The collagen field undoubtedly regains
The collagen field undoubtedly regains attractiveness with the emergence of new concepts (cell niche, regulation of growth factor bioavailability, mechanotransduction, cell microenvironment sensing) and new tools to investigate their in vivo functions (atomic force microscopy (AFM), second harmonic generation (SHG) imaging, genome editing, gene reporter lines….). Growing evidence showed that they sustain functions that would have been difficult to predict based on in vitro and in cellulo data. In that, the use of diverse animal models has proven to be useful to fully understand collagen function in their entire diversity, from the simplest invertebrates like hydra to high vertebrates . This has been beautifully illustrated with the story of the discovery of the collagen XVIII in vivo function. While the knockout of Col18a1 in mice showed no obvious phenotype, lack of h89 of its orthologs, cle-1 in C. elegans or Mp in drosophila, was strikingly coupled to axon guidance defects [6,7]. This established a link with the Knobloch syndrome in humans and further allowed revisiting the Col18a1-null mouse phenotype .
More recently, zebrafish (Danio rerio) has proven to be a powerful animal model to interrogate ECM function in development and disease [, , ]. It came out as no surprise since, in a mere two decades, zebrafish has become one of the most popular model organisms used in basic and medical research . Besides the genetic tractability of this model, zebrafish are unmatched in their maintenance, breeding, raising and feeding. Moreover, zebrafish is highly amenable to live imaging approaches because of its small size, its external development and the transparency of embryos and larvae . Lastly but not the least, this small vertebrate represents a good alternative model to implement the “3R” principles of ethical experimentations in disease modeling and drug screening . While zebrafish has largely gained its reputation on its amenability to forward genetics in the 90’s, the development of reverse genetic approaches in this small vertebrate proved challenging. The real revolution arose from recent advances on the so-called Clustered Regulatory Interspaced Short Palindromic Repeats (CRISPR)/Cas9 genome editing technology that offered unprecedented possibilities for genome manipulation and the investigation of specific physiological functions not only during embryonic and larval development but also during adulthood and aging [15,16].
We believe that the recent advances in gene editing methods and the increasing availability of “omics” data together with our recent characterization of the zebrafish matrisome showing that ECM proteins are well conserved between mammals and fish  are factors, among others, propitious to rapidly propel zebrafish as one of the leading models in ECM research. Here we specifically discuss mounting evidence highlighting the suitability of the zebrafish model to investigate the collagen functions in development, tissue regeneration and disease.
The zebrafish extracellular matrix Two main types of ECM organization can be observed in tissues: interstitial ECM and basement membrane (BM) that are both easily identifiable with transmission electron microscopy (TEM) in vertebrates, and zebrafish is not an exception (Fig. 1). Collagens are central components of these highly structured and tissue-specific supra-molecular aggregates. Interstitial matrices and BM found in zebrafish tissues and organs are generally similar in composition and structure than the mammalian ones, but zebrafish ECM also exhibit a few unique features. Specifically, skin ECM displays striking particularities. At early stage, epidermis is only composed of two cell layers and starts to thicken from 15 dpf (day post-fertilisation) to become multilayered in adults . As in mammals, the epidermis is separated from the subjacent dermis by a BM whose structure and organization is indistinguishable at TEM from the mammalian one (Fig. 1B). Along this line, a global time-course transcriptomic analysis of BM formation during caudal fin regeneration revealed that BM gene expression kinetics recapitulates that the one described in mammals . However, unlike other vertebrates, the lamina lucida contains electron-dense small aggregates aligned regularly between the inner surface of the lamina lucida and the epidermal cell surface (Fig. 1B) [17,19]. These structures, termed adepidermal granules, are not present in adults . The epidermal cells are most likely responsible for their production, but their composition is unknown.