- Email: [email protected]
DEVELOPMENTAL EXPRESSION OF FRONTOTEMPORAL DEMENTIA PROTEIN C9ORF72 IN VIVO AND IN VITRO
Poster Presentations: P2
P487
Table 1 Number of neurons (cell), volume (mm3) and cell density (cell/mm3) of the Substantia Nigra (SN) in 15 human cases. Number of neurons Case Age Genre Side per side Total number of neurons Volume per side Total volume Neuron density per side Mean neuron density 1
50
F
2
50
M
3
52
M
4
55
F
5
57
F
6
60
F
7
65
M
8
66
M
9
71
M
10
74
F
11
77
M
12
78
F
13
82
F
14
83
M
15
91
F
R L R L R L R L R L R L R L R L R L R L R L R L R L R L R L
576,431 738,047 686,869 503,704 839,058 730,640 588,047 605,724 616,701 811,179 732,660 608,755 707,071 579,798 763,637 867,341 622,222 947,475 367,677 311,111 315,960 544,916 763,636 650,505 494,545 700,606 694,950 597,980 405,387 565,657
1,314,478 1,190,573 1,569,698 1,193,771 1,427,880 1,341,415 1,286,869 1,630,978 1,569,697 678,788 860,876 1,414,141 1,195,151 1,292,930 971,044
88.8 102.0 116.28 92.82 107.1 93.1 96.6 98.35 90.44 110.96 127.6 102.4 66.15 67.2 136.5 130.9 89.25 106.4 81.9 80.85 69.02 69.02 94.5 79.8 70.72 83.64 100.4 89.6 91.0 97.3
190.8 209.1 200.2 194.95 201.4 230.0 133.35 267.4 195.65 162.75 138.04 174.3 154.36 190.0 188.3
7234.50 7235.75 5907.02 5426.67 7834.33 7847.90 6087.44 6158.86 6818.88 7310.54 5741.84 5944.86 10688.89 8627.94 5594.40 6625.97 6971.67 8904.83 4489.33 3848.00 4577.79 7895.04 8080.80 8151.69 6993.00 8376.44 6921.80 6673.88 4454.80 5813.53
7235.13 5666.85 7841.12 6123.15 7064.71 5843.35 9658.42 6110.19 7938.25 4168.67 6236.42 8116.25 7684.72 6797.84 5134.17
Abbreviations: R, Right side; L, Left side
Paulo, Brazil; 2University of Sao Paulo, Sao Paulo, Brazil; 3University of Sao Paulo Medical School, Sao Paulo, Brazil; 4University of S~ao Paulo Medical School, Sao Paulo, Brazil; 5University of Sao Paulo, Sao Paulo, Brazil; 6University of Sao Paulo, S~ao Paulo, Brazil; 7University of Sao Paulo Medical School, S~ ao Paulo, Brazil; 8University of S~ao Paulo, S~ao 9 Paulo, Brazil; University of S~ao Paulo, S~ao Paulo, Brazil; 10Wuerzburg University, Wuerzburg, Germany; 11University of California San Francisco, San Francisco, California, United States. Contact e-mail: anatealho@ gmail.com Background: The human brain changes along the life span; however, these changes are heterogeneuous. During aging process, some areas loses volumes whereas other remain intact. The substantia nigra (SN) is main dopamine producer in the brain, and vulnerable to several neurodegenerative diseases (Alzheimer’s, Parkinson’s and Frontotemporal lobar degeneration) from their early stages. Aim: To analyze changes in volume and neuronal number and density in the SN during human aging, using stereological methods and techniques of 3D reconstruction. Methods: Intact SNs were provided by the Brain Bank of the Brazilian Aging Brain Study Group. Fifteen subjects were included according these criteria: Older than 50 years, no cognitive impairment, or parkinsonism evaluated by a clinical interview applied to an informant. The samples were fixed in formalin, embedded in celloidin, cut in 400mm horizontal serial sections. Blockface images were used for 3D reconstruction. Slides were stained with gallocyanin. Number of neurons were estimated using the optical dissector method, following the princi-
ples of design based stereology. Volume and neuronal density were estimated. Results: SN volume ranged from 133.4 mm3 and 267.4 mm3 (variation > 100%). Significant correlation between older age and decreased SN volume was found (p¼0.04) great variability was also found regarding neuronal number and density. Number of neurons also varied by more than 100%, ranging between 678,788 cells and 1630,978cells, however numbers remained stable during aging. 3D reconstruction showed considerable interindividual variability and asymmetry. Conclusions: SN loses volume during again process despite maintaining neuronal numbers. Loss of non-pigmented cells of SN and variation in neurons size can contribute for this volume decrease. In addition, inter-individual and inter-hemispheric differences are pronounced. It is not clear how these changes affect function. Studies with a larger sample size are needed to create normative basis for SN studies in aging individuals and comparison with disease states. P2-048
DEVELOPMENTAL EXPRESSION OF FRONTOTEMPORAL DEMENTIA PROTEIN C9ORF72 IN VIVO AND IN VITRO
Anna King1, Rachel Atkinson1, James Vickers2, 1Wicking Dementia Research and Education Centre, Hobart, Australia; 2Wicking Dementia Research and Education Centre, Hobart, Australia. Contact e-mail: [email protected] Background: Hexanucleotide repeat expansion in a non-coding region of the C9ORF72 gene is the most common genetic cause of
P488
Poster Presentations: P2
frontotemporal dementia. It is still unclear how this mutation causes disease, however altered expression of C9ORF72 transcripts are present in C9ORF72 patients suggesting that loss of function is a potential pathogenic mechanism of disease. C9ORF72 encodes a protein of unknown function. In this study we have examined the expression and localization of C9ORF72 over a developmental timecourse both in vivo and in vitro in order to provide some insight into the normal role of the protein. Methods: Primary neuron and glia cultures were derived from embryonic and postnatal rodents, grown for up to 21 days and fixed in PFA or harvested for Western blotting over a timecourse of 1, 3,7 14 and 21 days. Brains were harvest from mice at E15, E18, P1, P7, P14 and P28 and processed for Western blotting or fixed in PFA. Immunolabelling and Western blotting were performed with antibodies against C9ORF72 in addition to cytoskeletal proteins, axonal and dendritic markers and cellular organelles. Results: C9ORF72 was consistently expressed in both neurons and glia from early in development. C9ORF72 was predominantly absent from the nucleus but had a punctate expression in the cytoplasm and throughout the neuropil in vivo. In vitro the protein was localized to discrete vesicles (approximate diameter 0.6mm), which extended into both axons and dendrites. C9ORF72 immunoreactive-vesicles extended beyond the microtubule cytoskeleton into actin rich structures, such as filapodia. C9ORF72 was not localized to Golgi apparatus or mitochondria but shared some co-localization with components of the endosome/lysosome system. C9ORF72 immunoreactivity rarely localized to synaptic puncta in vitro or in vivo. Conclusions: These data suggest that C9ORF72 is a vesicular protein that is expressed throughout development and into adulthood. Protein expression is present in both neuron and glial populations. C9ORF72 immunoreactive vesicles are found throughout the cell including in actin rich structures and may be involved in membrane trafficking. Determining the normal role of C9ORF72 protein may help to determine the role it plays in disease. P2-049
MELATONIN TREATMENT ATTENUATES GLUCOSE DYSREGULATION AND GLIOSIS IN HYPERGLYCEMIC NON-TG MICE
Alice Robinson1, Ayodeji A. Asuni1, Rania Salman2, 1University of Southampton, Southampton, United Kingdom; 2University of Southampton, Southampton, United Kingdom. Contact e-mail: [email protected] Background: Brain glucose metabolism shows regional variability and synchronized circadian rhythm. Hyperglycemia-mediated disruption of these coordinated rhythms may contribute to metabolic disease and exacerbation of cognitive impairments observed in diabetics. Glucose metabolism is also impaired in AD brain, and appears to be causative, rather than consequential for neurodegeneration. Hyperglycemia-induced overproduction of reactive oxygen species (ROS), associated neuroinflammation and generation of advanced glycation end products mechanistically promote cellular injury in the CNS. The impact of hyperglycemia on the differential expression of CLOCK genes and nutrient transporters in the brain of diabetics is uncertain. We have examined whether neuropathology in hyperglycemic non-Tg animals can be modulated by treatment with the antioxidant hormone, melatonin (5-methoxy- N -acetyltryptamine); a potent endogenous scavengers with proven capacity to modulate cellular stress and inflammation. Methods: Cohorts of normoglycemic and insulin-deficient streptozotocin (STZ)-induced hyperglycemic non-Tg C57BL6 animals were compared. Body weight and glucose levels were assessed and the mice were killed after 90 days, and their brains were portioned to measure biochemical and histological indices of hyper-
glycemia and insulin resistance. Transcript analysis was also carried with selected primer pairs to reveal evidence for modulation of glucose metabolism (glut-1), gliosis (Cd11b, GFAP), and clock genes (clock, timeless). Results: There was progressive increase in gliosis in the brains of hyperglycemic compared to normoglycemic non-Tg mice. Hyperglycemic non-Tg animals exhibited disrupted mRNA expression of genes involved in controlling circadian rhythmicity, and the brain expression of glucose transporter type 1 was dysregulated. Treatment with melatonin rescued glucose dyshomeostasis, and attenuated neuropathological indices of hyperglycemiamediated neurodegeneration. Conclusions: Our findings suggest that aberrant glucose dyshomeostasis and associated CNS circadian clock dyssynchrony contribute to hyperglycemia-mediated neurodegeneration, which is attenuated by melatonin. These data support the further exploration of melatonin augmentation as an adjunctive treatment option for diabetes and related neurodegenerative diseases.
P2-050
CHANGES IN THE HUMAN HIPPOCAMPAL PROTEOME DURING ALZHEIMER’S DISEASE
David Hondius1, Pim van Nierop2, Ka Wan Li2, Jeroen Hoozemans1, Roel van der Schors2, Elise van Haastert1, Saskia M. van der Vies3, Annemieke Rozemuller1, Guus Smit2, 1VU University Medical Center, Amsterdam, Netherlands; 2VU University, Amsterdam, Netherlands; 3VU University Medical Center, Amsterdam, Netherlands. Contact e-mail: d. [email protected] Background: Because of the current lack of effective treatment and early presymptomatic diagnostic markers of Alzheimer’s disease (AD), detailed insight into disease mechanisms involved at the various stages of AD is highly important. The pathological hallmarks, amyloid beta aggregation into plaques and intra cellular aggregation of hyperphosphorylated tau in neurofibrillary tangles, are used to objectively stage the disease. In this study we have analysed the proteome of the hippocampus, and in particular the CA1 region and subiculum, as this is one of the most vulnerable and early affected regions in AD, with the aim to identify changes in the human proteome and underlying disease mechanisms. Methods: A total of 56 patients representing all stages of AD (Braak 0 to VI) were selected. Hippocampal regions CA1 and subiculum were isolated from human post-mortem brain tissue using laser capture microdissection and protein lysates were prepared. After separation of the proteins by SDS-PAGE and in-gel trypsin digestion, the extracted peptides were analysed by LC-MS/MS using Orbitrap mass spectrometry. MaxQuant software was used for protein identification and quantification. Cluster analysis was used to identify proteins that are (co-) regulated during the disease course. Results: Analysis of the mass spectrometry data resulted in the identification of a total of 3283 proteins of which a subset of 218 proteins were found significantly regulated. Known proteins such as GFAP and tau were increased with increased Braak stage, as expected. We also identified a range of proteins that have not been associated with AD before. Interestingly one group exhibited an "early up late down" pattern. Results were validated by immunoblotting and immunohistochemical analysis. Conclusions: Combining laser capture microdissection and quantitative proteomics enabled us to obtain an extensive representation of the human proteome of the hippocampal subregions at different stages of AD pathogenesis. The validity and reliability of our approach is illustrated by the identification of several proteins previously shown to change during the progression of AD. Newly identified proteins and pathways will provide insight into the complex biology of AD and may lead to new biomarkers and/or therapeutic drug targets.
P487
Table 1 Number of neurons (cell), volume (mm3) and cell density (cell/mm3) of the Substantia Nigra (SN) in 15 human cases. Number of neurons Case Age Genre Side per side Total number of neurons Volume per side Total volume Neuron density per side Mean neuron density 1
50
F
2
50
M
3
52
M
4
55
F
5
57
F
6
60
F
7
65
M
8
66
M
9
71
M
10
74
F
11
77
M
12
78
F
13
82
F
14
83
M
15
91
F
R L R L R L R L R L R L R L R L R L R L R L R L R L R L R L
576,431 738,047 686,869 503,704 839,058 730,640 588,047 605,724 616,701 811,179 732,660 608,755 707,071 579,798 763,637 867,341 622,222 947,475 367,677 311,111 315,960 544,916 763,636 650,505 494,545 700,606 694,950 597,980 405,387 565,657
1,314,478 1,190,573 1,569,698 1,193,771 1,427,880 1,341,415 1,286,869 1,630,978 1,569,697 678,788 860,876 1,414,141 1,195,151 1,292,930 971,044
88.8 102.0 116.28 92.82 107.1 93.1 96.6 98.35 90.44 110.96 127.6 102.4 66.15 67.2 136.5 130.9 89.25 106.4 81.9 80.85 69.02 69.02 94.5 79.8 70.72 83.64 100.4 89.6 91.0 97.3
190.8 209.1 200.2 194.95 201.4 230.0 133.35 267.4 195.65 162.75 138.04 174.3 154.36 190.0 188.3
7234.50 7235.75 5907.02 5426.67 7834.33 7847.90 6087.44 6158.86 6818.88 7310.54 5741.84 5944.86 10688.89 8627.94 5594.40 6625.97 6971.67 8904.83 4489.33 3848.00 4577.79 7895.04 8080.80 8151.69 6993.00 8376.44 6921.80 6673.88 4454.80 5813.53
7235.13 5666.85 7841.12 6123.15 7064.71 5843.35 9658.42 6110.19 7938.25 4168.67 6236.42 8116.25 7684.72 6797.84 5134.17
Abbreviations: R, Right side; L, Left side
Paulo, Brazil; 2University of Sao Paulo, Sao Paulo, Brazil; 3University of Sao Paulo Medical School, Sao Paulo, Brazil; 4University of S~ao Paulo Medical School, Sao Paulo, Brazil; 5University of Sao Paulo, Sao Paulo, Brazil; 6University of Sao Paulo, S~ao Paulo, Brazil; 7University of Sao Paulo Medical School, S~ ao Paulo, Brazil; 8University of S~ao Paulo, S~ao 9 Paulo, Brazil; University of S~ao Paulo, S~ao Paulo, Brazil; 10Wuerzburg University, Wuerzburg, Germany; 11University of California San Francisco, San Francisco, California, United States. Contact e-mail: anatealho@ gmail.com Background: The human brain changes along the life span; however, these changes are heterogeneuous. During aging process, some areas loses volumes whereas other remain intact. The substantia nigra (SN) is main dopamine producer in the brain, and vulnerable to several neurodegenerative diseases (Alzheimer’s, Parkinson’s and Frontotemporal lobar degeneration) from their early stages. Aim: To analyze changes in volume and neuronal number and density in the SN during human aging, using stereological methods and techniques of 3D reconstruction. Methods: Intact SNs were provided by the Brain Bank of the Brazilian Aging Brain Study Group. Fifteen subjects were included according these criteria: Older than 50 years, no cognitive impairment, or parkinsonism evaluated by a clinical interview applied to an informant. The samples were fixed in formalin, embedded in celloidin, cut in 400mm horizontal serial sections. Blockface images were used for 3D reconstruction. Slides were stained with gallocyanin. Number of neurons were estimated using the optical dissector method, following the princi-
ples of design based stereology. Volume and neuronal density were estimated. Results: SN volume ranged from 133.4 mm3 and 267.4 mm3 (variation > 100%). Significant correlation between older age and decreased SN volume was found (p¼0.04) great variability was also found regarding neuronal number and density. Number of neurons also varied by more than 100%, ranging between 678,788 cells and 1630,978cells, however numbers remained stable during aging. 3D reconstruction showed considerable interindividual variability and asymmetry. Conclusions: SN loses volume during again process despite maintaining neuronal numbers. Loss of non-pigmented cells of SN and variation in neurons size can contribute for this volume decrease. In addition, inter-individual and inter-hemispheric differences are pronounced. It is not clear how these changes affect function. Studies with a larger sample size are needed to create normative basis for SN studies in aging individuals and comparison with disease states. P2-048
DEVELOPMENTAL EXPRESSION OF FRONTOTEMPORAL DEMENTIA PROTEIN C9ORF72 IN VIVO AND IN VITRO
Anna King1, Rachel Atkinson1, James Vickers2, 1Wicking Dementia Research and Education Centre, Hobart, Australia; 2Wicking Dementia Research and Education Centre, Hobart, Australia. Contact e-mail: [email protected] Background: Hexanucleotide repeat expansion in a non-coding region of the C9ORF72 gene is the most common genetic cause of
P488
Poster Presentations: P2
frontotemporal dementia. It is still unclear how this mutation causes disease, however altered expression of C9ORF72 transcripts are present in C9ORF72 patients suggesting that loss of function is a potential pathogenic mechanism of disease. C9ORF72 encodes a protein of unknown function. In this study we have examined the expression and localization of C9ORF72 over a developmental timecourse both in vivo and in vitro in order to provide some insight into the normal role of the protein. Methods: Primary neuron and glia cultures were derived from embryonic and postnatal rodents, grown for up to 21 days and fixed in PFA or harvested for Western blotting over a timecourse of 1, 3,7 14 and 21 days. Brains were harvest from mice at E15, E18, P1, P7, P14 and P28 and processed for Western blotting or fixed in PFA. Immunolabelling and Western blotting were performed with antibodies against C9ORF72 in addition to cytoskeletal proteins, axonal and dendritic markers and cellular organelles. Results: C9ORF72 was consistently expressed in both neurons and glia from early in development. C9ORF72 was predominantly absent from the nucleus but had a punctate expression in the cytoplasm and throughout the neuropil in vivo. In vitro the protein was localized to discrete vesicles (approximate diameter 0.6mm), which extended into both axons and dendrites. C9ORF72 immunoreactive-vesicles extended beyond the microtubule cytoskeleton into actin rich structures, such as filapodia. C9ORF72 was not localized to Golgi apparatus or mitochondria but shared some co-localization with components of the endosome/lysosome system. C9ORF72 immunoreactivity rarely localized to synaptic puncta in vitro or in vivo. Conclusions: These data suggest that C9ORF72 is a vesicular protein that is expressed throughout development and into adulthood. Protein expression is present in both neuron and glial populations. C9ORF72 immunoreactive vesicles are found throughout the cell including in actin rich structures and may be involved in membrane trafficking. Determining the normal role of C9ORF72 protein may help to determine the role it plays in disease. P2-049
MELATONIN TREATMENT ATTENUATES GLUCOSE DYSREGULATION AND GLIOSIS IN HYPERGLYCEMIC NON-TG MICE
Alice Robinson1, Ayodeji A. Asuni1, Rania Salman2, 1University of Southampton, Southampton, United Kingdom; 2University of Southampton, Southampton, United Kingdom. Contact e-mail: [email protected] Background: Brain glucose metabolism shows regional variability and synchronized circadian rhythm. Hyperglycemia-mediated disruption of these coordinated rhythms may contribute to metabolic disease and exacerbation of cognitive impairments observed in diabetics. Glucose metabolism is also impaired in AD brain, and appears to be causative, rather than consequential for neurodegeneration. Hyperglycemia-induced overproduction of reactive oxygen species (ROS), associated neuroinflammation and generation of advanced glycation end products mechanistically promote cellular injury in the CNS. The impact of hyperglycemia on the differential expression of CLOCK genes and nutrient transporters in the brain of diabetics is uncertain. We have examined whether neuropathology in hyperglycemic non-Tg animals can be modulated by treatment with the antioxidant hormone, melatonin (5-methoxy- N -acetyltryptamine); a potent endogenous scavengers with proven capacity to modulate cellular stress and inflammation. Methods: Cohorts of normoglycemic and insulin-deficient streptozotocin (STZ)-induced hyperglycemic non-Tg C57BL6 animals were compared. Body weight and glucose levels were assessed and the mice were killed after 90 days, and their brains were portioned to measure biochemical and histological indices of hyper-
glycemia and insulin resistance. Transcript analysis was also carried with selected primer pairs to reveal evidence for modulation of glucose metabolism (glut-1), gliosis (Cd11b, GFAP), and clock genes (clock, timeless). Results: There was progressive increase in gliosis in the brains of hyperglycemic compared to normoglycemic non-Tg mice. Hyperglycemic non-Tg animals exhibited disrupted mRNA expression of genes involved in controlling circadian rhythmicity, and the brain expression of glucose transporter type 1 was dysregulated. Treatment with melatonin rescued glucose dyshomeostasis, and attenuated neuropathological indices of hyperglycemiamediated neurodegeneration. Conclusions: Our findings suggest that aberrant glucose dyshomeostasis and associated CNS circadian clock dyssynchrony contribute to hyperglycemia-mediated neurodegeneration, which is attenuated by melatonin. These data support the further exploration of melatonin augmentation as an adjunctive treatment option for diabetes and related neurodegenerative diseases.
P2-050
CHANGES IN THE HUMAN HIPPOCAMPAL PROTEOME DURING ALZHEIMER’S DISEASE
David Hondius1, Pim van Nierop2, Ka Wan Li2, Jeroen Hoozemans1, Roel van der Schors2, Elise van Haastert1, Saskia M. van der Vies3, Annemieke Rozemuller1, Guus Smit2, 1VU University Medical Center, Amsterdam, Netherlands; 2VU University, Amsterdam, Netherlands; 3VU University Medical Center, Amsterdam, Netherlands. Contact e-mail: d. [email protected] Background: Because of the current lack of effective treatment and early presymptomatic diagnostic markers of Alzheimer’s disease (AD), detailed insight into disease mechanisms involved at the various stages of AD is highly important. The pathological hallmarks, amyloid beta aggregation into plaques and intra cellular aggregation of hyperphosphorylated tau in neurofibrillary tangles, are used to objectively stage the disease. In this study we have analysed the proteome of the hippocampus, and in particular the CA1 region and subiculum, as this is one of the most vulnerable and early affected regions in AD, with the aim to identify changes in the human proteome and underlying disease mechanisms. Methods: A total of 56 patients representing all stages of AD (Braak 0 to VI) were selected. Hippocampal regions CA1 and subiculum were isolated from human post-mortem brain tissue using laser capture microdissection and protein lysates were prepared. After separation of the proteins by SDS-PAGE and in-gel trypsin digestion, the extracted peptides were analysed by LC-MS/MS using Orbitrap mass spectrometry. MaxQuant software was used for protein identification and quantification. Cluster analysis was used to identify proteins that are (co-) regulated during the disease course. Results: Analysis of the mass spectrometry data resulted in the identification of a total of 3283 proteins of which a subset of 218 proteins were found significantly regulated. Known proteins such as GFAP and tau were increased with increased Braak stage, as expected. We also identified a range of proteins that have not been associated with AD before. Interestingly one group exhibited an "early up late down" pattern. Results were validated by immunoblotting and immunohistochemical analysis. Conclusions: Combining laser capture microdissection and quantitative proteomics enabled us to obtain an extensive representation of the human proteome of the hippocampal subregions at different stages of AD pathogenesis. The validity and reliability of our approach is illustrated by the identification of several proteins previously shown to change during the progression of AD. Newly identified proteins and pathways will provide insight into the complex biology of AD and may lead to new biomarkers and/or therapeutic drug targets.