Bioinformatics. Why do some neurons regenerate axon more successfully than others, and how can we boost regeneration in adult CNS neurons? We employ a range of techniques and strategies to tackles these questions. Click each for more details.
- Expression profiling (RNA-seq)
Dr. Ishwariya Venkatesh, drives this work with an impressive array of leading-edge bioinformatic skills
RNA-seq. A natural place to start understanding differences in axon growth between cell types is to look for differences in gene expression. Many labs have performed transcriptional profiling in different regeneration-competent cell types (peripheral neurons, embryonic neurons, zebrafish neurons, etc.), searching for genes that explain this ability. We surveyed many of these efforts in a review article (Blackmore, 2012). In our lab we mostly focus on developmental changes in gene expression, because immature neurons in the cortex display spectacular growth ability compared to their adult counterparts. Some examples of RNA-seq datasets from the lab are in our recent manuscripts (Wang et al 2018, Venkatesh et al 2018).
Network analysis. Our experience is that simply creating lists of differently expressed genes isn't terribly useful. What we really wan to understand is the underlying logic and especially the underlying control mechanisms - that's what we need to know in order to intervene. A critical step is network analysis, in which sets of genes are grouped according to known interactions between them. These could be physical interactions, transcriptional interactions, common regulatory pathways - there are many options. This analysis creates sub-networks, or modules. Each module tends to be highly enriched for different functions, which is a first step in understanding how the overall gene list might be functioning. We also find that sub-networks are enriched
for motifs that are bound by specific transcription factors. This motif analysis is fairly standard, but the key point is that you need to break the data into meaningful sub-networks before the underlying regulators emerge - their enrichment is essentially diluted out otherwise. Why transcription factors? Because they are powerful levers and the place to start if you want to modify gene expression in a cell. Stem cell biologists discovered long ago that the right set of transcription factors can turn a fibroblast into a stem cell or a neuron. We believe that in the same way, the right cocktail of transcription factors can turn a non-regenerating neuron into a successful regenerator. Defining subnetworks of genes important for axon growth, then identifying the transcription factors that regulate each, is one way we hope to identify the right combination of transcription factors to orchestrate pro-regenerative gene activity.
ATAC-seq footprinting. As alluded to above, we think a lot about how to find the right combination of transcription factors to promote axon regeneration. Transcription factors can interact in many ways, and each mechanism has important implications for the best way to discover interacting factors and the best way to use them. We wrote a recent review about this (Venkatesh and Blackmore, 2016).
As we surveyed the various mechanisms and their implications, one simple idea that rose to the top is the notion of co-occupancy. Basically, transcription factors that work together to control expression will tend to bind close to one another on shared regulatory DNA across the genome. So if you could somehow look across the whole genome and determine where each factor binds, then you could start to detect instances in which certain factors tend to cluster. It is now possible to do exactly that, using ATAC-seq datasets of DNA accessibility. Small DNA motifs that are stably bound by TFs create "footprints", small protected areas, which can be detected. Using this approach, for instance, we recently found that two factors called KLF6 and STAT3 tend to bind in close proximity. Not only that, the genes that are co-occupied by KLF6 and STAT3 are highly enriched for pro-regenerative functions. You can get more details in a recent manuscript.
DNA accessibility. Chromatin accessibility is emerging as a key regulatory mechanism in many areas of cell biology, and we believe it plays a critical role in axon regeneration as well. ATAC-seq datasets now enable genome-wide insight into DNA accessibility. We have prepared a publication (now in review, and available as a pre-print here), that uses an ATAC-seq approach to examine changes in DNA accessibility as neurons mature in early postnatal development, which is when they lose regenerative ability. We found widespread changes in accessibility, and pronounced closure of DNA in gene networks associated with axon growth functions. Maybe even more interestingly, we looked at four pro-regenerative transcription factors that we have used to try to promote axon regeneration in the adult cortex. Two of these (KLF6/7 and Sox11) are somewhat effective, and two others (STAT3 and JUN) were ineffective in our hands. It turned out that according to ATACseq datasets from adult cortex, the target genes of KLF7 and Sox11 were still accessible, but the targets of STAT3 and JUN were closed. Overall, the picture that is emerging is that chromatin restriction likely contributes to the developmental loss of axon growth, and may constrain the ability of therapeutic transcription factors to promote regeneration. We have some ideas about how to overcome these chromatin obstacles, for instance we think pioneer factors are going to be critical. Check out a recent manuscript for more details on this.
