The GLIA-Baek Hyun-Sook/Frank Kirchhoff team found that a group of OPCs did not express Olig2, and that acute brain injury and learning activities promoted the formation of this group of cells

In the central nervous system, oligodendrocyte (OL) is the only glial cell
that can form a myelin sheath.
Oligodendrocyte precursor cells (OPCs), as precursor cells of OL, continue to proliferate and differentiate into OL[1] throughout life ]
。 Olig2 is a basic helix-loop-helix (bHLH) transcription factor that is widely expressed in OL lineage cells, including OPC and mature OL), which plays an important role in the development and differentiation of OL lineage cells, and its role is different at different stages of the lineage [2].

Therefore, Olig2 is often used as a marker for OL lineage cells
.
NG2 (nerve/glia antigen 2) is a chondroitin sulphate proteoglycan 4 (CSPG4), under physiological conditions OPC and pericyte specific expression
.
Recent studies have shown that there is a group of cells in the brain that are both positive for NG2 and negative for Olig2 [3, 4].

Our previous work has shown that a subset of microglia also express NG2 protein
under acute brain injury conditions.
In addition, the outer end of NG2 protein is easily hydrolyzed and adhered to other cell surfaces under pathological conditions [5].

Therefore, the immunoreactivity of the NG2 protein itself does not necessarily mean whether the cells express the NG2 protein, let alone that these cells are OPCs
.
Therefore, it is unclear whether OPCs that do not express Olig2 are actually present in the brain.

If so, it is unclear whether their numbers remain constant or can be regulated? Do they function differently than Olig2-positive OPCs?

On October 29, 2022, Hyun-sook Baek and Frank Kirchhoff (Academician of the European Academy of Sciences) from the Department of Molecular Physiology at Saarland University in Germany were in the team A subset of OPCs do not express Olig2 during development which can be increased in the adult by brain injuries and complex motor was published in the GLIA publication learning"
.
Fang Lirun doctoral student of Saarland University in Germany is the first author of the paper, and the leader of Bai Xianshu's research group at Saarland University and Professor Frank Kirchhoff are the co-corresponding authors
of the paper.
The study used the OPC-specific antibody platelet derived growth factor receptor alpha (PDGFR α α) to bind transgenic mice It was found that in the brain of normal mice, a group of oligodendrocytes did not express Olig2, and these cells were enriched in the brain of young mice, and their number decreased with age and almost disappeared
.
Interestingly, the number of this particular group of OPCs increased
significantly during brain injury or under study training.

The authors tagged OPCs with the oligodendropod progenitor cell (OPC)-specific antibody PDGFRα, which was found to occur after birth (postnatal,).
p) PDGFRα pos can be detected in different brain regions within the 14 mouse brain, including cortex, corpus callosum, hippocampal CA1, and dentate gyrus Cells of Olig2^neg (Figure 1).

To confirm that Olig2 negatives were not caused by antigen disruption caused by PFA fixation, the authors used three different antigen retrieval methods
.
After antigen retrieval by different methods, about 3.
5% of PDGFRα^pos cells were detected in the cortex not expressing Olig2, prompting Olig2^negPDGFRα^pos cells do exist in the brains of healthy mice.

Figure 1 In the brain of juvenile mice, Olig2-negative oligodendrogliome cells are present

(Source: Fang LP, et al.
, GLIA, 2022).

To further confirm that these PDGFRα^posOlig2^neg cells are indeed OPCs, the authors performed immunostaining using specific antibodies for different cell types.
All of these particular OPCs were found to express the NG2 protein (OPC-specific marker) but hardly another OL Lineage marker Sox10
.
At the same time, these cells do not express specific markers for other types of cells, including neurons (NeuN), astrocyte (GFAP/glutamine synthetase, GS), microglia ( IBA1/CD68), precursor cells (Nestin/Sox2/double cortin, DCX), oligodendrocytes (CC1).

This proves that these PDGFRα^posOlig2^neg cells are indeed OPCs
.
To further validate this conclusion, the authors used NG2-CreERT2xRosa26-fSTOPf-tdTomato double-transgenic mice to label OPCs
.
On day 14 after birth, the mice also detected tdTomato^posPDGFRα pos Olig2 neg cells in their brains (Figure 2), thus demonstrating that Olig2 is indeed not expressed in the brains of healthy mice OPC of transcription factors.

Figure 2The cells of PDGFRα^posOlig2^neg are indeed oligodendrocytes

(Source: Fang LP, et al.
, GLIA, 2022).

To explore whether these cell populations persist throughout life, the authors collected mouse brains of different ages for staining analysis
.
The authors found that Olig2^negOPC was detectable in multiple brain regions of p5 young mice, and its cell density increased with age, peaking at p14
。 Since then, however, the density has decreased with age, and such cells are hardly visible in the brains of adult mice (Figure 3).

Remarkably, in embryonic mouse brains, the authors also observed the presence of Olig2^negOPC in about 7 percent
.

Fig.
3Olig2-negative oligodendrocytes appeared in large numbers in the brains of young mice

(Source: Fang LP, et al.
, GLIA, 2022).

To explain why the number of these cells changed, the authors first immunostained these cells with a marker of apoptosis (cleaved caspase 3, CC-3) and a microglial phagocytosis marker (CD68), but CC-3 was not detected in these cells CD68-positive phagosomes were also not observed around cells, ruling out a decrease in cell density due to apoptosis or microglial phagocytosis.

Next, the authors used the BrdU assay to test the proliferative capacity
of such cells.
To distinguish between rapid and slow proliferation, the authors used two BrdU labeling methods, namely two-hour rapid labeling and three-day long-term labeling
.
When the authors gave p14 mice intraperitoneal single injection of BrdU two hours after immunostaining analysis, unlike 7% Olig2^posOPC None of the Olig2^negOPCs are marked with BrdU
.
When the authors injected BrdU at p12-14 for three consecutive days, there was about 10% Olig2^neg OPC incorporates BrdU, however Olig2^posOPC has a labeling rate of around 29% ( Figure 4).

These results suggest that the proliferation rate of Olig2 neg OPC itself may be very slow, and it is more likely to be converted from Olig2 pos OPC with a faster value-added rate
.
To further confirm whether Olig2 neg OPC proliferates or not, the authors examined whether Olig2^negOPC expresses a mitotic marker (Ki67).

The staining results showed that about 10% of Olig2 ^posOPCs in all brain regions (except for about 30% in the corpus callosum) were in the mitotic phase, while almost all Olig2^neg None of the OPCs showed Ki67 immunopositive, which confirmed that the proliferation rate of Olig2 neg OPC was significantly lower than that of Olig2 ^posOPC , presumably Olig2 neg OPC comes from the differentiation of Olig2^posOPC
.

Fig.
4Olig2-negative oligodendrogliome cells have extremely low proliferative activity

(Source: Fang LP, et al.
, GLIA, 2022).

Since Olig2 plays a crucial role in the differentiation of OPCs, can this group of Olig2-negative OPCs continue to differentiate into OLs? ？ During differentiation, OPCs extend more protrusions and their cell morphology becomes more complex
.
Therefore, the authors used the Imaris image analysis software to analyze and compare Olig2 ^neg and Olig2^pos Morphological structure
of OPC.
The authors found that Olig2^negOPC has shorter cell protrusions and fewer bifurcations, and is morphologically more inclined to precursor cells
.
These results suggest that Olig2^neg OPC is a group of OPC subsets that proliferate very slowly and prefer to remain in the precursor cell state.

Neural network setup occurs between the first and fourth weeks of life, and OPC is also involved in neural network establishment by forming synaptic connections with neurons (since p5) [6].

。 This time period coincides with
the emergence of Olig2 neg OPC.
To verify the causal relationship between the two, the authors used a model of acute brain injury and motor learning to change brain activity from two perspectives to observe changes in the number of Olig2^negOPCs
.
3 days after surgery for acute brain injury, the number of Olig2^negOPC at the site of cortical injury increases substantially, and in (at least) the postoperative period The presence of a large number of Olig2^negOPCs was still observed at the site of injury for 10 days (Figure 5).

。 In both stroke models and Kainic acid-induced epilepsy models, large amounts of Olig2^negOPC were also observed at the site of
injury.
These results show that in the brain of adult mice, altered brain activity can induce the formation
of Olig2 neg OPCs.

Since such cells are almost undetectable in the brains of healthy adult mice, and they appear rapidly under acute brain injury conditions, where do these cells come from? The above BrdU experimental results suggest that such cells may be transformed from Olig2^posOPC
.
To confirm this hypothesis, the authors chose to treat NG2-CreERT2 x Rosa26-fSTOPf-GCaMP3 transgenic mice (labelable OPCs) for acute brain injury
.
Mice drink BrdU-containing water for two weeks from the sixth week after birth (i.
e.
, two weeks before injury), so that cells (including OPCs) that have proliferated or have proliferated during this period will integrate BrdU
。 At 3 days postoperatively, the authors observed that all Olig2^negOPCs in the contralateral cortex were not labeled with BrdU, but had About 47.
5% of Olig2 pos OPC is labeled with BrdU, again proving Olig2^negOPC Does not proliferate or proliferates very slowly
.
However, in the cortex on the side of injury, the authors found that about 7 percent of Olig2 ^negOPCs were labeled with BrdU, suggesting that such cells are highly likely to be from Olig2^pos OPC transforms rather than multiplies
on its own.
To further validate the very low donor ability of Olig2 neg OPC, the authors examined whether Olig2neg OPC at the injury site expressed the mitotic phase marker Ki67
。 The immunostaining results showed that these cells did not express Ki67, confirming that Olig2^negOPC at the site of injury also exhibited very low proliferative activity It is possible that the cells are derived from Olig2^posOPC differentiation
.
This also shows that when the extracellular environment changes, Olig2 ^posOPC can shift its phenotype and reduce Olig2 expression levels
.

Figure 5: Acute brain injury induced a sharp increase in the number of Olig2-negative oligodendrogliome cells

(Source: Fang LP, et al.
, GLIA, 2022).

Since the emergence of Olig2 neg OPC is related to changes in the activity of neural circuits in the brain, can physiological conditions also promote the formation of Olig2^negOPC? To answer this question, the authors trained adult mice to perform complex movements on the Erasmus Lader for three weeks, that is, the mice learned to successfully walk through the Erasmus Lader
by overcoming roadblocks with or without warning.
During the learning process, the neurons in the hippocampal region of the mice are constantly stimulated by motor learning and form new neural circuits, and eventually these mice can quickly and efficiently overcome roadblocks and become "super-mice"
.
After three weeks of study, the number of Olig2^negOPCs increased significantly in the hippocampal region of these "superrats", especially in the dentate gyrus (Figure 6).

Thus, it was confirmed once again that the appearance of Olig2^negOPC was closely related
to the activity of neural circuits in the brain.

Figure 6 Motor learning increases the number of Olig2^negOPCs in the dentate gyrus of the hippocampus

(Source: Fang LP, et al.
, GLIA, 2022).

Figure 7 Downregulation of gene expression related to cell differentiation and myelination within Olig2^negOPC

(Source: Fang LP, et al.
, GLIA, 2022).

Unexpectedly, the authors analyzed the published single-cell sequencing results and found that there was also a group of OPCs that did not express Olig2
at the RNA level.
Gene Ontology analysis found that genes involved in cell differentiation and myelination were downregulated in Olig2^neg cells
.