Natural Product Discovery with LC-MS/MS Diagnostic Fragmentation Filtering: Application for Microcystin Analysis

Natural products are often biosynthesized as a class of compounds, and not as a single product. Diagnostic fragmentation filtering is a simple approach for detecting all structurally-related compounds within a complex mixture. Diagnostic fragmentation filtering, implemented within the MZmine software, allows the analyst to explore mass spectrometry data sets for new natural products that would otherwise be missed.

This includes new bioproducts, and also new toxins. Diagnostic fragmentation filtering, applied to toxin classes that contaminate food and water, can inform exposure assessments. When applied to microbial and plant extracts, it can result in the discovery of new compounds.

Using diagnostic fragmentation filtering requires an understanding of the key tandem mass spectrometry features that define a chemical class. This means that structurally similar compounds will generate characteristic ions. My name is Shawn Hoogstra.

I am a research technician and software developer at our center, and I will be demonstrating the procedure. To prepare a sample for microcystin analysis, inoculate 30 milliliters of sterile cyanobacteria growth medium to approximately five-times-10-to-the-fifth cells per milliliter, under aseptic conditions, in 250-milliliter Erlenmeyer flasks, monitoring the cell density with a hemocytometer. After 26 days of photoautotrophic culture, illuminated with cool white fluorescent light using a 12-hour light-dark regime at 27 degrees Celsius, with swirling once daily, use 47-milliliter diameter GF/C glass microfiber filter papers to separate the cells from the culture medium by vacuum filtration.

Add three milliliters of 80%methanol to the harvested cells in 14-milliliter test tubes, and vortex and subsequently sonicate each tube for 30 seconds. After sonication, lyse the samples with three consecutive one-hour, minus-20-degree-Celsius freeze, and 15-minute room temperature thaw cycles, and filter each resulting cyanobacteria cell extract through individual 0.22-micrometer polytetrafluoroethylene syringe filters. Using a gentle stream of nitrogen gas, dry the extracts with an evaporator at 30 degrees Celsius, and store the extracts dry at minus-20 degrees Celsius.

For liquid chromatography tandem mass spectrometry analysis, reconstitute the dried residue to be analyzed with 500 microliters of 90%methanol, and vortex the samples for 30 seconds. Transfer to an amber high-pressure liquid chromatography vial. Then analyze the cyanobacteria extract using a data-dependent acquisition method on a high-resolution mass spectrometer according to standard protocols.

Most microcystins contain an ATA residue that produces two diagnostic product ions in tandem mass spectrometry, M through Z 135.0803, and 163.1114. For non-targeted liquid chromatography tandem mass spectrometry data set preparation, open MZmine 2, and select Raw data import"under Raw data methods"drop-down menu. Then select the data files from the microcystin analysis, and select the peak picking filter to apply a vendor-supplied centroiding algorithm.

For diagnostic fragmentation filtering, or DFF, of the imported data-dependent acquisition files, use the cursor to highlight the data files, and open the Visualization"drop-down menu to select the DFF option. In the DFF dialogue box that appears, input the range of retention times and minutes when the targeted class of analytes will elute, and to find the mass-to-charge ratio range of the targeted class of analytes, including the possibility for multiple charged compounds when appropriate. Input the achievable tandem mass spectrometry mass accuracy of the mass spectrometry instrument, and input the class-specific product ions of the mass-to-charge ratio.

Enter the class-specific neutral losses, and define the minimum intensity for diagnostic product ions and/or neutral losses to be considered as a percentage of the base peak of the tandem mass spectrometry spectra. Then select a path and file name to output the results and click OK"to start the DFF. A DFF plot will appear upon a successful completion of the setup.

The analyst must interpret tandem mass spectra of known compounds within a compound class to identify the product ions and/or neutral losses that are diagnostic for the entire class of compounds. This DFF plot was generated following the analysis of M.aeruginosa CPCC 300. The x-axis represents this mass-to-charge ratio of the precursor ions that satisfied the defined DFF criteria, while the y-axis shows the mass-to-charge ratio of all of the product ions within the microcystins tandem mass spectrometry spectra.

For this analysis, the criteria for microcystin detection included precursor ions within the mass-to-charge ratio of 400 to 1, 200, and retention times between two to six minutes. Most importantly, these tandem mass spectrometry spectra contained both mass-to-charge ratios above the defined 15%base peak intensity threshold. Of the 4, 116 tandem mass spectrometry spectra acquired during the analysis, 26 satisfied the DFF criteria and were detected in the M.aeruginosa CPCC 300 extract.

The major microcystins detected could be confidently assigned as microcystin leucine arginine and demethylated aspartic acid microcystin leucine arginine. As new compounds are discovered using diagnostic fragmentation filtering, the organisms can be grown in large scale for isolation and characterization of the compounds by NMR. When applied to microcystins, diagnostic fragmentation filtering allows us to determine if traditional targeted screening methods are detecting the major toxins present within a sample.