Titre and affinity of propylthiouracil-induced anti-myeloperoxidase antibodies are closely associated with the development of clinical vasculitis

Abstract

Substantial evidences suggested that propylthiouracil (PTU) could induced anti-myeloperoxidase (MPO) antibodies in sera from patients with hyperthyroidism, however, only a subgroup of the PTU-induced anti-MPO antibody positive patients developed clinical evident vasculitis. The aim of this study is to compare the titres and affinities of PTU induced anti-MPO antibodies in sera from patients with hyperthyroidism with and without clinical vasculitis. Anti-MPO antibody positive sera from patients diagnosed hyperthyroidism with (n = 13) and without (n = 14) clinical evident vasculitis were collected. The titre was determined by MPO-ELISA and expressed as logarithm value (lgT). The affinity constant (aK) of anti-MPO IgG was measured by antigen inhibition assay. The titre and aK values were compared between patients with and without vasculitis. In patients with vasculitis, the mean lgT of anti-MPO antibodies was 3·62 ± 0·66; the median aK was 4·47 × 10⁷M⁻¹. In patients without vasculitis, the mean lgT was 2·54 ± 0·29; the median aK was 0·14 × 10⁷M⁻¹, and both were significant lower than those in patients with vasculitis (t = 5·464; P = 0·000 & z = −4·373; P= 0·000, respectively). We concluded that the titre and affinity of anti-MPO antibodies might be associated with the development of clinical vasculitis in patients with PTU-induced ANCA.

Introduction

Anti-neutrophil cytoplasmic antibodies (ANCA) are important serological markers of certain small vessel vasculitides. Anti-proteinase 3 (PR3) antibodies produce cytoplasmic fluorescence and are found in patients with Wegener's granulomatosis (WG); anti-myeloperoxidase (MPO) antibodies produce perinuclear fluorescence and are found in microscopic polyangiitis (MPA). It is suggested that ANCA play a pathogenetic role in ANCA associated vasculitis [1]. Propylthiouracil (PTU) is a common antithyroid drug, which has been known to induce ANCA-positive vasculitis [2–6]. In most of the cases of PTU-induced ANCA-positive vasculitis, a higher prevalence of positive MPO-ANCA was reported. However, our previous studies revealed that some patients with PTU induced MPO-ANCA had no clinical manifestations of vasculitis [7]. We speculated that the amount and the binding capacity of the anti-MPO antibodies might be associated with appearance of clinical vasculitis. In the current study, we compared the titres and affinities of PTU induced anti-MPO antibodies in sera from patients diagnosed hyperthyroidism with and without clinical vasculitis.

Materials and methods

Patients and sera

PTU-induced anti-MPO antibody positive sera from patients with hyperthyroidism with (n = 13) and without (n = 14) clinical evident vasculitis, diagnosed in Peking University First Hospital during December 1999 to December 2004, were collected at presentation and were stored at −20 °C until use. Clinical data of patients were summarized in Table 1 and Table 2. The Birmingham Vasculitis Activity Score (BVAS) was used to assess the clinical activity of vasculitis [8].

Table 1. Clinical and immunological data of patients with vasculitis.

No	Age	Sex	Organ involvement	Serum creatinine (µmol/l)	MPO (µg/ml)	aK (×107 M−1)	Titre	lgT	BVAS
1	20	F	ENT, Ey, J, K, L, M, S	44·0	3·13	4·47	1/25 600	4·41	31
2	37	M	Ey, K, L, M, S	60·0	50	0·28	1/3200	3·51	20
3	61	F	K, L, S	731	1·2	11·7	1/800	2·9	20
4	20	F	C, G, K, O, S	42·0	12·5	1·12	1/1600	3·2	17
5	20	F	C, Ey, ENT, J, K, M, S	62·9	18·7	0·75	1/25 600	4·41	19
6	27	F	G, K, M	86·7	6·25	2·24	1/25 600	4·41	16
7	30	F	ENT, K	82·0	12·5	1·12	1/1600	3·2	13
8	42	F	J, K, L	70·0	<0·1	>140	1/800	2·9	20
9	30	F	ENT, Ey, J, K, S	51·0	50	0·28	1/400	2·6	17
10	13	F	G, K	64·0	25	0·56	1/3200	3·51	17
11	21	F	J, K, L, S	289	0·4	35·0	1/3200	3·51	16
12	9	F	M, L, K, S	68·0	0·2	70·0	1/12 800	4·11	17
13	43	F	ENT, K, L, S	242	0·1	140	1/25 600	4·41	16

C, cutaneous involvements; ENT, ear, nose and throat; Ey, eyes; G, gastrointestinal involvement; J, joint; K, kidney; L, lung; M, musculoskeletal involvement; N, neuritis; O, oral involvement; S, systemic symptoms (fever, malaise, weigh loss); MPO, MPO concentration resulting in 50% inhibition of antibody binding; aK, affinity constant of anti-MPO IgG; lgT, logarithm value of titre; BVAS, The Birmingham Vasculitis Activity Score.

Table 2. Titre and affinity of patients without vasculitis.

Patient	Age	Sex	MPO (µg/ml)	aK(×107 M−1)	Titre	lgT
1	60	F	>100	<0·14	1/400	2·6
2	65	F	>100	<0·14	1/100	2
3	19	F	50	0·28	1/200	2·3
4	19	M	100	0·14	1/200	2·3
5	52	F	>100	<0·14	1/800	2·9
6	38	F	100	0·14	1/400	2·6
7	29	F	>100	<0·14	1/400	2·6
8	46	F	25	0·56	1/400	2·6
9	34	M	100	0·14	1/400	2·6
10	31	M	>100	<0·14	1/200	2·3
11	24	F	100	0·14	1/400	2·6
12	56	F	100	0·14	1/400	2·6
13	27	F	100	0·14	1/1600	3·2
14	9	F	100	0·14	1/200	2·3

MPO, MPO concentration resulting in 50% inhibition of antibody binding; aK, affinity constant of anti-MPO IgG; lgT, logarithm value of titre.

Detection of titre of anti-MPO antibodies

The titres of anti-MPO antibodies were measured by enzyme-linked immunosorbent assay (ELISA). In brief, highly purified human native MPO [9] were coated to Costar microtitre plates (Data Packaging Corporation, MA, USA) at a concentration of 2·0 µg/ml in coating buffer (0·05 mol/l bicarbonate buffer, pH 9·6). The volume in each well was 100 µl in this step and subsequent steps and every sample was added in duplication, all incubations were carried out at 37æ for 1 h, and the plates were washed three times with phosphate buffer solution (PBS) containing 0·1% Tween 20 (PBST) between stages. Sera from patients were diluted from 1 : 50 to 1 : 25 600 with PBST and were incubated for 1 h at 37 °C and the binding was revealed with a horseradish peroxidase-conjugated goat anti-human IgG (Jackson Immunoresearch, USA) diluted at 1 : 10 000, followed by addition of diaminobenzidine. The absorbance was recorded at 490 nm. Every plate contained a positive control, a negative control and blank controls. The titre of anti-MPO antibodies was defined as the maximum serum dilution giving a positive binding and was expressed as logarithm value (lgT).

Detection of functional affinity of anti-MPO antibodies

The functional affinity constant (aK) was determined as described [10,11] as the reciprocal value of molar concentration of MPO in the liquid phase resulting in 50% inhibition of antibody binding. Briefly, the serum concentration required for competitive assay was first determined for each patient as the serum dilution giving about 70% of the maximum binding in the standard MPO ELISA. The competitive binding assay was performed by incubating the diluted patients' sera with increasing amounts of MPO (from 0·1 µg/ml to 100 µg/ml) in PBST, for 2 h at 37 °C. The diluted sera with and without MPO inhibition were then both transferred to MPO-coated plates for the standard ELISA procedure. The anti-MPO IgG binding was expressed as the percentage of control binding determined in absence of fluid phase MPO.

Statistical analysis

The independent Student's t-test and the Mann–Whitney U-test were used to compare quantitative parameters, significant difference was considered as P < 0·05.

Results

Demographic data

There were 13 patients with PTU induced ANCA positive vasculitis, 12 were female and one was male with an average age at 28·7 ± 14·2 (9–61) years. All patients had multisystemic involvement. The mean of BVAS was 18·4 ± 4·3 (13–31). There were 14 patients without clinical vasculitis, 11 were female and three were male with an average age at 36·3 ± 17·1 (9–65) years, no significant difference could be found in age and gender between the two groups (Tables 1 and 2).

Titre and affinity of anti-MPO antibodies

In patients with vasculitis, the mean lgT of anti-MPO antibodies was 3·62 ± 0·66 and the median aK was 4·47 × 10⁷M⁻¹ (range, 0·28 × 10⁷M⁻¹ to >140 × 10⁷M⁻¹). In patients without vasculitis, the mean lgT was 2·54 ± 0·29; the median aK was 0·14 × 10⁷M⁻¹ (range, < 0·14 × 10⁷M⁻¹ to 0·56 × 10⁷M⁻¹), and both were significantly lower than that in patients with vasculitis (t = 5·464; P = 0·000 & z = −4·373; P = 0·000, respectively).

Discussion

ANCA was more than a serological marker of disease and could stimulate leucocytes to undergo a respiratory burst and degranulate primary granular constituents in a wide variety of ways resulting in the release of reactive oxygen species, granule proteins, cytokines, chemokines, and adhesion molecules. Leucocytes activated by ANCA could also adhere to endothelium and cause endothelial cell damage [12,13]. These supported a direct pathogenic role for ANCA in development of vasculitis.

In patients with ANCA associated vasculitis, higher autoantibody titres could be observed at the onset of the disease and during relapse [14,15]. Jayne et al. [16] suggested that ANCA could be undetectable in clinical remission after treatment and most relapses were preceded by a rise in ANCA titre.

Antibodies with high binding capacity could react with antigens more quickly and strongly and might result in more potent inflammatory injury. Studies suggested that the binding capacity of antibodies might also play a pathogenic role. Kokolina et al. [11] showed that the titre and affinity of anti-MPO antibodies decreased after immunosuppressive therapy in sera from patients with primary vasculitis.

The pathogenesis of PTU induced ANCA associate vasculitis is still not clear yet. Our previous work suggested that antiendothelial cell antibodies might play an important role in the pathogenesis of PTU-induced vasculitis [17]. However, substantial evidences suggested that the interaction between PTU and the target ANCA antigen of ANCA might also attribute to the pathogenesis of PTU-induced vasculitis. Lee et al. [18] demonstrated that the structure of MPO could be partially changed by the repeated administration of PTU and Jiang et al. [19] suggested that PTU could serve as a MPO substrate and the metabolites might induce autoimmunity by exposing autoreactive lymphocytes to abnormal forms of self-material. It was reasonable to speculate that anti-MPO antibodies might play an important role in the pathogenesis of PTU-induced vasculitis. Harper et al. [20] and our previous study [7] demonstrated that PTU-induced ANCA were due to polyclonal activation of B cells, one of our further studies revealed that anti-MPO IgG3 subclass could not be detected in sera from both active phase and remission, which contrasted to the finding in ANCA associated systemic vasculitis; however, the titre of anti-MPO IgG4 subclass decreased more quickly than that in primary vasculitis [21]. Recently, we also demonstrated that the epitopes were overlapped between the two groups, the epitopes of anti-MPO antibodies from patients with PTU-induced vasculitis were more restricted than that from patients with primary ANCA associated systemic vasculitis [22]. These studies suggested that characteristics of anti-MPO antibodies in PTU-induced vasculitis and primary systemic vasculitis might be different.

Although, PTU could induce ANCA in a substantial percentage of patients with hyperthyroidism, however, most patients had no manifestation at all [7,20]. In the current study, the titre and affinity of anti-MPO antibodies in patients with clinical vasculitis were significant higher than that in patients without clinical vasculitis, which suggested that the amount and the binding capacity of anti-MPO antibodies might be also associated the development of clinical vasculitis.