Download Morphologic Alteration of Cultured Arterial Smooth Muscle Cells by

JOURNAL
OF SURGICAL
RESEARCH
35, 490-497 (1983)
Morphologic Alteration of Cultured Arterial Smooth
Muscle Cells by Cyclic Stretching’
VIKROM S. SOTTIURAI,*
PETER KOLLROS, SEYMOUR GLAGOV,~
CHRISTOPHER
K. ZARINS, AND MARTIN B. MATHEWS
Departments of Pathology, Pediatrics, and Surgery of The Pritzker School of Medicine at
The University of Chicago, Chicago, Illinois 60637
Submitted for publication March 8, 1983
Cyclic stretching of smooth muscle cells in culture resulted in a two- to fivefold increase in protein
and collagen synthesis. The same in vitro system was utilized to relate changes in smooth muscle cell
morphology to mechanical stress. Smooth muscle cells, grown in culture from rabbit aorta explants,
were transferred to purified elastic membranes derived from bovine aorta. The membranes were either
subjected to stretching and relaxation 52 times per minute or stretched and held stationary for 8, 48,
or 56 hr. Profiles of rough endoplasmic reticulum (RER) were counted and myohlament content
estimated from electron micrographs of 100 cells for each experiment. Cells from cyclically stretched
preparations were compared with stationary cells derived from the same subculture. Myolilaments were
largely replaced by RER in cyclically stretched cells and there was a reciprocal relationship between
RER and myotilament content in individual cells. In cells from stationary preparations, myolilament
content also diminished with time but RER profiles were few. At 56 hr, RER profiles numbered 16.7
+ 1.7 in stretched cells compared with 3.6 + 1.3 in stationary cells (P -Z 0.05). Cyclically stretched cells
formed numerous intercellular contacts and showed little evidence of cytoplasmic degradation while
stationary cells showed few contacts and contained numerous cytosomes and lamellar bodies. The
results suggestthat cyclic stretching resulted in the formation of RER or the preservation of myofilaments
and that immobility resulted in the disappearance of myofilaments and cytoplasmic degradation. We
conclude that cyclic stretching of arterial cells in culture under a variety of mechanical and metabolic
conditions may reveal relationships among excessive mechanical stress,hormonal status, and metabolic
environment in the modification of cell structure and matrix elaboration. Such studies could help to
identify factors involved in the formation of arterial dysplasias.
Arterial fibrodysplasia is a pathologic entity
of undetermined etiology. It consists of abnormalities of the artery wall characterized by
altered morphology and disproportion and/
or disarray of its usual cellular and matrix
components. Mural ischemia, mechanical
stress, and hormonal effects have been proposed as factors in the pathogenesis of this
disorder [5, 7, 12, 131. A role for ischemia
was supported by recent experimental findings
’ The work was supported by Grant HL 23022-03 of
the National Institutes of Health and was presented in
part at the Association for Academic Surgery on November
11, 1981 in Chicago.
* Present address: Department of Surgery, L.S.U. Medical Center, 1542 Tulane Avenue, New Orleans, La. 70 112.
’ To whom reprint requests should be addressed: Department of Pathology, University of Chicago, Chicago,
Ill. 60637
0022-4804/83 %I .50
Copyright 0 1983 by Academic Press, Inc.
All rights of reproduction
in any form resewed.
that occlusion of vasa vasorum by thrombin
gel resulted in medial fibrosis [ 191 and that
hypoxia induced morphologic alterations in
cultured smooth muscle cells [ 111. There has
been no direct evidence that reaction of the
media to physical torsion or excessive tension
is a contributing factor, although this hypothesis is attractive in view of the focal nature of
the morphologic changes in many instances
[20]. A system for investigating the effects of
mechanical stress on cells in culture has been
developed by Leung et al. [8]. Utilizing this
apparatus, they found that cyclic stretching
markedly increased the synthesis of collagen
and certain acid mucopolysaccharides by rabbit aortic smooth muscle cells but had no effect
on cell proliferation [8,9]. The purpose of the
present report is to document the changes in
cell morphology which correspond to the
490
SOTTIURAI
ET AL.: CYCLIC STRETCHING
demonstrated biosynthetic effects of cyclic
stretching on arterial smooth muscle cells in
culture. The findings indicate that this may
be a useful in vitro model for isolating and
probing the interaction between specific modes
of mechanical stress and other metabolic conditioning factors in determining the structure
and function of arterial smooth muscle cells.
MATERIALS
AND METHODS
Methods for cell preparation, a description
of the apparatus, and the procedure used to
subject arterial smooth muscle cells to cyclic
stretching in culture have been presented in
detail elsewhere [9]. In brief, cells were grown
from explants of rabbit aortic media and subcultured in modified Eagle’s medium supplemented with 10% fetal calf serum. The initial
explants were grown to confluence in 2-3
weeks. Elastin membranes served as the mechanical substrate for the cells. These were
produced by slicing samples of bovine aorta
in a cryostat and autoclaving the resulting
slices to remove cells and collagen. The purified elastin membranes were mounted on
Teflon frames designed to permit cyclic
OF SMOOTH
MUSCLE
CELLS
491
stretching and relaxation or to hold them stationary at any desired length. The frames with
their membranes were mounted in individual
culture dishes and the entire assembly sterilized by autoclaving. Smooth muscle cells (approximately 2 X 106) were then plated onto
the membranes and the dishes were transferred
to an incubator
modified to permit the
mounting frames to be connected by a train
of hinged connecting rods to a motor outside
the incubator. The incubator used for the preparatory cultures, and the modified incubator
in which the stretching experiments were conducted, were gassed with 10% CO*, 90% air,
and maintained at 37°C. For each of the experiments which form the basis of this report,
cells from the same subculture were transferred
to two membranes. After 5 days of further
growth, the dishes were transferred to the
modified incubator.
One membrane was subjected to stretching
to 110% of original length and relaxation to
original length at the rate of 52 cycles per
minute, while the other was elongated to 110%
of its original length and remained stationary.
Experiments were carried out for 8, 48, and
56 hr. Two trials were performed at each time
FIG. 1. Electron micrograph of a typical rabbit aortic smooth muscle cell following cyclic stretching on
a purified bovine e&tin membrane in culture for 56 hr. Only peripheral myofilaments (mf) remain. The
cytoplasm is occupied predominantly by rough endoplasmic reticulum (er). Numerous cytoplasmic processes
are present. Some of these processes interdigitate with those of the adjacent cell (arrows). Basal lamina (bl)
is prominent. There is little evidence of organelle degradation. X9600.
492
JOURNAL
OF SURGICAL
RESEARCH: VOL. 35, NO. 6, DECEMBER
interval. At the conclusion of each experiment,
samples of the membranes were fixed by immersion in 3% phosphate-buffered glutaraldehyde at pH 7.4 for 2 hr. Care was taken to
remove and fix samples with the membranes
in their maximally stretched positions. This
was accomplished by placing two metal ring
washers, one above and one below the membrane, clamping the washers together with
screws, and cutting out a circle of membrane
at the outside edge of the washers. A radial
marking on the upper washer was oriented so
as to indicate the direction of stretch. The
tissue sample within the washer center space
was then cut into blocks, postfixed in 1% 0~0~
for 1 hr, followed by ethanol dehydration,
Epon embedding, thin-sectioning, and staining
for transmission electron microscopy (TEM).
1983
All of the cells present on sections from
each of three randomly chosen blocks from
each preparation were examined. Cell profiles
were then photographed for assessment of cytoplasmic components. The criterion for eligibility for quantitative study was that the cell
profile show an uninterrupted cell membrane
and a cross-sectional area at least two-thirds
as large as the largest cell profile found for
each experiment. The first 100 cell profiles to
fulfill these criteria were photographed and
prints prepared at identical magnifications.
Rough endoplasmic reticulum (RER) was
quantitated by counting the RER profiles and
semiquantitative
estimates of myofilament
content were made by sorting the photographs
repeatedly and assigning grades of 0 to 3+
according to increasing prominence of the
FIG. 2. Electron micrograph of a rabbit aortic smooth muscle cell at&r 48 hr of cyclic stretching in
culture. Cytoplasmic processes (cp) are numerous. Pro6les of rough endoplasmic ieticulum (er) were less
abundant at 48 hr than at 56 hr. Residual myofilaments (mf) are foucd at the periphery bf the cell. These
were most prominent in the cyclically stretched cells which contained the fewest profiles of endoplasmic
reticulum. Mitochondria (m), vacuole (v). X20,000.
SOTTIURAI
ET AL.: CYCLIC STRETCHING
myofilaments in the cell profiles. Cells from
cyclically stretched preparations were compared with cells grown on stationary membranes derived from the same subculture.
RESULTS
Characteristic features of smooth muscle
cell morphology,
i.e., myofilaments,
basal
lamina, dense bodies, and micropinocytotic
vesicles, became less distinct with time in culture in all of the experiments but differences
between cyclically stretched and stationary
preparations were apparent at all of the time
intervals. Smooth muscle cells subjected to
cyclic stretching (Figs. l-3) were usually elongated, with their long axes oriented mainly in
the direction of stretching. Polar cytoplasmic
processes in contact with, or interdigitating
with, adjacent cells were numerous. Ribosomes and RER were abundant and were most
prominent in the cells which were subjected
to prolonged cyclic stretching (Fig. 1). Strips
OF SMOOTH
MUSCLE
CELLS
493
of myotilaments were found in most of the
cells and basal lamina was often extensive.
Cells on stationary membranes (Figs. 4 and
5) showed only occasional cell interdigitations
or junctions and lacked any specific pattern
of cell orientation. Profiles of basal lamina
were frequently interrupted, attenuated, or
absent. RER, mitochondria,
ribosomes, and
Golgi complexes were sparse. Myofilaments
were greatly reduced, particularly in central
portions of the cell bodies, but there was no
consistent change with time during the experimental period. Unlike cyclically stretched
cells, stationary cells contained numerous
myelin figures, cytosomes, and laminar bodies
as well as vacuoles.
Quantitation
revealed that RER profiles
were five times more abundant in stretched
than in nonstretched cells (Table 1) and appeared to increase with duration of cyclic
stretching. Cells that were stretched for 56 hr
had 25% more rough endoplasmic reticulum
than cells stretched for 8 hr. In addition, di-
FIG. 3. Electron micrograph of a smooth muscle cell following 8 hr of cycling stretching. Lipid droplets
(1) and cytosomes (cs) were seen in both stretched and stationary preparations, but were much more
abundant in stationary preparations. Rough endoplasmic reticulum (er), mitochondria (m), and cytoplasmic
processes are evident (cp). X8000.
FIG. 4. Electron micrograph of a smooth muscle cell following 8 hr of culture on a stationary elastin
membrane. Myofilaments (mf) are evident as well as peripheral dense bodies (db). Some rough endoplasmic
reticulum (er) and a few ribosomes (r) are. also present as well as occasional cytosomes (cs). Mitochondria
(m).XlO,OOO.
FIG. 5. Electron micrograph of a typical rabbit aortic smooth muscle cell after 56 hr of culture as a
sparse. Cytoplasmic processes are few. Cytosomes (cs) and lamellar bodies (lb) are numerous.
494
X
10,000.
SOTTIURAI
ET AL.: CYCLIC STRETCHING
TABLE I
PROFILESOFROUGH ENWPLASMIC RETICULUMIN
RABBITAORTICSM~IOOTH MUSCLE CELLCULTURES
Duration
in culture
(W
RER profiles/cell section’
Stretched
Nonstretched
P
8
48
56
11.8 rt 1.2
14.2 f 1.4
16.7 f 1.7
1.9 f 0.6
3.1 f 0.7
3.6 + 1.3
co.05
-co.05
-co.05
Note. Profiles of rough endoplasmic reticulum in cyclically stretched and in stationary cell smooth muscle
cell cultures. Each determination is based on the examination of 100 cells.
0 Mean k SD for 100 cells for each determination.
lated cistemae were more evident in cells following the more extended periods of stretching
(Fig. 6). In cyclically stretched cells, the abundance of myofilaments tended to be inversely
related to that of other cellular organelles, but
particularly in relation to RER and Golgi
complexes. In most of the cells with more
than 10 RER profiles, myofilaments
were
graded 0 or 1+. Cells with fewer than 6 RER
profiles usually contained 2+ or 3+ myofilaments. Cells grown on stationary membranes
had only 0 or l+ grades of myofilaments despite the sparsity of RER profiles.
DISCUSSION
In the present study morphologic differences
between cyclically stretched and stationary
smooth muscle cells were apparent at 8 hr and
became increasingly prominent
with prolonged periods of cyclic stretching. Myofilaments were generally diminished in both situations but evidences of organelle degradation
were more evident in cells grown on stationary
membranes. Cytosomes and myelin figures
were especially prominent in the stationary
preparations. In contrast, the cytoplasm of
smooth muscle cells subjected to cyclic
stretching contained abundant rough endoplasmic reticulum and dilated cistemae. RER
profiles were 4-6 times more abundant in cyclically stretched cells depending on duration
OF SMOOTH
MUSCLE
CELLS
495
of stretching. This observation is consonant
with the two- to fivefold increase in protein
and collagen synthesis by cyclically stretched
aortic smooth muscle cells reported by Leung
et al. [8, 91 using the same system. This close
correspondence indicates that quantitative
correlations can be established among cell
morphology, biosynthesis of matrix fibers, and
mechanical stimuli for smooth muscle cells
under these experimental conditions.
Worthy of note is the finding that the cyclically stretched cells showed evidence of
modulation from a predominantly contractile
to a predominantly
biosynthetic mode with
minimal evidence of organelle degradation
and a reciprocal relationship between myofilament and RER content. The stationary
smooth muscle cells, not subjected to cyclic
changes in tensile stress, showed both a
marked reduction in myofilaments and relatively little increase in rough endoplasmic reticulum. The absence of tensile stimulation
apparently reduced cell function to basal
maintenance levels in the stationary preparations and favored the involutional and degenerative changes usually noted in culture,
while cyclic stretching induced increased matrix biosynthesis or maintained a complement
of myofilaments. Such findings indicate that
cultured arterial smooth muscle cells may be
capable of a range of responses to tensile stimuli. One should be able to find the level of
stretch amplitude and frequency, in vitro,
which maintains the cell in a predominantly
contractile or a predominantly
biosynthetic
mode for extended periods.
The ability of an arterial wall to withstand
the tensile forces associated with changes in
size during growth and with increases in blood
pressure, and to recover from mechanical injury [2 11, requires cell proliferation as well as
modifications in matrix composition and architecture of the media. The mechanisms by
which this adaptation is accomplished are not
clear. There is, however, evidence that the rate
of connective tissue production by smooth
muscle cell during normal growth [lo] and in
response to hypertension [22] is closely related
to the magnitude of tensile stress imposed on
496
JOURNAL
OF SURGICAL
RESEARCH: VOL. 35, NO. 6, DECEMBER
1983
FIG. 6. Smooth muscle cell 56 hr after cyclic stretching. Note abundant Golgi complexes (gc), and dilated
cistemae of abundant rough endoplasmic reticulum (er). Myolilaments are absent and there is no evidence
of organelle degeneration. Fibrillar structures are evident in the widened cistemae (arrows). A centriole (c)
is present. X9600.
the vessel wall. Under abnormal metabolic
conditions and in response to chronic mechanical trauma, arterial smooth muscle cells
may respond without maintaining
normal
proportions of matrix or normal architecture
depending on size and location of the vessel.
Following
physical injury, for example,
smooth muscle cells of the coronary artery
tend to synthesize mainly collagen, while those
of the thoracic aorta increase elastin production predominantly
[ 171. Hyperlipidemia
has
been shown to interfere with healing of the
media following an acute transmural injury
[2]. Modification of, or interference with, normal biosynthetic [ 15, 161 regulatory and reparative [ 141 mechanisms by excessive and
prolonged mechanical deformations, altered
hormonal status, and/or hypoxia may contribute to the development of artery wall dysplasias. Ross and Klebanoff have reported that
estrogen enhances protein synthesis by uterine
smooth muscle cells [ 181. Similar effects of
hormones on vessel walls are suggested by the
findings of Bo et al. [ 11. Reduction in nutrient
supply to the vessel wall has been shown to
increase accumulation of collagen and certain
mucopolysaccharides with concomitant morphologic alteration of the smooth muscle cells
[ 12, 13, 191. Arterial changes in pulmonary
hypertension [6], as well as intimal fibrocellular hyperplasia in vein grafts [3,4], are probably examples of reactions to prolonged, abnormally increased medial stresses. Age may
also play a role. Although caution must be
SOTTIURAI
ET AL.: CYCLIC STRETCHING
exercised in extrapolating the results of mechanical stretching of smooth muscle cells in
vitro to mural dysplasia in vivo, our system
has the advantage of permitting the independent evaluation of many of the variables which
may modulate smooth muscle response. Results of such experiments could help to establish the pathogenic significance of hormonal, nutritional, and mechanical factors in
the development of the morphological features
associated with arterial dysplasias.
REFERENCES
1. Bo, W. J., Odor, D. L., and Rothrock, M. L. Ultrastructure of uterine smooth muscle following progesterone or progesterone-estrogen treatment. Anat.
Rec. 163: 121, 1968.
2. Bomberger, R. A., Zarins, C. K, and Glagov, S. Medial
injury and hyperlipidemia in development of aneurysms or atherosclerotic plaques. Surg. Forum 31:
338, 1980.
3.
Brody, W. R., Kosek, J. C., and Angell, W. W. Changes
in vein grafts following aortocoronary bypass induced
by pressure and ischemia. J. Thorac. Cardiovasc. Surg.
64 841,
1972.
Curtis, J. J., Stoney, W. S., Alford, W. C., Jr., Burrus,
G. R., and Thomas, C. S., Jr. Intimal hyperplasia.
Ann. Thou. Surg. 20: 628, 1975.
5. Dimpoulos, C., and Kehayas, P. Nephroptosis. Luncet
4.
2: 667,
1972.
Esterly, J. A., Glagov, S., and Ferguson, D. J. Morphogenesis of intimal obliterative hype&&a of small
arteries in experimental pulmonary hypertension.
Amer. J. Pathol. 62: 325, 1968.
7. Kaufman, J. J., and Maxwell, M. H. Upright aortography in the study of nephroptosis, stenotic lesions
of the renal artery and hypertension. Surgery 53: 736,
6.
1963.
Leung, D. Y. M., Glagov, S., and Mathews, M. B.
Cyclic stretching stimulates synthesis of matrix components by arterial smooth muscle cells in vitro. Science 191: 415, 1976.
9. Leung, D. Y. M., Glagov, S., and Mathews, M. B. A
8.
10.
1*.
t2
’
OF SMOOTH
MUSCLE
CELLS
497
new in vitro system for studying cell response to mechanical stimulation. Exp. Cell Rex 109: 285, 1977.
Leung, D. Y. M., Glagov, S., and Mathews, M. B.
Elastin and collagen accumulation in rabbit ascending
aorta and pulmonary trunk during postnatal growth:
Correlation of cellular synthetic response with medial
tension. Circ. Rex 41: 316, 1917.
May, J. H., Paule, W. J., and Blankenhom, D. H.
Effect of varying oxygen levels on cultured aortic
smooth muscle cells. Anat. Rec. 178: 413, 1974.
Nakata, Y., and Shionoya S. Vascular lesions due to
obstruction of the vasa vasorum. Nature (London)
212: 1258,
1966.
13. Nakata, Y. An experimental study on the vascular
lesions caused by obstruction of the vasa vasorum.
Japan. Circ. J. 31: 275, 1967.
14. Poole, J. C. F., Cromwell, S. B., and Bend&t, E. P.
Behavior of smooth muscle cells and formation of
extracellular structures in the reaction of arterial walls
to injury. Amer. J. Pathol. 62: 39 1, 197 1.
15. Ross, R., and Klebanoff SJ. The smooth muscle cell
and in vivo synthesis of connective tissue protein. J.
Cell Biol. 50: 159, 1971.
16. Ross, R. The smooth muscle cell. II. Growth of smooth
muscle in culture and formation of elastic fibers. J.
Cell Biol. SO: 172, 197 1.
17. Ross, R. The arterial smooth muscle cell. In R. Wissler
and J. Geer (Eds.), The Puthogenesis ofdtherosclerosis, Baltimore: Williams & Wilkins, 1972. P. 147.
18. Ross, R., and Klebanoff, S. J. Fine structural changes
in uterines smooth muscle and fibroblasts in response
to estrogen. J. Cell Biol. 32: 155, 1967.
19. Sottiurai, V., Fry, W. J., and Stanley, J. C. Ultrastructural characteristics of experimental arterial medial fibroplasia induced by vasa vasorum occlusion.
J. Surg. Rex 24: 169, 1978.
20. Stanley, J. C., Gewerta, B. C., Bove, E. L., Sottiurai,
V., and Fry, W. J. Arterial fibrodysplasia. Histopathologic character and current etiologic concepts. Arch.
Surg. 110: 561, 1975.
2 1. Glagov, S.,and Ts’ao, C. H. Restitution of aortic wall
after sustained necrotizing transmural ligation injury:
Role of blood cells and artery cells. Amer. J. Pathol.
19: I, 1915.
22.
Wolinsky, H. Response of the rat aortic media to
hypertension: Morphological and chemical studies.
Circ. Rex 26: 507, 1970.