Subcellular localization of wild-type and Parkinson's disease-associated mutant alpha -synuclein in human and transgenic mouse brain

Abstract

Mutations in the alpha-synuclein (alphaSYN) gene are associated with rare cases of familial Parkinson's disease, and alphaSYN is a major component of Lewy bodies and Lewy neurites. Here we have investigated the localization of wild-type and mutant [A30P]alphaSYN as well as betaSYN at the cellular and subcellular level. Our direct comparative study demonstrates extensive synaptic colocalization of alphaSYN and betaSYN in human and mouse brain. In a sucrose gradient equilibrium centrifugation assay, a portion of betaSYN floated into lower density fractions, which also contained the synaptic vesicle marker synaptophysin. Likewise, wild-type and [A30P]alphaSYN were found in floating fractions. Subcellular fractionation of mouse brain revealed that both alphaSYN and betaSYN were present in synaptosomes. In contrast to synaptophysin, betaSYN and alphaSYN were recovered from the soluble fraction upon lysis of the synaptosomes. Synaptic colocalization of alphaSYN and betaSYN was directly visualized by confocal microscopy of double-stained human brain sections. The Parkinson's disease-associated human mutant [A30P]alphaSYN was found to colocalize with betaSYN and synaptophysin in synapses of transgenic mouse brain. However, in addition to their normal presynaptic localization, transgenic wild-type and [A30P]alphaSYN abnormally accumulated in neuronal cell bodies and neurites throughout the brain. Thus, mutant [A30P]alphaSYN does not fail to be transported to synapses, but its transgenic overexpression apparently leads to abnormal cellular accumulations.

Fig. 1.

Selective detection of αSYN and βSYN.A, A Western blot prepared from 5 μg of purified human brain αSYN and βSYN was probed with antiserum 6485 (left) and then stripped and reprobed with antiserum 6482 (right). Antiserum 6485 specifically recognized βSYN, whereas antiserum 6482 raised against the C terminus of αSYN showed some cross-reactivity toward βSYN. The positions of prestained molecular weight markers are indicated to the left of each blot. B, Western blots were prepared from 5 μg (top 2 panels) or 2 μg (bottom 3 panels) of purified recombinant human αSYN, βSYN, and γSYN. C, Heat-stable supernatant of mouse (mus) or human (hum) brain cytosol (100 μg) was lyophilized and subjected to 15% SDS-PAGE. Blots were probed with the antibodies specified to the right.

Fig. 2.

Subcellular localization of synucleins in brain. Sucrose gradient (20–48%) floatation assays were performed with 200 μl of mouse postnuclear supernatant directly (A) and after 16,000 × gcentrifugation (C). Fractions were separated by SDS-PAGE (12.5%), and Western blots were sequentially probed for αSYN (top panel), βSYN (middle panel), and synaptophysin (bottom panel). Two additional experiments revealed the same results. B, Sucrose gradient (30–48%) floatation assays were performed with postnuclear supernatant from rapidly processed temporal cortex gray matter of epilepsy patients after lobotomy. Fractions were subjected to Tris–tricine-PAGE (4–20% gradient), and Western blots were sequentially probed with Mc42 anti-αSYN (top panel) and anti-synaptophysin (bottom panel). ECL was used as chemiluminescence substrate. Tissue from an additional patient revealed the same result.D, Schematic representation of the subcellular fractionation steps. The postnuclear supernatant of one mouse brain (E) (representative for 3 independent experiments) or biopsied temporal cortex gray matter of an epilepsy patient (F) was subjected to subcellular fractionation. SDS-PAGE (12.5%) was performed with 50 μg (only 25 μg were available of the synaptosomal pellet P2) (E) or 20 μg (only 10 μg were available of the synaptic vesicle pellet LP2) (F) of each fraction. The corresponding Western blot was sequentially probed for αSYN (top panels; Mc42 in E, 15G7 inF), βSYN (middle panels), and synaptophysin (bottom panels). ECLplus was used as chemiluminescence substrate in F.

Fig. 3.

Colocalization of αSYN and βSYN in human cerebellum. Antibodies 15G7 anti-αSYN (A), 6485 anti-βSYN (B), and anti-synaptophysin (C) all stained a punctate pattern in the molecular layer. The immunoreactivity in the granule cell layer showed a patchy distribution corresponding to labeling of cerebellar glomeruli. Scale bar: A–C, 100 μm. Double-labeled immunofluorescent confocal microscopy revealed a colocalization of αSYN (D) and βSYN (E) in the cerebellar glomeruli of the granule cell layer, which resulted in a yellow signal in the superimposed digital picture (F). A similar colocalization is seen with synaptophysin (G) and βSYN (H). I, Superimposed digital picture. Scale bar: D–I, 10 μm.

Fig. 4.

Expression of [A30P]αSYN in transgenic mouse brain. A, Schematic drawing of the transgenic construct (not drawn to scale). Hatched box, mThy-1.2 promoter region. Open boxes, Thy-1 exonic sequences; (truncated) exon IV contains the polyadenylation signals. Solid line, Thy-1 intron A. Start and stop codons of the open reading frame for human [A30P]αSYN (filled box; *, A30P mutagenesis site) are directly flanked by XhoI restriction sites (X). Dashed line, 3′-region of the Thy-1 gene. N,NotI restriction sites used to linearize construct and remove vector sequences before microinjection. B, A mixture of two probes specific for the human αSYN transgene and a probe for the mouse β-actin gene was hybridized to a Northern blot of poly(A⁺) RNA from [A30P]αSYN mice, as indicated. The sizes of the transcripts were ∼1.8 and 2.1 kb, respectively.C–E, Lyophilized heat-stable supernatants of whole brain cytosol (200 μg) from 6- to 10-week-old individuals of the indicated [A30P]αSYN mouse lines were subjected to 15% SDS-PAGE. Equal loading was demonstrated by Brilliant blue staining of the gels after transfer. Western blots were sequentially probed with the mouse-specific antiserum 7544 (C), Mc42 (D), and the human-specific antibody 3400 (E). Bands were quantified by densitometric scanning (bottom panels). The data are representative for at least three animals per line screened for [A30P]αSYN protein expression. αSYN-immunoreactive double bands (asterisks) comigrating with the 29 kDa standard, a position consistent with the molecular mass of a dimer, were observed at variable intensity. These putative dimeric species could be well resolved in large gels (P10DS; Owl Separation Systems, Portsmouth, NH) (F). Whole brain cytosol (600 μg) from a 10-month-old line 18 mouse was directly subjected to 15% SDS-PAGE, without any concentration step. Western blot was probed with antibody 3400.

Fig. 5.

Synaptosomal localization of [A30P]αSYN in transgenic mouse brain. Postnuclear supernatant from a 4-month-old male line 18 mouse was processed as described in Figure2A. The Western blot was sequentially probed with antiserum 3400 (top panel), Mc42 (middle panel), and anti-synaptophysin (bottom panel).

Fig. 6.

Distribution of transgenic and endogenous αSYN in cerebellar sections. The human-specific antibody 15G7 did not react with nontransgenic mouse cerebellar sections (A) but showed strong labeling of the molecular layer in [wt]αSYN mice (B) and line 31 [A30P]αSYN (C) mice. In addition, diffuse cytosolic immunoreactivity was observed in Purkinje cells of both transgenic mice. The mouse-specific antiserum 7544 did not show a specific signal with human cerebellar sections (D). A normal synaptic staining pattern of endogenous αSYN with labeling of the molecular layer and cerebellar glomeruli was observed in [wt]αSYN mice (E) and line 31 [A30P]αSYN mice (F). Note that no cytosolic staining was observed with the antiserum 7544 in the transgenic mice.

Fig. 7.

Accumulation of human αSYN in neuronal cell bodies and neurites of transgenic mouse brain. Abnormal accumulation of human [A30P]αSYN was detected in cell bodies and bulbous neurites in most brain regions, including frontal cortex [A(line 31) and B (line 18), 15G7]. Some [A30P]αSYN-filled neurites (arrows) emanated from neuronal cell bodies with accumulated [A30P]αSYN (A,B). In contrast, expression of βSYN (C, antiserum 6485) and endogenous αSYN (D, antiserum 7544) was restricted to the neuropil and was not found in neuronal cell bodies and neurites accumulating [A30P]αSYN. Pathological bulbous αSYN-positive neurites (arrows) could also be observed in the dentate nucleus of [wt] (E) and [A30P] [line 31 (F) and line 18 (G); arrows] mice (15G7 staining). H, Lewy neurites (arrows) stained with 15G7 in the hippocampal CA2/3 region of a patient with LB dementia.