The capabilities of all organisms are circumscribed by a living code written in DNA. This code is unique to each organism and defines the capacities of the cell, providing the blueprints which enable complex organisms to flourish. Evolutionary forces converged on DNA due to its remarkable balance of stability and flexibility; it can persist for eons yet is mutable enough to allow for the breathtaking diversity that epitomizes life on Earth. Despite these extraordinary properties, DNA is not without its faults; its mutability, so critical for diverse mechanisms of survival, is a double edged sword that can hinder as easily as it can help. Changes to DNA can occur through a wide variety of mechanisms, which we broadly call mutagens, and the spectra of mutations arising from these mutagenic forces can have a devastating impact on the cell or, more rarely, on the organism as a whole.


The Baldwin Lab is interested in understanding the role of mutations in iPSCs and neurons. Because iPSCs are used in disease modeling studies, and increasingly in clinical contexts, it is important to understand the scope and source of mutations among iPSC line derived from various reprogramming methods and sources. Previously published work from our lab showed that blood derived iPSCs exhibited more mutations than iPSCs derived from younger patients. Curiously, however, iPSCs derived from extremely old (>90 yrs) patients showed fewer mutations than expected (LoSardo et. al, Nature Biotech 2017). Subsequent work on iPSC genomes has revealed that different reprogramming methods have varying impacts on mutation rates. Interestingly, it appears as though the commonly used episomal reprogramming method is more mutagenic than the integrating lentiviral reprogramming method. Additionally, different reprogramming methods give rise to unique mutational spectra which differ from somatic mutations (Duran and LoSardo et. al. Manuscript in prep)

Genome integrity is widely recognized as playing a crucial role in cell health and efficacy. What is less understood is how somatic genome variance between cells in a population impacts the function of the whole system. This question is particularly relevant in the case of neurons, where many individual cells each assume an essential role in a neural network. Indeed, there are several lines of evidence indicating that maintaining genome integrity is critically important to neurons. An average wild-type neuron may be required to maintain genome integrity for decades or even, in rare cases, for over a century. This longevity raises the question of whether or not a failure to maintain genome integrity over time may in some way contribute to age-associated loss of cognitive function. Thus, whole genome sequencing (WGS) data of single young and aged neurons could implicate mechanisms of mutagenesis which underlie age-associated phenotypes.

One of the main challenges in addressing neuronal mutations is the senescent state of mature neurons. Conducting WGS experiments on post-mitotic cell types typically requires error-prone single cell whole genome amplification (SC-WGA) technologies. To circumvent this, our lab utilizes somatic cell nuclear transfer (SCNT) in the mouse, which allows us to amplify the genome evenly in a way that is minimally mutagenic. Although there are clever strategies for mitigating the drawbacks of SC-WGA, to the best of our knowledge SCNT remains the only method to assay all classes of mutation across the entire genome. SCNT thus provides a high-resolution map of individual genomes which complement the high-throughput studies performed in human neurons by SC-WGA. Our lab previously published on this approach applied to mitral and tufted neurons, finding over 100 unique neuronal mutations per single mature neuron (Hazen and Faust et. al. Neuron 2016). More recently, we have extend our SCNT approach to assay genomes of rod photoreceptors of varying ages (p6-2.5 years), which are exposed to UV and are responsible for a highly polygenic set of retinal disorders. This work provides evidence that (1) neurons in mice accumulate mutations at an annual rate comparable to humans, (2) there are both recurring and unique mechanisms of mutation across different neuronal subtypes, and (3) the most deleterious classes of mutation occur in highly expressed genes (Duran et. al. Manuscript in prep)


Figure 1 MCNT-ESCs and Their Whole-Genome Sequences

(A) Dissociated MT neurons shown with injection pipette

(B) tdTomato-positive blastocysts generated from MT neurons

(C) tdTomato-positive MCNT-ESCs

(D) Schematic of Pcdh21/Cre-Ai9 donor animals and the MCNT-ESC lines and control tissues sequenced from each animal

(E) Representative PCR subclone validation for two structural variants (SVs)

(F and G) Observed mutations (black/red bars) and estimated mutational burden based on the false negative rate. For SVs, observed and predicted values for breakpoints are plotted.