DWE, SalA and SalB reduced aggregation of Aβ42 in vitro
To verify whether DWE and its components SalA and SalB influence the aggregation of Aβ42, they were first subjected to a primary screening via Thioflavin T (ThT). ThT binds to fibrillated aggregates whereby the dye experiences a characteristic red shift of its emission spectrum (Groenning, 2010). At specific time points, the fluorescence generated by ThT binding to amyloid fibrils were then measured and quantified. Decreasing fluorescence of ThT signified the test compound’s ability to inhibit Aβ42 aggregation. Morin was added as a positive control in the assay (Kapoor & Kakkar, 2012).
At the time point of 3600 seconds, Morin reduced the RFU readings by 71.2% compared to DMSO. At the same time point, addition of DWE to Aβ42 peptides decreased RFU readings by 36.9% while Aβ42 peptides incubated with SalA and SalB experienced 65.9% and 50.8% decrement respectively in RFU readings when compared to the DMSO (Figure 1). This showed that DWE, SalA and SalB affected the aggregation of Aβ42 by reducing the fibrillation.
DWE, SalA and SalB protected PC12 cells from Aβ42-induced cell death
To test the efficacy of DWE, SalA and SalB in cells, we tested the compounds using PC12 rat pheochromocytoma cells. The presence of Aβ42 peptides decreased PC12 cell viability to 40% compared to the unexposed control (p=0.0022). However, this toxic effect was rescued with the supplementation of DWE and its components, SalA and SalB (Figure 2). The best rescue effect was exhibited by incubation with SalB in which cell viability was increased to 95.4% cell viability based on the control when compared to the Aβ42-incubated PC12 cells supplemented with vehicle control (p=0.004). This was followed by SalB with a cell viability of 84% (p=0.0093) and DWE with cell viability of 73.5% (p=0.011). This demonstrated the ability of DWE, SalA and SalB to protect neuronal cells from the neurotoxicity effect of Aβ42.
SalA and SalB were detected in the brains and bodies of Drosophila after DWE feeding
In order to test the effect of these compounds on a whole organism, we chose Drosophila melanogaster as our model organism. As SalA and SalB are the two most abundant components in DWE (Ai & Li, 1988; Lian-niang et al., 1984), their presence in various parts of DWE-fed Drosophila were analysed using UPLC. The retention time of SalA and SalB were 21.9 minutes (Figure 3Ai) and 20.7 minutes (Figure 3Aii) respectively. By comparing the retention time of both reference standards, DWE with a concentration of 10mg/ml were found to have 43.0 μg/ml of SalA and 571 μg/ml of SalB.
Subsequently, one hundred Drosophila were fed with DWE and the heads, bodies and faeces were harvested and analysed through UPLC. For SalA, 0.85 μg/ml was detected in the heads, 4.3 μg/ml in the bodies and 37.3 μg/ml in the faeces. Likewise, the concentration of SalB in DWE-fed Drosophila heads, bodies and faeces were determined to be 2.6 μg/ml, 12.9 μg/ml and 407.1 μg/ml, respectively. The data confirmed the presence of SalA and SalB in the heads and bodies of Drosophila after feeding of DWE.
Effect of DWE, SalA and SalB on Drosophila melanogaster AD model
The rough eye phenotype (REP) screening system was used to observe the effects of DWE, SalA and SalB on Drosophila, the rough eye phenotype (REP) (Kumar, 2012). The Drosophila eye is a model system to understand developmental neurobiology due to its simple neuroectoderm structure consisting of photoreceptors and accessory cells. Each eye comprised 800 hexagonal-shaped components known as ommatidium, organized in a crystalline array akin to the honeycomb cells of a beehive. These ommatidia are positioned in columns across the eye resulting in a concave “egg-like” formation (Figures 5A and 5A’). The mechano-sensory bristles extending at alternating vertices of each ommatidium that are directed at precise angles give an additional sensory field (R. Cagan, 2009; R. L. Cagan & Ready, 1989; Kumar, 2012). As each ommatidium contains seven photoreceptor nerve cells, any distortion in the eye morphology could be attributed to abnormalities in the neurons (Basler et al., 1991; Tomlinson et al., 1987).
By using light micrographs, a strongly observable REP was seen when human Aβ42 was ectopically expressed in the Drosophila eyes via the pan-retinal GMR-GAL4 driver (Figures 3B and 3B’) (Finelli et al., 2004). When compared to the control GMR-OreR.DMSO (Figures 4A and 4A’), the GMR-Aβ42.DMSO adult eyes were severely malformed with merged ommatidia (Figures 4B and 4B’)
The supplementation of transgenic Drosophila with different concentrations of DWE resulted in the partial rescue of eye deformation compared to that of GMR-Aβ42.DMSO albeit at varying degrees (Figures 4C-E). As the dosage of DWE increased, the rescue effect on the eyes of the Aβ42-expressing Drosophila fed with 10 mg/mL (Figure 4D) was more apparent showing similarity to those of the control GMR-OreR.DMSO (Figure 4A). However, the dosage of 20 mg/mL did not give an observable improvement on the eye (Figures 4E). Similarly for both SalA (Figure 4F-3H) and SalB (Figure 4I-3K) treatments, there was a recovery in eye morphology when transgenic Drosophila were cultured in 100 μM of compounds compared to 50 μM with 100 μM fed eyes having less fused ommatidia (indicated by dotted circles) and 500 μM feeding did not improve rectification of the eyes further. Thus, the optimum concentration for amelioration of the REP in GMR-Aβ42 transgenic flies for DWE, SalA and SalB are 10 mg/mL, 100 μM and 100 μM, respectively.
Scanning electron microscopy (SEM) was used to analyse the phenotype at higher magnification (Figure 5). GMR-Aβ42.DMSO eyes (Figure 5B and 5B’) showed merged and bulged ommatidia that had perforated holes which gave the eye a “glazed” exterior compared to GMR-OreR.DMSO’s (Figure 5A) “egg-like” shaped eye. In addition, the overall inter-ommatidial bristles were fewer than the Control and exhibited distorted polarities.
The extent of rescue effect on Aβ42’s toxicity was assessed using the Flynotyper software (Iyer et al., 2018, 2016). The quantification of morphological disfigurements in the Drosophila eye was evaluated in the form of the phenotypic score (P-score). The higher the P-score, the more distorted that specific eye is, and thus the more severe the REP is (Figure 5F). Parallel to SEM images (Figures 5A-G and 5A’-G’), the P-score for GMR-Aβ42.DMSO was significantly higher (p=0.0088) than GMR-OreR.DMSO, demonstrating Aβ42’s adverse effects on Drosophila eyes when not supplemented with any extra extracts or compounds. Feeding of DWE, SalA and SalB reduced the P-score on the eyes with GMR-Aβ42.DWE having the lowest P-score among compound-fed Drosophila compared to GMR-Aβ42.DMSO (p=0.0028), followed by SalB (p=0.0015) and SalA (p=0.011). These results implied that DWE and its components SalA and SalB showed amelioration towards Aβ42-induced neurodegeneration in Drosophila in a dosage dependent manner, with 10 mg/mL DWE being the optimum dosage.
DWE, SalA and SalB extended the lifespan of the Drosophila AD model
The effect of prolonged exposure to 10 mg/mL DWE, 100 µM SalA and 100 µM SalB and their influence on longevity were investigated. In order to do this, the Actin5C-GAL4 driver which drives ubiquitous expression in muscle tissue was employed. There was no significant difference when Actin5C-OreR Drosophila was fed with DWE, SalA and SalB versus the same line cultured with only vehicle control (0.5% DMSO) for both males (Table 2, Figure 6A) and females (Table 3, Figure 6C). This indicated that DWE and its compounds did not exert any ill effects on the lifespan of control Drosophila. On the other hand, Actin5C-Aβ42.DMSO Drosophila was significantly different from Actin5C-OreR.DMSO with average median lifespans of male and female dropping by 42.8% and 47.6% respectively. Moreover, Actin5C-Aβ42 Drosophila lines with DWE, SalA and SalB consumptions were significantly different compared to Actin5C-Aβ42.DMSO. The average median lifespans for male Aβ42-expressing Drosophila increased by 75%, 37.5% and 37.5% after feeding of 10 mg/mL DWE, 100 µM SalA and 100 µM SalB respectively when compared to vehicle control-fed Actin5C-Aβ42.DMSO (Table 2, Figure 6B). Likewise, Actin5C-Aβ42.DMSO females experienced 63.6%, 36.7% and 72.7% extension in average median lifespan after being cultured in 10 mg/mL DWE, 100 µM SalA and 100 µM SalB respectively (Table 2, Figure 6D). To investigate the magnitude of rescue effect of the compounds, comparison of the ratio of Actin5C-Aβ42.compound:Actin5C-Aβ42.DMSO was made. As the same Actin5C-Aβ42.DMSO triplicate line was utilized to compare all of the compound-fed fly lines, the length of lifespan for Actin5C-Aβ42.DMSO remained static. The determining element depended on the lifespan length of compound-fed lines. Longer lifespans of compound-fed fly lines resulted in higher ratio values which indicated the compounds had a higher potency in prolonging lifespan. Based on the males, the ratios of Actin5C-Aβ42.DWE:Actin5C-Aβ42.DMSO for the restricted mean and average maximum lifespan were 1.75 and 1.74, respectively. For SalA, the ratios were 1.39 and 1.84, respectively while SalB males had ratios of 1.51 and 2.00, respectively. On the other hand, DWE females exhibited ratios of 1.64 and 1.57, respectively. Females of SalA had ratios of 1.40 and 1.48, respectively while SalB females showed ratios of 1.73 and 1.90. Here, we showed that DWE, SalA and SalB were able to prolong Actin5C-Aβ42 Drosophila’s severely shortened lifespan. When compared within sexes, DWE was the most effective in lengthening the lifespan of male Actin5C-Aβ42 Drosophila, followed by SalB and SalA. Alternatively, female Actin5C-Aβ42 Drosophila benefit most from SalB consumption followed by DWE and finally SalA.
Table 2: Mean and median of male experimented Drosophila melanogaster lines fed with or without DWE, SalA or SalB
Name
|
No. of subjects
|
Restricted mean
|
|
Age in days at % mortality
|
Days
|
Std. error
|
95% C.I.
|
|
25%
|
50%
|
75%
|
90%
|
100%
|
95% Median C.I.
|
|
Actin5C-OreR.DMSO
|
156
|
17.14
|
0.76
|
15.65 ~ 18.63
|
|
10
|
14
|
24
|
31
|
41
|
13.0 ~ 15.0
|
Actin5C-OreR.DWE
|
182
|
17.77
|
0.73
|
16.35 ~ 19.19
|
|
10
|
15
|
26
|
33
|
41
|
13.0 ~ 16.0
|
Actin5C-OreR.SalA
|
156
|
17.14
|
0.76
|
15.65 ~ 18.63
|
|
10
|
14
|
24
|
31
|
41
|
13.0 ~ 15.0
|
Actin5C-OreR.SalB
|
154
|
16.26
|
0.79
|
14.72 ~ 17.80
|
|
8
|
13
|
23
|
32
|
42
|
12.0 ~ 15.0
|
Actin5C- Aβ42.DMSO
|
161
|
8.81
|
0.31
|
8.20 ~ 9.42
|
|
6
|
8
|
11
|
15
|
19
|
8.0 ~ 8.0
|
Actin5C- Aβ42.DWE
|
168
|
15.50
|
0.60
|
14.32 ~ 16.68
|
|
9
|
14
|
21
|
28
|
33
|
13.0 ~ 15.0
|
Actin5C- Aβ42.SalA
|
163
|
12.21
|
0.52
|
11.20 ~ 13.22
|
|
7
|
11
|
16
|
22
|
35
|
10.0 ~ 11.0
|
Actin5C- Aβ42.SalB
|
199
|
13.33
|
0.55
|
12.26 ~ 14.40
|
|
7
|
11
|
19
|
25
|
38
|
10.0 ~ 12.0
|
Table 3: Mean and median of female experimented Drosophila melanogaster lines fed with or without DWE, SalA or SalB
Name
|
No. of subjects
|
Restricted mean
|
|
Age in days at % mortality
|
Days
|
Std. error
|
95% C.I.
|
|
25%
|
50%
|
75%
|
90%
|
100%
|
95% Median C.I.
|
|
Actin5C-OreR.DMSO
|
194
|
21.01
|
0.68
|
19.66 ~ 22.35
|
|
13
|
21
|
29
|
34
|
41
|
18.0 ~ 22.0
|
Actin5C-OreR.DWE
|
198
|
21.11
|
0.75
|
19.65 ~ 22.57
|
|
12
|
20
|
30
|
36
|
41
|
17.0 ~ 22.0
|
Actin5C-OreR.SalA
|
154
|
19.25
|
0.89
|
17.50 ~ 20.99
|
|
9
|
17
|
28
|
36
|
44
|
15.0 ~ 19.0
|
Actin5C-OreR.SalB
|
159
|
19.65
|
0.91
|
17.87 ~ 21.44
|
|
11
|
17
|
29
|
38
|
45
|
15.0 ~ 18.0
|
Actin5C- Aβ42.DMSO
|
153
|
11.48
|
0.39
|
10.71 ~ 12.24
|
|
8
|
11
|
15
|
19
|
23
|
10.0 ~ 11.0
|
Actin5C- Aβ42.DWE
|
171
|
18.78
|
0.62
|
17.56 ~ 20.00
|
|
12
|
18
|
25
|
30
|
36
|
16.0 ~ 19.0
|
Actin5C- Aβ42.SalA
|
159
|
16.10
|
0.65
|
14.83 ~ 17.37
|
|
10
|
15
|
22
|
28
|
34
|
14.0 ~ 16.0
|
Actin5C- Aβ42.SalB
|
154
|
19.87
|
0.74
|
18.42 ~ 21.32
|
|
13
|
19
|
26
|
32
|
44
|
18.0 ~ 20.0
|
Increased average climbing speed (mm/s) in compound-fed AD Drosophila
In Drosophila, accumulation of Aβ42 peptides contributes to locomotor dysfunction (Iijima et al., 2004). Hence, the possibility of these compounds having significant effects on the mobility of the AD Drosophila was investigated. In this experiment, Actin5C-Aβ42 Drosophila were cultured with and without compounds (DWE, SalA and SalB) and the negative geotaxis assay was performed. This assay recorded the average climbing speed (mm/s) of the flies.
For male AD Drosophila fed with and without DWE (Figure 7A), there was a significant difference in the average climbing speed between Actin5C-Aβ42.DWE and Actin5C-Aβ42.DMSO. Actin5C-Aβ42.DWE had a higher average climbing speed than Actin5C-Aβ42.DMSO at all three time points. Similar to the males, there was a significant difference in the average climbing speed of female Actin5C-Aβ42.DWE and Actin5C-Aβ42.DMSO (Figure 7B). Actin5C-Aβ42.DWE showed higher average climbing speed than Actin5C-Aβ42.DMSO at all three time points.
For male AD Drosophila fed with and without SalA (Figure 7C), the male Actin5C-Aβ42.SalA showed significantly higher average climbing speed compared to Actin5C-Aβ42.DMSO at all three timepoints. Contrary to the male Actin5C-Aβ42.SalA (Figure 7D), female Actin5C-Aβ42.SalA showed no significant difference in average climbing speed between Actin5C-Aβ42.SalA and Actin5C-Aβ42.DMSO for all three time points.
For male AD Drosophila fed with and without SalB (Figure 7E), there was a significant difference between the Actin5C-Aβ42.SalB and Actin5C-Aβ42.DMSO for all three time points. Female Actin5C-Aβ42.SalB (Figure 7F) was also shown to have significant difference in average climbing speed when compared to Actin5C-Aβ42.DMSO but only on the 10th and 15th day.
To study the extent of rescue effect of the compounds, we compared the ratio between the average speed of AD Drosophila fed with and without the compounds, whereby a higher ratio indicates a better effect on the mobility. It was suggested that SalB had the best effect for male AD Drosophila. For the female counterpart, DWE had the best effect on all three time points, suggesting it to be the best compound even at an early stage.