Spatial organization of astrocyte clones: The role of developmental progenitor timing

Astrocytes represent a diverse and morphologically complex group of glial cells critical for shaping and maintaining nervous system homeostasis, as well as responding to injuries. Understanding the origins of astroglial heterogeneity, originated from a limited number of progenitors, has been the focus of many studies. Most of these investigations have centered on protoplasmic and pial astrocytes, while the clonal relationship of fibrous astrocytes or juxtavascular astrocytes has remained relatively unexplored. In this study, we sought to elucidate the morphological diversity and clonal distribution of astrocytes across adult cortical layers, with particular emphasis on their ontogenetic origins. Using the StarTrack lineage tracing tool, we explored the characteristics of adult astroglial clones derived from single and specific progenitors at various embryonic stages. Our results revealed a heterogeneous spatial distribution of astroglial clones, characterized by variations in location, clonal size, and rostro–caudal dispersion. While a considerable proportion of clones were confined within specific cortical layers, others displayed sibling cells crossing layer boundaries. Notably, we observed a correlation between clone location and developmental stage at earlier embryonic stages, although this relationship diminished in later stages. Fibrous astrocyte clones were exclusively confined to the corpus callosum. In contrast, protoplasmic or juxtavascular clones were located in either the upper or lower cortical layers, with certain clones displayed sibling cells distributed across both regions. Our findings underscore the developmental origins and spatial distribution of astroglial clones within cortical layers, providing new insights into the interplay between their morphology, clonal sizes, and progenitor heterogeneity.


| INTRODUCTION
Astrocytes are a type of glial cell with a complex morphology characterized by their star shape and extensive branching.These cells extend numerous processes that envelop adjacent neurons and blood vessels occupying discrete domains where their processes do not overlap.
The morphological diversity of astrocytes in various brain regions was originally described by Cajal over a century ago (Ramón y Cajal, 1913).Recent studies have further revealed significant heterogeneity, in terms of morphology, functional properties, and clonal arrangement.Astrocytes can be also classified such as protoplasmic, fibrous, juxtavascular, and pial astrocytes (Astiz et al., 2022;Bribián et al., 2016;Clavreul et al., 2019;García-Marqués & López-Mascaraque, 2013;Holt, 2023;Khakh & Deneen, 2019;Shen et al., 2021).Additionally, astrocytes differ in the expression of different markers.Protoplasmic astrocytes in the cortex are identified by the expression of GFAP and S100β, whereas hippocampal and pial astrocytes predominantly express GFAP (Zhang et al., 2019).Moreover, astrocytes also display variation in their transcriptional regulation of gene expression.In transgenic mice, the deletion of Olig2 affects astrocyte production differently in white and gray matter (Cai et al., 2007).Furthermore, immunoreactivity for dopamine receptors D1R and D4R has been observed to be more pronounced in pial astrocytes and layer I astrocytes, compared with protoplasmic astrocytes within cortical layers II-IV (Oda & Funato, 2023).This heterogeneity among astrocytes can be also attributed to their origins.The StarTrack methodology, utilizing the GFAP-promoter, has revealed distinct subtypes of sibling astrocytes emerging from different progenitors (García-Marqués & López-Mascaraque, 2013).However, studies using clonal method approaches such as UbC-(GFAP-PB)-StarTrack (Ojalvo-Sanz & López-Mascaraque, 2021) and MAGIC (Clavreul et al., 2019) demonstrated a common progenitor for pial and protoplasmic astrocytes.Astrocyte precursors colonize the neocortex in a nonlayered manner during perinatal stages (Clavreul et al., 2019), unlike the layered pattern observed in neurons (Bayraktar et al., 2020).The ontogenic relationship of astrocytes may have functional implications.For instance, subpopulations of juxtavascular astrocytes express Kiv4.3 based on their developmental origin, influencing their electrophysiological properties (Götz et al., 2021).Moreover, sibling astrocytes display a preference for establishing GAP junctions (Gutiérrez et al., 2019), which could contribute to the clonal astrocytic response to injury (Bribian et al., 2018).
While previous studies have explored astroglial clonal relationships, the distribution of sibling astrocytes across cortical layers, particularly for fibrous and juxtavascular astrocytes, remains understudied.To address this gap, our research explores the clonal origins of the less-investigated juxtavascular and fibrous astrocytes, extending beyond the predominant focus on protoplasmic and pial astrocytes.After targeting individual pallial neural progenitor cells (NPCs) at various embryonic stages, our analyses in adult mice brains (P30) reveal that diversity among astroglial clones extends to their compositions, spatial distribution, clonal size, and rostro-caudal (R-C) dispersion.These findings significantly enhance our understanding of astrocyte diversity, particularly within understudied subtypes.

| StarTrack plasmids
The StarTrack system facilitated the targeting of NPCs and the tracking of their cell progeny.This approach, based on PiggyBac technology, introduces an inheritable "color code" into NPCs and their offspring via 12 plasmids encoding six fluorescent proteins, both nuclear and cytoplasmic (García-Marqués & López-Mascaraque, 2013).
In brief, the transposase recognizes the inverted terminal repeats of the StarTrack plasmids and integrates the fluorophore sequence into NPCs in a stable and heritable manner.This stochastic integration results in over 16 million theoretical combinations, considering distinct color code, fluorophore cellular location, and intensity fluorescence levels (Figueres-Onãte et al., 2016).Thus, sibling cells derived from the same NPC can be identified according to their fluorophore code, fluorophore localization, and intensity, providing a potent tool for studying the fate and cell potential of single progenitors.
Additionally, plasmids were coelectroporated with Cre-ERT2 plasmid to remove episomal copies.

| In utero electroporation and tamoxifen administration
In utero electroporations were performed at different embryonic stages (E12, E14, and E16).Pregnant mice were anesthetized with 3% isofluorane/O2 and maintained at 2% in a sterile surgical area at 37 C. Subcutaneous injections of the antibiotic Baytril (5 mg/kg; Bayer) and the anti-inflammatory/analgesic Meloxicam (300 μg/kg: VITA Laboratories) were administrated.A skin incision was made in the abdominal area after cleaning with ethanol 70% and saline 0.9%.
The uterine horns were exposed, and the StarTrack plasmid mixture was injected into the lateral ventricle (LV, 0.5 μg/μL in distilled water).
Fast Green (0.2%) was used to confirm the successful filling of the LV.To facilitate plasmid entry into the cells, five square electric pulses of 50 ms were delivered at different voltages (33 mV at E12, 35 mV at E14, and 37 mV at E16) with electrode tweezers at 950 ms intervals between pulses.Pallial progenitors were targeted with the positive electrode.After the procedure, the embryos were placed back into the abdominal area, and the incision was closed.The embryos were allowed to develop until P30 for subsequent analysis.At P8, a single intraperitoneal injection of tamoxifen (Sigma-Aldrich T5648-1G, 20 mg/mL in corn oil) was administered at a dose of 7.5 mg/kg body weight to induce Cre recombinase activity and to remove the nonintegrated plasmid copies.
After blocking with 5% normal goat serum in PBS-T 0.1%, the sections were incubated overnight at 4 C with primary antibodies (see Table 1 for details).The primary antibodies used included S100 calcium-binding protein beta (S100β) as an astrocyte marker, Lycopersicon esculentum (tomato) Lectin (TL) as markers for blood vessels, and chicken ovalbumin upstream promoter transcription factor-interacting proteins 2 (CTIP-2) as a marker for cortical V-VI layers.Subsequently, the sections were washed again and incubated in the corresponding secondary antibody for up to 2 h (see Table 1).

| Image analysis and processing
To verify the StarTrack labeling, sections were examined under an epifluorescence microscope.The six fluorescent proteins were captured, in separate channels, for each section to avoid overlapping.
The imaging was performed using a TCS-SP5 confocal microscope (Leica, TCS-SP5).The confocal lines were set between 15% and 30% of intensity, and the excitation (Ex) and emission (Em) wavelengths in nanometers were as follows: mT-Sapphire (Ex: 405; Em:

| Statistical analysis
Statistical comparisons and data visualization were carried out using RStudio (version 1.4.1106) and GraphPad Prism (version 6.0).The normality of the data was evaluated using the Kolmogorov-Smirnov test with Dallal Wilkinson-Liffie for p-value calculation.T-test was applied for comparing two groups, while the one-way ANOVA test was used for comparisons involving more than two groups.When data display a non-Gaussian distribution, nonparametric tests were employed.The Mann-Whitney test was applied for comparing two groups, while the Kruskal-Wallis test was used for comparisons involving more than two groups.Boxplot graphs include all data points and represent the minimum to maximum range, along with the median.A confidence interval of 95% was adopted, and statistical significance was determined using specific critical values: * for p < .05,** for p < .01,*** for p < .001,and **** for p < .0001.
T A B L E 1 List of primary and secondary antibodies for the molecular characterization of neural cells.

| Morphological diversity of astroglial clones in cortical layering organization
To explore the distribution of sibling astrocytes across cortical layers, we used the lineage tracing tool StarTrack, which allows targeting single progenitors and tracking their entire cell progeny.We used the UbC-(NG2 or GFAP-PB)-StarTrack to tag NPCs due to no significant differences were observed in the clonal size of astroglial clones derived from NG2-and GFAP-progenitors (Figure S1).By targeting NPCs at various developmental stages (E12, E14, and E16), we analyzed astroglial clones in the adult mice brain (P30).For each animal, we selected astroglial clusters in the upper and the lower layers, as well as pial and fibrous astrocytes.We examined all sibling astrocytes belonging to a clone regardless of their locations along the R-C axis in the 3D reconstructed brains.Our analysis revealed a morphological heterogeneity, among astroglial clones (Figure 1A), including protoplasmic astrocytes (Figure 1aA-D), fibrous astrocytes (Figure 1aE), juxtavascular astrocytes (Figure 1 aH), and pial astrocytes (Figure 1aD,F,G).
Within some clones, a clear correlation between morphology and clonal identity was evident.Sibling fibrous astrocytes predominantly resided in the white matter and displayed elongated, thin, and unbranched processes (Figure 1aE).Protoplasmic astrocytes, on the other hand, occupied the gray matter and exhibited numerous branching processes.Given the distinct characteristics in cell density, synaptic connections, and neuronal function across cortical layers, we differentiate between upper and lower layers, recognizing their potential role in astroglial clonal generation.In certain cases, protoplasmic sibling astrocytes were exclusively located in the upper cortical layer (Figure 1aA), which corresponds with regions over or in the granular layer (layers I-IV) characterized by a higher cell density and extensive synaptic connections.In other cases, protoplasmic sibling astrocytes were placed in the lower cortical layer (Figure 1aB) termed infragranular layers that contain larger pyramidal neurons and play a crucial role in transmitting signals to subcortical structures.Despite the initial selection of astroglial clusters within distinct regions, some sibling cells transcended these boundaries and were distributed across both upper and lower layers (Figure 1aC).Pial astrocytes were in contact with the pial surface, forming a tangentially oriented sheet with a fibroblast-like appearance (Figure 1aF).These cells formed exclusively pial astrocyte clones but others also had sibling protoplasmic astrocytes placed just in layer I (Figure 1aG) or spanning across upper and/or lower layers (Figure 1aD).Furthermore, sibling juxtavascular astrocytes enveloped some blood vessels (Figure 1aH).This diversity highlights the intricate organization of sibling astrocytes within cortical layers.
We analyzed a total of 210 astroglial clones from 18 adult mice.
In each animal, astroglial clones were preferentially restricted to one region (p = .003,Figure 1b), meaning that sibling cells along the R-C axis were confined to a single region.Out of the total number of clones, 60% (N = 127) exhibited regional confinement (Figure 1b,c) to the upper cortical layers, lower cortical layers, corpus callosum, or pial surface.Conversely, the remaining 40% of clones (N = 83) displayed a lack of regional confinement, extending across these anatomical boundaries (Figure 1b,c).These areas included both upper and lower layers, layer I with contact to the pia or upper and/or lower layers while maintaining contact with the pial surface.

| Cortical layering distribution of astroglial clones according to their ontogenetic origins
We further investigated clones generated from NPCs at various developmental stages.We analyzed a total of 68 clones from E12-progenitors (Figure 2a), 62 clones from E14-progenitors (Figure 2b), and 80 clones from E16-progenitors (Figure 2c).In each animal astroglial clones derived from E12 progenitors exhibited a preference for regional restriction (p = .023,Figure 2a).However, in the case of those arising from E14 and E16-progenitors (Figure 2b,c), comparable numbers of clones were regionally restricted, or dispersed across various areas.Our analysis revealed no statistically significant differences in the regional pattern distribution (E14: p = .400and E16: p = .119).
Out of the total number of clones derived from E12-progenitors, the 68% of the clones (N = 46) were limited to one region (Figure 2a,d).
Within this latter group, half of the clones (55%, N = 12) consisted of both pial and protoplasmic astrocytes, while the remaining half (45%, N = 10) represented astroglial clones with sibling cells dispersed across the upper and lower cortical layers (Figure 2d).Astrocytes derived from E14 progenitors exhibited a comparable distribution of clones, showing either localized confinement or dispersion across multiple locations.
Specifically, 56% (N = 35) of clones were regionally restricted, while 44% (N = 27) were nonrestricted clones (Figure 2b,e).Within the nonrestricted group, the proportions resembled those observed with E12 progenitors.Notably, 56% (N = 15) of these clones contained both pial and protoplasmic astrocytes, while 44% (N = 12) represented astroglial clones with sibling cells spanning the upper and lower cortical layers (Figure 2e).Continuing with the analyses from E16-progenitors, the results mirrored those obtained from E14-progenitors.An equivalent quantity of clones with regional restrictions or dispersed across diverse regions was identified.Specifically, 58% (N = 46) of the clones exhibited regional restriction, while the remaining 43% (N = 34) constituted nonrestricted clones (Figure 2c,f).Within the nonrestricted group, the proportions were similar to data observed from E12 and E14 progenitors.In particular, half of these clones (53%, N = 18) were astroglial clones with sibling cells distributed across the upper and lower cortical layers, while the other half (47%, N = 16) were composed of both pial and protoplasmic astrocytes (Figure 2f).

| Smaller clonal size of regionally restricted astroglial clones
The clonal size of astroglial clones, which represents the number of sibling cells within each clone, was evaluated.Regionally restricted clones displayed a significantly lower number of sibling astrocytes ( p < .0001)when compared with regionally nonrestricted clones (Figure 3a).Specifically, regionally restricted clones, confined within a single region, consisted of 2 to 27 astrocytes per clone, with a median of 5 cells/clone (Figure 3a,b).In contrast, clones with cells spanning more than one region comprised 2 to 45 cells per clone, with a median of 8 cells/clone (Figure 3a,c).
Notably, no significant differences were observed in the clonal size of either regionally-restricted ( p = .335,Figure 3d) or

| Larger size and dispersion of nonrestricted juxtavascular astroglial clones
Sibling cells of juxtavascular astroglial clones are closely associated with blood vessels.In Figure 5a, sibling astrocytes surrounding a blood vessel are located in two consecutive slices (100 μm).These types of clones were found exclusively in either the upper or the lower layers, or in both upper and lower layers.In each animal, juxtavascular clones exhibited distinct preference for confinement within a specific region ( p = .0147,Figure 5b).Most juxtavascular clones exhibited regional restrictions (66%; N = 37), while the rest were nonrestricted (34%; 5b, inset).When considering the developmental stage of the progenitors, a clear restricted pattern was observed in clones from E14-progenitors (p = .031,Figure 5b).In this group, 80% of clones (N = 13) were confined to one region, while 20% (N = 3) were dispersed across upper and lower layers.Conversely, clones from earlier (E12) and later (E16) stages did not show a preference for restriction or nonrestriction ( p = .250and p = .596,respectively, Figure 5c).
The clonal size exhibits an increment correlated with the R-C dispersion (Figure 5d) in both restricted and nonrestricted clones, although nonrestricted clones exhibited larger sizes (p < .0001, Figure 5e).
Restricted clones had a median of 6 cells/clone, while nonrestricted clones reach to 12 cells/clone (Figure 5e).Restricted clones had a consistent clonal size regardless of the progenitor stages (Figure 5f, purple).
Similarly, nonrestricted neither had differences in the clonal size according to their progenitor timing (Figure 5f, green).Considering the total number of clones originated from specific progenitor time points (Figure 5f), nonrestricted was larger than restricted ones in E12 (p = .0003)and E16 (p = .013)but not in E14 (p = .446).In relation to the R-C dispersion, nonrestricted clones showed greater dispersion (p = .005,Figure 5g,h) than restricted clones.However, there were not significant differences related to the progenitor stage (Figure 5h).

| DISCUSSION
In this article, we present a comprehensive analysis of astroglial clones and their morphological distribution within cortical layers, with a particular emphasis on the detailed examination of fibrous and juxta- The investigation into the clonal relationships of fibrous or juxtavascular astrocytes has remained relatively unexplored.Addressing this gap, our findings reveal diverse aspects of astroglial clones, significantly advancing our understanding of astrocytes diversity, and highlighting the less-explored juxtavascular and fibrous astrocytes.Our approach involved the labeling of individual pallial-NPCs at various embryonic stages (E12, E14, and E16).Subsequently, we examined astroglial clones in the brains of adult mice at P30, as substantial cellular rearrangements are known not to occur before this period (Clavreul et al., 2019).
Astroglial clones displayed a variety of patterns, consistent with previous findings (Clavreul et al., 2019;García-Marqués & López-Mascaraque, 2013;Ojalvo-Sanz & López-Mascaraque, 2021;Shen et al., 2021).While most astrocytes derived from the same progenitor tended to be localized within a single cortical layer, sibling astrocytes were not exclusively confined to one specific region (Figure 6).Different studies have proposed different arrangements for astroglial clones, including ordered columnar distributions (Magavi et al., 2012) and dispersion into multiple clusters (Clavreul et al., 2019;Shen et al., 2021).According to these results, restricted clones appeared to possibly following a columnar arrangement.This spatial organization could potentially carry functional implications, similar to the functional units observed in neuronal columnar structures (Noctor et al., 2001).
It is important to note that molecular and morphological differences exist between astrocytes located in the upper and lower layers, and these disparities have been linked to their interactions with synapses.
These interactions could potentially impact processes such as glutamate clearance and synaptic plasticity modulation (Lanjakornsiripan et al., 2018).
The presence of progenitors with limited differentiation capacity could explain the diverse spatial arrangement of astroglial clones and their clonal sizes (Clavreul et al., 2019;García-Marqués & López-Mascaraque, 2013;Ojalvo-Sanz & López-Mascaraque, 2021;Shen et al., 2021).Previous studies have indicated that an intermediate pro- genitor typically generates a unit of around 2-3 astrocytes (Shen et al., 2021), and each clone can consist of two or three cell units (Clavreul et al., 2019).The relatively smaller number of astrocytes within a clone specifically localized to a cortical layer could arise from having fewer cell units.Clones restricted to specific regions appear to be composed of one or two clusters, while those spanning regions (nonrestricted clones) may be originated from intermediate progenitors, that, through dynamic expansion of sibling cells, generate more than two clusters.Although we did not observe significant differences between clones exclusively placed in the upper or lower layers, certain studies suggest a lower number of astrocytes in clones located within lower layers (Clavreul et al., 2019).In contrast, clones within the upper or lower cortical layers were larger than those committed to the corpus callosum.Notably, astrocytes in the gray matter exhibit different characteristics compared with those in white matter, particularly in terms of their interaction with neurons, glutamate, and energy metabolism (Köhler et al., 2019).For example, while a single fine astroglial process of astrocytes in gray matter can establish contact with as many as 200,000 synapses, in white matter, astrocytes interact with Ranvier nodes (Köhler et al., 2019;Oberheim et al., 2006;Sofroniew & Vinters, 2010).This disparity contributes to the elevated expression of glutamate enzyme metabolism in gray matter astrocytes, resulting in more rapid glutamate clearance (Hassel et al., 2003;Regan et al., 2007).These greater energy demands could potentially influence the proliferation of astroglial progenitors in gray matter, leading to the generation of more sibling cells.The functional differences between astrocytes in these two zones might be attributed to differences in the origin of protoplasmic and fibrous astrocytes (Bribián et al., 2016;García-Marqués & López-Mascaraque, 2013;Luskin & McDermott, 1994).
Since their initial descriptions, astrocytes have been known to establish intimate connections with blood vessels (Ramón y Cajal, 1896;Virchow, 1846).Recent investigations suggest that most gray matter astrocytes link with at least one blood vessel (Hösli et al., 2022).However, emerging studies have also highlighted the heterogeneity in astrocyte end-feet architecture, suggesting diverse modes of interaction with blood vessels (Kameyama et al., 2023;Kubotera et al., 2019).Intriguingly, following the disruption of end feet or the stalk of astrocytic processes, some astrocytes exhibit the capacity to regenerate these structures while others do not (Kubotera et al., 2019).Furthermore, detailed analysis using mass spectrometry has unveiled a spectrum of sizes and molecular components within astrocyte end feet, indicating potential functional diversity in their interaction with blood vessels (Kameyama et al., 2023).This underscores a crucial distinction between astrocytes capable of developing end feet for direct contact with vessels and those whose somas maintain tight associations with blood vessels (Kubotera et al., 2019).
Given that astrocyte generation parallels blood vessel establishment during development (Paredes et al., 2018), ontogeny likely plays a pivotal role in determining the formation of juxtavascular astrocytes.
Notably, accumulating evidence, including this study and previous research, supports the existence of a shared progenitor for juxtavascular astrocytes (Bribián et al., 2016;García-Marqués & López-Mascaraque, 2013;Götz et al., 2021).In response to injury, such as a stab wound, the dynamic behavior of astrocytes plays a crucial role in the brain's response to damage.Studies have shown that a subset of astrocytes, particularly those closely associated with blood vessels, demonstrates heightened proliferation in reaction to injury, likely contributing to the repair process (Bardehle et al., 2013).This proliferation response suggests a specialized role for these juxtavascular astrocytes in injury recovery, possibly involving the reinforcement of the blood-brain barrier and the provision of support to damaged neurons.Additionally, a specific subset of astrocytes expressing Mlc-1, a marker associated with vascularity, may possess specialized functions related to vascular regulation or response to vascular damage (Morales et al., 2022).However, given the heterogeneity of astrocytes, further experiments are needed to rule out a shared origin with protoplasmic astrocytes.
In the gray matter, certain astrocytes establish contact with the pial surface and are referred to as pial astrocytes or interlaminar astrocytes (Falcone et al., 2021;García-Marqués & López-Mascaraque, 2013).These cells have been identified in rodents and primates.Initially, it was suggested that pial astrocytes originated from a distinct lineage separate from protoplasmic astrocytes (García-Marqués & López-Mascaraque, 2013).
However, subsequent experiments revealed a common progenitor for both pial and protoplasmic astrocytes (Clavreul et al., 2019;Ojalvo-Sanz & López-Mascaraque, 2021).This apparent contradiction may be attributed to the use of piggyBac plasmids under the GFAP promoter used in the initial study (García-Marqués & López-Mascaraque, 2013).It is known that while GFAP is strongly expressed in pial astrocytes, not all protoplasmic astrocytes are labeled with GFAP (Zhang et al., 2019).Two distinct patterns have been observed in this type of clone.In many instances, protoplasmic astrocytes are localized in layer I, while in other cases, astrocytes are also found in upper and/or lower layers.A recent study in the rat cerebral cortex revealed that pial and protoplasmic astrocytes in layer I share a stronger immunoreactivity for D1R and D4R, which could indicate the potential influence of the dopaminergic system on the activity of these astrocytes (Oda & Funato, 2023).
An increasing number of studies have revealed significant differences between various subclusters of astrocytes (Batiuk et al., 2020;Bayraktar et al., 2015;Götz et al., 2021;Lee et al., 2006;Morel et al., 2017;Oberheim et al., 2012;Oda & Funato, 2023;Zhang & Barres, 2010).For example, single-cell transcriptomic analyses of the cortex have identified five different types of astrocytes, each with a specific genetic signature (Batiuk et al., 2020).These studies have also revealed astroglial transcriptomic layering that differs from neuronal layering (Bayraktar et al., 2020).The heterogeneous morphological distribution of astroglial clones, as well as the expression of specific proteins and genes, might be linked to distinct functional roles.Interestingly, Olig2, a traditional marker for oligodendroglial lineage, has been observed in certain subpopulations of astrocytes (Ohayon et al., 2021;Wang et al., 2021).Studies involving transgenic mice have shown that the deletion of Olig2 affects the generation of astrocytes in the white matter while promoting their production in the gray matter (Cai et al., 2007).Furthermore, Pax6 and Nkx6.1 expression have been identified in subsets of astrocytes in the white matter of the spinal cord (Hochstim et al., 2008).The combination of these markers along with reelin and slit has enabled the definition of three distinct populations of astrocytes (Tsai et al., 2012).These molecular differences among astrocyte subpopulations could be influenced by their ontogeny and could potentially modulate their responses.For instance, the shared expression of Kiv4.3 by sibling juxtavascular astrocytes has been shown to modify their electrophysiological properties (Götz et al., 2021).Additionally, in a multiple sclerosis model, astrocytes exhibited a clonal response to a lesion (Bribian et al., 2018), which might be mediated by preferential communication through GAP junctions between sibling astrocytes (Gutiérrez et al., 2019).
In summary, the study highlights the significant variability present among cortical sibling astrocytes in terms of their composition, organization, and clonal size, which are shaped by the timing of their progenitors.These findings enhance our understanding of the complex layer-specific arrangement of astroglial cortical clones and shed light on the diverse nature of both astrocytes and their precursor cells.
Additionally, the research contributes valuable insights into the characteristics of fibrous and juxtavascular astrocytes.While further research is necessary to uncover the functional roles of these distinct astrocyte populations, a more comprehensive comprehension of astrocyte development lays the foundation for potential future interventions aimed at modulating the behavior of specific astrocyte subtypes.
Wild-type C57BL/6 mice were raised at the Cajal Institute animal facility.The study was conducted following the guidelines of the Spanish Animal Care and Use Committee according to the European Union (2010/63/EU).All experiments were approved by the Bioethics Committee of the Community of Madrid (PROEX 314/19).Mice were on a 12-h light-dark cycle, and room temperature was maintained at 22 C.The gestation period in mice lasted 19 days and the day of vaginal plug detection was defined as the first embryonic day 0 (E0) and the day of birth was defined as postnatal day 0 (P0).A total of 18 mice from 12 different litters were included in the study.For each stage, n = 6 animals were utilized in the experiments, sourced from at least three distinct litters.
520-535), mCerulean (Ex: 458; Em: 468-480), EGFP (Ex: 488; Em: 498-510), YFP (Ex:514; Em: 525-535), mKO (Ex: 514; Em: 560-580), mCherry (Ex: 561; Em: 601-620), and Alexa Fluor 633/647 (Ex: 633; Em: 650-760).The maximum projection of the images was generated using LasX software (Leica Application Suite X, Version 3.5.1),and a 3D reconstruction was created for each animal.Supervised clonal analysis was performed using a Fiji macro developed in the Scientific and Microscopy Image Unit of the Cajal Institute (Madrid, Spain).A barcode was generated for each cell based on the presence or absence of the fluorescent proteins (YFP, mKO, mCerulean mCherry, mTSapphire, and EGFP).The nuclear or cytoplasmic location of the fluorophore was determined visually.Cells with the same fluorophores code and fluorophores location were classified as sibling cells.Cell types were determined based on morphological and immunohistochemical characteristics.The region spanning from the first to the last section containing fluorescent cells was defined as the electroporated area.The first slice with StarTrack labeled cells was designated as "0 μm," and subsequent slices were incremented by 50 μm (corresponding to the section thickness).
nonrestricted ( p = .177,Figure3e) astroglial clones, after targeting NPCs at different developmental stages.However, significant differences emerged among various types of regionally restricted astroglial clones (p < .0001,Figure3f).Clones located in the lower or upper layers were statistically larger than the fibrous corpus callosum (CC) astroglial clones ( p = .0002and p < .0001,respectively).Astroglial clones confined to the CC consisted of 2 to 8 cells per clone (median: 3 cells/clone), while those in either lower or upper layers, F I G U R E 1 Cortical layering organization of sibling astrocytes.(a) Scheme representing the morphological characteristics of astroglial clones.(A) Sibling fibrous astrocytes.(B) Protoplasmic astroglial clone exclusively located in the upper layers.(C) Protoplasmic astroglial clone located in the lower layers.(D) Clone of protoplasmic astrocytes with sibling cells in upper and lower cortical layers.(E) Pial astrocytes clone.(F) Clone of astrocytes in contact with the pial surface with siblings in layer I. (G) Astroglial clone with sibling cells occupying upper and lower layers along with pial astrocytes.(H) Clone of juxtavascular astrocytes located in upper layers.(I) Sibling juxtavascular astrocytes distributed across the upper and lower cortical layers.(b) Distribution of regionally restricted or nonrestricted clones.(c) Distribution of sibling cells for both restricted (clones 1 to 9) and nonrestricted (clones 10 to 14) astroglial clones.Each data point represents an individual cell.I: layer I, II-IV: upper layers, V-VI: lower layers.CC, corpus callosum.Scale bar: 50 μm.rangedfrom 2 to 14 or 2 to 27 cells, respectively, with a median of 5 cells/clone (Figure3f).Additionally, clones exclusively composed of pial astrocytes contained 4 to 10 cells per clone (median: 6 cells/ clone).Regarding nonrestricted clones, significant differences (p < .0001)were identified between nonrestricted clones composed of pial+protoplasmic astrocytes and astroglial clones dispersed across the upper and lower layers (Figure3g).Clones comprising pial+protoplasmic astrocytes contained 2 to 45 cells per clone (median: 7 cells/clone), whereas upper+lower astroglial clones comprised 3 to 33 cells per clone (median: 11 cells/clone).Distinct spatial patterns were observed within clones composed of pial and protoplasmic astroglial clones.Some clones were predominantly composed of pial astrocytes with protoplasmic astrocytes primarily located in layer I, while others displayed protoplasmic astrocytes in different cortical layers (upper/lower, data not shown).These spatial variations correlated with differences (p = .0069)in clonal sizes.Pial+layer I astroglial clones tended to be smaller, with 2-12 cells per clone (median: 5 cells/clone), whereas pial +upper/lower protoplasmic astroglial clones exhibited a larger size, with 3-45 cells per clone (median: 8 cell/clone, data not shown).

3. 4 |
Regionally restricted astroglial clones are less R-C dispersed than nonrestricted Upon performing 3D reconstruction of mouse brains, we measured the distance between the first and last sibling astrocyte, within a given clone, to assess the R-C dispersion, as illustrated in Figure 4a.Despite the extension of the electroporated region spanning seven slices, the blue cells formed a clone that extended across only four slices, resulting in an R-C dispersion of 200 μm.Astroglial clones with sibling cells regionally restricted to a single layer along the R-C axis displayed significantly lower dispersion (p = .0016)compared with regionally nonrestricted clones (Figure 4b).Most regionally restricted clones exhibited a dispersion of approximately 50 μm (Figure 4b,c), whereas F I G U R E 2 Diversity in astroglial clonal distribution according to their ontogeny.(a-c) Number of clones regionally restricted or nonrestricted after targeting progenitors at different embryonic stages.(d) Representation of sibling astrocytes in regionally restricted or nonrestricted clones derived from E12-RGPs.Each data point represents a cell in a clone.Column represents the distribution of sibling cells in a clone.(e) Representation of sibling astrocytes in regionally restricted or nonrestricted clones derived from E14-RGPs.(f) Representation of sibling astrocytes in regionally restricted or nonrestricted clones derived from E16-RGPs.nonrestricted clones displayed a dispersion of about 100 μm (Figure 4b,d).Notably, differences were observed (p = .036)in the R-C dispersion among regionally restricted clones derived from NPCs at different developmental stages (Figure4e).In clones generated from E14-NPCs, the median R-C dispersion was 100 μm, while clones derived from E16 were significantly less spread (median: 50μm.p = .038).Additionally, astroglial clones restricted to the CC, lower layers, upper layers, or those in contact with the pial surface displayed similar dispersions ( p = .372,Figure4f).Specifically, clones within the CC showed a median dispersion of 50 μm, clones in lower layers exhibited 100 μm, clones in upper layers 50 μm, and those in contact with the pial surface had a dispersion of 100 μm.Regarding nonrestricted clones, the R-C dispersion did not vary depending on the developmental stage ( p = .184,Figure4g), displaying a median dispersion of 100 μm in all stages.However, the R-C dispersion of astroglial clones, containing both pial+upper/lower astrocytes, differed from nonrestricted astroglial clones located in the upper and lower layer ( p = .0006,Figure4h).Clones composed of pial and protoplasmic astrocytes exhibited a median dispersion of 50 μm, whereas astroglial clones with sibling cells in the upper and lower layers displayed a median dispersion of 100 μm.

F
I G U R E 3 Clonal sizes of regionally restricted and nonrestricted astroglial clones.(a) Comparison of clonal sizes between regionally restricted and nonrestricted sibling astrocytes.(b) Representative image showing a regionally restricted astroglial clone.(c) Representative image of a regionally nonrestricted astroglial clone.(d) Change in clonal sizes in regionally restricted astroglial clones during development.(e) Variation in clonal size in regionally nonrestricted astroglial clones during development.(f) Clonal size distribution of regionally restricted astroglial clones categorized by their cortical layer locations.(g) Clonal size distribution of regionally nonrestricted astroglial clones categorized by their cortical layer locations.LL, lower layers; UL, upper layers.Scale bar: 50 μm.*p < .05,**p < .01,***p < .001,****p < .00001.F I G U R E 4 Rostro-caudal (R-C) dispersion of regionally restricted and nonrestricted astroglial clones.(a) Schematic illustration of R-C dispersion.(b) Comparison of R-C dispersion between regionally restricted and nonrestricted sibling astrocytes.(c) Microscopic image showing regionally restricted sibling astrocytes.(d) Microscopic image illustrating an astroglial clone with cells distributed across upper and lower layers.(e) R-C dispersion of regionally restricted astroglial clones derived from progenitor at different developmental stages.(f) R-C dispersion of regionally restricted astroglial categorized by their cortical layer locations.(g) R-C dispersion of regionally nonrestricted astroglial clones derived from progenitor at different developmental stages.(h) R-C dispersion of regionally nonrestricted astroglial clones categorized by their cortical layer locations.II-IV: Upper L. (Upper layers); V-VI: Lower L. (Lower layers).Pia denotes astrocytes in contact with the pial surface, R-C dispersion.CC, corpus callosum.Scale bar: 50 μm.*p< .05,***p < .001.F I G U R E 5 Regionally restricted and nonrestricted juxtavascular astroglial clones.(a) Distribution of juxtavascular astroglial clones, categorized as regionally restricted or nonrestricted.(b) The number of regionally restricted or nonrestricted juxtavascular astroglial clones after targeting RGPs at different developmental stages.(c) Correlation between clonal sizes and rostro-caudal (R-C) dispersion in juxtavascular astroglial clones.(d) Comparative clonal sizes of regionally restricted and nonrestricted juxtavascular astroglial clones.(e) Clonal sizes of regionally restricted juxtavascular astroglial clones across various stages of development.(f) Clonal sizes of regionally nonrestricted juxtavascular astroglial clones across various stages of development.(g) R-C dispersion of regionally restricted and nonrestricted juxtavascular astroglial clones.(h) R-C dispersion of regionally restricted juxtavascular astroglial clones across different developmental stages.(i) R-C dispersion of regionally nonrestricted juxtavascular astroglial clones across various stages of development.
be constituted by a more cohesive group of cells, whereas nonrestricted clones often exhibited a distribution across several clusters, F I G U R E 6 Clonal size and cortical layer arrangement of astroglial clones.Schematic representation of clonal size and cortical layer arrangement of regionally restricted and nonrestricted astroglial clones derived from E12-NPCs (a), E14-NPCs (b), and E16-NPCs (c).