Hemodynamic Support by Left Ventricular Assist Devices Reduces Cardiomyocyte DNA Content in the Failing Human Heart
Background— Whether adult cardiomyocytes have the capacity to regenerate in response to injury and, if so, to what extent are still issues of intense debate. In human heart failure, cardiomyocytes harbor a polyploid genome. A unique opportunity to study the mechanism of polyploidization is provided through the setting of hemodynamic support by left ventricular assist devices. Hence, the cardiomyocyte DNA content, nuclear morphology, and number of nuclei per cell were assessed before and after left ventricular assist device support.
Methods and Results— In 23 paired myocardial samples, cardiomyocyte ploidy was investigated by DNA image cytometry, flow cytometry, and in situ hybridization. Nuclear cross-sectional area and perimeters were measured morphometrically, and the binucleated cardiomyocytes were counted. The median of the cardiomyocyte DNA content and the number of polyploid cardiomyocytes both declined significantly from 6.79 c to 4.7 c and 40.2% to 23%, whereas a significant increase in diploid cardiomyocytes from 33.4% to 50.3% and in binucleated cardiomyocytes from 4.5% to 10% after unloading was observed.
Conclusions— The decrease in polyploidy and increase in diploidy after left ventricular assist device suggest a numeric increase in diploid cardiomyocytes (eg, through cell cycle progression with completion of mitosis or by increased stem cells). The cardiac regeneration that follows may serve as a morphological correlate of the recovery observed in some patients after unloading.
Received July 18, 2008; accepted December 4, 2009.
Whether adult human cardiomyocytes have the capacity to regenerate by mitotic cell division in vivo and, if so, to what extent and how has been intensely studied and controversially discussed in the last decades. Most cardiomyocytes have been assumed to be terminally differentiated with an extremely limited, if any, capacity to reenter the cell cycle with full progression to mitosis. This view has been challenged recently.1 It was demonstrated that in human myocardial infarction a proportion of adult cardiomyocytes enter in mitosis and proceed toward karyokinesis/cytokinesis and formation of daughter cells.2 Cardiomyocytes at the periphery of myocardial infarction undergo endomitosis and polyploidization instead of conventional mitosis with subsequent cell division.3 In a rapid pacing-induced chronic heart failure (CHF) model in dogs, Setoguchi and coworkers4 were able to demonstrate cell cycle reentry and cell cycle progression of cardiomyocytes, including mitosis in mononucleated and binucleated cells. A new aspect in this issue was provided by the demonstration of the existence of cardiac stem cells,5–7 to which some of the previous observations on cardiomyocyte regeneration have been attributed. Moreover, by “genetic fate mapping” experiments in a transgenic mouse model, it was recently suggested that precursor cells may participate in the generation of new cardiomyocytes after injury but not during normal aging.8
Editorial see p 957
Clinical Perspective on p 996
In CHF, left ventricular assist devices (LVADs) lead to numerous molecular changes in cardiomyocytes and decreased cardiac hypertrophy (“reverse cardiac remodeling”).9,10 This process involves transcriptional, molecular, and signal transduction pathways that regulate myocardial growth, function, regeneration, and death.11 A decrease in cardiomyocyte nuclear size and chromatin density after unloading with DNA cytometry and morphometry was reported in 2 patients.12
In the present study, we have addressed the possibility of myocardial regeneration by investigating the plasticity and dynamics of cardiomyocyte DNA content and ploidy, as well as nuclear size and number of nuclei before and after mechanical support by LVAD. We show that there are dramatic changes in mean cardiomyocyte DNA content per nucleus and per individual cardiomyocyte and in the nuclear size and number of nuclei after hemodynamic unloading, suggesting that changes in nuclear size and ploidy are crucially involved in reverse cardiac remodeling.
Twenty-three patients (22 male patients, 1 female patient; 11 with Novacor N100, Baxter Healthcare Corp, Novacor Division, Oakland, Calif; 3 with MicroMed DeBakey axial flow pump system, 3 with Incor, and 6 with Heartmate 1000 IP, Thermo Cardiosystems Inc, Woburn, Mass) who underwent LVAD implantation for end-stage CHF as a bridge to transplantation were studied. The causes underlying CHF included dilated cardiomyopathy (n=14), ischemic cardiomyopathy (n=7), and myocarditis (n=2). The median patient age was 47 years (minimum, 3 years; maximum, 60 years). The median duration of ventricular support by LVAD was 146 days (minimum, 17 days; maximum, 426 days). Five hearts from donors without a history of cardiovascular disease that were not used for transplantation served as controls. Informed consent was obtained from every patient, and the study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki.
Determination of Cardiomyocyte Diameter
Cardiac muscle specimens from each patient removed at the time of LVAD implantation (from the cardiac apex) and at transplantation (from the ventricular wall cranial to the apex to avoid sampling of scar tissue) were stained with the periodic acid Schiff reaction. With the use of an image analysis program (KS 300, Zeiss, Oberkochen, Germany), the diameters of at least 100 cardiomyocytes were determined in randomly selected visual fields at a 400-fold magnification.
Determination of Binucleated Cardiomyocytes
Differentiation of cardiomyocyte nuclei from fibroblast nuclei is difficult in hematoxylin and eosin–stained slides because cell margins often blend with the interstitium. To discriminate cardiomyocyte from fibroblast nuclei, slides were subjected to immunohistochemistry with antibodies directed against pan-cadherin (Sigma, Munich, Germany; 1:10 000 with pretreatment by heating in citrate buffer at pH 6 at 90°C), which helps highlight the intercalated disks and allows discrimination of the longitudinal margins. The cardiac tissue was simultaneously stained by diastase-periodic acid Schiff reaction for sarcoplasmic membrane staining and removal of glycogen to allow discrimination of cell margins other than intercalated disks. With the combination of the 2 staining methods, the cell margins of the cardiomyocytes could be clearly seen in the shape of a “frame” in areas in which the cardiomyocytes were longitudinally oriented. In these sections, the number of binucleated cardiomyocytes was determined in 200 cardiomyocytes in each patient before and after LVAD.
Determination of Cardiomyocyte DNA Content
DNA Image Cytometry on Isolated Cardiomyocyte Nuclei
For cell separation, 2 to 3 sections 50- to 100-μm thick from 23 patients before and after LVAD were dewaxed and rehydrated according to standard procedures at room temperature on a shaker. The specimens were digested with a pepsin solution (0.025 g in 5 mL 0.07 mol/L HCl at 37°C) for 30 minutes. The reaction was stopped by 5 mL cold phosphate buffered saline buffer. The suspended nuclei underwent Feulgen staining, and DNA cytometry was done at a 400-fold magnification measuring 300 control cardiomyocytes and 30 to 40 diploid reference cells (fibroblasts). Isolated single cardiomyocyte nuclei have been identified according to their size, nuclear shape, and chromatin pattern. To verify the assignment of nuclei as of either cardiomyocyte or noncardiomyocyte/fibroblast origin, immunofluorescence labeling of the nuclei was performed with the cardiomyocyte-specific marker c-troponin 1. Isolated nuclei from paired myocardial samples were stained with antibodies directed against human c-troponin 1 (1:50, MAB 3152, Millipore GmbH, Schwalbach, Germany) visualized with fluorescein isothiocyanate (polyclonal rabbit anti-mouse, DakoCytomation, Hamburg, Germany). With this approach, it can be demonstrated that cardiomyocyte nuclei showing a more rotund, vesicular, and less chromatin-dense morphology are positively labeled by antibodies directed against c-troponin 1 indicating cardiomyocyte origin, whereas fibroblast nuclei showing slender, longitudinal, and more chromatin-dense morphology remain unlabeled by c-troponin 1 (Figure 1A and 1B).
DNA Image Cytometry on Histological Sections
To investigate whether there are changes in the DNA content per individual cardiomyocyte (including binucleated cardiomyocytes), Feulgen staining as described above was applied to histological slides of 10 paired myocardial samples. Using light microscopy with partially closed condenser lens, we could reliably distinguish cardiomyocytes from noncardiomyocytes. The DNA content in ≥1 nuclei per individual cardiomyocyte was counted in cardiomyocytes with clearly visible sarcoplasmic margins as described above.
In Situ Hybridization of Chromosomal Centromeres
As a different independent approach to assess changes in the cardiomyocyte DNA content, in situ hybridization of chromosomal centromeres was chosen to verify the findings obtained by optical density DNA cytometry. Paraffin-embedded tissue slides were dewaxed by xylol (3 times for 5 minutes) and 100% ethanol (2 times for 5 minutes) and air-dried. The deparaffinized slides were stored in HCl (0.2 N) for 20 minutes and rinsed with distilled water and saline–sodium citrate buffer. The slides were then incubated in the pretreatment solution (Paraffin Pretreatment Kit II, Abbott, Wiesbaden, Germany) for 30 minutes at 80°C. After rinsing with saline–sodium citrate buffer, the slides were incubated with protease for 20 minutes at 37°C. Finally, the slides were incubated for hybridization with 5 μL CEP8 (No. 6J 3708, Abbott) probes directed against the centromeric region of chromosome 8 at 73°C for 5 minutes for melting and at 42°C for 22 hours for hybridization. After hybridization, the coverslips were removed and the slides were washed with 2× saline–sodium citrate buffer and formamide. Nuclei were visualized by 4′,6-diamidino-2-phenylindole staining. Evaluation was performed by counting the signals in 100 cardiomyocyte nuclei before and after unloading; results are given as percentages (Figure 1C).
Flow Cytometry–Based Evaluation of Cardiomyocyte DNA Content
To test the validity of the data obtained by optical density DNA image cytometry based on Feulgen staining and in situ hybridization of chromosomal centromeres, we used flow cytometry on 7 frozen paired myocardial samples. Dissociation of small tissue pieces (a few cubic millimeters each) of fresh-frozen paired myocardial samples was performed as previously published.13 Disaggregated cells were indirectly stained anti-MEF2 (clone 17785, 1:10, Santa Cruz Biotechnology, Inc, Santa Cruz, Calif; fluorescein isothiocyanate–conjugated secondary antibody, 1:20, Dako GmbH, Hamburg, Germany). Samples were measured on a FACSCanto-II analyzer (BD Biosciences, Heidelberg, Germany). Data were recorded and analyzed with FACSDiva software (version 6.1.1, BD Biosciences). MEF2-negative events were excluded from the analysis. Subfractions of diploid, tetraploid, octaploid, and hexadecaploid particles were quantified with a signal-area histogram display on positive MEF2 gating. The distribution of aneuploid particles before and after LVAD support versus the main diploid fraction was identified by quantification of the mean fluorescence intensity. The distribution of individual aneuploid versus diploid fractions was quantified with histogram regions and was finally normalized to 100%.
Determination of Cardiomyocyte Nuclear Profile Area and Perimeter
To determine changes in nuclear size associated with changes in DNA content, the profile area (cross-sectional area, or area of the interior of the structure) and perimeter (circumference, or length of the contour of the structure) of cardiomyocyte nuclei were measured with an image analysis program (KS 300, Zeiss, Oberkochen, Germany). One hundred nuclei in each patient before and after LVAD, including controls, were measured, and the profile area and perimeter were calculated automatically. Moreover, mitotic figures (metaphases) were intensely sought throughout the whole slides examined.
All of the data were analyzed and are expressed as median with minimum, maximum, and range and depicted as box plots. For evaluation of statistical significance of difference between samples before and after LVAD, the nonparametric Wilcoxon test for paired samples and Spearman ρ bivariate correlation analysis were used (SPSS, version 17.0, SPSS Inc, Chicago, Ill). Intergroup differences were calculated by 1-way ANOVA, followed by posthoc analysis according to Duncan. Values of P≤0.05 were considered significant.
Differences in Nuclear Size and Number of Nuclei per Cardiomyocyte Before and After LVAD Support
A marked observation throughout our studies was the clear difference in nuclear size in matched myocardial samples before and after LVAD from the same patient: The cardiomyocyte nuclei before LVAD support were much larger than afterward (Figure 1D). In addition, it appeared that there were more binucleated cardiomyocytes in LVAD-supported hearts (Figure 1D, inset). To assess these apparent differences in nuclear size and number of binucleated cardiomyocytes, we examined quantitatively nuclear profile area and perimeter and counted the number of binucleated cardiomyocytes.
Cardiomyocyte Nuclear Size Decreases After LVAD Support
Before LVAD, the median of the cardiomyocyte nuclear profile area was 102.13 μm2 (minimum, 43.78 μm2; maximum, 175.35 μm2; range, 131.57 μm2). After ventricular unloading, the nuclear profile area decreased significantly to a median of 74.53 μm2 (minimum, 19.28 μm2; maximum, 151.76 μm2; range, 132.48 μm2; P<0.05; Figure 2A). There was a statistically significant difference in the cardiomyocyte nuclear profile area in failing and unloaded hearts and in control donor hearts (median, 54.61 μm2; minimum, 27.09 μm2; maximum, 66.71 μm2; range, 39.62 μm2). Accordingly, the median of the cardiomyocyte nuclear perimeter was measured as 39.93 μm (minimum, 25.69 μm; maximum, 51.24 μm; range, 25.55 μm). After unloading, it was significantly diminished to 34.61 μm (minimum, 12.51 μm; maximum, 49.20 μm; range, 36.69 μm; P<0.05). The nuclear perimeters of failing, supported, and control hearts (median, 29.02 μm; minimum, 20.78 μm; maximum, 32.22 μm; range, 11.44 μm) were also statistically different (data not shown).
Number of Cardiomyocytes With ≥2 Nuclei Increases After LVAD
Before and after hemodynamic unloading, cardiomyocytes with 2 nuclei can be observed. Occasionally, cardiomyocytes with 3 to 5 nuclei are seen, but they constitute only a small proportion of cardiomyocytes with >1 nucleus. Before unloading, the median number of binucleated and multinucleated cardiomyocytes per 100 counted cardiomyocytes was 4.5 (minimum, 1.5; maximum, 11.5; range, 10.0), but it increased significantly after support (median, 10; minimum, 2.5; maximum, 22.5; P<0.05). There were no significant differences between the number of binucleated cardiomyocytes before LVAD and in control hearts (Figure 2B).
Cardiomyocyte Mean DNA Content Is Increased in Failing Hearts and Decreases Again After LVAD
By examining at least 300 single isolated cardiomyocyte nuclei (identified as cardiomyocyte nuclei according to their size, nuclear shape, and chromatin pattern; the assignment of origin by morphological criteria was controlled by immunofluorescence with the cardiomyocyte-specific marker c-troponin 1; Figures 1A, 1B, 3A, and 3⇓⇓B), we measured the median of the cardiomyocyte DNA value as 6.79 c (minimum, 3.60 c; maximum, 13.60 c; range, 10.0 c) in failing hearts before LVAD. After LVAD, the median of the DNA value decreased significantly to 4.70 c (minimum, 2.29 c; maximum, 10.9 c; range, 8.61; P<0.05). There was no difference in DNA content between unloaded and control hearts (median, 2.97 c; minimum, 2.41 c; maximum, 3.96 c; range, 1.55 c; Figure 3⇓C). In situ hybridization signals with a human chromosome 8 centromeric probe showed a significant decrease in cardiomyocyte DNA after unloading in line with the results obtained by optical DNA densitometry (Wilcoxon test for paired samples, P<0.05). Before LVAD, the median of the percentage of chromosome 8 signals was 3.9% (minimum, 2.3%; maximum, 6.1%; range, 3.8%) and significantly decreased to 2.7% (minimum, 1.5%; maximum, 4.2%; range, 2.7%) after unloading. Post hoc analysis according to Duncan showed a significant difference between before and after unloading, whereas there was no significant difference between unloaded hearts and controls (median, 2.1%; minimum, 2%; maximum, 2.5%; range, 0.5%), indicating that the DNA content decreased to the level of normal control hearts (Figure 3⇓D). Moreover, there was a significant positive correlation between the cardiomyocyte mean DNA content and the numbers of centromeres per nucleus (P<0.05, r=0.467; Figure 3⇓E).
To investigate whether the decrease in DNA content in isolated nuclei was due merely to karyokinesis, the DNA content was determined in cardiomyocytes (including binucleated cardiomyocytes) on histological sections from 10 paired myocardial samples by DNA optical densitometry. In accordance with the results obtained from single isolated nuclei, a significant decrease in DNA content per individual cardiomyocyte was noted after unloading (P<0.05). Before unloading, the median of the DNA content per individual cardiomyocyte was 5.9 c (minimum, 3.5 c; maximum, 10.6 c; range, 7.1 c) and decreased significantly to 4.5 c after unloading (minimum, 2.8 c; maximum, 8.1 c; range, 5.3 c; Figure 3⇑F).
Cardiomyocyte Polyploidy Is Reduced Whereas Diploidy Is Increased After Ventricular Unloading
Using the raw data obtained by DNA cytometry, we discriminated cardiomyocytes according to their ploidy status (2, 4, 8, 16, and >32 c) before and after LVAD and calculated percent differences. There was a significant increase in the percentage of diploid cardiomyocytes in LVAD-supported hearts (from 33.4% to 50.3%; P<0.05), whereas the percentage of cardiomyocytes with 8, 16, and 32 c was significantly decreased after LVAD (from 31.8% to 19.5%, 6.9% to 3.3%, and 1.5% to 0.2%, respectively; P<0.05; Figure 3⇑G). No numeric changes in tetraploid cardiomyocytes were observed.
Moreover, flow cytometry was performed to verify the data obtained by optical density DNA cytometry and in situ hybridization showed an increase in diploid and a decrease in polyploid cardiomyocyte nuclei. Because of the limited number of samples, no statistical analysis was done (data not shown).
The decreases in mean cardiomyocyte DNA content, nuclear perimeter and profile area, and cardiomyocyte diameters were independent of the type of CHF, the type of LVAD implanted, and age; in particular, there was no difference between the pediatric (<20 years of age) and adult (>20 years of age) age groups. Changes in the above parameters were also independent of the duration of mechanical ventricular support (data not shown).
For detailed information concerning individual patient data, directionality and dimension, and exact probability values, refer to Tables I and II in the online-only Data Supplement.
Several studies have shown that cardiomyocytes can reenter the cell cycle and complete cell division on divergent stimuli.14–17 Focal cardiomyocyte DNA synthesis, followed by polyploidy and occasional karyokinesis, has been observed in the periphery of myocardial infarctions in several animal models.2–4 In human patients, the majority of cardiomyocytes in congestive heart failure and cardiac hypertrophy harbor a polyploid genome.18–20 Neither the mechanisms nor the consequences of such polyploidy are clear, however, especially with respect to whether it is related to actual cell division. The ultimate proof of cell division taking place—an increased cardiomyocyte turnover—is impossible to address by such studies. Their inherent problem is the inability to provide more than just a snapshot of a putative cell division process without the possibility of follow-up. Bergmann and coworkers21 used radiocarbon dating of DNA to analyze the age of the cardiomyocyte DNA and compared it with the patients’ chronological age to estimate the long-term cardiomyocyte turnover in humans. Using this approach, the authors calculated that ≈50% of cardiomyocytes are changed during a normal lifespan. Another unique way to investigate the fate of cardiomyocytes over time is to study matched samples from the same human failing heart before and after its relief from hemodynamic overload by LVAD. This technique allows the monitoring of dynamic changes in the same heart under treatment. Our study with such samples provides circumstantial evidence for karyokinesis and/or cytokinesis occurring on a large scale after relief of hemodynamic overload in the failing human heart.
In our study, we have confirmed the existence of vast polyploidy in the failing human heart before LVAD with >70% of cardiomyocytes harboring a genome ≥4 c and reaching 32 c. Most striking, however, was our observation of a dramatic 2-fold reduction in the mean cardiomyocyte DNA content after LVAD with an ≈2-fold increase in the number of diploid cardiomyocytes, together with a significant decrease in the number of polyploid cardiomyocytes. There may be explanations for some of these aspects individually, but there is only one explanation for all of them together. Hypertrophic stimuli present in CHF have been shown to induce cardiomyocyte transition through the G1/S restriction point and entry into the S phase via D-type cyclins, suggesting a connection between adaptive hypertrophic growth and DNA synthesis.22 Accordingly, the decrease in cardiomyocyte DNA content in our study may be explained by fewer cardiomyocytes entering the S phase because of amelioration of hypertrophic stimuli by hemodynamic relief; however, this explanation does not account for the increase in diploid and binucleated cardiomyocytes. Tea and coworkers23 reported that during hypertrophy regression, on administration of antihypertensive drugs in spontaneously hypertensive rats, apoptosis occurred, although transiently, on a large scale 1 to 2 weeks after the antihypertensive medication was started. Some antihypertensive drugs were found to regulate changes in heart mass and cardiomyocyte DNA content independently of blood pressure reduction, so the trophic changes in the heart were not secondary to hemodynamic changes. A timely, limited, pronounced apoptosis may explain the decreased polyploid cardiomyocytes and could account for the (percent) increase in diploid cardiomyocytes. Increased apoptotic cardiomyocyte cell death could not be demonstrated in human hearts after ventricular unloading.24–27
The hypothesis we favor takes into account both the decrease in polyploidy and the increase in diploidy. We suggest that the vast polyploidy of cardiomyocytes in the failing heart is a result of hypertrophic growth associated with repeated rounds of DNA synthesis with an incomplete attempt to enter mitosis that has been blocked after completion of DNA replication. When hemodynamic stress is relieved by LVAD and the hypertrophic stimulation is decreased, one could hypothesize that this blockade is released, which may cause karyokinesis, resulting in the formation of binucleated cardiomyocytes, cytokinesis, and real cell duplication. The decrease in mean DNA content per cardiomyocyte (including binucleated cells) suggests not only that karyokinesis with a decreased DNA content per nucleus is taking place but also that cell division or progenitor cell proliferation occurs after unloading, resulting in a reduction in total DNA per cardiomyocyte. The simultaneous occurrence of karyokinesis and cytokinesis with cell division in the myocardium would plausibly explain the decrease in polyploid and the increase in diploid cardiomyocytes after unloading and would account for the increase in binucleated cardiomyocytes as readout of karyokinesis before cytokinesis.
Regardless of the underlying mechanism(s), our data further strengthen the notion that at least a proportion of cardiomyocytes are not terminally differentiated and thus are capable of reentering the cell cycle during regenerative processes. Our findings suggest a dynamic and plastic regulation of cardiomyocyte DNA content in heart failure by hemodynamic support. They also suggest that failing cardiomyocytes may accumulate polyploid DNA as a result of several rounds of DNA replication without consecutive mitosis because of a cardiomyocyte-specific cell cycle blockade. How relief of hemodynamic strain releases such a blockade and whether it alone may cause it are exciting new questions in the study of cardiac regeneration.
The mechanisms underlying this increase in diploid cardiomyocytes in unloaded hearts are unclear. The challenging view we put forward is that the decrease in DNA content after unloading may be due to a real completion of the cell cycle and/or karyokinesis with the potential of true cardiac regeneration. This may be an explanation for the still enigmatic existence of permanent cardiac recovery after LVAD weaning in some heart failure patients.11
The skillful technical assistance of Dorothe Möllmann, Antje Deichmann, Mareike Müller, Peter Babioch (University of Duisburg-Essen), and Elisabeth Schmidt-Brücken (University of Regensburg) is highly appreciated.
Source of Funding
This study was supported by the Deutsche Forschungsgemeinschaft (Ba 1730/9–1, BA 1730/10–1, Ba 1730/11–1 to Dr Baba).
Anversa P, Kajstura J. Ventricular myocytes are not terminally differentiated in the adult mammalian heart. Circ Res. 1998; 83: 1–14.
Meckert PC, Rivello HG, Vigliano C, Gonzalez P, Favaloro R, Laguens R. Endomitosis and polyploidization of myocardial cells in the periphery of human acute myocardial infarction. Cardiovasc Res. 2005; 67: 116–123.
Leri A, Kajstura J, Anversa P. Cardiac stem cells and mechanisms of myocardial regeneration. Physiol Rev. 2005; 85: 1373–1416.
Zafeiridis A, Jeevanandam V, Houser SR, Margulies KB. Regression of cellular hypertrophy after left ventricular assist device support. Circulation. 1998; 98: 656–662.
Wohlschlaeger J, Schmitz KJ, Schmid C, Schmid KW, Keul P, Takeda A, Weis S, Levkau B, Baba HA. Reverse remodeling following insertion of left ventricular assist devices (LVAD): a review of the morphological and molecular changes. Cardiovasc Res. 2005; 68: 376–386.
Engel FB, Hsieh PC, Lee RT, Keating MT. FGF1/p38 MAP kinase inhibitor therapy induces cardiomyocyte mitosis, reduces scarring, and rescues function after myocardial infarction. Proc Natl Acad Sci U S A. 2006; 103: 15546–15451.
Engel FB, Schebesta M, Duong MT, Lu G, Ren S, Madwed JB, Jiang H, Wang Y, Keating MT. p38 MAP kinase inhibition enables proliferation of adult mammalian cardiomyocytes. Genes Dev. 2005; 19: 1175–1187.
Woo YJ, Panlilio CM, Cheng RK, Liao GP, Atluri P, Hsu VM, Cohen JE, Chaudhry HW. Therapeutic delivery of cyclin A2 induces myocardial regeneration and enhances cardiac function in ischemic heart failure. Circulation. 2006; 114 (suppl): I-206–I-213.
Bergmann O, Bhardwaj RD, Bernard S, Zdunek S, Barnabe-Heider F, Walsh S, Zupicich J, Alkass K, Buchholz BA, Druid H, Jovinge S, Frisen J. Evidence for cardiomyocyte renewal in humans. Science. 2009; 324: 98–102.
Busk PK, Bartkova J, Strom CC, Wulf-Andersen L, Hinrichsen R, Christoffersen TE, Latella L, Bartek J, Haunso S, Sheikh SP. Involvement of cyclin D activity in left ventricle hypertrophy in vivo and in vitro. Cardiovasc Res. 2002; 56: 64–75.
Tea BS, Dam TV, Moreau P, Hamet P, deBlois D. Apoptosis during regression of cardiac hypertrophy in spontaneously hypertensive rats: temporal regulation and spatial heterogeneity. Hypertension. 1999; 34: 229–235.
de Jonge N, van Wichen DF, van KJ, Kirkels H, Lahpor JR, Gmelig-Meyling FH, van den Tweel JG, de Weger RA. Cardiomyocyte death in patients with end-stage heart failure before and after support with a left ventricular assist device: low incidence of apoptosis despite ubiquitous mediators. J Heart Lung Transplant. 2003; 22: 1028–1036.
Flesch M, Margulies KB, Mochmann HC, Engel D, Sivasubramanian N, Mann DL. Differential regulation of mitogen-activated protein kinases in the failing human heart in response to mechanical unloading. Circulation. 2001; 104: 2273–2276.
Baba HA, Stypmann J, Grabellus F, Kirchhof P, Sokoll A, Schafers M, Takeda A, Wilhelm MJ, Scheld HH, Takeda N, Breithardt G, Levkau B. Dynamic regulation of MEK/Erks and Akt/GSK-3beta 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.
Chronic heart failure remains a major cause of mortality and morbidity. Recent scientific work has focused on the possibility of manipulating cardiomyocytes for regeneration as a therapy for chronic heart failure. The issue of whether cardiomyocytes are terminally differentiated or able to regenerate during normal aging or in response to injury is still an object of intense debate. Evidence from animal and human studies shows that cardiomyocytes may reenter the cell cycle under certain conditions with subsequent mitotic cell division. It is known that after unloading by left ventricular assist devices, cardiomyocytes show numerous molecular changes, and in many instances, cardiac function can partially recover. This study demonstrates that substantial changes in the DNA content of cardiomyocytes occur after left ventricular unloading with a decrease in the number of polyploid and an increase in the number of diploid cardiomyocytes. These data indicate that DNA content is highly dynamic even in the dysfunctional cardiomyocytes of heart failure patients and suggest that nuclear division and cell division may be dissociated in this and other contexts. Further studies are necessary to explore the relationships between these findings and potential changes in cardiomyocyte numbers after left ventricular assist device. A better understanding of the complex interplay between nuclear and cardiomyocyte division in health and disease is essential for advances in cardiac regenerative medicine.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.108.808071/DC1.
↵*Drs Wohlschlaeger and Levkau contributed equally to this work.