© 2004 by European Society of Cardiology
Copyright © 2004, European Society of Cardiology
Experimental and clinical parameters of cardiac reverse remodeling after LVAD support: a call for uniformity
aInstitute of Pathology, University of Essen, Hufelandstrasse 55, 45122 Essen, Germany
bDepartment of Pathophysiology, University of Essen, Hufelandstrasse 55, 45122 Essen, Germany
* Corresponding author. Institut für Arterioskleroseforschung an der Universität Münster Domagkstr. 3 48149 Münster, Germany. Tel.: +49-251-835-8625; fax: +49-251-835-6088. levkau{at}uni-muenster.de
Received 15 October 2003; accepted 23 October 2003
We are grateful for the comments of Razeghi et al. elicited by our recent report [1] on the important issue of comparability of experimental data raised from myocardial tissue samples among different studies on cardiac reverse remodeling under LVAD support. We completely agree that there are a number of non-standardized parameters among the different studies published thus far that make the direct comparison of experimental data difficult. These include patient characteristics and the underlying cause of heart failure as well as the make, duration and degree of LVAD support. In addition, we wish to emphasize another critical issue missing in most studies published thus far in the literature (with exceptions such as a study by Ragehzi et al. [2]): the lack of positive or negative correlative data between experimental findings and the clinical condition of the patients before and during LVAD support. Cardiac function parameters such as echocardiography, pulmonary artery and capillary wedge pressure are rarely mentioned, and are not correlated with the experimental data from myocardial tissue samples. Neither is the clinical follow-up or outcome of the LVAD-supported patients elaborated in most studies. However, it is just this attempt to associate clinical with experimental findings that is crucial for differentiating between causal and epiphenomenal observations in the pathophysiological process of cardiac reverse remodeling.
The issue of myocardial sample control Razeghi et al. refer to as an explanation between their study [3] and ours [1] may, indeed, be another major potential cause of incongruence. Transmural heterogeneity is not uncommon in the heart and has been shown for stress proteins [4,5], transcription factor activity [6] and fetal gene reactivation [7]. There are several ways we have tried to control for sample uniformity and comparability with other published studies. All tissue samples in our study were taken transmurally from the same anatomical part of the heart both prior to and after LVAD support (with pre-LVAD samples stemming from the apex and post-LVAD samples from the myocardium adjacent to the LVAD device). Avoiding differences in tissue composition is very important when gene or protein expression is to be compared, not only to ensure similar cell numbers present in the sample but also to control for the myocyte/non-myocyte ratio. Still, data from Western blots or quantitative PCR/gene array technology always represent the mean sum of gene or protein regulation in all different cell types present in the sample. Therefore, such data need to be independently evaluated by immunohistochemistry, in situ hybridization or single cell analysis using laser dissecting microscopy. Taking this into consideration, we have confirmed the regulation of both Erks and Akt phosphorylation in our study at the single cardiomyocyte level by immunohistochemistry and by defining an immunohistochemical cytoplasmic and nuclear staining score for phospho-Erks. The necessity to do this became also obvious from the strong staining for phosphorylated Erks in non-myocytes in the failing human hearts. With respect to the differences in the Western blot data on Akt and GSK-3β phosphorylation between the study of Razeghi et al. [3] and ours [1], it should be noted that, among the patients included in our study, which were twice the number included in the former study, there were two patients with only minor changes in Akt phosphorylation and one with an even opposite regulation. Therefore, a higher number of samples may be needed to detect statistically significant changes in Akt phosphorylation than Erk phosphorylation. The microarray gene profiling study by Chen and co-workers [8] that also examined Akt phosphorylation under LVAD support included only patients with idiopathic dilated cardiomyopathy (n = 6) in contrast to our study, which included both patients with dilated and ischemic cardiomyopathy (n = 5 each). Interesting in this context is that Chen and co workers refer to markedly different patterns of gene expression between ischemic and dilative cardiomyopathy although they do not show the data [8].
In addition to the descriptive measurements of Akt phosphorylation under LVAD support in all three studies including ours, we attempted to simulate the in vivo situation by applying mechanical stretch followed by release to cultured neonatal myocytes in vitro and Langendorff-perfused hearts in whole organ experiments. In both cases, Akt and GSK-3β phosphorylation/dephosphorylation was a reproducible and sensitive molecular readout of the applied stimulus. Therefore, we believe that regulation of these kinases is associated with the changes in wall tension and stretch occurring during LVAD support. The biological meaning of this regulation is still unclear as we haven't been able to associate it with cell size, hypertrophy or apoptosis, which all decrease after LVAD support [1]. A recent study by Haq et al. has shown that cardiomyocytes exposed to hypertrophic stimuli stabilize β-catenin via Akt phosphorylation and GSK-3β inactivation, which results in an increase of β-catenin levels, its translocation to the nucleus, and transcriptional activation of genes involved in hypertrophy and growth [9]. We have examined a number of cell–cell contact proteins prior to and after LVAD support but did not detect any differences in the β-catenin levels or subcellular localization after mechanical unloading (unpublished data).
Nevertheless, the major point by Razeghi et al. with respect to the need for myocardial sample control in all LVAD-related studies using myocardial tissues in order to ensure comparability remains indisputable, and we strongly support this notion. We are very much aware that collecting LVAD samples for analysis is often not controlled by the investigator but rather by the clinical or surgical situation and donor heart availability. However, we believe that careful documentation of all parameters currently known to influence gene and protein expression is of crucial importance and may be one approach to ensure comparability. This documentation should cover the broad spectrum of parameters ranging from those that are easier to monitor such as anatomical site of sampling and cellular/tissue composition to the much more complex parameters such as circadian rhythm and phase of the cardiac cycle [10,11]. In addition, attempts should be undertaken to set up a general nomenclature of such parameters for all investigators working in the field. This may render studies at individual centers more difficult to perform by limiting the number of comparable cases, but, on the other hand, it would greatly promote collaborations aimed at examining uniform subsets of patient samples. The latter would be of enormous help for the understanding of the obviously present but biologically still elusive process of cardiac reverse remodeling.
 |
References |
|---|
 Top References |
|---|
- Baba H.A., Stypmann J., Grabellus F., et al. Dynamic regulation of MEK/Erks and Akt/GSK-3β in human end-stage heart failure after left ventricular mechanical support: myocardial mechanotransduction-sensitivity as a possible molecular mechanism. Cardiovasc. Res. (2003) 59:390–399.
[Abstract/Free Full Text] - Razeghi P., Young M.E., Ying J., Depre C., Uray I.P., Kolesar J., et al. Downregulation of metabolic gene expression in failing human heart before and after mechanical unloading. Cardiology (2002) 97:203–209.[CrossRef][Web of Science][Medline]
- Razeghi P., Bruckner B.A., Sharma S., et al. Mechanical unloading of the failing human heart fails to activate the protein kinase b/akt/glycogen synthase kinase-3beta survival pathway. Cardiology (2003) 100:17–22.[CrossRef][Web of Science][Medline]
- Baba H.A., Grabellus F., August C., et al. Reversal of metallothionein expression is different throughout the human myocardium after prolonged left ventricular mechanical support. J. Heart Lung Transplant. (2000) 19:668–674.[CrossRef][Web of Science][Medline]
- Grabellus F., Schmid C., Levkau B., Breukelmann D., Halloran P.F., August C., et al. Reduction of hypoxia-inducible heme oxygenase-1 in the myocardium after left ventricular mechanical support. J. Pathol. (2002) 197:230–237.[CrossRef][Web of Science][Medline]
- Grabellus F., Levkau B., Sokoll A., Welp H., Schmid C., Deng M.C., et al. Reversible activation of nuclear factor-kappaB in human end-stage heart failure after left ventricular mechanical support. Cardiovasc Res. (2002) 53:124–130.
[Abstract/Free Full Text] - Altemose G.T., Gritsus V., Jeevanandam V., Goldman B., Margulies K.B. Altered myocardial phenotype after mechanical support in human beings with advanced cardiomyopathy. J. Heart Lung Transplant. (1997) 16:765–773.[Web of Science][Medline]
- Chen Y., Park S., Li Y., et al. Alterations of gene expression in failing myocardium following left ventricular assist device support. Physiol. Genomics (2003) 14:251–260.
[Abstract/Free Full Text] - Haq S., Michael A., Andreucci M., et al. Stabilization of beta-catenin by a Wnt-independent mechanism regulates cardiomyocyte growth. Proc. Natl. Acad. Sci. U. S. A. (2003) 100:4610–4615.
[Abstract/Free Full Text] - Young M.E., Razeghi P., Cedars A.M., Guthrie P.H., Taegtmeyer H. Intrinsic diurnal variations in cardiac metabolism and contractile function. Circ. Res. (2001) 89:1199–1208.
[Abstract/Free Full Text] - Yamamoto K., Dang Q.N., Maeda Y., Huang H., Kelly R.A., Lee R.T. Regulation of cardiomyocyte mechanotransduction by the cardiac cycle. Circulation (2001) 103:1459–1464.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||