Spinal versus brain microglial and macrophage activation traits determine the differential neuroinflammatory responses and analgesic effect of minocycline in chronic neuropathic pain

Abstract

Substantial evidence indicates involvement of microglia/macrophages in chronic neuropathic pain. However, the temporal-spatial features of microglial/macrophage activation and their pain-bound roles remain elusive. Here, we evaluated microglia/macrophages and the subtypes in the lumbar spinal cord (SC) and prefrontal cortex (PFC), and analgesic-anxiolytic effect of minocycline at different stages following spared nerve injury (SNI) in rats. While SNI enhanced the number of spinal microglia/macrophages since post-operative day (POD)3, pro-inflammatory MHCII spinal microglia/macrophages were unexpectedly less abundant in SNI rats than shams on POD21. By contrast, less abundant anti-inflammatory CD172a (SIRPα) microglia/macrophages were found in the PFC of SNI rats. Interestingly in naïve rats, microglial/macrophage expression of CD11b/c, MHCII and MHCII/CD172a ratio were higher in the SC than the cortex. Consistently, multiple immune genes involved in anti-inflammation, phagocytosis, complement activation and M2 microglial/macrophage polarization were upregulated in the spinal dorsal horn and dorsal root ganglia but downregulated in the PFC of SNI rats. Furthermore, daily intrathecal minocycline treatment starting from POD0 for two weeks alleviated mechanical allodynia most robustly before POD3 and attenuated anxiety on POD9. Although minocycline dampened spinal MHCII microglia/macrophages until POD13, it failed to do so on cortical microglia/macrophages, indicating that dampening only spinal inflammation may not be enough to alleviate centralized pain at the chronic stage. Taken together, our data provide the first evidence that basal microglial/macrophage traits underlie differential region-specific responses to SNI and minocycline treatment, and suggest that drug treatment efficiently targeting not only spinal but also brain inflammation may be more effective in treating chronic neuropathic pain.

Topography of microglial activation in sensory‐ and affect‐related brain regions in chronic pain

“…This study shows that a peripheral nerve lesion results in significant microglial activation in brain regions associated with pain and affect, including thalamus, sensory cortex, and several limbic regions. This complements previous studies that have demonstrated robust microglial activation in the spinal cord after nerve injury and suggests that pain‐induced neuroinflammation occurs widely throughout the neuroaxis (Li et al, ; Taves et al, ). This premise is supported by a previous study showing that chemotherapy‐induced pain resulted in significant microglial activation throughout many brain regions, including the periaqueductal gray (PAG), thalamus, anterior cingulate cortex, secondary sensory cortex, and medium forebrain bundle (Di Cesare Mannelli et al, ).…”

Differential Spinal and Supraspinal Activation of Glia in a Rat Model of Morphine Tolerance

“…In our investigation, the increased proportion of M2microglia in the SC could be interpreted as an effort to resolve the neuroinflammation caused by chronic morphine administration. Analogously, a rise in the proportion of similarly defined M2-microglia in the SC has been detected seven days after peripheral nerve injury in the rat (Li et al, 2016). In accordance with this hypothesis, parthenolide, known to affect intracellular microglial inflammation pathways, like p-38 and ERK1/2 (Popiolek-Barczyk et al, 2015), has promoted spinal M2 microglia/macrophage polarization and relieved pain in a rat model of neuropathy (Popiolek-Barczyk et al, 2015).…”

Microglial regional heterogeneity and its role in the brain

“…Another recent study detecting the level of 14 C in genomic DNA of microglia isolated from the human brain showed that 28% of human microglia, with an average life-span of 4.2 years, were renewed every year and most of which continued to do so indefinitely [47]. Using a multicolor fluorescent mapping [30,31,33,34]; Lower in human cerebral GM than WM [39] High [30,31]; Lower in CA3 than CA1 [21] High in CVOs [49] Average [30,31] or high [32]; Higher in SNr than VTA [65] Low [30,31,33,34]; Higher in cerebellar nuclei and granular layers [30,40]; Less CD68+ & MHCII+ in human cerebellum [38,39] Low [136,137] Morphology (ramification) High [33,34] High [33,34] Amoeboid [30,49,52] Higher in SNr than VTA [65] Low [33,34] Average; Cell smaller in dorsal horn [138] Molecular expression CX3CR1…”