Electromembrane Extraction of Unconjugated Fluorescein

Subscriber access provided by UNIV OF LOUISIANA

Article

Electromembrane extraction of unconjugated fluorescein isothiocyanate from solutions of labeled proteins prior to flow induced dispersion analysis Magnus Saed Restan, Morten Enghave Pedersen, Henrik Jensen, and Stig Pedersen-Bjergaard Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.9b00730 • Publication Date (Web): 30 Apr 2019 Downloaded from http://pubs.acs.org on April 30, 2019

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 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 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.

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 19 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

2 3 4 5

Manuscript for Analytical Chemistry

Electromembrane extraction of unconjugated fluorescein isothiocyanate from solutions of labeled proteins prior to flow induced dispersion analysis

6 7 8 9 10 11 12

Magnus Saed Restana, Morten E. Pedersenb , Henrik Jensenb,c, Stig Pedersen-Bjergaarda,c aDepartment

of Pharmacy, University of Oslo, P.O. Box 1068 Blindern, 0316 Oslo, Norway ApS, c/o University of Copenhagen, Universitetsparken 2 2100 Copenhagen, Denmark cDepartment of Pharmacy, Faculty of Health and Medical Sciences, University of Copenhagen, Universitetsparken 2, 2100 Copenhagen, Denmark

bFIDA-Tech

13 14 15

Abstract

16

In this initial research on feasibility, removal of unconjugated fluorescein isothiocyanate (FITC)

17

after fluorescent labeling of human serum albumin (HSA) with electromembrane extraction (EME)

18

was investigated for the first time. A 100 µL solution of 0.1 mg/mL HSA was fluorescently labeled

19

with 0.01 mg/mL FITC in a molar ratio of 10:1 in an Eppendorf tube for 30 min under agitation

20

and absence of light. Then the labeled solution was transferred to 96-well EME with 3 µL 0.1 %

21

(w/w) Aliquat 336 in 1-octanol as the supported liquid membrane (SLM) and 200 µL 10 mM NaOH

22

as waste solution. EME was performed for 10 min with a voltage of 50 V, with the anode in the

23

waste solution and at 900 rpm agitation. Negatively charged and unconjugated FITC was extracted

24

electro-kinetically into the SLM and to the waste solution. Analysis of purified samples, by Taylor

25

dispersion analysis (TDA), showed a 92 % removal of unconjugated FITC (FITC clearance: 92 %,

26

RSD: 3 %), while 79 % of the HSA/FITC complex remained in the sample (Protein retention: 79 %,

27

RSD: 18 %). Conserved functionality of the HSA/FITC complex after EME was proven by a binding

28

affinity study with anti-HSA using flow induced dispersion analysis (FIDA). In this real sample, the

29

dissociation constant (Kd) and hydrodynamic radius of the complex were determined to be 0.8 µM

30

and 5.87 nm respectively, which was in concordance with previously reported values.

ACS Paragon Plus Environment

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 2 of 19

Introduction

31 32 33

Electromembrane extraction (EME) is a three-phase microextraction technique developed for

34

analytical applications, and was presented for the first time in 20061. The technique is a

35

combination of liquid phase microextraction (LPME) and electrophoresis, and has been reported

36

for extraction of chargeable analytes from aqueous samples such as biological fluids2-4,

37

environmental waters5,6, and beverages7. In EME, analytes are extracted through a supported

38

liquid membrane (SLM) and into aqueous acceptor phase. The SLM is an organic solvent

39

immobilized inside the pores of a thin polymeric membrane. The main driving force for mass

40

transfer in EME is an electrical field sustained across the SLM. Two electrodes connected to an

41

external power supply are introduced into the system, one in the sample and one in the acceptor

42

phase. The electrical field force charged analytes to migrate towards the electrode of opposite

43

polarity. For the extraction of basic (cationic) analytes, the cathode is placed in the acceptor phase

44

while the anode is placed in the sample. For acidic analytes, the direction of the electrical field is

45

reversed. Since the electrical field only affects charged compounds, pH of the sample and acceptor

46

phase has to be controlled to a level where the analytes of interest are fully charged (i.e two-three

47

pH units from their pKa). EME can typically be completed in 5-10 minutes1,8, and the adaptation of

48

EME to a 96-well format open for high-throughput operation9. In the current literature, EME has

49

been used for analytical extraction of target analytes from complex samples. Target analytes have

50

been transferred electro-kinetically from the original sample solution, through the SLM, and into

51

pure aqueous solution containing acid or base as acceptor phase. Subsequently the analytes in the

52

acceptor phase have been measured by chromatography, mass spectrometry, or related

53

instrumental techniques.

54

Recently, EME was utilized for purification for the first time. In one paper, EME was used to

55

remove and recycle template molecules after polymerization of molecularly imprinted polymers10.

56

In the second paper, EME was used for rapid and efficient salt removal from minute volumes of

57

saline samples11. From a conceptual point of view, these papers represented a new direction for

58

EME and new types of applications may be developed. To explore this potential, fundamental

59

research is requested, and in the present work we test for the first time EME for removal of

60

unconjugated fluorescent reagent from protein samples. Fluorescence labeling is a common way

61

to enhance the sensitivity and detectability of proteins and involves labeling of purified protein

62

with a fluorescent reagent and subsequently detection by fluorescence spectrophotometry12. To

63

ensure effective labeling of the protein, surplus fluorescent reagent is added to the protein

64

solution. However, in order to suppress the background level of fluorescence, the samples need to

65

be purified to remove unconjugated fluorescent reagent

13-15.


This is of general importance in

Page 3 of 19 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


66

experiments with fluorescence detection, including those conducted by flow induced dispersion

67

analysis (FIDA). In FIDA, the in-capillary diffusivity of an analyte is used to calculate the analyte

68

concentration16-18. FIDA can also be used in binding studies of antibodies to fluorescently labeled

69

protein, and removal of unconjugated fluorescent reagent is mandatory for accurate calculation

70

of binding constants through the change in diffusivity.

71

The most common fluorescent reagents are derivatives of fluorescein, witch have different

72

fluorescence, stability and protein binding properties. Fluorescein isothiocyanate (FITC) is one

73

example, and is frequently used19. Existing techniques for removal of unconjugated FITC include

74

gel filtration20, dialysis21, and purification columns for dye removal22,23. The mutual goal for these

75

techniques is (1) to remove as much as possible of unconjugated fluorescent reagent (high reagent

76

clearance) and (2) to avoid loss of labeled protein (high protein retention). The existing

77

techniques have proven successful in the separation of labeled protein and unconjugated

78

fluorescent reagent24,25 but can be time and labor intensive26, and with a loss of labeled protein26.

79

The current paper is the first report on EME removal of unconjugated fluorescent reagent from

80

protein solutions. The experimental work was fundamental and intended as an initial research on

81

feasibility. Although real samples were included in the current experiments, it should be

82

emphasized that more work is required before the concept can generally be applied as part of

83

experimental work with proteins. FITC was selected as model substance for the fluorescent

84

reagent. FITC is an ideal candidate for EME (Figure S1). The carboxylic group can be ionized in

85

neutral and alkaline solution, which is required for EME. In addition, mass transfer is favored by

86

the small size (389.4 Da) and hydrophobic character (log P 4.8) of FITC. The common proteins

87

cytochrome C, human serum albumin (HSA), and myoglobin were selected as model proteins. First

88

EME was optimized for removal of FITC (high FITC clearance), and for retention of protein in the

89

sample (high protein retention). Second, samples of FITC labeled HSA was processed with EME

90

and subjected to Taylor dispersion analysis (TDA)27-30, to further verify EME performance and to

91

document the stability of the HSA/FITC complex during EME. Finally, EME was combined FIDA

92

for a real sample application, where binding and affinity studies of HSA and anti-HSA antibody

93

were performed.



Experimental

94 95

Chemicals and solutions

96

FITC,sodium tetraborate decahydrate, boric acid, 2-nitrophenyl octyl ether (NPOE), 1-octanol,

97

methanol, acetone, horse heart myoglobin, bovine heart cytochrome C, HSA, and monoclonal

98

antibody against HSA (mouse, IgG1) were purchased from Sigma Aldrich (St Louis, Missouri, USA).

99

Sodium chloride and sodium hydroxide (NaOH) were purchased from Merck (Darmstadt,

100

Germany). Aliquat 366 was purchased from Cognis Corporation (Cincinnati, Ohio, USA). Deionized

101

water was obtained from a Milli-Q water purification system (Molsheim, France)

102

A stock solution of 1 mg/mL FITC dissolved in acetone was prepared freshly each day and diluted

103

with either 50 mM borate buffer pH 9.20 or the reaction buffer (50 mM borate buffer pH 9.20 with

104

150 mM NaCl) to the working concentration of 0.01 mg/mL (2.57·10-5M). Solutions of each of the

105

proteins were made to a concentration of 0.1 mg/mL for the optimization studies. For the FITC

106

labelling reaction, 10 mg HSA was weighed and dissolved in 10 mL 50 mM borate buffer pH 9.20

107

providing a molar concentration of 1.50·10-5 M. All solutions were stored in darkness at < 8°C.

108

Electromembrane extraction (EME)

109

Equipment used for EME is illustrated in Supporting Information Figure S2. A laboratory built 96-

110

well stainless steel plate with 0.5 mL wells was used as compartment for the waste solutions and

111

as anode (96-well waste reservoir plate). A 96-well MultiScreen-IP filter plate with polyvinylidene

112

fluoride (PVDF) filter membranes with 0.45 µm pore size (Merck Millipore Ltd, Carrigtwohill,

113

Ireland) was used as SLM and compartment for the samples (96-well filter plate). The volume of

114

each well in the 96-well filter plate was 350 μL and this defined the maximum possible volume of

115

sample. A laboratory built aluminum plate with 96 rods tailor-made for the wells of the

116

MultiScreen plate, was used as cathode (96-electrode plate). A model ES 0300-0.45 (Delta

117

Elektronika BV, Zierikzee, The Netherlands) was used as power supply, and a Vibramax 100

118

Heidolph shaking board (Kellheim, Germany) was used to agitate the extraction system.

119

EME was performed according to the following procedure for each sample: a) 200 μL of 10 mM

120

NaOH was pipetted into the 96-well waste reservoir plate, b) 3 µL organic solvent was pipetted

121

on to the filter membrane of the 96-well filter plate, and c) 100 µL sample was pipetted into the

122

96-well filter plate. The 96-well waste reservoir plate and the 96-well filter plate were clamped,

123

and the 96-electrode plate was located on top. The rod electrodes of the electrode plate was then

124

in contact with the samples. The waste reservoir plate (anode) and the electrode plate (cathode)

125

were connected to the power supply. The extraction time varied from 1 to 10 minutes, the voltage

126

was between 1 and 100 V, and agitation was set to 900 rpm. For a couple of experiments, the

127

direction of the electrical field was reversed.


Page 4 of 19

Page 5 of 19 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


128

FITC clearance and protein retention

129

FITC clearance is the percentage of FITC removed from the sample after EME compared to the

130

non-extracted sample, as defined by equation (1). FITC clearance of 100% indicates complete

131

removal of FITC from the sample after EME. Protein retention is the percentage of protein left in

132

the sample after EME compared to the non-extracted sample, as defined by equation (2). Protein

133

retention of 100% indicates no protein loss after EME.

134

𝐹𝐼𝑇𝐶 𝑐𝑙𝑒𝑎𝑟𝑎𝑛𝑐𝑒 = 1 ― [𝐹𝐼𝑇𝐶 ]𝑠𝑎𝑚𝑝𝑙𝑒,𝑏𝑒𝑓𝑜𝑟𝑒 𝐸𝑀𝐸 ∙ 100%

135

𝑃𝑟𝑜𝑡𝑒𝑖𝑛 𝑟𝑒𝑡𝑒𝑛𝑡𝑖𝑜𝑛 =

136

FITC labeling protocol

137

The fluorescent labeling was performed on HSA and the reaction was done with a molar ratio of

138

10:1 FITC to HSA31. Calculation of the volume of 0.01 mg/mL FITC (VFITC(uL)) needed to obtain

139

molar ratio of 10:1, was done with the general formula:

140

𝑉𝐹𝐼𝑇𝐶(𝜇𝑙) = 𝑀𝑃𝑟𝑜𝑡𝑒𝑖𝑛(𝑔/𝑚𝑜𝑙) ∙ 1000 𝑚𝑔/𝑚𝑙 ∙ 𝐶𝐹𝐼𝑇𝐶 (𝑚𝑜𝑙/𝑙) ∙ 106 𝜇𝑙/𝑙

141

The HSA solution was prepared to a molar concentration of 1.50·10-5 M and set to a volume of 100

142

µL. The volume of 0.01 mg/mL FITC needed to obtain a molar ratio of 10:1 was 0.585 mL.

143

The labelling reaction was performed in a 1.5 mL low-bind Eppendorf tube (Eppendorf, Hamburg,

144

Germany). 100 µL of HSA solution was mixed with 585 µL of a freshly made 0.01 mg/mL FITC

145

solution (in reaction buffer) in the Eppendorf tube. The tube was wrapped in aluminium foil, for

146

light protection, and was agitated for 30 minutes at 900 rpm. After fluorescent labeling, 100 µL of

147

the solution was transferred to EME.

148

To avoid saturation of the fluorescence detector in the TDA and FIDA analysis, both the extracted

149

and un-extracted samples were diluted with 50 mM borate buffer pH 9.20 to a theoretical

150

concentration of 10 nM or 20 nM for the HSA/FITC –complex.

151

Flow injection analysis with UV-detection

152

In experiments reported in section 3.1 and 3.2, FITC and proteins were individually measured

153

using a 3000 Ultimate HPLC-UV (Thermo Fisher, Waltham, MA, USA) for automated flow injection

154

analysis with UV-detection. Thus, the HPLC column was replaced with PEEK tubing. Mobile phase

155

consisted of a 50 mM borate buffer pH 9.20 and the injection volume was 20 µL. The flow of mobile

156

phase was 0.1 mL/min. UV-detection was performed at 214 nm for Myoglobin and Cytochrome C,

(

[𝐹𝐼𝑇𝐶 ]𝑠𝑎𝑚𝑝𝑙𝑒, 𝑎𝑓𝑡𝑒𝑟 𝐸𝑀𝐸

(

[𝑃𝑟𝑜𝑡𝑒𝑖𝑛]𝑠𝑎𝑚𝑝𝑙𝑒,𝑎𝑓𝑡𝑒𝑟 𝐸𝑀𝐸

)

(1)

) ∙ 100%

(2)

[𝑃𝑟𝑜𝑡𝑒𝑖𝑛]𝑠𝑎𝑚𝑝𝑙𝑒,𝑏𝑒𝑓𝑜𝑟𝑒 𝐸𝑀𝐸

𝑉𝑃𝑟𝑜𝑡𝑒𝑖𝑛(𝑚𝑙) ∙ 𝐶𝑃𝑟𝑜𝑡𝑒𝑖𝑛(𝑚𝑔/𝑚𝑙)

𝑀𝑜𝑙𝑎𝑟 𝑟𝑎𝑡𝑖𝑜


(3)


157

216 nm for HSA, and 495 nm for FITC. The system was washed with a pure methanol solution

158

prior to each injection to avoid any carryover.

159

Taylor Dispersion Analysis conditions

160

The TDA experiments were conducted on a capillary electrophoresis instrument (Agilent 3D CE,

161

Agilent Technologies, Waldbronn, Germany) using light emitting diode fluorescence detection

162

(ZETALIF LED, Picometrics, Labège, France) with excitation wavelength 480 nm. A standard fused

163

silica capillary (inner diameter: 75 µm, outer diameter: 375 µm, length total: 100 cm, length to

164

detection window: 88 cm) was used. The capillary cassette was temperature-controlled to 25°C,

165

and the capillary inlet, outlet and sample vials were indirectly thermostated to 24°C via the lab

166

heating, ventilation and air conditioning system.

167

The capillary was rinsed and equilibrated prior to each sample analysis with 1 M NaOH at 1 bar

168

for 5 min followed by 50 mM Borate buffer pH 9.20 at 1 bar for 7 min. The HSA/FITC–complex

169

sample was injected at 50 mbar for 10 s (corresponding to 1 % of the capillary volume).

170

Subsequently, the injected sample was mobilized towards the detector with running buffer (50

171

mM borate buffer, pH 9.20) at 50 mbar for 20 min. All samples were analyzed in triplicate.Flow-

172

Induced Dispersion Analysis conditions

173

The FIDA experiments were conducted on a FIDAlyzer instrument (FIDA-Tech ApS, Copenhagen,

174

Denmark) using light emitting diode fluorescence detection (ZETALIF LED, Picometrics) with

175

excitation wavelength 480 nm. A fused silica capillary (inner diameter: 75 µm, outer diameter:

176

375 µm, length total: 100 cm, length to detection window: 90 cm) with an inner coating of

177

poly(ethylene glycol) was used. The capillary was temperature-controlled to 25°C inside the

178

FIDAlyzer instrument. Capillary inlet and samples were also temperature-controlled to 25°C.

179

The HSA/FITC–complex, purified via EME was diluted with 67 mM phosphate buffer pH 7.4 to a

180

fixed indicator concentration of 20 nM in varying analyte concentrations of anti-HSA antibody (0

181

– 5 µM). All samples were pre-incubated to reach equilibrium prior to analysis.

182

The capillary was flushed and equilibrated prior to each sample analysis with 67 mM phosphate

183

buffer at 1500 mbar for 5 min. The analyte sample (anti-HSA at varying concentrations) was

184

injected at 1500 mbar for 45 s, subsequently the indicator sample (HSA/FITC–complex, mixed

185

with anti-HSA at varying concentrations) was injected at 50 mbar for 10 s (39 nL, corresponding

186

to 1 % of the capillary volume). The injected indicator sample was then mobilized towards the

187

detection point with the analyte sample at 400 mbar for 180 s. All samples were analyzed in

188

triplicate.

189


Page 6 of 19

Page 7 of 19 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


190

TDA and FIDA data analysis

191

The Taylorgrams were processed using FIDA Data Analysis Software, version 01 (FIDA-Tech ApS,

192

Copenhagen, Denmark) in order to calculate the hydrodynamic radii of two species in one signal

193

with identical residence time, i.e. unconjugated FITC and HSA/FITC complex. Subsequently the

194

software was used for plotting the binding curve. The software utilized equations (S1), (S2), (S3)

195

and (S4) (Supporting Information)18.



196

Results and discussion

197 198

Concept and principle

199

The purpose of this fundamental research was to investigate the potential of EME for removal of

200

unconjugated fluorescent reagent after protein labeling. The workflow included (1) labeling of

201

protein with fluorescent reagent, (2) EME and (3) final measurement by TDA or flow-induced

202

dispersion analysis (FIDA). The EME device is presented in Figure S2, and consisted of (a) a 96-

203

well waste reservoir plate, (b) a 96-well filter plate, (c) a 96-electrode plate, (d) an agitator, and (e)

204

an external power supply. The three 96-plates (a,b,c) were sandwiched, and provided 96

205

individual extraction channels. This enabled simultaneous processing of up to 96 samples with

206

labeled protein (one sample per channel).

207

EME involved the following four steps: (1) loading the supported liquid membrane (SLM), (2)

208

sample loading, (3) waste solution loading, and (4) extraction. During SLM loading (1), 3 μL organic

209

solvent was pipetted on to a filter on the 96-well filter plate (b). This procedure was repeated for

210

the number of samples with labeled protein to be processed. The organic solvent rapidly diffused

211

into the pore volume of the filter and was immobilized by capillary forces, creating the SLM.

212

During sample loading (2), samples were pipetted above the loaded SLMs in the 96-well filter

213

plate (b). The samples contained labelled protein and fluorescent reagent. During waste solution

214

loading (3), a waste solution was pipetted into the 96-well waste reservoir (a) for each sample,

215

and the three plates (a,b,c) were clamped together and placed in the agitator (d). Finally, during

216

EME (4) the electrical field was applied by connecting the 96-well waste reservoir plate and the

217

96-electrode plate to the external power supply (5). The anode (positive potential) was located

218

in the waste solution while the cathode was in the samples.

219

FITC was selected as model fluorescent reagent. Three model proteins were selected to cover

220

different molecular sizes and isoelectric point, namely HSA (MW = 66.5 kDa, pI = 5.8)32, equine

221

heart myoglobin (MW = 17 kDa, pI = 6.7 (1) and 7.2(2))33 and bovine heart cytochrome C (MW =

222

12kDa, pI = 10.5)34.

223 224

Operational parameters for efficient FITC clearance and protein retention

225

Major operational EME parameters were considered to be (1) the chemical composition of the

226

SLM, (2) voltage, and (3) time. 1-octanol and NPOE are both commonly used as SLM solvents in

227

EME, mainly for the extraction of hydrophobic (log P > 1.5) acidic and basic drug compounds,


Page 8 of 19

Page 9 of 19 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


228

respectively8,35. In initial EME experiments, FITC clearance was investigated with 1-octanol and

229

NPOE. However, both 1-octanol and NPOE proved to be ineffective for extraction, and FITC

230

clearance was limited to 5-10 % and

Electromembrane Extraction of Unconjugated Fluorescein

Recommend Documents