PART I THE APPLICATION AND MECHANISMS TO TREAT
III. RESULTS
1. Establishment of inducible polyQ stable cell lines
PolyQ diseases are caused by mutant proteins containing more than 30-40 glutamine repeats [39]. Expanded polyQ protein in cells tend to induce protein aggregation dots [40]. The established stable clones, Q36 and Q79, were induced with doxycycline to express the EGFP-conjugated chimeric proteins, ΔC-TBP-Q36-EGFP (Q36-EGFP) and ΔC-TBP-Q79-EGFP (Q79-ΔC-TBP-Q79-EGFP), respectively. (Figure 1A). Afterinduction, the recombinant EGFP chimera was expressed in whole cells to form
aggregates in Q79 cells as shown by fluorescence microscopy detection (Figure 1B). The lysate was collected of both Q36 and Q79 cells
following induction with Dox. In Western blot, the recombinant proteins Q36 and Q79 from cell lysatesshowed different GFP recombinant protein with different sizes as detected by anti-GFP primary antibody (Figure 1C). The results suggested that the inducible polyQ disease cell models can be established for high-content screening.
2. Autophagic enhancer screening
To identify candidate chemicals by high-content imaging, a panel of synthetic triazole compounds were screened in cell models. The
autophagy fluxby enhancing lysosomal acidification can be detected by LysoTracker staining [41, 42]. Q79 cells were treated with different triazole compounds at 20 μM for 48 h and stained with LysoTracker (Red) for detection by fluorescence high-content imaging and quantitated.
The positive control trehalose is a natural disaccharide that enhances
28
autophagy is known being able to reduce polyQ aggregation [43]. Among them, compounds OC-13, OC-14, YR-21 and YR-72 significantly
induced LysoTracker intensities (Figure 2A), while the polyQ aggregation intensities were decreased by OC-13 and YR-72 only (Figure 3A) without loss of cell numbers (Figure 3B). The results concluded that OC-13 and YR-72 are potential autophagic enhancers without affecting viabilities in Q79 stable cell line.
To further select better autophagic enhancer, the candidate drugs was further analyzed by Western blot for autophagy determination. LC3-I conversion to LC3-II is an autophagic marker [8]. Compounds OC-14, YR-21 and YR-72 increased the intensities of LC3-II in both Q36 and Q79 cell line. Only OC-13 induced LC3-II conversion in Q79 cells (Figure 4A). More analysis of Western blot showed that OC-13 is the only compound that induced LC3-II/LC3-I intensities inQ79 cells. The results demonstrated the specificity of OC-13 in inducing autophagy in cells with expanded glutamine without affecting cell viabilities.
3. More evidence of OC-13-activated autophagy.
To investigate the induced autophagic flux, both Q36 and Q79 cells were treated with different concentration of OC-13 (5, 10, 20 μM) or DMSO control for 48 h. OC-13 (Figure 5A) not affect cell viabilities (Figure 5B) and enhanced intensities of LysoTracker fluorescence (Figure 6A). The effects were time- and concentration-dependent (Figure 6B and 6C).Beclin-1 is a protein for initiation of autophagy and p62 for
29
degradation by autophagy [9, 11]. Western blot analysis showed that OC-13 increased Beclin-1 and LC3-II/LC3-I ratios and decreased of p62 in Q79 cells (Figure 7A and 7B). Baf A1 is a late phase of autophagy inhibitor by interfering degradation of p62 and LC3-II [44]. The accumulated LC3-II and p62 significantly blocked autophagic flux (Figure 8A and 8B). Formation of double-membrane autophagosome puncta was induced by OC-13 (Figure 9A and 9B). The data suggested that OC-13 induces autophagic flux in Q79 cells.
4. Amelioration of Q79-EGFP aggregates by OC-13
The autophagic enhancer OC-13 dissipate the accumulated polyQ.
The aggregation dots were decreased significantly by OC-13 as the increasing of concentrations (Figure 10A and 10B).The effective
concentration that reduced 50% of aggregation by OC-13 was determined at 14 μM (Figure 10B). Aggregation protein was accumulated at the interface of the stack gel and separation gel in Western blot analysis [45].
The lysates of OC-13-treated cells were analyzed by filter trap assay [46]
and Western blot analysis by incubating the membrane blot with GFP antibody. The insoluble Q79-EGFP aggregates were dissipated by OC-13 (Figure 11A). The levels of the soluble EGFP protein were increased at the expense of the cumulated aggregates protein in Q79 cells (Figure 11B and 11C). These data supported that polyQ aggregates can be eliminated by the autophagic enhancer OC-13.
30
5. The autophagic clearance is related to JNK signaling pathway activation
To find out the mechanism in OC-13-induced autophagy, the signaling pathways of autophagy was analyzed by Western blot.
Activation of S6K is a downstream for mTOR [47]. OC-13 increased Akt, S6K, and JNK1/2 phosphorylation in Western blot (Figure 12A). Up-regulated JNK phosphorylation induces autophagic flux [15]. Activation of Akt/mTOR pathway suppresses autophagic flux [14, 48]. The results suggested that OC-13 induces activation of JNK signaling pathway.
6. Clearance of the Q79-EGFP aggregates by JNK-mediated autophagy
The polyQ aggregation degradation co-localized with the induced autophagosome. It is necessary to understand the mechanism for
degradation. SP600125 is an inhibitor for repressing JNK
phosphorylation [49].3-MA is an early autophagy inhibitor [50]. Insulin activates Akt phosphorylation [51]. MG-132 is a proteasome inhibitor [52]. Pretreatment of JNK inhibitor SP600125, and autophagy inhibitors Baf A1 and 3-MA impaired clearance of the aggregates, while insulin and MG-132 did not as shown in fluorescence microscopy (Figure 13A and B). On the other hand, clearance of insoluble polyQaggregation were repressed by SP600125, Baf A1 and 3-MA in Western blot analysis (Figure 14A and B). Both SP600125 and 3-MA inhibited drug-mediated autophagosome formation and Baf A1 blocked degradation of
autophagosome (Figure 15A and B). These finding suggested that OC-13
31
induces autophagic clearance of the Q79-EGFP aggregates that involved JNK signaling pathway.
7. Exclusion of nucleus polyQ aggregation
The nuclei aggregation can be eliminated as detected by confocal microscopy. More experiments were used to determine how nucleus aggregates were eliminated by drug-induced autophagy. The aggregates were decreased by OC-13 (20 μM) in confocal microscopy (Figure 16A).
The number of aggregations in nucleus with EGFP positive cells were counted after 48 h treatment. OC-13 eliminated aggregates more efficient in nucleus than in cytoplasm (Figure 16B). In Western blot analysis, the insoluble polyQ aggregation in nucleus was decreased more than
cytoplasm (Figure 16C and D). The previous study showed that nucleus polyQ aggregation is degraded by proteasome [53]. Pretreatment of SP600125, Baf A1 and 3-MA impaired clearance of the nucleus aggregates in Q79 cell. But the proteasome inhibitor MG-132 did not repress OC-13-mediated clearance of nucleus aggregates (Figure 17A). In Western blot, degradation of nucleus insoluble polyQ aggregation was repressed by 3-MA and SP600125. Taken all together, OC-13 eliminates aggregation in nucleus and cytoplasm by JNK-mediated autophagy.
8. Transient Httex1 polyQ transfection
To further understand if OC-13 eliminates HD aggregates, plasmids encoding of Htt exon1 including 25 or 97 CAG repeats connected with GFP were obtained to establish Httex1 polyQ cell models (Figure 18A).
32
Human neuroblastoma cells SK-N-SH were transfected with Httex-Q25 and -Q97, respectively, for 24 h and the chimeric proteins were found expressed. Httex1-Q97 transfected cells exhibited accumulated more aggregation compared to Httex1-Q25 transfected cells as indicated in fluorescence microscopy (Figure 18 B). The data indicated that the established cell models expressing the chimeric proteins Httex1 exon1 Q25 or Q97-GFP protein can be established with polyQ accumulation in
Httex1-Q97 transfected cells that can be used to test the efficiency of
OC-13.9. OC-13-induced autophagy in cells transfected with Httex1-Q97
Previous data showed that OC-13 induced autophagic flux in ΔC-TBP-Q79 stable cell line and autophagic flux by activation of lysosome.To identify GFP-positive cells, both Httex1-Q25 and Httex1-Q97
transfected cells were analyzed by flow cytometry (Figure 19A). OC-13 induced Httex1-Q97 transfected cells LysoTracker intensity (Figure 19B and 19C). Compared with DMSO control, the lysosome fluorescence was increased by OC-13 after 12 h. The images shown enhanced Httex1-Q97 transfected cells LysoTracker intensities as observed by fluorescence microscopy in cells transfected with the Httex1 construct (Figure 20A) and can be quantitated (Figure 20B). The results suggested that OC-13 activates lysosome in cells with expanded polyglutamine.
Next, whether lysosome activator OC-13 assisted in autophagosome formation in Httex1-Q97 transfected cells. The confocal images revealed
33
that OC-13 increased autophagosome puncta in cells expressing Httex1-Q97 (Figure 21A). OC-13 decreased p62 (Figure 22A) and induced LC3-I to LC3-LC3-ILC3-I conversion (Figure 22B) in Httex1-Q97 transfected cells. PLC3-I is an fluorescent dye to label dead cells [54]. To gate GFP-positive cells,
Httex1-Q25 or Httex1-Q97 transfected cells were stained with PI and
analyzed by flow cytometry (Figure 23A). PI positive cells remained not increase in Httex1-Q25 or Httex1-Q97 transfected cells (Figure 23B), indicating OC-13 did not affect cell viability in Httex1 transfected cells (Figure 23C). This results suggested that OC-13 induced autophagosome formation and activated lysosome in HD cell model without affecting cell viabilities.10. Phosphorylation of JNK and Akt are increased by OC-13 in Httex1-Q97 transfected cells
OC-13-induced autophagy in Q79 cells is activated by JNK.
Whether, the autophagic signaling pathway can be activated in HD cell model needs to be addressed. OC-13 induced phosphorylated JNK, Akt and S6K ratios in HD disease cell model (Figure 24). The results showed that JNK-mediated autophagy pathway, but not Akt/mTOR-dependent pathway was activated by OC-13.
11. OC-13 decreased Httex1-Q97 aggregations
To study the clearance of polyQ aggregation protein, more experiments tested the hypothesis if the autophagy enhancer OC-13 decreased polyQ aggregation protein. By counting fluorescent
34
aggregation in 500 Httex1-Q97 expressing cells, the aggregates were eliminated (Figure 25A) by increasing OC-13 concentrations (Figure 25B). OC-13 decreased the insoluble polyQ accumulation at the interface between the running and stacking gels [17, 55] as detected by GFP
antibody (Figure 26A). The disappearance of the insoluble polyQ aggregation was further asserted by filter retardation assay [56]. The aggregated dots intensities were reduced by OC-13 (Figure 26B) and the difference to eliminate insoluble Httex1-Q97 aggregation was significant (Figure 26C).
12. Elimination of Httex1-Q97 aggregation by OC-13-mediated autophagy
Since the autophagy inhibitor 3-MA inhibits PI3KCIII activation and downregulates autophagy [50], cells were pretreated with 3-MA for 1 h before OC-13 treatment. By counting the numbers of aggregates in 500 GFP positive cells, 3-MA significantly repressed OC-13-mediated
clearance of fluorescent Httex1-Q97 aggregates (Figure 27A and 27B). In Western blot, the eliminated Httex1-Q97 aggregation was significantly reverted by 3-MA treatment (Figure 28A) as proved by densitometric analysis (Figure 28B). The results suggested that OC-13-mediated autophagy is involved to eliminate Httex1-Q97 aggregates in cells transfected with Httex1-Q97 construct.
13. Elimination of aggregates by JNK-mediated autophagy
Todetermine the mechanism for OC-13 for aggregation elimination,
35
it is necessary to find out if JNK pathway mediated autophagy. The drug-induced elimination of Httex1-Q97 aggregates were repressed by
autophagic inhibitor and JNK inhibitor (Figure 29A). By counting of aggregates in 500 GFP positive cells, 3-MA and Baf A1, and SP60025 blocked the drug-induced polyQ degradation (Figure 29B) and insulin and proteasome inhibitor MG-132 did not repress degradation of aggregates by OC-13 (Figure 29B). Both inhibitors for autophagy and JNK pathway blocked OC-13-induced LC3 puncta formation (Figure 30A). The results showed that Httex1-polyQ aggregations were decreased by JNK-mediated autophagy.