Lumora
A two-step therapy for atrophic acne scars, from a final-year project I ran alone. Scroll to go from the skin surface into a single cell. Drag the model to turn it.
built and tested in vitroA serum goes on first
This is an atrophic scar: the surface has sunk because collagen in the dermis below has been lost, and the ridges between epidermis and dermis have flattened. The serum is acidic (pH 3.69). The idea is that alpha hydroxy acids loosen dead surface skin so what follows can get in.
A hydrogel patch bridges the scar
Chitosan, alginate and marine collagen I extracted from fish skin in my own lab. Patch surface pH is 5.45, much closer to skin.
What went wrong: adding calcium chloride too fast made the gel lumpy. Adding it dropwise fixed it. Ascorbic acid kept degrading, so I cut its oxygen exposure and switched to amber glass.
Fibroblasts build the scaffold
Spindle-shaped fibroblasts sit between woven collagen bundles. Under the scar the bundles are sparse and thin, so nothing holds the skin up.
Collagen is a relay
TGF-β binds a receptor on the cell surface. SMAD proteins carry the message to the nucleus, where the COL1A1 gene is read out. The mRNA is built into collagen chains on the rough ER.
In human keloid, a scarring model, my analysis found these turned up versus normal skin (padj < 0.05, 4 keloid vs 6 normal, GSE158395):
Fold, pack, secrete
P4HA1 and P4HA2 fold the new chains and need vitamin C, which is why ascorbic acid is in the serum. Golgi packs procollagen, vesicles carry it out, and fibrils assemble outside. Asiaticoside, Pal-KTTKS and GHK-Cu are reported in the literature to push fibroblasts toward more collagen. I did not test that.
I docked 8 compounds against wound-healing targets with AutoDock Vina: 11 of 13 pairs docked, -3.59 to -13.68 kcal/mol. Acemannan was too large to dock. Predictions, not measured binding.
What the patch actually did
What this does not show
Everything here is lab-bench and computational. No cells, animals or people were treated. The mouse dataset was too small to compare with the human one, and my regeneration propensity score is still in progress. Wet-lab and in vivo testing would be needed to confirm any of it.
The 3D scenes are schematic illustrations of real anatomy and biochemistry, not microscope images.