| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Mount Sinai Bone Program and Department of Medicine and Geriatrics, Mount Sinai School of Medicine, and Division of Endocrinology and Geriatric Research Education and Clinical Center (GRECC), Veterans Affairs Medical Center, New York, New York, USA;
* Trudeau Institute, Lake Placid, New York, USA;
The Laboratory of Physiology, University of Cambridge, CB2 3EG, UK; and
Department of Pathology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
1Correspondence: Division of Endocrinology, Box 1055, Mount Sinai School of Medicine, One Gustave Levy Place, New York, NY 10029, USA. E-mail: mone.zaidi{at}mssm.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
fourfold increase in osteoclast formation in response to incubation for 6 days with RANK-L and M-CSF. The osteoclasts so formed in these cultures showed a 2.5-fold increase in resorptive activity compared with wild-type cells. However, when adherent bone marrow stromal cells were allowed to mature into alkaline phosphatase-positive colony-forming units (CFU-Fs), those derived from CD38-/- mice showed a significant reduction in differentiation compared with wild-type cells. Real-time RT-PCR on mRNA isolated from osteoclasts at day 6 showed a significant reduction in IL-6 and IL-6 receptor mRNA, together with significant decreases in the expression of all calcineurin A isoforms, 
, ß, and 
. These findings establish a critical role for CD38 in osteoclast formation and bone resorption. We speculate that CD38 functions as a cellular NAD+ "sensor," particularly during periods of active motility and secretion.—Sun, L., Iqbal, J., Dolgilevich, S., Yuen, T., Wu, X.-B., Moonga, B. S., Adebanjo, O. A., Bevis, P. J. R., Lund, F., Huang, C. L. H., Blair, H. C., Abe, E., Zaidi, M. Disordered osteoclast formation and function in a CD38 (ADP-ribosyl cyclase) -deficient mouse establishes an essential role for CD38 in bone resorption.
Key Words: Ca2+ channel • osteoclast • ryanodine receptor • bone resorption • osteoporosis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
4
5)
. Cyclic ADPr activates ryanodine receptors (RyRs), a family of ubiquitously expressed Ca2+ release channels (6)
.
CD38 is distributed widely in hematopoetic cells, including B and T lymphocytes, thymocytes, plasma cells, macrophages, and erythrocytes, as well as in kidney, cardiac, pancreatic, spleen, lung, and liver cells (7
, 8)
. It has been implicated in the control of several cellular processes. Through its generation of cADPr, the enzyme is thought to play a critical role in Ca2+-induced insulin release from pancreatic ß cells (9
10
11)
. CD38-deficient mice thus display impaired glucose tolerance (12
, 13)
. Additional proposed roles for CD38 include the modulation of cell proliferation, apoptosis, T lymphocyte signaling, neutrophil migration, and neurotransmission (14
15
16
17)
. On a more general biological front, CD38 has been shown recently to gate nucleoplasmic Ca2+ influx across the inner nuclear membrane through the activation of its resident RyRs (18
, 19)
. The physiological significance of this new finding to the process of nuclear Ca2+ homeostasis, however, remains unclear.
Recently, we provided preliminary evidence that CD38 plays a role in remodeling of the adult skeleton, particularly in regulating bone resorption by the osteoclast (20)
. Remodeling results in constant skeletal renewal through the precise integration of osteoclastic bone resorption and osteoblastic bone formation. Increased osteoclastic activity relative to bone formation results in low bone mass (osteoporosis), and vice versa. Through molecular cloning and immunodetection, we found that CD38 was expressed in abundance in the osteoclast (20
, 21)
. We also showed that an agonist anti-CD38 antibody that triggered cytosolic Ca2+ release inhibited bone resorption, but increased IL-6 expression (20)
.
We and others have suggested that CD38 may function as a metabolic "sensor" (22)
. In this role, it could potentially sense changes in cellular NAD+ and transduce these into changes in cytosolic Ca2+ via RyR activation. This is particularly relevant to the osteoclast that normally experiences surges in metabolic activity because of its intense motility and secretory function (23)
. The evidence is therefore persuasive for a role for CD38 in osteoclastic bone resorption through cADPr generation and RyR activation.
It is unclear, however, whether osteoblasts that also possess abundant CD38 and RyRs at both endoplasmic reticular and nuclear membrane locations respond to CD38 activation (18
, 22)
. A time-dependent increase in cell number has been noted after extracellular application of NAD+ to human osteoblasts (24)
. Nonetheless, it is uncertain whether enhanced cellular maturation and bone-forming activity accompany the increased osteoblastic proliferation induced by NAD+.
Our argument for a role of CD38 in osteoclastic bone resorption prompted us to characterize the skeletal phenotype of the CD38-/- mouse. We show significant decreases in bone mineral density (BMD) associated with increased resorptive surfaces in situ in the CD38-/- mouse, consistent with an enhanced number and function of osteoclasts. Ex vivo cultures showed evidence for increased formation of, and enhanced resorption by osteoclasts isolated from CD38-/- mice. This was associated with a reduction in the expression of IL-6 and its receptor. Finally, the expression of calcineurin, a calmodulin/Ca2+-activated phosphatase, was lower in CD38-/- osteoclasts than in wild-type cells. The functional and molecular defects in CD38-/- osteoclasts demonstrated in vivo and in vitro confirm a critical role of CD38 in bone resorption, possibly in coupling the intense metabolism of this cell with its Ca2+ signaling activity.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. CD38+/- ES clones were injected into C57BL/6J blastocysts, blastocysts were implanted into pseudopregnant B6.CBA F1 females, and the resulting chimeras were crossed to C57BL/6J mice to establish germline transmission. CD38+/- F1 mice were intercrossed and F2 breeding pairs were established from CD38+/+ and CD38-/- mice. 129Ola x C57BL/6J F3 mice were used for all experiments shown. Both homologous recombinant clones (236 and 2832) transmitted the CD38 mutation to the germline CD38-/- mice, which grew and survived normally. BMD measurements were performed at 3, 4, and 5 months of age using a Lunar Piximus with a precision of <1.5%. Mice were anesthetized by the intraperitoneal administration of chloral hydrate (380 mg/kg). After whole body BMD measurements, with the cranium excluded, a region of interest was selected manually for measurements at specific sites, notably the spine (L4-L6), right and left femur, and right and left tibia. The instrument was calibrated each time before use by a phantom per manufacturer’s recommendation.
The mice were then killed per our IACUC approved protocol. The vertebral column, femur, and tibia were cleaned of excess muscle and soft tissue and processed for histological examination. The bones were placed overnight in phosphate-buffered paraformaldehyde (5%, v/v), then decalcified with HCl, dehydrated, and embedded in paraffin. Sections (6 µm) were cut and stained with hematoxylin and eosin. In separate experiments, calcein labeling was performed to examine the extent and rate of osteoblastic bone formation. Calcein (10 mg/kg body weight) was administered intraperitoneally on days -7 and -2. The mice were killed on day 0; long bones were immersed in OCT Compound (Electron Microscopy Science, Washington, PA, USA) rapidly, then frozen in liquid nitrogen and sectioned at 6 µm using a Reichert-Jung sledge microtome. The calcein labels were examined using a 380–425 nm excitation filter and a 450 nm emission filter. Morphometry was performed using the Optimetrix System to determine the extent of labeling and distance between the two labels (interlabel distance). Mineralizing surface was corrected for total surface; other measurements were from hematoxylin and eosin decalcified stained sections by digital histomorphometry, as described. Nonstandard units were as defined by Parfitt et al. (26)
.
In vitro studies
To examine for differences in the formation and function of osteoclasts derived from CD38-/- and CD38+/+ mice, the respective bone marrow cells were first isolated as described previously (27)
. The cells were then cultured for 2 days with 5 ng/mL human macrophage colony-stimulating factor (M-CSF). Nonadherent cells were collected, purified by Ficoll-PackPlus (Amersham Pharmacia Biotech Inc., Arlington Height, IL, USA), and counted. For osteoclast formation experiments, the cells were incubated with M-CSF (30 ng/mL) and receptor activator for NF-
B ligand (RANK-L) (60 ng/mL) for 4 to 6 days, followed by staining for tartarate-resistant acid phosphatase (TRAP) using a kit (Sigma, St. Louis, MO, USA) per manufacturer’s instruction. The number of TRAP-positive cells was counted.
To examine resorption, equal counts of CD38+/+ and CD38-/- hematopoietic bone marrow cells were dispersed on slices of devitalized cortical bone substrate, then cultured with M-CSF (30 ng/mL) and RANK-L (60 ng/mL) for 6 days (28)
. For quantitation, the slices were bleached with NaOCl (5 min) before air drying and staining for toluidine blue to allow for visualization of the osteoclastic excavations (pits). Their number was determined by light microscopy and expressed as means ± SE. Student’s unpaired t test was used to compare resorption in the two groups of mice, which was considered significant at P<0.05. The latter method does not discriminate between effects on formation vs. sole effects on bone resorptive ability, but unlike the rat, it is difficult to isolate fresh osteoclasts from the mouse for overnight cultures for resorptive activity.
To examine for differences in osteoblast differentiation, we cultured bone marrow cells in the presence of ß-glycerophosphate (10 mM) and ascorbic acid (50 µM). At 3 days, multicellular fibroblastoid colonies appeared that became increasingly alkaline phosphatase positive over the following week. These are termed CFU-Fs, and their alkaline phosphatase positivity was assessed using a kit (Sigma) per manufacturer’s instructions. The number of alkaline phosphatase CFU-F colonies was counted.
We next examined for gene expression in osteoclast cultures from CD38+/+ and CD38-/- mice that had been incubated for 6 days with RANK-L and M-CSF. Total RNA was purified from the cultured osteoclasts using a StrataPrep Total RNA miniprep kit (Stratagene, La Jolla, CA, USA) per manufacturer’s protocol. Expression levels of various transcripts were determined by quantitative real-time RT-PCR using a previously described protocol. Briefly, 5 µg total RNA was converted into cDNA and 1/200 (500 pg) was used for a 40 cycle, three-step PCR in an ABI Prism 7900HT (Applied Biosystems, Foster City, CA, USA) in 20 mM Tris (pH 8.4), 50 mM KCl, 5 mM MgCl2, 200 µM dNTPs, 0.5x SYBR Green I (Molecular Probes, Eugene, OR, USA), 200 nM of each primer, and 0.5 U PlatinumTaq DNA polymerase (Invitrogen, San Diego, CA, USA). Amplicon size and reaction specificity were confirmed by agarose gel electrophoresis. Each transcript in each sample was assayed three times and the median CT values were used to calculate the fold change between experimental and control samples for each gene. Data validation by modeling reaction efficiency and analysis of precision were performed as described elsewhere.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). At 4 months, only the substantial reduction in BMD at the lumbar spine persisted (P=0.016). At 5 months, however, there were no significant differences in BMD between the CD38-/- mice and their wild-type littermates at any site. In fact, there appeared to be a trend for increased BMDs, but this was not significant statistically at any site or through total body measurements.
|
We then performed a full histological assessment on hematoxylin/eosin sections and made histomorphometric measurements of selected parameters in calcein labeled bones. Figure 2
A compares the growth plate and primary spongiosa of the CD38-/- mice (bottom) and wild-type littermates (top). A striking histological finding was that whereas the primary spongiosa extended 50 to 100 µm from the growth plate in the wild-type mouse, its thickness was decreased by 50% in the CD38-/- mouse. This finding was consistent with increased osteoclastic activity. Figure 2B
directly demonstrates the striking increase in osteoclastic activity, as evidenced by large scalloped resorption lacunae (arrows) in vertebral cancellous bone of CD38-/- mice (bottom panel) compared with wild-type littermates (upper panel). This increased resorption was also obvious in histomorphometric analyses, where an 80% increase in resorbed surface ([OC.S]) was evident, though insignificant statistically (Table 1
). Calcein labeling studies revealed no difference between CD38-/- and wild-type littermates in terms of interlabel distance ([Ir.L.Th]), mineralizing surface ([(dLS + sLS/2)/BS]) or bone area ([B.Ar]) (Table 1)
. Together, the findings suggested a significant enhancement in bone resorption but a relatively intact extent and rate of bone formation in CD38-/- mice, at least in vivo.
|
|
Figure 3
shows results from ex vivo osteoclast cultures where hematopoetic stem cells from CD38-/- and CD38+/+ mice were cultured in the presence of RANK-L and M-CSF for 6 days. There was a significant (P<0.01) fourfold increase in TRAP-positive osteoclast formation in these cultures (Fig. 3A
). Note that the morphology or size of the cells was not different between the two groups (plates). When cells were cultured on bone and allowed to resorb in the presence of RANK-L and M-CSF, CD38-/- osteoclasts showed evidence of a 2.5-fold increase in resorption. There was no obvious difference in the morphology or size of the pits: only pit number per slice was increased (Fig. 3B
).
|
We next examined for differences in osteoblast differentiation in CD38-/- mice and their wild-type littermates. Bone marrow stromal cells were allowed to differentiate in the presence of ß-glycerophosphate (10 mM) and ascorbic acid (50 µM) for 10 days. CFU-F formation, assessed using alkaline phosphatase staining, revealed a significant (P<0.05) 80% reduction in CFU-F number per well in cultures from CD38-/- mice (Fig. 4
). It appeared that although early osteoblast differentiation was impaired in ex vivo cultures, subsequent stages of differentiation were likely well compensated to defray an in vivo bone formation phenotype in the CD38-/- mouse.
|
Finally, we probed into molecular mechanisms contributing to the observed phenotype. Our in vitro studies had demonstrated that production of cytokines such as IL-6 was enhanced upon activation of CD38 by an agonist antibody (20)
. We speculated that CD38 deletion might inhibit IL-6 expression, and we therefore used real-time RT-PCR to examine the expression of both IL-6 and its receptor. We have recent evidence that calcineurin A, a Ca2+/calmodulin-activated phosphatase, inhibits osteoclastic bone resorption (29)
. We hypothesized that CD38 may regulate the expression of one or all of the three calcineurin A isoforms, , ß, and , in osteoclasts. Again, hematopoetic stem cells were cultured for 6 days with RANK-L and M-CSF. The RT-PCR data shown in Fig. 5
have been normalized to ribosomal protein S11 (control) and represent fold changes with respect to wild-type osteoclasts. We found a significant reduction in the expression of IL-6, IL-6 receptor, and the calcineurin , ß, and isoforms. Expression of one of the regulatory subunit of calcineurin, calcineurin B, was also diminished. Although we cannot derive any causal relationships, it is clear that these calcineurin A isoforms are regulated by CD38.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The high extracellular Ca2+ concentration is sensed by a Ca2+ sensor, which we believe is a RyR-like molecule resident in the osteoclast’s plasma membrane (31
32
33)
. RyRs are also located at their traditional microsomal membrane site, where they gate Ca2+ release in response to cADPr production (34)
. Cyclic ADPr is generated through the cyclization of the intermediary metabolite NAD+ by the multifunctional cyclase under study, CD38. CD38 is expressed in abundance in the osteoclast and its activation by an agonist antibody elicits Ca2+ release through RyRs (20)
. This signaling is accompanied by a reduction in bone resorption, but with an enhanced expression of IL-6; these effects are reminiscent of those elicited by a high extracellular Ca2+ (35
, 36)
. Thus, the evidence is compelling that CD38 has a role in Ca2+ signaling and bone resorption in the osteoclast.
However, unlike most other eukaryotic cells wherein Ca2+ activates cellular processes, the osteoclast is unique in that a rise in cytosolic Ca2+ is associated with an inhibition of bone resorption, possibly as a feedback mechanism for osteoclast control (31)
; thus, the activation of CD38 by its agonist antibody, which also elevates cytosolic Ca2+ reduces bone resorption (20)
. By implication, therefore, the absence of CD38 in osteoclasts should stimulate bone resorption. This was indeed the case. Furthermore, osteoblast formation was also elevated by fourfold, indicating that CD38, in addition to regulating the resorptive function of mature osteoclasts, also controls osteoclastogenesis. An enhanced formation and function of osteoclasts would together result in an overt osteoporotic phenotype with accompanying histological and histomorphometric changes, which is what we observe in vivo. However, as the animals became older the phenotype became less pronounced, with a low BMD significant only at the lumbar spine. At 5 months, none of the sites displayed reduced BMD, indicating full compensation and a trend toward osteopetrosis. The mechanism of this compensation is unclear; however, it is possible that CD157, a molecule with a similar spectrum of cyclase activity may be up-regulated to assume the function of CD38 (37
, 38)
.
We know from earlier results that IL-6 expression in the osteoclast is enhanced upon CD38 activation in vitro (20)
. Consistent with this is our observation that CD38-/- osteoclasts express significantly lower levels of IL-6 and the IL-6 receptor than corresponding wild-type littermates. Reduced IL-6 expression is seemingly paradoxical to elevated osteoclastogenesis, but is not inconsistent with the requirement of intracellular Ca2+ to enhance IL-6 expression, as we have shown before in vitro (20
, 35)
. Taken together, the findings had indeed predicted that deletion of the CD38 gene would limit (or abolish) cADPr production in the osteoclast, thus reducing RyR activation. Reduced Ca2+ release would follow; this, in turn, could potentially limit IL-6 expression, while stimulating osteoclastic resorption.
How does CD38 function in bone resorption? We speculate that CD38 is a metabolic sensor that monitors intracellular (and extracellular) levels of NAD+ generated as a result of the intense metabolic activity experienced during resorption. Glycolysis results in the generation of NAD+; the ATP formed from NAD+ is used for maintaining the activity of the H+-ATPase that is essential for acid secretion. It would thus not be unreasonable to assume that the NAD+ so produced is also a substrate for CD38 and that an inverse relationship exists between the CD38-mediated cytosolic conversion of NAD+ to cADPr (which will inhibit resorption) and the mitochondrial production of ATP (which should facilitate resorption).
Quite unexpectedly, functionally intact CD38, with its NH2-terminal 49 amino acids comprising the transmembrane and intracellular tail deleted, localizes to the mitochondria (22)
. This is in contrast to the localization of wild-type CD38: the enzyme localizes to the plasma membrane, endoplasmic reticular, and nuclear membranes. At both the latter sites, it colocalizes with RyRs (22)
. The molecular basis underlying the aberrant mitochondrial localization of the CD38 mutant is unclear. However, it is possible that the enzymatically active CD38 mutant traps NAD+ to generate cADPr within mitochondria to trigger Ca2+ release. In parallel, we have shown previously that CD38, localized to the inner nuclear membrane of osteoblasts, catalyzes NAD+ cyclization to cADPr intranuclearly and thus gates Ca2+ release through inner nuclear membrane RyRs (18
, 39)
. We had also found that when applied extracellularly, cADPr, although impermeant, elevates cytosolic Ca2+ (34)
. We have attributed this effect to CD38 functioning as a transmembrane cADPr carrier, as has been shown with other cells (40
41
42)
.
As we have established before, osteoblasts also contain abundant CD38, and we had predicted that a bone formation defect would result from CD38 gene deletion. To our surprise, however, we found no significant changes in bone formation rate on histomorphometric analysis. There was a significant decrease in osteoblast differentiation; nonetheless, it is known that a single defective step in the osteoblast maturation pathway does not necessarily alter the eventual outcome of bone formation.
In conclusion, we have shown previously that CD38 is expressed in abundance in the osteoclast and its activation results in a rise in cytosolic Ca2+, increased IL-6 expression, and inhibition of bone resorption (20)
. We now provide firm evidence that CD38 gene deletion results in enhanced osteoclast formation and function, resulting in an overt osteoporotic phenotype, but with reduced IL-6 expression. We speculate therefore that CD38 may couple the cell’s intense metabolic activity by sensing cellular NAD+ with Ca2+ signaling through its generation of cADPr. Finally, it is unlikely that CD38 has as critical a role in regulating bone formation.
|
ACKNOWLEDGMENTS |
|---|
Received for publication March 5, 2002. Accepted for publication November 22, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Inhibition of bone resorption. Am. J. Physiol. 259In press
This article has been cited by other articles:
|
|
 |
|
  F. Malavasi, S. Deaglio, A. Funaro, E. Ferrero, A. L. Horenstein, E. Ortolan, T. Vaisitti, and S. Aydin Evolution and Function of the ADP Ribosyl Cyclase/CD38 Gene Family in Physiology and Pathology Physiol Rev, July 1, 2008; 88(3): 841 - 886. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. B. Yaroslavskiy, A. C. Sharrow, A. Wells, L. J. Robinson, and H. C. Blair Necessity of inositol (1,4,5)-trisphosphate receptor 1 and {micro}-calpain in NO-induced osteoclast motility J. Cell Sci., August 15, 2007; 120(16): 2884 - 2894. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Iqbal and M. Zaidi CD38 is required for priming by TNF-{alpha}: a mechanism for extracellular coordination of cell fate Am J Physiol Renal Physiol, April 1, 2007; 292(4): F1283 - F1290. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. G. P. Guedes, J. Paulin, L. Rivero-Nava, H. Kita, F. E. Lund, and M. S. Kannan CD38-deficient mice have reduced airway hyperresponsiveness following IL-13 challenge Am J Physiol Lung Cell Mol Physiol, December 1, 2006; 291(6): L1286 - L1293. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Johnson, E. L. Ford, E. Bernal-Mizrachi, K. L. Kusser, D. S. Luciani, Z. Han, H. Tran, T. D. Randall, F. E. Lund, and K. S. Polonsky Suppressed insulin signaling and increased apoptosis in CD38-null islets. Diabetes, October 1, 2006; 55(10): 2737 - 2746. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Iqbal, K. Kumar, L. Sun, and M. Zaidi Selective upregulation of the ADP-ribosyl cyclases CD38 and CD157 by TNF but not by RANK-L reveals differences in downstream signaling Am J Physiol Renal Physiol, September 1, 2006; 291(3): F557 - F566. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Deshpande, T. A. White, A. G. P. Guedes, C. Milla, T. F. Walseth, F. E. Lund, and M. S. Kannan Altered Airway Responsiveness in CD38-Deficient Mice Am. J. Respir. Cell Mol. Biol., February 1, 2005; 32(2): 149 - 156. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
