Search

Additional file 2: Supplementary Figure 2. Comparisons of pentameric C-reactive protein (pCRP) (A–C), monomeric (m)CRP (D–F), and mCRP/pCRP ratio (G–I) demonstrated between normal vs. subnormal estimated glomerular filtration rate (eGFR) (A, D, G), presence vs. absence of hematuria (B, E, H) and presence vs. absence albuminuria (C, F, I) in patients with ANCA-associated vasculitis. Abnormal eGFR

Additional file 3: Supplementary Figure 3. Standard curves and reactivity of the monoclonal detection antibody 8C10 towards monomeric C-reactive protein (mCRP) and pentameric (p)CRP: (A) standard curve for 8C10 antibody with mCRP; (B) reactivity of 8C10 antibody with pCRP; and (C) standard curve for samples in the present study.

Additional file 1: Supplementary Figure 1. Comparisons of pentameric C-reactive protein (pCRP) (A–D), monomeric (m)CRP (E–H), and mCRP/pCRP ratio (I–L) in 160 patients with systemic lupus erythematosus with lupus nephritis vs. no nephritis (A, E, I), normal vs. subnormal estimated glomerular filtration rate (eGFR) (B, F, J), presence vs. absence of hematuria (C, G, K) and presence vs. absence of albuminuria (D, H, L). Abnormal eGFR

Additional file 3: Supplementary Figure 3. Standard curves and reactivity of the monoclonal detection antibody 8C10 towards monomeric C-reactive protein (mCRP) and pentameric (p)CRP: (A) standard curve for 8C10 antibody with mCRP; (B) reactivity of 8C10 antibody with pCRP; and (C) standard curve for samples in the present study.

Additional file 3: Supplementary Table 3. Baseline characteristics in CCP2-positive and -negative RA. The table show baseline characteristics for the EIRA patients included in the study, based on anti-CCP2 IgG status; p-values indicate differences between subsets with respect to: age, female-to-male ratio, number of smokers, HLA-DRB1 SE-positivity, PTPN22-positivity, DAS28 and CRP.

Additional file 7: Supplementary Table 6. Associations between different ACPA fine-specificities and HLA-DRB1 SE, in anti-CCP2-positive and anti-CCP2-negative RA. Odds ratios with 95% confidence intervals are shown for associations between HLA-DRB1 SE and presence/absence of 19 different ACPA fine-specificities, in anti-CCP2-positive and -negative RA: p-values indicate differences in ORs between ACPA fine-specificity-positive and -negative subsets in the anti-CCP2-positive subset.

Additional file 6: Supplementary Table 5. Associations between different autoantibodies and PTPN22 polymorphism, in anti-CCP2-negative RA. Odds ratios with 95% confidence intervals are shown for associations between PTPN22 polymorphism and presence/absence of ACPA, RF, other autoantibodies, or "any autoantibody" in anti-CCP2-negative RA.

Additional file 6: Supplementary Table 5. Associations between different autoantibodies and PTPN22 polymorphism, in anti-CCP2-negative RA. Odds ratios with 95% confidence intervals are shown for associations between PTPN22 polymorphism and presence/absence of ACPA, RF, other autoantibodies, or "any autoantibody" in anti-CCP2-negative RA.

Additional file 7: Supplementary Table 6. Associations between different ACPA fine-specificities and HLA-DRB1 SE, in anti-CCP2-positive and anti-CCP2-negative RA. Odds ratios with 95% confidence intervals are shown for associations between HLA-DRB1 SE and presence/absence of 19 different ACPA fine-specificities, in anti-CCP2-positive and -negative RA: p-values indicate differences in ORs between ACPA fine-specificity-positive and -negative subsets in the anti-CCP2-positive subset.

Additional file 4: Supplementary Figure 1. ACPA fine-specificity levels in anti-CCP2-negative RA and controls. ACPA levels for 19 individual ACPA fine-specificities are shown in separate graphs with AU values on the Y-axes. Only ACPA levels above cutoff for positivity are shown.

Additional file 1.

Additional file 4: Supplementary Figure 1. ACPA fine-specificity levels in anti-CCP2-negative RA and controls. ACPA levels for 19 individual ACPA fine-specificities are shown in separate graphs with AU values on the Y-axes. Only ACPA levels above cutoff for positivity are shown.

Additional file 8: Supplementary Table 7. Associations between RF isotypes and smoking, in anti-CCP2-positive and anti-CCP2-negative RA. Odds ratios with 95% confidence intervals are shown for associations between smoking and presence/absence of IgM, IgG or IgA RF, in anti-CCP2-positive and -negative RA; p-values indicate differences in ORs between RF isotype-positive and -negative subsets.

Additional file 8: Supplementary Table 7. Associations between RF isotypes and smoking, in anti-CCP2-positive and anti-CCP2-negative RA. Odds ratios with 95% confidence intervals are shown for associations between smoking and presence/absence of IgM, IgG or IgA RF, in anti-CCP2-positive and -negative RA; p-values indicate differences in ORs between RF isotype-positive and -negative subsets.

Illustration of a conditional inference tree for developing RA in pre-symptomatic individuals compared with control subjects. CCP2 anti-CCP2 antibodies, Filaggrin anti-filaggrin antibodies, CEP-1 = anti-α-enolase antibodies, IgM-RF immunoglobulin M rheumatoid factor, HLA-SE human leukocyte antigen shared epitope, n number of individuals, Pr probability of detecting a pre-symptomatic individual, + = positive, − = negative. (DOCX 29 kb)

Prevalence, sensitivity, specificity and OR of numbers of RF isotypes, alone and/or in combination with ten different ACPA specificities, anti-CCP2 antibodies and anti-carbamylated protein antibodies in pre-symptomatic individuals and population controls. (DOCX 17kb)

Illustration of a conditional inference tree for developing RA in pre-symptomatic individuals compared with control subjects. CCP2 anti-CCP2 antibodies, Filaggrin anti-filaggrin antibodies, CEP-1 = anti-α-enolase antibodies, IgM-RF immunoglobulin M rheumatoid factor, HLA-SE human leukocyte antigen shared epitope, n number of individuals, Pr probability of detecting a pre-symptomatic individual, + = positive, − = negative. (DOCX 29 kb)

Figure S1. The correlation between RF-IgM and all proteins analyzed. (PDF 284 kb)

Additional file 1: Supplementary Table 1. Citrullinated peptide antigens on the multiplex microarray. The table lists citrullinated antigens used on the multiplex microarray, including name, protein of origin, and amino acid sequence, as well as references.

A more detailed description of the MS proteomic method is described. (PDF 289 kb)

A more detailed description of the affinity-based proteomic method is described. (PDF 370 kb)

Additional file 5: Supplementary Table 4. Other autoantibodies in anti-CCP2-positive and -negative RA and controls. Frequencies of 17 other autoantibodies in anti-CCP2-positive RA, anti-CCP2-negative RA and controls are shown, as well as median antibody levels in anti-CCP2-positive and -negative RA (p-values indicate differences between anti-CCP2-negative and anti-CCP2-positive RA subsets).

Table S2. Medications for the patients are reported. (PDF 526 kb)

Additional file 9: Supplementary Figure 2. Antibody levels in EIRA RA cases based on smoking status. IgM, IgG and IgA RF levels, as well as anti-CCP2 IgG levels, are shown for never, current and former smokers in all EIRA RA patients where smoking data was available.

Additional file 1: Supplementary Table 1. Citrullinated peptide antigens on the multiplex microarray. The table lists citrullinated antigens used on the multiplex microarray, including name, protein of origin, and amino acid sequence, as well as references.

Additional file 5: Supplementary Table 4. Other autoantibodies in anti-CCP2-positive and -negative RA and controls. Frequencies of 17 other autoantibodies in anti-CCP2-positive RA, anti-CCP2-negative RA and controls are shown, as well as median antibody levels in anti-CCP2-positive and -negative RA (p-values indicate differences between anti-CCP2-negative and anti-CCP2-positive RA subsets).

Additional file 2: Supplementary Table 2. Other (non-citrullinated) autoantigens on the multiplex microarray. The table lists the other antigens used on the multiplex microarray, including name, protein of origin, and major associated disease(s).

Figure S1. The correlation between RF-IgM and all proteins analyzed. (PDF 284 kb)

Additional file 9: Supplementary Figure 2. Antibody levels in EIRA RA cases based on smoking status. IgM, IgG and IgA RF levels, as well as anti-CCP2 IgG levels, are shown for never, current and former smokers in all EIRA RA patients where smoking data was available.

Additional file 1: Figure 1. Sequence, structure and major immunogenic region (mir) of hnRNP-D and hnRNP-DL. A, Schematic representation of hnRNP-D (isoform p45), hnRNP-DL and the different recombinant hnRNP-DL variants studied. The main structural features are highlighted. Mir-region is the major immunogenic region, RBD1 and RBD2 are RNA-binding domains 1 and 2, Gly-rich is the C-terminal glycine-rich region of the proteins. B, Global amino acid sequence alignment of hnRNP-D and hnRNP-DL1 (isoform 1). HnRNP-D and -DL share 89.1% similarity by sequenc e[1]. Regions “mir”, “RBD1”, “RBD2” and “Gly-rich” are highlighted. Figure 2. Characterisation of autoantibodies against, A, citrullinated α-hnRNP-DLmir (cit-DL), B, α-hnRNP-DLmir (DL) and C, ∆OD between cit-DL and DL (ΔDL) determed by ELISA in sera of other diseases (n=127; MS n=20, reA n=7, Sclero n=20, Sjö n=20, PsA n=20, MB n=20, OA n=20). The dotted lines markes the cutoff vs. other diseases (except systemic lupus erythematosus) or healthy controls with 98% specificity each. OD, optical density; nm, nano meter; vs., versus; MS, multiple sclerosis; reA, reactive arthritis; Sclero, scleroderma; Sjö, Sjögren´s syndrome; PsA, psoriasis arthritis; MB, ankylosing spondylitis; OA. Osteoarthritis. Table 1. Mann Whitney U-test of (cit) α-hnRNP-DLmir-OD signals of seropositive and seronegative data sets of RA-cohorts. Table 2. Mann Whitney U-test of cit α-hnRNP-DLmir-OD signals of seronegative data sets of RA-cohorts and data sets of other inflammatory diseases. Figure 3. XY-Plot and Spearman Correlation of citrullinated or native α-hnRNP-DLmir versus ΔhnRNP-DLmir for the early RA cohort EIRA (A/D; n=404), the seropositive EIRA sera (B/E; n=202) and the seronegative EIRA sera (C/F; n=202). Table 3. Spearman correlation of the early RA sera of the EIRA cohort (n=404). The results are given as R value (left of slash) with the corresponding p-value (right of slash). Table 4. Spearman correlation of the 242 EIRA sera treated with MTX (α-CCP2 positive n=133, α-CCP2 negative n=109). The results are given as R value (left of slash) with the corresponding p-value (right of slash). Table 5. Spearman correlation of the established RA sera of the Predict cohort (n=94; RF IgM and/or α-CCP2 positive n=64, RF IgM and α-CCP2 negative n=30). The results are given as R value (left of slash) with the corresponding p-value (right of slash). Table 6. ROC analysis of native hnRNP-DLmir of MTX-treated EIRA patients (n=192; seropositive n=93, seronegative n=99). Table 7. Negative CNDL-index of MTX-treated EIRA patients n=192 (Resp. n=161, non-Resp. n=31). Table 8. ROC analysis of native hnRNP-DLmir of Enbrel®-treated Predict patients (n=94; seropositive n=63, seronegative n=31). Figure 4. High baseline titer against α-hnRNP-DLmir (DL) is rather present in 6-month EULAR Responder RA patients who had received MTX or α-TNF inhibitor therapy (Enbrel®). A-C, Citrullinated α-hnRNP-DLmir (citDL) (A), α-hnRNP-DLmir (DL) (B) and ∆ OD between citDL and DL (ΔDL) (C) were measured by ELISA in patient sera from the EIRA cohort treated with MTX (n=192) with 161 EULAR Responder and 31 EULAR non-Responder among 6 months. The evaluation was done according to the cutoff versus other diseases. D, α-DL were measured by ELISA in patient sera from the Predict cohort treated with α-TNF inhibitor therapy with 6-month EULAR response data (n=94, responder n=63, non-Responder n=31). Based on the signals, a response-cutoff (dotted line, OD 0.174) was determined, from which only responders are recognized as positive. OD, optical density; nm, nano meter; RA, rheumatoid arthritis; SLE, systemic lupus erythematosus; MTX, Methotrexate; Resp., 6-month EULAR Responder. Figure 5. A, Influence of cytokines on hnRNP-DL expression determined by immunoblotting. Cellular extracts from unstimulated, IL1α- or TNFα-stimulated HeLa cells and from unstimulated and IL6-stimulated HepG2 cells were probed with α-hnRNP-DL1/2-peptide specific rabbit serum. B, Citrullination of hnRNP-DL in synovial tissue from a patient with rheumatoid arthritis was investigated with an α-deiminated arginine antibody and an α-hnRNP-DL antibody. Both positive bands were labled with hnRNP-DL, which isoforms were not analysed.

Table S1. The number of missing data points for each variable is shown. (PDF 668 kb)

Additional file 1: Figure 1. Sequence, structure and major immunogenic region (mir) of hnRNP-D and hnRNP-DL. A, Schematic representation of hnRNP-D (isoform p45), hnRNP-DL and the different recombinant hnRNP-DL variants studied. The main structural features are highlighted. Mir-region is the major immunogenic region, RBD1 and RBD2 are RNA-binding domains 1 and 2, Gly-rich is the C-terminal glycine-rich region of the proteins. B, Global amino acid sequence alignment of hnRNP-D and hnRNP-DL1 (isoform 1). HnRNP-D and -DL share 89.1% similarity by sequenc e[1]. Regions “mir”, “RBD1”, “RBD2” and “Gly-rich” are highlighted. Figure 2. Characterisation of autoantibodies against, A, citrullinated α-hnRNP-DLmir (cit-DL), B, α-hnRNP-DLmir (DL) and C, ∆OD between cit-DL and DL (ΔDL) determed by ELISA in sera of other diseases (n=127; MS n=20, reA n=7, Sclero n=20, Sjö n=20, PsA n=20, MB n=20, OA n=20). The dotted lines markes the cutoff vs. other diseases (except systemic lupus erythematosus) or healthy controls with 98% specificity each. OD, optical density; nm, nano meter; vs., versus; MS, multiple sclerosis; reA, reactive arthritis; Sclero, scleroderma; Sjö, Sjögren´s syndrome; PsA, psoriasis arthritis; MB, ankylosing spondylitis; OA. Osteoarthritis. Table 1. Mann Whitney U-test of (cit) α-hnRNP-DLmir-OD signals of seropositive and seronegative data sets of RA-cohorts. Table 2. Mann Whitney U-test of cit α-hnRNP-DLmir-OD signals of seronegative data sets of RA-cohorts and data sets of other inflammatory diseases. Figure 3. XY-Plot and Spearman Correlation of citrullinated or native α-hnRNP-DLmir versus ΔhnRNP-DLmir for the early RA cohort EIRA (A/D; n=404), the seropositive EIRA sera (B/E; n=202) and the seronegative EIRA sera (C/F; n=202). Table 3. Spearman correlation of the early RA sera of the EIRA cohort (n=404). The results are given as R value (left of slash) with the corresponding p-value (right of slash). Table 4. Spearman correlation of the 242 EIRA sera treated with MTX (α-CCP2 positive n=133, α-CCP2 negative n=109). The results are given as R value (left of slash) with the corresponding p-value (right of slash). Table 5. Spearman correlation of the established RA sera of the Predict cohort (n=94; RF IgM and/or α-CCP2 positive n=64, RF IgM and α-CCP2 negative n=30). The results are given as R value (left of slash) with the corresponding p-value (right of slash). Table 6. ROC analysis of native hnRNP-DLmir of MTX-treated EIRA patients (n=192; seropositive n=93, seronegative n=99). Table 7. Negative CNDL-index of MTX-treated EIRA patients n=192 (Resp. n=161, non-Resp. n=31). Table 8. ROC analysis of native hnRNP-DLmir of Enbrel®-treated Predict patients (n=94; seropositive n=63, seronegative n=31). Figure 4. High baseline titer against α-hnRNP-DLmir (DL) is rather present in 6-month EULAR Responder RA patients who had received MTX or α-TNF inhibitor therapy (Enbrel®). A-C, Citrullinated α-hnRNP-DLmir (citDL) (A), α-hnRNP-DLmir (DL) (B) and ∆ OD between citDL and DL (ΔDL) (C) were measured by ELISA in patient sera from the EIRA cohort treated with MTX (n=192) with 161 EULAR Responder and 31 EULAR non-Responder among 6 months. The evaluation was done according to the cutoff versus other diseases. D, α-DL were measured by ELISA in patient sera from the Predict cohort treated with α-TNF inhibitor therapy with 6-month EULAR response data (n=94, responder n=63, non-Responder n=31). Based on the signals, a response-cutoff (dotted line, OD 0.174) was determined, from which only responders are recognized as positive. OD, optical density; nm, nano meter; RA, rheumatoid arthritis; SLE, systemic lupus erythematosus; MTX, Methotrexate; Resp., 6-month EULAR Responder. Figure 5. A, Influence of cytokines on hnRNP-DL expression determined by immunoblotting. Cellular extracts from unstimulated, IL1α- or TNFα-stimulated HeLa cells and from unstimulated and IL6-stimulated HepG2 cells were probed with α-hnRNP-DL1/2-peptide specific rabbit serum. B, Citrullination of hnRNP-DL in synovial tissue from a patient with rheumatoid arthritis was investigated with an α-deiminated arginine antibody and an α-hnRNP-DL antibody. Both positive bands were labled with hnRNP-DL, which isoforms were not analysed.

Table S1. The number of missing data points for each variable is shown. (PDF 668 kb)

A more detailed description of the affinity-based proteomic method is described. (PDF 370 kb)

Table S3. The correlation between RF-IgM and the 25 proteins in Table 2 are shown. (PDF 256 kb)

A more detailed description of the MS proteomic method is described. (PDF 289 kb)

Table S3. The correlation between RF-IgM and the 25 proteins in Table 2 are shown. (PDF 256 kb)

Table S2. Medications for the patients are reported. (PDF 526 kb)