| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online May 9, 2001 as doi:10.1096/fj.00-0594fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Biology, University of York, York, YO10 5YW, UK
2Correspondence: Department of Biology, University of York, Heslington, York, YO10 5YW, UK. E-mail: pg5{at}york.ac.uk
SPECIFIC AIMS
Previous evidence that osteoblasts and osteoclasts express diverse glutamate receptors similar to those described at glutamatergic synapses in the central nervous system (CNS) suggests that glutamate may also act as a signaling molecule in bone; however, the origin of the glutamate stimulus is unclear. Since the source of glutamate for signaling in bone is central to our understanding of this novel communication pathway, the aim of this study was to determine whether osteoblasts released glutamate actively in a manner similar to presynaptic neurons and to identify the regulatory inputs and functional significance of osteo-glutamatergic signaling.
PRINCIPAL FINDINGS
1. Localization of glutamate signaling proteins in osteoblastic
cells
In neuronal cells, glutamate release is initiated by
depolarization of the presynaptic nerve terminal with the subsequent
exocytotic fusion of glutamate-loaded vesicles with the plasma
membrane, under the direction of specific protein–protein
interactions. The released glutamate then binds to and activates
postsynaptic glutamate receptors. We have already shown that
osteoblasts express several SNARE (soluble N-ethyl maleimide-sensitive
factor attachment protein receptor) and related vesicular signaling
proteins that colocalize with immunoreactive glutamate. To identify the
spatial patterning of proteins associated with pre- and postsynaptic
glutamate signaling in osteoblast cultures, we coimmunolocalized
synapsin I, a neurospecific signaling protein expressed on presynaptic
transmitter-containing vesicles, and NMDAR1, the essential NMDA-type
glutamate receptor subunit. Both synapsin I and NMDAR1 were coexpressed
in the same cellular microenvironment in individual osteoblastic cells
within intracellular cytoplasmic pools and at the cell periphery.
Although NMDAR1 and synapsin I were expressed within the same
microenvironment at the plasma membrane, the lack of red-green
alignment by double exposure photography demonstrated little evidence
of NMDAR1/synapsin I colocalization.
2. Spontaneous glutamate release in osteoblastic cells
We developed an enzyme-linked fluorimetric glutamate assay, based
on previously published methods used for neuronal cells, to determine
exocytotic activity in a range of osteoblastic cells. This modified
assay allows for microplate analysis and high throughput screening of
multiple replicated samples, relying on the principle that in the
presence of glutamate dehydrogenase (GDH) and ß-nicotinamide adenine
dinucleotide phosphate (NADP+), released
glutamate is oxidized to 
-ketoglutarate with the fluorimetric
production of NADPH. Addition of GDH to cultures of MG-63 osteosarcoma
cells caused a rapid increase in fluorescence that was dependent on the
presence of NADP+ in the reaction buffer.
Fluorescence levels peaked after 
6 min and remained constant over
the remainder of the assay period, demonstrating spontaneously
glutamate release activity in these cells. Permeabilization of the
cells by the addition of 1% Triton X-100 caused a further rapid
increase in fluorescence corresponding to intracellular free glutamate
concentrations. Further experiments demonstrated that steady-state
levels of spontaneous glutamate release from osteoblastic cells (MG63,
SaOS-2, MC3T3-E1, primary mouse and human osteoblasts) ranged from
2.5 to 7 nmol glutamate per mg protein, equivalent to or greater
than reported levels of glutamate release from depolarized neurons.
Vesicle recycling activity was investigated in osteoblastic cells using
the styryl dye FM1–43, which becomes incorporated into the vesicular
membrane during fusion and endocytosis phases of the vesicle cycle.
Time course studies demonstrated that all osteoblastic cells examined
(MG-63, SaOS-2, MC3T3-E1) rapidly accumulated the fluorescent dye over
a 20 min incubation period. After removal of FM1–43 from the bathing
medium, the cells destained within 30 min indicating constitutive
endocytotic and exocytotic vesicle recycling, consistent with
spontaneous glutamate release activity. Osteocyte-like MLO-Y4 cells did
not release detectable levels of glutamate.
3. Effect of depolarization of glutamate release activity in
osteoblastic cells
In neurons, glutamate release is stimulated by depolarization of
the presynaptic nerve terminal. We investigated the effects of
depolarizing K+ ion concentrations on glutamate
release in osteoblastic cells. In the presence of 30 mM and 60 mM KCl,
glutamate release in MG-63 cells was significantly reduced by 60–70%
compared with glutamate levels measured in the presence of
physiological KCl concentrations (P<0.01). KCl (60 mM)
caused significant (P<0.05) inhibition of glutamate release
in SaOS-2 cells, but had no effect on exocytotic activity in MC3T3-E1
cells or MLO-Y4 cells.
The calcium dependency of glutamate release was investigated further in MG-63 cells. In the absence of extracellular Ca2+ ions, depolarization with 60 mM KCl had no effect, whereas basal levels of released glutamate in the presence of physiological KCl (3 mM), were significantly elevated (P<0.005) compared with calcium-containing controls. In the presence of Ca2+ but the absence of extracellular Mg2+ ions, the inhibitory effect of depolarization with 60 mM KCl was maintained (P<0.001). Depolarization with 60 mM KCl had no significant effect on glutamate release in the absence of both extracellular Ca2+ and Mg2+ ions whereas basal glutamate release levels were again significantly elevated (P<0.001).
4. Glutamatergic regulation of glutamate release in osteoblastic
cells
Steady-state levels of glutamate release were maintained and
achieved without depletion of intracellular free glutamate stores
measured in solubilized cells. We investigated possible glutamatergic
feedback mechanisms in osteoblastic cells by determining the effects of
glutamate receptor blockade on regulation of constitutive glutamate
release. In the presence of the NMDA channel antagonist MK-801 (100
µM), glutamate release was moderately, but significantly inhibited
(P=0.038). However, blockade of AMPA-type glutamate
receptors with the highly selective noncompetitive antagonist CFM-2
caused a significant dose-dependent inhibition of glutamate release in
MG-63 cells at 5–50 µM concentrations (P<0.005).
However, depolarization-induced inhibition of glutamate release was not
significantly affected by CFM-2.
5. Glutamate release during osteoblastic differentiation
We observed that glutamate release in primary mouse and human
osteoblasts and MC3T3-E1 cells was relatively low compared with the
osteosarcoma cell lines examined. However, these experiments were
performed after the cells had been in culture for only 24 h, at
which time they display predominantly preosteoblastic characteristics.
Glutamate release activity was therefore investigated in MC3T3-E1 cells
that were grown under osteogenic conditions for up to 8 days in culture
to promote their differentiation toward a more osteoblastic phenotype.
During this time, where a sevenfold increase in alkaline phosphatase
activity was observed, basal glutamate release increased from 2
nmol/mg protein on day 3 of culture to 5 nmol/mg protein on day 8, and
intracellular glutamate increased from 8 to 17 nmol/mg protein. In
addition, during the early stages of culture (days 3 and 5),
depolarization with 60 mM KCl had no significant effect on glutamate
release activity, whereas on days 7 and 8, 60 mM KCl induced a
significant inhibition of glutamate release (P<0.005).
Parallel cultures of MC3T3-E1 cells grown over the same time period,
but without osteogenic ascorbate or ß-glycerophosphate supplements,
showed no significant differences in basal glutamate release levels or
intracellular glutamate concentrations, and depolarization with 60 mM
KCl had no effect on glutamate release at any time point.
6. Functional role of glutamate release in osteoblasts
Riluzole is an anticonvulsant and acts by inhibiting glutamate
release at central synapses. We found that riluzole induced a
significant inhibition of glutamate release in osteoblastic cells in a
dose-dependent manner. Riluzole similarly inhibited
1,25(OH)2 vitamin
D3-induced osteocalcin secretion in MG-63 cells
at concentrations of 1–50 µM. In cultures of MC3T3-E1 cells grown
under osteogenic conditions for up to 6 days, we demonstrated that
riluzole significantly inhibited alkaline phosphatase activity in a
dose-dependent manner; however, MTT assays revealed that exposure to
riluzole at concentrations 
25 µM for longer than 24 h
compromised cell viability. Riluzole similarly reduced viable cell
numbers in cultures of MG-63 and SaOS-2 cells over 24–48 h with
significant effects observed at concentrations of 5 µM (for MG-63
cells) and 25 µM (SaOS-2). Riluzole-induced cell death was
accompanied by cell shrinkage and membrane blebbing; after 24 h
exposure to 25 µM riluzole, DAPI and TUNEL staining identified
nuclear and DNA fragmentation in MG-63 cells. Similar observations were
also made in SaOS-2 cells, MC3T3-E1 cells, and primary human
osteoblasts. Addition of exogenous glutamate (5 µM–1 mM)
significantly increased survival rates of primary human osteoblasts
cultured under glutamate/aspartate/serum-free conditions; exposure to
proinflammatory cytokines (TNF- and IFN-
), which have previously
been shown to have proapoptotic effects, significantly inhibited
glutamate release in human osteoblasts.
CONCLUSIONS
We have demonstrated that osteoblasts and osteoblast-like cells
actively release glutamate at concentrations that are sufficient to
activate receptors expressed on bone cell surfaces, providing
convincing evidence for an intrinsic osteo-glutamatergic signaling
mechanism (see Fig. 1
). The precise mechanism by which osteoblasts
secrete glutamate is not yet clear, though distinct similarities with
neuronal exocytotic pathways exist. Although glutamate may be released
by the reversal of glutamate transporters, we found that preloading
osteoblastic cells for 1 h with the glutamate transporter
inhibitor L-trans-pyrrollidine-2,4-dicarboxylic acid in
order to preferentially occupy intracellular transporter sites and
retard glutamate efflux had no significant effect on glutamate
release.
|
All the osteoblastic cells examined spontaneously released glutamate at
concentrations (2–7nmol/mg protein) equivalent to or greater than
reported levels of glutamate released from depolarized neurons and
astrocytes. We have previously shown that osteoblastic cells express
the glutamate transporter GLAST (EAAT1) and rapidly accumulate
[3H]-glutamate, providing a molecular mechanism
for intracellular glutamate accumulation. However, GLAST expression
appears to be down-regulated as osteoblasts further differentiate into
osteocytes; in sections of neonatal and adult rat bone, only recently
embedded osteocytes near to bone surfaces express GLAST, and this may
explain why osteocyte-like MLO-Y4 cells do not release detectable
levels of glutamate.
Osteoblasts including MG-63, SaOS-2, and MC3T3-E1 cells express a range of voltage-dependent calcium channels and are susceptible to depolarization by elevated extracellular K+ ion concentrations. We demonstrated that continued depolarization by 30–60 mM KCl induces a significant calcium-dependent inhibition of glutamate release, suggesting that unlike neurons, glutamate exocytosis in osteoblastic cells is negatively regulated by voltage-dependent calcium entry. During the osteoblastic differentiation of MC3T3-E1 cells over several days in culture, we identified increased levels of glutamate exocytosis, elevated intracellular free glutamate, and increased sensitivity to depolarization-induced inhibition of glutamate release, suggesting that osteoblasts attain a more active glutamatergic phenotype as they differentiate.
The glutamate release inhibitor riluzole (1–10 µM) significantly
inhibits alkaline phosphatase activity in MC3T3-E1 cultures without
affecting cell viability. However, riluzole at a concentration of 25
µM, which is neuroprotective in the CNS, induced morphological and
biochemical characteristics of apoptosis in osteoblastic cells. It has
been proposed that between 50 and 70% of osteoblasts present at active
remodeling sites eventually die by apoptosis under the influence of
locally produced cytokines. The possibility that exocytosed glutamate
also acts as an osteoblast survival factor warrants further
investigation and would explain the requirement for continual glutamate
release by osteoblastic cells. This hypothesis is supported by our
evidence that addition of exogenous glutamate (50 µM-1 mM) promotes
osteoblast survival under serum-free conditions. We were also able to
demonstrate that glutamate signaling in osteoblasts is regulated by the
cytokines TNF- and IFN-, which are known to have potent effects
on bone remodeling. These cytokines have been shown to induce apoptosis
in osteoblasts, and our evidence suggests that this may be mediated at
least in part through decreased glutamate release.
Further investigations aimed at broadening our understanding of osteo-glutamatergic signaling will advance current concepts of intercellular communication during bone remodeling and may uncover novel therapeutic targets for the treatment of bone disorders such as osteoporosis.
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.00-0594fje ; to cite this
article, use FASEB J. (May 9, 2001) 10.1096/fj.00-0594fje 
This article has been cited by other articles:
|
|
 |
|
  E. Hinoi, T. Takarada, K. Uno, M. Inoue, Y. Murafuji, and Y. Yoneda Glutamate Suppresses Osteoclastogenesis through the Cystine/Glutamate Antiporter Am. J. Pathol., April 1, 2007; 170(4): 1277 - 1290. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Wang, E. Hinoi, A. Takemori, N. Nakamichi, and Y. Yoneda Glutamate Inhibits Chondral Mineralization through Apoptotic Cell Death Mediated by Retrograde Operation of the Cystine/Glutamate Antiporter J. Biol. Chem., August 25, 2006; 281(34): 24553 - 24565. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
