Lactulose and Melibiose Inhibit α-Synuclein Aggregation and Up-Regulate Autophagy to Reduce Neuronal Vulnerability

Parkinson’s disease (PD) is a neurodegenerative disease characterized by selective dopaminergic (DAergic) neuronal degeneration in the substantia nigra (SN) and proteinaceous α-synuclein-positive Lewy bodies and Lewy neuritis. As a chemical chaperone to promote protein stability and an autophagy inducer to clear aggregate-prone proteins, a disaccharide trehalose has been reported to alleviate neurodegeneration in PD cells and mouse models. Its trehalase-indigestible analogs, lactulose and melibiose, also demonstrated potentials to reduce abnormal protein aggregation in spinocerebellar ataxia cell models. In this study, we showed the potential of lactulose and melibiose to inhibit α-synuclein aggregation using biochemical thioflavin T fluorescence, cryogenic transmission electron microscopy (cryo-TEM) and prokaryotic split Venus complementation assays. Lactulose and melibiose further reduced α-synuclein aggregation and associated oxidative stress, as well as protected cells against α-synuclein-induced neurotoxicity by up-regulating autophagy and nuclear factor, erythroid 2 like 2 (NRF2) pathway in DAergic neurons derived from SH-SY5Y cells over-expressing α-synuclein. Our findings strongly indicate the potential of lactulose and melibiose for mitigating PD neurodegeneration, offering new drug candidates for PD treatment.


Introduction
Parkinson's disease (PD) is the second most common neurodegenerative disorder affecting 1% people older than 60 years old. The symptoms commonly seen in PD patients are resting tremor, rigidity, bradykinesia and postural instability. These symptoms result predominantly from a massive loss of dopaminergic (DAergic) neurons located in the pars compacta of the substantia nigra (SN) 3 of 21 cDNA (NM_000345) was amplified using sense (5 -CCATGGATGGATGTATTCATGAAA-GGAC-3 ) and antisense (5 -CTCGAGGGCTTCAGGTTCGTAGTC-3 ) primers and cloned into pGEM-T Easy vector (Promega, Fitchburg, WI, USA) and sequenced. The 432-bp amplified SNCA cDNA fragment was excised with NcoI and XhoI and ligated into the corresponding sites of pET-28a(+) (Novagen, Madison, WI, USA). The resulting pET28/SNCA plasmid was transformed into Escherichia coli (E. coli) BL21(DE3) (Novagen), selected with kanamycin (30 µg/mL), and His-tagged SNCA protein expression was induced with 0.4 mM isopropyl-β-d-thiogalactopyranoside (IPTG) for 3 h at 37 • C. Bacterial cells were then harvested and the SNCA-His protein was purified using His-Bind resins (Novagen) according to the supplier's instructions. Both bacterial cell lysates and purified SNCA-His protein were examined by Coomassie blue staining of SDS-polyacrylamide gel electrophoresis (SDS-PAGE) gels (10%). The purified SNCA-His proteins were concentrated and solvent-exchanged using Amicon Ultra-4 centrifugal filters with 10-kDa molecular mass cutoff (Millipore, Temecula, CA, USA). Aliquots of protein were stored at −20 • C.
To examine α-synuclein aggregation, the formed fibrillar samples-without disaccharide treatment on day one to three and with disaccharide (4 µM) treatment on day three-were imaged by cryo-TEM. Samples were placed on a 200-Mesh copper (holey-carbon) grid and plunged into liquid nitrogen. They were stored frozen until visualization, in which samples were inserted into a Gatan CP3 cryoholder and viewed at 200 kV by JEM-2100F transmission electron microscope (JEOL, Tokyo, Japan). Images were recorded by using a Gatan 1024 × 1024 CCD camera (Gatan, Inc., Pleasanton, CA, USA).
For filter trap assay to quantify α-synuclein aggregates, the purified SNCA-His protein was incubated with tested disaccharide (4 µM) at 37 • C for three days as described. Protein (0.5 µg) was diluted in 2% SDS in PBS and filtered through a cellulose acetate membrane (0.2-µm pore size; Merck, Kenilworth, NJ, USA) pre-equilibrated in 2% SDS in PBS on a slot-blot filtration unit (GE Healthcare, Chicago, IL, USA). After three washes with 2% SDS buffer, the membrane was blocked in PBS containing 5% non-fat dried milk and stained with anti-α-synuclein antibody (1:1000; BD Biosciences #610787). The immune complexes on the filter were detected as described.

α-Synuclein Aggregation in E. coli Monitored by Split Venus BiFC and Filter Trap Assays
The E. coli cells with induced complementary halves of Venus were treated with tested disaccharide (1-1000 µM) for 1 h and VN 1-211 -SNCA and SNCA-VC 212-239 protein expressions were induced with 0.4 mM IPTG for 3 h at 37 • C. The complementary Venus fluorescence was measured using a fluorometer (Bio-TeK FLx800) with 485 ± 10 nm excitation and 528 ± 10 nm emission. In addition, bacterial cell lysates from disaccharide (1 mM)-treated E. coli cells were prepared and proteins (1 µg) subjected to filter trap assay as described.
Briefly, SH-SY5Y cells in Dulbecco's modified Eagles medium (DMEM)/F12 with 10% fetal bovine serum (FBS) were seeded at 5 × 10 4 cells in 12-well plates. The next day, cell supernatant was replaced by 0.25 mL of fresh medium containing 8 µg/mL of polybrene (Sigma-Aldrich) and 0.01 multiplicity of infection (MOI) of the lentivirus. After 6 h, the medium was changed into fresh media. The next day, the infected cells were passaged into a 10-cm dish, followed by addition of blasticidin (5 µg/mL; InvivoGen, San Diego, CA, USA) on the next day to select for stable transfectants. Fresh blasticidin-containing medium was added every 3-4 days until the un-transduced control cells were completely dead (about 3-4 weeks). Single cell clones were picked and plated in a 96-well plate for 10 days, followed by further expansion for an additional three weeks under blasticidin selection. The established blasticidin-resistant SH-SY5Y clones were examined for doxycycline-induced (10 µg/mL; Sigma-Aldrich) SNCA-GFP expression using anti-α-synuclein (1:1000; BD Biosciences #610787) and anti-GFP (1:500; Santa Cruz Biotechnology #sc-9996) antibodies.

Reactive Oxygen Species (ROS) Assessment
The SNCA-GFP SH-SY5Y cells were incubated at 37 • C for 30 min in the fluorogenic CellROX™ Deep Red Reagent (5 µM; Molecular Probes, Eugene, OR, USA), which is designed to measure ROS reliably in live cells. Subsequently, the cells were washed with PBS and analyzed for red (ROS) fluorescence on a flow cytometry system (Becton−Dickinson, Franklin Lakes, NJ, USA), with excitation (helium-neon laser)/emission wavelengths at 633/661 ± 8 nm. Each sample contained 2 × 10 4 cells.

Caspase 1 and 3 Activities and Lactate Dehydrogenase (LDH) Release Assays
The SNCA-GFP SH-SY5Y cells were pretreated with each test disaccharide and induced SNCA-GFP expression in the presence of α-synuclein fibrils as described. For LDH release assay, cell culture media were collected on day 14 and the release of LDH was examined by using LDH cytotoxicity assay kit (Cayman, Ann Arbor, MI, USA). The absorbance was read at 490 nm with Multiskan GO microplate reader. For caspase 1 and 3 activity assays, cells were lysed in 1 × lysis buffer by repeated cycles of freezing and thawing. Caspase 1 and 3 activities were measured using the caspase 1 (BioVision) and caspase 3 (Sigma-Aldrich) fluorimetric assay kits, with excitation/emission wavelengths of 420 ± 25/485 ± 10 nm (caspase 1 assay) or 360 ± 20/460 ± 20 nm (caspase 3 assay) (FLx800 fluorescence microplate reader, Bio-Tek).

Statistical Analysis
For each set of values, three independent experiments were performed and data were expressed as the means ± standard deviation (SD). Differences between groups were evaluated by Student's t-test or one-way ANOVA (analysis of variance) with a post hoc Tukey test where appropriate. All p values were two-tailed, with values of p < 0.05 considered significant.

DAergic Differentiation of SH-SY5Y Cells with Induced α-Synuclein-GFP Expression
Lentivirus containing GFP-tagged SNCA (pAS4.1w.Pbsd-aOn-SNCA-GFP, Figure 4A) was used to transduce human neuroblastoma SH-SY5Y cells. The expanded single cell clones with good resistance to blasticidin were examined for SNCA-GFP expression after 10 µg/mL doxycycline addition for two days ( Figure 4B). Clone 3 with the greatest amount SNCA-GFP was further examined for TH expression, a DAergic neuronal marker, with 120 nM TPA treatment ( Figure 4C). After 13 days of TPA treatment, cells displayed the properties of DAergic neurons, with increased expression of TH by immunoblot ( Figure 4D), and 95% of cells being TH-positive by immunostaining ( Figure 4E).

DAergic Differentiation of SH-SY5Y Cells with Induced α-Synuclein-GFP Expression
Lentivirus containing GFP-tagged SNCA (pAS4.1w.Pbsd-aOn-SNCA-GFP, Figure 4A) was used to transduce human neuroblastoma SH-SY5Y cells. The expanded single cell clones with good resistance to blasticidin were examined for SNCA-GFP expression after 10 μg/mL doxycycline addition for two days ( Figure 4B). Clone 3 with the greatest amount SNCA-GFP was further examined for TH expression, a DAergic neuronal marker, with 120 nM TPA treatment ( Figure 4C).  After 13 days of TPA treatment, cells displayed the properties of DAergic neurons, with increased expression of TH by immunoblot ( Figure 4D), and 95% of cells being TH-positive by immunostaining ( Figure 4E).

Discussion
The α-synuclein is prone to aggregate formation [41]. Several lines of evidence have shown the important role of misfolded α-synuclein in the pathogenesis of PD [42] and degradation of misfolded α-synuclein has been suggested to be one of the therapeutic strategies for PD [43]. Apart from targeting on ubiquitin-proteasome system (UPS), chaperone-mediated autophagy (CMA), and macroautophagy, the development of potent chemical chaperones could be a therapeutic approach for neurodegenerative diseases caused by the misfolded protein [44][45][46]. Several molecules including osmolytes have been identified as chemical chaperones that can refold proteins with conformation change [47]. As an autophagy inducer, trehalose has promising therapeutic effects on cellular and animal models of aggregation-prone neurodegenerative diseases [17][18][19][20][21]48,49]. In addition to its autophagy-inducing function, trehalose has been shown to have chemical chaperone activity [30,45,50]. Lactulose is composed of galactose and fructose. The indigestible lactulose has been in medical use for over 40 years, mainly in the treatment of portosystemic encephalopathy and of constipation [51]. Melibiose, formed by partial hydrolysis of raffinose, can be broken down into galactose and glucose by the enzyme α-galactosidase. We hypothesized that, similar to trehalose [13,52], both lactulose and melibiose may stabilize aggregation-prone proteins by refolding the abnormal protein conformation, which may be supported by the thioflavin T fluorescence, cryo-TEM and split Venus BiFC assays. By using the thioflavin T fluorescence assay and cryo-TEM, we Autophagy induction and oxidative stress reduction of tested disaccharides in SNCA-GFP-expressing SH-SY5Y cells. On day eight, TPA-differentiated SH-SY5Y cells were pretreated with 100 µM trehalose, lactulose, or melibiose for 8 h followed by induction of SNCA-GFP expression and addition of α-synuclein fibrils for six days. Relative (A) LC3-I and LC3-II and (B) NRF2, NQO1, GCLC protein levels were analyzed (n = 3). GAPDH was included in immunoblot as an internal control. To normalize, expression level in cells uninduced and without preformed fibril addition (−Dox/−α-Syn fibrils) was set at 100%. p values: comparisons between −Dox/−α-Syn fibrils vs. +Dox/+α-Syn fibrils ( ## : p < 0.01, ### : p < 0.001), between +Dox/−α-Syn fibrils vs. +Dox/+α-Syn fibrils ( & : p < 0.05, && : p < 0.01), or between +Dox/+α-Syn fibrils-cells with and without disaccharide treatment (*: p < 0.05, **: p < 0.01; one-way ANOVA with a post hoc Tukey test).

Discussion
The α-synuclein is prone to aggregate formation [41]. Several lines of evidence have shown the important role of misfolded α-synuclein in the pathogenesis of PD [42] and degradation of misfolded α-synuclein has been suggested to be one of the therapeutic strategies for PD [43]. Apart from targeting on ubiquitin-proteasome system (UPS), chaperone-mediated autophagy (CMA), and macroautophagy, the development of potent chemical chaperones could be a therapeutic approach for neurodegenerative diseases caused by the misfolded protein [44][45][46]. Several molecules including osmolytes have been identified as chemical chaperones that can refold proteins with conformation change [47]. As an autophagy inducer, trehalose has promising therapeutic effects on cellular and animal models of aggregation-prone neurodegenerative diseases [17][18][19][20][21]48,49]. In addition to its autophagy-inducing function, trehalose has been shown to have chemical chaperone activity [30,45,50]. Lactulose is composed of galactose and fructose. The indigestible lactulose has been in medical use for over 40 years, mainly in the treatment of portosystemic encephalopathy and of constipation [51]. Melibiose, formed by partial hydrolysis of raffinose, can be broken down into galactose and glucose by the enzyme α-galactosidase. We hypothesized that, similar to trehalose [13,52], both lactulose and melibiose may stabilize aggregation-prone proteins by refolding the abnormal protein conformation, which may be supported by the thioflavin T fluorescence, cryo-TEM and split Venus BiFC assays. By using the thioflavin T fluorescence assay and cryo-TEM, we demonstrated that both lactulose and melibiose are effective in reducing α-synuclein fibrillation. Similarly, Yu et al. also showed trehalose decreased A53T α-synuclein fibrillation by using thioflavin T fluorescence assay [53]. They further demonstrated that trehalose changed β-sheet structure of α-synuclein to random coil by using circular dichroism spectroscopy. The similar results were also found by another study using synchrotron radiation circular dichroism spectroscopy, where trehalose could interact with α-synuclein, affecting its folding property in dose-dependent manner [54]. Regarding the mechanism of how trehalose refolds the abnormal structure of α-synuclein, although remains unclear, water-layer with preferential exclusion is proposed by Yu et al. that the interaction between trehalose and A53T α-synuclein is stronger than that of inter-hydrogen bonding within A53T α-synuclein, thus preventing self-association of A53T α-synuclein [53]. Although we did not show the structure change by using circular dichroism spectroscopy, we used cryo-TEM to show the aggregates and fibrils were decreased by treatment with tested disaccharides.
Oligomers, fibrils, and aggregates of α-synuclein have been suggested to cause neurotoxicity of PD. Recently, BiFC has been used as an important technique to visualize protein-protein interactions in different models [55]. BiFC assay is based on reconstitution of an intact fluorescent protein when two complementary non-fluorescent fragments are brought closely to proximity by a pair of interacting proteins or by protein self-association. This technique has been also used to image oligomerization, fibrillation, and exosomal cell-to-cell transmission of α-synuclein in cellular and mouse models [5,[56][57][58]. This technique is important not only for pathogenesis investigation but also for screening of drugs that can inhibit α-synuclein oligomerization, fibrillation, and aggregation. In the present study, by applying the split Venus BiFC assay we have demonstrated the effects of trehalose, lactulose and melibiose on α-synuclein aggregation inhibition.
It is important to note that trehalose is readily digested by trehalase in the gut of humans [25], which implicates trehalase-indigestible analogs rather than trehalose as the potential treatments for aggregation-associated neurodegenerative diseases. Previously, we have found that lactulose and melibiose indigestible by trehalase have anti-aggregation and neuroprotection effects in SCA3 and SCA17 cell models, mainly through autophagy-activation [18,19]. Recently, in vitro studies showed that treatment with trehalose significantly reduced oxidative stress induced by chloroquine or cadmium via activating the kelch-like ECH-associated protein 1 (KEAP1)/NRF2 pathway, suggesting it is being a strong antioxidant [59,60]. In this study, using TPA-induced neuronal differentiation of SH-SY5Y cells that express α-synuclein, we have shown potent effects of lactulose and melibiose in reducing α-synuclein aggregation aggravated by addition of α-synuclein fibril and in promoting neurite outgrowth. We then showed that lactulose and melibiose, in addition to trehalose, rescued the decreased LC3-II/LC3-I ratio and increased LDH release, ROS level and caspase 1/3 activity, which were resulted from expression of SNCA-GFP plus α-synuclein fibrils. We further demonstrated that trehalose, lactulose and melibiose increased NRF2 expression and its downstream genes, NQO1 and GCLC, leading to decreased oxidative stress. These results support our and other researchers' previous findings that trehalose, lactulose and melibiose promote degradation of aggregates and decrease oxidative stress via enhancing autophagy and NRF2 pathway to provide neuroprotection effects [18,19,59]. It is noted that misfolded α-synuclein in either mutant or wild type form has been shown to cause neurotoxicity of PD, although mutant form such as A53T may be more prone to aggregation [41,61]. Since our SNCA-GFP-expressing SH-SY5Y cell model has shown significant aggregates and neurotoxicity, we did not establish another cell model expressing A53T SNCA. However, future studies are warranted to test the effects of lactulose and melibiose on a more toxic disease model expressing A53T SNCA. Substantial evidence has shown that trehalose can be detected in the brain homogenates of HD transgenic mice administered orally with 2% trehalose in drinking water, suggesting its ability to cross the blood-brain barrier (BBB) [13]. Similarly, we propose that lactulose and melibiose indigestible by trehalase, may also penetrate the BBB to exert their neuroprotection effect in the brain. However, lactulose is poorly absorbed into the blood (~3% at most) in humans and this will undoubtedly limit its potential clinical use. Therefore, it is important to develop other delivery methods that can increase the lactulose concentration in animal and human brain in the future. Future studies in PD animal models are warranted to further consolidate the neuroprotection effect of lactulose.
In conclusion, our results show that trehalose, lactulose and melibiose inhibit α-synuclein aggregation and up-regulate autophagy to reduce neurotoxicity. Given that lactulose and melibiose are trehalase-indigestible, we propose that lactulose and melibiose may have potential as the future therapeutics for human PD.