Page 36 - Read Online

Page 32 Benusa et al. Neuroimmunol Neuroinflammation 2020;7:23-39 I http://dx.doi.org/10.20517/2347-8659.2019.28

neurons in sham injured micro pigs [Figure 1C] [172] . These PC microglia also appeared activated without
falling into the morphological categories of phagocytic or rod microglia [172] . Another recent study found
that MPC onto cell bodies of injured neurons is associated with protection. Specifically inhibition of this
[37]
MPC increased ischemia-induced lesion volume, behavioral morbidity, and calcium influx . Additionally,
ischemic injury results in microglial process contacts with injured synapses that are nearly 10 times longer
in duration than the 4-5-min-long contacts observed in non-injured animals using a thinned-skull live
[9]
imaging approach . Therefore, it is likely that MPC could involve an increase in both the number of
microglial processes as well as the duration of these contacts onto injured axons.

The mechanisms involved in regulating MPC onto neuronal and axonal segments has primarily been
studied in mouse models of epilepsy. The number of microglial process contacts appear to be directly
related to the level of neuronal activity, in that MPC was significantly reduced upon reduction in neuronal
activity via either temperature reduction or tetrodotoxin administration in thinned-skull live-imaging
[9]
studies . Induction of neuronal hyperexcitability to the point of excitotoxicity also promoted MPC [173] .
Hyperexcitability-induced MPC resulted in reduced neuronal activity and overall increased neuronal survival
in the face of otherwise excitotoxic events that were not seen following microglia elimination or inhibition of
2+
MPC [173,174] . Neuronal excitation precipitates higher extracellular and lower intracellular Ca concentrations
and increased extracellular ATP concentrations around the active neuron, which appear to be primary
molecular mediators of hyperexcitability-induced MPC [173-176] . ATP-mediated MPC was found to promote
polarization of microglial process outgrowth toward the location with high ATP levels [Figure 1C] [177] .
Elimination of the purinergic receptor P2Y12 or the fractalkine receptor CX3CR1 drastically reduced MPC
onto hyper-excitable neurons, indicating that microglial P2Y12 and CX3CR1 are required for ATP-mediated
MPC [176,178] . Excitatory neurons also release glutamate upon excitation. Concentration of extracellular
glutamate has also been found to mediate hyperexcitability-induced MPC potentially via activation of
N-methyl-D-aspartate (NMDA) receptors [174,178] . Glutamate/NMDA-mediated MPC was also found to require
microglial P2Y12, but not CX3CR1 . Glutamate-mediated MPC also promoted nonpolarized outgrowth
[177]
of microglial processes, indicating that different molecular mechanisms of hyperexcitability-induced MPC
may result in different forms of MPC [Figure 1C] [177] . Further, these mechanisms appear distinct from
those involved in microglial phagocytosis, as knocking out or inhibiting P2Y12 or NMDA inhibited
MPC without affecting phagocytosis [178] . While epilepsy and TBI are different CNS diseases with distinct
neuropathologies, the molecules and mechanisms discussed above are prime candidates for regulation of
TBI-induced MPC. In fact, the Jacobs group found that both axotomized and intact neurons demonstrate
hyperexcitability one day following TBI in mice that appears to resolve in the axotomized population, but
not the intact neurons, by two days post-injury [179,180] . While TBI-induced MPC onto the proximal axonal
segments of axotomized neurons has yet to be thoroughly investigated, these findings indicate the potential
that similar mechanisms might be at play in TBI and epilepsy-induced MPC.

It appears that TBI-induced MPC may be species dependent, as it was found that rats sustaining the same
[50]
central fluid percussion injury paradigm as their pig counterparts did not demonstrate MPC . Rather, at
the same time points following injury, there was a significant decrease in microglial contacts onto injured
proximal axonal segments in the rats, indicating microglial processes that diverged from injured axons
[50]
or microglia process divergence (MPD) . This MPD observed in rats is in alignment with previous
observations in injured rats and mice that activated microglia do not physically associate with proximal
segments of injured axons following brain injury [116,163] . TBI-associated MPD was also observed by a group
assessing the occurrence of microglia associations with the AIS, regardless of axonal injury, following
TBI in mice . They demonstrated that microglial contacts onto the AIS of axons significantly decreased
[45]
following TBI in mice, indicating MPD similar to that observed by the other groups following TBI in
rodents [45,50,116,163] . In contrast to those studies, however, these AIS-associating, or “AXIS”, microglia were not
specific to injured axonal segments and appeared ramified (morphologically not activated), indicating that