(Hypertension. 1999;34:1053-1059.)
© 1999 American Heart Association, Inc.
Scientific Contributions |
From the Division of Cardiology (M.B., J.C., P.D.), University Hospital of Geneva, Switzerland; and the Renal Division (S.R.P., W.E.M.), Emory University Medical School, Atlanta, Ga.
Correspondence to Marijke Brink, PhD, Department of Medicine, Division of Cardiology, University of Geneva, Ave de la Roseraie 64, CH-1211 Geneva, Switzerland. E-mail marijke.brink{at}dim.hcuge.ch
| Abstract |
|---|
|
|
|---|
Key Words: hypertrophy, cardiac heart failure growth substances angiotensin II immunohistochemistry
| Introduction |
|---|
|
|
|---|
Angiotensin (Ang) II is another important factor in this context; it causes hypertrophy not only through its blood pressureraising action, but probably also through direct trophic effects on the fibroblast or cardiomyocyte.14 Whereas the initial hypertrophic response to overload seems to support contractility of the heart, continuous exposure to high blood pressure will result in pathological hypertrophy. The beneficial effects of angiotensin-converting enzyme inhibitors or Ang II receptor antagonists in animal models of cardiac hypertrophy as well as in human cardiac disease strongly support that Ang II is involved in this pathological hypertrophy.
Diseases such as congestive heart failure and chronic renal failure frequently go hand in hand with cachexia.15 16 In fact, cachexia was shown to be an independent risk factor leading to increased mortality in congestive heart failure.17 Moreover, cachexia can be associated with low circulating levels of IGF-I.18 Conditions of GH or IGF-I deficiency in humans are associated with cardiac atrophy and impaired cardiac function.13 If it is true that IGF-I supports cardiac function, low circulating IGF-I levels may be one of the direct reasons for increased mortality in patients with congestive heart failure.
We have recently reported that infusion of rats with Ang II through osmotic minipumps results in significant body weight loss as well as a strong reduction in circulating levels of IGF-I independent of increased blood pressure.19 The present study uses the same model to investigate whether circulating IGF-I, cardiac IGF-I, and cardiac IGFI receptor(IGF-IR) are involved in the hypertrophic response to Ang II infusion. First, we determined whether the IGF-I and IGF-IR mRNAs were upregulated in the heart, despite the downregulation of circulating IGF-I. Next, we analyzed whether the observed changes in IGF-I mRNA were direct effects of Ang II at the level of the cardiomyocyte or whether they were caused by the blood pressureraising potential of Ang II using the vasodilator hydralazine. Finally, we investigated the effect of augmenting circulating IGF-I levels on cardiac mass by coinfusing rats with Ang II and IGF-I.
| Methods |
|---|
|
|
|---|
To determine the effect of blood pressure, control- or Ang IIinfused rats received drinking water with or without hydralazine (10 mg · kg-1 · d-1) or losartan (25 mg · kg-1 · d-1). Losartan was kindly provided by Dr R. Smith, DuPont Merck, Wilmington, Del. Both antihypertensive treatments were started 2 days before implantation of the minipumps.
Three, 7, or 14 days after implantation of the pumps, rats were anesthetized and euthanatized after aortic blood was taken, mixed with EDTA in prechilled glass tubes, and transferred to ice. Plasma was obtained after centrifugation for 10 minutes at 4°C and 2000 rpm and stored at -20°C until analysis by radioimmunoassay as described previously.19 The heart was excised and the left ventricle was dissected, weighed, and snap-frozen in liquid nitrogen.
IGF-I and IGF-IR mRNA Levels
Total RNA was prepared from frozen left ventricular
muscle with an acid phenol-guanidinium reagent (Tri-Reagent, Molecular
Research Center) according to the manufacturers instructions. RNA was
quantified and purity was assessed by measuring absorptions at 260 and
280 nm. Solution hybridization and RNase protection assays were
performed as previously described.2 Briefly, 20 µg of
total RNA was hybridized overnight in hybridization buffer (Ambion,
Inc) with 5x105 cpm riboprobe (see below). After
RNase digestion with 40 µg/mL ribonuclease A and 100 U/mL
ribonuclease T1 was complete, samples
underwent treatment with proteinase K, phenol extraction, ethanol
precipitation, and analysis on a 6% denaturing
polyacrylamide urea sequencing gel.
For determination of IGF-I mRNA levels, 2 riboprobes were used; initial studies were performed with a probe transcribed from a plasmid provided by Dr P. Rotwein, Washington University School of Medicine, St Louis, Mo, that included exon 4 and noncoding sequence as described previously.2 The 956-base riboprobe gives a 182-bp protected band after hybridization and RNase digestion. For later studies, a new plasmid was created by subcloning the full-length rat IGF-I cDNA sequence20 (provided by Dr G.I. Bell, Howard Hughes Medical Institute, Chicago, Ill) into the EcoRI site of pGEM-9Zf(-). Transcription with SP6 of the Hinf I linearized plasmid results in a full-length probe of 299 bases, and after hybridization and RNase digestion, the protected fragment is 263 bp long. The 2 probes gave identical results in terms of changes in IGF-I mRNA levels in response to Ang II infusion.
For determination of IGF-IR mRNA levels, a 203-bp EcoRI and KpnI rat IGF-IR cDNA fragment in pGEM-3Z was used to generate a radiolabeled antisense probe, as previously described.21 The full-length probe is 251 bp, and the protected fragment is 195 bp in length. As an additional control for RNA loading and to determine the specificity of observed changes in IGF-IR mRNA levels, a GAPDH riboprobe was included in the hybridization mixture. This probe gives a 133-bp protected band after RNase digestion. Autoradiographic signals for IGF-I and IGF-IR were quantified in arbitrary units by use of a phosphoimager (Molecular Dynamics), and signals were normalized by use of the corresponding GAPDH values to correct for variations in RNA loading.
Tissue Extraction and Radioimmunoassay for IGF-I
Tissue concentrations of IGF-I were determined as described by
DErcole et al.22 Briefly, left ventricles were
pulverized under liquid nitrogen, further disrupted using a Potter
homogenizer, and extracted in 1 mol/L acetic acid for 2
hours at 4°C. After centrifugation (10 minutes at
1000g) was completed, the supernatants were neutralized with
NaOH. Protein concentration of the tissue extract was determined by use
of the BioRad protein assay. Radioimmunoassays were performed as
described previously19 with 40 µg of protein
extracted from each left ventricle. IGF-I polyclonal antibody UB 286
(raised by Drs L.E. Underwood and J.J. van Wyk, University of
North Carolina, Chapel Hill) was donated by the US National Hormone and
Pituitary Program.
Immunohistochemistry
For immunocytochemistry, left ventricles were fixed with 4%
formaldehyde and embedded in paraffin with routine procedures.
Deparaffinized sections 6 µm thick were preincubated for 1 hour
with 5% normal goat serum (Vector Laboratories, Inc) in PBS, incubated
overnight with the polyclonal anti-IGF-I antibody described above
(dilution, 1:1000), washed 4x for 15 minutes with PBS, and then
incubated with a biotin-conjugated goat anti-rabbit antibody (Vector
Laboratories) for 1 hour. Finally, the sections were incubated for 30
minutes with an alkaline phosphataseconjugated biotin-streptavidin
complex (Vector Laboratories), following the suppliers instructions.
The antibody binding sites were visualized with fast red (Vector
Laboratories). The sections were counterstained with Harris
hematoxylin and mounted with GVA mounting solution (Zymed).
Statistical Analysis
All data represent means of 4 to 10 rats per group per
time point, and values are expressed as mean±SE. Results were
analyzed by unpaired Students t test when results
from 2 experimental groups were compared or by ANOVA when data from
3
groups were studied. For data analyzed by ANOVA, pairwise
comparisons were made by Tukeys test.
| Results |
|---|
|
|
|---|
|
Thus, Ang II infusion increases cardiac IGF-I mRNA expression, in contrast to its known suppressor effect on circulating IGF-I levels.19 To determine whether Ang II downregulation of circulating IGF-I resulted in autocrine induction of cardiac IGF-I through a feedback mechanism, we designed an additional experiment. We reversed the suppressive effect of Ang II on circulating IGF-I levels by coinfusing recombinant IGF-I with Ang II. Under this condition, cardiac IGF-I mRNA expression was significantly (P<0.05) increased compared with pair-fed rats (Figure 1B) at both time points. The difference in IGF-I mRNA levels between Ang IIinfused and Ang II and IGF-Iinfused rats at day 7 did not reach statistical significance. These data indicate that Ang II induces left ventricular IGF-I mRNA through a mechanism independent of food intake or changes in circulating IGF-I.
Effect of Ang II Infusion on IGF-I Protein Expression
To determine whether the Ang IIinduced increase in cardiac IGF-I
mRNA correlates with an increase in IGF-I protein, we prepared protein
extracts from the left ventricle of 5 Ang IIinfused rats and 6
control rats and measured IGF-I using a radioimmunoassay. Left
ventricles from Ang IIinfused rats contained significantly more IGF-I
than ventricles from control rats (23% increase; P<0.001;
Figure 1C). To localize IGF-I expression in the heart, we
performed immunohistochemistry using paraffin sections from control and
Ang IIinfused rats. As shown in Figure 2, IGF-I immunoreactivity was
predominantly located on the cardiomyocytes. Control
sections using nonimmune serum showed no staining (data not shown).
|
Effect of Antihypertensive Treatment on Ang IIInduced Changes in
Left Ventricular IGF-I mRNA
Pharmacological treatment of rats with either losartan or
hydralazine beginning 2 days before implantation of the
minipumps reduced blood pressure to control levels, as published
previously.19 Cardiac IGF-I mRNA expression in these
animals is shown in Figure 3
(representative RNase protection assay); quantification
of results from 4 to 6 animals in each experimental group demonstrated
that Ang II raised IGF-I mRNA levels (normalized with GAPDH)
1.7±0.1-fold at day 7 compared with ad libitum-fed controls
(P<0.01). This Ang IIinduced increase was blocked by
administration of losartan (Figure 3B). Losartan
alone did not affect IGF-I mRNA levels versus results from control
rats. Normalization of blood pressure by hydralazine also
prevented the increase in IGF-I mRNA levels in Ang IIinfused rats,
because rats treated with Ang II and hydralazine had IGF-I mRNA
levels that were indistinguishable from control rats receiving
hydralazine or from control rats not given pharmacological
treatment. In conclusion, both losartan and hydralazine
normalized blood pressure and prevented the induction of left
ventricular IGF-I mRNA expression. Thus, the increase in
left ventricular IGF-I mRNA is a response to the
hemodynamic action of Ang II, an
AT1 receptormediated effect.
|
Effect of Ang II and IGF-I Infusion on Left Ventricular
Weight
The upregulation of cardiac IGF-I mRNA and protein by Ang II
suggests that autocrine IGF-I may be involved in the hypertrophic
response of the left ventricle to increased hemodynamic
load. Circulating IGF-I produced by the liver may also be important for
cardiac growth, but we have recently shown that circulating IGF-I
levels are dramatically reduced in this model of hypertension. This
response could negatively affect cardiac growth processes. IGF-I has
been shown to be a potent survival factor for
myocytes.23 24 25 To investigate whether increasing
circulating IGF-I levels will enhance the hypertrophic response, we
coinfused Ang II and IGF-I. This resulted in a significant, 2.2-fold
increase in circulating IGF-I levels as measured by radioimmunoassay
and caused an increase in left ventricular mass: after 7
days, cardiac mass of IGF-I and Ang IIcoinfused rats was 14% higher
than in rats infused with Ang II alone (P<0.05). The
increase in left ventricular weight in rats coinfused with
Ang II and IGF-I was 26% more than pair-fed controls
(P<0.01). Figure 4 shows a
summary of the results after calculation of the hypertrophic index
(left ventricular weight in mg divided by body weight in
g). The Ang II and IGF-Iinduced increases in cardiac mass become
even more pronounced when expressed as hypertrophic indices, because
Ang II infusion caused a body weight loss. Thus, Ang II increased the
hypertrophic index by 20% at 7 days (P<0.01) and an
additional increase of 16% occurred coinfusion of IGF-I and Ang II
(P<0.01).
|
To determine whether cardiac hypertrophy was also detectable at lower doses of Ang II, we performed an experiment using 500, 350, and 200 ng · kg-1 · min-1. At 7 days, the hypertrophic index was increased by 43±7%, 20±4%, and 26±7% in Ang IIinfused rats, respectively, compared with pair-fed controls (P<0.01, P<0.01, and P<0.05, respectively). No difference in cardiac index between pair-fed and ad libitumfed rats was measured (data not shown).
Effect of Ang II Infusion on IGF-IR mRNA
We measured left ventricular IGF-IR mRNA by RNase
protection assay. Ang II caused an increase in IGF-IR mRNA both at 1
(Figure 5A) and 2 weeks of infusion
compared with vehicle-infused control animals that were fed ad libitum.
At 7 days, a 2.1±0.05-fold increase (P<0.01) occurred in
left ventricular IGF-IR mRNA (Figure 5B). A similar
increase in IGF-IR mRNA occurred in hearts of pair-fed control rats;
ie, no difference was detectable between Ang II and pair-fed control
animals, which indicates that the induction of receptor mRNA is due to
the anorexia induced by Ang II treatment. Thus, low food intake results
in increased IGF-IR mRNA expression in the left ventricle. The increase
in IGF-IR mRNA levels at 7 days is not due to low circulating IGF-I
levels, because increasing the circulating IGF-I protein levels by
coinfusing IGF-I and Ang II still resulted in elevated levels of IGF-IR
mRNA (Figure 5B).
|
| Discussion |
|---|
|
|
|---|
Notably, circulating IGF-I levels are dramatically downregulated in the Ang IIinfusion model.19 Despite this, Ang II induced an increase in cardiac mass, consistent with an active role for the local pressure-induced IGF-I mRNA. Direct evidence that circulating IGF-I also contributes to the hypertrophic response comes from the experiment in which coinfusion of Ang II and IGF-I resulted in an increase in left ventricular mass that was significantly greater than in rats infused with Ang II alone. This strongly suggests that both circulating as well as local IGF-I are important in the response to hemodynamic load.
The clinical relevance of our findings is related to the pathophysiology of hyperreninemic states such as congestive heart failure29 30 31 or chronic renal failure.16 These diseases are commonly complicated by cachexia, and it has been proposed that low IGF-I levels contribute to the increased mortality seen in heart failure patients.17 In this context, the other main finding of the present study, namely that IGF-IR mRNA is induced by anorexia, is important because the higher level of the receptor could facilitate trophic effects of IGF-I in the cardiomyocytes. The observation corresponds well with other studies that show that IGF-IR gene expression is highly regulated under many physiological and pathological conditions.32 In vivo, fasting increased IGF-Ispecific binding in several tissues, and these changes were accompanied by increases in IGF-IR mRNA abundance.33 Often an increase in IGF-IR expression seems secondary to a decrease in the local tissue IGF-I concentration, because in some of the tissues, the local levels of IGF-I are decreased after a reduction in caloric intake.34 Our results clearly show that this is not the case in the heart, because local levels of IGF-I are increased, and, moreover, increased circulating levels of IGF-I by coinfusion of Ang II and IGF-I does not abolish the upregulation of IGF-IR mRNA.
Our results that Ang II causes hypertrophy even at lower doses are consistent with previous reports.35 36 Data published by Kim et al,37 who have analyzed left ventricular dry weight of Ang IIinfused animals, support that the hypertrophy is due to actual growth. Others have reported increased cardiac protein synthesis,38 increased fibroblast proliferation,39 and an induction of fibronectin expression associated with fibrosis.40 It has been suggested that in experimental arterial hypertension, myocyte and nonmyocyte compartments are under separate controls: myocyte hypertrophy is most closely related to ventricular loading, whereas circulating Ang II regulates interstitial fibrosis.41 42 Our new finding that IGF-I is located on the cardiomyocytes suggests strongly that IGF-I is involved in the myocyte response to increased afterload.
Although the mechanism whereby IGF-I and its receptor increase left ventricular mass in the model of Ang II infusion remains to be elucidated, studies in animals or cultured cells suggest that it may involve the prevention of apoptosis,23 24 25 the stimulation of cells to reenter the cell cycle,43 or the stimulation of anabolic pathways,44 such as the enhancement of myofibril development45 46 or the reversion of hypertension-induced changes in cardiac protein expression.47 12 Regardless, in vivo evidence suggests that IGF-I may ameliorate myocyte growth and contractility and thereby sustain ventricular pump function. GH treatment improved cardiac function by increasing myocardial contractility in rats with postinfarction left ventricular dysfunction.48 IGF treatment resulted in improved hemodynamic parameters in rats with cardiac failure.45 In mice, treatment with GH and IGF-I induced hypertrophy and produced a positive inotropic effect without significant changes in expression of fetal and other selected myocardial genes, which suggests that treatment with GH and IGF I induces a physiological type of hypertrophy.12 In humans, a preliminary study of GH in the treatment of dilated cardiomyopathy demonstrated that myocardial mass was increased and hemodynamics and myocardial energy metabolism were improved.8 Recently, a study demonstrated that in patients with dilated cardiomyopathy given recombinant human GH, a significant increase occurs in left ventricular mass; however, no improved function was seen.49
In summary, our data demonstrate that Ang II infusion increases cardiac IGF-I mRNA expression by a hemodynamic mechanism and increases cardiac IGF-IR mRNA levels by a nonhemodynamic mechanism related to reduced food intake. The cardiac hypertrophic response to Ang II occurs despite a strong reduction in circulating IGF-I caused by Ang II. Coinfusion of IGF-I and Ang II produces an additional hypertrophic response, which suggests strongly that the reduction in circulating IGF-I induced by Ang II may partially blunt the cardiac hypertrophic response that accompanies induction of the autocrine cardiac IGF-I system by Ang II. The interaction between cardiac and systemic IGF-I regulation likely plays an important role in cardiac remodeling in response to Ang II.
| Acknowledgments |
|---|
Received February 22, 1999; first decision March 29, 1999; accepted July 9, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
B.-W. Wang, H. Chang, P. Kuan, and K.-G. Shyu Angiotensin II activates myostatin expression in cultured rat neonatal cardiomyocytes via p38 MAP kinase and myocyte enhance factor 2 pathway J. Endocrinol., April 1, 2008; 197(1): 85 - 93. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Abe, T. Matsunaga, K. Kameda, H. Tomita, T. Fujiwara, H. Ishizaka, H. Hanada, K. Fukui, I. Fukuda, T. Osanai, et al. Increased Level of Pericardial Insulin-Like Growth Factor-1 in Patients With Left Ventricular Dysfunction and Advanced Heart Failure J. Am. Coll. Cardiol., October 3, 2006; 48(7): 1387 - 1395. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. T. Nguyen, N. Cao, J. L. Short, and P. J. White Intravenous Insulin-like Growth Factor-I Receptor Antisense Treatment Reduces Angiotensin Receptor Expression and Function in Spontaneously Hypertensive Rats J. Pharmacol. Exp. Ther., September 1, 2006; 318(3): 1171 - 1177. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. E. Larkin, B. C. Frank, R. M. Gaspard, I. Duka, H. Gavras, and J. Quackenbush Cardiac transcriptional response to acute and chronic angiotensin II treatments Physiol Genomics, July 8, 2004; 18(2): 152 - 166. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Zahradka, B. Litchie, B. Storie, and G. Helwer Transactivation of the Insulin-Like Growth Factor-I Receptor by Angiotensin II Mediates Downstream Signaling from the Angiotensin II Type 1 Receptor to Phosphatidylinositol 3-Kinase Endocrinology, June 1, 2004; 145(6): 2978 - 2987. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. de Simone, F. Pasanisi, and F. Contaldo Link of Nonhemodynamic Factors to Hemodynamic Determinants of Left Ventricular Hypertrophy Hypertension, July 1, 2001; 38(1): 13 - 18. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Vecchione, S. Colella, L. Fratta, M. T. Gentile, G. Selvetella, G. Frati, B. Trimarco, and G. Lembo Impaired Insulin-Like Growth Factor I Vasorelaxant Effects in Hypertension Hypertension, June 1, 2001; 37(6): 1480 - 1485. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Brink, S. R. Price, J. Chrast, J. L. Bailey, A. Anwar, W. E. Mitch, and P. Delafontaine Angiotensin II Induces Skeletal Muscle Wasting through Enhanced Protein Degradation and Down-Regulates Autocrine Insulin-Like Growth Factor I Endocrinology, April 1, 2001; 142(4): 1489 - 1496. [Abstract] [Full Text] |
||||
![]() |
A. S. Pachori, H. Wang, C. H. Gelband, C. M. Ferrario, M. J. Katovich, and M. K. Raizada Inability to Induce Hypertension in Normotensive Rat Expressing AT1 Receptor Antisense Circ. Res., June 9, 2000; 86(11): 1167 - 1172. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1999 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |