Par-Cteno-Genesis or Cteno Par-Genesis

In bilaterians and cnidarians, embryonic and epithelial cell-polarity are regulated by the interactions between Par proteins, Wnt/PCP signaling pathway, and cell-cell adhesion. Par proteins are highly conserved across Metazoa, including ctenophores. But strikingly, ctenophore genomes lack components of the Wnt/PCP pathway and cell-cell adhesion complexes; raising the question if ctenophore cells are polarized by mechanisms involving Par proteins. Here, by using immunohistochemistry and live-cell imaging overexpression of specific mRNAs, we describe for the first time the subcellular localization of selected Par proteins in blastomeres and epithelial cells during the embryogenesis of the ctenophore Mnemiopsis leidyi. We show that these proteins distribute differently compared to what has been described for other animals. This differential localization might be related with the emergence of different junctional complexes during Metazoa evolution. Data obtained here challenge the ancestry of the apicobasal cell polarity and raise questions about the homology of epithelial tissue across the Metazoa.


INTRODUCTION
The emergence of epithelial tissues was arguably one of the most critical events in animal evolution. In bilaterians and cnidarians, a true epithelium is defined as a group of polarized cells joined by belt-like cell-cell junctions and supported by a basement membrane, that can regulate passage of molecules paracellularly. Epithelial cells are polarized along the apical-basal axis and form a planar sheet of cells that can undergo subsequent morphogenesis [1][2][3][4][5] . While the asymmetric cortical distribution of the Wnt Planar Cell Polarity (PCP) pathway components polarizes the cells along the tissue plane, the asymmetric cortical distribution of Par system components polarizes the cells along the apical-basal axis 2,3,[5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] . These mechanisms that organize cell-polarity are highly conserved in all animals that have been studied and most likely they must have been present in the most recent common ancestor (MRCA) of Cnidaria and Bilateria 3,5,16,18,[20][21][22][23] . However, little is known outside of those two clades ( Figure 1A).
Interestingly, ctenophores or comb jellies, currently hypothesized to be the sister taxa to all metazoans 24-28 , do not have the genes that encode the components of the Wnt/PCP pathway in their genomes 26 . Thus, the study of the subcellular organization of the Par system components in ctenophores and their potential role in apical-basal polarity is important to understand the evolution of tissue organization in Metazoa.

Antibody Specificity
Components Genome searches of M. leidyi showed there were only a single copy for both Par-6 (MlPar-6) and Par-1 (MlPar-1) genes. Rabbit polyclonal affinity-purified antibodies (Bethyl labs, Inc) designed against M. leidyi MlPar-6 and MlPar-1 were used to determine their spatial and temporal expression at different developmental stages. Each antibody was characterized by Western blotting to establish its specificity ( Figure 2A). Western blots of M. leidyi adult extracts showed that the antibodies recognized different bands for MlPar-6 (predicted size 33.3 KD; Figure   2A) and MlPar-1 (predicted size 84.7 KD; Figure 2A) indicating that each antibody recognizes different genes. Pre-adsorption of the MlPar-6 and MlPar-1 antibody with a tenfold molar excess of the respective antigenic peptide (used to generate and affinity purify the antibodies) resulted in the elimination of the appropriate-sized single band for MlPar-6 and MlPar-1 (Figure 2A). In addition, whole-mount immunohistochemistry pre-adsorption experiments were performed to test the specificity of the MlPar-6 and MlPar-1 antibodies. The staining pattern was strongly mitigated in early embryos when pre-incubated antibodies were used ( Figure 2B). Thus, both antibodies are specific to their intended targets and provide robust reagents to determine the subcellular localization of these proteins during ctenophore embryogenesis. structure and function of epithelial tissue in the Metazoa.

MlPar-6 Gets Localized to the Apical Cortex of Cells During Early M. leidyi Development
Using our specific MlPar-6 antibody we characterized the subcellular localization of the MlPar-6 protein during early M. leidyi development. In all of the over 100 specimens examined, 6 MlPar-6 expression polarizes to the animal cortex (determined by the position of the egg pronucleus; Figure 3A) of the single cell zygote and to the apical cell cortex (cell-contact-free regions facing the external media) during every cleavage stage (Figure 3). At the cortex, MlPar-6 localizes perpendicular to the cleavage furrow in cell-contact-free regions ( Figure 3B). In addition, as cleavage ensues, MlPar-6 becomes localized to the position of cell-cell contacts between blastomeres until gastrulation ( Figure 3A). When the embryo is gastrulating (3-7 hpf), MlPar-6 is not localized in cells undergoing cellular movements including the animal (4 hpf; Figure 4A) and vegetal ectoderm (5 hpf; Figure 4A) that are undergoing epibolic movements, syncytial endoderm, and mesenchymal 'mesoderm.' However, this protein remains apically polarized in 'static' ectodermal cells remaining at the animal pole (blastopore) and vegetal pole. By the end of gastrulation (8-9 hpf; Figure 5A), MlPar-6 becomes localized in the apical cortex of the ectodermal epidermal cells and is also asymmetrically localized in the future ectodermal pharyngeal cells that start folding inside the blastopore ( Figure 5A). Interestingly, no clear cortical localization was observable in later stages and the antibody signal is weak after 10 hpf. Contrary to what was expected, in these later stages, MlPar-6 is localized to the cytoplasm and does not localize in the cortex of epidermal cells and few epithelial and mesenchymal cells showed nuclear localization ( Figure 5A), suggesting the protein does not play a role in cell adhesion or cell polarity during those stages.
Similar results were obtained when the mRNA encoding for MlPar-6 fused to mVenus (MlPar-6-mVenus) was overexpressed and the in vivo localization of the protein was recorded in M.
leidyi embryos ( Figure 5B). In these experiments, the translated MlPar-6-mVenus was observed after 4 hours post injection into the uncleaved egg. M. leidyi develops rapidly with cell cycle times of 15 minutes between cleavages, and therefore, it was impossible at this time to observe the in vivo localization of this protein in early blastomeres. During gastrulation, MlPar-6-mVenus localizes to the apical cell cortex and displays enrichment at the level of cell-cell contacts ( Figure 5B). However, as we observed by antibody staining, this cortical localization is no longer observable during the cell 7 movements associated with gastrulation and MlPar-6-mVenus remains cytosolic ( Figure 5B). After 8 hpf, MlPar-6-mVenus localizes to the apical cortex of ectodermal epidermal and pharyngeal cells but is not observable in any other internal tissue ( Figure 5B). After 10 hpf, MlPar-6-mVenus remains in the cytosol and no cortical localization was detectable ( Figure 5B), confirming the antibody observations presented above.

MlPar-1 Remains Cytoplasmic During Early M. leidyi Development
In bilaterians and cnidarians, the apical localization of MlPar-6 induces the phosphorylation Nevertheless, and strikingly, MlPar-1 remains as punctate aggregations distributed uniformly in the cytosol, and in some cases, co-distributes with chromosomes during mitosis ( Figure 6 and Figure   7), but no asymmetric localization of MlPar-1 was observed in the cell cortex of M. leidyi embryos at any of the stages described above for MlPar-6.
These Similar results were observed in vivo when the mRNA encoding for MlPar-1 fused to mCherry (MlPar-1-mCherry) was overexpressed into M. leidyi embryos by microinjection ( Figure   8A). Similar to MlPar-6-mVenus mRNA overexpression, the MlPar-1-mCherry translated protein was observed after 4 hours post injection into the uncleaved egg and its early localization in blastomeres was too faint to detect by this method. Our in vivo observations confirm the localization pattern described above using MlPar-1 antibody at gastrula stages. MlPar-1-mCherry localizes uniformly and form aggregates in the cytosol during gastrulation (4-5 hpf; Figure 8A). This localization pattern remains throughout all recorded stages until cydippid larva where MlPar-1-8 mCherry remains cytosolic in all cells but is highly concentrated in the tentacle apparatus and underneath the endodermal canals (24 hpf; Figure 8A).
Interestingly, the punctuate aggregates of MlPar-1-mCherry are highly dynamics and move along the entire cytoplasm suggesting a potential association with cytoskeletal components (Movie 1). Unfortunately, we were not able to clone and express neither MlDlg nor MlLgl to observe if they resemble similar localization patterns in vivo.

MlPar-6 and MlPar-1 Structures are Conserved
In bilaterians To discount the possibility that the observations recorded in vivo for both MlPar-6-mVenus and MlPar-1-mCherry proteins are caused by a low-quality mRNA or lack of structural conservation, we overexpressed each ctenophore mRNA into embryos of the cnidarian Nematostella vectensis and followed their localization by in vivo live-cell imaging ( Figure 8B). In these experiments, both MlPar-6-mVenus and MlPar-1-mCherry translated proteins display the same pattern than the previously described endogenous N. vectensis proteins 5 . In N. vectensis embryos, MlPar-6-mVenus and MlPar-1-mCherry symmetrically localize during early cleavage stages and both protein asymmetric localization was observable only after blastula formation ( Figure 8B). These data not only suggest that the ctenophore MlPar-6 and MlPar-1 protein structure and function are conserved but also indicate that their subcellular localization in M. leidyi embryos is regulated by an interaction with other signaling pathways that is different than N. vectensis embryos.

MlCdc42 and MlaPKC
To expand our observations on the localization of Par system components, we separately overexpressed the mRNA encoding for MlCdc42 fused to mCherry (MlCdc42-mCherry) and MlaPKC fused to mVenus (MlaPKC-mVenus). Both MlCdc42-mCherry ( Figure 8C) and MlaPKC-mVenus ( Figure 8D) translated protein were observable and recorded after 3 to 4 hours post injection into the uncleaved egg. Surprisingly, MlCdc42-mCherry only localizes in the cytosol and form punctate aggregates without any clear cortical localization through all stages recorded ( Figure   8C). On the other hand, MlaPKC-mVenus displayed an apparent apical localization but the signal of the translated protein was not strong enough to discriminate this localization from the cytosolic protein ( Figure 8D). Even though these data give some insights on the localization of both MlCdc42-mCherry and MlaPKC-mVenus proteins, the development of specific antibodies is required to determine the localization of maternally loaded proteins and discard technical artifacts. Unfortunately, we were not able at this time to co-inject these proteins together in order to see their in vivo interactions during embryonic development. Further technical improvements beyond the purposes of this work are necessary.

Embryogenesis
The early polarization of the Par/aPKC complex seems to be a conserved characteristic of stereotyped cleavage patterns in bilaterians where the Par system has been studied 4,5,14,15,53,[56][57][58][59]63,66,68,69,[76][77][78][79][80] . In these species, the asymmetric localization of Par-6 and Par-1 has been used as an indicator of maternal partitioning of determinants 58,61,63,69,81,82  Recent studies have shown that ctenophore species do not have the molecular components to form SJs and lack a Scribble homolog 23,46 . This suggests that none of the lateral polarity proteins (Dlg, Lgl, and Par-1) can localize to the lateral cortex of the ctenophore cells and explains the cytosolic localization of MlPar-1 during the observed stages. The primary structure of MlPar-1 protein (a Serine/threonine-protein kinase) is highly conserved and contains all the domains (with the same amino acid length) required for its proper functioning in other metazoans 20,23 , and

Evolution of Cell Polarity in Metazoa
In bilaterians the asymmetric localization of the components of the Par system establishes apicobasal cell polarity at the very early stages of embryonic development 4,5,17,69 , controlling their characteristic stereotyped cleavage patterns. However, in the cnidarian N. vectensis, embryonic polarity is controlled by the Wnt signaling system 16,71,[98][99][100] . The asymmetric localization of Par system components is established after cell-cell junctions are already from in the blastula epithelium and has no role on the early cleavage patterns 5 . In spite of these differences, in both bilaterian and cnidarian species, Par-mediated apicobasal cell polarity is responsible for the maturation and maintenance of cell-cell adhesion in epithelial tissue 4,22 . This suggests that the polarizing activity of an apical cell polarization is present during early cleavage stages and differentially regulated during gastrulation, but it is absent from epithelial cells of later juvenile stages. Unless the embryonic cell polarity in ctenophores is mediated by a completely different mechanism than bilaterians (that is also highly likely), our data would imply that cnidarians, such as N. vectensis, lost the mechanism by which Par proteins asymmetrically localize during early cleavage stages and gain a mechanism to maintain apicobasal cell polarity in epithelial tissues.
Intriguingly, ctenophore genomes do not have the full set of cell-cell adhesion 23,26,46 and Wnt signaling pathway components 26,27,101 that control the activity of Par proteins in bilaterian and cnidarian embryos. For example, in bilaterians the Wnt/PCP signaling pathway antagonizes the action of the Par/aPKC complex 7,10,15,47,102,103 , so this may explain the lack of polarization. In addition, the establishment of mature AJs and SJs is essential for the maintenance of cell polarity in cnidarian and bilaterian cells, so the absence of these pathways in ctenophores implies that a new set of interactions emerged at least in the Cnidaria+Bilateria ancestor, and that, could have regulated the way by which the Par system polarizes embryonic and epithelial cells (Figure 9). In conclusion, we have shown that regardless of the high structural conservation of Par proteins across Metazoa, including ctenophores, ctenophore cells do not have other essential components to deploy the asymmetrical polarizing function of the Par system as in other studied metazoans. These results reinforce the idea that signaling pathways act as modules that interact with each other to organize the cells [104][105][106] . Recent genomic studies have shown that, during animal evolution, different signaling pathways components have emerged at different nodes of the metazoan tree 23,26,28 . Interestingly, ctenophore genomes suggest that new signaling pathways do not only emerge as individual components but also as a full set of protein complexes that bring a new set of interactions that remodel the cell structure and behavior. In agreement with genomic studies, our results challenge the conception of a deep homology of the epithelial structure in Metazoa and suggest that, regardless the genetic background, similar morphologies (e.g., epidermis and mesoderm) could be developed by similar cellular behaviors.

Culture and Spawning of M. leidyi
Spawning, gamete preparation, fertilization and embryo culturing of M. leidyi embryos was performed as previously described 107 . Adult M. leidyi were maintained at the Whitney Laboratory for Marine Bioscience of the University of Florida (USA). Spawning was induced by incubating the adults under a three to four-hour dark cycle at room temperature. The embryos were kept in glass dishes (to prevent sticking) in filtered 1x seawater at room temperature until the desired stage.

Western Blot
Western blots were carried out as described 5,22 using adult tissue lysates dissected by hand in order to discard larger amount of mesoglea. Antibody concentrations for Western blot were 1:1,000 for all antibodies tested.

Immunohistochemistry
All immunohistochemistry experiments were carried out using the previous protocol for M. leidyi 107 . Embryos were fixed on a rocking platform at room temperature. Embryos of different stages were fixed for 2 hours in fresh Fixative (100mM HEPES pH 6.9; 0.05M EGTA; 5mM MgSO4; 200mM NaCl; 1x PBS; 3.7% Formaldehyde; 0.2% Glutaraldehyde; 0.2% Triton X-100; and 1X fresh sea water). Fixed embryos were rinsed at least five times in PBT (PBS buffer plus 0.1% BSA and 0.2% Triton X-100) for a total period of 3 hours. PBT was replaced with 5% normal goat serum (NGS; diluted in PBT) and fixed embryos were blocked for 1 to 2 hours at room temperature with gentle rocking. Primary antibodies were diluted in 5% NGS to desired concentration. Blocking solution was removed and replaced with primary antibodies diluted in NGS. All antibodies incubations were conducted over night on a rocker at 4°C. After incubation of the primary antibodies, samples were washed at least five times with PBT for a total period of 3 hours.
Secondary antibodies were then applied (1:250 in 5% NGS) and samples were left on a rocker overnight at 4°C. Samples were then washed with PBT and left on a rocker at room temperature for an hour. Samples were then washed once with PBT and incubated with DAPI (0.1µg/µl in PBT; Invitrogen, Inc. Cat. # D1306) for 1 hour to allow nuclear visualization. Stained samples were rinsed again in PBS two times and dehydrated quickly into isopropanol using the gradient 50%, 75%, 90%, and 100%, and then mounted in Murray's mounting media (MMM; 1:2 benzyl benzoate:benzyl alcohol) for visualization. Note that MMM attenuates the DAPI signal from samples. We scored more than 1,000 embryos per each antibody staining and confocal imaged more than 50 embryos at each stage obtaining similar staining patterns for each case.

mRNA Microinjections
The coding region for each gene of interest was PCR-amplified and cloned into pSPE3-mVenus or pSPE3-mCherry using the Gateway system 108 . Eggs were injected directly after fertilization as previously described for N. vectensis studies 5,109,110 with the mRNA encoding one or more proteins fused in frame with reporter fluorescent protein (N-terminal tag) using final concentrations of 300 ng/µl for each gene. Fluorescent dextran was also co-injected to visualize the embryos. Live embryos were kept at room temperature and visualized after the mRNA of the FP was translated into protein (4-5 hours). Live embryos were mounted in 1x sea water for visualization. Images were documented at different stages. We injected and recorded 20 embryos for each injected protein and confocal imaged each specimen at different stages for detailed analysis of phenotypes in vivo. We repeated each experiment at least five times obtaining similar results for each case. The fluorescent dextran and primers for the cloned genes are listed in Key resources table.

Imaging of M. leidyi Embryos
Images of live and fixed embryos were taken using a confocal Zeiss LSM 710 microscope

DECLARATION OF INTERESTS
The authors declare no competing interests.