OUP user menu

Expression of secreted frizzled related proteins 3 and 4 in human ventricular myocardium correlates with apoptosis related gene expression

Heike Schumann, Jürgen Holtz, Hans-Reinhard Zerkowski, Mechthild Hatzfeld
DOI: http://dx.doi.org/10.1016/S0008-6363(99)00376-4 720-728 First published online: 1 February 2000


Objective: Overload-induced heart failure is associated with myocyte apoptosis induced by unknown mechanisms. Wnt genes encode secreted signaling molecules that bind to frizzled receptors and stabilize cytosolic β-catenin which is translocated into the nucleus, acts as transcriptional activator and imparts an apoptosis resistant phenotype. This signaling pathway is antagonized by secreted frizzled related proteins (sFRPs) which modulate apoptosis susceptibility in cell culture models. On the basis of these considerations, the present investigation compares myocardial mRNA expression of sFRPs and the level of soluble β-catenin in tissue samples from nonfailing and failing hearts. Methods: Nonischemic transmural samples from human failing left ventricles and from nonfailing donor ventricles were used in the present study. The mRNA concentration of the Wnt-antagonists sFRP 1–4 were determined by quantitative reverse transcription polymerase chain reaction (RT-PCR). The myocardial localization of sFRP 3 and 4 expression was investigated using in situ RT-PCR. The pool of soluble β-catenin was quantified by Western blot analysis of protein extracts. Results: The mRNA levels of proapoptotic sFRPs 3 and 4 but not of sFRP 1 and 2 were elevated in failing ventricles compared to donor hearts. There was no significant difference between patients suffering from a dilated cardiomyopathy or a coronary heart disease. sFRPs 3 and 4 were expressed in cardiomyocytes and their expression correlated with the mRNA expression of the proapoptotic Fas/Fas-antagonist ratio, but inversely with the mRNA levels of the antiapoptotic bcl-xL. The size of the pool of 0.1% Triton soluble β-catenin tended to decrease in myocardial samples with high sFRP 3 and 4 expression levels. Conclusions: The results support the hypothesis that in failing human myocardium the Wnt/β-catenin pathway is attenuated by enhanced expression of two endogenous Wnt-antagonists. This might contribute to an apoptosis susceptible phenotype of overloaded human myocardium.

  • Apoptosis
  • Cell communication
  • Heart failure
  • Receptors

Time for primary review 27 days.

1 Introduction

Distension-induced apoptosis of cardiomyocytes has been proposed as a mechanism contributing to overload-associated progression of heart failure [1]. Indeed, myocyte apoptosis has been demonstrated in myocardium with experimental chronic pressure overload [2–5], in overloaded, nonischemic myocardium from hearts with myocardial infarctions [6–9], and in explanted failing human myocardium [10–12]. Furthermore, stretching of papillary muscles or isolated cardiomyocytes in vitro induced apoptosis [13–15] and hemodynamic unloading of failing human hearts by ventricular assist devices was associated with a reduction of myocardial apoptosis [16]. In spite of this plethora of supportive data this attractive proposal [1] is still regarded as controversial for several reasons: overload-associated myocyte apoptosis in vivo is also affected by altered neuroendocrine activity (reviewed in Ref. [17]); induction of apoptosis in isolated papillary muscle requires a degree of distension [13] that never occurs in vivo; isolated cardiomyocytes in culture have an extremely high frequency of apoptosis far above that in failing hearts [14,15] and, most importantly, an undisputed cellular model for the transduction of the mechanical stimulus into an apoptotic signal is not yet available.

The putative distension-induced apoptosis suggests a contribution of the cytoskeleton in signal transduction. A potential candidate for such a function is the Wnt signaling cascade which is involved in regulation of cytoskeletal rearrangements [18–22] as well as of apoptosis and proliferation [23,24]. This pathway is regulated by a large family of secreted Wnt proteins, activating membrane receptors related to the Drosophila gene product frizzled (reviewed in Refs. [25,26]). Wnt-mediated stimulation of frizzled receptors can activate several signals: the β-catenin pathway, mitogen activated protein kinase (MAPK) cascades and G-protein dependent pathways [26]. β-Catenin is a member of the armadillo multigene family. These proteins are characterized by 45 amino acid repeats which provide numerous protein binding sites and are supposed to function as molecular adaptors [27]. Armadillo related proteins combine structural functions as cell contact and cytoskeleton associated proteins and signaling functions modulating gene expression (reviewed in Ref. [28]). The signaling function of the Wnt/frizzled pathway is antagonized by secreted frizzled related proteins (sFRPs) [29–35] which bind to either Wnts or frizzled receptors [36]. Some Wnt-antagonistic sFRPs lower the cytosolic levels of β-catenin and enhance the cellular susceptibility for apoptosis [37].

Wnts are a large family with at least 16 different proteins in human; at least eight different genes encoding human frizzled receptors are known [26] and six mammalian sFRPs have been identified so far. Little is known about the Wnt/frizzled pathway in cardiomyocytes and about its role in ventricular overload-hypertrophy and apoptosis. The mRNAs of two frizzled receptors have been shown to be expressed in rat ventricular myocardium, with high expression in fetal/neonatal hearts, lowered expression in adult hearts and an enhanced reexpression in overloaded myocardium [38,39]. In myofibroblasts of infarcted rat hearts frizzled 2 expression was considerably enhanced [40]. Based on the findings that (1) Wnt/frizzled is involved in regulation of the cytoskeleton, (2) frizzled 2 has been suggested to play a role in cardiac hypertrophy and (3) sFRPs can induce a proapoptotic phenotype we hypothesized that enhanced expression of antagonistic sFRPs in cardiomyocytes occurs in the apoptosis susceptible phenotype of the failing human heart. As a first step to test the plausibility of this assumption we have analyzed the myocardial mRNA expression of all known sFRPs expressed in the heart and of the frizzled 2 receptor (fz2). An enhanced expression of sFRPs 3 and 4 in failing myocardium and the correlation of these expression levels with certain apoptosis regulating genes is in agreement with our hypothesis.

2 Methods

2.1 Human heart tissue specimens

Left ventricular transmural samples without visible signs of infarction were obtained from explanted hearts of patients with endstage heart failure during cardiac transplantation. These patients included 13 male and one female persons with an age of 54±2 years. A dilated cardiomyopathy (DCM) was diagnosed in seven patients and seven had a coronary heart disease (CHD). The ejection fraction was 25±3%. Samples from the left ventricle of donor hearts which were not transplanted for technical reasons were from one female and five males (age 39±4 years) and served as nonfailing controls. The investigation conforms with the principles outlined in the Declaration of Helsinki, the local ethical committee approved the use of human cardiac tissue in this study.

In some of these samples the mRNA expression of apoptosis related proteins such as bcl-xL and Fas had been quantified previously [12,16].

2.2 RNA preparation

Total RNA was isolated from the left ventricular samples by mechanical crushing in liquid nitrogen, homogenization in guanidinium–isothiocyanate solution and centrifugation through a caesium chloride cushion [41].

2.3 Reverse transcription polymerase chain reaction (RT-PCR)

One μg of total RNA was reverse transcribed in a total volume of 20 μl using 50 U Expand™ Reverse Transcriptase (Roche, Mannheim, Germany) and random hexamer oligonucleotide primers (Life Technologies, Eggenstein, Germany). The number of PCR-cycles was adjusted to the linear range of amplification for each set of primers. The primer sequences, binding sites and the sizes of the resulting PCR products are summarized in Table 1. The PCR reactions (25 μl) contained the following components: first strand cDNA reaction (0.5 μl); 1×PCR-buffer; 0,25–2 mM MgCl2, 12 μM of each dNTP; 5 pmol of each primer; 2 U Taq DNA Polymerase (Invitek, Berlin, Germany). Using a thermocycler (Techne, Cambridge, UK) a suitable number of amplification cycles was performed after an initial 2 min denaturation step at 95°C: 30 s denaturation at 94°C, 30 s primer annealing at 60°C, 40 s extension at 72°C. For quantification of mRNA expression of sFRP 1–4 a standard calibrated competitive RT-PCR was established. Synthetic cDNA standards with primer binding sites identical to those in the target PCR product were prepared using the PCR MIMIC™ Construction Kit (Clontech, Palo Alto, CA, USA). A dilution series of the cDNA standard was included in the PCR reactions. The PCR products were separated by agarose gel electrophoresis. The identity of the PCR products was verified by sequence analysis using the Thermo Sequenase Dye Terminator Cycle Sequencing Kit (Amersham Pharmacia, Braunschweig, Germany) and an ABI automated sequencer (PE Applied Biosystems, Eggenstein, Germany).

View this table:
Table 1

Summary of the used primers of amplification in the RT-PCR

GeneAccessionPrimer sequence (5′→3′)PositionFragment-length
GAP-DHaJ02642CAT CAC CAT CTT CCA GGA GCG220–240442 bpb
sFRP 1AF017987TCT ACA CCA AGC CAC CTC AG454–473616 bp
sFRP 2AF017986CTC GCT GCT GCT GCT CTT C282–300499 bp
sFRP 3U91903ATG GTC TGC GGC AGC CCG G70–88431 bp
sFRP 4AF026692GTT CCT CTC CAT CCT AGT GG260–279574 bp
fz2AB017364GGT GCC ATC CTA TCT CAG CT656–675403 bp
  • a GAP-DH=glycerolaldehyd-3-phosphate dehydrogenase.

  • b bp=Base pair.

2.4 In situ RT-PCR

Five μm paraffin sections of myocardial samples were deparaffinized and digested with 7.500 U/ml pepsin (Roche) for 10 min at 37°C. The reaction mix containing the components of the Titan™ One Tube RT-PCR System (Roche) and a specific pair of primers (see Table 1) with a 5′-terminal extension for concatemerization [42] was applied to the sections. Using the GeneAmp System 1000 (PE Applied Biosystems) 55 amplification cycles were performed after reverse transcription at 55°C for 30 min: 40 s denaturation at 95°C, 45 s primer annealing at 60°C, 70 s extension at 68°C. Negative controls were performed without primers. The amplified DNA was labeled in one cycle of DNA denaturation at 95°C, annealing at 60°C and extension at 72°C using Taq DNA polymerase, Stoffel fragment (PE Applied Biosystems), PCR digoxigenin (DIG) Labeling Mix (Roche) and a set of internal primers (sFRP 3: CTG CTC TCT GCC TGC TCC and GGG CTT GAT GGG CTC GTG; sFRP 4: CCC AAC CAC CTG CAC CAC and AAA TGC ACA CGC CAC GGT C). Immunohistochemical detection of incorporated DIG-labeled nucleotides were achieved with antiDIG-alkaline phosphatase (Roche) applied in a dilution of 1:500 using the DIG Wash and Blocking Buffer Set (Roche) and nitro blue tetrazolium chloride/5-bromo-4-chloro-3-indolyl phosphate (NBT/BCIP) as substrate.

2.5 Protein extraction and Western blot

Myocardial tissue samples from 13 patients (four nonfailing and nine failing hearts) were homogenized in liquid nitrogen, resuspended in ice cold extraction buffer (10 mM Tris–HCl pH 7.5, 140 mM NaCl, 0.1% Triton X-100, proteinase inhibitor cocktail (Roche)) and further homogenized using an Ultraturrax (IKA-Werke, Staufen, Germany). The extracts were incubated for 30 min on ice. After centrifugation at 10 000 g for 10 min, the protein in the supernatant (cytosolic fraction) was precipitated with methanol and chloroform [43]. The protein pellet was dissolved in 1% sodium dodecyl sulfate (SDS) and the protein concentration was determined by Micro BCA Protein Assay Reagent Kit (Pierce, Rockford, IL, USA). Ten μg of total protein was loaded onto 7.5% SDS polyacrylamide gels, electrophoretically separated and blotted on nitrocellulose. A monoclonal antibody against β-catenin (clone 6F9: Sigma, Deisenhofen, Germany) was applied at a 1:5 000 dilution in TBS–T (10 mM Tris–HCl pH 8.0, 150 mM NaCl, 0.05% Tween 20). Bound β-catenin antibodies were detected using the ECL system and Hyperfilm (Amersham Pharmacia).

2.6 Data analysis

mRNA expression levels were determined by densitometric scanning of ethidium bromide stained agarose gels and evaluated using the AIDA software program (Aida beta, Raytest, Straubenhardt, Germany). All analyses were performed in duplicate. The mRNA data are given as mean±S.E.M. (molecules/μg total RNA). Student's t-test was used to assess the statistical significance of variations observed between patient groups. For quantification of β-catenin exposed Hyperfilms were evaluated using the AIDA software program.

3 Results

3.1 Unchanged expression of sFRPs 1, 2 and frizzled 2 in failing human myocardium

The sFRP 1 mRNA was detected in all samples. Expression levels varied considerably within each group of patients and the donors: four out of the seven DCM patients showed a strong expression as well as three out of seven CHD patients and three out of the six donors (Fig. 1). The quantification by competitive RT-PCR revealed that the average value at patients (n=14; 4.18±1.00·104) and donors (n=6; 3.66±1.18·104) were not significantly different. Results were similar for DCM and CHD patients.

Fig. 1

Quantification of the mRNA expression of GAP-DH, frizzled 2 (fz2) and sFRPs 1–4 in human left ventricular myocardium by semiquantitative RT-PCR. The figure shows PCR products after agarose gel electrophoretical separation. The PCR amplification was within the linear range for all mRNAs analyzed. GAP-DH was expressed at the same level in all samples. The size of the soluble pool of β-catenin was analyzed by Western blotting with a β-catenin specific antibody after extraction of tissue samples with 0.1% Triton and SDS gel electrophoretic separation of 10 μg of Triton soluble protein. Samples for protein extraction were not available from all donors and patients whose mRNA has been analyzed. DCM=dilative cardiomyopathy; CHD=coronary heart disease; M=100 base pair-ladder.

The mRNA expression of sFRP 2 was low in most samples of ventricular myocardium and was not detectable in one donor and in two CHD patients under the reaction conditions described. One donor and one DCM patient showed extremely high levels of sFRP 2 mRNA compared to other patients or donors (Fig. 1). The quantification revealed again considerable variations within each group but no statistically significant differences between failing (2.24±1.83·104) and nonfailing myocardium (1.10±0.85·104). There was no significant difference between the DCM and CHD groups.

Donor number 3 differed considerably from other donors and showed elevated levels of sFRP 1 and 2 expression. A left ventricular hypertrophy was diagnosed in this donor.

The mRNA encoding frizzled 2 was detected in all myocardial samples (Fig. 1), but its expression was not significantly upregulated in heart failure patients compared to donors.

3.2 Expression of sFRPs 3 and 4 is elevated in failing human myocardium and expression correlates with expression of apoptosis related genes

sFRP 3 was expressed in all myocardial samples (Fig. 1). All donors revealed relatively low expression levels, whereas some patients showed a strong increase in mRNA. In average, samples from failing myocardium showed a two- to threefold increase in expression compared to donors (Fig. 2A). Again, expression levels varied within each group of patients (Fig. 1) but no significant differences were observed between the CHD and DCM patients (Fig. 2A).

Fig. 2

The mRNA expression of sFRP 3 (A) and sFRP 4 (B) in human left ventricular myocardium of donor and terminally failing hearts. Correlation of mRNA ratio Fas/FasExo6Del (C) and of bcl-xL mRNA to sFRP 3 (D). * P=0.01; n.s. not significant (P=0.15).

The mRNA encoding sFRP 4 was detected in all samples except for one donor (Fig. 1). Three out of six donors showed relatively low expression of sFRP 4, with a mean value of 1.33±0.29·103 for all donors. sFRP 4 expression was strongly enhanced in most heart failure patients. However, in some of these patients expression levels were similar to those detected in donors. In average, the mRNA level tended to be elevated in failing hearts compared to donors (Fig. 2B). Expression of sFRP 4 in patients suffering from DCM and CHD was similar (Fig. 2B). Again, donor number 3 differed from the majority of other healthy donors by an elevated expression level of sFRP 4 and to a lesser extent of sFRP 3 suggesting that expression of these mRNAs is already elevated at an early stage of hypertrophy. The mRNA expression of sFRP 4 correlated positively (r=0.69, P<0.001) to the expression of sFRP 3, indicating coordinated regulation of the two antagonists.

The proapoptotic mRNA ratio of Fas per Fas-antagonist FasExo6Del (generated by alternative splicing of exon 6) [12] correlated positively with the expression of sFRP 3 (Fig. 2C) and tended to correlate with the expression of sFRP 4 (r=0.38, P=0.12). The expression of the antiapoptotic bcl-xL [16] correlated negatively with the expression of sFRP 3 (Fig. 2D).

3.3 sFRPs 3 and 4 are expressed in cardiomyocytes

Using in situ RT-PCR, the mRNA of sFRPs 3 and 4 was detected mainly in cardiomyocytes (Fig. 3). These mRNA expressions were evenly distributed throughout the tissue sections.

Fig. 3

Detection of sFRP 3 (A, B as negative control) and sFRP 4 (C, D as negative control) mRNA in cardiomyocytes of human myocardial tissue sections by in situ RT-PCR. RNA expression is visualized as a dark precipitate in the cells.

3.4 Cytosolic β-catenin levels are reduced in samples with high levels of sFRP 3

Western blot analysis of 0.1% Triton soluble protein extracts revealed a single 94 kilodalton band specifically reacting with the β-catenin antibody in all investigated myocardial samples (Fig. 1). Quantitative analysis showed a tendency for lower β-catenin in the cytosolic fraction of heart failure patients (n=9; 3.47±4.09) compared to donors (n=4; 6.84±3.28 relative units; n.s.). The amount of cytosolic β-catenin tended to correlate negatively to the mRNA expression levels of sFRP 3 (r=0.49, P=0.09).

4 Discussion

Our study shows that sFRPs 1–4 are expressed in ventricular myocardium from healthy donors and from patients with terminal heart failure. This finding confirms and extends previous reports on cardiac expression of sFRPs in adult mammals [29,33,37,44]. Moreover, we show enhanced expression of the Wnt-antagonists sFRP 3 and sFRP 4 in failing ventricular myocardium while the expression of the frizzled receptor fz2 is essentially unaltered in heart failure. These antagonistic sFRPs are expressed mainly in cardiomyocytes, and their expression correlates positively with the proapoptotic Fas/FasExo6Del mRNA ratio and negatively with the mRNA of the antiapoptotic bcl-xL. Interestingly, hemodynamic unloading of failing human myocardium by ventricular assist devices renormalizes this elevated Fas/FasExo6Del ratio and the lowered bcl-xL expression [16]. This indicates that overloaded and distended myocytes of the adult myocardium in situ shift their phenotype towards an apoptosis susceptible pattern and that the enhanced expression of sFRPs 3 and 4 is part of this overload pattern. This is in agreement with our proposal that endogenous inhibition of the Wnt signaling pathway contributes to myocyte apoptosis in cardiac overload. Ideally, mechanisms of distension-induced apoptosis should be analyzed in isolated cells. However, isolated cardiomyocytes in culture have a very high level of basal apoptosis [14,15] for unexplained reasons, and the process of cellular isolation severely interferes with the integrity of the cytoskeleton and the balance between the two cellular pools of β-catenin (see below), which should affect the Wnt signaling cascade. Our data show activation of putatively proapoptotic Wnt-antagonists in overloaded myocardium independently from artefacts due to cellular isolation.

By various experimental approaches, the sFRPs have been shown to play a role in induction of apoptosis. The sFRP 4 (with a tendency for enhanced expression in failing hearts, Fig. 2B) has been identified as an apoptosis relevant protein by differential display analysis from several involuting tissues undergoing apoptosis [44,45]. sFRP 1 (expression not elevated in failing hearts) and another family member (SARP3, not expressed in myocardium) have been identified as “secreted apoptosis related proteins” (SARPs) in conditioned cell culture media with apoptosis modulating activity [37]. Other members of the sFRP family (sFRP 2, not elevated, and sFRP 3, elevated in failing hearts, Fig. 2A) have been characterized as extracellular modulators of Wnt/frizzled signaling [29,30,33–35,37,46]. They modulate Wnt signaling by interacting either with Wnts thereby preventing Wnts from binding and activating their receptors or by directly interacting with frizzled receptors [36]. For sFRP 1, a connection between the apoptosis inducing phenotype and the Wnt/frizzled signaling pathway has been suggested since transfection of sFRP 1 into an epithelial cell line lowered cytosolic β-catenin and rendered cells susceptible to several proapoptotic stimuli [37]. In contrast, the putative Wnt modulator sFRP 2 (=SARP1) rendered transfected cells more resistant to apoptotic stimuli and elevated their cytosolic β-catenin levels [37]. The molecular mechanism underlying this effect is not yet understood although the data suggest an agonistic role of sFRP 2 in the Wnt signaling pathway.

Although frizzled receptors have been shown to activate not only β-catenin directed gene transcription, but also MAPK and G-protein dependent pathways [26], β-catenin seems to be the main target of most Wnts. Among Wnts that activate β-catenin are Wnt-1 and Wnt-8 which have been shown to interact with sFRP 3 and 4 [34,44]. There are two pools of β-catenin in the cell, a pool with a structural function that is associated with cell contacts and a soluble cytoplasmic pool that is involved in regulating gene expression patterns after nuclear import and interaction with transcription factors (Fig. 4, [47–50]). Cytosolic β-catenin as the central element of the Wnt signaling pathway has been related to apoptosis in epithelial cells: depletion of cytosolic β-catenin through recruitment into cell contacts induced apoptosis and suppressed proliferation [51] whereas stabilization of high levels of cytosolic β-catenin through consitutive stabilization resulted in elevated proliferation and reduced apoptosis [52]. The modulation of apoptosis and proliferation by β-catenin is thought to result from alterations in gene expression [23,24]. Once apoptosis has been started, cytosolic β-catenin is further depleted due to cleavage by activated caspase-3 [53].

Fig. 4

Current model of the Wnt/β-catenin signaling pathway. Wnt binds to frizzled receptors, which activate a cytoplasmic kinase related to Drosophila gene, dishevelled. Dishevelled phosphorylates glycogen synthase kinase 3 (GSK-3) which is inactive in its phosphorylated form. Thus, with frizzled activation by Wnt, β-catenin stays unphosphorylated and accumulates in the cytoplasm, binds to high mobility group (HMG)-box transcription factors and is involved in regulating gene expression in the nucleus. Wnt signaling is inhibited by sFRPs which can either bind to Wnts and prevent them from activating frizzled receptors or directly inactivate the receptors. In the absence of a Wnt signal, GSK-3 is active and phosphorylates β-catenin which becomes degraded. In the absence of a soluble pool of β-catenin, transcription regulated by certain HMG-box transcription factors is not activated. The pool of β-catenin bound to cadherin in cell contacts is not available for signaling.

Since β-catenin is regulated predominantly at the posttranslational level, we have analyzed protein levels in the soluble pool (0.1% Triton) by Western blot analysis. We show a tendency for reduction of soluble β-catenin in samples from failing hearts and in samples with high expression levels of sFRPs 3 and 4. These alterations do not reach the level of significance, probably because of the low number of tissues available and the uncertainties of extracting soluble proteins from intact tissues. However, these observations are in agreement with the assumption that sFRPs 3 and 4 antagonize Wnt signaling through depletion of cytosolic β-catenin thereby enhancing apoptosis susceptibility. Alternatively, sFRP 3 and 4 could also interact with other Wnts besides Wnt-1 and Wnt-8 or with frizzled receptors that do not activate the β-catenin signaling pathway although there is no experimental evidence supporting this hypothesis.

The substantial heterogeneity, both in myocardial sFRP expression and in cytosolic β-catenin levels (Fig. 1), indicates that multiple influences besides heart failure-associated myocyte distension affect these signaling systems in the human myocardium. An example might be donor 3 who had a left ventricular hypertrophy with high expression of all sFRPs and reduced levels of cytosolic β-catenin, suggesting that upregulation of sFRPs may be an early event during development of hypertrophy. On the other hand, the system might be activated in terminal heart failure not only by cellular distension but also by side-to-side slippage of myocardial fibers [54] with transient interruption and reconnection of cell contacts between adjacent myocytes. These mechanical factors as well as the enhanced neuroendocrine activity in the patients are intermittently modulated by therapy, the influence of which on this signaling system remains unknown.

5 Conclusions

We show that sFRP 3 and 4 expression is elevated in failing compared to nonfailing human myocardium and the size of the pool of soluble β-catenin tends to decrease in tissue samples with high expression of sFRPs 3 and 4. Expression of sFRPs has been associated with apoptosis in other cellular models. Our findings are compatible with the hypothesis that paracrine Wnt/frizzled signaling is inhibited by sFRPs in overloaded myocardium, and that this inhibition is associated with depletion of cytosolic β-catenin and the induction of an apoptosis-susceptible myocyte phenotype. The postulated model of myocyte apoptosis regulation by Wnt/frizzled/β-catenin signaling may also contribute to the enhanced myocyte apoptosis during new formation of intercellular contacts in failing myocardium with overload-induced dilative remodeling and side-to-side slippage of myocytes.


This study was supported by a grant from the BMBF (German Federal Minister for Education, Research and Technology) to M.H., No. 01ZZ9512. We deeply appreciate the cooperation and continuous support by the patients and the staff of the Halle heart transplant program at the Clinic of Cardiothoracic Surgery, Martin-Luther-University Halle–Wittenberg. We thank K. Bauer and Dr. R. Hinze for help with the paraffine sections and U. Vinzens for help with photographic documentation.


  • 1 Present address: Clinic for Cardiothoracic Surgery, University Clinic of Kantonspital Basel, Basel, Switzerland.


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
  44. [44]
  45. [45]
  46. [46]
  47. [47]
  48. [48]
  49. [49]
  50. [50]
  51. [51]
  52. [52]
  53. [53]
  54. [54]
View Abstract