Cyclooxygenase 2 (COX-2) expression is elevated in prefrontal cortex neurons, not microglia, following methamphetamine self-administration in male and female rats.

Methamphetamine (METH) use produces lasting elevations in peripheral and central inflammation that correspond to cognitive and behavioral deficits, which may contribute to the likelihood of relapse. Though a great deal of research over the last few decades has attempted to investigate pharmacotherapies to help facilitate long term abstinence from METH, there are still no FDA approved medications to treat METH use disorders. Our laboratory recently showed that markers of neuroinflammation persist three weeks into abstinence following prolonged (96 h/week for 3 weeks) access to METH self-administration. We also showed that medial prefrontal cortex (mPFC) mediated behavioral flexibility deficits observed at this timepoint are attenuated by the COX-2 inhibitor parecoxib. Given the role of microglia in central inflammatory processes, the current work utilized immunohistochemistry to determine whether COX-2 expression in microglia was elevated in the mPFC during abstinence from METH. We also sought to determine if METH-induced changes in microglial morphology could potentially be altered by COX-2 inhibition. We found that the number of COX-2 expressing cells was elevated in the mPFC of rats that self-administered METH compared to those that self-administered saline. Surprisingly, COX-2 immunoreactivity was absent in microglia but was predominantly observed in neurons. Most COX-2 immunoreactivity was detected in glutamatergic neurons in both sexes, while males exhibited a reduction in COX-2 expression in GABAergic neurons. COX-2 immunoreactivity was frequently absent from the infralimbic and cingulate cortices and was therefore not analyzed. Paired with our prior findings that COX-2 inhibition attenuates METH-induced behavioral deficits known to be mediated by the mPFC, these results suggest that altered neuronal COX-2 expression should be investigated for its influence on METH-induced deficits in mPFC function.