Development of a Paper Spray Mass Spectrometry ... - ACS Publications

Subscriber access provided by NEW YORK UNIV

Article

Development of a paper spray mass spectrometry cartridge with integrated solid phase extraction for bioanalysis Chengsen Zhang, and Nicholas Edward Manicke Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b00884 • Publication Date (Web): 22 May 2015 Downloaded from http://pubs.acs.org on May 23, 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 22

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

Development of a paper spray mass spectrometry cartridge with integrated solid phase extraction for bioanalysis Chengsen Zhang, Nicholas E. Manicke* Department of Chemistry and Chemical Biology, Indiana University-Purdue University Indianapolis, Indianapolis, Indiana 46202, United States

A novel paper spray cartridge with an integrated solid phase extraction (SPE) column is described. The cartridge performs extraction and preconcentration, as well as sample ionization by paper spray, from complex samples such as plasma. The cartridge allows for selective enrichment of target molecules from larger sample volumes and removal of the matrix, which significantly improved the signal intensity of target compounds in plasma samples by paper spray ionization. Detection limits, quantitative performance, recovery, ionization suppression and the effects of sample volume were evaluated for five drugs: carbamazepine, atenolol, sulfamethazine, diazepam, and alprazolam. Compared with direct paper spray analysis of dried plasma spots, paper spray analysis using the integrated solid phase extraction improved the detection limits significantly by a factor of 14 to 70 depending on the drug. The improvement in detection limits was due in large part to the capability of analyzing larger sample volumes. In addition, ionization suppression was found to be lower and recovery was higher for paper spray with integrated SPE as compared to direct paper spray analysis. By spiking an isotopically labeled internal standard into the plasma sample, a linear calibration curve for the drugs was obtained from the LOD to 1 µg/mL, indicating this method can be used for quantitative analysis. The paper spray cartridge with integrated SPE could prove valuable for analytes that ionize poorly, in applications where lower detection limits are required, or on portable mass spectrometers. The improved performance comes at the cost of requiring a more complex paper spray cartridge and requiring larger sample volumes than those used in typical direct paper spray ionization.

1 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 22

Paper spray mass spectrometry, first described in 2010, is a method for performing rapid, direct analysis of samples spotted on paper or another porous substrate.1,2 Much of the early work on paper spray MS has focused on the targeted quantitative analysis of drugs and drug metabolites directly from dried biofluids2-5, which is important in pharmaceutical research, clinical chemistry, and forensic toxicology. A number of other applications have been reported for paper spray MS as well. These include profiling of lipids in bacteria6 and microalgae7, online chemical monitoring of cell culture8, detection of chemical contaminants in food, including plasticizers, melamine, pharmaceuticals, and 4-methylimidazole9,10, analysis of acyl-carnitines from blood and urine11,12, as an ion source for a microfluidic chip13. Other recent developments include coupling paper spray to differential mobility spectrometry to improve selectivity14 and a 3D printed paper spray cartridge15. Sample analysis by paper spray MS is typically performed by depositing a liquid sample onto a paper substrate that has been cut to a sharp point. A solvent is added to the paper where it wicks through the paper and the sample by capillary action, extracting soluble components from the sample in the process. Analyte ionization is achieved in most cases by inducing an electrospray at the tip of the wet paper by applying a potential of several thousand volts. Paper spray has the potential to simplify and expand the utility of mass spectrometric assays. However, the detection limits of paper spray analyses from complex samples such as plasma are sometimes inadequate. While low or sub-ng/mL detection limits from dried blood or plasma spots can be achieved in the most favorable cases (i.e. hydrophobic and basic analytes analyzed on sensitive triple quadrupole mass spectrometers), detection limits are significantly higher for analytes that ionize poorly or cannot be efficiently extracted from the sample matrix.16 Additionally, detection limits are significantly higher on portable or miniature mass spectrometers due to size constraints limiting the MS performance.17,18 Simply increasing the sample amount in paper spray does not significantly increase signal intensity or improve detection limits. Improving detection limits requires matrix removal and/or concentration of the analyte. Sample preparation methods that accomplish this while maintaining the simplicity of the approach would be beneficial. There are a number of examples in the literature that describe combining simplified sample preparation techniques to direct analysis methods19-22 and using microfluidics platforms to perform automated sample preparation prior to MS analysis23,24. In this study, we describe a paper spray cartridge with an integrated solid phase extraction (SPE) column for immediate extraction and concentration of analytes from complex samples such as plasma. The cartridge also included the necessary components to perform sample ionization by paper spray. Sample extraction, preconcentration, and ionization from complex samples are all performed on a cartridge that could, in principal, be disposable due to its simple design and low-cost materials. Analysis required the same number of steps as typical paper spray (application of sample, sample drying, and application of extraction/spray solvent). Sample preconcentration and extraction occurred automatically, requiring no human action or secondary device. No pumping was required; both sample and extraction solvent were fed through the device passively by capillary action. Because the sample is stored as a dried sample, this paper spray cartridge may share some of the benefits of traditional dried matrix sampling such as being able 2 ACS Paragon Plus Environment

Page 3 of 22

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


to store and ship samples at room temperature. After sample application and drying, one step was needed to analyze the sample by MS: a solvent was added to the cartridge which wicked through the SPE material and the paper spray substrate by capillary action, recovering the analyte and generating gas-phase ions for MS analysis in a single step. We have created working prototype cartridges and tested them using animal blood plasma. Detection limits, recovery and ionization suppression, quantitative performance, and the effect of sample volume for a set of test drugs are reported.

Experimental Methods: Chemicals and Materials. Methanol (HPLC grade) and acetic acid (HPLC grade) were purchased from Fisher Scientific (Hampton, New Hampshire). All drugs and cellulose powder (50 µm) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Sulfamethazine-d4 was purchased from CDN Isotopes (Pointe-Claire, Quebec, Canada), and all other stable isotopically labeled internal standards (in solution) were purchased from Cerilliant (Reston, VA, USA) and stored at −20 °C. Bovine plasma with K2-EDTA anticoagulant was purchased from Lampire Biological Laboratories (Pipersville, PA, USA) and stored at −20 °C. Whatman grade 31ET-Chr paper and gel-blotting-paper were purchased from Whatman (Piscataway, NJ, USA). Supel Select HLB (hydrophilic lipophilic balance) SPE was purchased from Sigma-Aldrich (St. Louis, MO, USA). Sample Preparation. Bovine plasma was brought to room temperature prior to use. Drugs and internal standards were first diluted with methanol and then spiked into the plasma, keeping the organic fraction at one percent or less. Experiments were performed on the same day the plasma samples were prepared. Direct Paper Spray Ionization. The plasma sample was spotted onto a piece of triangular shaped 31-ET chromatography paper and stored as a dried spot. Typical sample volume was 3 µL, and the size of the paper triangle was about 5 mm x 8 mm (base x height). Analysis was performed by depositing 20 µL of 95:5 (v:v) methanol-water with 0.01% acetic acid to the rear of the paper so that the solvent wicked through the paper and sample by capillarity. The paper, which was cut to a sharp point, was positioned 5 millimeters away from the atmospheric pressure inlet of the mass spectrometer and a high voltage of 4.5 kV was then applied to the paper, inducing an electrospray at the tip of the paper. The solvent evaporates from the charged droplets generated by the electrospray process, leaving gas phase ions of the analyte molecules which can then be detected by a mass spectrometer.1 Paper Spray Ionization using Integrated SPE. The cartridges were made from Delrin® plastic (McMaster Carr, Elmhurst, IL) on a milling machine (Sherline, Vista, CA). As shown in figure 1, the cartridge consisted of two parts, a bottom part (LWH: 40mm x 26mm x 6mm) and a top part (LWH: 14mm x 22mm x 13mm) joined together using a tongue and groove. The bottom part had two separate recessed regions: one region to hold a piece of Whatman gel blotting paper (20 mm x 16mm x 0.7mm) which acted as an absorbent waste pad and another to hold the paper spray substrate (Whatman 31ET, 8mm x 5mm x 0.5mm). A slot measuring 5mm x 0.5mm was cut into the front side of the cartridge to allow the sharp tip of the spray substrate to protrude. The top part of the cartridge had a through hole with 3mm diameter to 3 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 4 of 22

contain the integrated SPE column and sample. Unless noted otherwise, the SPE material was a 1:1 mixture of Select HLB SPE and cellulose powder . The sample was applied to the top of the SPE column on the cartridge, and the sample was drawn through the SPE column by the capillary action of the waste pad. Samples were allowed to dry fully prior to analysis. Drug recovery and ionization was performed using 60 µL of 95:5 methanol:water with 0.01% acetic acid. The spray voltage was 4.5 kV. Additional details about the cartridge are found in the Results and Discussion section. Mass Spectrometry, Data Collection, and Data Processing. MS analysis was performed using on Thermo Fisher LTQ-XL mass spectrometer (Thermo Scientific Inc., San Jose, CA). The temperature of the MS capillary inlet was set at 275 °C. The tube lens voltage was set at 55 V. The voltage used for paper spray ionization was 4.5 kV in positive ion mode from instrument’s built-in power supply. Tandem mass spectra were recorded using collision-induced dissociation (CID). The most abundant fragment of each compound was selected for quantification. The precursor ions and fragment ions for the five test compounds were as follows: alprazolam (m/z 309 → 281), atenolol (m/z 267 → 225), carbamazepine (m/z 237 → 194), diazepam (m/z 285 → 257), and sulfamethazine (m/z 279 → 186). Quantitative analysis was performed by analyzing all five drugs simultaneously from one plasma sample by collecting sequential MS/MS spectra for each drug and internal standard. Internal standards in methanol were first diluted with water and then spiked into the plasma sample by 50-fold dilution (10 µL into 490 µL) prior to adding the sample to the SPE column. The final concentration of internal standard was 100 ng/mL for the SPE paper spray analyses and 500 ng/mL for direct paper spray. Quantitation was performed using the ratio of the average absolute signal intensity of analyte to that of the internal standard. The standard line was calculated using 1/x2 weighted least squares. Limits of detection were calculated as 3 times the standard deviation of the blank divided by the slope of the trend line.25 Recovery and Ionization Suppression. Analyte recovery and ionization suppression relative were determined for paper spray analysis of plasma with and without integrated SPE. Both recovery and ionization suppression were measured relative to blank paper at two different sample volumes, 3 µL and 100 µL. Paper spray with integrated SPE was performed exactly as described in the “Paper Spray Ionization using Integrated SPE” subsection above, except that a 3mm diameter disc of Whatman 31ET (lower disc) was placed below the SPE column after sample application and drying. The function of this disc is described below. Direct paper spray was performed using a device depicted in supplementary figure 1B. The Delrin plastic holder was machined with a slot milled into the side to hold the spray substrate and a 3mm diameter hole perpendicular to the top surface of the paper. One or more 3 mm diameter 31ET paper discs containing dried plasma were put into the 3 mm diameter hole prior to analysis. The paper discs were spotted with sample separately outside the cartridge and inserted after they were fully dried. Otherwise, at large volumes, the layer of dried plasma that formed was impenetrable to the spray solvent. The volume of spray solvent was 30 µL for 1 paper disc (corresponding to 3 µL plasma sample) and 90 µL for 32 paper discs (corresponding to 100 µL plasma sample). Ionization was induced directly from the paper substrate by applying a high voltage (4.5 kV typically) to the paper through a wire inserted in 4 ACS Paragon Plus Environment

Page 5 of 22

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


the device. All other experiment materials, parameters and conditions were as described in theo “Direct Paper Spray Ionization” subsection. To determine ionization suppression and recovery, three groups were prepared, each having a lower disc placed below an upper portion consisting of either paper discs containing dried plasma (direct paper spray) or the SPE column through which the sample had been extracted. The three groups were as follows. Group A: lower disc contained 0.6 ng of each analyte and 2.4 ng of each IS (neat) for the 3 µL sample size. For the 100 µL sample size, the amount of analyte was increased to 20 ng. For direct paper spray, the upper portion consisted of 1 blank paper disc for the 3 µL sample size or 32 blank paper discs for 100 µL sample. A blank SPE column was used for paper spray with SPE. Group B: lower disc was the same as group A. The upper portion consisted of 1 paper disc containing 3 µL of blank plasma or 32 paper discs containing 100 µL of blank plasma for the direct paper spray experiments. For paper spray with SPE, the upper portion consisted of the SPE column through which either 3 or 100 µL of blank plasma had been extracted. Group C: lower disc contained 2.4 ng of each IS. For direct paper spray, the upper portion consisted of either 1 disc containing 3 µL of plasma with 0.6 ng of each analyte or 32 discs containing 100 µL of plasma with 20 ng of each analyte. For paper spray with integrated SPE, the upper portion consisted of the SPE column through which either 3 or 100 µL of plasma with 0.2 ng/µL of each analyte had been extracted. Ionization suppression was defined as the percent change in analyte signal intensity between groups A and B. Recovery was defined as the ratio of the analyte to the internal standard determined for group C divided by group B and expressed as a percent.

RESULTS AND DISCUSSION Cartridge Design We designed and fabricated prototype paper spray cartridges with an integrated solid phase extraction column. The cartridge consists of two parts, as shown in figure 1(a). The bottom part had two separate recessed regions to hold an absorbent waste pad and the paper spray substrate. A wire inserted through the bottom of the cartridge provided electrical contact with the spray substrate. The top part of the cartridge had a hole bored through it to contain the SPE column. The SPE column used in these studies consisted of (from top to bottom) one layer of 31ET paper to help keep the SPE powder in place, a layer of 5 mg SPE material, and one layer of Whatman grade 2727 chromatography paper again to keep the SPE material in place. The top and bottom parts were assembled together using a tongue and groove system which allowed the top part to be held in close contact with bottom part while also allowing top part to be moved from the sample loading position to the sample recovery/ionization position. The cost of materials was less than 20 cents (US) per cartridge. Most of this cost was the Delrin plastic, which was selected because of its good machinability and solvent compatibility. The procedure for performing paper spray analysis using the integrated SPE column is depicted in figure 1(b) though 1(d). The plasma samples were added to the hole in the top part of the cartridge containing the SPE column. The sample volume was between 10 µL up to hundreds of microliters. At sample volumes less than 10 microliters, no extraction or concentration of the analyte occurred. The sample wicked through the SPE column and subsequently onto the absorbent pad contained within the bottom part of the cartridge. As the 5 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 6 of 22

sample passed through the SPE column, the target analytes were retained on the SPE column while the excess matrix was absorbed onto the waste pad. The sample was then allowed to dry, during which a cover was placed over the top part of the cartridge to prevent sample evaporation from the top of the SPE column, which would cause “backflow” of analyte depleted sample back up the SPE column (figure 1c). The target compounds were recovered from the SPE column and analyzed by sliding the top part of the cartridge over so that the SPE column was positioned over the pentagonal shaped paper spray substrate rather than the waste pad. The cartridge was positioned in front of the inlet to the mass spectrometer, and the extraction/spray solvent was added to the top of the SPE column. The solvent wicked through the SPE column, recovering the analytes in the process, and onto the spray substrate passively by capillary action. Ionization was induced directly from the paper substrate by applying a high voltage (4.5 kV typically) to the paper through a wire inserted in bottom part. Figure 2 shows photographs of a prototype paper spray cartridge with integrated SPE. For the SPE material to be usable in the format described here, it must have two general characteristics. First, the SPE material must be water-wettable. Typical reverse-phase SPE materials are not water-wettable, so pressure is normally used to force aqueous samples through the extraction material. A small amount of organic solvent is also sometimes added to aqueous samples as well. Neither of these two solutions is feasible in a simple, automatic on-cartridge extraction. Second, typical SPE materials must be conditioned with solvents and water prior to sample application and cannot be allowed to dry out before the sample is applied. Several manufacturers make polymeric SPE materials that are water wettable yet have reverse phase type retaining character. These materials are also not affected by drying out prior to sample application like traditional SPE materials. In this study, we used Supel Select Polymeric SPE with HLB (hydrophilic lipophilic balance) chemistry. The HLB SPE material has reverse phase retention character but is more water wettable than typical reverse phase SPE materials and, according to the manufacturer, can also be used while dry. We found, however, that this SPE material was still not sufficiently water wettable. To increase the wicking rate of aqueous samples, we combined the SPE powder with an equal mass of cellulose powder and mixed by vortexing. The SPE material was otherwise unmodified. All of the experiments reported here use this 1:1 mixture of SPE:cellulose, and the mass of SPE material cited include the mass of the SPE powder and the cellulose powder. Signal intensity and limits of detection The analyte chemical/physical properties and the sample matrix both significantly affect the limits of detection in paper spray MS. In one comparative analysis of numerous small molecules (MW range of 150 to 850) in blood samples with widely varying properties, the LOD varied over four orders of magnitude.16 The chemical matrix also significantly affects the LOD, with poorer signal intensity seen in more complex matrices such as urine, waste water, and plasma. In a typical paper spray MS analysis increasing the sample volume beyond a couple of microliters does not improve detection limits because the size of the paper substrate, the volume of extraction/spray solvent, and the amount of matrix all must increase as well.26 By employing a solid phase extraction, however, the target analytes can be selectively 6 ACS Paragon Plus Environment

Page 7 of 22

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


concentrated while removing some of the matrix. Tandem mass spectra for two of the compounds studied, sulfamethazine and diazepam, are shown in figure 3. Analysis was performed in plasma using both on-cartridge SPE with 100 µL of sample and by the typical paper spray approach directly on 3µL of plasma. Compared to direct analysis, mass spectra using the SPE cartridge showed higher signal to noise (S/N). The absolute signal intensities of characteristic fragment ions of each target compound using the SPE cartridge were about 40 times higher than those obtained from direct analysis. The change in the signal intensity of the characteristic fragment ions for 17 different pharmaceuticals analyzed from plasma is shown in Figure 4, with most improving by more than a factor of 20. The results indicate that the SPE cartridge could be utilized to improve the detection limits as a result of selective enrichment of target molecules from larger sample volumes or reduction of ion suppression. Signal intensity for wet plasma samples was significantly worse than for dried samples. Analysis of wet plasma samples gave low and unstable analyte signal for all of the drugs evaluated. All subsequent experiments were performed on dried plasma samples Limits of detection (LOD) were determined for five of the test compounds in plasma using both normal paper spray and the integrated SPE paper spray cartridge (Table 1). By using the SPE cartridge, the LODs of the five drugs each improved significantly, decreasing by a factor of 14 to 70, depending on the compound. In these experiments, 100 µL sample and 5 mg SPE material were used. Detection limits could likely be further improved. First, the sample volume and the amount of SPE material in the column could be increased. Second, in typical SPE methods, the SPE material is washed with water and often a low concentration organic/water mixture to remove impurities. While including these washing steps would have probably improved detection limits, we did not perform any washing steps in our experiments in order to keep the procedure simple. Finally, solvent optimization was not performed in this study, but would be expected to improve detection limits by improving recovery or increasing ionization efficiency. Sample volume In direct paper spray analysis, the sample volumes used are typically between 0.5 µL2 and 15 µL16, with little to no improvement in the analyte signal intensity when the sample volume is increased beyond a few microliters. Increasing the volume of the sample significantly in order to improve the detection limits is not feasible. The area of the paper triangle can be increased in proportion to the sample volume in order to accommodate more sample as shown in supplementary figure 1A. When solvent wicks through the sample to the tip of the paper, however, it does not interacted with the entire sample. It merely wicks through the paper in the shortest path to the tip. The effective amount of sample extracted by the spray solvent therefore does not increase in direct proportion to the sample volume. One way to get around this problem is to make a long, narrow spray substrate (supplementary figure 1A). However, the wicking rate through a horizontal substrate is inversely proportional to the square of the distance the solvent has traveled27. With a long spray substrate, particularly when wicking through plasma or blood, the wicking rate was not fast enough to sustain the spray. Finally, ionization suppression and relative recovery get worse as the sample volume increases, which 7 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 8 of 22

offsets the increased sample volume significantly. This topic is explored in more detail in the next section. For these reasons, increasing the sample size in direct paper spray is not straight-forward. In paper spray with integrated SPE, on the other hand, larger sample volumes can be applied to the cartridge to improve signal intensity and detection limits. Excess sample flows onto the waste pad while analytes are retained by the SPE material (depending on their respective properties). The signal intensities in MS/MS mode for five different drugs (200 ng/mL) in plasma at sample volumes ranging from 3 to 150 µL are shown in Figure 5. The amount of SPE material was fixed at 5 mg. Signal intensity for each drug increased as the sample volume was increased. At sample volumes larger than 10 µL, in particular, MS/MS signal intensity significantly improved due to preconcentration of the analyte on the SPE material, roughly in proportion to the sample volume. At larger volumes, analyte breakthrough would eventually occur, which could be counteracted by increasing the mass of SPE material. This data is also shown in supplementary table 1 as signal increase relative to direct paper spray analysis of a 3 µL plasma sample. Recovery and Ionization Suppression In HPLC-MS assays, ionization suppression is normally determined by comparing the analyte peak area obtained in the presence and absence of matrix. Likewise, recovery is determined by comparing analyte peak area when spiking before versus after extraction. In paper spray, extraction and ionization occur simultaneously. The decrease in analyte peak area in the presence of matrix is therefore due to a combination of ionization suppression and lower recovery. Another difference is that post-extraction spiking experiments are not possible in the traditional sense. We therefore determined the recovery and ionization suppression in paper spray analysis of plasma (both with and without integrated SPE) relative to spotting neat analyte on paper. This is a departure from the normal procedure in HPLC-MS assays in which ionization suppression and recovery are measured relative to having the analyte dissolved in solvent. Ionization suppression and recovery for 3 µL and 100 µL plasma samples by paper spray with and without integrated SPE are shown in Table 2. Ionization suppression is expressed as a percent decrease relative to blank paper, whereas recovery is epxressed as the percent recovered from plasma relative to blank paper. Overall, paper spray with integrated SPE had less ionization suppression and higher recovery compared to direct paper spray for each of the 5 drugs studied at both sample volumes. To scale up the sample volume in direct paper spray, we fabricated the device depicted in supplementary figure 1B. Sample volume was increased by adding any number of 3mm discs of paper, each containing 3 µL of sample. The punches had to be spotted with sample and dried separately in the open air, and then loaded into the holder. This was necessary because if a large volume of plasma was spotted together on the column of paper, the plasma concentrated at the exposed surface as the water evaporated and became impenetrable to the solvent after drying. This was a tedious process and is not a feasible approach to increase sample volume in paper spray. It was employed here only as a means of scaling up the sample volume for direct paper spray analysis for the sake of 8 ACS Paragon Plus Environment

Page 9 of 22

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


comparison. At a sample size of 3 µL, no preconcentration occurred in the SPE cartridge because there was insufficient volume to break through the SPE column and onto the waste pad. Nevertheless, ion suppression was greater for direct paper spray and recovery was lower compared to paper spray with integrated SPE. Apparently, recovering the analyte from paper is more difficult than from the SPE material for these 5 drugs. Likewise, we speculate that some matrix components which caused ion suppression were retained by the SPE material during the extraction/spray process, but not by the paper. When the sample size was increased from 3 µL to 100 µL for direct paper spray, the ionization suppression generally increased while the recovery decreased. For paper spray with integrated SPE, ionization suppression also seemed to increase at the larger sample volume. The magnitude of the increase was not as large as for direct paper spray, however, and the overall ionization suppression remained lower for paper spray with integrated SPE compared to direct paper spray. Recovery for paper spray with integrated SPE is more complex because it is a function of both initial analyte retention during sample application and analyte extraction efficiency when the spray solvent is applied. For two of the drugs (alprazolam and diazepam), the recovery decreased at the larger sample volume presumably due to poor retention during sample application. Two of the drugs actually showed substantially higher recovery at 100 µL compared to 3 µL. At 3 µL sample volume, a substantial portion of the analyte was likely contained within the paper disc placed on top of the SPE material. For 100 µL samples, on the other hand, the bulk of the analyte was in the SPE material. Because analyte elution from paper is less efficient than from SPE material, the 100 µL sample may show higher percent recovery provided that retention is good. Compared to more traditional sample preparation workflows, the ionization suppression and recovery of paper spray (with or without SPE) is relatively poor. While this has a negative impact on detection limits, it does not prevent the development of a robust quantitative assay. If matrix matched calibrators and stable isotopically labeled internal standards are used (as described in the next section), acceptable quantitative performance can be obtained. Quantitative analysis Paper spray has been shown to have acceptable quantitative performance for target compounds in complex biological samples, such as whole blood16 and urine.2 The SPE cartridge was evaluated for quantitative analysis of five drugs in plasma using stable isotopically labeled analogs as internal standards. The internal standards were mixed into the sample before adding the sample to the cartridge. A representative calibration curve for sulfamethazine is shown in Figure 6. The ratio of drug to internal standard demonstrated good linearity from the LOD to 1000 ng/mL for each drug, with the exception of atenolol which was linear up to 500 ng/mL. An R2 of >0.99 was obtained in all cases. For each drug, the imprecision of the measurement less than 10% at all standard concentrations above the estimated lower limit of quantitation (10*sB/m). The complete bias and imprecision data for all 5 drugs tested are shown in Table 3. The experimental data indicates that, with the use of 9 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 10 of 22

isotopically labeled internal standards, this method can be used for quantitative analysis.

CONCLUSION The aim of this work was to create an “all-in-one” disposable capable so that sample extraction, pre-concentration, and ionization could be performed from a single device. The analysis took the same number of steps as typical paper spray analysis (adding the sample, letting it dry, then adding the spray solvent), but improved the MS signal intensity and detection limits significantly. The five drugs tested showed good linearity when an isotopically labeled internal standard was spiked into the sample. Recovery and ionization suppression was compared to direct paper spray analysis of plasma, and paper spray analysis with integrated SPE was found to have lower levels of ionization suppression and better overall recovery. Compared to direct paper spray, the disadvantages of this approach are that a somewhat more complex cartridge is required and that larger samples volumes are required. Highly automated on-line SPE systems operating in direct injection mode would have similar throughput and probably better detection limits than the paper spray SPE cartridge. However, that level of automation is out of reach for many labs due to its cost and complexity. Using the paper spray cartridge with integrated SPE, dried samples could be shipped to the lab at room temperature and analyzed immediately without any additional handling or sample preparation by paper spray-MS. This is a significant simplification over current SPE workflows, and the detection limits are improved significantly relative to direct paper spray. The simple and automatic on-cartridge preparation could also enable more effective use of portable mass spectrometers or the use of mass spectrometers outside traditional analytical labs. The cartridge design shown here is a proof-of-concept prototype. Future efforts will focus on protecting the sample and spray tip from contamination or damage, which could occur during shipping. Additionally, the development of methods for incorporating the internal standard either at the point of sample collection or upon receipt at the laboratory are of interest to simplify the methodology and improve robustness. Finally, developing methods for analyzing wet plasma samples28 will be explored. The current implementation requires that the plasma samples be dried, which does not delay analysis if the paper spray cartridges are loaded with sample remotely and shipped to the lab. Methods for analyzing wet samples would decrease turn-around time for on-site or stat sample analysis. Acknowledgements This project was supported by Award No. 2014-R2-CX-K007 awarded by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice. The opinions, findings, and conclusions or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect those of the Department of Justice


Page 11 of 22

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


Table 1. Limits of Detection for Five Drugs by paper spray with integrated SPE compared with direct paper spray analysis drug

integrated SPE LOD (ng/mL)

Direct analysis LOD (ng/mL)

factor decrease

carbamazepine

0.34

7.9

23

atenolol

2.2

58

26

sulfamethazine

0.08

5.2

70

diazepam

6.1

121

20

alprazolam

1.3

18.5

14



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 22

Table 2. Ionization suppression and recovery for 5 different drugs in plasma. Analysis was performed with either direct paper spray analysis or using paper spray with integrated SPE. Three replicates were performed for each. 3 µL Sample Volume Direct PS PS on SPE cartridge

Alprazolam

Atenolol

Carbamaze.

Diazepam

Sulfametha.

Ion suppression

-72

-50%

-66%

-68%

-69%

Recovery

28%

30%

19%

30%

27%

Ion suppression

-49%

-23%

-42%

-48%

-50%

Recovery

70%

62%

29%

60%

57%

-88%

-76%

-83%

-80%

-92%

100 µL Sample Volume Direct PS PS on SPE cartridge

Ion suppression Recovery

18%

16%

22%

16%

24%

Ion suppression

-57%

-27%

-52%

-55%

-67%

Recovery

33%

65%

81%

29%

98%


Page 13 of 22

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


Table 3. Regression parameters, lower limit of quantitation, bias, and imprecision for analysis of calibration standards prepared in plasma Atenolol

Alprazolam

Carbamazepine

Diazepam

Sulfamethazine

Slope (m)

0.023

0.0075

0.013

0.010

0.0086

intercept LOD (ng/mL) a LLOQ (ng/mL) Standard (ng/mL) 0.1

0.0015

0.0022

0.000042

0.019

0.00021

2.2

1.3

0.3

6.1

0.08

7

4

1

20

.2

% Biasb

%CVc

% Bias

%CV

% Bias

%CV

% Bias

%CV

% Bias

%CV

< LOD

< LOD

< LOD

< LOD

< LOD

< LOD

< LOD

< LOD

0.6%

7%

1

< LOD

< LOD

< LOD

< LOD

3%

17%

< LOD

< LOD

-6%

8%

10

1%

4%

2%

9%

-13%

4%

1%

8%

6%

6%

100

-3%

1%

-2%

3%

-14%

2%

-2%

7%

-3%

7%

200

-8%

5%

-1%

5%

-11%

4%

-9%

2%

-2%

3%

500

10%

8%

1%

2%

-6%

5%

-1%

7%

-3%

2%

1000

>ULOQ

>ULOQ

8%

5%

14%

2%

11%

3%

8%

0.1%

a

estimated LLOQ. LLOQ = 10*sb/m % bias = (calculated concentration – actual concentration)/actual concentration*100% c % CV = relative standard deviation of replicate measurements of a single standard (N=3) b



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 22

Figure 1. Diagram depicting the workflow for paper spray analysis with integrated on-cartridge solid phase extraction.


Page 15 of 22

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. Photographs of a prototype paper spray cartridge with integrated SPE. (a) Cartridge is shown in the sample loading, extraction, and dying position above the waste absorption pad. SPE column is packed in the hole indicated with a red arrow. (b) Cartridge in the elution and detection position. Analytes are recovered and ionized in one step by adding 60 µL of an extraction/spray solvent to the SPE column, which wicks through the column and onto the paper spray substrate (c) Cartridge held in front of the mass spectrometer inlet for analysis. Ionization occurs by inducing an electrospray at the sharp tip of the paper near the MS inlet.



(b)

186.0 6

M=sulfamethazine m/z 279.0

4

[M+H]+

2

279.0

Intensity (×10)

Intensity (×103)

(a)

M=sulfamethazine 10

m/z 279.0

5

[M+H]+ 186.0

0

0 100

150

200

100

250

150

m/z

(c) 3

4

M=diazepam m/z 285.1

2

[M+H]+

1

285.1 0 100

150

200

250

(d)

257.0 4

200

279.0

m/z

250

300

Intensity (×102)

Intensity (×103)

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 22

3

257.0

M=diazepam m/z 285.1

2

[M+H]+

1

285.1 0 100

150

m/z

200

250

300

m/z

Figure 3. (a and c): Tandem mass spectra obtained for 200 ng/mL sulfamethazine and diazepam analyzed in 100 µL of plasma using paper spray with integrated SPE. The molecular ion and the primary fragment ion are labeled by m/z on the spectra. (b and d): MS/MS spectra obtained when analyzing the same plasma sample using the typical paper spray approach directly on 3µL of sample.


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


Signal Increase Relative to Direct Analysis

Page 17 of 22

50 on-cartridge SPE (100 µL plasma)

40

30

20

10

0

Figure 4. Relative increase in MS/MS signal intensity of the most intense fragment ion for 17 different drugs in plasma (200 ng/mL) analyzed by passing the 100 µL plasma sample through 5 mg of SPE material, followed by direct elution and analysis by paper spray MS, using the device shown in figures 1 and 2. Signal increase is relative to direct analysis of 3 µL of plasma by direct paper spray-MS/MS. Error bars are the standard deviation of the mean (N=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 18 of 22

Figure 5. Fragment ion intensities of five different drugs (200 ng/mL) in plasma obtained by on-cartridge SPE (5mg) paper spray ionization with plasma volumes of 3, 10, 50, 100, and 150 µL. Error bars show the standard deviation of the mean (N=8 at each volume)


Page 19 of 22

10

analyte:IS signal intensity ratio

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


0.009 0.006

8

0.003 6

0 0

0.5

1

4 2 0 0

200

400

600

800

1000

sulfamethazine concentration in plasma (ng/mL) Figure 6. Calibration curve for sulfamethazine from 100 µL plasma sample using paper spray with integrated SPE (5 mg) using an isotopically labeled internal standard. Inset shows the two lowest calibration standards (0.1 and 1 ng/mL). Error bars show the standard deviation of replicate measurements of a single standard (N=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 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 22

References (1) Wang, H.; Liu, J.; Cooks, R. G.; Ouyang, Z. Angewandte Chemie International Edition 2010, 49, 877. (2) Liu, J.; Wang, H.; Manicke, N. E.; Lin, J.-M.; Cooks, R. G.; Ouyang, Z. Analytical Chemistry 2010, 82, 2463. (3) Shi, R.-Z.; El Gierari, E. T. M.; Manicke, N. E.; Faix, J. D. Clinica Chimica Acta 2015, 441, 99. (4) Espy, R. D.; Teunissen, S. F.; Manicke, N. E.; Ren, Y.; Ouyang, Z.; van Asten, A.; Cooks, R. G. Analytical Chemistry 2014, 86, 7712. (5) Wang, H.; Ren, Y.; McLuckey, M. N.; Manicke, N. E.; Park, J.; Zheng, L. X.; Shi, R. Y.; Cooks, R. G.; Ouyang, Z. Analytical Chemistry 2013, 85, 11540. (6) Hamid, A. M.; Jarmusch, A. K.; Pirro, V.; Pincus, D. H.; Clay, B. G.; Gervasi, G.; Cooks, R. G. Analytical Chemistry 2014, 86, 7500. (7) Oradu, S. A.; Cooks, R. G. Analytical Chemistry 2012, 84, 10576. (8) Liu, W.; Wang, N. J.; Lin, X. X.; Ma, Y.; Lin, J. M. Analytical Chemistry 2014, 86, 7128. (9) Zhang, Z. P.; Cooks, R. G.; Ouyang, Z. Analyst 2012, 137, 2556. (10) Li, A. Y.; Wei, P.; Hsu, H. C.; Cooks, R. G. Analyst 2013, 138, 4624. (11) Yang, Q.; Manicke, N. E.; Wang, H.; Petucci, C.; Cooks, R. G.; Ouyang, Z. Anal. Bioanal. Chem. 2012, 404, 1389. (12) Naccarato, A.; Moretti, S.; Sindona, G.; Tagarelli, A. Anal. Bioanal. Chem. 2013, 405, 8267. (13) Zhang, Y. D.; Li, H. F.; Ma, Y.; Lin, J. M. Analyst 2014, 139, 1023. (14) Manicke, N. E.; Belford, M. Journal of The American Society for Mass Spectrometry 2015, 26, 701. (15) Salentijn, G. I. J.; Permentier, H. P.; Verpoorte, E. Analytical Chemistry 2014, 86, 11657. (16) Manicke, N. E.; Abu-Rabie, P.; Spooner, N.; Ouyang, Z.; Cooks, R. G. Journal of the American Society for Mass Spectrometry 2011, 22, 1501. (17)Cooks, R. G.; Manicke, N. E.; Dill, A. L.; Ifa, D. R.; Eberlin, L. S.; Costa, A. B.; Wang, H.; Huang, G.; Ouyang, Z. Faraday Discussions 2011, 149, 247. (18) Xu, W.; Manicke, N. E.; Cooks, G. R.; Ouyang, Z. Journal of the Association for Laboratory Automation 2010, 15, 433. (19) Kennedy, J. H.; Aurand, C.; Shirey, R.; Laughlin, B. C.; Wiseman, J. M. Analytical Chemistry 2010, 82, 7502. (20)Gómez-Ríos, G. A.; Pawliszyn, J. Angewandte Chemie International Edition 2014, 53, 14503. (21) Mirnaghi, F. S.; Pawliszyn, J. Analytical Chemistry 2012, 84, 8301. (22) Ren, Y.; McLuckey, M. N.; Liu, J.; Ouyang, Z. Angewandte Chemie International Edition 2014, 53, 14124. (23) Lafrenière, N. M.; Mudrik, J. M.; Ng, A. H. C.; Seale, B.; Spooner, N.; Wheeler, A. R. Analytical Chemistry 2015, 87, 3902. 20 ACS Paragon Plus Environment

Page 21 of 22

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


(24) Kirby, A. E.; Lafrenière, N. M.; Seale, B.; Hendricks, P. I.; Cooks, R. G.; Wheeler, A. R. Analytical Chemistry 2014, 86, 6121. (25) Long, G. L.; Winefordner, J. D. Analytical Chemistry 1983, 55, A712. (26) Yang, Q.; Wang, H.; Maas, J. D.; Chappell, W. J.; Manicke, N. E.; Cooks, R. G.; Ouyang, Z. International Journal of Mass Spectrometry 2012, 312, 201. (27) Washburn, E. W. Physical Review 1921, 17, 273. (28) Espy, R. D.; Manicke, N. E.; Ouyang, Z.; Cooks, R. G. Analyst 2012, 137, 2344.



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 22

For Table of Contents Only

Sample Extraction

Ionization


Development of a Paper Spray Mass Spectrometry ... - ACS Publications

Recommend Documents