- Email: [email protected]
Effects on splicing and protein function of three mutations in codon N296 of tau in vitro
Neuroscience Letters 323 (2002) 33–36 www.elsevier.com/locate/neulet
Effects on splicing and protein function of three mutations in codon N296 of tau in vitro Andrew Grover a,b,1, Michael DeTure 1c, Shu-Hui Yen c, Mike Hutton a,* a Laboratory of Neurogenetics, Mayo Clinic Jacksonville, 4500 San Pablo Road, Jacksonville, FL 32224, USA Department of Psychiatry, Division of Neuroscience, School of Medicine, Queen Elizabeth Psychiatric Hospital, University of Birmingham, Birmingham B15 2QZ, UK c Laboratory of Biochemistry, Neuroscience Department, Mayo Clinic Jacksonville, 4500 San Pablo Road, Jacksonville, FL 32224, USA b
Received 28 November 2001; received in revised form 17 January 2002; accepted 17 January 2002
Abstract Three Mutations were recently reported in the same codon (N296) in exon 10 of the tau gene. Two of these mutations, N296N and N296H, lead to a clinical syndrome similar to autosomal dominant fronto-temporal dementia with Parkinsonism linked to chromosome 17. In contrast the third mutation, delN296, gives rise to atypical progressive supranuclear palsy in individuals homozygous for the mutation, but in heterozygous individuals this mutation is incompletely penetrant and associated with a phenotype similar to idiopathic Parkinson’s disease. Functional assays were employed to determine the effects of these mutations on alternative splicing of exon 10, on microtubule assembly and self-aggregation of recombinant tau protein. We demonstrate that these mutations exhibit a spectrum of potentially pathogenic changes in tau function, and provide insight into the possible cause of the incompletely penetrant phenotype of the delN296 mutation. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Tau; Mutation; Splicing; Aggregation; Microtubule; Fronto-temporal dementia
Aggregated tau protein is the major component of the fibrillar lesions that characterize a group of neurodegenerative disorders referred to as tauopathies, which include fronto-temporal dementia with Parkinsonism linked to chromosome 17 (FTDP-17), progressive supranuclear palsy (PSP), Pick’s disease and cortical basal degeneration amongst others [7]. The recent discovery of mutations in the tau gene that cause FTDP-17 has demonstrated that perturbation of tau protein function is sufficient to cause neurodegeneration [6,11,12]. Tau is a soluble microtubuleassociated protein thought to play a role in the assembly and dynamics of microtubules in neurons and glia. Six major isoforms of tau are expressed in the brain, as a result of alternative splicing of exons 2, 3 and 10. Exons 9–12 encode four imperfect-repeat microtubule-binding domains; alternative splicing of exon 10 (E10 2 , E10 1 ) gives rise to isoforms with either three (3R) or four (4R) microtubulebinding domains, in an approximately 1:1 ratio in normal adult human brain [7]. * Corresponding author. Tel.: 11-904-953-0159; fax: 11-904953-7370. E-mail address: [email protected] (M. Hutton). 1 These authors contributed equally to this work.
Missense, deletion and silent mutations have been reported in the coding region of tau, with additional mutations in the intronic region close to the 5 0 splice site of exon 10 [7]. The majority of missense mutations have been shown to impair the ability of the tau protein to bind and polymerize microtubules [5] and to increase the ability of tau to self-aggregate into filamentous structures [3,9]. A second group of exon 10 mutations (S305S, L284L, S305N, N296N, delK280 and intronic mutations proximal to the 5 0 -splice site of exon 10) alter the ratio of tau mRNA with or without exon 10 (E101/E102), thereby changing the ratio of 3R and 4R tau isoforms in the brain of affected patients [7]. All but one of these splicing mutations (delK280) result in an increase in the proportion of 4R Tau [7]. Recently three mutations have been reported in codon Asparagine 296 of the tau gene, namely N296N [13], delN296 [10] and N296H [8]. The N296N and N296H mutations are associated with clinical and pathological phenotypes generally consistent with autosomal-dominant inherited FTDP-17 [1,8,13]. The proband in the delN296 family was homozygous for the deletion, one of two brothers with atypical PSP from the same 3rd degree consangui-
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 2) 00 12 4- 6
34
A. Grover et al. / Neuroscience Letters 323 (2002) 33–36
neous marriage. Two of six heterozygous relatives from the previous generation were diagnosed with 1-dopa responsive Parkinson’s disease (PD), previously unreported in any kindred with tau mutations. D’Souza et al. [2] demonstrated that a region of exon 10 including codon 296 was important for the functioning of an exonic splice suppressor (ESS), that negatively regulates the splicing of exon 10 (Fig. 1). Spillantini et al. [13] demonstrated that N296N causes an increase in splicing presumably through disruption of this ESS element. No splicing analysis has previously been published for the other N296 mutations. Given the variable phenotype of the N296 muta-
Fig. 1. Effects of mutations in Codon 296 on an in vitro splicing assay. (A) A diagram of Exon 10 showing the likely regulatory elements involved in regulation of alternative splicing of this exon. PPE: polypurine positive element; NE: negative element; ESS: exonic splice suppressor; SL/ISM: stem loop/intronic splice modulator. (B) The sequence of the exonic splice suppressor (ESS) showing codon 296 highlighted. The three mutations are shown above and below the sequence. (C) An image of a representative quantitative RT-PCR reaction, showing PCR product for each of the mutants and wild-type, run out on a 2% agarose gel stained with SYBR-green. (D) Quantitation of the RT-PCR product by densitometry of a gel image of duplicate RT-PCR reactions from three separate transfections of wild-type and each mutant, run in a random order. Error bars show one standard deviation.
tions we proposed to examine the splicing behavior of these tau mutants and compare the effects of the delN296 and N296H mutations on the ability of tau protein to assemble microtubules and aggregate into filaments in vitro. To assess the effect of each mutant on tau exon 10 splicing we used the in vitro splicing assay described in Grover et al. [4]. Mutations were generated in a wild-type tau exon 10 construct in the pSPL3b splicing vector by site-directed mutagenesis. Splicing assays were performed as previously described [4], except that the polymerase chain reaction (PCR) protocol for amplification of splicing products was modified. An initial stage of 5 min at 948C was followed by 25 cycles of 948C for 30 s, 608C for 30 s, 728C for 45 s, followed by a final extension at 728C for 10 min. PCR products were visualized on 2% agarose gels stained with SYBR-green. A digital image of the gel was taken and used for quantification of the products using the ImageQuant 5.0 program. For recombinant tau protein work the delN296 and N296H mutations were created by site-directed mutagenesis in pET30a vector incorporating the full-length (2N4R) tau cDNA. Recombinant tau protein was expressed and purified using standard techniques [14]. The ability of tau mutants to promote microtubule assembly in vitro was examined using light-scattering assays. Ice-cold tubulin dimer (25 mM: Cytoskeleton Inc.) was added to tau (2.5 mM) in assembly buffer (100 mM MES, 1 mM GTP and 1 mM DTT at pH 6.8) and immediately transferred to a quartz cuvette, in a spectrophotometer equilibrated to 378C reading absorbance at 350 nm (A350). Assembly assays were performed in quintuplicate. To assess the ability of the coding mutations to alter aggregation polymerization reactions were set up in 30 mM MOPS at pH 7.4 with 0.2 mg/ml tau and 0.02 mg/ml heparin. Samples were incubated at 378C and examined after 1–4 days. Filament formation was quantified from electron micrographs [9] and thioflavin-S binding fluorescence [14]. The results of the splicing assays (Fig. 1) show that N296N produces the largest increase in the ratio of E10 1 /E10 2 with a 4–5-fold increase in the proportion of E10 1 RNA. This is consistent with results reported by Spillantini et al. for this mutation [13]. The N296H mutation generates a 3-fold increase in E101/E102 ratio over wildtype, lower than the N296N mutant, but still highly significant (P , 0:0001, all P values shown are calculated by ttest) and similar to increases previously observed for other pathogenic E10 splicing mutations [4]. In contrast, the delN296 mutation gave E101/E102 splicing ratios similar to the wild-type construct (1.2-fold increase, P , 0:04). The N296H and delN296 mutant tau proteins show a marked (P , 0:0001) decrease in both the rate and extent of tubulin polymerization compared to wild-type tau, with delN296 producing the greatest reductions in polymerization (Fig. 2). The delN296 gave an assembly rate of less than 10% of wild-type, with N296H only 23% of wild-type. Likewise, delN296 had a .70% reduction in the extent of
A. Grover et al. / Neuroscience Letters 323 (2002) 33–36
Fig. 2. Effects of delN296 and N296H mutations on in vitro microtubule assembly reactions. Wild-type recombinant tau protein returned a maximal extent of 0.37 ^ 0.03 absorbance units (au), with a Vmax of 0.154 ^ 0.024 au/minute. delN296 and N296H showed a marked reduction in both maximal extent (0.10 ^ 0.02 and 0.21 ^ 0.03 au, respectively) and rate of polymerization (Vmax of 0.012 ^ 0.004 and 0.041 ^ 0.011 au/minute, respectively).
microtubule assembly, with N296H showing an ,45% reduction in the extent of assembly. Analysis of tau aggregation is shown in Fig. 3 after 2 days incubation. The total filament length/image generated by the wild-type protein averaged 87.8 ^ 31.0 mm taken from five representative grids, compared with 114.5 ^ 19.8 mm for the N296H mutation, and 296.0 ^ 6.6 mm for the delN296 mutation. Thus the N296H displayed a marginal (30%, P , 0:08) increase in filament formation over wild-type tau, whereas the delN296 mutation showed a greater than 3-fold increase (P , 0:001) in the amount of aggregated protein per grid [9]. These results are mirrored by the Thioflavin-S binding assay which shows the same relative order of tau protein aggregation (delN296 q N296H . WT). Morphological analysis showed that the number and length of the filaments are different between mutants and wild-type tau, although the filament diameters were indistinguishable. For wild-type the average number of filaments, counted from five grids, was 58 ^ 20 with an average length of
35
1.53 ^ 1.32 mm; for the N296H mutant 161 ^ 58 filaments per grid gave an average length of 0.71 ^ 0.53 mm, but for the delN296 mutation 805 ^ 130 filaments were counted with an average length of 0.37 ^ 0.20 mm (all values P , 0:001). Based on these results there appears to be increased filament numbers but reduced filament length for the N296H and delN296 mutants compared to wildtype tau. It is possible that this reflects increased nucleation rates for the delN296 and N296H mutant proteins. We have demonstrated that the three mutations in codon N296 exhibit a spectrum of effects on tau. The N296N mutation increases the inclusion of exon 10 in tau mRNA and therefore increases the ratio of 4R/3R tau protein. The delN296 is at the other end of the spectrum, with little or no effect on splicing, but causing a large reduction in the ability of tau to promote microtubule assembly and increasing aggregation of tau into filaments. Finally, the N296H mutation causes increased exon 10 splicing, reduces the ability of tau to promote tubulin polymerization and increases tau aggregation. Indeed this appears to be the first mutation that increases E10 splicing-in and also effects the properties of the tau protein. Heterozygous delN296 individuals display an incompletely-penetrant PD phenotype with a relatively late age of onset (age 60–70) and currently unknown disease duration. In contrast, the homozygous delN296 individual and his sibling develop an atypical PSP phenotype with an early age of onset (38–39) and short duration (3 years). Based on these studies it is unclear why this mutation should produce a relatively mild phenotype in individuals heterozygous for delN296. This may reflect genetic background or be a consequence of the specific nature of the delN296 mutation. Originally thought to alter exon 10 splicing, we have shown that this mutation appears to increase the nucleation of tau aggregation, but results in the formation of relatively short filaments in vitro; perhaps the formation of short filaments has a reduced impact on neuronal function in vivo.
Fig. 3. delN296 and N296H mutations impart a marked increase in the ability of recombinant tau protein to aggregate in vitro. Electron microscope images of: (A) wild-type; (B) N296H; and (C) delN296 recombinant protein aggregation reactions. It is clear from these images that N296H gives more numerous but shorter filaments than wild-type and that delN296 produces many more and much shorter filaments than either of the former. Scale bar ¼ 300 nm. (D) Thioflavin-S fluorescence intensity plot, showing that more tau filament is present in the delN296 (C); followed by the N296H (B); with wild-type (A) the lowest. This is a representative assay, using 10 mM thioflavin-S with lex ¼ 440 nm. ‘2ve’ represents ‘no heparin’ control for each mutant.
36
A. Grover et al. / Neuroscience Letters 323 (2002) 33–36
In contrast to delN296, the N296N and N296H mutations are associated with similar autosomal dominant FTDP-17like phenotypes with later ages of onset, 56 and 57, respectively, in the individuals described. However, clinical details of only one individual from each kindred are available, so any correlation between the relative in vitro effects of these mutations and disease phenotype is uncertain at this time. Codon 296 falls within a region shown by D’Souza et al. to regulate exon 10 alternative splicing in the manner of an ESS element [2]. Altering the sequence of codon 296, as in the N296H and N296N mutations, increases in the ratio of E10 1 /E10 2 mRNA, consistent with disruption of the ESS element. Deletion of the whole codon might also be expected to disrupt the function of this element, but no significant effect on splicing was seen. It is therefore possible that the N296N and N296H mutations alter splicing by creating ‘splice enhancer’ sequences, rather than destroying the putative splice-suppressor element. As demonstrated by Pastor et al., the asparagine at position 296 is highly conserved within microtubule-binding domains of tau and other MAPs [10]. Our data shows that the asparagine residue at position 296 is key to the microtubule polymerizing properties of tau, and that the reduction in microtubule assembly seen in these experiments is due to the disruption of the microtubule-binding domain. Interestingly, mutations that cause the greatest reduction in the ability of tau to polymerize tubulin (delN296 . N296H) also produce the greatest increase in tau self-aggregation (delN296 . N296H). This relative order of effect is consistent with data from previous reports [3,5,9]. This trio of mutations in the same codon of exon 10 represents a unique opportunity to compare and contrast the pathogenic mechanisms of tau mutations demonstrated to cause neurodegeneration. This work shows that a tau mutation can cause pathogenic changes at both RNA and protein levels, confirms the importance of this residue in normal tau function, and reinforces the idea that the relative effect of a mutation in vitro does not necessarily predict disease phenotype. This work was supported by the National Institute on Aging (PO1 grant to S.-H.Y & M.H.). [1] Brown, J., Lantos, P.L., Roques, P., Fidani, L. and Rossor, M.N., Familial dementia with swollen achromatic neurons and corticobasal inclusion bodies: a clinical and pathological study, J. Neurol. Sci., 135 (1996) 21–30. [2] D’Souza, I. and Schellenberg, G.D., Determinants of 4repeat tau expression. Coordination between enhancing and inhibitory splicing sequences for exon 10 inclusion, J. Biol. Chem., 275 (23) (2000) 17700–17709. [3] Goedert, M., Jakes, R. and Crowther, R.A., Effects of frontotemporal dementia FTDP-17 mutations on heparin-induced assembly of tan filaments, FEBS Lett., 450 (1999) 306–311.
[4] Grover, A., Houlden, H., Baker, M., Adamson, J., Lewis, J., Prihar, G., Pickering-Brown, S., Duff, K. and Hutton, M., 5 0 splice site mutations in tau associated with the inherited dementia FTDP-17 affect a stem-loop structure that regulates alternative splicing of exon 10, J. Biol. Chem., 274 (21) (1994) 15134–15143. [5] Hong, M., Thukareva, V., Vogelsburg-Ragaglia, V., Wszolek, Z., Reed, L., Miller, B.I., Geschwind, D.H., Bird, T.D., McKeel, D., Goate, A., Morris, J.C., Wilhelmsen, K.C., Schellenberg, G.D., Trojanowski, J.Q. and Lee, V.M.-Y., Mutation-specific functional impairments in distinct tan isoforms of hereditary FIDP-17, Science, 282 (1998) 1914–1917. [6] Hutton, M., Lendon, C.L., Rizzu, P., Baker, M., Froelich, S., Houlden, H., Pickering-Brown, S., Chakraverty, S., Isaacs, A., Grover, A., Hackett, J., Adamson, J., Lincoln, S., Dickson, D., Davies, P., Petersen, R.C., Stevens, M., de Graaff, E., Wauters, E., van Baren, J., Hillebrand, M., Joosse, M., Kwon, J.M., Nowotny, P., Che, L.K., Norton, J., Morris, J.C., Reed, L.A., Trojanowski, J., Basun, H., Lannfelt, L., Neystat, M., Fahn, S., Dark, F., Tannenberg, T., Dodd, P.R., Hayward, N., Kwok, J.B.J., Schofield, P.R., Andreadis, A., Snowden, J., Craufurd, D., Neary, D., Owen, F., Oostra, B.A., Hardy, J., Goate, A., van Swieten, J., Mann, D., Lynch, T. and Heutink, P., Association of missense and 5 0 splice-site mutations in tau with the inherited dementia FTDP-17, Nature, 393 (1998) 70–75. [7] Hutton, M., Missense and splice site mutations in tau associated with FTDP-17: multiple pathogenic mechanisms, Neurology, 56 (2001) S21–S25. [8] Iseki, E., Matsumura, T., Marui, W., Hino, H., Odawara, T., Sugiyama, N., Sawada, H., Arai, T. and Kosaka, K., Familial fronto-temporal dementia and Parkinsonism with a novel N296H mutation in exon 10 of the tau gene and widespread tau accumulation in the glial cells, Acta Neuropathol., 102 (2001) 285–292. [9] Nacharaju, P., Lewis, J., Easson, C., Yen, S., Hackett, J., Hutton, M. and Yen, S.-H., Accelerated filament formation from tau protein with specific FTDP-17 missense mutations, FEBS Lett., 447 (1998) 195–199. [10] Pastor, P., Pastor, E., Carnero, C., Vela, R., Garcia, T., Amer, G., Tolosa, E. and Oliva, R., Familial atypical progressive supranuclear palsy associated with homozygosity for the delN296 mutation in the tau gene, Ann. Neurol., 49 (2001) 263–267. [11] Poorkaj, P., Bird, T.D., Wijsman, E., Nemens, E., Garruto, R.M., Anderson, L., Andreadis, A., Wiederholt, W.C., Raskind, M. and Schellenberg, G.D., Tau is a candidate gene for chromosome 17 fronto-temporal dementia, Ann. Neurol., 43 (6) (1998) 815–825. [12] Spillantini, M.G., Murrell, J.R., Goedert, M., Farlow, M.R., Klug, A. and Ghetti, B., Mutation in the tau gene in familial multiple system tauopathy with presenile dementia, Proc. Natl. Acad. Sci. USA, 95 (1998) 7737–7741. [13] Spillantini, M.G., Yoshida, H., Rizzini, C., Lantos, P.L., Khan, N., Rosser, M.N., Goedert, M. and Brown, J., A novel tau mutation (N296N) in familial dementia with swollen achromatic neurons and corticobasal inclusion bodies, Ann. Neurol., 48 (2000) 939–943. [14] Zhang, E.Y., DeTure, M.A., Bubb, M.R., Caviston, T.L., Erdos, G.W., Whittaker, S.D. and Punch, D.L., Self-assembly of the brain MAP-2 microtubule-binding region into polymeric structures resembling Alzheimer filaments, Biochem. Biophys. Res. Commun., 229 (1996) 176–181.
Effects on splicing and protein function of three mutations in codon N296 of tau in vitro Andrew Grover a,b,1, Michael DeTure 1c, Shu-Hui Yen c, Mike Hutton a,* a Laboratory of Neurogenetics, Mayo Clinic Jacksonville, 4500 San Pablo Road, Jacksonville, FL 32224, USA Department of Psychiatry, Division of Neuroscience, School of Medicine, Queen Elizabeth Psychiatric Hospital, University of Birmingham, Birmingham B15 2QZ, UK c Laboratory of Biochemistry, Neuroscience Department, Mayo Clinic Jacksonville, 4500 San Pablo Road, Jacksonville, FL 32224, USA b
Received 28 November 2001; received in revised form 17 January 2002; accepted 17 January 2002
Abstract Three Mutations were recently reported in the same codon (N296) in exon 10 of the tau gene. Two of these mutations, N296N and N296H, lead to a clinical syndrome similar to autosomal dominant fronto-temporal dementia with Parkinsonism linked to chromosome 17. In contrast the third mutation, delN296, gives rise to atypical progressive supranuclear palsy in individuals homozygous for the mutation, but in heterozygous individuals this mutation is incompletely penetrant and associated with a phenotype similar to idiopathic Parkinson’s disease. Functional assays were employed to determine the effects of these mutations on alternative splicing of exon 10, on microtubule assembly and self-aggregation of recombinant tau protein. We demonstrate that these mutations exhibit a spectrum of potentially pathogenic changes in tau function, and provide insight into the possible cause of the incompletely penetrant phenotype of the delN296 mutation. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Tau; Mutation; Splicing; Aggregation; Microtubule; Fronto-temporal dementia
Aggregated tau protein is the major component of the fibrillar lesions that characterize a group of neurodegenerative disorders referred to as tauopathies, which include fronto-temporal dementia with Parkinsonism linked to chromosome 17 (FTDP-17), progressive supranuclear palsy (PSP), Pick’s disease and cortical basal degeneration amongst others [7]. The recent discovery of mutations in the tau gene that cause FTDP-17 has demonstrated that perturbation of tau protein function is sufficient to cause neurodegeneration [6,11,12]. Tau is a soluble microtubuleassociated protein thought to play a role in the assembly and dynamics of microtubules in neurons and glia. Six major isoforms of tau are expressed in the brain, as a result of alternative splicing of exons 2, 3 and 10. Exons 9–12 encode four imperfect-repeat microtubule-binding domains; alternative splicing of exon 10 (E10 2 , E10 1 ) gives rise to isoforms with either three (3R) or four (4R) microtubulebinding domains, in an approximately 1:1 ratio in normal adult human brain [7]. * Corresponding author. Tel.: 11-904-953-0159; fax: 11-904953-7370. E-mail address: [email protected] (M. Hutton). 1 These authors contributed equally to this work.
Missense, deletion and silent mutations have been reported in the coding region of tau, with additional mutations in the intronic region close to the 5 0 splice site of exon 10 [7]. The majority of missense mutations have been shown to impair the ability of the tau protein to bind and polymerize microtubules [5] and to increase the ability of tau to self-aggregate into filamentous structures [3,9]. A second group of exon 10 mutations (S305S, L284L, S305N, N296N, delK280 and intronic mutations proximal to the 5 0 -splice site of exon 10) alter the ratio of tau mRNA with or without exon 10 (E101/E102), thereby changing the ratio of 3R and 4R tau isoforms in the brain of affected patients [7]. All but one of these splicing mutations (delK280) result in an increase in the proportion of 4R Tau [7]. Recently three mutations have been reported in codon Asparagine 296 of the tau gene, namely N296N [13], delN296 [10] and N296H [8]. The N296N and N296H mutations are associated with clinical and pathological phenotypes generally consistent with autosomal-dominant inherited FTDP-17 [1,8,13]. The proband in the delN296 family was homozygous for the deletion, one of two brothers with atypical PSP from the same 3rd degree consangui-
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 2) 00 12 4- 6
34
A. Grover et al. / Neuroscience Letters 323 (2002) 33–36
neous marriage. Two of six heterozygous relatives from the previous generation were diagnosed with 1-dopa responsive Parkinson’s disease (PD), previously unreported in any kindred with tau mutations. D’Souza et al. [2] demonstrated that a region of exon 10 including codon 296 was important for the functioning of an exonic splice suppressor (ESS), that negatively regulates the splicing of exon 10 (Fig. 1). Spillantini et al. [13] demonstrated that N296N causes an increase in splicing presumably through disruption of this ESS element. No splicing analysis has previously been published for the other N296 mutations. Given the variable phenotype of the N296 muta-
Fig. 1. Effects of mutations in Codon 296 on an in vitro splicing assay. (A) A diagram of Exon 10 showing the likely regulatory elements involved in regulation of alternative splicing of this exon. PPE: polypurine positive element; NE: negative element; ESS: exonic splice suppressor; SL/ISM: stem loop/intronic splice modulator. (B) The sequence of the exonic splice suppressor (ESS) showing codon 296 highlighted. The three mutations are shown above and below the sequence. (C) An image of a representative quantitative RT-PCR reaction, showing PCR product for each of the mutants and wild-type, run out on a 2% agarose gel stained with SYBR-green. (D) Quantitation of the RT-PCR product by densitometry of a gel image of duplicate RT-PCR reactions from three separate transfections of wild-type and each mutant, run in a random order. Error bars show one standard deviation.
tions we proposed to examine the splicing behavior of these tau mutants and compare the effects of the delN296 and N296H mutations on the ability of tau protein to assemble microtubules and aggregate into filaments in vitro. To assess the effect of each mutant on tau exon 10 splicing we used the in vitro splicing assay described in Grover et al. [4]. Mutations were generated in a wild-type tau exon 10 construct in the pSPL3b splicing vector by site-directed mutagenesis. Splicing assays were performed as previously described [4], except that the polymerase chain reaction (PCR) protocol for amplification of splicing products was modified. An initial stage of 5 min at 948C was followed by 25 cycles of 948C for 30 s, 608C for 30 s, 728C for 45 s, followed by a final extension at 728C for 10 min. PCR products were visualized on 2% agarose gels stained with SYBR-green. A digital image of the gel was taken and used for quantification of the products using the ImageQuant 5.0 program. For recombinant tau protein work the delN296 and N296H mutations were created by site-directed mutagenesis in pET30a vector incorporating the full-length (2N4R) tau cDNA. Recombinant tau protein was expressed and purified using standard techniques [14]. The ability of tau mutants to promote microtubule assembly in vitro was examined using light-scattering assays. Ice-cold tubulin dimer (25 mM: Cytoskeleton Inc.) was added to tau (2.5 mM) in assembly buffer (100 mM MES, 1 mM GTP and 1 mM DTT at pH 6.8) and immediately transferred to a quartz cuvette, in a spectrophotometer equilibrated to 378C reading absorbance at 350 nm (A350). Assembly assays were performed in quintuplicate. To assess the ability of the coding mutations to alter aggregation polymerization reactions were set up in 30 mM MOPS at pH 7.4 with 0.2 mg/ml tau and 0.02 mg/ml heparin. Samples were incubated at 378C and examined after 1–4 days. Filament formation was quantified from electron micrographs [9] and thioflavin-S binding fluorescence [14]. The results of the splicing assays (Fig. 1) show that N296N produces the largest increase in the ratio of E10 1 /E10 2 with a 4–5-fold increase in the proportion of E10 1 RNA. This is consistent with results reported by Spillantini et al. for this mutation [13]. The N296H mutation generates a 3-fold increase in E101/E102 ratio over wildtype, lower than the N296N mutant, but still highly significant (P , 0:0001, all P values shown are calculated by ttest) and similar to increases previously observed for other pathogenic E10 splicing mutations [4]. In contrast, the delN296 mutation gave E101/E102 splicing ratios similar to the wild-type construct (1.2-fold increase, P , 0:04). The N296H and delN296 mutant tau proteins show a marked (P , 0:0001) decrease in both the rate and extent of tubulin polymerization compared to wild-type tau, with delN296 producing the greatest reductions in polymerization (Fig. 2). The delN296 gave an assembly rate of less than 10% of wild-type, with N296H only 23% of wild-type. Likewise, delN296 had a .70% reduction in the extent of
A. Grover et al. / Neuroscience Letters 323 (2002) 33–36
Fig. 2. Effects of delN296 and N296H mutations on in vitro microtubule assembly reactions. Wild-type recombinant tau protein returned a maximal extent of 0.37 ^ 0.03 absorbance units (au), with a Vmax of 0.154 ^ 0.024 au/minute. delN296 and N296H showed a marked reduction in both maximal extent (0.10 ^ 0.02 and 0.21 ^ 0.03 au, respectively) and rate of polymerization (Vmax of 0.012 ^ 0.004 and 0.041 ^ 0.011 au/minute, respectively).
microtubule assembly, with N296H showing an ,45% reduction in the extent of assembly. Analysis of tau aggregation is shown in Fig. 3 after 2 days incubation. The total filament length/image generated by the wild-type protein averaged 87.8 ^ 31.0 mm taken from five representative grids, compared with 114.5 ^ 19.8 mm for the N296H mutation, and 296.0 ^ 6.6 mm for the delN296 mutation. Thus the N296H displayed a marginal (30%, P , 0:08) increase in filament formation over wild-type tau, whereas the delN296 mutation showed a greater than 3-fold increase (P , 0:001) in the amount of aggregated protein per grid [9]. These results are mirrored by the Thioflavin-S binding assay which shows the same relative order of tau protein aggregation (delN296 q N296H . WT). Morphological analysis showed that the number and length of the filaments are different between mutants and wild-type tau, although the filament diameters were indistinguishable. For wild-type the average number of filaments, counted from five grids, was 58 ^ 20 with an average length of
35
1.53 ^ 1.32 mm; for the N296H mutant 161 ^ 58 filaments per grid gave an average length of 0.71 ^ 0.53 mm, but for the delN296 mutation 805 ^ 130 filaments were counted with an average length of 0.37 ^ 0.20 mm (all values P , 0:001). Based on these results there appears to be increased filament numbers but reduced filament length for the N296H and delN296 mutants compared to wildtype tau. It is possible that this reflects increased nucleation rates for the delN296 and N296H mutant proteins. We have demonstrated that the three mutations in codon N296 exhibit a spectrum of effects on tau. The N296N mutation increases the inclusion of exon 10 in tau mRNA and therefore increases the ratio of 4R/3R tau protein. The delN296 is at the other end of the spectrum, with little or no effect on splicing, but causing a large reduction in the ability of tau to promote microtubule assembly and increasing aggregation of tau into filaments. Finally, the N296H mutation causes increased exon 10 splicing, reduces the ability of tau to promote tubulin polymerization and increases tau aggregation. Indeed this appears to be the first mutation that increases E10 splicing-in and also effects the properties of the tau protein. Heterozygous delN296 individuals display an incompletely-penetrant PD phenotype with a relatively late age of onset (age 60–70) and currently unknown disease duration. In contrast, the homozygous delN296 individual and his sibling develop an atypical PSP phenotype with an early age of onset (38–39) and short duration (3 years). Based on these studies it is unclear why this mutation should produce a relatively mild phenotype in individuals heterozygous for delN296. This may reflect genetic background or be a consequence of the specific nature of the delN296 mutation. Originally thought to alter exon 10 splicing, we have shown that this mutation appears to increase the nucleation of tau aggregation, but results in the formation of relatively short filaments in vitro; perhaps the formation of short filaments has a reduced impact on neuronal function in vivo.
Fig. 3. delN296 and N296H mutations impart a marked increase in the ability of recombinant tau protein to aggregate in vitro. Electron microscope images of: (A) wild-type; (B) N296H; and (C) delN296 recombinant protein aggregation reactions. It is clear from these images that N296H gives more numerous but shorter filaments than wild-type and that delN296 produces many more and much shorter filaments than either of the former. Scale bar ¼ 300 nm. (D) Thioflavin-S fluorescence intensity plot, showing that more tau filament is present in the delN296 (C); followed by the N296H (B); with wild-type (A) the lowest. This is a representative assay, using 10 mM thioflavin-S with lex ¼ 440 nm. ‘2ve’ represents ‘no heparin’ control for each mutant.
36
A. Grover et al. / Neuroscience Letters 323 (2002) 33–36
In contrast to delN296, the N296N and N296H mutations are associated with similar autosomal dominant FTDP-17like phenotypes with later ages of onset, 56 and 57, respectively, in the individuals described. However, clinical details of only one individual from each kindred are available, so any correlation between the relative in vitro effects of these mutations and disease phenotype is uncertain at this time. Codon 296 falls within a region shown by D’Souza et al. to regulate exon 10 alternative splicing in the manner of an ESS element [2]. Altering the sequence of codon 296, as in the N296H and N296N mutations, increases in the ratio of E10 1 /E10 2 mRNA, consistent with disruption of the ESS element. Deletion of the whole codon might also be expected to disrupt the function of this element, but no significant effect on splicing was seen. It is therefore possible that the N296N and N296H mutations alter splicing by creating ‘splice enhancer’ sequences, rather than destroying the putative splice-suppressor element. As demonstrated by Pastor et al., the asparagine at position 296 is highly conserved within microtubule-binding domains of tau and other MAPs [10]. Our data shows that the asparagine residue at position 296 is key to the microtubule polymerizing properties of tau, and that the reduction in microtubule assembly seen in these experiments is due to the disruption of the microtubule-binding domain. Interestingly, mutations that cause the greatest reduction in the ability of tau to polymerize tubulin (delN296 . N296H) also produce the greatest increase in tau self-aggregation (delN296 . N296H). This relative order of effect is consistent with data from previous reports [3,5,9]. This trio of mutations in the same codon of exon 10 represents a unique opportunity to compare and contrast the pathogenic mechanisms of tau mutations demonstrated to cause neurodegeneration. This work shows that a tau mutation can cause pathogenic changes at both RNA and protein levels, confirms the importance of this residue in normal tau function, and reinforces the idea that the relative effect of a mutation in vitro does not necessarily predict disease phenotype. This work was supported by the National Institute on Aging (PO1 grant to S.-H.Y & M.H.). [1] Brown, J., Lantos, P.L., Roques, P., Fidani, L. and Rossor, M.N., Familial dementia with swollen achromatic neurons and corticobasal inclusion bodies: a clinical and pathological study, J. Neurol. Sci., 135 (1996) 21–30. [2] D’Souza, I. and Schellenberg, G.D., Determinants of 4repeat tau expression. Coordination between enhancing and inhibitory splicing sequences for exon 10 inclusion, J. Biol. Chem., 275 (23) (2000) 17700–17709. [3] Goedert, M., Jakes, R. and Crowther, R.A., Effects of frontotemporal dementia FTDP-17 mutations on heparin-induced assembly of tan filaments, FEBS Lett., 450 (1999) 306–311.
[4] Grover, A., Houlden, H., Baker, M., Adamson, J., Lewis, J., Prihar, G., Pickering-Brown, S., Duff, K. and Hutton, M., 5 0 splice site mutations in tau associated with the inherited dementia FTDP-17 affect a stem-loop structure that regulates alternative splicing of exon 10, J. Biol. Chem., 274 (21) (1994) 15134–15143. [5] Hong, M., Thukareva, V., Vogelsburg-Ragaglia, V., Wszolek, Z., Reed, L., Miller, B.I., Geschwind, D.H., Bird, T.D., McKeel, D., Goate, A., Morris, J.C., Wilhelmsen, K.C., Schellenberg, G.D., Trojanowski, J.Q. and Lee, V.M.-Y., Mutation-specific functional impairments in distinct tan isoforms of hereditary FIDP-17, Science, 282 (1998) 1914–1917. [6] Hutton, M., Lendon, C.L., Rizzu, P., Baker, M., Froelich, S., Houlden, H., Pickering-Brown, S., Chakraverty, S., Isaacs, A., Grover, A., Hackett, J., Adamson, J., Lincoln, S., Dickson, D., Davies, P., Petersen, R.C., Stevens, M., de Graaff, E., Wauters, E., van Baren, J., Hillebrand, M., Joosse, M., Kwon, J.M., Nowotny, P., Che, L.K., Norton, J., Morris, J.C., Reed, L.A., Trojanowski, J., Basun, H., Lannfelt, L., Neystat, M., Fahn, S., Dark, F., Tannenberg, T., Dodd, P.R., Hayward, N., Kwok, J.B.J., Schofield, P.R., Andreadis, A., Snowden, J., Craufurd, D., Neary, D., Owen, F., Oostra, B.A., Hardy, J., Goate, A., van Swieten, J., Mann, D., Lynch, T. and Heutink, P., Association of missense and 5 0 splice-site mutations in tau with the inherited dementia FTDP-17, Nature, 393 (1998) 70–75. [7] Hutton, M., Missense and splice site mutations in tau associated with FTDP-17: multiple pathogenic mechanisms, Neurology, 56 (2001) S21–S25. [8] Iseki, E., Matsumura, T., Marui, W., Hino, H., Odawara, T., Sugiyama, N., Sawada, H., Arai, T. and Kosaka, K., Familial fronto-temporal dementia and Parkinsonism with a novel N296H mutation in exon 10 of the tau gene and widespread tau accumulation in the glial cells, Acta Neuropathol., 102 (2001) 285–292. [9] Nacharaju, P., Lewis, J., Easson, C., Yen, S., Hackett, J., Hutton, M. and Yen, S.-H., Accelerated filament formation from tau protein with specific FTDP-17 missense mutations, FEBS Lett., 447 (1998) 195–199. [10] Pastor, P., Pastor, E., Carnero, C., Vela, R., Garcia, T., Amer, G., Tolosa, E. and Oliva, R., Familial atypical progressive supranuclear palsy associated with homozygosity for the delN296 mutation in the tau gene, Ann. Neurol., 49 (2001) 263–267. [11] Poorkaj, P., Bird, T.D., Wijsman, E., Nemens, E., Garruto, R.M., Anderson, L., Andreadis, A., Wiederholt, W.C., Raskind, M. and Schellenberg, G.D., Tau is a candidate gene for chromosome 17 fronto-temporal dementia, Ann. Neurol., 43 (6) (1998) 815–825. [12] Spillantini, M.G., Murrell, J.R., Goedert, M., Farlow, M.R., Klug, A. and Ghetti, B., Mutation in the tau gene in familial multiple system tauopathy with presenile dementia, Proc. Natl. Acad. Sci. USA, 95 (1998) 7737–7741. [13] Spillantini, M.G., Yoshida, H., Rizzini, C., Lantos, P.L., Khan, N., Rosser, M.N., Goedert, M. and Brown, J., A novel tau mutation (N296N) in familial dementia with swollen achromatic neurons and corticobasal inclusion bodies, Ann. Neurol., 48 (2000) 939–943. [14] Zhang, E.Y., DeTure, M.A., Bubb, M.R., Caviston, T.L., Erdos, G.W., Whittaker, S.D. and Punch, D.L., Self-assembly of the brain MAP-2 microtubule-binding region into polymeric structures resembling Alzheimer filaments, Biochem. Biophys. Res. Commun., 229 (1996) 176–181.