isoproterenol-induced myocardial fibrosis in relation to...

657

Isoproterenol-Induced Myocardial Fibrosis inRelation to Myocyte NecrosisIvor J. Benjamin, Jorge E. Jalil, L.B. Tan, Kathy Cho,

Karl T. Weber, and William A. Clark

Treatment of rats with the ^-adrenergic agonist isoproterenol results in cardiac hypertrophy,myocyte necrosis, and interstitial cell fibrosis. Our objectives in this study have been to examinewhether hypertrophy and fibrosis occur in a compensatory and reparative response to myocyteloss or whether either process may be occurring independently of myocyte loss and thus be areactive response to adrenergic hormone stimulation. We have examined this question byevaluating each of these responses in rats treated with different doses and forms of isoproterenoladministration. Myocyte necrosis was evaluated using in vivo labeling with monoclonalantimyosin for identification of myocytes with permeable sarcolemma, which was indicative ofirreversible injury. Myocardial fibrosis was evaluated by morphometric point counting ofGomori-stained tissue sections and by assessment of the stimulation of fibroblast proliferationby determination of increased levels of DNA synthesis. Stimulation of fibroblast DNA synthesiswas determined from DNA specific radioactivities and radioautography after pulse labeling with[3H]thymidine. The evidence provided by this study suggests that the degree and timing ofmyocardial hypertrophy does not follow the course of myocyte loss and, thus, appears to beeither a response to altered cardiac loading or a reactive response to /3-adrenergic hormonestimulation rather than a compensation for myocyte loss. Myocardial fibrosis, on the otherhand, appears to be more closely related to myocyte necrosis with respect to collagenaccumulation in the same areas of the heart, its dose-response relation to the amount ofisoproterenol administered, and the timing of increased DNA synthesis, or fibroblast prolifer-ation, after myocyte loss. {Circulation Research 1989;657-670)

A increase in collagen content and concentra-tion and an interstitial fibrosis have beenobserved in many instances of myocardial

hypertrophy resulting from various forms of ven-tricular pressure overload.1-9 Myocyte loss hasbeen suggested as a factor influencing the appear-ance of increased collagen concentration in a num-ber of cases,3-10"12 but the evidence has not consis-tently supported myocyte loss as the primarystimulus for increased collagen since there areinstances in which myocyte loss was not judged tobe a factor in the development of fibrosis.5'8'13 Inmyocardial hypertrophy resulting from volumeoverload,2'8-14-15 thyroxine treatment, and exercise

From the Cardiovascular Institute, Michael Reese Hospitaland Medical Center, University of Chicago Pritzker School ofMedicine, Chicago, Illinois.

Supported in part by a grant from the Robert Wood JohnsonFoundation (IJ.B.) and NHLBI grants RO1-HL31701 (K.T.W.)and RO1-HL32040 (W.A.C.). L.B.T. is a recipient of the TravelingFellowship of the Medical Research Council, United Kingdom.

Address for correspondence: Dr. William A. Clark, Depart-ment of Medicine, Northwestern University, Medical School,303 E. Chicago Ave, Chicago, IL 60611.

Received November 11, 1988; accepted February 28, 1989.

training,2-16 myocyte loss has not been observed,and an increase in collagen content occurred only tothe extent that overall myocardial mass increased.Collagen concentration and the interstitial volumefraction remained equal to or less than that ofcontrols.2 In focusing on the instances where colla-gen concentration and the interstitial volume frac-tion is increased above control levels, we havesought to determine if the initial rise in collagenconcentration arises solely as a result of myocyteloss, or alternatively, if a reactive process leads tothis result.17

In this connection, adrenergic hormones and theiranalogues also represent a potent stimulus for bothmyocardial hypertrophy18-23 and fibrosis.24'25 In moststudies26-29 and a recent review,30 significant myo-cyte necrosis occurs with these agents, and thus,hypertrophy and fibrosis have been characterizedas a compensatory response to myocyte loss.25

Although a-adrenergic stimulation has been shownto induce hypertrophy in tissue culture,31-33 wheremyocyte loss is not a factor, it is unclear whether a-and/or 0-adrenergic stimulation in vivo might directlyinfluence myocyte hypertrophy or fibrosis indepen-

by guest on April 19, 2018

http://circres.ahajournals.org/D

ownloaded from

http://circres.ahajournals.org/

658 Circulation Research Vol 65, No 3, September 1989

dently of a reparative component due to myocyteloss.

A major difficulty encountered with regard toevaluation of this question is related to the sensitiv-ity of commonly used methods for direct estimationof myocyte necrosis and fibrosis after treatmentwith either adrenergic hormones or pressure over-load. Thus, it has not been possible to establish thetemporal and spatial relation between the degree offibrosis resulting from these interventions with theextent of myocyte necrosis that preceded it. Inorder to examine this relation more closely, wehave used the technique of in vivo labeling ofnecrotic myocytes with monoclonal antimyosin forevaluation of myocyte necrosis. This determina-tion has been combined with an estimation ofincreased fibroblast proliferation using both theincorporation of [H]thymidine into myocardial DNAand estimation of the volume fraction of interstitialconnective tissue by morphometric point counting.In this manner we have compared both the temporaland spatial relations between myocardial necrosisand fibrosis in isoproterenol-treated rats.

The evidence provided by this study would seemto suggest that the degree' and timing of myocardialhypertrophy is not totally dependent upon myocyteloss, whereas the initial development of interstitialfibrosis is more closely coupled to injury. Thisconclusion is based on the levels of each response(myocyte necrosis, myocardial hypertrophy, andinterstitial fibrosis) that are generated after variousdoses and forms of isoproterenol administration andtheir temporal sequence.

Materials and MethodsExperimental Model

Experimental studies were conducted in 173 maleSprague-Dawley rats (200-400 g) that were main-tained on an ad libitum diet with free access towater. Isoproterenol hemisulfate (Sigma, St. Louis,Missouri) was prepared in isotonic phosphate-buffered saline (PBS) and administered by either oftwo methods: 1) daily subcutaneous injections of50-5,000 /xg/kg body wt in 0.2 ml solution or 2)continuous infusion of 1-22 mg/kg/day using mini-pumps (model 2002, Alza, Palo Alto, California)placed in the peritoneum. When minipumps wereused, isoproterenol was dissolved in PBS contain-ing 0.1% sodium metabisuLfite to prevent oxidation.Pumps were presoaked in sterile saline for 24 hoursbefore implantation to ensure that the drug wasbeing properly delivered. Control rats received 0.2ml saline instead of isoproterenol.

In some cases, rats were pretreated with 5 mg/kgpropranolol (/3-adrenergic antagonist) 1 hour beforeisoproterenol administration. After 3 hours to 10days of isoproterenol, animals were killed by CO2asphyxiation or after methohexital (35-40 mg/kghp.) anesthetic. The thorax was opened, and the

hearts were quickly excised, weighed, and furtherprocessed as described below.

Evaluation of Myocyte Necrosis WithMonoclonal Antimyosin

Myocyte necrosis resulting from isoproterenolwas assessed by in vivo labeling using monoclonalantimyosin. Necrotic myocytes that have devel-oped sarcolemmal permeability allow antimyosin toenter the cell and bind myofibrillar myosin. Intact,uninjured cells exclude the antibody and remainunlabeled.34

Both isoproterenol-treated and control rats weretreated with 1 mg immunoglobulin G (IgG) fractionof monoclonal antibody CCM-52 intraperitoneally24 hours before removal of heart. Animals killedless than 24 hours after isoproterenol treatmentwere pretreated with antimyosin. The antibody usedin this study has been shown to be specific forcardiac myosin.36 Antibody administered intraperi-toneally appears in the serum in less than an hour,reaches a maximal enzyme-linked immunosorbentassay serum titer by 3 hours, and remains at highserum titers for more than a week after a singleinjection (W.A. Clark, unpublished data).

After removal of the heart, 2-mm—thick midven-tricular cross sections were quickly frozen in isopen-tane cooled to the freezing point, -160° C, in liquidnitrogen. After cryostat sectioning of the heart, thesections were stained with fiuorescein isothiocya-nate (FITC)-conjugated rabbit anti-mouse IgG (Cap-pel Laboratories, Cochranville, Pennsylvania) tolocalize myocardial cells that bind antimyosin mono-clonal antibody.34 Immunofluorescent staining wasevaluated with a microscope (Carl Zeiss, Thorn-wood, New York) equipped with epifluorescentoptics optimized for FITC fluorescence.

Morphometric AnalysisA 2-mm-thick coronal section was taken from the

equator of each heart and fixed in neutral bufferedformalin. Formalin-fixed sections were dehydratedthrough a graded series of alcohol and xylene andembedded in paraffin. Five-micrometer paraffin sec-tions were stained with hematoxylin and eosin forevaluation of histology and Gomori trichrome fordistinguishing muscle and interstitial connectivetissue.

Myocardial fibrosis was evaluated in each sectionof heart tissue using a morphometric point-countingprocedure.37 Morphometry was conducted to eval-uate the increase in the subendocardial volumefraction of fibrous tissue in the heart after isopro-terenol treatment. To estimate interstitial volumefraction in the myocardium, cross sections ofGomori-stained tissue were viewed under a 441point-counting grid at xlOO magnification. The frac-tion of grid points over muscle (red staining) andinterstitial (blue staining) tissue was scored in eachfield. Four random fields, selected in each of thefour quadrants of the ventricular cross section,



ownloaded from


Benjamin et al Myocardial Fibrosls and Necrosis 659

were counted in each heart in the zone lying within1.4 mm of the ventricular endocardium. This count-ing protocol provided the greatest sensitivity incomparing fibrosis in different treatment groupssince the greatest degree of fibrosis was confined tothe subendocardium in all groups.

Evaluation of Myocardial DNA SynthesisStimulation of DNA synthesis in heart tissue was

analyzed after 48 hours of isoproterenol treatmentboth by estimation of total incorporation of[3H]thymidine into DNA and by radioautography.[Methyl-3H]thymidine (TRK.686, specific activ-ity>50 Ci/mmol, Amersham, Arlington Heights,Illinois) was administered by intraperitoneal injec-tion (1 ^Ci/g) 4 hours before removal of heart.Thymidine incorporation into DNA was determinedby a modification of the procedure described byMorkin and Ashford. Hearts were quickly excised,weighed, and placed in ice-cold saline. The heartswere divided into right ventricle and left (plusseptum) ventricles, finely minced with scissors, andhomogenized with a polytron (Brinkmann Instru-ments, Westbury, New York) in 2 ml ice-cold PBS.Two millimeters of cold 10% trichloracetic acid(TCA) was added to the homogenate to precipitateDNA and proteins. The tissue homogenates werecentrifuged, washed three times with 4 ml cold 5%TCA, and then given a final rinse of 2% sodiumacetate in 90% ethanol to remove unincorporated[H]thymidine. After centrifugation, the superna-tant was discarded, and the precipitates were resus-pended in 2 ml of 5% TCA. DNA was extracted byheating the samples at 90° C for 30 minutes. Aftercentrifugation, aliquots of the supernatant weretaken for determination of tritium radioactivity byscintillation counting and determination of DNAcontent by the Burton method. DNA standardswere prepared from calf thymus DNA, and theconcentrations of the standard were determined byits absorbance at 260 nm. Stimulation of DNAsynthesis was determined from the specific radioac-tivity (cpm//ig DNA) in each tissue sample.

Identification of cell types involved in DNA syn-thesis was achieved by radioautography. After hema-toxylin and eosin staining of paraffin sections of[JH]thymidine-labeled heart tissue, sections wereair dried from the aqueous phase. Slides were thencoated at 45° C with Kodak NTB-3 nuclear tractemulsion diluted 1:1 with distilled water. Afterdraining and drying, the coated slides were storeddesiccated for 7 days at 4° C. The radioautographswere developed in half-strength Kodak D-ll, fixed,washed, dehydrated, and mounted in permount.

Statistical AnalysisAll quantitative data generated in this study were

analyzed using the SPSS/PC+ statistical package.Data for all groups were compared simultaneouslyby analysis of variance. If a significant F value was

obtained, pairwise comparisons were performed usingthe Scheffe method at specified levels of significance.

ResultsAcute Myocyte Necrosis After a SingleDose of Isoproterenol

In vivo labeling with monoclonal anticardiac myo-sin provides a highly sensitive and specific meansfor characterizing injury resulting from treatmentwith isoproterenol. A single subcutaneous injectionof isoproterenol results in the appearance of necroticmyocytes within hours of treatment. Using mono-clonal antimyosin to assess the development ofmyocyte membrane permeability, we foundantibody-labeled fluorescent myocytes within 3 hours(Figure 1A). The level of antimyosin stainingappeared to be increased at 6-12 hours and maximalat 24 hours after isoproterenol treatment (Figure IBand 1C). During this period, the affected myocyteswere largely confined to the subendocardial regionof the heart. Isoproterenol produced a pattern ofscattered myocyte fluorescence involving isolatedcells and groups of cells with no well-defined zoneof injury (Figure IB and 1C).

At 48 hours after administration of a single doseof isoproterenol, the number of fluorescent necroticmyocytes was considerably decreased over thatobserved at 24 hours (Figure 2), with evidence ofextensive breakdown of the remaining labeled cells(Figure 2, inset). Hearts examined between 3 and 8days after treatment had no evidence of fluorescentmyocytes, and no myocyte labeling of any kind wasever observed in control rats that had been treatedwith monoclonal antibody without isoproterenol(Table 1).

These data were taken to suggest that after sub-cutaneous administration of a single dose of isopro-terenol, there was a rapid appearance of necroticmyocytes in the subendocardium. The cell debriscreated by this process was then cleared by ahealing process that had removed most traces ofnecrotic cells by 48 hours after treatment.

Myocyte Necrosis After Repeated DosingWith Isoproterenol

Repetitive treatment with isoproterenol results indown regulation of adrenergjc receptors with con-siderable attenuation of the effect of subsequentdosing. Since antibody-labeled myocytes were notobserved more than 3 days after injection of a singledose of isoproterenol, we examined whether repet-itive daily dosing with 1,000 /ig/kg isoproterenol for8 days would reduce the level of necrosis observed24 hours after the final treatment in comparisonwith that observed 24 hours after a single dose(Table 1). Substantial evidence of new necrosis wasobserved in rats treated according to this protocol(Figure 3). Unlike the pattern described after asingle dose of isoproterenol in which labeling wasconfined to discrete cells, continued dosing pro-



ownloaded from





ownloaded from


Benjamin et al Myocardial Fibrosis and Necrosis 661

FIGURE 1. Photomicrographs showing monoclonalantimyosin immunofluorescent labeling of necrotic car-diac myocytes in rats treated with isoproterenol andmonoclonal antimyosin. Panel A: Antimyosin labelingafter 3 hours of 1 mglkg s.c. isoproterenoL Magnifica-tion, x625. Panel B: Rat treated for 24 hours with 1mglkg s.c. isoproterenol. Magnification, x160. Panel C:Rat treated with 2 mglkg s.c. isoproterenol for 24 hours.Magnification, X240.

duced a more diffuse staining pattern (Figure 3,inset). Both intracellular and some extracellularstaining was observed, indicating a substantial leak-ing of myosin from injured myocytes. Thus, repeateddosing with isoproterenol appears to induce addi-tional necrosis for at least 8 days in some rats.

Development of Necrosis After ChronicLow-Dose Isoproterenol

A single injection of isoproterenol, even in dosesas low as 125 fig/kg, produces myocyte necrosis. Itwas reasoned that with a bolus injection of isopro-terenol, in which there was a marked increase inheart rate and a drop in systemic pressure, thehemodynamic effect on the myocardium might initself generate enough subendocardial ischemia togenerate myocyte necrosis. 18-39'40 Thus, we exam-

ined whether isoproterenol delivery via osmoticmini pumps, in which acute high levels were notexperienced, might be accomplished without thegeneration of myocyte necrosis. Contrary to expec-tations, antibody-labeled necrotic myocytes weredetected after 48 hours in rats receiving from 1 to 22mg/kg/day isoproterenol (0.2-5 ^.g/min) via osmoticminipumps (Table 1 and Figure 4).

Inhibition of Necrosis With PropranololPretreatment of rats with 5 mg/kg of the /3-

antagonist propranolol 1 hour before a single sub-cutaneous dose of 1 mg/kg isoproterenol substan-tially reduced the level of myocyte necrosis to nearthe detectable limit (Table 1).

Stimulation of Myocardial Hypertrophy:Dose-Response

In addition to the acute response of myocytenecrosis after isoproterenol, there was also an induc-tion of myocardial hypertrophy and fibrosis. Tocompare the relation of these processes with myo-cyte necrosis, we evaluated myocardial hypertro-phy and fibrosis after the same treatment protocolsdescribed above. Statistically significant (p<0.05)hypertrophy was observed within 2 days after treat-ment with 500 ^,g/kg/day s.c. isoproterenol (Table

t. .rnatomicrograpn snowing rat treatea wun l mg/Kg s.c. isoproterenol for 4$ nours. lhe number of labeledmyocytes is decreased over that at 24 hours, but necrotic zones are still evident. Magnification, x250. Inset: Detail innecrotic zone. Magnification, *340.



ownloaded from



TABLE 1. Frequency or Isoproterenol-Induced Necrosis Evalu-ated by In Vivo Labeling With Monoclonal Antlmyosin

Treatment

Myocyte labeling levels

1 2 3

Control:

Single ISO injection(1 mg/kg s.c.)<1 dav

1 days

3-10 days

PROP(5mgs.c.)+ISO(1 mg/kg s.c.)1 day

Daily ISO injections(1 mg/kg s.c.)8 days

ISO infusion with mini pump(1-22 mg/kg/day i.p.)

2 days

1 1 92 2 1

?1

1 1 . . 1

2 7

ISO, isoproterenol; PROP, propranolol. The frequency andlevels of immunofluorescent staining of cardiac myocytes withmonoclonal antimyosin were determined in rats treated withisoproterenol. Categories of fluorescent staining were as follows:0, no detectable fluorescence in any cells; 1, fewer than fivefluorescent positive cells per cross section; 2, a single group offive or more positive cells; 3, two to five clusters of positive cells;4, more than five clusters of positive cells.

2). Treatment with 2,500 /ig/kg/day for 2 daysresulted in an average of 27% myocardial hypertro-phy. Even greater hypertrophy (36%) was observedafter 10 days of daily subcutaneous dosing. Unlikethe results obtained after 2 days, there was noapparent dose-response over the range of 125-5,000/ig/kg/day isoproterenol (Table 3).

Stimulation of Myocardial Fibrosis:Dose-Response to Repeated Isoproterenol

Ten days of treatment with isoproterenol pro-duced marked connective tissue accumulation inthe subendocardium in both the left ventricle andseptum. This effect is well illustrated in the photo-micrograph shown in Figure 5A. Connective tissueaccumulation in this rat treated for 10 days withisoproterenol (1,000 /xg/kg) is clearly evident in thesubendocardium, whereas there is essentially noincreased connective tissue in the subepicardium.Cellular detail in the subendocardial region is shownin greater detail in Figure 5B. Isoproterenol appearsto stimulate both an increase in the fibroblast cellpopulation and the fibrous extracellular matrix.Figure 5C shows the same field as shown in Figure5B with filters adjusted to show collagen fiberstaining. Comparable areas of a control untreatedheart is shown in Figure 5D and 5E.

FIGURE 3. Photomicrograph showing widespread diffuse myocardial staining in rats treated with 1 mg/kg/day s.c.isoproterenol for 8 days. Magnification, x.250. Inset: Detail in necrotic zone. Magnification, x.340.



ownloaded from


Benjamin et al Myocardial Fibros is and Necrosis 663

FIGURE 4. Photomicrograph shdwing myocardial staining in rats with chronic treatment consisting of 11 mglkgldayisoproterenol via osmotic minipump for 2 days. Magnification, X250.

To quantitate isoproterenol stimulation of regionalfibroplasia in the heart, we used morphometricpoint counting. Subendocardial fibrosis averaged8-10 times greater than control levels after treat-ment with 125-5,000 fig/kg isoproterenol for 10

days (Figure 6). Subepicardial fibrosis was notsignificantly greater than control values in any of

3 . Myocardial Hypertrophy Induced by 10 Days of Treat-ment With Subcutaneous Isoproterenol

TABLE 2. Myocardial Hypertrophy Induced by 2ment With Subcutaneous Isoproterenol

Isoproterenol BW (g)

Control 232.0 + 9.6(«-12)

2-Day subcutaneous injection50 jig/kg 257.8 + 0.053

500 Mg/kg 250.0±12.1

(n-4)1,000 A»g/kg 265.7±14.8

2,500 fig/kg 243.0*7.5

(«=5)

HW(g)

0.69 + 0.04

0.85 + 0.05

0.88±0.(»

0.96±0.05*

0.90±0.01*

Values are mean ± SEM. BW, body weight; HWStatistical significance was derived by pairwise co

Days of Treat-

HW/BW(g/kg)

3.02±0.06

3.28 = 0.08

3.48±0.18*

3.62+0.05t

3.82±0.11t

, heart weight.omparison with

Isoproterenol BW(g)

Control 349 0+? 41

10-Day subcutaneous injection125 /ig/kg 354.4±7.2

(»=7)250 t i o n 355.8±4.7

500Mg/Xg

1,000 pgjkg

2,500 ^g/kg(«=5)

5,000 jtg/kg

361.6±7.8

332.0±5.8

353.2±8.6

344.0±4.0

HW(g)

1.19+0.03

1.61 ±0.07*

1.65+0.03*

1.64+0.05*

1.42±0.05

1.71+0.06*

1.74±0.03*

HW/BW(g/Tcg)

3 42 ±0.07

4.54±0.12*

4.65 ±0.06*

4.56+0.15*

4.23±0.13*

4.85 +0.21*

5.05±0.15*

g y p pcontrol values by use of the Scheffe method after analysis ofvariance.

*p<0.05.tp<0.01.

Values arc mean±SEM. Statistical significance was derivedby pairwise comparison with control values by use of the Scheffemethod after analysis of variance.



ownloaded from



, - - •

JFIGURE 5. Photomicrographs of sections of rat heart. Panel A: Gomori-stained rat heart (left ventricle removed) treatedfor 10 days with 1 mglkglday isoproterenol and photographed under a 546 nm interference filter (green). Interstitial cellsin the subendocardium appear light while muscle tissue appears dark. Magnification, x.17. Panel B: Cellular detail insubendocardium under 546 nm filter. Magnification, X175. Panel C: Same field as panel B and photographed underWratten LA filter to show enhanced collagen fiber contrast staining. Magnification, X175. Panel D: Gomori-staineduntreated control rat heart under 546 nm filter. Magnification, xl75. Panel E: Gomori-stained untreated control rat heartunder Wratten LA filter. Magnification, xl75.

the isoproterenol treated groups, as is shown inFigure 5A (morphometric data not shown). Anincrease in the degree of su ben do car dial fibrosiswas noted when the doses of isoproterenol wereincreased from 125 to 5,000 /Ag/kg (Figure 6).

Effect of Single Versus Repetitive Isoproterenolon the Degree of Fibrosis

To determine whether the amount of subendocar-dial fibrosis that developed with the initial treatment

with isoproterenol was enhanced by repetitive doses,we compared the degree of fibrosis that resultedafter either single or multiple doses at differentconcentrations. The results shown in Figure 7 clearlyshow that there is a marked cumulative effectregarding the fibrous tissue formation that is pro-duced by isoproterenol. A single subcutaneous injec-tion of 1,000 /ig/kg generated about 5.5% suben-docardial fibrosis in rats that were examined 10days after treatment. This is approximately twice



ownloaded from


Benjamin et al Myocardlal Fibrosis and Necrosis 665

BOO

Ctrl 5.0.13 .25 .50 1.0 2.5Dose ISO mg/kg BW

FIGURE 6. Bar graph showing effect of dose of isopro-terenol (ISO) on subenaocardial interstitial volume frac-tion after 10 days. BW, body weight. Rats were treatedwith 10 daily injections of the doses of isoproterenolindicated. After staining of ventricular cross sections ofthe heart with Gomori stain, the degree of fibrosis wasestimated by point counting. Statistical difference fromcontrol group is indicated by *p<0.01 and #p<0.05. Thenumber of cases included for each data point is the sameas given in Table 3, with the exception that in the controlgroup, n=8.

the control level of connective tissue. Repeatingthis treatment for 2 days doubles the amount offibrous tissue produced, even in rats examined 24hours after the second treatment rather than at 10days. Considerably greater fibrosis is observed inrats receiving 10 daily doses of isoproterenol. How-ever, rats that received 125 Mg/kg isoproterenolgiven in 10 daily doses had four times the level of

I-Trmt:Repeat:

Duration:

Ctrl 1000l x

lOd

10002x2x

1000lOxlOd

125lOxlOd

I+PplOxlOx

FIGURE 7. Bar graph showing evaluation of repetitivedosing and concentration of isoproterenol (I) on thedegree of interstitial fibrosis in the subendocardium. Ratswere treated with 125 or 1,000 pglkg body wt s.c.isoproterenol that was repeated daily for the times indi-cated. Data is also included for a group treated with 5mglkglday propranolol (Pp) 1 hour before receiving 1mglkglday isoproterenol. Interstitial fibrosis was evalu-ated in the subendocardium by point counting. ^Statisticaldifference from control group, p<0.01. The number ofcases for each group are 8,4,6,5,7, and 6 according totheir order of appearance on graph.

4

u

1)&,

•<

Q

400

300 -

200

100-

jra Daily sc

•Minipump

iCtrl 0.05 0.5 1.0 2.5 5.0 11

mg/kg/day IsoproterenolFIGURE 8. Bar graph showing the effect of dose ofisoproterenol on the stimulation of myocardial DNAsynthesis. Rats were treated with daily subcutaneous (sc)doses of isoproterenoL On the second day of treatment,the animals were pulse-labeled with [3H]thymidine for 4hours. The specific radioactivity of DNA cpm/fig is given.The number of cases in each group are n=4 for control,n=6for sham Alza pump, n=5-7 in the daily subcutane-ous groups, and n-2-3 in minipump groups.

fibrosis as compared with rats receiving 1,000in one dose. Thus, prolonged stimulation with lowerdoses of isoproterenol produces greater fibrosisthan a single higher dose. Development of suben-docardial fibrosis could be blocked by pretreatingthe rats with 5 mg of the /3-antagonist propranolol 1hour before administration of each daily subcutane-ous dose of 1,000 Mg/kg isoproterenol.

Myocardial DNA SynthesisTo better characterize the relation between myo-

cyte necrosis and stimulation of fibrosis during theperiod immediately after isoproterenol treatment,we measured the direct stimulation of DNA synthe-sis in the heart. The advantage of using DNAsynthesis for estimation of stimulation of myocar-dial fibrosis is twofold. First, the difference in DNAsynthesis rates in comparison with controls is greaterthan the difference in percent fibrosis measured bymorphometry. Secondly, peak stimulation of DNAsynthesis rates occurs within 48 hours of treatment.41

Thus, the timing of the peak activation of DNAsynthesis can be more closely compared with theonset of treatment than the development of fibrosis,and the temporal relation between necrosis andfibrosis can be more accurately determined thanthat achieved using morphometry alone.

Rats that were treated with 50-5,000 /tg/kg/dayisoproterenol for 2 days were pulse-labeled with[3H]thymidine for 4 hours before sacrifice. DNAwas extracted from the left ventricular, septal, andright ventricular samples of each heart, and thespecific radioactivity (cpm/Mg DNA) was deter-mined in each sample. As was observed in thedevelopment of fibrosis (Figure 6), stimulation ofmyocardial DNA synthesis appeared to be relatedto the dose of isoproterenol (Figure 8). The incor-



ownloaded from



poration of [3H]thymidine into DNA of the leftventricle and septum was increased more than 10times control values in rats receiving more than 1,000Mg/kg isoproterenol (Figure 8). Stimulation of DNAsynthesis was also observed in the right ventricleafter treatment with 1,000 M-g/kg isoproterenol(mean±SEM, 233±61 cpm/ju,g; «=6;/><0.01) versuscontrol right ventricle although levels were slightlylower than those observed in the left ventricle.

Administration of isoproterenol with minipumps,which presumably prevented the appearance ofacute high serum levels of isoproterenol, did notprevent the appearance of myocyte necrosis, asshown above (Figure 4) nor did it prevent thestimulation of myocardial DNA synthesis (Figure8). Although the same dose of 1 mg/kg/day isopro-terenol produced a considerably higher stimulationof DNA synthesis when given by a single injectionthan by minipump, higher levels of isoproterenoldelivered by minipump produced the same maximalstimulation of DNA synthesis (Figure 8). A propor-tionality was also observed between the level ofnecrosis and DNA synthesis in rats given isopro-terenol by minipump; two animals with the lowestnecrosis score in Table 1 received the lowest dose,1 mg/kg/day, via the minipump.

Radioautographic AnalysisTo further evaluate the regional differences in

the stimulation of DNA synthesis in the myocar-dium, as well as to determine whether this responsewas limited to the fibroblast population, we con-ducted radioautography on paraffin sections of[3H]thymidine-labeled hearts. Virtually all nuclearlabeling in both control and isoproterenol-treatedhearts was limited to the nonmyocyte cells. Agradient of labeling frequency was observed pro-gressing from the subendocardium to subepicar-dium in treated animals (Figure 9). Elevated label-ing indexes were also noted in the right ventricle(Figure 9).

DiscussionAdministration of the /3-adrcncrgic agonist, iso-

proterenol, results in cardiac hypertrophy, myocytenecrosis, fibroblast proliferation, and connectivetissue accumulation. Our objective in this study hasbeen to compare the temporal, spatial, and dose-response relation of the levels of necrosis, fibrosis,and hypertrophy to determine whether fibrosis andhypertrophy occur in reparation of injury and as aconsequence of myocyte loss. Alternatively, if thelevel of either of these responses was independentof myocyte loss, it could be suggestive of a reactiveresponse to adrenergic hormone stimulation.

To characterize the onset, timing, and extent ofmyocardial necrosis after isoproterenol treatment,we have used the technique of in vivo labeling withmonoclonal antimyosin as a sensitive and specificprobe for the development of myocyte necrosis.34

Monoclonal antimyosin is specific for myocytes

with permeable sarcolemma and remains active inthe circulation for more than a week. Rather thanbeing the sole result of a passive entry of the probeinto the injured cell, the antibody is concentrated bybinding to myofibrillar myosin. Thus, antibody label-ing provides a clear objective means to detectnecrotic cells down to a level of a single myocyteper cross section. This technique may also serve torule out myocyte macromolecular permeability incells that might appear abnormal by histologicalcriteria and yet remain unlabeled with antimyosin.

Using this approach, we have characterized thedevelopment of myocyte necrosis after isoprotere-nol treatment with a greater degree of sensitivityand specificity than has heretofore been possible.Our study confirms earlier reports that myocytepermeability develops within a very short periodafter isoproterenol administration.24-29 The level oflabeling rises to a maximum between 12-24 hoursafter treatment and then declines to control levelsby 72 hours. The rapid rise and fall of myocytelabeling suggests that necrosis was produced duringa single short-lived period.

Rona et al,24 using doses of isoproterenol con-siderably greater than those used here (85 mg/kg),reported a rapid appearance of grossly apparentnecrosis within 8 hours of treatment, peak devel-opment by 48 hours, and regression and disappear-ance of necrotic muscle by 72 hours. The signifi-cance of the present study lies in the ability todetect necrosis even earlier and with considerablygreater sensitivity. By identifying each new waveof necrosis after repeated injections or minipumpdosing, we have been able to closely explore thetemporal and spatial relation between myocytenecrosis, subsequent fibrogenesis, and myocardialhypertrophy.

The amount of subendocardial fibrosis presentafter 10 days of isoproterenol appears to be dosedependent. The full extent of fibrosis is not appar-ent by morphometric point counting, however, untilseveral days after treatment. Using incorporation of[3H]thymidine into myocardial DNA as a probe forstimulation of fibroblast proliferation, we were ableto assess the level of this response as early as 24hours after isoproterenol treatment. Maximal stim-ulation of DNA synthesis occurs approximately 48hours after a single dose of isoproterenol.41 Stimu-lation of DNA synthesis at 48 hours also follows anapproximate dose-response relation for the doses ofisoproterenol we used.

Given that neither this study nor previous reportshave been able to totally dissociate a fibroticresponse from the occurrence of myocytenecrosis,25 it is still possible to qualitatively com-pare the dose-response, timing, and response tosingle, pulse, and chronic administration of dif-ferent levels of isoproterenol to evaluate howclosely the necrosis and fibrosis responses arerelated. The findings appear to indicate that thereare at least three correlations between the devel-



ownloaded from




ownloaded from



opment of myocyte necrosis and myocardialfibrosis after isoproterenol administration.

First, myocyte necrosis and fibrosis both developin the same myocardial regions. Maximal myocytenecrosis and myocardial fibrosis are both confinedto the subendocardium and midwall.24-25 No signif-icant scarring developed in the subepicardium, andnecrotic myocytes were not observed in this regionof the heart.

Second, the appearance of necrosis and fibrosisare closely related in time. Necrosis reaches amaximum at 12-24 hours after a single dose ofisoproterenol. Stimulation of fibroblast DNA syn-thesis reaches a maximum at 48 hours.41

Third, there is a similarity in the dose-responserelation of both necrosis and fibrosis. Significantnecrosis and fibrosis were both observed in heartstreated with 125 /ug/kg/day s.c. isoproterenol.Increasing the doses or repetition of isoproterenoltreatment resulted in increased levels of observednecrosis, greater stimulation of DNA synthesis, andincreased myocardial fibrosis. If only a single doseof isoproterenol is given, the degree of subendocar-dial fibrosis observed 10 days later is only slightlygreater than in control hearts. Repeated treatmentsproduce additional myocyte necrosis and consider-ably greater levels of subendocardial fibrosis.

Since bolus subcutaneous injection results in acutehigh serum levels of isoproterenol, it was presumedthat a low dose of isoproterenol infused over aprolonged period might produce myocardial fibrosiswithout the generation of significant necrosis. Thisdid not appear to be the case. Myocyte necrosis andsignificant stimulation of DNA synthesis occurredeven when we administered 1 mg/kg i.p. isoprotere-nol over 24 hours with an osmotic minipump. How-ever, the levels of both necrosis and DNA synthesiswere less than that observed when the same dosewas given by daily subcutaneous injection.

The conclusion that initial fibrosis after isopro-terenol treatment is predominantly related to repa-ration of myocyte loss is based in part on thecomparison of dose-response levels of each pro-cess. While a quantitative estimation of the degreeof fibrosis is relatively direct, estimation of theamount of myocardial necrosis resulting from iso-proterenol is somewhat indirect, even when mono-clonal antimyosin labeling is used. Fibrosis is acumulative permanent response that may be evalu-ated in a number of ways at various periods after anintervention. Monoclonal labeling of necrotic myo-cytes, on the other hand, is a transient, time-varying process. Necrotic myocytes are probablylabeled with antimyosin within a few minutes afterdevelopment of membrane permeability. However,reparative processes immediately begin to breakdown the necrotic cell and remove it from the heart.Chronic or repeated dosing protocols stimulate newnecrosis and myocyte labeling while previouslylabeled cells are being degraded. Thus, monoclonalantibody labeling can only identify the necrotic

population during the period after development ofmembrane permeability and before removal fromthe heart. Direct quantification of the number orvolume fraction of antibody-labeled myocytes atany specific time would most likely provide a sig-nificant underestimate of the true fraction of myo-cyte loss. A quantitative estimate of total myocyteloss would probably require direct morphometriccounting procedures as has been used by Anversaet a l . «

Myocardial necrosis and fibrosis appear to berelated processes that have not been observed inde-pendently of each other. The question of whethermyocardial hypertrophy after isoproterenol treat-ment is a response related to myocyte loss25 or areactive process related to /3-adrenergic hormonestimulation20 has also been addressed in this study.Myocardial mass also increases significantly afterisoproterenol treatment. In contrast to necrosis andfibrosis, however, myocardial hypertrophy appearsto be less associated with the dose of isoproterenolover the ranges we evaluated than either myocytenecrosis or fibrosis. In our study, myocardial hyper-trophy showed a graded response to increasingdoses of isoproterenol after 2 days. However, unlikethe levels of fibrosis, maximal stimulation of hyper-trophy was observed over the range of 125-5,000)ttg/kg/day after 10 days. It should be noted thatboth Collins et al and Alderman and Harrisonfound that longer periods were required for devel-opment of maximal hypertrophy. However, theirstudies evaluated lower levels of isoproterenol thanwe studied. Stanton et al, using doses equivalentto those used in our study, found both a dose- andtime-dependent relation with regard to the degree ofdeveloped myocardial hypertrophy. Both Stantonet al and Collins et al have suggested that hyper-trophy develops in compensation for myocyte lossafter isoproterenol administration. However, in nei-ther of these studies was the level of myocytenecrosis directly determined.

The results of this study show that isoproterenolproduces myocyte necrosis, even when applied inlow doses and with chronic infusion to avoid tran-sient high levels. This necrosis is followed byincreased fibrosis and hypertrophy. The fibrotictissue response is graded and closely parallels necro-sis. Hypertrophy also developed in a measuredresponse related to isoproterenol dosage within 2days of treatment but reached maximal levels at alldoses with 10 daily repetitive treatments. Thus,hypertrophy may derive either from increased myo-cardial loading, which is a direct result of adrener-gic hormone stimulation of the heart, or as a com-pensation for myocyte loss. Thus, at this point, it isnot possible to totally rule out myocyte loss as abasic factor in any instance of increased myocardialfibrosis or increased collagen concentration (in con-trast to increased collagen content, which is propor-tional to the level of increased hypertrophy). Appli-cation of monoclonal antimyosin for detection of



ownloaded from


Benjamin et al Myocardlal Flbrosis and Necrosis 669

myocyte necrosis provides considerably greatersensitivity and, when applied in these instances,may provide significant new insight into both thecauses and direct effects of myocyte loss in myo-cardial hypertrophy.

References

8.

Morkin E, Ashford TP: Myocardial DNA synthesis in exper-imental cardiac hypertrophy. Am J Physiol 1968;215:1409-1413Bartosova D, Chvapil M, Korecky B, Poupa O, Rakusan K,Turek Z, Vizek M: The growth of the muscular and colla-genous parts of the rat heart in various forms of cardiomeg-aly. J Physiol (Lond) 1969;200:285-295Turto H: Collagen metabolism in experimental cardiac hyper-trophy in the rat and the effect of digitoxin treatment.Cardiovasc Res 1977;11:358-366Caspari PG, Newcomb M, Gibson K, Harris P: Collagen inthe normal and hypertrophied human ventricle. CardiovascRes 1977; 11:554-558Anversa P, Loud AV, Giacomelli F, Wiener J: Absolutemorphometric study of myocardial hypertrophy in experi-mental hypertension. II. Ultrastructure of myocytes andinterstitium. Lab Invest 1978;38:597-609Medugorac I: Collagen content in different areas of normaland hypertrophied rat myocardium. Cardiovasc Res 1980;14:551-554Cooper G, Tomanck RJ, Ehrhardt JC, Marcus ML: Chronicprogressive pressure overload of the cat right ventricle. OreRes 1981;48:488-497Michel JB, Salzmann JL, Ossondo Nlom M, Bmneval P,Barres D: Morphometric analysis of collagen network andplasma perfused capillary bed in the myocardium of ratsduring evolution of cardiac hypertrophy. Basic Res Cardiol1986;81:142-154

9. Weber KT, Janicki JS, Pick R, Abrahams C, Shroff SG,Bashey RI, Chen RM: Collagen in the hypertrophied, pres-sure overloaded myocardium. Circulation 1987;75:40-47

10 Grove D, Zak R, Nair KG, Aschenbrenner V: Biochemicalcorrelates of cardiac hypertrophy. IV. Observations on thecellular organization of growth during myocardial hypertro-phy in the rat. Ore Res 1969;25:473-485

11 Skosey JL, Zak R, Martin AF: Biochemical correlates ofcardiac hypertrophy. V. Labeling of collagen, myosin, andnuclear DNA during experimental myocardial hypertrophyin the rat. Ore Res 1972;31:145-157Bishop SP, Melsen LR: Myocardial necrosis, flbrosis, andDNA synthesis in experimental cardiac hypertrophy inducedby sudden pressure overload. Ore Res 1976;39:238-245

13. Weber KT, Janicki JS, Shroff SG, Pick R, Chen RM, BasheyRI: Collagen remodeling of the pressure overloaded, hyper-trophied nonhuman primate myocardium. Ore Res 1988;62:757-765Hatt PH, Rakusan K, Gastineau P, Laplace P: Morphometryand ultrastructure of heart hypertrophy induced by chronicvolume overload (aorto-caval fistula in the rat). J Mol CellCardiol 1979; 11:989-998Carey RA, Natarjan G, Bove AA, Santamore WP, Spann JF:Elevated collagen content in volume overload induced car-diac hypertrophy. J Mol Cell Cardiol 1980;12:929-936Holubarsch C, Holubarsch T, Jacob R, Medugorac I, Thiede-mann KU: Passive elastic properties of myocardium indifferent models and stages of hypertrophy: A study com-paring mechanical, chemical, and morphometric parameters,in Alpert NR (ed): Perspectives in Cardiovascular Research,vol 7: Myocardial Hypertrophy and Failure. New York,Raven Press, Publishers, 1983, pp 323-336Weber KT, Janicki JS, Shroff SG, Pick R, Abrahams C,Chen RM, Bashey RI: Collagen compartment remodeling inthe pressure overloaded left ventricle. JAppl Cardiol 1988;3:37-46

12

14

15

16

17

18. Stanton HC, Brenner G, Mayfield ED: Studies onisoproterenol-induced cardiomegaly in rats. Am Heart J1969;77:72-80

19. Alderman EL, Harrison DC: Myocardial hypertrophy result-ing from low dosage isoproterenol administration in rats.Proc Soc Exp Biol Med 1971;136:268-270

20. Gordon AL, Inchinosa MA, Lehr D: Isoproterenol-inducedcardiomegaly: Assessment of myocardial protein content,actomyosin ATPase and heart rate. / Mol Cell Cardiol 1972;4:543-557

21. Pagano VT, Inchinosa MA: Cardiomegaly produced bychronic beta-adrenergic stimulation in the rat: Comparisonwith alpha-adrenergic effects. Life Sci 1977;21:619-624

22. Tse J, Powell JR, Baste CA, Priest RE, Kuo JF: Isoproterenol-induced cardiac hypertrophy: Modifications in characteris-tics of beta-adrenergic receptor, adenylate cyclase, andventricular contraction. Endocrinology 1979;105:246-255

23. Genovese A, Chiariello M, Bozzaotre M, Latte S, De AlfieriW, Condorelli M: Adrenergic activity as modulating factor inthe genesis of myocardial hypertrophy in the rat. Exp MolPathol 1984;41:390-396

24. Rona G, Kahn DS, Chappel CI: Study on the healing ofcardiac necTosis in the rat. Am J Pathol 1961;39:473-489

25. Collins P, Billings CG, Barer GR, Daly JJ, Jolly A: Quanti-zation of isoprenaline-induced changes in the ventricularmyocardium. Cardiovasc Res 1975;9:797-806

26. Rona G, Chappel CI, Balazs T, Gaudry R: An infarct-likemyocardial lesion and other toxic manifestations producedby isoproterenol in the rat. AMA Arch Pathol 1959;67:99-111

27. Beznak M: Hemodynamics during the acute phase of myo-cardial damage caused by isoproterenol. Can J BiochemPhysiol 1962;40:25-30

28. Bloom S, Cancilla PA: Myocytolysis and mitochrondrialcalcification in rat myocardium after low doses of isoprotere-nol. Am J Pathol 1969;54:373-391

29. Todd GL, Cullan GE, Cullan GM: Isoproterenol-inducedmyocardial necrosis and membrane permeability alterationsin the isolated perfused rabbit heart. Exp Mol Pathol 1980;33:43-54

30. Rona G: Catecholamine cardiotoxicity (editorial review). /Mol Cell Cardiol 1985;17:291-301

31. Simpson P, McGrath A, Savion S: Myocyte hypertrophy inneonatal rat heart cultures and its regulation by serum andcatecholamines. Ore Res 1982;51:787-801

32. Simpson P, McGrath A: Norepinephrine-stimulated hyper-trophy of cultured rat myocardial cells is an alpha! adrener-gic response. / Clin Invest 1983;72:732-738

33. Simpson P, Bishopric N, Coughlin S, Karliner J, Ordahl C,Starksen N, Tsao T, White N, Williams L: Dual trophiceffects of the alpha] adrenergic receptor in cultured neonatalrat heart muscle cells. J Mol Cell Cardiol 1986;18:45-58

34. Nolan AC, Clark WA, Karwoski T, Zak R: Patterns ofcellular injury in myocardial ischemia determined by mono-clonal antimyosin. Proc Natl Acad Sci USA 1983;80:6046-6050

35. Hayes JS, Bowling N: Role of the alpha agonist activity ofdobutamine in mediating cardiac output: Effects of pro-longed isoproterenol infusion. J Pharmacol Exp Ther 1987;241:861-869

36. Clark WA, Chizzonite RA, Everett AW, Rabinowitz M, ZakR: Species correlations between cardiac isomyosins: Acomparison of electrophoretic and immunological proper-ties. JBiol Chem 1982;257:5449-5454

37. Weibel ER: Stereological Methods. New York, AcademicPress, 1979, vol 1: Practical Methods for Biologic Morphometry

38. Burton K: A study of the conditions and mechanism of thediphenylamine reaction for the colorimetric estimation ofdeoxyribonucleic acid. Biochem J 1956;62:315-323

39. Strubelt O, Siegers CP: Role of cardiovascular and ionicchanges in pathogenesis and prevention of isoprenaline-induced cardiac necrosis, in Fleckenstein A, Rona G (eds):Recent Advances in Studies on Cardiac Structure and Metab-olism, vol 6: Pathophysiology and Morphology of Myocar-



ownloaded from


670 Circulation Research VcA 65, No 3, September 1989

dial Celt Alteration, Baltimore, Md, University Park Press, 42. Anversa P, Hiler B, Ricci R, Guideri G, Olivetti G: Myocyte1975, pp 135-142 cell loss and myocyte hypertrophy in the aging rat heart. J

40. Yeager JC, lams SG: The hemodynamics of isoproterenol- Am Coll Cardiol 1986;8:1441-1448induced cardiac failure in the rat. Ore Shock 1981 ;8:151-163

41. Kizer DE, Howell BA: On relationships between synthesisof DNA and incorporation of deoxythymidinc into DNAduring myocardial infarctions induced in rats by isoprotere- KEY WORDS • hypertrophy • monoclonal antimyosin •nol. Chem Biol Interact 1971;4:251-264 fibroblasts • collagen • immunofluorescence



ownloaded from


I J Benjamin, J E Jalil, L B Tan, K Cho, K T Weber and W A ClarkIsoproterenol-induced myocardial fibrosis in relation to myocyte necrosis.

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1989 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.65.3.6571989;65:657-670Circ Res.

http://circres.ahajournals.org/content/65/3/657World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Research Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:



ownloaded from

http://circres.ahajournals.org/content/65/3/657

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circres.ahajournals.org//subscriptions/


isoproterenol-induced myocardial fibrosis in relation to...

Documents