The concept of electrical stimulation to elicit fracture-healing has
a long history, dating from 1812, when direct current was used to
elicit the healing of a nonunion of a fracture1.
Authors of other early reports have also described encouraging results
with galvanopuncture (galvanic stimulation delivered through insulated
needles) for the treatment of pseudarthrosis2,3.
Despite many successes, however, the technology disappeared from
mainstream medical research by the end of the nineteenth century
because claims regarding its efficacy had remained unsubstantiated.
Fukuda and Yasuda4 rekindled
interest in electrically induced bone growth in 1957 with a description
of electrical fields generated by mechanical stress on bone. They
suggested that stress on the crystalline components of bone produced
a current flow that triggers healing processes. Yasuda5 demonstrated that electrical signals
similar to those generated by mechanical stress could enhance fracture-healing.
These reports encouraged both laboratory and clinical research on electrically
induced bone formation and healing with use of various forms of
electrical stimulation6. The clinical
effectiveness of bone-growth stimulation proved to be easier to
demonstrate than did the mechanism or mechanisms of action of electrically
induced osteogenesis. Studies on the treatment of nonunion with
direct current7-9, inductive coupling10-13, and capacitive coupling14,15 as well as the treatment of spinal
fusions with inductive coupling16 and
capacitive coupling17 largely
predated information on the mechanism or mechanisms of action.
The current study was undertaken to determine the biochemical
pathways that are activated in signal transduction when various
types of electricity are applied to bone cells. The types of electrical
stimulation used were capacitive coupling, inductive coupling, and
combined electromagnetic fields. The hypothesis tested was that
if the initial transduction site during capacitive-coupling electrical
stimulation is at or within the cell membrane and the initial transduction
site during inductive-coupling or combined-electromagnetic-fields
electrical stimulation is intracellular, then the dose-response
of capacitive coupling as well as the signal transduction and biochemical
pathways activated by capacitive coupling will be different from
that of the other two signals.
Part A: Dose-Response of Three Signals (Capacitive
Coupling, Inductive Coupling, and Combined Electromagnetic Fields)
Cell Culture
MC3T3-E1 osteoblastic cells18 from
mice were cultured in 150-mm culture flasks in Dulbecco modified
Eagle medium (D-MEM; Life Technologies, Grand Island, New York)
supplemented with 10% newborn calf serum in 5% humidified
carbon dioxide at 37°C. Media were changed every three days. Prior
to confluence, the cells were subcultured and plated at a density
of 50,000 cells/cm2 either onto
35-mm tissue-culture dishes (Corning Glass Works, Corning, New York)
for use in both the combined-electromagnetic-fields and inductive-coupling
experiments or onto specially modified Cooper dishes (Falcon, Oxnard,
California) for use in the capacitive-coupling experiments. The cells
were grown until two days postconfluence, with the media changed
just prior to the beginning of the experimental treatments.
Capacitive Coupling
Capacitive-coupling electrical stimulation was performed as previously
described19-21. Bone cells were
plated in monolayer on the bottom of modified Cooper dishes, each
fitted with glass coverslips on the top and bottom to which stainless-steel
electrodes were attached, as previously described21.
The electrodes were connected to a custom-built function generator
with a blocking capacitor in the circuit and then to a power amplifier
(model XL-500; Hafler, Tempe, Arizona). The experimental cell cultures
throughout these studies were subjected to a 60-Hz sine-wave signal
with an output of 44.81 V peak to peak. This produced a calculated
electrical field strength in the culture medium of 2.0 V/m
with a current density of 300 A/cm2,20.
Control cell-culture dishes were identical to the stimulated dishes
except that the electrodes were not connected to the generator in
the case of capacitive coupling and the electromagnetic fields were
not turned on in the case of inductive coupling or combined electromagnetic
fields.
Inductive Coupling
Inductive-coupling stimulation was generated by a commercial
unit (EBI, Parsippany, New Jersey) that is used clinically. There
was an arched copper coil over the top of each unit. A 17.8 ¥ 20.2-cm
Plexiglas stage held six 35-mm tissue-culture dishes within the
unit. The stage was placed in the bottom of the unit at a distance
that is typical during clinical use. The pulsed electromagnetic
or inductively coupled field consisted of a 15-Hz burst of pulses
with twenty pulses per burst and a pulse frequency of 4.3 kHz. The
maximum value of the magnetic field amplitude generated at the culture
dish was 22.5 ± 2.5 G, and the electrical field
amplitude induced in the culture dish at 1.0 cm from the center
of the dish was 0.16 ± 0.02 V/m.
Combined Electromagnetic Fields
Combined electromagnetic fields were generated by a commercial
bone-growth stimulator (OL 1000; OrthoLogic, Phoenix, Arizona) that
is used clinically. The unit uses a pair of copper-coil transducers
to generate colinear static and time-varying magnetic fields. A
15 ¥ 17.5-cm Plexiglas stage held ten 35-mm tissue-culture
dishes within the unit. The stage was placed in the middle of the
unit, 9 cm from the top and bottom, a distance from the coils that
is typical during clinical use. The combined-electromagnetic-fields
condition is made up of a static or direct-current magnetic field
combined with a colinear alternating-current sine-wave electromagnetic
field. The measured magnetic field strengths used in these experiments
were 340 ± 140 mG for the static field and 370 ±
47 mG for the alternating-current sine-wave time-varying magnetic
field. The measured frequency of the sine wave was 76.6 Hz. The
value of the electric field amplitude induced in each culture dish
at a distance of 1.0 cm from the center of the dish was calculated
to be 0.89 ± 0.16 ¥ 104 V/m.
Experimental Design
For all three forms of stimulation, the experimental unit was designed
to allow for thirty minutes, two hours, six hours, or twenty-four
hours of stimulation. The units of each experiment were separated
in matched incubators and placed in the same position and orientation
in each of the incubators. A dual-channel thermometer (Fisher Scientific,
Pittsburgh, Pennsylvania) was used to measure temperature within
the sample dishes for each unit of each experiment. Preliminary experiments
revealed that the temperature in the culture dishes during stimulation
with combined electromagnetic fields and with inductive coupling,
but not with capacitive coupling, was approximately 0.2°C warmer
than that in the control dishes at thirty minutes. Accordingly,
during the experiment proper, the temperatures of the incubators
containing cells stimulated with combined electromagnetic fields or
inductive coupling were adjusted to 36.8°C during the entire treatment
period to compensate for the 0.2°C heating of the media by the coils.
The cells were stimulated for thirty minutes and for two, six, and
twenty-four hours and were harvested twenty-four hours after the
beginning of stimulation. For example, cells stimulated for thirty
minutes remained unstimulated in culture for another 23.5 hours.
Control dishes were incubated in the control unit for the same time-period
as the experimental dishes. Twenty-four hours after the beginning
of stimulation, the cells from the experimental and control groups
were harvested, by scraping, into phosphate-buffered saline solution
and were used for measuring total DNA content as an index of proliferation22. Ten dishes were used at each time-period
for each run with combined electromagnetic fields, and six dishes
were used at each time-period for each run with inductive coupling
and for each run with capacitive coupling. Each run was repeated three
or four times with each method of stimulation for the thirty-minute
and twenty-four-hour time-periods, and each run was repeated two,
three, or four times with each method of stimulation for the two
and six-hour time-periods.
Because the Cooper dishes were of a different size, shape, and material
(plastic and glass rather than plastic alone) than the 35-mm tissue-culture
dishes, an additional experiment was done to make sure that any
differences noted between capacitive coupling and inductive coupling
or combined electromagnetic fields were real. In this experiment,
all cultures, experimental and control, for assessment of all three
signals were performed in Cooper dishes. The cells were grown until two
days postconfluence as described above, and then all cultures were
stimulated for twenty-four hours with capacitive coupling, inductive
coupling, or combined electromagnetic fields as described above.
At the end of the twenty-four-hour stimulation period, the cells
were harvested and the DNA content per dish was determined as described
above. The experiment was run three times with a total of eighteen
dishes for capacitive coupling and twelve each for inductive coupling and
combined electromagnetic fields.
Part B: Transduction of the Three Signals
Signal Transduction Inhibitors
Six signal transduction inhibitors were used: verapamil (Sigma,
St. Louis, Missouri), which blocks voltage-gated calcium channels
in the cell membrane23; neomycin
(Pharma-Tek, Huntington, New York), which blocks the inositol phosphate
pathway in the cell membrane by inhibiting phospholipase C-meditated
hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2)24; bromophenacyl bromide (Sigma),
which inhibits phospholipase A in the cell membrane25; TMB-8 (Sigma), which inhibits Ca2+ release
from intracellular stores26; indomethacin
(Sigma), which inhibits prostaglandin synthesis in the cell membrane27; or N-(6-aminohexyl)-5-chloro-1-naphthalenesulfonamide hydrochloride
(W-7; Sigma), a calmodulin antagonist28.
The concentrations of the inhibitors used in this study, which were
the same as those used in previous studies, were 20 M of verapamil,
10 M of neomycin, 15 M of bromophenacyl bromide, 125 M of TMB-8,
4 g/mL of indomethacin, and 1 M of W-729,30.
At these concentrations, the inhibitors had no effect on cell proliferation
in control, nonstimulated cells. The total DNA content was determined
in the absence and presence of the specific intracellular inhibitors
in bone cells stimulated by the three signals.
Experimental Design
The bone cells were grown in the media and under the environmental
conditions described above until two days postconfluence. At that
time, the media were changed and each capacitive-coupling experimental
run was divided into four groups: control, stimulated, control and
inhibitor, and stimulated and inhibitor. Each combined-electromagnetic-fields and
inductive-coupling experimental run was divided into three groups:
control, control and inhibitor, and stimulated and inhibitor. Inhibitor
was added to the appropriate cultures at the concentrations presented
above, and 50 g/mL of sodium ascorbate was added to all
cultures. The cells were stimulated for two hours. Twenty-four hours
after the beginning of stimulation, the cells were harvested and
were used for total DNA analysis. For each run with six inhibitors,
six to twelve dishes were used for each time-period, and each run was
repeated two, three, or four times.
Statistical Analysis
All data were analyzed with use of one-way analysis of variance
and the Tukey-Kramer multiple-comparisons test for significant differences
between groups.
Part A: Effects of the Three Signals on Cell
Proliferation
All three signals produced significant increases in bone-cell DNA
per dish compared with that in the controls at all time-points (p < 0.05)
(Table I and Fig. 1). However, use
of capacitive coupling resulted in a significant, ever-increasing
DNA production (17%, 23%, 25%, and 49%)
compared with that in the controls at each time-period (thirty minutes,
two hours, six hours, and twenty-four hours, respectively), whereas
use of the other two signals resulted in only a minimal increase
in DNA production after the first thirty minutes of stimulation
(15%, 17%, 21%, and 22% for inductive
coupling and 21%, 22%, 25%, and 30% for
combined electromagnetic fields, respectively). The proliferative response
of bone cells to capacitive coupling after twenty-four hours of
stimulation increased the production of DNA by 123% compared
with that in the cells exposed to inductive coupling fields (p < 0.05)
and by 63% compared with that in the cells exposed to combined
electromagnetic fields (p < 0.05).
In the experiment in which bone cells exposed to the three different
signals for twenty-four hours were all grown in Cooper dishes in
order to rule out any influence that the size, structure, or material
of the culture dish might have on the results, the cells exposed
to the capacitively coupled signal showed a significant increase
in DNA content per dish compared with those stimulated with either
inductive coupling (p = 0.006) or combined electromagnetic
fields (p < 0.0001) (Table II).
Part B: The Effects of Signal Transduction
Inhibitors on Cell Proliferation
The increase in cellular proliferation caused by capacitive-coupling
stimulation was inhibited by blocking voltage-gated calcium channels
with verapamil, by blocking either phospholipase A2 with bromophenacyl
bromide or prostaglandin synthesis with indomethacin, or by blocking
Ca2+ activation of cytoskeletal
calmodulin with W-7. Neither blocking the release of Ca2+ from
intracellular stores with TMB-8 nor blocking the inositol phosphate
pathway in the cell membrane with neomycin had any effect on cell
proliferation (Table III).
None of the metabolic blockers (verapamil, bromophenacyl bromide,
indomethacin, or neomycin) that act within the cell membrane had
any effect on the increase in cellular proliferation produced by
inductive coupling or combined electromagnetic fields. Inhibiting
the release of Ca2+ with TMB-8
or inhibiting the activation of cytoskeletal calmodulin with W-7,
however, did block the increase in cellular proliferation otherwise
produced by either of these two signals (Table III).
The data above indicate that the signal transduction pathway activated
by the various forms of electrical stimulation is the pathway that
is inhibited by a specific metabolic inhibitor or inhibitors (Table III). Thus, for
capacitive coupling, signal transduction is by means of Ca2+ ion
translocation through cell membrane voltage-gated calcium channels
leading to increases in prostaglandin E2, cytosolic Ca2+,
and activated cytoskeletal calmodulin (Table IV and Fig. 2). For both inductive coupling and
combined electromagnetic fields, signal transduction is by means
of the intracellular release of Ca2+ leading
to increases in cytosolic Ca2+ and
activated cytoskeletal calmodulin (Table IV and Fig. 3).
The results in Part A on the effect of electrical stimulation
on cell proliferation indicated that all three electrical signals
produced a significant increase in total bone-cell DNA per dish compared
with that in the controls. However, the fact that capacitive-coupling
stimulation resulted in a significant, ever-increasing DNA production
at each time-period up to twenty-four hours, while combined-electromagnetic-fields
and inductive-coupling stimulation produced only minimal increases
in DNA production beyond the first thirty minutes, supported the concept
that the dose-response of capacitive coupling is different from
that of the other two signals. This encouraged us to proceed with
Part B of the study, in which various metabolic inhibitors were
used to block specific signal transduction pathways in order to
determine the mechanism of signal transduction for each form of
electricity used in the study.
The results in Part B indicated that transduction of a capacitively
coupled electrical signal is by means of Ca2+ ion
translocation through voltage-gated calcium channels (blocked by
verapamil) leading to an increase in phospholipase A2 (blocked by
bromophenacyl bromide) and to an increase in cytosolic Ca2+.
The increase in phospholipase A2 leads to an increase in prostaglandin
E2 synthesis (blocked by indomethacin), and the increase in cytosolic
Ca2+ leads to an increase in
activated (cytoskeletal) calmodulin (blocked by W-7) (Table IV). These findings
are in agreement with those in our previous report30. Activated calmodulin is known to
promote nucleotide synthesis and cellular proliferation31,32. Prostaglandin E2 acts as an
autocrine and/or paracrine factor to stimulate bone-cell
proliferation and possibly to increase intracellular calcium33.
The results in Part B also showed that the transduction of combined
electromagnetic fields and inductively coupled signals is by means
of the intracellular release of Ca2+ from
intracellular stores (blocked by TMB-8) leading to an increase in
cytosolic Ca2+ that, in turn,
leads to an increase in activated calmodulin (blocked by W-7) and
a subsequent increase in bone-cell proliferation. Thus, although
the initial transduction site with capacitively coupled stimulation
(voltage-gated calcium channel Ca2+ influx
into the cell) is different from that with stimulation with combined
electromagnetic fields and with inductive coupling (intracellular
release of Ca2+), all three
methods of stimulation have a common final pathway—that
is, an increase in cytosolic Ca2+ and
an increase in activated calmodulin. However, the precise mechanism
by which the electrical and electromagnetic fields are transduced
at these sites is not yet understood in terms of a rigorous model.
Also, one should be cautioned that an in vitro state, in which isolated
bone cells are grown under exacting conditions, is an artificial
environment; such conditioned cells may respond only in a limited
way by following limited biochemical pathways in response to limited
stimulation. The same cells in their natural setting in vivo are
exposed to a myriad of different upregulating and downregulating
signals and thus may respond differently from those described in
the present study.
It is interesting to note that bone cells also respond to mechanical
strain with an increase in intracellular Ca2+.
We demonstrated that bone cells subjected to a biaxial, cyclic mechanical
strain of 0.17% showed an increase in intracellular Ca2+ through
a release from intracellular stores that was due to activation of
the inositol phosphate cascade (blocked by neomycin) in the cell
membrane34. An increase in inositol
triphosphate stimulated an intracellular Ca2+ release
that, in turn, led to an increase in activated calmodulin (blocked
by W-7) and a subsequent increase in cellular proliferation (Fig. 4)34. More recently, a group of investigators
showed that a fluid shear-induced mechanical signal in osteoblasts
leads to increased expression of cyclooxygenase-2/c-Fos
through a mechanism that involves reorganization of the cytoskeleton35. The same group later showed that
these fluid shear-induced responses were due to inositol-triphosphate-mediated
intracellular Ca2+ release that
was blocked by neomycin36. Thus,
all three forms of electrical stimulation as well as mechanical
strain led, within the limitations of these experiments at least,
to the same common pathway, an increase in cytosolic Ca2+ and
an increase in activated cytoskeletal calmodulin.
It is apparent from the above discussion that the initial transduction
of a capacitive coupling signal is at or within the bone-cell membrane,
whereas the initial transduction of either combined electromagnetic
fields or inductive coupling is within or upon intracellular calcium
stores (for example, the endoplasmic reticulum). The time-varying
electromagnetic fields of the inductive-coupling and combined-electromagnetic-fields
signals pass through the bone-cell membrane to set up a time-varying
electrical field within the cytosol that, in turn, brings about
the release of intracellular Ca2+.
One possible explanation for the differences seen in the dose-response
curves of capacitive coupling compared with those of combined electromagnetic
fields and inductive coupling is that the intracellular store of
Ca2+ is limited compared with
the infinite amount of Ca2+ ions
in the extracellular fluid available to enter the bone cell by means
of activation of voltage-gated calcium channels in the cell membrane.
Several other investigators have studied bone-cell second messengers
that are activated by various electrical fields. Studies have shown
an increase in cAMP, little change in cAMP, or even a decrease in
cAMP when bone cells have been exposed to various electrical fields37-39. An increase in ornithine decarboxylase
following electrical stimulation of bone cells has also been recorded37. However, the most telling evidence
to date is the increase in prostaglandin E238 and
cytosolic calcium30,40,41 as the
predominant second messengers in electrically induced osteogenesis.
The current study certainly supports those findings.
Other investigators have looked farther downstream in the metabolic
pathway of electrically stimulated bone cells to assess the influence
of electricity on growth factors. Fitzsimmons et al.42-46 showed that low-amplitude, low-frequency
capacitively coupled signals or combined-electromagnetic-fields
signals led to an increase in insulin-like growth factor (IGF)-II
mRNA accumulation, IGF-11 secretion, and IGF-11 receptor number as
well as a net calcium flux in TE-85 osteosarcoma cells. In a previous
study, we showed that capacitively coupled electrical fields increased
transforming growth factor-β1 (TGF-β1)
mRNA in MC3T3-E1 bone cells and that this increase was blocked by
verapamil and W-747. This result
suggests that electrical stimulation delivered by capacitive coupling
induces an increase in TGF-β1 mRNA in osteoblastic
cells by a mechanism involving the cytosolic Ca2+/calmodulin
pathway.
The above studies provide insight into the biochemical events that
occur in the transduction of electrical signals used to stimulate
healing of fracture nonunions and to enhance spinal fusions. Our
data support the hypothesis that differences in the dose-response
(bone-cell proliferation) of the various forms of electrical stimulation
are due to differences in signal transduction.
There is now solid evidence that there are distinct transduction
pathways for mechanical stimulation and that electrical stimulation
with capacitive coupling, inductive coupling, and combined electromagnetic
coupling leads to a proliferative response of bone cells. Moreover,
the pathways are complementary in that they all lead to an increase
in cytosolic Ca2+ and activated
calmodulin. Electrical stimulation is finally moving beyond the "black
box" image that it has had for so many years. These studies
provide a theory of basic cellular mechanisms to augment the clinical
reports of the efficacy of electrical stimulation.
Note: The authors thank Terry Corbin of the University of Minnesota
Clinical Outcomes Research Center for research on the history of
electrical stimulation.