LBNL Masthead A-Z IndexBerkeley Lab mastheadU.S. Department of Energy logoPhone BookJobsSearch
A Key Enzyme to the Potency of an Anticancer Agent Print

Incorporation of halogen atoms into drug molecules often increases biological activity. This is the case with salinosporamide A (sal A), a natural product from the marine bacterium Salinispora tropica that is 500 times more active than sal B, its nonchlorinated analog. Sal A is in phase I human clinical trials for the treatment of multiple myeloma and solid tumors. A group of researchers, using diffraction data collected at ALS Beamline 8.2.2, discovered and characterized the chlorinating SalL enzyme, a crucial component in sal A biosynthesis that uses a unique chlorine-activating mechanism.

Chlorine-Loving Enzyme Speeds Things Up

The smell of chlorine is unmistakable. It's in the swimming pool, in household bleach, and sometimes the faint odor of chlorine rises up from your tap water. But chlorine is more than a purifier and a disinfectant. It's found almost everywhere in nature—in table salt, dissolved in the ocean, and as a key ingredient in biosynthesis, a vital part of the metabolism of all living organisms.

A biosynthesis product from Salinispora tropica, a marine bacterium, has been proving in clinical trials to be a highly effective cancer treatment. But researchers didn't have a complete picture of how the bacterium makes this compound. They knew it had to do with the incorporation of chlorine atoms and that chlorine is especially important for the compound’s effectiveness. Then they discovered a new chlorinase enzyme, SalL, in the bacterium's biosynthesis pathway. This enzyme doesn't activate chlorine through the expected oxidation routes. SalL's method is a substitution strategy that uses natural nonoxidized chlorine, like the kind you find in common table salt. This discovery will provide a new "road map" for furthering S. tropica’s drug development potential.

Reaction catalyzed by SalL, a chlorinase involved in salinosporamide A biosynthesis.

Over 2,000 chlorinated natural products have been discovered. Until now, common enzymatic methods for C–Cl bond formation have involved oxidation of ground-state chloride to form reactive electrophilic or radical species, which then function as halogenating agents. SalL, however, uses chloride to displace L-methionine from S-adenosyl-L-methionine (SAM) and generate 5'-chloro-5'deoxyadenosine (5'-ClDA) in a rarely observed nucleophilic substitution strategy, which is analogous to that of fluorinase from the fluoroacetate producer Streptomyces cattleya. Fluorinated compounds make up approximately 20% of all drugs in the market. Understanding structural similarities between SalL and flurorinase could lead to new halogenation pathways and more effective drug treatment strategies.

Closeup view of SalL's active site. (a) SalL single active-site mutant Y70T with trapped substrates SAM and chloride. (b) Reorientation of panel a, showing that chloride is well positioned for nucleophilic attack on C5 of SAM, i.e., angle of ~180° between Cl and S+-C bond is consistent with a SN2-type mechanism. The mutation allows water to enter the halide binding pocket, reducing the nucleophilicity of chloride and consequently the turnover of the enzyme (amount of substrate converted by the enzyme). (c) Wild-type SalL with co-purified products 5'-ClDA and L-methionine.

SalL chlorinase was characterized in vivo and in vitro. First, the researchers disrupted the salL gene in S. tropica, abolishing sal A production while still allowing accumulation of sal B. Addition of 5'-ClDA to the salL mutant culture selectively restored sal A assembly, confirming that 5'-ClDA is a biosynthesis intermediate. SalL recombinant protein purified from E. coli organizes as a homotrimer (30 kDa/monomer). Biochemical in vitro analyses showed it also accepts bromide and iodide as substrates but not fluoride. However, kinetic constants measured in vitro suggest the brominase and iodinase activities have no biological relevance in the marine environment since the Km for these halides is at least two orders of magnitude higher than their concentration in seawater.

Next, the researchers determined the crystal structures of SalL wild-type, and single (Y70T) and double (G131S/Y70T) active-site mutants co-purified with substrates or products by molecular replacement using an available fluorinase structure. SalL’s monomeric units are similarly organized to fluorinase's, comprising two domains connected by a well-defined loop of 17 residues. The catalytic machinery of these enzymes is located at the interface of adjacent monomers between their N- and C-termini with three active sites per trimer. Analysis of these structures revealed that SalL differs from fluorinase by the absence of a 23-amino-acid inserted loop in the N-terminal domain between residues 87 and 90, which decreases the buried area around the active site. Furthermore, fluorinase is a hexamer (a dimer of trimers).

Site-directed mutagenesis confirmed residues important for activity. Replacement of Tyr70 by a less bulky threonine reduces substrate turnover to 0.07%. Accordingly, the Y70T crystal structure shows SAM and chloride substrates trapped in the active site. Moreover, in contrast to the wild-type structure, Y70T allows water to enter the halide binding pocket, inhibiting nucleophilic substitution.

Despite mutations toward a more "fluorinase-like" active site (evidenced in SalL double mutant G131S/Y70T), no activity was detected in the presence of fluoride. A structural overlay of fluorinase and the double mutant shows the active sites are not identical (G131S/Y70T has a larger halide binding pocket). The extra 23-residue loop in the fluorinase N-terminal domain sits just above the active site, appearing to modify its architecture and residue-product interactions, thus influencing halide specificity.

Overlay of the active sites of SalL double mutant (off-white) and fluorinase (purple). Left: SalL on top, fluorinase below. In fluorinase, Ser158 forms a hydrogen bonding network with Thr75, the 2'-OH of the ribose, and the fluorine atom of 5'-FDA (dotted lines). The loop harboring Ser131 in the chlorinase SalL double mutant is displaced by 1.4 Å relative to the equivalent loop harboring Ser158 in fluorinase, disrupting the hydrogen bonding network. Right: Fluorinase on top; SalL below. The N-terminal region of fluorinase contains the inserted fluorinase loop (pink) that interacts with the N-terminal region of the adjacent monomer (domain 1) and packs itself against the C-terminal region of the same monomer (domain 2) and hence compresses the loop carrying Ser158.

SalL's mechanism for chlorine incorporation is orthogonal (unrelated) to known examples of biological chlorination, expanding our knowledge of nature's halogenating repertoire. The discovery of common committed steps for the biosynthesis of chlorinated and fluorinated natural products opens the door for engineering of new drug candidates. The team recently synthesized fluorosalinosporamide in a salL mutant by feeding 5'-FDA to the corresponding fluorinated product of the SalL reaction.

Research conducted by A.S. Eustáquio and B.S. Moore (University of California, San Diego), and F. Pojer and J.P. Noel (Salk Institute for Biological Studies, La Jolla, CA).

Publication about this research: A.S. Eustáquio, F. Pojer, J.P. Noel, and B.S. Moore, "Discovery and characterization of a marine bacterial SAM-dependent chlorinase," Nature Chemical Biology4, 69 (2007).

Research funding: The U.S. National Oceanic and Atmospheric Administration, the U.S. National Institutes of Health, the U.S National Science Foundation, the Life Sciences Research Foundation, Tularik, Inc., and the Deutsche Forschungsgemeinschaft. Operation of the ALS is supported by the U.S. Department of Energy, Office of Basic Energy Sciences (BES).