Oligomeric α-synuclein-specific degradation by HtrA2/Omi to bestow a neuroprotective function

Although the malfunction of HtrA2/Omi leads to Parkinson’s disease (PD), the underlying mechanism has remained unknown. Here, we showed that HtrA2/Omi specifically removed oligomeric α-Syn but not monomeric α-Syn to protect oligomeric α-Syn-induced neurodegeneration. Experiments using mnd2 mice indicated that HtrA2/Omi degraded oligomeric α-Syn specifically without affecting monomers. Transgenic Drosophila melanogaster experiments of the co-expression α-Syn and HtrA2/Omi and expression of genes individually also confirmed that pan-neuronal expression of HtrA2/Omi completely rescued Parkinsonism in the α-Syn-induced PD Drosophila model by specifically removing oligomeric α-Syn. HtrA2/Omi maintained the health and integrity of the brain and extended the life span of transgenic flies. Because HtrA2/Omi specifically degraded oligomeric α-Syn, co-expression of HtrA2/Omi and α-Syn in Drosophila eye maintained a healthy retina, while the expression of α-Syn induced retinal degeneration. This work showed that the bacterial function of HtrA to degrade toxic misfolded proteins is evolutionarily conserved in mammalian brains as HtrA2/Omi.

). 148 The transgenic hOmi Drosophila line was heterozygous for the dominantly marked CyO balancer 149 chromosome carrying a dominant mutation, CyO, which causes curly wings for easy detection.   Figure S5). Female α-Syn Drosophila were mated with male hOmi 156 Drosophila, and +/hOmi; α-Syn/+ flies were selected based on the dominant phenotypes of the Parkinsonism (Fig. 3). As shown in Fig. 3a, loss of climbing ability of α-Syn Drosophila was 166 completely rescued by co-expression with hOmi. The performance index of locomotion was 167 significantly lower in α-Syn Drosophila than hOmi/α-Syn Drosophila or hOmi Drosophila of the 168 same age. As previously reported, the survival rate of α-Syn Drosophila was significantly reduced 169 as a phenotype of Drosophila Parkinsonism. However, the survival rate of hOmi/α-Syn 170 Drosophila was increased as much as wild type, indicating that hOmi/HtrA2 completely rescued 171 the Drosophila Parkinsonism induced by α-Syn (Fig. 3b). Kaplan-Meier Survival analyses also 172 showed that hOmi/HtrA2 completely rescued the Drosophila Parkinsonism (Supplementary 173 Information, Figure S6). Overall, pan-neuronal co-expression of hOmi with α-Syn completely 174 rescued the Parkinsonism phenotypes of α-Syn Drosophila. Additionally, it is interesting to note 175 that pan-neuronal sole expression of hOmi (hOmi Drosophila) resulted in better performance in 176 the locomotor reaction and increased survival rate compared with the wild type control. 177 Human HtrA2/Omi rescued the α-Syn-Induced neurotoxicity in a Drosophila model of 178 Parkinson's disease by oligomeric α-Syn-specific degradation 179 To investigate how hOmi rescues α-Syn-induced neurotoxicity in Drosophila, histological 180 examinations were performed using the hOmi/α-Syn Drosophila line along with wild-type, hOmi 181 and α-Syn Drosophila lines. Immunohistochemical confocal microscopy using an oligomer-182 specific monoclonal antibody, anti-α-Syn (ASy05), on brain sections showed that co-expression 183 of hOmi and α-Syn completely eliminated the oligomeric α-Syn (Fig. 4a), which is consistent 184 with the in vitro and mouse experiments ( Fig. 1 and 2). Quantification of the green fluorescent 185 intensity in the flies definitively revealed a large quantity of oligomeric α-Syn accumulation only 186 in α-Syn Drosophila (Fig. 4b). However, there were no detectable α-Syn oligomers in hOmi/α-187 Syn Drosophila. We further confirmed the specific degradation of α-Syn oligomers by hOmi with total protein extracts of hOmi/α-Syn fly brains (Fig. 4c). Anti-α-Syn antibody detected both 189 oligomeric and monomeric α-Syn in α-Syn Drosophila. However, only monomeric α-Syn was 190 detected by western blotting of hOmi/α-Syn Drosophila. This in vivo result clearly confirmed that 191 HtrA2/Omi specifically recognized and degraded oligomeric α-Syn without affecting monomeric 192 α-Syn. Considering that oligomeric α-Syn has strong neurotoxicity to function as an etiological 193 agent for PD while monomeric α-Syn lacks neurotoxicity, rather playing an essential role in 194 maintaining a supply of synaptic vesicles in presynaptic terminals 32, 33 , this result shed light on 195 how hOmi provides a neuroprotective function in PD. 196 We further investigated Drosophila brains after immunostaining with anti-α-Syn and anti-197 HtrA2 antibodies. An age-dependent accumulation of α-Syn clearly caused the accumulation of 198 Lewy bodies in α-Syn Drosophila, whereas co-expression of hOmi with α-Syn in hOmi/α-Syn 199 Drosophila completely prevented the accumulation of Lewy bodies, and the overall integrity of 200 the brain tissue was the same as the normal control (Fig. 5a, b). H&E staining of the brain slices 201 of the flies further confirmed a clear neurodegeneration in the α-Syn Drosophila (Fig. 6a, b). The 202 integrity of the brain tissue was observed in both young and aged hOmi/α-Syn Drosophila as in 203 the case of control flies and hOmi Drosophila, but not in aged α-Syn Drosophila due to 204 neurodegeneration. The brains of 40-day-old α-Syn Drosophila showed clear neuronal loss with 205 astrocytosis and the appearance of Lewy bodies both in male (Fig. 6a) and female flies (Fig. 6b).

206
In accordance with the previous results concerning the function of hOmi in oligomeric α-Syn-207 specific degradation, this result again confirmed that hOmi rescued the α-Syn-induced 208 neurotoxicity in α-Syn Drosophila. 209 Human HtrA2/Omi counteracted the α-Syn-induced developmental defect in Drosophila eye It has been previously observed that the expression of α-Syn in the developing eye causes retinal 211 degeneration in Drosophila 34 . Since α-Syn-induced retinal degeneration well-represented α-Syn-212 induced neurotoxicity, we crossed the GMR-GAL4 driver line with the α-Syn, hOmi and hOmi/α-  Additionally, the expression of α-Syn led to developmental defects of the eyes, showing a loss 221 of general retinal tissue integrity and roughness of the eye (Fig. 7). Serious eye defects were 222 observed in both male and female α-Syn Drosophila, and the eye defects became more serious as 223 the flies aged (bottom panel of Fig. 7a, b). In contrast, hOmi/α-Syn Drosophila did not show any 224 eye defects and the eye phenotype was equivalent to the normal control (GMR-GAL4) and hOmi 225 Drosophila (Fig. 7a, b). Scanning electron microscopy also revealed serious defects of the α-Syn 226 Drosophila eye and the normal undamaged eye of hOmi/α-Syn Drosophila ( Fig. 7c and 7d).

227
Ommatidial disarray was significantly increased in α-Syn Drosophila compared with hOmi/α-228 Syn Drosophila, hOmi Drosophila or the normal control (GMR-GAL4), and the difference 229 became more evident as the flies increased in age (bottom panel of Fig. 7c, d). Furthermore, the    HtrA that specifically degrades misfolded proteins through its protease activity 45 . The bacterial homolog of HtrA, HtrA2/Omi, functions as a protease to specifically degrade a type of misfolded 299 protein, i.e., oligomeric α-Syn. Considering that the original function of HtrA was to degrade 300 misfolded protein through its protease activity, it is very interesting to note that the function of the 301 mammalian version of HtrA, HtrA2/Omi, is to remove oligomeric α-Syn through its protease 302 activity. Because oligomeric α-Syn is the misfolded version of native α-Syn, the original function 303 of HtrA seems to be perfectly conserved in mammals.

305
Reagents and antibodies 306 The reagents used in all experiments and antibodies used for western blot or immunohistochemical 307 analysis are listed in the Supplementary Information Table S1.  The protein samples were mixed with NuPAGE 4 × LDS sample buffer (Invitrogen, cat# NP007), 402 heated at 95℃ for 5 min and run on a 4-12% Bis-Tris gradient gel (Invitrogen, cat# NP#00322).

403
The proteins were visualized by staining with Coomassie blue.

404
For western blotting, the SDA-PAGE gels were transferred to PVDF membranes (Thermo  Four-week-old mnd2/mnd2 mice, mnd2/+ mice and aged match controls were sacrificed, and 550 the brain tissues were chopped into small pieces, followed by the addition of 1 mL of total protein 551 extraction kit TM buffer containing 0.1% SDS (Millipore, cat# 2140) to 0.2 g of brain samples.