You are here

| Direct Reprogramming

Who Needs Pluripotency? - Direct Lineage Conversion of Terminally Differentiated Hepatocytes to Functional Neurons

From Cell Stem Cell
By Stuart P. Atkinson

Direct conversion of one somatic cell to another somatic cell type, completely bypassing the pluripotent stage through the forced expression of lineage specific transcription factors has emerged as a large “splinter group” of research, taking many lessons from induced pluripotent stem cell (iPSC) technology. The direct generation of induced neuronal cells (iN) from human fibroblasts has been previously demonstrated in several papers (Ambasuhan et al, Caiazzo et al, Pang et al, Pfisterer et al, Qiang et al, Son et aland Yoo et al.) however fibroblasts represent a heterogeneous mixture of cells, potentially including cells of the neural crest, and so the reprogrammed cell of origin remains undefined. Therefore, researchers from the group of Marius Wernig at the Stanford University School of Medicine, USA, decided to attempt to identify a specific somatic cell type from one germ lineage and reprogram these cells across the germ layer barrier into iN cells. The study, published as a short article in Cell Stem Cell demonstrates the direct conversion of mouse hepatocytes to iN cells and analyses the reprogramming process, demonstrating the faithful silencing of the hepatocyte expression program and the expression of the neuronal expression program (Marro et al).

Primary liver cultures from postnatal days 2–5 wild-type and Tau-EGFP knock-in mice (with Tau expression to identify neuronal cells) were established and four days after isolation, the majority of cells showed a typical epithelial morphology and expressed liver associated genes (Alb, Afp and a-antitrypsin). Primary cultures were typically composed of 60% Albumin-positive hepatocytes with a general absence of neuronal or neural progenitor cell markers (Sox2, Brn2, Mtap2, and NeuN), but with rare (1/5000) Tuj1 positive cells, although these did not have typical neuronal cell morphology. Tau-EGFP-positive cells were not detectable, but upon replating of primary liver cultured cells, infection with doxycycline (dox)-inducible lentiviruses containing the cDNAs of Brn2 (B), Ascl1 (A) and Myt1l (M) and 13 days of dox treatment, Tau-EGFP/Tuj1-positive cells with a complex neuronal morphology were readily detected. These cells were also positive for PSA-NCAM (Poly-Sialated Neural Cell Adhesion Molecule), NeuN, Mtap2, and Synapsin with a fraction of cells positive for the vesicular glutamate transporter 1 (vGlut1), but were negative for Gad1, Th, Chat, or serotonin, suggesting that these induced neuronal cells from hepatocytes (Hep-iN cells) were excitatory neurons. QPCR analysis confirmed that 3 weeks after infection, the iN cells had not only induced neuronal transcripts, but efficiently silenced transcripts characteristic of the starting hepatocyte population.

Next, using a cross between Albumin-Cre mice, in which hepatocytes in both foetal and adult mice are labelled, and ROSA26-mTmG reporter mice, which express membranous tdTomato before and membranous EGFP after Cre-mediated recombination, it was definitively demonstrated that Hep-iN cells are derived from Albumin-expressing hepatocytes. From a starting population of 60% EGFP-positive (hepatocytes) and 40% tdTomato-positive cells, reprogramming with BAM factors and 13 days of dox induction produced both red and green fluorescent cells with neuronal morphology, with EGFP-positive cells also expressing the neuronal markers Tuj1 and PSA-NCAM.

Further, it was demonstrated that Hep-iN cells could acquire the functional properties of mature neurons. The average resting membrane potential of Hep-iN cells was around 50 mV, spontaneous action potentials were detected in half of the cells and all analyzed Hep-iN cells generated action potentials when depolarized by current injections, showing fast inactivation sodium current and outward potassium currents. When Hep-iN cells were FACS-sorted 7 days after dox treatment and cultured together with mouse cortical neuronal cultures for another 4 weeks, postsynaptic responses could be evoked by extracellular stimulation of surrounding neurons. At holding potentials of 70 mV, a small inward current was detected, presumably mediated by AMPA receptors and/or GABAA receptors, while at 60 mV a large outward current was evoked, presumably mediated by NMDA and/or GABAA receptors.

To allow analysis of iN reprogramming, triple transgenic mice (TauEGFP allele together with Albumin-Cre and a ROSA26-tdTomato reporter) were generated, allowing only Albumin-positive hepatocytes and their progeny (the iN cells) to express tdTomato. As expected, 13 days after BAM transduction Tau-EGFP/tdTomato-double positive Hep-iN cells appeared but, surprisingly, Tau-EGFP was expressed in some hepatocytes after only one day. Over time the generation of Tau-EGFP-positive cells steadily increased, with similar kinetics for hepatocytes observed for a control mouse embryonic fibroblast (MEF) reprogramming experiment and at day 13, efficiency of reprogramming was intermediate between postnatal fibroblasts (6%) and embryonic fibroblasts (20%). Additionally, it was also shown that iN cell reprogramming could be established in mature hepatocytes (1 year), although with decreased efficiency (2.7% ± 1.4%). The transcriptional profiles of iN cells from hepatocytes at different days post-reprogramming (Day 13 and Day 22), MEFs and tail tip fibroblasts (TTFs) were analysed, alongside primary neonatal cortical neurons (CNs) and neurons derived from fetal (E13.5) forebrain neural progenitor cells (NPCs) at 7 and 13 days after differentiation. These were sorted for Tau-EGFP expression and compared through unsupervised hierarchical clustering of all samples based on 12,275 genes. Most iN samples clustered with the primary neuron sample, although NPC-derived neurons were more similar to two iN cell populations (d22 MEF-iN and d22 TTF-iN) demonstrating that the transcriptional variability between two different primary neuronal populations was greater than that between iN cells and a specific population of primary neurons. The fact that on day 22 fibroblast-iN cells are more like primary neurons, d22 Hep-iN cells and d13 fibroblast-iN cells are more like primary neurons than to fibroblasts or hepatocytes, and d13 Hep-iN cells, are more similar to hepatocytes than they are to primary neurons, suggests that hepatocytes are harder to reprogram and need more time to induce a complete neuronal program when compared with fibroblasts.

Through analysis of donor-specific expression it was also shown that the iN cells also faithfully silence donor-expression programs while up-regulating neuronal-expression programs. The MEF signature (221 probes) showed a 95% (on day 13) and 91% (by day 22) downregulation at least 2-fold in MEF-iN, while the liver-specific signature showed a 76% (day 13) and 85%(day 22) downregulation in HEP-iN cells, again suggesting that hepatocytes may be more difficult to faithfully reprogram. However, Hep-iN cells completely lost hepatocyte-specific functional properties such as Albumin secretion and urea production and single cell reprogramming assays found robust expression of pan-neuronal markers in 27/28 Hep-iN cells (b-III-tubulin, Map2, Ncam). Surprisingly, many primary neurons expressed some of the eight analyzed liver signature genes, while Hep-iN cells were randomly positive for one or more liver markers illustrating the transcriptional noise of assumed cell type-specific genes. Thus, while some Hep-iN cells appear to be indistinguishable from primary neurons, there is a trend that Hep-iN cells continue to express low levels of liver-associated genes, suggesting an epigenetic memory of their cell of origin, but the lack of detectable hepatic functional properties suggests that this epigenetic memory has little if any functional consequence.

This paper demonstrates that direct programming can occur across germ layers by the creation of functional ectodermal cells from functional endodermal cells, bypassing the pluripotent stage completely. But is this bypass wise? A recent study has shown that aged cells that are reprogrammed are “given back their youth” (Lapasset et al), and that by directly reprogramming cells, this stage may be missed to the detriment of the cell type generated. Therefore, functional characterisation, survival and senescence need to be assessed for iN generated in this manner.



Ambasudhan, R., Talantova, M., Coleman, R. et al
Direct Reprogramming of Adult Human Fibroblasts to Functional Neurons under Defined Conditions.
Cell Stem Cell 9, 113–118, 2011.

Caiazzo, M., Dell’anno, M.T., Dvoretskova, E. et al
Direct generation of functional dopaminergic neurons from mouse and human fibroblasts.
Nature 476, 224–227, 2011

Lapasset L., Milhavet O., Prieur A. et al
Rejuvenating senescent and centenarian human cells by reprogramming through the pluripotent state
Genes Dev. November 1, 25: 2248-2253, 2011.

Pang, Z.P., Yang, N., Vierbuchen, T. et al.
Induction of human neuronal cells by defined transcription factors.
Nature 476, 220–223, 2011

Pfisterer, U., Kirkeby, A., Torper, O. et al.
Direct conversion of human fibroblasts to dopaminergic neurons.
Proc. Natl. Acad. Sci. USA 108, 10343–10348, 2011.

Qiang, L., Fujita, R., Yamashita, T. et al.
Directed conversion of Alzheimer’s disease patient skin fibroblasts into functional neurons.
Cell 146, 359–371, 2011.

Son, E.Y., Ichida, J.K., Wainger, B.J. et al.
Conversion of mouse and human fibroblasts into functional spinal motor neurons.
Cell Stem Cell 9, 205–218, 2011.

Yoo, A.S., Sun, A.X., Li, L. et al.
MicroRNA-mediated conversion of human fibroblasts to neurons.
Nature 476, 228–231, 2011.