Transgenic systems for unequivocal identification of cardiac myocyte nuclei and analysis of cardiomyocyte cell cycle status
- 2.2k Downloads
Even though the mammalian heart has been investigated for many years, there are still uncertainties in the fields of cardiac cell biology and regeneration with regard to exact fractions of cardiomyocytes (CMs) at different developmental stages, their plasticity after cardiac lesion and also their basal turnover rate. A main shortcoming is the accurate identification of CM and the demonstration of CM division. Therefore, an in vivo model taking advantage of a live reporter-based identification of CM nuclei and their cell cycle status is needed. In this technical report, we describe the generation and characterization of embryonic stem cells and transgenic mice expressing a fusion protein of human histone 2B and the red fluorescence protein mCherry under control of the CM specific αMHC promoter. This fluorescence label allows unequivocal identification and quantitation of CM nuclei and nuclearity in isolated cells and native tissue slices. In ventricles of adults, we determined a fraction of <20 % CMs and binucleation of 77–90 %, while in atria a CM fraction of 30 % and a binucleation index of 14 % were found. We combined this transgenic system with the CAG-eGFP-anillin transgene, which identifies cell division and established a novel screening assay for cell cycle-modifying substances in isolated, postnatal CMs. Our transgenic live reporter-based system enables reliable identification of CM nuclei and determination of CM fractions and nuclearity in heart tissue. In combination with CAG-eGFP-anillin-mice, the cell cycle status of CMs can be monitored in detail enabling screening for proliferation-inducing substances in vitro and in vivo.
KeywordsCardiac myocyte Cell cycle Genetically altered mice Proliferation miRNA
The mammalian heart is composed of different cell types such as endothelial cells, fibroblasts, smooth muscle cells, and cardiomyocytes (CMs). The latter provide the contractile properties of the heart muscle and CM loss in pathologies, such as myocardial infarction or cardiomyopathy, leads to functional impairment and eventually heart failure . While there is mutual consent that CMs make up the majority of the heart mass , there is a discrepancy regarding the fraction of CMs in the mammalian heart which is estimated to be composed of 15–35 % CMs [25, 37, 42], as addressed by histology but not specific cell markers . A recent report, using flow cytometry, claims that mouse hearts exceptionally consist of 55 % of CMs . Also identifying CM nuclei is challenging in cardiac sections in which CM cytoplasm is identified either by immunohistological staining for structural proteins or by transgenic CM-specific overexpression of fluorescence proteins. In those sections, non-CM nuclei in close proximity to CM cytoplasm cannot be distinguished from CM nuclei leading to false positives .
The majority of CMs are polyploid, caused either by an increase in nuclearity with ~90 % binuclear ventricular CMs in mouse [37, 43] and rat , and 25.5–61.9 % in humans [8, 29] or by increased nuclear ploidy up to 8N in humans  and 16N in mice . Polyploid cells are thought to be unable to undergo cell division, but are still able to enter the cell cycle and to perform DNA synthesis resulting in CMs with further increased ploidy [14, 23]. Heart regeneration is commonly analyzed with proliferation markers localized in the nucleus, such as PCNA, Ki-67, or pHH3 or Thymidine analogs such as BrdU or 3H-Thymidine. The results of these studies depend strongly on the unequivocal identification of CM nuclei, which is error-prone without the use of a specific nuclear CM marker . This could be one of the major reasons for the discrepancies concerning the estimates of annual turnover rates of adult mouse CMs, ranging from 0.76 %  up to 80 %  and the regeneration potential after experimental lesions, estimated from 0 % [15, 44] up to 15.3 % . Using a new in vivo proliferation marker, the eGFP-anillin system, we recently showed that adult CMs undergo endoreduplication (DNA replication without karyokinesis or cytokinesis) rather than cell division after cardiac lesion in mice .
Two challenges arise from the above-mentioned discrepancies, firstly the unambiguous identification of CM nuclei and secondly the confirmation of CM division and not mere cell cycle activity.
Here, we report the generation of a reporter system that directly visualizes CM nuclei in vitro and in vivo, by fusing the red fluorescence protein mCherry (mCh) to human histone 2B (H2B) and expressing it in CMs by use of the αMHC (Myh6) promoter. We demonstrate that this fusion protein accurately labels CM nuclei and allows the determination of CM portions and nuclearity during postnatal development of the mouse heart and after cardiac injury. We further combine αMHC-H2B-mCh mice with the CAG-eGFP-anillin mouse line for unambiguous visualization of cell cycle progression and division of postnatal CMs and for establishing a screening assay for cell cycle-modifying substances.
Generation of the αMHC-hH2BmCh-IRES-PuroR vector
The pIRES2-eGFP vector (CLONTECH Laboratories, Inc.) was digested with AseI and NheI, blunt-ended and religated. Thereby the CMV IE promoter was excised. EGFP was replaced by the PuromycinR (PuroR) cassette (HindIII-ClaI fragment) of the Cre-Pac vector. The resulting vector (pIRES2-PuroR) was opened with BglII, blunt-ended and subsequently digested with XhoI. The mouse αMHC promoter [5.5 kb BamHI(blunt)-SalI fragment] of the αMHC-pBK plasmid (kindly provided by J. Robbins, University of Cincinnati, Cincinnati, OH) was inserted into the opened pIRES2-PuroR-vector (pαMHC-IRES2-PuroR).
The pCAG-hH2B-mCh-IRES-PuroR expression vector was a gift from Heiko Lickert, Helmholtz Zentrum München. The vector was digested with NotI, blunt-ended, and cut with AccIII. The excised H2BmCh-IRES-Puro fragment was ligated into the EcoICRI-AccIII opened pαMHC-IRES-PuroR vector, thereby replacing the IRES2-sequence and part of the PuroR cDNA of the vector. The resulting vector αMHC-hH2B-mCh-IRES-PuroR (αMHC-H2B-mCh) was verified by restriction analysis and sequencing.
Generation and cultivation of transgenic ESC clones
G4 hybrid ESCs  were cultured in Knockout-Dulbecco’s modified Eagle’s medium (DMEM), high-glucose, supplemented with 15 % v/v fetal calf serum (FCS), 0.1 mM nonessential amino acids, 2 mg/ml l-Glutamine, 50 μg/ml each penicillin and streptomycin, 0.1 mM β-mercaptoethanol, and 500 U/ml Leucemia inhibitory factor (LIF). The ESCs were kept on irradiated neomycin-resistant mouse embryo fibroblasts (Millipore).
For generation of transgenic ESCs 5 × 106 cells were mixed with 30 µg of linearized plasmid DNA in PBS and electroporated at 250 V and 500 µF, 1 pulse, using a Bio-Rad Gene Pulser. The cells were plated on two 100-mm plates. Selection for neomycin-resistant cells started 2 days after electroporation by adding 165 µg/ml G418 to the medium. Resistant colonies were picked onto mouse embryonic fibroblast-coated 24-well plates, propagated and analyzed for mCh expression.
Differentiation of transgenic ESCs and purification of CMs
ESC differentiation was performed using mass culture. In brief, 2.0 × 106 ESCs were suspended in 10 ml differentiation medium (Iscove’s Modified Dulbecco’s Medium (IMDM), supplemented with 20 % FCS, 0.1 mM nonessential amino acids, 50 μg/ml each penicillin and streptomycin, 0.1 mM β-mercaptoethanol in the absence of LIF and cultivated on a horizontal shaker at 70–80 rpm, 37 °C, and 5 % CO2 for 2 days. Half of the EBs were further cultivated in a CELLSPIN 500 (INTEGRA biosciences) spinner flask. To select for CMs, puromycin treatment (0.5–1.5 µg/ml depending on the resistance of particular clone) was started on day 10 of differentiation in suspension culture. For isolation of CMs on d11 and d13, the EBs were dissociated with 0.125 % Trypsin/EDTA for 5 min. Treatment was stopped with differentiation medium and the cells were centrifuged (210 g, 5 min). ~105 cells were plated on coverslips, pre-treated with 10 µg/ml fibronectin in 24-well tissue plates, cultured for 48 h in differentiation medium, and for further purification of CMs puromycin was added. Cells were fixed with 4 % paraformaldehyde (PFA) at room temperature (RT) for 30 min.
Copy number determination by qPCR
The copy number of the H2B-mCh transgene was determined in different ESC clones applying a TaqMan-based qPCR assay as described by the manufacturer (Applied Biosystems). For the transgene detection, a Carboxyfluorescein (FAM)-coupled mCh probe (Applied Biosytems) was used and as genomic reference a VIC-coupled probe for the Transferrin receptor was chosen. Both probes were analyzed in triplicates for each gDNA (200 ng). QPCR was performed in a Rotor-Gene 6000 (Corbett).
Transgenic αMHC-H2BmCh-IRES-PuroR mice and myocardial infarction
All animal experiments were performed in accordance with the guidelines from Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes and were approved by the Animal Care Committees (8.87-50.10.31.08.199 and 8.87-50.10.37.09.238). Transgenic ESC clones were screened for a proper karyotype (40 chromosomes) and aggregated with diploid morula stage CD1 embryos as described previously . Ventricular cryolesions were generated in 10-week-old mice with 3.5-mm liquid-nitrogen-cooled copper probes as described earlier . Briefly, mice were anesthetized with 4 Vol% Isoflurane, 0.6 l/min O2, 0.4 l/min N2O, intubated and ventilated. The concentration of Isoflurane was reduced to 1–2 Vol%. The heart was exposed by thoracotomy. For local anesthesia, 0.1 ml of a 0.5 % Bupivacaine solution was injected into the intercostal musculature. Cryolesions at the anterolateral left ventricular wall were generated using liquid-nitrogen-cooled copper probes 3.5 mm in diameter. For ligation the left anterior descending (LAD) coronary artery was occluded directly distal of the left atrial auricle. For peri-operative analgesia, 20 mg/kg s.c Metamizol was used and 5 mg/kg s.c. Carprofen was used for analgesia post operation for 5 days. Neonatal mice were euthanized by decapitation and adult mice by cervical dislocation.
Left ventricular catheterization and hemodynamics
Left ventricular catheterization was performed on 15-week-old H2B-mCh and CD1 wt mice by a blinded investigator (for detail see ). Briefly, a 1.4 F Millar Aria1 catheter (Millar Instruments Inc., Houston, USA) was inserted under general anesthesia via the right carotid artery into the left ventricle, and pressure-volume (PV)-loops were recorded online. For analysis, Millar PVAN software was used. After the experiment, mice were sacrificed and hearts were harvested for histological analysis.
Dissociation of postnatal mouse hearts and miRNA transfection
Mice were sacrificed on d2/d3/d7 after birth; the hearts were prepared and analyzed for H2B-mCh expression by using a macroscope (AxioZoom.V16, Zeiss). Atria and ventricles of the transgenic hearts were separated and dissociated using the Neonatal heart dissociation kit (Miltenyi Biotec) without using the gentle MACS Dissociator. The heart tissue was dissected by slowly pipetting every 15 min. 7500 cells were seeded per well on a 0.001 % fibronectin-coated 384-well µ-clear microtiter plate (Greiner). Cells were incubated in differentiation medium (see above) at 37 °C and 5 % CO2. For analysis of the binucleation index, cells were fixated after overnight adhesion to the dish.
In order to transfect postnatal CMs, ventricles of αMHC-H2B-mCh/CAG-eGFP-anillin double transgenic hearts were dissociated as described above. 5 µl of 500 nM Stock solutions of hsa-miR-199a-3p or miR mimic, Negative control #1 (Ambion, Life Technologies) were incubated with 0.2 µl Lipofectamine RNAi Max (Invitrogen) in 14.8 µl OPTI-MEM on 0.001 % fibronectin-coated 384-well µ-clear microtiter plates (Greiner) for 20 min at RT. Afterward, 7500 dissociated cells were added in a volume of 60 µl differentiation medium. Medium was changed after 48 h and the cells were further incubated for 24 h.
Harvesting of mouse hearts
Heparinized adult mice were sacrificed by cervical dislocation and the hearts were harvested. After cannulation of the ascending aorta, the hearts were perfused with PBS, followed by perfusion with 4 % PFA, and further fixation overnight. Embryonic and postnatal hearts were excised and immersion fixated in 4 % PFA. Hearts were dehydrated in 20 % sucrose solution and mounted in O.C.T. compound.
Ventricular myocytes from adult mice were enzymatically isolated by Langendorff perfusion as previously described , with minor variations. Briefly, animals were killed by cervical dislocation. The heart was dissected and perfused using oxygenated Ca2+-free Tyrode solution (135 mM NaCl, 4 mM KCl, 1 mM MgCl2, 2.5 mM HEPES, 5 mM Glucose, pH 7.4) with 25 mM 2,3-Butanedione monoxime (BDM) for 5 min. The enzymatic digestion with 1 mg/ml (activity 0.199 U/mg) Collagenase B and 0.033 mg/ml Trypsin was performed in oxygenated Tyrode solution with 25 mM BDM and 50 µM CaCl2 for 8–12 min and after manually dissecting the tissue the reaction was stopped in stopping solution (oxygenated Tyrode solution with 25 mM BDM, 50 µM CaCl2 and 5 % FCS). The cell suspension was filtered through a 100-µm Cell strainer and centrifuged (700 rpm, 1 min). Isolated cells were plated on 8 µg/ml laminin-coated coverslips in 24-well plates in differentiation medium (see above) at 37 °C, 5 % CO2 for 5 h and fixated with 4 % PFA solution afterward.
Histology and immunofluorescence stainings
Following PFA fixation, all tissues and EBs were incubated in 20 % sucrose in PBS prior to cryopreservation in Tissue Tek O.C.T. compound (Sakura Finetek Europe B.V.). Sectioning of 10 µm/50 µm cryoslices was performed with a cryotome CM 3050S (Leica).
Fixated cells and tissue slices were stained for the following differentiation and proliferation markers (in 0.2 % Triton X in PBS, supplemented with 5 % donkey serum; 2 h at RT): α-actinin (1:400, Sigma–Aldrich A7811), cTropT (1:50, DSHB Hybridoma Product RV-C2), PCM-1 (1:200, Cell Signaling G2000), and Aurora B kinase (1:400, Sigma-Aldrich A5102). For the PCM-1 staining a short antigen-retrieval with 10 mM sodium citrate, pH 6.0, was performed in a KOS Microwave (Milestone, program: 4 min 70 °C, 3 min 85 °C, 4 min 90 °C, 5 min 93 °C), which allowed conservation of the H2B-mCh fluorescence. Primary antibodies were visualized by secondary antibodies conjugated to Cy5 (1:400, Jackson ImmunoResearch) diluted in 1 µg/ml Hoechst 33342 (nuclei staining) at RT for 1 h. Fluorescein Griffonia Simplificiosa Lectine (GSL, Vector Laboratories) was diluted 1:100 and stained at RT for 1 h. Immunostainings were documented with an inverted fluorescence microscope (Axiovert 200; Carl Zeiss MicroImaging, Inc.) equipped with a slider module (ApoTome; Carl Zeiss MicroImaging, Inc.). Sirius red, hematoxylin, and eosin stainings were performed using standard histology protocols. Pictures were acquired with an AxioStar Plus microscope (Zeiss).
For quantification of CM nuclei fractions in different regions of the heart, 10-µm thick transversal cryo-sections were analyzed (zone 1: section at atria level, zone 2: section through the middle of the ventricle, determined by length measurement of a macroscopic picture of each heart, zone 3: section in ~200 µm distance from the apex). The following 25× pictures of cross-sectioned CMs were analyzed per mouse (n = 3 per time point) and time points (P3, P7, 9 W): zone 1: left A atrium, LV left ventricle compact zone, LV trabecular zone, RV right ventricle compact zone, RV trabecular zone, zone 2: LV compact zone, LV trabecular zone, RV compact zone, RV trabecular zone, layer 3: LV compact zone, LV trabecular zone.
Determination of binuclearity and fractions of CM nuclei in thick slices
PFA-fixated, cryopreserved hearts were sliced into 50-µm thick sections with a cryotome CM 3050S (Leica) and treated with RNAse A (20 µg/ml) in wash buffer (0.5 M NaCl, 0.1 M Tris pH 7.5, 50 mM EDTA) for 1 h at 37 °C. Slices were incubated with 1 µM TO-PRO3 iodide (642/661) (Molecular Probes) and Fluorescein wheat germ agglutinine (WGA) (1:100, Vector Laboratories) at 4 °C overnight. Image stacks were acquired with an inverted confocal laser scanning microscope (Nikon Eclipse Ti) equipped with a ×40/1.15 NA water-dipping objective (ApoLWD 40× WI SDIC N2). Stacks with a z-step width of 0.5 µm were recorded. Excitation wavelengths for WGA, H2B-mCh, and TO-PRO 3 were 488, 543, and 642 nm, respectively. Binucleation was determined in z-stacks manually by scrolling through the different layers of the stack. Only cells that lay completely within the z-stack (WGA staining visible in every dimension) were quantified.
The number of CM nuclei (H2B-mCh+) and all nuclei (TO-PRO3+) was determined in 3D-reconstructions using the automatic 3D analysis module of NIS Elements. The result was verified by manually correcting for doublets, which were nuclei in very close proximity to each other, that were not correctly separated by the software. Virtual longitudinal and cross sections (0.5 µm) were generated using NIS Elements. Per thick slice ≥3 longitudinal sections with a distance of 10 µm to each other were manually analyzed for H2B-mCh signals and TO-PRO3 signals. In virtual cross sections, the distance between the analyzed sections (n ≥ 4 per thick slice) was 40 µm.
Nuclei isolation and flow cytometry
Nuclei isolation experiments were performed as reported earlier . In brief, frozen adult H2B-mCh hearts were cut into pieces manually and put in ice cold lysis buffer (in mM: 350 Sucrose; 5 CaCl2; 5 C4H6MgO4–4H2O; 2.0 EDTA; 0.5 EGTA; 10 Tris–HCl (pH 8.0); 1 DTT). The heart pieces were homogenized with a T-25 Ultra-Turrax® probe homogenizer (IKA) at 24 k rpm for 10 s and then using a glass dounce to further homogenize the tissue and to free the nuclei. The crude lysate was passed through a 100- and 70-μm nylon mesh cell strainer (BD Biosciences), consecutively, and then centrifuged at 700 g for 10 min in (4 °C). The nuclei pellet was re-suspended in sucrose buffer containing (in mM) the following: 2100 sucrose; 5 C4H6MgO4–4H2O; 1 DTT; 10 Tris–HCl (pH 8.0). By using gradient centrifugation, the sample was centrifuged at 13,000 g for 60 min in 4 °C. The nuclei pellet was re-suspended in nuclei storage buffer containing (in mM) the following: 440 sucrose; 70 KCl; 10 MgCl2; 10 Tris–HCl (pH 7.2); 1.5 spermine. Nuclei were stained with rabbit anti-mouse PCM-1 (Cell Signaling Tech.) overnight in 4 °C. Primary antibody was visualized with secondary antibody conjugated to FITC (Jackson ImmunoResearch) and DNA was stained with DAPI. Cytometric acquisition was performed on Aria III (BD Biosciences) using the 70-µm nozzle. Positive gates were set according to unstained controls, FMO controls, and single staining controls. Analysis was done using FlowJo™ Version V10.
Data are depicted as mean ± SEM. Statistical significance was determined by Student’s unpaired t test or 1way ANOVA with Bonferroni’s multiple comparison test. p < 0.05 was considered statistically significant.
Characterization of αMHC-hH2B-mCh-IRES-PuroR transgenic ESC clones
For generation of transgenic mice and in vitro analysis of H2B-mCh expression in CMs, we generated stably transfected mouse embryonic stem cell (mESC) lines and differentiated these into embryonic bodies (EBs). The expression of the reporter construct became visible around day 8 of differentiation (day 8). This correlated with the appearance of the first spontaneous beating cell clusters in the EBs and H2B-mCh expression was restricted to these areas (Fig. 1b, Suppl. video 1). Cryoslices of fixated EBs at day 10 revealed an exclusive expression of the H2B-mCh protein in the nuclei of α-actinin+ CMs (Fig. 1c, d). On day 10 puromycin treatment was started for 3 days to enrich for ESC-derived CMs. During puromycin treatment, EBs were dissociated and the cells replated (Fig. 1e), followed by quantification of H2B-mCh+/α-actinin+ CMs (Fig. 1f), which revealed a ~16-fold enrichment of CMs (Fig. 1g). The overlap of H2B-mCh expression and α-actinin staining was 96 ± 1.5 %, revealing high specificity of the transgene. In summary, the αMHC-H2B-mCh system can identify CM nuclei in vitro and CMs derived from the transgenic ESCs are viable and can be enriched by puromycin treatment.
H2B-mCh specifically marks CM nuclei in transgenic hearts
From a characterized αMHC-H2B-mCh ESC clone with only one transgene integration, we generated a transgenic mouse line by aggregation with diploid embryos from wildtype CD1 mice. The chimeric mice provided germ-line transmission and the transgenic progeny was viable, fertile, and had a normal lifespan.
As a definitive proof that H2B-mCh exclusively marks all CMs, we performed Langendorff dissociation of αMHC-H2B-mCh hearts (n = 6) to verify the specificity and the penetrance of our reporter system. All CMs with rod-shaped morphology and positive for α-actinin expressed H2B-mCh in their nuclei (Fig. 2d; n = 1610) proving complete penetrance, while nuclei from other cells were negative for H2B-mCh (Fig. 2d, arrows). The specificity to heart tissue was further demonstrated by analyzing sections from skeletal muscles such as M. tibialis anterior and diaphragm, which did not display any H2B-mCh expression (Fig. 2e).
We further investigated the specificity of the H2B-mCh expression pattern by staining with GSL to exclude expression in endothelial cells (Fig. 2f). There were several nuclei, which were covered by cross-striations and therefore seemingly belonged to CMs, but were clearly H2B-mCh negative (Fig. 2f, arrows) and vice versa (Fig. 2f, arrowheads). This underlines the difficulty to unequivocally identify CM nuclei on tissue sections.
Recent studies on cardiac turnover relied on the correct identification of CM nuclei by antibody staining for specific markers such as Troponin I and T  and PCM-1 , followed by flow cytometry analysis. To test, if our system can be used to identify CM nuclei out of total nuclei from hearts, we isolated cardiac nuclei from adult H2B-mCh mice and analyzed these by flow cytometry.
Hemodynamics of H2B-mCh hearts
Heart rate (bpm)
471.8 ± 42.6
499.7 ± 75.4
Maximum pressure (mmHg)
99.1 ± 7.9
99.6 ± 3.3
End-diastolic pressure (mmHg)
5.5 ± 5.9
4.9 ± 3.6
Stroke volume (µL)
25.5 ± 15.1
23.5 ± 7.1
Ejection fraction (%)
78.1 ± 12.4
69.4 ± 9.0
Cardiac output (µL/min)
11,787.5 ± 6295.5
11,413.9 ± 1549.7
dPdt max (mmHg/s)
9956.8 ± 1036.3
9425.9 ± 1445.6
dPdt min (mmHg/s)
−10,347.2 ± 1853.1
−10,052.9 ± 1290.4
Next, we analyzed transgene expression during embryonic development and neonatal stages. At E12.5, E15.5, and E18.5, the expression of H2B-mCh was limited mainly to the atria (Fig. 3d), probably due to the restricted activity of the αMHC promoter . Although there was some expression in the trabecular zone of the left and right ventricles (Fig. 3e), the H2B-mCh signals were only detectable in a subpopulation of CMs and expression was significantly weaker compared to atrial CMs.
After demonstrating the high specificity and complete penetrance of our transgene for CMs, we applied our system to several open biological questions concerning CM fractions, ploidy, and cell cycle status.
Determination of CM nuclei fractions at postnatal stages
One focus of our study was the precise determination of the CM nuclei fraction in different areas of the heart and at different developmental time points after birth, which is of particular relevance for studying cardiac regeneration. For our analysis, we have chosen an early postnatal time point at which the heart still has proliferative capacity (P3), a later postnatal time point at which the heart is just losing its potential to regenerate (P7) and adult age (9 weeks) when the heart is terminally differentiated. Figure 4a displays epifluorescence pictures of transgenic hearts at P0, P3, P7, and at adult stage [9 weeks (9 W)]. While the expression level of H2B-mCh increased in the ventricle during postnatal development, it did not change in atria (Fig. 4a, close-ups).
For quantification of CM nuclei fractions, we analyzed heart sections for H2B-mCh+ nuclei and total cardiac nuclei. We assumed that due to the occurrence of binucleated CMs, there would be a difference between quantification of CM nuclei based on either transverse or cross sections. In a cross section, only one of the two nuclei of a binucleated CM can be visible, while in longitudinal sections the chance to detect both nuclei is limited by section thickness. Therefore, both methods could be error-prone and either underestimate or overestimate the number of CMs. To determine the difference of counting transverse or cross sections on actual CM numbers, we first quantified the fraction of CM nuclei in volume 3D-reconstitutions of Z-stacks of thick (50 µm) slices from adult αMHC-H2B-mCh hearts without taking into account the degree of binucleation. Nuclei were stained with TO-PRO3 iodide, cell borders were visualized by WGA staining and CM nuclei were identified by nuclear H2B-mCh fluorescence (Fig. 4c). The effect of quantification of longitudinal or cross cardiac sections was determined by generating virtual sections from the 3D volume view (Fig. 4c). Interestingly, both methods of quantification gave very similar results compared to the fraction of CM nuclei obtained by analyzing the 3D reconstitution (Fig. 4d).
Determination of CM nuclearity and CM fractions at postnatal stages
To accurately determine the ratio of mononucleated to bi- or multinucleated CMs, we followed two different approaches. The number of nuclei per CM was counted in enzymatically dissociated transgenic hearts at P3, P7, and in Langendorff-dissociated hearts at 9 W. The quantification of isolated CMs revealed a strongly increased binucleation index in ventricular CMs from 11.6 ± 1.6 % at P3, to 65.5 ± 1.7 % at P7, and to 89.5 ± 2.0 % at 9 W (Fig. 6b). In atrial CMs, we determined only a slight increase in binuclearity from P3 (14.2 ± 2.2 %) to P7 (32.0 ± 1.2 %). The fraction of multinucleated CMs exceeding binucleation was <1 % in all cases. In parallel, confocal images of thick slices (50 µm) of transgenic hearts at 9 W were analyzed as Z-stacks to exclude a potential selective loss of mono- or binucleated CMs during the Langendorff isolation procedure. The slices were stained with WGA to visualize cell borders (Fig. 6c). Only CMs that lay completely within the z-stack (WGA staining visible in every dimension) were analyzed.
The quantifications of Z-stacks at adult stage allowed a separate analysis of the binucleation index in cells from the left (77.2 ± 4.4 %; n = 5) and right ventricle (77.8 ± 3.4 %; n = 5) and the atrium (14.4 ± 4.3 %; n = 4) (Fig. 6b), thereby supporting the results obtained by the analysis of the Langendorff dissociated CMs. The fraction of mononucleated cells in the ventricles, as determined by this method, was significantly higher (p = 0.034) than with enzymatic dissociation.
By taking into account the portion of binucleated CMs, as assessed by Z-stack analysis, enzymatic cell separation and Langendorff dissociation, we were able to calculate the portions of CM cells in atria and ventricles at the selected time points (Fig. 6d). Due to the low degree of binucleation in atria, the fraction of CMs in this compartment was significantly higher at all analyzed stages (P3: 48.7 ± 1.4 %; P7: 36.1 ± 3.9 %; 9 W: 30.1 ± 1.6 %) compared to ventricles at P3 (39.2 ± 2.6 %), P7 (25.8 ± 0.1 %), 9 W (16.1 ± 0.2 % dissociation) and LV (18.1 ± 0.6 % thick slices) and RV (17.3 ± 0.6 % thick slices), respectively.
We also explored, whether H2B-mCh expression correlates to DNA content and could therefore be used as a direct readout for endoreduplication. However, neither Hoechst staining of thick sections of the LV (Fig. 6e) nor isolation of CM nuclei followed by nuclear staining and flow cytometry (Fig. 6f–h) yielded a direct correlation between H2B-mCh signal intensity and DNA content.
Easy identification of the border zone after cryoinjury of the heart
The expression of the fusion protein further facilitates the identification of nuclei from border zone CMs. Especially in cross sections, the reliable identification of CM nuclei is limited without the use of a specific nuclear marker as the CM nuclei are not always centered within the cells (Fig. 7e, arrows). Similarly, individual CMs or small islands of surviving CMs, which are hard to detect within the lesioned area, are easily detectable by their H2B-mCh nuclear signal (Fig. 7f, arrows).
In summary, the αMHC-H2B-mCh mouse line facilitates the reliable identification of CM nuclei after myocardial infarction and eases the estimation of the lesion size.
An in vitro assay for screening for proliferation-inducing substances in CMs
We have generated a novel transgenic model for the direct visualization of CM nuclei in vitro and in vivo. This system is based on the expression of a fusion protein consisting of the human histone 2B and the red fluorescent protein mCh under transcriptional control of the CM specific Myh6 promoter, resulting in a nuclear signal. We showed the specific labeling of CM nuclei throughout our in vitro and in vivo experiments.
As the mammalian heart consists of mono- and binucleated CMs, it was of great interest to determine the fraction and nuclearity of CMs in different regions of the mouse heart. The contribution of CMs to total number of heart cells in adult humans and rodents has been estimated to be 15–20 % [37, 42]. However, for mice, even substantially higher estimates have been reported, assigning an exceptional role for this species . We can clearly exclude this and confirm earlier estimates of <15 % CMs, which correspond to <30 % CM nuclei [37, 43]. While previous quantifications were based on cell suspensions of dissociated hearts, our data are based on (1) flow cytometry of isolated nuclei, (2) quantification on tissue sections, (3) Langendorff isolation, as well as (4) 3D-reconstruction and (5) Z-stack analysis of cardiac slices. Cell isolation procedures in general cannot guarantee uniform extraction of all cell types. Cells that are more sensitive to enzymatic treatment are prone to be damaged, while cells surrounded by connective tissue are less likely to be dissolved from the tissue. By using Z-stacks of cardiac sections for quantitation, we found a significantly lower fraction of binuclear CMs as with enzymatic dissociation. A recent study in which histological sections were used also reported higher numbers of mononuclear CMs (24 % ) that match our observations of ~25 %. In atria of adult mice, we found an even higher fraction of mononucleated cells of ~86 %. The functional difference between mono- and binucleated CMs is unknown. However, there is evidence that mononucleated CMs are prone to enter the cell cycle after stimulation with neuregulin  or after cardiac lesion . Interestingly, atrial mononuclear CMs display cell cycle activity after experimental lesion in the left ventricle in rats, which was attributed to either cell division  or increased binucleation .
Additionally, we also addressed the issue of CM fractions in the context of cardiac growth and spatial distribution in the heart, which is of relevance for studying cardiac regeneration as it has been shown that neonate mammalian hearts have a remarkable regenerative capacity . We found a significantly higher proportion of CMs at P3 compared to P7 and adult mice which is most likely due to the extensive proliferation of fibroblasts taking place from P1 to P10 . The spatial distribution of CMs revealed only a difference in atria of P3, P7, and adult mice, in which the fraction of CMs was significantly higher compared to the ventricles. This is due to the significantly lower fraction of binucleated cells in atria of mouse, which was ~14 % in adults.
A promising strategy for regeneration of lost CMs is the re-induction of CM division by application of proliferation-inducing substances. However to identify such factors and to verify them in vivo, CM nuclei and cell division need to be unequivocally identified.
The advantages of our CM proliferation assessment assay, which is based on H2B-mCh/eGFP-anillin compound transgenic mice, are the immediate CM nuclei identification and quantification of cell cycle activity due to the fluorescent labels and in particular the possibility to live-monitor cell cycle progression of CMs. In contrast, most commonly used assays use BrdU and stainings with combinations of Ki-67, pHH3, PCNA, and AuroraB-kinase [7, 9, 10, 41] which require fixation and prevent the direct follow-up of cell fate. As a proof of concept, we were able to verify the proliferation-inducing effect of miR-199, as previously published . Due to the high resolution of M-phase, the eGFP-anillin assay also allows us to discriminate between cell division and cell cycle variations such as endoreduplication and acytokinetic mitosis leading to binucleation . Surprisingly, treatment with miR-199 led to an increase of binucleated CMs, indicating induction of cell cycle variations. In summary, the H2B-mCh/eGFP-anillin assay will provide crucial information about the true nature of cell cycle activity in the future.
Currently, identification of CMs is mostly based on staining of the contractile apparatus (cTropI, cTropT or α-actinin) or on the use of transgenic mouse models such as αMHC-mCh mice  or αMHC-Cre inducible mice  which provide CM-specific cytoplasmic expression of fluorescence reporters . As a caveat, these techniques do not allow a direct correlation of nuclei to the marked CMs. False positive identification of a non-CM nucleus as cell cycle active CM nucleus may strongly alter a conclusion concerning the regeneration potential of adult CMs as the numbers of CMs with cell cycle activity are described to be very low (1 in 180,000) . Indeed, a lack of proper identification of CM nuclei seems to be one of the major reasons for the high divergences in literature concerning the annual turnover rate of adult mouse CMs, ranging from 0.76 %  up to 80 % .
Currently only one further transgenic mouse model, MHC-nLAC, has been described which enables a specific identification of CM nuclei by staining for β-galactosidase activity . As a shortcoming, the nuclear signal is replaced by a diffuse cytoplasmic signal in CMs as soon as the nuclear envelope dissolves. This complicates quantifications of CM nuclei if cells enter M-Phase. The H2B-mCh mice described here have two advantages over the existing mouse lines, which use NLS or cytoplasmic expression of a CM specific marker: first, compared to lacZ as a marker  no additional staining step is needed, which is particulary important for live-imaging applications and second, due to the fusion to histone H2B, mCh stays connected to the chromosomes during all phases of the cell cycle, a prerequisite for automatic cell tracking in a screening assay . Furthermore, our mouse model can also be used to directly and specifically isolate CM nuclei for the analysis of epigenetic chromatin modifications, an approach recently established in CMs . Taken together, our transgenic system provides a new technology for the identification of CM nuclei without any further treatment and for the isolation of CM nuclei for downstream applications. In combination with the eGFP-anillin system, it is a novel tool to specifically resolve controversial questions in the cardiac field, such as the ongoing debate, if adult CMs have the potential to divide. This mouse line will be useful for assessing proliferation in models of cardiac regeneration, for the differentiation between atrial and ventricular CMs during embryonic development, identification of CMs in screening assays for proliferation-inducing substances, transplantation experiments, and for the analysis of plasticity of adult CMs in future studies.
We thank A. Nagy (Toronto, Canada) and M. Gertsenstein (Toronto, Canada) for providing the G4 mouse ESC line. We also would like to acknowledge S. Grünberg (Bonn, Germany), P. Freitag (Bonn, Germany), and J. Schmidt (Bonn, Germany) for technical assistance. This work was supported by the Deutsche Forschungsgemeinschaft [FOR 1352, FL276/6-1 to B.K.F., RO2279/2-2 to W.R] and the NRW International Graduate School BIOTECH-PHARMA [fellowship to A.R.].
Conflict of interest
Supplementary material 1 (AVI 8393 kb)
Supplementary material 2 (AVI 1906 kb)
Supplementary material 3 (AVI 11048 kb)
Supplementary material 4 (AVI 11207 kb)
- 13.Gilsbach R, Preissl S, Gruning BA, Schnick T, Burger L, Benes V, Wurch A, Bonisch U, Gunther S, Backofen R, Fleischmann BK, Schubeler D, Hein L (2014) Dynamic DNA methylation orchestrates cardiomyocyte development, maturation and disease. Nat Commun 5:5288. doi: 10.1038/ncomms6288 CrossRefPubMedCentralPubMedGoogle Scholar
- 15.Hesse M, Raulf A, Pilz GA, Haberlandt C, Klein AM, Jabs R, Zaehres H, Fugemann CJ, Zimmermann K, Trebicka J, Welz A, Pfeifer A, Roll W, Kotlikoff MI, Steinhauser C, Gotz M, Scholer HR, Fleischmann BK (2012) Direct visualization of cell division using high-resolution imaging of M-phase of the cell cycle. Nat Commun 3:1076. doi: 10.1038/ncomms2089 CrossRefPubMedCentralPubMedGoogle Scholar
- 17.Hosoda T, D’Amario D, Cabral-Da-Silva MC, Zheng H, Padin-Iruegas ME, Ogorek B, Ferreira-Martins J, Yasuzawa-Amano S, Amano K, Ide-Iwata N, Cheng W, Rota M, Urbanek K, Kajstura J, Anversa P, Leri A (2009) Clonality of mouse and human cardiomyogenesis in vivo. Proc Natl Acad Sci USA 106:17169–17174. doi: 10.1073/pnas.0903089106 CrossRefPubMedCentralPubMedGoogle Scholar
- 19.Jesty SA, Steffey MA, Lee FK, Breitbach M, Hesse M, Reining S, Lee JC, Doran RM, Nikitin AY, Fleischmann BK, Kotlikoff MI (2012) c-kit+ precursors support postinfarction myogenesis in the neonatal, but not adult, heart. Proc Natl Acad Sci USA 109:13380–13385. doi: 10.1073/pnas.1208114109 CrossRefPubMedCentralPubMedGoogle Scholar
- 24.Mollova M, Bersell K, Walsh S, Savla J, Das LT, Park SY, Silberstein LE, Dos Remedios CG, Graham D, Colan S, Kuhn B (2013) Cardiomyocyte proliferation contributes to heart growth in young humans. Proc Natl Acad Sci USA 110:1446–1451. doi: 10.1073/pnas.1214608110 CrossRefPubMedCentralPubMedGoogle Scholar
- 29.Olivetti G, Cigola E, Maestri R, Corradi D, Lagrasta C, Gambert SR, Anversa P (1996) Aging, cardiac hypertrophy and ischemic cardiomyopathy do not affect the proportion of mononucleated and multinucleated myocytes in the human heart. J Mol Cell Cardiol 28:1463–1477. doi: 10.1006/jmcc.1996.0137 CrossRefPubMedGoogle Scholar
- 34.Sohal DS, Nghiem M, Crackower MA, Witt SA, Kimball TR, Tymitz KM, Penninger JM, Molkentin JD (2001) Temporally regulated and tissue-specific gene manipulations in the adult and embryonic heart using a tamoxifen-inducible Cre protein. Circ Res 89:20–25. doi: 10.1161/hh1301.092687 CrossRefPubMedGoogle Scholar
- 41.Uosaki H, Magadum A, Seo K, Fukushima H, Takeuchi A, Nakagawa Y, Moyes KW, Narazaki G, Kuwahara K, Laflamme M, Matsuoka S, Nakatsuji N, Nakao K, Kwon C, Kass DA, Engel FB, Yamashita JK (2013) Identification of chemicals inducing cardiomyocyte proliferation in developmental stage-specific manner with pluripotent stem cells. Circ Cardiovasc Genet 6:624–633. doi: 10.1161/CIRCGENETICS.113.000330 CrossRefPubMedCentralPubMedGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.