Distribution of immunoglobulin G subclasses of anti-thyroid peroxidase antibody in sera from patients with Hashimoto's thyroiditis with different thyroid functional status

Abstract

The mechanism of disease progression in Hashimoto's thyroiditis (HT) is still unclear. Anti-thyroid peroxidase antibody (TPOAb), a diagnostic hallmark of HT, is principally of the immunoglobulin G (IgG) isotype, and it appears to be a response to thyroid injury. The aim of our study was to evaluate the distribution of IgG subclasses of TPOAb in sera from patients with HT with different thyroid functional status. Sera from 168 patients with newly diagnosed HT were collected and divided into three groups according to thyroid function: patients with hypothyroidism (n = 66), subclinical hypothyroidism (n = 60) and euthyroidism (n = 42). Antigen-specific enzyme-linked immunosorbent assay was used to detect the distribution of TPOAb IgG subclasses. The prevalence of TPOAb IgG subclasses in all patients’ sera with HT was IgG1 70·2%, IgG2 35·1%, IgG3 19·6% and IgG4 66·1% respectively. The prevalence of IgG2 in sera from patients with hypothyroidism (51·5%) was significantly higher than that of subclinical hypothyroidism (33·3%) (P < 0·05), and the latter was also significantly higher than that of euthyroidism (11·9%) (P < 0·05). The positive percentage of IgG2 subclass in sera from patients with hypothyroidism and subclinical hypothyroidism was significantly higher than that of euthyroidism (P < 0·05), the prevalence and positive percentage of IgG4 subclass in sera from patients with hypothyroidism and subclinical hypothyroidism was significantly higher than that of euthyroidism respectively (P < 0·05). The predominant TPOAb IgG subclasses in sera from patients with HT were IgG1 and IgG4. Patients with high levels of TPOAb IgG2, IgG4 subclasses might be at high risk of developing overt hypothyroidism.

Introduction

Autoimmune thyroid diseases (AITD) are among the most frequent organ-specific autoimmune diseases, including Hashimoto's thyroiditis (HT), Graves’ disease, atrophic thyroiditis, primary myxoedema and post-partum thyroiditis [1].

Anti-thyroid peroxidase antibody (TPOAb) is the hallmark of HT, which can be detected in almost 95% of HT patients’ sera. It is believed that thyroid lesion is, in part, a consequence of complement fixation [2–4] and antibody-dependent cell-mediated cytotoxicity (ADCC) [5–8] generated by TPOAb. Patients with HT have a great deal of clinical status; in general, it is an inconvertible process of evolving from euthyroidism to hypothyroidism. It is known that positive TPOAb increases the probability of developing hypothyroidism [9]. However, the progression rate of euthyroidism to subclinical, and even to overt, hypothyroidism is variable [9], and the progression mechanism of HT is still unclear.

The TPOAb found in Hashimoto’ sera are principally of the immunoglobulin G (IgG) subclasses. Because each subclass has a different biological function, the subclass distribution of TPOAb might be different in the progression of HT. One study reported that the IgG1/IgG4 ratio was significantly lower in patients with subclinical hypothyroidism than those with hypothyroidism [10]. We speculated that studies on immunological characteristics of TPOAb might provide useful information on the mechanism of disease progression in HT patients. The aim of our study was to investigate IgG subclasses distribution of TPOAb in sera from patients with HT with different thyroid functional status.

Patients and methods

Study population

A total of 168 patients with newly diagnosed HT in Peking University First Hospital during August 2005 to December 2007 were selected on the basis of high levels of TPOAb in the current study. All the patients had the classic (or goitrous) variant of thyroiditis. Diagnosis was based on the presence of a firm, symmetrically enlarged thyroid. None of the patients had other autoimmune diseases, such as pernicious anaemia, systemic lupus erythematosus, rheumatoid arthritis, type 1 diabetes or any evidence of co-existent gravidity or tumour. Patients receiving anti-thyroid drugs or l-thyroxine therapy were also excluded. There was no evidence of hereditary and acquired variations in the concentrations of thyroxine-binding globulin in sera from the patients. This study complied with the Helsinki declaration and was approved by the Ethics Committee of Peking University First Hospital. All patients gave written informed consent.

According to thyroid function, they were divided into three groups: patients with hypothyroidism (H) (n = 66, four males, 62 females), subclinical hypothyroidism (sH) (n = 60, 10 males, 50 females) and euthyroidism (E) (n = 42, two males, 40 females). There were no significant sex differences among the H, sH and E groups. The average patient age, in years, was similar for all three groups, i.e. H (46 ± 15), sH (50 ± 15) and E (44 ± 16).

Serological and clinical examinations

Serum samples were collected on diagnosis and kept frozen at −20°C until use. Chemiluminescent immunoassays were used to detect TPOAb, total triiodothyronine (TT3), total tetraiodothyronine (TT4) and thyrotropic-stimulating hormone (TSH) [TT3, TT4 and TSH by ADVIA Centaur (Bayer Healthcare Diagnostics, Tarrytown, NY, USA), TPOAb by IMMULITE 1000 (Diagnostic Products Corporation, Los Angeles, CA, USA)].

Enzyme-linked immunosorbent assay specific for IgG subclasses of TPOAb

Ninety-six-well plates (Costar, Cambridge, MA, USA) were coated with 0·5 µg/ml human thyroid peroxidase (TPO) (AppliChem Corporation, Ottoweg, Darmstadt, Germany) in 0·1 M carbonate/bicarbonate buffer, pH 9·6, at 4°C overnight. Serum samples were diluted (1:50) in phosphate-buffered saline (PBS) containing 0·1% Tween 20, and incubated for 30 min. After extensive washing, horseradish peroxidase-labelled mouse anti-human monoclonal antibodies were added. Monoclonal antibodies to IgG1 (4E3), IgG2 (HP6014), IgG3 (HP6050) and IgG4 (HP6025) (Southernbiotech, Birmingham, AL, USA) were used at dilutions of 1:2000, 1:800, 1:1000 and 1:1000 respectively. After incubation for 30 min and extensive washing, 0·4 mg/ml o-phenylenediamine and 1 µl/ml 3% H₂O₂ were finally added to each well and the reaction was stopped with 1 M hydrochloric acid after 20 min. Every plate contained positive, negative and blank controls (PBS + Tween). The volume in each well was 100 µl in all steps, and each sample was added in duplicate. The results were recorded as optical density at 490 nm and expressed as percentage of a known positive sample. Samples were considered positive if they exceeded mean + 3 standard deviations from 100 sera in normal blood donors (no clinical, autoantibody or ultrasonographic evidence of thyroid disease).

Statistical analysis

A non-parametric test was used to compare the total TPOAb levels in the three study groups. The prevalence of IgG subclasses was examined using the χ² test. The positive percentage of IgG subclasses was performed on log transformation, and comparison was performed using analysis of variance, followed by a group × group comparison using the Student–Neuman–Keuls test. The SPSS version 11·0 statistical analysis program (SPSS Inc., Chicago, IL, USA) was used. A P-value less than 0·05 was considered to have significant difference.

Results

Thyroid function and TPOAb levels

As shown in Table 1, TSH levels in the H group were significantly higher than those in the sH and E groups respectively (P < 0·001). TT3 and TT4 levels in the H group were significantly lower than those in the other two groups (P < 0·001). There were no significant differences between TSH, TT3 and TT4 levels in the sH and E groups (P > 0·05).

Table 1.

The levels of total triiodothyronine (TT3), total tetraiodothyronine (TT4) and thyrotropic-stimulating hormone (TSH) in sera from patients with hypothyroidism, subclinical hypothyroidism and euthyroidism.

Group	n	TSH (uIU/ml)	TT3 (nmol/l)	TT4 (nmol/l)
H	66	79·5 ± 50·8*,**	1·1 ± 0·6*,**	33·4 ± 25·9*,**
sH	60	12·2 ± 5·9	1·9 ± 0·6	85·9 ± 22·4
E	42	2·6 ± 1·3	1·8 ± 0·2	100·4 ± 15·5

^*P < 0·001 compared with E

^**P < 0·001 compared with sH. H, hypothyroidism; sH, subclinical hypothyroidism; E, euthyroidism. Reference value range: TT3: 0·92–2·79 nmol/l; TT4: 58·1–140·6 nmol/l; TSH: 0·35–5·5 uIU/ml.

Table 2 summarizes the median levels of TPOAb in three groups. Differences in TPOAb levels among the H, sH and E groups were not statistically significant (P > 0·05).

Table 2.

Anti-thyroid peroxidase antibody (TPOAb) levels in sera from patients with hypothyroidism, subclinical hypothyroidism and euthyroidism.

Group	n	TPOAb (IU/ml)
		Median	Interquartile range
H	66	1000·0	(427·0–1000)
sH	60	883·0	(284·8–1000)
E	42	1000·0	(332·8–1000)

H, hypothyroidism; sH, subclinical hypothyroidism; E, euthyroidism.

Distribution of TPOAb IgG subclasses

As shown in Table 3, the prevalence of TPOAb IgG subclasses in sera from all patients with HT was IgG1 70·2%, IgG2 35·1%, IgG3 19·6% and IgG4 66·1% respectively. The prevalence of IgG2 in the H group (51·5%) was significantly higher than that of the sH group (33·3%) (P < 0·05), and the latter was also significantly higher than that of the E group (11·9%) (P < 0·05). The prevalence of the IgG4 subclass in the H (72·7%) and sH (71·7%) groups was significantly higher than that of the E group (P < 0·05).

Table 3.

The prevalence of anti-thyroid peroxidase antibody (TPOAb) immunoglobulin G (IgG) subclasses in sera from patients with hypothyroidism, subclinical hypothyroidism and euthyroidism.

Group	n	IgG1 positive n (%)	IgG2 positive n (%)	IgG3 positive n (%)	IgG4 positive n (%)
H	66	47 (71·2)	34 (51·5*,**)	11 (16·7)	48 (72·7*)
sH	60	44 (73·3)	20 (33·3*)	10 (16·7)	43 (71·7*)
E	42	27 (64·3)	5 (11·9)	12 (28·6)	20 (47·6)
Total	168	118 (70·2)	59 (35·1)	33 (19·6)	111 (66·1)

^*P < 0·05 compared with E

^**P < 0·05 compared with sH. H, hypothyroidism; sH, subclinical hypothyroidism; E, euthyroidism.

Table 4 shows that the positive percentages of TPOAb IgG2 and IgG4 subclasses in the H and sH groups were significantly higher than that of the E group respectively (P < 0·05), but no statistical differences were found in the H and sH groups. There were no significant differences in the prevalence and positive percentage of IgG1 and IgG3 among the three groups.

Table 4.

The positive percentage of anti-thyroid peroxidase antibody (TPOAb) immunoglobulin G (IgG) subclasses in sera from patients with hypothyroidism, subclinical hypothyroidism and euthyroidism.

Group	n	TPOAb IgG subclasses
		IgG1 (%)		IgG2 (%)		IgG3 (%)		IgG4 (%)
		Median	Interquartile range	Median	Interquartile range	Median	Interquartile range	Median	Interquartile range
H	66	10·60	(3·90–11·5)	5·17*	(3·46–7·33)	2·71	(1·70–4·42)	6·95*	(3·48–12·8)
sH	60	9·32	(3·94–11·9)	4·28*	(2·89–6·93)	2·45	(1·65–3·53)	5·92*	(3·35–10·2)
E	42	8·08	(2·99–11·5)	3·05	(2·27–4·32)	2·61	(1·72–5·94)	3·80	(1·21–11·1)

^*P < 0·05 compared with E.

H, hypothyroidism; sH, subclinical hypothyroidism; E, euthyroidism.

Discussion

The TPOAb is principally of the IgG isotype. The major Fc-mediated functions of IgG include the activation of complement via the classical pathway, and the targeting of bound antigen for elimination through phagocytosis and ADCC [11]. Among the four IgG subclasses, IgG1 and IgG3 have the highest affinity for C1q [12] and the strongest ADCC [13]. IgG2 is less capable to fix complement and mediate ADCC [11]. IgG4 is poor at complement fixation, failing to bind C1q [13]. It has a low affinity for Fcγ receptors, making it very weak in mediation of ADCC [14], and is considered as a marker of chronic antigen exposure [15]. Because each subclass has a different biological function, studies on the IgG subclass distribution of TPOAb in HT patients’ sera with different thyroid functional status might provide useful information on the progression mechanism of HT.

Our study confirms the observation that TPOAb is not restricted to any particular isotype, but comprises all four IgG subclasses. Regardless of different thyroid functional status, the most represented IgG subclasses of TPOAb were IgG1 and IgG4, in agreement with some previous studies [16–18].

The study observed that the prevalence and positive percentage of TPOAb IgG1 were predominant in all three groups. As described above, IgG1 plays an important role in the inflammation process and is very powerful in mediating ADCC [19]. A human TPO-specific Fab converted to IgG1, but not IgG4, could mediate cytotoxic effects on human thyroid cells in vitro[5]. Davies et al.[20] have suggested that suppressor T cell control of total IgG secretion in patients with AITD was deficient in in vitroconditions. T cells of patients with HT were able to induce differentiation of B cells to secrete detectable IgG1, but suppression of IgG1 secretion was significantly less than that using cells from normal subjects [21]. These previous studies might explain the dominance of IgG1 regardless of thyroid functional status. In our study, there were no significant differences in the prevalence and positive percentage of TPOAb IgG1 subclass among the three groups, in agreement with the study by Silva et al.[10]. It is well known that HT is a chronic inflammation of the thyroid gland, associated with follicular destruction [22]. However, approximately 90% of the thyroid gland must be destroyed before hypothyroidism develops [23]. We speculated that at the early stage of HT (euthyroid stage) immune impairment had already started, although patients possibly had no symptoms. Therefore, high levels of the IgG1 subclass, which plays an important role in the inflammation process, might present regardless of thyroid functional status.

In our study, the highest prevalence and positive percentage of IgG2 was found in the H group, followed by that of the sH and E groups. Weetman et al.[18] found that IgG2 had the highest relative functional affinity. Previous studies have shown that the T helper (Th) lymphocytes infiltrating the thyroid gland were mainly of the Th1 type in HT [24], and interferon (IFN)-γ was an important Th1 type cytokine. Some studies have shown that IFN-γ secreted by CD4⁺ and CD8⁺ T cells in patients with hypothyroidism was significantly higher than that of euthyroidism [25], and IFN-γ was able to induce IgG2 expression [26]. These might be the reason for higher levels of IgG2 subclass in the H and sH groups. Our study indicated that IgG2 might be another important subclass involved in immune impairment in HT other than IgG1, and patients with high levels of TPOAb IgG2 subclass might be at high risk of developing overt hypothyroidism.

In clinical practice, many patients with HT have mild symptoms, and may be diagnosed as HT only during routine examination. Accordingly, although the patients in our study were all newly diagnosed, they might have a long history of HT. The prevalence and positive percentage of the IgG4 subclass in the H and sH groups in our study were significantly higher than that of the E group. As described above, IgG4 subclass was considered as a marker of chronic antigen exposure. Therefore, we speculated that repeated antigenic exposure and stimulation in immune impairment of the thyroid follicle might induce production of the TPOAb IgG4 subclass, and the longer that immune injury might be present, the higher IgG4 levels might be detected and the greater the possibility that patients with HT develop overt hypothyroidism. Further study needs to be conducted to confirm it.

In conclusion, the distributions of TPOAb IgG subclasses in sera from patients with HT did not restrict to any particular IgG subclass, predominantly IgG1 and IgG4. Patients with high levels of TPOAb IgG2, IgG4 subclasses might be at high risk of progressing to overt hypothyroidism. Although TPOAb appears to be a secondary response to thyroid injury, it might contribute to its development and chronicity [23]. Therefore, we speculate that immunological characteristics of TPOAb might play an important role in the progression of HT. Long-term follow-up studies on these patients are currently under way to verify this hypothesis.