Integrated Protein Microchip Assay with Dual Fluorescent- and MALDI

1. Introduction ... to engage in scientific and educational proteomics activities, and to assist in the coordination of public proteome initiatives th...
0 downloads 0 Views 417KB Size
Download PDF
Integrated Protein Microchip Assay with Dual Fluorescent- and MALDI Read-Out D. Finnskog,† A. Ressine,† T. Laurell,‡ and G. Marko-Varga*,‡ Department of Electrical Measurements, Lund Institute of Technology, Lund University, P.O. Box 118, S-221 00 Lund, Sweden and Department of Analytical Chemistry, Lund University, P.O. Box 124, S-221 00 Lund Sweden Received April 4, 2004

A pore chip protein array (PCPA) concept based on a dual readout configuration, fluorescence imaging, and MALDI-TOF MS has been developed. Highly packed, (>4000 spots/cm2), antibody arrays were dispensed on the porous chip by using a piezo-electric microdispenser. Sandwich assay was made after blocking by addition of a secondary antibody either IgG-FITC-labeled or anti-Ang II. The antigen in the first system was a large protein (IgG), and in the other system, a FITC marked peptide Angiotensin II (Ang II) was used. Ang II antibodies showed specificity for Ang II, while the Ang I antibodies showed binding properties for Ang I, II, and Renin. Fluorescence and MALDI TOF MS read-out was made for IgG and Ang II. A major advantage of the dual read-out PCPA approach is that both affinity binding and mass identity are derived. Detection limits for Ang II on the chip is as low as 500 zmol (Ang II). Keywords: matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF)

1. Introduction As an emerging field in life sciences, proteomics is based upon, and being developed beyond, genomics. Consequently, large efforts are now put into developing new bioanalytical platform to facilitate protein analysis in studies with complex biological samples. The traditional way of protein expression mapping by the use of 2D-gel electrophoresis has certain limitations such as limited detectability within the lower abundancy region. As a competitor to 2D-PAGE, multidimensional liquid-phase separations are now being developed and utilized.1-5 An important benefit of the multidimensional chromatography approach is the possibility of simultaneously obtaining both quantitative and qualitative information of the expressed proteins. Another rapidly emerging approach to protein mapping is the protein chip microarray technology field.6-8 In contrast to multidimensional chromatography, biologically complex samples in their native state with the intact proteins are commonly used. This area has a focus on two main developments: (a) the development of highly specific and stable affinity binders and (b) tailored surfaces with respect to surface chemistry as well as micro and nano morphology for the coupling of the affinity binder. Visions of the future possibilities that protein chip developments will offer to the proteomics area vary. However, it seems likely that protein microarrays will be important to bridge the gap between genomics and proteomics, being a complementary platform for deriving functional information on gene * To whom correspondence should be addressed. E-mail: [email protected]. † Department of Electrical Measurements, Lund Institute of Technology, Lund University. ‡ Department of Analytical Chemistry, Lund University.

988

Journal of Proteome Research 2004, 3, 988-994

Published on Web 09/11/2004

expression.9-12 There seem to be two directions that this development is taking: (i) a global approach, where most genetically coded human proteins will be mapped, and (ii) the focused approach, where a selected group of proteins are to be measured, such as a pathway, a protein class, e.g., kinases,13 or important regulatory proteins such as cytokines and chemokines. The latter group already exists as products from several instrumental vendors. These also offer instruments and chip consumables to make high throughput multiplexing analysis in a fully robotic setting. The ability to quantify samples with built-in algorithms that calculate accurate levels using internal standards and calibrants is impressive; however, to perform this operation on hundreds of thousands of proteins is still a vision. It is fair to say that the area of protein microchip research still has some development phases to pass before we are able to fully utilize the power of the high-density protein chip concept. Some recent developments within this area address new directions and openings that the field could take, both soon as well as for more long-term developments.14-18 One global project activity that addresses the technology drive to make proteomics analysis possible is the HUPO organization (www.HUPO.org). The mission of HUPO is to consolidate national and regional proteomics organizations into a worldwide network, to engage in scientific and educational proteomics activities, and to assist in the coordination of public proteome initiatives that stretch geographically over Europe, Asia, Oceania, and the Americas.19 Within HUPO, one of the five activities is specifically devoted to protein chip technology development.20 Recent papers have nicely illustrated the direction that high-density protein chip research is taking where the functional analysis of close to 6000 yeast proteins were analyzed.21 Global data analysis of protein activities using these proteome chips are available at: http://bioinfo.mbb.yale.edu/ 10.1021/pr0499287 CCC: $27.50

 2004 American Chemical Society

Integrated Protein Microchip Assay

research articles

Figure 1. Principles for the two approaches to assay protocols that were followed were (I) upper: large antigen detection of IgG with a-IgG directly deposited on the chip surface II) lower: small antigen detection, e.g., Ang II, by coupling a-AngII to a-IgG directly deposited on the chip surface.

proteinchip/db/. The potential of utlizing protein chips in disease context was recently demonstrated showing promising features in, e.g., cancer proteomics.22-24 One of the limitations of the conventional protein chip approach is the fact that the final read-out will not reveal whether a single or a multicomponent (i.e., multi-annotated) detection has been accomplished. To address this issue, this paper describes the development of a protein chip protocol that is based on a two step (dual read-out) analysis procedure on a single microchip, The described dual read-out-microchip technology offers high throughput protein array analysis with fluorescence detection in the first analysis dimension and in the second dimension MALDI mass identification of single or multiple antigen determination is applied. This principle can be applied to both small and large molecular weight antigens.

2. Experimental Section 2.1 Materials and Reagents. Silicon wafers (〈100〉 , 3′′, p-type, boron-doped, 10-15 Ωcm) were purchased from Topsil A/S (Denmark). Ethanol, 45% aqueous solution of HF were from Merck (Darmstad, Germany). Anti-rabbit IgG (anti-RIgG) and rabbit IgG-FITC (RIgG-FITC) were from Sigma (Aldrich, Sweden). Human plasma samples were obtained from healthy volunteers. Rabbit anti-Human Angiotensin II (anti-AngII) were from USBiological (Swampscott, USA) and Angiotensin II coupled to FITC by a Cystein linker at the C terminal were from Innovagen AB (Lund, Sweden). Rabbit Angiotensin I antiserum (anti-AngI), [Ile7]-Angiotensin III and Renin Tetradecapeptide were from Sigma (Aldrich, Sweden). 2.2 Pore Chip Protein Array (PCPA) Fabrication. Macroporous silicon supports were fabricated by anodic dissolution of silicon wafers in a mixture of hydrofluoric acid (HF, 45%) and dimethylformamide (DMF) (1:10, v/v) solution. This process was performed in a two compartment electrochemical cell. Current was passed through the silicon wafer to initiate and process the porous silicon layer formation. Macroporous silicon support surfaces were formed on p-type 10-15 Ωcm silicon wafers, 〈100〉 crystal orientation. The backside of the wafer was illuminated during anodization. A current density of 2 mA/cm2 was applied for 60 min. The porosified silicon wafer with the

macroporous layer on the top was diced into about 1 cm2 sized pieces, forming macroporous silicon chips. 2.3 Bioassay Protocol. To evaluate immobilization efficiency of the macroporous silicon chips two model antibody binding bioassays were performed. The capture antibodies Anti-rabbit IgG (aRIgG) were arrayed onto the macroporous silicon chip followed by a washing step in 10 mM PBS (pH 7.4) containing 0.05% v/v polyoxyethylene sorbitan monolaurate (Tween 20, PBS-tween). The capturing antibodies were adsorbed directly to the surface by means of physical adsorption. This adsorption is shown to be strong enough to avoid desorption of the molecules during washing protocols and to maintain the biological activity, that its ability to bind the antigens. Arrayed chips were blocked with PBS-tween containing 5 wt% nonfat milk for 20 min. After washing in PBS, the chips were incubated in a target antibody solution rabbit IgG-FITC (RIgG-FITC) for 1 h or with a secondary antibody rabbit antiAng II for 2 h incubation. Following the antibody incubations, the chips were washed twice in PBS-tween and then in pure water to remove salt from the surface. The Ang II assay was incubated with Ang II solution in PBS for 1 h before final washing. 2.4 Protein Microchip Handling. Microarrays were formed by spotting anti-RIgG, with a concentration of 1 mg/mL diluted in PBS, in 40 µm droplets onto the macroporous silicon chips using an in-house developed chip-based piezoelectric microdispenser (Figure 1) and a computer controlled arraying station. The diameter of single droplets was measured by stroboscopic imaging and the volume of a droplet was calculated to be about 100 pL. In all experiments, microarrays were formed by dispensing one droplet per spot to keep the spot area as small as possible. Fluorescence microscope imaging was performed in order to evaluate the level of bioactive capture antibodies immobilized onto the macroporous silicon. Images were grabbed via a confocal microscope. Mean fluorescence intensities of the spots were measured and compared for different amounts of immobilized capture antibodies and for different concentrations of probe molecule solution. Journal of Proteome Research • Vol. 3, No. 5, 2004 989

research articles

Finnskog et al.

Figure 2. Protocol for the dual PCPA readout as adapted for large molecule (protein - upper) and small molecule (peptide - lower) affinity binding. In the case of large proteins, a step of on-chip protein digestion is performed prior to adding MALDI matrix and the subsequent MS readout. For peptide antigens the antibody was coupled to the porous silicon surface via an IgG spacer, improving MS compatibility of the PCPA surface.

2.5 Fluorescence Readout and Analysis. The fluorescence readout was performed using a confocal microscope setup, Olympus, Fluoview 500. Excitation wavelength (495 nm) was selected to match the fluorophore used; FITC and collected images were analyzed using Fluoview FV 300 ver. 4.3 software from Olympus. Fluorescence data was derived from the average spot intensity. 2.6 Mass Spectrometry Identification. The MALDI-TOF instrument was a MATILDA II (Micromass, Manchester, UK). The instrument, equipped with a delayed extraction ion source, utilized a nitrogen laser at 337 nm and was operated in reflection mode at an accelerating voltage of 20 kV. Analysis was performed directly on the porous silicon target after careful addition of 0.5 µL of R-CHCA matrix consisting of a saturated solution in 50/50 ACN/0.2% TFA in purified water (v/v).

3. Results & Discussion 3.1 Dual Readout Microchip Array Principle. The core idea of the dual readout principle is in the first step to use a fluorescent protein micro array immunoassay to identify spots displaying an affinity interaction with analyte molecules in a sample. The fluorescence readout thus serves as a primary screen, selecting which spots to address in the subsequent the mass spectrometry (MALDI-TOF) step, thereby saving MS-runtime and reducing the time required for bioinformatics database search as non binding assay spots are not evaluated by MS. The dual readout principle, thereby, simultaneously provides both affinity and mass identity information. The protein microchip array concept proposed by our group, named PCPA (pore chip protein array), is based upon the development of a surface morphology tailored at the microand nano-scale level.25 The resulting protein chip surface is composed of a thin layer of macro and nanoporous silicon fabricated in monocrystaline wafers. As porous silicon (PS) is known to possess intrinsic fluorescence properties the process of manufacturing the PS layer was tailored not to display any interfering background fluorescence. At the same time, the surface also had to meet the requirements set by the MALDITOF MS application to encompass the dual read-out concept. The obtained PS surfaces provided an enlargement of the total surface area available per microarray spot and consequently made the binding molecule, i.e., the antibody monolayer coverage highly enlarged. The enlarged density coverage of each individual array spot conceptually allowed higher antigen binding which resulted in an increase in assay spot read-out as monitored by fluorescence imaging. 990

Journal of Proteome Research • Vol. 3, No. 5, 2004

The developed PCPA surfaces that evolved have proven to fulfill the requirements to perform well as both fluorescence based assay detection surfaces with advantageous read-out properties (described below) as well as MALDI target plate surfaces. An important feature of the dual readout protein chip approach is that not only is an affinity binding event identified, giving quantitative information. In the subsequent MS readout, qualitative information is also gained as the actual structure information is revealed, disclosing if a single analyte generated the binding event or if a series of antigen homologues were involved. To adapt to situations of small molecule antigens, peptides, and large molecule antigens, proteins, two protocols were developed for the dual readout PCPA. In the case of large molecule detection, e.g., IgG, anti-IgG was arrayed onto the chip surface and subsequently incubated with FITC-labeled IgG spiked samples. To access the bound antigen by means of MALDI-TOF MS, the bound protein had to be digested on-spot prior to MALDI analysis, as shown in Figure 2 (upper). In the case of a low molecular target analyte as exemplified herein, MALDI matrix was directly added to the PCPA-chip after a positive fluorescence readout, where after the MALDI analysis was performed, Figure 2 (lower). 3.2 Porous Silicon. The new protein chip surface used in this paper was derived as a further development of the PCPA concept earlier presented by Ressine et al.25 By anodizing the microchip surface in a hydrofluoric acid solution a highly differentiated porous silicon surface is obtained that offers a large surface area compiled in a small chip area. A consequence of this is that a high projected density of affinity binder can be immobilized on such a surface, hence, an increased bioassay readout sensitivity can be expected as also shown in ref 25. In this paper, the porous surface was tailored to meet the requirements set by the different assays performed with respect to: •microfluidic properties, i.e., hydrophobic properties to obtain sufficient spot confinement for increased affinity binder density and thus higher bioassay sensitivity read-out. •optical properties, i.e., (I) homogeneous fluorescence intensity spot profiles due to the micro-/nanostructured surface, (II) low background fluorescence •biocompatibility, i.e., (I) amenable to immobilization and/ or adsorption of biospecific binders with maintained affinity and selectivity (II) allowing protein digestion to be performed directly on the surface (III) allow complex sample analysis such as blood.

Integrated Protein Microchip Assay

Figure 3. Characteristics of macroporous silicon and the insert (top) shows the profile of a dispensed water droplet on the surface describing a drying sequence.

•amenable to matrix-assisted laser desorption ionization. These are requirements which indeed are multifaceted and are difficult to fulfill using conventional surfaces. However, by carefully tuning the porous silicon formation process, a nanostructured surface that fairly well stand-up to these demands has been accomplished, Figure 3 and insert lower left, showing the top view and cross-section of the PCPA chip, respectively. An issue that is especially important to raise is the microfluidic property of the surface, which acts hydrophobically (low wetting) on a bulk fluid scale, i.e., microdroplets deposited on the surface are confined with a relatively high contact angle, approximately 110°, which is attributed to the nano- and micromorphology of the PCPA surface. A consequence of this is that the low wetting properties of the porous surface provides smaller spot sizes for a given deposited fluid volume, which inherently gives a higher affinity binder density and also that smaller spot sizes provide improved assay kinetics.26 The confinement of the droplet and the drying sequence of a 5 nL aqueous volume is illustrated in the top insert Figure 3A. Physical adsorption of proteins to porous silicon is a complex reaction, involving a large number of factors such as: pore size, protein size, layer thickness, and protein structure and function.27-29 It was found, for example, that anti-RIgG absorbs well to the PCPA chip and keeps its binding properties for RIgGFITC. This is not the case for anti-Ang II. Either it does not adsorb to the surface or it does not remain its binding to Ang II-FITC, as no assay readout is accomplished when arraying the Ang II antibody directly onto the PCPA surface. Instead, the anti-Ang II molecule requires a preadsorbed anti-RIgG that it binds to, since it was developed in rabbit, and remains active. 3.3 Assays for Peptide and Bio-Macromolecule Antigens. To have a generic way of developing bioassays for both low as well as high molecular weight antigens the surface conditions need to be matched to accommodate for the different protocols that these assays require. The PCPA developed in this paper has been taken to a level where both peptide and macro molecule binding can be performed, still offering the benefits of the proposed dual read-out protocol. 3.3.1 Medical Background to Assay Development. The pathway around the “renin-angiotensin” system plays an important role in regulating blood volume, arterial pressure, and cardiac and vascular function. In disease areas, the pathways for the renin-angiotensin system have been found

research articles in a number of tissues. The most important site for renin release is the kidney.30 Renin is an enzyme that acts upon a circulating substrate, the angiotensinogen, which undergoes proteolytic cleavage to Angiotensin I. Ang II has several very important functions such as vessel constriction (via Ang II receptors) thereby increasing systemic vascular resistance and arterial pressure. Ang II also acts upon the adrenal cortex to release aldosterone, which in turn acts upon the kidneys to increase sodium level and fluid retention, also stimulating thirst centers within the brain. There has been a long standing interest for therapeutic manipulation of this pathway as it has become very important in treating hypertension and heart failure. For example Angiotensin converting enzyme (ACE) inhibitors and AII receptor blockers, can be used to decrease arterial pressure, ventricular afterload, blood volume and hence ventricular preload, as well as inhibit and reverse cardiac and vascular hypertrophy.30,31 3.3.2 Angiotensin Metabolism. Angiotensin, is an octapeptide (DRVYIHPF) originally found to be produced by kidneyderived renin from an R-2 hepatic globulin. Cell activation is driven by ligand binding to cell surface receptors AT1 and AT2. The original angiotensin Angiotensin I gives rise to Angiotensin II by removal of a C-terminal dipeptide. This reaction is mediated by an ACE in plasma, liver, and nerve tissues. Further removal of the aminoterminal amino acid from Angiotensin II by an aspartate amino peptidase yields Angiotensin III. The three forms are correspondingly found present in many clinical samples of disease interest. Angiotensin II is mainly known for its potent pharmacological activities within inflammation processes.32-34 This peptide is involved in the mechanism that drives pulmonary fibrosis,35,36 which in turn drives the tissue collagen turnover in an autocrine manner. These tissue alterations are also observed in heart cardiac myocyte necrosis.37,38 Genes regulated by AT1 receptorinhibition also have a functional effect such as induced vascular smooth muscle cell proliferation.39 To capture the activity of the Renin pathway, we developed two dual read-out assays to match the simplest multiplex configuration. We were interested in having a wide range antibody that mapped both forms, where total read-outs could be made as well as identification of specific immunoreagents that measured only one of the angiotensins in the pathway, namely Ang II. Correspondingly, the difference in absolute measurements within the multiplex assay will be the levels of Ang I. 3.3.3 Gobal Angiotensin Assay. The strategy in the Angiotensin protein chip development was to provide an assay readout that measured Renin pathway activity. The presence of any Angiotensins were detected by arraying an immunoreagent (serum anti Ang I) with broad specificity that has an epitope affinity that captured both Ang I, Ang II, Ang III, and Renin. Utilizing the fluorescence mode of the dual readout a positive binding event confirmed the presence of either Ang I, Ang II, Figure 4, and Ang III or Renin, Figure 5. One benefit of the dual readout is that rapid primary screening of large arrays can be performed in fluorescence mode, identifying positive read-outs, which require a level two analysis, preformed by mass spectrometry. When executed on a larger scale after the selection of only relevant spots (positive fluorescence read-out) for MS-analysis, a considerable amount of time and computer database matching efforts is saved. The advantage of the dual read-out PCPA approach is clearly illustrated in Figure 4 showing FITC-labeled Angiotensin I and II at a Journal of Proteome Research • Vol. 3, No. 5, 2004 991

research articles

Finnskog et al.

Figure 4. Assay for FITC-labeled angiotensin 1 and 2 at a concentration of 70 nM diluted in blood plasma. Angiotesin II has the m/z of 1509 and Angiotensin I 1759.

Figure 6. Fluorescence readout for an Ang II PCPA chip at a concentration of 70 nM (top). The lower figure shows the dynamic response of the assay, i.e., antigen concentration versus fluorescence readout.

Figure 5. Assay for [Ile7]-Angiotensin III and Renin Angiotensinogen 1-14 at concentrations of 70 nM diluted in blood plasma. Angiotensin III has the m/z of 898 and Renin 1762.

concentration of 70 nM in blood plasma. Angiotesin II was identified at m/z of 1508 and Angiotensin I at 1759. The dual read-out approach here shows both affinity binding and mass identity of the bound species. The same immunoassay also demonstrated specificity for [Ile7]-Angiotensin III (m/z 898) and Renin Angiotensinogen 1-14 (m/z ) 1762) at concentrations of 70 nM diluted in blood plasma as seen in Figure 5. 3.3.4 Specific Angiotensin II Assay. A peptide antibody against Angiotensin II was also used with the PCPA. Figure 6 illustrates a fluorescence array image of a PCPA surface where FITC-labeled Ang II has been determined. The amount of antibody consumed was 0.7 fmol/spot and the detection limits for fluorescence read-out was 3.5 nM for Ang II-FITC. When performing the subsequent MALDI-TOF MS readout the exact mass of the antigen bound to the capture antibody was determined. In Figure 7 the MS spectra of Angiotensin II corresponding to m/z 1509 in the assay is seen. The antigen was diluted in blood plasma to a concentration of 70 nM and a total amount of 20 µL was added to the chip giving a total of 1.4 pmol Angiotensin II in the incubation step. As a control experiment mass spectra were also derived from the identical antibody activated PCPA chip which was exposed to plasma without Ang II spiking. No measurable Ang II background was identified. The sensitivity of the fluorescently labeled AngII assay was found not to match the identical assay when performed on non labeled Ang II. It was thus possible to detect much lower 992

Journal of Proteome Research • Vol. 3, No. 5, 2004

Figure 7. MALDI spectra from the PCPA after performing the Angiotensin II assay in plasma samples, Ang II is seen at m/z of 1510.

amounts of unlabeled Ang II on a PCPA chip surface. To demonstrate the detection limit for Ang II an amount of 500 zeptomol was deposited directly on the chip surface and identified by MALDI TOF MS, Figure 8 (Ang II, non FITC labeled is seen at m/z ) 1046). It seems reasonable to assume that the FITC labeled Ang II used in the model assay had less favorable ionization properties than the unlabeled Ang II peptide which thus lowered the detection limit. The lowest detected amount of FITC-labeled Ang II was found at single fmol level. Interestingly, we did not see any peptide fragment signals generated from the Ang II antibody, which is a big advantage of the current dual readout protein chip approach. This is probably one reason for the low

research articles

Integrated Protein Microchip Assay

Figure 8. MALDI spectra of unlabeled Angiotensin II at 500 zeptomol. Unlabeled Ang II has a m/z of 1046 Da.

detection levels observed as in this case no ion suppression effects are induced by undesired antibody peptides. Also, the laser pulse energy level used in combination with the macro-, and nanoporous morphology of the PCPA (Figure 3) makes the adsorbed antibody yet adhere to the chip surface while the affinity captured antigen is successfully ionized. This effect has earlier also been observed when performing assays in nitrocellulose coated nanovials.40 3.3 IgG Assay. We have previously been able to show25 that our PCPA surfaces are highly suitable for fluorescent assays. One reason for the good sensitivity obtained is attributed to the porous 3-D structure combined with the wetting properties of the PCPA surface which confines the arrayed Ab to a small spot, providing a high surface density of affinity binder. In this paper, we have been able to demonstrate that by a further modification of the PCPA surface we are able to generate high contact angles (110°), Figure 3 top insert, which improves the overall sensitivity of the array concept. However, from the perspective of mass spectrometry analysis of antigens, large biomacro-molecule antigens have the disadvantage of being too large to be suitable for MALDI analysis with reasonable sensitivities. This was clear from our studies using IgG where the molecular weight is 140 kDa. To facilitate the identification and increase MS-sensitivity an on-chip digestion step was introduced. Each assay spot was covered with enzyme solution. The digestion was made by adding 20 ng Promega Trypsin diluted in 50 mM NH4HCO3 to the chip and was allowed to incubate for 4 h at 37 °C and humid conditions. The resulting mass spectra in Figure 9 shows four peptides (m/z ) 1737, 2189, 2368, and 2448) derived from the antigen. These identifications were confirmed by matching to peptide fingerprints derived from a pure antigen digest performed on the PCPA. The other peaks derive from either anti-RIgG or from milk used in the blocking step.

4. Conclusions The present work demonstrates the possibility of using porous silicon as a substrate for high sensitive protein chip applications in a dual readout mode. Tailored porous silicon surfaces, PCPA, have shown good properties both for quantitative fluorescence readout of FITC labeled biomolecules as well as for MALDI-TOF MS data generation, enabling structural identification.

Figure 9. Resulting MALDI spectra from a 70 nM IgG assay digested on the PCPA surface. The fragments marked with stars originate from the antigen as matched to peptide fingerprints derived from pure antigen digests, (m/z ) 1737, 2189, 2368, and 2448).

By a combination of fluorescence and MALDI readout, it would be possible to use fluorescence readout to make preselection of relevant spots based on a fluorescent microarray immunoassay that would reduce non informative MALDI analysis and lower the amounts of saved MS data in a second analysis step. The advantages of the PCPA surface for fluorescence readout can be summarized as the following: 1. Reduced spot size due to low wetting properties (enabling high-density arraying). 2. Homogeneous spots that can be read by automated methods, (fluorescence/MALDI). 3. Low fluorescence background. 4. Silicon offers a wide range of possibilities for coupling of biomolecules by physical adsorption or by chemical derivatization. 5. The PCPA surface showed advantageous properties in terms of peptide sensitivities for Ang I with good signal-tonoise ratio. It should be added though that the amount of sample within the matrix crystal at this level only allows a few MALDI laser shots before the sample is lost in the analysis. High-density antibody arrays (>4000 spots/cm2), were dispensed on the PCPA, using an in-house developed piezoelectric microarray station. The dual readout protein microarray is demonstrated on both peptides and proteins. A dedicated protocol for mapping inflammatory peptides belonging to the Renin pathway has been demonstrated in human plasma. Also, an example of large size protein identification was successfully performed on the PCPA by performing an on-target digest after completed fluorescence assay, followed by PMF MALDI-TOF MS. Abbreviations: ACN, acetonitrile; MALDI-TOF, matrix assisted laser desorption ionization - time-of-flight; PBS, phosphate buffered saline 10 mM pH 7.4; IgG, human immunoglobulin G; Ang II, Angiotensine II; R-CHCA, R-cyano-4-hydroxycinnamic acid; TFA, Trifluoroacetic acid; FITC, Flourescein 5(6)isothiocynate; DMF, Dimethylformamide.

Acknowledgment. We would like to thank SWEGENE Foundation for Strategic Research, Swedish Research Council and the Wallenberg foundation for financial support, as well Journal of Proteome Research • Vol. 3, No. 5, 2004 993

research articles as the kind support from Thorleif Lavold, Micromass Europe. The Swedish Science Council is also acknowledged for financial support.

References (1) Miliotis, T.; Ericsson, P.-O.; Marko-Varga, G.; Svensson, R.; Nilsson, J.; Laurell, T.; Bischoff, R. J. Chromatogr. 2001, 752, 323334. (2) Gygi, S. P.; Rist, B.; Griffin, T. J.; Eng, J.; Aebersold, R. J. Proteome Res. 2002, 1, 47-54. (3) Podtelejnikov, A. V.; Mann, M. Methods Enzymol. 2002, 351, 296321. (4) Peng, J.; Elias, J. E.; Thoreen, C. C.; Licklider, L. J.; Gygi, S. P. J. Proteome Res. 2003, 2, 43-50. (5) Leea, Wen-Chien*; Leeb, Kelvin H. Anal. Biochem. 2004, 324, 1-10. (6) Zhu, H.; Snyder, M. Curr. Opin. Chem. Biol. 2001, 5, 40. (7) de Wildt, R. M. T., et al. Nat. Biotechnol. 2000, 18 (9), p 989994. (8) MacBeath, G.; Schreiber, S. L. Science 2000, 289 (5485), p 17601763. (9) Cahill, D. J. J. Immunol. Methods 2001, 250 (1-2), p 81-91. (10) Cahill, D. J. Proteomics: A Trends Guide 2000, 47-51. (11) Albala, J. S.; Humphery-Smith, I. Curr. Opin. Pharm. Proteomics 1999, 680-684. (12) Emili, A. Q.; Cagney, G. Nat. Biotechnol. 2000, 18, 393-397. (13) Zhu, H.; Klemic, J. F.; Chang, S.; Bertone, P.; Casamayor, A.; Klemic, K. G.; Smith, D.; Gerstein, M.; Reed, M. A.; Snyder, M. Nat. Genet. 2000, 26, 283-289. (14) Agaton, C., et al. Mol. Cell. Proteomics 2003, 2, 405-414. (15) Weinberger, S. R.; Morris, T. R.; Pawlak, M. Pharmacogenomics 2000, 1, 395-416. (16) Zhu, H.; Bilgin, M.; Bangham, R.; Hall, D.; Casamayor, A.; Bertone, P.; Lan, N.; Jansen, R.; Bidlingmaier, S.; Houfek, T.; Mitchell, T.; Miller, P.; Dean, R. A.; Gerstein, M.; Snyder, M. Science Vol. 293, Issue 5537, 2101-2105, September 14, 2001. (17) Haab, B. B. Proteomics 2003, 11, 2116-2122. (18) Bouwman, K.; Qui, J.; Zhou, H.; Schotanus, M.; Mangold, L. A.; Vogt, R.; Erlandson, E.; Trenkle, J.; Partin, A. W.; Misek, D.; Omenn, G. S.; Haab, B. B.; Hanash, S. Proteomics 2003, 11, 22002207. (19) Hanash, S.; Celis, J. E. Mol. Cell. Proteomics 2002, 1, 413414. (20) Hanash, S. oral presentation at “Biotech 2003” Washington D. C., 25: th June, 2003.

994

Journal of Proteome Research • Vol. 3, No. 5, 2004

Finnskog et al. (21) Zhu, H.; Bilgin, M.; Bangham, Rh.; Hall, D.; Casamayor, A.; Bertone, P.; Lan, N.; Jansen, R.; Bidlingmaier, S.; Houfek, Th.; Mitchell, T.; Miller, P.; Dean, R. A.; Gerstein, M.; Snyder, M. Science 2001, 293, 2101-2105. (22) Bohr, D. F.; Webb, R. C. Am. J. Med. 1984, 77, 3-10. (23) Kawamura, M.; Terashita, Z.; Okuda, H.; Imura, Y.; Shino, A.; Nakao, M.; Nishikawa, K. J. Pharmacol. Exp. Ther. 1993, 266, 1664-1669. (24) Prescott, M. F.; Webb, R. L.; Reidy, M. A. Am. J. Pathol. 1991, 139, 1291-1296 Y. (25) Ressine, A.; Ekstrom, S.; Marko-Varga, G.; Laurell, T. Anal. Chem. 2003; 75 (24), 6968-6974. (26) Ekins, R. P. Clin. Chem. 1998, 44, 9, 2015-2030. (27) Arwin, H.; Gavutis, M.; Gustavsson, J.; Schultzberg, M.; Zangooie, S.; Tengvall, P. Phys. Stat. Sol. (a) 2000, 182, 515. (28) Tinsley-Bown, A. M.; Canham, L. T.; Hollings, M.; Anderson, M. H.; Reeves, C. L.; Cox, T. I.; Nicklin, S.; Squirrel, D.J.; Perkins, E.; Hutchinson, A. M. J. Sailor A. Wun, Phys. Stat. Sol. (a) 2000, 182, 547. (29) Karlsson, L. M.; Tengvall, P.; Lundstro¨m, I.; Arwin, H. Phys. Stat. Sol. (a) 2003, 197, 2, 326-330. (30) Hansson, L.; Lindholm, L. H.; Niskanen, L et al. Lancet 1999, 353, 611-116. (31) Zanchetti, A.; Omboni, S.; Biagio, Di. J. Human Hypertens. 1997, 11, 57-59. (32) Bohr, D. F.; Webb, R. C. Am. J. Med. 1984, 77, 3-10. (33) Kawamura, M.; Terashita, Z.; Okuda, H.; Imura, Y.; Shio, A.; Nakao, M.; Nishikawa, K. J. Pharmacol. Exp. Ther. 1993, 266, 1664-1669. (34) Prescott, M. F.; Webb, R. L.; Reidy, M. A. Am. J. Pathol. 1991, 139, 1291-1296. (35) Sun, J. Felix, Ramires, A.; Zhou, G.; Venkataseshu, K.; Ganjaman, K.; Weber, K. T. J. Mol. Cell. Cardiol. 1997, 29, 2001-2012. (36) Sun, Y.; Weber, K. T. J. Mol. Cell Cardiol. 1996, 28, 851-856. (37) Sun, Y.; Weber, K. T. J. Mol. Cell Cardiol. 1996, 28, 851-856. (38) Smits, J. M. F.; van Krimpen, C.; Schemaker, C. G.; Cleutjens, J. P. M. Daemen, J. P. M. J. Cardiovasc. Pharmacol. 1992, 20, 772778. (39) Viswnathan, M.; Stromberg, C.; Seltzer, A.; Saavedra, J. M. J. Clin. Invest. 1992, 90, 1707-1712. (40) Borrebaeck, C. A.; Ekstrom, S.; Hager, A. C.; Nilsson, J.; Laurell, T.; Marko-Varga, G. Biotechniques. 2001, 30 (5), 1126-1130, 1132.

PR0499287