Content uploaded by Wieland Gevers
Author content
All content in this area was uploaded by Wieland Gevers on Oct 22, 2018
Content may be subject to copyright.
Proc.
Natl.
Acad.
Sci.
USA
Vol.
76,
No.
1,
pp.
353-357,,
January
1979
Cell
Biology
Glycoprotein,
elastin,
and
collagen
secretion
by
rat
smooth
muscle
cells
(differentiation/extracellular
matrix/smooth
muscle
sublines)
PETER
A.
JONES*,
TIMOTHY
SCOTT-BURDENt,
AND
WIELAND
GEVERSt
Medical
Research
Council,
Unit
for
Molecular
and
Cellular
Cardiology,
University
of
Stellenbosch
Medical
School,
Box
63,
Tygerberg
7505,
South
Africa
Communicated
by
Melvin
Calvin,
October
25,1978
ABSTRACT
Smooth
muscle
cells
from
rat
heart
secreted
extracellular
matrix
components
at
high
rates
for
many
gener-
ations
in
culture.
The
matrix
proteins
remained
anchored
to
the
culture
dish
and
were
characterized
after
removal
of
cellular
material
with
sodium
dodecyl
sulfate.
Sequential
enzyme
di-
gestion
demonstrated
the
presence
of
at
least
three
components,
including
glycoprotein(s),
elastin,
and
collagen.
Prolonged
ex-
traction
of
the
matrix
with
detergent
under
reducing
conditions
solubilized
a
fucosylated
glycoprotein
having
an
apparent
molecular
weight
of
250,000
and
two
other
proteins
with
mo-
lecular
weights
of
72,000
and
45,000,
respectively.
Sublines
derived
from
discrete
colonies
of
smooth
muscle
cells
synthe-
sized
all
of
the
matrix
components,
and
the
proportion
of
col-
lagen
secreted
by
some
sublines
increased
with
time
in
culture.
The
biosynthesis
of
a
mixed
extracellular
matrix
and
the
rela-
tionships
among
the
component
proteins
were
therefore
studied
in
one
system
producing
milligram
quantities
of
material.
The
extracellular
matrix
proteins
are
of
great
importance
to
the
functioning
of
an
organism.
Present
culture
models
for
their
biosynthesis
are
not
altogether
satisfactory
because
most
cells
tend
to
lose
their
differentiated
properties
when
they
are
re-
moved
from
their
normal
environment
(1)-for
example,
3T3
and
3T6
cells,
which
are
often
used
for
studies
of
collagen
biosynthesis,
have
largely
lost
their
ability
to
synthesize
this
protein
(2,
3).
However,
it
has
recently
been
shown
that
primary
avian
tendon
cells
synthesize
physiological
amounts
of
collagen
under
the
correct
culture
conditions
(1).
The
current
interest
in
the
proteins
comprising
the
extra-
cellular
matrix
(4-7)
involves
not
only
collagen
and
elastin
but
also
the
high
molecular
weight
glycoproteins
which
probably
play
a
major
role
in
the
organization
and
properties
of
the
matrix.
The
microfibrillar
protein
of
the
elastic
fiber
has
an
apparent
molecular
weight
of
270,000
(8)
and
has
been
sug-
gested
to
be
involved
in
the
organization
of
the
fiber
(7).
Fi-
bronectin
(LETS
protein)
has
also
been
shown
to
be
a
pre-
dominantly
matrix
protein
(9),
to
bind
to
collagen
(10),
and
possibly
to act
as
a
bridge
in
the
binding
of
cells
to
collagen
(11).
Our
understanding
of
the
biosynthesis,
processing,
and
inter-
actions
of
these
proteins
has
been
retarded
not
only
by
the
lack
of
culture
systems
that
produce
them
rapidly
in
physiological
amounts
but
also
by
the
lack
of
quantitative
data
on
the
matrix
components.
There
is
little
information
in
the
literature
as
to
the
actual
quantities
of
material
secreted
by
cultured
cells,
and
no
studies
have
been
reported
on
the
biochemical
analysis
of
a
mixed
matrix
produced
in
vitro.
In
this
report
we
describe
a
culture
system
in
which
large
amounts
of
an
extracellular
matrix
are
formed
in
a
relatively
short
time.
Smooth
muscle
cells
cultured
from
rat
heart
form
a
multilayered
structure
containing
milligram
quantities
of
connective
tissue
proteins
in
a
crosslinked,
insoluble
form
within
2
weeks,
even
after
many
generations
in
culture.
These
proteins
remain
firmly
anchored
to
the
culture
dish
after
removal
of
the
cells
with
detergent,
allowing
their
quantitation
and
analysis
by
enzymatic
techniques.
We
therefore
studied
the
production
of
all
of
the
insoluble
matrix
proteins
in
one
culture
system
and
the
relationships
among
the
components.
MATERIALS
AND
METHODS
Heart
Cell
Cultures.
Twelve
hearts
obtained
from
3-day-old
rats
were
excised
and
cut
into
small
pieces
with
a
pair of
scissors.
These
pieces
were
washed
twice
in
calcium-
and
magnesium-
free
phosphate-buffered
saline
and
then
trypsinized
at
room
temperature
in
freshly
prepared
0.1%
trypsin
(1:250;
Difco).
The
first
harvest
of
cells,
obtained
after
10
min,
was
discarded.
The
next
four
successive
30-min
harvests
were
collected
and
the
cells
were
obtained
by
centrifugation.
They
were
then
cultured
in
Eagle's
minimal
essential
medium
containing
10%
calf
serum,
10%
tryptose
phosphate
broth
(Difco),
penicillin
(100
units/ml),
and
streptomycin
(100
,gg/ml).
The
cells
were
seeded
into
75-cm2
tissue
culture
flasks
(Falcon)
at
4-5
X
106
cells
per
flask
and
incubated
in
a
CO2
incubator
(95%
air/5%
C02)
at
370C
for
90
min.
The
supernatant
medium
was
then
poured
off,
the
attached
cells
were
washed
once,
and
fresh
medium
was
added.
The
cultures
were
passaged
shortly
before
confluence
at
a
ratio
of
1:4,
and
the
secondary
cultures
were
trypsinized
and
frozen
in
liquid
nitrogen
at
approximately
1.5
X
106
cells
per
2-ml
vial
in
medium
containing
10%
dimethyl
sulfoxide.
Each
vial
was
used
to
establish
one
75-cm2
flask,
and
the
cells
reached
confluence
6
days
after
seeding.
Growth
Curve
Analysis
and
Production
of
Matrix.
In
all
experiments
in
which
the
behavior
of
the
cells
was
being
studied
or
radioactive
matrix
was
being
prepared,
cultures
received
daily
additions
of
50
,g
of
ascorbic
acid
(Merck,
Darmstadt,
Germany)
per
ml;
this
was
added
as
a
100-fold
concentrated
stock
solution.
Cultures
were
initiated
at
105
cells
per
35-mm
culture
dish
(Corning)
and
the
medium
was
changed
every
2
days
thereafter.
The
cell
number
was
determined
in
a
Coulter
Counter
after
dispersion
of
the
cells
with
0.25%
Viokase
con-
taining
0.05%
collagenase.
Trypsin
was
not
adequate
for
this
task
because
the
cells
became
enmeshed
in
the
extracellular
matrix
which
was
only
partially
digested
by
trypsin.
When
production
of
radioactively
labeled
matrix
was
re-
quired,
isotopes
were
added
to
culture
medium
at
a
level
of
1
,uCi/ml
at
a
stage
when
matrix
started
to
appear
(days
6-7
of
culture).
The
isotopes
used
were
the
following:
L-(3,4(n)-
Abbreviations:
NaDodSO4,
sodium
dodecyl
sulfate;
BAPN,
0-amino-
propionitrile.
*
Present
address:
Division
of
Hematology-Oncology,
Childrens
Hospital
of
Los
Angeles,
4650
Sunset
Boulevard,
Los
Angeles,
CA
90027.
t
Present
address:
Department
of
Medical
Biochemistry,
Medical
School,
University
of
Cape
Town,
Observatory,
Cape
Town
7935,
South
Africa.
353
The
publication
costs
of
this
article
were
defrayed
in
part
by
page
charge
payment.
This
article
must
therefore
be
hereby
marked
"ad-
vertisement"
in
accordance
with
18
U.
S.
C.
§1734
solely
to
indicate
this
fact.
Proc.
Nat!.
Acad.
Sci.
USA
76
(1979)
[3H]proline,
30
Ci/mmol;
L-[1-3H]fucose,
3
Ci/mmol;
L-
[2,3,4-3H]valine,
15
Ci/mmol;
L-1[aS]cystine,
40
Ci/mmol;
L-[U-14C]valine,
225
mCi/mmol.
The
amounts
of
matrix
produced
were
determined
in
dishes
in
which
the
cellular
components
were
dissolved
with
sodium
dodecyl
sulfate
(NaDodSO4).
After
the
medium
was
removed,
the
cultures
were
washed
once
with
water
and
allowed
to
stand
for
30
min
with
1%
NaDodSO4.
The
first
wash
was
discarded,
and
the
dishes
then
were
treated
for
a
further
2-6
hr
with
the
detergent.
The
dishes
were
then
washed
with
a
stream
of
dis-
tilled
water,
rinsed
four
times
with
70%
ethanol,
and
allowed
to
dry.
The
amount
of
protein
present
on
the
dish
was
deter-
mined
by
the
method
of
Lowry
et
al.
(12)
after
overnight
dis-
solution
of
the
matrix
in
1
ml
of
2
M
NaOH
at
370C
in
a
humid
environment.
A
standard
curve
was
prepared
for
this
purpose
from
a
1:1
mixture
of
rat
tail
collagen
and
bovine
elastin
(Sigma)
dissolved
in
2
M
NaOH.
Alternatively,
the
protein
on
the
dish
was
stained
directly
with
Coomassie
brilliant
blue
R250
(Sigma).
Isolation
of
Sublines.
Fourth-passage
cells
were
plated
into
60-mm
dishes
(1-3
X
103
cells
per
dish)
and
allowed
to
grow
as
isolated
colonies
for
12-14
days.
The
plating
efficiency
was
approximately
3.5%,
and
70%
of
the
colonies
present
were
ac-
tively
secreting
matrix.
Discrete
colonies,
containing
vigorously
growing
cells,
were
then
ring-isolated,
trypsinized,
and
cultured
in
35-mm
dishes.
Two
sublines
were
grown
to
confluence
and
frozen
in
liquid
nitrogen
for
further
studies.
Enzymatic
Hydrolysis
of
Labeled
Matrix.
Dried,
ra-
dioactively
labeled
matrices
were
subjected
to
proteolytic
di-
gestion
by
trypsin
(Sigma
type
III,
pretreated
with
5
mg
of
bovine
elastin
per
ml
to
adsorb
contaminating
elastase)
elastase
(Worthington,
ESFF),
and
collagenase
(Worthington,
CLSPA).
The
enzymes
were
all
used
at
concentrations
of
10
Mg/ml
in
0.1
M
Tris.HCI,
pH
7.6/10
mM
CaC12.
The
progress
of
digestion
was
followed
by
removing
100
,gl
samples
at
various
times
and
determining
the
released
radioactivity
in
5
ml
of
Instagel
(Packard).
All
liquid
was
decanted
at
the
end
of
the
enzymatic
digestions,
and
the
dishes
were
washed
and
treated
with
2
M
NaOH
as
described
above.
This
solution
was
neutralized
and
the
radioactivity
in
200-IAI
samples
was
determined
as
above.
Electrophoresis.
The
electrophoretic
system
described
by
Porzio
and
Pearson
(13)
was
used
to
analyze
solubilized
matrix
components.
Gels
were
calibrated
by
using
a
preparation
of
myofibrillar
proteins
obtained
from
rat
heart
(13).
RESULTS
Growth
of
Cells
and
Production
of
Matrix.
The
primary
cultures
were
subjected
to
a
selection
procedure
to
eliminate
cardiomyocytes
which
take
longer
than
other
cell
types
to
attach
to
plastic
(14).
This
was
accomplished
by
washing
the
primary
cultures
90
min
after
plating;
subsequently,
little
or
no
beating
activity
was
seen.
Both
polygonal
and
fibroblastic
cells
were
present
in
the
cultures.
Electron
microscopy
showed
that
the
majority
of
the
adherent
cells
had
the
typical
appearance
of
smooth
muscle
cells
(Fig.
1).
The
cells
grew
as
tightly
packed
multilayers,
and
considerable
extracellular
material
including
collagen
fibrils
and
smooth
muscle
basal
laminae
was
often
apparent.
The
detergent-insoluble
matrix
had
a
three-dimen-
sional
network
appearance
(Fig.
2).
The
growth
rate
of
and
production
of
insoluble
extracellular
matrix
by
fourth-passage
cells
are
shown
in
Fig.
3.
The
cells
grew
with
a
doubling
time
of
approximately
48
hr
and formed
multilayered
structures
containing
1.5
X
106
cells
per
35-mm
culture
dish.
No
change
in
morphologic
appearance
consistent
with
fibroblast
overgrowth
was
observed,
and
the
cells
main-
tained
a
polygonal
shape
throughout
the
experiment.
Extra-
FIG.
1.
Rat
smooth
muscle
multilayer
showing
typical
smooth
muscle
morphology.
The
section
was
cut
at
right
angles
to
the
plane
of
the
culture
dish.
(X5000.)
cellular
material
began
to
appear
6-7
days
after
seeding
and
subsequently
rapidly
increased
in
quantity.
Because
as
much
as
33%
of
the
total
protein
in
the
layer
was
matrix
material,
the
cells
could
not
be
dissociated
by
trypsin
alone,
and
a
mixture
of
Viokase
and
collagenase
was
required
to
dissolve
the
structure
and
release
the
cells.
In
other
experiments,
up
to
3
mg
of
matrix
proteins
was
formed
per
35-mm
culture
dish
when
cultures
were
maintained
for
10
weeks.
The
deposition
of
insoluble
matrix
could
be
reversibly
inhibited
by
the
addition
to
the
culture
medium
of
the
lathyrogen
f3-aminopropionitrile
(BAPN),
a
compound
that
prevents
the
crosslinking of
con-
nective
tissue
proteins
(15).
The
composition
of
the
insoluble
matrix
was
probed
by
biosynthetic
labeling
with
radioactive
fucose,
cystine,
proline,
or
valine
followed
by
sequential
enzymatic
digestion.
Fucose
and
cystine
were
selected
for
their
abilities
to
preferentially
label
the
glycoprotein(s)
components
[these
two
compounds
are
t'ifw
-_
..
,
A
.NW
_
W
A
EW
~ ~
INd:~.
---
w
OF
*4
Adw
bI,
FIG.
2.
Scanning
electron
micrograph
of
NaDodSO4-insoluble
extracellular
matrix
produced
by
cultured
smooth
muscle
cells.
(X8500;
bar
=
1
Am.)
%0)
05-
4
Cell
Biology:
Jones
et
al.
Proc.
Nati.
Acad.
Sci.
USA
76
(1979)
355
1.0
5
10
~~~~~~~~~~~~~CL
~0.
0
5
1
0
Days
after
plating
FIG.
3.
Growth
curves
and
production
of
NaDodSO4-insoluble
matrix
proteins
by
rat
smooth
muscle
cells.
Cultures
were
initiated
in
35-mm
dishes
with
105
cells
each;
ascorbic
acid
(50
,ug/ml)
was
added
daily,
and
medium
was
changed
every
second
day.
Solid
sym-
bols,
cell
number;
open
symbols,
amounts
of
NaDodSO4-insoluble
material.
0,
0,
Control;
A,
*,
f,-aminopropionitrile
(BAPN)
(50
'gg/ml)
added
on
day
5;
o,
*,
BAPN
added
on
day
5
and
removed
on
day
9.
present
in
glycoproteins
but
largely
absent
from
elastin
and
collagen
(8,
16)].
Proline
was
chosen
as
a
label
for
all
compo-
nents,
and
valine
is
10
times
more
prevalent
in
elastin
than
in
collagen
(7,
17).
Because
crosslinked
elastin
is
not
hydrolyzed
by
trypsin,
but
elastase
digests
proteins
other
than
elastin
(18),
the
sequence
of
enzyme
treatment
was
always
trypsin
followed
by
elastase.
Collagenase
digestion
was
the
final
treatment
used
because
collagen
is
insensitive
to
most
proteases
(19)
and
its
digestion
by
this
enzyme
was
retarded
by
the
presence
of
the
other
matrix
compounds.
Trypsin
rapidly
solubilized
93%
of
the
fucose
and
69%
of
the
cysteine
radioactivity
from
the
matrix
(Fig.
4a
and
b).
In
con-
trast,
only
52%
of
the
proline
and
59%
of
the
valine
radioactivity
were
released
from
duplicate
matrices
by
the
same
treatment.
These
results
indicate
that
trypsin
preferentially
hydrolyses
fucose-labeled
glycoprotein
components
of
the
matrix.
It
is
also
likely
that
trypsin
removes
noncrosslinked
precursor
molecules
which
are
known
to
be
protease
sensitive
(20).
Elastase
treatment
released
a
further
32%
of
the
proline-
labeled
and
35%
of
the
valine-labeled
matrix
components
(Fig.
4c
and
d)
but
only
5%
of
the fucose
and
17%
of
the
cysteine
label.
This
suggests
that
the
trypsin-insensitive
material,
which
Table
1.
Solubility
of
extracellular
matrix
in
various
solvents
%
of
total
[3H]proline
solubilized
Solvent
6-day
cultures
13-day
cultures
1%
NaDodSO4
16
7
10
M
urea
7
2
8
M
urea
+
1%
NaDodSO4
+
1%
mercaptoethanol
28
9
5
M
guanidinium
chloride
+
10
mM
EDTA
+
1%
mercaptoethanol
25
11
[3H]Proline-labeled
matrices
were
prepared
by
detergent
treatment
of
6-day
or
13-day-old
cultures
of
fifth-passage
smooth
muscle
cells.
The
percentage
of
the
total
radioactivity
present
in
the
matrix
that
was
solubilized
by
the
indicated
solvents
in
5
days
at
371C
was
then
determined.
was
rapidly
solubilized
by
elastase,
represents
crosslinked
elastin.
The
fibrillar
material
remaining
after
trypsin
and
el-
astase
treatments
was
solubilized
by
purified
bacterial
colla-
genase
and,
because
it
contained
virtually
none
of
the
incor-
porated
fucose,
cysteine,
or
valine,
it is
therefore
collagen.
More
recent
experiments
have
shown
that
the
presence
of
collagen
and
hydroxyproline
in
the
matrix
is
absolutely
dependent
upon
the
addition
of
ascorbic
acid
to
the
growth
medium.
Further-
more,
the
percentage
of
collagen
can
be
readily
modulated
by
varying
the
ascorbic
acid
concentration
(unpublished
data).
Amino
acid
analyses
of
the
material
solubilized
by
sequential
trypsin,
elastase,
and
collagenase
treatments
were
in
good
agreement
with
published
values
for
the
microfibrillar
protein,
elastin,
and
collagen,
respectively
(unpublisied
data).
These
experiments
therefore
suggest
that
the
matrix
contained
at
least
three
components
including
glycoprotein(s),
elastin,
and
col-
lagen.
The
matrix
was
highly
insoluble
in
various
denaturing
sol-
vents
(Table
1).
The
solubility
in
all
solvents
decreased
with
increasing
maturity
of
the
cultures,
an
observation
consistent
with
an
increase
in
the
degree
of
crosslinking
with
time.
The
material
solubilized
by
the
urea/NaDodSO4
mixture
from
matrices
produced
after
6
days
in
culture
contained
three
major
polypeptide
species
when
analyzed
by
polyacrylamide
gel
3
6 9
3
6
9
Time,
h
3
6
9
3
6
9
FIG.
4.
Sequential
release
of
radioactivity
from
NaDodSO4-
insoluble
matrix
by
trypsin
(*-*),
pancreatic
elastase
(0-0),
and
bacterial
collagenase
(r-u).
Dishes
containing
labeled
matrix
were
incubated
in
0.1
M
Tris-HCl,
pH
7.6/10
mM
CaCl2
and
the
indicated
enzyme
(10 Jg/ml)
at
37°C.
Samples
were
withdrawn
at
the
indicated
times
for
radioactivity
determinations.
The
replacement
of
one
en-
zyme
for
another
after
the
3-hr
treatment
time
was
carried
out
after
rinsing
with
water.
(a)
Matrix
labeled
with
[3H]fucose;
(b)
matrix
labeled
with
[35S]
cysteine;
(c)
matrix
labeled
with
[3H]proline;
(d)
matrix
labeled
with
[3H]valine.
I
In0
10
20
30
40
50
60
Distance
from
origin,
mm
FIG.
5.
Polyacrylamide
gel
analysis
of
material
solubilized
by
8
M
urea/1%
NaDodSO41%
mercaptoethanol/0.5
M
Tris-glycine
buffer,
pH
8.8
from
matrix
produced
by
cells
after
6
days
in
culture.
Gels
stained
with
Coomassie
blue
were
scanned
on
a
Unicam
SP
1700
ap-
paratus
at
550
nm.
w
10
0
-W
%._
0
.0
q
m
6
0
._
o
,D
*0
a
b
c
d
Cell
Biology:
Jones
et
al.
Proc.
Natl.
Acad.
Sci.
USA
76
(1979)
Table
2.
Analysis
of
matrices
produced
by
smooth
muscle
sublines
%
of
total
radioactivity
solubilized
by
Subline
Passage
Trypsin
Elastase
Collagenase
R22CID
8
39
38
23
20 44
16
37
R22CIF
8
19
28
51
20
13
2
82
Sublines
derived
from
mass
cultures
of
rat
smooth
muscle
cells
were
grown
in
the
presence
of
[3H]proline
and
daily
supplements
of
as-
corbate.
The
matrices
produced
by
the
cells
were
then
analyzed
by
sequential
enzyme
hydrolysis.
electrophoresis
(Fig.
5).
Band
A,
which
was
sensitive
to
trypsin
digestion,
had
an
apparent
molecular
weight
of
250,000
and
contained
all
of
the incorporated
[3H]fucose
radioactivity
when
isolated
from
fucose-labeled
matrices.
Band
B,
molecular
weight
72,000,
contained
high
levels
of
[3H]proline
and
[14C]valine
when
isolated
from
cells
cultured
in
the
presence
of
these
two
isotopes
(7).
This
was
also
found
to
be
true
for
band
C,
molecular
weight
45,000.
A
number
of
less-well-defined
bands,
which
contained
high
levels
of
[3H]proline,
were
ap-
parent
between
bands
A
and
B,
with
molecular
weights
con-
sistent
with
those
reported
for
the
a
and
ft
chains
of
collagen
(5).
Analysis
of
the
matrices
produced
by
two
sublines
isolated
from
the
parental
cultures
was
undertaken
to
determine
whether
individual
cell
types
in
the
mass
cultures
secreted
all
components
of
the
matrix
or
whether
the
constituent
proteins
were
synthesized
by
different
cell
types
in
the
mixed
cultures.
Two
such
sublines
isolated
from
discrete
colonies
were
found
to
secrete
glycoprotein(s),
elastin,
and
collagen,
and
the
per-
centage
of
collagen
synthesized
increased
with
increasing
passage
level
(Table
2).
The
results
are
unlikely
to
be
due
to
overgrowth
of
one
cell
type
by
another
because
no
morphologic
changes
were
observed.
Fibroblast
contamination
was
also
unlikely
because
it
was
difficult
to
isolate,
from
the
parent
cultures,
colonies
that
grew
with
a
fibroblastic
appearance.
Not
all
of
the
sublines
tested
demonstrated
increases
in
the
collagen
component
with
increasing
passage
level,
but
no
diminution
in
the
total
amount
of
matrix
synthesized
by
the
cultures
at
high
passage
was
observed.
For
example,
one
of
the
parental
cell
strains
(R9)
secreted
400
,ug
of
matrix
per
35-mm
dish
at
passage
4
and
930
;zg
at
passage
90.
DISCUSSION
Any
definition
of
differentiation
must
include
not
only
the
quality
of
gene
expression
but
also
the
quantity.
This
has
be-
come
particularly
apparent
with
the
realization
that
certain
proteins
once
thought
to
be
the
products
of
specific
cells
(e.g.,
myosin
and
collagen)
are
synthesized
by
a
wide
variety
of
di-
verse
cell
types.
It
therefore
follows
that
culture
conditions
should
be
such
that
differentiated
cells
can
express
their
phe-
notype
to
an
extent
comparable
to
the
in
vivo
situation
(1).
The
smooth
muscle
cells
derived
from
rat
heart
synthesize
and
process
large
amounts
of
connective
tissue
proteins.
Al-
though
we
are
unable
to
relate
the
quantities
of
protein
pro-
duced
to
the
situation
in
the
intact
animal,
the
composition
of
the
matrix
is
similar
to
that
reported
for
the
aortic
media
(7).
The
matrix
materials
and
cells
actually
formed
a
"tissue"
in
culture
so
that
the
structure
was
resistant
to
trypsin
and
could
only
be
disaggregated
by
elastase
and
collagenase.
From
an
experimental
standpoint,
it
was
also
important
that
the
matrix
proteins
remained
firmly
attached
to
the
culture
dish
after
removal
of
the
cellular
material
with
detergent.
Two
other
properties
of
the
rat
heart
smooth
muscle
cell
cultures
commend
them
as
a
valuable
experimental
resource:
these
cells
main-
tained
their
differentiated
features
for
more
than
90
passages,
and
sublines
derived
from
individual
colonies
could
be
obtained
with
relative
ease.
These
features
have
enabled
us
to
obtain
quantitative
results
and
perform
compositional
analyses
on
the
complete
extracellular
matrix
elaborated
by
cultured
smooth
muscle
cells.
The
rat
system
therefore
has
many
advantages
over
other
smooth
muscle
culture
systems
currently
in
use
(21-24).
The
glycoprotein(s)
component
of
the
matrix
with
a
molec-
ular
weight
of
250,000
was
trypsin-sensitive
and
shares
many
of
the
properties
of
the
microfibrillar
protein
(8)
and
fibronectin
(9-11).
This
protein(s)
probably
plays
a
major
role
in
the
or-
ganization
and
structure
of
the
matrix.
The
elastin
component
is
highly
crosslinked
as
shown
by
its
resistance
to
denaturing
solvents
and
also
to
trypsin.
It
is
likely
that
the
protein
species
of
molecular
weight
72,000,
which
was
extractable
from
the
matrix
and
contained
a
high
ratio
of
valine
to
proline,
is
tro-
poelastin
(7).
The
collagen
in
the
matrix
contained
little
cysteine
or
valine,
was
resistant
to
trypsin
and
elastase,
and
was
highly
susceptible
to
purified
bacterial
collagenase.
Banded
collagen
fibrils
have
been
seen
in
electron
micrographs.
The
recent
re-
sults
of
Mayne
et
al.
(24)
show
that
monkey
smooth
muscle
cells
synthesize
types
I
and
III
collagen.
It
is
also
possible
that
the
peptide
with
molecular
weight
45,000
is
identical
to
the
CP45
species
secreted
by
monkey
cells
(24).
Chen
et
al.
(25)
identified
actin
in
the
matrix
prepared
from
chicken
fibroblasts
by
treatment
with
a
nonionic
detergent,
and
it
is
therefore
possible
that
actin
is
also
present
in
the
smooth
muscle
matrices.
The
extreme
insolubility
of
the
mature
matrix
in
strongly
denaturing
solvents
argues
against
any
large-scale
contamination
by
cellular
proteins,
particularly
because
no
protein
bands
could
be
resolved
by
gel
analysis
of
extracts
of
12
to
14-day-old
matrices.
Experiments
with
[14C]thymidine
(not
shown)
demonstrated
that
<19%
of
the
cellular
DNA
re-
mained
associated
with-the
matrix
(i.e.,
<2
tig
of
DNA
per
400
Aig
of
matrix
proteins).
Although
we
have
evidence
that
the
cells
produce
sulfated
proteoglycans,
these
molecules
do
not
remain
in
the
matrix
preparation
used
here.
The
use
of
an
ionic
de-
tergent
such
as
NaDodSO4
may
therefore
allow
the
preparation
of
a
more
insoluble
matrix
than
that
prepared
with
nonionic
detergents
(25,
26).
The
interaction
of
the
matrix
proteins
with
each
other
and
with
cells
is
of
fundamental
importance
to
the
properties
of
connective
tissue.
It
is
therefore
significant
that
sublines
of
smooth
muscle
cells
elaborated
both
the
elastin
and
collagen
components
of
the
matrix
and
that
the
composition
of
the
matrix
changed
in
certain
cultures
with
repeated
subculturing.
This
latter
observation
is
of
particular
significance
in
view
of
the
fact
that
the
extracellular
material
produced
in
athero-
sclerotic
plaques
is
mainly
collagen,
whereas
elastin
is
the
major
component
of
the
aortic
media
(27).
The
changes
in
collagen/
elastin
ratio
may
mimic
the
pathological
situation
and
it
will
be
of
interest
to
define
conditions
that
alter
the
relative
ratios
of
these
proteins.
It
is
also
interesting
that
the
cells
secrete
such
large
amounts
of
collagen
at
high
passage
in
contrast
to
other
cultured
cells
which
rapidly
lose
their
differentiated
properties
in
vitro
(1).
The
availability
of
a
matrix
radioactively
labeled
in
its
spe-
cific
components
has
made
it
possible
for
us
not
only
to
inves-
tigate
the
biosynthesis
of
all
the
insoluble
proteins
in
one
system
but
also
to
study
the
degradation
of
the
matrix
by
hydrolytic
enzymes
produced
by
cultured
cells.
We
have
been
able
to
356
Cell
Biology:
Jones
et
al.
Proc.
Natl.
Acad.
Sci.
USA
76
(1979)
357
demonstrate
the
breakdown
of
the
matrix
by
activated
mac-
rophages
and
malignant
cells.
These
studies
have
a
direct
bearing
on
inflammation,
tissue
remodeling,
and
tumor
inva-
sion.
Dr.
H.
Neustein
and
Mr.
D.
Sanan
are
thanked
for
their
help
with
the
electron
micrographs.
This
work
was
supported
by
the
National
Cancer
Association
of
South
Africa
and
the
South
African
Medical
Research
Council.
1.
Schwarz,
R.
I.
&
Bissell,
M.
J.
(1977)
Proc.
Natl.
Acad.
Sci.
USA
74,4453-4457.
2.
Peterkofsky,
B.
&
Prather,
W.
B.
(1974)
Cell
3,291-299.
3.
Evans,
C.
A.
&
Peterkofsky,
B.
(1976)
J.
Cell.
Physsol.
89,
355-368.
4.
Bressan,
G.
M.
&
Prockop,
D.
J.
(1977)
Biochemistry
16,
1406-1412.
5.
Miller,
E.
J.
(1976)
Mol.
Cell.
Biochem.
13,
165-192.
6.
Narayanan,
A.
S.,
Sandberg,
L.
B.,
Ross,
R.
&
Layman,
D.
L.
(1976)
J.
Cell
Biol.
68,
411-419.
7.
Rucker,
R.
B.
&
Tinker,
D.
(1977)
Int.
Rev.
Exp.
Pathol.
17,
1-47.
8.
Muir,
L.
W.,
Bornstein,
P.
&
Ross,
R.
(1976)
Eur.
J.
Biochem.
64,
105-114.
9.
Hedman,
K.,
Vaheri,
A.
&
Wartiovaara,
J.
(1978)
J.
Cell
Biol.
76,
748-760.
10.
Engvall,
E.
&
Ruoslahti,
E.
(1977)
Int.
J.
Cancer
20,
1-5.
11.
Pearlstein,
E.
(1976)
Nature
(London)
262,497-500.
12.
Lowry,
0.
H.,
Rosebrough,
N.
J.,
Farr,
A.
L.
&
Randall,
R.
J.
(1951)
J.
Biol.
Chem.
193,
265-275.
13.
Porzio,
M.
A.
&
Pearson,
A.
M.
(1977)
Biochim.
Biophys.
Acta.
490,27-34.
14.
Blondel,
B.,
Roigen,
I.
&
Cheneval,
J.
P.
(1971)
Experimentia
27,
356-358.
15.
Miller,
E.
J.,
Martin,
G.
R.,
Mecca,
C.
E.
&
Piez,
K.
A.
(1965)
J.
Biol
Chem.
240,3623-3627.
16.
Sear,
C.
H.
J.,
Grant,
M.
E.
&
Jackson,
D.
S.
(1977)
Biochem.
J.
168,91-103.
17.
Murphy,
L.,
Marsch,
M.,
Mori,
T.
&
Rosenbloom,
J.
(1972)
FEBS
Lett.
21,
113-117.
18.
Hartley,
B.
S.
&
Shotton,
D.
M.
(1971)
in
The
Enzymes,
ed.
Boyer,
P.
D.
(Academic,
New
York),
Vol.
3.
19.
Harris,
E.
D.
&
Krane,
S.
M.
(1974)
N.
Engl.
J.
Med.
291,
557-563,
605-609,
652-661.
20.
Mecham,
R.
P.
&
Foster,
J.
A.
(1977)
Biochemistry
16,
2825-
2831.
21.
Ross,
R.
(1971)
J.
Cell
Biol.
50,
172-186.
22.
Gimbrone,
M.
A.
&
Cotran,
R.
S.
(1975)
Lab.
Invest.
33,
16-
27.
23.
Foster,
J.
A.,
Mecham,
R.
P.,
Rich,
C.
B.,
Cronin,
M.
F.,
Levine,
A.,
Imberman,
M.
&
Salcedo,
L. L.
(1978)
J.
Biol.
Chem.
253,
2797-2803.
24.
Mayne,
R.,
Vail,
M.
S.
&
Miller,
E.
J.
(1978)
Biochemistry
17,
446-452.
25.
Chen,
L.
B.,
Murray,
A.,
Segal,
R.
A.,
Bushnell,
A.
&
Walsh,
M.
L.
(1978)
Cell
14,377-391.
26.
Birdwell,
C.
R.,
Gospodarowicz,
D.
&
Nicolson,
G.
L.
(1978)
Proc.
Natl.
Acad.
Sci.
USA
75,3273-3277.
27.
Benditt,
E.
P.
(1977)
Am.
J.
Pathol.
86,693-702.
Cell
Biology:
Jones
et
al.
