A coating copied from the mammalian egg kept transplanted islets working for 100 days in mice

The gel forms on the cell rather than around it. Instead of loading islets into a preformed bead, the Penn State group emulated the chemistry that hardens a fertilised egg's zona pellucida, calcium-driven cleavage and crosslinking of surface glycoproteins, so the shell assembles in place ^[12][3]. It does so in aqueous solution under physiological conditions, which is the basis for the team's claim of no loss of viability or function and 100% encapsulation efficiency ^[11][9]. Principal investigator Yong Wang and colleagues spent eight years getting a layer to lie against a curved cell surface without disturbing it ^[10].

The volume claim is checkable, and it checks out. Polymer in a 500-micrometre microcapsule around a 100-micrometre islet occupies the shell between the two radii: 250 cubed minus 50 cubed, about 15.5 million cubic micrometres. A 20-micrometre coat on the same islet gives 70 cubed minus 50 cubed, about 218,000. The ratio is 71, which is the figure the researchers report ^[4][14]. In thickness the coat is 75 times slimmer than the 1.5 mm capsules most of the field works with, and the reported protection is equivalent ^[1][8][13].

Thinness is not only about packing. The stated point of a semi-permeable capsule is to admit oxygen and nutrients while excluding immune cells ^[21], so a shorter diffusion path is a mechanistic reason to care, separate from the argument that less polymer means more islets fit in the implant site ^[15] and possibly a safety margin ^[16].

Now the denominators. Immunocompetent diabetic mice given BZP-U-coated islets were back to healthy glucose within a week, and the majority held past 100 days without systemic immunosuppression, against uncoated controls that could not stay in range beyond about seven days, roughly a fourteenfold difference in useful graft life ^[19][20][3]. The word doing the work is "majority": the account does not give group sizes, mouse strain, how diabetes was induced, how closely donor and recipient were matched, or any transport measurement ^[22]. The immune tests described are a foreign body response assay using polystyrene microparticles in the peritoneum, where coated beads provoked less of a response than bare ones and ultrapure alginate beat regular alginate ^[17][18].

That is a real result about tolerance of the material and function of the graft. What it cannot address is how the same coat behaves against a human immune repertoire over years. Keep the human number beside it: under the Edmonton protocol introduced in 2000 at the University of Alberta, 24% of recipients were insulin independent at 28 months ^[6][7], meaning 76% were not ^[2], and all of them carried the drug regimen that this coating is meant to remove ^[1]. A hundred days in a mouse and 28 months in a person are not the same clock, and nobody should convert one into the other.

Islet transplantation has been attempted in patients since 1974 ^[5], and the field's harder constraint has usually been what the recipient must take, not what the islet does. On that framing, the claim I would most want to see replicated first is the manufacturing one: a shell that generates itself on every cell in solution, with no functional loss, is the step that has to survive the jump from a mouse's islet mass to a person's ^[9][11][15].