Abstract 3707: Minocycline Treatment Inhibits Spontaneous Hemorrhage in a Mouse Model of Cerebral Amyloid Angiopathy

Abstract

Listen

Translate article

Background: Cerebral amyloid angiopathy (CAA), the accumulation of amyloid around arterioles, is a common cause of recurrent intracerebral hemorrhage (ICH) in the elderly. The precise mechanism leading to ICH is unclear, but several mouse models which mimic aspects of CAA have been created. These mice develop amyloid plaques, CAA, and microhemorrhages with advanced age. We have utilized one such mouse model (Tg2576 mice) to determine if inhibiting inflammatory responses with minocycline may impact spontaneous hemorrhage rates. Methods: Groups of aged-matched tg2576 mice, ranging from 15-20 months, were treated with saline (n=8) or minocycline 50 mg/kg (n=9), ip every other day for 2 months. After treatment, mice were sacrificed, brains were removed, sectioned and stained for amyloid (congo red and Abeta immunostaining), microhemorrhage (Perl’s Blue), astrogliosis (GFAP immunostaining), or microgliosis (CD11b immunostaining). CAA load and amyloid plaque load were measured based on area fraction of 4 regularly spaced brain sections from each mouse. Hemorrhages were counted and graded based on size (grade 1, 1-3 Perl’s blue puncta; grade 2, 4-10 puncta; grade 3, >10 puncta) in 30 regularly spaced brain sections from each mouse. Groups were compared using Student’s t-test or Mann-Whitney Rank Sum Test. Results: Mice tolerated minocycline treatment without adverse effects: average weights before, during, and after treatment did not differ between groups. Mice treated with minocylcine had significantly fewer hemorrhages than saline-treated mice (0.26 vs 0.43 hemorrhages/section, p=0.02). Grade 2 hemorrhages were significantly fewer in minocycline-treated mice (0.06 vs. 0.17 hemorrhages/section, p=0.005), but grade 1 and 3 hemorrhages were not statistically different between groups. Amyloid plaque load in the cortex (14 vs 13% area) and hippocampus (11 vs 8% area) did not differ between the two treatment groups. Total brain CAA load (0.12 vs 0.14% area) was also unchanged. Immunohistochemical analysis revealed far fewer activated microglia (CD11b immunostaining) in the minocycline-treated group compared to the saline-treated group, especially around amyloid deposits (both plaques and vascular amyloid); astrocytosis (GFAP immunostaining) was not altered. Conclusions: These data indicate that minocycline treatment reduced spontaneous hemorrhage rates in Tg2576 mice, a murine model of CAA and AD, without affecting amyloid load. Thus, it appears to inhibit downstream mechanisms involved in CAA-induced hemorrhage by decreasing microglial activation. Because minocycline is FDA-approved for a variety of clinical uses, its safety has been well-documented. A clinical trial using minocycline in CAA patients with ICH should be considered.