View Full Version : Bone Fluid Flow Workshop Summary

Susannah Fritton
09-28-1997, 08:02 AM
Summary of a Workshop on Bone Fluid Flow

A Bone Fluid Flow Workshop with the objective of summarizing the
state of the research on bone fluid flow and its role in the bone
tissue mechanosensory system was held on September 8th, 1997 at the
City College of the City University of New York (CCNY). The workshop
was sponsored by City College's Center for Biomedical Engineering and
was organized by Steve Cowin, Shelly Weinbaum, and Susannah Fritton.
More than 50 people attended the workshop, and the speakers were Shelly
Weinbaum (Center for Biomedical Engineering, CCNY), Yi-Xian Qin
(Musculo-Skeletal Research Lab, SUNY Stony Brook), Steve Doty
(Analytical Microscopy Core, Hospital for Special Surgery), Melissa
Knothe Tate (University and Swiss Federal Institute of
Technology-Zurich, AO Research Institute-Davos), Eric Nauman
(Orthopaedic Biomechanics Lab, UC Berkeley), Dajun Zhang (Center for
Biomedical Engineering, CCNY), P.J. Kelly (Orthopedics, Mayo Clinic,
Retired), Todd McAllister (Bioengineering, UCSD), Jenneke Klein-Nulend
(Dept. of Oral Cell Biology, Free University, Amsterdam), Sol Pollack
(Bioengineering, Penn) and Elisabeth Burger (Dept. of Oral Cell
Biology, Free University, Amsterdam). The contents of the presentations
of the eleven speakers are summarized in the following eleven

The first technical presentation was an Overview of Bone Fluid
Flow by Shelly Weinbaum from CCNY. The overview addressed fluid movement
in both the lacunar-canalicular porosity and blood flow in bone. A
review of bone blood flow was accompanied by the suggestion that the
periosteum acts as a high pressure membrane maintaining the mean pore
pressure in bone at about 50 mm of Hg. The possibility of fluid pores
within the mineralized matrix was examined and the arguments for and
against the mineralized matrix or the lacunar-canalicular porosity being
the site of the strain generated potential were presented. The role and
characterization of the gel-like matrix structure in the fluid annulus
surrounding the osteocytic processes was discussed and its importance
in the electromechanical coupling between the fluid flow in the annulus
and intracellular currents explored. The role of this matrix in
modulating the fluid shear stresses on osteocytes induced by mechanical
loading was analyzed and recent experiments investigating the
intracellular chemical response of bone cells to shear stress in tissue
culture were summarized.

Yi-Xian Qin, presently a post-doc from the Musculo-Skeletal
Research Laboratory at SUNY Stony Brook, spoke on The Interdependence
of Intracortical Fluid Flow and Loading Frequency, and Their Regulatory
Role in Bone Adaptation. The speaker noted that there exists an
increasing body of analytical and experimental evidence which
demonstrates that fluid flow may be an important mediator of bone cell
activity. He also observed that perturbation of this flow, via changes
in functional activity, may ultimately prove the key influence in the
plasticity of the skeleton. Should fluid flow control bone cell
activity, then loading regimens which are osteogenic should generate a
distribution of intracortical fluid pressure which correlates to the
remodeling response. In the experiments reported by Dr. Qin the
potential role of fluid flow in adaptive responses in bone was
investigated through the use of an in vivo animal model in which the
mechanical loading environment can be controlled, the adaptive response
quantified, and the fluid flow estimated through the resultant streaming
potentials and numerical modeling. The influence of surface leakage
boundary conditions was discussed and evaluated in the context of
analytical one-dimensional models. A strong correlation was found
between transcortical fluid flow and streaming potentials in two
distinct loading cases, with a strong dependence on the loading
frequency. The results suggested that intracortical fluid flow is a
product of both matrix strain gradients and intramedullary pressure, the
latter arising primarily through volumetric changes in the marrow
cavity. These two distinct sources for remodeling stimuli may explain,
at least in part, the differential remodeling responses observed as a
function of strain frequency.

Steve Doty from the Hospital for Special Surgery in New York
gave a talk entitled Morphologic and Tracer Studies of the Flow
Through Bone. Dr. Doty reported on a morphologic approach to
understanding flow through bone that was executed by using in-vivo
tracers of different size and chemistry. Vascular transport of
microperoxidase, native ferritin and tetracycline was described as they
flow from the blood vessels to surrounding osteocytes within compact
bone. He also considered the escape of these tracers from the vascular-
osteocyte system into the surrounding bone matrix. A comparison was
made between tracer flow through lamellar bone (rat) and Haversian
systems (rabbit). This information is being collected to aid in the
description of the flow path in bone and the size of "pores" in this
path which might regulate the flow rate. Dr. Doty also reported on a 3-
dimensional morphologic analysis whose objective was to describe the
vascular and osteocyte relationships and the osteocyte and their
canalicular systems. Methacrylate or resin casts of these structures
were made from compact bone and following polymerization, the bone is
etched away from the plastic to provide a 3-dimensional structure.
Scanning electron micrographs are being taken and relevant dimensions
and sizes of the cell structures will be collected. These electron
micrographs were considered by the audience to be quite striking and
revealing, although the results were consistent with the present state
of knowledge of these nano-scale structures.

Melissa Knothe Tate from Zurich and Davos reported on her computational
modelling and animal experiments in a presentation entitled Measurement
of Load-Induced Fluid Flow as a Function of Mechanical Loading
Parameters. She began with the observation that, due to the
inaccessibility of the minute spaces through which load-induced fluid
flow is believed to occur, it is inherently difficult to measure
directly load-induced fluid displacements using experimental methods.
She then reported on theoretical and newly developed experimental models
designed to elucidate interstitial fluid flow and its role in processes
associated with growth and functional adaptation. Data from her mass
transport and fluid displacement finite element models, ex vivo model of
the sheep forelimb, in vitro model of small, cylindrical compact bone
specimens and in vivo model of the rat tibia show significant
enhancement of molecular transport resulting from mechanical loading of
the poroelastic, fluid-filled bone tissue. In addition, observed
patterns of fluid flow and tracer transport were delineated as a
function of mechanical parameters (e.g. strain magnitude, number of
cycles, strain rate) as well as tracer molecule size. Finally, she
observed that, if specific tracer distributions caused by deformation
induced fluid flow can be related to cellular activity associated with
remodeling processes, the mechanisms for functional adaptation within
the context of Wolff's Law would have to be expanded to include effects
of load-induced fluid flow.

Eric Nauman, presently a graduate student working with Tony
Keaveny at Berkeley, presented their joint work: The Dependence of
Inter-Trabecular Permeability on Volume Fraction and Trabecular
Orientation. This research concerns the pressure gradients in the marrow
and interstitial fluid created by the mechanical loading of trabecular
bone which deforms the trabecular matrix. The resulting fluid flow
exerts shear stresses on the bone lining cells and may stimulate
remodeling. In addition, fluid flow plays an important role in the
integration of bone grafts and the hydraulic stiffening of trabecular
bone during impact loading. A fundamental parameter for characterizing
the flow through the inter-trabecular pores is the permeability. The
wide range of experimental values in the literature indicates that this
aspect of trabecular fluid flow is not well understood. Thus, there is a
need for development of a theoretical model that can be used to
interpret existing data. This model could also be used with poroelastic
models of trabecular bone to determine the physiological range of fluid
shear stresses exerted on the bone lining cells. The goal of this work
is to develop a simple cellular solid model that describes the
dependence of inter-trabecular permeability on volume fraction and
orientation. The model will then be validated by comparison with
experimentally obtained inter-trabecular permeabilities for a range of
anatomic sites.

Dajun Zhang, presently a post-doc in the Center for Biomedical
Engineering at CCNY, gave a talk entitled Modeling Electrical Signal
Transmission in the Bone Cell Network. Dr. Zhang noted that it is now
generally accepted that the weak strain generated electrical potentials
(SGPs) in wet bone are dominantly caused by the streaming potentials
established by strain-induced bone fluid flow. He reported his
calculation of the intracellular potential and current induced by the
load-driven streaming potentials within a representative osteocytic
process along the radius of a typical osteon. The streaming potential is
derived based on poroelasticity theory and electrokinetic theory and the
intracellular electrical response is evaluated through the cable theory.
Particularly, his results demonstrated that the SGP-induced variations
in the transmembrane potential at bone lining cells located along the
wall of the Haversian canal behave as a high-pass, low-pass filter with
respect to loading frequency. This strong frequency selectivity suggests
that intermediate-frequency (15 - 30 Hz), low-amplitude mechanical
loading, such as those contributed by muscle tone, may be important to
bone maintenance and remodeling.

Pat Kelly, Emeritus Professor of Orthopedics at the Mayo
Clinic, spoke on the topic Fluid Flow and Bone Formation. Dr. Kelly
described a venous tourniquet model in the canine that was employed to
study fluid flow, bone formation and pressure effects across the
capillary barrier (Kelly et al., Clinical Orthopedics and Related
Research, Vol. 254, 1990). The channels for fluid flow were
demonstrated in the presentation. Experiments on weight bearing and
non weight bearing canine tibiae show that less bone appears in non
weight bearing tibial defects than in weight bearing tibial defects.
Studies in the same model reveal that the interstitial fluid space (ISF)
is less on the non weight bearing side than on the weight bearing side.
The hypothesis offered is that less function results in decreased bone
formation because of a decrease in capillary filtration and a decrease
in perfusion of the osteoblast with important solutes that are needed
for osteoblastic activity; alternative explanations were offered.

Todd McAllister, presently a graduate student working with
John Frangos in Bioengineering at UCSD, presented their joint work:
Characteristics of Flow-Induced Nitric Oxide Release in Osteoblasts. In
their background remarks the authors noted that transcortical
interstitial fluid flow has been shown to be a potent stimulus for
osteogenic autocrine/paracrine factors. Previously the authors have
demonstrated that fluid flow-induced shear stress stimulates nitric
oxide (NO) and prostaglandin E2 (PGE2) release in cultured osteoblasts.
The purpose of the current study was to identify the role of calcium and
G-proteins in this flow-mediated signal transduction. Flow-induced NO
release in osteoblasts demonstrated a biphasic response, with an initial
burst (8.2 nmols/mg/hr) followed by a steady and sustained production
(2.2 nmols/mg/hr). Treatment with GDPbS (900 uM) or quin 2/AM (30uM)
inhibits this initial response, but does not significantly attenuate
sustained production. G-protein activation with GTPgS (300-900 uM)
stimulated a dose dependent and sustained release. Calcium ionophore
(1uM) stimulated an initial burst, but no sustained production. Taken
together, these data suggest that flow-induced NO production in
osteoblasts is regulated by two distinct mechanisms. Transients in
shear activate a G-protein and calcium dependent pathway, while steady
flow activates a calcium independent pathway.

Jenneke Klein-Nulend from the Dental School ACTA-Vrije Universiteit in
Amsterdam spoke on the topic Osteocyte Mechanosensitivity and
Prostaglandins. As a background and motivation, Dr. Klein-Nulend
observed that mechanical stress produces flow of interstitial fluid in
the bone lacunar-canalicular network along the surface of osteocytes and
lining cells. This flow has been postulated to provide the physiological
signal for bone cell adaptive responses in vivo. Over the last few
years, Dr. Klein-Nulend and colleagues have examined this theory
experimentally, using among others cell cultures of isolated osteocytes
from embryonic chicken calvariae. Their work has shown that bone cells,
in particular osteocytes, are extremely sensitive to mechanical stress,
and that osteocytes are much more sensitive to fluid flow than to
hydrostatic compression. The response of bone cells in culture of fluid
flow includes prostaglandin synthesis and induction of expression of
prostaglandin G/H synthase (PGHS-2 or inducible cyclooxygenase, COX-2),
an enzyme that mediates the induction of bone formation by mechanical
loading in vivo. This response was also observed in the presence of very
low serum concentration (0.1%). Disruption of the actin-cytoskeleton
abolishes the response to stress, suggesting that the cytoskeleton is
involved in cellular mechanotransduction. The data reported support the
hypothesis that stress on bone causes fluid flow in the
lacunar-canalicular system, which stimulates osteocytes to produce
prostaglandins that induce an osteogenic response.

Sol Pollack from Bioengineering at Penn gave a talk entitled
Fluid Flow Effects on Osteoblast Intracellular Calcium Concentration.
Professor Pollack began by noting that investigations of cellular
interactions with their local physical environment aim to elucidate the
mechanism by which physical forces are transduced into cytostolic and
nuclear events that ultimately determine the state and function of the
cell. Efforts in his laboratory have detailed a distinct dose-response
interaction involving fluid flow induced shear stress amplitude and an
increase in intracellular calcium concentration, [Ca2+], in primary
cultured osteoblast-like cells. The amplitude of the calcium response
was significantly increased by the presence of serum. By the use of
appropriate blockers they have identified that it is the inositol
phospholipid pathway that leads to the intracellular calcium
mobilization from the endoplasmic reticulum. However in the presence of
serum the flow transduction is dependent on pertussis toxin sensitive
G-proteins while the serum free transduction is not. Furthermore, the
amplitude of the calcium response in the absence of serum is reduced by
passing the primary cells, is non-existent in cloned and transformed
osteoblasts, is blocked by Gadolinium suggesting the involvement of
stretch receptors and is significantly reduced when calcium is
eliminated from the perfusate. Combined with the observed increase in
the calcium amplitude with serum concentration, Professor Pollack labels
the serum free mechanism "mechano-transduction" and the mechanism with
serum as a "mass transport" mechanism. Arguments for both were

Elisabeth Burger from the Dental School ACTA-Vrije Universiteit in
Amsterdam spoke on the topic Bone Cell Mechanosensitivity and
Osteoporosis. Professor Burger related the day's discussions to clinical
problems. She noted that recent studies address the issue of bone cell
mechanosensitivity in relation to the emerging awareness that
mechanical disuse may be an important determinant of bone weakness
as in osteoporosis. It is known that bone metabolism is changed in
osteoporotic (OP) patients but a relationship with abnormal
mechanosensitivity of bone tissue is unknown. As a first step to test
the hypothesis that a low bone cell mechanosensitivity may predispose an
individual for osteoporosis in later life, she compared the in vitro
response to stress of bone cells from OP patients with cells from age-
matched controls. Primary bone cell cultures from iliac bone biopsies of
9 OP patients (3 male, 6 females, 47-72y) and 6 controls (4 males, 2
females, 44-77y) were mechanically stressed for 1 h by pulsating fluid
flow (PFF, 0.7*0.03 Pa at 5 Hz, peak stress rate 12 Pa/sec). Both OP
and CO cells increased their release of prostaglandin E2 (PGE2) and
nitric oxide (NO) during 1 h PFF-treatment, in agreement with earlier
findings in mouse and chicken bone cells. However, at 24 h after 1 h
PFF treatment, the release of PGE2 was still enhanced by more than
two-fold in the CO cell cultures, but not in the OP cultures. As PGE2
is likely involved in the transduction of mechanical signals, these data
suggest that the long-term response of osteoporotic bone to mechanical
stress may be changed. She speculated that a disease-related abnormality
in the mechanosensitivity of bone cells may be involved in the
pathogenesis of osteoporosis.

Steve Cowin and Susannah Fritton, September 25th, 1997.