Cell biology of human vascular smooth muscle

Ann R Coll Surg Engl 1994; 76: 298-303
Cell biology
human vascular smooth
Philip Chan
Senior Surgical Registrar, Hon Lecturer
Department of Clinical Pharmacology and Vascular Surgery, St Mary's Hospital, London
Key words: Vascular smooth muscle cell; Intimal hyperplasia; Restenosis; Cell biology
Vascular smooth muscle is the cellular substrate of most
significant arterial diseases. Restenosis after angioplasty
and surgery mainly represents vascular smooth muscle
reaction to trauma, a process which is also significant in
the early stages of atherogenesis.
Empirical approaches, based on findings in animal
models of vascular injury, have notably failed to make
any impact on human restenosis. We have developed and
validated growth of the human VSMC in culture as a
model of restenosis.
Intimal hyperplastic lesions producing vascular restenosis contain cells that have reduced sensitivity to
physiological growth inhibition by heparin in cell culture
conditions, compared with cells from normal vascular
tissue. Undiseased saphenous vein obtained from patients
with intimal hyperplastic restenoses also contain cells
that are relatively resistant to heparin inhibition.
Arterial healing that progresses to restenosis may
have distinct and fundamental differences at the
cellular level from the normal process of arterial
healing after injury.
Why study the vascular smooth muscle cell? These cells
represent the most numerous cell population in the vessel
wall, but have largely been considered as a rather dull,
inert, structural component, albeit with responsibility for
vessel tone, in contrast to its glamorous, functional,
endothelial neighbour.
The surgical significance of the vascular smooth muscle
cell (VSMC) lies in its role in the response to vessel
injury. Haust et al. (1) and Geer et al. (2) first identified
VSMC in proliferating arterial lesions, both in human
atherosclerosis, and in experimental animal lesions after
Based on a Hunterian Lecture given at St Mary's Hospital,
London, on 13 May 1994
Correspondence to: Mr Philip Chan, Department of Vascular
Surgery, St Mary's Hospital, Praed Street, London W2 lNY
vascular injury. It has since become clear that VSMC
proliferation is the hallmark of vascular reaction to
trauma, and also to more subtle forms of mechanical
loading, such as systemic hypertension and arterialisation
of vein grafts (3). As such, the VSMC plays a central role
in important cardiovascular pathologies, such as atherogenesis, hypertensive vasculopathy, and transplantation
arteriopathy. However, the most urgent indication for
research into the VSMC arises from its role in restenosis.
The proliferative reaction of VSMC to vessel injury has
been noted since the early days of vascular surgery (4). It
has been referred to as 'fibrosis' of the artery or graft, and
been variously named as intimal, neointimal or fibrous
hyperplasia, early or accelerated atherosclerosis (5,6). My
preferred term is myointimal hyperplasia (MIH), which
recognises the role of smooth muscle in the process.
MIH is characterised by a substantial thickening of the
intimal layer of the vessel wall. The thickened intima is
composed of fibrocellular connective tissue of variable
cellularity, surrounded by extracellular matrix of collagen
and ground substance. In the early stages of MIH, the
cellular component is prominent; in mature lesions, more
extracellular matrix is observed. MIH may coexist with
atherosclerosis, which is a more complex lesion, with
elements of cell proliferation and fibrosis, but also
including regions of necrosis, and accumulation of
intracellular and extracellular lipid (7).
Thickening of the vessel wall may be self-limiting or, in
some cases, may progress to cause substantial narrowing
of the lumen and critical stenosis, and contribute to
occlusion by thrombosis. The proportion of cases in
which progressive hyperplasia occurs depends on the
nature of detection methods, the vascular insult, and
perhaps on patient-related risk factors. There is no
Cell biology of human vascular smooth muscle
difference in the histological or cytopathological appearances of MIH that progresses and MIH that does not;
furthermore, there are no detectable differences between
MIH produced at different sites by different procedures.
It would therefore appear that the proliferative VSMC
response that complicates a surgical or angioplasty
procedure in any particular vessel is essentially similar
to MIH in any other vessel, after any other procedure.
The current literature favours the term 'restenosis' for
progressive, stenotic MIH, whether or not there was a
stenosis pre-existing before the vascular procedure. Thus,
a vein graft is said to restenose, even though there was no
stenosis present in it before implantation. This nomenclature reflects the dominance of studies on coronary
angioplasty (in which true restenosis does occur) in this
The incidence of restenosis varies surprisingly little
between different interventions in different vascular beds.
Although restenosis rates for large arteries (eg iliac) are
low, relatively constant rates of 25-40% of restenoses are
observed for all procedures, surgical and endoluminal, in
all commonly operated arteries. This is all the more
remarkable given that definitions of restenosis can be
inconstant, no single definition being particularly
satisfactory in all situations (8).
There is a multitude of experimental models of
restenosis (9). Such models encompass every common
species of laboratory animal, with almost every imaginable
type of injury inflicted on every accessible vascular bed.
Although the relative merits of each model are arguable,
in general, animal models of vascular injury successfully
reproduce self-limiting, hyperplastic VSMC responses,
but do not produce progressive hyperplasia leading to
significant restenosis.
A large number of agents have been shown to be
effective in reducing VSMC proliferation in animal
models. No such agent has been able to translate this
effect into even a small significant reduction of human
restenosis, as measured by consistent results in controlled
clinical trials of adequate power (Table I). Most clinical
studies have been in the field of restenosis after coronary
angioplasty. Although many trials have been poorly
designed, often with inadequate dosage regimens, and
although there have been questions raised about the
reproducibility and error of detection with quantitative
coronary angiography (10), many clinicians and scientists
are now convinced that the disparity between animal and
human results reflects a more fundamental, biological
difference between adaptive VSMC hyperplasia on the
one hand, and maladaptive restenotic hyperplasia on the
other. The human appears to be the only species which
reliably reproduces restenosis, and then only in around
30% of cases.
Classic risk factors for atherosclerosis are not strikingly
or consistently related to the failure of vascular procedures (11,12). The inference can be drawn that these
factors are not risk factors for restenosis, and are probably unrelated to the aetiology of MIH. Nevertheless, it would be wrong to conclude that no risk factors
exist for restenosis. Clearly, some factor(s) must dis-
Table I.
Effective in animals
Ineffective in man
Aspirin/antiplatelet drugs
Fish oils
ACE inhibitors
Calcium channel blockers
Lipid lowering agents
ADP ribosyltransferase inhibitor
a-adrenergic blockers
Ornithine decarboxylase inhibitor
Antisense oligonucleotides
Fish oils
ACE inhibitors
Calcium channel
Lipid lowering
tinguish between the 30% of patients who restenose, and
the 70% who do not. There does appear to exist an
individual predilection for restenosis and graft occlusion,
with individuals failing several grafts sequentially. Most
of this data is anecdotal (6); however, a recent study of
repeat coronary angioplasty indicates that the risk of
repeated restenosis after re-angioplasty is 57%, nearly
double the primary restenosis rate of 30% (13); this
increased risk may persist even if the first restenosis
occurred in a vessel remote from the second angioplasty
(14). Epidemiological data also clearly indicates that a
powerful risk factor for restenosis after coronary
angioplasty is a previous restenosis, not necessarily at
the same site (15).
Human VSMC culture
In view of the failure of animal models, and the vital role
of the VSMC in restenosis, it appeared logical to study the
human VSMC. Noting the possibility of an individual
predilection to restenosis, I speculated that such a
predilection may be evident at a cellular level; more
specifically, that characteristics related to cell proliferation
may differ between cells derived from individuals who
restenose and those who do not.
Culture of human VSMC from saphenous vein had
been described previously (16), but was still rarely
reported in the literature (17-21). The major difference
between my approach and those described above is that
they have pursued a universal approach, whereas I have
adopted an individual one. That is, other investigators
have pooled together information derived from several
lines of VSMC, in order to define common characteristics
of 'the human VMSC'. In contrast, I have sought
differences between different specimens, in the belief
that such differences may reflect differences between
groups of patients from whom the specimens derive.
The method of human VSMC culture used by the
P Chan
groups at St Mary's and the Thrombosis Research
Institute have been described previously, and reproduced by several other groups in the UK and abroad
(22,23). The St Mary's experience of human VSMC
culture up to July 1993 extends to over 400 specimens
from cardiac or peripheral vascular surgery patients.
For all 254 specimens placed into culture between
November 1990 and November 1992, the success rate in
primary culture is 85.4% overall, with a rate of progression to the second subculture passage of 60.6%. These
figures do not exclude technical failures, and failure due to
infection, which was seen in 46 specimens. Cells derived
from restenotic lesions (n = 24) are not more successful
in culture than cells derived from normal artery or vein
(92% vs 89% success in primary culture, 58% vs 66%
progression to second passage, again without exclusions;
P=NS, x2 test). Isner's group in Boston, using a more
discriminating measure of success in culture, outgrowth
of VSMC from individual explants, have shown that
restenotic tissue is more likely to produce cells in primary
culture (24), but our data show that this 'irons out' in
further culture, and that other tissue can produce usable
cells in subculture as well as restenotic tissue.
Cell culture characteristics
Our group at St Mary's have painstakingly developed
reproducible assays of human VSMC growth (25). We
have focused on measurement of increase in cell number
over 14 days in response to stimulation by 15% fetal calf
serum (FCS), both in the presence and absence of heparin
at 100 jig/ml concentration (Fig. 1). We felt that FCS
represents a mixture of potent stimulants for VSMC
growth, similar to the in vivo stimulation of platelet
degranulation after vessel injury; heparin represents the
first described, and best characterised inhibitor of VSMC
growth (26), and may represent a mechanism of physiological growth control (27,28).
Briefly, we have shown that growth of VSMC measured
in this assay is
1 Consistent within a VSMC line in separate concurrent experiments.
2 Consistent within a VSMC line at subculture
passages between 2 and 6.
3 Consistent between VSMC lines derived from the
same tissue.
4 Consistent between VSMC lines derived from
different tissues (eg an artery and a vein) within the
same individual.
We have concluded that cell growth and heparin
inhibition is a characteristic of the individual patient,
which is preserved (or unmasked) in cell culture
conditions in a consistent and reproducible manner (25).
This validation work threw our further observations
into sharper focus, and put them on a firm statistical basis.
As there is a high degree of consistency of VSMC
growth characteristics within an individual, then variation
of such characteristics between VSMC from different
individuals might be inferred (in a statistical sense) to
Cell growth curve (15% FCS)
Cell Number
Time (days)
Figure 1. Cell growth curve. VSMC at passage 3 were
seeded at 10 000/well in 24 well plates in 15% FCS in
standard nutrient medium. After 24 h, they were growth
arrested in 0.4% FCS, and then stimulated with 15%
FCS. Triplicate wells were trypsinised at days 0, 3, 7, 10
and 14 after stimulation, and counted in a Coulter particle
counter. Medium was replenished in the remaining wells
at the same time points. Results are mean and SD for a
typical cell strain.
reflect real differences. Considered another way, it
appeared that differences between individuals are reflected in cellular behaviour in culture in a very real sense,
particularly as such behaviour is consistent within an
There are enormous differences in growth rates
between different cell strains, ranging from 300% to
2500% increase in number over 14 days. However, there
were no apparent differences in the overall growth rates
between cells from restenotic lesions and cells from
undiseased artery and vein (Fig. 2). This is in contrast
to previous reports, based on smaller samples, that cells
from restenoses were growing more rapidly (29).
Cell growth
Control patients
Restenosis patients
Cell type
Figure 2. Growth rates of human VSMC, classified by
tissue of origin. The percentage increase in cell number
over 14 days is calculated as (number at day 14/number at
day Ox 100%), for each cell strain. Results are mean and
SEM for all strains tested.
Cell biology of human vascular smooth muscle
Heparin resistant
Heparin sensitive
150000 .
Cell number
Cell number
I~~ ~
. .
Heparin 100 microg/m I
Figure 3. Growth curves of heparin-sensitive and heparin-resistant cell strains. The cell growth assay is identical to that
described in Fig. 1, with the addition of a second series of wells containing 100 gig/ml heparin (Paynes and Byrne) in 15%
FCS medium. Results are 14-day growth curves of the first two strains tested.
100 I_I
restenotc lesion
normal vein
undiseased vessel (stenosis pt.)
normal artery
Heparin sensitivity of
strain. Results
14 in
surgryPrimary bypass
VSMC, classified by tissue of origin. The percentage heparin sensitivity is
heparin containing wells/number
shown for all strains
tested; each point represents
14 in control
wells} x
heparin sensitivity for
1 00 %), for each cell
single cell strain.
P Chan
There are marked differences in response to inhibition
by heparin between different cell strains. A proportion of
VSMC cell lines were noted to be resistant to heparin at
100 gg/ml; this was certainly a novel observation, as
sensitivity to heparin had been considered to represent a
hallmark VSMC characteristic (Fig. 3). Furthermore,
heparin resistance was significantly associated with cell
lines derived from restenoses; and also with VSMC from
normal venous tissue from patients with restenoses undergoing corrective surgery (30). In comparison, VSMC
from patients undergoing primary bypass surgery (around
30% of whom might be expected to develop restenosis)
exhibited a range of heparin sensitivity, from 0-80%
inhibition (Fig. 4).
Heparin resistance
The interpretation of these findings has proved to be
controversial. Heparin resistance of VSMC in culture is
clearly associated with human restenosis, at least in
peripheral bypass grafts, which form the majority of
these specimens. It can be assumed, but has not been
shown with certainty, that some individuals with restenosis are not heparin resistant, and that others with cellular
heparin resistance might not develop restenosis after a
procedure. Cellular heparin resistance might then be
regarded as a risk factor, at the cellular level, for
restenosis. Prospective studies are under way to define
the predictive power of cellular heparin resistance for
peripheral bypass restenosis.
The finding of resistance to growth inhibition of even
undiseased saphenous vein in patients with proven
hyperplastic lesions is potentially of great importance.
VSMC from certain patients appear to be insensitive to
heparin whether they come from stenotic intimal
hyperplasia or from undiseased vessels which have notdeveloped into a stenotic lesion. This indicates that there
may be a population of patients who possess VSMC who
have the capacity for accelerated and uncontrolled growth;
that this growth potential pre-exists any vascular injury,
but a vascular injury will release that potential. The
resultant process is then qualitatively different from
normal vascular healing, and may be more likely to
progress to florid intimal hyperplasia and vascular
There is evidence from studies on rat neointimal lesions
(31), that VSMC hyperplasia is a function of a subset of
potentiated cells that exist a priori within the vessel wall.
These cells were identified within the rat arterial media by
distinct morphological features which are not seen in our
human VSMC cultures. Nevertheless, it is tempting to
identify the heparin-resistant cell as the equivalent of a
potentiated cell subset, preferentially selected by culture.
In particular, as heparin resistance is found in undiseased
vein in patients with restenosis, it could even represent a
genetic mechanism which is linked to, or even responsible
for restenosis. Our further work is concentrating on
clarifying this issue.
Implications of heparin resistance
Heparin resistance of VSMC may represent a mechanism
of aberrant growth control, leading to a propensity for
over-vigorous VSMC growth, as seen in restenosis.
Heparan sulphates, which are closely analogous to
heparin have been postulated as a physiological mechanism of growth inhibition, being secreted by endothelium
and by confluent VSMC (27,28). Resistance to growth
control is seen in cancer cell lines; although human
VSMC are not transformed, it is possible that they share
such mechanisms, but not others, with tumours. There is
an interesting correspondence between this speculation
and the clonal hypothesis of Benditt and Benditt (32),
which considers atherosclerotic lesions to be oligoclonal,
and similar to benign tumours.
Most VSMC derived from laboratory animals have
been shown to be sensitive to heparin inhibition. VSMC
from spontaneously hypertensive rats, which are prone to
proliferative vascular lesions, have been reported to be
heparin resistant (33). Quantitatively, minor subsets of
heparin-resistant VSMC (only two cell lines in all) have
been isolated from aortas of standard laboratory rats (34).
In contrast, we report a 30-40% incidence of heparin
resistance in human subjects prospectively undergoing
cardiovascular surgery.
The search for pharmacological treatments to prevent
restenosis has mainly tested agents on models of animal
vascular injury that represent the normal healing process,
as opposed to progressive restenosis (9). It is possible that
agents effective in inhibiting intimal hyperplasia in these
models may not affect abnormal, accelerated hyperplasia;
indeed we have presented evidence of this with regard to
heparin. The failure of a multitude of agents to translate
their efficacy in animal models to the prevention of human
vascular restenosis may be explained by the simple contention that the two processes are biologically different;
and this has now been established with respect to sensitivity to VSMC growth inhibition.
Furthermore, if cellular characteristics such as heparin
resistance are important determinants of restenosis, a
considerable confounding factor for randomised clinical
trials is introduced; that is, there is no practical method to
ensure that two randomised groups of patients contain
equal proportions of patients with cellular factors, such as
heparin resistance, that represent a high risk of restenosis.
This consideration may lead to type 1 false-negative error
in some existing trials that have reported negative results
as well as, theoretically, to false-positive error.
In conclusion, restenosis remains a significant clinical
problem, complicating all forms of vascular intervention
in all vascular beds. Empirical approaches, based on
findings in animal models of vascular injury have notably
failed to make any impact on human restenosis. The
human VSMC in culture offers a validated model to
examine biological processes that may differ between
animals and humans, and between humans who develop
restenosis and those who do not. Our findings have
indicated that arterial healing that progresses to restenosis
may have distinct and fundamental differences at the
Cell biology of human vascular smooth muscle
cellular level from the normal process of arterial healing
after injury.
Grateful thanks to Peter Sever, Michael Schachter and
Mahendra Patel, who gave me a chance; to Katy DemoliouMason and Sally Mill, with whom the smooth muscle adventure
started; to Laura Betteridge, Karen Gallagher and Euan Munro,
for their dedicated efforts in the lab; to Averil Mansfield,
Andrew Nicolaides and John Wolfe for their unfailing support;
and particularly to Institut des Recherches Servier, and the
British Heart Foundation.
1 Haust MD, More RH, Movat HZ. The role of smooth
muscle cells in the fibrogenesis of atherosclerosis. Am J
Pathol 1960; 37: 377-89.
2 Geer JC, McGill HC, Strong JD. The fine structure of the
human atherosclerotic lesion. Am J Pathol 1961; 38: 263-87.
3 Schwartz SM, Campbell GR, Campbell JH. Replication of
smooth muscle cells in vascular disease. Circ Res 1986; 58:
4 Carrel A, Guthrie CC. Uniterminal and biterminal venous
transplantations. Surg Gynecol Obstet 1906; 2: 266-86.
5 Imparato AM, Bracco A, Kim GE, Zeff R. Intimal and
neointimal fibrous proliferation causing failure of arterial
reconstructions. Surgery 1972; 72: 1007-17.
6 DeWeese JA. Anastomotic intimal hyperplasia. In: Sawyer
PN, Kaplitt MJ, eds. Vascular Grafts. Norwalk: AppletonCentury-Crofts, 1978.
7 Dilley RJ, McGeachie JK, Prendergast FJ. A review of the
histologic changes in vein-to-artery grafts, with particular
reference to intimal hyperplasia. Arch Surg 1988; 123: 691-6.
8 Beatt KJ, Serruys PW, Luitjen HE et al. Restenosis after
coronary angioplasty: the paradox of increased lumen
diameter and restenosis. J Am Coll Cardiol 1992; 19: 258-66.
9 Muller DW, Ellis SG, Topol EJ. Experimental models of
coronary artery restenosis. J Am Coll Cardiol 1992; 19: 41832.
10 Gould KL. Qualitative analysis of coronary artery stenosis
after coronary angioplasty-has the rose lost its bloom? J Am
Coll Cardiol 1992; 19: 946-7.
11 McBride W, Lange RA, Hillis LD. Restenosis after
successful coronary angioplasty. N Engl J Med 1988; 318:
12 Whittemore AD, Clowes AW, Couch NP, Mannick JA.
Secondary femoropopliteal reconstruction. Ann Surg 1981;
193: 35-42.
13 Bresee SJ, Jacobs AK, Garber GR et al. Prior restenosis
predicts restenosis after coronary angioplasty of a new
significant narrowing. Am i Cardiol 1991; 68: 1158-62.
14 Berger PB, Bell MR, Holmes DR, Hammes L, Kosanke JL,
Bresee SJ. Effect of restenosis after an earlier angioplasty at
another coronary site on the frequency of restenosis after a
subsequent coronary angioplasty. Am J Cardiol 1992; 69:
15 Holmes DR, Vlietstra RE, Smith HC et al. Restenosis after
percutaneous transluminal coronary angioplasty: a report
from the PTCA registry of the National Heart, Lung and
Blood Institute. Am J Cardiol 1984; 53: 77C-81C.
16 Chamley-Campbell JH, Campbell GR, Ross R. The smooth
muscle cell in culture. Physiol Rev 1979; 59: 1-61.
17 Gimbrone MA, Cotran RS. Human vascular smooth muscle
in culture. Lab Invest 1975; 33: 16-27.
18 Fager G, Hansson GK, Ottosson P, Dahllof B, Bondjers G.
Human arterial smooth muscle cells in culture: effects of
PDGF and heparin on growth in vitro. Exp Cell Res 1988;
176: 319-35.
19 Hosang M, Rouge M. Human vascular smooth muscle cells
have at least two distinct PDGF receptors and can secrete
PDGF-AA. J Cardiovasc Pharmacol 1989; 14: S22-S26.
20 Aoyagi M, Fukai N, Sakamoto H et al. Altered cellular
responses to serum mitogens, including PDGF, in cultured
SMC derived from arteries of patients with Moyamoya
disease. J Cell Physiol 1991; 147: 191-8.
21 Bjorkerud S. Effects of transforming growth factor Beta 1 on
human arterial SMC in vitro. Arterioscl Thromb 1991; 11:
22 Chan P, Mill S, Mulloy B, Kakkar V, Demoliou-Mason C.
Heparin inhibition of human vascular smooth muscle cell
proliferation in culture. Int Angiol 1992; 11: 261-7.
23 Chan P, Munro E, Patel M et al. Cellular biology of human
intimal hyperplastic stenosis. Eur J Vasc Surg 1993; 7: 12935.
24 Pickering JG, Weir L, Rosenfield K, Stetz J, Jekanowski J,
Isner JM. Smooth muscle cell outgrowth from human
atherosclerotic plaque: implications for the assessment of
lesion biology. J Am Coll Cardiol 1992; 20: 1430-9.
25 Munro E, Chan P, Patel M et al. Consistent responses of
human vascular smooth muscle in cell culture: implications
for restenosis. J Vasc Surg 1994 (in press).
26 Clowes AW, Karnovsky MJ. Suppression by heparin of
smooth muscle cell proliferation in injured arteries. Nature
1977; 265: 625-6.
27 Fritze L, Reilly C, Rosenberg R. An antiproliferative
heparan sulfate species produced by postconfluent smooth
muscle cells. J Cell Biol 1985; 100: 1041-9.
28 Castellot JJ, Addinizio ML, Rosenberg R, Karnovsky MJ.
Cultured endothelial cells produce a heparin-like inhibitor of
smooth muscle cell growth. J Cell Biol 1981; 90: 372-9.
29 Dartsch PC, Voisard R, Bauriedel G, Hofling B, Betz E.
Growth characteristics and cytoskeletal organisation of
cultured smooth muscle cells from human primary stenosing
and restenosing lesions. Arteriosclerosis 1990; 10: 62-75.
30 Chan P, Patel M, Betteridge L et al. Abnormal growth
regulation of vascular smooth muscle in patients with
restenosis. Lancet 1993; 341: 341-2.
31 Majesky MW, Giachelli CM, Reidy MA, Schwartz SM. Rat
carotid neointimal smooth muscle cells re-express a
developmentally regulated mRNA phenotype during repair
of arterial injury. Circ Res 1992; 71: 759-68.
32 Benditt EP, Benditt JM. Evidence for a monoclonal origin of
human atherosclerotic plaques. Proc Natl Acad Sci USA
1973; 70: 1753-6.
33 Resink TJ, Scott-Burden T, Baur U, Burgin M, Buhler FR.
Decreased susceptibility to cultured SMCs from spontaneously hypertensive rats to growth inhibition by heparin.
J Cell Physiol 1989; 138: 137-44.
34 Sanantonio JD, Karnovsky MJ, Ottlinger ME, Schillig R,
Pukac LA. Isolation of heparin insensitive aortic smooth
muscle cells. Arteriosci Thromrb 1993; 13: 748-57.
Received S January 1994