Are changes in synaptic proteins associated with amyloid beta deposition in transgenic mice and with dementia states in man?
By Johanna O’Connor
Research into dementia is necessary to aid the development of diagnostic tools as well as treatments and preventions for this debilitating condition. Several causes for dementia have been proposed and the amyloid cascade hypothesis is the leading theory for the pathophysiology of the dementia symptoms seen in Alzheimer’s Disease (AD), however, similar symptoms are seen in other dementias without the presence of senile plaques. Other features of dementia include synaptic loss which results in impaired neuronal transmission and may be the causative factor for symptoms such as memory loss and impaired cognition in patients with AD. Synaptic loss is not unique to AD and may be a contributing factor in other types of dementias such as Parkinson’s Disease Dementia (PDD) and dementia with Lewy Bodies (DLB). Both PDD and DLB are forms of α-synucleinopathies and are characterised by Lewy bodies (α-synuclein aggregates).
To understand the cellular mechanisms involved in synaptic loss several synaptic proteins were investigated in both a transgenic mouse model of AD and in human post-mortem tissue from AD, PDD, DLB and aged control patients. The following synaptic proteins were labelled in the transgenic mouse model of AD: SNAP25, PSD95, synaptophysin, neurogranin, SV2C and SNAP47. Synaptophysin levels were significantly reduced in the transgenic mouse model with an increase in Aβ deposition (p=0.022). There were no differences in SNAP25, PSD95, neurogranin, SV2C and SNAP47 between the young and old transgenic mice. In the human studies, SNAP47 labelling was shown to be significantly different between the AD, PDD and DLB patients, compared to the aged control group for the number (p=0.002) and area (p=0.001) of labelled features. When comparing individual dementia groups against control, AD showed a significant increase in SNAP47 labelling for both number (p=0.039) and area (p=0.038) of labelled features. There were no significant changes found when comparing SNAP47 labelling in PDD or DLB directly with control, there was also no significant correlations found when comparing SNAP47 scores with cognitive decline.
In England, 670,000 people live with dementia and this figure is believed to double by 2030 1. According to the WHO, dementia affects 47.5 million people worldwide and it is believed that this will increase to 75.6 million by 2030 2. Dementia is a progressive condition and not a normal part of ageing; it results in a decline in cognitive function that affects the following: memory, comprehension, language, planning, calculating, motor skills, social skills and behaviour. The economic impact of dementia in England is huge and costs over 19 billion pounds a year – greater than the cost of cancer, heart disease or stroke 1.
Synaptic transmission is essential for healthy nervous system function. Progressive synaptic loss is believed to be the cause of dementia and can be characterised by quantifying specific synaptic proteins or synaptic sites 3. Synaptic loss is also the best correlation for the severity of Alzheimer’s Disease (AD) and therefore quantifying specific synaptic proteins in demented / healthy patients may prove useful in understanding the mechanism of synaptic loss and ultimately the pathophysiology of dementia 4. Little is known, on the correlation of synaptic loss with other dementias such as Parkinson’s Disease dementia (PDD) and dementia with Lewy Bodies (DLB) although synaptic loss is known to occur independently of Aβ plaques, a main feature of AD. The synaptic proteins investigated in this paper are synaptosome-associated protein of 25,000 daltons (SNAP25), synaptophysin, synaptic-vesicle glycoprotein 2C (SV2C), synaptosome-associated protein of 47,000 daltons (SNAP47), neurogranin and post-synaptic density protein 95 (PSD95), see figure 1. The presynaptic proteins (SNAP25, SNAP47, synaptophysin and SV2C) are involved in vesicle fusion and subsequent exocytosis.
The postsynaptic proteins, neurogranin and PSD95 are involved in long-term potentiation (LTP) and synaptic plasticity which in turn is important for spatial and emotional learning 5, 6.
Figure 1: An illustration of the pre and post synaptic proteins studied in this paper. Synaptophysin and SV2C are synaptic vesicle proteins whilst SNAP25 and SNAP47 are associated with the Soluble NSF Attachment Protein Receptor (SNARE) complex and thus the docking and exocytosis of synaptic vesicles. PSD95 is a postsynaptic scaffolding protein associated with NMDA and AMPA receptors and neurogranin is a postsynaptic calmodulin binding protein. NMDA: N-Methyl-D-Aspartate receptor. AMPAR:α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor. 7, 8, 9, 10
Popular Hypotheses for Alzheimer’s Disease
Research into the pathophysiology of dementia has largely remained focused on AD as this is the most common form of dementia making up 60-80% of all dementia cases. There are three popular hypotheses for the pathophysiology of AD, the amyloid cascade hypothesis, the tau hypothesis and the cholinergic hypothesis.
Since the early 90’s, the amyloid cascade hypothesis has played a prominent role in understanding the pathophysiology of AD and for the targeting of novel therapeutics. The amyloid cascade hypothesis posits that amyloid-beta deposition is responsible for the pathological events that occur in AD: senile plaques, neurofibrillary tangles, neuronal cell death and ultimately dementia. However, senile plaques and neurofibrillary tangles can develop independently of amyloid-beta and randomised clinical trials that have tested drugs or antibodies targeting components of the amyloid pathway have remained inconclusive 11.
The tau hypothesis, on the other hand, states that abnormal phosphorylation of tau such that neurofibrillary tangles and paired helical filaments are formed results in the destabilisation of the neuronal cytoskeleton and subsequent neuronal injury 12. However, clinical trials of drugs targeting tau aggregation remain unfruitful, despite several reaching Phase 3 clinical trials 13.
The oldest hypothesis for dementia is the cholinergic hypothesis. The cholinergic hypothesis suggests that it is the deficiency in acetylcholine that causes the symptoms and deterioration seen in AD. Early therapeutic research was based on this hypothesis and acetylcholinesterases were used as first line anti-Alzheimer’s drugs. But, increasing acetylcholine in the brain does not reverse Alzheimer’s progression, nor does it prevent it.
This paper will look at the distribution of synaptic proteins in a transgenic mouse model of AD and in human post-mortem tissue of AD, PDD, DLB and age controlled patients in the hopes of understanding the pathophysiology of dementia and its potential therapeutic targets.
Materials and Methods
As a preliminary to the study of synaptic proteins in dementia, immunohistochemical (IHC) analyses of these proteins were conducted on tissue from a TAS10 x TPM (TASTPM) transgenic mouse model of AD. TASTPM mice were used as they demonstrate an increase in Aβ deposition with age due to the expression of human familial mutations in amyloid precursor protein and presenilin-1.
Sections used for the experiments described in this paper were obtained from TASTPM mice and from human brain tissue. TASTPM develop age-related amyloid pathology, memory deficits, gliosis and minimal neuronal loss but do not show synaptic loss, tangles or changes in long term potentiation (LTP) / long term depression (LTD). The TASTPM mice were provided by Glaxo Smith Kline. The human brain tissue used was supplied by the Thomas Wills Oxford Brain Collection and the London Neurodegenerative Disease Brain Bank which are part of the Brains for Dementia Research Network. A total of 5 Alzheimer patients, 6 Dense Lewy body patients, 5 Parkinson’s disease dementia patients, and 5 control patients were used for this investigation. Samples of human tissue used were from the cortical regions of the brain; neuropathological assessments were conducted by a consultant neuropathologist. Clinical classification of PDD was based on parkinsonism presenting at least a year before dementia, whilst DLB was classified as cognitive impairment/hallucinations a year or more before parkinsonism. All cases were diagnosed by experienced clinicians using validated clinical rating systems. AD cases were diagnosed using the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD), DLB according to international consensus criteria and PDD according to Movement Disorders Society criteria 14, 15. Cognitive impairment was measured by the Mini Mental State Examination (MMSE) which is a yearly assessment performed on the onset of dementia (but also used to screen for dementia). The final MMSE is the last MMSE performed before the death of the patient.
Primary antibodies used for investigation of synaptic proteins
|Glaxo Smith Kline||Mouse Monoclonal||1:10000|
|Cathepsin D||Santa Cruz Biotechnology||Goat Polyclonal||1:500|
Table 1: A list of antibodies used to stain for synaptic proteins during IHC investigations in TASTPM and human investigations. 488 / 568 Alexa Fluor used depending on experiment. Vector Laboratories, Inc. provided the DAB and NovaRED kit.
Post-mortem human brain samples for IHC
|Source||Disease||Age||Gender||Braak Stage||MMSE Final|
Table 2: Shows the human cases used for the SNAP47 investigation. Overall there were 21 patients used which included controls, AD, DLB and PDD patients.
Brain sections (7um) used were provided embedded in wax and thus were dewaxed using the Leica Autostainer XL (Leica Microsystems, Milton Keynes, UK). Antigen retrieval was then performed by submerging sections into citrate buffer (pH 6) and microwaving for a total of 20 minutes. After removing the sections from the autoclave they were replaced in the Leica Microsystem Autostainer to be washed in primary buffer solution (PBS). Once washed the sections were ready to receive relevant primary antibodies diluted in primary buffer (containing PBS, bovine antibody and triton) required to perform either fluorescent staining or visualisation with DAB or NovaRED. The sections were then incubated with primary antibody overnight at 4 ̊C. (For a detailed description of the protocol please refer to Appendix 3)
Fluorescent staining was achieved by directly applying the same species Alex Fluor to the primary antibody and waiting between 1.5 and 2 hours, or by applying the same species biotin to the primary antibody. Biotinylated sections were incubated for 1 hour before applying Streptavidin Alexa Fluor for 1.5 – 2 hours. After incubating with Alexa Fluor the sections were washed in PBS, immersed into Fisher Chemical Sudan Black at 0.1% in ethanol and then rinsed in distilled water. Finally, the sections were mounted with coverslips using Vectamount Aqueous mounting medium ready for imaging.
Diaminobenzidine immunohistochemistry / Novared
Appropriate biotinylated secondary antibodies diluted in secondary buffer (containing PBS and triton) were added to the sections and left to incubate for 1 hour. After the PBS wash Vector ABC solution was added at 80μl to each section and left to incubate for another hour. After incubation sections were ready to undergo either DAB or NovaRED preparation:
- DAB solution was added to each section and colour development ranged between 2-10 minutes
- NovaRED solution was added and colour development ranged between 5-15 minutes
After correct colour development, the sections were washed and dehydrated. The sections were then cover-slipped and left overnight before imaging.
All imaging was carried out and captured using the Leica Microsystems DMRB microscope fitted with a DFC420C digital camera. Alexa Fluor 488 produced a yellow / green fluorescence under a 488nm wavelength (blue light), Alexa Fluor 568 produced a red fluorescence when viewed under a 568nm wavelength (green light). For sections stained with DAB or NovaRED light microscopy was used. Images were focused by altering the exposure adjust and the histogram settings, however, all images used for comparison were captured using the same settings.
Images taken from mouse sections were taken at the hippocampal region whilst images from human tissue was taken from cortical regions (e.g. BA9, BA24).
To assess amyloid beta deposition in the mouse model a scoring system was used. The system involved scoring the amount of fluorescence seen in blinded sections to a scale of 0 to 3 (Figure 2).
The grading scale:
- 0 = no fluorescence
- 1 = little fluorescence
- 2 = moderate fluorescence
- 3 = heavy fluorescence
The SNAP47 study in humans was analysed using computer-assisted software – Image J. SNAP47 labelling allowed for computer-assisted analysis as it labelled astrocytes (defined by the software as particles) which could then be quantified using the ImageJ programme. This software calculated the number of particles in each image, the size of each particle as well as the total area of all particles and allowed for a more objective analysis of the data. Ten images were quantified per section. For ImageJ analysis, a threshold of >30 pixels was applied in to remove background fluorescence (e.g. minute dots) from the data.
Results obtained from the TASTPM experiment were inputted into SPSS where descriptives, as well as a One-Way ANOVA were performed. Results obtained from ImageJ were combined with patients details from the Brain Research UK database, descriptives and correlations were made using SPSS.
The TASTPM model was used to identify whether changes in Aβ deposition resulted in alterations in the following synaptic proteins: SNAP25, PSD95, neurogranin, SV2C, SNAP47 and synaptophysin. To compare the effects of Aβ deposition and synaptic protein alterations the mice were divided into two groups; mice less than 40-weeks old were used as a model of little to no Aβ deposition and were classified as young mice, whilst mice of greater than 20-months old were used as a model of high Aβ deposition and classified as old mice. The distribution of young (<40 weeks) and old (>20 months) mice for each synaptic protein investigation are shown in Table 3.
Table 3: Shows the distribution of young (<40 weeks) and old (>20 months) TASTPM mice in each of the synaptic protein experiments with the percentage distribution given alongside.
|Synaptic Protein Investigation||<40 weeks||>20 months|
|PSD95 n=12||8 (75.00%)||4 (25.00%)|
|SNAP25 n=13||8 (61.54%)||5 (38.46%)|
|Neurogranin n=16||5 (31.25%)||11 (68.75%)|
|Synaptophysin n=18||6 (33.33%)||12 (66.66%)|
|SV2C n=7||3 (42.86%)||4 (57.14%)|
|SNAP47 n=6||3 (50.00%)||3 (50.00%)|
Fluorescent labelling for SNAP25, PSD95, neurogranin, SV2C and SNAP47 between the young and old mice did not differ significantly. However, synaptophysin labelling in older mice decreased when compared to the young group of mice (p=0.002), as shown in Figure 3.
The IHC protocols that were used in the TASTPM mice were then transferred to the human post-mortem tissue. Quantifiable labelling was only observed with the SNAP47 antibody in the human tissue. A total of 21 different patients were used for the study, classified as AD, PDD, DLB or aged control patients. Both mean particle count of SNAP47 labelling and mean percentage area of SNAP47 labelling showed statistically significant differences between AD and control patients (Figure 4 & 5). However, SNAP47 labelling in different dementia states did not reveal any significant associations with final MMSE or Tau Braak stages.
Figure 4: Average Particle Count in the Different Dementia States shows the mean particle count of SNAP47 in the cortical regions of the brain. A One-Way ANOVA was performed and showed that the differences between the disease groups seen in this graph are statistically significant p=0.002. Statistical significance found between the Alzheimer’s Disease state and control p=0.039. Error bars: + / 1 standard error.
Figure 5: Average SNAP47 staining as an area percentage shows the mean area as a percentage of the amount of SNAP47 staining in the images analysed with ImageJ. A One-Way ANOVA was performed and showed that the differences in percentage area stained with SNAP47 in the different disease states are statistically significant P=0.01. Statistical significance found between the percentage area of SNAP47 labelling between Alzheimer Disease state and control (OneWay ANOVA: p=0.038). Error bars: +/- standard error.
Synaptic proteins in the TASTPM model
Synaptic loss plays an important role in AD, however, it would be useful to know whether changes in specific synaptic proteins are a feature of dementia or associated with severity. By investigating the levels of synaptic proteins in the TASTPM model of AD it is possible to investigate the sole effect of Aβ deposition on synapses.
In the late 80’s, Hammos et al revealed that synaptic loss was significantly associated with the progression of AD 16. Despite, over 20 years have passed since this discovery and the pathophysiology of AD is still unknown. More recent studies have focused on the roles of specific synaptic proteins to discover how exactly AD progression occurs. Clinical studies of synaptophysin have revealed that its level decreases with the progression of AD 17, 18 Whether these levels are reduced due to overall synaptic loss, or are a marker of synaptic damage prior to synaptic loss is yet to be determined. Using the TASTPM model, this study has shown that increased Aβ deposition is significantly associated with reduced synaptophysin levels. Surprisingly, increased Aβ deposition was not associated with significant changes in other synaptic proteins investigated (neurogranin, SNAP47, SV2C, SNAP25 or PSD95) and this suggests that synaptophysin loss is not simply the result of an overall synaptic loss. Thus, other pathological markers of dementia, such as neurofibrillary tangles of tau, reduced acetylcholine, reduced choline acetyltransferase and Lewy body depositions may play a specific role in synaptic breakdown as these features are not found in the TASTPM model. Therefore, synaptic loss may be the result of the cumulative effect of several independent pathophysiological processes which is in contrast with the popular hypotheses for dementia – the amyloid hypothesis and the tau hypothesis.
SNAP47 in AD, DLB, PDD and aged controls
Animal models of dementia allow specific markers of dementia to be studied in isolation and are therefore useful for understanding the pathophysiology of the disease. However, research on human brain tissue is more reliable albeit more challenging. Previous studies have revealed that PSD95 is reduced in patients with AD and has the potential to be used as a biomarker of AD 19. Whilst synaptophysin levels have been shown to be reduced in AD and patients with combined DLB and AD 20. This study attempted to identify further correlations between synaptic proteins and dementia, however, only SNAP47 labelling provided quantifiable data.
SNAP47 was introduced to this study through a Wolfson Laboratories collaboration with Bereczki et al in the Karolinska Laboratories in Stockholm. The preliminary findings from their mass spectrometry study (now published) of human post-mortem tissue suggested SNAP47 levels are reduced in patients with Parkinson’s Disease (PD) and DLB 21. Findings from this IHC study, conversely showed SNAP47 significantly increased in AD, insignificantly decreased in PDD and unchanged in DLB. Due to the conflicting results between this study and Bereczki et al, further research into the role of SNAP47 is necessary. However, other presynaptic proteins such as Rab3 and SNAP25 have recently been shown to be reduced during the cognitive decline of dementia through combined use of mass spectrometry and IHC. Rab3 was reduced in DLB, whilst SNAP25 was reduced in AD, both were in association with cognitive decline. Furthermore, the study revealed synaptic protein levels to be significantly different between the different forms of dementia (DLB, AD and PD) 22.
The TASTPM model does not show cognitive decline nor does it produce neurofibrillary tangles and may only illustrate familial forms of AD. The TASTPM study is also limited by its size. Limitations of the human study included its limited size, the variety and availability of cortical tissue, as well as the inability to control certain variables such as the time of post-mortem and the processing and fixation time. Assessment of synaptic proteins in both the TASTPM study and the human study had elements of observer bias.
The presence of amyloid plaques in AD was discovered by Alois Alzheimer in 1906. Over 100 years have passed since this discovery and the pathophysiology of AD, as well as other dementias, remain unknown. Current treatments for dementia, such as acetylcholinesterase inhibitors and NMDA receptors antagonists, provide only symptomatic relief and do not reverse, slow down or cure the underlying disease. Therapies targeting amyloid plaque deposition and tau aggregation have failed to provide benefit during clinical trials and highlights the need for further research into the pathophysiology of dementia. Although there is little research into the roles of synaptic proteins in dementia, recent studies have shown significant associations between synaptic protein levels and the progression of dementia. Further research into this field may result in the use of synaptic proteins as biomarkers for differentiating between different dementia states, predictors of cognitive decline as well as therapies aimed at restoring synaptic protein levels and potentially preventing synaptic loss and/or neuronal cell death.
I would like to thank David Howlett for being an immense support during this project.
King’s College London
- Department of Health. Prime Minister’s challenge on dementia. 2012. Retrieved 22nd of April 2016 from https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/215101/dh_133176.pdf
- World Health Organisation. 10 Facts on Dementia. 2015. Retrieved 22nd of April from http://www.who.int/features/factfiles/dementia/en/
- Davies CA, Mann DM., Sumpter PQ, Yates PO. A quantitative morphometric analysis of the neuronal and synaptic content of the frontal and temporal cortex in patients with Alzheimer’s disease. J Neurol Sci 1987 78(2):151-64.
- DeKosky ST & Scheff SW. Synapse loss in frontal cortex biopsies in Alzheimer’s disease: correlation with cognitive severity. Ann Neurol 1990 27(5):457-64.
- Diez-Guerra FJ. Neurogranin, a link between calcium/calmodulin and protein kinase C signaling in synaptic plasticity. IUBMB Life 2010;62(8):597-606.
- Chen X, Nelson CD, Li X, Winters CA, Azzam R, Sousa AA, et al. PSD-95 Is Required to Sustain the Molecular Organization of the Postsynaptic Density. J Neurosci 2011 04/27;31(17):6329.
- Sorensen JB., Matti U, Wei S, Nehring RB, Voets T, Ashery U, et al. The SNARE protein SNAP-25 is linked to fast calcium triggering of exocytosis. Proc Natl Acad Sci U S A 2001 12/17;99(3):1627-1632.
- Edelmann L, Hanson PI, Chapman ER, Jahn R. Synaptobrevin binding to synaptophysin: a potential mechanism for controlling the exocytotic fusion machine. EMBO J 1995 01/16;14(2):224-231.
- Holt M, Varoqueaux F, Wiederhold K, Takamori S, Urlaub H, Fasshauer D, et al. Identification of SNAP-47, a Novel Qbc-SNARE with Ubiquitous Expression. J Biol Chem 2006 06/23;281(25):17076-17083.
- Janz R, Suddhof TC. SV2C is a synaptic vesicle protein with an unusually restricted localization: anatomy of a synaptic vesicle protein family. Neuroscience 1999 11;94(4):1279-1290.
- Reitz C. Alzheimer’s Disease and the Amyloid Cascade Hypothesis: A Critical Review. International Journal of Alzheimer’s Disease 2012 01/03;2012:369808.
- Brion JP. Neurofibrillary Tangles and Alzheimer’s Disease. Eur Neurol 1998 40(3):130-40.
- Therapeutics: LMTM. Retrieved 17th of March 2017 from http://www.alzforum.org/therapeutics/lmtm
- McKeith IG, Dickson DW, Lowe J, Emre M, O’Brien JT, Feldman H et al. Diagnosis and management of dementia with Lewy bodies—third report of the DLB consortium. Neurology 2005 65:1863–1872.
- Emre M, Aarsland D, Brown R, Burn DJ, Duyckaerts C, Mizuno Y, et al. Clinical diagnostic criteria for dementia associated with Parkinson’s disease. Movement Disorders 2007;22(12):1689-1707.
- Hamos JE, DeGennaro LJ, & Drachman DA. Synaptic loss in Alzheimer’s disease and other dementias. Neurology 1989 39(3):355-61.
- Reddy PH, Mani G, Park BS, Jacques J, Murdoch G, Whetsell WJr, et al. Differential loss of synaptic proteins in Alzheimer’s disease: implications for synaptic dysfunction J Alzheimer’s Dis 2005 7(2):103–117
- Counts SE, Nadeem M, Lad SP, Wuu J, Mufson Differential expression of synaptic proteins in the frontal and temporal cortex of elderly subjects with mild cognitive impairment. JNeuropathol Exp Neurol 2006 65:592–601
- Whitfield DR, Vallortigara J, Alghamdi A, Howlett D, Hortobagyi T, Johnson M, et al. Assessment of ZnT3 and PSD95 protein levels in Lewy body dementias and Alzheimer’s disease: association with cognitive impairment. Neurobiol Aging 2017/03;35(12):2836-2844.
- Hansen LA, Daniel SE, Wilcock GK, Love S. Frontal cortical synaptophysin in Lewy body diseases: relation to Alzheimerâ€™s disease and dementia. J Neurol Neurosurg Psychiatr 1998 05/01;64(5):653.
- Bereczki E, Branca RMM, Francis PT, Baek JH, Lehtio J, Aarsland D. Synaptic biomarkers for cognitive decline in Lewy body dementias and Alzheimer’s disease. A mass spectrometry approach [abstract]. Mov Disord 2016; 31 (suppl 2). http://www.mdsabstracts.org/abstract/synaptic-biomarkers-for-cognitive-decline-in-lewy-body-dementias-and-alzheimers-disease-a-mass-spectrometry-approach/. Accessed March 16, 2017.
- Bereczki E, Francis PT, Howlett D, Pereira JB, Hoglund K, Bogstedt A, et al. Synaptic proteins predict cognitive decline in Alzheimer’s disease andLewy body dementia. Alzheimer’s & Dementia 2016 11;12(11):1149-1158.