Dilation of cortical capillaries is not related to astrocyte calcium signaling

Abstract

The brain requires an adequate supply of oxygen and nutrients to maintain proper function as neuronal activity varies. This is achieved, in part, through neurovascular coupling mechanisms that mediate local increases in blood flow through the dilation of arterioles and capillaries. The role of astrocytes in mediating this functional hyperemia response is controversial. Specifically, the function of astrocyte Ca²⁺ signaling is unclear. Cortical arterioles dilate in the absence of astrocyte Ca²⁺ signaling, but previous work suggests that Ca²⁺ increases are necessary for capillary dilation. This question has not been fully addressed in vivo, however, and we have reexamined the role of astrocyte Ca²⁺ signaling in vessel dilation in the barrel cortex of awake, behaving mice. We recorded evoked vessel dilations and astrocyte Ca²⁺ signaling in response to whisker stimulation. Experiments were carried out on WT and IP3R2 KO mice, a transgenic model where astrocyte Ca²⁺ signaling is substantially reduced. Compared to WT mice at rest, Ca²⁺ signaling in astrocyte endfeet contacting capillaries increased by 240% when whisker stimulation evoked running. In contrast, Ca²⁺ signaling was reduced to 9% of WT values in IP3R2 KO mice. In all three conditions, however, the amplitude of capillary dilation was largely unchanged. In addition, the latency to the onset of astrocyte Ca²⁺ signaling lagged behind dilation onset in most trials, although a subset of rapid onset Ca²⁺ events with latencies as short as 0.15 s occurred. In summary, we found that whisker stimulation-evoked capillary dilations occurred independent of astrocyte Ca²⁺ increases in the cerebral cortex.

Keywords: IP3R2 KO; astrocyte; awake mouse; calcium signaling; capillary dilation; cerebral blood flow; endfeet; processes.

Figure 1.

Measurement of vessel dilation and astrocyte endfoot Ca²⁺ responses. (a) Maximum projection of a Z stack of 2P images showing 0 through 4^th order blood vessels in the barrel cortex of the mouse. Vessel are labeled with Texas-Red (red) and astrocytes express Cyto-GCaMP6f (green). (b to d) 2P images from a time series showing dilation of a WT 4^th order capillary and Ca²⁺ increases in adjacent endfeet. The hashed lines around the capillary indicate the width of the vessel before stimulation. The time of acquisition, relative to the onset of whisker stimulation, is indicated in each panel. (b) Before stimulation. (c) Near the peak of vessel dilation. (d) Near the peak of the astrocyte Ca²⁺ increase. The boxed area illustrates the ROI used to measure the endfoot Ca²⁺ increase. (e) Traces showing capillary dilation (blue) and astrocyte Ca²⁺ increases (orange) in the trial shown in panels b to d. The traces are an average of 9 no movement trials from the vessel. The black bar indicates the time course of the whisker stimulus.

Figure 2.

Average responses of 4^th order capillaries and adjacent astrocyte endfeet to whisker stimulation. Mean ± SEM vessel dilations (blue) and astrocyte Ca²⁺ increases (orange) are shown. Black bars indicate the time course of the whisker stimulus. (a) Trials where WT mice did not run. N = 35 vessels. (b) Trials where the stimulus evoked running in WT mice. Running elicited increased astrocyte Ca²⁺ signaling compared to trials where mice did not run. N = 18 vessels. (c) Trials of IP3R2 KO mice where the animal did not run. The evoked astrocyte Ca²⁺ response is largely eliminated compared to WT trials. N = 19 vessels. (d) Trials of IP3R2 KO mice where the stimulus evoked running. N = 22. Despite the widely varying Ca²⁺ responses under the four conditions, vessel dilation remained largely unchanged.

Figure 3.

Summary of vessel dilation and astrocyte endfoot Ca²⁺ responses for all order vessels. Peak vessel dilation (blue) and astrocyte Ca²⁺ increases (orange) for WT, WT Running, IP3R2 KO and IP3R2 KO Running groups for all order vessels are shown as mean ± SEM plus individual vessel values. Astrocyte Ca²⁺ increases displayed are from Cyto-GCaMP6f recordings. Astrocyte Ca²⁺ increases in WT Running trials were significantly larger compared to no running WT trials. Compared to WT and WT Running, astrocyte Ca²⁺ increases in IP3R2 KO groups were significantly decreased. However, across all order vessels, vessel dilations remained largely unchanged despite the increased or decreased astrocyte Ca²⁺ signaling. Two-way ANOVA between genotype and movement conditions for each vessel order, Tukey-Kramer Post Hoc. Between the different orders, both 0 and 1^st order vessels (n = 63 and 70, respectively) dilated more than 2^nd, 3^rd and 4^th order vessels (n = 114, 100 and 94, respectively). One-way ANOVA, Tukey-Kramer post-hoc test, p < 0.05. * p < 0.05, ** p < 0.01, *** p < 0.001. Tables 1 to 4 contain a summary of the data displayed, including the n’s for each category.

Figure 4.

Astrocyte endfoot Ca²⁺ responses and resulting vessel dilations of 1^st through 4^th order capillaries. (a) Evoked Ca²⁺ responses in astrocyte endfeet measured with Cyto-GCaMP6f in WT (n = 129) and IP3R2 KO (n = 78) animals. Mean ± SEM of all 1^st through 4^th order capillary trials. (b) Evoked Ca²⁺ responses in astrocyte endfeet measured with Lck-GCaMP6f in WT (n = 96) and IP3R2 KO (n = 110) animals. Mean ± SEM of all 1^st through 4^th order capillary trials. (c) Summary of results, showing the decrease of evoked Ca²⁺ responses in IP3R2 KO animals measured with both Cyto- and Lck-GCaMP6f Ca²⁺ indicators. Two sample t-test of WT and IP3R2 KO capillary endfeet peak Ca²⁺ responses, *** p < 0.001. (d) Scatterplot showing relation between the evoked astrocyte Ca²⁺ increase and the corresponding capillary dilation for individual capillaries for all 1^st through 4^th order capillaries. Lines indicate least-squares fit for trials using the Cyto (blue, n = 129) and Lck (orange, n = 96) Ca²⁺ indicators. The R² coefficient of determination values are indicated.

Figure 5.

Astrocyte process Ca²⁺ responses near 0 order penetrating arterioles and 1^st through 4^th order capillaries. (a) Evoked Ca²⁺ responses in astrocyte processes near 0 order penetrating arterioles measured with Lck-GCaMP6f in WT (n = 21) and IP3R2 KO (n = 15) animals. Mean ± SEM of all Lck-GCaMP6f 0 order trials. (b) Evoked Ca²⁺ responses in astrocyte processes near 1^st through 4^th order capillaries measured with Lck-GCaMP6f in WT (n = 95) and IP3R2 KO animals (n = 108). Mean ± SEM of all Lck-GCaMP6f 1^st through 4^th order capillary trials. (c) Summary of results, showing the decrease of evoked Ca²⁺ responses in astrocyte processes near dilating blood vessels. Two sample t-tests of WT and IP3R2 KO animal. ** p < 0.01, *** p < 0.001.

Figure 6.

Latency to the onset of vessel dilation for all order vessels from WT and IP3R2 KO mice trials without running. 1^st order capillaries had a faster dilation onset compared to 0 order penetrating arterioles and 2^nd order capillaries. One-way ANOVA between the orders of each genotype, Tukey-Kramer Post Hoc. * p < 0.05, ** p < 0.01. Mean ± SEM plus individual vessels are shown. Tables 1 and 2 contain a summary of the data displayed, including the n’s for each category.

Figure 7.

Latency to the onset of Ca²⁺ increase in astrocyte endfeet contacting vessels of different orders from WT mice trials without running. (a) Summary of latencies for each vessel order of mice expressing either Cyto-GCaMP6f (left, n = 11, 13, 30, 31, 32 for 0, 1^st, 2^nd, 3^rd, 4^th orders, respectively) or Lck-GCaMP6f (right, n = 9, 12, 19, 16, 9 for 0, 1^st, 2^nd, 3^rd, 4^th orders, respectively). Faster onset latencies were seen in Lck-GCaMP6f trials compared to Cyto-GCaMP6f. Mean ± SEM plus individual vessels are shown. Two sample t-test between onset latency from combined orders of Cyto-GCaMP6f vs Lck-GCaMP6f, p = 9.0 × 10⁻⁶. (b & c) Heat maps showing Ca²⁺ responses from astrocyte endfeet contacting individual vessels. Records are sorted by latency. Ca²⁺ signals measured with Cyto-GCaMP6f (b, n = 117) and Lck-GCaMP6f (c, n = 65) are shown. ΔF/F for each vessel was normalized to its peak ΔF/F value. Trials where baseline variation exceeding 30% of peak ΔF/F in the 2 s prior to the stimulus were excluded. Times are relative to stimulus onset (white vertical lines). (d) Cumulative plots of the onset latencies of astrocyte Ca²⁺ responses measured with Cyto-GCaMP6f (blue, n = 117) and Lck-GCaMP6f (orange, n = 65). A greater proportion of fast Ca²⁺ responses were recorded with Lck-GCaMP6f. Two sample Kolmogorov-Smirnov test, p = 1.25 × 10⁻⁴.