Desmosomal cadherins in zebrafish epiboly and gastrulation
© Goonesinghe et al; licensee BioMed Central Ltd. 2012
Received: 8 August 2011
Accepted: 11 January 2012
Published: 11 January 2012
The desmosomal cadherins (DCs), desmocollin (Dsc) and desmoglein (Dsg), are the adhesion molecules of desmosomes, intercellular adhesive junctions of epithelia and cardiac muscle. Both the DCs and desmosomes have demonstrably essential roles in mammalian development. In order to initiate their study in a more tractable developmental system we have characterised zebrafish DCs and examined their roles in early zebrafish development.
We find that zebrafish possess one Dsc, the orthologue of mammalian Dsc1, which we designate zfDsc. Unlike mammalian Dscs, zfDsc exists only as the "a" form since it lacks the alternatively-spliced mini-exon that shortens the cytoplasmic domain to produce the "b" form. Zebrafish possess two Dsgs, designated zfDsgα and zfDsgβ, orthologues of mammalian Dsg2. They show 43.8% amino acid identity and the α form has a 43 amino acid glycine-rich sequence of unknown function in its extracellular domain. Both zfDsc and zfDsgα were present as maternal and zygotic transcripts whereas zfDsgβ was first expressed from 8 hours post-fertilisation (hpf). All three transcripts were present throughout subsequent stages of development. Morpholino knockdown of both zfDsc and zfDsgα expression produced similar defects in epiboly, axis elongation and somite formation, associated with abnormal desmosomes or reduced desmosome numbers.
These results demonstrate an important role for DCs and desmosomes in the early morphogenesis of the zebrafish embryo, provide a basis for more detailed analysis of their role and raise interesting questions relating to the evolution and functional significance of DC isoforms.
Cell-cell adhesion is a key mechanism that guides and co-ordinates the dynamic rearrangements of cell populations during animal development. Early zebrafish development is characterised by many such cell movements including epiboly, the spreading of the blastoderm over the yolk, ingression or involution at gastrulation, and convergence and extension, which lengthen the embryonic axis during and following gastrulation. Cell adhesion mediated by the Type 1 cadherin E-cadherin has been shown to play an essential co-ordinating role in each of these processes [1–5] and N-cadherin regulates posterior body formation [6–8].
Desmocollin (Dsc) and desmoglein (Dsg), the adhesion molecules of the intercellular junctions known as desmosomes, are also members of the cadherin superfamily [9–11]. Desmosomes maintain strong adhesion in adult epithelia and cardiac muscle but appear very early in mammalian development, for which their function is essential [12–15]. Ultrastructural evidence showed that desmosomes appear at the mid gastrula stage in the embryo of the teleost Fundulus heteroclitus and zebrafish [16, 17] so desmosomes could also contribute to fish early development, but this has not been investigated.
The desmosomal cadherins are transmembrane proteins whose extracellular domains mediate principally homophilic adhesion in the desmosomal interspace or desmoglea . Their cytoplasmic domains interact with linker or adaptor proteins plakoglobin, plakophilin and desmoplakin forming a dense desmosomal plaque, which is linked to the intermediate filament cytoskeleton, thus forming the desmosome-intermediate filament complex . In mammals both Dsc and Dsg are present as multiple genetic isoforms that show differential tissue expression. For example in human tissues there are three Dscs and four Dsgs . Dsc2 and Dsg2 are ubiquitous in tissues containing desmosomes but Dsc 1 and 3 and Dsg 1, 3, and 4 are largely confined to stratified epithelia where they show differentiation-specific expression [20, 21]. Desmosomes in cells expressing multiple isoforms contain a mixture of those isoforms [18, 22]. It is not clear why multiple isoforms of desmosomal cadherins are functionally necessary. Do they have specific adhesive functions or do they carry out specific roles in tissue differentiation and morphogenesis? The evidence from gene deletion and over-expression studies in mice suggests that they may have both adhesive and signalling functions (reviewed by [11, 23]).
The three mammalian Dsc isoforms also show alternative splicing of their cytoplasmic domains giving rise to 'a' and 'b' forms. The mammalian genes contain seventeen exons of which exon 16 is small and contains a stop codon. When this is spliced in the cytoplasmic domain of the 'a' form, roughly equal in size to those of Type 1 cadherins, is truncated giving the shorted 'b' form. Although it is usually present in roughly equal quantity to the 'a' form, the function of the 'b' form is unknown. The 'a' form alone appears sufficient to support desmosomal plaque formation .
In order to initiate the study of the role of desmosomes in early zebrafish development we have cloned zebrafish Dsc and Dsg, studied the timing of their expression and carried out knock down of their expression. We find that the desmosomal cadherins have a role in zebrafish development from epiboly onwards.
All primers for RT-PCR analysis of zfDsc, zfDsgα and zfDsgβ, and control primers were designed by Primer 3 program http://www.invitrogen.com/site/us/en/home/Products-and-Services/Product-Types/Primers-Oligos-Nucleotides/applied-biosystems-custom-primers-probes.html. The primers used for DSC1 were 5'-AAGGCGGTGTATGAGGTCAC-3' and 5'-GGTGCCTCTGTGTTGGATTT-3'; DSG2a 5'-CCAGTTCATGGTCATCGT-3' and 5'-GTCAGTGCAAAGTGTCTGG-3'; Dsg2b 5'-GGTGGAGGAAAACACCAGA-3' and 5'-GAGCATGGTGTCGCTGTCTA-3'; Beta-actin 5'-CCACGAGACCACCTTCAACT-3' and 5'-CATTGTGAGGAGGGCAAAGT-3'.
Rapid amplification of cDNA ends (RACE)
Primers for both 5' and 3' RACE were also designed by Primer 3 program. The primers used for RACE were:- 5'RACE, GSP 1, CTG TCA AAA GCC TTG CCT TC; GSP 2, ATG GGT GAA AGG GGG CTT TG; GSP 3, ATC CAC AGC CCC AGG ATT CA; 3'RACE, GSP, GAC TGC GGG GAA ATG GAC AG; GSP 1, TAG GAG CTG CGG GTT TCC TG. zfDsgα 5' sequences were obtained using the 5' RACE kit according to the manufacturer's instructions. Three antisense primers (GSP 1-3) were made based on the previously determined zfDsgα sequence. GSP 1 (Gene specific primer) was used for the first strand cDNA synthesis and GSP 2 and 3 were used for nested PCR. The PCR conditions of GSP 2 and GSP 3 were the same, with the cycling numbers of 35 and annealing temperature of 58°C. The 3' end of the zfDsgα was amplified using the 3' RACE kit according to the manufacturer's instructions. Two sense primers (GSP and GSP 1) for nested PCR were designed based on the known 3' sequence. Cycling numbers of GSP was 35 with the annealing temperature of 58°C, and cycling numbers of GSP 1 was 35 with the annealing temperature of 60°C.
Sequencing and bioinformatics
The ABI PRISM BigDye Terminator Cycle Sequencing Ready-reaction Kit was used for DNA sequencing as per the manufacturer's instructions. The sequencing reactions were processed in the Central Sequencing Facility (Faculty of Life Sciences, University of Manchester) using the AB377 sequencer (Applied Biosystems). The data from the sequencing were analyzed using BioEdit software. Zebrafish EST and YAC clone sequences were retrieved by searching the databases of NCBI http://www.ncbi.nlm.nih.gov/ and ZFIN http://zfin.org. ENSEMBL http://www.ensembl.org/ was used to predict gene structure and give a predicted cDNA, genomic and protein sequences. The chromosome location of zfDsc and zfDsgα was analyzed by ZFIN, VEGA http://vega.sanger.ac.uk/Danio_rerio/ and Tübingen http://wwwmap.tuebingen.mpg.de/. NCBI Blastp program and Workbench http://seqtool.sdsc.edu/CGI/BW.cgi/ were used to analyze protein homology, and NCBI Blastn program was used to analyze nucleotide homology. Ebi http://www.ebi.ac.uk and Expasy http://us.expasy.org were used to predict the peptide sequence of cDNA sequences. PSORT II http://psort.hgc.jp/form2.html was used to predict the signal peptide.
PCR amplification and electrophoresis
1 μl template cDNA was added to a mixture of 2 μl MgCl2 (25 mM, Promega), 2 μl 10× Buffer (Promega), 0.5 μl dNTPs (Bioline), 0.5 μl each of forward and reverse primers and 0.2 μl Taq polymerase (Promega). The final reaction mixture was made up to a volume of 20 μl with ddH2O, the reaction mixture was amplified in a thermal cycler using standard cycling conditions. In all PCR reactions a minus template control was used as a genomic contamination control.
The DNA in solution (mixed with 5× loading dye) and either Hyperladder I or IV (Bioline) were both run on a TAE agarose gel (containing ethidium bromide at 10 mg/ml) at varying concentrations depending on product size (< 100 bp on a 2% gel to > 10 KB in a 0.8% gel) in 1× TAE buffer at 15 V/cm of gel. The different bands were visualized by UV illumination and photographed. Comparison of the product band to Hyperladder I/IV™ (Bioline) allowed the determination of the product size.
RNA extraction, cDNA and 'rescue' mRNAsynthesis
Embryos in their experimental groups (50 embryos) were homogenised in 1 ml Trizol (Invitrogen) using a Teflon homogeniser. 200 μl chloroform was added to the samples and they were shaken vigorously and incubated for 2-3 minutes at RT, then centrifuged for 15 minutes at 12,000 × g at 4°C. The aqueous phase was removed into a fresh tube, 500 μl isopropanol was added and the samples were incubated for 10 minutes at RT and centrifuged for 10 minutes at 12,000 × g at 4°C. The supernatant was removed, the pellet was washed with 1 ml 75% ethanol and centrifuged for 5 minutes at 7,500 × g at 4°C. The ethanol was removed and the pellet was air dried and re-suspended in 40 μl RNase-free water, if the pellet had trouble dissolving the solution was incubated for 5 minutes at 60°C. The concentration of RNA was determined by using a NanoDrop™ spectrophotometer.
The Qiagen Omniscript cDNA synthesis kit and its components were used to synthesise cDNA from RNA samples. 1 μg RNA was added to 2 μl 10X RT Buffer, 2 μl dNTPs (0.5 μM each), 0.5 μl RND primers (Invitrogen), 0.5 μl of OligodTs (Invitrogen), 1 μl RNase inhibitor 10 units/μl, (Roche), 1 μl Omniscript RT and made up to 20 μl with RNase-free water. This was incubated at 37°C for 1 hour and then stored at -20°C.
The Ambion T7 mMessenger™ kit was used to generate 5' capped mRNA from the pT7Ts vector. These were diluted to the appropriate concentration before being co-injected with the corresponding morpholino.
In situhybridisation probe synthesis
To synthesize the probe a mixture of 2 μl linearised DNA (μg), 2 μl 10× transcription buffer, 2 μl DIG labelling mixture (Roche), 1 μl (20 units) of RNase inhibitor and 2 μl (40 units) T7 RNA polymerase made up to 20 μl with RNase free water was incubated for 2 hours at 37°C. Afterwards 1 μl of RNase free DNase I™ (Roche) was added and incubated for further 30 minutes and then the reaction was cleaned up using the Qiagen RNA easy mini kit as per manufacturer's instructions. The probe was then diluted 1:1 with formamide and stored at -80°C.
Fixed de-chorionated embryos were incubated for 4 hours in approximately 1 ml HYB buffer at 60°C. The HYB buffer was replaced with a mixture up to 500 μl of preheated HYB buffer and the corresponding probe (0.5 - 2.5 ng/μl). The embryos were then incubated o/n at 60°C. The next day the embryos were washed briefly in HYB wash buffer and then put through a series of 15 minute washes in mixtures of HYB buffer and SSC in different compositions (75% HYB buffer/25% 2 × SSC, 50% HYB buffer/50% 2 × SSC, 25% HYB buffer/75% 2 × SSC, 100% 2 × SSC) at 60°C, the embryos were then washed 2 × 30 minutes in 0.2 × SSC at 60°C before they were then put through a further series of 5 minute washes with a mixture of 0.2 × SSC and PBST (75% 0.2 × SSC/25% PBST, 50% 0.2 × SSC/50% PBST, 25% 0.2 × SSC/75% PBST). After this the embryos were washed 2 × 5 minutes in PBST. The embryos were then put in PBST with 2% blocking reagent for a minimum of 1 hour at RT, they were incubated for o/n at 4°C in blocking reagent with the diluted antibody, anti-dig (1.5 μl in 5000 μl blocking reagent).
The following day the embryos were washed 8 × 15 minutes in PBST and 3 × 5 minutes in freshly prepared NTMT solution, the embryos were then transferred into BM purple ready-to-use-solution (Roche) to perform the colour reaction.
Following the colour development under the microscope the colour reaction was stopped at the necessary point by washing the embryos 3 × 5 minutes with PBS. To reduce the background staining due to the BM purple the embryos were transferred through a series of glycerol/PBS-solutions (30 minutes in 20% glycerol/80% PBS, 30 min in 50% glycerol/50% PBS, then in 80% glycerol/20% PBS). Images were taken with a Zeiss Axioplan dissection microscope.
Zebrafish husbandry and collection of embryos
Zebrafish were maintained according to standard conditions described in the zebrafish handbook, . Embryos were cultured at 28.5°C in conditioned water (60 mg Instant Ocean salts, Tropic Marin® in 1 litre deionised H2O).
TL (Tübingen Longfin) strain zebrafish embryos were used in all experiments. To generate embryos, adult females and males were transferred into breeding tanks (two females and two males per tank) the night before the embryos were required. Males and females were separated by a plastic divider. The following morning the dividers were removed and the fish allowed to breed. Embryos were collected 30 minutes after removal of the dividers and the fertilised ones used for injection.
Special injections plates were prepared for the injections -1.5% agarose in conditioned water was set in a 94 mm Petri-dish with an inserted (custom made) mould, which was removed after the agarose had set, to create special grooves to hold the embryos. The collected embryos were lined up in the grooves and injected (a volume of approximately 3 nl) with of RNA, MO or MO, by using a gas driven microinjection apparatus and glass needles. The injections were carried out into the yolk before the 16-cell stage of development. After that the embryos were incubated in fresh Petri-dishes with conditioned water at 28.5°C. All embryos were destroyed at 96 hpf in accordance with Home Office regulations.
DNA Sequence (5' - 3')
Dsc 1 ATG block
GCG TTC ATA CAT CCT GAA GCG AGA G
Dsc 1 '2' ATG block
GAT GTG TCC GGT CTC CAC CAT AAA C
Dsg 2a ATG block
AAC AGG TGA AAT TCG CCG GGC C
Dsg 2a '2' ATG block
CCG GTA GAA CAC GAT ATT TCC TGA T
Dsg 2a splice block
TGC AGG TCA CAT ACC AAC AGC ACT G
CCT CTT ACC TCA GTT ACA ATT TAT A
De-chorionation, observations and DIC analysis of live embryos
Embryos were incubated in a pre-warmed pronase (Fluka) solution (2 mg/ml) at 37°C for 20 minutes. Afterwards the chorions were removed by passing the embryos several times gently through a glass Pasteur pipette. They were then transferred to an 8× chorion wash solution three times.
Observations of live embryos were carried out in the chorions or alternatively the embryos were manually de-chorionated and mounted in a 1.5% solution of methyl cellulose in 8× chorion water, held in a specially designed embryo mould and viewed using either a Zeiss Stereolumar dissecting stereoscope or Zeiss AxioPlan with Nomarski optics. For description of the stages and timing of embryonic development see .
To categorise the MO injected embryos into phenotypes, the embryos were left overnight at 28.5°C. The categorisation was carried out into wildtype, mild, moderate and severe phenotypes. To take images of the observed phenotypes by using a compound microscope with Nomarski differential interference contrast (DIC) optics as well as a stereo-microscope at 24 hpf, the embryos were de-chorionated by incubating them for 15 min in pre-warmed pronase solution (2 mg/ml) at 37°C. Afterwards the embryos were anaesthetised using MS222 and held in 8 × chorion solution.
SYTOX nuclear green stain is impermeable to living cells, but stains nuclei in a syncitium (or otherwise following membrane degradation). A 5 mM solution in DMSO was co-injected in a 50:50 mix with 1 mM MO.
Embryos were fixed overnight in 2% paraformaldehyde/2% glutaraldehyde in 0.1 M cacodylate buffer at pH 7.4, then post-fixed in 1% OsO4 in the same buffer for 2 hours, dehydrated and embedded in Spurr resin. Ultrathin sections were stained with 1% uranyl acetate and 0.3% lead citrate, then examined on a FEI Technai Biotwin electron microscope.
Isolation of zebrafish desmocollin and desmoglein sequences
The partial cDNA sequence for zebrafish Dsc (zfDsc) was obtained from four overlapping EST sequences from the TIGR Zebrafish Gene Index (Accession numbers: TC225818, TC130266, TC169171 and TC182641) and one EST sequence from GenBank (Accession number: CD283558.1) and used to obtain a predicted cDNA sequence from Ensembl. From this assembly, overlapping primer sets were designed to clone by PCR and sequence the full length zfDsc mRNA from 4 dpf zebrafish embryos. This yielded a cDNA of 3.4 kb with an ORF of 2676 bp encoding 892 amino acids (GenBank Accession number: JQ013460) (Additional file 1, Figure S1). All available EST sequences with homology to mammalian Dsc appeared to arise from a single gene.
A partial sequence for zfDsg was obtained from three overlapping TIGR EST sequences (Accession numbers: AI883817, AI397144 and BI878525) and used to predict the cDNA sequence from Ensembl. Five overlapping primer pairs were then used to obtain the sequence. The sequence obtained still lacked correct 5' and 3' ends so both ends were extended by RACE. Finally a full length cDNA of 3.6 kb was obtained containing an ORF of 3429 bp encoding a protein of 1143 amino acids (GenBank Accession number: JQ013461) (Additional file 1, Figure S2). This sequence was designated zfDsgα to distinguish it from a second sequence, designated zfDsgβ, indentified during nucleotide homology analysis from a NCBI EST sequence (Accession number: BI884774). A partial cDNA sequence for zfDsgβ was obtained by PCR (GenBank Accession number: JQ013462) (Additional file 1, Figure S3). These results indicate that zebrafish have one desmocollin isoform and two related desmoglein isoforms.
Genomic organisation of zebrafish desmosomal cadherins
Searching the zebrafish genome in Ensembl (Zv9) revealed that zfDsc cDNA was located on chromosome 20, at position 7,437,162-7,468,565 (reverse strand) (ENSDARG00000039677), so the zfDsc gene is about 30 kb in size. Again this analysis suggested the existence of only a single Dsc paralogue in zebrafish. The gene for zfDsgα (ENSDARG00000062750) is also located on chromosome 20 at position 16,990,224-17,010,459 also on the reverse strand. The gene for zfDsgβ (ENSDARG00000076426) is located on chromosome 2 at position 2,392,229-2,424,395. No other Dsg paralogues were detected.
Comparison of zfDsc exon sizes with three human desmocollin genes
Comparison of exon sizes of zfDsgα with those of human and mouse Dsg2
Zebrafish desmosomal cadherins are the orthologues of mammalian Dsc1 and Dsg2
Comparison of deduced amino acid sequences showed that zfDsgα to have the greatest homology with human and mouse Dsg 2, which shared 61% and 64% amino acid homology, respectively (Figure 1B). The zfDsgα protein precursor comprised a 16 aa signal sequence followed by a 21 aa pre-protein. Within the extracellular region, the high homology domains were EC1, 2 and 4. The CAR site in the EC1 domain is IAL rather than YAL in Dsg 2 of human and mouse. A glycine-rich protein sequence was found in the EC 3 domain. This showed no significant identity with any mammalian Dsg 2 or with any other protein in the databases. Its cDNA sequence did, however, show 100% identity to three zebrafish EST sequence (Accession number: BI881590.1, BE017197.1 and CK738901.1). The highest amino acid identity appeared in the zfDsgα TM domain, which shared 65% and 68% protein identity with human and mouse Dsg 2, respectively. The cytoplasmic region of zfDsgα was divided into IA, ICS, LD, RUD and TD sub-domain and, except the ICS region, the overall amino acid identity for this region was low (Figure 1D).
We were able to clone only part of the zfDsgβ sequence but were subsequently able to obtain a full length cDNA sequence from Ensembl (Accession number: ENSDART00000109073). The predicted amino acid sequence shows 43.8% identity with zfDsgα. We have used this sequence to generate the cladogram in Figure 1B. This suggests that zfDsgα and β are paralogues. Domain comparison between their predicted protein sequences is shown in Additional file 1, Figure S4.
Desmosomal cadherins are continuously expressed from early in zebrafish development
Knockdown of desmosomal cadherins affects epiboly, gastrulation and convergence-extension movements
Desmosome formation is reduced in morphant embryos
Our results show that the zebrafish possesses 3 desmosomal cadherin genes and that the desmosomal cadherins are of functional importance in embryonic development from epiboly and through gastrulation and axis formation.
The single zebrafish desmocollin gene shows two surprises. Firstly, it is the orthologue of mammalian Dsc1, which is principally expressed in the differentiated layers of stratified squamous epithelia, rather than the more widely expressed Dsc2. In mammals Dsc1 is involved in strengthening cell-cell adhesion in these epithelia, particularly epidermis , and, since it is the only desmocollin present, zfDsc presumably has a similar function wherever it is expressed. (Previous work has suggested the presence of Dsc2 in zebrafish and suggested that it has a role in heart development . However, while this work appears to relate to the gene we describe here as zfDsc, it difficult to compare that work with ours due to the lack of information regarding methods and sequences in .)
Secondly, zfDsc lacks a shortened 'b' form, which in mammals is produced by alternative splicing of a mini exon , also missing from zebrafish. Thus zfDsc corresponds to the longer 'a' form of mammals. This is interesting because the functional significance of the 'b' form is not clear, the 'a' form having been shown to be sufficient for desmosomal plaque formation, while the 'b' form does not support plaque formation . Apparently the zebrafish can form perfectly normal desmosomes in the completed absence of a 'b' form leaving the significance of the 'b' form even more in doubt.
There are two zebrafish desmoglein genes, both orthologues of the widely expressed mammalian Dsg2. Designated zfDsgα and zfDsgβ, they differ in that α contains a 40 amino acid, glycine-rich insert in its extracellular domain and share 43.8% amino acid identity. The insert is unique to zebrafish and of unknown functional significance. Both α and β are expressed early in zebrafish development though β is not represented in maternal transcripts. It may be suggested that both are of functional significance in early development, though this remains to be demonstrated for β.
It was a surprise to find that knock-down of desmosomal cadherins disrupted development from early epiboly because ultrastructural studies of developing fish embryos have not detected desmosomes at this stage (present work and [16, 17]). However, we showed that maternal transcripts for both zfDsc and zfDsgα were present from the start of development so these are clearly functional at this very early stage. (These early effects were obtained only with ATG morpholinos; maternal mRNA is spliced and as anticipated the splice site morpholinos were without effect (not shown).) The effects of knock-down were often quite severe, apparently involving detachment of superficial cells from the eYSL, which, in some cases continued epiboly movements. It has been shown previously in Fundulus that YSL movements continue independently after mechanical removal of the DC and EL . Our results suggest the possibility that desmosomal cadherins may be involved in adhesion of the superficial layers to the eYSL.
The role of desmosomal cadherins in epiboly appears to differ from those described for two other adhesion molecules, E-cadherin and Ep-CAM, which have been shown to be functionally important during this stage, because the phenotypes generated by mutations and or knock-down of these molecules are different from those described here [1, 2, 5, 17, 32]. E-cadherin appears to be important for intercalation of interior layer cells into the exterior layer of the epiblast and/or adhesion of deep cells to the underside of the enveloping layer [1, 2]. Ep-CAM is also involved in adhesion between deep cells and the enveloping layer as well as cell-cell adhesion within the enveloping layer . No effects on epiboly were reported for knock-down of the desmosomal plaque armadillo-family protein plakoglobin, which occurs in adherens junctions as well as in desmosomes . On the other hand morpholino knockdown of claudin-E, which is necessary for tight junction formation at the leading edge of the EVL, also caused a strong epiboly delay or arrest, similar to the phenotypes we have described here . Interestingly, the claudin-E morpholino also caused a delay in eYSL epiboly.
Our ultrastructural studies show that while desmosomes were absent during early epiboly, they were present between cells of the enveloping layer at the shield stage, where they form the basal components of typical epithelial junctional complexes. This timing appears to correspond well with the appearance on zygotic desmosomal cadherin transcripts and is slightly earlier than "mid gastrulation" as reported for Fundulus by Trinkaus and Lentz . Thereafter, desmosomes persisted between cells of the enveloping layer and by 24 hpf formed extensive arrays of ten or more individual junctions. Desmosomal cadherin knock-down was associated with the presence of fewer desmosomes and altered desmosome morphology, suggesting that desmosomal adhesion was reduced. (Ideally it would be desirable to confirm these observations by immunofluorescence and western blotting with appropriate antibodies. Screening of existing antibodies revealed none that reacted with zebrafish desmosomal cadherins and our attempts to raise such antibodies have so far proved unsuccessful.)
The changes in morphology that we have found in longer-surviving embryos namely, shortening of the embryonic axis, broadening of the pattern of gata1 expression and lateral enlargement of somites and of the field of myoD expression, are very similar to those found in a number of mutants that affect gastrulation and convergence-extension movements, including the mutant half baked which affect E-cadherin [2, 35]. This suggests that the desmosomal cadherins also play an important role in coordinating these movements. Since electron microscopy suggests that desmosomes are confined to the junction complexes of the EL, it appears that maintenance of normal adhesion in these complexes is crucial for gastrulation and convergence-extension movements.
Zebrafish have a single desmocollin, the orthologue of mammalian Dsc1, and two closely related desmogleins, orthologues of mammalian Dsg2. These are expressed early in zebrafish development where morpholino knockdown suggests that they play important role in epiboly, gastrulation and convergence-extension movements. This work provides a novel insight into zebrafish early development and a basis for further investigation of the role of desmosomal adhesion
We thank Samantha Forbes for assistance with electron microscopy and Jeroen Bakkers for the gift of riboprobes. We thank the Biotechnology and Biological Sciences Research Council for PhD support of AG. Financial support was provided by the Medical Research Council grant G9630879.
- Kane DA, McFarland KN, Warga RM: Mutations in half baked/E-cadherin block cell behaviors that are necessary for teleost epiboly. Development. 2005, 132: 1105-16. 10.1242/dev.01668.View ArticlePubMedGoogle Scholar
- Shimizu T, Yabe T, Muraoka O, Yonemura S, Aramaki S, Hatta K, Bae YK, Nojima H, Hibi M: E-cadherin is required for gastrulation cell movements in zebrafish. Mech Dev. 2005, 122: 747-63. 10.1016/j.mod.2005.03.008.View ArticlePubMedGoogle Scholar
- Montero JA, Carvalho L, Wilsch-Brauninger M, Kilian B, Mustafa C, Heisenberg CP: Shield formation at the onset of zebrafish gastrulation. Development. 2005, 132: 1187-98. 10.1242/dev.01667.View ArticlePubMedGoogle Scholar
- Solnica-Krezel L: Gastrulation in zebrafish -- all just about adhesion?. Curr Opin Genet Dev. 2006, 16: 433-41. 10.1016/j.gde.2006.06.009.View ArticlePubMedGoogle Scholar
- Babb SG, Marrs JA: E-cadherin regulates cell movements and tissue formation in early zebrafish embryos. Dev Dyn. 2004, 230: 263-77. 10.1002/dvdy.20057.View ArticlePubMedGoogle Scholar
- von der Hardt S, Bakkers J, Inbal A, Carvalho L, Solnica-Kreze Ll, Heisenberg CP, Hammerschmidt M: The Bmp gradient of the zebrafish gastrula guides migrating lateral cells by regulating cell-cell adhesion. Curr Biol. 2007, 17: 475-87. 10.1016/j.cub.2007.02.013.View ArticlePubMedGoogle Scholar
- Harrington MJ, Hong E, Fasanmi O, Brewster R: Cadherin-mediated adhesion regulates posterior body formation. BMC Dev Biol. 2007, 7: 130-10.1186/1471-213X-7-130.View ArticlePubMedGoogle Scholar
- Warga RM, Kane DA: A role for N-cadherin in mesodermal morphogenesis during gastrulation. Dev Biol. 2007, 310: 211-25. 10.1016/j.ydbio.2007.06.023.View ArticlePubMedGoogle Scholar
- Garrod DR, Merritt AJ, Nie Z: Desmosomal adhesion: structural basis, molecular mechanism and regulation (Review). Mol Membr Biol. 2002, 19: 81-94. 10.1080/09687680210132476.View ArticlePubMedGoogle Scholar
- Garrod DR, Merritt AJ, Nie Z: Desmosomal cadherins. Curr Opin Cell Biol. 2002, 14: 537-45. 10.1016/S0955-0674(02)00366-6.View ArticlePubMedGoogle Scholar
- Dusek RL, Godsel LM, Green KJ: Discriminating roles of desmosomal cadherins: beyond desmosomal adhesion. J Dermatol Sci. 2007, 45: 7-21. 10.1016/j.jdermsci.2006.10.006.View ArticlePubMedGoogle Scholar
- Fleming TP, Garrod DR, Elsmore AJ: Desmosome biogenesis in the mouse preimplantation embryo. Development. 1991, 112: 527-39.PubMedGoogle Scholar
- Gallicano GI, Kouklis P, Bauer C, Yin M, Vasioukhin V, Degenstein L, Fuchs E: Desmoplakin is required early in development for assembly of desmosomes and cytoskeletal linkage. J Cell Biol. 1998, 143: 2009-22. 10.1083/jcb.143.7.2009.View ArticlePubMedGoogle Scholar
- Den Z, Cheng X, Merched-Sauvage M, Koch PJ: Desmocollin 3 is required for pre-implantation development of the mouse embryo. J Cell Sci. 2006, 119: 482-9. 10.1242/jcs.02769.View ArticlePubMedGoogle Scholar
- Eshkind L, Tian Q, Schmidt A, Franke WW, Windoffer R, Leube RE: Loss of desmoglein 2 suggests essential functions for early embryonic development and proliferation of embryonal stem cells. Eur J Cell Biol. 2002, 81: 592-8. 10.1078/0171-9335-00278.View ArticlePubMedGoogle Scholar
- Trinkaus JP, Lentz TL: Surface specializations of Fundulus cells and their relation to cell movements during gastrulation. J Cell Biol. 1967, 32: 139-53. 10.1083/jcb.32.1.139.View ArticlePubMedGoogle Scholar
- Slanchev K, Carney TJ, Stemmler MP, Koschorz B, Amsterdam A, Schwarz H, Hammerschmidt M: The epithelial cell adhesion molecule EpCAM is required for epithelial morphogenesis and integrity during zebrafish epiboly and skin development. PLoS Genet. 2009, 5: e1000563-10.1371/journal.pgen.1000563.View ArticlePubMedGoogle Scholar
- Nie Z, Merritt A, Rouhi-Parkouhi M, Tabernero L, Garrod D: Membrane-impermeable cross-linking provides evidence for homophilic, isoform-specific binding of desmosomal cadherins in epithelial cells. J Biol Chem. 286: 2143-54.Google Scholar
- Garrod D, Chidgey M: Desmosome structure, composition and function. Biochim Biophys Acta. 2008, 1778: 572-87. 10.1016/j.bbamem.2007.07.014.View ArticlePubMedGoogle Scholar
- Whittock NV, Bower C: Genetic evidence for a novel human desmosomal cadherin, desmoglein 4. J Invest Dermatol. 2003, 120: 523-30. 10.1046/j.1523-1747.2003.12113.x.View ArticlePubMedGoogle Scholar
- Nuber UA, Schafer S, Stehr S, Rackwitz HR, Franke WW: Patterns of desmocollin synthesis in human epithelia: immunolocalization of desmocollins 1 and 3 in special epithelia and in cultured cells. Eur J Cell Biol. 1996, 71: 1-13.PubMedGoogle Scholar
- North AJ, Chidgey MA, Clarke JP, Bardsley WG, Garrod DR: Distinct desmocollin isoforms occur in the same desmosomes and show reciprocally graded distributions in bovine nasal epidermis. Proc Natl Acad Sci USA. 1996, 93: 7701-5. 10.1073/pnas.93.15.7701.View ArticlePubMedGoogle Scholar
- Thomason HA, Scothern A, McHarg S, Garrod DR: Desmosomes: adhesive strength and signalling in health and disease. Biochem J. 429: 419-33.Google Scholar
- Troyanovsky SM, Eshkind LG, Troyanovsky RB, Leube RE, Franke WW: Contributions of cytoplasmic domains of desmosomal cadherins to desmosome assembly and intermediate filament anchorage. Cell. 1993, 72: 561-74. 10.1016/0092-8674(93)90075-2.View ArticlePubMedGoogle Scholar
- Westerfield M: The Zebrafish Book. 1995, Eugene: University of Oregon PressGoogle Scholar
- Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF: Stages of embryonic development of the zebrafish. Dev Dyn. 1995, 203: 253-310. 10.1002/aja.1002030302.View ArticlePubMedGoogle Scholar
- Collins JE, Legan PK, Kenny TP, MacGarvie J, Holton JL, Garrod DR: Cloning and sequence analysis of desmosomal glycoproteins 2 and 3 (desmocollins): cadherin-like desmosomal adhesion molecules with heterogeneous cytoplasmic domains. J Cell Biol. 1991, 113: 381-91. 10.1083/jcb.113.2.381.View ArticlePubMedGoogle Scholar
- Thisse B, Pflumio S, Fürthauer M, Loppin B, Heyer V, Degrave A, Woehl R, Lux A, Steffan T, Charbonnier XQ, Thisse C: Expression of the zebrafish genome during embryogenesis (NIH R01 RR15402). ZFIN Direct Data Submission http://zfin.org. ZFIN Direct Data Submission. 2001, [http://zfin.org]Google Scholar
- Chidgey M, Brakebusch C, Gustafsson E, Cruchley A, Hail C, Kirk S, Merritt A, North A, Tselepis C, Hewitt J, et al: Mice lacking desmocollin 1 show epidermal fragility accompanied by barrier defects and abnormal differentiation. J Cell Biol. 2001, 155: 821-32. 10.1083/jcb.200105009.View ArticlePubMedGoogle Scholar
- Heuser A, Plovie ER, Ellinor PT, Grossmann KS, Shin JT, Wichter T, Basson CT, Lerman BB, Sasse-Klaassen S, Thierfelder L, et al: Mutant desmocollin-2 causes arrhythmogenic right ventricular cardiomyopathy. Am J Hum Genet. 2006, 79: 1081-8. 10.1086/509044.View ArticlePubMedGoogle Scholar
- Trinkaus JP: Cells into Organs: Forces that shape the embryo. 1984, London: Prentice Hall InternationalGoogle Scholar
- Tay HG, Ng YW, Manser E: A vertebrate-specific Chp-PAK-PIX pathway maintains E-cadherin at adherens junctions during zebrafish epiboly. PLoS One. 5: e10125-Google Scholar
- Martin ED, Moriarty MA, Byrnes L, Grealy M: Plakoglobin has both structural and signalling roles in zebrafish development. Dev Biol. 2009, 327: 83-96. 10.1016/j.ydbio.2008.11.036.View ArticlePubMedGoogle Scholar
- Siddiqui M, Sheikh H, Tran C, Bruce AE: The tight junction component Claudin E is required for zebrafish epiboly. Dev Dyn. 239: 715-22.Google Scholar
- Solnica-Krezel L, Stemple DL, Mountcastle-Shah E, Rangini Z, Neuhauss SC, Malicki J, Schier AF, Stainier DY, Zwartkruis F, Abdelilah S, et al: Mutations affecting cell fates and cellular rearrangements during gastrulation in zebrafish. Development. 1996, 123: 67-80.PubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.