Clin Rheumatol (2014) 33:1523–1526 DOI 10.1007/s10067-014-2764-2
BRIEF REPORT
Association of ferritin antibodies with Takayasu arteritis K. Große & T. Witte & F. Moosig & B. F. Hoyer & C. Lansche & R. E. Schmidt & N. T. Baerlecken
Received: 7 August 2014 / Accepted: 24 August 2014 / Published online: 2 September 2014 # International League of Associations for Rheumatology (ILAR) 2014
Abstract Takayasu arteritis (TA) is difficult to diagnose because diagnostic biomarkers have not yet been established. In a previous study, we detected autoantibodies against the human ferritin heavy chain protein (HFC) in the sera of patients with giant cell arteritis (GCA) and/or polymyalgia rheumatica (PMR). The aim of this study is to evaluate the frequency of autoantibodies against HFC in TA. We established seven ELISA assays for the detection of autoantibodies against HFC. We used the full-length recombinant HFC expressed in Escherichia coli or one of six different HFC peptides as autoantigens: 1-18Aa (98.8 % purity), 19-45Aa (98.8 % purity), 52-78Aa (98.3 % purity), 79-104Aa (98.8 % purity), 105143Aa (98.4 % purity) and 145-183Aa (98.5 % purity). We collected sera from 48 patients with TA, 36 patients with systemic lupus erythematosus (SLE), 35 patients with arteriosclerosis, 133 patients with febrile diseases, which are known to generate unspecific autoantibodies, and 50 blood donors, which served as controls. The best results were obtained using the ferritin peptides as antigens. By combining the results from the different ELISAs that detect autoantibodies against the HFC peptides 19-44A, 79-104A and 105-144A, we were able to detect ferritin peptide antibodies in 30/48 (62 %) of the K. Große : T. Witte : C. Lansche : R. E. Schmidt : N. T. Baerlecken (*) Department of Clinical Immunology and Rheumatology, Medical University Hannover, Carl-Neuberg Str. 1, 30625 Hannover, Germany e-mail: [email protected] F. Moosig Department of Rheumatology and Immunology, Bad Bramstedt/ University of Luebeck, Luebeck, Germany B. F. Hoyer Department of Rheumatology and Clinical Immunology, Charite Free University and Humboldt University Charité University Medicine Berlin, Berlin, Germany
TA patients. The frequency was lower than in early GCA and PMR (previous study showed up to 92 %). Positive results were observed in 0/50 (0 %) of the control blood donors, 10/ 36 (28 %) of the SLE patients, 4/35 (11 %) of the arteriosclerosis patients and 27/133 (20 %) of the fever patients. Considering the lack of biomarkers for TA, autoantibodies against HFC peptides could act as useful markers for TA. Keywords Autoantibodies . Epitope mapping . Ferritin . Takayasu arteritis
Introduction Takayasu arteritis (TA) has an incidence of 2.6 million cases per year in Western countries [1]. It occurs predominantly in females less than 20 years of age. An acute inflammatory phase with systemic and cardiovascular symptoms followed by the chronic course in which the disease slowly progresses occurs in 50 % of cases. Lack of pulse, vascular bruit, elevated blood pressure, heart failure and abnormal fundi sometimes occur. Non-specific symptoms such as asthenia, weight loss, fever, dyspnoea, headache and arthralgias can also occur. Laboratory findings show that greater than 80 % of patients have an elevated erythrocyte sedimentation rate (ESR). Gamma globulin levels may also be elevated. In Western countries, diagnosis is greatly delayed, ranging from 2 to 11 years after the onset of the first symptoms. TA is more common in Asia, which leads to faster diagnosis, ranging, for example, from 2.5 to 5 months in India [2, 3]. Early diagnosis and treatment can avert complications such as aortic regurgitation, congestive heart failure and cerebrovascular events [4, 5]. Contemporary diagnostic methods such as MRI, angiography and Doppler ultrasounds are however either insufficient or quite elaborate. Recently, we characterised antibodies against the human ferritin heavy
1524
chain (HFC) and its N-terminal peptide in the sera of giant cell arteritis (GCA)/polymyalgia rheumatica (PMR) patients. Because TA is also a large vessel vasculitis, we wanted to study whether antibodies against ferritin are a diagnostic marker of TA [6–9].
Clin Rheumatol (2014) 33:1523–1526
3/133 had SLE with sepsis, 2/118 had Sjögren’s syndrome with acute infection, 23/133 had different malignant diseases with acute infection and 35/133 had various infectious diseases without other underlying diseases. The patients with arteriosclerosis were collected from our in- and outpatient clinics. Rheumatoid arthritis was observed in 23/35 patients: 4/35 had SLE and 8/35 had osteoarthritis.
Patients, materials and methods ELISA assays In Hannover, Bad Bramstedt/Luebeck and Berlin (Germany), sera from 48 patients with TA were collected. Patients were selected according to the 1990 ACR criteria [10]. As controls, we tested sera from 74 patients with fever above 39 °C, 36 patients with systemic lupus erythematosus (SLE) and healthy blood donors [11]. The study was approved by our local ethics committee (“Ethik-Kommission der Medizinischen Hochschule Hannover”, head: Prof. Dr. Tröger) (project number 4928). All patients agreed to written informed consent. The demographic, laboratory and disease activity data were supplied from the Department of Clinical Immunology and Rheumatology of the medical University of Hannover and the Department of Rheumatology and Immunology, Bad Bramstedt/University of Luebeck, Department of Rheumatology and Clinical Immunology, Charité University Medicine Free University and Humboldt University of Berlin. All patient data are associated with the date when the blood samples were taken. The proportion of females was 42/48 (87 %) for TA patients, 31/36 (86 %) for SLE patients, 38/118 (32 %) for fever patients, 14/35 (40 %) for patients with arteriosclerosis and 33/50 (66 %) for blood donors. The median (range) age was 42 years (21–64) for TA patients, 43 years (21–74) for SLE patients, 52 years (21– 85) for fever patients, 62 years (42–86) for patients with arteriosclerosis and 37 years (21–37) for blood donors. The mean (SD) C-reactive protein (CRP) level was 15 mg/l (26) for TA patients and 4 mg/l (7) for SLE patients. The mean (SD) erythrocyte sedimentation rate (ESR) was 26 mmnW (20) for TA patients and 22 (21) for SLE patients. CRP was not available in five patients with TA. SLE patients were defined as active by the treating rheumatologist. The mean systemic lupus erythematosus disease activity index (SLEDAI) was 7 with a SD of 5. The mean anti-nuclear antibody (ANA) titre was 1:640, 28/ 36 (78 %), and these sera contained antibodies against dsDNA. GC treatment was administered to 36/36 (100 %) patients and 31/36 (86 %) received DMARDs. A proportion of patients with fever (24/118) had AIDS-related diseases, e.g. tuberculosis, non-Hodgkin lymphoma, toxoplasmosis, Pneumocystis jiroveci pneumonia or soor oesophagitis, 18/ 133 had tuberculosis including the HIV-infected patients, 20/ 133 patients had adult-onset systemic disease (AOSD), 4/118 had rheumatoid arthritis with sepsis, 2/133 had sarcoidosis,
We tested the same sera with all seven different ELISA assays. Each ELISA plate was coated with a human ferritin peptide or the entire human ferritin protein. The sequence of A1-18 is MTTASTSQVRQNYH QDSE, and its purity is 97.5 %. A19-45 has a sequence of AAINRQINLELYASYVYLSMSYYFDRD and a pu r i t y o f 9 8 .8 %. A 5 2- 7 8 ha s a se q ue nc e of FAKYFLHQSHEEREHAEKLMKLQNQRG and a purity of 98.3 %. The sequence of A79-104 is GRIFLQDIKK PDCDDWESGLNAMECA, and its purity is 98.8 %. The purity of A105-144 is 98.4 %, and its sequence is LHLEKNVNQSLLELHKLATDKNDPHLCDFIETHY LNEQVK. The sequence of A145-183 is AIKELGDHVT NLRKMGAPESGLAEYLFDKHTLGDSDNES, and its purity is 98.5 %. All peptides were synthesised by Biomatik, Wilmington, DE, USA. For those six peptides, we developed six different ELISAs for the detection of antibodies against the associated peptide. For the detection of antibodies against the full-length human ferritin heavy chain, we used the same protocol as in our previous publication [7]. ELISA protocol for the peptides First, 96-well plates (Maxisorp, Nunc, Roskilde, Denmark) were coated with 1 μg of the appropriate peptide per well in 300 μl phosphate saline buffer (PBS) at 4 °C overnight. The coating solution was subsequently discarded. Then, the plates were blocked with PBS for 30 min at room temperature to prevent the non-specific binding of antibodies. The PBS was also discarded. We filled each well with 100 μl of distilled water and 100 μl of diluted serum (1 μg per well) in PBS and incubated the plate for 30 min at room temperature. Then, the plates were washed four times with 200 μl of Tris buffered saline with 0.02 % Tween20 (TBS-T). Next, 100 μl of a horse radish peroxidase (HRP)-conjugated goat anti-human IgG secondary antibody (Jackson ImmunoResearch Europe Ltd., Newmarket Suffolk, UK) was added at a dilution of 1 μl in 40 ml of PBS and 10 ml of 5 % BSA. The plates were incubated for 20 min at room temperature and washed four times with TBS-T. The colourimetric reaction was performed using 3,3′,5,5′ tetramethyl benzidine (TMB) (Thermo Fisher Scientific, Roskilde, Denmark) for up to 5 min according to
Clin Rheumatol (2014) 33:1523–1526
1525
the manufacturer’s instructions. Then, the reaction was stopped by adding 100 μl of stop solution (Thermo Fisher) to each well. The ODs were measured at 450 nm in an ELISA reader. The ELISA cutoff was calculated as the mean signal from the blood donors’ sera plus 3 standard deviations. Each ELISA was performed with eight to 16 blood donors. The Wilcoxon test was used to evaluate significant differences in the evaluated data within a group. Comparisons between groups were evaluated using the Kruskal-Wallis test. All p values refer to two-tailed tests. Correlations were analysed in terms of Spearman’s correlation. Two-tailed p
BRIEF REPORT
Association of ferritin antibodies with Takayasu arteritis K. Große & T. Witte & F. Moosig & B. F. Hoyer & C. Lansche & R. E. Schmidt & N. T. Baerlecken
Received: 7 August 2014 / Accepted: 24 August 2014 / Published online: 2 September 2014 # International League of Associations for Rheumatology (ILAR) 2014
Abstract Takayasu arteritis (TA) is difficult to diagnose because diagnostic biomarkers have not yet been established. In a previous study, we detected autoantibodies against the human ferritin heavy chain protein (HFC) in the sera of patients with giant cell arteritis (GCA) and/or polymyalgia rheumatica (PMR). The aim of this study is to evaluate the frequency of autoantibodies against HFC in TA. We established seven ELISA assays for the detection of autoantibodies against HFC. We used the full-length recombinant HFC expressed in Escherichia coli or one of six different HFC peptides as autoantigens: 1-18Aa (98.8 % purity), 19-45Aa (98.8 % purity), 52-78Aa (98.3 % purity), 79-104Aa (98.8 % purity), 105143Aa (98.4 % purity) and 145-183Aa (98.5 % purity). We collected sera from 48 patients with TA, 36 patients with systemic lupus erythematosus (SLE), 35 patients with arteriosclerosis, 133 patients with febrile diseases, which are known to generate unspecific autoantibodies, and 50 blood donors, which served as controls. The best results were obtained using the ferritin peptides as antigens. By combining the results from the different ELISAs that detect autoantibodies against the HFC peptides 19-44A, 79-104A and 105-144A, we were able to detect ferritin peptide antibodies in 30/48 (62 %) of the K. Große : T. Witte : C. Lansche : R. E. Schmidt : N. T. Baerlecken (*) Department of Clinical Immunology and Rheumatology, Medical University Hannover, Carl-Neuberg Str. 1, 30625 Hannover, Germany e-mail: [email protected] F. Moosig Department of Rheumatology and Immunology, Bad Bramstedt/ University of Luebeck, Luebeck, Germany B. F. Hoyer Department of Rheumatology and Clinical Immunology, Charite Free University and Humboldt University Charité University Medicine Berlin, Berlin, Germany
TA patients. The frequency was lower than in early GCA and PMR (previous study showed up to 92 %). Positive results were observed in 0/50 (0 %) of the control blood donors, 10/ 36 (28 %) of the SLE patients, 4/35 (11 %) of the arteriosclerosis patients and 27/133 (20 %) of the fever patients. Considering the lack of biomarkers for TA, autoantibodies against HFC peptides could act as useful markers for TA. Keywords Autoantibodies . Epitope mapping . Ferritin . Takayasu arteritis
Introduction Takayasu arteritis (TA) has an incidence of 2.6 million cases per year in Western countries [1]. It occurs predominantly in females less than 20 years of age. An acute inflammatory phase with systemic and cardiovascular symptoms followed by the chronic course in which the disease slowly progresses occurs in 50 % of cases. Lack of pulse, vascular bruit, elevated blood pressure, heart failure and abnormal fundi sometimes occur. Non-specific symptoms such as asthenia, weight loss, fever, dyspnoea, headache and arthralgias can also occur. Laboratory findings show that greater than 80 % of patients have an elevated erythrocyte sedimentation rate (ESR). Gamma globulin levels may also be elevated. In Western countries, diagnosis is greatly delayed, ranging from 2 to 11 years after the onset of the first symptoms. TA is more common in Asia, which leads to faster diagnosis, ranging, for example, from 2.5 to 5 months in India [2, 3]. Early diagnosis and treatment can avert complications such as aortic regurgitation, congestive heart failure and cerebrovascular events [4, 5]. Contemporary diagnostic methods such as MRI, angiography and Doppler ultrasounds are however either insufficient or quite elaborate. Recently, we characterised antibodies against the human ferritin heavy
1524
chain (HFC) and its N-terminal peptide in the sera of giant cell arteritis (GCA)/polymyalgia rheumatica (PMR) patients. Because TA is also a large vessel vasculitis, we wanted to study whether antibodies against ferritin are a diagnostic marker of TA [6–9].
Clin Rheumatol (2014) 33:1523–1526
3/133 had SLE with sepsis, 2/118 had Sjögren’s syndrome with acute infection, 23/133 had different malignant diseases with acute infection and 35/133 had various infectious diseases without other underlying diseases. The patients with arteriosclerosis were collected from our in- and outpatient clinics. Rheumatoid arthritis was observed in 23/35 patients: 4/35 had SLE and 8/35 had osteoarthritis.
Patients, materials and methods ELISA assays In Hannover, Bad Bramstedt/Luebeck and Berlin (Germany), sera from 48 patients with TA were collected. Patients were selected according to the 1990 ACR criteria [10]. As controls, we tested sera from 74 patients with fever above 39 °C, 36 patients with systemic lupus erythematosus (SLE) and healthy blood donors [11]. The study was approved by our local ethics committee (“Ethik-Kommission der Medizinischen Hochschule Hannover”, head: Prof. Dr. Tröger) (project number 4928). All patients agreed to written informed consent. The demographic, laboratory and disease activity data were supplied from the Department of Clinical Immunology and Rheumatology of the medical University of Hannover and the Department of Rheumatology and Immunology, Bad Bramstedt/University of Luebeck, Department of Rheumatology and Clinical Immunology, Charité University Medicine Free University and Humboldt University of Berlin. All patient data are associated with the date when the blood samples were taken. The proportion of females was 42/48 (87 %) for TA patients, 31/36 (86 %) for SLE patients, 38/118 (32 %) for fever patients, 14/35 (40 %) for patients with arteriosclerosis and 33/50 (66 %) for blood donors. The median (range) age was 42 years (21–64) for TA patients, 43 years (21–74) for SLE patients, 52 years (21– 85) for fever patients, 62 years (42–86) for patients with arteriosclerosis and 37 years (21–37) for blood donors. The mean (SD) C-reactive protein (CRP) level was 15 mg/l (26) for TA patients and 4 mg/l (7) for SLE patients. The mean (SD) erythrocyte sedimentation rate (ESR) was 26 mmnW (20) for TA patients and 22 (21) for SLE patients. CRP was not available in five patients with TA. SLE patients were defined as active by the treating rheumatologist. The mean systemic lupus erythematosus disease activity index (SLEDAI) was 7 with a SD of 5. The mean anti-nuclear antibody (ANA) titre was 1:640, 28/ 36 (78 %), and these sera contained antibodies against dsDNA. GC treatment was administered to 36/36 (100 %) patients and 31/36 (86 %) received DMARDs. A proportion of patients with fever (24/118) had AIDS-related diseases, e.g. tuberculosis, non-Hodgkin lymphoma, toxoplasmosis, Pneumocystis jiroveci pneumonia or soor oesophagitis, 18/ 133 had tuberculosis including the HIV-infected patients, 20/ 133 patients had adult-onset systemic disease (AOSD), 4/118 had rheumatoid arthritis with sepsis, 2/133 had sarcoidosis,
We tested the same sera with all seven different ELISA assays. Each ELISA plate was coated with a human ferritin peptide or the entire human ferritin protein. The sequence of A1-18 is MTTASTSQVRQNYH QDSE, and its purity is 97.5 %. A19-45 has a sequence of AAINRQINLELYASYVYLSMSYYFDRD and a pu r i t y o f 9 8 .8 %. A 5 2- 7 8 ha s a se q ue nc e of FAKYFLHQSHEEREHAEKLMKLQNQRG and a purity of 98.3 %. The sequence of A79-104 is GRIFLQDIKK PDCDDWESGLNAMECA, and its purity is 98.8 %. The purity of A105-144 is 98.4 %, and its sequence is LHLEKNVNQSLLELHKLATDKNDPHLCDFIETHY LNEQVK. The sequence of A145-183 is AIKELGDHVT NLRKMGAPESGLAEYLFDKHTLGDSDNES, and its purity is 98.5 %. All peptides were synthesised by Biomatik, Wilmington, DE, USA. For those six peptides, we developed six different ELISAs for the detection of antibodies against the associated peptide. For the detection of antibodies against the full-length human ferritin heavy chain, we used the same protocol as in our previous publication [7]. ELISA protocol for the peptides First, 96-well plates (Maxisorp, Nunc, Roskilde, Denmark) were coated with 1 μg of the appropriate peptide per well in 300 μl phosphate saline buffer (PBS) at 4 °C overnight. The coating solution was subsequently discarded. Then, the plates were blocked with PBS for 30 min at room temperature to prevent the non-specific binding of antibodies. The PBS was also discarded. We filled each well with 100 μl of distilled water and 100 μl of diluted serum (1 μg per well) in PBS and incubated the plate for 30 min at room temperature. Then, the plates were washed four times with 200 μl of Tris buffered saline with 0.02 % Tween20 (TBS-T). Next, 100 μl of a horse radish peroxidase (HRP)-conjugated goat anti-human IgG secondary antibody (Jackson ImmunoResearch Europe Ltd., Newmarket Suffolk, UK) was added at a dilution of 1 μl in 40 ml of PBS and 10 ml of 5 % BSA. The plates were incubated for 20 min at room temperature and washed four times with TBS-T. The colourimetric reaction was performed using 3,3′,5,5′ tetramethyl benzidine (TMB) (Thermo Fisher Scientific, Roskilde, Denmark) for up to 5 min according to
Clin Rheumatol (2014) 33:1523–1526
1525
the manufacturer’s instructions. Then, the reaction was stopped by adding 100 μl of stop solution (Thermo Fisher) to each well. The ODs were measured at 450 nm in an ELISA reader. The ELISA cutoff was calculated as the mean signal from the blood donors’ sera plus 3 standard deviations. Each ELISA was performed with eight to 16 blood donors. The Wilcoxon test was used to evaluate significant differences in the evaluated data within a group. Comparisons between groups were evaluated using the Kruskal-Wallis test. All p values refer to two-tailed tests. Correlations were analysed in terms of Spearman’s correlation. Two-tailed p
