Autopsy material was obtained from five patients with diagnoses of combined AD/LBD. In addition, one patient had a clinical diagnosis of PD. There were 2 males and 3 females, with an age range of 68–85 y. The patients whose tissue was examined postmortem, with appropriate consent, were participants in the University of Arkansas for Medical Sciences National Institute on Aging-sponsored Alzheimer's Disease Center under University Institutional Review Board guidelines.

Secreted APP was purified from a recombinant expression system as described previously 15. Mouse IL-1ra was from R&D System (Minneapolis MN). Medium, serum, and B27 supplement for cell cultures were from Invitrogen (Grand Island NY).

Primary neuronal cultures were derived from the cerebral cortex of fetal Spraque-Dawley rats (embryonic day 18), as described previously 19. Experiments using primary neuronal cells were performed after 8–12 days in culture. Primary cultures of rat microglia were generated from the cortical tissue of neonatal (0–3 days) Sprague-Dawley rats, as described previously 15. When mixed cultures were used, primary microglia were added directly to cultures of primary neurons and exposed for 24 h to 0.0 to 10 ng/mL IL-1β. To test the potential role of factors derived from activated microglia in driving neuronal expression of substrates of AD- and PD-associated neuropathologies, cultures of primary neurons were incubated for 24 h with medium from unactivated primary microglia, or with medium derived from sAPP-activated (30 nM) microglia. To determine if such changes in expression were mediated by IL-1, some of the neuron cultures were treated with IL-1 receptor antagonist (IL-1ra, 50 ng/mL) before incubation with conditioned medium from activated microglia.

Pellets impregnated with IL-1β (100 ng of recombinant mouse IL-1β) or control pellets (containing 100 ng of acetone-extracted bovine serum albumin, BSA) were implanted 2.8 mm caudal to bregma;4.5 mm right of the midline, and 2.5 mm deep to the pial surface 12. Twenty-one male Sprague-Dawley rats weighing 264 ± 6 g were randomly divided into three groups. Eight rats received implants of IL-1-containing pellets, seven rats received pellets with BSA impregnation, and six rats served as un-operated controls. Twenty-one days after implantation, cortex from the hemisphere contralateral to the implant was collected for RNA isolation.

Total RNA was extracted from rat brain tissue with Tri-Reagent (Molecular Research, Cincinnati, OH). RT-PCR was performed as previously described 12. The level of α-synuclein PCR product in each sample was normalized to that of GAPDH in the same sample.

Western immunoblot analyses were performed as described previously 18. Briefly, equal amounts of total protein (determined by bicinchonic acid assay) were resolved on 8–15% SDS/polyacrylamide gels and transferred to nitrocellulose blots. Blots were probed overnight at 4°C with either a rabbit anti-α-synuclein polyclonal antibody, diluted 1:1000 (Chemicon, Temecula CA); or a rabbit anti-βAPP polyclonal antibody, diluted 0.5 μg/mL (Stressgen, San Diego, CA; Vector Lab, Burlingame, CA); or monoclonal phosphorylated-Tau antibody, diluted 1:1000 (AT8, Pierce, Rockford, IL); or MAPK-p38, diluted 1:1000 (BioLabs, Inc., Beverly, MA); or monoclonal synaptophysin antibody, diluted 1:2000 (Roche, Indianapolis, IN); or IL-1β, diluted 0.1 μg/mL (Chemicon, Temecula, CA) After washing, blots were reacted with HRP-conjugated goat anti-rabbit antibodies, for polyclonal antibodies or anti-mouse for monoclonal antibodies. Detection of antibody-antigen binding was performed with Western-Light™ Chemiluminescent Detection System (Applied Biosystems, Foster City, CA). For Western immunoblot analysis of culture medium to detect either sAPP or IL-1β, media samples were concentrated via filtration through Centricon YM-10 centrifugal concentrators (Millipore, Bedford, MA) before analysis as above.

Analysis was performed on 6-micron thick tissue sections from formalin-fixed, paraffin-embedded blocks of parahippocampal gyri from brains of patients with AD/LBD. Tissue sections were deparaffinized by passing through a series of xylenes, and rehydrated through a graded series of alcohols, to deionized water. For the triple-labeled immunoreactions in the present study, the four antibodies were combined as follows: AT8 and α-synuclein with either IL-1α or βAPP. All incubations were followed by washing in PBS and, unless stated otherwise, were performed at room temperatures. For immunoreactions that included IL1α, AT8, and α-synuclein, sections were pretreated with an antigen retrieval enzyme digestion system (Digest-All 2, Zymed, South San Francisco, CA), immunoreacted at overnight with polyclonal IL-1α antibody (Peprotech, Rocky Hill, NJ), diluted 1:10 in 2.5% normal horse serum in PBS, immunoreacted with an ImmPRESS Reagent anti-rabbit Ig-peroxidase-micropolymerized reporter enzyme staining system (Vector), followed by detection with DAB (Zymed) and a 30-min treatment with Double Stain Enhancer (Zymed). Then the sections were pretreated for antigen epitope retrieval by microwaving in boiling 1 mM EDTA buffer (pH 8) for 10 min. Next, the sections were immunoreacted with AT8 antibody overnight, followed by a 30-min incubation with biotinylated goat-anti mouse antibody, diluted 1:200 in 2% normal goat serum in PBS, and then a 30-min treatment with avidin D-conjugated β-galactosidase (Vector), diluted 1:200 in 2% normal goat serum in PBS, and finally, detection using X-Gal Substrate Set (KPL). This antigen detection was followed by α-synuclein antigen epitope retrieval using two pretreatments: first, a 10-min microwave pretreatment in boiling 1 mM EDTA buffer (pH 8), and second, a 2-min incubation in concentrated formic acid, followed by immunoreaction with a monoclonal anti-α-synuclein antibody (Vector), diluted 1:30, for 2.5 h. This was followed by incubation with polymer-alkaline-phosphatase-conjugated secondary antibody (DAKO, Envision Kit), and labeled with permanent red (DAKO). When the three antigens to be detected were AT8, βAPP, and α-synuclein, monoclonal anti-βAPP antibody (Pierce) was immunoreacted in the place of polyclonal anti-IL-1α antibody in the protocol above as follows: sections were immunoreacted 30 min with a monoclonal AT8 antibody (Pierce), diluted 1:1000 in 2.5% normal horse serum in PBS, and immunoreacted with an ImmPRESS Reagent anti-mouse Ig-peroxidase-micropolymerized reporter enzyme staining system (Vector), followed by detection with DAB (Zymed), and then by a 30-min treatment with Double Stain Enhancer (Zymed). This was followed by α-synuclein antibody immunoreaction and detection by the β-galactosidase-avidin D/X-Gal system. Finally, anti-βAPP was applied and detected with the alkaline-phosphatase-conjugated secondary antibody and permanent red.

The unpaired, two-tailed t-test was used to determine significance using the StatView 4.1 statistical analysis program.

To test specific degenerative feedback relationships between neurons and microglia, primary neuronal and microglial culture models were employed. IL-1β treatment of purified primary neurons, cultured alone, produced dose-dependent increases in sAPP release over an IL-1 concentration range of 0.1 and 1.0 ng/mL. These IL-1β effects were enhanced when the neurons were cultured together with microglia (Fig. 1). For instance, in such neuronal-microglial cultures, neuronal sAPP induction required as little as one third the amount of IL-1β required for a similar elevation of sAPP in the absence of microglia.

One possible explanation for the enhanced overexpression of sAPP in mixed cultures, relative to that seen in neurons cultured alone, would be a positive feedback mechanism involving (i) IL-1β-induced sAPP release from neurons, (ii) sAPP-induced activation of microglia with induction of further IL-1β, and (iii) further induction of neuronal sAPP release by the increased levels of IL-1β. In support of this possibility, 24-h exposures of purified microglial cultures to recombinant sAPPα resulted in dose-dependent elevations of IL-1β expression (Fig. 2).

To test the consequences of neuron-microglia interactions as they may apply to AD- and PD-related aspects of neurodegeneration, we assessed the role of IL-1 in the effects of conditioned medium from sAPP-activated microglia on neuronal cultures. Such exposure to sAPP resulted in an increase in the levels of α-synuclein that was mediated by IL-1, as demonstrated by suppression of α-synuclein expression by pretreatment of the neuron cultures with the natural IL-1 receptor antagonist IL-1ra (Fig. 3). α-Synuclein is a necessary component of the Lewy bodies that characteristize neuropathological features of both PD and dementia with Lewy bodies, and that often co-exist with the characteristic neuropathological features of AD. In addition to increased levels of α-synuclein, we found altered expression of other markers of neuropathological changes that have been found in AD, PD and AD/LBD. We found increases in the expression of i) βAPP – the precursor of Aβ and sAPP; ii) activated (phosphorylated) MAPK-p38, a tau kinase; and iii) phosphorylated tau, of the primary component of paired helical filaments and neurofibrillary tangles. Conversely, and coincident with these increases, we found iv) decreased levels of synaptophysin, a marker of synaptic integrity (Fig. 3). To test the role of IL-1 in these events, parallel cultures of neurons were pretreated for 1 h with IL-1ra (50 ng/mL) before application of the conditioned medium from sAPP-activated microglia. This pretreatment suppressed the increases in α-synuclein, βAPP, phosphorylated-tau, and activated MAPK-p38.

To more carefully characterize the effect of IL-1β on α-synuclein expression, dependencies on dose and time were assessed in cultures of neurons. As little as 3.0 ng/mL IL-1β was an adequate dose to effect a substantial elevation of α-synuclein levels in neurons (Fig. 4A, B). Such elevations were apparent within 6 h of treatment (Fig. 4C, D).

We next sought to extend our in vitro findings regarding IL-1-induced elevation of α-synuclein and of other proteins associated with neurodegenerative events in AD to an in vivo model. Pellets that slowly release IL-1β over a period of 21 days were implanted into the cerebral cortex of Sprague-Dawley rats; "sham" rats received pellets containing BSA 12. IL-1β administration resulted in significant elevations in α-synuclein mRNA levels relative to either the levels observed in BSA "sham" animals (p < 0.02) or unoperated controls (p = 0.0003) (Fig. 5). Other markers of neurodegeneration elevated by IL-1β in vitro also showed elevated expression in cortices of rats bearing IL-1β pellets (Fig. 5).

To determine whether the IL-1-mediated events observed in vitro and in brains of experimental animals are paralleled and mutually associated in AD/LBD, we examined histological sections from autopsy material. Figure 6 illustrates triple-label immunohistochemistry confirming colocalization of neurofibrillary tangles (anti-AT8) with Lewy body (anti-α-synuclein) staining (Fig. 6A), and the tell-tale accompaniment of microglial cell activation with over-expression of IL-1 (Fig. 6B).