Differentiation of osteoclast progenitors into mature osteoclasts

1α,25-Dihydroxyvitamin D3 enhances bone resorption
by stimulating differentiation and fusion of
mononuclear osteoclast progenitors into mature
multinucleated osteoclasts (Clohisy et al., 1987; Takahashi
et al., 1988). This process, osteoclastogenesis,
involves a complex interaction of osteoclast progenitors,
osteoblasts and bone marrow-derived stromal
cells (Suda et al., 1992a). Studies using VDR-ablated
mice showed that stimulation of osteoclast formation
by 1α,25(OH)2D3 requires the presence of VDR in
osteoblast-like cells but not in osteoclast precursor
cells; however, if VDR is absent in the osteoblastic
cells, PTH or interleukin-1α can stimulate osteoclast
formation (Takeda et al., 1999).
Stromal cells of the bone marrow control osteoclastogenesis
through the production of cytokines capable
of promoting the proliferation and differentiation
of osteoclast progenitors. One particular cytokine, interleukin-
11 (IL-11), is a rather specifi c product of the
mesenchymal cell lineage, which includes bone marrow
stromal cells and osteoblasts (Paul et al., 1990).
IL-11 is a potent inducer of osteoclast development
(Girasole et al., 1994) and its production by primary
osteoblasts can be stimulated by 1α,25(OH)2D3 among
other osteoclastogenic factors (Romas et al., 1996).
The addition of anti-IL-11 antibody to bone marrow
cell cultures suppresses the ability of 1α,25(OH)2D3 to
induce osteoclast development (Girasole et al., 1994),
suggesting that IL-11 is an essential factor for the osteoclastogenic
effect of 1α,25(OH)2D3. Another soluble
factor, macrophage-colony stimulating factor (MCSF),
is essential for osteoclast differentiation from
progenitors, but its production by osteoblasts/stromal
cells does not appear to be regulated by 1α,25(OH)2D3
(Suda et al., 1992a).
In 1997, two independent research groups reported
the discovery of novel cytokines which inhibited the
differentiation of osteoclast progenitors into mature
osteoclasts. These secreted proteins were named
osteoclastogenesis inhibitory factor (OCIF) and
osteoprotegerin (OPG) by the respective groups
(Tsuda et al., 1997; Simonet et al., 1997). Yasuda et
al. (1998a) found these two proteins to be identical,
hence the present term, OPG/OCIF. Transgenic mice
with over-expressed OPG/OCIF and mice injected
with OPG/OCIF exhibited profound yet non-lethal
osteopetrosis, coincident with arrested osteoclast
development in the later stages (Simonet et al., 1997).
Yasuda et al. (1998a) reported that the expression of
the OPG/OCIF gene in stromal cells is down-regulated
by 1α,25(OH)2D3 and up-regulated by calcium
ions. These results imply that OPG/OCIF regulates
osteoclastogenesis in response to stimulators of bone
resorption and calcium ions released at bone-resorbing
sites.
1α,25-Dihydroxyvitamin D3, along with PTH and a
number of other factors, stimulate osteoclastogenesis
through signal transduction pathways mediated by an
osteoclast differentiation factor (ODF) on the membrane
of osteoblasts/stromal cells (Suda et al., 1992b).
OPG/OCIF inhibits in vitro osteoclastogenesis by
directly binding to the ODF, thereby interrupting
ODF-mediated signalling from osteoblast/stromal
216 Vitamins: their role in the human body
cells to osteoclast progenitors for their differentiation
into mature osteoclasts (Tsuda et al., 1997). 1α,25-Dihydroxyvitamin
D3, by down-regulating OPG/OCIF,
permits osteoclastic bone resorption.
Yasuda et al. (1998b) showed that ODF is identical
to a regulator of T lymphocyte cells and dendritic
cells designated TRANCE or RANKL. Expression of
the TRANCE/RANKL/ODF gene was up-regulated by
osteotropic factors, including 1α,25(OH)2D3.
Sato et al. (1991) reported that mouse bone marrow-
derived stromal cells and primary osteoblastic
cells in vitro produce the third component of complement
(C3) in response to 1α,25(OH)2D3. This appears
to be a bone-specifi c effect as C3 production by hepatocytes
is not dependent on 1α,25(OH)2D3. Bone C3
is also induced tissue-specifi cally by 1α,25(OH)2D3
in vivo (Jin et al., 1992). Actinomycin D completely
inhibits the effect of 1α,25(OH)2D3 on both mRNA
expression and protein production of C3, indicating
that the hormone acts at the transcriptional level
(Hong et al., 1991). The addition of anti-C3 antibody
to mouse bone marrow cultures completely inhibits
the 1α,25(OH)2D3-induced formation of osteoclastlike
cells, suggesting that the C3 produced by stromal
cells in response to 1α,25(OH)2D3 is somehow involved
in osteoclast formation (Sato et al., 1991). The
production of C3 in stromal cells and osteoblastic
cells is also stimulated by local bone-resorbing agents,
such as interleukin-1, tumour necrosis factor-α and
lipopolysaccharides (Hong et al., 1991). Sato et al.
(1993) concluded that the bone C3, acting in concert
with other factors induced by 1α,25(OH)2D3, potentiates
proliferation of bone marrow cells and induces
differentiation of these cells into osteoclasts.
Attachment of osteoclasts to the bone matrix
1,25-Dihydroxyvitamin D3 activates the transcription
of integrin αv and β3 subunit genes, resulting in an
increased number of integrin αvβ3 receptors on the
surface of osteoclast progenitors (Medhora et al.,
1993; Mimura et al., 1994). These receptors recognize
and bind to the bone matrix proteins osteopontin
and bone sialoprotein. Expression of integrin αvβ3
coincides with the differentiation of progenitors into
osteoclasts and is essential to the resorptive process.
Activation of quiescent osteoclasts
1α,25-Dihydroxyvitamin D3 stimulates osteoclastic
bone resorption in tissue culture (Raisz et al., 1972)
and in rat bones in vivo (Holtrop et al., 1981). Mc-
Sheehy & Chambers (1987) reported that isolated osteoclasts
do not respond to 1α,25(OH)2D3 if incubated
alone, but they do so if incubated in the presence of
osteoblastic-like cells. Incubation of osteoblastic cells
in the presence of 1α,25(OH)2D3 produced a soluble
factor that stimulated osteoclastic bone resorption.
Jimi et al. (1996) demonstrated that cell-to-cell contact
between osteoblastic cells and osteoclast-like cells
was required to promote pit-forming activity. Mee et
al. (1996) were the fi rst to show that active human
osteoclasts in vivo possess mRNA for the VDR. It
seems, therefore, 1α,25(OH)2D3 can act directly on
osteoclasts to stimulate bone resorption as well as
indirectly through its effects on osteoblasts.
8.6.9 Calcium homeostasis
The control of calcium homeostasis involves three
major sites: bone, the kidneys and the intestine.
Bone is the major reservoir of calcium in the body,
storing around 99% of the total. The role of bone
in calcium homeostasis is to ‘buffer’ blood calcium
level, releasing Ca2+ to the blood when the blood
level decreases and taking Ca2+ back when the level
rises. Calcium homeo stasis is coordinately regulated
by 1α,25(OH)2D3 and parathyroid hormone (PTH),
with calcitonin playing an important supporting role.
PTH is a peptide hormone produced by the parathyroid
glands in response to low plasma calcium levels.
Its action is mediated by PKA via the second messenger
cyclic AMP (cAMP) (Horiuchi et al., 1977). In the
kidney, PTH activates the 25(OH)D3-1α-hydroxylase
by increasing the enzyme’s mRNA through effects on
gene transcription (Brenza et al., 1998; Murayama et
al., 1998). Simultaneously, PTH suppresses the renal
24R-hydroxylase (Shinki et al., 1992), the enzyme responsible
for catalysing the C-24 oxidation pathway.
This reciprocal regulation by PTH in the kidney allows
for 1α,25(OH)2D3 production with minimal
concomitant catabolism. The effect of 1α,25(OH)2D3
is to restore plasma calcium levels to normal. PTH
has no effect on intestinal 24R-hydroxylase activity
(Shinki et al., 1992), which is not surprising as the intestine
lacks PTH receptors. In osteoblasts, PTH acts
synergistically with 1α,25(OH)2D3 to increase transcription
of the 24R-hydroxylase gene (Armbrecht et
al., 1998). Thus PTH exerts opposite effects in kidney
and bone with respect to 24R-hydroxylase regulation.
Vitamin D 217
Activation of the 24R-hydroxylase in bone is a means
of regulating the resorptive action of 1α,25(OH)2D3.
Cycloheximide has no effect on the capacity of PTH to
increase 24R-hydroxylase mRNA levels in osteoblasts,
showing that the action of PTH does not require new
protein synthesis. One possible mechanism by which
PTH could increase 24R-hydroxylase promoter
activity in conjunction with 1α,25(OH)2D3 is phosphorylation
of the cAMP-response element binding
protein (CREB) by PKA and consequent binding of
the CREB to a cAMP-response element (CRE). Interaction
between CREB/CRE and VDR/VDRE complexes
would synergistically increase 24R-hydroxylase
promoter activity.
PTH and 1α,25(OH)2D3 mutually regulate each
other’s synthesis and/or secretion. PTH downregulates
VDR abundance in the kidney (Reinhardt
& Horst, 1990), while 1α,25(OH)2D3 lowers preproPTH
mRNA levels in the parathyroids (Silver et
al., 1985, 1986) with a subsequent reduction in PTH
secretion (Cantley et al., 1985). In osteoblastic cells,
1α,25(OH)2D3 reduces PTH-stimulated cAMP production
(Pols et al., 1986), while PTH up-regulates
VDR abundance (Krishnan et al., 1995).
The homeostatic control of calcium is represented
schematically in Fig. 8.10. In response to a lowered
concentration of serum Ca2+, the parathyroid glands
are stimulated, within minutes, to secrete PTH. This
hormone stimulates, within hours, the activity of
renal 25(OH)D-1α-hydroxylase. The 1α,25(OH)2D3
thus produced acts by itself to initiate active calcium
transport in the intestine. In bone, 1α,25(OH)2D3
acts synergistically with PTH to mobilize Ca2+ (accompanied
by Pi) from the bone fl uid compartment
into the bloodstream. The presence of both PTH and
the vitamin D hormone are required for this system to
operate in vivo (Jones et al., 1998). In the kidney, PTH
and 1α,25(OH)2D3 act in concert to cause the reabsorption
of the last 1% of the fi ltered load of calcium
into the plasma compartment. The resultant increase
in the concentration of circulating Ca2+ provides a
powerful negative feedback signal to the parathyroids,
suppressing the secretion of PTH and ultimately stopping
the renal production of 1α,25(OH)2D3.