Fast Surface Acoustic Wave-Matrix-Assisted Laser Desorption

Feb 5, 2013 - Lund University, Center for Molecular Protein Science, Biochemistry and Structural Biology, Box 124, 221 00 Lund, Sweden. §. Lund Unive...
0 downloads 0 Views 1MB Size
Subscriber access provided by UNIV PRINCE EDWARD ISLAND

Article

Fast SAW-MALDI MS of Cell Response from Islets of Langerhans Loreta Bllaci, Sven Kjellstrom, Lena Eliasson, James R. Friend, Leslie Y. Yeo, and Staffan Nilsson Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/ac3019125 • Publication Date (Web): 05 Feb 2013 Downloaded from http://pubs.acs.org on February 20, 2013

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.

Analytical 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 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

Fast SAW-MALDI MS of Cell Response from Islets

2

of Langerhans

3

Loreta Bllaci†, Sven Kjellström‡, Lena Eliasson ¥, James R. Friend§, Leslie Y. Yeo§, Staffan

4

Nilsson†*

5

†Lund University, Center for Chemistry and Chemical Engineering, Pure and Applied

6

Biochemistry, Box 124, 221 00 Lund, Sweden

7

‡Lund University, Center for Molecular Protein Science, Biochemistry and Structural Biology,

8

Box 124, 221 00 Lund, Sweden.

9

¥ Lund University, Diabetes Centre, Clinical Sciences Malmö, CRC 91-11, SE-205 02 Malmö,

10

Sweden.

11

§ RMIT University, Micro/Nanophysics Research Laboratory, School of Electrical & Computer

12

Engineering, Melbourne, VIC, 3000Australia.

13

KEYWORDS

14

Open chip, Surface Acoustic Wave atomizer, MALDI MS, Rapid cell releasate profiling

15 16

ACS Paragon Plus Environment

1

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

17

Page 2 of 24

ABSTRACT

18

A desire for higher speed and performance in molecular profiling analysis at a reduced cost is

19

driving a trend in miniaturization and simplification of procedures. Here we report the use of a

20

Surface Acoustic Wave (SAW) atomizer for fast sample handling in MALDI MS peptide and

21

protein profiling of Islets of Langerhans, for future Type 2 Diabetes (T2D) studies. Here the

22

SAW atomizer was used for ultrasound (acoustic) extraction of insulin and other peptide

23

hormones released from freshly prepared islets, stimulated directly on a membrane. A high

24

energy propagating SAW atomizes the membrane-bound liquid into approximately 2µm

25

diameter droplets, rich in cell-released molecules. Besides acting as a sample carrier, the

26

membrane provides a purification step by entrapping cell clusters and other impurities within its

27

fibers. A new SAW-based sample-matrix deposition method for MALDI MS was developed,

28

characterized by a strong insulin signal and a limit of detection (LOD) lower than 100 attomoles,

29

was achieved. Our results support previous work reporting the SAW atomizer as a fast and

30

inexpensive tool for ultrasound, membrane-based sample extraction.. When interfaced with

31

MALDI MS, the SAW-atomizer constitutes a valuable tool for rapid cell studies. Other

32

biomedical applications of SAW-MALDI MS are currently being developed, aiming at fast

33

profiling of biofluids. The membrane sampling is a simplistic and noninvasive collection method

34

of limited volume biofluids such as the gingival fluid and the tearfilm.

35 36

INTRODUCTION

37

Over the years, bio-analysis has continually benefited at various levels from miniaturization,

38

smaller sampling volumes1 and easier handling to dramatically shorten processing time. MALDI

39

MS, an indispensable tool in protein analysis,2-4 is usually preceded by multiple steps of sample

ACS Paragon Plus Environment

2

Page 3 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

40

pretreatment and purification which not only are time consuming but might also lead to MALDI

41

artifacts.5 Here we introduce the Surface Acoustic Wave (SAW) atomizer, a tool for fast

42

ultrasound e.g. “mechanical” sample extraction6 from living cell samples in MALDI MS

43

analysis.

44

The atomizer consists of a low-loss piezoelectric substrate where the sample droplet is placed

45

(Figure 1A) with two single-phase unidirectional transducers SPUDTs electrodes (Figure 1B)

46

driven at 30 MHz in this study by an electrical supply.6-8 When the Rayleigh SAW propagates

47

under the membrane-bound sample, SAW energy is transferred into the membrane (Figure 1C)

48

and capillary waves7 are generated in the fluid on the sample's surface. Due to the tremendous

49

accelerations induced by the 30 MHz frequency SAW atomizer, the capillary waves destabilize

50

and ejects droplets from its crests; the aerosol generated possesses a nearly monodisperse

51

diameter distribution centered at a value in the range of 2µm that depends on the fluid’s viscosity

52

and surface tension.7 Cavitation is absent due to the fact that the propensity of cavitation

53

nucleation is inversely dependent on the frequency, squared6-8 and at the frequencies used in the

54

SAW, the power required is at least three orders of magnitude greater than used for atomization.

55

The effect of fluid shear is minimized due to the short exposure time of only a few microseconds

56

and the fact that the time scale of SAW-driven oscillation is far away from the typical resonant

57

time scales of molecules that might incur shear-induced damage. An insignificant degradation of

58

less than 1% was reported for SAW atomizer-extracted Ovalbumin and BSA for SDS-PAGE

59

analysis.6 The open format of the SAW-atomizer,9 circumvents the use of sample transferring

60

micro-channels,10, 11 and related clogging risk. Finally, SAW atomization is simple; is essentially

61

instant-on, instant-off technology; economical with power consumption of less than 3W; and

62

requires minimal personnel training. It is but one example of the burgeoning field of fluid and

ACS Paragon Plus Environment

3

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 24

63

particle manipulation using acoustics that is solving a broad array of problems in microfluidics.6,

64

8

65 66

Figure 1. The SAW atomizer’s working principle and experimental setup. The atomizer consists

67

of a piezoelectric material with two SPUDT electrodes which generate unidirectional Rayleigh-

68

wave SAW that propagate towards and underneath the sample. (A) A resting droplet on the SAW

69

atomizer, on which no SAW excitation is present. (B) The SAW has reached the droplet,

70

destabilized and generated liquid aerosol with a centered monodisperse diameter of 2 µm. (C)

71

High energy SAW drive atomization even when the liquid is previously sampled on a membrane.

72

The resulting aerosol (B and C) are collected on a MALDI plate via a pinhole for MS analysis.

73

(D) Frontal image of the SAW atomizer (above); lateral image of the working atomizer (below).

ACS Paragon Plus Environment

4

Page 5 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

74

In combination with simple filter paper or with a purpose-selected membrane, upon application

75

of the living cell sample, the SAW atomizer offers a purification and extraction system;

76

impurities such as cell debris or cell clusters are likely to get entrapped within the membrane

77

fibers. Successful SAW membrane-extraction of low molecular weight analytes from blood and

78

water have already been demonstrated by interfacing SAW with ESI-MS.8,

79

better suited for direct analysis of complex biological samples due to the relatively high tolerance

80

of impurities and simplified spectra mostly consisting of single protonated species.13

81

Nevertheless, multiple experimental factors determine the spectral quality, of which, the sample

82

and matrix deposition on the MALDI plate is essential.14, 15 Ideally, the sample-matrix layer for

83

analysis contains small, homogeneous and equally distributed matrix crystals embedding the

84

sample analytes. Several deposition methods to date suffer from uneven analyte distribution

85

within the MALDI spot,16 therefore manually searching for "hot” spots—regions of the

86

preparation that yield higher signal intensities—is required. Thin matrix-sample layers are

87

reported to ionize easily and lead to high signal intensities.17-19 The SAW-membrane platform is

88

here used in fast screening of islet hormone secretion for future Type 2 Diabetes (T2D) studies in

89

parallel to its use as a tool for thin layer MALDI sample preparation.

12

MALDI MS is

90

The disease arises from the interplay of several factors of which imbalance of pancreatic

91

hormones, responsible for carbohydrate, fat, and protein metabolism regulation is a major

92

contributor. Pancreatic hormones are produced from cell clusters named islets of Langerhans

93

(Figure 2). Every islet harbors 1.000-3.000 cells of five different cell types named α, β, δ, PPi,

94

and ε-cells. In human, insulin-releasing β-cells constitute the main population (48-59%) followed

95

by the glucagon-releasing α-cells (33-46%).20, 21 Glucose is the main regulator of both insulin

96

and glucagon release, while hormones22-24 or neurotransmitters like acetylcholine25 modulate the

ACS Paragon Plus Environment

5

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 24

97

response from the islet’s cells. While impaired insulin release is an important factor for the

98

disease onset, T2D patients are characterized by elevated levels of circulating glucagon,22

99

suggesting that α-cells might also play a key role in the T2D pathophysiology. The contribution

100

of other cell types—(δ, PPi, and ε-cells) constituting a minority of the islet mass—in the disease

101

onset is even more obscure and requires further research.

102 103

Figure 2. Illustration of an islet of Langerhans in mice. The β-cells (A), depicted in green,

104

dominate the core of the islet, mostly surrounded by glucagon releasing α-cells (red). The

105

electron microscopy (EM) picture (B) shows one insulin-releasing β-cell. Higher magnification

106

of two adjacent β-cells shows numerous insulin-granules anchored in the plasma membrane. N-

107

nucleus; g-granule; PM-plasma membrane; m-mitochondria. The green line indicates the plasma

108

membrane. Scale bar 2 µm (B) and 0.5 µm (C).

109

Comparative proteomics of healthy versus diseased islets as endocrine units might help

110

identify T2D biomarkers and highlight T2D’s pathogenesis. In our forthcoming studies, we

111

intend to use the SAW atomizer as a tool for quick sample handling in MALDI MS in this work.

112

Besides performing an instant ultrasound-based, membrane purification/extraction of the sample,

113

the chip reduces the use of vials, containers, and related analysis cost.

ACS Paragon Plus Environment

6

Page 7 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

114

EXPERIMENTAL SECTION

115

Reagents. Acetylcholine chloride, ammonium citrate and bovine transferrin were purchased

116

from Sigma-Aldrich (Buchs SG, Switzerland), human insulin from Novo Nordisk, α-cyano-4-

117

hydroxycinnamic acid (CHCA) from Bruker Daltonics. TFA was purchased from Merck. Islets

118

were freshly prepared and supplied from the Clinical Research Center (CRC), Malmö.

119

Quantitative, medium-wide pores (5-6 µm pore size) filter papers, were provided by Munktell

120

Filter AB (Falun, Sweden). Acetonitrile and MilliQ water were of LC gradient grade. RPMI-

121

1640 (Roswell Park Memorial Institute medium) media was purchased from Sigma Aldrich

122

while

123

phosphoethanolamine, triethylammoniumsalt) from Invitrogen UK.

Fluorescein-DHPE(N-(fluorescein-5-thiocarbamoyl)-1,2-dihexadecanoyl-sn-glycero-3-

124

Biological Sample Preparation. Mice used in the experiments were female NMRI

125

(Crl:NMRI(Han) (outbred), 8 weeks old. Approximately 100 islets were isolated from each

126

mouse, allowing for 15 to 20 performed experiments. After isolation, islets were suspended in

127

RPMI 1640 (Roswell Park Memorial Institute medium). Prior to the analysis they were

128

transferred in HEPES-buffered medium, pH 7,4: 25 mM HEPES, 125 mM NaCl, 5.9 mM KCl,

129

12.8 mM CaCl2, 1.2 mM MgCl2, with various concentrations of stimulating glucose content

130

varying from 10 to 30 mM. For the control experiments, the buffer was supplied with 3 mM

131

glucose; a concentration below the insulin-release threshold (7 mM),26 and acetylcholine varying

132

in concentration (0, 1 and 100 µM). Acetylcholine chloride for stimulation of islets and bovine

133

transferrin (used as a matrix additive in the insulin LOD evaluation) were prepared at

134

concentrations of 100, 10 and 5 µM respectively.

135

Stimulation of islets on the atomizer. A volume of 2 µL HEPES buffer, containing typically

136

2-6 islets, was pipetted on a dry, un-pretreated membrane placed on the atomizer. An equivalent

ACS Paragon Plus Environment

7

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 24

137

volume of acetylcholine (aq) was subsequently applied on the islets located onto the membrane.

138

Although not being the main grounds of this work, stimulation of a single islet was performed on

139

the bare chip without employing the membrane. In this case, 1 µL HEPES buffer containing one

140

islet was placed on the chip followed by the addition of 1 µL acetylcholine (aq). Stimulation time

141

with acetylcholine was five minutes in all experiments. Stimulated islets, either on the membrane

142

or directly on the chip, were supplied every 2 minutes with 0.5µL water to compensate for the

143

evaporation.

144

Electron Micrscopy. The samples in the electron micrographs were fixed in 2.5%

145

glutaraldehyde and 1% osmiumtetroxide prior to embedding in AGAR100 as previously

146

described by Andersson et al.27 Images were taken using a JEM 1230 electron microscope.

147 148

Fluorescent Nanoparticles. Lipid-based nanoparticles were prepared as described by Nilsson28 with supplementary fluorescein DHPE.

149

Surface Acoustic Wave atomizer. The SAW atomizer consists of a low-loss piezoelectric,

150

single crystal lithium niobate 127.68° Y-rotated cut, X-propagating material sputtered with two

151

chromium–aluminum single-phase unidirectional transducer (SPUDT). When a sinusoidal

152

electrical signal is applied between the electrodes, Rayleigh surface acoustic waves of a few nm

153

in amplitude are produced and propagate unidirectionally along the X-axis of the substrate,

154

toward and underneath the membrane-bound sample. The transverse component of the SAW is

155

progressively absorbed into the fluid, generating a simple sound wave that propagates at the so-

156

called Rayleigh angle that is only dependent upon the relative difference in sound speed between

157

the SAW in the lithium niobate and the sound wave in the fluid.8 The sound wave propagating in

158

the fluid drives the formation of a capillary wave along the fluid present atop the membrane, and

159

due to the extreme accelerations of over 108 m/s2 present at the fluid interface, droplets are

ACS Paragon Plus Environment

8

Page 9 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

160

driven from the crests of the destabilized capillary wave without cavitation to form a micron-

161

order aerosol.6 The SAW-atomizer was fabricated in house by the Melbourne Centre for

162

Nanofabrication.

163

Matrix preparation. In our experiments, 2-5 mg/mL CHCA was dissolved in 50: 50: 0,1 %

164

with water, acetonitrile and trifluoroacetic acid (TFA) with supplementary 2mM ammonium

165

citrate. In the LOD evaluation experiments, Transferrin (5 µM and 10 µM ) was used as a matrix

166

additive29 for insulin signal enhancement (Figure 7, pink trace).

167

Stainless steel plate mask. A plate mask, set atop a plastic surface possessing a 10 mm

168

diameter opening corresponding to 10 neighbor spots on the MALDI plate, was used to prevent

169

the aerosol’s deposition throughout the plate's area. The membrane size was 5x 8 mm.

170

MALDI MS Instrument. MALDI-MS analyses of the crystallized samples were performed

171

on a 4700 Proteomics Analyzer MALDI-TOF/TOF™ mass spectrometer (Applied Biosystems,

172

Framingham, MA). MS data acquisition (3000 laser shots per spot) was performed in positive

173

linear mode. Stainless steel plates with 192 positions were used for the trapping of SAW-

174

generated aerosol.

175

Sample-matrix deposition method. Sample and matrix were consecutively SAW-extracted

176

from the membrane and deposited onto a controlled area of the MALDI plate. Every SAW

177

extraction/deposition event takes less than 15 seconds, therefore sample and matrix SAW

178

extraction/deposition takes approximately 30 seconds. The sample was prepared as previously

179

described (see, “Stimulation of islets on the atomizer”). A volume of 4 µL CHCA was used for

180

matrix-membrane saturation, following the first event of SAW extraction/deposition of the

181

sample. When a single islet was stimulated on the chip (no membrane used), a volume of 2 µL

182

CHCA was applied on the cleaned atomizer.

ACS Paragon Plus Environment

9

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

183 184

Page 10 of 24

SAW atomizer cleaning. After each experiment, the piezoelectric substrate was cleaned with acetone, isopropanol and rinsed with milliQ water as described by J. Ho.8

185 186

RESULTS AND DISCUSSION

187

Fluorescent microscopy visualization of SAW extracted fluorescent nanoparticles

188

First, a simple approach to demonstrate the SAW operation was carried out through the

189

extraction of fluorescent nanoparticles28. Nanoparticles extracted from the membrane via SAW

190

atomization were collected on a second membrane that was subjected to conventional fluorescent

191

microscopy (FM, Figure 3). The increases in nanoparticle concentration lead to enhanced FM

192

signaling suggesting the role of the SAW in the extraction process.

193

Figure 3. Visualization of SAW-extracted fluorescent nanoparticles atop a collection membrane.

194

Nanoparticles embedded in a membrane placed on the atomizer are extracted via SAW

195

atomization and collected on a second membrane positioned parallel to and above the atomizer,

196

in a fashion similar to the MALDI plate. The collection membrane shows numerous SAW-

197

extracted fluorescent nanoparticles; wherein the brighter spots represent aggregates of the

198

extracted nanoparticles.

199

ACS Paragon Plus Environment

10

Page 11 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

200

SAW-MALDI MS spectra of healthy islet secretion

201

Fast MALDI MS spectra of stimulated, intact islets from healthy mice are shown in Figure 4.

202

Islet’s chemical stimulation (30 mM glucose, 100 µM acetylcholine) has selectively triggered the

203

healthy β–cells to release insulin (Figure 4), therefore the acquired spectra are characterized by

204

intense peaks of insulin (relative intensity). Other co-secreted molecules such as C-peptide and

205

amylin are also detected. Acetylcholine can evoke insulin release under non-stimulatory glucose

206

(3mM) at high concentrations (100µM) only. However, in this case, insulin release is

207

considerably reduced (spectra not shown).

208

The second largest cell community of the islet, α-cells, respond completely opposite to β-cells

209

towards glucose concentrations. While high glucose above 7 mM triggers mouse β-cells to

210

release insulin, glucagon-secretion at these higher concentrations is suppressed.30 Low levels of

211

glucagon in healthy individuals keep the glycemic index in control by suppressing hepatic

212

glucose output. However T2D is characterized by reduced glucagon suppression,22 a subject

213

which has gained high interest in the recent years.

214

Other α-cells released peptides like glicentin-related polypeptide (GRPP), the incretin

215

glucagon-like 1 (GL-1), and δ-cell’s somatostatin-14 peptide (SMS-14) are also indicated in the

216

spectra (Figure 4).

ACS Paragon Plus Environment

11

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 24

217 218 219

Figure 4. SAW-MALDI MS spectra of stimulated islets of healthy mice. (A) Acquired spectra

220

from three islets in 30 mM Glucose HEPES using the SAW-spotting. Single-charged insulin

221

(5807 Da), glucagon (3482 Da), and SMS-14 (1640 Da) peptide are labeled. (B) A magnification

222

of the preceding spectra reveals several peaks, corresponding to single-charged peptides released

223

from α-cells: GRPP (3439 Da), GLP-1(4168 Da), β-cell released C-peptide (3121 Da), amylin

224

(3922 Da), somatostatin (SMS-14, 1640 Da) released from δ-cells. The double charged insulin is

225

also detected (2905 Da).

226 227

Sample-matrix deposition method

ACS Paragon Plus Environment

12

Page 13 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

228

In MALDI MS, informative and qualitative spectra acquisition is greatly depended on the

229

procedure how the sample and matrix is deposited on the MALDI plate.14, 31 Additives to the

230

matrix solution are also important; it was shown previously that the addition of phosphoric acid32

231

or transferrin29 to the matrix solution improves the detectability of intact protein. Our acquired

232

spectra, characterized by high signal intensities and “run to run” reproducibility have been

233

acquired with the “SAW-sampling/SAW-matrixing” on the plate.

234

According to our developed method, the membrane-bound sample is SAW-extracted and

235

deposited on the MALDI plate (Figure 5). In a second step the same volume of matrix is applied

236

on the membrane, subsequently, SAW-extracted and deposited atop the sample film on the

237

MALDI plate. The overall time of SAW sample/matrix deposition takes approximately 30

238

seconds, where every SAW extraction event takes less than 15 seconds.

239 240 241 242 243 244 245 246 247 248 249

Figure 5. Illustration of the SAW-deposition method. (A) The membrane-bound sample is first

250

extracted with SAW atomization and deposited on a controlled area (by a pinhole in the plastic

ACS Paragon Plus Environment

13

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 24

251

mask) of the MALDI plate. (B) Following sample extraction, the matrix solution is applied on

252

the same membrane and subsequently deposited on the plate using the SAW. (C) Depiction of

253

the area where the thin film of SAW-extracted sample and matrix are deposited on the MALDI

254

plate.

255

The SAW-deposition method resulted in thin layer formation of the sample and matrix that

256

ionizes easily and is characterized by high signal intensities.17-19 Spraying sample and matrix on

257

the MALDI plate with ad Electrospray needle and piezoelectric microdispenser33 has also been

258

reported to form homogenous matrix-sample crystals characterized by high signal enhancement

259

and reproducibility.34-37 Besides its obvious advantages, electrospraying the matrix-sample

260

mixture nevertheless has its downsides, including the necessity of extra equipment (e.g.,

261

transformers, etc.) and dangerously high voltages (3-5 kV). Thus, another advantage of the SAW

262

chip emerges in our study as an easy, fast and safe tool for thin matrix-sample crystal layer

263

deposition for MALDI MS.

264

Initially we tested the “Quick & Dirty”, the “Matrix-precoated layers” and the “Fast

265

evaporation” methods of sample/matrix deposition.15 In any method, the matrix: analyte volume

266

ratio38 (under matrix molar excess conditions) is of particular importance. However small matrix

267

pipetted volumes (0.5 µL) inundated the thin film of the SAW-extracted sample leading to loss

268

of signal.

269 270

Reproducibility and LOD

271

The SAW-deposition strategy favors thin sample layer formation, associated with high signal

272

intensities and reproducibility (Figure 6). The acquired MALDI spectra of stimulated intact islets

273

are characterized by high S/N values of single charged insulin.

ACS Paragon Plus Environment

14

Page 15 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

274 275 276 277 278 279 280 281 282 283 284 285

Figure 6. The MALDI MS spectra acquired and shown in this figure, were acquired by

286

stimulation of four Spectra acquired and shown in this figure, were acquired by stimulation of

287

four islets. Each MALDI spectra is characterized by a large peak corresponding to the single-

288

charged mouse insulin (MW 5807 Da) and a small peak of single charged glucagon (MW 3482

289

Da). The double-charged insulin (2905 Da) is also detected. In the inset a magnified view of the

290

islet’s response spectra is shown. Thorough investigation of low abundance peptide and proteins

291

is needed to find new T2D markers resulting in a better understanding of the molecular

292

mechanisms underlying the T2D disease.

293

The lower insulin LOD, estimated by serial dilutions of insulin stock solutions, using the

294

SAW-deposition method was less than 100 attomole, with 2 mg/mL CHCA as a matrix (Figure

295

7). The signal was significantly enhanced, by at least one order of magnitude, when the matrix

296

was premixed with 5 µM transferrin in a 5 to 1 ratio. Our results are in good agreement with

ACS Paragon Plus Environment

15

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 24

297

previous studies suggesting enhanced properties of transferrin when premixed with matrices.29

298

The enhancing properties of transferrin-premixed matrices on insulin and other protein/peptides

299

released by stimulated islets will require further study. 300 301 302 303 304 305 306 307

308

Figure 7. MALDI spectra obtained from SAW-spotting of 100 attomole human insulin (MW

309

5808 Da) without (blue trace) and with transferrin (red trace) in the matrix (2 mg/mL) premixed

310

in 5 to 1 ratio with 5 µM transferrin.

311 312

Islet response sampled on the bare chip

313

The membrane, as mentioned earlier, aids sample pre-purification, “desalting” and thin crystal

314

layer formation on the MALDI plate leading to qualitative spectra. However the atomization

315

could be carried out without the membrane when "pure sample” is used and thus decreasing the

316

risk of membrane sample retention that could be critical for detecting scarce analytes. Figure 8

317

shows the MALDI spectra from a freshly prepared and thoroughly washed single islet stimulated

318

directly on the bare chip. The acquired spectra are characterized by a good baseline; signals

319

corresponding to insulin, C-peptide, glucagon and other small peptides, are also detected.

ACS Paragon Plus Environment

16

Page 17 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

320

321 322

Figure 8. Single islet MALDI spectrum obtained by stimulating (100 µM acetylcholine, 30 mM

323

Glucose, HEPES buffer) the islet, present on the naked chip, e.g. without membrane purification.

324 325

CONCLUSIONS

326

The SAW atomizer can be used as a fast, ultrasound extraction tool for small volume, complex

327

samples from a membrane. We have successfully for the first time (to our knowledge)

328

hyphenated the SAW open chip with MALDI MS. In our experiments it was used for the

329

extraction of islet hormones released upon stimulation with glucose and acetylcholine for future

330

T2D studies. Since the sample here is minimally chemically pretreated, the risk of MALDI

331

artefacts is considerably reduced. Furthermore, we have developed a SAW-based sample-matrix

332

deposition method that generates MALDI spectra characterized by high signal intensities of the

333

analytes with high “run to run” reproducibility. The LOD for insulin was 100 attomole, however

334

when transferrin was premixed with the matrix the signal was enhanced to 10 attomoles.

335

Reduced sample handling analysis times, qualitative MALDI spectra acquisition, and membrane

336

sample preservation reduces the overall cost of biomolecular screening such as comparative MS

337

protein profiling for T2D biomarkers. Future use of SAW-MALDI MS in single cell studies can

ACS Paragon Plus Environment

17

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 24

338

be foreseen. We are currently investigating other applications of SAW-MALDI, such as relative

339

quantification and rapid peptide and protein profiling of membrane preserved, small volume (less

340

than 0.5 µL) of biofluid.

341 342

FIGURES

343

Figure 1. The SAW atomizer’s working principle and experimental setup. The atomizer consists

344

of a piezoelectric material with two SPUDT electrodes which generate unidirectional Rayleigh-

345

wave SAW that propagate towards and underneath the sample. (A) A resting droplet on the SAW

346

atomizer, on which no SAW excitation is present. (B) The SAW has reached the droplet,

347

destabilized and generated liquid aerosol with a centered monodisperse diameter of 2 µm. (C)

348

High energy SAW drive atomization even when the liquid is previously sampled on a membrane.

349

The resulting aerosol (B and C) are collected on a MALDI plate via a pinhole for MS analysis.

350

(D) Frontal image of the SAW atomizer (above); lateral image of the working atomizer (below).

351

Figure 2. Illustration of an islet of Langerhans in mice. The β-cells (A), depicted in green,

352

dominate the core of the islet, mostly surrounded by glucagon releasing α-cells (red). The

353

electron microscopy (EM) picture (B) shows one insulin-releasing β-cell. Higher magnification

354

of two adjacent β-cells shows numerous insulin-granules anchored in the plasma membrane. N-

355

nucleus; g-granule; PM-plasma membrane; m-mitochondria. The green line indicates the plasma

356

membrane. Scale bar 2 µm (B) and 0.5 µm (C).

357

Figure 3. Visualization of SAW-extracted fluorescent nanoparticles atop a collection membrane.

358

Nanoparticles embedded in a membrane placed on the atomizer are extracted via SAW

359

atomization and collected on a second membrane positioned parallel to and above the atomizer,

ACS Paragon Plus Environment

18

Page 19 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

360

in a fashion similar to the MALDI plate. The collection membrane shows numerous SAW-

361

extracted fluorescent nanoparticles; wherein the brighter spots represent aggregates of the

362

extracted nanoparticles.

363

Figure 4. SAW-MALDI MS spectra of stimulated islets of healthy mice. (A) Acquired spectra

364

from three islets in 30 mM Glucose HEPES using the SAW-spotting. Single-charged insulin

365

(5807 Da), glucagon (3482 Da), and SMS-14 (1640 Da) peptide are labeled. (B) A magnification

366

of the preceding spectra reveals several peaks, corresponding to single-charged peptides released

367

from α-cells: GRPP (3439 Da), GLP-1(4168 Da), β-cell released C-peptide (3121 Da), amylin

368

(3922 Da), somatostatin (SMS-14, 1640 Da) released from δ-cells. The double charged insulin is

369

also detected (2905 Da).

370

Figure 5. Illustration of the SAW-deposition method. (A) The membrane-bound sample is first

371

extracted with SAW atomization and deposited on a controlled area (by a pinhole in the plastic

372

mask) of the MALDI plate. (B) Following sample extraction, the matrix solution is applied on

373

the same membrane and subsequently deposited on the plate using the SAW. (C) Depiction of

374

the area where the thin film of SAW-extracted sample and matrix are deposited on the MALDI

375

plate.

376

Figure 6. The MALDI MS spectra acquired and shown in this figure, were acquired by

377

stimulation of four Spectra acquired and shown in this figure, were acquired by stimulation of

378

four islets. Each MALDI spectra is characterized by a large peak corresponding to the single-

379

charged mouse insulin (MW 5807 Da) and a small peak of single charged glucagon (MW 3482

380

Da). The double-charged insulin (2905 Da) is also detected. In the inset a magnified view of the

381

islet’s response spectra is shown. Thorough investigation of low abundance peptide and proteins

ACS Paragon Plus Environment

19

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 24

382

is needed to find new T2D markers resulting in a better understanding of the molecular

383

mechanisms underlying the T2D disease.

384

Figure 7. MALDI spectra obtained from SAW-spotting of 100 attomole human insulin (MW

385

5808 Da) without (blue trace) and with transferrin (red trace) in the matrix (2 mg/mL) premixed

386

in 5 to 1 ratio with 5 µM transferrin.

387

Figure 8. Single islet MALDI spectrum obtained by stimulating (100 µM acetylcholine, 30 mM

388

Glucose, HEPES buffer) the islet, present on the naked chip, e.g. without membrane purification.

389

Corresponding Author

390

Staffan Nilsson.

391

E-mail: [email protected]; Phone +46 46 222 81 77; fax:+46 46 222 46 11

392

Present Addresses

393

†Institute of Biochemistry and Molecular Biology, University of Southern Denmark, Campusvej

394

55, DK-5230 Odense, Denmark

395

Author Contributions

396

The manuscript was written through contributions of all authors. All authors have given approval

397

to the final version of the manuscript.

398

Notes

399

Restricted parts of this work have been presented in MS-Öresund (Sweden, 2011) and

400

Analysdagarna (Sweden, '2012).

401

ACS Paragon Plus Environment

20

Page 21 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

402 403

ACKNOWLEDGMENT

404

The Swedish Research Council (VR no 621-2010-5384 to S. Nilsson). We owe our gratitude to

405

Britt-Marie S Nilsson for islets preparation. LYY and JRF acknowledge support of the SAW

406

atomizer technology by the Australian Research Council (DP0985253, DP1092955, and

407

DP120100013) and the NHMRC for Development Grant 1000513. JRF is grateful to the

408

Melbourne Centre for Nanofabrication for an MCN Tech Fellowship and RMIT University for

409

the Vice-Chancellor’s Senior Research Fellowship. LYY is grateful for an Australian Research

410

Fellowship from the Australian Research Council. LE is a senior researcher at the Swedish

411

Research Council.

412 413

ABBREVIATIONS

414

SAW, surface acoustic wave;T2D, type 2 diabetes; LOD, limit of detection; SPUDT, single-

415

phase unidirectional transducer; CHCA, α-cyano-4-hydroxycinnamic acid; SMS, somatostatin;

416

GLP-1, glucagon-like peptide 1; SMS-14, somatostatin-14; GRPP, glicentin-related polypeptide.

417

Fluorescein-DHPE(N-(fluorescein-5-thiocarbamoyl)-1,2-dihexadecanoyl-sn-glycero-3-

418

phosphoethanolamine, triethylammoniumsalt) RPMI 1640 (Roswell Park Memorial Institute

419

medium).

420 421

REFERENCES

ACS Paragon Plus Environment

21

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

422 423

Page 22 of 24

(1) Santesson, S.; Degerman, E.; Rorsman, P.; Johansson, T.; Lemos, S.; Nilsson, S. Integr. Biol. 2009, 1, 595-601.

424

(2) Karas, M.; Bachmann, D.; Hillenkamp, F. Anal. Chem. 1985, 57, 2935-2939.

425

(3) Karas, M.; Bachmann, D.; Bahr, U.; Hillenkamp, F. Int. J. Mass Spectrom. Ion Processes

426

1987, 78, 53–68.

427

(4) Karas, M; Hillenkamp, F. Anal. Chem. 1988, 60, 2299–2301.

428

(5) Walch, A.; Rauser, S.; Deininger, S. O.; Höfler, H. Histochem. Cell Biol. 2008, 130, 421–

429

434.

430

(6) Qi, A.; Yeo, L. Y.; Friend, J. R.; Ho, J. Lab Chip 2010, 10, 470-476.

431

(7) Qi, A.; Yeo, L. Y.; Friend, J. R. Phys. Fluids 2008, 20, 074103.

432

(8) Ho, J. T.; M. K.; Go, D. B.; Yeo, L. Y.; Friend, J. R.; Chang, H. C. Anal. Chem. 2011, 83,

433

3260–3266.

434

(9) Yeo, L. Y.; Friend, J. R. Biomicrofluidics 2009, 3, 012002.

435

(10) Dishinger, J. F.; Reid, K. R.; Kennedy, R. T. Anal. Chem. 2009, 81, 3119-3127.

436

(11) Lenshof, A.; Magnusson, C.; Laurell, T. Lab Chip 2012, 12, 1210-1233.

437

(12) Heron, S. R.; Wilson, R.; Shaffer, S. A.; Goodlett, D. R.; Cooper, J. M. Anal. Chem. 2010,

438 439 440

82, 3985-3989. (13) Hummon, A. B.; Sweedler, J. V.; Corbin, R. C. TrAC, Trends Anal. Chem. 2003, 22, 515– 521.

441

(14) Cohen, S. L.; Chait, B. T. Anal. Chem. 1996, 68, 31-37.

442

(15) Sigma-Aldrich. AnalytiX 2001, 6, 1-6.

443

(16) Garden, R. W.; Sweedler, J. V. Anal. Chem. 2000, 72, 30-36.

444

(17) Knochenmuss, R. Anal. Chem. 2003, 75, 2199-2207.

ACS Paragon Plus Environment

22

Page 23 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

445

(18) Knochenmuss, R.; Zhigilei, L. V. J. Phys. Chem. 2005, 109, 22947-22957.

446

(19) Knochenmuss, R. Analyst (Cambridge, U. K.) 2006, 131, 966-986.

447

(20) Cabrera, O.; Berman, D. M.; Kenyon, N. S.; Ricordi, C.; Berggren, P.O.; Caicedo, A.

448 449 450 451 452 453 454 455 456 457 458

PNAS 2006, 103, 2334 –2339. (21) Brissova, M.; Fowler, M. J.; Nicholson, W. E.; Chu, A.; Hirshberg, B.; Harlan, D.; Power, A. C. J. Histochem. Cytochem. 2005, 53, 1087-1097. (22) Lund, A.; Vilsbøll, T.; Bagger, J. I.; Holst, J. J.; Knop, F. Am. J. Physiol. Endocrinol. Metab. 2011, 300, 1038-1046. (23) Hansen, K. B.; Vilsbøll, T.; Bagger, J. I.; Holst, J. J.; Knop, F. K. J. Clin. Endocrinol. Metab. 2012, 97, 1363-1370. (24) Knop, F. K.; Aaboe, K.; Vilsbøll, T.; Vølund, A.; Holst, J.; Krarup,T.; Madsbad, S. Obes. Metab. 2012, 14, 500-510. (25) Gromada, J.; Høy, M.; Renström, E.; Bokvist, K.; Eliasson, L.; Göpel, S.; Rorsman, P. J. Physiol. 1999, 518, 745-759.

459

(26) Gilon, P.; Henquin, J. C. Endocr. Rev. 2001, 22, 565–604.

460

(27) Andersson, S. A.; Pedersen, G.; Vikman, J.; Eliasson, L. Pflugers Arch. - Eur. J. Physiol.

461 462 463 464 465 466 467

2011, 462, 443-454. (28) Nilsson, C.; Harwigsson, I.; Becker, K; Bülow, L.; Birnbaum, S.; Nilsson, S. Anal. Chem. 2009, 81, 315-321. (29) Kobayashi, T.; Kawai, H.; Suzuki, T.; Kawanishi, T.; Hayakawa, T. Rapid Commun. Mass Spectrom. 2004, 18, 1156-1160. (30) MacDonald, P. E.; De Marinis, Y. Zh.; Ramracheya, R.; Salehi, A.; Ma, X.; Johnson, P. R. V.; Cox, R.; Eliasson, L.; Rorsman, P. PLoS Biol. 2007, 5, 1236-1247.

ACS Paragon Plus Environment

23

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 24

468

(31) Beavis, R. C.; Chait, B. C. Methods Enzymol. 1996, 270, 519-551.

469

(32) Kjellström, S.; Jensen, O. N. Anal. Chem. 2004, 76, 5109-5117.

470

(33) Miliotis, T.; Kjellstrom, S.; Nilsson, J.; Laurell, T.; Edholm, L. E.; Marko-Varga, G. .

471 472 473 474 475

Rapid Commun. Mass Spectrom. 2002, 16, 117-126. (34) Axelsson, J.; Hoberg, A. M.; Waterson, C.; Myatt, P.; Chield, G. L.; Varney, J.; Haddleton, D. M.; Derrick, J. Rapid Commun. Mass Spectrom. 1997, 11, 209-213. (35) Hensel, R. R.; King, R. C.; Owens, K. G. Rapid Commun. Mass Spectrom. 1997, 11, 1785-1793.

476

(36) Hanton, S.; Clark, P.; Owens, K. J. Am. Soc. Mass Spectrom. 1999, 10, 104-111.

477

(37) Yao, J.; Scott, J. R.; Young, M. K.; Wilkins, C. L. J. Am. Soc. Mass Spectrom. 1998, 9,

478 479 480

805-813. (38) Duncan, M. W.; Roder, H.; Hunusucker, S. W. Brief. Funct. Genomic. Proteomic. 2008, 7, 355-370.

481 482

ACS Paragon Plus Environment

24