SEARCH

SEARCH BY CITATION

Keywords:

  • Sulf1;
  • somite;
  • chondrogenesis;
  • mesenchymal condensation;
  • perichondrium;
  • joint formation

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

A very dynamic and localised spatiotemporal expression pattern of Sulf1 was observed in axial structures and different regions of developing quail somites that included myotomal and sclerotomal regions at specific levels. Sulf1 expression was also observed in not only the scapular and pelvic girdle forming regions of the quail limb that connect the appendicular skeleton to the body trunk but also the cartilage templates of the appendicular skeleton. The highest expression level of Sulf1 was observed in condensing mesenchyme, during the early differentiation stage of chondrogenesis, and highly dynamic expression was observed in the perichondrial and joint-forming regions. Overexpression of Sulf1 in quail micromass cultures enhanced aggregation and differentiation of prechondrocytes into chondrogenic lineage supporting its role in mesenchymal condensation and early differentiation of cartilaginous elements. The exposure of digital explants to high levels of Sulf1 expression in vitro led to increased growth of the original 1st phalange but complete inhibition of joint formation and generation of any further phalanges. Sulf1 thus plays a key role during multiple stages of cartilage development and joint formation. Developmental Dynamics 235:3327–3335, 2006. © 2006 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The modulation of the cell signalling, spatially or temporally, leading to patterning information is accomplished by regulating the expression of specific ligands and their receptors. However, developing organisms also utilise various other ways to regulate cell signalling, including the secretion of enzymes such as Sulf1 (Dhoot et al.,2001). Sulf1 is a member of an evolutionarily conserved sulfatase family related to the lysosomal N-acetly glucosamine sulfatases that catalyze the hydrolysis of 6-0 sulfates from N-acetyl glucosamines of heparin sulphate proteoglycans (HSPGs) (Robertson et al.,1992). Studies of Sulf1 have demonstrated a novel mechanism for modulating the activities of multiple growth factors and signalling molecules such as Wnt signalling (Dhoot et al.,2001; Ai et al.,2003), BMP signalling (Viviano et al.,2004), and fibroblast growth factor regulation (Wang et al.,2004), which are known to play major roles during embryogenesis, through the patterned expression of this secreted sulfatase. For example, the Wnt family in vertebrates, comprised of more than 20 members of secreted glycosylated factors (Tian et al.,2005), has been shown to play a range of important functions during development ranging from patterning individual structures to controlling differentiation, proliferation, and survival of cells. Wnt signals are transmitted through binding to transmembrane receptors, including the frizzled family of receptors (Tamai et al.,2000; Wehrli et al.,2000; Pandur and Kühl,2001), which can be further modified by co-receptors like Ror2 and LRP5/6 (Oishi et al.,2003; Billiard et al.,2005). In myogneic precursors, Sulf1 has been shown to regulate heparin-dependent Wnt signalling required for MyoD expression (Dhoot et al.,2001) by modulating Glypican 1, a heparan sulfate proteoglycan (Ai et al.,2003). In addition, research by other laboratories has shown that Sulf1 activity in ovarian cells decreases Noggin (a BMP antagonist) binding to the cell surface and hence enhances the activity of BMPs (Viviano et al.,2004), one of the key regulators of embryonic development. In contrast, Sulf1 function has been shown to inhibit FGF signalling during chick angiogenesis (Wang et al.,2004) and in ovarian cells (Lai et al.,2003), rapidly dampening the response of the cell to the FGF signal. Since Sulf1 has the potential to modulate the activities of key growth factors and signalling molecules, its highly specific and dynamic expression pattern could reveal its significance in the regulation of growth and development of multiple tissues, including limb development.

Positive or negative association of Sulf1 with specific signalling molecules in localised regions of developing embryos could further indicate its interaction with those signalling molecules. We, therefore, undertook to analyse the Sulf1 expression pattern during early embryonic development to investigate its role in the generation of musculoskeletal structures. Our earlier study showed Sulf1 to be co-expressed with the muscle specification genes Myf5 and MyoD in Shh-responsive epaxial muscle progenitors of newly formed somites (Dhoot et al.,2001). Sulf1 is also expressed in the notochord and floor plate, which produce Shh, and in the Shh-responsive ventral neural tube in the region of motor neuron specification (Liem et al.,1995; Dhoot et al.,2001). The present study investigated the detailed expression pattern of Sulf1 in developing cartilage and joint-forming regions, including the influence of Sulf1 overexpression on mesenchymal aggregation and phalange formation.

The initial patterning of the skeleton is regulated by molecular landmarks that establish spatial coordinates in the early limb bud, controlling the differentiation of the mesoderm. This early step in skeletal patterning is followed by the regulated expression of signalling molecules with precise spatial and temporal patterns of distribution (Reddi, 1995; Wozney and Rosen,1998) regulating the differentiation of limb mesoderm towards myogenic and the chondrogenic lineages. Development of the skeletal elements and joints begins with increased cell adhesion in a group of cells within the mesenchyme. These cells form prechondrogenic condensations that go on to produce matrix molecules such as aggrecan and type II collagen. (Tsonis and Walker 1991). Chondrocyte differentiation and ossification occur in a precise sequence and pattern following a wave of chondrocyte proliferation, differentiation, hypertrophy, and ossification initiating in the centre of the element, moving toward the ends. The perichondrium expresses parathyroid hormone-related peptide, which prevents terminal differentiation of the epiphyseal chondrocytes in the developing skeleton and also contributes to appositional growth of the skeletal elements (Iwamoto et al.,1999). Therefore, the perichondrium is an important signalling layer coordinating skeletal growth and differentiation.

The current study demonstrated the highly dynamic nature of Sulf1 expression in not only the somites and axial structures but also during different phases of cartilage development and joint formation. In vitro overexpression studies showed Sulf1 to enhance condensation of prechondrocytes and early differentiation into chondrocytes while inhibiting joint formation and further phalange formation at high levels of expression.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Sulf1 Expression in Developing Somites and the Axial Structures

The earliest site of Sulf1 expression in the developing embryo using a whole-mount in situ hybridisation procedure was detected in the notochord of stage-8 embryo (Fig. 1A, A1–A3) with its high level expression persisting until stage 18 (Fig. 1D1, D2) following later expression in the floor plate and motor neuron lineages of the neural tube (Fig. 1B1–B3, C1–C3, D1–D3), both of which are induced by Shh signalling (Roelink et al.,1995). Sulf1 expression is also detected first in the ventral somite of anterior to middle somites of stage-8 embryos (Fig. 1A, A2, A3). Further embryonic growth, e.g., at stage 12, detected Sulf1 somitic expression restricted to only myotome with little or no expression in the dermatome or sclerotome at any level in either anterior or posterior somites (Fig. 1B, B1, B2, B3). In stage-12 embryos, Sulf1 is activated in the medial domain of newly formed somites, along the dorsal-ventral axis, which by stage 14 becomes intensely expressed in newly formed somites (Fig. 1C1). Sulf1 expression then localizes to the MyoD-expressing myotomal muscle progenitors in the dorsal medial somite as shown in our earlier study of stage-12 embryos (Dhoot et al.,2001), where expression persists in myotomal cells during subsequent development, e.g., at stages 14 (Fig. 1C3) and 18 (Fig. 1D2, D3). While Sulf1 expression in the anterior 13–14 somites remains restricted to myotome up to stage 12–13, increased expression of Sulf1 in posterior somites (at the mesonephros level) was observed spreading to sclerotome by stage 14 (Fig. 1C and C1). The high level expression in these posterior somites was always accompanied by expression in the adjacent mesonephros (Fig. 1C1, D1, D2) even before this structure had ventralised itself (Fig. 1C1). Sulf1 in somites is thus dynamically expressed in domains that are under the control of Shh signalling, initially in the Pax-1-expressing sclerotome progenitors in the ventral somite in early embryos, prior to MyoD activation. The sclerotomal expression is considerably downregulated by stage 18 (Fig. 1).

thumbnail image

Figure 1. The expression pattern of Sulf1 in developing quail somites and axial structures of stage 8, 12, 14, and 18 quail embryos. AD: Whole-mount in situ hybridisation staining of quail embryos at stages 8, 12, 14 and 18. A1D1: Sections of such stained embryos through posterior regions. A2D2: Sections through the middle regions of such embryos. A3D3: Sections through the anterior somites. nc, notochord; nt, neural tube; dm, dermamyotome; d, dermatome; m, myotome; sc, scleratome; msn, mesonephros; pn, pronephros. Arrows point to mesonephros.

Download figure to PowerPoint

Sulf1 Is Expressed in Two Domains in the Early Limb Bud With the Posterior Proximal Domain Appearing Earlier Than the Anterior Proximal Domain

Sulf1 in the early limb bud is expressed in posterior and anterior proximal domains (Fig. 2A–C) with generally similar patterns of expression in both the wing and leg buds. However, the expression of Sulf1 in the leg bud (not shown) was slightly earlier than that in the wing bud. It was first detected in the posterior proximal region of the leg bud at stage 23 and in a similar location in the wing bud at stage 24 (Fig. 2A). Longitudinal sections across the whole limb bud at this stage showed Sulf1 expression restricted mainly to the ventral half of the limb including the ventral ectoderm (Fig. 2A1, B1) with no detection in the dorsal region. As embryos developed, Sulf1 transcripts also appeared in the anterior proximal region of the leg bud at stage 24 and in the wing bud at stage 25 (Fig. 2B). By stage 26 in the wing and stage 25 in the leg bud, the expression level of Sulf1 in the posterior proximal region had already declined while expression was maintained in the anterior limb bud (Fig. 2B,C). Unlike the posterior proximal expression domain, the expression of Sulf1 in the anterior proximal region extended throughout the dorsal and ventral regions of the mesoderm with no expression in the overlying ectoderm. The sectioning of limb buds further confirmed restriction of Sulf1 expression to mesoderm in the anterior proximal domain (Fig. 2C1) but to both mesoderm and ectodermal lining in the posterior proximal domain.

thumbnail image

Figure 2. Sulf1 expression in the anterior and posterior proximal regions of developing quail wings at different developmental stages. The anterior proximal regions of Sulf1 staining are indicated by green arrows while posterior proximal regions are indicated by broken red arrows. AC (left): Whole-mount in situ hybridisation staining for Sulf1 at developmental stages 24, 25, and 26. A1C1: Vibratome sections of such stained limbs.

Download figure to PowerPoint

The Transient Expression of Sulf1 in Condensing Mesenchyme, Cartilage Template, Perichondrium, and the Developing Joint

Following Sulf1 expression in posterior and anterior proximal domains, Sulf1 transcripts during subsequent limb development were detected in condensing mesenchyme of all limb cartilaginous elements including the developing digital rays (Figs. 3A–E, 4A1, A2, B1). The expression of Sulf1 in all cartilaginous elements, however, was transient, leading to rapid down-regulation in the central core of the developing cartilage (Fig. 4A2.1, B2, 5A3). The down-regulation in the central core of all developing cartilaginous elements followed a proximo-distal progression. While Sulf1 was rapidly down-regulated in the central core of cartilaginous elements, its expression persisted in the peripheral regions, corresponding to the entire perichondrial lengths during earlier stages (Fig. 5A1, A3) followed by reduction or down-regulation in the proximal perichondrium of the developing digital rays during later stages of development (Figs. 3F–K, 4A3–A4, B2–B4, 5B, C1, C2).

thumbnail image

Figure 3. Sulf1 expression pattern in the developing quail limb and autopod of hindlimb during different stages of development (stage 25 to 35) examined by a whole-mount in situ hybridisation procedure. Both the condensing mesenchyme and the perichondrial regions of long bones and phalanges as well as joint formation show dynamic spatial and temporal changes in Sulf1 expression.

Download figure to PowerPoint

thumbnail image

Figure 4. Transverse and longitudinal sections of developing hindlimb/autopod stained for Sulf1 expression using a whole-mount in situ hybridisation procedure. Transverse sections through the developing quail cartilage and skeletal elements at developmental stages 25, 27, 29, and 31 (A1A4) demonstrate Sulf1 expression in the condensing mesenchyme with its down-regulation in the maturing chondrocytes but persistence in the perichondrium. High Sulf1 expression is apparent during early digit formation while Sulf1 expression is down-regulated in the centre of the digital ray following the subsequent phalange formation. Longitudinal sections (B1B4) demonstrate the expression pattern of Sulf1 in developing quail digits and during joint formation at developmental stages 26, 29, 30, and 32, determined by a whole-mount in situ hybridisation procedure. The stage-32 autopod in B4 is identical to the same stage autopod shown in Figure 3I.

Download figure to PowerPoint

thumbnail image

Figure 5. Sulf1 expression in the perichondrium of quail long bones (A1A3) at developmental stage 25 and developing digital rays at developmental stages 30 (B) and 33 (C1, C2). Except for A1 and A2, images are paraffin sections of limbs at stage 25 (A3) or autopods at stages 30 and 33 (B, C1 and C2) of quail embryos stained by a whole-mount in situ hybridisation procedure. The solid arrows indicate the positive Sulf1 staining of the distal phalangeal perichondrium while the broken arrows indicate the negative Sulf1 staining of the proximal phalangeal perichondrium. The partly formed distalmost phalange of digit 3 (B), in contrast, shows the positive proximal perichondrial staining for Sulf1. The arrowheads point to the transverse joint lines showing a moderate level of Sulf1 expression in this region at stage 30 but with little or no expression at stage 33.

Download figure to PowerPoint

Detailed examination of the autopod showed that as the embryo develops, the Sulf1 expression in the cartilage was down-regulated with sustained expression in the perichondrium (Figs. 3F–K, 4A3–A4, B2–B4) and the distal ends of the 1st condensing mesenchymal phalangeal template (Figs. 3I, 4A4.2, B4). As the digits elongated further to generate additional phalanges, Sulf1 expression was observed in only the perichondrial sheath and the tip of the already formed phalange with no expression in the partly generated proximal half of the 2nd phalange (Figs. 3F–H, 4A3, B2, B3). The newly generated digit, however, eventually expressed high levels of Sulf1 expression at its distal end (Fig. 4B2–B4). Sulf1 expression in the developing autopod thus followed the proximo-distal addition of further phalanges until the final number of phalanges for each digit had been achieved. The identity of the final phalange for each digit became clearly apparent from the much higher and broader region of Sulf1 expression in its distal tip marking the cessation of further phalange formation (Fig. 4B4). Even the early phase of the last phalange became apparent from its high level expression in the proximal perichondrium (Fig. 4B3), which did not persist in the other phalanges.

Every time a new phalange was generated, the joint line became apparent from the Sulf1 expression appearing as a transverse line at the tip of the first phalange followed by its appearance at the tips of subsequent phalanges. The broader transverse lines observed during earlier stages (Figs. 3F, 5A1, A2, A3), however, must represent condensing mesenchyme while the narrower transverse lines appearing during later stages (Fig. 3G–J, 4B2–B4, 5B, C1, C2) could represent the joint line. A precise distinction, however, would require more detailed analysis using in situ hybridisation staining of thin paraffin sections of these regions. Down regulation of Sulf1 in the joint line during further development was initiated in the centre of the transverse line while retaining some expression for a little longer in its peripheral edges. Sulf1 expression became undetectable during the later phase of joint development (Fig. 5B, C1, C2).

The expression pattern of Sulf1 in this study clearly demonstrated sequential addition of phalanges. We never observed the formation of a whole cartilaginous element before segmentation. It was always a sequential addition of each cartilaginous element or a phalange starting from the proximal to distal end.

Sulf1 Overexpression in Micromass Cell Culture Promotes Aggregation/Condensation and Early Differentiation of Cartilage Progenitors

To investigate the role of Sulf1 in chondrogenesis, full-length wild-type Sulf1 (WT) and mutated Sulf1 (AA) DNAs were integrated into adenoviral vectors for overexpression studies using empty virus (66) as a control. We overexpressed these different adenoviral integrated constructs in high-density micromass cultures isolated from stage-24 quail embryonic limb mesenchyme. After 7 days of incubation, the cultured cells were stained with anti-collagen type I and type II antibodies to characterize different stages of chondrogenesis. Collagen type I is known to recognize prechondrocytes while collagen type II is typically associated with early cartilage differentiation (Onodera et al.,2005; Tsonis and Walker,1991). Overexpression of wild type (WT) Sulf1 induced an increase in the number of cartilaginous aggregates (Fig. 6). The expression levels of collagen type I and type II were also higher in WT compared with mutated (AA) and the control virus (66). The statistic analysis showed significant difference between WT and AA and WT and 66 (P < 0.001) by Student's t-test with no apparent difference between AA and 66 (P > 0.05).

thumbnail image

Figure 6. Micromass cultures prepared from stage 24 quail limb buds cultured for 7 days in the presence of wild type (WT), mutated Sulf1 (AA) and control empty virus (66), stained with collagen type I (A, A1, B, B1, C, C1) and collagen type II (D, D1, E, E1, F, F1) antibodies using an immunoperoxidase procedure. Sulf1 overexpression in micromass cultures promotes aggregation/condensation and early differentiation into chondrocytes.

Download figure to PowerPoint

Ectopic Sulf1 Overexpression During Digital Explant Culture Increases Phalangeal Length But Inhibits Joint Formation and the Generation of Any Further Phalanges

To investigate the effect of Sulf1 on digital growth and joint formation, we cultured third digits isolated from the autopods of stage-27 quail leg buds in the absence of any additions and in the presence of empty viral control (66) and wild type Sulf1 (WT). After 2 days of culture in the presence of ectopic WT Sulf1, whole mount in situ hybridization detected high levels of Sulf1 expression in the distal ends of the digits (Fig. 7C), reminiscent of the distal-most tip of the last phalange in all digital rays in vivo although the Sulf1 level in vitro appeared considerably higher compared with the levels detected in normally developing digits in vivo. The morphological observations also showed a considerable increase (33 ± 4%) in the length of the original phalange but lacked joint formation resulting in failure to generate any additional phalanges in explants exposed to high levels of wild type Sulf1 (Fig. 7). In contrast, the digital explants cultured in the normal medium (Fig. 7A) or in the presence of empty adenovirus (Fig. 7B) showed normal growth and joint formation including the detection of Sulf1 expression in the prospective joint regions.

thumbnail image

Figure 7. Digits III, isolated from stage-27 quail embryos, were cultured for 48 hr with no additions to the culture medium (A) or in the presence of empty adenovirus 66 (B) or adenovirus containing wild type (WT) Sulf1 construct (C). Unlike the explants cultured in the control medium and in the presence of empty virus (66), digits III cultured in the presence of wild type Sulf1 led to enhanced elongation of the original phalange (labelled as 1) present at the onset of the culture (C) indicated by the broken blue bars. Joint formation in this digit, however, was inhibited with no generation of further phalanges (C) although the length of the original phalange had increased. The red arrowheads point to the normal expression of Sulf1 in joint line in control samples (A, B). The red arrow indicates the much higher level of Sulf1 expression in the distal end of the original phalange (C) in the presence of ectopic wild type Sulf1 that prevents the joint formation and generation of further phalanges. Phalange number 2 is clearly present in the explants cultured in the control medium (A) and in the presence of control virus (B) but absent in the presence of overexpressed Sulf1 (C).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Sulf1 expression in the early embryo was highly dynamic and localized to sites of Sonic hedgehog signalling, which included the somite, notochord, neural tube floor plate, and mesonephros. Sulf1 expression in stage-12 embryos has been reported to be required for MyoD activation (Dhoot et al.,2001). The present investigation was extended to both earlier and later stages of development, demonstrating the requirement of Sulf1 for not only myogenesis but also chondrogenesis. Wnt, BMP, and/or FGF signalling are likely targets for Sulf-1 regulation of both myogenesis and chondrogenesis in developing somites (Dhoot et al.,2001; Ai et al.,2003; Wang et al.,2004; Viviano et al.,2004). Wnt and FGF signals are regulated by sulfate-dependent interactions with HSPGs. HSPGs regulate the transmission of Wnt signals in Drosophila (Lin and Perrimon,2000) and vertebrate embryos (Ai et al.,2003), and Wnts produced by the neural tube and surface ectoderm are required, in combination with Shh, to promote MyoD activation in somites (Munsterberg et al.,1995). Desuflation of HSPGs by Sulf1 facilitates presentation of Wnts to dorsal somite cells to activate MyoD and initiate myogenic progenitor cell specification. Sulf-1 expression in somites could also regulate chondrogenesis as its expression was observed in sclerotome.

The formation of the cartilaginous skeleton is a central event of vertebrate limb morphogenesis, which involves the participation of many signalling molecules with precise spatial and temporal patterns of expression (Reddi,1994; Wozney and Rosen,1998). Sulf1 during early limb development showed two expression domains in anterior proximal and posterior proximal regions of the limb bud. The anterior domain of Sulf1 expression appeared to overlap with the expression of Pax1 in this region (Gruss and Walther,1992; Noll,1993). Fate map studies have shown this region of forelimb and hindlimb mesenchyme to contribute to the formation of pelvic and shoulder girdles (Vargesson et al.,1997). The shoulder and pelvic girdles represent the proximal bones of the appendicular skeleton that connect the anterior and posterior limbs to the body trunk. Pax1 functions in their formation since Hofmann et al. (1998) have shown BMP inhibition of Pax1 expression in the anterior limb bud leading to shoulder girdle defects. Furthermore, mice that have defects in the Pax1 locus also show shoulder girdle malformations (Balling et al.,1992). While Pax1 expression in the somites marks the sclerotomal cells, Pax1 expression is not observed in the appendicular cartilage at any stage of development. This difference in staining patterns may be related to their different cell origins. For example, unlike the appendicular skeleton that is derived from the lateral plate mesoderm, the pelvic girdle and the scapula are believed to have a dermamyotomal origin (Huang et al.,2000) and may explain Pax1 expression in this region as it is regarded a sclerotomal marker. Unlike Pax1, Sulf1 expression, however, is observed in not only some sclerotomal regions but also the appendicular cartilaginous templates. Sulf1 may regulate BMP activities in the shoulder and pelvic girdle–forming regions since many BMPs such as BMP2, 4, and 7 are expressed in the developing limb with BMP4 expression (Francis et al.,1994) being particularly close to the anterior proximal domain of Sulf1 expression. The expression of Sulf1 in the anterior proximal region of the limb bud thus indicates its involvement in the formation of the shoulder/pelvic girdle through its possible interaction with BMPs and Pax1. Although, a 2nd patch of Pax1 has also been reported in the posterior proximal region during later stages (27/28) of development, this expression is much later than the expression of Sulf1. Unlike Pax1, Sulf1 shows its posterior proximal expression domain earlier than the anterior proximal expression domain. The precise location of this Pax1-positive region is also different from Sulf1 expression in the posterior proximal region and its correlation with Pax1 in this region, therefore, is not clear. Another homeobox transcription factor called Emx2, however, is expressed in both anterior proximal and posterior proximal domains overlapping with Sulf1 expression. Similar to Pax1, Emx2 during the development of wing has been shown to be a marker for the prospective scapular blade region (Prols et al.,2004) as would also appear to be the case for Sulf1 in scapular formation although it remains to be determined how it interacts with other molecular intermediates in cartilage formation. Emx2, nevertheless, has been shown to have binding sites for Tcf and Smad proteins, the transcriptional mediators of the Wnt and BMP signalling pathways. Since Sulf1 can regulate Wnt and BMP signalling (Dhoot et al.,2001; Viviano et al.,2004), it may play a part in regulating the expression of Emx2 in the limb.

The limb skeleton is derived from mesenchymal cells that condense and form cartilage templates prefiguring the arrangement of skeletal elements. The limb cartilage elements form in a temporal proximo-distal sequence showing spatial and temporal expression of Sulf1 fully compatible with proximo-distal limb development. The expression of Sulf1 in early cartilage templates as well as its expression in the distal tips of the cartilaginous elements or phalanges indicates its involvement in mesenchymal condensation of presumptive chondrocytes that could be accomplished through increased adhesion of the cells. Sulf1 expression persists and is also observed during early differentiation of the chondrocytes leading to rapid down-regulation following further maturation as is observed in the central core of the cartilaginous elements. Cells deeper in the core of such condensations initiate chondrocyte differentiation that go on to deposit extensive ECM while a layer of fibroblast-like cells generates a perichondrial sheath on its periphery. The pattern of Sulf1 expression indicates that it is not involved in chondrocyte maturation but only the early differentiation of chondrocytes. BMPs are also required for prechondrogenic condensation and differentiation into chondrocytes (Pizette and Niswander,2000). Sulf1 could co-operate and enhance BMP activities by inhibiting BMP antagonists (Viviano et al.,2004). To investigate its role in mesenchymal condensation, Sulf1 was overexpressed in limb bud micromass cultures. Overexpression of wild type Sulf1 resulted in increased aggregation of mesenchymal cells as indicated by an increase in the number of both Collagen type I- and Collagen type II-positive aggregates supporting our in vivo observations of early developing limb that Sulf1 promotes condensation and early differentiation of chondrocytes since collagen type II is an early differentiation marker while collagen type I has been reported in the perichondrium and articular cartilage, the areas that also express Sulf1 (Swalla et al., 1998; Tsonis and Walker,1991; Onodera et al.,2005).

The perichondrial staining of Sulf1 was continuous along its entire length during early development but was disrupted during later phase of development. The precise basis of this selective down-regulation in parts of the perichondrial sheath during later stages is not clear since in the digital rays, it was usually down-regulated in the proximal halves compared with the distal halves that extended far beyond the articular perichondrial length. A high level of Sulf1 expression in the proximal perichondrium was observed in only the first and the last phalange of each digit that also expressed high levels in its distal mesenchymal tip. Sulf1-expressing regions may represent high levels of proliferation in the distal halves of the cartilage/skeletal elements although it remains to be determined whether the perichondrial cells lining the distal cartilaginous elements are indeed more proliferative than the perichondrial cells lining the diaphyses or proximal articular cartilage. Compared with the proximal end, the leading edge of the growing cartilage, particularly the last phalange expressed higher levels of Sulf1 that could mark mesenchymal condensation. While Sulf1 during early stages could also be associated with increased proliferation, the very high Sulf1 expression in the digital tip of the last phalange inhibits joint formation and thus prevents the generation of any additional phalanges.

Sulf1 expression in the limb overlaps with the expression of several Wnts in cartilage and joint development where Sulf1 could regulate their activities by the release of heparin sulphate bound Wnt (Dhoot et al.,2001, Ai et al., 2003). For example, Wnt4, Wnt5a, Wnt5b, Wnt11, and Wnt14 have been shown to mediate distinct effects on the initiation of chondrogenesis and differentiation of chondrocytes with restricted patterns of expression in developing cartilage and joint formation (Hartmann and Tabin,2000, 2001; Church et al.,2002). Joint formation, however, is a complex multi-step process, initiated with the inhibition of chondrogenic differentiation in the prospective joint line leading to changes in the matrix followed by cell death and the formation of the articular cartilage and joint capsule around the joint cavity to generate the mature joint. Sulf1 shows a very transient expression in the early joint line followed by its rapid down-regulation during later stages of joint development indicating its inhibitory role during later stages of joint formation. Wnt 4 and Wnt14 expression has been reported in developing limb joints (Hartmann and Tabin,2000, 2001; Church et al.,2002) where Sulf1 could regulate their activities. Wnt14 has also been shown to up-regulate the BMP anatagonist, chordin, in the early joint interzone. The very transient appearance of Sulf1 in the joint line may enhance Wnt signalling or moderate the action of chordin, as has been demonstrated for another BMP antagonist, Noggin (Viviano et al.,2004).

During early development, Sulf1 was expressed in the perichondrium with only transient expression in the joint line. Sulf1 expression was specifically excluded from the later stages of joint formation in the developing digits. The role of Sulf1 in joint development was, therefore, further examined by ectopic overexpression of Sulf1 on the development of digital explants in vitro. Stage-27 digital explants cultured over a 48-hr time period grew well in vitro showing normal development in the presence of mutated and empty control virus. In contrast to this, wild type Sulf1 showed a marked effect on both phalangeal growth and joint formation. These explants showed increased length of the original 1st phalange but complete inhibition of further phalangeal segments, which may be due to high Sulf1 accumulation in the distal tip. The relatively high Sulf1 expression in the distal tip in vivo was only observed in the distal-most phalange marking the end of further phalange formation. A high level of Sulf1 expression in vitro thus inhibited joint formation completely. It is possible that Sulf1 stimulates joint formation at low levels during an early phase of its development as indicated by its low level transient expression in vivo. The ectopic overexpression of Sulf1 in vitro led to total inhibition of joint formation. Sulf1 could play a key role in regulating the length and number of phalanges and the spacing of the joints by altering its level of expression in the developing autopod. Sulf1 thus regulates not only the mesenchymal condensation and differentiation of prechondrocytes into chondrocytic lineage but also growth and joint formation.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Whole-Mount In Situ Hybridisation Procedure

Fertilized quail eggs obtained from Rosedean Farm, Cambridge, were incubated to the desired embryonic stages at 38°C. Embryos staged according to Hamburger and Hamilton (1951) were washed in PBS and fixed in 4% formaldehyde overnight at 4°C. As was the case in our earlier study (Dhoot et al.,2001), a riboprobe was prepared for an in situ hybridisation procedure using a DNA fragment, designated S1–17 corresponding to the 5′-end of the gene that includes the key catalytic domain (0.2–2.2kb). This DNA fragment was linearised with Xho1 for amplification with T3 RNA polymerase. A sense probe was prepared by linearization with EcoR1 and RNA amplification using T7 RNA polymerase. An an in situ hybridization procedure was performed generally as described by Nieto et al. (1996). Briefly, fixed embryos were washed in PBST (PBS with Tween 20) and treated with proteinase K (20 μg/ml) for different lengths of time. After 3 rinses with PBST, embryos were refixed in 4% formaldehyde and 0.25% glutaraldehyde for 10–30 minutes at 4°C. After fixation, embryos were washed in PBST and prehybridized in hybridization buffer at 68°C for 2–24 hr followed by 1–3-day hybridisation with the riboprobe at 68°C. Embryos were washed several times in 2× SSC and 0.2 × SSC buffers to remove unbound probe before room temperature incubation with 15% FCS (fetal calf serum) for 2–6 hr. Embryos were then incubated in alkaline phosphatase-conjugated Fab fragments against digoxigenin (Roche) diluted 1:4,000 in 15% FCS overnight. Embryos were stained at room temperature in alkaline phosphatase substrate buffer (0.1 M Tris, pH 9.5, 50 mM MgCl2, 0.1 M NaCl, 1% Tween20) with 4.5 μl NBT and 3.5 μl BCIP/1.5 ml buffer. Unlike antisense riboprobe, sense probe showed no staining. When required, some stained embryos were embedded in 4% agar and 50–100-μm-thick sections were cut using a vibratome. Some tissues were also embedded for paraffin sectioning for visualisation at higher magnifications.

Preparation of Adenovirus

The coding sequences of full-length wild type and mutated Sulf1 cDNAs (Dhoot et al.,2001) were subcloned into a commercially available shuttle vector, pDC515 (Microbix Biosystems Canada) downstream from the mouse cytomegalovirus promoter and replication-deficient adeno viruses generated by site-specific FLP-mediated recombination of the cotransfected shuttle and genomic plasmids in 293 cells as described previously (Sala-Newby et al.,2005).

Micromass Cell Culture and Immunohistochemistry

The limb buds were excised from the body wall of stage-24 embryos collected in PBS. To minimise the inclusion of myogenic cells in these cultures, only distal 1/3 of the limb buds were dissected out from which the ectodermal layer was removed by trypsinisation and mesenchymal tissues dissociated into single cell suspension at 1.5 × 107 cells/ml. A 10-μl drop was plated onto each well (24-well tissue culture dish) and cultured for 1 hr at 37°C before flooding with 1 ml of culture medium composed of 45% Dulbecco's Modified Eagle Medium (DMEM), 45% F12 Medium, 10% FCS and 100 μg/ml ascorbic acid. To investigate the function of Sulf1 in vitro, 1 μl of adenovirus (2×1010/ml) carrying full-length Sulf1 cDNA (WT), mutated Sulf1 cDNA (AA), and empty adenovirus vectors (66) at similar concentrations were added separately to three groups of cell cultures and incubated at 37°C for 7 days. The cell cultures were washed with PBS before fixation with 100% methanol for 10 min at room temperature. Anti-chick Collagen type I and type II antibodies were used to detect cell aggregation and differentiation into chondrocytes by a peroxidase immunostaining procedure described previously (Zhao and Dhoot,2000). The number of Collagen type I- and type II-positive cell clusters were counted using an inverted light microscope. The results are expressed as means ± SD. Data from three different groups were compared by One-Way analysis of variance (ANOVA), followed by a Student Newman-Keul test. The comparison between control and WT and control and AA was performed by Student's t-test. All statistical comparisons were made at the 0.05 significance level and were performed with the programme SPSS for Windows (SPSS Inc., Chicago).

Explant Culture

The autopods from stage 27 quail hindlimbs were dissected into ice cold PBS and incubated in 0.5% trypsin for 30 min on ice before the surface ectoderm and interdigital mesoderm could be dissected out using tungsten needles. Only the third digit was used for all explant culture analyses. The length of the digit from the root to the tip was measured after rinsing with PBS. The digits were then transferred onto the filter paper for 48 hr in 24-well culture plates with 1 ml medium to which either no additions or 1 μl of adenovirus (2×1010pfu/ml) carrying full-length Sulf1 DNA (WT) or empty adenovirus (66) at the same concentration was added. The medium was composed of 45% DMEM (Dulbecco's Modified Eagle Medium), 45% F12 Medium, and10% fatal calf serum. After 48 hr incubation at 37°C, the explants were washed with PBS and the length of individual phalanges measured. The measurements were taken from 12 explants each (4 samples in three independent experiments) for each of the three groups. The PBS washed explants were fixed overnight at 4°C in 4% formaldehyde for in situ hybridization analysis as described above.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank Dr. Steve Allen for his critical reading of the manuscript.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES