Thyroid peroxidase and thyroglobulin expression in normal human thyroid glands

The objective of the present study was to investigate thyroid peroxidase (TPO) and thyroglobulin (Tg) expression in normal thyroid tissue. Thirty-eight normal thyroids from fetuses, children, adults, and elderly subjects were evaluated. TPO and Tg expression was determined by analyzing at least 200 cells/specimen. A value of 80% was considered to be the threshold for TPO positivity. In terms of age, TPO expression occurs in fetuses along with follicle differentiation and is absent in solid follicles. TPO staining was cytoplasmic, and was more intense in the apical membrane in tissue from fetuses, children, and young adults. A perinuclear ring was found in tissue from older adults and from the elderly subjects. There was concomitant TPO and Tg expression in all groups studied. Tg expression was observed in all normal thyroids from all age groups. Tg expression was extremely variable (10–100% of the cells) and fainter than TPO expression. Presented at the XXI International Congress of the International Academy of Pathology and 12th World Congress of Academic and Environmental Pathology, Budapest. Hungary, October 20–25, 1996.     

Anti-thyroglobulin autoantibodies in sera from patients with chronic thyroiditis and from healthy subjects: differences in cross-reactivity with thyroid peroxidase.

A significant portion (about 12.7%) of healthy subjects was found to contain anti-thyroglobulin (anti-Tg) antibodies in their sera. We compared the binding activities of these antibodies and of anti-Tg autoantibodies from sera of patients with chronic thyroiditis with human thyroid peroxidase (TPO). The results obtained by ELISA indicated that out of 10 healthy subjects with anti-Tg antibodies, only four had anti-Tg antibodies capable of binding to TPO, whereas anti-Tg autoantibodies from almost all patients with chronic thyroiditis possessed high binding activities to TPO. By use of the immunoprecipitation method, it was also shown that although all anti-Tg autoantibodies from patients precipitated TPO, a majority of anti-Tg antibodies from healthy subjects could not precipitate TPO. Such findings cannot be ascribed to the differences in levels of anti-Tg autoantibodies and anti-TPO autoantibodies in sera and the differences in avidities of anti-Tg antibodies in sera between healthy subjects and patients with chronic thyroiditis. Thus, it can be concluded that anti-Tg antibodies from healthy subjects differ from those of patients with chronic thyroiditis with respect to TPO binding, probably due to difference in fine specificities of these anti-Tg antibodies.     

Antibodies cross-reacting with thyroglobulin and thyroid peroxidase are induced by immunization of rabbits with an immunogenic thyroglobulin 20mer peptide

Thyroglobulin (Tg) and thyroid peroxidase (TPO) are two major autoantigens in autoimmune thyroid diseases (AITD). Cross-reactive anti-Tg/TPO antibodies have been identified in patients with AITD and in mice immunized with Tg or TPO. In the present study, we investigated the production of anti-Tg/TPO antibodies in rabbits immunized with human Tg and with a highly immunogenic Tg peptide (namely TgP41, sequence 2651-2670 of human Tg), by noncompetitive and competitive ELISA. TgP41 was found previously to induce intramolecular epitope spreading. We found that Tg-immunized rabbits developed a serological immune response to TPO due to cross-reactivity with Tg, since serum TPO reactivity was inhibited by soluble Tg and affinity-purified anti-Tg antibodies cross-reacted with TPO. Moreover, TgP41-immunized rabbits responded to Tg and TPO. This serological response was attributed to anti-Tg/TPO antibodies, based on the observation that serum TPO reactivity was again inhibited by soluble Tg, and affinity-purified anti-Tg antibodies, induced by TgP41-immunization, cross-reacted with TPO. Purified anti-TgP41 antibodies did not react with TPO, suggesting that a putative common antigenic determinant is not included in the peptide sequence. We propose that intermolecular spreading of reactivity to TPO observed after administration of the Tg-peptide is a result of intramolecular epitope spreading to determinant(s) responsible for Tg/TPO cross-reactivity.     

Thyroid peroxidase and the induction of autoimmune thyroid disease

Animal models of autoimmune thyroid disease are associated with thyroglobulin (Tg) as autoantigen whereas in man the autoimmune response to microsomal antigen/thyroid peroxidase (TPO) appears to play a major role in thyroiditis. Consequently, we have compared the ability of TPO and Tg to induce thyroid autoantibodies and thyroid damage in mice known to be susceptible (CBA/J) or resistant (BALB/c) to thyroiditis induced using murine Tg. Groups of three to five mice were immunized twice using Freund's complete adjuvant with 80-100 micrograms highly purified porcine (p) TPO, pTg, rat (r) Tg, human Tg, bovine serum albumin (BSA) or BSA + 0.2 micrograms pTg (the level of Tg contamination of TPO). Four weeks after immunization with TPO, plasma from CBA/J (but not BALB/c) mice contained IgG class antibodies which bound to TPO-coated tubes in the presence or absence of excess Tg (and could therefore be clearly distinguished from Tg antibodies) but there was no evidence of thyroiditis in either strain of mice. In contrast, in CBA/J mice immunized with rTg and, to a lesser extent in mice that had received pTg, thyroid tissue was infiltrated with lymphoid cells and/or neutrophils and antibodies to pTg (but not pTPO) were present. Our observations demonstrate that induction of TPO antibody alone is insufficient to lead to thyroiditis in CBA/J mice. Further, these studies emphasize the complex interactions between MHC and different thyroid antigens in the processes leading to thyroid destruction.     

Value of combined clinical information and thyroid peroxidase antibodies in pregnancy for the prediction of postpartum thyroid dysfunction

PROBLEM: To investigate the utility of thyroid peroxidase antibodies (TPOAb) in early pregnancy combined with clinical information for prediction of postpartum thyroid dysfunction (PPTD) within 1 year postpartum. METHOD OF STUDY: We studied 98 pregnant women by determining their TPOAb levels in early pregnancy, as well as their serum thyrotropin and free thyroid (fT4) levels at 6 and 12 months postpartum. Furthermore, they answered a questionnaire and physical examination was performed by only one examiner. RESULTS: Of the 98 women, 10 were positive TPOAb in early pregnancy. The overall risk of PPTD within 1 year of follow-up was 10.2% (95% CI 4.1-16.3). Risk of PPTD was significantly higher among women with a family history of thyroid disease, TPOAb positive and presenting goiter in early pregnancy. The sensitivity, specificity and positive predictive value of TPOAb in PPTD prediction were 60.0%, 95.5% and 60%. Restricting screening to women with a family history of thyroid disease or presenting goiter increases the positive predictive value from 60% to 82.4%. CONCLUSION: Our results suggest that TPOAb could be used as a screening test for PPTD prediction at least among women who present a high risk of developing PPTD.     

Use of recombinant epitopes to study the heterogeneous nature of the autoantibodies against thyroid peroxidase in autoimmune thyroid disease.

Microsomal antigen is often recognized by the sera from patients with autoimmune thyroid disease (AITD). Human thyroid peroxidase (hTPO) is the main component of this antigen. In a previous study, we expressed hTPO cDNA as fusion proteins in prokaryotic vector; we thereby defined seven antigenic peptides by using two rabbit polyclonal anti-hTPO antibodies. In the present study we used the seven epitopes and three widened peptides to define the reactivity pattern of 61 sera from patients with AITD. Thirty-eight of them reacted against at least one of the seven hTPO-restricted epitopes; 14 were negative against the seven determinants but recognized one or two of the extended peptides. Thus, the antibody response against hTPO appeared to be highly heterogeneous in AITD patient sera. Moreover, we demonstrated that the immunodetection of the hTPO on Western blotting with deoxycholate solubilized microsomes can be perfectly correlated with the recognition of one of the epitopes in the region 554-735.     

Anti-thyroid antibodies and thyroid dysfunction in rheumatoid arthritis: Prevalence and clinical value

Objectives: The aim of this study was to assess thyroid function as well as the prevalence and clinical value of anti-thyroid antibodies in patients with rheumatoid arthritis (RA).

Methods: Seventy patients with active RA (ACR criteria), 9 males and 61 females, mean age 47 years (range 15–77) were analyzed. Anti-thyroperoxidase (TPOAb) and anti-thyroglobulin antibodies (TgAb) were tested using radioimmunoassay. Free thyroxine (FT4) and free triiodothyronine (FT3) and thyroid-stimulating hormone (TSH) serum levels were measured using electro-immunochemiluminescence (ECLIA, Elecsys Roche). Clinical variables, including tender and swollen joint count, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), rheumatoid factor (RF), anti-cyclic citrullinated peptide antibody (anti-CCP) and antinuclear antibodies (ANA) were also evaluated. Statistics were performed by the SPSS statistical software for Windows.

Results: Twenty-six patients (37%) with RA were positive for TPOAb and 16 (23%) for TgAb. In 5 (7.1%) patients TSH level was slightly elevated, ranging between 4.52 and 15.65 UI/ml. The increase of TSH levels was associated with normal FT4 in 3 cases (4.2%) and with reduced FT4 in 2 cases (2.8%). One patient (1.5%) had low TSH serum value along with normal FT4. No differences in clinical and serological data between anti-thyroid positive and negative patients were observed.

Conclusion: Our study shows an increased prevalence of anti-thyroid antibodies in RA patients with a low prevalence of hormonal alterations. However, anti-thyroid antibodies do not seem to identify any peculiar RA phenotype.     

T cell responses to synthetic thyroid peroxidase peptides in autoimmune thyroid disease.

Sixteen peptides, representing four different extracellular regions of thyroid peroxidase (TPO) predicted to contain a high proportion of potential T cell epitopes, were synthesized to investigate which parts of this autoantigen may be targets for the T cell response in thyroid autoimmunity. Compared with 25 controls, peripheral blood T cells from 23-37% of 30 patients with Graves' disease or autoimmune hypothyroidism were stimulated significantly by three peptides, representing amino acids 415-432, 439-457 and 463-481 of the TPO sequence; T cells from individual patients were also stimulated by several other peptides. These results indicate that the T cell response to TPO is directed against several epitopes which may be recognized by different patients.     

Significance of thyroglobulin antibodies cross-reactive with thyroperoxidase (TGPO antibodies) in individual patients and immunized mice.

Thyroglobulin (TG) and thyroperoxidase (TPO), both involved in thyroid hormone synthesis, represent major autoantigens in thyroid autoimmune disease. Despite numerous studies, the emergence, pathophysiological significance and role of autoantibodies to TG and TPO remain elusive. The recent identification of a new category of thyroid-specific autoantibody interacting with both TG and TPO (TGPO autoantibodies) offers a new opportunity in the study of thyroid autoimmunity. To gain a better insight into the significance of these TGPO autoantibodies, measurement in individual samples appeared necessary. The unique property of TGPO autoantibodies, simultaneous binding to TG and TPO, was used to set up a sandwich method which combined coated TG and radio-iodinated TPO. This method was found to be strictly specific for TGPO autoantibodies and sensitive enough to assay TGPO autoantibodies in serum. In humans, TGPO autoantibodies were found in most of the sera with high TG and TPO autoantibody titres, but not in sera negative for TG autoantibodies, whatever the TPO autoantibody titre. Furthermore, high TGPO autoantibody titres were found in sera strongly cytotoxic for cultured porcine thyroid cells. However, significant correlation of TGPO autoantibody titre was observed neither with TG and TPO autoantibody titres (n = 48) nor with complement-dependent cytotoxicity (n = 50). TGPO antibody assay was also performed in individual plasma of CBA/J mice immunized with either human TG (n = 6) or human TPO (n = 6). Immunization with TG induced high levels of not only TG but also TGPO antibodies, which exhibited a strong reactivity for TPO and whose binding to TG and TPO was fully inhibited by TG. In contrast, immunization with TPO induced high levels of only specific TPO antibodies accompanied by low levels of specific TG antibodies. In this case TGPO antibodies were not detected. Of note, TG- and TPO-immunized mice mounted an immune response against their own TG, but did not exhibit histological signs of thyroiditis. Large panels of TG and TPO MoAbs were also investigated with this method: 18/25 TG MoAbs and only 1/13 TPO MoAbs were found cross-reactive. Taken together, these data provide evidence that TGPO antibodies are effectively present in individual patients and TG-immunized mice, are different from specific TG and TPO antibodies, and may derive from natural B cell repertoire by autoimmune processes involving TG and not TPO.     

Evaluation of conformational epitopes on thyroid peroxidase by antipeptide antibody binding and mutagenesis

Autoantibodies to thyroid peroxidase (TPO) recognize predominantly conformational epitopes, which are restricted to two distinct determinants, termed immunodominant domain region (IDR) A and B. These dominant determinants reside in the region with structural homology to myeloperoxidase (MPO)-like domain and may extend into the adjacent complement control protein (CCP) domain. We have explored the location of these determinants on the MPO-like domain of the structural model of TPO, by identifying exposed hydrophilic loops that are potential candidates for the autoantigenic sites, generating rabbit antipeptide antisera, and competing with well characterized murine monoclonal antibodies (mabs) specific for these two IDRs. We recently defined the location of IDR-B, and here report our findings on the location of IDR-A and its relationship to IDR-B, defined with a new panel of 15 antipeptide antisera. Moreover, in combination with single amino acid replacements by in vitro mutagenesis, we have defined the limits of the IDR-B region on the TPO model. The combination of antisera to peptides P12 (aa 549-563), P14 (aa 599-617) and P18 (aa 210-225) inhibited the binding of the mab specific for IDR-A (mab 2) by 75%. The same combination inhibited the binding of autoantibodies to native TPO from 67 to 94% (mean 81.5%) at autoantibody levels of 5 IU. Fabs prepared from the antipeptide IgG and pooled in this combination were also effective in competition assays, thus defining the epitopes more precisely. IDR-A was found to lie immediately adjacent to IDR-B and thus the two immunodominant epitopes form an extended patch on the surface of TPO. Finally, by single amino acid mutagenesis, we show that IDR-B extends to residue N642, thus further localizing the boundary of this autoantigenic region on the structural model.     

Antigen-specific T cell recognition of affinity-purified and recombinant thyroid peroxidase in autoimmune thyroid disease.

The T cell proliferative responses of peripheral blood lymphocytes from 20 patients with autoimmune thyroid disease (AITD) and 20 healthy controls were analysed to immunoaffinity-purified thyroid peroxidase (TPO) and recombinant antigen preparations generated in Escherichia coli as glutathione-s-transferase fusion proteins. The epitope specificity of the T cell response was investigated using a selection of eight discrete recombinant fragments encompassing the whole of the extracellular region of the TPO molecule. Significant differences in the proliferative responses between patients and controls were observed to the full length, affinity-purified TPO molecule (P less than 0.002) as well as to the recombinant fragments R1c (residues 145-250) (P less than 0.001) and R2b (residues 457-589) (P less than 0.001) suggesting the presence of at least two distinct T cell determinants on this autoantigen. One of these T cell epitopes, localized within the region R1c, has not previously been identified by studies with synthetic peptides.     

Anti-thyroid peroxidase antibodies in sera from healthy subjects and from patients with chronic thyroiditis: differences in the ability to inhibit thyroid peroxidase activities.

A significant percentage (6.4%) of healthy subjects was found to contain anti-thyroid peroxidase (TPO) antibodies in their sera. However, in contrast with IgG from sera of patients with chronic thyroiditis, IgG from sera of healthy subjects did not inhibit TPO activities both in guaiacol and iodide assays. In addition, anti-TPO antibodies from healthy subjects did not block the inhibition of enzyme activities by anti-TPO antibodies from patients. These findings suggest that anti-TPO antibodies from healthy subjects do not bind to the epitopes relating to substrate-combining sites of TPO. Thus, the specificities of anti-TPO antibodies in healthy subjects may differ from those in cases of chronic thyroiditis.     

甲状腺球蛋白抗体和甲状腺过氧化物酶抗体定量检测在甲状腺肿大患者病因诊断上的应用

目的探讨甲状腺球蛋白抗体(TGAb)和甲状腺过氧化物酶抗体(TPOAb)定量检测在甲状腺肿大患者病因诊断上的应用。方法对291例甲状腺肿大患者血清和86例正常人血清用直接化学发光定量检测TGAb和TPOAb,并对甲状腺肿大患者检测甲功5项。结果甲状腺肿大患者TGAb和TPOAb浓度分别为(254±417)U/ml和(1001±1108)U/ml,有124例确诊为自身免疫性甲状腺炎(AITT),90例诊断为单纯性甲状腺肿大;AITT患者TGAb和TPOAb浓度和抗体检出率明显高于非AITT患者,P<0.01。结论定量检测血清TGAb和TPOAb对甲状腺肿大患者在病因诊断上有重要意义。     

Application of mouse monoclonal antibodies for identification of antigen regions of human thyroid peroxidase in adolescents with Graves' disease and non-toxic multinodular goiter by flow cytometry

Thyroid peroxidase (TPO) is the major thyroid autoantigen recognized by serum autoantibodies from patients with Graves' disease (GD) or Hashimoto's thyroiditis directed to two immunodominant conformational regions termed A and B. The epitopes of human TPO have been defined using a panel of mouse monoclonal antibodies (mAbs).

The aim of this study was to estimate the expression of chosen surface antigen regions of TPO (1, 18, 30, 64 epitopes) on thyroid cells in 15 patients with non-toxic multinodular goiter (NTMG) and 15 patients with GD. The thyrocytes were identified by indirect method: in the first stage we added mouse monoclonal autoantibodies specific for TPO regions and in the second stage we conjugated this complex with rabbit anti-mouse antibodies IgG (Fab')₂ with FITC. All investigations were performed by flow cytometry using Coulter EPICS XL apparatus. The percentages of thyrocytes with expression of epitopes 1, 18, 30, 64 TPO were measured in relation to the respective anti-TPO concentrations: 50–1600 μg/ml.

The analysis of epitopes located in immunodominant regions (IDR) of TPO revealed higher percentages of thyrocytes in cases with GD in comparison to NTNG. The most predominant difference was observed for mAb 64 epitope (48 vs 7%, p < 0.019; 39 vs 5%, p < 0.017) at the concentration of 100–200 μg/ml mAbs. The expression of 18 epitope on thyrocytes was also statistically higher in Graves' patients than in the NTMG (14 vs 6%, p < 0.025) at concentration of 400 μg/ml mAbs. However, this expression was much less pronounced.

In all the cases, the percentages of thyrocytes with epitopes 1 and 30 were in low detection (8–15% of positive cells).

In conclusion, our findings suggest that the elevated expression of TPO epitopes 18 and 64 in young patients with thyroid autoimmune diseases increase stimulation and activation of thyroid cells during inflammatory reaction within the thyroid gland. In addition, predominant expression of 64 TPO epitope that recognizes B domain in GD patients could be a useful marker of the immune process in the thyroid gland.     

Localization of key amino acid residues in the dominant conformational epitopes on thyroid peroxidase recognized by mouse monoclonal antibodies

Autoantibodies to thyroid peroxidase (TPO), the major target autoantigen in autoimmune thyroid diseases, recognize conformational epitopes limited to two immunodominant regions (IDRs) termed IDR-A and -B. The apparent restricted heterogeneity of TPO autoantibodies was discovered using TPO-specific mouse monoclonal antibodies (mAbs) and later confirmed by human recombinant Fabs. In earlier studies we identified key amino acids crucial for the interaction of human autoantibodies with TPO. Here we show the critical residues that participate in binding of five mAbs to the conformational epitopes on the TPO surface. Using ELISA we tested the reactivity of single and multiple TPO mutants expressed in CHO cells with a panel of mAbs specifically recognizing IDR-A (mAb 2 and 9) and IDR-B (mAb 15, 18, 64). We show that antibodies recognizing very similar regions on the TPO surface may interact with different sets of residues. We found that residues K713 and E716 contribute to the interaction between mAb 2 and TPO. The epitope for mAb 9 is critically dependent on residues R646 and E716. Moreover, we demonstrate that amino acids E604 and D630 are part of the functional epitope for mAb 15, and amino acids D624 and K627 for mAb 18. Finally, residues E604, D620, D624, K627, and D630 constitute the epitope for mAb 64. This is the first detailed study identifying the key resides for binding of mAbs 2, 9, 15, 18, and 64. Better understanding of those antibodies' specificity will be helpful in elucidating the properties of TPO as an antigen in autoimmune disorders.     

Immunopathological studies on thyroid immunity. IX. Thyroid and renal amyloidosis in thyroglobulin immunized rabbits

In serial studies of immunopathologic changes in an animal model of thyroiditis 52 rabbits were immunized with either bovine, porcine or human thyroglobulin (Tg) in Freund's complete adjuvant, while another 47 animals served as noninjected or adjuvant injected controls. The immunized animals were divided into two groups, one receiving an initial series of only three Tg injections while the other received, in addition, challenging injections over an 8-week period. The immunized animals were killed over a period of 6-34 months after the last Tg injection, and untreated controls were killed at comparable ages. In Tg immunized animals, lymphocytic thyroiditis was encountered in 25 per cent and thyroid amyloid in 17 per cent; glomerular amyloid was encountered in 44 per cent with diffuse lesions in 8 per cent, nodular lesions in 17 per cent and a mixture of the two in 19 per cent. That the thyroid and glomerular hyaline deposits contained amyloid was shown by various histochemical criteria, as well as by the presence of typical fibrils on electron microscopy. Immunohistochemical studies indicated that the amyloid was predominantly of the AA type. Rabbits receiving challenging Tg injections, in addition to the initial series, showed only thyroiditis and nodular glomerular lesions most of which were amyloid. Whilst the vast majority of rabbits with lymphocytic or amyloidotic responses showed both thyroid and renal lesions, a small percentage of animals showed only a lesion of one or the other of these two organs. It is of interest that the thyroid and renal amyloid lesions, described for the first time with induced thyroglobulin immunity, were not detected in other earlier short term investigations.     

重组嵌合抗人CD22四价基因工程抗体在毕赤酵母中的表达及活性检测

目的:采用酵母多拷贝分泌表达载体pPIC9K表达重组嵌合抗人CD22四价基因工程抗体cRFB4FC和cRFB4CH3,并进行活性检测。方法:采用基因克隆技术获得两抗体的基因,然后将其克隆入表达载体pPIC9K中,构建重组质粒pPIC9K/cRFB4FC和pPIC9K/cRFB4CH3,经测序鉴定正确,电转化入毕赤酵母SMD1163中,将经G418抗性筛选出的重组酵母菌株进行PCR鉴定,经甲醇诱导,表达产物经SDS-PAGE分析和Western blot鉴定,经ELISA检测生物学活性。采用正交试验优化重组酵母菌株的表达条件以提高抗体的表达量。结果:获得重组表达质粒pPIC9K/cRFB4FC和pPIC9K/cRFB4CH3,经G418筛选及PCR鉴定得到高拷贝整合的重组酵母菌株,经甲醇诱导表达出cRFB4FC和cRFB4CH3蛋白。它们在SDS-PAGE上分别表现为相对分子质量(Mr)约326 000和295 000的蛋白区带,在Western blot分析中均可被山羊抗人IgG Fc单克隆抗体(mAb)识别,经ELISA分析它们都具有较高的生物学活性。优化表达条件后,抗体的表达量分别提高了196.4%和151.7%。结论:抗体cRFB4FC和cRFB4CH3在酵母菌SMD1163中成功获得高效分泌表达,并有免疫学活性。     

纯化标签不同融合端的选择对蝎毒肽色谱行为的影响

目的考察组氨酸标签置于抗癫痫肽AEP肽链不同末端对重组肽亲合层析的影响。方法同源建模构建AEP三维结构,预测空间构象对纯化标签的屏蔽效应;构建组氨酸标签不同末端融合表达载体;对比两者相同色谱条件下的层析行为及生物活性。结果 AEP构象分析显示,肽链C末端空间屏蔽效应小;在两者表达量相当的前提下,经亲和层析纯化,每升发酵液可获色谱纯样品,AEP-His6为3.3mg,Met-Gly-His6-AEP为0.8mg;HistagC端融合对rAEP活性影响较N端融合小。结论对于蝎毒这类空间结构简单的小分子活性肽,组氨酸标签融合端的选择对重组肽色谱行为及活性有显著影响。     

Primary and recurrent cytomegalovirus infections have different effects on human herpesvirus-6 antibodies in immunosuppressed organ graft recipients: absence of virus cross-reactivity and evidence for virus interaction.

The relationship between serum antibodies to human herpesvirus-6 (HHV-6) and cytomegalovirus (CMV) infection was studied in immunosuppressed adult organ graft recipients all of whom had IgG to both HHV-6 and Epstein-Barr virus capsid antigen (EBVCA) before operation and who had received an organ or organs from HHV-6 seropositive donors. In primary CMV infection the titre of IgG to HHV-6 rose substantially (between 32- and 512-fold) in eight out of eight patients whereas IgG to EBVCA only rose 32-fold in two patients. Moreover, the HHV-6 responses coincided closely with the CMV seroconversion. Serum absorption studies gave no evidence for antibody cross-reaction between CMV and HHV-6 because the CMV antibody titre could be reduced specifically without affecting HHV-6 antibody titres and vice versa. In recurrent CMV infection, HHV-6 antibody levels rose (32-fold) in three out of eight patients but these changes did not coincide with the CMV antibody response. Similarly, in the complete absence of CMV infection, five out of eight patients showed antibody rises to HHV-6 (between four- and 16-fold). IgG titres to EBVCA were stable in both these groups of patients. It is concluded that there is serological evidence (rising titre greater than or equal to four-fold) for genuine HHV-6 reactivation or, alternatively, for reinfection in 16 out of the 24 patients. This phenomenon was most frequent in primary CMV infection where the largest HHV-6 antibody responses were seen probably because of an, as yet, undetermined interaction between the two viruses.