Sage Journals: Discover world-class research

Abstract

This study aims to investigate the expression of metalloproteinases (MMPs) and their tissue inhibitors (TIMPs) in chronic doxorubicin cardiomyopathy in a rabbit model and to evaluate the effects of bone marrow-derived mesenchymal stem cell (MSC) transplantation in this disease. Thirty-nine 3-month-old New Zealand rabbits were divided into 4 groups: group 1 (n = 9) was the untreated control. Groups 2–4 were treated with 6 weeks of doxorubicin (3 mg/kg). Group 2 (n = 6) received no further treatment. In group 3 (n = 9), animals were treated with culture medium (CM) alone. In group 4 (n = 15), autologous MSCs (1.5–2.0 X 10⁶/ml) were injected in the left ventricular (LV) wall. Hearts were stained with HE and picrosirius red. MMP-1, −2, −3 and −9 and TIMP-2 and −3 were detected immunohistochemically. The mRNA levels were determined by real-time polymerase chain reaction. The results confirmed that doxorubicin treatment resulted in minimal myocardial fibrosis and showed that expression of MMPs increased and TIMP-3 decreased. The injection procedure resulted in increased myocardial fibrosis in groups 3 and 4. After MSC injection, MMP-1, MMP-2, and TIMP-3 expression was higher than that in group 2. CM injection led to more fibrosis, elevated TIMP-3, but diminished MMP-1 and MMP-2 expression compared with MSC injection. The mRNA levels of MMPs and TIMPs were not significantly different among all groups. In conclusion, chronic doxorubicin cardiomyopathy was characterized by increased MMP and decreased TIMP-3 expression. MSCs injection into the LV resulted in marked differences of collagen content and MMP/TIMP expression in the whole heart, although significant numbers of living MSCs were not detected after 4 weeks.

Keywords

Cardiomyopathy doxorubicin extracellular matrix heart immunohistochemistry matrix metalloproteinase rabbit stem cell

Stem cell therapy is gaining importance as a potential new therapy for patients with advanced heart failure.^4, ^14, ²⁰ Induction of differentiation of bone marrow–derived mesenchymal stem cells (MSCs) in vitro into cells with beating myotubules by treatment with 5-azacytidine was first reported in 1995 by Wakitani et al.³⁵ Even without pretreatment, MSCs have recently been shown to home to the site of myocardial infarction,^3, ⁴ differentiate into cardiomyocyte-like cells,^31, ³² and integrate functionally and morphologically in the myocardium.^4, ³⁹ Since then, the body of evidence has been growing that transdifferentiation plays a minor role, if any, in the therapeutic effects of stem cell transplantation. Using echocardiography, Scorsin et al.²⁸ observed improvement of fractional shortening 1 month after transplantation of fetal cardiomyocytes into doxorubicin-treated hearts in mice. Interestingly, this improvement of functional parameters was observed, although no evidence of cell survival over this course was found. It seems that as yet unidentified paracrine mechanisms may be involved in this therapeutic process.²⁷

In dilative cardiomyopathy, not only disturbances in the contractile apparatus of the cardiomyocytes but also alterations in interstitial architecture of the myocardium play a role in the pathogenesis of heart failure.^6, ^10, ^11, ^12, ¹⁵ Extracellular matrix (ECM) integrity is maintained by a balance between the activity of matrix metalloproteinases (MMPs), a family of enzymes that degrade ECM proteins,^18, ¹⁹ and their tissue inhibitors (TIMPs).^6, ^12, ¹⁵ Cardiomyocytes, fibrocytes, and endothelial cells express 1 or more types of MMP/TIMP.^12, ³³ The activity of MMPs and TIMPs varies with the type, etiology, and stage of the cardiac disease;^10, ^24, ^30, ³¹ but in human idiopathic dilative cardiomyopathy in general, MMP levels are high and TIMP levels are low.³⁸ Doxorubicin is commonly used in anticancer therapy in humans and animals, but dilative cardiomyopathy is an undesired side effect. To our knowledge, MMPs and TIMPs have not been studied in doxorubicin cardiomyopathy in any species.

We hypothesized that alterations of ECM metabolism play a role in doxorubicin-induced cardiomyopathy and may be influenced by stem cell injection. The validity of the doxorubicin cardiomyopathy model has been established by previous studies in rabbits,² mice,²⁸ rats,²⁶ and dogs.²⁵ Thus, the aim of this study was first to investigate the expression of MMPs and TIMPs in chronic doxorubicin cardiomyopathy in the rabbit model and, second, to evaluate the effects of MSC transplantation in this disease.

Material and Methods

All experiments were conducted in accordance with guidelines published in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health (USA) and were under protocols approved by the Institutional Animal Care and Use Committee at Leipzig University and Regierungspraesidium (Germany).

Animals

Three-month-old male healthy New Zealand white rabbits (Charles River, Germany), housed individually under conventional caging conditions, were used in this study. After a 2-week acclimatization period, animals were randomly assigned either to the healthy untreated control (group 1, n = 9) or to the doxorubicin treatment group (n = 30). Doxorubicin hydrochloride (Ribodoxo-L 50) (3 mg/kg of body weight) was administered once a week for 6 weeks intravenously followed by an interval of 14 days without treatment. These treated animals were then randomly subdivided into 3 groups: 1 with no further therapy (group 2, n = 6), a group with intramyocardial injection of culture medium (CM) (group 3, n = 9), and a group receiving autologous MSCs (group 4, n = 15). The time course of the experiment is summarized in Table 1.

Table 1.

Experimental design for examination of the effects of mesenchymal bone marrow–derived stem cells in doxorubicin-induced cardiomyopathy.

	Days∗
Group	0–42	43–56	57–84
1	—	—	—
2	Doxorubicin treatment	—	—
3	Doxorubicin treatment	—	Day 57: culture medium injection
4	Doxorubicin treatment	Day 50: bone marrow aspiration, Day 50–56: isolation, expansion, and labeling of stem cells	Day 57: stem cell injection

∗Echocardiograms were obtained on days 0, 42, 57, and 84.

— = no treatment or application

Isolation, Expansion, and Labeling of MSCs

Harvest and expansion of mesenchymal bone marrow stem cells were achieved in accordance with previously described methods.²² To avoid immune rejection, we gave the animals of group 4 autologous femoral bone marrow cells (BMCs).

On day 50 of the experiment, a femoral bone marrow aspirate was performed aseptically in the animals of group 4. Marrow cells were flushed out with phosphate buffered saline, pH 7.4, and ethylenediaminetetraacetic acid (2 mM) (PBS-EDTA) to obtain a maximum amount of cells. The BMCs were fractionated over 1.073-g/ml Ficoll solution (Sigma, Steinheim, Germany), and mononuclear cells were obtained according to Tomita et al.³² After 2 washings, BMCs were resuspended in 12-ml Dulbecco's Modified Eagle Medium (DMEM, Sigma) with 10% fetal bovine serum, 100-U/ml penicillin G, and 100-μg/ml streptomycin (Sigma) added. To induce differentiation, we cultured cells with 10-μM 5-azacytidine (Sigma) for 24 hours; we then washed them, seeded them on gelatin-coated Petri dishes, and cultured them at 37°C in a humidified atmosphere with 5% CO₂. The MSCs were isolated on the basis of their ability to adhere to the culture plate. Hematopoietic and other nonadherent cells were removed twice by a medium change on days 52 and 54. Prior to cell transplantation, adherent cells were detached by trypsin-EDTA, passaged twice, and labeled with 30-μl/ml Vybrant cell-labeling solution (1,1′-dioctadecyl-3,3,3′,3′,-tetramethylindocarbocyanine perchlorate [DIL]; Invitrogen, Paisley, UK) for 30 minutes under light protection at 37°C, followed by 2 washes with PBS. DIL is a red lipophilic fluorochromatic dye which accumulates in the membrane of rough endoplasmic reticulum.⁸ Cells were further passaged twice, harvested, and used for cell transplantation.

CM Injection and Stem Cell Transplantation

On day 57, thoracotomy was performed in animals of groups 3 and 4, under general anesthesia through the left intercostal space to expose the heart. The free lateral wall of the left ventricle (LV) was identified, and either 1-ml pure CM (group 3) or 1-ml CM containing 1.5–2.0 × 10⁶ stem cells/ml was injected transepicardially into the myocardium in a circular manner at 4 locations (diameter of the injection area, ∼1.5 cm) within 2 minutes. A thoracic drain was inserted before closing the chest, and the rabbits were treated with antibiotics and analgesics.

Clinical Observations

Clinical observations together with cardiac auscultation were performed regularly.

Echocardiographic Studies

Echocardiograms were obtained on days 0, 42, 57, and 84. Results were evaluated by statistical methods.

Gross Examination

On day 84, animals in all groups were euthanized by heart explantation under general anesthesia. The hearts were inspected grossly, and transverse sections were made through the injection area. A representative section was fixed in 10% buffered formol saline. One section was snap-frozen in liquid nitrogen and stored at −80°C for mRNA analyses.

Histologic Analysis

Formalin-fixed specimens were embedded routinely in paraffin wax, sectioned at 3 μm, and stained with HE. Picrosirius red–stained slides⁹ were used for fibrillar collagen measurements. For representative measurements, heart cross-sections were divided into equal areas (0.12 cm²) (LV wall, n = 5; right ventricular [RV] wall, n = 5; interventricular septum [IVS], n = 5) and digitized. The mean value of collagen deposition was generated. The area (percentage) of extracellular red-stained collagen fibers was computed from the fields, providing a composite measure of interstitial, perivascular, and replacement fibrosis. Additionally, investigations of picrosirius red–stained slides were viewed with polarized light to demonstrate the collagen types I (yellow-red) and III (green).³⁷ Deparaffined unstained slides were examined by fluorescence microscopy to identify the DIL-labeled stem cells.

Immunohistochemistry

Using the peroxidase-antiperoxidase (PAP) technique, serial sections were immunolabeled with the following primary mouse monoclonal antibodies: MMP-1 (Chemicon, MAB 3307, Hofheim, Germany), MMP-2 (Chemicon, MAB 3308, Hofheim, Germany), MMP-3 (Chemicon, MAB 3312, Hofheim, Germany), MMP-9 (Chemicon, MAB 3309, Hofheim, Germany), TIMP-2 (Lab Vision, MS1485, Fremont, California), and TIMP-3 (Chemicon, MAB 3318, Hofheim, Germany). The secondary antibody was rat anti-mouse immunoglobulin G (Dianova, Hamberg, Germany), and mouse PAP complex (Dako, Hamberg, Germany) was used. Diaminobenzidinehydrochloride (Sigma-Aldrich Chemie, München, Germany) was used as chromogen, and slides were counterstained with Papanicolaou's stain (Merck, Darmstedt, Germany). Negative controls were obtained by substituting irrelevant mouse antibody as the primary antibody.

All slides were read blind. The staining intensity of the cardiomyocytes was evaluated semiquantitatively and graded as follows: (±) minimal, (+) mild, (++) moderate, and (+++) marked. Fibrocytes with a positive or negative signal were counted digitally (Imaging System 3.2, Olympus DP-Soft, Olympus, Hamburg, Germany). For a representative measurement, 5 high-power fields (each 0.1 mm²) excluding the injection zone were randomly chosen from the LV, RV, and IVS. The mean number of cells counted in these 5 fields was generated.

Assay of mRNA Profiles

To assess the mRNA expression of MMP-1, -2, -3, -9 and TIMP-2, -3, we performed real-time polymerase chain reaction (PCR). Frozen tissue samples of the RV, LV, and IVS from 6 animals in each group were prepared with an Ultra-Turrax (IKA Werke, Detmold, Germany) using Trizol Reagent (Invitrogen, Karlsruhe, Germany). An aliquot of total RNA (500 ng) was reversely transcribed into cDNA using a Sensiscript Kit (Qiagen, Hilden, Germany) and random primers (Invitrogen) according to the instructions of the manufacturers. To quantify mRNA expression levels, we used 1-μl aliquot of the cDNA in real-time PCR containing the gene-specific primers (Table 2) and LightCycler FastStart DNA Master SYBR-Green I reaction mix (Roche Diagnostics, Mannheim, Germany). The specific expression levels were analyzed using LightCycler software and expressed in relation to the individual 18S rRNA intensity in relative units. Each PCR was performed in triplicate with a variability <10%. PCR conditions for each of the MMP and TIMP mRNA were as follows: annealing temperature 60°C and elongation time 30–40 seconds at 21–40 PCR cycles.

Table 2.

Sequences of primers for MMPs and TIMPs in real time PCR.∗

	Sense Primers	Antisense Primers
MMP-1	TAGCTGGTTCAACTGCAGGA	GAACTTGGCCTTCCTTAGCC
MMP-2	AATCCACGATGGAGAAGCTG	GCATTGCCATACTTCACACG
MMP-3	GTTCCGCCTTTCTCAAGATG	CTTGCGCCAGAAATACCTGT
MMP-9	ACACACACGACGTCTTCCAG	AAGAAAGGCGAGGAGAGAGG
TIMP-2	CCCTTGACAAAGAGGACAGA	CTCTTGATGGGGTTGCCATA
TIMP-3	TTCTGCAACTCCGACATCGT	CATCTTGCCTTCATACACGC

∗MMP = metalloproteinase; TIMP = metalloproteinases tissue inhibitor; PCR = polymerase chain reaction.

Statistical Analysis

Statistical analyses included the Shapiro-Wilk test and 1-way analysis of variance. For post hoc analyses, the Levene and Bonferroni tests were applied. A paired t-test was performed as appropriate. Results were considered statistically significant if p ≤ .05. All data analyses were performed by using SPSS software, Version 11.5 (SSPS Inc. Headquarters, Chicago, IL, USA).

Results

Clinical Observations

Doxorubicin administration caused clinical signs of heart failure (cyanosis, dyspnea) from treatment day 20 on, which was confirmed by echocardiography on day 42. Serial echocardiography detected a decline in the ejection fraction and fractional shortening in doxorubicin-failing hearts (groups 2–4) compared with those in the untreated controls (group 1). MSC injection (group 4), but not CM injection (group 3), resulted in a significantly better heart function as reported previously.⁷

Gross Findings

Macroscopically, moderate LV dilation was seen in all doxorubicin-treated animals. The injection sites of CM or MSCs were characterized by focal mild-to-moderate epicarditis (Fig. 1).

Fig. 1.

Heart; rabbit No. 28, group 4. Macroscopically, focal epicarditis (asterisk) on the left ventricle is present in the area of transepicardial stem cell injection.

Histopathologic Findings

The hearts of the untreated controls (group 1) were normal: The epicardium was composed of only a few type I collagen fibers and several adipocytes around coronary vessels. Cardiomyocytes were arranged in the typical pattern with few intermingled type I collagen fibers and single type III collagen fibers. Fibrocytes were characterized by slender hyperchromatic nuclei and small amounts of cytoplasm. Interstitial and perivascular collagen fibers (Fig. 2) constituted ∼7% of the myocardial area (Table 3), and fibrocytes numbered 65 ± 10 per high-power field. The endocardium consisted of endothelial cells with few underlying type I collagen fibers.

Table 3.

Area of myocardial type I and III collagen fibers(mean ± SD) in the different groups and heart locations.

Group∗	Left Ventricle	Right Ventricle	Interventricular Septum	P Value (heart locations)
1	6.8 ± 1.8	8 ± 1.0	6.1 ± 1.9
2	10.2 ± 5.0	7.9 ± 3.5	7 ± 3.7
3	14.5 ± 4.7†	15 ± 3.0†	12.2 ± 4.5†
4	12.4 ± 4.0†	11.1 ± 3.0‡	9.1 ± 1.9	LV–IVS; P = .002
P	G1–G3; .008	G1–G3; .001	G1–G3; .005
	G1–G4; .05	G2–G3; .001	G2–G3; .022
		G4–G3; .019

∗Group 1 (G1) = untreated control; group 2 (G2) = doxorubicin treatment; group 3 (G3) = doxorubicin treatment and culture medium injection; group 4 (G4) = doxorubicin treatment and stem cells injection. Only values of significant differences are presented in the table.

†Significant vs. G1.

‡Significant G4 vs. G3.

Table 4.

Intensity of expression of MMPs and TIMPs in the cardiomyocytes(exclusive to the injection site) in the different groups detected by immunohistochemistry.∗

	Group 1†	Group 2	Group 3	Group 4
MMP-1‡	+	++	++	+++
MMP-2	(±)	++	++	+++
MMP-3	−	−	−	−
MMP-9	+	+++	+++	+++
TIMP-2	+	+	+	+
TIMP-3	++	++	+++	++

∗Staining intensity in cardiomyocytes: – negative; (±) = minimal; + = mild; ++ = moderate; +++ = marked.

†Group 1 = untreated control; group 2 = doxorubicin treatment; group 3 = doxorubicin treatment and culture medium injection; group 4 = doxorubicin treatment and stem cells injection.

‡MMP = metalloproteinase; TIMP = metalloproteinase tissue inhibitor.

Fig. 2.

Heart; rabbit No. 1, group 1. In the untreated controls, intramural coronary vessels are surrounded by normal small amounts of collagen fibers. Picrosirius red stain.

In the hearts of all doxorubicin-treated animals (groups 2–4), single lymphocytes and macrophages were distributed in the interstitium. Mild angiosclerosis was apparent in some coronary vessels. In group 2, type I and III collagen fibers were not significantly (P = .852) increased compared with those in the untreated controls (group 1). In group 3, collagen fibers were mainly perivascular (Fig. 3), significantly increased compared with those in group 1. In group 4, the collagen content was significantly increased in the LV only. In the RV, the fibrotic area in group 3 was significantly greater than that in group 4 (Table 3). The number of cardiac fibrocytes (70 ± 5) was not significantly different among the groups.

Fig. 3.

Heart; rabbit No. 17, group 3. Injection of culture medium induces marked perivascular fibrosis spreading into the surrounding myocardium. Picrosirius red stain.

At injection sites, focal mild-to-moderate chronic fibroblastic inflammation was noted in the epicardium (group 3: mild, n = 3; moderate, n = 6; group 4: mild, n = 9; moderate, n = 6). Well-demarcated fibrotic foci extending from the epicardium into the myocardium were interpreted as injection canals (Fig. 4). These were mainly composed of type III collagen fibers, with ∼10% type I collagen fibers, large spindle-shaped fibroblasts, a few macrophages and lymphocytes, and single neutrophilic granulocytes.

Fig. 4.

Heart; rabbit No. 21, group 3. The injection site is characterized by focal fibroblastic epicarditis and a focal myocardial scar extending from the epicardium into the myocardium. Picrosirius red stain.

Stem cells could not be identified in HE-stained slides. Using fluorescence microscopy, we saw single large cells (1–2 cells/high-power field), with oval nuclei showing a specific intracytoplasmic signal deriving from DIL loading in the sites of MSC injection (group 4); however, from their morphology, it was unclear whether they were original MSCs or whether they were macrophages with phagocytosed remnants of MSCs. In the sites of CM injection (group 3), no cells with a specific intracytoplasmic signal were identified.

Immunohistochemical Findings

The group-specific findings of MMP and TIMP expression in cardiomyocytes and MMP-2 and TIMP-3 expression in fibrocytes are presented in Tables 4– 7. In all groups, the staining intensity of MMPs and TIMPs in cardiomyocytes was slightly higher in the RV than in the LV or IVS. Only TIMP-3 expression in fibrocytes was significantly different among the RV, LV, and IVS in most groups (Table 6). MMP-1 and MMP-9 expression was present in cardiomyocytes, but not in fibrocytes, endothelia, and vascular smooth muscle cells. MMP-2, TIMP-2, and TIMP-3 were expressed in cardiomyocytes as well as in fibrocytes, endothelia, and vascular smooth muscle cells. Macrophages and neutrophilic granulocytes were positive for MMP-9 and TIMP-3, and lymphocytes expressed TIMP-3.

Table 5.

Metalloproteinase-2 expressing myocardial fibrocytes (%) exclusive to the injection sites in the different groups and heart locations (mean ± SD).

Group∗	Left Ventricle	Right Ventricle	Interventricular Septum
1	16.4 ± 7.5	17.3 ± 7.3	11.6 ± 4.3
2	35.0 ± 16.7	41.2 ± 14.1†	40.5 ± 19.4†
3	17.0 ± 10.1	21.8 ± 7.6‡	16.0 ± 12.7‡
4	26.6 ± 11.9	33.1 ± 5.4	29.6 ± 13.2
P	No significant differences	G1–G2; .012	G1–G2; .008
		G2–G3; .002	G2–G3; .022

∗Group 1 (G1) = untreated control; group 2 (G2) = doxorubicin treatment; group 3 (G3) = doxorubicin treatment and culture medium injection; group 4 = doxorubicin treatment and stem cells injection. Only values of significant differences are presented in the table.

†Significant vs. G1.

‡Significant G2 vs. G3.

Table 6.

Metalloproteinase tissue inhibitor-3 expressing myocardial fibrocytes (%) exclusive to the injection site in the different groups and heart locations (mean ± SD).

Group∗	LV	RV	IVS	P Value (heart locations)
1	19.2 ± 7.2‡	27.5 ± 10.9	17.3 ± 9.6†	RV–LV, P = .035
				RV–IVS, P = .021
2	7.5 ± 3.5	20.5 ± 5.1	7.0 ± 4.9	RV–LV, P = .033
				RV–VS, P = .011
3	19.4 ± 11.0‡	27.4 ± 13.9	16.7 ± 8.2‡	No significant differences
4	16.4 ± 11.3	25.2 ± 9.1	17.4 ± 9.8‡	RV–LV, P = .015
				RV–IVS, P = .018
P	G1–G2; .005	No significant differences	G1–G2; .04
	G2–G3; .03		G2–G3; .03
			G2–G4; .04

†LV = left ventricle; RV = right ventricle; IVS = interventricular septum.

‡Significant vs. G2.

Table 7.

Immunohistochemical detection of MMP and TIMP expression in the epicardial and myocardial fibrocytes at the injection sites of culture medium(group 3) and stem cells (group 4).

	Group 3		Group 4
	Epicardium Injection Sites	Myocardium Injection Canal	Epicardium Injection Sites	Myocardium Injection Canal
MMP-1∗	−	++	−	+
MMP-2	−	+	−	(+)
MMP-3	−	−	−	−
MMP-9	−	−	−	−
TIMP-2	+++	+++	+++	+++
TIMP-3	+++	+++	+++	+++

∗MMP = metalloproteinases; TIMP = metalloproteinases tissue inhibitors. (+) = <10% of fibrocytes are positive; + = <20% of fibrocytes are positive; ++ = <50% of fibrocytes are positive; +++ = >50% of fibrocytes are positive; − = all fibrocytes are negative.

Group 1: In the untreated controls, MMP-1 (Fig. 5) and MMP-9 (Fig. 6) expression in cardiomyocytes was mild and diffuse (Table 3). MMP-2 was minimally expressed in perivascular cardiomyocytes and was also mildly expressed in ∼15% of the fibrocytes (Table 5). TIMP-2 and TIMP-3 (Fig. 7) were mildly and moderately, respectively, expressed in all cardiomyocytes and moderately expressed in ∼20% of fibrocytes (Table 6).

Fig. 5.

Heart; rabbit No. 4, group 1. In untreated controls, expression intensity of metalloproteinase-1 (MMP-1) in cardiomyocytes is minimal and absent in fibrocytes and vessels. Immunohistochemistry, MMP-1, Papanicolaou's counterstain.

Fig. 6.

Heart; rabbit No. 4, group 1. In untreated controls, metalloproteinase-9 (MMP-9) expression in cardiomyocytes is mild and absent in fibrocytes and vessels. Immunohistochemistry, MMP-9, Papanicolaou's counterstain.

Fig. 7.

Heart; rabbit No. 6, group 1. In untreated controls, expression of metalloproteinase tissue inhibitors-3 (TIMP-3) in cardiomyocytes is mild but intense in numerous fibrocytes (arrows) and in endothelia. Immunohistochemistry, TIMP-3, Papanicolaou's counterstain.

Group 2: Doxorubicin treatment resulted in mildly increased expression of MMP-1 and MMP-2 in perivascular cardiomyocytes. In contrast to the untreated controls (group 1), marked MMP-9 expression was seen diffusely in cardiomyocytes. MMP-3 expression was limited to a few small foci of the myocardium. When compared with all groups, MMP-2-positive fibrocytes (∼40%) were increased, and TIMP-3-expressing fibrocytes (∼10%) were decreased.

Group 3: Injection of CM induced moderate MMP-1 and MMP-2 expression intensity in perivascular cardiomyocytes, while MMP-9 was marked in all cardiomyocytes. Compared with group 2, TIMP-3 staining in cardiomyocytes was increased and ∼20% of fibrocytes expressed TIMP-3 and MMP-2.

Group 4: MSC injection resulted in marked MMP-1 (Fig. 8) and MMP-2 expression in perivascular cardiomyocytes, and ∼30% of fibrocytes expressed MMP-2. MMP-9 expression was also marked and was noted in all cardiomyocytes (Fig. 9). Moderate expression of TIMP-2 and TIMP-3 was also seen in all cardiomyocytes and in ∼20% of fibrocytes.

Fig. 8.

Heart; rabbit No. 29, group 4. After stem cell injection, the expression intensity of metalloproteinase-1 (MMP-1) is markedly increased in perivascular cardiomyocytes. Immunohistochemistry, MMP-1, Papanicolaou's counterstain.

Fig. 9.

Heart; rabbit No. 29, group 4. After stem cell injection, the expression of metalloproteinase-9 (MMP-9) in all cardiomyocytes is moderately increased, but fibrocytes and vessels are negative. Immunohistochemistry, MMP-9, Papanicolaou's counterstain.

Injection sites (groups 3 and 4): In general, in cardiomyocytes bordering the injection site, there was an increased expression of MMP-1 and MMP-2 compared with the remaining myocardium. A greater number of MMP-1- and MMP-2-positive fibrocytes was seen in CM injection sites than in MSC injection sites (Table 7). Furthermore, all fibrocytes within the injection sites expressed TIMP-2 and TIMP-3 (Fig. 10) intensively.

Fig. 10.

Heart; rabbit No. 18, group 3. In the culture medium injection site, the fibroblasts express metalloproteinase tissue inhibitor-3 (TIMP-3) intensively, and in the periphery of the scar, the expression intensity in cardiomyocytes is increased. Immunohistochemistry, TIMP-3, Papanicolaou's counterstain.

On the basis of the semiquantitative and statistical results, it can be summarized that doxorubicin treatment results in minimally increased myocardial fibrosis but marked increase of MMP-1, -2 and -9 and decreased TMP-3 expression.

Injection of either CM or MSCs increased myocardial fibrosis. CM injection resulted in increased TIMP-3 expression and a higher degree of fibrosis. MMP expression was similar compared with that in group 2, whereas MSC injection led to a higher MMP-1 and MMP-2 and normal TIMP-3 expression.

mRNA Assays

The mRNA values for MMP-3 were very low compared with the values of other MMPs and TIMPs. All mRNA levels of MMPs (Figs. 11, 12) and TIMPs (data not shown) were not significantly different among the 4 groups or among the 3 locations of the heart (RV, LV, and IVS) (data not shown). No correlation between the mRNA levels and the immunohistochemical findings, which showed marked group-specific differences, could be detected.

Fig. 11.

Transcription levels of metalloproteinases (MMPs) in the left ventricle (LV), group 1 (white) compared with group 2 (black). MMP-3 levels are markedly lower than other MMP levels; no significant differences are seen in the transcription levels of all MMPs between the untreated controls (group 1) and the doxorubicin-treated animals (group 2).

Fig. 12.

Transcription levels of metalloproteinases (MMPs) in the left ventricle (LV), group 3 (white) compared with group 4 (black). MMP-3 levels are markedly lower than other MMP levels; no significant differences are seen in all MMPs among the rabbits with culture medium injection (group 3) and stem cell injection (group 4).

Discussion

In this study, we investigated first the expression of MMPs and TIMPs in chronic doxorubicin cardiomyopathy in a rabbit model and, second, the effects of stem cell injection in this disease.

In accordance with the literature,^33, ³⁶ in normal hearts (group 1), the expression intensity of MMPs and TIMPs was very mild. Cardiomyocytes were positive for MMP-1, -2, -9 and TIMP-2 and -3, whereas in fibrocytes, MMP-2 and TIMP-2 and -3 expression was detected. Endothelia expressed MMP-2 and TIMP-2 and -3.

Successful induction of heart failure in the doxorubicin-treated rabbits was confirmed by echocardiographic data⁷ and ß-receptor analysis.⁵ Corresponding to the literature,^13, ³⁴ collagen fibers were minimally increased mainly in perivascular areas in doxorubicin cardiomyopathy (group 2). In vitro studies in rat heart–derived embryonic myocytes (H9c2) showed that doxorubicin-induced a rapid increase of MMP-2 and MMP-9 expression and activation.²⁹ This was confirmed in vivo by our rabbit model. MMP-1, MMP-2, and MMP-9 were increased in cardiomyocytes, and the number of TIMP-3-positive fibrocytes decreased.

Myocardial fibrosis was increased after CM injection (group 3) and MSC injection (group 4). Interestingly, the fibrosis of the RV was significantly higher in group 3 than in group 4, although injection was into the LV. The relationship of this diffuse fibrosis to the surgical intervention and local injection lesion is not readily explained. MSC injection, however, resulted in less myocardial fibrosis than injection of CM alone and was accompanied by higher perivascular MMP-1 and MMP-2 and lower fibrocyte TIMP-3 expression.

Significant differences among the groups were detected only by immunohistochemistry and not by PCR, possibly due to the fact that the investigations were carried out in the chronic phase of the disease. Post-transcriptional mechanisms controlling the protein translation rate, the half-life of specific proteins or mRNAs, and the intracellular location and molecular association of the protein products of expressed genes may be insufficiently sensitive to predict protein expression levels from quantitative mRNA data. Similar difficulties, reported by Peterson et al.,²¹ were that mRNA levels and total amounts of protein were not comparable in all cases. In our study, the clinical findings correlated with the morphologic and immunohistochemical differences among the groups but not with the PCR results.

We conclude that there are marked differences of extracellular matrix metabolism between medium and stem cell injection groups at the local injection site as well as in the whole heart. In MSC injection sites, local MMP-1 and -2 expression of fibrocytes was less intensive than in CM injection sites, while TIMP-2 and -3 expression in fibrocytes was intensive in the injection sites in both groups; however, the repair process in CM injection sites seemed more active than that in stem cell injection sites. To our knowledge, no comparable studies have been reported to date. Further investigations using in situ zymography to identify the cells that are involved in the remodeling processes are indicated.

Only a few cells were identified as integrated original stem cells or more likely phagocytosed debris of MSC so that it is unlikely that MSCs had a direct effect on the heart overall. While Agbulut et al.¹ identified single stem cells from 2 days up to 2 weeks after injection of unpurified BMCs in acute doxorubicin cardiomyopathy in transgenic mice, Scorsin et al.²⁸ found no evidence of transplanted cell survival over the follow-up period, although they observed an improvement of fractional shortening by echocardiography 1 month after transplantation of fetal cardiomyocytes into doxorubicin-treated hearts. As discussed in other studies, paracrine mechanisms appear to be more likely than direct effects of stem cells.^20, ²³ Dhein et al.⁵ demonstrated that down-regulation of ß-adrenoceptors was attenuated, and noradrenaline and adrenaline plasma levels were normal in the stem cell–treated rabbits of our study. To our knowledge, the effects of stem cell injection on the extracellular matrix metabolism have not been investigated in detail in any other study. MMPs are regulated at multiple levels, including transcription, translation, secretion, and activation.¹⁷ Plasmin, TGFß1, TNFa, and IL-1ß are some examples for the mediators that are involved in the MMP regulation;¹⁶ however, it remains unclear which are secreted by the stem cells and how the beneficial effects are induced.

In conclusion, doxorubicin cardiomyopathy is characterized by minimally increased myocardial fibrosis but a marked increase of MMP-1, -2, and -9 and decreased TMP-3 expression. MSC injection improved clinical heart function⁷ and ß-receptor expression⁵ but induced myocardial fibrosis and increased MMP expression. In contrast, injection of CM resulted not in better heart function⁷ but in a higher degree of myocardial fibrosis and TIMP-3 expression than MSC injection did. A relevant number of stem cells could not be identified after 4 weeks; thus, paracrine mechanisms of stem cells in doxorubicin cardiomyopathy have to be considered.

Footnotes

Acknowledgements

This work was supported by the German Heart Foundation “Deutsche Stiftung für Herzforschung” grant F/08/03 held by Stefan Dhein. We thank K. Schneider and C. Ullmann (Experimental Laboratory Department of Cardiac Surgery, Heart Center, Leipzig, Germany), A. Richter (statistics), and J. Kacza (Morphometry, Veterinary Anatomy).

References

1 Agbulut

Menot

M-L

Marotte

Paulin

Hagege

Chomienne

Samuel

J-L

Menasche

: Temporal patterns of bone marrow cell differentiation following transplantation in doxorubicin-induced cardiomyopathy. Cardiovasc Res 58:451–459, 2003

2 Arnolda

McGrath

Cocks

Sumitran

Johnston

: Adriamycin: cardiomyopathy in the rabbit—an animal model of low output cardiac failure with activation of vasoconstrictor mechanisms. Cardiovasc Res 19:378–382, 1985

3 Barbach

Chouraqui

Baron

Feinberg

Etzion

Tesson

Miller

Guetta

Zipori

Kedes

Kloner

Leor

: Systemic delivery of bone marrow-derived mesenchymal stem cells to the infarcted myocardium. Circulation 108:863–868, 2003

4 Chiu

: Bone-marrow stem cells as a source for cell therapy. Heart Fail Rev 8:247–251, 2003

5 Dhein

Garbade

Rouabah

Abraham

Ungemach

F-R

Schneider

Ullmann

Aupperle

Gummert

J-F

Mohr

F-W

: Effects of autologous bone marrow stem cell transplantation on beta-adrenoceptor density in a rabbit model of non-ischemic heart failure. J Cardiothorac Surg 1: 17, 2006

6 Fedak

PWM

Altamentova

Weisel

Nili

Ohno

Verma

Lee

Kiani

Mickle

DAG

Strauss

R-K

: Matrix remodeling in experimental and human heart failure: a possible regulatory role for TIMP-3. Am J Physiol Heart Circ Physiol 284: H626–634, 2003

7 Garbade

Schubert

Lipinski

Aupperle

Walther

Gummert

Dhein

Mohr

: Stem cell transplantation enhances functional and myocardial remodelling of cardiomyocytes in chronic non-ischaemic heart disease. Europ Heart J 26: abstr suppl: 530, 2005 [Abstract]

8 Godement

Vanselow

Thanos

Bonhoeffer

: A study in developing visual systems with a new method of staining neurones and their processes in fixed tissue. Development 101:697–713, 1987

9 Grüninger

Schoon

Menger

Klug

: Incidence and morphology of endometrial angiopathies in mares with relationship to age and parity. J Comp Pathol 119:293–309, 1998

10.

10 Herpel

Singer

Flechtenmacher

Pritsch

Sack

F-U

Hagl

Katus

Haass

Otto

Schnabel

: Extracellular matrix proteins and matrix metalloproteinases differ between right and left ventricular sites in end-stage cardiomyopathies. Virchows Arch 446:369–378, 2005

11.

11 Janicki

Brower

Gardner

Chancey

Stewart

: The dynamic interaction between matrix metalloproteinase activity and adverse myocardial remodelling. Heart Fail Rev 9:33–42, 2004

12.

12 Kassiri

Khokha

: Myocardial extra-cellular matrix and its regulation by metalloproteinases and their inhibitors. Thromb Haemost 93:212–219, 2005

13.

13 Klimtova

Simunek

Mazurova

Hrdina

Gersl

Adamcova

: Comparative study of chronic toxic effects of daunorubicin and doxorubicin in rabbits. Hum Exp Toxicol 21:649–657, 2002

14.

14 Leor

Battler

: Editorial: regeneration hope to grow a new heart muscle. Heart Fail Rev 8:197–199, 2003

15.

15 Li

Simon

Bocan

TMA

Peterson

: MMP/TIMP expression in spontaneously hypertensive heart failure rats: the effect of ACE- and MMP-inhibition. Cardiovasc Res 46:298–306, 2000

16.

16 Li

McTiernan

Feldman

: Interplay of matrix metalloproteinases, tissue inhibitors of metal-loproteinases and their regulators in cardiac matrix remodelling. Cardiovasc Res 46:214–224, 2000

17.

17 Lindsey

Gannon

Aikawa

Schoen

Rabkin

Lopresti-Morrow

Crawford

Black

Libby

Mitchell

Lee

: Selective matrix metalloproteinase inhibition reduces left ventricular remodelling but does not inhibit angiogenesis after myocardial infarction. Circulation 105:753–758, 2002

18.

18 Nagase

Visse

Murphy

: Structure and function of matrix metalloproteinases and their TIMPs. Cardiovasc Res 69:562–573, 2006

19.

19 Newby

: Matrix metalloproteinases regulate migration, proliferation, and death of vascular smooth muscle cells by degrading matrix and non-matrix substrates. Cardiovasc Res 69:614–624, 2006

20.

20 Perin

Dohmann

HFR

Borojevic

Silva

Sousa

ALS

Mesquita

Rossi

MID

Carvallho

Dutra

Dohmann

HJF

Silva

Belem

Vivacqua

Rangel

FOD

Esporcatte

Geng

Vaughn

Assad

JAR

Mesquita

Willerson

: Transendocardial, autologous bone marrow cell transplantation for severe, chronic ischemic heart failure. Circulation 107:2294–2302, 2003

21.

21 Peterson

Dillon

Braynt

: Evolution of matrix metalloproteinase and tissue inhibitor expression during heart failure progression in the infarcts rat. Cardiavasc Res 46:307–315, 2000

22.

22 Pittenger

Mackay

Beck

Jaiswal

Douglas

Mosca

Moorman

Simonetti

Craig

Marshak

: Multilineage potential of adult human mesenchymal stem cells. Science 284:143–147, 1999

23.

23 Scharz

Kornowski

: Autologous stem cells for functional myocardial repair. Heart Fail Rev 8:237–245, 2003

24.

24 Schulze

Wang

Suarez-Pinzon

Sawicka

Sawicki

Schulz

: Imbalance between tissue inhibitor of metalloproteinases during acute myocardial ischemia-reperfusion injury. Circulation 107:2487–2492, 2003

25.

25 Schupp

Huck

Sykora

Flechtenmacher

Gorenflo

Koch

Sack

F-U

Haass

Katus

Ulmer

Hagl

Otto

Schnabel

: Right ventricular expression of extracellular matrix proteins, matrix metalloproteinases, and their inhibitors over a period of 3 years after heart transplantation. Virchows Arch 448:184–194, 2006

26.

26 Schwarz

Pollick

Dow

Patterson

Birnbaum

Kloner

: A small animal model of non-ischemic cardiomyopathy and its evaluation by transthoracic echocardiography. Cardiovasc Res 39:216–223, 1998

27.

27 Schwartz

Kornowski

: Autologous stem cells for functional myocardial repair. Heart Fail Rev 8:237–245, 2003

28.

28 Scorsin

Hagege

Dolizy

Marotte

Mirochnik

Copin

Barnoux

le Bert

Samuel

Rappaport

Menasche

: Can cellular transplantation improve function in doxorubicin-induced heart failure? Circulation 98(suppl 19):II151–156, 1998

29.

29 Spallarossa

Altieri

Garibaldi

Ghigliotti

Barisione

Manca

Fabbi

Ballestero

Brunelli

Barsotti

: Matrix metalloproteinase-2 and −9 are induced differently by doxorubicin in H0c2 cells: the role of MAP kinases and NAD(P)H oxidase. Cardiovasc Res 69:736–745, 2006

30.

30 Spinale

Coker

Thomas

Walker

Mukherjee

Hebbar

: Time-dependent changes in matrix metalloproteinase activity and expression during the progression of congestive heart failure. Circ Res 82:482–495, 1998

31.

31 Toma

Pittenger

Cahill

Byrne

Kessler

: Human mesenchymal stem cells differentiate to a cardiomyocyte phenotype in the adult murine heart. Circulation 105:93–98, 2002

32.

32 Tomita

R-K

Weisel

Mickle

DAG

Kim

E-J

Sakai

Fabbi

Ballestero

Brunelli

Barsotti

: Autologous transplantation of bone marrow cells improves damaged heart function. Circulation 100(suppl II):247–256, 1999

33.

33 Vanhoutte

Schellings

Pinto

Heymans

: Relevance of matrix metalloproteinases and their inhibitors after myocardial infarction: a temporal and spatial window. Cardiovasc Res 69:604–613, 2006

34.

34 Van Vleet

Ferrans

: Clinical and pathological features of chronic adriamycin toxicosis in rabbits. Am J Vet Res 41:1462–1469, 1980

35.

35 Wakitani

Saito

Caplan

: Myogenic cells derived from rat bone marrow mesenchymal stem cells exposed to 5-azacytidine. Muscle Nerve 18:1417–1426, 1995

36.

36 Wang

Schulze

Suarez-Pinzon

Dyck

JRB

Sawicki

Schulz

: Intracellular action of matrix metalloproteinase-2 accounts for acute myocardial ischemia and reperfusion injury. Circulation 106:1543–1549, 2002

37.

37 Whittaker

Kloner

Boughner

Pickering

: Quantitative assessment of myocardial collagen with picrosirius red staining and circularly polarized light. Basic Res Cardiol 89:397–410, 1994

38.

38 Yokoseki

Yazaki

Suzuki

Imamura

Takenaka

Isobe

: Association of matrix metalloproteinase expression and left ventricular function in idiopathic dilated cardiomyopathy. Jpn Circ J 64:52–57, 2000

39.

39 Yoo

K-J

R-K

Weisel

Mickle

DAG

Jia

Z-Q

Kim

E-J

Tomita

Yau

: Heart cell transplantation improves heart function in dilated cardiomyopathic hamsters. Circulation 102(suppl III):204–209, 2000

Effects of Autologous Stem Cells on Immunohistochemical Patterns and Gene Expression of Metalloproteinases and Their Tissue Inhibitors in Doxorubicin Cardiomyopathy in a Rabbit Model

Abstract

Keywords

Material and Methods

Animals

Isolation, Expansion, and Labeling of MSCs

CM Injection and Stem Cell Transplantation

Clinical Observations

Echocardiographic Studies

Gross Examination

Histologic Analysis

Immunohistochemistry

Assay of mRNA Profiles

Statistical Analysis

Results

Clinical Observations

Gross Findings

Histopathologic Findings

Immunohistochemical Findings

mRNA Assays

Discussion

Footnotes

Acknowledgements

References