Development and Validation of a Liquid Chromatography-Tandem

Oct 15, 2015 - Development and Validation of a Liquid Chromatography-Tandem Mass Spectrometry Method for the Quantitation of Microcystins in Blue-Gree...
0 downloads 14 Views 765KB Size
Subscriber access provided by NEW YORK MED COLL

Article

Development and Validation of a Liquid ChromatographyTandem Mass Spectrometry Method for the Quantitation of Microcystins in Blue-Green Algal Dietary Supplements Christine H. Parker, Whitney L. Stutts, and Stacey L DeGrasse J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.5b04292 • Publication Date (Web): 15 Oct 2015 Downloaded from http://pubs.acs.org on October 22, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Agricultural and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 38

Journal of Agricultural and Food Chemistry

1

Development and Validation of a Liquid

2

Chromatography-Tandem Mass Spectrometry

3

Method for the Quantitation of Microcystins in

4

Blue-Green Algal Dietary Supplements

5

Christine H. Parker,* Whitney L. Stutts, and Stacey L. DeGrasse

6 7

U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition,

8

5100 Paint Branch Parkway, College Park, MD 20740

9 10 11 12 13

*To whom correspondence should be addressed:

14

Phone: (240) 402-2019

15

Fax: (301) 436-1052

16

Email: [email protected]

17

ACS Paragon Plus Environment

1

Journal of Agricultural and Food Chemistry

18 19

Page 2 of 38

ABSTRACT A liquid chromatography-tandem mass spectrometry (LC-MS/MS) method was

20

developed for the simultaneous detection and quantitation of seven microcystin congeners (1–7)

21

and nodularin-R (8) in blue-green algal dietary supplements. Single-laboratory method validation

22

data were collected in four supplement matrices (capsule, liquid, powder, and tablet) fortified at

23

toxin concentrations from 0.25–2.00 µg/g (ppm). Average recoveries and relative standard

24

deviations (RSD) using matrix-corrected solvent calibration curves were 101% (6% RSD) for all

25

congeners and supplements investigated. Limits of detection (0.006–0.028 µg/g) and quantitation

26

(0.018–0.084 µg/g) were sufficient to confirm the presence of microcystin contamination at the

27

Oregon-mandated guidance concentration of 1.0 µg microcystin-LReq/g. Quantitated

28

concentrations of microcystin contamination in market-available Aphanizomenon flos-aquae

29

blue-green algal supplements ranged from 0.18–1.87 µg microcystin-LReq/g for detected

30

congeners microcystin-LR, microcystin-LA, and microcystin-LY (3–5). Microcystin-RR, -YR, -

31

LW, -LF, and nodularin-R (1–2, 6–8) were not detected in the supplements examined.

32 33 34 35 36 37 38 39

KEYWORDS: microcystin, cyanotoxin, blue-green algae, dietary supplements, A. flos-aquae,

40

tandem mass spectrometry

ACS Paragon Plus Environment

2

Page 3 of 38

41 42

Journal of Agricultural and Food Chemistry

INTRODUCTION Cyanobacteria (previously classified as blue-green algae) are a diverse group of photo-

43

autotrophic organisms which are found in terrestrial and aquatic environments. While serving as

44

a nutrient dense resource in many ecosystems, under euphotic conditions massive blooms of

45

toxic cyanobacteria occur in both freshwater and marine environments. Cyanobacterial toxins are

46

structurally classified as cyclic peptides (microcystin and nodularin), alkaloids (anatoxin,

47

saxitoxin, and cylindrospermopsin), and lipopolysaccharides.1,2 Of the cyanobacterial toxins,

48

microcystins are the most abundant known toxins of bloom-forming cyanobacteria.

49

Microcystins pose a major threat to drinking and irrigation water supplies, vegetation,

50

aquatic wildlife, terrestrial animals, and humans that have been in contact with or consume

51

products harvested from contaminated waters. The most severe human exposure to microcystins

52

was reported in 1996, when contaminated water was inadvertently distributed to 126 renal

53

dialysis patients in Brazil, resulting in 60 deaths.3,4 While acute and short-term hepatotoxic

54

effects have been associated with human exposure to microcystin contaminated water, the

55

potential exposure to cyanotoxins through organic cyanobacteria- and algae-derived dietary

56

supplements pose a risk to consumers of blue-green algal supplements.

57

Cyanobacteria- and algae-derived dietary supplements can be divided into three main

58

categories: Aphanizomenon flos-aquae, Spirulina (Arthrospira platensis and Arthrospira

59

maxima), and Chlorella pyrenoidosa products.5,6 Marketed to children and adults, blue-green

60

algal supplements are advertised for treatment of fatigue, anxiety, depression, attention deficit-

61

hyperactivity disorder (ADHD), diabetes, and high cholesterol while supporting weight loss,

62

stimulating immune function, and elevating energy.6–9 Whereas Spirulina was initially

63

commercialized as a non-toxic dietary supplement sourced from constructed ponds, A. flos-

ACS Paragon Plus Environment

3

Journal of Agricultural and Food Chemistry

64

aquae entered the market as an organic cyanobacterial source harvested from natural lakes. The

65

remarkably stable abundance and highly available biomass of A. flos-aquae in Upper Klamath

66

Lake, Klamath Falls, Oregon renders this natural lake one of the largest viable commercial

67

sources for harvesting of cyanobacterial-derived dietary supplements.

68

Page 4 of 38

Microcystins are characterized by a common chemical structure containing three D-amino

69

acids [D-glutamic acid, D-alanine, and D-MeAsp (D-erythro-β-methylaspartic acid)], two non-

70

proteinogenic amino acids [Mdha (N-methyldehydroalanine) and Adda (2S,3S,8S,9S-3-amino-9-

71

methoxy-2,6,8-trimethyl-10-phenyldeca-4E,6E-dienoic acid)], and two variable L-amino acids.10

72

A generalized chemical structure is shown in Figure 1A where structural variants are named

73

according to the one-letter abbreviation of the two variable amino acids at positions R1 and R2.

74

Among the 94 variants of microcystins that have been reported,11 microcystin-RR, microcystin-

75

YR, and microcystin-LR (1–3) are the most common in cyanobacterial blooms worldwide.12,13

76

While extensive toxicological data are available for microcystin-LR (3), the relative potencies of

77

cyanotoxin congeners are primarily dependent on L-amino acid substituted chemical variants.14

78

Similar in structure to microcystin, the cyclic pentapeptide nodularin-R is commonly isolated

79

from the cyanobacterium Nodularia spumigena15,16 and most abundant in cyanobacterial blooms

80

from brackish water bodies in the Baltic Sea and estuaries of Australia.17,18 Nodularin-R is

81

characterized by a chemical structure of D-glutamic acid, N-methyldehydrobutyrine (Mdhb), D-

82

MeAsp, Adda, and L-arginine (R2) (Figure 1B).

83

To date, all reported tests of algal dietary supplements produced from Klamath Lake have

84

failed to detect the presence of anatoxin, cylindrospermopsin, nodularin-R, and saxitoxin

85

cyanotoxins.19–23 The risk of chronic exposure to microcystins, however, led the Oregon Health

86

Division and the Oregon Department of Agriculture to establish a regulatory limit of 1 µg

ACS Paragon Plus Environment

4

Page 5 of 38

Journal of Agricultural and Food Chemistry

87

microcystin-LReq/g for microcystins in blue-green algal products.24 Health Canada followed in

88

1999, releasing a precautionary advisory statement recommending the discontinued consumption

89

of blue-green algal supplements for children. Currently, the levels of algal toxins in food

90

supplements are unregulated at the federal level in the United States. The U.S. Environmental

91

Protection Agency (EPA-820R15102) and Harmful Algal Bloom and Hypoxia Research and

92

Control Amendments Act of 2014, however, have emphasized a need for research-driven action

93

strategies to mitigate stakeholder response to freshwater harmful algal blooms and safeguard

94

consumers from potential exposure to toxin contaminated sources.

95

Current methods for microcystin detection can be characterized into approaches for

96

screening and quantitation including: enzyme-linked immunosorbent assays (ELISA),25–26

97

protein phosphatase inhibition assays,27,28 liquid chromatography (LC) combined with ultraviolet

98

(UV)29,30 or mass spectrometric (MS) detection,31–33 polymerase chain reaction (PCR)

99

taxonomic-based assays,34–35and surface plasmon resonance (SPR) biosensors.36–37 Although

100

biochemical and physiochemical methods for microcystin detection are suitable for general

101

monitoring purposes, limitations in sensitivity and specificity constrain quantitation of individual

102

structural variants. LC-MS-based approaches provide a powerful technology capable of meeting

103

required needs for sensitivity while allowing simultaneous quantitation and structural

104

characterization of multiple microcystin analogs. The contamination of A. flos-aquae-based blue-

105

green algal dietary supplements with microcystins, primarily microcystin-LR (3) and

106

microcystin-LA (4), has been confirmed for products surveyed worldwide with concentrations

107

ranging from 95% by high performance

126

liquid chromatography (HPLC). Leucine enkephalin (Protea Biosciences, Morgantown, WV) and

127

angiotensin I (Sigma Aldrich, St. Louis, MO) were reconstituted to stock concentrations of

128

9 fmol/µL and 100 pmol/µL, respectively, and added to each sample prior to injection as a

129

quality control standard and background matrix surrogate, respectively. Optima grade solvents

130

for sample reconstitution, sample pretreatment, and LC analysis were purchased from Fisher

131

Scientific (Pittsburg, PA). All sample preparation was performed using Eppendorf Protein

132

LoBind microcentrifuge tubes (Fisher Scientific). LC-MS certified clear glass total recovery

ACS Paragon Plus Environment

6

Page 7 of 38

Journal of Agricultural and Food Chemistry

133

vials with presplit PTFE/silicone septa (Waters Corporation, Manchester, UK) were utilized as

134

sample injection vials for all samples.

135

Calibration Standard Preparation

136

Lyophilized microcystin and nodularin-R standards were reconstituted to a stock

137

concentration of 10 µg/mL in methanol. Sample dilutions were made with calibrated pipettes in

138

80% methanol, 20% water, 9 fmol/µL leucine enkephalin, and 1 pmol/µL angiotensin I. To

139

prevent solvent evaporation and sample concentration, exposure of the solutions to air was

140

minimized. Standard solutions were stored in the dark at −20 oC or colder (up to six months).

141

Neat calibration solutions were analyzed in a randomized order of injection over the range of

142

1.5–197 pg/µL (1.5, 3.1, 6.2, 12.3, 24.6, 49.3, 98.5, and 197 pg/µL). The 12.3 pg/µL (0.19 µg/g)

143

standard was evaluated in 3 h intervals for quality control performance.

144

Preparation of Extracts from Blue-Green Algal Dietary Supplements

145

Five commercially purchased blue-green algal dietary supplements were investigated for

146

total microcystin content. Supplements, with corresponding daily serving quantities and

147

excipients in parenthesis, included: A. flos-aquae capsule (1 g; plant cellulose and water), A. flos-

148

aquae liquid (1 Tbsp), A. flos-aquae powder (1 g), A. flos-aquae Lot A and Lot B tablet (1 g;

149

microcrystalline cellulose), and Spirulina powder (7 g). Each homogenized powdered

150

supplement (0.100 + 0.004 g) was weighed into a 2.0 mL microcentrifuge tube. Twelve tablets

151

(500 mg each) from each sample lot (60 count) were ground into a fine homogenized powder in

152

a 50 mL Nalgene high-speed centrifuge tube with polypropylene screw cap (Fisher Scientific)

153

using a 3/8” stainless steel grinding ball (SPEX, Metuchen, NJ) and SPEX Sample Prep 2010

154

Geno/Grinder (1400 rpm × 1 min × 5 cycles and 1500 rpm × 1 min × 1 cycle). Similarly,

155

powdered contents from twelve capsules (500 mg each) were emptied and homogenized into a

ACS Paragon Plus Environment

7

Journal of Agricultural and Food Chemistry

Page 8 of 38

156

uniform mixture (500 count). Aliquots (1 mL) of the mixed thawed liquid supplement (16 oz)

157

were dried by vacuum centrifugation for 12–16 h to an equivalent dry weight of 0.050 + 0.010 g.

158

The liquid supplement was stored at 2–8 oC for 7–10 d according to manufacturer

159

recommendations.

160

Weighed powdered supplements were extracted in 1.0 mL of 80% methanol, 20% water

161

[1:10 supplement to solvent (w/v) extraction]. Samples were vortex mixed (1400 rpm) using an

162

Eppendorf ThermoMixer for 15 min at room temperature (23 oC), rotated end-over-end (0.23 ×

163

g) at room temperature, and centrifuged at 14,000 × g for 10 min at 20 oC. The supernatant was

164

transferred to a new 1.5 mL microcentrifuge tube. Liquid supplement sample extractions were

165

performed in a total volume of 500 µL to maintain a 1:10 (w/v) ratio.

166

Five independent sample extractions were performed for each matrix source. Fortified matrix

167

samples were prepared at 0.25, 0.50, 1.00, and 2.00 µg/g concentrations for each analyte by

168

spiking equivalent volumes of a neat cyanotoxin standard solution into each matrix prior to

169

extraction. Fortified concentrations were based upon the Oregon established regulatory limit of

170

1 µg microcystin-LReq/g for products containing blue-green algae.24

171

Sample Pretreatment of Blue-Green Algal Extracts

172

Sample extract pretreatment procedures were evaluated to optimize sample recovery

173

without analyte bias and to limit matrix interference. The following sorbents were assessed in

174

method development: Phenomenex (Torrance, CA) Strata C18-E SPE (1 mL; 50 mg sorbent

175

mass), Phenomenex Strata-X Polymeric SPE (1 mL; 30 mg sorbent mass), and Pierce (Thermo

176

Fisher Scientific, Waltham, MA) Graphite Spin Columns (500 µL; 10 mg sorbent mass).

177

Preparatory columns were used according to manufacturer recommendations with slight

178

modifications to accommodate sample extract conditions and optimize sample recovery.

ACS Paragon Plus Environment

8

Page 9 of 38

Journal of Agricultural and Food Chemistry

179

Phenomenex SPE cartridges were activated in 100% methanol (2 × 1 mL) and equilibrated in

180

94.5% water, 5% methanol, and 0.5% trifluoroacetic acid (TFA) (2 × 1 mL). Sample extracts

181

(400 µL) were combined 1:4 (v/v) with 100% water (1600 µL) to facilitate analyte binding in

182

SPE. Diluted sample extracts were added to the SPE column in two aliquots (1000 µL each) and

183

passed through the sorbent a second time to enhance binding capacity. The column was washed

184

with 94.5 % water, 0.5%TFA (2 × 1 mL) and sample eluted in 80% methanol, 19.9% water, and

185

0.1% formic acid (2 × 100 µL). Pierce graphite spin columns were utilized as filters for pigment

186

and auxiliary contaminant removal. Columns were activated and equilibrated according to

187

manufacturer recommendations. Sample extracts (100 µL) were applied to the prepared resin bed

188

and centrifuged at 1000 × g for 3 min without incubation or vortex mixing. The graphite resin

189

was washed with a 50 µL aliquot of 80% methanol, 20% water and centrifuged for an additional

190

3 min at 1000 × g.

191

Cyanotoxin Infusion and Fragmentation Identification

192

Infusions of cyanotoxin standards were performed on a hybrid LTQ-Orbitrap Elite

193

(Thermo Scientific, San Jose, CA) with a Digital PicoView (New Objective, Woburn, MA)

194

nanospray source operated at a flow rate of 300 nL/min. Individual analyte solutions were

195

prepared in 50% methanol, 49.9% water, and 0.1% formic acid. The purity of individual

196

microcystin standards was evaluated by full scan MS over the mass range m/z 400–1250. Higher-

197

energy collisional dissociation (HCD) spectra were acquired for each microcystin congener and

198

nodularin-R at 10 microscans/spectrum with a maximum injection time of 100 ms. Each high-

199

resolution Fourier transform (FT) MS/MS spectrum was collected over a scan range m/z 100–

200

1250 for a specified precursor mass (isolation width m/z 2.0) at a resolving power of 120,000.

201

Normalized collision energies (activation time 0.100 ms) were optimized for each analyte and

ACS Paragon Plus Environment

9

Journal of Agricultural and Food Chemistry

202

empirically adjusted based upon precursor charge so as to assign primary structural

203

fragmentation while limiting pathways of secondary fragmentation. An automatic gain control

204

(AGC) target allowed accumulation of up to 5 × 104 ions for FT MS/MS scans.

205

Page 10 of 38

In FT MS3 scans, precursor masses were first isolated for ion trap collision-induced

206

dissociation (CID) at a precursor ion isolation width of m/z 2.0, using an AGC target of 1 × 104,

207

10 microscans per spectrum, and a maximum ion accumulation time of 100 ms. Fragmentation

208

was performed with experimentally determined normalized collision energies and an activation

209

time of 10 ms (Qact = 0.250). Directly following each MS/MS experiment, a user-specified

210

product ion was isolated and fragmented by HCD (isolation width m/z 2.0; activation time

211

0.100 ms) to obtain a MS3 spectrum over the scan range m/z 100–1000.

212

Instrumental Analysis for Quantitation

213

An ACQUITY UPLC system with 150 mm × 1 mm i.d., 1.7 µm (130 Å), ACQUITY

214

UPLC C18 BEH analytical column (Waters Corporation) was used for reverse phase separation

215

at 40 oC. Column flow rate was maintained at 50 µL/min. Mobile phase A was prepared with

216

0.1% (v/v) formic acid in water and mobile phase B with 0.1% (v/v) formic acid in acetonitrile.

217

Weak needle wash and strong needle wash solvent compositions matched that of mobile phase A

218

and B, respectively. Sample injections were made in partial loop mode at 2 µL volumes (10 µL

219

sample loop) and the autosampler temperature was thermostated to 8 oC.

220

Cyanotoxin congeners were eluted with a step gradient of 35–45% B in 5 min and 45–

221

75% B from 5 to 6 min. The gradient was ramped to 90% B and re-equilibrated at initial starting

222

conditions for a total run time of 14 min. Mass spectrometric analyses were accomplished using

223

a 5500 QTRAP (AB Sciex, Framingham, MA) operated in positive ionization mode. Turbo V

224

ion source parameters (ion spray voltage, source temperature, gas flows) were optimized

ACS Paragon Plus Environment

10

Page 11 of 38

Journal of Agricultural and Food Chemistry

225

collectively for all analytes under the chromatographic conditions specified above. The curtain

226

gas was set at 20 au, CAD gas at High (12 au), ion spray voltage at 5000 V, source temperature

227

at 400 oC, gas 1 pressure at 40 au, gas 2 pressure at 30 au, and entrance potential at 10 V. Using a

228

syringe pump for infusion, compound dependent parameters (declustering potential, collision

229

energy, and collision cell exit potential) were optimized for solvent standards of each analyte at a

230

flow rate of 7 µL/min. Consensus values for declustering potential (80 V), entrance potential

231

(10 V), and collision cell exit potential (18 V) were determined for each analyte. Individual

232

compound transitions are shown in Table 1 with corresponding retention times, optimized

233

collision energy voltages, and approximate analyte relative abundances. Retention times for the

234

target compounds were determined by analyzing a mixed solution of cyanotoxin standards, under

235

the conditions described above, without scheduling. The MS/MS data for all validation samples

236

were collected in scheduled multiple reaction monitoring (MRM) mode with low resolution for

237

Q1, unit resolution for Q3, a 5 ms pause between mass ranges, a MRM detection window of

238

120 s and a targeted scan time of 1 s. Quantitative data analysis was performed using Skyline v

239

2.6.39

240 241

RESULTS AND DISCUSSION

242

Optimization of Sample Pretreatment

243

Dietary supplement extractions were based upon previous literature reports to optimize

244

conditions for microcystin recovery.19,40,41 Representative A. flos-aquae-based (capsule, liquid,

245

powder, and tablet) and Spirulina-based (powder) blue-green algal dietary supplements were

246

selected for method development. Spirulina, commercialized as a non-toxic algal supplement,

247

was evaluated in the context of this work as a potential matrix reference blank for quantitation.

ACS Paragon Plus Environment

11

Journal of Agricultural and Food Chemistry

Page 12 of 38

248

Extraction efficiencies were evaluated for various supplement-solvent ratios, methanol-water

249

ratios, acidities, mechanical agitations, and sequential re-extractions. A 1:10 dietary supplement

250

to solvent (w/v) extraction with 80% methanol, 20% water using gentle agitation (vortex and

251

rotation) was selected based upon method efficacy and recovery. To minimize sample-surface

252

contact and facilitate the provisioning of numerous fortified samples, a 1.0 mL sample extraction

253

volume was prepared for each 100 mg sample of homogenized supplement.

254

The stability of sample extracts was evaluated at varying temperatures (−20 oC and 4 oC)

255

and methanol-water ratios (8% methanol and 80% methanol) over the course of five consecutive

256

days. Sample stability was maximized for samples stored in sealed, light-shielded vials at −20 oC

257

and freshly aliquoted prior to LC-MS analysis. The effect of methanol concentration was

258

demonstrated to have a noteworthy influence on analyte recovery with an average increase in

259

peak area of 77 + 3% for samples prepared and injected in 80% methanol.

260

Solid-phase sorbents were investigated for contaminant removal from supplement

261

extracts. Analyte recoveries were compared between pre- and post-fortified sample extracts for a

262

C18 silica-based SPE cartridge, a polymeric SPE cartridge, and a graphitized carbon spin

263

column. C18 sorbents are traditionally employed in the literature for the concentration of

264

microcystins from various aqueous matrices of contaminated water, fish, and blue-green algal

265

dietary supplements.19,31,42,43 The C18 silica- and polymeric-based SPE sorbents, however,

266

yielded reduced recoveries (