|
|
||||||||
Department of Neurology, University of Vermont, Burlington, Vermont 05405, USA
1Correspondence: Department of Neurology, Given C423, University of Vermont College of Medicine, Burlington VT 05405, USA. E-mail: hlangevi{at}zoo.uvm.edu
| ABSTRACT |
|---|
|
|
|---|
Key Words: collagen mechanotransduction meridian
| INTRODUCTION |
|---|
|
|
|---|
According to traditional Chinese theory, the needling of acupuncture
points has specific therapeutic effects believed to occur either
locally or at a distance via the system of acupuncture meridians
(6)
. Mechanistic models of acupuncture based on laboratory
experiments performed over the past 30 years have mostly abandoned
these traditional concepts in favor of viewing the effects of
acupuncture as taking place essentially through the nervous system
(7)
. Rather than forming an integrated system, acupuncture
points are thought to represent discrete locations on the body where
manual or electrical stimulation can activate appropriate neural
pathways (8
, 9)
. A fundamental distinction therefore
currently exists between traditional and scientific views of
acupuncture. A mechanistic explanation incorporating both classic
acupuncture theory and available scientific evidence would therefore be
a major unifying step in the field.
An important key to such a mechanism may be the characteristic
reaction to acupuncture needling known as de qi (1
, 6
, 7
, 10
, 11)
. De qi is elicited by brief manual manipulation (e.g.,
rotation, up-and-down motion) of the inserted acupuncture needle. This
method is used to elicit de qi whether or not electrical stimulation is
subsequently applied (1
, 7)
. De qi has a sensory component
perceived by the patient as an ache or heaviness in the area
surrounding the needle and a simultaneous biomechanical component,
needle grasp, perceived by the acupuncturist. During needle grasp, the
acupuncturist feels as though the tissue is grasping the needle, such
that there is increased resistance to further motion of the manipulated
needle (1
, 7
, 10)
. This tug on the needle is
classically described as "like a fish biting on a fishing line"
(12)
. Needle grasp has been described in traditional
acupuncture texts for more than 2000 years (13)
and is
still widely considered essential to acupunctures therapeutic effect
(1
, 6
, 7
, 10
, 11)
, yet its underlying mechanism is
unknown.
We hypothesize that, during de qi, the needle is being grasped by connective tissue as a result of collagen and elastic fibers winding and tightening around the needle during needle rotation. In this manner, a mechanical coupling is developed between needle and tissue. We further hypothesize that needle manipulation transmits a mechanical signal into connective tissue via this needle/tissue coupling. The subsequent transduction of this mechanical signal to a cellular response may underlie some of the therapeutic effects of acupuncture both locally and at remote locations.
| PROPOSED MECHANISM OF NEEDLE GRASP |
|---|
|
|
|---|
|
To investigate the role of the skin and/or subcutaneous (s.c.)
connective tissues in needle grasp, we used rat abdominal wall explants
as an experimental model (Fig. 2
). Using tissue explants allowed us to examine histological changes
associated with acupuncture needle manipulation in large viable tissue
samples that could be rapidly fixed after needle manipulation.
Full-thickness 4 x 4 cm rat abdominal wall samples including
dermis, subcutaneous (s.c.) muscle, s.c. tissue, and abdominal wall
muscles were excised from normal male Wistar rats (250275 g)
immediately after death and placed in 37°C HEPES buffer, pH 7.4. Five
minutes after excision, a stainless steel acupuncture needle (Seirin,
Tokyo, Japan; 0.25 mm diameter) was inserted in the center of each
sample and either rotated in one direction for 32 revolutions (Fig. 2b
) or not rotated (Fig. 2a
). After 1 min,
tissues were immersion-fixed in formalin. Fixed samples were
paraffin-embedded, sectioned parallel to the needle track, and
processed for histology. We observed that needle rotation was
accompanied by marked thickening of the s.c. connective tissue layer in
the area surrounding the needle (Fig. 2b
). There was no
structural change in dermis, s.c. muscle, or abdominal wall muscles
other than displacement by the thickened s.c. tissue layer. Masson
trichrome staining showed collagen winding around the needle track with
acupuncture needle rotation (Fig. 2c
, 2d
).
Consistent with these findings is an electron microscopy study of
debris found on acupuncture needles after insertion, manipulation, and
removal in humans, revealing elastic and collagen fibers entwined
around the needle (15)
. Together, these observations
support the hypothesis that connective tissue winds around the needle
during needle rotation.
|
In a study of the biomechanical response to acupuncture needling in
humans, we quantified needle grasp by measuring the force required to
pull out an inserted acupuncture needle (pullout force). Pullout force
was significantly greater with needle rotation compared with needle
insertion without rotation (16)
. This result demonstrates
that needle rotation enhances needle grasp. Pullout forces of 100300
g were routinely observed after needle rotation, and occasionally
pullout forces that saturated our 500 g load sensor were seen.
These are substantial loads considering the small diameter (250 µ) of
the needle.
The winding of strands of material around a rotating drum or shaft,
resulting in progressive tightening, is a common phenomenon observed in
many settings. Simple friction resists sliding of the material around
the shaft. Because the material is wrapped around the shaft, the
compressive force between the shaft and the material augments as the
tension in the strands increases. This in turn increases the friction
force, allowing more tension to be developed in the strands. In simple
systems, such as a cable wrapping around a winch drum, the maximum
tension that can be developed in the free end of the cable before the
cable slips on the drum increases exponentially with increased winding
(17)
. The winding amplifies the inherent friction between
the cable and drum. This exponential increase occurs when the cable
does not wrap over itself. If the material is allowed to wrap over
itself, as we believe happens in the case of connective tissue winding
around an acupuncture needle, the material can become self-locking in
several revolutions. That is, the friction force is amplified so much
that the material cannot slip on the shaft no matter how much tension
is applied. This self-locking winding of connective tissue is familiar
to surgeons using rotating equipment such as drills or reamers.
Winding of connective tissue around the needle results in a marked
amplification of the mechanical coupling between the needle and the
local connective tissue. Some initial coupling must be present,
however, for the tissue to begin wrapping around the rotating needle
shaft. We believe that the initial mechanical coupling results from
attractive forces between needle and tissue, such forces likely being
surface tension and electrical attraction. Connective tissue is
composed of cells embedded in extracellular matrix made of interwoven
collagen and elastic fibers associated with glycoproteins and
negatively charged proteoglycans (18)
. Electrical
attraction may therefore occur between the metal needle and fixed
tissue charges. Such attractive forces are likely to be relatively
weak, but strong enough to cause initial winding of tissue around the
rotating needle. This is made easier by the small diameter of the
needle. Once some wrapping has occurred, frictional forces take over.
The collagen and elastic fiber network is likely to have some initial
laxity, allowing the initial wrapping to occur without having to
overcome large tensile forces.
In our human experiments, we typically observed that the torque
required to rotate the needle increases continuously as needle rotation
proceeds (Fig. 3
). This is consistent with our hypothesis of connective tissue winding
around the needle. The torque reflects the tension developed in the
tissue during winding. The torque curve was similar in most cases
except for a variable duration between the start of needle rotation and
the beginning of the steep torque increase. We believe that this
represented a prewinding period during which the tissue had not yet
become caught on the needle, allowing the needle to slip.
|
Once the acupuncture needle becomes coupled to tissue, movements of the
needle (rotation or pistoning) may send a signal through connective
tissue via deformation of the extracellular matrix. To investigate this
hypothesis, we examined the orientation of collagen fibers in rat s.c.
tissue with and without needle rotation. Collagen bundles were
straighter and more nearly parallel to each other after needle rotation
(Fig. 4b
) than after needle insertion without rotation (Fig. 4a
), clearly demonstrating local alignment of tissue with
needle rotation. The importance of this effect is that pulling of
collagen fibers during needle manipulation may transmit a mechanical
signal, through deformation of the extracellular matrix, to cells such
as fibroblasts that are abundant in connective tissue. The subsequent
signal transduction events may contribute to the therapeutic effect of
de qi.
|
Mechanical signal transduction
In many cell types such as fibroblasts, endothelial cells, and
sensory neurons, focal adhesions form a mechanical link between
extracellular collagen matrix and intracellular cytoskeleton (19
, 20)
. The mechanism of mechanical load detection is thought to be
a mechanosensory complex composed of extracellular
matrix-integrin-cytoskeletal components linked to a kinase cascade
(21)
. In this model, load deformation displaces matrix
molecules tethered to clustered integrins at focal adhesions
(22)
. The cell membrane displacement is transduced by an
integrin to an integrin binding protein such as talin and then to
associated proteins such as vinculin, tensin, paxillin, Src, and focal
adhesion kinase (23)
. In addition, one or more of these
proteins can undergo a conformation change in response to displacement
and initiate a series of phosphorylation and binding reactions in the
protein complex (24)
. Therefore, the result of mechanical
load deformation of an integrin molecule via extracellular matrix
attachment is activation of a signaling cascade leading to a wide range
of cellular responses, including changes in the actin cytoskeleton with
formation of stress fibers (24
25
26)
.
Using rat s.c. tissue explants, we have found that acupuncture needle
rotation caused fibroblasts to become aligned with collagen fibers and
change shape from a rounded appearance (Fig. 4a
, 4c
, 4e
) to a more spindle-like shape (Fig. 4b
, 4d
, 4f
). Increased
cytoplasmic staining for polymerized filamentous (F-) actin can be seen
in fibroblasts 1 min after needle rotation (Fig. 4d
, 4f
) compared with needle insertion only (Fig. 4c
, 4e
). Redistribution of polymerized actin is
known to occur in cultured fibroblasts and endothelial cells within
minutes of applying a force to the cell surface using magnets or
mechanical traction (27
28
29)
. The pulling of collagen
fibers induced by acupuncture needle manipulation appears to have a
similar effect on connective tissue fibroblasts via their attachment to
collagen fibers at focal adhesion complexes.
These observations suggest that the mechanical signal created by
acupuncture needle manipulation can induce intracellular cytoskeletal
rearrangements in fibroblasts and possibly in other cells present
within connective tissue, such as capillary endothelial cells.
Cytoskeletal reorganization in response to mechanical load signals has
been shown to induce cell contraction, migration, and protein synthesis
(26
, 30)
. Potentially powerful effects may derive from
this mechanical signal transduction, including autocrine and paracrine
cellular effects, with modification of the surrounding extracellular
matrix (31)
.
Figure 4g
, h
illustrates our proposed mechanism for needle
grasp involving mechanical signaling through connective tissue:
1) winding of connective tissue around the acupuncture
needle, 2) pulling of collagen fibers and matrix
deformation, 3) transduction of the mechanical signal into
fibroblasts and/or other cells attached to collagen fibers at focal
adhesions, and 4) cellular response, including cytoskeletal
rearrangement, with potentially therapeutic downstream effects.
| POSSIBLE DOWNSTREAM EFFECTS OF NEEDLE GRASP |
|---|
|
|
|---|
B, which may promote the transcription of specific
stress-responsive genes such as collagen XII, tenascin-C, and
platelet-derived growth factor (24
A variety of mechanical stimuli (mechanical stretch, shear stress,
pulsatile stress) has been shown to alter the expression of
proto-oncogenes [c-fos (35
36
37
38)
, c-jun (38)
,
Fra-1 (38)
], as well as genes coding for extracellular
matrix components [tenascin-C (34)
, collagen XII
(34)
], enzymes [cyclooxygenase 2 (39)
,
nitric oxide synthetase (40)
], membrane proteins
(connexin 43; ref 35
), peptides (parathyroid
hormone-related peptide; ref 41
), and cytokines [PDGF
(42)
, TGF-ß1 (43
, 44)
, and tissue
plasminogen activator (45)
]. The increase in
cyclooxygenase 2 mRNA expression in response to fluid shear was blocked
by cytochalasin D (39)
, demonstrating that this response
involves actin cytoskeleton reorganization. Most of these studies were
carried out in cultured fibroblasts, endothelial, and smooth muscle
cells using either cyclical stretch or prolonged shear stress over
several hours. Although less information is available on the effect of
brief mechanical stimuli, several reports indicate that mechanical
stimuli lasting 3 s (impulse flow, endothelial cells; ref
35
), 1 min (mechanical stretch of intestinal smooth muscle
cells; 43
) and 15 min (mechanical stretch of vascular
smooth muscle cells; ref 38
) induced significant increases
in cFos (35
, 38)
, Cx43 (35)
, and TGF-ß1
(43)
mRNAs at 30 min, 90 min, and 4 h, respectively.
Increased gene expression in response to mechanical stress has also
been demonstrated in tissue explants (36
, 46)
and in vivo
(41)
.
Downstream effects of the mechanical signal generated by acupuncture
needle manipulation therefore potentially include synthesis and local
release of growth factors, cytokines, vasoactive substances,
degradative enzymes, and structural matrix elements. Release of these
substances may influence the extracellular milieu surrounding
connective tissue cells. Changes in matrix composition, in turn, can
further modulate signal transduction to and within the cell
(47)
.
Interstitial connective tissue network
The effect of mechanical forces on mesenchymal tissues has been
studied extensively in tendon, ligaments, joint capsules, dermis,
cartilage, and bone (48
49
50
51)
. Interstitial connective
tissues, on the other hand, have so far received relatively little
attention. These connective tissues constitute a network throughout the
body, including intermuscular and s.c. tissue planes, and are
continuous with more specialized connective tissues such as perimysium,
periosteum, pleura, and peritoneum. Interstitial connective tissues
also constitute the milieu surrounding nerves, blood vessels, and
lymphatics. Modification of interstitial connective tissue therefore
may have important biomechanical, vasomotor and neuromodulatory
effects.
Contraction of interstitial connective tissue has been documented
during wound healing, tissue remodeling, and fibrotic processes. During
these types of contractions, fibroblasts undergo phenotypic changes
occurring over days to years involving the expression of different
actin isoforms and formation of myofibroblasts (52)
.
Rapid reversible contraction of fibroblasts, accompanied by phenotypic
changes occurring over minutes, is well documented in vitro and
involves polymerization of soluble actin and formation of actin stress
fibers (53)
. These cytoskeletal changes are also thought
to occur in vivo, as a step toward the formation of myofibroblasts
(54)
, or in a reversible manner in response to temporary
changes in tissue strain (55)
. During acupuncture needle
manipulation, pulling of collagen may cause reversible contraction of
large numbers of fibroblasts near the acupuncture needle. This is
supported by the phenotypic change in connective tissue fibroblasts
after needle rotation shown in Fig. 4
. Local tissue contraction may
contribute to the phenomenon of needle grasp and to the tugging
sensation felt by the acupuncturist. Furthermore, the contraction of
fibroblasts itself would cause further pulling of collagen fibers,
resulting in a wave of matrix deformation and cell contraction
spreading away from the needle through interstitial connective tissue
(Fig. 5
). Patients frequently report a slow spreading of de qi sensation along
acupuncture meridians (6)
. Acupuncture points and
meridians typically are located between muscles or between a muscle and
a tendon or bone (1
, 56)
. The ancient maps of acupuncture
points and meridians may essentially be a guide to insert the needle
into connective tissue. Spreading of matrix deformation and cell
activation along connective tissue planes thus may mediate acupuncture
effects remote from the acupuncture needle site (Fig. 5)
.
|
Sensory afferent stimulation
Different types of sensory receptors may be stimulated directly as
a result of the matrix deformation generated by acupuncture needle
manipulation. A study by Wang et al. (57)
reported an
association between stimulation of group III muscle afferents and
sensations of heaviness, distention, and aching typically associated
with de qi. This study has been quoted as evidence that the sensation
perceived during de qi is due to a contraction of muscle
(7)
. However, group III muscle afferents are found in
perimuscular fascia and the adventitia of muscle blood vessels and
respond to a variety of stimuli including pressure, stretch, and
mechanical stimulation of the muscle surface (58)
, which
could occur as a result of mechanical connective tissue matrix
deformation. Group III muscle afferents, along with various other types
of sensory receptors present within connective tissue, may therefore be
activated directly as a result of the mechanical signal generated by
needle manipulation. In addition to group III muscle afferents, other
types of sensory receptors and primary afferent nerve fibers are known
to transmit mechanosensory information in s.c. and deep connective
tissues, and may be stimulated by connective tissue matrix deformation.
Both slow-adapting (SA II) and fast-adapting (FA II) mechanoreceptive
afferents have been described in dermis, s.c. tissue, and interstitial
connective tissue planes. SA II receptors respond to both pressure and
stretch and consist of bundles of collagen fibers with sensory axons
branching between collagen fibrils (Ruffini endings). FA II receptors
are associated with Pacinian corpuscles and most effectively transmit a
sensation of vibration. Two main types of nociceptors can also transmit
mechanosensory information in skin and deep connective tissues. A
mechanical nociceptors give rise to small myelinated fibers and are
thought to respond to damaging mechanical stimuli, though the threshold
for these receptors vary, many being in the innocuous range
(59)
. Finally, C polymodal nociceptors give rise to small
unmyelinated fibers and respond to noxious mechanical, thermal, and
chemical stimuli.
Acupuncture needle manipulation therefore may (via connective tissue matrix deformation) cause stimulation of a wide variety of sensory mechanoreceptors and/or nociceptors.
The importance of this effect is that 1) connective tissue matrix deformation may not be restricted to the area of the needle, but may spread along interstitial connective tissue planes; 2) a wave of sensory receptor activation occurring over seconds to minutes may simultaneously follow the mechanical signal away from the needle site; 3) a second wave of cellular activation, followed by altered gene expression, protein synthesis, and extracellular matrix modification, may ensue after a certain time delay and last hours to days; and 4) subsequent stimulation of these connective tissue sensory receptors by body movement may be modulated by this sequence of events.
Neuromodulation
Whether the sensation evoked by stimulation of various types of
sensory receptors is experienced as pain (or not) depends not only on
the type of receptor, but also on the status of the tissue and of
synaptically related spinal cord neurons. Peripheral sensitization of
primary afferents or changes in central synapses can contribute to an
increased pain sensation (60)
. It is now widely accepted
that target organs can influence the neurons that innervate them.
Peripheral tissue factors known to influence sensory input include
tissue perfusion and inflammatory mediators (61)
. Tissue
perfusion is itself regulated by a complex interplay of autonomic,
hormonal, and local controls (62
, 63)
. In connective
tissue, fibroblasts and collagen matrix are the underlying milieu in
which these regulatory events take place. This connective tissue milieu
has the property of responding to mechanical signals such as those
produced by acupuncture needle manipulation. The effect of acupuncture
needle manipulation on blood flow, cytokines, and/or growth factors may
result in long-term modulation of sensory information. This may
influence whether or not sensations generated by stretching of
connective tissue during body movements are perceived as pain. The
delayed cellular and molecular events triggered in connective tissue by
acupuncture needle manipulation may therefore modulate processing of
mechanical sensory stimuli that occur hours to days later.
Link to measurable clinical effects
Pain is a subjective symptom that is notoriously difficult to
quantify, and the therapeutic effect of acupuncture in general has been
difficult to study under placebo-controlled conditions (3
, 5)
. Changes in tissue perfusion, pH, cytokines, or growth
factors, however, can be studied objectively using techniques such as
laser Doppler fluxmetry, isotope washout methods and microdialysis.
These techniques may provide objective evidence of prolonged connective
tissue changes after acupuncture treatments both near and distant from
the needle along connective tissue planes. The connective tissue
environment surrounding nerves, blood vessels, and lymphatics may play
an important and largely unexplored role in various types of chronic
pain. The effects of acupuncture on connective tissue may therefore be
important from the point of view of 1) establishing a
mechanism linking acupuncture needle manipulation to a therapeutic
effect, 2) providing biological markers of the effect of
acupuncture that can be used in clinical trials, and 3)
understanding the role played by connective tissue in the pathogenesis
of chronic pain syndromes.
In summary, the insertion and manipulation of acupuncture needles may
have both local and remote therapeutic effects based on the same
underlying mechanism: mechanical coupling of needle to connective
tissue, winding of tissue around the needle, generation of a mechanical
signal by pulling of collagen fibers during needle manipulation, and
mechanotransduction of the signal into cells. Downstream effects of
this mechanical signal may include cell secretion, modification of
extracellular matrix, amplification and propagation of the signal along
connective tissue planes, and modulation of afferent sensory input via
changes in the connective tissue milieu (Fig. 5)
.
We propose that mechanical signal transduction is a common mechanism
underlying the effects of a variety of acupuncture needling methods.
Modern acupuncture techniques using electrical stimulation may have
additional effects through prolonged stimulation of nerves or muscle.
However, de qi is a common denominator to both traditional and modern
acupuncture treatments. Indeed, documentation of de qi is used as a
criterion for evaluating the adequacy of both manual and electrical
acupuncture treatments in clinical trials (64
, 65)
.
Acupuncture needle rotation (either uni- or bidirectional) may be
important to initiate needle grasp, but other types of needle
manipulation such as pistoning may also effectively transmit a
mechanical signal to cells once needle grasp has been initiated.
Transduction of the mechanical signal into cells with subsequent
cellular response and downstream effects may explain the perplexing
claim that acupuncture treatments have long term effects lasting for
days to weeks and even permanently.
Traditional acupuncture theory is based on empirical observations first made over 2000 years ago that so far have remained without solid scientific validity. The field of mechanotransduction may now provide scientific grounding for this ancient form of therapy. In return, acupuncture may provide an important clinical application for the current explosion in basic knowledge of the powerful and diverse biological effects of mechanical signaling.
| ACKNOWLEDGMENTS |
|---|
Received for publication March 13, 2001.
Revision received June 20, 2001.
| REFERENCES |
|---|
|
|
|---|
1(I) collagen expression in fetal human intestinal smooth muscle cells. Am. J. Physiol. 40,G1074-G1080
This article has been cited by other articles:
![]() |
D. D. Tang and Y. Anfinogenova Physiologic Properties and Regulation of the Actin Cytoskeleton in Vascular Smooth Muscle Journal of Cardiovascular Pharmacology and Therapeutics, June 1, 2008; 13(2): 130 - 140. [Abstract] [PDF] |
||||
![]() |
S.-M. Wang, Z. N. Kain, and P. F. White Acupuncture Analgesia: II. Clinical Considerations Anesth. Analg., February 1, 2008; 106(2): 611 - 621. [Abstract] [Full Text] [PDF] |
||||
![]() |
S.-M. Wang, R. T. Constable, F. S. Tokoglu, D. A. Weiss, D. Freyle, and Z. N. Kain Acupuncture-Induced Blood Oxygenation Level-Dependent Signals in Awake and Anesthetized Volunteers: A Pilot Study Anesth. Analg., August 1, 2007; 105(2): 499 - 506. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Berman A 60-Year-Old Woman Considering Acupuncture for Knee Pain JAMA, April 18, 2007; 297(15): 1697 - 1707. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. M. Langevin, N. A. Bouffard, G. J. Badger, J. C. Iatridis, and A. K. Howe Dynamic fibroblast cytoskeletal response to subcutaneous tissue stretch ex vivo and in vivo Am J Physiol Cell Physiol, March 1, 2005; 288(3): C747 - C756. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. M. Langevin, D. L. Churchill, J. R. Fox, G. J. Badger, B. S. Garra, and M. H. Krag Biomechanical response to acupuncture needling in humans J Appl Physiol, December 1, 2001; 91(6): 2471 - 2478. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |