Investigating the Uptake of Arsenate by Chlamydomonas reinhardtii

Article Cite This: Anal. Chem. XXXX, XXX, XXX−XXX

pubs.acs.org/ac

Investigating the Uptake of Arsenate by Chlamydomonas reinhardtii Cells and its Effect on their Lipid Profile using Single Cell ICP−MS and Easy Ambient Sonic-Spray Ionization−MS Emmanouil Mavrakis,†,§ Leonidas Mavroudakis,†,§ Nikos Lydakis-Simantiris,‡ and Spiros A. Pergantis*,†

Downloaded via KEAN UNIV on July 23, 2019 at 08:04:38 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

Environmental Chemical Processes Laboratory, Department of Chemistry, University of Crete, Voutes Campus, Heraklion 70013, Greece ‡ Laboratory of Environmental Chemistry and of Biochemical Processes, Department of Agriculture, Hellenic Mediterranean University, Chania 73133, Greece S Supporting Information *

ABSTRACT: The complementary use of single cell atomic mass spectrometry (MS) and ambient molecular MS allowed for the indepth study of arsenate uptake by Chlamydomonas reinhardtii cells and of the effect this toxic metalloid species has on their lipid profile. Compared to conventional inductively coupled plasma mass spectrometry (ICP−MS) analysis, in which case hundreds of thousands of cells are digested and then analyzed, it is demonstrated that single cell (SC) ICP−MS provides uptake data that are potentially of greater biological relevance. This includes the arsenic mass distribution within the cell population, which fits to a lognormal probability function, the most frequently contained arsenic mass within the cells (1.5−1.8 fg As per cell), and the mean arsenic uptake value (ranging from 2.7 to 4.1 fg As per cell for the three arsenate incubation concentrations, that is, 15, 22.5, and 30 μg As per mL) derived from the log-normal arsenic mass distribution within the cell population. The SC approach also allows for differentiating the arsenic present in and/or adsorbed on the cells, from the arsenic present in the extracellular solution, in a single analysis. In a similar fashion, ambient molecular MS in the form of desorption easy ambient sonic spray ionization (EASI) -MS was used to rapidly profile cell membrane lipids from cells spotted directly on a glass slide. EASI−MS analysis revealed that cells grown in the presence of increasing concentrations of arsenate exhibited changes in the degree of saturation of their membrane lipids, as was observed by the increasing intensity ratio of lipids with less unsaturated acyl chains to the same type of lipids with more unsaturated fatty acid chains. Thus, indicating “homeoviscous adaptation” of extraplastidial and thylakoid cell membranes, induced by the presence of arsenate.

T

regarding the variability of As uptake within a cell population (cell-to-cell variation), and thus reveal cell heterogeneity with respect to metal uptake. Information which is potentially useful to better understand cell behavior in the presence of toxic metal species or metallodrugs. In order to overcome these limitations, considerable effort has been invested in developing a single cell (SC) metal determination approach. This involves using inductively coupled plasma mass spectrometry (ICP−MS), a highly sensitive atomic spectrometric technique, for the analysis of dilute cell suspensions in order to determine the metal content of individual cells as individual cell detection events. Extremely short detector dwell times, on the ms to μs time scale, are used in SC-ICP−MS in order to maximize the probability of

o date, several studies have reported on the uptake of arsenic (As) compounds by various strains of Chlamydomonas reinhartdii.1−7 In most cases, conventional analytical schemes were used in order to determine As uptake by this unicellular photosynthetic green algae. This included acid digestion of large numbers of cells, followed by the use of an atomic spectrometric technique for As determination.1,3−6 Even though this is a well-established approach, validated for its quantitative accuracy, it has several limitations with respect to providing further insight into the actual uptake behavior of individual cells. So even though the approach provides a mean As amount per cell, if cell numbers are known, it does not allow for the assessment of the biological relevance of this value. In fact, the mean value may not be representative at all of how much As is actually present in each cell, particularly if the analyzed cell population has a broad As distribution. It is therefore evident that when using a conventional analytical approach for cell analysis we are unable to gain information © XXXX American Chemical Society

Received: February 19, 2019 Accepted: July 4, 2019 Published: July 4, 2019 A

DOI: 10.1021/acs.analchem.9b00917 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

ambient sonic-spray ionization mass spectrometry45 (EASI− MS, also known as desorption sonic-spray ionization mass spectrometry, DeSSI−MS). EASI−MS was the ambient MS technique used for lipid profiling throughout this study. Chlamydomonas reinhadtii was selected in this study as a model organism due to its extensive use in many types of biochemical studies,46 and is therefore representative of what can be achieved when using the proposed combined SC-ICP− MS/EASI−MS approach. Here we wish to demonstrate the rapid nature of the combined approach, which avoids timeconsuming sample preparation steps for the analysis of single cells or multiple cells in a state that is as close to their native state as possible; thus allowing for the evaluation of As uptake and lipid profile changes directly from untreated intact cells. It is also of significance to demonstrate that this extensive cell analysis can be achieved using limited numbers of cells, that is, approximately 18 000 cells per SC-ICP−MS analysis and about 10 000−50 000 cells for each EASI−MS sample spot.

detecting single cell events and avoid multiple cell events within the same dwell time. Once a single cell enters the ICP, it produces a cloud of atomic ions originating from most of its constituent elements. As a result, detection of the element or metal of interest allows for its quantitative determination on a per cell basis. Although the approach is still being developed, it has already demonstrated great potential and several advantages for advancing the determination of metals in cells.8−21 Advantages include its extremely high sensitivity and capability to provide information about elemental mass distributions in cell populations, as well as its requirement for limited sample preparation, which is typically only cell dilution prior to analysis. However, in several biochemical studies, gaining insight into how metal uptake affects cellular metabolism is also important, i.e., how the metal affects the cell’s proteome, metabolome or lipidome. Molecular MS has been used extensively to study metabolic changes in a wide range of cellular-based studies.22−24 Due to sensitivity issues, however, molecular MS is still in its infancy with respect to single cell analysis.25 Nevertheless, molecular MS has achieved considerable progress through the advent and rapid development of ambient MS techniques, which do not require substantial sample preparation for cell analysis. Therefore, they can be used for rapid, high throughput, and sensitive cell analysis.25,26 In order to obtain both detailed cellular metal uptake and lipidomic data, which is the objective of the present study, state-of-the-art atomic and molecular MS techniques must be used in combination, that is, SC-ICP−MS and ambient MS. The present study involved cultivating Chlamydomonas reinhardtii cells in elevated arsenate (iAs(V)) concentrations, in order to investigate As uptake and membrane lipid profile changes. It has been reported that C. reinhardtii cells will uptake arsenate and continue to grow, providing arsenate is present at sublethal concentrations.1−3,6 Growth and photosynthesis inhibition are toxic effects that are induced by the presence of arsenate.4,6 Arsenate can enter cells through phosphate transporters due to their structural similarity and the first step toward detoxification is its reduction to arsenite with subsequent methylation.1,2,4 However, it seems that cell membrane lipids are affected by these toxic arsenic species as evident by studies on arsenite-oxidizing bacteria,27 arsenicresistant bacteria,28 and human erythrocytes29 in which the degree of saturation of membrane lipids increased as a result of the increased content of saturated membrane lipids, a mechanism known as “homeoviscous adaptation”. In another report, Chlorella vulgaris cells were incubated with arsenate and a fluorescent probe (calcein), in which case it was observed that calcein was accumulated in the cells; indicating that cell membranes were more permeable.30 To the best of our knowledge, there have been no reports so far on the effect of arsenate on the membrane lipids of C. reinhardtii cells. Among the most common methods for lipid analysis and profiling of cells are mass spectrometric techniques coupled to either liquid (LC−MS)31−35 or gas chromatography (GC− MS).36−38 Despite the high sensitivity and quantitative accuracy of these techniques, lipids must be extracted from cells in a time-consuming sample preparation procedure. Other methods that require fewer sample preparation steps include single cell matrix-assisted laser desorption ionization mass spectrometry (SC-MALDI−MS)39 and ambient ionization techniques such as desorption electrospray ionization mass spectrometry (DESI−MS),40−42 nano-DESI−MS43,44 and easy

■

EXPERIMENTAL SECTION Cell Cultivation. The cells that were used in this study were the wild-type strain Chlamydomonas reinhardtii CC-1690 purchased from Chlamydomonas Resource Center, University of Minnesota. C. reinhardtii cell colonies were transferred from agar plates into 3 L of cultivation medium (tris-acetate phosphate, TAP, pH 7.0, supplemented with acetic acid as a carbon source) and cultivated for 5 days under continuous illumination (∼2500 lx) and stirring at 25 °C. Subsequently, the cell culture was examined under a microscope for the presence of organisms other than C. reinhardtii. Axenic cultures were used to inoculate fresh TAP media, which were spiked with appropriate volumes of sodium arsenate solution so that the final calculated concentration would be 0 (control), 15, 22.5, and 30 μg As mL−1. All the cell cultures were grown under identical conditions for 5 days, at which point cells were harvested by centrifugation (4000g, 5 min, 4 °C). Remainings of the TAP media were removed by a second centrifugation under the same conditions and cells were resuspended in 150 mM NaCl, 4 mM MgCl2·6H2O, 20 mM Trizma base, pH 7.0. Finally, cells were centrifuged and transferred into a highdensity sucrose solution (0.8 M sucrose, 50 mM Trizma base, pH 7.0), used as a cryoprotectant, and stored in small aliquots at −80 °C until used. Samples that had been thawed for analysis, were not refrozen and reanalyzed. Chemicals and Reagents. Acetonitrile (ACN) of HPLC CHROMASOLV grade and N,N-dimethylformamide (DMF) were obtained from Sigma-Aldrich (Seelze, Germany). Deionized (d.i.) water (18.2 ΜΩ cm−1, PURELAB Option-S, Elga Labwater, UK) was provided by an in-house water purification system. All the reagents used for the cultivation of cells were of the highest grade possible. Sodium arsenate heptahydrate, AsHNa2O4·7H2O was obtained from SigmaAldrich. Single Cell ICP−MS Analysis. Analysis was carried out on cells prepared as described above (unwashed cells) and on cells which were subsequently treated with EDTA solution (50 mM Trizma, 1 mM EDTA, pH 7) in order to remove any iAs(V) adsorbed to the cell wall (washed cells). For quantitation of the cell populations, a Neubauer hemocytometer was used. Subsequently, appropriate volumes of stock cell suspensions were diluted in deionized water at a number concentration of 105 cells mL−1 and analyzed within 10 min of dilution. Arsenic determination in cells was carried out using a NexION 300 X B


Article


As is the case in single particle ICP−MS, a threshold value should be determined in order to discriminate between the continuous signal, deriving from the background or dissolved element, and the cell events. Among the most widely used strategies is the iterative 3σ approach introduced by Pace et al.,47,48 according to which all signals above the threshold, determined by triple the standard deviation (3σ) added to the background mean (m), correspond to nanoparticles. The validity of this approach relies on the assumption that the dissolved or background signal is normally distributed. Yet in this study, the use of short dwell time (50 μs) and the extremely low concentration levels of dissolved As in the cell suspensions deviate significantly from the assumed normal distribution of the blank and thus render the 3σ approach for calculating the Limit of Detection (LD) inappropriate. The 75 As intensity acquisition over time for the control C. reinhardtii cells (SI Figure S2A), illustrating simply the background 75As intensity, and the corresponding intensity histogram (SI Figure S2B) dictate the need for a more suitable formula for LD calculation that would incorporate Poisson statistics. Herein, the Limit of Detection is calculated as

ICP−MS (PerkinElmer, Shelton, CT) equipped with a HighEfficiency Introduction System appropriate for Single-Cell analysis. The introduction system comprised of a PFA nebulizer (Elemental Scientific Inc.), fitted onto the Asperon spray chamber (PerkinElmer). The nebulizing and spray chamber makeup gas flow were 0.3 and 0.52 L min−1, respectively. A peristaltic pump was used to deliver samples at a flow rate of 19.4 μL min−1. The instrument was operated in Time-Resolved Analysis mode with a dwell time of 50 μs. Bulk analysis of cells was carried out using conventional ICP−MS, following cell digestion. Data acquisition during SC-ICP−MS analysis, as well as most data processing operations, were performed using the Syngistix Single-cell Application Module Software v1.2 (PerkinElmer). The software’s signal integration is based on an iterative algorithm in which cell events are discriminated from the dissolved element signal and integrated. An essential attribute of this software is the option to manually set the threshold signal, above which cell events are counted and integrated. Details about how this was set in the present study are provided in the following “SC ICP−MS parameter selection and limit of detection determination” section. In addition, the data for constructing histograms and investigating detected cell event profiles were treated with OriginPro and Microsoft Excel software. In order to quantitate cellular As, a mass flux calibration curve47,48 was established using dissolved standards of 1, 10, and 50 ng As per mL and applying a transport efficiency value, which was determined to be 9.9 ± 0.9% when using 60 nm Au NPs (NIST 8013). The obtained calibration parameters allowed for the use of the following equation for quantitating As in each cell detection event: mcell =

PA m

L D = b + 2.71 + 3.29 b

(2)

b standing for the average background intensity (counts). The above formula has been established by IUPAC for “wellknown” blanks, and has been derived from the normal approximation of the Poisson distribution.50,51 More specifically, it has been established based upon the hypothesis of a continuous signal distribution, that is, Gaussian, but with a nonconstant variance, being dependent upon the magnitude of the observed signal. The latter is the case for Poisson statistics, which evidently govern the observed mass spectrometer signals at the low count range.52 EASI−MS Analysis. C. reinhardtii cell samples were thawed and washed three times with d.i. water to remove media components, which caused ionization suppression, by centrifugation of the cells at 6000 rpm for 4 min, resuspension in 1 mL of d.i. water and vortexing. The resulting cell pellets were resuspended in 20 μL of d.i. water, and 1 μL of the cell suspension was placed on glass slide and left to dry under ambient conditions for approximately 25 min prior to EASI− MS analysis. The nebulizer used was constructed in-house. It consisted of 1/8 in. polyether ether ketone (PEEK) T-union and nuts. Whereas PEEK tubing was used to secure the fused silica capillaries (Polymicro Technologies, Phoenix, AZ). The outer fused silica capillary had an internal diameter (I.D.) of 540 μm and outer diameter (o.d.) of 690 μm, whereas the inner fused silica capillary had an i.d. of 50 μm and o.d. of 360 μm. The spraying solvent was a mixture of ACN/DMF 1:1 v/v delivered at a flow rate of 5 μL min−1 using a syringe pump (Cole-Parmer, IL, USA). The nebulizing gas was nitrogen (N2) supplied from a compressed gas cylinder (99.99% purity) at a backpressure of 8 bar. The nebulizer tip was positioned appropriately using a x-y-z stage; angled at 40° relative to the sample plate, and located approximately 2 mm above the sample surface and 6.5−7.5 mm from the inlet of the mass spectrometer. The sample glass slide was placed 1 mm below the inlet of the mass spectrometer. The mass spectrometer used was a quadrupole ion trap LCQ Advantage (Thermo Fisher Scientific, San Jose, CA) operated in both positive and negative ion mode. Mass spectra were recorded in the m/z 500−1000 range in the full scan mode.

(1) −1

where mcell is the mass per cell event (fg As cell ), PA is the integrated cell event intensity (counts) and m (counts fg−1) is the slope of the mass flux calibration curve (Supporting Information (SI) Figure S1). SC ICP−MS Parameter Selection and Limit of Detection Determination. The fundamental condition for conducting SC-ICP−MS analysis is that each detected cell event corresponds to a single cell, and therefore no cell overlapping takes place. Meeting this condition is directly dependent on the instrument dwell time and the number of cells introduced into the plasma per unit of time. A detector dwell time of 50 μs, with no settling time between consecutive data points, was selected for maximum instrument capability in terms of temporal resolution and signal-to-background (S/B) ratios. Cell suspensions were prepared at concentrations of 105 cells mL−1 in order to keep the occurrence of multiple cell events at low probability. Prediction of multiple cell events was based on Poisson statistics, as has been done for particle events upon nebulization of particle suspensions using a pneumatic nebulizer.49 If a cell event lasts no more than 1 ms, then given the sample uptake rate (19.4 μL min−1) and the cell number concentration (105 cells mL−1) there is a 3.1% and 0.05% probability to register single and double cell events, respectively. Considering that these probabilities further decrease when a transport efficiency (TE) below 100% is applied, it seems highly unlikely to detect multiple cell events. However, such statistics do not account for cell aggregation due to biological or cell culture growth conditions. C


Article


Figure 1. Panels (A) and (B): Time resolved ICP−MS signals obtained from washed C. reinhardtii cell suspensions, for cells grown in medium containing: (A) arsenate at 22.5 μg As mL−1, (B) no arsenate for control cells. Panel (C) shows the time profile of cell events chosen from Panel A with durations close to the average (●), narrower than the average (▲) and broader than the average (■), data points obtained every 50 μs, and panel (D) a histogram showing the distribution of cell event duration times for 123 detected cell events, with m.f: most frequent duration time; mu: mean duration time.

mean duration of approximately 250−350 μs as shown in Figure 1C for three representative cell events. The determination of all cell event duration times resulted in the frequency histogram shown in Figure 1D. So far, most SCICP−MS published studies have used relatively long detector dwell times (1−5 ms), and therefore the time profile of a cell event could not be determined. In the present study, we have used 50 μs detector dwell time, with no dead time, making it possible to determine the time profile of each cell event. Also, even though some studies have used 50 μs dwell times, such cell event duration profiles have not been studied in detail.13,17 Upon examination of the cell duration histogram it becomes clear that cell duration times are similar to those already observed in ICP−MS for 50 nm Au nanoparticles.54 This seems to indicate that the metal content of a C. reinhardtii cell forms a similarly sized nanoparticle upon drying in the plasma, or alternatively that signal duration for As in algal cells is governed by the element’s diffusion behavior in the plasma. The latter has been previously reported for other elements.15 Each cell event signal (signal spike) was integrated (Figure 1C), and the resulting integration values were transformed into

Ions were recorded with manually performed MSn experiments. The mass spectrometer settings used in the positive ion were: ion transfer capillary voltage, 20 V; tube lens offset, 20 V; whereas in the negative ion mode were capillary voltage, −34 V; tube lens offset, −20 V. Ion transfer capillary temperature was set at 300 °C for both ion polarity modes. Mass spectra were recorded using the Xcalibur 2.0.7 software (Thermo Fisher Scientific). Data were processed using Microsft Excel and OriginPro software. Nonpaired two sample Welch t test was conducted using R53 in order to determine if the lipid intensity ratio differences observed for control and arsenate incubated cells were statistically significant.

■

RESULTS AND DISCUSSION Arsenic Determination in C. reinhardtii by SC-ICP− MS. The detection of signal spikes for As ions (m/z 75), following the analysis of C. reinhardtii cell suspensions, revealed the presence of As in cells that had grown in arsenate-containing medium (Figure 1A), and its absence from cells cultivated with no arsenate added (Figure 1B). The detected signal spikes, corresponding to cell events, had a D


Article


function to the mass histograms in all cases (Figure 2A−C). This type of log-normal As mass uptake was also observed for unwashed algal cells (SI Figure S3), indicating that the As adsorbed to the cell walls follows the same type of distribution. Since previous studies have reported on the log-normal size distribution for C. reinhardtii cells,55 it is likely that As uptake per cell is influenced by cell size, as would also be expected for cell wall adsorption. Algal cells grown in 15, 22.5, and 30 μg As mL−1 as arsenate were determined by SC-ICP−MS to contain a log-normal mean As mass of 2.67, 4.07, and 2.80 fg respectively (Figure 2A−C), corresponding to 2.1−3.3 × 107 As atoms per cell. The observation that cells cultivated in the highest As (arsenate) concentration did not exhibit the highest As uptake was counterintuitive, however, the same trend was also observed when using conventional ICP−MS for bulk cell analysis. In this case, the mean As uptake values where determined to be 3.2, 7.2, and 4.8 fg As, respectively (Figure 3A); thus confirming that the highest As cultivation concentration did not correspond to the highest As uptake. This observation may be attributed to some extent to the slower cell growth rates observed for the cells grown in iAs(V) at 30 μg As mL−1 (Figure 4). After 5 days of cultivation under these conditions, the cells achieved approximately 50% growth,

As mass (fg) per detected cell event. The resulting mass distribution histograms for washed cells, showing As mass per individual cell (fg Ag per cell) per incubation concentration, that is, 15, 22.5, and 30 μg As per mL, are displayed in Figure 2. It was not possible to construct such histograms for control

Figure 2. As mass distribution histograms for C. reinhardtii cells grown in 15 (A), 22.5 (B) and 30 μg mL−1 As (C), binned at 0.8 fg. The log-normal probability function has been fitted to the mass histogram data in all three cases. Values for mu and m.f denote the log-normal mean and most frequently observed As mass per cell, respectively.

samples because no cell events were detected during their analysis (Figure 1B). Each of the cell event data series, displayed as a mass distribution histogram (Figure 2), were obtained from 9 min acquisitions resulting from the addition of three technical replicates (3 min per replicate, data shown in SI Table S1) carried out for each cell suspension. The obtained data clearly show the uptake of fg amounts of As by individual C. reinhardtii cells cultivated in the presence of 15, 22.5, and 30 μg As mL−1 as arsenate. The mass distribution of As among the analyzed cell population provides novel information regarding As uptake by the analyzed C. reinhardtii cells (Figure 2). High As mass dispersivity among the washed algal cells is clearly observed by the excellent fit of the log-normal probability

Figure 3. Column chart illustrating the mean As amount (fg/cell) obtained by using SC-ICP−MS (green) compared to conventional ICP−MS (orange) for washed (A) and unwashed (B) cells. Blue bars represent the dissolved As determined by using SC-ICP−MS and normalized per cell. Error bars of the mean determined by SC-ICP− MS represent the uncertainty of the log-normal mean. Error bars for conventional ICP−MS were calculated based on the 95% confidence interval of the dissolved As calibration curve. The error in the normalized dissolved As mass per cell is the standard deviation of the 3 independent replicates carried out for each cell suspension. E


Article


result in potential inaccuracies of the determined mass flux entering the ICP per time. Also, changes in the condition of cell suspension over time, that is, arsenate or other ions excreted from the cell due to osmotic imbalance, which will be detected by SC-ICP−MS as dissolved extracellular As, but not detected when using conventional ICP−MS analysis. This is because the conventional mode cannot discriminate between dissolved extracellular metal and the metal in the cell. Also, when conventional ICP−MS is used to determine the metal amount per cell, the analyzed cell number must be known with high accuracy. Supposing this number is underestimated, then the metal amount determined per cell will be biased to higher values. Nevertheless, the discrepancies observed in the present study between SC and conventional ICP−MS analysis are similar in magnitude to those that have been reported in existing SC-ICP−MS literature.10,19,21 Differences regarding As uptake and mass distribution behavior were also observed between washed and unwashed cells cultivated under the same conditions. Figure 5 shows the

Figure 4. Growth curves of C. reinhardtii cells grown in 15 (red), 22.5 (blue), 30 (green) μg As mL−1 and control (black).

compared to control cells and cells grown in the lower two As concentrations, i.e., 15 and 22.5 μg As mL−1. This clearly shows that the presence of arsenate at high levels causes substantial cellular stress, thus retarding cell growth rate. This is further supported by Wang et al., who showed that the median (50%) effect concentration (EC50) of arsenic for C. reinhardtii cells, cultivated in arsenate, was 33.5 μg As per mL.1 Similar growth inhibition has been observed in other studies for arsenate, in which case cell stress was linked to the overexpression of stress-related proteins.22 It is also of interest to note that a new analytical metric for metal determination in cells can now be determined using SCICP−MS, that is, the most frequently detected amount of As. This was determined to be 1.81, 1.73, and 1.53 fg As for iAs(V) exposure concentrations of 15, 22.5, and 30 μg As mL−1, respectively (Figure 2). It is interesting to note that the most frequently detected As mass was similar for all three As incubation concentrations. Therefore, it may represent a steady state or saturation mass that has been reached for the majority of the cell population at each growth condition. If this is the case, then once again the mean As uptake value, as determined from the bulk analysis of cells by conventional ICP−MS, seems to have little relevance with respect to how individual cells or cell populations actually behave once grown in the presence of As. The determined by conventional ICP−MS As mean values of approximately 3−8 fg seem to indicate that all cells uptake higher amounts of As than those determined using SC-ICP− MS. Also, the higher As uptake amounts determined for only a limited number of cells by SC-ICP−MS, as seen by their lognormal uptake behavior, may be providing insight into the occurrence of cell bloating, cell aggregation, or just reflecting normal algal cell size variations. Much broader As mass distribution was mainly observed for cells cultivated in 22.5 μg As mL−1; indicating the presence of a significant population of cells that had uptaken higher amounts of As. We envision that following further development of SC-ICP−MS these processes will be studied extensively, and will thus be better understood. For all the previously discussed biological reasons, it is evident that it is difficult to compare the mean values determined using SC-ICP−MS to those obtained by conventional ICP−MS (Figure 3). Furthermore, from an analytical point of view, additional reasons may be contributing to the differences observed between the two approaches. One such reason may be differences in the determined transport efficiency using Au nanoparticles and the actual transport efficiency achieved for C. reinhardtii cells. Such differences

Figure 5. Arsenic mass distribution in washed (blue line) and unwashed (green line) C. reinhardtii cells incubated in 22.5 μg As mL−1.

log-normal fit for the mass distribution for both types of cells grown under the same As concentrations (22.5 μg As mL−1). From this analysis we observe a shift of the As amount present in the unwashed cells to lower mean amounts for washed cells, that is, the mean drops from 6.1 to 4.1 fg As per cell; clearly indicating that a significant amount of As is wall bound. Effect of Arsenate on Membrane Lipids. With respect to lipid profiling, it is known that glycerolipids are the main constituents of photosynthetic membranes in green algae that have been conserved throughout the course of evolution. The chloroplasts of these cells contain low amounts of phosphatidylglycerol (PG) and phosphatidylinositol (PI) and higher amounts of monogalactosyldiacylglycerol (MGDG), digalactosyldiacylglycerol (DGDG) and sulfoquinovosyldiacylglycerol (SQDG) lipids. MGDG and DGDG lipids are neutral galactoglycerolipids and being the most abundant in photosynthetic membranes, they possess vital roles in the stabilization of the membrane lipid bilayer.56 The integrity of the membrane bilayer is based on a sufficient proportion of neutral galactoglycerolipids and a smaller proportion of charged polar lipids such as SQDG and PG.56,57 Apart from stabilizing the membrane lipid bilayer, these lipids also possess vital roles in biological activities of the cells such as secretion, transportation, signal transduction, photosynthetic light F


Article


Figure 6. Univariate box-whisker plots of DGTS (A), PG (B), and SQDG (C, D) lipid species intensity ratio for control and arsenate-grown cells. Arithmetic mean values are represented as small box with ±1 SD (standard deviation) as larger box and ±1.96 SD as whiskers. Horizontal lines inside the larger boxes represent the median values, whereas the X symbols represent the minimum and maximum data points. Asterisks indicate statistically significant values compared to the control (one asterisk, p < 0.01; two asterisks, p < 0.001; nonpaired two sample Welch t test).

harvesting and electron transfer.45,56 In Chlamydomonas reinhardtii a unique class of lipids is also present, that of 1,2diacylglyceryl −3-O-4′- (N,N,N-trimethyl) -homoserine (DGTS), which are constituents of extraplastidial membranes and substitute for phosphatidylcholine (PC) that is normally present in these membranes.35,58−60 Using desorption EASI−MS, it was possible to rapidly obtain lipid profiles from intact C. reinhardtii cells without the need for prior cell lysis and extractions. Mass spectra obtained from the EASI−MS analysis of cells in both positive and negative ion mode (SI Figure S4) provided information on the membrane lipid composition of the examined algal cells. More specifically, detected lipids included DGTS, DGDG, PG, SQDG, and PI. Identification of lipids was based on literature results35,45,57,59 and manual interpretation of acquired MS/MS spectra. Also, NIST MS Search 2.0 with mass spectra libraries61 was used to compare experimental MS/MS spectra. All identified lipids are summarized in SI Table S1. Only relative quantitation of membrane lipids was possible using desorption EASI−MS. This allowed us to monitor the relative lipid changes without the need to know the number of cells that were sampled during each spraying event. It should be noted that sample spotting and drying does not guarantee cell number homogeneity, because during the drying process the cells are distributed heterogeneously over the spot as they tend to concentrate in its outer ring (“coffee ring” effect). In Figure 6, changes in intensity ratios of certain membrane lipids are shown. In the DGTS (C18/C18) series of lipids, it was evident that with increasing concentration of iAs(V) the

intensity ratio of DGTS (36:3)/DGTS(36:6) also increased, i.e., the intensity of the less unsaturated DGTS (36:3) increased. DGTS lipids are part of extraplastidial membranes, and in C. reinhardtii cells substitute PC which is usually found in membranes of plant cells.35,58−60 This trend was also apparent for other lipid classes such as PG (C18/C16), SQDG (C18/C16), and SQDG (C20/C16) (Figure 6) which are part of the photosynthetic membrane lipid bilayer.56,57 These data suggest possible reorganization of membranes induced by the presence of arsenate in the growth medium. As the media concentration of iAs(V) increased, for the aforementioned lipid classes, conversion from more to less unsaturated fatty acid chains in the membrane lipids seems to have occurred. As a result of this, the membranes comprised of these lipids must have become more rigid due to the increased proportion of less unsaturated fatty acid containing lipids. It has been reported that for C. reinhardtii cells the parent lipid molecules initially contain saturated fatty acids such as 16:0/16:0, monounsaturated (16:0/18:1 and 18:0/18:1) and diunsaturated (18:1/18:1).59 Subsequently, C16 and C18 fatty acids are desaturated by enzymes called desaturases that convert a single C−C bond into a double bond, resulting in fatty acids with a higher degree of unsaturation.59,60 Additionally, remodeling of membrane lipids was also observed in ironstarved C. reinhardtii cells, where higher degree of saturation was observed for membrane glycerolipids.62 The authors concluded that this was the result of iron withdrawal from the cultivation medium as fatty acid desaturases are diiron enzymes. A possible explanation for our observed results G



■

could be that as iAs(V) enters the cells through phosphate transporters, it can be reduced to As(III)2 which can interact with thiol compounds.5 Thus, we speculate that As(III) may have a direct effect on enzymes that are involved in the fatty acid desaturation pathway. Cell Integrity Before and After EASI−MS. The integrity of the cells that were washed with d.i. water and spotted onto the glass slide along with the integrity of those obtained after spraying with ACN/DMF 1:1 v/v solution was investigated. Cells washed with H2O were observed under a microscope to remain intact, without any signs of cell bursting or swelling. Also, cells that had been dried on a glass slide were carefully resuspended in a droplet of water for subsequent electron microscopy and fluorescence analysis. The obtained images shown in SI Figure S5 revealed that the cells were intact. Finally, the desorption process was mimicked but the aerosol produced after spraying the cells was collected on a glass slide, resuspended in a small volume of water and observed once again under an electron microscope. Microscopy and fluorescence images indicate that the cells had lysed after desorbing them with the spraying solvent and collecting them (SI Figure S5).

■

Article

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.9b00917. Additional information on the step-by-step transformation of the intensity registered for a cell event into As mass, 75As+ intensity distribution for control C. reinhardtii cells and their As mass distribution in unwashed cells, EASI mass spectra of C. reinhardtii cells in positive and negative ion mode and microscopy images before and after EASI−MS analysis, identified lipids molecules in C. reinhardtii cells (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Spiros A. Pergantis: 0000-0002-9077-7870 Author Contributions §

These authors contributed equally. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

CONCLUSIONS

Notes

The authors declare no competing financial interest.

■

Current state-of-the-art analytical methodologies for studying metal uptake by cells involve analyzing bulk cell samples after they have been digested and analyzed using atomic spectrometric techniques. Whereas for complementary molecular analysis, bulk cell samples must be extracted and the extracts purified prior to their mass spectrometric analysis. However, apart from the requirement for the growth of large amounts of cells and the tedious sample preparation procedures (digestion, extraction, and cleanup) these methodologies have several other limitations as well. In order to overcome them we have applied two mass spectrometric techniques, currently still being developed, that allow for cell analysis in a state as close as possible to their native state. These two techniques are SC-ICP−MS for As determination in single cells, and ambient MS in the form of EASI−MS for lipidomic analysis of cell spots directly from a solid surface. Both techniques eliminate the need for most of the aforementioned sample preparation steps, except for cell rinsing, and only require small numbers of cells. In addition, these two techniques provide additional analytical metrics. In the case of SC-ICP−MS, these include metal mass per analyzed cell, mass distribution in cell populations, most frequent As mass amount in the analyzed cells, simultaneous determination of cell-containing As and extracellular As in a single analysis step. Whereas for EASI−MS both negative and positive mass spectra can be obtained without any need to reoptimize source conditions since no high voltages are applied to the nebulizer or the source entrance. Thus, only the ion optics and analyzer voltage polarities need to be switched in order to monitor both negative and positive ions. This type of switching has no effect on the spray or ion formation processes. Finally, both techniques can potentially provide useful analytical data only minutes after the cell samples have been thawed, diluted, and analyzed.

ACKNOWLEDGMENTS We thank PerkinElmer for the generous sponsoring of the single cell sample introduction system and the single cell data processing software. Also, we acknowledge the support provided by Elemental Scientific Incorporated for the construction of the EASI−MS source. Finally, we acknowledge the assistance of Dr. Th. Nazos and the Electron Microscopy Laboratory “Vasilis Galanopoulos”, University of Crete, for the optical microscopy images.

■

REFERENCES

(1) Wang, N. X.; Li, Y.; Deng, X. H.; Miao, A. J.; Ji, R.; Yang, L. Y. Water Res. 2013, 47 (7), 2497−2506. (2) Yin, X.; Wang, L.; Duan, G.; Sun, G. J. Environ. Sci. 2011, 23 (7), 1186−1193. (3) Kaise, T.; Fujiwara, S.; Tsuzuki, M.; Sakurai, T.; Saitoh, T.; Mastubara, C. Appl. Organomet. Chem. 1999, 13 (2), 107−111. (4) Murota, C.; Matsumoto, H.; Fujiwara, S.; Hiruta, Y.; Miyashita, S.; Shimoya, M.; Kobayashi, I.; Hudock, M. O.; Togasaki, R. K.; Sato, N.; Suzuki, M. Planta 2012, 236, 1395−1403. (5) Wang, N. X.; Huang, B.; Xu, S.; Wei, Z. B.; Miao, A. J.; Ji, R.; Yang, L. Y. Aquat. Toxicol. 2014, 157, 167−174. (6) Kobayashi, I.; Fujiwara, S.; Saegusa, H.; Inouhe, M.; Matsumoto, H.; Tsuzuki, M. Mar. Biotechnol. 2006, 8 (1), 94−101. (7) Miyashita, S.; Fujiwara, S.; Tsuzuki, M.; Kaise, T. Biosci., Biotechnol., Biochem. 2011, 75 (3), 522−530. (8) Li, F.; Armstrong, D. W.; Houk, R. S. Anal. Chem. 2005, 77 (5), 1407−1413. (9) Ho, K.; Chan, W. J. Anal. At. Spectrom. 2010, 25, 1114−1122. (10) Wei, X.; Hu, L. L.; Chen, M. L.; Yang, T.; Wang, J. H. Anal. Chem. 2016, 88 (24), 12437−12444. (11) Corte Rodríguez, M.; Á lvarez-Fernández García, R.; Blanco, E.; Bettmer, J.; Montes-Bayón, M. Anal. Chem. 2017, 89, 11491−11497. (12) Meyer, S.; López-Serrano, A.; Mitze, H.; Jakubowski, N.; Schwerdtle, T. Metallomics 2018, 10 (1), 73−76. (13) Merrifield, R. C.; Stephan, C.; Lead, J. R. Environ. Sci. Technol. 2018, 52 (4), 2271−2277. H


Article

Analytical Chemistry (14) Tsang, C. N.; Ho, K. S.; Sun, H.; Chan, W. T. J. Am. Chem. Soc. 2011, 133 (19), 7355−7357. (15) Shigeta, K.; Koellensperger, G.; Rampler, E.; Traub, H.; Rottmann, L.; Panne, U.; Okino, A.; Jakubowski, N. J. J. Anal. At. Spectrom. 2013, 28 (5), 637−645. (16) Zheng, L. N.; Wang, M.; Wang, B.; Chen, H. Q.; Hong, O.; Zhao, Y. L.; Chai, Z. F.; Feng, W. Y. Talanta 2013, 116, 782−787. (17) Miyashita, S. I.; Groombridge, A. S.; Fujii, S. I.; Minoda, A.; Takatsu, A.; Hioki, A.; Chiba, K.; Inagaki, K. J. Anal. At. Spectrom. 2014, 29 (9), 1598−1606. (18) Verboket, P. E.; Borovinskaya, O.; Meyer, N.; Dittrich, P. S. Anal. Chem. 2014, 86 (12), 6012−6018. (19) Wang, H.; Wang, B.; Wang, M.; Zheng, L.; Chen, H.; Chai, Z.; Zhao, Y.; Feng, W. Analyst 2015, 140 (2), 523−531. (20) Zheng, L.-N.; Wang, M.; Zhao, L.-C.; Sun, B.-Y.; Wang, B.; Chen, H.-Q.; Zhao, Y.-L.; Chai, Z.-F.; Feng, W.-Y. Anal. Bioanal. Chem. 2015, 407 (9), 2383−2391. (21) Wang, H.; Wang, M.; Wang, B.; Zheng, L.; Chen, H.; Chai, Z.; Feng, W. Anal. Bioanal. Chem. 2017, 409 (5), 1415−1423. (22) Walliwalagedara, C.; Keulen, H. Van; Willard, B.; Wei, R. Am. J. Plant Sci. 2012, 3, 764−772. (23) Watrous, J.; Roach, P.; Heath, B.; Alexandrov, T.; Laskin, J.; Dorrestein, P. C. Anal. Chem. 2013, 85 (21), 10385−10391. (24) Watrous, J.; Roach, P.; Alexandrov, T.; Heath, B. S.; Yang, J. Y.; Kersten, R. D.; van der Voort, M.; Pogliano, K.; Gross, H.; Raaijmakers, J. M.; Moore, B. S.; Laskin, J.; Bandeira, N.; Dorrestein, P. C. Proc. Natl. Acad. Sci. U. S. A. 2012, 109 (26), E1743−E1752. (25) Duncan, K. D.; Fyrestam, J.; Lanekoff, I. Analyst 2019, 144, 782−793. (26) Hsu, C. C.; Dorrestein, P. C. Curr. Opin. Biotechnol. 2015, 31, 24−34. (27) Ghosh, D.; Bhadury, P.; Routh, J. MicrobiologyOpen 2018, 7 (5), 1−12. (28) Pepi, M.; Heipieper, H. J.; Fischer, J.; Ruta, M.; Volterrani, M.; Focardi, S. E. Extremophiles 2008, 12 (3), 343−349. (29) Zhang, T. L.; Gao, Y. X.; Lu, J. F.; Wang, K. J. Inorg. Biochem. 2000, 79 (1−4), 195−203. (30) Quoc, L.; Thi, T.; Huong, T.; Thi, P.; Hong, A.; Kawakami, T. Toxicol. In Vitro 2008, 22, 1632−1638. (31) Marques, A. S.; Bedia, C.; Lima, K. M. G.; Tauler, R. Anal. Bioanal. Chem. 2016, 408 (21), 5829−5841. (32) Shan, Y.; Liu, Y.; Yang, L.; Nie, H.; Shen, S.; Dong, C.; Bai, Y.; Sun, Q.; Zhao, J.; Liu, H. J. Sep. Sci. 2016, 39 (19), 3745−3753. (33) Plohnke, N.; Seidel, T.; Kahmann, U.; Rögner, M.; Schneider, D.; Rexroth, S. Mol. Cell. Proteomics 2015, 14 (3), 572−584. (34) Tomčala, A.; Kyselová, V.; Schneedorferová, I.; Opekarová, I.; Moos, M.; Urajová, P.; Kručinská, J.; Oborník, M. J. Sep. Sci. 2017, 40 (17), 3402−3413. (35) Yang, D.; Song, D.; Kind, T.; Ma, Y.; Hoefkens, J.; Fiehn, O. PLoS One 2015, 10 (9), 1−16. (36) Biller, P.; Ross, A. B. Algal Res. 2014, 6 (PA), 91−97. (37) Wang, J.; Zhang, X.; Shi, M.; Gao, L.; Niu, X.; Te, R.; Chen, L.; Zhang, W. Funct. Integr. Genomics 2014, 14 (2), 431−440. (38) Furuhashi, T.; Nakamura, T.; Fragner, L.; Roustan, V.; Schön, V.; Weckwerth, W. Biotechnol. J. 2016, 11 (10), 1262−1267. (39) Krismer, J.; Steinhoff, R.; Zenobi, R. Chim. Int. J. Chem. 2013, 33 (4), 589−609. (40) Cichon, A. B. T.; Suder, P. Cytotechnology 2014. (41) Song, Y.; Talaty, N.; Tao, W. A.; Pan, Z.; Cooks, R. G. Chem. Commun. 2007, 1, 61−63. (42) Zhang, J. I.; Talaty, N.; Costa, A. B.; Xia, Y.; Tao, W. A.; Bell, R.; Callahan, J. H.; Cooks, R. G. Int. J. Mass Spectrom. 2011, 301 (1− 3), 37−44. (43) Lanekoff, I.; Geydebrekht, O.; Pinchuk, G. E.; Konopka, A. E.; Laskin, J. Analyst 2013, 138 (7), 1971−1978. (44) Bergman, H. M.; Lanekoff, I. Analyst 2017, 142 (19), 3639− 3647.

(45) Liu, Y.; Zhang, J.; Nie, H.; Dong, C.; Li, Z.; Zheng, Z.; Bai, Y.; Liu, H.; Zhao, J. Anal. Chem. 2014, 86 (14), 7096−7102. (46) Hanikenne, M. New Phytol. 2003, 159 (2), 331−340. (47) Pace, H. E.; Rogers, N. J.; Jarolimek, C.; Coleman, V. A.; Higgins, C. P.; Ranville, J. F. Anal. Chem. 2011, 83, 9361−9369. (48) Pace, H. E.; Rogers, N. J.; Jarolimek, C.; Coleman, V. A.; Higgins, C. P.; Ranville, J. F. Anal. Chem. 2012, 84, 4633−4633a. (49) Olesik, J. W.; Gray, P. J. J. Anal. At. Spectrom. 2012, 27 (7), 1143. (50) Currie, L. A. Pure Appl. Chem. 1995, 67 (10), 1699−1723. (51) Currie, L. A. Anal. Chem. 1968, 40 (3), 586−593. (52) Tanner, M.; Günther, D. Anal. Chim. Acta 2009, 633 (1), 19− 28. (53) R Core Team. R: A Language and Environment for Statistical Computing; Vienna, Austria, 2018. (54) Gschwind, S.; Flamigni, L.; Koch, J.; Borovinskaya, O.; Groh, S.; Niemax, K.; Günther, D. J. Anal. At. Spectrom. 2011, 26, 1166− 1174. (55) Garz, A.; Sandmann, M.; Rading, M.; Ramm, S.; Menzel, R.; Steup, M. Biophys. J. 2012, 103 (5), 1078−1086. (56) Boudière, L.; Michaud, M.; Petroutsos, D.; Rébeillé, F.; Falconet, D.; Bastien, O.; Roy, S.; Finazzi, G.; Rolland, N.; Jouhet, J.; Block, M. A.; Maréchal, E. Biochim. Biophys. Acta, Bioenerg. 2014, 1837 (4), 470−480. (57) Vieler, A.; Wilhelm, C.; Goss, R.; Süß, R.; Schiller, J. Chem. Phys. Lipids 2007, 150 (2), 143−155. (58) Moore, T. S.; Du, Z.; Chen, Z. Plant Physiol. 2001, 125 (1), 423−429. (59) Giroud, C.; Gerber, A.; Eichenberger, W. Plant Cell Physiol 1988, 29 (4), 587−595. (60) Li-Beisson, Y.; Beisson, F.; Riekhof, W. Plant J. 2015, 82 (3), 504−522. (61) Kind, T.; Liu, K. H.; Lee, D. Y.; Defelice, B.; Meissen, J. K.; Fiehn, O. Nat. Methods 2013, 10 (8), 755−758. (62) Urzica, E. I.; Vieler, A.; Hong-Hermesdorf, A.; Page, M. D.; Casero, D.; Gallaher, S. D.; Kropat, J.; Pellegrini, M.; Benning, C.; Merchant, S. S. J. Biol. Chem. 2013, 288, 30246−30258.

I


Investigating the Uptake of Arsenate by Chlamydomonas reinhardtii

Recommend Documents