Research Article: Mechanisms of Gravitational Sensitivity of Osteogenic Precursor Cells

Date Published: April , 2010

Publisher: A.I. Gordeyev

Author(s): L.B. Buravkova, P.M. Gershovich, J.G. Gershovich, A.I. Grigorʹev.



This report is a detailed review of the current data on the mechanic and gravitational
sensitivity of osteoblasts and osteogenic precursor cells in vitro. It
summarizes the numerous responses of cells with an osteoblastic phenotype and osteogenic
precursor cells and especially their responses to the alteration of their mechanic or
gravitational surroundings. The review also discusses the osteogenic cell’s pathways of
signal transduction and the mechanisms of gravitational sensitivity. It was shown that the
earliest multipotent stromal precursor cells of an adult organism’s bone marrow can sense
changes of intensity in a gravitational or mechanic field in model conditions, which may play a
certain role in the development of osteopenia in microgravity.

Partial Text

During the evolution, the skeletal system of land vertebrates adapted itself to an environment
in which one of the most prominent and constant factors is gravity. This factor has determined
the morphogenesis and structure of all land animals. Certain elements of the skeleton have
evolved for maintaining posture and achieving active locomotion and thus are constantly
experiencing static and dynamic strain as a result of “defying the gravitational
force.” Since humans have started exploring the outer space, the effect of microgravity
on the skeletal system has become an important issue, as a lack of mechanic stress
(microgravity, hypokinesia, hypodynamia, immobilization) can lead to the loss of bone mass
caused by insufficient mechanic impulses and gravity–induced deformations, which are not
capable of supporting the integrity of the skeletal remodeling processes [1, 2].

Comparison of the results obtained in in vitro experiments, with the
changes that take place in a human organism under the influence of microgravity, provides an
opportunity to differentiate and establish the role of cellular reactions in forming
physiological responses, since it allows to factor out the effects of the integral regulating
systems of the human organism. The development of the views on cellular gravitational
sensitivity per se can be seen in a series of reports [16–20]. Discussions of whether an
in vitro single cell or a cell population can sense changes in the
gravitational field are still very heated. Despite this, an enormous body of experimental data
undoubtedly indicates that several types of cultured cells are sensitive to gravity. In
particular, it was demonstrated that microgravity causes multiple and often reversible
morpho–functional alterations, including remodeling of the cytoskeleton, change of gene
expression and a mosaic rearrangement of the intracellular regulatory machinery. These
alterations are reviewed in detail in [5, 19, 21, 22].

For a long time, osteocytes and the mature inactive osteoblasts were widely accepted to be the
most likely candidates for a mechanosensor in the bone tissue [14, 15]. It was supposed that this
process was performed via cell–cell junctions, formed by integrins, which interact with
elements of the actin cytoskeleton (actin, vinculin, etc.) inside the cell and with various
proteins of the bone matrix outside the cell, thus forming a continuous network which
encompasses osteocytes and the bone matrix. It was thought that this ever–present and
all–encompassing structure could sense and potentiate the effect of even miniscule
mechanical stimuli [26].

The question of whether the immunophenotype of precursor cells remains intact under conditions
of altered gravity may be of much importance for several reasons. First, the main
CD–clusters, which are expressed on the MMSC membrane, regulate various aspects of
precursor cell functioning. Since they are surface receptors for growth factors and thus
mediate the interactions between MMSC and hemopoetic precursors and lymphocytes, they modulate
the maturation and activity of the latter and take part in the interaction of cells with
molecules of the extracellular matrix [11, 13, 44]. Second, the role of some antigens in the
realization of unique stem cell differentiation potentials is still unknown. Instances of the
effect of model microgravity on the expression of specific MMSC surface markers are rare and
controversial. Specifically, one study determined that a 7–day incubation in a 3D–
с linostat caused an increase of the population ratio of human MMSC cells expressing
stromal cell antigens CD44+, CD90+, CD29+ [39]. Another
study showed that a 6–day incubation in a horizontal с linostat decreased the
number of cells bearing the CD105 and HLA A,B,C antigens in a culture of human bone marrow MMSC
[45]. Our own studies show that a 5–day incubation
of MMSC on a RPM causes an increase in the number of cells expressing integrin CD49b, but does
not affect the percentage of cells, expressing CD29 [46].

Studies that focus on the various parameters of collagen biosynthesis of the so–called
mechanocytes (fore mostly fibroblasts and bone cells) under conditions of elevated or decreased
gravity are of especial interest. Hypergravity usually results in increased type I collagen
biosynthesis [54], while microgravity or their modeling
suppress the expression of this protein [4, 55]. Our study found that MMSC, which were committed to
osteogenesis simultaneously with the transfer of cells into simulated microgravity, displayed a
decrease in the production rate of extracellular collagen matrix (type I collagen) [56].

Runx2/PEBP2aA/Cbfa1, the main regulator of mesenchymal cell osteogenic differentiation, which
can respond to the effect of osteogenic growth factors, was first identified in the course of
studies connected with osteogenic differentiation of pluripotent mesenchymal
precursor–cells of the C2C12 mouse line [62].
Full–fledged osteoblast differentiation and expression of specific osteogenic genes
requires the cooperation of Runx2 and Smad molecules, which are activated by BMP–2. It
was also discovered that growth factor BMP–7 induced the expression of Runx2 mRNA
earliear than osteocalcin expression, and furthermore, transfection by an isoform of
Runx2 led to the osteogenic differentiation of non–osteogenic cells [63].

The organism possess a highly surprising connection between osteogenesis and adipogenesis,
which is preserved in the cultured precursor cells as well. Probably, these unusual reciprocal
interactions between the two differentiation lineages of MMSC are determined by shared
signaling pathways and regulating mechanisms, which prioritize the development of one
differentiation path at the expense the other one, basing this choice on the signals received
by the cells. At least some of these mechanisms have recently been elucidated.

The reciprocal suppression of the two differentiation pathways of MMSC may be attributed to
the existence of other regulatory mechanisms, including those of autocrine and paracrine
nature. For instance, the products of one of the differentiation pathways may inhibit the
production of compunds which are needed for the formation of the other phenotype. Studies have
shown that the lipoprotein lipase produced by adipocytes could bind sortilin, the expression of
which was induced during osteogenic differentiation of MMSC, since this receptor protein was
needed for the normal mineralization of the bone matrix. Moreover, sortilin itself was able to
mediate the endocytosis of lipoprotein lipase [85]. It
has also been shown that the increase in adipogenic differentiation of MMSC’s obtained
from osteoporosis patients was caused by an abnormal response of the cells to the leptin
cytokine, which usually suppressed PPAR γ by phosphorylation [86].

Recently, there are more and more observations giving strength to the idea that cytoskeletal
structures and cell surface receptors connected to them play an imporatant role in the
regulation of the differentiation potential of stem cells, which is affected by signals from an
“external mechanical field” (Fig. 3). Also,
changes of shape and of the inner cytoskeletal architecture are common cell responses under
conditions of real [22] or simulated microgravity [26,
46, 92]. It has been determined that changes in the morphological characteristics of cells, or
modulation of the Rho family proteins activity (GTPases that regulate actin cytoskeleton) can
lead to the modification of the differentiation potential of MMSCs. For instance, activation of
Rho–kinase (ROCK) by the upstream RhoA GTPase can induce the myogenic MMSC
differentiation pathway and inhibit the adipogenic pathway even in the presence of the
insulin–like IGF–I factor [93]. It is
proposed that cell shape can act as a mechanical stimulus and plays an important role in the
determination of the differentiation pathway of the precursor cells. It was shown that well
spread cells were inclined to differentiate down the osteogenic pathway, while round unspread
cells tended to take the adipogenic fate. Expression of the dominant negative RhoA
caused differentiation into adipocytes, while overexpression of the wild–type
gene led to osteogenesis. The authors found that normal actin–myosin tension was required
for the correct activation of Rho–kinases by RhoA and suggested that the cytoskeleton and
the regulatory proteins coupled to it could act as an integral regulatory system that
controlled cell differentiation decisions, which were mainly defined through mechanical signals
[94]. Interestingly, cultivation of MMSC under simulated
microgravity caused changes in the actin cytoskeleton, up to a complete absence of filamentous
actin in the cell after a 7–day incubation. Another effect was a strong drop in the
activity of RhoA–kinase. Moreover, transfection of the cells by a viral vector, which
expressed a constitutively active RhoA , prevented the described
cytoskeleton alterations and neutralized the development of adipogenic features in the cells
[92]. Direct interaction between ERK1/2MAPK
with the integrin–mediated signaling pathway and also with the activity of several
cytoskeletal effector proteins was demonstrated by switching–off of one of the actin
cytoskeleton remodeling proteins (Rho), which caused the inactivation of the MAP–kinase
cascade [95].