Toni Kline

A beta-glucuronide-based linker for attaching cytotoxic agents to monoclonal antibodies (mAbs) was designed and evaluated. We employed the cytotoxic auristatin derivatives MMAE (1a) and MMAF (1b) and doxorubicin propyloxazoline (DPO, 2) to give the beta-glucuronide drug-linkers 9a, 9b, and 17, respectively. Cysteine-quenched derivatives of 9b and 17 were determined to be substrates for E. coli beta-glucuronidase, resulting in facile drug release. The beta-glucuronide MMAF drug-linker 9b was highly stable in rat plasma with an extrapolated half-life of 81 days. Each drug-linker when conjugated to mAbs c1F6 (anti-CD70) and cAC10 (anti-CD30) gave monomeric antibody-drug conjugates (ADCs) with as many as eight drugs per mAb and had high levels of immunologically specific cytotoxic activity on cancer cell lines. cAC10-9a displayed pronounced antitumor activity in a subcutaneous Karpas 299 lymphoma tumor model. A single dose treatment led to cures in all animals at the 0.5 mg/kg dose level and above, and the conjugate was well tolerated at 100 mg/kg. In mice with subcutaneous renal cell carcinoma xenografts, the MMAF conjugate c1F6-9b was tolerated at 25 mg/kg and efficacious at 0.75 mg/kg. These results demonstrate that the beta-glucuronide linker system is an effective strategy for targeting cytotoxic agents providing ADCs with high degrees of efficacy at well-tolerated doses.

Deacetylation of uridyldiphospho-3-O-(R-hydroxydecanoyl)-N-acetylglucosamine by LpxC is the first committed step in the Pseudomonas aeruginosa biosynthetic pathway to lipid A; homologous enzymes are found widely among Gram-negative bacteria. As an essential enzyme for which no inhibitors have yet been reported, the P. aeruginosa LpxC represents a highly attractive target for a novel antibacterial drug. We synthesized several focused small-molecule libraries, each composed of a variable aromatic ring, one of four heterocyclic/spacer moieties, and a hydroxamic acid and evaluated the LpxC inhibition of these compounds against purified P. aeruginosa enzyme. To ensure that the in vitro assay would be as physiologically relevant as possible, we synthesized a tritiated form of the specific P. aeruginosa glycolipid substrate and measured directly the enzymatically released acetate. Several of our novel compounds, predominantly those having fluorinated substituents on the aromatic ring and an oxazoline as the heterocyclic moiety, demonstrated in vitro IC(50) values less than 1 microM. We now report the synthesis and in vitro evaluation of these P. aeruginosa LpxC inhibitors.

LpxC [UDP-3-O-(R-3-hydroxymyristoyl)-GlcNAc deacetylase] is a metalloamidase that catalyzes the first committed step in the biosynthesis of the lipid A component of lipopolysaccharide. A previous study (H. R. Onishi, B. A. Pelak, L. S. Gerckens, L. L. Silver, F. M. Kahan, M. H. Chen, A. A. Patchett, S. M. Galloway, S. A. Hyland, M. S. Anderson, and C. R. H. Raetz, Science 274:980-982, 1996) identified a series of synthetic LpxC-inhibitory molecules that were bactericidal for Escherichia coli. These molecules did not inhibit the growth of Pseudomonas aeruginosa and were therefore not developed further as antibacterial drugs. The inactivity of the LpxC inhibitors for P. aeruginosa raised the possibility that LpxC activity might not be essential for all gram-negative bacteria. By placing the lpxC gene of P. aeruginosa under tight control of an arabinose-inducible promoter, we demonstrated the essentiality of LpxC activity for P. aeruginosa. It was found that compound L-161,240, the most potent inhibitor from the previous study, was active against a P. aeruginosa construct in which the endogenous lpxC gene was inactivated and in which LpxC activity was supplied by the lpxC gene from E. coli. Conversely, an E. coli construct in which growth was dependent on the P. aeruginosa lpxC gene was resistant to the compound. The differential activities of L-161,240 against the two bacterial species are thus the result primarily of greater potency toward the E. coli enzyme rather than of differences in the intrinsic resistance of the bacteria toward antibacterial compounds due to permeability or efflux. These data validate P. aeruginosa LpxC as a target for novel antibiotic drugs and should help direct the design of inhibitors against clinically important gram-negative bacteria.