|Home | About | Journals | Submit | Contact Us | Français|
Direct conversion of cardiac fibroblasts (CFs) into induced cardiomyocytes (iCMs) holds great potential for regenerative medicine by offering alternative strategies for treatment of heart disease. This conversion has been achieved by forced expression of defined factors such as Gata4 (G), Mef2c (M) and Tbx5 (T). Traditionally, iCMs are generated by a cocktail of viruses expressing these individual factors. However, reprogramming efficiency is relatively low and most of the in vitro G,M,T-transduced fibroblasts do not become fully reprogrammed, making it difficult to study the reprogramming mechanisms. We recently have shown that the stoichiometry of G,M,T is crucial for efficient iCM reprogramming. An optimal stoichiometry of G,M,T with relative high level of M and low levels of G and T achieved by using our polycistronic MGT vector (hereafter referred to as MGT) significantly increased reprogramming efficiency and improved iCM quality in vitro. Here we provide a detailed description of the methodology used to generate iCMs with MGT construct from cardiac fibroblasts. Isolation of cardiac fibroblasts, generation of virus for reprogramming and evaluation of the reprogramming process are also included to provide a platform for efficient and reproducible generation of iCMs.
Cardiovascular disease remains the leading cause of death worldwide, accounting for 17.3 million deaths per year1. Loss of cardiomyocytes resulting from myocardial infarction (MI) or progressive heart failure is a major cause of morbidity and mortality2. Due to limited regenerative capacity, adult mammalian hearts usually suffer from impaired pump function and heart failure following injury3–6. As such, efficient (re)generation of cardiomyocytes in vivo and in vitro for treatment of heart disease and for disease modeling is a critical issue needing to be addressed.
Recent development of direct reprogramming, which directly reprograms cells from one differentiated phenotype to another without transitioning through the pluripotent state, offers a promising alternative approach for regenerative medicine. The mammalian heart contains abundant cardiac fibroblasts (CFs), which account for approximately half of the cells in heart and massively proliferate upon injury7–9. Thus, the vast pool of CFs could serve as an endogenous source of new CMs for regenerative therapy if they could be directly reprogrammed into functional CMs. It has been shown that a combination of transcription factors, such as Gata4 (G), Mef2c (M) and Tbx5 (T), with or without microRNAs or small molecules can reprogram fibroblasts into iCMs10–26. Importantly, this conversion can also be induced in vivo, and results in an improvement in cardiac function and a reduction in scar size in an infarcted heart16,27–29. These studies indicate that direct cardiac reprogramming may be a potential avenue to heal an injured heart. However, the low efficiency of iCM reprogramming has become a major hurdle for further mechanistic studies. In addition, the reproducibility of cardiac reprogramming is another controversial issue of this technology11,30,31.
Very recently, we generated a complete set of polycistronic constructs encoding G,M,T in all possible splicing orders with identical 2A sequences in a single mRNA. These polycistronic constructs yielded varied G, M and T protein expression levels, which led to significantly different reprogramming efficiency25. The most efficient construct, named MGT, which showed a relatively high Mef2c and low Gata4 and Tbx5 expression, significantly improved reprogramming efficiency and produced large amounts of iCMs with CM markers expression, robust calcium oscillation and spontaneous beating25. Moreover, by using MGT polycistronic construct, our study avoided the use of multiple vectors and generated cells with homogenous expression ratio of G,M,T, thus providing an improved platform for cardiac reprogramming research. To increase experimental reproducibility, here we describe in detail how to isolate fibroblasts, produce retrovirus carrying MGT cassette, generate iCMs and evaluate the reprogramming efficiency.
The protocol outlined here uses neonatal mice. Animal care and experiments are performed in accordance with the guidelines established by The Division of Laboratory Animal Medicine (DLAM) at University of North Carolina, Chapel Hill.
Note: Perform the following steps in a BSL2 Biological Safety Cabinet under sterile conditions. The proper disposal of transfected cells, pipette tips and tubes is recommended to avoid risk of environmental and health hazards.
The reprogramming steps are summarized by schematic in Figure 1. After MGT transduction, GFP expression in reprogramming cells could be detected as early as day 3. Puromycin selection of transduced cells starts from day 3 and is maintained during the first two weeks if pMx-puro-MGT construct is used. By day 10 to day 14, expression of cardiac markers like cTnT and αActinin could be detected by both ICC (Figure 2B, step 5) and FACS (Figure 2A, step 6), indicating that the starting fibroblasts are undergoing reprogramming toward a cardiac cell fate.
For successful iCM generation when using this protocol, there are several important factors that have an impact on the overall efficiency. Particularly the conditions of starting fibroblasts and the quality of retrovirus encoding MGT can greatly affect the reprogramming efficiency.
It is important to generate fibroblasts as fresh and healthy as possible. For explant culture method, fibroblasts can be used before seven days after the explants were plated on dishes. For enzyme digestion method, fibroblasts can be used as early as the next day of isolation. Freezing or passaging the fibroblasts for reprogramming is not recommended. These additional treatments of fibroblasts will significantly decrease the reprogramming efficiency. The seeding density is another important factor that affects the reprogramming efficiency. If cells are sparsely seeded, they tend to become unhealthy with irregular cell morphology and in large size. Seldom could those cells be converted into iCMs. If cells are seeded too densely, the MOI (multiplicity of infection) for viral infection is decreased. Overgrowth of uninfected fibroblasts significantly dilutes the reprogramming events thus resulting in a low percentage of iCMs. We also notice the emergence of smaller cells in cobblestone-like morphology if more cells are seeded for reprogramming. They could hardly be reprogrammed and thus result in reduced reprogramming efficiency. When modifying the protocol for different size of tissue culture dishes, it is recommended that seeding cell numbers be adjusted proportionally to the surface area of the culture dish.
The purity of fibroblasts is critical for reprogramming. The high purity of fibroblasts could be achieved either by FACS sorting15 or MACS enrichment as we described here. Compared to FACS sorting, MACS sorting is easier, gentler, and more convenient. Cell purity after MACS could reach over 90% when strictly following the protocol. Low purity of fibroblasts may decrease reprogramming efficiency since the contaminated cells are harder to be reprogrammed than fibroblasts and that they proliferate much faster than reprogramming cells. It is also highly recommended to seed the cells right after MACS and perform viral infection overnight after the seeding. The sorted fibroblasts may partially become senescent after growing in dish for a long time, which decreases uptake of the retrovirus that only infects the proliferating cells. Tail-tip fibroblasts (TTFs) and embryonic fibroblasts (MEFs) have been reported to be converted into iCMs with some other transcription factors10,12,19,26. Our protocol here only focuses on cardiac fibroblasts. The application of this protocol to other fibroblast types may be further optimized due to the inherent variations of fibroblast signatures such as the proliferation status, genetic and epigenetic landmarks.
The quality of retrovirus for transduction is of utmost importance. Low titer viruses fail to convert fibroblasts into iCMs. Whereas viruses at excessive amount will cause cytotoxicity that directly leads to fibroblast cell death. It is necessary to determine the viral titer and optimize the viral dosage for reprogramming based on laboratory set-up and experimental conditions. The method for viral titration has been thoroughly described before23. In fact we found that the growth condition of Plat-E cells directly relates to viral quality. It is recommended to thaw Plat-E cells at lower passage (<30) each time. Cells obtained at passage three are best for packaging virus at a density around 4–5 x 106 cells per 10 cm culture dish. Additionally, the freshly harvested viruses yield higher reprogramming efficiency than frozen viruses. Please note that co-infection of MGT virus with another pMxs vector based retrovirus will decrease the reprogramming efficiency possibly due to the competition between the two viruses of the same type. Two different transfection methods are provided here. Both methods produce comparable retrovirus. While commercial Lipofectamine is costly but stable, transfection with calcium phosphate is cheaper. Preparation of HBS buffer needs to take great caution because the pH of HBS buffer significantly affects the transfection efficiency.
There are some limitations of this protocol that need to be addressed. One of the limitations lies in the usage of retrovirus, which inevitably renders to biosafety issues. The heterogeneity of fibroblasts and the lack of specific cardiac fibroblast marker for isolation also affect the purity of cells that are reprogrammable. Variation of reprogramming efficiency may be observed due to batch variability inherent to fibroblast condition and virus production. In addition, although the reprogrammed cells express sarcomere structure proteins such as cardiac troponin T and αActinin, they are still not mature cardiomyocytes characterized by functional properties such as contractility and electrical propagation.
In summary, the methodology described here allows efficient generation of iCMs based on retroviral delivery of M,G,T transcription factors in a single construct. Our protocol provides a reproducible and valuable platform for iCM research, and will facilitate ongoing efforts on high-throughput screening and mechanistic studies of iCM reprogramming, and ultimately move iCM reprogramming field closer to clinical applications.
We are grateful for expert technical assistance from the UNC Flow Cytometry Core and UNC Microscopy Core. We thank members of the Qian lab and the Liu lab for helpful discussions and critical reviews of the manuscript. This study was supported by NIH/NHLBI R00 HL109079 grant to Dr. Liu and American Heart Association (AHA) Scientist Development Grant 13SDG17060010 and the Ellison Medical Foundation (EMF) New Scholar Grant AG-NS-1064-13 to Dr. Qian.
The video component of this article can be found at http://www.jove.com/video/53426/
The authors have nothing to disclose.