AGE-induced neuronal cell death is enhanced in G2019S LRRK2 mutation with increased RAGE expression

HEK293 cells were culture in Dulbecco’s modified Eagle’s medium (DMEM, Gibco) containing a high glucose concentration supplemented with 10% fetal bovine serum (FBS) and penicillin/streptomycin. For the transfection, Fugene HD (Roche) was used. Primary neurons from cortex and striatum were prepared from newborn pups (postnatal day 0) [17]. Briefly, cells were dissociated by papain (Worthington Biochemical Corp) solution and then placed in poly-D-lysine (BD Bioscience) plate in Basal Eagle Medium (Sigma-Aldrich) supplemented with 1 mM L-glutamine, B27, N2, and penicillin/streptomycin (Invitrogen). To prevent glial cell growth, arabinosylcytosine (Sigma-Aldrich) was used. The medium was changed every 2 days.

To prepare AGE, BSA (50 mg/mL) was incubated with 0.5 M glucose in 0.2 M sodium phosphate buffer pH 7.4 for 10 weeks at 37 °C. The control sample of albumin was also incubated under same conditions but without glucose. All incubations were performed under sterile environments.

To determine the number of viable HEK293 cells, trypan blue dye exclusion assay was performed. It is based on the principle that live cells possess intact cell membrane that excludes dye whereas dead cells do not. A 1:1 dilution of HEK293 cell suspension was mixed with 0.4% Trypan Blue solution (Bio-Rad) at room temperature for 1-2 min. Viable cells showed clear cytoplasm, whereas nonviable cells were blue. Cell mixtures were injected into the hemocytometer chamber, and counted under the microscope. To assess the survival of primary cultured neurons, we seeded the same number of neurons in each culture and randomly captured the images for about 100 MAP2-stained neurons in vehicle and AGR-treated groups. We counted the number of survived neurons based on the appearance of normal dendritic morphology, mainly the continuous and elaborated dendritic trees.

The cDNA fragments encoding full-length human G2019S LRRK2 mutant was inserted into a tetracycline operator-regulated gene expression vector, pPrP-tetP. To facilitate protein identification, C-termini of human G2019S LRRK2 protein were tagged by hemagglutinin (HA) epitope. The F1 transgenic mice were crossed with CaMKII-tTA mice to achieve high expression of LRRK2 G2019S in the forebrain [18]. The mice were fed regular diet ad libitum and housed in a 12 h light/dark cycle. All mouse work followed the guidelines approved by the Institutional Animal Care and Use Committees of the National Institute of Child Health and Human Development.

For immunocytochemistry, cultured neurons were fixed with 4% paraformaldehyde (PFA) and permeabilized with 0.1% Triton X-100 in PBS. After blocking nonspecific staining using 10% goat serum (Sigma-Aldrich) for 1 h, cells were incubated with primary antibodies overnight. For immunohistochemistry, mice were perfused via cardiac infusion with 4% PFA in cold PBS. To prepare frozen sections, brain was removed and submerged in 30% sucrose for 24 h and then sectioned at 40 μm thickness using cryostat (Leica CM1950). Primary antibodies used included MAP2 (SantaCruz), GFAP (Sigma-Aldrich), and NeuN (Millipore). To detect RAGE protein, primary antibodies from SantaCruz and R&D were used. Anti-LRRK2 polyclonal antibodies (4EC9E and 4C84E) were kindly provided from Dr. Jean-Marc Taymans. Alexa 488 or 568-conjugated secondary antibody was incubated to visualize the staining. Fluorescent images were captured by a Zeiss confocal microscope (LSM 510 META).

Mouse brain tissue was homogenized with 10 volumes of sucrose buffer (0.32 M sucrose, 1 mM MgCl₂, 1 mM NaHCO₃, and 0.5 mM CaCl₂) containing protease and phosphatase inhibitor cocktail (Roche). To obtain crude membrane fraction, homogenized brain samples were centrifuged at 1000 × g for 10 min to remove nuclear fraction and then supernatants were centrifuged at 20,000 × g for 20 min. The pellet fraction was dissolved in RIPA buffer by sonication and concentrations were measured by BCA assay. Proteins were analyzed by 4-12% NuPage BisTris-polyacrylamide gel electrophoresis (Invitrogen) in MOPS running buffer (Invitrogene) and transferred to polyvinylidene difluoride (PVDF) membranes. The signals were visualized by enhanced chemiluminescence development (Pierce) and quantified using Scion Image System (Frederick, MD).

Striatal tissues were obtained from the brain bank of Johns Hopkins University School of Medicine. Subjects or their legal representatives signed informed consents approved by the Johns Hopkins Institutional Review Boards. The diagnosis of PD was made based on the clinicopathological criteria including characteristic clinical features and on the presence of Lewy bodies within the pigmented neurons lost in the substantia nigra. The average age at death for the PD subjects included in this study was 80 + 6.9 years old, while for the NPC subjects it was 85 + 6.6 years old.

Statistical analysis was performed with GraphPad Prism 5 (GraphPad Software). Statistical significances were obtained by comparing means of different case using t test or ANOVA followed by Tukey’s honestly significant difference test. Error bars indicate SD. ^*P < 0.05; ^**P < 0.01; ^***P < 0.001.

To test the toxic effects of AGEs generated for the cell viability assay, we first treated HEK293 cells with increasing dose of AGEs (Fig 1a). After treated with AGEs for 24 h, more HEK293 cells were lost at higher AGE dosages, whereas cells incubated with BSA showed no obvious cell death (Fig. 1a). Since LRRK2 is more abundant in the striatum [19], we cultured neurons from striatum of postnatal day 0 (P0) pups for AGE treatment. After 15 days in vitro culture (15 DIV), the cultured striatal neurons were incubated with AGEs at the concentrations of 0.0125, 0.025, 0.05, 0.1, and 0.2 μg/μl for 24 h and the cells were immuno-stained with anti-MAP2 and anti-GFAP antibodies to count the surviving neurons and astrocytes, respectively (Fig 1b). Cultured striatal neurons showed a number of neurites and healthy morphologies at the 0.0125 to 0.05 μg/μl AGE treatment. At the 0.1 μg/μl AGE-treated cases, we observed fragmented neurites but not neuronal loss. By contrast, almost all MAP2-positive neurons were dead after incubated with 0.2 and 0.4 μg/μl AGE for 24 h. On the other hand, GFAP-positive astrocytes were less sensitive to AGEs on the same condition. Based on these findings, in the later experiments, we treated cultured striatal neurons with AGE at the concentrations of 0.1~ 0.15 μg/μl for 24 h.

Next, we cultured the striatal neurons (DIV 15) from the striatum of P0 G2019S LRRK2 transgenic (Tg) and wild-type (WT) LRRK2 Tg pups to investigate whether these neurons exhibit different vulnerability to AGEs at 0.1 and 0.15 μg/μl after 24 h treatment (Fig. 2a). We found more than 60% loss of G2019S LRRK2-expressing neurons compared to the WT LRRK2 cultures after treated with 0.1 μg/μl AGEs for 24 h (Fig. 2a, b). The surviving G2019S LRRK2-expressing neurons exhibited severe neurite fragmentation and condensed cell body, while control neurons showed only slight neurite fragmentation (Fig. 2a). After treated with AGEs at 0.15 μg/μl for 24 h, more severe loss of G2019S LRRK2-expressing neurons as well as WT neurons was observed (Fig. 2a, b). These data indicate that G2019S LRRK2-expressing neurons are much more sensitive to AGE –induced toxicity compared to neurons expressing WT LRRK2.

We then investigated whether AGE-induced neuronal cell death in LRRK2 G2019S neurons is mediated by any specific intracellular signaling pathways. Receptor of AGE (RAGE) is reported to be expressed in aged brains. Therefore, we hypothesized that neuronal cell death triggered by AGE is mediated by RAGE. To test the role of RAGE, we neutralized RAGE function using anti-RAGE antiserum in LRRK2 G2019S neurons (Fig. 3a). As a result, severe cell death induced by 0.1 μg/μl AGE treatment was remarkably mitigated by pre-incubation of anti-RAGE antiserum (Fig. 3a, b). This event indicates that RAGE-related pathway participates AGE-induced neuronal cell death in LRRK2 G2019S mutation.

To test whether RAGE protein levels are regulated by G2019S LRRK2 expression, HEK293 cells were transfected with doxycycline (DOX)-activated G2019S LRRK2 expression constructs (Fig. 4a). We found enhanced RAGE protein levels in cells treated by DOX (Fig. 4a, b). Also, we observed increased levels of RAGE proteins in the G2019S LRRK2 expressing striatal neuron cultures compared to control neurons (Fig. 4c, d). Next, the striatal tissues from 3-month-old G2019S LRRK2 transgenic and littermate non-transgenic (nTG) mice were dissected and total proteins were extracted for analyses of RAGE expression. Since RAGE is a type I transmembrane protein, crude membrane proteins were also prepared to examine the RAGE protein levels. Significantly increased RAGE levels were presented in both total lysates and crude membrane extracts of the brains from G2019S LRRK2 mutant mice (Fig. 4e-g). Furthermore, immuno-staining with anti-RAGE antibody also confirmed the increase of RAGE levels in the cultured striatal neurons from G2019S LRRK2 P0 pups compared to the WT LRRK2 (Fig. 4h, i). These results provide strong evidence of upregulated RAGE protein expression in the G2019S LRRK2-expressing mouse brains.

Finally, we examined the levels of RAGE proteins in the human brain samples from sporadic PD patients and age-matched control subjects. We tested total protein lysates extracted from striatal tissues. As expected, significantly increased RAGE expression in the striatum from sporadic PD patients was detected by the western blotting (Fig. 5). This upregulation of RAGE protein levels in the striatum of PD patients supports the involvement of AGE-RAGE pathway in PD pathogenesis.