Accumulation of amyloid-β (Aβ) peptides in the brain is one of the central pathogenic events in Alzheimer's disease (AD). However, why and how Aβ aggregates within the brain of AD patients remains elusive. Previously, we demonstrated hemoglobin (Hb) binds to Aβ and co-localizes with the plaque and vascular amyloid deposits in post-mortem AD brains. In this study, we further characterize the interactions between Hb and Aβ in vitro and in vivo and report the following observations: 1) the binding of Hb to Aβ required iron-containing heme; 2) other heme-containing proteins, such as myoglobin and cytochrome C, also bound to Aβ; 3) hemin-induced cytotoxicity was reduced in neuroblastoma cells by low levels of Aβ; 4) Hb was detected in neurons and glial cells of post-mortem AD brains and was up-regulated in aging and APP/PS1 transgenic mice; 5) microinjection of human Hb into the dorsal hippocampi of the APP/PS1 transgenic mice induced the formation of an envelope-like structure composed of Aβ surrounding the Hb droplets. Our results reveal an enhanced endogenous expression of Hb in aging brain cells, probably serving as a compensatory mechanism against hypoxia. In addition, Aβ binds to Hb and other hemoproteins via the iron-containing heme moiety, thereby reducing Hb/heme/iron-induced cytotoxicity. As some of the brain Hb could be derived from the peripheral circulation due to a compromised blood-brain barrier frequently observed in aged and AD brains, our work also suggests the genesis of some plaques may be a consequence of sustained amyloid accretion at sites of vascular injury.

No abstract

No abstract

J. Neurochem. (2010) 112, 1138–1146. Abstract Stress and corticosterone level are thought to negatively associate with neurogenesis in mammalian brains. Social support can diminish many adverse effects of stress. The present study examined the modulating effect of social support on stress‐decreased cell proliferation and neuronal differentiation in a mouse model. A randomly‐scheduled foot shock followed by restraint in water was used as a profound stress‐provoking regimen. Bromodeoxyuridine (BrdU) staining was used to indicate newly mitotic cells and doublecortin (DCx) staining was used to reveal immature neurons. This stress‐provoking regimen rapidly decreased BrdU‐ and BrdU/DCx‐labeled cells in the dentate gyrus. However, such a stress‐provoking regimen did not affect the number of these labeled cells in the subventricular zone. Familiar and unfamiliar mice’ company throughout the stress regimen completely reversed the stress‐decreased cell proliferation and neuronal genesis in the dentate gyrus. Likewise, both odor‐familiar (from their home cages) and ‐unfamiliar (from cages other than their home cages) wooden blocks completely reversed the stress‐decreased BrdU/DCx‐labeled cells in the dentate gyrus. In contrast, wooden blocks free of any odor and camphor odor alone failed to affect the stress‐decreased BrdU‐ or BrdU/DCx‐labeled cells. Finally, we showed that conspecifics or their odors during the stress regimen reversed the stress‐decreased cell proliferation and neuronal differentiation in the dentate gyrus via a corticosterone‐independent mechanism. We conclude that stress and familiarity distinctively affect neurogenesis in the dentate gyrus and subventricular zone. Conspecific companions or presence of their odors reverse stress‐decreased neurogenesis in the dentate gyrus, suggesting that social support during stress exposure may improve neurogenesis‐related psychological functions.

Reactivation of cocaine-associated memories plays a critical role in reinstating the cocaine-seeking behavior and causing relapse. Cocaine-induced conditioned place preference (CPP) was used as a behavioral paradigm indicative of cocaine-associated memory and repeated cocaine-free preference tests served as a behavioral procedure to retrieve such a memory in this study. Since D-cycloserine was reported to eradicate drug-associated memories, two other N-methyl-D-aspartate (NMDA) receptor agonists were assessed for their efficacy on facilitating the extinction of cocaine-induced CPP. Although D-cycloserine (30 mg/kg) abolished cocaine (10 mg/kg)-induced CPP, sarcosine (300 and 600 mg/kg) and D-serine (600 mg/kg) diminished the expression of such a cocaine memory. Sarcosine (600 mg/kg) and D-serine (600 mg/kg) did not affect the storage of this cocaine memory. It was of interest to note that D-cycloserine facilitated the extinction of cocaine-induced CPP in a fast and early-onset manner, while sarcosine and D-serine decreased cocaine-induced CPP expression in a delay-onset manner. D-cycloserine (30 mg/kg), D-serine (600 mg/kg) and sarcosine (600 mg/kg) did not affect the consolidation of cocaine (5 mg/kg)-induced CPP. Finally, sarcosine (at 600 mg/kg/day for 3 consecutive days) and D-serine (at 600 mg/kg/day for 3 consecutive days) did not produce observable aversive effect associated with their administration in a conditioned place aversion paradigm. Likewise, a similar dosing regimen of sarcosine or D-serine did not cause evident activity-impairing effect. In addition to D-cycloserine treatment, our results indicate that long-term treatment with D-serine and sarcosine may afford a therapeutic advance in suppressing the expression of cocaine-associated memory.