-
605
SPONTANEOUS AND EXPERIMENTAL MYOCARDITIS:THE EFFECT OF FREUND'S
ADJUVANT ON THE HEART
AND OTHER ORGANS*
A. LAUFER, E. ROSENMANN AND A. M. DAVIESFrom the Rothschild
Hadassah University Hospital, Departments of Pathology
and Medical Ecology, Hebrew University-Hadassah Medical
School,Jerusalem, Israel
Received for publication May 27, 1966.
FREUND'S mineral-oil mixture and its variations are widely used
as adjuvantsin the stimulation of antibody response to a variety of
antigens (Freund, 1951;Kaplan, 1958; Kaplan and Craig, 1963; Brown,
Glynn and Holborow, 1963;Laufer, Ginsburg, Gery and Davies, 1963;
Davies, Laufer, Gery and Rosenmann,1964; and Cuppage, 1965). That
the mycobacterial adjuvant by itself canproduce a variety of
pathological responses, ranging from arthritis to
generalizeddisease in the rat (Pearson, 1956; Pearson, Waksman and
Sharp, 1961; Pearsonand Wood, 1963) granulomatous lesions of the
lungs (Amemori and Altschul,1963; Allegretti and Vitale, 1965) and
of other organs (Thal, Laufer and Behar,1958; Laufer, Thal and
Behar, 1959; Steiner, Langer and Schatz, 1960; Jahieland Koffler,
1961; Brown et al., 1963; Geduldig, Reubner and Iber, 1964) of
avariety of animals, with, in mice, amyloid deposits, (Laufer et
al., 1959; Thal,Laufer and Zlotnick, 1964) is well established.
The production of experimental myocarditis in laboratory animals
by injectionof heart muscle extracts in Freund's complete adjuvant
(Davies et al., 1964), theoccasional appearance of lesions in
control animals and the prevalence of " spon-taneous " myocarditis
in laboratory stocks, prompted this study of the effects ofthe
adjuvant itself on the heart and other organs.
MATERIAL AND METHODSAnimals.-Outbred albino rats of the Hebrew
University strain weighing 60-110 g.,
albino mice (35-40 g.) and pen bred guinea-pigs (170-370 g.)
were used in these experiments.Mongrel rabbits (about 2-5 kg.) were
obtained from a dealer and Syrian hamsters (90-120 g.)from the
laboratories of the Beilinson Hospital. The numbers examined and
the treatmentgiven are shown in Table I. Control animals, matched
for weight but not for sex, werereceived at the same time as the
test animals and were kept in adjacent cages receiving thesame diet
but no treatment.
Freund's complete adjuvant.-(Difco) was emulsified with an equal
volume of saline andinjected i.c. in the stated dosages divided
between 2 sites, weekly for 6 weeks.
Method&.-One week after the last injection, the animals were
killed by exanguinationfrom the neck vessels while under ether
anaesthesia, except for the rabbits which were killedby a blow on
the back of the neck. Immediately after death, whole organs, or
samplepieces in larger animals, were removed and fixed in neutral
formalin.
Paraffin sections of the organs cut at 10,u were stained with
haematoxylin and eosin aswell as with special stains for amyloid
where indicated.
* This work was supported by Research Grant HE-05396 from the
U.S. National Institutes ofHealth.
-
A. LAUFER, E. ROSENMANN AND A. M. DAVIS
Sera were examined for the presence of anti-heart antibodies in
the passive haemagglutina-tion test using human group 0
erythrocytes coated with extracts of homologous and hete-rologous
heart (Gery and Davies, 1961).
RESULTSSerological findings
None of the animals, test or control, showed antiheart
antibodies of a titrehigher than 1: 5, i.e. all sera were "
negative " by definition (Gery and Davies,1961).Histopathological
findings (Tables I and II)
Heart.-The histological changes in the myocardium were the same
in treatedand untreated animals but varied in distribution and
intensity in the different
TABLE I.-Study Design and Summary of ResultsNumber of animals
with pathological lesions*
Treatment
Adjuvant-0. 5 ml.weekly for 6 weeksNone-controls
Adjuvant-0 5 ml.weekly for 6 weeks
None-controls
Adjuvant-0-5 ml.weekly for 6 weeks
None-controls
Adjuvant-0- 7 ml.weekly for 6 weeks
Heart Lungs
4/25 14/20
4/27 111220/15 10/14
0/8 8/85/44 35/42
3/31 20/2513/26 23/25
Spleen
20/24Liver
20/24
3/23 12/260/15 5/14
Pancreas andmesenteric
fat
15/22 5/25
0/210/14
1/8 1/8 0/80/40 8/42 4/43
0/2710/25
5/268/24
0/2813/19
14 . None-controls
* No. with pathological changesNo. animals examined
Number of
7/14 8/14 3/14 0/12 0/13
Fre,und's Adjuvant*Pancreas
andmesenteric
Species animals Heartt Lungs Spleen Liver fat KidneyiMouse. 25 .
(4/25) . 1/20 . 12/244 . 14/24 . 15/22 0/25Hamster . 15 . 0/15 .
0/14 . 0/16 . 0/14 . 0/14 . 0114Rat . 44 . (5/44) . 3/42 . 0/40
2/42 . 4/43 . 0/42Guinea pig . 26 . (13/26) . 11/26 10/25 . 6/24 .
13/19 . 0/25
* Numbers with lesions/numbers examined.t See text. The number
ofanimals with cardiac lesions were similar in treated and control
animals.t Of the 12 affected mice, 3 showed presence of
amyloid.
Guinea pig270 100 g.
Speciesand weight
Mouse37 2g.
Hamster105 + 15 g.
Rat .85 25g.
Number ofanimals
25
27
15
8
44
Kidneys
31
26
5/250/14
0/81/42
1/271/25
TABLE II.-Pathological Lesions Attributable to
1/12
606
8
-
SPONTANEOUS AND EXPERIMENTAL MYOCARDITIS
species. The characteristic feature was an interstitial
infiltration of small ormedium sized round cells, usually at
single, but sometimes multiple, foci or scatteredbetween the muscle
fibres. In rats and guinea pigs, the infiltrate displaced,
andsometimes replaced, the muscle fibres at some points (Figs. 1,
2, 3, 4, 5). None ofthe hamsters showed myocardial lesions while
the rats were unusual in thattreated animals showed an interstitial
infiltration deep between the myocardialfibres while in the control
group, the infiltrate was subepicardial in position.
Foam cells were not seen in any of the preparations.ILungs.-Only
mice, rats and guinea pigs which had received adjuvant, showed
granulomatous lesions (Fig. 6). The granuloma consisted of
epitheloid cells witha few giant cells and foam cells, fat vacuoles
surrounded by lymphoid cells beingpresent in some of them. No
necrosis was present in the granulomata nor wereacid fast bacilli
detected. The granulomata were single or conglomerated andsituated
either in the parenchyma or near small vessels or bronchioles.
Peri-vascular cuffing with lymphoid cells was seen in several of
the rats and guineapigs.
Areas of congestion and haemorrhage and interstitial pneumonia
were seen insome of the animals-both treated and not.
Liver.-The treated animals showed capsular and/or parenchymal
changes ofvarying distribution and intensity. The capsular lesions
consisted of granulomatawith macrophages and fat vacuoles (Fig 7).
Similar granulomata were presentin the liver parenchyma of mice,
rats and guinea pigs in varying degrees (Fig. 8).A round cell
infiltration varying in intensity was present in the portal spaces
orin the sinusoids extending into the parenchyma, sometimes
accompanying granu-lomatous lesions (Fig. 9). The control group
showed, rarely, a minimal roundcell infiltration in the portal
spaces, but no granulomata.
Kidneys.-Interstitial round cell infiltrations were present in
some of thetreated and untreated animals; there were no glomerular
lesions.
Pancrea8.-Lesions in the pancreas and peripancreatic fat were
seen in treatedmice, rats and guinea pigs. They consisted of round
cell infiltration in theinterlobular septa and peripancreatic fat
tissue with granulomata in the peri- andinter-pancreatic septa
(Fig. 10). The granulomatous lesions consisted of epitheloidcells
surrounded or mingled with foam cells or large fat vacuoles, round
cellinfiltration and marked proliferation of fibroblasts (Fig. 11).
Neutrophils wererare but there were many capillaries and distended
vessels lined by flattenedendothelial cells. The pancreatic
parenchyma was intact. The control animalsdid not show any changes
in the pancreas.
Spleen.-Histological changes appeared only in mice and guinea
pigs givenadjuvant and were present both in the splenic tissue and
capsule. The follicles,in some animals, were large without clear
boundaries and the red pulp showedmarked histiocytic proliferation.
In the mouse spleen there were abundant giantcells and this
species, alone, showed amyloidosis, of a perifollicular and/or
pulpardistribution (Fig. 12). The amyloid showed characteristic
properties, i.e. stainingwith PAS and Congo-red, metachromasia with
gentian violet and dichroism inpolarized light.
While the splenic capsule was usually thin, local thickening was
found in someanimals. This consisted ofround cell infiltration with
fibroblasts, fibrocytes withincollagen fibres as well as foamy
epitheloid cells, forming a granulomatous reactionwithout necrosis.
Many dilated lymph vessels were present, filled with small
607
-
A. LAUFER, E. ROSENMANN AND A. M. DAVIS
lymphocytes and lined by one or two layers of elongated or
cuboidal cells, here andthere accumulating into giant cells. In
some places there were adhesions to theinflamed peri-pancreatic
fatty tissue.
Splenic lesions, present in some of the control animals, were
limited to anoccasional large follicle.
DISCUSSION
The histological changes in the hearts of injected animals and
untreatedcontrols are morphologically identical but vary in
intensity and distribution.The one exception, in the rat
myocardium, lies in the localization of the cellularinfiltrate
which is interstitial in treated animals, rather than subepicardial
incontrols. The amounts of Freund's adjuvant received by these
animals weremany times larger than in other series (Pearson, 1956;
Pearson et al., 1961;Pearson and Wood, 1963) and thus one would
have expected more or differentlesions in injected animals. Their
similarity to the controls suggests that thelesions observed are
those of " spontaneous myocarditis " although the possibilityof
minimal damage by the adjuvant cannot be excluded. The results
obtained inother organs by injection of adjuvant by different
routes, alone and in combinationwith antisera (Roitt, Jones and
Doniach, 1961; Geduldig et al., 1964; Watson,Dixon and Feldman,
1965; Cuppage, 1965) indicate a damaging effect of theadjuvant
itself although leaving unsolved the precise mode of action.
" Spontaneous myocarditis " has been reported by several
authors, usually inthe sense of myocarditis of unknown cause. Up to
two-thirds of the apparentlyhealthy mice of different colonies,
have been affected (Miller, 1924; Wilens andSproul, 1938a, b; Loewe
and Lenke, 1940; Pearson and Wood, 1963) andwe have noted similar
changes, though less frequently, in guinea pigs, rats andrabbits
(Davies et al., 1964). The same histological picture is produced
byendotoxin (Kovats, 1961; Davies, Gery, Rosenmann and Laufer,
1963) and
EXPLANATION OF PLATESFIG. 1. Rat-injected with adjuvant.
Interstitial focus of myocarditis replacing
muscle fibres and consisting of small and medium sized round
cells. H and E X 405.FIG. 2.-Guinea pig-injected with adjuvant.
Interstitial focus of myocarditis, consistingof mainly small round
cells mingled with some histiocytic cells. H and E X 396.
FIG. 3.-Rat, control group-Diffuse focus of interstitial
myocardial infiltration.HandE x405.
FIG. 4.-Guinea pig, control group-small interstitial focus of
infiltrationdisplacing the surrounding muscle fibres. H and E x
405.
FIG. 5.-Mouse intracutaneously injected with Freund's
adjuvant.Small interstitial focus of myocarditis. H and E x
405.
FIG. 6.-Guinea pig-injected with Freund's adjuvant.
Granulomatous lesionspresent in the lung parenchyma. H and E x
410.
FIG. 7.-Mouse-injected with Freund's adjuvant. Liver capsule,
thickened, infiltratedwith numerous granulomata. Note the presence
of dilated lymphatics. H and E X 38.
FIG. 8.-Mouse-injected with Freund's adjuvant. Granuloma present
in the liverparenchyma. Note the presence of large fat vacuoles. H
and E x 405.
FIG. 9.-Rat-injected with Freund's adjuvant. Round cell
infiltration in the portalspace of the liver extending into the
parenchyma. H and E x 100.
FIG. 10.-Mouse-injected with Freund's adjuvant. Peri-pancreatic
fat granuloma;mild interstitial round cell infiltration. H and E x
405.
FIG. 11.-Mouse-injected with Freund's adjuvant. Fat granulomain
the pancreatic septa. H and E x 405.
FIG. 12.-Mouse-injected with Freund's adjuvant.Diffuse
amyloidosis in spleen. H and E x 405.
608
-
BRITISH JOURNAL OF EXPERIMENTAL PATHOLOGY. Vol. XLVII, No.
6.
1*
'
,
Laufer, Rosenmann and Davies.
-067!&W.. 4ftJW.'C4.4 ."e,
-4w" -IrM
-
Vol. XLVII, No. 6.BRITISH JOURNAL OF EXPERIMEN-TAL
PATHOLOGY.
rn__ I
I . A..f,*A
Laufer, Rosenmann and Davies
-
BRITISH JOURNAL OF EXPERIMENTAL PATHOLOGY.
-'
wt-LA,.
W VWW& m
F PI
Ii,
.,11
V
A4Q
Laufer, Rosenmann and Davies.
VOl. XLVII, NO. 6.
.:.;... IPW .-It -W
..x
Aulyllo.
. 4 Lmp
.NW
-
SPONTANEOUS AND EXPERIMENTAL MYOCARDITIS
although no microorganisms were found in sections of the heart,
the animalbreeding stocks suffered from endemic salmonellosis
(mainly Salmonella typhi-murium) and these organisms in the gut
might well have released endotoxin intothe circulation (Davies et
al., 1963). Another aetiological possibility is virusinfection and
coxsackie virus among others, may cause lesions
indistinguishablefrom those seen (Moore, Ridge, Huntington, Hall,
Riffith and Knowles, 1947;Laruelle and Reumont, 1952). The "
spontaneous myocarditis " seen in therat is usually mild (Davies et
al., 1963) while the hamster, in this and previousseries, was never
affected. This fact emphasizes the differences in susceptibilityof
different species and suggests the hamster to be a suitable animal
for studieson experimental myocarditis. The guinea pig showed the
greatest sensitivityto Freund's adjuvant as illustrated by the
lesions present in lung, liver, pancreasand peripancreatic fat and
splenic capsule. That this might be a reflection of thisspecies'
known sensitivity to toxic and immunological stimuli is suggested
by therelatively high prevalence of myocarditis in control as well
as in treated animals.
It is likely that the adjuvant elicits the cellular reactions in
one of two waysdepending on the amount and route of injection
(Steiner et al., 1960; Dalle, 1960,1961; Schoenberg and Moore,
1961). Dispersal of the oil phase as microemboliwill produce direct
tissue damage with resulting granuloma formation. Stimula-lation of
the reticuloendothelial system in a way not fully understood
butconnected with the mycobacterial fractions will result in a "
hypersensitivity"reaction with many manifestations (Freund, 1951,
1957; Laufer et al., 1959;Pearson et al., 1961; Steiner et al.,
1960; Pearson and Wood, 1963; Thal et al.,1964).
Thus the pancreatic lesions could result from damage of the cell
membrane bycontact with the oil droplets or the mycobacterial
antigen releasing proteolyticand lipolytic enzymes. Here again, the
species difference was marked, lesions ofthe pancreas being
frequent in guinea pigs and mice, rare in rats and absent
inhamsters. Granulomata of lung, liver and pancreas are noted and
have beenpreviously described (Freund, 1951 ; Thal et al., 1958;
Laufer et al., 1964;and Geduldig et al., 1964) while Amemori and
Altschul (1963) describe foam likecells in the myocardium after
i.v. injection of adjuvant.
The hypersensitivity reactions described have been elicited with
small dosesof mycobacterial adjuvant (Allegretti and Vitale, 1965)
and microemboli were notfound in the vicinity of, for instance,
arthritic joints (Pearson, 1956; Pearson et al.,1961; and Pearson
and Wood, 1963). None ofhundreds ofsections ofmyocardiumexamined in
this and previous series has shown foam cells or the presence
ofadjuvant droplets adjacent to focal myocarditis. The resemblance
between thelesions observed and those of the myocarditis of
homologous disease (Stastny,Stembridge, Vischer and Ziff, 1965)
suggest an immune mechanism. We havesuggested elsewhere however
(Laufer and Davies, 1966) that in the heart, theeffects of "
hypersensitivity ", as those of infection, endotoxin or anoxia,
aremediated through a final common path, involving changes in
cellular permeability,lysosome damage and autolysis and there is a
mounting body of supportingevidence (Laufer and Davies, 1966;
Gazenfeld, Rosenmann, Davies, and Laufer,1966). The absence of
circulating antibodies is in accord with this suggestion(Davies, et
al., 1963, 1964; Laufer and Davies, 1966).
It is, in fact, doubtful whether such an entity as " spontaneous
myocarditis"exists and the lesions so described are most likely the
standard histological response
609
-
610 A. LAUFER, E. ROSENMANN AND A. M. DAVIS
through different pathways to one or other of a variety of
agents. The verymonotony of the cellular response (Laufer and
Davies, 1966) demands the mostcritical analysis and the use of
adequate controls, in all experiments involvinginterpretation of
the lesions of experimental myocarditis.
SUMMARY
The effect of Freund's adjuvant on experimental myocarditis in
differentspecies was studied in relation to " spontaneous
myocarditis ". Treated animalsshowed lesions of lungs, spleen,
liver and their serosal surfaces, pancreas andperipancreatic fat,
that were attributable to the action of the adjuvant.Pancreatic
lesions were more prominent in mice and guinea pigs, amyloidosis
waspresent only in mice while hamsters showed no specific lesions
indicating a speciessusceptibility.
The possible pathogenesis of " spontaneous myocarditis " was
discussed in thelight of the probability that different causes
mediate through a final commonpathway to produce a standard
histological response.
REFERENCESALLEGRETTI, N. AND VITALE, B.-(1965) Immunol., 9,
11.AMEMORI, T. AND ALTSCHUL, R.-(1963) Zbl. allg. Path. path.
Anat., 104, 352.BROWN, P. C., GLYNN, L. E. AND HOLBOROW, E.
J.-(1963) J. Path. Bact., 86, 505.CUPPAGE, F. E.-(1965) Lab.
Invest., 14, 514.DALLE, M. M.-(1960) Br. J. exp. Path., 41,
86-(1961) Br. J. exp. Path., 42, 297.DAVIES, A. M., LAUFER, A.,
GERY, I., ROSENMANN, E. AND LAUFER, A.-(1963) Proc.
Soc. exp. Biol. Med., 114, 520.DAVIES, A. M., GERY, I. AND
ROSENMANN, E.-(1964) Arch. Path., 78, 369.FREUND, J.-(1951) Am. J.
clin. Path., 21, 645.-(1957) J. Allergy, 28, 18.GAZENFELD, E.,
ROSENMANN, E., DAVIES, A. M. AND LAUFER, A.-(1966) Immunol.,
10, 193.GEDULDIG, M. M., REUBNER, B. AND IBER, F. L.-(1964)
Gastroenterology, 46, 175.GERY, I. AND DAVIES, A. M.-(1961) J.
Immun., 87, 351.JAHIEL, R. I. AND KOFFLER, D.-(1961) Br. J. exp.
Path., 42, 338.KAPLAN, M. H.-(1958) J. Immun., 80, 254.KAPLAN, M.
H. AND CRAIG, J. M.-(1963) J. Immun., 90, 725.KOVATS, T.G.-(1961)
Naturiwissenschaften, 48, 572.LAUFER, A. AND DAVIES, A. M.-(1966)
In " Methods and Achievements in Experimental
Pathology ", Vol. II, Ed. Bajusz, E. and Jasmin G. Basel:
(Karger) (in press).LAUFER, A., GINSBURG, I., GERY, I. AND DAVIES,
A. M.-(1963) Path. Biol., 11, 769.LAUFER, A., THAL, C. AND BEHAR,
A. J.-(1959) Br. J. exp. Path., 40, 1.LARUELLE AND REUMONT-(1952)
Ann. Inst. Pasteur, 83, 543.LOEWE, L. AND LENKE, S. E.-(1940) J.
exp. Med., 71, 89.MILLER, P.-(1924) J. exp. Med., 40, 543.MOORE, J.
F., RIDGE, G. K., HUNTINGTON, R. W., HALL, E. M., RIFFITH, G.
ANI)
KNOWLES, R. G.-(1947) Proc. Soc. exp. Biol. Med., 65,
102.PEARSON, C. M.-(1956) Proc. Soc. exp. Biol. Med., 91,
95.PEARSON, C. M., WAKSMAN, B. H. AND SHARP, J. T.-(1961) J. exp.
Med., 113, 485.PEARSON, C. M. AND WOOD, F. D.-(1963) Am. J. Path.,
42, 73.ROITT, I. M., JONES, H. E. H. AND DONIACH, DEBORAH-(1961) In
lInd International
Symposium on Immunopathology, Ed. Grabar, P. and Miescher, P.,
Stuttgart.(Benno Schwabe), p. 174.
-
SPONTANEOUS AND EXPERIMENTAL MYOCARDITIS 611
SCHOENBERG, M. D. AND MOORE, R. D.-(1961) Arch. Path., 72,
424.STASTNY, P., STEMBRIDGE, V. A., VISCHER, T. AND ZIFF, M.-(1965)
J. exp. Med., 122,
681.STEINER, J. W., LANGER, B. AND SCHATZ, D. L.-(1960) Arch.
Path., 70, 424.THAL, C., LAUFER, A. AND BEHAR, A. J.-(1958) Br. J.
exp. Path., 39, 159.THAL, C., LAUFER, A. AND ZLOTNICK, A.-(1964)
Br. J. exp. Path., 45, 323.WATSON, J. T., DIXoN, F. J. AND FELDMAN,
J. D.-(1965) Lab. Invest., 14, 1559.WILENS, S. L. AND SPROUL, E.
E.-(1938a) Am. J. Path., 14, 177-(1938b) Am. J. Path.,
14, 201.
57