Putting Skin in the Game

This “skin on a chip” bioreactor can help researchers study and treat keloid disease and other forms of extreme scarring.

by John Kosowatz
January 16, 2018

Not satisfied with in vitro models for studying skin disorders, researchers at the University of Pennsylvania leveraged advances in tissue engineering and the development of microfluidic chips to mimic human skin structure. The bioreactor they developed will be used as a platform for studying the role of mechanical stretch in keloid disease and scarring of the skin.

Existing models for skin lack its physical components, such as defined vascular networks and mechanical stretch, says Megan Farrell, a bioengineering postdoctoral fellow in the Biologically Inspired Engineering Systems Laboratory (BIOLines) at the University of Pennsylvania. The one she designed, dubbed "skin on a chip," produced multilayered tissue with physiological spatial arrangement that resembles the epidermis and its vascular system. She then mimicked skin stretching by integrating the model with a computer-controlled mechanical actuation system.

“There aren’t many micro-engineered skin models,” says Dongeun Huh, BIOLines principal investigator. “What we were looking to achieve is to make a model more realistic, to resemble in vitro skin more closely.”

In a paper recently published in Stem Cell Reviews and Reports, Dutch researchers reviewed the types and availability of in vitro skin models and microfluidic devices. “Skin disease modeling, substance testing, and ultimately personalized medicine would be enabled by an ideal in vitro 3D skin model containing vasculature, immune cells and appendages,” they wrote. “Until now, full-thickness skin models based on primary cells were most common, even available commercially. Commercial models have limited physiological relevance for risk assessment and testing mode of action of novel actives.”

Farrell’s differs from previous versions because of its multiple layers. The upper and lower channels are separated by a 10-micrometer membrane with a 10-micormeter porosity. The upper channel was seeded with epidermal cells, human adult keratinocytes. Dermal fibroblast cells were cultured in hydrogel in the lower channel. Human umbilical vein endothelial cells were also cultured and added to the dermal, or bottom, layer. Two flanking channels run along the gel channel for access and feeding the cultures.

“It is a fairly sophisticated model, especially when you compare it to what [private] companies sell,” says Huh. “Ours offers the unique capability to culture and create a number of cells. It may be unique because it creates a vascular structure.”

Farrell says it typically takes about three weeks to grow and differentiate skin cells on her chip. Researchers found that after the cells were grown and stabilized, they began to stratify and form the physical properties of human skin. Researchers exposed the upper layer of epithelial cells to air and found they dramatically changed shape and function. “They became stratified and changed their biochemical process,” Huh says.

She was also successful in mimicking skin stretching through the creation and integration of a mechanical bioreactor, which includes a motorized mechanical arm with a custom-designed holder, or gripper, that grabs onto the chip. It moves in lateral directions, basically pulling the entire device, and is controlled through computer code written by Farrell.

The device measures linear strain and was applied in the range of 5 percent to 15 percent, to mimic the mechanical motion and strain of skin tissue. Researchers were able to quantify tissue deformation—cell elongation—in the epidermal and dermal layers using image-based analysis.

Replicating strain at the cellular level is important for the next step, furthering research into keloid disease, a rare and incurable disease where scar tissue grows extensively over a wound. The tissue forms fibrous, hard growths called keloids and are often much larger than the original wound. They are not life-threatening but are often a cosmetic concern, especially if they appear on the face or an area of high visibility. But they can be painful and lesions often develop in areas of high mechanical tension, such as the chest, shoulder and back, suggesting that keloid cells respond differently to mechanical stress, says Huh.

Animal models and cells are commonly used in research, but work on keloid disease is limited because it is found only in humans. There are no animal models on which to base research.

This microfluidic device offers a new platform for gaining knowledge of the disease and its progression, Huh says. For the next step, Farrell is working to obtain keloid cells from patients and grow them in the device. After replicating the mechanoresponsiveness of keloid cells to dynamic stretch, the UPenn team will try to determine if strain-induced effects can be mitigated. That would open up research into treatment of the formation of keloid lesions.

Read more about Medical Devices on AABME.org.