| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Antithyroid Peroxidase Antibodies from Different Tissues in Hashimoto’s Thyroiditis1
Department of Medicine, University of Sheffield Clinical Sciences Center, Northern General Hospital (R.S.M., M.S.A., E.H.K., P.F.W., A.P.W.), Sheffield, United Kingdom S5 7AU; the Department of Biochemistry, Medical Center of Postgraduate Education (A.G.), Warsaw 01 813, Poland; and the Department of Medicine, Kings College School of Medicine (J.P.B.), Denmark Hill, London, United Kingdom SE5 8RX
Address all correspondence and requests for reprints to: Prof. A. P. Weetman, Department of Medicine, University of Sheffield Clinical Sciences Center, Northern General Hospital, Sheffield, United Kingdom S5 7AU.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
(IgG) and IgG
phage display
combinatorial libraries from the cervical (thyroid-draining) lymph
nodes of 2 Hashimoto’s thyroiditis patients and from the thyroid of 1
patient. After selection with purified recombinant human TPO, up to 10
high affinity IgG clones from each tissue source were analyzed
further. No IgG Fab were detected in the patient with the highest
TPO Ab titer. Sequence analysis of the clones showed restricted heavy
and light chain usage, similar to that in previously published
TPO-reactive Fabs. This was despite the substantially larger sizes of
the initial libraries, the use of lymph node tissue to generate
libraries, and the analysis of the repertoire in patients with
Hashimoto’s thyroiditis rather than Graves’ disease. There was
overall similarity in sequences obtained from lymph node and thyroid
libraries, with higher levels of somatic hypermutation in the former.
The Fab inhibited binding of serum TPO Ab from five patients by
55–95%. These data together with those from previous reports indicate
that although there is no unique Ab gene usage, there is the recurrent
presence of certain variable regions in the high affinity TPO Ab
response. |
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Binding and sequence analysis of thyroid lymphocyte-derived TPO-reactive Fabs have been previously described by two groups (2, 4, 6), principally from patients with Graves’ disease (GD), with evidence of substantial sequence restriction, both within and between patients. These studies were based on the known importance of the thyroid as a site of thyroid autoantibody (AAb) synthesis (7). However, extrathyroidal sites of thyroid AAb synthesis, particularly the draining cervical lymph nodes (LN), also contribute (7). Such specialized lymphoid tissue, as a probable site of antigen-driven somatic hypermutation, may produce a different pattern of restriction from that displayed by the thyroid infiltrate. Only four TPO Ab have been described from two Hashimoto’s thyroiditis (HT) patients (6, 8, 9). We hypothesized that restriction would be less likely to occur in HT, in which there is a more chronic disease process than in GD, more frequent lymphoid follicle formation within the thyroid, and no exposure to antithyroid drugs, which reduce the severity of the thyroid lymphocytic infiltrate and thyroid AAb titers in GD (10, 11).
We have, therefore, investigated the restriction of TPO AAb in HT with
the phage display combinatorial library technique, using both thyroid
tissue and cervical LN tissue draining the thyroid. We have studied a
panel of 37 IgG anti-TPO Fabs from two HT patients, allowing
comparative analysis of the TPO Ab response in different tissues and of
the development of the TPO Ab response. The results show that within
each patient, there appears to be a restricted response to TPO, that
the response to TPO is similar in patients with GD and HT, and that
this response appears to be subject to continuous somatic
hypermutation, even in patients with long standing disease.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Samples of cervical LN tissue were obtained from two female
patients (HT126 and HT131) with HT undergoing subtotal thyroidectomy
for enlarging goiter; thyroid tissue was also obtained from one
patient, but in the other it was not available as the entire material
was submitted for pathological examination. Thyroid histology in both
cases confirmed uncomplicated HT, with lymphocytic infiltration and
germinal center formation present in both patients. The patients had
been diagnosed as suffering from HT for 6 months (HT126) and 6 yr
(HT131) before surgery, were euthyroid during T4 treatment,
and were aged 31 (HT126) and 73 (HT131) yr at the time of surgery.
Patient serum Ab was analyzed by hemagglutination. For patient HT126,
the results were: TgAb negative, 1:16,000 TPO Ab; for patient HT131,
they were 1:16,000 Tg Ab, 1:6,400 TPO Ab. TPO AAb containing light
chains were analyzed on purified TPO-coated enzyme-linked immunosorbent
assay (ELISA) plates (supplied by Cogent Diagnostics, Edinburgh, UK)
using 1:100 dilutions of patient and control sera, with detection by a
1:5,000 dilution of alkaline phosphatase-conjugated anti--chain
(Sigma, Poole, UK).
PCR amplification and Ig chain cloning
Intrathyroidal lymphocytes were prepared using previously described methods (12). LN lymphocytes were prepared by mechanical disruption of the tissue, followed by Ficoll-Hypaque density gradient centrifugation (13). Total ribonucleic acid and first strand complementary DNA were prepared, and PCR amplification was carried out as previously described (13). Primer sequences for Ig gene PCR amplification have been previously described (13). After purification (Magic PCR preps, Promega, Southampton, UK), Ig PCR products were digested with restriction enzymes and agarose gel purified as described previously (13). They were then ligated into digested pComb3 vector and transformed by electroporation as described previously (13). Recombinant phage were prepared and stored in sterile PBS at -20 C (13, 14).
Library screening and Fab solubilization
Library screening was carried out as described previously (13, 14). Briefly, ELISA wells were coated overnight at 4 C with 100 µL 50 µg/mL purified recombinant human TPO (13, 14, 15). After blocking and washing, 100 µL phage library were added (typically 1011 plaque-forming units) and incubated for 2 h at 37 C. Unbound phage were removed, and wells were washed 10 times with excess PBS with 0.05% (vol/vol) Tween-20 over a period of 1 h at room temperature. Adherent phage were eluted and propagated as described previously (13, 14). Libraries were selected between three and five times. After Fab solubilization, transformation, and plating, individual clones were grown, and plasmid DNA was prepared as described previously (13). For Fab preparation, clones were grown at 37 C for 6 h in 2 mL super broth medium containing 50 µg/mL ampicillin (13), and Fab synthesis was induced by addition of 18 mL super broth medium containing 50 µg/mL ampicillin and 1 mmol/L isopropylthio-ß-D-galactoside (Sigma) followed by overnight incubation at 28 C. The culture was centrifuged at 1000 x g for 10 min at 4 C, and the supernatant was removed as a source of Fab. Additional Fab was prepared by resuspension of the bacterial pellet in 800 µL PBS containing 2 µg/mL aprotinin, 1 µg/mL leupeptin, 1 µg/mL pepstatin, and 0.1 mmol/L phenylmethylsulfonylfluoride (all from Sigma) and disruption of the cells by three freeze-thaw cycles (-80 C/37 C). Debris was pelleted by centrifugation at 3000 x g, and the supernatant was combined with the culture supernatant for analysis of Fab reactivity.
Determination of Fab specificity and approximate affinity
Initial screening of Fab specificity was carried out by ELISA using purified human TPO-coated ELISA plates (Cogent Diagnostics), detected using alkaline phosphatase-conjugated antihuman IgG (Fab-specific; Sigma), and compared to 1:100 dilutions of patient and control sera. Cross-reactivity of the Fab was determined by competitive ELISA against purified human Tg (13) and bovine lactoperoxidase (Sigma), both coated at 10 µg/mL. Inhibition of patient serum TPO Ab binding by Fab supernatants was determined by ELISA, using both commercial TPO-coated plates (Cogent Diagnostics) and plates coated with 100 µL 1 µg/mL purified human TPO (Calbiochem, Nottingham, UK). After an initial 2-h preincubation at 37 C with 100 µL of a 1:10 dilution of Fab, residual binding of 1:1,000 dilutions of serum (from five patients with high anti-TPO titers) was measured using antihuman IgG (Fc-specific) conjugated to alkaline phosphatase (1:40,000 dilution; Sigma). Control wells indicated that the antihuman IgG (Fc-specific) conjugate showed no reactivity to the Fab. Inhibition of patient serum Ab binding to Tg by Fab supernatants was similarly determined by ELISA using wells coated with purified human Tg at 10 µg/mL.
Approximate Fab affinity was determined using two methods. Duplicate
aliquots of Fab were diluted 1:10 in assay buffer (0.15 mol/L NaCl; 10
mmol/L Tris-HCl, pH 7.5; and 0.5% BSA) and incubated with 10,000 cpm
125I-labeled purified human TPO (RSR, Cardiff, UK),
dilutions of purified human TPO (10-11-10-8
mol/L final concentration; Calbiochem), and mouse antihuman light
chain (Serotec, Oxford, UK) in a total volume of 100 µL. After 1
h at room temperature, 100 µL donkey antimouse Sac-cel (IDS, Boldon,
Tyne and Wear, UK) was added, and the incubation was continued for an
additional 30 min. The immune complexes were sedimented by
centrifugation at 1,000 x g for 5 min and then
counted. Alternatively, Fab were diluted 1:10 in PBS, and affinity was
determined by ELISA in the presence of dilutions of purified human TPO
on purified human TPO-coated ELISA plates, with bound Fab detected
using antihuman IgG (Fab-specific) conjugate. Approximate Fab affinity
was defined as the concentration of unlabeled TPO that caused 50%
inhibition of maximum labeled TPO binding or Fab binding.
Sequencing of heavy and light chains
DNA was manually sequenced in both directions using Sequenase
version 2.0 and [
-35S]deoxy-ATP (both from Amersham,
Aylesbury, UK), and using automated sequencing (Taq Dye
deoxy Terminator cycle sequencing kit, Applied Biosystems, Foster City,
CA) using an Applied Biosystems model 373A. V, D, and J regions were
assigned using the GCG analysis package on the EMBL database (16).
Where possible, VH and V region nomenclature
conform to those of Matsuda et al. (17) and Schäble
and Zachau (18) respectively.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The heavy chain libraries contained 1.38 x 106,
0.44 x 106, and 2.1 x 106
recombinant colony-forming units (CFU; HT126 thyroid, HT126 LN, and
HT131 LN libraries, respectively). The final IgG phage libraries
contained 90.5 x 106, 7.4 x 106,
and 53.0 x 106 double recombinant CFU (HT126 thyroid,
HT126 LN, and HT131 LN, respectively). IgG libraries were also made,
containing 38.0 x 106, 10.1 x 106,
and 35.5 x 106 double recombinant CFU (HT126 thyroid,
HT126 LN, and HT131 LN, respectively). IgG libraries were panned
three times (HT126 thyroid 3' and HT126 LN; sequencing codes 126A-J and
126TP, respectively) or five times (HT126 thyroid 5' and HT131 LN;
sequencing codes 126TO and 131TP, respectively).
After selection, between 8–10 anti-TPO Fab from each TPO-selected
library were analyzed further (Table 1
);
all selected Fab had absorbances of greater than 50% of a 1:100
dilution of HT126 patient serum, with over half having absorbances
greater than a 1:100 dilution of serum. Simultaneous screening of the
patient HT126 IgG libraries failed to detect any TPO-reactive Fab
even after five rounds of antigen selection; this observation was
consistent with the small amount of detectable IgG TPO Ab in this
patient [control sera (n = 5), 0.155 ± 0.037; HT patient
sera (n = 9), 0.221 ± 0.037; patient HT126, 0.191).
|
Sequences were grouped according to similarity of V, (D), and J
region usage and similarity across the V/(D)/J junction (Table 1
); each
group, therefore, represented the products of somatic hypermutation
from a single V(D)J recombination event. For each group, minimum and
maximum numbers of somatic mutation events from the putative
originating V(D)J sequence are shown in Table 1. Amino acid alignments
for the heavy chain V and DJ regions are presented in Figs. 1 and 2,
respectively; where sequences are identical, only a single sequence is
shown.
|
|
). From
patient HT131, all Fab contained VH3 heavy chains, with all
but one Fab (131TP14; VH3DP58/D2/JH6) falling
into a single group, VH3–23/D3/JH6 (group H3B;
Table 1). All heavy chain sequences differed substantially from the
putative most homologous germline variable genes (Fig. 1). The patterns
of mutation in the heavy chains from patient HT126 were consistent with
somatic hypermutation. Although replacement/silent mutation analysis of
mutations in the heavy chains from patient HT131 were not entirely
typical of those caused by somatic hypermutation, the presence of
several mismatches in the J segment did indicate active somatic
hypermutation (Fig. 2). The VH3 sequences from the two
patients were derived from different VH3 genes and showed
little similarity in either the pattern of somatic hypermutation or the
junction sequence (Figs. 1 and 2). Different heavy chain sequences were obtained from the three HT126 TPO-selected libraries; in particular, only group H3A sequences were detected in the thyroid 3' library. In addition, there was typically a greater degree of somatic hypermutation in the sequences from the LN and thyroid 5' library than in the thyroid 3' library. Analysis of the small section of constant region sequence available using the previously described sequencing primers (13) indicated that all Fab were of the IgG1 subclass. Further sequencing of the constant regions of three Fab (126H, 126TO2, and 131TP7) confirmed this result.
-chain sequence analysis
The light chain sequences were more restricted than those of the
heavy chains, with all but one sequence (131TP14;
VIII/J3) being VI, and the
majority of sequences were J4 (Table 1). From patient
HT126, two principal groups of VI/J4
sequence (groups VIB and VID) were found,
each associated with one of the two major heavy chain groups. In
addition, there were a number of minor
VI/J4 sequence groups, each derived from
a distinct V/J recombination event (Table 1 and Figs. 3 and 4).
All but three of the sequences from patient HT126 were
J4; the remainder were J5. Within group
VIB, there were examples of both germline and
somatically hypermutated sequences, with germline sequences
predominating (Table 1). Only somatically hypermutated species were
present in group VID. There was a greater degree of
somatic hypermutation in the Fab from the LN and thyroid 5' libraries,
although much less than that observed in the heavy chains.
|
|
I groups, both with a high degree of homology to
patient HT126 group VIB (sharing the germline
VO12; Table 1 and Fig. 3). The remaining HT131 sequence,
from clone 131TP14, was the only non-VI sequence present
(Table 1 and Fig. 3). Four of the eight sequences from patient HT131
contained J1, and four contained J3
(Table 1 and Fig. 4); thus, J homology was not shared
between the patients, unlike V homology. Characterization of anti-TPO Fab reactivity
The Fab showed no detectable cross-reactivity with either purified
human Tg or bovine lactoperoxidase (data not shown). The approximate
affinity of the Fab was determined by two methods, because although Fab
containing the VO12 light chain bound the
125I-labeled purified human TPO preparation used, those not
containing the VO12 light chain did not (data not
shown). Abolition of binding of the murine monoclonal anti-TPO Ab, mAb
9, by iodination of tyrosine residues has been described previously,
and this mAb binds to a TPO domain to which patient AAb bind (3). For
the Fab showing no reactivity toward the labeled TPO, an alternative
method to establish affinity was used, with the affinity of a single
Fab (126TP13) determined using both methods to allow comparison (Table 2). Approximate affinity for TPO varied
over the range 10-10-10-8 mol/L (Table 2),
similar to data previously reported (2, 4, 6, 8). Affinity was
determined for the majority of Fab to allow correlation between the
degree of heavy chain somatic mutation and affinity for TPO, but no
clear pattern emerged from this analysis. This was probably due at
least in part to differences in light chain sequence masking the
effects of heavy chain differences and vice versa.
|
IB; 126B and 126TO10
from group H3A/VID; and 131TP2, 131TP7, and 131TP15 from
groups H3B/VIH and H3B/VII. These Fab
represented the two predominant classes of Fab detected in patients
HT126 and HT131, respectively, and displayed binding characteristics
different from those of iodinated TPO, suggesting that they may bind
different epitopes on TPO. The Fab inhibited serum TPO Ab binding to
purified human TPO-coated ELISA plates by between 55.0–95.3% (Fig. 5). Similar results were obtained from
both types of TPO-coated plates (data not shown). Inhibition of TPO Ab
binding was similar for single Fab when tested individually, typically
with no increased inhibition of binding in the presence of a second
Fab, suggesting that the Fab were interacting with the same or
neighboring epitopes on TPO. The TPO-reactive Fab did not inhibit
binding of patient sera to Tg (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
anti-TPO Fabs
from two HT patients, using a previously unexamined tissue source, the
LN draining the thyroid. In one patient, we compared anti-TPO sequences
from LN and thyroid and also compared Fab from three and five rounds of
antigen selection from the thyroid library. In this patient, sequences
from the thyroid 5' library most closely resembled those from the LN
library, possibly indicating the outgrowth of less frequent clones that
were more common in the starting LN than thyroid libraries. In
addition, sequences from the HT126 LN library tended to show a greater
number of somatic mutations than those from the thyroid library.
Overall, however, the sequences obtained form the two tissue sources
overlapped considerably, with two principal heavy chain types, each
associated with a distinct VI region. The repertoires in
the two patients showed some similarities, but in other respects they
were distinct. TPO-reactive Fab sequences have been published previously, with several heavy/light chain pairings described by two groups studying patients with GD and HT (2, 4, 6, 8, 9). The current study differs in four respects from those previously published; the use of a probable site of TPO Ab somatic hypermutation as a tissue source, the use of combinatorial libraries greater than 10 times larger than those used previously, the use of a more complete panel of Ig primers for library construction, and analysis of patients with HT, from whom few sequences were available. Despite this, there is a remarkable similarity between our results and those published previously, suggesting that the TPO Ab repertoires in GD and HT patients overlap considerably. The only reported differences in TPO Ab reactivity between patients with GD and HT is toward linear epitopes of TPO (5, 27). Like those published previously, these Fab did not bind denatured TPO (28).
In common with these sequences, the majority of previously published
sequences have been either VH1 or VH3 and
VI or VIII (2, 4, 6, 9). VH1
and VH3, and VI and VIII are
the most common V segments, in terms of both numbers of germline genes
and the expressed adult repertoire (29, 30). Reported J region usage
has been more diverse, with the JH4, JH6, and
J2 regions found in most of the previously published
sequences, and the JH4, JH5, JH6,
J1, J3, J4, and
J5 regions found in the Fab reported here.
JH4, JH5, and JH6, and
J1, J2, and J4 are the
most common J regions in the adult repertoire (30, 31). The TPO Ab
repertoire described to date, therefore, primarily contains the most
commonly used genetic elements. In addition, no general CDR3 (V/(D)/J
junction) motifs could be discerned; in particular, the VH1
CDR3 glycine-rich motif reported in Fab SP4.6 and 6F (6) was not
present in any of the Fab reported here. Several V segments are shared
between these and previously reported sequences (2). In particular, the
VI germline gene VO12 was also present in
TPO-specific Fab from five of six AITD patients (2).
There was a substantial amount of somatic hypermutation present, particularly in the heavy chain sequences; where somatic hypermutation was evident, the putative originating sequences were generally not detected. This suggests that somatic mutation of these germline sequences is necessary to generate high affinity anti-TPO responses, and that the germline gene pairings might be expected to have a relatively lower affinity for TPO. In agreement with previous observations (2, 4, 6, 9), we found a greater degree of somatic hypermutation in heavy chain sequences than in light chain sequences; indeed, several of the light chain sequences showed no evidence of somatic hypermutation. Analysis of individual somatic hypermutation and germline-derived differences between previously reported TPO Fab sequences and those reported here reveals that most differences are unique to individual patients.
In agreement with previous studies, we showed evidence of a restricted
pattern of pairing between heavy and light chains in the
VH1 and VH3 Fabs, particularly apparent in
patient HT126. Despite the fact that both major heavy chain groups of
Fabs from this patient had VI sequences, there was no
overlap in light chain sequence between the groups. In a study forcing
recombination between heavy and light chains from Fab specific for
different domains of TPO, the resulting Fab did not bind TPO,
suggesting specific (nonpromiscuous) pairing of heavy and light chains
specific for different domains on TPO (32).
No IgG Fab reactive to TPO were detected in either of the two
patient HT126 libraries screened, even after five rounds of panning.
This was probably not due to a problem inherent in the libraries
themselves, as high affinity IgG Fab (reactive to thyroglobulin) had
previously been obtained from the patient HT131 LN library (13).
However, patient HT126 serum IgG anti-TPO Ab were of a much lower
apparent titer than those of the IgG light chain type, which appears
to be a common pattern in patients with AITD (33, 34), and only four
TPO-reactive IgG Fab have been described (2).
Although they are very common in AITD, the evidence suggesting that TPO Ab are directly pathogenic is currently inconclusive (1, 2). TPO Ab are not usually found in either spontaneous or induced animal models of AITD (1). IgG1 and IgG4 typically account for more 80% of patient serum TPO Ab activity, with IgG3 generally being undetectable (33, 34). Although the functional affinity of IgG1 TPO Ab is reported as being generally less than that of IgG4 TPO Ab (34), they are nevertheless complement fixing and, therefore, potentially more damaging. Although the amount of ADCC detected using thyroid cells and AITD patient sera did not correlate with TPO Ab titers, and preincubation of sera with purified TPO did not remove ADCC activity from the sera (35), direct analysis of IgG1 and IgG4 TPO Ab in ADCC assays has indicated that IgG1 are more damaging (36). The Fab in this study were all from the IgG1 subclass, indicating that they may well represent pathologically active species. In addition to direct pathogenic effects, B cells producing TPO AAb may influence the T cell response to TPO, for example by effects on TPO processing and the altered display of T cell epitopes (37). A greater understanding of TPO Ab structure and function may, therefore, provide important information on their role in immunopathogenesis.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Division of Molecular and Cellular Immunology,
Department of Clinical Laboratory Sciences, Floor A, West Block,
Queen’s Medical Center, Nottingham, United Kingdom NG7 2UH.
3 Sponsored by the Ministry of Education, Pakistan.
Received October 23, 1996.
Revised July 2, 1997.
Accepted July 23, 1997.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
and IgG
anti-thyroglobulin autoantibodies from a patient with Hashimoto’s
thyroiditis: evidence for in vivo antigen-driven repertoire
selection. J Immunol. 157:927–935.[Abstract]
type. J Mol Biol. 176:189–196.[CrossRef][Medline]
immunoglobulin gene connecting the J and C regions reveals a matrix
association region (MAR) next to the enhancer. Nucleic Acids Res. 20:4929–4930. anti-thyroid peroxidase antibodies in Hashimoto’s
thyroiditis: epitope mapping analysis. J Clin Endocrinol Metab. 82:2639–2644. genes. Ann NY Acad Sci. 764:74–83.[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  M. Dubska, J. P. Banga, D. Plochocka, G. Hoser, E. H. Kemp, B. J. Sutton, A. Gardas, and M. Gora Structural Insights into Autoreactive Determinants in Thyroid Peroxidase Composed of Discontinuous and Multiple Key Contact Amino Acid Residues Contributing to Epitopes Recognized by Patients' Autoantibodies Endocrinology, December 1, 2006; 147(12): 5995 - 6003. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Stamatopoulos, C. Belessi, A. Hadzidimitriou, T. Smilevska, E. Kalagiakou, K. Hatzi, N. Stavroyianni, A. Athanasiadou, A. Tsompanakou, T. Papadaki, et al. Immunoglobulin light chain repertoire in chronic lymphocytic leukemia Blood, November 15, 2005; 106(10): 3575 - 3583. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Latrofa, P. Pichurin, J. Guo, B. Rapoport, and S. M. McLachlan Thyroglobulin-Thyroperoxidase Autoantibodies Are Polyreactive, Not Bispecific: Analysis Using Human Monoclonal Autoantibodies J. Clin. Endocrinol. Metab., January 1, 2003; 88(1): 371 - 378. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Chapal, T. Chardes, D. Bresson, M. Pugniere, J.-C. Mani, B. Pau, M. Bouanani, and S. Peraldi-Roux Thyroid Peroxidase Autoantibodies Obtained from Random Single Chain Fv Libraries Contain the Same Heavy/Light Chain Combinations as Occur in Vivo Endocrinology, November 1, 2001; 142(11): 4740 - 4750. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Jirholt, L. Strandberg, B. Jansson, E. Krambovitis, E. Soderlind, C. A.K. Borrebaeck, R. Carlsson, L. Danielsson, and M. Ohlin A central core structure in an antibody variable domain determines antigen specificity Protein Eng. Des. Sel., January 1, 2001; 14(1): 67 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Hobby, A. Gardas, R. Radomski, A. M. McGregor, J. P. Banga, and B. J. Sutton Identification of an Immunodominant Region Recognized by Human Autoantibodies in a Three-Dimensional Model of Thyroid Peroxidase Endocrinology, June 1, 2000; 141(6): 2018 - 2026. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 , , , , , , , , , and Human Anti-Thyroid Peroxidase Single-Chain Fragment Variable of Ig Isolated from a Combinatorial Library Assembled In-Cell: Insights into the In Vivo J. Immunol., April 15, 2000; 164(8): 4162 - 4169. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Metcalfe, R. S. McIntosh, F. Marelli-Berg, G. Lombardi, R. Lechler, and A. P. Weetman Detection of CD40 on Human Thyroid Follicular Cells: Analysis of Expression and Function J. Clin. Endocrinol. Metab., April 1, 1998; 83(4): 1268 - 1274. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
