|Home | About | Journals | Submit | Contact Us | Français|
Parkinson's disease (PD) is the most common neurodegenerative movement disorder. A pathological hallmark of PD is the presence of intraneuronal inclusions composed of fibrillized α-synuclein (α-syn) in affected brain regions. Mutations in the gene, PARK7, which encodes DJ-1, can cause autosomal recessive early-onset PD. Although DJ-1 has been shown to be involved in diverse biological processes, several in vitro studies suggest that it can inhibit the formation and protect against the effects of α-syn aggregation. We previously established and characterized transgenic mice expressing pathogenic Ala53Thr human α-syn (M83 mice) that develop extensive α-syn pathologies in the neuroaxis resulting in severe motor impairments and eventual fatality. In the current study, we have crossbred M83 mice on a DJ-1 null background (M83-DJnull mice) in efforts to determine the effects of the lack of DJ-1 in these mice. Animals were assessed and compared for survival rate, distribution of α-syn inclusions, biochemical properties of α-syn protein, demise and function of nigral dopaminergic neurons, and extent of gliosis in the neuroaxis. M83 and M83-DJnull mice displayed a similar onset of disease and pathological changes, and none of the analyses to assess for changes in pathogenesis revealed any significant differences between M83 and M83-DJnull mice. These findings suggest that DJ-1 may not function to directly modulate α-syn nor does DJ-1 appear to play a role in protecting against the deleterious effects of expressing pathogenic Ala53Thr α-syn in vivo. It is possible that α-syn and DJ-1 mutations may lead to PD via independent mechanisms.
In 1997, the seminal identification of a missense mutation (Ala53Thr) in α-synuclein (α-syn) in several kindred with Parkinson's disease (PD) (1) lead to subsequent studies showing that α-syn is a major component of several types of brain amyloidogenic pathological inclusions (2–5). For example, it is now established that Lewy bodies, characteristic neuronal inclusions of PD, are predominantly composed of α-syn polymerized into 10–15 nm fibrils (2–4). The spectrum of neurodegenerative disorders with α-syn pathological inclusions are termed synucleinopathies (2–6). α-Syn is a 140 amino acid protein that is predominantly expressed in the brain and which localizes to presynaptic nerve terminals (3,4,7). The physiological role for α-syn is not fully understood, but several functions including the abilities to act as an auxiliary molecular chaperone and to play a role in maintaining synaptic nerve terminal integrity have been suggested (2,7–9). Although there is substantial evidence supporting the toxic nature of aberrant α-syn aggregation, mutations (missense or gene multiplication) in the α-syn gene (SNCA) provide the most direct evidence for a pathogenic role of α-syn (1–5).
PD is the most common neurodegenerative movement disorder (10,11). The clinical features of PD include bradykinesia, postural instability, resting tremor and rigidity, which result from the progressive loss of dopaminergic neurons in the substantia nigra pars compacta (12–16), as well as a range of non-motor symptoms (17,18). While for most patients the cause for PD is idiopathic, mutations in genes at multiple loci, designated PARK1 through PARK 13, result in parkinsonian phenotypes with distinct features (19–22). DJ-1 was initially associated with PD pathogenesis when homozygous recessive mutations in the PARK7 gene encoding DJ-1 protein were identified in patients with early-onset PD (23). Subsequent to this initial report, various autosomal recessive mutations in DJ-1, including missense, splice-site, frameshift and large deletions have been discovered (23–29) in 1–2% of PD patients with early to mid age of onset (26,30,31). DJ-1 mutations are thought to cause PD due to a loss of functional DJ-1 protein, although the natural role for DJ-1 as it relates to sporadic PD is not known (32–35). In addition, no autopsies have been performed on individuals with DJ-1 mutations; therefore, the exact neuropathological manifestations of disease in patients harboring DJ-1 mutations remains to be determined.
DJ-1 encodes a 189 amino acid protein which is a member of the ThiJ/PfPI superfamily based on its structure (36–39). It is expressed in both neurons and astrocytes in the brain (40–44), but it is also expressed in many other organs (45–47). DJ-1 has been shown to protect against a variety of insults including oxidation, inflammation, mitochondrial inhibition and proteasome dysfunction (48–56). More specifically, in vitro studies have suggested that DJ-1 may act to directly prevent α-syn aggregation (57,58) and several groups have reported that DJ-1 can ameliorate the harmful effects of mutant α-syn in vitro and in cell culture studies (57,59,60). Interestingly, elevated levels of oxidized DJ-1 protein are present in the brains of patients with sporadic PD (61) and DJ-1 associates with inclusions in various other synucleinopathies (62,63). Thus, it is plausible to hypothesize that DJ-1 may physiologically act to protect against the formation or the harmful effects of aggregated α-syn.
We previously reported a transgenic mouse of synucleinopathies that was generated by expressing human Ala53Thr α-syn in the nervous system using the mouse prion protein promoter (64). These mice developed an age-dependent severe movement disorder which is associated with abundant neuronal α-syn inclusions in the neuraxis and axonal degeneration (64). As DJ-1 has been postulated to have several protective functions, including anti-α-syn aggregation properties, we sought to study the effects of the lack of DJ-1 in these mice. We hypothesize that the loss of DJ-1 may exacerbate the extent or promote the onset of disease in these mice, either by promoting α-syn aggregation or the consequences of α-syn inclusions. In the current study, transgenic mice homozygotically expressing human Ala53Thr α-syn (‘M83 mice’) were crossed with genetically altered null DJ-1 mice in order to generate homozygous Ala53Thr α-syn transgenic mice on a DJ-1 null background (‘M83-DJnull mice’). M83-DJnull mice were analyzed and compared with M83 mice as it relates to survival rate, distribution of α-syn pathologies, biochemical properties of the α-syn protein, and extent of gliosis in the neuroaxis.
DJ-1 null mice were generated as described in detail in ‘Materials and Methods’ in order to create a loss-of-function DJ-1 mouse model. The disruption of DJ-1 expression was demonstrated with several DJ-1 antibodies by western blot analysis of total protein lysates that were extracted from the brain cortices of DJ-1 null, heterozygous (Het) and wild-type (WT) mice (Fig. 1C). The protein signal for DJ-1 in DJ-1 Het mice was ~0.6-fold the intensity compared with WT mice and was completely absent in samples from DJ-1 null mouse tissues. Similar results were obtained using other tissues including cerebellum, brainstem, spinal cord, liver and lung (data not shown). As reported by others for other DJ-1 null mice, extensive histological analyses did not demonstrate any evidence of degeneration in the nervous system (data not shown and see Fig. 3C) (65–69).
In attempts to understand the role for DJ-1 in modulating α-syn pathology in vivo, the previously established transgenic mouse line M83 expressing human Ala53Thr α-syn (64) was cross-bred with the DJ-1 deficient mouse line described above in order to generate mice homozygous for both the α-syn transgene and the DJ-1 null allele (‘M83-DJnull’ mice). The double homozygous transgenic genotype was confirmed as described in Materials and Methods section. The lack of DJ-1 was confirmed by western blot analysis that also demonstrated that the deficiency of DJ-1 did not alter the expression levels of human α-syn (Fig. 1D). It is known that M83 mice exhibit a severe motor impairment with mid to late age of onset and which eventually results in fatality (64). To determine whether a DJ-1 deficiency would alter the disease motor phenotype and onset of disease, a cohort of M83-DJnull mice and M83 mice were housed in parallel. It was observed that the M83-DJnull mice presented symptoms of weight loss, neglect of grooming, decreased mobility, paralysis and eventual fatality, symptoms similar to those characterized for the M83 mice described previously (64). The severity of disease symptoms was comparable between M83-DJnull mice and M83 mice. Further, although the age range for the onset of disease in the M83-DJnull mice was 4–16 months of age (Fig. 2), this was not significantly different from the age of disease onset in M83 mice, which was 8–17 months of age. To confirm this observation, statistical analyses comparing the survival rates between mice revealed that the median age of disease onset for both mouse genotypes was 11 months of age.
Although M83-DJnull mice did not exhibit enhanced motor impairment or decreased survival rate in comparison to M83 mice, it is possible that DJ-1 deficient animals may display an altered distribution of α-syn inclusions, perhaps affecting areas of the nervous system that were not previously observed in M83 mice. Certain neuronal populations, including the hippocampus, the olfactory bulb, dopaminergic neurons of the substantia nigra and Purkinje cells of the cerebellum are completely devoid of α-syn inclusions in the M83 mouse (64). Thus, post-mortem immunohistochemical and immunofluorescence analyses were conducted on brain and spinal cord tissues from sick M83-DJnull mice in order to assess the distribution of α-syn inclusions. These analyses revealed that affected M83-DJnull mice demonstrate the same distribution of somatodendritic α-syn neuronal inclusions and dystrophic α-syn neurites throughout the neuraxis as seen in M83 mice. A high density of inclusions was observed in the spinal cord, throughout the brainstem, the deep cerebellar nuclei, and some regions of the thalamus, such as the medioventral, ventromedial and paracentral nuclei (Fig. 3A and data not shown). In the cortex, α-syn inclusions were predominantly observed in the motor cortex. Conversely, regions that were devoid of pathology in M83 mice, such as the olfactory bulb, hippocampus or Purkinje neurons were also spared of inclusions in M83-DJnull mice (Fig. 3A and data not shown). In representative images showing the similarities between the mouse genotypes, inclusions were immunoreactive with antibody Syn 514 which detects pathological α-syn (Fig. 3A). Ubiquitin, which is known to modify α-syn in the inclusions in M83 mice (64,70), also localized to most of the α-syn inclusions in M83-DJnull animals (Fig. 3A). It was previously shown that α-syn inclusions that are present in the diseased tissues of M83 mice are highly phosphorylated at Ser129 in α-syn (71). Similarly, the majority of the α-syn inclusions in M83-DJnull mice were immunoreactive for an antibody specific for this modification (Fig. 3A). Notably, although α-syn inclusions were present in a few of the tyrosine hydroxylase (TH)-positive neurons of the locus coeruleus (Fig. 3B), none were present in the TH-positive neurons of the substantia nigra in M83-DJnull animals (Fig. 3B), a phenomenon which was also previously observed in M83 mice (64). Further, stereological analyses of the substantia nigra in WT, M83, DJ-1 null and M83-DJnull animals did not reveal any significant differences in the total number of dopaminergic neurons between mouse genotypes when animals were analyzed at 6 and 12 months of age (Fig. 3C). However, similar to M83 mice, a significant number of the α-syn inclusions in M83-DJnull mice were positive for the amyloid binding dye Thioflavin-S (Fig. 3D) (64).
Previous studies of M83 mice revealed that the formation of α-syn inclusions is reflected by the accumulation of insoluble α-syn protein in the respective tissues analyzed (64). Although M83-DJnull mice did not reveal histological differences in comparison to M83 mice, biochemical studies may reveal more subtle differences. Western blot analyses were performed using extracts from brain cortical, cerebellar and brainstem/spinal cord (BS/SC) tissues which had been fractionated into buffers of increasing solubilization strengths. Analyses between age-matched sick M83-DJnull and M83 mice were compared. These analyses revealed that the abundance and biochemical distribution of human α-syn was similar between the M83-DJnull and M83 mice in all of the tissues analyzed. In the cerebral cortex, human α-syn was most abundantly extracted into the high salt (HS), HS-triton (HS-TX) and radioimmunoprecipitation assay (RIPA) fractions for both mouse genotypes (Fig. 4A). In the cerebellum of M83 and M83-DJnull mice, α-syn was distributed in a similar manner (Fig. 4B). However, α-syn exhibited slightly reduced solubility into SDS fractions in cerebellar tissues of some of the mice analyzed (Fig. 4B). This observation was not due to a DJ-1 deficiency, however, since it was variable from mouse to mouse, irrespective of genotype (data not shown). In the BS/SC regions of affected M83 and M83-DJnull mice, biochemical analysis revealed a similar accumulation of RIPA-insoluble/SDS-soluble human α-syn (Fig. 4C). Additionally, although there were no differences in the abundance or extractability of the α-syn present in the BS/SC tissues of the predicted ~17 kDa size, when comparing between M83 and M83-DJnull mice, several insoluble, higher molecular weight α-syn species were detected on some immunoblots (Fig. 4C). These species were previously shown to represent ubiquitinated forms of α-syn (70). However, this phenomenon was not due to a DJ-1 deficiency, because it was variable and independent of DJ-1 expression (data not shown) (70).
In a previous report, diseased M83 mice exhibited increased levels of insoluble α-syn protein which was highly phosphorylated specifically at amino acid residue Ser129 (71). To determine whether a DJ-1 deficiency could alter the levels of the Ser129-specific phosphorylation of α-syn, brain cortical, cerebellar and BS/SC tissue extracts from age-matched diseased M83 and M83-DJnull mice were compared by western blot analysis with the antibody pSer129. α-Syn that was extracted from BS/SC tissues from M83-DJnull mice were highly phosphorylated in the SDS-soluble fraction but barely detectable in the HS, HS-TX or RIPA fractions in all of the mice analyzed (Fig. 4C). However, there were no differences between M83 and M83-DJnull mice when the levels of phosphorylated α-syn were compared.
It is well known that reactive gliosis can follow brain tissue damage and is typically accompanied by inflammation (72). A recent study by Waak et al. (52) showed that the disruption of DJ-1 expression in astrocytes resulted in enhanced neurotoxcity which was due to an elevated neuroinflammatory response. Diseased M83 mice exhibit astrocytic gliosis in the brain regions where α-syn inclusions are the most abundant (64). It is possible that a DJ-1 deficiency in M83-DJnull mice would result in an aberrant inflammatory response in these animals. To test this hypothesis, M83 and M83-DJnull mouse brain tissues were evaluated by immunohistochemistry analysis with the reactive microglial antibody marker, IBA-1 as well as the astrocyte marker, glial fibrillary acidic protein (GFAP). The distributions of the microglia and astrocytes present in the mouse tissues were compared. Although these analyses revealed that reactive microglia were primarily localized to the brainstem and spinal cord tissue regions, the distribution and abundance of microglia were similar in M83 and M83-DJnull mouse tissues (Fig. 5A and data not shown). Astrogliosis also was observed in areas of the neuroaxis such as the brainstem and spinal cord where abundant α-syn pathological inclusions were present; however, the prevalence was similar in M83 and M83-DJnull mice (Fig. 5A and data not shown). This observation was further confirmed by biochemical fractionation/western blot analysis for IBA-1 (Fig. 5B) and GFAP (Fig. 5C).
During the course of previous proteomics analysis, we discovered that affected M83 mice display increased levels of peroxiredoxin (Prx) 6 that is substantiated by western blot analysis (Fig. 6A) and this may reflect an attempt to activate a protective mechanism. In addition, DJ-1 is thought to be able to act as an atypical Prx-like peroxidase (65) and to protect against oxidative insults (48,59,65,73–77). It is possible that the disruption of DJ-1 expression may induce further reactive stress in M83-DJnull mice. Therefore, the effect of the paucity of DJ-1 on the levels of Prx6 induced by α-syn pathology was assessed. However, no major differences between M83 and M83-DJnull mice were detected as it related to the extent of Prx6 expression or biochemical distribution.
As some studies suggest that DJ-1 and α-syn may modulate TH and l-3,4-dihydroxyphenylalanine (l-DOPA) decarboxylase enzyme activities (78–81), it is possible that M83-DJnull mice could display altered levels of dopamine (DA) and its metabolites. To test this hypothesis, striatum tissues were dissected from WT, M83, DJ-1 null and M83-DJnull mice at 6 and 12 months of age. Tissues were analyzed by high performance liquid chromatography-electrochemical detection (HPLC-ECD) to quantify and compare the total levels of l-DOPA, DA and 3,4-dihydoxyphenylacetic acid (DOPAC) between mouse genotypes. However, there were no significant differences between animals for either of the age groups analyzed (Fig. 7), indicating normal DA-related biochemical activities in M83-DJnull mice.
Some in vitro studies and cell culture paradigms indicate that DJ-1 may be able to modulate the aggregation of α-syn and/or act to mitigate the toxicity of pathogenic forms of α-syn (57–60). However, these properties of DJ-1 have not been directly assessed in vivo. M83 transgenic mice are previously described animals that express human Ala53Thr α-syn and developed a severe age-dependent motor phenotype associated with the formation of α-syn pathological inclusions in the neuroaxis (64). These inclusions were demonstrated to be composed of fibrillized aggregates of insoluble α-syn protein with a high abundance in the brainstem and spinal cord, while inclusions were absent in some regions such as the olfactory bulb, hippocampus and dopaminergic neurons of the substantia nigra (64). In the current study, in efforts to ascertain whether DJ-1 modulates α-syn aggregation and pathology in vivo, double homozygous M83-DJnull transgenic mice were generated which express human Ala53Thr α-syn on a DJ-1 null background. If DJ-1 normally acts to mitigate α-syn aggregation or has protective functions against the deleterious effects of α-syn aggregation, it was hypothesized that the DJ-1 deficient M83-DJnull mice would exhibit an exacerbated phenotype in comparison to that observed in M83 mice. Statistical analyses comparing the survival rates revealed that the disease onset was not earlier in M83-DJnull mice compared with M83 mice. The median age of survival for both groups of animals was 11 months of age. This suggested that DJ-1 may not play a major role in ameliorating the harmful effects of expressing Ala53Thr mutant α-syn in vivo. However, to further assess the differences between these mice, several analyses were conducted. Nevertheless, no major differences between M83 and M83-DJnull mice were observed in any of the analyses that were performed. Immunohistochemical analyses comparing the distributions of α-syn pathologies throughout the neuroaxis failed to detect any overt variations between M83 and M83-DJnull mice. Both mouse groups consistently exhibited abundant pathologies in the same tissue regions, while the absence of inclusions in other areas was remarkably similar between mice. These analyses also revealed that most of the inclusions in diseased M83-DJnull animals were composed of ubiquitinated, fibrillized α-syn protein that was highly phosphorylated at amino acid residue Ser129 similar to inclusions in diseased M83 mice (64,71) and in patients (71,82–85). The formation of inclusions in M83 and M83-DJnull mice was further validated by biochemical and western blot analyses to assess the soluble extractability. However, the studies herein did not reveal any changes relating to α-syn aggregation in M83-DJnull animals. In addition, these studies did not reveal any differences in the levels of the phosphorylation of Ser129. Thus, these findings suggest that DJ-1 does not directly alter α-syn in vivo, as it relates to protein fibrillization, phosphorylation or aggregation. Neither does DJ-1 appear to directly modulate the expression of mutant α-syn as similar levels of the mutant protein were observed between mice in the brain regions analyzed.
Although the present study suggests that DJ-1 does not act to directly regulate Ala53Thr mutant α-syn in vivo, it is possible that DJ-1 can mitigate secondary deleterious effects of α-syn aggregation in mice such as inflammation and oxidative stress. It was previously established that M83 mice exhibited astrocytic gliosis associated with the formation of α-syn inclusions (64). Additionally, recent studies using a primary neuron and astrocyte co-cultured model reveal that a DJ-1 deficiency is causal of an enhanced neuroinflammatory response that results in increased neurotoxicity (52). To determine whether DJ-1 modulates the gliosis associated with α-syn aggregation in vivo, diseased M83 and M83-DJnull mice were analyzed by immunohistochemistry with astrocyte-specific and reactive microglia-specific markers. However, these analyses did not reveal any differences between mouse types relating to the distribution and/or extent of gliosis. Additionally, the biochemical and immunoblot assessments of protein extracts from M83 and M83-DJnull mouse tissues also did not identify any differences. Thus, as gliosis was independent of DJ-1 expression, DJ-1 may not act to regulate gliosis in M83 mice.
As DJ-1 and α-syn may regulate TH promoter activation and have an effect on TH enzyme activity (78–81,86), it was relevant to assess for the levels of striatal DA in M83-DJnull mice in comparison to WT, DJ-1 null and M83 animals. However, no differences were revealed for any of the genotypes when mice were analyzed for the levels of l-DOPA, DA or DOPAC. This supports previous findings in the literature which report normal DA levels and its metabolites in DJ-1 null animals (65,66,68,69). Further, in the studies herein, stereological analyses revealed no change in the number of nigral dopaminergic neurons in both a young and old cohort of M83-DJnull mice. Our finding is also in agreement with other reports which do not detect degeneration of these neurons in DJ-1 null mice (65–68,87,88).
In a variety of model systems, the expression of mutant Ala53Thr α-syn has been shown to result in increased levels of oxidative stress (89–91). Although DJ-1 has been shown to be involved in diverse biological processes, several studies demonstrate its ability to mitigate the toxic effects of aberrant oxidation (48,59,65,73,75–77,92). In addition, in some studies, DJ-1 was reported to be able to act as an atypical Prx-like peroxidase (65). The current study demonstrated that when compared with WT mice and healthy M83 mice, sick M83 animals exhibited increased steady-state expression of Prx6. This phenomenon specifically occurred in the BS/SC tissues of M83 mice where α-syn pathologies were most abundant. Comparatively, no alterations in Prx6 were observed in the cortex in diseased M83 animals (data not shown). This suggests that the increased levels of Prx6 observed in sick M83 mouse BS/SC tissues may be a protective attempt against α-syn aggregation. It is known that Prx6 levels are increased in PD and that this increase is associated with the recruitment of Prx6 to Lewy body inclusions (93). Additionally, as Prx6 has been shown to be up-regulated in response to oxidation (94), it could imply that α-syn inclusion formation is accompanied by increased oxidative stress in sick M83 mice. To determine whether these observed alterations in Prx6 were modulated by DJ-1 in vivo, BS/SC tissues from sick M83 and M83-DJnull mice were analyzed by western blot analysis and compared for immunoreactivity with the Prx6 antibody. However, these analyses revealed that diseased M83 and M83-DJnull mice expressed Prx6 at similar levels.
Herein, a loss-of-function DJ-1 mouse model was employed in attempts to ascertain the role for DJ-1 in mitigating mutant α-syn in vivo. Many groups have utilized DJ-1 deficient animals in efforts to define a physiological role for DJ-1 protein in vivo, as it relates to PD (65,66,68,69,87,95–97). However, although DJ-1 null animals have been shown to exhibit mild behavioral changes, these were not associated with any observed degeneration in the nervous system (65–67,69,95). Neither have DJ-1 null mice exhibited increased signs of oxidative stress (67,95). A few studies have shown that DJ-1 deficient animals are more vulnerable to paraquat (87), MPTP (68,97) and rotenone toxicity (96), suggesting a role for DJ-1 in protecting against oxidative stress and mitochondrial impairments.
In the present study, DJ-1 deficient animals were not more vulnerable to the toxic effects of mutant Ala53Thr α-syn. These studies suggest that DJ-1 may not have a significant role in mitigating the formation of α-syn inclusions and their deleterious effects in vivo. Alternatively, there may be compensatory mechanisms in the mouse which act to mimic the function of the lost DJ-1 protein. For example, recent studies by other groups suggest a possible genetic interaction between DJ-1 and other PD-related genes, PINK1 and parkin (98,99). It is of interest to note that in the current study, no alterations in the steady-state levels of parkin protein were observed in the brains of M83 and M83-DJnull mice when they were analyzed by western blot (data not shown). Neither did PINK1 protein levels appear to be modulated by DJ-1 in the mice analyzed (data not shown). Further, triple knockout mice that are deficient for DJ-1, parkin and PINK1 do not have decreased life spans nor do they exhibit phenotypes that are comparable to any of the symptoms of PD (88). It is possible that yet unidentified candidate genes may be acting to compensate for the loss of DJ-1 in vivo. Identifying such candidate genes would give new insights into plausible biochemical mechanisms that underlie the pathogenesis of PD.
Anti-IBA1 is a rabbit polyclonal antibody raised against ionized calcium-binding adaptor molecule 1 (Iba1), a marker for activated microglial cells (Wako Chemicals USA, Inc., Richmond, VA, USA). Anti-GFAP is a rabbit polyclonal antibody against glial fibrillary acidic protein, a specific marker for astrocytes (Promega Corporation, Madison, WI, USA). LB509 is a mouse monoclonal antibody that specifically reacts with human α-syn (100,101). Syn 303 and 514 are monoclonal antibodies raised against oxidized forms of human α-syn and that preferentially recognize pathological forms of the protein (102,103). pSer129 is a monoclonal antibody that specifically recognizes α-syn that is phosphorylated at the Ser129 (71). Anti-ubiquitin (clone 1510) is a mouse monoclonal antibody that reacts with conjugated and unconjugated forms of ubiquitin (Millipore, Billerica, MA, USA). Anti-actin (clone C4) is an affinity purified monoclonal antibody that reacts with all vertebrate isoforms of actin (Millipore). Anti-TH antibody (Millipore) is an affinity purified rabbit polyclonal antibody. An affinity purified rabbit polyclonal antibody raised against a synthetic peptide corresponding to amino acid residues 196–210 of human Prx6 was obtained from Sigma-Aldrich (Saint Louis, MO, USA). 691 is a rabbit polyclonal antibody raised against recombinant human DJ-1 protein but that reacts with DJ-1 from various species (63). DJ-1 (N-20) is an affinity purified goat polyclonal antibody raised against a peptide corresponding to the N-terminus of human DJ-1 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). Anti-β-tubulin (clone TUB2.1) is a monoclonal antibody raised against β-tubulin (Sigma). Goat anti-mouse IgG Alexa Fluor 488 conjugated antibody and goat anti-rabbit or anti-mouse IgG Alexa Fluor 594-conjugated antibodies were purchased from Molecular Probes (Eugene, OR, USA).
Murine embryonic stem (ES) cells (clone XE726) with a recombination resulting in the disruption of the DJ-1 gene were obtained from BayGenomics (San Francisco, CA, USA). This cell line was generated by random exon trapping and it was determined by DNA sequencing that the recombination event occurred after exon 6 in the murine DJ-1 gene (Fig. 1A). These ES cells were injected into C57BL/6J blastocysts as a service of the University of Pennsylvania Transgenic and Chimeric Mouse Facility and chimeric mice were generated. The germ line transmission of the null-gene was determined by breeding chimeras with C57BL/6J mice and PCR analyses of tail DNA with primers specific for the neomycin gene. Mice homozygous for the null-allele were generated by crossing F1 mice, and homozygous null mice were determined by PCR analyses of tail DNA and western blot analysis of mouse tail protein samples (dissolved in 4% SDS, 8 m urea) using an antibody to DJ-1.
The previously established homozygous transgenic mice line M83 expressing Ala53Thr human α-syn (64) was bred with the homozygous DJ-1 null mouse line described in the previous section. The double Het offspring were mated to generate double homozygous transgenic mice. Mouse genotypes were confirmed by Southern blot analysis of tail genomic DNA with a probe for human α-syn. Homozygous α-syn transgenic lineages were identified by quantitative Southern blot analysis and verified by backcrossing. Null DJ-1 homozygosity was determined by PCR against the neo gene and western blot analysis of mouse tail protein samples as described in the previous section. In some mice, immunoblotting of brain tissue extracts with 691 and LB509 antibodies were used to confirm the loss of DJ-1 protein and the expression of human pathogenic α-syn protein, respectively. Mice were sacrificed by CO2 euthanization as approved by the University of Pennsylvania Institutional Animal Care and Use Committee.
Mice were sacrificed with CO2 euthanization and perfused with PBS/heparin, followed by perfusion with either 70% ethanol/150 mm NaCl or PBS-buffered formalin. The brain and spinal cord were then removed and fixed for 24 h in the respective fixatives used for perfusion. As previously described tissues were dehydrated at room temperature through a series of ethanol solutions, followed by xylene and then were infiltrated with paraffin at 60°C (104). The tissues were then embedded into paraffin blocks which were then cut into 6 µm sections. Immunostaining of the sections was then performed using previously described methods (105).
Paraffin-embedded tissue sections were deparaffinized and hydrated through a series of graded ethanol solutions followed by 0.1 m Tris, pH 7.6. The sections were incubated simultaneously with the Syn 303 and anti-TH primary antibodies diluted in 5% dry milk/0.1 m Tris, pH 7.6. After extensive washing, sections were incubated with goat anti-mouse secondary conjugated to Alexa 488 and goat anti-rabbit secondary conjugated to Alexa 594. After washing, the sections were coverslipped with VectaShield-DAPI mounting medium (Vector Laboratories, Burlingame, CA, USA) and visualized using an Olympus BX51 microscope.
The number of anti-TH immunoreactive neurons in the entire substantia nigra was assessed for WT, M83, DJ-1 null and M83-DJnull mice at ages 6 and 12 months using similar methods described by Kitada et al. (88) with a few exceptions. Brains were fixed in 70% ethanol/150 mm NaCl and sectioned at 10 µm thickness. Every fifth section was stained with anti-TH polyclonal antibody (Millipore). Three brains were analyzed for each of the indicated mouse strains at each of the respective ages.
Paraffin-embedded tissue sections were incubated with antibody Syn303 followed by anti-mouse conjugated Alexa Fluor 594 as described above. Sections were stained with thioflavin-S by immersing in freshly made 0.0125% thioflavin-S/40% EtOH/60% PBS and differentiated in 50% EtOH/50% PBS. The sections were coverslipped and visualized as described above.
Cerebral cortical tissues were harvested from 6-month-old DJ-1 null, DJ-1 Het and WT mice. Tissues were sonicated in three tissue volumes of 2% SDS/8 m urea. Total protein extracts were then quantified using the bicinchoninic acid (BCA) assay (Pierce Biotechnology, Rockford, IL, USA) and bovine serum albumin as the standard. Equal amounts of protein extracts were resolved by SDS–PAGE and analyzed by western blot.
The cortex, cerebellum, brainstem and spinal cord were dissected from mice. For each mouse, the brainstem and spinal cord were pooled together and then all tissues were weighed and homogenized with a pellet pestle motor in three tissue volumes of HS buffer [50 mm Tris (pH 7.5), 750 mm NaCl, 5 mm EDTA, with a protease inhibitor cocktail at 1:1000 and PMSF at 1:500, and the phosphatase inhibitors, 20 µm NaF, 1 µm NaVO4, and 1 µm okadaic acid] followed by sedimentation at 100 000g for 20 min. Supernatants were saved as the HS fraction. Pellets were homogenized in three tissues volumes of HS-TX buffer (HS buffer with 1% Triton X-100) and sedimented at 100 000g for 20 min. The supernatants were saved as the HS-TX fraction. The pellets were subjected to a sucrose myelin float by homogenizing in three pellet volumes of sucrose buffer (HS buffer/1 m sucrose) and after centrifugation, the myelin-rich supernatants were discarded. Pellets were then homogenized in two tissue volumes of RIPA buffer [50 mm Tris (pH 8.0), 150 mm NaCl, 5 mm EDTA, 1% NP40, 0.5% sodium deoxycholate and 0.1% SDS with the protease and phosphatase inhibitors], sedimented at 100 000g for 20 min, and the supernatants were saved as the RIPA fraction. Pellets were then sonicated in one pellet volume of 2% SDS/8 m urea and were designated to be the SDS fractions. The fractions were then quantified using the BCA assay, resolved by SDS–PAGE and analyzed by immunoblot as previously described above.
Striatal catecholamine content was assessed in WT, M83, DJ-1 null and M83-DJnull mice. Mice were analyzed at 6 and 12 months of age. In order to isolate the striatum, the brains were submerged into ice-cold phosphate buffer and dissected into 1 mm-thick coronal sections using a vibratome (Series 1000, R.L. Slaughter, Essex, UK). Dissected tissues were weighed and homogenized by sonication in 10 volumes of 1 µm 3,4 dihydroxybenzylamine (DHBA) in 0.1 m perchloric acid. Following centrifugation at 16 000g for 10 min at 4°C, 20 µl of the supernatants were injected onto an 1100 series Agilent HPLC system controlled by Chemstation software (Agilent, Palo Alto, CA, USA). The mobile phase consisted of 72 mm citric acid, 28.4 mm sodium phosphate, 2% methanol, pH 2.8. Catechols were resolved at a flow rate of 1 ml/min on a reverse-phase C18 Luna column (150 × 4.6 mm, 5 µm; Phenomenex, Torrance, CA, USA) and detected with a Coularray detector (ESA Biosciences, Chemsford, MA, USA) with the following working potentials (in mV): −200, +50, +300 and +400. Chemstation software (version 1.04, ESA Biosciences) was used for the quantification by comparing to the peak areas of known concentrations of standards. Acid-precipitated protein pellets were extracted in 2% SDS/50 mm Tris–Cl, pH 7.4 and the protein concentrations were determined using the microBCA kit (Peirce, Rockford, IL, USA). Analyte levels were normalized to DHBA and protein concentration.
This work was supported by the National Institute on Aging [AG09215] and the National Institute of Neurological Disorders and Stroke [NS053488]. C.P.R. is supported by a pre-doctoral NRSA fellowship from the National Institute on General Medicine Sciences [GM082026].
We would like to thank Dr Thomas Chou for his help in the proteomics studies of α-syn transgenic mice.
Conflict of Interest statement. None declared.