Pathogenic Huntingtin Repeat Expansions in Patients with Frontotemporal Dementia and Amyotrophic Lateral Sclerosis

Abstract

We examined the role of repeat expansions in the pathogenesis of frontotemporal dementia (FTD) and amyotrophic lateral sclerosis (ALS) by analyzing whole-genome sequence data from 2,442 FTD/ALS patients, 2,599 Lewy body dementia (LBD) patients, and 3,158 neurologically healthy subjects. Pathogenic expansions (range, 40-64 CAG repeats) in the huntingtin (HTT) gene were found in three (0.12%) patients diagnosed with pure FTD/ALS syndromes but were not present in the LBD or healthy cohorts. We replicated our findings in an independent collection of 3,674 FTD/ALS patients. Postmortem evaluations of two patients revealed the classical TDP-43 pathology of FTD/ALS, as well as huntingtin-positive, ubiquitin-positive aggregates in the frontal cortex. The neostriatal atrophy that pathologically defines Huntington's disease was absent in both cases. Our findings reveal an etiological relationship between HTT repeat expansions and FTD/ALS syndromes and indicate that genetic screening of FTD/ALS patients for HTT repeat expansions should be considered.

Keywords: amyotrophic lateral sclerosis; frontotemporal dementia; huntingtin; repeat expansions; whole-genome sequencing.

Figure 1.. Repeat sizing analysis identifies FTD/ALS patients harboring HTT CAG repeat expansions.

(A) An ideogram of chromosome 4 showing the HTT gene location at 4p16.3, the gene transcript (exon 1 in red), and representative repeat-primed PCR chromatograms depicting wild-type and HTT CAG repeat expansions. (B) The distributions of HTT CAG repeat expansions in the FTD/ALS (n = 2,442), LBD (n = 2,599), and control (n = 3,158) discovery cohorts. Insets are magnified views showing the number of cases carrying CAG repeat expansions ε 36 repeats. (C) Ages at symptom onset among FTD/ALS patients versus the size of their HTT repeat expansions. The curve represents the estimated onset age and corresponding standard deviation based on the number of CAG repeats (Langbehn et al., 2010). (D) The allelic structure of patients carrying HTT repeat expansions. The pathogenic repeat sequence is represented by [CAG]_n, where n corresponds to the number of repeats. The trailing CAG-CAA glutamine sequence, the CCG-CCA proline sequence, and the [CCT]_n codons are also shown (Ciosi et al., 2019). See also Figures S1, S2, S3, and S6.

Figure 2.. Quantitative analysis of somatic HTT mosaicism across brain regions in FTD/ALS and Huntington’s disease patients.

(A-C) Measurements of somatic mosaicism across brain regions obtained at autopsy for Patient #5, Patient #8, and a Huntington’s disease patient. (D) Heat map of the expansion indices aggregated for the three patients, with tissues displayed in order of mean expansion index. Grey boxes indicate regions that were not assessed. M Ctx, Motor Ctx = motor cortex; GP/Put = globus pallidus/putamen; Corp C, Corpus C = corpus callosum; P Ctx, Pariet Ctx = parietal cortex; F Ctx, Frontal Ctx = superior frontal cortex; Cd/Acb/Put = caudate/accumbens/putamen; Temp Pole = temporal pole; Cing Gyrus = cingulate gyrus; Hipp head = head of the hippocampus; Mb/SN = midbrain/substantia nigra; Hipp = hippocampal formation; L Pons = lower pons; Subthal N = subthalamic nucleus; U Pons = upper pons; SC Cerv = cervical spinal cord; Dent Gyrus = dentate gyrus; SC Lumb = lumbar spinal cord; SC Thor = thoracic spinal cord. See also Figure S4.

Figure 3.. Neuropathologic changes observed in an ALS patient carrying a full-penetrance pathogenic HTT repeat expansion (Patient #5).

(A) A representative cervical cord section showing pallor of the lateral (*) and anterior corticospinal tracts (**) with ventral horn atrophy. (B) The loss of motor neurons of the anterior horns was severe. (C) Nucleocytoplasmic TDP-43 translocation (arrows) in the prefrontal cortex (BA9 area). (D) Frequent p62 (red arrow) and huntingtin (black arrow) dystrophic neurites (Insert), intranuclear huntingtin (black arrow), and p62 (red arrow) inclusions were noted within the prefrontal cortex. (E) The neostriatum was apparently normal, for example, at the level of the nucleus accumbens, and neither neuronal loss nor reactive gliosis was detectable. (F) & (G) Occasional huntingtin aggregates were seen within the neuropil of the nucleus accumbens. (H) The tail of the caudate nucleus was not atrophic, and the neuronal density was normal and without reactive gliosis. (I) & (J) Rare huntingtin aggregates involve the neuropil of the tail of the caudate nucleus (arrows). Scale bars: A: 1 mm, and C-D: 50 microns. See also Figure S5.

Figure 4.. Neuropathologic changes observed in an ALS patient carrying a full-penetrance pathogenic HTT repeat expansion (Patient #8).

(A) A coronal section of the fresh brain showing that the caudate, putamen, and globus pallidus was intact with no evidence of atrophy. (B) Luxol fast blue/hematoxylin and eosin staining of the caudate nucleus showed no neuronal loss or gliosis. (C) Ubiquitin immunostaining of the caudate nucleus showed extranuclear aggregates (arrow) and rare intranuclear inclusions (arrowhead). (D-E) Immunohistochemistry for polyglutamine expansions showed occasional extranuclear inclusions within the caudate nucleus (D) and the peri-Rolandic cortex (E, arrows). (F) There was severe motor neuron loss within the anterior horn of the spinal cord (Luxol fast blue/hematoxylin and eosin). (Inset) A remaining motor neuron with a TDP-43 cytoplasmic inclusion. Scale bars: B: 50 microns, C-D: 20 microns, and F: 100 microns.