Abstract

Background: There is growing
demand for human-based assays to test compounds/therapeutics for adverse immune
reactions prior to entering clinical trials. Human skin equivalent models are
useful in vitro testing platforms. We
describe a unique full-thickness 3D autologous skin equivalent model made from
primary human tissue which is representative of normal human skin and an
autologous platform for testing therapeutics under development.

Methods: Primary human
fibroblasts and keratinocytes were grown from healthy volunteer tissue. The
full-thickness skin equivalent model was generated by first forming a dermal
equivalent by culturing fibroblasts on a scaffold before adding donor-matched keratinocytes
and culturing at the air-liquid interface. Histology and immunofluorescence for
protein markers of epidermal differentation (involucrin and cytokeratin 14) and
for dermal collagen (Picro Sirius red) was completed and compared to normal
skin. We co-cultured the model with activated peripheral blood mononuclear
cells (PBMCs) and observed similar immune damage to that in our skin explant
assay. Immunofluorescence staining for heat shock protein 70 (HSP70) as a
marker of apoptosis was used to confirm these findings. We also used the model
to screen for adverse immune events to known positive and negative control
therapeutic monoclonal antibodies, similar to our skin explant assay, but
replacing the skin explant with the 3D autologous skin equivalent model.

Results: Our model was
representative of normal human skin, showing similar structure observed by
haematoxylin and eosin staining and positive immunofluorescence staining  for protein markers of epidermal
differentiation and skin structure (involucrin, cytokeratin 14 and Picro Sirius
red). Our data also suggest this model could be used to detect adverse immune
events to positive control therapeutic monoclonal antibodies with damage in the
model as shown by positive HSP70 staining. No HSP70 staining was observed in
the 3D skin model treated with negative control monoclonal antibody.

Conclusions: We have generated
an autologous 3D skin equivalent model, representative of normal human skin.
Our data has shown this to be a useful platform for the detection of advserse
immune events when testing biologics under development.