Printer Friendly

Monomeric calgranulins measured by SELDI-TOF mass spectrometry and calprotectin measured by ELISA as biomarkers in arthritis.

Members of the S100 protein family are acidic proteins of low molecular mass characterized by cell type-specific production and the presence of 2 EFh- and calcium-binding domains (1). Of these proteins, S100A8 (also known as myeloid-related protein 8 and calgranulin A), S100A9 (myeloid-related protein 14, calgranulin B), and S100A12 (calgranulin C) have prompted particular interest among rheumatologists because they are valuable markers of phagocyte activation (2). The secretion of S100A8 and S100A9 by neutrophils and macrophages is stimulated during these cells' interaction with activated endothelial cells (3, 4). A hallmark of the S100A8 and S100A9 proteins is to form dimers, with a heterodimer-homodimer ratio of about 10:1 (5, 6). The S100A8/[(S100A9).sub.2] heterotrimer, better known as calprotectin, has a mass of 36.5 kDa, although trimer formation has never been confirmed by nuclear magnetic resonance analysis (5, 7). Nevertheless, the term calprotectin continues to be used to refer to the S100A8/S100A9 heterocomplex, and the ELISA technique has almost been used almost exclusively to quantify calprotectin status in disease states. Finally, tetramers [(S100A8/S100A9).sub.2] have also been described (8). In inflammatory conditions, S100A8 and S100A9 are produced independently; however, only S100A9 is found in acute inflammatory conditions such as gingivitis, with the production of S100A8 being restricted to chronic inflammation (9). Increased production of S100A8 and S100A9 has been observed in serum, synovial fluid, and synovium samples for many inflammatory rheumatic diseases, including rheumatoid arthritis (RA), [6] psoriatic arthritis (PsA), and ankylosing spondylitis (AS) (10-13), and in the serum of individuals with nonrheumatic inflammatory diseases, such as inflammatory bowel diseases (IBDs) (14). The serum concentration of calprotectin has been correlated with many variables associated with disease activity in RA, PsA, and AS, including such clinical variables as the number of swollen joints (15), the Ritchie index (10), the Disease Activity Score ([DAS.sub.28]) (16), and such biological variables as C-reactive protein (CRP) concentration and erythrocyte sedimentation rate (10, 15, 16).

S100A12 is produced mainly by granulocytes upon inflammatory activation and acts independently of S100A8 and S100A9 (17). Its interaction with the receptor for advanced glycation end products (RAGE) induces proinflammatory signals in endothelium and in cells of the immune system (18). Increased S100A12 concentrations have been found in the serum, synovium, and synovial fluid of patients with RA (11, 19), PsA (19), and AS (19), as well as in the serum of IBD patients (20). Serum amyloid A (SAA), a key promoter of inflammatory events in RA and shown to be produced by inflamed synovial tissue, also induces RAGE activation (21).

The identification of the S100 and SAA groups of proteins as biomarkers is therefore challenging. In the absence of ELISAs specific for the monomeric forms of S100A8 and S100A9 and of a commercially available test for S100A12, we have hypothesized that mass spectrometry (MS) might be a possible approach for detecting these proteins (22). S100A8 has already been identified by 2-dimensional gel electrophoresis to be present in synovium (13), but not in serum (12). Two-dimensional liquid chromatography-coupled tandem MS has also been used successfully to identify several S100 proteins, including S100A8, in serum samples from a few RA patients (11); however, the labor-intensive nature of these 2 proteomics technologies allow the investigation of only small numbers of biological samples. Accordingly, we postulated SELDI-TOF MS technology to be a more appropriate proteomic approach for detecting low molecular weight proteins (<20 kDa), because such an approach permits rapid analysis of hundreds of serum samples at a time (23). The potential of this technology for discovering biomarkers has been demonstrated for such chronic inflammatory conditions as RA and PsA (24), as well as for IBD (25). We therefore investigated S100A8, S100A9, S100A12, and SAA status in RA, PsA, and AS patients, with IBD patients used as a positive inflammatory control (IC) group. We then evaluated the data generated for these markers with respect to the plasma calprotectin concentration and several clinical and biological variables associated with arthritis activity.

Patients and Methods


Blood samples were prospectively collected from 139 Caucasian patients after they had provided informed consent. The study protocol was approved by the ethics committee of our academic hospital (CHU de Liege). Beginning in 2002, we collected blood samples into 10-mL serum separator Vacutainer Tubes (BD Medical Systems). We allowed the blood to clot for 30 min at room temperature and centrifuged the samples at 700g for 10 min. All serum samples were aliquoted and immediately frozen at -80[degrees]C until thawed for SELDI-TOF MS analysis. We defined 5 sets of patients: (a) a noninflammatory control (NIC) group of 36 individuals (16 healthy individuals and 20 osteoarthritis patients), (b) 34 RA patients, (c) 22 PsA patients, (d) 19 AS patients, and (e) an IC group of 28 patients with IBD (14 with Crohn disease and 14 with ulcerative colitis).

The epidemiologic characteristics of the patient groups are summarized in Table 1. RA patients fulfilling the 1987 American College of Rheumatology criteria (26) had a median [DAS.sub.28] of 6.3 (range, 3.5-8.8), with 86% of the scores >5.1 (high disease activity). PsA patients had active disease with at least 3 tender and swollen joints. The AS patients [modified New York criteria (27)] had active disease, as indicated by a median Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) of 6/10 (range, 4/10-10/10). Diagnosis of IBD patients was made according to validated criteria (28). Active Crohn disease was defined by a Harvey-Bradshaw index [greater than or equal to]7, and active ulcerative colitis was defined by both clinical and endoscopic signs of activity.

The preparation and reproducibility of ProteinChip arrays (Vermillion/Ciphergen Biosystems) are described in the supplemental methods in the Data Supplement that accompanies the online version of this article at


Peaks were detected with ProteinChip Biomarker Wizard software (version 3.0; Bio-Rad Laboratories). We analyzed the data by 2 approaches, nonparametric Mann-Whitney U tests and a machine-learning algorithm called random forests (29), after we had completed various preprocessing steps (24). Random forests is a decision-tree multivariate analysis that estimates the relevance or relative contribution of each peak to the classification of 2 groups (29). The latter approach allows m/z values to be ranked according to their relevance for differentiating the 2 groups based on quantitative estimates of the percentage of information (% of info) supplied. P values <0.05 were considered to be statistically significant.


S100A8, S100A9, S100A12, and SAA were assessed by Western blot (WB) analysis. In brief, 2 [micro]L of serum was run on 12% NuPAGE Bis-Tris polyacrylamide gels (Invitrogen), transferred, and incubated with anti-S100A8 monoclonal antibody (20 [micro]L diluted in 10 mL; BMA Biomedicals), anti-S100A9 polyclonal antibody (10 [micro]L diluted in 10 mL; Santa Cruz Biotechnology), anti-S100A12 polyclonal antibody (20 [micro]L diluted in 10 mL: Santa Cruz Biotechnology), or anti-SAA monoclonal antibody (5 [micro]L diluted in 10 mL; Abcam). We then incubated with a mouse secondary antibody (2 [micro]L diluted in 10 mL;GEHealthcare/Amersham Biosciences) to specifically detect S100A8 and SAA, with a rabbit secondary antibody (2 [micro]L diluted in 10 mL; Dako) to detect S100A9, or with a goat secondary antibody (2 [micro]L diluted in 10 mL; Santa Cruz Biotechnology) to detect S100A12. Proteins were revealed with an enhanced chemiluminescence detection method according to the manufacturer's instructions (GE Healthcare/Amersham Biosciences).

The supplemental methods in the online Data Supplement describe the immunoprecipitation and 1-dimensional gel electrophoresis methods as well as LC-MS/MS identification of SAA.


We carried out ELISAs for anti- cyclic citrullinated peptide 2 (anti-CCP2) antibody (cutoff, 5 000 relative units/L; Euroimmun), calprotectin (range, 1.6-100 [micro]g/L; Hycult biotechnology), and matrix metalloproteinase 3 (MMP-3) (range, 1.25-20 [micro]g/L; BioSource) (30) as recommended by the respective manufacturers. We used serum samples or, in the case of calprotectin assays, the corresponding plasma samples.


We used ranked data to calculate Spearman correlation coefficients and used the [chi square] test to compare qualitative data. P values <0.05 were considered statistically significant.



We loaded 139 serum samples in duplicate on weak cation-exchange ProteinChip (CM10) arrays and collected 278 spectra. We first compared the spectra for the RA and NIC groups. The Bio-Rad software resolved 200 peaks. Four peaks detected at m/z values of 10 444, 10 835, 12 688, and 13 272 were differentiated with high statistical significance (P values <[10.sup.-6]) by the Mann-Whitney U test. The intensities of all 4 peaks were significantly increased in RA patients. The peak intensities of the healthy control individuals and the osteoarthritis patients were not significantly different, allowing us to pool these 2 groups into a single NIC group (Table 2). A decision-tree random-forests analysis also showed that these 4 peaks distinguished the RA group from the NIC group with high statistical significance. These results allowed these variables to be ranked among the top 10 most discriminatory m/z values: 10 444 (% of info, 3.8%; rank, 7), 10 835 (% of info, 3.7%; rank, 8), 12 688 (% of info, 4.4%; rank, 5), and 13 272 (% of info, 5.3%; rank, 3). These peaks were thought to be potentially related to 3 S100 proteins: S100A8 (calculated mass, 10 835 Da), S100A9 (calculated masses, 12 691 Da and 13 242 Da), and S100A12 (calculated mass, 10 444 Da).

We also loaded the 139 serum samples in triplicate onto immobilized metal affinity capture ProteinChip (IMAC-[Cu.sup.2+]) arrays and collected 417 spectra. Both statistical approaches detected 3 peaks as significantly increased in RA patients compared with the NIC group. These peaks had m/z values of 11 438 (P<[10.sup.-6]; %of info, 1.2%; rank, 20), 11 528 (P<[10.sup.-9]; % of info, 3.7%; rank, 5), and 11 680 (P<[10.sup.-9]; % of info, 6.5%; rank, 1). These peaks were thought to be related to 3 SAA variant proteins: SAA (calculated mass, 11 682 Da), SAA des-Arg (SAA-R) (calculated mass, 11 526 Da) and SAA des-Arg/des-Ser (SAA-RS) (calculated mass, 11 439 Da). Fig. 1 shows these potential biomarkers on spectra collected on CM10 and IMAC[Cu.sup.2+] arrays of serum samples from patients with various forms of arthritis and from NIC group individuals.


We confirmed the identities of the S100A8, S100A9, and S100A12 proteins by comparing WB and SELDI-TOF MS results for the serum samples (Fig. 2). The WB analysis was performed with 8 serum samples (positions 1-8) from each of the 5 sets of patients (NIC, RA, PsA, AS, and IC). The same RA sample was run in position 9 on each gel as a positive control. SELDI-TOF MS spectra for the same serum samples were monitored at m/z values of 10 835, 13 272, and 10 444 and compared with the corresponding WB results. We obtained similar profiles in the WB and SELDI analyses for each protein tested (Fig. 2, A-C). These data suggest that the peaks at 10 835, 13 272, and 10 444 m/z values are the S100A8, S100A9, and S100A12 proteins, respectively. According to these 2 semiquantitative approaches, some RA, PsA, AS, and, to a lesser extent, IC serum samples were positive for the 3 S100 proteins. We detected 2 S100A9 variants. The m/z value of 13 272 agreed well with the calculated mass for an oxidized form of S100A9 (13 242 Da plus 32 Da), and the 12 688 m/z value likely represents an S100A9 variant known as S100A9 *, which has previously been characterized by ultraviolet MALDI MS (31). S100A9 * results from translation beginning at amino acid residue 5 and acetylation of amino acid residue 6 (a Ser residue), yielding a calculated mass of 12 691 Da.


We also confirmed the identities of the S100A8 and S100A9 proteins by eluting the 2 proteins from their immunocomplexes on NP20 arrays after immunodepletion of a serum sample from an RA patient (see Fig. 1 in the online Data Supplement). We also detected a slight cross-reactivity with S100A12, S100A9 *, and S100A9 on the NP20 spectra after eluting S100A8 (the sequence homologies of the S100A8, S100A9, and S100A12 proteins are around 40%) (32). We were unable to immunoprecipitate the S100A12 protein.


To identify the proteins responsible for the 11 438, 11 528, and 11 680 m/z peak values in the RA spectra on IMAC-[Cu.sup.2+] arrays, we collected serum samples from a healthy control individual and an RA patient, depleted the samples of albumin and IgG, and ran them on a 1-dimensional SDS-PAGE gel (see Fig. 2 in the online Data Supplement). After silver staining, we excised a band from the RA sample on the gel with an apparent molecular weight of 11 kDa and subjected it to LC-MS/MS analysis (see Fig. 2 in the online Data Supplement). We also excised another band from the gel at the same position for the serum sample from the healthy control individual. We digested these bands with trypsin and analyzed the resulting peptides by tandem MS (see Fig. 2 in the online Data Supplement). Sequencing analysis of 4 major tryptic fragments (sequence coverage, 51%; total score, 186) revealed the excised protein to be SAA.


We confirmed the identity of SAA by comparing the WB results with the SELDI-TOF MS results for the serum samples (Fig. 2D). Each set of patient samples (but not those of the NIC group) was positive for SAA in the WB analysis. This result confirmed the identity of the peak at m/z 11 680 to be the SAA protein (calculated mass, 11 682 Da). The peaks at m/z values of 11 438, 11 528, and 11 680 were clustered on each spectrum. We therefore hypothesized that the 11 438 and 11 528 m/z peak values represented 2 variants of the original SAA protein. The 11 528 m/z value corresponds to the calculated mass for the SAA protein without its first N-terminal Arg residue (-156 Da); this peak represents the SAA-R protein (calculated mass, 11 526 Da). Similarly, the 11 438 m/z value corresponds to the calculated mass of the SAA protein truncated at the N-terminal end by 2 residues, Arg and Ser (-243 Da); this peak represents the SAA-RS protein (calculated mass, 11 439 Da). We also confirmed the identities of these proteins by eluting them from their immunocomplexes on NP20 after immunodepleting a serum sample from an RA patient (see Fig. 1 in the online Data Supplement).


The discriminatory power of the peaks at m/z values of 10 835 (S100A8), 12 688 (S100A9 *), 13 272 (S100A9), 10 444 (S100A12), 11 680 (SAA), 11 528 (SAA-R), and 11 438 (SAA-RS) detected in the spectra of samples from each of the arthritis groups was assessed with the Mann-Whitney U test (Table 2). In brief, S100 proteins were found to be highly significantly effective (P < [10.sup.-9]) for distinguishing RA, PsA, and AS patients, but not IC patients, from NIC individuals. Similarly, the SAA, SAA-R, and SAA-RS proteins were effective for distinguishing RA, PsA, and IC patients (but not AS patients) from NIC individuals. All of the S100 proteins significantly distinguished RA, PsA, and AS patients from individuals in the IC group, whereas the SAA, SAA-R, and SAA-RS proteins did not, with the exception of AS patients. Lastly, a comparison within the rheumatic inflammatory disease groups revealed that the S100 proteins weakly distinguished PsA and AS patients, whereas the S100A12 protein was the only variable that significantly distinguished RA patients from PsA patients. SAA, SAA-R, and SAA-RS proteins uniformly and weakly distinguished RA from AS patients.

We next analyzed the percentages of serum samples that were positive for the 7 biomarkers. These percentages were calculated according to the peak intensities measured at m/z values of 10 444 (S100A12), 10 835 (S100A8), 13 272 (S100A9), and 12 688 (S100A9 *) on CM10 arrays, and at m/z values of 11 680 (SAA), 11 528 (SAA-R), and 11 438 (SAA-RS) on IMAC-[Cu.sup.2+] arrays. Peak intensity values were averaged. The cutoff value was defined as the highest peak intensity of the corresponding m/z values in the spectra for the NIC serum samples. Similarly, the percentages of plasma samples positive for calprotectin were calculated with the cutoff defined as the highest calprotectin concentration observed in the NIC group. Increased S100 protein peak intensities and increased calprotectin concentrations were detected in 43%-89% and 33%-56% of arthritis serum samples, respectively, and in 15%-34% and 33% of IC patients, respectively (Table 3). The positivity rates among the 4 S100 proteins were significantly linked [[chi square] (9)=84; P<0.0001] and were significantly linked with the positivity rates for calprotectin [[chi square] (7)=38; P<0.0001]. Increased intensities for SAA proteins were detected in 29%-44% of RA patients, 22% of PsA patients, 16% of AS patients, and 33%-42% of IC patients. No significant differences were found between SAA, SAA-R, and SAA-RS intensities in any of the sample sets. The positivities of these variants were also significantly linked [[chi square] (3) = 113; P < 0.0001].

Statistically significant correlations were found between the various S100 proteins, SAA, calprotectin, and other evaluated variables in the RA, PsA, AS, and IC patient groups (Table 4). In brief, values for the 4 S100 proteins were both highly intercorrelated and correlated with plasma calprotectin concentration in all of the patient groups. S100 protein peaks were also correlated with SAA peaks, but only in the RA and IC groups. In the RA group, S100 protein peaks and the calprotectin concentration were correlated with CRP, log MMP-3, and anti-CCP2 antibody (except for S100A12) serum concentrations and with the [DAS.sub.28], but not with the number of tender or swollen joints (data not shown). RA patients who produced S100 proteins or calprotectin had longer disease durations than S100-negative and calprotectin-negative patients (132 months vs 17 months and 117 months vs 16 months, respectively; P < 0.05). S100 protein peaks were not correlated in PsA or AS patient groups with the CRP, SAA, or log MMP-3 (except for S100A12 in the PsA group) serum concentration, or with clinical variables, including the number of swollen or tender joints and the BASDAI (data not shown). S100 protein peaks were also correlated with log MMP-3 serum concentration in the IC patient group.

SAA was correlated with the serum concentration of CRP in all patient groups and with [DAS.sub.28] and logarithm MMP-3 serum concentration in the RA patient group. The plasma calprotectin concentration was correlated with log MMP-3 serum concentration in the RA, PsA, and IC patient groups and with the SAA protein in the RA, AS, and IC patient groups.


In this study, we used the SELDI-TOF MS approach to detect S100A8, S100A9, an S100A9 variant (S100A9 *), S100A12, SAA, SAA-R, and SAA-RS, and we identified these proteins by WB, immunodepletion, and nano-LC-MS/MS. In a previous study, we had already demonstrated byWBanalysis that the peak at m/z 10 835 in one RA serum sample was S100A8 (24). We confirmed the identity of this peak in the present study, not only in serum samples from RA patients but also in samples from patients with other inflammatory diseases. The mass of the protein for the peak identified as S100A9 * was in good agreement with that of a truncated form of S100A9, a variant that has already been detected by ultraviolet MALDI MS in human buffy coats (31), in head and neck squamous cell carcinoma (33), and in neutrophils from pediatric cystic fibrosis patients (34). The SAA variants had previously been identified in renal cancer, but not in arthritis or in patients with IBD. SELDI-TOF MS is therefore a very efficient technology for detecting such arthritis biomarkers as S100 and SAA proteins in their monomeric, truncated, or posttranslationally modified forms.

The highly significant linear correlations between the peak intensities for the S100 proteins and the calprotectin plasma concentration as measured by ELISA confirm the reliability of this new proteomic approach for investigating these inflammation-related proteins. We conclude that these proteins are up-regulated in each of the diseases we studied because the S100 proteins and calprotectin are correlated both qualitatively, as shown by the strong concordance between WB positivity and SELDI-TOF MS positivity, and quantitatively, as shown by the significant linear correlations in peak intensities. Furthermore, the peak intensities of S100A8, S100A9, S100A9 *, and S100A12, as well as the calprotectin concentration, are correlated with variables that reflect the biological and clinical activity of RA, such as the [DAS.sub.28] and serum concentrations of CRP, MMP-3, and anti-CCP2 antibody. These results provide strong support for the clinical relevance of proteomic detection of calgranulins. The anti-CCP2 antibody may also be related to radiologically observed structural damage in RA because anti-CCP antibodies have been demonstrated to be independent predictors of joint damage (35), as is the serum concentration of calprotectin (16). We could not address this question directly because we conducted no x-ray-imaging studies; however, we did find that patients with increased S100 protein or calprotectin concentrations had significantly longer disease durations, which are expected to be associated with greater radiographically detectable damage.

We observed a discrepancy between RA patients and PsA and AS patients in that the 4 S100 proteins were significantly correlated with CRP and MMP-3 concentrations in RA patients, but not in the other 2 groups of patients. We attribute this finding to the fact that AS and PsA patients exhibit abnormal CRP and MMP-3 concentrations less frequently than RA patients (30). The strong correlation between the 4 S100 proteins and the serum concentration of MMP-3 in the IC group agrees with findings of abundant production of both types of proteins in inflamed intestinal tissue (36).

We also observed that the 4 S100 proteins distinguished arthritis conditions from nonarthritis conditions (NIC and IC groups), whereas SAA patterns distinguished inflammatory diseases (RA, PsA, and IC groups) from noninflammatory conditions (the NIC group). It is also interesting that within the arthritis groups the peak intensities for the 4 S100 proteins were correlated with the SAA peak intensities in the RA group but not in the AS and PsA groups. These results mimic what we had already observed with the CRP and MMP-3 variables. We therefore conclude that the arthritis process is correlated with the inflammatory process in RA, in which the acute-phase response is well developed, but not in AS or PsA, in which it is weaker. This conclusion thus suggests that the regulatory pathways at the sites of inflammation and the patterns of local production of S100 and SAA proteins in RA patients are different from those in AS and PsA patients (37).

SELDI-TOF MS is a unique method for distinguishing S100 monomers from multimeric forms and for detecting SAA variants, which are not possible with current ELISAs. The identification of the various SAA forms may be important because they may have different pathophysiological roles. The recent identification via SELDI-TOF analysis of several truncated forms of S100A8 and S100A12 in cystic fibrosis patients suggests that C-terminal truncations affect protein function (34). The presence of S100A8 and S100A12, but not calprotectin, have been found to be characteristic of intra-amniotic inflammation (22), and SAA variants with different properties, such as differential susceptibility to matrix metalloproteinase digestion, have been described (38). These findings demonstrate the relevance of developing new proteomic approaches sufficiently powerful for investigating proteins in their modified or monomeric forms.

In conclusion, we have used SELDI-TOF MS technology to identify several relevant arthritis biomarkers that are correlated with several biological or clinical variables associated with disease activity. We could not address the functional role of these biomarkers in the pathophysiology of arthritis, because we selected serum samples from individuals with the respective disease characteristics and not from individuals at different stages of each disease. Studies that take advantage of the SELDI-TOF MS technology to evaluate the effects of disease stage on these biomarkers may shed light on this question.

Grant/Funding Support: This research was supported by the National Fund for Scientific Research (FNRS, Belgium) and the Fonds d'Investissement pour la Recherche Scientifique (FIRS), CHU de Liege, Belgium.

Financial Disclosures: None declared.

Acknowledgments: The authors thank Aline Desoroux and Gael Cobraiville for their expert technical assistance. M.F. is a Research Associate, and E.L. and M.P.M. are Senior Research Associates at FNRS (National Fund for Scientific Research).

Received November 6, 2007; accepted March 28, 2008. Previously published online at DOI: 10.1373/clinchem.2007.099549


(1.) Foell D, Wittkowski H, Vogl T, Roth J. S100 proteins expressed in phagocytes: a novel group of damage-associated molecular pattern molecules. J Leukoc Biol 2007;81:28-37.

(2.) De Rycke L, Baeten D, Foell D, Kruithof E, Veys EM, Roth J, et al. Differential expression and response to anti-TNF[alpha] treatment of infiltrating versus resident tissue macrophage subsets in autoimmune arthritis. J Pathol 2005;206:17-27.

(3.) Frosch M, Strey A, Vogl T, Wulffraat NM, Kuis W, Sunderkotter C, et al. Myeloid-related proteins 8 and 14 are specifically secreted during interaction of phagocytes and activated endothelium and are useful markers for monitoring disease activity in pauciarticular-onset juvenile rheumatoid arthritis. Arthritis Rheum 2000;43:628-37.

(4.) Sunahori K, Yamamura M, Yamana J, Takasugi K, Kawashima M, Yamamoto H, et al. The S100A8/A9 heterodimer amplifies proinflammatory cytokine production by macrophages via activation of nuclear factor kappa B and p38 mitogen-activated protein kinase in rheumatoid arthritis. Arthritis Res Ther 2006;8:R69.

(5.) Hunter MJ, Chazin WJ. High level expression and dimer characterization of the S100 EF-hand proteins, migration inhibitory factor-related proteins 8 and 14. J Biol Chem 1998;273:12427-35.

(6.) Propper C, Huang X, Roth J, Sorg C, Nacken W. Analysis of the MRP8-MRP14 protein-protein interaction by the two-hybrid system suggests a prominent role of the C-terminal domain of S100 proteins in dimer formation. J Biol Chem 1999; 274:183-8.

(7.) Strupat K, Rogniaux H, Van Dorsselaer A, Roth J, Vogl T. Calcium-induced noncovalently linked tetramers of MRP8 and MRP14 are confirmed by electrospray ionization-mass analysis. J Am Soc Mass Spectrom 2000;11:780-8.

(8.) Korndorfer IP, Brueckner F, Skerra A. The crystal structure of the human [(S100A8/S100A9).sub.2] heterotetramer, calprotectin, illustrates how conformational changes of interacting [alpha]-helices can determine specific association of two EF-hand proteins. J Mol Biol 2007;370:887-98.

(9.) Zwadlo G, Bruggen J, Gerhards G, Schlegel R, Sorg C. Two calcium-binding proteins associated with specific stages of myeloid cell differentiation are expressed by subsets of macrophages in inflammatory tissues. Clin Exp Immunol 1988;72: 510-5.

(10.) Kane D, Roth J, Frosch M, Vogl T, Bresnihan B, FitzGerald O. Increased perivascular synovial membrane expression of myeloid-related proteins in psoriatic arthritis. Arthritis Rheum 2003;48: 1676-85.

(11.) Liao H, Wu J, Kuhn E, Chin W, Chang B, Jones MD, et al. Use of mass spectrometry to identify protein biomarkers of disease severity in the synovial fluid and serum of patients with rheumatoid arthritis. Arthritis Rheum 2004;50: 3792-803.

(12.) Sinz A, Bantscheff M, Mikkat S, Ringel B, Drynda S, Kekow J, et al. Mass spectrometric proteome analyses of synovial fluids and plasmas from patients suffering from rheumatoid arthritis and comparison to reactive arthritis or osteoarthritis. Electrophoresis 2002;23:3445-56.

(13.) Tilleman K, Van Beneden K, Dhondt A, Hoffman I, De Keyser F, Veys E, et al. Chronically inflamed synovium from spondyloarthropathy and rheumatoid arthritis investigated by protein expression profiling followed by tandem mass spectrometry. Proteomics 2005;5:2247-57.

(14.) Lugering N, Stoll R, Kucharzik T, Schmid KW, Rohlmann G, Burmeister G, et al. Immunohistochemical distribution and serum levels of the [Ca.sup.2+]-binding proteins MRP8, MRP14 and their heterodimeric form MRP8/14 in Crohn's disease. Digestion 1995;56:406-14.

(15.) Brun JG, Jonsson R, Haga HJ. Measurement of plasma calprotectin as an indicator of arthritis and disease activity in patients with inflammatory rheumatic diseases. J Rheumatol 1994;21:733-8.

(16.) Hammer HB, Odegard S, Fagerhol MK, Landewe R, van der Heijde D, Uhlig T, et al. Calprotectin (a major leucocyte protein) is strongly and independently correlated with joint inflammation and damage in rheumatoid arthritis. Ann Rheum Dis 2007;66:1093-7.

(17.) Vogl T, Propper C, Hartmann M, Strey A, Strupat K, van den Bos C, et al. S100A12 is expressed exclusively by granulocytes and acts independently from MRP8 and MRP14. J Biol Chem 1999; 274:25291-6.

(18.) Hofmann MA, Drury S, Fu C, Qu W, Taguchi A, Lu Y, et al. RAGE mediates a novel proinflammatory axis: a central cell surface receptor for S100/ calgranulin polypeptides. Cell 1999;97:889-901.

(19.) Foell D, Kane D, Bresnihan B, Vogl T, Nacken W, Sorg C, et al. Expression of the pro-inflammatory protein S100A12 (EN-RAGE) in rheumatoid and psoriatic arthritis. Rheumatology (Oxford) 2003; 42:1383-9.

(20.) Foell D, Kucharzik T, Kraft M, Vogl T, Sorg C, Domschke W, et al. Neutrophil derived human S100A12 (EN-RAGE) is strongly expressed during chronic active inflammatory bowel disease. Gut 2003;52:847-53.

(21.) Cai H, Song C, Endoh I, Goyette J, Jessup W, Freedman SB, et al. Serum amyloid A induces monocyte tissue factor. J Immunol 2007;178: 1852-60.

(22.) Buhimschi IA, Buhimschi CS, Weiner CP, Kimura T, Hamar BD, Sfakianaki AK, et al. Proteomic but not enzyme-linked immunosorbent assay technology detects amniotic fluid monomeric calgranulins from their complexed calprotectin form. Clin Diagn Lab Immunol 2005;12:837-44.

(23.) Tang N, Tornatore P, Weinberger SR. Current developments in SELDI affinity technology. Mass Spectrom Rev 2004;23:34-44.

(24.) de Seny D, Fillet M, Meuwis MA, Geurts P, Lutteri L, Ribbens C, et al. Discovery of new rheumatoid arthritis biomarkers using the surface-enhanced laser desorption/ionization time-of-flight mass spectrometry ProteinChip approach. Arthritis Rheum 2005;52:3801-12.

(25.) Meuwis MA, Fillet M, Geurts P, de Seny D, Lutteri L, Chapelle JP, et al. Biomarker discovery for inflammatory bowel disease, using proteomic serum profiling. Biochem Pharmacol 2007;73: 1422-33.

(26.) Arnett FC, Edworthy SM, Bloch DA, MacShane DJ, Fries JF, Cooper N. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315-24.

(27.) Dougados M, van der Linden S, Juhlin R, Huitfeldt B, Amor B, Calin A, et al. The European Spondylarthropathy Study Group preliminary criteria for the classification of spondylarthropathy. Arthritis Rheum 1991;34:1218-27.

(28.) Sands BE. From symptom to diagnosis: clinical distinctions among various forms of intestinal inflammation. Gastroenterology 2004;126:1518-32.

(29.) Geurts P, Fillet M, de Seny D, Meuwis MA, Malaise M, Merville MP, et al. Proteomic mass spectra classification using decision tree based ensemble methods. Bioinformatics 2005;21: 3138-45.

(30.) Ribbens C, Martin y Porras M, Franchimont N, Kaiser MJ, Jaspar JM, Damas P, et al. Increased matrix metalloproteinase-3 serum levels in rheumatic diseases: relationship with synovitis and steroid treatment. Ann Rheum Dis 2002;61: 161-6.

(31.) Vogl T, Roth J, Sorg C, Hillenkamp F, Strupat K. Calcium-induced noncovalently linked tetramers of MRP8 and MRP14 detected by ultraviolet matrix-assisted laser desorption/ionization mass spectrometry. J Am Soc Mass Spectrom 1999;10: 1124-30.

(32.) Foell D, Roth J. Proinflammatory S100 proteins in arthritis and autoimmune disease. Arthritis Rheum 2004;50:3762-71.

(33.) Roesch Ely M, Nees M, Karsai S, Magele I, Bogumil R, Vorderwulbecke S, et al. Transcript and proteome analysis reveals reduced expression of calgranulins in head and neck squamous cell carcinoma. Eur J Cell Biol 2005;84:431-44.

(34.) McMorran BJ, Ouvry Patat SA, Carlin JB, Grimwood K, Jones A, Armstrong DS, et al. Novel neutrophil-derived proteins in bronchoalveolar lavage fluid indicate an exaggerated inflammatory response in pediatric cystic fibrosis patients. Clin Chem 2007;53:1782-91.

(35.) Syversen SW, Gaarder PI, Goll GL, Odegard S, Haavardsholm EA, Mowinckel P, et al. High anti-CCP levels and an algorithm of four variables predict radiographic progression in patients with rheumatoid arthritis: results from a 10-year longitudinal study. Ann Rheum 2008; 2008;67: 212-7.

(36.) Louis E, Ribbens C, Godon A, Franchimont D, De Groote D, Hardy N, et al. Increased production of matrix metalloproteinase-3 and tissue inhibitor of metalloproteinase-1 by inflamed mucosa in inflammatory bowel disease. Clin Exp Immunol 2000;120:241-6.

(37.) Mullan RH, Bresnihan B, Golden-Mason L, Markham T, O'Hara R, FitzGerald O, et al. Acute-phase serum amyloid A stimulation of angiogenesis, leukocyte recruitment, and matrix degradation in rheumatoid arthritis through an NF-[kappa]B-dependent signal transduction pathway. Arthritis Rheum 2006;54:105-14.

(38.) Stix B, Kahne T, Sletten K, Raynes J, Roessner A, Rocken C. Proteolysis of AA amyloid fibril proteins by matrix metalloproteinases-1, -2, and -3. Am J Pathol 2001;159:561-70.

Dominique de Seny, [1] * Marianne Fillet, [2] Clio Ribbens, [1] Raphael Maree, [3] Marie-Alice Meuwis, [2] Laurence Lutteri, [2] Jean-Paul Chapelle, [2] Louis Wehenkel, [4] Edouard Louis, [5] Marie-Paule Merville, [2] and Michel Malaise [1]

[1] Laboratory of Rheumatology, GIGA Research, CHU, University of Liege; [2] Laboratory of Clinical Chemistry, GIGA Research, University of Liege; [3] GIGA Bioinformatics Platform, University of Liege; [4] Bioinformatics and Modeling Unit, Department of Electrical Engineering & Computer Science, GIGA Research, University of Liege; [5] Laboratory of Hepato-Gastroenterology, CHU, University of Liege, Liege, Belgium.

* Address correspondence to this author at: Laboratory of Rheumatology, Tour GIGA +2, CHU, 4000 Liege, Belgium. Fax (32) 4 366 45 34; e-mail

[6] Nonstandard abbreviations: RA, rheumatoid arthritis; PsA, psoriatic arthritis; AS, ankylosing spondylitis; IBD, inflammatory bowel disease; [DAS.sub.28], Disease Activity Score; CRP, C-reactive protein; RAGE, receptor for advanced glycation end products; SAA, serum amyloid A; MS, mass spectrometry; IC, inflammatory control; NIC, noninflammatory control; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; WB, western blot; anti-CCP2, anti-cyclic citrullinated peptide 2; MMP-3, matrix metalloproteinase 3; CM10, weak cation-exchange ProteinChip; % of info, percentage of information; IMAC-[Cu.sup.2+], immobilized metal affinity capture ProteinChip; SAA-R, SAA des-Arg; SAA-RS, SAA des-Arg/ des-Ser.
Table 1. Epidemiologic characteristics of the patients and
controls. (a)

Characteristic NIC (n = 36) RA (n = 34)

Women patients, % 66 65
Age, years 57 (21-77) 55 (27-79)
Disease duration, years - 8.7 (0.1-19)
Concomitant therapy
 Methotrexate treatment, % 0 62
 Methotrexate, mg/week 0 12.5 (5-20)
 Prednisolone treatment, % 0 68
 Prednisolone, mg/day 0 10 (2.5-40)
CRP, mg/L 2.6 (0-9) 12 (1-69)
Increased CRP (>6 mg/L), % 5 65
MMP-3,[micro] g/L 6.8 (3-22) 25.5 (3-175)
Increased MMP-3, %b 0 66
Increased CRP and MMP-3, % 0 56
Calprotectin,[micro] g/L 272 (107-542) 607 (145-3387)
IgM RFcpositivity, % 0 85
Positivity for anti-CCP2, % 0 82
28-Joint counts (tender)d 0 12 (3-28)
28-Joint counts (swollen) 0 10 (3-23)

Characteristic PsA (n = 22) AS (n = 19)

Women patients, % 55 26
Age, years 46 (21-80) 41 (30-58)
Disease duration, years 5.3 (0.1-31) 17 (0.7-30)
Concomitant therapy
 Methotrexate treatment, % 36 10
 Methotrexate, mg/week 15 (12.5-25) 15 (15-15)
 Prednisolone treatment, % 23 16
 Prednisolone, mg/day 7.5 (5-7.5) 5 (2.5-15)
CRP, mg/L 8 (1-112) 9 (1-41)
Increased CRP (>6 mg/L), % 59 58
MMP-3,[micro] g/L 18.7 (4-171) 14.5 (5-52)
Increased MMP-3, %b 56 16
Increased CRP and MMP-3, % 39 10
Calprotectin,[micro] g/L 717 (154-2690) 297 (79-2239)
IgM RFcpositivity, % 0 0
Positivity for anti-CCP2, % 0 0
28-Joint counts (tender)d 10 (3-52) 0
28-Joint counts (swollen) 4 (3-22) 0

Characteristic IC (n = 28)

Women patients, % 41
Age, years 36 (21-59)
Disease duration, years 6.8 (0.5-27)
Concomitant therapy
 Methotrexate treatment, % 0
 Methotrexate, mg/week 0
 Prednisolone treatment, % 0
 Prednisolone, mg/day 0
CRP, mg/L 13 (1-228)
Increased CRP (>6 mg/L), % 65
MMP-3,[micro] g/L 10.5 (2-161)
Increased MMP-3, %b 17
Increased CRP and MMP-3, % 0
Calprotectin,[micro] g/L 563 (113-1975)
IgM RFcpositivity, % 0
Positivity for anti-CCP2, % 0
28-Joint counts (tender)d 0
28-Joint counts (swollen) 0

(a) Values are presented as the median (range) except where otherwise

(b) Abnormal MMP-3 values were as defined in Ribbens et al. (30),
with cutoffs of 14 [micro]g/L in women and 34 [micro]g/L in men.

(c) RF, rheumatoid factor.

(d) The 28 joints are the following: metacarpophalangeal, 10; proximal
interphalangeal, 10; wrists, 2; elbows, 2; shoulders, 2; knees, 2.

Table 2. Discriminatory power of selected peaks as assessed with
the Mann-Whitney U test and represented as P values.

 P values

 NIC vs

Array m/z Protein RA PsA

CM10 10 835 S100A8 <[10.sup.-9] <[10.sup.-9]
CM10 12 688 S100A9 * <[10.sup.-9] <[10.sup.-9]
CM10 13 272 S100A9 <[10.sup.-9] <[10.sup.-9]
CM10 10 444 S100A12 <[10.sup.-9] <[10.sup.-9]
IMAC-[Cu.sup.-2+] 11 680 SAA <[10.sup.-8] <[10.sup.-4]
IMAC-[Cu.sup.-2+] 11 528 SAA-R <[10.sup.-9] <[10.sup.-2]
IMAC-[Cu.sup.-2+] 11 438 SAA-RS <[10.sup.-6] <[10.sup.-2]

 NIC vs IC vs

Array AS IC RA

CM10 <[10.sup.-9] 0.12 <[10.sup.-5]
CM10 <[10.sup.-9] <[10.sup.-2] <[10.sup.-4]
CM10 <[10.sup.-9] 0.37 <[10.sup.-6]
CM10 <[10.sup.-9] 0.09 <[10.sup.-9]
IMAC-[Cu.sup.-2+] 0.12 <[10.sup.-8] 0.33
IMAC-[Cu.sup.-2+] 0.33 <[10.sup.-4] 0.80
IMAC-[Cu.sup.-2+] ND (a) <[10.sup.-4] 0.33

 IC vs RA vs

Array PsA AS PsA

CM10 <[10.sup.-3] <[10.sup.-7] 0.38
CM10 <[10.sup.-3] <[10.sup.-6] 0.67
CM10 <[10.sup.-4] <[10.sup.-8] 0.25
CM10 <[10.sup.-3] <[10.sup.-7] <[10.sup.-2]
IMAC-[Cu.sup.-2+] 0.02 <[10.sup.-3] 0.17
IMAC-[Cu.sup.-2+] 0.16 <[10.sup.-2] 0.02
IMAC-[Cu.sup.-2+] 0.04 <[10.sup.-3] 0.15

 RA vs PsA vs

Array AS AS

CM10 0.07 <[10.sup.-2]
CM10 0.01 <[10.sup.-2]
CM10 0.14 <[10.sup.-2]
CM10 0.86 <[10.sup.-2]
IMAC-[Cu.sup.-2+] <[10.sup.-3] 0.05
IMAC-[Cu.sup.-2+] <[10.sup.-4] 0.08
IMAC-[Cu.sup.-2+] <[10.sup.-3] 0.05

(a) ND, no peak detected.

Table 3. Percentages of serum samples positive for S100A12, S100A8,
S100A9, S100A9 *, SAA, SAA-R, and/or SAA-RS proteins. (a)


 S100A12/ S100A8/ S100A9/ S100A9 */
Group CM10 CM10 CM10 CM10

NIC, % 0 0 0 0
RA, % 70 56 50 56
PsA, % 52 56 43 52
AS, % 84 89 52 84
IC, % 19 27 15 34


 Calprotectin/ IMAC- IMAC- IMAC-
Group ELISA [Cu.sup.2+] [Cu.sup.2+] [Cu.sup.2+]

NIC, % 0 0 0 0
RA, % 56 44 37 29
PsA, % 54 22 22 22
AS, % 33 16 16 16
IC, % 33 42 33 42

(a) Percentages of serum samples with MS peaks indicating positivity
for S100A12 (m/z = 10 444), S100A8 (m/z = 10 835), S100A9
(m/z = 13 272), S100A9 * (m/z = 12 688), SAA (m/z = 11 680),
SAA-R ( m/z = 11 528), and SAA-RS ( m/z = 11 439). Cutoff values were
defined as the highest peak intensity found in the spectra of serum
samples from the NIC group. The concentration of calprotectin in serum
samples was assessed by ELISA, and the cutoff value was defined as the
highest value obtained in samples from the NIC group.

Table 4. Correlation coefficients for S100 proteins, SAA, and clinical
variables for serum samples collected from RA, PsA, AS, and IC
patients. (a)

 S100A8 S100A9 S100A9 * S100A12
 S100A8 - 0.97 *** 0.99 *** 0.92 ***
 S100A9 - - 0.97 *** 0.89 ***
 S100A9 * - - - 0.93 ***
 S100A12 - - - -
 Calprotectin - - - -
 SAA - - - -
 S100A8 - 0.99 *** 0.96 *** 0.92 ***
 S100A9 - - 0.95 *** 0.93 ***
 S100A9 * - - - 0.92 ***
 S100A12 - - - -
 Calprotectin - - - -
 SAA - - - -
 S100A8 - 0.96 *** 0.95 *** 0.86 ***
 S100A9 - - 0.94 *** 0.88 ***
 S100A9 * - - - 0.79 ***
 S100A12 - - - -
 Calprotectin - - - -
 SAA - - - -
 S100A8 - 0.95 *** 0.96 *** 0.80 ***
 S100A9 - - 0.97 *** 0.79 ***
 S100A9 * - - - 0.80 ***
 S100A12 - - - -
 Calprotectin - - - -
 SAA - - - -

 Calprotectin SAA CRP Log MMP-3
 S100A8 0.83 *** 0.55 ** 0.53 ** 0.45 *
 S100A9 0.84 *** 0.54 * 0.55 ** 0.46 *
 S100A9 * 0.81 *** 0.53 ** 0.52 * 0.40 *
 S100A12 0.81 *** 0.47 ** 0.48 ** 0.39 *
 Calprotectin - 0.55 ** 0.54 * 0.55 **
 SAA - - 0.73 *** 0.44 *
 S100A8 0.51 * 0.39 0.31 0.29
 S100A9 0.49 * 0.34 0.32 0.3
 S100A9 * 0.52 * 0.35 0.3 0.33
 S100A12 0.51 * 0.18 0.25 0.47 **
 Calprotectin - 0.28 0.46 * 0.40 *
 SAA - - 0.71 *** 0.14
 S100A8 0.43 * 0.22 0.09 0.04
 S100A9 0.44 * 0.23 0.14 0.06
 S100A9 * 0.54 * 0.16 0.22 0.24
 S100A12 0.37 0.26 0.09 0.07
 Calprotectin - 0.42 * 0.26 0.09
 SAA - - 0.42 * 0.39
 S100A8 0.93 *** 0.49 * 0.08 0.45 *
 S100A9 0.90 *** 0.41 * 0.03 0.50 *
 S100A9 * 0.94 *** 0.47 ** 0.12 0.52 *
 S100A12 0.85 *** 0.36 * 0.05 0.52 *
 Calprotectin - 0.40 * 0.07 0.52 *
 SAA - - 0.73 *** 0.36 *

 Anti-CCP2 RF (b) [DAS.sub.28]
 S100A8 0.36 * 0.28 0.50 *
 S100A9 0.36 * 0.30 0.40 *
 S100A9 * 0.35 * 0.28 0.51 *
 S100A12 0.32 0.30 0.39 *
 Calprotectin 0.37 * 0.33 0.48 **
 SAA 0.25 0.07 0.52 *
 S100A9 *
 S100A9 *
 S100A9 *

(a) *P<0.05; ** P<0.01; ***P<0.001.

(b) RF, rheumatoid factor.
COPYRIGHT 2008 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2008 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:de Seny, Dominique; Fillet, Marianne; Ribbens, Clio; Maree, Raphael; Meuwis, Marie-Alice; Lutteri, L
Publication:Clinical Chemistry
Date:Jun 1, 2008
Previous Article:Aspirin responsiveness in healthy volunteers measured with multiple assay platforms.
Next Article:Effect of collection tube type and preanalytical handling on myeloperoxidase concentrations.

Related Articles
Cardiovascular proteomics; methods and protocols.
Quality control of serum albumin depletion for proteomic analysis.
Putting the "bio" back into biomarkers: orienting Proteomic discovery toward biology and away from the measurement platform.
Reproducibility in protein profiling by MALDI-TOF mass spectrometry.
Mass spectrometric analysis of protein markers for ovarian cancer.
Validation of breast cancer biomarkers identified by mass spectrometry.
Direct-tissue SELDI-TOF mass spectrometry analysis: a new application for clinical proteomics.
Point proteomic patterns in biological fluids: do they represent the future of cancer diagnostics?

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters