Author contributions: A.A.R. and D.A.P.: conception and design, collection and/or assembly of data, data analysis and interpretation, manuscript writing, and final approval of manuscript; A.A.F.V.: conception and design, data analysis and interpretation, manuscript writing, and final approval of manuscript; M.J.S.: financial support, data analysis and interpretation, manuscript writing, and final approval of manuscript; D.E.A.: conception and design, financial support, data analysis and interpretation, manuscript writing, and final approval of manuscript.
Disclosure of potential conflicts of interest is found at the end of this article.
First published online in STEM CELLSEXPRESS September 17, 2012.
Cardiomyogenic differentiation of stem cells can be accomplished by coculture with cardiomyocytes (CMCs). To facilitate their identification, stem cells are often labeled through viral transduction with a fluorescent protein. A second marker to distinguish stem cell-derived CMCs from native CMCs is rarely used. This study aimed to investigate the occurrence of secondary transduction of unlabeled neonatal rat (nr) CMCs after coculture with human cells that had been transduced 0, 7, or 14 days earlier with a vesicular stomatitis virus (VSV) G protein-pseudotyped lentiviral vector (LV) encoding enhanced green fluorescent protein (GFP). To reduce secondary LV transfer, GFP-labeled cells were incubated with non-heat-inactivated human serum (NHI) or with VSV-neutralizing rabbit serum (αVSV). Heat-inactivated human serum and normal rabbit serum were used as controls. Immunostaining showed substantial GFP gene transfer to nrCMCs in cocultures started at the day of transduction indicated by the presence of GFP-positive/human lamin A/C-negative nrCMCs. The extent of secondary transduction was significantly reduced in cocultures initiated 7 days after GFP transduction, while it was completely abolished when human cells were added to nrCMCs 14 days post-transduction. Both NHI and αVSV significantly reduced the occurrence of secondary transduction compared to their controls. However, under all circumstances, GFP-labeled human cells had to be passaged for 14 days prior to coculture initiation to prevent any horizontal GFP gene transfer to the nrCMCs. This study emphasizes that differentiation experiments involving the use of viral vector-marked donor cells should be interpreted with caution and describes measures to reduce/prevent secondary transduction. STEM CELLS 2012;30:2830–2834
Whether somatic stem cells (SSCs) can undergo cardiomyogenic differentiation without genetic intervention is a topic of much debate [1, 2]. At least part of the confusion may relate to application of different criteria and methods to identify SSC-derived cardiomyocytes (CMCs) and to the use of SSCs from different sources and differently aged donors [2–4]. Intramyocardial transplantation and coculture with CMCs are commonly used to investigate cardiomyogenic differentiation potential of SSCs. However, these studies often do not include the use of species-, strain-, or gender-specific markers to unambiguously demonstrate derivation of a CMC from a stem cell. Instead, stem cells are transduced with a viral vector encoding a fluorescent protein and the appearance of cells coexpressing the fluorescent protein and one or more CMC markers is taken as proof for their cardiomyogenic differentiation [5–11]. An often neglected pitfall of these studies is secondary transduction of CMCs by viral vector-marked stem cells. Pan et al. previously showed that hematopoietic target cell-associated vesicular stomatitis virus (VSV) G protein-pseudotyped lentiviral vector (LV) particles can transduce neighboring cells . In this study, secondary transduction of neonatal rat (nr) CMCs by adult human cells that had previously been transduced with an enhanced green fluorescent protein (GFP)-encoding LV was studied. Also, options to abolish secondary transduction were investigated including incubation of LV-GFP-treated cells with VSV-specific antiserum or with normal human serum, which contains natural antibodies that can neutralize VSV in a complement-dependent manner .
MATERIALS AND METHODS
Human tissues were obtained with donors' written informed consent and with approval of the Medical Ethics Committee of Leiden University Medical Center (LUMC). The study conformed to the principles of the Declaration of Helsinki. Animal experiments were approved by LUMC's Animal Experiments Committee and conformed to the Guide for Care and Use of Laboratory Animals (10236) .
Adult bone marrow (BM)- or adipose tissue (AT)-derived human mesenchymal stem cells (hMSCs) and human skin fibroblasts (hSFs) were transduced with GFP (multiplicity of infection 18 HeLa cell-transducing units per cell) using the VSV G protein-pseudotyped human immunodeficiency virus type 1 vector CMVPRES (hereinafter referred to as LV-GFP) [15, 16]. This specific vector dose resulted in transduction efficiencies of nearly 100% without causing overt cytotoxicity. The aforementioned cell types and nrCMCs were isolated and cultured as previously described [17–19]. nrCMC cultures typically contained ±10% cardiac fibroblasts and were mitomycin C-treated to prevent proliferation of the latter cell type . For a schematic overview of the experimental setup refer Figure 1. After 9 days of coculture, cells were subjected to immunostaining as previously described . Primary antibodies specific for human lamin A/C (clone 636, VP-L550, Vector laboratories, Burlingame, CA, http://www.vectorlabs.com), α-actinin (clone EA53, A7811), and connexin43 (C6219) (both from Sigma-Aldrich, Zwijndrecht, The Netherlands, http://www.sigmaaldrich.com) were used. These were visualized with Alexa Fluor-conjugated secondary antibodies, while nuclei were stained using Hoechst 33342 (all from Invitrogen, Breda, The Netherlands, http://www.invitrogen.com). Image J software (Institutes of Health, Bethesda, MD) was used to determine the number of GFP-positive/human lamin A/C-negative nrCMCs in randomly chosen regions (≥3,000 cells analyzed per condition). Human-specific quantitative reverse transcription-polymerase chain reaction (qRT-PCR) analyses were used to detect possible cardiomyogenic differentiation of the human cells after 9 days of coculture with nrCMCs .
Experimental results were expressed as mean ± SD for a given number (n) of observations. Data were analyzed by Student's t test for direct comparisons. Analysis of variance followed by appropriate post hoc analysis was performed for multiple comparisons. Statistical analysis was performed using SPSS 16.0 for Windows (SPSS, Chicago, IL, http://www.ibm.com). Differences were considered statistically significant at p < .05.
Cocultures initiated with freshly transduced adult BM hMSCs, AT hMSCs, or hSFs contained 51.2% ± 2.1%, 40.6% ± 3.8%, and 8.9% ± 0.6% of GFP-labeled nrCMCs, respectively (Fig. 2A, 2D). GFP gene transfer to nrCMCs was strongly reduced in cocultures started with human cells 7 days post-transduction (BM hMSCs: 11.1% ± 0.55%, AT hMSCs: 2.60% ± 0.2%, hSFs: 2.10% ± 0.1%; p < .01) (Fig. 2B, 2D) and completely abolished using human cells that had been transduced 14 days earlier (Fig. 2C, 2D). In none of the cocultures, expression of human ACTN2, TNNT2, NPPA, MYL2, MYL7, GATA4, or NKX2-5 was observed by qRT-PCR confirming previous findings that hSFs and adult hMSCs do not undergo cardiomyogenesis when coincubated with nrCMCs (Fig. 2E) .
Secondary transduction was inhibited by treatment of GFP-labeled adult BM hMSCs with non-heat-inactivated human serum (NHI) or VSV-neutralizing rabbit serum (αVSV) prior to coculture initiation. Compared to controls (heat-inactivated human serum and normal rabbit serum, respectively), incubation of the hMSCs with NHI or αVSV significantly reduced GFP gene transfer to nrCMCs in cocultures started at the day of transduction (p < .01) (Figs. 3Aa, 3Ba, 3C, 4Aa, 4Ba, 4C) or 7 days later (p < .01) (Figs. 3Ab, 3Bb, 3C, 4Ab, 4Bb, 4C), while hMSCs that had been transduced 14 days before coculture initiation did not give rise to any secondary transduction of nrCMCs (Figs. 3Ac, 3Bc, 3C, 4Ac, 4Bc, 4C).
DISCUSSION AND CONCLUSIONS
The key findings of this study are (a) hMSCs and, to a lesser extent, hSFs labeled with GFP by lentiviral gene transfer serve as a reservoir of functional LV particles causing secondary transductions of cocultured nrCMCs. (b) Passaging of the LV-GFP-transduced cells and frequent refreshment of culture medium prior to coculture initiation reduce and ultimately abolish GFP gene transfer to nrCMCs over time. (c) Incubation of the LV-GFP-transduced hMSCs with NHI or αVSV before coculture with nrCMCs greatly decreases GFP gene transfer to nrCMCs but does not shorten the time needed to completely eliminate secondary transduction. The strong inhibitory effect of NHI and αVSV on the frequency of GFP-labeled nrCMCs indicates that pseudotransduction or vesicle uptake is not the major contributor to the occurrence of GFP-positive nrCMCs in our coculture system . Furthermore, the absence in the cocultures of human lamin A/C-positive cells expressing CMC markers excludes heterocellular fusion as cause for the appearance of GFP-labeled nrCMCs. Our results are in line with those of Pan et al., who showed secondary transduction resulting from release of VSV G protein-pseudotyped LV particles by murine whole BM cells but did not investigate the effects of repeated passaging or treatment with αVSV or NHI of LV-transduced cells on horizontal gene transfer .
Cardiomyogenic Differentiation Versus Secondary Transduction
The ability of SSCs to undergo cardiomyogenesis remains a controversial topic. Coincubation with CMCs and intramyocardial transplantation are frequently used to investigate cardiomyogenic differentiation of SSCs [5–11]. In these experiments, the SSCs are often labeled with GFP using viral vectors before coincubation with nrCMCs and their cardiomyogenic differentiation is inferred from the occurrence of GFP-positive cells expressing CMC markers. In our laboratory, an antibody detecting human lamin A/C has been used as a second identifier besides GFP positivity to distinguish human cells from cocultured nrCMCs [17, 18]. Also, qRT-PCR with human-specific primer pairs can be used to assess cardiomyogenesis in human cells cocultured with nrCMCs [17, 18]. Using this approach, human CMC-specific gene expression was not observed even not under conditions that gave rise to high percentages of GFP- and α-actinin-double-positive cells. Our finding that coincubation of LV-GFP-treated human cells with untransduced nrCMCs leads to considerable secondary transduction of the latter cells highlights the importance of using an endogenous marker to assess stem cell differentiation in the presence of the differentiated cell type(s) to be generated. This is particularly relevant when the combination of viral vector and target cell does not allow elimination of residual gene transfer activity by repeated cell passaging/prolonged cell culture as is the case when using episomal viral vectors (e.g., adenoviral vectors) or postmitotic target cells.
This research forms part of Project P1.04 SMARTCARE of the BioMedical Materials (BMM) program, which is cofunded by the Dutch Ministry of Economic Affairs, Agriculture, and Innovation. This work was supported by the Dutch Heart Foundation (NHS), the Smart Mix Program (D.E.A.), and of the Netherlands Organisation for Scientific Research (NWO, Veni Grant 91611070, D.A.P.). We thank the Virology Division, Department of Infectious Diseases and Immunology, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands for providing us with VSV-neutralizing rabbit serum.
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
The authors indicate no potential conflicts of interest.