (Irvine and Wieschaus, 1994; Zallen and Wieschaus, 2004). Pare et ah (2014) identified three Toll family receptors as downstream targets of the pair-rule genes during embryo elongation by comparing the transcriptomes of mutant embryos lacking the pair-rule genes eve and runt to wild-type Drosophila embryos. These leucine-rich receptors are differentially expressed in stripes of cells along the AP axis. Based on Toll mutant phenotypes, they propose a model in which heterotypic binding between adjacent stripes drives the differential accumulation of intracellular effector molecules, e.g., MyoII, to those cell interfaces (Pare et ah, 2014). There is evidence that basolateral protrusions also depend on proper AP axial patterning since basolateral rosette formation was suppressed in eve and Toll RNAi embryos (Sun et ah, 2017).