Characterization of Polymer Membranes by MALDI Mass

Apr 17, 2013 - Table 1 summarizes the masses of the ion signal series around the fragment m/z 618.363 ... PS signals were overlaid by a signal pattern...
1 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/ac

Characterization of Polymer Membranes by MALDI MassSpectrometric Imaging Techniques Katharina Krueger,† Cindy Terne,†,‡ Carsten Werner,§ Uwe Freudenberg,§ Vera Jankowski,† Walter Zidek,† and Joachim Jankowski*,† †

Charité-Universitaetsmedizin Berlin, Medizinische Klinik IV, Hindenburgdamm 30, D-12200 Berlin, Germany Helmholtz Virtual Institute − Multifunctional Materials in Medicine, Berlin and Teltow, Germany § Leibniz Institute of Polymer Research, Dresden, Germany ‡

S Supporting Information *

ABSTRACT: For physical and chemical characterization of polymers, a wide range of analytical methods is available. Techniques like NMR and X-ray are often combined for a detailed characterization of polymers used in medical applications. Over the past few years, MALDI massspectrometry has been developed as a powerful tool for space-resolved analysis, not least because of its mass accuracy and high sensitivity. MALDI imaging techniques combine the potential of mass-spectrometric analysis with imaging as additional spatial information. MALDI imaging enables the visualization of localization and distribution of biomolecules, chemical compounds, and other molecules on different surfaces. In this study, surfaces of polymeric dialyzer membranes, consisting of polysulfone (PS) and polyvinylpyrrolidone (PVP), were investigated, regarding chemical structure and the compound’s distribution. Flat membranes as well as hollow fiber membranes were analyzed by MALDI imaging. First, analysis parameters like laser intensity and laser raster step size (spatial resolution in resulting image) were established in accordance with polymer’s characteristics. According to the manufacturing process, luminal and abluminal membrane surfaces are characterized by differences in chemical composition and physical characteristics. The MALDI imaging demonstrated that the abluminal membrane surface consists more of polysulfone than polyvinylpyrrolidone, and the luminal membrane surface displayed more PVP than PS. The addition of PVP as hydrophilic modifier to polysulfone-based membranes increases the biocompatibility of the dialysis membranes. The analysis of polymer distribution is a relevant feature for characterization of dialysis membranes. In conclusion, MALDI imaging is a powerful technique for polymer membrane analysis, regarding not only detection and identification of polymers but also localization and distribution in membrane surfaces.

D

For characterization of polymers, a wide spectrum of analytical techniques is available. For example, polymers are analyzed by spectroscopy methods like NMR,5 X-ray photoelectron spectroscopy, IR spectroscopy, atomic force microscopy,6−8 and other microscopy techniques.9 These analytical techniques are focused on physical or chemical characteristics of the polymer membranes. In general, several techniques have to be combined to provide a detailed polymer characterization.10 Although the polymer distribution is essential for both biocompatibility and efficiency, the distribution of the membrane compounds is not well characterized until now. In recent years, mass spectrometry (MS) has been established as a powerful tool in polymer analysis because of its high sensitivity and mass accuracy.11−14 By MALDI massspectrometric imaging, both identification of compounds and their localization and distribution can be determined. Thin sections of polymer membranes are transferred onto glass slides and overlaid with matrix solution. Subsequently, the matrixcovered sample is placed in an ion source; spectra are acquired

ialyzer polymer membranes are grossly classified as cellulose-based and synthetically produced membranes. Although cellulose membrane’s clearance and mechanical properties qualify it for use in dialyzer membranes, hydroxyl groups of these membranes lead to strong activation of the complement system.1 Therefore, recent membranes are manufactured from synthetic polymers. Common polymers for membranes used in medical applications are polyacrylonitrile, polymethylmethacrylate, polyamide, and polysulfone.2,3 Polymers are characterized by the chemical composition of its repeating units, end groups, and individual oligomer’s molecular weight distribution. The majority of dialysis membranes mainly consist of polysulfone (PS). Polysulfone is a thermoplastic polymer with high thermal and chemical stability.4 Because of its hydrophobicity, a hydrophilic modifier, e.g., polyvinylpyrrolidone (PVP), is added to PS to increase the membrane’s hydrophilicity. The use of PS and PVP results in polymer membranes with excellent biocompatibility and well-defined pore-structure with a fine pored luminal and a wide pored abluminal surface. The luminal surface corresponds to the inner surface of the dialysis hollow fiber membrane; the abluminal membrane surface corresponds to the outer surface of a dialysis hollow fiber. Molecular structures of PS and PVP are shown in Figure 1A,B. © 2013 American Chemical Society

Received: January 20, 2013 Accepted: April 17, 2013 Published: April 17, 2013 4998

dx.doi.org/10.1021/ac4002063 | Anal. Chem. 2013, 85, 4998−5004

Analytical Chemistry

Article

Figure 1. Chemical structures of (A) polysulfone (PS) and (B) polyvinylpyrrolidone (PVP).

without the possibility to analyze and optimize the distribution of these compounds by an analytic method before using the membranes in physiologic assays. However, methods for detection of these substances with high accuracy and high specificity on the membrane surface are not available; the accuracy and in parts the specificity of time of flight-secondary ion mass spectrometry (TOF-SIMS) technology is limited because of the physical limitations of the TOF analyzer. The MALDI imaging techniques in combination with the FT technique lead to a more accurate identification, characterization, and quantification of components than TOFSIMS technology. Therefore, we used in the present study the MALDI massspectrometric imaging techniques for detection of PVP and PS on polymer membrane surfaces. The results characterize MALDI mass-spectrometric imaging as a useful tool for molecular characterization of polymer membrane surfaces.

at different positions distributed across the sample surface defining an x, y grid. Each mass spectrum reflects the composition of the components at a defined position. The intensity of any given mass signal or the combination of signals can be visualized in images, allowing analysis of compound distribution.15 Distribution of polymers in biomaterials is essential for the biocompatibility of the materials. For example, distribution of PVP is crucial for the biocompatibility of dialyzer membranes; the detection of the polymers is essential for analyzing and increasing biocompatibility. Exposure of blood to hemodialysis membranes results in numerous interactions between the blood components and the membrane having the potential to induce an inflammatory response and to lead to numerous long-term clinical sequel that are in part determined by the degree of membrane biocompatibility. A biocompatible membrane has traditionally been defined as “one that elicits the least amount of inf lammatory response in patients exposed to it”.16 The Consensus Conference on Biocompatibility recommended the following criteria to assess material biocompatibility: (i) evaluation of thrombosis formation upon contact with blood with the artificial material (in this particular case, clot formation in the ECC); (ii) signs of coagulation system activation; and (iii) signs of thrombocyte activation. Polysulfone-based membranes are widely used for production of conventional hemodialysis membranes since these membranes fulfill most of these criteria.16 Polysulfone-based biomaterials are presently the gold standard in the production of biocompatible hemodialyzers. Since polysulfone-based biomaterials are hydrophobic, these membranes have to be blended with hydrophilic polymers, such as polyvinylpyrrolidone4 to increase biocompatibility and wettability. The resulting membranes are characterized by physical and chemical techniques regarding membrane structure, defined in terms of size, form, and distribution of the pores at the luminal separating layer.4 Second, the physical and chemical properties of the bloodcontacting surface should minimize the interaction of blood and biomaterial neither affecting the membrane function nor causing adverse reactions for the patient. However, there is no standard technique for the measurement of compounds like polyvinylpyrrolidone distribution on the luminal and abluminal membrane side available until now. As a result, the biocompatibility is solely determined by physiologic assays

■

MATERIALS AND METHODS Preparation of Membranes Used in the Study. Two different types of polymer membranes were analyzed within this study, flat membranes and hollow fiber membranes. Asymmetric flat membranes were produced on laboratory scale. Briefly, a solution of 15% polysulfone and 3.75% PVP in dimethylacetamide was spread out by a scalpel and precipitated by a mixture of 70% dimethylacetamide and 30% water. The precipitation results in an asymmetric flat membrane with a fine porous surface area formed at the site of first contact with the precipitation liquid and a larger pore surface at the opposite side. The freshly formed membrane was incubated in distilled water overnight and dried subsequently. During this process, membrane sides present different surfaces corresponding to luminal and abluminal surfaces of hollow fiber membranes. Membranes were cut in small pieces (approximately 1 cm × 1 cm). For analysis, both membrane sides were fixed with adhesive tape onto glass slides (SuperfrostPlus, Thermo Scientific, Germany). Hollow fiber membranes were taken from the commercial dialyzer FX-800 (Fresenius Medical Care, Germany) and were placed onto glass slides. For analysis of luminal membrane surfaces, hollow fiber membranes were opened with a longitudinal cut by a scalpel. Membrane samples were fixed 4999

dx.doi.org/10.1021/ac4002063 | Anal. Chem. 2013, 85, 4998−5004

Analytical Chemistry

Article

Figure 2. SEM images of (A) flat membrane surfaces: abluminal side (left image) and luminal side (right image); (B) hollow fiber membrane surfaces: abluminal side (left image) and luminal side (right image).

GmbH, Witten, Germany). While this processes, the substance in a liquid body crosses the boundary from liquid to gas and the liquid changes into gas at a finite rate, while the amount of liquid decreases. By using the critical point pressure chamber for all fluid exchanges, loss of material and air-drying artifacts due to specimen handling are reduced.17 Samples were then sputter-coated with gold (Sputter Coating Device SCD 050, BAL-TEC GmbH) and examined at 10 kV accelerating voltage in an environmental scanning electron microscope (PHILIPS XL 30 ESEM FEG; Philips, Eindhoven, The Netherlands).

with the luminal surface upward onto glass slides. 2,5Dihydroxybenzoic acid was dissolved at a concentration of 40 mg/mL in 50% ethanol (Merck; Germany) and 0.1% trifluoroacetic acid (TFA, Sigma-Aldrich; Germany). αCyano-4-hydroxycinnamic acid (Sigma-Aldrich; Germany) was dissolved at a concentration of 10 mg/mL in 50% acetonitrile (Sigma-Aldrich; Germany) and 0.1% TFA. Sinapinic acid (Sigma-Aldrich; Germany) was dissolved at a concentration of 20 mg/mL in 50% acetonitrile (SigmaAldrich; Germany) and 0.1% TFA. Matrix deposition was achieved by a MALDI spotter (Sunchrom; Germany). Fixed membranes were coated with 12 thin layers matrix solution with an increasing flow rate from 20 to 40 μL/min. Following each matrix deposition cycle, the membrane surfaces were allowed to dry at room temperature for 30 s. MALDI Mass Spectrometric Imaging. Mass spectrometric imaging was performed by a LTQ Orbitrap XL with an MALDI ion source (Thermo Fisher Scientific; Germany). It is equipped with a nitrogen laser (MNL-100; LTB Lasertechnik Berlin; Germany) at 337 nm operating at a 60 Hz repetition rate and 3 ns pulse rate. All data acquisitions were performed in the centroid mode and using the mass-range of 500−2.000 (m/ z), since the S/N was increased compared to higher mass-range as is common. Fourier-transform mass-spectrometric (FTMS) data were acquired in a measuring grid across the membrane area with a resolution of 60.000. The sample areas were 1.000 × 2.000 μm for flat membranes and 900 × 500 μm for hollow fiber membranes. The raster distances varied between 20 and 100 μm. The laser intensity varied between 20 and 60 μJ, corresponding to laser fluences of approximately 120 to 370 J/m2. ThermoImageQuest (Thermo Fisher Scientific, Germany) was used for visualization of PVP and PS distributions on membrane surfaces. For FTMS imaging, selected polymer masses were normalized by the total ion current (TIC). Scanning Electron Microscopy. Membranes were washed, and the liquids were removed in a precise and controlled way by using the critical point-dried techniques using a CO2 system (Critical Point Dryer CPD 030, BAL-TEC

■

RESULTS The luminal and abluminal surfaces of dialysis membranes differ in pore structure, pore distribution, and polymer composition. Thus, both membrane surfaces were separately analyzed by MALDI FT-Orbitrap mass-spectrometric imaging. Figure 2A,B presents scanning electron microscope (SEM) images of flat membrane surfaces (above) and hollow fiber membrane surfaces (below). The left images show the abluminal membrane side with evenly distributed wide pores. The right images display the luminal surface of the membrane oriented to the lumen. The pore diameters of the luminal surface are smaller compared to pores of the abluminal surface. SEM analysis shows pore diameters of 20 to 60 nm for luminal surface and 600 nm to 3 μm for abluminal surface pores. Different matrixes were investigated for polymer characterization by mass spectrometry in pilot studies. Beside DHB, αcyano-4-hydroxy cinnamic acid and sinapinic acid as matrix for polymer analysis were tested. DHB was the best choice for measurement of polysulfone and polyvinylpyrrolidone (data not shown). Besides sample preparation, laser intensity as well as spatial resolution (raster step size) is an essential parameter of MALDI imaging to increase the signal-to-noise (S/N) ratio and resolution. Therefore, the influence of both parameters on the measurement of polymers was investigated. First, laser intensity was varied between 20 and 60 μJ at laser raster step size of 100 μm. These parameters resulted in an x−y grid of 100 μm for the resulting MALDI FTMS images. All mass 5000

dx.doi.org/10.1021/ac4002063 | Anal. Chem. 2013, 85, 4998−5004

Analytical Chemistry

Article

507.295 (Figure 3C, mass description for PVP fragment masses; see Table1). The green color represents the masssignal intensity of m/z 507.295. Figure 3D presents the overlay of both images of PS and PVP, showing the different localization of PS and PVP on the identical membrane surface. Whereas polysulfone was evenly distributed at the membrane surface, PVP was condensed in clusters of the membrane, detectable by the defined regions of mass signals of PVP (green spots). In contrast to the abluminal membrane surface, measurement of the luminal membrane showed no specific mass signals at laser intensities of 20 and 40 μJ (data not shown). However, masses of PVP were detected by using laser intensity at 60 μJ (Figure 4A). There are the repeating units of 111 m/z of PVP

spectra of the abluminal membrane surface taken with different laser intensities showed intensive signals at m/z 907.241, m/z 1349.368, and m/z 1792.496 (Figure 3A, recorded with 60 μJ).

Figure 3. (A) MALDI mass spectrum of abluminal membrane surface, measured with a laser raster step size of 100 μm and laser intensity of 60 μJ. The associated MALDI images of the distribution of (B) PS (red color) and (C) PVP (green color) on the same abluminal membrane and (D) its overlay.

The signal-to-signal distance of 442 Da of the main signals corresponds to the polysulfone repeating unit. The most intensive signal at m/z 907.241 represents two polysulfone repeating units with a Na+ adduct. An increase in laser energy to 40 and 60 μJ resulted in increased signal intensities and signal-to-noise ratios. The analysis of pure polymer polysulfone showed a similar mass spectrum with masses m/z 907.241, m/z 1349.368, and m/z 1792.496 (see supplementary figure, Supporting Information). The two-dimensional MALDI image given in Figure 3B shows the localization and distribution of the polysulfone oligomer (m/z 907.241), accumulated by a laser energy of 60 μJ with a raster step size of 100 μm. The red color represents the mass-signal intensity of m/z 907.241. Although mass-signal intensities for polyvinylpyrrolidone were low, we were able to visualize the distribution of the polyvinylpyrrolidone mass m/z

Figure 4. (A) MALDI mass spectrum obtained from a luminal membrane surface at 100 μm laser raster step size and laser intensity of 60 μJ (the inset shows four different signal series for PVP). The associated MALDI images of distribution of (B) PS (red color) and (C) PVP (green color) on the same luminal membrane and (D) the overlay of both.

shown. The mass spectrometric measurement of PVP showed at least four different oligomeric signal series for PVP, each with

Table 1. Overview about Mass Signals, Corresponding to Observed Four Mass Signal Series for PVP, Demonstrated by the Example of Mass Signal m/z 618.363

5001

dx.doi.org/10.1021/ac4002063 | Anal. Chem. 2013, 85, 4998−5004

Analytical Chemistry

Article

a signal-to-signal distance of 111 Da (as shown in the example of mass signal m/z 618 in the expanded spectrum in the inset in Figure 4A). It is obvious that additional mass signals represent different polymer fragments caused by breakdown of higher molecular weight polymer chains within the membrane. Table 1 summarizes the masses of the ion signal series around the fragment m/z 618.363 with possible corresponding chemical structures of these polymer fragments. According to this the PVP mass, m/z 618.363 consists of four repeating units of 111 Da, a sodium adduct, and the C2H2• radical. A comparable mass spectrum was acquired by analyzing authentic PVP (see supplementary figure, Supporting Information). In addition to PVP signals, the analysis of a luminal membrane surface showed mass signals for PS at m/z 907.241 and m/z 1349.365. Figure 4B demonstrates the distribution of polysulfone (m/z 907.241), using a laser intensity of 60 μJ and laser grid of 100 μm. Figure 4C shows the MALDI image of PVP mass signal distribution (m/z 507.295) on the membrane surface. The overlay of these images demonstrates the different distribution of PVP and PS (Figure4D). PVP is obviously more uniformly distributed than the PS as indicated by the mass signals. Another essential aspect in analysis of polymer membranes is the spatial resolution. For this analysis, we varied the laser raster size from 20 to 100 μm. To maximize the MALDI signal intensity and resolution, the laser intensity was set to 60 μJ. In the case of the abluminal membrane analysis, a laser raster size of 20 μm showed low S/N for PS signals. PS signals were overlaid by a signal pattern with mass distance of 24 Da (Figure 5A). The inset in this figure displays the PS signal m/z 1349.365, which is overlaid by the pattern with distance of 24 Da. This pattern is not derived from fragmentation of PS; however, the origin of these fragments is unknown. A significant decrease in laser energy with the same laser raster did not show this signal pattern (data not shown). However, analyses with setting laser raster step sizes of 50 and 100 μm resulted in a significant increase in the S/N of polysulfone. Figure 5B,C clearly demonstrates the increase of the signal intensities of PS fragments m/z 907.241, m/z 1349.365, and m/z 1792.496. Figure 5D,E shows the corresponding MALDI images for PS m/z 907.241 (Figure 5D), for PVP m/z 507.295 (Figure 5E), and the overlay of both at grid distance of 50 μm (Figure 5F). In comparison to MALDI-FT mass-spectrometric images resulting by a grid distance of 100 μm (Figure 3B−D), an increased spatial resolution is performed by a 50 μm raster step size, resulting in an increased accuracy of mass signal localization. The mass spectrometric measurement of the luminal membrane surface showed comparable mass spectra, recorded with small as well as wide raster step size. In comparison to a laser raster step size of 20 μm (Figure 6A), an increased S/N for PVP signals was obtained using a grid distance of 50 μm (Figure 6B). A raster step size of 100 μm did not result in a further increase in S/N for PVP oligomers (Figure 6C). Figure 6D−F shows the MALDI mass-spectrometric images of the distribution of mass signals m/z 907.241 for PS and m/z 507.295 for PVP and the corresponding overlay. Mass spectrometric measurement using a raster step size of 50 μm shows increased local resolution compared to a raster step size of 100 μm (Figure 4B−D). In the following, on the basis of the above results, we used a laser intensity of 60 μJ and a raster step size of 50 μm for the mass-spectrometric analysis of commercially available hollow fiber membranes. The resulting data are presented in Figure 7.

Figure 5. MALDI mass spectra of abluminal membrane surface, measured with laser intensity of 60 μJ and depending on laser raster step size: (A) 20 μm, (B) 50 μm, and (C) 100 μm (the inset in A shows an unknown polymeric signal series at low laser raster size). MALDI images of distribution of (D) PS (red color) and (E) PVP (green color) on the same abluminal membrane, obtained with settings of 60 μJ laser intensity and a 50 μm grid distance. (F) The overlay of both.

Optical images of both measured membrane areas are given in Figure 7A (abluminal membrane side) and Figure 7B (luminal membrane side). Figure 7C shows a MALDI mass spectrum obtained from the abluminal membrane surface. While the characteristic mass-signal pattern with a distance of 442 Da for polysulfone was detected with S/N > 10, the S/N of PVP was lower than 3. In contrast, the mass spectrum of the luminal surface shows the characteristic mass-signal series for PVP with a signal-to-signal distance of 111 Da and additionally the PS signal pattern (Figure 7D). Both mass spectra clearly show the different chemical composition of abluminal and luminal membrane. These mass-spectrometric data result in unique images (Figure 7E−J). Imaging of both membrane surfaces show that polysulfone is present more on the abluminal membrane surface than on the luminal membrane surface (Figure 7E vs Figure 7F). In contrast, polyvinylpyrrolidone amount is increased on the luminal surface compared to the abluminal surface (Figure 7G vs Figure 7H). Furthermore, the distribution of PS is more homogeneous on the abluminal membrane side, while the PVP mass signals are more evenly distributed on the luminal membrane surface. As shown in Figure 7I,J, both overlay images represent differences in polymer distribution and composition in abluminal and luminal membrane surfaces, comparable with flat membrane structures. Polyvinylpyrrolidone is present more in the luminal membrane 5002

dx.doi.org/10.1021/ac4002063 | Anal. Chem. 2013, 85, 4998−5004

Analytical Chemistry

Article

Figure 6. MALDI mass spectra obtained from luminal membrane surface at laser intensity of 60 μJ and depending on laser raster step size: (A) 20 μm, (B) 50 μm, and (C) 100 μm. MALDI images of distribution of (D) PS (red color) and (E) PVP (green color) on the same luminal membrane with settings of 60 μJ laser intensity and a grid distance of 50 μm. (F) The overlay of both.

Figure 7. Optical images in (A) abluminal membrane side and (B) luminal membrane surface of hollow fiber membranes. MALDI mass spectra, obtained from (C) abluminal membrane surface and (D) luminal membrane surface at 60 μJ laser intensity and a laser raster step size of 50 μm. MALDI images of distribution of PS (red color) in (E) and (F) and PVP (green color) in (G) and (H), respectively, shown for abluminal and luminal membrane surface. (I, J) The corresponding overlays of both polymers.

surface than in the abluminal membrane surface, whereas polysulfone is much more detectable in the abluminal membrane surface.

■

DISCUSSION Mass spectrometry is a powerful tool for investigation of polymers, concerning their molecular masses and mass distribution. The number of applications of MALDI mass spectrometry for polymer characterization has increased significantly.18 The present study focused on MALDI-FTOrbitrap imaging as a new method for the characterization of polymers. MALDI imaging allowed for the first time the simultaneous analysis of composition and distribution of polymers on membrane surfaces. Choosing an optimal matrix and sample preparation, as well as optimizing application parameters, is important to obtain high quality mass spectra.19 In this study, polymer membranes were investigated by MALDI FTMS imaging. Membranes were overlaid with several thin layers of different matrixes, with the flow rate being increased in each cycle. It is essential to start at low flow-rate cycles to achieve an equal and fine crystallized matrix distribution and to get high-quality spectra. Furthermore, matrix deposition by spraying ensures favorable spot-to-spot and shot-to-shot reproducibility within mass-spectrometric analysis.20 Sodium adducts were detected in each spectrum, since sodium is ubiquitous in glassware, solvents, reagents, etc.

and works in the mass spectrometric analysis of polymers as cationization agent.21 After DHB was identified as the polymer of interest for analysis, laser power as well as laser raster step size affect S/N of the spectra and the image’s spatial resolution. Therefore, both parameters were investigated to get maximal signal intensity and high spatial resolution. It was shown that PS and PVP indicated differences in its ionization characteristics. While PS fragments were identified with low laser intensities, a characterization of PVP fragments is preferentially achievable at high laser power. According to this, both polymers show different fragmentation behavior. On the basis of the mass spectra, there is a clear difference in the signal pattern of the polymers. A single signal series is monitored with signal to signal of 442 Da for PS. The analysis of PVP displayed at least four different signal series with a mass difference of 111 Da. The series are assigned as different fragments of PVP during the ionization process, which correspond to evaluated structures. Trimpin et al. already described a theoretic PVP fragment, determined by MALDI-TOF mass spectrometry.22 They demonstrated that an increase in laser power caused decreased signal intensities for this unknown PVP adduct. A further 5003

dx.doi.org/10.1021/ac4002063 | Anal. Chem. 2013, 85, 4998−5004

Analytical Chemistry

Article

increase of laser intensity to 60 μJ showed no signal reduction for PVP adducts in this case. For determination of polymer localization and distribution, optical resolution is an essential additional feature. By variation of laser step size, the optimal optical resolution is reached. In this study, grid distances between 20 and 100 μm were investigated. A laser raster step size with a distance of 50 μm provided mass spectra with lower signal intensities in contrast to 100 μm but with an increased signal-to-noise ratio. Furthermore, the spatial resolution of resulting images is much better than with a grid of 100 μm. An additional decrease of grid distances is feasible, but in parallel, the laser beam itself is a limiting factor.23 Lasers with beam sizes in the range of 100 μm are typically used in MALDI mass spectrometers. The use of a raster step size less than the size of laser beam area results in a surface overlap of adjacent sample positions. This results in a decrease in signal intensity, since overlapping area results in no or reduced signal intensities. This effect is confirmed by the analyses of polysulfone on abluminal membrane sides with raster step sizes of 50 and 100 μm. Mass spectra, acquired with 20 μm laser raster, show low signal intensities for PS, overlaid by a pattern with a signal-to-signal distance of 24 Da. Although the source of these signals is still unknown, PS as the parent molecule is unlikely. This signal pattern is not detected using low laser energies. Compared with PS on the abluminal membrane surface, the analyses of the luminal membrane surface presented signals for PVP for the investigated grid distances. There is no significant decrease in signal intensity observed. Even the analysis with a laser raster step size of 20 μm showed intensive PVP signals. In conclusion, MALDI imaging in combination with an Orbitrap analyzer is an appropriate method for characterization of polymer membranes and in the future in addition to bounded peptides or proteins.

■

(5) Raith, K.; Kuhn, A. V.; Rosche, F.; Wolf, R.; Neubert, R. H. Pharm. Res. 2002, 19, 556−560. (6) Hayama, M.; Yamamoto, K.; Kohori, F.; Sakai, K. J. Membr. Sci. 2004, 234, 41−49. (7) Wanakule, N. S.; Nedoma, A. J.; Robertson, M. L.; Fang, Z.; Jackson, A.; Garetz, B. A.; Balsara, N. P. Macromolecules 2008, 41, 471−477. (8) Heigl, N, P. C.; Rainer, M.; Najam-ul-Haq, M.; Vallant, R. M.; Bakry, R.; Bonn, G. K.; Huck, C. W. J. Near Infrared Spectrosc. 2007, 15, 269−282. (9) Magonov, S. N.; Reneker, D. H. Annu. Rev. Mater. Sci. 1997, 27, 175−222. (10) Ariza, M. J.; Pradanos, P.; Rico, R.; Rodriguez-Castellon, E.; Benavente, J. Surf. Interface Anal. 2003, 35, 360−368. (11) Ellison, S. T.; Gies, A. P.; Hercules, D. M.; Morgan, S. L. Macromolecules 2009, 42, 3005−3013. (12) Weidner, S. M.; Falkenhagen, J. Anal. Chem. 2011, 83, 9153− 9158. (13) Trimpin, S.; Clemmer, D. E. Anal. Chem. 2008, 80, 9073−9083. (14) Hanton, S. D. Chem. Rev. 2001, 101, 527−569. (15) Cornett, D. S.; Reyzer, M. L.; Chaurand, P.; Caprioli, R. M. Nat. Methods 2007, 4, 828−833. (16) Grooteman, M. P.; Nube, M. J. Nephrol., Dial., Transplant. 2004, 19, 2965−2970. (17) Brown, J. N. J. Microsc. 1977, 111, 351−358. (18) Mautjana, N. A.; Pasch, H. Macromol. Symp. 2012, 313−314, 157−161. (19) Schwartz, S. A.; Caprioli, R. M. Methods Mol. Biol. 2003, 656, 3− 19. (20) Axelsson, J.; Hoberg, A.-M.; Waterson, C.; Myatt, P.; Shield, G. L.; Varney, J.; Haddleton, D. M.; Derrick, P. J. Rapid Commun. Mass Spectrom. 1997, 11, 209−213. (21) Chen, H.; He, M.; Pei, J.; He, H. Anal. Chem. 2003, 75, 6531− 6535. (22) Trimpin, S.; Eichhorn, P.; Rader, H. J.; Mullen, K.; Knepper, T. P. J. Chromatogr., A 2001, 938, 67−77. (23) Jurchen, J. C.; Rubakhin, S. S.; Sweedler, J. V. J. Am. Soc. Mass Spectrom. 2005, 16, 1654−1659.

ASSOCIATED CONTENT

S Supporting Information *

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

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: (+49)-30-450525 567. Fax: (+49)-30-450-525 923. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors were supported by a grant from the Federal Ministry of Education and Research (BMBF-NET; 13N11416), Federal Ministry of Economics and Technology (KF2263201SB9), and Helmholtz Foundation to J.J. and by the Else-Kroener Foundation (P40/09//A29/09) to V.J.

■

REFERENCES

(1) Cruz, D. N.; de Cal, M.; Ronco, C. Contrib. Nephrol. 2008, 161, 89−98. (2) Subramanian, S.; Venkataraman, R.; Kellum, J. A. Kidney Int. 2002, 62, 1819−1823. (3) Hakim, R. M.; Held, P. J.; Stannard, D. C.; Wolfe, R. A.; Port, F. K.; Daugirdas, J. T.; Agodoa, L. Kidney Int. 1996, 50, 566−570. (4) Bowry, S. K.; Gatti, E.; Vienken, J. Contrib. Nephrol. 2011, 173, 110−118. 5004

dx.doi.org/10.1021/ac4002063 | Anal. Chem. 2013, 85, 4998−5004