| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Endocrinology Division (V.K., Y.M., S.F., C.A.F.), Eli Lilly & Co., Lilly Corporate Center, Indianapolis, Indiana 46285; St. Vincent’s Institute of Medical Research (J.M.M.), Victoria 3065, Australia
Address all correspondence and requests for reprints to: Venkatesh Krishnan, Endocrinology Division, Eli Lilly & Co., Lilly Corporate Center, Indianapolis, Indiana 46285. E-mail: Krishnan_Gary{at}lilly.com
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Indian hedgehog (Ihh) is a member of the vertebrate family of Hedgehog genes that include Sonic hedgehog (Shh) and Desert hedgehog (Dhh) (2). Indian hedgehog is expressed in a discrete portion of the developing growth cartilage at the border of maturing and hypertrophic chondrocytes (3). It is believed that Ihh’s effect on chondrocyte growth and differentiation is primarily mediated by PTH-related protein (PTHrP) and its cognate receptor (4). A recent report by Pathi and co-workers has revealed yet another BMP-2/4 related pathway that may be downstream of Ihh (5). Prior use of bone explants to study the role of Ihh and PTHrP have revealed an inhibitory effect of PTHrP and Ihh on cartilage differentiation as measured by collagen X expression (6). However, recent studies in mice null for the Ihh gene reveal that these factors may also have a growth and differentiation promoting role during endochondral ossification (7). Ihh and Shh are highly homologous and bind the receptor ptc with similar affinity and are capable of being used interchangeably in experimental systems (3). In this report, we use a neonatal rat metatarsal culture model (8, 9) to study the effect of the N-terminal 19-kDa Shh protein fragment (Shh-N) on longitudinal growth. We observe an anabolic effect for both Shh-N and BMP-4 and these two pathways were found to work in concert in the model. Noggin is a secreted protein that antagonizes the BMP-2/4 signaling by tethering the ligand and preventing it from binding to its cognate receptor (10). We provide evidence that noggin is up-regulated by treatment with 200 ng/ml BMP-4, and is capable of blocking the anabolic effect of both Shh-N and BMP-4. Furthermore, we show an increase in osteocalcin expression after treatment of metatarsals with 1 nM Shh-N, 25 ng/ml BMP-4 and BMP-4 + Shh-N. Finally, with the use of antisense Cbfa1/Osf-2 oligonucleotides, we show that both Shh-N and BMP-4 cannot mediate their effect in the absence of Cbfa1 expression. Collectively, this report points to a BMP-2/4 and Cbfa1-dependent pathway of endochondral ossification that is induced by Shh-N, and presumably Ihh, in cultured neonatal rat metatarsals.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human Ihh protein
Human Ihh was cloned from human liver messenger RNA (mRNA) using
RT-PCR with a specific set of primers and its sequence was subsequently
verified (2). A construct containing the nucleotides to
generate an N-terminal fragment (amino acid residues 28–202) was
synthesized using a pBluescript KSII vector (Stratagene,
La Jolla, CA). This construct was used to generate in vitro
translated radiolabeled N-terminal human Ihh that was
immunoprecipitated (IP’d) using the mShh-N antibody from R&D Systems.
In vitro translation was performed using the Promega Corp. (Madison, WI) T3/T7 reticulocyte lysate kit.
Immunoprecipitation was performed using protein A Sepharose beads
conjugated with 100 µl of 1:10 diluted Shh-N antibody ± Shh-N 2
µg/ml, or control IgG, for 2 h. Fifty microliters of the
reticulocyte lysate was incubated at 30 C along with the conjugate for
2 h. Five or 10 µl of the supernatant was boiled in the presence
of SDS loading buffer, run onto a 10% SDS-PAGE and visualized by
autoradiography. The Shh-N (2 µg/µl) precleared antibody was used
to ascertain specificity of the protein product.
Rat neonate metatarsal culture
Metatarsals were cultured as previously described
(1). Newborn Sprague Dawley rats (Harlan Sprague Dawley, Inc., Indianapolis, IN) were killed at day 0, and the
metatarsals were surgically isolated and placed in BGJ media
(Life Technologies, Inc., Rockville, MD), without serum
containing 100 µg/ml antibiotic/antimycotic solution (Life Technologies, Inc.). Metatarsals were cultured for 7 days in
this media in a 96-well round bottom Petri dish (Nunclon, Nunc
Products, Denmark) under 5% CO2 at 37 C. Wells
were replaced with fresh media containing appropriate treatments every
24 h for 7 days. The metatarsals were imaged under a light
microscope on day 0 and day 7 and changes in mineralization were
quantified using Image Pro analysis software.
Calcein staining and histopathological analysis
At the end of the culture, metatarsals were fixed in 10%
buffered formalin for 24 h at 4 C. Half of the metatarsals were
routinely processed (11) and embedded in plastic. The
other half of the metatarsals were subsequently decalcified in
Decalcifier II (Surgipath Medical Industries Inc., Richmond, IL) for
4 h at 4 C and then processed for paraffin embedding. Longitudinal
5-µm sections were stained with either toluidine blue or hematoxylin
and eosin. One set of metatarsals were stained for 2 h, before
embedding in plastic, with Calcein (500 ng/ml) and were subsequently
sectioned and imaged under a fluorescent microscope.
cAMP assay
This assay was performed as described earlier (12).
PTHrP ± 3F5 was incubated overnight at 4 C and 100 µl of this
mix was added to 900 µl of assay reaction buffer.
mRNA analysis
Metatarsals were pulverized after culture in a guanidinium
thiocyanate solution as previously described (13). The
lysate was placed onto a sucrose density gradient and total RNA was
pelleted after ultracentrifugation at 25 C for 20 h at 32,000 rpm
in a SW41Ti (Beckman Coulter, Inc.) rotor. RNA pellets
were precipitated using isopropanol and subsequently washed with 70%
ethanol 30% water. Total RNA was dissolved in water and 250 ng was
used in a RT reaction using the Superscript II kit (Life Technologies, Inc.) and random hexamer primers. Subsequently, an
aliquot of the RT reaction was subjected to 30 cycles of PCR using
Platinum Taq polymerase and specific primers for noggin,
osteocalcin (rat and mouse) and GAPDH. PCR were annealed at 60 C, 54 C
and 54 C, respectively, in a 25 µl reaction. Ten microliters were
loaded onto a 1.5% TAE gel and the products were analyzed after
ethidium bromide staining.
Primer information
All custom primers were obtained from Life Technologies, Inc.
Rat noggin primers. 241 GGAGGCATGGAGCGCTGCCC (forward) 990 GGATCCATCAAGTGTCTGG (reverse)
Human Ihh primers. 1 CCGGCGCCTCATGACCCAGCGCTGCAAGGA (forward) 1211 GGAGTCGCCGTGCCAGCCTCAAGGTCTC (reverse)
Rat GAPDH primers. 301 TCGTGGAGTCTACTGGCGTCTT (forward) 1075 CCTCTCTCTTGCTCTCAGTATC (reverse)
Rat osteocalcin primers. 474 AGTCACCAACCACAGCATCC (forward) 801 TTTGTCCCTTCCCTTCTGCC (reverse)
Quantitative PCR analysis
Poly A mRNA was isolated from rat neonate metatarsals treated
with 25 ng/ml BMP-4, 1 nM Shh-N, and Shh-N+ BMP-4 for
72 h by using RNA isolation kit from Promega Corp.
Osteocalcin and GAPDH was measured by real time PCR that allows a
quantitative evaluation of the steady-state mRNA levels.
Mouse and rat GAPDH. 200 GGCAAATTCAACGGCACAGT (sense) 269 AGATGGTGATGGGCTTCCC (antisense) 221 AAGGCCGAGAATGGGAAGCTTGTCATC (taqman probe)
Rat osteocalcin. 1131 ATGAGGACCCTCTCTCTG (sense) 1193 TGCCAGGTCAGAGAGGC (antisense) 1151 CACTCTGCTGGCCCTGACTGCA (taqman probe)
The taqman probe consists of an oligonucleotide with a 5'-reporter dye and a 3-quencher dye. The fluorescent reporter dye FAM (6carboxyfluorescein) was covalently linked to the 5' end of the oligonucleotide, and the reporter was quenched by TAMRA (6-carboxytetramethylrhodamine) that is located in the 3' end. The quencher suppressed fluorescence until the PCR when the probe was cleaved which released the fluorescent radicals. The resulting increase in fluorescence was measured and reflects the amount of specific mRNA present in each tube. The results are shown as fold or percent changes over vehicle controls.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Furthermore, at higher
magnification (Fig. 1B) plump cuboidal osteoblasts (OB) and
multinucleated osteoclasts (OC) can be seen which suggests, that after
7 days of culture the organ is viable and is capable of responding to
bone modeling or anabolic agents. To identify the mineralized region
within the metatarsal that is being quantified we visualized the
mineralization after calcein staining of the metatarsal. Figure 2 (brightfield) shows the presence of a
calcified matrix and the measured value (shown in brackets) can be
overlaid onto the calcein stained (Fig. 2; darkfield) area of the
metatarsal section. Anabolic activity of a particular treatment is the
increase in endochondral ossification which was defined by longitudinal
extensions within this calcified region comparing the day 0 and day 7
images of the same metatarsal.
|
|
). There are two bands which correspond
to full-length (
19 kDa) and a lower molecular weight band as a
result of aberrant 3' translation using the in vitro system.
The antibody was raised against a N termini peptide and can detect full
length and 3' truncated proteins. In contrast, neither the Shh-N
precleared antibody nor the IgG could immunoprecipitate these bands
(Fig. 3). Twenty-five microliters of this reticulocyte lysate was found
to increase secreted alkaline phosphatase activity by 3-fold in
C3H10T1/2 cells after 3 days of treatment (data not shown). Earlier
reports have used both Ihh and Shh-N interchangeably (3)
and since recombinant mShh-N protein gave us consistent results and
could block the detection of the in vitro translated Ihh
protein, subsequent experiments were performed using the recombinant
Shh-N protein.
|
). It
has been reported that PTHrP is a target gene for Ihh signaling and
that PTHrP acts as a feedback inhibitor of Ihh signaling. To study the
role of PTHrP in mediating this effect we used the 3F5 PTHrP-blocking
antibody (5 µg/ml) together with PTHrP and Shh-N. Cotreatment with 10
µg/ml of 3F5 decreases the PTHrP induced dose-dependent increase in
cAMP levels, observed in UMR 106.01 cells (Table 1). The actual dose for the antibody
seems to be slightly different for the two assays. As shown in Fig. 4B, 1
nM rPTHrP (1–34) cotreatment with 5 µg/ml 3F5 antibody
results in complete reversal of PTHrP-mediated anabolic activity.
However, cotreatment of metatarsals with 1 or 0.1 nM Shh-N
and the 3F5 antibody does not result in a significant change in
Shh-N-mediated anabolic activity. In contrast, IgG cotreated with 1
nM rPTHrP and 1 nM or 0.1 nM Shh-N
does not block the anabolic activity of both PTHrP and Shh-N. This
result suggests that Shh-N- induced anabolic activity in these
metatarsals is not mediated by PTHrP.
|
|
). To investigate the reason why this
anabolic effect was lost at higher doses, mRNA from these metatarsals
were subjected to a RT-PCR using two sets of primers for GAPDH and
noggin. It has been reported that noggin expression is under the
control of BMP-2/4 signaling and noggin can inhibit the BMP-2/4
pathway, which provides a negative feedback loop for BMP-2/4 signaling
(10). Accordingly, we found an increase in noggin mRNA as
the concentration of BMP-4 was increased from the anabolic 25 ng/ml to
the restrictive 200 ng/ml amount (Fig. 6;
compare lanes 2 through 8). Hence, we hypothesize that the lack of
anabolic activity at higher doses of BMP-4 may result from a negative
feedback control mechanism which is mediated by noggin.
|
|
). Furthermore, an increased expression of noggin was observed in
these metatarsals (Fig. 6, compare lanes 2 and 12). Cotreatment of 25
ng/ml hBMP-4 and 1 nM Shh-N did not result in an additional
increase in endochondral ossification than either treatment alone (Fig. 5). Cotreatment of metatarsals with 1 nM Shh-N and 25 ng/ml
BMP-4 or 200 ng/ml of BMP-4, completely reversed the anabolic activity
of Shh-N at the higher dose (200 ng/ml) of BMP-4, but it did not
further increase the anabolic activity at the lower dose (25 ng/ml) of
BMP-4 (Fig. 5). Finally, addition of exogenous amounts of recombinant
noggin (20 µg/ml) to 1 nM Shh-N or 25 ng/ml BMP-4,
reversed the anabolic activity of both agents (Fig. 7). In contrast the anabolic activity of
1 nM PTHrP is not significantly altered in the presence of
20 µg/ml of recombinant noggin (Fig. 7).
|
To determine the role of Cbfa1 in mediating the anabolic activity
of BMP-4, we used a previously characterized antisense oligonucleotide
(17) that attenuates Cbfa1 expression, along with a
control scrambled sense oligonucleotide that does not affect Cbfa1
levels. These antisense oligonucleotides have been directed against the
nucleotides immediately 5' of the runt domain and have been shown to
decrease alkaline phosphatase activity in primary rat osteoblasts
(17). These oligonucleotides were added to the culture
1 h before treatment of metatarsals with BMP-4 (25 ng/ml). To
confirm the effect of this antisense oligonucleotide approach, we
measured the mRNA levels of osteocalcin, an osteoblast specific marker,
in both scrambled and antisense oligonucleotide-treated metatarsals. We
saw a significant decrease ( 85%) in osteocalcin mRNA compared with
minimal change in GAPDH mRNA (5%) as measured by real time PCR
after day 3 of treatment with 10 µM of antisense
oligonucleotide (Fig. 9A). In contrast, the scrambled sense
oligonucleotide did not have an effect on osteocalcin mRNA (Fig. 8A). Furthermore, only the antisense
oligonucleotide and not the scrambled sense oligonucleotide, resulted
in complete blockade of BMP-4 and Shh-N mediated anabolic activity
(Fig. 8B). Finally, we measured osteocalcin mRNA, a downstream target
of Cbfa1, using real time PCR in metatarsals treated with BMP-4 (25
ng/ml), Shh-N 10 nM, and Shh-N+ BMP-4. An approximately
3-fold increase in osteocalcin mRNA was observed with BMP-4 treatment
and an approximately 2-fold increase was observed with 10
nM Shh-N. As seen earlier cotreatment with both agents did
not increase the osteocalcin mRNA levels over and above that of BMP-4
alone (Fig. 9). Collectively, these
results suggest a mechanistic link between Shh-N/Ihh-N, BMP-2/4, and
Cbfa1/Osf-2 in mediating the anabolic affects of Shh-N/Ihh-N.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Earlier reports have shown that treatment of chick bone explants with
PTHrP (retroviral overexpression) leads to decreased collagen X
expression, a marker for hypertrophic cartilage, suggesting an
inhibitory role for PTHrP and Ihh in cartilage differentiation
(18). However, in the same study it was suggested that Ihh
is also important for chondrocyte differentiation, as exemplified by
the short-limb dwarfism phenotype in the Ihh knockout mouse
(19). Osteoblasts are primarily responsible for laying
down the collagen Ienriched matrix that accounts for most of the
mineral present in the mid-diaphysis region (20). It is
noteworthy that treatment of these metatarsals with 1 to 100
nM calcitonin did not affect their mineralization (data not
shown), in spite of the presence of multinucleated osteoclasts (refer
to Fig. 1B). The ability of recombinant BMP-4 to elicit an anabolic
response provided initial validation that this model is responsive to a
known anabolic agent. Interestingly, the appearance of noggin mRNA at
higher doses of BMP-4 suggests that this anabolic activity can be
controlled by a negative feedback loop mediated by noggin
(21). This observation may have important implications
when targeting modulators of BMP-2/4 activity for the treatment of
osteoporosis.
The process of fracture healing is shown to result in the formation of a pseudo growth plate, which provides for the recapitulation of the endochondral ossification within the microenvironment of the fracture (22). Correspondingly, it has been reported that BMP-2/4 transcripts along with that of Ihh have been identified in such microenvironments (18). Hence, there is precedence for these two signaling molecules to be coexpressed in areas undergoing new bone formation. In Ihh null mice, a specific effect on osteoblast development is observed, such that the bone collar which is derived from diaphyseal osteoblasts is lacking in these mice. This observation points to chondrocyte maturation-independent effects of Ihh and is consistent with the loss of Cbfa1/Osf-2 expression in these mice (19). Because PTHrP blocking antibody fails to inhibit this Shh-N–induced anabolic activity, it can be inferred that Shh-N/Ihh targets areas other than the PTHrP-responsive tissues, in this model. In support of this hypothesis it has been reported that the ROS17/2.8 cell line model with some features of a mature rat osteoblast, expresses Ihh and its cognate receptor patched. However, it remains to be seen whether this can effectively explain all of Shh-N's anabolic activity (23). In addition, the inability of a constitutively active PTHrP receptor (PTHRI) to rescue the short-limbed phenotype of Ihh null mice, when expressed as a transgene, is indicative of nonPTHrP pathways mediating Ihh action (7). Based on our results using Cbfa1 antisense oligonucleotides it can be inferred that Shh-N/Ihh anabolic action is mediated by osteoblasts and chondrocytes that express Cbfa1 (20, 24, 25). The precise details of mechanism will become more apparent with the availability of more sensitive techniques to detect Ihh protein as it is found to migrate across the critical cell layers during endochondral ossification.
Our results indicate that BMP-4 signaling pathways mediate bone anabolic activity of Shh-N/Ihh, in this organ culture model. This observation provides additional evidence that these two signaling molecules are intricately intertwined during embryonic and postnatal limb development (26, 27, 28). This observation that Shh/Ihh, BMP-2/4 and noggin signaling are intertwined has been previously implicated in somitogenesis (29, 30). However, the relationship between Shh-N and BMP-4 has been antagonistic unlike our current report, which seems to indicate that BMP-2/4 signaling may mediate the anabolic action of Shh/Ihh during endochondral ossification. The recent evidence that BMP-4 has a role in cartilage differentiation may shed some light (5) on the mechanism of this anabolic activity in the metatarsal model. An increase in chondrocyte differentiation caused by Shh-N/Ihh or BMP-4 could lead to increase in the presence of chondro-osteoblast precursors or to an increased scaffold laid down by the maturing chondrocytes. This would eventually result in increased mineralization. However, treatment of these metatarsals with 1, 10 nM Shh-N did not alter the level of collagen X mRNA as detected by RT-PCR (data not shown). This result would suggest that the anabolic activity for Shh-N/Ihh is not solely mediated by increase in the number of hypertrophic chondrocytes. Finally the observation that Cbfa1 antisense oligonucleotides block the anabolic activity of Shh-N and BMP-4 indicates that compromising the ability of an osteoblast to deposit bone matrix will interfere with endochondral ossification. Evidence for this role is provided by the finding that expression of a dominant negative Cbfa1 in mature osteoblasts severely affects bone volume without affecting osteoblast number (31).
In conclusion, this report has identified a mechanistic link between Shh-N/Ihh, BMP-2/4, and Cbfa1 signaling in mediating bone anabolic activity in an organ culture model. This activity of Shh-N/Ihh is independent of PTHrP but is dependent on BMP-2/4 and Cbfa1/Osf-2 signaling. Future studies on the mechanism of this anabolic action and the role of the individual cell types in mediating this anabolic action will shed light on the complex pathway of endochondral ossification.
|
Acknowledgments |
|---|
Received April 27, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Mukherjee and P. Rotwein Insulin-Like Growth Factor-Binding Protein-5 Inhibits Osteoblast Differentiation and Skeletal Growth by Blocking Insulin-Like Growth Factor Actions Mol. Endocrinol., May 1, 2008; 22(5): 1238 - 1250. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Dwyer, N. Sever, M. Carlson, S. F. Nelson, P. A. Beachy, and F. Parhami Oxysterols Are Novel Activators of the Hedgehog Signaling Pathway in Pluripotent Mesenchymal Cells J. Biol. Chem., March 23, 2007; 282(12): 8959 - 8968. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Yaar, C. Wu, H.-Y. Park, I. Panova, G. Schutz, and B. A. Gilchrest Bone Morphogenetic Protein-4, a Novel Modulator of Melanogenesis J. Biol. Chem., September 1, 2006; 281(35): 25307 - 25314. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. K. Chan, D. Miao, R. Deckelbaum, I. Bolivar, A. Karaplis, and D. Goltzman Parathyroid Hormone-Related Peptide Interacts with Bone Morphogenetic Protein 2 to Increase Osteoblastogenesis and Decrease Adipogenesis in Pluripotent C3H10T1/2 Mesenchymal Cells Endocrinology, December 1, 2003; 144(12): 5511 - 5520. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Canalis, A. N. Economides, and E. Gazzerro Bone Morphogenetic Proteins, Their Antagonists, and the Skeleton Endocr. Rev., April 1, 2003; 24(2): 218 - 235. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Krishnan, T. L. Moore, Y. L. Ma, L. M. Helvering, C. A. Frolik, K. M. Valasek, P. Ducy, and A. G. Geiser Parathyroid Hormone Bone Anabolic Action Requires Cbfa1/Runx2-Dependent Signaling Mol. Endocrinol., March 1, 2003; 17(3): 423 - 435. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
