Panel A shows fluorescence microscopy images of microglia from wild type, C3 deficient, and CD47 deficient mice stained for Iba1, ChAT, and VGlut2. Panel B shows corresponding fluorescence microscopy images stained for Iba1, ChAT, and GAD65/67. Panel C shows a bar graph of the percentage of microglia containing VGlut2 residues, with differences among wild type, C3 deficient, and CD47 deficient mice. Panel D shows a bar graph of the percentage of microglia containing GAD65/67 residues, with no significant differences between genotypes. Panel E shows a fluorescence microscopy image of a ChAT positive motor neuron region containing labeled synaptic structures, with a boxed region enlarged in panel F. Panel F shows an enlarged fluorescence microscopy image of a tagged synapse labeled for CD47, SIRP alpha, and VGlut1. Panel G shows a line intensity profile graph demonstrating overlapping signals for VGlut1, CD47, ChAT, and SIRP alpha along the indicated line. Panel H shows individual fluorescence microscopy channels for ChAT, CD47, SIRP alpha, and VGlut1 at the tagged synapse, with arrows indicating the corresponding signals. Panel I shows a bar graph comparing the total number of VGlut1 synapses with the number of VGlut1 synapses tagged by SIRP alpha and CD47.
Microglia engulfment of excitatory VGluT2+, but not inhibitory GAD65/67+ synapses, is affected by deletion of C3 or CD47, and SIRPα tags proprioceptive synapses on motor neurons. (A) Confocal images of microglia (Iba1; in red) in the vicinity of L4 motor neurons (ChAT; in blue) and VGluT2+ synapses (in white) from WT (top), C3−/− (middle), and CD47−/− (bottom) mice. Scale bar: 10 μm. (B) Single-plane confocal images showing microglia (Iba1, in red), ChAT (blue) and GAD65/67+ (in white) synapses engulfed within microglia. Scale bar: 10 μm. (C) Percentage of microglia (in the vicinity of motor neurons) containing remnants of VGluT2+ synapses. WT, N = 3 mice; C3−/−, N = 3; CD47−/−, N = 3; *P < 0.05, **P < 0.01; one-way ANOVA, post hoc Bonferroni’s test. (D) Percentage of microglia containing GAD65/67 residues. WT, N = 3 mice; C3−/−, N = 3; CD47−/−, N = 3. Each data point corresponds to the average obtained from a single mouse. Only microglia within the motor neuron area was analyzed. ns: no significance. One-way ANOVA, post hoc Bonferroni’s test. (E) Confocal image showing quadruple immunoreactivity against CD47 (red), ChAT (blue), SIRPα (green), and VGluT1+ (white) in L4 WT motor neurons. Scale bar: 10 μm. (F) Inset from E, showing at higher magnification a VGluT1+ synapse contacting a motor neuron while being tagged by SIRPα and CD47. Scale bar: 1 μm. (G) Yellow dotted line indicates the line intensity profile shown in the graph (G) for all four fluorochromes confirming tagging of VGluT1 synapse by CD47 and SIRPα (overlap of green, red, and white lines). (H1–H4) Individual images from inset are shown in F. White arrows indicate point of tagging. (I) Grey data points in blue bar are the total number of VGluT1 synapses per motor neuron. Colored data points are the average per mouse. Grey data points in orange bar are the number of VGluT1+ synapses tagged by both CD47 and SIRPa for the same motor neurons analyzed (WT mice at P5). Colored data points represent the average number of synapses per animal (N = 3). ***P < 0.001, unpaired t test.