Rabbit anti-β-amyloid (22–35) antibody (anti-Aβ[22–35]) and protease inhibitor cocktail were from Sigma-Aldrich (St. Louis, MO, USA). Anti-β-amyloid 1–16 (6E10) mouse monoclonal antibody (anti-Aβ[1–16]) was from Covance (Princeton, NJ, USA). Rabbit polyclonal anti-actin antibody was from Santa Cruz Biotechnology (Santa Cruz, CA, USA). [Methyl-³H]Choline chloride (88.5 Ci/mmol) was from PerkinElmer Life Sciences (Boston, MA, USA). SH-SY5Y human neuroblastoma cells were from American Type Culture Collection (Manassus, VA, USA), and Invitrogen (Burlington, ON, Canada) supplied fetal bovine serum (FBS), Lipofectamine 2000, OptiMEM, culture media and reagents, AlexaFluor-488 donkey anti-mouse IgG and AlexaFluor-647 goat anti-rabbit IgG antibodies. Enhanced ChemiLuminescence (ECL) immunoblot reagent, Protein G Sepharose and Biodegradable Scintillant were from GE Healthcare Life Sciences (Baie d’Urfé, QC, Canada). Clarity ECL was from BioRad (Mississauga, ON, Canada). Polyclonal CHT antibody was raised in rabbits to the antigenic peptide DVDSSPEGSGTEDNLQ conserved at the C-terminus of human and rat CHT (Genemed Synthesis, San Antonio, TX, USA); this peptide was conjugated to KLH carrier protein by an N-terminal cysteine residue (Pinthong et al., 2008). CHT-specific IgG was affinity-purified in our laboratory from crude antiserum on NHS-Sepharose (GE Healthcare) to which antigenic peptide had been coupled as the binding element. Peroxidase-conjugated goat anti-rabbit IgG and peroxidase-conjugated goat anti-mouse IgG were from Jackson ImmunoResearch Laboratories (West Grove, PA, USA).

Full-length rat CHT cDNA ligated to pSPORT was a gift from Dr. T. Okuda (Okuda and Haga, 2000); a FLAG-epitope tag (DYKDDDDK) was added to the amino-terminus by PCR and the resulting cDNA ligated to pcDNA3.1. SH-SY5Y cells were transfected with FLAG-CHT plasmid by Lipofectamine 2000. Stable transformants (SY5Y-CHT cells) were selected using 500 μg/ml geneticin (G418) for 4 weeks, and then grown in complete medium (DMEM containing 10% FBS, 100 U/ml penicillin, and 100 μg/ml each of streptomycin and G418). SH-SY5Y cell differentiation was induced by addition of 10 μM RA (all-trans-retinoic acid) for 3 days, during which time cells underwent morphological and biochemical differentiation. For transient transfection, cells were treated with RA for 3 days, then immediately before transfection culture medium was changed to complete medium without antibiotics. At the time of transfection, a ratio of 1 μg plasmid DNA in 100 μl OptiMEM was added to 100 μl OptiMEM containing 2.5 μl Lipofectamine 2000, then incubated for 20 min at room temperature. This mixture was added to cell monolayers in antibiotic-free medium and incubated for 4–6 h. At the end of this incubation, culture medium was replaced with complete medium containing RA and grown for an additional 24 h.

SY5Y-CHT cells were grown to near confluency on 100 mm dishes in complete medium with 10 μM RA for 3 days. Cells were transiently transfected with 9 μg per dish of either APP_Swe plasmid DNA or the empty vector pcDNA3.1 using Lipofectamine 2000. The full-length human isoform 695 APP_Swe plasmid, generated by Dr. D. Selkoe (Young-Pearse et al., 2007), was obtained from Addgene (plasmid 30145). Following transfection, culture medium was replaced with 5 mL complete medium containing 10 μM RA per 100 mm dish and grown for an additional 24 h to condition the medium. This CM collected from vector-expressing cells (CM-vector) and APP_Swe-expressing cells (CM-APP_Swe) was cleared of cells and cellular debris by centrifugation at 300× g at 4°C for 10 min and either used immediately or stored at −80°C. Storage at −80°C does not alter the Aβ concentration in CM based on measurements using a human Aβ_1–42 ELISA or by Aβ immunoblot profile. Two separate batches each of CM-vector and CM-APP_Swe were collected from successive passages of cells (250 mL total per collection from 50 culture plates) for use in these studies. The consistency in Aβ concentration and Aβ immunoblot profile was confirmed between CM batches using Aβ_1–42 ELISA to measure Aβ_1–42 concentration and Aβ immunoprecipitation from CM to assess the amount and apparent molecular masses of the Aβ peptides recovered.

In some experiments, Aβ peptides were immunoprecipitated from CM-vector and CM-APP_Swe. CM was first pre-cleared with 15 μL/mL of washed Protein G Sepharose for 1 h at 4°C, then Protein-G Sepharose and non-specifically bound proteins were removed from CM by centrifugation at 2500× g for 5 min. Cleared CM supernatant was incubated with 5 μg/mL of either negative control anti-HA antibody, anti-Aβ[1–16] or anti-Aβ[22–35] for 1 h at 4°C. Washed Protein-G Sepharose (15 μL/mL) was then added to samples and mixed by rotation for 24 h at 4°C. Protein-G Sepharose with bound proteins were collected by centrifugation and washed three times with lysis buffer to remove non-specifically bound proteins. Proteins were eluted by incubation for 10 min at 55°C with Aβ Laemmli sample buffer (2% SDS, 40% glycerol, 200 mM Tris-HCl, pH 6.8, 0.04% bromophenol blue and 2% β-mercaptoethanol), then separated on 12% SDS-PAGE gels and transferred to polyvinylidene difluoride (PVDF) membranes. Membranes were blocked in 8% non-fat dry milk in wash buffer (phosphate-buffered saline (PBS) with 0.15% Triton X-100) for 1 h, then incubated overnight at 4°C with anti-Aβ[1–16] antibody (1:1000). After washing, membranes were incubated for 1 h in wash buffer containing 8% milk and peroxidase-conjugated goat anti-mouse IgG secondary antibody. Immunoreactive proteins on membranes were detected by chemiluminescence using a Chemidoc Imaging System (BioRad). Membranes were stripped for 20 min at 55°C followed by 5 min at room temperature in stripping buffer (62.5 mM Tris-HCl, pH 6.7, 2% SDS, 0.78% 2-mercaptoethanol), and then washed five times for 30 min in wash buffer before being re-probed with anti-Aβ[22–35] antibody (1:1000).

In experiments where Aβ peptides were neutralized in CM-vector and CM-APP_Swe, CM was incubated with 5 μg/mL of either negative control anti-HA antibody, anti-Aβ[1–16] antibody or anti-Aβ[22–35] antibody for 24 h at 4°C. This medium was then used to treat SY5Y-CHT cells that had been grown in complete medium containing 10 μM RA for 3 days for a period of 24 h at 37°C.

The amount of human Aβ_1–42 released by cells was measured in CM-vector and CM-APP_Swe at 24 h following transfection using the human Aβ_1–42 ELISA kit (Invitrogen), according to the manufacturer’s protocols. In some experiments, CM was incubated for an additional 24 h at 4°C with either anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody, then Aβ_1–42 content was measured.

Choline uptake activity was evaluated in SY5Y-CHT cells grown for 24 h in either CM-vector or CM-APP_Swe that had been pre-incubated for 24 h with either anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody. Monolayers of cells were rinsed with warm Krebs-Ringer-HEPES (KRH) buffer (mM: NaCl, 124; KCl, 5; MgSO₄, 1.3; CaCl₂, 1.5; glucose, 10; HEPES-NaOH, 20; pH 7.4), then incubated in KRH at 37°C for 15 min. This buffer was aspirated from cells and choline uptake initiated by the addition of KRH buffer containing 0.5 μM [³H]choline (0.5 Ci/mmol) either with or without 1 μM HC-3. Uptake was stopped after 5 min by placing cells on ice and washing with ice-cold KRH. Cells were solubilized in 0.1 M NaOH, then aliquots taken for tritium quantification by liquid scintillation counting and protein measurement.

SY5Y-CHT cells plated on 100 mm dishes were grown for 24 h in either CM-vector and CM-APP_Swe that was immuno-depleted by 24 h treatment with either anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody. Cells were then washed twice with cold HBSS and placed on ice under cold HBSS to stop protein trafficking. Plasma membrane proteins were biotinylated on ice by incubating with 1 mg/ml sulfo-NHS-SS-biotin in HBSS for 1 h. Unbound biotin was quenched by washing and incubating cells in cold 100 mM glycine in HBSS. After two further washes with HBSS, cells were lysed on ice for 45 min in 1% Triton X-100 lysis buffer. Neutravidin beads were incubated with cell lysates for 1 h at 4°C with gentle mixing to bind biotin-labeled proteins to allow their separation from the non-biotinylated proteins. Beads were then washed with lysis buffer three times and bound proteins were eluted by incubation for 10 min at 55°C with Laemmli sample buffer. Aliquots of biotinylated proteins and total cell lysates were separated on 10% SDS-PAGE gels and transferred to PVDF membranes. Membranes were blocked in 8% non-fat dry milk in wash buffer for 1 h, and then incubated overnight with either anti-CHT (1:3000), anti-actin (1:3000), or anti-calnexin (1:1000) antibody in wash buffer with 8% non-fat milk. After washing, membranes were incubated for 1 h with either peroxidase-conjugated goat anti-rabbit IgG (1:10,000) or peroxidase-conjugated goat anti-mouse IgG (1:10,000) secondary antibody in wash buffer containing 8% milk, and washed again. Immunoreactive proteins were detected by chemiluminescence using a Chemidoc Imaging System (BioRad). Immunopositive bands were quantified by densitometry.

Digital images of fixed cells were acquired with a Zeiss LSM510-Meta laser-scanning confocal microscope using a 63× magnification oil-immersion objective and magnified three times. Untransfected SY5Y-CHT cells plated on 35 mm dishes were grown for 24 h in CM collected from either vector-expressing or APP_Swe-expressing SY5Y-CHT cells. Immediately prior to cell treatment, CM was incubated for 24 h with either anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody. Cells were then washed twice in warm HBSS and formaldehyde-fixed. Fixed cells were incubated first with anti-CHT (1:1000) or anti-LAMP1 (1:200) primary antibodies, then incubated with secondary antibodies rabbit AlexaFluor 647 (1:1000), to label CHT, and mouse AlexaFluor 488 (1:1000), to label LAMP1. Images were then acquired using 488 nm excitation and 505–530 nm emission wavelengths for LAMP1 and 647 nm excitation and 650 nm emission using a long pass filter for CHT. Co-localization analysis was performed on confocal images using Imaris software version 7.7.0 with the Imaris Co-localization module (Bitplane) to examine the co-localization of the brightest 2% of pixels in each channel, as described previously (Lorenzen et al., 2010; Cuddy et al., 2012).

Data are presented as the mean ± SEM with n values representing the number of independent experiments performed on separate populations of cells; each n value was obtained from the average of multiple sample replicates in each experiment. Replicate experiments were performed on cells cultured in successive passages as much as possible to minimize inter-experiment variability, and intra-experiment variability between replicate samples was minimal. GraphPad Prism 5 was used for data analysis. Statistical significance was determined by paired Student’s t-test, or between groups using repeated measures one-way analysis of variance (ANOVA) with Tukey’s post hoc multiple comparison test or by two-way ANOVA, as appropriate.

The purpose of the present study was to investigate the effect of endogenously produced Aβ peptides on CHT function. To this end, we measured CHT activity and trafficking in untransfected SY5Y-CHT cells treated with Conditioned Media (CM) derived from SY5Y-CHT cells that transiently expressed either empty vector or APP_Swe (CM-vector and CM-APP_Swe, respectively). Therefore, we determined the amount and apparent molecular masses of soluble Aβ peptides present in CM-vector and CM-APP_Swe. Aβ peptides were recovered from CM-vector and CM-APP_Swe by immunoprecipitation using either anti-HA antibody as a negative control, anti-Aβ antibody directed at amino acids 1–16 at the N-terminus of Aβ (anti-Aβ[1–16] antibody) or anti-Aβ antibody directed at amino acids 22–35 in the mid-region of Aβ (anti-Aβ[22–35] antibody). Figure 1A shows a representative immunoblot that was probed for Aβ peptides using anti-Aβ[1–16] antibody (top panel), then stripped and re-probed for Aβ peptides using anti-Aβ[22–35] antibody (bottom panel). No Aβ peptides were immunoprecipitated from CM-vector or CM-APP_Swe with the anti-HA antibody, whereas strong Aβ immunoreactive bands with masses of approximately 4-kDa were recovered from CM-APP_Swe immunoprecipitated with either anti-Aβ[1–16] or anti-Aβ[22–35] antibody (Figure 1A). Faint immunoreactive Aβ bands with masses of about 4-kDa were also recovered by immunoprecipitation from CM-vector with either the anti-Aβ[1–16] or the anti-Aβ[22–35] antibody (Figure 1A). Immunoreactive Aβ bands with higher molecular masses of approximately 8-kDa and 12-kDa were recovered from CM-vector and CM-APP_Swe immunoprecipitated with the anti-Aβ[22–35] antibody coupled with probing immunoblots with anti-Aβ[22–35] (Figure 1A, lower panel). Faint immuoreactive Aβ peptide bands with masses of approximately 16-kDa were recovered from CM-APP_Swe immunoprecipitated with either anti-Aβ[1–16] or anti-Aβ[22–35] antibody coupled with immunoblots probed with anti-Aβ[1–16] antibody (Figure 1A, top panel). We next measured the concentration of Aβ_1–42 in CM-vector and CM-APP_Swe using a human Aβ_1–42 ELISA. As expected, the amount of Aβ_1–42 present in CM-APP_Swe is significantly greater than that present in CM-vector (11.7 ± 1.3 and 101.9 ± 3.9 pM Aβ_1–42 for CM-vector and CM-APP_Swe, respectively; p ≤ 0.05; Figure 1B).

It has been demonstrated previously that N-terminal Aβ antibodies can neutralize soluble synaptotoxic forms of Aβ and inhibit adverse effects of Aβ in an epitope-specific manner (Buttini et al., 2005; Shankar et al., 2008; Jin et al., 2011; Zago et al., 2012). Both anti-Aβ[1–16], an N-terminal-directed antibody, and anti-Aβ[22–35], a mid-region directed antibody, effectively recovered Aβ peptides from CM-APP_Swe by immunoprecipitation (Figure 1A). We next compared the ability of anti-Aβ[1–16] and anti-Aβ[22–35] antibodies to neutralize Aβ_1–42 peptides present in CM-APP_Swe. To determine this, CM-vector and CM-APP_Swe were incubated for 24 h with either anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody, then Aβ_1–42 concentrations in antibody-treated media were measured using a human Aβ_1–42 ELISA that recognizes the N-terminus and C-terminus of Aβ_1–42. As shown in Figure 1C, incubating CM-APP_Swe for 24 h with anti-Aβ[1–16] antibody significantly reduces Aβ_1–42 concentration in comparison to that found in CM-APP_Swe incubated for 24 h with either anti-HA or anti-Aβ[22–35] antibody (97.8 ± 2.5, 18.2 ± 2.8 and 204.5 ± 78 pM Aβ_1–42 for CM-APP_Swe incubated with anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody, respectively; *p ≤ 0.05). While the anti-Aβ[22–35] antibody bound to and could immunoprecipitate Aβ peptides from CM-APP_Swe (Figure 1A), the Aβ_1–42 concentration was not decreased by this antibody possibly because it does not bind to either the N-terminal or C-terminal amino acids of Aβ_1–42 that are detected in the ELISA. While not statistically significant, the mean Aβ_1–42 concentration was higher in CM-APP_Swe incubated with the anti-Aβ[22–35] antibody compared to CM-APP_Swe incubated with the anti-HA antibody. Aβ_1–42 levels were not measurably changed in CM-vector incubated with either anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody (12.1 ± 0.8, 8.0 ± 2.3 and 8.7 ± 1.8 pM Aβ_1–42, respectively; p ≤ 0.05; Figure 1C).

CM-vector and CM-APP_Swe used for cell treatments in this study were collected from successive passages of SY5Y-CHT cells transiently expressing either vector or APP_Swe. The data above represent CM-vector and CM-APP_Swe collected from a single passage of SY5Y-CHT cells. Similar Aβ concentrations and Aβ immunoblot profiles between CM-vector and CM-APP_Swe collected from successive cell passages were confirmed by Aβ_1–42 ELISA to measure Aβ_1–42 concentration in the medium and immunoprecipitation of Aβ to assess the amount and molecular masses of the recovered Aβ peptides.

The N-terminal directed anti-Aβ[1–16] and mid-region directed anti-Aβ[22–35] antibodies both effectively bind Aβ in CM-APP_Swe at different epitopes (Figure 1A). The purpose of the next experiment was to compare the effect of neutralizing Aβ in CM-APP_Swe with either anti-Aβ[1–16] or anti-Aβ[22–35] antibody on high-affinity choline uptake in SY5Y-CHT cells. To accomplish this, CM-vector was incubated with anti-HA antibody and CM-APP_Swe was incubated with either anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody for 24 h. SY5Y-CHT cells were then incubated with the antibody-treated CM-vector or CM-APP_Swe, and [³H]choline uptake activity was measured. Choline uptake activity is significantly decreased in SY5Y-CHT cells treated with CM-APP_Swe containing anti-HA antibody when compared to cells treated with CM-vector containing anti-HA antibody (p ≤ 0.05; Figure 2A). Interestingly, this significant decrease in high-affinity choline uptake activity was attenuated by incubation of CM-APP_Swe with anti-Aβ[1–16] antibody, but not with anti-Aβ[22–35] antibody (Figure 2A; p ≤ 0.05). High-affinity choline uptake activity in SY5Y-CHT cells treated with CM-vector did not differ statistically from that measured in cells treated with CM-APP_Swe containing anti-Aβ[1–16] antibody. Total CHT protein levels were equivalent across the treatment groups (Figure 2B).

Since the anti-Aβ[1–16] antibody neutralizes soluble Aβ peptides and prevents the decrease in high-affinity choline uptake seen in SY5Y-CHT cells treated with CM-APP_Swe, we predicted that binding of anti-Aβ[1–16] antibody to Aβ would similarly protect against Aβ-mediated loss of CHT cell surface expression. Thus, we performed cell surface protein biotinylation experiments using SY5Y-CHT cells treated for 24 h with CM-vector that had been incubated with anti-HA antibody or with CM-APP_Swe that had been incubated with either anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody for 24 h. Plasma membrane proteins were then biotinylated using membrane impermeable sulfo-NHS-biotin at 4°C. Representative immunoblots in Figure 3A show the cell surface levels of (biotinylated) CHT and calnexin proteins and the total amount of CHT, calnexin and actin proteins in cell lysates. Quantitative analysis of cell surface CHT protein immunoblots (top panel) reveals that CHT cell surface levels are significantly lower in SY5Y-CHT cells treated with CM-APP_Swe containing either anti-HA or anti-Aβ[22–35] antibody, but not in SY5Y-CHT cells treated with CM-APP_Swe containing anti-Aβ[1–16] antibody when compared to CM-vector treated SY5Y-CHT cells (Figure 3B; p ≤ 0.05). Total CHT protein levels are equal between cells treated with either CM-vector or CM-APP_Swe incubated with either anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody (middle panel). CHT cell surface levels did not differ between SY5Y-CHT cells treated with CM-vector and cells treated with CM-APP_Swe containing anti-Aβ[1–16] (Figure 3A, lanes 1 and 3, respectively). The absence of calnexin immunoreactivity in biotinylated cell surface fractions and its presence in total cell lysate fractions confirmed the isolation of cell surface proteins.

Plasma membrane CHT levels are maintained by constitutive recycling of CHT proteins between endocytic compartments and the cell surface, thereby regulating choline uptake activity. CHT proteins internalize rapidly from the plasma membrane to early endosomes by a clathrin-mediated process, thus we investigated whether Aβ present in CM-APP_Swe alters CHT localization to early endosomes. To this end, we used confocal microscopy to visualize the co-localization of CHT and early endosome maker EEA1 in SY5Y-CHT cells treated with either CM-vector containing anti-HA antibody or CM-APP_Swe containing either anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody. To assess the extent of co-localization between CHT and EEA1, we used a quantitative approach using Imaris software (Bitplane) to set threshold fluorescence intensities that filter the brightest 2% of pixels of CHT (shown in green) that also fall within the brightest 2% of pixels of EEA1 (shown in red). This analysis is described further in Hutcheon et al. (2000) and Lorenzen et al. (2010). The co-localized pixels are identified in a separate co-localization channel and shown as white in the right overlay panels of Figure 4A. As shown in Figure 4B, analysis of the quantified pixels revealed that CHT co-localizes significantly less with EEA1 in cells treated with CM-APP_Swe incubated with either anti-HA or anti-Aβ[22–35] antibody (31.3 ± 1.0% and 28.9 ± 1.0%, respectively) compared to cells treated with either CM-vector or CM-APP_Swe containing with anti-Aβ[1–16] antibody (35.6 ± 1.0% and 41.5 ± 1.1%, respectively). Unexpectedly, CHT co-localizes significantly more with EEA1 in cells treated with CM-APP_Swe containing anti-Aβ[1–16] compared to cells treated to CM-vector treated cells. The reason for this is unknown, but could be due to low physiological levels of Aβ present in CM-vector and not present in CM-APP_Swe containing anti-Aβ[1–16], thereby regulating CHT recycling between early endosomes and the cell surface.

CHT proteins internalize to early endosomes from the plasma membrane and either recycle back to the cell surface or move through late endosomes to lysosomes for degradation (Cuddy et al., 2012). Our results reveal that Aβ causes a significant decrease in CHT activity and cell surface expression that corresponds with a significant loss of CHT proteins from early endosomes. Thus, we hypothesize that the Aβ-mediated loss of CHT proteins from early endosomes, and the corresponding decrease in CHT cell surface expression and activity, could be related to an increased movement of CHT to lysosomes for degradation. We next used confocal microscopy to visualize the co-localization of CHT and lysosome marker LAMP-1 in SY5Y-CHT cells treated with either CM-vector containing anti-HA antibody or CM-APP_Swe containing either anti-HA, anti-Aβ[1–16] or anti-Aβ[22–35] antibody. To assess the extent of co-localization between CHT and LAMP-1, we used the quantitative approach described above to set threshold fluorescence intensities that filter the brightest 2% of pixels of CHT (shown in green) that also fall within the brightest 2% of pixels of LAMP-1 (shown in red). The co-localized pixels are identified in a separate co-localization channel and shown as white in the right overlay panels of Figure 5A. As shown in Figure 5B, analysis of the quantified pixels revealed that CHT co-localizes significantly less with LAMP-1 in cells treated with CM-APP_Swe incubated with either anti-HA or anti-Aβ[22–35] antibody (32 ± 1.2% and 33 ± 1.6%, respectively) compared to cells treated with either CM-vector or CM-APP_Swe containing anti-Aβ[1–16] antibody (38 ± 1.4% and 45 ± 1.6%, respectively). Consistent with our observation that CHT co-localizes more with EEA1 in cells treated with CM-APP_Swe containing anti-Aβ[1–16] compared to cells treated to CM-vector treated cells, CHT also co-localizes significantly more with LAMP-1 in cells treated with CM-APP_Swe containing anti-Aβ[1–16] compared to cells treated to CM-vector treated cells.

Aβ peptides present in CM-APP_Swe decrease the amount of CHT localizing to both early endosomes and lysosomes (Figures 4, 5, respectively). To investigate whether the Aβ-mediated loss of CHT proteins from early endosomes and lysosomes is related to increased CHT degradation by the lysosome, we blocked proteolytic activity of the lysosome pharmacologically using BafA₁ (which blocks acidification of the lysosome), and measured CHT co-localization with LAMP-1 by confocal microscopy. In this experiment, SY5Y-CHT cells were treated with either CM-vector or CM-APP_Swe containing either vehicle or BafA₁ and the extent of co-localization between CHT and LAMP-1 was assessed using the quantitative approach described above. Co-localization between CHT and LAMP-1 was observed in cells treated with either CM-vector or CM-APP_Swe containing either vehicle or BafA₁ (Figure 6A). Co-localized pixels of CHT and LAMP-1 appear white in the overlay panels and co-localized pixels panels of these images. As shown in Figure 6B, and consistent with our findings above (Figure 5), analysis of the quantified pixels revealed that CHT co-localizes significantly less with LAMP-1 in cells treated with CM-APP_Swe containing vehicle compared to cells treated with CM-vector containing vehicle (41.6 ± 1.5% and 48.1 ± 1.4%, respectively). Importantly, treatment of cells with CM-APP_Swe containing BafA₁ attenuated the effect of CM-APP_Swe, with co-localization of CHT with LAMP-1 not differing significantly between these treatment groups (47.4 ± 1.7% and 47.1 ± 1.9%, respectively).

We investigated whether Aβ inhibits CHT function by increasing lysosomal degradation of CHT proteins by blocking proteolytic activity of lysosomes using the inhibitor BafA₁, then measuring CHT cell surface expression and activity using cell surface protein biotinylation and choline uptake assays, respectively.

To determine the effect of lysosome inhibition on CHT cell surface expression, SY5Y-CHT cells were treated with either CM-vector or CM-APP_Swe containing either vehicle or BafA₁ for 24 h. Plasma membrane proteins were then biotinylated using membrane impermeable sulfo-NHS-biotin at 4°C. Representative immunoblots in Figure 7A show the level of cell surface (biotinylated) CHT and calnexin proteins and the amount of total CHT, calnexin and actin protein. Total CHT protein levels (middle panel) are equal between cells treated with either CM-vector or CM-APP_Swe containing vehicle, while total CHT protein levels are substantially increased in cells treated with CM containing BafA₁ compared to cells treated with CM containing vehicle. Quantitative analysis was carried out on immunoblots of biotinylated cell surface CHT protein levels (top panel). Statistical analysis by two-way ANOVA reveals no interaction between media treatment and drug (BafA₁) treatment of cells (P = 0.46), but there is a statistically significant difference related to BafA₁ treatment of cells (P = 0.0018) with this lysosome inhibitor attenuating the decrease in cell surface CHT protein levels observed in cells treated with CM-APP_Swe containing vehicle. In SY5Y-CHT cells treated with CM-APP_Swe containing vehicle, cell surface CHT protein levels are decreased by 28% when compared to cells treated with CM-vector containing vehicle (Figure 7B); this is comparable to the statistically significant 22% decrease in CHT cell surface levels for the same treatment groups shown in Figure 3. BafA₁ treatment attenuated the effect of CM-APP_Swe, with CHT cell surface levels being the same for cells treated with CM-vector and CM-APP_Swe. The absence of calnexin immunoreactivity in biotinlyated cell surface fractions and presence in total cell lysate fractions confirmed the isolation of cell surface proteins.

Since BafA₁ prevents the Aβ-mediated decrease in CHT cell surface expression in SY5Y-CHT cells treated with CM-APP_Swe, we predicted that BafA₁ would also block an Aβ-mediated decrease in high-affinity choline uptake activity. To test this, cells were treated with either CM-vector or CM-APP_Swe containing either vehicle or BafA₁ for 24 h, then [³H]choline uptake activity was measured (Figure 7C). Statistical analysis by two-way ANOVA revealed no interaction between media treatment and drug (BafA₁) treatment (P = 0.06), and while there was no difference related to BafA₁ treatment (P = 0.18), a significant difference was found for media treatment (P = 0.0005). In SY5Y-CHT cells treated with CM-APP_Swe containing vehicle, choline uptake activity is decreased by 37% when compared to cells treated with CM-vector containing vehicle (Figure 7C); this is comparable to the statistically significant 31% decrease in choline uptake activity for the same treatment groups shown in Figure 2. Interestingly, high-affinity choline uptake activity was not increased in SY5Y-CHT cells treated with CM-APP_Swe containing BafA₁ when compared to vehicle. Figure 7D is a representative experiment showing that total sample CHT and actin protein levels were equivalent across the treatment groups.