| |
|
|
Roland
Baron, DDS, PhD
Prof. Orthopaedics and Cell Biology
|
 |
Research
Summary
Our laboratory studies several of the molecular mechanisms
that regulate the formation of bone by osteoblasts and bone
resorption by osteoclasts. Our major effort focuses on signaling
mechanisms in osteoclasts that involve the nonreceptor tyrosine
kinase c-Src. Transgenic mice that |
lack the gene for Src are osteopetrotic (have excessive
bone) as a consequence of the inability of the Src-/-
osteoclasts to resorb bone. Using adenovirus constructs
to express mutant proteins in osteoclasts, we have shown
that Src acts downstream of the vitronectin receptor (VnR)
or avb3 integrin, the most common integrin on osteoclasts,
to regulate the assembly and disassembly of podosomes,
the highly dynamic actin-rich structures that mediate
osteoclast (OC) attachment and migration. Following the
binding of the VnR on osteoclasts to matrix proteins,
the tyrosine kinases Pyk2 and Src are sequentially recruited
to the membrane, along with the adaptor protein Cbl.
Autophosphorylation of Pyk2 creates a binding site for
Src’s SH2 domain. Overexpressing a form of Pyk2
that has the binding site mutated inhibits in vitro bone
resorption in a dominant-negative manner, indicating that
the recruitment of Src to the Pyk2-containing complex
is a critical part of Src signaling in osteoclasts. Src
kinase activity is equally important, since the overexpression
of catalytically inactive Src also acts in a dominant-negative
manner to inhibit in vitro bone resorption. We are currently
identifying and characterizing the molecules that interact
with the Pyk2-Src complex. One of these is the ubiquitin
ligase and adaptor protein Cbl, which binds to Src’s
SH3 domain as well as to several other proteins, most
notably phosphatidylinositol 3-kinase (PI3K), a protein
that is independently known to be required for normal
osteoclast function and for macrophage motility. The interaction
of Cbl and PI3K appears to be a critically important event
downstream of Pyk2 and Src, since overexpressing a Cbl
mutant that lacks the PI3K-binding site strongly inhibits
in vitro bone resorption. The GTPase dynamin is another
protein that appears to be involved in Src-related mechanisms
that regulate osteoclast motility and bone resorbing activity.
Dynamin is essential for endocytosis and we have shown
it to also be involved in actin remodeling and podosome
turnover in osteoclasts. Dynamin interacts indirectly
with Cbl, possibly via the adaptor protein Grb2. The Cbl-dynamin
complex is destabilized by catalytically active Src and
stabilized by kinase-inactive Src. We recently found that
Pyk2 also associates indirectly with dynamin. The Pyk2-dynamin
association doesn’t require Src, but dynamin destabilizes
the Src-Pyk2 complex in a yet undetermined way. We are
currently studying the mutual regulation of the intermolecular
interactions of Pyk2, Src, Cbl and dynamin, in particular
the roles of Pyk2 and Src kinase activities and dynamin
GTPase activity, and how these proteins interact to regulate
osteoclast morphology and motility, podosome turnover
and bone resorption.
|
In a related project, we are characterizing the roles
of Cbl and the related gene product, Cbl-b in osteoclasts.
The two proteins share numerous domains, some of which
have nearly identical sequences, and the embryonic death
of Cbl/Cbl-b double-knock out mice indicates a redundancy
of key functions. However, there are a number of specific
binding motifs that are present only in one or the other
protein, and Cbl and Cbl-b single knockout mice exhibit
different osteoclast phenotypes, with the Cbl-b deletion
leading to increased differentiation and activity of osteoclasts
together with increased surface expression and signaling
activity of the osteoclastogenic receptor RANK, whereas
deletion of c-Cbl reduces OC motility. Our
hypothesis is that the differences in structural motifs
and the resulting molecular interactions mediate specific
functions of each of the two proteins that cannot be compensated
by the other. We are working to identify the specific
motifs and residues within Cbl-b whose absence cause the
osteopenia of the Cbl-b-/- mouse and the molecular associations
that they mediate. We are approaching this problem by
generating chimeric proteins that replace specific regions
of Cbl-b with the corresponding non-identical Cbl sequence
and determining how the overexpression of the chimeric
proteins affect osteoclast differentiation and function.
|
Calcitonin
(CT) is a peptide hormone that regulates calcium homeostasis
by potently inhibiting bone resorption by osteoclasts and
promoting calcium excretion by the kidney. Understanding
the mechanisms by which CT inhibits osteoclast bone resorbing
activity has been a longstanding interest of our group.
Following the initial cloning of the CT receptor (CTR),
we identified an unusual splice variant that lacked much
of the seventh transmembrane domain. We showed that the
novel splice variant coupled to the cAMP pathway but not
to pathways involving the activation of phospholipase C
and consequent Ca2+ signaling, and that the relative levels
of expression of this variant and the more common unspliced
form vary in a tissue-specific manner. More recently, we
have shown that the unspliced isoform constitutively cycles
between the cell surface and intracellular compartments
in a mechanism that depends on its binding to the actin
cytoskeleton-associated protein filamin A, and that the
alternatively spliced isoform regulates the trafficking
of the unspliced variant to the cell surface in a dominant-negative
fashion. We have also found that CT regulates several podosome-associated
proteins, including Pyk2, talin and filamin, by mechanisms
that involve Src and the calcium-dependent protease calpain.
We are currently investigating the mechanisms that couple
the CTR to these intracellular signaling effectors and to
the regulation of podosome assembly and disassembly.
Our
current investigation of osteoblast regulation began with
the observation that overexpressing the AP-1 family transcription
factor DFosB in transgenic mice led to a severe and progressive,
but reversible increase in bone formation and bone mass
leading to an osteosclerotic phenotype. We showed that DFosB
had the same effects when it was specifically expressed
in osteoblasts using the osteocalcin promoter, demonstrating
that the protein acts directly in osteoblasts. Transcription
factors such as DFosB act by binding to DNA and to other
DNA-regulating proteins, suggesting that the overexpressed
DFosB is disturbing the balance of transcription factor
complexes in a way that increases osteoblast differentiation
and bone formation. We therefore focused our efforts on
identifying other transcription factors that bind to DFosB.
A yeast two-hybrid screen identified a number of proteins
that bound to DFosB, including Runx2, which is known to
play a key role in promoting bone formation, and a novel
transcription factor composed of 30 zinc finger domains,
denoted ZFP 521. ZFP 521 is normally expressed at high levels
in primary calvarial osteoblasts as well as in brain, heart,
lung, and skeletal muscle and at lower levels in fat, spleen,
kidney and liver. In situ hybridization studies demonstrated
that during bone development, it is expressed in mesenchymal
condensations, in the preosteoblastic layers, and in the
growth plate, indicating a potentially important role in
the regulation of osteoblast and chondrocyte development.
Consistent with that possibility, ZFP 521 expression in
several mesenchymal cell lines is reduced by treating with
the osteogenic agent BMP-2.
|
Overexpression
of ZFP 521 in primary osteoblasts and osteoblastic cell
lines decreases alkaline phosphatase activity, inhibits
the Runx2-iduced activation of Runx2 and osteocalcin promoters
in luciferase assays and suppresses the DFosB-induced increase
in Runx2 expression. We are currently continuing to characterize
the functional interaction of ZFP 521 with Runx2, DFosB
and other transcriptional regulatory proteins and testing
the hypothesis that ZFP 521 acts as a Runx2 repressor during
osteoblast development. |
|
|
Last
Edited
11/07/2005
JBS |
|
Figure
Legends
Fig. 1 An osteoclast
Fig. 2 The Bone Remodeling Sequence
Fig. 3 Src kinase inhibition blocks bone resorption and stimulates
bone formation
Fig. 4 Cbl Binds to Src and negatively regulates its kinase activity
and cell adhesion
Fig. 5 Osteoclast motility and activity requires podosome assembly
and disassembly (Sanjay
et al., J Cell Biol 2001)
Fig.
6 & 7 Transgenic mice are osteosclerotic (Sabatakos et al.,
Nature Med, 2000)
|
