In Vitro Cell.Dev.Biol.—Animal DOI 10.1007/s11626-015-9880-4
A fetal human heart cardiac-inducing RNA (CIR) promotes the differentiation of stem cells into cardiomyocytes Andrei Kochegarov & Ashley Moses-Arms & Larry F. Lemanski
Received: 13 December 2014 / Accepted: 11 February 2015 / Editor: T. Okamoto # The Society for In Vitro Biology 2015
Abstract A specific human fetal heart RNA has been discovered, which has the ability to induce myocardial cell formation from mouse embryonic and human-induced pluripotent stem cells in culture. In this study, commercially obtained RNA from human fetal heart was cloned, sequenced, and synthesized using standard laboratory approaches. Molecular analyses of the specific fetal cardiac-inducing RNA (CIR), revealed that it is a fragment of N-sulfoglucosaminesulfohydrolase and the caspase recruitment domain family member 14 precursor. Stem cells transfected with CIRs often form into spindleshaped cells characteristic of cardiomyocytes,and express the cardiac-specific contractile protein marker, troponin-T, in addition to tropomyosin and α-actinin as detected by immunohistochemical staining. Expression of these contractile proteins showed organization into sarcomeric myofibrils characteristic of striated cardiac muscle cells. Computer analyses of the RNA secondary structures of the active CIR show significant similarities to a RNA from salamander or myofibrilinducing RNA (MIR), which also promotes non-muscle cells to differentiate into cardiac muscle. Thus, these two RNAs, salamander MIR and the newly discovered human-cloned CIR reported here, appear to have evolutionarily conserved secondary structures suggesting that both play major roles in vertebrate heart development and, particularly, in the differentiation of cardiomyocytes from non-muscle cells during development. Keywords Cardiomyocytes . Myofibrillogenesis . Cardiac-inducing RNA (CIR) . RNA secondary structure . Stem cells A. Kochegarov : A. Moses-Arms : L. F. Lemanski (*) Department of Biological and Environmental Sciences, Texas A&M University-Commerce, Commerce, TX 75429-3011, USA e-mail: [email protected]
Introduction In our earlier studies on mechanisms of myofibrillogenesis and cardiac myocyte differentiation, we discovered a unique RNA in the Mexican axolotl (a salamander), which can turn non-muscle cells into vigorously contracting cardiomyocytes with normal myofibrils (Zhang et al. 2003, 2009). We named it MIR for myofibril-inducing RNA. Our studies were published using the cardiac mutant axolotl embryonic heart organ culture bioassay system (Lemanski et al. 1996; Zhang et al. 2003; Rueda-de-León et al. 2011; Kochegarov et al. 2013). Non-contracting mutant embryo hearts, lacking organized myofibrils, were placed in organ cultures with MIR from normal axolotl embryonic hearts or endoderm and after 48–72 h, the mutant hearts begin contracting rhythmically, and immunofluorescent staining with contractile protein antibodies (e.g., anti-tropomyosin, anti-cardiac troponin-T, anti-αactinin) revealed well-organized sarcomeric myofibrils of normal morphology. Subsequently, a US Patent (US Issued Patent [9/18/07] 60/462 171, Lemanski and Zhang, Inventors) was issued for the specific MIR primary sequence from axolotls that has the ability to rescue the mutant axolotl hearts by converting non-muscle cells in the mutant heart myocardial wall into vigorously contracting cardiomyocytes (Lemanski and Zhang 2007; Zhang et al. 2009). Also, in earlier preliminary studies, we found evidence that commercially purchased RNA from human adult or fetal heart was able to rescue the non-function mutant axolotl embryonic hearts by converting the non-muscle cells in their heart walls into contracting myocardial tissue with well-organized sarcomeric myofibrils (Rueda-de-León et al. 2011). Since no sequence homologues for the axolotl MIR were found in the human genome using the NCBI databases, we designed further experiments to determine whether human fetal heart RNA might have rescuing capabilities through functional
KOCHEGAROV ET AL.
homologues to the axolotl MIR. We have since found a RNA derived from human fetal heart RNA obtained commercially, which has the same ability as the MIR from axolotl to promote the rescue of mutant axolotl hearts (Kochegarov et al. 2013). In addition, this human fetal RNA promotes the differentiation of mouse embryonic stem cells and humaninduced pluripotent stem cells (from WiCell, Inc., Madison, WI) to form into cardiomyocytes in culture. We call this human-derived RNA, cardiac-inducing RNA (CIR). We have characterized these mouse and human stem-cell-derived cardiomyocytes as functional cardiac cells, since they are able to express cardiac markers (cardiac troponin-T, tropomyosin, α-actinin) and contract vigorously and rhythmically when they clump together to form into cardiac cell aggregations. In the present study, we randomly cloned 396 RNAs expressed in human fetal heart and tested them to identify individual RNAs which might promote cardiomyocyte differentiation. Using this approach, we identified and characterized the active clone of CIR from the fetal human heart which induces the differentiation of mouse embryonic stem cells and human-induced pluripotent stem cells to form into definitive cardiomyocytes. We further have found that the human CIR secondary structure is very similar to our previously described RNA (MIR) from salamander heart (Zhang et al. 2003) although there is not a significant sequence homology. We believe that the similarities in secondary structures of the MIR and the newly discovered human RNA (CIR) account for their abilities to promote myofibrillogenesis and rescue cardiac mutant axolotl hearts as well as promote mouse and human stem cells to form into well-defined cardiomyocytes.
kanamycin and incubated overnight at 37 °C on a shaker at 225–250 rpm. Plasmids with clones were extracted according to the standard Miniprep Plasmid DNA Isolation Protocol in the online archive of the Institute of Bioinformatics and Applied Biotechnology. PCR. The T7 RNA polymerase-binding site was added to the 5′ end of the forward and reverse M13 primers indicated below. Forward primer: 5′-TAATACGACTCACTATAGGGGT AAAACGACGGCCAG-3′ Reverse primer 5′-TAATACGACTCACTATAGGGCA GGAAACAGCTATGAC-3′ PCR was performed using a MyTaq™ Red Mix kit (Bioline USA, Inc., BIO-25043, Taunton, MA) including polymerase and dNTP plus the above primers and DNA templates. The reaction included denaturation at 95°C for 15 s followed by annealing at 55°C for 15 s and elongation at 72°C for 15 s during 30 cycles. The resulting DNA was purified by 5 M sodium chloride salt and isopropanol precipitation. Pellets were washed with 70% ethanol and re-suspended in 1× Tris-EDTA buffer. RNA synthesis. The transcription reaction mixture was assembled from the MAXIscript® T7 Kit (Ambion No. AM1314M, Life Technologies, Grand Island, NY). We added 1 μg of PCR product DNA, 2 μL of 10× transcription buffer, 2 μL of T7 Enzyme Mix, and 1 μL of each (10 mM) dNTP and adjusted the volume to 20 μL with nuclease-free water. The reaction mixture was incubated at 37°C for 2 h. RNA was purified using ammonium acetate and ethanol precipitation and resuspended in nuclease-free water. The concentration of RNA was determined spectrophotometrically at 260 nm using a Synergy HT (BioTek, Winooski, VT) platereader.
Materials and Methods Cloning. For cloning, a total of 2 μg of human fetal heart RNA (No. 540165, Agilent Technologies, Inc, Santa Clara, CA) was used for each reaction. The CloneMiner™ II cDNA Library Construction Kit (No. A11180, Invitrogen, Carlsbad, CA) was used to create individual clones. First and second DNA strands were synthesized from template RNAs and ligated into the pDONR222 vector. The pDONR222 vector contains a kanamycin resistance gene which allows for selection of transfected bacteria and the ccdB gene which interferes with Escherichia coli DNA gyrase allowing for negative selection of the donor vector in E. coli following recombination and transformation. Plated cells were incubated overnight at 37°C. Individual colonies containing the vector with cloned genes (over 4000 genes from fetal human heart) were collected and transferred into snap-cap tubes with 2 mL of 2xYT medium containing 50 μg/mL of
qRT-PCR. RNA was extracted using a NucleoSpin RNAII Kit (Macherey-Nagel, Bethlehem, PA) from differentiated cells treated with the active RNA, and one control untreated (treated only with lipofectamine). Quantitative reverse transcription polymerase chain reaction (qRT-PCR) was performed with a Rotor-Gene machine using a Rotor-Gene SYBR PCR kit (No. 204074, Qiagen, Valencia, CA) with primers as designed in our earlier studies (Zhang et al. 2009). Stem cell culture and differentiation protocol. Human-induced pluripotent stem cells (iPSCs), DF19-9–11T.H, from WiCell, Inc. (Madison, WI) and mouse embryonic stem cells (mESCs), Strain 129, OriCell from Cyagen Biosciences, Inc. (Santa Clara, CA) were incubated and grown at 37°C in a humidified 5% CO2 atmosphere, and passaged routinely according to our routine protocols (Lemanski et al. 2012). To promote the formation of embryoid bodies, 20uL hanging
RNA PROMOTES CARDIOGENESIS
drops of cell suspensions were micropipetted onto the inner surface of a Petri dish lid and cultured for 2 d, then washed in medium and transferred to gelatin-coated dishes. When the cells began to proliferate and spread, they were transfected with active clones of RNA mixed with Lipofectamine RNA:Max (Life Technologies) diluted to a concentration of 50 ng/ul in OPTI-MEM medium (Life Technologies) and incubated for 6 h. Fixation, staining, and confocal microscopy of cultured stem cells. CIR-treated and non-inducing RNA-treated or untreated control cells in culture were fixed in 4% paraformaldehyde for 30 min, rinsed in phosphate-buffered saline (PBS) with 3% bovine serum albumin (BSA) for 3 min, permeabilized in 0.1% Tween-20 and 3% BSA, and stained with primary antibody diluted 1:75 with PBS overnight. The cells were then rinsed with PBS and 3% BSA for 3 min and stained with the secondary anti-mouse antibody diluted to 1:75 with PBS for 1 h. The cells were immunofluorescently stained for tropomyosin, cardiac troponin-T, or α-actinin. The primary antibodies used for all three proteins were monoclonal antibodies from mouse, and the secondary antibodies were Goat F (ab) antimouse polyclonal antibodies with a fluorescein isothiocyanate (FITC) tag excited at 490 nm (Abcam, Cambridge, MA). The cells were analyzed using a confocal microscope to identify and localize the presence of tropomyosin, cardiac troponin-T, and α-actinin in the cells, including those proteins present in organized sarcomeric myofibrils. The immunofluorescently stained cells were analyzed using an Olympus BX62 scanning laser confocal microscope. Sequencing and secondary structure prediction. Specific RNA molecules were determined to have rescuing abilities initially by using our mutant axolotl heart bioassay system (Zhang et al. 2009). That specific cloned RNA then was sent to Functional Biosciences (Madison, WI) to determine its original DNA sequence. The DNA sequence associated with the rescuing RNA was screened for vector contamination using the protocols on the NIH Website http://www.ncbi. nlm.nih.gov/VecScreen/VecScreen.html, and then it was trimmed to remove contamination. The online NCBI BLAST program was utilized to determine the exact sequence within the human genome, and the sequence editor database at http://www.fr33.net/seqedit.php was used to convert the DNA sequence into the RNA sequence. The resulting RNA sequence was entered into the Genebee RNA computational software secondary structure prediction program developed in the Belozersky Institute at Moscow State University, Russia. This software was used to predict likely secondary structures for the RNA. The RNA’s
secondary structure was then compared with the secondary structure of the axolotl MIR to determine similarities.
Results Analysis of pluripotency in human-induced pluripotent stem cells and mouse embryonic stem cells in culture. When the human-induced pluripotent stem cells, DF19-9–11T.H (WiCell, Inc.) and mouse embryonic stem cells, Strain 129 (OriCell from Cyagen Biosciences, Inc.), were placed in culture and incubated with the human iPSCs (Fig. 1A) and mouse ESCs (Fig. 1B), they formed monolayer growth colonies that were rounded in shape and tightly packed together with distinct, well-defined edges characteristic of undifferentiated cells. When stained with antibodies for the human iPSCs Oct-3/4 pluripotency factor, both human iPSCs (Fig. 1C) and mouse (not illustrated) stain red. When the cells are also double stained with the blue nuclear stain, DAPI (Fig. 1D), it is clear from the merged image (Fig. 2E) that most of the cells (appearing purple) express the Oct-3/4 in the nuclei, thus confirming pluripotency of the stem cells at this early stage of culture. Formation of embryoid bodies. The stem cell suspensions placed on the inner surface of Petri dish lids in Bhanging drops^ (Fig. 2A) form into distinct clumps (embryoid bodies) after 2 d in the hanging drop cultures (Fig. 2B, C). The embryoid bodies then are washed in medium and transferred to gelatin-coated culture dishes, and after 4–5 d in these cultures, the cells from the embryoid bodies begin to attach to the substratum (Fig. 2D). At this point, the cells are transfected with active clones of RNA mixed with Lipofectamine RNA:Max to ensure uniform entry into the cells. Mutant axolotl heart bioassays to identify active human RNA clones. Initial bioassays were conducted by combining and pooling groups of RNAs in which mutant salamander nonbeating hearts were organ cultured (Moses-Arms et al. 2014). Control experiments involved organ culturing mutant hearts with no treatment and normal hearts with and without treatment. Pooled RNA groups that caused the treated mutant hearts to begin beating by the 2nd–4th d of treatment were separated into smaller testing groups until each individual RNA had been checked for rescuing ability. Within the 396 clones tested, one clone that showed significant rescuing ability and caused the non-beating mutant hearts to beat vigorously was clone No. 6 (now called CIR). In the present study, we describe and characterize the CIR derived from clone No. 6 and test for the CIR’s ability to promote the differentiation of non-muscle mouse embryonic and human-induced pluripotent stem cells into definitive cardiomyocytes.
KOCHEGAROV ET AL. Figure 1. A, B Nondifferentiated colonies of human iPSCs (A) and mouse ESCs (B). C Immunohistochemical staining in red (rhodamine) of human iPSCs with antibodies for the pluripotency marker Oct-3/4. D Nuclei stain in blue with DAPI in these human iPSCs. E Combined images of C Oct-3/4 and D DAPI show that the majority of human iPSCs express the factor of pluripotency Oct-3/4 which colocalizes in most of the cell nuclei stained with DAPI (purple shows co-localization).
Confocal microscopy of positive control axolotl hearts. Analysis of untreated normal axolotl embryonic hearts show well-organized sarcomeric cardiac myofibrils (Fig. 3A) Figure 2. A Plating of stem cells in hanging drops on the inside of a Petri dish lid. B Day 2. Formation of clumps (embryoid bodies) in the suspension. C High magnification of EB in the suspension. D Attachment and spreading of cells on culture dish after Day 5.
with antibody staining for tropomyosin. Untreated mutant hearts cultured in Holtfreter’s solution without CIR treatment showed virtually no staining with anti-tropomyosin
RNA PROMOTES CARDIOGENESIS
Figure 3. Tropomyosin expression revealed by immunochemical staining followed by confocal imaging. A Normal non-treated embryonic axolotl heart. B.Mutant non-treated embryonic axolotl heart. C Mutant embryonic axolotl heart after transfection with CIR derived
from fetal human heart RNA clones. The normal and CIR-treated mutant hearts show extensive staining for tropomyosin. The untreated mutant heart shows essentially no staining.
antibodies, and no myofibrillar structures could be identified in these untreated, undifferentiated mutant heart cells (Fig. 3B). Mutant axolotl hearts treated with the CIR fetal heart human RNA showed well-organized myofibrils to the extent that these rescued mutant myocardial cells (Fig. 3C) appeared indistinguishable from the normal control hearts compare (Fig. 3A, C). Thus, the CIR clearly has caused the cardiac mutant axolotl heart cells to form an undifferentiated embryonic, non-cardiac phenotype into normally appearing, vigorously contracting differentiated cardiac muscle cells containing myofibrils of normal morphology. The CIR derived from clone No. 6 rescues the mutant embryonic axolotl hearts in a manner very similar to the MIR derived from normal (+/+ or +/c) axolotl embryonic anterior endoderm that has been previously described (Zhang et al. 2009). It is very clear that the axolotl MIR and the human clone No. 6 CIR serve as functional homologues in the cardiac mutant axolotl rescue bioassay and both RNAs have the ability to promote the differentiation of beating cardiac tissue from undifferentiated mutant embryonic heart cells.
into the online NCBI BLAST program to determine for what protein the sequence coded, and it was identified as a fragment of N-sulfoglucosaminesulfohydrolase and the caspase recruitment domain family member 14 precursor. Using a sequence converter, the DNA sequence was converted to the corresponding RNA sequence (Fig. 4B). The possible secondary structures of the comparable regions of the human CIR (Fig. 5A), the normal axolotl MIR (Fig. 5B), and the mutant axolotl MIR (Fig. 5C) were determined using the Genebee RNA secondary structure prediction model. In comparing the active CIR from human (Fig. 5A) and the active MIR from normal (+/+ and +/c) axolotl (Fig. 5B), which have very similar secondary structures, with the non-active c/c mutant axolotl MIR (Fig. 5C), both human and normal axolotl have major differences in secondary RNA structure from the mutant. The human CIR and normal MIR both promote myofibril formation in and rescue mutant hearts. The c/c mutant RNA, with its significantly different secondary structure, lacks the capability of promoting myofibrillogenesis and inducing differentiation of the mutant embryonic hearts. Thus, the secondary structures of the RNAs appear to be a critical factor in the human CIR’s and the normal axolotl MIR’s abilities to induce cardiac myofibril cell formation and rescue the mutant hearts.
Sequencing and secondary structure of the active clone. PCR products and plasmids containing the DNA that corresponded to the CIR (clone No. 6 RNA) were sent to Functional Biosciences (Madison, WI) for determination of its DNA sequence (Fig. 4A). The trimmed DNA sequence was entered Figure 4. Sequences of CIR cloned as DNA (A) and converted into RNA (B).
KOCHEGAROV ET AL.
Figure 5. Secondary structure of CIR (A), active normal salamander MIR (B), and non-active mutant axolotl MIR (C). The CIR and active normal axolotl MIR have regions of very similar structure. Both differ from the non-active mutant RNA. Comparable areas are shown in boxes.
Analysis of CIR in promoting cardiomyocyte differentiation from mouse embryonic stem cells and human-induced pluripotent stem cells. Spontaneous cardiomyocyte differentiation of mouse and human stem cells has been described previously by Mummery et al. (2002). In the present study, stem cells were differentiated into cardiomyocytes after formation of embryoid bodies (EBs) by culturing the cells in the presence of cloned CIR. We tested this approach and found that
spontaneous Bbackground^ cardiomyocyte differentiation without RNA treatment was approximately 9–10%. We passaged stem cells and plated them into small drops as Bhanging drops^ on Petri dish lids and incubated them for 2 d. On the second day, stem cells aggregated and formed EBs. We plated the EBs on collagen-coated dishes and allowed them to attach. Cells from EBs started to grow and proliferate. At this stage, we treated them with the CIR and in 7–8 d, 70–80% had differentiated into cell shapes characteristic for cardiomyocytes (Fig. 6A, B) and expressed cardiac-specific troponin-T as revealed by antibody staining methods (Fig. 7A, B). Non-treated control human iPSCs or mouse ESCs showed only 9–10% of the cells with cardiac morphological traits (Fig. 6C, D) at this 7–8-d stage in culture; moreover, there was very little detectable cardiac staining for cardiac-specific troponin-T in any of the non-treated human or mouse stem cells (Fig. 7C, D). High-resolution confocal imaging illustrated myofibril organization in the stem cell-derived cardiomyocytes from iPSCs and from mouse embryonic stem cells (ESC) transfected with active CIR when stained with antibodies against cardiac-specific troponin-T, tropomyosin, or αactinin (Fig. 8A–C). When the RNAs from stem cell-derived cardiomyocytes and from non-differentiated stem cells were extracted and expression of cardiac-specific messenger RNA (mRNA) quantified with qRT-PCR (Fig. 9), it was shown that the CIR treatment very significantly increased expression of the cardiac markers, cardiac troponin-T, and tropomyosin, in comparison with the untreated cells in ESC or iPSC. The expression of cardiac-specific troponin-T in human stem cells and mouse embryonic stem cells far exceeded the expression in untreated mouse or human cells. In fact, it was in the range of 7–8-fold higher in the CIR-treated cells after only 7–8 d in culture (Fig. 9). Tropomyosin was also significantly higher in the CIR-treated stem cells, approximately 5-fold higher (Fig. 9). Also, when we screened the cloned sequence in the human genome with BLAST at the NCBI database, we found two high score matches with a portion of exon 8 of the human N-sulfoglucosaminesulfohydrolase (SGSH) gene on the sense strand of DNA and with the caspase recruitment domain family, member 14 (CARD14), on the antisense strand. These genes, SGSH and CARD14, partially overlap and belong to opposite DNA strands (forward and reverse) on human chromosome 17.
Discussion We have discovered a specific RNA, originally identified from commercially obtained human fetal heart RNA, which has the ability to turn undifferentiated non-muscle cells into definitive
RNA PROMOTES CARDIOGENESIS Figure 6. Human iPSCs (A) and mouse ESCs (B) differentiated into cells with morphologies characteristic of cardiomyocytes after transfection with CIR. Human iPSC (C) and mouse ESC (D) controls, without transfection with CIR do not show myocardial cell morphologies.
cardiomyocytes. This was accomplished by initially preparing 396 human fetal RNA clones and analyzing them in pooled groups to evaluate their ability to promote myocardial cell Figure 7. Human iPSC-derived (A) and mouse ESC-derived (B) cardiomyocytes after transfection with CIR and immunostained for cardiac troponin-T; human iPSC (C) and mouse ESC (D) controls immunostained without transfection of CIR. The CIRtreated cells show an intense reaction for the cardiac-specific troponin-T while the untreated cultures show very little staining.
formation using our published cardiac mutant axolotl heart rescue bioassay system (Zhang et al. 2003, 2009). After these initial experiments to confirm the positive cardiomyogenic
KOCHEGAROV ET AL. Figure 8. Differentiated cardiomyocytes derived from human iPSCs treated with CIR and stained with antibodies against cardiac-specific troponinT (A), tropomyosin (B), and αactinin (C). The cells stain for all of these muscle-related proteins.
effect of clone No. 6 RNA (now called CIR), we tested CIR on mammalian stem cells in culture, both mouse ESC from Cyagen Biosciences, Inc. and human iPSC from WiCell, Inc.. In the present study, our results confirm that CIR has the ability to promote the differentiation of both mouse embryonic and human-induced pluripotent stem cells into definitive cardiac muscle cells. Immunoflourescent confocal microscopy demonstrates the presence of cardiac myofibrils that contain cardiac-specific troponin-T as well as tropomyosin in 70–80% of the CIR-treated cells. In addition, CIR-treated cardiomyocytes contain well-organized myofibrils often with definitive sarcomeric structures, both strong indicators that these cells indeed are myocardiocytes. Only 9–10% of untreated stem cells express cardiac-specific troponin-T and tropomyosin. It is very clear from our results that the CIR has a strong inducing effect in converting these non-muscle stem cells into definitive cardiac muscle cells. The mechanism for this CIR cardiogenic process will require further analysis to be fully understood. However, one interesting observation that suggests a possible mechanism for the action of the CIR relates to our finding that the cloned sequence in the human genome with BLAST at the NCBI database, has two high score matches with a portion of exon 8 of the human N-sulfoglucosaminesulfohydrolase (SGSH) Figure 9. Expression of cardiacspecific mRNA quantified with qRT-PCR in human iPSC-derived (A) and mouse ESC-derived (B) cardiomyocytes after transfection with CIR (CIR) compared with non-transfected controls (Con); p
A fetal human heart cardiac-inducing RNA (CIR) promotes the differentiation of stem cells into cardiomyocytes Andrei Kochegarov & Ashley Moses-Arms & Larry F. Lemanski
Received: 13 December 2014 / Accepted: 11 February 2015 / Editor: T. Okamoto # The Society for In Vitro Biology 2015
Abstract A specific human fetal heart RNA has been discovered, which has the ability to induce myocardial cell formation from mouse embryonic and human-induced pluripotent stem cells in culture. In this study, commercially obtained RNA from human fetal heart was cloned, sequenced, and synthesized using standard laboratory approaches. Molecular analyses of the specific fetal cardiac-inducing RNA (CIR), revealed that it is a fragment of N-sulfoglucosaminesulfohydrolase and the caspase recruitment domain family member 14 precursor. Stem cells transfected with CIRs often form into spindleshaped cells characteristic of cardiomyocytes,and express the cardiac-specific contractile protein marker, troponin-T, in addition to tropomyosin and α-actinin as detected by immunohistochemical staining. Expression of these contractile proteins showed organization into sarcomeric myofibrils characteristic of striated cardiac muscle cells. Computer analyses of the RNA secondary structures of the active CIR show significant similarities to a RNA from salamander or myofibrilinducing RNA (MIR), which also promotes non-muscle cells to differentiate into cardiac muscle. Thus, these two RNAs, salamander MIR and the newly discovered human-cloned CIR reported here, appear to have evolutionarily conserved secondary structures suggesting that both play major roles in vertebrate heart development and, particularly, in the differentiation of cardiomyocytes from non-muscle cells during development. Keywords Cardiomyocytes . Myofibrillogenesis . Cardiac-inducing RNA (CIR) . RNA secondary structure . Stem cells A. Kochegarov : A. Moses-Arms : L. F. Lemanski (*) Department of Biological and Environmental Sciences, Texas A&M University-Commerce, Commerce, TX 75429-3011, USA e-mail: [email protected]
Introduction In our earlier studies on mechanisms of myofibrillogenesis and cardiac myocyte differentiation, we discovered a unique RNA in the Mexican axolotl (a salamander), which can turn non-muscle cells into vigorously contracting cardiomyocytes with normal myofibrils (Zhang et al. 2003, 2009). We named it MIR for myofibril-inducing RNA. Our studies were published using the cardiac mutant axolotl embryonic heart organ culture bioassay system (Lemanski et al. 1996; Zhang et al. 2003; Rueda-de-León et al. 2011; Kochegarov et al. 2013). Non-contracting mutant embryo hearts, lacking organized myofibrils, were placed in organ cultures with MIR from normal axolotl embryonic hearts or endoderm and after 48–72 h, the mutant hearts begin contracting rhythmically, and immunofluorescent staining with contractile protein antibodies (e.g., anti-tropomyosin, anti-cardiac troponin-T, anti-αactinin) revealed well-organized sarcomeric myofibrils of normal morphology. Subsequently, a US Patent (US Issued Patent [9/18/07] 60/462 171, Lemanski and Zhang, Inventors) was issued for the specific MIR primary sequence from axolotls that has the ability to rescue the mutant axolotl hearts by converting non-muscle cells in the mutant heart myocardial wall into vigorously contracting cardiomyocytes (Lemanski and Zhang 2007; Zhang et al. 2009). Also, in earlier preliminary studies, we found evidence that commercially purchased RNA from human adult or fetal heart was able to rescue the non-function mutant axolotl embryonic hearts by converting the non-muscle cells in their heart walls into contracting myocardial tissue with well-organized sarcomeric myofibrils (Rueda-de-León et al. 2011). Since no sequence homologues for the axolotl MIR were found in the human genome using the NCBI databases, we designed further experiments to determine whether human fetal heart RNA might have rescuing capabilities through functional
KOCHEGAROV ET AL.
homologues to the axolotl MIR. We have since found a RNA derived from human fetal heart RNA obtained commercially, which has the same ability as the MIR from axolotl to promote the rescue of mutant axolotl hearts (Kochegarov et al. 2013). In addition, this human fetal RNA promotes the differentiation of mouse embryonic stem cells and humaninduced pluripotent stem cells (from WiCell, Inc., Madison, WI) to form into cardiomyocytes in culture. We call this human-derived RNA, cardiac-inducing RNA (CIR). We have characterized these mouse and human stem-cell-derived cardiomyocytes as functional cardiac cells, since they are able to express cardiac markers (cardiac troponin-T, tropomyosin, α-actinin) and contract vigorously and rhythmically when they clump together to form into cardiac cell aggregations. In the present study, we randomly cloned 396 RNAs expressed in human fetal heart and tested them to identify individual RNAs which might promote cardiomyocyte differentiation. Using this approach, we identified and characterized the active clone of CIR from the fetal human heart which induces the differentiation of mouse embryonic stem cells and human-induced pluripotent stem cells to form into definitive cardiomyocytes. We further have found that the human CIR secondary structure is very similar to our previously described RNA (MIR) from salamander heart (Zhang et al. 2003) although there is not a significant sequence homology. We believe that the similarities in secondary structures of the MIR and the newly discovered human RNA (CIR) account for their abilities to promote myofibrillogenesis and rescue cardiac mutant axolotl hearts as well as promote mouse and human stem cells to form into well-defined cardiomyocytes.
kanamycin and incubated overnight at 37 °C on a shaker at 225–250 rpm. Plasmids with clones were extracted according to the standard Miniprep Plasmid DNA Isolation Protocol in the online archive of the Institute of Bioinformatics and Applied Biotechnology. PCR. The T7 RNA polymerase-binding site was added to the 5′ end of the forward and reverse M13 primers indicated below. Forward primer: 5′-TAATACGACTCACTATAGGGGT AAAACGACGGCCAG-3′ Reverse primer 5′-TAATACGACTCACTATAGGGCA GGAAACAGCTATGAC-3′ PCR was performed using a MyTaq™ Red Mix kit (Bioline USA, Inc., BIO-25043, Taunton, MA) including polymerase and dNTP plus the above primers and DNA templates. The reaction included denaturation at 95°C for 15 s followed by annealing at 55°C for 15 s and elongation at 72°C for 15 s during 30 cycles. The resulting DNA was purified by 5 M sodium chloride salt and isopropanol precipitation. Pellets were washed with 70% ethanol and re-suspended in 1× Tris-EDTA buffer. RNA synthesis. The transcription reaction mixture was assembled from the MAXIscript® T7 Kit (Ambion No. AM1314M, Life Technologies, Grand Island, NY). We added 1 μg of PCR product DNA, 2 μL of 10× transcription buffer, 2 μL of T7 Enzyme Mix, and 1 μL of each (10 mM) dNTP and adjusted the volume to 20 μL with nuclease-free water. The reaction mixture was incubated at 37°C for 2 h. RNA was purified using ammonium acetate and ethanol precipitation and resuspended in nuclease-free water. The concentration of RNA was determined spectrophotometrically at 260 nm using a Synergy HT (BioTek, Winooski, VT) platereader.
Materials and Methods Cloning. For cloning, a total of 2 μg of human fetal heart RNA (No. 540165, Agilent Technologies, Inc, Santa Clara, CA) was used for each reaction. The CloneMiner™ II cDNA Library Construction Kit (No. A11180, Invitrogen, Carlsbad, CA) was used to create individual clones. First and second DNA strands were synthesized from template RNAs and ligated into the pDONR222 vector. The pDONR222 vector contains a kanamycin resistance gene which allows for selection of transfected bacteria and the ccdB gene which interferes with Escherichia coli DNA gyrase allowing for negative selection of the donor vector in E. coli following recombination and transformation. Plated cells were incubated overnight at 37°C. Individual colonies containing the vector with cloned genes (over 4000 genes from fetal human heart) were collected and transferred into snap-cap tubes with 2 mL of 2xYT medium containing 50 μg/mL of
qRT-PCR. RNA was extracted using a NucleoSpin RNAII Kit (Macherey-Nagel, Bethlehem, PA) from differentiated cells treated with the active RNA, and one control untreated (treated only with lipofectamine). Quantitative reverse transcription polymerase chain reaction (qRT-PCR) was performed with a Rotor-Gene machine using a Rotor-Gene SYBR PCR kit (No. 204074, Qiagen, Valencia, CA) with primers as designed in our earlier studies (Zhang et al. 2009). Stem cell culture and differentiation protocol. Human-induced pluripotent stem cells (iPSCs), DF19-9–11T.H, from WiCell, Inc. (Madison, WI) and mouse embryonic stem cells (mESCs), Strain 129, OriCell from Cyagen Biosciences, Inc. (Santa Clara, CA) were incubated and grown at 37°C in a humidified 5% CO2 atmosphere, and passaged routinely according to our routine protocols (Lemanski et al. 2012). To promote the formation of embryoid bodies, 20uL hanging
RNA PROMOTES CARDIOGENESIS
drops of cell suspensions were micropipetted onto the inner surface of a Petri dish lid and cultured for 2 d, then washed in medium and transferred to gelatin-coated dishes. When the cells began to proliferate and spread, they were transfected with active clones of RNA mixed with Lipofectamine RNA:Max (Life Technologies) diluted to a concentration of 50 ng/ul in OPTI-MEM medium (Life Technologies) and incubated for 6 h. Fixation, staining, and confocal microscopy of cultured stem cells. CIR-treated and non-inducing RNA-treated or untreated control cells in culture were fixed in 4% paraformaldehyde for 30 min, rinsed in phosphate-buffered saline (PBS) with 3% bovine serum albumin (BSA) for 3 min, permeabilized in 0.1% Tween-20 and 3% BSA, and stained with primary antibody diluted 1:75 with PBS overnight. The cells were then rinsed with PBS and 3% BSA for 3 min and stained with the secondary anti-mouse antibody diluted to 1:75 with PBS for 1 h. The cells were immunofluorescently stained for tropomyosin, cardiac troponin-T, or α-actinin. The primary antibodies used for all three proteins were monoclonal antibodies from mouse, and the secondary antibodies were Goat F (ab) antimouse polyclonal antibodies with a fluorescein isothiocyanate (FITC) tag excited at 490 nm (Abcam, Cambridge, MA). The cells were analyzed using a confocal microscope to identify and localize the presence of tropomyosin, cardiac troponin-T, and α-actinin in the cells, including those proteins present in organized sarcomeric myofibrils. The immunofluorescently stained cells were analyzed using an Olympus BX62 scanning laser confocal microscope. Sequencing and secondary structure prediction. Specific RNA molecules were determined to have rescuing abilities initially by using our mutant axolotl heart bioassay system (Zhang et al. 2009). That specific cloned RNA then was sent to Functional Biosciences (Madison, WI) to determine its original DNA sequence. The DNA sequence associated with the rescuing RNA was screened for vector contamination using the protocols on the NIH Website http://www.ncbi. nlm.nih.gov/VecScreen/VecScreen.html, and then it was trimmed to remove contamination. The online NCBI BLAST program was utilized to determine the exact sequence within the human genome, and the sequence editor database at http://www.fr33.net/seqedit.php was used to convert the DNA sequence into the RNA sequence. The resulting RNA sequence was entered into the Genebee RNA computational software secondary structure prediction program developed in the Belozersky Institute at Moscow State University, Russia. This software was used to predict likely secondary structures for the RNA. The RNA’s
secondary structure was then compared with the secondary structure of the axolotl MIR to determine similarities.
Results Analysis of pluripotency in human-induced pluripotent stem cells and mouse embryonic stem cells in culture. When the human-induced pluripotent stem cells, DF19-9–11T.H (WiCell, Inc.) and mouse embryonic stem cells, Strain 129 (OriCell from Cyagen Biosciences, Inc.), were placed in culture and incubated with the human iPSCs (Fig. 1A) and mouse ESCs (Fig. 1B), they formed monolayer growth colonies that were rounded in shape and tightly packed together with distinct, well-defined edges characteristic of undifferentiated cells. When stained with antibodies for the human iPSCs Oct-3/4 pluripotency factor, both human iPSCs (Fig. 1C) and mouse (not illustrated) stain red. When the cells are also double stained with the blue nuclear stain, DAPI (Fig. 1D), it is clear from the merged image (Fig. 2E) that most of the cells (appearing purple) express the Oct-3/4 in the nuclei, thus confirming pluripotency of the stem cells at this early stage of culture. Formation of embryoid bodies. The stem cell suspensions placed on the inner surface of Petri dish lids in Bhanging drops^ (Fig. 2A) form into distinct clumps (embryoid bodies) after 2 d in the hanging drop cultures (Fig. 2B, C). The embryoid bodies then are washed in medium and transferred to gelatin-coated culture dishes, and after 4–5 d in these cultures, the cells from the embryoid bodies begin to attach to the substratum (Fig. 2D). At this point, the cells are transfected with active clones of RNA mixed with Lipofectamine RNA:Max to ensure uniform entry into the cells. Mutant axolotl heart bioassays to identify active human RNA clones. Initial bioassays were conducted by combining and pooling groups of RNAs in which mutant salamander nonbeating hearts were organ cultured (Moses-Arms et al. 2014). Control experiments involved organ culturing mutant hearts with no treatment and normal hearts with and without treatment. Pooled RNA groups that caused the treated mutant hearts to begin beating by the 2nd–4th d of treatment were separated into smaller testing groups until each individual RNA had been checked for rescuing ability. Within the 396 clones tested, one clone that showed significant rescuing ability and caused the non-beating mutant hearts to beat vigorously was clone No. 6 (now called CIR). In the present study, we describe and characterize the CIR derived from clone No. 6 and test for the CIR’s ability to promote the differentiation of non-muscle mouse embryonic and human-induced pluripotent stem cells into definitive cardiomyocytes.
KOCHEGAROV ET AL. Figure 1. A, B Nondifferentiated colonies of human iPSCs (A) and mouse ESCs (B). C Immunohistochemical staining in red (rhodamine) of human iPSCs with antibodies for the pluripotency marker Oct-3/4. D Nuclei stain in blue with DAPI in these human iPSCs. E Combined images of C Oct-3/4 and D DAPI show that the majority of human iPSCs express the factor of pluripotency Oct-3/4 which colocalizes in most of the cell nuclei stained with DAPI (purple shows co-localization).
Confocal microscopy of positive control axolotl hearts. Analysis of untreated normal axolotl embryonic hearts show well-organized sarcomeric cardiac myofibrils (Fig. 3A) Figure 2. A Plating of stem cells in hanging drops on the inside of a Petri dish lid. B Day 2. Formation of clumps (embryoid bodies) in the suspension. C High magnification of EB in the suspension. D Attachment and spreading of cells on culture dish after Day 5.
with antibody staining for tropomyosin. Untreated mutant hearts cultured in Holtfreter’s solution without CIR treatment showed virtually no staining with anti-tropomyosin
RNA PROMOTES CARDIOGENESIS
Figure 3. Tropomyosin expression revealed by immunochemical staining followed by confocal imaging. A Normal non-treated embryonic axolotl heart. B.Mutant non-treated embryonic axolotl heart. C Mutant embryonic axolotl heart after transfection with CIR derived
from fetal human heart RNA clones. The normal and CIR-treated mutant hearts show extensive staining for tropomyosin. The untreated mutant heart shows essentially no staining.
antibodies, and no myofibrillar structures could be identified in these untreated, undifferentiated mutant heart cells (Fig. 3B). Mutant axolotl hearts treated with the CIR fetal heart human RNA showed well-organized myofibrils to the extent that these rescued mutant myocardial cells (Fig. 3C) appeared indistinguishable from the normal control hearts compare (Fig. 3A, C). Thus, the CIR clearly has caused the cardiac mutant axolotl heart cells to form an undifferentiated embryonic, non-cardiac phenotype into normally appearing, vigorously contracting differentiated cardiac muscle cells containing myofibrils of normal morphology. The CIR derived from clone No. 6 rescues the mutant embryonic axolotl hearts in a manner very similar to the MIR derived from normal (+/+ or +/c) axolotl embryonic anterior endoderm that has been previously described (Zhang et al. 2009). It is very clear that the axolotl MIR and the human clone No. 6 CIR serve as functional homologues in the cardiac mutant axolotl rescue bioassay and both RNAs have the ability to promote the differentiation of beating cardiac tissue from undifferentiated mutant embryonic heart cells.
into the online NCBI BLAST program to determine for what protein the sequence coded, and it was identified as a fragment of N-sulfoglucosaminesulfohydrolase and the caspase recruitment domain family member 14 precursor. Using a sequence converter, the DNA sequence was converted to the corresponding RNA sequence (Fig. 4B). The possible secondary structures of the comparable regions of the human CIR (Fig. 5A), the normal axolotl MIR (Fig. 5B), and the mutant axolotl MIR (Fig. 5C) were determined using the Genebee RNA secondary structure prediction model. In comparing the active CIR from human (Fig. 5A) and the active MIR from normal (+/+ and +/c) axolotl (Fig. 5B), which have very similar secondary structures, with the non-active c/c mutant axolotl MIR (Fig. 5C), both human and normal axolotl have major differences in secondary RNA structure from the mutant. The human CIR and normal MIR both promote myofibril formation in and rescue mutant hearts. The c/c mutant RNA, with its significantly different secondary structure, lacks the capability of promoting myofibrillogenesis and inducing differentiation of the mutant embryonic hearts. Thus, the secondary structures of the RNAs appear to be a critical factor in the human CIR’s and the normal axolotl MIR’s abilities to induce cardiac myofibril cell formation and rescue the mutant hearts.
Sequencing and secondary structure of the active clone. PCR products and plasmids containing the DNA that corresponded to the CIR (clone No. 6 RNA) were sent to Functional Biosciences (Madison, WI) for determination of its DNA sequence (Fig. 4A). The trimmed DNA sequence was entered Figure 4. Sequences of CIR cloned as DNA (A) and converted into RNA (B).
KOCHEGAROV ET AL.
Figure 5. Secondary structure of CIR (A), active normal salamander MIR (B), and non-active mutant axolotl MIR (C). The CIR and active normal axolotl MIR have regions of very similar structure. Both differ from the non-active mutant RNA. Comparable areas are shown in boxes.
Analysis of CIR in promoting cardiomyocyte differentiation from mouse embryonic stem cells and human-induced pluripotent stem cells. Spontaneous cardiomyocyte differentiation of mouse and human stem cells has been described previously by Mummery et al. (2002). In the present study, stem cells were differentiated into cardiomyocytes after formation of embryoid bodies (EBs) by culturing the cells in the presence of cloned CIR. We tested this approach and found that
spontaneous Bbackground^ cardiomyocyte differentiation without RNA treatment was approximately 9–10%. We passaged stem cells and plated them into small drops as Bhanging drops^ on Petri dish lids and incubated them for 2 d. On the second day, stem cells aggregated and formed EBs. We plated the EBs on collagen-coated dishes and allowed them to attach. Cells from EBs started to grow and proliferate. At this stage, we treated them with the CIR and in 7–8 d, 70–80% had differentiated into cell shapes characteristic for cardiomyocytes (Fig. 6A, B) and expressed cardiac-specific troponin-T as revealed by antibody staining methods (Fig. 7A, B). Non-treated control human iPSCs or mouse ESCs showed only 9–10% of the cells with cardiac morphological traits (Fig. 6C, D) at this 7–8-d stage in culture; moreover, there was very little detectable cardiac staining for cardiac-specific troponin-T in any of the non-treated human or mouse stem cells (Fig. 7C, D). High-resolution confocal imaging illustrated myofibril organization in the stem cell-derived cardiomyocytes from iPSCs and from mouse embryonic stem cells (ESC) transfected with active CIR when stained with antibodies against cardiac-specific troponin-T, tropomyosin, or αactinin (Fig. 8A–C). When the RNAs from stem cell-derived cardiomyocytes and from non-differentiated stem cells were extracted and expression of cardiac-specific messenger RNA (mRNA) quantified with qRT-PCR (Fig. 9), it was shown that the CIR treatment very significantly increased expression of the cardiac markers, cardiac troponin-T, and tropomyosin, in comparison with the untreated cells in ESC or iPSC. The expression of cardiac-specific troponin-T in human stem cells and mouse embryonic stem cells far exceeded the expression in untreated mouse or human cells. In fact, it was in the range of 7–8-fold higher in the CIR-treated cells after only 7–8 d in culture (Fig. 9). Tropomyosin was also significantly higher in the CIR-treated stem cells, approximately 5-fold higher (Fig. 9). Also, when we screened the cloned sequence in the human genome with BLAST at the NCBI database, we found two high score matches with a portion of exon 8 of the human N-sulfoglucosaminesulfohydrolase (SGSH) gene on the sense strand of DNA and with the caspase recruitment domain family, member 14 (CARD14), on the antisense strand. These genes, SGSH and CARD14, partially overlap and belong to opposite DNA strands (forward and reverse) on human chromosome 17.
Discussion We have discovered a specific RNA, originally identified from commercially obtained human fetal heart RNA, which has the ability to turn undifferentiated non-muscle cells into definitive
RNA PROMOTES CARDIOGENESIS Figure 6. Human iPSCs (A) and mouse ESCs (B) differentiated into cells with morphologies characteristic of cardiomyocytes after transfection with CIR. Human iPSC (C) and mouse ESC (D) controls, without transfection with CIR do not show myocardial cell morphologies.
cardiomyocytes. This was accomplished by initially preparing 396 human fetal RNA clones and analyzing them in pooled groups to evaluate their ability to promote myocardial cell Figure 7. Human iPSC-derived (A) and mouse ESC-derived (B) cardiomyocytes after transfection with CIR and immunostained for cardiac troponin-T; human iPSC (C) and mouse ESC (D) controls immunostained without transfection of CIR. The CIRtreated cells show an intense reaction for the cardiac-specific troponin-T while the untreated cultures show very little staining.
formation using our published cardiac mutant axolotl heart rescue bioassay system (Zhang et al. 2003, 2009). After these initial experiments to confirm the positive cardiomyogenic
KOCHEGAROV ET AL. Figure 8. Differentiated cardiomyocytes derived from human iPSCs treated with CIR and stained with antibodies against cardiac-specific troponinT (A), tropomyosin (B), and αactinin (C). The cells stain for all of these muscle-related proteins.
effect of clone No. 6 RNA (now called CIR), we tested CIR on mammalian stem cells in culture, both mouse ESC from Cyagen Biosciences, Inc. and human iPSC from WiCell, Inc.. In the present study, our results confirm that CIR has the ability to promote the differentiation of both mouse embryonic and human-induced pluripotent stem cells into definitive cardiac muscle cells. Immunoflourescent confocal microscopy demonstrates the presence of cardiac myofibrils that contain cardiac-specific troponin-T as well as tropomyosin in 70–80% of the CIR-treated cells. In addition, CIR-treated cardiomyocytes contain well-organized myofibrils often with definitive sarcomeric structures, both strong indicators that these cells indeed are myocardiocytes. Only 9–10% of untreated stem cells express cardiac-specific troponin-T and tropomyosin. It is very clear from our results that the CIR has a strong inducing effect in converting these non-muscle stem cells into definitive cardiac muscle cells. The mechanism for this CIR cardiogenic process will require further analysis to be fully understood. However, one interesting observation that suggests a possible mechanism for the action of the CIR relates to our finding that the cloned sequence in the human genome with BLAST at the NCBI database, has two high score matches with a portion of exon 8 of the human N-sulfoglucosaminesulfohydrolase (SGSH) Figure 9. Expression of cardiacspecific mRNA quantified with qRT-PCR in human iPSC-derived (A) and mouse ESC-derived (B) cardiomyocytes after transfection with CIR (CIR) compared with non-transfected controls (Con); p
