Response by Chang and Sansing to Letter Regarding Article, “Bexarotene Enhances Macrophage Erythrophagocytosis and Hematoma Clearance in Experimental Intracerebral Hemorrhage”

In Response

We thank Drs. Shao and Gao for their comments and interest in our work and welcome the opportunity to respond. This work was a continuation of our previous study showing that efferocytosis is involved in hematoma resolution and recovery through promoting a reparative macrophage activation state¹. To our knowledge, bexarotene is the only FDA-approved drug that upregulates Axl expression, a key receptor involved in efferocytosis, and thus is an important candidate to test in ICH. In this study, we found that bexarotene increases macrophage Axl and Cd36 expression as well as erythrophagocytosis and decreases Tnf expression in vitro. We also found that bexarotene promotes hematoma clearance and neurobehavioral recovery in vivo². We fully agree that the M1-M2 dichotomy is inadequate in describing the functional heterogeneity and phenotypic spectrum of macrophages in vivo and we do not use this nomenclature in the work. In fact, our previous work has demonstrated that murine blood-derived macrophages do not upregulate most traditional M2 markers (as defined by stimulation with IL-4, IL-10, immune complexes, glucocorticoids, and/or TGF-β) while they contribute to neurological recovery after ICH¹. Rather, we conceptualize functions of macrophages in the context of the specific inflamed tissue, in this case the brain after ICH. In this proof-of-concept study, we reasoned that instead of extensively measuring protein markers based on nomenclature derived from in vitro stimulation paradigms, we would assess a key function of macrophages in aiding recovery and restoring tissue homeostasis after bexarotene treatment.

Drs. Shao and Gao note the value of utilizing time-related markers to characterize the macrophage response. We agree entirely: our previous work demonstrated the temporal changes in (decreasing) Tnf and (increasing) Axl expression in vivo after ICH¹, which led us to focus on these key effector molecules in the present study. As they suggest, evaluating the time course and levels of Axl expression in the ICH brain after bexarotene treatment will be useful. We also agree that broadly quantifying the molecular responses of macrophages after bexarotene will be beneficial for understanding the potential of bexarotene as an ICH treatment. With its ability to activate retinoic-X-receptor (RXR), bexarotene may exert pleiotropic effects on immune cells, and the upregulation of Axl and CD36 may be indicative of broader functional changes to macrophages after treatment. Now that our work has demonstrated a beneficial effect of macrophages after bexarotene², we are very interested in learning whether bexarotene can induce a previously unexplored reparative macrophage activation state in the brain that contributes to ICH recovery and share their enthusiasm for exploring specific mechanisms.

Recent work showed that microglia and macrophages phagocytose apoptotic neurons and aid in the resolution of CNS inflammation after ischemic stroke and that these pathways are dependent on RXR-alpha. Deletion of RXR-alpha in phagocytes abolished the protective effects of microglia and macrophages in response to bexarotene after brain injury³. This finding emphasizes the essential roles of these two myeloid populations in bexarotene-facilitated brain repair. We agree with Drs. Shao and Gao that future work to identify the in vivo response of microglia in the ICH brain after bexarotene treatment will be important. We suggest that such studies should not be limited to studying microglia erythrophagocytosis, but other anti-inflammatory, pro-reparative, and neurorestorative functions of myeloid cells are also exciting areas to explore.