One-Step Detection of Major Lipid Components in ... - ACS Publications

Subscriber access provided by - Access paid by the | UCSB Libraries

Article

One-Step Detection of Major Lipid Components in Submicroliter Volumes of Unpurified Liposome and Cell Suspensions Ssu-Ying Chen, Ching-Yi Wu, Yu-Chie Chen, and Pawel L Urban Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b01740 • Publication Date (Web): 23 Jun 2016 Downloaded from http://pubs.acs.org on June 24, 2016

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 31

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

One-Step Detection of Major Lipid Components in Sub-microliter Volumes of Unpurified Liposome and Cell Suspensions

Ssu-Ying Chen, Ching-Yi Wu, Yu-Chie Chen and Pawel L. Urban*

Department of Applied Chemistry, National Chiao Tung University 1001 University Rd., Hsinchu, 300, Taiwan

Word count: 5623 + 1250 (5 single-column figures × 250) = 6873

* Corresponding author: Dr. P.L. Urban ([email protected])

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 31

ABSTRACT

Liposomes and cells have high lipid contents, which are the main components of the external and internal membranes. Mass spectrometry (MS) is widely used in the analysis of the lipids present in the biological matrices. However, MS analysis of liposome and cell suspensions is challenging due to the presence of other highabundance matrix components (e.g. salts, buffers, and growth media) that cause ion suppression. These interfering species would normally be removed by dialysis or centrifugation. Here we propose a simple and fast method to detect major lipid components in cells and cell suspensions by MS while circumventing dialysis and centrifugation. Capillary hydrodynamic chromatography (HDC) has been coupled online with the aid of an electrospray ionization (ESI) interface to an ion-trap mass spectrometer. Complex samples containing bioparticles and a large amount of potential interferents (buffer, inorganic salts, amino acids) hydrodynamically,

detected

optically

(by

light

were separated

absorption/scattering),

and

immediately transferred to the MS interface. Liposomes and animal cells are disintegrated during electrospray, and the constituent lipids are ionized. The signal-tonoise ratios are ∼10× higher in HDC-ESI-MS than in direct infusion ESI-MS experiments (with or without dilution). This method has been tested on liposomes (containing

phosphatidylcholine

and

phosphatidylglycerol)

and

four

types

of

animal/human cells, i.e. mouse macrophages (RAW 264.7), human breast cancer cells (T47D and Hs578T), and mouse pre-adipocyte cells (3T3-L1). We suggest that HDC-

2 ACS Paragon Plus Environment

Page 3 of 31

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


ESI-MS can be used in quality control analyses of bioparticle suspensions in the fields of biotechnology, molecular biology, drug discovery, and cosmetics.

INTRODUCTION

Modern bioscience requires efficient assays to characterize bioparticles such as cells,1-3 viruses,4-6 and liposomes.7,8 Mass spectrometry (MS) is a powerful technique for direct chemical profiling of bioparticles including cell suspensions.9,10 However, complex bioparticle suspensions often contain molecules and ions11,12 that interfere with mass spectrometric detection.13 For example, cultured animal cells are suspended in growth media containing high concentrations of salts, amino acids, and carbohydrates,11,12 while liposome suspensions normally contain large amounts of residual hydration electrolytes and supplemented with the compounds being encapsulated.14-16 Thus, purification of complex bioparticle samples before MS analysis is required in most cases. One of the first separations via hydrodynamic chromatography (HDC) was conducted by Pedersen in the 1960s.17 The development of HDC is attributed to Small who conducted separation of colloidal particles in 1974.18 In HDC, particles are separated because of the differences in their hydrodynamic diameters. In a typical HDC operation regime, large particles move closer to the center of the separation channel and experience faster mobile phase flow. On the other hand, small particles move slower because they remain closer to the wall of the channel where the mobile phase velocities are relatively low.19 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 4 of 31

accepted model of HDC retention, the dimensionless retention times (τ) of particles with different radii are computed using the following formulae:19 λ=

ோ౛౜౜

(eq. 1)

ோౙ ௧౦

τ = ௧ = (1 + 2λ - Cλ2)-1

(eq. 2)

ౣ

where λ is the aspect ratio derived from the effective radii of the analyte particles (Reff ) and the capillary channel radius (Rc), while tp and tm are the retention times of the analyte and a “point-like particle” (PLP), respectively. The constant C is referred to as quadratic modifier. Various values of C have been used to model separations of different kinds of particles under different conditions. For example, in the case of permeable spheres, the C values may range from ∼ 1 to 3.20 HDC enables fast separations of particle mixtures. Such separations can be conducted using large columns packed with non-porous particles17 or open tubular capillary columns.20,21 Using capillary columns, very small (sub-microliter) volumes of samples can be analyzed. HDC is widely used to separate oligo- and poly-nucleotides.22-30 It has also been used to separate proteins,21 liposomes,31,32 cells,33 as well as nanoparticles.34,35 HDC separations have been coupled to diverse types of off-line and on-line detectors.33-39 For example, microbial cell aggregates, present in probiotic drinks, could be readily separated by capillary HDC and detected by UV absorption/scattering detection.33 Cylindrical illumination confocal spectroscopy was used to detect small quantities of DNA that was separated in thin capillaries.26 HDC was also coupled on-line with inductively coupled plasma MS to investigate nanoparticles and their agglomerates in environmental and artificial matrices.34,39


Page 5 of 31

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


Moreover, off-line coupling of HDC with matrix-assisted laser desorption/ionization MS for analysis of liposome-type nanoparticles has been demonstrated.32 In this report, we demonstrate the usefulness of the on-line coupling of capillary HDC with electrospray ionization (ESI) interface in MS. The proposed apparatus enables separation of bioparticles (liposomes, and cells) from potential interferents (media containing buffers, inorganic salts, and amino acids), and immediate detection by MS.

EXPERIMENTAL SECTION

Materials Ammonium acetate (LC grade), 9-aminoacridine (9-AA), 1,2-dioleoyl-sn-glycero-3phosphocholine (DOPC), 1-(3-sn-phosphatidyl)-rac-glycerol (PG; 34:1) sodium salt, RPMI 1640 and DMEM were purchased from Sigma-Aldrich (St. Louis, MO, USA). 1,2-Dioleoylsn-glycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl; DOPE-rho) was purchased from Avanti Lipid Polar (Alabaster, AL, USA). Water (LC grade) was purchased from Merck (Darmstadt, Germany). Chloroform was purchased from J.T. Baker (Center Valley, PA, USA). All animal/human cells were purchased from Bioresource and Collection Research Center (Hsinchu, Taiwan).



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

Coupling

hydrodynamic

chromatography

with

Page 6 of 31

electrospray

ionization

mass

spectrometry The parameters of the home-made capillary hydrodynamic chromatography device were evaluated in preliminary experiments (Figure S1). The fused silica capillary column (ID: 75 µm; OD: 375 µm; Polymicro Technologies, Phoenix, AZ, USA) length was fixed at 320 cm. The column was directly coupled to a UV imaging absorption detector (ActiPix D100, Paraytec, York, UK) set to ~200 nm with an ion-trap mass spectrometer (amaZon speed; Bruker Daltonics, Bremen, Germany; Figure 1) equipped with an electrospray ion source (Apollo II). The ESI nebulizer gas pressure was ~4 × 104 Pa, the MS inlet voltage was ±4500 V, and the temperature and flow rate of dry gas (N2) were 250 °C and 8 L min−1, respectively. The protrusion of the ESI needle from the nebulizer tube was set to ∼ 0.4 mm. The aqueous bioparticle suspension samples and mobile phase (aqueous 5 mM ammonium acetate solution) were delivered via pressurizing gas (nitrogen; overpressure: ∼ 100 kPa) in the headspace of the inlet vial. A pinch valve controlled the injection time by determining the direction of gas flow. The injection and separation processes were guided by an Arduino Pro Micro microcontroller (Torino, Italy) executing a custom-written script (see ref.33). The mobile phase flow rate was estimated based on the retention time of a chromatographic zone containing small molecules to be ∼ 1.5 µL min-1. This value is similar to the value calculated using the Hagen-Poiseuille equation40 (~ 1.6 µL min-1). Based on these two values, we further estimated the underpressure at the outlet of the ESI nebulizer (∼ -1.1 kPa, relative to the atmospheric pressure). This contributed slightly to the flow. The Reynolds


Page 7 of 31

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


number was estimated to ∼20 and indicating that the hydrodynamic flow inside the separation capillary is mostly laminar (unless perturbed by the injected bioparticles).

Synthesis of liposomes To synthesize liposomes (Figure S2), a chloroform-based solution of 3.18 mM DOPC, 3.19 mM PG (34:1), and 7.69 µM DOPE-rho was prepared. DOPE-rho is a fluorescent molecule integrated with the lipid membrane due to its fatty acyl groups. It enabled observation of liposomes via fluorescent microscopy. Phosphatidylcholine (DOPC) is widely used for the synthesis of liposomes.41,42 Phosphatidylglycerol enhances the stability of the liposome structure (partly due to the negative charge at neutral pH).43 The solvent was evaporated from a 100-µL aliquot of the solution in the stream of nitrogen gas. The resulting lipid cake was subsequently hydrated with 100 µL of 5 mM ammonium acetate solution (pH value: ∼ 7) containing a fluorescent dye (1.5 mM 9-AA). Before HDC-ESI-MS analysis, the liposome suspension was diluted 100× with mobile phase (aqueous 5 mM ammonium acetate solution) to prevent overloading the MS detector with 9-AA ions. The diameters of the synthesized liposomes were within the range of 1-27 µm, while the median value was ~ 11 µm (Figure S3A).

Cell culture and sample preparation Four cell suspensions were tested: RAW 264.7 (mouse macrophage cells); T47D (human breast cancer cells); 3T3-L1 (mouse pre-adipose cells); and Hs578T (human breast cancer cells). The T47D cells were cultured in RPMI medium,12 and the other types of cells were



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

cultured in the DMEM.11 Before analysis, trypsin-EDTA was used to dissociate the cell monolayer from the flask wall. Thus, the cells were suspended in culture media without exerting a substantial mechanical force. Subsequently, aliquots of the cell suspensions were analyzed on the HDC-ESI-MS system without any additional preparation or dilution. The median diameters of RAW 264.7, T47D, 3T3-L1, and Hs578T cells are ∼ 14, 19, 15, and 20 µm, respectively (Figures S3B-E).

Data treatment MS data sets were acquired in the positive-ion mode unless otherwise noted. The MS data acquisition rate was 32,500 u s-1. The total ion current (TIC) and extracted ion current (EIC) data sets were exported using Bruker’s software (DataAnalysis, version 4.1) to ASCII files. Similarly, the UV data sets were exported from Paraytec’s software to ASCII files. Exponential smoothing was applied to the MS chromatograms (time constant: 11.8 s) to reduce temporal noise. In the case of the repeatability test (Figure S4), the chromatographic peaks were integrated using the PeakFit software (version 4.12; Systat Software, San Jose, CA, USA). The baseline was assumed to be linear. The peaks were fitted with the HaarhoffVan der Linde function before integration. To evaluate sensitivity and detection limits (Figure S5), the peak areas were integrated with the OriginPro software (version 9; OriginLab Corporation, Northampton, MA, USA). In the latter method of data treatment, all the irregular chromatographic features contributed by the DOPC ions (m/z 787) were considered. The diameters of liposomes and cells were measured by overlaying the micrograph features with circles using the CorelDraw software (version X6; Corel


Page 8 of 31

Page 9 of 31

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


Corporation, Ottawa, Canada). This was especially helpful while sizing microscopic features with irregular shapes.

RESULTS AND DISCUSSION

Proof of concept This study analyzed aqueous suspensions of liposomes and cells in concentrated multicomponent media by MS. Nonetheless, at the beginning, it was not obvious whether: (i) such bioparticles could give raise to gas-phase ions without prior disruption, extraction, or addition of a makeup solvent; (ii) the high abundance sub-micron species present in the suspension medium would suppress ionization of the bioparticle components and contaminate the instrument. In one of the first attempts, we implemented a simple nanoESI-MS system (without a nebulizer gas) to detect liposomes in post-hydration solution (Figure S6A). This home-made setup incorporated a silver-coated and sharpened capillary that drove the suspension of liposomes towards the orifice of the mass spectrometer and enabled desolvation of microdroplets containing liposomes. However, the lipid signal (1-(3-sn-phosphatidyl)-racglycerol; m/z 747; negative-ion mode) could barely be recorded (Figure S6D). To address that technical problem, we switched to a conventional ESI emitter with nebulizer gas and pumped the liposome suspension hydrodynamically toward the MS orifice (Figure S7). In that case, the lipid signal was more intense and stable. However, using ESIMS alone could not eliminate the ion suppression due to the presence of concentrated post-



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

synthesis matrix. Moreover, direct infusion of post-synthesis suspensions of liposomes to MS for a long time (or when analyzing multiple samples) resulted in quick contamination of the ion optical elements of the mass spectrometer. To reduce ion suppression and contamination of vulnerable parts of the mass spectrometer, we coupled hydrodynamic chromatographic separation with the ESI interface (Figure 1). The resulting HDC-ESI-MS system enabled rudimentary separation of liposomes and components of the medium (e.g. liposome hydration mixture). In this system, micrometer-scale particles are transferred to the ESI source before the arrival of PLPs (small and large molecules with nanometer scale size). Larger particles (liposomes, cells) move faster because they can pick up momentum of the mobile phase laminae more than smaller particles (e.g. buffer, fluorescent dye and contaminant molecules present in the sample). In principle, HDC fractions can be obtained at the column outlet and analyzed off-line by MS.27 However, on-line coupling of HDC with ESI-MS is more straightforward because unpurified suspensions can be analyzed in one step to circumvent fraction collection. The repeatability of the HDC-UV-ESI-MS analysis was tested by conducting 10 consecutive injections of a caffeine standard sample (5 × 10-5 M; Figure S4). The RSD values of peak area are 17% and 12% in the UV and MS, respectively (Table S1). This is satisfactory considering that a prototype HDC instrument was used.

Separation of liposomes and a low-molecular-weight dye The on-line analysis of a post-hydration liposome suspension by HDC-UV-ESI-MS shows good separation of major lipids (i.e. DOPC; m/z 787; positive-ion mode) and small molecules


Page 10 of 31

Page 11 of 31

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


(fluorescent dye: 9-AA) in both UV and MS chromatograms (Figure 2). Baseline separation was achieved in both cases. The lower resolution in the MS chromatogram (1.5) than the UV chromatogram (3.4) is probably related to transient adsorption of liposome-fragments on the surfaces around the MS inlet. However, this may also be due to the irregular shapes of the temporal features in the MS chromatograms and the inaccuracy of the curve fitting. Moreover, different species are detected by UV and MS detectors (light-scattering particles vs. lipid ions produced from disintegrated liposomes). The suspension of newly synthesized liposomes is polydisperse in terms of size and structure (Figures 3A and S3A). The polydispersity of the liposome sample explains the irregular shape of the chromatographic feature corresponding to liposome in Figures 2A and 2B. The noise in the MS chromatograms (Figures 2B and 2C) could probably be reduced by replacing the ion-trap with a triple-quadrupole—a mass analyzer that is more suitable for detection of chromatographic output and quantitative analyses. When analyzing varied numbers of liposomes by HDC-ESI-MS, one can observe that the peak area of the lipid component in liposomes (EIC at m/z 787; DOPC) correlates somewhat with the estimated number of liposomes in the injection plug (Figure S5). Because the liposome suspension is non-homogeneous, precipitation and injection bias may occur. Thus, the relationship between the MS signal and the predicted number of injected liposomes is not linear. The limit of detection for liposomes was estimated to be ∼ 200 liposomes per injection (∼ 100 nL) according to the criterion of signal-to-noise ratio (S/N) = 3. However, a larger sample volume (>10 µL) had to be used to enable injection by dipping the capillary column inlet into the sample and applying overpressure. It has been estimated44,45 that every



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

liposome (average diameter: ∼ 11 µm) contains ∼ 1.4 picograms of lipids. Thus, 200 liposomes carry ~ 0.28 nanograms of the lipid component to the ESI source. The synthesized liposomes possess electric charge (resulting from the phosphate groups), are sensitive to the environment (pH, buffer type),46 and are vulnerable to electrostatic forces in the proximity of another electrically charged material (e.g. inner wall of silica capillary). Thus, they may deform or even become damaged. However, we found that the liposomes can be readily driven along a microscale silica capillary by hydrodynamic flow without obvious disruption of their structure (Figure 3).

Separation of animal cells and low-molecular-weight residues To further demonstrate the usefulness of this method, suspensions containing four cell lines were tested in the HDC-ESI-MS system. The resulting chromatograms show the separation of cells and free molecules (including plastic-related contaminant species) present in the growth medium (Figure 4). The cells were readily separated from PLPs (medium components; cf. Figure S8). The majority of the cellular lipids could be detected despite the high concentration of salts in the cell growth media (∼ 100 mM).11 This is because the cells left the separation column before the concentrated salt zone. In fact, the cell suspension was not diluted before injection. Due to the hydrodynamic separation in the capillary column, ion suppression due to the salt-rich matrix was minimized. Moreover, the mobile phase slightly dilutes the salt-rich matrix due to the so-called miscible dispersion, which is related to convection and molecular diffusion present in the laminar flow47. The local dilution of the salt-rich matrix zone may further minimize ion suppression. Similar to the liposome


Page 12 of 31

Page 13 of 31

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


chromatograms (Figure 2B), the cell chromatograms show irregular lipid features (Figure 4, middle traces). This is due to the polydispersity of the cell suspensions. The EICs from cell separations (Figure 4, middle) are simpler than the EIC from liposome separation (Figure 2B) because the cultured animal cells are less polydisperse than the liposomes used in this study (Figure S3). Overall, in this low-pressure separation system, the HDC step essentially sorts matrix components into two classes: (i) bioparticles with diameters of several to several tens of micrometers; and (ii) point-like particles (< 1 µm). When the former are liposomes or animal cells, they give raise to lipid-related signals. In fact, membrane lipids are major components of such bioparticles. In ESI, they form ions within the m/z 500-1000 in the negative-ion mode (due to deprotonation) and positive-ion mode (due to adduction of small cations such as H+, NH4+, Na+, or K+). For example, the feature in the T47D cell chromatogram within a time range of 451-550 s corresponds to the MS signal at the m/z 761 among others. Fragmentation of the ion at m/z 761 during direct infusion of T47D cell suspension to the ESI-ion-trap mass spectrometer produced phosphocholine ion (C5H15NO4P+) at the m/z 184 (Figure S9A). This result suggests that the parent ion (m/z 761) is formed from phosphatidylcholine. In another control experiment, lipids were extracted from the cells via the modified Bligh-Dyer extraction.48 Fragmentation of the parent ions at the m/z 761 and 787 (same as those displayed in Figure 4) during direct infusion of ESI-triple-quadrupole MS/MS analyses of the extracts from the four types of cells also yielded phosphocholine ions (m/z 184; Figure S10). The extracts were subsequently analyzed by reversed phase (RP) liquid chromatography (LC)49 coupled to a triple quadrupole mass spectrometer operated in 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

precursor-ion scan mode (fragment ion m/z: 184). Although no absolute quantification was performed in this control experiment, the RP-LC-MS result (Figure S11) suggests that the target ions shown in Figure 4 are among the most abundant phosphatidylcholines present in these cells. In fact, phosphatidylcholines are abundant in animal cell membranes.50 The chromatographic feature from 551 to650 s contains numerous sub-micron species including small molecules such as the plasticizer (diisooctyl phthalate) producing ions at the m/z 413 (fragmenting to the m/z 171 and 301; Figure S9B). This compound can be readily detected by ESI-MS.51 It may originate from the pipette tips, plastic containers and microcentrifuge tubes used to handle solutions and cell suspensions. It might be useful to couple HDC-ESI with a high-resolution mass analyzer to expedite identification of molecules present in the bioparticles separated by HDC. It would also be practical to couple HDC-ESI with an ion mobility MS instrument to enable real-time separations of individual lipids and their isomers immediately after elution from the HDC column. Comparison of the S/N ratios of the MS signals within the cell lipid-related chromatographic zones and the S/N ratios of the MS signals recorded during continuous injection of cell suspensions (without and with dilution) shows an increase in the S/N ratios ∼ 10× (Figure 5). This increase is attributed to the removal of ion-suppressing interferents by HDC. In this comparative study, the noise was estimated by averaging the intensities of signals present in the mass spectra within the m/z ranges: 408-411 and 416-419; 750-753 and 765-768; and 777-780 and 792-795 in the case of the three target analytes (represented by MS peaks at the m/z 413, 761, and 787). In continuous injection ESI-MS as well as the HDC-ESIMS, every data point was based on a 20-s average spectrum within a 100-s interval. Labels (a-


Page 14 of 31

Page 15 of 31

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


e) in Figure 5 indicate the consecutive 20-s intervals within the 100-s window containing that chromatographic feature of interest (at the m/z 413, 761, or 787). Notably, the data points labeled with (b) and (c) represent the highest values because they correspond to the retention times closest to the apexes of the investigated chromatographic features. Thus, for qualitative or semi-quantitative analysis of a given bioparticle sample—after collecting MS data for the whole separation—one can select a time interval within the MS chromatogram that provides the highest S/N ratio and compute the average mass spectrum for that interval. The retention times of liposomes and animal cells are different because their diameters are slightly different (cf. Table S2 and Figure S3). In a rudimentary estimation—and considering the τ and λ computed for the five lipid-rich particles included in this study—we did not find a consistent C value that could be used with eq. 2. This is because the particles are quite large (> 10 µm) relative to the capillary diameter (75 µm), and their separation from PLPs (e.g. unbound small molecules and polymers) cannot be predicted with the standard HDC model (eqs. 1 and 2). Moreover, because the diameters of some of the analyzed bioparticles are larger than 20% of the chromatographic capillary diameter (∼ 15 µm vs. 75 µm), they considerably disturb the Poiseuille flow while traversing the HDC column.19 Although retention times cannot be accurately predicted using the standard HDC model because of this disruption of Poiseuille flow, the HDC-ESI-MS method still provides a convenient way to reduce ion-suppression while analyzing complex bioparticle suspensions by MS.



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

CONCLUDING REMARKS

We have presented a prototype system combining capillary hydrodynamic chromatography and electrospray ionization mass spectrometry. It enables separation and simultaneous analysis of samples containing liposomes, cells, components of growth medium, contaminants, and residual molecules left after encapsulation. To the best of our knowledge, this is the first time in which small molecules and cells or liposomes have been separated and detected by online ESI-MS in one run. This method offers two main advantages: First, the main components of the bioparticles can be detected in one step to circumvent centrifugation or dialysis. The high-abundance ion suppressing molecules (e.g. salts, growth medium components) are eluted from the HDC column after elution of the analyzed bioparticles. Second, the method allows one to study the disproportionation of high-abundance molecules between the solution (unbound form) and the bioparticles (bound form). Due to the heterogeneous nature of bioparticle suspensions, and the related injection variability, the method can be regarded as qualitative or semi-quantitative. The assembly of the HDC-ESI-MS system is straightforward. Unlike in capillary electrophoresis, separation does not require electric voltage. Thus,interfacing HDC column with ESI is simple and uses inexpensive parts. The capillary columns can be readily replaced if damaged, clogged, or contaminated. Maintenance is simple. The volumes of samples and mobile phases consumed in every analysis are very small (nanoliter range). We believe that HDC-ESI-MS can be further used in quality control analyses of bioparticle suspensions in the fields of biotechnology, molecular biology, drug discovery, and cosmetics. Miniature mass spectrometers are increasingly used


Page 16 of 31

Page 17 of 31

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


for quality control, following little or no sample preparation. Ion sources (including ESI) in some commercial devices are consumable items. Implementing HDC prior to ESI-MS in such on-site analyses can reduce buildup of sample matrix on the surfaces near the MS inlet and increase the lifetime of those consumable parts. This will reduce the cost of replacement and maintenance. At the same time, complex samples with lipid-rich particles and high salt concentrations can be analyzed.

ASSOCIATED CONTENT Supporting Information Contents: Supporting tables (S1, S2) and supporting figures (S1-S11) The Supporting Information is available free of charge on the ACS Publications website.

ACKNOWLEDGEMENTS We thank Mr Hsu Ting and Mr Yu-Ching Lo for conducting preliminary tests on liposomes, Dr Jie-Bi Hu and Ms Ewelina P. Dutkiewicz for discussions, as well as Mr Chun-Hsien Li, Ms Sin-Ge Wang and Mr Kai-Chieh Li for providing us various cell samples. We thank Paraytec Ltd (York, UK) for providing us the ActiPix D100 UV imaging detector. We also acknowledge the National Chiao Tung University and the Ministry of Science and Technology, Taiwan (formerly, National Science Council;



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

grant numbers NSC 102-2113-M-009-004-MY2 and MOST 104-2628-M-009-003MY4) for the financial support of this work.


Page 18 of 31

Page 19 of 31

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


REFERENCES 1.

Ince, T. A.; Sousa, A. D.; Jones, M. A.; Harrell, J. C.; Agoston, E. S.; Krohn, M.; Selfors, L. M.; Liu, W.; Chen, K.; Yong, M.; Buchwald, P.; Wang, B.; Hale, K. S.; Cohick, E.; Sergent, P.; Witt, A.; Kozhekbaeva, Z.; Gao, S.; Agoston, A. T.; Merritt, M. A.; Foster, R.; Rueda, B. R.; Crum, C. P.; Brugge, J. S.; Mills, G. B. Nat. Commun. 2015, 6, 7419.

2.

Martí, M.; Mulero, L.; Pardo, C.; Morera, C.; Carrió, M.; Laricchia-Robbio, L.; Esteban, C. R.; Belmonte, J. C. I. Nat. Protoc. 2013, 8, 223-253.

3.

Hao, H.-C. Y., D.-J. Curr. Top. Med. Chem. 2015, 15, 1543-1550.

4.

Dienstag, J. L. Yale J. Biol. Med. 1980, 53, 61-69.

5.

Guan, Y.; Zheng, B. J.; He, Y. Q.; Liu, X. L.; Zhuang, Z. X.; Cheung, C. L.; Luo, S. W.; Li, P. H.; Zhang, L. J.; Guan, Y. J.; Butt, K. M.; Wong, K. L.; Chan, K. W.; Lim, W.; Shortridge, K. F.; Yuen, K. Y.; Peiris, J. S. M.; Poon, L. L. M. Science 2003, 302, 276278.

6.

Verma, M. S.; Rogowski, J. L.; Jones, L.; Gu, F. X. Biotechnol. Adv. 2015, 33, 666-680.

7.

Ruozi, B.; Belletti, D.; Tombesi, A.; Tosi, G.; Bondioli, L.; Forni, F.; Vandelli, M. A. Int. J. Nanomedicine 2011, 6, 557-563.

8.

Chen, C.; Zhu, S.; Huang, T.; Wanga, S.; Yan, X. Anal. Methods 2013, 5, 2150-2157.

9.

Tobias, H. J.; Pitesky, M. E.; Fergenson, D. P.; Steele, P. T.; Horn, J.; Frank, M.; Gard, E. E. J. Microbiol. Methods 2006, 67, 56-63.



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

10. Bendall, S. C.; Simonds, E. F.; Qiu, P.; Amir, E. D.; Krutzik, P. O.; Finck, R.; Bruggner, R. V.; Melamed, R.; Trejo, A.; Ornatsky, O. I.; Balderas, R. S.; Plevritis, S. K.; Sachs, K.; Péer, D.; Tanner, S. D.; Nolan, G. P. Science 2011, 332, 687-696. 11.

Dulbecco, R.; Freeman, G. Virology 1959, 8, 396-397.

12.

Moore, G. E.; Gerner, R. E.; Franklin, H. A. JAMA 1967, 199, 519-524.

13.

Kandiah, M.; Urban, P. L. Chem. Soc. Rev. 2013, 42, 5299-5322.

14.

Buttino, I.; De Rosa, G.; Carotenuto, Y.; Ianora, A.; Fontana, A.; Quaglia, F.; La Rotonda, M. I.; Miralto, A. J. Exp. Biol. 2006, 209, 801-809.

15.

Nomura, S. M.; Kondoha, S.; Asayama, W.; Asada, A.; Nishikawa, S.; Akiyoshia, K. J. Biotechnol. 2008, 133, 190-195.

16.

Kamiya, K.; Tsumoto, K.; Yoshimura, T.; Akiyoshi, K. Biomaterials 2011, 32, 98999907.

17.

Pedersen, K. O. Arch. Biochem. Biophys. 1962, Suppl. 1, 157-168.

18.

Small, H. J. Colloid and Interface Sci. 1974, 48, 147-161.

19.

Striegel, A. M.; Brewer, A. K. Annu. Rev. Anal. Chem. 2012, 5, 15-34.

20.

Tijssen, R.; Bos, J.; Vankreveld, M. E. Anal. Chem. 1986, 58, 3036-3044.

21. Kraak, J. C.; Oostervink, R.; Poppe, H.; Esser, U.; Unger, K. K. Chromatographia 1989, 27, 585-590. 22.

Hirabayashi, J.; Kasai, K.-I. Anal. Biochem. 1989, 178, 336-341.

23.

Iki, N.; Kim, Y.; Yeung, E. S. Anal. Chem. 1996, 68, 4321-4325.

24.

Peyrin, E.; Guillaume, Y. C.; Villet, A.; Favier, A. Anal. Chem. 2000, 72, 853-857.


Page 20 of 31

Page 21 of 31

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.

Wang, X.; Veerappan, V.; Cheng, C.; Jiang, X.; Allen, R. D.; Dasgupta, P. K.; Liu, S. J. Am. Chem. Soc. 2010, 132, 40-41.

26.

Liu, K. J.; Rane, T. D.; Zhang, Y.; Wang, T.-H. J. Am. Chem. Soc. 2011, 133, 68986901.

27.

Zhu, Z.; Chen, H.; Wang, W.; Morgan, A.; Gu, C.; He, C.; Lu, J. J.; Liu, S. Angew. Chem. Int. Ed. 2013, 52, 5612-5616.

28.

Liu, L.; Veerppan, V.; Pu, Q.; Cheng, C.; Wang, X.; Lu, L.; Allen, R. D.; Guo, G. Anal. Chem. 2014, 86, 729-736.

29.

Chen, H.; Zhu, Z. Lu, J. J.; Liu, S. Anal. Chem. 2015, 87, 1518-1522.

30.

Friedrich, S. M.; Liu, K. J.; Wang, T.-H. J. Am. Chem. Soc. 2016, 138, 319-327.

31.

Lucas, F. J. M.; Buso, A. O. V.; Bosch, P. D.; Alejano, J. E. F. J. High Resol. Chromatogr. 1989, 12, 560-561.

32.

Helsper, J. P. F. G.; Peter, R. J. B.; Brouwer, L.; Weigel, S. Anal. Bioanal. Chem. 2013, 405, 1181-1189.

33.

Tang, Y.-R.; Huang, H.-Y.; Hu, J.-B.; Rattinam, R.; Lee, C.-H.; Chen, Y.-C.; Urban, P. L. Anal. Chim. Acta 2016, 910, 75-83.

34.

Pergantis, S. A.; Jones-Lepp, T. L.; Heithmar, E. M. Anal. Chem. 2012, 84, 6454-6462.

35.

Rakcheev, D.; Philippe, A.; Schaumann, G. E. Anal. Chem. 2013, 85, 10643-10647.

36.

Thompson, J. W.; Lieberman, R. A.; Jorgenson, J. W. J. Chromatogr. A 2009, 1216, 7732-7738.

37.

Umehara, R.; Miyahara, H.; Okino, A.; Harada, M.; Okada, T. Anal. Sci. 2012, 28, 359365.



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

38.

Edam, R.; Eeltink, S.; Vanhoutte, D. J. D.; Th. Kok, W.; Schoenmakers, P. J. J. Chromatogr. A 2011, 1218, 8638-8645.

39.

Philippe, A.; Schaumann, G. E. PLoS One 2014, 9, e90559.

40.

Pfitzner, J. Anaesthesia 1976, 31, 273-275.

41.

Beaune, G.; Ménager, C. J. Colloid Interface Sci. 2010, 343, 396-399.

42. Boudier, A.; Castagnos, P.; Soussan, E.; Beaune, G.; Belkhelfa, H.; Ménager, C.; Cabuil, V.; Haddioui, L.; Roques, C.; Rico-Lattes, I.; Blanzat, M. Int. J. Pharm. 2011, 403 230236. 43.

Marsh, D. Biophys. J. 2012, 102, 1079-1087.

44.

Güven, A.; Ortiz, M.; Constanti, M.; O'Sullivan, C. K. J. Liposome Res. 2009, 19, 148154.

45.

Encapsula

Nano

Science.

http://www.liposomes.org/2009/01/number-of-lipid-

molecules-per-liposome.html. (view on 06/14/2016) 46.

Hamada, T.; Hagihara, H.; Morita, M.; Vestergaard, M. C.; Tsujino, Y.; Tsujino, M. J. Phys. Chemi. Lett. 2012, 3, 430-435.

47.

Probstein, R. F. Physicochemical Hydrodynamic . 2nd, ed. Wiley: Hoboken. 1995.

48.

Bligh, E. G.; Dyer, W. J. Can. J. Biochem. Physiol. 1959, 37, 911-917.

49.

Hewelt-Belka, W.; J. N., Belka, M.; Bączek, T.; Namiésnik, J.; Kot-Wasik, A. J. Chromatogr. A 2014, 1362, 62-74.

50.

Li, Z.; Agellon,L. B.; Allen, T. M.; Umeda, M.; Jewell, L.; Mason, A.; Vance, D. E. Cell Metab. 2006, 3, 321-331.

51.

Keller, B. O.; Sui, J.; Young, A. B.; Whittal, R. M. Anal. Chim. Acta 2008, 627, 71-81.


Page 22 of 31

Page 23 of 31

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


FIGURE CAPTIONS

Figure 1. On-line coupling of hydrodynamic chromatography with mass spectrometry. (A) Schematic of the HDC-ESI-MS setup. (B) Bioparticle separation in the HDC capillary. (C) Putative fate of the separated bioparticles during the subsequent electrospray process.

Figure 2. Positive-ion mode HDC-ESI-MS result demonstrating separation of liposomes from other components of the liposome suspension (e.g. fluorescent dye, buffer). (A) UV absorption/scattering chromatogram (wavelength: 200 nm); (B) EIC at the m/z 787 revealing the lipid component (e.g. DOPC) in the synthesized liposomes; (C) EIC at the m/z 195 revealing 9-aminoacridine (a low-molecular-weight fluorescent dye). The yellow and grey highlights match the corresponding features in the UV and MS chromatograms while considering the time gap between the UV and MS detection. Blue bar (absorbance) represents 2 × 10-4 while black bars represent ion intensity (2 × 104 a.u). Exponential smoothing has been applied to MS experimental datasets (time constant = 11.8 s). Fused silica capillary column: total length, 320 cm; length from the inlet to the UV detector, 295 cm; length from the UV detector to ESI, 25 cm; OD, 375 µm; ID, 75 µm.

Figure 3. Micrographs of rhodamine-labeled and 9-aminoacridine encapsulated inside liposome suspension used as the test sample. (A) Fresh liposome suspension before



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

HDC (λex / λem = 510-560 / 572-648 nm); (B) HDC fraction collected from 3-m capillary column for 1 min (during separation, 4.5-5.5 min; λex/λem = 510-560 / 572-648 nm). The liposomes are intact after traversing the HDC column. (C) Empty silica capillary column (ID = 75 µm) viewed in bright field. (D) Superimposed fluorescence images revealing liposomes (containing 9-aminoacridine and rhodamine) inside a section of capillary column (λex/λem = 460-500 / 510-560 and 510-560 / 572-648 nm). Image brightness has been adjusted. Scale bars: 20 µm.

Figure 4. Positive-ion mode HDC-ESI-MS results obtained for four types of cultured cells: (A) RAW 264.7 (4.2 × 105 cells mL-1) in DMEM; (B) T47D (2.0 × 106 cells mL-1) in RPMI; (C) 3T3-L1 (3.8 × 106 cells mL-1) in DMEM; and (D) Hs578T (6.4 × 105 cells mL-1) in DMEM. Chromatograms (from top to bottom): UV absorption/scattering (wavelength: 200 nm); EIC of an abundant phosphatidylcholine present in the cells (m/z 761 or 787); and the EIC of a plasticizer (m/z 413) present in the culture medium. The dotted grey lines match the corresponding features in the UV and MS chromatograms considering the time gap between the UV and MS detection (delay time: ∼ 100 s). Blue bars (absorbance) represent 2 × 10-2 while black bars (ion intensity) represent 5 × 103 a.u. Exponential smoothing has been applied to MS experimental datasets (time constant = 11.8 s). Fused silica capillary column: total length, 320 cm; length from the inlet to the UV detector, 295 cm; length from the UV detector to ESI, 25 cm; OD, 375 µm; ID, 75 µm. PLP: point-like particle.


Page 24 of 31

Page 25 of 31

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


Figure 5. Comparison of signal-to-noise ratios obtained in direct continuous injection to ESIMS of the same samples (with and without dilution with 20 mM ammonium acetate solution) and in HDC-ESI-MS (positive-ion mode). The noise values were determined by averaging intensities of baseline signals in the m/z ranges: 408-411 and 416-419; 750-753 and 765-768; as well as 777-780 and 792-795 in the case of the analyte signals at the m/z 413 ± 0.5, 761 ± 0.5 and 787 ± 0.5, respectively. In both continuous injection ESI-MS and HDC-ESI-MS, every data point is based on a 20-s average spectrum within a 100-s interval. The time period of 20-120 s was selected for continuous injection ESIMS. Letters (a-e) indicate the consecutive 20-s intervals within the 100-s windows containing that chromatographic features in the HDC-ESI-MS data sets; the (b) and (c) labels represent the highest S/N values because they are obtained from the time intervals closest to the peak apexes.



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

Figure 1


Page 26 of 31

Page 27 of 31

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


Figure 2



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

Figure 3


Page 28 of 31

Page 29 of 31

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


Figure 4



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

Figure 5


Page 30 of 31

Page 31 of 31

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


Graphical abstract


One-Step Detection of Major Lipid Components in ... - ACS Publications

Recommend Documents