Date of Award

Spring 5-15-2015

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Immunology)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Osteoporosis is a skeletal disease characterized by bone resorption in excess of bone formation. The osteoclast is the sole bone resorbing cell and therefore provides an important avenue for novel therapeutic development. The mature osteoclast is derived from the monocyte/macrophage lineage, and its proper differentiation and function requires the interaction of receptor activator of NF-kB Ligand (RANKL) with its receptor RANK. RANKL exists as a homotrimer, with classical models of receptor activation assuming trimeric clustering of receptor as necessary for signal transduction. We have designed a novel single-chain RANKL (scRANKL) construct to test the assumption that the recruitment of three RANK receptors is necessary for downstream signaling. scRANKL consists of all three RANKL monomers connected by short, protease-resistant linkers, enabling individual manipulation of each RANK binding site and assessment of the receptor clustering state necessary for downstream signaling. We used our co-crystal structure of RANK/RANKL to design mutations in RANKL that abolish binding to RANK. After inserting these variants at one scRANKL site, to create a single-blocked ligand capable of recruiting only two RANK receptors, we still observe RANK signaling and osteoclast formation, although only at high ligand concentrations. In contrast, double-blocked scRANKL does not form osteoclasts. Hence, dimeric engagement of RANK by RANKL is sufficient and necessary for downstream signaling. Although double-blocked scRANKL may potentially antagonize RANK/RANKL signaling, it competes poorly with endogenous RANKL due to loss of receptor avidity. Therefore, we identified mutations in RANKL, using yeast surface display, that dramatically increase its affinity for RANK. Concurrently, we sorted for loss of binding to OPG, the RANKL endogenous decoy receptor, with the hypothesis that this trait would allow scRANKL to more efficiently function as a competitive antagonist, in vivo. When inserted at the intact RANK binding sites in single- or double-blocked scRANKL, these novel constructs are highly efficient inhibitors of wild-type RANKL-induced osteoclastogenesis. Our work demonstrates that the manipulation of RANK receptor oligomerization, and the subsequent blockade in signaling, may provide a novel avenue to develop anti-resorptive drugs. Additionally, the design of single-chain TNFSF ligands with individually manipulated receptor binding sites can more broadly inform our knowledge of how these receptors, with diverse biological functions, initiate signal transduction.


English (en)

Chair and Committee

Steven L Teitelbaum

Committee Members

Marco Colonna, Roberta Faccio, Daved Fremont, Philip Stahl, Wojciech Swat


Permanent URL:

Included in

Biology Commons