Serum PEDF Levels Are Decreased in a Spontaneous Animal Model

Dec 29, 2008 - also in the periphery of the eye proves PEDF as a promising uveitis ... enter the inner eye leading to uveitis.8,12 Experiments reveale...
0 downloads 0 Views 7MB Size
Download PDF
Serum PEDF Levels Are Decreased in a Spontaneous Animal Model for Human Autoimmune Uveitis Johanna K. Zipplies,† Stefanie M. Hauck,‡ Stephanie Schoeffmann,‡ Barbara Amann,† Manfred Stangassinger,† Marius Ueffing,‡,§,# and Cornelia A. Deeg*,†,# Institute of Animal Physiology, Department of Veterinary Sciences, LMU Munich, Veterina¨rstr. 13, D-80539 Munich, Germany, Department of Protein Sciences, Helmholtz Zentrum Mu ¨ nchen, German Research Center for Environmental Health (GmbH), Ingolstaedter Landstr. 1, D-85764 Neuherberg, Germany, and Institute of Human Genetics, Technical University of Munich, Trogerstr. 32, D-81675 Munich, Germany Received September 1, 2008

Identification of biomarkers is of critical relevance toward improving diagnosis and therapy of autoimmune disorders. Serum markers are a desirable choice as sera are easily accessible and the development of assays for routine clinical detection prompts feasible. Autoimmune uveitis, a recurrent disease affecting the eye, is characterized by returning inflammatory attacks of the inner eye followed by variable periods of quiescent stages. Spontaneous equine recurrent uveitis (ERU) is the equine equivalent and serves as a model for the human disease. To identify potential biomarker candidates, we first systematically compared the proteomes of individual ERU cases with healthy controls by proteomic profiling using 2-D difference-gel-electrophoresis (2-D DIGE) followed by tandem mass spectrometry. A total of seven differentially expressed proteins were identified. Besides the upregulation of IgG and the significant lower expression of albumin, Antithrombin III, and Vitamin D binding protein, we found complement components C1q and C4, to be downregulated in uveitic state. Interestingly, Pigment epithelium-derived factor (PEDF), a marker already detected by 2DE differential proteome analysis in ERU target tissues, vitreous and retina, was found to be also significantly downregulated in sera. The lower expression of PEDF in sera of horses with uveitis could be verified in a cohort of 116 ERU cases and 115 healthy controls. Our findings of a significant lower PEDF expression in ERU cases also in the periphery of the eye proves PEDF as a promising uveitis biomarker. Keywords: Pigment epithelium derived factor (PEDF) • serum • biomarker • Fluorescence twodimensional difference gel electrophoresis • uveitis

1. Introduction Equine recurrent uveitis (ERU) is a highly prevalent spontaneous autoimmune disease in horses with a characteristic remitting-relapsing nature.1 ERU serves as a valuable model for human autoimmune uveitis2-5 and other intermittent autoimmune diseases. Autoreactive T cells enter the inner eye in alternating intervals and cause intraocular inflammation with destruction of retinal architecture resulting in blindness.1,6,7 Molecular targets of these attacks are retina- and brain-specific proteins such as S-antigen (S-Ag), interphotoreceptor retinoid binding protein (IRBP) or cellular retinoid binding protein (CRALBP).1,8-10 Interestingly, only activated T cells are able to pass the blood-retinal barrier11 but the triggers of this activation are yet a puzzle in organ-specific autoimmune diseases. This is especially the case as the target organs of the inflam* To whom correspondence should be addressed. Institute of Animal Physiology, Veterina¨rstr. 13, D-80539 Munich. E-mail: [email protected]. Telephone: 00498921801630. Fax: 00498921802554. † Institute of Animal Physiology, LMU Munich. ‡ German Research Center for Environmental Health (GmbH). § Technical University of Munich. # Both authors share equal senior authorship.

992 Journal of Proteome Research 2009, 8, 992–998 Published on Web 12/29/2008

matory attack are separated from the immune system by the blood-retinal or the blood-brain barrier. Injection of uveitis autoantigens (emulsified in complete Freunds Adjuvans (CFA) to ensure a T helper 1 response) in the neck of experimental horses leads to a considerable immune reaction to these autoantigens that is measurable in peripheral blood several days after injection. T cells then overcome the physiological barrier for immune cells, the retinal pigment epithelium, and enter the inner eye leading to uveitis.8,12 Experiments revealed that autoreactive T cells are activated in the spleen13 and therefore home from spleen to the inner eye directly before the inflammatory attack. As a result, these cells can be detected in the peripheral blood straight before disease outbreak in experimental uveitis with known kinetics12,14 and by chance in patients sampled immediately before recurrence of spondyloarthritis.15 This phenomenon presents a rationale toward searching for biomarkers as well as specific biomarker signatures in the peripheral blood, which could indicate an approaching relapse and allow monitoring of autoaggressive activity. Such prediction of individual risks would probably allow earlier and personalized therapeutic intervention during onset and progression of disease. Our preceding studies 10.1021/pr800694y CCC: $40.75

 2009 American Chemical Society

Serum Biomarkers in Uveitis comparing the uveitis target tissue proteomes already proved the value of proteomics in unraveling differentially expressed proteins associated with uveitis pathogenesis.2,4 As the composition of the serum proteome has been shown to mirror most biological processes, as most proteins and their fragments are carried into the peripheral blood, protein profiles reflecting discrete physiological as well as pathological processes are to be expected in serum, though at a much diluted level. Serum or plasma proteomes, though rather complex in composition, are promising specimens for biomarker analysis because of their accessibility and easy storage for up to several years, allowing also retrospective validation of novel detected disease predictive markers. The goal of our study has been the discovery of biomarker candidates for uveitis. After thorough validation, such markers can add to and improve existing paraclinical tools, most importantly surrogate markers such as autoantibody tests in patient sera. Given an eventual prognostic value, they can as well be suited to indicate an approaching attack. In addition, they bear the chance to point toward discrete pathways involved in uveitis pathogenesis. Hence, we chose systematic exploration and comparison of the serum proteomes of healthy and uveitic cases using the advantage of multiplex analysis. Differential-in-gel-analysis (DIGE) overcomes some major drawbacks of conventional 2-D gel electrophoresis, such as the gel-to-gel variation observed with the separation profile even between the same samples.16 Inclusion of an internal standard17 and running different samples on the same gel using quantitative staining with CyDyes is particularly useful to identify true differentially regulated candidates.18

2. Experimental Procedures 2.1. Serum Samples. For the initial DIGE screening experiment comparing the serum proteomes of healthy horses and ERU cases, we processed sera of five healthy horses and five ERU cases. For validation of PEDF as a biomarker, we used 116 additional ERU cases and 115 controls. ERU was diagnosed according to clinical criteria as described.19 2.2. Two-Dimensional Gel Electrophoresis (2DE). Serum samples were stabilized with protease inhibitors (Roche) and stored at -80 °C. Protein content was quantified with the Bradford assay (BioRad). For the DIGE experiment, proteins were labeled with ester derivatives of Cy3 and Cy5 after adjusting pH to 8.5 according to the supplier’s instructions (GE Healthcare) in 2DE DIGE buffer pH 8.5 (30 mM Tris-HCl, 7 M urea, 2 M thiourea, 4% CHAPS). Additionally, an internal standard (a pool of all the samples within an experiment) was included in all experiments. This standard was labeled with ester derivative of Cy2 (GE Healthcare). Proteins were labeled at a ratio of 400 pmol Cy2, Cy3 or Cy5 per 50 µg of protein (minimal labeling) in dimethylformamide (Sigma) for 30 min on ice. The reaction was quenched by addition of 1 µL of 10 mM lysine per sample, and after additional 10 min incubation, samples were combined to sample sets each containing ERU, healthy and internal standard proteins. The total volume of each pool was adjusted to 460 µL with DIGE buffer and dithiothreitole was added at a final concentration of 35 mM. Immobiline dry strips pH 3-11 NL, 24 cm (GE-Healthcare) were rehydrated overnight with the respective sample pools and additional 1% Pharmalyte pH 3-10 (GE-Healthcare) and 0.5% bromphenole blue. Isoelectric focusing was performed on a Multiphor (GE-Healthcare) for 45 kVh at 20 °C, followed by separation on SDS-PAGE gels (12%)

research articles at constant 45 V per gel. Gels were then scanned with the Typhoon Trio Scanner (GE-Healthcare) with 8 bit/600 dpi resolution using different wavelengths for the respective CyDyes. Afterward, gels were silver stained20 or blue silver stained21 and differentially expressed spots were cut and processed for mass spectrometry. 2.3. Image Analysis and Detection of Differentially Expressed Proteins. Images of DIGE gels were imported to 2DE analysis software (DeCyder; release 6.5; GE Healthcare) after scanning. Gel analysis started with differential in-gel analysis (DIA) for intragel analysis.22 The DIA module starts with a spot detection algorithm for each gel (estimated number of spots: 5000). Then the dye tag is assigned to the images (Cy2-internal standard, Cy3 or Cy5-ERU or control, since reverse labeling was performed). After spot detection, a normalization of Cy3 and Cy5 volumes in relation to the Cy2 labeled internal standard (Cy3/ Cy2 and Cy5/Cy2) was performed. The following inclusion and exclusion criteria for spot detection were defined in order to exclude false positive spots: spot slope 2, minimum spot volume 30 000 and threshold of 2.5. Spots were then detected by the software, but detection was manually checked and corrected if necessary. For each gel, the number and distribution of spots according to their volume ratio are displayed in a histogram. The histogram indicates the type of the detected spot: decreased/unchanged/increased. Next, gels were analyzed with the biological variation analysis (BVA) allowing intergel analysis.22 Standardized abundances of all matched protein spots were compared across the five analytical gels per experiment, and Student’s t test was performed with the DeCyder BVA (biological variation analysis) module to validate the significance of the detected differences between spot volumes from uveitic sera and those of negative control sera (t test P-values set at e0.05). Respective spots were displayed on a graphical map which was then printed on a transparency in order to redetect these spots on the blue silver or silver stained gels to excise for mass spectrometry. 2.4. Mass Spectrometry for Protein Identification. Selected spots from 2DE were excised, destained and processed by proteolysis with trypsin as described before23,24 and analyzed by MALDI-TOF peptide mass fingerprinting and MS/MS on a MALDI-TOF/TOF tandem mass spectrometer (ABI 4700 Proteomics Analyzer, Applied Biosystems). For positive-ion reflector mode spectra, 2500 laser shots were averaged and processed with external calibration. PMF spectra were not smoothed and background was not subtracted. Monoisotopic peak masses were automatically determined within the mass range 800-4000 kDa with a signal-to-noise ratio minimum set to 5 and the local noise window width m/z 200. Up to seven of the most intense ion signals with signal-to-noise ratio above 30 were selected as precursors for MS/MS acquisition excluding common trypsin autolysis peaks and matrix ion signals. In MS/MS positive ion mode, 4000 spectra were averaged with 1 kV collision energy, collision gas air at a pressure of 1.6 × 10-6 Torr and default calibration. Monoisotopic peak masses were automatically determined with a signal-to-noise ratio minimum set to 10 and the local noise window width m/z 200. Combined PMF and MS/MS queries were performed using the MASCOT Database search engine v1.925 (Matrix Science Ltd.) embedded into GPS-Explorer Software (Applied Biosystems) on the Swiss-Prot database (version 20070531; 270 778 sequences; 99 412 397 residues), the MSDB metadatabase (version 20061115; 3 239 079 sequences; 1 079 594 700 residues) or the horse genome database (retrieved from UCSC, version 20070101; 39 612 sequences; 252 365 residues) with the Journal of Proteome Research • Vol. 8, No. 2, 2009 993

research articles

Zipplies et al. a

Table 1. Differentially Expressed Proteins in Sera of ERU Cases

theoretical spot ID

protein name

129

Immunoglobulin Gamma 7 hc Immunoglobulin Gamma 4 hc Immunoglobulin Gamma 4 hc Vitamin D binding Protein Antithrombin III Antithrombin III Pigment epithelium-derived factor Albumin Albumin Immunoglobulin Gamma 7 hc Albumin Albumin Complement factor I Complement C4 gamma chain IGLV3-25

130 131 132 133 134 135 136 137 138 139 140 141 142 143

species

accession number

MW (Da)

pI

protein score

regulation factors

expression pattern In ERU

Equus caballus

AAS18414

35721

7.69

59

1.28

v

Equus caballus

AAS18415

36211

7.71

70

1.52

v

Equus caballus

AAS18415

36211

7.71

66

1.64

v

Equus caballus

XP_001489400

54327

5.46

278

1.36

V

Pongo pygmaeus Bos taurus Homo sapiens

ANT3_PONPY A61435 PEDF_HUMAN

53061 49437 46342

6.32 6.02 5.97

335 168 142

1.46 1.45 1.22

V V V

Equus caballus Equus caballus Equus caballus

ALBU_HORSE ALBU_HORSE AAS18415

70550 70550 36211

5.95 5.95 7.71

103 198 91

1.25 1.29 1.45

V V v

Equus caballus Equus caballus Equus caballus

ALBU_HORSE ALBU_HORSE XP_001502956

70550 70550 66104

5.95 5.95 7.33

67 151 137

1.22 1.32 1.37

V V V

Equus caballus

XP_001492943

32856

6.68

95

1.54

V

Equus caballus

XP_001492872

29005

8.54

287

1.28

v

a Mass spectrometric identifications of differentially regulated proteins in ERU sera. Proteins listed have been identified with a probability score that is significant with p < 0.05. The expression pattern of the proteins in recurrent uveitis is indicated by arrows (v upregulated in ERU, V downregulated in ERU).

following parameter settings (if applicable): entries restricted to mammalian, 65 ppm mass accuracy for the precursor mass, 0.3 Da fragment mass tolerance, trypsin cleavage, one missed cleavage allowed, carbamidomethylation was set as fixed modification and oxidation of methionines was allowed as variable modification. A protein was regarded as identified (Table 1), if the probability based MOWSE score26 was significant for the respective database (protein scores greater than 58 were significant, p < 0.05 for Swiss-Prot, scores greater 67 for MSDB and scores greater 54 for the horse genome database), if the matched peptide masses were abundant in the spectrum and if the theoretical masses of the significant hit fit the experimentally observed values. 2.5. Quantification of PEDF Downregulation with ELISA. For validation of PEDF downregulation in uveitic state, we used a sandwich ELISA. PolySorb plates (Nunc) were first coated with 25 µL/mL Steptavidin (Roche) in carbonate buffer (pH 9.6). The capture antibody, a monoclonal mouse-anti-PEDF antibody (5 µg/mL, Chemicon) was preincubated.with anti-mouse IgG Biotin (1:1000, Linaris) for 1 h and then applied to the ELISA plate for 1 h. After blocking with 4% BSA (Applichem), sera were added in a 1:10 dilution in PBS-T. A polyclonal goat anti-PEDF antibody (1:5000, Santa Cruz) was used as detection antibody, followed by an anti-goat IgG POD labeled antibody (1:20 000, Linaris). The reaction was visualized with tetramethylbenzidine, stopped with 1 M sulfuric acid and the absorbance was measured at 450 nm using a microplate reader. As a positive control, a dilution series of recombinant human PEDF (Acris) was performed on every plate. PEDF abundances between ERU cases and controls were statistically analyzed using the Mann-Whitney test and the free software package Past (http:// folk.uio.no/ohammer/past/).

3. Results 3.1. Proteomic Map of Normal Equine Serum. We first obtained a map of normal equine serum proteins (Figure 1; blue silver stain) by 2 DE separation. Serum 2DE patterns were highly 994

Journal of Proteome Research • Vol. 8, No. 2, 2009

Figure 1. Representative blue silver-stained 2DE of healthy equine serum comprising pH gradient from 3 to 11. The proteome was resolved by 2-dimensional gel electrophoresis. Protein spot numbers refer to mass spectrometric identifications, as given in supplemental table.

reproducible between experiments and among sampled cases. Major landmark proteins could be subsequently and conclusively identified by mass spectrometry (see supplemental table in Supporting Information). A total of 128 spots were successfully identified by mass spectrometry resulting in the identification of 33 different proteins (supplemental table). 3.2. Seven Differentially Expressed Proteins Identified in Sera of ERU Cases. The DIGE experiment comparing controls (n ) 5) (Figure 2a, blue spots) and uveitic sera (n ) 5; Figure 2a, green spots) revealed several differentially regulated spots between both states (Figure 2c, differentially expressed spots labeled and Figure 3, displaying individual expression patterns).

research articles

Serum Biomarkers in Uveitis

Figure 2. Differentially regulated proteins in complete sera of cases with spontaneous uveitis. (a) Overlay image of fluorescently labeled 2DE with control serum (blue; labeled with Cy3, scanned at 532 nm) and serum of ERU case (green; labeled with Cy5, scanned at 633 nm) comprising pH gradient from 3 to 11. Turquoise stained spots were equally expressed in both states (control and ERU), whereas blue spots were higher expressed in control sera and green spots were higher expressed in uveitic state; (b) Silver-staining of the same 2DE gel; (c) survey map created with DeCyder BVA software indicating differentially regulated candidates (red; Student’s t-test p e 0.05); mass spectrometrically identified spots are numbered as in Table 1. Yellow labeled spots were higher abundant in ERU; green labeled spots were lower abundant in uveitic condition. Spot numbers refer to mass spectrometric identifications given in Table 1.

Seven proteins (Figure 3) with different abundance were unambiguously identified by mass spectrometry (Table 1, identifications with MALDI-TOF/TOF of spots referring to Figure 2c). One protein, Immunoglobulin G (IgG heavy chains 4 and 7 and IG light chain V3-25; corresponding to spots 129, 130, 131, 138 and 143 of Figure 2c and Table 1) was upregulated in serum of ERU cases compared to healthy state. In contrast, six proteins were downregulated in ERU sera. These were several albumin fragments (corresponding to spots 136, 137, 139 and 140), Vitamin D binding protein (corresponding to spot 132), Antithrombin III (corresponding to spots 133 and 134), Pigment epithelium-derived factor (PEDF, corresponding to spot 135), Complement factor 1q (corresponding to spot 141) and Complement C4 heavy chain (corresponding to spot 142). Additional proteins differed significantly in their abundance, but could not be identified by mass spectrometry (Red encircled spots in Figure 2c). 3.3. Validation of PEDF-Downregulation in Serum in Uveitic State. Since PEDF was already identified as a downregulated marker in uveitis target tissue,4 PEDF seemed the most promising ERU biomarker candidate in our opinion. Therefore, we examined PEDF expression profiles with a second technique, a PEDF-specific sandwich ELISA. We compared PEDF levels in a cohort of 116 ERU cases (Figure 4, represented by the light gray box) and 115 healthy sera (dark gray box). PEDF levels in horses with uveitis (median 0.08) was significantly lower than in healthy controls (median 0.11), as revealed by statistical analysis (Mann-Whitney test, p < 0.002).

4. Discussion We used 2D-DIGE for detection of quantitative differences between the serum proteomes of normal and uveitic condition and identified seven differentially regulated proteins between both states. Immunoglobulin G expression showed that two isotypes varied in the expression between control and uveitis sera. ERU cases had higher abundant IgG 4 and 7 isotypes (Table 1, Figures 2 and 3). The principle of different Ig classes (IgM, IgG, IgA, IgE, and IgD) sharing individual effector functions during the immune response is generally conserved between mammalian species.27 IgG isotypes can indicate a Th1 or Th2 response to an antigen. Although it is not exactly known which isotypes are associated with autoimmune reactions in horses, IgG 4 antibodies against S-antigen were developed after inducing a strong Th1 response to this autoantigen in experimental horses.14 IgG 4 and IgG 7 share high sequence homologies at both the nucleotide and amino acid sequence level, indicating that they duplicated most recently during evolution of IgG genes of the horse.27 An intraocular expression of IgG 4 was also demonstrated in target tissue retina itself.4 Another differentially expressed protein in ERU is serum albumin corresponding to spots 136, 137, 139 and 140 (Table 1; Figures 2 and 3) in sera. However, decreased expression was not observed for full-length albumin, but for several fragments with lower molecular masses. The function of these fragments is unknown and so the meaning of their lowered expression remains unclear. Decreased abundance of these albumin fragments may be indicative for an interesting decreased proteolytic activity in ERU serum. Since the main function of Journal of Proteome Research • Vol. 8, No. 2, 2009 995

research articles

Zipplies et al.

Figure 3. Differentially regulated candidates in sera of spontaneous uveitis cases according to DeCyder analysis. Each panel includes (a) enlarged two-dimensional picture of respective spots corresponding to the survey map in Figure 2c in ERU, spot number corresponds to the mass spectrometry Table 1, (b) corresponding spots in healthy control serum, (c) 3D graph of spot volume in ERU, (d) 3D graph of spot volume in healthy control serum. Graphs on the right side of each panel show the difference in abundance for spot pairs from all gels. The red dots correspond to the spot pair shown in a-d. The yellow diamonds represent the internal standard.

albumin is transportation of molecules, it would be interesting to study the binding specifity of the downregulated fragments. A lower expressed candidate in uveitic state compared to negative controls is antithrombin III (Table 1, Figures 2 and 3). Antithrombin III is the most important serine protease inhibitor in plasma that regulates the blood coagulation cascade. It inhibits thrombin as well as factors IXa, Xa and Xia.28 The meaning of the changed antithrombin III expression in sera of uveitis cases is unclear. A difference in serum generation between both groups cannot be excluded, but sera were sampled using the same procedure and processing specimen of both states at the same time. Further experiments are needed to detect the role of antithrombin III in this context, since this is a protein with many functions involved in several different cascades.29 Vitamin D binding protein (DBP), an extracellular multifunctional protein with vitamin D transporter activity, was found to be downregulated in ERU (Table 1, Figures 2 and 3). Only 5% of circulating DBP is actually complexed with vitamin 996

Journal of Proteome Research • Vol. 8, No. 2, 2009

D metabolites, leaving a considerable amount of the protein available for functions in macrophage activation,30,31 T cell modulation,31 chemotaxis32 or its role as part of the actinscavenger system.32 Further, Vitamin D shifts the immune response to a more anti-inflammatory response and enhances regulatory T cell functionality.31 There is increasing evidence for a role of the vitamin D endocrine system in the pathogenesis of autoimmune diseases.31,33 The exact mechanism is unclear at present and could be due to the strong immune modulating potential of vitamin D or the role of DBP in clearing tissueleaking proteins from the blood-stream. Therefore, the significant lowered expression of DBP in ERU is an interesting finding that needs further analysis in disease pathogenesis. Complement factor I was lower expressed in all ERU sera tested (Table 1, Figures 2 and 3). This factor inactivates complement subcomponents C3b, iC3b and C4b by proteolytic cleavage.34 Additionally, Complement C4 gamma chain was also lower expressed in uveitic state (Table 1, Figures 2 and 3). Deficiencies of C1q and C4 in humans34 and guinea pigs35 were

research articles

Serum Biomarkers in Uveitis

arkers shows more potential if a combined approach is used, evaluating differentially regulated candidates in diseased tissue and serum. Only PEDF overlapped also in experiments using target tissues retina and vitreous2,4 as a biomarker. The possible predictive value of PEDF downregulation for ERU for clinical routine now merits closer examination. Abbreviations: Apo, apolipoprotein; AMD, age-related macular degeneration; CFA, complete Freund’s adjuvans; CRALBP, cellular retinoid binding protein; DIGE, difference-gel-electrophoresis; DBP, Vitamin D-binding protein; ELISA, Enzyme Linked Immunosorbent Assay; ERU, equine recurrent uveitis; IgG, Immunoglobulin G; IGLV3-25, Immunoglobulin lambda variable 3-25; IRBP, interphotoreceptor retinoid binding protein; PEDF, pigment epithelium derived factor; RA, rheumatoid arthritis; S-Ag, S-antigen. Figure 4. Difference of PEDF levels in sera of 116 horses with uveitis and 115 healty controls was validated with a sandwich ELISA. PEDF levels in sera of horses with uveitis (light gray box) was significantly (**p < 0.002) lower compared to the healthy controls (dark gray box). For statistical analysis, Mann-Whitney test was used (free software package Past). The light gray box represents ERU cases, while the dark gray box represents healthy controls; results are given as adsorbance units at 450 nm (OD 450 nm). The spacing between the boxes indicates the degree of spread in the data. The black line in the boxes represents the median (ERU 0.11; controls 0.08).

Acknowledgment. This work was supported by the Deutsche Forschungsgemeinschaft (DFG) SFB 571 A5 and DE 719/2-1, by EU grant EVI-GENORET LSHG-CT-2005-512036 and by funding from the German Federal Ministry of Education and Research: BMBF-QuantPro 0316865A. The authors thank Kay Junghanns, Erich Grosskopf and Sven Reese for critical and helpful discussions.

found to predispose for autoimmune conditions, indicating that the complement system may moreover be involved in the induction and/or maintenance of tolerance at the humoral level. Notably, individuals, who are deficient in C3 have a low level of susceptibility to systemic lupus erythematosus SLE compared with individuals who are deficient in C1 or C4, which are major susceptibility factors.34 A correlation exists between deficiency in C1q or C4 and impaired clearance of apoptotic debris that could result in accumulation of self-antigens in sites such as the lymphoid compartment, where they could become immunogenic.34 A further known role of C4 in the humoral immune response is the binding of environmental antigens and, through its interaction with CR1, the retention of these antigens on follicular dendritic cells.34 Most interesting, we observed a significant downregulation of pigment epithelium-derived protein (PEDF) in this experiment, consistent with our earlier observations in ERU vitreous2 and retina.4 Therefore, we chose to validate PEDF as the most promising biomarker candidate in our opinion, suggesting that the downregulation of PEDF under inflammatory conditions is not only limited to the eye. With a large-scale validation approach using a second technique with a quantitative PEDFElisa, we could here confirm the significant lower PEDF expression in peripheral blood of ERU cases (Figure 4). PEDF is a neurotrophic factor and a potent inhibitor of angiogenesis.36 It was shown that the PEDF concentration found in blood is sufficient to have functional significance.37 In rats with endotoxin-induced uveitis, drastically decreased PEDF levels are detectable in retina and plasma, which suggests that PEDF is a negative acute-phase protein.38 PEDF operates as a regulator of inflammatory factors and suppresses endothelial permeability by protecting tight junction proteins.38 Our findings of a significant lower PEDF expression in ERU cases not only in the target organ, the eye,2,4 but also in the serum proves PEDF as a promising uveitis biomarker. Further, we demonstrated that the identification of disease specific serum biom-

References

Supporting Information Available: Table of identified proteins in normal equine serum. This material is available free of charge via the Internet at http://pubs.acs.org.

(1) Deeg, C. A.; Amann, B.; Raith, A. J.; Kaspers, B. Inter- and intramolecular epitope spreading in equine recurrent uveitis. Invest. Ophthalmol. Visual Sci. 2006, 47 (2), 652–6. (2) Deeg, C. A.; Altmann, F.; Hauck, S. M.; Schoeffmann, S.; Amann, B.; Stangassinger, M.; Ueffing, M. Downregulation of PEDF in uveitic lesions associates with focal VEGF expression and breakdown of the blood retinal barrier. Proteomics 2007, 7, 1540–1548. (3) Deeg, C. A.; Hauck, S. M.; Amann, B.; Kremmer, E.; Stangassinger, M.; Ueffing, M. Major retinal autoantigens remain stably expressed during all stages of spontaneous uveitis. Mol. Immunol. 2007, 44, 3291–3296. (4) Hauck, S. M.; Schoeffmann, S.; Amann, B.; Stangassinger, M.; Gerhards, H.; Ueffing, M.; Deeg, C. A. Retinal Mueller Glial Cells Trigger the Hallmark Inflammatory Process in Autoimmune Uveitis. J. Proteome Res. 2007, 6, 2121–2131. (5) Deeg, C. A.; Raith, A. J.; Amann, B.; Crabb, J. W.; Thurau, S. R.; Hauck, S. M.; Ueffing, M.; Wildner, G.; Stangassinger, M. CRALBP is a highly prevalent autoantigen for human autoimmune uveitis. Clin. Dev. Immunol. 2007, 39245. (6) Singh, V. K.; Rai, G.; Agarwal, S. S. Role of cytokines in experimental and clinical uveitis. Indian J. Ophthalmol. 2001, 49 (2), 81–90. (7) Nussenblatt, R. B. Bench to bedside: new approaches to the immunotherapy of uveitic disease. Int. Rev. Immunol. 2002, 21 (2-3), 273–289. (8) Deeg, C. A.; Pompetzki, D.; Raith, A. J.; Hauck, S. M.; Amann, B.; Suppmann, S.; Goebel, T. W.; Olazabal, U.; Gerhards, H.; Reese, S.; Stangassinger, M.; Kaspers, B.; Ueffing, M. Identification and functional validation of novel autoantigens in equine uveitis. Mol. Cell. Proteomics 2006, 5 (8), 1462–1470. (9) Caspi, R. R.; Roberge, F. G.; Chan, C. C.; Wiggert, B.; Chader, G. J.; Rozenszajn, L. A.; Lando, Z.; Nussenblatt, R. B. A new model of autoimmune disease. Experimental autoimmune uveoretinitis induced in mice with two different retinal antigens. J. Immunol. 1988, 140 (5), 1490–1495. (10) de Smet, M. D.; Bitar, G.; Mainigi, S.; Nussenblatt, R. B. Human S-antigen determinant recognition in uveitis. Invest. Ophthalmol. Visual Sci. 2001, 42 (13), 3233–3238. (11) Caspi, R. R. Immune mechanisms in uveitis. Springer Semin. Immunopathol. 1999, 21 (2), 113–124. (12) Deeg, C. A.; Thurau, S. R.; Gerhards, H.; Ehrenhofer, M.; Wildner, G.; Kaspers, B. Uveitis in horses induced by interphotoreceptor retinoid-binding protein is similar to the spontaneous disease. Eur. J. Immunol. 2002, 32 (9), 2598–2606. (13) Streilein, J. W.; Okamoto, S.; Hara, Y.; Kosiewicz, M.; Ksander, B. Blood-borne signals that induce anterior chamber-associated

Journal of Proteome Research • Vol. 8, No. 2, 2009 997

research articles (14) (15)

(16) (17)

(18) (19)

(20) (21) (22) (23)

(24)

(25)

998

immune deviation after intracameral injection of antigen. Invest. Ophthalmol. Visual Sci. 1997, 38 (11), 2245–2254. Deeg, C. A.; Reese, S.; Gerhards, H.; Wildner, G.; Kaspers, B. The uveitogenic potential of retinal S-antigen in horses. Invest. Ophthalmol. Visual Sci. 2004, 45 (7), 2286–2292. Marker-Hermann, E.; Meyer zum Buschenfelde, K. H.; Wildner, G. HLA-B27-derived peptides as autoantigens for T lymphocytes in ankylosing spondylitis. Arthritis Rheum. 1997, 40 (11), 2047– 2054. Unlu, M.; Morgan, M. E.; Minden, J. S. Difference gel electrophoresis: a single gel method for detecting changes in protein extracts. Electrophoresis 1997, 18 (11), 2071–2077. Alban, A.; David, S. O.; Bjorkesten, L.; Andersson, C.; Sloge, E.; Lewis, S.; Currie, I. A novel experimental design for comparative two-dimensional gel analysis: two-dimensional difference gel electrophoresis incorporating a pooled internal standard. Proteomics 2003, 3 (1), 36–44. Marouga, R.; David, S.; Hawkins, E. The development of the DIGE system: 2D fluorescence difference gel analysis technology. Anal. Bioanal. Chem. 2005, 382 (3), 669–678. Deeg, C. A.; Kaspers, B.; Gerhards, H.; Thurau, S. R.; Wollanke, B.; Wildner, G. Immune responses to retinal autoantigens and peptides in equine recurrent uveitis. Invest. Ophthalmol. Visual Sci. 2001, 42 (2), 393–398. Blum, H.; Beier, H.; Gross, H. J. Improved silver staining of plant proteins, RNA and DNA in polyacrylamide gels. Electrophoresis 1987, 8, 93–99. Braun, R. J.; Kinkl, N.; Zischka, H.; Ueffing, M. 16-BAC/SDS-PAGE analysis of membrane proteins of yeast mitochondria purified by free flow electrophoresis. Methods Mol. Biol. 2008, in press. Tannu, N. S.; Hemby, S. E. Two-dimensional fluorescence difference gel electrophoresis for comparative proteomics profiling. Nat. Protoc. 2006, 1 (4), 1732–1742. Hauck, S. M.; Ekstrom, P. A. R.; Ahuja-Jensen, P.; Suppmann, S.; Paquet-Durand, F.; van Veen, T.; Ueffing, M. Differential modification of phosducin protein in degenerating rd1 retina is associated with constitutively active Ca2+/calmodulin kinase II in rod outer segments. Mol. Cell. Proteomics 2006, 5 (2), 324–336. Hauck, S. M.; Schoeffmann, S.; Deeg, C. A.; Gloeckner, C. J.; Swiatek-de Lange, M.; Ueffing, M. Proteomic analysis of the porcine interphotoreceptor matrix. Proteomics 2005, 5 (14), 3623– 3636. Perkins, D. N.; Pappin, D. J.; Creasy, D. M.; Cottrell, J. S. Probabilitybased protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 1999, 20 (18), 3551–3567.

Journal of Proteome Research • Vol. 8, No. 2, 2009

Zipplies et al. (26) Pappin, D. J. Peptide mass fingerprinting using MALDI-TOF mass spectrometry. Methods Mol. Biol. 1997, 64, 165–173. (27) Wagner, B.; Miller, D. C.; Lear, T. L.; Antczak, D. F. The complete map of the Ig heavy chain constant gene region reveals evidence for seven IgG isotypes and for IgD in the horse. J. Immunol. 2004, 173 (5), 3230–3242. (28) Bode, W. The structure of thrombin: a janus-headed proteinase. Semin. Thromb. Hemost. 2006, 32 (1), 16–31. (29) Taylor, K. R.; Gallo, R. L. Glycosaminoglycans and their proteoglycans: host-associated molecular patterns for initiation and modulation of inflammation. FASEB J. 2006, 20 (1), 9–22. (30) Pani, M. A.; Regulla, K.; Segni, M.; Hofmann, S.; Hufner, M.; Pasquino, A. M.; Usadel, K.-H.; Badenhoop, K. A polymorphism within the vitamin D-binding protein gene is associated with Graves’ disease but not with Hashimoto’s thyroiditis. J. Clin. Endocrinol. Metab. 2002, 87 (6), 2564–2567. (31) Smolders, J.; Damoiseaux, J.; Menheere, P.; Hupperts, R. Vitamin D as an immune modulator in multiple sclerosis, a review. J. Neuroimmunol. 2008, 194 (1-2), 7–17. (32) Otterbein, L. R.; Cosio, C.; Graceffa, P.; Dominguez, R. Crystal structures of the vitamin D-binding protein and its complex with actin: structural basis of the actin-scavenger system. Proc. Natl. Acad. Sci. U.S.A. 2002, 99 (12), 8003–8008. (33) Ongagna, J. C.; Pinget, M.; Belcourt, A. Vitamin D-binding protein gene polymorphism association with IA-2 autoantibodies in type 1 diabetes. Clin. Biochem. 2005, 38 (5), 415–419. (34) Carroll, M. C. A protective role for innate immunity in systemic lupus erythematosus. Nat. Rev. Immunol. 2004, 4 (10), 825–831. (35) Bottger, E. C.; Hoffmann, T.; Hadding, U.; Bitter-Suermann, D. Guinea pigs with inherited deficiencies of complement components C2 or C4 have characteristics of immune complex disease. J. Clin. Invest. 1986, 78 (3), 689–695. (36) Steele, F. R.; Chader, G. J.; Johnson, L. V.; Tombran-Tink, J. Pigment epithelium-derived factor: neurotrophic activity and identification as a member of the serine protease inhibitor gene family. Proc. Natl. Acad. Sci. U.S.A. 1993, 90 (4), 1526–1530. (37) Petersen, S. V.; Valnickova, Z.; Enghild, J. J. Pigment-epitheliumderived factor (PEDF) occurs at a physiologically relevant concentration in human blood: purification and characterization. Biochem. J. 2003, 374 (Pt 1), 199–206. (38) Zhang, S. X.; Wang, J. J.; Gao, G.; Shao, C.; Mott, R.; Ma, J.-x. Pigment epithelium-derived factor (PEDF) is an endogenous antiinflammatory factor. FASEB J. 2006, 20 (2), 323–325.

PR800694Y