Final Report Summary - AAA+LON (Mechanistic insights into protein degradation by Lon, a AAA+ protease)
To study the mechanism of substrate degradation by Lon, we decided to isolate inhibitors that interfere with Lon activity and to use them to elucidate key steps in substrate processing. To isolate such inhibitors, an assay was developed to enable high throughput screening of a small molecule library. The assay is based on the use of a fluorescein conjugated variant of the titinI27 protein that is fused to the sul20C Lon degradation tag. Using this substrate, degradation can be detected via monitoring a decrease in fluorescence anisotropy (Fig. 1A-C). This assay can be performed in a 384-well format and was indeed used by us to screen the NCI library of 1468 molecules. We managed to identify several compounds, one of which, 60C3, presented itself as a potent Lon inhibitor (Fig. 2A&B). Indeed, we found a Ki of 0.3 M for the inhibition of Lon by this compound. Next, we examined the inhibition mechanism of our identified compound. We found that the inhibitor acts at the level of ATP hydrolysis by Lon. Namely, it inhibits the ATPase activity of the protease (Fig. 3). Interestingly, it does not block the ATPase completely under inhibitor concentrations that are sufficient for complete inhibition of substrate degradation. These findings suggest that either, in addition to inhibition, this compound decouples the ATPase activity from substrate processing, or that it act in a yet unidentified manner. These issues are currently investigated in the lab.
Based on the structure of 60C3, compounds with similar structures were tested for Lon inhibition, some of which presented an inhibitory effect in vitro (not shown). To generate an in vivo Lon degradation assay, the Mycobacterium smegmatis zurA gene was fused to a 5' FLAG tag coding sequence and to a 3' sul20C coding sequence (Fig. 4A). The whole gene fusion was cloned into pBAD24 and can therefore be expressed via L-Arabinose supplementation (0.2 % (w/v) to the growth media. The resulting chimeric protein, FLAG-Zur-sul20C, is rapidly degraded by Lon and can be detected by Western analysis using anti-FLAG antibodies. To detect Lon degradation, FLAG-Zur-sul20C is expressed for 30 min in exponentially growing cells. Then, following protein synthesis block by rifampicin (200 mg/L), samples are collected for western analysis (Fig. 4B). Using this assay, we identified, among our set of molecules, a compound (6279) that inhibited Lon degradation in vivo (Fig. 5A). For comparison, 60C3 and three other compounds, A, B and C, that were examined did not elicit the same inhibitory effect (Fig. 5B). Altogether, these results indicate that compound 6279 can cross the bacterial cell envelope and inhibit Lon in the E. coli cytoplasm.
As an additional approach for elucidating the effect of substrates of Lon activity, we tested degradation rates of circular permutants of a Lon substrate that we previously identified (Gur & Sauer, 2008). This substrate constitutes residues 3-93 of the Escherichia coli -galactosidase (hereafter, -gal3- 93), contains a Lon recognition sequence between residues 49-68 and lacks a defined tertiary structure. By introducing cysteins at desired locations along-gal3-93 and following disulfide bond formation, several circular permutants were generated (Fig. 6). We tested their degradation in the presence and in the absence of DTT, thus allowing Lon to attempt degradation both in the circular and in the linear form of these substrates. These experiments revealed that although the recognition sequence was accessible in all permutants, a free N-terminal tail was required for substrate degradation (Fig. 6). These results are consistent with a model by which the degradation tag facilitates initial interaction with the protease, but recruitment of an unfolded region by the pore is a second necessary step. In our substrate, however, it seems that engagement of the N-nerminal tail is much more efficient than engagement of the C-terminal tail, suggesting that having an unstructured tail is necessary but not sufficient. Rather, that the sequence of the tail matters. These conclusions will be tested in future studies.
In general, the achievements of this project can be summarized as follows:
1. A better mechanistic understanding of the interaction of Lon with its substrates was gained
2. Molecules were identified that specifically inhibit Lon in vitro and in vivo. The molecules have a potential to become antibiotic medicine and, as such, can help to treat infectious diseases
Based on the structure of 60C3, compounds with similar structures were tested for Lon inhibition, some of which presented an inhibitory effect in vitro (not shown). To generate an in vivo Lon degradation assay, the Mycobacterium smegmatis zurA gene was fused to a 5' FLAG tag coding sequence and to a 3' sul20C coding sequence (Fig. 4A). The whole gene fusion was cloned into pBAD24 and can therefore be expressed via L-Arabinose supplementation (0.2 % (w/v) to the growth media. The resulting chimeric protein, FLAG-Zur-sul20C, is rapidly degraded by Lon and can be detected by Western analysis using anti-FLAG antibodies. To detect Lon degradation, FLAG-Zur-sul20C is expressed for 30 min in exponentially growing cells. Then, following protein synthesis block by rifampicin (200 mg/L), samples are collected for western analysis (Fig. 4B). Using this assay, we identified, among our set of molecules, a compound (6279) that inhibited Lon degradation in vivo (Fig. 5A). For comparison, 60C3 and three other compounds, A, B and C, that were examined did not elicit the same inhibitory effect (Fig. 5B). Altogether, these results indicate that compound 6279 can cross the bacterial cell envelope and inhibit Lon in the E. coli cytoplasm.
As an additional approach for elucidating the effect of substrates of Lon activity, we tested degradation rates of circular permutants of a Lon substrate that we previously identified (Gur & Sauer, 2008). This substrate constitutes residues 3-93 of the Escherichia coli -galactosidase (hereafter, -gal3- 93), contains a Lon recognition sequence between residues 49-68 and lacks a defined tertiary structure. By introducing cysteins at desired locations along-gal3-93 and following disulfide bond formation, several circular permutants were generated (Fig. 6). We tested their degradation in the presence and in the absence of DTT, thus allowing Lon to attempt degradation both in the circular and in the linear form of these substrates. These experiments revealed that although the recognition sequence was accessible in all permutants, a free N-terminal tail was required for substrate degradation (Fig. 6). These results are consistent with a model by which the degradation tag facilitates initial interaction with the protease, but recruitment of an unfolded region by the pore is a second necessary step. In our substrate, however, it seems that engagement of the N-nerminal tail is much more efficient than engagement of the C-terminal tail, suggesting that having an unstructured tail is necessary but not sufficient. Rather, that the sequence of the tail matters. These conclusions will be tested in future studies.
In general, the achievements of this project can be summarized as follows:
1. A better mechanistic understanding of the interaction of Lon with its substrates was gained
2. Molecules were identified that specifically inhibit Lon in vitro and in vivo. The molecules have a potential to become antibiotic medicine and, as such, can help to treat infectious diseases
