XB-ART-37008Mech Dev March 1, 2008; 125 (3-4): 353-72.
H,K-ATPase protein localization and Kir4.1 function reveal concordance of three axes during early determination of left-right asymmetry.
Consistent laterality is a fascinating problem, and study of the Xenopus embryo has led to molecular characterization of extremely early steps in left-right patterning: bioelectrical signals produced by ion pumps functioning upstream of asymmetric gene expression. Here, we reveal a number of novel aspects of the H+/K+-ATPase module in chick and frog embryos. Maternal H+/K+-ATPase subunits are asymmetrically localized along the left-right, dorso-ventral, and animal-vegetal axes during the first cleavage stages, in a process dependent on cytoskeletal organization. Using a reporter domain fused to molecular motors, we show that the cytoskeleton of the early frog embryo can provide asymmetric, directional information for subcellular transport along all three axes. Moreover, we show that the Kir4.1 potassium channel, while symmetrically expressed in a dynamic fashion during early cleavages, is required for normal LR asymmetry of frog embryos. Thus, Kir4.1 is an ideal candidate for the K+ ion exit path needed to allow the electroneutral H+/K+-ATPase to generate voltage gradients. In the chick embryo, we show that H+/K+-ATPase and Kir4.1 are expressed in the primitive streak, and that the known requirement for H+/K+-ATPase function in chick asymmetry does not function through effects on the circumferential expression pattern of Connexin43. These data provide details crucial for the mechanistic modeling of the physiological events linking subcellular processes to large-scale patterning and suggest a model where the early cytoskeleton sets up asymmetric ion flux along the left-right axis as a system of planar polarity functioning orthogonal to the apical-basal polarity of the early blastomeres.
PubMed ID: 18160269
PMC ID: PMC2346612
Article link: Mech Dev
Genes referenced: actl6a atp1a1 gja1 kcnj10 kcnj2
Article Images: [+] show captions
|Fig. 1. H+/K+-ATPase is dynamically and asymmetrically localized in the early frog embryo. mRNA encoding the Î± subunit of the H+/K+-ATPase is asymmetrically localized at the 2-cell stage (A) and revealed to be right-sided by the 4-cell stage (B) although significant variability of in situ signal was observed in embryos from different females. Antibodies developed to the Xenopus H+/K+-ATPaseÎ± subunit give a single clean band of the predicted size in Western blots (C) and were used to examine protein localization (which does not always match mRNA because of maternal contributions and post-transcriptional regulation) by immunohistochemistry using alkaline-phosphatase chromogenic detection on sections (purple signal). â No primaryâ negative controls gave the predicted lack of signal (D). Localization of H+/K+-ATPaseÎ± subunit protein was detected circumferentially (E) in a region at the vegetal pole (F) of unfertilized eggs. At fertilization, a specific spot of H+/K+-ATPaseÎ± protein was detected at 180Â° opposite the sperm entry point (blue arrow, G), and H+/K+-ATPaseÎ± protein became spread out towards the anterior pole, then occupying more than half of the embryo (H). At the 2-cell stage, asymmetric localization of the H+/K+-ATPaseÎ± subunit was detected at the cell membrane on sections taken through the animal hemisphere (I) and throughout the cytoplasm in sections taken through the vegetal hemisphere (J). By the 4-cell stage, the localization was confined to the right ventral hemisphere (K, taken more vegetally, and L, taken more animally). The H+/K+-ATPaseÎ² subunit was likewise asymmetric at the 2-cell stage (M). The asymmetric localization was abolished (bilateral cell membrane staining) by exposure to the actin disruptor Latrunculin (N). In contrast, the microtubule disruptor Nocodazole did not interfere with asymmetric localization (O), but did loosen the association of the protein with the membrane. Section O had to be taken in a less apical plane than sections for untreated embryos, because these drugs interfered with the normal animal-pole localization of these proteins. In addition to the localization along the LR and dorso-ventral axes, we detected dynamic localization of the H+/K+-ATPase along the animalâ vegetal axis. While the H+/K+-ATPaseÎ± protein is tightly localized to the vegetal cortex before fertilization (P), the band of protein rises towards the animal pole after fertilization (Q) and ultimately settles under the animal pole surface (R). This localization was not disrupted by Blebbistatin, an inhibitor of myosin (S), but was abolished by Nocodazole (T). Inhibition of Dynein (U) or disruption of the actin cytoskeleton (V) had no effect on the normal animal-pole localization of H+/K+-ATPaseÎ± subunit. Note that the embryos in (T)â (V) were at the same age (equivalent to the 2-cell stage) as those in R: the cell cleavage plane is not apparent in these embryos because the efficient disruption of the cytoskeleton and motor protein machinery interferes with cytokinesis. Red arrows indicate positive signal; white arrows indicate lack of signal. Yellow ovals in each panel indicate plane of section, as illustrated in (W); orientation of sections taken perpendicular to the AV axis is shown in (Wâ ²).|
|Fig. 4. Localization and function of Kir4.1 in the frog embryo. Using antibodies to Kir4.1 that reveal clean bands of the predicted size (corresponding to single subunits and dimers) (A), immunohistochemistry revealed that the unfertilized egg localized Kir4.1 protein throughout the vegetal half, excluding a sickle-shaped (bowl shaped, in three dimensions) area under the animal pole (B). In contrast, after fertilization, the protein is located throughout the animal hemisphere (C). At the 2-cell stage, Kir4.1 protein localization exhibits a range of patterns consistent with finger-like projections which can be observed at different stages of a movement from an equatorial pool towards the animal pole (D–F). The patterns are suggestive of a movement that bends to follow the contour of the animal pole edge and then returns to the equatorial pool, forming triangle-like areas including at the cell membrane (F–H). At the 2-cell stage, localization is symmetrical along the LR axis and results in Kir4.1 presence at the cell cortex in sections taken at the animal pole (I). Sections taken in the vegetal hemisphere exhibit a more complex localization, being present throughout the cytoplasm of the ventral cells and in rod-like patterns in the dorsal cells (blue arrows, J). The normal sequence of animal pole-directed movement of Kir4.1 protein (B–H) was disrupted by manipulation of the cytoplasmic transport machinery. Inhibition of myosin by Blebbistatin often introduced an asymmetry along the LR axis and abolished the appearance of well-defined fingers (L). Depolarization of microtubules somewhat disrupted the well-formed pattern (M). Inhibition of dynein motor protein activity and disruption of the actin cytoskeleton resulted in much more drastic changes in the normal pattern, resulting in most of the Kir4.1 protein localizing to the animal pole in bulk (N and O). Red arrows indicate positive signal. Yellow ovals in each panel indicate plane of section, as indicated in (K); orientation of sections taken perpendicular to the AV axis is shown in (K′). Functional roles for Kir4.1 in left–right patterning were tested by misexpression of a dominant negative construct designed to trap endogenous Kir4.1 protein subunits in the ER at the 1-cell stage, and scoring laterality of the heart, gut, and gall-bladder at st. 43. The raw data are shown in Table 2. In contrast to control (uninjected) embryos and embryos injected with Kir2.1-ER mRNA, all of which exhibited >97% normal laterality (P), embryos injected with Kir4.1-ER mRNA exhibited 25% incidence of independent randomization of the three organs assayed (heterotaxia), including reversals of the heart (Q), and gut and gall-bladder (R). Red arrowheads indicate apex of the heart; yellow arrowheads indicate direction of gut coil vertex; green arrowhead indicates position of the gall bladder.|
References [+] :
Abe, Microtubule defects and cell morphogenesis in the lefty1lefty2 tubulin mutant of Arabidopsis thaliana. 2004, Pubmed