Inhibition of the Burkholderia mallei acyl-homoserine lactone synthase BmaI1
Quorum sensing is a signalling and regulatory phenomenon in bacteria that controls many behaviors including virulence. In many species of Proteobacteria quorum sensing is mediated by a diffusible acyl-homoserine lactone (acyl-HSL) signal. We have developed a screen to target the octanoyl-HSL synthase of Burkholderia mallei, BmaI1. Our new screen is a cell-free assay that couples a prduct of the more ..
BioActive Compounds: 44
Quorum sensing is a signalling and regulatory phenomenon in bacteria that controls many behaviors including virulence. In many species of Proteobacteria quorum sensing is mediated by a diffusible acyl-homoserine lactone (acyl-HSL) signal. We have developed a screen to target the octanoyl-HSL synthase of Burkholderia mallei, BmaI1. Our new screen is a cell-free assay that couples a prduct of the reaction to generation of a fluorescence signal. There were technical hurdles in the development of the screen. First, the acyl substrate for BmaI1 is octanoyl-acyl carrier protein (ACP) of fatty acid biosynthesis. We needed to purify sufficient ACP and modify it with octanoyl-phosphopantetheine to make octanoyl-ACP. Second, the amino donor for the reaction is S-adenosylmethionine (SAM) and commercial preparations are not sufficiently pure. We synthesized SAM enzymatically for this screen. The coupling enzyme mixture is commercially available from Cayman Chemical. In the final step of the coupling system, horseradish peroxidase reduces 10-Acetyl-3,7-dihydroxyphenoxazine (ADHP or Amplex Red) to the fluorescent compound resorufin. We also perform a counterscreen assay containing only the coupling system, which is initiated by the product of the HSL synthase: 5'-methylthioadenosine (MTA).
384-well assay plates (Corning 3820) were filled with Mix 1 at 10 microL per well. Mix 1 contained 133.3 microM potassium phosphate pH 7.5, 133.3 microM manganese sulfate, 5% Cayman enzyme mix, and 2 microM BmaI1. 300 nL of each experimental compound were pin-transferred to assay wells. 3.33 microL of 40 mM acetovanillone in DMSO was added to positive control wells before the reaction was started.
The primary assay was initiated with 3.33 microL Mix 2. Mix 2 contained 100 microM octanoyl-ACP, 800 microM 10-acetyl-3,7-dihydroxyphenoxazine (ADHP), and 140 microM S-adenosylmethionine (SAM). The secondary assay was initiated by addition of 3.33 microL of Mix 4. Mix 4 contained 133.3 microM potassium phosphate pH 7.5, 133.3 microM manganese sulfate, and 4% Cayman enzyme mix. The assay was allowed to proceed for 40 to 60 minutes, at which time it was stopped by the addition of 3.33 microL of 40 mM acetovanillone to all wells except the positive control wells.
Plates were read on a PerkinElmer EnVision (525/25 excitation filter, 590/20 emission filter, 555 mirror. Gain 165, excitation light 5%, 10 flashes).
Library plates were screened using a primary assay and a counterscreen assay, with all assay plates for a given library plate prepared on the same day. For every compound library plate, there were four daughter plates. Primary plates A & B were prepared as described; counterscreen plates CNTL-A and CNTL-B contained the same components except for 5'-Deoxy-5'-(methylthio)adenosine instead of BmaI1, acyl-ACP, and SAM. Counterscreen plates also contained 3% enzyme mixture instead of 3.5%.
Positive control: All wells of column 24 contained the final enzymatic reaction mix, with acetovanillone added before the assay was started, and no experimental compound.
Negative control: All wells in column 23 contained only the final enzymatic reaction mix (no experimental compound).
For each non-excluded experimental well, a robust Z-score was calculated based on the fluorescence intensity median and normalized MAD for each replicate in the primary assay and the counterscreen assay. The absolute value of each experimental screen well Z-score was subtracted from the absolute value of plate average positive control Z-score. Wells with a robust Z-score < 2 units from the plate average positive control Z-score for both primary screen replicates, and a robust Z-score < 3 for at least one counterscreen replicate, were considered positive.
Activity scores were calculated for both the primary screen and the counterscreen using normalized percent of control values. For each well (both replicates), the positive control plate average fluorescence intensity was subtracted from the well fluorescence intensity, divided by the difference between plate average negative and positive control fluorescence intensity, subtracted from 1, and multiplied by 100. Some plates lacked a positive control; the positive control average of all plates was used for these plates. Replicate normalized percent of control values were averaged for the primary screen and counterscreen separately to determine activity scores for each. Values less than 0 were set to 0 and values greater than 100 were set to 100.
Data Table (Concise)