Aluminum-induced Nonenzymatic Phospho-incorporation into Human ...

11 downloads 0 Views 5MB Size Report
Jan 12, 1993 - However, controversy exists over .... Protein amounts in the collected filtrates ... aluminum, progressively reduced amounts were observed to.
THEJOURNAL OF BIOLOGICAL CHEMISTRY Q 1993 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 268, No. 16,Issue of June 5 , pp. 1197611981,1993 Printed in U.S.A.

Aluminum-induced Nonenzymatic Phospho-incorporationinto Human Tau and OtherProteins* (Received for publication, January 12, 1993)

Mossaad Abdel-Ghany, Abdel KhalekEl-SebaeS, and David Shalloway$ From the Section of Biochem&y, Molecular and Cell Biology, Cornell University, Zthuca, New York 14853

Incubation of purified recombinant human tau pro- in these diseases, it has been demonstrated that aluminum tein with aluminum salts at concentrations r l O O p~ salts can induce a variety of physiological effects in culture induces aggregation of tau that prevents its entry into and animalsystems. For example, in vivo aluminum treatment SDS-polyacrylamidegels and filtration through nylon can: 1) alter neurofilament protein phosphorylation (19-21); membranes. Thiseffect is noncovalent and can be re- 2) induce cytoskeletal changes in human neuronal cultures versed by addition of EDTA. However, when incubated (22, 23); 3) reduce expression of mRNA coding for neurofilaalong with ATP, GTP, or CTP, aluminum catalyzes a ment subunits (24); 4)induce expression of an immunological covalent linkage that results in incorporationof the a- epitope associated with Alzheimer’s disease on human neuroand 7-phosphates into the tau protein (phospho-incorporation). Thesensitivity to phosphatases and partial blastoma cells (25, 26); and 5) induce neurofibrillary tangles hydrolysis and the labeling observed with ATP con- in cultured rat neurons (13,27,28). These in vivo effects may taining radioisotopes at different positions suggest a be related to one or more of the biochemical effects of aluThese include its ability to: 1) promote tubulin novel reaction in which the entire triphosphate moiety minum in vitro. is transferred from ATP and linked to tau via an 0- polymerization (29); 2)promote the aggregation (30) and alter linkage to the a-phosphate. The aggregation and tri- the electrophoretic properties (31) of highly phosphorylated phosphorylation phenomenawere not catalyzed bydi- neurofilament subunits from brainextracts; 3) inhibitdevalent or quadrivalent cations, but similar effects were phosphorylation of tau protein in synaptosomal cytosol fracobserved with some other trivalent cations. They oc- tions (32); and 4) as aluminum fluoride, to activate G proteins’ curredataluminumconcentrationssimilartothose and modulate adenylate cyclase activity (33-35). found in human brainswith Alzheimer’s disease, sugAluminum is transported in the blood bybinding to human gesting the possibility that related reactions may haveserum transferrin and serum albumin (36), and it can be physiological significance in vivo. concentratedin neural tissues at surprisingly highlevels, particularly in Alzheimer’s disease patients. AI3+ has been Aluminum is one of the most abundant elements in the earth’s crust, and biological systems probably evolved in the presence of appreciable concentrations (1). The unavoidable exposure to aluminum is subject to considerable quantitative fluctuations resulting from variations in geographic, technological, cultural, medical, and other factors (2). Aluminum appears to be closely associated with several adverse neurological andskeletal effects (3). In particular, it has been suggested that elevated levels of aluminum in brain tissues correlate with the occurrence of neurological brain disorders such as Alzheimer’s disease, Down’s Syndrome, and Parkinsonism dementia (4-7). However, controversy exists over whether aluminum has a role in the etiology of Alzheimer’s disease (8-13). Many studiesusing electron probe (14),proton probe ( E ) , and nuclear microscopy (16) have failed to find aluminum in @-amyloidplaques. Stern et al. (17) also failed to find aluminum or silicon in theplaque cores, despite using a technique (laser microprobe mass analysis) capable of parts/ million sensitivity. In contrast, they consistently found aluminum in neurofibrillary tangles (18). While a causal role for aluminum has not been established

* This work was supported by Grant IIRG-90-038 from the Alzheimer’s Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Current address: Dept. of Chemistry, Texas SouthernUniversity, Houston, T X 77004. 5 To whom correspondence should be addressed.

found at 0-0.6 ppm (dry weight) in normal human brain; in brains from Alzheimer’s patients it has been found at 0-1.5 ppm in nondegenerated regions and at 6-12 ppm in the areas of neurofibrillary degeneration (37). The presence ofA13+ in neurons is associated with neurofibrillary tangle formation, and may thus represent a pathogenetic factor for the development of Alzheimer’s disease (13). The neurofibrillary tangles are composed of paired helical filaments whose main protein constituent is tau, a protein that is normally associated with microtubules (38). Tau contains multiple phosphorylation sites (39-41), can be phosphorylated in vitroby various protein kinases (42-45), and its function in vivomay be controlled, at least in part, by phosphorylation and dephosphorylation reactions (46-48). Tau in Alzheimer’s paired helical filaments is hyperphosphorylated at multiple sites and may undergo other modifications that result in the appearance of immunologically distinct forms (48-51). While hyperphosphorylation of tau appears to be a distinguishing criterion of Alzheimer’s disease (48), it is not known if this plays a causal role in the disease etiology. These observations suggest the possibility that aluminum may play some role in neurofibrillary tangle formation by acting on tau. To explore this hypothesis, we began investigating the ability of aluminum to potentiate the phosphorylation of human tau by a variety of protein kinases, In thecourse of these studies, we discovered that aluminum alone, without any kinase, could modify human tau in two distinct ways that mimic modifications that The abbreviations used are: G protein, guanine nucleotide-binding protein; DTT, dithiothreitol; PAGE, polyacrylamide gel electrophoresis.

11976

Aluminum and Human Tau occur in Alzheimer's disease: 1) aluminum inducesaggregation 2) aluminum catalyzes of human tau (and other proteins); and nonenzymatic covalent transfer of the triphosphate group from ATP to tau.

1

2

11977 3

4

5

6

7

EXPERIMENTALPROCEDURES Materials-ATP, histone 1 (type IIIS), @-casein from bovine milk, rabbit muscle enolase, and dithiothreitol were from Sigma. [Y-~'P] ATP, [a-32P]ATP,and [2,8,5'-3H]ATPfrom Du Pont-New England FIG. 1. SDS-PAGE of purified human tau following incuNuclear; [y3'P]GTP and [cY-~'P]GTP from Amersham; [~u-~'P]CTPbation with aluminum. Purified human tau (3 pg) was incubated fromICN,calf intestinal alkalinephosphatase fromBoehringer in 50 pl of 20 mM Hepes (pH 7.4), 1 mM DTT without (lane 1 ) and Mannheim; bacterial alkaline phosphatase from Worthington; and with different concentrations of AlC& (lanes: 2, 50 p M ; 3, 100 mM; 4, aluminum chloride from Allied Chemicals. 300 pM; 5,500 p ~6,; 1 mM) for 30 min, 30 "C, and analyzed by 10% Purification of HumanTau-Escherichiacolicarryingplasmid SDS-PAGE. Lane7, same as lane6 except that EDTA was added to PET-N1234C was grown and induced with isopropyl-1-thio-@-D-ga- a final concentrationof 50mM following the preincubation and before lactopyranoside. Tau waspurifiedfrom Triton X-100 lysates by sample loading. The gel was stained with Coomassie Blue following sequential DE52 (Whatman) and P11 (Whatman) ion-exchangechro- electrophoresis. Molecular massstandards are shown in kDa. matography as described (52). Tau Aggregation Assays-10% SDS-PAGE was performed accordTABLE I ing to Laemmli et al.(53). For filtration assays, sampleswere loaded Effect of preincubation with AiCk onentry of proteins into wells of a Hoefer PR600 Slot Blot apparatus equipped with a into 10%SDS-polyacrylamide gels Nylon-Plus 0.45-pm membrane, and filtered undervacuum with two Experiments like that shown in Fig. 1 were conducted with the washes of 200 pl of dH2O. Protein amounts in the collected filtrates indicatedproteins. The opticaldensities of the CoomassieBluewere assayed by the method of Bradford (54). Phospho-incorporationAssay-Assays were performed in a final stained protein bands were estimated by computer-assisted video to calculatethe volume of 50 pl of 20 mM Hepes (pH 7.4), 1 mM DTT, 50 p M [y3'p] scanning(BioImageVisage110;MilliporeCorp.) ATP (2000-3000 cpm/pmol) containing 3 pg of purified tau or other amounts of protein entering the gel. All values are normalized to the indicated substrates, and AlCl3 at the indicated concentrations (the amount entering following a 30-min incubation with buffer alone. final pH ofthe reaction assaywas 7.0.).Followinga 30-min incubation Values are averages oftwo experiments; standard deviations were at 30 "C, reactionswere stopped by adding 12.5 pl of5 X SDS sample