(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Lmo4 synergizes with Fezf2 to promote direct in vivo reprogramming of upper layer cortical neurons and cortical glia towards deep-layer neuron identities [1]

['Torsten Felske', 'Université Côte D Azur', 'Cnrs', 'Inserm', 'Ibv', 'Nice', 'Chiara Tocco', 'Sophie Péron', 'Research Group', 'Adult Neurogenesis']

Date: 2023-08

In vivo direct neuronal reprogramming relies on the implementation of an exogenous transcriptional program allowing to achieve conversion of a particular neuronal or glial cell type towards a new identity. The transcription factor (TF) Fezf2 is known for its role in neuronal subtype specification of deep-layer (DL) subcortical projection neurons. High ectopic Fezf2 expression in mice can convert both upper-layer (UL) and striatal projection neurons into a corticofugal fate, even if at low efficiency. In this study, we show that Fezf2 synergizes with the nuclear co-adaptor Lmo4 to further enhance reprogramming of UL cortical pyramidal neurons into DL corticofugal neurons, at both embryonic and early postnatal stages. Reprogrammed neurons express DL molecular markers and project toward subcerebral targets, including thalamus, cerebral peduncle (CP), and spinal cord (SC). We also show that co-expression of Fezf2 with the reprogramming factors Neurog2 and Bcl2 in early postnatal mouse glia promotes glia-to-neuron conversion with partial hallmarks of DL neurons and with Lmo4 promoting further morphological complexity. These data support a novel role for Lmo4 in synergizing with Fezf2 during direct lineage conversion in vivo.

Funding: This work was funded by an AFM-Telethon grant (20899 to MS), by the French Government (National Research Agency, ANR) through the ‘Investments for the Future’ programs LABEX SIGNALIFE (ANR-11-LABX-0028-01 and IDEX UCAJedi ANR-15-IDEX-01 to MS), by an ERA-NET Neuron grant (Brain4Sight) (ANR-21-NEU2-0003-03 to MS; 01EW2202 to BB); by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (101021560, IMAGINE to BB), by the Wellcome Trust (206410/Z/17/Z to BB), by the German Research Foundation (BE 4182/11-1, 357058359 to BB), by the research initiative of Rheinland-Pfalz at the Johannes Gutenberg University Mainz (ReALity to BB); by a SIGNALIFE Ph.D. contract to TF and by the Inneruniversitäre Forschungsförderung Stufe I of the Johannes Gutenberg University Mainz to SP. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Aiming at improving direct reprogramming efficiency and subtype specificity, first in immature neuronal cells and then in developing glial cells’ we used Fezf2 as a well-established subcerebral determinant gene together with the co-adaptor Lmo4. Lmo4 is known to work as an epigenetic and subtype-specific factor by acting through an HDAC-dependent mechanism in de-repressing the Ctip2 locus [ 27 – 31 ]. We show that Lmo4 synergizes with Fezf2 in converting UL neurons into DL subcortical projection neurons at a higher efficiency than previously reported [ 2 , 4 ]. We also find that Fezf2 directs Neurog2/Bcl2-mediated reprogramming of early cortical glia into Ctip2-expressing iNs, while Lmo4 further promotes their morphological complexity. Together, our data show that Fezf2 synergizes with the co-adaptor Lmo4 in lineage reprogramming in both neurons and glia towards a DL cortical neuron fate.

To date, generating different neuronal subtypes of fully functional mature cells constitutes the major challenge in direct reprogramming [ 6 , 23 ]. TFs commonly used in direct neuronal reprogramming typically possess pioneering activity, such as transiently engaging closed chromatin to initiate transcriptional programs leading to cell fate changes [ 24 , 25 ]. However, they often fail to activate genes that are silenced by specific DNA and chromatin modifications [ 26 ]. Identifying co-factors that facilitate the binding of lineage-specific TFs on less accessible chromatin sites might improve reprogramming efficiency and specificity and stabilize neuronal identity. Therefore, one of the major aims of direct reprogramming becomes to identify novel factors conducive for generating specific neuronal subtypes.

Although direct lineage reprogramming still inspires cell replacement therapy, it also includes major limitations that need to be overcome. First, molecular and cellular features of the starter cell type may ease or impede the conversion process, and lineage-related cells might be easier to convert into each other as they share a common origin. For example, glial cells can be converted in vitro into functional neurons by overexpression of a single TF [ 8 , 9 ], whereas in vivo, single-factor reprogramming is much more limited, often requiring additional stimuli such as tissue injury and glial cell reactivation [ 10 – 13 ]. Moreover, the same reprogramming factor can trigger different outcomes when induced in different cell types. For example, Neurog2 has been shown to convert cortical astrocytes into glutamatergic-like cortical neuros [ 9 , 14 ], while thalamic and spinal astrocytes acquire thalamic relay neurons and spinal interneuron-like signatures, respectively [ 15 , 16 ], and fibroblasts even adopt a cholinergic motor neuron-like cell fate [ 17 ]. Moreover, successful reprogramming may require co-factors that negotiate critical transitions in the cellular metabolism [ 18 , 19 ]. Conversion of one postmitotic neuron subtype into another appears to be more complex and limited to the very early stages of postnatal life. For example, embryonic (E) 14.5 callosal projection neurons of cortical upper layers (ULs) II to IV could be converted into deep layers (DLs) V/VI subcortical projection neurons via the forced expression of the TF Fezf2, but failed at later stages, i.e., after postnatal (P) day 3 [ 2 , 4 , 20 , 21 ]. These observations point to the existence of a tight crosstalk between reprogramming factors and the cellular context in which they operate and suggest that the epigenetic signature of the starting cell population may limit cellular plasticity and their response to reprogramming factors [ 22 ]. Thus, current research aims at identifying reprogramming roadblocks whose removal could improve reprogramming efficiency and accuracy [ 7 ]. For example, during glia-to-neuron reprogramming, increased production of reactive oxygen species (ROS) can lead to cell death of induced neurons (iNs) per ferroptosis [ 18 ]. This has been partially overcome by allowing the expression of anti-cell death regulators, such as Bcl2, or by pharmacological treatments aimed at reducing ROS. Simultaneous expression of Neurog2 and Bcl2 induced the reprogramming of non-neuronal cells into immature DL pyramidal neurons [ 18 ].

While mechanisms driving the acquisition of specific neuronal class and subtype identities are increasingly investigated, the maintenance of such identity and conversely their degree of plasticity remains rather enigmatic [ 1 ]. Work over the last decade has challenged the view that neural cell identity is irrevocably fixed by demonstrating that fate-restricted neuronal progenitors and even early postmitotic neurons can be coaxed into neurons of distinct identities when appropriate transcriptional cues are provided [ 2 – 4 ], an experimental approach referred to as direct lineage reprogramming [ 5 ]. Likewise, different classes of glial cells, i.e., astrocytes, oligodendrocyte progenitor cells, and microglia can be converted into induced neurons by forced expression of neurogenic transcription factors (TFs) or regulatory RNAs, known to act as key regulators of cell fate during development [ 6 , 7 ]. However, it is still unclear to which extent induced neurons generated by direct lineage reprogramming acquire authentic molecular signatures of the desired neuronal subtype sharing similar developmental trajectories, function, and connectivity with subtype-specific projections.

Results

Reprogramming of upper into layer V neurons by Fezf2 and Lmo4 is more efficient in primary motor than somatosensory cortex We next investigated whether the cortical environment could impact UL to DL conversion. Since the primary motor area (M1) differs in its cytoarchitecture from S1 cortex by containing a higher number of layer V Ctip2+ neurons [41,42], we hypothesized that M1 might provide an even more conducive environment for layer V reprogramming of UL neurons. To examine this possibility, E14.5 embryos were electroporated with cGFP, cLmo4, cFezf2, or cFezf2/cLmo4 plasmids into the presumptive motor area and brains analyzed at P7 (Fig 2A). Targeted GFP+ M1 postmitotic UL neurons (Fig 2B) were assessed for expression of UL and DL markers and then compared to S1-electroporated neurons (Fig 2C–2E). As for S1, ectopic expression of cLmo4 alone did not result in any lineage conversion, but the combined expression of cFezf2 with cLmo4 in M1 resulted in a 92% reduction of the GFP/Cux1+ cells and an almost 80% increase of the GFP/Ctip2+ and 60% of the GFP/Pcp4+ cell populations, compared to the presence of 78% Cux1+ UL neurons versus 34% Ctip2+ and 23% Pcp4+ DL-like neurons in cFezf2-electroporated brains (Fig 2C and S2 Data). No statistical differences were observed in the number of layer VI Fog2+ and Darpp32+ cells between cFezf2 and cFezf2/cLmo4 GFP+ electroporated cells (Fig 2C and S2 Data), strongly supporting a synergistic effect of Lmo4 with Fezf2 in reconverting UL into layer V-like projection neurons. In addition, our data show that cell lineage conversion was more efficient in M1 than S1, particularly regarding the GFP/Ctip2+ cells that increased from 64% in S1 to almost 80% in M1 in double cFezf2/cLmo4 UL GFP+ cells. No significant changes between S1 and M1 were observed for Fog2+ cells (Fig 2D and 2E and S2 Data). These data strongly indicate that Lmo4/Fezf2 co-expression leads to higher reprogramming efficiency in M1 than in S1. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Reprogramming of ULs into layer V neurons by Fezf2 and Lmo4 is more efficient in motor than somatosensory cortex. (A) Schematic representation of the experimental procedure. cGFP, cLmo4 (cL), cFezf2 (cF) or cFezf2 and cLmo4 (cF+cL) plasmids were electroporated into E14.5 motor (M1) cortices. Brains were collected at P7. (B) IF of GFP, UL marker Cux1, and DL V marker Ctip2 on a coronal slice of a cF+cL-electroporated brain. The white box indicates the magnification image on the right side. (C) Percentage of M1 electroporated-UL neurons expressing UL vs. DL markers. (D, E) Percentage of S1 vs. M1-electroporated UL neurons expressing layer V marker Ctip2 (D) and layer VI marker Fog2 (E). Scale bars: B = 1,000 μm (left, macro image), 200 μm (right, magnified image). Results are represented as mean ± SEM. Two-way ANOVA with Tukey’s post hoc correction (2C) or two-way ANOVA with Sidak’s post hoc correction (2D-E) was used for statistical analysis. *p < 0.5, **p < 0.01, ***p < 0.001, ns = not significant. n = 3 brains for each plasmid. Extended data and statistics are listed in S2 Data. DL, deep layers; GFP, green fluorescent protein; IF, immunofluorescence; UL, upper layers. https://doi.org/10.1371/journal.pbio.3002237.g002 This suggests that either the environment and/or the intrinsic cell competence of M1 are more conducive towards Fezf2-dependent layer V lineage conversion, in line with its expanded expression and a larger representation of subcerebral layer V projection neurons in M1 than in S1 during physiological development [43]. Even though M1 produced a slightly better conversion, S1 remained our choice of preference for the following experiments due to its high accessibility that grants more reliable and reproducible electroporation sites, and linked analyses.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002237

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.