New antibiotics are needed to combat rising resistance, with new Mycobacterium tuberculosis (Mtb) drugs of highest priority. Conventional whole-cell and biochemical antibiotic screens have failed. We developed a novel strategy termed PROSPECT (PRimary screening Of Strains to Prioritize Expanded Chemistry and Targets) in which we screen compounds against pools of strains depleted for essential bacterial targets. We engineered strains targeting 474 Mtb essential genes and screened pools of 100-150 strains against activity-enriched and unbiased compounds libraries, measuring > 8.5-million chemical-genetic interactions. Primary screens identified > 10-fold more hits than screening wild-type Mtb alone, with chemical-genetic interactions providing immediate, direct target insight. We identified > 40 novel compounds targeting DNA gyrase, cell wall, tryptophan, folate biosynthesis, and RNA polymerase, as well as inhibitors of a novel target EfpA. Chemical optimization yielded EfpA inhibitors with potent wild-type activity, thus demonstrating PROSPECT's ability to yield inhibitors against novel targets which would have eluded conventional drug discovery.

CYP199A4 (RPB3613) from Rhodopseudomonas palustris HaA2 is a heme monooxygenase that catalyzes the hydroxylation of para-substituted benzoic acids. Monooxygenase activity of CYP199A4 can be reconstituted in a Class I electron transfer chain with an associated [2Fe-2S] ferredoxin, HaPux, (RPB3614) and the flavin-dependent reductase, HaPuR, (RPB3656) that is not associated with a CYP gene. CYP199A4 and the ferredoxin HaPux are produced in greater quantities using recombinant Escherichia coli expression systems when compared to the equivalent proteins in the closely related CYP199A2-Pux-PuR Class I system from R. palustris CGA009. HaPuR and HaPux can also replace PuR and Pux in supporting the CYP199A2 enzyme turnover with high activity. Whole-cell in vivo substrate oxidation systems for CYP199A4 and CYP199A2 with HaPux and HaPuR as the electron transfer proteins have been constructed. These E. coli systems were capable of selectively demethylating veratric acid at the para position to produce vanillic acid at rates of up to 15.3 microM (g-cdw)(-1) min(-1) and yields of up to 1.2 g L(-1).

The crystal structures of the 4-methoxybenzoate bound forms of cytochrome P450 enzymes CYP199A2 and CYP199A4 from the Rhodopseudomonas palustris strains CGA009 and HaA2 have been solved. The structures of these two enzymes, which share 86% sequence identity, are very similar though some differences are found on the proximal surface. In these structures the enzymes have a closed conformation, in contrast to the substrate-free form of CYP199A2 where an obvious substrate access channel is observed. The switch from an open to a closed conformation arises from pronounced residue side-chain movements and alterations of ion pair and hydrogen bonding interactions at the entrance of the access channel. A chloride ion bound just inside the protein surface caps the entrance to the active site and protects the substrate and the heme from the external solvent. In both structures the substrate is held in place via hydrophobic and hydrogen bond interactions. The methoxy group is located over the heme iron, accounting for the high activity and selectivity of these enzymes for oxidative demethylation of the substrate. Mutagenesis studies on CYP199A4 highlight the involvement of hydrophobic (Phe185) and hydrophilic (Arg92, Ser95 and Arg243) amino acid residues in the binding of para-substituted benzoates by these enzymes.

CYP199A2 from Rhodopseudomonas palustris CGA009 is a heme monooxygenase that catalyzes the oxidation of para-substituted benzoic acids. CYP199A2 activity is reconstituted by a class I electron transfer chain consisting of the associated [2Fe-2S] ferredoxin palustrisredoxin (Pux) and a flavoprotein palustrisredoxin reductase (PuR). Another [2Fe-2S] ferredoxin, palustrisredoxin B (PuxB; RPA3956) has been identified in the genome. PuxB shares sequence identity and motifs with vertebrate-type ferredoxins involved in Fe-S cluster assembly but also 50% identity with Pux and it mediates electron transfer from PuR to CYP199A2, albeit with lower steady-state turnover activity: 99 nmol (nmol P450)(-1)min(-1) for 4-methoxybenzoic acid oxidation compared with 1,438 nmol (nmol P450)(-1 )min(-1) for Pux. This difference mainly arises from weak CYP199A2-PuxB binding (K (m) 34.3 vs. 0.45 microM for Pux) rather than slow electron transfer (k (cat) 19.1 vs. 37.9 s(-1) for Pux). Comparison of the 2.0-A-resolution crystal structure of the PuxB A105R mutant with other vertebrate-type, P450-associated ferredoxins revealed similar protein folds but also significant differences in some loop regions. Therefore, PuxB offers a platform for studying ferredoxin-P450 recognition in class I P450 systems. Substitution of PuxB residues at key locations with those in Pux shows that Ala42, Cys43, and Ala44 in the [2Fe-2S] cluster binding loop and Met66 are important in electron transfer from PuxB to CYP199A2, whereas Phe73 and the C-terminal Ala105 were involved in both protein binding and electron transfer.

Cytochrome P450-199A2 from Rhodopseudomonas palustris oxidizes para-substituted benzoic acids and may play a role in lignin and aromatic acid degradation pathways in the bacterium. CYP199A2 has an associated [2Fe-2S] ferredoxin, palustrisredoxin (Pux) but not a ferredoxin reductase. A genome search identified the palustrisredoxin reductase (PuR) gene. PuR was produced in Escherichia coli and shown to be a flavin-dependent protein that supports efficient electron transfer from NADH to Pux, thus reconstituting CYP199A2 monooxygenase activity (k(cat) = 37.9 s(-1) with 4-methoxybenzoic acid). The reduction of Pux by PuR shows K(m) = 4.2 microM and k(cat) = 262 s(-1) in 50 mM Tris, pH 7.4. K(m) is increased to 154 microM in the presence of 200 mM KCl, indicating the importance of ionic interactions in PuR/Pux binding. The crystal structure of PuR has been determined at 2.2 A resolution and found to be closely related to that of other oxygenase-coupled NADH-dependent ferredoxin reductases. Residues on the surface that had been proposed to be involved in ferredoxin reductase-ferredoxin binding are conserved in PuR. However, Lys328 in PuR lies over the FAD isoalloxazine ring and, together with His11 and Gln41, render the electrostatic potential of the surface more positive and may account for the greater involvement of electrostatic interactions in ferredoxin binding by PuR. Consistent with these observations the K328G mutation weakened Pux binding and virtually eliminated the dependence of PuR/Pux binding on salt concentration, thus confirming that the FAD si side surface in the vicinity of Lys328 is the ferredoxin binding site.

A ferredoxin associated with biological Fe-S cluster assembly has been remodelled to transfer electrons to a P450 enzyme and support substrate oxidation at 80% of the physiological ferredoxin activity, opening up the possibility of tailoring ferredoxins to reconstitute the activity of P450 enzymes for which the electron transfer partner proteins are not known.

The ubiquitin–proteasome system targets selected proteins for degradation by the 26S proteasome. Rpn12 is an essential component of the 19S regulatory particle and plays a role in recruiting the extrinsic ubiquitin receptor Rpn10. In the present paper we report the crystal structure of Rpn12, a proteasomal PCI-domain-containing protein. The structure helps to define a core structural motif for the PCI domain and identifies potential sites through which Rpn12 might form protein–protein interactions. We demonstrate that mutating residues at one of these sites impairs Rpn12 binding to Rpn10 in vitro and reduces Rpn10 incorporation into proteasomes in vivo.

The efficacies of all antibiotics against tuberculosis are eventually eroded by resistance. New strategies to discover drugs or drug combinations with higher barriers to resistance are needed. Previously, we reported the application of a large-scale chemical-genetic interaction screening strategy called PROSPECT (PRimary screening Of Strains to Prioritize Expanded Chemistry and Targets) for the discovery of new Mycobacterium tuberculosis inhibitors, which resulted in the identification of the small molecule BRD-8000, an inhibitor of a novel target, EfpA [Nature201951772]. Leveraging the chemical genetic interaction profile of BRD-8000, we identified BRD-9327, another structurally distinct small molecule EfpA inhibitor. We show that the two compounds are synergistic and display collateral sensitivity because of their distinct modes of action and resistance mechanisms. High-level resistance to one increases the sensitivity to and reduces the emergence of resistance to the other. Thus, the combination of BRD-9327 and BRD-8000 represents a proof-of-concept for the novel strategy of leveraging chemical genetics in the design of antimicrobial combination chemotherapy in which mutual collateral sensitivity is exploited.