FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Boutte, A. M.
Right arrow Articles by Montine, T. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Boutte, A. M.
Right arrow Articles by Montine, T. J.
(The FASEB Journal. 2006;20:1473-1483.)
© 2006 FASEB

Selectively increased oxidative modifications mapped to detergent-insoluble forms of Aß and ß-III tubulin in Alzheimer’s disease

Angela M. Boutte*, Randall L. Woltjer, Lisa J. Zimmerman{dagger}, Sheryl L. Stamer{dagger}, Kathleen S. Montine, Michael V. Manno, Patrick J. Cimino, Daniel C. Liebler{dagger} and Thomas J. Montine1

* Center for Molecular Neuroscience and

{dagger} Department of Biochemistry and Mass Spectrometry Research Center, Vanderbilt University, Nashville, Tennessee, USA; and 3Department of Pathology, University of Washington, Seattle, Washington, USA

1Correspondence: Department of Pathology, University of Washington, Harborview Medical Center, Box 359791, Seattle, WA 98104, USA. E-mail: tmontine{at}u.washington.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Deleterious post-translational modifications (PTMs) to the neuronal cytoskeleton are a proposed mechanistic link between accumulation of amyloid (A) ß peptides and subsequent abnormalities of tau and neurodegeneration in Alzheimer’s disease (AD). Here we tested the hypothesis that PTMs on neuronal tubulins selectively accumulate in a pathological protein fraction in AD. We used new software, P-MOD, to identify comprehensively and map PTMs using mass spectral data from soluble (normal) and detergent-insoluble (pathological) protein fractions from AD, as well as total extracts from controls, for selected proteins: Aß, tau, apolipoprotein (apo) E, glial fibrillary acidic protein (GFAP), {alpha}-III tubulin, and ß-III tubulin. Our results confirmed direct observations of others by identifying methionine (M) sulfoxides at Aß position 35 and numerous sites of tau phosphorylation in detergent-insoluble protein from AD, while no PTMs were enriched on primarily astrocyte-derived apoE or GFAP in this fraction. P-MOD mapped several abundant M sulfoxides to neuron-enriched ß-III tubulin but not its heterodimeric partner, neuron-enriched {alpha}-III tubulin, a result confirmed by selective suppression of CNBr-mediated cleavage of ß-III tubulin. These findings are the first comprehensive assessment of PTMs in AD and point to oxidative modification of ß-III tubulin as a potential contributor to the neuronal cytoskeletal disruption that is characteristic of AD.—Boutte, A. M., Woltjer, R. L., Zimmerman, L. J., Stamer, S. L., Montine, K. S., Manno, M. V., Cimino, P. J., Liebler, D. C., Montine, T. J. Selectively increased oxidative modifications mapped to detergent-insoluble forms of Aß and ß-III tubulin in Alzheimer’s disease.


Key Words: PTMs • gene expression • paired helical filament • dendrites


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
THE AMYLOID HYPOTHESIS of Alzheimer’s disease (AD) holds that accumulation of amyloid (A) ß peptides in some form sets in motion a cascade of pathogenic events that culminate in the full expression of AD (1) , prominently including morphological and biochemical changes indicative of damage to the neuronal cytoskeleton; such changes include loss of synapses and dendrites, synaptic vesicles that fail to reach the terminal compartment, vesicle accumulation in neuron soma, increased mitochondrial elements in lysosomes, and formation of a highly modified paired helical filament (PHF) form of the microtubule-associated protein (MAP) tau that is the major component of neurofibrillary tangles (NFTs) (2 3 4 5 6 7 8) . While there are several transgenic mouse models that reflect the Aß accumulation facet of autosomal dominant forms of AD, so far there is no animal model that fully reflects the neuronal cytoskeletal changes characteristic of AD, let alone the neuron loss that ultimately accrues. Therefore, it remains unclear what molecular mechanisms connect accumulation of Aß peptides with neuronal cytoskeletal disruption.

We and others have proposed that oxidative post-translational modification (PTM) of neuronal cytoskeletal protein is one potential link between pathogenic processes initiated by Aß accumulation and the cytoskeletal abnormalities that are closely related to neuron dysfunction and degeneration in AD. Given the limitations of current animal models, many have pursued this hypothesis in tissue from patients with pathologically confirmed AD where numerous studies have documented many abnormal PTMs, not only from oxidative mechanisms. However, the vast majority of these studies have quantified a type of PTM from digests but have not determined the specific proteins involved. Notable exceptions are oxidation of methionine (M) 35 of Aß peptides to its sulfoxide and phosphorylation of tau (9 , 10) . Recent studies have focused on determining the proteome that harbors specific PTMs, such as protein carbonyls or specific glycations (11 12 13) . However, even these elegant proteomic studies have limitations: they are restricted to a particular PTM, have not mapped PTMs to specific amino acids within proteins, and do not address the pathophysiologic relevance of the specific PTM investigated.

Here we used a new discovery tool for MS-MS data, P-MOD, in a thorough assessment of PTMs on selected proteins that accumulate in a pathophysiologically relevant fraction in AD brain. P-MOD is new software that comprehensively identifies and maps mass shifts by analyzing MS-MS data against in silico generated peptides from a known protein sequence (14) . P-MOD assigns a P value to each sequence-to-spectrum match that is scaled to the number of comparisons, so error rates do not increase with expanded search lists, and also accounts for the possibility that modifications to lysine or arginine residues could block tryptic digestion. Two recent applications of P-MOD have been to map protein hydroxylation and Cu2+-induced histidine and methionine oxidation of Aß peptides in vitro (15 , 16) . Despite these successful applications to in vitro samples, P-MOD has yet to be employed in the analysis of MS data from tissue. We used liquid chromatography (LC) -MS-MS and P-MOD analysis as a novel means to discover and map PTMs on soluble (normal) and detergent-insoluble (disease-associated) forms of six proteins in the temporal cortex, a diseased region of the AD brain. Of these, the first two were primarily neuron-derived proteins with at least one previously demonstrated disease-associated PTM (Aß peptides and tau) to act as validation of our approach. The next two were primarily astrocyte-derived proteins (apolipoprotein E or apoE and glial fibrillary acidic protein or GFAP) to examine cellular specificity of PTMs. The last two were neuron-enriched tubulins, {alpha}-III and ß-III, to test the hypothesis that pathological forms of neuronal cytoskeletal proteins in AD selectively accumulate specific PTMs in functionally relevant domains.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Human brain tissue
Use of human tissue was approved by the University of Washington (UW) Institutional Review Board. Human brain samples were obtained from the Neuropathology Core of the Alzheimer Disease Research Center (ADRC) at UW after appropriate informed consent, flash frozen in liquid nitrogen at time of autopsy and stored at –80°C. Patients with AD were volunteers in the UW ADRC, where they were diagnosed during life with probable AD and shown by neuropathologic examination to have AD (17) . Control individuals also were volunteers in the UW ADRC, were never diagnosed during life with disease of the central nervous system, and had age-related changes only by neuropathologic examination.

Extractions
Superior and middle temporal gyri cortex was used in all experiments. Tissue from five patients with AD and five controls were serially extracted first with first buffer A (10 µl/g tissue in 10 mM Tris, 1 mM EGTA, 1 mM DTT, 10% sucrose, pH 7.5) to yield extract A (XA), then buffer B (buffer A plus 1% triton) to yield extract B (XB), then buffer C (buffer A plus 1% n-laurylsarcosyl) to yield extract C (XC), and finally 70% formic acid (FA) to yield extract D (XD) exactly as described previously by us (18 , 19) ; this was repeated with three separate pieces of tissue from each AD patient to generate three sets of AD-XA and AD-XD extracts for each of the five patients. Tissue from the same controls was also extracted directly into 70% formic acid to yield extract F (XF) (19 , 20) ; this was repeated to yield 2 C-XF sets for each of the five controls. Extracts from individuals were used for Western blots. For liquid chromatography (LC) -tandem mass spectrometry (MS-MS), equal amounts (by original wet tissue wt) of AD-XA, AD-XD, or C-XF from each set were combined to make a pooled sample from each of the five AD patients or five controls. Each pooled C-XF (n=2), AD-XA (n=3), and AD-XD (n=3) sample was desalted, alkylated with iodoacetamide, cleaved with trypsin, and prepared for LC-MS-MS exactly as described (18 , 19) .

Mass spectrometry
Each pooled sample of digested peptides was separately purified with a C18 solid-phase extraction column (OasisR MCX, Milford, MT, USA), then analyzed using a LCQ (Thermo Electron, San Jose, CA, USA) with 12-step 2-dimensional microcapillary high performance LC (strong cation exchange column with two alternating reversed-phase C18 columns) followed by MS-MS exactly as described previously (18 , 19) . The error in mass measurement by LCQ is ± 0.2 amu. MS-MS data files were searched against the International Protein Database using SEQUEST. The six proteins selected for investigation had ProteinProphet probability scores greater than 0.95 in all preparations (18 , 19 , 21) .

P-Mod analysis
Each of the MS-MS data files for the triplicate AD-XA, triplicate AD-XD, and duplicate C-XF sets was evaluated separately by P-MOD searching against in silico-generated tryptic peptide fragments (5 to 30 amino acids) with and without missed lysines from the following six proteins sequences obtained from the SWISS PROT database: antibody (Ab)42 (P05067), apoE (P02649) {alpha}-III tubulin (Q71U36), ß-III tubulin (Q13509), tau (P10636), and GFAP (P14136); P-MOD output files were then combined for each of the six proteins. Only P-MOD sequence-to-spectrum matches with P < 0.01 were retained for further analysis. In an attempt to control for the amount of material being evaluated, the prevalence of mass shifts was expressed as the number of P-MOD identified peptides with a particular mass shift (P<0.01) divided by the total number of scorable spectra reported by P-MOD in that extract.

CNBr cleavage and Western blot
Individual AD-XD and C-XF preparations were dried, resuspended in water, and protein concentration was determined using the Bio-Rad DC Reagent Kit (Bio-Rad, Hercules, CA, USA). 50 µl of each was dried under vacuum, then resuspended in either 500 µl 70% formic acid or 100 mM CNBr plus 70% formic acid per 12.4 µg total protein; this achieved a 100-fold molar excess of CNBr to tubulin that was estimated to be ~10% total cellular protein (22 23 24) . Samples were digested overnight, dried, and resuspended in Laemmli sample buffer containing 200 mM DTT for separation by SDS-PAGE using Tris-Tricine Ready Gels (Bio-Rad). Western blots were performed exactly as described previously (25) using monoclonal anti-{alpha}-tubulin Ab (Sigma Chemical Co., St. Louis, MO, USA), monoclonal anti-ß-III tubulin Ab (Covance Research Products, Babco, CA, USA), monoclonal 4G8 against Aß peptide amino acids17 to 24 (Signet Laboratories, Inc., Dedham, MA, USA), and affinity isolated polyclonal antibody to tau phosphorylated at position 231 (Sigma Chemical Co.). Films from Western blots were digitized and band density integrated with ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Statistical analyses were performed using GraphPad Prism (San Diego, CA, USA).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Cortex from the superior and middle temporal gyri from 13 individuals was used in these experiments (Table 1 ). Representative Western blots for Aß-, apoE-, and P-tau231-immunoreactive proteins in serial extracts from two AD patients and two controls are presented in Fig. 1 and show disease-specific accumulation of all three proteins in XD. We previously demonstrated that other proteins are selectively present in XD from AD patients but not control individuals, including tubulins and GFAP (19) .


View this table:
[in this window]
[in a new window]
  		Table 1. Information on individuals whose tissue was used in these experimentsa


Figure 1
View larger version (51K):
[in this window]
[in a new window]
  		Figure 1. Representative Western blots of extracts from temporal cortex of two patients with AD and two controls, denoted by subscripts (see Table 1 ). Tissue was serially extracted in aqueous buffer (A), nonionic detergent (B), ionic detergent (C), and finally any remaining protein was extracted by 70% formic acid (D). Three blots are shown that correspond to primary antibodies directed at Aß peptides, P-tau231, or apoE. Note immunoreactive material in extract D from both AD patients that is not present in either control individual.

Among the LC-MS-MS spectra from the eight sets of pooled extracts from five AD patients or five controls (3 AD-XA, 3 AD-XD, 2 C-XF), P-MOD reported a total of 327,297 scorable spectra; the average number (±SD) of scorable spectra for P-MOD analysis for each set was 35877 ± 7052 (n=2) for C-XF, 33626 ± 1360 (n=3) for AD-XA, and 51554 ± 1037 (n=3) for AD-XD (ANOVA for these data had P<0.05). Among these scorable spectra, P-MOD identified 8618 peptides with sequence-to-spectrum matches having P < 0.01 for the six proteins under investigation (Table 2 ). We observed P-MOD-identified mass shifts in these two extracts ranging from small losses (<100 Da) up to additions that reached almost 1200 Da. All six proteins had extensive coverage (see Table 2 ) except for Aß in C-XF, where no sequence-to-spectrum match with P < 0.01 was identified; this perhaps is not surprising given the low abundance of Aß peptide in healthy brain. Since the number of scorable spectra varied significantly among extracts, we divided the number of P-MOD matches for each mass shift by the number of scorable spectra in that extract to give an estimate of prevalence of each P-MOD-identified mass shift. The ratio of the prevalence in AD-XD to AD-XA was calculated for each mass shift identified by P-MOD. This relative prevalence in insoluble vs. soluble fractions from the same tissue was transformed by log2 and plotted against mass shift. For all six proteins, the vast majority of peptides had prevalence ratios in AD-XD vs. AD-XA of ~1 (log2 1=0).


View this table:
[in this window]
[in a new window]
  		Table 2. Protein coverage and number of sequence-to-spectrum matches by P-MOD with P < 0.01a

In analogy to gene expression data, we selected upper and lower limits for the ratio of mass shift relative prevalence; in this case, log2 of +3 or –3 is an 8-fold increase or decrease in AD-XD relative to AD-XA (Fig. 2 ). ApoE and GFAP, both of the primarily astrocyte-derived proteins, and {alpha}-III tubulin, one of the two neuron-enriched tubulins, were not included in Fig. 2 because no mass shift on any of these three proteins achieved log2 of relative prevalence in AD-XD vs. AD-XA that was <–3 or >3. In contrast, we did observe relative prevalence that exceeded our cut points in AD-XD vs. AD-XA for nine mass shifts on Aß, tau, and ß-III tubulin; potential identification of these most prevalent mass shifts was assigned using Delta Mass (www.abrf.org) and UniMod (www.unimod.org) web engines (Table 3 ). The only mass shift on Aß that exceeded our cut points was an increase of +16 Da in AD-XD, indicative of oxidation or hydroxylation of detergent-insoluble forms of these peptides relative to their soluble forms from the same AD tissue. The most prevalent mass shifts on ß-III tubulin also were an increase in +16 Da as well as +70 Da in AD-XD, the latter consistent with adduction by crotonaldehyde, a product of lipid peroxidation (26 27 28) . Tau contained the largest number of prevalent PTMs, including + 70 Da, but also a number of other PTMs related to glycation and glycosylation, processes previously localized to NFTs in AD brain; three of these were decreased in AD-XD relative to AD-XA (10) . Although falling slightly below our cutoff of log2 > 3, apparent phosphorylation (+80 Da) on tau had log2 = 2.6 for its ratio in AD-XD vs. AD-XA, and contingency analysis had P < 0.0001 for the proportion of tau peptides observed that had P-MOD-identified +80 Da in AD-XD vs. AD-XA or C-XF.


Figure 2
View larger version (20K):
[in this window]
[in a new window]
  		Figure 2. Plot of mass shifts (Da) vs. log2 of their relative prevalence in AD-XD (sarcosyl-insoluble fraction) compared to AD-XA (soluble in aqueous buffer) for Aß, ß-III tubulin, and tau. Prevalence was calculated by dividing the number peptides with each mass shift by total scorable spectra for that sample. Relative prevalence is the ratio of a mass shift’s prevalence in AD-XD divided by it’s prevalence in AD-XA. A few instances of low prevalence, high molecular mass shifts in AD-XD with no corresponding mass shift detected in AD-XA were not plotted. Dotted lines mark log2 of 3 (8-fold increase in AD-XD vs. AD-XA) and –3 (8-fold decrease in AD-XD vs. AD-XA); no mass shift on apoE, GFAP, or {alpha}-III tubulin exceeded these cut points.


View this table:
[in this window]
[in a new window]
  		Table 3. Mass shift, log2 of relative prevalence in AD-XD vs. AD-XA, and tentative identificationa

We focused on mass shifts of +16 Da in part because they are relatively straightforward to interpret and were most prevalent in AD-XD for two of the six proteins investigated (Table 4 ). We expanded our analysis to include +32 Da to capture possible methionine sulfone formation and performed contingency analyses for all six proteins; these showed that the proportion of peptides with +16 Da shifts increased significantly in AD-XD compared to AD-XA or C-XF for Aß (P<0.01) and ß-III tubulin (P<0.0001), but not apoE, GFAP, tau, or {alpha}-III tubulin. No protein had a significant change in the proportion of +32 Da shifts across the three fractions.


View this table:
[in this window]
[in a new window]
  		Table 4. Number of peptides with mass shift of +16 or +32 Da in C-XF, AD-XA, and AD-XD compared to total number of sequence-to-spectrum matches for that protein by P-MOD with P < 0.01a

P-MOD not only identifies peptide mass shifts but also maps them to specific amino acids. Figure 3 shows a representative m/z spectrum identified by PMOD as Aß1–42 tryptic peptide G29-A42 with a +16 mass shift. The amino acid sequence m/z for the expected daughter (b- and y-) ions, actual m/z detected, and calculated mass shifts are shown. PMOD maps the mass shift by determining where the mass difference between expected and observed becomes greater than 1 for both the b- or y-ions, in this case at M35. A map of all +16 Da mass shifts identified by P-MOD for AD-XA and AD-XD (no matches to Aß tryptic peptides with P < 0.01 were identified in C-XF) is presented in Fig. 4 ; here we have plotted the normalized frequency at each position by dividing the number of times + 16 Da was mapped to that position by the number of peptides observed for that protein. Confirming the direct biochemical data of others, the +16 Da shift on Aß peptides identified by P-MOD centered on M35 (9) . This characteristic oxidative modification of Aß peptides was restricted exclusively to AD-XD and was not observed in AD-XA. Lower abundance +16 Da shifts were observed in AD-XA but on different amino acids.


Figure 3
View larger version (29K):
[in this window]
[in a new window]
  		Figure 3. Example of P-MOD-generated data for mapping Aß peptides with M + 16 Da. P-MOD sequence-to-spectrum matches to b-ion series are in blue and y-ion series are in red for a tryptic fragment of Aß1–42 (G29-A42) from AD-XD (m/z 589.63 is M0 + H2O). The table shows the amino acid sequence, expected daughter ions m/z, the actual P-MOD-detected daughter ions m/z, and the corresponding mass shifts. Note that the b- or y-ions containing M35 had a mass shift of ~+16 Da. ND means not detected.


Figure 4
View larger version (14K):
[in this window]
[in a new window]
  		Figure 4. Map of normalized frequency of +16 Da shift identified by P-MOD on Aß peptides from AD-XA and AD-XD. Normalized frequency at each position was determined by dividing the number of times +16 Da was mapped to that position by the number of peptides observed for that protein. There were no P-MOD sequence-to-spectra matches with P < 0.01 for tryptic peptides of Aß in C-XF.

Although not a focus of this work, we also examined P-MOD mapping of +80 Da on tau as an additional means of validating our approach. P-MOD mapped the most abundant +80 Da mass shifts on tau in AD-XD to S214, T231, S262, S285, S356, S396, and S400 (not shown), confirming the results of others who used phospho-specific antibodies to identify these sites of tau phosphorylation in AD brain (10) .

While methionine oxidation has been reported for Aß, to our knowledge this is the first report of selective M oxidation of the neuron-enriched cytoskeletal protein ß-III tubulin. As shown in Table 4 , ß-III tubulin but not {alpha}-III tubulin had a significantly increased proportion of +16 Da shifts in AD-XD identified by P-MOD. Figure 5 contains a sample m/z spectrum of a tryptic peptide fragment from ß-III tubulin (A63-R77) in AD-XD, m/z for the expected daughter ions, actual m/z for the daughter ions detected, and calculated mass shifts. Note that detected b- or y-ions containing M73 had mass differences of ~+16 Da. Figure 6 A to 6C show full-length ß-III tubulin maps for all +16 Da shifts identified by P-MOD in the three fractions. +16 Da on P272 and D295 are present at similar normalized frequencies in all three extracts and likely represent aspects of normal physiology, likely hydroxylations of these amino acids. As far as we are aware, this is a novel observation for tubulins. Low frequency oxidation of M257 was observed in C-XF but not AD-XA; however, several new M sulfoxides were mapped in AD-XD (M73, M104, M267, M293, M330, M338) as well as a marked increase in the normalized frequency of M257 + 16 Da. A composite map of amino acids 250 to 300 (Fig. 6D ) clarifies the distribution of +16 Da shifts in this crowded region of ß-III tubulin.


Figure 5
View larger version (26K):
[in this window]
[in a new window]
  		Figure 5. Example of P-MOD-generated data for mapping mass shifts for ß-III tubulin with M + 16 Da. P-MOD sequence-to-spectrum matches to b-ion series are in blue and y-ion series are in red for a tryptic fragment of ß-III tubulin (A63-R77) from AD-XD. The table shows the sequence, expected daughter ions m/z, actual P-MOD-detected daughter ions m/z, and the corresponding mass shifts. Note that the b- or y-ions containing M73 had a mass difference of ~+16 Da. ND means not detected.


Figure 6
View larger version (12K):
[in this window]
[in a new window]
  		Figure 6. Maps of normalized frequency (calculated as described in Fig. 4 legend) of +16 Da shift identified by P-MOD on full-length ß-III tubulin in A) C-XF, B) AD-XA, and C) AD-XD as well as D) a composite map for all three preparations from amino acid 250 to 300. +16 Da shifts on P272 and D295 likely represent hydroxylations and are present at similar frequency in all three preparations. All remaining + 16 Da shifts were mapped to methionines as indicated and likely represent sulfoxide formation.

While we validated the power of P-MOD to identify and map known PTMs on Aß and tau, our novel observation of selective methionine oxidation of ß-IIItubulin required biochemical confirmation. To do this, we pursued independent verification of the most prevalent adduct, methionine oxidation, on ß-III tubulin. This was done by determining the efficiency of CNBr-mediated cleavage, which is blocked by oxidized methionine (29) . We used AD-XD from the same 5 AD patients whose tissue was used in P-MOD analysis and compared CNBr cleavage to C-XF from temporal cortex of three individuals who died without clinical evidence of neurological disease and who had age-related changes only in brain by neuropathologic examination (Table 1) . Protein extracts from controls and AD patients were subjected to cleavage with CNBr, separated by SDS-PAGE, and probed by Western blots with antibodies against {alpha} and ß-III tubulin; antibodies specific to neuron-enriched {alpha}-III tubulin are not available (Fig. 7 ). Our results showed that CNBr treatment cleaved virtually all {alpha} and ß-III tubulin extracted from control tissue, while cleavage of ß-III tubulin was selectively and significantly reduced in extracts from AD tissue (P≤0.01).


Figure 7
View larger version (52K):
[in this window]
[in a new window]
  		Figure 7. Cleavage of {alpha} and ß-III tubulins with CNBr. AD-XD from the same five AD patients analyzed by P-MOD and C-XF from three controls (Table 1) were prepared from temporal cortex, subjected to cleavage by CNBr, separated by SDS-PAGE, probed with antibodies to {alpha} or ß-III tubulin, and integrated band densities were determined from all samples. A) Data are the percent reduction in band density from samples run in parallel without CNBr. Two-way ANOVA had P < 0.05 for {alpha} vs. ß-III tubulin, P < 0.01 for C-XF vs. AD-XD, and P < 0.05 for interaction between these two terms. Bonferroni-corrected post-tests had P < 0.01 for AD-XD but P > 0.05 for C-XF. B) Representative Western blot of C-XF and AD-XD fractions after incubation with (+) or without (–) CNBr and probed with ß-III tubulin Ab. Note the smearing of ß-III tubulin immunoreactivity that is observed for many proteins in AD-XD but not in C-XF extracts.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Here we tested the hypothesis that abnormal PTMs specifically accumulate within known functional domains of neuron-enriched tubulins. Our study is unique in several respects: we used P-MOD to perform a comprehensive search of PTMs on selected proteins, estimated their relative enrichment in the pathological detergent-insoluble fraction of AD brain, and mapped some of these PTMs that were significantly disproportionately observed in the detergent-insoluble fraction. Using this approach, we discovered selective methionine oxidations on detergent-insoluble ß-III tubulin to the exclusion of its heterodimeric partner, and confirmed these findings by independent biochemical means.

A critical issue is distinguishing physiological from pathophysiologic PTMs. We approached this by comparing not only control tissue to AD tissue, but soluble (normal) protein and detergent-insoluble (pathological) protein obtained from the same AD tissue. This strategy enabled us to identify pathological PTMs and enhanced the likelihood that the identified PTMs contribute to disease pathogenesis. Comparing fractions from the same tissue also removed variance from interindividual differences as well as from tissue procurement, storage, and preparation. Accumulation of multiple detergent-insoluble proteins is a fundamental biochemical abnormality that, with the exception of tau, occurs even before the onset of dementia in AD patients and likely reflects some combination of misfolding, aggregation, and PTM along with damage to cellular lipid; this fundamental feature of AD is lacking in current transgenic models (19) . Recent data have highlighted a likely central role for soluble Aß peptides in initiating neurotoxicity. However, these findings do not discount the likely key pathophysiologic role for modifications on neuronal cytoskeletal proteins that lead to their detergent insolubility and thereby contribute to neuronal dysfunction and death.

Our data for Aß M35 sulfoxide formation and tau hyperphosphorylation at multiple sites confirmed direct biochemical data of others and thus strongly validated of our approach of using P-MOD to survey PTMs in AD tissue (9 , 10 , 30 , 31) . In contrast to these known modifications of Aß and tau, GFAP and apoE served as controls for cell specificity; indeed, results from both primarily astrocyte-derived proteins suggested that AD-associated PTMs were accumulating largely on neuronal proteins, a finding consonant with numerous immunohistochemical studies of AD brain (32) . Nevertheless, apoE colocalizes to pathological structures in AD and it accumulates in the detergent-insoluble pool (19) . As noted above, there are several proposed mechanisms by which proteins may accumulate in this pathological fraction; our data suggest that mechanisms other than PTM may be the major contributors to the accumulation of detergent-insoluble apoE as previously suggested by others (33) . In combination, these results strongly indicate that the novel approach developed here was an accurate survey of PTMs in AD brain that performed within expected parameters set by direct biochemical and immunochemical observations made by others and us.

In addition to confirming direct biochemical observations about Aß M35 oxidation and tau phosphorylation, we also made novel observations with respect to Aß and tau PTMs in AD. To our knowledge, ours is the first demonstration that Aß M35 sulfoxide is largely restricted to the detergent-insoluble pool. Moreover, our data were the first demonstration that, in contrast to Aß peptides, detergent-insoluble tau proteins were significantly and extensively decorated with a variety of PTMs that were enriched in the detergent-insoluble fraction to levels comparable to phosphorylation. The functional significance of M35 oxidation of Aß peptides is complex, with data from different in vitro and cell culture models suggesting both increased and decreased neurotoxicity (34) . Similarly, whether increased tau phosphorylation in vivo is the cause or consequence of neuronal cytoskeletal disruption in AD remains unclear, although some of the sites of increased phosphorylation in detergent-insoluble tau identified by P-MOD, S262, T231, and S235 have been shown to inhibit tau binding to microtubules by up to 35% (35) . Although not shown, several of the tau PTMs mapped within the microtubule binding regions and therefore have the potential to modify its ability to stabilize microtubules (36 37 38 39) .

Another novel finding was P-MOD’s identification of prevalent methionine oxidation (+16 Da) and other PTMs consistent with crotonaldehyde adducts on ß-III tubulin but not its herterodimeric partner {alpha}-III tubulin; we confirmed selective modification of methionines in ß-III tubulin with CNBr cleavage followed by Western blots. The tentative identification of selective crotonaldehyde adducts on ß-III tubulin and tau will require further investigation; however, it is supported by several lines of biochemical and immunohistochemical data (32) . The potential functional significance of specific of ß-III tubulin methionine oxidations was considered using bovine and porcine brain heterodimer models submitted to the Protein Data Bank (40 , 41) (http://www.rcsb.org/pdb/index.html). Oxidation of ß-III tubulin methionines was observed at increased frequency in AD-XD fractions in domains that subserve GTPase or protein-protein interaction functions of ß-III tubulin (A63-R77, V155-R162, I163-K175, L263-R276, and A283-K297), as well as a region (L253-R262) adjacent to the MAP binding region. M in region L263-R276 is normally inaccessible in ß-III tubulin (40 , 42) , suggesting that misfolding may have preceded methionine oxidation in some regions of ß-III tubulin.

Earlier data from mouse models and our previous observations with tissue from patients with prodromal AD suggest that Aß and ß-III tubulin accumulate as detergent-insoluble forms very early in the course of AD pathogenesis whereas tau does so late in AD (6 , 19 , 43) . Combined with results from the current work, we speculate that direct oxidation reactions with Aß peptides yielded M35 sulfoxide as well as oxidized surrounding protein and lipid (44 , 45) and that these reactions relatively selectively targeted ß-III tubulin, which likely contributed to destabilization of the neuronal cytoskeleton. These findings are a potential link between the oxidative stress and neuronal cytoskeletal disruption that are characteristic of AD.


   ACKNOWLEDGMENTS
 
This work was supported by the Nancy and Buster Alvord Endowment as well as grants from the National Institutes of Health, including AG24011, AG05136, AG022040, and AG023483.

Received for publication February 6, 2006. Accepted for publication March 14, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Hardy, J., Selkoe, D. J. (2002) The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science 297,353-356[Abstract/Free Full Text]
  2. Praprotnik, D., Smith, M. A., Richey, P. L., Vinters, H. V., Perry, G. (1996) Filament heterogeneity within the dystrophic neurites of senile plaques suggests blockage of fast axonal transport in Alzheimer’s disease. Acta Neuropathol. (Berlin) 91,226-235[CrossRef][Medline]
  3. Hirai, K., Aliev, G., Nunomura, A., Fujioka, H., Russell, R. L., Atwood, C. S., Johnson, A. B., Kress, Y., Vinters, H. V., Tabaton, M., Shimohama, S., Cash, A. D., Siedlak, S. L., Harris, P. L., Jones, P. K., Petersen, R. B., Perry, G., Smith, M. A. (2001) Mitochondrial abnormalities in Alzheimer’s disease. J. Neurosci. 21,3017-3023[Abstract/Free Full Text]
  4. Terry, R. D. (1996) The pathogenesis of Alzheimer disease: an alternative to the amyloid hypothesis [see comments]. J. Neuropathol. Exp. Neurol. 55,1023-1025[Medline]
  5. Scheff, S. W., DeKosky, S. T., Price, D. A. (1990) Quantitative assessment of cortical synaptic density in Alzheimer’s disease. Neurobiol. Aging 11,29-37[CrossRef][Medline]
  6. Stokin, G. B., Lillo, C., Falzone, T. L., Brusch, R. G., Rockenstein, E., Mount, S. L., Raman, R., Davies, P., Masliah, E., Williams, D. S., Goldstein, L. S. (2005) Axonopathy and transport deficits early in the pathogenesis of Alzheimer’s disease. Science 307,1282-1288[Abstract/Free Full Text]
  7. Iqbal, K., Alonso, A., Gong, C., Khatoon, S., Kudo, T., Singh, T., Grundke-Iqbal, I. (1993) Molecular pathology of Alzheimer neurofibrillary degeneration. Acta Neurobiol. Exp. (Warsaw) 53,325-335
  8. Buee, L., Hamdane, M., Delobel, P., Sambo, A. V., Begard, S., Ghestem, A., Sergeant, N., Delacourte, A. (2002) [Tau story: from frontotemporal dementia to other tauopathies]. J. Soc. Biol. 196,103-108[Medline]
  9. Naslund, J., Schierhorn, A., Hellman, U., Lannfelt, L., Roses, A. D., Tjernberg, L. O., Silberring, J., Gandy, S. E., Winblad, B., Greengard, P., et al (1994) Relative abundance of Alzheimer A beta amyloid peptide variants in Alzheimer disease and normal aging. Proc. Natl. Acad. Sci. U. S. A. 91,8378-8382[Abstract/Free Full Text]
  10. Gong, C. X., Liu, F., Grundke-Iqbal, I., Iqbal, K. (2005) Post-translational modifications of tau protein in Alzheimer’s disease. J. Neural Transm. 112,813-838[CrossRef][Medline]
  11. Castegna, A., Aksenov, M., Aksenova, M., Thongboonkerd, V., Klein, J. B., Pierce, W. M., Booze, R., Markesbery, W. R., Butterfield, D. A. (2002) Proteomic identification of oxidatively modified proteins in Alzheimer’s disease brain. Part I: creatine kinase BB, glutamine synthase, and ubiquitin carboxy-terminal hydrolase L-1. Free Radic. Biol. Med. 33,562-571[CrossRef][Medline]
  12. Castegna, A., Aksenov, M., Thongboonkerd, V., Klein, J. B., Pierce, W. M., Booze, R., Markesbery, W. R., Butterfield, D. A. (2002) Proteomic identification of oxidatively modified proteins in Alzheimer’s disease brain. Part II: dihydropyrimidinase-related protein 2, alpha-enolase and heat shock cognate 71. J. Neurochem. 82,1524-1532[CrossRef][Medline]
  13. Pamplona, R., Dalfo, E., Ayala, V., Bellmunt, M. J., Prat, J., Ferrer, I., Portero-Otin, M. (2005) Proteins in human brain cortex are modified by oxidation, glycoxidation, and lipoxidation. Effects of Alzheimer disease and identification of lipoxidation targets. J. Biol. Chem. 280,21522-21530[Abstract/Free Full Text]
  14. Hansen, B. T., Davey, S. W., Ham, A. J., Liebler, D. C. (2005) P-Mod: an algorithm and software to map modifications to peptide sequences using tandem MS data. J. Proteome Res. 4,358-368[CrossRef][Medline]
  15. Schiewe, A. J., Margol, L., Soreghan, B. A., Thomas, S. N., Yang, A. J. (2004) Rapid characterization of amyloid-beta side-chain oxidation by tandem mass spectrometry and the scoring algorithm for spectral analysis. Pharm. Res. 21,1094-1102[CrossRef][Medline]
  16. Vanacore, R. M., Friedman, D. B., Ham, A. J., Sundaramoorthy, M., Hudson, B. G. (2005) Identification of S-hydroxylysyl-methionine as the covalent cross-link of the noncollagenous (NC1) hexamer of the alpha1alpha1alpha2 collagen IV network: a role for the post-translational modification of lysine 211 to hydroxylysine 211 in hexamer assembly. J. Biol. Chem. 280,29300-29310[Abstract/Free Full Text]
  17. . The National Institute on Aging and Reagan Institute Working Group on Diagnostic Criteria for the Neuropathological Assessment of Alzheimer’s Disease (1997) Consensus recommendations for the postmortem diagnosis of Alzheimer’s disease. Neurobiol. Aging 18,S1-S2[CrossRef][Medline]
  18. Wang, Q., Woltjer, R. L., Cimino, P. J., Pan, C., Montine, K. S., Zhang, J., Montine, T. J. (2005) Proteomic analysis of neurofibrillary tangles in Alzheimer disease identifies GAPDH as a detergent-insoluble paired helical filament tau binding protein. FASEB J. 19,869-871[Abstract/Free Full Text]
  19. Woltjer, R. L., Cimino, P. J., Boutte, A. M., Schantz, A. M., Montine, K. S., Larson, E. B., Bird, T., Quinn, J. F., Zhang, J., Montine, T. J. (2005) Proteomic determination of widespread detergent insolubility, including Abeta but not tau, early in the pathogenesis of Alzheimer’s disease. FASEB J. 19,1923-1925, Epub 2005 Aug 29[Abstract/Free Full Text]
  20. Zhou, Y., Gu, G., Goodlett, D. R., Zhang, T., Pan, C., Montine, T. J., Montine, K. S., Aebersold, R. H., Zhang, J. (2004) Analysis of alpha-synuclein-associated proteins by quantitative proteomics. J. Biol. Chem. 279,39155-39164[Abstract/Free Full Text]
  21. Zhang, J., Goodlett, D. R., Peskind, E. R., Quinn, J. F., Zhou, Y., Wang, Q., Pan, C., Yi, E., Eng, J., Aebersold, R. H., Montine, T. J. (2005) Quantitative proteomic analysis of age-related changes in human cerebrospinal fluid. Neurobiol. Aging 26,207-227[CrossRef][Medline]
  22. Hollemeyer, K., Heinzle, E., Tholey, A. (2002) Identification of oxidized methionine residues in peptides containing two methionine residues by derivatization and matrix-assisted laser desorption/ionization mass spectrometry. Proteomics 2,1524-1531[CrossRef][Medline]
  23. Kaiser, R., Metzka, L. (1999) Enhancement of cyanogen bromide cleavage yields for methionyl-serine and methionyl-threonine peptide bonds. Anal. Biochem. 266,1-8[CrossRef][Medline]
  24. McLaughlin, L., Zemlan, F. P., Dean, G. E. (1997) Identification of microtubule-associated protein tau isoforms in Alzheimer’s paired helical filaments. Brain Res. Bull. 43,501-508[CrossRef][Medline]
  25. Neely, M. D., Boutte, A., Milatovic, D., Montine, T. J. (2005) Mechanisms of 4-hydroxynonenal-induced neuronal microtubule dysfunction. Brain Res. 1037,90-98[CrossRef][Medline]
  26. Esterbauer, H., Schaur, R. J., Zollner, H. (1991) Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic. Biol. Med. 11,81-128[CrossRef][Medline]
  27. Shibata, N., Nagai, R., Uchida, K., Horiuchi, S., Yamada, S., Hirano, A., Kawaguchi, M., Yamamoto, T., Sasaki, S., Kobayashi, M. (2001) Morphological evidence for lipid peroxidation and protein glycoxidation in spinal cords from sporadic amyotrophic lateral sclerosis patients. Brain Res. 917,97-104[CrossRef][Medline]
  28. Ichihashi, K., Osawa, T., Toyokuni, S., Uchida, K. (2001) Endogenous formation of protein adducts with carcinogenic aldehydes: implications for oxidative stress. J. Biol. Chem. 276,23903-23913[Abstract/Free Full Text]
  29. Shechter, Y., Burstein, Y., Patchornik, A. (1975) Selective oxidation of methionine residues in proteins. Biochemistry 14,4497-4503[CrossRef][Medline]
  30. Kuo, Y. M., Kokjohn, T. A., Beach, T. G., Sue, L. I., Brune, D., Lopez, J. C., Kalback, W. M., Abramowski, D., Sturchler-Pierrat, C., Staufenbiel, M., Roher, A. E. (2001) Comparative analysis of amyloid-beta chemical structure and amyloid plaque morphology of transgenic mouse and Alzheimer’s disease brains. J. Biol. Chem. 276,12991-12998[Abstract/Free Full Text]
  31. Dong, J., Atwood, C. S., Anderson, V. E., Siedlak, S. L., Smith, M. A., Perry, G., Carey, P. R. (2003) Metal binding and oxidation of amyloid-beta within isolated senile plaque cores: Raman microscopic evidence. Biochemistry 42,2768-2773[CrossRef][Medline]
  32. Montine, T. J., Neely, M. D., Quinn, J. F., Beal, M. F., Markesbery, W. R., Roberts, L. J., 2nd, Morrow, J. D. (2002) Lipid peroxidation in aging brain and Alzheimer’s disease. Free Radic. Biol. Med. 33,620-626[CrossRef][Medline]
  33. Christie, R. H., Chung, H., Rebeck, G. W., Strickland, D., Hyman, B. T. (1996) Expression of the very low-density lipoprotein receptor (VLDL-r), an apolipoprotein-E receptor, in the central nervous system and in Alzheimer’s disease. J. Neuropathol. Exp. Neurol. 55,491-498[Medline]
  34. Butterfield, D. A., Boyd-Kimball, D. (2005) The critical role of methionine 35 in Alzheimer’s amyloid beta-peptide (1–42)-induced oxidative stress and neurotoxicity. Biochim. Biophys. Acta 1703,149-156[Medline]
  35. Sengupta, A., Kabat, J., Novak, M., Wu, Q., Grundke-Iqbal, I., Iqbal, K. (1998) Phosphorylation of tau at both Thr 231 and Ser 262 is required for maximal inhibition of its binding to microtubules. Arch. Biochem. Biophys. 357,299-309[CrossRef][Medline]
  36. Lee, G., Neve, R. L., Kosik, K. S. (1989) The microtubule binding domain of tau protein. Neuron 2,1615-1624[CrossRef][Medline]
  37. Gustke, N., Trinczek, B., Biernat, J., Mandelkow, E. M., Mandelkow, E. (1994) Domains of tau protein and interactions with microtubules. Biochemistry 33,9511-9522[CrossRef][Medline]
  38. Gamblin, T. C., Berry, R. W., Binder, L. I. (2003) Tau polymerization: role of the amino terminus. Biochemistry 42,2252-2257[CrossRef][Medline]
  39. Perez, M., Arrasate, M., Montejo De Garcini, E., Munoz, V., Avila, J. (2001) In vitro assembly of tau protein: mapping the regions involved in filament formation. Biochemistry 40,5983-5991[CrossRef][Medline]
  40. Nogales, E., Wolf, S. G., Downing, K. H. (1998) Structure of the alpha beta tubulin dimer by electron crystallography. Nature 391,199-203[CrossRef][Medline]
  41. Nogales, E. (1999) A structural view of microtubule dynamics. Cell. Mol. Life Sci. 56,133-142[CrossRef][Medline]
  42. de Pereda, J. M., Andreu, J. M. (1996) Mapping surface sequences of the tubulin dimer and taxol-induced microtubules with limited proteolysis. Biochemistry 35,14184-14202[CrossRef][Medline]
  43. Oddo, S., Caccamo, A., Kitazawa, M., Tseng, B. P., LaFerla, F. M. (2003) Amyloid deposition precedes tangle formation in a triple transgenic model of Alzheimer’s disease. Neurobiol. Aging 24,1063-1070[CrossRef][Medline]
  44. Ali, F. E., Separovic, F., Barrow, C. J., Cherny, R. A., Fraser, F., Bush, A. I., Masters, C. L., Barnham, K. J. (2005) Methionine regulates copper/hydrogen peroxide oxidation products of Abeta. J. Pept. Sci. 11,353-360[CrossRef][Medline]
  45. Barnham, K. J., Ciccotosto, G. D., Tickler, A. K., Ali, F. E., Smith, D. G., Williamson, N. A., Lam, Y. H., Carrington, D., Tew, D., Kocak, G., Volitakis, I., Separovic, F., Barrow, C. J., Wade, J. D., Masters, C. L., Cherny, R. A., Curtain, C. C., Bush, A. I., Cappai, R. (2003) Neurotoxic, redox-competent Alzheimer’s beta-amyloid is released from lipid membrane by methionine oxidation. J. Biol. Chem. 278,42959-42965[Abstract/Free Full Text]
  46. Mirra, S. S. (1997) The CERAD neuropathology protocol and consensus recommendations for the postmortem diagnosis of Alzheimer’s disease: a commentary. Neurobiol. Aging 18,S91-S94[CrossRef][Medline]
  47. Braak, H., Braak, E. (1991) Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol. (Berlin) 82,239-259[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Boutte, A. M.
Right arrow Articles by Montine, T. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Boutte, A. M.
Right arrow Articles by Montine, T. J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS