Notes on Direct Conversion of Human Fibroblasts into Neural Progenitors

Direct Conversion of Human Fibroblasts into Neural Progenitors

Introduction to the Research

This article, "Direct Conversion of Human Fibroblasts into Neural Progenitors Using Transcription Factors Enriched in Human ESC-Derived Neural Progenitors," published in Stem Cell Reports (ISSCR 2, OPEN ACCESS), details a method for streamlining the generation of neural progenitor cells (NPCs) for therapeutic and modeling applications.

The Challenge: Timelines for NSC Treatment

Traditional iPSC-Mediated NSC Generation Pipeline

Converting fibroblasts to Neural Stem Cells (NSCs) via induced Pluripotent Stem Cells (iPSCs) is a lengthy process with a typical timeline of approximately $112$ days (or ~$3.7$ months):

1. Fibroblast Isolation and Expansion: $14$ days

2. Fibroblast Quality Assessment: $7$ days

   • Karyotype analysis.

   • Sterility testing (e.g., bacteria, fungi, mycoplasma).

3. iPSC Reprogramming: $14$ days

4. iPSC Expansion & Banking: $21$ days (for $15$ million cells).

5. iPSC Gene Expression Analysis, Flow Cytometry: $7$ days

6. NSC Differentiation (from iPSCs): $14$ days

7. NSC Expansion & Banking: $21$ days (for $15$ million cells; $10$ million for treatment/$5$ million for testing).

8. NSC Gene Expression Analysis, Flow Cytometry: $7$ days

9. NSC Quality Assessment: $7$ days

   • Karyotype analysis.

   • Sterility testing.

The "Golden Window" for Stroke Treatment

Stroke is a critical condition:

• Second leading cause of death globally.

• Number one cause of long-term disability in the US.

• Someone has a stroke every $40$ seconds.

• Someone dies of stroke every $4$ minutes.

There are two main types: Ischemic stroke and Hemorrhagic stroke.
The golden window for stroke treatment and recovery is typically considered to be within $3$ months. The traditional iPSC-mediated NSC generation timeline of ~$3.7$ months is too long to meet this critical therapeutic window.

Proposed Solution: Direct Reprogramming for Time Savings

Direct reprogramming technologies could significantly reduce this timeline. One proposed method seeks to save $42$ days, reducing the total time to approximately $70$ days, thus falling within the $3$ month golden window for acute conditions.

Diseases with Less Critical Timing

While acute conditions like stroke demand rapid intervention, other neurodegenerative diseases, such as Huntington’s disease (HD) and Alzheimer’s disease (AD), have a less critical timing window for general intervention. These conditions are often selected for disease modeling and drug screening due to their genetic basis, allowing researchers to create patient-specific cell lines to study disease mechanisms and test potential treatments.

Key Elements from the Article's Introduction

iPSC Technology: Advantages and Concerns

• Advantages: iPSCs can be readily reprogrammed from patients' somatic cells (e.g., fibroblasts) to generate patient- and disease-specific cell types. These are invaluable for disease modeling and drug development (e.g., HD iPSC Consortium, 2012).

• Concerns: The use of iPSCs carries risks of tumorigenicity and uncontrolled spontaneous differentiation, which are significant hurdles for clinical application.

Induced Neuron (iN) Technology: A Safer Alternative

• Advantages: The iN technology (direct conversion) offers a fast and simple method to generate specific neuronal subtypes. Critically, it may avoid problems associated with hiPSCs, such as uncontrolled cell differentiation and tumor formation.

• Many research groups have successfully demonstrated the direct reprogramming of fibroblasts into neural stem cells (NSCs) in both mouse and human systems.

Focus on Embryonic Neural Progenitors (ENPs) and Broader Differentiation Potential

• Most studies using iN technology have concentrated on differentiating NSCs/NPs into Central Nervous System (CNS) cell types.

• This study aims to generate Embryonic Neural Progenitors (ENPs). ENPs possess a greater level of plasticity compared to more mature NSCs/NPs, allowing for a wider range of differentiation outcomes.

• The research also seeks to address peripheral nerve cell types, which are often overlooked in current studies.

Overview of CNS and PNS Cell Types

• Central Nervous System (CNS) Cell Types and Functions:

  • Neurons: Conduct electrical impulses; enable cognition, sensation, motor control.

  • Astrocytes: Regulate the blood–brain barrier, ion balance, and neurotransmitter recycling.

  • Oligodendrocytes: Form myelin sheaths around axons in the CNS.

  • Microglia: Act as resident immune cells of the CNS (analogous to macrophages).

• Peripheral Nervous System (PNS) Cell Types and Functions:

  • Neurons: Sensory and motor neurons connecting the CNS to limbs and organs.

  • Schwann cells: Myelinate axons in the PNS (analogous to oligodendrocytes).

  • Satellite glial cells: Support and insulate neuron cell bodies in ganglia.

  • Macrophage-like immune cells: Provide immune surveillance in the PNS (non-microglial).

Degrees of Plasticity

Cellular plasticity refers to the ability of a cell to differentiate into other cell types:

• Totipotent Stem Cells: Can differentiate into any cell type, including placental cells, forming a complete organism (e.g., two-cell stage, four-cell stage, eight-cell stage embryos).

• Pluripotent Stem Cells: Can differentiate into any cell type of the three germ layers (endoderm, mesoderm, ectoderm), but cannot form a complete organism (e.g., Inner Cell Mass cells of the blastocyst, human embryonic stem cells (hESCs), Induced Pluripotent Stem Cells (iPSCs)).

• Multipotent Stem Cells: Can differentiate into a limited number of specialized cell types within a specific lineage (e.g., hematopoietic stem cells, neural stem cells).

• Differentiated Cells: Terminally specialized cells (e.g., Lung, Pancreas from Endoderm; Heart Muscle, Red blood cells from Mesoderm; Skin, Neuron from Ectoderm).

Differentiation and Reprogramming: Epigenetic Changes

Differentiation as an Epigenetic Change

Epigenetic modifications are heritable changes in gene expression or cellular phenotype that do not involve alterations to the underlying DNA sequence. These changes primarily affect gene activity.

Most common types of epigenetic modifications:

1. DNA Methylation:

   • Involves the addition of a methyl group to DNA, typically at CpG sites.

   • Enzymes like DNA methyltransferases (DNMTs), specifically DNMT3, catalyze this transfer.

   • Methylation often leads to compacted chromatin and gene silencing, suppressing gene expression.

2. Histone Modifications (e.g., Blocking/Acetylation):

   • Histone acetylation involves the transfer of an acetyl functional group from acetyl coenzyme A to histone proteins, catalyzed by "histone acetyltransferases" (HATs).

   • Histone deacetylation is the reverse reaction, removing acetyl groups, catalyzed by "histone deacetylases" (HDACs).

   • Acetylation removes the positive charge on histones, weakening their interaction with the negatively charged DNA. This results in a more relaxed chromatin structure, which is associated with increased gene transcription.

   • Deacetylation leads to condensed chromatin, which is generally associated with reduced gene transcription.

Reprogramming as the Reverse of Differentiation

Reprogramming, as highlighted by the 2012 Nobel Prize in Physiology or Medicine, is essentially the reversal of epigenetic modifications that occur during differentiation. It involves inducing cells to shed their differentiated identity and revert to a more pluripotent or multipotent state by manipulating these epigenetic marks.

Major Objectives and Findings Claimed by the Authors

1. Efficient Conversion of Fibroblasts to iENPs: The researchers defined two specific transcription factor (TF) combinations that can efficiently convert human fibroblasts (FBs) directly into multipotent induced Embryonic Neural Progenitors (iENPs).

2. Varying Proliferative Features and Regional Differentiation Preferences: They found that these different TF combinations induce iENP populations that exhibit distinct proliferative characteristics and preferential differentiation into specific brain regions or cell types.

3. Disease Modeling: Neurons derived from iENPs generated from patients with Alzheimer’s disease (AD) and Huntington’s disease (HD) successfully recapitulated the major pathological features of these diseases in vitro.

Experimental Figures and Findings Explained

Figure 1: Identification of Key Transcription Factors and Direct Conversion Strategy

• Figure 1a & b: TF Selection

  • Heatmap analysis of global gene expression profiles was performed on human Embryonic Stem Cell-derived ENPs (hESC-ENPs, NP1 and NP2) and human fibroblasts (FBs, FB1, FB2, FB3).

  • This analysis identified $25$ transcription factors (TFs) that were highly expressed in hESC-ENPs but not in FBs, indicating their potential role in neural progenitor identity.

• Figure 1B: Experimental Strategy

  • A schematic depicts the strategy for directly converting FBs into iENPs (termed "iENP-25F") by introducing the $25$ selected TFs.

  • PAX6:EGFP and SOX1:EGFP reporters were used. These reporters link the expression of Enhanced Green Fluorescent Protein (EGFP) to the promoters of neural progenitor genes, PAX6 and SOX1.

    • A gene promoter is a DNA sequence that acts as a binding site for proteins, initiating and controlling gene transcription (making RNA from DNA). These reporters are crucial for tracking and enriching for cells that successfully acquire neural progenitor identity.

• Figure 1C: Infection and Sorting

  • Fibroblasts were infected with lentiviruses encoding the $25$ hESC-ENP TFs and the neural reporter (e.g., PAX6:EGFP).

  • Cells were then grown and sorted using FACS (Fluorescence-Activated Cell Sorting) to isolate EGFP-positive neural progenitors.

  • UbC:EGFP infected cells served as controls.

• Figure 1D: Immunocytochemistry (ICC)

  • ICC analysis of iENP-25F clusters (which resembled neural progenitor colonies/spheres) using antibodies against key neural progenitor markers such as NESTIN, SOX1, and PAX6 confirmed their identity.

• Figure 1E: RT-PCR Analysis

  • RT-PCR (Reverse Transcription Polymerase Chain Reaction) was used to analyze the expression of indicated genes in iENP-25F.

    • Standard PCR amplifies specific DNA sequences.

    • RT-PCR first converts RNA into complementary DNA (cDNA) using reverse transcriptase, then amplifies specific sequences from the cDNA, allowing for the detection and quantification of RNA.

    • Primers are specific DNA probes for gene targets.

    • Gel electrophoresis separates DNA/RNA/protein fragments by size and charge. Ethidium bromide (EtBr) is used to visualize DNA under UV light. Primer design predicts fragment size.

  • Controls in RT-PCR (Figure 1E):

    • Positive Controls (P): hESC-ENP samples, expected to show expression of neural progenitor genes.

    • Negative Controls (N): Original fibroblast (FB) samples, expected to show no expression of neural progenitor genes, and "No Template Control" (NC) to check for contamination.

Figure 2 and 3: Refining TF Combinations and Characterizing iENPs

These figures focus on reducing the number of TFs required for efficient conversion and characterizing the resulting iENPs.

• Figure 2 (PAX6:EGFP-driven iENPs):

  • Figure 2A: Strategy for TF Reduction: Schematic illustrates the stepwise reduction of TFs from $25$ to a minimal set, using PAX6:EGFP as a reporter to monitor successful neural progenitor induction.

  • Figure 2B & C: Stepwise TF Dropout: Individual TFs were systematically removed from the $25$-TF and then $15$-TF combinations. The relative percentage of PAX6:EGFP+ cells was measured after each dropout to identify the most potent and essential factors.

  • Figure 2D: Efficiency Comparison: Comparison of iENP induction efficiency (PAX6:EGFP+ cells from FBs) using $25$-, $15$-, and $6$-TF combinations (measured by FACS). The $6$-TF combination (iENP-6F) achieved a significant percentage of PAX6:EGFP+ cells (10.54 ext{%} ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ }) with reduced factors.

  • Figure 2E: Global Gene Expression: Heatmap (microarray analysis) showed global gene expression profiles of FB, hESC-ENP, iENP-6F, and iENP-15F, confirming transcriptional similarity of iENPs to hESC-ENPs.

  • Figure 2F: ICC Staining: Immunocytochemistry of iENP-6F using antibodies against NP markers (e.g., PAX6, SOX1, NESTIN, ZO1) further validated their neural progenitor identity.

  • Figure 2G & H: Endogenous and Exogenous Gene Expression: RT-PCR assessed both endogenous and exogenous expression of the $6$ TFs, along with other neural genes, confirming successful reprogramming and activation of neural programs.

• Figure 3 (SOX1:EGFP-driven iENPs):

  • A parallel experiment to Figure 2, but focused on the SOX1:EGFP reporter. This led to a different, but similarly efficient, minimal TF combination: $7$ TFs (iENP-7F). It achieved 11.22 ext{%} ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } ext{ } SOX1:EGFP+ cells.

  • Similar characterizations (TF reduction, efficiency, global gene expression, ICC, RT-PCR) were performed for iENP-7F, confirming its neural progenitor identity.

Figure 4: Differentiation Potential and Functional Characterization

This figure demonstrates the multipotency of the generated iENPs and the functionality of their differentiated neuronal derivatives.

• Figure 4A-E: Multipotency - Glial and Neuronal Markers:

  • ICC staining of differentiated iENP-6F showed expression of glial markers like GFAP (astrocyte marker) and GALC (oligodendrocyte marker), as well as various neuronal markers (e.g., MAP2, NEUND, NFH, TUJ1), and the synapse marker SYN.

  • This indicates that iENPs can differentiate into multiple neural cell types (neurons, astrocytes, oligodendrocytes).

• Figure 4F: Quantification of Cell Types: Quantification confirmed the presence of TUJ1+ (neurons), GFAP+ (astrocytes), and GALC+ (oligodendrocytes) cells in differentiated hESC-NPs, iENP-6F, and iENP-15F, showing comparable differentiation capabilities.

• Figure 4G-M: CNS and PNS Neuronal Subtypes:

  • ICC staining revealed that iENP-6F could differentiate into both CNS (e.g., GABAergic, TBR1+ cortical, TH+ dopaminergic) and PNS neuronal antigens (e.g., PRPH+ neurons).

  • This demonstrates the broad regional differentiation capacity of these iENPs.

• Figure 4N: Lineage-Specific Differentiation Cues:

  • 4Na: Schematic depicting experimental procedures using lineage-specific cues (e.g., N2B27+SHH+Noggin+DKK1+XAV979+LDN93189+SB431542 for cortical neurons) to induce specific neuronal subtypes from iENP-6F.

  • 4Nb: ICC characterization confirmed successful generation of specific subtypes (e.g., TBR1+ cortical, TH+ dopaminergic, PRPH+ PNS neurons).

  • 4Nc: Quantification showed improved generation of specific neuronal subtypes when specific inducers (GF+) were applied compared to conditions without them (GF-).

• Figure 4O: Functional Neuronal Activity (Patch Clamp):

  • Whole-cell patch-clamp recording of iENP-6F-derived neurons demonstrated functional electrophysiological properties.

  • 4Oa: Current recordings showed spontaneous electrical activity.

  • 4Ob: Action potentials (nerve impulses) were successfully induced by current steps ($50$ to $+120$ pA), indicating that the derived neurons are electrically excitable.

    • Neuron Action Potential: A brief electrical impulse that travels along a neuron's axon. It is an "all-or-none" event triggered when a stimulus causes the membrane potential to reach a threshold, leading to rapid changes in voltage due to ion (Na+, K+) flow across the membrane.

  • 4Oc: Inward Na+ currents and outward Ca2+ currents were observed. The inward Na+ currents were specifically blocked by tetrodotoxin (TTX), confirming the presence of voltage-gated sodium channels critical for action potential generation and thus, functional neuronal activity.

• Figure 4P: In Vivo Transplantation:

  • iENP-6F cells were transplanted into a host brain to assess their survival, integration, and differentiation in vivo.

  • 4Pa: IHC staining with human nuclear antigen (HuNu) confirmed successful engraftment of human cells in the corpus callosum.

  • 4Pb-i: IHC analysis at $12$ weeks post-transplantation showed that the transplanted iENP-6F cells differentiated into various neural cell types in vivo, including astrocytes (GFAP+), oligodendrocytes (MBP+, NG2+), and neurons (TUJ1+, MAP2+), further demonstrating their therapeutic potential and multipotency in a complex environment.

  • 4Pj: Scheme illustrates the relative positions of differentiated cells after transplantation.

Figure 6: Differential Properties of iENP-6F and iENP-7F

This figure compares the proliferative features and regional differentiation preferences between iENP-6F (PAX6-driven) and iENP-7F (SOX1-driven).

• Figure 6A & B: Gene Expression and Proliferation/Cell Death:

  • 6A: Heatmap analysis of global gene expression profiles of undifferentiated iENP-6F, iENP-7F, and FBs showed overall similarities but also significant differences.

  • 6B(a): Analysis of Gene Ontology (GO) terms revealed that iENP-7F showed lower expression of genes related to cell cycle and division compared to iENP-6F.

  • 6B(b): Ingenuity Pathway Analysis (IPA) indicated that cell-death-associated pathways were activated in iENP-7F compared to iENP-6F. Overall, $170$ genes were found to be significantly different (>2-fold change) between the two iENP types.

  • 6C(c): Growth Curve Analysis: Confirmed that iENP-7F had a slower proliferation rate than iENP-6F.

  • 6D(d): BrdU and TUNEL Assays:

    • The BrdU (bromodeoxyuridine) assay measures cell proliferation by detecting BrdU incorporation into newly synthesized DNA during the S phase of the cell cycle, using a BrdU-specific antibody. iENP-7F showed fewer BrdU+ cells.

    • The TUNEL (Terminal deoxynucleotidyl transferase dUTP Nick-End Labeling) assay detects DNA fragmentation, a hallmark of apoptosis. The enzyme TdT adds fluorescein-labeled dUTP to $3' ext{-OH}$ ends of DNA fragments. iENP-7F showed more TUNEL+ cells, indicating higher apoptosis.

• Figure 6C & D: Regional Markers in iENPs and iENP-Derived Neurons:

  • 6C(a-b): ICC staining and quantification showed preferential expression of specific brain regional antigens (e.g., BF1 for forebrain, PITX3 for midbrain, HOXB4 for hindbrain, ISL1 for motor neurons, P75 for PNS neurons) in undifferentiated iENP-6F and iENP-7F. This suggests an intrinsic bias in their developmental potential.

  • 6D: Quantification showed these brain regional markers were also expressed in iENP-derived neurons, confirming that the regional preferences persisted after differentiation.

• Figure 6E & F: Regional Gene Expression Differences:

  • 6E: A pie chart depicted the proportion of brain regional subtype-associated genes that were up- or down-regulated between iENP-7F and iENP-6F.

  • 6F: RT-qPCR analysis further detailed the relative expression of these regional-associated genes. iENP-6F displayed more forebrain (FB) and midbrain (MB) characteristics (CNS), while iENP-7F leaned more towards spinal cord (SC) and peripheral nervous system (PNS) characteristics.

• Figure 6G: Model of Differential Differentiation Propensities:

  • This model summarizes that using different combinations of hESC-ENP-TFs and neural reporters (e.g., PAX6:EGFP for iENP-6F vs. SOX1:EGFP for iENP-7F) can directly convert fibroblasts into iENPs with distinct differentiation propensities.

Figure 7: Disease Modeling with Patient-Derived iENPs

This figure demonstrates the capacity of patient-derived iENPs to recapitulate disease pathologies in vitro.

• Generation of Disease-Specific iENPs:

  • iENPs were generated from fibroblasts of AD patients (one with APOE4/E4 mutation (AD1), two with PSEN1 mutation (AD2, AD3 familial AD - fAD)) and HD patients (male and female, 41 CAG repeats in the HTT gene (Huntingtin gene)).

• Figure 7A: Morphology and Nestin Expression in Disease iENPs:

  • Representative images showed typical iENP morphology and positive ICC staining for Nestin in both AD-iENPs and HD-iENPs, confirming their neural progenitor identity.

• Figure 7B: Differentiation into Neural Cell Types:

  • Phase-contrast images and ICC staining for GFAP, GALC, and TUJ1 confirmed that both AD-iENP and HD-iENP derivatives could differentiate into glia and neurons, similar to control iENPs.

• Figure 7C: Alzheimer's Disease Pathology - Amyloid Beta (Aeta):

  • Elevated amyloid $b$ (Aeta) and phosphorylated TAU (pTAU) are major pathological features of AD. Researchers measured extracellular Aeta40 and Aeta42 levels in conditioned media from neurons differentiated from AD- or control-iENPs using ELISA (Enzyme-Linked Immunosorbent Assay).

    • ELISA is a biochemical assay to detect and quantify proteins. It utilizes specific antibodies, a enzyme-linked secondary antibody, and a substrate that generates a detectable signal.

  • Results showed that neurons derived from AD-iENPs (especially those carrying PSEN1 mutations like AD2 and AD3) exhibited increased levels of Aeta isoforms, particularly an elevated Aeta42/40 ratio, mimicking AD pathology.

• Figure 7D: Alzheimer's Disease Pathology - Phosphorylated TAU (pTAU):

  • 7D(a): ICC staining for pTAU expression (using AT8 antibody) in AD-iENP-derived neurons confirmed significant aggregation of pTAU.

  • 7D(b): Quantification showed that treatment of AD-iENP-derived neurons with GSK3eta inhibitors (SB415286 and 1-Aza) significantly reduced pTAU aggregation, compared to DMSO-treated controls. This demonstrates the utility of the iENP model for drug screening and validation of therapeutic targets for AD.

• Figure 7E: Huntington's Disease Pathology - DNA Damage:

  • 7E(a): ICC staining and 7E(b): quantification of oldsymbol{ ext{gH2AX+}} cells in vehicle (DMSO) and CGS 21680-treated control and HD-iENPs.

    • oldsymbol{ ext{gH2AX}} (gamma H2AX) is a key molecular marker for DNA double-strand breaks (DSBs). Increased oldsymbol{ ext{gH2AX}} indicates DNA damage.

  • Previous research established that HD-iPSC-derived neurons are vulnerable to DNA damage, and A2AR (adenosine A2A receptor) stimulation reduces it. This study confirmed that HD-iENPs and their neuronal derivatives recapitulate this vulnerability.

  • Treatment with the selective A2AR agonist, CGS21680, significantly reduced DNA damage (fewer oldsymbol{ ext{gH2AX+}} cells) in HD-iENPs, demonstrating their suitability for modeling HD pathology and testing potential therapeutic compounds.

Conclusion: Summary of Author's Claims

The authors successfully provided data to support their major claims:

1. "We defined two TF combinations, the overexpression of which can efficiently convert human FBs into multipotent iENPs."

   • Supported by Figures 1, 2, and 3, which demonstrate the efficient generation and characterization of iENP-6F (PAX6-driven) and iENP-7F (SOX1-driven) populations from human fibroblasts using specific TF combinations.

2. "Importantly, we found that different combinations of TFs can induce iENP populations with varying proliferative features and regional differentiation preferences."

   • Supported by Figure 6, where iENP-6F and iENP-7F displayed different growth rates, cell death profiles, and distinct biases towards CNS (forebrain/midbrain) or PNS regional identities, both in their progenitor state and after differentiation into neurons.

3. "We also demonstrated that neurons derived from AD- and HD- iENPs, recapitulated the major disease pathological features in vitro."

   • Supported by Figure 7, which shows that AD-iENP-derived neurons exhibited increased Aeta levels and pTAU aggregation, while HD-iENP-derived neurons displayed increased DNA damage, all consistent with the respective disease pathologies and responsive to pathogenic modulators.

Broader Significance

This research aligns with the global critical need for transplantable cells and tissues to address organ failure and the increasing impact of degenerative age-related human diseases with limited or no treatments. The generation of human-induced pluripotent stem cells (hiPSCs) and, by extension, directly converted iENPs, offers a unique opportunity to obtain a virtually unlimited supply of specialized cells for therapeutic applications, disease modeling, and drug development, potentially circumventing some of the challenges associated with traditional iPSC-based approaches.