Id1, Id2, Id3, and Id4 in adult mouse brain lateral wall V-SVZ neural stem cells

Version 465 2024-01-25, 03:30

Version 464 2024-01-03, 03:27

Version 463 2024-01-03, 03:26

Version 462 2023-12-22, 03:27

Version 461 2023-12-14, 04:04

Version 460 2023-11-12, 17:06

Version 459 2023-11-12, 03:38

Version 458 2023-11-12, 03:34

Version 457 2023-09-22, 01:01

Version 456 2023-09-21, 23:34

Version 455 2023-08-28, 02:38

Version 454 2023-07-21, 00:19

Version 453 2023-07-19, 03:23

Version 452 2023-03-19, 06:10

Version 451 2023-03-18, 17:32

Version 450 2023-01-19, 05:36

Version 449 2023-01-19, 04:26

Version 448 2023-01-19, 04:21

Version 447 2023-01-09, 23:12

Version 446 2023-01-09, 06:19

figure

posted on 2021-07-27, 03:37 authored by Hyung-song NamHyung-song Nam, Mario Capecchi, Robert Benezra

An R script to analyze the Mizrak et al., 2019 dataset.

Clusters. Selected marker genes RNA in clusters.

Markers for selected subclusters identified by Seurat.

To validate the AverageExpression command results, RT-qPCR measurements from Morizur et al., 2018 (https://doi.org/10.1016/j.stemcr.2018.06.005) were utilized. The trends were very similar, suggesting that normalized scRNA-seq datasets can also reveal the average RNA expression level of cell populations otherwise obtained by FACS sorting and RT-qPCR.

The relative RNA abundance in the lateral wall V-SVZ qNSC's of Id1, Id2, Id3, and Id4 seems to be Id2 > Id3 > Id4 > Id1. How the RNA levels correlate to protein levels is unclear, although the Id1 protein in particular was measured directly by knock-in C-terminal fusion of the Venus yellow fluorescent protein (https://doi.org/10.1016/j.stem.2009.08.017, https://doi.org/10.1016/j.ccr.2011.11.025, https://doi.org/10.1242/jcs.096198, https://doi.org/10.1016/j.stemcr.2014.09.012, https://doi.org/10.7554/elife.01197, https://doi.org/10.1016/j.devcel.2019.05.032).

==============

The material from my PhD thesis work I put up here after going through them again after more than ten years are (1) copied and pasted from my PhD thesis, (2) data from my PhD thesis work (Immunofluorescence images, Southern blots, PCR gels, restriction digests, and flow data), and (3) re-analysis of the data using modern software (flow data and sequence analyses for the knock-in allele targeting vectors and Southern blots).

Mirax epifluorescence scans of coronal sections of >6 week old wildtype mouse brain. Id1 (green). Ki-67 or Mcm2 (red). DAPI (blue). Id1 and Ki-67 or Mcm2 merged.

A confocal z-stack from coronal section of a wildtype mouse brain V-SVZ, >6 week old. A maximum projection image montage and a movie of the z-stack. Id1 (green). Mcm2 (red). DAPI (blue).

A confocal z-stack from coronal section of a wildtype mouse brain V-SVZ, >6 week old. Perfused 48 h after the cessation of Ara-C infusion for 6 days, as described in Nam and Benezra, 2009. EdU was injected 30 min before the perfusion. A maximum projection image montage and a movie of the z-stack. Id1 (green). EdU (red). DAPI (blue). The small dots are probably non-specific protein precipitates from the block buffer.

A confocal z-stack from coronal section of a wildtype mouse brain V-SVZ, >6 week old. A maximum projection image montage and movies of the z-stack. Id1 (green). S100beta (red). DAPI (blue).

Some self-explanatory images.

Production of the Id1-floxed allele mouse (http://www.informatics.jax.org/allele/MGI:4366910), the Id1-Venus allele mouse (http://www.informatics.jax.org/allele/MGI:4366905), the Id1-IRES-creERT2 allele mouse (http://www.informatics.jax.org/allele/MGI:4366863), and the ROSA26-StLa allele mouse (http://www.informatics.jax.org/allele/MGI:4366911).

There were two variants of the Id1 targeting vector. The first variant utilized a long 5' arm and a DTA cassette. The second variant utilized a shorter 5' arm and DTA and TK cassettes. The lengths of the homology arms in the first variant were from Yan et al., 1997 (https://doi.org/10.1128/mcb.17.12.7317). The actual genomic DNA was subcloned from RPCI-23 C57BL/6J mouse genome BAC library using recombineering.

I would put up the Sanger sequencing files of the targeting vectors if I could, but I'm not sure if I have the intellectual property rights to do that. Regardless, I have re-analyzed the sequencing results of the Id1 targeting vectors, and did not find any errors. I did not go through the ROSA26-StLa targeting vector sequencing again.

By the way, these targeting vectors were constructed on a high copy backbone. With the large genomic insert, the plasmid DNA of the final construct was very difficult to prepare because the E. coli harboring it grew very slowly, and the plasmid was prone to rearrangements by recombination. This sometimes happens - see https://www.addgene.org/61580/ for comments from the Hongkui Zeng laboratory. To make it work, plasmid prep from every culture of the clone had to be checked for recombination by restriction digests before using. Preps of recombined plasmids had to be discarded. The low copy backbone plasmids are much easier to work with because they don't behave this way. For anybody still doing gene targeting, I would recommend them (see https://doi.org/10.6084/m9.figshare.c.5277341.v12 for links to Addgene).

Flow cytometry data. The figure pdf shows the gates that were utilized to detect Id1-Venus+ cells from dissociated V-SVZ cells as well as Id1-Venus+ cells in cultured V-SVZ cells. The zip files are the actual FCS files. The file names were modified to be descriptive.

Tried a wider gate for Id1-Venus and Gfap detection. Could see more in the double-positive gate, but the larger cells might be more autofluorescent.

As in the paper, there were neural Id1-Venus+ cells as well as endothelial Id1-Venus+ cells in the V-SVZ (mice >6 weeks of age). Culturing the V-SVZ cells in neurosphere-forming media with EGF and FGF-2 was a quick way to enrich for the neural lineage cells without FACS, etc. In the cultures of neural lineage cells of the V-SVZ thus obtained, there were Id1-Venus+ and Id1-Venus- cells that could be discerned. The Id1-Venus-high cells formed self-renewing neurospheres, etc as described in the paper.

As an aside, the Id1-Venus fusion protein reporter actually generated a very sensitive read-out of the Id1 protein. These are why I think so. The Id1 protein was very difficult to detect in the brain tissue (i.e., no signal with conventional indirect immunofluorescence in both cell types). I could only visualize it after Tyramide Signal Amplification. The labeling efficiency with the Id1^IRES-creERT2 allele was also somewhat low. Although other reasons are also possible, these suggested to me low protein expression level. Then, there may not be so many copies of the Id1-Venus protein in the cell. Yet, it was detectable with flow cytometers.

If the Discussion of the Nam and Benezra, 2009 paper wasn't clear, I note here that even though the absolute level of the Id1 protein may be low, its level is highest in the stem/progenitor cells if one considers the relative levels along the neurogenic lineage.

Id1-Venus fluorescence from adherent NSPC cultures.

Two confocal images of YFP+ cells from ventricular wall whole mount of Id1^{IRES-creERT2/+}; ROSA26^LSL-YFP/+ mice that were scored to be "B1" cells from Mirzadeh et al., 2008 (https://doi.org/10.1016/j.stem.2008.07.004).

Previously, the whole mount immunfluorescence technique wasn't as optimized as it is now, so the images weren't so clear. However, putting the previous work in context with my more recent work, what is clear is that the cells labeled by the Id1^IRES-creERT2 allele were different from the cells labeled by the Lrig1^T2A-iCreERT2 allele (see https://doi.org/10.6084/m9.figshare.12731900.v8). So how is it possible to get Lrig1 out of Id1^high cells when they label cells with different morphologies? Although it wasn't apparent with the previous technique, more recent results suggest that Id1 also labels the cells labeled by Lrig1 (unpublished preliminary observation). So, a working model is that there are at least two different types of stem cells: Id1 reveals both, whereas Lrig1 reveals mostly one subset.

==============

Id1+ cells outside the V-SVZ.

Two coronal section images from hippocampus of Id1^{IRES-creERT2/+}; ROSA26^StLa/+ mouse. 1 month after tamoxifen inductions. Non-endothelial X-gal+ cells (from tauLacZ) in the dentate gyrus.

A confocal image from coronal section of hippocampus of Id1^{IRES-creERT2/+}; ROSA26^StLa/+ mouse. 1 month after tamoxifen inductions. Tau-b-gal (green), NeuN (red), and DAPI (blue).

A confocal image from a coronal section of wildtype mouse embryo (~E12.5) spinal cord. Id1 (green), Nestin (red), and DAPI (blue).