Bone Cell Mechanoresponsiveness - Skeletal Aging and Osteoporosis: Biomechanics and Mechanobiology - page 180

on medial side to 700 le on lateral side at 1 Hz for 8 min) upon [ 3 H]thymidine
(index of proliferation) and [ 3 H]proline (matrix formation) incorporation com-
pared to estrogen or load alone [ 20 ]. Extensions of this work have helped to
elucidate the role of estrogens and the estrogen receptors in mediating the adaptive
response to load (c.f., [ 2 , 19 , 27 , 50 , 51 , 67 , 75 ]).
Because osteogenic cells appear to retain a memory of their donor environment
[ 49 , 86 ], one would expect osteoblastic cells derived from post-menopausal sub-
jects to demonstrate a differential mechanoresponse compared to cells derived
from pre-menopausal subjects. While such a direct comparison has not been made,
investigators have demonstrated that estrogen augments the response to mechan-
ical load. Joldersma et al. revealed that two days of culture in 10 pM estrogen
significantly enhanced PGE 2 release from osteoblastic cells from elderly woman
(age 56-75 years, non-osteoporotic) compared to cells exposed to pulsatile fluid
flow without estrogen pre-treatment [ 43 ]. Similarly, Bakker et al. demonstrated
that 24 h of 10 pM estrogen provided a synergistic response to pulsatile fluid
flow in osteoblastic cells from osteoporotic women (mean age 82; range

62-90 years) [ 4 ].
What these studies do not show, unfortunately, are age-related decreases in a
biological response to loading that can be rescued by estrogen treatment. Nor do
they address a mechanism whereby estrogen exerts an additive effect upon bone
cell mechanoresponsiveness. Interestingly, Armstrong et al. showed that the
estrogen receptors (ER) plays an obligate role in Wnt signaling [ 3 ], as ER
antagonism with ICI 182,780 or tamoxifen prevented nuclear accumulation of
b-catenin. While not demonstrated to date, one would predict decreased
Wnt signaling in cells derived from post-menopausal subjects compared to pre-
menopausal subjects. The importance of this pathway, through both canonical and
non-canonical mechanisms, in skeletal development and adaptation to load is well-
documented [ 31 , 45 , 56 , 57 , 66 , 81 , 90 , 91 ].
Aguirre et al. have provided a model for ligand-independent ER function in
mediating mechanotransduction [ 2 ]. They observed attenuated strain-induced
ERK1/2 phosphorylation in cells derived from ERa -/- or ERb -/- mice.
Remarkably, transfection with the ligand-binding domain of either receptor
restored the ability of these cells to increase ERK1/2 phosphorylation in response
to strain, as did transfection with an ERa mutant that is unable to bind estrogens.
The role of ERa and ERb in mediating these processes appears to involve non-
genomic signaling, based upon the rapid time of response (10 min) and because
mechanoresponsiveness was restored using a plasmalemma-targeted, but not
nuclear-targeted, ER. These findings would suggest, therefore, that bone cell
mechanoresponsiveness would not be affected by estrogen status, if ER expression
does not change with aging. However, there is evidence that ER expression also
decreases with age [ 7 , 11 , 12 , 34 ], perhaps reducing the potential for mechano-
transduction via any estrogen-independent ER mechanism.