QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 3 December 2003
doi: 10.1242/dev.00901

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in Development Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Nakayama, N. Articles by Zhang, K. Search for Related Content PubMed PubMed Citation Articles by Nakayama, N. Articles by Zhang, K.

A novel chordin-like BMP inhibitor, CHL2, expressed preferentially in chondrocytes of developing cartilage and osteoarthritic joint cartilage

Naoki Nakayama^1,*,, Chun-ya E. Han¹, Linh Cam², Jae I. Lee¹, Jim Pretorius³, Seth Fisher⁴, Robert Rosenfeld⁴, Sheila Scully³, Ryuichi Nishinakamura⁵, Diane Duryea³, Gwyneth Van³, Brad Bolon³, Takashi Yokota⁵ and Ke Zhang²

¹ Department of Metabolic Disorders, Amgen, One Amgen Center Drive, Thousand Oaks, CA 91320, USA
² Department of Cancer Biology, Amgen, One Amgen Center Drive, Thousand Oaks, CA 91320, USA
³ Department of Pathology, Amgen, One Amgen Center Drive, Thousand Oaks, CA 91320, USA
⁴ Department of Protein Science, Amgen, One Amgen Center Drive, Thousand Oaks, CA 91320, USA
⁵ Department of Stem Cell Regulation, Institute of Medical Science, The University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8639, Japan

View larger version (50K): [in a new window]   Fig. 1. Primary structure of CHL2. (A) Schematic representation of chordin, CHL1(l), and CHL2. SP stands for signal peptide. The CR1 and CR3 regions in CHL1 and CHL2 (black boxes) are most homologous to CR3 of chordin (also in black). The chordin CR1 (in gray) and CR3 possess the BMP-binding capability (Larrain et al., 2000). Putative BMP1/Tolloid cleavage sites are indicated with an asterisk, while actual Tolloid cleavage sites (Scott et al., 1999) are shown by vertical arrows. The CHL1 ORF had two sites with amino acid sequence variations (dE and d5) (Nakayama et al., 2001). (B) Amino acid sequences of mouse, rat and human CHL2 protein precursors. The three CRs are indicated by boxes. The vertical arrow indicates the NH2-terminal amino acid of mature mCHL2-FLAG (Leu26), as determined by amino acid sequencing of purified recombinant protein. (C) Amino acid sequence alignment showing sequence similarities between CR1 or CR3 of mouse CHL1 and CHL2, and CR3 of mouse chordin. Ten conserved cysteines (highlighted in black) are found in the spacing typical of vertebrate chordins. Other conserved amino acids are highlighted in gray.

View larger version (46K): [in a new window]   Fig. 2. Direct interaction of mCHL2 with BMPs, and inhibition of BMP4 binding to BMP receptor ectodomain by mCHL2. (A) FLAG-tagged CHL2 protein. Proteins in the peak eluate from the hydroxyapatite column chromatography were separated by SDS-polyacrylamide gel electrophoresis under reducing conditions and then silver stained (Sambrook et al., 1989). The mCHL2-FLAG band was excised and the NH2-terminal amino acid sequence (vertical arrow in Fig. 1B) determined. (B) Immunoprecipitation/western blot analysis of mCHL2-FLAG individually mixed with BMP2 (a), BMP4 (b), BMP5 (c), BMP6 (d), BMP7 (e), GDF5 (f), activin A (g), TGFß1 (h), TGFß2 (i) or TGFß3 (j), followed by treatment with CHL2-COOH (lanes underlined). Immunocomplexes were detected using the corresponding antibodies (upper panels). Reactions only with mCHL2-FLAG, BMP, GDF, activin or TGFß were also performed as negative controls. Each blot was further developed with M2 to confirm the presence of precipitated mCHL2-FLAG (lower panels). The TGFß immunocomplexes (h-j) were separated into two sets; one was loaded on a non-reducing gel to visualize TGFß (upper panels), and the other on a reducing gel to detect CHL2 (lower panels). Lanes not underlined were directly loaded with the indicated amount (ng) of mCHL2-FLAG, BMP, GDF, activin or TGFß (for standards). (C) Inhibition of BMP4 binding to BMPR1B ectodomain by mCHL2-FLAG. The indicated amount of mCHL2-FLAG was first mixed with or without BMP4, and then BMPR1B-Fc or IgG-Fc was added. Protein complexes containing BMPR1B-Fc or IgG-Fc were selectively precipitated with protein A and subjected to western blot analysis (lanes underlined). Upper panel: bound BMP4 visualized with anti-BMP4 antibody. Middle panel: co-precipitation of mCHL2-FLAG checked with M2. Lower panel: precipitation of BMPR1B-Fc/IgG-Fc confirmed with anti-IgG-Fc antibody. For the standards, 0.04 µg of mCHL2-FLAG and 0.04 µg of BMP4 were loaded directly.

View larger version (35K): [in a new window]   Fig. 3. Inhibitory effects of mCHL2 on BMP-dependent differentiation of C2C12 cells in vitro. Dose-dependent inhibition by CHL2, chordin or noggin of BMP-dependent alkaline phosphatase (AP) induction in C2C12 cells. (A) Cells were cultured in differentiation medium for 2 days with various concentrations of BMP4 (plus sign), BMP6 (circle), or BMP7 (triangle). (B-D) Cells were also differentiated with a constant concentration of BMP4 (10 nM; B), BMP6 (15 nM; C) or BMP7 (19 nM; D) and various concentrations of mCHL2-FLAG (plus sign), noggin-Fc (triangle), or mCHD-His (circle) for 2 days. Mean specific activity of AP (triplicate assays) is shown; vertical bars denote s.d. Data are representative of three independent experiments.

View larger version (93K): [in a new window]   Fig. 4. mCHL2 mRNA expression in cartilage. In situ hybridization (ISH) for CHL2 in normal mouse embryos, depicted in paired bright-field (left) and dark-field (right) panels. (A) Expression of CHL2 mRNA at the costochondral junction, at E17.5. Note CHL2 in chondrocytes on both sides of the junction. (B) Transverse section through sternum, at E18.5. Signal is present in areas where ribs converge. (C,D) ISH for CHL2 in the adult knee. Weak CHL2 expression is present over the articular cartilage surface of the femoral head and patella, but not in growth plate chondrocytes. Boxed areas in C shown at a higher magnification in D in which the signal can be seen in chondrocytes on both sides of the joint. (E,F) Expression of CHL2 in adult vertebral articulation. Signal occurs in superficial articular chondrocytes on both sides of the zygapophyseal or facet joint. The boxed area in E is shown at a higher magnification in F, revealing signal localization over the superficial zone chondrocytes. dv, dorsal (superior) vertebra; vv, ventral (inferior) vertebra. Arrowheads indicate CHL2-positive chondrocytes.

View larger version (85K): [in a new window]   Fig. 5. Expression of CHL2 mRNA in soft tissues. (A) Northern blot analyses with human multiple tissue blots IV (lanes 1-8) and I (lanes 9-16) are shown. Blots were first probed with a DNA containing the hCHL2 ORF (CHL2), and then re-probed with human ß-actin (Clontech). Essentially identical results were achieved with three different blot batches. Lane 1, spleen; lane 2, thymus; lane 3, prostate; lane 4, testis; lane 5, uterus (without endometrium); lane 6, small intestine; lane 7, colon; lane 8, peripheral blood leukocytes; lane 9, heart; lane 10, whole brain; lane 11, placenta; lane 12, lung; lane 13, liver; lane 14, skeletal muscle; lane 15, kidney; lane 16, pancreas. (B) In situ hybridization for CHL2 in adult mouse tissues, shown in paired bright-field (left) and dark-field (right) panels. (a) Uterus (transverse section) showing CHL2 expression primarily in the myometrium. (b) Colon, showing weak CHL2 expression in serosal cells – a different pattern to that of CHL1 (Nakayama et al., 2001). muc, submucosa; mus, muscularis; end, endometrium; myo, myometrium.

View larger version (67K): [in a new window]   Fig. 6. Expression of CHL2 mRNA in diseased cartilage. (A) In situ hybridization for CHL2 in human adult knee joints, depicted in paired bright-field (left) and dark-field (middle) panels, and accompanied by a higher magnification view (bright field panel, right). Expression of CHL2 mRNA in normal knee cartilage (femur, 55-year old female). Note that CHL2 is localized on articular chondrocytes (right). However, CHL2+ chondrocytes are scattered throughout the cartilage. Weak signals are also visible in superficial zone chondrocytes. (B,C) Expression of CHL2 in osteoarthritic (OA) knee cartilage (B: 73-year-old male with OA, degenerative joint disease (DJD), and chronic proliferative synovitis; C: 48-year-old female with DJD). Signal is strongly induced in middle zone chondrocytes, but absent in superficial zone chondrocytes. (D) Expression of CHL2 in rheumatoid arthritic knee cartilage (80-year-old female with RA). Weakly positive chondrocytes are scattered throughout the cartilage. (E,F) Expression of CHL2 mRNA in hind paws of rats with collagen-induced arthritis, depicted in paired bright-field (left) and dark-field (middle) panels. CHL2 signal is weakly expressed in normal chondrocytes (E), and is not substantially elevated in animals with severe immune-mediated disease (F). Arrowheads indicate CHL2-positive chondrocytes. sz, superficial zone; mz, middle zone; dz, deep zone.

View larger version (74K): [in a new window]   Fig. 7. Effect of CHL2 on in vitro chondrogenesis. (A) CHL mRNA expression during chondrogenic culture. hMSCs were grown in pellet culture in the presence of TGFß3. On the indicated day, RNA was extracted from particles, treated with DNase I, and subjected to RT-PCR using primers for aggrecan, COMP, COL2, CHL1, CHL2 and GAPD. The CHL2 signal was confirmed with a 33P-labeled XbaI-SalI fragment from pSPORThCHL2. (B) hMSCs were grown in pellet culture in the presence of TGFß3 alone (a,d), or TGFß3 with either 0.2 µg/ml mCHL2-FLAG (b), 2 µg/ml mCHL2-FLAG (c), 0.1 µg/ml noggin-Fc (e), or 1 µg/ml noggin-Fc (f). On day 21, particles were harvested and stained with Toluidine Blue. Data are representative of five independent experiments. Addition of IgG-Fc did not affect growth and maturation of cartilage-containing particles (not shown). Note that cartilage nodules, in which well-separated cells were embedded, stained more intensely with Toluidine Blue. (C) Effect of CHL2 on in vitro mineralization of EB cell-derived cartilage. FACS-purified FLK1–PDGFR+ EB cells were grown in pellet micromass culture to produce hyaline cartilage particles. On day 18, the medium was changed to a hypertrophic differentiation medium in the absence (a-c) or presence of 3 µg/ml mCHL2-FLAG (d-f) or 2 µg/ml noggin-Fc (g,h). On day 24, particles were harvested, stained with von Kossa (a,d,g), and immunostained with X53 for COL10 (b,e,h) or 2B1.5 for COL2 (c,f). Data are representative of four independent experiments. The Scale bar for B and C is shown in C.