| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online August 29, 2005 as doi:10.1096/fj.05-4263fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Pathology, Division of Neuropathology, University of Washington, Seattle, WA,
* Department of Neurology, Oregon Health & Sciences University, Portland, OR
Correspondence: Department of Pathology, Harberview Medical Center, Seattle, WA 98104, USA. E-mail: tmontine{at}u.washington.edu
SPECIFIC AIMS
Classical biochemical characterization of the major detergent-insoluble proteins that comprise hallmark histopathologic lesions initiated the molecular era of Alzheimer’s disease (AD) research. Here, we re-investigated detergent-insoluble proteins in AD using liquid chromatography (LC) -tandem mass spectrometry (MS) -based proteomics and samples from late-onset AD (LOAD), prodromal dementia, autosomal dominant AD, APPswe mice, and controls.
PRINCIPAL FINDINGS
1. Discovery: widespread detergent-insoluble protein in late-onset AD (LOAD)
We performed serial extractions of gray matter from superior and middle temporal gyrus from LOAD patients and controls (Fig. 1
). Western blots showed the expected accumulation of aggregated immunoreactivity (IR) for P-tau and Aß species in the detergent-insoluble fraction D of LOAD patients. We performed proteomic analysis on fraction D pooled from 5 LOAD patients using a liquid chromatography (LC) tandem mass spectrometry (MS) method. We discovered 125 proteins, robustly identified by the consensus of 2 or more unique peptides, in pooled fraction D. Of these, more than one-third were within pathogenic pathways already implicated in LOAD: dendritic spines or synapses, mitochondrial enzymes, cytoskeletal proteins, chaperonins, ubiquitination pathway, oxidative stress response, extracellular matrix, and prostaglandin synthesis. Some of the proteins identified have demonstrated protein-protein interactions with elements in the Aß or tau pathways, are components of the synaptic scaffolding that contribute to synaptic fragility in neurodegeneration, or have genetic linkages to increased risk of LOAD (Table 1
).
|
|
2. Verification of proteomic results
We verified our proteomic results by Western blot with the use of commercially available antibodies to selected proteins of interest among the 125 robustly identified in our survey. As expected, two proteins, Aß and tau, were present in the fraction D samples of all 5 LOAD patients as smears of multiple species spanning a wide range of molecular masses. The following proteins were also verified in fraction D of all 5 LOAD samples, showing discrete bands of appropriate molecular mass rather than a smeared pattern: 14-3-3 protein, actin, apoE, collagen IV, cytochrome c, GAPDH, GFAP, histone H2B, HSC70, MBP, tubulin ß isoform, and UCHL1. Connexin 43 also showed discrete banding on Western blot; however, it was detected in only 3 of the 5 LOAD fraction D samples. Thus we verified by Western blot 15 of 15 proteins identified with 2 or more unique peptides by our proteomic method.
3. Accumulation of multiple detergent-insoluble proteins, but not tau, precedes clinical diagnosis of dementia
We investigated a prodromal group, who had mild reduction in cognitive performance but did not have dementia, to determine the extent of detergent-insoluble proteins that might be involved early in the course of AD pathogenesis. Protein concentration in prodromal fraction D was in between LOAD and control groups (µg/mL): LOAD = 131.5 +3.4, prodromal = 44.1 + 2.6, and control = 20.1 + 1.8 (ANOVA had P<0.0001 with Bonferroni-corrected repeated pair comparisons having P<0.05 for all combinations). Since total protein in fraction D varied significantly among groups, quantitative comparisons of verified proteins used IR normalized to protein loaded into the lane on the gel. Remarkably, while detergent-insoluble Aß IR increased 100- to 1000-fold in prodromal and LOAD groups compared with controls (P<0.05 for both comparisons), detergent-insoluble tau IR normalized to total protein in prodromal individuals was not different from controls (P>0.05). As far as we are aware this is the first demonstration of selective increase in detergent-insoluble Aß IR species but not tau in individuals with prodromal dementia.
The protein-normalized IR of 7 other proteins identified and verified in our proteomic analysis followed a pattern like Aß, not like tau, and were significantly elevated in both prodromal and LOAD groups (P<0.05 for all repeated pair-corrected comparisons); these were: apoE, tubulins, 14-3-3, GAPDH, cytochrome c, glial fibrillary acidic protein, and myelin basic protein. The one exception was collagen IV that was relatively more abundant in controls compared with LOAD and prodromal groups, suggesting that the normally detergent-insoluble collagen IV (presumably derived from intraparenchymal blood vessels) was a greater proportion of fraction D protein in controls than LOAD or prodromal groups. These data demonstrated widespread decreased protein solubility in the prodromal group that was similar to LOAD, indicating that formation of multiple detergent-insoluble proteins, excluding tau, occurred early in the course of LOAD and even preceded the clinical diagnosis of dementia.
4. Accumulation of multiple detergent-insoluble proteins in autosomal dominant AD but not in a mouse model of autosomal dominant AD
We repeated this analysis for the same region of brain from patients who died from autosomal dominant AD. Our results showed that, late in the course of their illness, patients with autosomal dominant forms of AD derived from mutations in PSEN1 and PSEN2 had a similar pattern of widespread protein insolubility as patients with advanced LOAD. We also performed the same analysis on cerebrum from 5- or 17-month-old APPswe mice. In fraction D, Aß peptides were below the limit of detection in 17-month-old littermate controls, near the limit of detection in 5-month old APPswe, and abundant in 17-month-old APPswe mice; none of the other 6 proteins assayed were detected in fraction D from littermate controls or APPswe mice of either age.
CONCLUSIONS AND SIGNIFICANCE
We investigated detergent-insoluble proteins in AD using modern proteomic techniques. Using LC-MS-MS-based proteomics we robustly identified 125 proteins in the detergent-insoluble fraction of LOAD temporal cortex that included synaptic, mitochondrial, cytoskeletal, chaperonins, and ubiquitination pathway proteins; we estimated that these represented 
80% of detergent-insoluble proteins in AD that were detectable by our method. We showed that Aß and 7 other newly identified detergent-insoluble proteins were present at low levels in control fraction D and were increased in prodromal and AD groups; however, detergent-insoluble tau was disproportionately elevated only in AD patients. While this pattern of widespread protein insolubility was present in LOAD and autosomal dominant forms of AD, a transgenic mouse model of autosomal dominant AD had much more restricted protein insolubility. Our results are consistent with the amyloid hypothesis that proposes pathologic changes in tau occur relatively late in the pathogenesis of AD. However, our results also extend the amyloid hypothesis to include widespread protein insolubility, not exclusively Aß, in the temporal isocortex early in AD pathogenesis even before the onset of clinical dementia.
Genetic linkages to disease establish the relevance of specific candidate molecules to disease pathogenesis, but in isolation often provide limited insight into pathogenetic mechanisms; consider, for example, the genetic identifications of APP, apoE, AS, Huntingtin, PS-1, PS-2, and tau as proteins that contribute to neurodegenerative disease or the risk of its development, even as the search for a clear picture of pathogenic mechanisms that relate these genetic results to each other and to other observations continues. While some proteins may become insoluble only as part of a widespread decreased solubility, proteomic analysis of a pathologically relevant fraction, such as detergent-insoluble proteins, may achieve a result similar to genetic studies by identifying novel candidate proteins that may be involved in this pathologic process. Likewise, our data may help in the substantiation of newly implicated susceptibility genes, like GAPDH or APOD, or pathways shared by products of susceptibility genes, like UBQLN1. Like the initial biochemical identification of Aß and tau, we hope that discovery of these abnormal proteins and the processes that underlie their alterations in AD will aid others in designing experiments to unravel this facet of AD pathogenesis.
|
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-4263fje;
This article has been cited by other articles:
|
|
 |
|
  W. Yang, R. L. Woltjer, I. Sokal, C. Pan, Y. Wang, M. Brodey, E. R. Peskind, J. B. Leverenz, J. Zhang, D. P. Perl, et al. Quantitative Proteomics Identifies Surfactant-Resistant {alpha}-Synuclein in Cerebral Cortex of Parkinsonism-Dementia Complex of Guam but Not Alzheimer's Disease or Progressive Supranuclear Palsy Am. J. Pathol., September 1, 2007; 171(3): 993 - 1002. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Boutte, R. L. Woltjer, L. J. Zimmerman, S. L. Stamer, K. S. Montine, M. V. Manno, P. J. Cimino, D. C. Liebler, and T. J. Montine Selectively increased oxidative modifications mapped to detergent-insoluble forms of A{beta} and {beta}-III tubulin in Alzheimer's disease FASEB J, July 1, 2006; 20(9): 1473 - 1483. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
