One reason that foreign tissue sparks an attack by the host is that it brings along a specific type of immune cells called antigen presenting cells (APCs), which are also found in the recipient. APCs carry bits of invading pathogens on their surfaces to alert T cells, which then multiply and attack the intruder. A pig's APCs are similar enough to those of other mammals to trigger a response in a transplant recipient. In this case, however, the host turns around and attacks the pig tissue. But there is an Achilles' heel. In order to be activated, a recipient's T cells need to be "costimulated" by a molecule, dubbed CD86, that also sits on the APCs.
A team led by immunologist Robert Lechler of the Imperial College of Science, Technology, and Medicine in London tried to exploit this weakness. The idea was to make the mouse produce antibodies that would destroy the pig CD86. To stimulate these antibodies, the group first produced hybrid molecules whose back half mimicked a string of amino acids from pig CD86, while the front half contained an amino acid stretch from ovalbumin, which coaxes the mouse's immune system into producing antibodies. The researchers immunized eight diabetic mice with the hybrid molecules. Then they took islets from the pancreas of a pig and transplanted them into the rodents. In vaccinated mice, the pig islets kept producing insulin between 21 and 47 days, versus 11 to 20 days for unvaccinated controls. And the more antibodies against CD86 the mice produced, the longer their xenograft survived.
"It's a neat twist," says Stanford immunologist Garrison Fathman. But because all the xenografts were ultimately rejected, he cautions that much more work will be needed before the technique can make it to the clinic.