Accelerating the Analysis of Cyanotoxins Sébastien Sauvé Environmental

Accelerating the Analysis of Cyanotoxins Sébastien Sauvé Environmental Analytical Chemistry – Université de Montréal sebastien. sauve@umontreal. ca Dipankar Ghosh Director for Environmental & Food safety Thermo Fisher Scientific Dipankar. ghosh@thermofisher. com

Contributors Audrey Roy-Lachapelle Khadija Aboulfadl Pascal Lemoine Sherri Macleod Liza Viglino Arash Zamyadi Michèle Prévost

Context • Microcystins are hepatotoxins produced by cyanobacteria (Bluegreen Algae) • These cyanotoxins are found in fresh waters and in drinking water reservoirs. Cyanobacterial bloom • A bloom can occur in warm, shallow, undisturbed surface water rich in nutrients. Microcystis aeruginosa http: //www. aquarius-systems. com/Entries/View/349/bluegreen_algae. aspx http: //www. plingfactory. de/index. html K. Sivonen, G. Jones, in: 1. Chorus, J. Bartram (Eds. ), Toxic Cyanobacteria in Water: A Guide to their Public Health Consequences, London, 1999.

Objectives • Multi-toxin online SPE-LC-MS/MS method • Ultrafast laser diode thermal desorption methods (LDTD-APCI-MS/MS) • Anatoxin-A • Sum of microcystins

Online SPELC-MS/MS high pressure multi-toxins method Sébastien Sauvé, Département de chimie

SPE: Enrichmment (solid phase extraction)

Automated SPE Extraction (Online SPE) 1. 0 ml SPE Waste Chromatography MS/MS The whole mass of analytes within the 1. 0 ml sample will ne injected into the MS detector

LC-MS/MS Detection: Tandem mass spectrometry (selected reaction monitoring - SRM) Thermo. Electron TSQ Quantum Ultra EQuan MAX System

Tandem Mass Spectrometry (MS/MS) Argon-induced Fragmentation m/z=156. 0 m/z=108. 0 Sulfamethoxazole+H+ m/z= 254. 0 m/z=92. 0

Cyanotoxins using LC-MS/MS Challenge is to combine varied compounds into a single method for the simultaneous determination of different cyanotoxins.

Target compounds

Challenge Speed!! • Eliminate off line SPE • Separate phenylanaline from anatoxin a (same SRM) http: //fav. me/dsk 92 w

Anatoxine-a and phenylalanin • Separation of isobars using chromatography • Quantification of specific fragment for anatoxin-a (166. 10 > 43. 3) anatoxine-a 166. 10 > 43. 3 quantification phénylalanine 166. 10 > 120. 0 anatoxine-a 166. 10 > 131. 1 anatoxine-a 166. 10 > 149. 1

Specific conditions for cyanotoxin determinations

Chromatograms obtained using SPEUPLC/MSMS-ESI, in Milli-Q water spiked @ 1 µg/l Chromatograms obtained using SPEUPLC/MSMS-ESI, in real sample (lab culture)

Preliminary estimates of performance

Even faster? • LDTD

Laser diode thermal desorption (LDTD) § Principles of the LDTD-APCI source: ü technique that combines thermal desorption (laser diode) and APCI ü sample is spotted (1 -10 μL) into a 96 -well plate and air-dried for 2 min ü uncharged analytes are thermally desorbed into the gas phase ü ionization takes place in the corona discharge region by APCI and the charged molecules will be transferred toformationinlet Primary ion the MS → e- + N 2 → N 2+. + 2 e. Secondary ion formation → N 2+. + H 2 O → N 2 + H 2 O+. → H 2 O+. + H 2 O → H 3 O+ + HO. Proton transfer → H 3 O+ + M → (M+H)+ + H 2 O Source: www. chm. bris. ac. uk/ms/theory/apci-ionisation. html

Laser diode thermal desorption (LDTD) LDTD (980 nm, 20 W) IR Laser Installation Au to -sa 96 0 s m ple am r ple s Can ramp up to 3000 o. C/sec. Laser power is defined in % Normally ~100 -150 o. C Corona needle position (APCI)

Laser diode thermal desorption (LDTD)

LDTD Process (2) (5) (1) (3) (4) (1) Infrared laser (980 nm, 20 W) (2) Laz. Well Plate (96 wells): analyte desorption (1 -10 µL spotted) (3) Transfer tube transporting the neutrally desorded analytes to the APCI region (4) Corona needle discharge region (APCI) (5) MS inlet

LDTD Optimization § No need to optimize liquid chromatography - it has been completely eliminated! § Optimization for MS (precursor) and MS/MS (SRM transitions) conditions in NI and PI mode. A minimum of 2 SRM transitions were selected + their ion ratios § Parameters of the LDTD-APCI source are optimized for signal intensity : solvent choice for analyte deposition in the well cavities ü laser power (%) ü carrier gas flow rate (L/min) ü mass deposition (deposition volume in µL) into plate well ü laser pattern ü

Results / challenge Only anatoxin-a can be vaprized and ionized by LDTD-APCI. 166. 03>149. 06 Interference from phénylalanine: Different desorption patern (signal intensity vs laser power). SRM Optimisation (main SRM identical). 166. 06>149. 02

Separation using a gradient of LDTD laser power

Performances (anatoxin-a)

Microcystins Adda MMPB Leucine (L) Microcystin-LR Arginine (R) • The are over 80 known microcystins. • A unique structural feature: Adda (3 -amino-9 -methoxy-2, 6, 8 trimethyl-10 -phenyldeca-4, 6 -dienoic acid) which plays an important role in its toxicity. X. Wu, C. Wang, B. Xiao, Y. Wang, N. Zheng, J. Liu. Analytical Chimica Acta, 709 (2012) 66 -72.

Context • The presence of microcystins can pose an health risk for humans and animals: –Skin irritation, vomiting, diarrhea, asthma, headache, fever, and muscle weakness. –Inhibiting protein phosphatases in tissues, causing serious damage to the liver from bioaccumulation. The World Health Organisation (WHO) recomends a guideline for MC-LR of 1 mg l. L-1 in drinking water. K. Sivonen, G. Jones, in: 1. Chorus, J. Bartram (Eds. ), Toxic Cyanobacteria in Water: A Guide to their Public Health Consequences, London, 1999.

Alternatives Need of robust detection methods to evaluate and control the risks due to the presence of microcystins in water. Analytical Methods Advantages Disadvantages • Specific analysis • Time consuming • Stantards limitation • Expensive GC-MS • Total MC analysis • More steps (need of derivatization) • Time consuming ELISA • Fast and easy • Unexpensive • Total MC analysis • Binding constants of the MC with the anti-body may vary • Cross-selectivity HPLC-UV HPLC-MS D. O. Mountfort, P. Holland, J. Sprosen. Toxicon 45 (2005) 199 -206 K. Kaya, T. Sano. Analytica Chimica Acta 386 (1999) 107 -112.

Objective • Objective: Analysis of total microcystins using LDTD-APCI-MS/MS technology. • The method provides: –Instant information about risks of contamination –Information about the whole spectrum of cyanobacterial peptide toxins congeners

Oxydation Experimental workflow: –Lemieux oxidation of microcystins into MMPB –Liquid-liquid extraction (Ethyl acetate) –Desorption by LDTD –Negative ionisation by APCI –Detection with a TSQ Quantum Ultra AM triple quadrupole mass spectrometer erythro-2 -Methyl-3 -methoxy-4 -phenylbutyric Acid MMPB X. Wu, C. Wang, B. Xiao, Y. Wang, N. Zheng, J. Liu. Analytical Chimica Acta, 709 (2012) 66 -72. M-R. Neffling, E. Lance, J. Meriluoto. Environmental Pollution, 158 (2012) 948 -952

Lemieux oxidation Adda KMn. O 4 + Na. IO 4 Microcystin MMPB • 0, 05 M Potassium permanganate (KMn. O 4) and 0, 05 M Sodium periodate (Na. IO 4) • Oxidation, at room temperature and p. H 9 for 1 hour • Reaction quenched with saturated sodium bisulfite • Use of sulfuric acid 10% to reach p. H 2 T. Sano, K. Nohara, F. Shiraishi, K. Kaya. J. Environ. Anal. Chem. , 49 (1992) 163 -170.

Microcystins Oxidation Optimisation Reagents concentrations KMn. O 4 and Na. IO 4 optimised at 0, 05 M

Microcystins Oxidation Optimisation Oxidation time Optimal oxidation time at 1 h

LDTD parameter optimisation Laser power Best laser power at 35%

Microcystins Oxidation Optimisation p. H during oxidation Optimised conditions at p. H 9

Microcystin detection and quantification Quantification of MMPB by internal calibration with 4 -phenylbutyric acid APCI (-) Scan time: 0, 005 s Q 1 width: 0, 70 amu Q 3 width: 0, 70 amu MMPB 4 -phenylbutyric acid (4 -PB) (Internal standard) Optimal Selected Reaction Monitoring (SRM) parameters for the analysis of MMPB and 4 -PB by MS/MS Compound MMPB 4 -PB Precursor ion (m/z) 207. 1 163. 1 Product ion (m/z) Collision energy (V) 131* - 15 175 - 13 91* - 16 119 - 12 Tube Lens (V) - 25 - 22

Analysis of MMPB with LDTD-APCI-MS/MS Internal Calibration (MMPB / 4 -PB ratio) Method Validation n=6 R 2 : 0, 9995 Linearity range: 1 – 500 mg/L LOD: 1 mg/L LOQ: 3 mg/L Standards Avg. RSD < 9% Calibration curve showing the linearity of the LDTD experiment Oxidation reaction yield of Microcystins : 111% MMPB recovery yield : 48% WHO Guideline: 1 mg/L

Conclusions • An 8 -min automated online SPE-LC-MS/MS method for many toxins (but excluding saxitoxins) • Ultrafast laser diode thermal desorption methods (LDTD-APCI-MS/MS) (15 sec per sample but with simple oxydation for MC) • Anatoxin-A • Sum of microcystins

Acknowledgements Parterns and funding agencies:

Questions? Dipankar. ghosh@thermofisher. com sebastien. sauve@umontreal. ca

Analysis with LDTD-APCI-MS/MS LDTD Source (0, 5 -3 L/min) (980 nm, 20 W) http: //ldtd. phytronix. com/

Analysis with LDTD-APCI-MS/MS LDTD Source 10 -plate sample loader LDTD a sample introduction method using thermal desorption –Minimal sample preparation –Small volume of sample needed (1 -5 m. L) – 15 sec / sample (no chromatographic separation) 10 plates in the loader = 960 samples –No carryover –Combined with atmospheric ionisation (APCI) –High-thoughput http: //ldtd. phytronix. com/ Laz. Well sample plate

LDTD parameter optimisation Laser desorption parameters Laser Power: 35% Gas Flow: 3 L/min Deposition volume: 2 m. L Laser pattern duration: 6 s Laser pattern LDTD peak shape

LDTD parameter optimisation Ethyl Acetate is the best deposition solvent Deposition solvent Plate well Sample residue Carrier gas flow rate Gaz flow at 3, 0 L/min