Behavioural and dopaminergic changes in double mutated human A30P*A53T alpha-synuclein transgenic mouse model of Parkinson´s disease

Alpha-synuclein (aSyn) is the main component of Lewy bodies, the histopathological marker in Parkinson’s disease (PD), and point mutations and multiplications of the aSyn coding SNCA gene correlate with early onset PD. Therefore, various transgenic mouse models overexpressing native or point-mutated aSyn have been developed. Although these models show highly increased aSyn expression they rarely capture dopaminergic cell loss and show a behavioural phenotype only at old age, whereas SNCA mutations are risk factors for PD with earlier onset. The aim of our study was to re-characterize a transgenic mouse strain carrying both A30P and A53T mutated human aSyn. Our study revealed decreased locomotor activity for homozygous transgenic mice starting from 3 months of age which was different from previous studies with this mouse strain that had behavioural deficits starting only after 7–9 months. Additionally, we found a decreased amphetamine response in locomotor activity and decreased extracellular dopaminergic markers in the striatum and substantia nigra with significantly elevated levels of aSyn oligomers. In conclusion, homozygous transgenic A30P*A53T aSyn mice capture several phenotypes of PD with early onset and could be a useful tool for aSyn studies.

proteins and aSyn aggregation redistributes them leading to decreased DA release in a mouse line expressing truncated aSyn 17,18 .
Transgenic (tg) mice overexpressing human aSyn have been a common tool to study aSyn toxicity and aSyn targeting drug therapies. Several mouse lines overexpressing wildtype (wt) aSyn or mutated forms of aSyn have been developed but these models mainly lack DAergic neuronal cell loss despite excessive aSyn expression (reviewed in 19 ). Richfield et al. 20 introduced a mouse model expressing double mutated human aSyn gene with both A30P and A53T point mutations under the rat tyrosine hydroxylase (TH) promoter (C57BL/6J-Tg (TH-SNCA*A30P*A53T)39Eric/J) that combined two well-characterized familial site mutations of SNCA to model PD. These tg mice expressed human aSyn in cell bodies, axons, and terminals in the nigrostriatal pathway, and had decreased locomotor activity at age 7-9 and 13-23 months, and lowered concentration of DA and its metabolites in the striatal tissue at age 16-18 months. Similar to several other aSyn tg mice, younger mice did not have significant changes in locomotor activity or in striatal DA concentration. However, both A30P and A53T familial point mutations in SNCA are a risk factor for early onset PD 19 but these features were not captured in the earlier study. Therefore, the aim of this study was to breed a homozygous A30P*A53T aSyn tg mouse strain, and characterize if this animal model would capture the phenotype of early-onset PD. We designed PCR oligonucleotides and a new genotyping protocol to distinguish between wt, heterozygous, and homozygous animals in order to characterize behavioural and DAergic changes in homozygous A30P*A53T aSyn tg mice. Interestingly, we found several behavioural and histological changes that were not described in the original publication.
During the 22-hour locomotor activity monitoring, differences were found between tg and wt mice in total distance travelled, vertical count, jump count, and average speed ( Fig. 2A-D). Total distance travelled was significantly lower in 9 ( Fig  activity was altered between aSyn transgenic (tg) and wild type (wt) littermates in all age groups. At 3 months, significant changes between tg and wt animals were not seen (A). 6, 9, and 12 months old tg mice were more active compared to wt littermates right after the lights were turned off but less active than wt mice between 22:00 and 1:00 (B-D). 18 months old tg mice were more active during the dark time compared to wt littermates (E). When comparing locomotor activity during the first hour of the locomotor experiment, tg mice showed increased activity only at the age of 12 months (F). Data are expressed as mean ± SEM, n = 7-23. Repeated measures 2-way-ANOVA, (A-E); Student's t-test (F); *p < 0.05, **p < 0.005, ***p < 0.001.  Comparison of total distance travelled, vertical counts, jump count, and average speed during 22 h locomotor activity measurements between aSyn transgenic mice (tg) and wild type littermates (wt). Total distance travelled was decreased in 9 months and increased in 18 months old tg mice compared to wt mice (A). 3 and 6 months old tg mice showed significantly less vertical counts compared to littermates (B). 12 old months tg mice had more vertical counts compared to wt mice (B). Jump counts were significantly lower in 3, 6, and 9 months old tg mice compared to wt littermates (C). Average speed of tg mice was significantly lowered in all age groups (D). Data are expressed as mean ± SEM, n = 7-23. Student's t-test, *p < 0.05, **p < 0.005, ***p < 0.001, ****p < 0.0001. (2019) 9:17382 | https://doi.org/10.1038/s41598-019-54034-z www.nature.com/scientificreports www.nature.com/scientificreports/ t-test) and SNpc (Fig. 6C, 12 months: t = 2.499, p = 0.019; 18 months: t = 3.175, p = 0.0067, Student's t-test) compared to wt littermates. Accumulation of aSyn oligomers was identified by immunohistochemistry (IHC) in the SNpc and striatum. Both 12 and 18 months old tg mice had significantly increased immunoreactivity for aSyn oligomers in striatum (immunostained by aSynO5 antibody; Fig. 6A,D, 12 months: t = 6.36, p < 0.0001; 18 months: t = 7.716, p < 0.0001) and in SNpc (immunostained by aSynO5 antibody; Fig. 6A,E, 12 months: t = 8.084, p < 0.0001; 18 months: t = 11.53, p < 0.0001). OD analysis for total aSyn in the striatum revealed no significant differences in immunoreactivities in 12 and 18 months old tg mice compared to wt mice (see Supplementary Fig. S1).

Figure 3.
Amphetamine-induced locomotor activity was increased in the aSyn transgenic animals (tg) compared to wild-type (wt) littermates. After amphetamine administration, increased activity in tg mice was seen at the 6 and 18 month time-points, while only initial locomotor activity was increased in 9 months old tg animals (A-E). Total travelled distance of tg mice was increased in all age groups during the initial 5 minutes of the locomotor activity test compared to wt littermates (F). Data are expressed as mean ± SEM; n = 7-23. Repeated measures two-way ANOVA(A-E), Student's t-test (F); *p < 0.05, **p < 0.01, ***p < 0.001.

Discussion
In this study, our aim was to perform additional behavioural and biochemical characterization of the homozygous C57BL/6J-Tg(TH-SNCA*A30P*A53T)39Eric/J tg mice line and establish whether the mouse strain generates aSyn aggregate accumulation in the nigrostriatal tract. Our results show that homozygous tg mice had www.nature.com/scientificreports www.nature.com/scientificreports/ significant differences in locomotor activity in all age groups starting from 3 month-old animals, altered amphetamine response, and increased aSyn oligomer levels and decreased TH immunostaining in the nigrostriatal pathway. In an earlier study by Richfield et al. 20 with the same mouse strain, the tg mice had age-dependent changes www.nature.com/scientificreports www.nature.com/scientificreports/ in locomotor activity starting from 7 months of age, significantly later than in our study. Therefore, we propose that a genotyping protocol that was able to separate heterozygous and homozygous mice was needed to generate homozygous animals. Figure 6. aSyn transgenic (tg) mice had significantly decreased TH and increased oligomer specific aSyn (aSyn OS) immunoreactivity in the striatum (STR) and substantia nigra (SN). Representative images of TH and aSyn OS staining from the striatum and substantia nigra of 12 and 18 month old wt and tg mice (A). TH optical density (OD) was significantly decreased in the STR and SN of 12 and 18 month old tg mice (B,C). OD of aSyn OS was significantly increased in both STR and SN at 12 and 18 months (D,E). Data are expressed as mean + SEM; n = 8-18. Student's t-test,* p < 0.05, **p < 0.005, ****p < 0.0001.
Our detailed analysis of locomotor activity revealed that tg mice had decreased dark time activity but their highest activity period was shifted earlier compared to wt mice indicating abnormal locomotor behaviour. Interestingly, the most distinct differences in vertical and jump counts were observed in 3 and 6 months old animals, which could be considered to model motor symptoms of early-onset PD. However, at 12 months of age, the difference was not as clear as at earlier time points, and at 18 months tg mice were hyperactive in the locomotor test compared to wt littermates. This has been described earlier with A53T-aSyn overexpression tg mice strains that have shown to exhibit hyperactivity and anxiety, explained by alterations in function of DAT and an increased amount of D1 receptors [21][22][23] . Additionally, we observed significantly elevated glutamate levels in the striatal tissue of 18 months old tg mice compared to wt littermates which could partly explain the hyperactivity. It has been reported that the C57BL/6 mice strain has reduced motor activity at 12 and 22 months of age compared to 3 months old animals 24 , supporting reduced motor activity seen in 18 months old wt animals.
Richfield´s study (2002) 20 demonstrated a reduced locomotor response to amphetamine in tg mice, however our study was contrary since 6, 9, and 18 months old tg mice had an increased response to amphetamine in the locomotor activity assay compared to wt littermates. Increased amphetamine-induced locomotor activity and decreased amphetamine-induced DA release in tg mice can indicate dysfunction of DAT. Interestingly, elevated amphetamine-induced DA release was only observed in 12 months old tg mice compared to wt littermates but not in 18 months old tg mice. This can be explained by the finding that the wt mice had decreased amphetamine-induced DA release at 18 months old compared to 12 months old wt mice, indicating that age-related changes in DAergic function in wt mice could cover the difference that was observed between the 12 months old wt and tg mice. This promotes the role of aSyn aggregation and toxicity in this mouse model since aSyn is known to regulate the function of DAT. In its normal state, aSyn is considered as a negative regulator of DAT and aSyn overexpression has been shown to modify basal and amphetamine-induced DA efflux 16 .
The earlier study also reported that striatal tissue concentrations of DA, DOPAC, and HVA were declined in 16-18 months old tg mice compared to wt littermates 20 . This was consistent with our study as the same effect was seen for 18 months old mice. Striatal DA was not yet changed in 12 months old tg mice compared to wt littermates although striatal and nigral TH were already decreased at 12 months. However, impairment in DAergic function of 12 months old mice was also observed in the microdialysis experiment as amphetamine-induced DA release was decreased and behaviour altered in tg mice compared to wt mice. Previous studies have shown that aSyn binds to TH 25 and A53T mutated aSyn and aSyn aggregation and phosphorylation abolishes aSyn's impact on TH 26,27 . Therefore, the elevated level of aSyn oligomers seen in tg mice could possibly explain reduced TH. Additionally, striatal tissue concentrations of 5-HT and GABA were increased which may arise from decreased DA regulation 28 . As a further support for this, striatal microdialysis revealed elevated baseline levels of 5-HIAA in tg mice compared to wt mice indicating increased 5-HT metabolism. Age-dependent decreases in striatal GABA and glutamate concentrations were observed in wt mice but not in tg mice. Similar findings in wt mice have been reported earlier by 29 . Decreased striatal DA can increase GABA and glutamate since DA downregulates striatal GABA and glutamate via DA receptor D2 28,30,31 . Additionally, accumulation of aSyn has been reported to induce enlargement of glutamatergic nerve terminals in the mouse striatum 32 . In conclusion, decreased DAergic function combined with the effect of aSyn on glutamatergic neurons is probably causing the difference in striatal GABA and glutamate between 18 months old wt and tg mice.
After aSyn aggregation was revealed as a potential key player in PD pathophysiology, several tg mouse lines overexpressing human aSyn with a A30P or A53T point mutation have been described. Mice expressing A30P aSyn have, in several studies, failed to show differences in locomotor activity, and in DA and TH levels despite accumulation of aSyn in several brain regions [33][34][35] . A53T mutant aSyn expressing mice usually have more severe motor impairments starting at older age, but malfunction of the DAergic system has not been clearly determined [36][37][38] . This has been one of the major problems when using aSyn tg mice since they do not model the most important feature in PD, the degeneration of DAergic system, very effectively. Additionally, these aSyn point mutations are a risk factor for early onset PD which occurs before the age of 40 to 50 years in humans 39 , and generally behavioural deficits are seen only in aged tg animals (>12 months). Our study demonstrated that A30P*A53T aSyn tg mice did not only have early behavioural changes but also changes in their DAergic system and a decreased amount of TH positive cells compared to wt littermates. These effects are most likely caused by toxicity from increased aSyn oligomers 40 that we reported here in 12-and 18-month age groups. Behavioural changes indicate that aSyn toxicity starts earlier since locomotor deficits were already observed in 3 months old tg mice, and another double mutant A30P*A53T aSyn mouse line with a different promoter (Thy-1 promoter) described motor impairment starting from 3 months 41 , similar to our findings. Although A30P and A53T double mutation in SNCA has not been described clinically, our results indicate that this models early onset PD better than other tg mouse models.
In conclusion, there is still a lack of a mouse model for PD that displays motor and non-motor deficits typical for PD, alterations in the DAergic system and DAergic cell loss together with aSyn propagation and formation of aSyn-rich Lewy bodies. Such a research tool would be particularly critical when developing novel disease-modifying therapies targeting causes of PD. Our current study with homozygous double mutant A30P*A53T aSyn tg mice does not fulfil all of these requirements but it has early onset and age-dependent changes in locomotor activity and in the striatal DAergic function together with aSyn oligomer formation, and it could be a useful tool to model early onset PD with familial SNCA mutations. Genotyping. While we bred tg mice with wt mice to create homozygous mice, we found that homozygous male animals have a phenotypical feature where length of hair is much longer compared to the heterozygous and wt animals ( Supplementary Fig. S2). Homozygous mice with the long-haired phenotypical feature and wt littermates were selected for sequencing. Sequencing service and genotyping primer design for the differentiation of the wt, heterozygous, and homozygous tg animals was provided by the University of Helsinki Genomic Core facility (Helsinki, Finland). Primer pairs provided were tested and all of the PCR products were sequence verified by the University of Helsinki Genomic Core facility. List of the best genotyping primer pairs and sequences can be found in Table 1. Routine DNA extraction and PCR were performed using REDExtract-N-Amp ™ Tissue PCR kit (#XNAT-1000RXN, Sigma-Aldrich, Saint Louis, MO, USA) with touchdown PCR cycling conditions provided in Table 2 and PCR product size was verified by agarose gel electrophoresis. tissue processing. At the age of 12 or 18 months mice were sacrificed by cervical dislocation followed by dissection of the whole brain. The hemispheres were separated by using a brain matrix. The left hemispheres were frozen in isopentane on dry ice to collect striatal and nigral tissue samples for HPLC analysis as described in 42 The right hemispheres were postfixed for 24 h in fresh 4% paraformaldehyde (PFA) at +4 °C and transferred to 10% sucrose in PBS (pH 7.4; 137 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 , 1.8 mM KH 2 PO 4 ) overnight at +4 °C. On the next day, tissue was transferred to 30% sucrose solution in PBS until brains sank. Brains were frozen on dry ice and were kept at −80 °C until sectioning. Frozen brains were sectioned as 30 μm free-floating sections on a cryostat (Leica CM3050) and kept in a cryoprotectant solution (30% ethylene glycol and 30% glycerol in 0.5 M phosphate buffer).

Immunohistochemistry (IHC).
Tyrosine hydroxylase (TH) IHC was done as described in 43 . In short, after blocking endogenous peroxidase activity sections were incubated for 30 min in 10% normal goat serum to block nonspecific binding, after which the sections were incubated overnight in rabbit anti-TH primary antibody (1:2000; AB152, RRID:AB_390204, Merck, Darmstadt, Germany). Subsequently, the sections were placed in goat anti-rabbit biotin-conjugated secondary antibodies (1:500; BA1000, RRID:AB_2313606, Vector Laboratories, Peterborough, UK). The signal was enhanced with the avidin-biotin complex method (Standard Vectastain ABC kit, RRID: AB_2336819, Vector Laboratories) and visualized with 3,3′-diaminobenzidine (DAB). Oligomer-specific -aSyn IHC was performed as described in 43   Microscopy and optical density analyses. The optical densities (ODs) of TH and oligomer-specific aSyn from striatum and SNPc were determined as described earlier in 43 . Digital images were scanned at 40x magnification with a Pannoramic Flash II Scanner (3DHISTECH, Budapest, Hungary), and three coronal sections from each mouse were processed for further analyses with Pannoramic Viewer (version 1.15.3. RRID:SCR_014424, 3DHISTECH). Images were converted to grayscale and inverted, and line analysis tools for striatum or freehand for SN in ImageJ (1.48b; RRID:SCR_003070, NIH) was used to measure the OD of immunoreactivity. Corpus callosum was used to subtract the background optical density of each section and then normalized to the control mice.
Behavioral assessments. Locomotor activity was measured every three months from 3 to 12 months and 18 months using automated open field locomotor activity chambers (Activity monitor, SOF-812, Med Associates inc, Georgia, USA). Total photobeam breaks were recorded for 22 h (starting at 10:00) for horizontal, vertical, and ambulatory movements. Amphetamine-induced locomotor activity was assessed every third month. Mice were habituated in locomotor boxes for 30 minutes before amphetamine (3 mg/kg i.p) was administered. Locomotor activity was measured for 90 minutes immediately after the amphetamine administration.

Microdialysis.
Microdialysis was performed in the 12 months old and 18 months old tg mice and their wt littermates as described earlier in 42 . Shortly, a microdialysis probe (1-mm cuprophan membrane, o.d. 0.2 mm, 6 kDa cut-off; AT4.9.1.Cu, AgnTho's) was inserted into the guide cannula 2 h before the experiment, and the probe was perfused with a modified Ringer solution (147 mM NaCl, 1.2 mM CaCl 2 , 2.7 mM KCl, 1.0 mM MgCl 2 , and 0.04 mM ascorbic acid) at a flow rate of 2.0 μL/min. Four baseline samples were collected (20 min/40 μL/sample) after the stabilization period. After the collection of baseline samples, the probe was perfused 2 × 20 min with 10 μM and 30 μM d-amphetamine sulphate with 2 × 20 min recovery time between the concentrations. The concentrations of DA, its metabolites, DOPAC and HVA, and 5-HIAA as well as GABA in dialysates were measured using the HPLC methods that have been described earlier in 42 .
Tissue HPLC analysis. Striatal tissue samples were punched below corpus callosum +0.74 mm from bregma to 2 mm depth by using sample corer (i.d. of 2 mm) with a plunger (Stoelting Co, Wood Dale, IL, USA) on a cryostat (Leica CM3050). Tissue processing was done as earlier described in 42 . The concentration of DA, its metabolites DOPAC and HVA, 5-HT, its metabolite 5-HIAA, GABA and glutamate in the tissue samples of striatum were analyzed with an HPLC as earlier described in 42 . The concentrations were calculated as nanograms per milligram of brain tissue.

Data availability
All materials, data and associated protocols are available to readers.