https://orcid.org/0000-0001-7273-1847
Abstract
Alzheimer’s disease (AD) is the most prevalent type of dementia globally. The accumulation of amyloid-beta (Aβ) extracellular senile plaques in the brain is one of the hallmark mechanisms found in AD. Aβ42 is the most damaging and aggressively aggregating Aβ isomer produced in the brain. Although Aβ42 has been extensively researched as a crucial peptide connected to the development of the characteristic amyloid fibrils in AD, the specifics of its pathophysiology are still unknown. Therefore, the main objective was to identify novel compounds that could potentially mitigate the negative effects of Aβ42. 3-[[(3S)-1,2,3,4-Tetrahydroisoquinoline-3-carbonyl]amino]propanoic acid (THICAPA) was identified as a ligand for Aβ42 and for reducing fibrillary Aβ42 aggregation. THICAPA also improved cell viability when administered to PC12 neuronal cells that were exposed to Aβ42. Additionally, this compound diminished Aβ42 toxicity in the current AD Drosophila model by rescuing the rough eye phenotype, prolonging the life span, and enhancing motor functions. Through next-generation RNA-sequencing, immune response pathways were downregulated in response to THICAPA treatment. Thus, this study suggests THICAPA as a possible disease-modifying treatment for AD.
Aging, a process of growing older and senescence, increases the risk of age-related illnesses like dementia (1). With a current average life expectancy of 72.6 years for women and 67.2 years for men (2), the number of older individuals worldwide is projected to rise from 1 billion in 2020 to 2.1 billion by 2050 (3). Due to this, age-associated neurodegenerative illnesses are becoming significant public health challenges. The most prevalent type of dementia, Alzheimer’s disease (AD), is characterized by a progressive and gradual loss of cognitive ability (4). Although memory loss is the disease’s most well-known symptom, AD patients frequently experience a variety of other symptoms, including behavioral changes, motor impairment, and eventually the inability to perform basic daily tasks (4). The disease is multifactorial, with numerous theories proposed on AD pathogenesis. The deposition of amyloid-beta (Aβ) in the brain is the most widely accepted theory (5,6). Among the Aβ present in the brain, Aβ42 is the most neurotoxic and aggressively aggregating species (7). Despite decades of research, there is yet to be successful treatments for AD. The available U.S. Food and Drug Administration-approved AD medications merely aid to delay the onset of symptoms (8). As the world population shifts toward older ages, the demand for new and improved AD treatments becomes more pressing.
AD therapeutics discovery commonly focuses on areas such as probiotic studies (9,10) and phytotherapy (11,12). Here, the primary goal was to identify novel compounds with the potential to counteract the adverse effects of Aβ42. First, in vitro techniques were carried out to identify Aβ42 ligands that reduce Aβ42 fibrillation. PC12 neuronal cells were used as a screening platform to obtain ligands that protect against Aβ42-associated toxicity. Next, a transgenic Drosophila melanogaster model that carries the human Aβ42 gene was used to test the potential anti-AD effects of the compounds. In this study, 3-[[(3S)-1,2,3,4-Tetrahydroisoquinoline-3-carbonyl]amino]propanoic acid (THICAPA) was identified as an Aβ42 ligand that ameliorated the toxic effects of Aβ42 in vitro and in neuronal cell culture, as well as in the AD Drosophila. To further understand the anti-AD mechanism of this compound, RNA-sequencing was performed to obtain a concise review of the differentially expressed genes (DEGs) associated when THICAPA was fed to the AD Drosophila. From this, we propose THICAPA as a possible novel treatment for AD.
Materials and Methodology
Chemical Array Assay
In-house histidine-tagged Aβ42 peptides were prepared for the chemical array assay. Genomic DNA from D. melanogaster carrying the human Aβ42 gene (Bloomington Stock Centre, #33769) was extracted. The Aβ42 region was amplified via polymerase chain reaction (PCR) using primers (forward primer: 5ʹ-GCAGTAGCATATGGATGCAGAATTCCGACA-3ʹ, reverse primer: 5ʹ-TAATCTCGAGCGCAATGACAA-3ʹ) and inserted into the pET21a(+) plasmid (Novagen, Cat no.: 69740-3) at the NdeI and XhoI sites to produce pET21a-Aβ42. The pET21a-Aβ42 plasmid was transformed into Escherichia coli BL21 and grown in a lysogeny broth medium. Fermentation was performed in shake flasks with 100 μg/mL ampicillin at 37° C and 180 rpm. When cultures achieved an OD600 of 0.4–0.6, isopropyl β-D-1-thiogalactopyranoside was added at a final concentration of 1 mM to induce the production of the Aβ42 peptide. The pellet was harvested by centrifugation after 5 hours of induction.
For histidine-tagged Aβ42 peptide purification, the cell pellet was suspended in phosphate buffer (625 mM sodium chloride, 60 mM Na2HPO4) and lysed via sonication. The bacterial lysate was centrifuged at 8 000g for 10 minutes at 4° C. HisTALON Gravity (TaKaRa) column with Talon Metal Affinity Resin (TaKaRa) was incubated at room temperature for an hour. The column and sample pellet were equilibrated and washed with Buffer A (8 M urea, 50 mM Tris). The sample was loaded into the column and the flow-through was collected in prechilled conical centrifuge tubes. The resin was eluted with phosphate buffer containing 100 mM imidazole. Collected flow-through containing purified histidine-tagged Aβ42 peptide was lyophilized and stored at −80° C for further use.
For the chemical array assay (13), a chip laced with 22 097 compounds from the RIKEN (Wako, Saitama, Japan) Natural Products Depository (NPDepo) was coated with a layer of the in-house produced histidine-tagged Aβ42 and incubated for 60 minutes at 37° C. The chip was then washed to eliminate any residue histidine-tagged Aβ42 that failed to bind to any compound. Subsequently, immobilized compounds that adhered to the histidine-tagged Aβ42 were identified via immunostaining with anti-histidine antibodies.
Thioflavin T (ThT) Aβ42 Aggregation Assay
The ThT assay used the SensoLyte ThT β-Amyloid (1–42) Aggregation Kit (Anaspec, Cat no: AS-72214) with alterations to the manufacturer’s protocol. In a black 96 well-plate, 42.5 µM of Aβ42 (Anaspec, Cat no: AS-72216) was added to respective compounds (50 µg/mL each). In a dark room, 20 µM of ThT was added to each well. The fluorescence reading was read via the microplate reader (Biotek Synergy 2 SLFP Multimode, Biotek) every 300 seconds for a total of 3 600 seconds at an excitation/emission of 440 nm/484 nm with pulsed shaking at 37° C. Morin (final concentration = 50 µg/mL) was used as the positive control.
Cell Viability Determination by ATP Assay
Rat pheochromocytoma PC12 cells (RIKEN Cell Bank, Tsukuba, Japan) were cultured in Dulbecco’s Modified Eagle Media (DMEM) (Gibco, Cat no.: C11995500BT) added with 10% Fetal bovine serum (Sigma Chemical Co. St Louis, Cat no.: 172012) and 10% Horse serum (Gibco, Cat no.: 26050-070) at 37° C and 5% CO2. To induce the differentiation of PC12 cells, cultures were supplemented with DMEM containing 10% Horse serum and 100 ng/mL Nerve Growth Factor (NGF; Sigma Chemical Co., Cat no.: H9666-10UG).
Aβ42 (Anaspec, Cat no: AS-72216) was added to respective compounds and aged for 72 hours at 37° C (14). Cells were seeded in white 96 clear bottom well plates at 5 × 103 cells/100 µL with DMEM containing 10% Fetal bovine serum and 10% Horse serum. After 24 hours, the available media was replaced with DMEM containing 10% Horse serum and 100 ng/mL NGF. To determine if PC12 cells have differentiated after NGF treatment, morphological changes were observed (Supplementary Figure 1). Prior to the supplementation of NGF, PC12 cells had round to ovoid structures. After incubation with NGF for 72 hours, the cells grew branching dendrites. Aged Aβ42 at a final concentration of 10 µM with 50 µM of respective compounds were added to the wells and incubated for 24 hours. Equal volumes of CellTiter-Glo (Promega, Cat no: G7570) were added. The plate was shaken for 2 minutes and left to stand at room temperature for 10 minutes. Luminescence reading was taken using the microplate reader (Varioskan LUX multimode). Each experiment was performed in triplicate.
Drosophila Stocks and Husbandry
All Drosophila stocks (Supplementary Tables 1 and 2) used in this study are listed at http://flybase.bio.indiana.edu. All stocks were cultured at 25° C whereas crosses were maintained at 29° C. To facilitate mating, 5–10 virgin female flies (Gal4 or UAS line) and 3 to 5 male flies of the corresponding parent line were placed into culture vials supplemented with solid feed. For wild-type controls, Oregon-R was mated with the GAL4 line used during the particular analysis to yield GAL4-OreR (15,16). On the other hand, UAS-Aβ42 was mated with the specific GAL4 line to yield the transgenic fly line GAL4-Aβ42 that expressed Aβ42 in desired tissues.
Solid feed was prepared with 10% brown sugar, 5% cornmeal, 5% inactivated yeast, 4% corn starch, 3% nipagin, 0.7% agar, 0.7% propionic acid, and 0.5% dimethyl sulfoxide (DMSO) either unaccompanied or with THICAPA. The CApillary FEeder (CAFE) (17) assay was adjusted and applied to both the negative geotaxis assay and life span analysis. Liquid feed for the CAFE assay followed the same recipe as solid feed without cornmeal or agar.
Rough eye Phenotype (REP) Assay
AD Drosophila fed with different concentrations of THICAPA or untreated (0.5% DMSO) were fixed in McDowell-Trump (Sigma-Aldrich, Missouri, USA) overnight at 4° C. Following 0.1 M phosphate buffer (Sigma-Aldrich) rinsing, the samples were post-fixed for an hour in 1% osmium tetroxide (Sigma-Aldrich). The samples were rinsed once with distilled water, dehydrated for 15 minutes each in increasing concentrations of ethanol, dried for 10 minutes in hexamethyldisilazane (Sigma-Aldrich), and placed in a desiccator to air dry overnight. Before the scanning electron microscopy (SEM; SU8010; Hitachi Ltd., Tokyo, Japan) examination, samples were mounted and coated with gold. Triplicate SEM images of each sample were taken for analysis using the Flynotyper (https://flynotyper.sourceforge.net) (18).
Life-span Assay
Experimental Drosophila was gathered within 24 hours of eclosure and moved to CAFE vials (<20 per container) at 29° C and humidity of 60% (19,20). Daily counts of dead Drosophila were made. Surviving Drosophila were moved daily to sterile vials with fresh liquid feed containing either THICAPA or untreated (0.5% DMSO). The maximum life span is the day when the final Drosophila in a cohort expires. Triplicate of each line was performed.
Negative Geotaxis Assay
The average climbing speed of experimental Drosophila was used to gauge its mobility. The Drosophila were transferred from CAFE vials to climbing vials and allowed to acclimatize. mina HiSeq PE150 platformThe Drosophila’s ascent of the vials was recorded and analyzed through the Toxtrac program (https://sourceforge.net/projects/toxtrac/) (21). The experiment was done in triplicate.
Total RNA Extraction and RNA-seq
TRIzol (Invitrogen, Massachusetts, USA) combined with the RNeasy Mini kit (Qiagen, Hilden, Germany) was used to extract total RNA from samples of the AD Drosophila strains (Actin5C-Aβ42.DMSO and Actin5C-Aβ42.TH100). Ten experimental Drosophila were quickly homogenized in 300 µL of TRIzol. A total of 80 µL of chloroform was added and vortexed for 20 seconds. To separate the layers, the mixture was centrifuged at 10 000g for 10 minutes at 4° C. After transferring the aqueous layer to a fresh tube, the sample was mixed with 200 µL of absolute, ice-cold isopropanol. The MinElute Cleanup Kit (Thermo Fisher Scientific, Massachusetts, USA) and Turbo DNase Kit (Thermo Fisher Scientific) were used to clean samples following the manufacturer’s instructions. Gel electrophoresis, NanoDrop 2000/2000c (Thermo Fisher Scientific), and 2100 Bioanalyzer (Agilent, California, USA) were used to assess the total RNA quality. For library construction, high-quality total RNA (minimum 5 μg; minimum 200 ng/μL and OD260/280 = 1.8–2.2) was used. Samples were collected in triplicate with 5 males and 5 females for every replicate. Apical Group (Singapore) constructed the libraries using TruSeq RNASample Preparation (Illumina, California, USA) and carried out sequencing using the Illumina HiSeq PE150 platform. Raw data were processed by In-house Perl scripts. The library construction parameters are: Q20 over 95%, Q30 over 90%, 30%–70% GC content, effective rate over 90%, the error rate of 0.01%, and a sequencing depth of minimum of 30 million reads per sample.
Differential Expression Analysis
Reads were uploaded onto the Galaxy server (https://usegalaxy.org/). Using RNA STAR (Galaxy Version 2.7.8a + galaxy0) with the “Paired-end” and “unstranded strand” information, they were mapped to the reference genome for D. melanogaster (https://zenodo.org/record/1185122/files/Drosophila_melanogaster.BDGP6.87.gtf) (22). The mapped reads were counted using featureCounts (Galaxy Version 2.0.1 + galaxy2) (23). The differential expression analysis used DESeq2 R tool (version 3.6.3) (24). For each gene, a filtering criterion with a threshold of more than 5 counts for at least 2 samples was used. To control the false discovery rate, the Benjamini and Hochberg technique was used to produce adjusted p values (padj). Genes were classified as DEGs if their padj value was <.05 and their log2FoldChange (log2FC) value >1 and <1.
The Gene Ontology (GO) enrichment analysis of the DEGs was executed using g:profiler version e107_eg54_p17_bf42210 (https://biit.cs.ut.ee/gprofiler/gost) (25), which also integrated results from Kyoto Encyclopedia of Genes and Genomes (KEGG) Pathway database. For further functional analysis and visualization, REVIGO (http://revigo.irb.hr/) was used (26).
Quantitative Reverse Transcription PCR (RT–qPCR)
Quantitative real–time PCR (qRT–PCR) was used to confirm the transcriptome analysis using a housekeeping gene, and 5 downregulated DEGs (Supplementary Table 3). Following the manufacturer’s protocol, cDNA was synthesized from total RNA using the iScript Reverse Transcription Supermix (Bio-rad, California, USA). qRT–PCR was carried out using iTaq Universal SYBR Green Supermix (Bio-rad) in a CFX96 (Bio-rad) real-time thermal cycler. The experiment was carried out in triplicate per sample.
Statistical Analysis
Student’s t test was generally used to statistically analyze data obtained. The log-rank test was used to analyze the significance of data for the life-span assay. p Values represent the statistical significance of the observed difference between data. Error bars represent standard deviation (SD).
Results
In Vitro Screening of Ligands Showed 7 Compounds That Reduced Aβ42 Aggregation
The new expression vector, pET21a-Aβ42, with a size of 5 493 bp (Figure 1A), contained the Aβ42 sequence and hexa-histidine tag at the C-terminal. The recombinant plasmid was transformed into E. coli strain BL21 culture, and the pellets were purified after 5 hours of 1 mM isopropyl β-D-1-thiogalactopyranoside induction to obtain histidine-tagged Aβ42 peptides for the chemical array assay. It was found that 86 compounds were able to retain the histidine-tagged Aβ42. Figure 1B shows the setup of the chemical array assay.
In vitro screening of compounds. (A) pET21a-Aβ42 vector map (B) Schematic representation depicting the mechanism of the chemical array adapted from (27) (C) Second round of ThT Aβ42 aggregation assay of Aβ42 ligands. Data represents one experiment. (D) ATP cell viability assay. “Control” bar denotes control cells without the incubation of Aβ42 and compounds. “Vehicle control” and compound bars denote cells with vehicle control (0.5% DMSO) or designated compounds, respectively, in the presence of 10 μM Aβ42. The hashtag (#) signifies p-values against the black-bar control, whereas the asterisk (*) signifies p-values against the vehicle control. p-values show significance at *p < 0.05, #p < 0.05 ##p < 0.005, ###p < 0.0005. Data represents mean ± SD of three biological replicates. (E) Chemical structure depiction of THICAPA.
Subsequently, the Aβ42 ligands were subjected to a second in vitro screening using ThT. When ThT binds to aggregated fibrils, ThT undergoes a distinguishing red shift in its emission spectrum (Groenning, 2010). Compounds that resulted in at least 30% lower fluorescence readings than the vehicle control (0.5% DMSO) at the time-point of approximately 3 600 seconds (indicated by dotted lines in Supplementary Figure 2 and Figure 1C) when incubated with Aβ42 indicated the compound’s ability to inhibit Aβ42 aggregation. Out of the 86 compounds, 7 Aβ42 ligands were able to meet the benchmark (Supplementary Figure 2). These 7 compounds were repeated for a second round using the ThT Aβ42 aggregation assay to test for reproducibility, and all 7 Aβ42 ligands achieved the benchmark (Figure 1C).
THICAPA Increased the Cell Viability of Aβ42-Incubated Cells
To investigate the protective impact of the 7 screened compounds against neuronal degeneration brought on by the addition of Aβ42 peptides in neurons, PC12 rat pheochromocytoma neuronal cells were used. In Figure 1D, the presence of Aβ42 peptides in the vehicle control significantly reduced cell viability to 40% compared to the control (without 10 μM Aβ42). When compared against the vehicle control, only THICAPA was significantly different with a cell viability of 92%, while the other 6 compounds did not show any significant differences. Contrarily, when compared against the control (without 10 μM Aβ42), only THICAPA was not significantly different. This demonstrated THICAPA’s capacity to improve cell viability of Aβ42-incubated cells.
THICAPA (Figure 1E) has an IUPAC compound name of 3-[[(3S)-1,2,3,4-Tetrahydroisoquinoline-3-carbonyl]amino]propanoic acid , molecular formula of C13H16N2O3 and a molecular weight of 248.28 g/mol. Further details of the compound can be found in PubMed (https://pubchem.ncbi.nlm.nih.gov/compound/7094666).
THICAPA Ameliorated the REP of Drosophila AD Model
The model organism D. melanogaster was used to observe how THICAPA affects an entire organism. First, phenotypes of the AD Drosophila eyes were scrutinized. As each of the 800 ommatidia in a Drosophila eye contains 8 photoneurons, any irregularities in the neurons may manifest as deformations in the eye morphology called the REP (28). GMR-GAL4 was used to drive the expression of Aβ42 to the eyes. Light microscopy was first used to observe the eyes of GMR-Aβ42 fed with THICAPA to identify the ideal concentration needed to improve the REP. When Aβ42 was expressed in the untreated GMR-Aβ42.DMSO (Figure 2B) Drosophila, the eyes were severely deformed with perforation and fused ommatidia in comparison to the wild-type control GMR-OreR.DMSO (Figure 2A). Administration of various concentrations of THICAPA onto GMR-Aβ42 resulted in the dose-dependent recovery of the REP (Figure 2C–E). GMR-Aβ42.TH100 showed the best rescue effect in eye structures when fed 100 μM of THICAPA with the least merged and distorted ommatidia (area indicated by a dotted circle) compared to 50 μM and 500 μM.
Rough eye phenotype analysis. (A)–(E) Comparison of Drosophila eye light micrographs viewed at 11.5× magnification. Areas with a high density of merged or distorted ommatidia are indicated by dotted circles. (F)–(H) Comparison of Drosophila eye scanning electron micrographs viewed at 200× magnification. (Fʹ)–(Hʹ) Comparison of Drosophila eye scanning electron micrographs viewed at 1 500× magnification. (I) P-scores of Drosophila eyes. The hashtag (#) signifies p values against the wild-type control, whereas the asterisk (*) signifies p values against the untreated. p Values show significance at *p < .05, #p < .05, and ##p < .005. Data represents mean ± SD of 3 Drosophila per line.
To further evaluate the REPs at higher magnifications, SEM images of the eyes were inspected (Figure 2F–H). Quantification of the eye deformities was measured as a phenotypic score (P-score) ratio; the more severe the REP, the higher the ratio (Figure 2I). When compared to GMR-OreR.DMSO, GMR-Aβ42.DMSO possessed a significantly higher average P-score of 1.87 (p = .0068). This established the toxicity of Aβ42 in the eyes of the Drosophila AD model when untreated. Conversely, eyes fed with 100 μM THICAPA were significantly different (p = .033) with an average P-score of 1.44 compared to GMR-Aβ42.DMSO. This supported the light micrograph findings that THICAPA reduced Aβ42-related REP in the Drosophila AD model in a concentration-dependent manner, with 100 μM THICAPA being the most beneficial concentration.
THICAPA Prolonged Life Span of AD Drosophila
The long-term influence of 100 μM THICAPA on the life span of Aβ42-expressing Drosophila was assessed. This was accomplished using the Actin5C-GAL4 driver, which promotes widespread expression. Results from Supplementary Tables 4 and 5, Figure 3A and C proved that THICAPA was safe for consumption and had no significant negative impacts on either the male or female life spans of Actin5C-OreR.TH100 wild-type control. In contrast, untreated Actin5C-Aβ42.DMSO had shorter life spans compared to the wild-type control. The feeding of THICAPA to both male and female Aβ42-expressing Drosophila prolonged the average maximum life spans by 47.4% (Figure 3B) and 69.6% (Figure 3D), respectively, compared to their respective Actin5C-Aβ42.DMSO counterparts. To evaluate the significance of life span differences among the Drosophila lines, a Log-rank test was used. Although Actin5C-OreR.DMSO had an extremely low p value (p < .0005 for both males and females) compared to the untreated Actin5C-Aβ42.DMSO, it was less significantly different against Actin5C-Aβ42.TH100 for both males and females with p values of p = 5.5E−8 and p = 9.7E−9 values, respectively. This data implied that THICAPA ameliorated the life span of Aβ42-expressing Drosophila.
The effect of THICAPA on the life span of transgenic AD Drosophila. (A) and (C) show the longevity graphs of Actin5C-OreR males and females, respectively. (B) and (D) depict the graphs of Actin5C-Aβ42 males and females, respectively. Data represents mean ± SD of 3 biological replicates with approximately 50 Drosophila each.
THICAPA Alleviated Aβ42-associated Locomotor Impairment in the Drosophila AD During Middle-age
Because locomotor dysfunction is a symptom of AD (29), THICAPA’s therapeutic effects on the mobility of AD Drosophila were investigated. The locomotive assay involved measuring the climbing speeds of Drosophila at 5, 10, and 15 days after eclosure (dae), which correspond to the “early,” “middle,” and “old” phases, respectively, of the untreated Actin5C-Aβ42.DMSO life span (30).
Both males (Figure 4A) and females (Figure 4B) exhibited similar climbing trends. Even though Actin5C-Aβ42.TH100 continuously outperformed Actin5C-Aβ42.DMSO in terms of average climbing speeds over the course of the 3 time points, the average climbing speed was significantly different only on 10 dae. This showed that the protective effect of THICAPA on Aβ42-related mobility loss was most effective during the middle-age phase for both males and females.
Average climbing speeds (mm/s) of AD Drosophila at 3 time-points (5 days, 10 days, and 15 days) for (A) males and (B) females. The hashtag (#) signifies p values against the wild-type control, whereas the asterisk (*) signifies p values against the untreated. p Values show significance at *p < .05, **p < .005, #p < .05, ##p < .005, and ###p < .0005. Data represents mean ± SD of 3 biological replicates with approximately 20 Drosophila each.
RNA-seq Analysis Showed Downregulated GO Terms and KEGG Pathway Pertaining to the Innate Immunity
From Supplementary Table 5, at 15 dae, untreated Actin5C-Aβ42.DMSO was at 75% (females) and 90% (males) mortality. Thus, 15 dae represented the “old” age phase for Actin5C-Aβ42.DMSO (30). Because AD is a progressive neurodegenerative disorder that affects older adults, 15 dae was chosen as the time point for RNA collection. The 2 lines of Actin5C-Aβ42.DMSO (denoted as Untreated following this) and Actin5C-Aβ42.TH100 (denoted as Treated following this) were collected in triplicate (biological replicate). A volcano plot (Figure 5A) was used to provide a general understanding of the transcriptome variations between the Untreated and Treated samples. From a total of 13 624 genes, 13 (0.10%) were significantly DEGs (padj > .05; Supplementary Table 6).
RNA-seq analysis between Untreated and Treated. (A) Volcano plot. (B) Verification of RNA-sequencing results with quantitative RT–qPCR. p Values show significance between Untreated and Treated at *p < .05, ** p < .005, and *** p < .0005. Data represents mean ± SD of 3 biological replicates with 10 Drosophila each. (C) Downregulated biological process GO terms. (D) Downregulated cellular component GO terms. (E) Downregulated molecular function GO terms. (F) Enriched KEGG pathways.
A total of 34 GO terms (Supplementary Table 7) were acquired from the downregulated DEGs. There were 24 terms under biological process (BP; Figure 5C), which included terms related to the innate immune response and catabolic processes of compounds. Two cellular components (CC) terms (Figure 5D) were also obtained that were associated with the extracellular component, while the 8 molecular function (MF) terms (Figure 5E) attained were associated with aldehyde dehydrogenase activity. Alternatively, KEGG analysis (Figure 5F; Supplementary Table 7) of the downregulated DEGs resulted in 2 pathways: the Toll and Imd signaling pathway and the 1 carbon pool by folate pathway.
To validate the RNA-seq results, RT–qPCR (Figure 5B) was used to examine the expression of 5 genes. All assessed genes displayed profiles that matched the RNA-seq, thus verifying the accuracy of the results.
Discussion
According to the etiology of AD amyloidogenesis, Aβ40 and Aβ42 peptides are the main components of the neurotoxic amyloid plaques discovered in AD patients’ brains, with Aβ42 being the most aggressively aggregating and toxic species (7). Due to the lack of successful treatments against AD, the main objective of this research was to discover novel compounds that could potentially offset the negative effects of Aβ42.
Here, the extensive RIKEN NPDepo library was used. Due to its enormity, the entire library of 22 097 compounds was prescreened to identify ligands of Aβ42 using the chemical array. THICAPA was identified to adhere to Aβ42. Because the chemical array only detects compounds that are able to bind to Aβ42 but not the compounds’ properties, the next move was to screen the compounds for their activity in inhibiting Aβ42 aggregation. ThT was used as the Aβ42 fibril aggregation marker in this test. Out of the 86 Aβ42 ligands, only the addition of 7 compounds, including THICAPA, to Aβ42 resulted in RFU readings lower than the vehicle control; hence, establishing THICAPA as a ligand of Aβ42 that suppressed Aβ42 aggregation. This also led to the theory that an Aβ42 ligand may not exhibit activities indicative of being an inhibitor of Aβ42 aggregation, despite being able to bind to Aβ42.
THICAPA, with an IUPAC name of 3-[[(3S)-1,2,3,4-Tetrahydroisoquinoline-3-carbonyl]amino]propanoic acid and synonym name of N-[(3S)-1,2,3,4-tetrahydroisoquinolin-3-ylcarbonyl]-beta-alanine has a molecular formula of C13H16N2O3 (Molecular weight of 248.28). It is a derivative of 1,2,3,4-tetrahydroisoquinoline and beta-alanine. The numerous biological functions of tetrahydroisoquinoline derivatives, including their anti-inflammatory action (31), affinity as ligands for central nervous system (CNS) receptors (32), and anticancer characteristics, have led to their widespread usage in medicinal chemistry (33). Moreover, compounds with a tetrahydroisoquinoline moiety were discovered to have neuroprotective characteristics and anti-inhibitory effects in Parkinson’s disease models, as well as inhibitory effects on inflammatory responses (34).
THICAPA was then evaluated on pheochromocytoma PC12 neuronal cells, which are extremely sensitive to Aβ-associated neurodegeneration (35,36). Supplementation of THICAPA significantly increased cell viability of Aβ42-incubated cells. This supported the hypothesis that THICAPA might confer protective properties on neural cells against Aβ42 toxicity, which may be due to the antioxidative qualities of THICAPA’s beta-alanine structure. When PC12 cells were exposed to Aβ, there was an increased amount of reactive oxygen species/reactive nitrogen species (ROS/RNS) followed by mitochondrial dysfunction and apoptosis. Damage to the mitochondria led to ATP depletion and an increase in ROS, which further triggered apoptotic cell death (37). Beta-alanine has been shown to reduce Aβ toxicity, inhibit the production of ROS, scavenge hydroxyl radicals and reactive aldehydes, and inhibit protein glycation as seen in immortalized rat brain endothelial cells and cerebrospinal fluid of AD patients (38,39). In humans, higher serum beta-alanine levels were discovered to be significantly linked with decreased risks of All-cause dementia and AD (40). Besides this, THICAPA may be promoting cell proliferation in PC12 cells; therefore, increasing total cell number, which led to increased cell viability. In addition, THICAPA might have aided in maintaining or restoring cellular energy metabolism, possibly by enhancing metabolic flux, which may result in higher ATP levels. THICAPA may also exert an indirect impact on cell protection by activating cellular pathways such as signaling cascades that defend against Aβ42-associated injury.
Next, the compound was evaluated on a whole organism. Due to its capacity to mimic many behavioral facets of AD, D. melanogaster was selected. Using the GMR-GAL4 driver, Aβ42 was expressed in the differentiating retinal tissues of Drosophila, which leads to an impaired eye development phenotype termed REP (28). This was clear in the current data, which showed that the eyes of the untreated AD Drosophila had a badly structured arrangement with punctured and merged ommatidia. Based on light micrographs, administration of 100 μM THICAPA had the most amelioration in eye phenotype with the least drastic REP when compared to eyes fed with 50 μM and 500 μM THICAPA. This was validated again by SEM whereby 100 μM THICAPA significantly reduced the severity of REP in AD Drosophila, consequently proving THICAPA’s capacity to protect eye tissues from the negative effects of Aβ42. As GMR-GAL4 drives expression to both neuronal and non-neuronal cells of the Drosophila eye, it is plausible that THICAPA is protective on non-neuronal cells as well. Following the results of augmented cell viability, THICAPA may have indirectly influenced cellular development processes that increased the differentiation of neuronal and non-neuronal cells. THICAPA could also have induced compensatory mechanisms such as cellular repair processes, inflammation regulation, or activation of alternative pathways that alleviate Aβ42’s toxic effects, leading to reduced REP.
Similarly, this rescue trend was seen in the longevity assay, which used Actin5C-GAL4 to drive ubiquitous Aβ42 expression. THICAPA was verified to be safe for Drosophila consumption as THICAPA feeding did not significantly affect the life span of the wild-type control. Conversely, the untreated AD Drosophila had shortened life span than the wild-type control, which recapitulated the AD effect in humans (41,42). This decrease in life span was lengthened when AD Drosophila was fed with THICAPA.
For the negative geotaxis assay that tested motor impairment induced by Aβ42 expression, the experimental Drosophila were monitored at 3 time points that paralleled the “young,” “middle,” and “old” ages of the untreated AD Drosophila life span. Throughout these time-points, locomotor activity gradually decreased in the treated and untreated AD Drosophila lines, with the treated AD Drosophila moving consistently faster than the untreated counterpart. Despite this, only at the “middle” age (10 dae) did the treated AD Drosophila’s average climbing speed significantly outpace the untreated AD Drosophila. Although THICAPA provided possible protection against Aβ42-associated neurodegeneration, which resulted in an overall extension of the life spans for both males and females expressing Aβ42, it did not affect muscle deterioration caused by Aβ42 during “old” age; hence, indicating that THICAPA was likely more of a preventive medication for AD symptoms rather than a treatment to AD. Moreover, THICAPA may have other broad effects on overall Drosophila health and vitality, such as boosting immune responses, enhancing metabolic balance, or hormonal signaling, which indirectly influence neuronal function and overall life span. THICAPA treatment might also activate cellular defense mechanisms, such as stress response pathways or antioxidant systems, which could aid in mitigating Aβ42 toxicity and promote cellular resilience.
To further elucidate the mechanism behind THICAPA’s protective abilities, the transcriptome of the AD Drosophila lines (untreated Actin5C-Aβ42.DMSO vs treated Actin5C-Aβ42.TH100) was analyzed. The Toll and Imd pathway that controls most of Drosophila immune-associated genes was the KEGG pathway that was enriched from the downregulated genes in THICAPA-treated samples. The Imd pathway is induced primarily by Gram-negative bacteria infection and regulates antibacterial peptide gene production, whereas the Toll pathway is primarily triggered by Gram-positive bacteria and fungi (43). A Drosophila AD model that expressed human Aβ42 in photoreceptor cells revealed that Toll pathway components mediate Aβ42-associated neurotoxicity (44). This model proved that damage from Aβ42 expression could be mitigated through the suppression of gene activities in the Toll pathway such as Dorsal and Dif. Additionally, while these immune responses serve as the body’s initial line of defense against infections, inflammation can also trigger their production (45). In fact, Aβ42 has been recognized as an antimicrobial peptide whose aggregates induce CNS inflammation (46). Interestingly, our previous study showed increased expression levels of antibacterial humoral response-associated genes as well as upregulation of antimicrobial peptides genes in the same Untreated AD Drosophila model as the current study when compared to Drosophila that lacked the Aβ42 gene (30). This correlates with the downregulation in GO terms linked to the antimicrobial humoral response in the current results, which further strengthens the notion that THICAPA could be an anti-inflammatory agent. In addition, these downregulated DEGs might be indirect consequences of THICAPA’s altercation with Aβ42 by inhibiting the aggregation of the peptide. THICAPA may also have protective effects via various molecular mechanisms, such as oxidative stress, protein misfolding, or neural plasticity.
In light of these discoveries, several research directions can be explored to complement the experiments of this work and further comprehend THICAPA’s effects on Aβ42 toxicity. Although the chemical array assay offers preliminary info on compound interactions, it does not clarify the precise mechanisms of action. To fully understand the precise binding locations and modalities of interaction between THICAPA and Aβ42, structural studies such as X-ray crystallography or molecular dynamics simulations would prove useful. Next, the cell viability assay revealed that THICAPA significantly increased viability; however, it does not directly demonstrate THICAPA’s impact on Aβ42 toxicity or amelioration against cell death. Experiments by examining mitochondrial activity or cellular markers of apoptosis would benefit in supplementing this assay. Furthermore, as the eye assay is not direct evidence of neuronal cell protection, future studies can focus on immunohistochemical staining using specific neuronal markers or in vivo imaging techniques including confocal microscopy to visualize neuronal activity in live Drosophila brains. It would also be advantageous to investigate age-dependent transcriptomic expression changes of “young” and “middle”-aged AD Drosophila and perform a comparison of all 3 time points for a more concise overview of THICAPA’s anti-Aβ42 mechanism throughout the AD Drosophila’s life span. Moreover, given the downregulation of immune response pathways discovered from the RNAseq data and validated by RT–qPCR, further investigation on THICAPA’s effects on microglial activation, neuroinflammation, and immune cell infiltration would aid in understanding the larger implications of its actions on Aβ42-related neurodegeneration.
Conclusion
This work used various research methodologies to screen compounds for their capacity to prevent the uncontrolled aggregation of Aβ42 peptides. Through extensive in vitro testing, THICAPA was discovered to be an Aβ42 ligand that suppressed fibrillary Aβ42 aggregation. Moreover, THICAPA provided amelioration against Aβ42-induced toxicity to PC12 neuronal cells. In a whole model organism, THICAPA reduced Aβ42 toxicity in the present AD Drosophila model by diminishing the REP, extending longevity, and improving motor capabilities. Additionally, transcriptomic evidence exhibited downregulation of immune defense genes due to THICAPA treatment. As of the time of publication, this study is the first to examine THICAPA’s protective effects on neurological disorders, specifically AD. Therefore, the current study recommends THICAPA as a potential treatment for AD.
Funding
This work was supported by the Ministry of Higher Education Malaysia for Transdisciplinary Research Grant Scheme (TRGS) for the project titled “Elucidating the molecular pathway of THICAPA and POET using Drosophila melanogaster Alzheimer’s disease models” (TRGS/1/2020/USM/02/3/1). F.H.P.T. was a recipient of the RIKEN International Program Associate and is a USM Post-doctoral Fellow (FPD).
Conflict of Interest
None.
Acknowledgments
We would like to thank all our collaborators and colleagues for the discussion and the work conducted in USM PPSK Lab 418 and RIKEN CSRS. We also thank RIKEN CSRS’s Emiko Sanada, Tatsuro Kawamura, and Kaori Honda for their vital support of the project.
Author Contributions
F.H.P.T.: contributed to the visualization and conceptualization of the manuscript, drafted the methodology, formal analysis, investigation, and writing (original draft, review, and editing). A.C.J.T.: contributed to the conceptualization and visualization of the manuscript, formal analysis, and writing (original draft, review, and editing). G.A.: supervised the conceptualization and research methodology, reviewed and edited the draft, as well as funded the research. N.N.: supervised the conceptualization and research methodology, reviewed and edited the draft, as well as funded the research. N.W.: supervised the conceptualization and research methodology, reviewed and edited the draft, as well as funded the research. S.S.: supervised the project and funded the research. A.Z.: supervised the project and funded the research. H.O.: supervised the project and funded the research.
