Download T Gulick, SJ Pieper, MA Murphy, LG Lange and GF

A new method for assessment of cultured cardiac myocyte contractility detects immune
factor-mediated inhibition of beta-adrenergic responses.
T Gulick, S J Pieper, M A Murphy, L G Lange and G F Schreiner
Circulation. 1991;84:313-321
doi: 10.1161/01.CIR.84.1.313
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1991 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/84/1/313
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/
Downloaded from http://circ.ahajournals.org/ by guest on May 1, 2014
313
A New Method for Assessment of Cultured
Cardiac Myocyte Contractility Detects
Immune Factor-Mediated Inhibition of
j3-Adrenergic Responses
Tod Gulick, MD; Stephen J. Pieper, MD; Margaret A. Murphy;
Louis G. Lange, MD, PhD; and George F. Schreiner, MD, PhD
Background. Potentially reversible congestive heart failure accompanies disease states
associated with an immune cell myocardial infiltrate such as cardiac allograft rejection and
inflammatory myocarditis. We therefore examined the hypothesis that immune cells can
produce noncytotoxic alterations in cardiac function.
Methods and Results. A novel system to evaluate cultured cardiac myocyte contractility was
developed using neonatal rat cardiocytes grown on human amniotic membrane segments.
Spontaneous synchronous cell beating produced macroscopic distortion of these membranes.
Movement of free-floating membranes anchored within a perfusion chamber was visualized
under low-power microscopy and measured from recordings of the rhythmic displacement of
membrane-adherent markers. Additions of graded concentrations of isoproterenol to the
perfusate produced up to threefold increases in the initial contractile phase velocity (contractile
index), with an EC50 of 107 M. When the extracellular Ca2+ concentration was increased from
0.9 to 3.6 mM, 2.43-fold increases in this index occurred. Myocytes incubated for 72 hours in
the presence of dilutions of medium conditioned by activated rat splenic macrophages and
lymphocytes exhibited an isoprotcrenol contractile index inhibited by 62% compared with
control cells. In contrast, responses of supernatant-exposed and control cells to increased
extracellular Ca2+ concentrations were not significantly different. Parallel studies of increases
in myocyte intracellular adenosine 3' 5'-cyclic monophosphate concentrations in response to
isoproterenol stimulation demonstrated correlative inhibition that was specific for exposure to
medium conditioned by immune cells.
Conclusions. Thus, a new method of in vitro cardiac contractility assessment that has
significant advantages over existing systems has been developed and characterized. This new
method has enabled description of an inhibitor of cardiac contractile function produced by
activated immune cells. (Circulation 1991;84:313-321)
rofound suppression of cardiac contractile function can accompany inflammatory myocarditis
associated with idiopathic congestive cardiomyopathy12 and cardiac allograft rejection.3 This functional impairment occurs in a setting of infiltration of
the myocardium by activated lymphocytes and macrophages.4,5 Restitution of or improvement in contractile
performance occurs in a significant subset of patients,
and this functional improvement correlates temporally
with resolution of the mononuclear cell infiltrate.1,2,4,6
The histopathologic features of inflammatory cardiac disease are quite variable. To complicate matters, despite recent attempts to reach a consensus for
the diagnosis and classification of myocarditis,7 inter-
From the Department of Medicine (T.G., S.J.P., M.A.M.,
L.G.L.), Cardiovascular Division, Jewish Hospital of St. Louis,
Washington University Medical Center and the Departments of
Medicine and Pathology (G.F.S.), Renal Division, Washington
University School of Medicine, Barnes Hospital, St. Louis.
Supported by Public Health Service Institutional NRSA grant
5-T32-HL07275 (T.G.), Public Health Service Individual NRSA
grant 1-F32-GM12392-01 (T.G.), National Institute on Alcohol
Abuse and Alcoholism grant R01-AA06989-03 (L.G.L.), a grant
from the Alcoholic Beverage Medical Research Foundation
(L.G.L.), an Established Investigatorship from the American
Heart Association (G.F.S.), and a grant from the Communities
Foundation of Texas (G.F.S.).
Address for correspondence: George F. Schreiner, MD,
PhD, Washington University School of Medicine, Department
of Pathology, Box 8118, 660 South Euclid Avenue, St. Louis,
MO 63110.
Received June 29, 1989; revision accepted February 26, 1991.
P
Downloaded from http://circ.ahajournals.org/ by guest on May 1, 2014
314
Circulation Vol 84, No 1 July 1991
pretation of endomyocardial biopsy results is subject
to rather dramatic interobserver variability.8 While
myocyte necrosis and fibrosis are often seen on
histopathologic analysis of myocardial biopsy specimens obtained from inflamed hearts, some investigators consider a focal interstitial cell infiltrate without
myocyte degeneration or necrosis to be sufficient for
diagnosis.9 Support for the latter perspective is derived from several studies in which fibrosis and
myocytolysis were absent or minimal in the vast
majority of cases of myocarditis and idiopathic dilated cardiomyopathy, despite concomitant severe
cardiac dysfunction.10'11 The lack of fibrosis and
myocytolysis and a reversible dysfunction in some
patients with inflammatory heart disease led us to
hypothesize that activated immune cells or their
soluble products can alter cardiac function without
cytotoxicity.
To examine this question, we developed a new
system to evaluate cardiac myocyte contractility using
cells established in monolayer culture on flexible
biological membrane substrata.12 These preparations
exhibited rhythmic motion resulting from the
summed activity of spontaneously and synchronously
beating cells that was readily observed visually. Videotapes of membrane-adherent markers under lowpower inverted microscopy permitted precise tracking and measurement of myocyte-mediated
membrane movements.
Methods
Myocyte Cell Culture
Neonatal rat cardiac myocytes were isolated and
cultured using a modification of standard techniques.13 Briefly, hearts were removed aseptically
from 1-day-old Sprague-Dawley rats and subjected to
digestion with 200 units/ml collagenase (Wako, Dallas,
Tex.) in divalent cation-free isotonic buffer (137 mM
NaCl, 2.7 mM KCl, 8 mM Na2HPO4, 1.5 mM KH2PO4,
and 5.5 mM dextrose) supplemented with 2% (by
volume) fetal bovine serum (FBS) (Hyclone Laboratories Inc., Logan, Utah). Intact viable myocytes were
separated from nonmyocytes by centrifugation
through a Percoll (Pharmacia Diagnostics Inc., Fairfield, N.J.) step gradient, effecting purification to
greater than 95%. Myocytes were washed and suspended in culture medium consisting of Ham's F-12
nutrient solution supplemented with 10% (by volume)
FBS, 10 mM N-[2-hydroxyethyl]piperazine-N'-[2ethanesulfonic acid] (HEPES), 50 ,ug/ml streptomycin, and 50 units/ml penicillin. Cells were plated within
1-cm-diameter cloning rings on amniotic membranes
in 35-mm petri dishes (contractility assay) or in 96well Primaria (Becton Dickinson, Lincoln Park, N.J.)
tissue culture plates (adenosine 3': 5'-cyclic monophosphate [cAMP]assay) and incubated at 37°C in 5%
CO2 and humidified air. Attachment to substrata and
formation of syncytia required 48 hours, after which
experimental manipulations were initiated.
Contractility Assay
Fresh human placental amniotic membrane14 was
separated from the chorion by blunt dissection and
thoroughly rinsed of blood prior to incubation for 24
hours in 4 gm% deoxycholate at 4°C. Epithelial cells
were removed from the underlying basement membrane by scraping with gauze pads, after which the
membrane was washed and stored in Dulbecco's
phosphate buffered saline (PBS) containing penicillin and streptomycin. Membrane pieces (2x2 cm)
were placed stromal side down/basement membrane
side up in 35-mm petri dishes. Isolated myocytes
were then plated within 1-cm-diameter cloning rings
placed on the membranes. Cloning rings were made
by slicing 1Ox75-mm centrifuge tubes with a hot
knife. A synchronously contracting monolayer that
produced macroscopic displacement of the membrane was apparent after 48-72 hours. Following
syncytium formation, there was a window of 4 days
during which the simultaneous contractions of plated
cells produced membrane displacements that could
be easily and accurately measured under low-power
microscopy. Decrements in myocyte-mediated membrane motion that produced this temporal window
appeared to result from a gradual decrease in the
compliance of the membrane accompanying the proliferation of the small population of nonmyocyte
cells.
Following stated incubation conditions, small aliquots of a suspension containing 0.9-,um cationic
fluorescent beads (Covalent Technology, Ann Arbor,
Mich.) were added within the cloning rings and
permitted to adhere to cells and interstices. For
assays, individual membranes were transferred to a
perfusion chamber within a circulating water bath
mounted on a Zeiss inverted microscope stage
(Thornwood, N.Y.). Membranes were tethered on six
wire pins extending up from the bottom of the
perfusion chamber arranged in a 2-cm-diameter circle. The chamber was perfused with modified Krebs
buffer (128 mM NaCI, 4 mM KCI, 0.9 mM CaCI2, 1.0
mM MgCI2, 0.9 mM NaH2PO4, 22 mM HEPES, and
5.5 mM dextrose) supplemented with 1% (by volume) FBS, 1 mM ascorbate, and stated concentrations of isoproterenol or calcium ion (added as
CaCl2). Beads were then visualized under low-power
microscopy and videotaped using a Hamamatsu fluoresence-enhancing video camera (Bridgewater,
N.J.). Time-dependent linear displacement of single
beads was measured directly from a video screen.
Immune Cell Cultures
Activated splenocyte culture supernatants were obtained from cultures of adult rat splenocytes suspended
in medium containing 5 ,ug/ml concanavalin A. Medium was harvested after 24 hours, after which lectin
was removed by batch adsorption with Sephadex
G-25.15 Rat bidirectional primary mixed lymphocyte
cultures were established using splenocytes harvested
from Lewis strain and outbred Sprague-Dawley rats.
Downloaded from http://circ.ahajournals.org/ by guest on May 1, 2014
Gulick et al Immune Factor Inhibition of Cardiac Myocyte Contractility
Cellular activation was confirmed by documenting accelerated proliferation in mixed lymphocyte cultures
compared with syngeneic splenocyte cultures using
[3H]thymidine incorporation rates. Culture supernatants were harvested after 96 hours. All conditioned
media were clarified by centrifugation at 2,000g for 10
minutes, sterile-filtered (0.2 gm), and stored in aliquots
at -20°C. Immune cell cultures were established in
medium identical to that used for myocyte cultures.
Media contained less than 50 pg/ml lipopolysaccharide.
Cardiac Nonmyocyte Cell Culture
Acutely isolated neonatal cardiac cells were plated
in 100-mm petri dishes at a density of 5x105 cells/
cm2. Dishes were washed exhaustively 30 minutes
after plating, at which time selective adherence of
nonmyocyte cells was achieved.16 Medium identical
to that used for myocyte cultures was then added and
harvested after 3 days. Supernatants were clarified by
centrifugation at 2,000g for 10 minutes, sterile filtered, and stored in aliquots at -20°C.
cAMP Assay
Following the incubation of myocytes under stated
control and experimental conditions for indicated
periods, culture medium was aspirated from wells
and replaced with assay buffer consisting of PBS, 10
mM HEPES, 5.5 mM dextrose, 1 mM ascorbate, and
stated concentrations of isoproterenol. After incubation for 10 minutes at 37°C, samples were quenched
and deproteinized with perchloric acid (final concentration 0.6 M) and cooled to 4°C. Aliquots of samples
were neutrilized with KHCO3 and analyzed for
cAMP content by radioimmunoassay.17 Duplicate
assays of quadruplicate well myocyte cAMP concentrations were performed for each experimental condition. Total cellular protein in culture wells was
determined by Bradford assay18 following solubilization with 0.1N NaOH.
Statistics
The significance of differences between control
and immune cell culture supernatant-exposed cellular responses to agonist stimulation were determined
by Student's t test for unpaired data. Data are
reported as mean±SEM unless stated otherwise.
Results
Purification of Cardiac Myocytes by Silica Sol
Gradient Centrifugation
Established methods to purify acutely isolated
cardiac myocytes rely on differential kinetics of cell
attachment to substrata.19'20 Contaminating fibroblasts and endothelial, mesothelial, and smooth muscle cells preferentially adhere to tissue culture substrata after plating such that the nonadherent
myocytes remain in suspension and are proportionately enriched. Using this differential attachment
technique, we were able to consistently obtain myocyte cell suspensions that included 10-20% nonmyo-
100 r
0
315
100
0
80-
80
0
9
0
0
0
60 p
A
.6-
40 5
4010
0
20 F
20 D
Ocp`.
o
p
t04 1.05 1tO6 tO7 1.0B 1.09
Percoll Density (gImL)
Density Gradient Centrifugation
..Ocus4r
,
FIGURE 1. Purification of cardiac myocytes by selective
adherence versus Percoll step gradient centnifugation. Cardiac
cells were obtained from collagenase digestion of minced
neonatal rat hearts. Initial cell suspension containing contaminating cell elements was plated within 60-mm Petri dishes,
with nonadherent cells harvested after 30 minutes (three
cycles), or divided into aliquots and centrifuged through
isotonic divalent cation-free Percoll of stated densities. For all
samples, myocyte purity and yield were determined by hemocytometer counting of trypan blue-excluding cells.
cytes, often with a loss of 40% of the original myocyte
population. Because of our particular interest in a
potential functional myocyte response to immune
factors, we sought to obtain a more homogeneous
population of contracting myocytes.
Acutely isolated cardiac cells were subjected to
density gradient centrifugation through Percoll silica
sol. Preliminary experiments demonstrated that myocytes exhibited a buoyant density of approximately
1.06-1.08 g/ml in isotonic fluid (data not shown).
Subsequently, aliquots of a suspension of cardiac
cells were centrifuged over Percoll, with the density
in different tubes ranging from 1.040 to 1.090 g/ml.
As shown in Figure 1, cells migrating through Percoll
solutions of greater than 1.070 g/ml density were
uniformly 95-99% viable myocytes. Myocyte yield at
the lowest density that effected this degree of purification was 45% of the initial cell number. These data
compare favorably with those obtained by the selective adherence technique, with which only 80% purity
could be achieved. For all studies presented here,
contaminating blood cells were eliminated from the
myocyte suspension by centrifugation over a Percoll
step gradient (1.070 over 1.095 g/ml). Myocytes sedimented to the Percoll interface and segregated from
both low-density nonmyocyte contaminants and highdensity erythrocytes and leukocytes.
Measurement of Myocyte-Mediated Amniotic
Membrane Displacement
In vitro assessment of cultured myocyte contractile
function has been complicated by the necessity to
Downloaded from http://circ.ahajournals.org/ by guest on May 1, 2014
Circulation Vol 84, No 1 July 1991
316
200
,5150
0
E 100
0
0
K~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.Ais 50
Uln A
-0
0 0.8 1.C 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
0.2 0.40 0.6
Time (sec)
200
addition, beat-to-beat temporal variability and displacement variability are minimal. Maximal displacements varied by less than 5% under a given perfusion
condition.
Despite the well-behaved motion of single points
on a given membrane, three factors contributed to
variability in myocyte-mediated membrane motion
between different membranes and between different
points on the same membrane. First, the placental
membrane segments varied in thickness and compliance, and compliance decreased with time after
myocyte syncytium formation. Second, peripheral
tethering necessarily altered the imposed load, and
this could not be experimentally fixed. Third, the
central portion of the circular area on which cells
were plated experienced forces generated by contracting cells on all sides, while forces imposed on the
peripheral regions were more unidirectional. Thus,
we concluded that determinations of contractile
force could not be compared between different membrane segments, even of the same preparation. However, for a single point on a given membrane preparation, a relative change in contractile state in
response to a stimulus could be measured as long as
the load and mass of the system were not altered.
-
150 [_
0
E 100-
0k
.na 50
0.8 1.0
1.2
1.4
1.6
1.8
2.4 2.6
Time (sec)
FIGURE 2. Myocyte-mediated time-dependent movement of
amniotic membrane substratum. Punfied cardiac myocytes
plated on basement membrane side of acellular human amnion segments within cloning rings were permitted to adhere
and form syncytium. After 72 hours, membranes were transferred to and tethered within perfusion chamber maintained at
37 °C on inverted microscope stage. Membrane displacement,
visualized by low-power microscopy of membrane-adherent
microspheres and recorded on videotape, was measured directly from screen for each video frame. Top: Time-dependent
linear displacement of membrane substratum produced by
spontaneously and synchronously contracting cells under basal
conditions during perfusion with modified Krebs buffer. Bottom: Movement of same membrane after addition of 10'7M
isoproterenol to perfusate (A). Results depicted are of single
typical preparation.
relatively small myocyte sarcolemma displacements as an index of cellular contraction.21 23
Because of myocyte attachment to rigid substrata in
these studies, the degree of sarcolemma displacement is highly dependent on the membrane location
being monitored.
To circumvent these methodological difficulties,
we used human placental membrane as a flexible
biological substrate on which to culture myocytes.
The synchronous activity of spontaneously contracting cells caused the circular area on which cells were
plated to contract regularly. For measurement, membranes were transferred to a perfusion chamber for
recording of motion under low-power microscopy.
Linear displacement of areas on the membrane was
visualized readily by tethering several points on the
periphery of the otherwise free-floating membrane.
The time-dependent displacement of one point on a
representative preparation, measured from a videotape, is depicted in Figure 2, top. Of particular note,
maximal membrane displacement is nearly 200 ,um,
equivalent to approximately four cell lengths. In
measure
Myocyte Inotropic Response to
,[3Adrenergic Stimulation
To examine the effect of ,3-adrenergic agonist
stimulation on myocyte contractile responses using
this system, myocyte-mediated membrane motion
was recorded before and after addition of isoproterenol to the perfusate. The motion of a single point on
each membrane was recorded before and after agonist exposure without intervening manipulation of
the membrane to ensure that load was not altered.
As shown in Figure 2, bottom, isoproterenol produced increases in beating rate and maximal displacement.
To quantify changes in contractile velocity, the
average rate of displacement during the initial two
frames of the contractile phase of five consecutive
beats was determined before (vo) and after (vag)
agonist perfusion and expressed as a ratio (vag/vo) to
provide an index of contractile response to the agonist. This two-frame time interval was selected to
maximize the detection of changes in maximal contractile velocity and to reduce error. Comparison of
displacement measurements from longer intervals
underestimated true changes in contractile velocity
because of inflection in the displacement versus time
curves. Subjectivity in assignment of the frame in
each cycle during which contractile motion could first
be detected and variation in the actual duration of
motion in the first video frame, since motion could be
initiated at any time during that frame, constituted
the major sources of error. Determination of displacement over a two-frame interval and averaging
the readings of five consecutive beats reduced the
contribution of these error terms.
Downloaded from http://circ.ahajournals.org/ by guest on May 1, 2014
Gulick et al Immune Factor Inhibition of Cardiac Myocyte Contractility
4
x
3.0 r
x
a)
3
~0
C-0
317
2.5 2.0 F
21
cd
0
0
o
1
1.5 F
1.0
9
8
7
Basal
6
-Log [Isoproterenol] (M)
FIGURE 3. Isoproterenol-mediated increases in cultured cardiac myocyte contractility. Time-dependent cardiac myocytemediated displacement of amniotic membrane segments was
determined under basal conditions and after 5 minutes of
perfusion with isoproterenol. Measurements offive consecutive
beats were averaged under each condition. Ratio of displacement duringfirst two video frames of contraction in each cycle
after isoproterenolperfusion to displacement before isoproterenol perfusion was calculated for each preparation to provide
index of contractile response to agonist (contractile index).
Four different membrane preparations perfused with single
isoproterenol concentration were used to generate data for
each point, depicted as mean ±SEM.
Importantly, the basal contractile state did not
change during the time required for measurements.
Thus, the index did not differ significantly from unity
when measurements were obtained from preparations
perfused without agonist over a 10-minute period.
This confirmed the legitimacy of the contractile index
and ensured that responses to agonist exposure would
not be blunted by decrements in basal contractile
function during the assay period. The index increased
in a monotonic fashion to 3.0 with increasing isoproterenol concentrations, with an EC50 of approximately
i0` M (Figure 3). Somewhat blunted responses were
demonstrated when a single membrane preparation
was exposed to graded increases in isoproterenol
concentration (which required up to 45 minutes)
compared with responses of individual myocyte preparations challenged a single time with one isoproterenol concentration. In the former case, contributions
from desensitization and from decrements in basal
contractile state could not be segregated. The latter
protocol was judged to be more accurate and was used
in subsequent experiments.
Inhibition of 3-Adrenergic Responsiveness by
Activated Splenic Cell Culture Supematants
Myocyte preparations were cultured in the presence of dilutions of medium conditioned by mitogenactivated splenocytes or control medium for 3 days
prior to the assessment of contractile responses. No
gross differences in basal contractions were seen
.
Control Exposea
Control Exposed
Isoproterenol
t [Ca +]
FIGURE 4. Immune factor inhibition of cardiac myocyte
inotropic responses to isoproterenol. Cardiac myocytes established on amniotic membrane segments were incubated with
1: 4 dilution of concanavalin A-activated rat splenocyte culture supematant or control medium for 72 hours. Mean
inotropic responses to these preparations to 10` M isoproterenol and to 3.6 mM extracellular Ca2' are depicted. Values are
mean +SEM of eight determinations for each condition.
between control and supernatant-exposed preparations. In contrast, responses of splenocyte supernatant-exposed myocytes to isoproterenol stimulation
were markedly impaired. At a stimulating isoproterenol concentration of 1 gM, supernatant-exposed
preparations exhibited a 62% inhibited contractile
index of 1.58±0.06 compared with a control response
of 2.54±0.16 (p<0.0005) (Figure 4). To examine
whether these cells were capable of an augmented
contractile state, the response to increased extracellular Ca concentrations was determined for both
control and supernatant-exposed myocytes. Index
changes in response to 3.6 mM Ca21 were not significantly different (2.43±0.16 versus 2.33 ±0.29, control
versus supernatant-exposed, p>0.25). Thus, the impaired response induced by immune cell-conditioned medium appeared at least partially selective
for ,B-adrenergic stimulation.
Myocyte chronotropic responses to adrenergic stimulation were also impaired in cells preincubated with
immune cell supematants. Control cells exhibited a
2.07-fold increase in beating rate when stimulated
with 1 ,M isoproterenol, from 57±7 to 118±21
(mean±SD) beats/min. Under otherwise identical
conditions, myocytes preincubated with a 1: 4 dilution
of lectin-activated splenocyte supernatant exhibited a
1.52-fold increase, from 60+6 to 91±22 beats/min.
The basal beating rates of control and supernatantexposed cells did not differ significantly (p>0.25).
Thus, immune cell culture supernatants inhibited the
myocyte chronotropic response to isoproterenol by
52%, significant at the 0.005 level.
Inhibition of contractile responsiveness was apparent after 3 and 4 days of supernatant exposure.
Because of the 4-day window between syncytium
formation at day 3 and limiting decrements in myocyte-mediated membrane motion at day 7, it was not
Downloaded from http://circ.ahajournals.org/ by guest on May 1, 2014
318
Circulation Vol 84, No 1 July 1991
TABLE 1. Inhibition of Isoproterenol-Stimulated Cardiac Myocyte Intracellular cAMP Concentration Increases by
Conditioned Media
Myocyte cAMP
Concentration
Inhibition (%)
Medium supplement
(mean+SD pmol/mg)
p
117+11
...
None
64
43+2
<0.001
Splenocyte supernatant (20% by volume)
Fibroblast supernatant
>0.25
119+16
...
20% by volume
0.11
...
134+19
50% by volume
cAMP, adenosine 3:5' =cyclic monophosphate.
possible to measure the effect of longer periods of
supernatant exposure. Similarly, these same constraints precluded analysis of reversibility of this
physiological effect.
Correlative Inhibition of f3-Adrenergic
Agonist-Mediated Myocyte cAMP Concentration
Increases by Supematants
Because /3-adrenergic agonists produce enhanced
contractile function via stimulation of adenyl cyclase24 and consequent increases in intracellular
cAMP concentrations, correlative metabolic effects
of immune cell factors on this system in the myocyte
were investigated. Medium conditioned by lectinactivated rat splenic cells inhibited myocyte cAMP
concentration increases in response to J3-adrenergic
stimulation. Isoproterenol increased the control cell
intracellular cAMP concentration from a basal level
of 10 to 145 pmol/mg protein. Supernatant exposure
for 72 hours produced inhibition of cAMP concentration increases by up to 60%, with an EC50 of
10-20%. Detailed analysis of the time-dependence
of this cAMP-suppressive activity revealed that the
effect was first manifest after 24-48 hours, persisted
for up to 7 days, and was reversible within 3 days
upon discontinuation of supernatant exposure.25
To eliminate attribution of this immune cell culture supernatant effect to medium exhaustion or
nutrient depletion, medium harvested from cultures
of cardiac nonmyocytes (fibroblasts and mesothelial,
endothelial, and smooth muscle cells) was tested for
myocyte cAMP-suppressive activity. As shown in
Table 1, myocytes that were susceptible to splenocyte
supernatant-induced inhibition of agonist-mediated
cAMP concentration increases were not affected by
fibroblast-conditioned media. Myocytes incubated in
50% (by volume) fibroblast-conditioned medium had
a 134±19 pmol/mg intracellular cAMP concentration
following stimulation with isoproterenol, which was
not significantly different from the control cell cAMP
concentration (p =0.11). Studies using supernatants
from other cell culture sources indicated that
myocyte cAMP-suppressive bioactivity was produced
exclusively by activated macrophages and/or lymphocytes.26 This bioactivity was immunologically nonrestricted since conditioned media from Lewis, Buffalo,
Brown-Norway, and outbred Sprague-Dawley adult
rats (including mothers of the animals killed for
myocyte preparations) all contained virtually equivalent bioactivities.
Preliminary characterization studies revealed that
the activity was proportionately increased by the
concentration of supernatants by pressurized filtration (Amicon YM-10 membrane [Beverly, Mass.],
10-kDa cutoff). As shown in Figure 5, serum-free
medium conditioned by concanavalin A-activated rat
splenic cells had myocyte cAMP-suppressive activity.
After fivefold concentration, this same medium was
significantly more potent, producing 31% inhibition
at a concentration of 6% by volume compared with
14% inhibition produced by the 10% by volume
unconcentrated supernatant. Thus, activity could be
attributed to a substance with a molecular mass
greater than 10 kDa.
40
^
s
C
< 0
.0
*-
.-c
U1)
0-~
0
2
4
6
8
10
[Supernatant]
(% vIv)
FIGURE 5. Immune factor inhibition of increases in cardiac
myocyte adenosine 3' :5'-cyclic monophosphate (cAMP) concentrations in response to isoproterenol stimulation (IsoStim.). Cardiac myocytes established in monolayer culture in
96-well plates were incubated with control medium or stated
dilutions of unconcentrated (o) orfivefold concentrated (10kDa-cutoff membrane) (A) concanavalin A -activated rat
splenocyte culture supematant for 72 hours. Myocyte intracellular cAMP concentration following stimulation with 10-7 M
isoproterenol was determined by radioimmunoassay for quadruplicate wells for each condition and each was then normalized to wellprotein concentrations. Data were transformed
to reflect inhibition of myocyte cAMP concentration increases
compared with cells incubated in control medium.
Downloaded from http://circ.ahajournals.org/ by guest on May 1, 2014
Gulick et al Immune Factor Inhibition of Cardiac Myocyte Contractility
Discussion
Determination of the contractile state of muscle is
complicated by difficulties in controlling preload and
afterload, particularly in vivo. This is of particular
concern in the assessment of cardiac inotropic response to a mediator that has secondary or other
primary effects on vascular tone and therefore on
load. Efforts to circumvent this have included the in
vitro direct measurement of tension development in
isolated atrial27 and papillary muscle28,29 strips,
where load can be experimentally manipulated and
fixed. While these systems have enabled accurate
assessment of physiological responses, the limited
time of viability of these tissues precludes the assessment of physiological effects that are manifest only
after extended periods of exposure. In addition,
delivery of mediators requires passive diffusion
through numerous nonperfused cell layers.
Physiological responses of isolated cultured cardiac cells have also been examined.21-23 These systems permit monitoring of physiological characteristics over extended periods and have the advantage of
direct delivery of mediators to cells, which exist in
monolayers in culture. However, complex equipment
capable of detecting minute displacements is required. In addition, the movement of a single cell
sarcolemma has been used as an index of cellular
contractile function. Because cells are grown on
fixed, nondeformable substrata, the extent of sarcolemmal displacement is highly dependent on the
sarcolemma location being monitored with respect
to adherent cell borders and adjacent cell attachment sites.
The present method represents an improvement in
existing techniques for the in vitro assessment of
cardiac myocyte inotropic responses. Use of a flexible
biological substratum amplifies the physiological response by reflecting the cumulative contractile activity of numerous beating cells. Thus, a point on the
membrane substratum moves several times the linear
dimension of a single cell, or across approximately
one half of a low-power microscopic field. The system
shares the advantages of an ability to directly deliver
mediators to cells and carries the potential for fixing
load. Unlike established systems in which contractile
responses of cultured cells are determined by measurements of maximal sarcolemmal displacement,
this method permits the determination of contractile
phase velocity.
In inflammatory myocardial diseases, a major subpopulation experiences reversible and often profound congestive heart failure.4,6,9 In early and mild
or moderate disease, histopathologic analysis of myocardium reveals little muscle necrosis or scarring.11'30
Functional reversibility and minimal cell death implicate a noncytotoxic process. A sparse distribution of
mononuclear leukocytes, which are not present in
normal myocardium, exists in inflammatory disease.
These cells or their soluble products are likely candidates as mediators of noncytotoxic cardiac dysfunc-
319
tion. Humoral and cytotoxic cellular immune responses in myocarditis and allograft rejection have
been demonstrated in humans30'31 and animal models,32'33 but there exists no unifying scheme to explain
immune-mediated myocardial injury in the absence
of myocyte necrosis.
These studies demonstrate that a soluble product
of activated immune cells is capable of interfering
with cardiac myocyte /3-adrenergic responsiveness.
Exposure of cells to immune cell soluble factors
significantly decreased contractile velocity and chronotropic responses to isoproterenol stimulation. As a
caveat, it is necessary to note that, since supernatant
exposure resulted in an inhibited chronotropic response to stimulation, it is possible that the observed
inhibition of augmentation of contractile phase velocity was due exclusively to a frequency-force relation and not to a true alteration in contractility.
Definitive evidence would require measurements of
cells paced at a rate exceeding the maximal agoniststimulated rate. Nevertheless, supernatant exposure
clearly interfered with myocyte responsiveness to
adrenergic stimulation.
Sterin-Borda et a127 reported that lectin-activated
human lymphocytes produced positive chronotropic
and inotropic effects on isolated rat atria. That study
differed significantly from our work in several crucial
aspects, in addition to the obvious methodological
differences. In that study, immune cells were coincubated directly with atrial strips, the time course of the
effect was acute (10-50 minutes), the effect was seen
on otherwise unstimulated muscle and ,3-adrenergic
antagonists did not influence the effect, and the
stimulation of inotropy and chronotropy was sensitive
to inhibitors of lipoxygenase and appeared specific
for a subpopulation of T cells. Given these numerous
differences, those results are not in conflict with ours.
The inhibition of physiological responses was associated with a corresponding decrease in adrenergically stimulated intracellular cAMP concentration
increases. Importantly, the basal contractile function
and cAMP concentration of immune-conditioned
myocytes did not differ from that of controls. This is
consistent with previous observations that these cells
do not differ from normal with respect to general
cellular metabolism.26 Temporal limitations of the
system used to measure cultured cell contractile
responses precluded assessment of reversibility of the
immune factor-induced inhibitory effect. However,
parallel studies have demonstrated reversal of myocyte cAMP suppression upon discontinuation of supernatant exposure.25
Since catecholamines are the principal physiological mediators of augmented cardiac contractility,
these observations may have important implications
for understanding the congestive heart failure of
inflammatory heart disease. There are several potential mechanisms by which immune factors might
produce blunted adrenergic responses without affecting basal metabolism and function. We have
shown that ,8-receptor density and affinity for ligands
Downloaded from http://circ.ahajournals.org/ by guest on May 1, 2014
320
Circulation Vol 84, No 1 July 1991
are not altered in myocytes exposed to immune
cell-conditioned media.26,34 However, other possible
explanations include immune factor modulation of
receptor internalization, interference with the formation of functionally active high-affinity receptor-G
protein complexes, alterations in membrane G protein density or stoichiometry, changes in adenyl cyclase activity, or acceleration of cAMP phosphodiesterase activity. The mechanism of immune cytokine
inhibition of ,B-adrenergic agonist stimulation of myocyte intracellular cAMP concentration increases is
the subject of another report.25
Lectin-activated splenocytes release numerous soluble cytokines with myriad effects.1535 Studies designed to purify and characterize the responsible
factor(s) have demonstrated that interleukin-1 and
tumor necrosis factor-a each have activity in suppressing isoproterenol-stimulated increases in intracellular cAMP concentration.34 However, there is
bioactivity in complex immune cell culture supernatants that cannot be attributed to these two defined
cytokinins, and experiments are underway to further
segregate and identify all of the active constituents of
the complex conditioned media. Once complete, this
will permit the exploration of interactions between
effectors vis-a-vis myocyte metabolism, including efforts to determine whether each is a primary or
secondary mediator. Detailed examination of myocyte contractile responses following exposure to each
factor will then be undertaken.
In summary, a novel system to evaluate cultured
contractile cell function was developed and characterized. Positive contractile velocity and chronotropic responses of cardiac myocytes to ,B-adrenergic agonist
stimulation were markedly impaired in cells preincubated in medium conditioned by activated immune
cells. Parallel studies demonstrated correlative reversible supernatant-induced inhibition of isoproterenolstimulated intracellular cAMP concentration increases
in a setting of normal general cellular metabolism. This
inhibitor of cardiac contractile function may have relevance in disease states characterized by impaired contractility, alterations in /3-adrenergic receptor expression and function,36-38 and the presence of leukocytes
within the myocardium.4-11
References
1. Kawai C, Matsumori A, Fujiwara H: Myocarditis and dilated
cardiomyopathy. Annu Rev Med 1987;38:221-239
2. Kopecky SL, Gersh MB: Dilated cardiomyopathy and myocarditis: Natural history, etiology, clinical manifestations, and
management. Curr Probl Cardiol 1987;12:569-647
3. Murdock DK, Collins EG, Lawless CE, Molnar Z, Scanlon PJ,
Pifarre R: Rejection of the transplanted heart. Heart Lung
1987;16:237-245
4. Hammond EH, Menlove RL, Anderson MD: Predictive value
of immunofluorescence and electron microscopic evaluation of
endomyocardial biopsies in the diagnosis and prognosis of
myocarditis and idiopathic dilated cardiomyopathy. Am Heart
J 1987;114:1055-1065
5. Marboe CC, Schierman SW, Rose E, Reemtsma K, Fenoglio
JJ: Characterization of mononuclear cell infiltrates in human
cardiac allografts. Transplant Proc 1984;16:1598-1599
6. Fenoglio JJ, Ursell PC, Kellogg CF, Drusin RE, Weiss MB:
Diagnosis and classification of myocarditis by endomyocardial
biopsy. N Engl J Med 1983;308:12-18
7. Aretz HT, Billingham ME, Edwards WD: Myocarditis: A
histopathologic definition and classification. Am J Cardiovasc
Pathol 1987;1:3-14
8. Shanes JG, Ghali J, Billingham ME: Interobserver variability
in the pathologic interpretation of endomyocardial biopsy
results. Circulation 1987;75:401-405
9. Dec GW, Palacios MD, Fallon JT, Aretz HT, Mills J, Lee
DCS, Johnson RA: Active myocarditis in the spectrum of
acute dilated cardiomyopathies. N Engl J Med 1985;312:
885-890
10. Rose AG, Beck W: Dilated cardiomyopathy: A syndrome of
severe cardiac dysfunction with remarkably few morphological
features of myocardial damage. Histopathology 1985;9:367-379
11. Tazelaar HD, Billingham ME: Leukocyte infiltrates in idiopathic dilated cardiomyopathy. Am J Surg Pathol 1986;10:
405-412
12. Gulick T, Pieper SJ, Chung MK, Lange LG, Schreiner GF: An
immune cell secretory protein inhibits cultured cardiac myocyte
contractility (abstract). Circulation 1988;78(suppl II):II-178
13. Harary I, Farley B: In vitro studies on single beating rat heart
cells. Exp Cell Res 1963;29:451-454
14. Liotta LA, Lee CW, Morakis DJ: New method for preparing
large surfaces of intact human basement membrane for tumor
invasion studies. Cancer Lett 1980;11:141-152
15. Kappler J, Marrack P: Lymphokines, in Weir DM (ed):
Handbook of Experimental Immunology, Volume 2: Cellular
Immunology. Boston, Blackwell Scientific, 1986, pp 57.1-57.23
16. Mohamed SNW, Holmes R, Hartzell CR: A serum-free
chemically defined medium for function and growth of primary
neonatal rat heart cell cultures. In Vitro 1983;6:471-478
17. Tamayo J, Bellorin-Font E, Sicard G, Anderson C, Martin KJ:
Desensitization to parathyroid hormone in the isolated perfused kidney: Reversal of altered receptor-adenylate cyclase
system by guanosine triphosphate in vitro. Endocrinology 1982;
111:1311-1317
18. Bradford MM: A rapid and sensitive method for quantitation
of microgram quantities of protein utilizing the principle of
protein-dye binding. Anal Biochem 1976;72:248-254
19. Claycomb WC: Cultures of cardiac muscle cells in serum-free
media. Exp Cell Res 1980;131:231-236
20. Libby P: Long-term culture of contractile mammalian heart
cells in a defined serum-free medium that limits non-muscle
cell proliferation. JMol Cell Cardiol 1984;16:803-811
21. Marsh JD, Barry WH, Smith TW: Desensitization to the
inotropic effect of isoproterenol in cultured ventricular cells. J
Pharmacol Exp Ther 1982;223:60-67
22. Marsh JD, Lachance D, Kim D: Mechanisms of l8-adrenergic
receptor regulation in cultured chick heart cells. Circ Res
1985;57:171-181
23. Wabanow W, Wollenberger A: Time parameters of contraction of inotropically responding cultured heart cells in relation
to cellular cyclic AMP levels. Life Sci 1982;31:1603-1612
24. Evans DB: Modulation of cAMP: Mechanism for positive inotropic action. J Cardiovasc Pharmacol 1986;8(suppl 9):S22-S28
25. Chung MK, Gulick TS, Rotondo RE, Schreiner GF, Lange
LG: Mechanism of cytokine inhibition of l3-adrenergic agonist
stimulation of cyclic AMP in rat cardiac myocytes: Impairment
of signal transduction. Circ Res 1990;67:753-763
26. Gulick T, Chung MK, Pieper SJ, Schreiner GF, Lange LG:
Immune cytokine inhibition of 83-adrenergic agonist stimulated cyclic AMP generation in cardiac myocytes. Biochem
Biophys Res Commun 1988;150:1-9
27. Sterin-Borda L, Borda E, Fink S, de Bracco ME: Effect of
phytohemagglutinin-stimulated human lymphocytes on isolated rat atria. Arch Pharmacol 1983;324:58-63
28. Murray JJ, Dobson JG, Reed PW: Effects of divalent cation
ionophore A23187 on cardiac contractile parameters. Am J
Physiol 1985;249:H1195-H1203
29. Marban E, Kusuoka H, Yue DT, Weisfeldt ML, Wier WG:
Maximal Ca'+-activated force elicited by tetanization of ferret
papillary muscle and whole heart: Mechanism and character-
Downloaded from http://circ.ahajournals.org/ by guest on May 1, 2014
Gulick et al Immune Factor Inhibition of Cardiac Myocyte Contractility
30.
31.
32.
33.
34.
istics of steady contractile activation in intact myocardium.
Circ Res 1986;59:262-269
Cramer DV: Cardiac transplantation: Immune mechanisms
and alloantigens involved in graft rejection. CRC Crit Rev
Immunol 1987;7:1-30
Maisch B, Deeg P, Liebau G, Kochsiek K: Diagnostic relevance of humoral and cytotoxic immune reactions in primary
and secondary dilated cardiomyopathy. Am J Cardiol 1983;52:
1072-1078
Harkiss GD, Cave P, Brown DL, English TAH: Anti-heart
antibodies in cardiac allograft recipients. Int Arch Allergy
Immunol 1984;73:18-22
Huber S, Lodge P: Coxsackie B-3 myocarditis: Identification of
different pathogenic mechanisms in DBA/2 and Balb/c mice.
Am J Pathol 1986;122:284-291
Gulick T, Chung MK, Pieper SJ, Lange LG, Schreiner GF:
Interleukin-1 and tumor necrosis factor inhibit cardiac myocyte 3-adrenergic responsiveness. Proc Natl Acad Sci U S A
1989;86:6753-6757
321
35. Werb Z, Banda MJ, Takemura R, Gordon S: Secreted proteins of resting and activated macrophages, in Weir DM (ed):
Handbook of Experimental Immunology, Volume 2: Cellular
Immunology. Boston, Blackwell Scientific, 1986, pp 47.1-47.29
36. Fowler MB, Laser JA, Hopkins GL, Minobe BS, Bristow MR:
Assessment of the f-adrenergic receptor pathway in the intact
failing human heart: Progressive receptor down-regulation
and subsensitivity to agonist response. Circulation 1986;74:
1290-1302
37. Limas CJ, Limas C: Beta-adrenergic hormone-receptor interactions in the hypertrophied and failing myocardium. Eur
Heart J 1984;5(suppl F):329-337
38. Lurie KG, Bristow MR, Reitz BA: Increased 83-adrenergic
receptor density in an experimental model of cardiac transplantation. J Thorac Cardiovasc Surg 1983;86:195-201
KEY WORDS * cardiac transplantation * adenosine 3: 5'-cyclic
monophosphate * myocarditis * leukocytes
Downloaded from http://circ.ahajournals.org/ by guest on May 1, 2014