Preventing Wine Spoilage - IonSense · Elizabeth Crawford1,2, Paola Domizio3,4, Brian Musselman1,...
Transcript of Preventing Wine Spoilage - IonSense · Elizabeth Crawford1,2, Paola Domizio3,4, Brian Musselman1,...
Elizabeth Crawford1,2, Paola Domizio3,4, Brian Musselman1, C. M. Lucy Joseph3, Linda F. Bisson3, Bart C. Weimer5 and Richard Jeannotte5,6
11th June 2013 2013 ASMS Annual Meeting
Minneapolis, MN, USA
Preventing Wine Spoilage: Rapid Screening & Quantitative Analysis of Off-flavor Phenolic Compounds by DART Mass Spectrometry
1IonSense, Inc, Saugus, MA, USA 2Dept. of Food Analysis & Nutrition, Institute of Chemical Tech., Prague, Czech Republic 3Dept. of Viticulture & Enology, Univ. of California-Davis, Davis, CA, USA 4Dipart. di Gestione Sistemi Agrari, Alimentari e Forestali (GESAAF), Univ. degli Studi di Firenze, Italy 5Dept. of Health & Reproduction, School of Veterinary Med, Univ. of California-Davis 6Facultad de Ciencias, Univ. de Tarapacá, Arica, Chile
Brettanomyces
“Brett” Wheel http://heysmartbeerdude.files.wordpress.com/2013/04/brett-aroma-wheel.jpeg (Access: 10 June 2013)
• Budding yeast found widely distributed in nature
• Discovered in beer in 1904 (Claussen), in wine (Krumbholz & Tauschanoff,1930) and again in 1940 (Custers)
• Produces a wide array of aromatic compounds
• Wine cellar contamination was widespread
• “Brett” characters can compete with varietal characters for dominance of wine profile
Background:
When Is It Spoilage? • High concentration, dominating wine profile • Conflict with wine matrix characters • Suppression of varietal character • Enhancement of off-notes • Lactic acid bacteria often found in wines with
Brettanomyces
Slide courtesy of Dr. L. Bisson, Dept. of Viticulture & Enology, Univ. of California-Davis
Recovery Thresholds: • Chatonnet* has defined spoilage as:
• >426 ppb of 4-ethylphenol (4-EP) and 4-ethylguaiacol (4-EG) • >620 ppb of 4-EP
• 50% of tasters can detect 605 ppb in wine or 440 ppb in water of 4-EP
* Chatonnet, P.; Boidron, J. N.; Dubourdieu, D. Influence des conditions d’ élevage et de sulfitage des vins rouges en barriques sur leur teneur en acide acétique et en éthyl-phenols. J. Int. Sci. Vigne Vin. 2003, 27, 277-298.
Main Aromas and Incidence of Spoilage
Country >426ppb >620ppb
France 36% 28%
Italy 49% 19%
Australia 59% 46%
Portugal 42% 27%
Wines may contain up to 50 ppm (!) of 4-EP
• Band-Aid (4-Ethyl Phenol) • Earthy (Geosmin) • Horsy • Leather
• Putrid • Soy • Tobacco
Associated Aromas:
Slide courtesy of Dr. L. Bisson, Dept. of Viticulture & Enology, Univ. of California-Davis
Evolving Open Air Ionization
Venter, A.; Nefliu, M. Cooks, R.G., “Ambient Desorption Ionization
Mass Spectrometry”, Trends in Anal. Chem., 27, 284-290, 2008.
Classic vs. Ambient Ionization
• Direct sample analysis under ambient conditions
• No separation of sample components (chromatographic) • Minimal sample preparation requirements (sample concentration) • Analysis time of 3 minutes per sample
GC-MS Method: Sample Preparation: • Total runtime = ~50 minutes
• Total sample prep time = 15 minutes with 100% analyst involvement • GC-MS runtime of 32 minutes
• Internal standards (4-EP-d4; 4-EG-d5) spiked into wines (5 mL), vortexed, solvent added (t-butyl ether), centrifugation, 5 µL injection onto GC
(Adapted form Rayne & Eggers (2008) (Am. J. Enol. Vitic. 59:92-97))
DART Ambient Ionization Method:
Slide courtesy of Prof. Jana Hajšlová, ICT Prague, Czech Republic
(GC-MS)
Key DART Source Parameters • Source Parameters:
• Ionization gasses • N2 (Standby) • He (Run Mode)
• Temperature of ionization gas • Sample introduction speed
• Ions typically observed in DART–MS mass spectra: • [M+H]+, [M–H]–
• M+·, M–· (ionic compounds) • Use of dopants to promote
ionization: • [M+NH4]+ (ammonia) • [M+Cl]– (e.g. CH2Cl2) • [M+CF3COO]– (trifluoroacetic
acid)
Figure courtesy of Prof. Jana Hajšlová ICT Prague, Czech Republic
Targeted MS2 Settings •Scan Parameters:
•Negative Ion Mode
•Resolution: 35,000 FWHM at m/z 200 •NCE Fragmentation: 45 (4-EP); 25 (4-EG) •AGC Target: 2e5 charges •Max IT: 250 ms •Isolation Width (m/z): 3.0
•Q Exactive Inlet Parameters: •Capillary Temp: 200° C •S Lens: 50
•External Mass Calibration •All of the following parameters were set to zero:
•Sheath Gas Flow, Aux Gas Flow, Sweep Gas Flow, Spray Voltage
DART-SVP Settings
DART-SVP Coupled with Q Exactive
• DART Source:
• Negative Ion Mode
• Heater Temperature Gradient: 100 - 300° C
* SVP = Standardized Voltage and Pressure
Classic DART Sampling: Wine on Glass Tips
100 105 110 115 120 125 130 135 140 145 150 155 160
m/z
0
20
40
60
80
100
0
20
40
60
80
100
Re
lative
Ab
un
da
nce
121.02972R=47028
100.96769R=51560 122.03308
R=47276106.04254R=47475
136.05320R=42411
151.04036R=41581
137.05645R=41068
121.02967R=46360
108.02191R=48937
150.00222R=41958
101.02456R=50686
124.01674R=47195
NL: 1.95E6
Twister_WineStdCurve_750ppb_Bar1_12hrspin_25MAR13_Run01#318-414 RT: 3.00-3.74 AV: 8 F: FTMS - p NSI Full ms2 [email protected] [95.00-145.00]
NL: 1.01E6
Twister_WineStdCurve_750ppb_Bar1_12hrspin_25MAR13_Run01#357-445 RT: 3.37-4.00 AV: 7 F: FTMS - p NSI Full ms2 [email protected] [95.00-175.00]
DART_tMS2_ChiantiWine_4-EP&4-EG_0.5-1... 1/25/2013 3:53:20 PM 4-EP 45NCE;4-EG 35NCE Blue 3; 300 C3 uL spot DIP-it Method; He; UHV 3.45e-9RT: 0.00 - 4.26
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Time (min)
0
20
40
60
80
100
0
20
40
60
80
100
Re
lative
Ab
un
da
nce
3.232.93 3.70
3.563.39 3.903.034.082.89
3.202.69
2.652.20 2.361.931.87
2.18 2.541.56 1.841.21 1.511.180.83
2.933.21
3.243.38
3.693.042.89 3.55 3.88 4.082.692.68
2.20 2.341.881.96
1.56 1.851.19 1.50 1.990.980.83
NL: 3.91E6
m/z= 106.04061-106.04167 F: FTMS - p NSI Full ms2 [email protected] [50.00-145.00] MS DART_tMS2_ChiantiWine_4-EP&4-EG_0.5-100ppm_300C_DIP-it_01
NL: 1.06E7
m/z= 136.05114-136.05250 F: FTMS - p NSI Full ms2 [email protected] [50.00-175.00] MS DART_tMS2_ChiantiWine_4-EP&4-EG_0.5-100ppm_300C_DIP-it_01
100 105 110 115 120 125 130 135 140 145 150 155 160
m/z
0
20
40
60
80
100
0
20
40
60
80
100
Re
lative
Ab
un
da
nce
121.06476R=46679
106.04122R=48831
108.02054R=48979
136.05195R=44520
152.03447R=42152121.02839
R=47115
NL: 6.93E6
DART_tMS2_ChiantiWine_4-EP&4-EG_0.5-100ppm_300C_DIP-it_01#1482-1532 RT: 3.58-3.69 AV: 25 F: FTMS - p NSI Full ms2 [email protected] [50.00-145.00]
NL: 4.15E6
DART_tMS2_ChiantiWine_4-EP&4-EG_0.5-100ppm_300C_DIP-it_01#1474-1534 RT: 3.56-3.70 AV: 31 F: FTMS - p NSI Full ms2 [email protected] [50.00-175.00]
0.5 ppm 1 ppm 5 ppm
10 ppm
50 ppm 100 ppm
Spiked Wine Sampled Direct on Glass
Targeted MS/MS experiment looking at major fragment ions, with direct sampling from merlot wine spiked with 4-EP and 4-EG
δ mass accuracy: 1.2 ppm
δ mass accuracy: 1.6 ppm
Negative Ion Mode
!
4-EP
4-EG
Gerstel Twister Stir Bar: Sample Concentration
Twister was spun overnight in orange juice with 50 ppb
spiking level of 10 pesticides
Real Wine Sample - Direct Twister Preliminary Results - No Glass Chamber
DART_tMS2_Wine_4-EP&4-EG_Wine without... 1/25/2013 4:20:00 PM 4-EP 45NCE;4-EG 35NCE Blue 3; 300 CGerstel Twisters 24hr spin; He; UHV 3.45e-9RT: 0.59 - 2.70
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
Time (min)
0
20
40
60
80
100
0
20
40
60
80
100
Re
lative
Ab
un
da
nce
2.352.04
2.31
2.142.151.91
1.84 2.291.831.99 2.382.06
1.821.120.98 1.15 1.310.750.68 1.65 2.572.201.440.86 2.421.59
2.32 2.35
2.04
1.912.36
2.151.99
1.84 2.112.291.83 2.38
0.97 1.11 1.15 1.270.770.68 1.761.691.35 2.24 2.46 2.591.42 1.530.85
NL: 2.20E6
m/z= 106.04061-106.04167 F: FTMS - p NSI Full ms2 [email protected] [50.00-145.00] MS DART_tMS2_Wine_4-EP&4-EG_Wine without & with Brett_300C_Twisters_24hr spin_01
NL: 2.55E6
m/z= 136.05114-136.05250 F: FTMS - p NSI Full ms2 [email protected] [50.00-175.00] MS DART_tMS2_Wine_4-EP&4-EG_Wine without & with Brett_300C_Twisters_24hr spin_01
100 105 110 115 120 125 130 135 140 145 150 155 160
m/z
0
20
40
60
80
100
0
20
40
60
80
100
Re
lative
Ab
un
da
nce
121.06469R=47167
106.04118R=49117
108.02049R=50078
136.05193R=44447
152.03443R=42515
122.03618R=45526
117.03532R=48099
108.02048R=48727
NL: 2.49E6
DART_tMS2_Wine_4-EP&4-EG_Wine without & with Brett_300C_Twisters_24hr spin_01#953-981 RT: 2.29-2.35 AV: 15 F: FTMS - p NSI Full ms2 [email protected] [50.00-145.00]
NL: 1.12E6
DART_tMS2_Wine_4-EP&4-EG_Wine without & with Brett_300C_Twisters_24hr spin_01#954-981 RT: 2.29-2.35 AV: 14 F: FTMS - p NSI Full ms2 [email protected] [50.00-175.00]
Wine No “Brett”
Wine With “Brett”
Signal from real wine sample could not be detected directly, need to concentrate analytes.
4-EP
4-EG
Wine Sampled with PDMS Twister
Twister stir bar (PDMS, 0.5 mm thickness) introduced into 2 mL wine
Minimum 30 minutes unattended stirring
Dr. Kathy Loftin http://etd.fcla.edu/CF/CFE0002714/Loftin_Kathleen_B_200908_PhD.pdf
MAJOR IMPROVEMENT Twister placed in glass chamber = COMPLETE desorption
Typical DART-MS Chronogram from Stir Bar Final Results - Glass Chamber Yields
Much Improved Sampling Reproducibility
RT: 0.00 - 5.60
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Time (min)
0
50
100
0
50
100
0
50
100
Rela
tive A
bund
ance 0
50
1002.04
3.312.361.601.15 3.62
2.392.06 2.70 2.91 3.441.18 3.75
3.46 3.783.152.732.41
1.98
4.55 5.314.880.63
3.913.493.182.862.542.121.90
1.571.23
NL: 2.04E4
m/z= 106.04135-106.04347 F: FTMS - p NSI Full ms2 [email protected] [95.00-145.00] MS Twister_WineStdCurve_25ppb_Bar1_12hrspin_25MAR13_Run01
NL: 5.78E4
m/z= 110.06642-110.06862 F: FTMS - p NSI Full ms2 [email protected] [95.00-145.00] MS Twister_WineStdCurve_25ppb_Bar1_12hrspin_25MAR13_Run01
NL: 3.59E4
m/z= 136.05162-136.05434 F: FTMS - p NSI Full ms2 [email protected] [95.00-175.00] MS Twister_WineStdCurve_25ppb_Bar1_12hrspin_25MAR13_Run01
NL: 3.38E5
m/z= 141.08450-141.08732 F: FTMS - p NSI Full ms2 [email protected] [95.00-180.00] MS Twister_WineStdCurve_25ppb_Bar1_12hrspin_25MAR13_Run01
Major improvement using glass chamber for even desorption from a single Twister stir bar
25 ppb Spiking Level of Phenolics
4-EP
4-EP-d4
4-EG
4-EG-d5
Twister Stir Bar: Quantitative Results
y = 0.0025x + 0.0165 R² = 0.9990
0
0.5
1
1.5
2
2.5
0 100 200 300 400 500 600 700 800
Avg
. No
rmal
ized
Pea
k A
rea
Spiking Level (µg/mL)
Merlot Spiked with 4-Ethylguaiacol (4-EG)
y = 0.0027x + 0.0576 R² = 0.9998
0
0.5
1
1.5
2
2.5
0 100 200 300 400 500 600 700 800
Avg
. No
rmal
ized
Pea
k A
rea
Spiking Level (µg/mL)
Merlot Spiked with 4-Ethylphenol (4-EP)
10 – 750 µg/L
10 – 750 µg/L
4-EP Conc.
(µg/L)
Low High
50 500
1 47.7 492.2
2 58.7 509.2
3 52.1 N/A
4 41.0 N/A
Mean 49.9 500.7
%CV 14.9 2.4
%Bias -0.3 0.1
n 4 2
DART MS/MS Method: Figures of Merit
Wine Sample
DART HRAM MS/MS GC MS
4-EP (µg/L)
4-EG (µg/L)
4-EP (µg/L)
4-EG (µg/L)
Sample 04 854 * 197 845 203
Sample 05 518 157 563 161
Sample 06 52 ND 129 14
Sample 09 ND ND 110 13
Sample 14 2774 * 492 2534 433
Yellow = Brettanomyces
Compare: Calculated Levels of 4-EP & 4-EG
* Levels above selected calibration range
Twister: Effect of Time on Concentration
Wine Sample 4-EP
(µg/L) 4-EG
(µg/L)
Sample 44 (30 min) 3775* 287
Sample 44 (1 hour) 3276* 295
Sample 44 (12 hour) 3587* 309
RSD % 7.0 3.6
Sample 19 (30 min) 2230* 450
Sample 19 (12 hour) 2342* 440
RSD % 3.5 1.6
Sample 36 (30 min) 1443* 289
Sample 36 (12 hour) 1277* 296
RSD % 8.6 1.7
* Levels above selected calibration range
• Remarkable correlation between classic GC-MS method and novel DART-MS/MS method
• Effective and labor-free and solvent-free analyte concentration onto the Gerstel Twister sorptive stir bars • Direct screening of volatile phenolic compounds at low levels (10 - 50 ppb
range) with 3 minute sample analysis time with DART ionization
• The stir bar DART HRAM MS/MS method is quantitative over the targeted concentration range with good precision and accuracy • Limits of quantification between 10 - 50 ppb
• Minimized analyst interaction with the samples with DART method: • Reduces analyst error • Increases productivity and sample throughput • Real-time monitoring = quick response to changes during wine
fermentation to prevent wine spoilage (minimize revenue loss)
• Future work: Apply this method for broader phenolic characterization of wines for rapid wine characteristic fingerprinting
Summary
Haunschmidt, M. et al. Determination of organic UV filters in water by stir bar sorptive extraction and direct analysis in real-time mass spectrometry. Anal. Bioanal. Chem., 2010, 397(1): 269-275
Questions?
Email: [email protected]
Many thanks to…
Thermo Fisher Scientific • Dr. Catharina Crone • Dr. Markus Kellmann • Dr. Tabiwang Arrey
Recent Publications • Direct analysis in real time mass spectrometry and multivariate data analysis: A novel approach to rapid
identification of analytical markers for quality control of traditional Chinese medicine preparation. Zeng S, Wang L,
Chen T, Wang Y, Mo H and Qu H. Analytica Chimica Acta 733:38–47, 2012
• Rapid detection of alkaloids in Ipecac by direct analysis in real time tandem mass spectrometry (DART-MS/MS).
Sun L, Hu X, Liu L, Jin H and Lin R. Zhongguo Zhong Yao Za Zhi 37:1426–30, 2012 (Article in Chinese)
• Metabolomic fingerprinting employing DART-TOF MS for authentication of tomatoes and peppers from organic
and conventional farming. Novotna H, Kmiecik O, Gałązka M, Krtkova V, Hurajová A, Schulzová V, Hallmann E,
Rembiałkowska E and Hajšlova J. Food Additives & Contaminants: Part A 29(9):1335–1346, 2012
• Rapid identification of synthetic cannabinoids in herbal samples via direct analysis in real time mass
spectrometry. Musah R, Domin M, Walling M and Shepard J. Rapid Commun. Mass Spectrom. 26(9):1109–1114,
2012
• Analysis of isoflavones in soybeans employing direct analysis in real-time ionization–high-resolution mass
spectrometry. Lojza J, Cajka T, Schulzova V, Riddellova K and Hajslova J. J. Sep. Science. 35(3):476–481, 2012
• Confined direct analysis in real time ion source and its applications in analysis of volatile organic compounds of
Citrus limon (lemon) and Allium cepa (onion). Li Y. Rapid Commun. in Mass Spectrom. 26(10):1194–1202, 2012
• Chemometric Classification of Morphologically Similar Umbelliferae Medicinal Herbs by DART-TOF-MS
Fingerprint. Lee SM, Kim HJ and Jang YP. Phytochem. Anal. 23(5):508-512, 2012.
http://www.ionsense.com/pdfs/DARTFoodBook18Sep12sm.pdf