reductase activity for TECR (GPSN2), and it will be interesting to determine the Langlois, V.S., Zhang, D.

s, V.S., Zhang, D., Cooke, G.M., and Trudeau,

confirming an earlier study, which had
identified fatty acids as the substrates of
this steroid 5α-reductase family member
(Figure 1B) (Moon and Horton, 2003).
The work of Cantagrel and coworkers
is creatively exhaustive and nails down
every aspect of a difficult research prob-
lem from yeast cultures to the clinic.
Where could they possibly go from here?
For one, their studies in mice with muta-
tions in Srd5a3 suggest a regulatory
crosstalk between the mevalonate path-
way and N-linked protein glycosylation,
mechanism underlying this interplay. In
addition, the substrate and physiological
function of SRD5A2L are still unknown.
Will this steroid 5α-reductase function in
glycosylation or hormone production, or
will it play a completely distinct role from
that of its namesake?
References
Cantagrel, V., Lefeber, D.J., Ng, B.G., Guan, Z.,
Silhavy, J.L., Bielas, S.L., Lehle, L., Hombauer,
H., Adamowicz, M., Swiezewska, E., et al. (2010).
Cell, this issue.
V.L. (2010). Gen. Comp. Endocrinol. 166, 489–497.
Li, J., Biswas, M.G., Chao, A., Russell, D.W., and
Chory, J. (1997). Proc. Natl. Acad. Sci. USA 94,
3554–3559.
Moon, Y.-A., and Horton, J.D. (2003). J. Biol.
Chem. 278, 7335–7343.
Russell, D.W., and Wilson, J.D. (1994). Annu. Rev.
Biochem. 63, 25–61.
Uemura, M., Tamura, K., Chung, S., Honma, S.,
Okuyama, A., Nakamura, Y., and Nakagawa, H.
(2008). Cancer Sci. 99, 81–86.
Wilson, J.D. (1978). Annu. Rev. Physiol. 40,
279–306.

No Bones About It:

Insulin Modulates Skeletal Remodeling
Clifford J. Rosen1,* and Katherine J. Motyl1
1
Maine Medical Center Research Institute, Scarborough, ME 04074, USA
*Correspondence: rosenc@mmc.org
DOI 10.1016/j.cell.2010.07.001

Advancing the hypothesis that bone remodeling is intimately linked to metabolic homeostasis,
Fulzele et al. (2010) and Ferron et al. (2010) present evidence that insulin signaling promotes
the activation of bone-forming osteoblasts and enhances production of osteocalcin, a secreted
mediator of insulin sensitivity, through modulation of bone resorption.

Bone remodeling is a highly conserved
and regulated process that controls
calcium homeostasis within a very tight
range and maintains the structural integ-
rity of the skeleton. In this issue of Cell,
tandem articles from independent labo-
ratories (Fulzele et al., 2010; Ferron et
al., 2010) provide fresh support for the
hypothesis that the insulin regulatory sys-
tem mediates communication between
metabolic control and bone remodeling.
Fulzele et al. (2010) provide compelling in
vivo evidence, using conditional deletion
of the insulin receptor in osteoblasts,
that insulin signaling promotes bone
formation by suppressing an inhibi-
tor of osteoblast development, Twist2,
and enhances expression of osteocal-
cin, a mediator of insulin sensitivity and
secretion. Ferron et al. (2010) show that
insulin signaling in osteoblasts mediates
glucose homeostasis by stimulating the
production of an inactive form of osteo-
calcin and by promoting osteoclast-
mediated bone resorption, which then
releases the active form of osteocalcin.
In this putative feed-forward loop, the
complete bone remodeling process is
involved, thereby linking skeletal homeo-
stasis to energy regulation. Given that
the findings from these papers may have
profound implications for understanding
and ultimately treating several metabolic
conditions, a closer examination of these
two reports is necessary to clarify where
and how this regulatory loop is situated
in the hierarchy of control mechanisms
that define insulin signaling and bone
turnover.
It has only been a decade since Ducy
et al. (2000) demonstrated that leptin (an
adipokine that regulates appetite, energy
expenditure, and insulin secretion) influ-
ences bone remodeling through hypo-
thalamic modulation of the sympathetic
nervous system. Prior to that, skeletal
remodeling was considered within its
essential context of calcium conserva-
tion, a homeostatic process regulated
by calciotropic hormones (such as
parathyroid hormone and vitamin D),
gonadal steroids (such as estrogen), and
local growth factors (such as insulin-like
growth factor I and transforming growth
factor β). Subsequent work has provided
stronger support for the tenet that lep-
tin is a major, albeit indirect, mediator
of skeletal turnover through neuronal
circuits in the central nervous system
(Lee et al., 2007; Yoshizawa et al., 2009).
In addition the osteoblast-specific pro-
tein osteocalcin is reported to promote
insulin secretion (Hinoi et al., 2008; Fer-

198  Cell 142, July 23, 2010 ©2010 Elsevier Inc.

ron et al., 2008). Interest- cates that these changes
ingly, osteocalcin undergoes are due to altered expres-
gamma carboxylation in a sion of osteoblast-derived
vitamin K-dependent process osteoprotegerin (OPG), a
prior to its secretion, although secreted inhibitor of osteo-
a relatively small fraction of clast maturation and activity.
osteocalcin is undercarboxy- Although osteoclasts do not
lated and is released into make osteocalcin, it has long
the circulation. It is the lat- been recognized that matrix-
ter form that is postulated to derived osteocalcin can enter
be active on adipocytes and the circulation in substantial
insulin-producing pancreatic amounts during bone resorp-
beta cells (Figure 1) (Lee et tion. The new data from Fer-
al., 2007). ron et al. clearly show that
Several lines of evidence the acidic environment of the
from the two new papers resorption pit is sufficient to
strengthen the proposal that decarboxylate, and there-
osteocalcin is indeed the fore activate, osteocalcin. In
skeletal mediator of energy support of this notion, when
balance. For example, mice RANKL (receptor activator
lacking osteocalcin have a for nuclear factor κB ligand),
metabolic phenotype nearly the most powerful endog-
identical to that of mice with Figure 1. Energy Regulation and Bone Turnover by the Insulin/­ enous stimulators of resorp-
conditional deletion of the Osteocalcin Axis tion by osteoclasts, is admin-
A putative feed-forward regulatory loop ties bone turnover to energy regulation
insulin receptor in osteo- as proposed by Ferron et al. (2010) and Fulzele et al. (2010). Insulin activates istered to wild-type mice,
blasts. In addition, Fuzele et skeletal remodeling (that is, increases bone formation by osteoblasts and re- undercarboxylated osteocalcin
al. improve insulin sensitivity sorption by osteoclasts), which in turn releases uncarboxylated osteocalcin increases 3-fold. In contrast,
from the skeletal matrix into the circulation. This enhances insulin secretion
in mice lacking expression of and increases the insulin sensitivity of adipocytes. A tyrosine phosphatase mice with nonfunctional osteo-
the insulin receptor in osteo- OST-PTP, which is encoded by the Esp gene, binds the insulin receptor (IR) clasts have reduced levels of
blasts with a 2 week infusion and suppresses its activation through dephosphorylation. The transcription undercarboxylated osteocalcin
factor Twist2 is a critical downstream suppressor of osteoblast differentiation.
of undercarboxylated osteo- Osteoprotegerin (OPG) is an osteoblast-specific inhibitor of RANKL, acting and exhibit a nearly identical
calcin. Despite this treatment, as a decoy receptor to block bone resorption. Hydroxyapatite is the mineral metabolic phenotype to mice
the mice remain glucose intol- component of bone. lacking osteocalcin. Moreover,
erant, most likely as a result reduction of osteoclast resorp-
of a decrease in beta cell mass from substrate trapping, Ferron et al. now tion with bisphosphonates rescues the
chronic insulin resistance. Determin- identify a direct interaction between the enhanced insulin secretion and glucose
ing whether chronic infusions of under- insulin receptor and OST-PTP resulting tolerance observed in OST-PTP-deficient
carboxylated osteocalcin could fully in dephosphorylation and subsequent mice.
rescue both the metabolic and obesity inactivation of the insulin receptor. In A third concern raised from earlier
phenotypes of these mice will be critical mice lacking OST-PTP, insulin sensitiv- data relates to the absence of a human
for placing this protein in the regulatory ity and secretion is markedly enhanced homolog for the Esp gene. Ferron again
hierarchy that controls insulin secretion as a result of constitutive activation of use substrate trapping with PTP1B, a
and will provide an indicator of its future the receptor. Osteoblast-specific dele- tyrosine phosphatase in humans that
therapeutic utility. tion of one allele of the insulin recep- dephosphorylates the insulin receptor,
The work by Ferron et al. rein- tor ameliorates the high levels of active and show that decreasing expression
forces the findings of Fuzele et al. and undercarboxylated osteocalcin and of PTP1B with a small interfering RNA
addresses some, but not all, of the normalizes the insulin tolerance of OST- suppresses osteoprotegerin expres-
questions raised since the Karsenty PTP deficient mice. sion in vitro. Finally the authors present
group first introduced the concept that A second issue has been the absence data from osteopetrotic patients (that is,
the skeleton could regulate energy of a clearly defined link between bone individuals with impaired bone resorp-
metabolism (Lee et al., 2007; Hinoi et turnover by insulin receptor activation tion) showing reduced serum undercar-
al., 2008). In prior work, osteocalcin has and undercarboxylated osteocalcin. boxylated osteocalcin levels compared
been shown to be regulated by the Esp Intriguingly, in the conditional insu- to controls, as well as impaired insulin
gene, which encodes osteotesticular lin receptor null mice, bone resorp- secretion after a meal.
protein tyrosine phosphatase (OST- tion is markedly reduced, whereas Despite these remarkable data, ques-
PTP). However, it has been unknown bone resorption is high in mice lacking tions remain. First and foremost, there
how OST-PTP suppresses levels of OST-PTP. A prior report by Ferron and is still no known osteocalcin receptor.
undercarboxylated osteocalcin. Using colleagues (Ferron et al., 2008) indi- Hence, the mechanism of how under-

Cell 142, July 23, 2010 ©2010 Elsevier Inc.  199

carboxylated osteocalcin increases
pancreatic insulin production and
secretion is not clear. Second, if bone
resorption is so critical to the posttrans-
lational modification of osteocalcin and
therefore insulin secretion, are individu-
als on antiresorptive therapy for osteo-
porosis more likely to be insulin resis-
tant and glucose intolerant? In a small
observational study, Kaji et al. (2009)
report that fasting glucose levels are
higher in women treated for 2 years with
alendronate, the most commonly used
treatment for osteoporosis. However, fat
mass does not differ in these patients,
and the statistical relationship between
bone density and fasting glucose dis-
appears when corrected for changes in
lean mass. In addition, previous human
investigations clearly demonstrate that
an oral glucose load that increases
insulin secretion suppresses markers of
bone resorption by 50% (Clowes et al.,
2003). Most importantly, and yet much
Calcium and Energy:
Making the Cake and Eating It too?
Douglas R. Green1,* and Ruoning Wang1
1
Department of Immunology, St. Jude Children’s Research Hospital, Memphis, TN 92105, USA
*Correspondence: douglas.green@stjude.org
DOI 10.1016/j.cell.2010.07.007
Mitochondrial calcium ions promote a number of events that sustain ATP levels in the cell. Cardenas
et al. (2010) now demonstrate that the inositol 1,4,5-triphosphate receptor at the endoplasmic
reticulum constitutively provides calcium for mitochondria. In the absence of this calcium transfer,
cells use autophagy to sustain survival.
Life is all about capturing energy and
putting it to use seeking more energy,
avoiding being someone else’s energy,
building and repairing parts, and mak-
ing more life. In eukaryotic cells, the
mitochondria are the processing plants
where the major forms of energy cur-
rency, such as ATP, are generated by
catabolism of nutrients and consump-
tion of oxygen. Much of mitochon-
drial research of the last 50 years has
200  Cell 142, July 23, 2010 ©2010 Elsevier Inc.
more difficult to address, is whether the
relationship between bone and energy
metabolism provides an evolutionary
advantage and, if so, where skeletal
control fits into the complex hierarchy
of acute and chronic regulation of insu-
lin secretion during physiologic states.
Notwithstanding these unanswered
questions, these two studies provide a
strong framework for understanding the
emerging role of the skeleton in meta-
bolic homeostasis.
Acknowledgments
This work was supported by a grant from NIAMS:
AR45433 and NIDDK-DK 84970.
References
Clowes, J.A., Allen, H.C., Prentis, D.M., Eastell, R.,
and Blumsohn, A. (2003). J. Clin. Endocrinol. Me-
tab. 88, 4867–4873.
Ducy, P., Amling, M., Takeda, S., Priemel, M., Schil-
ling, A.F., Beil, F.T., Shen, J., Vinson, C., Rueger,
focused on how this flow and con-
version of energy is controlled. But
it is only in the last few years that we
have realized that this process is an
important node in the complex signal
transduction machinery that extends
throughout the cell and connects it to
the external environment. In a sense,
we are seeking the links between infor-
mation and energy in the cell. Now, a
paper in this issue provides one such
J.M., and Karsenty, G. (2000). Cell 100, 197–207.
Ferron, M., Hinoi, E., Karsenty, G., and Ducy,
P. (2008). Proc. Natl. Acad. Sci. USA 105,
5266–5270.
Ferron, M., Wei, J., Yoshizawa, T., Del Fattore, A.,
DePinho, R.A., Teti, A., Ducy, P., and Karsenty, G.
(2010). Cell, this issue.
Fulzele, K., Riddle, R.C., DiGirolamo, D.J., Cao,
X., Wan, C., Chen, D., Faugere, M.-C., Aja, S.,
Hussain, M.A., Brüning, J.C., and Clemens, T.L.
(2010). Cell, this issue.
Hinoi, E., Gao, N., Jung, D.Y., Yadav, V., Yoshiza-
wa, T., Myers, M.G., Jr., Chua, S.C., Jr., Kim, J.K.,
Kaestner, K.H., and Karsenty, G. (2008). J. Cell
Biol. 183, 1235–1242.
Kaji, H., Hisa, I., Inoue, Y., Naito, J., Sugimoto, T.,
and Kasuga, M. (2009). J. Bone Miner. Metab. 27,
76–82.
Lee, N.K., Sowa, H., Hinoi, E., Ferron, M., Ahn,
J.D., Confavreux, C., Dacquin, R., Mee, P.J.,
McKee, M.D., Jung, D.Y., et al. (2007). Cell 130,
456–469.
Yoshizawa, T., Hinoi, E., Jung, D.Y., Kajimura,
D., Ferron, M., Seo, J., Graff, J.M., Kim, J.K.,
and Karsenty, G. (2009). J. Clin. Invest. 119,
2807–2817.
link: Cardenas et al. (2010) report that
calcium constitutively released from
the endoplasmic reticulum (ER) by
the inositol 1,4,5-triphosphate recep-
tor (IP 3R) is taken up by mitochondria
where it is required for efficient oxygen
consumption and ATP production.
When eukaryotic cells are starved of
nutrients, they engage the process of
autophagy to catabolize themselves to
survive, and this is generally triggered by