Conditional nmy-1 and nmy-2 alleles establish that non-muscle myosins are required for late C. elegans embryonic elongation

The elongation of C. elegans embryos allows examination of mechanical interactions between adjacent tissues. Muscle contractions during late elongation induce the remodelling of epidermal circumferential actin filaments through mechanotransduction. We investigated the possible role of the non-muscle myosins NMY-1 and NMY-2 in this process using nmy-1 and nmy-2 thermosensitive alleles. Our findings suggest these myosins act redundantly in late elongation, and that they are involved in the multi-step process of epidermal remodeling. When inactivated, NMY-1 was seen to form protein aggregates.

Morphogenesis refers to the process by which organisms gradually develop a characteristic 3D form.This can involve an increase in the number of cells, or a change in the configuration of their relative positions, which is accompanied by modifications in their cell-cell membrane adhesions (Lecuit, 2005).Multicellular organisms need to spatiotemporally coordinate the morphogenesis of multiple tissues in time and space (GILMOUR et al. 2017; GOODWIN AND NELSON 2021).It has become clear that communication through mechanical inputs plays a key role in ensuring the smooth development of adjacent tissues (AIGOUY et al. 2010;ZHANG et al. 2011;COLLINET et al. 2015;LYE et al. 2015).A classical paradigm is that a protein associated with a transmembrane receptor, such as integrin or E-cadherin, undergoes a conformational change that favours the binding of additional proteins.In turn, this can influence protein trafficking, the orientation of planar polarity, junction remodelling, cytoskeleton dynamics, or the translocation of transcription factors to the nucleus, to name a few (DEL RIO et al. 2009;AIGOUY et al. 2010;LE DUC et al. 2010;YONEMURA et al. 2010;LEVAYER et al. 2011;ZHANG et al. 2011;LARDENNOIS et al. 2019).
Although, several of the proteins relaying mechanical stress within a cell have been identified, we are far from a detailed understanding of all mechanotransductive pathways (MOORE et al. 2010;HU et al. 2017;YAP et al. 2018;NIETHAMMER 2021).In particular, tissue morphogenesis generally involves repeated mechanical inputs resulting in progressive shape changes (MARTIN et al. 2009;SOLON et al. 2009;AIGOUY et al. 2010;RAUZI et al. 2010;ZHANG et al. 2011;MAITRE et al. 2015;LARDENNOIS et al. 2019), but the mechanisms involved in stabilizing cell shapes are only beginning to be discovered.Recent results have emphasized the importance of permanent viscoplastic changes induced by repeated mechanical inputs and the key role of the actomyosin cortex (BONAKDAR et al. 2016;DOUBROVINSKI et al. 2017;KHALILGHARIBI et al. 2019;LARDENNOIS et al. 2019;STADDON et al. 2019; MOLNAR AND LABOUESSE 2021).
C. elegans represents a powerful system to analyze the consequences of mechanical inputs.
During C. elegans embryonic morphogenesis, an ellipsoidal cell aggregate elongates fourfold into a vermiform shape (Priess & Hirsh, 1986).This occurs in two distinct stages and relies on epidermal cell shape change (Vuong-Brender et al., 2016).The mechanical input comes from the epidermal actomyosin cortex and from the muscles for the early and late stages of elongation, respectively (WILLIAMS AND WATERSTON 1994;PIEKNY et al. 2003;GALLY et al. 2009;ZHANG et al. 2011).
The early stage involves two distinct groups of epidermal cells, the lateral and dorsal/ventral cells.In the lateral cells, there is a greater concentration of non-muscle myosin and a disordered actin network (PIEKNY et al. 2003;GALLY et al. 2009).Contractions from these two zones flanking the embryo on either side provide the force from the one-fold to roughly 2-fold stages, at which point the muscles become active (VUONG-BRENDER et al. 2017).Equally important are the actin filaments in the dorsal/ventral cells, which are arranged circumferentially and make bundles of a few distinct filaments, which provide integrity along the body and cause the force generated by the lateral cells to be directed to the tips of the embryo (VUONG-BRENDER et al. 2017).
The late stage begins when the muscles are in place and begin to contract (WILLIAMS AND WATERSTON 1994), at which point the actomyosin cortex in the lateral cells has achieved a circumferential pattern as well (GILLARD et al. 2019).Muscles are arranged into four bands just underneath the epidermis to which they are tightly attached, such that muscle contractions induce deformation of epidermal cells (ZHANG AND LABOUESSE 2010;ZHANG et al. 2011) .In particular, muscle activity transiently bends the actin bundles beyond a critical angle (Fig. 1A), shown in vitro to induce severing by cofilin (MCCULLOUGH et al. 2011); in C. elegans the severing proteins are villin and gelsolin instead (LARDENNOIS et al. 2019).Actin filaments are then restabilized by the p21-activated kinase PAK-1, the alpha-spectrin SPC-1 and the atypical formin FHOD-1 (Fig. 1A) (LARDENNOIS et al. 2019).In the absence of PAK-1 and SPC-1, or FHOD-1 and SPC-1, embryos elongate up to the 1.5-fold stage and then regress to their initial lima bean shape (roughly 1.2-fold stage) due to the loss of actin filament integrity (LARDENNOIS et al. 2019).Thereby muscles contribute to progressively shorten actin filaments, promote embryo elongation and decrease their diameter.
The model described above assumes that the ends of severed actin filaments do not get much further apart before they get restabilized by the PAK-1/SPC-1/FHOD-1 complex.However, this might not be the case since hydrostatic pressure exerts a radial force that could be expected to pull those actin ends if they are not held together.Since muscles are arranged orthogonally to actin bundles, they cannot contribute to bring actin ends closer.We thus wondered whether there might be additional players holding actin filaments once severed, and/or bringing them closer before handing them over to PAK-1/SPC-1/FHOD-1 (Fig. 1).
Although spectrins are actin-binding proteins that could fulfil the ascribed function of holding actin ends (LENNE et al. 2000;LAW et al. 2003; CHOI AND WEIS 2011), we considered the possibility that additional proteins could be required.We focused on non-muscle myosins, because they are obvious motor proteins working on actin.Non-muscle myosins are heterohexamers consisting of two myosin heavy chains, two regulatory light chains (MLC-4 in C. elegans), which must be phosphorylated to enable motor activity, and two essential light chains (MLC-5 in C. elegans) (Vicente-Manzanares et al., 2009).Importantly, several of those hexamers can assemble in a multi-subunit complex of opposite polarity through the C-terminal coiled-coil of the heavy chain.C. elegans has two non-muscle heavy chains, NMY-1 and NMY-2.NMY-1 is expressed at the time of elongation, is partially required for elongation and is essential for fertility (PIEKNY et al. 2003;KOVACEVIC et al. 2013).NMY-2 is expressed in early embryos, and is essential to establish early embryonic polarity and for cytokinesis (GUO AND KEMPHUES 1996;SHELTON et al. 1999;MUNRO et al. 2004;LIU et al. 2010).Both isoforms appear to act redundantly in embryonic morphogenesis, although the precise stage at which they are acting together has not been determined due to the genetic tools used at the time (PIEKNY et al. 2003).
To define more precisely how the two heavy chains act, we used temperature sensitive mutants to inactivate NMY-1/-2 at different stages.In addition, we combined them with other players involved in actin dynamics downstream of the muscle-induced actin severing.Our data establish that NMY-1 and NMY-2 are jointly required once muscles become active.We discuss their role compared to spectrins.

Strains and genetic methods
The C. elegans control strain N2 and other strains were maintained under standard conditions (Brenner 1974), and were propagated at 20°C unless noted otherwise.A complete list of strains and associated genotypes used in this study are included in Table S1.

Generation of a nmy-1 thermosensitive allele
We introduced a mutation in nmy-1 by CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) at the position corresponding to the allele nmy-2(ne3409ts) affecting changing Leucine-981 to Proline.The corresponding leucine is conserved among non-muscle myosins (see Fig. S1).It was generated using an oligonucleotide repair template (GAAACCGTCCGTGATCTCGAGGAGCAACTCGAGCAAGAtGAACAAGCTAGACAG AAACTGCTTccGGATAAGACGAATGTTGACCAGAGACTTCGAAACCTGGAAGAGC G) carrying two mutations, one to create a non-functional PAM (protospacer adjacent motif) site (AGG to ATG), and one to introduce the desired mutation (TTG to CCG).We co-injected the plasmid encoding Cas9 (CRISPR-associated endonuclease 9) and the sgRNA (singlemolecule guide RNA) for nmy-1 (5'GAGGAGCAACTCGAGCAAG) at 50 ng/µl, the nmy-1 oligonucleotide repair template at 20 ng/µl, along with the plasmids pRF4 and pBSK2 each at 100 ng/µl in the strain ML2540 (10.7554/eLife.23866)carrying a CRISPR-generated NMY-1::GFP fusion, so as to track the putative mutant protein.We subsequently picked 88 Roller animals and screened by PCR for the presence of the mutation using the primers 5'TCAAGCTCACCGCTTTAATTATGAAC and 5'CCCATTTTCTCGGCCAAGTGATCT to find one positive hit that could be recovered, which we named nmy-1(mc90ts).The allele nmy-1(mc90ts) had two mutations, as verified by sequencing the nmy-1 locus from homozygous animals, one corresponding to the desired L>P change and one corresponding to the PAM mutation, such that the protein sequence 959-QEEQARQKLLL became 959-QDEQARQKLLP.Homozygous nmy-1(mc90ts) animals were sterile at 25°C but not at 15°C (see results).We assume that thermosensitivity was due to the Leu to Pro change rather than to the upstream synonymous Glu to Asp change at the PAM site in the repair template.

Hatch count and arrest count protocols
Strains carrying a theremosensitive mutation were stored at 15˚C on agar plates.Embryos at different developmental stages on the plates were removed with an eyelash tool, washed in M9, and put on a 5% agarose pad with M9 on a slide sealed in paraffin oil.The slide was then imaged using a Roper Scientific spinning disk system with an immersion oil 40x objective (Zeiss Axio Observer Z1 microscope, Yokogawa CSUX1-A1 spinning disk confocal head, Photometrics Evolve 512 EMCCD camera, Metamorph software).Images were initially taken at 20˚C until the incubation chamber attached to the microscope was set to 25˚C, and imaging was continued overnight (8 hrs).A count was then performed to determine how many embryos of a particular initial developmental stage could make it to hatching, and at what stage they had arrested, if so.

Fluorescence microscopy
Spinning disk fluorescence imaging was performed with a 63× or 100× oil-immersion objective, NA=1.4.The temperature of the microscopy room was maintained at 20°C.Images of embryos were acquired using time-lapse mode with a 110 ms exposure at intervals depending on the experiment.Laser power and exposure times were kept constant throughout the experiments for specific strains and their controls.Images of Fig. 1F were acquired with a LSM 980 Airyscan2 Zeiss confocal system equipped with 488 nm and 561 nm excitation laser lines and an oil immersion objective with NA=1.4.We performed a simultaneous acquisition for the green and red channels in order to get both stainings in the same time in confocal mode.The images were deconvolved with the Huygens software.

Image processing and quantification
All images were treated by first subtracting the background with a rolling ball radius of 50 pixel.Any stacks were projected using maximum intensity.In Fig. 4, the average displacement was obtained by tracking stable points of high intensity in the epidermis and calculating the displacement using the following equation: The images in Fig. 4 were three-plane projections from stacks taken with an inter-plane distance of 0.8 μm to capture the epidermis.The distribution of the signal was obtained by measuring the mean intensity and standard deviation of a square of 45X33 pixels inside the H1 cell of the lateral epidermis, distribution = σ/<I>.Images with an even distribution of signal (meaning no bright spots) have a low standard deviation σ, and therefore a low distribution.In the case of bright spots against a black background, the value of σ increases, and therefore so will the distribution.

The concentration of non-muscle myosins increases until the 2-fold stage
To better understand when non-muscle myosins are required during elongation, we examined the fluorescence of three CRISPR knockin strains marking the essential light chain MLC-5, NMY-2 and NMY-1.We imaged randomly picked MLC-5::GFP embryos and examined both their stage and the GFP fluorescence level.We observed that the fluorescence increased between the one-and 2-fold stages, and was followed by a slight plateau in the lateral cells, and a mostly steady value in the DV cells (Fig. 1B-C).Using an NMY-2::mKate; NMY-1::GFP double knockin strain, we observed that only NMY-2 started to decline after the beginning of lima bean stage, whereas NMY-1 increased after that stage (Fig. 1D-E).At higher magnification, we observed that NMY-2 was faint with no clear pattern in the lateral cells but formed aligned puncta in DV cells, whereas NMY-1 was enriched in the lateral epidermis cells as aligned puncta and colocalized in DV epidermal cells with NMY-2 puncta (Fig. 1F).These results indicate that the myosin population is not static.Furthermore, they are compatible with the possibility that the two non-muscle myosin motors could be acting redundantly with one another during late elongation.

Inhibition of NMY-1 and NMY-2 arrests late elongation
To test the functions of the two non-muscle myosin heavy chains, we used conditional alleles.
A temperature sensitive nmy-2 allele had previously been described, nmy-2(ne3409ts), which changes a conserved Leucine residue among heavy chains into a Proline, NMY-2(L981P) (LIU et al. 2010).We engineered an NMY-1(L970P) change by CRISPR at the position homologous to that of nmy-2(ne3409); the strategy also involved a change of E960 into D (see Methods and Fig. S1).The resulting allele, named nmy-1(mc90ts), induced a thermosensitive sterile phenotype but very little lethality (Table 1; Fig. 2A,2B-line 4).We assume that the sterility was due to the L970P mutation rather than to the synonymous E960D change, because the nmy-2(ne3409ts) at the homologous position is conditional, but cannot exclude additive effects of both changes.At the non-permissive temperature, the allele nmy-1(mc90ts) behaved like the previously known nmy-1 missense allele sb113, which induced <10% lethality and partial sterility, but was not as severe as the presumptive null allele sb115, which induced over 50% embryonic and larval lethality (PIEKNY et al. 2003).Since the mutant protein NMY-2(ne3409ts) unfolds and becomes very rapidly inactive when animals are shifted to 25°C (LIU et al. 2010), we expected the mutant protein NMY-1(mc90ts) to behave likewise upon a temperature shift.
Using these two conditional alleles, we investigated at which stage they are required during elongation by raising mutant embryos at 20°C and then shifting them to the non-permissive temperature (Fig. 2A).Homozygous nmy-2(ne3409ts) embryos displayed embryonic lethality when shifted at an early elongation stage, however they could elongate if shifted at or beyond the 2-fold stage (Fig. 2B-line8).By contrast, double nmy-2(ne3409ts); nmy-1(mc90ts) mutants displayed 100% embryonic lethality irrespective of the time at which they were shifted to 25°C (Fig. 2B-line10), but could in general hatch if maintained at 20°C (Fig. 2B-line9).
We further examined at which elongation stage non-muscle myosin mutants arrested.When nmy-2(ne3409ts) embryos were shifted to 25°C at the pre-comma stage, about 40% remained pre-comma and 40% made it to the comma stage (Fig. 2D).When shifted at the comma stage, 50% nmy-2(ne3409ts) progressed to the 2-fold stage.When double nmy-2(ne3409ts); nmy-1(mc90ts) mutants were shifted to 25°C, they immediately stopped elongation at the stage at which the temperature had been raised (Fig. 2C-D).Importantly, they did not regress to an earlier body morphology as we had observed in spc-1 pak-1 or fhod-1; spc-1 double mutants (Fig. 2C).These results show that NMY-2 is required during elongation, but not when muscles become required.Moreover, we conclude that NMY-1 and NMY-2 are continuously required for elongation at all stages, that they act redundantly, but that they are not required to maintain body shape.These observations are also consistent with the notion that both non-myosin mutant proteins become very rapidly inactive at 25°C.The arrest phenotype observed upon an early elongation shift resembles that observed in strong let-502, mlc-4 or mlc-5 deficient embryos, as had previously been reported (PIEKNY et al. 2003).However, in contrast to the arrest phenotype observed when nmy-2(ne3409ts); nmy-1(mc90ts) are shifted to 25°C at the 2-fold stage, let-502(sb92ts) embryos shifted to 25°C at the 2-fold stage did not arrest during elongation (DIOGON et al. 2007).One possibility could be that another kinase such as PAK-1 or MRCK-1 acts in parallel to LET-502/Rho-kinase to phosphorylate MLC-4/RMLC at that stage (GALLY et al. 2009).
Our results confirm and much extend an earlier study suggesting that the two non-muscle myosins NMY-1 and NMY-2 are redundantly required during embryonic elongation (PIEKNY et al. 2003).Those previous experiments involved RNA interference against nmy-2 in the background of the non-conditional and putative null alleles nmy-1(sb113) or nmy-1(sb115).As RNAi against nmy-2 had to be mild to allow cytokinesis, it did not enable the authors to define when NMY-1 and NMY-2 act during elongation.Our results establish that neither NMY-1 nor NMY-2 alone is required, but that they are constantly and redundantly required during late embryonic elongation, which depends on muscle input.Moreover, we conclude that NMY-2 has a more critical role during early elongation since all single nmy-2 mutants, but only a few single nmy-1 mutants, arrested when shifted to 25°C during early elongation.Consistent with these results, NMY-1, NMY-2, and the myosin essential light chain MLC-5 are expressed in the epidermis throughout elongation.
Nevertheless, these embryos could maintain normal muscle activity for several hours after the temperature shift.We thus wondered whether elongation could resume after an arrest.To test this possibility, we shifted nmy-2 (ne3409ts); nmy-1(mc90ts) embryos at 25°C for 45 min, using the incubator on the microscope to control the temperature, then then back to 15°C for 5hrs in a different incubator before imaging.As described in Fig. 3A, 60% of 2-fold nmy-2 (ne3409ts); nmy-1(mc90ts) embryos were able to hatch, as compared to 0% when left at 25°C.The resumption of elongation after arrest suggests that the temporary absence of NMY-1 and NMY-2 did not permanently damage any of the molecular-level components involved in this developmental stage.Finally, we investigated for how long could elongation be paused and then successfully restarted.To test this, we progressively increased the duration of the pause and found that the curve was sigmoidal, with a mid-height value of about 2,5 hours, and that it levelled off around 5 hours (Fig. 3B).

nmy-2 (ne3409ts); nmy-1(mc90ts) has normal muscle activity at 25°C
We considered two explanations to account for the fact that nmy-2(ne3409ts); nmy-1(mc90ts) embryos stop elongation when shifted to 25°C at the 2-fold stage.One possibility would be that these non-muscle myosins are required in muscles, for instance that in their absence muscles do not contract strongly enough or do not transmit properly the contractions to initiate actin filament bending and fail to trigger their reorganisation.Another possibility would be that they act in the epidermis to help reorganise actin filaments.
Analysis of NMY-1::GFP videograms showed that embryonic movements can be decomposed into two fundamental movements: rotation and twitching.The first was to follow the rotations of the cylindrical body within the eggshell of 90°, which is triggered by muscle activity (YANG 2017).The second was measured by tracking landmarks in the lateral epidermal cells inbetween body rotations (Fig. 4A, lower panels).We found that both control and double nonmuscle myosin mutants at 25°C could rotate on average every 12 seconds (Fig. 4A-B upper panels; Fig. 4D), and contract locally over 0.73 ± 0.06 µm/sec, as compared to the control nmy-1::gfp, 0.81 ± 0.06 µm/sec when placed at 25°C (Fig. 4E).By contrast the muscle deficient strain exhibited no rotations and contracted 0.21 ± 0.01 µm/sec (Fig. 4C-E).We conclude that the double mutant is not lacking the mechanical input from the muscles needed to drive elongation.Furthermore, because we measured the behaviour of the epidermis, the muscles are also properly transmitting forces to the neighbouring tissue.
The results described above thus suggest that both non-muscle myosins are required in the epidermis.Which function could non-muscle myosins perform in the epidermis?As recalled above (see introduction), we previously suggested based on genetic and imaging data that the repeated muscle contractions induce the severing and shortening of circumferential actin filaments (LARDENNOIS et al. 2019).We could not at the time define what happens with severed actin ends.One possibility is that the hydrostatic pressure building up during elongation could be expected to pull them apart.Although spectrin may keep actin filaments together, since spectrins are considered as springs and ß-spectrin can bind actin (LENNE et al. 2000;LAW et al. 2003; CHOI AND WEIS 2011), our data posit non-muscle myosins as perfect candidates to keep actin ends together.
Non-muscle myosins have two well-characterized activities, actin binding and actin pulling through their power stroke, which requires myosin regulatory light chain  phosphorylation.We considered the possibility that muscle contractions could locally activate MLC-4 and attempted to test this idea by examining a wild-type MLC-4::GFP marker in control embryos, but failed to record any obvious such event (data not shown).We did not directly test in which cells NMY-1 and NMY-2 are required.However, we note that when wild-type MLC-4 is expressed under a dorso-ventral epidermal promoter in mlc-4 null mutant embryos, the embryos elongate up to the 2.5-fold but do not reach full elongation (GALLY et al. 2009), indicating that MLC-4 activity, and hence NMY-1/2 activity, is most likely also required in dorso-ventral cells.In part because the precise organization and polarity of actin filaments within dorso-ventral actin bundles is unknown (COSTA et al. 1997), we cannot currently determine whether non-muscle myosins are only required to maintain actin ends together or if they pull on severed actin filaments to bring them closer before FHOD-1 activity.If actin filaments all have the same polarity, it would tend to exclude a pulling function, since pulling of actin filaments by non-muscle myosin relies on filaments of opposite polarity.One strategy to discriminate between both models would be to introduce a secondary mutation in NMY-1 preventing ATP binding (OSORIO et al. 2019), which is beyond the scope of this study.

Inactivated NMY-1 forms aggregates
During the course of the experiments described above, we noticed that the NMY-1(mc90ts)::GFP protein aggregated in the epidermis at the non-permissive temperature (Fig. 5A).The distribution of these aggregates was quantified in the lateral cells (see Methods), whereby a large value of the ratio between the standard deviation σ over the mean intensity (<I>) indicated a less uniform distribution.Using this approach, we measured a value of 0.5 ± 0.1 at 25˚C for control NMY-1::GFP embryos, the most uniform distribution of all the strains.
We also investigated the long-term fate of these aggregates and found that after the initial inactivation, the many small clusters began to fuse for roughly 1.5 hours, until a sudden transition occurred within 3 minutes.At that point, the aggregates disappeared in a highly coordinated fashion.Note that the images used for the quantification of the distribution of the aggregates were taken on average one hour after the temperature shift had occurred, and thus correspond to roughly the half-point in the time-evolution of the particles.Our findings are consistent with the literature, indicating that non-functional non-muscle myosins in both C. elegans oocytes, and human thrombocytes tend to aggregate (ALTHAUS AND GREINACHER 2009; SUN et al. 2020).

Conclusion
Altogether, our data identify the presence of both NMY-1 and NMY-2 in the epidermis during late elongation.We found that either NMY-1 or NMY-2 can support late elongation but that if both are absent elongation beyond the 2-fold stage failed.Not only did their combined absence cause a hatching failure, but the arrest in development was immediate, which is also the case for earlier developmental stages in the embryo.Importantly, this arrest could not be attributed to less muscle movement, and thus to abnormal mechanotransduction in the epidermis.We found that the arrest could be reversed by returning the embryos to < 20˚C, indicating that the absence of both non-muscle myosin heavy chains did not permanently affect actin integrity, nor any other epidermal structure.Our data are compatible with the possibility that this myosin pair acts to reduce the length of the epidermal actin filaments by holding or pulling the severed ends together.

LEGENDS
Figure 1.The two non-muscle myosins are present in the epidermis throughout morphogenesis (A) Cross-section through an embryo illustrating the main anatomical features of the embryo: the dorsal and ventral epidermis (pink, red), the lateral epidermis (yellow) and the muscles (orange); other tissues not represented for clarity.Muscles are attached to the apical extracellular matrix (not illustrated) surrounding the embryo through hemidesmosomes in the epidermis (blue dots).Actin filaments (green) run circumferentially as bundles.The current model holds that in a typical contraction cycle, muscles will locally displace the actin filaments (black arrows in 2 nd drawing), which will trigger their severing (3 rd drawing), before they eventually get stabilized by a complex of actin-binding proteins (dark green stars).At the end of the cycle, when the underlying muscles relax, the embryo has become thinner and longer (4 th drawing).It is unclear whether the actin-binding complex can hold actin ends to counter the hydrostatic pressure (blue arrows in 3 rd drawing) if they drift apart.(B) Fluorescence micrographs showing the distribution of the essential myosin light chain MLC-5 (marked by a GFP knockin) at three different stages: lima bean, 2-fold, 3-fold.Note that MLC-5 is enriched in the lateral seam cells (yellow arrow).(C) Quantification of MLC-5::GFP fluorescence over time in the lateral and dorso-ventral epidermis.(D) Fluorescence micrographs showing the distribution of the two large non-muscle myosin chains NMY-1 (marked by a GFP knockin) and NMY-2 (marked by a mCherry knockin) at different stages; timing starts approximately 125 minutes after the 1-cell stage (black line in E; see video1).(E) Quantification of NMY-1::GFP and NMY-2::mKate fluorescence starting at the lima bean stage.Note the lack of a clear myosin cable along cell contours in the 2-fold embryo (+300 min image).(F) Deconvoluted confocal micrographs of NMY-1::GFP and NMY-2::mKate in the lateral seam (yellow box)and dorso-ventral (pink box) epidermis of a 3-fold embryo.Higher magnification of the area marked by a dotted rectangle are shown on the left for seam cells (three consecutive focal planes; 1.75x magnification), or in the top right corner for the dorso-ventral epidermis (4.5x magnification) revealing short circumferential alignments.Scale bars, 10 µm (B, D, F), except 4 µm for the enlargements.Note that NMY-1 and NMY-2 form parallel bands, especially in the dorso-ventral epidermis, but outside the area located above muscles.