Association Between Tau H2 Haplotype and Age at Onset in Frontotemporal Dementia

ORIGINAL CONTRIBUTION Association Between Tau H2 Haplotype and Age at Onset in Frontotemporal Dementia

Barbara Borroni, MD; Despina Yancopoulou, PhD; Miho Tsutsui, MD, PhD; Alessandro Padovani, MD, PhD; Stephen J. Sawcer, MD; John R. Hodges, MD; Maria Grazia Spillantini, PhD

Background: The frontotemporal dementia (FTD) syn- Results: No difference in the distribution of the tau dromes have been associated with the microtubule- H1 and H2 haplotypes between FTD cases and con- associated tau protein since tau gene mutations have been trols was observed, whereas the ApoE ε4 allele was demonstrated to be the cause of FTD and parkinsonism more frequent in FTD cases. The presence of at least 1 linked to chromosome 17. In cases of FTD without tau tau H2 allele was found to be significantly associated gene mutations, however, it is unclear whether genetic with an earlier age of onset in patients with FTD. The variability in the tau gene is associated with the devel- association between the H2 allele and age at onset was opment or modulation of FTD. not related to family history, clinical presentation, or Objective: To determine whether genetic variability in ApoE genotype. tau and apolipoprotein E (ApoE) modulates and contributes to the development of FTD. Conclusion: These findings support a role of tau pro- Design and Patients: The distribution of tau gene tein in modulating disease phenotype by influencing the haplotypes and the ApoE genotype were investigated age at onset in these FTD cases. in 86 patients with well-characterized FTD and 50 control subjects. Arch Neurol. 2005;62:1419-1422

HE TAU GENE HAS BEEN IM- patients.8 In FTD, however, no gene has plicated in the develop- so far been definitely associated with dis- ment of frontotemporal de- ease modulation, and the role of ApoE ε4 mentia (FTD) syndromes, remains controversial.9-12 as several pathogenic mu- To determine whether genetic vari- tationsT have been found in FTD with par- ability in tau and ApoE modulates kinsonism linked to chromosome 17.1 and contributes to the heterogeneity Most familial FTD cases have been observed in FTD, we assessed the tau proved to be clinically, neuropathologi- and ApoE genotypes in a cohort of spo- cally, and genetically heterogeneous and radic and familial cases of FTD without are not associated with tau gene muta- tau gene mutations. We found a similar tions, suggesting that additional genetic, distribution of the H1 and H2 haplotypes epigenetic, and environmental factors may between FTD cases and control subjects. be implicated in the pathogenesis or modu- However, our results show that the pres- lation of the disease. ence of the tau H2 allele leads to a sig- Indeed, recent studies have investi- nificantly earlier age of onset in FTD. To gated the effect of genetic variability in tau our knowledge, this is the first report of Author Affiliations: Brain an influence of the tau H2 haplotype on and apolipoprotein E (ApoE) on the de- Repair Centre (Drs Borroni, FTD phenotype. Yancopoulou, Tsutsui, and velopment of FTD. Although still contro- Spillantini) and Neurology Unit versial,2-5 an association between FTD and (Drs Sawcer and Hodges), the tau H1 haplotype alone or in combi- METHODS Department of Clinical nation with ApoE ε4 and ε2 alleles6,7 has Neurosciences, University of been suggested. Cambridge, Cambridge, SUBJECTS England; and Department of Apolipoprotein E modulates disease Medical Sciences, Neurological phenotype in Alzheimer disease, in that the Tau exon sequences and ApoE genotype were Clinic, University of Brescia, presence of the ApoE ε4 allele signifi- determined in 86 well-characterized white pa- Brescia, Italy (Dr Padovani). cantly decreases the age of onset in these tients with FTD (hereafter referred to as FTD

(REPRINTED) ARCH NEUROL / VOL 62, SEP 2005 WWW.ARCHNEUROL.COM 1419

©2005 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 Table. Demographic and Clinical Characteristics of Patients With FTD Grouped According to Clinical Presentation*

Subtypes of FTD Total No. of FTD Cases fvFTD SD PNFA Characteristics (N = 86) (n = 51) (n = 27) (n=8) Sex, No. M/F 43/43 18/33 20/7 5/3 Family history, No. (%) 18 (20.9) 14 (27.4) 2 (7.4) 2 (25.0) Frequency of *H1 vs *H2, No. (%) 47 (54.7) vs 39 (45.3) 25 (49.0) vs 26 (51.0) 16 (59.3) vs 11 (40.7) 6 (75.0) vs 2 (25.0) ApoE ε4 genotype, No. (%) 32 (37.2) 18 (35.3) 10 (37.0) 4 (50.0) Age at onset, y 57.2 ± 7.7 (86) 55.5 ± 7.8 (51) 59.0 ± 6.9 (27) 58.3 ± 8.4 (8) Age at onset with *H1, y† 59.2 ± 6.0 (47) 57.7 ± 6.0 (25) 60.4 ± 5.3 (16) 61.3 ± 7.0 (6) Age at onset with *H2, y 54.1 ± 8.7 (39) 53.2 ± 8.9 (26) 56.9 ± 8.6 (11) 49.5 ± 6.3 (2)

Abbreviations: ApoE, apoliprotein E; FTD, frontotemporal dementia; fvFTD, frontal variant FTD; *H1, absence of the H2 allele (H1/H1); *H2, presence of at least 1 H2 allele (H1/H2 or H2/H2 ); PNFA, progressive nonfluent aphasia; SD, semantic dementia. *Unless otherwise indicated, data are expressed as mean ± SD (number of cases). †Significant between *H1 and *H2 carriers (PϽ.003); otherwise no significant differences were seen between FTD, SD, and PNFA subtypes in the demographic and clinical characteristics.

patients or FTD cases) (Table). All patients were from East action was performed as previously described.16 The following Anglia, an area that covers about 100 miles around Cam- polymorphisms were used to determine tau H1 and H2 hap- bridge, England, with more than 2 million inhabitants. The FTD lotypes in patients and controls: proline 176 in exon 7 and ala- patients underwent assessment in the Early Onset Dementia nine 227 and asparagine 255 in exon 9. Clinic at Addenbrookes Hospital, Cambridge, from January 1, 1998, through December 31, 2004. Clinical evaluation of FTD ApoE GENOTYPING consisted of neurological and neuropsychological testing, which included an interview of family members or caregivers. Struc- Genetic variation at the ApoE locus was determined by restric- tural neuroimaging studies using magnetic resonance imag- tion isotyping using polymerase chain reaction amplification ing were also performed in all cases. All patients fulfilled in- and subsequent digestion with restriction enzyme HhaI(New ternational consensus criteria for frontotemporal lobar England Biolabs, Ipswich, Mass). degeneration.13 In keeping with other studies from the Cam- bridge group (that included one of us [J.R.H.]), we retained the term frontotemporal dementia as a general superordinate la- STATISTICAL ANALYSIS bel with a subsequent subdivision into the following 3 major clinical subtypes: behavioral or frontal variant FTD (fv FTD), Genotype and allele frequencies between FTD patients and con- semantic dementia (SD), and progressive nonfluent aphasia trols were compared by means of the Pearson ␹2 test. We com- (PNFA).14,15 pared clinical subgroups using the Pearson ␹2 test, unpaired t Patients who were considered to have a positive family his- test, and analysis of variance (ANOVA). We used the Bonfer- tory were those who had a first-degree relative with dementia, roni test for post hoc analysis to avoid type II error.17 We also parkinsonism, or motor neuron disease (amyotrophic lateral performed analysis of covariance (ANCOVA) to account for pos- sclerosis). To date, 18 of the 86 FTD patients have died. Au- sible confounds. Statistical significance was assumed at PϽ.05. topsy was performed in 13 of these patients, and the results re- vealed the characteristic macroscopic and microscopic neuro- RESULTS pathological features of FTD. For comparison, we also tested the tau and ApoE genes in 50 age- and sex-matched white controls who had no record of In total, 86 FTD patients underwent screening. A posi- neurological or psychiatric symptoms. Controls were re- tive family history was recorded in 18 cases, but no mu- cruited from the same geographical area as the patients as part tations were identified in the tau gene in any of these cases. of large population studies on FTD, Alzheimer disease, and mul- Genotype frequencies were compared between cases tiple sclerosis (performed by J.R.H. and S.J.S.). Patients and con- and controls, but no statistically significant difference was trols had a similar range of education. The study was con- observed, with the following results: H1/H1, 54.7% (n=47) ducted in accordance with local ethical regulations, and an vs 48% (n=24); H1/H2, 40.7% (n=35) vs 48% (n=24); informed consent was obtained from all subjects and care- and H2/H2, 4.6% (n=4) vs 4% (n=2) (␹2, P=.78). The givers. same was true for the allele frequencies, with both groups showing a similar distribution of H1 and H2 alleles and DNA EXTRACTION H1 being the most common allele (␹2 test, P=.82). The AND TAU SEQUENCE ANALYSIS genotype and allele frequencies were similar to those observed in previous studies.4 Genomic DNA was extracted from blood samples of patients In this study, *H2 denotes the presence of at least 1 using a commercially available DNA extraction kit (Qiagen Inc, Valencia, Calif). All 13 tau exons and exon-intron boundaries H2 allele, ie, H1/H2 or H2/H2; *H1, the absence of the ␹2 were sequenced in the familial FTD patients. Tau exons 7 and H2 allele, ie, H1/H1. Results of testing showed that no 9 through 13 and exon-intron junctions were analyzed in spo- demographic characteristics, ie, sex or family history of radic FTD cases. Tau exons 7 and 9 were sequenced in con- dementia, were associated with *H1 or *H2 (␹2 test, P=.75 trols. DNA amplification with the use of polymerase chain re- and P=.53, respectively). However, the presence of *H2

(REPRINTED) ARCH NEUROL / VOL 62, SEP 2005 WWW.ARCHNEUROL.COM 1420

©2005 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 (n=39 carriers) was found to be significantly associated FTD, and the remaining 5 cases have not yet undergone with earlier age at onset compared with *H1 (n=47 car- analysis. Although some of the cases that have not un- riers) (mean±SD age for *H1, 59.2±6.0 years; for *H2, dergone autopsy might not have pathological FTD, de- 54.1±8.7 years; t test, P=.003). Further analysis using spite patients’ clinical record, the fact that FTD was patho- ANOVA and Bonferroni post hoc tests to compare the 3 logically confirmed so far in all of the cases that underwent groups (H1/H1, H1/H2, and H2/H2) confirmed the effect autopsy, together with the imaging study results and the of the H2 allele on decreasing the age at onset and re- expertise of the neurology clinic on differential diag- vealed that the 4 homozygous H2/H2 cases showed a fur- noses of FTD, makes it reasonable to assume that these ther decrease of age at onset compared with H1/H1 car- cases represent a largely homogeneous FTD group. riers (mean±SD age in H2/H2 carriers, 44.5±10.8 years; An interesting question that remains to be answered ANOVA and Bonferroni post hoc analysis, PϽ.001). concerns the relationship between the presence of tau H1 Among *H2 carriers, 11 (28.2%) of 39 individuals showed and H2 haplotypes and disease duration in these FTD pa- an age at onset younger than 50 years compared with only tients. 3 (6.3%) of 47 of *H1 carriers. Although we have the age at presentation/onset for all The following numbers of FTD patients were in the 3 of the patients, which corresponds to the date on which clinical subtypes: 51 in fvFTD patients, 27 in SD pa- they were first referred to and seen by the neurologist or tients, and 8 in PNFA patients.14,15 Demographic and clini- the neuropsychologist, only 18 have died so far and the cal characteristics of the subgroups are shown in the Table. disease duration (2-14 years) in these 18 is similar be- We performed ANCOVA to evaluate the effect of the H2 tween patients carrying tau H1 and H2 haplotypes. The allele when adjusting for subgroup diagnosis, and the same remaining 68 patients who are still alive have a disease trend for the anticipation of age of onset in *H2 carriers duration from presentation that ranges from 1 to 12 years. was found (ANCOVA, PϽ.01). In fact, in any of the 3 At present, we cannot conclude how long they will have clinical subtypes (fvFTD, SD, or PNFA), *H2 carriers the disease. showed an earlier age of onset compared with *H1 car- No significant association between FTD and the tau H1 riers (ANOVA, P=.02). Mean values are given in the Table. or H2 haplotypes was detected, a finding in agreement with A significant difference in the distribution of ApoE ε4 those of previous reports.3,4 Furthermore, our results show isoforms was seen between patients and controls. The an increased frequency of the ApoE ε4 allele in all FTD ApoE ε4 genotype was significantly more frequent in pa- subgroups, as also reported by Gustafson et al.9 However, tients compared with controls (38.6% vs 20.0%; ␹2 test like other results,18 ours do not confirm previous reports P=.03). The 3 FTD subtypes did not differ in ApoE iso- of an association of the ApoE ε4 allele with earlier age of form frequency (␹2 test, P=.78). The ApoE genotype was onset in FTD10,11 and do not support a synergistic effect not associated with age of onset (t test, mean±SD age for between tau and the ApoE ε4 allele.6,19 non–ApoE ε4 carriers vs ApoE ε4 carriers, 56.8±6.4 vs We performed our study in a group of white patients 58.7±7.8 years; P=.24) or family history (␹2 test, P=.66). from the same geographical area, and it will be impor- Finally, no other demographic or clinical variables cor- tant to extend these studies to a larger cohort of patients related significantly with age of onset (␹2 test for sex, from a wide range of different geographical locations, P=.92; for family history of dementia, P=.17; for FTD population groups, and ethnic origins. phenotype, P=.14). Our findings demonstrate a clear association between the presence of the H2 allele and earlier age at onset in FTD and indicate a key role of the tau gene vari- COMMENT ability in modulating FTD, even in patients who do not carry known tau gene mutations. Although the presence of tau gene mutations in FTD with The mechanism by which tau gene polymorphisms parkinsonism linked to chromosome 17 has proved the contribute to the modulation of FTD is currently un- important association of tau with FTD, the effects of the clear; it could be related to an effect on tau expression genetic variability of tau and other genes on the patho- or to an association with other disease-modifying fac- genesis of familial and sporadic FTD cases without tau tors. Indeed, a recent study investigating the effect of tau gene mutations remain unclear. haplotypes on tau expression showed that the H1 and H2 In our study of tau and ApoE genotype frequencies in alleles have different transcriptional activity in human 86 well-characterized FTD patients, the presence of the cell lines, with H1 being more efficient at driving tau gene tau H2 allele was associated with a significant decrease expression.20 (on average, 5 years) in the age of onset in FTD pa- The fact that the H2 haplotype in our patients was as- tients. A further decrease in the age of onset was ob- sociated with earlier onset in all 3 FTD subtypes sug- served in patients carrying 2 H2 alleles. To our knowl- gests the existence of common tau-related pathways in- edge, the role of tau haplotypes as a modulator of age of volved in the development of different clinical FTD onset in FTD has not been previously reported. In our subtypes. study, earlier onset in FTD patients carrying the tau H2 allele was seen throughout our data analysis, and it was Accepted for Publication: January 7, 2005. independent of clinical presentation (fvFTD, SD, or PNFA) Correspondence: Maria Grazia Spillantini, PhD, Brain Re- and familial history. pair Centre, Department of Clinical Neurosciences, For- At present, 18 of the 86 patients have died. Results of vie Site, Robinson Way, Cambridge, CB2 2PY, England autopsy in 13 of them have confirmed the diagnosis of ([email protected]).

(REPRINTED) ARCH NEUROL / VOL 62, SEP 2005 WWW.ARCHNEUROL.COM 1421

©2005 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 Author Contributions: Study concept and design: Borroni, 6. Ingelson M, Froelich Fabre S, Lilius L, et al. Increased risk for frontotemporal Yancopoulou, Sawcer, Hodges, and Spillantini. Acquisi- dementia through interaction between tau polymorphisms and apolipoprotein E ε4. Neuroreport. 2001;12:905-909. tion of data: Borroni, Yancopoulou, Tsutsui, Hodges, and 7. Verpillat P, Camuzat A, Hannequin D, et al. Association between the extended Spillantini. Analysis and interpretation of data: Borroni, tau haplotype and frontotemporal dementia. Arch Neurol. 2002;59:935-939. Yancopoulou, Tsutsui, Padovani, Sawcer, and Spillantini. 8. Corder EH, Sanders AM, Strittmatter WG, et al. Gene dose of apolipoprotein type Drafting of the manuscript: Borroni, Yancopoulou, and 4 allele and the risk of Alzheimer’s disease in late onset families. Science. 1993; Spillantini. Critical revision of the manuscript for impor- 261:921-923. 9. Gustafson L, Abrahamson M, Grubb A, Nilsson K, Fex G. Apolipoprotein-E geno- tant intellectual content: Yancopoulou, Tsutsui, Padovani, typing in Alzheimer’s disease and frontotemporal dementia. Dement Geriatr Cogn Sawcer, Hodges, and Spillantini. Statistical analysis: Disord. 1997;8:240-243. Borroni, Yancopoulou, and Tsutsui. Obtained funding: 10. Farrer LA, Abraham CR, Volicer L, et al. Allele ε4 of apolipoprotein E shows a Spillantini. Administrative, technical, and material sup- dose effect on age at onset of Pick disease. Exp Neurol. 1995;136:162-170. port: Tsutsui, Sawcer, Hodges, and Spillantini. Study su- 11. Minthon L, Hesse C, Sjogren M, Englund E, Gustafson L, Blennow K. The apolipoprotein E ε4 allele frequency is normal in fronto-temporal dementia, but cor- pervision: Hodges and Spillantini. relates with age at onset of disease. Neurosci Lett. 1997;226:65-67. Funding/Support: This study was supported by the UK 12. Boccardi M, Sabattoli F, Testa C, Beltramello A, Soininen H, Frisoni GB. APOE Medical Research Council, London, England. and modulation of Alzheimer’s and frontotemporal dementia. Neurosci Lett. 2004; Additional Information: Drs Borroni and Yancopoulou 356:167-170. contributed equally to this work. 13. Neary D, Snowden JS, Gustafson L, et al. Frontotemporal lobar degeneration: a consensus on clinical diagnostic criteria. Neurology. 1998;51:1546-1554. 14. Hodges JR. Frontotemporal dementia (Pick’s disease): clinical features and REFERENCES assessment. Neurology. 2001;56:S6-S10. 15. Hodges JR, Davies R, Xuereb J, Kril J, Halliday G. Survival in frontotemporal 1. Yancopoulou D, Spillantini MG. Tau protein in familial and sporadic diseases. dementia. Neurology. 2003;61:349-354. Neuromolecular Med. 2003;4:37-48. 16. Spillantini MG, Murrell JR, Goedert M, Farlow MR, Klug A, Ghetti B. Mutation in 2. Morris HR, Janssen JC, Bandmann O, et al. The tau gene A0 allele and progres- the tau gene in familial multiple system tauopathy with presenile dementia. Proc sive supranuclear palsy and related neurodegenerative diseases. J Neurol Neu- Natl Acad Sci U S A. 1998;95:7737-7741. rosurg Psychiatry. 1999;66:665-667. 17. Keppel G. Design and Analysis: A Researcher’s Handbook. Englewood Cliffs, NJ: 3. Panegyres PK, Zafiris-Toufexis K. Polymorphisms in the tau gene in sporadic Prentice-Hall International Inc; 1991. frontotemporal dementia and other neurodegenerative disorders. Eur J Neurol. 18. Pickering-Brown SM, Owen F, Isaacs A, et al. Apolipoprotein E ε4 allele has no 2002;9:485-489. effect on age at onset or duration of disease in cases of frontotemporal demen- 4. Sobrido MJ, Miller BL, Havlioglu N, et al. Novel tau polymorphisms, tau haplo- tia with Pick- or microvacuolar-type histology. Exp Neurol. 2000;163:452-456. types, and splicing in familial and sporadic frontotemporal dementia. Arch Neurol. 19. Short RA, Graff-Radford R, Adamson J, Baker M, Hutton M. Differences in tau 2003;60:698-702. and apolipoprotein E polymorphism frequencies in sporadic frontotemporal lo- 5. Hughes A, Mann D, Pickering-Brown S. Tau haplotype frequency in frontotem- bar degeneration syndromes. Arch Neurol. 2002;59:611-615. poral lobar degeneration and amyotrophic lateral sclerosis. Exp Neurol. 2003; 20. Kwok JB, Teber ET, Loy C, et al. Tau haplotypes regulate transcription and are 181:12-16. associated with Parkinson’s disease. Ann Neurol. 2004;55:329-334.

Visit www.archneurol.com.

As an individual subscriber to Archives of Neurology, you have full-text online access to the journal from 1998 forward. In addition, you can find abstracts to the journal as far back as 1975.

(REPRINTED) ARCH NEUROL / VOL 62, SEP 2005 WWW.ARCHNEUROL.COM 1422