| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The Sohnis Family Research Laboratory for the Regeneration of Functional Myocardium and the Rappaport Family Institute for Research in the Medical Sciences, The Bruce Rappaport Faculty of Medicine, Technion- Israel Institute of Technology, Haifa, Israel
1Correspondence: The Bruce Rappaport Faculty of Medicine, Technion- Israel Institute of Technology, 2 Efron St., P.O.B. 9649, 31096 Haifa, Israel, E-mail: mdlior{at}tx.technion.ac.il
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: myogenesis • myosin light chain 2-V • promoter • selectable marker • embryoid body • cardiac differentiation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Myocardial cell replacement therapy is emerging as a novel therapeutic paradigm for myocardial repair (2
3
4
5)
but is hampered by the paucity of cell sources for human cardiomyocytes. The recently described human embryonic stem cell (hESC) lines may provide a possible solution for this cell sourcing problem (6
, 7)
. These unique pluripotent stem cell lines, derived from the inner cell mass of human blastocysts, can be propagated continuously in culture in the undifferentiated state and coaxed to differentiate into a variety of cell lineages (6
7
8
9
10
11
12
13
14
15
16)
.
Among the different cell types generated during hESC differentiation, cardiomyocyte induction is demonstrated by the appearance of spontaneously contracting areas in three-dimensional differentiating cell aggregates termed embryoid bodies; EBs (15
, 17
, 18)
. Cells isolated from these beating areas displayed molecular, structural, and functional properties of early stage cardiomyocytes (15
16
17
18
19)
. More recently, we demonstrated that the hESC derived cardiomyocytes can form a functional syncytium (20)
and can integrate structurally and functionally with preexisting cardiac tissue both in vitro (in coculturing studies) and in vivo (by serving as a biological pacemaker in animal models of slow heart rate) (21
, 22)
.
Despite these initial encouraging results, application of hESC in cardiovascular regenerative medicine and in several other cardiac research areas is limited by the inability to identify and derive pure populations of differentiated cardiomyocytes (23)
. Given the heterogeneous cell mixture within the EBs, derivation of a homogenous cardiac cell populations will ultimately depend on specificity of the cell selection process. Although mechanical dissection of the contracting areas (20)
and percoll gradient centrifugation (17)
has been used to enrich cardiomyocyte selection, the degree of purity that is achieved will probably be insufficient for clinical or research purposes.
In the current study we describe the generation of transgenic hESC lines that allow identification and selection of differentiating cardiomyocytes. This approach is based on using a cardiac-specific promoter to drive the expression of a selectable marker or reporter gene. Using the human myosin light chain 2v (MLC-2v) promoter we generated stable transgenic hESC lines in which eGFP-expressing cells, appearing during in vitro differentiation, can be identified and sorted. The eGFP-expressing cells displayed structural, molecular, and functional properties of early stage human cardiomyocytes and could form stable cell grafts following in vivo myocardial transplantation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 25)
. The second transcriptional unit contained the PGK promoter driving the expression of aminoglycoside phosphotransferase (PGK-NeoR), which was meant to allow selection of the transfected undifferntiaed hESC cells. Eventually we did not use this antibiotic selection strategy because, in our preliminary studies, we encountered difficulties in titrating the antibiotic dose (relatively high numbers of false positive clones were found), while the alternative strategy of generating single-cell clones (followed by PCR analysis of the generated clones) provided superior results mostly due to the relatively high transduction efficacy of the lentiviral vector. The PGK-NeoR cDNA was subcloned from pMSCV Neo (Clontech). The MLC-2v-HygR–eGFP and PGK-NeoR fragments were then subcloned to the lentivirus vector pTK113, using BamHI and XhoI restriction sites.
Establishment of the transgenic hESC lines
To generate lentivirus particles, human embryonic kidney (HEK) 293T cells were transfected with 15, 10, and 5 µg of the lentivirus vector, the packaging cassette expression plasmid (
NRF), and the VSV-G envelope expression plasmid, respectively. HEK 293T cells were transfected using the conventional calcium phosphate transient transfection method. To collect the virus particles, the HEK 293T cell media was harvested 55 h after transfection and centrifuged at 2000 rpm for 7 min. Supernatant was filtered through a 45 µm filter and then concentrated using Vivaspin (membrane cut-off 100,000; Vivascience, Goettingen, Germany). The concentrated virus particles containing media were supplemented with 6 µg/ml polybrene and added to the undifferentiated hESC culture medium (as clumps of 
200 cells obtained at the time of routine passage using mechanical and enzymatic dissociation). The hESC clone H9.2 (26)
(passage 40) was used in the current study. After additional 16 h, the virus particles were collected again and hESC were infected again as described above. The transduced hESC clumps were then washed briefly with PBS and replated on a fresh mouse feeder layer at a concentration of one small single colony per well. These transgenic colonies were continuously cultured, and the lines that demonstrating robust, stable, long-term, and homogenous expression of the transgene were propagated.
Creation of single-cell transgenic hESC clones
The transgenic hESC colonies established were digested using 0.25% trypsin-EDTA solution (Biological Industries, Kibbutz Beit Haamek, Israel) for 10 min. Cells were then counted and diluted to give 1 cell/ml and plated in MEF-covered 24 wells plates at a concentration of a single cell/well. The single-cell derived clones were grown as described above, and the ones demonstrating robust, stable, long-term, and homogenous expression of the transgene were chosen for propagation.
Pluripotent properties of the transgenic lines
To verify that the transgenic hESC lines and clones generated retain the unique properties of the parental hESC lines, the following studies were performed.
Staining for specific hESC markers
Colonies of undifferentiated stem cells derived from the transgenic lines were stained for specific hESC markers (SSEA-4, Oct-4, Tra-1–60) as described below.
Teratoma formation
Undifferentiated hESC were harvested using 1 mg/ml collagenase IV (Life Technologies, Grand Island, NY, USA) and injected into the hind limbs of severe combined immunodeficient SCID/beige mice (5x106 cells per injection). The teratomas were palpable after 6–7 wk and were harvested for histological examination.
hESC propagation and in vitro cardiomyocyte differentiation
Undifferentiated hESC were grown on a mitotically inactivated mouse embryonic fibroblast feeder layer (MEF) as described previously (15
, 26)
. The culture medium consisted of 20% FBS (HyClone, Logan, UT, USA), 80% knockout DMEM supplemented with 1 mM L-glutamine, 0.1 mM mercaptoethanol, and 1% nonessential amino acids (all from Life Technologies). To induce differentiation, hESC were dispersed to small clamps (3–20 cells) using collagenase IV (1 mg/mL, Life Technologies) and were transferred to plastic Petri dishes a cell density of 5 x 106 cells in a 58 mm dish, where they were cultured in suspension for 7–10 days. During this stage the cells aggregated to form EBs, which were then plated on 0.1% gelatin-coated 24-well plates, at a density of 7 EBs per well and observed for the appearance of spontaneous contractions.
The contracting areas within the EBs, generated during differentiation, were identified by light microscopy; their presence and location were compared with the spatial distribution of eGFP expression using epifluorescent microscopy by two independent investigators. The eGFP expressing areas were then mechanically dissected for the phenotypic characterization studies described below. In some of these studies, the EBs were dispersed to single cells by enzymatic dissociation as described previously (19)
.
FACS sorting
EBs were digested using 0.25% trypsin-EDTA solution (Biological Industries) for 10 min. Due to the need for a relatively large number of eGFP cells for the FACS sorting studies, we mainly took EBs from wells showing an increased rate of contracting EBs. In other studies, aiming to examine the entire population of eGFP-expressing cells, the entire population of differentiating EBs were used. Cells were then resuspended in the culture medium in a concentration of 10 cells/ml. Flow cytometric analysis was performed using a FACS sorter (Becton Dickinson Immunocytometry Systems, Franklin Lakes, NJ, USA). A 530/30 nm bandpass filter was used to measure EGFP fluorescence intensity excited with the 488 nm line of an argon ion laser. Detector settings were calibrated with untransfected hESC-derived EBs that were digested by the same method. The FACS sorted cells were plated on gelatin coated 24-wells culture plates at a density of 105 cells/well.
Immunostaining
Cells or whole EBs were fixed using 4% paraformaldehyde with sucrose, washed with PBS, and permabilized with 1% Triton-X-100. In the in vivo studies, the hearts were harvested, frozen in liquid nitrogen, and cryo-sectioned. Cells were then blocked with 5% normal goat serum and incubated overnight at 4°C with primary antibodies for cardiac troponin I (cTnI), connexin-43 (Cx43), human mitochondria (all from Chemicon, Temecula, CA, USA), sarcomeric 
-actinin (Sigma, St. Louis, MO, USA), MLC-2v, Oct-4 (both from Santa Cruz Biotechnology, Santa Cruz, CA, USA), SSEA-4, and Tra-1–60 (both from Chemicon). Secondary antibodies were Cy3 and Cy2 conjugated donkey and goat anti-mouse IgG antibodies and Cy3 and Cy5 conjugated anti-rabbit IgG antibodies (Jackson, Westgrove, PA, USA) at dilution 1:200 for 1 h at room temperature. Nuclei were counterstained by ToPro3 (Molecular Probes, Eugene, OR, USA) or DAPI (Sigma). To overcome the possible problem of autofluorescence artifact at the injection site, we performed additional immunostainings using polyclonal antibodies for eGFP (1:100) (MBL) in a similar manner to the protocol described above. In these studies, we have also used a number of secondary antibodies (with different excitation/emission spectra) as well as performed negative control stainings, in which the secondary antibodies were added without the primary antibody, in an attempt to assess the degree of autofluorescence. Confocal microscopy was performed using a Nikon (Tokyo, Japan) Eclipse E600 microscope and Bio-Rad (Hercules, CA, USA) Radiance 2000 scanning system.
RT-PCR analysis
Total RNA was isolated from undifferentiated hESC, unfractionated dispersed cells derived from the differentiating EBs, FACS-sorted eGFP-expressing cells, and the nonsorted cells using the high-pure RNA isolation kit (Roche, Mannheim, Germany). cDNA was synthesized using access RT-PCR introductory system (Promega) and subjected to PCR with primers for cardiac specific genes, pluripotent markers (Oct4), endodermal (-fetoprotein) and ectodermal (beta-III-tubulin) markers (Table 1
).
|
Patch-clamp studies
For single-cell action-potential analysis, the whole-cell configuration of the patch-clamp technique was used as described previously (19)
. After dissociation with collagenase B (1 mg/mL, Roche), cells were replated for 1–3 days on gelatin-coated glass coverslips. The patch pipette solution consisted of (in mM): 120 KCl, 1 MgCl2, 3 Mg-ATP, 10 HEPES, 10 EGTA, pH 7.3. The bath recording solution consisted of (in mM): 140 NaCl, 5.4 KCl, 1.8 CaCl2, 1 MgCl2, 10 HEPES, 10 glucose, pH 7.4. On seal formation and following patch-break analog capacitance compensation was used. Axopatch 200B, Digidata1322, and pClamp8 (Axon, Burlingame, CA, USA) were used for data amplification, acquisition, and analysis. A cardiac phenotype was assigned to the examined cells if it displayed cardiac action potential or ionic currents in the current-clamp or voltage-clamp modes, respectively. A total of 33 eGFP-cells was studied.
Multielectrode array recordings
The electrophysiological properties of the eGFP-expressing cell-clusters were examined using a microelectrode array (MEA) data acquisition system (Multichannel Systems, Reutlingen, Germany) as described previously (20
, 27)
. The MEA plates consist of a matrix of 60 electrodes with an interelectrode distance of 100 or 200 µm allowing simultaneous recording of the extracellular potentials at a sampling rate of 10 KHz. All recordings were performed at 37°C and a pH of 7.4. Local activation time (LAT) at each electrode was determined by the timing of the maximal negative intrinsic deflection (dV/dtmin). This information was then used for the generation of color-coded activation maps by interpolating the LAT values between the electrodes using MATLAB standard two-dimensional plotting function.
Myocardial engraftment of the eGFP-expressing cells
All animal experiments were approved by the Animal Board and Safety Committee of the Technion’s Faculty of Medicine. For the transplantation studies, the eGFP-expressing areas within the differentiating EBs (20–40 days of differentiation) were carefully microdissected with a curved 23G needle and were then dissociated into small cell clusters (20–100 cells) by incubation with 1mg/ml of collagenase B (Roche) for 45 min. From our preliminary studies, this protocol results in the best survival rate of the grafted cells. Male Sprague-Dawley rats weighing 200–250 g were then anesthetized using a ketamine/xylasin preparation and mechanically ventilated with a Harvard small-animal mechanical respirator. Through a left thoracotomy, the eGFP-expressing cell clusters were then grafted to a left ventricular site using a 28 g needle. The cells were suspended prior to injection in 300 µl serum-free media. Following the procedure, the animals were treated by daily injections of cyclosporine-A (10 mg/kg) and methylprednisolone (2 mg/kg) to prevent immune rejection. Three days and four weeks following cell grafting, the hearts were harvested for pathological examination.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Five stable transgenic hESC lines were generated using the aforementioned vector. PCR analysis of selected colonies confirmed the presence of both transcriptional units in the transfected lines (data not shown). The transgenic hESC lines were analyzed and compared to the parental lines from which they were derived. We found no differences in their immunostaining results for the presence of typical undifferentiated hESC markers (Oct-4, SSEA-4, Tra-1–60) and their ability to form teratomas when injected into SCID mice (Fig. 1
).
|
We next sought to determine whether the established transgenic hESC lines can be utilized to identify and select for the differentiating cardiomyocytes. Following selection and expansion of the transfected undifferentiated colonies, the hESC were allowed to differentiate using the EB differentiating system as described previously (15)
. Thus, following 7–10 days of cultivation in suspension, the EBs were plated on gelatin-coated culture plates and observed using light and epifluorescent microscopy (by two independent investigators) for the appearance of spontaneous contraction and eGFP expression respectively.
Stable transfection of the hESC lines did not seem to affect their cardiomyocyte differentiating capabilities with 10% of the EBs showing spontaneous contracting areas (in a similar manner to the wild-type lines). An excellent spatial correlation was noted between the location of the beating areas in the contracting EBs and the presence of eGFP expression (Fig. 2
A, Supplement Movie). Hence, 91% of the 899 EBs that were scanned and demonstrated to contain contracting zones in the MLC-2V-transgenic lines also displayed positive eGFP fluorescence in the same regions (ranging from 67% to 98% in the 5 different MLC-2V lines). More importantly, prospective analysis of 129 EBs showing condensed eGFP fluorescence demonstrated that 98% also displayed spontaneous beating in the same areas. Importantly, eGFP expression could also be noted and maintained in differentiating EBs (undergoing long-term culturing) that were derived from transgenic hESC lines that have undergone multiple passages (31 passages, the longest period analyzed). This indicates the lack of promoter shutdown (which may be a significant limiting factor in the genetic modification of hESC) in the transgenic lines that were chosen for propagation. Interestingly, we could begin to detect the expression of eGFP at around the time (or 1–2 days prior to) the initiation of spontaneous contractions and eGFP expression was continuous for up to 50 days post-plating (the longest period studied).
|
The variability observed between the different hESC transgenic lines generated may result from the fact that the established lines were not truly single-cell clones but rather were derived from individually transfected colonies and, hence, not all cells in the differentiating EBs may show the same level of transgene expression. To test this hypothesis, we continued to create single-cell clones from one of the transgenic MLC-2V lines. A total of 14 single-cell clones were established, of which 3 were continuously propagated (those that displayed continuous, homogeneous, and robust eGFP fluorescence during EB differentiation).
The single-cell clones were characterized by the same unique undifferentiated properties, pluripotency, and capacity to differentiate into cardiomyocytes as the parental lines. Yet, they were also characterized by a more homogeneous expression of the transgene during in vitro EB differentiation. This was manifested by an increase in the percentage of beating EBs showing eGFP fluorescence (100% in the single-cell clones vs. 67% to 98% in the regular transgenic lines). Similarly, the pattern and level (intensity) of eGFP fluorescence within a single beating area was more homogeneous in the EBs generated from the single-cell clones when compared to the parental transgenic hESC line (compare Figs. 2C-D to A-B
).
Characterization of the eGFP expressing cells
Immunostaining studies of the EBs (Fig. 2A
) demonstrated that the eGFP-expressing cells were positively stained for cardiac-specific markers. Note, for example, in Fig. 2A
at both low (middle tracing) and high (lower tracing) magnifications the cells expressing eGFP within the EBs were also stained positively with anticTnI antibodies and that these cells demonstrated an early striated pattern, which is typical of early stage cardiomyocytes. The cardiac specificity of the MLC-2v promoter-driven eGFP expression could be more clearly demonstrated in immunocytostaining studies of dispersed cells, isolated from the beating areas (Fig. 2B
). Quantitative analysis of these colocalization immunocytostaining experiments (Table 2
) demonstrated that 91% (429/474, n=5) of the eGFP-expressing cells were also stained positively for cardiac-specific markers (using antisarcomeric -actinin antibodies). In contrast, 14% of the alpha-actinin positive cells were eGFP negative
|
The EBs derived from the single-cell clones were characterized by a more intense and homogeneous eGFP expression than the parental transgenic hESC lines (Figs. 2C-D
). The eGFP-expressing cells derived during the differentiation of these single-cell clones were stained positive for cardiac-specific markers (cTnI and MHC) either as dispersed cells (Fig. 2C
) or as whole EB immunostaining studies (Fig. 2D
).
Our next step was to develop strategies for the selection of the eGFP-expressing cells. To this end, we used FACS sorting of dispersed cells (obtained using enzymatic dissociation of the differentiating EBs) to select for the eGFP-expressing cells (Fig. 3
). These studies also demonstrated a significant improvement in the single-cell clones over the parental polyclonal lines (Fig. 3A
, right vs. middle panel). Thus, pooled FACS analysis characterizing the differentiating EBs demonstrated that the number of eGFP-expressing cells (normalized per a single beating EB) was higher (339) in the single-cell clone vs. the parental transgenic line (169).
|
We next analyzed the viability of the cells following the FACS-sorting procedure (using trypan blue staining). We found that 95% of the cells were viable prior to FACS sorting (following the enzymatic dispersion) and that 85% of the sorted eGFP-cells remained viable following this FACS procedure.
The selected eGFP-expressing cardiomyocytes could then be maintained and would remain viable for several weeks in culture (Fig. 3B
, bottom panel). The sorted eGFP-expressing cells were stained positively for different cardiac-specific markers, including anti-MLC-2v antibodies (Fig. 4
A). We next quantified the fraction of cells that continued to express eGFP following the FACS selection procedure and the percentage of these cells that express cardiac-specific markers. We noted that 96.8% (244 out of 252, n=4) of the sorted cells continued to express eGFP and that 93.4% (228/244, n=4) of these cells were also stained positively for cardiac-specific markers. In contrast, plating of the unfractionated dispersed cells, derived from similar stage contracting EBs, without FACS selection, resulted in the majority of these cultured cells not having a myocyte phenotype (Fig. 3B
, top panel). Moreover, the eGFP-based FACS selection strategy was found to be significantly better (P<0.01) than that of microdissection of the contracting areas from wild-type EBs (n=6), in which 58.8% of the cells were found to be positively stained for cardiac-specific markers.
|
Similar to the immunostaining studies, RT-PCR studies of the selected eGFP-cells demonstrated the expression of cardiac-specific genes. Figure 4B
depicts the results of these RT-PCR studies in four populations of cells: undifferentiated hESC, unfractionated cells derived from the differentiating EBs (prior to FACS), the sorted eGFP-expressing cells, and the nonselected cells. Note the expression of the pluripotent marker, Oct4, in the undifferentiated hESC and its significant down-regulation in all differentiated progeny. Also note that the highest expression of the cardiac-specific genes (MLC-2v, MLC-2a, and -MHC) was found in the eGFP-sorted population with a lower degree of expression in the unfractionated cells and even a lower degree in the nonsorted population (eGFP-). Similarly, the expression of nonmyocyte markers, such as -fetoprotein (an endodermal marker) and beta-III-tubulin (an ectodermal marker), could be identified in the unfractionated population with a significant diminution in the eGFP-selected cells.
We next sought to determine whether the eGFP-expressing cells also demonstrate functional properties typical of cardiomyocytes. The eGFP-expressing areas within the EBs were mechanically dissected and plated on top of a microelectrode array (MEA) mapping technique (Fig. 5
, left). The MEA consists of 60 electrodes (spaced 100 µm apart) that allow assessment of the electrical activity with extremely high spatial and temporal resolutions. An excellent spatial correlation was noted between the location of the eGFP-expressing area in the EB and the recording of electrical activity. Hence, local extracellular potentials could be recorded only in electrodes directly underlying the eGFP-expressing cells (Fig. 5)
. Determination of the LAT at each electrode allowed the construction of detailed activation maps depicting the spread of electrical activation within the eGFP-expressing region (Fig. 5
, right). These studies also showed that both the areas initiating the electrical activity (pacemaker areas) as well as the areas in which the action potential is propagated consist of eGFP-expressing cells.
|
Whole cell patch-clamp studies of the eGFP-expressing cells, either in small clumps or as isolated cells, demonstrated the presence of cardiac-specific action-potentials (Fig. 6
). Interestingly, we could observe the presence of a cardiac-specific electrophysiological signature also in eGFP-expressing cells that were not beating spontaneously. Out of the 33 eGFP-expressing cells studied, 32 were determined to be cardiomyocytes, based on either the presence of cardiac-specific action-potentials or ionic transients using the current- or voltage-clamp modes, respectively.
|
The action-potential properties of the eGFP-expressing cells were compared with the cardiomyocytes isolated from similar stage EBs derived from wild-type hESC lines (Fig. 6)
. We did not note any significant differences in the action-potential morphologies recorded from the eGFP cells and wild-type cells at similar developmental stages (Fig. 6
, bottom, right). In all cases an "embryonic"-like phenotype was identified. The action-potential measurements were also comparable between the cells with the maximal diastolic potentials (MDP) recorded being: –54.2 ± 3.2 mV and –55.3 ± 5.1 mV in the MLC-2V and wild-type derived hESC derived cardiomyocytes, respectively, and APD90 averaging 253 ± 33 ms and 288 ± 65 ms, respectively.
In vivo grafting
Finally, proof-of-concept studies were performed to test the ability of the eGFP-expressing cells, derived from the MLC-2V transgenic line, to form stable intracardiac cell grafts. Three days (n=4) and four weeks (n=3) following cell grafting, the animals were sacrificed and the hearts were harvested for pathological examination. As can be seen in Fig. 7
, the grafted cells survived and could be identified, in all animals studied, as relatively small, eGFP-expressing, cells that were interspersed isotropically within host rat myocardium. The cardiomyocyte and human phenotype of the grafted eGFP-expressing cells was verified by costaining for cardiac-specific (Fig. 7C
) and human-specific (Fig. 7E
) markers, respectively. Quantitative assessment of the histological specimens (n=4) demonstrated that 95% of the eGFP-expressing cells (273/287 cells) were also stained positively for cardiac-specific markers. Immunostaining studies for both undifferentiated markers (Oct-4) and endodermal markers (-fetoprotein) failed to show any positive staining within the cell graft.
|
The transplanted cells could still be identified within the host myocardium as long as 4 wk following cell grafting (the longest period studied, Fig. 7E-F
). Interestingly, some form of structural maturation seemed to have occurred during this period, with the grafted cells increasing in size and showing a more elongated morphology (Fig. 7E
). Importantly, the grafted cells were also demonstrated to form gap junctions with host myocardial cells (positive punctuated immunostaining for Cx43 indicated by the arrows in Fig. 7F
).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The use of tissue-specific promoters for selection of specific cell lineages was previously described in the murine ESC model and allowed identification and selection of pancreatic beta cells (28)
, neurons (29
, 30)
, and cardiomyocytes (31
, 32)
. This strategy has also been fine-tuned and utilized in the mouse ESC model to identify early cardiomyocyte precursor cells (33
, 34)
and even subpopulations of cardiomyocytes such as ventricular (35)
, atrial (36)
, or pacemaker cells (36
, 37)
. However, significant interspecies differences have prevented the development of similar strategies in the hESC model. These difficulties stem from the inability to achieve high stable transfection efficiencies in the hESC, the low clonal efficiency of these cells, the phenomenon of promoter shutdown during hESC propagation and differentiation, and the heterogeneous and prolonged differentiation process in the hESC model (38)
. Thus, despite the fact that more than 7 yr have passed since the original derivation of the hESC lines, this is the first report of the ability to specifically label, identify, and characterize purified populations of hESC derived cardiomyocytes.
We chose to use the MLC-2v promoter because, in our previous studies, we noted robust expression of this gene in cardiomyocytes generated during the in vitro differentiation of the hESC (15)
. A 560 bp fragment of the MLC-2v promoter, upstream of the transcriptional start of the human hz15030 gene (which encodes the MLC-2v protein) was utilized. We chose to use this fragment because of its homology to a 250 bp fragment in the rat MLC-2 promoter, that was previously found to be sufficient for cardiac-specific expression (24
, 25)
. The rat MLC-2v promoter was characterized previously in detail by a number of studies and cardiac-specific regulatory elements in the 5' flanking region of this gene were recognized (24
, 25)
. The aforementioned 250 bp fragment within the MLC-2 promoter contains CArG motifs and additional conserved regions, two of which (HF-1 and HF-2) were found to play an important role in the cardiac-specific expression (25)
.
Interestingly, the generation of the transgenic hESC lines did not have any negative impact on the ability to propagate these cells in the undifferentiated state (with the undifferentiated colonies stained positively for the specific markers Oct-4, SSEA-4, and Tra-I-60), on their pluripotent properties (their ability to form teratomas in immunosupressed mice), or on their ability to differentiate into spontaneously beating cardiomyocytes when compared to the wild-type lines. An excellent spatial correlation was found between the contracting areas within the EBs and eGFP expression. Detailed immunocytostaining and patch-clamp analysis of dispersed cells isolated from the differentiating EBs and of eGFP-expressing cells selected following FACS sorting, confirmed the cardiac-specificity of the MLC-2V promoter-driven eGFP-expression with >93% of these cells displaying a cardiac-specific phenotype. This eGFP-based selection strategy was found to be significantly better than that of microdissection of the beating areas of wild-type EBs.
Since the transgenic lines generated were derived from individually transfected colonies, rather than from single-cells, we expected that their might be some variations in the degree of transgene expression within the cells in the differentiating EBs. This was manifested by the presence of some heterogeneity in these polyclonal cell lines in the number of cardiomyocytes expressing eGFP as well as in the level of eGFP expression by these cardiomyocytes. We, therefore, continued to generate single-cell clones from the parental transgenic hESC lines. The single-cell clones retained the properties of the parental transgenic lines in terms of their unique undifferentiated properties, their pluripotency, and their ability to differentiate into cardiomyocytes. More importantly, transgene expression in the differentiating cardiomyocyte was more homogeneous in these single-cell clone derivatives in terms of the percentage of cardiomyocytes expressing eGFP as well as in the degree and intensity of eGFP expression.
Interestingly, the molecular, structural, and functional properties (patch-clamp data) of the early stage human cardiomyocytes identified based on the activity of the MLC-2v promoter did not differ significantly from those previously reported by our group in nonselected cardiomyocyte populations at similar stage of development that were derived from the wild-type hESC lines (15
, 19
, 20)
. The cells were relatively small, had a high nuclear/cytoplasm ratio, and displayed an early striated pattern, which indicated a relatively early stage ultrastructural maturation. Similarly, the eGFP-cells expressed simultaneously a number of cardiac-specific genes including MLC-2v, MLC-2a, and ANP (data not shown), which are considered to be chamber-specific in the adult heart. This phenomenon is not surprising, since it is well established in the literature that ANP expression, for example, is not limited to the atria but is ubiquitously expressed also in embryonic and fetal ventricular tissue (39
40
41)
, as well is in early stage hESC-derived cardiomyocytes (15)
. Similarly, our previous reports demonstrate robust expression of these genes in the early stage wild-type hESC derived cardiomyocytes (15)
.
Detailed patch-clamp studies of the pMLC-2V-eGFP cells revealed a typical "embryonic" phenotype similar to that found in the unselected wild-type hESC derived cardiomyocytes at similar maturation stages (19)
. Our results are somewhat different than those reported by He et al. (18)
, which found atrial-, ventricular-, and nodal-like action potential morphologies in contracting EBs generated from the hESC. These potential discrepancies may stem from the different developmental stage of the EBs studied (with later-stage EBs analyzed in the He study), the recording methods (sharp electrodes in whole EBs vs. whole-cell patch-clamp of single cells in our study), and differences in the lines used and culturing techniques. Still, note that 14% of the -actinin positive cells, derived from the differentiating EBs, were eGFP negative (and therefore could not be analyzed in the patch-clamp studies); this finding suggests that this population of cells may possess a different electrophysiological phenotype. A more detailed electrophysiological characterization of the cells should be focused, in the future, to analyze the type of ionic currents expressed by the cells, since it provides a better "signature" of the electrophysiological phenotype of the cells than action-potential morphology alone.
The similar properties of the cardiomyocytes identified suggest that at this relatively early stage of maturation (up to 30–40 days of differentiation), the differentiating cardiomyocytes, selected based on the activity of MLC-2V promoter, probably represent the same, relatively homogenous, population of early stage embryonic cardiomyocytes. These results are somewhat different than those reported in the mouse ES model, in which specification into chamber-specific phenotypes occurs relatively early during EB differentiation (35
, 42)
and in which selection based on the MLC-2V promoter can specifically enrich for ventricular cardiomyocytes with a more adult ventricular phenotype (35)
. These interspecies differences may stem from the differences in the length of the gestational period between the two species. In fact, preliminary results from our lab (data not shown), using the wild-type differentiating system suggest the beginning of electrophysiological maturation in the hESC-derived cardiomyocytes after more then 3 months in culture. It will be interesting, in the future, to perform similar, technical demanding, long-term studies also in the transgenic lines to asses for possible differences in the electrophysiological properties of the eGFP-expressing cardiomyocytes selected based on the activity of the MLC-2v promoter.
The ability to identify and select for specific cell types by using tissue-restrictive promoters may have important research and clinical applications. In the cardiovascular field this may allow to identify and study early cardiac precursor cells that can not be recognized due to the lack of cross-striation and spontaneous beating. Identification of even earlier cardiac precursor cells can be achieved by using an even earlier-expressed cardiac promoter [such as Nkx2.5 (33
, 34)
] and may allow us to gain important insights into the mechanisms involved in early human cardiac differentiation and functional maturation. Similarly, the ability to identify, online, the generation of viable cardiomyocytes (using eGFP expression) provides a unique tool to robustly quantify the extent of hESC cardiomyocyte differentiation in studies examining the effects of different interventions on cardiomyocyte yield.
Most importantly, the ability to select and generate pure cardiomyocyte cultures may be of specific importance for the possible future utilization of hESC in the emerging field of myocardial cell replacement therapy, to avoid the presence of noncardiomyocyte cell derivatives and to ensure the absence of any remaining pluripotent stem cells in the cell grafts. The latter issue may be crucial to prevent the possible generation of hESC-related tumors such as teratomas (23)
. In this respect, the cell grafting results of the present study (demonstrating the feasibility of intramyocardial engraftment and identification of the eGFP-expressing cells as well as the continuous expression of the transgene and maintenance of the cardiac phenotype by these cells), together with recent reports showing the ability of the hESC derived cardiomyocytes to integrate functionally and structurally with host cardiomyocytes both in vitro and in vivo, are encouraging (21
, 22)
. Although fusion cannot be completely excluded as a mechanism for the cardiac phenotype of the engrafted eGFP expressing cells, it is unlikely to play a major role due to the presence of the cardiac phenotype of the grafted cells already prior to cell transplantation and the fact that the eGFP-expressing cells are smaller than host rat cardiomyocytes and display a more immature sracomeric pattern.
Taken together, this report describes the generation of unique transgenic hESC that allow, for the first time, identification, labeling, and sorting of pure populations of human cardiomyocytes. This ability may have important implication for several cardiovascular research fields, including basic developmental studies, pharmacological and physiological studies, cell therapy, and tissue engineering. More broadly the results described in the current report may also be applicable for identification and selection of other hESC derived cell lineages provided that adequate tissue-specific promoters can be identified.
|
ACKNOWLEDGMENTS |
|---|
Received for publication January 23, 2006. Accepted for publication March 1, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. M. Pedrotty, R. Y. Klinger, N. Badie, S. Hinds, A. Kardashian, and N. Bursac Structural coupling of cardiomyocytes and noncardiomyocytes: quantitative comparisons using a novel micropatterned cell pair assay Am J Physiol Heart Circ Physiol, July 1, 2008; 295(1): H390 - H400. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Liu, J. D. Fu, C. W. Siu, and R. A. Li Functional Sarcoplasmic Reticulum for Calcium Handling of Human Embryonic Stem Cell-Derived Cardiomyocytes: Insights for Driven Maturation Stem Cells, December 1, 2007; 25(12): 3038 - 3044. [Abstract] [Full Text] [PDF]
|
|
