Alzheimer's disease (AD) and Parkinson's disease (PD) are the two most common age-dependent disorders of the central nervous system 1 conferring a detrimental neurodegenerative phenotype on the individual. In both diseases a subset of cases has been linked to autosomal dominant mutations in specific genes. While in inherited cases of AD, genetic mutations in amyloid precursor protein (APP) and presenilins (PS1 and PS2) lead to the pathological features of extracellular Aβ plaques and intracellular neurofibrillary tangles 2, mutations in the α-synuclein gene (Snca) confer familial PD, and its protein product is a component of the pathological hallmark, the intracellular Lewy bodies 34. More than a decade ago evidence emerged that the Aβ plaques contain a peptide then termed "non-amyloid component of Aβ plaques" (NAC) 56, which was later identified as a fragment of α-synuclein. The 35 amino acid fragment found in the plaques was determined to be extremely hydrophobic 7, and multiple lines of evidence indicated that α-synuclein aids the aggregation of Aβ42 in vitro and might be involved in the formation of Aβ plaques in the brains of patients with AD and dementia with Lewy bodies (DLB) 8. Studies by Yang et al. 9 suggested that α-synuclein was the major component accumulating in the neurites of aged Tg2576 APP transgenic mice and suggested that this aggregation might be caused by Aβ peptides. Other studies have generated in vivo data using overexpression of APP, leading to an increase of hydrophobic accumulations in a mouse model also overexpressing α-synuclein 10. Together these published data indicate that α-synuclein might interact with other fibrillogenic proteins and enhance their accumulation and in return is positively influenced to aggregate as well. However, this detrimental function of α-synuclein has also been called into question as a few publications failed to establish any association between α-synuclein and Aβ pathology 11.

Although multiple α-synuclein transgenic lines have been created, which mimic some aspects of PD pathology 12, no studies have been reported on how the lack of α-synuclein affects the progression of neurodegenerative diseases. Several lines of α-synuclein (Snca) knockout mice have been established, most of which show only a very mild phenotype, suggesting genetic redundancy 131415. These studies suggested that α-synuclein is not an essential component of the neuronal system under normal circumstances or that other proteins may compensate for the loss of α-synuclein 16.

Based on published reports, we hypothesized that α-synuclein in an AD mouse model overexpressing APP would likely support the aggregation of Aβ plaques and its absence would result in a decreased number of plaques and a belated onset of pathology. We thus bred the Snca knockout mice 14 with Tg2576 17 mice overexpressing mutant human APP that develop amyloid plaque pathology at 9–10 months of age, and the number of plaques increased dramatically with age 17.

In our study we used aged Tg2576 animals, herein referred to as APP animals, and compared their plaque pathology to that of APP animals lacking α-synuclein (APP/Snca^-/-) at various ages. In general the mice are indistinguishable from their littermates in terms of appearance and viability. In order to determine whether we could detect differences in the production of two main Aβ species we performed Sandwich ELISA experiments at an early age. At 6 months of age neither the APP nor the APP/Snca^-/-exhibit Aβ plaque pathology (data not shown). As expected, no difference in Aβ40 or Aβ42 levels could be detected by Sandwich ELISA (Figure 1) when comparing 6-month-old APP and APP/Snca^-/- animals. We then focused our study to determine the effects of α-synuclein on the onset of plaque deposition, and the progression of the pathology by comparing these parameters between APP and APP/Snca^-/-.

In order to determine whether the age of onset of plaque deposition is affected, we studied 9-month-old APP and APP/Snca^-/- animals as that is the earliest time when plaques can be detected in APP mice. Using immunohistochemistry with 6E10 antibody to human APP, we detected a slight increase in the number of plaques (Figure 2A), but the difference failed to reach statistical significance (Figure 2B). The number of plaques increased in both APP and APP/Snca^-/- mice at 12 months (Figure 2). Similar to 9-month-old animals, there was a slight, but not statistically significant, difference between the two genotypes in the number or size of plaques (Figure 2B). As a homologous protein to α-synuclein, we studied the protein levels of β-synuclein as well as another synaptic vesicle protein, synaptophysin, in the brain homogenates of APP and APP/Snca^-/- mice. We were unable to detect significant differences in the expression of β-synuclein or synaptophysin in brain homogenates, limiting the possibility that these presynaptic proteins play a compensatory role for the loss of α-synuclein (Figure 2C). Overall, it appears that α-synuclein is not a large player in initiation/seeding of the plaque pathology in APP mice.

At 18 months of age, however, differences in the number of plaques became obvious. When performing 6E10 and thioflavine S staining of comparable coronal sections, we observed a vastly increased number of plaques in the APP/Snca^-/- animals when compared to APP animals (Figure 3A). Because the number of plaques in APP/Snca^-/- mice reached a level that was difficult to be counted manually, we chose to quantify the plaque density by assessing the signal intensity of matched areas. Statistical analysis revealed that both total (6E10-positive) and dense-core plaque (thioflavin S-positive) load in the APP/Snca^-/- animals were increased 3–4 fold when compared to the number of plaques found in age-matched Tg2576 animals (p < 0.001).

As expected, Western blot analysis showed that α-synuclein protein was missing from the brain homogenates of the APP/Snca^-/- animals (Figure 4A) and the levels of full-length APP were not affected by the absence of α-synuclein. Once again we were unable to detect an increase in the β-synuclein (Figure 4A). Unexpectedly, we found that synaptophysin, which was unchanged at 12 months, was up-regulated by two-fold in APP/Snca^-/- animals when compared to APP animals (Figure 4B, p < 0.01). This could be attributed to a compensatory mechanism activated due to α-synuclein deficiency. However, the reason for the selective increase of synaptophysin, but not β-synuclein, is not clear.

Genetic mutations and gene amplification of α-synuclein (Snca) are linked to familial Parkinson's disease, and aggregation of the protein is the pathological hallmark of the disease. Numerous studies have focused on the effect that the overexpression of α-synuclein has on the dopaminergic system. In this pathogenic state, even as little as a two-fold increase in protein expression leads to the pathological feature of PD 12. However, its normal function is much less understood. Early studies of knockout animals lacking either α- or β-synuclein or both revealed little impact on the mammalian system. Mice lacking the protein are phenotypically normal, and only close investigation reveals slight deficits such as a reduction in dopamine levels 1213.

It is well established that α-synuclein localizes to the presynaptic terminals and associates with synaptic vesicles. In addition the protein has also been shown to preferentially associate with lipid rafts at these sites 1819. Lipid rafts are enriched in sphingolipids and cholesterol and are the site of localization for the γ-secretase complex 17, which is the enzyme responsible for the intramembranous cleavage of APP leading to the production of Aβ40 and Aβ42, among others 20). Since α-synuclein localizes to the exact microdomain where APP processing takes place 17, it is a distinct possibility that the two proteins interact. Furthermore, research has suggested that α-synuclein expression is up-regulated in AD cases 21. Overall these papers suggest that α-synuclein may be the primary insult leading to aggregation of the protein and subsequent neurodegeneration. While there exist several contradictory reports as to whether α-synuclein associates with Aβ containing plaques 9 it is likely that α-synuclein plays a contributing role in AD, PD and other diseases.

Over the years several publications have called into question the direct correlation between plaque and Lewy body pathology in AD and PD and the respective neuronal loss 22). In addition, numerous papers recently suggested that a small increase in α-synuclein expression is the cellular response to oxidative stresses and confers a higher viability to the cells 2324. This sheds light on a different, neuroprotective property of α-synuclein in which the α-synuclein indeed is called upon in the times of cellular distress to help clear the insult 25. In fact, our data that the amyloid plaques are increased in the absence of α-synuclein at old age suggests that α-synuclein may indeed serve as a chaperone helping the cells to clear protein deposits. While we were unable to detect any effect on the onset of plaque deposition in the APP animals, the distinct effect that α-synuclein has at a later stage of the pathology (18 months) is striking. Hence, α-synuclein does not appear to be involved in the onset of plaque deposition, but it is clearly involved in prevention of further accumulation in later stages of the pathology, maybe in clearing the Aβ fragments, possibly from the site of generation, to prevent further pathological development. Although many studies have suggested that α-synuclein is the culprit leading to neuronal cell death, it seems equally likely that under physiological conditions the protein presents a deterrent to the accumulation of other proteins whose accumulation might be more toxic to the cell. To date, α-synuclein has been shown to interact with a variety of proteins and is involved in a variety of diseases such as Parkinson's disease, AD, as well as Sandhoff disease 26, and it appears that the protein might play a more basic role in cellular defense from diseases. The role of the soluble protein might be to identify toxic proteins and bind to them to induce a cellular defense mechanism. In this scenario, under pathological conditions the amount of α-synuclein-containing aggregates is simply too much for the cellular machinery to handle, leading to intracellular aggregates as in the case of PD and Sandhoff disease or extracellular aggregates as in the case of AD. While many studies over the past year have indicated that α-synuclein has a protective role in neurons in vitro, to our knowledge this is the first report investigating the effect that the absence of α-synuclein has on the disease progression under pathological circumstances. There is no doubt that the aggregation of an otherwise neuroprotective protein could lead to pathological conditions if its natural response leads to protein aggregation and up-regulation of stress induced pathways. It is possible that the fact that α-synuclein was first identified as the protein whose overexpression leads to Parkinson's disease has masked its true role as a neuroprotector.

Our study investigates the effect of the absence of α-synuclein in the progression of a pathological hallmark of AD. In this study we identified α-synuclein as a protein influencing the progression of plaque deposition in a transgenic mouse model of AD. Our data support a protective role of α-synuclein against amyloid pathology at a later stage, and offer an alternative explanation to the aggregation of α-synuclein under pathological conditions.

AD: Alzheimer's disease; PD: Parkinson's disease; Aβ : β-amyloid; APP: amyloid precursor protein; PS: presenilin; DLB: Dementia with Lewy Bodies.

The author(s) declare that they have no competing interests.

VK generated the transgenic mice, carried out biochemical and immunohistochemical analyses, and wrote the first draft of the manuscript. EP assisted VK in the immunostaining, performed quantification of plaque load, and edited the manuscript. HZ is responsible in designing and overseeing the experiments and, together with VK, wrote the manuscript.