Division of Bone and Mineral Diseases Washington University School of Medicine St. Louis, Missouri 63110
Address all correspondence and requests for reprints to: Roberto Pacifici, MD, Division of Bone and Mineral Diseases, Barnes/Jewish Hospital, North Campus, 216 S. Kings Highway, St. Louis, Missouri 63110.
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() |
---|
Calcium (Ca) balance and calcium absorption studies have demonstrated that most patients with hypercalciuria have increased intestinal Ca absorption. The terms absorptive hypercalciuria and dietary hypercalciuria are frequently used to describe those patients who exhibit an abnormal response to an oral Ca load and who excrete normal amount of Ca after prolonged fasting and/or on a low Ca diet. The terms resorptive hypercalciuria or true idiopathic hypercalciuria are used to describe those stone formers who excrete increased amounts of Ca even when maintained on a low Ca diet or during the fasting state, pointing to the skeleton as an additional source of urinary Ca.
Studies have revealed that patients with fasting IH have decreased axial and peripheral bone density (2, 3) and increased bone resorption (4). Hypercalciuric stone formers are often advised to adhere to strict low Ca diets. Thus, chronic calcium deprivation may be a contributing factor. However, the finding of a more severe bone loss in patients with fasting IH strongly suggests that an abnormality in bone remodeling may be responsible, at least in part, for the excessive Ca excretion characteristic of these patients.
Although primary hyperparathyroidism or a renal "Ca leak" resulting in a tendency to hypocalcemia and secondary hyperparathyroidism may account for increased bone resorption and bone loss in some patients, PTH levels are usually normal or low in patients with IH. Increased levels of 1,25 (OH)2D3 are also frequently found in these patients. However, increased calcitriol production is unlikely to account for increased bone resorption because 1,25 (OH)2D3 levels are generally similar in patients with absorptive and fasting hypercalciuria, and 1,25 (OH)2D3 levels are positively correlated with bone. Histological studies have revealed that patients with IH typically exhibit high bone turnover, the hallmark of postmenopausal osteoporosis and of that induced by androgen deficiency in men. High bone turnover has also been documented as a typical feature of a syndrome described by Perry et al., (5) characterized by the presence of hypercalciuria and osteoporosis in eugonadic men.
Postmenopausal osteoporosis (and its equivalents in men) is a heterogeneous disorder characterized by a progressive loss of bone tissue that begins after natural or surgical menopause and leads to fracture within 1520 yr from the cessation of the ovarian function. The bone-sparing effect of sex steroids is mainly related to their ability to block bone resorption, although stimulation of bone formation is likely to play a contributory role. Sex steroid-dependent inhibition of bone resorption is, in turn, the result of both decreased osteoclastogenesis and a diminished resorptive activity of mature osteoclasts.
In recent years our understanding of the mechanism of sex steroid
action in bone has grown considerably. This is mainly a result of the
recognition that sex steroid regulates bone remodeling by modulating
the production of cytokines and growth factors from bone marrow and
bone cells (6, 7). Among these factors are IL-1 and ß, IL-6, tumor
necrosis factor
and ß (TNF), M-CSF, and GM-CSF. In addition to
osteoporosis, an increasing number of common clinical entities have
been ascribed to an abnormal production of bone resorbing cytokine by
bone and/or bone marrow cells. Among them are Pagets disease, primary
hyperparathyroidism, Ghoram-Stouts disease, post-transplant
osteoporosis, thyroid hormone induced bone loss, endometriosis,
multiple myeloma, and rheumatoid arthritis.
![]() |
Cytokines and Bone Remodeling |
---|
![]() ![]() ![]() ![]() ![]() |
---|
The interpretation of the biological effects of IL-1 is further
complicated by the fact that the binding of IL-1 to the type I receptor
is antagonized not only by IL-1ra, but also by soluble type I (sIL-1
RI) and type II IL-1 receptor (sIL-1 RII) anti IL-1 autoantibodies
and IL-1ß binding proteins. Moreover, while sIL-1 RI antagonizes the
effects of IL-1ra, sIL-1 RII binds IL-1ß, but does not bind IL-1ra.
Thus, sIL-1 RII can compete with cell-associated receptors for IL-1ß
and potentiate the inhibitory action of IL-1ra. Recognition that the
biological effects of IL-1 are not only a function of net concentration
of IL-1 molecules, but rather of the fine balance between agonist and
antagonist molecules, may facilitate the interpretation of contrasting
data obtained measuring IL-1 activity by bioassay and IL-1
concentrations by enzyme linked immunoabsorsent assays or
immunoradiometric assays. For example, an association between estrogen
deficiency and increased IL-1 activity was demonstrated in studies
conducted by measuring IL-1 activity by bioassay. Conversely, this
association was not observed when IL-1 was measured by enzyme linked
immunoabsorbent assay or immunoradiometric assay.
Another cytokine that has received considerable attention for its proosteoclastogenic effects is IL-6. This factor exerts its effects via a cell surface receptor that consists of a ligand binding chain (IL-6R) and a signal transducing chain known as gp130. Although IL-6 alone does not stimulate osteoclast formation, when bound to soluble IL-6R, IL-6 stimulates the early stages of osteoclastogenesis in human and murine cultures, presumably by forming a complex with gp130 expressed on either stromal cell or osteoblasts. Interestingly, corticosteroids have been found to increase the stromal cell expression of IL-6R, suggesting the possibility that steroid-induced bone loss may be caused, at least in part, by an increase in stromal cell responsiveness to IL-6. IL-6 increases in vitro bone resorption in systems rich in osteoclast precursors, such as the mouse fetal metacarpal assay, whereas it has no effect in organ cultures where more mature cells predominate, such as murine fetal radii. This suggests that IL-6 is more potent in increasing the formation of osteoclasts from hemopoietic precursors than in activating mature osteoclasts. Nevertheless, the effects of IL-6 on bone resorption in vivo remain controversial, because blocking of IL-6 does not decrease in vivo bone resorption, because IL-6 levels do not correlate with indices of bone turnover in postmenopausal women, and because osteoporosis is not a feature of transgenic mice overexpressing IL-6 or IL-6R. However, this cytokine does cause hypercalcemia in nude mice.
The formation of osteoclasts in bone marrow cultures is also increased by GM-CSF. This factor stimulates the early stages of osteoclastogenesis in cooperation with IL-3. In humans, GM-CSF is indeed the most potent proosteoclastogenic factor among the known growth factors and cytokines. Thus, although in the mouse osteoclast formation is completely blocked by anti-M-CSF but not anti-GM-CSF antibodies, GM-CSF is critical for the proliferation and differentiation of human osteoclast precursors.
Although most bone cell-targeting cytokines are produced by either bone and bone marrow cells, mononuclear cells of the monocyte/macrophage lineage are recognized as the major source of IL-1 and TNF. In contrast, proosteoclastogenic "downstream" cytokines are mainly produced by stromal cells and osteoblasts. Thus, osteoclastogenesis requires the hierarchical interaction of mononuclear cells, stromal cells, and/or osteoblasts and hematopoietic osteoclast precursors, as well as the conditioning effect of numerous cytokines.
![]() |
Cytokines and idiopathic hypercalciuria |
---|
![]() ![]() ![]() ![]() ![]() |
---|
Interestingly, IL-6 levels were not significantly increased in subjects with IH. This finding is not unexpected because IL-6 does not stimulate in vivo bone resorption, and inhibition of IL-6 activity with anti-IL-6 antibody does not prevent ovariectomy-induced bone loss (9). Thus, information obtained in other experimental models is consistent with the notion that IL-6 does not modulate the increase in bone resorption of IH patients.
Ghazzali and colleagues (8) also reported that bone density was significantly lower in IH patients than in age-matched controls. Interestingly, the study also revealed the lack of a correlation between bone density and spontaneous cytokine production, although an inverse relationship was found between bone density and LPS-induced IL-6 secretion, and a positive correlation was noted between bone density and LPS-induced GM-CSF levels. Several factors should be taken into consideration in interpreting the apparent complex relationship between cytokine levels and bone mass. First, cytokine levels reflects the rate of bone remodeling at the time of sample collection, whereas bone density measurements provide values reflecting all past and current events capable of influencing skeleton development and involution. Thus, cross-sectional studies are rarely informative with respect to the effects of cytokines on bone mass. Second, the secretion of some cytokines, such as IL-6, increases with age (10), whereas the production of other factors, such as IL-1 and TNF does not. Thus, the opposite effects of aging on IL-6 production and bone mass could account for the existence of a correlation between these two variables independently of the effects of IL-6 in bone.
The interesting study by Ghazzali et al. (6) confirms and
extends previous reports on a possible role for cytokines in the
pathogenesis of IH (4, 11). Studies by us had demonstrated that
patients with fasting IH have decreased bone mass and an increased
monocytic production of IL-1 and increased bone resorption (4).
However, we did not examine the production of other critical bone
resorbing factors. In a subsequent study Weisenger et al.
(11) demonstrated that monocytes from patients with IH have an
increased expression of IL-1 m RNA and produce larger amounts of
IL-1
. They also confirmed the existence of a correlation between
age-normalized bone density values and IL-1
levels. The study of
Ghazzali et al. (8) not only confirmed these observations,
but also provides insights on the contribution of TNF and GM-CSF.
However, it should be recognized that none of these three studies demonstrate the existence of a cause/effect relationship between increased production of cytokines, bone loss, and hypercalciuria. The possibility that increased amounts of cytokines are produced in response to increased bone resorption should be considered. Collagen and other matrix proteins released into the bone microenvironment during bone resorption are, in fact, known to bind to integrin receptors expressed in monocytes and stimulate cytokine production (12).
Although a direct link between increased cytokine production and IH
remains to be demonstrated, an increased monocytic production of IL-1
and TNF has been shown to play an important causal role in
postmenopausal osteoporosis (6). Studies conducted with specific
inhibitors of IL-1 and TNF have, in fact, demonstrated that the
functional block of these cytokines prevent bone loss and block
osteoclast formation and in vivo bone resorption in
ovariectomized rats and mice (13, 14). Thus, it is tempting to
speculate (Fig. 1) that both IH and postmenopausal bone
loss may result from a primary increase in bone resorption induced by
an overproduction of critical cytokines such as IL-1, TNF, and GM-CSF
by bone marrow cells. The different outcome of these two clinical
entities (stone formation vs. spontaneous fractures) may be
conditioned by factors such as intestinal absorption of calcium or
calcitriol serum levels, which are both high in IH and low in
postmenopausal subjects. It should also be noted that, while sex
steroids have been demonstrated to directly regulate cytokine
production via an effect on cytokine gene expression, the mechanism by
which cytokine production is upregulated in IH remains to be
determined.
|
Received October 21, 1996.
Accepted October 23, 1996.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() |
---|