|Home | About | Journals | Submit | Contact Us | Français|
Amyloid beta (Aβ) was shown to bind the 75 kD neurotrophin receptor (p75NTR) to induce neuronal death. We synthesized a p75NTR antagonistic peptide (CATDIKGAEC) that contains the KGA motif that is present in the toxic part of Aβ and closely resembles the binding site of NGF for p75NTR. In vivo injections of Aβ into the cerebral cortex of B57BL/6 mice together with the peptide produced significantly less inflammation than simultaneous injections of Aβ and a control (CKETIADGAC, scrambled) peptide injected into the contralateral cortex. These data suggest that blocking the binding of Aβ to p75NTR may reduce neuronal loss in Alzheimer’s disease.
Neuronal death in the presence of beta amyloid (Aβ) characterizes Alzheimer’s disease (AD). We have demonstrated (Yaar et al. 1997, 2002) and others have confirmed (Kuner et al. 1998; Perini et al. 2002; Tsukamoto et al. 2003) that Aβ induces apoptosis/necrosis of neurons by binding the 75 kD transmembrane neurotrophin receptor (p75NTR), activating nuclear NFκB (Kuner et al. 1998) as well as capsases 8, 9, and 3 (Perini et al. 2002). Our previous in vitro work has demonstrated that a p75NTR antagonistic cyclic peptide (CATDIKGAEC) interferes with Aβ signaling and rescues neurons from Aβ-induced toxicity (Yaar et al. 2007).
p75NTR interacts with receptors of the trk family, increasing receptor affinity for neurotrophins and leading to several beneficial effects on neurons including survival [reviewed in (Ibanez 2002). However, certain ligands like Aβ bind and activate p75NTR alone, without coordinate binding to receptors of the trk family, and induce cell death (Yaar et al. 1997, 2002).
Ibanez et al. (1992) showed that amino acids 29–35 (TDIKGKE) of NGF, that form one of its β hairpin loop structures, mediate p75NTR binding, and if K (lysine) in position 34 is replaced by A (alanine), the resulting NGF mutant still binds p75NTR but with 50% lower affinity (Ibanez et al. 1992). Interestingly, amino acid residues 28–30 of Aβ are KGA, and computerized structure analysis of Aβ suggests that these residues have a high probability of being in a β-loop configuration (White et al. 1994), constituting a receptor binding site. Indeed, a cyclic peptide containing the KGA motif that specifically binds p75NTR but not p140trkA, prevents Aβ-induced apoptotic signaling and prevents Aβ-induced neuronal death in vitro (Yaar et al. 2007).
We now report that in vivo injections of Aβ together with this p75NTR antagonistic cyclic peptide into the cerebral cortex of mice produced significantly less inflammation than simultaneous injections of Aβ and a control peptide. Blocking Aβ binding to p75NTR may provide a novel approach to arresting the progression of AD.
The cyclic peptides were purchased from BioSource International. Aβ (1–42) (Quality Controlled Biochemicals, Hopkinton, MA) was dissolved in sterile distilled water (0.2 mM) and incubated at 37°C for 1 day before use to allow peptide aggregation. About 14–16-week-old male C57BL/6 wild type (WT) or p75NTR-null (p75NTR−/−) mice (Charles River Laboratories, Wilmington, MA) were used. Animal experiments were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by both the Veteran’s Affairs and Boston University Animal Care committees.
Two experiments were performed. In the first experiment mice received stereotactic injections of Aβ (1–42, 0.2 mM) and diluent, or Aβ (0.2 mM) plus antagonistic cyclic peptide (0.25 μM, 1 μM, or 10 μM) into the cerebral cortexes as described (Kowall et al. 1992). Animals were sacrificed 7 days later and brains were processed for analysis. The entire brain was sectioned (each section 35 μm thick) and at least 18 sections including a negative (lesion-free) section at each end were collected and stained with cresyl violet (CV) to identify inflammatory cells (Kowall et al. 1992). In the second experiment, mice (n = 15) were injected with Aβ and p75NTR antagonistic cyclic peptide (CATDIKGAEC) to one hemisphere as above, or Aβ and a scrambled peptide (CKETIADGAC) as control to the contralateral hemisphere. Brain sections were processed as above, the extent of inflammatory infiltrate was identified manually in each section and the total area occupied by inflammatory cells was determined [Macintosh scanner IPlab spectrum software (Signal Analytics, Vienna, VA)].
Visual inspection of CV-stained sections showed that the inflammatory infiltrate surrounding brain areas injected with Aβ alone was consistently larger than that produced in contralateral sites injected with Aβ and the cyclic peptide (Fig. 1). Brain lesions in hemispheres injected with the antagonistic cyclic peptide in addition to Aβ were significantly smaller than those produced by injections of Aβ and the scrambled peptide (P < 0.025, paired t-test).
To further confirm the role of p75NTR in Aβ-induced inflammation, wild type (WT, n = 18) or p75NTR-null (p75NTR−/−, n = 22) C57BL/6 mice were injected with Aβ (0.2 mM) to one cortical hemisphere and diluent to the contralateral hemisphere, and total lesion area was determined as above. In WT mice brain lesions in Aβ injected hemispheres were 75 ± 24% (mean ± SEM) larger than in diluent injected hemispheres (P < 0.02, paired t-test). However, in p75NTR−/−mice, brain lesions in Aβ injected hemispheres were only 27 ± 21% (mean ± SEM) larger than in diluent injected hemispheres (P = 0.56, paired t-test). Thus, in mice lacking p75NTR there was no significant difference in inflammation between Aβ and diluent alone.
Neuroinflammation in Alzheimer’s disease contributes to the brain’s response to Aβ deposits and to the ensuing neuronal apoptosis and necrosis (reviewed in (Stege and Bosman 1999; Tan et al. 1998; Yaar et al. 1997)). Activated microglia and astrocytes are recruited to the area and assemble within and around Aβ plaques (El Khoury et al. 1996). Together with T lymphocytes, these cells are activated to produce inflammatory mediators including cytokines IL-1β, IL-6, TNF-α, and monocyte chemoattractant protein 1 (Stege and Bosman 1999). Additionally, these inflammatory cells produce reactive oxygen species (ROS) (Tan et al. 1998) and nitric oxide (Goodwin et al. 1995), ultimately increasing neuronal injury and inflammation. Interestingly, inflammatory cell cytokines like IL-1 enhance Aβ synthesis (Goldgaber et al. 1989) and are then activated by newly synthesized Aβ to produce more inflammatory mediators, generating a positive feedback cycle that further aggravates neuronal injury (Klegeris et al. 1994). Although innate immune responses to injurious stimuli is generally beneficial, a prolonged response such as that present in AD is detrimental, adversely affecting vulnerable neurons and leading to further neuronal degeneration (Craft et al. 2005).
Injections of Aβ into murine cortex was reported to produce areas positive for glial fibrillary acidic protein immunoreactivity, increased astrocyte density, and increased level of markers for oxidative damage (Klein et al. 1999). In aged primates, Aβ intracerebral injections produced lesions containing argyrophilic, thioflavine S fluorescent, Alz 50, and ubiquitin immunoreactive neurons (McKee et al. 1998), demonstrating Aβ neurotoxicity, and possibly identifying the initial damage caused by Aβ deposition in the brain. Here we report that a p75NTR antagonistic cyclic peptide, known to block Aβ-induced neuronal death in vitro(Yaar et al. 2007), decreases inflammation in cerebral cortices of mice injected with Aβ together with the antagonistic cyclic peptide as compared to those injected with Aβ and a control scrambled peptide.
Senile plaques in AD are characterized by extra-cellular accumulation of Aβ containing primarily Aβ42 (Knauer et al. 1992). Although in humans these plaque evolve over many years, when Aβ42 is injected into murine cortex it rapidly leads to AD-like changes, primarily recruitment of activated glial cells and increased level of ROS (determined by immunostaining for 8-hydroydeoxyguanosine) (Klein et al. 1999; Kowall et al. 1991), leading to neuronal degeneration (Kowall et al. 1991). Using this murine model, we show a beneficial anti-inflammatory effect of a p75NTR antagonistic cyclic peptide, not observed with an irrelevant scrambled cyclic peptide or in mice lacking p75NTR. The studies, however, do not provide information on the cyclic peptide ability to prevent neuronal death in vivo or on its long-term effects in preventing the reported pathological, biochemical, and behavioral decrements associated with Alzheimer’s disease. However, in combination with the prior in vitro observation that the cyclic peptide prevents Aβ-mediated neuronal apoptosis/necrosis (Yaar et al. 2007) the data suggest that the recruitment of microglia and other inflammatory cells that would have otherwise been activated in response to neuronal injury (Stege and Bosman 1999; Tan et al. 1998; Yaar et al. 1997). Indeed, degenerating neurons are known to release pro-inflammatory mediators such as heat shock proteins and interleukins (IL-1 and -8) to recruit and activate inflammatory cells (Gupta et al. 2006). Alternately, or in addition, microglia and astrocytes are known to express p75NTR (Heese et al. 1998; Perini et al. 2002) and to be activated by Aβ to produce a variety of inflammatory mediators (El Khoury et al. 1996; Goodwin et al. 1995; Klegeris et al. 1994). Some of these mediators like TNF-α and IL-1β synergize with Aβ to enhance neuronal degeneration (Perini et al. 2002). Thus, by interfering with Aβ binding to p75NTR in glial cells, p75NTR antagonistic cyclic peptide may prevent their activation and thus decrease the inflammatory response.
Regardless of mechanism, our studies suggest that inhibiting the binding of Aβ to p75NTR may have the beneficial effect of slowing the progression of AD.
This work was supported by pilot grant from the Massachusetts Alzheimer’s Disease Research Center and from Boston University Community Technology Fund (MY) and the Department of Veteran’s Affairs, Department of Defense (98228055), and NIH (AG13846, AG12922) (NWK).
Conflict of Interest Statement Aspects of the work reported in this correspondence pertain to patents for which M.Y. and B.A.G. are co-inventors. These patents are assigned to the Trustees of Boston University (their employer) and licensed to SemaCo, Inc., a for-profit company in which M.Y. and B.A.G. hold equity, created to commercialize intellectual property arising out of their laboratory.
Mina Yaar, Department of Dermatology, Boston University School of Medicine, 715 Albany Street, Boston, MA 02118, USA. Department of Dermatology, Boston University School of Medicine, 715 Albany Street, Boston, MA 02118, USA.
Bennet L. Arble, Department of Dermatology, Boston University School of Medicine, 715 Albany Street, Boston, MA 02118, USA.
Kenneth B. Stewart, Department of Dermatology, Boston University School of Medicine, 715 Albany Street, Boston, MA 02118, USA.
Nazer H. Qureshi, Department of Veterans Affairs, Geriatric Research Education and Clinical Center, VA Medical Center, 200 Springs Road, Bedford, MA 01730, USA.
Neil W. Kowall, Department of Dermatology, Boston University School of Medicine, 715 Albany Street, Boston, MA 02118, USA. Department of Veterans Affairs, Geriatric Research Education and Clinical Center, VA Medical Center, 200 Springs Road, Bedford, MA 01730, USA. Department of Neurology, Boston University School of Medicine, 715 Albany Street, Boston, MA 02118, USA.
Barbara A. Gilchrest, Department of Dermatology, Boston University School of Medicine, 715 Albany Street, Boston, MA 02118, USA. Department of Pathology, Boston University School of Medicine, 609 Albany Street, J-Building, Boston, MA 02118-2394, USA.