| Abstract |
Changes in bone homeostasis and development in states of myeloproliferative neoplasms (MPNs)
are complex. Continued research is needed to elucidate mechanisms that govern such changes. In
the current Ph.D. dissertation, I focused on primary myelofibrosis (PMF), a form of MPN that is
characterized by excess proliferation of megakaryocytes (MKs) and bone marrow (BM) fibrosis.
Osteosclerosis has been reported as a serious complication ofsome MPNs. Although osteosclerosis
contributes to BM failure, and is associated with severe joint and bone pain, its manifestation and
etiology in PMF have remained elusive. The increased number of MKs could be involved in bone
changes since MKs have been reported to influence bone cells differentiation.
Several investigations point to increased bone volume in PMF, but comparison according to sex
differences has not been fully performed. In Aims 1 and 2 of our study, we used a humanized
transgenic mouse model of PMF carrying the most common mutation associated with this disease,
namely the human JAK2V617F hyper-activating mutation, under the control of the Vav promoter.
These mice were used to study influences of the mutation on bone density and strength in the two
sexes (Aim 1 of the dissertation). Overall, micro-computed tomography (micro-CT) revealed
significant phenotypes in both males and females, but the effect of the mutation seemed to have
opposite effects on trabecular bone in the two sexes. It increased trabecular bone volume in
females, while decreasing it in males, as compared to matching controls. At the femur mid-shaft,
the transgenic male mice had much greater porosity and lower mineral density and cortical
thickness, compared to controls. There were also trends towards JAK2V617F male mice having
lower moments of inertias, which is one of the main indicators of bending strength. For this reason,
we proceeded to mechanical testing (three-point bending), and observed that male JAK2V617F mice
had similar or slightly inferior bone mechanical properties relative to their controls. Based on the
micro-CT and three-point bending results, we next assessed the bones at tissue level to examine
differences in the mineralization process. This analysis revealed the presence of more osteoid in
tissues from JAK2V617F male mice compared to controls, which is the unmineralized organic
portion of the bone matrix that is formed prior to the maturation of bone tissue. This indicated that
the mineralization process is hampered in male JAK2V617F mice. The next step (Aim 2 of the
dissertation) was to investigate how MKs derived from male mice that bear the JAK2V617F mutation
influence the differentiation of osteoblasts (OBs) that are responsible for the mineralization process. This was carried out with bone marrow Mesenchymal Stem Cells (MSC)-MK co-cultures,
using MKs from male mice and culture conditions that promote OB differentiation (therefore,
referred to also as an OB-MK co-culture). The presence of JAK2V617F MKs and/or their
conditioned medium significantly decreased OB differentiation.
In search for possible candidates responsible for the effect of JAK2V617F mutated MKs on
osteogenesis, we used a mouse mRNA osteogenesis array, and several transcripts were identified
as differentially expressed between the mutated MKs and control ones. There was increased
expression of Noggin, Chordin, Alpha-2-HS-glycoprotein, Collagen type IV alpha 1 and Collagen
type XIV alpha 1 (mostly known to inhibit bone differentiation), and decreased expression of
Alkaline phosphatase, Vascular cell adhesion molecule 1, Sclerostin, Distal-less homeobox 5 and
Collagen type III alpha 1 (associated with osteogenesis) in JAK2V617F MKs from male mice,
compared to controls. For each gene or protein differentially displayed in JAK2V617F mutated MKs
compared to controls, we contemplated cause-and-effect studies to examine their contribution to
the inhibitory effect of male MKs on OB differentiation. Yet, it became apparent that the mutated
MKs undergo a change in global gene programming involving several pathways linked to
inhibition of osteogenesis. Hence, we propose that a cluster of such changes, and not a single one,
is promoting the inhibition observed in OB differentiation by mutated MKs.
Contrary to the observation in male JAK2V617F mice, micro-CT analysis of mutated female mice
showed greater bone volume relative to control mice. Based on mechanical testing analysis it was
concluded that this mutation-acquired phenotype is likely due to differences in bone geometry and
not differences in the intrinsic material properties of the bone. This suggested that in females there
are no detrimental effects of JAK2V617F expression on either whole bone strength or intrinsic bone
tissue quality. Initial OB-MK co-culture experiments showed a trend for an inhibitory or no effect
of JAK2V617F MKs from female mice on OB differentiation. An osteogenesis gene expression array
was used to test JAK2V617F MKs from female mice in comparison to control MKs, showing that
the majority of the differentially expressed transcripts pointed to an OB inhibitory profile as well.
However, a vast increase in the expression of the bone differentiation promoting factor, Bone
Morphogenetic Protein Receptor type-1B could contribute to the bone gain observed in the
JAK2V617F female mice, compared to their matching controls. Among other factors, lysyl oxidase (LOX) was found to be highly elevated in MKs from PMF
patients and in mouse models of PMF. This enzyme has been studied both in context of bone
formation and fibrosis. In Aim 3 of this dissertation, we also examined the innovative contention
that MK LOX plays a significant role in controlling bone integrity under normal conditions, which
if found to be the case, would motivate future studies on the role of MK LOX in shaping bone
changes in PMF. Previous studies examined the role of LOX in bone formation. However, the role
of MK LOX in MK and bone development had not been investigated. To this end, we generated
transgenic mice that have LOX deleted in their MKs, by utilizing the Cre-lox system with clustered
regularly interspaced short palindromic repeats (CRISPR/Cas9) technology. Deletion of LOX in
MKs did not affect MK development, since MK number and ploidy did not change. However, MK
LOX affected bone volume in a sex-dependent way. There was increased bone volume in male
mice that have LOX specifically deleted in their MKs compared to control mice, while no
difference was observed in the females compared to matching controls. The upregulation of Bone
Morphogenetic Protein 2, Alkaline phosphatase and Fibronectin 1 (associated with osteogenesis)
in the MKs of the male transgenic animals could account for the increased bone volume. The male
transgenic mice had inferior apparent bone mechanical properties compared to their controls,
which could be explained by the disrupted collagen profile that was observed in their bones. The
increased expression of Matrix Metallopeptidase 8 (which is involved in collagen degradation) in
these animals further supports these findings. These intriguing results will serve as basis for future
mechanistic investigations.
Together, this research dissertation provides important new knowledge regarding sex-dependent
changes in bone volume and properties in PMF, enabled by JAK2V617F
-induced reprograming of
MKs towards expression of OB differentiation-promoting factors. As mentioned earlier, literature
suggests that osteosclerosis is a serious complication of some forms of MPNs. Our new findings
that JAK2V617F mutated male mice present with significant bone loss highlight a gap in human
studies, since comparison according to patients’ sex was not reported in the existing literature.
Further, we made the new discovery that LOX specifically expressed in MKs contributes to bone
homeostasis in male mice. Both aims of study provide ground for future investigation of effects of
sex hormone-related changes on bone in PMF.
|
Collections
