«Left-Right Asymmetry in Animal Embryogenesis Michael Levin Cell Biology dept. Bldg. C1, rm. 403 Harvard Medical School 200 Longwood Ave. Boston, MA ...»
in Animal Embryogenesis
Cell Biology dept.
Bldg. C1, rm. 403
Harvard Medical School
200 Longwood Ave.
Boston, MA 02115
Tel. (617) 432-0044
Fax. (617) 432-1144
The geometrical invariance known as symmetry is a striking feature of developmental
morphology during embryogenesis. There are several types, such as translational symmetry
(repeated units such millipede segments), and reflectional symmetry (two or more sections of an organism looking the same to some level of detail on either side of a symmetry line). Animal bodyplans occur in a wide variety of symmetries (see Figure 1). Vertebrates have a generally bilaterallysymmetrical body-plan, but this symmetry is broken further into a pseudo-symmetry by the consistently asymmetric placement of various internal organs such as the heart, liver, spleen, and gut, or an asymmetric development of paired organs (such as brain hemispheres or lungs).
Symmetries are repeatedly broken during development. For example, the radial symmetry of the early chick blastoderm (see Fig. 2) is broken into a bilateral symmetry by the appearance of Köhler's sickle and then the primitive streak. This is further broken into a definitive pseudosymmetry by the right-sided looping of the heart tube. In contrast, the sea-urchin develops from a bilaterally-symmetric larva into an adult with a five-fold radial symmetry.
Arguably, the most interesting asymmetry in vertebrate development is that along the leftright (LR) axis. I limit this discussion to include only invariant (i.e., consistent among all normal individuals of a given type) differences between the left and right sides of an animal's morphology.
This specifically exclude pseudo-random characteristics such as animal coat colors, and minor stochastic deviations due to developmental noise.
The LR axis itself follows automatically from the definition of the AP and DV axes, as it is perpendicular to both; however, consistently imposed asymmetry across it is fundamentally different from patterning along the other two axes. Firstly, while the AP and DV axes can be set by exogenous cues such as gravity, or sperm entry point, there is no independent way to pick out the left (or right) direction, since no obvious macroscopic aspect of nature differentiates left from right.
Secondly, all normal members of a given species are asymmetrical in the same direction. However, animals with complete mirror reversal of internal organs can arise (situs inversus) and are otherwise phenotypically unimpaired. Thus, while it is possible to come up with plausible evolutionary reasons for why organisms might be asymmetric in the first place (optimal packing of viscera, etc.), there is no obvious reason for why they should all be asymmetric in the same direction. It is, after all, much easier to imagine a developmental mechanism for generating asymmetry (such as positive-feedback and amplification of stochastic biochemical differences) than for biasing it to a given direction. The left-right axis thus presents several unique and deeply interesting theoretical issues.
A priori, one can imagine several ways to generate consistent LR asymmetry in an embryo.
One way would be to orient the embryo within the mother organism (Fig. 3A); thus the asymmetry would derive directly from LR-asymmetric influences applied by the already asymmetric maternal organism. While this might be plausible in mammals, the consistent asymmetry in free-developing organisms argues against the necessity for this kind of mechanism. Another possibility is the generation of prepattern in the egg (Fig. 3B). Thus, if the oocyte was asymmetrically loaded with determinants by the maternal ovary, these determinants could then go on to elaborate the LR asymmetry during later development. However, the regulative nature of development argues against this mechanism; for example, blastomeres of mouse embryos can be scrambled, split, or added to (i.e., 2 embryos made to aggregate together) and still result in phenotypically normal organisms.
Another interesting mechanism makes use of a fundamental force of physics to orient the LR axis relative to the other two axes (Fig. 3C). Huxley and deBeer 1 proposed that LR asymmetry was oriented during embryonic development by an electric current running down the length of the notochord, which would generate a magnetic field pointing R or L, if measured at the dorsal or ventral sides. There is, however, no good evidence for such a mechanism. The most plausible mechanism for generating left-right asymmetry makes use of molecular chirality (Fig. 3D). If a chiral molecule (shown as a 2-dimensional “F”, after Brown and Wolpert, 1990 2) has a directional activity (such as transport of subcellular components, or nucleation of directed microtubules) and is oriented within the cell relative to the other 2 axes, it can generate a consistent asymmetry.
Whatever mechanism initially differentiates L from R represents the first basic step of LR asymmetry (Fig. 4). The next step consists of elaborating this information into multi-cellular fields of asymmetric gene expression. As will be discussed below, there has been identified a cascade of genes which are expressed only on the left or the right side of the body, and regulate (turn on or off) each other’s expression. In the final step, tissues which are forming asymmetric organs such as hearts and stomachs take cues from asymmetrically expressed genes and undergo sided morphogenesis. Several labs have made significant progress in working out the details of the regulatory interactions between the known asymmetric genes, and elucidating mechanisms by which organ primordia interact with such genes; thus, we are well on our way to understanding stages 2 and 3. The most fascinating questions however concern stage 1, and are still almost completely open. These will be touched upon at the end of this paper.
Besides the intrinsic interest to those working on fundamental morphogenetic mechanisms, LR asymmetry is also relevant to medical considerations of several fairly common human birth defects: syndromes as Kartagener's and Ivemark's 3, dextrocardia, situs inversus (a complete mirror-image reversal of the sidedness of asymmetrically positioned organs and asymmetric paired organs), heterotaxia (where each organ makes an independent decision as to its situs), and right or left isomerism (where the organism is completely symmetrical, for example, polysplenia or asplenia). Of these, only the complete (and rare) situs inversus is not associated with physiological difficulties. The rest, especially heterotaxia, often result in serious health problems for the patient.
Pre-molecular data While molecular mechanisms underlying antero-posterior and dorso-ventral asymmetry have been studied in detail, the mechanistic basis for LR asymmetry was, until recently, completely unknown. The bilateral body plan is thought to have originated with the eumetazoa. The LR axis is specified after the anterior-posterior (AP) and dorso-ventral (DV) axes, and is determined with respect to them 4,5. Currently, several morphological markers of LR asymmetry are apparent in vertebrates: heart, direction of embryo rotation, gut, liver, lungs, etc. The organs possessing asymmetries, as well as the direction of their asymmetry, are evolutionarily well conserved. The heart is asymmetrically located in the mollusks 6; the situs of the stomach and the liver 7 is the same among fish, reptiles, birds, and mammals.
Neville 8 presents an extensive and fascinating survey of various animal asymmetries.
Besides the above-mentioned internal organs, beetles consistently fold one wing under the other, many crustaceans have specialized right and left fore-limbs, some flatfish consistently settle on and undergo eye migration to one side, and there is even a species of parasite which lives only on one side of host shrimp. Meanwhile, there has been little information shedding light on the mechanisms determining the sidedness of the asymmetries. Selection for LR asymmetries in Drosophila, in hopes of generating a genetically-tractable mutant, failed 9.
Several experiments have shed light on the timing of LR asymmetry specification. Chick heart sidedness has been experimentally demonstrated to be determined during gastrulation 10;
studies on LR inversions induced by drugs likewise suggest that in mammals, a critical period in LR biasing occurs before late gastrulation 11. Thus it is clear that decisions fundamental to LR asymmetry are made long before any overt signs of morphological asymmetry, and long before the morphogenesis of asymmetric organs.
Several kinds of mollusks undergo spiral cleavage and secrete an exoskeleton shaped like a conical spiral. In 3D space, such spirals can have two possible variants: a left-handed and a righthanded helix (which are otherwise identical). Each particular species of snail has invariant (consistent) chirality, but there are species which utilize each type of coiling. Murray and Clarke 12 found that the direction of coiling of P. suturalis is maternally inherited and sinistrality is dominant to dextrality. Freeman and Lundelius 13, studying a different species, found that dextrality is dominant; interestingly, the dextral gene apparently functions via a cytoplasmic product since it is possible to transfer (by micro-pipette) cytoplasm from the dextral variant of the snail into the sinistral variety, and rescue the dextral coiling phenotype. The biochemical nature of this activity has not yet been identified.
There is a variety of drugs which cause defects in a LR-asymmetric manner or randomize asymmetry (Table 1). These form a basically unrelated group, which includes even such simple substances as cadmium. The drugs which cause worse limb defects on one side were suggested 14 to be due to a differential blood supply to the two limbs (due to asymmetry in blood vessels exiting the heart). This is made somewhat unlikely by the fact that cadmium causes opposite-sided defects in rats and mice 15,16, while cardiac anatomy and relative vessel size of both species are extremely similar. This suggests a fundamental difference between left and right limbs. The pharmacology of these drugs has not yet suggested anything about the normal mechanisms of LR patterning, except that an adrenergic pathway may be involved 11.
Several mammalian mutants are known which display either defects in basic LR patterning or phenotypes which differentially affect the left or right sides of the body (Table 2). For example, iv 17 results in racemic offspring (50% being phenotypically situs inversus), while inv 18 mice have 100% of the offspring showing mirror image inversions of the internal organs (although in the context of other heterotaxia-like phenotypes, 19). Mutants such as legless 20 exhibit limb phenotypes which are more pronounced on one side of the body. In crosses with iv, the side affected is shown to reverse with the organ situs.
Regulatory cascade of asymmetric gene expression A number of asymmetrically-expressed genes have now been described (see Table 3).
These include a variety of signaling molecules and transcription factors. Figure 5 illustrates the expression pattern of three such genes as assayed by in situ hybridization with riboprobes to the relevant genes: Sonic Hedgehog (Shh), Nodal, and PTC 21. Beginning with the studies of Levin et al. 21, it was discovered that Shh is expressed only on the left side of Hensen’s node in the gastrulating chick embryo (Figure 5A). Shortly thereafter, Nodal and PTC are expressed also on the left side.
Once a set of asymmetrically-expressed genes was identified, their location and relative timing of expression suggested a possible pathway of sequential inductions and repressions.
Using artificial retroviruses bearing the gene of interest and protein-coated beads, a pathway was constructed. For example, it was found that misexpressing the normally left-sided gene Shh on the right side caused the ectopic right-sided expression of Nodal, which is normally also confined to the left side. This cascade (summarized in Figure 6) begins when activin βB becomes expressed on the right side of Hensen's node (st. 3). This soon induces the expression of cAct-RIIa in the right side, and shuts off the right-side expression of Shh (which was previously expressed throughout the node). Soon thereafter, Shh (which at that point is expressed only on the left side of the node and in the notochord) induces nodal in a small domain of cells adjacent to the left side of the node.
This is soon followed by a much larger domain in the lateral plate mesoderm.
Most importantly, the early asymmetrically-expressed genes are not merely markers of inherent laterality, but play an active role in LR patterning. Misexpression of activin or Shh (which result in missing or bilateral nodal expression respectively) specifically randomize heart situs in the chick 21. Moreover, nodal, which is in direct contact with cardiac precursor cells, can reverse heart situs or cause symmetric hearts 22. Thus, though there is no consensus on what causes cardiac looping in the first place, it is plausible that nodal is instructing heart looping by providing an asymmetric signal to one side of the cardiac primordia, and affecting the proliferation, migration, or cytoskeletal organization of cardiac precursors. The fact that morphologically normal hearts form in the absence of Shh and nodal expression (albeit with randomization of heart situs) indicates that the genes in this cascade are neither responsible for inducing heart formation nor for instructing its morphogenesis. Rather, they seem to provide a pivotal influence determining the handedness of the heart. Interestingly, the other organs besides the heart likewise take their cues from this genetic cascade 22.
Laterality Disturbances in Twins and the Midline Barrier The identification and characterization of several players in LR patterning has enabled models explaining the finding that conjoined twins of armadillo 23, fish 24, frog 25, and man 3,26, often exhibit alterations of situs in one of the twins. As early as 1919, Spemann and Falkenberg 27 reported that producing conjoined twins by tying a hair between the two blastomeres of amphibian eggs results in situs inversus in one of the twins. Levin et al. (1995) suggest that an explanation for the association of laterality defects and twinning might be found in consideration of interactions between signaling molecules in two closely aligned primitive streaks.