It’s easy to take for granted how our physical environment shapes our behavior. An example is how our behavior changes in a bus as it becomes increasingly crowded. The impacts are multiple: our mobility decreases, our affinity for people we know increases, and certain areas become difficult to access (often times the exit). When it comes to chromatin in a cell’s nucleus, the basic premise is the same: the physical environment plays a critical role in controlling molecular behavior, even if the exact kinetics and functional effects of chromatin are distinct from the chemical reactions governing gene expression. Born of this idea, we set out to build a chemical-reactions framework for transcription that considers the complex effects of the physical organization of chromatin on gene expression, and therefore, cellular behavior. The work is described in our paper published in Nature Biomedical Engineering.

The challenge with chromatin and gene expression is one of length scales. While we can easily perceive our environment, the relevant scales within the cell nucleus are as small as 20 nm. Until recently, it was challenging to probe structures within this range in eukaryotic cells, and nearly impossible within live cells. To overcome this, we developed a suite of imaging technologies, the most prominently employed within this work being Partial Wave Spectroscopic (PWS) microscopy. PWS microscopy detects and quantifies the subdiffractional (<350 nm) organizational structure within cells by measuring alterations in backscattered light. As the variations in the backscattered spectrum increase, so does the scaling of mass density and the DNA packing heterogeneity. As a consequence, direct information on chromatin packing can be collected from within the cell nucleus of live cells without the need for labels (Video 1). Using these capabilities, we could then ask questions about how alterations in chromatin packing (and therefore crowding) impact gene expression.

Video 1: Label-free PWS microscopy allows detection and quantification of nanoscopic changes in cells over time.

Unlike a crowded bus, chromatin is unique in that it is both the determinant physical element as well as the functional unit involved in transcriptional reactions. Therefore, with respect to our analogy, chromatin is simultaneously the passengers and the bus. To overcome the challenge this posed, we used an integrative multiscale analysis approach that combines Brownian Dynamics, Monte Carlo simulations and systems modeling to probe how transcriptional reactions are influenced by crowding conditions. Since chromatin is an inherently heterogeneous structure formed by the spatial distribution of polymeric DNA, we employed and derived information about how changes to this polymeric structure would influence the distribution of mass within the cell nucleus as well as the overall accessibility of genes to transcription factors and polymerases. Remarkably, when we tested this model against experimental data, we found strong agreement, especially with respect to the collective behaviors of tens-to-hundreds of genes. In particular, we found that physical heterogeneity directly mapped to transcriptional heterogeneity.

As a consequence of this finding, and of previous studies showing a near-universal link between altered chromatin structure and carcinogenesis, we hypothesized that the physical structure of chromatin could serve as a major unidentified lever in the adaptive potential of cancer cells. In essence, we set out to test whether tuning the variations in chromatin packing, and therefore the variations in gene expression, would be a critical factor in cellular fitness (Figure 1). Using chemoevasion as a model, we tested if decreased variations in chromatin packing would render cancer cells more sensitive to chemotherapeutic agents. As we demonstrate, this is indeed the result (Figure 2). In sum, we provide a means to target physical forces within the cell nucleus to predictively control the expression of many genes simultaneously. Using these capabilities, we further demonstrate that this approach can be used to treat complex diseases, such as cancer.

Our paper: Almassalha, L. M.*, Bauer, G. M.*, Wu, W.* et al. Macrogenomic engineering via modulation of the scaling of chromatin packing density. Nat. Biomed. Eng. doi:10.1038/s41551-017-0153-2.

*Authors contributed equally to this work