What does physics have to do with cancer?

Abstract

Large-scale cancer genomics, proteomics and RNA-sequencing efforts are currently mapping in fine detail the genetic and biochemical alterations that occur in cancer. However, it is becoming clear that it is difficult to integrate and interpret these data and to translate them into treatments. This difficulty is compounded by the recognition that cancer cells evolve, and that initiation, progression and metastasis are influenced by a wide variety of factors. To help tackle this challenge, the US National Cancer Institute Physical Sciences-Oncology Centers initiative is bringing together physicists, cancer biologists, chemists, mathematicians and engineers. How are we beginning to address cancer from the perspective of the physical sciences?

Figure 1. Similarities among soap bubbles, cells in the Drosophila Nature Reviews eye and loss of Cancer tissue organization in cancer

The normal ommatidium structure within the Drosophila melanogaster retina is shown (part a). Note the beautifully regular pattern of six cells, four in the middle and two around, which constitutes the basic organizational unit. The geometry of the ommatidium mirrors that of four soap bubbles (part b), suggesting that surface tension has an important role in defining the shape of the ommatidium. Cells misexpressing N-cadherin are marked with green fluorescent protein (GFP; false-coloured purple) (part c). Cell outlines are visualized with fluorescent E-cadherin (green). A simplified form of an equation is shown, which was used by Hilgenfeldt et al. to model the interplay of physics (for example, membrane mechanics) and biology (for example, N- and E-cadherin levels). Kronecker delta function terms representing homophilic interactions are not shown for clarity. Equation terms are explained in the text. Alterations in cell and tissue architecture in colorectal cancer are shown (part d). The arrows represent normal and dysplastic crypts (with diameters of 75 microns and 150 microns, respectively). Parts a–c are reproduced, with permission, from REF. (2004) Macmillan Publishers Ltd. All rights reserved. Part d is courtesy of B. Vogelstein, John’s Hopkins University, USA.

Figure 2. Tissues are complex dynamic systems that feature multiscale mechanochemical coupling

Progress has been made particularly in delineating the molecular mechanisms of force generation by the cytoskeleton, the details of cell–cell adhesion and force sensing by proteins such as talin 1 and vinculin. However, mechanical effects in biology are inherently multiscale, in the sense that single cells can generate stresses and strains that contribute to the mechanics and the organization of entire tissues, and in turn, millimetre-scale tension fields within tissues can provide signals that are sensed by potentially millions of cells within that tissue. Understanding this interplay will require new types of experiments that can interrogate all relevant scales simultaneously, and broad conceptual and theoretical advances. ECM, extracellular matrix; FAK, focal adhesion kinase; P, phosphorylation; ROCK, Rho-associated, coiled-coil containing protein kinase; SFK, SRC family kinase.

Figure 3. Physical sciences shed light onto nucleosome and transcription factor competition and chromosome packaging

Schematic illustration of the partial wave spectroscopy (PWS) experiment (part a). Cells from the tissue of interest are brushed from the tissue and studied ex vivo. Each cell is individually scanned in a two-dimensional lattice of diffraction-limited pixels (strictly voxels because of the thickness of the cell). The cells are illuminated with light of wavelength λ, and backscattered waves propagating along one-dimensional paths within each voxel are measured as functions of lateral position over the cell (x, y) and of wavelength λ. Wavelength-dependent variability in reflectivity R at each pixel arises from the interference of photons scattered from refractive index fluctuations within the cell at that location, providing information about internal cellular heterogeneity on subwavelength length scales. The magnitude of the spectral fluctuations at each pixel in the image are represented by a disorder strength, L_d = <Δn²> × l_c, where Δn is the local fluctuations in intracellular refractive index and l_c is the correlation length of these fluctuations. For many different cancer types, the measured disorder strength increases in cells that that are located some distance away from a cancerous lesion, even though the cells being analysed are themselves not cancerous. Parts b–g illustrate the free energy landscape from a typical section of genomic DNA. The 10 bp oscillations in interaction energy arise from the ~10 bp helical symmetry of DNA, the approximately circular DNA wrapping in the nucleosome, and the mechanical nature of the nucleosome–DNA interaction (part b). Parts c–g illustrate nucleosome–transcription factor competition. The purple curves in part d show the corresponding distribution of nucleosome start probabilities with no transcription factor present (from the solution of equation 3 in BOX 2), and the top two rows of orange ovals (part e) represent two of the many different nucleosome configurations, all of which have significant probability in the equilibrium distribution. The purple curves in part g show the corresponding distribution of nucleosome occupancy (summed over the full set of allowed configurations, from the solution of equation 4 in BOX 2). The free energy landscape of a transcription factor, shown in part c, highlights two specific binding sites with equal energies (affinities), as indicated by the two sharp valleys in energy (the two green lines). The transcription factor also has slightly varying affinities for nonspecific sequences (indicated here by the thickness of the green bar covering the remainder of the landscape). In this example, the model is solved for a transcription factor concentration that is sufficiently high so as to give high occupancy at one of the binding sites (probability approaching 1; green curve in part d). However, because this site overlaps a preferred nucleosome location, binding at this site requires redistribution of the local nucleosome organization, to a new distribution of nucleosome start probabilities (orange curves in part d). The bottom two rows of orange and green ovals in part f represent two of the many different configurations of nucleosomes and bound factors, respectively, which have significant probability in the resulting equilibrium distribution. The corresponding nucleosome and transcription factor occupancies are shown in orange and green, respectively, in part g. Even though the two transcription factor sites have identical intrinsic affinities for the transcription factor, the required nucleosome redistribution is energetically inexpensive for the right-hand transcription factor binding site, compared with the left-hand transcription factor site, which allows a higher occupancy at the right-hand site for a given transcription factor concentration. The image of the nucleus in part a is reproduced, with permission, from REF. (1997) National Academy of Sciences. The graph in part a is reproduced, with permission, from REF. . Parts b–g are reproduced, with permission, from REF. . a.u., arbitrary units.

Figure 4. The application of physical science approaches for understandingNatur Reviews Cancer deregulated transport in cancer

This is an illustration of the efforts to develop a broader understanding of the physical barriers and the biological factors that are involved in the progression of tumours and the efforts to design novel biocompatible delivery carriers that can overcome or take advantage of these barriers with favourable pharmacokinetics and tissue distribution profiles for highly efficient delivery of novel therapeutic and imaging agents. A physics- and biology-driven, and a mathematics-based, design of the engineered drug delivery vectors multiplies the probability of recognition of the novel targets, thus providing a synergistic solution for the imaging and therapy of tumours at the interface of physics, engineering, mathematics and cancer biology.