Live long and prosper: the enterprise of understanding diseased epithelium

Abstract

The epithelium is a particularly complicated and dynamic tissue, and dysregulation of epithelial structure and function is a hallmark of several lung diseases. Motivated by the life and recent passing of Leonard Nimoy, we highlight several recent studies that explore the nuanced relationship between the epithelium and disease progression. Specifically, we focus on recent innovative and integrative approaches that shed new light on epithelial wounding, healing, and development.

Generations of scientists have been motivated to pursue their careers through role models in popular culture. The science fiction genre in particular has inspired curiosity, enthusiasm and passion to understand the world around us, and perhaps the most widely known example of this is in the Star Trek franchise. Beginning in 1966 with The Original Series, the Star Trek universe has showcased the complex relationship between humanity and technology, and has provided a broadly accessible point of entry for the public to reflect upon issues such as space exploration, cloning, climate change, and nanotechnology. Perhaps the most widely known scientist archetype has been developed in the character of Mr Spock, the half-Human, half-Vulcan science officer on the original Enterprise. An exploration of his internal conflict between his ‘Human’ emotions, and ‘Vulcan’ logic is a recurring theme, and has made the character… fascinating. Perhaps it is in emulating this dichotomy that successful scientists can combine the Human artistry of developing creative hypotheses, with the Vulcan pursuit of rigorous logical analysis. It is likely that Leonard Nimoy, the actor who portrayed Mr Spock, has inspired more young people towards science than many practising engineers and scientists. Sadly, Mr Nimoy passed away on February 27, 2015, from complications arising from chronic obstructive pulmonary disease (COPD). The characteristic feature of this disease is a progressive narrowing of the small airways, making it difficult to breathe, and ultimately resulting in a breakdown of lung tissue. As in many lung diseases, it is now well established that COPD pathophysiology is mediated by chronic inflammation and abnormal remodelling of the lung epithelium, which forms the cellular barrier between vascularized lung tissue and breathable air.¹ Understanding the mechanisms that drive pathological epithelial dysfunction in the lungs remains a challenge as the intricate geometry, small scale, and dynamic microenvironment of lung alveoli make it difficult to design and interpret controlled in vivo experiments. Here, we honour Mr Nimoy's indirect, yet substantial contributions to science, by highlighting recent advances in our understanding of the physical mechanisms underlying epithelial pathology, through interdisciplinary studies of epithelial wounding, healing and development.

Localized wounding of the epithelium is emerging as a key factor in initiating and driving lung diseases such as COPD² and idiopathic pulmonary fibrosis (IPF).³ The currently accepted hypothesis is that epithelial wounding drives an epithelial-to-mesenchymal transition in the otherwise-quiescent cells, which then triggers matrix invasion, degradation and other inflammatory processes. Wounding of the epithelium has previously been shown to occur due to rupture of liquid occlusions within the airways which also causes the characteristic respiratory crackling sounds in late-stage lung disease.⁴ However, the causes for wounding during early-stage disease remain poorly defined. Clinical evidence that microaspiration of acidified gastric contents into the airway network is associated with disease progression, suggests that acidic conditions may cause these initial wounds. However, creating localized acidic wounds in model culture systems is challenging. Conventional wounding assays use pipette tips to scratch epithelial monolayers, creating wounds substantially larger than the micron-scale epithelial injuries seen in IPF, and cannot control the chemical nature of the wounding stimuli. These issues were recently addressed by Guenat and colleagues in a recent issue of Integrative Biology, who built upon previous microfluidic wound-healing assays by incorporating a novel aqueous two-phase system to conduct acidic wounding studies.³ Laminar microfluidic flows limit mixing across streams, and an injuring liquid (such as enzymatic trypsin) can be contained within a narrow stream to create microwounds in cultured epithelia.⁵ However, it is not possible to maintain an acidic pH by relying solely on laminar flow, as the small acidic H⁺ ions diffuse rapidly across streams. Instead, a thermodynamically separated aqueous two-phase system consisting of an acidic pepsin-containing polyethylene glycol phase and a dextran phase was used to maintain an acidified wounding flow stream. Through precise control of injection rates, microwounds ranging from 40 μm to 200 μm in size were made in lung epithelia, and the authors found that the combination of pepsin and HCl leads to microinjuries that were morphologically different from those created by trypsin–EDTA, and histologically similar to those seen in IPF (Fig. 1). This study provides evidence that acid-based microwounding is distinct from regular wounding, indicating that the mechanism of injury in diseased environments can have subtle but significant effects on biology. More broadly, this technological approach provides the key infrastructure necessary to begin studies of physiologically relevant epithelial wounding mechanisms.

Fig. 1 (a) Schematic and (b) micrographs of parallel injuries made by hydrodynamic flow-focusing of trypsin–EDTA within a monolayer of alveolar cells. Alveolar monolayers exposed to (c) pepsin–HCl, and (d) trypsin–EDTA, reveal clear morphological differences between the two wounding agents. Figure adapted with permission from Felder et al.³

While this two-phase microfluidic wounding assay focuses on wounds at the multicellular scale, a recent discovery by Casares et al. in Nature Materials indicates that epithelial wounding can also occur at the sub-cellular level for cells cultured on physiologically-relevant porous hydrogel matrices.⁶ Lung epithelial sheets function under challenging mechanical conditions involving repeated cyclic stretch. At high strains, cracks can form in the epithelium, presumably when epithelial tension reaches the stress fracture limit.⁷ In a series of well-designed experiments, Casares et al. demonstrate that these fractures are not caused by elevated elastic strains, but instead by transient hydraulic pressures within the hydrogel substrates. Epithelial islands expressing fluorescently-labelled filamentous actin were micropatterned onto hydrogels attached to stretchable elastomeric membranes (Fig. 2). This culture system was placed under homogeneous biaxial stretch, and cracks characterized by discontinuities in cortical actin formed only when substrate tension was released. Monolayer traction force microscopy confirmed that fissures formed when cellular tensions were lowest, suggesting that stretch-induced epithelial crack formation is not a result of increased tension in the epithelial sheet. Instead, epithelial cracks are associated with flux of the liquid component through the deforming porous hydrogel, and epithelial islands attached to the hydrogels form a boundary that is impermeable to rapid liquid flow. This creates a hydraulic pressure difference that is greatest immediately after releasing a strained hydrogel. Further experiments demonstrate significantly more cracks form on stiff gels than on soft ones, and that crack healing occurs more slowly on thicker gels than on thin ones, consistent with the idea that transient fluid pressures rather than structural deformation can cause epithelial damage. Considering that COPD is characterized by increased thickness and stiffness of the alveolar environment, it now seems likely that in vivo hydraulic pressures are creating an environment in which the epithelium is predisposed to cracking and to undergoing sustained and repeated injury. This new understanding of epithelial fracture mechanics provides a novel way to consider COPD disease progression and demonstrates the importance of using realistic hydrogel-based culture systems, rather than tissue culture plastic or even silicone rubbers.

Fig. 2 (a) Schematic of epithelial cell stretching device. (b) LifeAct-GFP-expressing epithelial cell islands before, during, and after application of biaxial stretch. Arrow heads point out the epithelial cracks that have only appeared after stretching has stopped. Figure adapted with permission from Casares et al.⁶

While these studies show that the dynamics of wounding are both subtle and non-intuitive, an equally enigmatic aspect of lung disease progression is in the dysregulation of epithelial healing mechanisms. Studying wound healing mechanisms is essential, because while these processes are essential to maintain healthy epithelial homeostasis, dysregulated healing is associated with disease progression. Currently, there are two generally-accepted and distinct mechanisms that work together to close a wound: (1) actin accumulation around the wound edge creates a contractile actin purse-string that closes wounds; and (2) cells migrate by extending lamellipodia into the wound gap. Both these mechanisms are active in wound healing processes, and decoupling their individual component contributions can be challenging. Furthermore, there are many in vivo situations in which healing occurs over gaps with reduced or absent ECM adhesion proteins, suggesting that purse-string healing may be even more important than previously considered. An extremely simple yet powerful microfabricated approach by Ladoux and colleagues was recently described in Nature Communications to analyze purse-string healing in isolation of lamellipodial forces.⁸ Reasoning that lamellipodial dynamics relies on cell–substrate interactions, they eliminated these interactions by preventing cell adhesion at defined locations. Epithelial cells were micropatterned onto fibronectin-coated surfaces that enclose a circular non-adhesive region. Suspended epithelial bridges formed across these non-adhesive gaps, mediated by the formation and contraction of a supracellular actin ring around the wound edge, which provided sufficient tension to anisotropically close gaps up to 100 μm in diameter (Fig. 3). Epithelial cells derive a large portion of the necessary tension to close wounds through cell–substrate interactions at the leading edge. In the case of non-adherent wound closure, these interactions are not possible, indicating that the requisite tension must come from another source. Monolayer traction force microscopy and live-cell microscopy revealed that actin contractility was reinforced by the adherent cells around the gap through the reorganization of actin cables directed tangentially from the wound edge. This suggests that this tissue-wide reorganization of the cytoskeleton plays a crucial role for providing sufficient force for wound closure. This idea is further supported by Trepat and colleagues in Nature Physics who used a combination of laser beam ablation and monolayer traction force microscopy to map the forces required for adherent wound healing.⁹ They modeled an actin ring structure that also transmits its contractile tension to the underlying substrate tangentially to the wound edge. Although actin relocalization and contraction play important roles for both adherent and non-adherent wound closure, non-adherent gap closure occurs more slowly than adherent wound healing. Since characteristic features of lung disease include a discontinuous alveolar epithelium, these healing processes may play a significant role in mitigating disease progression.

Fig. 3 (a) Schematic of non-adherent gap wound closure. Z-projection confocal images in the xy and yz planes of epithelial cell clusters stained for filamentous actin (green) patterned onto fibronectin (red) during (b) early, (c) intermediate, (d) late stages of wound closure. Figure adapted with permission from Vedula et al.⁸

Whereas disease processes destroy epithelial structure and function, developmental processes work to establish these relationships from biological matter that lacks inherent structure and function. Perhaps greater insight into these wounding and healing mechanisms can be obtained by studying developmental programs involved in the formation of functional epithelia. Epithelial morphogenesis can be challenging to study, as the resulting tissue structures are dynamic and spatially complex, and undergo elongation, budding, bending, branching, involution and invagination. In particular, the formation of segregating boundaries between distinct cell types is necessary for the ultimate compartmentalization of different tissue types, and hence must play a critical role in the healthy formation of alveolar sac structures. Furthermore, these boundaries are hypothesized to form secondary signalling centers that drive morphogenetic pattern formation at the local level, making boundary formation crucial to the successful development of any organism.¹⁰ However, the mechanisms that maintain and guide boundary formation remain unclear, mainly due to the lack of in vitro platforms capable of studying these dynamic tissue processes. In a recent issue of Integrative Biology, McGuigan and co-workers present an in vitro model that allows the formation and manipulation of stable boundary layers between two distinct cell types, in a controlled in vitro culture system.¹¹ Retinal cells from the ectoderm and bronchiole epithelial cells from the mesoderm were co-cultured upon micropatterned adhesive substrates, and boundary formation was analyzed over several days. The patterned boundaries faithfully recapitulated many in vivo boundary parameters, such as a smooth boundary network and enriched actomyosin bundles aligned parallel to the interface. Using wounding scratch assays they discovered that migration, actin cytoskeletal polarization, and modulation of tensile forces are critical to create and maintain stable boundary formation between distinct cell types. It is particularly interesting to note that these same mechanisms that work seamlessly to maintain a functional boundary in a developmental environment are likely disrupted in a diseased environment and possibly contribute to the aberrant alveolar modifications in lung disease. Further studies using this in vitro system can be aimed at understanding the mechanical and chemical changes that derail normal epithelial boundary formation and can provide fundamental insight as to how diseased environments might impact epithelial biology.

In parallel with these integrative approaches, more conventional analyses of protein signalling mechanisms have recently been used to analyze multicellular collective migration. Understanding the regulation of coordinated collective cell migration is crucial, as it is a necessary component of both normal developmental processes and healthy wound healing. A recent study by Das et al. in Nature Cell Biology identified the cytoskeletal protein Merlin as a mechanochemical transducer that coordinates collective sheet migration by relocalizing from intercellular junctions to the cytoplasm through tensile modulation.¹² These findings suggest a novel target mechanism for epithelial dynamics in disease progression, and demonstrate the importance of integrating fundamental molecular biology approaches with advanced integrative technologies to understand the mechanisms and consequences of a diseased environment on epithelial regulation in lung disease.

It is clear that the relationship between epithelium and its environment is complex, nuanced and challenging to simulate in vitro. The need for innovative and integrative strategies that bridge the fields of mechanical, chemical, and materials engineering with fundamental studies of cell biology are necessary to understand epithelial-related disease processes. Similar to the premise of the Star Trek franchise, the need for exploration of these relatively uncharted regions is evident. And just as Leonard Nimoy once did in his field, this particular set of problems is ripe for creative yet rigorous approaches to boldly go where no one has gone before.

References

1. J. C. Hogg and W. Timens, Annu. Rev. Pathol.: Mech. Dis., 2009, 4, 435–459.
2. J. M. Perotin, D. Adam, J. Vella-Boucaud, G. Delepine, S. Sandu, A. C. Jonvel, A. Prevost, G. Berthiot, C. Pison, F. Lebargy, P. Birembaut, C. Coraux and G. Deslee, Respir. Res., 2014, 15, 151.
3. M. Felder, A. O. Stucki, J. D. Stucki, T. Geiser and O. T. Guenat, Integr. Biol., 2014, 6, 1132–1140.
4. D. Huh, H. Fujioka, Y. C. Tung, N. Futai, R. Paine, J. B. Grotberg and S. Takayama, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 18886–18891.
5. M. Murrell, R. Kamm and P. Matsudaira, PLoS One, 2011, 6, e24283.
6. L. Casares, R. Vincent, D. Zalvidea, N. Campillo, D. Navajas, M. Arroyo and X. Trepat, Nat. Mater., 2015, 14, 343–351.
7. A. R. Harris, L. Peter, J. Bellis, B. Baum, A. J. Kabla and G. T. Charras, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 16449–16454.
8. S. R. Vedula, G. Peyret, I. Cheddadi, T. Chen, A. Brugues, H. Hirata, H. Lopez-Menendez, Y. Toyama, L. N. de Almeida, X. Trepat, C. T. Lim and B. Ladoux, Nat. Commun., 2015, 6, 6111.
9. A. Brugues, E. Anon, V. Conte, J. H. Veldhuis, M. Gupta, J. Colombelli, J. J. Munoz, G. W. Brodland, B. Ladoux and X. Trepat, Nat. Phys., 2014, 10, 683–690.
10. C. Dahmann, A. C. Oates and M. Brand, Nat. Rev. Genet., 2011, 12, 43–55.
11. S. Javaherian, E. D'Arcangelo, B. Slater, T. Zulueta-Coarasa, R. Fernandez-Gonzalez and A. P. McGuigan, Integr. Biol., 2015, 7, 298–312.
12. T. Das, K. Safferling, S. Rausch, N. Grabe, H. Boehm and J. P. Spatz, Nat. Cell Biol., 2015, 17, 276–287.

---