Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Feb 6;188(3):623-639.e19.
doi: 10.1016/j.cell.2024.11.038. Epub 2025 Jan 16.

Long somatic DNA-repeat expansion drives neurodegeneration in Huntington's disease

Robert E Handsaker  1 , Seva Kashin  2 , Nora M Reed  3 , Steven Tan  3 , Won-Seok Lee  3 , Tara M McDonald  3 , Kiely Morris  4 , Nolan Kamitaki  3 , Christopher D Mullally  3 , Neda R Morakabati  4 , Melissa Goldman  3 , Gabriel Lind  3 , Rhea Kohli  3 , Elisabeth Lawton  4 , Marina Hogan  3 , Kiku Ichihara  3 , Sabina Berretta  5 , Steven A McCarroll  6
Affiliations

Long somatic DNA-repeat expansion drives neurodegeneration in Huntington's disease

Robert E Handsaker et al. Cell. .

Abstract

In Huntington's disease (HD), striatal projection neurons (SPNs) degenerate during midlife; the core biological question involves how the disease-causing DNA repeat (CAG)n in the huntingtin (HTT) gene leads to neurodegeneration after decades of biological latency. We developed a single-cell method for measuring this repeat's length alongside genome-wide RNA expression. We found that the HTT CAG repeat expands somatically from 40-45 to 100-500+ CAGs in SPNs. Somatic expansion from 40 to 150 CAGs had no apparent cell-autonomous effect, but SPNs with 150-500+ CAGs lost positive and then negative features of neuronal identity, de-repressed senescence/apoptosis genes, and were lost. Our results suggest that somatic repeat expansion beyond 150 CAGs causes SPNs to degenerate quickly and asynchronously. We conclude that in HD, at any one time, most neurons have an innocuous but unstable HTT gene and that HD pathogenesis is a DNA process for almost all of a neuron's life.

Keywords: CAG; DNA repeats; Huntington’s disease; neurodegeneration; repeat instability; single-nucleus RNA-seq; somatic expansion; striatal projection neurons; triplet repeat disorders.

PubMed Disclaimer

Conflict of interest statement

Declaration of interests Patent applications filed by the Broad Institute of MIT and Harvard related to this work include subsets of the authors as inventors. S.A.M. has received compensation for scientific advice to Roche, Pfizer, Biogen, Vertex, and LoQus23 Therapeutics.

Figures

Figure 1.
Figure 1.
SPN loss in persons with HD. (A) Cell-type proportions in the striatum (anterior caudate) of each donor. (B) SPN loss with HD progression (increasing CAP score). Unaffected controls in white; among persons with HD, darker shades of gray represent increasing CAP score. These same data are shown on a log scale in Figure S2A. (C) Decline in iSPNs (D2 SPNs, y-axis) and dSPNs (D1 SPNs, x-axis) with HD progression. Gray shading as in B. (D,E) Expression of HTT transcripts (units: UMIs per 100k) in the nuclei of (D) striatal cell types and (E) SPN subtypes, among 53 control (unaffected) donors. Boxes represent the interquartile range; whiskers extend beyond the hinges by 1.5 times the interquartile range.
Figure 2.
Figure 2.
Single-cell analysis of HTT CAG-repeat length and genome-wide RNA expression in the same nuclei. (A) Molecular approach. Two sequencing libraries are prepared from the same set of barcoded nuclear cDNAs. The first is a conventional snRNA-seq library. The second library samples the CAG-repeat sequence in HTT gene transcripts and is analyzed by long-read sequencing. The presence of shared cell barcodes in the two libraries allows each CAG-repeat sequence to be matched to the RNA-expression profile of the nucleus from which it was sampled. (B) Concordance between pairs of measurements of CAG-repeat length from different HTT RNA transcripts (with different UMIs) in the same nucleus (same cell barcode). For each such transcript-pair, the longer of the two CAG-repeat measurements is shown on the y-axis. Nuclei in which both measurements are from the long (HD-causing) allele (orange) make it possible to measure precision and error rate. (C) Distributions of CAG-repeat length measurements by donor and cell type. (D) Distributions of CAG-repeat length measurements in SPNs (for each donor in C), showing only the long (HD-causing) allele and the much-wider range of CAG-repeat lengths that HD-causing alleles attain in SPNs. Note that the mode/peak in D corresponds to the distribution for the long (HD-causing) allele in C. (E) The full length distributions for the HD-causing CAG repeat in SPNs for each donor in C. Blue shaded areas are smoothed density estimates of the repeat-length distribution. Overplotted black points show the measurements in individual SPNs. In all six donors, the repeat-length distribution exhibits an armadillo-like shape, in which the DNA repeat in most of a donor’s SPNs has undergone modest expansion (up to 100 CAGs) but in a small fraction of SPNs has undergone greater expansion (up to 500+ CAGs). Figure S3 and Methods S1, section 4 contain additional visualizations.
Figure 3.
Figure 3.
An apparently high length threshold for effects of the HTT CAG repeat on SPN biology. (A) Gene-expression comparisons of sets of SPNs (from the same tissue sample) grouped into deciles based on the CAG-repeat length of their HD-causing HTT allele. Gray scale: magnitude of gene-expression difference (one minus the correlation coefficient); black indicates maximal difference observed in any comparison; white indicates no difference. Figure S4A shows similar analyses of SPNs from six persons with HD. (B,C) Comparisons (volcano plots) of gene expression between sets of SPNs, sampled from the same person with HD but with CAG-repeat lengths in different ranges. p-values (y-axis) are derived from a Wilcoxon test across the individual SPNs in each group. Fold-changes (x-axis) are the log2-ratio of the group averages (positive fold-change indicates higher expression in the SPNs with longer repeats). Analyses for all six donors are in Figure S4B–S4E. (D) Consistency of long(150)-repeat-expansion associated SPN gene-expression changes among individual persons with HD. Each panel is a pairwise comparison of SPN gene-expression data involving two persons with HD (x- and y-axes), in which the values on the two axes are the log2-fold-changes in gene expression when comparing an individual’s SPNs with >150 CAGs to the same individual’s SPNs with <150 CAGs. Genes whose expression levels change significantly with repeat expansion in at least one of the donors are shown. More analyses are in Figure S4F; see also Methods S1, section 3.
Figure 4.
Figure 4.
Gene-expression changes in SPNs with somatic CAG-repeat expansion beyond 150 CAGs. (A) On each plot, one donor’s individual SPNs are ordered from left to right by the length of their HTT CAG repeat (the columns of the heatmap). Each row shows expression data for a specific gene in these SPNs. (The genes shown are genes found to change in expression with repeat expansion.) Shades of each facet show the level of expression of that gene in that SPN, relative to the average SPN with repeat length < 150 in that donor. Example trajectories for individual genes are in Figure S6A. (B) As in A, on each plot, individual SPNs are ordered from left to right by their CAG-repeat length. Each SPN is represented by both a blue and a green point: blue points show the median fold-change of a set of 192 genes that decrease in expression with CAG-repeat expansion (C− genes). Green points show the median fold-change of a set of 274 genes that increase in expression with CAG-repeat expansion (C+ genes). See also Table S2. (C) Genes that decline in expression during Phase C (C− genes) are genes that are more strongly expressed in healthy SPNs than in striatal interneurons. Gray points: all genes (cell-type-specific expression levels in unaffected individuals). Colored circles: Genes whose expression levels decline (blue) or increase (green) in SPNs with HTT CAG-repeat expansion beyond 150 units (in phase C).
Figure 5.
Figure 5.
De-repression crisis (phase D) and subsequent SPN elimination (phase E). (A) De-repression of genes that are normally silent in SPNs. Points represent individual SPNs; red and pink points are those SPNs whose phase C expression changes (Figure 4B) have progressed beyond the threshold values shown in the legend. y-axis: de-repression score, the number of transcripts (UMIs) detected from 107 “phase D” genes that are normally silent in SPNs. Additional visualizations of this relationship are in Figures S7C–S7D. (B) Fraction of SPNs exhibiting this de-repression phenotype, in relationship to the increasing dysregulation (reduced expression) of the phase C− genes. Error bars represent 95% binomial confidence intervals. (C) Fraction of SPNs exhibiting this de-repression phenotype, in relation to CAG-repeat length. Error bars represent 95% binomial confidence intervals. (D) Expression of HOX cluster genes (left) and CDKN2A (right) in SPNs of persons with HD. (E,F) SPN loss and transcriptionopathy as HD progresses. (E) Relationship of SPN survival (shown on a logarithmic scale) to HD progression as indexed by CAP score. The dashed curve shows a logistic function fit to the SPN-survival data; its slope (derivative) estimates average rates of SPN loss as HD progresses. Error bars represent 95% binomial confidence intervals. (F) Relationship of the frequency of SPN phase C transcriptionopathy (fraction of SPNs, y-axis) to HD progression as indexed by CAP score (x-axis) for the same donors in E. The dashed curve is the negative of the derivative of the SPN-survival curve from E (i.e., is proportional to the estimated rate of SPN loss). In E and F donors were excluded if they had fewer than 25 surviving SPNs (n=8) or were beyond the plot range (CAP score >800, n=2).
Figure 6.
Figure 6.
Computational modeling of somatic CAG-repeat expansion dynamics. (A) This schematic illustrates mechanisms (established in earlier work) for non-replicative DNA-repeat expansion in post-mitotic cells. Extrahelical DNA extrusions (“slip-out” structures) form from mispairing within the CAG-repeat tract after strand separation (e.g. due to transcription). The MutSβ complex can bind to these transient structures, initiating DNA excision and resynthesis, which (when initiated on the strand opposite the slip-out) can result in the incorporation of an extra repeat unit. Occurring many times across a human lifetime, this mutational process results in a progressive expansion of the DNA repeat. (B) Cumulative distributions of CAG-repeat length measurements in SPNs from six deeply sampled donors. The gray shaded region highlights the range (70–90 CAGs) over which somatic expansion appears to greatly accelerate. (C) Distributions of CAG-repeat length measurements in SPNs from these same donors (black) overlaid with the results of stochastic models (orange) for which a few key parameters (such as mutation rate) have been fitted to each donor’s repeat-length and SPN-loss data. (D) Effect of changing a single variable (inherited/germline CAG-repeat length) in the model for a typical donor, keeping the other fitted parameters fixed. Each curve indicates the predicted CAG-repeat length distribution for surviving SPNs at each decade (ages 10 to 80). (E) Model-estimated relationship between inherited germline CAG-repeat length and age at clinical motor onset. As a proxy for age of onset, we used the predicted time at which 25% of a donor’s SPNs have been lost. We estimated this age-of-onset proxy at different hypothetical inherited repeat lengths. The shapes of the resulting curves approximate the known relationship between inherited repeat length and age of HD onset. (F) Observed CAG-repeat length distributions for two donors (donors 7 and 8, Table S1) with HD-causing alleles who passed away prior to onset of clinical motor symptoms, based on review of their medical records. Data from a typical symptomatic donor (donor 5) is shown at bottom for comparison (distributions for several other symptomatic donors are in Figure 2E).
Figure 7.
Figure 7.
ELongATE (extra-long repeats acquire toxic effect): a model for SPN pathology in HD. (A) Individual neurons pass asynchronously through five key pathological phases, spending >95% of their lives in a long period of DNA-repeat expansion (a “ticking DNA clock”, phases A and B) with a biologically harmless (but unstable) HTT gene. Individual neurons asynchronously exit phase A and proceed through the subsequent, faster phases. (B) Prediction of the fraction of SPNs in each of the five phases, across the latent, peri-onset and then progressive stages of HD. The estimated trajectories are based on the data from a representative donor. The indicated ranges for clinical motor onset and escalating symptoms are approximate. The illustrated onset range, representing loss of 20% to 50% of a donor’s SPNs, was inferred from available age-of-onset data for the brain donors whose tissue we analyzed. The time estimates listed for each phase are for persons who inherit the more common HD-causing alleles (40 to 45 CAGs).
See this image and copyright information in PMC

References

    1. Huntington’s Disease Collaborative Research Group (1993). A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes. Cell 72, 971–983. - PubMed
    1. Gusella JF, Lee J-M, and MacDonald ME (2021). Huntington’s disease: nearly four decades of human molecular genetics. Hum. Mol. Genet. 30, R254–R263. - PMC - PubMed
    1. Langbehn DR, Brinkman RR, Falush D, Paulsen JS, Hayden MR, and International Huntington’s Disease Collaborative Group (2004). A new model for prediction of the age of onset and penetrance for Huntington’s disease based on CAG length. Clin. Genet. 65, 267–277. - PubMed
    1. Scahill RI, Zeun P, Osborne-Crowley K, Johnson EB, Gregory S, Parker C, Lowe J, Nair A, O’Callaghan C, Langley C, et al. (2020). Biological and clinical characteristics of gene carriers far from predicted onset in the Huntington’s disease Young Adult Study (HD-YAS): a cross-sectional analysis. Lancet Neurol. 19, 502–512. - PMC - PubMed
    1. Tabrizi SJ, Schobel S, Gantman EC, Mansbach A, Borowsky B, Konstantinova P, Mestre TA, Panagoulias J, Ross CA, Zauderer M, et al. (2022). A biological classification of Huntington’s disease: the Integrated Staging System. Lancet Neurol. 21, 632–644. - PubMed

LinkOut - more resources