1. health and fitness
2. disease

Cardiovascular Research 33 Ž1997. 469–477
Insulin-like growth factor-II delays myocardial infarction in
experimental coronary artery occlusion
Achim M. Vogt, Patrik Htun, Angelika Kluge, Rene´ Zimmermann, Wolfgang Schaper
)
Max Planck Institute for Physiological and Clinical Research, (W.G. Kerckhoff Institute), Department of Experimental Cardiology, Benekestrasse 2,
D-61231 Bad Nauheim, Germany
Received 24 June 1996; accepted 26 September 1996
Objective: We have previously shown that short pulses of myocardial ischemia cause increased mRNA expression of the insulin-like
growth factor II ŽIGF-II. gene. The expression of IGF-II precedes the expression of its binding protein 5 ŽIGFBP-5.. The cardioprotective
actions of the IGF-II peptide and of its binding protein 5 as well as the underlying mechanisms were investigated in this study. Methods
and results: Human recombinant IGF-II Ž0.25 mgrml. was applied by means of direct intramyocardial infusion ŽIM. for 60 min prior to
a 60 min LAD occlusion and 120 min reperfusion. Myocardial infarction, compared to the region at risk, was significantly decreased by
IGF-II treatment, whereas infusion of Krebs-Henseleit buffer did not show any protective effect ŽIGF-II, 78.75 " 1.51%; control, 100%;
P - 0.005.. A comparable degree of cardioprotection was observed after infusion of an equipotent concentration of recombinant human
insulin Ž0.02 IUrml; 88.25 " 1.45%; P - 0.05.. Lavendustin A Ž100 mM., an inhibitor of protein tyrosine kinases, prevented the
observed cardioprotection. The protective effect of IGF-II was lost when IGFBP-5 was simultaneously infused. Conclusion: IGF-II, a
peptide that binds to the insulin receptor and whose mRNA is rapidly transcribed by cardiac myocytes following ischemic stress, is
cardioprotective and mimics ischemic preconditioning. Its observed actions are probably based on its metabolic effects and are mediated
by the insulin or the IGF-I receptor. IGFBP-5, whose expression follows IGF-II’s expression with a short delay, inhibits the
cardioprotection afforded by IGF-II and may thus account for the limited temporal duration of ischemic preconditioning.
Keywords: Insulin-like growth factor II; Insulin; Myocardial protection; Preconditioning; Myocardial infarction
1. Introduction
Repetitive coronary occlusions of short duration render
the heart resistant against a subsequent long period of
ischemia. This experimental intervention, termed ‘ischemic
preconditioning’ w1x, is the most powerful protection against
myocardial infarction for the in situ beating heart.
Previous work from our laboratory had demonstrated
significant changes in myocardial gene expression in myocardium subjected to ischemic preconditioning w2x.
Amongst other genes, insulin-like growth factor II ŽIGF-II.
and insulin-like growth factor binding protein 5 ŽIGFBP-5.
w3x were upregulated by ischemiarreperfusion but also by
surgical stress. The insulin-rinsulin-like growth factor system is an endocrinerparacrine physiologic system controlling metabolism, growth, differentiation, and the response
to stress ŽFig. 1. w4,5x. Recent observations in neuronal
models suggest trophic and protective actions of the IGFs
in the setting of ischemia and reperfusion w1,6–9x. Like
adult myocytes, neurons are terminally differentiated cells
that exhibit pathways of endogenous protection similar to
ischemic preconditioning in the heart w10–14x. In skeletal
muscle, it was demonstrated that the IGFs stimulate glucose uptake by their insulin-like activity w15,16x. It is
therefore tempting to speculate that in times of metabolic
stress, as in ischemia, the heart expresses ‘its own insulin’
by making use of the metabolic properties of IGF-II. Since
the expression of IGF-II is followed by the expression of
one of its binding proteins which terminates its action, we
hypothesize that ischemic preconditioning can, at least in
part, be explained on the basis of changes of the IGF tissue
concentration.
)
Corresponding author. Tel. q49 6032 705-402; Fax q49 6032
705-419.
Time for primary review 21 days.
0008-6363r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved.
PII S 0 0 0 8 - 6 3 6 3 Ž 9 6 . 0 0 2 1 2 - X
Downloaded from http://cardiovascres.oxfordjournals.org/ by guest on June 9, 2014
Abstract
470
A.M. Vogt et al.r CardioÕascular Research 33 (1997) 469–477
2. Methods
The experimental protocol described in this study was
approved by the Bioethical Committee of the District of
Darmstadt, Germany. Furthermore, all animals in this study
were handled in accordance with the guiding principles in
care and use of animals as approved by the American
Physiological Society and the investigation conformed with
the Guide for Care and Use of Laboratory Animals published by the US National Institutes of Health.
2.1. Animal preparation
Sixteen castrated male German landrace-type domestic
pigs with body weights between 27.5 and 35.5 kg Ž32.6 "
0.22 kg. were premedicated with 2 mg P kgy1 BW i.m.
azaperone ŽStresnil, Janssen Pharmaceutica, Neuss, Germany. and 2 mg P kgy1 BW s.c. piritramid ŽDipidolor,
Janssen Pharmaceutica, Neuss, Germany. 30 min prior to
the initiation of the anesthesia with 10 mg P kgy1 BW
metomidate ŽHypnodil, Janssen Pharmaceutica, Neuss,
Germany.. After either oral or endotracheal intubation, a
bolus of 25 mg P kgy1 BW of a-chloralose ŽSigma,
Deisenhofen, Germany. was given intravenously. Anesthe-
sia was maintained by a continuous intravenous infusion of
25 mg P kgy1 BW P hy1 a-chloralose. The animals were
ventilated artificially with a pressure-controlled respirator
ŽStephan Respirator ABV, F. Stephan GmbH, Quickborn,
Germany. with room air enriched with 2 l P miny1 oxygen.
Arterial blood gases were analyzed frequently to guide
adjustment of the respirator settings. Additional doses of
piritramid Ž10 mg. were given i.v. every 60 min.
Both internal jugular veins were cannulated with polyethylene tubes for administration of saline, piritramid and
a-chloralose. Arterial sheath catheters Ž7F. were inserted
into both common carotid arteries. To measure aortic
blood pressure, the left sheath was advanced into the aortic
arch and connected with a Statham transducer ŽP23XL,
Statham, Puerto Rico.. A 5F high-fidelity catheter-tipped
manometer ŽMillar Instruments, Houston, TX, USA. was
inserted via the right common carotid artery into the left
ventricle to measure left ventricular pressure and to calculate its first derivative ŽLV dPrdt.. The chest was opened
by a midsternal thoracotomy and the heart was suspended
in a pericardial cradle. A loose silk ligature was placed
halfway around the left anterior descending coronary artery
ŽLAD., and was subsequently tightened to occlude the
vessel. After preparation, a stabilization period of 20 min
was allowed and the different experimental protocols were
started.
Index ischemia was a 60 min LAD occlusion followed
by 120 min reperfusion. If ventricular fibrillation occurred,
a defibrillator with paddles placed onto the external chest
wall was used for conversion. Data from DC-amplifiers
including surface ECG ŽHeinemann, FrankfurtrMain, Germany. were digitized by an analog-digital converter ŽMacLabr8, AD Instruments Pty Ltd., Castle Hill, Australia.
and continuously recorded onto the hard disc of a personal
computer ŽPower Book 180, Apple Computer Inc., Cupertino, CA, USA.. Body temperature was maintained within
the physiological range by thermal isolation of the animals.
Of the 16 animals investigated, 5 experienced ventricular
fibrillation during LAD occlusion and were successfully
defibrillated. To avoid the side-effects of systemic injections of either IGF-II or insulin on systemic glucose
metabolism w22x we treated the myocardium of interest
only locally by intramyocardial infusion as described in
detail previously w23,24x.
Eight 26-gauge needles with a free length of 5 mm that
were connected by tubing with a peristaltic pump ŽMinipuls, Gilson, Germany. were placed in pairs into the
subsequently ischemic myocardium perpendicular to the
epicardial surface. The substances were infused at 20
mlrminrneedle, amounts far below those causing distension of the myocardium.
2.2. Experimental groups
The experimental design is demonstrated in Fig. 2. Five
groups of animals were studied Žin a sixth group of
animals a different design and methodology was used..
Downloaded from http://cardiovascres.oxfordjournals.org/ by guest on June 9, 2014
Since the appearance of protein may follow the increased mRNA content with only a short delay w17x, the
aim of the present study was to investigate whether the
increased expressions of IGF-II and its binding protein 5
may be operative in increasing myocardial resistance
against infarction.
To test its cardioprotective actions, human recombinant
IGF-II was directly infused into porcine myocardium prior
to coronary occlusion and reperfusion normally causing
myocardial infarction. To further elucidate the underlying
mechanism, an equipotent dose concerning IGF-II’s
metabolic effects of recombinant human insulin was tested
in a second group. The receptors involved were further
characterized by simultaneous infusion of lavendustin A to
block the protein tyrosine kinase domain of either the
IGF-I or the insulin receptor w18,19x.
There is very little knowledge available yet about the
physiologic functions of the IGF-binding proteins. Recent
investigations have demonstrated that the physiologic actions of the IGFs Ži.e., cellular glucose uptake. can be
inhibited by their binding proteins w16,20x. In our porcine
model of repeated short coronary occlusions, the expression of IGFBP-5 follows the expression of IGF-II with a
short delay w1,3x. Due to the good correlations of the time
courses of the appearance of IGFBP-5 mRNA and the
vanishing protection of ischemic preconditioning w21x we
hypothesized that the increased content of IGFBP-5 with
prolonged reperfusion periods will inhibit IGF-II’s cardioprotective actions.
Possible interference of the expression of IGFBP-5 was
investigated by a simultaneous infusion of IGFBP-5 with
IGF-II prior to ischemia and reperfusion.
A.M. Vogt et al.r CardioÕascular Research 33 (1997) 469–477
471
Fig. 2. Experimental groups. Six experimental groups were investigated. After stabilization Ž20 min., groups 1 to 3 consisted of animals receiving
microinfusions of Krebs-Henseleit buffer, IGF-II, or insulin respectively. The animals of groups 4 and 5 were treated with simultaneous infusions of
IGF-IIrinsulin and lavendustin A. Group 6 was subjected to microinfusions of IGF-II and IGFBP-5. In all animals, index ischemia causing myocardial
infarction was 1 h LAD occlusion followed by 2 h of reperfusion.
Downloaded from http://cardiovascres.oxfordjournals.org/ by guest on June 9, 2014
Fig. 1. Signal transduction pathway of the insulinrIGF system. Insulin, IGF-I and IGF-II bind to their own receptors with high Žbold arrows. and to the
other receptors with lower Žthin arrows. affinity. Insulin does not bind to the IGF-II receptor. The insulin and the IGF-I receptor are homodimers consisting
of extracellular Ž a ., transmembrane and intracellular domains Žb .. The intracellular domains have protein tyrosine kinase activity that can be blocked with
lavendustin A. In general, the insulin receptor mediates metabolic effects whereas anabolic and proliferative effects are transmitted by the IGF-I receptor.
The IGF-II receptor that was not detectable in cardiac tissue by Northern blot analysis w3x is a G-protein-coupled receptor. Its exact function remains to be
elucidated.
472
A.M. Vogt et al.r CardioÕascular Research 33 (1997) 469–477
2.3. Exclusion criteria
Perfusion sites were excluded from evaluation if systolic-diastolic cardiac movements caused dislocation of
microinfusion needles or if protected areas were not totally
surrounded by pNBT-negative, necrotic tissue. Successful
DC countershock treatment of ventricular fibrillation was
not a criterion for exclusion. Since intramyocardial infusion was finished before the onset of index ischemia, a
possible dislocation of the microinfusion needles at this
time point could not interfere with the result and was
hence not an exclusion criterion.
2.4. Measurement of infarct size
At the end of the experimental protocol, the left anterior
descending coronary artery was occluded again and 1 g
fluorescein-sodium ŽFluka, Neu-Ulm, Germany, dissolved
in 10 ml saline. was injected as a bolus intravenously.
After the overall distribution of the dye the animals were
sacrificed with an intravenous bolus of 20% potassium
chloride Ž30 ml. to achieve cardiac arrest. The heart was
excised and both atria and the right ventricle were removed. The left ventricle was cut into slices along the
pairwise inserted microinfusion needles perpendicular to
the LAD. After weighing the slices, the ischemic area was
identified as the non-fluorescent area by blacklight examination. The infarcted area was demarcated using p-nitroblue-tetrazolium ŽpNBT. staining. The results of the stainings were documented on transparency film for further
planimetric evaluation. Due to the onset of rigor mortis the
heart slices sometimes shrink between drawing of the
fluorescein outlines and the tetrazolium incubation, which
makes it sometimes difficult to match the outlines of the
risk region with that of the infarct. Since the fluorescein
leaks out during incubation, it is impossible to determine
the risk region and the infarct size after the tetrazolium
bath. We have therefore replaced in some animals the
fluorescein method with fluorescent microspheres that do
not leak out during incubation. The color transparencies of
the risk region and of the infarcts were scanned into an
Apple Quadra computer and a composite image was created using ‘Photoshop’ graphics software.
2.5. Planimetry and quantification of myocardial protection
Since only a fraction of the ischemic myocardium was
reached by the infused substances, quantification of myocardial protection was done as follows:
Myocardial infarcts in the pig after 60 min coronary
occlusion and 120 min reperfusion are homogenous and
transmural. Fig. 3 demonstrates that the substances were
infused into a region that, according to our experience,
becomes necrotic afterwards. Intramyocardial infusions of
fluorescein sodium Ždone in several animals. helped us to
identify the perfused myocardial areas, as determined by
blacklight examination. PNBT-positive Žviable. myocardial
areas in the tissue close to the needles Ž3b., totally surrounded by pNBT-negative Žnecrotic. tissue, were regarded as myocardial areas protected by intramyocardial
infusion. Description and statistical evaluation of myocardial protection was done in two ways as follows. Using a
nominal scale, evaluation was done by differentiating between the infusion sites showing ‘protection’ and ‘no
protection’. These results are shown in Table 1. The other
method consisted of comparing the infarcted areas by
planimetry: The border zones between ischemic and nonischemic myocardium were not reached by our microinfusion technique, as determined by dye infusions Ž3a.. Thus,
we made sure that the outline of the infarct remained
unaffected by the active compounds. In this way, every
myocardial slice planimetred served as its own control:
Downloaded from http://cardiovascres.oxfordjournals.org/ by guest on June 9, 2014
In all animals index ischemia was a 60 min LAD
occlusion followed by 120 min reperfusion. The control
group Ž1. consisted of animals receiving intramyocardial
infusions of Krebs-Henseleit buffer for 60 min prior to
index ischemia. In group 2, recombinant human IGF-II
Ž250 ngrml, Biomol, Hamburg, Germany. was infused for
60 min. The animals of group 3 received infusions of
recombinant human insulin Ž0.02 IUrml, H-insulin
Hoechst w , Hoechst AG, Frankfurt, Germany. in an equipotent concentration with regard to IGF-II’s metabolic effects. In groups 4 and 5, either IGF-II or insulin were
co-infused with lavendustin A Ž100 mM., a potent and
selective inhibitor of protein tyrosine kinases w25x. To
make sure that protein tyrosine kinases were inhibited at
the onset of IGF-IIrinsulin infusion, lavendustin A was
infused alone 30 min before adding the other substances.
To test the expected inhibitory action of IGFBP-5, the
animals of group 6 received co-infusions of IGF-II and
recombinant human IGFBP-5 ŽAustral Biologicals, San
Ramon, CA, USA. in equimolar concentrations.
The concentrations used were determined in previous
pilot experiments. Concerning IGF-II’s cardioprotective
actions, a dose–response testing indicated no protection at
the concentrations of 2.5 and 25 ngrml. At 50 ngrml, a
small, inconstant effect was observed whereas 250 ngrml
showed reproducible, marked protective effects. IGF-II
stimulates metabolic insulin effects with a potency of
about 25% of the action of insulin itself w15,26x. For this
reason we have chosen the insulin concentration as described above Ži.e., 25% of IGF-II’s molar concentration..
IGFBP-5, a protein showing two bands at 31 and 32 kD in
SDS-PAGE w27x, was tested in concentrations of 50, 100
and 200% relative to IGF-II’s molar concentration. The
cardioprotective effect of IGF-II was inhibited at equimolar concentrations and when IGFBP-5 was co-infused.
Since IGFBP-5 shows no affinity to insulin w28x, this
experimental situation was not tested.
A.M. Vogt et al.r CardioÕascular Research 33 (1997) 469–477
473
previously described w3,29,30x. Quantification and referencing of the Northern signal were carried out by PhosphorImager as previously described w3,29,30x.
2.7. Statistical analysis
All data are given as mean values with standard error of
mean. Statistical comparisons between groups were done
by ANOVA ŽScheffe´ test. unless mentioned otherwise. A
P-value smaller than 0.05 was considered statistically significant.
3. Results
3.1. Hemodynamic data
Fig. 4 shows the hemodynamic data recorded throughout the experiments. Compared to baseline values Ž t s 0
min., intramyocardial microinfusion for 60 min did not
affect systemic hemodynamics. The time course of hemodynamics during ischemia and reperfusion follows the
First, we determined the infarcted area that would have
occurred without protection Ži.e., the total area of the
transmural infarction.. This area was set as 100%. In the
second step, we determined the infarcted area Žtransmural
area minus protected myocardial area. and related this
value to the area obtained in step one. Per animal and
substance, one mean ratio Žin percent risk region. was
determined and subjected to further statistical evaluation.
2.6. RNA isolation and Northern hybridization
Six additional animals were anesthetized, intubated and
artificially ventilated as described. In 3 animals a PTCA
catheter was advanced into the left anterior descending
coronary artery that was occluded by balloon inflation for
10 min. Catheter inflation was performed at pressures that
occluded but did not stretch the arteries. Occlusion was
checked by fluoroscopy and contrast medium injection.
The catheter was thereafter deflated and withdrawn for 30
min and again advanced under fluoroscopic guidance into
the LAD for another round of 10 min occlusion and 30
min of reperfusion. Thereafter the animal was sacrificed by
an overdose of the anesthetic plus a bolus of concentrated
KCl. The heart was removed and quickly dissected into the
ischemic-reperfused LAD region and the normal left ventricle. In 3 other pigs only the arterial cutdown was
performed and the hearts were removed after 80 min of
anesthesia, an identical amount of contrast medium injection and after sacrifice with nembutal and KCl. These
hearts served as controls; they were dissected in an identical way to that of the PTCA hearts.
RNA extraction, electrophoresis and cDNA probe fashioning were done according to standard protocols and as
Fig. 4. Hemodynamic data. Systemic hemodynmic data from all experiments shown as mean"standard error of mean. Minutes y30 to 60:
intramyocardial microinfusion. Minutes 61 to 120: LAD occlusion. Minutes 121 to 240: reperfusion. LVP s left ventricular pressure; AOP s
aortic pressure; dPrdt s first derivative of left ventricular pressure; s s
systolic; d s diastolic; HR s heart rate.
Downloaded from http://cardiovascres.oxfordjournals.org/ by guest on June 9, 2014
Fig. 3. Intramyocardial microinfusion. The microinfusion needles were
placed in pairs into the LAD territory that will become ischemic afterwards Ž3a.. Tissue distribution Žas determined by dye infusions. indicated
by dark areas. After pNBT staining to detect myocardial infarction Žlower
row., myocardial protection was seen by pNBT-positive, viable tissue
surrounding the needles Žstriped areas in 3b. which did not occur if the
infused substances were without protective effect. Myocardial protection
was expressed by relating the actual infarcted area Žwhite area in 3b, left.
to the infarcted area that would have become necrotic assuming transmural infarction Žwhite area in 3b, right..
474
A.M. Vogt et al.r CardioÕascular Research 33 (1997) 469–477
Fig. 5. Areas at risk. Bar graph showing constant areas at risk Žischemic
areas. in the different experimental groups.
3.2. Myocardial protection
Table 1 demonstrates myocardial protection achieved in
the different experimental groups. Statistical analysis of
the data by contingency table analysis shows that the
P-value would be ) 0.001 in the case of one dropout Ži.e.,
a change in the wrong direction.. Since we had no dropouts,
the P-value would be even smaller. IGF-II as well as
insulin displayed cardioprotection at every infusion site
investigated Ž n s 16. in 4 animals. These effects were
inhibited by the simultaneous infusion of lavendustin A, an
inhibitor of protein tyrosine kinases, blocking the signal
transduction of IGF-II and insulin at the receptor level.
Likewise, simultaneous infusion of IGFBP-5 in all experiments prevented protection by IGF-II.
Fig. 6. Infarct size. Infarct size expressed as percentage of control,
transmural infarction. Open symbols represent single experiments; mean
values with standard errors of mean are given by filled symbols. C s
control; IGF s IGF-II microinfusions; INS s insulin; LAV A s
lavendustin A; BP 5 s IGFBP-5. Significances between the experimental groups are given by the symbols $, %, and a ŽANOVA..
Fig. 7. Infarct imaging. Representative experiment showing the salvaging
influence of intramyocardial infusion. The non-ischemic region is in red
with light dots from the microsphere injection with the LAD re-closed at
the end of the experiment. Infarcted tissue is yellow green and ischemic
tissue surrounding the 2 infusion needles is dark blue–magenta.
Fig. 8. IGF-II mRNA expression in closed-chest pigs. Northern hybridization of blotted RNA obtained from normal and ischemic preconditioned
regions Žocclusions by balloon catheter. that were probed with an IGF-II
cDNA probe. Two experiments are shown with strong expression of the
2.8 and 6 kb bands typical of ischemia. The induction was 3.5-fold and
5-fold above control.
Downloaded from http://cardiovascres.oxfordjournals.org/ by guest on June 9, 2014
pattern as previously described w31x and is not shown here
in detail.
A.M. Vogt et al.r CardioÕascular Research 33 (1997) 469–477
Table 1
Myocardial protection in the different experimental groups using a nominal scale
Group
Protection
Yes
No
Control
IGF
INS
IGFqLAV
INSqLAV
IGFqBP5
0
16r4
16r4
0
0
0
16r4
0
0
16r4
16r4
16r4
The first number indicates the slices evaluated, the second number the
animals investigated. For further explanation, see text.
3.3. IGF-II expression
IGF-II mRNA is 4-fold overexpressed following coronary occlusion by balloon inflation in two animals Žsee
Fig. 8.. The third balloon-occluded animal died of ventricular fibrillation. The data of 2 control animals are not
shown because no increased IGF mRNA expression was
noted. In 1 control animal the RNA was degraded. No
statistical analysis was attempted because of the small
number of animals and the large differences in mRNA
expression between balloon-occluded and control tissue.
4. Discussion
In a previous paper w3x we described increased and very
fast IGF-II mRNA expression following repeated brief
ischemia under open-chest conditions in the pig. Since the
expression of the IGF binding protein 5, which interferes
with the action of IGF-II, was also upregulated, albeit with
a phase lag of about 30 min, we hypothesized that the
protection caused by ischemic preconditioning can also be
explained by the action of the IGF system where protection is initiated by IGF-II and terminated by IGFBP-5.
Since this hypothesis cannot be proven by mRNA data
alone and since available antibodies directed at human
recombinant IGF-II do not cross-react with the porcine
homolog, we tested this idea by direct intramyocardial
infusion of the peptides involved. Our results show that
indeed IGF-II delays the progression of an experimental
infarct just like ischemic preconditioning and IGFBP-5
neutralizes this effect. We show further that the protective
effect is mediated via the insulin receptor which belongs to
the tyrosine kinase class of receptors and that the effect
can be neutralized by the tyrosine kinase inhibitor lavendustin. IGF-II binds either to the IGF-I receptor, the
IGF-IIrmannose 6-phosphate or to the insulin receptor and
has hence either a mitogenic or a metabolic effect. Our
results suggest that the cardioprotection is based on the
metabolic effects of IGF-II because:
Ø insulin itself produced a similar degree of protection,
Ø insulin does not bind to the IGF-II and only weakly to
the IGF-I receptor,
Ø neither the IGF-I nor the IGF-II receptor produces a
metabolic effect w5,32x and
Ø the IGF-II receptor is not expressed under the experimental conditions that we have created w33x, which is
compatible with the low number of IGF-II receptors
reported for the adult heart w1,34x. The mRNAs of the
insulin and the IGF-I receptor, on the other hand, are
abundantly expressed in porcine heart.
To the best of our knowledge, this is the first report
describing cardioprotection by one of the insulin-like
growth factors. In contrast to the heart, the role of the
IGF’s is better investigated in neuronal models. Recent
investigations reported a beneficial role for the IGF’s in
brain ischemia w6,9x, and the effect was also produced by
its metabolic effects w6,7x. In the heart, influencing glucose
metabolism has been shown to be beneficial in
ischemiarreperfusion settings w35–37x and our preliminary
experiments with positron emission tomography Žtogether
with Schwaiger, unpublished. showed increased focal glucose uptake after intramyocardial IGF-II infusion, from
which we conclude that this was one of the underlying
mechanisms responsible for the increased ischemia resistance of the myocardium w15,16x. Interestingly, increased
myocardial glucose uptake is a characteristic feature also
of ‘hibernating myocardium’ w1,38x to protect itself from
sustained ischemic stress w39x. Thus, the increased mRNA
content for IGF-II may be interpreted to take place as a
self-defense mechanism of myocardium subjected to repeated episodes of ischemia. Thereby the heart circumvents its inability to synthesize insulin itself. Instead, it
increases its production of insulin-like growth factor II,
‘the heart’s own insulin’, to increase its tolerance against
subsequent ischemic episodes.
Since IGF-II mRNA is markedly overexpressed in preconditioned porcine myocardium and IGF-II peptide was
demonstrated to be protective in the setting of cardiac
ischemia and reperfusion, an involvement of IGF-II in
preconditioning’s cardioprotection appears likely since the
translation of small-peptide hormones may happen quickly,
even in ischemia w17x. The delay that is necessary for
ischemic preconditioning to become manifest may correspond to the time necessary to translate IGF-II mRNA.
Although recent reports in the rabbit suggest ischemic
preconditioning to be independent of de noÕo protein
synthesis w40x, the conclusion rested on the assumption that
the dose of cycloheximide was sufficient to inhibit global
protein synthesis. This may not apply to small peptides
like the IGFs. The 30 min delay with which IGFBP-5
mRNA expression follows the expression of IGF-II w3x
could explain why the memory effect of ischemic precon-
Downloaded from http://cardiovascres.oxfordjournals.org/ by guest on June 9, 2014
Fig. 5 demonstrates that the data for areas at risk did
not differ significantly in the groups investigated. The
conventional way of expressing infarct size is shown in
Fig. 6. An example of infarct imaging is given in Fig. 7.
475
476
A.M. Vogt et al.r CardioÕascular Research 33 (1997) 469–477
inhibits the cardioprotection afforded by IGF-II and thereby
possibly accounts for the limited protective time frame of
ischemic preconditioning. IGF-II and adenosine cooperate
by increasing the glucose uptake in ischemic-reperfused
myocardium and they may act via translocation of the
glucose transporter.
4.1. Limitations of the study
The method of direct infusion of tool drugs into the
myocardium was developed for the study of focal arrhythmias w46x. We show here and in a previous paper w24x that
it is also useful for the application of hormones that act
mainly locally and that cannot be applied systemically in
large animals due to cost and to serious side-effects. The
disadvantage of the method is that the concentration of the
hormone at some distance from the infusion needle is not
known. With continuous infusion the gradients of concentration will decrease, but the final space of distribution
remains unknown. Experiments with dyes showed that at a
rate of infusion of 20 mlrmin a volume of approximately
1 cm3 is reached within 10 min. Longer infusion times
increase the volume of distribution, but we do not at
present know where the saturation point is.
References
w1x Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium.
Circulation 1986;74:1124–1136.
w2x Schaper W, Zimmermann R, Kluge A, et al. Patterns of myocardial
gene expression after cycles of brief coronary occlusion and reperfusion. Ann NY Acad Sci 1994;723:284–291.
w3x Kluge A, Zimmermann R, Munkel
B, Verdouw P, Schaper J,
¨
Schaper W. Insulin-like growth factor II is an experimental stress
inducible gene in a porcine model of brief coronary occlusions.
Cardiovasc Res 1995;29:708–716.
w4x Cochik WS, Clemmons DR. The insulin-like growth factors. Annu
Rev Physiol 1993;55:131–153.
w5x Sara VR, Hall K. Insulin-like growth factors and their binding
proteins. Physiol Rev 1990;70:591–614.
w6x Zhu CZ, Auer RN. Intraventricular administration of insulin and
IGF-1 in transient forebrain ischemia. J Cereb Blood Flow Metab
1994;14:237–242.
w7x Zhu CZ, Auer RN. Centrally administered insulin and IGF-1 in
transient forebrain ischemia in fasted rats. Neurol Res 1994;16:116–
120.
w8x Strong AJ, Fairfield JE, Monteiro E, et al. Insulin protects cognitive
function in experimental stroke. J Neurol Neurosurg Psychiatry
1990;53:847–53.
w9x Hamilton MG, Tranmer BI, Auer RN. Insulin reduction of cerebral
infarction due to transient focal ischemia. J Neurosurg 1995;82:262–
268.
w10x Aoki M, Abe K, Kawagoe J, Nakamura S, Kogure K. The preconditioned hippocampus accelerated HSP70 heat shock gene expression
following transient ischemia in the gerbil. Neurosci Lett
1993;155:7–10.
w11x Aoki M, Abe K, Kawagoe J, Nakamura S, Kogure K. Acceleration
of HSP70 and HSC70 heat shock gene expression following transient ischemia in the preconditioned gerbil hippocampus. J Cereb
Blood Flow Metab 1993;13:781–788.
Downloaded from http://cardiovascres.oxfordjournals.org/ by guest on June 9, 2014
ditioning Ži.e., the fact that it ‘remembers’ short occlusions. is ‘forgotten’ after less than 1 hour following the
last conditioning short occlusion and that cardioprotection
is no longer present after that time w21x. By extending the
time period after the repetitive coronary occlusions the
increase in IGFBP-5 content will terminate preconditioning’s cardioprotection: i.e., the protection observed is the
resultant of the myocardial content of IGF-II Žprotective.
and IGFBP-5 Žinhibitory..
This proposed mechanism explains the observation that
in our porcine model ischemic preconditioning is not yet
optimal when shorter protocols of repetitive coronary occlusions w41x are used Ž2 occlusions of 10 min or 4
occlusions of 5 min are needed for protection, but only the
first produced significant amounts of adenosine. in contrast to other species w24,42x. This would suggest that the
adenosine hypothesis in the porcine heart can only partially explain the protective effect of short occlusions and
that an additional factor has to be assumed. This is most
likely the IGF system: the multiple occlusions needed for
protection in our model produce the necessary amounts of
IGF-II. The functional synergism between adenosine and
insulin with regard to glucose uptake that is stimulated by
both agents is indeed very helpful for cytoprotection
w43,44x. Although it was shown that neither insulin nor
adenosine increase glucose uptake during prolonged ischemia, the reperfusion periods in between brief occlusions may stimulate glucose uptake. Other properties of
IGF-II-like inhibition of protein breakdown and preservation of translation may add to the cytoprotective profile of
the hormone w45x.
One argument encountered during the review of our
previous paper was that the upregulation of IGF-II mRNA
expression may not be specific for ischemia but may also
have been caused by the surgical stress of the open-chest
situation, because a statistically non-significant trend
seemed to indicate that. Furthermore, there was no difference in IGF-mRNA between ischemic-reperfused and adjacent non-ischemic myocardium. This argument can now be
placed into proper perspective with our new data that show
upregulation of IGF-II mRNA expression in a closed-chest
model only with ischemia and in the ischemic region Žsee
Fig. 8.. We conclude from these and from the previous
findings that both surgical stress and ischemia proper are
able to upregulate IGF-II mRNA. On the basis of mRNA
data alone it cannot be inferred that IGF-II may have a
protective effect because it could have been an epiphenomenon related to stress. However, our direct infusion
studies show clearly the protective potency of the hormone.
In summary, our data describe a cardioprotective effect
of IGF-II, a tissue hormone with insulin-like actions whose
mRNA is 4- to 5-fold overexpressed in preconditioned
myocardium. The cardioprotection caused by IGF-II is
mediated via the insulin receptor. IGFBP-5, whose expression follows the expression of IGF with a short delay,
A.M. Vogt et al.r CardioÕascular Research 33 (1997) 469–477
w29x Kluge A, Zimmermann R, Munkel
B, et al. Insulin-Like Growth
¨
Factor II ŽIGF-II. ist ein Stress-Gen und ein trophischer Faktor im
Herzmuskel des Schweines. Z Kardiol 1993;82ŽSuppl 1.:521ŽAbstract..
w30x Kluge A, Zimmermann R, Munkel
B, Verdouw PD, Schaper J,
¨
Schaper W. Localization of insulin-like growth factor II ŽIGF-II., a
stress-response gene, in ischemic porcine heart. J Mol Cell Cardiol
1993;25Žsuppl I.:III-P-4ŽAbstract..
w31x Schott RJ, Rohmann S, Braun ER, Schaper W. Ischemic preconditioning reduces infarct size in swine myocardium. Circ Res
1990;66:1133-1142.
w32x Kornfeld S. Structure and function of the mannose 6-phosphaterinsulin like growth factor II receptors. Annu Rev Biochem
1992;61:307–330.
w33x Kluge A, Zimmermann R, Weihrauch D, et al. Coordinate expression of the insulin-like growth factor system after microembolisation
in porcine heart. Cardiovasc Res 1996;in press.
w34x Roth RA, Kiess W. Insulin-like growth factor: recent developments
and new methodologies. Growth Regul 1994;1:31–38.
w35x Apstein C, Gravino F, Haudenschild C. Determinants of a protective
effect of glucose and insulin on the ischemic myocardium. Circ Res
1983;52:515–526.
w36x Opie L, Bruyneel K, Owen P. Effects of glucose, insulin and
potassium infusion on tissue metabolic changes within first hour of
myocardial infarction in baboon. Circulation 1975;52:49–57.
w37x Maroko PR, Kjekshus JK, Sobel BE, et al. Factors influencing
infarct size following experimental coronary artery occlusion. Circulation 1971;43:67–82.
w38x Rahimtoola SH. Chronic myocardial hibernation. Circulation
1994;89:1907.
w39x Camici P, Ferranini E, Opie LH. Myocardial metabolism in ischemic
heart disease: basic principles and application to imaging by positron
emission tomography. Prog Cardiovasc Dis 1989;32:217–238.
w40x Thornton J, Striplin S, Liu GS, et al. Inhibition of protein synthesis
does not block myocardial protection afforded by preconditioning.
Am J Physiol 1990;259:H1822–H1825.
w41x Strasser R, Arras M, Elsasser
A, et al. Preconditioning of porcine
¨
myocardium: How much ischemia is required for induction? What is
its duration? Is a renewal effect possible? Circulation 1994;90Žpart
2.:579ŽAbstract..
w42x Lawson CS, Downey JM. Preconditioning: state of the art myocardial protection. Cardiovasc Res 1993;27:542–550.
w43x Law WR, McLane MP. Adenosine enhances myocardial glucose
uptake only in the presence of insulin. Metabolism 1991;40:947–952.
w44x Law W, McLane M, Raymond R. Adenosine is required for myocardial insulin responsiveness in vivo. Diabetes 1988;37:842–845.
w45x Dale WE, Hale CC, Kim HD, Rovetto MJ. Myocardial glucose
utilization: failure of adenosine to alter it and inhibiton by the
adenosine analogue N 6 -ŽL-2-phenylisoprophyl. adenosine. Circ Res
1991;69:791–799.
w46x Podzuweit T. Catecholamine-cyclic AMP-Ca2q-induced ventricular
tachycardia in the intact pig heart. Basic Res Cardiol 1980;75:772–
779.
Downloaded from http://cardiovascres.oxfordjournals.org/ by guest on June 9, 2014
w12x Liu Y, Kato H, Nakata N, Kogure K. Temporal profile of heat shock
protein-70 synthesis in ischemic tolerance induced by preconditioning ischemia in rat hippocampus. Neuroscience 1993;56:921–927.
w13x Gidday JM, Fitzgibbons JC, Shah AR, Park TS. Neuroprotection
from ischemic brain injury by hypoxic preconditioning in the neonatal rat. Neurosci Lett 1994;168:221–224.
w14x Kato H, Liu Y, Kogure K, Kato K. Induction of 27-kDa beat shock
protein following cerebral ischemia in a rat model of ischemic
tolerance. Brain Res 1994;634:235–244.
w15x Burguera B, Elton CW, Tapscott EB, Pories WJ, Dimarchi R,
Sakano K. Stimulation of glucose uptake by insulin-like growth
factor IIrmannose 6-phosphate receptor. Biochem J 1994;300:781–
785.
w16x Bevan SJ, Parry-Billings M, Opara E, Liu CT, Dunger DB, Newsholme EA. The effect of insulin-like growth factor II on glucose
uptake and metabolism in the rat skeletal muscle in vitro. Biochem J
1992;286:561–565.
w17x Tonnessen T, Giaid A, Saleh D, Neass PA, Yanagisawa M, Christensen G. Increased in vivo expression and production of endothelin-1
by porcine cardiomyocytes subjected to ischemia. Circ Res
1995;76:767–772.
w18x Draznin B, Chang L, Leitner JW, Takata Y, Olefsky JM. Insulin
activates p21Ras and guanine nucleotide releasing factor in cells
expressing wild type and mutant insulin receptors. J Biol Chem
1993;268:19998–20001.
w19x O’Dell TJ, Kandel ER, Grant SG. Long-term potentiation in the
hippocampus is blocked by tyrosine kinase inhibitors. Nature
1991;353:558–560.
w20x Hellenius M, Brismar K, Gerglund B, De Faire U. Effects on
glucose tolerance insulin secretion, insulin-like growth factor 1 and
its binding protein, IGFBP-1, in a randomized controlled diet and
exercise study in healthy, middle aged men. J Intern Med
1995;238:121–130.
w21x Sack S, Mohri M, Arras M, Schwarz ER, Schaper W. Ischaemic
preconditioning—time course of renewal in the pig. Cardiovasc Res
1993;27:551–555.
w22x Lewitt MS. Role of the insulin-like growth factors in the endocrine
control of glucose homeostasis. Diabetes Res Clin Pract 1994;23:3–
15.
w23x Vogt A, Barancik M, Weihrauch D, Arras M, Podzuweit T, Schaper
W. Protein kinase C inhibitors reduce infarct size in pig hearts in
vivo. Circulation 1994;90:3486ŽAbstract..
w24x Vogt AM, Htun P, Arras M, Elsasser
A, Podzuweit T, Schaper W.
¨
Intramyocardial infusion of tool drugs for the study of molecular
mechanisms in ischemic preconditioning. Basic Res Cardiol
1996;91:389–400.
w25x Burke TJ. Protein-tyrosine kinases: potential targets for anticancer
drug development. Stem Cells Dayt 1994;12:1–6.
w26x Froesch ER, Schmid C, Schwander J, Zapf J. Actions of insulin-like
growth factors. Annu Rev Physiol 1985;47:443–467.
w27x Cohik W, Clemmons D. The insulin-like growth factors. Annu Rev
Physiol 1985;55:131–153.
w28x Herington AC. Insulin-like growth factors: biochemistry and physiology. Baillieres Clin Endocrinol Metab 1991;5:531–551.
477