News & Views | Published:

Tissue engineering

Blood vessels on a chip

Nature volume 488, pages 465466 (23 August 2012) | Download Citation

To understand how blood vessels form and function, scientists require reproducible systems that mimic living tissues. An innovative approach based on microfabricated vessels provides a key step towards this goal.

The establishment of blood-vessel networks is a matter of life and death for tissues and organisms. Failure to form a functional vascular network causes early death of embryos, and dysfunction of endothelial cells (which line the inside of blood vessels) contributes to many diseases, including stroke, thrombosis and atherosclerosis1. Therefore, substantial resources are being directed towards research into the cellular, molecular and physical factors that regulate the formation, stability and functional responses of the vasculature. Writing in Proceedings of the National Academy of Sciences, Zheng et al.2 describe how they created living, three-dimensional microvascular networks that recapitulate some of the functions that are typical of blood vessels.

Animal models take centre stage in the study of many aspects of blood vessels, such as genetic regulation, developmental patterning and diseases. But in vitro culture of tissues and vascular cells provides the basis for our understanding of other aspects, such as endothelial cell biology, cell-shape regulation and the vessels' responses to physical forces. Linking in vivo with in vitro research has not always been successful and represents a major challenge in the field. In particular, most in vitro models lack the three-dimensional complexity, blood flow, cell–cell interactions or proper extracellular (matrix) environment that are typical of living tissues.

Microfluidic devices, consisting of submillimetre-scale channels or tunnels through which liquid flows in a controlled manner, are an attractive alternative for studying blood-vessel development and function. Scientists can build precast three-dimensional micronetworks using various polymers, and can seed endothelial cells that form tubular microvessels in which blood flow can be mimicked and regulated. However, previous attempts3 — mostly using simple silicone polymers such as polydimethylsiloxane — have failed to implement, and therefore to address, crucial steps of angiogenesis4 (formation of new blood vessels). Moreover, research using microfluidic devices has generally concentrated on studying the influence of variations in flow on the function and permeability of endothelial cells, or on the interactions between these cells and blood cells.

Using silicone moulds together with casting gels made out of collagen, Zheng and co-workers created microvessels that replicated some aspects of angiogenesis (Fig. 1). Collagen gels have been used before for this purpose5, but the resulting devices never delivered the level of endothelial functionality that the authors' system has now achieved. The researchers report that the microvessels were beautifully lined with continuous endothelium and did not leak. Moreover, the vessels — when activated with appropriate biochemical signals — produced new branches in a process known as sprouting angiogenesis, and recruited pre-seeded mural cells, which normally associate with blood vessels and affect their functions. As such, the device provides the important advance of not only allowing the researchers to shape the network, but also permitting the biological elements (cells and their products) to reshape or remodel the system dynamically.

Figure 1: In vitro microvessels.
Figure 1

a, Zheng et al.2 used a temporary silicone scaffold (not shown) and a collagen gel to create a miniaturized tubular network that resembled blood vessels. In this system, some of the cell types that are normally associated with blood vessels (such as mural cells) can be incorporated in the gel, whereas others (endothelial cells) can be added in the culture medium that flows through the channels. b, The authors replicated in their device several processes commonly associated with blood vessels, such as the formation of branches (sprouting) and the interaction of endothelial cells with mural cells. (Modified from ref. 2.)

As an example of the possible uses of their in vitro microvessels, the authors used whole blood to analyse platelet aggregation, a process that, when uncontrolled, can lead to the formation of a blood clot that interrupts the flow in a vessel (thrombosis). In agreement with an earlier study6, they found that adding a proinflammatory compound to the in vitro vessels induced changes in the endothelium ('activation') that promoted platelet aggregation and formation of blood clots. The mechanism involved two proteins, protein kinase C and von Willebrand factor. Protein kinase C stimulated endothelial cells to secrete von Willebrand factor, which then formed fibres on the cells' surface. The fibres promoted platelet adhesion and accumulation, leading to vessel occlusion. These simple results, although preliminary, highlight the potential for microfluidic devices to improve our knowledge of fundamental aspects of blood-vessel physiology.

However, the remodelling of the microvascular network demonstrated by Zheng et al. is still rudimentary, and does not fulfil all of the requirements for the replacement of in vivo models. Sprouting in their system, and the additional remodelling of the vascular structures that occurs afterwards, remain unsatisfactory, and improvements in culture conditions will be required to emulate the dynamic complexity seen in vivo. Yet the authors' report demonstrates the possibility of creating an open and flexible assay in which blood, plasma or culture medium flows through vessels that can be stimulated and manipulated, and that allows real-time observation of the dynamics of endothelial-cell behaviour through the different phases of angiogenesis. We envisage that improved versions of Zheng and colleagues' device will permit quick and accurate manipulation of plasma flow and composition, as well as the delivery of experimental molecules to screen their effects on endothelial function.

Today, the full pattern of vascular remodelling is still beyond the reach of existing microfluidic devices. Moreover, achieving an adequate flow in complex vascular networks in vitro remains problematic; most current systems rely on gravity-driven flow, which provides insufficient shear forces to mimic in vivo conditions. Looking ahead, from an experimentalist's point of view, the ideal microfluidic device should mimic conditions in living organisms and be easy to assemble. It should also enable the tight control and manipulation of all parameters relevant to angiogenesis, be scalable to high-throughput analysis, and facilitate the application of standard methods of molecular analysis and live imaging.

Developments such as the one reported by Zheng and colleagues are closing the technological gaps, and we foresee a strengthening of microfluidic devices as platforms to investigate blood-vessel function and the interaction of blood vessels with tumours. Moreover, the establishment of a functional, hierarchically branched vascular network in vitro that could support the formation of new tissues from stem cells would be an extraordinary achievement that would boost the field of tissue engineering. Time will tell whether the vessel remodelling observed in the authors' microfluidic chamber can be driven far enough to develop systems conducive to large-scale tissue engineering for transplantation.

References

  1. 1.

    Nature 438, 932–936 (2005).

  2. 2.

    et al. Proc. Natl Acad. Sci. USA 109, 9342–9347 (2012).

  3. 3.

    , , , & J. Biomed. Biotechnol. 2009, 8231–8248 (2009).

  4. 4.

    , & Cell 146, 873–887 (2011).

  5. 5.

    et al. Biomaterials 31, 6182–6189 (2010).

  6. 6.

    et al. J. Clin. Invest. 122, 408–418 (2012).

Download references

Author information

Affiliations

  1. Claudio Franco and Holger Gerhardt are at the Vascular Biology Laboratory, London Research Institute, Cancer Research UK, Lincoln's Inn Fields Laboratories, London WC2A 3LY, UK.

    • Claudio Franco
  2. H.G. is also at the Vascular Patterning Laboratory, Vesalius Research Center, Leuven, Belgium.

    • Claudio Franco
    •  & Holger Gerhardt

Authors

  1. Search for Claudio Franco in:

  2. Search for Holger Gerhardt in:

Corresponding author

Correspondence to Holger Gerhardt.

About this article

Publication history

Published

DOI

https://doi.org/10.1038/488465a

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing