4.0000
#23-OR
HLAMATCHMAKER AND THE RECOGNITION OF ALLOEPITOPES ON HLA MOLECULES.
Rene J. Duquesnoy, PhD., Medhat Askar, MD PhD, Alin Girnita, MD, Adriana Zeevi, PhD and Marilyn Marrari, BA CHT. University of Pittsburgh Medical Center Transplantation Pathology.

---

HLAMatchmaker is a computer algorithm to determine HLA compatibility at the structural level. It uses linear sequences of three amino acids (triplets) as the basis for epitopes recognized by alloantibodies. This program is especially useful for identifying suitable donors for allosensitized patients. We have reexamined the structure of alloepitopes with the Entrez Molecular Modeling Database of the National Center for Biotechnology Information (NCBI). Their Cn3D program provides detailed three-dimensional structural information for eight class I HLA antigens that have been analyzed by crystallography. We will show which polymorphic amino acid residues are exposed on the molecular cell surface and therefore accessible to alloantibodies and which residues are hidden. The α1 and α2 domains of the HLA class I molecule have six “clusters” of polymorphic residues and we have identified triplets with nonlinear sequences within each cluster. The α3 domains contain four clusters of polymorphism and all consisted of linear triplets. This new information has led to a refinement of the HLAMatchmaker algorithm and a better determination of the role of triplets in the immunogenicity (ability to induce antibodies) and the antigenicity (ability to react with antibodies) of alloepitopes. Evidence will be presented that alloepitope-antibody interactions require a second contact site on the HLA molecule. This modified version of HLAMatchmaker will permit a better structural assessment of HLA compatibility.
(Supported by NIH grant AI-55933)