Matrix Biology 28 (2009) 490–502
Contents lists available at ScienceDirect
Matrix Biology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m a t b i o
Craniofacial cartilage morphogenesis requires zebrafish col11a1 activity Dominique Baas 1, Maryline Malbouyres, Zofia Haftek-Terreau 1, Dominique Le Guellec, Florence Ruggiero ⁎ Université de Lyon; Université Lyon 1, France Institut de Biologie et Chimie des Protéines, CNRS UMR 5086, Université Lyon-1, France IFR 128 BioSciences-Gerland, 7 passage du Vercors 69367 Lyon cedex 7, France
a r t i c l e
i n f o
Article history: Received 16 January 2009 Received in revised form 14 July 2009 Accepted 17 July 2009 Keywords: Type XI collagen col11a1 Zebrafish Extracellular matrix Chondrogenesis col2a1 Craniofacial cartilage Neurocranium Pharyngeal cartilage Notochord Chondrodysplasia
a b s t r a c t The zebrafish ortholog of the human COL11A1 gene encoding the cartilage collagen XI proα1 chain was characterized to explore its function in developing zebrafish using the morpholino-based knockdown strategy. We showed that its expression in zebrafish is developmentally regulated. A low expression level was detected by real-time PCR during the early stages of development. At 24 hpf, a sharp peak of expression was observed. At that stage, in situ hybridization indicated that col11a1 transcripts are restricted to notochord. At 48 hpf, they were exclusively detected in the craniofacial skeleton, endoskeleton of pectoral fins and in otic vesicles. Collagen XI α1deficient zebrafish embryos developed defects in craniofacial cartilage formation and in notochord morphology. Neural crest specification and mesenchymal condensation occurred normally in morpholino-injected embryos. Col11a1 depletion affected the spatial organization of chondrocytes, the shaping of cartilage elements, and the maturation of chondrocytes to hypertrophy. Knockdown of col11a1 in embryos stimulated the expression of the marker of chondrocyte differentiation col2a1, resulting in the deposit of abnormally thick and sparse fibrils in the cartilage extracellular matrix. The extracellular matrix organization of the perichordal sheath was also altered and led to notochord distortion. The data underscore the importance of collagen XI in the development of a functional cartilage matrix. Moreover, the defects observed in cartilage formation resemble those observed in human chondrodysplasia such as the Stickler/Marshall syndrome. Zebrafish represent a novel reliable vertebrate model for collagen XI collagenopathies. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Collagens form a super family of 28 distinct matrix proteins that provide structural support and strength to the connective tissues. Beside their strong contribution to matrix biomechanical properties, collagens are also closely involved in a variety of morphogenetic events and display highly specialized functions (Ricard-Blum and Ruggiero, 2005). A sub-class of the collagen superfamily is referred to as fibrillar collagens that assemble into highly ordered polymers, the striated fibrils. The structure and composition of the fibrils strongly contribute to the function and biomechanical properties of connective tissues such as bone, skin, tendon and cartilage. Cartilage plays an important role in the formation and growth of the vertebrate skeleton through a mechanism called endochondral ossification. During embryogenesis, mesenchymal cells condense and go through differentiation along the chondrocyte cell lineage. This transition is characterized by the expression of extracellular matrix markers such as the fibrillar collagen II, a major component of the cartilaginous extracellular matrix. In mammals, collagens II and XI are expressed by chondrocytes as they start to differentiate and these two collagens associate extracel⁎ Corresponding author. Tel.: +33 472722657; fax: +33 472722602. E-mail address:
[email protected] (F. Ruggiero). 1 Present address: ENS-Lyon, 46 Allée d'Italie, 69364 Lyon cedex 07, France. 0945-053X/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.matbio.2009.07.004
lularly along with the non fibrillar collagen IX to form the heterotypic cartilage collagen fibrils (Eyre, 2004). Collagen XI is a heterotrimer composed of three chains: α1(XI), α2(XI) and α3(XI), the latter being the product of the collagen II gene, col2a1. Although minor in quantity, collagen XI plays a critical role in collagen fibril assembly and growth (for review Fichard et al., 1995; Eyre, 2004). Mutations in the human collagen XI genes, COL11A1 and COL11A2, can lead to osteoarthritis and to a variety of chondrodysplasia and have dramatic effects on cell differentiation and skeleton morphogenesis (Vikkula et al., 1995, Kuivaniemi et al., 1997; Spranger, 1998; Jakkula et al., 2005). Mouse strains with spontaneous or targeted mutations in collagen XI genes have provided animal models for collagen XI collagenopathies. The Col11a2 null mice showed a mild cartilage phenotype, consistent with human disorders caused by mutations in COL11A2 gene such as the non-ocular Stickler syndrome also called oto-spondylomegaepiphyseal dysplasia (OSMED) (Vikkula et al., 1995; Li et al., 2001), as well as the non-syndromic form of deafness called DFNA13 (McGuirt et al., 1999). COL11A1 human gene mutations have been associated with Marshall syndrome that is an autosomal dominant craniofacial disorder similar to the more common disorder Stickler syndrome (Griffith et al., 1998; Majava et al., 2007). Chondrodysplasia (cho) mice present a spontaneous mutation in Col11a1 gene that results in premature termination of the proα1(XI) chain translation (Li et al., 1995). The lack of functional Col11a1 gene in mice altered the chondrocyte columnar alignment in the
D. Baas et al. / Matrix Biology 28 (2009) 490–502
growth plates of all long bones but had no effect on chondrocyte differentiation. As a consequence of the lack of collagen XI expression, the cho/cho mice die at birth and show important cartilage alterations leading to disproportionate dwarfism, short snout and cleft palate. The cartilage presents thicker fibrils, supporting the role of collagen XI in the control of collagen II fibril growth. Heterozygous mice survive normally beyond birth and show no obvious signs of chondrodyspasia. The value of zebrafish model for human disease is well recognized (Ingham, 2009). The rapid external development of the transparent embryo allowing direct visual analysis of phenotypic defects coupled with the ease to perform protein knockdown in developing embryos make zebrafish an attractive vertebrate model for exploring protein function. We previously showed that the morpholino-knockdown strategy is efficient and reliable to decipher collagen function (PagnonMinot et al., 2008). The skeleton of zebrafish develops in the same way as in humans and the molecular mechanisms controlling skeleton development are highly conserved (Yelick and Schilling, 2002). To further investigate the role of col11a1 in cartilage formation, we thus performed a functional analysis using morpholino antisense oligonucleotides to knockdown col11a1 expression in developing zebrafish embryos. With this aim, we have identified and characterized the cDNA of the zebrafish col11a1 gene. In situ hybridization showed that col11a1 is expressed in notochord and in developing cartilage. In addition to abnormalities in notochord shape, knockdown experiments demonstrated that the lack of col11a1 altered the chondrocyte stacking into orderly arrays and the proper formation and spatial organization of craniofacial cartilages but did not affect early cartilage patterning. Our data add support to the concept that collagen XI plays an important role in cartilage morphogenesis that likely extends beyond its contribution to fibrillogenesis process. Moreover, the col11a1 morphants provide a new reliable vertebrate model system to elucidate molecular mechanisms underlying type XI collagenopathies. 2. Results 2.1. Identification and characterization of zebrafish col11a1 gene Cloning of zebrafish collagen XI cDNA was carried out using a reported zebrafish genomic sequence coding for a collagen identified in database as the ortholog of human COL11A2 gene (GenBank accession number AL672176). The 5′ end of the partial reported sequence was retrieved using 5′RACE amplification. A 580 nt fragment was obtained that comprises the 5′UTR region, the start codon, the signal peptide sequence and the lacking N-terminal end of the protein. From the 5′ RACE data, the partial genomic sequence AL672176 and a genomic sequence reported later (XM685088), we determined the full-length of the zebrafish cDNA. The sequence spans 5086 bp predicting a protein of 1695 amino acids (accession number AM287214). The deduced collagen pro-chain starts with a typical hydrophobic signal peptide sequence with a putative cleavage site at position A25-Q26. The mature protein has the structural features of the minor fibrillar collagen XI chains and consists of a large N-terminal propeptide, a central large triple helix of 1014 residues (COL1), and a C-propeptide (NC1). The Nterminal domain contains a TSPN domain (241 amino-acids) with 4 cysteines at conserved positions, a short variable region (VR, 60 aminoacids), a short triple helix domain which is double interrupted and consists of continuous 17 Gly-Xaa-Yaa triplets (COL2) and a linker region (NC2) (Fig. 1A). In order to unambiguously identify the reported collagen pro-chain, the carboxyl terminal domain of the zebrafish collagen reported as the proα2(XI) chain in database was aligned with the corresponding sequences of the human proα1(XI), proα2(XI) and proα1(V) chains and the red seabream pro-α1(V/XI) chain (Touhata et al., 2001) (Fig. 1B). The number and location of cysteine residues in the Cpropeptide were reported to be unique, and thus this structural feature can be used to identify the type and chain species of fibrillar collagens
491
(Morvan-Dubois et al., 2003). The length of the NC1 domain and the substitution of the cysteine 2 by a serine in the zebrafish C-propeptide sequence both indicate that the identified cDNA encodes a collagen prochain more closely related to the mammalian proα1(XI) chain than to the proα2(XI) chain as reported in database. The TSPN domain of proα1(V) and proα1(XI) chains is removed by BMP-1 during proteolytic processing (Imamura et al., 1998; Medeck et al., 2003). The cleavage site is conserved in various species and is present in the proα2(XI) chain. Alignments of human proα1(XI), proα2 (XI) and proα1(V) N-propeptide cleavage sites with the corresponding region of fish collagen XI chains (medaka, tetraodon and zebrafish) show that the glutamine at the P2 and P1′ positions and the proline at the P3′ position are conserved (Fig. 1C). Interestingly, the aspartate residue at position P2′ that was recently shown to be crucial for efficient BMP-1 cleavage of the human proα1(V) chain (Bonod-Bidaud et al., 2007) is present in fish species. The sequence upstream the cleavage site flanked by two cysteines is also strictly conserved in fish collagen XI chains, human proα1(XI) and proα1(V) chains but not in proα2(XI) (Fig. 1C). Altogether, our data indicate that the zebrafish collagen XI sequence reported in the database as col11a2 gene corresponds most likely to the col11a1 gene. 2.2. Col11a1 is developmentally regulated and is primarily expressed in notochord and cartilaginous tissues To examine the timing and the level of col11a1 expression we performed real-time RT-PCR on RNA extracted from embryos at different stages (Fig. 2). Col11a1 transcripts were readily detectable at 3 h postfertilization (hpf) but at a very low level. The low expression level was maintained during the early stages of development. At 20 hpf, col11a1 transcript level increased and a sharp peak of expression (5 fold compared to that of 20 hpf embryos) was observed at 24 hpf. Then, expression decreased progressively but a significant col11a1 expression persisted until 5 days post-fertilization (dpf) that is the last stage analyzed. The spatial distribution of col11a1 during zebrafish development was analyzed with in situ hybridization on whole embryos at different stages (Fig. 3). In mouse and human, splice variants of col11a1 have been reported. A probe encompassing the TSPN domain was prepared in order to detect all the possible splice variants. Specificity of hybridization signal was attested by using sense RNA probe complementary to the specific probe (data not shown). A strong signal was detected at 20 hpf (Fig. 3A) in the notochord and became restricted to the caudal region at 30 hpf (Fig. 3B,C) and 36 hpf (Fig. 3D). At 48 hpf, expression in the notochord disappeared and cartilaginous tissues were strongly stained (Fig. 3E, F–J). An intense signal was detected in the otic capsule and in the pectoral fins (Fig. 3E). Serial cross-sections of embryo head showed that col11a1 messages accumulated in the developing craniofacial cartilage elements, the neurocranium and the pharyngeal skeleton (Fig. 3F–J). Signals are detected in all different cartilage elements of the zebrafish skeleton along the anteroposterior axis: Meckel's (Fig. 3F,G), ethmoid plate and palatoquadrate (Fig. 3G,H), trabeculae and ceratohial, basihal, hyosympletic (Fig. 3I) and parachordal (Fig. 3J). The eyes fit into the grooves along the sides of the ethmoid plate and trabeculae (Fig. 3H, I). The otic vesicles are composed of the cartilaginous ear capsules and fits into the large cavities formed, to either sides, by the parachordal cartilage (Fig. 3J). 2.3. Morpholino-based targeted knockdown of col11a1 To investigate the role of col11a1 in zebrafish development, an antisense morpholino targeting the ATG start codon of col11a1 (col11a1MO) was designed. Serial dilutions of col11a1MO were injected in 1 to 2-cell embryos. Injection of quantities superior or equal to 20 pmol resulted in a lethal phenotype at approximately 2 dpf whereas injection of 10 pmol of col11a1MO resulted in embryos that survived until approximately 7 dpf. 24-hpf embryos injected with the
492
D. Baas et al. / Matrix Biology 28 (2009) 490–502
Fig. 1. Schematic diagram of the primary structure of zebrafish collagen XI proα1 chain (A) and sequence alignments (B). (A) The structure of zebrafish collagen proα1 chain is identical to the structure of the human ortholog. It consists of a large triple helix (COL1) flanked by a C-terminal domain (NC1) responsible for the trimerization of the molecule and a N-terminal extension. The N-terminal extension is composed of a non collagenous domain (NC3) sub-divided into the globular TSPN domain and the elongated variable region (VR), a small triple helix domain (COL2) and a linker region (NC2). (B) Multiple alignments of the carboxy terminal domains of human proα1(V) and proα1(XI) chains, red sea-bream proα1(V/XI) and zebrafish proα1(XI) chains. Gaps are introduced to maximize the alignment and are indicated by spaces. Cysteine residues (in bold) are numbered from the Nterminal end of the C-propeptide (1 to 8). Cys 2 in the zebrafish C-propeptide sequence is lacking in zebrafish sequence (grey box). (C) Multiple alignments of the conserved BMP-1 cleavage site (arrow, grey box) located at the C-terminal end of the TSPN domain. Conserved residues are in bold. Aspartate at position P2′ is underlined.
5-mismatch control morpholino, mcol11a1MO (Fig. 4B), were not phenotypically distinguishable from 24 hpf wild type embryos (Fig. 4A). Col11a1MO embryos exhibited abnormal head morphology visible from 24 hpf: the head was smaller and flattened and the eyes were mispositioned (Fig. 4D). The specificity of the col11a1 morpholino was verified using a morpholino targeted to the ATG start codon of col1a1 that encodes the collagen I α1 chain, which is expressed in bones but not in cartilage. The head morphology of embryos injected with col1a1MO appeared identical to wild type embryos at 24 hpf (Fig. 4C). To confirm the specificity of the col11a1-morpholinos, Western blots were performed on lysates of 24 hpf col11a1MO and control embryos using human antibodies to collagen XI. The antibodies recognize the α1(XI)
and α2(XI) chains of pepsinized mammalian collagen XI (Fig. 4E, lane 1). Three bands were detected in zebrafish wild type lysates, referred to as a, b and c (Fig. 4E, lane 2) which can correspond to collagen XI α1 chain intermediate processed forms. The position of band b corresponds to the molecular weight of the mature collagen XI α1 chain. Based on their electrophoresis pattern, band c likely corresponds to the unprocessed proα1(XI) chain and band a to the triple helix domain of the α1(XI) chain. The intensity of the bands was markedly reduced in col11a1MOembryo lysates (Fig. 4E, lane 3) but unchanged in mcol11a1MO-embryo lysates (Fig. 4E, lane 4). We conclude that the col11a1 morpholinos provide a significant and specific reduction of collagen XI α1 translation. 2.4. Collagen XI α1 knockdown results in important defects in craniofacial cartilage development
Fig. 2. Real time quantitative RT-PCR. Real time quantitative RT-PCR assay for expression of col11a1 in zebrafish embryos at different stages as indicated. Relative cycle threshold (Ct) values are correlated with histone Ct.
The zebrafish cartilaginous head skeleton consists of the neurocranium, protecting the brain and the sensory organs and the pharyngeal skeleton, supporting the feeding and gill-breathing structures (Kimmel et al., 2001). The external morphology of the larvae was analyzed at 3 dpf by stereomicroscopy (Fig. 5A,B) and by scanning electron microscopy (Fig. 5C–F). Compared to wild type (Fig. 5A), col11a1MO-injected embryos showed abnormal head morphology and the mouth was not visible (Fig. 5B). As observed in Fig. 4D, the head of col11a1MO-injected embryos was flat and small (Fig. 5B). The ears (Fig. 5B) and pectoral fins (not shown), that expressed col11a1 in developing embryos (Fig. 3), appeared normal. By scanning electron microscopy, the protuberant pharyngeal arches were clearly visible at 3 dpf in wild type (Fig. 5C), and
D. Baas et al. / Matrix Biology 28 (2009) 490–502
493
Fig. 3. Expression pattern of col11a1. A–F, In situ hybridization of whole mount zebrafish embryos at 20, 30, 36 and 48 hpf with the col11a1 antisense RNA probe. (A) col11a1 is expressed in notochord (arrowheads) at 20 hpf prior to be restricted to the caudal notochord at 30 hpf (B, C) and 36 hpf (D). (C) shows a cross-section of the tail of a 30 hpf embryo. A strong signal is observed in the notochord (n). (nt) and (a) indicate the neural tube and the aorta respectively. By 48 hpf expression in the notochord disappears and col11a1 messages accumulate in otic vesicles (E, arrow) and in pectoral fins (E, arrowhead). At this stage, a strong signal is also detected in the developing craniofacial cartilage as observed in serial cross-sections of the embryo head (F–J). Staining of the elements of the neurocranium and the pharyngeal cartilage is observed along the anteroposterior axis of the embryo head: F, G Meckel's cartilage (m); G, H ethmoid plate (ep) and palatoquadrate (pq); I, trabulae (tr); ceratohial (ch), hyosympletic (hs); basihal (bh); J, parachordal (pch). A, B, E, lateral views; D dorsal view; C, E-I cross-sections.
in collagen I and 5-mismatch MO controls (Fig. 5E,F), but were missing in age-matched col11a1MO-injected larvae (Fig. 5D). The lack of the mouth opening (Fig. 5D) was confirmed in 60% of col11a1MO-injected larvae. The other 40% exhibited an atrophied or misshapen mouth. Alcian blue staining was used to reveal head cartilage patterning of 3 dpf larvae (Fig. 6A–G). In col11a1 morphants, most of pharyngeal (Fig. 6B) and neurocranium (Fig. 6D) cartilage elements were malformed or severely reduced compared to wild type (Fig. 6A,C). An anteroposterior gradient of severity was observed. The first arch ventral Meckel's (m) was the most drastically altered in morphants: it appeared smaller, mispositioned and severely malformed (Fig. 6B,F,G). The ceratohyal second arch (ch) and traberculae (tr) were also generally misshapen and/or mispositioned (Fig. 6B,F,G). The five ceratobranchial arches were generally present but severely reduced (Fig. 6B) compared to wild types (Fig. 6A). Reproducible morphological phenotypes were observed following several sets of injections, 100% of morphants showed alteration of the first arch, 57% also showed defects in the second arch and/or in the 3–7 arches (ceratobranchials). The anterior neurocranium (palate) is also altered in col11a1MO-injected larvae. In wild type, the anterior trabeculae (tr) fused in the midline and extended anteriorly to form the ethmoid plate (ep) (Fig. 6C). In col11a1MOinjected larvae (Fig. 6D dorsal view; Fig. 6F and G, lateral views) the ethmoid plate did not extend anteriorly and was smaller than in wild type (Fig. 6C dorsal view; Fig. 6E lateral view). Lateral views confirm that lack of collagen XI expression impaired the formation of the mouth (Fig. 6F,G). Otic vesicles and pectoral fin cartilage also retained alcian blue staining. No external phenotypic defect was observed in pectoral appendages (Fig. 6D) compared to wild type (Fig. 6C). As observed on live larvae (Fig. 5A,B), otic capsules appeared normal in morphants (Fig. 6D,F,G) compared to wild type larvae (Fig. 6C,E). Zebrafish possess dermal and cartilage replacement bones. Dermal bones develop as intramembranous bones and cartilage-replacement bones form by endochondral ossification of cartilage precursors. To
study the consequence of col11a1 depletion on cartilage-replacement bone formation, we stained larvae at 7 dpf with the vital stain Quercetin. At that stage most endochondral and dermal bones are mineralized. We focused our observations on the ceratobranchial 5 (cb5) that anchors the pharyngeal teeth. The adjacent teeth served as a control of efficient staining. In wild type larvae, pharyngeal teeth and cb5 are ossified and showed strong staining (Fig. 6H). In morphants, reduced or no staining was observed in the remnant cb5 (Fig. 6I,J) whereas the teeth showed intense staining and appeared correctly formed. 2.5. Expression of chondrogenesis markers in col11a1MO-injected embryos The cranial neural crest cell (NCC) streams migrate from the hindbrain and posterior midbrain to form most of the pharyngeal skeleton while more anterior NCCs contribute to the formation of the neurocranium. Although col11a1 expression analyzed by real-time PCR was low (Fig. 2) and diffuse (not shown) in early development, it was important to examine whether the different steps of the developmental pathway from neural crest to cartilage are perturbed in absence of col11a1 expression. In order to detect possible defects in hindbrain segmentation, in situ hybridization experiments with the rhombomerespecific marker krox-20 were performed in col11a1MO-injected embryos and wild type at 20 hpf (Fig. 7A,B). The expression domain of krox-20 that is normally expressed in rhombomere 3 (r3) and 5 (r5) was identical in morphants (Fig. 7B) and wild type embryos (Fig. 7A). The migration of neural crest cells into the pharyngeal arches can be visualized by in situ hybridization with the homeobox-containing gene dlx2a. Dlx2a is expressed in the three neural crest streams (Fig. 7C,D). The most anterior stream (s1) and the second stream (s2) correspond to the neural crest cells that migrate into the mandibular arch and the hyoid arch, respectively. The third stream (s3) represents neural crest cells that contribute to the five posterior pharyngeal arches (Piotrowski and Nusslein-Volhard, 2000). In wild type embryos, dlx2a was detected
494
D. Baas et al. / Matrix Biology 28 (2009) 490–502
expression (Fig. 8A,B). In col11a1MO-injected larvae, the misshapen posterior pharyngeal arches retained col2a1 expression (Fig. 8G,H). A substantial increase in col2a1 expression was found in morphants (Fig. 8G,H) compared to wild type embryos (Fig. 8E,F). This result was confirmed at the protein level using polyclonal antibodies to collagen II (Fig. 8I,J). Collagen II expression was strongly increased in craniofacial chondrocytes of 6 dpf col11a1-injected larvae (Fig. 8J) compared to wild type (Fig. 8I). The staining also revealed the alteration of chondrocytes stacking in col11a1MO-injected larvae (compare Fig. 8I,J). 2.6. Collagen XI α1 knockdown affects proper chondrocyte stacking and shaping of pharyngeal cartilage
Fig. 4. Col11a1 morpholino specificity. A–D, Lateral views of live wild type (A), mcol11a1MO (B), col1a1MO (C) and col11a1MO (D) at 24 hpf. Defects in head morphology are merely observed in col11a1MO-injected embryos: the head is smaller and flattened and eyes are mispositioned (D). E, Western blot analysis of lysates of 24 hpf wild type (lane 2), col11a1MO (lane 3) and mcol11a1MO (lane 4) embryos probed with antibodies to collagen XI (upper panel) and with actin antibody for protein loading control (lower panel). Lane 1 shows bovine pepsinized collagen XI probed with antibodies to collagen XI. The polyclonal antibodies recognize the α1(XI) and α2(XI) chains. In wild type lysates, three bands (a, b and c) are detected (lane 2). In col11a1MO embryos (lane 3), the intensity of the three bands a, b and c are less intense than in controls (lanes 2 and 4).
in the three neural crest streams at 15 hpf (Fig. 7C). At 30 hpf, dlx2a was expressed in the migrating neural crest cells that populate the different pharyngeal pouches (Fig. 7G). At 15 hpf (Fig. 7D) and 20 hpf (Fig. 7F), morphants exhibited reduced dlx2a expression compared to agematched wild type embryos (Fig. 7C and E). However, the 30 hpf morphants (Fig. 7H) and 20 hpf wild type embryos (Fig. 7E) showed similar dlx2a expression. To ensure that the reduced expression of dlx2a in morpholino-injected embryos was due to a developmental delay, the expression of marker genes present in different structures of the head from 15 hpf to 30 hpf (fgf8a, pax2a, flh) was analyzed. No difference was observed in col11a1-morphants compared to wild type embryos (data not shown). We thus cannot completely exclude that the mild reduction of dlx2a expression is solely due to a developmental delay. However, the formation of the pharyngeal arches was further ascertained using the monoclonal antibody Zn8 that stains the endodermal pouches (in green), and a cell nuclei staining (in blue) (Fig. 7I,J). At 33 hpf, Zn8 staining showed no distinguishable difference in morphants (Fig. 7J) compared to controls (Fig. 7I). The transcription factor sox9a is expressed at chondrogenesis sites. Its expression slightly precedes and regulates the expression of the major chondrocyte marker col2a1. A robust expression of sox9a was observed in the pharyngeal cartilage elements of 3 dpf wild type larvae (Fig. 8A,B). In col11a1MO-injected larvae, some cartilage elements of the head are reduced, misshaped or missing (Fig. 8C,D) as observed with alcian-blue staining (Fig. 6). At 3 dpf, the expression domain of col2a1 in head skeleton (Fig. 8E,F) coincided with sox9a
In 6 dpf-wild type larvae, the pharyngeal arches form a long rod of successive cartilage elements (Fig. 9A). At that stage, chondrocytes have already started to differentiate into hypertrophic chondrocytes, but orderly stacks of chondrocytes were still clearly visible (Fig. 9C). In contrast, morphants failed to undergo these major morphogenetic processes (Fig. 9B,D). First, compared to wild type larvae (Fig. 9A), the number, the shaping and organization of the cartilage elements were deeply altered in col11a1MO-injected larvae (Fig. 9B). The pharyngeal cartilage elements were malformed and mispositioned and, in some cases, individualization of arches failed to occur (Fig. 9B,D). The spatial organization of ceratobranchial arches (cb1–cb5) that are separated from each other by gill clefts in wild type larvae (Fig. 9A) and the shape of the ethmoid plate (ep), that forms the palate, were severely perturbed in col11a1 morphants (compare Fig. 9A,B). Consequently, the craniofacial skeleton is deeply altered in col11a1MO-injected larvae and the mouth did not protrude from the anterior-most region of the head (Fig. 9A,B). Secondly, col11a1MO larvae failed to form orderly stacks of chondrocytes (Fig. 9D) as in wild type larvae (Fig. 9C). In col11a1MO-injected embryos, pectoral fins that expressed col11a1 (Fig. 3), were found slightly shorter than in wild type but appeared otherwise normal. The organization and formation of the endochondral disc was not altered in col11a1MO-injected larvae (see Supplementary data 1). 2.7. Collagen XI α1 knockdown provokes an increase in cartilage fibril diameter In human and mice, collagen XI regulates the diameter of collagen II fibrils (Eyre, 2004). We thus analyzed the collagen fibrils structure in col11a1MO-injected larvae using transmission electron microscopy (TEM). At 50 hpf, an abundant network of thin, randomly organized fibrils was observed in the pharyngeal cartilage elements of wild type embryos (Fig. 10C) whereas only few sparse filamentous structures were observed in morphants (Fig. 10D). In the 50 hpf col11a1 morphants, TEM observations confirmed the defects in the columnar organization of chondrocytes and their abnormal round shape (compare Fig. 10A,B). At 6 dpf, chondrocytes in wild type larvae resembled hypertrophic chondrocytes (Fig. 10E), whereas in the col11a1MOinjected larvae, chondrocyte differentiation appeared to be developmentally delayed (Fig. 10F). An abundant extracellular matrix was deposited but the collagen fibrils were much larger (Fig. 10H) than those present in age-matched wild type larvae (Fig. 10G). Contrary to wild type larval fibrils (Fig. 10G), the banding pattern of the morphant fibrils was clearly visible. Less proteoglycan aggregates were found to be associated with collagen fibrils in morphants (Fig. 10H) compared to wild type (Fig. 10G). 2.8. Collagen XI α1 knockdown results in notochord distortion Given that col11a1 was transiently expressed in the notochord of developing embryos (Fig. 3), its structure was analyzed using alcian blue staining. The distortion of the notochord was clearly observed in
D. Baas et al. / Matrix Biology 28 (2009) 490–502
495
Fig. 5. Head morphology. A, B, Lateral photographs of live 3 dpf wild type (A) and col11a1MO (B) larvae. The head of col11a1MO larvae is smaller and malformed and the mouth is not visible (arrow indicates the position of the mouth). The otic vesicle is not altered (asterisk). C–F, Scanning electron microscopy of wild type (C) and col11a1MO (D) larvae at 3 dpf, and the corresponding controls, col1a1MO (E) and mcol11a1MO (F). Ventral views show that in wild type (C), col1a1MO (E) and mcol11a1MO (F) embryos, the mouth (C, E) and the protuberant pharyngeal arches (C, E, F) are clearly visible and properly formed, whereas in col11a1MO-injected embryos (D), the mouth is absent and the pharyngeal arches are not visible. mo, mouth; op, olfactory pits; pa, pharyngeal arches. Scale bar = 50 µm (C–F).
34 hpf embryos injected with 15 pmol of col11a1 morpholinos (compare Fig. 11A,B). The notochord was vacuolated normally in morphants indicating that lack of collagen XI expression may primarily affect the shaping and not the differentiation of the notochord. In zebrafish, the notochord is surrounded by an extracellular matrix sheath, which is composed of three different layers (Parsons et al., 2002). Adjacent to the basement membrane that surrounds the notochord (Fig. 11C,D, arrow), a dense fibrous layer runs circularly and perpendicularly to the chordal axis (Fig. 11C,D, arrowhead). The third layer is composed of an outermost fibrous layer that runs parallel to the body axis (Fig. 11C, D, asterisk). In col11a1 morphants (Fig. 11D), the basement membrane adjacent to notochord cells (arrows) showed interruptions and the cell membrane appeared less closely apposed to the basement membrane (Fig. 11D). The thickness and the density of the fibrous layer (arrowhead) were markedly reduced (compare Fig. 11C,D) and the fibril organization, structure and density were also considerably altered in the outermost layer (asterisk) compared to wild type embryos (Fig. 11C, D). 3. Discussion In cartilage, each chondrocyte is surrounded by a specialized extracellular matrix that is composed of thin fibrils made of collagen II and XI.
The extracellular matrix formation and chondrocyte differentiation are thus closely connected. The present study demonstrates that col11a1 activity is indispensable to proper cartilage development in zebrafish. 3.1. Characterization and expression pattern of zebrafish col11a1 In the current study, we have identified and characterized the zebrafish ortholog of a human COL11A1 gene referred to as the col11a2 gene in database. The two human chains, proα1(XI) and proα2(XI), show identical structural organization and are highly homologous. However, at least two features justify the designation of this sequence as col11a1 gene: the lack of the cysteine 2 in the Cpropeptide and the normal size of C-propeptide (Fichard et al., 1995). Interestingly, the unusual cleavage site for BMP-1 responsible for the release of the TSPN domain during collagen V/XI maturation (Medeck et al., 2003; Bonod-Bidaud et al., 2007) is conserved in zebrafish as well as in other fish species. The N-terminal processing of collagen V/XI by BMP-1 is an important event that is involved in the control of fibril diameter. Collagen II and XI form heterotypic fibrils in cartilage extracellular matrix and their expression patterns are intimately correlated in mammals (Yoshioka et al., 1995; Eyre, 2004). The col11a1 expression pattern reported herein coincides with the col2a1
496
D. Baas et al. / Matrix Biology 28 (2009) 490–502
Fig. 6. Alcian-blue and Quercetin staining. A–G, Photographs of alcian-stained 3 dpf wild type (A, C, E) and col11a1MO (B, D, F, G) embryos. Ventral (A,B) and dorsal (C,D) views show that most of cranial cartilage elements, pharyngeal arches and neurocranium, are misshapen and reduced in col11a1MO injected embryos (B, D) compared to wild type (A, C). Arrows point to the pectoral fins of wild type (C) and morphants (D). Lateral views (E–G) confirm that in col11a1-morphants (F, G) cranial cartilages are mispositioned and malformed compared to wild type (E). Arrowheads mark the ear capsule (E–G). m, Meckel's cartilage; pq, palatoquadrate; ch, ceratohyal; cb, ceratobranchial; ep, ethmoid plate; tr, trabeculae; ov, otic; vesicles. H–J, Ventral views of 7 dpf larvae stained with Quercetin. Ossification was observed in pharyngeal teeth (t) and in ceratobranchial 5 (cb5) of wild type larvae (H). In col11a1MO-injected larvae, the ceratobranchial 5 adjacent to the teeth shows reduced (I) or no staining (J). cb5; ceratobranchial 5; t, tooth.
expression pattern in developing zebrafish (Yan et al., 1995). As described for col2a1, substantial expression level was observed at 20 hpf when zebrafish col11a1 is transiently expressed in a non chondrogenic tissue, the notochord. Both Col2a1 and Col11a transcripts were shown to accumulate in few other non-cartilaginous sites in developing mouse (Cheah et al., 1991; Yoshioka et al., 1995). 3.2. Collagen XI α1 knockdown in zebrafish affects craniofacial chondrocyte stacking and cartilage shaping Most of the neurocranium and all pharyngeal skeleton are derived from cranial neural crest cells. During neural crest migration, dlx2a is expressed in three major streams of cells corresponding to a mandibular, hyoid and a posterior domain that forms the ceratobranchial arches (Piotrowski and Nusslein-Volhard, 2000). The migration of crest cells as visualized by dlx2a expression, was slightly delayed in 15 hpf and 20 hpf morphants suggesting a defect in NCC migration or a general developmental delay of the col11a1MO-injected embryos. The latter hypothesis is more probable since the 30 hpf morphants and 20 hpf wild type embryos showed similar dlx2a expression. Moreover, the endodermal pouches properly developed and neural crest cells populated the arches as in wild type. The first step of chondrocyte differentiation, after prechondrogenic condensation, consists in producing an abundant cartilaginous matrix that surrounds cells. Although collagen XI is minor in quantity, its crucial role in cartilage development was first supported by the analysis of the mice homozygous for the cho mutation (Li et al., 1995). We showed that the col11a1MO-injected larvae exhibited obvious facial defects and recapitulate both molecularly and phenotypically the cho mouse (Li et al., 1995). As observed in cho/+ mice, loss of col11a1 expression in developing zebrafish results in defects in the craniofacial cartilage
morphogenetic events that accompany the onset of chondrocyte differentiation: chondrocyte stacking and shaping, individuation and spatial organization of cartilage elements. Expression of sox9a, a key transcription factor in cartilage formation, marks differentiation of chondrocytes and is described as a potent activator of col2a1 expression in mammals (Bell et al., 1997; de Combrughe et al., 2000) and in zebrafish (Yan et al., 2002; Yan et al., 2005). In mouse, SOX9 also acts as a stimulator of the transcription of Col11a2 (Liu et al., 2000). Although SOX9-binding enhancer element has not been identified in col11a1 promoter so far, col11a1-morphants showed overlapping cartilage morphogenetic defects with the zebrafish jellyfish (jef) mutants harboring two mutations that disrupt sox9a (Yan et al., 2002). Yan and coworkers suggested that the failure of jef mutant to activate cartilage col2a1 expression underlies the defects in cartilage morphogenesis, particularly those affecting chondrocytes stacking and cartilage shaping. We showed here that cartilage formation is seemingly altered in col11a1-morphants albeit collagen II fibrils are deposited in the cartilage matrix (see below) suggesting that the composition of the extracellular matrix is subtly adapted to tissue formation and function. Inappropriate composition of the extracellular matrix may directly affect chondrocyte signaling. Integrins are the major collagen receptors. Loss of α10β1 integrin expression in mouse was shown to disturb columnar arrangement and shape of chondrocytes (Bengtsson et al., 2005), a phenotype very similar to the defects observed in col11a1 morphants. Although col11a1 expression was detected in pectoral fins at 48 hpf, no morphological defect was observed in the pectoral fins of the col11a1-morphants. The absence of phenotype can be explained by the fact that pectoral fin endoskeleton and craniofacial cartilage chondrocytes are from different embryonic origins (Grandel and SchulteMerker, 1998).
D. Baas et al. / Matrix Biology 28 (2009) 490–502
497
Fig. 7. Pharyngeal markers. A–H, In situ hybridizations with krox-20 (A, B) and dlx2a (C–H). At 20 hpf, krox-20 expression is similar in wild type (A) and in col11a1MO embryos (B). At 15 hpf, dlx2a is expressed in the three neural crest streams in wild type embryos (C) and in morphants (D) though reduced signal is observed. In 20 hpf wild type (E), dlx2a expressing neural crest cells have populated the arches to form the pharyngeal pouches (p) clearly observed at 30 hpf (G). Note that the dlx2a expression domain observed in wild type at 20 hpf (E), corresponds to that observed at 30 hpf in morphants (H). I, J Zn-8 antibody is used as a marker of the formation of pharyngeal pouches (in green). Confocal microscopy stacks of 33 hpf wild type and col11a1MO-injected embryos reveal the presence of the five pharyngeal pouches. Three of them are shown in I (wild type) and J (morphant). Nuclei (in blue) are stained with TOTO®-3. A–H, Dorsal views, posterior side to the right. I,J Lateral views, posterior side to the right. s1–3, neural crest streams; p1–4, pharyngeal pouches; end pouches, endodermal pouches.
3.3. Collagen XI α1 knockdown affects zebrafish extracellular matrix organization It has been shown that collagen XI regulates the nucleation and the lateral growth of cartilage heterotypic fibrils containing collagen II (Blaschke et al., 2000). Our data add support to these observations. Very few fibrils were found in cartilage elements of 50 hpf morphants compared to wild type. This is consistent with the finding that collagen XI nucleates the formation of heterotypic fibrils. At 6 dpf, along with a noticeable increase in collagen II expression, the cartilage extracellular matrix of coll11a1-morphants exhibited abnormally thick collagen fibrils with a distinct banded pattern. The control of fibril growth is mediated by the incorporation of collagen XI molecules into collagen II
fibrils during fibril formation (Mendler et al., 1989; Eyre, 2004). In vitro fibrillogenesis studies showed that control of lateral growth is observed for mixtures of collagens II and XI as long as the molar excess of collagen II was less than 8-fold (Blaschke et al., 2000). In agreement to this observation, overexpression of collagen II transgenic mice resulted in the presence of abnormally wide fibrils with a strong banding pattern in the growth cartilage (Garofalo et al., 1993). Our data underscore the importance of the collagen II/XI ratio for proper fibrillogenesis in cartilage. Such abnormal large fibrils were also described in the cho cartilage (Seegmiller and Monson, 1982; Li et al., 1995). In cho mice, an alternative hybrid collagen XI molecule was shown to assemble in which the closely related collagen V α1 chain replaces the missing collagen XI α1 chain (Fernandes et al., 2007). They concluded that the
498
D. Baas et al. / Matrix Biology 28 (2009) 490–502
Fig. 8. Chondrocyte differentiation markers. A–H, In situ hybridizations of 3 dpf wild type (A, B, E, F) and col11a1MO (C, D, G, H) larvae with sox9a (A–D) and col2a1 (E–H) probes. sox9a (A, B) and col2a1 (E, F) are both strongly expressed in wild type pharyngeal cartilages. Dorsal (C, G) and lateral (D, H) views of col11a1MO-injected embryos show that the remnant pharyngeal cartilages are sox9a (C, D) and col2a1 (G, H) positive. Note that the level of col2a1 expression is strongly increased in morphants (G, H) compared to wild type (E, F) larvae. I, J Immunofluorescence staining of 6 dpf wild type larvae (I) and morphants (J) with polyclonal antibodies to collagen II. An intense staining is observed in chondrocytes of the remaining col11a1MO pharyngeal arches (J) compared to wild type (I). The staining also reveals that, in contrast to wild type, chondrocyte stacking is defective in morphant cartilage. ov, otic vesicle, pf, pectoral fin.
native type XI collagen molecules containing the collagen XI α1 chain are required to form uniformly thin fibrils. The ortholog of mammalian col5a1 gene has not been described in zebrafish yet. It would have been interesting to see whether a similar mechanism can occur in col11a1morphants. Consistent with previous observations in cho mice (Li et al., 1995), less proteoglycan aggregates were observed in the cartilage matrix of col11a1-morphants. Collagen XI α1 chain contains three different heparin binding sites that can bind heparan sulfate proteoglycans (Delacoux et al., 2000; Vaughan-Thomas et al., 2001; Blum et al., 2006). It is thus conceivable that the lack of collagen XI α1 alters retention of proteoglycans in cartilage tissues and, consequently affects interactions
with other matrix components. Altogether, our data support the concept (Hansen and Bruckner, 2003) that the formation of inappropriate collagen composites and consequently abnormal fibril organization underlies the functional deficiencies in collagen-related diseases. 3.4. Collagen XI α1 knockdown affects zebrafish perichordal sheath structure Our in situ hybridizations revealed that col11a1 is transiently expressed in zebrafish developing notochord as described for col2a1 (Yan et al., 1995). We showed that knockdown of col11a1 expression in
Fig. 9. Histological analysis of craniofacial cartilages. A–D, Longitudinal sections through 6 dpf wild type (A, C) and col11a1MO-injected larvae (B, D) stained with Methylen/blueAzur II. In wild type, pharyngeal arches are separated from each other by the gill clefts and are well differentiated (A). In contrast, the organization of cartilage elements and gill clefts are completely disorganized in col11a1MO-injected embryos (B). Most of the cartilage elements are present in morphant larvae (B) compared to wild type (A) but they are smaller and misshapen. At higher magnification, the chondrocytes that are hypertrophic in wild type (C), appeared less differentiated and are not organized into stacks. (A, B) Asterisk indicates the location of the mouth opening. ch, ceratohyal; ep, ethmoid plate; cb1–cb5, ceratobranchial arches 1 to 5; ov, otic vesicle. Scale bars = 100 µm (A, B), 20 µm (C, D).
D. Baas et al. / Matrix Biology 28 (2009) 490–502
499
Fig. 10. Ultrastructure of the pharyngeal cartilage. A–H, Ultrathin sections of ceratobranchial cartilage of wild type (A, C, E, G) and col11a1MO-injected embryos (B, D, F, H) at 50 hpf (A–D) and 6 dpf (E–H). Well-organized chondrocyte stacks are observed in 50 hpf wild type embryos (A) whereas chondrocytes in col11a1MO-injected embryos fail to form rows (B). Only few sparse filamentous structures are present in the extracellular space of col11a1MO morphant cartilage (D) while an abundant network of thin, randomly organized fibrils is observed in wild type embryos (C). At 6 dpf, chondrocytes in wild type larvae are hypertrophic (E) and the matrix network displays thin unbanded fibrils entrapping proteoglycans aggregates (G). Chondrocyte differentiation in col11a1MO-injected larvae (F) appears to be delayed compared to wild type (E). Collagen fibers are thicker in col11a1MO-injected larvae and banding pattern is visible. Sparse proteoglycan aggregates are observed (H). Scale bars = 2 µm (A, B, E, F), 500 nm (C, D), 200 nm (G, H).
zebrafish results in notochord distortion likely due to defects in perichordal extracellular matrix structure. The pressure generated by the vacuolated cells and the perichordal sheath both contribute to the notochord stiffness (Stemple, 2005). Laminin 1 and collagen IV are expressed in the surrounding basement membrane as in mammals. More recently, we have reported the expression of the basement membrane collagen type XV in zebrafish notochord (Pagnon-Minot et al., 2008). The loss of one of these components disrupts the basement membrane organization and notochord differentiation (Stemple, 2005; Paulus and Halloran, 2006; Pagnon-Minot et al., 2008). Although we did not observe gross defects in notochord differentiation in collagen XI α1 morphants, it is not excluded that collagen XI depletion may slightly affect the final stages of notochord development. The long-term consequence of the notochord defects in the formation of the vertebral column cannot be approached using knockdown strategy.
3.5. Appraisal of col11a1-morphants as a model for chondrodysplasia Mutations in the human gene COL11A1 have been reported to cause the autosomal dominant craniofacial disorders, the Stickler/Marshall syndromes (Griffith et al., 1998; Majava et al., 2007). Whereas Stickler syndrome exhibits genetic heterogeneity, Marshall syndrome phenotype was associated with mutations in COL11A1 gene only. Clinical changes include facial anomalies, cleft palate and hearing defects. The formation of teeth is generally not altered in Sticker/Marshall syndrome. The Marshall syndrome is distinct from the Stickler syndrome because of the relative rarity of cleft palate. Facial abnormalities were similarly observed in col11a1MO-injected embryos. Although cleft palate was not observed at the morpholino dose used in this study, the skeletal palate was malformed and mouth formation was clearly altered in all col11a1morphants. Pharyngeal teeth were present in col11a1-morphants as in
500
D. Baas et al. / Matrix Biology 28 (2009) 490–502
Fig. 11. Structure of the notochord. A, B, Lateral views of alcian blue staining of 34 hpf wild type (A) and col11a1MO-injected (B) embryos revealing the structure of the notochord. Contrary to the rectilinear structure of wild type notochord (A), the notochord of the morphant is distorted and exhibits folds (B). C, D TEM observations of ultrathin sections through the tail of wild type (A) and col11a1MO-injected (D) 50 hpf embryos. The perichordal sheath is composed of three different layers: an internal basement membrane (arrows), a middle fibrillar layer (arrowhead), and an external granular layer (asterisk). In col11a1MO-injected (D), the basement membrane exhibits clear interruptions (open arrowheads), the thickness of fibrillar layer (arrowhead) and the granular layer (asterisk) are reduced and the fibrils are loosely organized compared to wild type (C). C, D, scale bar = 500 nm.
flathead and jef zebrafish craniofacial mutants, despite the absence of cb5 arch cartilage that supports these teeth (Schilling et al., 1996). Our data reinforce the view that teeth formation occurs independently of properly differentiated cartilage or bone attachment. The Marshall syndrome is associated with sensorineural hearing loss (Griffith et al., 1998). It has been proposed that hearing loss caused by COL11A2 gene mutations in humans can result from deleterious effects on extracellular matrix organization in the tectorial membrane where collagen XI is expressed along with collagen II (Vikkula et al., 1995; McGuirt et al., 1999). Coll11a1 haploinsufficiency did not cause significant hearing loss in cho/+ mouse (Szymko-Bennett et al., 2003). Zebrafish does not possess outer or middle ears but have an inner ear whose development and function appear to be well conserved (Whitfield, 2002). The otoliths which are composed of a stroma-like structure is involved in perception of equilibrium and sound. Although zebrafish col11a1 is expressed in otic vesicle, the development of ear appeared phenotypically unaffected in col11a1-morphants. Direct response to auditory stimulus can be performed in adult zebrafish (Whitfield, 2002). Unfortunately, the morpholino-knockdown technique does not permit assessment of the long-term effect of gene inactivation. In conclusion, the defects in cartilage formation reported here are in good agreement with human disorders caused by COL11A1 gene mutations (Griffith et al., 1998; Majava et al., 2007) and with the phenotypic defects observed in cho mice in which a spontaneous mutation in col11a1 gene has been identified (Li et al., 1995). The homozygous cho mice die at birth and most studies were performed on heterozygous mice. Col11a1-morphants provide an alternative animal model for human disorders caused by mutations in collagen XI genes and for the development of therapeutics to repair cartilage defects in collagen humans. Moreover, our study may further help in the identification of a col11a1 mutant from zebrafish screen for craniofacial mutants. 4. Experimental procedures 4.1. Maintenance of zebrafish General maintenance, collection and staging of zebrafish were carried out at the zebrafish facilities (PRECI, IFR 128) as described
(Westerfield, 2001). The developmental stages are given in hour postfertilization (hpf) and day post-fertilization (dpf) at 28.5 °C, according to morphological criteria. 4.2. Cloning of zebrafish collagen XI and sequence analysis Zebrafish genomic clones (GenBank accession numbers AL672176 and XM_685088) reported as partial sequences of the col11a2 were found in database. Both sequences showed high homology with col11a2 and with col11a1 human genes (GenBank accession numbers P13942 and P12107 respectively). The sequence reported as AL672176 lacked the N-terminal part of the collagen XI pro-chain. To cover the entire coding part, 5′RACE was employed using GeneRacer 5′ primer CGACTGGAGCACGAGGACACTGA-3′ (Invitrogen) and a reverse primer sequence derived from the reported nucleotide sequence (AL672176) 5′-ACCAGAGTCATGATGGAGAAGTTCTCCGGG-3′. A 580 nt cDNA fragment was obtained including the UTR region, the missing 5′ region and a region overlapping the reported sequence. From these data (that were confirmed by sequences reported afterward, XM_685088 and CO958254), the full-length cDNA of the pro-chain of collagen XI was deduced (5426 nt) and submitted to databases (accession number AM287214). To determine whether the identified zebrafish sequence encodes proα2(XI) or proα1(XI), alignments of the deduced amino acid sequence of the zebrafish proα1/α2(XI) with the human proα2(XI), proα1(XI), proα1(V) chain sequences (GenBank accession numbers P13942, P12107 and P20908 respectively) and the pro-α1(V/XI) chain sequence of red sea-bream (AB045975) were performed using CLUSTALW (NCBI). 4.3. Quantitative RT-PCR Total RNA was prepared from wild type embryos at different stages using a published modification of the guanidinium isothiocyanate procedure (Chanut-Delalande et al., 2004). Reverse transcription (RT) of 1 µg of RNA was performed using Expand reverse transcriptase (Roche). Real-time PCR was performed to quantify the relative abundance of col11a1 transcripts in wild-type developing embryos at different stages. The assays were carried out with a DNA Engine Opticon2 (MJ research)
D. Baas et al. / Matrix Biology 28 (2009) 490–502
using intercalation of SYBR Green as fluorescent reporter following the manufacturer's protocol (Quantitect SYBR Green PCR kit, Qiagen). The primers used were specific for col11a1 (forward primer 5'CCCGGAGAACTTCTCCATCATGACTCTGGT-3′; reverse primer 5′CGATGCTGGTCCTCATACAG -3′), and histone (forward primer 5′- CCTCGAGCTGGCCGGGAA-3′; reverse primer 5′-CTCGGACTAGCTGCGTTT-3′). Reactions were performed in triplicate from three separate RNA preparations and thermal cycling conditions consisted of an initial denaturation step of 94 °C for 2 min, 11 cycles of 94 °C for 15 s, annealing and extension at 60 °C for 30 s, followed by a final elongation step of 68 °C for 3 min. Relative gene expression was determined using the 2−ΔΔCT method as previously described (Chanut-Delalande et al., 2004). Mean fold changes in col11a1 gene expression were calculated for the different stages. Variations were normalized to the relative expression obtained at stage 24 hpf, and resulting values were plotted as histograms. 4.4. Morpholinos Morpholinos (MO) were designed with sequence complementary to col11a1 cDNA in a location just upstream to the initial start codon and purchased from Gene Tools (Philomath, OR). The sequences of the col11a1 morpholino antisense oligonucleotides used were: col11a1MO 5′-TCCGCTTCTTCCGAATATCCATAGT-3′ (predicted start codon is underlined) and mcol11a1MO (five-base mismatches, in bold) 5′-TCGGGTTCTTCCCAATATCGATACT-3′. The sequence of the col1a1 morpholino used was: col1a1MO 5′-gCCAGCAGAATATCCACAAAGCTGA-3′. Col11a1MO was tagged with fluorescein at the 3′ end to monitor success of injection and even distribution of morpholinos in the embryos. 1- to 2- cell stage embryos were injected with different quantities of morpholinos (5 to 40 pmol) in 1X Danieau buffer (58 mM NaCl, 0.7 mM KCl, 0.4 mM MgSO4, 0.6 mM Ca(NO3)2, 5 mM Hepes pH 7.6). The morphology of the morphants was observed with a Leica stereoscope equipped with a digital camera. In all experiments, MO-injected embryos were compared to mcol11a1MO-injected embryos and with non-injected embryos. When none indicated, col11a1MO-injected embryos will refer to embryos injected with 10 pmol. 4.5. Whole-mount in situ hybridization Whole-mount in situ hybridizations were performed as previously described (Pagnon-Minot et al., 2008). Riboprobe for col11a1 was transcribed from a template comprising 300 bp from nt 251-nt 551 localized within the N-terminal domain referred to as TSPN. The following primers was used to amplify the probe from zebrafish cDNA (forward: 5′-CCCGGAGAACTTCTCCATCATGACTCTGGT-3′ and reverse 5′CCTGGAAAACCTCCTCATCCAGGAGACGGGCTC-3′). A sense RNA probe complementary to the specific col11a1 probe was prepared as negative control. Sox9a and col2a1 RNA probes correspond to 572 nt and 447 nt respectively and were prepared using the following primers: sox9a forward 5′-GCCAGAGCGAATCTGAAGAC-3′, sox9a reverse 5′TGCTGTAATGCTGGAGATGC-3′, col2a1 forward 5′- GAGAGTGTTTTCGATACCTCAC-3′, col2a1 reverse 5′- CACCATCCAGCTGAACTTCCTC-3′. The myoD (Weinberg et al., 1996), krox-20 (Taneja et al., 1996), and dlx2a (Borday-Birraux et al., 2006) RNA probes were prepared as described (Bader et al., 2008). Embryos were observed with a Leica stereomicroscope. For sections, embryos hybridized with col11a1 probe were embedded in epoxy resin as described below and 10–15 µm sections were performed. 4.6. Confocal microscopy Immunofluorescence labeling of frozen sections was performed as previously described (Bader et al., 2008). Purified polyclonal antibodies against human collagen II and collagen XI were purchased from Novotec (Lyon, France). For whole-mount immunofluorescence labeling, embryos were fixed overnight at 4 °C in 4% paraformaldehyde. Fixed
501
embryos were blocked and permeabilized using a solution containing 5% goat serum, 0.5% Triton X100 in PBS. Embryos were incubated with Zn8 or Zn1 zebrafish neuronal marker monoclonal antibodies (Developmental studies, Hybridoma Bank, University of Iowa, USA). After an overnight incubation with primary antibody at 4 °C, embryos were intensively washed in PBS at room temperature. Embryos were then incubated with the secondary antibody conjugated to Cy-3 (Jackson ImmunoReseach) at room temperature for 1 h, treated with RNase and finally incubated with TOTO®-3 iodide (642/660) (molecular probes) to visualize nuclei. Sections and whole-mount embryos were observed with a Zeiss LSM510 confocal microscope (PLATIM, IFR128). Images were processed using Abode photoshop and Imaris softwares. 4.7. Skeletal staining For cartilage analysis, larvae were fixed overnight at 4 °C in 4% paraformaldehyde and stained overnight in 0.1% alcian Blue in methanol/glacial acetic acid (7:3). After gradual re-hydration, larvae tissues were cleared with glycerol and defects in cartilage formation were analyzed according to Kimmel et al. (2001). Skeletal bones were visualized using the vital stain Quercetin (Sigma) that stains calcium. 7 dpf larvae were incubated in Quercetin (0.1 mg/ml) for 24 h and mineralized bones were visualized at 288 nm with a Zeiss LSM510 confocal microscope. 4.8. Transmission electron microscopy Whole embryos were fixed in a mixture of aldehydes and embedded in Epoxy as previously described (Le Guellec et al., 2004). Thin sections were stained with Methylene blue/Azur II (vol: vol) and observed with a Leica light microscope equipped with a digital camera (Nikon). Ultra thin sections were contrasted with uranyl acetate and lead citrate, and examined with a Philips CM 120 electron microscope. 4.9. Western blot Total proteins from 24 hpf wild type and morphant embryos were extracted in Tris/HCl 100 mM pH 6.8, 2% SDS overnight at 4 °C and centrifuged to eliminate insoluble materials. Protein concentration of supernatants was determined using the Bradford protein determination kit (BioRad). Equal protein quantities of samples were analyzed by 6% SDS-PAGE under reducing conditions followed by electrotransfer onto polyvinylidene difluoride membranes (Immobilon-P, Millipore) overnight in 10 mM CAPS (pH 11), 5% methanol. After saturation, membranes were incubated with rabbit polyclonal antibodies against human pepsinized collagen XI (Novotec, France), or with actin antibody (AC15, Sigma) for protein loading control, followed with the secondary antibodies conjugated to horseradish peroxidase. The immunoreactive bands were revealed using a chemiluminescence detection kit (ECL plus, Amersham, Biosciences). Acknowledgments We thank Sylvain Cogne (IBCP, Lyon) for his technical assistance with real-time PCR and Dr. Bernard and Christine Thisse for kindly providing in situ hybridization probes. This work was supported by grants from the Emergence Research Program (Région Rhône-Alpes) and from the University Lyon 1 (BQR). We thank Laure Bernard (PRECI, IFR 128 Biosciences Gerland, Lyon) for breeding zebrafish and providing embryos. Confocal microscopy observations and PCR-Q experiments were performed respectively at the PLATIM and the “Plateau d'Analyse Génétique” core facilities of the IFR 128 Biosciences-Gerland (Lyon). Electron microscopy (TEM and SEM) samples were observed at the “Centre Technique des Microstructures” (Université Lyon 1, Villeurbanne, France).
502
D. Baas et al. / Matrix Biology 28 (2009) 490–502
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.matbio.2009.07.004. References Bader, H.L., Charvet, B., Veit, G., Driever, W., Koch, M., Ruggiero, F., 2008. The zebrafish ortholog of the col12a1 gene is ubiquitously expressed in embryonic connective tissue sheaths (fascia). Matrix Biol. 28, 32–43. Bell, D.M., Leung, K.K., Wheatley, S.C., Ng, L.J., Zhou, S., Ling, K.W., Sham, M.H., Koopman, P., Tam, P.P., Cheah, K.S., 1997. SOX9 directly regulates the type-II collagen gene. Nat. Genet. 16, 174–178. Bengtsson, T., Aszodi, A., Nicolae, C., Hunziker, E.B., Lundgren-Akerlund, E., Fässler, R., 2005. Loss of alpha10beta1 integrin expression leads to moderate dysfunction of growth plate chondrocytes. J. Cell Sci. 118, 929–936. Blaschke, U.K., Eikenberry, E.F., Hulmes, D.J., Galla, H.J., Bruckner, P., 2000. Collagen XI nucleates self-assembly and limits lateral growth of cartilage fibrils. J. Biol. Chem. 275, 10370–10378. Bonod-Bidaud, C., Beraud, M., Vaganay, E., Delacoux, F., Font, B., Hulmes, D.J.S., Ruggiero, F., 2007. Enzymatic cleavage specificity of the proα 1(V) chain processing analyzed by site-directed mutagenesis. Biochem. J. 405, 299–306. Borday-Birraux, V., Van der Heyden, C., Debiais-Thibaud, M., Verreijdt, L., Stock, D.W., Huysseune, A., Sire, J.Y., 2006. Expression of Dlx genes during the development of the zebrafish pharyngeal dentition: evolutionary implications. Evol. Dev. 8, 130–141. Chanut-Delalande, H., Bonod-Bidaud, C., Cogne, C., Malbouyres, M., Ramirez, F., Fichard, A., Ruggiero, F., 2004. Development of a functional skin matrix requires deposition of collagen V heterotrimers. Mol. Cell. Biol. 24, 6049–6057. Cheah, K.S., Lau, E.T., Au, P.K., Tam, P.P., 1991. Expression of the mouse alpha 1(II) collagen gene is not restricted to cartilage during development. Development 111, 945–953. de Combrughe, B., Lefebvre, V., Behringer, R.R., Bi, W., Murakami, S., Huang, W., 2000. Transcriptional mechanisms of chondrocyte differentiation. Matrix Biol. 19, 389–394. Delacoux, F., Fichard, A., Cogne, S., Garrone, R., Ruggiero, F., 2000. Unraveling the amino-acid sequence crucial for heparin binding to collagen V. J. Biol. Chem. 275, 29377–29382. Eyre, D.R., 2004. Collagens and cartilage matrix homeostasis. Clin. Orthop. Relat. Res. 427, 118–122. Fernandes, R.J., Weis, M., Scott, M.A., Seegmiller, R.E., Eyre, D.R., 2007. Collagen XI chain misassembly in cartilage of the chondrodysplasia (cho) mouse. Matrix Biol. 26, 597–603. Fichard, A., Kleman, J.P., Ruggiero, F., 1995. Another look at collagen V and XI molecules. Matrix Biol. 14, 515–531. Garofalo, S., Metsäranta, M., Ellard, J., Smith, C., Horton, W., Vuorio, E., de Crombrugghe, B., 1993. Assembly of cartilage collagen fibrils is disrupted by overexpression of normal type II collagen in transgenic mice. Proc. Natl. Acad. Sci. 90, 3825–3829. Grandel, H., Schulte-Merker, S., 1998. The development of the paired fins in the zebrafish (Danio rerio). Mech. Dev. 79, 99–120. Griffith, A.J., Sprunger, L.K., Sirko-Osadsa, D.A., Tiller, G.E., Meisler, M.H., Warman, M.L., 1998. Marshall syndrome associated with a splicing defect at the COL11A1 locus. Am. J. Hum. Genet. 62, 816–823. Hansen, U., Bruckner, P., 2003. Macromolecular specificity of collagen fibrillogenesis: fibrils of collagens I and XI contain a heterotypic alloyed core and a collagen I sheath. J. Biol. Chem. 278, 37352–37359. Imamura, Y., Steiglitz, B.M., Greenspan, D.S., 1998. Bone morphogenetic protein-1 processes the NH2-terminal propeptide, and a furin-like proprotein convertase processes the COOH-terminal propeptide of pro-alpha1(V) collagen. J. Biol. Chem. 273, 27511–27517. Ingham, P.W., 2009. The power of the zebrafish for disease analysis. Hum. Mol. Genet. 18, R107–R112. Jakkula, E., Melkoniemi, M., Kiviranta, I., Lohiniva, J., Raina, S.S., Perala, M., Warman, M.L., Ahonen, K., Kroger, H., Goring, H.H., Ala-Kokko, L., 2005. The role of sequence variations within the genes encoding collagen II, IX and XI in non-syndromic, earlyonset osteoarthritis. Osteoarthr. Cartil. 6, 497–507. Kimmel, C.B., Miller, C.T., Moens, C.B., 2001. Specification and morphogenesis of the zebrafish larval head skeleton. Dev. Biol. 233, 239–257. Kuivaniemi, H., Tromp, G., Prockop, D.J., 1997. Mutations in fibrillar collagens (types I, II, III, and XI), fibril-associated collagen (type IX), and network-forming collagen (type X) cause a spectrum of diseases of bone, cartilage, and blood vessels. Hum. Mutat. 9, 300–315. Le Guellec, D., Morvan-Dubois, G., Sire, J.Y., 2004. Skin development in bony fish with particular emphasis on collagen deposition in the dermis of the zebrafish (Danio rerio). Int. J. Dev. Biol. 48, 217–231. Li, Y., Lacerda, D.A., Warman, M.L., Beier, D.R., Yoshioka, H., Ninomiya, Y., Oxford, J.T., Morris, N.P., Andrikopoulos, K., Ramirez, F., Wardell, B.B., Lifferth, G.D., Teuscher, C., Woodward, S.R., Taylor, B.A., Seegmiller, R.E., Olsen, B.R., 1995. A fibrillar collagen gene, Col11a1, is essential for skeletal morphogenesis. Cell 80, 423–430. Li, S.W., Takanosu, M., Arita, M., Bao, Y., Ren, Z.X., Maier, A., Prockop, D.J., Mayne, R., 2001. Targeted disruption of Col11a2 produces a mild cartilage phenotype in transgenic mice: comparison with the human disorder otospondylomegaepiphyseal dysplasia (OSMED). Dev. Dyn. 222, 141–152. Liu, Y., Li, H., Tanaka, K., Tsumaki, N., Yamada, Y., 2000. Identification of an enhancer sequence within the first intron required for cartilage-specific transcription of the alpha2(XI) collagen gene. J. Biol. Chem. 275, 12712–12718.
McGuirt, W.T., Prasad, S.D., Griffith, A.J., Kunst, H.P., Green, G.E., Shpargel, K.B., Runge, C., Huybrechts, C., Mueller, R.F., Lynch, E., King, M.C., Brunner, H.G., Cremers, C.W., Takanosu, M., Li, S.W., Arita, M., Mayne, R., Prockop, D.J., Van Camp, G., Smith, R.J., 1999. Mutations in COL11A2 cause non-syndromic hearing loss (DFNA13). Nat. Genet. 23, 413–419. Majava, M., Hoornaert, K.P., Bartholdi, D., Bouma, M.C., Bouman, K., Carrera, M., Devriendt, K., Hurst, J., Kitsos, G., Niedrist, D., Petersen, M.B., Shears, D., StolteDijkstra, I., Van Hagen, J.M., Ala-Kokko, L., Männikkö, M., Mortier, G.R., 2007. A report on 10 new patients with heterozygous mutations in the COL11A1 gene and a review of genotype–phenotype correlations in type XI collagenopathies. Am. J. Med. Genet. 143, 258–264. Medeck, R.J., Sosa, S., Morris, N., Oxford, J.T., 2003. BMP-1-mediated proteolytic processing of alternatively spliced isoforms of collagen type XI. Biochem. J. 376, 361–368. Mendler, M., Eich-Bender, S.G., Vaughan, L., Winterhalter, K.H., Bruckner, P., 1989. Cartilage contains mixed fibrils of collagen types II, IX, and XI. J. Cell Biol. 108, 191–197. Morvan-Dubois, G., Le Guellec, D., Garrone, R., Zylberberg, L., Bonnaud, L., 2003. Phylogenetic analysis of vertebrate fibrillar collagen locates the position of zebrafish alpha3(I) and suggests an evolutionary link between collagen alpha chains and hox clusters. J. Mol. Evol. 57, 501–514. Pagnon-Minot, A., Malbouyres, M., Haftek-Terreau, Z., Kim, R., Sasaki, T., Thisse, C., Thisse, B., Ingham, P.W., Ruggiero, F., Le Guellec, D., 2008. Collagen XV, a novel factor in zebrafish notochord differentiation and muscle development. Dev. Biol. 316, 21–35. Parsons, M.J., Pollard, S.M., Saude, L., Feldman, B., Coutinho, P., Hirst, E.M.A., Stemple, D.L., 2002. Zebrafish mutants identify an essential role for laminins in notochord formation. Development 129, 3137–3146. Paulus, J.D., Halloran, M.C., 2006. Zebrafish bashful/laminin-alpha 1 mutants exhibit multiple axon guidance defects. Dev. Dyn. 235, 213–224. Piotrowski, T., Nusslein-Volhard, C., 2000. The endoderm plays an important role in patterning the segmented pharyngeal region in zebrafish (Danio rerio). Dev. Biol. 225, 339–356. Ricard-Blum, S., Ruggiero, F., 2005. The collagen superfamily: from the extracellular matrix to the cell membrane. Pathol. Biol. 53, 430–442. Ricard-Blum, S., Beraud, M., Raynal, N., Farndale, R.W., Ruggiero, F., 2006. Structural requirements for heparin/heparan sulfate binding to type V collagen. J. Biol. Chem. 281, 25195–25204. Schilling, T.F., Piotrowski, T., Grandel, H., Brand, M., Heisenberg, C.P., Jiang, Y.J., Beuchle, D., Hammerschmidt, M., Kane, D.A., Mullins, M.C., van Eeden, F.J., Kelsh, R.N., FurutaniSeiki, M., Granato, M., Haffter, P., Odenthal, J., Warga, R.M., Trowe, T., Nüsslein-Volhard, C., 1996. Jaw and branchial arch mutants in zebrafish I: branchial arches. Development 123, 329–344. Seegmiller, R.E., Monson, C.B., 1982. Scanning electron microscopy of cartilage in mice with hereditary chondrodysplasia. Scan. Electron Microsc. 3, 1259–1267. Stemple, D.L., 2005. Structure and function of the notochord: an essential organ for chordate development. Development 132, 2503–2512. Spranger, J., 1998. The type XI collagenopathies. Pediatr. Radiol. 28, 745–750. Szymko-Bennett, Y.M., Kurima, K., Olsen, B., Seegmiller, R., Griffith, A.J., 2003. Auditory function associated with Col11a1 haploinsufficiency in chondrodysplasia (cho) mice. Hear Res. 175, 178–182. Taneja, R., Thisse, B., Rijli, F.M., Thisse, C., Bouillet, P., Dolle, P., Chambon, P., 1996. The expression pattern of the mouse receptor tyrosine kinase gene MDK1 is conserved through evolution and requires Hoxa-2 for rhombomere-specific expression in mouse embryos. Dev. Biol. 177, 397–412. Touhata, K., Tanaka, H., Yokoyama, Y., Sakaguchi, M., Toyohara, H., 2001. Structure of a full-length cDNA clone for the pro-alpha1(V/XI) collagen chain of red seabream. Biochim. Biophys. Acta 1517, 323–326. Vaughan-Thomas, A., Young, R.D., Phillips, A.C., Duance, V.C., 2001. Characterization of type XI collagen-glycosaminoglycan interactions. J. Biol. Chem. 276, 5303–5309. Vikkula, M., Mariman, E.C., Lui, V.C., Zhidkova, N.I., Tiller, G.E., Goldring, M.B., van Beersum, S.E., de Waal Malefijt, M.C., van den Hoogen, F.H., Ropers, H.H., Mayne, R., Cheah, K.S.E., Olsen, B.R., Warman, M.L., Brunner, H.G., 1995. Autosomal dominant and recessive osteochondrodysplasias associated with the COL11A2 locus. Cell 80, 431–437. Weinberg, E.S., Allende, M.L., Kelly, C.S., Abdelhamid, A., Murakami, T., Andermann, P., Doerre, O.G., Grunwald, D.J., Riggleman, B., 1996. Developmental regulation of zebrafish MyoD in wild-type, no tail and spadetail embryos. Development 122, 271–280. Westerfield, M., 2001. The zebrafish book. A guide for the laboratory use of zebrafish (Danio rerio). University of Oregon Press, Eugene, Oregon. Whitfield, T.T., 2002. Zebrafish as a model for hearing and deafness. J. Neurobiol. 53, 157–171. Yan, Y.L., Hatta, K., Riggleman, B., Postlethwait, J.H., 1995. Expression of a type II collagen gene in the zebrafish embryonic axis. Dev. Dyn. 203, 363–376. Yan, Y.L., Miller, C.T., Nissen, R.M., Singer, A., Liu, D., Kirn, A., Draper, B., Willoughby, J., Morcos, P.A., Amsterdam, A., Chung, B.C., Westerfield, M., Haffter, P., Hopkins, N., Kimmel, C., Postlethwait, J.H., 2002. A zebrafish sox9 gene required for cartilage morphogenesis. Development 129, 5065–5079. Yan, Y.L., Willoughby, J., Liu, D., Crump, J.G., Wilson, C., Miller, C.T., Singer, A., Kimmel, C., Westerfield, M., Postlethwait, J.H., 2005. A pair of Sox: distinct and overlapping functions of zebrafish sox9 co-orthologs in craniofacial and pectoral fin development. Development 132, 1069–1083. Yelick, P.C., Schilling, T.F., 2002. Molecular dissection of craniofacial development using zebrafish. Crit. Rev. Oral Biol. Med. 13, 308–322. Yoshioka, H., Iyama, K., Inoguchi, K., Khaleduzzaman, M., Ninomiya, Y., Ramirez, F., 1995. Developmental pattern of expression of the mouse alpha1(XI) collagen gene (Col11a1). Dev. Dyn. 204, 41–47.
