Effects of the mechanical properties of collagen gel on the in vitro formation of microvessel networks by endothelial cells
Nahoko Yamamura, M.S., Ryo Sudo, Ph.D., Mariko Ikeda, Ph.D., and Kazuo Tanishita, Ph.D.
Overview:
In this experiment, the effects of the mechanical properties of collagen gel on the formation of 3D capillary networks were studied. Bovine pulmonary microvascular ECs (BPMECs) were cultured in collagen gels along with an angiogenic factor (added 24 hours after seeding) to promote growth of the networks. The gel stiffness was varied by using several different pHs of the collagen polymerization solution, keeping the collagen concentration constant. Within pH values between 5 and 8, the relaxation moduli of collagen gel increases linearly with pH (i.e. “flexible gels” polymerized at pH 5; “rigis gels” polymerized at pH 9). Microscope imaging showed that the formation of 3D microvessel networks differed based on the stiffness or flexibility of the gel in which the tissue was cultured. Also, vinculin expression varied along with the gel’s mechanical properties, indicating that microvessel network morphogenesis involves a vinculin-mediated interaction between the ECs and ECM substrate.
Methods:
In order to study network growth, the cultures were photographed under a phase-contrast microscope every 10 minutes for 2 days. To study network configuration, the cells were fluorescently labeled with CellTracker Green BODIPY (after 7 days of culturing) and observed using confocal laser scanning microscopy. Imaging software was used to calculate the area, length, and density of the EC networks. To examine thin vertical sections of the network, they were stained and transmission electron microscopy was used. And finally, to study actin and vinculin expression, the cells were incubated overnight with mouse anti-human vinculin antibody, anti-mouse IgG antibody, and phalloidin, and then observed using a confocal laser scanning microscope. Successive vertical section images were reconstructed to form a 3D image.
Results:
ECs in the flexible gel grew by elongating into thin vessels and interlinking with other networks. Cells invaded the gel individually and formed few clumps, creating thin and dense netrworks. In the rigid gel, however, ECs formed aggregates, which grew while filamentous cytoplasmic processes extended out and formed branched networks. Thicker, more sparsely distributed clumps formed in the rigid gel (in addition to forming deeper networks). In general, network area decreased with depth in all gels, but networks at lower depths seemed to grow best in the rigid gel. Also, intracellular vacuoles were observed in networks in the flexible gel, while luminal structures were observed in the rigid gel. In the flexible gel, faint expression of vinculin was observed, whereas intensive actin filaments and numerous clumps of vinculin were observed in the rigid gel.
In conclusion, the formation, configuration, and distribution of microvessel networks depend on the gel’s mechanical properties, which has an effect on EC morphogenesis by regulating the proliferation, migration, and differentiation of ECs.
Importance:
This experiment is important because while 2D tissues (such as skin) can be reconstructed, 3D tissues (such as liver tissue) are still hard to grow and manipulate. 3D tissue organoids can be grown in vitro, but they must be vascularized. In order to do this, the interaction between endothelial cells (ECs) and extra cellular matrix (ECM) must be better understood. This experiment gives us further insight into the relationship between ECM mechanical properties and EC proliferation, migration, and differentiation.
It is also important because it shows that the mechanical properties of the adhesion substrate must be considered, as they have an effect on microvessel network configuration. Focal adhesion proteins can be used as mechanosensors, because they respond differently to different gel stiffnesses.
8 comments:
I like this article since it integrates our lab techniques into an experiment with which we could work.
The difficulty in 3D culture is an appealing challenge and is something Napolitano et al. tried to reduce(http://www.liebertonline.com/doi/pdf/10.1089/ten.2006.0190).
Testing mechanical properties at the cellular level is complicated, as the same mechanical properties may alter dramatically given larger cell clusters/tissue formations.
My project group is thinking of ways to test mechanical properties of our cell cultures, but keep running into the obstacle of realizing there are not many known benchmarks to which we can even compare our measurements (given the assumption that small clusters of cells will have different mechanical properties than the associated tissue). Well, this article is helpful.
Could the work on microvessels in this paper be applied to studying arteries (just wondering)?
This is interesting, but using just one ECM protein (collagen) may not be representative of the entire ECM environment. Perhaps the authors could create other types of gels with varied ECM compositions (but that'll probably be difficult).
This is an interesting article in that it definitely shows how significant the mechanical properties of the gel that simulates the EM can be, considering the fact that manipulating the Young's modulus of EM easily affects the differentiation of mesenchymal stem cells. I assume that this article further supports the fact that we can alter the stiffness or flexibility of the gel to regulate the proliferation, migration, and differentiation of ECs, giving us a better control for implanting tissues.
To angelee:
are you talking about the mechanical properties of the collagen gel or of the culture itself? Another thing that was considered that I did not mention is the diameter of the collagen fibrils at different pH's. And yes, the properties would change with larger tissue formations, but this paper is only looking at microvessels, so only small clusters are considered anyways.
To rustin:
I don't know how applicable this study is to arteries. The purpose was to investigate the vascularization of 3D tissue engineered organoids, which is a problem (it has already been done for 2D in vivo). The study would be more applicable to liver tissue engineering rather than arteries.
To lizhi:
I agree, collagen is not entirely representative of the ECM environment. It seems like the difficulties would lie in creating gels with desired mechanical and biochemical properties.
To ben:
Yes, I think this paper does support the idea of altering stiffness in order to change cellular properties. It also gives us an idea of how to vary mechanical properties of the gel in order to achieve a specific Young's modulus. It is definitely a step in solving the challenge of engineering 3D tissues that can be effectively vascularized such that they may be implanted into the body.
Imagine the cells were cultured on a substrate consisting of a mechanical stiffness gradient from soft to stiff. How do you think this kind of multi-stiffness substrate would affect the formation, configuration, and distribution of resulting microvessel networks?
Were the experiments done in this paper purely investigational or are there actual applications of this technique? Does the collagen in the body have different mechanical properties throughout?
Post a Comment