Analysis of short tandem repeats linked to polyglutamine diseases from whole-genome sequencing reveals intermediate alleles of HTT associated with an early disease onset in C9orf72 carriers

Abstract

Carriers of the GGGGCC pathogenic expansion in C9orf72 can develop symptoms of frontotemporal dementia and/or amyotrophic lateral sclerosis, with variable and unpredictable ages at onset. Previous studies aiming to decipher the genetic bases of the clinical variability in this rare disease included bi-allelic polymorphisms, excluding short tandem repeats. Whole-genome sequencing data of 195 C9orf72 patients were used to consider all short tandem repeats linked to polyglutamine disorders as potential genetic modifiers given the existing links between C9orf72 and polyglutamine diseases. Intermediate alleles of HTT encoding huntingtin were associated with an earlier age at onset among C9orf72 carriers in the discovery cohort (n = 195, P = 0.0003) and in a European replication cohort (n = 145, P = 0.006). In the merged cohort (n = 340), the average difference of age at disease onset was 9.42 ± 2.14 years (P = 1.3 × 10^e-5) between carriers and non-carriers of HTT-intermediate alleles. Neuropathology of one C9orf72 case heterozygous for HTT-intermediate allele showed typical TDP-43 inclusions related to the C9orf72 pathogenic expansion and was negative for polyglutamine inclusion. No somatic expansion of HTT was detected in blood of all C9orf72exp/HTT-intermediate carriers. If this study reinforces potential biological links between huntingtin and C9orf72 that remain to be explored, the results also illustrate the interest of considering short tandem repeats from whole-genome data in association studies which paves the way to more exhaustive approaches to explore the trait heritability due to short-tandem-repeats still hidden in the genome.

Keywords: C9orf72; amyotrophic lateral sclerosis; frontotemporal dementia; huntingtin; whole-genome sequencing.

Graphical Abstract

Figure 1

Age at disease onset in function of HTT CAG repeat number (longest allele) in C9orf72 patients. (A) discovery cohort (n = 195); (B) replication cohort (n = 145); (C) merged cohort (n = 340). LOWESS curves are appearing for each graph. Each data point represents the longest number of CAG in HTT for each C9orf72 patient.

Figure 2

HTT CAG repeat profiles in C9orf72 patients’ blood carrying IA-HTTs from PCR-fragment analyses. X-axis, top of the figure: PCR length (bp, same scale for all profiles), Y-axis: fluorescent units. Genotypes inferred from these profiles are indicated and in accordance with estimates from short-read whole-genome sequencing.

Figure 3

Neuropathology of a C9orf72 case carrying one intermediate allele of HTT (CAG_n = 26/31). (A, B) Cerebellar grain layer showing discrepancy between the presence of p62+ inclusions (A, arrows) and the absence of loss of normal TDP43+ labelling of nuclei (B); (C) Skein-like inclusions in motor neurons of the anterior horn of the spinal cord. (D, E) Striatum. Oligodendroglial inclusions seen with anti-p62 (D, arrows) and anti-TDP43 (E, arrows). Scale bars = 20 μm.

Figure 4

Anti-HTT immunohistochemistry in twelve regions of the central nervous system of the patient carrying C9orf72exp and IA-HTT. (A) lower mesencephalon; (B) inferior olivary nucleus; (C) cerebellum; (D) cervical spinal cord; (E) thalamus; (F) hippocampus; (G) putamen; (H) caudate nucleus; (I) nucleus basalis of Meynert; (J) midfrontal cortex; (K) temporal pole; (L) calcarine sulcus. Diffuse intracytoplasmic positivity were observed in the inferior olivary olive (B). No intranuclear inclusions were found in any region examined. Scale bars = 40 μm.