AmericanChemicalSociety.com

The radical SAM (RS) superfamily of enzymes is a newly described class of metalloproteins that employ a unique [4Fe–4S]2+/+–S-adenosyl-L-methionine (SAM) cofactor to initiate and carry out catalysis by mechanisms involving obligate radical intermediates. Three of the iron atoms of the [4Fe–4S] cluster are coordinated by one of three conserved cysteinyl residues residing in a CxxxCxxC motif, common to almost all enzymes within this superfamily. The fourth iron, which is not ligated by a protein-derived cysteinyl residue, contains open coordination sites to which SAM binds via its α-amino and α-carboxylate groups. The function of the iron–sulfur (Fe/S) cluster is to inject an electron into the sulfonium group of SAM, causing it to fragment into L-methionine and a 5'-deoxyadenosyl 5'-radical, an obligate intermediate in RS enzymes. This intermediate then initiates catalysis by abstracting a hydrogen atom from the appropriate substrate. Relatively little attention has been given to understanding the energetics of forming the high-energy 5'-dA• that is responsible for initiating catalysis. Redox potential determinations have only been carried out on a few RS enzymes, and the midpoint potentials of the bound Fe/S clusters are typically ~-500 mV, which is significantly higher (i.e., less favorable) than that required to inject an electron into the sulfonium of SAM (~-1.8 V), suggesting that compensatory interactions involving the enzyme and the relevant substrates and cofactors must be operative. The goal of the grant from the ACS-PRF was to try to determine what these compensatory interactions are for several RS enzymes, notably lipoyl synthase (LS) and biotin synthase (BS).

Several specific aims were noted:

1. To conduct X-ray absorption spectroscopy (XAS) studies on LS and BS in collaboration with Dr. Robert Scott (University of Georgia) to assess whether the sulfur of methionine becomes coordinated to one of the irons of the Fe/S cluster concomitant with reductive cleavage of SAM, and to determine whether coordination correlates with the ability of the substrate to induce cleavage of SAM.

2. To assess whether the binding of SAM and SAM analogs to the [4Fe–4S] cluster in LS and BS alters the midpoint potential of the cluster.

3. To use ENDOR spectroscopy to probe the relative orientation of the methionine group of SAM in the active site of LS using analogs of SAM that are unable to form the N/O chelate.

As described in the previous report specific aim 1 has been conducted, and we have moved on to specific aims 2 and 3. In order to best conduct these aims we opted to clone and express the gene that encodes another radical SAM enzyme that appears to be significantly more tractable than LS and BS. This enzyme, BtrN, catalyzes a radical-dependent dehydrogenation reaction, oxidizing the alcohol at carbon 3 of 2-deoxy-scyllo-inosamine (DOIA) to a ketone, affording amino-2-deoxy-scyllo-inosose (amino-DOI). There are several advantages that this enzyme offers over LS and BS. First, it is one of the smallest known radical SAM enzymes in size, having a molecular mass of ~30 kDa. Second, it is extremely soluble and well behaved. We have been able to concentrate the protein to ~4 mM. Third, we have shown that the protein contains two [4Fe–4S] clusters; however, only one is readily reduced by the required reducing agent. Moreover, unlike LS, reduction takes place stoichiometrically. Last, however, we have generated crystals of the protein, as a first step in obtaining an X-ray crystal structure of it. At present, the crystals diffract to ~8 Å; however, we are in the process of fine-tuning the conditions to get better crystals.

Over the past year we have conducted a study on BtrN, which we have submitted for publication in the journal Biochemistry, with acknowledgement to ACS-PRF for funding. This paper describes the characterization of the iron–sulfur clusters on the protein. Previously, it had been reported that the protein contained only one [4Fe–4S] cluster. We used an assortment of physical techniques, including Mössbauer spectroscopy, electron paramagnetic spectroscopy, UV-vis spectroscopy, and analytical determinations of iron and sulfide in combination with site-directed mutagenesis to show that the protein contains two. This work complements our work on another class of these proteins, which we also showed contains multiple [4Fe–4S] clusters. We postulated that the second cluster found in these proteins coordinates to the substrate to allow facile electron transfer from a substrate-radical intermediate generated via abstraction of a substrate-derived hydrogen atom by the 5'-deoxyadenosyl radical. We have already initiated XAS and ENDOR studies with Dr. Robert Scott (University of Georgia) and Dr. Brian Hoffmann (Northwestern) to probe for substrate coordination. These studies will be done in concert with Specific Aims 2 and 3 of the original proposal; however, we will use BtrN instead of LS or BS.