SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Objective

To identify pathogenic and/or disease-specific short peptides in sera from patients with systemic sclerosis (SSc).

Methods

Serum samples from 40 patients with SSc, 30 patients with systemic lupus erythematosus, 21 patients with rheumatoid arthritis, 30 patients with osteoarthritis, and 26 healthy donors were tested. Short peptides with molecular weights of smaller than ∼3 kd, purified from the sera by magnetic bead–based hydrophobic interaction chromatography 18, were detected and their amino acid sequences determined using matrix-assisted laser desorption ionization–time-of-flight mass spectrometry. Effects of the identified peptides on fibroblasts and microvascular endothelial cells were tested using synthesized peptides and sera containing the peptides.

Results

A group of peptides with mass/charge (m/z) values of 1,865, 1,778, 1,691, 1,563, and 1,450 were detected predominantly in the SSc sera. These peptides were identified as family members of complement C3f-des-arginine (DRC3f) derived from C3b. The level of DRC3f (m/z 1,865) was related to vascular involvement in SSc and to SSc disease activity. The synthesized peptides of DRC3f and C3f, as well as the filtrated sera containing DRC3f, enhanced proliferation of microvascular endothelial cells, but not fibroblasts. Both DRC3f and C3f increased production of transforming growth factor β1 by dermal microvascular endothelial cells.

Conclusion

This comprehensive peptidomics analysis revealed the predominance of DRC3f in the sera of patients with SSc. Investigation of DRC3f may be a useful tool for the diagnosis and evaluation of disease activity in SSc. Moreover, its demonstrated effects on endothelial cells suggest a potential role for DRC3f in the pathophysiologic mechanisms of SSc.

Systemic sclerosis (SSc) is an autoimmune disorder of the connective tissue, characterized by widespread vascular lesions and fibrosis. Raynaud's phenomenon, increased thickness of the vascular wall, devascularization, and thickness of the basement membrane are typical features of SSc. The digital arteries of patients with SSc exhibit marked intimal proliferation, resulting in severe narrowing and occlusion of arterial lumen and limited vasodilative responses to vasodilator therapies (1).

Thus far, the contributions of many molecules to the pathogenesis and diagnosis of SSc have been reported (2–9). Investigation of these known molecules is of great value for the understanding of SSc pathogenesis; however, it would also be important to identify novel SSc-related molecules by hypothesis-free comprehensive surveillance. In this regard, genomic surveillance using single-nucleotide polymorphisms, DNA arrays for messenger RNA expression, and proteomic surveillance have been used recently (10). These techniques can detect various molecules that may be involved in the pathogenesis of SSc or that may be useful for diagnosis. However, the targets of surveillance have been limited to genes or relatively large proteins. Peptides with low molecular weights (smaller than ∼3 kd) generally cannot be investigated by these techniques.

Peptides with such low molecular weights often play a central role in biologic and pathologic processes. Typical of these peptides would be bradykinin, which is a peptide that is only 9 amino acid residues in length and produced from kininogen by specific proteolysis. Another example would be substance P, in a neuropeptide of 11 amino acid residues. Peptides with such low molecular weights are estimated to be produced in large amounts by proteolysis of large proteins, which thereby generates various bioactive and/or diagnostically useful short peptides in addition to the known peptides in the body. However, there are only a few ways to survey such peptides efficiently. Quite recently, mass spectrometry methods that directly detect and identify peptides with low molecular weights and with low concentrations (referred to as peptidomics analysis) have been developed. This offers a promising approach to the discovery of novel short peptides that could be useful in the diagnosis and treatment of diseases (11–13).

In the present study, we aimed to identify low molecular weight peptides in the serum of patients with SSc in order to understand the role of these peptides in disease pathogenesis, which could lead to better diagnosis and treatment of SSc. By combining a microamount peptide-separating method with magnetic beads and matrix-assisted laser desorption ionization–time-of-flight (MALDI-TOF) mass spectrometry, we were able to comprehensively detect short peptides in the sera of patients with SSc, those with non-SSc rheumatic diseases, and healthy donors. In comparing the results between each group of samples, we found that the sera of patients with SSc carried complement C3f-des-arginine (DRC3f) and its degraded smaller fragments at very high levels and with high frequency, compared with the sera of patients with non-SSc rheumatic diseases or healthy donors. Furthermore, we demonstrated that these peptides promoted proliferation of human microvascular endothelial cells (HMVECs) in vitro. The results of this study will open a new field of investigation into the pathophysiologic processes of SSc.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Patients.

Forty patients with SSc (1 man and 39 women) whose diagnosis fulfilled the American College of Rheumatology (ACR; formerly, the American Rheumatism Association) criteria for SSc (14) were studied. The mean age of these SSc patients was 56.4 years (range 29–83 years). The duration of SSc was defined as the period from the date of onset of the disease to the date of blood sampling. Appearance of the first SSc symptom was defined as the first presentation of a symptom, even if it was before the date of diagnosis. The mean disease duration among the patients with SSc was 6.83 years (range 3–441 months). Skin changes were recorded using the modified Rodnan total skin thickness score (15).

Twenty-six serum samples from age- and sex-matched healthy donors (1 man and 25 women) were selected as controls for the SSc samples. The mean age of the control group was 55.6 years (range 27–84 years). In addition, 30 patients (3 men and 27 women) with systemic lupus erythematosus (SLE) (16) were enrolled. The mean age of the SLE patients was 47.2 years (range 23–83 years), and the mean SLE Disease Activity Index (SLEDAI) score (scale 0–105) (17) in these patients was 13.4 (range 7–28). Twenty-one patients with rheumatoid arthritis (RA) (2 men and 19 women) were also enrolled. The mean age of the RA patients was 60.1 years (range 36–78 years). Ten of the RA patients had an erythrocyte sedimentation rate (ESR) of higher than 30 mm/hour, and 18 RA patients had an elevated C-reactive protein (CRP) level. Finally, 30 patients (6 men and 24 women) with osteoarthritis (OA) were also enrolled in this study. The mean age of the OA patients was 64.4 years (range 46–84 years), and these patients were diagnosed according to the ACR criteria for knee OA (18).

The serum samples were stored at −80°C until used. All samples were obtained after subjects provided their informed consent, and the study was approved by the local institutional ethics committee.

Separation and purification of serum peptides.

Serum peptides were separated and purified with a purification kit of magnetic beads, using magnetic bead–based hydrophobic interaction chromatography 18 (MB-HIC18) (Bruker Daltonics, Ettlingen, Germany) according to the manufacturer's instructions (19). Briefly, 5 μl of magnetic beads was mixed with 10 μl of MB-HIC18 binding solution and 5 μl of each serum sample, and the beads were then collected by a magnetic beads separator. After washing 4 times in an MB-HIC18 wash solution, we eluted the bound peptides off the beads into 5 μl of 50% acetonitrile, and then diluted 1 μl of the eluted peptide solution in 10 μl of matrix solution (0.3 mg/ml α-cyano-4-hydroxycinnamic acid in ethanol:acetone 2:1). Finally, 1 μl of the diluted eluate was loaded onto 600-μm–diameter spots on a metal plate for mass spectrometry analysis (AnchorChip; Bruker Daltonics).

Serum samples from the patients as well as from the healthy controls were filtered using 3.0-kd filters (Microcon; Millipore, Bedford, MA) at 15,000 revolutions per minute for 2 hours. Peptides in the flow-through fraction were concentrated with Ziptip-C18 (Millipore) and then eluted off the tip in 1 μl of 50% acetonitrile. The eluate was diluted in 10 μl of matrix solution, and 1 μl of the eluate–matrix mixture was then loaded on the AnchorChip target plate.

Mass spectrometry, identification of peptide sequences, and comparison of peptide profiles.

The eluted peptides were detected with a MALDI-TOF mass spectrometer (Ultraflex; Bruker Daltonics). The mass spectra of the peptide peaks were initially detected in the automatic linear positive mode, for simple comparison between sample groups. Mass spectrometry analysis was then performed using the reflector mode, to obtain accurate masses of the peptides. Finally, MALDI-TOF analysis and a subsequent sequence search using the search engine Mascot (www.matrixscience.com) were performed to identify the sequences of the peptides of interest.

A comparative analysis of the mass spectra of the peptide peaks between SSc patients and healthy controls as well as between SSc patients and patients with other rheumatic diseases was performed using ClinPro Tools (CPT) software, version 1.0 (Bruker Daltonics) as previously described (19). To evaluate the relative levels of peptides in each serum sample, we defined the upper limit of normal range as the mean peak intensity + 3 SD of healthy control sera. Peptide levels in the serum from each patient, expressed in arbitrary units, were calculated according to the formula: arbitrary units = individual peak intensity/(mean peak intensity of healthy controls + 3 SD) × 100. The cutoff point to define a relative fold increase in peptide levels was set at 100 arbitrary units.

Preparation of peptides.

The peptides of complement C3f, whose sequence is NH2-Ser-Ser-Lys-Ile-Thr-His-Arg-Ile-His-Trp-Glu-Ser-Ala-Ser-Leu-Leu-Arg-COOH (17 amino acid residues with peptide mass of 2,021 daltons), and DRC3f, whose sequence is NH2-Ser-Ser-Lys-Ile-Thr-His-Arg-Ile-His-Trp-Glu-Ser-Ala-Ser-Leu-Leu-COOH (16 amino acid residues with peptide mass of 1,865 daltons), were synthesized by solid-phase peptide synthesis and purified by RP-18 high-performance liquid chromatography as previously described (20). Fibrinopeptide A (FPA), purchased from Bachem (Bubendorf, Switzerland), was used as a control peptide.

Cell culture.

Lung HMVECs and adult dermal HMVECs (both from Cell Systems, Kirkland, WA) were cultured in CS-C serum-free complete medium with a growth factor (SF-4Z0-500; Cell Systems) on type I collagen–coated Cellware plates (Becton Dickinson, Bedford, UK). Normal human dermal fibroblasts (Cambrex, Long Beach, CA) were cultured in RPMI 1640 medium (Sigma-Aldrich, Poole, UK) containing 10% fetal bovine serum (FBS) on tissue culture plates (Becton Dickinson). All cells were cultured at 37°C in a humidified atmosphere of 5% CO2. The medium was replaced every other day.

Measurement of growth factors.

Lung and adult dermal HMVECs were cultured on 12-well type I collagen–coated plates. After the cell density reached 70% confluence, the cells were further cultured for 72 hours in CS-C serum-free medium (SF-4Z0-500; Cell Systems) containing different amounts of C3f, DRC3f, and FPA (0, 15.6, 31.25, 62.5, 125, 250, and 500 ng/ml) with or without 20 ng/ml recombinant human insulin-like growth factor 1 (IGF-1) (Promega, Madison, WI). Similarly, normal human dermal fibroblasts were cultured on 12-well plates in RPMI 1640 medium containing 10% FBS. After the cell density reached 70% confluence, the cells were further cultured for 72 hours with serum-free RPMI 1640 medium containing different amounts of C3f, DRC3f, and FPA with or without IGF-1. The supernatant was collected and stored at −20°C until used.

Quantikine immunoassay kits (R&D Systems, Minneapolis, MN) were used for quantifying human transforming growth factor β1 (TGFβ1), vascular endothelial growth factor (VEGF), and epidermal growth factor (EGF). Concentrations of TGFβ1, VEGF, and EGF in the supernatants of the 3 cell lines were measured according to the manufacturer's instructions.

Cell proliferation assay.

Lung and adult dermal HMVECs were cultured in growth factor–free medium on type I collagen–coated 96-well plates at a cell density of 4,000/well. Normal human dermal fibroblasts were cultured in RPMI 1640 medium containing 1% FBS at a cell density of 3,000/well. C3f, DRC3f, or FPA, variously diluted in medium, was then added into the wells with or without IGF-1. Six different wells were used for each condition to calculate the mean cell numbers.

Proliferation of lung and adult dermal HMVECs was also quantitated using, in place of synthesized peptides, whole sera or filtered serum samples that were negative or positive for DRC3f. Specifically, serum samples filtrated with a 0.2-μm filter (Millex; Millipore) were added to the culture medium up to a final concentration of 5% (5 μl per well). Similarly, serum samples filtrated with a 3.0-kd filter (Microcon; Millipore) were added into the medium up to a final concentration of 15% (15 μl per well). Cells were cultured on 96-well plates for 48 hours at 37°C in a humidified atmosphere of 5% CO2.

The CellTiter 96 AQueous One Solution Cell Proliferation Assay (Promega) was used for evaluating cell proliferation, according to the manufacturer's instructions. Briefly, 20 μl of CellTiter 96 AQueous One Solution reagent was added into each well of the plate, followed by incubation for 2 hours. Absorbance at 490 nm was then measured using a 96-well microplate reader. The fold increase in mean cell number was calculated according to the formula: fold increase = mean absorbance value in a single well/mean absorbance value in unstimulated cell wells.

Statistical analysis.

Levels of peptide peaks, clinical data, and age of the patients were expressed as the mean ± SD. Student's t-test was used for comparisons between the mean values. Chi-square test was used to evaluate differences in frequency. The relationship of DRC3f levels with clinical and laboratory data were analyzed using linear correlation analysis. Effects of stimulation by C3f and DRC3f on cell proliferation as well as on production of TGFβ1, VEGF, and EGF were expressed as the relative change from baseline (nonstimulated values), using repeated-measures analysis of variance and Student-Newman-Keuls tests for pairwise comparisons. P values less than 0.05 were considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Detection of predominant peptide peaks in SSc sera.

To find disease-related peptides in sera from patients with SSc, we first purified peptides from the serum samples using MS-HIC18 magnetic beads, and then detected profiles of the purified peptides in each of the serum samples by MALDI-TOF mass spectrometry. We compared the peptide profiles between patients with SSc and patients with other diseases (SLE, RA, and OA), using CPT analysis. As a result, several peptide peaks, such as those with mass/charge (m/z) values of 1,865, 1,778, 1,691, 1,563, and 1,450, were detected predominantly in the SSc sera (Figure 1).

thumbnail image

Figure 1. Spectra of serum peptide peaks in patients with systemic sclerosis (SSc) as compared with healthy controls (HC) (a), patients with systemic lupus erythematosus (SLE) (b), patients with rheumatoid arthritis (RA) (c), or patients with osteoarthritis (OA) (d). The mean peak intensities for the peptide spectra were compared using ClinPro Tools software, in 40 serum samples from SSc patients, 30 from SLE patients, 21 from RA patients, 30 from OA patients, and 26 from healthy control subjects. A range of mass/charge (m/z) values, from 1,000 daltons to 3,000 daltons, is shown. Upper panels show comparisons of the mean peak spectra between groups, with results expressed in arbitrary units (arb. u). Lower panels show the peptide peak spectra in the serum of each patient by 2-dimensional density plot in Pseudo Gel View; the peptide peak intensities were converted to a gel separation image to facilitate visualization. Spect. = spectrum.

Download figure to PowerPoint

We next compared the mean peak intensity levels between patients with SSc and the other groups of patients or healthy controls. As shown in Figure 2a, the mean intensities of peptide peaks for the peptides with m/z of 1,450, 1,563, 1,778, and 1,865 were significantly higher in SSc sera than in sera from SLE, RA, and OA patients or healthy controls.

thumbnail image

Figure 2. Genetic algorithms analysis of peptide peaks. a, Comparison of mean peak peptide areas between patients with SSc and the other groups, for peptide peak areas with m/z of 1,450, 1,563, 1,691, 1,778, 1,865, and 2,021, by ClinPro Tools paired analysis. b, Comparison of the frequency of peaks for peptides with m/z of 1,450, 1,563, 1,691, 1,778, 1,865, and 2,021 between patients with SSc and the other groups. Frequency values are shown in parentheses. The solid horizontal line indicates the cutoff point of 100 arbitrary units (the mean peak intensity + 3 SD in healthy controls). Circles indicate individual values. Broken horizontal lines indicate the mean arbitrary units for each group. ∗ = P < 0.05; ∗∗ = P < 0.01. See Figure 1 for definitions.

Download figure to PowerPoint

To further evaluate the levels of these peaks, we compared the frequency of peptide-positive serum samples between SSc patients and the healthy controls and patient groups by assessing each of the peptide peaks for peptides with m/z of 1,450, 1,563, 1,691, 1,778, and 1,865. As shown in Figure 2b, the frequencies of peptide-positive serum samples for peptides with m/z of 1,450, 1,563, and 1,778 were significantly higher in the SSc group than in all other groups. For peptides with m/z of 1,691 and 1,865, the frequencies of peptide-positive serum samples were significantly higher in the SSc patients than in the RA patients, OA patients, and healthy control subjects; however, the differences in these peptide frequencies between patients with SSc and patients with SLE were not statistically significant. In addition, only small differences in mean peak intensity and frequency of peptide positivity for the peptide with m/z of 2,021 were observed among the tested groups.

We then calculated the fold increase in mean peak area of these peptide peaks among the SSc, SLE, RA, and OA groups (Table 1). The results further confirmed that the increases in these peptide peaks were highest in SSc sera. In particular, the mean peak intensities for peptides with m/z of 1,865 and 1,778 in the SSc sera were >18 times as high as those in the healthy control sera. Our findings suggest that these predominant peptides could be linked to the pathophysiologic processes of SSc.

Table 1. Peptides identified by mass spectrometry and the fold increase in mean peak area in the sera of patients with systemic sclerosis (SSc) compared with patients with other rheumatic diseases*
Peptide m/zPeptide familyAmino acid sequenceAmino acid positionAccession no.Mean peak area fold increase versus healthy controls
StartEndSScSLERAOA
  • *

    m/z = mass/charge; SLE = systemic lupus erythematosus; RA = rheumatoid arthritis; OA = osteoarthritis; ITIH4 = inter-α trypsin inhibitor heavy-chain H4 precursor.

  • P < 0.05 versus SLE, RA, and OA groups.

  • P < 0.05 versus RA and OA groups.

  • §

    P < 0.05 versus SLE and OA groups.

1,449.77Complement C3fTHRIHWESASLL5161413205A7.632.121.850.67
1,562.86Complement C3fITHRIHWESASLL4161413205A8.564.031.130.46
1,690.98Complement C3fKITHRIHWESASLL3161413205A16.1412.210.461.60
1,777.96Complement C3fSKITHRIHWESASLL2161413205A18.135.771.521.19
1,865.01Complement C3fSSKITHRIHWESASLL1161413205A18.987.021.020.76
2,021.11Complement C3fSSKITHRIHWESASLLR1171413205A1.241.060.280.45
2,551.25Complement C4TLEIPGNSDPNMIPDGDFNSYVR957979AAB679800.970.740.880.96
2,565.55Complement C4TLEIPGNSDPNMIPDGDFNSYVR+ Me-ester (DE)957979AAB679808.542.271.820.29
1,895.96Complement C4NGFKSHALQLNNRQIR1,3371,352AAB679803.59§1.762.440.27
1,530.08ClusterinRPHFFFPKSRIV214226P1090911.42.182.790.94
2,184.08ITIH4QLGLPGPPDVPDHAAYHPFR669688Q146244.940.391.440.41
2,229.04AlbuminDAHKSEVAHRFKDLGEENF2543AAH392353.441.491.050.99
1,927.86Apolipoprotein A-IV precursorSLAELGGHLDQQVEEFR288304AAA967312.841.831.041.47
2,755.43Apolipoprotein A-IV precursorGNTEGLQKSLAELGGHLDQQVEEFR280304AAA967315.274.481.150.81

Identification of SSc-predominant peptides as derivatives of C3f.

We next determined the amino acid sequences of the detected peptides by MALDI-TOF mass spectrometry, followed by a de novo sequencing method. As a result, the peptide with m/z of 1,865 was identified as DRC3f, a degraded derivative of C3f produced by removal of the C-terminal arginine with carboxypeptidase N (21). The other peptides, with m/z of 1,778, 1,691, 1,563, and 1,450, were identified as shorter derivatives of DRC3f (Table 1). We also identified the peptide with m/z of 2,021 as C3f itself. It is very interesting that the relative concentration of DRC3f and its degraded peptides, with m/z of 1,778, 1,691, 1,563, and 1,450, were much higher in the SSc group than in the other groups, whereas the relative concentration of C3f itself, with m/z of 2,021, in the SSc group did not differ substantially from that in the other groups (Table 1).

In addition to DRC3f and its degraded peptides, we identified other peptides that were detected predominantly in the SSc sera (shown in Table 1). However, the peak levels of these peptides in the sera were much lower than those of DRC3f. Therefore, in the present study, we focused solely on DRC3f.

Relationship of DRC3f to vascular involvement and disease activity in SSc.

We next investigated whether the elevated levels of DRC3f (m/z 1,865) were associated with clinical features in the patients with SSc. Based on the predefined cutoff point of 100 arbitrary units (as shown in Figure 2b), we classified the patients with SSc into 2 groups according to whether their serum levels of DRC3f were elevated (DRC3f-elevated; n = 28) or were within the normal range (DRC3f-normal; n = 12). We then identified the clinical features that showed an association with elevated DRC3f levels (Table 2).

Table 2. Comparison of clinical and laboratory data between DRC3f-elevated and DRC3f-normal groups of patients with SSc*
 Total (n = 40)DRC3f-normal (n = 12)DRC3f-elevated (n = 28)
  • *

    Except where indicated otherwise, values are the number/total number (%) of patients. Patients were grouped according to whether the levels of complement C3F-des-arginine (DRC3f) were elevated or within the normal range. Groups were compared by chi-square test for frequency differences or Student's t-test for pairwise comparisons of mean values. dcSSc = diffuse cutaneous systemic sclerosis; lcSSc = limited cutaneous systemic sclerosis; ILD = interstitial lung disease; ANA = antinuclear antibody; anti-ssDNA = anti–single-stranded DNA; anti-dsDNA = anti–double-stranded DNA; RF = rheumatoid factor; AU = arbitrary units; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate; SP-D = surfactant protein D; PGI2 = prostaglandin I2; ACEI/ARB = angiotensin-converting enzyme inhibitor/angiotensin II type I receptor blocker; NSAIDs = nonsteroidal antiinflammatory drugs.

  • P < 0.05 versus DRC3f-normal group.

  • P < 0.02 versus DRC3f-normal group.

  • §

    Determined in 6 samples from the DRC3f-normal group and 9 samples from the DRC3f-elevated group.

Clinical   
 Age at disease onset, mean ± SD years49.6 ± 10.154.3 ± 11.947.6 ± 9.2
 Disease duration, mean ± SD years6.83 ± 6.31.25 ± 1.2029.21 ± 7.45
 No. female/no. male39/111/128/0
 dcSSc18 (45)4 (33)14 (50)
 lcSSc22 (55)8 (67)14 (50)
 Skin thickness score, mean ± SD19 ± 9.1515.6 ± 9.520.4 ± 8.61
 Raynaud's phenomenon37 (93)10 (83)27 (96)
 Digital pitting scar11 (28)011 (39)
 Active digital ulcer4 (10)04 (14)
 ILD24 (60)1 (8)23 (82)
 Pulmonary hypertension1 (3)01 (4)
 Esophageal involvement25 (63)3 (25)22 (79)
 Sicca symptoms17 (43)2 (17)15 (54)
 Proteinuria2 (5)02 (7)
 Anemia8 (20)1 (8)7 (25)
Laboratory   
 ANA positive37 (93)10 (83)27 (96)
  Titer, mean ± SD1,739.7 ± 452.7736.67 ± 452.72,169.6 ± 1,771.7
 Anti–topo I positive11/36 (31)1/36 (3)10/36 (28)
  Titer, mean ± SD1.6 ± 0.660.36 ± 0.462.32 ± 1.97
 Anticentromere positive20/30 (67)4/30 (13)16/30 (53)
  Titer, mean ± SD60.1 ± 86.3108.9 ± 83.141.9 ± 58.4
 Anti–U1 RNP positive8/36 (22)0/108/36 (22)
 Anti-ssDNA positive3/26 (12)1/26 (4)2/26 (8)
 Anti-dsDNA positive3/20 (15)1/20 (5)2/20 (10)
 Anti-SSA positive3/36 (8)0/363/36 (8)
 Anti-SSB positive0/360/360/36
 Anti-Sm positive0/330/330/33
 Anti–Jo-1 positive0/160/160/16
 RF, mean ± SD AU/ml44.7 ± 35.427.58 ± 32.755.11 ± 54.8
  >6 AU/ml35 (88)8 (67)27 (96)
 C3, mean ± SD mg/dl90.66 ± 10.75100.46 ± 9.5989.71 ± 8.7
  <60 mg/dl7 (18)07 (25)
 C4, mean ± SD mg/dl20.7 ± 3.3324.05 ± 3.919.45 ± 5.59
  <15 mg/dl7 (18)07 (25)
 CH50, mean ± SD units/ml43.2 ± 9.5347.61 ± 9.241.59 ± 4.78
  >40 units/ml29 (73)8 (67)21 (75)
 CRP, mean ± SD mg/dl0.75 ± 1.620.24 ± 0.3360.57 ± 0.54
  >0.2 mg/dl25 (63)5 (42)20 (71)
 ESR, mean ± SD mm/hour36.9 ± 19.426.33 ± 16.8539.11 ± 24.42
  >30 mm/hour20 (50)6 (50)14 (50)
 IgG, mean ± SD mg/dl1,898.2 ± 432.41,583.8 ± 432.42,033.03 ± 597.12
 IgA, mean ± SD mg/dl406.2 ± 132.5296.75 ± 132.5453.11 ± 174.34
 IgM, mean ± SD mg/dl183.9 ± 67.9164.16 ± 67.9192.39 ± 74.04
 Albumin, mean ± SD gm/dl4.21 ± 0.334.42 ± 0.334.12 ± 0.34
 Creatinine, mean ± SD mg/dl0.57 ± 0.090.56 ± 0.090.57 ± 0.11
 KL-6, mean ± SD§540.9 ± 340.3886.5 ± 620.2500.5 ± 226.7
 SP-D, mean ± SD§119.1 ± 95.3104.01 ± 58.3125.15 ± 82.3
Treatment   
 Prednisone23 (58)6 (50)17 (61)
  Dosage, mean ± SD mg/day6.91 ± 4.568.75 ± 5.126.11 ± 4.1
 Immunosuppressor3 (7.5)1 (8.3)2 (7.1)
  PGI223 (58)2 (17)21 (75)
  ACEI/ARB5 (13)1 (8)4 (14)
  NSAIDs12 (30)2 (17)10 (36)

Our results revealed that disease duration was significantly longer in the DRC3f-elevated group than in the DRC3f-normal group (P < 0.05). Raised levels of DRC3f were observed in 12 (54.5%) of the 22 patients with a disease duration shorter than 5 years compared with 16 (88.9%) of the 18 patients with a disease duration of ≥5 years. However, among only the patients in the DRC3f-elevated group, the mean DRC3f level was higher in those with a disease duration of <5 years than in those with a disease duration of ≥5 years.

In addition, the frequencies of interstitial lung disease (ILD), symptoms of sicca syndrome, and esophageal involvement were significantly higher in the DRC3f-elevated group than in the DRC3f-normal group (all P < 0.05). Interestingly, pitting scars and active digital ulcers were observed in only the DRC3f-elevated group. The frequency of use of prostaglandin I2–maintaining therapy was also higher in the DRC3f-elevated group.

The mean titer of anticentromere antibodies, a biomarker of mild SSc, was significantly higher in the DRC3f-normal group (P < 0.05), whereas the mean titer of anti–topoisomerase I antibodies was significantly higher in the DRC3f-elevated group (P < 0.05). Patients with raised DRC3f levels had a higher mean skin thickness score, but the difference compared with patients with DRC3f levels in the normal range was not statistically significant. These results suggest that DRC3f may be associated with the diffuse form of SSc as well as with long-lasting disease and vascular involvement in SSc.

With regard to laboratory data, the DRC3f-elevated group showed decreased levels of serum C3 and C4 more frequently than did the DRC3f-normal group (Table 2) (both P < 0.05). Moreover, the mean concentrations of serum C3, C4, and CH50 were significantly lower in the DRC3f-elevated group than in the DRC3f-normal group (all P < 0.05) (Table 2). DRC3f levels were negatively correlated with the levels of C3 and C4 (r = −0.423 and −0.397, respectively, both P < 0.05) (results not shown). These findings suggest that DRC3f could be an indicator of activation of the complement system in SSc.

In addition, the concentration of CRP was higher and the ESR was more elevated in patients with raised DRC3f levels than in patients with normal-range DRC3f levels (both P < 0.05) (Table 2). DRC3f levels were positively correlated with the ESR (r = 0.272, P < 0.05) (results not shown).

Finally, we found that the serum IgG and IgA levels were much higher in the DRC3f-elevated group than in the DRC3f-normal group of SSc patients (Table 2). DRC3f levels were positively correlated with IgA levels (r = 0.432, P < 0.05) (results not shown). These findings suggest a role for DRC3f in SSc disease activity.

We next analyzed the relationship between DRC3f levels and disease activity in patients with SLE and patients with RA. We found that the DRC3f level was elevated in 16 patients with SLE (53.3%) (Figure 2), and that the DRC3f-elevated group of SLE patients had lower levels of serum C3 and C4 than did the DRC3f-normal group of SLE patients. However, the SLEDAI score was not correlated with DRC3f levels. In fact, DRC3f levels in SLE patients with a SLEDAI score >21 were lower than those observed in SLE patients with SLEDAI scores <10 or 10–20 (mean ± SD DRC3f arbitrary units 84.8 ± 15.2 in patients with a SLEDAI score >21 versus 209.4 ± 22.2 and 287.4 ± 325.31 in patients with a SLEDAI score <10 and SLEDAI scores of 10–20, respectively; both P < 0.05).

Among patients with RA, the levels and frequency of DRC3f were much lower than in patients with SSc or patients with SLE (Figure 2). We found no significant difference in the clinical features and RA-related laboratory findings between the DRC3f-elevated and DRC3f-normal groups of patients with RA (results not shown).

Promotion of HMVEC proliferation in vitro by DRC3f and C3f.

In a previous study, the hexapeptide HWESAS, which is included in the DRC3f amino acid sequence, was found to promote mitogenic activities of IGF (22). We thus expected that DRC3f might promote cell proliferation via IGF, and thereby play a role in the pathogenesis of SSc. Because skin fibroblasts and HMVECs have been demonstrated to play crucial roles in SSc, we investigated the proliferative effects of synthesized DRC3f and C3f peptides on these cells.

We found that both DRC3f and C3f enhanced proliferation of HMVECs of both a dermal cell line and a lung cell line, in the presence and in the absence of IGF (Figures 3a and b). However, neither DRC3f nor C3f affected proliferation of normal human dermal fibroblasts, even in the presence of IGF (Figures 3a and b). In addition, DRC3f showed a stronger proliferative effect than C3f (Figure 3c). These results suggest that DRC3f and C3f promote proliferation of HMVECs, but not fibroblasts.

thumbnail image

Figure 3. Influence of C3f and C3f-des-arginine (DRC3f) on cell proliferation and production of transforming growth factor β1 (TGFβ1) by normal human adult dermal microvascular endothelial cells (ADM), normal human lung microvascular endothelial cells (LME), and normal human dermal fibroblasts (NHDF). In proliferation tests, cells were cultivated in 96-well plates (3,000 cells per well) and stimulated for 48 hours with a serially diluted concentration of complement C3f (a) or DRC3f (b) in the absence or presence of insulin-like growth factor 1 (IGF-1) (20 ng/ml). Fibrinopeptide A (FPA) was used as a control. ∗ = P < 0.05 between 2 adjacent concentrations of C3f or DRC3f. The influence on cell proliferation was compared between C3f and DRC3f (c). ∗ = P < 0.05, C3f versus DRC3f stimulation with or without IGF. Values in a–c are the mean and SD fold increase in mean cell number in a single well for each concentration of stimulant (6 wells for each concentration) relative to that in unstimulated wells. To determine the influence of C3f and DRC3f on production of growth factors, the cells were cultured in 12-well plates and stimulated with different concentrations of C3f and DRC3f for 48 hours. The concentration of TGFβ1 (d), vascular endothelial growth factor (not shown), and epidermal growth factor (not shown) was determined in supernatants by enzyme-linked immunosorbent assay. The centrifuged supernatants of ADM and LME cells were diluted 1:4, while the supernatant of NHDFs was used directly after centrifuging without dilution. The minimum detectable dose of TGFβ1 was <7.0 pg/ml. Values in d are the mean and SD. ∗ = P < 0.05 between 2 adjacent concentrations of C3f or DRC3f.

Download figure to PowerPoint

To confirm the above-described effect of serum DRC3f on cell proliferation, we next stimulated adult dermal and lung HMVECs with DRC3f-elevated and DRC3f-normal sera from patients with SSc, patients with SLE, and healthy donors. As a result, the DRC3f-elevated sera enhanced proliferation of both HMVEC lines to a much greater extent than did the DRC3f-normal sera (Figure 4). This result was reproduced when sera that contained only low molecular weight molecules were used. Specifically, for the latter, we filtered out and removed molecules with molecular weights larger than ∼3 kd, and then checked the effects with the low molecular weight peptide–containing sera. As a result, the filtered sera from the DRC3f-elevated group still enhanced proliferation of the cells to a greater extent than did the similarly filtrated DRC3f-normal sera (Figure 4). These results suggest that DRC3f functions in a manner similar to a growth factor.

thumbnail image

Figure 4. Proliferation of human microvascular endothelial cells (HMVECs) stimulated with either fresh or filtered sera from patients with systemic sclerosis (SSc) as compared with patients with systemic lupus erythematosus (SLE) and healthy controls (HC). Lung HMVECs (3,000 cells/well) were stimulated with fresh sera (5 μl/well) (a) or filtered sera (15 μl/well, filtered with a 3.0-kd filter)(b) from each group (n = 3 per group). Serum samples in which C3f-des-arginine (DRC3f) was in the normal range (DRC3f-normal; open bars) and serum samples in which DRC3f was detected at high levels (DRC3f-elevated; solid bars) were used to stimulate the cells. Values are the mean and SD fold increase in mean cell number relative to that in unstimulated (Unst) wells. ∗ = P < 0.05; ∗∗ = P < 0.01 versus DRC3f-normal sera. Levels of DRC3f and its derivatives were assessed by matrix-assisted laser desorption ionization–time-of-flight (MALDI-TOF) mass spectrometry, and results were compared between DRC3f-normal and DRC3f-elevated filtered serum samples (c). The peptides in 3.0-kd filtered sera were separated and concentrated using Ziptip-C18 and then eluted with 50% acetonitrile. The eluate was loaded on the AnchorChip target plate and analyzed by MALDI-TOF mass spectrometry. The representative MALDI peak spectra of paired DRC3f-normal and DRC3f-elevated samples from each group are shown. The peak at mass/charge 1,865 (DRC3f) and its smaller fragments were detected in the filtered DRC3f-elevated sera, but not in the filtered DRC3f-normal sera.

Download figure to PowerPoint

Increased production of TGFβ1 in endothelial cells by DRC3f and C3f.

We next investigated whether the growth factor–like function of both DRC3f and C3f was mediated by other growth factors secreted by the DRC3f- or C3f-stimulated cells. We measured concentrations of TGFβ1, VEGF, and EGF in the supernatant of these cell lines, with or without stimulation by DRC3f and C3f. As a result, both DRC3f and C3f enhanced TGFβ1 production by adult dermal HMVECs, and the highest concentration of TGFβ1 reached was 125 ng/ml. This maximal concentration of TGFβ1 stimulated by DRC3f and C3f was consistent with the maximal proliferative response of adult dermal HMVECs (as shown in Figure 3b) (F = 6.43 and F = 36.83, respectively; both P < 0.01).

In contrast, DRC3f and C3f did not enhance release of TGFβ1 from lung HMVECs (F = 0.93 and F = 2.08, respectively; both P < 0.05) (Figure 3d). Similarly, the line of normal human dermal fibroblasts used in our experiments produced very little TGFβ1, in spite of the addition of DRC3f and C3f.

VEGF and EGF were not produced by any of the 3 cell lines, even with stimulation by DRC3f and C3f (results not shown). Thus, the positive effects of DRC3f and C3f on proliferation of lung HMVECs appeared to be independent of their effects on growth factors, even though the DRC3f- and C3f-induced proliferation of adult dermal HMVECs was dependent, at least in part, on the increase in TGFβ1 production. These findings indicate that DRC3f appears to work directly as a growth factor–like molecule and, in part, indirectly via the induction of TGFβ1.

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

This study is the first to comprehensively survey disease-related short peptides in patients with SSc, since there have been no efficient ways to approach this in a comprehensive manner thus far. By combining purification of short peptides with C18-bound magnetic beads and peptidomics analysis by a mass spectrometry–based technique, we successfully detected and identified a series of short peptides that exist predominantly in the sera of patients with SSc. Our findings were as follows. 1) The short peptides detected predominantly in patients with SSc as compared with those with SLE, RA, or OA and healthy donors were identified as DRC3f, a fragment produced by the inactivation process of C3b, and its degraded derivatives. 2) The relative concentration of DRC3f was related to several clinical features and parameters of vascular involvement and disease activity in SSc. 3) Functionally, synthesized peptides of DRC3f and C3f were demonstrated to enhance proliferation of HMVECs and increase HMVEC production of TGFβ1.

The complement system has been shown to be involved in the pathogenesis of SSc. Genetically, the null allele of C4A*Q0 was reported to have a strong association with SSc (23, 24). Increased serum concentrations of C1q, C2, C5, C6, C7, C9, and factor B and a decreased concentration of C4 have been observed in patients with SSc (25). Increased concentrations of C3d and C4d and increased ratios of C3d:C3 and C4d:C4 were reported to have a positive relationship with the severity of SSc (26). Another study also showed that an increased C3d level and markedly decreased complement level were dependent on prevention of immune precipitation in patients with SSc (27). In that study, immune precipitation was suppressed significantly in patients who carried the C4A*Q0 allotype (27). More recently, the activated complement complex C5b-9 and the C5a receptor were detected in the microvasculature of SSc patients, both in the early and in the late stages of the disease (28, 29).

These findings suggest that both abnormal complement activation and complement heredity are involved in the pathogenesis of SSc. Activated complement may impair the membranes of endothelial cells directly by C5b-9, resulting in endothelial cell death and increased permeability of the endothelium. Furthermore, fragments released from complement activation, such as C3a and C5a, are strong chemotaxins, which attract leukocytes into sites of inflammation (30). Other fragments produced in the process of activation/inactivation of complement, such as C3b, iC3b, and C3d, may have biologic functions. However, it remains uncertain whether these degraded fragments of complement play a pathogenic role in SSc.

Complement C3 is a key molecule in the formation of C5 convertase. Complement activation includes 3 main pathways of the classical pathway, mannose-binding lectin pathway, and alternative pathway. In healthy conditions, only low-level spontaneous C3 activation occurs, via the alternative pathway. C3b, produced by C3 activation, is degraded to form inactive iC3b with the help of complement receptor 1 (CR1) or factor H, while, simultaneously, C3f of a 17-amino-acid peptide is cleaved from C3b. C3f is further degraded to form DRC3f via carboxypeptidase N, which leads to release of the C-terminal arginine, the main form of C3f derivatives (20). Thus, DRC3f might also be a marker of complement activation. In fact, our results showed that the dramatic increase in serum levels of DRC3f in SSc patients had a negative correlation with the serum C3 and C4 levels.

However, it remains unclear why the DRC3f levels were much higher in patients with SSc than in patients with SLE, the classic immune complex disease. The increased basic levels of complement components, including C3, in those with SSc might have partially contributed to elevations in the DRC3f level (25), whereas the reduced levels of complement regulators, including CR1, and quick depletion of the complement components (31) might have contributed to the lower level of DRC3f in those with SLE. It would be of value to investigate the activity of complement regulatory proteins, including CR1, factor H, and factor I, in SSc.

We found that the DRC3f level had a significant correlation with the level of IgA. In a previous study, IgA-containing immune deposits were detected at vascular sites in patients with SSc (32). Moreover, IgA in immune aggregates deposited at vascular sites was demonstrated to trigger activation of the complement cascade through the alternative pathway (33, 34). Both of these features of IgA may play important pathologic roles in SSc. Functionally, both C3f and DRC3f have been demonstrated to enhance vascular permeability and induce smooth muscle contraction as a weak spasmogen (20). The hexapeptide HWESAS, a sequence encompassed within C3f, has recently been found to potentiate the sulfation and mitogenic activities of the IGFs (22). Based on these findings in addition to our own results, we hypothesized that DRC3f could play a role in the pathogenesis of SSc.

To test this point, we synthesized both C3f peptides and DRC3f peptides and observed the functions of both peptides in vitro. Our results showed that the synthesized C3f and DRC3f promoted proliferation of HMVECs independent of any role of IGF. In addition, we showed that both the DRC3f-containing serum samples and the filtered DRC3f-containing serum samples (containing only molecules with molecular weights of lower than ∼3 kd) enhanced the proliferation of HMVECs (Figure 4). These results suggest that DRC3f is one of the low molecular weight growth factors in sera (22, 35). However, morphologic abnormalities in the vessels, defective angiogenesis, and apoptosis instead of proliferation of endothelial cells are common features of the microvasculature in SSc. This appears to be consistent with the increased DRC3f levels observed in the present study.

In contrast with the findings in skin HMVECs, lung HMVECs have been demonstrated to proliferate in bleomycin-induced pulmonary fibrosis in rats (36), in SSc patients with fibrosing alveolitis (37, 38), and in SSc patients with pulmonary hypertension (39). We found that HMVECs from a lung cell line showed a higher frequency of proliferative responses to DRC3f and C3f than did HMVECs from a skin cell line. Moreover, the DRC3f level was associated with ILDs in patients with SSc. Thus, DRC3f may play a role in the pathogenesis of SSc in association with ILDs and/or pulmonary hypertension.

It has been demonstrated that growth factors, including TGFβ, are responsible for proliferation of dermal fibroblasts and small artery smooth muscle cells, and for excessive production of extracellular matrix components such as types I, III, V, and VII collagens and fibronectin (40). We therefore also investigated the influence of DRC3f and C3f on the production of TGFβ1, VEGF, and EGF, using the same cell lines as above. Both DRC3f and C3f markedly enhanced production of TGFβ1 by the HMVECs of a skin cell line, but not by the lung HMVECs or a human dermal fibroblast cell line. None of the 3 cell lines were found to produce detectable amounts of VEGF and EGF after stimulation with either DRC3f or C3f. Thus, in our study, DRC3f and C3f failed to show a growth factor–like potential, at least with regard to the expression of TGFβ1, VEGF, and EGF, even though a part of this potential could be attributed to expression of TGFβ1 in one of the tested cell lines. Alternatively, DRC3f and C3f may have a direct influence on HMVECs. To resolve this issue, we are currently investigating target molecules of DRC3f and C3f, using a comprehensive proteomics approach.

With regard to clinical features, the DRC3f level was associated with higher levels of antinuclear antibodies, rheumatoid factor, IgG, IgA, and CRP, lower levels of C3, C4, and CH50, an accelerated ESR, and presence of anti–topoisomerase I and anti-RNP antibodies. These findings indicate that DRC3f could be linked to pathologic immune reactions in SSc. In addition, the DRC3f level was associated with ILDs, digital pitting scars, sicca symptoms, and esophageal involvement. DRC3f therefore appears to be linked to the activity and severity of SSc, including vascular involvement. However, the DRC3f level did not correlate significantly with the modified Rodnan total skin thickness score, which is a measure of the main feature of SSc.

In addition, the level of DRC3f was associated with long disease duration, but the activity of SSc is often higher in the early stages of the disease. It is also puzzling that DRC3f levels were associated with the presence of ILDs but not with KL-6 and surfactant protein D. Therefore, we must conclude, based on the available evidence, that the linkage of DRC3f to the activity and severity of SSc might be only partial. To resolve this issue, future investigations into the mechanisms by which DRC3f contributes to the clinical features of SSc are needed.

In summary, DRC3f and its smaller derivatives, produced in the process of inactivation of C3b, were detected predominantly in the sera of patients with SSc in association with disease activity and vascular involvement. We further demonstrated DRC3f to be a low molecular weight growth factor that enhanced proliferation of HMVECs and, in some cells, enhanced TGFβ1 production. These results suggest that the activation of complement and the effects of DRC3f each play a crucial role in the pathogenesis of SSc. Future studies should focus on the functions of DRC3f in SSc sera, since this would provide us with valuable information regarding the pathophysiologic mechanisms of the disease.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Dr. Xiang had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Xiang, Nakamura, Masuko, Yudoh, Nishioka, Kato.

Acquisition of data. Matsui, Matsuo, Shimada, Tohma.

Analysis and interpretation of data. Xiang, Matsui, Matsuo, Shimada, Tohma, Nakamura, Masuko, Yudoh, Nishioka, Kato.

Manuscript preparation. Xiang, Kato.

Statistical analysis. Xiang, Kato.

Acknowledgements

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

We thank Ms Mie Kanke and Ms Mayumi Tamaki for their technical assistance.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES
  • 1
    Rodnan GP, Myerowitz RL, Justh GO. Morphologic changes in the digital arteries of patients with progressive systemic sclerosis (scleroderma) and Raynaud phenomenon. Medicine (Baltimore) 1980; 59: 393408.
  • 2
    Pignone A, Scaletti C, Matucci-Cerinic M, Vazquez-Abad D, Meroni PL, Del Papa N, et al. Anti-endothelial cell antibodies in systemic sclerosis: significant association with vascular involvement and alveolo-capillary impairment. Clin Exp Rheumatol 1998; 16: 52732.
  • 3
    Kantor TV, Friberg D, Medsger TA Jr, Buckingham RB, Whiteside TL. Cytokine production and serum levels in systemic sclerosis. Clin Immunol Immunopathol 1992; 65: 27885.
  • 4
    Sato S, Hanakawa H, Hasegawa M, Nagaoka T, Hamaguchi Y, Nishijima C, et al. Levels of interleukin 12, a cytokine of type 1 helper T cells, are elevated in sera from patients with systemic sclerosis. J Rheumatol 2000; 27: 283842.
  • 5
    Sato S, Nagaoka T, Hasegawa M, Tamatani T, Nakanishi T, Takigawa M, et al. Serum levels of connective tissue growth factor are elevated in patients with systemic sclerosis: association with extent of skin sclerosis and severity of pulmonary fibrosis. J Rheumatol 2000; 27: 14954.
  • 6
    Yamane K, Miyauchi T, Suzuki N, Yuhara T, Akama T, Suzuki H, et al. Significance of plasma endothelin-1 levels in patients with systemic sclerosis. J Rheumatol 1992; 19: 156671.
  • 7
    Nordenbaek C, Johansen JS, Halberg P, Wiik A, Garbarsch C, Ullman S, et al. High serum levels of YKL-40 in patients with systemic sclerosis are associated with pulmonary involvement. Scand J Rheumatol 2005; 34: 2937.
  • 8
    Sato S, Fujimoto M, Hasegawa M, Komura K, Yanaba K, Hayakawa I, et al. Serum soluble CTLA-4 levels are increased in diffuse cutaneous systemic sclerosis. Rheumatology (Oxford) 2004; 43: 12616.
  • 9
    Sato S, Komura K, Hasegawa M, Fujimoto M, Takehara K. Clinical significance of soluble CD31 in patients with systemic sclerosis (SSc): association with limited cutaneous SSc. J Rheumatol 2001; 28: 24605.
  • 10
    Feghali-Bostwick CA. Genetics and proteomics in scleroderma. Curr Rheumatol Rep 2005; 7: 12934.
  • 11
    Schrader M, Schulz-Knappe P. Peptidomics technologies for human body fluids. Trends Biotechnol 2001; 19 Suppl 10: S5560.
  • 12
    Schrader M, Selle H. The process chain for peptidomic biomarker discovery. Dis Markers 2006; 22: 23545.
  • 13
    Soloviev M, Finch P. Peptidomics: current status. J Chromatogr B Analyt Technol Biomed Life Sci 2005; 815: 1124.
  • 14
    Subcommittee for Scleroderma Criteria of the American Rheumatism Association Diagnostic and Therapeutic Criteria Committee. Preliminary criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum 1980; 23: 58190.
  • 15
    Clements P, Lachenbruch P, Siebold J, White B, Weiner S, Martin R, et al. Inter and intraobserver variability of total skin thickness score (modified Rodnan TSS) in systemic sclerosis. J Rheumatol 1995; 22: 12815.
  • 16
    Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982; 25: 12717.
  • 17
    Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang DH, and the Committee on Prognosis Studies in SLE. Derivation of the SLEDAI: a disease activity index for lupus patients. Arthritis Rheum 1992; 35: 63040.
  • 18
    Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of criteria for the classification and reporting of osteoarthritis: classification of osteoarthritis of the knee. Arthritis Rheum 1986; 29: 103949.
  • 19
    Zhang X, Leung SM, Morris CR, Shigenaga MK. Evaluation of a novel, integrated approach using functionalized magnetic beads, bench-top MALDI-TOF-MS with prestructured sample supports, and pattern recognition software for profiling potential biomarkers in human plasma. J Biomol Tech 2004; 15: 16775.
  • 20
    Ganu VS, Muller-Eberhard HJ, Hugli TE. Factor C3f is a spasmogenic fragment released from C3b by factors I and H: the heptadeca-peptide C3f was synthesized and characterized. Mol Immunol 1989; 26: 93948.
  • 21
    Harrison RA, Farries TC, Northrop FD, Lachmann PJ, Davis AE. Structure of C3f, a small peptide specifically released during inactivation of the third component of complement. Complement 1988; 5: 2732.
  • 22
    Dousset B, Straczek J, Maachi F, Nguyen DL, Jacob C, Capiaumont J, et al. Purification from human plasma of a hexapeptide that potentiates the sulfation and mitogenic activities of insulin-like growth factors. Biochem Biophys Res Commun 1998; 247: 58791.
  • 23
    Briggs DC, Welsh K, Pereira RS, Black CM. A strong association between null alleles at the C4A locus in the major histocompatibility complex and systemic sclerosis. Arthritis Rheum 1986; 29: 12747.
  • 24
    Takeuchi F, Nabeta H, Hong GH, Matsuta K, Tokunaga K, Tanimoto K, et al. C4A and C4B null alleles are genetic markers of different types of systemic sclerosis in Japanese patients. Clin Exp Rheumatol 1998; 16: 5560.
  • 25
    Benbassat C, Schlesinger M, Luderschmidt C, Valentini G, Tirri G, Shoenfeld Y. The complement system and systemic sclerosis. Immunol Res 1993; 12: 3126.
  • 26
    Senaldi G, Lupoli S, Vergani D, Black CM. Activation of the complement system in systemic sclerosis: relationship to clinical severity. Arthritis Rheum 1989; 32: 12627.
  • 27
    Arason GJ, Geirsson AJ, Kolka R, Vikingsdottir T, Valdimarsson H. Deficiency of complement-dependent prevention of immune precipitation in systemic sclerosis. Ann Rheum Dis 2002; 61: 25760.
  • 28
    Magro CM, Nuovo G, Ferri C, Crowson AN, Giuggioli D, Sebastiani M. Parvoviral infection of endothelial cells and stromal fibroblasts: a possible pathogenetic role in scleroderma. J Cutan Pathol 2004; 31: 4350.
  • 29
    Sprott H, Muller-Ladner U, Distler O, Gay RE, Barnum SR, Landthaler M, et al. Detection of activated complement complex C5b-9 and complement receptor C5a in skin biopsies of patients with systemic sclerosis (scleroderma). J Rheumatol 2000; 27: 4024.
  • 30
    Sallusto F, Mackay CR. Chemoattractants and their receptors in homeostasis and inflammation. Curr Opin Immunol 2004; 16: 72431.
  • 31
    Ross GD, Yount WJ, Walport MJ, Winfield JB, Parker CJ, Fuller CR, et al. Disease associated loss of erythrocyte complement receptors (CR1, C3b receptors) in patients with SLE and other diseases involving autoantibodies and/or complement activation. J Immunol 1985; 135: 200514.
  • 32
    Evans DJ, Cashman SJ, Walport M. Progressive systemic sclerosis: autoimmune arteriopathy. Lancet 1987; 1: 4802.
  • 33
    Schifferli JA, Steiger G, Polla L, Didierjean L, Saurat JH. Activation of the alternative pathway of complement by skin immune deposits. J Invest Dermatol 1985; 85: 40711.
  • 34
    Hiemstra PS, Gorter A, Stuurman ME, Van Es LA, Daha MR. Activation of the alternative pathway of complement by human serum IgA. Eur J Immunol 1987; 17: 3216.
  • 35
    Heulin MH, Rajoelina J, Artur M, Geschier C, Straczek J, Lasbennes A, et al. Isolation and characterization of a low molecular weight growth-promoting factor from human plasma. Life Sci 1987; 41: 297304.
  • 36
    Kawamoto M, Fukuda Y. Cell proliferation during the process of bleomycin-induced pulmonary fibrosis in rats. Acta Pathol Jpn 1990; 40: 22738.
  • 37
    Renzoni EA, Walsh DA, Salmon M, Wells AU, Sestini P, Nicholson AG, et al. Interstitial vascularity in fibrosing alveolitis. Am J Respir Crit Care Med 2003; 167: 43843.
  • 38
    Fujimoto T, Matsumoto T, Shoji S. Disorganizing-fibrosing processes in alveolar walls of interstitial pneumonia, “alveolopneumonitis”: a morphopathological study. Tohoku J Exp Med 1989; 158: 23751.
  • 39
    Ramirez A, Varga J. Pulmonary arterial hypertension in systemic sclerosis: clinical manifestations, pathophysiology, evaluation, and management. Treat Respir Med 2004; 3: 33952.
  • 40
    Postlethwaite AE. Connective tissue metabolism including cytokines in scleroderma. Curr Opin Rheumatol 1995; 7: 53540.