Activity-induced Ca2+ signaling in perisynaptic Schwann cells is mediated by P2Y1 receptors and regulates muscle fatigue

Perisynaptic glial cells respond to neural activity by increasing cytosolic levels of calcium, but the functional significance of this pathway is unclear. Terminal/persiynaptic Schwann cells (TPSCs) are a perisynaptic glial cell at the neuromuscular junction. Here, we provide genetic evidence that neural activity-induced intracellular calcium accumulation in neonatal TPSCs is mediated exclusively by P2Y1 receptors. In P2ry1 mutant mice lacking these responses, postsynaptic, rather than presynaptic, function was altered in response to nerve stimulation. This impairment was correlated with a greater susceptibility to activity-induced muscle fatigue. Interestingly, fatigue in P2ry1 mutants was exacerbated by exposure to high potassium to a greater degree than in control mice. High potassium itself increased cytosolic levels of calcium in TPSCs, a response which was also reduced P2ry1 mutants. These results suggest that activity-induced calcium responses in perisynaptic glia at the NMJ regulate postsynaptic function and muscle fatigue by influencing the levels of perisynaptic potassium.

Results

208
In order to study the Ca 2+  GCaMP3, which encodes a green fluorescent protein (GFP)-conjugated calmodulin that 218 fluoresces upon Ca 2+ binding (Zariwala et al., 2012). In whole-mounts of P7 diaphragm muscle 219 incubated with GFP antibodies to label GCaMP3, S100 antibodies to detect Schwann cells, and 220 -BTX to visualize NMJs, we observed expression of GCaMP3 in all TPSCs ( Figure 1B).  period was usually less than 10% of that at the initial peak. GCaMP3-expressing TPSCs 237 responded to multiple stimulations, providing a useful tool to measure the effects of different 238 stimuli on the same TPSCs. Similar to a previous study (Darabid et al., 2013), the peak 239 intensity of Ca 2+ transients observed after a subsequent 45s bout of 40Hz stimulation was lower 240 than that of the first (33 + 4.1 vs. 26.5 + 6.1 dB, first vs. second stim, n=5; p<0.01).

241
Interestingly, analyses of transients in TPSCs across large regions of the diaphragm from low-242 magnification videos showed substantial variability in the onset after nerve stimulation (from 243 0.8s to 4.3s). These results are not related to heterogeneity in the onset of neurotransmitter 244 release, as individual muscle fibers across the entire diaphragm exhibited shortening at nearly 245 the same onset after nerve stimulation (Supplemental Video 1).

246
Although the diaphragm is activated by short bursts of tonic stimulation of the phrenic 247 nerve during several behaviors (e.g., expulsive maneuvers such as wretching; Hodges and 248 Gandevia, 2000), the more native physiological pattern of stimulation that occurs during 249 respiration is phasic with a duty cycle between 25 and 50%, in which each period of activation is . At lower frequencies of stimulation, we observed similar differences of Ca 2+ signals in response to phasic vs. tonic patterns. Interestingly, in response to 10Hz 257 stimulation, the lowest phasic rate that produced a measurable response, Ca 2+ transients were 258 much slower in onset after nerve stimulation than after 40Hz stimulation ( Figure 2F).

259
We next examined whether lower frequencies were capable of inducing Ca 2+ transients 260 in TPSCs, similar to astrocytes (Sun et al., 2014

318
We investigated which nerve-derived, P2Y 1 R-stimulating ligands were capable of

353
We next evaluated the effect of eliminating activity-induced Ca 2+ signaling on the 354 presynaptic release of neurotransmitter, based on results obtained in previous studies 355 (Robitaille, 1998;Castonguay and Robitaille, 2001 Figure 3A).

426
Fatigue was similarly enhanced, using both optical and tension measurements (Supplemental 427 Figure 3B). However, when we examined NMJs by neurofilament immunohistochemistry to 428 detect the numbers of innervating axons at individual NMJs at several ages between P7 and 429 P15, we were unable to observe any differences (Supplemental Figure 3C,D).    (Figure 6C,D)

533
The current study represents the first evaluation of Ca 2+ responses using genetically

552
Our results fail to support the idea that TPSC Ca 2+ responses affect the presynaptic 553 release of ACh, in contrast to previous studies (Robitaille, 1998    markedly reduced in these mutants, providing indirect evidence that K + uptake was impaired.

591
Interestingly, elevation of [K + ] o to 20 mM also induced a robust increase of intracellular Ca 2+ in 592 cultured astrocytes (Duffy and MacVicar, 1994). However, in contrast to the current study, this 593 response was mediated entirely by external influx through voltage-gated Ca 2+ channels (VGCC).

594
Therefore, K + -induced Ca 2+ responses in TPSCs do not appear to result from depolarization-595 mediated ingress through VGCC, but rather by the release from intracellular stores. Together, 596 these data suggest that activity stimulates perisynaptic K + uptake in TPSCs by both feedforward 597 Ca 2+ responses initiated by neurotransmitter and feedback signals initiated by high [K + ] o itself.

598
The intracellular uptake of K + against its concentration gradient has been demonstrated 599 in Müller glia in the retina (Newman et al., 1984) and is mediated by several mechanisms, 600 including inwardly-rectifying potassium channels (Kir), Na + , K + ATPases, and Na + , K + Cl -

702
The diaphragm of Wnt1-GCaMP3 mice was illuminated with a Spectra X light engine 703 (Lumencor). In order to quantify maximal fluorescence (F max ) exhibited by GCaMP3 in Schwann 704 cells, 30μM CPA was added to deplete sarcoplasmic reticular Ca 2+ stores (Heredia et al., 2016).

705
Image sequences were captured using an Andor Neo sCMOS camera and a Windows-based 706 PC using Nikon NIS Elements 4.1. Image sequences were recorded at 25 frames per second, 707 and were exported as 8-bit TIFF files into custom-written software (Volumetry G8d). A Gauss 708 filter (3x3 pixel, sd = 1.0) was applied to reduce camera noise, and motion-correction routines  to the application of the stimulus (0.5-1.0s) extending to 60s was calculated. SD maps were 721 color coded using a "Fire" CLUT. Traces of fluorescence intensity were generated from movies and presented as changes in fluorescence with respect to initial fluorescence (ΔF/F avg(prestim) or 723 ΔF/F o ). Peak intensities of Ca 2+ transients were calculated as a signal-to-noise ratio (SNR) in 724 which peak standard deviation values were divided by the prestim standard deviation value.

725
The log 10 of this ratio was generated and multiplied by 20 to standardize the SNR as decibels

752
For experiments with altered KCl, NaCl was adjusted accordingly to maintain the same Cl -753 concentration before being perfused in.

773
Statistics: Power analyses were performed using G*power 3.010 to determine the numbers of P2ry1 wild-775 type (WT) and mutant mice required. For example, to determine the number of mice (n) and 776 cells (c) to analyze for electrophysiological recordings, a power of 0.8, significance or alpha of 777 0.05 and effect size or Pearson's r of 3.6 was used. Thus, for these experiments, data was 778 generated from c=3 per animal or more, and from n=3 or more. In this case, each c and each n 779 are biological replicates. Differences between P2ry1 mutant and WT mice were assessed by 780 unpaired Student's t-tests between two independent means, assuming equal variance. A P < 781 0.05 was considered significant .   782  783  784  785  786  787  788  789  790  791  792  793  794  795  796  797  798  799  800  801  802  803  804  805  806  807  808  809  810  811  812  813  814  815  816  817  818  819  820  821  822  823  824  825 National Institute of General Medical Sciences (8P20 GM103513-09).

839
All authors approve of the final version and agree to be held accountable for all aspects of the 840 work.

844
The authors declare no competing interests.

846
Correspondence and requests for materials should be addressed to T.W. Gould