CRYSTAL STRUCTURE OF ADP-RIBOSYL CYCLASE. C. D. Stout, G. Sridhar Prasad. E. A. Stura, D. E. McRee, Department of Molecular Biology, The Scripps Research Institute, La Jolla, CA 92037, D. G. Levitt, H. C. Lee, Department of Physiology, University of Minnesota, Minneapolis, MN 55455

The crystal structure of ADP-ribosyl (ADPR) cyclase reveals a novel dimer in which the deep active site clefts of the monomers face toward the local two-fold axis. The monomers associate in such a way that a solvent filled tunnel connects the active sites. ADPR cyclase catalyses the synthesis of cyclic ADP-ribose (cADPR) from NAD in a reaction that requires displacement of nicotinamide followed by refolding of the nucleotide such that the N1 of adenine is covalently bonded to the C1' carbon of the terminal ribose with retention of configuration (1). The structure implies that the dual nature of the cyclase active sites is critical to carrying out this reaction. Soaking experiments coupled with modeling of difference Fourier maps in progress may define the binding site of the substrate, intermediates or product. These results may infer an enzyme mechanism.

cADPR is emerging as an endogenous regulator of Ca²⁺-induced Ca²⁺ release in cells (2). ADPR cyclase is abundant in Aplysia ovotestes and this source has been used for obtaining crystals (3). The enzyme was discovered in sea urchin eggs and is ubiquitous in tissues of marine invertebrates, amphibians, avians, and mammals, including humans (2). ADPR cyclase exhibits significant sequence homology to CD38, a lymphocyte differentiation antigen, which is a bifunctional ectozyme, also catalyzing the hydrolysis of cADPR.

The ADPR cyclase L-shaped monomer is comprised of a N-terminal helical domain and a C-terminal [[beta]]-sheet containing domain resembling flavodoxin. There are 5 disulfides. The structure, and alignment of ADPR cyclase and CD38 sequences, suggests that the active site resides in the cleft between domains. Key residues for activity appear to be Trp77, Tyr81, His85, Thr96, Glu98, Asp99, GlylO3, TyrlO4, AsnlO7, SerlO8 and Trpl40. The structure was solved using a NCS averaged MIR map based on 6 derivatives. The current R-factor for all data in the range 8.0-2.4Å is 0.22 (Rfree 0.31).

1. H. C. Lee et al., Nat. Struc. Biol. 1, 143 (1994).

2. H. C. Lee et al., Vitamins and Hormones 48, 199 (1994).

3. G. S. Prasad, et al., Proteins 24, 138 (1996).

Supported by NSF grant 95-13421.