|Radical Acceleration of Nuclear Reprogramming By Chromatin Remodeling with the Transactivation Domain of MyoD|
From the September Edition of Stem Cells
Generation of induced pluripotent stem cells (iPSCs) is both slow and inefficient; the route from somatic target cell generally takes a minimum of 4 weeks and only 1 in a thousand target cells being reprogrammed. The proper reconfiguration of the chromatin landscape is deemed a potential obstacle in the reprogramming process; so much so that small molecule inhibitors which promote a more open chromatin configuration are becoming common place in many reprogramming protocols (Huangfu, Maehr et al and Huangfu, Osafune et al). One potential problem with this approach is the lack of specific chromatin changes; instead these inhibitors promote global chromatin changes. Specific chromatin changes occur due to the specific recruitment of epigenetic regulators to specific loci by transcription factors; and so this suggests that currently used transcription factors in reprogramming (Oct4, Sox2, Klf4, Myc and Nanog) may have limited means to reconfigure chromatin. Myod1 is a master transcription factor for skeletal myogenesis and can directly convert one cell type into another, as exemplified by its ability to generate myotubes from pigmented retinal epithelial cells (Choi et al). This suggests that Myod1 has a more potent ability to recruit epigenetic modifiers leading to activation of suppressed genes embedded in closed chromatin. This hypothesis has been now been tested by the laboratory of Nobuaki Kikyo from the Stem Cell Institute, University of Minnesota, USA and Laboratory of Animal Reproduction, Kinki University, Nara, Japan and is presented in the September Edition of Stem Cells (Hirai et al). Full-length mouse Oct4 (O) was fused with various fragments of mouse Myod1, excluding the basic helix-loop-helix (bHLH) domain to avoid activation of Myod1 target genes, and were co-transduced with a polycistronic retroviral vector encoding mouse Sox2, Klf4, and Myc (SKM) into MEFs derived from Oct4-GFP mice, allowing for the monitoring of the reprogramming process. Using this system, the authors demonstrate that that a specific chimeric Oct4-Myod1 protein can reprogram MEFs to an iPSC state more efficiently than OSKM.
Control experiments (MEFs transduced with OSKM) gave GFP positive colonies around day 10 with an estimated 0.08+0.09% of MEFs being converted into GFP-positive cells by day 18. The M3 region of MyoD encompasses the acidic transactivation domain (TAD) (amino acids 3-56) (Weintraub et al) and transduction of an M3-Oct4 fusion protein (M3O) alongside SKM drastically increased the percentage of GFP-positive colonies, with 5.10+0.85% of MEFs being transformed into GFP-positive cells by day 15, and approximate 50-fold increase. Further, GFP-positive colonies emerged while using M3O on day 5 and by day 12 3.6+0.5% of transduced cells formed GFP-positive colonies. 97% of M3O-SKM colonies were homogenously GFP-positive by day 7 compared to around 5% of colonies derived with OSKM obtained on day 12. Expression of endogenous Oct4, endogenous Sox2, and Nanog gradually increase during the initial week after emergence of GFP-positive colonies in the control transduction. In comparison, induction of M3O-SKM led to a massive increase in the levels of endogenous Oct4, endogenous Sox2, and Nanog which reached or exceeded levels observed in ESCs at the time of the emergence of GFP-positive colonies and remained at similar levels until day 30. This differential efficiency of transcriptional reprogramming was also evident upon analysis of three fibroblast-enriched genes (Thy1, Col6a2, and Fgf7), with these genes being more highly downregulated in a shorted period of time in the M3O-SKM cells compared to control. The pluripotency of M3O-iPSCs was verified by genome-wide transcript analysis which demonstrated highly similar gene expression in M3O-iPSCs and hESCs, teratoma formation in NOD/SCID mouse with the presence of various tissues derived from the three germ layers and the formation of chimeric mice, with some evidence of germline transmission.
Comparisons of transcription factor binding and changes to chromatin structure were then examined to understand how M3O-SKM facilitated nuclear reprogramming at the molecular level. This demonstrated that that chromatin at the Oct4 and Sox2 loci was remodelled toward an ESC pattern in the majority of M3O-induced cells during the first ten days after transduction, while chromatin in the majority of controls was not significantly altered. Remodelling the chromatin landscape from a somatic cell fate, MEFs, to an ESC state entails the removal of repressive chromatin marks and the deposition of permissive marks leading to a more open chromatin configuration. Initial transcription factor binding studies showed that the binding of Oct4 and Sox2 to the distal enhancer of Oct4 was low in controls at day 9 suggesting a more closed chromatin state not amenable for transcription factor binding. However, at day 3 in M3O-SKM cells, Oct4 was already bound to the Oct4 distal enhancer when no GFP-positive colonies had yet appeared, suggesting that the modulation of the chromatin environment to a more open configuration is an early event. Oct4- and Sox2-binding eventually reached a level comparable to that seen in ESCs on day 9 in the M3O-SKM cells. Increased chromatin remodelling towards an ESC like chromatin environment was observed in the M3O-iPSCs compared to control as indicated by higher sensitivity to DNAses, as a more open chromatin configuration would lead to increased amounts of non-nucleosomal DNA which becomes a target for DNAses. DNA methylation analysis demonstrated that cells induced with M3O-SKM showed two-fold more unmethylated CpG sites at the Oct4 promoter on day 9 compared to control, consistent with higher GFP expression and a more open chromatin configuration. The Paf1 complex, which is recruited to the distal enhancer of the Oct4 gene through binding to the Oct4 protein (Ding et al and Ponnusamy et al) and then generally moves to the coding region of the gene (Gerber and Shalatifard), displayed a gradual increase of binding to the distal enhancer and coding region of the Oct4 gene in M3O-iPSCs, but not in control, between days 3 and 9 following transduction. The Paf1 complex recruits the histone methyltransferase complex COMPASS, which catalyzes trimethylation of lysine 4 on histone H3 and this permissive modification, alongside two other markers for active genes, acetylation of lysines 9 and 14 on histone H3, were also increased specifically in M3O-iPSCs in the coding regions of the Oct4 and Sox2 genes. In addition, two histone modifications linked to gene repression, trimethylation of lysine 9 and lysine 27 on histone H3, were more drastically decreased in M3O-iPSCs compared to control. Again similar results were observed for Sox2.
Lastly, the effect of c-Myc removal from the reprogramming cocktail was tested. Previous studies have reported that iPSCs can be established without c- Myc, but OSK induction generally requires 3-4 weeks in the presence of feeders and gives a very low efficiency of iPSC production (0.01%) (Nakagawa et al and Wernig et al). However, the use of M3O alongside SK (M3OSK) led to a large increase in the speed and efficiency with which GFP-positive colonies form, with GFP-positive colonies evident without feeder cells by day 7 at an efficiency of 0.44%; over 40-fold more efficient than OSK. Analysis at day 9 demonstrated that 3.16% of M3OSK cells were GFP-positive but did not significantly decrease the overall level of DNA methylation. Oct4, Sox2, and one protein of the Paf1 complex, Cdc73, were still bound to the Oct4 enhancer, but the binding of Cdc73, Leo1, and Paf1, both also part of the Paf1 complex, to the initiation site of Oct4 was weak without c-Myc. Consistent with this partial assembly of the Paf1 complex at the Oct4 gene, the level of trimethylation of lysine 27 remained low without c-Myc, as did the levels of acetylation of lysine 9. Removal of repressive marks was also effected, with trimethylation of lysine 27 on histone H3 more resistant to demethylation whereas trimethylation of lysine 9 was effectively decreased. In comparison, Sox2 promoter dynamics remained similar between M3O-SK and M3OSKM. Together, these chromatin studies indicate that while M3O could facilitate formation of GFP-positive colonies without c-Myc, chromatin remodelling was retarded, suggesting that c-Myc has a major role in the modulation of the chromatin environment towards an ESC-like state during reprogramming.
M3O could also facilitate generation of human iPSCs, with NANOG-positive human ESC-like colonies emerging around day 8 with the number increasing by around day 15 with an efficiency of 0.30+0.033%. In contrast, when OSKM was transduced, NANOG-positive colonies did not emerge until around day 12 with an efficiency of 0.0052+0.0018%. Furthermore, while less than 10% of the colonies that appeared with OSKM were NANOG positive, more than 90% of the colonies produced with M3O-SKM were NANOG-positive. Human M3OSKM-iPSCs also expressed endogenous OCT4 and surface markers SSEA4, TRA-1-60 and TRA-1- 81 on day 28, demonstrated normal karyotypes and formed teratomas when injected into an NOD/SCID mouse proving pluripotency of the cells.
Overall, this exciting piece of research shows that the fusion of the transactivation domain to Oct4 increases the reprogramming process by promoting the reconfiguration of the chromatin landscape at the Oct4 and Sox2 gene loci. Fusion proteins such as this which boost the chromatin reconfiguration capability of a transcription factor promises to have wide reaching consequences. In iPSC production, it promises to boost efficiency, especially useful when the source of fibroblasts is in short supply or grow poorly due to an associated disease/syndrome and also to obviate the necessity for small molecule inhibitors of epigenetic modifying drugs. This is important as is stops non-specific global chromatin changes which could potentially activate genes non-beneficial to the reprogramming process. Further, it may also be important for direct differentiation strategies which do not go through the iPSC stage or reprogramming. Before these next steps, further analysis of global gene expression and chromatin changes are merited in both human and mouse iPSCs.
Choi J, Costa ML, Mermelstein CS, Chagas C, Holtzer S, Holtzer H.
Ding L, Paszkowski-Rogacz M, Nitzsche A, Slabicki MM, Heninger AK, de Vries I, Kittler R, Junqueira M, Shevchenko A, Schulz H, Hubner N, Doss MX, Sachinidis A, Hescheler J, Iacone R, Anastassiadis K, Stewart AF, Pisabarro MT, Caldarelli A, Poser I, Theis M, Buchholz F.
Gerber M, Shilatifard A.
Hirai H, Tani T, Katoku-Kikyo N, Kellner S, Karian P, Firpo M, Kikyo N.
Huangfu D, Osafune K, Maehr R, Guo W, Eijkelenboom A, Chen S, Muhlestein W, Melton DA.
Huangfu D, Maehr R, Guo W, Eijkelenboom A, Snitow M, Chen AE, Melton DA.
Nakagawa M, Koyanagi M, Tanabe K, Takahashi K, Ichisaka T, Aoi T, Okita K, Mochiduki Y, Takizawa N, Yamanaka S.
Ponnusamy MP, Deb S, Dey P, Chakraborty S, Rachagani S, Senapati S, Batra SK.
Weintraub H, Dwarki VJ, Verma I, Davis R, Hollenberg S, Snider L, Lassar A, Tapscott SJ.
Wernig M, Meissner A, Cassady JP, Jaenisch R.