High Speed Size Sorting of Subcellular Organelles by Flow Field-Flow

Subscriber access provided by NEW YORK UNIV

Article

High Speed Size Sorting of Subcellular Organelles by Flow Field-Flow Fractionation Joon Seon Yang, Ju Yong Lee, and Myeong Hee Moon Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b01207 • Publication Date (Web): 25 May 2015 Downloaded from http://pubs.acs.org on May 28, 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.

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 33

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

1 2 3

May 22, 2015

High Speed Size Sorting of Subcellular Organelles by Flow FieldFlow Fractionation

4 5

Joon Seon Yang, Ju Yong Lee, Myeong Hee Moon*

6 7 8 9 10

Department of Chemistry Yonsei University Seoul, 120-749 South Korea

11 12 13

*Corresponding author:

Myeong Hee Moon

14 15 16 17 18

Department of Chemistry Yonsei University Seoul, 120-749, South Korea Phone: (82) 2 2123 5634 Fax: (82) 2 364 7050

19 20

Email: [email protected]

21

ACS Paragon Plus Environment


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 33

2

22

ABSTRACT

23

Separation/isolation of subcellular species, such as mitochondria, lysosomes, peroxisomes,

24

Golgi apparatus and others, from cells is important for gaining an understanding of the

25

cellular functions performed by specific organelles. This study introduces a high speed, semi-

26

preparative scale, biocompatible size sorting method for the isolation of subcellular organelle

27

species from homogenate mixtures of HEK 293T cells using flow field-flow fractionation

28

(FlFFF). Separation of organelles was achieved using asymmetrical FlFFF (AF4) channel

29

system at the steric/hyperlayer mode in which nuclei, lysosomes, mitochondria, and

30

peroxisomes were separated in a decreasing order of hydrodynamic diameter without

31

complicated pre-processing steps. Fractions in which organelles were not clearly separated

32

were re-injected to AF4 for a finer separation using the normal mode, in which smaller sized

33

species can be well fractionated by an increasing order of diameter. The subcellular species

34

contained in collected AF4 fractions were examined with scanning electron microscopy to

35

evaluate their size and morphology, Western blot analysis using organelle specific markers

36

was used for organelle confirmation, and proteomic analysis was performed with nanoflow

37

liquid chromatography-tandem mass spectrometry (nLC-ESI-MS/MS). Since FlFFF operates

38

with biocompatible buffer solutions, it offers great flexibility in handling subcellular

39

components without relying on a high concentration sucrose solution for centrifugation or

40

affinity- or fluorescence tag-based sorting methods. Consequently, the current study provides


Page 3 of 33

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


3

41

an alternative, competitive method for the isolation/purification of subcellular organelle

42

species in their intact states.

43

44 45

Keywords: organelle separation, flow field-flow fractionation, size sorting, Western blot of organelles, organelle proteome, nLC-ESI-MS/MS

46

47 48

INTRODUCTION

49

Organelles are subcellular, membrane-bounded structures found in eukaryotic cells and

50

include the nucleus, Golgi complex, mitochondrion, endoplasmic reticulum (ER), lysosome,

51

peroxisome, ribosome, proteasome, and others.1,2,3 Since organelles have their own distinct

52

functions, abnormalities of these organelles may influence the development of diseases

53

directly or indirectly. For instance, it is reported that somatic mitochondrial DNA mutation

54

caused by oxidative damage may cause neurodegenerative disorders such as Alzheimer’s

55

disease and Parkinson’s disease.4,5,6 A number of diseases are known to be related with

56

different organelles: typical lysosomal storage disorders (LSD)7 such as Gaucher and Fabry

57

diseases, peroxisome biogenesis disorders (PBD)8 such as Zellweger syndrome and neonatal

58

adrenoleukodystrophy, and ER stress9 such as obesity, insulin resistance, and type 2 diabetes.



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 33

4

59

While advances in proteomic analyses offer the capability to unveil the functions of some of

60

the regulatory proteins, these proteins are often expressed in specific subcellular locations

61

and are very low in abundance.3 Therefore, selective isolation or purification of organelles is

62

an important challenge for the development of both biomarkers and therapeutic methods.

63

Isolation of organelles is traditionally achieved with centrifugation methods, such as

64

differential centrifugation and density gradient centrifugation. While centrifugation-based

65

methods are simple to perform, they are rather time consuming and it is difficult to achieve a

66

pure fraction when the organelles of interests have similar densities.2,10,11,12 Affinity

67

purification utilizes an antibody which has a specific binding affinity to a protein on the

68

surface of the organelle, yielding a relatively pure fraction of the specific organelle. However,

69

this method relies strictly on the binding specificity of the antibody to the organelle and,

70

moreover, available antibodies can be limiting.2,13,14 Fluorescence-activated organelle sorting

71

(FAOS) utilizes flow cytometry to selectively sort organelles labeled with specific fluorescent

72

chemicals. In the literature, it has been reported for the enrichment of endosomes,15

73

mitochondria,16 and secretory granules.17 While FAOS offers a relatively fast (< 1 h) sorting

74

of organelles with lower levels of contamination by other organelles compared to density

75

gradient centrifugation, it requires the expression of organelle-specific fluorescent proteins

76

and the sizes of the organelles that can be sorted are limited.

77

Flow field-flow fractionation (FlFFF) is an elution-based separation method that is capable


Page 5 of 33

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


5

78

of size-sorting particles and biological macromolecules like proteins, DNA, cells, and so on,

79

without relying on a partition or an interaction of sample components with the packing

80

materials.18,19,20 Separation in FlFFF is carried out in an empty, thin rectangular channel space

81

using two different flow streams: a migration flow carrying the sample components along the

82

channel axis to the detector and a crossflow moving across the channel cross-section to drive

83

the sample components in the vicinity of one wall of the channel. Particles with smaller

84

diameters diffuse faster, protrude farther against the channel wall in a direction opposite to

85

the crossflow, and migrate down the channel faster than larger particles since the flow

86

velocity in the parabolic flow profile increases as it moves away from the channel wall.

87

Therefore, smaller particles elute earlier than do larger ones, according to the normal mode of

88

FlFFF separation.18 However, for particles larger than ~ 1 µm, the particles’ diffusion

89

becomes negligible and the particles migrate at or above a certain distance from the channel

90

wall. In this case, the center of gravity (cg) of the particles determines the relative height of

91

the particles against the wall, and thus the migration speed, resulting in an elution of large

92

particles first described as the steric/hyperlayer mode of separation.21 Since separation in

93

FlFFF takes place in an unobstructed channel with biological buffer solutions, a gentle but

94

high speed separation of biological macromolecules can be achieved without shear induced

95

deformation or degradation. FlFFF has been widely used for the separation of biological

96

particles

97

lipoproteins,26,27,28 red blood cells,29 mitochondria,30 exosomes,31 and other cell components.

such

as

bacterial

ribosome

subunits,22,23

whole


bacterial

cells,21,24,25


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 33

6

98

While FlFFF has been employed for isolation of mitochondria and exosomes,30,31 these

99

studies were based on the size fractionation of subcellular species already separated by

100

centrifugation-based methods, followed by proteomic analysis of the collected fractions.

101

This study aimed to develop a high speed and soft sorting method using asymmetrical

102

FlFFF (AF4) channel for the isolation of subcellular species from cell homogenates in which

103

cell debris and heavy subcellular species were removed using gentle centrifugation.

104

Fractionation of cell homogenates by AF4 was first optimized using the steric/hyperlayer

105

mode, in which the separation of subcellular species in a decreasing order of hydrodynamic

106

diameter can be achieved. The organelle fractions were collected for examination of

107

morphology using scanning electron microscopy (SEM), organelles were confirmed by

108

Western blot analysis using organelle-specific markers, and proteomic analysis of each

109

fraction was performed using nanoflow liquid chromatography-electrospray ionization-

110

tandem mass spectrometry (nLC-ESI-MS/MS), as depicted in Figure 1. Fractions containing

111

early eluting species were collected and re-injected into the AF4 for a fine separation of

112

nuclei, lysosomes, and microtubules using different run conditions. The present study

113

demonstrates that AF4 can be utilized for the high speed and selective isolation of organelles

114

in an aqueous buffer solution without disrupting the subcellular species. This protocol can be

115

used as a complimentary method for studying subcellular proteins and lipids for future

116

biomarker development.


Page 7 of 33

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


7

117

118

EXPERIMENTAL

119

Materials and reagents

120

Protein standards and the following chemicals were purchased from Sigma-Aldrich (St. Louis,

121

MO, USA): bovine serum albumin (BSA 66 kDa), thyroglobulin (670 kDa), NaCl, HEPES,

122

MgCl2, KCl, protease inhibitor cocktail, osmium tetroxide (OsO4), Triton X-100, Tris-HCl,

123

Trizma Base, sodium dodecyl sulfate (SDS), dithiothreitol (DTT), glycerol, bromophenol

124

blue, Tween-20, ammonium persulfate, TEMED, glycine, urea, iodoacetamide (IAM), and

125

cysteine. Polystyrene (PS) standards with nominal diameters of 1.999 µm, 4.000 µm, and

126

6.007 µm (hereafter referred to as 2, 4, and 6 µm) were from Thermo Fisher Scientific

127

(Waltham, MA, USA). The blocking reagent used for Western blots and the protein

128

quantification solution used in Bradford assays were obtained from Bio-Rad Laboratories, Inc.

129

(Hercules, CA, USA). Primary antibodies including rabbit-α-LAMP2, rabbit-α-fibrillarin,

130

mouse-α-ATPB (ATP synthase subunit beta), rabbit-α-tubulin, mouse-α-58K, rabbit-α-

131

calnexin, and rabbit-α-catalase were purchased from Abcam Plc. (Cambridge, UK).

132

Secondary antibodies including anti-rabbit-IgG (HRP-linked) and anti-mouse-IgG (HRP-

133

linked) were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). The

134

protein ladder used in polyacrylamide gel electrophoresis was PageRulerTM from Pierce

135

Biotechnology (Rockford, IL, USA). For chemiluminescence detection, the EZ-Western



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 33

8

136

Lumi Femto solution from Daeil Lab Service Co.Ltd. (Seoul, Korea) was used. HPLC

137

solvents (acetonitrile, methanol, and water) were from J.T. Baker, Inc. (Phillipsburg, NJ,

138

USA). For proteolysis, proteomics grade trypsin from Promega Corp. (Madison, WI, USA)

139

was used.

140

141

Cell culture & homogenization

142

The Human Embryonic Kidney 293T cell line (HEK 293T) was obtained from Severance

143

Hospital (Seoul, Korea). Cells were grown in 100 mm culture flasks (72.3 cm2) in Dulbecco’s

144

Modified Eagle’s Medium (DMEM) from Invitrogen (Carlsbad, CA, USA) with 10% heat

145

inactivated fetal bovine serum (FBS) and 1% penicillin streptomycin added, in a 37oC

146

incubator with 5% CO2. When cells reached 90% confluency within 48-72 h, cells were

147

detached by adding 0.25% trypsin-EDTA (Invitrogen) to allow for subculturing.

148

Cells (about 1 x 107 cells/mL) detached from the culture flask were centrifuged at 200 x g

149

for 4 min to remove the trypsin-EDTA, mixed with 4 mL of 0.1 M PBS solution, and

150

centrifuged for 4 min at 200 x g. The remaining cells were dispersed in 5 mL of a hypotonic

151

lysis buffer solution32 containing 10 mM HEPES, pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5

152

mM DTT and protease inhibitor cocktail and incubated at 4oC for 5 min. The mixture was

153

transferred to a 7 mL Dounce homogenizer (30 strokes with a tight pestle) from Wheaton


Page 9 of 33

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


9

154

(Millville, NJ, USA). The homogenate was centrifuged at 15000 x g for 10 min to remove

155

heavy organelles and the supernatant solution (~ 4 mL) was collected for organelle separation

156

by FlFFF. The homogenate used in FlFFF is stored at 4 oC and only used within 24 h to

157

prevent potential damage of organelles due to its low osmolality.

158

159

FlFFF

160

The FlFFF channel system used in this study the model LC (Long Channel, 275 mm length),

161

an asymmetrical type of FlFFF (AF4) channel from Wyatt GmbH (Dernbach, Germany). The

162

channel space was cut in a trapezoidal design: an inlet breadth of 2.2 cm decreased to 0.6 cm

163

at the outlet, 26.6 cm long in a 250 µm thick Teflon sheet with a geometrical volume of 0.90

164

mL. For the channel membrane, a NADIR® regenerated cellulose membrane sheet (MWCO

165

10 kDa) from Microdyn-Nadir GmbH (Wiesbaden, Germany) was utilized. For sample

166

injection, a model 7125 injector (100 µL loop) from Rheodyne (Cotati, CA, USA) was placed

167

before the sample inlet of the AF4 channel, as shown in Figure S1 in Supporting Information.

168

Delivery of the carrier liquid to the AF4 channel was accomplished using a Model SP930D

169

HPLC pump from Young-Lin Instrument (Seoul, Korea). Carrier solutions were prepared

170

with deionized water containing 0.1% FL-70 from Thermo Fisher Scientific (Waltham, MA,

171

USA) added with 0.02% NaN3 for PS separation and 0.1 M PBS solution for the separation of

172

cell homogenates. Separation of organelles was performed using the following procedures.



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 10 of 33

10

173

The composition of FL-70 is in Supporting Information. Cell homogenate (100 µL) was

174

loaded in the injector and delivered to the AF4 channel with a carrier liquid using the

175

focusing/relaxation mode in which the 2.5 mL/min of carrier liquid was split to the inlet and

176

outlet at a ratio of 1:9 using both 3-way and 4-way valves and all liquid exited through the

177

channel wall, as shown in the dotted line configuration in Figure S1. Sample components

178

were focused and accumulated at equilibrium at or around a position, 2.6 cm from the

179

channel inlet, for 5 min. After focusing/relaxation, the flow direction was changed to the

180

configuration depicted by the solid line so that the pump flow was delivered only to the

181

channel inlet at an increased flow rate, V&in = 4.5 mL/min, and separation began. During the

182

separation mode, the outflow rate ( V&out ) was adjusted to 2.0 mL/min so that the crossflow rate

183

( V&c ) was maintained at 2.5 mL/min. Monitoring of eluting subcellular species was performed

184

with a model UV730D UV detector from Young-Lin at a wavelength of 280 nm and the

185

detected signals were recorded using Autochro-Win 2.0 plus software from Young-Lin.

186

Organelle fractions collected during FlFFF separation were used in secondary analyses with

187

SEM, Western blot, and nLC-ESI-MS/MS.

188

189

190

SEM of organelle fractions

During the AF4 separation of cell homogenates, four fractions were collected at time


Page 11 of 33

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


11

191

intervals of 0.7-1.7, 1.7-3.7, 5.2-8.0, and 8.5-12.0 min. Five AF4 runs were performed to

192

accumulate organelle fractions for SEM analysis. Each fraction was concentrated to about 1

193

mL by using an Amicon® Ultra-15 centrifugal filter unit (10 kDa NMWL) from Millipore

194

(Danvers, MA, USA) centrifuged at 2000 x g for 10 min at 4oC. Each concentrated solution

195

of the organelle fraction was slowly dropped on a (7 x 7 mm) polycarbonate membrane (pore

196

size of 0.1 µm) from Millipore placed on a filter unit, the bottom part of which was connected

197

with a syringe to pull the liquid in so that the dropped liquid was easily filtered through the

198

membrane. After filtration, 100 µL of 2% OsO4 solution was added to the filtrate on the

199

membrane for fixation and an hour was allowed for the natural draining of the OsO4 solution.

200

After fixation, the OsO4 fixed specimens were washed twice with 200 µL water and then

201

dehydrated by flushing with a series of methanol solutions (30, 50, 70, and 100%). For each

202

flush, 200 µL of each methanol solution was applied for 15 min and filtered by suction. After

203

the dehydration steps, specimens on the membranes were dried and then sputtered with Pt for

204

120 s. SEM examination was performed at 5.0 kV of acceleration voltage using a JEOL-

205

6701F Field Emission Scanning Electron Microscopy from JEOL Ltd. (Tokyo, Japan).

206

207

Western blot analysis of organelle fractions

208

For Western blot analysis of the organelle fractions collected from the AF4 runs, organelle

209

proteins were released by adding Triton-X100 to each collected fraction followed by tip



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 33

12

210

sonication (10 s pulse duration with 2 s intervals in 5 min).30,31 Concentration of Triton-X 100

211

was kept at 0.1% (v/v) in order to filtrate organelle proteins efficiently in the purification step

212

and to reduce interferences in UV absorbance measurement in Bradford assay. After lysis, the

213

protein concentration of each fraction was measured using the Bradford method. From each

214

fraction, 10 µg of protein was mixed with 5X Laemmli buffer solution (pH 6.8) containing 10%

215

(w/v) SDS, 10 mM DTT, 50% (v/v) glycerol, 0.2 M Tris-HCl, and 0.05% (w/v) bromophenol

216

blue and denatured at 90oC for 5 min. The volume of the denaturation buffer solution used

217

was 1/4 of the volume of the lysed protein solution. Gel electrophoresis of the denatured

218

proteins was carried out on a 10% polyacrylamide gel using the Mini-PROTEAN® Tetra Cell

219

system from Bio-Rad Laboratories, Inc. (Hercules, CA, USA).

220

protein migration was 80 V in the stacking gel and was increased to 150 V in the running gel.

221

After electrophoresis, proteins were transferred to a nitrocellulose membrane (Bio-Rad) at

222

200 V for 1 h and the membrane was washed with 0.1 M TBS buffer with 0.1% Tween-20 for

223

5 min. Then, the membrane was blocked with a blocking solution (5% (w/v) skim milk in

224

TBST) for 1 h and incubated with primary antibody at 4oC overnight. After incubating with

225

primary antibody, the membrane was washed with TBS for 30 min and incubated with

226

secondary antibody (1/5000 (v/v)) at room temperature for 1 h. Chemiluminescence detection

227

was performed with an LAS-4000 Mini detector from GE Healthcare (Little Chalfont, UK).

228


The applied voltage used for

Page 13 of 33

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


13

229

nLC-ESI-MS/MS of organelle proteins

230

Each AF4 fraction was digested in-solution for proteomic analysis. Details of proteolysis

231

can be found in Supporting Information. Protein analysis of the AF4 fractions was carried out

232

with nLC-ESI-MS/MS: a model 1260 Capillary LC system equipped with an autosampler

233

from Agilent Technologies (Waldbronn, Germany) and an LTQ Velos ion trap mass

234

spectrometer from Thermo Finnigan (San Jose, CA, USA). For nLC separation prior to ESI, a

235

trap column (2 cm x 200 µm-i.d., 5 µm-200Å Magic C18AQ from Michrom Bioresources Inc.

236

(Auburn, CA, USA)) and an analytical column (7 cm x 75 µm-i.d., 5 µm-100Å Magic

237

C18AQ) were connected via a PEEK microcross from IDEX (Oak Harbor, CA, USA), as

238

shown in Figure S1. Columns were prepared in the laboratory using capillaries from

239

Polymicro Technology, LLC (Phoenix, AZ, USA). Details of the packing procedure can be

240

found in the literature.33 Details of nLC-ESI-MS/MS can be found in Supporting Information.

241

242

RESULTS & DISCUSSION

243

The separation of organelles by AF4 was evaluated using various flow rates and channel

244

thicknesses. Figure 2 shows the fractograms of the HEK 293T cell homogenates obtained at

245

different crossflow rates ( V&c ), but using a fixed outflow rate, V&out =2.0 mL/min., in 0.1 M

246

PBS as a carrier solution. The fractogram of the cell homogenates at V&c = 1.5 mL/min (top)



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 33

14

247

shows a substantial, large void peak, which is expected to contain some non-retained

248

components such as large subcellular species or cytoplasmic proteins contained in

249

homogenates, followed by two overlapping peaks. As the crossflow rate was raised to 2.5 and

250

3.5 mL/min, the intense peak at early retention times was divided into two peaks (a sharp

251

void peak at the beginning and a large peak) presumably containing membrane debris or

252

proteins. Separation of the two overlapping peaks found in the 3-10 min range of the top

253

fractogram was improved and better resolution was obtained when the field strength, which

254

refers to the strength of an external force generated by crossflow in FlFFF, was raised to V&c =

255

2.5 mL/min. However, the intensities of these two peaks were reduced significantly when V&c

256

was further increased to 3.5 mL/min. In order to check for the presence of any remaining

257

components in the channel, the field strength was set to zero at 25 min of elution. The small

258

peaks that appear after 25 min in the top two fractograms were thought to be mostly from the

259

system pulse due to the pressure change when crossflow rate was turned off; a finding which

260

was confirmed by a blank run using the same run condition shown in the middle of Figure 2.

261

This system pulse was increased significantly in the bottom fractogram which may originate

262

from the larger pressure drop but it cannot exclude the possibility that a considerable amount

263

of subcellular species could be dragged by the channel surface (membrane) under a strong

264

field strength. While the channel thickness used for Figure 2 was 250 µm, we also tested

265

channels of 190 µm and 150 µm since separation in steric/hyperlayer mode of FlFFF can be


Page 15 of 33

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


15

266

enhanced by reducing channel thickness, resulting in a decrease in zone broadening. However,

267

separation was not improved much, and no additional elution of strongly retained

268

components was seen. As the channel thickness was increased to 350 µm, the second and

269

third peaks were not successfully resolved, as shown at the bottom of Figure S2a. Thus, the

270

run condition using V&out / V&c =2.0/2.5 mL/min in a 250 µm thick channel was selected for

271

subsequent separation of the cell homogenates.

272

Figure 3a shows the superimposed fractograms of cell homogenates from five repeated

273

injections at V&out / V&c =2.0/2.5 mL/min, showing good reproducibility of retention time and

274

resolution together with consistent peak intensities of the transient peaks after turning off the

275

field. Fractions of cell homogenates collected from AF4 separations were accumulated for 5,

276

10, and 10 repeated runs for SEM, Western blot, and proteomic analysis using nLC-ESI-

277

MS/MS, respectively. For each run, 100 µL of homogenate sample was injected. Collection

278

time interval for each fraction is listed in Table 1. In order to confirm that the eluted species

279

in the F3 and F4 fractions were subcellular components, the cell homogenate sample used for

280

the AF4 separation was tip-sonicated for 5 min to disrupt organelles and the sonicated

281

mixture was then injected to the AF4 channel at the same run condition used in Figure 3a.

282

The resulting fractogram, shown in Figure 3b, demonstrates that the peaks observed in the 5-

283

13 min range in Figure 3a disappeared with the increase in the peak intensity of the first peak,

284

supporting the idea that the species in F3 and F4 were membrane-based subcellular particles.



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 33

16

285

The possibility of elution of very high MW but free unbound proteins in the F3 and F4 peaks

286

is relatively small, since MWs of most proteins in cells are less than a million. Based on the

287

separation of protein standards (BSA, 66 kDa and thyroglobulin, 670 kDa) in the fractogram

288

shown in Figure 3c, it can be estimated that the huge peak in the Figure 3b fractogram may

289

have originated from the increased amount of proteins (< 100 kDa) caused by the disruption

290

of organelles by sonication. The fractogram in Figure 3d represents a steric/hyperlayer

291

separation of polystyrene (PS) standard latex particles (6, 4, and 2 µm in diameter) and

292

provides an approximate size range of the species eluted in F3 and F4. PS separation was

293

obtained using the same flow rate condition but using a different carrier solution (0.1% FL-70

294

plus 0.02% NaN3 for particle separation).

295

After the calibration using PS particles, the channel membrane was replaced with a new one,

296

and each of the four fractions (F1-F4) was collected from five repeated FlFFF runs and was

297

examined by SEM, shown in Figure 4a. The SEM images of fractions F1 and F2 show

298

relatively large species (>> 1 µm) with irregular shapes, except for a few spheres found in F1,

299

which are likely agglomerations of cell wall debris, proteins possibly derived from ER, and a

300

few sub-micrometer-sized subcellular species that were swept out without being resolved in

301

the steric/hyperlayer separation conditions. While the first two fractions showed non-specific

302

shaped species, fractions F3 and F4 clearly contain organelles, with decreased sizes in the

303

later fraction. The average diameter of particles measured in each fraction is listed in Table 1,


Page 17 of 33

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

304

and reveals that the average diameter or apparent length of particles decreased as retention

305

time increased, thereby supporting the idea that these fractions were separated by the

306

steric/hyperlayer elution mode of FFF. The contents of all fractions were further confirmed

307

with Western blot by using seven organelle specific markers, especially focusing on

308

membrane bound organelles (anti-tubulin for the presence of microtubules in the cytoplasm,

309

anti-fibrillarin for nuclei, anti-LAMP2 for lysosomes, anti-ATPB for mitochondria, anti-

310

catalase for peroxisomes, anti-58K for Golgi apparatus, and anti-calnexin for ER). The

311

Western blot results from fractions F1 and F2, presented in Figure 4b, demonstrate the

312

presence of tubulin (supporting the presence of cytoplasmic proteins) in both fractions.

313

Lysosomes were enriched in F1, while nuclei (typically the largest organelle in cells) were

314

found in both fractions. Since the homogenate sample was centrifuged at 15,000 x g for 10

315

min during lysate preparation, most nuclei would be expected to have been removed.

316

Therefore, the bands of the nuclei marker in Figure 4b are very dim compared to the bands of

317

other markers. Moreover, anti-tubulin is a marker for tubulins in the cytoplasm34or

318

microtubules that are long hollow cylinders made up of polymerized α/β-tubulin dimers

319

thought to elute in the first peak due to their long dimensions. It is noteworthy that although

320

lysosomes are known to be similar in size to mitochondria and peroxisomes,2,35,36 the

321

lysosomes in our experiment are clearly separated from the mitochondria and peroxisomes.

322

We presumed that lysosomes appeared in this experiment were either much larger or even

323

smaller than the other two organelles as they eluted at short retention times according to the



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 33

18

324

steric/hyperlayer mode of separation. From the SEM image and a report that lysosomes may

325

increase in size by aggregation,37 they were thought to be aggregated in the first AF4

326

separation. This is supported from the additional experiments in Figure 5 and will be

327

discussed later. In order to further examine these two fractions (F1 and F2), they were

328

collected for re-injection into the AF4 channel using a different flow rate condition, as will be

329

shown later.

330

were confirmed to be mitochondria in Figure 4b, smaller oblong particles were found in F4

331

together with spherical particles which were presumed to be the peroxisomes identified by

332

the Western blot results of the F4 fractions. Therefore, according to Western blot analysis, the

333

major components of F3 and F4 are expected to be mitochondria and peroxisomes,

334

respectively. While mitochondria and peroxisomes are known to be similar in size,2 both were

335

clearly separated in the AF4 experiment due to their differences in shape. Components of

336

Golgi and ER were found in all fractions, as shown in Figure 4b, but they appeared to be

337

more concentrated in F2 and F4, respectively. There is no clear explanation as to why they

338

appeared throughout the fractions. The Golgi apparatus and ER are known to be non-specific

339

in size and shape.36 Instead, it is possible that these organelles were associated with other

340

organelles, resulting in their co-elution, or aggregation of proteins may result in the presence

341

of ER throughout the fractions.

342

While the organelles seen in the micrographs of F3 are mostly rod shape and

Confirmation of organelles was obtained by proteomic analysis of each fraction by using


Page 19 of 33

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


19

343

nLC-ESI-MS/MS and the number of identified proteins belonging to each subcellular

344

location was grouped into two groups (F1 plus F2 vs. F3 plus F4) in Table 2. The number of

345

proteins that belong to the cytoplasm, nucleus, and lysosome (including endosome and

346

exosome) identified by nLC-ESI-MS/MS as being exclusively present in fractions F1+F2 is

347

larger than those present in F3+F4 (Table 2), while mitochondrial and peroxisomal proteins

348

are exclusively enriched in fractions F3+F4. The numbers listed for each organelle are

349

represented in two ways; the number of unique proteins found only in either fraction is paired

350

together with the number in parentheses which represents proteins counted in multiple

351

locations. The data in Table 2 supported the expectations derived from the Western blot

352

results of the collected fractions from the AF4 separation. Details of the proteins belonging to

353

lysosomes, mitochondria, and peroxisomes are presented in Tables S1, S2, and S3 of

354

Supporting Information, respectively. It is noted that lysosomal proteins and exsosomal

355

proteins are found mostly from F1 and F1+F2, respectively in Table S1. Moreoever,

356

mitochondrial and peroxisomal proteins are enriched in F3 and F4, respectively in Table S2

357

and S3.

358

To achieve a better separation of the organelle species found in fractions F1 and F2, both

359

fractions were collected from 10 additional runs and a mixture of the F1 and F2 fractions was

360

re-injected into the AF4 channel using different run conditions of V&out / V&c = 0.5 / 2.5

361

mL/min, which is a much lower outflow rate compared to that used in Figure 3 and similar to



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 33

20

362

the run condition for normal mode of separation.26 The small peak which appeared after the

363

field-off at 15 min in Figure 5 was the system pulse which had a peak area nearly the same as

364

that of a blank run. In addition, the eluent from the system pulse was collected for Bradford

365

assay and used to confirm that neither proteins nor any organelles were present in the system

366

pulse. The intense peak was clearly separated from a void peak (at ~ 1 min) as the outflow

367

rate was reduced in the re-injection run. Moreover, it appeared to be resolved better than the

368

former separation condition in Figure 3, therefore the four fractions (R1: 0-2.0 min, R2: 2.0-

369

4.5, R3: 4.5-7.0, and R4: 7.0-15 min) were collected and analyzed by Western blot. The

370

inserted micrographs of Western blots in Figure 5 show that nuclei eluted with the void peak.

371

This is because nuclei are much larger than other organelles, resulting in their being swept

372

along with the void peak, rather than being retained, due to size effects in this run condition.

373

Cytoplasmic proteins, including tubulins, were confirmed to be present in fraction R2 and

374

clearly separated from nuclei. Moreover, lysosomes were found to be more concentrated in

375

fraction R3 and Golgi apparatus, which was present in all fractions in the original run, were

376

now more concentrated in R4. Since lysosomes are spherical and less than ~1 µm in size,

377

their separation in Figure 5 appears to be accomplished by the normal mode of FlFFF in

378

which proteins of smaller sizes (less than 100 kDa typically) eluted earlier than larger ones.

379

This may be explained as non-spherical particles, such as rods, which are usually less

380

retained in FlFFF than spheres with diameter equivalent to the length of rods, due to the

381

increase in the steric/entropic effect.38,39 While fractions R2 and R3 were confirmed to


Page 21 of 33

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


21

382

contain tubulins and lysosomes, it cannot be excluded that the intense peak (R2 and R3) was

383

largely contributed by the detection of cytoplasmic proteins.

384

385

CONCLUSION

386

This study provides a new guideline for the fast isolation/enrichment of subcellular

387

organelles, without the use of a sucrose solution, allowing the collected fractions to be used

388

in additional biological experiments for other purposes and it expands the potential utility of

389

asymmetrical flow field-flow fractionation (AF4) for size sorting of subcellular organelles in

390

cell homogenate mixtures. The present study demonstrated that a careful selection of the AF4

391

run conditions using a consecutive separation by re-injecting fractions collected from one run

392

can be useful in the isolation/enrichment of target organelles of interests. In addition, it was

393

shown that the successful fractionation of organelles can be easily confirmed with SEM,

394

Western blot, and proteomic analysis using nLC-ESI-MS/MS, without inducing any serious

395

destruction of the organelles. While the recent centrifuge methods using sophisticated

396

gradient media (iodinated or colloid-based) are known to provide efficient fractionation of

397

organelles,11,12 the separation of organelles by AF4 provides a speed in separation, a

398

flexibility in handling cell homogenates with any biologically compatible buffer solution, and

399

an advantage of fractionating organelles by size and shape factors which are different from

400

mass based separation in centrifugation methods. While the current study demonstrates the



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 22 of 33

22

401

potential to sort organelles by size, further optimizations are needed to improve the resolution

402

of AF4 separations. It would be helpful to design experiments in such a way to increase

403

throughput by constructing a preparative scale AF4 channel or employing a multiplexed

404

hollow fiber FlFFF (MxHF5) so that sufficient amount of subcellular fractions can be

405

retrieved.

406

407

Acknowledgements

408 409 410

This study was supported by the grant NRF-2010-0014046 from the National Research Foundation (NRF) of Korea.

411 412 413 414 415

Supporting Information Additional information as noted in text. This information is available free of charge via the internet at http://pubs.acs.org

416 417


Page 23 of 33

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


23

418

Reference

419 420 421

(1) Yates, J. R., 3rd; Gilchrist, A.; Howell, K. E.; Bergeron, J.J. Nat. Rev. Mol. Cell Biol. 2005, 6, 702-714.

422

(2) Satori, C. P.; Kostal, V.; Arriaga, E. A. Anal. Chim. Acta. 2012, 753, 8-18.

423

(3) Huber, L. A.; Pfaller, K.; Vietor, I. Circ. Res. 2003, 92, 962-968.

424

(4) Reddy, P. H.; Beal, M. F. Brain Res. Rev. 2005, 49, 618-632.

425

(5) Chan, D. C. Cell 2006, 125, 1241-1252.

426

(6) Lin, M. T.; Beal, M. F. Nature 2006, 443, 787-795.

427

(7) Meikle, P. J.; Hopwood, J. J.; Clague, A. E.; Carey, W. F. J. Am. Med. Assoc. 1999, 281,

428

249-254.

429

(8) Gould, S. J.; Valle, D. Trends Genet. 2000, 16, 340-345.

430

(9) Özcan, U.; Cao, Q.; Yilmaz, E.; Lee, A. H.; Iwakoshi, N. N.; Özdelen, E.; Tuncman, G.;

431 432 433

Görgun, C.; Glimcher, L. H.; Hotamisligil, G. S. Science 2004, 306, 457-461. (10) Schmidt, H.; Gelhaus, C.; Lucius, R.; Nebendahl, M.; Leippe, M.; Janssen, O. BMC Immunol. 2009, 10, 41-52.

434

(11) Sims, N. R.; Anderson, M. F. Nat. Protoc. 2008, 3, 1228-1239.

435

(12) Segle, P. O.; Brinchmann, M. F. Autophagy 2010, 6, 542-547.

436

(13) Hornig-Do, H. T.; Günther, G.; Bust, M.; Lehnartz, P.; Bosio, A.; Wiesner, R. J. Anal.

437

Biochem. 2009, 389, 1-5.



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 33

24

438

(14) Wang, Y.; Taylor, T. H.; Arriaga, E. A. Anal. Bioanal. Chem. 2012, 402, 41-49.

439

(15) Wilson, R. B.; Murpphy, R. F. Methods Cell Biol. 1989, 31, 293-317.

440

(16) Cossarizza, A.; Ceccarelli, D.; Masini, A. Exp. Cell Res. 1996, 222, 84-94.

441

(17) Gauthier, D. J.; Sobota, J. A.; Ferraro, F.; Mains, R. E.; Lazure, C. Proteomics 2008, 8,

442

3848-3861.

443

(18) Giddings, J. C. Anal. Chem. 1981, 53, 1170A-1175A.

444

(19) Wahlund, K.-G.; Litzén, A. J. Chromatogr. 1989, 461, 73-87.

445

(20) Giddings, J. C. Science 1993, 260, 1456-1465.

446

(21) Reschiglian. P.; Zattoni, A.; Roda, B.; Casolari, S.; Moon, M. H.; Lee, J.; Jung, J.;

447 448 449

Rodmalm, K.; Cenacchi, G. Anal. Chem. 2002, 74, 4895-4904. (22) Nilsson, M.; Birnbaum, S.; Wahlund, K.-G. J. Biochem. Biophys. Methods 1996, 33, 923.

450

(23) Nilsson, M.; Wahlund, K.-G.; Bulow, L. Biotechnol. Tech. 1998, 12, 477-480.

451

(24) Lee, H.; Williams, S. K.; Wahl, K. L.; Valentine, N.B. Anal. Chem. 2003, 75, 2746-2752.

452

(25) Reschiglian, P.; Zattoni, A.; Roda, B.; Cinque, L.; Melucci, D.; Min, B. R.; Moon, M. H.

453 454 455 456 457

J. Chromatogr A, 2003, 985, 519-529. (26) Park, I.; Paeng, K.-J.; Yoon, Y.; Song, J.-H.; Moon, M. H. J. Chromatogr. B. 2002, 780, 415-422. (27) Rambaldi, D.C.; Zattoni, A.; Casolari, S.; Reschiglian, P.; Roessner, D.; Johann, J. Clin. Chem. 2007, 53, 2026-2029.


Page 25 of 33

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


25

458

(28) Lee, J.Y.; Choi, D.; Johan, C.; Moon, M.H. J. Chromatogr. A. 2011, 1218, 4144-4148.

459

(29) Barman, B. N.; Ashwood, E. R.; Giddings, J.C. Anal. Biochem. 1993, 212, 35-42.

460

(30) Kang, D.; Oh, S.; Reschiglian, P.; Moon, M.H. Analyst 2008, 133, 505-515.

461

(31) Kang, D.; Oh, S.; Ahn, S.-M.; Lee, B.-H.; Moon, M. H. J. Proteome Res. 2008, 7, 3475-

462

3480.

463

(32) Brunner, E.; Ahrens, C.H.; Mohanty, S.; Baetschmann, H.b.; Loevenich, S.; Potthast, F.;

464

Deutsch, E.W.; Panse, C.b.; De Lichtenberg, U.; Rinner, O.; Lee, H.; Pedrioli, P.G.A.;

465

Malmstrom, J.; Koehler, K.; Schrimpf, S.; Krijgsveld, J.; Kregenow, F.; Heck, A.J.R.;

466

Hafen, E.; Schlapbach, R.; Aebersold, R. Nat. Biotechnol. 2007, 25, 576-583.

467

(33) Kang, D.; Moon, M. H. Anal. Chem. 2006, 78, 5789-5798.

468

(34) Kurita, M.; Kuwajima, T.; Nishimura, I.; Yoshikawa, K. J. Neurosci 2006, 26, 12003-

469 470 471 472 473 474 475

12013 (35) Rockstroh, M.; Müller S. A.; Jende, C.; Kerzhner, A.; von Bergen, M.; Tomm, J. M J. Integr. OMICS. 2011, 1, 135-143. (36) Satori, C. P.; Henderson, M. M.; Krautkramer, E. A.; Kostal, V.; Distefano, M. M.; Arriaga, E. A. Chem. Rev. 2013, 113, 2733-2811. (37) Wang, H.-J.; Zhang, D.; Tan, Y.-Z; Li, T. Am. J. Physiol. Cell Physiol. 2013, 304, C617C626.

476

(38) Beckett, R.; Giddings, J. C. J. Colloid Interface Sci. 1997, 186, 53-59.

477

(39) Alfi, M.; Park, J. J. Sep. Sci. 2014, 37, 876-883.

478



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 26 of 33

26

479

Figure Captions

480 481

Figure 1. Scheme for organelle sorting by FlFFF and the process for confirming the presence of organelles in the collected fractions.

482

Figure 2. FlFFF fractograms of HEK 293T cell homogenate by varying crossflow rates ( V&c ) from 1.5 mL/min. to 3.5 mL min at a fixed outflow rate ( V&out ), 2.0 mL/min in 0.1 M PBS solution. The crossflow rate was changed to 0 mL/min at 25 minutes.

483 484 485 486 487 488 489

491

Figure 3. (a) Superimposed fractograms of HEK 293T cell homogenate obtained by five repeated FlFFF runs. Four fractions were collected at time intervals of 0.7-1.7, 1.7-3.7, 5.57.9, 9.0-12.0 min, respectively. (b) Fractogram of the cell homogenate sample after tipsonication, (c) separation of BSA and thyroglobulin, and (d) separation of PS standard mixtures. The crossflow rate was turned off at 17 minutes in a-c. Carrier solutions were 0.1 M PBS for runs a-c and 0.1% FL-70 with 0.02% NaN3 for run d. All runs were obtained at V&out / V& =2.0/2.5 mL/min.

492 493 494

Figure 4. (a) SEM images of each fraction collected during FlFFF separation showing the different sizes and shapes of the observed particles. (b) Western blot analysis using organelle specific antibodies. Each lane was loaded with 10 µg of protein extract from each fraction.

495 496 497

Figure 5. Fractogram of the combined fractions (F1 and F2) that were re-injected into the FlFFF channel to resolve organelles using the normal mode of separation and the corresponding Western blot results from the four fractions (R1-R4). Flow rate conditions were V&out / V&c = 0.5 / 2.5 mL/min.

490

498

c

499 500 501


Page 27 of 33

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


27

502 503

Table 1. Sizes of the observed subcellular species in each fraction measured from SEM images. fraction #

time

length

width

number

(shape)

(min.)

(µm)

(µm)

counted

F1

0.7 – 1.7

6.23 ± 2.08

3.65 ± 0.91

10

F2

1.7 – 3.7

4.36 ± 1.09

1.60 ± 0.72

11

F3 (rods)

5.5 – 7.9

1.88 ± 0.46

0.45 ± 0.09

34

F4 (rods)

9.0 – 12.0 0.83 ± 0.14

0.59 ± 0.17

15

F4 (spheres) 9.0 – 12.0 0.67 ± 0.14 504

a

a

21

diameter

505 506 507 508 509

Table 2. Number of proteins belonging to different subcellular locations identified in fractions F1+F2, F3+F4, and all fractions. A pair of numbers representing the protein numbers classified as being present in a single location, with the numbers in parentheses indicating proteins present in multiple locations.

510

Subcellular Location

F1+F2 only

F3+F4 only

(Multiple Location)

Both

Net

Fractions

Cytoplasm

172(52)

86(6)

98(12)

356(77)

Nucleus

87(51)

3(4)

2(7)

92(62)

Lysosome, Endosome, Exosome

12(10)

2(1)

0(2)

14(13)

Mitochondrion

10(2)

57(4)

23(7)

90(13)

Peroxisome

0(0)

12(2)

1(1)

13(3)

ER

16(5)

3(1)

15(1)

34(7)

Golgi Apparatus

9(3)

4(0)

11(2)

24(5)

Others

71(3)

42(2)

23(2)

138(7)

Total number of unique proteins

377(63)

209(11)

173(16)

759(90)

511 512 513 514



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 28 of 33

28

515

TOC Graphic only


Page 29 of 33

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


293T Cells

remove cell debris

Flow FFF Field

SEM (size & shape) Western blot (organelle confirmation)

homogenization

organelle mixtures

organelle separation


nLC-MS/MS (proteomics)

Figure 1


Vout / Vc = 2.0/1.5 (mL/min)

Vc  0 Detector Response

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

Page 30 of 33

= 2.0/2.5

blank run = 2.0/3.5

0

5 10 15 20 25 30 Retention time (min)


Figure 2

Page 31 of 33

Fractions: F1 F2 F3

Detector Response

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


F4

Vc  0

a.

b.

after tip sonication

BSA (66 kDa)

c.

Thy (670 kDa) 6 µm 4 µm Void

0

PS 2 µm

d.

5 10 15 Retention time (min)

20


Figure 3


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

a.

b.

F1

Page 32 of 33

F2

F3

F4 nucleus (anti-fibrillarin) cytoplasmic tubulins (anti-tubulin) lysosome (anti-LAMP2)

F1 1 µm

F2

mitochondria (anti-ATPB)

F3

peroxisome (anti-catalase) Golgi app. (anti-58K)

F4

ER (anti-calnexin)


Figure 4

Page 33 of 33

R1 R2 R3 R4 R1

nucleus

R2 R3

tubulins lysosome

Detector Response

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


Golgi app. ER

R4

Vc  0 blank

0

5

10

15

20

Retention time (min)


Figure 5

High Speed Size Sorting of Subcellular Organelles by Flow Field-Flow

Recommend Documents