Simulations and parameter estimation for individual persons with HD

Huntington’s Disease is caused by inheritance of an expanded DNA repeat in exon 1 of the Huntingtin (HTT) gene. In unaffected individuals, a repeated triplet of the DNA letters “CAG” appears less than 35 times within this gene, typically in the range of 17 to 20 times for most people. People who inherited an allele with 36 or more copies of this CAG triplet are at high risk for Huntington’s Disease, with the age of symptomatic onset strongly predicted by the length of the inherited repeat (the longer the repeat, the earlier symptom onset occurs).

It has been known that this DNA repeat is not only variable across individuals, but that in individuals that inherit a long repeat, the repeat will then be unstable across a person’s lifetime, with a propensity for the repeat to expand somatically with age. The degree of somatic expansion had been observed to differ between different tissues and across different regions of the brain.

In our recent work on Huntington’s Disease, we generated high-resolution data on the amount of somatic expansion of this DNA repeat in individuals neurons in the human brain. We found that somatic expansion is dramatic, with some neurons reaching repeat lengths of up to 1000 CAG units. In every patient, however, the majority of the neurons were always observed to have more modest expansion (up to 80 or so CAG units) with only a minority exhibiting very long expansions. The evidence also suggests that the neurons with the really long expansions are lost as the disease progresses.

We combined these insights and this high-resolution data to construct mathematical models and simulations to explain the observed data and to better understand the temporal dynamics of the somatic repeat expansion in the striatal projection neurons in the anterior caudate, one of the most vulnerable types of neuronons in Huntington’s Disease.

The first plot shows the observed distribution of the repeat lengths from individual neurons from the brain of one Huntington’s Disease patient (black) as the results of fitting these models to the observed data (orange). Although the model is fit to a single timepoint at which we measured the patient’s brain tissue, the models predict the entire temporal dynamics of the process, from birth onwards, based on the data from this single donor. This is shown in the animated simulation in the second figure, where each dot represent the somatic expansion of a single neuron from among the population of neuron’s in the pathttps://mccarrolllab.org/dna_repeat_expansion_simulations/?preview=trueient’s caudate.

Effect of inherited CAG-repeat length

Using a model of somatic expansion fitted to a single patient, it is possible to ask how the model for the expansion dynamics would change under the assumption that the patient has been born with a CAG allele of a different length. The plots below show how the trajectory of disease progression would be predicted to change for one typical donor given four different assumptions about their inherited CAG allele length, along with animations illustrating the dynamic somatic expansion process for each of these hypothetical scenarios.

Dynamics of dSPN and iSPN somatic repeat expansion and loss

There are two main subtypes of striatal projection neurons (SPNs) found in the caudate. Direct pathway SPNs (dSPNs) project axons to the interface between the basal ganglia and the rest of the brain, whereas the indirect pathway SPNs (iSPNs) project axons only within the basal ganglia, thus connecting indirectly to the rest of the brain. It has been observed that the indirect pathway SPNs appear to be lost somewhat earlier than direct SPNs during Huntington’s disease.

Based on our experimental data, we modeled the difference in the dynamics of the repeat expansion between direct and indirect SPNs.

The plot above shows the model of the difference in survival of these two types of neurons over time, based on the data from one patient. The animation shows the relative rates of somatic expansion in these two types of cells via a simulation of the somatic expansion process.