Versatile and simple approach to determine astrocyte territories in mouse neocortex and hippocampus

Abstract

Background: Besides their neuronal support functions, astrocytes are active partners in neuronal information processing. The typical territorial structure of astrocytes (the volume of neuropil occupied by a single astrocyte) is pivotal for many aspects of glia-neuron interactions.

Methods: Individual astrocyte territorial volumes are measured by Golgi impregnation, and astrocyte densities are determined by S100β immunolabeling. These data are compared with results from conventionally applied methods such as dye filling and determination of the density of astrocyte networks by biocytin loading. Finally, we implemented our new approach to investigate age-related changes in astrocyte territories in the cortex and hippocampus of 5- and 21-month-old mice.

Results: The data obtained by our simplified approach based on Golgi impregnation were compared to previously published dye filling experiments, and yielded remarkably comparable results regarding astrocyte territorial volumes. Moreover, we found that almost all coupled astrocytes (as indicated by biocytin loading) were immunopositive for S100β. A first application of this new experimental approach gives insight in age-dependent changes in astrocyte territorial volumes. They increased with age, while cell densities remained stable. In 5-month-old mice, the overlap factor was close to 1, revealing little or no interdigitation of astrocyte territories. However, in 21-month-old mice, the overlap factor was more than 2, suggesting that processes of adjacent astrocytes interdigitate.

Conclusion: Here we verified the usability of a simple, versatile method for assessing astrocyte territories and the overlap factor between adjacent territories. Second, we found that there is an age-related increase in territorial volumes of astrocytes that leads to loss of the strict organization in non-overlapping territories. Future studies should elucidate the physiological relevance of this adaptive reaction of astrocytes in the aging brain and the methods presented in this study might be a powerful tool to do so.

Figure 1. Comparison of two methods to assess astrocyte territorial volume.

For description of the z-stack method, see the Methods chapter of the text. (A) The area of maximal extension of a hippocampal astrocyte was recorded on a Golgi-impregnated brain slice from an adult mouse using the reflection mode. The approximate borders of an astrocyte were encircled only considering clearly attached compartments (red line), and the enclosed area was calculated using the LSM software and was treated as that of a virtual circle, representing the central section of a virtual sphere. Next, the radius of this virtual circle could be calculated. This was taken to get the volume of the astrocyte territory. .Scale bar, 10 µm. (B) Two methods were compared by determining volumes of astrocyte territories on the same slices from adult mice (n = 4) for both cortex and hippocampus. Although astrocyte volume tended to be slightly underestimated when the calculation was based on the area of maximal extension, no significant difference was found. Each bar represents the values from 20–48 cells. Neoc., neocortex; Hippo., hippocampus.

Figure 2. Morphometric analysis of astrocytes in the neocortex and in the hippocampus.

(A) Representative scans from hippocampal astrocytes (stratum moleculare). Astrocytes were revealed by GFAP and S100β immunolabeling or by Golgi impregnation. Scale bars, 20 µm. (B) Volume of astrocyte territories calculated from the area of maximal extension in Golgi-impregnated brain slices. (C) The number of branchings per main process within 15 µm from the cell soma was determined. *P<0.05, **P<0.01 vs.values of adult mice. Each column includes cells from 3–5 mice. Absolute cell numbers are given in each column.

Figure 3. Astrocyte cell densities in hippocampal and cortical brain slices.

(A) Verification of S100β as suitable astrocyte marker to quantify cell densities. An individual astrocyte was filled with biocytin to reveal coupling within the network of protoplasmic astrocytes in the hippocampus. The same slice was labeled for S100β and GFAP. Scale bar, 50 µm. (B) The number of astrocytes stays constant between 5 and 21 months of age.

Figure 4. Overlap factor determined from neocortical and hippocampal brain slices.